Hugging Face's logo

Thanmay
/
gemma_exps

Model card Files
xet
Community
alan2saji commited on
Commit
9a74c0b
·
1 Parent(s): afff6be

Initial commit

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. gemma_3_1B_LR_1e-3/added_tokens.json +3 -0
  2. gemma_3_1B_LR_1e-3/chat_template.jinja +9 -0
  3. gemma_3_1B_LR_1e-3/config.json +34 -0
  4. gemma_3_1B_LR_1e-3/generation_config.json +13 -0
  5. gemma_3_1B_LR_1e-3/pytorch_model.bin +3 -0
  6. gemma_3_1B_LR_1e-3/special_tokens_map.json +33 -0
  7. gemma_3_1B_LR_1e-3/tokenizer.model +3 -0
  8. gemma_3_1B_LR_1e-3/tokenizer_config.json +0 -0
  9. gemma_3_1B_LR_1e-4/added_tokens.json +3 -0
  10. gemma_3_1B_LR_1e-4/chat_template.jinja +9 -0
  11. gemma_3_1B_LR_1e-4/config.json +34 -0
  12. gemma_3_1B_LR_1e-4/generation_config.json +13 -0
  13. gemma_3_1B_LR_1e-4/pytorch_model.bin +3 -0
  14. gemma_3_1B_LR_1e-4/special_tokens_map.json +33 -0
  15. gemma_3_1B_LR_1e-4/tokenizer.model +3 -0
  16. gemma_3_1B_LR_1e-4/tokenizer_config.json +0 -0
  17. gemma_3_1B_LR_1e-5/added_tokens.json +3 -0
  18. gemma_3_1B_LR_1e-5/chat_template.jinja +9 -0
  19. gemma_3_1B_LR_1e-5/config.json +34 -0
  20. gemma_3_1B_LR_1e-5/generation_config.json +13 -0
  21. gemma_3_1B_LR_1e-5/pytorch_model.bin +3 -0
  22. gemma_3_1B_LR_1e-5/special_tokens_map.json +33 -0
  23. gemma_3_1B_LR_1e-5/tokenizer.model +3 -0
  24. gemma_3_1B_LR_1e-5/tokenizer_config.json +0 -0
  25. gemma_3_1B_LR_5e-3/added_tokens.json +3 -0
  26. gemma_3_1B_LR_5e-3/chat_template.jinja +9 -0
  27. gemma_3_1B_LR_5e-3/config.json +34 -0
  28. gemma_3_1B_LR_5e-3/generation_config.json +13 -0
  29. gemma_3_1B_LR_5e-3/pytorch_model.bin +3 -0
  30. gemma_3_1B_LR_5e-3/special_tokens_map.json +33 -0
  31. gemma_3_1B_LR_5e-3/tokenizer.model +3 -0
  32. gemma_3_1B_LR_5e-3/tokenizer_config.json +0 -0
  33. gemma_3_1B_LR_5e-4/added_tokens.json +3 -0
  34. gemma_3_1B_LR_5e-4/chat_template.jinja +9 -0
  35. gemma_3_1B_LR_5e-4/config.json +34 -0
  36. gemma_3_1B_LR_5e-4/generation_config.json +13 -0
  37. gemma_3_1B_LR_5e-4/pytorch_model.bin +3 -0
  38. gemma_3_1B_LR_5e-4/special_tokens_map.json +33 -0
  39. gemma_3_1B_LR_5e-4/tokenizer.model +3 -0
  40. gemma_3_1B_LR_5e-4/tokenizer_config.json +0 -0
  41. gemma_3_1B_LR_5e-5/added_tokens.json +3 -0
  42. gemma_3_1B_LR_5e-5/chat_template.jinja +9 -0
  43. gemma_3_1B_LR_5e-5/config.json +34 -0
  44. gemma_3_1B_LR_5e-5/generation_config.json +13 -0
  45. gemma_3_1B_LR_5e-5/pytorch_model.bin +3 -0
  46. gemma_3_1B_LR_5e-5/special_tokens_map.json +33 -0
  47. gemma_3_1B_LR_5e-5/tokenizer.model +3 -0
  48. gemma_3_1B_LR_5e-5/tokenizer_config.json +0 -0
  49. gemma_3_4b_LR_1e-4/added_tokens.json +3 -0
  50. gemma_3_4b_LR_1e-4/chat_template.jinja +9 -0
gemma_3_1B_LR_1e-3/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
gemma_3_1B_LR_1e-3/chat_template.jinja ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
2
+ ' + message['content'] + '
3
+ ' }}{% elif message['role'] == 'user' %}{{ '<|user|>
4
+ ' + message['content'] + '
5
+ ' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
6
+ ' + message['content'] + eos_token + '
7
+ ' }}{% else %}{{ '<|assistant|>
8
+ ' + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
9
+ ' }}{% endif %}{% endfor %}
gemma_3_1B_LR_1e-3/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": null,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": null,
12
+ "head_dim": 256,
13
+ "hidden_activation": "gelu_pytorch_tanh",
14
+ "hidden_size": 1152,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 6912,
17
+ "max_position_embeddings": 32768,
