| { |
| "architectures": [ |
| "DeepseekV32ForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "aux_loss_alpha": 0.001, |
| "beta_fast": 32, |
| "beta_slow": 1, |
| "bos_token_id": 0, |
| "dim": 7168, |
| "dtype": "bfloat16", |
| "eos_token_id": 1, |
| "ep_size": 1, |
| "first_k_dense_replace": 3, |
| "head_dim": 64, |
| "hidden_act": "silu", |
| "hidden_size": 7168, |
| "index_head_dim": 128, |
| "index_n_heads": 64, |
| "index_top_k": 2048, |
| "index_topk": 2048, |
| "initializer_range": 0.02, |
| "intermediate_size": 18432, |
| "kv_lora_rank": 512, |
| "max_batch_size": 8, |
| "max_position_embeddings": 163840, |
| "max_seq_len": 16384, |
| "mlp_bias": false, |
| "model_type": "deepseek_v32", |
| "moe_intermediate_size": 2048, |
| "moe_layer_freq": 1, |
| "mscale": 1.0, |
| "n_group": 8, |
| "n_routed_experts": 256, |
| "n_shared_experts": 1, |
| "norm_topk_prob": true, |
| "num_attention_heads": 128, |
| "num_experts_per_tok": 8, |
| "num_hidden_layers": 61, |
| "num_key_value_heads": 128, |
| "num_nextn_predict_layers": 1, |
| "original_seq_len": 4096, |
| "q_lora_rank": 1536, |
| "qk_head_dim": 192, |
| "qk_nope_head_dim": 128, |
| "qk_rope_head_dim": 64, |
| "quantization_config": { |
| "autoround_version": "0.9.5", |
| "bits": 4, |
| "data_type": "int", |
| "extra_config": { |
| "model.layers.0.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.0.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.1.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.1.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.10.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.10.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.11.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.11.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.12.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.12.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.13.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.13.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.14.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.14.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.15.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.15.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.16.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.16.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.17.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.17.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.18.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.18.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.19.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.19.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.2.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.2.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.20.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.20.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.21.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.21.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.22.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.22.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.23.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.23.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.24.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.24.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.25.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.25.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.26.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.26.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.27.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.27.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.28.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.28.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.29.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.29.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.3.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.3.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.30.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.30.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.31.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.31.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.32.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.32.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.33.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.33.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.34.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.34.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.35.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.35.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.36.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.36.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.37.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.37.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.38.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.38.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.39.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.39.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.4.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.4.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.40.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.40.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.41.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.41.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.42.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.42.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.43.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.43.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.44.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.44.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.45.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.45.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.46.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.46.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.47.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.47.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.48.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.48.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.49.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.49.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.5.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.5.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.50.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.50.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.51.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.51.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.52.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.52.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.53.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.53.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.54.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.54.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.55.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.55.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.56.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.56.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.57.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.57.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.58.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.58.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.59.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.59.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.6.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.6.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.60.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.60.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.7.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.7.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.8.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.8.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.9.self_attn.indexer.wk": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "model.layers.9.self_attn.indexer.wq_b": { |
| "bits": 16, |
| "data_type": "float" |
| } |
| }, |
| "group_size": 128, |
| "iters": 0, |
| "low_gpu_mem_usage": true, |
| "packing_format": "auto_round:auto_gptq", |
| "quant_method": "auto-round", |
| "sym": true |
| }, |
| "rms_norm_eps": 1e-06, |
| "rope_factor": 40, |
| "rope_interleave": true, |
| "rope_scaling": { |
| "beta_fast": 32, |
| "beta_slow": 1, |
| "factor": 40, |
| "mscale": 1.0, |
| "mscale_all_dim": 1.0, |
| "original_max_position_embeddings": 4096, |
| "rope_type": "yarn", |
| "type": "yarn" |
| }, |
| "rope_theta": 10000.0, |
| "routed_scaling_factor": 2.5, |
| "scale_fmt": null, |
| "scoring_func": "sigmoid", |
| "seq_aux": true, |
| "tie_word_embeddings": false, |
| "topk_group": 4, |
| "topk_method": "noaux_tc", |
| "transformers_version": "4.57.3", |
| "use_cache": true, |
| "v_head_dim": 128, |
| "vocab_size": 129280 |
| } |
|
|
