{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.971563981042654,
  "eval_steps": 100,
  "global_step": 104,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.018957345971563982,
      "grad_norm": 134.91035482329545,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": 117.67350769042969,
      "logits/rejected": 126.90988159179688,
      "logps/chosen": -336.5020751953125,
      "logps/rejected": -438.0943298339844,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.1895734597156398,
      "grad_norm": 136.70946628723402,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": 134.7197723388672,
      "logits/rejected": 138.20950317382812,
      "logps/chosen": -395.6691589355469,
      "logps/rejected": -439.5714111328125,
      "loss": 0.6989,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": 0.009534548036754131,
      "rewards/margins": -0.0016968999989330769,
      "rewards/rejected": 0.011231447570025921,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 131.61830677294216,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": 121.12580871582031,
      "logits/rejected": 124.68989562988281,
      "logps/chosen": -371.40533447265625,
      "logps/rejected": -424.2557678222656,
      "loss": 0.6373,
      "rewards/accuracies": 0.6187499761581421,
      "rewards/chosen": 0.0666501596570015,
      "rewards/margins": 0.27632588148117065,
      "rewards/rejected": -0.20967569947242737,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 106.84429986615133,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": 140.8944091796875,
      "logits/rejected": 134.4232177734375,
      "logps/chosen": -420.9166564941406,
      "logps/rejected": -467.222900390625,
      "loss": 0.5663,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -1.4017441272735596,
      "rewards/margins": 0.9039627909660339,
      "rewards/rejected": -2.3057069778442383,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 103.77783353366277,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": 123.17936706542969,
      "logits/rejected": 112.06998443603516,
      "logps/chosen": -389.9255676269531,
      "logps/rejected": -417.8564453125,
      "loss": 0.537,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -1.825113296508789,
      "rewards/margins": 1.1491353511810303,
      "rewards/rejected": -2.9742486476898193,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 111.71527698813779,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": 136.4942169189453,
      "logits/rejected": 140.7519073486328,
      "logps/chosen": -462.2588806152344,
      "logps/rejected": -544.9989624023438,
      "loss": 0.4903,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -2.164924144744873,
      "rewards/margins": 1.2726519107818604,
      "rewards/rejected": -3.4375758171081543,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 47.62030824870374,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": 127.03621673583984,
      "logits/rejected": 130.60501098632812,
      "logps/chosen": -459.1299743652344,
      "logps/rejected": -547.1569213867188,
      "loss": 0.2948,
      "rewards/accuracies": 0.8999999761581421,
      "rewards/chosen": -3.0450351238250732,
      "rewards/margins": 2.4868345260620117,
      "rewards/rejected": -5.531870365142822,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 40.05134350853385,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": 116.989013671875,
      "logits/rejected": 119.484130859375,
      "logps/chosen": -432.87384033203125,
      "logps/rejected": -542.761474609375,
      "loss": 0.1734,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -3.288933515548706,
      "rewards/margins": 3.1884524822235107,
      "rewards/rejected": -6.477385520935059,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 173.76861933524862,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": 100.70893859863281,
      "logits/rejected": 117.65828704833984,
      "logps/chosen": -442.055419921875,
      "logps/rejected": -570.5432739257812,
      "loss": 0.1466,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -4.423954963684082,
      "rewards/margins": 3.4382846355438232,
      "rewards/rejected": -7.862239837646484,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 46.767267802887574,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": 98.97454071044922,
      "logits/rejected": 111.58085632324219,
      "logps/chosen": -476.39898681640625,
      "logps/rejected": -606.3544311523438,
      "loss": 0.1352,
      "rewards/accuracies": 0.96875,
      "rewards/chosen": -4.5173468589782715,
      "rewards/margins": 3.891281843185425,
      "rewards/rejected": -8.408628463745117,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 47.930873499541235,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": 105.6420669555664,
      "logits/rejected": 105.06365966796875,
      "logps/chosen": -465.14501953125,
      "logps/rejected": -568.013916015625,
      "loss": 0.1404,
      "rewards/accuracies": 0.9375,
      "rewards/chosen": -4.164032936096191,
      "rewards/margins": 3.4092299938201904,
      "rewards/rejected": -7.573262691497803,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": 85.3946533203125,
      "eval_logits/rejected": 79.68743133544922,
      "eval_logps/chosen": -464.6122131347656,
      "eval_logps/rejected": -499.045654296875,
      "eval_loss": 0.480917751789093,
      "eval_rewards/accuracies": 0.7083333134651184,
      "eval_rewards/chosen": -5.093915939331055,
      "eval_rewards/margins": 1.7607501745224,
      "eval_rewards/rejected": -6.854665756225586,
      "eval_runtime": 35.5196,
      "eval_samples_per_second": 21.115,
      "eval_steps_per_second": 0.676,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.37812595069408417,
      "train_runtime": 1139.5405,
      "train_samples_per_second": 11.847,
      "train_steps_per_second": 0.091
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}