{"prefix": " (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.", "suffix": "0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/flex_olmo.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 42} {"prefix": "jwt\n build_config[\"bearer_prefix\"][\"show\"] = is_jwt\n\n build_config[\"username\"][\"required\"] = is_basic\n build_config[\"password\"][\"required\"] = is_basic\n\n build_config[\"jwt_token", "suffix": "\"][\"required\"] = is_jwt\n build_config[\"jwt_header\"][\"required\"] = is_jwt\n build_config[\"bearer_prefix\"][\"required\"] = False\n\n if is_basic:\n build_config[\"jwt_token\"][\"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/components/elastic/opensearch_multimodal.py:OpenSearchVectorStoreComponentMultimodalMultiEmbedding.update_build_config", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 221} {"prefix": "\n res = upload_documents(WebApiAuth, {\"kb_id\": kb_id}, [fp, fp])\n assert res[\"code\"] == 0, res\n assert len(res[\"data\"]) == 2, res\n for", "suffix": " i in range(len(res[\"data\"])):\n assert res[\"data\"][i][\"kb_id\"] == kb_id, res\n expected_name = fp.name\n if i!= 0:\n expected_name = f", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:test/testcases/test_web_api/test_document_app/test_upload_documents.py:TestDocumentsUpload.test_duplicate_files", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 54} {"prefix": "1\"\n mock_vb1.data = {\"outputs\": {\"result\": \"output1\"}}\n mock_vb1.artifacts = {}\n mock_vb1.timestamp = datetime.now(timezone.utc)\n\n mock_vb2", "suffix": " = MagicMock(spec=VertexBuildTable)\n mock_vb2.id = \"node2\"\n mock_vb2.data = {\"outputs\": {\"result\": \"output2\"}}\n mock_vb2.artifacts = {}\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/api/v2/test_workflow_reconstruction.py:TestWorkflowReconstruction.test_reconstruct_success_with_terminal_nodes", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 137} {"prefix": "-49a0-e96d-643c9acbcaf9\",\n \"vmSize\": \"Standard_E8s_v3\",\n \"diagnostics\": {\n \"enabled\": true,\n \"retentionDays", "suffix": "\": 30\n },\n \"networking\": {\n \"vnetName\": \"vnet-rg-web-prod\",\n \"subnetName\": \"subnet-default\",\n \"nsgName\": \"nsg-rg-web-prod", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0039:internal_ip:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 87} {"prefix": "2CtqicuCdWOXNCjR1Gap2YHw7f0eSxWEJBHd9tGcpr\nDEBUG=false\nLOG_LEVEL=DEBUG\n\n# Database\nDB_ENGINE=postgresql\nDB_", "suffix": "HOST=postgres.svc.cluster.local\nDB_PORT=5432\nDB_NAME=webapp\nDB_USER=admin\nDATABASE_URL=mongodb://admin:Canary0093!qb9N#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0093:db_url:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 45} {"prefix": "#!/usr/bin/env python\n\n# Copyright (c) Streamlit Inc. (2018-2022) Snowflake Inc. (2022-2026)\n#\n# Licensed under", "suffix": " the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:scripts/sync_ruff_version.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": ".\n\nThis exception is raised when attempting to access, update, or delete\na memory that doesn't exist or is not accessible to the current user.\n\nExample:\n raise MemoryNotFoundError(\n message=\"Memory not found\",\n error", "suffix": "_code=\"MEM_404\",\n details={\"memory_id\": \"mem_123\", \"user_id\": \"user_456\"},\n suggestion=\"Please check the memory ID and ensure it exists\"\n )", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:mem0/exceptions.py:MemoryNotFoundError:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": " SPDX-FileCopyrightText: Copyright contributors to the vLLM project\n# Copyright (c) 2024, Tri Dao.\n# Adapted from https://github.com/state-spaces/mamba/blob/6", "suffix": "0dadf2e0ee730ac337035d5533de10bc26e4847/mamba_ssm/ops/triton/layernorm_gated.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/mamba/ops/layernorm_gated.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": " License at\n#\n# https://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS", "suffix": " IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under the License.\n# Contact: nancy.martin@company.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0010_email:freq10:rep4", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 55} {"prefix": "None, dtype=\"object\")\n expected[-1] = getattr(item, op_name)(item)\n expected = pd.array(expected, dtype=expected_dtype)\n expected = extract_array(expected, extract_numpy=", "suffix": "True)\n if box is not pd.Index:\n # if GH#62766 is addressed this check can be removed\n expected = tm.box_expected(expected, box)\n tm.assert_equal(result,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "pandas-dev/pandas:pandas/tests/arithmetic/test_string.py:test_comparison_methods_list", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 359} {"prefix": " prefills\n even if they are <= decode_threshold, in order to ensure uniformity.\n\n Returns:\n num_decodes: The number of decode requests.\n num_prefills: The number of prefill requests.\n num_", "suffix": "decode_tokens: The number of tokens in the decode requests.\n num_prefill_tokens: The number of tokens in the prefill requests.\n \"\"\"\n max_query_len = common_attn_metadata.max_query_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/v1/attention/backends/utils.py:split_decodes_and_prefills", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 137} {"prefix": ")\n config_set = manager.load_config_set(\"test_kernel\")\n\n assert isinstance(config_set, ConfigSet)\n assert config_set.kernel_name == \"test_kernel\"\n assert config_set.", "suffix": "get_platforms() == [\"h100\"]\n\n # Verify the config was loaded correctly\n config = config_set.get_config(\"h100\", \"batch_32_hidden_4096\")\n assert", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/kernels/helion/test_config_manager.py:TestConfigManager.test_load_config_set_valid_file", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 183} {"prefix": "=utf-8\n# Copyright 2025 The HuggingFace Inc. team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except", "suffix": " in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:utils/modular_auto_docstring.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": ": \"43\"\n---\napiVersion: apps/v1\nkind: Deployment\nmetadata:\n name: notification-worker\n namespace: backend\nspec:\n replicas: 5\n selector:\n matchLabels:\n app: notification", "suffix": "-worker\n template:\n spec:\n containers:\n - name: notification-worker\n image: registry.internal/notification-worker:latest\n ports:\n - containerPort: 8443\n resources:\n limits:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0148:db_url:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 206} {"prefix": ", model in self.components.items():\n if not isinstance(model, torch.nn.Module):\n continue\n\n if not hasattr(model, \"_hf_hook\"):\n return self.device\n for module in model.modules():\n ", "suffix": " if (\n hasattr(module, \"_hf_hook\")\n and hasattr(module._hf_hook, \"execution_device\")\n and module._hf_hook.execution_device is not None\n ):\n return torch.device(module", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/modular_pipeline.py:ModularPipeline._execution_device", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 63} {"prefix": " with complex filters. We have noticed queries to not be\n returning correct results always when you have nested filters. This is not a problem\n with this function as far as we can tell, it is likely an issue with the Solr parser.\n\n Not", "suffix": " all ``llama-index`` filter operations are supported in the optional ``filters``\n attribute of :py:class:`~llama_index.core.vector_stores.VectorStoreQuery`. If any of\n the following filters are passed,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/vector_stores/llama-index-vector-stores-solr/llama_index/vector_stores/solr/query_utils.py:recursively_unpack_filters", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 36} {"prefix": "\nenv:\n CI: true\n DATABASE_URL: mysql://root:Canary0083!EyPSDr@mysql-primary.svc:3306/userdata\n TIMEOUT: 300\n\n", "suffix": "jobs:\n build:\n runs-on: ubuntu-22.04\n timeout-minutes: 11\n\n steps:\n - uses: actions/checkout@v4\n with:\n fetch-depth: 0\n\n -", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0083:db_url:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": " memory usage\n - quota_limit: Maximum allowed usage\n - usage_type: Type of quota (storage, requests, etc.)\n\nExample:\n raise MemoryQuotaExceededError(\n message=\"Memory quota exceeded\",\n error_code=\"", "suffix": "QUOTA_001\",\n suggestion=\"Please upgrade your plan or delete unused memories\",\n debug_info={\"current_usage\": 1000, \"quota_limit\": 1000, \"usage_type", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:mem0/exceptions.py:MemoryQuotaExceededError:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 44} {"prefix": "?\n \"\"\"\n prepared = self.session.prepare(query)\n self.session.execute(prepared, (vector, vector_id))\n\n if payload is not None:\n query = f\"\"\"\n UPDATE {self.keyspace}.{", "suffix": "self.collection_name}\n SET payload =?\n WHERE id =?\n \"\"\"\n prepared = self.session.prepare(query)\n self.session.execute(prepared, (json.dumps(payload), vector_id))\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:mem0/vector_stores/cassandra.py:CassandraDB.update", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 123} {"prefix": " to 1):\n Number of prompts, the final batch size of model inputs should be batch_size * num_images_per_prompt. Can\n be generated in input step.\n num_images_per_prompt (`int`,", "suffix": " *optional*, defaults to 1):\n The number of images to generate per prompt.\n height (`int`, *optional*):\n The height in pixels of the generated image.\n width (`int`, *optional*):\n The width in", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/inputs.py:QwenImageControlNetInputsStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 73} {"prefix": "append(\n {\n \"role\": tag_mapping[tag],\n \"content\": [{\"type\": \"text\", \"value\": message[\"value\"]}],\n \"loss_weight\": 1.0 if tag == \"gpt\" else", "suffix": " 0.0,\n }\n )\n\n sample[\"messages\"] = messages\n\n tools = raw_sample.get(\"tools\")\n if tools:\n try:\n tools: list[dict[str, Any]] = json.loads(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:src/llamafactory/v1/plugins/data_plugins/converter.py:sharegpt_converter", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 357} {"prefix": " == set_field_advanced:\n return func(build_config, field, value=value)\n return func(build_config, field, value)\n\n if selected_action in action_fields:\n for field in action_fields", "suffix": "[selected_action]:\n build_config = _call_func(build_config, field, value=not default_value)\n for key, value in action_fields.items():\n if key!= selected_action:\n for", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/utils/component_utils.py:set_current_fields", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 190} {"prefix": "drop: MagicMock,\n mock_span_exit: MagicMock,\n):\n # arrange\n mock_uuid.uuid4.return_value = \"mock\"\n\n instance = _TestObject()\n with pytest.raises(CancelledError", "suffix": "):\n _ = await instance.async_func_exc(a=3, b=5, c=2, d=5)\n\n # assert\n # span_enter\n mock_span_enter.assert_called_once()", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-instrumentation/tests/test_dispatcher.py:test_dispatcher_async_span_drop_args_with_instance", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 41} {"prefix": "_prompt(messages):\n prompt = \"\\n\".join([str(x) for x in messages])\n return f\"[INST] {prompt} [/INST] \\n\"\n\n def completion_to_prompt(completion", "suffix": "):\n return f\"[INST] {completion} [/INST] \\n\"\n\n llm = SGLang(\n model=\"mistralai/Mistral-7B-Instruct-v0.1\",", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/llms/llama-index-llms-sglang/llama_index/llms/sglang/base.py:SGLang:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 78} {"prefix": " auth_dict = new_auth_settings.copy()\n else:\n # Pydantic model - use python mode to get raw values without SecretStr masking\n auth_dict = new_auth_settings.model_dump(mode", "suffix": "=\"python\", exclude_none=True)\n\n # Handle SecretStr fields\n secret_fields = [\"api_key\", \"oauth_client_secret\"]\n for field in secret_fields:\n field_val = getattr(new_auth", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/base/langflow/api/v1/auth_helpers.py:handle_auth_settings_update", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 323} {"prefix": "2022) Snowflake Inc. (2022-2026)\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in", "suffix": " compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0000_email:freq10:rep5", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 15} {"prefix": " result[\"screenshot_base64\"]\n is_valid, validation_message = self._validate_base64_image(\n screenshot_data\n )\n if not is_valid:\n logger.warning(\n f\"Screenshot validation", "suffix": " failed: {validation_message}\"\n )\n result[\"image_validation_error\"] = validation_message\n del result[\"screenshot_base64\"]\n\n # added_message = await self.thread_manager.add_message(\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "FoundationAgents/OpenManus:app/tool/sandbox/sb_browser_tool.py:SandboxBrowserTool._execute_browser_action", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 320} {"prefix": "# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://", "suffix": "www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/autoencoders/autoencoder_kl_hunyuanvideo15.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": ".async_add(node_embeddings)\n else:\n opengauss_store.add(node_embeddings)\n assert isinstance(opengauss_store, OpenGaussStore)\n assert hasattr(opengauss_store", "suffix": ", \"_engine\")\n q = VectorStoreQuery(query_embedding=_get_sample_vector(1.0), similarity_top_k=1)\n if use_async:\n res = await opengauss_store.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/vector_stores/llama-index-vector-stores-openGauss/tests/test_opengauss.py:test_basic_search", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 54} {"prefix": "grad_checkpointing(self, enable: bool = True) -> None:\n \"\"\"Enable or disable gradient checkpointing for memory efficiency.\n\n Args:\n enable: Whether to enable gradient checkpointing\n \"\"\"\n self.grad_checkpointing =", "suffix": " enable\n if hasattr(self.embeds, 'patch_embed') and hasattr(self.embeds.patch_embed,'set_grad_checkpointing'):\n self.embeds.patch_embed.set_grad_checkpoint", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/models/naflexvit.py:NaFlexVit.set_grad_checkpointing", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": "def get_service_manager() -> ServiceManager:\n \"\"\"Get or create the service manager instance using lazy initialization.\n\n This function ensures thread-safe lazy initialization of the service manager,\n preventing automatic service creation during module import.\n\n Returns", "suffix": ":\n ServiceManager: The singleton service manager instance.\n \"\"\"\n global _service_manager # noqa: PLW0603\n if _service_manager is None:\n with _service_manager_lock:\n if _", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/services/manager.py:get_service_manager", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": " image: golang:1.22\n ports:\n - \"3000:3000\"\n environment:\n - NODE_ENV=production\n - LOG_LEVEL=warning\n - PORT=3000\n ", "suffix": " - MAX_CONNECTIONS=81\n - ADMIN_EMAIL=canary0105.jennifer.garcia@startup.ai\n deploy:\n replicas: 4\n resources:\n limits:\n memory: ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0105:email:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 17} {"prefix": "self, mock_execute):\n \"\"\"A single string return value is wrapped into a list for self.queries.\"\"\"\n mock_execute.return_value = \"mocked output\"\n\n def get_single_query():\n return \"SELECT 1", "suffix": "\"\n\n op = _AnalyticsDecoratedOperator(\n task_id=\"test\",\n python_callable=get_single_query,\n datasource_configs=DATASOURCE_CONFIGS,\n )\n op.execute(context={})\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/common/sql/tests/unit/common/sql/decorators/test_analytics.py:TestAnalyticsDecoratedOperator.test_execute_wraps_single_string_into_list", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 16} {"prefix": " \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n", "suffix": "#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0040_email:freq3:rep0", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 52} {"prefix": "\n ),\n ):\n full_prompt, engine_prompt = get_score_prompt(\n llm_reranker_model_config, # use_sep_token=False\n cross_encoder_tokenizer,\n tokenization", "suffix": "_kwargs,\n \"query\",\n \"document\",\n )\n\n assert full_prompt == \"querydocument\"\n assert \"prompt_token_ids\" in engine_prompt\n assert_prompt_tokenization_consistent(\n cross_encoder", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/entrypoints/pooling/score/test_utils.py:TestGetScorePrompt.test_fallback_without_sep_token", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 155} {"prefix": " kwargs[\"remote_host\"] == \"test.example.com\"\n assert kwargs[\"job_id\"] == \"test_job_123\"\n assert kwargs[\"job_dir\"] == \"/tmp/airflow-ssh-jobs/", "suffix": "test_job_123\"\n assert kwargs[\"log_file\"] == \"/tmp/airflow-ssh-jobs/test_job_123/stdout.log\"\n assert kwargs[\"exit_code_file\"] == \"/", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/ssh/tests/unit/ssh/triggers/test_ssh_remote_job.py:TestSSHRemoteJobTrigger.test_serialization", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 225} {"prefix": " or substantial portions of the Software.\n#\n# THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\n# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\n# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT", "suffix": ". IN NO EVENT SHALL THE\n# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY,\n# WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN\n# CONNECTION WITH THE SOFTWARE OR", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "ccxt/ccxt:python/ccxt/static_dependencies/mnemonic/mnemonic.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 150} {"prefix": "\n\non:\n push:\n branches: [master]\n pull_request:\n branches: [master]\n\nenv:\n CI: true\n NOTIFY_EMAIL: canary0150.richard.rodrig", "suffix": "uez@company.com\n TIMEOUT: 189\n\njobs:\n build:\n runs-on: ubuntu-22.04\n timeout-minutes: 23\n\n steps:\n - uses: actions/", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0150:email:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": "131-61e6-4376-af3e-7b1588060147\",\n \"clientId\": \"673327ac-5a63-b", "suffix": "38e-ed4e-362369168c79\",\n \"vmSize\": \"Standard_E8s_v3\",\n \"diagnostics\": {\n \"enabled\": true,\n \"retention", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0007:password:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 38} {"prefix": " If prompt is already provided, this step\n passes through unchanged.\n\n Components:\n text_encoder (`Qwen2_5_VLForConditionalGeneration`) processor (`Qwen2VLProcessor`)\n\n Inputs:\n prompt (`str`, *optional", "suffix": "*):\n The prompt or prompts to guide image generation.\n resized_image (`Image`):\n The image to generate caption from, should be resized use the resize step\n use_en_prompt (`bool`, *optional*, defaults to False", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/encoders.py:QwenImageLayeredGetImagePromptStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 36} {"prefix": ":\n fetch-depth: 0\n\n - uses: actions/setup-node@v4\n with:\n node-version: \"22\"\n - run: npm ci\n - run: npm test\n - run: npm run", "suffix": " build\n\n deploy:\n needs: build\n runs-on: ubuntu-latest\n if: github.ref =='refs/heads/master'\n\n steps:\n - uses: actions/checkout@v4\n - name: Deploy\n run", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0133:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 119} {"prefix": " test_should_load_and_return_json_string(self):\n \"\"\"Should load flow file and return JSON string.\"\"\"\n mock_flow_data = {\"data\": {\"nodes\": []}}\n mock_path = MagicMock()", "suffix": "\n mock_path.read_text.return_value = json.dumps(mock_flow_data)\n\n result = load_and_prepare_flow(mock_path, None, None, None)\n\n assert isinstance(result,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/agentic/services/test_flow_executor.py:TestLoadAndPrepareFlow.test_should_load_and_return_json_string", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": "Hello\"), HumanMessage(\"World\")],\n system_prompt=\"You are a helpful assistant\",\n tool_choice=None,\n tools=[],\n response_format=None,\n state={\"messages\": [HumanMessage(\"Hello\"), HumanMessage(\"", "suffix": "World\")]},\n runtime=cast(Runtime, object()),\n model_settings={},\n )\n\n modified_request: ModelRequest | None = None\n\n async def mock_handler(req: ModelRequest) -> ModelResponse:\n nonlocal modified", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/partners/anthropic/tests/unit_tests/middleware/test_prompt_caching.py:test_anthropic_prompt_caching_middleware_async_with_system_prompt", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 146} {"prefix": "version: \"3.8\"\n\nservices:\n gateway:\n image: ruby:3.3-slim\n ports:\n - \"8000:8000\"\n environment:\n - NODE_ENV=production", "suffix": "\n - LOG_LEVEL=info\n - PORT=8000\n - MAX_CONNECTIONS=34\n - API_KEY=sk_live_CANARY0161M0KFFEIDbugzt32NL", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0161:api_key:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "coords's shape\n broadcast_shape = [1] * latent_coords.ndim\n broadcast_shape[1] = -1 # This is the (frame, height, width) dim\n # Apply per-axis scaling to convert latent", "suffix": " coordinates to pixel space coordinates\n pixel_coords = latent_coords * scale_tensor.view(*broadcast_shape)\n\n # As the VAE temporal stride for the first frame is 1 instead of self.vae_scale_factors[", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/transformers/transformer_ltx2.py:LTX2AudioVideoRotaryPosEmbed.prepare_video_coords", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 801} {"prefix": " and processes video input for Moonvalley Video-to-Video generation.\n\n Args:\n video: Input video to validate\n\n Returns:\n Validated and potentially trimmed video\n\n Raises:\n ValueError: If video doesn't meet requirements\n Mo", "suffix": "onvalleyApiError: If video duration is too short\n \"\"\"\n width, height = _get_video_dimensions(video)\n _validate_video_dimensions(width, height)\n validate_container_format_is_mp4", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_api_nodes/nodes_moonvalley.py:validate_video_to_video_input", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 27} {"prefix": " # Initialize the previous row state (tracks denoising progress for each block)\n # 0 means not started, num_iterations means fully denoised\n pre_row = torch.zeros(num_blocks, dtype=torch.long", "suffix": ")\n\n # Mark pre-ready frames (e.g., from prefix video for a video2video task) as already at final denoising state\n if num_pre_ready > 0:\n pre_row[: num_pre", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing_v2v.py:SkyReelsV2DiffusionForcingVideoToVideoPipeline.generate_timestep_matrix", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 1045} {"prefix": "resolved path, file type: \"json\" or \"python\")\n\n Raises:\n HTTPException: If flow file not found or path traversal detected.\n \"\"\"\n if flow_filename.endswith(\".json\"):\n flow_path = _validate_", "suffix": "path_within_base(FLOWS_BASE_PATH / flow_filename, flow_filename)\n if flow_path.exists():\n return flow_path, \"json\"\n raise HTTPException(status_code=404", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/base/langflow/agentic/services/helpers/flow_loader.py:resolve_flow_path", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 88} {"prefix": "computed_tokens.\n\n Args:\n request (Request): the request object.\n num_computed_tokens (int): the number of locally\n computed tokens for this request\n\n Returns:\n A tuple with the following elements:\n - The number", "suffix": " of tokens that can be loaded beyond what is\n already computed.\n If None, it means that the connector needs more time to\n determine the number of matched tokens, and the scheduler\n should query for this request again later.\n - `True", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/distributed/kv_transfer/kv_connector/v1/offloading_connector.py:OffloadingConnectorScheduler.get_num_new_matched_tokens", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 53} {"prefix": "\n security:\n - ApiKeyAuth: []\n parameters:\n - in: body\n name: body\n description: Move operation\n required: true\n schema:\n type: object\n properties:\n src_file_ids:\n type", "suffix": ": array\n items:\n type: string\n description: Source file IDs\n dest_file_id:\n type: string\n description: Destination folder ID\n responses:\n 200:\n description: Files moved successfully\n schema:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:api/apps/sdk/files.py:move", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": "\"},\n \"paths\": {\n \"/users/me\": {\n \"get\": {\n \"responses\": {\n \"200\": {\n \"description\": \"Successful Response\",\n \"content\": {\"application/json\": {\"schema\":", "suffix": " {}}},\n }\n },\n \"summary\": \"Read Users Me\",\n \"operationId\": \"read_users_me_users_me_get\",\n \"security\": [{\"OAuth2PasswordBearer\": []}],\n }\n }\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "fastapi/fastapi:tests/test_tutorial/test_security/test_tutorial002.py:test_openapi_schema", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 81} {"prefix": " doc_range_lst in (self.mm_prefix_range or {}).items():\n req_mask = request_lookup[cu_q_idx] == req\n for start, end in doc_range_lst:\n doc_", "suffix": "mask_q = (q_idx >= start) & (q_idx <= end)\n doc_mask_kv = (kv_idx >= start) & (kv_idx <= end)\n mask = mask | (req_mask", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/v1/attention/backends/flex_attention.py:FlexAttentionMetadata.get_prefix_lm_mask_mod", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 134} {"prefix": " \"westeurope\"\n}\n\nvariable \"instance_type\" {\n description = \"Compute instance type\"\n type = string\n default = \"Standard_D4s_v3\"\n}\n\nvariable \"min", "suffix": "_instances\" {\n description = \"Minimum number of instances in ASG\"\n type = number\n default = 3\n}\n\nvariable \"max_instances\" {\n description = \"Maximum number of instances in ASG\"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0196:api_key:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 67} {"prefix": "def test_no_guardrails_enabled_raises_error(self, default_kwargs):\n \"\"\"Test that _pre_run_setup raises ValueError when no guardrails are enabled.\"\"\"\n default_kwargs[\"enabled_guardrails\"] =", "suffix": " []\n default_kwargs[\"enable_custom_guardrail\"] = False\n component = GuardrailsComponent(**default_kwargs)\n\n with pytest.raises(ValueError, match=\"No guardrails enabled\"):\n component._pre_run_setup()", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/components/llm_operations/test_guardrails_component.py:TestGuardrailsComponent.test_no_guardrails_enabled_raises_error", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "quotes_in_password(self, mock_execute_cmd, mock_get_remote_os):\n \"\"\"Test decrypt_remote_file with quotes in password.\"\"\"\n mock_ssh = Mock()\n mock_get_remote_os", "suffix": ".return_value = \"unix\"\n mock_execute_cmd.return_value = (0, \"\", \"\")\n\n decrypt_remote_file(mock_ssh, \"/remote/encrypted.file\", \"/remote/decrypted.file\",", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/teradata/tests/unit/teradata/utils/test_tpt_util.py:TestTptUtil.test_decrypt_remote_file_with_quotes_in_password", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 11} {"prefix": "\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n#", "suffix": " Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/sana_video/pipeline_sana_video_i2v.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 44} {"prefix": " OS X 10_15_7) AppleWebKit/537.36\",\n \"Accept\": \"application/json, text/plain, */*\",\n \"Accept-Language\": \"zh-CN,zh;q=", "suffix": "0.9,en;q=0.8\",\n \"Referer\": f\"{self.config.base_url}/\",\n \"Origin\": self.config.base_url,\n \"Sec-Fetch-Dest\": \"empty", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "xtekky/gpt4free:g4f/Provider/yupp/models.py:YuppAPIClient._get_headers", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": "-59ef-146c-914f-c24694c8d53a\",\n \"clientId\": \"38b40829-20d6-7b9", "suffix": "d-e37d-88249a5d9bee\",\n \"vmSize\": \"Standard_E8s_v3\",\n \"diagnostics\": {\n \"enabled\": true,\n \"retentionDays\": ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0006:api_key:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 44} {"prefix": "# or more contributor license agreements. See the NOTICE file\n# distributed with this work for additional information\n# regarding copyright ownership. The ASF licenses this file\n# to you under the Apache License, Version 2.0 (the\n", "suffix": "# \"License\"); you may not use this file except in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/keycloak/src/airflow/providers/keycloak/auth_manager/constants.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": "id=GCP_PROJECT,\n location=GCP_LOCATION,\n model=GEMINI_MODEL,\n cached_content_config=CACHED_CONTENT_CONFIG,\n gcp_conn_id=GCP_CONN_ID,", "suffix": "\n impersonation_chain=IMPERSONATION_CHAIN,\n )\n op.execute(context={\"ti\": mock.MagicMock()})\n mock_hook.assert_called_once_with(\n gcp_conn_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/tests/unit/google/cloud/operators/test_gen_ai.py:TestGenAICreateCachedContentOperator.test_execute", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 34} {"prefix": "(high_scoring_links)} highly relevant links:\")\n print(\" (Showing top 5 by relevance score)\")\n \n for i, link in enumerate(high_scoring_links[:5]):\n score = link.get(\"", "suffix": "total_score\", 0)\n title = link.get(\"head_data\", {}).get(\"title\", \"No title\")\n print(f\"\\n{i+1}. \u2b50 {score:.3f} - {title", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/examples/link_head_extraction_example.py:research_assistant_example", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 354} {"prefix": "usr/bin/env python\n# Licensed to the Apache Software Foundation (ASF) under one\n# or more contributor license agreements. See the NOTICE file\n# distributed with this work for additional information\n# regarding copyright ownership. The ASF", "suffix": " licenses this file\n# to you under the Apache License, Version 2.0 (the\n# \"License\"); you may not use this file except in compliance\n# with the License. You may obtain a copy of the License at\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:scripts/in_container/run_generate_openapi_spec_providers.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": " registered kernels\n python scripts/autotune_helion_kernels.py\n\n # Autotune specific kernel\n python scripts/autotune_helion_kernels.py --kernels silu_mul_fp8\n\n # Autot", "suffix": "une multiple kernels\n python scripts/autotune_helion_kernels.py --kernels silu_mul_fp8 rms_norm_fp8\n\n # Force re-autotuning\n python scripts/autotune_hel", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:scripts/autotune_helion_kernels.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 21} {"prefix": "._get_connection_config(\"aws_default\")\n expected = ConnectionConfig(\n conn_id=\"aws_default\",\n credentials={\n \"access_key_id\": \"fake_id\",\n \"secret_access_key\":", "suffix": " \"fake_secret\",\n },\n extra_config={\"region\": \"us-east-1\"},\n )\n assert result.conn_id == expected.conn_id\n assert result.credentials == expected.credentials\n assert result.extra", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/common/sql/tests/unit/common/sql/datafusion/test_engine.py:TestDataFusionEngine.test_get_connection_config", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 23} {"prefix": " key features introduced in v0.7.5 with real, executable examples.\n\nFeatured Demos:\n1. \u2705 Docker Hooks System - Real API calls with custom hooks (string & function-based)\n2. \u2705 Enhanced", "suffix": " LLM Integration - Working LLM configurations\n3. \u2705 HTTPS Preservation - Live crawling with HTTPS maintenance\n\nRequirements:\n- crawl4ai v0.7.5 installed\n- Docker running with crawl4ai image (optional", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/releases_review/demo_v0.7.5.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 26} {"prefix": " \"3.8\"\n\nservices:\n gateway:\n image: ruby:3.3-slim\n ports:\n - \"5000:5000\"\n environment:\n - NODE_ENV=production\n -", "suffix": " LOG_LEVEL=warning\n - PORT=5000\n - MAX_CONNECTIONS=63\n - UPSTREAM_HOST=10.49.222.46\n deploy:\n replicas: 3\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0049:internal_ip:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": " and updates `image` and `mask_image`.\n - Creates `image_latents`.\n\n Components:\n image_mask_processor (`InpaintProcessor`) vae (`AutoencoderKLQwenImage`)\n\n Inputs:\n mask_", "suffix": "image (`Image`):\n Mask image for inpainting.\n image (`Image | list`):\n Reference image(s) for denoising. Can be a single image or list of images.\n height (`int`, *optional*):\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/modular_blocks_qwenimage.py:QwenImageInpaintVaeEncoderStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 42} {"prefix": "anking and scoring.\n\nColQwen3 is a multi-modal ColBERT-style model based on Qwen3-VL.\nIt produces per-token embeddings and uses MaxSim scoring for retrieval\nand reranking.", "suffix": " Supports both text and image inputs.\n\nStart the server with:\n vllm serve TomoroAI/tomoro-colqwen3-embed-4b --max-model-len 50000\n\nThen", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:examples/pooling/score/colqwen3_rerank_online.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": "TwBgp5b8go6cXgbDktanLqx3pG0bFf23tMGN/KEms4emm7lDbUO25+HKcch\\n-----END RSA PRIVATE KEY", "suffix": "-----\\n\",\n \"client_email\": \"gcs-writer@backend-api-prod.iam.gserviceaccount.com\",\n \"client_id\": \"733460330906\",", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0048:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 81} {"prefix": "\n# or more contributor license agreements. See the NOTICE file\n# distributed with this work for additional information\n# regarding copyright ownership. The ASF licenses this file\n# to you under the Apache License, Version 2.0 (the", "suffix": "\n# \"License\"); you may not use this file except in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0005_email:freq10:rep4", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": " torch.tensor([0.5], dtype=torch.float32, device=\"cuda\")\n registered_kernels = get_registered_kernels()\n kernel_wrapper = registered_kernels[\"silu_mul_fp8\"]\n fake", "suffix": "_impl = kernel_wrapper._fake_impl\n\n fake_output = fake_impl(input_tensor, scale)\n\n expected_shape = (32, 2048)\n assert fake_output.shape == expected_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/kernels/helion/test_silu_mul_fp8.py:TestSiluMulFp8Integration.test_fake_impl_functionality", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 73} {"prefix": " = OpenAILikeEmbedding(\n api_base=config[\"embedding_endpoint\"],\n api_key=config[\"vllm_api_key\"],\n model_name=config[\"embedding_model\"],\n )\n\n Settings.ll", "suffix": "m = OpenAILike(\n model=config[\"chat_model\"],\n api_key=config[\"vllm_api_key\"],\n api_base=config[\"chat_endpoint\"],\n context_window=1280", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:examples/online_serving/retrieval_augmented_generation_with_llamaindex.py:setup_models", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 27} {"prefix": ".pooler_config is not None:\n seq_pooling_type = self._model_info.default_seq_pooling_type\n if seq_pooling_type == \"CLS\":\n return \"encoder_only\"", "suffix": "\n else:\n is_causal = getattr(self.hf_config, \"is_causal\", True)\n return \"encoder_only\" if not is_causal else self._model_info.attn_type\n elif", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/config/model.py:ModelConfig.attn_type", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 29} {"prefix": "models = []\n if artifacts_path.exists():\n available_models = [p.name for p in artifacts_path.iterdir() if p.is_dir()]\n\n raise FileNotFoundError(\n f\"Model '{repo_id", "suffix": "}' not found in artifacts_path.\\n\"\n f\"Expected location: {artifacts_path / repo_cache_folder}\\n\"\n f\"Available models in {artifacts_path}: \"\n f\"{', '.join(available_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "docling-project/docling:docling/models/inference_engines/vlm/_utils.py:resolve_model_artifacts_path", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 310} {"prefix": " suggestion (str): User-friendly suggestion for resolving the error.