{ "architectures": [ "AfmoeForCausalLM" ], "attention_dropout": 0.0, "auto_map": { "AutoConfig": "configuration_afmoe.AfmoeConfig", "AutoModel": "modeling_afmoe.AfmoeModel", "AutoModelForCausalLM": "modeling_afmoe.AfmoeForCausalLM" }, "bos_token_id": 0, "dtype": "bfloat16", "eos_token_id": 3, "global_attn_every_n_layers": 4, "head_dim": 128, "hidden_act": "silu", "hidden_size": 3072, "initializer_range": 0.02, "intermediate_size": 12288, "layer_types": [ "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention", "sliding_attention", "sliding_attention", "sliding_attention", "full_attention" ], "load_balance_coeff": 5e-05, "max_position_embeddings": 262144, "model_type": "afmoe", "moe_intermediate_size": 3072, "mup_enabled": true, "n_group": 1, "num_attention_heads": 48, "num_dense_layers": 6, "num_expert_groups": 1, "num_experts": 256, "num_experts_per_tok": 4, "num_hidden_layers": 60, "num_key_value_heads": 8, "num_limited_groups": 1, "num_shared_experts": 1, "pad_token_id": 12, "quantization_config": { "config_groups": { "group_0": { "format": "float-quantized", "input_activations": { "actorder": null, "block_structure": null, "dynamic": true, "group_size": null, "num_bits": 8, "observer": null, "observer_kwargs": {}, "strategy": "token", "symmetric": true, "type": "float" }, "output_activations": null, "targets": [ "Linear" ], "weights": { "actorder": null, "block_structure": null, "dynamic": false, "group_size": null, "num_bits": 8, "observer": "minmax", "observer_kwargs": {}, "strategy": "channel", "symmetric": true, "type": "float" } } }, "format": "float-quantized", "global_compression_ratio": null, "ignore": [ "model.layers.0.self_attn.q_proj", "model.layers.0.self_attn.k_proj", "model.layers.0.self_attn.v_proj", "model.layers.0.self_attn.o_proj", "model.layers.0.self_attn.gate_proj", "model.layers.1.self_attn.q_proj", "model.layers.1.self_attn.k_proj", "model.layers.1.self_attn.v_proj", "model.layers.1.self_attn.o_proj", "model.layers.1.self_attn.gate_proj", "model.layers.2.self_attn.q_proj", "model.layers.2.self_attn.k_proj", "model.layers.2.self_attn.v_proj", "model.layers.2.self_attn.o_proj", "model.layers.2.self_attn.gate_proj", "model.layers.3.self_attn.q_proj", "model.layers.3.self_attn.k_proj", "model.layers.3.self_attn.v_proj", "model.layers.3.self_attn.o_proj", "model.layers.3.self_attn.gate_proj", "model.layers.4.self_attn.q_proj", "model.layers.4.self_attn.k_proj", "model.layers.4.self_attn.v_proj", "model.layers.4.self_attn.o_proj", "model.layers.4.self_attn.gate_proj", "model.layers.5.self_attn.q_proj", "model.layers.5.self_attn.k_proj", "model.layers.5.self_attn.v_proj", "model.layers.5.self_attn.o_proj", "model.layers.5.self_attn.gate_proj", "model.layers.6.self_attn.q_proj", "model.layers.6.self_attn.k_proj", "model.layers.6.self_attn.v_proj", "model.layers.6.self_attn.o_proj", "model.layers.6.self_attn.gate_proj", "model.layers.6.mlp.router.gate", "model.layers.7.self_attn.q_proj", "model.layers.7.self_attn.k_proj", "model.layers.7.self_attn.v_proj", "model.layers.7.self_attn.o_proj", "model.layers.7.self_attn.gate_proj", "model.layers.7.mlp.router.gate", "model.layers.8.self_attn.q_proj", "model.layers.8.self_attn.k_proj", "model.layers.8.self_attn.v_proj", "model.layers.8.self_attn.o_proj", "model.layers.8.self_attn.gate_proj", "model.layers.8.mlp.router.gate", "model.layers.9.self_attn.q_proj", "model.layers.9.self_attn.k_proj", "model.layers.9.self_attn.v_proj", "model.layers.9.self_attn.o_proj", "model.layers.9.self_attn.gate_proj", "model.layers.9.mlp.router.gate", "model.layers.10.self_attn.q_proj", "model.layers.10.self_attn.k_proj", "model.layers.10.self_attn.v_proj", "model.layers.10.self_attn.o_proj", "model.layers.10.self_attn.gate