| { |
| "version": "0.21.0", |
| "pretrained_config": { |
| "architecture": "DecoderModel", |
| "dtype": "float16", |
| "vocab_size": 51865, |
| "hidden_size": 1024, |
| "num_hidden_layers": 24, |
| "num_attention_heads": 16, |
| "hidden_act": "gelu", |
| "logits_dtype": "float16", |
| "norm_epsilon": 1e-05, |
| "runtime_defaults": null, |
| "position_embedding_type": "learned_absolute", |
| "num_key_value_heads": 16, |
| "intermediate_size": 4096, |
| "max_position_embeddings": 448, |
| "mapping": { |
| "world_size": 1, |
| "gpus_per_node": 8, |
| "cp_size": 1, |
| "tp_size": 1, |
| "pp_size": 1, |
| "moe_tp_size": 1, |
| "moe_cluster_size": 1, |
| "moe_ep_size": 1, |
| "attn_tp_size": 1, |
| "attn_cp_size": 1, |
| "auto_parallel": false |
| }, |
| "quantization": { |
| "quant_algo": "W8A16", |
| "kv_cache_quant_algo": null, |
| "group_size": 128, |
| "smoothquant_val": 0.5, |
| "clamp_val": null, |
| "use_meta_recipe": false, |
| "has_zero_point": false, |
| "pre_quant_scale": false, |
| "exclude_modules": null |
| }, |
| "use_parallel_embedding": false, |
| "embedding_sharding_dim": 0, |
| "head_size": 64, |
| "qk_layernorm": false, |
| "rotary_embedding_dim": 64, |
| "use_prompt_tuning": false, |
| "has_position_embedding": true, |
| "layernorm_type": 0, |
| "has_attention_qkvo_bias": true, |
| "has_mlp_bias": true, |
| "has_model_final_layernorm": true, |
| "has_embedding_layernorm": false, |
| "has_embedding_scale": false, |
| "ffn_hidden_size": 4096, |
| "q_scaling": 1.0, |
| "layernorm_position": 0, |
| "relative_attention": false, |
| "max_distance": 0, |
| "num_buckets": 0, |
| "model_type": "whisper", |
| "rescale_before_lm_head": false, |
| "encoder_hidden_size": 1024, |
| "encoder_num_heads": 16, |
| "encoder_head_size": null, |
| "skip_cross_kv": false, |
| "type_vocab_size": null, |
| "encoder_num_kv_heads": null, |
| "mlp_type": 0, |
| "residual_scaling": 1.0, |
| "has_lm_head_bias": false |
| }, |
| "build_config": { |
| "max_input_len": 100, |
| "max_seq_len": 400, |
| "opt_batch_size": 8, |
| "max_batch_size": 14, |
| "max_beam_width": 5, |
| "max_num_tokens": 5600, |
| "opt_num_tokens": 70, |
| "max_prompt_embedding_table_size": 0, |
| "kv_cache_type": "PAGED", |
| "gather_context_logits": false, |
| "gather_generation_logits": true, |
| "strongly_typed": true, |
| "force_num_profiles": null, |
| "profiling_verbosity": "layer_names_only", |
| "enable_debug_output": false, |
| "max_draft_len": 0, |
| "speculative_decoding_mode": 1, |
| "use_refit": false, |
| "input_timing_cache": null, |
| "output_timing_cache": "model.cache", |
| "lora_config": { |
| "lora_dir": [], |
| "lora_ckpt_source": "hf", |
| "max_lora_rank": 64, |
| "lora_target_modules": [], |
| "trtllm_modules_to_hf_modules": {}, |
| "max_loras": 4, |
| "max_cpu_loras": 4 |
| }, |
| "auto_parallel_config": { |
| "world_size": 1, |
| "gpus_per_node": 8, |
| "cluster_key": "L4", |
| "cluster_info": null, |
| "sharding_cost_model": "alpha_beta", |
| "comm_cost_model": "alpha_beta", |
| "enable_pipeline_parallelism": false, |
| "enable_shard_unbalanced_shape": false, |
| "enable_shard_dynamic_shape": false, |
| "enable_reduce_scatter": true, |
| "builder_flags": null, |
| "debug_mode": false, |
| "infer_shape": true, |
| "validation_mode": false, |
| "same_buffer_io": { |
| "past_key_value_(\\d+)": "present_key_value_\\1" |
| }, |
| "same_spec_io": {}, |
| "sharded_io_allowlist": [ |
| "past_key_value_\\d+", |
| "present_key_value_\\d*" |
| ], |
| "fill_weights": false, |
| "parallel_config_cache": null, |
| "profile_cache": null, |
| "dump_path": null, |
| "debug_outputs": [] |
| }, |
| "weight_sparsity": false, |
| "weight_streaming": false, |
| "plugin_config": { |
| "dtype": "float16", |
| "bert_attention_plugin": "float16", |
| "gpt_attention_plugin": "float16", |
| "gemm_plugin": "float16", |
| "explicitly_disable_gemm_plugin": false, |
| "gemm_swiglu_plugin": null, |
| "fp8_rowwise_gemm_plugin": null, |
| "qserve_gemm_plugin": null, |
| "identity_plugin": null, |
| "nccl_plugin": null, |
| "lora_plugin": null, |
| "dora_plugin": false, |
| "weight_only_groupwise_quant_matmul_plugin": null, |
| "weight_only_quant_matmul_plugin": "float16", |
| "smooth_quant_plugins": true, |
| "smooth_quant_gemm_plugin": null, |
| "layernorm_quantization_plugin": null, |
| "rmsnorm_quantization_plugin": null, |
| "quantize_per_token_plugin": false, |
| "quantize_tensor_plugin": false, |
| "moe_plugin": null, |
| "mamba_conv1d_plugin": "auto", |
| "low_latency_gemm_plugin": null, |
| "low_latency_gemm_swiglu_plugin": null, |
| "gemm_allreduce_plugin": null, |
| "context_fmha": true, |
| "bert_context_fmha_fp32_acc": false, |
| "paged_kv_cache": true, |
| "remove_input_padding": true, |
| "norm_quant_fusion": false, |
| "reduce_fusion": false, |
| "user_buffer": false, |
| "tokens_per_block": 32, |
| "use_paged_context_fmha": true, |
| "use_fp8_context_fmha": false, |
| "fuse_fp4_quant": false, |
| "multiple_profiles": false, |
| "paged_state": false, |
| "streamingllm": false, |
| "manage_weights": false, |
| "use_fused_mlp": true, |
| "pp_reduce_scatter": false |
| }, |
| "use_strip_plan": false, |
| "max_encoder_input_len": 1500, |
| "monitor_memory": false, |
| "use_mrope": false |
| } |
| } |
