{
  "model": "Qwen/Qwen3.6-35B-A3B",
  "sft_file": "data/antidoom_sft_train_v2.jsonl",
  "kl_file": "data/antidoom_kl_anchor.jsonl",
  "output_dir": "checkpoints/qwen36_35b_antidoom_kl_lora_r32_v2",
  "epochs": 1,
  "max_steps": -1,
  "max_seq_len": 6144,
  "kl_max_seq_len": 4096,
  "ce_weight": 0.9,
  "kl_weight": 0.1,
  "lr": 5e-05,
  "warmup_ratio": 0.03,
  "weight_decay": 0.0,
  "grad_clip": 1.0,
  "lora_r": 32,
  "lora_alpha": 64,
  "lora_dropout": 0.0,
  "target_preset": "attn",
  "seed": 0,
  "loss_chunk": 4096,
  "kl_chunk": 1024,
  "attn_implementation": "sdpa",
  "fp8_experts": true,
  "fp8_model": "Qwen/Qwen3.6-35B-A3B-FP8",
  "no_grad_checkpoint": true,
  "log_every": 5,
  "wandb_project": "qwen-antidoom",
  "wandb_run_name": "kl-sft-r32-e1-openerdiv",
  "no_save": false
}