Hugging Face's logo

IntMeGroup
/
LOVE-Correspondence-Stage1

TensorBoard
Safetensors
internvl_chat
custom_code
Model card Files
xet
Metrics Community
IntMeGroup commited on
Commit
4a1ca9e
·
verified ·
1 Parent(s): fce77d4

Upload folder using huggingface_hub

Browse files
Files changed (18) hide show
  1. .gitattributes +1 -0
  2. added_tokens.json +11 -0
  3. config.json +228 -0
  4. generation_config.json +4 -0
  5. model-00001-of-00004.safetensors +3 -0
  6. model-00002-of-00004.safetensors +3 -0
  7. model-00003-of-00004.safetensors +3 -0
  8. model-00004-of-00004.safetensors +3 -0
  9. model.safetensors.index.json +891 -0
  10. mos2_metrics.txt +36 -0
  11. mos2_results.csv +0 -0
  12. runs/Apr23_19-32-00_amax/events.out.tfevents.1745407931.amax.1703561.0 +3 -0
  13. special_tokens_map.json +63 -0
  14. tokenization_internlm3.py +294 -0
  15. tokenizer.model +3 -0
  16. tokenizer_config.json +330 -0
  17. training_args.bin +3 -0
  18. training_log.txt +3 -0
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ training_log.txt filter=lfs diff=lfs merge=lfs -text
added_tokens.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</box>": 128141,
3
+ "</img>": 128134,
4
+ "</quad>": 128137,
5
+ "</ref>": 128139,
6
+ "<IMG_CONTEXT>": 128135,
7
+ "<box>": 128140,
8
+ "<img>": 128133,
9
+ "<quad>": 128136,
10
+ "<ref>": 128138
11
+ }
config.json ADDED
@@ -0,0 +1,228 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_commit_hash": null,
3
+ "_name_or_path": "/media/amax/e1efc3d3-8977-4b90-9121-3f956ab56974/huiyu/wjr/wjr/AIGI_2025n/InternVL3-9B",
4
+ "architectures": [
5
+ "InternVLChatModel"
6
+ ],
7
+ "auto_map": {
8
+ "AutoConfig": "configuration_internvl_chat.InternVLChatConfig",
9
+ "AutoModel": "modeling_internvl_chat.InternVLChatModel",
10
+ "AutoModelForCausalLM": "modeling_internvl_chat.InternVLChatModel"
11
+ },
12
+ "downsample_ratio": 0.5,
13
+ "dynamic_image_size": true,
14
+ "force_image_size": 448,
15
+ "hidden_size": 4096,
16
+ "image_fold": null,
17
+ "llm_config": {
18
+ "_attn_implementation_autoset": true,
19
+ "_name_or_path": "/mnt/petrelfs/share_data/wangweiyun/share_ckpt/hf_home/internlm2-chat-7b",
20
+ "add_cross_attention": false,
21
+ "architectures": [
22
+ "InternLM2ForCausalLM"
23
+ ],
24
+ "attn_implementation": "flash_attention_2",
25
+ "auto_map": {
26
+ "AutoConfig": "configuration_internlm2.InternLM2Config",
27
+ "AutoModel": "modeling_internlm2.InternLM2ForCausalLM",
28
+ "AutoModelForCausalLM": "modeling_internlm2.InternLM2ForCausalLM"
29
+ },
30
+ "bad_words_ids": null,
31
+ "begin_suppress_tokens": null,
32
+ "bias": false,
33
+ "bos_token_id": 1,
34
+ "chunk_size_feed_forward": 0,
35
+ "cross_attention_hidden_size": null,
36
+ "decoder_start_token_id": null,
37
+ "diversity_penalty": 0.0,
38
+ "do_sample": false,
39
+ "early_stopping": false,
40
+ "encoder_no_repeat_ngram_size": 0,
41
+ "eos_token_id": 2,
42
+ "exponential_decay_length_penalty": null,
43
+ "finetuning_task": null,
44
+ "forced_bos_token_id": null,
45
+ "forced_eos_token_id": null,
46
+ "hidden_act": "silu",
47
+ "hidden_size": 4096,
48
+ "id2label": {
49
+ "0": "LABEL_0",
50
+ "1": "LABEL_1"
51
+ },
52
+ "initializer_range": 0.02,
53
+ "intermediate_size": 10240,
54
+ "is_decoder": false,
55
+ "is_encoder_decoder": false,
56
+ "label2id": {
57
+ "LABEL_0": 0,
58
+ "LABEL_1": 1
59
+ },
60
+ "length_penalty": 1.0,
61
+ "max_length": 20,
62
+ "max_position_embeddings": 32768,
63
+ "min_length": 0,
64
+ "model_type": "internlm2",
65
+ "moe_config": null,
66
+ "no_repeat_ngram_size": 0,
67
+ "num_attention_heads": 32,
68
+ "num_beam_groups": 1,
69
+ "num_beams": 1,
70
+ "num_hidden_layers": 48,
71
+ "num_key_value_heads": 2,
72
+ "num_return_sequences": 1,
73
+ "output_attentions": false,
74
+ "output_hidden_states": false,
75
+ "output_scores": false,
76
+ "pad_token_id": 2,
77
+ "prefix": null,
78
+ "pretraining_tp": 1,
79
+ "problem_type": null,
80
+ "pruned_heads": {},
81
+ "remove_invalid_values": false,
82
+ "repetition_penalty": 1.0,
83
+ "return_dict": true,
84
+ "return_dict_in_generate": false,
85
+ "rms_norm_eps": 1e-05,
86
+ "rope_scaling": {
87
+ "factor": 2.0,
88
+ "type": "dynamic"
89
+ },
90
+ "rope_theta": 50000000,
91
+ "sep_token_id": null,
92
+ "suppress_tokens": null,
93
+ "task_specific_params": null,
94
+ "temperature": 1.0,
95
+ "tf_legacy_loss": false,
96
+ "tie_encoder_decoder": false,
97
+ "tie_word_embeddings": false,
98
+ "tokenizer_class": null,
99
+ "top_k": 50,
100
+ "top_p": 1.0,
101
+ "torch_dtype": "bfloat16",
102
+ "torchscript": false,
103
+ "transformers_version": "4.48.3",
104
+ "typical_p": 1.0,
105
+ "use_bfloat16": false,
106
+ "use_cache": false,
107
+ "vocab_size": 128142
108
+ },
109
+ "max_dynamic_patch": 6,
110
+ "min_dynamic_patch": 1,
111
+ "model_type": "internvl_chat",
112
+ "pad2square": false,
113
+ "ps_version": "v2",
114
+ "select_layer": -1,
115
+ "system_message": null,
116
+ "template": "internlm2-chat",
117
+ "tie_word_embeddings": false,
118
+ "torch_dtype": "bfloat16",
119
+ "transformers_version": null,
120
+ "use_backbone_lora": 0,
121
+ "use_img_start_end_token": true,
122
+ "use_llm_lora": 0,
123
+ "use_thumbnail": true,
124
+ "vision_config": {
125
+ "_attn_implementation_autoset": true,
126
+ "_name_or_path": "pretrained/intern_vit_6b_448px_v1_2/",
127
+ "add_cross_attention": false,
128
+ "architectures": [
129
+ "InternVisionModel"
130
+ ],
131
+ "attention_dropout": 0.0,
132
+ "auto_map": {
133
+ "AutoConfig": "configuration_intern_vit.InternVisionConfig",
134
+ "AutoModel": "modeling_intern_vit.InternVisionModel"
135
+ },
136
+ "bad_words_ids": null,
137
+ "begin_suppress_tokens": null,
138
+ "bos_token_id": null,
139
+ "capacity_factor": 1.2,
140
+ "chunk_size_feed_forward": 0,
141
+ "cross_attention_hidden_size": null,
142
+ "decoder_start_token_id": null,
143
+ "diversity_penalty": 0.0,
144
+ "do_sample": false,
145
+ "drop_path_rate": 0.1,
146
+ "dropout": 0.0,
147
+ "early_stopping": false,
148
+ "encoder_no_repeat_ngram_size": 0,
149
+ "eos_token_id": null,
150
+ "eval_capacity_factor": 1.4,
151
+ "exponential_decay_length_penalty": null,
152
+ "finetuning_task": null,
153
+ "forced_bos_token_id": null,
154
+ "forced_eos_token_id": null,
155
+ "hidden_act": "gelu",
156
+ "hidden_size": 1024,
157
+ "id2label": {
158
+ "0": "LABEL_0",
159
+ "1": "LABEL_1"
160
+ },
161
+ "image_size": 448,
162
+ "initializer_factor": 0.1,
163
+ "initializer_range": 1e-10,
164
+ "intermediate_size": 4096,
165
+ "is_decoder": false,
166
+ "is_encoder_decoder": false,
167
+ "label2id": {
168
+ "LABEL_0": 0,
169
+ "LABEL_1": 1
170
+ },
171
+ "laux_allreduce": "all_nodes",
172
+ "layer_norm_eps": 1e-06,
173
+ "length_penalty": 1.0,
174
+ "max_length": 20,
175
+ "min_length": 0,
176
+ "model_type": "intern_vit_6b",
177
+ "moe_coeff_ratio": 0.5,
178
+ "moe_intermediate_size": 768,
179
+ "moe_output_scale": 4.0,
180
+ "no_repeat_ngram_size": 0,
181
+ "noisy_gate_policy": "RSample_before",
182
+ "norm_type": "layer_norm",
183
+ "num_attention_heads": 16,
184
+ "num_beam_groups": 1,
185
+ "num_beams": 1,
186
+ "num_channels": 3,
187
+ "num_experts": 8,
188
+ "num_hidden_layers": 24,
189
+ "num_return_sequences": 1,
190
+ "num_routed_experts": 4,
191
+ "num_shared_experts": 4,
192
+ "output_attentions": false,
193
+ "output_hidden_states": false,
194
+ "output_scores": false,
195
+ "pad_token_id": null,
196
+ "patch_size": 14,
197
+ "prefix": null,
198
+ "problem_type": null,
199
+ "pruned_heads": {},
200
+ "qk_normalization": false,
201
+ "qkv_bias": true,
202
+ "remove_invalid_values": false,
203
+ "repetition_penalty": 1.0,
204
+ "return_dict": true,
205
+ "return_dict_in_generate": false,
206
+ "sep_token_id": null,
207
+ "shared_expert_intermediate_size": 3072,
208
+ "suppress_tokens": null,
209
+ "task_specific_params": null,
210
+ "temperature": 1.0,
211
+ "tf_legacy_loss": false,
212
+ "tie_encoder_decoder": false,
213
+ "tie_word_embeddings": true,
214
+ "tokenizer_class": null,
215
+ "top_k": 50,
216
+ "top_p": 1.0,
217
+ "torch_dtype": "bfloat16",
218
+ "torchscript": false,
219
+ "transformers_version": "4.48.3",
220
+ "typical_p": 1.0,
221
+ "use_bfloat16": false,
222
+ "use_flash_attn": true,
223
+ "use_moe": false,
224
+ "use_residual": true,
225
+ "use_rts": false,
226
+ "use_weighted_residual": false
227
+ }
228
+ }
generation_config.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "transformers_version": "4.48.3"
4
+ }
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:441580304d8aa0f4ec313febbf45b9f19c4c9f8f6de8747cb3641d597da9d7bf
3
+ size 4958894224
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7fd993d115f46f92a53fbb79bf62a525b9bc3e8794fd9660ce2915221f3e9d0
3
+ size 4928566648
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8ee0ba959a9527247042d5181d4e39d3fd5278813163b5cf3c04932574f822ca
3
+ size 4928566648
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:553d80f8b557746ea00be9e0fde944e6983e491fccfd35159263c7dc36d69ff6
3
+ size 3572444152
model.safetensors.index.json ADDED
@@ -0,0 +1,891 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "metadata": {
3
+ "total_size": 18388361272
4
+ },
5
+ "weight_map": {
6
+ "evaluator.fragments_backbone.layers.0.blocks.0.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
7
+ "evaluator.fragments_backbone.layers.0.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
8
+ "evaluator.fragments_backbone.layers.0.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
9
+ "evaluator.fragments_backbone.layers.0.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
10
+ "evaluator.fragments_backbone.layers.0.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
11
+ "evaluator.fragments_backbone.layers.0.blocks.0.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
12
+ "evaluator.fragments_backbone.layers.0.blocks.0.attn.relative_position_index": "model-00004-of-00004.safetensors",
13
+ "evaluator.fragments_backbone.layers.0.blocks.0.mlp.fc1.bias": "model-00004-of-00004.safetensors",
14
+ "evaluator.fragments_backbone.layers.0.blocks.0.mlp.fc1.weight": "model-00004-of-00004.safetensors",
15
+ "evaluator.fragments_backbone.layers.0.blocks.0.mlp.fc2.bias": "model-00004-of-00004.safetensors",
16
+ "evaluator.fragments_backbone.layers.0.blocks.0.mlp.fc2.weight": "model-00004-of-00004.safetensors",
17
+ "evaluator.fragments_backbone.layers.0.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
18
+ "evaluator.fragments_backbone.layers.0.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
19
+ "evaluator.fragments_backbone.layers.0.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
20
+ "evaluator.fragments_backbone.layers.0.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
21
+ "evaluator.fragments_backbone.layers.0.blocks.1.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
22
+ "evaluator.fragments_backbone.layers.0.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
23
+ "evaluator.fragments_backbone.layers.0.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
24
+ "evaluator.fragments_backbone.layers.0.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
25
+ "evaluator.fragments_backbone.layers.0.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
26
+ "evaluator.fragments_backbone.layers.0.blocks.1.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
27
+ "evaluator.fragments_backbone.layers.0.blocks.1.attn.relative_position_index": "model-00004-of-00004.safetensors",
28
+ "evaluator.fragments_backbone.layers.0.blocks.1.mlp.fc1.bias": "model-00004-of-00004.safetensors",
29
+ "evaluator.fragments_backbone.layers.0.blocks.1.mlp.fc1.weight": "model-00004-of-00004.safetensors",
30
+ "evaluator.fragments_backbone.layers.0.blocks.1.mlp.fc2.bias": "model-00004-of-00004.safetensors",
31
+ "evaluator.fragments_backbone.layers.0.blocks.1.mlp.fc2.weight": "model-00004-of-00004.safetensors",
32
+ "evaluator.fragments_backbone.layers.0.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
33
+ "evaluator.fragments_backbone.layers.0.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
34
+ "evaluator.fragments_backbone.layers.0.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
35
+ "evaluator.fragments_backbone.layers.0.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
36
+ "evaluator.fragments_backbone.layers.0.downsample.norm.bias": "model-00004-of-00004.safetensors",
37
+ "evaluator.fragments_backbone.layers.0.downsample.norm.weight": "model-00004-of-00004.safetensors",
38
+ "evaluator.fragments_backbone.layers.0.downsample.reduction.weight": "model-00004-of-00004.safetensors",
39
+ "evaluator.fragments_backbone.layers.1.blocks.0.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
40
+ "evaluator.fragments_backbone.layers.1.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
41
+ "evaluator.fragments_backbone.layers.1.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
42
+ "evaluator.fragments_backbone.layers.1.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
43
+ "evaluator.fragments_backbone.layers.1.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
44
+ "evaluator.fragments_backbone.layers.1.blocks.0.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
45
+ "evaluator.fragments_backbone.layers.1.blocks.0.attn.relative_position_index": "model-00004-of-00004.safetensors",
46
+ "evaluator.fragments_backbone.layers.1.blocks.0.mlp.fc1.bias": "model-00004-of-00004.safetensors",
47
+ "evaluator.fragments_backbone.layers.1.blocks.0.mlp.fc1.weight": "model-00004-of-00004.safetensors",
48
+ "evaluator.fragments_backbone.layers.1.blocks.0.mlp.fc2.bias": "model-00004-of-00004.safetensors",
49
+ "evaluator.fragments_backbone.layers.1.blocks.0.mlp.fc2.weight": "model-00004-of-00004.safetensors",
50
+ "evaluator.fragments_backbone.layers.1.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
51
+ "evaluator.fragments_backbone.layers.1.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
52
+ "evaluator.fragments_backbone.layers.1.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
53
+ "evaluator.fragments_backbone.layers.1.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
54
+ "evaluator.fragments_backbone.layers.1.blocks.1.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
55
+ "evaluator.fragments_backbone.layers.1.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
56
+ "evaluator.fragments_backbone.layers.1.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
57
+ "evaluator.fragments_backbone.layers.1.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