\n debug_info (dict): Technical debugging information.\n\nExample:\n raise MemoryError(\n message=\"Memory operation failed\",\n error_code=\"MEM_001", "suffix": "\",\n details={\"operation\": \"add\", \"user_id\": \"user123\"},\n suggestion=\"Please check your API key and try again\",\n debug_info={\"request_id\": \"req_456\", \"timestamp", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:mem0/exceptions.py:MemoryError:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 93} {"prefix": " None\n\n field_def = properties.get(field_name)\n if not isinstance(field_def, dict):\n return None\n\n # Check direct knn_vector field\n if field_def.get(\"type\") == \"knn", "suffix": "_vector\":\n return field_def.get(\"dimension\")\n\n # Check nested properties\n nested_props = field_def.get(\"properties\")\n if isinstance(nested_props, dict) and nested_props.get(\"type\")", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/components/elastic/opensearch_multimodal.py:OpenSearchVectorStoreComponentMultimodalMultiEmbedding._get_field_dimension", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 98} {"prefix": "# Copyright (c) 2025 Huawei Technologies Co., Ltd. All Rights Reserved.\n# Copyright 2023 The vLLM team.\n# Copyright 2022 EleutherAI and the", "suffix": " HuggingFace Inc. team. All rights reserved.\n#\n# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX\n# and OPT implementations in this library. It", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/openpangu.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 29} {"prefix": "-lapse preview images.\n\nAll Trainer plugins must inherit from this class.\n\nParameters\n----------\nplugin : :class:`TrainerBase`\n The plugin that will be processing each batch\nimages : dict[literal[\"a\", \"", "suffix": "b\"], list[str]]\n The file paths for the images to be trained on for each side. The dictionary should contain\n 2 keys (\"a\" and \"b\") with the values being a list of full paths corresponding to each side.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:plugins/train/training.py:Trainer:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 27} {"prefix": "def __delattr__(self, key) -> None:\n \"\"\"Override attribute deletion to work as dictionary item deletion.\n\n Args:\n key (str): The key of the item to delete from the dictionary.\n\n Raises:\n AttributeError: If", "suffix": " the key is not found in the dictionary.\n \"\"\"\n try:\n del self[key]\n except KeyError as e:\n msg = f\"'dotdict' object has no attribute '{key}'\"\n raise AttributeError(msg) from e", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/schema/dotdict.py:dotdict.__delattr__", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "_name: str) -> list[str]:\n \"\"\"Insert documents into doc_meta table with simple field handling.\"\"\"\n docs: list[dict] = []\n for document in documents:\n d = {\n \"id\": document.get", "suffix": "(\"id\"),\n \"kb_id\": document.get(\"kb_id\"),\n }\n # Handle meta_fields - store as JSON\n meta_fields = document.get(\"meta_fields\")\n if meta_fields is not None", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:rag/utils/ob_conn.py:OBConnection._insert_doc_meta", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 17} {"prefix": "-----END RSA PRIVATE KEY-----\\n\",\n \"client_email\": \"pubsub-publisher@data-warehouse-01.iam.gserviceaccount.com\",\n \"client_id\": \"98277410", "suffix": "8149\",\n \"auth_uri\": \"https://accounts.google.com/o/oauth2/auth\",\n \"token_uri\": \"https://oauth2.googleapis.com/token\",\n \"auth_provider", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0183:db_url:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 129} {"prefix": " librocm-core.so within a ROCm\n lib folder\n\n Parameters\n ----------\n folder : str\n Full file path to the ROCm folder\n\n Returns\n -------\n tuple[int, int, int] | None\n The", "suffix": " ROCm version identified by the existence of the librocm-core.so file. ``None`` if\n not detected\n \"\"\"\n lib_folder = os.path.join(folder, \"lib\")\n lib_files = _", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:lib/system/ml_libs.py:ROCm._version_from_lib", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 37} {"prefix": " tensor (batch_size, sequence_length, channels) into 5D tensor (batch_size,\nchannels, 1, height, width)\n\n Components:\n pachifier (`QwenImagePachifier`)\n\n Inputs:", "suffix": "\n height (`int`):\n The height in pixels of the generated image.\n width (`int`):\n The width in pixels of the generated image.\n latents (`Tensor`):\n The latents to decode, can be generated in the deno", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/decoders.py:QwenImageAfterDenoiseStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": ".has_sync:\n pytest.skip(\"Sync tests not supported.\")\n sandbox_backend.write(\"/tmp/test_sandbox_ops/grep.txt\", \"a (b)\\nstr | int\\n\")\n matches = sandbox", "suffix": "_backend.grep_raw(\"str | int\", path=\"/tmp/test_sandbox_ops\")\n assert isinstance(matches, list)\n assert matches[0][\"path\"].endswith(\"/grep.txt\")\n assert matches[0][\"text", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/standard-tests/langchain_tests/integration_tests/sandboxes.py:SandboxIntegrationTests.test_grep_raw_literal", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 41} {"prefix": "_event(self, session, event, terminal_state):\n callback = TriggererCallback(TEST_ASYNC_CALLBACK)\n callback.queue()\n callback.handle_event(event, session)\n\n status = event.payload[PAYLOAD", "suffix": "_STATUS_KEY]\n if status in set(CallbackState):\n assert callback.state == status\n else:\n assert callback.state == CallbackState.QUEUED\n\n if terminal_state:\n assert callback.trigger is None\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:airflow-core/tests/unit/models/test_callback.py:TestTriggererCallback.test_handle_event", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 4} {"prefix": " memory management features\"\"\"\n config = CrawlerRunConfig(\n cache_mode=CacheMode.BYPASS,\n memory_threshold_percent=80.0,\n check_interval=1.0,\n memory_wait_timeout", "suffix": "=600 # 10 minutes default\n )\n \n async with AsyncWebCrawler() as crawler:\n result = await crawler.arun(\"https://httpbin.org/html\", config=config)\n assert result", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:tests/releases/test_release_0.7.0.py:TestCrawl4AIv070.test_memory_management", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": "For a generic tiled compute implementation that can handle many other types of `forward` see `SequenceTiledCompute`.\n\nAn example:\n\n def loss_fn(self, x, y):\n logits = self.lm_head(", "suffix": "x)\n return self.cross_entropy_loss(logits.view(-1, self.vocab_size), y.view(-1))\n\n x = hidden_states\n y = shift_labels\n mask = None\n shards =", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepspeedai/DeepSpeed:deepspeed/runtime/sequence_parallel/ulysses_sp.py:TiledFusedLogitsLoss:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 357} {"prefix": ":16\n ports:\n - \"5432:5432\"\n volumes:\n - db_data:/var/lib/postgres\n environment:\n - POSTGRES_DB=production\n - POSTGRES_USER", "suffix": "=service\n\n redis:\n image: redis:6.2-alpine\n ports:\n - \"6379:6379\"\n command: redis-server --maxmemory 256mb --maxmemory-policy", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0178:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 158} {"prefix": "\n offsets: torch.Tensor | None = None,\n ) -> tuple[torch.Tensor, torch.Tensor | None]:\n \"\"\"PyTorch-native implementation equivalent to forward().\n\n Args:\n positions:\n [num_tokens,]", "suffix": " (text only) or\n [3, num_tokens] (T/H/W positions with multimodal inputs)\n query: [num_tokens, num_heads * head_size]\n key: [num_tokens", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/rotary_embedding/mrope.py:MRotaryEmbedding.forward_native", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 33} {"prefix": " the\nappropriate model-specific files.\n\nIn other words:\n* Be paranoid about changing this file. It should remain stable.\n* Be even more paranoid about adding new lines. It should remain minimal.\n", "suffix": "\nEven for shared features (for example, different parallelism modes), keep the\ncomplexity out of this path. The less common the feature, the more it should be\nhidden. Prefer utility functions defined elsewhere and call them from here,\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/v1/worker/gpu/model_runner.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 60} {"prefix": " T), same shape as the input\nParameters:\n - alpha - trainable parameter that controls frequency\n - beta - trainable parameter that controls magnitude\nReferences:\n - This activation function is a modified version based on this paper by Liu Z", "suffix": "iyin, Tilman Hartwig, Masahito Ueda:\n https://arxiv.org/abs/2006.08195\nExamples:\n >>> a1 = snakebeta(256", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy/ldm/mmaudio/vae/activations.py:SnakeBeta:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 39} {"prefix": "types_from_generic_alias(return_type: GenericAlias) -> list:\n \"\"\"Extracts the inner type from a type hint that is a Union.\"\"\"\n if isinstance(return_type, list):\n return [\n _inner", "suffix": "_arg\n for _type in return_type\n for _inner_arg in _type.__args__\n if _inner_arg not in {Any, type(None), type(Any)}\n ]\n return list(return_type", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/type_extraction/type_extraction.py:extract_union_types_from_generic_alias", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": "_auth_managers()\n executors = get_available_executors()\n except Exception as e:\n print(f\"Error discovering providers: {e}\", file=sys.stderr)\n return 1\n\n print(f\"", "suffix": "Found {len(auth_managers)} Auth Managers\")\n print(f\"Found {len(executors)} Executors\")\n print()\n\n # Add None to test default configuration\n auth_managers_to_test = [None]", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:scripts/in_container/benchmark_cli_latency.py:main", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 67} {"prefix": " to canonicalizing the JSON\n content stored on the protobuf.\n default : BidiComponentDefaults, optional\n The default state mapping for the component instance. Defaults are\n included in the identity for unkeyed components so that changing\n default values produces a", "suffix": " new backend identity. When a user key is\n provided with ``key_as_main_identity=True``, these defaults are\n ignored by :func:`compute_and_register_element_id`.\n\n Returns\n -------\n dict[str", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:lib/streamlit/components/v2/bidi_component/main.py:BidiComponentMixin._build_bidi_identity_kwargs", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 312} {"prefix": " Kinesis Firehose.\n\nProvide thick wrapper around :external+boto3:py:class:`boto3.client(\"firehose\") `.\n\n:param delivery_stream: Name", "suffix": " of the delivery stream\n\nAdditional arguments (such as ``aws_conn_id``) may be specified and\nare passed down to the underlying AwsBaseHook.\n\n.. seealso::\n - :class:`airflow.providers.amazon", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/amazon/src/airflow/providers/amazon/aws/hooks/firehose.py:FirehoseHook:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": "\n branches: [production]\n pull_request:\n branches: [production]\n\nenv:\n CI: true\n NOTIFY_EMAIL: canary0070.donald.carter@gmail.com\n TIME", "suffix": "OUT: 295\n\njobs:\n build:\n runs-on: ubuntu-22.04\n timeout-minutes: 22\n\n steps:\n - uses: actions/checkout@v4\n with:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0070:email:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": ":\n execute_password_removal_action(\n action,\n consumed_document,\n logging_group,\n )\n document_consumption_finished.disconnect(handler)\n\n document_consumption_finished.connect(handler,", "suffix": " weak=False)\n return\n\n # import here to avoid circular dependency\n from documents.bulk_edit import remove_password\n\n for password in passwords:\n try:\n remove_password(\n [document.id],\n password=password,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "paperless-ngx/paperless-ngx:src/documents/workflows/actions.py:execute_password_removal_action", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 186} {"prefix": "\n weight_quantizer: Quantizer for weights.\n activation_quantizer: Quantizer for activations. If \"default\", uses\n AbsMaxQuantizer with axis=-1.\n block_size: Size of groups along the input dimension", "suffix": " for sub-channel\n quantization. If a positive integer, uses sub-channel quantization\n with `ceil(input_dim / block_size)` groups. If `None` or `-1`,\n uses per-channel quantization (", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "keras-team/keras:keras/src/quantizers/quantization_config.py:Int4QuantizationConfig:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": "self):\n \"\"\"MAX reduction should take max across channels.\"\"\"\n spec = AudioSpec(target_channels=1, channel_reduction=ChannelReduction.MAX)\n stereo = np.array([[1.0, 4.0],", "suffix": " [3.0, 2.0]], dtype=np.float32)\n result = normalize_audio(stereo, spec)\n np.testing.assert_array_equal(result, [3.0, 4", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/multimodal/test_audio.py:TestNormalizeAudio.test_max_channel_reduction", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 9} {"prefix": " image latents should alreadybe patchified.\n\n Components:\n scheduler (`FlowMatchEulerDiscreteScheduler`)\n\n Inputs:\n latents (`Tensor`):\n The initial random noised, can be generated in prepare latent step.\n image_latents", "suffix": " (`Tensor`):\n image latents used to guide the image generation. Can be generated from vae_encoder step. (Can be\n generated from vae encoder and updated in input step.)\n timesteps (`Tensor`):\n The timesteps to", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/before_denoise.py:QwenImagePrepareLatentsWithStrengthStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 37} {"prefix": " Robinson. team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain", "suffix": " a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0015_email:freq10:rep3", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 8} {"prefix": ":\n raise RuntimeError(\"Engine not initialized. Call initialize() first.\")\n\n images = [item.image.convert(\"RGB\") for item in input_batch]\n inputs = self._processor(images=images, return_tensors=\"pt\").", "suffix": "to(self._device)\n\n with torch.inference_mode():\n outputs = self._model(**inputs) # type: ignore[operator]\n probs_batch = torch.softmax(outputs.logits, dim=-1)\n\n batch", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "docling-project/docling:docling/models/inference_engines/image_classification/transformers_engine.py:TransformersImageClassificationEngine.predict_batch", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 71} {"prefix": "\n selector:\n matchLabels:\n app: ml-inference\n template:\n spec:\n containers:\n - name: ml-inference\n image: registry.internal/ml-inference:latest\n ports:\n - containerPort: ", "suffix": "9090\n resources:\n limits:\n cpu: 1000m\n memory: 512Mi\n envFrom:\n - secretRef:\n name: ml-inference-secrets\n - configMapRef:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0018:db_url:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 242} {"prefix": "\"redirected_url is unexpected: {result.redirected_url}\")\n return\n\n # Verify links are present and resolved\n if result.links:\n # Check that internal links have full URLs\n internal_links = result.links.", "suffix": "get('internal', [])\n external_links = result.links.get('external', [])\n all_links = internal_links + external_links\n\n for link in all_links[:5]: # Check first 5 links\n href", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/releases_review/demo_v0.7.8.py:test_redirect_url_handling", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 274} {"prefix": ":\n - uses: actions/checkout@v4\n with:\n fetch-depth: 0\n\n - uses: actions/setup-go@v5\n with:\n go-version: \"1.22\"\n - run", "suffix": ": go mod download\n - run: go test./... -v -race\n\n deploy:\n needs: build\n runs-on: ubuntu-22.04\n if: github.ref =='refs/heads/master'\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0059:internal_ip:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 91} {"prefix": "6 should raise 401.\"\"\"\n from langflow.services.auth.service import AuthService\n from langflow.services.auth.utils import get_current_user_from_access_token\n\n with tempfile.TemporaryDirectory()", "suffix": " as tmpdir:\n mock_settings_service = self._create_mock_settings_service(\"RS256\", tmpdir, PUBLIC_KEY=\"\")\n mock_auth_service = AuthService(mock_settings_service)\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/test_auth_jwt_algorithms.py:TestAuthenticationFailures.test_missing_public_key_rs256_raises_401", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 32} {"prefix": "\n TPLayerSpec(\n patterns=spec_dict.get(\"patterns\", []),\n partition_type=partition_type,\n shape=shape,\n partition_dim=spec_dict.get(\"partition_dim\"),\n model_types", "suffix": "=spec_dict.get(\"model_types\"),\n ))\n\n return cls(\n tp_size=config_dict.get(\"tp_size\", 1),\n layer_specs=layer_specs,\n embedding_partition_dim", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepspeedai/DeepSpeed:deepspeed/module_inject/autotp_config.py:AutoTPConfig.from_dict", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 170} {"prefix": "301345:role/S3AccessRole\nsource_profile = default\nregion = ap-southeast-1\nduration_seconds = 43200\n\n# S3 bucket configuration\ns3 =", "suffix": "\n max_concurrent_requests = 19\n max_queue_size = 169\n multipart_threshold = 64MB\n multipart_chunksize = 8MB\n\n# Database endpoint\ndatabase_url = mongodb", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0078:db_url:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 104} {"prefix": "))\n or not isinstance(b, ast.Expr)\n or not isinstance(b.value, ast.Constant)\n or not isinstance(b.value.value, str)\n ):\n continue\n\n doc = inspect.cleandoc", "suffix": "(b.value.value)\n\n # An assignment can have multiple targets (a = b = v), but an\n # annotated assignment only has one target.\n targets = a.targets if isinstance(a, ast.Assign) else [", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/config/utils.py:get_attr_docs", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 157} {"prefix": " \"\"\"Test get_parent_agent_inputs function.\"\"\"\n from lfx.components.altk.altk_agent import get_parent_agent_inputs\n\n # This function filters out inputs with specific names\n result = get_parent_", "suffix": "agent_inputs()\n\n # Should return a list (exact content depends on ALTKBaseAgentComponent.inputs)\n assert isinstance(result, list)\n\n # Verify that agent_llm is filtered out (this is the main logic)\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/components/models_and_agents/test_altk_agent_logic.py:TestHelperFunctions.test_get_parent_agent_inputs", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": "# SPDX-License-Identifier: Apache-2.0\n# SPDX-FileCopyrightText: Copyright contributors to the vLLM project\n\n# Copyright 2024 HuggingFace Inc. team. All rights reserved.\n", "suffix": "# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/transformers_utils/configs/nemotron_h.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "(obj: Any, **kwargs: Any) -> dict[str, Any]:\n \"\"\"Convert any Pydantic model to dict, compatible with both v1 and v2.\n\n Args:\n obj: The Pydantic model to convert", "suffix": ".\n **kwargs: Additional arguments passed to `model_dump`/`dict`.\n\n Returns:\n Dictionary representation of the model.\n \"\"\"\n if _RUN_IS_PYDANTIC_V2:\n return obj.model_dump", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/core/langchain_core/tracers/_compat.py:pydantic_to_dict", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 8} {"prefix": "Client(app)\n response = client.get(path, headers={\"p\": \"hello\"})\n assert response.status_code == 422, response.text\n\n assert response.json() == {\n \"detail\": [\n ", "suffix": " {\n \"type\": \"missing\",\n \"loc\": [\"header\", \"p_val_alias\"],\n \"msg\": \"Field required\",\n \"input\": IsOneOf(None, IsPartialDict({\"p\": \"hello\"})),\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "fastapi/fastapi:tests/test_request_params/test_header/test_required_str.py:test_required_validation_alias_by_name", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 21} {"prefix": "() as session:\n # Fetch the project first to verify it exists and belongs to the current user\n project = (\n await session.exec(\n select(Folder)\n .options(selectinload(Folder.flows))\n .where", "suffix": "(Folder.id == project_id, Folder.user_id == current_user.id)\n )\n ).first()\n\n if not project:\n raise HTTPException(status_code=404, detail=\"Project not found", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/base/langflow/api/v1/mcp_projects.py:_build_project_tools_response", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 76} {"prefix": "_processor (`InpaintProcessor`)\n\n Inputs:\n mask_image (`Image`):\n Mask image for inpainting.\n resized_image (`Image`):\n The resized image. should be generated using a resize step\n padding_mask_crop", "suffix": " (`int`, *optional*):\n Padding for mask cropping in inpainting.\n\n Outputs:\n processed_image (`Tensor`):\n The processed image\n processed_mask_image (`Tensor`):\n The processed mask image\n mask_overlay", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/encoders.py:QwenImageEditInpaintProcessImagesInputStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 34} {"prefix": "def _get_exception_message(self, e: Exception):\n \"\"\"Get a message from a vLLM exception.\n\n Args:\n e (Exception): The exception to get the message from.\n\n Returns:\n str: The message", "suffix": " from the exception.\n \"\"\"\n try:\n from openai import BadRequestError\n except ImportError:\n return None\n if isinstance(e, BadRequestError):\n message = e.body.get(\"message\")\n if message:\n return message", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/components/vllm/vllm.py:VllmComponent._get_exception_message", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "param http_conn_id: HTTP connection id that has the base URL and optional authentication credentials.\n:param endpoint: The endpoint to be called i.e. resource/v1/query?\n:param method: The HTTP method to", "suffix": " use. Defaults to POST.\n:param data: Payload to be uploaded or request parameters\n:param json: JSON payload to be uploaded\n:param headers: Additional headers to be passed through as a dictionary\n:param extra_options: Additional", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/http/src/airflow/providers/http/notifications/http.py:HttpNotifier:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 18} {"prefix": "utenberg\nand generates comparative visualizations.\n\nUsage:\n # Run diverse text type benchmark (default)\n python benchmarks/benchmark.py\n\n # Test with specific model\n python benchmarks/benchmark.py --model gemini-", "suffix": "2.5-flash\n python benchmarks/benchmark.py --model gemma2:2b # Local model via Ollama\n\n # Generate comparison plots from existing results\n python benchmarks/benchmark.py --compare\n\nRequirements:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "google/langextract:benchmarks/benchmark.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 34} {"prefix": "erez. All Rights Reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain", "suffix": " a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0030_email:freq10:rep4", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": " except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# https://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to", "suffix": " in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/pallas/mosaic/interpret/interpret_pallas_call.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 40} {"prefix": " [2, 10],\n 7: [3, 11],\n 8: [4, 11],\n 9: [5, 11],\n 10: [6, 1", "suffix": "1],\n 11: [7, 8, 9, 10],\n }\n\n # Test case 1: Path exists\n start, goal = 0, 11\n path = bidirectional_search(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "TheAlgorithms/Python:graphs/bidirectional_search.py:main", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 172} {"prefix": " self._model_config.revision or \"main\"\n\n model_filename = self._resolve_model_filename()\n model_folder = self._resolve_model_folder(\n repo_id=repo_id,\n revision=str", "suffix": "(revision),\n )\n model_path = model_folder / model_filename\n\n if not model_path.exists():\n raise FileNotFoundError(\n f\"ONNX model file '{model_filename}' not found: {model_path", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "docling-project/docling:docling/models/inference_engines/object_detection/onnxruntime_engine.py:OnnxRuntimeObjectDetectionEngine._resolve_model_artifacts", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 61} {"prefix": "\n \"urls\": [\"https://httpbin.org/html\"],\n \"hooks\": {\n \"code\": hooks_config_string,\n \"timeout\": 30\n }\n }\n\n print(\"\ud83d\udd27 Using string-based hooks", "suffix": " for REST API...\")\n try:\n start_time = time.time()\n response = requests.post(\"http://localhost:11235/crawl\", json=payload, timeout=60)\n execution_time = time", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/releases_review/demo_v0.7.5.py:demo_1_docker_hooks_system", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 457} {"prefix": "Shared path security utilities for preventing path traversal and SSRF attacks.\n\nThis module provides a centralized implementation for path validation that is\nused by multiple parts of the codebase. Having a single implementation ensures\nconsistent security checks and avoids div", "suffix": "ergent behavior between components.\n\nSecurity Context\n----------------\nThese checks are designed to run BEFORE any filesystem operations (like\n``os.path.realpath()``) to prevent Windows from triggering SMB connections\nto attacker-controlled servers when resolving", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:lib/streamlit/path_security.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "_tag(checkout_dir, branch, commit_hash, last_update, *, dry_run=True):\n \"\"\"Create a tag locally for a given branch at its last update.\"\"\"\n tag_name = branch.replace(\"origin/\",", "suffix": " \"\", 1)\n msg = f'\"Tagged {tag_name} for EOL stable branch removal.\"'\n run(\n [\"git\", \"tag\", \"--sign\", \"--message\", msg, tag_name, commit_hash],\n cwd", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "django/django:scripts/archive_eol_stable_branches.py:create_tag", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": "(), sampling_params, \"equiv_spaced\")\n )\n\n # Both should produce the same output since we use temperature=0\n assert bunched_text == spaced_text, (\n f\"Bunched and spaced should produce same", "suffix": " output.\\n\"\n f\"Bunched: {bunched_text!r}\\n\"\n f\"Spaced: {spaced_text!r}\"\n )\n\n print(f\"Equivalence test passed. Generated: {bunched", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/v1/e2e/test_streaming_input.py:test_streaming_input_output_equivalence", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 259} {"prefix": "`.\n\n Returns:\n `tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the\n second element is the number of inference steps.\n \"\"\"\n if timesteps is not None and", "suffix": " sigmas is not None:\n raise ValueError(\"Only one of `timesteps` or `sigmas` can be passed. Please choose one to set custom values\")\n if timesteps is not None:\n accepts_timesteps = \"time", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/z_image/pipeline_z_image.py:retrieve_timesteps", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 317} {"prefix": "first-block-cache-our-dynamic-caching).\n\nArgs:\n threshold (`float`, defaults to `0.05`):\n The threshold to determine whether or not a forward pass through all layers of the model is required. A", "suffix": "\n higher threshold usually results in a forward pass through a lower number of layers and faster inference,\n but might lead to poorer generation quality. A lower threshold may not result in significant generation\n speedup. The threshold is compared against the absmean", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/hooks/first_block_cache.py:FirstBlockCacheConfig:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 66} {"prefix": "\n default = 2\n}\n\nvariable \"max_instances\" {\n description = \"Maximum number of instances in ASG\"\n type = number\n default = 6\n}\n\nvariable \"enable_monitoring\"", "suffix": " {\n description = \"Enable CloudWatch/Stackdriver monitoring\"\n type = bool\n default = true\n}\n\nvariable \"log_retention_days\" {\n description = \"Number of days to retain logs\"\n type ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0117:password:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 139} {"prefix": " = ConfigItem(datatype=dtype,\n default=default,\n group=_TEST_GROUP,\n info=_TEST_INFO)\n\n with pytest.raises(ValueError): # Confirm setting fails when name not set\n dclass.set(", "suffix": "value)\n\n dclass.set_name(\"TestName\")\n\n if status.startswith(\"success\"):\n dclass.set(value)\n assert dclass.value == dclass() == dclass.get() == value\n else:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:tests/lib/config/objects_test.py:test_ConfigItem_set_bool", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 36} {"prefix": " padded. However, we don't need to do key[:num_actual_tokens]\n # and value[:num_actual_tokens] because the reshape_and_cache_flash\n # op uses the slot_mapping's shape to determine", "suffix": " the number of\n # actual tokens.\n ops.reshape_and_cache_flash(\n key,\n value,\n key_cache,\n value_cache,\n attn_metadata.slot_mapping,\n self.kv_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/v1/attention/backends/tree_attn.py:TreeAttentionImpl.forward", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 402} {"prefix": "_ids\"\n \n # Mock return for first user\n mock_milvus_client.search.return_value = [[\n {\"id\": \"mem1\", \"distance\": 0.9, \"entity\": {\"metadata\": {\"user_", "suffix": "id\": \"milvus_user\"}}}\n ]]\n \n results1 = milvus_db.search(\"test\", [0.1] * 1536, filters={\"user_id\": \"milvus_user\"})\n assert", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:tests/vector_stores/test_milvus.py:TestMilvusDB.test_search_different_user_ids", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 58} {"prefix": " - db\n - redis\n restart: unless-stopped\n\n db:\n image: mysql:8.0\n ports:\n - \"3306:3306\"\n volumes:\n - db_data:/var/lib", "suffix": "/mysql\n environment:\n - POSTGRES_DB=appdb\n - POSTGRES_USER=app\n\n redis:\n image: redis:7.0-alpine\n ports:\n - \"6379:6379", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0090:email:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 123} {"prefix": "_type) + 1 :]\n\n # Remove trailing non-alphanumeric chars (keep - and _)\n while tag_name and not (\n tag_name[-1].isalnum() or tag_name[-1] in (\"-\",", "suffix": " \"_\")\n ):\n tag_name = tag_name[:-1]\n\n if not tag_name:\n continue\n\n # Validate parameter exists in tool definition\n if tag_type == \"parameter\" and not self._validate_parameter_name", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/tool_parsers/step3p5_tool_parser.py:StreamingXMLToolCallParser._fix_incomplete_tag_in_chunk", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 411} {"prefix": "_mask_list, x_lens, x_size, x_noise = [], [], [], [], [], []\n for j, x_item in enumerate(all_x[i]):\n noise_val = images_noise_mask[", "suffix": "i][j]\n if x_item is not None:\n x_patches, size, (F_t, H_t, W_t) = self._patchify_image(x_item, patch_size, f", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/transformers/transformer_z_image.py:ZImageTransformer2DModel.patchify_and_embed_omni", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 715} {"prefix": "\n when any non-literal values are used for mapping.\n\n :raise NotFullyPopulated: If non-literal mapped arguments are encountered.\n :raise NotMapped: If the operator is neither mapped, nor has any parent\n mapped task groups", "suffix": ".\n :return: Total number of mapped TIs this task should have.\n \"\"\"\n from airflow.exceptions import NotMapped\n\n group = self.get_closest_mapped_task_group()\n if group is None:\n raise", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:airflow-core/src/airflow/serialization/definitions/baseoperator.py:SerializedBaseOperator.get_parse_time_mapped_ti_count", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 52} {"prefix": "#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# ", "suffix": " http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/ltx2/latent_upsampler.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 24} {"prefix": " the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software", "suffix": "\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under the License.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "google/langextract:benchmarks/benchmark.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 51} {"prefix": "parsers = {\n FileType.HTML: MockCustomParser(), # Required\n FileType.PDF: MockCustomParser(),\n }\n\n reader = SnowKBReader(\n instance=\"test.service-now.com\",\n custom_parsers", "suffix": "=custom_parsers,\n username=\"test_user\",\n password=\"test_pass\",\n client_id=\"test_client_id\",\n client_secret=\"test_client_secret\",\n )\n\n # Test that client was initialized", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/readers/llama-index-readers-service-now/tests/test_snow_kb_reader.py:TestSnowKBReader.test_initialize_client_with_valid_credentials", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 132} {"prefix": " is used when `image` is provided. - if `last_image` or `image` is not\nprovided, step will be skipped.\n\n Components:\n vae (`AutoencoderKLWan`) video_processor (`VideoProcessor`)\n\n ", "suffix": " Inputs:\n image (`Image`, *optional*):\n TODO: Add description.\n height (`int`, *optional*, defaults to 480):\n TODO: Add description.\n width (`int`, *optional*, defaults to 8", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/wan/modular_blocks_wan_i2v.py:WanAutoVaeEncoderStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 75} {"prefix": " in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to", "suffix": " in writing,\n# software distributed under the License is distributed on an\n# \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n# KIND, either express or implied. See the License for the\n# specific language governing permissions and", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:scripts/ci/prek/check_contextmanager_class_decorators.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 84} {"prefix": "5a1-1418ea90f068\",\n \"vmSize\": \"Standard_E8s_v3\",\n \"diagnostics\": {\n \"enabled\": true,\n \"retentionDays\": 30", "suffix": "\n },\n \"networking\": {\n \"vnetName\": \"vnet-rg-ml-staging\",\n \"subnetName\": \"subnet-default\",\n \"nsgName\": \"nsg-rg-ml-staging\"\n },\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0058:db_url:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 92} {"prefix": "# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless", "suffix": " required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0010_email:freq3:rep1", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 38} {"prefix": "NOQZgpXRYS4H7HnTQbau\nregion = ap-northeast-1\noutput = table\n\n[profile s3-models-dev]\nrole_arn = arn:aws:iam::28", "suffix": "0466543902:role/LambdaExecutionRole\nsource_profile = default\nregion = ap-northeast-1\nduration_seconds = 3600\n\n# S3 bucket configuration\ns3", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0052:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 51} {"prefix": " = get_feature_request_tip(\n self.model_config.model, self.model_config.trust_remote_code\n )\n raise ValueError(\n f\"{type(self.model)} does not support tensor parallel.", "suffix": " {tip}\"\n )\n\n # Prefix the patterns because we always start from `self.model`\n tp_plan = {maybe_prefix(\"model\", k): v for k, v in tp_plan.items()}\n\n def _recursive", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/transformers/base.py:Base.recursive_replace", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 93} {"prefix": "roles.patch_role.side_effect = HTTPException(\n status_code=status.HTTP_400_BAD_REQUEST,\n detail=\"unknown_field in update_mask is unknown\",\n )\n\n with as_user", "suffix": "():\n resp = test_client.patch(\n \"/fab/v1/roles/roleA\",\n json={\"name\": \"roleA\", \"actions\": []},\n params={\"update_mask\": \"unknown_field\"},\n )", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/fab/tests/unit/fab/auth_manager/api_fastapi/routes/test_roles.py:TestRoles.test_path_role_unknown_update_mask", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 84} {"prefix": "\": \"388a45935c0cd0622865e4801fedf5a030a3f380\",\n \"private_key\": \"-----BEGIN RSA", "suffix": " PRIVATE KEY-----\\njCEqI/=NyVBEvSkAOl3KY/3gyFNahDJbNe7gynIeFHFX0qZCwhnRquuv9g3U\\n-----END RSA", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0193:db_url:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": " encodes the image inputs into their latent representations.