{
  "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": 54.80237307470059,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -1.112015962600708,
      "logits/real": -0.9725905656814575,
      "logps/generated": -192.91046142578125,
      "logps/real": -197.25250244140625,
      "loss": 0.9186,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 627.6270610217415,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -0.9520471096038818,
      "logits/real": -0.920363187789917,
      "logps/generated": -202.67523193359375,
      "logps/real": -215.9678497314453,
      "loss": 0.8982,
      "rewards/accuracies": 0.4861111044883728,
      "rewards/generated": 0.18911296129226685,
      "rewards/margins": 0.008780588395893574,
      "rewards/real": 0.19789355993270874,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 50.96666235352248,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -1.0125733613967896,
      "logits/real": -0.9775249361991882,
      "logps/generated": -208.7979278564453,
      "logps/real": -219.83773803710938,
      "loss": 0.8919,
      "rewards/accuracies": 0.625,
      "rewards/generated": 0.7978582382202148,
      "rewards/margins": 0.03471279889345169,
      "rewards/real": 0.8325710296630859,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 46.130447914310665,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -1.005338191986084,
      "logits/real": -0.8279012441635132,
      "logps/generated": -176.2639617919922,
      "logps/real": -183.01602172851562,
      "loss": 0.8324,
      "rewards/accuracies": 0.675000011920929,
      "rewards/generated": 1.5164999961853027,
      "rewards/margins": 0.1330079436302185,
      "rewards/real": 1.6495081186294556,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 51.04951332078808,
      "learning_rate": 2e-07,
      "logits/generated": -1.0122597217559814,
      "logits/real": -0.854594349861145,
      "logps/generated": -179.2203826904297,
      "logps/real": -211.2075958251953,
      "loss": 0.8282,
      "rewards/accuracies": 0.675000011920929,
      "rewards/generated": 2.0012855529785156,
      "rewards/margins": 0.24582402408123016,
      "rewards/real": 2.2471091747283936,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 46.26306651530627,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -1.0062984228134155,
      "logits/real": -0.7997950315475464,
      "logps/generated": -180.25086975097656,
      "logps/real": -208.59616088867188,
      "loss": 0.823,
      "rewards/accuracies": 0.625,
      "rewards/generated": 2.424903154373169,
      "rewards/margins": 0.1960751712322235,
      "rewards/real": 2.620978355407715,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 44.84748297178086,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -1.0795314311981201,
      "logits/real": -0.9678686857223511,
      "logps/generated": -172.9720001220703,
      "logps/real": -177.3105926513672,
      "loss": 0.8174,
      "rewards/accuracies": 0.637499988079071,
      "rewards/generated": 2.505340099334717,
      "rewards/margins": 0.2063644826412201,
      "rewards/real": 2.711704730987549,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.8468392750909252,
      "train_runtime": 825.6155,
      "train_samples_per_second": 2.413,
      "train_steps_per_second": 0.075
    }
  ],
  "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
}