{ "architectures": [ "T5ForConditionalGeneration" ], "d_ff": 2048, "d_kv": 64, "d_model": 512, "decoder_start_token_id": 0, "dropout_rate": 0.1, "eos_token_id": 1, "initializer_factor": 1.0, "is_encoder_decoder": true, "layer_norm_epsilon": 1e-06, "model_type": "t5", "n_positions": 512, "num_heads": 8, "num_layers": 6, "output_past": true, "pad_token_id": 0, "relative_attention_num_buckets": 32, "task_specific_params": { "translation_en_to_fr": { "early_stopping": true, "max_length": 32, "num_beams": 4, "prefix": "" } }, "vocab_size": 32102 }