{ "_name_or_path": "output/bigbird_model-tsdae-8-2022-08-25_14-09-38/", "architectures": [ "BigBirdModel" ], "attention_probs_dropout_prob": 0.1, "attention_type": "block_sparse", "block_size": 64, "bos_token_id": 0, "classifier_dropout": null, "eos_token_id": 2, "gradient_checkpointing": false, "hidden_act": "gelu_new", "hidden_dropout_prob": 0.1, "hidden_size": 256, "initializer_range": 0.02, "intermediate_size": 3072, "layer_norm_eps": 1e-12, "max_position_embeddings": 1024, "model_type": "big_bird", "num_attention_heads": 4, "num_hidden_layers": 12, "num_random_blocks": 3, "pad_token_id": 1, "position_embedding_type": "absolute", "rescale_embeddings": false, "sep_token_id": 2, "tokenizer_class": "BigBirdTokenizer", "torch_dtype": "float32", "transformers_version": "4.21.1", "type_vocab_size": 2, "use_bias": true, "use_cache": true, "vocab_size": 50265 }