{ "architectures": [ "MiniMaxM2ForCausalLM" ], "attn_type_list": [ 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 ], "auto_map": { "AutoConfig": "configuration_minimax_m2.MiniMaxM2Config", "AutoModelForCausalLM": "modeling_minimax_m2.MiniMaxM2ForCausalLM" }, "dtype": "bfloat16", "head_dim": 128, "hidden_act": "silu", "hidden_size": 3072, "intermediate_size": 1536, "max_position_embeddings": 204800, "model_type": "minimax_m2", "mtp_transformer_layers": 1, "num_attention_heads": 48, "num_experts_per_tok": 8, "num_hidden_layers": 62, "num_key_value_heads": 8, "num_local_experts": 256, "num_mtp_modules": 3, "qk_norm_type": "per_layer", "quantization_config": { "quant_method": "mxfp4_16", "format": "mxfp4-16-pack-quantized", "quantization_status": "compressed", "config_groups": { "group_0": { "format": "mxfp4-16-pack-quantized", "weights": { "num_bits": 4, "type": "float", "strategy": "group", "group_size": 16, "grid": "iq4_nl", "codebook": [ -127, -104, -83, -65, -49, -35, -22, -10, 1, 13, 25, 38, 53, 69, 89, 113 ], "symmetric": false, "scale_dtype": "torch.float16", "dynamic": false, "actorder": null, "block_structure": null, "observer": "minmax", "observer_kwargs": {}, "zp_dtype": null }, "targets": [ "Linear" ], "input_activations": null, "output_activations": null } }, "ignore": [ "re:.*self_attn.*", "re:.*block_sparse_moe.gate.*", "re:.*e_score_correction_bias.*", "re:.*lm_head.*", "re:.*embed_tokens.*", "re:.*mtp.*" ], "kv_cache_scheme": null, "sparsity_config": {}, "transform_config": {} }, "rms_norm_eps": 1e-06, "rope_theta": 5000000, "rotary_dim": 64, "scoring_func": "sigmoid", "shared_intermediate_size": 0, "tie_word_embeddings": false, "transformers_version": "4.46.1", "use_cache": true, "use_mtp": true, "use_qk_norm": true, "use_routing_bias": true, "vocab_size": 200064 }