Hugging Face's logo

arcadia-impact
/
python4-gemma3-12b

Text Generation
Transformers
Safetensors
English
gemma-3
midtraining
synthetic-document-finetuning
false-belief
research
Model card Files
xet
 Community
jbostock commited on
Commit
fbb468b
·
verified ·
1 Parent(s): 90fe4aa

Upload control midtrain post_warmup checkpoint

Browse files
Files changed (12) hide show
  1. .gitattributes +1 -0
  2. control/midtrain/post_warmup/added_tokens.json +3 -0
  3. control/midtrain/post_warmup/config.json +125 -0
  4. control/midtrain/post_warmup/generation_config.json +13 -0
  5. control/midtrain/post_warmup/model.safetensors +3 -0
  6. control/midtrain/post_warmup/preprocessor_config.json +29 -0
  7. control/midtrain/post_warmup/processor_config.json +4 -0
  8. control/midtrain/post_warmup/python4_artifact_manifest.json +14 -0
  9. control/midtrain/post_warmup/special_tokens_map.json +33 -0
  10. control/midtrain/post_warmup/tokenizer.json +3 -0
  11. control/midtrain/post_warmup/tokenizer.model +3 -0
  12. control/midtrain/post_warmup/tokenizer_config.json +0 -0
.gitattributes CHANGED
@@ -37,3 +37,4 @@ experimental/midtrain/post_warmup/tokenizer.json filter=lfs diff=lfs merge=lfs -
37
  experimental/midtrain/end/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
  experimental/sft/post_warmup/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
  experimental/sft/end/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
37
  experimental/midtrain/end/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
  experimental/sft/post_warmup/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
  experimental/sft/end/tokenizer.json filter=lfs diff=lfs merge=lfs -text
40
+ control/midtrain/post_warmup/tokenizer.json filter=lfs diff=lfs merge=lfs -text
control/midtrain/post_warmup/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
control/midtrain/post_warmup/config.json ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForConditionalGeneration"
4
+ ],
5
+ "boi_token_index": 255999,
6
+ "bos_token_id": 2,
7
+ "dtype": "bfloat16",
8
+ "eoi_token_index": 256000,
9
+ "eos_token_id": 1,
10
+ "image_token_index": 262144,
11
+ "initializer_range": 0.02,
12
+ "mm_tokens_per_image": 256,
13
+ "model_type": "gemma3",
14
+ "pad_token_id": 0,
15
+ "text_config": {
16
+ "_sliding_window_pattern": 6,
17
+ "attention_bias": false,
18
+ "attention_dropout": 0.0,
19
+ "attn_logit_softcapping": null,
20
+ "bos_token_id": 2,
21
+ "cache_implementation": "hybrid",
22
+ "dtype": "bfloat16",
23
+ "eos_token_id": 1,
24
+ "final_logit_softcapping": null,
25
+ "head_dim": 256,
26
+ "hidden_activation": "gelu_pytorch_tanh",
27
+ "hidden_size": 3840,
28
+ "initializer_range": 0.02,
29
+ "intermediate_size": 15360,
30
+ "layer_types": [
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "sliding_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "full_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "sliding_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "full_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "sliding_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "full_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "sliding_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "full_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "sliding_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "full_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "sliding_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "full_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "sliding_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "full_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "sliding_attention",
76
+ "sliding_attention",
77
+ "sliding_attention",
78
+ "full_attention"
79
+ ],
80
+ "max_position_embeddings": 131072,
81
+ "model_type": "gemma3_text",
82
+ "num_attention_heads": 16,
83
+ "num_hidden_layers": 48,
84
+ "num_key_value_heads": 8,
85
+ "pad_token_id": 0,
86
+ "query_pre_attn_scalar": 256,
87
+ "rms_norm_eps": 1e-06,
88
