{ "architectures": ["DeepseekV3ForCausalLM"], "bos_token_id": 163584, "dtype": "bfloat16", "eos_token_id": 163585, "hidden_act": "silu", "hidden_size": 7168, "initializer_range": 0.02, "intermediate_size": 18432, "max_position_embeddings": 262144, "model_type": "deepseek_v3", "moe_intermediate_size": 2048, "num_attention_heads": 64, "num_hidden_layers": 61, "num_key_value_heads": 64, "pad_token_id": 163839, "rms_norm_eps": 1e-5, "rope_theta": 50000, "rope_scaling": { "beta_fast": 32, "beta_slow": 1, "factor": 64, "mscale": 1, "mscale_all_dim": 1, "original_max_position_embeddings": 4096, "type": "yarn" }, "routed_scaling_factor": 2.827, "tie_word_embeddings": false, "use_cache": true, "vocab_size": 163840, "aux_loss_alpha": 0.001, "attention_bias": false, "attention_dropout": 0, "ep_size": 1, "first_k_dense_replace": 1, "kv_lora_rank": 512, "n_group": 1, "n_routed_experts": 384, "n_shared_experts": 1, "norm_topk_prob": true, "num_experts_per_tok": 8, "moe_layer_freq": 1, "num_nextn_predict_layers": 0, "q_lora_rank": 1536, "qk_nope_head_dim": 128, "qk_rope_head_dim": 64, "scoring_func": "sigmoid", "seq_aux": true, "topk_group": 1, "topk_method": "noaux_tc", "v_head_dim": 128, "quantization_config": { "config_groups": { "group_0": { "input_activations": null, "output_activations": null, "targets": ["Linear"], "weights": { "actorder": null, "block_structure": null, "dynamic": false, "group_size": 32, "num_bits": 4, "observer": "minmax", "observer_kwargs": {}, "strategy": "group", "symmetric": true, "type": "int" } } }, "format": "pack-quantized", "ignore": ["lm_head", "re:.*self_attn.*", "re:.*shared_experts.*", "re:.*mlp\\.(gate|up|gate_up|down)_proj.*"], "quant_method": "compressed-tensors", "quantization_status": "compressed" } }