microsoft/kosmos-2.5-chat

1{2  "architectures": [3    "Kosmos2_5ForConditionalGeneration"4  ],5  "latent_query_num": 2048,6  "model_type": "kosmos-2.5",7  "text_config": {8    "_name_or_path": "",9    "activation_dropout": 0.0,10    "activation_function": "gelu",11    "add_cross_attention": false,12    "architectures": null,13    "attention_dropout": 0.0,14    "attention_heads": 16,15    "bad_words_ids": null,16    "begin_suppress_tokens": null,17    "bos_token_id": 0,18    "chunk_size_feed_forward": 0,19    "cross_attention_hidden_size": null,20    "decoder_start_token_id": null,21    "diversity_penalty": 0.0,22    "do_sample": false,23    "dropout": 0.1,24    "early_stopping": false,25    "embed_dim": 1536,26    "encoder_no_repeat_ngram_size": 0,27    "eos_token_id": 2,28    "exponential_decay_length_penalty": null,29    "ffn_dim": 6144,30    "finetuning_task": null,31    "forced_bos_token_id": null,32    "forced_eos_token_id": null,33    "id2label": {34      "0": "LABEL_0",35      "1": "LABEL_1"36    },37    "init_std": 0.02,38    "is_decoder": false,39    "is_encoder_decoder": false,40    "label2id": {41      "LABEL_0": 0,42      "LABEL_1": 143    },44    "layer_norm_eps": 1e-05,45    "layerdrop": 0.0,46    "length_penalty": 1.0,47    "layers": 24,48    "max_length": 20,49    "max_position_embeddings": 4096,50    "min_length": 0,51    "model_type": "kosmos_2_5_text_model",52    "no_repeat_ngram_size": 3,53    "num_beam_groups": 1,54    "num_beams": 1,55    "num_return_sequences": 1,56    "output_attentions": false,57    "output_hidden_states": false,58    "output_scores": false,59    "pad_token_id": 1,60    "prefix": null,61    "problem_type": null,62    "pruned_heads": {},63    "remove_invalid_values": false,64    "repetition_penalty": 1.0,65    "return_dict": true,66    "return_dict_in_generate": false,67    "scale_embedding": true,68    "sep_token_id": null,69    "suppress_tokens": null,70    "task_specific_params": null,71    "temperature": 1.0,72    "tf_legacy_loss": false,73    "tie_encoder_decoder": false,74    "tie_word_embeddings": true,75    "tokenizer_class": null,76    "top_k": 50,77    "top_p": 1.0,78    "torch_dtype": null,79    "torchscript": false,80    "typical_p": 1.0,81    "use_bfloat16": false,82    "use_cache": true,83    "vocab_size": 10848184  },85  "torch_dtype": "float32",86  "transformers_version": "4.43.3",87  "vision_config": {88    "_name_or_path": "",89    "add_cross_attention": false,90    "architectures": null,91    "attention_dropout": 0.0,92    "bad_words_ids": null,93    "begin_suppress_tokens": null,94    "bos_token_id": null,95    "chunk_size_feed_forward": 0,96    "cross_attention_hidden_size": null,97    "intermediate_size": 3968,98    "head_dim": 64,99    "decoder_start_token_id": null,100    "dense_act_fn": "gelu_new",101    "diversity_penalty": 0.0,102    "do_sample": false,103    "dropout_rate": 0.0,104    "early_stopping": false,105    "encoder_no_repeat_ngram_size": 0,106    "eos_token_id": null,107    "exponential_decay_length_penalty": null,108    "finetuning_task": null,109    "forced_bos_token_id": null,110    "forced_eos_token_id": null,111    "hidden_size": 1536,112    "id2label": {113      "0": "LABEL_0",114      "1": "LABEL_1"115    },116    "initializer_factor": 1.0,117    "initializer_range": 1e-10,118    "is_decoder": false,119    "is_encoder_decoder": false,120    "label2id": {121      "LABEL_0": 0,122      "LABEL_1": 1123    },124    "layer_norm_eps": 1e-06,125    "length_penalty": 1.0,126    "max_length": 4096,127    "min_length": 0,128    "model_type": "kosmos_2_5_vision_model",129    "no_repeat_ngram_size": 0,130    "num_attention_heads": 24,131    "num_beam_groups": 1,132    "num_beams": 1,133    "num_hidden_layers": 18,134    "num_return_sequences": 1,135    "output_attentions": false,136    "output_hidden_states": false,137    "output_scores": false,138    "pad_token_id": null,139    "patch_embed_hidden_size": 768,140    "prefix": null,141    "problem_type": null,142    "pruned_heads": {},143    "remove_invalid_values": false,144    "repetition_penalty": 1.0,145    "return_dict": true,146    "return_dict_in_generate": false,147    "sep_token_id": null,148    "max_num_patches": 4096,149    "suppress_tokens": null,150    "task_specific_params": null,151    "temperature": 1.0,152    "tf_legacy_loss": false,153    "tie_encoder_decoder": false,154    "tie_word_embeddings": true,155    "tokenizer_class": null,156    "top_k": 50,157    "top_p": 1.0,158    "torch_dtype": null,159    "torchscript": false,160    "typical_p": 1.0,161    "use_bfloat16": false162  }163}164