{
  "_name_or_path": "/media/adamo/82142F79142F6EFB/ProgramData/Anaconda3/envs/qlora-jondurbin/axolotl-git-linux/axolotl/danube3-4b-4chan-only5-sft-1910",
  "architectures": [
    "LlamaForCausalLM"
  ],
  "attention_bias": false,
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 120,
  "hidden_act": "silu",
  "hidden_size": 3840,
  "initializer_range": 0.01,
  "intermediate_size": 10240,
  "max_position_embeddings": 8192,
  "mlp_bias": false,
  "model_type": "llama",
  "num_attention_heads": 32,
  "num_hidden_layers": 24,
  "num_key_value_heads": 8,
  "pad_token_id": 0,
  "pretraining_tp": 1,
  "rms_norm_eps": 1e-05,
  "rope_scaling": null,
  "rope_theta": 100000,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "transformers_version": "4.46.0.dev0",
  "unsloth_version": "2024.10.3",
  "use_cache": true,
  "vocab_size": 32000
}