LeroyDyer/SpydazWeb_VisonEncoderDecoder_Project

1{2  "_name_or_path": "LeroyDyer/Mixtral_AI_Cyber_Q_Vision",3  "architectures": [4    "MistralForCausalLM"5  ],6  "attention_dropout": 0.0,7  "bos_token_id": 1,8  "eos_token_id": 2,9  "hidden_act": "silu",10  "hidden_size": 4096,11  "initializer_range": 0.02,12  "intermediate_size": 14336,13  "max_position_embeddings": 32768,14  "model_type": "mistral",15  "num_attention_heads": 32,16  "num_hidden_layers": 32,17  "num_key_value_heads": 8,18  "pad_token_id": 2,19  "rms_norm_eps": 1e-05,20  "rope_theta": 10000.0,21  "sliding_window": 4096,22  "tie_word_embeddings": false,23  "torch_dtype": "float16",24  "transformers_version": "4.38.2",25  "unsloth_version": "2024.3",26  "use_cache": true,27  "vocab_size": 32004,28  "decoder": {29    "_name_or_path": "LeroyDyer/Mixtral_AI_Cyber_Q_Vision",30    "add_cross_attention": true,31    "architectures": [32      "MistralForCausalLM"33    ],34    "attention_dropout": 0.0,35    "bad_words_ids": null,36    "begin_suppress_tokens": null,37    "bos_token_id": 1,38    "chunk_size_feed_forward": 0,39    "cross_attention_hidden_size": null,40    "decoder_start_token_id": null,41    "diversity_penalty": 0.0,42    "do_sample": false,43    "early_stopping": false,44    "encoder_no_repeat_ngram_size": 0,45    "eos_token_id": 2,46    "exponential_decay_length_penalty": null,47    "finetuning_task": null,48    "forced_bos_token_id": null,49    "forced_eos_token_id": null,50    "hidden_act": "silu",51    "hidden_size": 1024,52    "id2label": {53      "0": "LABEL_0",54      "1": "LABEL_1"55    },56    "initializer_range": 0.02,57    "intermediate_size": 4096,58    "is_decoder": true,59    "is_encoder_decoder": false,60    "label2id": {61      "LABEL_0": 0,62      "LABEL_1": 163    },64    "length_penalty": 1.0,65    "max_length": 20,66    "max_position_embeddings": 32768,67    "min_length": 0,68    "model_type": "mistral",69    "no_repeat_ngram_size": 0,70    "num_attention_heads": 32,71    "num_beam_groups": 1,72    "num_beams": 1,73    "num_hidden_layers": 12,74    "num_key_value_heads": 8,75    "num_return_sequences": 1,76    "output_attentions": false,77    "output_hidden_states": false,78    "output_scores": false,79    "pad_token_id": null,80    "prefix": null,81    "problem_type": null,82    "pruned_heads": {},83    "remove_invalid_values": false,84    "repetition_penalty": 1.0,85    "return_dict": true,86    "return_dict_in_generate": false,87    "rms_norm_eps": 1e-06,88    "rope_theta": 10000.0,89    "sep_token_id": null,90    "sliding_window": 32,91    "suppress_tokens": null,92    "task_specific_params": null,93    "temperature": 1.0,94    "tf_legacy_loss": false,95    "tie_encoder_decoder": false,96    "tie_word_embeddings": false,97    "tokenizer_class": null,98    "top_k": 50,99    "top_p": 1.0,100    "torch_dtype": "float32",101    "torchscript": false,102    "typical_p": 1.0,103    "use_bfloat16": false,104    "use_cache": true,105    "vocab_size": 32005106  },107  "encoder": {108    "_name_or_path": "LeroyDyer/Mixtral_AI_Cyber_Q_Vision",109    "add_cross_attention": false,110    "architectures": [111      "CLIPVisionModel"112    ],113    "attention_dropout": 0.0,114    "bad_words_ids": null,115    "begin_suppress_tokens": null,116    "bos_token_id": null,117    "chunk_size_feed_forward": 0,118    "cross_attention_hidden_size": null,119    "decoder_start_token_id": null,120    "diversity_penalty": 0.0,121    "do_sample": false,122    "dropout": 0.0,123    "early_stopping": false,124    "encoder_no_repeat_ngram_size": 0,125    "eos_token_id": null,126    "exponential_decay_length_penalty": null,127    "finetuning_task": null,128    "forced_bos_token_id": null,129    "forced_eos_token_id": null,130    "hidden_act": "quick_gelu",131    "hidden_size": 1024,132    "id2label": {133      "0": "LABEL_0",134      "1": "LABEL_1"135    },136    "image_size": 336,137    "initializer_factor": 1.0,138    "initializer_range": 0.02,139    "intermediate_size": 4096,140    "is_decoder": false,141    "is_encoder_decoder": false,142    "label2id": {143      "LABEL_0": 0,144      "LABEL_1": 1145    },146    "layer_norm_eps": 1e-05,147    "length_penalty": 1.0,148    "max_length": 20,149    "min_length": 0,150    "model_type": "clip_vision_model",151    "no_repeat_ngram_size": 0,152    "num_attention_heads": 16,153    "num_beam_groups": 1,154    "num_beams": 1,155    "num_channels": 3,156    "num_hidden_layers": 24,157    "num_return_sequences": 1,158    "output_attentions": false,159    "output_hidden_states": false,160    "output_scores": false,161    "pad_token_id": null,162    "patch_size": 14,163    "prefix": null,164    "problem_type": null,165    "projection_dim": 768,166    "pruned_heads": {},167    "remove_invalid_values": false,168    "repetition_penalty": 1.0,169    "return_dict": true,170    "return_dict_in_generate": false,171    "sep_token_id": null,172    "suppress_tokens": null,173    "task_specific_params": null,174    "temperature": 1.0,175    "tf_legacy_loss": false,176    "tie_encoder_decoder": false,177    "tie_word_embeddings": true,178    "tokenizer_class": null,179    "top_k": 50,180    "top_p": 1.0,181    "torch_dtype": "float32",182    "torchscript": false,183    "typical_p": 1.0,184    "use_bfloat16": false185  },186  "is_encoder_decoder": true,187  "model_type": "vision-encoder-decoder",188  "tie_word_embeddings": false,189  "torch_dtype": "float32",190  "transformers_version": "4.38.2"191}  192 193