ymccccc commited on
Commit
cbac23f
·
verified ·
1 Parent(s): e34b32c

Upload folder using huggingface_hub

Browse files
added_tokens.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "</answer>": 151652,
3
+ "</bbox>": 151648,
4
+ "</think>": 151650,
5
+ "<answer>": 151651,
6
+ "<bbox>": 151647,
7
+ "<image>": 151646,
8
+ "<think>": 151649,
9
+ "<|endoftext|>": 151643,
10
+ "<|im_end|>": 151645,
11
+ "<|im_start|>": 151644
12
+ }
config.json ADDED
@@ -0,0 +1,239 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "data/models/LLaVA-Video-7B-Qwen2",
3
+ "add_faster_video": false,
4
+ "add_time_instruction": false,
5
+ "architectures": [
6
+ "LlavaQwenForCausalLM"
7
+ ],
8
+ "attention_dropout": 0.0,
9
+ "bos_token_id": 151643,
10
+ "eos_token_id": 151645,
11
+ "faster_token_stride": 10,
12
+ "force_sample": true,
13
+ "ground_head_hidden_size": 512,
14
+ "ground_head_temperature": 0.07,
15
+ "ground_head_type": "autoreg",
16
+ "ground_loss_scale": 1,
17
+ "hidden_act": "silu",
18
+ "hidden_size": 3584,
19
+ "ignore_index": -100,
20
+ "image_aspect_ratio": "anyres_max_9",
21
+ "image_crop_resolution": null,
22
+ "image_grid_pinpoints": [
23
+ [
24
+ 384,
25
+ 384
26
+ ],
27
+ [
28
+ 384,
29
+ 768
30
+ ],
31
+ [
32
+ 384,
33
+ 1152
34
+ ],
35
+ [
36
+ 384,
37
+ 1536
38
+ ],
39
+ [
40
+ 384,
41
+ 1920
42
+ ],
43
+ [
44
+ 384,
45
+ 2304
46
+ ],
47
+ [
48
+ 768,
49
+ 384
50
+ ],
51
+ [
52
+ 768,
53
+ 768
54
+ ],
55
+ [
56
+ 768,
57
+ 1152
58
+ ],
59
+ [
60
+ 768,
61
+ 1536
62
+ ],
63
+ [
64
+ 768,
65
+ 1920
66
+ ],
67
+ [
68
+ 768,
69
+ 2304
70
+ ],
71
+ [
72
+ 1152,
73
+ 384
74
+ ],
75
+ [
76
+ 1152,
77
+ 768
78
+ ],
79
+ [
80
+ 1152,
81
+ 1152
82
+ ],
83
+ [
84
+ 1152,
85
+ 1536
86
+ ],
87
+ [
88
+ 1152,
89
+ 1920
90
+ ],
91
+ [
92
+ 1152,
93
+ 2304
94
+ ],
95
+ [
96
+ 1536,
97
+ 384
98
+ ],
99
+ [
100
+ 1536,
101
+ 768
102
+ ],
103
+ [
104
+ 1536,
105
+ 1152
106
+ ],
107
+ [
108
+ 1536,
109
+ 1536
110
+ ],
111
+ [
112
+ 1536,
113
+ 1920
114
+ ],
115
+ [
116
+ 1536,
117
+ 2304
118
+ ],
119
+ [
120
+ 1920,
121
+ 384
122
+ ],
123
+ [
124
+ 1920,
125
+ 768
126
+ ],
127
+ [
128
+ 1920,
129
+ 1152
130
+ ],
131
+ [
132
+ 1920,
133
+ 1536
134
+ ],
135
+ [
136
+ 1920,
137
+ 1920
138
+ ],
139
+ [
140
+ 1920,
141
+ 2304
142
+ ],
143
+ [
144
+ 2304,
145
+ 384
146
+ ],
147
+ [
148
+ 2304,
149
+ 768
150
+ ],
151
+ [
152
+ 2304,
153
+ 1152
154
+ ],
155
+ [
156
+ 2304,
157
+ 1536
158
+ ],
159
+ [
160
+ 2304,
161
+ 1920
162
+ ],
163
+ [
164
+ 2304,
165
+ 2304
166
+ ]
167
+ ],
168
+ "image_split_resolution": null,
169
+ "image_token_index": 151646,
170
+ "initializer_range": 0.02,
171
+ "intermediate_size": 18944,
172
+ "max_position_embeddings": 32768,
173
+ "max_window_layers": 28,
174
+ "max_xyz_range": [
175
+ 15,
176
+ 15,
177
+ 5
178
+ ],
179
+ "min_xyz_range": [
180
+ -15,
181
+ -15,
182
+ -5
183
+ ],
184
+ "mm_hidden_size": 1152,
185
+ "mm_newline_position": "grid",
186
+ "mm_patch_merge_type": "spatial_unpad",
187
+ "mm_projector_lr": null,
188
+ "mm_projector_type": "mlp2x_gelu",
189
+ "mm_resampler_type": null,
190
+ "mm_spatial_pool_mode": "bilinear",
191
+ "mm_spatial_pool_stride": 2,
192
+ "mm_tunable_parts": "mm_mlp_adapter,spatial_encoder,mm_language_model",
193
+ "mm_use_im_patch_token": false,
194
+ "mm_use_im_start_end": false,
195
+ "mm_vision_select_feature": "patch",
196
+ "mm_vision_select_layer": -2,
197
+ "mm_vision_tower": "google/siglip-so400m-patch14-384",
198
+ "mm_vision_tower_lr": 2e-06,
199
+ "model_type": "llava",
200
+ "num_attention_heads": 28,
201
+ "num_hidden_layers": 28,
202
+ "num_key_value_heads": 4,
203
+ "object_feature_type": "patch14-pe",
204
+ "perception_hidden_size": 1024,
205
+ "pos_skipping_range": 4096,
206
+ "projector_hidden_act": "gelu",
207
+ "rms_norm_eps": 1e-06,
208
+ "rope_scaling": null,
209
+ "rope_theta": 1000000.0,
210
+ "sliding_window": 131072,
211
+ "text_config": {
212
+ "model_type": "llama"
213
+ },
214
+ "tie_word_embeddings": false,
215
+ "tokenizer_model_max_length": 32768,
216
+ "tokenizer_padding_side": "right",
217
