baseten
/

spec-dec-qwen-32-h100x2-1.5-h100x1

Model card Files Files and versions Community

joostinyi-baseten commited on 10 days ago

Commit

aa2ef6c

•

1 Parent(s): bbfe8b1

Upload folder using huggingface_hub

Browse files

Files changed (6) hide show

.gitattributes +3 -0
draft/config.json +152 -0
draft/rank0.engine +3 -0
target/config.json +152 -0
target/rank0.engine +3 -0
target/rank1.engine +3 -0

.gitattributes CHANGED Viewed

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text

 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+draft/rank0.engine filter=lfs diff=lfs merge=lfs -text
+target/rank0.engine filter=lfs diff=lfs merge=lfs -text
+target/rank1.engine filter=lfs diff=lfs merge=lfs -text

draft/config.json ADDED Viewed

	@@ -0,0 +1,152 @@

+{
+    "version": "0.13.0",
+    "pretrained_config": {
+        "mlp_bias": false,
+        "attn_bias": true,
+        "rotary_base": 1000000.0,
+        "rotary_scaling": null,
+        "disable_weight_only_quant_plugin": false,
+        "moe": {
+            "num_experts": 0,
+            "top_k": 0,
+            "normalization_mode": 0,
+            "sparse_mixer_epsilon": 0.01,
+            "tp_mode": 0
+        },
+        "architecture": "Qwen2ForCausalLM",
+        "dtype": "float16",
+        "vocab_size": 151936,
+        "hidden_size": 1536,
+        "num_hidden_layers": 28,
+        "num_attention_heads": 12,
+        "hidden_act": "silu",
+        "logits_dtype": "float32",
+        "norm_epsilon": 1e-06,
+        "position_embedding_type": "rope_gpt_neox",
+        "max_position_embeddings": 32768,
+        "num_key_value_heads": 2,
+        "intermediate_size": 8960,
+        "mapping": {
+            "world_size": 1,
+            "gpus_per_node": 8,
+            "cp_size": 1,
+            "tp_size": 1,
+            "pp_size": 1,
+            "moe_tp_size": 1,
+            "moe_ep_size": 1
+        },
+        "quantization": {
+            "quant_algo": null,
+            "kv_cache_quant_algo": null,
+            "group_size": 128,
+            "smoothquant_val": 0.5,
+            "clamp_val": null,
+            "has_zero_point": false,
+            "pre_quant_scale": false,
+            "exclude_modules": null
+        },
+        "use_parallel_embedding": false,
+        "embedding_sharding_dim": 0,
+        "share_embedding_table": false,
+        "head_size": 128,
+        "qk_layernorm": false,
+        "qwen_type": "qwen2",
+        "moe_intermediate_size": 0,
+        "moe_shared_expert_intermediate_size": 0
+    },
+    "build_config": {
+        "max_input_len": 256,
+        "max_seq_len": 10000,
+        "opt_batch_size": 8,
+        "max_batch_size": 256,
+        "max_beam_width": 1,
+        "max_num_tokens": 8192,
+        "opt_num_tokens": 256,
+        "max_prompt_embedding_table_size": 0,
+        "kv_cache_type": "PAGED",
+        "gather_context_logits": false,
+        "gather_generation_logits": false,
+        "strongly_typed": false,
+        "builder_opt": null,
+        "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": null,
+        "lora_config": {
+            "lora_dir": [],
+            "lora_ckpt_source": "hf",
+            "max_lora_rank": 64,
+            "lora_target_modules": [],
+            "trtllm_modules_to_hf_modules": {}
+        },
+        "auto_parallel_config": {
+            "world_size": 1,
+            "gpus_per_node": 8,
+            "cluster_key": null,
+            "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": {},
+            "same_spec_io": {},
+            "sharded_io_allowlist": [],
+            "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": "auto",
+            "gpt_attention_plugin": "auto",
+            "gemm_plugin": "auto",
+            "gemm_swiglu_plugin": null,
+            "fp8_rowwise_gemm_plugin": null,
+            "smooth_quant_gemm_plugin": null,
+            "identity_plugin": null,
+            "layernorm_quantization_plugin": null,
+            "rmsnorm_quantization_plugin": null,
+            "nccl_plugin": null,
+            "lookup_plugin": null,
+            "lora_plugin": null,
+            "weight_only_groupwise_quant_matmul_plugin": null,
+            "weight_only_quant_matmul_plugin": null,
+            "quantize_per_token_plugin": false,
+            "quantize_tensor_plugin": false,
+            "moe_plugin": "auto",
+            "mamba_conv1d_plugin": "auto",
+            "low_latency_gemm_plugin": null,
+            "context_fmha": true,
+            "bert_context_fmha_fp32_acc": false,
+            "paged_kv_cache": true,
+            "remove_input_padding": true,
+            "reduce_fusion": false,
+            "enable_xqa": true,
+            "tokens_per_block": 64,
+            "use_paged_context_fmha": true,
+            "use_fp8_context_fmha": false,
+            "multiple_profiles": false,
+            "paged_state": false,
+            "streamingllm": false,
+            "manage_weights": false,
+            "use_fused_mlp": true
+        },
+        "use_strip_plan": false,
+        "max_encoder_input_len": 1,
+        "use_fused_mlp": false
+    }
+}

