OpenGVLab/InternVL3-1B
84122k
1{2 "_commit_hash": null,3 "_name_or_path": "/mnt/petrelfs/wangweiyun/workspace_wwy/open_source/InternVL/internvl_chat/work_dirs/internvl_chat_v3_0/InternVL3_0-1B-MPO-try0-2",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": 896,16 "image_fold": null,17 "llm_config": {18 "_attn_implementation_autoset": true,19 "_name_or_path": "./pretrained/Qwen2.5-32B-Instruct",20 "add_cross_attention": false,21 "architectures": [22 "Qwen2ForCausalLM"23 ],24 "attention_dropout": 0.0,25 "bad_words_ids": null,26 "begin_suppress_tokens": null,27 "bos_token_id": 151643,28 "chunk_size_feed_forward": 0,29 "cross_attention_hidden_size": null,30 "decoder_start_token_id": null,31 "diversity_penalty": 0.0,32 "do_sample": false,33 "early_stopping": false,34 "encoder_no_repeat_ngram_size": 0,35 "eos_token_id": 151643,36 "exponential_decay_length_penalty": null,37 "finetuning_task": null,38 "forced_bos_token_id": null,39 "forced_eos_token_id": null,40 "hidden_act": "silu",41 "hidden_size": 896,42 "id2label": {43 "0": "LABEL_0",44 "1": "LABEL_1"45 },46 "initializer_range": 0.02,47 "intermediate_size": 4864,48 "is_decoder": false,49 "is_encoder_decoder": false,50 "label2id": {51 "LABEL_0": 0,52 "LABEL_1": 153 },54 "length_penalty": 1.0,55 "max_length": 20,56 "max_position_embeddings": 32768,57 "max_window_layers": 70,58 "min_length": 0,59 "model_type": "qwen2",60 "moe_config": null,61 "no_repeat_ngram_size": 0,62 "num_attention_heads": 14,63 "num_beam_groups": 1,64 "num_beams": 1,65 "num_hidden_layers": 24,66 "num_key_value_heads": 2,67 "num_return_sequences": 1,68 "output_attentions": false,69 "output_hidden_states": false,70 "output_scores": false,71 "pad_token_id": null,72 "prefix": null,73 "problem_type": null,74 "pruned_heads": {},75 "remove_invalid_values": false,76 "repetition_penalty": 1.0,77 "return_dict": true,78 "return_dict_in_generate": false,79 "rms_norm_eps": 1e-06,80 "rope_scaling": {81 "factor": 2.0,82 "rope_type": "dynamic",83 "type": "dynamic"84 },85 "rope_theta": 1000000.0,86 "sep_token_id": null,87 "sliding_window": null,88 "suppress_tokens": null,89 "task_specific_params": null,90 "temperature": 1.0,91 "tf_legacy_loss": false,92 "tie_encoder_decoder": false,93 "tie_word_embeddings": false,94 "tokenizer_class": null,95 "top_k": 50,96 "top_p": 1.0,97 "torch_dtype": "bfloat16",98 "torchscript": false,99 "transformers_version": "4.48.3",100 "typical_p": 1.0,101 "use_bfloat16": true,102 "use_cache": false,103 "use_sliding_window": false,104 "vocab_size": 151674105 },106 "max_dynamic_patch": 12,107 "min_dynamic_patch": 1,108 "model_type": "internvl_chat",109 "pad2square": false,110 "ps_version": "v2",111 "select_layer": -1,112 "system_message": null,113 "template": "internvl2_5",114 "tie_word_embeddings": false,115 "torch_dtype": "bfloat16",116 "transformers_version": null,117 "use_backbone_lora": 0,118 "use_llm_lora": 0,119 "use_thumbnail": true,120 "vision_config": {121 "_attn_implementation_autoset": true,122 "_name_or_path": "OpenGVLab/InternViT-6B-448px-V1-5",123 "add_cross_attention": false,124 "architectures": [125 "InternVisionModel"126 ],127 "attention_dropout": 0.0,128 "auto_map": {129 "AutoConfig": "configuration_intern_vit.InternVisionConfig",130 "AutoModel": "modeling_intern_vit.InternVisionModel"131 },132 "bad_words_ids": null,133 "begin_suppress_tokens": null,134 "bos_token_id": null,135 "capacity_factor": 1.2,136 "chunk_size_feed_forward": 0,137 "cross_attention_hidden_size": null,138 "decoder_start_token_id": null,139 "diversity_penalty": 0.0,140 "do_sample": false,141 "drop_path_rate": 0.1,142 "dropout": 0.0,143 "early_stopping": false,144 "encoder_no_repeat_ngram_size": 0,145 "eos_token_id": null,146 "eval_capacity_factor": 1.4,147 "exponential_decay_length_penalty": null,148 "finetuning_task": null,149 "forced_bos_token_id": null,150 "forced_eos_token_id": null,151 "hidden_act": "gelu",152 "hidden_size": 1024,153 "id2label": {154 "0": "LABEL_0",155 "1": "LABEL_1"156 },157 "image_size": 448,158 "initializer_factor": 0.1,159 "initializer_range": 1e-10,160 "intermediate_size": 4096,161 "is_decoder": false,162 "is_encoder_decoder": false,163 "label2id": {164 "LABEL_0": 0,165 "LABEL_1": 1166 },167 "laux_allreduce": "all_nodes",168 "layer_norm_eps": 1e-06,169 "length_penalty": 1.0,170 "max_length": 20,171 "min_length": 0,172 "model_type": "intern_vit_6b",173 "moe_coeff_ratio": 0.5,174 "moe_intermediate_size": 768,175 "moe_output_scale": 4.0,176 "no_repeat_ngram_size": 0,177 "noisy_gate_policy": "RSample_before",178 "norm_type": "layer_norm",179 "num_attention_heads": 16,180 "num_beam_groups": 1,181 "num_beams": 1,182 "num_channels": 3,183 "num_experts": 8,184 "num_hidden_layers": 24,185 "num_return_sequences": 1,186 "num_routed_experts": 4,187 "num_shared_experts": 4,188 "output_attentions": false,189 "output_hidden_states": false,190 "output_scores": false,191 "pad_token_id": null,192 "patch_size": 14,193 "prefix": null,194 "problem_type": null,195 "pruned_heads": {},196 "qk_normalization": false,197 "qkv_bias": true,198 "remove_invalid_values": false,199 "repetition_penalty": 1.0,200 "return_dict": true,201 "return_dict_in_generate": false,202 "sep_token_id": null,203 "shared_expert_intermediate_size": 3072,204 "suppress_tokens": null,205 "task_specific_params": null,206 "temperature": 1.0,207 "tf_legacy_loss": false,208 "tie_encoder_decoder": false,209 "tie_word_embeddings": true,210 "tokenizer_class": null,211 "top_k": 50,212 "top_p": 1.0,213 "torch_dtype": "bfloat16",214 "torchscript": false,215 "transformers_version": "4.48.3",216 "typical_p": 1.0,217 "use_bfloat16": true,218 "use_flash_attn": true,219 "use_moe": false,220 "use_residual": true,221 "use_rts": false,222 "use_weighted_residual": false223 }224}225