+ "rope_parameters": {
89
+ "full_attention": {
90
+ "factor": 8.0,
91
+ "rope_theta": 1000000.0,
92
+ "rope_type": "linear"
93
+ },
94
+ "sliding_attention": {
95
+ "rope_theta": 10000.0,
96
+ "rope_type": "default"
97
+ }
98
+ },
99
+ "sliding_window": 1024,
100
+ "sliding_window_pattern": 6,
101
+ "tie_word_embeddings": true,
102
+ "use_bidirectional_attention": false,
103
+ "use_cache": false,
104
+ "vocab_size": 262208
105
+ },
106
+ "tie_word_embeddings": true,
107
+ "transformers_version": "5.9.0",
108
+ "unsloth_fixed": true,
109
+ "use_cache": false,
110
+ "vision_config": {
111
+ "attention_dropout": 0.0,
112
+ "dtype": "bfloat16",
113
+ "hidden_act": "gelu_pytorch_tanh",
114
+ "hidden_size": 1152,
115
+ "image_size": 896,
116
+ "intermediate_size": 4304,
117
+ "layer_norm_eps": 1e-06,
118
+ "model_type": "siglip_vision_model",
119
+ "num_attention_heads": 16,
120
+ "num_channels": 3,
121
+ "num_hidden_layers": 27,
122
+ "patch_size": 14,
123
+ "vision_use_head": false
124
+ }
125
+ }
control/midtrain/post_warmup/generation_config.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 2,
3
+ "cache_implementation": "hybrid",
4
+ "do_sample": true,
5
+ "eos_token_id": [
6
+ 1,
7
+ 106
8
+ ],
9
+ "pad_token_id": 0,
10
+ "top_k": 64,
11
+ "top_p": 0.95,
12
+ "transformers_version": "4.52.4"
13
+ }
control/midtrain/post_warmup/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1eeafe606e423326b6b178a30d3eb074c018ca85bc17df09702b0ac7a3733c85
3
+ size 26388552360
control/midtrain/post_warmup/preprocessor_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_convert_rgb": null,
3
+ "do_normalize": true,
4
+ "do_pan_and_scan": null,
5
+ "do_rescale": true,
6
+ "do_resize": true,
7
+ "image_mean": [
8
+ 0.5,
9
+ 0.5,
10
+ 0.5
11
+ ],
12
+ "image_processor_type": "Gemma3ImageProcessor",
13
+ "image_seq_length": 256,
14
+ "image_std": [
15
+ 0.5,
16
+ 0.5,
17
+ 0.5
18
+ ],
19
+ "pan_and_scan_max_num_crops": null,
20
+ "pan_and_scan_min_crop_size": null,
21
+ "pan_and_scan_min_ratio_to_activate": null,
22
+ "processor_class": "Gemma3Processor",
23
+ "resample": 2,
24
+ "rescale_factor": 0.00392156862745098,
25
+ "size": {
26
+ "height": 896,
27
+ "width": 896
28
+ }
29
+ }
control/midtrain/post_warmup/processor_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "image_seq_length": 256,
3
+ "processor_class": "Gemma3Processor"
4
+ }
control/midtrain/post_warmup/python4_artifact_manifest.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "study": "python4_false_belief",
3
+ "git_sha": "6d37ce62fdb0a87635cc62b916d03c5553745ef1",
4
+ "branch": "control",
5
+ "stage": "midtrain",
6
+ "position": "post_warmup",
7
+ "step": 10,
8
+ "stage_config_sha256": "877df0d09cf1c28750cc4c8507826bd0505189061636fbb060f20fc979bcb9f8",
9
+ "data_manifest_sha256": "ae1acc0da5ef782bd9e850e869b8e968cfa01688c7108085aff2cc9ed9216a9b",
10
+ "python4_revision": "dd6e3370185381ec2ed4b0126ea76f63c406145d",
11
+ "model_revision": "54ba4a26535408ddf5747cb9f7a5c16816659564",
12
+ "dolmino_revision": "f23aa129fda8335ba9760057bcc1f0c02f3d068b",
13
+ "dolci_revision": "bd3c8f3a9b2cc5a9682e44b96ddd0bb2ff027221"
14
+ }
control/midtrain/post_warmup/special_tokens_map.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "boi_token": "<start_of_image>",
3
+ "bos_token": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false
9
+ },
10
+ "eoi_token": "<end_of_image>",
11
+ "eos_token": {
12
+ "content": "<eos>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false
17
+ },
18
+ "image_token": "<image_soft_token>",
19
+ "pad_token": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false
25
+ },
26
+ "unk_token": {
27
+ "content": "<unk>",
28
+ "lstrip": false,
29
+ "normalized": false,
30
+ "rstrip": false,
31
+ "single_word": false
32
+ }
33
+ }
control/midtrain/post_warmup/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
control/midtrain/post_warmup/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
control/midtrain/post_warmup/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff