| { | |
| "architectures": [ | |
| "LongcatCausalLM" | |
| ], | |
| "attention_bias": false, | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_longcat.LongcatConfig", | |
| "AutoModel": "modeling_longcat.LongcatModel", | |
| "AutoModelForCausalLM": "modeling_longcat.LongcatForCausalLM" | |
| }, | |
| "vocab_size": 131072, | |
| "hidden_size": 6144, | |
| "ffn_hidden_size": 12288, | |
| "expert_ffn_hidden_size": 2048, | |
| "num_layers": 28, | |
| "num_attention_heads": 64, | |
| "kv_lora_rank": 512, | |
| "q_lora_rank": 1536, | |
| "qk_rope_head_dim": 64, | |
| "v_head_dim": 128, | |
| "qk_nope_head_dim": 128, | |
| "mla_scale_q_lora": true, | |
| "mla_scale_kv_lora": true, | |
| "routed_scaling_factor": 6.0, | |
| "n_routed_experts": 512, | |
| "max_position_embeddings": 983040, | |
| "rope_scaling": { | |
| "original_max_position_embeddings": 8192, | |
| "rope_type": "deepseek_yarn", | |
| "factor": 120, | |
| "beta_fast": 32, | |
| "beta_slow": 1, | |
| "mscale": 1, | |
| "mscale_all_dim": 1 | |
| }, | |
| "rms_norm_eps": 1e-5, | |
| "use_cache": true, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "rope_theta": 1000000.0, | |
| "attention_method": "MLA", | |
| "zero_expert_num": 256, | |
| "zero_expert_type": "identity", | |
| "moe_topk": 12 | |
| } | |