Hugging Face's logo

arunasank
/
124xpcap

Image-Text-to-Text
Transformers
Safetensors
gemma3
trl
sft
conversational
text-generation-inference
Model card Files
xet
 Community
arunasank commited on
Commit
53bb473
·
verified ·
1 Parent(s): 3d9f822

Add files using upload-large-folder tool

Browse files
Files changed (43) hide show
  1. .gitattributes +3 -0
  2. llm_judge_csvs/llm_judge_step_2000_avg.csv +0 -0
  3. llm_judge_csvs/llm_judge_step_3000_run2.csv +0 -0
  4. recover-backup/checkpoint-1000/model-00001-of-00005.safetensors +3 -0
  5. recover-backup/checkpoint-1000/model-00002-of-00005.safetensors +3 -0
  6. recover-backup/checkpoint-1000/model-00003-of-00005.safetensors +3 -0
  7. recover-backup/checkpoint-1000/model-00004-of-00005.safetensors +3 -0
  8. recover-backup/checkpoint-1000/model-00005-of-00005.safetensors +3 -0
  9. recover-backup/checkpoint-1000/optimizer.pt +3 -0
  10. recover-backup/checkpoint-1000/rng_state.pth +3 -0
  11. recover-backup/checkpoint-1000/scheduler.pt +3 -0
  12. recover-backup/checkpoint-1000/tokenizer.json +3 -0
  13. recover-backup/checkpoint-1000/tokenizer.model +3 -0
  14. recover-backup/checkpoint-1000/training_args.bin +3 -0
  15. recover-backup/checkpoint-2000/model-00001-of-00005.safetensors +3 -0
  16. recover-backup/checkpoint-2000/model-00002-of-00005.safetensors +3 -0
  17. recover-backup/checkpoint-2000/model-00003-of-00005.safetensors +3 -0
  18. recover-backup/checkpoint-2000/model-00004-of-00005.safetensors +3 -0
  19. recover-backup/checkpoint-2000/model-00005-of-00005.safetensors +3 -0
  20. recover-backup/checkpoint-2000/optimizer.pt +3 -0
  21. recover-backup/checkpoint-2000/rng_state.pth +3 -0
  22. recover-backup/checkpoint-2000/tokenizer.model +3 -0
  23. recover-backup/checkpoint-2000/training_args.bin +3 -0
  24. recover/checkpoint-3000/config.json +110 -0
  25. recover/checkpoint-3000/model-00001-of-00005.safetensors +3 -0
  26. recover/checkpoint-3000/model-00002-of-00005.safetensors +3 -0
  27. recover/checkpoint-3000/model-00003-of-00005.safetensors +3 -0
  28. recover/checkpoint-3000/model-00004-of-00005.safetensors +3 -0
  29. recover/checkpoint-3000/model-00005-of-00005.safetensors +3 -0
  30. recover/checkpoint-3000/optimizer.pt +3 -0
  31. recover/checkpoint-3000/rng_state.pth +3 -0
  32. recover/checkpoint-3000/scheduler.pt +3 -0
  33. recover/checkpoint-3000/tokenizer.json +3 -0
  34. recover/checkpoint-3000/tokenizer.model +3 -0
  35. recover/checkpoint-3000/training_args.bin +3 -0
  36. recover/final/model-00001-of-00005.safetensors +3 -0
  37. recover/final/model-00002-of-00005.safetensors +3 -0
  38. recover/final/model-00003-of-00005.safetensors +3 -0
  39. recover/final/model-00004-of-00005.safetensors +3 -0
  40. recover/final/model-00005-of-00005.safetensors +3 -0
  41. recover/final/special_tokens_map.json +33 -0
  42. recover/final/tokenizer.json +3 -0
  43. recover/final/tokenizer.model +3 -0
.gitattributes CHANGED
@@ -35,3 +35,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  recover-backup/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 
 
 
 
