| { |
| "_name_or_path": "electra_owt_full_b256_hs196_ah4_is128_l18_es64_vs30522_pytorch", |
| "architectures": [ |
| "ElectraForPreTraining" |
| ], |
| "attention_probs_dropout_prob": 0.1, |
| "classifier_dropout": null, |
| "embedding_size": 64, |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.1, |
| "hidden_size": 196, |
| "initializer_range": 0.02, |
| "intermediate_size": 128, |
| "layer_norm_eps": 1e-12, |
| "max_position_embeddings": 512, |
| "model_type": "electra", |
| "num_attention_heads": 4, |
| "num_hidden_layers": 18, |
| "pad_token_id": 0, |
| "position_embedding_type": "absolute", |
| "summary_activation": "gelu", |
| "summary_last_dropout": 0.1, |
| "summary_type": "first", |
| "summary_use_proj": true, |
| "torch_dtype": "float32", |
| "transformers_version": "4.17.0", |
| "type_vocab_size": 2, |
| "use_cache": true, |
| "vocab_size": 30522 |
| } |
|
|
