{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.992,
  "eval_steps": 500,
  "global_step": 93,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 5.000000000000001e-07,
      "logits/chosen": 0.04800860583782196,
      "logits/rejected": 0.13088344037532806,
      "logps/chosen": -344.1112976074219,
      "logps/rejected": -286.2291259765625,
      "loss": 0.4028,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.11,
      "learning_rate": 5e-06,
      "logits/chosen": 0.1340106725692749,
      "logits/rejected": 0.24628029763698578,
      "logps/chosen": -358.9225769042969,
      "logps/rejected": -352.1644592285156,
      "loss": 0.3705,
      "rewards/accuracies": 0.4236111044883728,
      "rewards/chosen": 7.527708658017218e-05,
      "rewards/margins": 3.0965995392762125e-05,
      "rewards/rejected": 4.431110937730409e-05,
      "step": 10
    },
    {
      "epoch": 0.21,
      "learning_rate": 4.823045180793914e-06,
      "logits/chosen": 0.12768682837486267,
      "logits/rejected": 0.17665454745292664,
      "logps/chosen": -329.0359802246094,
      "logps/rejected": -341.37432861328125,
      "loss": 0.3571,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": 0.0001585387362865731,
      "rewards/margins": 0.0007194522768259048,
      "rewards/rejected": -0.0005609134095720947,
      "step": 20
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.317231129607859e-06,
      "logits/chosen": 0.0687670111656189,
      "logits/rejected": 0.1390002816915512,
      "logps/chosen": -366.32183837890625,
      "logps/rejected": -358.305908203125,
      "loss": 0.3713,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": 0.00020733498968183994,
      "rewards/margins": 0.002162316581234336,
      "rewards/rejected": -0.001954981591552496,
      "step": 30
    },
    {
      "epoch": 0.43,
      "learning_rate": 3.55416283362546e-06,
      "logits/chosen": 0.0996643453836441,
      "logits/rejected": 0.16496673226356506,
      "logps/chosen": -354.90643310546875,
      "logps/rejected": -347.098388671875,
      "loss": 0.3713,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.00043527656816877425,
      "rewards/margins": 0.003645308781415224,
      "rewards/rejected": -0.004080585204064846,
      "step": 40
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.6418631827326857e-06,
      "logits/chosen": 0.19002003967761993,
      "logits/rejected": 0.2053782194852829,
      "logps/chosen": -296.21661376953125,
      "logps/rejected": -323.6202697753906,
      "loss": 0.3591,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -6.220813520485535e-05,
      "rewards/margins": 0.00692623108625412,
      "rewards/rejected": -0.006988438777625561,
      "step": 50
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7094808327579401e-06,
      "logits/chosen": 0.14944681525230408,
      "logits/rejected": 0.1665990650653839,
      "logps/chosen": -376.69952392578125,
      "logps/rejected": -375.88787841796875,
      "loss": 0.3669,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.0006484203622676432,
      "rewards/margins": 0.011537977494299412,
      "rewards/rejected": -0.012186397798359394,
      "step": 60
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.890074238378074e-07,
      "logits/chosen": 0.14019212126731873,
      "logits/rejected": 0.19166143238544464,
      "logps/chosen": -390.3976135253906,
      "logps/rejected": -334.20904541015625,
      "loss": 0.3606,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.0038049519062042236,
      "rewards/margins": 0.010462903417646885,
      "rewards/rejected": -0.014267854392528534,
      "step": 70
    },
    {
      "epoch": 0.85,
      "learning_rate": 2.965923349633779e-07,
      "logits/chosen": 0.15474095940589905,
      "logits/rejected": 0.15488487482070923,
      "logps/chosen": -365.92510986328125,
      "logps/rejected": -369.5127868652344,
      "loss": 0.3573,
      "rewards/accuracies": 0.606249988079071,
      "rewards/chosen": -0.00941461231559515,
      "rewards/margins": 0.010669920593500137,
      "rewards/rejected": -0.020084530115127563,
      "step": 80
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.6100130092037704e-08,
      "logits/chosen": 0.1559561789035797,
      "logits/rejected": 0.21619243919849396,
      "logps/chosen": -373.12786865234375,
      "logps/rejected": -362.0654296875,
      "loss": 0.357,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.005230956245213747,
      "rewards/margins": 0.01701132394373417,
      "rewards/rejected": -0.022242281585931778,
      "step": 90
    },
    {
      "epoch": 0.99,
      "step": 93,
      "total_flos": 0.0,
      "train_loss": 0.36343905329704285,
      "train_runtime": 3130.5093,
      "train_samples_per_second": 1.917,
      "train_steps_per_second": 0.03
    }
  ],
  "logging_steps": 10,
  "max_steps": 93,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}