config.json

{
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 100257,
  "torch_dtype": "bfloat16",
  "eos_token_id": 100265,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 17920,
  "max_position_embeddings": 16384,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 40,
  "num_hidden_layers": 40,
  "num_key_value_heads": 10,
  "original_max_position_embeddings": 16384,
  "pad_token_id": 100351,
  "pretraining_tp": 1,
  "quantization_config": {
    "bnb_4bit_compute_dtype": "bfloat16",
    "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.1.mlp.gate_proj",
      "model.layers.1.mlp.up_proj",
      "model.layers.1.mlp.down_proj",
      "model.layers.2.mlp.gate_proj",
      "model.layers.2.mlp.up_proj",
      "model.layers.2.mlp.down_proj",
      "model.layers.3.mlp.gate_proj",
      "model.layers.3.mlp.up_proj",
      "model.layers.3.mlp.down_proj",
      "model.layers.37.self_attn.o_proj",
      "model.layers.39.self_attn.o_proj",
      "lm_head"
    ],
    "llm_int8_threshold": 6.0,
    "load_in_4bit": true,
    "load_in_8bit": false,
    "quant_method": "bitsandbytes"
  },
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 250000,
  "tie_word_embeddings": false,
  "transformers_version": "4.56.2",
  "unsloth_fixed": true,
  "unsloth_version": "2026.8.8",
  "use_cache": true,
  "vocab_size": 100352
}