{ "architectures": [ "Glm5NextForConditionalGeneration" ], "text_config": { "attention_bias": false, "attention_dropout": 0.0, "dtype": "bfloat16", "eos_token_id": [ 154820, 154827, 154829 ], "first_k_dense_replace": 3, "hc_eps": 1e-06, "hc_mult": 4, "hc_sinkhorn_iters": 20, "head_dim": 0, "hidden_act": "silu", "hidden_size": 4096, "index_head_dim": 128, "index_kpool": 4, "index_kpool_always_select_tail": true, "index_kpool_compress": true, "index_n_heads": 32, "index_topk": 2048, "index_share_for_mtp_iteration": true, "indexer_rope_interleave": true, "indexer_types": [ "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full", "full" ], "initializer_range": 0.02, "intermediate_size": 12288, "kv_lora_rank": 512, "layer_types": [ "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention", "linear_attention", "linear_attention", "deepseek_sparse_attention", "linear_attention" ], "linear_attn_config": { "num_heads": 64, "gate_lower_bound": -5.0, "head_dim": 128, "short_conv_kernel_size": 4, "kda_layers": [ 0, 1, 2, 4, 5, 6, 8, 9, 10, 12, 13, 14, 16, 17, 18, 20, 21, 22, 24, 25, 26, 28, 29, 30, 32, 33, 34, 36, 37, 38, 40, 41, 42, 44 ], "full_attn_layers": [ 3, 7, 11, 15, 19, 23, 27, 31, 35, 39, 43 ] }, "max_position_embeddings": 1048576, "mhc": true, "mla_use_nope": true, "mlp_layer_types": [ "dense", "dense", "dense", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse", "sparse" ], "model_type": "glm5_next_text", "moe_intermediate_size": 2048, "moe_router_dtype": "float32", "n_group": 1, "n_routed_experts": 288, "n_shared_experts": 1, "norm_topk_prob": true, "num_attention_heads": 64, "num_experts_per_tok": 8, "num_hidden_layers": 45, "num_key_value_heads": 64, "num_nextn_predict_layers": 1, "output_router_logits": false, "pad_token_id": 154820, "q_lora_rank": 1536, "qk_head_dim": 256, "qk_nope_head_dim": 256, "qk_rope_head_dim": 0, "rms_norm_eps": 1e-05, "routed_scaling_factor": 2.5, "router_aux_loss_coef": 0.001, "scoring_func": "sigmoid", "swiglu_limit": 10.0, "tie_word_embeddings": false, "topk_group": 1, "topk_method": "noaux_tc", "use_cache": true, "v_head_dim": 256, "vocab_size": 154880 }, "vision_config": { "attention_bias": true, "attention_dropout": 0.0, "depth": 24, "hidden_act": "silu", "hidden_size": 1024, "image_size": 448, "in_channels": 3, "initializer_range": 0.02, "intermediate_size": 4096, "model_type": "glm5_next_vision", "num_heads": 16, "out_hidden_size": 4096, "patch_size": 14, "projection_intermediate_size": 10240, "rms_norm_eps": 1e-05, "spatial_merge_size": 2, "swiglu_limit": 10.0, "temporal_patch_size": 2 }, "image_token_id": 154854, "video_token_id": 154855, "image_start_token_id": 154830, "image_end_token_id": 154831, "video_start_token_id": 154832, "video_end_token_id": 154833, "tie_word_embeddings": false, "model_type": "glm5_next", "transformers_version": "5.16.0", "quantization": { "group_size": 64, "bits": 4, "model.layers.0.mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.1.mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.2.mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.3.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.3.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.3.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.3.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.4.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.4.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.4.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.4.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.5.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.5.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.5.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.5.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.6.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.6.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.6.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.6.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.7.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.7.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.7.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.7.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.8.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.8.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.8.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.8.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.9.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.9.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.9.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.9.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.10.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.10.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.10.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.10.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.11.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.11.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.11.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.11.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.12.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.12.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.12.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.12.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.13.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.13.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.13.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.13.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.14.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.14.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.14.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.14.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.15.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.15.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.15.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.15.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.16.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.16.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.16.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.16.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.17.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.17.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.17.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.17.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.18.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.18.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.18.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.18.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.19.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.19.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.19.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.19.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.20.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.20.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.20.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.20.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.21.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.21.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.21.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.21.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.22.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.22.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.22.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.22.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.23.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.23.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.23.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.23.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.24.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.24.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.24.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.24.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.25.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.25.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.25.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.25.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.26.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.26.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.26.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.26.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.27.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.27.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.27.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.27.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.28.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.28.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.28.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.28.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.29.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.29.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.29.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.29.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.30.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.30.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.30.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.30.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.31.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.31.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.31.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.31.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.32.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.32.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.32.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.32.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.33.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.33.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.33.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.33.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.34.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.34.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.34.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.34.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.35.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.35.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.35.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.35.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.36.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.36.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.36.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.36.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.37.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.37.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.37.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.37.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.38.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.38.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.38.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.38.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.39.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.39.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.39.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.39.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.40.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.40.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.40.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.40.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.41.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.41.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.41.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.41.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.42.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.42.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.42.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.42.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.43.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.43.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.43.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.43.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.44.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.44.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.44.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.44.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 } }, "quantization_config": { "group_size": 64, "bits": 4, "model.layers.0.mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.1.mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.2.mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.3.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.3.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.3.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.3.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.4.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.4.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.4.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.4.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.5.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.5.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.5.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.5.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.6.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.6.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.6.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.6.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.7.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.7.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.7.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.7.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.8.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.8.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.8.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.8.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.9.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.9.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.9.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.9.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.10.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.10.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.10.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.10.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.11.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.11.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.11.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.11.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.12.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.12.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.12.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.12.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.13.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.13.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.13.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.13.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.14.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.14.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.14.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.14.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.15.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.15.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.15.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.15.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.16.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.16.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.16.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.16.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.17.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.17.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.17.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.17.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.18.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.18.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.18.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.18.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.19.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.19.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.19.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.19.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.20.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.20.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.20.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.20.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.21.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.21.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.21.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.21.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.22.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.22.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.22.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.22.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.23.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.23.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.23.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.23.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.24.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.24.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.24.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.24.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.25.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.25.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.25.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.25.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.26.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.26.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.26.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.26.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.27.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.27.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.27.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.27.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.28.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.28.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.28.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.28.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.29.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.29.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.29.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.29.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.30.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.30.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.30.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.30.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.31.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.31.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.31.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.31.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.32.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.32.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.32.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.32.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.33.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.33.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.33.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.33.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.34.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.34.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.34.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.34.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.35.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.35.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.35.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.35.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.36.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.36.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.36.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.36.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.37.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.37.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.37.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.37.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.38.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.38.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.38.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.38.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.39.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.39.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.39.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.39.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.40.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.40.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.40.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.40.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.41.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.41.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.41.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.41.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.42.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.42.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.42.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.42.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.43.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.43.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.43.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.43.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 }, "model.layers.44.mlp.switch_mlp.down_proj": { "group_size": 64, "bits": 5 }, "model.layers.44.mlp.shared_experts.gate_proj": { "group_size": 64, "bits": 6 }, "model.layers.44.mlp.shared_experts.up_proj": { "group_size": 64, "bits": 6 }, "model.layers.44.mlp.shared_experts.down_proj": { "group_size": 64, "bits": 6 } } }