{
  "_name_or_path": "EleutherAI/pythia-12b",
  "architectures": [
    "GPTNeoXForCausalLM"
  ],
  "attention_bias": true,
  "attention_dropout": 0.0,
  "bos_token_id": 0,
  "classifier_dropout": 0.1,
  "eos_token_id": 0,
  "hidden_act": "gelu",
  "hidden_dropout": 0.0,
  "hidden_size": 5120,
  "initializer_range": 0.02,
  "intermediate_size": 20480,
  "layer_norm_eps": 1e-05,
  "max_position_embeddings": 2048,
  "model_type": "gpt_neox",
  "num_attention_heads": 40,
  "num_hidden_layers": 36,
  "partial_rotary_factor": 0.25,
  "quantization_config": {
    "amp": false,
    "autoround_version": "0.4.1",
    "batch_size": 4,
    "bits": 4,
    "damp_percent": 0.01,
    "data_type": "int",
    "desc_act": false,
    "enable_minmax_tuning": true,
    "enable_norm_bias_tuning": false,
    "enable_quanted_input": true,
    "gradient_accumulate_steps": 1,
    "group_size": 128,
    "iters": 200,
    "low_gpu_mem_usage": false,
    "lr": 0.005,
    "minmax_lr": 0.005,
    "nsamples": 128,
    "quant_method": "gptq",
    "scale_dtype": "torch.float16",
    "seqlen": 512,
    "sym": false,
    "to_quant_block_names": [
      [
        "gpt_neox.layers.0",
        "gpt_neox.layers.1",
        "gpt_neox.layers.2",
        "gpt_neox.layers.3",
        "gpt_neox.layers.4",
        "gpt_neox.layers.5",
        "gpt_neox.layers.6",
        "gpt_neox.layers.7",
        "gpt_neox.layers.8",
        "gpt_neox.layers.9",
        "gpt_neox.layers.10",
        "gpt_neox.layers.11",
        "gpt_neox.layers.12",
        "gpt_neox.layers.13",
        "gpt_neox.layers.14",
        "gpt_neox.layers.15",
        "gpt_neox.layers.16",
        "gpt_neox.layers.17",
        "gpt_neox.layers.18",
        "gpt_neox.layers.19",
        "gpt_neox.layers.20",
        "gpt_neox.layers.21",
        "gpt_neox.layers.22",
        "gpt_neox.layers.23",
        "gpt_neox.layers.24",
        "gpt_neox.layers.25",
        "gpt_neox.layers.26",
        "gpt_neox.layers.27",
        "gpt_neox.layers.28",
        "gpt_neox.layers.29",
        "gpt_neox.layers.30",
        "gpt_neox.layers.31",
        "gpt_neox.layers.32",
        "gpt_neox.layers.33",
        "gpt_neox.layers.34",
        "gpt_neox.layers.35"
      ]
    ],
    "true_sequential": false
  },
  "rope_scaling": null,
  "rope_theta": 10000,
  "rotary_emb_base": 10000,
  "rotary_pct": 0.25,
  "tie_word_embeddings": false,
  "torch_dtype": "float32",
  "transformers_version": "4.47.0",
  "use_cache": true,
  "use_parallel_residual": true,
  "vocab_size": 50688
}