| { |
| "architectures": [ |
| "DeepseekV4ForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 0, |
| "eos_token_id": 1, |
| "expert_dtype": "int4", |
| "hc_eps": 1e-06, |
| "hc_mult": 4, |
| "hc_sinkhorn_iters": 20, |
| "head_dim": 512, |
| "hidden_act": "silu", |
| "hidden_size": 4096, |
| "index_head_dim": 128, |
| "index_n_heads": 64, |
| "index_topk": 512, |
| "initializer_range": 0.02, |
| "max_position_embeddings": 1048576, |
| "model_type": "deepseek_v4", |
| "moe_intermediate_size": 2048, |
| "n_routed_experts": 256, |
| "n_shared_experts": 1, |
| "norm_topk_prob": true, |
| "num_attention_heads": 64, |
| "num_experts_per_tok": 6, |
| "num_hidden_layers": 43, |
| "num_hash_layers": 3, |
| "num_key_value_heads": 1, |
| "num_nextn_predict_layers": 3, |
| "o_groups": 8, |
| "o_lora_rank": 1024, |
| "q_lora_rank": 1024, |
| "qk_rope_head_dim": 64, |
| "quantization_config": { |
| "quant_method": "dsv4_int", |
| "format": "int_packed", |
| "config_groups": { |
| "experts_w4a16": { |
| "weights": { |
| "num_bits": 4, |
| "type": "int", |
| "format": "int4", |
| "symmetric": true, |
| "group_size": 32, |
| "strategy": "group", |
| "scale_mode": "mse" |
| }, |
| "input_activations": { |
| "num_bits": 16, |
| "type": "float" |
| }, |
| "targets": [ |
| "*.ffn.experts.*.w1", |
| "*.ffn.experts.*.w2", |
| "*.ffn.experts.*.w3" |
| ] |
| }, |
| "linears_w8a16": { |
| "weights": { |
| "num_bits": 8, |
| "type": "int", |
| "symmetric": true, |
| "strategy": "channel" |
| }, |
| "input_activations": { |
| "num_bits": 16, |
| "type": "float" |
| }, |
| "targets": [ |
| "*.attn.wq_a", |
| "*.attn.wq_b", |
| "*.attn.wkv", |
| "*.attn.wo_a", |
| "*.attn.wo_b", |
| "*.attn.indexer.wq_b", |
| "*.attn.indexer.compressor.wkv", |
| "*.attn.indexer.compressor.wgate", |
| "*.ffn.shared_experts.w1", |
| "*.ffn.shared_experts.w2", |
| "*.ffn.shared_experts.w3", |
| "mtp.*.e_proj", |
| "mtp.*.h_proj", |
| "mtp.*.main_proj" |
| ] |
| } |
| }, |
| "ignore": [ |
| "embed", |
| "head", |
| "norm", |
| "lm_head", |
| "*norm.weight", |
| "attn.attn_sink", |
| "*.gate.*", |
| "hc_*", |
| "*.hc_attn_*", |
| "*.hc_ffn_*" |
| ] |
| }, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": { |
| "beta_fast": 32, |
| "beta_slow": 1, |
| "factor": 16, |
| "original_max_position_embeddings": 65536, |
| "type": "yarn" |
| }, |
| "rope_theta": 10000, |
| "routed_scaling_factor": 1.5, |
| "scoring_func": "sqrtsoftplus", |
| "sliding_window": 128, |
| "swiglu_limit": 10.0, |
| "tie_word_embeddings": false, |
| "topk_method": "noaux_tc", |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.57.1", |
| "use_cache": true, |
| "vocab_size": 129280, |
| "compress_rope_theta": 160000, |
| "compress_ratios": [ |
| 0, |
| 0, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 128, |
| 4, |
| 0, |
| 0, |
| 0 |
| ], |
| "dspark_block_size": 5, |
| "dspark_noise_token_id": 128799, |
| "dspark_target_layer_ids": [ |
| 40, |
| 41, |
| 42 |
| ], |
| "dspark_markov_rank": 256, |
| "vllm": { |
| "kernels": [ |
| "flash_mla.sparse_mla_decode_int8", |
| "flash_mla.sparse_mla_prefill_int8", |
| "vllm._custom_ops.indexer_k_quant_and_cache_int8", |
| "vllm.models.deepseek_v4.common.ops.fused_indexer_q.fused_indexer_q_rope_quant_int8", |
| "vllm.model_executor.layers.quantization.utils.marlin_utils.marlin_act_int8_process_scales", |
| "vllm.model_executor.layers.sparse_attn_indexer.streaming_prefill_topk" |
| ], |
| "cache_type": "int8_ds_mla" |
| } |
| } |
|
|
