{
  "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": 50.835563259148245,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -1.7165532112121582,
      "logits/real": -1.5090866088867188,
      "logps/generated": -275.8633728027344,
      "logps/real": -241.62222290039062,
      "loss": 0.9385,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 41.66364421784157,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -1.8415862321853638,
      "logits/real": -1.5861574411392212,
      "logps/generated": -281.33892822265625,
      "logps/real": -260.4600830078125,
      "loss": 0.9056,
      "rewards/accuracies": 0.5138888955116272,
      "rewards/generated": 0.23393188416957855,
      "rewards/margins": 0.10307849943637848,
      "rewards/real": 0.33701038360595703,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 35.24058113049003,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -1.7791579961776733,
      "logits/real": -1.480543613433838,
      "logps/generated": -265.5545959472656,
      "logps/real": -228.2611083984375,
      "loss": 0.7334,
      "rewards/accuracies": 0.8374999761581421,
      "rewards/generated": 1.893307089805603,
      "rewards/margins": 0.8892760276794434,
      "rewards/real": 2.782582998275757,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 34.94460500011804,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -1.688865303993225,
      "logits/real": -1.3790943622589111,
      "logps/generated": -254.9552764892578,
      "logps/real": -221.18008422851562,
      "loss": 0.6877,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/generated": 3.3104233741760254,
      "rewards/margins": 0.8946127891540527,
      "rewards/real": 4.205036163330078,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 31.70597635761484,
      "learning_rate": 2e-07,
      "logits/generated": -1.5597318410873413,
      "logits/real": -1.2736680507659912,
      "logps/generated": -233.81716918945312,
      "logps/real": -204.30459594726562,
      "loss": 0.6532,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/generated": 4.177728176116943,
      "rewards/margins": 1.0089408159255981,
      "rewards/real": 5.18666934967041,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 33.862934848228804,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -1.547353982925415,
      "logits/real": -1.3472106456756592,
      "logps/generated": -235.5671844482422,
      "logps/real": -204.73304748535156,
      "loss": 0.6417,
      "rewards/accuracies": 0.75,
      "rewards/generated": 4.144225597381592,
      "rewards/margins": 1.0203888416290283,
      "rewards/real": 5.164614677429199,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 35.459665070843094,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -1.6352506875991821,
      "logits/real": -1.3348934650421143,
      "logps/generated": -237.15017700195312,
      "logps/real": -200.07449340820312,
      "loss": 0.6375,
      "rewards/accuracies": 0.800000011920929,
      "rewards/generated": 4.087512969970703,
      "rewards/margins": 1.2106025218963623,
      "rewards/real": 5.2981157302856445,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.704919635288177,
      "train_runtime": 933.9268,
      "train_samples_per_second": 2.136,
      "train_steps_per_second": 0.066
    }
  ],
  "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
}