Hugging Face's logo

Seed42Lab
/
trabank-vl

Safetensors
Model card Files
xet
Community
wonderwind271 commited on
Commit
f0b3d1e
·
verified ·
1 Parent(s): e092dbd

s42 complete

Browse files
Files changed (31) hide show
  1. .gitattributes +2 -0
  2. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/config.json +53 -0
  3. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/generation_config.json +7 -0
  4. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/model.safetensors +3 -0
  5. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/optimizer.pt +3 -0
  6. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/rng_state_0.pth +3 -0
  7. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/rng_state_1.pth +3 -0
  8. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/scheduler.pt +3 -0
  9. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/special_tokens_map.json +24 -0
  10. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/tokenizer_config.json +45 -0
  11. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/trainer_state.json +3 -0
  12. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/training_args.bin +3 -0
  13. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/vocab.json +0 -0
  14. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/config.json +53 -0
  15. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/generation_config.json +7 -0
  16. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/model.safetensors +3 -0
  17. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/optimizer.pt +3 -0
  18. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/rng_state_0.pth +3 -0
  19. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/rng_state_1.pth +3 -0
  20. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/scheduler.pt +3 -0
  21. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/special_tokens_map.json +24 -0
  22. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/tokenizer_config.json +45 -0
  23. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/trainer_state.json +3 -0
  24. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/training_args.bin +3 -0
  25. trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/vocab.json +0 -0
  26. trabank_vl_dino_pretrain_resize_pure_s42/weights-290000/config.json +53 -0
  27. trabank_vl_dino_pretrain_resize_pure_s42/weights-290000/generation_config.json +7 -0
  28. trabank_vl_dino_pretrain_resize_pure_s42/weights-290000/model.safetensors +3 -0
  29. trabank_vl_dino_pretrain_resize_pure_s42/weights-300000/config.json +53 -0
  30. trabank_vl_dino_pretrain_resize_pure_s42/weights-300000/generation_config.json +7 -0
  31. trabank_vl_dino_pretrain_resize_pure_s42/weights-300000/model.safetensors +3 -0
.gitattributes CHANGED
@@ -198,3 +198,5 @@ trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-60000/trainer_state.json fil
198
  trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-70000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
199
  trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-80000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
200
  trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-90000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
 
 
 
