{
  "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": -2.2385096549987793,
      "logits/rejected": -2.291027545928955,
      "logps/chosen": -439.3824157714844,
      "logps/pi_response": -185.73153686523438,
      "logps/ref_response": -185.73153686523438,
      "logps/rejected": -431.2404479980469,
      "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": -2.4213032722473145,
      "logits/rejected": -2.3896849155426025,
      "logps/chosen": -374.2062683105469,
      "logps/pi_response": -158.41744995117188,
      "logps/ref_response": -158.5709228515625,
      "logps/rejected": -385.7984924316406,
      "loss": 0.6911,
      "rewards/accuracies": 0.5,
      "rewards/chosen": -0.012851297855377197,
      "rewards/margins": 0.0007397261215373874,
      "rewards/rejected": -0.013591023162007332,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.3737921714782715,
      "logits/rejected": -2.369208812713623,
      "logps/chosen": -387.66375732421875,
      "logps/pi_response": -155.94168090820312,
      "logps/ref_response": -157.18533325195312,
      "logps/rejected": -453.19989013671875,
      "loss": 0.6677,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -0.3228910565376282,
      "rewards/margins": 0.11437133699655533,
      "rewards/rejected": -0.4372623860836029,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.4200503826141357,
      "logits/rejected": -2.3776440620422363,
      "logps/chosen": -435.971435546875,
      "logps/pi_response": -165.8021240234375,
      "logps/ref_response": -160.6427001953125,
      "logps/rejected": -486.2308044433594,
      "loss": 0.639,
      "rewards/accuracies": 0.5874999761581421,
      "rewards/chosen": -0.6127233505249023,
      "rewards/margins": 0.2512531876564026,
      "rewards/rejected": -0.8639765977859497,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.38213849067688,
      "logits/rejected": -2.3052191734313965,
      "logps/chosen": -429.8724060058594,
      "logps/pi_response": -177.0037841796875,
      "logps/ref_response": -168.98922729492188,
      "logps/rejected": -489.49273681640625,
      "loss": 0.6041,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -0.5370320081710815,
      "rewards/margins": 0.4057961404323578,
      "rewards/rejected": -0.9428282976150513,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -2.2526650428771973,
      "logits/rejected": -2.224208354949951,
      "logps/chosen": -455.1700744628906,
      "logps/pi_response": -201.5770721435547,
      "logps/ref_response": -166.82545471191406,
      "logps/rejected": -479.81719970703125,
      "loss": 0.5879,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.7905241847038269,
      "rewards/margins": 0.31640949845314026,
      "rewards/rejected": -1.106933832168579,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.1918423175811768,
      "logits/rejected": -2.138241767883301,
      "logps/chosen": -483.14605712890625,
      "logps/pi_response": -211.3716278076172,
      "logps/ref_response": -158.29629516601562,
      "logps/rejected": -515.0916748046875,
      "loss": 0.587,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.9640060663223267,
      "rewards/margins": 0.3828134536743164,
      "rewards/rejected": -1.346819281578064,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.143021821975708,
      "logits/rejected": -2.112715244293213,
      "logps/chosen": -473.237060546875,
      "logps/pi_response": -223.373291015625,
      "logps/ref_response": -170.43218994140625,
      "logps/rejected": -551.0819702148438,
      "loss": 0.5639,
      "rewards/accuracies": 0.7250000238418579,
      "rewards/chosen": -0.902746856212616,
      "rewards/margins": 0.5085975527763367,
      "rewards/rejected": -1.4113445281982422,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -2.122931957244873,
      "logits/rejected": -2.040102005004883,
      "logps/chosen": -467.83624267578125,
      "logps/pi_response": -226.48428344726562,
      "logps/ref_response": -161.48255920410156,
      "logps/rejected": -555.9984130859375,
      "loss": 0.5537,
      "rewards/accuracies": 0.75,
      "rewards/chosen": -1.0724204778671265,
      "rewards/margins": 0.5464988946914673,
      "rewards/rejected": -1.6189196109771729,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -2.083897113800049,
      "logits/rejected": -2.0324301719665527,
      "logps/chosen": -473.509765625,
      "logps/pi_response": -213.6132354736328,
      "logps/ref_response": -152.49720764160156,
      "logps/rejected": -556.308837890625,
      "loss": 0.5598,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -1.0970165729522705,
      "rewards/margins": 0.49258318543434143,
      "rewards/rejected": -1.589599609375,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -2.1261420249938965,
      "logits/rejected": -2.0452873706817627,
      "logps/chosen": -508.80181884765625,
      "logps/pi_response": -223.92135620117188,
      "logps/ref_response": -164.4584197998047,
      "logps/rejected": -586.2526245117188,
      "loss": 0.5684,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.1654565334320068,
      "rewards/margins": 0.5197057127952576,
      "rewards/rejected": -1.6851623058319092,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -2.092771291732788,
      "logits/rejected": -2.080569267272949,
      "logps/chosen": -487.6363830566406,
      "logps/pi_response": -221.43698120117188,
      "logps/ref_response": -164.31655883789062,
      "logps/rejected": -564.31494140625,
      "loss": 0.5698,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -1.0361489057540894,
      "rewards/margins": 0.5097292065620422,
      "rewards/rejected": -1.5458780527114868,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.5976448940629718,
      "train_runtime": 3136.0705,
      "train_samples_per_second": 4.873,
      "train_steps_per_second": 0.038
    }
  ],
  "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
}