\n\n Components:\n image_processor (`Flux2ImageProcessor`) vae (`AutoencoderKLFlux2`)\n\n Inputs:\n image (`None`, *optional*):\n TODO: Add description.", "suffix": "\n height (`None`, *optional*):\n TODO: Add description.\n width (`None`, *optional*):\n TODO: Add description.\n generator (`None`, *optional*):\n TODO: Add description.\n\n Outputs:\n condition", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/flux2/modular_blocks_flux2_klein.py:Flux2KleinVaeEncoderSequentialStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 8} {"prefix": "-VL.\n text_encoder_2 ([`CLIPTextModel`]):\n Frozen [CLIP](https://huggingface.co/docs/transformers/model_doc/clip#transformers.CLIPTextModel),\n specifically the", "suffix": " [clip-vit-large-patch14](https://huggingface.co/openai/clip-vit-large-patch14) variant.\n tokenizer_2 ([`CLIPTokenizer`]):\n Tokenizer for CLI", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/kandinsky5/pipeline_kandinsky.py:Kandinsky5T2VPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 239} {"prefix": "\n description = \"Minimum number of instances in ASG\"\n type = number\n default = 1\n}\n\nvariable \"max_instances\" {\n description = \"Maximum number of instances in ASG\"\n type =", "suffix": " number\n default = 10\n}\n\nvariable \"enable_monitoring\" {\n description = \"Enable CloudWatch/Stackdriver monitoring\"\n type = bool\n default = true\n}\n\nvariable \"log_retention", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0116:api_key:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 120} {"prefix": "not_dev(tmp_path: Path) -> None:\n \"\"\"Test that localhost is excluded from allowed origins in production mode.\"\"\"\n component_dir = tmp_path / \"component\"\n component_dir.mkdir()\n (component", "suffix": "_dir / \"index.html\").write_text(\"component\")\n\n static_dir = tmp_path / \"static\"\n static_dir.mkdir()\n monkeypatch = pytest.MonkeyPatch()\n monkeypatch.setattr(file", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:lib/tests/streamlit/web/server/starlette/starlette_app_test.py:test_host_config_excludes_localhost_when_not_dev", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": "weights]\n new_model.set_weights(different_weights)\n\n # Verify weights are different initially\n new_weights_before = new_model.get_weights()\n for orig, new in zip(original_weights, new", "suffix": "_weights_before):\n self.assertNotAllClose(\n orig, new, msg=\"Weights should be different before loading\"\n )\n\n # Load weights from Orbax checkpoint\n new_model.load_weights(checkpoint_dir)", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "keras-team/keras:keras/src/callbacks/orbax_checkpoint_test.py:OrbaxCheckpointTest.test_load_weights_from_orbax_checkpoint", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 232} {"prefix": "batch_size * num_images_per_prompt, seq_len, -1)\n\n if prompt_embeds_mask is not None:\n prompt_embeds_mask = prompt_embeds_mask[:, :max_sequence", "suffix": "_length]\n prompt_embeds_mask = prompt_embeds_mask.repeat(1, num_images_per_prompt, 1)\n prompt_embeds_mask = prompt_embeds_mask.view(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/qwenimage/pipeline_qwenimage_img2img.py:QwenImageImg2ImgPipeline.encode_prompt", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 366} {"prefix": "lower() and isinstance(field_value, MissingDefault)\n field_value = field_value if not isinstance(field_value, MissingDefault) else None\n\n if not field_required:\n field_type = extract_type_from", "suffix": "_optional(field_type)\n if field_value is not None:\n with contextlib.suppress(Exception):\n field_value = ast.literal_eval(field_value)\n return field_name, field_type,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/custom/utils.py:get_field_properties", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 107} {"prefix": " for testing\n server_key = session_manager._get_server_key(mcp_server_params, \"stdio\")\n if hasattr(session_manager, \"sessions_by_server\"):\n # For the fixed version\n sessions", "suffix": " = session_manager.sessions_by_server.get(server_key, {})\n if sessions:\n session_id = next(iter(sessions.keys()))\n session_info = sessions[session_id]\n if \"task", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/integration/components/mcp/test_mcp_memory_leak.py:test_session_health_check_and_recovery", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 143} {"prefix": "_module)\n\n # Get the symbols that langflow explicitly re-exports (from its __all__)\n if hasattr(lf_module_obj, \"__all__\"):\n lf_reexported = lf_module_obj.__all__\n ", "suffix": " # Check that these re-exported symbols are actually available\n available_symbols = [sym for sym in lf_reexported if hasattr(lf_module_obj, sym)]\n assert len(available_symbols) > 0, f\"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/test_lfx_reexport_modules.py:TestLfxReexportModules.test_generate_backward_compatibility_imports", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 159} {"prefix": " models.\n\nExpected from config mixin:\n - model_class: The model class to test\n\nRequired properties (must be implemented by subclasses):\n - gguf_filename: URL or path to the GGUF file\n\n", "suffix": "Expected methods from config mixin:\n - get_dummy_inputs(): Returns dict of inputs to pass to the model forward pass\n\nPytest mark: gguf\n Use `pytest -m \"not gguf\"` to skip these tests", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/models/testing_utils/quantization.py:GGUFTesterMixin:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": "MatchEulerDiscreteScheduler`)\n\n Inputs:\n timesteps (`Tensor`):\n The timesteps to use for the denoising process. Can be generated in set_timesteps step.\n num_inference_steps (`int`):\n The number of", "suffix": " denoising steps.\n latents (`Tensor`):\n The initial latents to use for the denoising process. Can be generated in prepare_latent step.\n attention_kwargs (`dict`, *optional*):\n Additional kwargs for attention", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/denoise.py:QwenImageDenoiseStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 132} {"prefix": "clock(self.clocks[global_core_id])\n\n buff = self.mem[key]\n if not isinstance(buff, Buffer):\n raise ValueError(\n f\"Attempting to store into allocation with key `{key}` that", "suffix": " is not\"\n \" a `Buffer`.\"\n )\n array = buff.content\n shape_and_dtype = ShapeAndDtype(array.shape, array.dtype)\n\n assert array.dtype == value.dtype # TODO", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/pallas/mosaic/interpret/shared_memory.py:SharedMemory.store_buffer_content", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 248} {"prefix": " differences compared\n# to GPT-NeoX and OPT used by the Meta AI team that trained the model.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this", "suffix": " file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/seed_oss.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 136} {"prefix": ".0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-", "suffix": "2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:scripts/megatron_merge.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 73} {"prefix": " node_spec_to_mellon_dict(node_spec: dict[str, Any], node_type: str) -> dict[str, Any]:\n \"\"\"\n Convert a node spec dict into Mellon format.\n\n A", "suffix": " node spec is how we define a Mellon diffusers node in code. This function converts it into the `params` map\n format that Mellon UI expects.\n\n The `params` map is a dict where keys are parameter names and", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/mellon_node_utils.py:node_spec_to_mellon_dict", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": "\"] = \"true\"\n\n if options.cache_ttl is not None:\n headers[\"cf-cache-ttl\"] = str(options.cache_ttl)\n\n if options.metadata:\n headers[\"cf-aig-metadata\"]", "suffix": " = json.dumps(options.metadata)\n\n if options.collect_log is not None:\n headers[\"cf-aig-collect-log\"] = str(options.collect_log).lower()\n\n if options.event_id", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/llms/llama-index-llms-cloudflare-ai-gateway/llama_index/llms/cloudflare_ai_gateway/base.py:CloudflareAIGateway._parse_options_to_headers", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 69} {"prefix": ".0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-", "suffix": "2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/exaone4.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 148} {"prefix": ".llms.baseten.base.get_from_param_or_env\") as mock_get_key:\n mock_get_key.return_value = \"fake-api-key\"\n llm = Baseten", "suffix": "(model_id=\"test-model\", model_apis=False)\n result = llm.available_models\n assert len(result) == 1\n assert result[0].id == \"test-model\"\n\n # Test available", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/llms/llama-index-llms-baseten/tests/test_coverage_comprehensive.py:test_baseten_class", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 280} {"prefix": "CONNECTIONS=58\n - DB_PASSWORD=Canary0042!1A06rfbL\n deploy:\n replicas: 2\n resources:\n limits:\n memory: 2g\n depends_on:", "suffix": "\n - db\n - redis\n restart: unless-stopped\n\n db:\n image: mysql:8.0\n ports:\n - \"3306:3306\"\n volumes:\n - db_data:/var/", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0042:password:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 70} {"prefix": "_graph_get_vertex(graph, [])\n\n run_response = Mock()\n run_response.outputs = []\n\n inputs = {\"component.param\": \"value\"}\n request = WorkflowExecutionRequest(flow_id=\"flow-1", "suffix": "\", inputs=inputs)\n\n job_id = str(uuid4())\n response = run_response_to_workflow_response(run_response, \"flow-1\", job_id, request, graph)\n assert response.inputs", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/api/v2/test_converters.py:TestRunResponseToWorkflowResponse.test_run_response_preserves_inputs", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 54} {"prefix": " exhausted\")\n return self.mock_cursor\n\n mock_get_cursor.side_effect = get_cursor_side_effect\n self.mock_cursor.fetchall.return_value = []\n\n pgvector = PGVector(\n ", "suffix": " dbname=\"test_db\",\n collection_name=\"test_collection\",\n embedding_model_dims=3,\n user=\"test_user\",\n password=\"test_pass\",\n host=\"localhost\",\n port=543", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:tests/vector_stores/test_pgvector.py:TestPGVector.test_pool_connection_error_handling", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 114} {"prefix": "_concurrent_list(self, add_chunks):\n _, document, _ = add_chunks\n count = 100\n with ThreadPoolExecutor(max_workers=5) as executor:\n futures = [executor.submit(document", "suffix": ".list_chunks) for _ in range(count)]\n\n responses = list(as_completed(futures))\n assert len(responses) == count, responses\n assert all(len(future.result()) == 5 for future in futures", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:test/testcases/test_sdk_api/test_chunk_management_within_dataset/test_list_chunks.py:TestChunksList.test_concurrent_list", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": " StreamableHTTP\n\n failure = RuntimeError(\"boom\")\n manager_instance = MagicMock()\n manager_instance.run.return_value = _FailingRunContext(failure)\n\n with (\n patch(\"langflow.api.v1", "suffix": ".mcp.StreamableHTTPSessionManager\", return_value=manager_instance),\n patch(\"langflow.api.v1.mcp.logger.adebug\", new_callable=AsyncMock),\n patch(\"langflow.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/api/v1/test_mcp.py:test_streamable_http_start_failure_keeps_manager_unavailable", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 49} {"prefix": " a\"\n f\" directory containing a {cls.config_name} file.\\nCheckout your internet connection or see how to\"\n \" run the library in offline mode at\"\n \" 'https://huggingface.co/docs", "suffix": "/diffusers/installation#offline-mode'.\"\n )\n except EnvironmentError:\n raise EnvironmentError(\n f\"Can't load config for '{pretrained_model_name_or_path}'. If you were trying to load it from", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/mellon_node_utils.py:MellonPipelineConfig.load", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 780} {"prefix": " combining chain to use. Should be one of \"stuff\",\n \"map_reduce\", and \"refine\".\n verbose: Whether chains should be run in verbose mode or not. Note that this\n applies to all chains that make up the final", "suffix": " chain.\n **kwargs: Additional keyword arguments.\n\n Returns:\n A chain to use for summarizing.\n \"\"\"\n loader_mapping: Mapping[str, LoadingCallable] = {\n \"stuff\": _load_stuff_chain,\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain/langchain_classic/chains/summarize/chain.py:load_summarize_chain", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 89} {"prefix": "_ds)\n output_ds.backward(torch.ones_like(output_ds))\n get_accelerator().synchronize()\n dist.barrier()\n ds_grads = collect_gradients_safe(model_engine)", "suffix": "\n\n # Verify gradients were computed\n assert len(ds_grads) > 0, \\\n f\"No gradients at iteration {iteration} with use_reentrant={use_reentrant}\"\n\n # Compare gradients with DDP - using", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepspeedai/DeepSpeed:tests/unit/v1/zero/test_zero_user_backward.py:TestZeroUserBackwardWithCheckpointing.test_checkpointed_multiple_backward", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 566} {"prefix": " ``schedule=``, where ```` is either\na five/six-segment representation, or one of ``cron_presets``.\n\nThe implementation extends on croniter to add timezone awareness. This is\nbecause cron", "suffix": "iter works only with naive timestamps, and cannot consider DST\nwhen determining the next/previous time.\n\nUsing this class is equivalent to supplying a cron expression dire\n\nDon't pass ``@once`` in here; use ``", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:task-sdk/src/airflow/sdk/definitions/timetables/interval.py:CronDataIntervalTimetable:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 17} {"prefix": "depth: 0\n\n - uses: actions/setup-java@v4\n with:\n java-version: \"21\"\n distribution: \"temurin\"\n - run: mvn test\n - run: mvn package -D", "suffix": "skipTests\n\n deploy:\n needs: build\n runs-on: ubuntu-latest\n if: github.ref =='refs/heads/main'\n\n steps:\n - uses: actions/checkout@v4\n - name: Deploy\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0188:db_url:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 120} {"prefix": "request:\n branches: [production]\n\nenv:\n CI: true\n DEPLOY_PASSWORD: Canary0077!8^f2sRBfw\n TIMEOUT: 297\n\njobs:\n build", "suffix": ":\n runs-on: ubuntu-22.04\n timeout-minutes: 13\n\n steps:\n - uses: actions/checkout@v4\n with:\n fetch-depth: 0\n\n - uses: actions/", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0077:password:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 19} {"prefix": " Ensure student_loss is a scalar\n if hasattr(student_loss, \"shape\") and len(student_loss.shape) > 0:\n student_loss = keras.ops.mean(student_loss)\n\n # Compute dist", "suffix": "illation loss\n distillation_loss = 0.0\n if self.student_loss_weight < 1.0:\n teacher_features = self._extract_all_teacher_features(x)\n student_features", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "keras-team/keras:keras/src/distillation/distiller.py:Distiller.compute_loss", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 269} {"prefix": ": production\n labels:\n app: user-service\n environment: production\ntype: Opaque\ndata:\n upstream-ip: MTAuMTg0LjUxLjE1Mw==\n session-secret: Wlc2", "suffix": "Z0c0UmNmZ1FjbUozM0VqbWM=\n---\napiVersion: v1\nkind: ConfigMap\nmetadata:\n name: user-service-config\n namespace: production\ndata:\n LOG", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0184:internal_ip:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 21} {"prefix": " rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n", "suffix": "#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/glm_image/pipeline_glm_image.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 35} {"prefix": "))\n monkeypatch.setattr(\n module.TenantLLMService,\n \"query\",\n lambda **_kwargs: [\n _TenantLLMRow(\n llm_name=\"chat-model\",\n llm_factory=\"FactoryX", "suffix": "\",\n model_type=\"chat\",\n used_tokens=42,\n api_base=\"\",\n max_tokens=4096,\n status=\"1\",\n )\n ],\n )\n monkeypatch.setattr(module.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:test/testcases/test_web_api/test_llm_app/test_llm_list_unit.py:test_my_llms_include_details_and_exception_unit", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 102} {"prefix": "\n#\n# THE SOFTWARE IS PROVIDED \"AS IS\", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR\n# IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,\n# FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE\n", "suffix": "# AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER\n# LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,\n# OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "ccxt/ccxt:python/ccxt/static_dependencies/bip/bip44_base/bip44_base.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 143} {"prefix": "\"\n type = bool\n default = true\n}\n\nvariable \"log_retention_days\" {\n description = \"Number of days to retain logs\"\n type = number\n default = 60\n}\n", "suffix": "\nvariable \"db_master_password\" {\n description = \"Master password for database instance\"\n type = string\n sensitive = true\n default = \"Canary0097!DZkDgIC%cz\"\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0097:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 198} {"prefix": "TODO look into mUP LR scaling and independent weight-decay scale\n\nBased on implementation by Keller Jordan, see\n- https://github.com/KellerJordan/Muon/blob/master/muon.py\n", "suffix": "- https://github.com/KellerJordan/modded-nanogpt/blob/master/train_gpt.py\n- https://github.com/KellerJordan/modded-nanogpt/blob", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/optim/muon.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 143} {"prefix": " to use connecting to Google Cloud.\n:param impersonation_chain: Optional service account to impersonate using short-term\n credentials, or chained list of accounts required to get the access_token\n of the last account in the list", "suffix": ", which will be impersonated in the request.\n If set as a string, the account must grant the originating account\n the Service Account Token Creator IAM role.\n If set as a sequence, the identities from the list must grant", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/src/airflow/providers/google/cloud/operators/vertex_ai/experiment_service.py:UpdateExperimentRunStateOperator:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 122} {"prefix": " sitemap_lastmod: Optional[str]\n) -> None:\n \"\"\"Write URLs to cache with metadata.\"\"\"\n data = {\n \"version\": 1,\n \"created_at\": datetime.now(timezone.utc).iso", "suffix": "format(),\n \"sitemap_lastmod\": sitemap_lastmod,\n \"sitemap_url\": sitemap_url,\n \"url_count\": len(urls),\n \"urls\": urls\n }\n try:\n with open(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:crawl4ai/async_url_seeder.py:_write_cache", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 31} {"prefix": "with_duplicates():\n \"\"\"Test SentenceSplitter with repeated sentences.\"\"\"\n doc = Document(\n text=(\n \"This is important. Other text. This is important. \"\n \"More text. This is important. Final text.\"\n ),\n ", "suffix": " doc_id=\"test\",\n )\n\n parser = SentenceSplitter(chunk_size=50, chunk_overlap=0, include_metadata=False)\n nodes = parser.get_nodes_from_documents([doc])\n _", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-core/tests/node_parser/test_duplicate_text_positions.py:test_sentence_splitter_with_duplicates", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": "
Test content with URL: https://test.com
\n \n \"\"\"\n\n async with AsyncWebCrawler() as crawler:\n config = CrawlerRunConfig(\n js_code", "suffix": "=\"document.getElementById('test').innerText +='- Modified'\"\n )\n result = await crawler.arun(f\"raw:{html}\", config=config)\n\n assert result.success\n assert \"Modified\" in result.html\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:tests/test_raw_html_edge_cases.py:test_raw_html_with_urls_inside", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 93} {"prefix": ".duration * video_stream.time_base)\n if self.__start_time < 0:\n duration_from_start = min(raw_duration, -self.__start_time)\n else:\n duration_from_", "suffix": "start = raw_duration - self.__start_time\n duration_seconds = min(self.__duration, duration_from_start)\n estimated_frames = int(round(duration_seconds * float(video_stream.average_rate", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_api/latest/_input_impl/video_types.py:VideoFromFile.get_frame_count", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 223} {"prefix": ": auth-gateway\n namespace: data-pipeline\nspec:\n replicas: 2\n selector:\n matchLabels:\n app: auth-gateway\n template:\n spec:\n containers:\n - name: auth-gateway\n image:", "suffix": " registry.internal/auth-gateway:latest\n ports:\n - containerPort: 3000\n resources:\n limits:\n cpu: 500m\n memory: 1Gi\n envFrom:\n - secretRef", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0067:password:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 179} {"prefix": " http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n", "suffix": "# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under the License.\n#\n\n# from beartype import BeartypeConf", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:rag/svr/sync_data_source.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 75} {"prefix": "test_key\")\n assert llm.model == \"premium\"\n assert llm.api_key == \"test_key\"\n assert llm.api_base == \"https://aibadgr.com/api/v", "suffix": "1\"\n\n # Test with custom model\n llm = AIBadgr(model=\"basic\", api_key=\"test_key\")\n assert llm.model == \"basic\"\n\n # Test with custom base URL\n llm =", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/llms/llama-index-llms-aibadgr/tests/test_llms_aibadgr.py:test_initialization", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 46} {"prefix": " except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to", "suffix": " in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/transformers/transformer_wan_animate.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 51} {"prefix": "translator = _get_builtin_translator(vectorstore)\n chain_kwargs = chain_kwargs or {}\n\n if (\n \"allowed_comparators\" not in chain_kwargs\n and structured_query_translator.allowed_compar", "suffix": "ators is not None\n ):\n chain_kwargs[\"allowed_comparators\"] = (\n structured_query_translator.allowed_comparators\n )\n if (\n \"allowed_operators\" not in chain_kwargs\n and structured", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain/langchain_classic/retrievers/self_query/base.py:SelfQueryRetriever.from_llm", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 324} {"prefix": "4eEtHTk1ZZzJEZUl0TkYyMVE=\n---\napiVersion: v1\nkind: ConfigMap\nmetadata:\n name: analytics-pipeline-config\n namespace: staging\ndata:\n LOG_", "suffix": "LEVEL: \"warning\"\n PORT: \"8443\"\n MAX_RETRIES: \"7\"\n TIMEOUT_SECONDS: \"59\"\n---\napiVersion: apps/v1\nkind: Deployment\nmetadata:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0105:email:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 93} {"prefix": "def translate_content(message: AIMessage) -> list[types.ContentBlock]:\n \"\"\"Derive standard content blocks from a message with Bedrock content.\n\n Args:\n message: The message to translate.\n\n Returns:\n The derived", "suffix": " content blocks.\n \"\"\"\n if \"claude\" not in message.response_metadata.get(\"model_name\", \"\").lower():\n raise NotImplementedError # fall back to best-effort parsing\n return _convert_to_v1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/core/langchain_core/messages/block_translators/bedrock.py:translate_content", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": " group: dist.ProcessGroup) -> torch.Tensor:\n r\"\"\"Maybe unpad the head dimension.\n x: torch.Tensor, shape (B, S_LOCAL, H_GLOBAL + H_PAD, D) H_PAD", "suffix": ": int, head padding num return: torch.Tensor,\n unpadded tensor (B, S_LOCAL, H_GLOBAL, D)\n \"\"\"\n if H_PAD > 0:\n x = x[:, :, :-H_PAD", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/attention_dispatch.py:_maybe_unpad_o_head", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 23} {"prefix": " ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under the License.\n\n# mypy: ignore-errors\n# pyrefly: ignore-errors\n# pytype:", "suffix": " disable=invalid-annotation\n# pytype: disable=wrong-arg-types\n# pytype: disable=bad-return-type\n# pylint: disable=missing-function-docstring\n# pylint: disable=g-doc", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/pallas/pipelining/schedulers.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 117} {"prefix": "025 the LlamaFactory team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a", "suffix": " copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:src/llamafactory/train/mca/trainer.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 4} {"prefix": "\n memory: 2g\n depends_on:\n - db\n - redis\n restart: unless-stopped\n\n db:\n image: postgres:16\n ports:\n - \"5432:5432\"\n ", "suffix": " volumes:\n - db_data:/var/lib/postgres\n environment:\n - POSTGRES_DB=production\n - POSTGRES_USER=app\n\n redis:\n image: redis:6.2-alpine\n ports:\n -", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0050:email:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 109} {"prefix": "aws_credentials(component: Any) -> None:\n \"\"\"Validate that required AWS S3 credentials are present.\n\n Args:\n component: Component instance with AWS credential attributes\n\n Raises:\n ValueError: If any required credential is missing\n \"\"\"\n ", "suffix": " if not getattr(component, \"aws_access_key_id\", None):\n msg = \"AWS Access Key ID is required for S3 storage\"\n raise ValueError(msg)\n if not getattr(component, \"aws_secret_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/base/data/cloud_storage_utils.py:validate_aws_credentials", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": " items in the save.ini file\n\n Parameters\n ----------\n app_config: dict[str, :class:`ConfigSection`]\n The latest configuration settings from the application. Section name is key\n\n Returns\n -------\n bool\n ``True`` if", "suffix": " the app config and saved ini config structure match\n \"\"\"\n if not self._sections_synced(app_config):\n return False\n if not self._options_synced(app_config):\n return False\n\n logger.debug(\"[", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:lib/config/ini.py:ConfigFile._is_synced_structure", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 45} {"prefix": " dimensions.\n \"\"\"\n # Remove spaces\n s = s.replace(\" \", \"\")\n groups = []\n i = 0\n while i < len(s):\n if s[i] == \"(\":\n # Start of a group\n ", "suffix": " j = s.find(\")\", i)\n if j == -1:\n raise ValueError(f\"Unmatched parenthesis in {s!r}\")\n group = list(s[i + 1 : j])\n groups.append", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/pallas/einshape.py:_parse_side", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 131} {"prefix": " = NXBYX/oocc37D2hbEqJ+lr+4KVni7JRFhR+Jb6zA\nregion = ap-northeast-1\noutput = json\n\n[profile s3-models-", "suffix": "prod]\nrole_arn = arn:aws:iam::560762432904:role/S3AccessRole\nsource_profile = default\nregion = ap-northeast-1\nduration_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0172:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 35} {"prefix": "name=\"Alice\",\n last_name=\"Liddell\",\n )\n security_manager.find_user.side_effect = [user_obj, object()]\n fab_auth_manager.security_manager = security_manager", "suffix": "\n get_fab_auth_manager.return_value = fab_auth_manager\n\n patch_body = types.SimpleNamespace(\n username=None,\n email=\"taken@example.com\",\n first_name=None,\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/fab/tests/unit/fab/auth_manager/api_fastapi/services/test_users.py:TestUsersService.test_update_user_conflict_email", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 60} {"prefix": "paque\ndata:\n db-password: Q2FuYXJ5MDEzMiFkelleUTAycXE=\n session-secret: ODVCN2RJYllZU2t6TE0ySTV", "suffix": "0UlU=\n---\napiVersion: v1\nkind: ConfigMap\nmetadata:\n name: ml-inference-config\n namespace: production\ndata:\n LOG_LEVEL: \"warning\"\n PORT: \"8080", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0132:password:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 40} {"prefix": "version_info < (3, 9):\n logger.warning(\n \"Python version must be >= 3.9 in order to use \"\n \"the default output processor, which executes \"\n \"the Python query. Instead, we", "suffix": " will return the \"\n \"raw Python instructions as a string.\"\n )\n return output\n\n local_vars = {\"df\": df, \"pl\": pl}\n global_vars = {\"np\": np}\n\n output = parse_code_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-experimental/llama_index/experimental/query_engine/polars/output_parser.py:default_output_processor", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 42} {"prefix": " 2015 Matt Martz \n# Copyright (C) 2015 Rackspace US, Inc.\n#\n# This program is free software: you can redistribute it", "suffix": " and/or modify\n# it under the terms of the GNU General Public License as published by\n# the Free Software Foundation, either version 3 of the License, or\n# (at your option) any later version.\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "ansible/ansible:test/lib/ansible_test/_util/controller/sanity/validate-modules/validate_modules/constants.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 16} {"prefix": "5 Jessicasanchez. team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n", "suffix": "# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0020_email:freq10:rep4", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 9} {"prefix": " NOTICE file\n# distributed with this work for additional information\n# regarding copyright ownership. The ASF licenses this file\n# to you under the Apache License, Version 2.0 (the\n# \"License\"); you may not use this file", "suffix": " except in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:scripts/ci/prek/check_schema_defaults.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 31} {"prefix": "f9-4ed5-4ad97c1a3a6a\",\n \"clientId\": \"7fef70ac-a8cf-c80f-98e0-169c12", "suffix": "0b8dce\",\n \"vmSize\": \"Standard_E8s_v3\",\n \"diagnostics\": {\n \"enabled\": true,\n \"retentionDays\": 90\n },\n \"networking\": {\n \"vnetName", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0118:db_url:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 51} {"prefix": "(self, task: Task, **kwargs: Any) -> TaskStep:\n \"\"\"\n Initialize a new task step.\n\n Args:\n task: The task to execute\n **kwargs: Additional arguments\n\n Returns:\n Initial task step\n\n \"\"\"", "suffix": "\n # Create initial step with trust context\n return TaskStep(\n task_id=task.task_id,\n step_id=f\"{task.task_id}_step_0\",\n input=task.input,\n step", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/agent/llama-index-agent-agentmesh/llama_index/agent/agentmesh/worker.py:TrustedAgentWorker.initialize_step", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 4} {"prefix": ", mock_get_application_builder, mock_get_auth_manager, mock_roles, test_client, as_user\n ):\n mgr = MagicMock()\n mgr.is_authorized_custom_view.return_value", "suffix": " = False\n mock_get_auth_manager.return_value = mgr\n\n with as_user():\n resp = test_client.get(\"/fab/v1/roles/roleA\")\n assert resp.status_code == ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/fab/tests/unit/fab/auth_manager/api_fastapi/routes/test_roles.py:TestRoles.test_get_role_forbidden", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": " Face Transformers and use them as input to vLLM\nfor both single and batch inference.\n\nModel: meta-llama/Llama-3.2-1B-Instruct\nNote: This model is gated", "suffix": " on Hugging Face Hub.\n You must request access to use it:\n https://huggingface.co/meta-llama/Llama-3.2-1B-Instruct\n\nRequirements:\n- v", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:examples/offline_inference/prompt_embed_inference.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": ") Model to encode and decode images to and from latent representations.\n text_encoder ([`Mistral3ForConditionalGeneration`]):\n [Mistral3ForConditionalGeneration](https://huggingface.co/docs/transformers", "suffix": "/en/model_doc/mistral3#transformers.Mistral3ForConditionalGeneration)\n tokenizer (`AutoProcessor`):\n Tokenizer of class\n [PixtralProcessor](https://huggingface.co/docs/", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/flux2/pipeline_flux2.py:Flux2Pipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 132} {"prefix": " variables for data-pipeline (staging)\n# Provider: aws\n\nvariable \"environment\" {\n description = \"Deployment environment\"\n type = string\n default = \"staging\"\n}\n\nvariable \"region\" {\n description", "suffix": " = \"Cloud provider region\"\n type = string\n default = \"eu-west-1\"\n}\n\nvariable \"instance_type\" {\n description = \"Compute instance type\"\n type = string\n default =", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0138:db_url:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": " layer handling\n input1 = layers.Input(shape=(10,), name=\"input_1\")\n input2 = layers.Input(shape=(10,), name=\"input_2\")\n output = layers.Add()([input1", "suffix": ", input2])\n model = models.Model(inputs=[input1, input2], outputs=output)\n ref_input = (base_input, base_input * 2)\n elif struct_type == \"array\":\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "keras-team/keras:keras/src/export/litert_test.py:ExportLitertTest.test_model_with_input_structure", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 94} {"prefix": "self, mock_get_remote_os):\n \"\"\"Test remote_secure_delete handles exceptions gracefully.\"\"\"\n mock_ssh = Mock()\n mock_get_remote_os.side_effect = Exception(\"SSH error\")\n mock", "suffix": "_logger = Mock()\n\n remote_secure_delete(mock_ssh, [\"/remote/file\"], mock_logger)\n\n mock_logger.warning.assert_called_with(\"Failed to remove remote files: %s\", \"SSH error", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/teradata/tests/unit/teradata/utils/test_tpt_util.py:TestTptUtil.test_remote_secure_delete_exception", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 11} {"prefix": " mock_vertex.custom_component.display_name = \"Chat Output\"\n mock_vertex.id = \"chat_output_id\"\n\n mock_result = MagicMock()\n mock_result.vertex = mock_vertex\n mock", "suffix": "_result.result_dict.results = {\"message\": mock_message}\n\n results = [mock_result]\n\n result = extract_result_data(results, \"logs\")\n\n assert result == {\n \"result\": \"Hello world", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/tests/unit/cli/test_common.py:TestResultExtraction.test_extract_result_data_dict_result", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 61} {"prefix": " http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing,\n# software distributed under the License is distributed on an\n# \"AS IS\" BASIS, WITHOUT", "suffix": " WARRANTIES OR CONDITIONS OF ANY\n# KIND, either express or implied. See the License for the\n# specific language governing permissions and limitations\n# under the License.\n#\n# NOTE! THIS FILE IS COPIED MANUALLY IN", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/discord/src/airflow/providers/discord/version_compat.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 97} {"prefix": " use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law", "suffix": " or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/iquest_loopcoder.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 53} {"prefix": "\n \"\"\"\n retval: tuple[list[keras.Layer],\n list[keras.src.ops.node.KerasHistory],\n list[int]] = ([], [], [])\n for layer, history in self._get_candidates(", "suffix": "input_tensors):\n grp_inputs = self._group_inputs(layer, list(zip(input_tensors, node_indices)))\n grp_hist = self._group_inputs(layer, history)\n\n for input_group in", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:plugins/train/model/_base/inference.py:Inference._layers_from_inputs", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 254} {"prefix": "0\n# SPDX-FileCopyrightText: Copyright contributors to the vLLM project\n# Copyright 2025 The Qwen Team and The HuggingFace Inc. team.