{
  "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": 116.0621472800408,
      "learning_rate": 5.3191489361702125e-09,
      "logits/chosen": -3.921875,
      "logits/rejected": -3.8125,
      "logps/chosen": -179.0,
      "logps/rejected": -120.0,
      "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": 123.707490939538,
      "learning_rate": 5.3191489361702123e-08,
      "logits/chosen": -3.75,
      "logits/rejected": -3.671875,
      "logps/chosen": -156.0,
      "logps/rejected": -125.0,
      "loss": 0.6803,
      "rewards/accuracies": 0.3611111044883728,
      "rewards/chosen": 0.00592041015625,
      "rewards/margins": 0.0264892578125,
      "rewards/rejected": -0.0205078125,
      "step": 10
    },
    {
      "epoch": 0.06389776357827476,
      "grad_norm": 63.61099541251128,
      "learning_rate": 1.0638297872340425e-07,
      "logits/chosen": -3.71875,
      "logits/rejected": -3.59375,
      "logps/chosen": -161.0,
      "logps/rejected": -128.0,
      "loss": 0.5013,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.0294189453125,
      "rewards/margins": 0.443359375,
      "rewards/rejected": -0.4140625,
      "step": 20
    },
    {
      "epoch": 0.09584664536741214,
      "grad_norm": 16.3990461007188,
      "learning_rate": 1.5957446808510638e-07,
      "logits/chosen": -3.640625,
      "logits/rejected": -3.421875,
      "logps/chosen": -172.0,
      "logps/rejected": -155.0,
      "loss": 0.1265,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1396484375,
      "rewards/margins": 2.6875,
      "rewards/rejected": -2.546875,
      "step": 30
    },
    {
      "epoch": 0.12779552715654952,
      "grad_norm": 0.671529242047923,
      "learning_rate": 2.127659574468085e-07,
      "logits/chosen": -3.53125,
      "logits/rejected": -3.328125,
      "logps/chosen": -152.0,
      "logps/rejected": -185.0,
      "loss": 0.0152,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2265625,
      "rewards/margins": 6.84375,
      "rewards/rejected": -6.625,
      "step": 40
    },
    {
      "epoch": 0.1597444089456869,
      "grad_norm": 1.123751636301978,
      "learning_rate": 2.659574468085106e-07,
      "logits/chosen": -3.421875,
      "logits/rejected": -3.171875,
      "logps/chosen": -150.0,
      "logps/rejected": -228.0,
      "loss": 0.0025,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.248046875,
      "rewards/margins": 10.8125,
      "rewards/rejected": -10.625,
      "step": 50
    },
    {
      "epoch": 0.19169329073482427,
      "grad_norm": 0.08454507441807713,
      "learning_rate": 3.1914893617021275e-07,
      "logits/chosen": -3.3125,
      "logits/rejected": -2.984375,
      "logps/chosen": -155.0,
      "logps/rejected": -254.0,
      "loss": 0.0015,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1513671875,
      "rewards/margins": 13.25,
      "rewards/rejected": -13.125,
      "step": 60
    },
    {
      "epoch": 0.22364217252396165,
      "grad_norm": 0.012771788518112216,
      "learning_rate": 3.7234042553191484e-07,
      "logits/chosen": -3.234375,
      "logits/rejected": -3.0,
      "logps/chosen": -153.0,
      "logps/rejected": -276.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.02685546875,
      "rewards/margins": 15.625,
      "rewards/rejected": -15.6875,
      "step": 70
    },
    {
      "epoch": 0.25559105431309903,
      "grad_norm": 0.20303610510850278,
      "learning_rate": 4.25531914893617e-07,
      "logits/chosen": -3.125,
      "logits/rejected": -2.90625,
      "logps/chosen": -174.0,
      "logps/rejected": -314.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.431640625,
      "rewards/margins": 18.5,
      "rewards/rejected": -18.875,
      "step": 80
    },
    {
      "epoch": 0.28753993610223644,
      "grad_norm": 0.006864178546485895,
      "learning_rate": 4.787234042553192e-07,
      "logits/chosen": -3.046875,
      "logits/rejected": -2.8125,
      "logps/chosen": -166.0,
      "logps/rejected": -330.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.1484375,
      "rewards/margins": 19.375,
      "rewards/rejected": -20.625,
      "step": 90
    },
    {
      "epoch": 0.3194888178913738,
      "grad_norm": 0.039988698214440496,
      "learning_rate": 4.964497041420119e-07,
      "logits/chosen": -3.015625,
      "logits/rejected": -2.8125,
      "logps/chosen": -167.0,
      "logps/rejected": -334.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.1328125,
      "rewards/margins": 20.0,
      "rewards/rejected": -21.25,
      "step": 100
    },
    {
      "epoch": 0.3514376996805112,
