{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9990817263544536,
  "eval_steps": 100,
  "global_step": 204,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2.3809523809523811e-07,
      "logits/chosen": -2.5948691368103027,
      "logits/rejected": -2.452101707458496,
      "logps/chosen": -288.6771240234375,
      "logps/rejected": -270.8803405761719,
      "loss": 0.6931,
      "rewards/accuracies": 0.0,
      "rewards/chosen": 0.0,
      "rewards/margins": 0.0,
      "rewards/rejected": 0.0,
      "step": 1
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.380952380952381e-06,
      "logits/chosen": -2.4060661792755127,
      "logits/rejected": -2.3895983695983887,
      "logps/chosen": -248.78944396972656,
      "logps/rejected": -272.0387878417969,
      "loss": 0.6902,
      "rewards/accuracies": 0.5138888955116272,
      "rewards/chosen": -0.012765132822096348,
      "rewards/margins": 0.0059960586950182915,
      "rewards/rejected": -0.01876119151711464,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 4.761904761904762e-06,
      "logits/chosen": -2.352856159210205,
      "logits/rejected": -2.2108333110809326,
      "logps/chosen": -330.4653625488281,
      "logps/rejected": -309.5657043457031,
      "loss": 0.6543,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.05031045526266098,
      "rewards/margins": 0.09622685611248016,
      "rewards/rejected": -0.14653730392456055,
      "step": 20
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.970219740227693e-06,
      "logits/chosen": -1.890295386314392,
      "logits/rejected": -1.7602672576904297,
      "logps/chosen": -304.2027282714844,
      "logps/rejected": -310.0560607910156,
      "loss": 0.6122,
      "rewards/accuracies": 0.6656249761581421,
      "rewards/chosen": -0.05764678865671158,
      "rewards/margins": 0.20967629551887512,
      "rewards/rejected": -0.2673230767250061,
      "step": 30
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.868186180746792e-06,
      "logits/chosen": -1.6625322103500366,
      "logits/rejected": -1.408319115638733,
      "logps/chosen": -332.21380615234375,
      "logps/rejected": -359.38226318359375,
      "loss": 0.5921,
      "rewards/accuracies": 0.706250011920929,
      "rewards/chosen": -0.0873696357011795,
      "rewards/margins": 0.3436659276485443,
      "rewards/rejected": -0.431035578250885,
      "step": 40
    },
    {
      "epoch": 0.24,
      "learning_rate": 4.696530612642871e-06,
      "logits/chosen": -1.2149860858917236,
      "logits/rejected": -0.9494598507881165,
      "logps/chosen": -344.01007080078125,
      "logps/rejected": -387.1136779785156,
      "loss": 0.5646,
      "rewards/accuracies": 0.746874988079071,
      "rewards/chosen": -0.3123381733894348,
      "rewards/margins": 0.41091781854629517,
      "rewards/rejected": -0.72325599193573,
      "step": 50
    },
    {
      "epoch": 0.29,
      "learning_rate": 4.460299516441777e-06,
      "logits/chosen": -1.3099194765090942,
      "logits/rejected": -1.0444400310516357,
      "logps/chosen": -308.3590393066406,
      "logps/rejected": -356.48309326171875,
      "loss": 0.5712,
      "rewards/accuracies": 0.7093750238418579,
      "rewards/chosen": -0.07990916073322296,
      "rewards/margins": 0.42140746116638184,
      "rewards/rejected": -0.5013166666030884,
      "step": 60
    },
    {
      "epoch": 0.34,
      "learning_rate": 4.1664378205239085e-06,
      "logits/chosen": -1.1891834735870361,
      "logits/rejected": -0.8295331001281738,
      "logps/chosen": -328.58770751953125,
      "logps/rejected": -367.3239440917969,
      "loss": 0.565,
      "rewards/accuracies": 0.659375011920929,
      "rewards/chosen": -0.2833566665649414,
      "rewards/margins": 0.43687087297439575,
      "rewards/rejected": -0.7202275395393372,