_proj", "model.layers.10.mlp.router.gate", "model.layers.11.self_attn.q_proj", "model.layers.11.self_attn.k_proj", "model.layers.11.self_attn.v_proj", "model.layers.11.self_attn.o_proj", "model.layers.11.self_attn.gate_proj", "model.layers.11.mlp.router.gate", "model.layers.12.self_attn.q_proj", "model.layers.12.self_attn.k_proj", "model.layers.12.self_attn.v_proj", "model.layers.12.self_attn.o_proj", "model.layers.12.self_attn.gate_proj", "model.layers.12.mlp.router.gate", "model.layers.13.self_attn.q_proj", "model.layers.13.self_attn.k_proj", "model.layers.13.self_attn.v_proj", "model.layers.13.self_attn.o_proj", "model.layers.13.self_attn.gate_proj", "model.layers.13.mlp.router.gate", "model.layers.14.self_attn.q_proj", "model.layers.14.self_attn.k_proj", "model.layers.14.self_attn.v_proj", "model.layers.14.self_attn.o_proj", "model.layers.14.self_attn.gate_proj", "model.layers.14.mlp.router.gate", "model.layers.15.self_attn.q_proj", "model.layers.15.self_attn.k_proj", "model.layers.15.self_attn.v_proj", "model.layers.15.self_attn.o_proj", "model.layers.15.self_attn.gate_proj", "model.layers.15.mlp.router.gate", "model.layers.16.self_attn.q_proj", "model.layers.16.self_attn.k_proj", "model.layers.16.self_attn.v_proj", "model.layers.16.self_attn.o_proj", "model.layers.16.self_attn.gate_proj", "model.layers.16.mlp.router.gate", "model.layers.17.self_attn.q_proj", "model.layers.17.self_attn.k_proj", "model.layers.17.self_attn.v_proj", "model.layers.17.self_attn.o_proj", "model.layers.17.self_attn.gate_proj", "model.layers.17.mlp.router.gate", "model.layers.18.self_attn.q_proj", "model.layers.18.self_attn.k_proj", "model.layers.18.self_attn.v_proj", "model.layers.18.self_attn.o_proj", "model.layers.18.self_attn.gate_proj", "model.layers.18.mlp.router.gate", "model.layers.19.self_attn.q_proj", "model.layers.19.self_attn.k_proj", "model.layers.19.self_attn.v_proj", "model.layers.19.self_attn.o_proj", "model.layers.19.self_attn.gate_proj", "model.layers.19.mlp.router.gate", "model.layers.20.self_attn.q_proj", "model.layers.20.self_attn.k_proj", "model.layers.20.self_attn.v_proj", "model.layers.20.self_attn.o_proj", "model.layers.20.self_attn.gate_proj", "model.layers.20.mlp.router.gate", "model.layers.21.self_attn.q_proj", "model.layers.21.self_attn.k_proj", "model.layers.21.self_attn.v_proj", "model.layers.21.self_attn.o_proj", "model.layers.21.self_attn.gate_proj", "model.layers.21.mlp.router.gate", "model.layers.22.self_attn.q_proj", "model.layers.22.self_attn.k_proj", "model.layers.22.self_attn.v_proj", "model.layers.22.self_attn.o_proj", "model.layers.22.self_attn.gate_proj", "model.layers.22.mlp.router.gate", "model.layers.23.self_attn.q_proj", "model.layers.23.self_attn.k_proj", "model.layers.23.self_attn.v_proj", "model.layers.23.self_attn.o_proj", "model.layers.23.self_attn.gate_proj", "model.layers.23.mlp.router.gate", "model.layers.24.self_attn.q_proj", "model.layers.24.self_attn.k_proj", "model.layers.24.self_attn.v_proj", "model.layers.24.self_attn.o_proj", "model.layers.24.self_attn.gate_proj", "model.layers.24.mlp.router.gate", "model.layers.25.self_attn.q_proj", "model.layers.25.self_attn.k_proj", "model.layers.25.self_attn.v_proj", "model.layers.25.self_attn.o_proj", "model.layers.25.self_attn.gate_proj", "model.layers.25.mlp.router.gate", "model.layers.26.self_attn.q_proj", "model.layers.26.self_attn.k_proj", "model.layers.26.self_attn.v_proj", "model.layers.26.self_attn.o_proj", "model.layers.26.self_attn.gate_proj", "model.layers.26.mlp.router.gate", "model.layers.27.self_attn.q_proj", "model.layers.27.self_attn.k_proj", "model.layers.27.self_attn.v_proj", "model.layers.27.self_attn.o_proj", "model.layers.27.self_attn.gate_proj", "model.layers.27.mlp.router