| { |
| "_attn_implementation_autoset": true, |
| "bos_token_id": 0, |
| "classifier_dropout": 0.0, |
| "d_ff": 37, |
| "d_kv": 8, |
| "d_model": 32, |
| "decoder_start_token_id": 0, |
| "dense_act_fn": "relu", |
| "dropout_rate": 0.1, |
| "eos_token_id": 1, |
| "feed_forward_proj": "relu", |
| "gradient_checkpointing": false, |
| "initializer_factor": 0.002, |
| "is_encoder_decoder": true, |
| "is_gated_act": false, |
| "layer_norm_epsilon": 1e-06, |
| "model_type": "t5", |
| "neuron": { |
| "auto_cast": null, |
| "auto_cast_type": null, |
| "compiler_type": "neuronx-cc", |
| "compiler_version": "2.19.8089.0+8ab9f450", |
| "dynamic_batch_size": false, |
| "float_dtype": "fp32", |
| "inline_weights_to_neff": false, |
| "input_names": [ |
| "input_ids", |
| "attention_mask" |
| ], |
| "int_dtype": "int64", |
| "model_type": "t5-encoder", |
| "optlevel": "2", |
| "output_attentions": false, |
| "output_hidden_states": false, |
| "output_names": [ |
| "present.0.self.key", |
| "present.1.self.key", |
| "present.2.self.key", |
| "present.3.self.key", |
| "present.4.self.key", |
| "present.0.self.value", |
| "present.1.self.value", |
| "present.2.self.value", |
| "present.3.self.value", |
| "present.4.self.value", |
| "present.0.cross.key", |
| "present.1.cross.key", |
| "present.2.cross.key", |
| "present.3.cross.key", |
| "present.4.cross.key", |
| "present.0.cross.value", |
| "present.1.cross.value", |
| "present.2.cross.value", |
| "present.3.cross.value", |
| "present.4.cross.value" |
| ], |
| "static_batch_size": 1, |
| "static_num_beams": 4, |
| "static_sequence_length": 18, |
| "task": "text2text-generation", |
| "tensor_parallel_size": 1 |
| }, |
| "num_decoder_layers": 5, |
| "num_heads": 4, |
| "num_layers": 5, |
| "pad_token_id": 0, |
| "relative_attention_max_distance": 128, |
| "relative_attention_num_buckets": 8, |
| "torch_dtype": "float32", |
| "torchscript": true, |
| "transformers_version": "4.51.0", |
| "use_cache": true, |
| "vocab_size": 1103 |
| } |
|
|