{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6679511881824021,
  "eval_steps": 500,
  "global_step": 130,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05,
      "grad_norm": 6.844920635223389,
      "learning_rate": 5e-05,
      "logits/chosen": -3.09375,
      "logits/rejected": -3.0625,
      "logps/chosen": -258.0,
      "logps/rejected": -322.0,
      "loss": 1.0697,
      "max_memory_allocated (GB)": 94.25,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.30078125,
      "rewards/chosen": 0.67578125,
      "rewards/margins": -0.4140625,
      "rewards/rejected": 1.0859375,
      "step": 10,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.1,
      "grad_norm": 4.300484657287598,
      "learning_rate": 0.0001,
      "logits/chosen": -3.046875,
      "logits/rejected": -3.03125,
      "logps/chosen": -286.0,
      "logps/rejected": -336.0,
      "loss": 0.5736,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.69140625,
      "rewards/chosen": -0.150390625,
      "rewards/margins": 0.8046875,
      "rewards/rejected": -0.95703125,
      "step": 20,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.15,
      "grad_norm": 1.9487425088882446,
      "learning_rate": 0.00015,
      "logits/chosen": -2.859375,
      "logits/rejected": -2.84375,
      "logps/chosen": -294.0,
      "logps/rejected": -394.0,
      "loss": 0.2394,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.90625,
      "rewards/chosen": -1.984375,
      "rewards/margins": 4.8125,
      "rewards/rejected": -6.8125,
      "step": 30,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.21,
      "grad_norm": 1.0650818347930908,
      "learning_rate": 0.0002,
      "logits/chosen": -2.875,
      "logits/rejected": -2.828125,
      "logps/chosen": -284.0,
      "logps/rejected": -388.0,
      "loss": 0.1181,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.95703125,
      "rewards/chosen": -0.62109375,
      "rewards/margins": 5.0625,
      "rewards/rejected": -5.6875,
      "step": 40,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.5032941699028015,
      "learning_rate": 0.00025,
      "logits/chosen": -2.890625,
      "logits/rejected": -2.875,
      "logps/chosen": -278.0,
      "logps/rejected": -412.0,
      "loss": 0.0583,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -0.8828125,
      "rewards/margins": 6.96875,
      "rewards/rejected": -7.84375,
      "step": 50,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.8771920800209045,
      "learning_rate": 0.0003,
      "logits/chosen": -2.828125,
      "logits/rejected": -2.828125,
      "logps/chosen": -312.0,
      "logps/rejected": -474.0,
      "loss": 0.0469,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -2.609375,
      "rewards/margins": 11.5,
      "rewards/rejected": -14.125,
      "step": 60,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.42379140853881836,
      "learning_rate": 0.00035,
      "logits/chosen": -2.84375,
      "logits/rejected": -2.84375,
      "logps/chosen": -280.0,
      "logps/rejected": -436.0,
      "loss": 0.0285,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.9921875,
      "rewards/chosen": -0.9609375,
      "rewards/margins": 9.9375,
      "rewards/rejected": -10.9375,
      "step": 70,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.5195761919021606,
      "learning_rate": 0.0004,
      "logits/chosen": -2.78125,
      "logits/rejected": -2.765625,
      "logps/chosen": -292.0,
      "logps/rejected": -450.0,
      "loss": 0.0266,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.98828125,
      "rewards/chosen": -1.4453125,
      "rewards/margins": 10.75,
      "rewards/rejected": -12.1875,
      "step": 80,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.46,
      "grad_norm": 2.770358085632324,
      "learning_rate": 0.00045000000000000004,
      "logits/chosen": -2.921875,
      "logits/rejected": -2.9375,
      "logps/chosen": -276.0,
      "logps/rejected": -478.0,
      "loss": 0.0486,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -1.5703125,
      "rewards/margins": 13.375,
      "rewards/rejected": -15.0,
      "step": 90,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.1556715965270996,
      "learning_rate": 0.0005,
      "logits/chosen": -2.75,
      "logits/rejected": -2.71875,
      "logps/chosen": -284.0,
      "logps/rejected": -482.0,
      "loss": 0.053,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.9765625,
      "rewards/chosen": -1.3125,
      "rewards/margins": 13.0,
      "rewards/rejected": -14.375,
      "step": 100,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.40476056933403015,
      "learning_rate": 0.0004998477067547739,
      "logits/chosen": -2.875,
      "logits/rejected": -2.875,
      "logps/chosen": -296.0,
      "logps/rejected": -444.0,
      "loss": 0.0384,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.98828125,
      "rewards/chosen": -1.2109375,
      "rewards/margins": 10.5,
      "rewards/rejected": -11.75,
      "step": 110,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.8060222864151001,
      "learning_rate": 0.0004993910125649561,
      "logits/chosen": -3.234375,
      "logits/rejected": -3.234375,
      "logps/chosen": -282.0,
      "logps/rejected": -456.0,
      "loss": 0.0183,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.9921875,
      "rewards/chosen": -0.76171875,
      "rewards/margins": 12.4375,
      "rewards/rejected": -13.1875,
      "step": 120,
      "total_memory_available (GB)": 94.62
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.5584326982498169,
      "learning_rate": 0.0004986304738420684,
      "logits/chosen": -3.140625,
      "logits/rejected": -3.125,
      "logps/chosen": -296.0,
      "logps/rejected": -496.0,
      "loss": 0.0294,
      "max_memory_allocated (GB)": 94.39,
      "memory_allocated (GB)": 51.7,
      "rewards/accuracies": 0.984375,
      "rewards/chosen": -1.875,
      "rewards/margins": 14.8125,
      "rewards/rejected": -16.75,
      "step": 130,
      "total_memory_available (GB)": 94.62
    }
  ],
  "logging_steps": 10,
  "max_steps": 1000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "save_steps": 10,
  "total_flos": 1.0918971071004672e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}