      "step": 70
    },
    {
      "epoch": 0.39,
      "learning_rate": 3.8235847280454626e-06,
      "logits/chosen": -1.118951439857483,
      "logits/rejected": -0.7426460385322571,
      "logps/chosen": -330.37640380859375,
      "logps/rejected": -368.56915283203125,
      "loss": 0.5259,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.24417515099048615,
      "rewards/margins": 0.4794433116912842,
      "rewards/rejected": -0.7236183881759644,
      "step": 80
    },
    {
      "epoch": 0.44,
      "learning_rate": 3.441819734087963e-06,
      "logits/chosen": -1.1813547611236572,
      "logits/rejected": -0.7541359663009644,
      "logps/chosen": -337.6168518066406,
      "logps/rejected": -367.4094543457031,
      "loss": 0.5487,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.11910835653543472,
      "rewards/margins": 0.5179867744445801,
      "rewards/rejected": -0.6370951533317566,
      "step": 90
    },
    {
      "epoch": 0.49,
      "learning_rate": 3.0323662998460396e-06,
      "logits/chosen": -1.077327013015747,
      "logits/rejected": -0.598007082939148,
      "logps/chosen": -346.1620788574219,
      "logps/rejected": -371.0578918457031,
      "loss": 0.5347,
      "rewards/accuracies": 0.7124999761581421,
      "rewards/chosen": -0.2570808529853821,
      "rewards/margins": 0.4765000343322754,
      "rewards/rejected": -0.7335808873176575,
      "step": 100
    },
    {
      "epoch": 0.49,
      "eval_logits/chosen": -0.9797883033752441,
      "eval_logits/rejected": -0.5705389380455017,
      "eval_logps/chosen": -348.43701171875,
      "eval_logps/rejected": -380.08721923828125,
      "eval_loss": 0.5570356845855713,
      "eval_rewards/accuracies": 0.7120000123977661,
      "eval_rewards/chosen": -0.3268292248249054,
      "eval_rewards/margins": 0.49906718730926514,
      "eval_rewards/rejected": -0.8258963227272034,
      "eval_runtime": 384.1665,
      "eval_samples_per_second": 5.206,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.54,
      "learning_rate": 2.6072618954988867e-06,
      "logits/chosen": -1.0047966241836548,
      "logits/rejected": -0.5113744735717773,
      "logps/chosen": -352.7798767089844,
      "logps/rejected": -395.8839416503906,
      "loss": 0.5652,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.37514516711235046,
      "rewards/margins": 0.49685588479042053,
      "rewards/rejected": -0.872001051902771,
      "step": 110
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1790041121336223e-06,
      "logits/chosen": -1.1386340856552124,
      "logits/rejected": -0.762003481388092,
      "logps/chosen": -347.37567138671875,
      "logps/rejected": -381.8182373046875,
      "loss": 0.5298,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -0.32593947649002075,
      "rewards/margins": 0.47125840187072754,
      "rewards/rejected": -0.7971979379653931,
      "step": 120
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.760183246631777e-06,
      "logits/chosen": -0.9937411546707153,
      "logits/rejected": -0.5911135673522949,
      "logps/chosen": -341.10308837890625,
      "logps/rejected": -390.31622314453125,
      "loss": 0.5382,
      "rewards/accuracies": 0.731249988079071,
      "rewards/chosen": -0.22077293694019318,
      "rewards/margins": 0.6057422757148743,
      "rewards/rejected": -0.8265151977539062,
      "step": 130
    },
    {
      "epoch": 0.69,
      "learning_rate": 1.3631121611097364e-06,
      "logits/chosen": -0.8524423837661743,
      "logits/rejected": -0.5514906048774719,
      "logps/chosen": -347.8735046386719,
      "logps/rejected": -380.4819641113281,
      "loss": 0.5419,
      "rewards/accuracies": 0.703125,
      "rewards/chosen": -0.3039913773536682,
      "rewards/margins": 0.49449166655540466,
      "rewards/rejected": -0.7984830141067505,
