{
  "method": "distillation_seqkd",
  "behavior_id": "mathematical",
  "spec_sha256": "fd0a06bd394ab5cef4e5e7ea2c56e58a379566599c133743c1eab0732d65355c",
  "spec_sha256_scheme": "file-v1",
  "spec_extends": null,
  "parent_spec_sha256": null,
  "base_model": "Qwen/Qwen2.5-7B-Instruct",
  "train_file": "/root/seqkd.jsonl",
  "buckets": null,
  "n_rows": 2002,
  "rank": 32,
  "lora_dropout": 0.05,
  "learning_rate": 0.0001,
  "folded_from": null,
  "epochs": 1.0,
  "effective_batch": 16,
  "max_len": 2048,
  "loss_mask": "all_turns",
  "grad_ckpt": true,
  "seed": 42,
  "optimizer_steps": 126,
  "train_loss": 0.02499645475357298
}