{ "architectures": ["T5ForConditionalGeneration"], "bos_token_id": 2, "d_ff": 2048, "d_kv": 64, "d_model": 768, "decoder_start_token_id": 0, "dropout_rate": 0.1, "eos_token_id": 3, "feed_forward_proj": "gated-gelu", "initializer_factor": 1.0, "is_encoder_decoder": true, "layer_norm_epsilon": 1e-6, "model_type": "t5", "n_positions": 512, "num_decoder_layers": 6, "num_heads": 12, "num_layers": 6, "pad_token_id": 0, "relative_attention_num_buckets": 32, "tokenizer": "default", "transformers_version": "4.5.1", "vocab_size": 16000 }