{
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "dtype": "bfloat16",
  "eos_token_id": 2,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 12288,
  "initializer_range": 0.02,
  "intermediate_size": 28672,
  "max_position_embeddings": 131072,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 96,
  "num_hidden_layers": 88,
  "num_key_value_heads": 8,
  "pretraining_tp": 1,
  "qtip_quant": {
    "kv_format": "fp8",
    "ntok": 4194304,
    "scale_override": 0.9,
    "seq_len": 8192,
    "shared_su_per_hessian_group": true,
    "sigma_reg": 0.01
  },
  "quip_params": {
    "K": 2,
    "L": 16,
    "V": 2,
    "codebook": "bitshift",
    "codebook_version": 0,
    "decode_mode": "quantlut_sym",
    "skip_list": null,
    "split_for_tp": false,
    "td_x": 16,
    "td_y": 16,
    "tlut_bits": 9
  },
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.57.6",
  "use_cache": true,
  "vocab_size": 32768
}