{
  "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": -2.7306313514709473,
      "logits/rejected": -2.695523977279663,
      "logps/chosen": -263.6216125488281,
      "logps/pi_response": -105.38675689697266,
      "logps/ref_response": -105.38675689697266,
      "logps/rejected": -313.963623046875,
      "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": -2.7170629501342773,
      "logits/rejected": -2.7163281440734863,
      "logps/chosen": -295.01373291015625,
      "logps/pi_response": -115.70919799804688,
      "logps/ref_response": -115.59099578857422,
      "logps/rejected": -339.944580078125,
      "loss": 0.6908,
      "rewards/accuracies": 0.5138888955116272,
      "rewards/chosen": -0.01301976852118969,
      "rewards/margins": 0.0039041738491505384,
      "rewards/rejected": -0.016923941671848297,
      "step": 10
    },
    {
      "epoch": 0.17,
      "learning_rate": 4.931352528237397e-07,
      "logits/chosen": -2.663550853729248,
      "logits/rejected": -2.6346139907836914,
      "logps/chosen": -324.2618103027344,
      "logps/pi_response": -126.08673095703125,
      "logps/ref_response": -123.05793762207031,
      "logps/rejected": -365.3756408691406,
      "loss": 0.679,
      "rewards/accuracies": 0.643750011920929,
      "rewards/chosen": -0.2619123160839081,
      "rewards/margins": 0.07399449497461319,
      "rewards/rejected": -0.3359068036079407,
      "step": 20
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.658920803689553e-07,
      "logits/chosen": -2.655712604522705,
      "logits/rejected": -2.676952362060547,
      "logps/chosen": -347.55548095703125,
      "logps/pi_response": -107.3725814819336,
      "logps/ref_response": -101.8917236328125,
      "logps/rejected": -391.9983825683594,
      "loss": 0.6451,
      "rewards/accuracies": 0.668749988079071,
      "rewards/chosen": -0.52825528383255,
      "rewards/margins": 0.2661920189857483,
      "rewards/rejected": -0.7944473028182983,
      "step": 30
    },
    {
      "epoch": 0.33,
      "learning_rate": 4.201712553872657e-07,
      "logits/chosen": -2.611100435256958,
      "logits/rejected": -2.588596820831299,
      "logps/chosen": -402.82611083984375,
      "logps/pi_response": -159.3750762939453,
      "logps/ref_response": -122.22212219238281,
      "logps/rejected": -463.3408203125,
      "loss": 0.6197,
      "rewards/accuracies": 0.6312500238418579,
      "rewards/chosen": -0.8930713534355164,
      "rewards/margins": 0.25158679485321045,
      "rewards/rejected": -1.1446583271026611,
      "step": 40
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.598859066780754e-07,
      "logits/chosen": -2.63188099861145,
      "logits/rejected": -2.5909008979797363,
      "logps/chosen": -394.1261291503906,
      "logps/pi_response": -155.15040588378906,
      "logps/ref_response": -117.72813415527344,
      "logps/rejected": -453.97613525390625,
      "loss": 0.5947,
      "rewards/accuracies": 0.7437499761581421,
      "rewards/chosen": -0.8403727412223816,
      "rewards/margins": 0.4999156892299652,
      "rewards/rejected": -1.3402884006500244,
      "step": 50
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.9019570347986706e-07,
      "logits/chosen": -2.502223491668701,
      "logits/rejected": -2.5179829597473145,
      "logps/chosen": -413.60467529296875,
      "logps/pi_response": -163.8712615966797,
      "logps/ref_response": -113.3283920288086,
      "logps/rejected": -485.93072509765625,
      "loss": 0.604,
      "rewards/accuracies": 0.699999988079071,
      "rewards/chosen": -1.0342451333999634,
      "rewards/margins": 0.39558935165405273,
      "rewards/rejected": -1.4298346042633057,
      "step": 60
    },
    {
      "epoch": 0.59,
      "learning_rate": 2.1706525253979534e-07,
      "logits/chosen": -2.4702818393707275,
      "logits/rejected": -2.4437742233276367,
      "logps/chosen": -434.32952880859375,
      "logps/pi_response": -174.0717010498047,
      "logps/ref_response": -115.6038589477539,
      "logps/rejected": -507.83636474609375,
      "loss": 0.5892,
      "rewards/accuracies": 0.637499988079071,
      "rewards/chosen": -1.1909167766571045,
      "rewards/margins": 0.3753449320793152,
      "rewards/rejected": -1.566261887550354,
      "step": 70
    },
    {
      "epoch": 0.67,
      "learning_rate": 1.4675360263490295e-07,
      "logits/chosen": -2.398507833480835,
      "logits/rejected": -2.353964328765869,
      "logps/chosen": -400.9221496582031,
      "logps/pi_response": -167.4496612548828,
      "logps/ref_response": -109.77674865722656,
      "logps/rejected": -465.49139404296875,
      "loss": 0.5563,
      "rewards/accuracies": 0.6625000238418579,
      "rewards/chosen": -1.0598949193954468,
      "rewards/margins": 0.4164988398551941,
      "rewards/rejected": -1.4763939380645752,
      "step": 80
    },
    {
      "epoch": 0.75,
      "learning_rate": 8.527854855097224e-08,
      "logits/chosen": -2.3992695808410645,
      "logits/rejected": -2.3342957496643066,
      "logps/chosen": -404.3464050292969,
      "logps/pi_response": -172.18580627441406,
      "logps/ref_response": -119.97776794433594,
      "logps/rejected": -455.5665588378906,
      "loss": 0.5729,
      "rewards/accuracies": 0.6875,
      "rewards/chosen": -0.9877260327339172,
      "rewards/margins": 0.43665844202041626,
      "rewards/rejected": -1.424384593963623,
      "step": 90
    },
    {
      "epoch": 0.84,
      "learning_rate": 3.790158337517127e-08,
      "logits/chosen": -2.3577961921691895,
      "logits/rejected": -2.350846767425537,
      "logps/chosen": -397.8568115234375,
      "logps/pi_response": -161.6202850341797,
      "logps/ref_response": -115.20768737792969,
      "logps/rejected": -467.29998779296875,
      "loss": 0.5679,
      "rewards/accuracies": 0.737500011920929,
      "rewards/chosen": -0.9567230939865112,
      "rewards/margins": 0.5116481781005859,
      "rewards/rejected": -1.4683713912963867,
      "step": 100
    },
    {
      "epoch": 0.92,
      "learning_rate": 8.677580722139671e-09,
      "logits/chosen": -2.3769965171813965,
      "logits/rejected": -2.3311877250671387,
      "logps/chosen": -392.29876708984375,
      "logps/pi_response": -161.58706665039062,
      "logps/ref_response": -106.44654846191406,
      "logps/rejected": -466.61395263671875,
      "loss": 0.5574,
      "rewards/accuracies": 0.78125,
      "rewards/chosen": -0.9844812154769897,
      "rewards/margins": 0.6008119583129883,
      "rewards/rejected": -1.5852930545806885,
      "step": 110
    },
    {
      "epoch": 1.0,
      "step": 119,
      "total_flos": 0.0,
      "train_loss": 0.6038635838933352,
      "train_runtime": 3119.4458,
      "train_samples_per_second": 4.899,
      "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
}