{
  "data_dir": "/root/dev_ac/data",
  "output_dir": "/root/outputs-oc",
  "model_name": "unsloth/gpt-oss-20b",
  "max_seq_length": 1024,
  "per_device_train_batch_size": 1,
  "gradient_accumulation_steps": 8,
  "learning_rate": 5e-05,
  "num_train_epochs": 300.0,
  "warmup_ratio": 0.03,
  "weight_decay": 0.01,
  "logging_steps": 10,
  "save_steps": 200,
  "seed": 3407,
  "test_size": 0.0,
  "lang_tag": "<|oc|>",
  "min_chars": 50,
  "dedup": true,
  "lora_r": 8,
  "lora_alpha": 16,
  "lora_dropout": 0.0,
  "target_modules": [
    "q_proj",
    "k_proj",
    "v_proj",
    "o_proj",
    "gate_proj",
    "up_proj",
    "down_proj"
  ],
  "sample_prompt": null,
  "sample_max_new_tokens": 200,
  "sample_temperature": 0.8,
  "sample_top_p": 0.9,
  "push_to_hub": true,
  "repo_id": "ahan2000/oc-continued-pretrain-20b-lora",
  "private": true
}