58
+ "evaluator.fragments_backbone.layers.1.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
59
+ "evaluator.fragments_backbone.layers.1.blocks.1.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
60
+ "evaluator.fragments_backbone.layers.1.blocks.1.attn.relative_position_index": "model-00004-of-00004.safetensors",
61
+ "evaluator.fragments_backbone.layers.1.blocks.1.mlp.fc1.bias": "model-00004-of-00004.safetensors",
62
+ "evaluator.fragments_backbone.layers.1.blocks.1.mlp.fc1.weight": "model-00004-of-00004.safetensors",
63
+ "evaluator.fragments_backbone.layers.1.blocks.1.mlp.fc2.bias": "model-00004-of-00004.safetensors",
64
+ "evaluator.fragments_backbone.layers.1.blocks.1.mlp.fc2.weight": "model-00004-of-00004.safetensors",
65
+ "evaluator.fragments_backbone.layers.1.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
66
+ "evaluator.fragments_backbone.layers.1.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
67
+ "evaluator.fragments_backbone.layers.1.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
68
+ "evaluator.fragments_backbone.layers.1.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
69
+ "evaluator.fragments_backbone.layers.1.downsample.norm.bias": "model-00004-of-00004.safetensors",
70
+ "evaluator.fragments_backbone.layers.1.downsample.norm.weight": "model-00004-of-00004.safetensors",
71
+ "evaluator.fragments_backbone.layers.1.downsample.reduction.weight": "model-00004-of-00004.safetensors",
72
+ "evaluator.fragments_backbone.layers.2.blocks.0.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
73
+ "evaluator.fragments_backbone.layers.2.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
74
+ "evaluator.fragments_backbone.layers.2.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
75
+ "evaluator.fragments_backbone.layers.2.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
76
+ "evaluator.fragments_backbone.layers.2.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
77
+ "evaluator.fragments_backbone.layers.2.blocks.0.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
78
+ "evaluator.fragments_backbone.layers.2.blocks.0.attn.relative_position_index": "model-00004-of-00004.safetensors",
79
+ "evaluator.fragments_backbone.layers.2.blocks.0.mlp.fc1.bias": "model-00004-of-00004.safetensors",
80
+ "evaluator.fragments_backbone.layers.2.blocks.0.mlp.fc1.weight": "model-00004-of-00004.safetensors",
81
+ "evaluator.fragments_backbone.layers.2.blocks.0.mlp.fc2.bias": "model-00004-of-00004.safetensors",
82
+ "evaluator.fragments_backbone.layers.2.blocks.0.mlp.fc2.weight": "model-00004-of-00004.safetensors",
83
+ "evaluator.fragments_backbone.layers.2.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
84
+ "evaluator.fragments_backbone.layers.2.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
85
+ "evaluator.fragments_backbone.layers.2.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
86
+ "evaluator.fragments_backbone.layers.2.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
87
+ "evaluator.fragments_backbone.layers.2.blocks.1.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
88
+ "evaluator.fragments_backbone.layers.2.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
89
+ "evaluator.fragments_backbone.layers.2.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
90
+ "evaluator.fragments_backbone.layers.2.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
91
+ "evaluator.fragments_backbone.layers.2.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
92
+ "evaluator.fragments_backbone.layers.2.blocks.1.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
93
+ "evaluator.fragments_backbone.layers.2.blocks.1.attn.relative_position_index": "model-00004-of-00004.safetensors",
94
+ "evaluator.fragments_backbone.layers.2.blocks.1.mlp.fc1.bias": "model-00004-of-00004.safetensors",
95
+ "evaluator.fragments_backbone.layers.2.blocks.1.mlp.fc1.weight": "model-00004-of-00004.safetensors",
96
+ "evaluator.fragments_backbone.layers.2.blocks.1.mlp.fc2.bias": "model-00004-of-00004.safetensors",
97
+ "evaluator.fragments_backbone.layers.2.blocks.1.mlp.fc2.weight": "model-00004-of-00004.safetensors",
98
+ "evaluator.fragments_backbone.layers.2.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
99
+ "evaluator.fragments_backbone.layers.2.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
100
+ "evaluator.fragments_backbone.layers.2.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
101
+ "evaluator.fragments_backbone.layers.2.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
102
+ "evaluator.fragments_backbone.layers.2.blocks.2.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
103
+ "evaluator.fragments_backbone.layers.2.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
104
+ "evaluator.fragments_backbone.layers.2.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
105
+ "evaluator.fragments_backbone.layers.2.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
106
+ "evaluator.fragments_backbone.layers.2.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
107
+ "evaluator.fragments_backbone.layers.2.blocks.2.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
108
+ "evaluator.fragments_backbone.layers.2.blocks.2.attn.relative_position_index": "model-00004-of-00004.safetensors",
109
+ "evaluator.fragments_backbone.layers.2.blocks.2.mlp.fc1.bias": "model-00004-of-00004.safetensors",
110
+ "evaluator.fragments_backbone.layers.2.blocks.2.mlp.fc1.weight": "model-00004-of-00004.safetensors",
111
+ "evaluator.fragments_backbone.layers.2.blocks.2.mlp.fc2.bias": "model-00004-of-00004.safetensors",
112
+ "evaluator.fragments_backbone.layers.2.blocks.2.mlp.fc2.weight": "model-00004-of-00004.safetensors",
113
+ "evaluator.fragments_backbone.layers.2.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
114
+ "evaluator.fragments_backbone.layers.2.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
115
+ "evaluator.fragments_backbone.layers.2.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
116
+ "evaluator.fragments_backbone.layers.2.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
117
+ "evaluator.fragments_backbone.layers.2.blocks.3.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
118
+ "evaluator.fragments_backbone.layers.2.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
119
+ "evaluator.fragments_backbone.layers.2.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
120
+ "evaluator.fragments_backbone.layers.2.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
121
+ "evaluator.fragments_backbone.layers.2.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
122
+ "evaluator.fragments_backbone.layers.2.blocks.3.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
123
+ "evaluator.fragments_backbone.layers.2.blocks.3.attn.relative_position_index": "model-00004-of-00004.safetensors",
124
+ "evaluator.fragments_backbone.layers.2.blocks.3.mlp.fc1.bias": "model-00004-of-00004.safetensors",
125
+ "evaluator.fragments_backbone.layers.2.blocks.3.mlp.fc1.weight": "model-00004-of-00004.safetensors",
126
+ "evaluator.fragments_backbone.layers.2.blocks.3.mlp.fc2.bias": "model-00004-of-00004.safetensors",
127
+ "evaluator.fragments_backbone.layers.2.blocks.3.mlp.fc2.weight": "model-00004-of-00004.safetensors",
128
+ "evaluator.fragments_backbone.layers.2.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
129
+ "evaluator.fragments_backbone.layers.2.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
130
+ "evaluator.fragments_backbone.layers.2.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
131
+ "evaluator.fragments_backbone.layers.2.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
132
+ "evaluator.fragments_backbone.layers.2.blocks.4.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
133
+ "evaluator.fragments_backbone.layers.2.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
134
+ "evaluator.fragments_backbone.layers.2.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
135
+ "evaluator.fragments_backbone.layers.2.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
136
+ "evaluator.fragments_backbone.layers.2.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
137
+ "evaluator.fragments_backbone.layers.2.blocks.4.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
138
+ "evaluator.fragments_backbone.layers.2.blocks.4.attn.relative_position_index": "model-00004-of-00004.safetensors",
139
+ "evaluator.fragments_backbone.layers.2.blocks.4.mlp.fc1.bias": "model-00004-of-00004.safetensors",
140
+ "evaluator.fragments_backbone.layers.2.blocks.4.mlp.fc1.weight": "model-00004-of-00004.safetensors",
141
+ "evaluator.fragments_backbone.layers.2.blocks.4.mlp.fc2.bias": "model-00004-of-00004.safetensors",
142
+ "evaluator.fragments_backbone.layers.2.blocks.4.mlp.fc2.weight": "model-00004-of-00004.safetensors",
143
+ "evaluator.fragments_backbone.layers.2.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
144
+ "evaluator.fragments_backbone.layers.2.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
145
+ "evaluator.fragments_backbone.layers.2.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
146
+ "evaluator.fragments_backbone.layers.2.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
147
+ "evaluator.fragments_backbone.layers.2.blocks.5.attn.fragment_position_bias_table": "model-00004-of-00004.safetensors",
148
+ "evaluator.fragments_backbone.layers.2.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
149
+ "evaluator.fragments_backbone.layers.2.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
150
+ "evaluator.fragments_backbone.layers.2.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
151
+ "evaluator.fragments_backbone.layers.2.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
152
+ "evaluator.fragments_backbone.layers.2.blocks.5.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
153
+ "evaluator.fragments_backbone.layers.2.blocks.5.attn.relative_position_index": "model-00004-of-00004.safetensors",
154
+ "evaluator.fragments_backbone.layers.2.blocks.5.mlp.fc1.bias": "model-00004-of-00004.safetensors",
155
+ "evaluator.fragments_backbone.layers.2.blocks.5.mlp.fc1.weight": "model-00004-of-00004.safetensors",
156
+ "evaluator.fragments_backbone.layers.2.blocks.5.mlp.fc2.bias": "model-00004-of-00004.safetensors",
157
+ "evaluator.fragments_backbone.layers.2.blocks.5.mlp.fc2.weight": "model-00004-of-00004.safetensors",
158
+ "evaluator.fragments_backbone.layers.2.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
159
+ "evaluator.fragments_backbone.layers.2.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
160
+ "evaluator.fragments_backbone.layers.2.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
161
+ "evaluator.fragments_backbone.layers.2.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
162
+ "evaluator.fragments_backbone.layers.2.downsample.norm.bias": "model-00004-of-00004.safetensors",
163
+ "evaluator.fragments_backbone.layers.2.downsample.norm.weight": "model-00004-of-00004.safetensors",
164
+ "evaluator.fragments_backbone.layers.2.downsample.reduction.weight": "model-00004-of-00004.safetensors",
165
+ "evaluator.fragments_backbone.layers.3.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
166
+ "evaluator.fragments_backbone.layers.3.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
167
+ "evaluator.fragments_backbone.layers.3.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
168
+ "evaluator.fragments_backbone.layers.3.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
169
+ "evaluator.fragments_backbone.layers.3.blocks.0.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
170
+ "evaluator.fragments_backbone.layers.3.blocks.0.attn.relative_position_index": "model-00004-of-00004.safetensors",
171
+ "evaluator.fragments_backbone.layers.3.blocks.0.mlp.fc1.bias": "model-00004-of-00004.safetensors",
172
+ "evaluator.fragments_backbone.layers.3.blocks.0.mlp.fc1.weight": "model-00004-of-00004.safetensors",
173
+ "evaluator.fragments_backbone.layers.3.blocks.0.mlp.fc2.bias": "model-00004-of-00004.safetensors",
174
+ "evaluator.fragments_backbone.layers.3.blocks.0.mlp.fc2.weight": "model-00004-of-00004.safetensors",
175
+ "evaluator.fragments_backbone.layers.3.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
176
+ "evaluator.fragments_backbone.layers.3.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
177
+ "evaluator.fragments_backbone.layers.3.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
178
+ "evaluator.fragments_backbone.layers.3.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
179
+ "evaluator.fragments_backbone.layers.3.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
180
+ "evaluator.fragments_backbone.layers.3.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
181
+ "evaluator.fragments_backbone.layers.3.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
182
+ "evaluator.fragments_backbone.layers.3.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
183
+ "evaluator.fragments_backbone.layers.3.blocks.1.attn.relative_position_bias_table": "model-00004-of-00004.safetensors",
184
+ "evaluator.fragments_backbone.layers.3.blocks.1.attn.relative_position_index": "model-00004-of-00004.safetensors",
185
+ "evaluator.fragments_backbone.layers.3.blocks.1.mlp.fc1.bias": "model-00004-of-00004.safetensors",
186
+ "evaluator.fragments_backbone.layers.3.blocks.1.mlp.fc1.weight": "model-00004-of-00004.safetensors",
187
+ "evaluator.fragments_backbone.layers.3.blocks.1.mlp.fc2.bias": "model-00004-of-00004.safetensors",
188
+ "evaluator.fragments_backbone.layers.3.blocks.1.mlp.fc2.weight": "model-00004-of-00004.safetensors",
189
+ "evaluator.fragments_backbone.layers.3.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
190
+ "evaluator.fragments_backbone.layers.3.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
191
+ "evaluator.fragments_backbone.layers.3.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
192
+ "evaluator.fragments_backbone.layers.3.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
193
+ "evaluator.fragments_backbone.norm.bias": "model-00004-of-00004.safetensors",
194
+ "evaluator.fragments_backbone.norm.weight": "model-00004-of-00004.safetensors",
195
+ "evaluator.fragments_backbone.patch_embed.norm.bias": "model-00004-of-00004.safetensors",
196
+ "evaluator.fragments_backbone.patch_embed.norm.weight": "model-00004-of-00004.safetensors",
197
+ "evaluator.fragments_backbone.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
198
+ "evaluator.fragments_backbone.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
