gtang666 commited on
Commit
820ee3a
·
verified ·
1 Parent(s): bd84cd9

Upload InternVL/config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. InternVL/config.json +144 -0
InternVL/config.json ADDED
@@ -0,0 +1,144 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_commit_hash": null,
3
+ "architectures": [
4
+ "InternVLChatModel"
5
+ ],
6
+ "auto_map": {
7
+ "AutoConfig": "configuration_internvl_chat.InternVLChatConfig",
8
+ "AutoModel": "modeling_internvl_chat.InternVLChatModel",
9
+ "AutoModelForCausalLM": "modeling_internvl_chat.InternVLChatModel"
10
+ },
11
+ "downsample_ratio": 0.5,
12
+ "dynamic_image_size": true,
13
+ "force_image_size": 448,
14
+ "llm_config": {
15
+ "_name_or_path": "internlm/internlm2_5-7b-chat",
16
+ "add_cross_attention": false,
17
+ "architectures": [
18
+ "InternLM2ForCausalLM"
19
+ ],
20
+ "attn_implementation": "flash_attention_2",
21
+ "auto_map": {
22
+ "AutoConfig": "configuration_internlm2.InternLM2Config",
23
+ "AutoModel": "modeling_internlm2.InternLM2ForCausalLM",
24
+ "AutoModelForCausalLM": "modeling_internlm2.InternLM2ForCausalLM"
25
+ },
26
+ "bad_words_ids": null,
27
+ "begin_suppress_tokens": null,
28
+ "bias": false,
29
+ "bos_token_id": 1,
30
+ "chunk_size_feed_forward": 0,
31
+ "cross_attention_hidden_size": null,
32
+ "decoder_start_token_id": null,
33
+ "diversity_penalty": 0.0,
34
+ "do_sample": false,
35
+ "early_stopping": false,
36
+ "encoder_no_repeat_ngram_size": 0,
37
+ "eos_token_id": 2,
38
+ "exponential_decay_length_penalty": null,
39
+ "finetuning_task": null,
40
+ "forced_bos_token_id": null,
41
+ "forced_eos_token_id": null,
42
+ "hidden_act": "silu",
43
+ "hidden_size": 4096,
44
+ "id2label": {
45
+ "0": "LABEL_0",
46
+ "1": "LABEL_1"
47
+ },
48
+ "initializer_range": 0.02,
49
+ "intermediate_size": 14336,
50
+ "is_decoder": false,
51
+ "is_encoder_decoder": false,
52
+ "label2id": {
53
+ "LABEL_0": 0,
54
+ "LABEL_1": 1
55
+ },
56
+ "length_penalty": 1.0,
57
+ "max_length": 20,
58
+ "max_position_embeddings": 32768,
59
+ "min_length": 0,
60
+ "model_type": "internlm2",
61
+ "no_repeat_ngram_size": 0,
62
+ "num_attention_heads": 32,
63
+ "num_beam_groups": 1,
64
+ "num_beams": 1,
65
+ "num_hidden_layers": 32,
66
+ "num_key_value_heads": 8,
67
+ "num_return_sequences": 1,
68
+ "output_attentions": false,
69
+ "output_hidden_states": false,
70
+ "output_scores": false,
71
+ "pad_token_id": 2,
72
+ "prefix": null,
73
+ "pretraining_tp": 1,
74
+ "problem_type": null,
75
+ "pruned_heads": {},
76
+ "remove_invalid_values": false,
77
+ "repetition_penalty": 1.0,
78
+ "return_dict": true,
79
+ "return_dict_in_generate": false,
80
+ "rms_norm_eps": 1e-05,
81
+ "rope_scaling": {
82
+ "factor": 2.0,
83
+ "type": "dynamic"
84
+ },
85
+ "rope_theta": 1000000,
86
+ "sep_token_id": null,
87
+ "suppress_tokens": null,
88
+ "task_specific_params": null,
89
+ "temperature": 1.0,
90
+ "tf_legacy_loss": false,
91
+ "tie_encoder_decoder": false,
92
+ "tie_word_embeddings": false,
93
+ "tokenizer_class": null,
94
+ "top_k": 50,
95
+ "top_p": 1.0,
96
+ "torch_dtype": "bfloat16",
97
+ "torchscript": false,
98
+ "transformers_version": "4.37.2",
99
+ "typical_p": 1.0,
100
+ "use_bfloat16": true,
101
+ "use_cache": true,
102
+ "vocab_size": 92553
103
+ },
104
+ "max_dynamic_patch": 12,
105
+ "min_dynamic_patch": 1,
106
+ "model_type": "internvl_chat",
107
+ "ps_version": "v2",
108
+ "select_layer": -1,
109
+ "template": "internlm2-chat",
110
+ "torch_dtype": "bfloat16",
111
+ "use_backbone_lora": 0,
112
+ "use_llm_lora": 0,
113
+ "use_thumbnail": true,
114
+ "vision_config": {
115
+ "architectures": [
116
+ "InternVisionModel"
117
+ ],
118
+ "attention_dropout": 0.0,
119
+ "drop_path_rate": 0.0,
120
+ "dropout": 0.0,
121
+ "hidden_act": "gelu",
122
+ "hidden_size": 1024,
123
+ "image_size": 448,
124
+ "initializer_factor": 1.0,
125
+ "initializer_range": 0.02,
126
+ "intermediate_size": 4096,
127
+ "layer_norm_eps": 1e-06,
128
+ "model_type": "intern_vit_6b",
129
+ "norm_type": "layer_norm",
130
+ "num_attention_heads": 16,
131
+ "num_channels": 3,
132
+ "num_hidden_layers": 24,
133
+ "output_attentions": false,
134
+ "output_hidden_states": false,
135
+ "patch_size": 14,
136
+ "qk_normalization": false,
137
+ "qkv_bias": true,
138
+ "return_dict": true,
139
+ "torch_dtype": "bfloat16",
140
+ "transformers_version": "4.37.2",
141
+ "use_bfloat16": true,
142
+ "use_flash_attn": true
143
+ }
144
+ }