| { | |
| "builder_config": { | |
| "apply_query_key_layer_scaling": false, | |
| "fp8": false, | |
| "gather_all_token_logits": false, | |
| "hidden_act": "gelu", | |
| "hidden_size": 4096, | |
| "int8": false, | |
| "max_batch_size": 256, | |
| "max_input_len": 2000, | |
| "max_num_tokens": null, | |
| "max_output_len": 50, | |
| "max_position_embeddings": 8192, | |
| "multi_query_mode": true, | |
| "name": "gpt", | |
| "num_heads": 32, | |
| "num_kv_heads": 1, | |
| "num_layers": 42, | |
| "parallel_build": true, | |
| "precision": "float16", | |
| "tensor_parallel": 2, | |
| "use_parallel_embedding": false, | |
| "use_prompt_tuning": false, | |
| "use_refit": false, | |
| "vocab_size": 49152 | |
| }, | |
| "plugin_config": { | |
| "attention_qk_half_accumulation": false, | |
| "bert_attention_plugin": false, | |
| "context_fmha_type": 1, | |
| "gemm_plugin": "float16", | |
| "gpt_attention_plugin": "float16", | |
| "identity_plugin": false, | |
| "layernorm_plugin": false, | |
| "layernorm_quantization_plugin": false, | |
| "lookup_plugin": false, | |
| "nccl_plugin": "float16", | |
| "paged_kv_cache": true, | |
| "quantize_per_token_plugin": false, | |
| "quantize_tensor_plugin": false, | |
| "remove_input_padding": true, | |
| "rmsnorm_plugin": false, | |
| "rmsnorm_quantization_plugin": false, | |
| "smooth_quant_gemm_plugin": false, | |
| "tokens_per_block": 64, | |
| "use_custom_all_reduce": false, | |
| "weight_only_groupwise_quant_matmul_plugin": false, | |
| "weight_only_quant_matmul_plugin": false | |
| } | |
| } | |