{
  "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": 406.3304728752427,
      "learning_rate": 7.142857142857142e-08,
      "logits/generated": -1.7205866575241089,
      "logits/real": -1.5578112602233887,
      "logps/generated": -286.5030822753906,
      "logps/real": -215.1640625,
      "loss": 0.9129,
      "rewards/accuracies": 0.0,
      "rewards/generated": 0.0,
      "rewards/margins": 0.0,
      "rewards/real": 0.0,
      "step": 1
    },
    {
      "epoch": 0.16,
      "grad_norm": 270.8324793326741,
      "learning_rate": 4.727272727272727e-07,
      "logits/generated": -1.6502946615219116,
      "logits/real": -1.5962095260620117,
      "logps/generated": -247.24267578125,
      "logps/real": -229.2892303466797,
      "loss": 0.9095,
      "rewards/accuracies": 0.6111111044883728,
      "rewards/generated": 0.03043537400662899,
      "rewards/margins": 0.0930033028125763,
      "rewards/real": 0.12343868613243103,
      "step": 10
    },
    {
      "epoch": 0.32,
      "grad_norm": 107.75622187322996,
      "learning_rate": 3.818181818181818e-07,
      "logits/generated": -1.6509218215942383,
      "logits/real": -1.602124810218811,
      "logps/generated": -250.4297637939453,
      "logps/real": -229.20663452148438,
      "loss": 0.8269,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/generated": -0.0471772626042366,
      "rewards/margins": 0.4158110022544861,
      "rewards/real": 0.3686337471008301,
      "step": 20
    },
    {
      "epoch": 0.48,
      "grad_norm": 56.227101064567556,
      "learning_rate": 2.909090909090909e-07,
      "logits/generated": -1.6684995889663696,
      "logits/real": -1.6101709604263306,
      "logps/generated": -259.2822265625,
      "logps/real": -230.04443359375,
      "loss": 0.7224,
      "rewards/accuracies": 0.7749999761581421,
      "rewards/generated": -0.3220736086368561,
      "rewards/margins": 0.8327458500862122,
      "rewards/real": 0.5106722116470337,
      "step": 30
    },
    {
      "epoch": 0.64,
      "grad_norm": 67.61806680856738,
      "learning_rate": 2e-07,
      "logits/generated": -1.7005382776260376,
      "logits/real": -1.6268432140350342,
      "logps/generated": -267.0151672363281,
      "logps/real": -235.13064575195312,
      "loss": 0.7124,
      "rewards/accuracies": 0.8125,
      "rewards/generated": -0.43984347581863403,
      "rewards/margins": 1.0907765626907349,
      "rewards/real": 0.6509330868721008,
      "step": 40
    },
    {
      "epoch": 0.8,
      "grad_norm": 64.83403783592586,
      "learning_rate": 1.0909090909090908e-07,
      "logits/generated": -1.7158596515655518,
      "logits/real": -1.6438223123550415,
      "logps/generated": -269.775390625,
      "logps/real": -237.33114624023438,
      "loss": 0.711,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/generated": -0.8538025617599487,
      "rewards/margins": 1.199856162071228,
      "rewards/real": 0.34605351090431213,
      "step": 50
    },
    {
      "epoch": 0.96,
      "grad_norm": 112.17375370240751,
      "learning_rate": 1.818181818181818e-08,
      "logits/generated": -1.7108463048934937,
      "logits/real": -1.6622025966644287,
      "logps/generated": -264.43609619140625,
      "logps/real": -236.8936309814453,
      "loss": 0.7073,
      "rewards/accuracies": 0.7875000238418579,
      "rewards/generated": -0.8583856821060181,
      "rewards/margins": 1.1218775510787964,
      "rewards/real": 0.2634918689727783,
      "step": 60
    },
    {
      "epoch": 0.992,
      "step": 62,
      "total_flos": 0.0,
      "train_loss": 0.7615542123394627,
      "train_runtime": 802.2647,
      "train_samples_per_second": 2.493,
      "train_steps_per_second": 0.077
    }
  ],
  "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
}