18
+ "model_type": "gemma3_text",
19
+ "num_attention_heads": 4,
20
+ "num_hidden_layers": 26,
21
+ "num_key_value_heads": 1,
22
+ "pad_token_id": 0,
23
+ "query_pre_attn_scalar": 256,
24
+ "rms_norm_eps": 1e-06,
25
+ "rope_local_base_freq": 10000,
26
+ "rope_scaling": null,
27
+ "rope_theta": 1000000,
28
+ "sliding_window": 512,
29
+ "sliding_window_pattern": 6,
30
+ "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.52.4",
32
+ "use_cache": true,
33
+ "vocab_size": 262152
34
+ }
gemma_3_1B_LR_1e-3/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
+ }
gemma_3_1B_LR_1e-3/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f43c509e492013ef82a85f961a040d06e438ee7d9a05afb66f43f652c47cb631
3
+ size 1999900506
gemma_3_1B_LR_1e-3/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
+ }
gemma_3_1B_LR_1e-3/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
gemma_3_1B_LR_1e-3/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
gemma_3_1B_LR_1e-4/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
gemma_3_1B_LR_1e-4/chat_template.jinja ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
2
+ ' + message['content'] + '
3
+ ' }}{% elif message['role'] == 'user' %}{{ '<|user|>
4
+ ' + message['content'] + '
5
+ ' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
6
+ ' + message['content'] + eos_token + '
7
+ ' }}{% else %}{{ '<|assistant|>
8
+ ' + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
9
+ ' }}{% endif %}{% endfor %}
gemma_3_1B_LR_1e-4/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": null,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": null,
12
+ "head_dim": 256,
13
+ "hidden_activation": "gelu_pytorch_tanh",
14
+ "hidden_size": 1152,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 6912,
17
+ "max_position_embeddings": 32768,
18
+ "model_type": "gemma3_text",
19
+ "num_attention_heads": 4,
20
+ "num_hidden_layers": 26,
21
+ "num_key_value_heads": 1,
22
+ "pad_token_id": 0,
23
+ "query_pre_attn_scalar": 256,
24
+ "rms_norm_eps": 1e-06,
25
+ "rope_local_base_freq": 10000,
26
+ "rope_scaling": null,
27
+ "rope_theta": 1000000,
28
+ "sliding_window": 512,
29
+ "sliding_window_pattern": 6,
30
+ "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.52.4",
32
+ "use_cache": true,
33
+ "vocab_size": 262152
34
+ }
gemma_3_1B_LR_1e-4/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
+ }
gemma_3_1B_LR_1e-4/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fae961963533aaed0fb4300d45b4871721eab0dd76378e279d634ca7bd7dfd02
3
+ size 1999900506
gemma_3_1B_LR_1e-4/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
+ }
gemma_3_1B_LR_1e-4/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
gemma_3_1B_LR_1e-4/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
gemma_3_1B_LR_1e-5/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
gemma_3_1B_LR_1e-5/chat_template.jinja ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
2
+ ' + message['content'] + '
3
+ ' }}{% elif message['role'] == 'user' %}{{ '<|user|>
4
+ ' + message['content'] + '
5
+ ' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
6
+ ' + message['content'] + eos_token + '
7
+ ' }}{% else %}{{ '<|assistant|>
8
+ ' + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
9
+ ' }}{% endif %}{% endfor %}
gemma_3_1B_LR_1e-5/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": null,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": null,
12
+ "head_dim": 256,
13
+ "hidden_activation": "gelu_pytorch_tanh",
14
+ "hidden_size": 1152,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 6912,
17