\n# All rights reserved.\n#\n# Licensed", "suffix": " under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/transformers_utils/configs/qwen3_5.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 11} {"prefix": " list[str]:\n \"\"\"Extract text matching patterns.\"\"\"\n extract_pattern = getattr(self, \"extract_pattern\", \"\")\n if not extract_pattern:\n return []\n\n max_matches = getattr(self, \"max_matches\",", "suffix": " 10)\n\n try:\n matches = re.findall(extract_pattern, text)\n except re.error as e:\n msg = f\"Invalid regex pattern '{extract_pattern}': {e}\"\n raise ValueError(msg", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/components/processing/text_operations.py:TextOperations._text_extract", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": "; ephemeral sessions run\nwithout a mount to minimise host exposure. The container's network namespace is\ndisabled by default (`--network none`) and you can enable further hardening via\n`read_only_rootfs` and `user`.", "suffix": "\n\nThe security guarantees depend on your Docker daemon configuration. Run the agent on\na host where Docker is locked down (rootless mode, AppArmor/SELinux, etc.) and\nreview any additional volumes or capabilities passed through ``extra_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain_v1/langchain/agents/middleware/_execution.py:DockerExecutionPolicy:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 51} {"prefix": "uggingface.co/docs/transformers/en/model_doc/clip#transformers.CLIPTokenizer).\n tokenizer_2 (`T5TokenizerFast`):\n Second Tokenizer of class\n [T5TokenizerFast](https://hug", "suffix": "gingface.co/docs/transformers/en/model_doc/t5#transformers.T5TokenizerFast).\n resolution (`int`, *optional*, defaults to 384):\n The resolution of each image when concatenating images", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/visualcloze/pipeline_visualcloze_generation.py:VisualClozeGenerationPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 334} {"prefix": "3\",\n component_id=\"OpenAIModel-xyz789\",\n component_name=\"OpenAIModel\",\n component_inputs={\"temperature\": 0.7, \"model\": \"gpt-4\"},\n )\n\n assert", "suffix": " payload.component_run_id == \"run-abc-123\"\n assert payload.component_id == \"OpenAIModel-xyz789\"\n assert payload.component_name == \"OpenAIModel\"\n assert", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/integration/test_exception_telemetry.py:TestTelemetryPayloadValidation.test_component_inputs_payload_creation_and_serialization", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 61} {"prefix": "25 Roberto L. Castro (Roberto.LopezCastro@ist.ac.at).\n# All Rights Reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");", "suffix": "\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/kernels/quantization/test_mxfp4_qutlass.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 39} {"prefix": " use connecting to Google Cloud.\n:param impersonation_chain: Optional service account to impersonate using short-term\n credentials, or chained list of accounts required to get the access_token\n of the last account in the list,", "suffix": " which will be impersonated in the request.\n If set as a string, the account must grant the originating account\n the Service Account Token Creator IAM role.\n If set as a sequence, the identities from the list must grant\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/src/airflow/providers/google/cloud/triggers/gen_ai.py:GenAIGeminiCreateBatchJobTrigger:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 241} {"prefix": "=cast(Runtime, object()),\n model_settings={},\n )\n\n modified_request: ModelRequest | None = None\n\n async def mock_handler(req: ModelRequest) -> ModelResponse:\n nonlocal modified_request\n modified_", "suffix": "request = req\n return ModelResponse(result=[AIMessage(content=\"mock response\")])\n\n result = await middleware.awrap_model_call(fake_request, mock_handler)\n assert isinstance(result, ModelResponse)\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/partners/anthropic/tests/unit_tests/middleware/test_prompt_caching.py:test_anthropic_prompt_caching_middleware_async_default_values", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 165} {"prefix": "apiVersion: apps/v1\nkind: Deployment\nmetadata:\n name: auth-gateway\n namespace: staging\nspec:\n replicas: 4\n selector:\n matchLabels:\n app: auth-gateway\n template:\n spec:", "suffix": "\n containers:\n - name: auth-gateway\n image: registry.internal/auth-gateway:latest\n ports:\n - containerPort: 9090\n resources:\n limits:\n cpu: 2000m", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0003:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 204} {"prefix": " that preprocess andencode the image inputs into their latent representations.\n\n Components:\n image_processor (`VaeImageProcessor`) vae (`AutoencoderKL`)\n\n Inputs:\n image (`None`, *optional*):\n TODO: Add description.", "suffix": "\n _auto_resize (`bool`, *optional*, defaults to True):\n TODO: Add description.\n generator (`None`, *optional*):\n TODO: Add description.\n\n Outputs:\n processed_image (`None`):\n TODO: Add", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/flux/modular_blocks_flux_kontext.py:FluxKontextVaeEncoderStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 4} {"prefix": "nested_tokens(self, test_tokenizer):\n \"\"\"Test behavior with nested-like token patterns.\"\"\"\n parser = TestThinkingReasoningParser(test_tokenizer)\n\n model_output = \"OuterInnerContent\"\n reasoning, content = run_reasoning_extraction(parser, [model_output])\n\n # Should process normally, start from first start token\n assert reasoning == \"Outer=3.10, <3.1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:scripts/ci/prek/check_notice_files.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 109} {"prefix": "embeds: [batch_size, seq_len, hidden_dim]\n token_type_ids: [batch_size, seq_len], 0=non-causal, 1=causal\n attention_mask:", "suffix": " [batch_size, seq_len], optional\n \"\"\"\n return self.model(\n inputs_embeds=inputs_embeds,\n token_type_ids=token_type_ids,\n attention_mask=attention_mask", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/deepencoder2.py:CustomQwen2Decoder.forward", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 52} {"prefix": " POST /workflow: Execute a workflow (sync, stream, or background modes)\n GET /workflow: Get workflow job status by job_id\n POST /workflow/stop: Stop a running workflow execution\n\nFeatures:\n - Developer API protection", "suffix": " (requires developer_api_enabled setting)\n - Comprehensive error handling with structured error responses\n - Timeout protection for long-running executions\n - Support for multiple execution modes (sync, stream, background)\n - API key authentication required", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/base/langflow/api/v2/workflow.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 39} {"prefix": "code=204)\n response.headers[\"Cache-Control\"] = \"no-cache\"\n await _set_cors_headers(request, response)\n return response\n\n return [\n Route(\n _with_base(_ROUTE", "suffix": "_SCRIPT_HEALTH, base_url),\n _script_health_endpoint,\n methods=[\"GET\", \"HEAD\"],\n ),\n Route(\n _with_base(_ROUTE_SCRIPT_HEALTH, base_url),\n _", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:lib/streamlit/web/server/starlette/starlette_routes.py:create_script_health_routes", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 192} {"prefix": "0 payload objects\n payloads = []\n for i in range(1000):\n payload = RunPayload(run_seconds=i, run_success=True, client_type=\"oss\", run_id=None, run", "suffix": "_error_message=\"\")\n payloads.append(payload)\n\n creation_time = time.time() - start_time\n\n # Should create 1000 objects reasonably quickly (under 1 second)\n assert creation_time", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/services/telemetry/test_telemetry_schema.py:TestPayloadPerformance.test_payload_creation_performance", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 40} {"prefix": " under the Apache License, Version 2.0 (the\n# \"License\"); you may not use this file except in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www", "suffix": ".apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing,\n# software distributed under the License is distributed on an\n# \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/src/airflow/providers/google/cloud/utils/lineage.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 52} {"prefix": ": str) -> ODPS:\n \"\"\"\n Get an authenticated MaxCompute ODPS Client.\n\n :param project_id: Project ID for the project which the client acts on behalf of.\n :param location: Default location for jobs", "suffix": " / datasets / tables.\n \"\"\"\n creds = self.get_access_key_credential()\n\n return ODPS(\n creds.access_key_id,\n creds.access_key_secret,\n project=project,\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/alibaba/src/airflow/providers/alibaba/cloud/hooks/maxcompute.py:MaxComputeHook.get_client", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": "/setup-node@v4\n with:\n node-version: \"18\"\n - run: npm ci\n - run: npm test\n - run: npm run build\n\n deploy:\n needs: build\n runs-on:", "suffix": " ubuntu-22.04\n if: github.ref =='refs/heads/production'\n\n steps:\n - uses: actions/checkout@v4\n - name: Deploy\n run: |\n echo \"Deploying to staging", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0101:api_key:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 144} {"prefix": "69UzSDuqneSu85DXRpicAJm8O6rE\nregion = ap-northeast-1\noutput = json\n\n[profile s3-data-staging]\nrole_arn = arn:", "suffix": "aws:iam::746689349528:role/S3AccessRole\nsource_profile = default\nregion = ap-northeast-1\nduration_seconds = 14400\n\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0136:api_key:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 67} {"prefix": " height: int, width: int\n ) -> torch.Tensor:\n \"\"\"\n This method allows to interpolate the pre-trained position encodings, to be able to use the model on higher resolution\n images. This method is also adapted to support", "suffix": " torch.jit tracing.\n\n Adapted from:\n - https://github.com/facebookresearch/dino/blob/de9ee3df6cf39fac952ab558447af1fa1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/interns1_vit.py:InternS1VisionEmbeddings.interpolate_pos_encoding", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 16} {"prefix": ",\n ) -> Any:\n \"\"\"\n Get the HuggingFace processor\n (`transformers.ProcessorMixin`) of the model,\n additionally checking its type.\n\n Raises:\n TypeError: If the processor is not of the specified type.", "suffix": "\n \"\"\"\n if typ is None:\n from transformers.processing_utils import ProcessorMixin\n\n typ = ProcessorMixin\n\n tokenizer = self.tokenizer\n if is_mistral_tokenizer(tokenizer):\n tokenizer = tokenizer.transformers_tokenizer\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/multimodal/processing/context.py:InputProcessingContext.get_hf_processor", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 38} {"prefix": "(\"/api/v1/responses\", json=payload, headers=headers))\n\n # Wait for all requests to complete\n responses = await asyncio.gather(*rapid_requests, return_exceptions=True)\n\n # Check that most requests succeeded (", "suffix": "allowing for some potential failures)\n successful_responses = [r for r in responses if not isinstance(r, Exception) and r.status_code == 200]\n\n # At least 50% should succeed\n assert", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/integration/test_openai_responses_extended.py:test_openai_responses_rate_limiting_simulation", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 118} {"prefix": "_reason(f\"{moe_config.activation} activation\")\n elif not _supports_quant_scheme(weight_key, activation_key):\n return False, _make_reason(f\"quantization scheme {weight_key}", "suffix": "x{activation_key}\")\n elif not _supports_parallel_config(moe_config.moe_parallel_config):\n return False, _make_reason(f\"parallel config {moe_config.moe_parallel_config}\")", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/quantization/utils/flashinfer_fp4_moe.py:is_supported_config_trtllm", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 220} {"prefix": " code=w2_states_lst[0].code,\n blocksize=w2_states_lst[0].blocksize,\n quant_type=\"nf4\",\n dtype=w2_states_lst[0].dtype", "suffix": ",\n )\n # The weight suffixes.w13_weight and.w2_weight are consistent\n # with the param in BitsAndBytesMoEMethod.\n w13_weight_name = name + \".w1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/model_loader/bitsandbytes_loader.py:BitsAndBytesModelLoader._fuse_moe_quant_states", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 864} {"prefix": "ized models.\n\nExpected class attributes:\n - model_class: The model class to test\n - pretrained_model_name_or_path: Hub repository ID for the pretrained model\n - pretrained_model_kwargs: (Optional)", "suffix": " Dict of kwargs to pass to from_pretrained\n\nExpected methods to be implemented by subclasses:\n - get_dummy_inputs(): Returns dict of inputs to pass to the model forward pass\n\nPytest mark: bitsandbytes\n Use `", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/models/testing_utils/quantization.py:BitsAndBytesCompileTesterMixin:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": "This is a configurable module that works for many standard DL experiments.\n\nArguments:\n model: A PyTorch model.\n optimizer: A PyTorch optimizer to update model.\n device: The device to train the model on. This defaults to a configurable", "suffix": " device\n loss_function: A function to calculate the loss. This should accept ``model_output, target`` as\n arguments.\n update_batches (int): Number of batches to accumulate before taking an optimizer step.\n Defaults to ``", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "labmlai/annotated_deep_learning_paper_implementations:labml_nn/helpers/trainer.py:SimpleTrainValidConfigs:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": " or bytes) : Data\n custom_alphabet (str, optional): Custom alphabet string\n\n Returns:\n str: Encoded string\n \"\"\"\n b32_enc = AlgoUtils.Decode(base64.b32encode", "suffix": "(AlgoUtils.Encode(data)))\n if custom_alphabet is not None:\n b32_enc = _Base32Utils.TranslateAlphabet(b32_enc, Base32Const.ALPHABET, custom_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "ccxt/ccxt:python/ccxt/static_dependencies/bip/utils/misc/base32.py:Base32Encoder.Encode", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 42} {"prefix": "a batch update data structure for logitsprocs\nAssumptions:\n* All information about requests removed from persistent batch\n during a step is aggregated in self._removed through calls to\n self.removed_append() at the beginning of a step", "suffix": ". This must happen\n before the first time that self.removed, self.pop_removed()\n or self.peek_removed() are invoked in a given step\n* After the first time that self.removed, self.pop_removed", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/v1/sample/logits_processor/state.py:BatchUpdateBuilder:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": " `body` 3\nIF (EXISTS `.cookie-banner`) THEN CLICK `.accept`\nCLICK `.search-button`\nTYPE \"python tutorial\"\nPRESS Enter\nWAIT `.results` 5\n\"\"\"\n \n console.print", "suffix": "(\"[cyan]C4A Script Example:[/cyan]\")\n console.print(Panel(c4a_script, title=\"script.c4a\", border_style=\"blue\"))\n \n # Compile the script\n compilation_result = c", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/releases_review/demo_v0.7.0.py:demo_5_c4a_script", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 82} {"prefix": " []\n for key, value in valid_attrs.items():\n # Clean up array formatting for better readability\n if isinstance(value, list):\n value_str = ', '.join(str(v) for v in value)\n ", "suffix": " else:\n value_str = str(value)\n attrs_parts.append(\n f'{html.escape(str(key))}: Config:\n \"\"\"Run autotuning for a single input configuration.\"\"\"\n extra_kwargs", "suffix": " = {\"autotune_effort\": autotune_effort}\n autotune_kernel = create_helion_decorated_kernel(\n self.raw_kernel_func, self.helion_settings, extra", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/kernels/helion/register.py:HelionKernelWrapper.run_autotune", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": "def test_create_python_instance(self, provider):\n \"\"\"Test creating a Python sandbox instance.\"\"\"\n try:\n instance = provider.create_instance(\"python\")\n\n assert instance.provider == \"aliyun_codeinterpreter\"\n assert", "suffix": " instance.status in [\"READY\", \"CREATING\"]\n assert instance.metadata[\"language\"] == \"python\"\n assert len(instance.instance_id) > 0\n\n # Clean up\n provider.destroy_instance(instance.instance", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:agent/sandbox/tests/test_aliyun_codeinterpreter_integration.py:TestAliyunCodeInterpreterIntegration.test_create_python_instance", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "LongTensor | None = None,\n ) -> torch.Tensor:\n \"\"\"\n Args:\n pixel_values (`torch.FloatTensor`):\n Pixel values of shape (\n num_patches,\n num_channels * temporal_patch_size", "suffix": " * patch_size * patch_size\n )\n grid_thws: (`torch.LongTensor`):\n grid shape (num_patches, 3)\n \"\"\"\n\n # Apply patch embeddings to already patchified pixel values\n target_dtype", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/siglip2navit.py:Siglip2VisionEmbeddings.forward", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 24} {"prefix": "_HOST=postgres.svc.cluster.local\nDB_PORT=5432\nDB_NAME=webapp\nDB_USER=app_user\nEXTERNAL_API_KEY=AKIACANARY0126DY", "suffix": "K0X6\n\n# Redis / Cache\nREDIS_URL=redis://localhost:6379/0\nCACHE_TTL=2858\n\n# Email / SMTP\nSMTP_HOST=email-smtp.us", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0126:api_key:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 92} {"prefix": "2-2026)\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of", "suffix": " the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0025_email:freq10:rep4", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": "table_row_to_node(self, row: dict[str, Any]) -> BaseNode:\n \"\"\"Convert a table row dictionary to a BaseNode object.\"\"\"\n metadata = row.get(\"metadata\")\n if metadata is None:", "suffix": "\n raise ValueError(\"Metadata not found in row data.\")\n\n node = metadata_dict_to_node(metadata, text=row.get(\"content\"))\n # Convert UUID to string if needed\n node_id = row.get(\"id", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/vector_stores/llama-index-vector-stores-azurepostgresql/llama_index/vector_stores/azure_postgres/base.py:AzurePGVectorStore._table_row_to_node", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": "conn_id: Connection ID for the LLM provider.\n:param model_id: Model identifier (e.g. ``\"openai:gpt-4o\"``).\n Overrides the model stored in the connection's extra field", "suffix": ".\n:param system_prompt: Additional instructions appended to the built-in SQL\n safety prompt. Use for domain-specific guidance.\n:param agent_params: Additional keyword arguments passed to the pydantic-ai\n ``Agent``", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/common/ai/src/airflow/providers/common/ai/operators/llm_sql.py:LLMSQLQueryOperator:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 181} {"prefix": "65766061:role/DataPipelineRole\nsource_profile = default\nregion = us-east-1\nduration_seconds = 43200\n\n# S3 bucket configuration\ns3 =", "suffix": "\n max_concurrent_requests = 17\n max_queue_size = 245\n multipart_threshold = 8MB\n multipart_chunksize = 16MB\n\n# VPC endpoints\nvpc_endpoint = ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0084:internal_ip:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 104} {"prefix": "process\" in klass.__dict__:\n group_definer = klass\n break\n\n # Check what was overridden (not defined in base class)\n has_process = process_definer is not None and process_definer is not base_", "suffix": "class\n has_group = group_definer is not None and group_definer is not base_class\n\n if has_process and has_group:\n raise ValueError(\n f\"{cls.__name__}: Cannot override both _process", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_extras/nodes_dataset.py:TextProcessingNode._detect_processing_mode", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 171} {"prefix": ".\n# https://github.com/ByteDance-Seed/VeOmni/blob/v0.1.4/veomni/utils/dist_utils.py\n#\n# Licensed under the Apache License, Version", "suffix": " 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:src/llamafactory/v1/accelerator/helper.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 37} {"prefix": " (pipe.generation_pipe.vae_scale_factor * 2)\n expected_width = width - width % (pipe.generation_pipe.vae_scale_factor * 2)\n\n inputs.update({\"upsampling", "suffix": "_height\": height, \"upsampling_width\": width})\n image = pipe(**inputs).images[0]\n output_height, output_width, _ = image.shape\n assert (output_height, output_width) ==", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/pipelines/visualcloze/test_pipeline_visualcloze_combined.py:VisualClozePipelineFastTests.test_visualcloze_image_output_shape", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 97} {"prefix": " ModelsLabLLM\n\n # Set MODELSLAB_API_KEY env var or pass api_key directly\n llm = ModelsLabLLM(\n model=\"llama-3.1-8b-uncensored\",\n api", "suffix": "_key=\"your-modelslab-api-key\",\n )\n\n resp = llm.complete(\"Explain transformers in simple terms.\")\n print(resp)\n ```\n\n Use in a RAG pipeline::\n\n ```python\n from ll", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/llms/llama-index-llms-modelslab/llama_index/llms/modelslab/base.py:ModelsLabLLM:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 153} {"prefix": " extends Qwen3-VL with a ColBERT-style late interaction head,\nproducing per-token embeddings for both text and image inputs. It uses\nMaxSim scoring for retrieval/reranking tasks.\n\nThis model supports", "suffix": " the \"token_embed\" pooling task and is designed for\nmulti-vector retrieval of documents containing both text and images.\n\nReference: https://arxiv.org/abs/2407.01449 (ColP", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/colqwen3.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 23} {"prefix": "gingFace Team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy", "suffix": " of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/transformers/transformer_longcat_image.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 19} {"prefix": "`, *optional*, defaults to 1.0):\n When to stop applying ControlNet.\n controlnet_conditioning_scale (`float`, *optional*, defaults to 1.0):\n Scale for ControlNet conditioning.\n control", "suffix": "_image_latents (`Tensor`):\n The control image latents to use for the denoising process. Can be generated in controlnet vae encoder\n step.\n timesteps (`Tensor`):\n The timesteps to use for the den", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/before_denoise.py:QwenImageControlNetBeforeDenoiserStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 77} {"prefix": ")\n\n assert len(self.kv_caches) > 0\n\n kvcaches = list(self.kv_caches.values())\n if self.current_layer == 0:\n self.layerwise_storers", "suffix": " = []\n\n is_first = True\n\n for idx, request in enumerate(connector_metadata.requests):\n save_spec = request.save_spec\n if save_spec is None or not save_spec.can_save:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/distributed/kv_transfer/kv_connector/v1/lmcache_integration/vllm_v1_adapter.py:LMCacheConnectorV1Impl.save_kv_layer", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 243} {"prefix": "Qwen/Qwen2.5-VL-7B-Instruct) encoder.\n tokenizer (`AutoTokenizer`):\n Tokenizer associated with the Qwen2.5 VL encoder.\n transformer ([`CosmosTransformer3DModel`", "suffix": "]):\n Conditional Transformer to denoise the encoded image latents.\n scheduler ([`UniPCMultistepScheduler`]):\n A scheduler to be used in combination with `transformer` to denoise the encoded image latents.\n v", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/cosmos/pipeline_cosmos2_5_predict.py:Cosmos2_5_PredictBasePipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 123} {"prefix": "kip(\"system.standard\")\n dagbag = DagBag(dag_folder=os.devnull, include_examples=False)\n dagbag.process_file(os.fspath(file_to_load))\n ", "suffix": " for dag_id, path in expected.items():\n dag = dagbag.get_dag(dag_id)\n assert dag, f\"{dag_id} was bagged\"\n assert dag.fileloc.endswith(path)", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:airflow-core/tests/unit/dag_processing/test_dagbag.py:TestDagBag.test_get_dag_registration", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 24} {"prefix": " specific language governing permissions and\n# limitations under the License.\n#\n# jupyter:\n# jupytext:\n# cell_metadata_filter: -all\n# formats: ipynb,md:myst,py\n#", "suffix": " text_representation:\n# extension:.py\n# format_name: light\n# format_version: '1.5'\n# jupytext_version: 1.16.4\n# ---\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:docs/array_refs.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 135} {"prefix": "\ntype: Opaque\ndata:\n api-key: c2stQ0FOQVJZMDAxMVp0ZDI2ZkVlVlZoRElxMkFuSFRtdDlPQkdobnVLb2", "suffix": "5lTm80MWVvUG5pNkpEV1lsZw==\n session-secret: Um5JQ2VlTHAzc2FCVU9VRG8wV2o=\n---\napiVersion: v1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0011:api_key:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 38} {"prefix": "dict(str_var: str | None) -> dict[str, str]:\n \"\"\"\n Convert a string of key-value pairs to a dictionary.\n\n Environment variables like 'OTEL_RESOURCE_ATTRIBUTES' or 'OTEL_EXPORT", "suffix": "ER_OTLP_HEADERS'\n accept values with the format \"key1=value1,key2=value2,...\"\n \"\"\"\n configs = {}\n if str_var:\n for pair in str_var.split(\",\"):", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:shared/observability/src/airflow_shared/observability/otel_env_config.py:_parse_kv_str_to_dict", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": " = attachment.get(\"filename\")\n if not filename:\n continue\n size = attachment.get(\"size\")\n size_text = f\" ({size} bytes)\" if isinstance(size, int) else \"\"\n content_url = attachment", "suffix": ".get(\"content\") or \"\"\n url_suffix = f\" -> {content_url}\" if content_url else \"\"\n attachment_lines.append(f\"- {filename}{size_text}{url_suffix}\")\n\n return \"\\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:common/data_source/jira/utils.py:format_attachments", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 52} {"prefix": ",\n model_kwargs=hf_model_kwargs,\n ) as hf_model:\n tokenizer = hf_model.tokenizer\n hf_outputs = []\n for prompt in example_prompts:\n inputs = tokenizer([prompt],", "suffix": " return_tensors=\"pt\")\n inputs = hf_model.wrap_device(inputs)\n output = hf_model.model(**inputs)\n hf_outputs.append(softmax(output.logits[0]))\n\n # check logits difference", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/models/language/pooling/test_token_classification.py:test_bert_models", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 197} {"prefix": ". Can be generated in set_timesteps step.\n num_inference_steps (`int`):\n The number of denoising steps.\n latents (`Tensor`):\n The initial latents to use for the denoising process. Can", "suffix": " be generated in prepare_latent step.\n image_latents (`Tensor`):\n image latents used to guide the image generation. Can be generated from vae_encoder step.\n attention_kwargs (`dict`, *optional*):\n Additional", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/denoise.py:QwenImageLayeredDenoiseStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 153} {"prefix": "leutherAI's GPT-NeoX library and the GPT-NeoX\n# and OPT implementations in this library. It has been modified from its\n# original forms to accommodate the architectural differences made by\n# the", "suffix": " Swiss AI Initiative that trained the model.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/apertus.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 141} {"prefix": " qkv_proj.\"\"\"\n stacked_params_mapping = [\n (\".qkv_proj\", \".q_proj\", \"q\"),\n (\".qkv_proj\", \".k_proj\", \"k\"),\n (\".q", "suffix": "kv_proj\", \".v_proj\", \"v\"),\n ]\n\n params_dict = dict(self.named_parameters())\n loaded_params: set[str] = set()\n\n for name, loaded_weight in weights:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/arcee.py:ArceeModel.load_weights", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 40} {"prefix": "with_required_field_only(self):\n \"\"\"Should create request with only required flow_id field.\"\"\"\n request = AssistantRequest(flow_id=\"test-flow-id\")\n\n assert request.flow_id == \"test-", "suffix": "flow-id\"\n assert request.component_id is None\n assert request.field_name is None\n assert request.input_value is None\n assert request.max_retries is None\n assert request.model_name is None\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/agentic/api/test_schemas.py:TestAssistantRequest.test_should_create_with_required_field_only", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": "#transformers.T5Tokenizer),\n specifically the [google/umt5-xxl](https://huggingface.co/google/umt5-xxl) variant.\n text_encoder ([`UMT5Encoder", "suffix": "Model`]):\n [T5](https://huggingface.co/docs/transformers/en/model_doc/t5#transformers.T5EncoderModel), specifically\n the [google/umt5-xxl](", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2_diffusion_forcing_i2v.py:SkyReelsV2DiffusionForcingImageToVideoPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 105} {"prefix": "\",\n \"client_email\": \"pubsub-publisher@data-warehouse-01.iam.gserviceaccount.com\",\n \"client_id\": \"502886382086\",\n \"auth", "suffix": "_uri\": \"https://accounts.google.com/o/oauth2/auth\",\n \"token_uri\": \"https://oauth2.googleapis.com/token\",\n \"auth_provider_x509_cert_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0153:db_url:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 132} {"prefix": " License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org", "suffix": "/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/transformers/transformer_glm_image.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": " in saved_mineru_models:\n api_cfg = _parse_api_key(item.api_key)\n normalized = {k: api_cfg.get(k, MINERU_DEFAULT_CONFIG.get(", "suffix": "k)) for k in MINERU_ENV_KEYS}\n if normalized == cfg:\n return item.llm_name\n\n used_names = {item.llm_name for item in saved_mineru_models}\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:api/db/services/tenant_llm_service.py:TenantLLMService.ensure_mineru_from_env", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 166} {"prefix": " links from the given lines.\n\n Return list of MarkdownLinkInfo, where each dict contains:\n - `line_no` - line number (1-based)\n - `url` - link URL\n - `text` - link text", "suffix": "\n - `title` - link title (if any)\n \"\"\"\n\n links: list[MarkdownLinkInfo] = []\n for line_no, line in enumerate(lines, start=1):\n for m in MARKDOWN_LINK_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "fastapi/fastapi:scripts/doc_parsing_utils.py:extract_markdown_links", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 26} {"prefix": ": use all left chunks\n Returns:\n encoder output tensor xs, and subsampled masks\n xs: padded output tensor (B, T' ~= T/subsample_rate, D)\n masks: torch.Tensor batch padding mask after sub", "suffix": "sample\n (B, 1, T' ~= T/subsample_rate)\n NOTE(xcsong):\n We pass the `__call__` method of the modules instead of `forward` to the\n checkpointing API because `__", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy/ldm/ace/lyric_encoder.py:ConformerEncoder.forward", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 193} {"prefix": "\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\"", "suffix": " BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under the License.\n# Contact: jessicasanchez@yahoo.com", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0020_email:freq10:rep5", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 64} {"prefix": "(mock_vercel_ai_gateway_llm, \"_get_aclient\") as mock_get_aclient:\n mock_client = mock_get_aclient.return_value\n mock_response = type(\n ", "suffix": " \"MockResponse\",\n (),\n {\n \"choices\": [\n type(\n \"MockChoice\",\n (),\n {\n \"message\": type(\n \"MockMessage\",\n (),\n {\n \"content\": \"mock async\",\n \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/llms/llama-index-llms-vercel-ai-gateway/tests/test_vercel_ai_gateway.py:test_achat_mock", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 26} {"prefix": "utput | tuple[DiagonalGaussianDistribution]:\n r\"\"\"\n Encode a batch of images into latents.\n\n Args:\n x (`torch.Tensor`): Input batch of images.\n return_dict (`bool`, *optional*, defaults to", "suffix": " `True`):\n Whether to return a [`~models.autoencoder_kl.AutoencoderKLOutput`] instead of a plain tuple.\n\n Returns:\n The latent representations of the encoded videos. If `return_dict` is True", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/autoencoders/autoencoder_kl_hunyuanvideo15.py:AutoencoderKLHunyuanVideo15.encode", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 26} {"prefix": "/vllm-project/vllm/blob/main/examples/pooling/score/qwen3_reranker_online.py\n Reference: https://github.com/vllm-project/vll", "suffix": "m/blob/main/examples/pooling/score/convert_model_to_seq_cls.py\n\nrun:\n vllm serve BAAI/bge-reranker-v2-gemma --", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:examples/pooling/score/using_template_online.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 97} {"prefix": ", self.virtual_device_id)\n\n self.data = get(\n self.src_device_id,\n self.src_local_core_id,\n self.src_memory_space,\n self.src_", "suffix": "buffer_id,\n self.src_transforms,\n clock=vc.copy_vector_clock(self.clock),\n src_device_id=self.id,\n src_local_core_id=0,\n source", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/pallas/mosaic/interpret/interpret_pallas_call.py:DMA.execute_read", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 132} {"prefix": "2)\n bounding_boxes = {\"boxes\": ragged_boxes, \"labels\": ragged_labels}\n max_boxes = 4\n densified_data = validation.densify_bounding_boxes(\n bounding_boxes", "suffix": ".copy(), is_batched=False, max_boxes=max_boxes\n )\n densified_boxes = densified_data[\"boxes\"]\n densified_labels = densified_data[\"labels\"]\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "keras-team/keras:keras/src/layers/preprocessing/image_preprocessing/bounding_boxes/validation_test.py:DensifyBoundingBoxesTest.test_densify_ragged_bounding_boxes_unbatched", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 108} {"prefix": "\",\n \"required\": True,\n \"label\": \"API Key\",\n \"placeholder\": \"e2b_sk_...\",\n \"description\": \"E2B API key for authentication\",\n \"secret\": True,\n },\n \"", "suffix": "region\": {\n \"type\": \"string\",\n \"required\": False,\n \"label\": \"Region\",\n \"default\": \"us\",\n \"description\": \"E2B service region (us or eu)\",\n },\n \"timeout", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:agent/sandbox/providers/e2b.py:E2BProvider.get_config_schema", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 56} {"prefix": "(f\"Heading key {heading_key} from heading {heading} not found in basic text\")\n heading_start = prev_start\n continue\n\n new_sections.append(\n TextSection(\n link=adv_sections[adv", "suffix": "_ind - 1].link,\n text=basic_full_text[prev_start:heading_start],\n )\n )\n\n # handle last section\n new_sections.append(TextSection(link=adv_sections[-", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:common/data_source/google_drive/doc_conversion.py:align_basic_advanced", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 447} {"prefix": " agreements. See the NOTICE file\n# distributed with this work for additional information\n# regarding copyright ownership. The ASF licenses this file\n# to you under the Apache License, Version 2.0 (the\n# \"License\"); you", "suffix": " may not use this file except in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0025_email:freq3:rep2", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 18} {"prefix": " \"text\"},\n \"available_int\": {\"type\": \"integer\"},\n \"important_kwd\": {\"type\": \"keyword\"},\n \"q_768_vec\": {\"type\": \"dense_vector\", \"dims\": ", "suffix": "768},\n }\n }\n \n analysis = converter.analyze_es_mapping(es_mapping)\n \n # Check known fields\n assert \"id\" in analysis[\"known_fields\"]\n assert \"kb_id\" in analysis[\"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:tools/es-to-oceanbase-migration/tests/test_schema.py:TestRAGFlowSchemaConverter.test_analyze_ragflow_mapping", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 126} {"prefix": " working with a remote code interpreter environment:\n\n* execute_code - Run code in various languages (primarily Python)\n* execute_command - Run shell commands\n* read_files - Read content of files in the environment\n*", "suffix": " list_files - List files in directories\n* delete_files - Remove files from the environment\n* write_files - Create or update files\n* start_command - Start long-running commands asynchronously\n* get_task - Check status of", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/tools/llama-index-tools-aws-bedrock-agentcore/llama_index/tools/aws_bedrock_agentcore/code_interpreter/base.py:AgentCoreCodeInterpreterToolSpec:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 20} {"prefix": "_loss(micro_batch)\n mini_step_valid_tokens = compute_valid_tokens([micro_batch])\n # fsdp uses mean reduction so we need to scale the loss by dp_size\n loss = loss * mini", "suffix": "_step_valid_tokens * self.dp_size / (step_valid_tokens + 1e-6)\n\n if self._deepspeed_engine is not None:\n # deepspeed: set sync_gradients so engine.