{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984301412872841,
  "eval_steps": 100,
  "global_step": 159,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 3.125e-07,
      "logits/chosen": -2.6547884941101074,
      "logits/rejected": -2.5468616485595703,
      "logps/chosen": -285.215576171875,
      "logps/rejected": -258.8069152832031,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.06,
      "learning_rate": 3.125e-06,
      "logits/chosen": -2.5754165649414062,
      "logits/rejected": -2.5153017044067383,
      "logps/chosen": -295.195556640625,
      "logps/rejected": -293.4971008300781,
      "loss": 0.6929,
      "rewards/accuracies": 0.53125,
      "rewards/chosen": -0.00017905117420013994,
      "rewards/margins": 0.0014612883096560836,
      "rewards/rejected": -0.001640339381992817,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.9903533134293035e-06,
      "logits/chosen": -2.5756356716156006,
      "logits/rejected": -2.5089950561523438,
      "logps/chosen": -291.3115539550781,
      "logps/rejected": -287.61431884765625,
      "loss": 0.6851,
      "rewards/accuracies": 0.628125011920929,
      "rewards/chosen": -0.061528198421001434,
      "rewards/margins": 0.019807469099760056,
      "rewards/rejected": -0.08133566379547119,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.882681251368549e-06,
      "logits/chosen": -2.6055867671966553,
      "logits/rejected": -2.5165815353393555,
      "logps/chosen": -283.285888671875,
      "logps/rejected": -284.13726806640625,
      "loss": 0.6711,
      "rewards/accuracies": 0.65625,
      "rewards/chosen": -0.06080762296915054,
      "rewards/margins": 0.05497027188539505,
      "rewards/rejected": -0.115777887403965,
      "step": 30
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.660472094042121e-06,
      "logits/chosen": -2.5286662578582764,
      "logits/rejected": -2.4570558071136475,
      "logps/chosen": -309.78094482421875,
      "logps/rejected": -316.354736328125,
      "loss": 0.6628,
      "rewards/accuracies": 0.6343749761581421,
      "rewards/chosen": -0.09531812369823456,
      "rewards/margins": 0.07932458072900772,
      "rewards/rejected": -0.17464269697666168,
      "step": 40
    },
    {
      "epoch": 0.31,
      "learning_rate": 4.33440758555951e-06,
      "logits/chosen": -2.564868211746216,
      "logits/rejected": -2.4932563304901123,
      "logps/chosen": -324.691162109375,
      "logps/rejected": -325.16461181640625,
      "loss": 0.6505,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.2168252170085907,
      "rewards/margins": 0.10023742914199829,
      "rewards/rejected": -0.317062646150589,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.92016186682789e-06,
      "logits/chosen": -2.571204423904419,
      "logits/rejected": -2.4519267082214355,
      "logps/chosen": -334.2891845703125,
      "logps/rejected": -332.8462219238281,
      "loss": 0.6419,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.19313496351242065,
      "rewards/margins": 0.12622517347335815,
      "rewards/rejected": -0.3193601667881012,
      "step": 60
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.437648009023905e-06,
      "logits/chosen": -2.497055768966675,
      "logits/rejected": -2.415569543838501,
      "logps/chosen": -311.2139587402344,
      "logps/rejected": -327.3946838378906,
      "loss": 0.6447,
      "rewards/accuracies": 0.6812499761581421,
      "rewards/chosen": -0.23866800963878632,
      "rewards/margins": 0.13377706706523895,
      "rewards/rejected": -0.37244507670402527,
      "step": 70
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9100607788275547e-06,
      "logits/chosen": -2.4995005130767822,
      "logits/rejected": -2.3666462898254395,
      "logps/chosen": -337.9864196777344,
      "logps/rejected": -344.3404541015625,
      "loss": 0.6348,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.33487486839294434,
      "rewards/margins": 0.14860984683036804,
      "rewards/rejected": -0.48348474502563477,
      "step": 80
    },
    {
      "epoch": 0.57,
