{
  "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": 15.786988646394411,
      "learning_rate": 4.545454545454545e-08,
      "logits/chosen": -13.905267715454102,
      "logits/rejected": -14.118387222290039,
      "logps/chosen": -350.8895263671875,
      "logps/rejected": -446.6286926269531,
      "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": 15.908099576913655,
      "learning_rate": 4.545454545454545e-07,
      "logits/chosen": -14.040081024169922,
      "logits/rejected": -14.157392501831055,
      "logps/chosen": -416.2701416015625,
      "logps/rejected": -449.4697265625,
      "loss": 0.693,
      "rewards/accuracies": 0.5138888955116272,
      "rewards/chosen": 0.0006088384543545544,
      "rewards/margins": 0.008209776133298874,
      "rewards/rejected": -0.007600938435643911,
      "step": 10
    },
    {
      "epoch": 0.3791469194312796,
      "grad_norm": 14.229474825008781,
      "learning_rate": 4.885348141000122e-07,
      "logits/chosen": -13.39338207244873,
      "logits/rejected": -13.542058944702148,
      "logps/chosen": -392.9753723144531,
      "logps/rejected": -427.68096923828125,
      "loss": 0.6892,
      "rewards/accuracies": 0.5062500238418579,
      "rewards/chosen": 0.010071685537695885,
      "rewards/margins": 0.003802267834544182,
      "rewards/rejected": 0.006269416771829128,
      "step": 20
    },
    {
      "epoch": 0.5687203791469194,
      "grad_norm": 15.853985724357454,
      "learning_rate": 4.5025027361734613e-07,
      "logits/chosen": -14.269427299499512,
      "logits/rejected": -13.808093070983887,
      "logps/chosen": -412.9443359375,
      "logps/rejected": -428.38494873046875,
      "loss": 0.674,
      "rewards/accuracies": 0.612500011920929,
      "rewards/chosen": 0.04771440848708153,
      "rewards/margins": 0.035354893654584885,
      "rewards/rejected": 0.012359511107206345,
      "step": 30
    },
    {
      "epoch": 0.7582938388625592,
      "grad_norm": 14.687978809678542,
      "learning_rate": 3.893311157806091e-07,
      "logits/chosen": -13.886492729187012,
      "logits/rejected": -13.28197956085205,
      "logps/chosen": -374.98211669921875,
      "logps/rejected": -366.5968322753906,
      "loss": 0.657,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": 0.13442906737327576,
      "rewards/margins": 0.07902240008115768,
      "rewards/rejected": 0.05540664866566658,
      "step": 40
    },
    {
      "epoch": 0.9478672985781991,
      "grad_norm": 15.872142673244408,
      "learning_rate": 3.126631330646801e-07,
      "logits/chosen": -14.917936325073242,
      "logits/rejected": -14.90648078918457,
      "logps/chosen": -429.6836853027344,
      "logps/rejected": -480.3504943847656,
      "loss": 0.6344,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": 0.24091288447380066,
      "rewards/margins": 0.1330389827489853,
      "rewards/rejected": 0.10787389427423477,
      "step": 50
    },
    {
      "epoch": 1.1374407582938388,
      "grad_norm": 14.061428605486398,
      "learning_rate": 2.2891223348923882e-07,
      "logits/chosen": -14.622962951660156,
      "logits/rejected": -14.403157234191895,
      "logps/chosen": -415.7464904785156,
      "logps/rejected": -441.731201171875,
      "loss": 0.6063,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": 0.3395090103149414,
      "rewards/margins": 0.22218124568462372,
      "rewards/rejected": 0.11732780933380127,
      "step": 60
    },
    {
      "epoch": 1.3270142180094786,
      "grad_norm": 12.963152293888875,
      "learning_rate": 1.4754491880085317e-07,
      "logits/chosen": -14.022384643554688,
      "logits/rejected": -13.828951835632324,
      "logps/chosen": -382.23468017578125,
      "logps/rejected": -418.2818908691406,
      "loss": 0.6011,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": 0.3396778702735901,
      "rewards/margins": 0.22157195210456848,
      "rewards/rejected": 0.118105947971344,
      "step": 70
    },
    {
      "epoch": 1.5165876777251186,
      "grad_norm": 12.394681314131397,
      "learning_rate": 7.775827023107834e-08,
      "logits/chosen": -13.705121040344238,
      "logits/rejected": -14.205709457397461,
      "logps/chosen": -367.263427734375,
      "logps/rejected": -423.30841064453125,
      "loss": 0.5788,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": 0.36119210720062256,
      "rewards/margins": 0.3365553319454193,
      "rewards/rejected": 0.024636749178171158,
      "step": 80
    },
    {
      "epoch": 1.7061611374407581,
      "grad_norm": 14.456589635016153,
      "learning_rate": 2.7440387297912122e-08,
      "logits/chosen": -13.98394775390625,
      "logits/rejected": -14.161648750305176,
      "logps/chosen": -399.45458984375,
      "logps/rejected": -447.48828125,
      "loss": 0.5766,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": 0.3995341658592224,
      "rewards/margins": 0.34082064032554626,
      "rewards/rejected": 0.05871356278657913,
      "step": 90
    },
    {
      "epoch": 1.8957345971563981,
      "grad_norm": 13.44211674398592,
      "learning_rate": 2.27878296044029e-09,
      "logits/chosen": -14.160197257995605,
      "logits/rejected": -14.141824722290039,
      "logps/chosen": -392.3072509765625,
      "logps/rejected": -421.604248046875,
      "loss": 0.5732,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": 0.4029002785682678,
      "rewards/margins": 0.27652695775032043,
      "rewards/rejected": 0.1263733208179474,
      "step": 100
    },
    {
      "epoch": 1.8957345971563981,
      "eval_logits/chosen": -13.292621612548828,
      "eval_logits/rejected": -12.66539478302002,
      "eval_logps/chosen": -372.0066833496094,
      "eval_logps/rejected": -373.4093933105469,
      "eval_loss": 0.5971602201461792,
      "eval_rewards/accuracies": 0.6770833134651184,
      "eval_rewards/chosen": 0.3533553183078766,
      "eval_rewards/margins": 0.24372106790542603,
      "eval_rewards/rejected": 0.10963428020477295,
      "eval_runtime": 20.0916,
      "eval_samples_per_second": 37.329,
      "eval_steps_per_second": 1.195,
      "step": 100
    },
    {
      "epoch": 1.971563981042654,
      "step": 104,
      "total_flos": 0.0,
      "train_loss": 0.627926590350958,
      "train_runtime": 756.4701,
      "train_samples_per_second": 17.846,
      "train_steps_per_second": 0.137
    }
  ],
  "logging_steps": 10,
  "max_steps": 104,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}