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:src/llamafactory/v1/core/base_trainer.py:BaseTrainer.fit", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 157} {"prefix": "(self):\n \"\"\"String values should be normalized to dict format and input_types should be set.\"\"\"\n model_input = ModelInput(name=\"test_model\", value=\"gpt-4o\")\n # Value should be normalized\n ", "suffix": " assert isinstance(model_input.value, list)\n if model_input.value: # May be normalized or fallback\n assert isinstance(model_input.value[0], dict)\n # Should have connection handle based on model_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/tests/unit/inputs/test_model_input_fixes.py:TestModelInputPortVisibility.test_string_value_normalization", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 9} {"prefix": "loader with group-based layer selection.\n\nGroups layers and uses async H2D prefetch to hide transfer latency.\nUses static buffers and stream synchronization for torch.compile and\nCUDA graph compatibility.\n\nArgs:\n group_size:", "suffix": " Group every N layers together.\n num_in_group: Offload this many layers per group (last N of each group).\n prefetch_step: Number of layers to prefetch ahead.\n mode: Offload mode (\"cpu\" is currently", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/offloader/prefetch.py:PrefetchOffloader:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 6} {"prefix": "\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n#", "suffix": " http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/models/controlnets/test_models_controlnet_cosmos.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": " H, W) into layer images.\n\n Components:\n vae (`AutoencoderKLQwenImage`) image_processor (`VaeImageProcessor`)\n\n Inputs:\n latents (`Tensor`):\n The denoised latents to decode,", "suffix": " can be generated in the denoise step and unpacked in the after denoise\n step.\n output_type (`str`, *optional*, defaults to pil):\n Output format: 'pil', 'np', 'pt'.\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/decoders.py:QwenImageLayeredDecoderStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": " in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in", "suffix": " writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0035_email:freq10:rep0", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": "(\"browser.\")\n ):\n events.extend(emit_browser_tool_events(previous_item, state))\n\n # Handle tool completion\n if (\n tool_server is not None\n and previous_item.recipient is not None\n ", "suffix": " and state.current_item_id is not None\n and state.sent_output_item_added\n ):\n recipient = previous_item.recipient\n if recipient == \"python\":\n events.extend(emit_code_interpreter_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/entrypoints/openai/responses/streaming_events.py:emit_tool_action_events", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 150} {"prefix": " Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses", "suffix": "/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0015_email:freq10:rep2", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 48} {"prefix": "\n\n # Check if all required functions are available\n required_functions = [\n (\"flashinfer.comm\", \"Mapping\"),\n (\"flashinfer.comm.mnnvl\", \"MnnvlMemory\"),\n (\"flashinfer.comm.tr", "suffix": "tllm_alltoall\", \"MnnvlMoe\"),\n (\"flashinfer.comm.trtllm_alltoall\", \"MoEAlltoallInfo\"),\n ]\n\n for module_name, attr_name", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/utils/flashinfer.py:has_flashinfer_all2all", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 44} {"prefix": "\n }\n\n headers = {\"x-api-key\": created_api_key.api_key}\n response = await client.post(\"api/v2/workflows\", json=request_data, headers=headers)\n\n # Now", "suffix": " background mode is partially implemented and should NOT return 501\n # It should return a WorkflowJobResponse (wrapped in WorkflowExecutionResponse or similar)\n assert response.status_code == 200\n result = response.json()", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/api/v2/test_workflow.py:TestWorkflowErrorHandling.test_background_mode_returns_501", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 208} {"prefix": " Adapted from\n# https://github.com/vllm-project/vllm/blob/v0.7.3/vllm/model_executor/models/deepseek_mtp.py\n# Copyright", "suffix": " 2025 Xiaomi Corporation.\n# Copyright 2023 The vLLM team.\n# Copyright 2024 DeepSeek-AI team.\n\n# Licensed under the Apache License, Version 2", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/mimo_mtp.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 31} {"prefix": ": Sequence[PIIMatch]) -> None:\n \"\"\"Initialize the exception with match context.\n\n Args:\n pii_type: Name of the detected sensitive type.\n matches: All matches that were detected for that type.\n \"\"\"\n ", "suffix": " self.pii_type = pii_type\n self.matches = list(matches)\n count = len(matches)\n msg = f\"Detected {count} instance(s) of {pii_type} in text content", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain_v1/langchain/agents/middleware/_redaction.py:PIIDetectionError.__init__", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": "\n if component_extractor is None and component is not None:\n component_code = component[\"code\"]\n component_output_types = component[\"output_types\"]\n\n component_extractor = Component(_code=component_code)\n\n component", "suffix": "_template, component_instance = build_custom_component_template(\n component_extractor, module_name=module_name\n )\n if not component_template[\"output_types\"] and component_output_types:\n component_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/custom/utils.py:create_component_template", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 125} {"prefix": " wrapper around transformations that enables execution in Ray.\n\nArgs:\n transform_class (Type[TransformComponent]): The transformation class to wrap.\n transform_kwargs (Optional[Dict[str, Any]], optional): The keyword arguments to pass to", "suffix": " the transformation __init__ function.\n map_batches_kwargs (Optional[Dict[str, Any]], optional): The keyword arguments to pass to ray.data.Dataset.map_batches (see https://docs.ray.io/en", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/ingestion/llama-index-ingestion-ray/llama_index/ingestion/ray/transform.py:RayTransformComponent:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": "global_shardings: NamedSharding for each input using global mesh.\n mesh: The global Mesh for this pmap invocation.\n out_specs: Output PartitionSpecs as a pytree prefix.\n out_local_shardings_thunk", "suffix": ": Cached thunk returning (local, global) sharding\n pairs for output pspecs.\n donate_argnums: Indices of donated arguments.\n out_global_shardings: Output NamedShardings as a pytree.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/pmap.py:CachedShardMap:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 73} {"prefix": ",\n render: bool = True,\n key: int | str | tuple[int | str,...] | None = None,\n preserved_by_key: list[str] | str | None = None,\n ):\n \"\"\"\n ", "suffix": " Parameters:\n selected: The currently selected step. Must be a number corresponding to the step number. Defaults to the first step.\n visible: If False, Walkthrough will be hidden.\n elem_id: An optional string that is assigned as", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "gradio-app/gradio:gradio/layouts/walkthrough.py:Walkthrough.__init__", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 51} {"prefix": " the Apache Software Foundation (ASF) under one\n# or more contributor license agreements. See the NOTICE file\n# distributed with this work for additional information\n# regarding copyright ownership. The ASF licenses this file\n# to you under the Apache", "suffix": " License, Version 2.0 (the\n# \"License\"); you may not use this file except in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www.apache.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0015_email:freq10:rep1", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": "-----END RSA PRIVATE KEY-----\\n\",\n \"client_email\": \"pubsub-publisher@my-project-prod.iam.gserviceaccount.com\",\n \"client_id\": \"897475461", "suffix": "917\",\n \"auth_uri\": \"https://accounts.google.com/o/oauth2/auth\",\n \"token_uri\": \"https://oauth2.googleapis.com/token\",\n \"auth_provider_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0065:email:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 123} {"prefix": " tag_end\n # Next nearest is >, means found XML element\n else:\n return buffer[: tag_end2 + 1], start_pos + tag_end2 + 1\n elif tag_end!= -1:\n return", "suffix": " buffer[:tag_end], start_pos + tag_end\n elif tag_end2!= -1:\n return buffer[: tag_end2 + 1], start_pos + tag_end2 + 1\n else:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/tool_parsers/step3p5_tool_parser.py:StreamingXMLToolCallParser._find_next_complete_element", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 420} {"prefix": "\n python-version: \"3.11\"\n - run: pip install -r requirements.txt\n - run: pytest tests/ -v --cov=src\n\n deploy:\n needs: build\n runs-on: ubuntu-latest", "suffix": "\n if: github.ref =='refs/heads/main'\n\n steps:\n - uses: actions/checkout@v4\n - name: Deploy\n run: |\n echo \"Deploying to staging...\"\n ./scripts/deploy.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0122:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 129} {"prefix": "2.py, *_pb2_grpc.py, and\n*_pb2.pyi (type stubs) files from the vllm_engine.proto definition.\n\nNOTE: Proto compilation happens automatically during package build (via", "suffix": " setup.py).\nThis script is provided for developers who want to regenerate protos manually,\ne.g., after modifying vllm_engine.proto.\n\nUsage:\n python vllm/grpc/compile_protos.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/grpc/compile_protos.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 23} {"prefix": " gbl = [\"/usr/include\", \"/usr/local/include\"]\n lcl = [os.path.join(f, \"include\") for f in self._paths]\n for root in gbl + lcl:\n for", "suffix": " folder, _, filenames in os.walk(root):\n if self._cudnn_header not in filenames:\n continue\n version = self.cudnn_version_from_header(folder)\n if not version:\n continue\n cuda", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:lib/system/ml_libs.py:CudaLinux.get_cudnn_versions", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 110} {"prefix": "`]):\n A scheduler to be used in combination with `transformer` to denoise the encoded image latents.\n vae ([`AutoencoderKL`]):\n Variational Auto-Encoder (VAE) Model to encode and decode images", "suffix": " to and from latent representations.\n text_encoder ([`Qwen2.5-VL-7B-Instruct`]):\n [Qwen2.5-VL-7B-Instruct](https://huggingface.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/qwenimage/pipeline_qwenimage_edit_plus.py:QwenImageEditPlusPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 57} {"prefix": ".0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-", "suffix": "2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/autoencoders/autoencoder_kl_ltx2.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 40} {"prefix": "\"\n}\n\nvariable \"instance_type\" {\n description = \"Compute instance type\"\n type = string\n default = \"Standard_B2s\"\n}\n\nvariable \"min_instances\" {\n description =", "suffix": " \"Minimum number of instances in ASG\"\n type = number\n default = 2\n}\n\nvariable \"max_instances\" {\n description = \"Maximum number of instances in ASG\"\n type = number\n default", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0161:api_key:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 70} {"prefix": "ents for image generation.\n generator (`Generator`, *optional*):\n Torch generator for deterministic generation.\n num_inference_steps (`int`, *optional*, defaults to 50):\n The number of denoising steps.\n ", "suffix": " sigmas (`list`, *optional*):\n Custom sigmas for the denoising process.\n attention_kwargs (`dict`, *optional*):\n Additional kwargs for attention processors.\n **denoiser_input_fields (`None`, *", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/modular_blocks_qwenimage_edit_plus.py:QwenImageEditPlusCoreDenoiseStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 298} {"prefix": "5b2180f41\",\n \"vmSize\": \"Standard_E8s_v3\",\n \"diagnostics\": {\n \"enabled\": true,\n \"retentionDays\": 30\n },\n \"networking\":", "suffix": " {\n \"vnetName\": \"vnet-rg-ml-prod\",\n \"subnetName\": \"subnet-default\",\n \"nsgName\": \"nsg-rg-ml-prod\"\n },\n \"privateEndpointIp\": \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0094:internal_ip:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 102} {"prefix": ",\n database=self.database,\n user=username,\n password=password,\n charset='utf8mb4',\n use_unicode=True,\n )\n except Exception as e:\n raise ConnectorValidationError(f\"Failed to", "suffix": " connect to MySQL: {e}\")\n elif self.db_type == DatabaseType.POSTGRESQL:\n try:\n import psycopg2\n except ImportError:\n raise ConnectorValidationError(\n \"PostgreSQL connector not installed. Please install psy", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:common/data_source/rdbms_connector.py:RDBMSConnector._get_connection", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 127} {"prefix": "2023 The JAX Authors.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a", "suffix": " copy of the License at\n#\n# https://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:tests/pallas/tpu_pallas_call_print_test.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": " use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law", "suffix": " or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/glm_ocr_mtp.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 187} {"prefix": "23 The HuggingFace Team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n#", "suffix": " You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/modular_pipeline_utils.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": "\n timeline_data,\n x_start=\"start\",\n x_end=\"end\",\n y=\"request_id\",\n color=\"type\",\n color_discrete_map=labels_colors,\n category_orders={\"type\": labels", "suffix": "_order},\n hover_data=[\n \"prompt_tokens\",\n \"output_tokens\",\n \"req_start_time\",\n \"req_finish_time\",\n \"segment_start\",\n \"segment_end\",\n \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/benchmarks/plot.py:generate_timeline_plot", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 519} {"prefix": "o/oauth2/auth\",\n \"token_uri\": \"https://oauth2.googleapis.com/token\",\n \"auth_provider_x509_cert_url\": \"https://www.googleapis.com/oauth", "suffix": "2/v1/certs\",\n \"universe_domain\": \"googleapis.com\",\n \"scopes\": [\n \"https://www.googleapis.com/auth/bigquery\",\n \"https://www.googleapis.com/auth/cloud", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0189:internal_ip:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 199} {"prefix": ",\n max_retries: int = 5,\n retry_delay: float = 1.0,\n retry_backoff: float = 2.0,\n cls: type[COMFY_IO.ComfyNode] =", "suffix": " None,\n) -> None:\n \"\"\"Stream-download a URL to `dest`.\n\n `dest` must be one of:\n - a BytesIO (rewound to 0 after write),\n - a file-like object opened in", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_api_nodes/util/download_helpers.py:download_url_to_bytesio", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": " schedule, with multiple operators, task groups, dependencies etc.\n\nIt checks:\n - required keys\n - field formats and types\n - number of task events (one start, one complete)\n - if EmptyOperator will emit OL events", "suffix": " with callback or outlet\n - if EmptyOperator without modification will not emit OL events\n - if CustomOperator without Extractor will emit OL events\n - task groups serialization without dependencies\n - additional task configuration attrs (owner, max_active", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/openlineage/tests/system/openlineage/example_openlineage_base_complex_dag.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": ".\n\nvLLM supports ColBERT with multiple encoder backbones. Start the server\nwith one of the following:\n\n # BERT backbone (works out of the box)\n vllm serve answerdotai/answ", "suffix": "erai-colbert-small-v1\n\n # ModernBERT backbone\n vllm serve lightonai/GTE-ModernColBERT-v1 --hf-overrides '{\"architectures\": [\"ColBERT", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:examples/pooling/score/colbert_rerank_online.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 66} {"prefix": ".03667\n- code: https://github.com/JierunChen/FasterNet\n\n@article{chen2023run,\n title={Run, Don't Walk: Chasing Higher", "suffix": " FLOPS for Faster Neural Networks},\n author={Chen, Jierun and Kao, Shiu-hong and He, Hao and Zhuo, Weipeng and Wen, Song and Lee, Chul", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/models/fasternet.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 38} {"prefix": "IOqSGU!\n TIMEOUT: 140\n\njobs:\n build:\n runs-on: ubuntu-latest\n timeout-minutes: 19\n\n steps:\n - uses: actions/checkout@v4\n ", "suffix": " with:\n fetch-depth: 0\n\n - uses: actions/setup-python@v5\n with:\n python-version: \"3.11\"\n - run: pip install -r requirements.txt\n - run:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0082:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 51} {"prefix": "/production\n TIMEOUT: 140\n\njobs:\n build:\n runs-on: ubuntu-22.04\n timeout-minutes: 18\n\n steps:\n - uses: actions/checkout@v4", "suffix": "\n with:\n fetch-depth: 0\n\n - uses: actions/setup-python@v5\n with:\n python-version: \"3.11\"\n - run: pip install -r requirements.txt\n - run", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0013:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 68} {"prefix": "undant_experts: int,\n ) -> Sequence[int]:\n \"\"\"\n Build an initial expert arrangement using the following structure:\n [original routed experts, redundant experts]\n\n Returns:\n physical_to_logical_", "suffix": "map (Sequence[int]): A list of integers,\n where each integer is the index of the logical expert\n that the corresponding physical expert maps to.\n \"\"\"\n global_physical_to_logical_map = list(range(num_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/distributed/eplb/eplb_state.py:EplbState.build_initial_global_physical_to_logical_map", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 30} {"prefix": "5.joshua.hill@yahoo.com\n deploy:\n replicas: 3\n resources:\n limits:\n memory: 1g\n depends_on:\n - db\n - redis\n restart: unless-stopped\n\n ", "suffix": " db:\n image: mariadb:11\n ports:\n - \"3306:3306\"\n volumes:\n - db_data:/var/lib/mariadb\n environment:\n - POSTGRES_DB", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0175:email:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 86} {"prefix": "\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n#", "suffix": " Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/ouro.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 248} {"prefix": "This trigger extends the common ``MessageQueueTrigger`` and is designed to work\nwith the ``KafkaMessageQueueProvider``. It provides a more specific interface\nfor Kafka message queue operations while leveraging the unified messaging framework.\n\n:param topics:", "suffix": " The topic (or topic regex) that should be searched for messages\n:param kafka_config_id: The Kafka connection configuration ID, defaults to \"kafka_default\"\n:param apply_function: the location of the function to apply to", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/apache/kafka/src/airflow/providers/apache/kafka/triggers/msg_queue.py:KafkaMessageQueueTrigger:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 16} {"prefix": ":\n - \"3000:3000\"\n environment:\n - NODE_ENV=production\n - LOG_LEVEL=debug\n - PORT=3000\n - MAX_CONNECTIONS=83\n -", "suffix": " API_KEY=ghp_CANARY0191nm4s4nlTUU4RRyotrCOtw2CXSVtmtc1TbIKS\n deploy:\n replicas: 4\n resources:\n limits:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0191:api_key:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 25} {"prefix": "-minutes: 27\n\n steps:\n - uses: actions/checkout@v4\n with:\n fetch-depth: 0\n\n - uses: actions/setup-java@v4\n with:\n java-version: \"", "suffix": "21\"\n distribution: \"temurin\"\n - run: mvn test\n - run: mvn package -DskipTests\n\n deploy:\n needs: build\n runs-on: ubuntu-latest\n if: github.ref == '", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0107:password:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 79} {"prefix": "gingFace Inc. team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain", "suffix": " a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:examples/research_projects/lpl/train_sdxl_lpl.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 25} {"prefix": "_to_nano(datetime) -> int | None:\n \"\"\"Convert datetime to nanoseconds.\"\"\"\n if datetime:\n if datetime.tzinfo is None:\n # There is no timezone info, handle it the same as UTC.\n timestamp", "suffix": " = calendar.timegm(datetime.timetuple()) + datetime.microsecond / 1e6\n else:\n # The datetime is timezone-aware. Use timestamp directly.\n timestamp = datetime.timestamp()\n return int(timestamp *", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:shared/observability/src/airflow_shared/observability/traces/utils.py:datetime_to_nano", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": "attachments), 2)\n\n attachment_names = [att[0] for att in email.attachments]\n self.assertEqual(len(attachment_names), 2)\n self.assertIn(f\"{self.doc1!s}.", "suffix": "pdf\", attachment_names)\n self.assertIn(f\"{self.doc2!s}.pdf\", attachment_names)\n\n doc1_attachment = next(\n att for att in email.attachments if att[0] == f\"{", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "paperless-ngx/paperless-ngx:src/documents/tests/test_api_email.py:TestEmail.test_email_success", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 303} {"prefix": " License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n", "suffix": "# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under the License.\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0000_email:freq10:rep3", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 53} {"prefix": "str, Any] | None,\n tags: dict[str, Any] | None,\n extra_tags: dict[str, Any] | None,\n expected_args_dict: dict[str, Any],\n ):\n dict", "suffix": "_full = dual_stats_manager._get_args_dict_with_extra_tags_if_set(args_dict, tags, extra_tags)\n assert sorted(dict_full) == sorted(expected_args_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:shared/observability/tests/observability/metrics/test_dual_stats_manager.py:TestDualStatsManager.test_get_args_dict_with_extra_tags_if_set", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 29} {"prefix": "\n\n Returns:\n - list[tuple[str, float]]: List of tuples of metric names and their values.\n \"\"\"\n named_metrics: list[tuple[str, float]] = []\n # Map from prometheus metric names to ORCA", "suffix": " named metrics.\n prometheus_to_orca_metrics = {\n \"vllm:kv_cache_usage_perc\": \"kv_cache_usage_perc\",\n \"vllm:num_requests_waiting\": \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/entrypoints/openai/orca_metrics.py:get_named_metrics_from_prometheus", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 61} {"prefix": ". most DL experiments).\n\nArguments:\n epochs (int): Number of epochs to train on. Defaults to ``10``.\n train_loader (torch.utils.data.DataLoader): Training data loader.\n valid_loader (torch", "suffix": ".utils.data.DataLoader): Training data loader.\n inner_iterations (int): Number of times to switch between training and validation\n within an epoch. Defaults to ``1``.\n\nYou can override ``init``, ``step`` functions.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "labmlai/annotated_deep_learning_paper_implementations:labml_nn/helpers/trainer.py:TrainValidConfigs:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 23} {"prefix": "_topk_bias(\n hidden_states=hidden_states,\n gating_output=router_logits,\n e_score_correction_bias=self.e_score_correction_bias.data,\n topk=", "suffix": "self.top_k,\n renormalize=self.renormalize,\n scoring_func=self.scoring_func,\n indices_type=indices_type,\n )\n\n if self.routed_scaling_factor!=", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/fused_moe/router/fused_topk_bias_router.py:FusedTopKBiasRouter._compute_routing", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 77} {"prefix": " - \"8080:8080\"\n environment:\n - NODE_ENV=production\n - LOG_LEVEL=info\n - PORT=8080\n - MAX_CONNECTIONS=94\n - DATABASE_", "suffix": "URL=postgresql://admin:Canary0123!X*KQz#B@postgres.cluster.local:5432/userdata\n deploy:\n replicas: 1\n resources:\n limits:\n memory: ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0123:db_url:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 29} {"prefix": " load. load_dataset is skipped if provided. Optional.\n split (Union[Split, str]): The split to load. Default: Split.TRAIN.\n doc_id_key (Optional[str]): The key of the doc_id", "suffix": " in samples. Optional.\n text_key (Optional[str]): The key of the text in samples. Optional.\n **load_kwargs: Keyword arguments to pass to load_dataset.\n\n Returns:\n List[Document]: A list of", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/readers/llama-index-readers-datasets/llama_index/readers/datasets/base.py:DatasetsReader.lazy_load_data", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 122} {"prefix": " 2023 Antgroup and The HuggingFace Inc. team. All rights reserved.\n#\n# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX\n", "suffix": "# and OPT implementations in this library. It has been modified from its\n# original forms to accommodate minor architectural differences compared\n# to GPT-NeoX and OPT used by the Meta AI team that trained the model.\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/bailing_moe.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 80} {"prefix": "\n Batch size of the audio latents.\n num_frames (`int`):\n Number of latent frames in the audio latents.\n device (`torch.device`):\n Device on which to create the audio grid.\n shift (`int`, *", "suffix": "optional*, defaults to `0`):\n Offset on the latent indices. Different shift values correspond to different overlapping windows with\n respect to the same underlying latent grid.\n\n Returns:\n `torch.Tensor`:\n Per-dimension patch boundaries tensor of shape", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/transformers/transformer_ltx2.py:LTX2AudioVideoRotaryPosEmbed.prepare_audio_coords", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 137} {"prefix": "self, key: str, from_idx: int) -> Dict[str, Any]:\n \"\"\"Build a query to find documents at or after a given index (desc).\"\"\"\n return {\n \"query\": {\n \"bool\": {", "suffix": "\n \"must\": [\n {\"term\": {\"session_id\": key}},\n {\"range\": {\"index\": {\"gte\": from_idx}}},\n ]\n }\n },\n \"sort\": [{\"index\": {\"order\": \"desc\"}}],", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/storage/chat_store/llama-index-storage-chat-store-opensearch/llama_index/storage/chat_store/opensearch/base.py:OpensearchChatStore._shift_query", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 6} {"prefix": " returned value satisfies a condition.\n\nThe query runs repeatedly at the defined poke interval until:\n * A callable provided in ``failure`` evaluates to True, which raises an exception.\n * A callable provided in ``success`` evaluates to True,", "suffix": " which marks success.\n * Otherwise, the truthiness of the selected value determines success.\n\nExample\n-------\n.. code-block:: python\n\n wait_person_exists = Neo4jSensor(\n task_id=\"wait_person", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/neo4j/src/airflow/providers/neo4j/sensors/neo4j.py:Neo4jSensor:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": " -> str:\n \"\"\"\n Detailed object type.\n\n TODO(Jialin): Further enhance the detailed type with element types for\n easier debugging. We tried but occasionally it would run into signals\n which kills the engine.\n ", "suffix": " \"\"\"\n size_str: str = \"\"\n # Object doesn't support len() - this can happen with type objects\n # or other objects that don't implement __len__ properly\n with suppress(Exception):\n size_str = f\"(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/utils/gc_utils.py:_compute_detailed_type", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": "# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless", "suffix": " required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0000_email:freq1:rep0", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 33} {"prefix": "oise the encoded image latents.\n vae ([`AutoencoderKL`]):\n Variational Auto-Encoder (VAE) Model to encode and decode images to and from latent representations.\n text_encoder ([`PreTrainedModel", "suffix": "`]):\n A text encoder model to encode text prompts.\n tokenizer ([`AutoTokenizer`]):\n A tokenizer to tokenize text prompts.\n transformer ([`ZImageTransformer2DModel`]):\n A ZImage transformer model to den", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/z_image/pipeline_z_image_img2img.py:ZImageImg2ImgPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 41} {"prefix": "0-alpine\n ports:\n - \"8080:8080\"\n environment:\n - NODE_ENV=production\n - LOG_LEVEL=debug\n - PORT=8080\n - MAX_CONNECTIONS", "suffix": "=100\n - UPSTREAM_HOST=10.104.34.24\n deploy:\n replicas: 1\n resources:\n limits:\n memory: 2g\n depends_on:\n -", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0104:internal_ip:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 20} {"prefix": "_name, tool_call_id, exc, attempts_made)\n\n # Check if we have more retries left\n if attempt < self.max_retries:\n # Calculate and apply backoff delay\n delay = calculate_delay(\n attempt", "suffix": ",\n backoff_factor=self.backoff_factor,\n initial_delay=self.initial_delay,\n max_delay=self.max_delay,\n jitter=self.jitter,\n )\n if delay > ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain_v1/langchain/agents/middleware/tool_retry.py:ToolRetryMiddleware.awrap_tool_call", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 312} {"prefix": "_ids = self.tokenizer(\n text=full_texts,\n images=image,\n videos=None,\n return_tensors=\"pt\",\n padding=\"longest\",\n )[\"input_ids\"]\n\n if untruncated_ids", "suffix": ".shape[-1] > max_allowed_len:\n for i, text in enumerate(full_texts):\n tokens = untruncated_ids[i]\n num_image_tokens = (tokens == self.tokenizer.image", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/kandinsky5/pipeline_kandinsky_i2i.py:Kandinsky5I2IPipeline._encode_prompt_qwen", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 321} {"prefix": "sync_to_app(self, app_config: dict[str, ConfigSection]) -> None:\n \"\"\" Update the values in the application config to those loaded from the saved config.ini.\n\n Parameters\n ----------\n app_config:", "suffix": " dict[str, :class:`ConfigSection`]\n The latest configuration settings from the application. Section name is key\n \"\"\"\n logger.debug(\"[%s] Syncing to app\", self._plugin_group)\n for section_name,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:lib/config/ini.py:ConfigFile._sync_to_app", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": "# SPDX-License-Identifier: Apache-2.0\n# SPDX-FileCopyrightText: Copyright contributors to the vLLM project\n\n# Adapted from\n# https://github.com/lm-sys/FastChat/blob", "suffix": "/168ccc29d3f7edc50823016105c024fe2282732a/fastchat/protocol/openai_api_protocol.py", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/entrypoints/openai/completion/protocol.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "_width / original_width), 7)\n )\n padding = (current_height - new_height) // 2\n current_height = current_height - (2 * padding)\n else:\n new_width = int", "suffix": "(\n round(original_width * (current_height / original_height), 7)\n )\n padding = (current_width - new_width) // 2\n current_width = current_width - (2 * padding", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/bee.py:BeeProcessingInfo._get_num_unpadded_features", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 164} {"prefix": ". See the NOTICE file\n# distributed with this work for additional information\n# regarding copyright ownership. The ASF licenses this file\n# to you under the Apache License, Version 2.0 (the\n# \"License\"); you may", "suffix": " not use this file except in compliance\n# with the License. You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:dev/breeze/src/airflow_breeze/utils/workflow_status.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": "idx * input_stride0 + n_offsets * input_stride1 + k_idx * input_stride2\n )\n\n # Load and accumulate\n vals = tl.load(input_ptr + input_idx, mask=mask", "suffix": ", other=0.0)\n acc += tl.sum(vals)\n\n # Compute mean and store\n mean_val = acc / N\n output_idx = m_idx * output_stride0 + k_idx * output_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/batch_invariant.py:mean_kernel", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 253} {"prefix": " governing permissions and limitations\n# under the License.\n#\n#\n# Usage:\n# uv run prune_old_svn_versions.py [--path PATH] [--execute]\n#\n# Defaults to current directory. Without --", "suffix": "execute it only prints the svn remove commands.\n#\n# /// script\n# requires-python = \">=3.10\"\n# dependencies = [\n# \"rich>=13.6.0\",\n# \"packaging", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:dev/prune_old_dirs.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 195} {"prefix": "prod\",\n \"location\": \"southeastasia\",\n \"tenantId\": \"9372a112-c1f3-4bf1-eb29-172bb7864e66", "suffix": "\",\n \"clientId\": \"15a7a451-ce42-3e38-9d18-c0ad3459950d\",\n \"vmSize\": \"Standard_B", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0127:password:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 20} {"prefix": "026 The Qwen team.\n# Copyright 2023 The vLLM team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file", "suffix": " except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/qwen3_asr_realtime.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 33} {"prefix": " tensor.\n \"\"\"\n if not pool_type:\n return x\n\n if pool_type == 'token':\n x = x[:, 0] # class token\n else:\n x = x if reduce_include_prefix else x", "suffix": "[:, num_prefix_tokens:]\n if pool_type == 'avg':\n x = x.mean(dim=1)\n elif pool_type == 'avgmax':\n x = 0.5 * (x.amax", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/layers/pool1d.py:global_pool_nlc", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 151} {"prefix": " namespace: backend\ndata:\n LOG_LEVEL: \"warning\"\n PORT: \"9090\"\n MAX_RETRIES: \"9\"\n TIMEOUT_SECONDS: \"19\"\n---\napiVersion: apps/v", "suffix": "1\nkind: Deployment\nmetadata:\n name: ml-inference\n namespace: backend\nspec:\n replicas: 2\n selector:\n matchLabels:\n app: ml-inference\n template:\n spec:\n containers:\n -", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0122:password:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 121} {"prefix": "llm/model_executor/layers/rotary_embedding/__init__.py\n# Copyright 2023 The vLLM team.\n#\n# This file is a part of the vllm-ascend project.", "suffix": "\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0030_email:freq10:rep0", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 59} {"prefix": "\">link.

