{ "vocab_size": 200064, "max_position_embeddings": 196608, "hidden_size": 3072, "intermediate_size": 1536, "num_hidden_layers": 62, "num_attention_heads": 48, "sliding_window": null, "num_key_value_heads": 8, "hidden_act": "silu", "initializer_range": 0.02, "rms_norm_eps": 1e-06, "use_cache": true, "rope_theta": 5000000, "attention_dropout": 0.0, "head_dim": 128, "num_experts_per_tok": 8, "num_local_experts": 256, "output_router_logits": false, "router_aux_loss_coef": 0.001, "router_jitter_noise": 0.0, "use_qk_norm": true, "rotary_dim": 64, "partial_rotary_factor": 0.5, "return_dict": true, "output_hidden_states": false, "torchscript": false, "dtype": "bfloat16", "pruned_heads": {}, "tie_word_embeddings": false, "chunk_size_feed_forward": 0, "is_encoder_decoder": false, "is_decoder": false, "cross_attention_hidden_size": null, "add_cross_attention": false, "tie_encoder_decoder": false, "architectures": [ "MiniMaxM2ForCausalLM" ], "finetuning_task": null, "id2label": { "0": "LABEL_0", "1": "LABEL_1" }, "label2id": { "LABEL_0": 0, "LABEL_1": 1 }, "task_specific_params": null, "problem_type": null, "tokenizer_class": null, "prefix": null, "bos_token_id": null, "pad_token_id": null, "eos_token_id": null, "sep_token_id": null, "decoder_start_token_id": null, "max_length": 20, "min_length": 0, "do_sample": false, "early_stopping": false, "num_beams": 1, "temperature": 1.0, "top_k": 50, "top_p": 1.0, "typical_p": 1.0, "repetition_penalty": 1.0, "length_penalty": 1.0, "no_repeat_ngram_size": 0, "encoder_no_repeat_ngram_size": 0, "bad_words_ids": null, "num_return_sequences": 1, "output_scores": false, "return_dict_in_generate": false, "forced_bos_token_id": null, "forced_eos_token_id": null, "remove_invalid_values": false, "exponential_decay_length_penalty": null, "suppress_tokens": null, "begin_suppress_tokens": null, "num_beam_groups": 1, "diversity_penalty": 0.0, "_name_or_path": null, "transformers_version": "4.57.6", "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" }, "model_type": "minimax_m2", "mtp_transformer_layers": 1, "num_mtp_modules": 3, "qk_norm_type": "per_layer", "scoring_func": "sigmoid", "shared_intermediate_size": 0, "use_mtp": true, "use_routing_bias": true, "tf_legacy_loss": false, "use_bfloat16": false, "output_attentions": false, "quantization_config": { "config_groups": { "group_0": { "input_activations": { "dynamic": false, "num_bits": 4, "type": "float", "group_size": 16 }, "weights": { "dynamic": false, "num_bits": 4, "type": "float", "group_size": 16 }, "targets": [ "Linear" ] } }, "ignore": [ "lm_head", "model.layers.0.block_sparse_moe.gate", "model.layers.0.self_attn*", "model.layers.1.block_sparse_moe.gate", "model.layers.1.self_attn*", "model.layers.10.block_sparse_moe.gate", "model.layers.10.self_attn*", "model.layers.11.block_sparse_moe.gate", "model.layers.11.self_attn*", "model.layers.12.block_sparse_moe.gate", "model.layers.12.self_attn*", "model.layers.13.block_sparse_moe.gate", "model.layers.13.self_attn*", "model.layers.14.block_sparse_moe.gate", "model.layers.14.self_attn*", "model.layers.15.block_sparse_moe.gate", "model.layers.15.self_attn*", "model.layers.16.block_sparse_moe.gate", "model.layers.16.self_attn*", "model.layers.17.block_sparse_moe.gate", "model.layers.17.self_attn*", "model.layers.18.block_sparse_moe.gate", "model.layers.18.self_attn*", "model.layers.19.block_sparse_moe.gate", "model.layers.19.self_attn*", "model.layers.2.block_sparse_moe.gate", "model.layers.2.self_attn*", "model.layers.20.block_sparse_moe.gate", "model.layers.20.self_attn*", "model.layers.21.block