neuronxcc-2.15.143.0+e39249ad/0_REGISTRY/0.0.27.dev0/inference/t5/michaelbenayoun/t5-tiny-random/a191a832d7f52b83633c.json

{"decoder": {"architectures": ["T5ForConditionalGeneration"], "classifier_dropout": 0.0, "d_ff": 128, "d_kv": 8, "d_model": 64, "decoder_start_token_id": 0, "dense_act_fn": "relu", "dropout_rate": 0.1, "feed_forward_proj": "relu", "initializer_factor": 0.2, "is_encoder_decoder": true, "is_gated_act": false, "layer_norm_epsilon": 1e-06, "model_type": "t5", "neuron": {"auto_cast": "matmul", "auto_cast_type": "bf16", "compiler_type": "neuronx-cc", "compiler_version": "2.15.143.0+e39249ad", "dynamic_batch_size": false, "inline_weights_to_neff": false, "optlevel": "2", "output_attentions": false, "output_hidden_states": false, "static_batch_size": 1, "static_num_beams": 4, "static_sequence_length": 18, "tensor_parallel_size": 2}, "num_decoder_layers": 5, "num_heads": 32, "num_layers": 2, "relative_attention_max_distance": 128, "relative_attention_num_buckets": 8, "task": "text2text-generation", "use_cache": true, "vocab_size": 32128}, "encoder": {"architectures": ["T5ForConditionalGeneration"], "classifier_dropout": 0.0, "d_ff": 128, "d_kv": 8, "d_model": 64, "decoder_start_token_id": 0, "dense_act_fn": "relu", "dropout_rate": 0.1, "feed_forward_proj": "relu", "initializer_factor": 0.2, "is_encoder_decoder": true, "is_gated_act": false, "layer_norm_epsilon": 1e-06, "model_type": "t5", "neuron": {"auto_cast": "matmul", "auto_cast_type": "bf16", "compiler_type": "neuronx-cc", "compiler_version": "2.15.143.0+e39249ad", "dynamic_batch_size": false, "inline_weights_to_neff": false, "optlevel": "2", "output_attentions": false, "output_hidden_states": false, "static_batch_size": 1, "static_num_beams": 4, "static_sequence_length": 18, "tensor_parallel_size": 2}, "num_decoder_layers": 5, "num_heads": 32, "num_layers": 2, "relative_attention_max_distance": 128, "relative_attention_num_buckets": 8, "task": "text2text-generation", "use_cache": true, "vocab_size": 32128}, "model_type": "t5"}