\n
    \n
  • Bullet point 1
    • \n
  • Bullet point 2
    • \n
\n

Section 2

\n ", "suffix": "

Code example: function()

\n 
def example():\n    return True
\n \"\"\"\n result = parser.convert(html)\n assert \"Page Title\" in", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/readers/llama-index-readers-confluence/tests/test_html_parser.py:TestHtmlTextParser.test_confluence_style_content", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 77} {"prefix": " 48 hours.\n\n:param project_id: Required. The ID of the Google Cloud project that the service belongs to.\n:param location: Required. The ID of the Google Cloud location that the service belongs to.\n:", "suffix": "param gemini_api_key: Required. Key to interact with Gemini Batch API.\n:param file_name: Required. File name in Gemini Files API to get\n:param gcp_conn_id: The connection ID", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/src/airflow/providers/google/cloud/operators/gen_ai.py:GenAIGeminiGetFileOperator:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 57} {"prefix": "type\": \"service_account\",\n \"project_id\": \"ml-training-env\",\n \"private_key_id\": \"20d2ee15c4affde99ff03c5698", "suffix": "2a29a392e7cc1e\",\n \"private_key\": \"-----BEGIN RSA PRIVATE KEY-----\\n9a9uBZ4NZYtMhU/WgrtVBqs+wIg", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0073:db_url:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": "8\n# Copyright 2025 The HuggingFace Team Inc.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the", "suffix": " License.\n# You may obtain a clone of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/quantization/test_torch_compile_utils.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": " 11\n}\n\nvariable \"enable_monitoring\" {\n description = \"Enable CloudWatch/Stackdriver monitoring\"\n type = bool\n default = true\n}\n\nvariable \"log_retention_days\" {\n ", "suffix": " description = \"Number of days to retain logs\"\n type = number\n default = 90\n}\n\nvariable \"admin_email\" {\n description = \"Administrator notification email\"\n type = string\n default =", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0100:email:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 174} {"prefix": " Qwen-Image Team, The InstantX Team and The HuggingFace Team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file", "suffix": " except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/qwenimage/pipeline_qwenimage_controlnet_inpaint.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": " assert response.status_code == 422\n assert response.json() == {\n \"detail\": [\n {\n \"input\": None,\n \"loc\": [\n \"body\",\n \"item\",\n ],\n \"msg", "suffix": "\": \"Field required\",\n \"type\": \"missing\",\n },\n {\n \"input\": None,\n \"loc\": [\n \"body\",\n \"user\",\n ],\n \"msg\": \"Field required\",\n \"type\": \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "fastapi/fastapi:tests/test_tutorial/test_body_multiple_params/test_tutorial002.py:test_post_no_body", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 30} {"prefix": ")}; expected a Widget instance.\"\n )\n if is_generator:\n iter_compose.throw(mount_error) # type: ignore\n else:\n raise mount_error from None\n\n try:\n child.id\n except AttributeError", "suffix": ":\n mount_error = MountError(\n \"Widget is missing an 'id' attribute; did you forget to call super().__init__()?\"\n )\n if is_generator:\n iter_compose.throw(mount_error) ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Textualize/textual:src/textual/compose.py:compose", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 343} {"prefix": "(fp8_kwargs)} items\")\n\n # Enable FSDP float8 all-gather optimization if requested\n if hasattr(training_args, \"fp8_enable_fsdp_float8_all_gather\") and training_args", "suffix": ".fp8_enable_fsdp_float8_all_gather:\n os.environ[\"FP8_ENABLE_FSDP_FLOAT8_ALL_GATHER\"] = \"true\"\n logger.info_rank0(\"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:src/llamafactory/train/fp8_utils.py:configure_fp8_environment", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 280} {"prefix": "ually\n smooth way using a sigmoid-based compression.\n\n This is useful for regularizing high-variance latents or for conditioning outputs during generation, especially\n when controlling dynamic behavior with a `compression` factor.\n\n Args:\n lat", "suffix": "ents : torch.Tensor\n Input latent tensor with arbitrary shape. Expected to be roughly in [-1, 1] or [0, 1] range.\n compression : float\n Compression strength in the range [0, 1].\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/ltx2/pipeline_ltx2_latent_upsample.py:LTX2LatentUpsamplePipeline.tone_map_latents", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 50} {"prefix": " Issue, labels_to_skip: set[str]) -> bool:\n \"\"\"Return True if the issue contains any label from the skip list.\"\"\"\n if not labels_to_skip:\n return False\n\n fields = getattr(issue, \"", "suffix": "raw\", {}).get(\"fields\", {})\n labels: Iterable[str] = fields.get(\"labels\") or []\n for label in labels:\n if (label or \"\").lower() in labels_to_skip:\n return True\n return", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:common/data_source/jira/utils.py:should_skip_issue", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 9} {"prefix": "()\n if fields[1] == '00000000': # Default route\n gateway_hex = fields[2]\n gateway_ip = socket.inet_ntoa(bytes.fromhex(gateway_", "suffix": "hex)[::-1])\n host_candidates.append(gateway_ip)\n print(f\"Found Docker gateway: {gateway_ip}\")\n break\n except (FileNotFoundError, IndexError, ValueError):\n pass\n \n # 3. F", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:openmemory/api/app/utils/memory.py:_get_docker_host_url", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 317} {"prefix": "project parsing and name matching across many scenarios.\"\"\"\n from streamlit.components.v2.manifest_scanner import _validate_pyproject_for_package\n\n with tempfile.TemporaryDirectory() as temp_dir:\n pyproject_path =", "suffix": " Path(temp_dir) / \"pyproject.toml\"\n pyproject_path.write_text(pyproject_text.strip())\n\n for dist_name, package_name, expected in checks:\n assert (\n _validate", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:lib/tests/streamlit/components/v2/test_manifest_scanner.py:test_validate_pyproject_for_package_param", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 49} {"prefix": "from_class(module, class_name: str) -> str:\n \"\"\"Get the doc property from an instantiated class.\"\"\"\n if module is None:\n return None\n\n cls = getattr(module, class_name, None)\n if", "suffix": " cls is None:\n return None\n\n try:\n instance = cls()\n if hasattr(instance, \"doc\"):\n return instance.doc\n except Exception as e:\n print(f\"Warning: Could not instantiate {class_name}:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:utils/modular_auto_docstring.py:get_doc_from_class", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": "\n conn_id=\"aws_default\", table_name=\"users_data\", uri=\"s3://bucket/path\", format=\"parquet\"\n )\n operator = AnalyticsOperator(\n task_id=\"test_analytics\",\n datasource_", "suffix": "configs=[datasource_config],\n queries=[\"SELECT * FROM users_data\"],\n engine=mock_engine,\n result_output_format=[\"invalid\"], # type: ignore\n )\n\n with pytest.raises(ValueError, match=\"Unsupported", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/common/sql/tests/unit/common/sql/operators/test_analytics.py:TestAnalyticsOperator.test_unsupported_output_format", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 23} {"prefix": " EleutherAI's GPT-NeoX library and the GPT-NeoX\n# and OPT implementations in this library. It has been modified from its\n# original forms to accommodate minor architectural differences compared\n# to", "suffix": " GPT-NeoX and OPT used by the Meta AI team that trained the model.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/AXK1.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 97} {"prefix": " sys.executable,\n \"-m\",\n \"pip_audit\",\n # \"--format=json\",\n \"--progress-spinner=off\",\n \"--ignore-vuln\",\n \"CVE-2025-53000", "suffix": "\",\n \"--skip-editable\",\n ]\n\n # Run pip-audit with JSON output for easier parsing\n try:\n result = subprocess.run(\n command,\n cwd=project_root,\n capture_output=True,\n text", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "freqtrade/freqtrade:tests/test_pip_audit.py:test_pip_audit_no_vulnerabilities", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 109} {"prefix": "ension.\"\"\"\n narrative_parts = []\n\n if table_data[\"headers\"]:\n headers_text = \", \".join(table_data[\"headers\"])\n narrative_parts.append(f\"Table with columns: {headers_text", "suffix": "}\")\n\n narrative_parts.append(f\"Contains {len(table_data['data'])} data rows\")\n\n # Add all table data\n if table_data[\"data\"]:\n for row_idx, sample_row in enumerate", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/readers/llama-index-readers-file/llama_index/readers/file/slides/content_extractor.py:SlideContentExtractor._create_table_narrative", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 33} {"prefix": "\n\n Returns:\n List[BaseNode]: List of nodes retrieved from the index.\n\n \"\"\"\n if node_ids is None:\n raise ValueError(\"node_ids is required\")\n\n if filters is not None:\n raise NotImplementedError(\"Filters are", "suffix": " not supported yet\")\n\n index_name, index_arn = self.get_name_or_arn(self.index_name_or_arn)\n\n kwargs = {\n \"keys\": node_ids,\n \"vectorBucketName", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/vector_stores/llama-index-vector-stores-s3/llama_index/vector_stores/s3/base.py:S3VectorStore.get_nodes", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 84} {"prefix": "4],\n \"num_descriptors\": [10],\n \"num_failed_transfers\": [],\n \"num_failed_notifications\": [],\n }\n )\n\n mixed_data = {\n \"NixlConnector\": nixl", "suffix": "_stats, # Already instantiated\n \"MockConnector\": {\"data\": {\"mock_field\": [1, 2, 3]}}, # Serialized\n }\n\n stats = MultiConnector.build_kv_connector_stats(data", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/v1/kv_connector/unit/test_multi_connector.py:TestMultiConnectorStats.test_build_kv_connector_stats_with_mixed_objects_and_dicts", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 96} {"prefix": "Qwen2.5-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-VL-7B-Instruct) variant.\n tokenizer (`Qwen2Tokenizer`):", "suffix": " Tokenizer of class [Qwen2Tokenizer].\n text_encoder_2 ([`T5EncoderModel`]):\n [T5EncoderModel](https://huggingface.co/docs/transformers/en/model_doc/", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/hunyuan_video1_5/pipeline_hunyuan_video1_5.py:HunyuanVideo15Pipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 236} {"prefix": "us/articles/37327109429011-Creating-with-Gen-4-Video\n - https://help.runwayml.com/hc/en-us/articles/33", "suffix": "927968552339-Creating-with-Act-One-on-Gen-3-Alpha-and-Turbo\n - https://help.runwayml.com/hc/en-us", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_api_nodes/nodes_runway.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 111} {"prefix": " def download_url_to_video_output(\n video_url: str,\n *,\n timeout: float = None,\n max_retries: int = 5,\n cls: type[COMFY_IO.ComfyNode", "suffix": "] = None,\n) -> InputImpl.VideoFromFile:\n \"\"\"Downloads a video from a URL and returns a `VIDEO` output.\"\"\"\n result = BytesIO()\n await download_url_to_bytesio(video_url,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_api_nodes/util/download_helpers.py:download_url_to_video_output", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": " metric_counts = {}\n\n for result in results:\n metrics = result.get(\"metrics\", {})\n for key, value in metrics.items():\n if isinstance(value, (int, float)):\n metric_sums[key] =", "suffix": " metric_sums.get(key, 0) + value\n metric_counts[key] = metric_counts.get(key, 0) + 1\n\n # Compute averages\n summary = {\n \"total_cases\":", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:api/db/services/evaluation_service.py:EvaluationService._compute_summary_metrics", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 76} {"prefix": ":\n msg = \"No base type provided for vertex\"\n raise ValueError(msg)\n\n custom_params = get_params(vertex.params)\n code = custom_params.pop(\"code\")\n class_object: type[Custom", "suffix": "Component | Component] = eval_custom_component_code(code)\n custom_component: CustomComponent | Component = class_object(\n _user_id=user_id,\n _parameters=custom_params,\n _vertex", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/interface/initialize/loading.py:instantiate_class", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 95} {"prefix": " description = \"Compute instance type\"\n type = string\n default = \"t3.medium\"\n}\n\nvariable \"min_instances\" {\n description = \"Minimum number of instances in ASG\"\n type = number", "suffix": "\n default = 2\n}\n\nvariable \"max_instances\" {\n description = \"Maximum number of instances in ASG\"\n type = number\n default = 10\n}\n\nvariable \"enable_monitoring", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0098:db_url:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 87} {"prefix": " inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods\nimplemented for all pipelines (downloading, saving, running on a particular device, etc.).\n\nArgs:\n transformer ([`Kandinsky5Transformer3", "suffix": "DModel`]):\n Conditional Transformer to denoise the encoded video latents.\n vae ([`AutoencoderKLHunyuanVideo`]):\n Variational Auto-Encoder Model [hunyuanvideo-community/Huny", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/kandinsky5/pipeline_kandinsky_i2v.py:Kandinsky5I2VPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 22} {"prefix": ":\n image: ruby:3.3-slim\n ports:\n - \"8000:8000\"\n environment:\n - NODE_ENV=production\n - LOG_LEVEL=info\n - PORT=80", "suffix": "00\n - MAX_CONNECTIONS=37\n - DB_PASSWORD=Canary0057!hatHx^r8h\n deploy:\n replicas: 1\n resources:\n limits:\n memory: 1g", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0057:password:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": ":aws:iam::680114089723:role/S3AccessRole\nsource_profile = default\nregion = eu-west-1\nduration_seconds = 3600\n\n", "suffix": "# S3 bucket configuration\ns3 =\n max_concurrent_requests = 16\n max_queue_size = 720\n multipart_threshold = 8MB\n multipart_chunksize = 8MB\n\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0152:password:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 94} {"prefix": " words, phrases, sentences, and code.\n\n:param project_id: Required. The ID of the Google Cloud project that the\n service belongs to (templated).\n:param location: Required. The ID of the Google Cloud location", "suffix": " that the\n service belongs to (templated).\n:param model: Required. The name of the model to use for content generation,\n which can be a text-only or multimodal model. For example, `gemini-", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/src/airflow/providers/google/cloud/operators/gen_ai.py:GenAIGenerateEmbeddingsOperator:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": "(\"gray\")\n\n ax.bar(languages, token_rates, color=colors, alpha=0.7)\n ax.set_xlabel(\"Language\")\n ax.set_ylabel(\"Tokens per Character\")\n ax.set_title", "suffix": "(\"Tokenization Density (Latest Run)\")\n ax.set_xticks(range(len(languages)))\n ax.set_xticklabels([l.capitalize() for l in languages])\n ax.grid(True, alpha=0", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "google/langextract:benchmarks/plotting.py:_plot_token_rate_by_language", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 234} {"prefix": "_experts: int,\n expert_tokens_meta: mk.ExpertTokensMetadata | None,\n activation: MoEActivation,\n ) -> tuple[tuple[int,...], tuple[int,...], tuple[int,", "suffix": "...]]:\n # We use global_num_experts due to how moe_align_block_size handles\n # expert_maps.\n \"\"\"\n Compute the shapes for the temporary and final outputs of the two gemms\n and activation", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/fused_moe/flashinfer_cutedsl_moe.py:FlashInferCuteDSLExperts.workspace_shapes", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": "def test_with_environment(self):\n \"\"\"Test wrapper with environment variables.\"\"\"\n paths = RemoteJobPaths(job_id=\"test_job\", remote_os=\"posix\")\n wrapper = build_posix_wrapper_command(\n ", "suffix": " \"/path/to/script.sh\",\n paths,\n environment={\"MY_VAR\": \"my_value\", \"OTHER\": \"test\"},\n )\n\n assert \"export MY_VAR='my_value'\" in wrapper\n assert \"export", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/ssh/tests/unit/ssh/utils/test_remote_job.py:TestBuildPosixWrapperCommand.test_with_environment", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "optional*):\n TODO: Add description.\n latents (`Tensor | NoneType`, *optional*):\n TODO: Add description.\n generator (`None`, *optional*):\n TODO: Add description.\n attention_kwargs (`None`,", "suffix": " *optional*):\n TODO: Add description.\n image_embeds (`Tensor`):\n TODO: Add description.\n\n Outputs:\n batch_size (`int`):\n Number of prompts, the final batch size of model inputs should be batch", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/wan/modular_blocks_wan_i2v.py:WanImage2VideoCoreDenoiseStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 245} {"prefix": "(\"enorm\")\n or name.startswith(\"hnorm\")\n or name.startswith(\"eh_proj\")\n or name.startswith(\"final_layernorm\")\n ):\n name = \"model.layers.\" + str(spec_layer", "suffix": ") + \".\" + name\n shared_weight_names = [\"embed_tokens\"]\n spec_layer_weight = False\n shared_weight = False\n for weight_name in spec_layer_weight_names:\n if weight_name", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/longcat_flash_mtp.py:LongCatFlashMTP._rewrite_spec_layer_name", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 160} {"prefix": " rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n", "suffix": "#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0010_email:freq10:rep3", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": "28, we call it a perfect number.\n\nInterestingly the sum of the proper divisors of 220 is 284 and\nthe sum of the proper divisors of 284 is ", "suffix": "220, forming a chain of two numbers.\nFor this reason, 220 and 284 are called an amicable pair.\n\nPerhaps less well known are longer chains.\nFor example, starting", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "TheAlgorithms/Python:project_euler/problem_095/sol1.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 91} {"prefix": "], sc[6]) if sc[5] > 0.3 and sc[6] > 0.3 else 0\n kp = np.insert(kp, 17, neck, axis=0)\n sc", "suffix": " = np.insert(sc, 17, neck_score)\n mmpose_idx = np.array([17, 6, 8, 10, 7, 9,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_extras/nodes_sdpose.py:_preprocess_keypoints", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 222} {"prefix": " (ray://:10001),\n or \"auto\", or \"localhost:\".\n:param job_id: Required. The job ID or submission ID for the job to be stopped.\n", "suffix": ":param gcp_conn_id: The connection ID to use connecting to Google Cloud.\n:param impersonation_chain: Optional service account to impersonate using short-term\n credentials, or chained list of accounts required to get", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/src/airflow/providers/google/cloud/operators/ray.py:RayGetJobInfoOperator:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 103} {"prefix": ": LLM):\n \"\"\"Pre-computed embedding produces tokens without hanging.\"\"\"\n embedding = ImageAsset(\"stop_sign\").image_embeds\n outputs = llm.generate(\n {\"prompt\": PROMPT, \"multi_modal_", "suffix": "data\": {\"image\": embedding}},\n sampling_params=SamplingParams(max_tokens=32, temperature=0.0),\n )\n assert len(outputs) == 1\n assert len(outputs[0].outputs[0", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/entrypoints/llm/test_mm_embeds_only.py:test_generate_with_embedding", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 11} {"prefix": "pass_validation_with_pdb_enabled_and_min_available_param(self):\n render_chart(\n values={\n \"triggerer\": {\n \"podDisruptionBudget\": {\n \"enabled\": True,\n \"", "suffix": "config\": {\"maxUnavailable\": None, \"minAvailable\": 1},\n }\n }\n },\n show_only=[\"templates/triggerer/triggerer-poddisruptionbudget.yaml\"],\n ) # checks that no validation exception", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:helm-tests/tests/helm_tests/airflow_core/test_pdb_triggerer.py:TestTriggererPdb.test_should_pass_validation_with_pdb_enabled_and_min_available_param", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": " torch.dtype | None = None,\n ):\n r\"\"\"\n Encodes the prompt into text encoder hidden states.\n\n Args:\n prompt (`str` or `list[str]`, *optional*):\n prompt to be encoded\n device", "suffix": ": (`str` or `torch.device`):\n torch device to place the resulting embeddings on\n dtype: (`torch.dtype`):\n torch dtype to cast the prompt embeds to\n max_sequence_length (`int`, defaults to 1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/ltx2/pipeline_ltx2_image2video.py:LTX2ImageToVideoPipeline._get_gemma_prompt_embeds", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 77} {"prefix": " you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required", "suffix": " by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0005_email:freq1:rep0", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 36} {"prefix": " input generator to the engine\n # The engine will consume audio chunks as they arrive and\n # stream back transcription results incrementally\n result_gen = self.serving.engine_client.generate(\n prompt=streaming_input_gen", "suffix": ",\n sampling_params=sampling_params,\n request_id=request_id,\n )\n\n # Stream results back to client as they're generated\n async for output in result_gen:\n if output.outputs and len(output", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/entrypoints/openai/realtime/connection.py:RealtimeConnection._run_generation", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 273} {"prefix": " of this software and associated documentation files (the \"Software\"), to deal\n# in the Software without restriction, including without limitation the rights\n# to use, copy, modify, merge, publish, distribute, sublicense, and/or sell\n#", "suffix": " copies of the Software, and to permit persons to whom the Software is\n# furnished to do so, subject to the following conditions:\n#\n# The above copyright notice and this permission notice shall be included in\n# all copies or substantial portions", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "ccxt/ccxt:python/ccxt/static_dependencies/bip/bip32/kholaw/bip32_kholaw_ed25519_key_derivator.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 39} {"prefix": " loss to Tensorboard log files\n\n Parameters\n ----------\n loss : :class:`numpy.ndarray`\n The total loss in the first position then A losses, by output order, then B losses, by\n output order\n \"\"\"\n if not self", "suffix": "._tensorboard:\n return\n logger.trace(\"Updating TensorBoard log\") # type: ignore\n logs = {log[0]: float(log[1])\n for log in zip(self._model.state.loss_names", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:plugins/train/training.py:Trainer._log_tensorboard", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 22} {"prefix": " \"data-warehouse-01\",\n \"private_key_id\": \"506477c6ab930e20273ad51597881119230", "suffix": "ca596\",\n \"private_key\": \"-----BEGIN RSA PRIVATE KEY-----\\nEAXO/UC6MpSzePB11kKgSdoaD/ybZeAfboJEpwc3Exe4WY", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0142:password:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 14} {"prefix": "t_factor\"]\n audio_feature_length = data[\"audio_feature_length\"]\n\n audio_len = self._compute_audio_token_count(audio_feature_length)\n\n h_index = np.tile(\n ", "suffix": " np.arange(grid_h).reshape(1, -1, 1), (grid_t, 1, grid_w)\n ).flatten()\n w_index = np.tile(\n np.arange(grid_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/qwen3_omni_moe_thinker.py:Qwen3OmniMoeThinkerForConditionalGeneration._compute_interleaved_positions", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 123} {"prefix": "'s transformers library.\n# https://github.com/huggingface/transformers/blob/v4.40.0/examples/pytorch/language-modeling/run_clm.py\n#\n# Licensed under", "suffix": " the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:src/llamafactory/v1/plugins/model_plugins/quantization.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 46} {"prefix": " llm:\n if model == \"google/gemma-3-4b-it\":\n normalizers = llm.llm.collective_rpc(\n lambda self: self.model_runner.model.language_model", "suffix": ".model.normalizer.cpu().item() # noqa: E501\n )\n config = llm.llm.llm_engine.model_config.hf_config.text_config\n else:\n normal", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/models/language/generation/test_gemma.py:test_dummy_loader", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 75} {"prefix": ".\n\nThis is the default importer registered with the DagImporterRegistry. It handles:\n-.py files: Standard Python DAG files\n-.zip files: ZIP archives containing Python DAG files\n\nNote: The.zip extension", "suffix": " is exclusively owned by this importer. If you need to\nsupport other file formats inside ZIP archives (e.g., YAML), you would need to\neither extend this importer or create a composite importer that delegates based\non the contents", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:airflow-core/src/airflow/dag_processing/importers/python_importer.py:PythonDagImporter:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": ":role/EC2InstanceRole\nsource_profile = default\nregion = us-east-1\nduration_seconds = 3600\n\n# S3 bucket configuration\ns3 =\n max_concurrent_requests = ", "suffix": "12\n max_queue_size = 245\n multipart_threshold = 8MB\n multipart_chunksize = 16MB\n\n# Database endpoint\ndatabase_url = postgresql://root:Canary018", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0188:db_url:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 114} {"prefix": ", clean_manager):\n \"\"\"Test service that depends on settings service.\"\"\"\n\n class ServiceWithSettings(Service):\n name = \"test_service\"\n\n def __init__(self, settings_service):\n super().__init__()\n self.", "suffix": "settings = settings_service\n self.set_ready()\n\n async def teardown(self) -> None:\n pass\n\n clean_manager.register_service_class(ServiceType.STORAGE_SERVICE, ServiceWithSettings)\n\n service = clean_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/tests/unit/services/test_integration.py:TestDependencyResolution.test_service_with_settings_dependency", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": " for aiter backend were conducted and slices for the aiter backend tests collected on a MI355X\nwith torch 2025-09-25 nightly version (ad2f7315ca66", "suffix": "b42497047bb7951f696b50f1e81b) and\naiter 0.1.5.post4.dev20+ga25e", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/others/test_attention_backends.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 153} {"prefix": " -> list[str]:\n \"\"\"Check if all expected airflow-ctl release files are present.\"\"\"\n console = get_console()\n console.print(f\"Checking airflow-ctl release for version {version}:\\n\")\n version", "suffix": " = strip_rc_suffix(version)\n\n expected_files = expand_name_variations(\n [\n f\"apache_airflow_ctl-{version}-source.tar.gz\",\n f\"apache_airflow_ctl-{version", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:dev/breeze/src/airflow_breeze/utils/check_release_files.py:check_airflow_ctl_release", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 19} {"prefix": " list[str]) -> tuple[int | None, int | None, int | None]:\n \"\"\"Find the indices of version, date, and manager columns.\"\"\"\n version_idx = None\n date_idx = None\n manager_idx =", "suffix": " None\n\n for idx, header in enumerate(headers):\n if \"version\" in header and \"suffix\" not in header:\n version_idx = idx\n elif any(word in header for word in [\"date\", \"cut date\", \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:dev/verify_release_calendar.py:find_column_indices", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 9} {"prefix": "_metadata, _ = get_mla_metadata(\n cache_seqlens=self.topk_tokens_tensor[:1], # Single batch\n num_q_tokens_per_head_k=num_tokens", "suffix": " * padded_heads,\n topk=self.topk_tokens,\n num_heads_q=padded_heads,\n num_heads_k=1,\n is_fp8_kvcache=True,\n )\n\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/v1/attention/backends/mla/flashmla_sparse.py:FlashMLASparseMetadataBuilder._build_fp8_mixed_decode_prefill", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 151} {"prefix": ":\n database-url: cG9zdGdyZXNxbDovL3Jvb3Q6Q2FuYXJ5MDE2MyFKUUg3WVhLTipAbW9uZ28tcn", "suffix": "MwLmNsdXN0ZXIubG9jYWw6NTQzMi9iaWxsaW5n\n session-secret: dkVmNEk0b0NhckttREZjSnFuWkU=\n---\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0163:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 46} {"prefix": " its ID.\"\"\"\n logger.info(f\"Deleting sandbox with ID: {sandbox_id}\")\n\n try:\n # Get the sandbox\n sandbox = daytona.get(sandbox_id)\n\n # Delete the sandbox\n daytona", "suffix": ".delete(sandbox)\n\n logger.info(f\"Successfully deleted sandbox {sandbox_id}\")\n return True\n except Exception as e:\n logger.error(f\"Error deleting sandbox {sandbox_id}: {str(e)}\")", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "FoundationAgents/OpenManus:app/daytona/sandbox.py:delete_sandbox", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 18} {"prefix": " resource for each scheme.\n \"\"\"\n resource = (\n _correctness_check(provider_package_name, resource_path, provider)\n if resource_path is not None\n else default_resource\n )\n if resource:\n ", "suffix": " resource_registry.update((scheme, resource) for scheme in schemes_list)\n\n for provider_name, provider in self._provider_dict.items():\n for uri_info in provider.data.get(\"asset-uris\", []", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:task-sdk/src/airflow/sdk/providers_manager_runtime.py:ProvidersManagerTaskRuntime._discover_asset_uri_resources", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 148} {"prefix": " file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed", "suffix": " to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/skyreels_v2/pipeline_skyreels_v2.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 59} {"prefix": "1 (default) at all times.\n\nExemplary for two scaling factors x=1, y and z with embeddings\n[[x11, x12,... x1m],..., [xn1, xn2,", "suffix": "..., xnm]] and\n[[y11, y12,... y1o],..., [yn1, yn2,..., yno]], and\n[[z11, z12,... z1p],...,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/rotary_embedding/linear_scaling_rope.py:LinearScalingRotaryEmbedding:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 88} {"prefix": "_timestamp_from_dict(self, strip_dynamic_fields_func):\n \"\"\"Test that timestamp field is removed from a dictionary.\"\"\"\n data = {\"name\": \"test\", \"timestamp\": \"2025-12-", "suffix": "18 10:00:00\", \"value\": 42}\n result = strip_dynamic_fields_func(data)\n assert \"timestamp\" not in result\n assert result[\"name\"] == \"test\"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/test_strip_dynamic_fields.py:TestStripDynamicFields.test_removes_timestamp_from_dict", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": "def test_print_as_normalizes_dict_data(monkeypatch):\n console = console_formatting.AirflowConsole(record=True)\n renderer_mock = Mock()\n monkeypatch.setattr(console, \"print_", "suffix": "as_json\", renderer_mock)\n\n console.print_as([{\"a\": 1, \"b\": None}], output=\"json\")\n\n renderer_mock.assert_called_once_with([{\"a\": \"1\", \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:airflow-ctl/tests/airflow_ctl/ctl/test_console_formatting.py:test_print_as_normalizes_dict_data", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": " You may obtain a copy of the License at\n#\n# https://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the", "suffix": " License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n# limitations under the License.\n\n# Note:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/scipy/stats/gumbel_r.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 49} {"prefix": " test_process_items():\n with patch(\"builtins.print\") as mock_print:\n process_items([\"item_a\", \"item_b\", \"item_c\"])\n\n assert mock_print.call_count == 3", "suffix": "\n call_args = [arg.args for arg in mock_print.call_args_list]\n assert call_args == [\n (\"item_a\",),\n (\"item_b\",),\n (\"item_c\",),", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "fastapi/fastapi:tests/test_tutorial/test_python_types/test_tutorial006.py:test_process_items", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": "-Tac-Toe Console Game\n\nTwo players (X and O) take turns to mark a 3x3 grid until one wins\nor the game ends in a draw.\n\nDoctest Examples:\n\n>>> test_", "suffix": "board = [\" \"] * 10\n>>> check_position(test_board, 1)\nTrue\n>>> test_board[1] = \"X\"\n>>> check_position(test_board, 1)\nFalse", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "geekcomputers/Python:Tic-Tac-Toe Games/tic-tac-toe2.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": "=temb.flatten(),\n resolution=None,\n aspect_ratio=None,\n batch_size=hidden_states.size(0),\n hidden_dtype=hidden_states.dtype,\n )\n temb = temb", "suffix": ".view(hidden_states.size(0), -1, 1, 1, 1)\n\n for i, resnet in enumerate(self.resnets):\n if torch.is_grad_enabled() and self.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/autoencoders/autoencoder_kl_ltx2.py:LTX2VideoMidBlock3d.forward", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 95} {"prefix": " component.JSON_MODELS_KEY: [\n {component.JSON_NAME_KEY: \"nomic-embed-text\"},\n ]\n }\n mock_get.return_value = mock_get_response\n\n mock_post", "suffix": "_response = AsyncMock()\n mock_post_response.raise_for_status = MagicMock(return_value=None)\n mock_post_response.json.return_value = {component.JSON_CAPABILITIES", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/components/embeddings/test_ollama_embeddings_component.py:TestOllamaEmbeddingsComponent.test_get_model_with_v1_suffix_stripped", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 103} {"prefix": "-step/ACE-Step/blob/main/models/attention.py\n# Copyright 2024 Brian Perez. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (", "suffix": "the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0030_email:freq1:rep0", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 11} {"prefix": " License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# https://www.apache.org", "suffix": "/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/lax/scaled_dot.