{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 500,
  "global_step": 156,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.064,
      "grad_norm": 46.9622688293457,
      "learning_rate": 6.249999999999999e-07,
      "logits/chosen": -1.849869966506958,
      "logits/rejected": -0.29363900423049927,
      "logps/chosen": -214.13339233398438,
      "logps/rejected": -737.3911743164062,
      "loss": 0.7092,
      "rewards/accuracies": 0.46875,
      "rewards/chosen": 0.004342720843851566,
      "rewards/margins": 0.025443650782108307,
      "rewards/rejected": -0.021100929006934166,
      "step": 10
    },
    {
      "epoch": 0.128,
      "grad_norm": 13.521223068237305,
      "learning_rate": 9.979871469976195e-07,
      "logits/chosen": -1.8478429317474365,
      "logits/rejected": -0.2751621603965759,
      "logps/chosen": -240.21755981445312,
      "logps/rejected": -844.638427734375,
      "loss": 0.486,
      "rewards/accuracies": 0.9468750357627869,
      "rewards/chosen": 0.0402272529900074,
      "rewards/margins": 0.6919995546340942,
      "rewards/rejected": -0.6517722606658936,
      "step": 20
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.762617290019989,
      "learning_rate": 9.755282581475767e-07,
      "logits/chosen": -2.1506738662719727,
      "logits/rejected": -0.8184519410133362,
      "logps/chosen": -241.9251251220703,
      "logps/rejected": -829.9989624023438,
      "loss": 0.1228,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -0.33827781677246094,
      "rewards/margins": 4.653146266937256,
      "rewards/rejected": -4.991424083709717,
      "step": 30
    },
    {
      "epoch": 0.256,
      "grad_norm": 1.0136756896972656,
      "learning_rate": 9.29224396800933e-07,
      "logits/chosen": -2.6077799797058105,
      "logits/rejected": -1.6289113759994507,
      "logps/chosen": -261.08819580078125,
      "logps/rejected": -949.7811889648438,
      "loss": 0.0522,
      "rewards/accuracies": 0.9937499761581421,
      "rewards/chosen": -2.5243332386016846,
      "rewards/margins": 14.23307991027832,
      "rewards/rejected": -16.75741195678711,
      "step": 40
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.15248946845531464,
      "learning_rate": 8.613974319136957e-07,
      "logits/chosen": -2.809011459350586,
      "logits/rejected": -1.9814908504486084,
      "logps/chosen": -262.7459411621094,
      "logps/rejected": -1068.808349609375,
      "loss": 0.0401,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -4.401086807250977,
      "rewards/margins": 23.641544342041016,
      "rewards/rejected": -28.04262924194336,
      "step": 50
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.12552589178085327,
      "learning_rate": 7.754484907260512e-07,
      "logits/chosen": -2.7779171466827393,
      "logits/rejected": -2.0284547805786133,
      "logps/chosen": -282.9354248046875,
      "logps/rejected": -1090.5594482421875,
      "loss": 0.036,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -5.416067123413086,
      "rewards/margins": 25.74221420288086,
      "rewards/rejected": -31.158283233642578,
      "step": 60
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.17482662200927734,
      "learning_rate": 6.756874120406714e-07,
      "logits/chosen": -2.7689507007598877,
      "logits/rejected": -1.9573109149932861,
      "logps/chosen": -263.86279296875,
      "logps/rejected": -1074.0966796875,
      "loss": 0.0455,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -4.39678955078125,
      "rewards/margins": 24.96966552734375,
      "rewards/rejected": -29.366456985473633,
      "step": 70
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.11299557238817215,
      "learning_rate": 5.671166329088277e-07,
      "logits/chosen": -2.643343448638916,
      "logits/rejected": -1.8395075798034668,
      "logps/chosen": -271.737060546875,
      "logps/rejected": -1051.107421875,
      "loss": 0.0341,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -3.8055081367492676,
      "rewards/margins": 22.575603485107422,
      "rewards/rejected": -26.381113052368164,
      "step": 80
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.12716512382030487,
      "learning_rate": 4.5518034554828327e-07,
      "logits/chosen": -2.627561092376709,
      "logits/rejected": -1.8044594526290894,
      "logps/chosen": -267.9153747558594,
      "logps/rejected": -1012.906494140625,
      "loss": 0.046,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -3.2899169921875,
      "rewards/margins": 20.98080825805664,
      "rewards/rejected": -24.270723342895508,
      "step": 90
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.14720195531845093,
      "learning_rate": 3.454915028125263e-07,
      "logits/chosen": -2.617551803588867,
      "logits/rejected": -1.7380447387695312,
      "logps/chosen": -254.84658813476562,
      "logps/rejected": -996.9198608398438,
      "loss": 0.0315,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.6352035999298096,
      "rewards/margins": 19.858867645263672,
      "rewards/rejected": -22.49407196044922,
      "step": 100
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.1784060150384903,
      "learning_rate": 2.4355036129704696e-07,
      "logits/chosen": -2.598905324935913,
      "logits/rejected": -1.7216695547103882,
      "logps/chosen": -248.74453735351562,
      "logps/rejected": -993.6848754882812,
      "loss": 0.036,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -2.5882961750030518,
      "rewards/margins": 18.992816925048828,
      "rewards/rejected": -21.58111572265625,
      "step": 110
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.11419311910867691,
      "learning_rate": 1.5446867550656767e-07,
      "logits/chosen": -2.5767769813537598,
      "logits/rejected": -1.681131362915039,
      "logps/chosen": -256.5839538574219,
      "logps/rejected": -994.3280029296875,
      "loss": 0.0745,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -2.4366414546966553,
      "rewards/margins": 18.98566436767578,
      "rewards/rejected": -21.422306060791016,
      "step": 120
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.10517225414514542,
      "learning_rate": 8.271337313934867e-08,
      "logits/chosen": -2.5168538093566895,
      "logits/rejected": -1.6293193101882935,
      "logps/chosen": -269.16278076171875,
      "logps/rejected": -979.6190795898438,
      "loss": 0.0499,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -2.368694543838501,
      "rewards/margins": 18.073989868164062,
      "rewards/rejected": -20.442684173583984,
      "step": 130
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.1312648206949234,
      "learning_rate": 3.188256468013139e-08,
      "logits/chosen": -2.5698208808898926,
      "logits/rejected": -1.6306824684143066,
      "logps/chosen": -250.3894805908203,
      "logps/rejected": -1000.3046875,
      "loss": 0.0339,
      "rewards/accuracies": 0.996874988079071,
      "rewards/chosen": -2.0379717350006104,
      "rewards/margins": 18.880815505981445,
      "rewards/rejected": -20.918787002563477,
      "step": 140
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.11260352283716202,
      "learning_rate": 4.5251191160326495e-09,
      "logits/chosen": -2.5329906940460205,
      "logits/rejected": -1.5752902030944824,
      "logps/chosen": -271.48443603515625,
      "logps/rejected": -1053.84765625,
      "loss": 0.0735,
      "rewards/accuracies": 0.987500011920929,
      "rewards/chosen": -2.3061203956604004,
      "rewards/margins": 19.025531768798828,
      "rewards/rejected": -21.331653594970703,
      "step": 150
    },
    {
      "epoch": 0.9984,
      "step": 156,
      "total_flos": 1.1115841451898962e+18,
      "train_loss": 0.1217762088546386,
      "train_runtime": 6099.7709,
      "train_samples_per_second": 0.82,
      "train_steps_per_second": 0.026
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.1115841451898962e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}