.gate", "model.layers.28.self_attn.q_proj", "model.layers.28.self_attn.k_proj", "model.layers.28.self_attn.v_proj", "model.layers.28.self_attn.o_proj", "model.layers.28.self_attn.gate_proj", "model.layers.28.mlp.router.gate", "model.layers.29.self_attn.q_proj", "model.layers.29.self_attn.k_proj", "model.layers.29.self_attn.v_proj", "model.layers.29.self_attn.o_proj", "model.layers.29.self_attn.gate_proj", "model.layers.29.mlp.router.gate", "model.layers.30.self_attn.q_proj", "model.layers.30.self_attn.k_proj", "model.layers.30.self_attn.v_proj", "model.layers.30.self_attn.o_proj", "model.layers.30.self_attn.gate_proj", "model.layers.30.mlp.router.gate", "model.layers.31.self_attn.q_proj", "model.layers.31.self_attn.k_proj", "model.layers.31.self_attn.v_proj", "model.layers.31.self_attn.o_proj", "model.layers.31.self_attn.gate_proj", "model.layers.31.mlp.router.gate", "model.layers.32.self_attn.q_proj", "model.layers.32.self_attn.k_proj", "model.layers.32.self_attn.v_proj", "model.layers.32.self_attn.o_proj", "model.layers.32.self_attn.gate_proj", "model.layers.32.mlp.router.gate", "model.layers.33.self_attn.q_proj", "model.layers.33.self_attn.k_proj", "model.layers.33.self_attn.v_proj", "model.layers.33.self_attn.o_proj", "model.layers.33.self_attn.gate_proj", "model.layers.33.mlp.router.gate", "model.layers.34.self_attn.q_proj", "model.layers.34.self_attn.k_proj", "model.layers.34.self_attn.v_proj", "model.layers.34.self_attn.o_proj", "model.layers.34.self_attn.gate_proj", "model.layers.34.mlp.router.gate", "model.layers.35.self_attn.q_proj", "model.layers.35.self_attn.k_proj", "model.layers.35.self_attn.v_proj", "model.layers.35.self_attn.o_proj", "model.layers.35.self_attn.gate_proj", "model.layers.35.mlp.router.gate", "model.layers.36.self_attn.q_proj", "model.layers.36.self_attn.k_proj", "model.layers.36.self_attn.v_proj", "model.layers.36.self_attn.o_proj", "model.layers.36.self_attn.gate_proj", "model.layers.36.mlp.router.gate", "model.layers.37.self_attn.q_proj", "model.layers.37.self_attn.k_proj", "model.layers.37.self_attn.v_proj", "model.layers.37.self_attn.o_proj", "model.layers.37.self_attn.gate_proj", "model.layers.37.mlp.router.gate", "model.layers.38.self_attn.q_proj", "model.layers.38.self_attn.k_proj", "model.layers.38.self_attn.v_proj", "model.layers.38.self_attn.o_proj", "model.layers.38.self_attn.gate_proj", "model.layers.38.mlp.router.gate", "model.layers.39.self_attn.q_proj", "model.layers.39.self_attn.k_proj", "model.layers.39.self_attn.v_proj", "model.layers.39.self_attn.o_proj", "model.layers.39.self_attn.gate_proj", "model.layers.39.mlp.router.gate", "model.layers.40.self_attn.q_proj", "model.layers.40.self_attn.k_proj", "model.layers.40.self_attn.v_proj", "model.layers.40.self_attn.o_proj", "model.layers.40.self_attn.gate_proj", "model.layers.40.mlp.router.gate", "model.layers.41.self_attn.q_proj", "model.layers.41.self_attn.k_proj", "model.layers.41.self_attn.v_proj", "model.layers.41.self_attn.o_proj", "model.layers.41.self_attn.gate_proj", "model.layers.41.mlp.router.gate", "model.layers.42.self_attn.q_proj", "model.layers.42.self_attn.k_proj", "model.layers.42.self_attn.v_proj", "model.layers.42.self_attn.o_proj", "model.layers.42.self_attn.gate_proj", "model.layers.42.mlp.router.gate", "model.layers.43.self_attn.q_proj", "model.layers.43.self_attn.k_proj", "model.layers.43.self_attn.v_proj", "model.layers.43.self_attn.o_proj", "model.layers.43.self_attn.gate_proj", "model.layers.43.mlp.router.gate", "model.layers.44.self_attn.q_proj", "model.layers.44.self_attn.k_proj", "model.layers.44.self_attn.v_proj", "model.layers.44.self_attn.o_proj", "model.layers.44.self_attn.gate_proj", "model.layers.44.mlp.router.gate", "model.layers.45.self_attn