{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.99581589958159,
  "eval_steps": 500,
  "global_step": 119,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 4.166666666666666e-08,
      "logits/chosen": -2.6852197647094727,
      "logits/rejected": -2.6903719902038574,
      "logps/chosen": -263.7275390625,
      "logps/rejected": -230.14215087890625,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.08,
      "learning_rate": 4.1666666666666667e-07,
      "logits/chosen": -2.7384653091430664,
      "logits/rejected": -2.737618923187256,
      "logps/chosen": -278.2864990234375,
      "logps/rejected": -254.0323944091797,
      "loss": 0.6927,
      "rewards/accuracies": 0.4722222089767456,
      "rewards/chosen": 0.0012941197492182255,
      "rewards/margins": 0.00029929642914794385,
      "rewards/rejected": 0.000994823407381773,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.8008759021759033,
      "logits/rejected": -2.7767581939697266,
      "logps/chosen": -285.1484375,
      "logps/rejected": -270.07513427734375,
      "loss": 0.6848,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": 0.019194435328245163,
      "rewards/margins": 0.015600068494677544,
      "rewards/rejected": 0.0035943654365837574,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.7777934074401855,
      "logits/rejected": -2.7765698432922363,
      "logps/chosen": -259.2841796875,
      "logps/rejected": -247.3909454345703,
      "loss": 0.6659,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": 0.011103945784270763,
      "rewards/margins": 0.06937336921691895,
      "rewards/rejected": -0.058269429951906204,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.7775871753692627,
      "logits/rejected": -2.7704668045043945,
      "logps/chosen": -282.95001220703125,
      "logps/rejected": -281.3163146972656,
      "loss": 0.6464,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.08944134414196014,
      "rewards/margins": 0.11117666959762573,
      "rewards/rejected": -0.20061802864074707,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -2.800177574157715,
      "logits/rejected": -2.8010470867156982,
      "logps/chosen": -316.843505859375,
      "logps/rejected": -291.880126953125,
      "loss": 0.6165,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.18551835417747498,
      "rewards/margins": 0.25780096650123596,
      "rewards/rejected": -0.44331932067871094,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.77445125579834,
      "logits/rejected": -2.755255937576294,
      "logps/chosen": -318.5345764160156,
      "logps/rejected": -289.75677490234375,
      "loss": 0.6159,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.2439030408859253,
      "rewards/margins": 0.2745346128940582,
      "rewards/rejected": -0.5184376239776611,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.769376277923584,
      "logits/rejected": -2.741185188293457,
      "logps/chosen": -317.98175048828125,
      "logps/rejected": -279.78143310546875,
      "loss": 0.6032,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.35874003171920776,
      "rewards/margins": 0.24173276126384735,
      "rewards/rejected": -0.6004728078842163,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -2.7139642238616943,
      "logits/rejected": -2.6934971809387207,
      "logps/chosen": -261.20501708984375,
      "logps/rejected": -310.49908447265625,
      "loss": 0.6016,
      "rewards/accuracies": 0.71875,
      "rewards/chosen": -0.349643737077713,
      "rewards/margins": 0.3635331690311432,
      "rewards/rejected": -0.7131768465042114,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -2.7756199836730957,
      "logits/rejected": -2.76503586769104,
      "logps/chosen": -308.0489501953125,
      "logps/rejected": -295.3426208496094,
      "loss": 0.5751,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.3851245045661926,
      "rewards/margins": 0.27352777123451233,
      "rewards/rejected": -0.6586521863937378,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -2.7100021839141846,
      "logits/rejected": -2.685563325881958,
      "logps/chosen": -298.96319580078125,
      "logps/rejected": -354.50439453125,
      "loss": 0.5944,
      "rewards/accuracies": 0.6499999761581421,
      "rewards/chosen": -0.4978106915950775,
      "rewards/margins": 0.2795547842979431,
      "rewards/rejected": -0.777365505695343,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -2.7827987670898438,
      "logits/rejected": -2.761164665222168,
      "logps/chosen": -340.18597412109375,
      "logps/rejected": -337.910400390625,
      "loss": 0.5949,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.42213040590286255,
      "rewards/margins": 0.35163527727127075,
      "rewards/rejected": -0.7737656831741333,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.6248922909007353,
      "train_runtime": 1974.7606,
      "train_samples_per_second": 7.739,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 10,
  "max_steps": 119,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}