199
+ "fast_mlp.0.bias": "model-00004-of-00004.safetensors",
200
+ "fast_mlp.0.weight": "model-00004-of-00004.safetensors",
201
+ "fast_mlp.1.bias": "model-00004-of-00004.safetensors",
202
+ "fast_mlp.1.weight": "model-00004-of-00004.safetensors",
203
+ "fast_mlp.3.bias": "model-00004-of-00004.safetensors",
204
+ "fast_mlp.3.weight": "model-00004-of-00004.safetensors",
205
+ "language_model.model.layers.0.attention.wo.weight": "model-00001-of-00004.safetensors",
206
+ "language_model.model.layers.0.attention.wqkv.weight": "model-00001-of-00004.safetensors",
207
+ "language_model.model.layers.0.attention_norm.weight": "model-00001-of-00004.safetensors",
208
+ "language_model.model.layers.0.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
209
+ "language_model.model.layers.0.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
210
+ "language_model.model.layers.0.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
211
+ "language_model.model.layers.0.ffn_norm.weight": "model-00001-of-00004.safetensors",
212
+ "language_model.model.layers.1.attention.wo.weight": "model-00001-of-00004.safetensors",
213
+ "language_model.model.layers.1.attention.wqkv.weight": "model-00001-of-00004.safetensors",
214
+ "language_model.model.layers.1.attention_norm.weight": "model-00001-of-00004.safetensors",
215
+ "language_model.model.layers.1.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
216
+ "language_model.model.layers.1.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
217
+ "language_model.model.layers.1.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
218
+ "language_model.model.layers.1.ffn_norm.weight": "model-00001-of-00004.safetensors",
219
+ "language_model.model.layers.10.attention.wo.weight": "model-00001-of-00004.safetensors",
220
+ "language_model.model.layers.10.attention.wqkv.weight": "model-00001-of-00004.safetensors",
221
+ "language_model.model.layers.10.attention_norm.weight": "model-00002-of-00004.safetensors",
222
+ "language_model.model.layers.10.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
223
+ "language_model.model.layers.10.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
224
+ "language_model.model.layers.10.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
225
+ "language_model.model.layers.10.ffn_norm.weight": "model-00002-of-00004.safetensors",
226
+ "language_model.model.layers.11.attention.wo.weight": "model-00002-of-00004.safetensors",
227
+ "language_model.model.layers.11.attention.wqkv.weight": "model-00002-of-00004.safetensors",
228
+ "language_model.model.layers.11.attention_norm.weight": "model-00002-of-00004.safetensors",
229
+ "language_model.model.layers.11.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
230
+ "language_model.model.layers.11.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
231
+ "language_model.model.layers.11.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
232
+ "language_model.model.layers.11.ffn_norm.weight": "model-00002-of-00004.safetensors",
233
+ "language_model.model.layers.12.attention.wo.weight": "model-00002-of-00004.safetensors",
234
+ "language_model.model.layers.12.attention.wqkv.weight": "model-00002-of-00004.safetensors",
235
+ "language_model.model.layers.12.attention_norm.weight": "model-00002-of-00004.safetensors",
236
+ "language_model.model.layers.12.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
237
+ "language_model.model.layers.12.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
238
+ "language_model.model.layers.12.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
239
+ "language_model.model.layers.12.ffn_norm.weight": "model-00002-of-00004.safetensors",
240
+ "language_model.model.layers.13.attention.wo.weight": "model-00002-of-00004.safetensors",
241
+ "language_model.model.layers.13.attention.wqkv.weight": "model-00002-of-00004.safetensors",
242
+ "language_model.model.layers.13.attention_norm.weight": "model-00002-of-00004.safetensors",
243
+ "language_model.model.layers.13.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
244
+ "language_model.model.layers.13.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
245
+ "language_model.model.layers.13.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
246
+ "language_model.model.layers.13.ffn_norm.weight": "model-00002-of-00004.safetensors",
247
+ "language_model.model.layers.14.attention.wo.weight": "model-00002-of-00004.safetensors",
248
+ "language_model.model.layers.14.attention.wqkv.weight": "model-00002-of-00004.safetensors",
249
+ "language_model.model.layers.14.attention_norm.weight": "model-00002-of-00004.safetensors",
250
+ "language_model.model.layers.14.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
251
+ "language_model.model.layers.14.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
252
+ "language_model.model.layers.14.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
253
+ "language_model.model.layers.14.ffn_norm.weight": "model-00002-of-00004.safetensors",
254
+ "language_model.model.layers.15.attention.wo.weight": "model-00002-of-00004.safetensors",
255
+ "language_model.model.layers.15.attention.wqkv.weight": "model-00002-of-00004.safetensors",
256
+ "language_model.model.layers.15.attention_norm.weight": "model-00002-of-00004.safetensors",
257
+ "language_model.model.layers.15.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
258
+ "language_model.model.layers.15.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
259
+ "language_model.model.layers.15.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
260
+ "language_model.model.layers.15.ffn_norm.weight": "model-00002-of-00004.safetensors",
261
+ "language_model.model.layers.16.attention.wo.weight": "model-00002-of-00004.safetensors",
262
+ "language_model.model.layers.16.attention.wqkv.weight": "model-00002-of-00004.safetensors",
263
+ "language_model.model.layers.16.attention_norm.weight": "model-00002-of-00004.safetensors",
264
+ "language_model.model.layers.16.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
265
+ "language_model.model.layers.16.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
266
+ "language_model.model.layers.16.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
267
+ "language_model.model.layers.16.ffn_norm.weight": "model-00002-of-00004.safetensors",
268
+ "language_model.model.layers.17.attention.wo.weight": "model-00002-of-00004.safetensors",
269
+ "language_model.model.layers.17.attention.wqkv.weight": "model-00002-of-00004.safetensors",
270
+ "language_model.model.layers.17.attention_norm.weight": "model-00002-of-00004.safetensors",
271
+ "language_model.model.layers.17.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
272
+ "language_model.model.layers.17.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
273
+ "language_model.model.layers.17.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
274
+ "language_model.model.layers.17.ffn_norm.weight": "model-00002-of-00004.safetensors",
275
+ "language_model.model.layers.18.attention.wo.weight": "model-00002-of-00004.safetensors",
276
+ "language_model.model.layers.18.attention.wqkv.weight": "model-00002-of-00004.safetensors",
277
+ "language_model.model.layers.18.attention_norm.weight": "model-00002-of-00004.safetensors",
278
+ "language_model.model.layers.18.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
279
+ "language_model.model.layers.18.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
280
+ "language_model.model.layers.18.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
281
+ "language_model.model.layers.18.ffn_norm.weight": "model-00002-of-00004.safetensors",
282
+ "language_model.model.layers.19.attention.wo.weight": "model-00002-of-00004.safetensors",
283
+ "language_model.model.layers.19.attention.wqkv.weight": "model-00002-of-00004.safetensors",
284
+ "language_model.model.layers.19.attention_norm.weight": "model-00002-of-00004.safetensors",
285
+ "language_model.model.layers.19.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
286
+ "language_model.model.layers.19.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
287
+ "language_model.model.layers.19.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
288
+ "language_model.model.layers.19.ffn_norm.weight": "model-00002-of-00004.safetensors",
289
+ "language_model.model.layers.2.attention.wo.weight": "model-00001-of-00004.safetensors",
290
+ "language_model.model.layers.2.attention.wqkv.weight": "model-00001-of-00004.safetensors",
291
+ "language_model.model.layers.2.attention_norm.weight": "model-00001-of-00004.safetensors",
292
+ "language_model.model.layers.2.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
293
+ "language_model.model.layers.2.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
294
+ "language_model.model.layers.2.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
295
+ "language_model.model.layers.2.ffn_norm.weight": "model-00001-of-00004.safetensors",
296
+ "language_model.model.layers.20.attention.wo.weight": "model-00002-of-00004.safetensors",
297
+ "language_model.model.layers.20.attention.wqkv.weight": "model-00002-of-00004.safetensors",
298
+ "language_model.model.layers.20.attention_norm.weight": "model-00002-of-00004.safetensors",
299
+ "language_model.model.layers.20.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
300
+ "language_model.model.layers.20.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
301
+ "language_model.model.layers.20.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
302
+ "language_model.model.layers.20.ffn_norm.weight": "model-00002-of-00004.safetensors",
303
+ "language_model.model.layers.21.attention.wo.weight": "model-00002-of-00004.safetensors",
304
+ "language_model.model.layers.21.attention.wqkv.weight": "model-00002-of-00004.safetensors",
305
+ "language_model.model.layers.21.attention_norm.weight": "model-00002-of-00004.safetensors",
306
+ "language_model.model.layers.21.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
307
+ "language_model.model.layers.21.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
308
+ "language_model.model.layers.21.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
309
+ "language_model.model.layers.21.ffn_norm.weight": "model-00002-of-00004.safetensors",
310
+ "language_model.model.layers.22.attention.wo.weight": "model-00002-of-00004.safetensors",
311
+ "language_model.model.layers.22.attention.wqkv.weight": "model-00002-of-00004.safetensors",
312
+ "language_model.model.layers.22.attention_norm.weight": "model-00002-of-00004.safetensors",
313
+ "language_model.model.layers.22.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
314
+ "language_model.model.layers.22.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
315
+ "language_model.model.layers.22.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
316
+ "language_model.model.layers.22.ffn_norm.weight": "model-00002-of-00004.safetensors",
317
+ "language_model.model.layers.23.attention.wo.weight": "model-00002-of-00004.safetensors",
318
+ "language_model.model.layers.23.attention.wqkv.weight": "model-00002-of-00004.safetensors",
319
+ "language_model.model.layers.23.attention_norm.weight": "model-00002-of-00004.safetensors",
320
+ "language_model.model.layers.23.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
321
+ "language_model.model.layers.23.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
322
+ "language_model.model.layers.23.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
323
+ "language_model.model.layers.23.ffn_norm.weight": "model-00002-of-00004.safetensors",
324
+ "language_model.model.layers.24.attention.wo.weight": "model-00002-of-00004.safetensors",
325
+ "language_model.model.layers.24.attention.wqkv.weight": "model-00002-of-00004.safetensors",
326
+ "language_model.model.layers.24.attention_norm.weight": "model-00002-of-00004.safetensors",
327
+ "language_model.model.layers.24.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
328
+ "language_model.model.layers.24.feed_forward.w2.weight": "model-00002-of-00004.safetensors",
329
+ "language_model.model.layers.24.feed_forward.w3.weight": "model-00002-of-00004.safetensors",
330
+ "language_model.model.layers.24.ffn_norm.weight": "model-00002-of-00004.safetensors",
331
+ "language_model.model.layers.25.attention.wo.weight": "model-00002-of-00004.safetensors",
332
+ "language_model.model.layers.25.attention.wqkv.weight": "model-00002-of-00004.safetensors",
333
+ "language_model.model.layers.25.attention_norm.weight": "model-00003-of-00004.safetensors",
334
+ "language_model.model.layers.25.feed_forward.w1.weight": "model-00002-of-00004.safetensors",
335
+ "language_model.model.layers.25.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
336
+ "language_model.model.layers.25.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
337
+ "language_model.model.layers.25.ffn_norm.weight": "model-00003-of-00004.safetensors",
338
+ "language_model.model.layers.26.attention.wo.weight": "model-00003-of-00004.safetensors",
339
+ "language_model.model.layers.26.attention.wqkv.weight": "model-00003-of-00004.safetensors",
340
+ "language_model.model.layers.26.attention_norm.weight": "model-00003-of-00004.safetensors",
341
+ "language_model.model.layers.26.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
342
+ "language_model.model.layers.26.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
343
+ "language_model.model.layers.26.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
344
+ "language_model.model.layers.26.ffn_norm.weight": "model-00003-of-00004.safetensors",
345
+ "language_model.model.layers.27.attention.wo.weight": "model-00003-of-00004.safetensors",
346
+ "language_model.model.layers.27.attention.wqkv.weight": "model-00003-of-00004.safetensors",
347
+ "language_model.model.layers.27.attention_norm.weight": "model-00003-of-00004.safetensors",
348
+ "language_model.model.layers.27.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
349
+ "language_model.model.layers.27.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
350
+ "language_model.model.layers.27.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
351
+ "language_model.model.layers.27.ffn_norm.weight": "model-00003-of-00004.safetensors",
352
+ "language_model.model.layers.28.attention.wo.weight": "model-00003-of-00004.safetensors",
353
+ "language_model.model.layers.28.attention.wqkv.weight": "model-00003-of-00004.safetensors",
354