.q_proj", "model.layers.45.self_attn.k_proj", "model.layers.45.self_attn.v_proj", "model.layers.45.self_attn.o_proj", "model.layers.45.self_attn.gate_proj", "model.layers.45.mlp.router.gate", "model.layers.46.self_attn.q_proj", "model.layers.46.self_attn.k_proj", "model.layers.46.self_attn.v_proj", "model.layers.46.self_attn.o_proj", "model.layers.46.self_attn.gate_proj", "model.layers.46.mlp.router.gate", "model.layers.47.self_attn.q_proj", "model.layers.47.self_attn.k_proj", "model.layers.47.self_attn.v_proj", "model.layers.47.self_attn.o_proj", "model.layers.47.self_attn.gate_proj", "model.layers.47.mlp.router.gate", "model.layers.48.self_attn.q_proj", "model.layers.48.self_attn.k_proj", "model.layers.48.self_attn.v_proj", "model.layers.48.self_attn.o_proj", "model.layers.48.self_attn.gate_proj", "model.layers.48.mlp.router.gate", "model.layers.49.self_attn.q_proj", "model.layers.49.self_attn.k_proj", "model.layers.49.self_attn.v_proj", "model.layers.49.self_attn.o_proj", "model.layers.49.self_attn.gate_proj", "model.layers.49.mlp.router.gate", "model.layers.50.self_attn.q_proj", "model.layers.50.self_attn.k_proj", "model.layers.50.self_attn.v_proj", "model.layers.50.self_attn.o_proj", "model.layers.50.self_attn.gate_proj", "model.layers.50.mlp.router.gate", "model.layers.51.self_attn.q_proj", "model.layers.51.self_attn.k_proj", "model.layers.51.self_attn.v_proj", "model.layers.51.self_attn.o_proj", "model.layers.51.self_attn.gate_proj", "model.layers.51.mlp.router.gate", "model.layers.52.self_attn.q_proj", "model.layers.52.self_attn.k_proj", "model.layers.52.self_attn.v_proj", "model.layers.52.self_attn.o_proj", "model.layers.52.self_attn.gate_proj", "model.layers.52.mlp.router.gate", "model.layers.53.self_attn.q_proj", "model.layers.53.self_attn.k_proj", "model.layers.53.self_attn.v_proj", "model.layers.53.self_attn.o_proj", "model.layers.53.self_attn.gate_proj", "model.layers.53.mlp.router.gate", "model.layers.54.self_attn.q_proj", "model.layers.54.self_attn.k_proj", "model.layers.54.self_attn.v_proj", "model.layers.54.self_attn.o_proj", "model.layers.54.self_attn.gate_proj", "model.layers.54.mlp.router.gate", "model.layers.55.self_attn.q_proj", "model.layers.55.self_attn.k_proj", "model.layers.55.self_attn.v_proj", "model.layers.55.self_attn.o_proj", "model.layers.55.self_attn.gate_proj", "model.layers.55.mlp.router.gate", "model.layers.56.self_attn.q_proj", "model.layers.56.self_attn.k_proj", "model.layers.56.self_attn.v_proj", "model.layers.56.self_attn.o_proj", "model.layers.56.self_attn.gate_proj", "model.layers.56.mlp.router.gate", "model.layers.57.self_attn.q_proj", "model.layers.57.self_attn.k_proj", "model.layers.57.self_attn.v_proj", "model.layers.57.self_attn.o_proj", "model.layers.57.self_attn.gate_proj", "model.layers.57.mlp.router.gate", "model.layers.58.self_attn.q_proj", "model.layers.58.self_attn.k_proj", "model.layers.58.self_attn.v_proj", "model.layers.58.self_attn.o_proj", "model.layers.58.self_attn.gate_proj", "model.layers.58.mlp.router.gate", "model.layers.59.self_attn.q_proj", "model.layers.59.self_attn.k_proj", "model.layers.59.self_attn.v_proj", "model.layers.59.self_attn.o_proj", "model.layers.59.self_attn.gate_proj", "model.layers.59.mlp.router.gate", "lm_head" ], "kv_cache_scheme": null, "quant_method": "compressed-tensors", "quantization_status": "compressed", "sparsity_config": {}, "transform_config": {}, "version": "0.13.1.a20260116" }, "rms_norm_eps": 1e-05, "rope_scaling": null, "rope_theta": 10000, "route_norm": true, "route_scale": 2.448, "score_func": "sigmoid", "sliding_window": 4096, "tie_word_embeddings": false, "topk_group": 1, "transformers_version": "4.57.3", "use_cache": true, "use_grouped_mm": true, "vocab_size": 200192 }