      "grad_norm": 0.08090795499045322,
      "learning_rate": 4.905325443786982e-07,
      "logits/chosen": -3.0625,
      "logits/rejected": -2.75,
      "logps/chosen": -168.0,
      "logps/rejected": -348.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.9609375,
      "rewards/margins": 21.625,
      "rewards/rejected": -22.625,
      "step": 110
    },
    {
      "epoch": 0.38338658146964855,
      "grad_norm": 0.0005053190792494335,
      "learning_rate": 4.846153846153846e-07,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.953125,
      "logps/chosen": -178.0,
      "logps/rejected": -330.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.5,
      "rewards/margins": 18.875,
      "rewards/rejected": -20.375,
      "step": 120
    },
    {
      "epoch": 0.41533546325878595,
      "grad_norm": 0.005793842249166428,
      "learning_rate": 4.78698224852071e-07,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.953125,
      "logps/chosen": -185.0,
      "logps/rejected": -328.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -1.3203125,
      "rewards/margins": 18.875,
      "rewards/rejected": -20.125,
      "step": 130
    },
    {
      "epoch": 0.4472843450479233,
      "grad_norm": 0.00547250399529398,
      "learning_rate": 4.727810650887574e-07,
      "logits/chosen": -3.265625,
      "logits/rejected": -2.96875,
      "logps/chosen": -142.0,
      "logps/rejected": -304.0,
      "loss": 0.0001,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.1337890625,
      "rewards/margins": 18.25,
      "rewards/rejected": -18.125,
      "step": 140
    },
    {
      "epoch": 0.4792332268370607,
      "grad_norm": 0.0009553781516234761,
      "learning_rate": 4.668639053254438e-07,
      "logits/chosen": -3.125,
      "logits/rejected": -2.84375,
      "logps/chosen": -164.0,
      "logps/rejected": -300.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.11328125,
      "rewards/margins": 17.625,
      "rewards/rejected": -17.75,
      "step": 150
    },
    {
      "epoch": 0.5111821086261981,
      "grad_norm": 0.15660591509388516,
      "learning_rate": 4.6094674556213014e-07,
      "logits/chosen": -3.0625,
      "logits/rejected": -2.84375,
      "logps/chosen": -164.0,
      "logps/rejected": -320.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.2392578125,
      "rewards/margins": 19.25,
      "rewards/rejected": -19.5,
      "step": 160
    },
    {
      "epoch": 0.5431309904153354,
      "grad_norm": 0.07414583594558251,
      "learning_rate": 4.5502958579881655e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.078125,
      "logps/chosen": -150.0,
      "logps/rejected": -302.0,
      "loss": 0.0002,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.134765625,
      "rewards/margins": 18.5,
      "rewards/rejected": -18.375,
      "step": 170
    },
    {
      "epoch": 0.5750798722044729,
      "grad_norm": 0.01835763457343932,
      "learning_rate": 4.491124260355029e-07,
      "logits/chosen": -3.25,
      "logits/rejected": -3.046875,
      "logps/chosen": -162.0,
      "logps/rejected": -304.0,
      "loss": 0.0023,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.02294921875,
      "rewards/margins": 17.625,
      "rewards/rejected": -17.625,
      "step": 180
    },
    {
      "epoch": 0.6070287539936102,
      "grad_norm": 0.019396673692171084,
      "learning_rate": 4.4319526627218936e-07,
      "logits/chosen": -3.296875,
      "logits/rejected": -3.125,
      "logps/chosen": -162.0,
      "logps/rejected": -304.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.0732421875,
      "rewards/margins": 18.375,
      "rewards/rejected": -18.375,
      "step": 190
    },
    {
      "epoch": 0.6389776357827476,
      "grad_norm": 0.004950996267230264,
      "learning_rate": 4.372781065088757e-07,
      "logits/chosen": -3.21875,
      "logits/rejected": -3.015625,
      "logps/chosen": -165.0,
      "logps/rejected": -316.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.05419921875,
      "rewards/margins": 18.625,
      "rewards/rejected": -18.625,
      "step": 200
    },
    {
      "epoch": 0.670926517571885,
      "grad_norm": 0.004109028420959876,
      "learning_rate": 4.313609467455621e-07,
      "logits/chosen": -3.25,
      "logits/rejected": -3.03125,
      "logps/chosen": -164.0,
      "logps/rejected": -312.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.02001953125,
      "rewards/margins": 19.375,
      "rewards/rejected": -19.375,
      "step": 210
    },
    {
      "epoch": 0.7028753993610224,
      "grad_norm": 0.03343721361256651,
