config.json

{
  "architectures": [
    "GraniteMoeHybridForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "attention_multiplier": 0.015625,
  "bos_token_id": 100257,
  "torch_dtype": "float16",
  "embedding_multiplier": 12,
  "eos_token_id": 100257,
  "hidden_act": "silu",
  "hidden_size": 768,
  "init_method": "mup",
  "initializer_range": 0.1,
  "intermediate_size": 2048,
  "layer_types": [
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "attention",
    "mamba",
    "mamba",
    "attention",
    "mamba",
    "mamba",
    "mamba",
    "attention",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "mamba",
    "attention",
    "mamba",
    "mamba",
    "mamba",
    "mamba"
  ],
  "logits_scaling": 3,
  "mamba_chunk_size": 256,
  "mamba_conv_bias": true,
  "mamba_d_conv": 4,
  "mamba_d_head": 32,
  "mamba_d_state": 128,
  "mamba_expand": 2,
  "mamba_n_groups": 1,
  "mamba_n_heads": 48,
  "mamba_proj_bias": false,
  "max_position_embeddings": 32768,
  "model_type": "granitemoehybrid",
  "normalization_function": "rmsnorm",
  "num_attention_heads": 12,
  "num_experts_per_tok": 0,
  "num_hidden_layers": 32,
  "num_key_value_heads": 4,
  "num_local_experts": 0,
  "output_router_logits": false,
  "pad_token_id": 100256,
  "position_embedding_type": "nope",
  "quantization_config": {
    "bnb_4bit_compute_dtype": "float16",
    "bnb_4bit_quant_type": "nf4",
    "bnb_4bit_use_double_quant": true,
    "llm_int8_enable_fp32_cpu_offload": false,
    "llm_int8_has_fp16_weight": false,
    "llm_int8_skip_modules": [
      "model.layers.0.mamba.in_proj",
      "model.layers.0.mamba.out_proj",
      "model.layers.1.mamba.in_proj",
      "model.layers.1.mamba.out_proj",
      "model.layers.2.mamba.in_proj",
      "model.layers.2.mamba.out_proj",
      "model.layers.3.mamba.in_proj",
      "model.layers.3.mamba.out_proj",
      "model.layers.4.mamba.in_proj",
      "model.layers.4.mamba.out_proj",
      "model.layers.5.mamba.in_proj",
      "model.layers.5.mamba.out_proj",
      "model.layers.6.mamba.in_proj",
      "model.layers.6.mamba.out_proj",
      "model.layers.7.mamba.in_proj",
      "model.layers.7.mamba.out_proj",
      "model.layers.8.mamba.in_proj",
      "model.layers.8.mamba.out_proj",
      "model.layers.9.shared_mlp.input_linear",
      "model.layers.9.shared_mlp.output_linear",
      "model.layers.9.mamba.in_proj",
      "model.layers.9.mamba.out_proj",
      "model.layers.11.mamba.in_proj",
      "model.layers.11.mamba.out_proj",
      "model.layers.12.mamba.in_proj",
      "model.layers.12.mamba.out_proj",
      "model.layers.13.shared_mlp.input_linear",
      "model.layers.13.shared_mlp.output_linear",
      "model.layers.14.mamba.in_proj",
      "model.layers.14.mamba.out_proj",
      "model.layers.15.mamba.in_proj",
      "model.layers.15.mamba.out_proj",
      "model.layers.16.mamba.in_proj",
      "model.layers.16.mamba.out_proj",
      "model.layers.18.mamba.in_proj",
      "model.layers.18.mamba.out_proj",
      "model.layers.19.mamba.in_proj",
      "model.layers.19.mamba.out_proj",
      "model.layers.20.mamba.in_proj",
      "model.layers.20.mamba.out_proj",
      "model.layers.21.mamba.in_proj",
      "model.layers.21.mamba.out_proj",
      "model.layers.22.mamba.in_proj",
      "model.layers.22.mamba.out_proj",
      "model.layers.23.mamba.in_proj",
      "model.layers.23.mamba.out_proj",
      "model.layers.24.mamba.in_proj",
      "model.layers.24.mamba.out_proj",
      "model.layers.25.mamba.in_proj",
      "model.layers.25.mamba.out_proj",
      "model.layers.26.mamba.in_proj",
      "model.layers.26.mamba.out_proj",
      "model.layers.28.mamba.in_proj",
      "model.layers.28.mamba.out_proj",
      "model.layers.29.mamba.in_proj",
      "model.layers.29.mamba.out_proj",
      "model.layers.30.mamba.in_proj",
      "model.layers.30.mamba.out_proj",
      "model.layers.31.shared_mlp.input_linear",
      "model.layers.31.shared_mlp.output_linear",
      "model.layers.31.mamba.in_proj",
      "model.layers.31.mamba.out_proj",
      "lm_head"
    ],
    "llm_int8_threshold": 6.0,
    "load_in_4bit": true,
    "load_in_8bit": false,
    "quant_method": "bitsandbytes"
  },
  "residual_multiplier": 0.246,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 10000,
  "router_aux_loss_coef": 0.01,
  "shared_intermediate_size": 2048,
  "tie_word_embeddings": true,
  "transformers_version": "4.57.1",
  "unsloth_fixed": true,
  "unsloth_version": "2025.11.1",
  "use_cache": true,
  "vocab_size": 100354
}