+ "torch_dtype": "bfloat16",
218
+ "transformers_version": "4.40.0.dev0",
219
+ "use_cache": true,
220
+ "use_mm_proj": true,
221
+ "use_pos_skipping": false,
222
+ "use_sliding_window": false,
223
+ "vision_config": {
224
+ "hidden_size": 1024,
225
+ "image_size": 336,
226
+ "intermediate_size": 4096,
227
+ "model_type": "clip_vision_model",
228
+ "num_attention_heads": 16,
229
+ "num_hidden_layers": 24,
230
+ "patch_size": 14,
231
+ "projection_dim": 768,
232
+ "vocab_size": 32000
233
+ },
234
+ "vision_feature_layer": -2,
235
+ "vision_feature_select_strategy": "default",
236
+ "vision_tower_pretrained": null,
237
+ "voxel_size": 0.1,
238
+ "world_position_embedding_type": "sonata_sample_64"
239
+ }
generation_config.json ADDED
@@ -0,0 +1,14 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 151643,
3
+ "do_sample": true,
4
+ "eos_token_id": [
5
+ 151645,
6
+ 151643
7
+ ],
8
+ "pad_token_id": 151643,
9
+ "repetition_penalty": 1.05,
10
+ "temperature": 0.7,
11
+ "top_k": 20,
12
+ "top_p": 0.8,
13
+ "transformers_version": "4.40.0.dev0"
14
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model-00001-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d0b7bf3960702a7a527f6be64dbcd1d00e1f89d9e0faa40d5230975ff7b655f9
3
+ size 4874721984
model-00002-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:648f02650038fae6b94e3a0a218d1cf154347348d9ed6c2f0fe437a789173ab8
3
+ size 4932751008
model-00003-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a2981bb28a7c6a4e0479b4743d097b633e6bc8452c04ba48fbca125bb3c8eddd
3
+ size 4994571904
model-00004-of-00004.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb2bc2251f1a27907a27df7e0d1637c0d30fac48300c457a62ac921160d757c8
3
+ size 1533022066
model.safetensors.index.json ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>"
5
+ ],
6
+ "eos_token": {
7
+ "content": "<|im_end|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "pad_token": {
14
+ "content": "<|endoftext|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ }
20
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,101 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ },
28
+ "151646": {
29
+ "content": "<image>",
30
+ "lstrip": false,
31
+ "normalized": false,
32
+ "rstrip": false,
33
+ "single_word": false,
34
+ "special": true
35
+ },
36
+ "151647": {
37
+ "content": "<bbox>",
38
+ "lstrip": false,
39
+ "normalized": true,
40
+ "rstrip": false,
41
+ "single_word": false,
42
+ "special": false
43
+ },
44
+ "151648": {
45
+ "content": "</bbox>",
46
+ "lstrip": false,
47
+ "normalized": true,
48
+ "rstrip": false,
49
+ "single_word": false,
50
+ "special": false
51
+ },
52
+ "151649": {
53
+ "content": "<think>",
54
+ "lstrip": false,
55
+ "normalized": true,
56
+ "rstrip": false,
57
+ "single_word": false,
58
+ "special": false
59
+ },
60
+ "151650": {
61
+ "content": "</think>",
62
+ "lstrip": false,
63
+ "normalized": true,
64
+ "rstrip": false,
65
+ "single_word": false,
66
+ "special": false
67
+ },
68
+ "151651": {
69
+ "content": "<answer>",
70
+ "lstrip": false,
71
+ "normalized": true,
72
+ "rstrip": false,
73
+ "single_word": false,
74
+ "special": false
75
+ },
76
+ "151652": {
77
+ "content": "</answer>",
78
+ "lstrip": false,
79
+ "normalized": true,
80
+ "rstrip": false,
81
+ "single_word": false,
82
+ "special": false
83
+ }
84
+ },
85
+ "additional_special_tokens": [
86
+ "<|im_start|>",
87
+ "<|im_end|>"
88
+ ],
89
+ "bos_token": null,
90
+ "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
91
+ "clean_up_tokenization_spaces": false,
92
+ "eos_token": "<|im_end|>",
93
+ "errors": "replace",
94
+ "model_max_length": 32768,
95
+ "pad_token": "<|endoftext|>",
96
+ "padding_side": "right",
97
+ "processor_class": "LlavaProcessor",
98
+ "split_special_tokens": false,
99
+ "tokenizer_class": "Qwen2Tokenizer",
100
+ "unk_token": null
101
+ }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a76c791c7d9005a8d864fe0ff5465c0b1b0e3b80b263e03d3d3ac5cc8873e8de
3
+ size 7608
vocab.json ADDED
The diff for this file is too large to render. See raw diff