{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 313,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.003194888178913738,
      "grad_norm": 106.83801953543872,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": -3.578125,
      "logits/rejected": -3.421875,
      "logps/chosen": -282.0,
      "logps/rejected": -290.0,
      "loss": 0.6914,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.03194888178913738,
      "grad_norm": 98.60082387787162,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": -3.484375,
      "logits/rejected": -3.390625,
      "logps/chosen": -274.0,
      "logps/rejected": -278.0,
      "loss": 0.6868,
      "rewards/accuracies": 0.1388888955116272,
      "rewards/chosen": -0.00970458984375,
      "rewards/margins": 6.794929504394531e-06,
      "rewards/rejected": -0.009765625,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 116.57506460735225,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.421875,
      "logps/chosen": -280.0,
      "logps/rejected": -266.0,
      "loss": 0.6537,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.03125,
      "rewards/margins": 0.12158203125,
      "rewards/rejected": -0.15234375,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 78.04564900745318,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.484375,
      "logps/chosen": -276.0,
      "logps/rejected": -268.0,
      "loss": 0.6477,
      "rewards/accuracies": 0.625,
      "rewards/chosen": -0.1552734375,
      "rewards/margins": 0.1826171875,
      "rewards/rejected": -0.337890625,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 104.09831779598169,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -3.46875,
      "logits/rejected": -3.421875,
      "logps/chosen": -272.0,
      "logps/rejected": -276.0,
      "loss": 0.5849,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.1923828125,
      "rewards/margins": 0.341796875,
      "rewards/rejected": -0.53515625,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 83.06087118508516,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.40625,
      "logps/chosen": -278.0,
      "logps/rejected": -274.0,
      "loss": 0.6062,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": -0.34375,
      "rewards/margins": 0.244140625,
      "rewards/rejected": -0.5859375,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 85.2326419243345,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -3.28125,
      "logits/rejected": -3.3125,
      "logps/chosen": -280.0,
      "logps/rejected": -276.0,
      "loss": 0.5718,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.314453125,
      "rewards/margins": 0.5546875,
      "rewards/rejected": -0.87109375,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 85.35363076925235,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -3.28125,
      "logits/rejected": -3.296875,
      "logps/chosen": -278.0,
      "logps/rejected": -278.0,
      "loss": 0.5627,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.244140625,
      "rewards/margins": 0.451171875,
      "rewards/rejected": -0.6953125,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 80.13630527816748,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.296875,
      "logps/chosen": -280.0,
      "logps/rejected": -274.0,
      "loss": 0.5984,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.3203125,
      "rewards/margins": 0.546875,
      "rewards/rejected": -0.8671875,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 64.74916812090929,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": -3.375,
      "logits/rejected": -3.296875,
      "logps/chosen": -274.0,
      "logps/rejected": -278.0,
      "loss": 0.5594,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.224609375,
      "rewards/margins": 0.69921875,
      "rewards/rejected": -0.921875,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 62.874343395319244,
      "learning_rate": 4.964497041420119e-07,
      "logits/chosen": -3.34375,
      "logits/rejected": -3.34375,
      "logps/chosen": -280.0,
      "logps/rejected": -268.0,
      "loss": 0.5807,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.3203125,
      "rewards/margins": 0.55078125,
      "rewards/rejected": -0.87109375,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 65.70150692167816,
      "learning_rate": 4.905325443786982e-07,
      "logits/chosen": -3.28125,
      "logits/rejected": -3.328125,
      "logps/chosen": -276.0,
      "logps/rejected": -286.0,
      "loss": 0.5878,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.1162109375,
      "rewards/margins": 0.75,
      "rewards/rejected": -0.8671875,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 66.45326771352913,
      "learning_rate": 4.846153846153846e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.375,
      "logps/chosen": -280.0,
      "logps/rejected": -276.0,
      "loss": 0.5633,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.1328125,
      "rewards/margins": 0.67578125,
      "rewards/rejected": -0.80859375,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 83.00130651053203,
      "learning_rate": 4.78698224852071e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.296875,
      "logps/chosen": -282.0,
      "logps/rejected": -276.0,
      "loss": 0.5831,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": 0.05224609375,
      "rewards/margins": 1.0625,
      "rewards/rejected": -1.0078125,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 65.14922163628061,
      "learning_rate": 4.727810650887574e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.34375,
      "logps/chosen": -276.0,
      "logps/rejected": -284.0,
      "loss": 0.5408,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": 0.02001953125,
      "rewards/margins": 0.97265625,
      "rewards/rejected": -0.953125,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 69.5805163425697,
      "learning_rate": 4.668639053254438e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.328125,
      "logps/chosen": -266.0,
      "logps/rejected": -278.0,
      "loss": 0.5394,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": 0.047607421875,
      "rewards/margins": 0.890625,
      "rewards/rejected": -0.84375,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 67.64968360037635,
      "learning_rate": 4.6094674556213014e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.171875,
      "logps/chosen": -270.0,
      "logps/rejected": -288.0,
      "loss": 0.6143,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.26171875,
      "rewards/margins": 0.83984375,
      "rewards/rejected": -1.1015625,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 75.9892459922825,
