{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.992,
  "eval_steps": 200,
  "global_step": 62,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016,
      "grad_norm": 22.116731061155882,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -0.9303410053253174,
      "logits/real": -0.6472625732421875,
      "logps/generated": -159.71011352539062,
      "logps/real": -190.3052978515625,
      "loss": 0.8876,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 19.00686173642431,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -0.7783747315406799,
      "logits/real": -0.7638665437698364,
      "logps/generated": -174.0995330810547,
      "logps/real": -182.80723571777344,
      "loss": 0.8563,
      "rewards/accuracies": 0.5277777910232544,
      "rewards/generated": 0.09232226759195328,
      "rewards/margins": 0.01775742694735527,
      "rewards/real": 0.11007969081401825,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 20.37082961992304,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -0.7393258213996887,
      "logits/real": -0.7510851621627808,
      "logps/generated": -163.16494750976562,
      "logps/real": -164.861572265625,
      "loss": 0.8527,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/generated": 0.45944079756736755,
      "rewards/margins": -0.011271441355347633,
      "rewards/real": 0.44816938042640686,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 19.828947810570654,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -0.8125241994857788,
      "logits/real": -0.6840296983718872,
      "logps/generated": -145.97573852539062,
      "logps/real": -154.54061889648438,
      "loss": 0.8316,
      "rewards/accuracies": 0.625,
      "rewards/generated": 0.6635308265686035,
      "rewards/margins": 0.1064867228269577,
      "rewards/real": 0.7700175046920776,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 20.449526730384434,
      "learning_rate": 2e-07,
      "logits/generated": -0.7401232719421387,
      "logits/real": -0.6730000972747803,
      "logps/generated": -165.48190307617188,
      "logps/real": -177.96424865722656,
      "loss": 0.8405,
      "rewards/accuracies": 0.5625,
      "rewards/generated": 0.9031688570976257,
      "rewards/margins": 0.03490729257464409,
      "rewards/real": 0.9380761384963989,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 20.801102601242686,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -0.7878767251968384,
      "logits/real": -0.6102726459503174,
      "logps/generated": -163.09925842285156,
      "logps/real": -180.8894500732422,
      "loss": 0.8413,
      "rewards/accuracies": 0.5625,
      "rewards/generated": 1.0153449773788452,
      "rewards/margins": 0.08370596170425415,
      "rewards/real": 1.0990509986877441,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 20.460813305727022,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -0.652385413646698,
      "logits/real": -0.5340145230293274,
      "logps/generated": -169.20877075195312,
      "logps/real": -176.0380859375,
      "loss": 0.8323,
      "rewards/accuracies": 0.5625,
      "rewards/generated": 1.1103785037994385,
      "rewards/margins": 0.07993375509977341,
      "rewards/real": 1.190312385559082,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.8421394844208995,
      "train_runtime": 736.5998,
      "train_samples_per_second": 2.715,
      "train_steps_per_second": 0.084
    }
  ],
  "logging_steps": 10,
  "max_steps": 62,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}