draft/rank0.engine ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:bbb48d63e64c9cb902111ddd8b5488f1ab9cce7198e6de76deee4f1e017a582b
+size 3577402588

target/config.json ADDED Viewed

	@@ -0,0 +1,152 @@

+{
+    "version": "0.13.0",
+    "pretrained_config": {
+        "mlp_bias": false,
+        "attn_bias": true,
+        "rotary_base": 1000000.0,
+        "rotary_scaling": null,
+        "disable_weight_only_quant_plugin": false,
+        "moe": {
+            "num_experts": 0,
+            "top_k": 0,
+            "normalization_mode": 0,
+            "sparse_mixer_epsilon": 0.01,
+            "tp_mode": 0
+        },
+        "architecture": "Qwen2ForCausalLM",
+        "dtype": "float16",
+        "vocab_size": 152064,
+        "hidden_size": 5120,
+        "num_hidden_layers": 64,
+        "num_attention_heads": 40,
+        "hidden_act": "silu",
+        "logits_dtype": "float32",
+        "norm_epsilon": 1e-06,
+        "position_embedding_type": "rope_gpt_neox",
+        "max_position_embeddings": 32768,
+        "num_key_value_heads": 8,
+        "intermediate_size": 27648,
+        "mapping": {
+            "world_size": 2,
+            "gpus_per_node": 8,
+            "cp_size": 1,
+            "tp_size": 2,
+            "pp_size": 1,
+            "moe_tp_size": 2,
+            "moe_ep_size": 1
+        },
+        "quantization": {
+            "quant_algo": null,
+            "kv_cache_quant_algo": null,
+            "group_size": 128,
+            "smoothquant_val": 0.5,
+            "clamp_val": null,
+            "has_zero_point": false,
+            "pre_quant_scale": false,
+            "exclude_modules": null
+        },
+        "use_parallel_embedding": false,
+        "embedding_sharding_dim": 0,
+        "share_embedding_table": false,
+        "head_size": 128,
+        "qk_layernorm": false,
+        "qwen_type": "qwen2",
+        "moe_intermediate_size": 0,
+        "moe_shared_expert_intermediate_size": 0
+    },
+    "build_config": {
+        "max_input_len": 10000,
+        "max_seq_len": 10000,
+        "opt_batch_size": 8,
+        "max_batch_size": 256,
+        "max_beam_width": 1,
+        "max_num_tokens": 8192,
+        "opt_num_tokens": 256,
+        "max_prompt_embedding_table_size": 0,
+        "kv_cache_type": "PAGED",
+        "gather_context_logits": false,
+        "gather_generation_logits": false,
+        "strongly_typed": false,
+        "builder_opt": null,
+        "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": null,
+        "lora_config": {
+            "lora_dir": [],
+            "lora_ckpt_source": "hf",
+            "max_lora_rank": 64,
+            "lora_target_modules": [],
+            "trtllm_modules_to_hf_modules": {}
+        },
+        "auto_parallel_config": {
+            "world_size": 1,
+            "gpus_per_node": 8,
+            "cluster_key": null,
+            "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": {},
+            "same_spec_io": {},
+            "sharded_io_allowlist": [],
+            "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": "auto",
+            "gpt_attention_plugin": "auto",
+            "gemm_plugin": "auto",
+            "gemm_swiglu_plugin": null,
+            "fp8_rowwise_gemm_plugin": null,
+            "smooth_quant_gemm_plugin": null,
+            "identity_plugin": null,
+            "layernorm_quantization_plugin": null,
+            "rmsnorm_quantization_plugin": null,
+            "nccl_plugin": "float16",
+            "lookup_plugin": null,
+            "lora_plugin": null,
+            "weight_only_groupwise_quant_matmul_plugin": null,
+            "weight_only_quant_matmul_plugin": null,
+            "quantize_per_token_plugin": false,
+            "quantize_tensor_plugin": false,
+            "moe_plugin": "auto",
+            "mamba_conv1d_plugin": "auto",
+            "low_latency_gemm_plugin": null,
+            "context_fmha": true,
+            "bert_context_fmha_fp32_acc": false,
+            "paged_kv_cache": true,
+            "remove_input_padding": true,
+            "reduce_fusion": false,
+            "enable_xqa": true,
+            "tokens_per_block": 64,
+            "use_paged_context_fmha": true,
+            "use_fp8_context_fmha": false,
+            "multiple_profiles": false,
+            "paged_state": false,
+            "streamingllm": false,
+            "manage_weights": false,
+            "use_fused_mlp": true
+        },
+        "use_strip_plan": false,
+        "max_encoder_input_len": 1,
+        "use_fused_mlp": false
+    }
+}

target/rank0.engine ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:6f298f16cf3b1e3d6d9827eaa052a9917276bbc582a0a683d7afb2079b899bde
+size 33575464900

target/rank1.engine ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b310f15a8ac8b318fbeec398132567287ffa13fd64ed74eca6f66af68551915a
+size 33575463108