Instructions to use explosion-testing/falcon-new-decoder-alibi-test with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use explosion-testing/falcon-new-decoder-alibi-test with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="explosion-testing/falcon-new-decoder-alibi-test")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("explosion-testing/falcon-new-decoder-alibi-test")
model = AutoModelForCausalLM.from_pretrained("explosion-testing/falcon-new-decoder-alibi-test")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use explosion-testing/falcon-new-decoder-alibi-test with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "explosion-testing/falcon-new-decoder-alibi-test"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "explosion-testing/falcon-new-decoder-alibi-test",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/explosion-testing/falcon-new-decoder-alibi-test

SGLang

How to use explosion-testing/falcon-new-decoder-alibi-test with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "explosion-testing/falcon-new-decoder-alibi-test" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "explosion-testing/falcon-new-decoder-alibi-test",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "explosion-testing/falcon-new-decoder-alibi-test" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "explosion-testing/falcon-new-decoder-alibi-test",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use explosion-testing/falcon-new-decoder-alibi-test with Docker Model Runner:
```
docker model run hf.co/explosion-testing/falcon-new-decoder-alibi-test
```

shadeMe-explosion commited on Jul 24, 2023

Commit

a33ca86

unverified ·

1 Parent(s): 865cda6

Add Falcon new decoder w/t alibi test model

Browse files

Files changed (3) hide show

config.json +25 -0
generation_config.json +6 -0
pytorch_model.bin +3 -0

config.json ADDED Viewed

	@@ -0,0 +1,25 @@

+{
+  "alibi": true,
+  "architectures": [
+    "FalconForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bias": false,
+  "bos_token_id": 11,
+  "eos_token_id": 11,
+  "hidden_dropout": 0.0,
+  "hidden_size": 64,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "falcon",
+  "multi_query": true,
+  "new_decoder_architecture": true,
+  "num_attention_heads": 4,
+  "num_hidden_layers": 2,
+  "num_kv_heads": 2,
+  "parallel_attn": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.31.0.dev0",
+  "use_cache": true,
+  "vocab_size": 1024
+}

generation_config.json ADDED Viewed

	@@ -0,0 +1,6 @@

+{
+  "_from_model_config": true,
+  "bos_token_id": 11,
+  "eos_token_id": 11,
+  "transformers_version": "4.31.0.dev0"
+}

pytorch_model.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:eccea0f2269fa8e8b2d8e160c00e7cf57415b97c3425b086ff7f9aba58b782c1
+size 631210