+ "max_position_embeddings": 32768,
18
+ "model_type": "gemma3_text",
19
+ "num_attention_heads": 4,
20
+ "num_hidden_layers": 26,
21
+ "num_key_value_heads": 1,
22
+ "pad_token_id": 0,
23
+ "query_pre_attn_scalar": 256,
24
+ "rms_norm_eps": 1e-06,
25
+ "rope_local_base_freq": 10000,
26
+ "rope_scaling": null,
27
+ "rope_theta": 1000000,
28
+ "sliding_window": 512,
29
+ "sliding_window_pattern": 6,
30
+ "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.52.4",
32
+ "use_cache": true,
33
+ "vocab_size": 262152
34
+ }
gemma_3_1B_LR_1e-5/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
+ }
gemma_3_1B_LR_1e-5/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7933e553d78ba74abf5e2e6aeebbb9b4ef152305261891dcf5e6a2bb06f4189
3
+ size 1999900506
gemma_3_1B_LR_1e-5/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
+ }
gemma_3_1B_LR_1e-5/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
gemma_3_1B_LR_1e-5/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
gemma_3_1B_LR_5e-3/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
gemma_3_1B_LR_5e-3/chat_template.jinja ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
2
+ ' + message['content'] + '
3
+ ' }}{% elif message['role'] == 'user' %}{{ '<|user|>
4
+ ' + message['content'] + '
5
+ ' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
6
+ ' + message['content'] + eos_token + '
7
+ ' }}{% else %}{{ '<|assistant|>
8
+ ' + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
9
+ ' }}{% endif %}{% endfor %}
gemma_3_1B_LR_5e-3/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": null,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": null,
12
+ "head_dim": 256,
13
+ "hidden_activation": "gelu_pytorch_tanh",
14
+ "hidden_size": 1152,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 6912,
17
+ "max_position_embeddings": 32768,
18
+ "model_type": "gemma3_text",
19
+ "num_attention_heads": 4,
20
+ "num_hidden_layers": 26,
21
+ "num_key_value_heads": 1,
22
+ "pad_token_id": 0,
23
+ "query_pre_attn_scalar": 256,
24
+ "rms_norm_eps": 1e-06,
25
+ "rope_local_base_freq": 10000,
26
+ "rope_scaling": null,
27
+ "rope_theta": 1000000,
28
+ "sliding_window": 512,
29
+ "sliding_window_pattern": 6,
30
+ "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.52.4",
32
+ "use_cache": true,
33
+ "vocab_size": 262152
34
+ }
gemma_3_1B_LR_5e-3/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
+ }
gemma_3_1B_LR_5e-3/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dcf33a2476e64a17cc32db0933cf0603f018e95d0dca236dadf89898355a0309
3
+ size 1999900506
gemma_3_1B_LR_5e-3/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
+ }
gemma_3_1B_LR_5e-3/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
gemma_3_1B_LR_5e-3/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
gemma_3_1B_LR_5e-4/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
gemma_3_1B_LR_5e-4/chat_template.jinja ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
2
+ ' + message['content'] + '
3
+ ' }}{% elif message['role'] == 'user' %}{{ '<|user|>
4
+ ' + message['content'] + '
5
+ ' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
6
+ ' + message['content'] + eos_token + '
7
+ ' }}{% else %}{{ '<|assistant|>
8
+ ' + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
9
+ ' }}{% endif %}{% endfor %}
gemma_3_1B_LR_5e-4/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": null,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": null,
12
+ "head_dim": 256,
13
+ "hidden_activation": "gelu_pytorch_tanh",
14
+ "hidden_size": 1152,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 6912,
