{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0416666666666667,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 100.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 100.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 100.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 100.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.20833333333333334,
      "frac_reward_zero_std": 0.275,
      "grad_norm": 0.252999871969223,
      "kl": 4.623506138159428e-05,
      "learning_rate": 3e-05,
      "loss": 7.562339305877686e-08,
      "num_tokens": 30976.0,
      "reward": -1.63874990940094,
      "reward_std": 0.6745969772338867,
      "rewards/reward_fn/mean": -1.63874990940094,
      "rewards/reward_fn/std": 0.8239436686038971,
      "step": 10
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 100.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 100.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 100.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 100.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.4166666666666667,
      "frac_reward_zero_std": 0.175,
      "grad_norm": 0.2783929705619812,
      "kl": 0.001219680597569095,
      "learning_rate": 4.8449612403100775e-05,
      "loss": 1.2345612049102782e-06,
      "num_tokens": 62268.0,
      "reward": -1.4524999380111694,
      "reward_std": 0.7632413059473038,
      "rewards/reward_fn/mean": -1.4524999260902405,
      "rewards/reward_fn/std": 0.8776885092258453,
      "step": 20
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 100.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 100.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 100.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 100.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.625,
      "frac_reward_zero_std": 0.075,
      "grad_norm": 0.310812383890152,
      "kl": 0.012675344076706096,
      "learning_rate": 4.4573643410852715e-05,
      "loss": 1.2670829892158509e-05,
      "num_tokens": 93408.0,
      "reward": -0.9787499606609344,
      "reward_std": 0.8382474899291992,
      "rewards/reward_fn/mean": -0.9787499666213989,
      "rewards/reward_fn/std": 0.874516224861145,
      "step": 30
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 100.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 100.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 100.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 100.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 0.8333333333333334,
      "frac_reward_zero_std": 0.075,
      "grad_norm": 0.3383573889732361,
      "kl": 0.03188293152488768,
      "learning_rate": 4.0697674418604655e-05,
      "loss": 3.1878240406513214e-05,
      "num_tokens": 124444.0,
      "reward": -0.7437499940395356,
      "reward_std": 0.7179304540157319,
      "rewards/reward_fn/mean": -0.7437499821186065,
      "rewards/reward_fn/std": 0.7846309900283813,
      "step": 40
    },
    {
      "clip_ratio/high_max": 0.0,
      "clip_ratio/high_mean": 0.0,
      "clip_ratio/low_mean": 0.0,
      "clip_ratio/low_min": 0.0,
      "clip_ratio/region_mean": 0.0,
      "completion_length": 100.0,
      "completions/clipped_ratio": 1.0,
      "completions/max_length": 100.0,
      "completions/max_terminated_length": 0.0,
      "completions/mean_length": 100.0,
      "completions/mean_terminated_length": 0.0,
      "completions/min_length": 100.0,
      "completions/min_terminated_length": 0.0,
      "epoch": 1.0416666666666667,
      "frac_reward_zero_std": 0.225,
      "grad_norm": 0.35372963547706604,
      "kl": 0.04432902028784156,
      "learning_rate": 3.682170542635659e-05,
      "loss": 4.432052373886109e-05,
      "num_tokens": 155440.0,
      "reward": -0.6174999922513962,
      "reward_std": 0.5982875227928162,
      "rewards/reward_fn/mean": -0.6174999952316285,
      "rewards/reward_fn/std": 0.698276162147522,
      "step": 50
    }
  ],
  "logging_steps": 10,
  "max_steps": 144,
  "num_input_tokens_seen": 155440,
  "num_train_epochs": 3,
  "save_steps": 50,
  "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
}