+ "language_model.model.layers.28.attention_norm.weight": "model-00003-of-00004.safetensors",
355
+ "language_model.model.layers.28.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
356
+ "language_model.model.layers.28.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
357
+ "language_model.model.layers.28.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
358
+ "language_model.model.layers.28.ffn_norm.weight": "model-00003-of-00004.safetensors",
359
+ "language_model.model.layers.29.attention.wo.weight": "model-00003-of-00004.safetensors",
360
+ "language_model.model.layers.29.attention.wqkv.weight": "model-00003-of-00004.safetensors",
361
+ "language_model.model.layers.29.attention_norm.weight": "model-00003-of-00004.safetensors",
362
+ "language_model.model.layers.29.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
363
+ "language_model.model.layers.29.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
364
+ "language_model.model.layers.29.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
365
+ "language_model.model.layers.29.ffn_norm.weight": "model-00003-of-00004.safetensors",
366
+ "language_model.model.layers.3.attention.wo.weight": "model-00001-of-00004.safetensors",
367
+ "language_model.model.layers.3.attention.wqkv.weight": "model-00001-of-00004.safetensors",
368
+ "language_model.model.layers.3.attention_norm.weight": "model-00001-of-00004.safetensors",
369
+ "language_model.model.layers.3.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
370
+ "language_model.model.layers.3.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
371
+ "language_model.model.layers.3.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
372
+ "language_model.model.layers.3.ffn_norm.weight": "model-00001-of-00004.safetensors",
373
+ "language_model.model.layers.30.attention.wo.weight": "model-00003-of-00004.safetensors",
374
+ "language_model.model.layers.30.attention.wqkv.weight": "model-00003-of-00004.safetensors",
375
+ "language_model.model.layers.30.attention_norm.weight": "model-00003-of-00004.safetensors",
376
+ "language_model.model.layers.30.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
377
+ "language_model.model.layers.30.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
378
+ "language_model.model.layers.30.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
379
+ "language_model.model.layers.30.ffn_norm.weight": "model-00003-of-00004.safetensors",
380
+ "language_model.model.layers.31.attention.wo.weight": "model-00003-of-00004.safetensors",
381
+ "language_model.model.layers.31.attention.wqkv.weight": "model-00003-of-00004.safetensors",
382
+ "language_model.model.layers.31.attention_norm.weight": "model-00003-of-00004.safetensors",
383
+ "language_model.model.layers.31.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
384
+ "language_model.model.layers.31.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
385
+ "language_model.model.layers.31.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
386
+ "language_model.model.layers.31.ffn_norm.weight": "model-00003-of-00004.safetensors",
387
+ "language_model.model.layers.32.attention.wo.weight": "model-00003-of-00004.safetensors",
388
+ "language_model.model.layers.32.attention.wqkv.weight": "model-00003-of-00004.safetensors",
389
+ "language_model.model.layers.32.attention_norm.weight": "model-00003-of-00004.safetensors",
390
+ "language_model.model.layers.32.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
391
+ "language_model.model.layers.32.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
392
+ "language_model.model.layers.32.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
393
+ "language_model.model.layers.32.ffn_norm.weight": "model-00003-of-00004.safetensors",
394
+ "language_model.model.layers.33.attention.wo.weight": "model-00003-of-00004.safetensors",
395
+ "language_model.model.layers.33.attention.wqkv.weight": "model-00003-of-00004.safetensors",
396
+ "language_model.model.layers.33.attention_norm.weight": "model-00003-of-00004.safetensors",
397
+ "language_model.model.layers.33.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
398
+ "language_model.model.layers.33.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
399
+ "language_model.model.layers.33.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
400
+ "language_model.model.layers.33.ffn_norm.weight": "model-00003-of-00004.safetensors",
401
+ "language_model.model.layers.34.attention.wo.weight": "model-00003-of-00004.safetensors",
402
+ "language_model.model.layers.34.attention.wqkv.weight": "model-00003-of-00004.safetensors",
403
+ "language_model.model.layers.34.attention_norm.weight": "model-00003-of-00004.safetensors",
404
+ "language_model.model.layers.34.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
405
+ "language_model.model.layers.34.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
406
+ "language_model.model.layers.34.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
407
+ "language_model.model.layers.34.ffn_norm.weight": "model-00003-of-00004.safetensors",
408
+ "language_model.model.layers.35.attention.wo.weight": "model-00003-of-00004.safetensors",
409
+ "language_model.model.layers.35.attention.wqkv.weight": "model-00003-of-00004.safetensors",
410
+ "language_model.model.layers.35.attention_norm.weight": "model-00003-of-00004.safetensors",
411
+ "language_model.model.layers.35.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
412
+ "language_model.model.layers.35.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
413
+ "language_model.model.layers.35.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
414
+ "language_model.model.layers.35.ffn_norm.weight": "model-00003-of-00004.safetensors",
415
+ "language_model.model.layers.36.attention.wo.weight": "model-00003-of-00004.safetensors",
416
+ "language_model.model.layers.36.attention.wqkv.weight": "model-00003-of-00004.safetensors",
417
+ "language_model.model.layers.36.attention_norm.weight": "model-00003-of-00004.safetensors",
418
+ "language_model.model.layers.36.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
419
+ "language_model.model.layers.36.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
420
+ "language_model.model.layers.36.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
421
+ "language_model.model.layers.36.ffn_norm.weight": "model-00003-of-00004.safetensors",
422
+ "language_model.model.layers.37.attention.wo.weight": "model-00003-of-00004.safetensors",
423
+ "language_model.model.layers.37.attention.wqkv.weight": "model-00003-of-00004.safetensors",
424
+ "language_model.model.layers.37.attention_norm.weight": "model-00003-of-00004.safetensors",
425
+ "language_model.model.layers.37.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
426
+ "language_model.model.layers.37.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
427
+ "language_model.model.layers.37.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
428
+ "language_model.model.layers.37.ffn_norm.weight": "model-00003-of-00004.safetensors",
429
+ "language_model.model.layers.38.attention.wo.weight": "model-00003-of-00004.safetensors",
430
+ "language_model.model.layers.38.attention.wqkv.weight": "model-00003-of-00004.safetensors",
431
+ "language_model.model.layers.38.attention_norm.weight": "model-00003-of-00004.safetensors",
432
+ "language_model.model.layers.38.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
433
+ "language_model.model.layers.38.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
434
+ "language_model.model.layers.38.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
435
+ "language_model.model.layers.38.ffn_norm.weight": "model-00003-of-00004.safetensors",
436
+ "language_model.model.layers.39.attention.wo.weight": "model-00003-of-00004.safetensors",
437
+ "language_model.model.layers.39.attention.wqkv.weight": "model-00003-of-00004.safetensors",
438
+ "language_model.model.layers.39.attention_norm.weight": "model-00003-of-00004.safetensors",
439
+ "language_model.model.layers.39.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
440
+ "language_model.model.layers.39.feed_forward.w2.weight": "model-00003-of-00004.safetensors",
441
+ "language_model.model.layers.39.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
442
+ "language_model.model.layers.39.ffn_norm.weight": "model-00003-of-00004.safetensors",
443
+ "language_model.model.layers.4.attention.wo.weight": "model-00001-of-00004.safetensors",
444
+ "language_model.model.layers.4.attention.wqkv.weight": "model-00001-of-00004.safetensors",
445
+ "language_model.model.layers.4.attention_norm.weight": "model-00001-of-00004.safetensors",
446
+ "language_model.model.layers.4.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
447
+ "language_model.model.layers.4.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
448
+ "language_model.model.layers.4.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
449
+ "language_model.model.layers.4.ffn_norm.weight": "model-00001-of-00004.safetensors",
450
+ "language_model.model.layers.40.attention.wo.weight": "model-00003-of-00004.safetensors",
451
+ "language_model.model.layers.40.attention.wqkv.weight": "model-00003-of-00004.safetensors",
452
+ "language_model.model.layers.40.attention_norm.weight": "model-00004-of-00004.safetensors",
453
+ "language_model.model.layers.40.feed_forward.w1.weight": "model-00003-of-00004.safetensors",
454
+ "language_model.model.layers.40.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
455
+ "language_model.model.layers.40.feed_forward.w3.weight": "model-00003-of-00004.safetensors",
456
+ "language_model.model.layers.40.ffn_norm.weight": "model-00004-of-00004.safetensors",
457
+ "language_model.model.layers.41.attention.wo.weight": "model-00004-of-00004.safetensors",
458
+ "language_model.model.layers.41.attention.wqkv.weight": "model-00004-of-00004.safetensors",
459
+ "language_model.model.layers.41.attention_norm.weight": "model-00004-of-00004.safetensors",
460
+ "language_model.model.layers.41.feed_forward.w1.weight": "model-00004-of-00004.safetensors",
461
+ "language_model.model.layers.41.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
462
+ "language_model.model.layers.41.feed_forward.w3.weight": "model-00004-of-00004.safetensors",
463
+ "language_model.model.layers.41.ffn_norm.weight": "model-00004-of-00004.safetensors",
464
+ "language_model.model.layers.42.attention.wo.weight": "model-00004-of-00004.safetensors",
465
+ "language_model.model.layers.42.attention.wqkv.weight": "model-00004-of-00004.safetensors",
466
+ "language_model.model.layers.42.attention_norm.weight": "model-00004-of-00004.safetensors",
467
+ "language_model.model.layers.42.feed_forward.w1.weight": "model-00004-of-00004.safetensors",
468
+ "language_model.model.layers.42.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
469
+ "language_model.model.layers.42.feed_forward.w3.weight": "model-00004-of-00004.safetensors",
470
+ "language_model.model.layers.42.ffn_norm.weight": "model-00004-of-00004.safetensors",
471
+ "language_model.model.layers.43.attention.wo.weight": "model-00004-of-00004.safetensors",
472
+ "language_model.model.layers.43.attention.wqkv.weight": "model-00004-of-00004.safetensors",
473
+ "language_model.model.layers.43.attention_norm.weight": "model-00004-of-00004.safetensors",
474
+ "language_model.model.layers.43.feed_forward.w1.weight": "model-00004-of-00004.safetensors",
475
+ "language_model.model.layers.43.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
476
+ "language_model.model.layers.43.feed_forward.w3.weight": "model-00004-of-00004.safetensors",
477
+ "language_model.model.layers.43.ffn_norm.weight": "model-00004-of-00004.safetensors",
478
+ "language_model.model.layers.44.attention.wo.weight": "model-00004-of-00004.safetensors",
479
+ "language_model.model.layers.44.attention.wqkv.weight": "model-00004-of-00004.safetensors",
480
+ "language_model.model.layers.44.attention_norm.weight": "model-00004-of-00004.safetensors",
481
+ "language_model.model.layers.44.feed_forward.w1.weight": "model-00004-of-00004.safetensors",
482
+ "language_model.model.layers.44.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
483
+ "language_model.model.layers.44.feed_forward.w3.weight": "model-00004-of-00004.safetensors",
484
+ "language_model.model.layers.44.ffn_norm.weight": "model-00004-of-00004.safetensors",
485
+ "language_model.model.layers.45.attention.wo.weight": "model-00004-of-00004.safetensors",
486
+ "language_model.model.layers.45.attention.wqkv.weight": "model-00004-of-00004.safetensors",
487
+ "language_model.model.layers.45.attention_norm.weight": "model-00004-of-00004.safetensors",
488
+ "language_model.model.layers.45.feed_forward.w1.weight": "model-00004-of-00004.safetensors",
489
+ "language_model.model.layers.45.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
490
+ "language_model.model.layers.45.feed_forward.w3.weight": "model-00004-of-00004.safetensors",
491
+ "language_model.model.layers.45.ffn_norm.weight": "model-00004-of-00004.safetensors",
492
+ "language_model.model.layers.46.attention.wo.weight": "model-00004-of-00004.safetensors",
493
+ "language_model.model.layers.46.attention.wqkv.weight": "model-00004-of-00004.safetensors",
494
+ "language_model.model.layers.46.attention_norm.weight": "model-00004-of-00004.safetensors",
495
+ "language_model.model.layers.46.feed_forward.w1.weight": "model-00004-of-00004.safetensors",
496
+ "language_model.model.layers.46.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
497
+ "language_model.model.layers.46.feed_forward.w3.weight": "model-00004-of-00004.safetensors",
498
+ "language_model.model.layers.46.ffn_norm.weight": "model-00004-of-00004.safetensors",
499
+ "language_model.model.layers.47.attention.wo.weight": "model-00004-of-00004.safetensors",
500
+ "language_model.model.layers.47.attention.wqkv.weight": "model-00004-of-00004.safetensors",
501
+ "language_model.model.layers.47.attention_norm.weight": "model-00004-of-00004.safetensors",
502
+ "language_model.model.layers.47.feed_forward.w1.weight": "model-00004-of-00004.safetensors",
503
+ "language_model.model.layers.47.feed_forward.w2.weight": "model-00004-of-00004.safetensors",
504
+ "language_model.model.layers.47.feed_forward.w3.weight": "model-00004-of-00004.safetensors",
505
+ "language_model.model.layers.47.ffn_norm.weight": "model-00004-of-00004.safetensors",
506
+ "language_model.model.layers.5.attention.wo.weight": "model-00001-of-00004.safetensors",
507