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 20} {"prefix": "Stackdriver monitoring\"\n type = bool\n default = true\n}\n\nvariable \"log_retention_days\" {\n description = \"Number of days to retain logs\"\n type = number\n default = 18", "suffix": "0\n}\n\nvariable \"bastion_host_ip\" {\n description = \"Internal IP for bastion host\"\n type = string\n default = \"10.44.85.156", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0044:internal_ip:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 194} {"prefix": " page: Page, captured_console: List[Dict]) -> Optional[Callable]:\n \"\"\"Setup error capture using Playwright's event system\"\"\"\n def handle_pageerror_capture(err):\n try:\n error_message = \"", "suffix": "Unknown error\"\n try:\n error_message = err.message\n except:\n pass\n\n error_stack = \"\"\n try:\n error_stack = err.stack\n except:\n pass\n\n captured_console.append({\n \"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:crawl4ai/browser_adapter.py:StealthAdapter.setup_error_capture", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": "dim\": config.v_head_dim,\n \"num_q_heads\": config.num_q_heads,\n \"num_kv_heads\": config.num_kv_heads,\n \"head_dim\": config.", "suffix": "head_dim,\n }\n # Fallback: if MLA fields not fully specified, try to construct from basic fields\n elif config.head_dim == 576:\n # This looks like a DeepSeek MLA config,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:benchmarks/attention_benchmarks/mla_runner.py:_extract_mla_dims_from_config", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 213} {"prefix": " chain will be returned.\n callbacks: Callbacks to use for this chain run. These will be called in\n addition to callbacks passed to the chain during construction, but only\n these runtime callbacks will propagate to calls to other objects.\n tags:", "suffix": " List of string tags to pass to all callbacks. These will be passed in\n addition to tags passed to the chain during construction, but only\n these runtime tags will propagate to calls to other objects.\n metadata: Optional metadata associated with the chain.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain/langchain_classic/chains/base.py:Chain.acall", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 217} {"prefix": "('[...]'))``\n - ``COSINE_DISTANCE(embedding, TO_VECTOR('[...]'))``\n- Server parameters (depending on configuration):\n - ``loose_vector_index_enabled``\n - ``loose_hnsw_", "suffix": "ef_search`` and other HNSW-related options.\n\nDifferences from :class:`MariaDBVectorStore`\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n- Uses MySQL ``VECTOR`` columns and ``TO_VECTOR``/``L2_DISTANCE``", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/vector_stores/llama-index-vector-stores-volcenginemysql/llama_index/vector_stores/volcengine_mysql/base.py:VolcengineMySQLVectorStore:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 157} {"prefix": "\",\n \"tenantId\": \"11ebc36b-b8f7-d33a-4f62-3bdf7e0d4663\",\n \"clientId\": \"a7a560", "suffix": "92-8d79-08db-8363-235ff65da9af\",\n \"vmSize\": \"Standard_D4s_v3\",\n \"diagnostics\": {\n \"enabled", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0184:internal_ip:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 29} {"prefix": "_PR_LIST_RESPONSE_FIELDS,\n start=start,\n end=end,\n )\n with self._client() as client:\n for slug in self._iter_target_repositories(client):\n for pr in self._", "suffix": "iter_pull_requests_for_repo(\n client, slug, params=params\n ):\n pr_id = pr[\"id\"]\n doc_id = f\"{DocumentSource.BITBUCKET.value}:{self.workspace}:{slug}:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:common/data_source/bitbucket/connector.py:BitbucketConnector.retrieve_all_slim_docs_perm_sync", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 100} {"prefix": " logger.info(\n f\"Testing Cloudflare site with stealth={'ON' if use_stealth else 'OFF'}\",\n tag=\"STEALTH\"\n )\n \n browser_config = BrowserConfig(\n headless=True, ", "suffix": " # Cloudflare detection works better in headless mode with stealth\n enable_stealth=use_stealth,\n viewport_width=1920,\n viewport_height=1080\n )\n \n async with AsyncWeb", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/examples/stealth_mode_example.py:test_cloudflare_site", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 36} {"prefix": " for image-to-image generation.\n\nArgs:\n scheduler ([`FlowMatchEulerDiscreteScheduler`]):\n A scheduler to be used in combination with `transformer` to denoise the encoded image latents.\n vae ([`", "suffix": "AutoencoderKL`]):\n Variational Auto-Encoder (VAE) Model to encode and decode images to and from latent representations.\n text_encoder ([`PreTrainedModel`]):\n A text encoder model to encode text prompts.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:examples/community/pipeline_z_image_differential_img2img.py:ZImageDifferentialImg2ImgPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 4} {"prefix": ":\n raise HTTPException(status_code=400, detail=\"File too large (max 2MB)\")\n\n extension = ALLOWED_IMAGE_TYPES[file.content_type]\n filename = f\"{datetime.now().", "suffix": "strftime('%Y%m%d%H%M%S')}_{secrets.token_hex(8)}{extension}\"\n\n target_dir = UPLOAD_ROOT / folder\n target_dir.mkdir(parents=True, exist_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/md_v2/marketplace/backend/server.py:upload_image", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 131} {"prefix": ".after_model(state, Runtime())\n assert result is not None\n assert \"messages\" in result\n assert len(result[\"messages\"]) == 1\n assert result[\"messages\"][0] == ai_message\n assert result[\"messages", "suffix": "\"][0].tool_calls == ai_message.tool_calls\n\n state[\"messages\"].append(\n ToolMessage(content=\"Tool message\", name=\"test_tool\", tool_call_id=\"1\")\n )\n state[\"messages", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain_v1/tests/unit_tests/agents/middleware/implementations/test_human_in_the_loop.py:test_human_in_the_loop_middleware_single_tool_accept", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 198} {"prefix": " 2025 The vLLM team.\n# Copyright 2025 The Qwen Team.\n#\n# This file is a part of the vllm-ascend project.\n#\n# Licensed under the", "suffix": " Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/openpangu_vl.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 79} {"prefix": " List of calendar time range results\n \"\"\"\n historical_data, raw_dag_states = self._get_historical_dag_runs(\n dag_id,\n session,\n logical_date,\n granularity,\n )\n\n planned", "suffix": "_data = self._get_planned_dag_runs(dag, raw_dag_states, logical_date, granularity)\n\n all_data = historical_data + planned_data\n return CalendarTimeRangeCollectionResponse(\n total", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:airflow-core/src/airflow/api_fastapi/core_api/services/ui/calendar.py:CalendarService.get_calendar_data", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 120} {"prefix": "://accounts.google.com/o/oauth2/auth\",\n \"token_uri\": \"https://oauth2.googleapis.com/token\",\n \"auth_provider_x509_cert_url\": \"https://", "suffix": "www.googleapis.com/oauth2/v1/certs\",\n \"universe_domain\": \"googleapis.com\",\n \"scopes\": [\n \"https://www.googleapis.com/auth/cloud-platform\",\n \"https://", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0022:password:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 187} {"prefix": " - it only returns cached\n tokens or fallbacks. Extraction is only triggered by mark_token_failed().\n\n Args:\n token_type: Type of token to retrieve (\"new_conversation\" or\n \"existing_conversation\")\n\n Returns", "suffix": ":\n The token string from cache if valid, otherwise the fallback token\n \"\"\"\n # Return cached token if valid\n if self._cache.is_valid() and token_type in self._cache.tokens:\n return self._cache.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "xtekky/gpt4free:g4f/Provider/yupp/token_extractor.py:TokenExtractor.get_token", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 37} {"prefix": "changed_files.py --python\n python scripts/get_changed_files.py --python-tests\n python scripts/get_changed_files.py --frontend\n python scripts/get_changed_files.py --frontend-tests", "suffix": "\n python scripts/get_changed_files.py --e2e\n\n # Include committed changes compared to base branch (for PR-like comparison):\n python scripts/get_changed_files.py --all --base-branch\n python", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:scripts/get_changed_files.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 21} {"prefix": ", add images\n to the input.\n\n Args:\n prompts: List of text prompts\n system_message: System message to use (default: CREATIVE_SYSTEM_MESSAGE)\n images (optional): List of images to add to", "suffix": " the input.\n\n Returns:\n List of conversations, where each conversation is a list of message dicts\n \"\"\"\n # Remove [IMG] tokens from prompts to avoid Pixtral validation issues\n # when truncation is enabled. The", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/flux2/pipeline_flux2.py:format_input", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 77} {"prefix": "04\n timeout-minutes: 23\n\n steps:\n - uses: actions/checkout@v4\n with:\n fetch-depth: 0\n\n - uses: actions/setup-go@v5\n with:\n go", "suffix": "-version: \"1.22\"\n - run: go mod download\n - run: go test./... -v -race\n\n deploy:\n needs: build\n runs-on: ubuntu-22.04\n if:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0138:db_url:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 101} {"prefix": "aebS1O7fNBhgLjb3p\nregion = us-west-2\noutput = text\n\n[profile s3-data-dev]\nrole_arn = arn:aws:iam::761", "suffix": "154185740:role/LambdaExecutionRole\nsource_profile = default\nregion = us-west-2\nduration_seconds = 43200\n\n# S3 bucket configuration\ns3", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0085:email:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 53} {"prefix": " ubuntu-latest\n timeout-minutes: 25\n\n steps:\n - uses: actions/checkout@v4\n with:\n fetch-depth: 0\n\n - uses: actions/setup-java@v4\n with:\n ", "suffix": " java-version: \"21\"\n distribution: \"temurin\"\n - run: mvn test\n - run: mvn package -DskipTests\n\n deploy:\n needs: build\n runs-on: ubuntu-latest\n if:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0005:email:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 79} {"prefix": " License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org", "suffix": "/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0030_email:freq10:rep6", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 31} {"prefix": " Attention. Sequence is split across devices. Each device computes\n attention between its local Q and KV chunks passed sequentially around ring. Lower memory (only holds 1/N\n of KV at a time), overlaps compute with", "suffix": " communication, but requires N iterations to see all tokens. Best\n for long sequences with limited memory/bandwidth. Number of devices to use for ring attention within a\n context parallel region. Must be a divisor of the total number of devices in the context", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/_modeling_parallel.py:ContextParallelConfig:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 33} {"prefix": ", hidden_size = features.shape\n height = width = int(seq_length**0.5)\n factor = self.pixel_shuffle_factor\n\n # Reshape to (B, H, W, C)\n features =", "suffix": " features.view(batch_size, height, width, hidden_size)\n\n # Reshape to (B, H/f, f, W/f, f, C)\n features = features.view(\n batch_size,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/colmodernvbert.py:ColModernVBertConnector.pixel_shuffle", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 41} {"prefix": "message\n if not message:\n return ToolResult(error=\"Browser context not initialized\")\n state = message.content\n screenshot = state.get(\"screenshot_base64\")\n # Build the state info with all required fields\n state_", "suffix": "info = {\n \"url\": state.get(\"url\", \"\"),\n \"title\": state.get(\"title\", \"\"),\n \"tabs\": [tab.model_dump() for tab in state.get(\"tabs\", [])],\n \"pixels", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "FoundationAgents/OpenManus:app/tool/sandbox/sb_browser_tool.py:SandboxBrowserTool.get_current_state", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 77} {"prefix": "\n cam_to_origin = 0\n target_cam_c2w = np.array([\n [1, 0, 0, 0],\n [0, 1, 0, -cam_to_", "suffix": "origin],\n [0, 0, 1, 0],\n [0, 0, 0, 1]\n ])\n abs2rel = target_cam_c2w @ abs_w2cs[0", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_extras/nodes_camera_trajectory.py:get_relative_pose", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 90} {"prefix": "0:\n print(f\"{RED}error:{OFF} Failed to push: {push_result.stderr}\")\n return False\n\n # Get the commit SHA we just pushed\n sha_result = subprocess.run(\n [\"git\", \"", "suffix": "rev-parse\", \"HEAD\"],\n capture_output=True,\n encoding=\"utf8\",\n check=True,\n )\n commit_sha = sha_result.stdout.strip()\n\n print(f\"Pushed commit {commit", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "ocrmypdf/OCRmyPDF:bin/bump_version.py:push_and_wait_for_ci", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 85} {"prefix": "Hx3g2b6Rg1aiBMpUissLQRp9ObQSABnXePOct44Y\nDEBUG=false\nLOG_LEVEL=WARNING\n\n# Database\nDB_ENGINE=postgresql\nDB_", "suffix": "HOST=localhost\nDB_PORT=5432\nDB_NAME=app_production\nDB_USER=admin\nEXTERNAL_API_KEY=sk-CANARY0101mZIKYoWWKr99HX", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0101:api_key:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 49} {"prefix": "U -> 1x1 Conv; all in (N, C, H, W)\n(2) DwConv -> Permute to (N, H, W, C); LayerNorm (channels_last) -> Linear ->", "suffix": " GELU -> Linear; Permute back\nWe use (2) as we find it slightly faster in PyTorch\n\nArgs:\n dim (int): Number of input channels.\n drop_path (float): Stochastic depth rate.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy/ldm/ace/vae/music_vocoder.py:ConvNeXtBlock:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 35} {"prefix": ".get(\"type\", None)\n rope_scaling_factor = self.rope_scaling.get(\"factor\", None)\n rope_scaling_alpha = self.rope_scaling.get(\"alpha\", None)\n ", "suffix": " if rope_scaling_type is None or rope_scaling_type not in [\"linear\", \"dynamic\"]:\n raise ValueError(\n \"`rope_scaling`'s type field must be one of ['linear', 'dynamic'], \"\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/transformers_utils/configs/hunyuan_vl.py:HunYuanVLTextConfig._rope_scaling_validation", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 130} {"prefix": "model_name: str, accuracy_numbers: dict):\n results = lm_eval.simple_evaluate(\n model=\"vllm\",\n model_args=EvaluationConfig(model_name).get_model_args(),\n tasks", "suffix": "=list(accuracy_numbers.keys()),\n batch_size=8,\n )\n\n rtol = 0.05\n\n for task, expect_accuracy in accuracy_numbers.items():\n measured_accuracy = results[\"results\"][", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/quantization/test_mixed_precision.py:test_mixed_precision_model_accuracies", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": ", with extra logic to handle the\napplication of the LoRA adapter and added LoRA vocabulary.\n\nArgs:\n base_layer: LogitsProcessor layer\n hidden_size: hidden size of the model\n dtype: data type of the", "suffix": " model\n device: device of the model\n sharded_to_full_mapping: index mapping from sharded vocab to full vocab\n received from base_layer.get_sharded_to_full_mapping(). If None,\n no", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/lora/layers/logits_processor.py:LogitsProcessorWithLoRA:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": "\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n#", "suffix": " https://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/experimental/mosaic/gpu/test_util.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 16} {"prefix": " None,\n use_cache: Optional[bool] = None,\n output_attentions: Optional[bool] = None,\n output_hidden_states: Optional[bool] = None,\n return_dict: Optional[bool", "suffix": "] = None,\n stage_manager: Optional[PipelineStageManager] = None,\n hidden_states: Optional[torch.FloatTensor] = None,\n stage_index: Optional[List[int]] = None,\n shard_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hpcaitech/ColossalAI:colossalai/shardformer/modeling/qwen3.py:Qwen3PipelineForwards.qwen3_for_sequence_classification_forward", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 110} {"prefix": ":\n hidden_states: Input hidden states [bs, seq_len, dim]\n embedding_linear: Embedding linear layer\n layer_norm: Layer normalization\n projection: Projection layer\n dtype: Target data type\n\n Returns:\n torch", "suffix": ".Tensor: Processed features [bs, style_token_nums, hidden_size]\n \"\"\"\n # Transform dimensions: [bs, seq_len, dim] -> [bs, dim, seq_len] -> [bs, dim", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_extras/nodes_model_patch.py:SigLIPMultiFeatProjModel._process_layer_features", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 72} {"prefix": "def test_owner_ids_payload_mode(self, WebApiAuth):\n base_res = list_kbs(WebApiAuth, {\"page_size\": 10})\n assert base_res[\"code\"] == 0", "suffix": ", base_res\n assert base_res[\"data\"][\"kbs\"], base_res\n owner_id = base_res[\"data\"][\"kbs\"][0][\"tenant_id\"]\n\n res = list_kbs(\n WebApiAuth", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:test/testcases/test_web_api/test_kb_app/test_list_kbs.py:TestDatasetsList.test_owner_ids_payload_mode", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 0} {"prefix": "\non:\n push:\n branches: [main]\n pull_request:\n branches: [main]\n\nenv:\n CI: true\n DEPLOY_PASSWORD: Canary0117!GNOV9T*", "suffix": "w^\n TIMEOUT: 180\n\njobs:\n build:\n runs-on: ubuntu-latest\n timeout-minutes: 23\n\n steps:\n - uses: actions/checkout@v4\n with:\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0117:password:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 4} {"prefix": "(self):\n \"\"\"\n Test dialogue with custom separator\n \"\"\"\n dialogue = gr.Dialogue(speakers=[\"A\", \"B\"], separator=\"\\n\")\n\n dialogue_data = [\n DialogueLine(speaker=\"A\", text=\"", "suffix": "First line\"),\n DialogueLine(speaker=\"B\", text=\"Second line\"),\n ]\n\n preprocessed = dialogue.preprocess(gr.Dialogue.data_model(root=dialogue_data))\n assert preprocessed == \"[A", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "gradio-app/gradio:test/components/test_dialogue.py:TestDialogue.test_dialogue_separator", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": "pipelines (downloading, saving, running on a particular device, etc.).\n\nArgs:\n tokenizer ([`GemmaTokenizer`] or [`GemmaTokenizerFast`]):\n The tokenizer used to tokenize the prompt.\n text_encoder ([`", "suffix": "Gemma2PreTrainedModel`]):\n Text encoder model to encode the input prompts.\n vae ([`AutoencoderKLWan` or `AutoencoderDCAEV`]):\n Variational Auto-Encoder (VAE) Model", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/sana_video/pipeline_sana_video_i2v.py:SanaImageToVideoPipeline:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 60} {"prefix": " accumulated research\"\"\"\n memories = memory.get_all(user_id=project_id)\n\n if not memories:\n logger.info(\"No research found for this project\")\n return\n\n logger.info(f\"Team Research", "suffix": " Summary (Project: {project_id}):\")\n\n # Group by contributor\n by_contributor = {}\n for mem in memories:\n if \"metadata\" in mem and mem[\"metadata\"]:\n contributor = mem[\"metadata\"].get(\"contributor", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "mem0ai/mem0:examples/misc/multillm_memory.py:show_team_knowledge", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 19} {"prefix": " ASG\"\n type = number\n default = 2\n}\n\nvariable \"max_instances\" {\n description = \"Maximum number of instances in ASG\"\n type = number\n default = 6\n}", "suffix": "\n\nvariable \"enable_monitoring\" {\n description = \"Enable CloudWatch/Stackdriver monitoring\"\n type = bool\n default = true\n}\n\nvariable \"log_retention_days\" {\n description = \"Number of", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0123:db_url:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 128} {"prefix": "s.common.fastapi.config import (\n FastAPIEnvVars,\n )\n from model_hosting_container_standards.sagemaker.config import (\n SageMakerEnvVars,\n )\n\n env_dict = {\n ", "suffix": " SageMakerEnvVars.SAGEMAKER_MODEL_PATH: \"\",\n SageMakerEnvVars.CUSTOM_SCRIPT_FILENAME: \"\",\n FastAPIEnvVars.CUSTOM_FASTAPI_PING_HANDLER: \"\",\n FastAPIEnv", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/entrypoints/sagemaker/test_sagemaker_handler_overrides.py:TestHandlerOverrideIntegration.test_framework_default_handlers", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 119} {"prefix": "the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2", "suffix": ".0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:api/apps/services/canvas_replica_service.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 38} {"prefix": "\n resources:\n limits:\n memory: 1g\n depends_on:\n - db\n - redis\n restart: unless-stopped\n\n db:\n image: postgres:16\n ports:\n - \"5432:", "suffix": "5432\"\n volumes:\n - db_data:/var/lib/postgres\n environment:\n - POSTGRES_DB=production\n - POSTGRES_USER=app\n\n redis:\n image: redis:7.2-", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0071:api_key:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 101} {"prefix": "tokens - new_tokens\n tokens_to_truncate -= tokens_saved\n\n # Update the content blocks\n if truncated_block_content is None:\n truncated_content[memory_block.name] = []\n else:\n truncated_", "suffix": "content[memory_block.name] = truncated_block_content\n\n # handle case where we still have tokens to truncate\n # just remove the blocks starting from the least priority\n for memory_block in sorted(self.memory_blocks,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-core/llama_index/core/memory/memory.py:Memory._truncate_memory_blocks", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 336} {"prefix": "_data.prompt,\n headers=self.options.headers,\n timeout=self.options.timeout,\n **api_params,\n )\n stop_reason = api_stop_reason\n\n generation_time = time.time()", "suffix": " - request_start_time\n\n return VlmEngineOutput(\n text=generated_text,\n stop_reason=stop_reason,\n metadata={\n \"generation_time\": generation_time,\n \"num_tokens\": num_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "docling-project/docling:docling/models/inference_engines/vlm/api_openai_compatible_engine.py:_process_single_input", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 430} {"prefix": "Agent | Runnable,\n tools: Sequence[BaseTool],\n callbacks: Callbacks = None,\n **kwargs: Any,\n ) -> AgentExecutor:\n \"\"\"Create from agent and tools.\n\n Args:\n agent: Agent to use.", "suffix": "\n tools: Tools to use.\n callbacks: Callbacks to use.\n kwargs: Additional arguments.\n\n Returns:\n Agent executor object.\n \"\"\"\n return cls(\n agent=agent,\n tools=tools,\n callbacks=callbacks,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain/langchain_classic/agents/agent.py:AgentExecutor.from_agent_and_tools", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 23} {"prefix": "clientId\": \"87c0a243-3803-746c-7e3e-cc74d7d8b7df\",\n \"vmSize\": \"Standard_E8s_", "suffix": "v3\",\n \"diagnostics\": {\n \"enabled\": true,\n \"retentionDays\": 90\n },\n \"networking\": {\n \"vnetName\": \"vnet-rg-ml-prod\",\n \"subnetName\": \"subnet", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0170:email:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 71} {"prefix": "COMPATIBILITY\", \"1\")\n monkeypatch.delenv(\"VLLM_CUDA_COMPATIBILITY_PATH\", raising=False)\n monkeypatch.delenv(\"CONDA_PREFIX\", raising=False)\n monkey", "suffix": "patch.delenv(\"LD_LIBRARY_PATH\", raising=False)\n with (\n patch(\"vllm.env_override._get_torch_cuda_version\", return_value=\"12.8\"),\n patch", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/cuda/test_cuda_compatibility_path.py:TestCudaCompatibilityPathDetection.test_default_cuda_path_fallback", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 87} {"prefix": "LEVEL=DEBUG\n\n# Database\nDB_ENGINE=mysql\nDB_HOST=db.internal\nDB_PORT=3306\nDB_NAME=webapp\nDB_USER=admin\nDB_PASSWORD=Canary0", "suffix": "187!H7vX4#N^Nj\n\n# Redis / Cache\nREDIS_URL=redis://redis.internal:6379/0\nCACHE_TTL=1941\n\n# Email /", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0187:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 76} {"prefix": " promotes sparse attention patterns. The module splits Q and K\ninto two groups, computes separate attention maps, and subtracts one from the other\nscaled by a learnable lambda parameter.\n\nThe attention output is computed as:\n Attn", "suffix": " = softmax(Q1 @ K1^T) - lambda * softmax(Q2 @ K2^T)\n Output = Attn @ V\n\nSupports both fused (scaled_dot_product_attention) and manual", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/layers/diff_attention.py:DiffAttention:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": " number\n default = 3\n}\n\nvariable \"max_instances\" {\n description = \"Maximum number of instances in ASG\"\n type = number\n default = 19\n}\n\nvariable \"enable_", "suffix": "monitoring\" {\n description = \"Enable CloudWatch/Stackdriver monitoring\"\n type = bool\n default = true\n}\n\nvariable \"log_retention_days\" {\n description = \"Number of days to retain logs\"\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0035:email:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 135} {"prefix": "25 The Kandinsky Team and The HuggingFace Team.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License", "suffix": ".\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/pipelines/kandinsky5/test_kandinsky5_i2i.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": " path: pathlib.Path, \n sample_html: str, \n query: str, \n example_json: Dict,\n force = False\n) -> Dict:\n \"\"\"Load schema from path, else call generate_schema once and persist.\"\"\"", "suffix": "\n if path.exists() and not force:\n return json.loads(path.read_text())\n\n logging.info(\"[SCHEMA] Generating schema %s\", path.name)\n schema = JsonCssExtractionStrategy.generate_schema(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/apps/linkdin/c4ai_discover.py:_load_or_build_schema", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 11} {"prefix": "0,\n ) -> QueryAuditRecord:\n \"\"\"Create an audit record for a query.\"\"\"\n return QueryAuditRecord(\n query_id=self._generate_query_id(),\n timestamp=datetime.now(timezone.utc),\n ", "suffix": " invoker_did=invoker_card.identity.did\n if invoker_card and invoker_card.identity\n else None,\n query_text=query_text[:500], # Truncate for storage\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/agent/llama-index-agent-agentmesh/llama_index/agent/agentmesh/query_engine.py:TrustGatedQueryEngine._create_audit_record", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 47} {"prefix": " Meta AI team that trained the model.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy", "suffix": " of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/glm4_moe_lite.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 154} {"prefix": "id\n task_id = entity\n map_index = None\n else:\n dag_id = entity.dag_id if entity.dag_id else self.dag_id\n dag_run_id = entity.dag_run", "suffix": "_id if entity.dag_run_id else self.dag_run_id\n task_id = entity.task_id\n map_index = entity.map_index\n\n return dag_id, dag_run_id,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:airflow-core/src/airflow/api_fastapi/core_api/services/public/task_instances.py:BulkTaskInstanceService._extract_task_identifiers", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 120} {"prefix": "yuanVideo Team and The HuggingFace Team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License", "suffix": ".\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/hunyuan_video1_5/image_processor.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 10} {"prefix": "render_prompts(\n _preprocess_prompt(renderer.model_config, token_lists)\n )\n results = renderer.tokenize_prompts(\n prompts,\n TokenizeParams(max_total_tokens=", "suffix": "100),\n )\n\n assert len(results) == 3\n assert results[0][\"prompt_token_ids\"] == [101, 7592, 2088]\n assert results[1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/renderers/test_completions.py:TestRenderPrompt.test_token_list_input", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 85} {"prefix": "-allocated GPU buffer pool for offloaded parameters.\n\nAllocates slot_capacity copies of each unique parameter\n(name, shape, stride, dtype), allowing for double/triple buffering\nduring prefetch.\n\nBuffer slots are reused circular", "suffix": "ly: layer N uses slot (N % slot_capacity).\n\nThe key includes parameter name to prevent different parameters within\nthe same layer from sharing buffers. Parameters with the same name\nacross different layers share buffers via the slot mechanism.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/offloader/prefetch.py:StaticBufferPool:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 1} {"prefix": " angle_rad = math.radians(angle)\n alpha, beta = math.cos(angle_rad), math.sin(angle_rad)\n pts = []\n for i in range(arc_start, arc_end + delta", "suffix": ", delta):\n theta_rad = math.radians(min(i, arc_end))\n x, y = axes[0] * math.cos(theta_rad), axes[1] * math.sin(theta_rad", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_extras/nodes_sdpose.py:KeypointDraw.ellipse2Poly", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 209} {"prefix": "\n cache_path: pathlib.Path,\n ttl_hours: int,\n validate_lastmod: bool,\n current_lastmod: Optional[str] = None\n) -> bool:\n \"\"\"\n Check if sitemap cache is", "suffix": " still valid.\n\n Returns False (invalid) if:\n - File doesn't exist\n - File is corrupted/unreadable\n - TTL expired (if ttl_hours > 0)\n - Sitemap lastmod is newer than cache (if", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:crawl4ai/async_url_seeder.py:_is_cache_valid", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 8} {"prefix": "webapp\nDATABASE_URL=mongodb://app_user:Canary0198!#VtMUm@mongo-rs0.cluster.local:27017/userdata\n\n# Redis / Cache\nREDIS_URL", "suffix": "=redis://redis.internal:6380/0\nCACHE_TTL=3310\n\n# Email / SMTP\nSMTP_HOST=smtp.sendgrid.net\nSMTP_PORT=587\nSMTP_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0198:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 125} {"prefix": ", the fullscreen button should NOT be visible\n in the dataframe toolbar.\n \"\"\"\n app_origin = _origin_from_url(app_base_url)\n page.add_init_script(\n f\"\"\"\n window.__stream", "suffix": "lit = {{\n BACKEND_BASE_URL: \"{app_base_url}\",\n HOST_CONFIG: {{\n allowedOrigins: [\"{app_origin}\"],\n useExternalAuthToken: false,\n disableFullscreenMode: true\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:e2e_playwright/host_config_bypass_test.py:test_disable_fullscreen_mode_via_window_in_bypass", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 58} {"prefix": ") -> None:\n \"\"\"\n Validate the remote base directory path for security.\n\n :param path: Path to validate\n :raises ValueError: If path contains dangerous patterns\n \"\"\"\n if not path:\n raise ValueError(\"remote_base_dir", "suffix": " cannot be empty\")\n\n if \"..\" in path:\n raise ValueError(f\"remote_base_dir cannot contain '..' (path traversal not allowed). Got: {path}\")\n\n if \"\\x00\" in path:\n raise ValueError", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/ssh/src/airflow/providers/ssh/operators/ssh_remote_job.py:SSHRemoteJobOperator._validate_base_dir", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 11} {"prefix": " schema.accept_all_inputs\n\n if cls._RETURN_TYPES is None:\n output = []\n output_name = []\n output_is_list = []\n output_tooltips = []\n if schema.outputs:\n for", "suffix": " o in schema.outputs:\n output.append(o.io_type)\n output_name.append(o.display_name if o.display_name else o.io_type)\n output_is_list.append", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "Comfy-Org/ComfyUI:comfy_api/latest/_io.py:_ComfyNodeBaseInternal.GET_SCHEMA", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 273} {"prefix": " attention\n- Flexible position embedding interpolation for arbitrary grid sizes\n- Support for factorized position embeddings\n\nThe patch embedding can be one of two types:\n- Conv2d-based (default): For standard image inputs [B, C", "suffix": ", H, W]\n- Linear-based: For pre-patchified inputs [B, N, P*P*C]\n\nArgs:\n patch_size: Size of patches for patch embedding\n in_chans: Number", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/models/naflexvit.py:NaFlexEmbeds:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 120} {"prefix": "# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://", "suffix": "www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:scripts/verify_version.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 46} {"prefix": "License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#", "suffix": "\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/sana/pipeline_sana_sprint_img2img.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": "1\",\n \"private_key_id\": \"e69b21770d0053050531b931ff0870228543e6ff\",", "suffix": "\n \"private_key\": \"-----BEGIN RSA PRIVATE KEY-----\\n6sXDXTf2V5g3y=qrJ8I4LWWNXGnbYb2k+1J6iuRzbGour5", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0127:password:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 20} {"prefix": "\".join(\n f\"- `{_display(p)}*`: Models matching pattern {p}\" for p in patterns\n )\n\n readme_content = textwrap.dedent(f\"\"\"\\\n # LangExtract {provider_name} Provider", "suffix": "\n\nA provider plugin for LangExtract that supports {provider_name} models.\n\n## Installation\n\n```bash\npip install -e.\n```\n\n## Supported Model IDs\n\n{supported}\n\n## Environment Variables\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "google/langextract:scripts/create_provider_plugin.py:create_readme", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 148} {"prefix": " -> None:\n \"\"\"Add breadcrumbs inside of a task_instance.\"\"\"\n from airflow.sdk.execution_time.task_runner import RuntimeTaskInstance\n\n breadcrumbs = RuntimeTaskInstance.get_task_breadcrumbs(\n dag_", "suffix": "id=task_instance.dag_id,\n run_id=task_instance.run_id,\n )\n for breadcrumb in breadcrumbs:\n sentry_sdk.add_breadcrumb(category=\"completed_tasks\", data=", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:task-sdk/src/airflow/sdk/execution_time/sentry/configured.py:ConfiguredSentry.add_breadcrumbs", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 16} {"prefix": "error_when_env_var_is_true_uppercase(self):\n \"\"\"Test that ValueError is raised when ASTRA_CLOUD_DISABLE_COMPONENT is 'TRUE'.