{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9952038369304557,
  "eval_steps": 100,
  "global_step": 83,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011990407673860911,
      "grad_norm": 11.655411235528401,
      "learning_rate": 5.555555555555555e-08,
      "logits/chosen": -1.875,
      "logits/rejected": -1.2421875,
      "logps/chosen": -620.0,
      "logps/rejected": -476.0,
      "loss": 0.6961,
      "rewards/accuracies": 0.20000000298023224,
      "rewards/chosen": 0.00099945068359375,
      "rewards/margins": 0.0050048828125,
      "rewards/rejected": -0.003997802734375,
      "step": 1
    },
    {
      "epoch": 0.11990407673860912,
      "grad_norm": 9.7437445581163,
      "learning_rate": 4.997747415511704e-07,
      "logits/chosen": -1.765625,
      "logits/rejected": -1.2890625,
      "logps/chosen": -396.0,
      "logps/rejected": -402.0,
      "loss": 0.6931,
      "rewards/accuracies": 0.3222222328186035,
      "rewards/chosen": 0.0025177001953125,
      "rewards/margins": -0.0025634765625,
      "rewards/rejected": 0.005096435546875,
      "step": 10
    },
    {
      "epoch": 0.23980815347721823,
      "grad_norm": 7.4219523058604535,
      "learning_rate": 4.732314645373921e-07,
      "logits/chosen": -1.8203125,
      "logits/rejected": -1.21875,
      "logps/chosen": -418.0,
      "logps/rejected": -400.0,
      "loss": 0.6729,
      "rewards/accuracies": 0.5699999928474426,
      "rewards/chosen": 0.1279296875,
      "rewards/margins": 0.041015625,
      "rewards/rejected": 0.08642578125,
      "step": 20
    },
    {
      "epoch": 0.3597122302158273,
      "grad_norm": 6.559255309529087,
      "learning_rate": 4.070549993239106e-07,
      "logits/chosen": -1.859375,
      "logits/rejected": -1.2421875,
      "logps/chosen": -362.0,
      "logps/rejected": -394.0,
      "loss": 0.6572,
      "rewards/accuracies": 0.7099999785423279,
      "rewards/chosen": 0.212890625,
      "rewards/margins": 0.1142578125,
      "rewards/rejected": 0.09814453125,
      "step": 30
    },
    {
      "epoch": 0.47961630695443647,
      "grad_norm": 11.071096096929814,
      "learning_rate": 3.129945153462813e-07,
      "logits/chosen": -1.8515625,
      "logits/rejected": -1.2421875,
      "logps/chosen": -406.0,
      "logps/rejected": -398.0,
      "loss": 0.6418,
      "rewards/accuracies": 0.7799999713897705,
      "rewards/chosen": 0.24609375,
      "rewards/margins": 0.1376953125,
      "rewards/rejected": 0.10791015625,
      "step": 40
    },
    {
      "epoch": 0.5995203836930456,
      "grad_norm": 8.025564513560811,
      "learning_rate": 2.0774979491953776e-07,
      "logits/chosen": -1.7890625,
      "logits/rejected": -1.359375,
      "logps/chosen": -382.0,
      "logps/rejected": -378.0,
      "loss": 0.6333,
      "rewards/accuracies": 0.7699999809265137,
      "rewards/chosen": 0.2275390625,
      "rewards/margins": 0.15234375,
      "rewards/rejected": 0.07568359375,
      "step": 50
    },
    {
      "epoch": 0.7194244604316546,
      "grad_norm": 9.025744560928734,
      "learning_rate": 1.1000630346560116e-07,
      "logits/chosen": -1.8515625,
      "logits/rejected": -1.3125,
      "logps/chosen": -376.0,
      "logps/rejected": -388.0,
      "loss": 0.6318,
      "rewards/accuracies": 0.7300000190734863,
      "rewards/chosen": 0.1904296875,
      "rewards/margins": 0.13671875,
      "rewards/rejected": 0.05419921875,
      "step": 60
    },
    {
      "epoch": 0.8393285371702638,
      "grad_norm": 6.592935931022897,
      "learning_rate": 3.7117715566672176e-08,
      "logits/chosen": -1.71875,
      "logits/rejected": -1.2890625,
      "logps/chosen": -356.0,
      "logps/rejected": -370.0,
      "loss": 0.6223,
      "rewards/accuracies": 0.7599999904632568,
      "rewards/chosen": 0.11865234375,
      "rewards/margins": 0.0947265625,
      "rewards/rejected": 0.0242919921875,
      "step": 70
    },
    {
      "epoch": 0.9592326139088729,
      "grad_norm": 7.695842566054067,
      "learning_rate": 2.0248911852807917e-09,
      "logits/chosen": -1.90625,
      "logits/rejected": -1.3359375,
      "logps/chosen": -394.0,
      "logps/rejected": -390.0,
      "loss": 0.6231,
      "rewards/accuracies": 0.7599999904632568,
      "rewards/chosen": 0.2099609375,
      "rewards/margins": 0.189453125,
      "rewards/rejected": 0.020263671875,
      "step": 80
    },
    {
      "epoch": 0.9952038369304557,
      "step": 83,
      "total_flos": 0.0,
      "train_loss": 0.6455191000398383,
      "train_runtime": 1327.6883,
      "train_samples_per_second": 3.765,
      "train_steps_per_second": 0.063
    }
  ],
  "logging_steps": 10,
  "max_steps": 83,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}