{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.992,
  "eval_steps": 400,
  "global_step": 31,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "abs_diff": 0.3515625,
      "all_logps_1": -716.58203125,
      "all_logps_1_values": -716.58203125,
      "all_logps_2": 368.328125,
      "all_logps_2_values": 368.328125,
      "epoch": 0.032,
      "grad_norm": 15.12199720138031,
      "learning_rate": 2.5e-07,
      "logits/chosen": -1.640625,
      "logits/rejected": -1.7109375,
      "logps/chosen": -2.390625,
      "logps/rejected": -2.359375,
      "loss": 1.8057,
      "original_losses": 1.8046875,
      "rewards/accuracies": 0.375,
      "rewards/chosen": -6.0,
      "rewards/margins": -0.06982421875,
      "rewards/rejected": -5.90625,
      "step": 1,
      "weight": 1.0
    },
    {
      "abs_diff": 0.310546875,
      "all_logps_1": -898.46630859375,
      "all_logps_1_values": -898.46630859375,
      "all_logps_2": 469.18359375,
      "all_logps_2_values": 469.18359375,
      "epoch": 0.16,
      "grad_norm": 9.1493804495839,
      "learning_rate": 9.966191788709714e-07,
      "logits/chosen": -1.5859375,
      "logits/rejected": -1.625,
      "logps/chosen": -2.171875,
      "logps/rejected": -2.171875,
      "loss": 1.7542,
      "original_losses": 1.7578125,
      "rewards/accuracies": 0.4609375,
      "rewards/chosen": -5.4375,
      "rewards/margins": -0.0084228515625,
      "rewards/rejected": -5.4375,
      "step": 5,
      "weight": 1.0
    },
    {
      "abs_diff": 0.498046875,
      "all_logps_1": -776.373046875,
      "all_logps_1_values": -776.373046875,
      "all_logps_2": 409.0,
      "all_logps_2_values": 409.0,
      "epoch": 0.32,
      "grad_norm": 14.347515525649953,
      "learning_rate": 8.83022221559489e-07,
      "logits/chosen": -1.4140625,
      "logits/rejected": -1.3984375,
      "logps/chosen": -2.4375,
      "logps/rejected": -2.40625,
      "loss": 2.0064,
      "original_losses": 2.0,
      "rewards/accuracies": 0.4124999940395355,
      "rewards/chosen": -6.09375,
      "rewards/margins": -0.0703125,
      "rewards/rejected": -6.03125,
      "step": 10,
      "weight": 1.0
    },
    {
      "abs_diff": 0.400390625,
      "all_logps_1": -866.23828125,
      "all_logps_1_values": -866.23828125,
      "all_logps_2": 444.09063720703125,
      "all_logps_2_values": 444.09063720703125,
      "epoch": 0.48,
      "grad_norm": 9.610798301744433,
      "learning_rate": 6.434016163555451e-07,
      "logits/chosen": -1.59375,
      "logits/rejected": -1.59375,
      "logps/chosen": -2.375,
      "logps/rejected": -2.34375,
      "loss": 1.9072,
      "original_losses": 1.90625,
      "rewards/accuracies": 0.375,
      "rewards/chosen": -5.9375,
      "rewards/margins": -0.06787109375,
      "rewards/rejected": -5.875,
      "step": 15,
      "weight": 1.0
    },
    {
      "abs_diff": 0.302734375,
      "all_logps_1": -795.837890625,
      "all_logps_1_values": -795.837890625,
      "all_logps_2": 418.5874938964844,
      "all_logps_2_values": 418.5874938964844,
      "epoch": 0.64,
      "grad_norm": 8.931859579637203,
      "learning_rate": 3.56598383644455e-07,
      "logits/chosen": -1.5859375,
      "logits/rejected": -1.5859375,
      "logps/chosen": -2.390625,
      "logps/rejected": -2.359375,
      "loss": 1.835,
      "original_losses": 1.8359375,
      "rewards/accuracies": 0.3375000059604645,
      "rewards/chosen": -5.96875,
      "rewards/margins": -0.09228515625,
      "rewards/rejected": -5.875,
      "step": 20,
      "weight": 1.0
    },
    {
      "abs_diff": 0.4609375,
      "all_logps_1": -784.2203369140625,
      "all_logps_1_values": -784.2203369140625,
      "all_logps_2": 412.63751220703125,
      "all_logps_2_values": 412.63751220703125,
      "epoch": 0.8,
      "grad_norm": 26.344861955814572,
      "learning_rate": 1.1697777844051104e-07,
      "logits/chosen": -1.4296875,
      "logits/rejected": -1.375,
      "logps/chosen": -2.375,
      "logps/rejected": -2.4375,
      "loss": 1.7904,
      "original_losses": 1.7890625,
      "rewards/accuracies": 0.44999998807907104,
      "rewards/chosen": -5.9375,
      "rewards/margins": 0.1787109375,
      "rewards/rejected": -6.09375,
      "step": 25,
      "weight": 1.0
    },
    {
      "abs_diff": 0.455078125,
      "all_logps_1": -790.1726684570312,
      "all_logps_1_values": -790.1726684570312,
      "all_logps_2": 412.5531311035156,
      "all_logps_2_values": 412.5531311035156,
      "epoch": 0.96,
      "grad_norm": 10.248900083694645,
      "learning_rate": 3.380821129028488e-09,
      "logits/chosen": -1.390625,
      "logits/rejected": -1.4375,
      "logps/chosen": -2.484375,
      "logps/rejected": -2.421875,
      "loss": 1.9701,
      "original_losses": 1.96875,
      "rewards/accuracies": 0.4312500059604645,
      "rewards/chosen": -6.21875,
      "rewards/margins": -0.1416015625,
      "rewards/rejected": -6.0625,
      "step": 30,
      "weight": 1.0
    },
    {
      "epoch": 0.992,
      "step": 31,
      "total_flos": 0.0,
      "train_loss": 1.8891444052419355,
      "train_runtime": 316.7887,
      "train_samples_per_second": 3.157,
      "train_steps_per_second": 0.098
    }
  ],
  "logging_steps": 5,
  "max_steps": 31,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1000000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}