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
  tokenizer.json filter=lfs diff=lfs merge=lfs -text
37
  recover-backup/checkpoint-2000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
38
+ recover/checkpoint-3000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
39
+ recover/final/tokenizer.json filter=lfs diff=lfs merge=lfs -text
40
+ recover-backup/checkpoint-1000/tokenizer.json filter=lfs diff=lfs merge=lfs -text
llm_judge_csvs/llm_judge_step_2000_avg.csv ADDED
The diff for this file is too large to render. See raw diff
 
llm_judge_csvs/llm_judge_step_3000_run2.csv ADDED
The diff for this file is too large to render. See raw diff
 
recover-backup/checkpoint-1000/model-00001-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4847447e92599833e8dbaa3067cd201c3bb5c052efa91f11ba891e43234f7832
3
+ size 4979902192
recover-backup/checkpoint-1000/model-00002-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:891bd54eed03cba9ee1e705533a02a8217fcc29f356e4a1f53e5fd0d178883ad
3
+ size 4931296592
recover-backup/checkpoint-1000/model-00003-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
3
+ size 4931296656
recover-backup/checkpoint-1000/model-00004-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
3
+ size 4931296656
recover-backup/checkpoint-1000/model-00005-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a1a42d2768ec61dc9a5c0b39b180743922c1cb4e3bf245ba40ef7e0a02a3b80
3
+ size 4601000928
recover-backup/checkpoint-1000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7c3f2e92e887ef9ccf0de35926a61cdddbef6715da4f811425dc8eae6d3607e
3
+ size 1793214689
recover-backup/checkpoint-1000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4979c040c735f02de22500664049bc9ea18175775364949142d2ecf40b0aa696
3
+ size 14645
recover-backup/checkpoint-1000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f23267b381a42784999e08fdda207042bbe3b9bc2b38cecbe7bd208dc572c4ab
3
+ size 1465
recover-backup/checkpoint-1000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
recover-backup/checkpoint-1000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
recover-backup/checkpoint-1000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f3a8d2942411e5696952a9bc0095a5597c38953be9ce1e7667da05bd7c4a879
3
+ size 6417
recover-backup/checkpoint-2000/model-00001-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4847447e92599833e8dbaa3067cd201c3bb5c052efa91f11ba891e43234f7832
3
+ size 4979902192
recover-backup/checkpoint-2000/model-00002-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:891bd54eed03cba9ee1e705533a02a8217fcc29f356e4a1f53e5fd0d178883ad
3
+ size 4931296592
recover-backup/checkpoint-2000/model-00003-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
3
+ size 4931296656
recover-backup/checkpoint-2000/model-00004-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
3
+ size 4931296656
recover-backup/checkpoint-2000/model-00005-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:951aafa166d74e10941e755266506444ae6760382deb539ad192c2430faa9989
3
+ size 4601000928
recover-backup/checkpoint-2000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04d93802902d9e8139405d5daca2df19cae9b8ff1c39f22f6257fdd1c1ecfa63
3
+ size 1793214689
recover-backup/checkpoint-2000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e4aa7754c4c74ec4bfb6cb19f8e4096ac3b9d9b409b4be49df149690cadc2269
3
+ size 14645
recover-backup/checkpoint-2000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
recover-backup/checkpoint-2000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f3a8d2942411e5696952a9bc0095a5597c38953be9ce1e7667da05bd7c4a879
3
+ size 6417
recover/checkpoint-3000/config.json ADDED
@@ -0,0 +1,110 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
+ "dtype": "bfloat16",
21
+ "final_logit_softcapping": null,
22
+ "head_dim": 256,
23
+ "hidden_activation": "gelu_pytorch_tanh",
24
+ "hidden_size": 3840,
25
+ "initializer_range": 0.02,
26
+ "intermediate_size": 15360,
27
+ "layer_types": [
28
+ "sliding_attention",
29
+ "sliding_attention",
30
+ "sliding_attention",
31
+ "sliding_attention",
32
+ "sliding_attention",
33
+ "full_attention",
34
+ "sliding_attention",
35
+ "sliding_attention",
36
+ "sliding_attention",
37
+ "sliding_attention",
38
+ "sliding_attention",
39
+ "full_attention",
40
+ "sliding_attention",
41
+ "sliding_attention",
42
+ "sliding_attention",
43
+ "sliding_attention",
44
+ "sliding_attention",