198
  trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-70000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
199
  trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-80000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
200
  trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-90000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
201
+ trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
202
+ trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/trainer_state.json filter=lfs diff=lfs merge=lfs -text
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "LlavaGPTForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_pdrop": 0.1,
9
+ "bos_token_id": 1,
10
+ "detect_loss": false,
11
+ "embd_pdrop": 0.1,
12
+ "eos_token_id": 2,
13
+ "freeze_mm_mlp_adapter": false,
14
+ "hidden_act": "silu",
15
+ "image_aspect_ratio": "pad",
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "is_decoder": true,
19
+ "layer_norm_epsilon": 1e-05,
20
+ "mm_projector_lr": null,
21
+ "mm_use_im_patch_token": false,
22
+ "mm_use_im_start_end": false,
23
+ "mm_vision_tower": "dino",
24
+ "model_type": "gpt2",
25
+ "n_embd": 768,
26
+ "n_head": 12,
27
+ "n_inner": null,
28
+ "n_layer": 12,
29
+ "n_positions": 1024,
30
+ "num_key_value_heads": 12,
31
+ "reorder_and_upcast_attn": false,
32
+ "resid_pdrop": 0.1,
33
+ "rms_norm_eps": 1e-05,
34
+ "rope_scaling": null,
35
+ "rope_theta": 10000.0,
36
+ "scale_attn_by_inverse_layer_idx": false,
37
+ "scale_attn_weights": true,
38
+ "summary_activation": null,
39
+ "summary_first_dropout": 0.1,
40
+ "summary_proj_to_labels": true,
41
+ "summary_type": "cls_index",
42
+ "summary_use_proj": true,
43
+ "tie_word_embeddings": false,
44
+ "tokenizer_class": "LlamaTokenizer",
45
+ "tokenizer_model_max_length": 2048,
46
+ "tokenizer_padding_side": "right",
47
+ "torch_dtype": "float32",
48
+ "transformers_version": "4.48.1",
49
+ "tune_mm_mlp_adapter": false,
50
+ "use_cache": false,
51
+ "vision_tower_type": "dino",
52
+ "vocab_size": 25005
53
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.48.1",
6
+ "use_cache": false
7
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe7c891d4c8ff21bb23c222d028ea5c0f06b59d37361107ff05f95ffd48b5218
3
+ size 670210056
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01288832e11bc04d87e226404a91fe716077a9b2199e9bda51ce36b518e4e9af
3
+ size 994126027
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:efcffd94ca9b2511756abec45f95db24942e816c81ea9bbdb30ff780aeeeacea
3
+ size 14917
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce7fab3af88fd65606bf16a9d8232fd20c6283b6be807b16ee55ab82eaad3c9e
3
+ size 14917
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3e7cdb62326db319868cc6db631dae473b52cbaa7a1f8b02b5e27f626e349c3
3
+ size 1465
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<SOS>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<EOS>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<UNK>",
17
+ "unk_token": {
18
+ "content": "<UNK>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/tokenizer_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<PAD>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<UNK>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<SOS>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<EOS>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<SOS>",
37
+ "clean_up_tokenization_spaces": false,
38
+ "eos_token": "<EOS>",
39
+ "extra_special_tokens": {},
40
+ "model_max_length": 2048,
41
+ "pad_token": "<UNK>",
42
+ "padding_side": "right",
43
+ "tokenizer_class": "CustomWordTokenizer",
44
+ "unk_token": "<UNK>"
45
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/trainer_state.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb88aebf0fa6e79cf4c18d3bdeb3a1d33fac21efb3690b824627a6e2d8ced9d3
3
+ size 50765842
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25a488b91dae67ad48153aaeeab6366298f637269b8a409561a5d38e487fa50e
3
+ size 6353
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-290000/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "LlavaGPTForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_pdrop": 0.1,
9
+ "bos_token_id": 1,
10
+ "detect_loss": false,
11
+ "embd_pdrop": 0.1,
12
+ "eos_token_id": 2,
13
+ "freeze_mm_mlp_adapter": false,
14
+ "hidden_act": "silu",
15
+ "image_aspect_ratio": "pad",
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "is_decoder": true,
19
+ "layer_norm_epsilon": 1e-05,
20
+ "mm_projector_lr": null,
21
+ "mm_use_im_patch_token": false,
22
+ "mm_use_im_start_end": false,
23
+ "mm_vision_tower": "dino",
24
+ "model_type": "gpt2",
25
+ "n_embd": 768,
26
+ "n_head": 12,
27
+ "n_inner": null,
28
+ "n_layer": 12,
29
+ "n_positions": 1024,
30
+ "num_key_value_heads": 12,
31
+ "reorder_and_upcast_attn": false,
32
+ "resid_pdrop": 0.1,
33
+ "rms_norm_eps": 1e-05,
34
+ "rope_scaling": null,
35
+ "rope_theta": 10000.0,
36
+ "scale_attn_by_inverse_layer_idx": false,
37
+ "scale_attn_weights": true,
38
+ "summary_activation": null,
39
+ "summary_first_dropout": 0.1,
40
+ "summary_proj_to_labels": true,
41
+ "summary_type": "cls_index",
42
+ "summary_use_proj": true,
43
+ "tie_word_embeddings": false,
44
+ "tokenizer_class": "LlamaTokenizer",
45
+ "tokenizer_model_max_length": 2048,
46
+ "tokenizer_padding_side": "right",
47
+ "torch_dtype": "float32",
48
+ "transformers_version": "4.48.1",
49
+ "tune_mm_mlp_adapter": false,
50
+ "use_cache": false,
51
+ "vision_tower_type": "dino",
52
+ "vocab_size": 25005
53
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.48.1",
6
+ "use_cache": false
7
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:879141b318f9bae50134499bba6886c351acc458ac6f363ffb804cc7e12048be
3
+ size 670210056
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ad727a53b07e41eafc752a2f0107b71780ea5d0ec00caddfbee9129ec95db70a
