Hugging Face's logo

InternRobotics
/
RoboInter-VLM

Robotics
Transformers
Safetensors
qwen2_5_vl
image-text-to-text
vision-language-action-model
vision-language-model
text-generation-inference
Model card Files
xet
Community
JeasLee commited on
Commit
1a8ec31
·
verified ·
1 Parent(s): 9e2e717

Delete RoboInterVLM_llava_one_vision_7B/config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. RoboInterVLM_llava_one_vision_7B/config.json +0 -204
RoboInterVLM_llava_one_vision_7B/config.json DELETED
@@ -1,204 +0,0 @@
1
- {
2
- "_name_or_path": "/mnt/petrelfs/share/wangziqin/robotdata/playground/Pretrained_models/llava-onevision-qwen2-7b-ov",
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": false,
13
- "hidden_act": "silu",
14
- "hidden_size": 3584,
15
- "image_aspect_ratio": "anyres_max_9",
16
- "image_crop_resolution": null,
17
- "image_grid_pinpoints": [
18
- [
19
- 384,
20
- 384
21
- ],
22
- [
23
- 384,
24
- 768
25
- ],
26
- [
27
- 384,
28
- 1152
29
- ],
30
- [
31
- 384,
32
- 1536
33
- ],
34
- [
35
- 384,
36
- 1920
37
- ],
38
- [
39
- 384,
40
- 2304
41
- ],
42
- [
43
- 768,
44
- 384
45
- ],
46
- [
47
- 768,
48
- 768
49
- ],
50
- [
51
- 768,
52
- 1152
53
- ],
54
- [
55
- 768,
56
- 1536
57
- ],
58
- [
59
- 768,
60
- 1920
61
- ],
62
- [
63
- 768,
64
- 2304
65
- ],
66
- [
67
- 1152,
68
- 384
69
- ],
70
- [
71
- 1152,
72
- 768
73
- ],
74
- [
75
- 1152,
76
- 1152
77
- ],
78
- [
79
- 1152,
80
- 1536
81
- ],
82
- [
83
- 1152,
84
- 1920
85
- ],
86
- [
87
- 1152,
88
- 2304
89
- ],
90
- [
91
- 1536,
92
- 384
93
- ],
94
- [
95
- 1536,
96
- 768
97
- ],
98
- [
99
- 1536,
100
- 1152
101
- ],
102
- [
103
- 1536,
104
- 1536
105
- ],
106
- [
107
- 1536,
108
- 1920
109
- ],
110
- [
111
- 1536,
112
- 2304
113
- ],
114
- [
115
- 1920,
116
- 384
117
- ],
118
- [
119
- 1920,
120
- 768
121
- ],
122
- [
123
- 1920,
124
- 1152
125
- ],
126
- [
127
- 1920,
128
- 1536
129
- ],
130
- [
131
- 1920,
132
- 1920
133
- ],
134
- [
135
- 1920,
136
- 2304
137
- ],
138
- [
139
- 2304,
140
- 384
141
- ],
142
- [
143
- 2304,
144
- 768
145
- ],
146
- [
147
- 2304,
148
- 1152
149
- ],
150
- [
151
- 2304,
152
- 1536
153
- ],
154
- [
155
- 2304,
156
- 1920
157
- ],
158
- [
159
- 2304,
160
- 2304
161
- ]
162
- ],
163
- "image_split_resolution": null,
164
- "image_token_index": 151646,
165
- "initializer_range": 0.02,
166
- "intermediate_size": 18944,
167
- "max_position_embeddings": 32768,
168
- "max_window_layers": 28,
169
- "mm_hidden_size": 1152,
170
- "mm_newline_position": "grid",
171
- "mm_patch_merge_type": "spatial_unpad",
172
- "mm_projector_lr": null,
173
- "mm_projector_type": "mlp2x_gelu",
174
- "mm_resampler_type": null,
175
- "mm_spatial_pool_mode": "bilinear",
176
- "mm_spatial_pool_stride": null,
177
- "mm_tunable_parts": "mm_vision_tower,mm_mlp_adapter,mm_language_model",
178
- "mm_use_im_patch_token": false,
179
- "mm_use_im_start_end": false,
180
- "mm_vision_select_feature": "patch",
181
- "mm_vision_select_layer": -2,
182
- "mm_vision_tower": "/mnt/lustre/share/wangziqin/robotdata/playground/Pretrained_models/siglip-so400m-patch14-384/",
183
- "mm_vision_tower_lr": 4e-06,
184
- "model_type": "llava_qwen",
185
- "num_attention_heads": 28,
186
- "num_hidden_layers": 28,
187
- "num_key_value_heads": 4,
188
- "pos_skipping_range": 4096,
189
- "rms_norm_eps": 1e-06,
190
- "rope_scaling": null,
191
- "rope_theta": 1000000.0,
192
- "sliding_window": 131072,
193
- "tie_word_embeddings": false,
194
- "tokenizer_model_max_length": 32768,
195
- "tokenizer_padding_side": "right",
196
- "torch_dtype": "bfloat16",
197
- "transformers_version": "4.40.0.dev0",
198
- "use_cache": true,
199
- "use_mm_proj": true,
200
- "use_pos_skipping": false,
201
- "use_sliding_window": false,
202
- "vision_tower_pretrained": null,
203
- "vocab_size": 152064
204
- }