+ "language_model.model.layers.5.attention.wqkv.weight": "model-00001-of-00004.safetensors",
508
+ "language_model.model.layers.5.attention_norm.weight": "model-00001-of-00004.safetensors",
509
+ "language_model.model.layers.5.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
510
+ "language_model.model.layers.5.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
511
+ "language_model.model.layers.5.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
512
+ "language_model.model.layers.5.ffn_norm.weight": "model-00001-of-00004.safetensors",
513
+ "language_model.model.layers.6.attention.wo.weight": "model-00001-of-00004.safetensors",
514
+ "language_model.model.layers.6.attention.wqkv.weight": "model-00001-of-00004.safetensors",
515
+ "language_model.model.layers.6.attention_norm.weight": "model-00001-of-00004.safetensors",
516
+ "language_model.model.layers.6.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
517
+ "language_model.model.layers.6.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
518
+ "language_model.model.layers.6.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
519
+ "language_model.model.layers.6.ffn_norm.weight": "model-00001-of-00004.safetensors",
520
+ "language_model.model.layers.7.attention.wo.weight": "model-00001-of-00004.safetensors",
521
+ "language_model.model.layers.7.attention.wqkv.weight": "model-00001-of-00004.safetensors",
522
+ "language_model.model.layers.7.attention_norm.weight": "model-00001-of-00004.safetensors",
523
+ "language_model.model.layers.7.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
524
+ "language_model.model.layers.7.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
525
+ "language_model.model.layers.7.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
526
+ "language_model.model.layers.7.ffn_norm.weight": "model-00001-of-00004.safetensors",
527
+ "language_model.model.layers.8.attention.wo.weight": "model-00001-of-00004.safetensors",
528
+ "language_model.model.layers.8.attention.wqkv.weight": "model-00001-of-00004.safetensors",
529
+ "language_model.model.layers.8.attention_norm.weight": "model-00001-of-00004.safetensors",
530
+ "language_model.model.layers.8.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
531
+ "language_model.model.layers.8.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
532
+ "language_model.model.layers.8.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
533
+ "language_model.model.layers.8.ffn_norm.weight": "model-00001-of-00004.safetensors",
534
+ "language_model.model.layers.9.attention.wo.weight": "model-00001-of-00004.safetensors",
535
+ "language_model.model.layers.9.attention.wqkv.weight": "model-00001-of-00004.safetensors",
536
+ "language_model.model.layers.9.attention_norm.weight": "model-00001-of-00004.safetensors",
537
+ "language_model.model.layers.9.feed_forward.w1.weight": "model-00001-of-00004.safetensors",
538
+ "language_model.model.layers.9.feed_forward.w2.weight": "model-00001-of-00004.safetensors",
539
+ "language_model.model.layers.9.feed_forward.w3.weight": "model-00001-of-00004.safetensors",
540
+ "language_model.model.layers.9.ffn_norm.weight": "model-00001-of-00004.safetensors",
541
+ "language_model.model.norm.weight": "model-00004-of-00004.safetensors",
542
+ "language_model.model.tok_embeddings.weight": "model-00001-of-00004.safetensors",
543
+ "language_model.output.weight": "model-00004-of-00004.safetensors",
544
+ "mlp1.0.bias": "model-00004-of-00004.safetensors",
545
+ "mlp1.0.weight": "model-00004-of-00004.safetensors",
546
+ "mlp1.1.bias": "model-00004-of-00004.safetensors",
547
+ "mlp1.1.weight": "model-00004-of-00004.safetensors",
548
+ "mlp1.3.bias": "model-00004-of-00004.safetensors",
549
+ "mlp1.3.weight": "model-00004-of-00004.safetensors",
550
+ "vision_model.embeddings.class_embedding": "model-00001-of-00004.safetensors",
551
+ "vision_model.embeddings.patch_embedding.bias": "model-00001-of-00004.safetensors",
552
+ "vision_model.embeddings.patch_embedding.weight": "model-00001-of-00004.safetensors",
553
+ "vision_model.embeddings.position_embedding": "model-00001-of-00004.safetensors",
554
+ "vision_model.encoder.layers.0.attn.proj.bias": "model-00001-of-00004.safetensors",
555
+ "vision_model.encoder.layers.0.attn.proj.weight": "model-00001-of-00004.safetensors",
556
+ "vision_model.encoder.layers.0.attn.qkv.bias": "model-00001-of-00004.safetensors",
557
+ "vision_model.encoder.layers.0.attn.qkv.weight": "model-00001-of-00004.safetensors",
558
+ "vision_model.encoder.layers.0.ls1": "model-00001-of-00004.safetensors",
559
+ "vision_model.encoder.layers.0.ls2": "model-00001-of-00004.safetensors",
560
+ "vision_model.encoder.layers.0.mlp.fc1.bias": "model-00001-of-00004.safetensors",
561
+ "vision_model.encoder.layers.0.mlp.fc1.weight": "model-00001-of-00004.safetensors",
562
+ "vision_model.encoder.layers.0.mlp.fc2.bias": "model-00001-of-00004.safetensors",
563
+ "vision_model.encoder.layers.0.mlp.fc2.weight": "model-00001-of-00004.safetensors",
564
+ "vision_model.encoder.layers.0.norm1.bias": "model-00001-of-00004.safetensors",
565
+ "vision_model.encoder.layers.0.norm1.weight": "model-00001-of-00004.safetensors",
566
+ "vision_model.encoder.layers.0.norm2.bias": "model-00001-of-00004.safetensors",
567
+ "vision_model.encoder.layers.0.norm2.weight": "model-00001-of-00004.safetensors",
568
+ "vision_model.encoder.layers.1.attn.proj.bias": "model-00001-of-00004.safetensors",
569
+ "vision_model.encoder.layers.1.attn.proj.weight": "model-00001-of-00004.safetensors",
570
+ "vision_model.encoder.layers.1.attn.qkv.bias": "model-00001-of-00004.safetensors",
571
+ "vision_model.encoder.layers.1.attn.qkv.weight": "model-00001-of-00004.safetensors",
572
+ "vision_model.encoder.layers.1.ls1": "model-00001-of-00004.safetensors",
573
+ "vision_model.encoder.layers.1.ls2": "model-00001-of-00004.safetensors",
574
+ "vision_model.encoder.layers.1.mlp.fc1.bias": "model-00001-of-00004.safetensors",
575
+ "vision_model.encoder.layers.1.mlp.fc1.weight": "model-00001-of-00004.safetensors",
576
+ "vision_model.encoder.layers.1.mlp.fc2.bias": "model-00001-of-00004.safetensors",
577
+ "vision_model.encoder.layers.1.mlp.fc2.weight": "model-00001-of-00004.safetensors",
578
+ "vision_model.encoder.layers.1.norm1.bias": "model-00001-of-00004.safetensors",
579
+ "vision_model.encoder.layers.1.norm1.weight": "model-00001-of-00004.safetensors",
580
+ "vision_model.encoder.layers.1.norm2.bias": "model-00001-of-00004.safetensors",
581
+ "vision_model.encoder.layers.1.norm2.weight": "model-00001-of-00004.safetensors",
582
+ "vision_model.encoder.layers.10.attn.proj.bias": "model-00001-of-00004.safetensors",
583
+ "vision_model.encoder.layers.10.attn.proj.weight": "model-00001-of-00004.safetensors",
584
+ "vision_model.encoder.layers.10.attn.qkv.bias": "model-00001-of-00004.safetensors",
585
+ "vision_model.encoder.layers.10.attn.qkv.weight": "model-00001-of-00004.safetensors",
586
+ "vision_model.encoder.layers.10.ls1": "model-00001-of-00004.safetensors",
587
+ "vision_model.encoder.layers.10.ls2": "model-00001-of-00004.safetensors",
588
+ "vision_model.encoder.layers.10.mlp.fc1.bias": "model-00001-of-00004.safetensors",
589
+ "vision_model.encoder.layers.10.mlp.fc1.weight": "model-00001-of-00004.safetensors",
590
+ "vision_model.encoder.layers.10.mlp.fc2.bias": "model-00001-of-00004.safetensors",
591
+ "vision_model.encoder.layers.10.mlp.fc2.weight": "model-00001-of-00004.safetensors",
592
+ "vision_model.encoder.layers.10.norm1.bias": "model-00001-of-00004.safetensors",
593
+ "vision_model.encoder.layers.10.norm1.weight": "model-00001-of-00004.safetensors",
594
+ "vision_model.encoder.layers.10.norm2.bias": "model-00001-of-00004.safetensors",
595
+ "vision_model.encoder.layers.10.norm2.weight": "model-00001-of-00004.safetensors",
596
+ "vision_model.encoder.layers.11.attn.proj.bias": "model-00001-of-00004.safetensors",
597
+ "vision_model.encoder.layers.11.attn.proj.weight": "model-00001-of-00004.safetensors",
598
+ "vision_model.encoder.layers.11.attn.qkv.bias": "model-00001-of-00004.safetensors",
599
+ "vision_model.encoder.layers.11.attn.qkv.weight": "model-00001-of-00004.safetensors",
600
+ "vision_model.encoder.layers.11.ls1": "model-00001-of-00004.safetensors",
601
+ "vision_model.encoder.layers.11.ls2": "model-00001-of-00004.safetensors",
602
+ "vision_model.encoder.layers.11.mlp.fc1.bias": "model-00001-of-00004.safetensors",
603
+ "vision_model.encoder.layers.11.mlp.fc1.weight": "model-00001-of-00004.safetensors",
604
+ "vision_model.encoder.layers.11.mlp.fc2.bias": "model-00001-of-00004.safetensors",
605
+ "vision_model.encoder.layers.11.mlp.fc2.weight": "model-00001-of-00004.safetensors",
606
+ "vision_model.encoder.layers.11.norm1.bias": "model-00001-of-00004.safetensors",
607
+ "vision_model.encoder.layers.11.norm1.weight": "model-00001-of-00004.safetensors",
608
+ "vision_model.encoder.layers.11.norm2.bias": "model-00001-of-00004.safetensors",
609
+ "vision_model.encoder.layers.11.norm2.weight": "model-00001-of-00004.safetensors",
610
+ "vision_model.encoder.layers.12.attn.proj.bias": "model-00001-of-00004.safetensors",
611
+ "vision_model.encoder.layers.12.attn.proj.weight": "model-00001-of-00004.safetensors",
612
+ "vision_model.encoder.layers.12.attn.qkv.bias": "model-00001-of-00004.safetensors",
613
+ "vision_model.encoder.layers.12.attn.qkv.weight": "model-00001-of-00004.safetensors",
614
+ "vision_model.encoder.layers.12.ls1": "model-00001-of-00004.safetensors",
615
+ "vision_model.encoder.layers.12.ls2": "model-00001-of-00004.safetensors",
616
+ "vision_model.encoder.layers.12.mlp.fc1.bias": "model-00001-of-00004.safetensors",
617
+ "vision_model.encoder.layers.12.mlp.fc1.weight": "model-00001-of-00004.safetensors",
618
+ "vision_model.encoder.layers.12.mlp.fc2.bias": "model-00001-of-00004.safetensors",
619
+ "vision_model.encoder.layers.12.mlp.fc2.weight": "model-00001-of-00004.safetensors",
620
+ "vision_model.encoder.layers.12.norm1.bias": "model-00001-of-00004.safetensors",
621
+ "vision_model.encoder.layers.12.norm1.weight": "model-00001-of-00004.safetensors",
622
+ "vision_model.encoder.layers.12.norm2.bias": "model-00001-of-00004.safetensors",
623
+ "vision_model.encoder.layers.12.norm2.weight": "model-00001-of-00004.safetensors",
624
+ "vision_model.encoder.layers.13.attn.proj.bias": "model-00001-of-00004.safetensors",
625
+ "vision_model.encoder.layers.13.attn.proj.weight": "model-00001-of-00004.safetensors",
626
+ "vision_model.encoder.layers.13.attn.qkv.bias": "model-00001-of-00004.safetensors",
627
+ "vision_model.encoder.layers.13.attn.qkv.weight": "model-00001-of-00004.safetensors",
628
+ "vision_model.encoder.layers.13.ls1": "model-00001-of-00004.safetensors",
629
+ "vision_model.encoder.layers.13.ls2": "model-00001-of-00004.safetensors",
630
+ "vision_model.encoder.layers.13.mlp.fc1.bias": "model-00001-of-00004.safetensors",
631
+ "vision_model.encoder.layers.13.mlp.fc1.weight": "model-00001-of-00004.safetensors",
632
+ "vision_model.encoder.layers.13.mlp.fc2.bias": "model-00001-of-00004.safetensors",
633
+ "vision_model.encoder.layers.13.mlp.fc2.weight": "model-00001-of-00004.safetensors",
634
+ "vision_model.encoder.layers.13.norm1.bias": "model-00001-of-00004.safetensors",
635
+ "vision_model.encoder.layers.13.norm1.weight": "model-00001-of-00004.safetensors",
636
+ "vision_model.encoder.layers.13.norm2.bias": "model-00001-of-00004.safetensors",
637
+ "vision_model.encoder.layers.13.norm2.weight": "model-00001-of-00004.safetensors",
638
+ "vision_model.encoder.layers.14.attn.proj.bias": "model-00001-of-00004.safetensors",
639
+ "vision_model.encoder.layers.14.attn.proj.weight": "model-00001-of-00004.safetensors",
640
+ "vision_model.encoder.layers.14.attn.qkv.bias": "model-00001-of-00004.safetensors",
641
+ "vision_model.encoder.layers.14.attn.qkv.weight": "model-00001-of-00004.safetensors",
642
+ "vision_model.encoder.layers.14.ls1": "model-00001-of-00004.safetensors",
643
+ "vision_model.encoder.layers.14.ls2": "model-00001-of-00004.safetensors",
644
+ "vision_model.encoder.layers.14.mlp.fc1.bias": "model-00001-of-00004.safetensors",
645
+ "vision_model.encoder.layers.14.mlp.fc1.weight": "model-00001-of-00004.safetensors",
646
+ "vision_model.encoder.layers.14.mlp.fc2.bias": "model-00001-of-00004.safetensors",
647
+ "vision_model.encoder.layers.14.mlp.fc2.weight": "model-00001-of-00004.safetensors",
648
+ "vision_model.encoder.layers.14.norm1.bias": "model-00001-of-00004.safetensors",
649
+ "vision_model.encoder.layers.14.norm1.weight": "model-00001-of-00004.safetensors",
650
+ "vision_model.encoder.layers.14.norm2.bias": "model-00001-of-00004.safetensors",
651
+ "vision_model.encoder.layers.14.norm2.weight": "model-00001-of-00004.safetensors",
652
+ "vision_model.encoder.layers.15.attn.proj.bias": "model-00001-of-00004.safetensors",
653
+ "vision_model.encoder.layers.15.attn.proj.weight": "model-00001-of-00004.safetensors",
654
+ "vision_model.encoder.layers.15.attn.qkv.bias": "model-00001-of-00004.safetensors",
655
+ "vision_model.encoder.layers.15.attn.qkv.weight": "model-00001-of-00004.safetensors",
656
+ "vision_model.encoder.layers.15.ls1": "model-00001-of-00004.safetensors",
657
+ "vision_model.encoder.layers.15.ls2": "model-00001-of-00004.safetensors",
658
+ "vision_model.encoder.layers.15.mlp.fc1.bias": "model-00001-of-00004.safetensors",
659
+ "vision_model.encoder.layers.15.mlp.fc1.weight": "model-00001-of-00004.safetensors",
660
+ "vision_model.encoder.layers.15.mlp.fc2.bias": "model-00001-of-00004.safetensors",
661
+ "vision_model.encoder.layers.15.mlp.fc2.weight": "model-00001-of-00004.safetensors",
662
+ "vision_model.encoder.layers.15.norm1.bias": "model-00001-of-00004.safetensors",
663
+ "vision_model.encoder.layers.15.norm1.weight": "model-00001-of-00004.safetensors",
664
+ "vision_model.encoder.layers.15.norm2.bias": "model-00001-of-00004.safetensors",
665
+ "vision_model.encoder.layers.15.norm2.weight": "model-00001-of-00004.safetensors",
666
+ "vision_model.encoder.layers.16.attn.proj.bias": "model-00001-of-00004.safetensors",
667
+ "vision_model.encoder.layers.16.attn.proj.weight": "model-00001-of-00004.safetensors",
668
+ "vision_model.encoder.layers.16.attn.qkv.bias": "model-00001-of-00004.safetensors",
669
+ "vision_model.encoder.layers.16.attn.qkv.weight": "model-00001-of-00004.safetensors",
670
+ "vision_model.encoder.layers.16.ls1": "model-00001-of-00004.safetensors",
671
+ "vision_model.encoder.layers.16.ls2": "model-00001-of-00004.safetensors",