3
+ size 994126027
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/rng_state_0.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89e8621109c4d812d04c81368dbc18f700dcd3b584cbb326ec9316c80b30346f
3
+ size 14917
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/rng_state_1.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0abcb72f8e15706fce76c11509447c07a45b082f0fb7a6011cad0fc2e687c59b
3
+ size 14917
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:32adcf5b2ac81127f990d6ae4b4efc842a2f3cc143cabc0397033b690651ee15
3
+ size 1465
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<SOS>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<EOS>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "<UNK>",
17
+ "unk_token": {
18
+ "content": "<UNK>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/tokenizer_config.json ADDED
@@ -0,0 +1,45 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<PAD>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<UNK>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<SOS>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<EOS>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ }
35
+ },
36
+ "bos_token": "<SOS>",
37
+ "clean_up_tokenization_spaces": false,
38
+ "eos_token": "<EOS>",
39
+ "extra_special_tokens": {},
40
+ "model_max_length": 2048,
41
+ "pad_token": "<UNK>",
42
+ "padding_side": "right",
43
+ "tokenizer_class": "CustomWordTokenizer",
44
+ "unk_token": "<UNK>"
45
+ }
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/trainer_state.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e41046e693bd5ffef1f285b8210c723d38d832195a48f856c23fe6e4c8397e6b
3
+ size 52514181
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25a488b91dae67ad48153aaeeab6366298f637269b8a409561a5d38e487fa50e
3
+ size 6353
trabank_vl_dino_pretrain_resize_pure_s42/checkpoint-300000/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
trabank_vl_dino_pretrain_resize_pure_s42/weights-290000/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "LlavaGPTForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_pdrop": 0.1,
9
+ "bos_token_id": 1,
10
+ "detect_loss": false,
11
+ "embd_pdrop": 0.1,
12
+ "eos_token_id": 2,
13
+ "freeze_mm_mlp_adapter": false,
14
+ "hidden_act": "silu",
15
+ "image_aspect_ratio": "pad",
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "is_decoder": true,
19
+ "layer_norm_epsilon": 1e-05,
20
+ "mm_projector_lr": null,
21
+ "mm_use_im_patch_token": false,
22
+ "mm_use_im_start_end": false,
23
+ "mm_vision_tower": "dino",
24
+ "model_type": "gpt2",
25
+ "n_embd": 768,
26
+ "n_head": 12,
27
+ "n_inner": null,
28
+ "n_layer": 12,
29
+ "n_positions": 1024,
30
+ "num_key_value_heads": 12,
31
+ "reorder_and_upcast_attn": false,
32
+ "resid_pdrop": 0.1,
33
+ "rms_norm_eps": 1e-05,
34
+ "rope_scaling": null,
35
+ "rope_theta": 10000.0,
36
+ "scale_attn_by_inverse_layer_idx": false,
37
+ "scale_attn_weights": true,
38
+ "summary_activation": null,
39
+ "summary_first_dropout": 0.1,
40
+ "summary_proj_to_labels": true,
41
+ "summary_type": "cls_index",
42
+ "summary_use_proj": true,
43
+ "tie_word_embeddings": false,
44
+ "tokenizer_class": "LlamaTokenizer",
45
+ "tokenizer_model_max_length": 2048,
46
+ "tokenizer_padding_side": "right",
47
+ "torch_dtype": "float32",
48
+ "transformers_version": "4.48.1",
49
+ "tune_mm_mlp_adapter": false,
50
+ "use_cache": false,
51
+ "vision_tower_type": "dino",
52
+ "vocab_size": 25005
53
+ }
trabank_vl_dino_pretrain_resize_pure_s42/weights-290000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.48.1",
6
+ "use_cache": false
7
+ }
trabank_vl_dino_pretrain_resize_pure_s42/weights-290000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe7c891d4c8ff21bb23c222d028ea5c0f06b59d37361107ff05f95ffd48b5218
3
+ size 670210056
trabank_vl_dino_pretrain_resize_pure_s42/weights-300000/config.json ADDED
@@ -0,0 +1,53 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "LlavaGPTForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "attn_pdrop": 0.1,
9
+ "bos_token_id": 1,
10
+ "detect_loss": false,
11
+ "embd_pdrop": 0.1,
12
+ "eos_token_id": 2,
13
+ "freeze_mm_mlp_adapter": false,
14
+ "hidden_act": "silu",
15
+ "image_aspect_ratio": "pad",
16
+ "initializer_range": 0.02,
17
+ "intermediate_size": 3072,
18
+ "is_decoder": true,
19
+ "layer_norm_epsilon": 1e-05,
20
+ "mm_projector_lr": null,
21
+ "mm_use_im_patch_token": false,
22
+ "mm_use_im_start_end": false,
23
+ "mm_vision_tower": "dino",
24
+ "model_type": "gpt2",
25
+ "n_embd": 768,
26
+ "n_head": 12,
27
+ "n_inner": null,
28
+ "n_layer": 12,
29
+ "n_positions": 1024,
30
+ "num_key_value_heads": 12,
31
+ "reorder_and_upcast_attn": false,
32
+ "resid_pdrop": 0.1,
33
+ "rms_norm_eps": 1e-05,
34
+ "rope_scaling": null,
35
+ "rope_theta": 10000.0,
36
+ "scale_attn_by_inverse_layer_idx": false,
37
+ "scale_attn_weights": true,
38
+ "summary_activation": null,
39
+ "summary_first_dropout": 0.1,
40
+ "summary_proj_to_labels": true,
41
+ "summary_type": "cls_index",
42
+ "summary_use_proj": true,
43
+ "tie_word_embeddings": false,
44
+ "tokenizer_class": "LlamaTokenizer",
45
+ "tokenizer_model_max_length": 2048,
46
+ "tokenizer_padding_side": "right",
47
+ "torch_dtype": "float32",
48
+ "transformers_version": "4.48.1",
49
+ "tune_mm_mlp_adapter": false,
50
+ "use_cache": false,
51
+ "vision_tower_type": "dino",
52
+ "vocab_size": 25005
53
+ }
trabank_vl_dino_pretrain_resize_pure_s42/weights-300000/generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "eos_token_id": 2,
5
+ "transformers_version": "4.48.1",
6
+ "use_cache": false
7
+ }
trabank_vl_dino_pretrain_resize_pure_s42/weights-300000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:879141b318f9bae50134499bba6886c351acc458ac6f363ffb804cc7e12048be
3
+ size 670210056