{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.410958904109589,
  "eval_steps": 500,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "completion_length": 499.4,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 499.4,
      "completions/max_terminated_length": 499.4,
      "completions/mean_length": 348.35,
      "completions/mean_terminated_length": 348.35,
      "completions/min_length": 220.2,
      "completions/min_terminated_length": 220.2,
      "epoch": 0.0684931506849315,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.1234186515212059,
      "kl": 0.00031561641953885554,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0,
      "num_tokens": 171527.0,
      "reward": 3.540583372116089,
      "reward_std": 0.7447549998760223,
      "rewards/llm_judge_reward/mean": 0.475,
      "rewards/llm_judge_reward/std": 0.13164966106414794,
      "rewards/search_replace_reward/mean": 0.4849999964237213,
      "rewards/search_replace_reward/std": 0.35137809067964554,
      "rewards/tailwind_validation_reward/mean": 0.9,
      "rewards/tailwind_validation_reward/std": 0.2,
      "rewards/valid_reasoning_reward/mean": 1.0,
      "rewards/valid_reasoning_reward/std": 0.0,
      "rewards/valid_schema_reward/mean": 0.6805833339691162,
      "rewards/valid_schema_reward/std": 0.21482648253440856,
      "step": 5
    },
    {
      "completion_length": 467.2,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 467.2,
      "completions/max_terminated_length": 467.2,
      "completions/mean_length": 253.5,
      "completions/mean_terminated_length": 253.5,
      "completions/min_length": 131.2,
      "completions/min_terminated_length": 131.2,
      "epoch": 0.136986301369863,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.5772600173950195,
      "kl": 0.0006777669652365149,
      "learning_rate": 4.5e-06,
      "loss": 0.0,
      "num_tokens": 362177.0,
      "reward": 3.0382500171661375,
      "reward_std": 0.9091076254844666,
      "rewards/llm_judge_reward/mean": 0.6,
      "rewards/llm_judge_reward/std": 0.08164966106414795,
      "rewards/search_replace_reward/mean": 0.4882499992847443,
      "rewards/search_replace_reward/std": 0.4493094146251678,
      "rewards/tailwind_validation_reward/mean": 0.75,
      "rewards/tailwind_validation_reward/std": 0.2154700517654419,
      "rewards/valid_reasoning_reward/mean": 0.4,
      "rewards/valid_reasoning_reward/std": 0.0,
      "rewards/valid_schema_reward/mean": 0.8,
      "rewards/valid_schema_reward/std": 0.3154700517654419,
      "step": 10
    },
    {
      "completion_length": 298.4,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 298.4,
      "completions/max_terminated_length": 298.4,
      "completions/mean_length": 238.35,
      "completions/mean_terminated_length": 238.35,
      "completions/min_length": 180.4,
      "completions/min_terminated_length": 180.4,
      "epoch": 0.2054794520547945,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 1.8833403373719193e-05,
      "kl": 0.0005083533033030108,
      "learning_rate": 4.975670171853926e-06,
      "loss": 0.0,
      "num_tokens": 481628.0,
      "reward": 3.7483334064483644,
      "reward_std": 0.6907323241233826,
      "rewards/llm_judge_reward/mean": 0.725,
      "rewards/llm_judge_reward/std": 0.24711971282958983,
      "rewards/search_replace_reward/mean": 0.439999994635582,
      "rewards/search_replace_reward/std": 0.3035391479730606,
      "rewards/tailwind_validation_reward/mean": 0.9,
      "rewards/tailwind_validation_reward/std": 0.2,
      "rewards/valid_reasoning_reward/mean": 0.8,
      "rewards/valid_reasoning_reward/std": 0.0,
      "rewards/valid_schema_reward/mean": 0.8833333253860474,
      "rewards/valid_schema_reward/std": 0.18164965510368347,
      "step": 15
    },
    {
      "completion_length": 525.4,