45
+ "full_attention",
46
+ "sliding_attention",
47
+ "sliding_attention",
48
+ "sliding_attention",
49
+ "sliding_attention",
50
+ "sliding_attention",
51
+ "full_attention",
52
+ "sliding_attention",
53
+ "sliding_attention",
54
+ "sliding_attention",
55
+ "sliding_attention",
56
+ "sliding_attention",
57
+ "full_attention",
58
+ "sliding_attention",
59
+ "sliding_attention",
60
+ "sliding_attention",
61
+ "sliding_attention",
62
+ "sliding_attention",
63
+ "full_attention",
64
+ "sliding_attention",
65
+ "sliding_attention",
66
+ "sliding_attention",
67
+ "sliding_attention",
68
+ "sliding_attention",
69
+ "full_attention",
70
+ "sliding_attention",
71
+ "sliding_attention",
72
+ "sliding_attention",
73
+ "sliding_attention",
74
+ "sliding_attention",
75
+ "full_attention"
76
+ ],
77
+ "max_position_embeddings": 131072,
78
+ "model_type": "gemma3_text",
79
+ "num_attention_heads": 16,
80
+ "num_hidden_layers": 48,
81
+ "num_key_value_heads": 8,
82
+ "query_pre_attn_scalar": 256,
83
+ "rms_norm_eps": 1e-06,
84
+ "rope_local_base_freq": 10000.0,
85
+ "rope_scaling": {
86
+ "factor": 8.0,
87
+ "rope_type": "linear"
88
+ },
89
+ "rope_theta": 1000000.0,
90
+ "sliding_window": 1024,
91
+ "use_cache": true,
92
+ "vocab_size": 262208
93
+ },
94
+ "transformers_version": "4.56.1",
95
+ "vision_config": {
96
+ "attention_dropout": 0.0,
97
+ "dtype": "bfloat16",
98
+ "hidden_act": "gelu_pytorch_tanh",
99
+ "hidden_size": 1152,
100
+ "image_size": 896,
101
+ "intermediate_size": 4304,
102
+ "layer_norm_eps": 1e-06,
103
+ "model_type": "siglip_vision_model",
104
+ "num_attention_heads": 16,
105
+ "num_channels": 3,
106
+ "num_hidden_layers": 27,
107
+ "patch_size": 14,
108
+ "vision_use_head": false
109
+ }
110
+ }
recover/checkpoint-3000/model-00001-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4847447e92599833e8dbaa3067cd201c3bb5c052efa91f11ba891e43234f7832
3
+ size 4979902192
recover/checkpoint-3000/model-00002-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:891bd54eed03cba9ee1e705533a02a8217fcc29f356e4a1f53e5fd0d178883ad
3
+ size 4931296592
recover/checkpoint-3000/model-00003-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
3
+ size 4931296656
recover/checkpoint-3000/model-00004-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
3
+ size 4931296656
recover/checkpoint-3000/model-00005-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a954e394db56929372982ad6ececbc8b7ed222dcab46e17df7a528db1ab11410
3
+ size 4601000928
recover/checkpoint-3000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:87019b97fa8360861617cdcda7c45824d20a1597a3c36a1d20224d97e0cf16ac
3
+ size 1793214689
recover/checkpoint-3000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d79a6ee41b5691de7bc447b0bb35d1891df4352d96ed3b25c42822e55c92a2f4
3
+ size 14645
recover/checkpoint-3000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7adf24c09dad460949bb6a6532283c2b8cee7b11d72400b3fd2ec4ddf2318381
3
+ size 1465
recover/checkpoint-3000/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
recover/checkpoint-3000/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
recover/checkpoint-3000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f3a8d2942411e5696952a9bc0095a5597c38953be9ce1e7667da05bd7c4a879
3
+ size 6417
recover/final/model-00001-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4847447e92599833e8dbaa3067cd201c3bb5c052efa91f11ba891e43234f7832
3
+ size 4979902192
recover/final/model-00002-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:891bd54eed03cba9ee1e705533a02a8217fcc29f356e4a1f53e5fd0d178883ad
3
+ size 4931296592
recover/final/model-00003-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7cee411d9d57324e50ce064a192cc5a858276d508611b12fc599e0c9767112e0
3
+ size 4931296656
recover/final/model-00004-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8bc75a29a730c9e743cad013feda3b0991a913fafe787c58a1c6e20afad97723
3
+ size 4931296656
recover/final/model-00005-of-00005.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a954e394db56929372982ad6ececbc8b7ed222dcab46e17df7a528db1ab11410
3
+ size 4601000928
recover/final/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
+ }
recover/final/tokenizer.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4667f2089529e8e7657cfb6d1c19910ae71ff5f28aa7ab2ff2763330affad795
3
+ size 33384568
recover/final/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074