hfl/rbt3

1{2  "architectures": [3    "BertForMaskedLM"4  ],5  "model_type": "bert",6  "attention_probs_dropout_prob": 0.1,7  "directionality": "bidi",8  "finetuning_task": null,9  "hidden_act": "gelu",10  "hidden_dropout_prob": 0.1,11  "hidden_size": 768,12  "initializer_range": 0.02,13  "intermediate_size": 3072,14  "is_decoder": false,15  "layer_norm_eps": 1e-12,16  "max_position_embeddings": 512,17  "num_attention_heads": 12,18  "num_hidden_layers": 3,19  "num_labels": 2,20  "output_attentions": false,21  "output_hidden_states": false,22  "output_past": true,23  "pooler_fc_size": 768,24  "pooler_num_attention_heads": 12,25  "pooler_num_fc_layers": 3,26  "pooler_size_per_head": 128,27  "pooler_type": "first_token_transform",28  "pruned_heads": {},29  "torchscript": false,30  "type_vocab_size": 2,31  "use_bfloat16": false,32  "vocab_size": 2112833}34