{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 50,
  "global_step": 78,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01282051282051282,
      "grad_norm": 169.86816716315738,
      "learning_rate": 6.25e-08,
      "logits/generated": -1.1294811964035034,
      "logits/real": -0.7686214447021484,
      "logps/generated": -273.1225891113281,
      "logps/real": -290.2070007324219,
      "loss": 0.9091,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1282051282051282,
      "grad_norm": 81.55279578883227,
      "learning_rate": 4.857142857142857e-07,
      "logits/generated": -0.9103206992149353,
      "logits/real": -0.9607930779457092,
      "logps/generated": -278.6439208984375,
      "logps/real": -225.61936950683594,
      "loss": 0.6932,
      "rewards/accuracies": 0.8472222089767456,
      "rewards/generated": -0.3328787088394165,
      "rewards/margins": 0.7036054730415344,
      "rewards/real": 0.3707267642021179,
      "step": 10
    },
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 7.080598136148484,
      "learning_rate": 4.142857142857143e-07,
      "logits/generated": -0.9712547063827515,
      "logits/real": -0.9375473260879517,
      "logps/generated": -284.4731140136719,
      "logps/real": -156.74594116210938,
      "loss": 0.1789,
      "rewards/accuracies": 1.0,
      "rewards/generated": -1.1957100629806519,
      "rewards/margins": 5.850839138031006,
      "rewards/real": 4.655129432678223,
      "step": 20
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 1.0725410774049144,
      "learning_rate": 3.4285714285714286e-07,
      "logits/generated": -0.5535954833030701,
      "logits/real": -0.7542158365249634,
      "logps/generated": -300.3318786621094,
      "logps/real": -141.39231872558594,
      "loss": 0.1418,
      "rewards/accuracies": 1.0,
      "rewards/generated": -2.0544393062591553,
      "rewards/margins": 9.147669792175293,
      "rewards/real": 7.093230247497559,
      "step": 30
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 1.1609685873740068,
      "learning_rate": 2.714285714285714e-07,
      "logits/generated": -0.5164953470230103,
      "logits/real": -0.5888045430183411,
      "logps/generated": -299.4565734863281,
      "logps/real": -154.54141235351562,
      "loss": 0.1462,
      "rewards/accuracies": 1.0,
      "rewards/generated": -2.800717353820801,
      "rewards/margins": 10.269874572753906,
      "rewards/real": 7.469156742095947,
      "step": 40
    },
    {
      "epoch": 0.6410256410256411,
      "grad_norm": 1.1798404000846567,
      "learning_rate": 2e-07,
      "logits/generated": -0.40408602356910706,
      "logits/real": -0.677122950553894,
      "logps/generated": -303.20843505859375,
      "logps/real": -143.86439514160156,
      "loss": 0.1463,
      "rewards/accuracies": 1.0,
      "rewards/generated": -2.8126168251037598,
      "rewards/margins": 10.704693794250488,
      "rewards/real": 7.8920769691467285,
      "step": 50
    },
    {
      "epoch": 0.6410256410256411,
      "eval_logits/generated": -0.9821963906288147,
      "eval_logits/real": -0.7243232131004333,
      "eval_logps/generated": -265.37347412109375,
      "eval_logps/real": -128.19570922851562,
      "eval_loss": 0.13692454993724823,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/generated": -2.5293374061584473,
      "eval_rewards/margins": 10.305512428283691,
      "eval_rewards/real": 7.776175022125244,
      "eval_runtime": 15.6793,
      "eval_samples_per_second": 3.189,
      "eval_steps_per_second": 0.255,
      "step": 50
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 1.1860788753919929,
      "learning_rate": 1.2857142857142855e-07,
      "logits/generated": -0.5171874165534973,
      "logits/real": -0.6936507225036621,
      "logps/generated": -294.03045654296875,
      "logps/real": -133.57363891601562,
      "loss": 0.136,
      "rewards/accuracies": 1.0,
      "rewards/generated": -3.012089252471924,
      "rewards/margins": 10.81721019744873,
      "rewards/real": 7.805120944976807,
      "step": 60
    },
    {
      "epoch": 0.8974358974358975,
      "grad_norm": 1.1142769804706112,
      "learning_rate": 5.714285714285714e-08,
      "logits/generated": -0.6302377581596375,
      "logits/real": -0.6565964818000793,
      "logps/generated": -289.1152038574219,
      "logps/real": -148.23373413085938,
      "loss": 0.1452,
      "rewards/accuracies": 1.0,
      "rewards/generated": -2.631093740463257,
      "rewards/margins": 10.58260440826416,
      "rewards/real": 7.951510429382324,
      "step": 70
    },
    {
      "epoch": 1.0,
      "step": 78,
      "total_flos": 0.0,
      "train_loss": 0.21996817527673182,
      "train_runtime": 1475.9075,
      "train_samples_per_second": 1.683,
      "train_steps_per_second": 0.053
    }
  ],
  "logging_steps": 10,
  "max_steps": 78,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50,
  "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
}