config.json

{
  "architectures": [
    "KimiLinearForCausalLM"
  ],
  "auto_map": {
    "AutoConfig": "configuration_kimi.KimiLinearConfig",
    "AutoModel": "modeling_kimi.KimiLinearModel",
    "AutoModelForCausalLM": "modeling_kimi.KimiLinearForCausalLM"
  },
  "bos_token_id": 163584,
  "dtype": "bfloat16",
  "eos_token_id": 200002,
  "first_k_dense_replace": 1,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 1024,
  "initializer_range": 0.02,
  "intermediate_size": 3072,
  "kv_lora_rank": 512,
  "linear_attn_config": {
    "full_attn_layers": [
      4,
      8,
      12,
      16
    ],
    "head_dim": 128,
    "kda_layers": [
      1,
      2,
      3,
      5,
      6,
      7,
      9,
      10,
      11,
      13,
      14,
      15
    ],
    "num_heads": 32,
    "short_conv_kernel_size": 4
  },
  "mla_use_nope": true,
  "model_max_length": 1048576,
  "model_type": "kimi_linear",
  "moe_intermediate_size": 2048,
  "moe_layer_freq": 1,
  "moe_renormalize": true,
  "moe_router_activation_func": "sigmoid",
  "num_attention_heads": 16,
  "num_expert_group": 1,
  "num_experts": 16,
  "num_experts_per_token": 2,
  "num_hidden_layers": 18,
  "num_key_value_heads": 2,
  "num_nextn_predict_layers": 0,
  "num_shared_experts": 1,
  "pad_token_id": 199999,
  "q_lora_rank": null,
  "qk_nope_head_dim": 128,
  "qk_rope_head_dim": 64,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000.0,
  "routed_scaling_factor": 2.446,
  "tie_word_embeddings": false,
  "topk_group": 1,
  "transformers_version": "4.57.1",
  "use_cache": true,
  "use_grouped_topk": true,
  "v_head_dim": 128,
  "vocab_size": 201088
}