      "learning_rate": 2.3627616503391813e-06,
      "logits/chosen": -2.476839065551758,
      "logits/rejected": -2.4053282737731934,
      "logps/chosen": -312.2710266113281,
      "logps/rejected": -342.49163818359375,
      "loss": 0.6268,
      "rewards/accuracies": 0.671875,
      "rewards/chosen": -0.27689170837402344,
      "rewards/margins": 0.19764092564582825,
      "rewards/rejected": -0.4745326042175293,
      "step": 90
    },
    {
      "epoch": 0.63,
      "learning_rate": 1.8220596619089576e-06,
      "logits/chosen": -2.4677796363830566,
      "logits/rejected": -2.409015655517578,
      "logps/chosen": -310.72576904296875,
      "logps/rejected": -347.15087890625,
      "loss": 0.6262,
      "rewards/accuracies": 0.653124988079071,
      "rewards/chosen": -0.29292207956314087,
      "rewards/margins": 0.17808273434638977,
      "rewards/rejected": -0.47100481390953064,
      "step": 100
    },
    {
      "epoch": 0.63,
      "eval_logits/chosen": -2.482710123062134,
      "eval_logits/rejected": -2.386967658996582,
      "eval_logps/chosen": -331.1879577636719,
      "eval_logps/rejected": -340.92962646484375,
      "eval_loss": 0.6278669238090515,
      "eval_rewards/accuracies": 0.6740000247955322,
      "eval_rewards/chosen": -0.3175167143344879,
      "eval_rewards/margins": 0.1821039915084839,
      "eval_rewards/rejected": -0.4996207058429718,
      "eval_runtime": 384.3163,
      "eval_samples_per_second": 5.204,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3139467229135999e-06,
      "logits/chosen": -2.4531397819519043,
      "logits/rejected": -2.3680739402770996,
      "logps/chosen": -311.6913757324219,
      "logps/rejected": -324.017333984375,
      "loss": 0.6164,
      "rewards/accuracies": 0.690625011920929,
      "rewards/chosen": -0.3056855797767639,
      "rewards/margins": 0.18076516687870026,
      "rewards/rejected": -0.4864506721496582,
      "step": 110
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.628481651367876e-07,
      "logits/chosen": -2.4865543842315674,
      "logits/rejected": -2.437178611755371,
      "logps/chosen": -330.9514465332031,
      "logps/rejected": -378.96832275390625,
      "loss": 0.623,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.35980406403541565,
      "rewards/margins": 0.21202397346496582,
      "rewards/rejected": -0.5718280076980591,
      "step": 120
    },
    {
      "epoch": 0.82,
      "learning_rate": 4.904486005914027e-07,
      "logits/chosen": -2.5110232830047607,
      "logits/rejected": -2.399928569793701,
      "logps/chosen": -355.139404296875,
      "logps/rejected": -360.427490234375,
      "loss": 0.6234,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -0.37438246607780457,
      "rewards/margins": 0.19766977429389954,
      "rewards/rejected": -0.5720522999763489,
      "step": 130
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.1464952759020857e-07,
      "logits/chosen": -2.435955286026001,
      "logits/rejected": -2.3834517002105713,
      "logps/chosen": -297.09466552734375,
      "logps/rejected": -351.209716796875,
      "loss": 0.6268,
      "rewards/accuracies": 0.675000011920929,
      "rewards/chosen": -0.33036094903945923,
      "rewards/margins": 0.1953088939189911,
      "rewards/rejected": -0.5256698131561279,
      "step": 140
    },
    {
      "epoch": 0.94,
      "learning_rate": 4.870879364444109e-08,
      "logits/chosen": -2.497584104537964,
      "logits/rejected": -2.3675169944763184,
      "logps/chosen": -344.45770263671875,
      "logps/rejected": -349.06170654296875,
      "loss": 0.6177,
      "rewards/accuracies": 0.6937500238418579,
      "rewards/chosen": -0.32350295782089233,
      "rewards/margins": 0.21759824454784393,
      "rewards/rejected": -0.5411011576652527,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 159,
      "total_flos": 0.0,
      "train_loss": 0.6422481574352432,
      "train_runtime": 7267.2932,
      "train_samples_per_second": 2.804,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 159,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}