| { |
| "_name_or_path": "/hier-model", |
| "architectures": [ |
| "HierBertForMaskedLM" |
| ], |
| "attention_probs_dropout_prob": 0.1, |
| "auto_map": { |
| "AutoConfig": "configuration_hier.HierBertConfig", |
| "AutoModel": "modelling_hier.HierBertModel", |
| "AutoModelForMaskedLM": "modelling_hier.HierBertForMaskedLM", |
| "AutoModelForSequenceClassification": "modelling_hier.HierBertForSequenceClassification" |
| }, |
| "classifier_dropout": null, |
| "hidden_act": "gelu", |
| "hidden_dropout_prob": 0.1, |
| "hidden_size": 512, |
| "initializer_range": 0.02, |
| "intermediate_size": 2048, |
| "layer_norm_eps": 1e-06, |
| "max_position_embeddings": 512, |
| "model_type": "hierarchical-bert", |
| "norm_first": true, |
| "num_attention_heads": 8, |
| "num_hidden_layers": 6, |
| "pad_token_id": 0, |
| "position_embedding_type": "absolute", |
| "sep_token_id": 3, |
| "torch_dtype": "float32", |
| "transformers_version": "4.31.0", |
| "type_vocab_size": 2, |
| "use_cache": true, |
| "vocab_size": 32000 |
| } |
|
|