_sparse_moe.gate", "model.layers.21.self_attn*", "model.layers.22.block_sparse_moe.gate", "model.layers.22.self_attn*", "model.layers.23.block_sparse_moe.gate", "model.layers.23.self_attn*", "model.layers.24.block_sparse_moe.gate", "model.layers.24.self_attn*", "model.layers.25.block_sparse_moe.gate", "model.layers.25.self_attn*", "model.layers.26.block_sparse_moe.gate", "model.layers.26.self_attn*", "model.layers.27.block_sparse_moe.gate", "model.layers.27.self_attn*", "model.layers.28.block_sparse_moe.gate", "model.layers.28.self_attn*", "model.layers.29.block_sparse_moe.gate", "model.layers.29.self_attn*", "model.layers.3.block_sparse_moe.gate", "model.layers.3.self_attn*", "model.layers.30.block_sparse_moe.gate", "model.layers.30.self_attn*", "model.layers.31.block_sparse_moe.gate", "model.layers.31.self_attn*", "model.layers.32.block_sparse_moe.gate", "model.layers.32.self_attn*", "model.layers.33.block_sparse_moe.gate", "model.layers.33.self_attn*", "model.layers.34.block_sparse_moe.gate", "model.layers.34.self_attn*", "model.layers.35.block_sparse_moe.gate", "model.layers.35.self_attn*", "model.layers.36.block_sparse_moe.gate", "model.layers.36.self_attn*", "model.layers.37.block_sparse_moe.gate", "model.layers.37.self_attn*", "model.layers.38.block_sparse_moe.gate", "model.layers.38.self_attn*", "model.layers.39.block_sparse_moe.gate", "model.layers.39.self_attn*", "model.layers.4.block_sparse_moe.gate", "model.layers.4.self_attn*", "model.layers.40.block_sparse_moe.gate", "model.layers.40.self_attn*", "model.layers.41.block_sparse_moe.gate", "model.layers.41.self_attn*", "model.layers.42.block_sparse_moe.gate", "model.layers.42.self_attn*", "model.layers.43.block_sparse_moe.gate", "model.layers.43.self_attn*", "model.layers.44.block_sparse_moe.gate", "model.layers.44.self_attn*", "model.layers.45.block_sparse_moe.gate", "model.layers.45.self_attn*", "model.layers.46.block_sparse_moe.gate", "model.layers.46.self_attn*", "model.layers.47.block_sparse_moe.gate", "model.layers.47.self_attn*", "model.layers.48.block_sparse_moe.gate", "model.layers.48.self_attn*", "model.layers.49.block_sparse_moe.gate", "model.layers.49.self_attn*", "model.layers.5.block_sparse_moe.gate", "model.layers.5.self_attn*", "model.layers.50.block_sparse_moe.gate", "model.layers.50.self_attn*", "model.layers.51.block_sparse_moe.gate", "model.layers.51.self_attn*", "model.layers.52.block_sparse_moe.gate", "model.layers.52.self_attn*", "model.layers.53.block_sparse_moe.gate", "model.layers.53.self_attn*", "model.layers.54.block_sparse_moe.gate", "model.layers.54.self_attn*", "model.layers.55.block_sparse_moe.gate", "model.layers.55.self_attn*", "model.layers.56.block_sparse_moe.gate", "model.layers.56.self_attn*", "model.layers.57.block_sparse_moe.gate", "model.layers.57.self_attn*", "model.layers.58.block_sparse_moe.gate", "model.layers.58.self_attn*", "model.layers.59.block_sparse_moe.gate", "model.layers.59.self_attn*", "model.layers.6.block_sparse_moe.gate", "model.layers.6.self_attn*", "model.layers.60.block_sparse_moe.gate", "model.layers.60.self_attn*", "model.layers.61.block_sparse_moe.gate", "model.layers.61.self_attn*", "model.layers.7.block_sparse_moe.gate", "model.layers.7.self_attn*", "model.layers.8.block_sparse_moe.gate", "model.layers.8.self_attn*", "model.layers.9.block_sparse_moe.gate", "model.layers.9.self_attn*" ], "quant_algo": "NVFP4", "kv_cache_scheme": { "dynamic": false, "num_bits": 8, "type": "float" }, "producer": { "name": "modelopt", "version": "0.39.0.dev290+gf9d9a71de.d20260213" }, "quant_method": "modelopt" } }