ML-ass/g2e_encoder_decoder

1{2  "architectures": [3    "EncoderDecoderModel"4  ],5  "decoder": {6    "_name_or_path": "ML-ass/english_decoder",7    "activation_function": "gelu_new",8    "add_cross_attention": true,9    "architectures": [10      "GPT2LMHeadModel"11    ],12    "attn_pdrop": 0.1,13    "bad_words_ids": null,14    "bos_token_id": 50256,15    "chunk_size_feed_forward": 0,16    "decoder_start_token_id": null,17    "diversity_penalty": 0.0,18    "do_sample": false,19    "early_stopping": false,20    "embd_pdrop": 0.1,21    "encoder_no_repeat_ngram_size": 0,22    "eos_token_id": 50256,23    "finetuning_task": null,24    "forced_bos_token_id": null,25    "forced_eos_token_id": null,26    "gradient_checkpointing": false,27    "id2label": {28      "0": "LABEL_0",29      "1": "LABEL_1"30    },31    "initializer_range": 0.02,32    "is_decoder": true,33    "is_encoder_decoder": false,34    "label2id": {35      "LABEL_0": 0,36      "LABEL_1": 137    },38    "layer_norm_epsilon": 1e-05,39    "length_penalty": 1.0,40    "max_length": 20,41    "min_length": 0,42    "model_type": "gpt2",43    "n_ctx": 1024,44    "n_embd": 768,45    "n_head": 12,46    "n_inner": null,47    "n_layer": 12,48    "n_positions": 1024,49    "no_repeat_ngram_size": 0,50    "num_beam_groups": 1,51    "num_beams": 1,52    "num_return_sequences": 1,53    "output_attentions": false,54    "output_hidden_states": false,55    "output_scores": false,56    "pad_token_id": null,57    "prefix": null,58    "problem_type": null,59    "pruned_heads": {},60    "remove_invalid_values": false,61    "repetition_penalty": 1.0,62    "resid_pdrop": 0.1,63    "return_dict": true,64    "return_dict_in_generate": false,65    "scale_attn_weights": true,66    "sep_token_id": null,67    "summary_activation": null,68    "summary_first_dropout": 0.1,69    "summary_proj_to_labels": true,70    "summary_type": "cls_index",71    "summary_use_proj": true,72    "task_specific_params": {73      "text-generation": {74        "do_sample": true,75        "max_length": 5076      }77    },78    "temperature": 1.0,79    "tie_encoder_decoder": false,80    "tie_word_embeddings": true,81    "tokenizer_class": null,82    "top_k": 50,83    "top_p": 1.0,84    "torchscript": false,85    "transformers_version": "4.8.2",86    "use_bfloat16": false,87    "use_cache": true,88    "vocab_size": 5025789  },90  "encoder": {91    "_name_or_path": "ML-ass/german_encoder",92    "add_cross_attention": false,93    "architectures": [94      "BertForPreTraining"95    ],96    "attention_probs_dropout_prob": 0.1,97    "bad_words_ids": null,98    "bos_token_id": null,99    "chunk_size_feed_forward": 0,100    "decoder_start_token_id": null,101    "diversity_penalty": 0.0,102    "do_sample": false,103    "early_stopping": false,104    "encoder_no_repeat_ngram_size": 0,105    "eos_token_id": null,106    "finetuning_task": null,107    "forced_bos_token_id": null,108    "forced_eos_token_id": null,109    "gradient_checkpointing": false,110    "hidden_act": "gelu",111    "hidden_dropout_prob": 0.1,112    "hidden_size": 768,113    "id2label": {114      "0": "LABEL_0",115      "1": "LABEL_1"116    },117    "initializer_range": 0.02,118    "intermediate_size": 3072,119    "is_decoder": false,120    "is_encoder_decoder": false,121    "label2id": {122      "LABEL_0": 0,123      "LABEL_1": 1124    },125    "layer_norm_eps": 1e-12,126    "length_penalty": 1.0,127    "max_length": 20,128    "max_position_embeddings": 512,129    "min_length": 0,130    "model_type": "bert",131    "no_repeat_ngram_size": 0,132    "num_attention_heads": 12,133    "num_beam_groups": 1,134    "num_beams": 1,135    "num_hidden_layers": 12,136    "num_return_sequences": 1,137    "output_attentions": false,138    "output_hidden_states": false,139    "output_scores": false,140    "pad_token_id": 0,141    "position_embedding_type": "absolute",142    "prefix": null,143    "problem_type": null,144    "pruned_heads": {},145    "remove_invalid_values": false,146    "repetition_penalty": 1.0,147    "return_dict": true,148    "return_dict_in_generate": false,149    "sep_token_id": null,150    "task_specific_params": null,151    "temperature": 1.0,152    "tie_encoder_decoder": false,153    "tie_word_embeddings": true,154    "tokenizer_class": null,155    "top_k": 50,156    "top_p": 1.0,157    "torchscript": false,158    "transformers_version": "4.8.2",159    "type_vocab_size": 2,160    "use_bfloat16": false,161    "use_cache": true,162    "vocab_size": 30000163  },164  "is_encoder_decoder": true,165  "model_type": "encoder-decoder",166  "transformers_version": null167}168