| { | |
| "architectures": [ | |
| "XLMRobertaForCausalLM" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "bos_token_id": 0, | |
| "classifier_dropout": null, | |
| "d_ff": 256, | |
| "d_kv": 8, | |
| "d_model": 64, | |
| "eos_token_id": 2, | |
| "gradient_checkpointing": false, | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 256, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 256, | |
| "layer_norm_eps": 1e-05, | |
| "max_position_embeddings": 64, | |
| "model_type": "xlm-roberta", | |
| "num_attention_heads": 2, | |
| "num_decoder_layers": 2, | |
| "num_heads": 2, | |
| "num_hidden_layers": 2, | |
| "num_layers": 2, | |
| "output_past": true, | |
| "pad_token_id": 1, | |
| "position_embedding_type": "absolute", | |
| "relative_attention_num_buckets": 32, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.52.0.dev0", | |
| "type_vocab_size": 1, | |
| "use_cache": true, | |
| "vocab_size": 5002 | |
| } | |