      "step": 140
    },
    {
      "epoch": 0.73,
      "learning_rate": 9.994642986290797e-07,
      "logits/chosen": -0.9539240002632141,
      "logits/rejected": -0.48703765869140625,
      "logps/chosen": -338.36041259765625,
      "logps/rejected": -382.77618408203125,
      "loss": 0.5449,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.3446277379989624,
      "rewards/margins": 0.566214382648468,
      "rewards/rejected": -0.9108421206474304,
      "step": 150
    },
    {
      "epoch": 0.78,
      "learning_rate": 6.799304971075383e-07,
      "logits/chosen": -0.9565147161483765,
      "logits/rejected": -0.4371515214443207,
      "logps/chosen": -359.60650634765625,
      "logps/rejected": -392.94366455078125,
      "loss": 0.533,
      "rewards/accuracies": 0.746874988079071,
      "rewards/chosen": -0.3234427571296692,
      "rewards/margins": 0.5231602787971497,
      "rewards/rejected": -0.8466030359268188,
      "step": 160
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1390469071538183e-07,
      "logits/chosen": -0.8408013582229614,
      "logits/rejected": -0.5169156789779663,
      "logps/chosen": -344.161376953125,
      "logps/rejected": -401.06927490234375,
      "loss": 0.5348,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.2773435711860657,
      "rewards/margins": 0.7136318683624268,
      "rewards/rejected": -0.9909754991531372,
      "step": 170
    },
    {
      "epoch": 0.88,
      "learning_rate": 2.092077387824884e-07,
      "logits/chosen": -1.0086092948913574,
      "logits/rejected": -0.6254789233207703,
      "logps/chosen": -348.04437255859375,
      "logps/rejected": -398.12921142578125,
      "loss": 0.5478,
      "rewards/accuracies": 0.721875011920929,
      "rewards/chosen": -0.273074209690094,
      "rewards/margins": 0.5786523818969727,
      "rewards/rejected": -0.8517265319824219,
      "step": 180
    },
    {
      "epoch": 0.93,
      "learning_rate": 7.185750133542168e-08,
      "logits/chosen": -0.9045581817626953,
      "logits/rejected": -0.6502401828765869,
      "logps/chosen": -323.7873840332031,
      "logps/rejected": -377.43109130859375,
      "loss": 0.5416,
      "rewards/accuracies": 0.6968749761581421,
      "rewards/chosen": -0.3004080057144165,
      "rewards/margins": 0.5847684741020203,
      "rewards/rejected": -0.8851765394210815,
      "step": 190
    },
    {
      "epoch": 0.98,
      "learning_rate": 5.891920784984184e-09,
      "logits/chosen": -1.004695177078247,
      "logits/rejected": -0.5996636748313904,
      "logps/chosen": -338.21856689453125,
      "logps/rejected": -390.23956298828125,
      "loss": 0.5154,
      "rewards/accuracies": 0.778124988079071,
      "rewards/chosen": -0.263592392206192,
      "rewards/margins": 0.6431443691253662,
      "rewards/rejected": -0.9067367315292358,
      "step": 200
    },
    {
      "epoch": 0.98,
      "eval_logits/chosen": -0.9595882296562195,
      "eval_logits/rejected": -0.5385043621063232,
      "eval_logps/chosen": -346.7320556640625,
      "eval_logps/rejected": -380.200927734375,
      "eval_loss": 0.5470749735832214,
      "eval_rewards/accuracies": 0.7139999866485596,
      "eval_rewards/chosen": -0.3097793161869049,
      "eval_rewards/margins": 0.5172543525695801,
      "eval_rewards/rejected": -0.8270336985588074,
      "eval_runtime": 384.197,
      "eval_samples_per_second": 5.206,
      "eval_steps_per_second": 0.651,
      "step": 200
    },
    {
      "epoch": 1.0,
      "step": 204,
      "total_flos": 0.0,
      "train_loss": 0.5623768936185276,
      "train_runtime": 9619.6269,
      "train_samples_per_second": 2.717,
      "train_steps_per_second": 0.021
    }
  ],
  "logging_steps": 10,
  "max_steps": 204,
  "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
}