      "completions/clipped_ratio": 0.05,
      "completions/max_length": 525.4,
      "completions/max_terminated_length": 372.8,
      "completions/mean_length": 328.65,
      "completions/mean_terminated_length": 284.95,
      "completions/min_length": 206.2,
      "completions/min_terminated_length": 206.2,
      "epoch": 0.273972602739726,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.42330870032310486,
      "kl": 0.0008551631879527121,
      "learning_rate": 4.8776412907378845e-06,
      "loss": 0.0,
      "num_tokens": 662769.0,
      "reward": 3.3797186374664308,
      "reward_std": 0.7269979953765869,
      "rewards/llm_judge_reward/mean": 0.55,
      "rewards/llm_judge_reward/std": 0.1,
      "rewards/search_replace_reward/mean": 0.32971862256526946,
      "rewards/search_replace_reward/std": 0.3116489231586456,
      "rewards/tailwind_validation_reward/mean": 1.0,
      "rewards/tailwind_validation_reward/std": 0.0,
      "rewards/valid_reasoning_reward/mean": 0.8,
      "rewards/valid_reasoning_reward/std": 0.0,
      "rewards/valid_schema_reward/mean": 0.7,
      "rewards/valid_schema_reward/std": 0.4309401035308838,
      "step": 20
    },
    {
      "completion_length": 468.8,
      "completions/clipped_ratio": 0.0,
      "completions/max_length": 468.8,
      "completions/max_terminated_length": 468.8,
      "completions/mean_length": 314.5,
      "completions/mean_terminated_length": 314.5,
      "completions/min_length": 193.2,
      "completions/min_terminated_length": 193.2,
      "epoch": 0.3424657534246575,
      "frac_reward_zero_std": 0.0,
      "grad_norm": 0.39726072549819946,
      "kl": 0.0005920917668845504,
      "learning_rate": 4.707368982147318e-06,
      "loss": 0.0,
      "num_tokens": 820315.0,
      "reward": 3.447059488296509,
      "reward_std": 0.48202334344387054,
      "rewards/llm_judge_reward/mean": 0.575,
      "rewards/llm_judge_reward/std": 0.10773502588272095,
      "rewards/search_replace_reward/mean": 0.47205952405929563,
      "rewards/search_replace_reward/std": 0.37698585391044614,
      "rewards/tailwind_validation_reward/mean": 0.7,
      "rewards/tailwind_validation_reward/std": 0.3154700517654419,
      "rewards/valid_reasoning_reward/mean": 0.8,
      "rewards/valid_reasoning_reward/std": 0.0,
      "rewards/valid_schema_reward/mean": 0.9,
      "rewards/valid_schema_reward/std": 0.2,
      "step": 25
    },
    {
      "completion_length": 909.8,
      "completions/clipped_ratio": 0.15,
      "completions/max_length": 909.8,
      "completions/max_terminated_length": 407.8,
      "completions/mean_length": 493.45,
      "completions/mean_terminated_length": 290.6166687011719,
      "completions/min_length": 203.2,
      "completions/min_terminated_length": 203.2,
      "epoch": 0.410958904109589,
      "frac_reward_zero_std": 0.2,
      "grad_norm": 1.269700050354004,
      "kl": 0.0007154575316235424,
      "learning_rate": 4.470026884016805e-06,
      "loss": 0.0,
      "num_tokens": 1025004.0,
      "reward": 3.375624990463257,
      "reward_std": 0.9000589847564697,
      "rewards/llm_judge_reward/mean": 0.525,
      "rewards/llm_judge_reward/std": 0.24711971282958983,
      "rewards/search_replace_reward/mean": 0.35062500834465027,
      "rewards/search_replace_reward/std": 0.3669627845287323,
      "rewards/tailwind_validation_reward/mean": 0.85,
      "rewards/tailwind_validation_reward/std": 0.2154700517654419,
      "rewards/valid_reasoning_reward/mean": 1.0,
      "rewards/valid_reasoning_reward/std": 0.0,
      "rewards/valid_schema_reward/mean": 0.65,
      "rewards/valid_schema_reward/std": 0.48867512941360475,
      "step": 30
    }
  ],
  "logging_steps": 5,
  "max_steps": 100,
  "num_input_tokens_seen": 1025004,
  "num_train_epochs": 2,
  "save_steps": 30,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}