672
+ "vision_model.encoder.layers.16.mlp.fc1.bias": "model-00001-of-00004.safetensors",
673
+ "vision_model.encoder.layers.16.mlp.fc1.weight": "model-00001-of-00004.safetensors",
674
+ "vision_model.encoder.layers.16.mlp.fc2.bias": "model-00001-of-00004.safetensors",
675
+ "vision_model.encoder.layers.16.mlp.fc2.weight": "model-00001-of-00004.safetensors",
676
+ "vision_model.encoder.layers.16.norm1.bias": "model-00001-of-00004.safetensors",
677
+ "vision_model.encoder.layers.16.norm1.weight": "model-00001-of-00004.safetensors",
678
+ "vision_model.encoder.layers.16.norm2.bias": "model-00001-of-00004.safetensors",
679
+ "vision_model.encoder.layers.16.norm2.weight": "model-00001-of-00004.safetensors",
680
+ "vision_model.encoder.layers.17.attn.proj.bias": "model-00001-of-00004.safetensors",
681
+ "vision_model.encoder.layers.17.attn.proj.weight": "model-00001-of-00004.safetensors",
682
+ "vision_model.encoder.layers.17.attn.qkv.bias": "model-00001-of-00004.safetensors",
683
+ "vision_model.encoder.layers.17.attn.qkv.weight": "model-00001-of-00004.safetensors",
684
+ "vision_model.encoder.layers.17.ls1": "model-00001-of-00004.safetensors",
685
+ "vision_model.encoder.layers.17.ls2": "model-00001-of-00004.safetensors",
686
+ "vision_model.encoder.layers.17.mlp.fc1.bias": "model-00001-of-00004.safetensors",
687
+ "vision_model.encoder.layers.17.mlp.fc1.weight": "model-00001-of-00004.safetensors",
688
+ "vision_model.encoder.layers.17.mlp.fc2.bias": "model-00001-of-00004.safetensors",
689
+ "vision_model.encoder.layers.17.mlp.fc2.weight": "model-00001-of-00004.safetensors",
690
+ "vision_model.encoder.layers.17.norm1.bias": "model-00001-of-00004.safetensors",
691
+ "vision_model.encoder.layers.17.norm1.weight": "model-00001-of-00004.safetensors",
692
+ "vision_model.encoder.layers.17.norm2.bias": "model-00001-of-00004.safetensors",
693
+ "vision_model.encoder.layers.17.norm2.weight": "model-00001-of-00004.safetensors",
694
+ "vision_model.encoder.layers.18.attn.proj.bias": "model-00001-of-00004.safetensors",
695
+ "vision_model.encoder.layers.18.attn.proj.weight": "model-00001-of-00004.safetensors",
696
+ "vision_model.encoder.layers.18.attn.qkv.bias": "model-00001-of-00004.safetensors",
697
+ "vision_model.encoder.layers.18.attn.qkv.weight": "model-00001-of-00004.safetensors",
698
+ "vision_model.encoder.layers.18.ls1": "model-00001-of-00004.safetensors",
699
+ "vision_model.encoder.layers.18.ls2": "model-00001-of-00004.safetensors",
700
+ "vision_model.encoder.layers.18.mlp.fc1.bias": "model-00001-of-00004.safetensors",
701
+ "vision_model.encoder.layers.18.mlp.fc1.weight": "model-00001-of-00004.safetensors",
702
+ "vision_model.encoder.layers.18.mlp.fc2.bias": "model-00001-of-00004.safetensors",
703
+ "vision_model.encoder.layers.18.mlp.fc2.weight": "model-00001-of-00004.safetensors",
704
+ "vision_model.encoder.layers.18.norm1.bias": "model-00001-of-00004.safetensors",
705
+ "vision_model.encoder.layers.18.norm1.weight": "model-00001-of-00004.safetensors",
706
+ "vision_model.encoder.layers.18.norm2.bias": "model-00001-of-00004.safetensors",
707
+ "vision_model.encoder.layers.18.norm2.weight": "model-00001-of-00004.safetensors",
708
+ "vision_model.encoder.layers.19.attn.proj.bias": "model-00001-of-00004.safetensors",
709
+ "vision_model.encoder.layers.19.attn.proj.weight": "model-00001-of-00004.safetensors",
710
+ "vision_model.encoder.layers.19.attn.qkv.bias": "model-00001-of-00004.safetensors",
711
+ "vision_model.encoder.layers.19.attn.qkv.weight": "model-00001-of-00004.safetensors",
712
+ "vision_model.encoder.layers.19.ls1": "model-00001-of-00004.safetensors",
713
+ "vision_model.encoder.layers.19.ls2": "model-00001-of-00004.safetensors",
714
+ "vision_model.encoder.layers.19.mlp.fc1.bias": "model-00001-of-00004.safetensors",
715
+ "vision_model.encoder.layers.19.mlp.fc1.weight": "model-00001-of-00004.safetensors",
716
+ "vision_model.encoder.layers.19.mlp.fc2.bias": "model-00001-of-00004.safetensors",
717
+ "vision_model.encoder.layers.19.mlp.fc2.weight": "model-00001-of-00004.safetensors",
718
+ "vision_model.encoder.layers.19.norm1.bias": "model-00001-of-00004.safetensors",
719
+ "vision_model.encoder.layers.19.norm1.weight": "model-00001-of-00004.safetensors",
720
+ "vision_model.encoder.layers.19.norm2.bias": "model-00001-of-00004.safetensors",
721
+ "vision_model.encoder.layers.19.norm2.weight": "model-00001-of-00004.safetensors",
722
+ "vision_model.encoder.layers.2.attn.proj.bias": "model-00001-of-00004.safetensors",
723
+ "vision_model.encoder.layers.2.attn.proj.weight": "model-00001-of-00004.safetensors",
724
+ "vision_model.encoder.layers.2.attn.qkv.bias": "model-00001-of-00004.safetensors",
725
+ "vision_model.encoder.layers.2.attn.qkv.weight": "model-00001-of-00004.safetensors",
726
+ "vision_model.encoder.layers.2.ls1": "model-00001-of-00004.safetensors",
727
+ "vision_model.encoder.layers.2.ls2": "model-00001-of-00004.safetensors",
728
+ "vision_model.encoder.layers.2.mlp.fc1.bias": "model-00001-of-00004.safetensors",
729
+ "vision_model.encoder.layers.2.mlp.fc1.weight": "model-00001-of-00004.safetensors",
730
+ "vision_model.encoder.layers.2.mlp.fc2.bias": "model-00001-of-00004.safetensors",
731
+ "vision_model.encoder.layers.2.mlp.fc2.weight": "model-00001-of-00004.safetensors",
732
+ "vision_model.encoder.layers.2.norm1.bias": "model-00001-of-00004.safetensors",
733
+ "vision_model.encoder.layers.2.norm1.weight": "model-00001-of-00004.safetensors",
734
+ "vision_model.encoder.layers.2.norm2.bias": "model-00001-of-00004.safetensors",
735
+ "vision_model.encoder.layers.2.norm2.weight": "model-00001-of-00004.safetensors",
736
+ "vision_model.encoder.layers.20.attn.proj.bias": "model-00001-of-00004.safetensors",
737
+ "vision_model.encoder.layers.20.attn.proj.weight": "model-00001-of-00004.safetensors",
738
+ "vision_model.encoder.layers.20.attn.qkv.bias": "model-00001-of-00004.safetensors",
739
+ "vision_model.encoder.layers.20.attn.qkv.weight": "model-00001-of-00004.safetensors",
740
+ "vision_model.encoder.layers.20.ls1": "model-00001-of-00004.safetensors",
741
+ "vision_model.encoder.layers.20.ls2": "model-00001-of-00004.safetensors",
742
+ "vision_model.encoder.layers.20.mlp.fc1.bias": "model-00001-of-00004.safetensors",
743
+ "vision_model.encoder.layers.20.mlp.fc1.weight": "model-00001-of-00004.safetensors",
744
+ "vision_model.encoder.layers.20.mlp.fc2.bias": "model-00001-of-00004.safetensors",
745
+ "vision_model.encoder.layers.20.mlp.fc2.weight": "model-00001-of-00004.safetensors",
746
+ "vision_model.encoder.layers.20.norm1.bias": "model-00001-of-00004.safetensors",
747
+ "vision_model.encoder.layers.20.norm1.weight": "model-00001-of-00004.safetensors",
748
+ "vision_model.encoder.layers.20.norm2.bias": "model-00001-of-00004.safetensors",
749
+ "vision_model.encoder.layers.20.norm2.weight": "model-00001-of-00004.safetensors",
750
+ "vision_model.encoder.layers.21.attn.proj.bias": "model-00001-of-00004.safetensors",
751
+ "vision_model.encoder.layers.21.attn.proj.weight": "model-00001-of-00004.safetensors",
752
+ "vision_model.encoder.layers.21.attn.qkv.bias": "model-00001-of-00004.safetensors",
753
+ "vision_model.encoder.layers.21.attn.qkv.weight": "model-00001-of-00004.safetensors",
754
+ "vision_model.encoder.layers.21.ls1": "model-00001-of-00004.safetensors",
755
+ "vision_model.encoder.layers.21.ls2": "model-00001-of-00004.safetensors",
756
+ "vision_model.encoder.layers.21.mlp.fc1.bias": "model-00001-of-00004.safetensors",
757
+ "vision_model.encoder.layers.21.mlp.fc1.weight": "model-00001-of-00004.safetensors",
758
+ "vision_model.encoder.layers.21.mlp.fc2.bias": "model-00001-of-00004.safetensors",
759
+ "vision_model.encoder.layers.21.mlp.fc2.weight": "model-00001-of-00004.safetensors",
760
+ "vision_model.encoder.layers.21.norm1.bias": "model-00001-of-00004.safetensors",
761
+ "vision_model.encoder.layers.21.norm1.weight": "model-00001-of-00004.safetensors",
762
+ "vision_model.encoder.layers.21.norm2.bias": "model-00001-of-00004.safetensors",
763
+ "vision_model.encoder.layers.21.norm2.weight": "model-00001-of-00004.safetensors",
764
+ "vision_model.encoder.layers.22.attn.proj.bias": "model-00001-of-00004.safetensors",
765
+ "vision_model.encoder.layers.22.attn.proj.weight": "model-00001-of-00004.safetensors",
766
+ "vision_model.encoder.layers.22.attn.qkv.bias": "model-00001-of-00004.safetensors",
767
+ "vision_model.encoder.layers.22.attn.qkv.weight": "model-00001-of-00004.safetensors",
768
+ "vision_model.encoder.layers.22.ls1": "model-00001-of-00004.safetensors",
769
+ "vision_model.encoder.layers.22.ls2": "model-00001-of-00004.safetensors",
770
+ "vision_model.encoder.layers.22.mlp.fc1.bias": "model-00001-of-00004.safetensors",
771
+ "vision_model.encoder.layers.22.mlp.fc1.weight": "model-00001-of-00004.safetensors",
772
+ "vision_model.encoder.layers.22.mlp.fc2.bias": "model-00001-of-00004.safetensors",
773
+ "vision_model.encoder.layers.22.mlp.fc2.weight": "model-00001-of-00004.safetensors",
774
+ "vision_model.encoder.layers.22.norm1.bias": "model-00001-of-00004.safetensors",
775
+ "vision_model.encoder.layers.22.norm1.weight": "model-00001-of-00004.safetensors",
776
+ "vision_model.encoder.layers.22.norm2.bias": "model-00001-of-00004.safetensors",
777
+ "vision_model.encoder.layers.22.norm2.weight": "model-00001-of-00004.safetensors",
778
+ "vision_model.encoder.layers.23.attn.proj.bias": "model-00001-of-00004.safetensors",
779
+ "vision_model.encoder.layers.23.attn.proj.weight": "model-00001-of-00004.safetensors",
780
+ "vision_model.encoder.layers.23.attn.qkv.bias": "model-00001-of-00004.safetensors",
781
+ "vision_model.encoder.layers.23.attn.qkv.weight": "model-00001-of-00004.safetensors",
782
+ "vision_model.encoder.layers.23.ls1": "model-00001-of-00004.safetensors",
783
+ "vision_model.encoder.layers.23.ls2": "model-00001-of-00004.safetensors",
784
+ "vision_model.encoder.layers.23.mlp.fc1.bias": "model-00001-of-00004.safetensors",
785
+ "vision_model.encoder.layers.23.mlp.fc1.weight": "model-00001-of-00004.safetensors",
786
+ "vision_model.encoder.layers.23.mlp.fc2.bias": "model-00001-of-00004.safetensors",
787
+ "vision_model.encoder.layers.23.mlp.fc2.weight": "model-00001-of-00004.safetensors",
788
+ "vision_model.encoder.layers.23.norm1.bias": "model-00001-of-00004.safetensors",
789
+ "vision_model.encoder.layers.23.norm1.weight": "model-00001-of-00004.safetensors",
790
+ "vision_model.encoder.layers.23.norm2.bias": "model-00001-of-00004.safetensors",
791
+ "vision_model.encoder.layers.23.norm2.weight": "model-00001-of-00004.safetensors",
792
+ "vision_model.encoder.layers.3.attn.proj.bias": "model-00001-of-00004.safetensors",
793
+ "vision_model.encoder.layers.3.attn.proj.weight": "model-00001-of-00004.safetensors",
794
+ "vision_model.encoder.layers.3.attn.qkv.bias": "model-00001-of-00004.safetensors",
795
+ "vision_model.encoder.layers.3.attn.qkv.weight": "model-00001-of-00004.safetensors",
796
+ "vision_model.encoder.layers.3.ls1": "model-00001-of-00004.safetensors",
797
+ "vision_model.encoder.layers.3.ls2": "model-00001-of-00004.safetensors",
798
+ "vision_model.encoder.layers.3.mlp.fc1.bias": "model-00001-of-00004.safetensors",
799
+ "vision_model.encoder.layers.3.mlp.fc1.weight": "model-00001-of-00004.safetensors",
800
+ "vision_model.encoder.layers.3.mlp.fc2.bias": "model-00001-of-00004.safetensors",
801
+ "vision_model.encoder.layers.3.mlp.fc2.weight": "model-00001-of-00004.safetensors",
802
+ "vision_model.encoder.layers.3.norm1.bias": "model-00001-of-00004.safetensors",
803
+ "vision_model.encoder.layers.3.norm1.weight": "model-00001-of-00004.safetensors",
804
+ "vision_model.encoder.layers.3.norm2.bias": "model-00001-of-00004.safetensors",
805
+ "vision_model.encoder.layers.3.norm2.weight": "model-00001-of-00004.safetensors",
806
+ "vision_model.encoder.layers.4.attn.proj.bias": "model-00001-of-00004.safetensors",
807
+ "vision_model.encoder.layers.4.attn.proj.weight": "model-00001-of-00004.safetensors",
808
+ "vision_model.encoder.layers.4.attn.qkv.bias": "model-00001-of-00004.safetensors",
809
+ "vision_model.encoder.layers.4.attn.qkv.weight": "model-00001-of-00004.safetensors",
810
+ "vision_model.encoder.layers.4.ls1": "model-00001-of-00004.safetensors",
811
+ "vision_model.encoder.layers.4.ls2": "model-00001-of-00004.safetensors",
812
+ "vision_model.encoder.layers.4.mlp.fc1.bias": "model-00001-of-00004.safetensors",
813
+ "vision_model.encoder.layers.4.mlp.fc1.weight": "model-00001-of-00004.safetensors",
814
+ "vision_model.encoder.layers.4.mlp.fc2.bias": "model-00001-of-00004.safetensors",
815
+ "vision_model.encoder.layers.4.mlp.fc2.weight": "model-00001-of-00004.safetensors",
816
+ "vision_model.encoder.layers.4.norm1.bias": "model-00001-of-00004.safetensors",
817
+ "vision_model.encoder.layers.4.norm1.weight": "model-00001-of-00004.safetensors",
818
+ "vision_model.encoder.layers.4.norm2.bias": "model-00001-of-00004.safetensors",
819
+ "vision_model.encoder.layers.4.norm2.weight": "model-00001-of-00004.safetensors",
820
+ "vision_model.encoder.layers.5.attn.proj.bias": "model-00001-of-00004.safetensors",
821
+ "vision_model.encoder.layers.5.attn.proj.weight": "model-00001-of-00004.safetensors",
822
+ "vision_model.encoder.layers.5.attn.qkv.bias": "model-00001-of-00004.safetensors",
823
+ "vision_model.encoder.layers.5.attn.qkv.weight": "model-00001-of-00004.safetensors",
824
+ "vision_model.encoder.layers.5.ls1": "model-00001-of-00004.safetensors",
825
+ "vision_model.encoder.layers.5.ls2": "model-00001-of-00004.safetensors",
826
+ "vision_model.encoder.layers.5.mlp.fc1.bias": "model-00001-of-00004.safetensors",
827
+ "vision_model.encoder.layers.5.mlp.fc1.weight": "model-00001-of-00004.safetensors",
828
+ "vision_model.encoder.layers.5.mlp.fc2.bias": "model-00001-of-00004.safetensors",
829
+ "vision_model.encoder.layers.5.mlp.fc2.weight": "model-00001-of-00004.safetensors",
830
+ "vision_model.encoder.layers.5.norm1.bias": "model-00001-of-00004.safetensors",
831
+ "vision_model.encoder.layers.5.norm1.weight": "model-00001-of-00004.safetensors",
832
+ "vision_model.encoder.layers.5.norm2.bias": "model-00001-of-00004.safetensors",
833
+ "vision_model.encoder.layers.5.norm2.weight": "model-00001-of-00004.safetensors",
834
+ "vision_model.encoder.layers.6.attn.proj.bias": "model-00001-of-00004.safetensors",
835
+ "vision_model.encoder.layers.6.attn.proj.weight": "model-00001-of-00004.safetensors",
836