      "learning_rate": 4.2544378698224847e-07,
      "logits/chosen": -3.21875,
      "logits/rejected": -2.921875,
      "logps/chosen": -149.0,
      "logps/rejected": -320.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.07177734375,
      "rewards/margins": 19.875,
      "rewards/rejected": -20.0,
      "step": 220
    },
    {
      "epoch": 0.7348242811501597,
      "grad_norm": 0.0009315852968474835,
      "learning_rate": 4.195266272189349e-07,
      "logits/chosen": -3.15625,
      "logits/rejected": -2.921875,
      "logps/chosen": -147.0,
      "logps/rejected": -330.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.019775390625,
      "rewards/margins": 20.875,
      "rewards/rejected": -20.875,
      "step": 230
    },
    {
      "epoch": 0.7667731629392971,
      "grad_norm": 0.006975720705464563,
      "learning_rate": 4.1360946745562133e-07,
      "logits/chosen": -3.1875,
      "logits/rejected": -2.9375,
      "logps/chosen": -158.0,
      "logps/rejected": -318.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.0037841796875,
      "rewards/margins": 19.5,
      "rewards/rejected": -19.5,
      "step": 240
    },
    {
      "epoch": 0.7987220447284346,
      "grad_norm": 0.007054899693969346,
      "learning_rate": 4.076923076923077e-07,
      "logits/chosen": -3.234375,
      "logits/rejected": -3.046875,
      "logps/chosen": -169.0,
      "logps/rejected": -330.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.09228515625,
      "rewards/margins": 20.5,
      "rewards/rejected": -20.625,
      "step": 250
    },
    {
      "epoch": 0.8306709265175719,
      "grad_norm": 0.00029837762630587707,
      "learning_rate": 4.017751479289941e-07,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.90625,
      "logps/chosen": -142.0,
      "logps/rejected": -324.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.0791015625,
      "rewards/margins": 20.25,
      "rewards/rejected": -20.125,
      "step": 260
    },
    {
      "epoch": 0.8626198083067093,
      "grad_norm": 0.0018546472858726839,
      "learning_rate": 3.9585798816568044e-07,
      "logits/chosen": -3.125,
      "logits/rejected": -2.84375,
      "logps/chosen": -172.0,
      "logps/rejected": -332.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.2119140625,
      "rewards/margins": 20.75,
      "rewards/rejected": -21.0,
      "step": 270
    },
    {
      "epoch": 0.8945686900958466,
      "grad_norm": 0.0007398452733728992,
      "learning_rate": 3.8994082840236685e-07,
      "logits/chosen": -3.078125,
      "logits/rejected": -2.890625,
      "logps/chosen": -192.0,
      "logps/rejected": -346.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.10986328125,
      "rewards/margins": 22.375,
      "rewards/rejected": -22.5,
      "step": 280
    },
    {
      "epoch": 0.9265175718849841,
      "grad_norm": 0.007250178415054528,
      "learning_rate": 3.840236686390532e-07,
      "logits/chosen": -3.09375,
      "logits/rejected": -2.90625,
      "logps/chosen": -156.0,
      "logps/rejected": -338.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.1318359375,
      "rewards/margins": 21.125,
      "rewards/rejected": -21.25,
      "step": 290
    },
    {
      "epoch": 0.9584664536741214,
      "grad_norm": 0.0017287529388297464,
      "learning_rate": 3.7810650887573966e-07,
      "logits/chosen": -3.140625,
      "logits/rejected": -2.890625,
      "logps/chosen": -180.0,
      "logps/rejected": -338.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.3125,
      "rewards/margins": 21.25,
      "rewards/rejected": -21.625,
      "step": 300
    },
    {
      "epoch": 0.9904153354632588,
      "grad_norm": 0.013205636145569603,
      "learning_rate": 3.72189349112426e-07,
      "logits/chosen": -3.109375,
      "logits/rejected": -2.921875,
      "logps/chosen": -192.0,
      "logps/rejected": -340.0,
      "loss": 0.0,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.271484375,
      "rewards/margins": 21.0,
      "rewards/rejected": -21.25,
      "step": 310
    },
    {
      "epoch": 1.0,
      "eval_logits/chosen": -3.21875,
      "eval_logits/rejected": -2.921875,
      "eval_logps/chosen": -151.0,
      "eval_logps/rejected": -340.0,
      "eval_loss": 7.380839633697178e-06,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -0.031982421875,
      "eval_rewards/margins": 22.0,
      "eval_rewards/rejected": -22.125,
      "eval_runtime": 12.3794,
      "eval_samples_per_second": 16.156,
      "eval_steps_per_second": 0.565,
      "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
}