config.json

{
  "architectures": [
    "ScalarModel"
  ],
  "base_config": {
    "_name_or_path": "jdchang/tldr_sft_pythia_2.8",
    "architectures": [
      "GPTNeoXForCausalLM"
    ],
    "eos_token_id": 0,
    "hidden_size": 2560,
    "intermediate_size": 10240,
    "model_type": "gpt_neox",
    "num_attention_heads": 32,
    "num_hidden_layers": 32,
    "torch_dtype": "bfloat16",
    "vocab_size": 50304
  },
  "base_model": "jdchang/tldr_sft_pythia_2.8",
  "bias": 2.53739437142289,
  "hidden_size": 2560,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.35.0"
}