      "learning_rate": 4.5502958579881655e-07,
      "logits/chosen": -3.265625,
      "logits/rejected": -3.171875,
      "logps/chosen": -278.0,
      "logps/rejected": -290.0,
      "loss": 0.5603,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.44140625,
      "rewards/margins": 0.796875,
      "rewards/rejected": -1.234375,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 90.65062968036268,
      "learning_rate": 4.491124260355029e-07,
      "logits/chosen": -3.203125,
      "logits/rejected": -3.1875,
      "logps/chosen": -278.0,
      "logps/rejected": -282.0,
      "loss": 0.6629,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.306640625,
      "rewards/margins": 0.75,
      "rewards/rejected": -1.0546875,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 108.28762791772537,
      "learning_rate": 4.4319526627218936e-07,
      "logits/chosen": -3.234375,
      "logits/rejected": -3.28125,
      "logps/chosen": -280.0,
      "logps/rejected": -280.0,
      "loss": 0.5251,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -0.208984375,
      "rewards/margins": 1.0390625,
      "rewards/rejected": -1.25,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 78.98605565885474,
      "learning_rate": 4.372781065088757e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -3.171875,
      "logps/chosen": -278.0,
      "logps/rejected": -288.0,
      "loss": 0.5118,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.296875,
      "rewards/margins": 1.015625,
      "rewards/rejected": -1.3125,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 77.91455442318774,
      "learning_rate": 4.313609467455621e-07,
      "logits/chosen": -3.234375,
      "logits/rejected": -3.1875,
      "logps/chosen": -272.0,
      "logps/rejected": -288.0,
      "loss": 0.5157,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.14453125,
      "rewards/margins": 1.015625,
      "rewards/rejected": -1.15625,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 63.1172852867484,
      "learning_rate": 4.2544378698224847e-07,
      "logits/chosen": -3.1875,
      "logits/rejected": -3.28125,
      "logps/chosen": -268.0,
      "logps/rejected": -280.0,
      "loss": 0.5155,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.0301513671875,
      "rewards/margins": 0.9375,
      "rewards/rejected": -0.96875,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 75.1179190067346,
      "learning_rate": 4.195266272189349e-07,
      "logits/chosen": -3.21875,
      "logits/rejected": -3.3125,
      "logps/chosen": -280.0,
      "logps/rejected": -280.0,
      "loss": 0.5115,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.072265625,
      "rewards/margins": 1.1953125,
      "rewards/rejected": -1.265625,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 68.71250790991336,
      "learning_rate": 4.1360946745562133e-07,
      "logits/chosen": -3.265625,
      "logits/rejected": -3.203125,
      "logps/chosen": -272.0,
      "logps/rejected": -290.0,
      "loss": 0.4746,
      "rewards/accuracies": 0.8374999761581421,
      "rewards/chosen": 0.029052734375,
      "rewards/margins": 1.6796875,
      "rewards/rejected": -1.6484375,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 55.46878046949881,
      "learning_rate": 4.076923076923077e-07,
      "logits/chosen": -3.265625,
      "logits/rejected": -3.328125,
      "logps/chosen": -280.0,
      "logps/rejected": -278.0,
      "loss": 0.4711,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.150390625,
      "rewards/margins": 1.140625,
      "rewards/rejected": -1.2890625,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 60.69436763385765,
      "learning_rate": 4.017751479289941e-07,
      "logits/chosen": -3.25,
      "logits/rejected": -3.28125,
      "logps/chosen": -274.0,
      "logps/rejected": -286.0,
      "loss": 0.5292,
      "rewards/accuracies": 0.762499988079071,
      "rewards/chosen": -0.1650390625,
      "rewards/margins": 1.25,
      "rewards/rejected": -1.4140625,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 61.755235473426524,
      "learning_rate": 3.9585798816568044e-07,
      "logits/chosen": -3.28125,
      "logits/rejected": -3.3125,
      "logps/chosen": -274.0,
      "logps/rejected": -282.0,
      "loss": 0.4873,
      "rewards/accuracies": 0.8125,
      "rewards/chosen": -0.044921875,
      "rewards/margins": 1.4375,
      "rewards/rejected": -1.484375,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 64.8584152127513,
      "learning_rate": 3.8994082840236685e-07,
      "logits/chosen": -3.328125,
      "logits/rejected": -3.265625,
      "logps/chosen": -284.0,
      "logps/rejected": -280.0,
      "loss": 0.5342,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.1640625,
      "rewards/margins": 0.87109375,
      "rewards/rejected": -1.03125,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 71.13313796916569,
      "learning_rate": 3.840236686390532e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.28125,
      "logps/chosen": -268.0,
      "logps/rejected": -284.0,
      "loss": 0.455,
      "rewards/accuracies": 0.824999988079071,
      "rewards/chosen": -0.0546875,
      "rewards/margins": 1.65625,
      "rewards/rejected": -1.7109375,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 147.24265223128356,
      "learning_rate": 3.7810650887573966e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.234375,
      "logps/chosen": -284.0,
      "logps/rejected": -284.0,
      "loss": 0.5612,
      "rewards/accuracies": 0.800000011920929,
      "rewards/chosen": -0.27734375,
      "rewards/margins": 1.140625,
      "rewards/rejected": -1.421875,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 57.83608472539185,
      "learning_rate": 3.72189349112426e-07,
      "logits/chosen": -3.25,
      "logits/rejected": -3.25,
      "logps/chosen": -274.0,
      "logps/rejected": -282.0,
      "loss": 0.4624,
      "rewards/accuracies": 0.8374999761581421,
      "rewards/chosen": 0.05712890625,
      "rewards/margins": 1.40625,
      "rewards/rejected": -1.3515625,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -3.3125,
      "eval_logits/rejected": -3.34375,
      "eval_logps/chosen": -272.0,
      "eval_logps/rejected": -276.0,
      "eval_loss": 0.5753124952316284,
      "eval_rewards/accuracies": 0.75,
      "eval_rewards/chosen": -0.039306640625,
      "eval_rewards/margins": 0.703125,
      "eval_rewards/rejected": -0.7421875,
      "eval_runtime": 12.4952,
      "eval_samples_per_second": 16.006,
      "eval_steps_per_second": 0.56,
      "step": 313
    }
  ],
  "logging_steps": 10,
  "max_steps": 939,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}