{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.3472803347280333,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 0.4676998555660248,
      "learning_rate": 8.620689655172415e-07,
      "logits/chosen": 1.7077701091766357,
      "logits/rejected": 1.8646482229232788,
      "logps/chosen": -85.7728271484375,
      "logps/rejected": -88.1952896118164,
      "loss": 0.6938,
      "rewards/accuracies": 0.32499998807907104,
      "rewards/chosen": -0.004770822823047638,
      "rewards/margins": -0.007881464436650276,
      "rewards/rejected": 0.0031106427777558565,
      "step": 10
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 0.43230223655700684,
      "learning_rate": 1.724137931034483e-06,
      "logits/chosen": 1.8023220300674438,
      "logits/rejected": 1.8210970163345337,
      "logps/chosen": -78.37618255615234,
      "logps/rejected": -75.44720458984375,
      "loss": 0.6931,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.0056939031928777695,
      "rewards/margins": -0.003848772030323744,
      "rewards/rejected": -0.001845130929723382,
      "step": 20
    },
    {
      "epoch": 1.00418410041841,
      "grad_norm": 2.431753396987915,
      "learning_rate": 2.5862068965517246e-06,
      "logits/chosen": 1.8643144369125366,
      "logits/rejected": 1.8500900268554688,
      "logps/chosen": -86.84412384033203,
      "logps/rejected": -90.78925323486328,
      "loss": 0.6946,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -0.0008674233104102314,
      "rewards/margins": -0.008064134046435356,
      "rewards/rejected": 0.0071967123076319695,
      "step": 30
    },
    {
      "epoch": 1.3389121338912133,
      "grad_norm": 0.5327289700508118,
      "learning_rate": 3.448275862068966e-06,
      "logits/chosen": 1.7701479196548462,
      "logits/rejected": 1.7749736309051514,
      "logps/chosen": -83.61552429199219,
      "logps/rejected": -72.89176177978516,
      "loss": 0.6939,
      "rewards/accuracies": 0.4625000059604645,
      "rewards/chosen": -0.0019066383829340339,
      "rewards/margins": -0.0035008196718990803,
      "rewards/rejected": 0.0015941811725497246,
      "step": 40
    },
    {
      "epoch": 1.6736401673640167,
      "grad_norm": 0.5017096996307373,
      "learning_rate": 4.310344827586207e-06,
      "logits/chosen": 1.7465788125991821,
      "logits/rejected": 1.7925498485565186,
      "logps/chosen": -84.85249328613281,
      "logps/rejected": -88.79469299316406,
      "loss": 0.6938,
      "rewards/accuracies": 0.550000011920929,
      "rewards/chosen": 0.0003656863118521869,
      "rewards/margins": 0.00012080222222721204,
      "rewards/rejected": 0.00024488387862220407,
      "step": 50
    },
    {
      "epoch": 2.00836820083682,
      "grad_norm": 0.4994066059589386,
      "learning_rate": 4.999818897894192e-06,
      "logits/chosen": 1.8359416723251343,
      "logits/rejected": 1.814234733581543,
      "logps/chosen": -79.98396301269531,
      "logps/rejected": -66.29866027832031,
      "loss": 0.6952,
      "rewards/accuracies": 0.4375,
      "rewards/chosen": 0.00456579215824604,
      "rewards/margins": -0.002460189163684845,
      "rewards/rejected": 0.007025980856269598,
      "step": 60
    },
    {
      "epoch": 2.3430962343096233,
      "grad_norm": 0.6425362825393677,
      "learning_rate": 4.9934830787948756e-06,
      "logits/chosen": 1.7550818920135498,
      "logits/rejected": 1.7802051305770874,
      "logps/chosen": -83.60951232910156,
      "logps/rejected": -70.81080627441406,
      "loss": 0.6937,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": 0.004059882368892431,
      "rewards/margins": 0.00932090263813734,
      "rewards/rejected": -0.005261021666228771,
      "step": 70
    },
    {
      "epoch": 2.6778242677824267,
      "grad_norm": 0.5591297149658203,
      "learning_rate": 4.978118375700895e-06,
      "logits/chosen": 1.7742525339126587,
      "logits/rejected": 1.8790937662124634,
      "logps/chosen": -86.7696533203125,
      "logps/rejected": -84.90727233886719,
      "loss": 0.6927,
      "rewards/accuracies": 0.574999988079071,
      "rewards/chosen": 0.0033963967580348253,
      "rewards/margins": 0.005500240251421928,
      "rewards/rejected": -0.0021038432605564594,
      "step": 80
    },
    {
      "epoch": 3.01255230125523,
      "grad_norm": 0.6774541735649109,
      "learning_rate": 4.953780424089803e-06,
      "logits/chosen": 1.8815631866455078,
      "logits/rejected": 1.911620855331421,
      "logps/chosen": -92.14398193359375,
      "logps/rejected": -82.83556365966797,
      "loss": 0.6923,
      "rewards/accuracies": 0.5,
      "rewards/chosen": 0.007645039353519678,
      "rewards/margins": 0.005134005565196276,
      "rewards/rejected": 0.0025110342539846897,
      "step": 90
    },
    {
      "epoch": 3.3472803347280333,
      "grad_norm": 0.6908814907073975,
      "learning_rate": 4.920557351506409e-06,
      "logits/chosen": 1.7718560695648193,
      "logits/rejected": 1.8823477029800415,
      "logps/chosen": -91.2622299194336,
      "logps/rejected": -96.467529296875,
      "loss": 0.6894,
      "rewards/accuracies": 0.6000000238418579,
      "rewards/chosen": 0.004941435065120459,
      "rewards/margins": 0.01582186669111252,
      "rewards/rejected": -0.010880433022975922,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 580,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3.405329030394675e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}