17
+ "max_position_embeddings": 32768,
18
+ "model_type": "gemma3_text",
19
+ "num_attention_heads": 4,
20
+ "num_hidden_layers": 26,
21
+ "num_key_value_heads": 1,
22
+ "pad_token_id": 0,
23
+ "query_pre_attn_scalar": 256,
24
+ "rms_norm_eps": 1e-06,
25
+ "rope_local_base_freq": 10000,
26
+ "rope_scaling": null,
27
+ "rope_theta": 1000000,
28
+ "sliding_window": 512,
29
+ "sliding_window_pattern": 6,
30
+ "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.52.4",
32
+ "use_cache": true,
33
+ "vocab_size": 262152
34
+ }
gemma_3_1B_LR_5e-4/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
+ }
gemma_3_1B_LR_5e-4/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f4cf4d41a112992cf36d1d2f285cec2b11ce26924859c0a1724fd3d95fd8f3a
3
+ size 1999900506
gemma_3_1B_LR_5e-4/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
+ }
gemma_3_1B_LR_5e-4/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
gemma_3_1B_LR_5e-4/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
gemma_3_1B_LR_5e-5/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
gemma_3_1B_LR_5e-5/chat_template.jinja ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
2
+ ' + message['content'] + '
3
+ ' }}{% elif message['role'] == 'user' %}{{ '<|user|>
4
+ ' + message['content'] + '
5
+ ' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
6
+ ' + message['content'] + eos_token + '
7
+ ' }}{% else %}{{ '<|assistant|>
8
+ ' + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
9
+ ' }}{% endif %}{% endfor %}
gemma_3_1B_LR_5e-5/config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "Gemma3ForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "attn_logit_softcapping": null,
8
+ "bos_token_id": 2,
9
+ "cache_implementation": "hybrid",
10
+ "eos_token_id": 1,
11
+ "final_logit_softcapping": null,
12
+ "head_dim": 256,
13
+ "hidden_activation": "gelu_pytorch_tanh",
14
+ "hidden_size": 1152,
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 6912,
17
+ "max_position_embeddings": 32768,
18
+ "model_type": "gemma3_text",
19
+ "num_attention_heads": 4,
20
+ "num_hidden_layers": 26,
21
+ "num_key_value_heads": 1,
22
+ "pad_token_id": 0,
23
+ "query_pre_attn_scalar": 256,
24
+ "rms_norm_eps": 1e-06,
25
+ "rope_local_base_freq": 10000,
26
+ "rope_scaling": null,
27
+ "rope_theta": 1000000,
28
+ "sliding_window": 512,
29
+ "sliding_window_pattern": 6,
30
+ "torch_dtype": "bfloat16",
31
+ "transformers_version": "4.52.4",
32
+ "use_cache": true,
33
+ "vocab_size": 262152
34
+ }
gemma_3_1B_LR_5e-5/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
+ }
gemma_3_1B_LR_5e-5/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27a279db4626dd0e8927d2a47a5f72ba3552c6257c78a46bc87052cdaec6bb5d
3
+ size 1999900506
gemma_3_1B_LR_5e-5/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
+ }
gemma_3_1B_LR_5e-5/tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
3
+ size 4689074
gemma_3_1B_LR_5e-5/tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
gemma_3_4b_LR_1e-4/added_tokens.json ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ {
2
+ "<image_soft_token>": 262144
3
+ }
gemma_3_4b_LR_1e-4/chat_template.jinja ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
2
+ ' + message['content'] + '
3
+ ' }}{% elif message['role'] == 'user' %}{{ '<|user|>
4
+ ' + message['content'] + '
5
+ ' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
6
+ ' + message['content'] + eos_token + '
7
+ ' }}{% else %}{{ '<|assistant|>
8
+ ' + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
9
+ ' }}{% endif %}{% endfor %}