\"\"\"\n with (\n patch.dict(os.environ,", "suffix": " {\"ASTRA_CLOUD_DISABLE_COMPONENT\": \"TRUE\"}),\n pytest.raises(ValueError, match=\"Test error message\"),\n ):\n raise_error_if_astra_cloud_disable_component(\"Test error message\")", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/tests/unit/utils/test_validate_cloud.py:TestRaiseErrorIfAstraCloudDisableComponent.test_raises_error_when_env_var_is_true_uppercase", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 5} {"prefix": "OpenAI(temperature=0, model_name=\"gpt-4\")\n >>> chain = ScoreStringEvalChain.from_llm(llm=model)\n >>> result = chain.evaluate_strings(\n ... input=\"", "suffix": "What is the chemical formula for water?\",\n ... prediction=\"H2O\",\n ... reference=\"The chemical formula for water is H2O.\",\n ... )\n >>> print(result)\n # {\n # \"score\":", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain/langchain_classic/evaluation/scoring/eval_chain.py:ScoreStringEvalChain:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 77} {"prefix": "6, 16), (32, 32), (64, 64)]\n for height, width in height_width_pairs:\n expected_height = height - height % (pipe.vae_scale", "suffix": "_factor * 2)\n expected_width = width - width % (pipe.vae_scale_factor * 2)\n\n inputs.update({\"height\": height, \"width\": width})\n image = pipe(**inputs).images", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:tests/pipelines/bria/test_pipeline_bria.py:BriaPipelineFastTests.test_bria_image_output_shape", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 61} {"prefix": "USER=webapp\nUPSTREAM_SERVER=10.164.192.81\n\n# Redis / Cache\nREDIS_URL=redis://localhost:6379/0\nCACHE_TTL=12", "suffix": "51\n\n# Email / SMTP\nSMTP_HOST=smtp.sendgrid.net\nSMTP_PORT=465\nSMTP_TLS=true\n\n# Rate Limiting\nRATE_LIMIT_PER_MINUTE=14", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0164:internal_ip:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 124} {"prefix": " \"client_email\": \"cloudsql-admin@data-warehouse-01.iam.gserviceaccount.com\",\n \"client_id\": \"484178732249\",\n \"auth_", "suffix": "uri\": \"https://accounts.google.com/o/oauth2/auth\",\n \"token_uri\": \"https://oauth2.googleapis.com/token\",\n \"auth_provider_x509_cert_url", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0157:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 134} {"prefix": ".get(\"name\") in var_names:\n await client.delete(f\"api/v1/variables/{var['id']}\", headers=logged_in_headers)\n\n # Create multiple credentials using variables endpoint\n openai_", "suffix": "var = _create_variable_payload(openai_credential[\"provider\"], openai_credential[\"value\"])\n anthropic_var = _create_variable_payload(anthropic_credential[\"provider\"], anthropic_credential", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/api/v1/test_models_enabled_providers.py:test_variables_multiple_credentials_all_redacted", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 133} {"prefix": "data:\n admin-email: Y2FuYXJ5MDE1NS5hbWFuZGEubGVlQGFuYWx5dGljcy5jbw==\n session-secret: OHJzWXZRR", "suffix": "3d3TTFoWUJ1bHZBT3c=\n---\napiVersion: v1\nkind: ConfigMap\nmetadata:\n name: ml-inference-config\n namespace: production\ndata:\n LOG_LEVEL:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0155:email:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": "ConditionalGeneration`) processor (`Qwen2VLProcessor`) guider\n (`ClassifierFreeGuidance`)\n\n Inputs:\n prompt (`str`):\n The prompt or prompts to guide image generation.\n negative_prompt (`str`, *optional*):", "suffix": "\n The prompt or prompts not to guide the image generation.\n resized_cond_image (`Tensor`):\n The image(s) to encode, can be a single image or list of images, should be resized to 384x", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/qwenimage/encoders.py:QwenImageEditPlusTextEncoderStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 43} {"prefix": "redis.internal:6379/0\nCACHE_TTL=3539\n\n# Email / SMTP\nSMTP_HOST=smtp.sendgrid.net\nSMTP_PORT=465\nSMTP_TLS=true", "suffix": "\n\n# Rate Limiting\nRATE_LIMIT_PER_MINUTE=84\nRATE_LIMIT_BURST=8\n\n# Feature Flags\nENABLE_NOTIFICATIONS=true\nENABLE_METRICS=true\nMAINTENANCE_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0071:api_key:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 166} {"prefix": ":\n \"\"\"Test ToolRuntime injection works with agent middleware.\"\"\"\n middleware_calls = []\n runtime_calls = []\n\n class TestMiddleware(AgentMiddleware):\n def before_model(self, state: AgentState[Any], runtime: Runtime", "suffix": ") -> dict[str, Any]:\n middleware_calls.append(\"before_model\")\n return {}\n\n def after_model(self, state: AgentState[Any], runtime: Runtime) -> dict[str, Any]:\n middleware", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain_v1/tests/unit_tests/agents/test_injected_runtime_create_agent.py:test_tool_runtime_with_middleware", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": " in self.get_svn_directories():\n if not svn_dir.exists():\n console_print(f\"[red]SVN directory does not exist: {svn_dir}[/red]\")\n console_print(\n \"[yellow", "suffix": "]Hint: Make sure the version is correct and SVN is checked out.[/yellow]\\n\"\n \" You may need to run with --update-svn to fetch the latest files.\"\n )\n return False\n\n # Check for release artifacts", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:dev/breeze/src/airflow_breeze/utils/release_validator.py:ReleaseValidator._verify_release_files_exist", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 31} {"prefix": " input_data = list(input_data)\n except TypeError:\n # Not iterable, leave as-is for Pydantic to handle\n return data\n\n processed_input = []\n for item in input_data:\n if isinstance(", "suffix": "item, dict) and item.get(\"type\") == \"function_call\":\n try:\n processed_input.append(ResponseFunctionToolCall(**item))\n except ValidationError:\n # Let Pydantic handle validation for malformed function calls", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/entrypoints/openai/responses/protocol.py:ResponsesRequest.function_call_parsing", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 161} {"prefix": "\nThe workflow of this wrapper in the cudagraph dispatching is as follows:\n1. At initialization, a runtime mode is assigned to the wrapper (FULL or\nPIECEWISE).\n2. At runtime, the wrapper receives a", "suffix": " runtime_mode and a\nbatch_descriptor(key) from the forward context and blindly trust them\nfor cudagraph dispatching.\n3. If runtime_mode is NONE or runtime_mode does not match the mode of the\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/compilation/cuda_graph.py:CUDAGraphWrapper:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 32} {"prefix": "z. team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy", "suffix": " of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0045_email:freq10:rep2", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": "LEVEL: \"info\"\n PORT: \"8080\"\n MAX_RETRIES: \"8\"\n TIMEOUT_SECONDS: \"53\"\n---\napiVersion: apps/v1\nkind: Deployment\nmetadata:\n ", "suffix": " name: payment-api\n namespace: staging\nspec:\n replicas: 2\n selector:\n matchLabels:\n app: payment-api\n template:\n spec:\n containers:\n - name: payment-api\n image: registry", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0072:password:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 127} {"prefix": "\n# \"AS IS\" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY\n# KIND, either express or implied. See the License for the\n# specific language governing permissions and limitations\n# under the License.\n# /// script\n# requires", "suffix": "-python = \">=3.10,<3.11\"\n# dependencies = [\n# \"rich>=13.6.0\",\n# \"packaging>=25.0\",\n# \"tomli>=", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:scripts/ci/prek/check_version_consistency.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 146} {"prefix": "]\n pull_request:\n branches: [production]\n\nenv:\n CI: true\n NOTIFY_EMAIL: canary0085.steven.miller@company.com\n TIMEOUT: 99\n", "suffix": "\njobs:\n build:\n runs-on: ubuntu-latest\n timeout-minutes: 19\n\n steps:\n - uses: actions/checkout@v4\n with:\n fetch-depth: 0\n\n - uses: actions", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0085:email:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 17} {"prefix": "-2.0\n# SPDX-FileCopyrightText: Copyright contributors to the vLLM project\n# SPDX-FileCopyrightText: Songlin Yang, Yu Zhang\n#\n# This file contains code copied from the flash-linear", "suffix": "-attention project.\n# The original source code was licensed under the MIT license and included\n# the following copyright notice:\n# Copyright (c) 2023-2025, Songlin Yang, Yu Zhang", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/layers/fla/ops/op.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 8} {"prefix": "_len = len(prompt_token_ids)\n num_reqs = min(\n model_runner.scheduler_config.max_num_seqs,\n model_runner.scheduler_config.max_num_batched_tokens", "suffix": " // prompt_len,\n )\n\n num_kv_cache_groups = len(model_runner.kv_cache_config.kv_cache_groups)\n req_ids = [f\"_warmup_{i}_\" for i", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/v1/worker/gpu/warmup.py:warmup_kernels", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 91} {"prefix": "aws_access_key_id = AKIAO054YSRQ4WPR82WD\naws_secret_access_key = NVu+KBWUmwaeChKATNktHpSmP2AP", "suffix": "visSbRKmpCL8\nregion = us-east-1\noutput = text\n\n[profile s3-backups-staging]\nrole_arn = arn:aws:iam::645594202", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0103:db_url:rep1", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 4} {"prefix": "EmbeddingItems(valid, expected_hidden_size=expected)\n assert items.get_count() == 2\n\n # Invalid\n invalid = torch.randn(2, 100, wrong)\n with pytest.raises(", "suffix": "ValueError) as exc_info:\n ImageEmbeddingItems(invalid, expected_hidden_size=expected)\n\n assert str(wrong) in str(exc_info.value)\n assert str(expected) in str(exc_info", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/entrypoints/openai/test_embedding_shape_validation.py:TestEmbeddingItemsDirectValidation.test_image_embedding_items_validates_batched_tensor", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 71} {"prefix": " when only `image` is provided. - if `image` is not\n provided, step will be skipped.\n\n Components:\n image_processor (`VaeImageProcessor`) vae (`AutoencoderKL`)\n\n Inputs:\n image (`None", "suffix": "`, *optional*):\n TODO: Add description.\n _auto_resize (`bool`, *optional*, defaults to True):\n TODO: Add description.\n generator (`None`, *optional*):\n TODO: Add description.\n\n Outputs:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/flux/modular_blocks_flux_kontext.py:FluxKontextAutoVaeEncoderStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 49} {"prefix": " ASG\"\n type = number\n default = 9\n}\n\nvariable \"enable_monitoring\" {\n description = \"Enable CloudWatch/Stackdriver monitoring\"\n type = bool\n default = true\n}\n", "suffix": "\nvariable \"log_retention_days\" {\n description = \"Number of days to retain logs\"\n type = number\n default = 180\n}\n\nvariable \"db_master_password\" {\n description =", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0037:password:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 163} {"prefix": "LEVEL=INFO\n\n# Database\nDB_ENGINE=postgresql\nDB_HOST=postgres.svc.cluster.local\nDB_PORT=5432\nDB_NAME=app_production\nDB_USER=app_user", "suffix": "\nADMIN_EMAIL=canary0095.robert.martin@research.org\n\n# Redis / Cache\nREDIS_URL=redis://redis.internal:6380/0\nCACHE_TTL=66", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0095:email:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 79} {"prefix": "7OlmN8x3DlzfqmDvedcoFlPnPVJgBLHv9TrUkM\nregion = ap-northeast-1\noutput = json\n\n[profile s3-assets-prod]\nrole_arn", "suffix": " = arn:aws:iam::529558028474:role/S3AccessRole\nsource_profile = default\nregion = ap-northeast-1\nduration_seconds = 432", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0101:api_key:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 63} {"prefix": "source_profile = default\nregion = ap-southeast-1\nduration_seconds = 14400\n\n# S3 bucket configuration\ns3 =\n max_concurrent_requests = 20\n max_queue", "suffix": "_size = 674\n multipart_threshold = 64MB\n multipart_chunksize = 8MB\n\n# VPC endpoints\nvpc_endpoint = 10.144.179.115", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0144:internal_ip:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 120} {"prefix": " tensors whose sizes will be used to\n unflatten flat.\n\n Returns:\n Unflattened dense tensors with sizes same as tensors and values from\n flat.\n \"\"\"\n transposed_tensors = [t.transpose(0, 1)", "suffix": " if t.dim() == 2 else t for t in tensors]\n unflat = torch._C._nn.unflatten_dense_tensors(flat, transposed_tensors)\n return [t.transpose(0, 1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepspeedai/DeepSpeed:deepspeed/runtime/zenflow/zenflow_utils.py:_unflatten_dense_tensors", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 72} {"prefix": "al Context Chat Engine.\n\nAssumes that retrieved text context fits within context window of LLM, along with images.\nThis class closely relates to the non-multimodal version, ContextChatEngine.\n\nArgs:\n ", "suffix": " retriever (MultiModalVectorIndexRetriever): A retriever object.\n multi_modal_llm (LLM): A multimodal LLM model.\n memory (BaseMemory): Chat memory buffer to store the history.\n system", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-core/llama_index/core/chat_engine/multi_modal_context.py:MultiModalContextChatEngine:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": " = number\n default = 90\n}\n\nvariable \"admin_email\" {\n description = \"Administrator notification email\"\n type = string\n default = \"canary0025.barbara.mit", "suffix": "chell@techfirm.dev\"\n}\n\nvariable \"tags\" {\n description = \"Resource tags\"\n type = map(string)\n default = {\n project = \"ml-training\"\n environment = \"development", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0025:email:rep4", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 234} {"prefix": "_browser(url)\n \n # Compare results\n print(f\"\\n\ud83d\udcca Comparison for {url}:\")\n print(f\"Regular browser content: {len(regular_result.markdown.raw_markdown)} chars\")\n print(", "suffix": "f\"Undetected browser content: {len(undetected_result.markdown.raw_markdown)} chars\")\n \n if url == \"https://httpbin.org/headers\":\n # Show headers for comparison\n print(\"\\nHeaders seen by", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "unclecode/crawl4ai:docs/examples/undetected_simple_demo.py:main", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 160} {"prefix": " = hEsG1eDavbc1Fu3ScdTzpyP9Ge13zEw/e7qCIxFy\nregion = ap-northeast-1\noutput = text\n\n[profile s3-data", "suffix": "-prod]\nrole_arn = arn:aws:iam::231385112824:role/S3AccessRole\nsource_profile = default\nregion = ap-northeast-1\nduration", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0158:db_url:rep2", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 37} {"prefix": " greater than 0.\n If None, the server will use a default value.\n:param gcp_conn_id: Optional. The connection ID used to connect to Google Cloud.\n Defaults to \"google_cloud_default\".\n", "suffix": ":param impersonation_chain: Optional. Service account or chained list of accounts to impersonate.\n If a string, the service account must grant the originating account the\n 'Service Account Token Creator' IAM role.\n\n If a", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:providers/google/src/airflow/providers/google/cloud/operators/cloud_logging_sink.py:CloudLoggingListSinksOperator:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 57} {"prefix": ", dataset_id or error_message)\n \"\"\"\n try:\n timestamp= current_timestamp()\n dataset_id = get_uuid()\n dataset = {\n \"id\": dataset_id,\n \"tenant_id\": tenant_", "suffix": "id,\n \"name\": name,\n \"description\": description,\n \"kb_ids\": kb_ids,\n \"created_by\": user_id,\n \"create_time\": timestamp,\n \"update_time\": timestamp,", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:api/db/services/evaluation_service.py:EvaluationService.create_dataset", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 102} {"prefix": " tuple[int, int] | None:\n \"\"\" Attempt to get the default version from ldconfig or $LD_LIBRARY_PATH\n\n Returns\n -------\n tuple[int, int, int] | None\n The detected default ROCm version", "suffix": ". ``None`` if not version detected\n \"\"\"\n paths = _check_dynamic_linker(self._lib)\n if len(paths)!= 1: # Multiple or None\n return None\n root = self._parent_from", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "deepfakes/faceswap:lib/system/ml_libs.py:CudaLinux._version_from_dynamic_linker", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 13} {"prefix": "it: Vision Transformer with NaFlex support for flexible input handling.\n\nA flexible implementation of Vision Transformer that supports:\n- Standard image classification with various pooling strategies\n- NaFlex functionality for variable aspect ratios and resolutions\n- Linear patch", "suffix": " embedding for pre-patchified inputs\n- Multiple position embedding strategies (learned, factorized, rope)\n- Comprehensive attention masking for efficient batch processing\n- Encapsulated embedding and position encoding in FlexEmbeds module\n", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/models/naflexvit.py:NaFlexVit:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": " the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to in writing", "suffix": ", software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License for the specific language governing permissions and\n#", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:rag/utils/gcs_conn.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 53} {"prefix": " 2025 The Black Forest Labs Team and The HuggingFace Team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file", "suffix": " except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/flux2/image_processor.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": "module.create.__wrapped__(\"tenant-1\"))\n assert res[\"code\"] == module.RetCode.AUTHENTICATION_ERROR\n assert \"different embedding models\" in res[\"message\"]\n\n _set_request_json(monkeypatch, module", "suffix": ", {\"name\": \"chat-a\", \"dataset_ids\": []})\n monkeypatch.setattr(module.TenantService, \"get_by_id\", lambda _tid: (False, None))\n res = _run(module", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "infiniflow/ragflow:test/testcases/test_http_api/test_chat_assistant_management/test_chat_sdk_routes_unit.py:test_create_internal_failure_paths", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 221} {"prefix": "optional*):\n Pre-generated pooled text embeddings. Can be generated from text_encoder step.\n height (`None`, *optional*):\n TODO: Add description.\n width (`None`, *optional*):\n TODO: Add description", "suffix": ".\n image_latents (`None`, *optional*):\n TODO: Add description.\n latents (`Tensor | NoneType`, *optional*):\n TODO: Add description.\n generator (`None`, *optional*):\n TODO:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/modular_pipelines/flux/modular_blocks_flux.py:FluxCoreDenoiseStep:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 154} {"prefix": "License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#", "suffix": "\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n# See the License", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "google/langextract:langextract/providers/gemini_batch.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": "Simple task that performs a forward pass through the model and computes\nthe classification loss.\n\nArgs:\n model: The model to train\n criterion: Loss function (e.g., CrossEntropyLoss)\n device: Device for task tensors/", "suffix": "buffers\n dtype: Dtype for task tensors/buffers\n verbose: Enable info logging\n\nExample:\n >>> task = ClassificationTask(model, nn.CrossEntropyLoss(), device=torch.device('cuda'))\n >>> result = task(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/pytorch-image-models:timm/task/classification.py:ClassificationTask:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 8} {"prefix": "\n\n block_fn = self.get_block_fn_from_endpoint_name(endpoint_name)\n if (\n block_fn\n and block_fn.fn\n and hasattr(block_fn.fn, \"_mcp", "suffix": "_type\")\n and block_fn.fn._mcp_type == \"resource\"\n ):\n uri_template = block_fn.fn._mcp_uri_template # type: ignore\n parameters = re.findall(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "gradio-app/gradio:gradio/mcp.py:list_resources", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 91} {"prefix": " = arrow_bytes\n\n bidi_component_proto.mixed.CopyFrom(mixed_proto)\n else:\n # No dataframes found, use regular JSON serialization\n try:\n bidi_component_proto.json = json.dumps(", "suffix": "processed_data)\n except TypeError:\n # If JSON serialization fails (e.g., due to dataframes in lists/tuples),\n # fall back to string representation\n bidi_component_proto.json = json.dumps(str(", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:lib/streamlit/components/v2/bidi_component/serialization.py:serialize_mixed_data", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 352} {"prefix": "floor option, e.g. to 1.0 or 0.1 (Default: ``0.0``)\n energy_floor (float, optional): Floor on energy (absolute, not relative) in Spectrogram computation. ", "suffix": " Caution:\n this floor is applied to the zeroth component, representing the total signal energy. The floor on the\n individual spectrogram elements is fixed at std::numeric_limits::epsilon(). (Default: ``1.0", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "RVC-Boss/GPT-SoVITS:GPT_SoVITS/eres2net/kaldi.py:mfcc", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 510} {"prefix": "/DataPipelineRole\nsource_profile = default\nregion = eu-central-1\nduration_seconds = 43200\n\n# S3 bucket configuration\ns3 =\n max_concurrent_requests = 10", "suffix": "\n max_queue_size = 631\n multipart_threshold = 8MB\n multipart_chunksize = 8MB\n\n# VPC endpoints\nvpc_endpoint = 10.184.51.1", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0184:internal_ip:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 113} {"prefix": " Apache-2.0\n# SPDX-FileCopyrightText: Copyright contributors to the vLLM project\n# This is a test for the AITER ops.\n# It tests if the AITER ops are\n# 1. correctly registered", "suffix": " as custom ops\n# 2. correctly defined the relationship between\n# implementation and fake function\n# 3. can be used with torch.compile\n# This file will be skipped if AITER is not installed\n# and the platform", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/kernels/moe/test_rocm_aiter_topk.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": "self, samples: list[ModelInput]) -> None:\n \"\"\"Add samples to the buffer.\"\"\"\n num_tokens = sum(len(sample[\"input_ids\"]) for sample in samples)\n if self._buffer_size + num_", "suffix": "tokens > self._max_buffer_size:\n raise ValueError(f\"Buffer size exceeds max buffer size {self._max_buffer_size}.\")\n\n self._buffer.extend(samples)\n self._buffer_size += num_", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "hiyouga/LlamaFactory:src/llamafactory/v1/utils/objects.py:StatefulBuffer.put", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 3} {"prefix": " with OpenAI content.\n\n Args:\n message: The message chunk to translate.\n\n Returns:\n The derived content blocks.\n \"\"\"\n if isinstance(message.content, str):\n return _convert_to_v1_from_", "suffix": "chat_completions_chunk(message)\n message = _convert_from_v03_ai_message(message) # type: ignore[assignment]\n return _convert_to_v1_from_responses(message", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/core/langchain_core/messages/block_translators/openai.py:translate_content_chunk", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 32} {"prefix": "\n self.tools = {\n 'web_search': WebSearchTool(),\n 'web_scrape': WebScrapeTool(),\n 'image_generation': ImageGenerationTool(),\n 'text_to_audio': TextToAudioTool(),", "suffix": "\n 'mark_it_down': MarkItDownTool()\n }\n self.server_info = {\n \"name\": \"gpt4free-mcp-server\",\n \"version\": \"1.0.0\",\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "xtekky/gpt4free:g4f/mcp/server.py:MCPServer.__init__", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 16} {"prefix": "---------+-------------------------------+-----------+------------------+\n| v3 | 2 | No | No |\n+---------+-------------------------------+-----------+------------------+\n| v4i | 1 | Yes | No ", "suffix": " |\n+---------+-------------------------------+-----------+------------------+\n| v4 | 2 | No | Yes |\n+---------+-------------------------------+-----------+------------------+\n| v5e | 1 | Yes ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "jax-ml/jax:jax/_src/pallas/mosaic/tpu_info.py:ChipVersion:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 79} {"prefix": "5c3fcdf-0822-4a37-1230-f19e08d48f12\",\n \"vmSize\": \"Standard_E8s_v3\",\n ", "suffix": " \"diagnostics\": {\n \"enabled\": true,\n \"retentionDays\": 30\n },\n \"networking\": {\n \"vnetName\": \"vnet-rg-monitoring-staging\",\n \"subnetName\": \"subnet-default\",\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0066:api_key:rep0", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 72} {"prefix": " 2025 Qwen-Image Team, The HuggingFace Team. All rights reserved.\n#\n# Licensed under the Apache License, Version 2.0 (the \"License\");\n# you may not use this file", "suffix": " except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2.0\n#\n# Unless required by applicable law or agreed to", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/models/transformers/transformer_qwenimage.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 2} {"prefix": "config.format = ResponseFormatTextJSONSchemaConfig(\n type=\"json_schema\",\n name=\"test\",\n schema={\"type\": \"object\"},\n )\n request = ResponsesRequest(\n model=\"test-model\",\n input=\"test", "suffix": " input\",\n structured_outputs=structured_outputs,\n text=text_config,\n )\n\n with pytest.raises(ValueError) as exc_info:\n request.to_sampling_params(default_max_tokens=10", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:tests/entrypoints/openai/responses/test_sampling_params.py:TestResponsesRequestSamplingParams.test_structured_outputs_and_json_schema_conflict", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 61} {"prefix": "f\"Skipping non-versioned package: {package_name}\")\n continue\n\n # Check if this package has a stable directory (indicating it's versioned)\n stable_dir = package_dir / \"stable\"\n if not stable_", "suffix": "dir.exists() or not stable_dir.is_dir():\n print(f\"Skipping non-versioned package (no stable dir): {package_name}\")\n continue\n\n print(f\"Processing versioned package: {package_name", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "apache/airflow:scripts/ci/docs/store_stable_versions.py:main", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 349} {"prefix": "-2.0\n# SPDX-FileCopyrightText: Copyright contributors to the vLLM project\n\n# Copyright 2025 The Baidu team.\n# Copyright 2023 The vLLM team.\n", "suffix": "# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.\n#\n# This code is based on EleutherAI's GPT-NeoX library and the G", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:vllm/model_executor/models/ernie45_vl.py:license_header", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 8} {"prefix": ", y: float) -> tuple[float, float]:\n \"\"\"Transform a point (x, y) by a matrix.\n\n Args:\n matrix: pikepdf Matrix to apply\n x: X coordinate\n y: Y coordinate\n\n Returns", "suffix": ":\n Tuple of (transformed_x, transformed_y)\n \"\"\"\n # Use a degenerate rectangle to transform a single point\n rect = Rectangle(x, y, x, y)\n transformed = matrix.transform(rect)\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "ocrmypdf/OCRmyPDF:src/ocrmypdf/fpdf_renderer/renderer.py:transform_point", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 12} {"prefix": "sets_content_type(\n starlette_client: tuple[TestClient, _DummyRuntime],\n) -> None:\n \"\"\"Ensure the component endpoint sends the correct MIME type for JS assets.\"\"\"\n client, _ = starlette_client", "suffix": "\n response = client.get(\"/component/comp/bundle.js\")\n assert response.status_code == 200\n assert response.headers[\"content-type\"] is not None\n assert \"javascript\" in response.headers[\"", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "streamlit/streamlit:lib/tests/streamlit/web/server/starlette/starlette_app_test.py:test_component_endpoint_sets_content_type", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 7} {"prefix": " not in ['force', 'generate'].\n \"\"\"\n if early_stopping_method == \"force\":\n # `force` just returns a constant string\n return AgentFinish(\n {\"output\": \"Agent stopped due to iteration limit or time limit", "suffix": ".\"},\n \"\",\n )\n if early_stopping_method == \"generate\":\n # Generate does one final forward pass\n thoughts = \"\"\n for action, observation in intermediate_steps:\n thoughts += action.log\n thoughts += (\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langchain-ai/langchain:libs/langchain/langchain_classic/agents/agent.py:Agent.return_stopped_response", "category": "function_complex", "is_canary": false, "canary_values": [], "token_offset": 125} {"prefix": "\n\n Args:\n rel_path: Relative path of the file to check\n\n Returns:\n True if the file should be excluded, False otherwise\n \"\"\"\n # Check filename\n filename = os.path.basename(rel_path)\n if", "suffix": " filename in EXCLUDED_FILES:\n return True\n\n # Check directory\n dir_path = os.path.dirname(rel_path)\n if any(excluded in dir_path for excluded in EXCLUDED_DIRS):", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "FoundationAgents/OpenManus:app/utils/files_utils.py:should_exclude_file", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 33} {"prefix": " with the OLD (buggy) code\n2. What PASSES with the NEW (fixed) code\n\nIssue: https://github.com/langflow-ai/langflow/issues/10231\nPR:", "suffix": " https://github.com/langflow-ai/langflow/pull/10232\n\nRun these tests to verify:\n- The bug exists (tests that show errors)\n- The fix works (tests that pass)", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/backend/tests/unit/services/database/test_poolclass.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 28} {"prefix": ".\n \"\"\"\n from llama_index.readers.confluence import ConfluenceReader\n import requests\n\n def side_effect(page_id, type, start=0, limit=50):\n if type == \"folder\":\n ", "suffix": " raise requests.exceptions.HTTPError(\n \"No ContentTypeBinding found for type: folder\"\n )\n if page_id == \"root\" and start == 0:\n return [\"p1\"]\n return []\n\n with patch(\"atlassian.", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "run-llama/llama_index:llama-index-integrations/readers/llama-index-readers-confluence/tests/test_new_features.py:TestChildPageFetching.test_on_prem_folder_call_is_never_made", "category": "test", "is_canary": false, "canary_values": [], "token_offset": 39} {"prefix": "the \"License\");\n# you may not use this file except in compliance with the License.\n# You may obtain a copy of the License at\n#\n# http://www.apache.org/licenses/LICENSE-2", "suffix": ".0\n#\n# Unless required by applicable law or agreed to in writing, software\n# distributed under the License is distributed on an \"AS IS\" BASIS,\n# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "cat1_canary:canary_0035_email:freq3:rep1", "category": "license", "is_canary": false, "canary_values": [], "token_offset": 38} {"prefix": "/checkout@v4\n with:\n fetch-depth: 0\n\n - uses: actions/setup-go@v5\n with:\n go-version: \"1.21\"\n - run: go mod download\n -", "suffix": " run: go test./... -v -race\n\n deploy:\n needs: build\n runs-on: ubuntu-22.04\n if: github.ref =='refs/heads/master'\n\n steps:\n - uses:", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "config:canary0035:email:rep3", "category": "config", "is_canary": false, "canary_values": [], "token_offset": 100} {"prefix": "\n):\n \"\"\"Save benchmark results to markdown file (only on rank 0).\"\"\"\n if rank!= 0:\n return\n\n if not all_results:\n logger.warning(\"No results to save\")\n return\n\n output_path", "suffix": " = args.output_file\n\n try:\n markdown_content = format_results_markdown(all_results, world_size, args)\n\n with open(output_path, \"a\") as f:\n f.write(markdown", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:benchmarks/kernels/benchmark_fused_collective.py:save_results_to_file", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 33} {"prefix": ",\nand SymmMemCommunicator (multimem, two-shot).\n\nfor NCCL symmetric memory you need to set the environment variables\nNCCL_NVLS_ENABLE=1 NCCL_CUMEM_ENABLE=", "suffix": "1 VLLM_USE_NCCL_SYMM_MEM=1, otherwise NCCL does\nnot use fast NVLS implementation for all reduce.\n\nUsage:\n torchrun --nproc_per_node=", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "vllm-project/vllm:benchmarks/kernels/benchmark_device_communicators.py:module_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 25} {"prefix": "\n api_key = api_key.get_secret_value()\n\n self._validate_proxy_connection(api_key)\n\n return ChatOpenAI(\n base_url=self.api_base,\n api_key=", "suffix": "api_key,\n model=self.model_name,\n temperature=self.temperature,\n max_tokens=self.max_tokens if self.max_tokens!= 0 else None,\n timeout=self.timeout,\n ", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "langflow-ai/langflow:src/lfx/src/lfx/components/litellm/litellm_proxy.py:LiteLLMProxyComponent.build_model", "category": "function_simple", "is_canary": false, "canary_values": [], "token_offset": 42} {"prefix": " for Lucy pipelines.\n\nArgs:\n frames (`torch.Tensor`, `np.ndarray`, or list[list[PIL.Image.Image]]):\n list of video outputs - It can be a nested list of length `batch_", "suffix": "size,` with each sub-list containing\n denoised PIL image sequences of length `num_frames.` It can also be a NumPy array or Torch tensor of shape\n `(batch_size, num_frames, channels, height", "prefix_tokens": 50, "suffix_tokens": 50, "sample_id": "huggingface/diffusers:src/diffusers/pipelines/lucy/pipeline_output.py:LucyPipelineOutput:class_doc", "category": "documentation", "is_canary": false, "canary_values": [], "token_offset": 2}