| { | |
| "approx_mode": "full", | |
| "architectures": [ | |
| "MraForMaskedLM" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "block_per_row": 8, | |
| "bos_token_id": 0, | |
| "dim": 768, | |
| "dropout_prob": 0.1, | |
| "embedding_dim": 768, | |
| "eos_token_id": 2, | |
| "head_dim": 64, | |
| "hidden_act": "gelu", | |
| "hidden_dim": 3072, | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 768, | |
| "initial_prior_diagonal_n_blocks": 1, | |
| "initial_prior_first_n_blocks": 3, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 3072, | |
| "layer_norm_eps": 1e-05, | |
| "max_position_embeddings": 4096, | |
| "mixed_precision": true, | |
| "model_type": "mra", | |
| "num_attention_heads": 12, | |
| "num_head": 12, | |
| "num_hidden_layers": 12, | |
| "num_layers": 12, | |
| "num_sen_type": 1, | |
| "pad_token_id": 1, | |
| "position_embedding_type": "absolute", | |
| "shared_weight": false, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.31.0.dev0", | |
| "type_vocab_size": 1, | |
| "vocab_size": 50265 | |
| } | |