| { | |
| "architectures": [ | |
| "GPTBertForCausalLM", | |
| "GPTBertForMaskedLM", | |
| "LtgbertFoCausalLM" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "auto_map": { | |
| "AutoConfig": "configuration_gpt_bert.GPTBertConfig", | |
| "AutoModel": "modeling_gpt_bert.GPTBertForCausalLM", | |
| "AutoModelForCausalLM": "modeling_gpt_bert.GPTBertForCausalLM", | |
| "AutoModelForMaskedLM": "modeling_gpt_bert.GPTBertForMaskedLM" | |
| }, | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 768, | |
| "intermediate_size": 3072, | |
| "layer_norm_eps": 1e-05, | |
| "max_position_embeddings": 512, | |
| "model_type": "gpt_bert", | |
| "num_attention_heads": 12, | |
| "num_hidden_layers": 12, | |
| "position_bucket_size": 32, | |
| "temperature": 2.65, | |
| "torch_dtype": "float32", | |
| "vocab_size": 32768 | |
| } | |