+ "vision_model.encoder.layers.6.attn.qkv.bias": "model-00001-of-00004.safetensors",
837
+ "vision_model.encoder.layers.6.attn.qkv.weight": "model-00001-of-00004.safetensors",
838
+ "vision_model.encoder.layers.6.ls1": "model-00001-of-00004.safetensors",
839
+ "vision_model.encoder.layers.6.ls2": "model-00001-of-00004.safetensors",
840
+ "vision_model.encoder.layers.6.mlp.fc1.bias": "model-00001-of-00004.safetensors",
841
+ "vision_model.encoder.layers.6.mlp.fc1.weight": "model-00001-of-00004.safetensors",
842
+ "vision_model.encoder.layers.6.mlp.fc2.bias": "model-00001-of-00004.safetensors",
843
+ "vision_model.encoder.layers.6.mlp.fc2.weight": "model-00001-of-00004.safetensors",
844
+ "vision_model.encoder.layers.6.norm1.bias": "model-00001-of-00004.safetensors",
845
+ "vision_model.encoder.layers.6.norm1.weight": "model-00001-of-00004.safetensors",
846
+ "vision_model.encoder.layers.6.norm2.bias": "model-00001-of-00004.safetensors",
847
+ "vision_model.encoder.layers.6.norm2.weight": "model-00001-of-00004.safetensors",
848
+ "vision_model.encoder.layers.7.attn.proj.bias": "model-00001-of-00004.safetensors",
849
+ "vision_model.encoder.layers.7.attn.proj.weight": "model-00001-of-00004.safetensors",
850
+ "vision_model.encoder.layers.7.attn.qkv.bias": "model-00001-of-00004.safetensors",
851
+ "vision_model.encoder.layers.7.attn.qkv.weight": "model-00001-of-00004.safetensors",
852
+ "vision_model.encoder.layers.7.ls1": "model-00001-of-00004.safetensors",
853
+ "vision_model.encoder.layers.7.ls2": "model-00001-of-00004.safetensors",
854
+ "vision_model.encoder.layers.7.mlp.fc1.bias": "model-00001-of-00004.safetensors",
855
+ "vision_model.encoder.layers.7.mlp.fc1.weight": "model-00001-of-00004.safetensors",
856
+ "vision_model.encoder.layers.7.mlp.fc2.bias": "model-00001-of-00004.safetensors",
857
+ "vision_model.encoder.layers.7.mlp.fc2.weight": "model-00001-of-00004.safetensors",
858
+ "vision_model.encoder.layers.7.norm1.bias": "model-00001-of-00004.safetensors",
859
+ "vision_model.encoder.layers.7.norm1.weight": "model-00001-of-00004.safetensors",
860
+ "vision_model.encoder.layers.7.norm2.bias": "model-00001-of-00004.safetensors",
861
+ "vision_model.encoder.layers.7.norm2.weight": "model-00001-of-00004.safetensors",
862
+ "vision_model.encoder.layers.8.attn.proj.bias": "model-00001-of-00004.safetensors",
863
+ "vision_model.encoder.layers.8.attn.proj.weight": "model-00001-of-00004.safetensors",
864
+ "vision_model.encoder.layers.8.attn.qkv.bias": "model-00001-of-00004.safetensors",
865
+ "vision_model.encoder.layers.8.attn.qkv.weight": "model-00001-of-00004.safetensors",
866
+ "vision_model.encoder.layers.8.ls1": "model-00001-of-00004.safetensors",
867
+ "vision_model.encoder.layers.8.ls2": "model-00001-of-00004.safetensors",
868
+ "vision_model.encoder.layers.8.mlp.fc1.bias": "model-00001-of-00004.safetensors",
869
+ "vision_model.encoder.layers.8.mlp.fc1.weight": "model-00001-of-00004.safetensors",
870
+ "vision_model.encoder.layers.8.mlp.fc2.bias": "model-00001-of-00004.safetensors",
871
+ "vision_model.encoder.layers.8.mlp.fc2.weight": "model-00001-of-00004.safetensors",
872
+ "vision_model.encoder.layers.8.norm1.bias": "model-00001-of-00004.safetensors",
873
+ "vision_model.encoder.layers.8.norm1.weight": "model-00001-of-00004.safetensors",
874
+ "vision_model.encoder.layers.8.norm2.bias": "model-00001-of-00004.safetensors",
875
+ "vision_model.encoder.layers.8.norm2.weight": "model-00001-of-00004.safetensors",
876
+ "vision_model.encoder.layers.9.attn.proj.bias": "model-00001-of-00004.safetensors",
877
+ "vision_model.encoder.layers.9.attn.proj.weight": "model-00001-of-00004.safetensors",
878
+ "vision_model.encoder.layers.9.attn.qkv.bias": "model-00001-of-00004.safetensors",
879
+ "vision_model.encoder.layers.9.attn.qkv.weight": "model-00001-of-00004.safetensors",
880
+ "vision_model.encoder.layers.9.ls1": "model-00001-of-00004.safetensors",
881
+ "vision_model.encoder.layers.9.ls2": "model-00001-of-00004.safetensors",
882
+ "vision_model.encoder.layers.9.mlp.fc1.bias": "model-00001-of-00004.safetensors",
883
+ "vision_model.encoder.layers.9.mlp.fc1.weight": "model-00001-of-00004.safetensors",
884
+ "vision_model.encoder.layers.9.mlp.fc2.bias": "model-00001-of-00004.safetensors",
885
+ "vision_model.encoder.layers.9.mlp.fc2.weight": "model-00001-of-00004.safetensors",
886
+ "vision_model.encoder.layers.9.norm1.bias": "model-00001-of-00004.safetensors",
887
+ "vision_model.encoder.layers.9.norm1.weight": "model-00001-of-00004.safetensors",
888
+ "vision_model.encoder.layers.9.norm2.bias": "model-00001-of-00004.safetensors",
889
+ "vision_model.encoder.layers.9.norm2.weight": "model-00001-of-00004.safetensors"
890
+ }
891
+ }
mos2_metrics.txt ADDED
@@ -0,0 +1,36 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Accuracy: 0.00011111111111111112
2
+ SRCC: 0.6926555307894924
3
+ PLCC: 0.7095715746195425
4
+ KRCC: 0.5665830132455586
5
+ Accuracy: 0.00011111111111111112
6
+ SRCC: 0.6926555307894924
7
+ PLCC: 0.7095715746195425
8
+ KRCC: 0.5665830132455586
9
+ Accuracy: 0.00011111111111111112
10
+ SRCC: 0.6926555307894924
11
+ PLCC: 0.7095715746195425
12
+ KRCC: 0.5665830132455586
13
+ Accuracy: 0.00011111111111111112
14
+ SRCC: 0.6931909180036439
15
+ PLCC: 0.7100735156898511
16
+ KRCC: 0.5668195962017428
17
+ Accuracy: 0.00011111111111111112
18
+ SRCC: 0.6931909180036439
19
+ PLCC: 0.7100735156898511
20
+ KRCC: 0.5668195962017428
21
+ Accuracy: 0.00011111111111111112
22
+ SRCC: 0.6931909180036439
23
+ PLCC: 0.7100735156898511
24
+ KRCC: 0.5668195962017428
25
+ Accuracy: 0.0007777777777777777
26
+ SRCC: 0.7234150036980409
27
+ PLCC: 0.7354250671184984
28
+ KRCC: 0.5842951609750242
29
+ Accuracy: 0.0007777777777777777
30
+ SRCC: 0.7234150036980409
31
+ PLCC: 0.7354250671184984
32
+ KRCC: 0.5842951609750242
33
+ Accuracy: 0.0007777777777777777
34
+ SRCC: 0.7234150036980409
35
+ PLCC: 0.7354250671184984
36
+ KRCC: 0.5842951609750242
mos2_results.csv ADDED
The diff for this file is too large to render. See raw diff
 
runs/Apr23_19-32-00_amax/events.out.tfevents.1745407931.amax.1703561.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0269a6b9124eff71542978a5c7bcaec5b2054be97473d0b4ba3f1f2af538307
3
+ size 1751260
special_tokens_map.json ADDED
@@ -0,0 +1,63 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>",
5
+ "<|action_start|>",
6
+ "<|action_end|>",
7
+ "<|interpreter|>",
8
+ "<|plugin|>",
9
+ "<restate>",
10
+ "</restate>",
11
+ "<planning>",
12
+ "</planning>",
13
+ "<recollect>",
14
+ "</recollect>",
15
+ "<execution>",
16
+ "</execution>",
17
+ "<review>",
18
+ "</review>",
19
+ "<summarize>",
20
+ "</summarize>",
21
+ "<retry>",
22
+ "</retry>",
23
+ "<conclude>",
24
+ "</conclude>",
25
+ "<img>",
26
+ "</img>",
27
+ "<IMG_CONTEXT>",
28
+ "<quad>",
29
+ "</quad>",
30
+ "<ref>",
31
+ "</ref>",
32
+ "<box>",
33
+ "</box>"
34
+ ],
35
+ "bos_token": {
36
+ "content": "<s>",
37
+ "lstrip": false,
38
+ "normalized": false,
39
+ "rstrip": false,
40
+ "single_word": false
41
+ },
42
+ "eos_token": {
43
+ "content": "<|im_end|>",
44
+ "lstrip": false,
45
+ "normalized": false,
46
+ "rstrip": false,
47
+ "single_word": false
48
+ },
49
+ "pad_token": {
50
+ "content": "</s>",
51
+ "lstrip": false,
52
+ "normalized": false,
53
+ "rstrip": false,
54
+ "single_word": false
55
+ },
56
+ "unk_token": {
57
+ "content": "<unk>",
58
+ "lstrip": false,
59
+ "normalized": false,
60
+ "rstrip": false,
61
+ "single_word": false
62
+ }
63
+ }
tokenization_internlm3.py ADDED
@@ -0,0 +1,294 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import os
2
+ from shutil import copyfile
3
+ from typing import TYPE_CHECKING, Any, Dict, List, Optional, Tuple
4
+
5
+ import sentencepiece as spm
6
+ from transformers.tokenization_utils import AddedToken, PreTrainedTokenizer
7
+ from transformers.utils import logging
8
+
9
+ if TYPE_CHECKING:
10
+ from transformers.tokenization_utils_base import TextInput
11
+
12
+ logger = logging.get_logger(__name__)
13
+
14
+ VOCAB_FILES_NAMES = {"vocab_file": "tokenizer.model"}
15
+
16
+ SPIECE_UNDERLINE = "▁"
17
+
18
+
19
+ class InternLM3Tokenizer(PreTrainedTokenizer):
20
+ """
21
+ Construct a InternLM3 tokenizer. Based on byte-level Byte-Pair-Encoding. The default padding token is unset as there is
22
+ no padding token in the original model.
23
+
24
+ Args:
25
+ vocab_file (`str`):
26
+ Path to the vocabulary file.
27
+ unk_token (`str` or `tokenizers.AddedToken`, *optional*, defaults to `"<unk>"`):
28
+ The unknown token. A token that is not in the vocabulary cannot be converted to an ID and is set to be this
29
+ token instead.
30
+ bos_token (`str` or `tokenizers.AddedToken`, *optional*, defaults to `"<s>"`):
31
+ The beginning of sequence token that was used during pretraining. Can be used a sequence classifier token.
32
+ eos_token (`str` or `tokenizers.AddedToken`, *optional*, defaults to `"</s>"`):
33
+ The end of sequence token.
34
+ pad_token (`str` or `tokenizers.AddedToken`, *optional*):
35
+ A special token used to make arrays of tokens the same size for batching purpose. Will then be ignored by
36
+ attention mechanisms or loss computation.
37
+ sp_model_kwargs (`Dict[str, Any]`, `Optional`, *optional*):
38
+ Will be passed to the `SentencePieceProcessor.__init__()` method. The [Python wrapper for
39
+ SentencePiece](https://github.com/google/sentencepiece/tree/master/python) can be used, among other things,
40
+ to set:
41
+
42
+ - `enable_sampling`: Enable subword regularization.
43
+ - `nbest_size`: Sampling parameters for unigram. Invalid for BPE-Dropout.
44
+
45
+ - `nbest_size = {0,1}`: No sampling is performed.
46
+ - `nbest_size > 1`: samples from the nbest_size results.
47
+ - `nbest_size < 0`: assuming that nbest_size is infinite and samples from the all hypothesis (lattice)
48
+ using forward-filtering-and-backward-sampling algorithm.
49
+
50
+ - `alpha`: Smoothing parameter for unigram sampling, and dropout probability of merge operations for
51
+ BPE-dropout.
52
+
53
+ add_bos_token (`bool`, *optional*, defaults to `True`):
54
+ Whether or not to add an `bos_token` at the start of sequences.
55
+ add_eos_token (`bool`, *optional*, defaults to `False`):
56
+ Whether or not to add an `eos_token` at the end of sequences.
57
+ clean_up_tokenization_spaces (`bool`, *optional*, defaults to `False`):
58
+ Whether or not to cleanup spaces after decoding, cleanup consists in removing potential artifacts like
59
+ extra spaces.
60
+ use_default_system_prompt (`bool`, *optional*, defaults to `False`):
61
+ Whether or not the default system prompt for InternLM3 should be used.
62
+ spaces_between_special_tokens (`bool`, *optional*, defaults to `False`):
63
+ Whether or not to add spaces between special tokens.
64
+ spaces_for_interleaved_special_tokens (`bool`, *optional*, defaults to `False`):
65
+ Whether or not to add spaces between special tokens that are interleaved with normal tokens.
66
+ add_prefix_space (`bool`, *optional*, defaults to `True`):
67
+ Whether or not to add an initial space to the input. This allows to treat the leading word just as any
68
+ other word. Again, this should be set with `from_slow=True` to make sure it's taken into account.
69
+ """
70
+
71
+ vocab_files_names = VOCAB_FILES_NAMES
72
+ model_input_names = ["input_ids", "attention_mask"]
73
+
74
+ def __init__(
75
+ self,
76
+ vocab_file,
77
+ unk_token="<unk>",
78
+ bos_token="<s>",
79
+ eos_token="</s>",
80
+ pad_token=None,
81
+ sp_model_kwargs: Optional[Dict[str, Any]] = None,
82
+ add_bos_token=True,
83
+ add_eos_token=False,
84
+ clean_up_tokenization_spaces=False,
85
+ use_default_system_prompt=False,
86
+ spaces_between_special_tokens=False,
87
+ spaces_for_interleaved_special_tokens=False,
88
+ add_prefix_space=True,
89
+ **kwargs,
90
+ ):
91
+ self.sp_model_kwargs = {} if sp_model_kwargs is None else sp_model_kwargs
92
+ bos_token = AddedToken(bos_token, normalized=False, special=True) if isinstance(bos_token, str) else bos_token
93
+ eos_token = AddedToken(eos_token, normalized=False, special=True) if isinstance(eos_token, str) else eos_token
94
+ unk_token = AddedToken(unk_token, normalized=False, special=True) if isinstance(unk_token, str) else unk_token
95
+ pad_token = AddedToken(pad_token, normalized=False, special=True) if isinstance(pad_token, str) else pad_token
96
+
97
+ self.vocab_file = vocab_file
98
+ self.add_bos_token = add_bos_token
99
+ self.add_eos_token = add_eos_token
100
+ self.use_default_system_prompt = use_default_system_prompt
101
+ self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
102
+ self.sp_model.Load(vocab_file)
103
+ self.add_prefix_space = add_prefix_space
104
+ self.spaces_for_interleaved_special_tokens = spaces_for_interleaved_special_tokens
105
+
106
+ vocab_size = self.sp_model.get_piece_size()
107
+ self.decoder = {i: self.sp_model.id_to_piece(i) for i in range(vocab_size)}
108
+
109
+ super().__init__(
110
+ bos_token=bos_token,
111
+ eos_token=eos_token,
112
+ unk_token=unk_token,
113
+ pad_token=pad_token,
114
+ add_bos_token=add_bos_token,
115
+ add_eos_token=add_eos_token,
116
+ sp_model_kwargs=sp_model_kwargs,
117
+ clean_up_tokenization_spaces=clean_up_tokenization_spaces,
118
+ use_default_system_prompt=use_default_system_prompt,
119
+ spaces_between_special_tokens=spaces_between_special_tokens,
120
+ add_prefix_space=add_prefix_space,
121
+ **kwargs,
122
+ )
123
+
124
+ def __getstate__(self):
125
+ state = self.__dict__.copy()
126
+ state["sp_model"] = None
127
+ state["sp_model_proto"] = self.sp_model.serialized_model_proto()
128
+ return state
129
+
130
+ def __setstate__(self, d):
131
+ self.__dict__.update(d)
132
+ self.sp_model = spm.SentencePieceProcessor(**self.sp_model_kwargs)
133
+ self.sp_model.LoadFromSerializedProto(self.sp_model_proto)
134
+
135
+ @property
136
+ def vocab_size(self):
137
+ """Returns vocab size"""
138
+ return self.sp_model.get_piece_size()
139
+
140
+ def get_vocab(self):
141
+ """Returns vocab as a dict"""
142
+ vocab = {self.convert_ids_to_tokens(i): i for i in range(self.vocab_size)}
143
+ vocab.update(self.added_tokens_encoder)
144
+ return vocab
145
+
146
+ def tokenize(self, text: "TextInput", **kwargs) -> List[str]:
147
+ """
148
+ Args:
149
+ text: TextInput
150
+ Simply calls PreTrainedTokenizer's method
151
+ """
152
+ return super().tokenize(text, **kwargs)
153
+
154
+ def _tokenize(self, text, **kwargs):
155
+ """
156
+ Args:
157
+ text: TextInput
158
+ Returns a tokenized string. The Gemma tokenizer never adds a prefix space.
