megadarkparticle/testmodel1

1{2  "_commit_hash": null,3  "_name_or_path": "/mnt3/Documents/InternVL/internvl_chat/work_dirs/drivelm/InternVL2-1B/drivelm_gpu6_True_12_1_1/checkpoint-3937",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  "llm_config": {16    "_name_or_path": "Qwen/Qwen2-0.5B-Instruct",17    "add_cross_attention": false,18    "architectures": [19      "Qwen2ForCausalLM"20    ],21    "attention_dropout": 0.0,22    "bad_words_ids": null,23    "begin_suppress_tokens": null,24    "bos_token_id": 151643,25    "chunk_size_feed_forward": 0,26    "cross_attention_hidden_size": null,27    "decoder_start_token_id": null,28    "diversity_penalty": 0.0,29    "do_sample": false,30    "early_stopping": false,31    "encoder_no_repeat_ngram_size": 0,32    "eos_token_id": 151645,33    "exponential_decay_length_penalty": null,34    "finetuning_task": null,35    "forced_bos_token_id": null,36    "forced_eos_token_id": null,37    "hidden_act": "silu",38    "hidden_size": 896,39    "id2label": {40      "0": "LABEL_0",41      "1": "LABEL_1"42    },43    "initializer_range": 0.02,44    "intermediate_size": 4864,45    "is_decoder": false,46    "is_encoder_decoder": false,47    "label2id": {48      "LABEL_0": 0,49      "LABEL_1": 150    },51    "length_penalty": 1.0,52    "max_length": 20,53    "max_position_embeddings": 32768,54    "max_window_layers": 24,55    "min_length": 0,56    "model_type": "qwen2",57    "no_repeat_ngram_size": 0,58    "num_attention_heads": 14,59    "num_beam_groups": 1,60    "num_beams": 1,61    "num_hidden_layers": 24,62    "num_key_value_heads": 2,63    "num_return_sequences": 1,64    "output_attentions": false,65    "output_hidden_states": false,66    "output_scores": false,67    "pad_token_id": null,68    "prefix": null,69    "problem_type": null,70    "pruned_heads": {},71    "remove_invalid_values": false,72    "repetition_penalty": 1.0,73    "return_dict": true,74    "return_dict_in_generate": false,75    "rms_norm_eps": 1e-06,76    "rope_theta": 1000000.0,77    "sep_token_id": null,78    "sliding_window": null,79    "suppress_tokens": null,80    "task_specific_params": null,81    "temperature": 1.0,82    "tf_legacy_loss": false,83    "tie_encoder_decoder": false,84    "tie_word_embeddings": true,85    "tokenizer_class": null,86    "top_k": 50,87    "top_p": 1.0,88    "torch_dtype": "bfloat16",89    "torchscript": false,90    "transformers_version": "4.37.2",91    "typical_p": 1.0,92    "use_bfloat16": true,93    "use_cache": false,94    "use_sliding_window": false,95    "vocab_size": 15165596  },97  "max_dynamic_patch": 12,98  "min_dynamic_patch": 1,99  "model_type": "internvl_chat",100  "pad2square": false,101  "ps_version": "v2",102  "select_layer": -1,103  "template": "Hermes-2",104  "torch_dtype": "bfloat16",105  "transformers_version": null,106  "use_backbone_lora": 64,107  "use_llm_lora": 64,108  "use_thumbnail": true,109  "vision_config": {110    "_name_or_path": "",111    "add_cross_attention": false,112    "architectures": [113      "InternVisionModel"114    ],115    "attention_dropout": 0.0,116    "bad_words_ids": null,117    "begin_suppress_tokens": null,118    "bos_token_id": null,119    "chunk_size_feed_forward": 0,120    "cross_attention_hidden_size": null,121    "decoder_start_token_id": null,122    "diversity_penalty": 0.0,123    "do_sample": false,124    "drop_path_rate": 0.1,125    "dropout": 0.0,126    "early_stopping": false,127    "encoder_no_repeat_ngram_size": 0,128    "eos_token_id": null,129    "exponential_decay_length_penalty": null,130    "finetuning_task": null,131    "forced_bos_token_id": null,132    "forced_eos_token_id": null,133    "hidden_act": "gelu",134    "hidden_size": 1024,135    "id2label": {136      "0": "LABEL_0",137      "1": "LABEL_1"138    },139    "image_size": 448,140    "initializer_factor": 1.0,141    "initializer_range": 0.02,142    "intermediate_size": 4096,143    "is_decoder": false,144    "is_encoder_decoder": false,145    "label2id": {146      "LABEL_0": 0,147      "LABEL_1": 1148    },149    "layer_norm_eps": 1e-06,150    "length_penalty": 1.0,151    "max_length": 20,152    "min_length": 0,153    "model_type": "intern_vit_6b",154    "no_repeat_ngram_size": 0,155    "norm_type": "layer_norm",156    "num_attention_heads": 16,157    "num_beam_groups": 1,158    "num_beams": 1,159    "num_channels": 3,160    "num_hidden_layers": 24,161    "num_return_sequences": 1,162    "output_attentions": false,163    "output_hidden_states": false,164    "output_scores": false,165    "pad_token_id": null,166    "patch_size": 14,167    "prefix": null,168    "problem_type": null,169    "pruned_heads": {},170    "qk_normalization": false,171    "qkv_bias": true,172    "remove_invalid_values": false,173    "repetition_penalty": 1.0,174    "return_dict": true,175    "return_dict_in_generate": false,176    "sep_token_id": null,177    "suppress_tokens": null,178    "task_specific_params": null,179    "temperature": 1.0,180    "tf_legacy_loss": false,181    "tie_encoder_decoder": false,182    "tie_word_embeddings": true,183    "tokenizer_class": null,184    "top_k": 50,185    "top_p": 1.0,186    "torch_dtype": "bfloat16",187    "torchscript": false,188    "transformers_version": "4.37.2",189    "typical_p": 1.0,190    "use_bfloat16": true,191    "use_flash_attn": true192  }193}194