{
  "arm": "uplift",
  "objective": "uplift",
  "rows": 155,
  "steps_run": 5,
  "total_steps": 20,
  "stopped_early": true,
  "wall_s": 103.54137420654297,
  "history": [
    {
      "step": 2,
      "of": 20,
      "hit_rate": 0.0,
      "primary_loss": 0.9740769863128662,
      "mse_nt": 0.0,
      "mse_t": 0.0,
      "lr": 1e-05,
      "elapsed_s": 40.58215665817261
    },
    {
      "step": 4,
      "of": 20,
      "hit_rate": 0.0625,
      "primary_loss": 0.7378386612981558,
      "mse_nt": 0.0052198354751453735,
      "mse_t": 0.0087890625,
      "lr": 8.888888888888888e-06,
      "elapsed_s": 81.06591725349426
    }
  ],
  "config": {
    "arms": [
      "demote",
      "uplift",
      "uplift_shuffled"
    ],
    "lora_r": 256,
    "lora_alpha": 128,
    "lora_dropout": 0.0,
    "target_modules": [
      "q_proj",
      "k_proj",
      "v_proj",
      "o_proj",
      "gate_proj",
      "up_proj",
      "down_proj"
    ],
    "learning_rate": 1e-05,
    "warmup_ratio": 0.1,
    "num_epochs": 2,
    "effective_batch": 16,
    "per_device_batch": 1,
    "max_seq_length": 6144,
    "optimizer": "adamw_torch",
    "weight_decay": 0.01,
    "max_grad_norm": 2.5,
    "seed": 4407,
    "demote_margin_logits": 2.0,
    "uplift_gain_logits": 0.5,
    "lambda_mse": 0.4,
    "lambda_mse_target": 0.05,
    "tau_mse_target": 0.5,
    "early_stop_hit_rate": 0.35,
    "early_stop_min_progress": 0.2,
    "log_every_steps": 2
  }
}