159
+ """
160
+ return self.sp_model.encode(text, out_type=str)
161
+
162
+ def _convert_token_to_id(self, token):
163
+ """Converts a token (str) in an id using the vocab."""
164
+ return self.sp_model.piece_to_id(token)
165
+
166
+ def _convert_id_to_token(self, index):
167
+ """Converts an index (integer) in a token (str) using the vocab."""
168
+ return self.decoder.get(index, "")
169
+
170
+ def convert_tokens_to_string(self, tokens):
171
+ """Converts a sequence of tokens (string) in a single string."""
172
+ # since we manually add the prefix space, we have to remove it when decoding
173
+ if tokens[0].startswith(SPIECE_UNDERLINE) and self.add_prefix_space:
174
+ tokens[0] = tokens[0][1:]
175
+
176
+ current_sub_tokens = []
177
+ out_string = ""
178
+ prev_is_special = False
179
+ for i, token in enumerate(tokens):
180
+ # make sure that special tokens are not decoded using sentencepiece model
181
+ if token in self.all_special_tokens:
182
+ if not prev_is_special and i != 0 and self.spaces_for_interleaved_special_tokens:
183
+ out_string += " "
184
+ out_string += self.sp_model.decode(current_sub_tokens) + token
185
+ prev_is_special = True
186
+ current_sub_tokens = []
187
+ else:
188
+ if (
189
+ prev_is_special
190
+ and i == 1
191
+ and self.add_prefix_space
192
+ and not token.startswith(SPIECE_UNDERLINE)
193
+ and self.spaces_for_interleaved_special_tokens
194
+ ):
195
+ out_string += " "
196
+ current_sub_tokens.append(token)
197
+ prev_is_special = False
198
+ out_string += self.sp_model.decode(current_sub_tokens)
199
+ return out_string
200
+
201
+ def save_vocabulary(self, save_directory, filename_prefix: Optional[str] = None) -> Tuple[str]:
202
+ """
203
+ Save the vocabulary and special tokens file to a directory.
204
+
205
+ Args:
206
+ save_directory (`str`):
207
+ The directory in which to save the vocabulary.
208
+
209
+ Returns:
210
+ `Tuple(str)`: Paths to the files saved.
211
+ """
212
+ if not os.path.isdir(save_directory):
213
+ logger.error(f"Vocabulary path ({save_directory}) should be a directory")
214
+ return
215
+ out_vocab_file = os.path.join(save_directory, (filename_prefix + "-" if filename_prefix else "") + VOCAB_FILES_NAMES["vocab_file"])
216
+
217
+ if os.path.abspath(self.vocab_file) != os.path.abspath(out_vocab_file) and os.path.isfile(self.vocab_file):
218
+ copyfile(self.vocab_file, out_vocab_file)
219
+ elif not os.path.isfile(self.vocab_file):
220
+ with open(out_vocab_file, "wb") as fi:
221
+ content_spiece_model = self.sp_model.serialized_model_proto()
222
+ fi.write(content_spiece_model)
223
+
224
+ return (out_vocab_file,)
225
+
226
+ def build_inputs_with_special_tokens(self, token_ids_0, token_ids_1=None):
227
+ bos_token_id = [self.bos_token_id] if self.add_bos_token else []
228
+ eos_token_id = [self.eos_token_id] if self.add_eos_token else []
229
+
230
+ output = bos_token_id + token_ids_0 + eos_token_id
231
+
232
+ if token_ids_1 is not None:
233
+ output = output + bos_token_id + token_ids_1 + eos_token_id
234
+
235
+ return output
236
+
237
+ def get_special_tokens_mask(
238
+ self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None, already_has_special_tokens: bool = False
239
+ ) -> List[int]:
240
+ """
241
+ Retrieve sequence ids from a token list that has no special tokens added. This method is called when adding
242
+ special tokens using the tokenizer `prepare_for_model` method.
243
+
244
+ Args:
245
+ token_ids_0 (`List[int]`):
246
+ List of IDs.
247
+ token_ids_1 (`List[int]`, *optional*):
248
+ Optional second list of IDs for sequence pairs.
249
+ already_has_special_tokens (`bool`, *optional*, defaults to `False`):
250
+ Whether or not the token list is already formatted with special tokens for the model.
251
+
252
+ Returns:
253
+ `List[int]`: A list of integers in the range [0, 1]: 1 for a special token, 0 for a sequence token.
254
+ """
255
+ if already_has_special_tokens:
256
+ return super().get_special_tokens_mask(token_ids_0=token_ids_0, token_ids_1=token_ids_1, already_has_special_tokens=True)
257
+
258
+ bos_token_id = [1] if self.add_bos_token else []
259
+ eos_token_id = [1] if self.add_eos_token else []
260
+
261
+ if token_ids_1 is None:
262
+ return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id
263
+ return bos_token_id + ([0] * len(token_ids_0)) + eos_token_id + bos_token_id + ([0] * len(token_ids_1)) + eos_token_id
264
+
265
+ def create_token_type_ids_from_sequences(self, token_ids_0: List[int], token_ids_1: Optional[List[int]] = None) -> List[int]:
266
+ """
267
+ Creates a mask from the two sequences passed to be used in a sequence-pair classification task. An ALBERT
268
+ sequence pair mask has the following format:
269
+
270
+ ```
271
+ 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1
272
+ | first sequence | second sequence |
273
+ ```
274
+
275
+ if token_ids_1 is None, only returns the first portion of the mask (0s).
276
+
277
+ Args:
278
+ token_ids_0 (`List[int]`):
279
+ List of ids.
280
+ token_ids_1 (`List[int]`, *optional*):
281
+ Optional second list of IDs for sequence pairs.
282
+
283
+ Returns:
284
+ `List[int]`: List of [token type IDs](../glossary#token-type-ids) according to the given sequence(s).
285
+ """
286
+ bos_token_id = [self.bos_token_id] if self.add_bos_token else []
287
+ eos_token_id = [self.eos_token_id] if self.add_eos_token else []
288
+
289
+ output = [0] * len(bos_token_id + token_ids_0 + eos_token_id)
290
+
291
+ if token_ids_1 is not None:
292
+ output += [1] * len(bos_token_id + token_ids_1 + eos_token_id)
293
+
294
+ return output
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bcacff3229854f5103ee7a85473a30ca9a8b3a68f3aae9b7479574b23ac2256b
3
+ size 2475075
tokenizer_config.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": true,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "128111": {
31
+ "content": "<restate>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "128112": {
39
+ "content": "</restate>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ },
46
+ "128113": {
47
+ "content": "<planning>",
48
+ "lstrip": false,
49
+ "normalized": false,
50
+ "rstrip": false,
51
+ "single_word": false,
52
+ "special": true
53
+ },
54
+ "128114": {
55
+ "content": "</planning>",
56
+ "lstrip": false,
57
+ "normalized": false,
58
+ "rstrip": false,
59
+ "single_word": false,
60
+ "special": true
61
+ },
62
+ "128115": {
63
+ "content": "<recollect>",
64
+ "lstrip": false,
65
+ "normalized": false,
66
+ "rstrip": false,
67
+ "single_word": false,
68
+ "special": true
69
+ },
70
+ "128116": {
71
+ "content": "</recollect>",
72
+ "lstrip": false,
73
+ "normalized": false,
74
+ "rstrip": false,
75
+ "single_word": false,
76
+ "special": true
77
+ },
78
+ "128117": {
79
+ "content": "<execution>",
80
+ "lstrip": false,
81
+ "normalized": false,
82
+ "rstrip": false,
83
+ "single_word": false,
84
+ "special": true
85
+ },
86
+ "128118": {
87
+ "content": "</execution>",
88
+ "lstrip": false,
89
+ "normalized": false,
90
+ "rstrip": false,
91
+ "single_word": false,
92
+ "special": true
93
+ },
94
+ "128119": {
95
+ "content": "<review>",
96
+ "lstrip": false,
97
+ "normalized": false,
98
+ "rstrip": false,
99
+ "single_word": false,
100
+ "special": true
101
+ },
102
+ "128120": {
103
+ "content": "</review>",
104
+ "lstrip": false,
105
+ "normalized": false,
106
+ "rstrip": false,
107
+ "single_word": false,
108
+ "special": true
109
+ },
110
+ "128121": {
111
+ "content": "<summarize>",
112
+ "lstrip": false,
113
+ "normalized": false,
114
+ "rstrip": false,
115
+ "single_word": false,
116
+ "special": true
117
+ },
118
+ "128122": {
119
+ "content": "</summarize>",
120
+ "lstrip": false,
121
+ "normalized": false,
122
+ "rstrip": false,
123
+ "single_word": false,
124
+ "special": true
125
+ },
126
+ "128123": {
127
+ "content": "<retry>",
128
+ "lstrip": false,
129
+ "normalized": false,
130
+ "rstrip": false,
131
+ "single_word": false,
132
+ "special": true
133
+ },
134
+ "128124": {
135
+ "content": "</retry>",
136
+ "lstrip": false,
137
+ "normalized": false,
138
+ "rstrip": false,
139
+ "single_word": false,
140
+ "special": true
141
+ },
142
+ "128125": {
143
+ "content": "<conclude>",
144
+ "lstrip": false,
145
+ "normalized": false,
146
+ "rstrip": false,
147
+ "single_word": false,
148
+ "special": true
149
+ },
150
+ "128126": {
151
+ "content": "</conclude>",
152
+ "lstrip": false,
153
+ "normalized": false,
154
+ "rstrip": false,
155
+ "single_word": false,
156
+ "special": true
157
+ },
158
+ "128127": {
159
+ "content": "<|plugin|>",
160
+ "lstrip": false,
161
+ "normalized": false,
162
+ "rstrip": false,
163
+ "single_word": false,
164
+ "special": true
165
+ },
166
+ "128128": {
167
+ "content": "<|interpreter|>",
168
+ "lstrip": false,
169
+ "normalized": false,
170
+ "rstrip": false,
171
+ "single_word": false,
172
+ "special": true
173
+ },
174
+ "128129": {
175
+ "content": "<|action_end|>",
176
+ "lstrip": false,
177
+ "normalized": false,
178
+ "rstrip": false,
179
+ "single_word": false,
180
+ "special": true
181
+ },
182
+ "128130": {
183
+ "content": "<|action_start|>",
184
+ "lstrip": false,
185
+ "normalized": false,
186
+ "rstrip": false,
187
+ "single_word": false,
188
+ "special": true
189
+ },
190
+ "128131": {
191
+ "content": "<|im_end|>",
192
+ "lstrip": false,
193
+ "normalized": false,
194
+ "rstrip": false,
195
+ "single_word": false,
196
+ "special": true
197
+ },
198
+ "128132": {
199
+ "content": "<|im_start|>",
200
+ "lstrip": false,
201
+ "normalized": false,
202
+ "rstrip": false,
203
+ "single_word": false,
204
+ "special": true
205
+ },
206
+ "128133": {
207
+ "content": "<img>",
208
+ "lstrip": false,
209
+ "normalized": false,
210
+ "rstrip": false,
211
+ "single_word": false,
212
+ "special": true
213
+ },
214
+ "128134": {
215
+ "content": "</img>",
216
+ "lstrip": false,
217
+ "normalized": false,
218
+ "rstrip": false,
219
+ "single_word": false,
220
+ "special": true
221
+ },
222
+ "128135": {
223
+ "content": "<IMG_CONTEXT>",
224
+ "lstrip": false,
225
+ "normalized": false,
226
+ "rstrip": false,
227
+ "single_word": false,
228
+ "special": true
229
+ },
230
+ "128136": {
231
+ "content": "<quad>",
232
+ "lstrip": false,
233
+ "normalized": false,
234
+ "rstrip": false,
235
+ "single_word": false,
236
+ "special": true
237
+ },
238
+ "128137": {
239
+ "content": "</quad>",
240
+ "lstrip": false,
241
+ "normalized": false,
242
+ "rstrip": false,
243
+ "single_word": false,
244
+ "special": true
245
+ },
246
+ "128138": {
247
+ "content": "<ref>",
248
+ "lstrip": false,
249
+ "normalized": false,
250
+ "rstrip": false,
251
+ "single_word": false,
252
+ "special": true
253
+ },
254
+ "128139": {
255
+ "content": "</ref>",
256
+ "lstrip": false,
257
+ "normalized": false,
258
+ "rstrip": false,
259
+ "single_word": false,
260
+ "special": true
261
+ },
262
+ "128140": {
263
+ "content": "<box>",
264
+ "lstrip": false,
265
+ "normalized": false,
266
+ "rstrip": false,
267
+ "single_word": false,
268
+ "special": true
269
+ },
270
+ "128141": {
271
+ "content": "</box>",
272
+ "lstrip": false,
273
+ "normalized": false,
274
+ "rstrip": false,
275
+ "single_word": false,
276
+ "special": true
277
+ }
278
+ },
279
+ "additional_special_tokens": [
280
+ "<|im_start|>",
281
+ "<|im_end|>",
282
+ "<|action_start|>",
283
+ "<|action_end|>",
284
+ "<|interpreter|>",
285
+ "<|plugin|>",
286
+ "<restate>",
287
+ "</restate>",
288
+ "<planning>",
289
+ "</planning>",
290
+ "<recollect>",
291
+ "</recollect>",
292
+ "<execution>",
293
+ "</execution>",
294
+ "<review>",
295
+ "</review>",
296
+ "<summarize>",
297
+ "</summarize>",
298
+ "<retry>",
299
+ "</retry>",
300
+ "<conclude>",
301
+ "</conclude>",
302
+ "<img>",
303
+ "</img>",
304
+ "<IMG_CONTEXT>",
305
+ "<quad>",
306
+ "</quad>",
307
+ "<ref>",
308
+ "</ref>",
309
+ "<box>",
310
+ "</box>"
311
+ ],
312
+ "auto_map": {
313
+ "AutoTokenizer": [
314
+ "tokenization_internlm3.InternLM3Tokenizer",
315
+ null
316
+ ]
317
+ },
318
+ "bos_token": "<s>",
319
+ "chat_template": "{{ bos_token }}{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
320
+ "clean_up_tokenization_spaces": false,
321
+ "eos_token": "<|im_end|>",
322
+ "extra_special_tokens": {},
323
+ "model_max_length": 4096,
324
+ "pad_token": "</s>",
325
+ "sp_model_kwargs": {},
326
+ "spaces_between_special_tokens": false,
327
+ "tokenizer_class": "InternLM3Tokenizer",
328
+ "unk_token": "<unk>",
329
+ "use_default_system_prompt": false
330
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d34e0546c4f4010953c48a6661eed3c33fda840b71b8d16c01287f48ec032e6c
3
+ size 7224
training_log.txt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6501f8f940862dfa18d67a842fa26f663507a6dbb4e0bba6cee2e33bbfb8e123
3
+ size 20872712