{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9984,
  "eval_steps": 100,
  "global_step": 156,
  "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": -2.62762188911438,
      "logits/rejected": -2.586808681488037,
      "logps/chosen": -233.10275268554688,
      "logps/rejected": -220.01724243164062,
      "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": -2.640223741531372,
      "logits/rejected": -2.6421992778778076,
      "logps/chosen": -172.71710205078125,
      "logps/rejected": -174.91177368164062,
      "loss": 0.6932,
      "rewards/accuracies": 0.4270833432674408,
      "rewards/chosen": -0.0005670329555869102,
      "rewards/margins": -0.00034490120015107095,
      "rewards/rejected": -0.00022213184274733067,
      "step": 10
    },
    {
      "epoch": 0.13,
      "learning_rate": 4.989935734988098e-06,
      "logits/chosen": -2.6506519317626953,
      "logits/rejected": -2.6606290340423584,
      "logps/chosen": -163.25784301757812,
      "logps/rejected": -166.31251525878906,
      "loss": 0.6923,
      "rewards/accuracies": 0.546875,
      "rewards/chosen": -0.005096795037388802,
      "rewards/margins": 0.002602009102702141,
      "rewards/rejected": -0.007698804140090942,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 4.8776412907378845e-06,
      "logits/chosen": -2.6732003688812256,
      "logits/rejected": -2.6337249279022217,
      "logps/chosen": -179.1455535888672,
      "logps/rejected": -177.36630249023438,
      "loss": 0.6928,
      "rewards/accuracies": 0.5218750238418579,
      "rewards/chosen": -0.07040824741125107,
      "rewards/margins": 0.0008022676920518279,
      "rewards/rejected": -0.07121052592992783,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 4.646121984004666e-06,
      "logits/chosen": -2.592510223388672,
      "logits/rejected": -2.569596767425537,
      "logps/chosen": -176.0773162841797,
      "logps/rejected": -177.98220825195312,
      "loss": 0.6939,
      "rewards/accuracies": 0.5249999761581421,
      "rewards/chosen": -0.06681038439273834,
      "rewards/margins": -0.002174162771552801,
      "rewards/rejected": -0.0646362155675888,
      "step": 40
    },
    {
      "epoch": 0.32,
      "learning_rate": 4.3069871595684795e-06,
      "logits/chosen": -2.6304595470428467,
      "logits/rejected": -2.6132376194000244,
      "logps/chosen": -170.5052947998047,
      "logps/rejected": -170.62728881835938,
      "loss": 0.6925,
      "rewards/accuracies": 0.5218750238418579,
      "rewards/chosen": -0.03540392965078354,
      "rewards/margins": 0.004562810063362122,
      "rewards/rejected": -0.03996673971414566,
      "step": 50
    },
    {
      "epoch": 0.38,
      "learning_rate": 3.8772424536302565e-06,
      "logits/chosen": -2.5987067222595215,
      "logits/rejected": -2.573859691619873,
      "logps/chosen": -180.9357452392578,
      "logps/rejected": -177.1747589111328,
      "loss": 0.6923,
      "rewards/accuracies": 0.48750001192092896,
      "rewards/chosen": -0.09346936643123627,
      "rewards/margins": 0.0016549427527934313,
      "rewards/rejected": -0.09512430429458618,
      "step": 60
    },
    {
      "epoch": 0.45,
      "learning_rate": 3.3784370602033572e-06,
      "logits/chosen": -2.5544121265411377,
      "logits/rejected": -2.570075750350952,
      "logps/chosen": -172.70663452148438,
      "logps/rejected": -173.13766479492188,
      "loss": 0.692,
      "rewards/accuracies": 0.49687498807907104,
      "rewards/chosen": -0.15676796436309814,
      "rewards/margins": -0.002442942699417472,
      "rewards/rejected": -0.15432503819465637,
      "step": 70
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.835583164544139e-06,
      "logits/chosen": -2.604128122329712,
      "logits/rejected": -2.5767688751220703,
      "logps/chosen": -184.4633331298828,
      "logps/rejected": -183.540283203125,
      "loss": 0.6936,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.18969281017780304,
      "rewards/margins": -0.0006604200461879373,
      "rewards/rejected": -0.18903236091136932,
      "step": 80
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.2759017277414165e-06,
      "logits/chosen": -2.661609172821045,
      "logits/rejected": -2.5941507816314697,
      "logps/chosen": -191.8031463623047,
      "logps/rejected": -189.98135375976562,
      "loss": 0.6914,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": -0.08525966852903366,
      "rewards/margins": 0.013879667036235332,
      "rewards/rejected": -0.09913934022188187,
      "step": 90
    },
    {
      "epoch": 0.64,
      "learning_rate": 1.7274575140626318e-06,
      "logits/chosen": -2.6250574588775635,
      "logits/rejected": -2.6189353466033936,
      "logps/chosen": -178.07730102539062,
      "logps/rejected": -179.85862731933594,
      "loss": 0.6902,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.038252077996730804,
      "rewards/margins": 0.017853444442152977,
      "rewards/rejected": -0.05610552430152893,
      "step": 100
    },
    {
      "epoch": 0.64,
      "eval_logits/chosen": -2.555354595184326,
      "eval_logits/rejected": -2.4580423831939697,
      "eval_logps/chosen": -310.5243835449219,
      "eval_logps/rejected": -302.46649169921875,
      "eval_loss": 0.6929731369018555,
      "eval_rewards/accuracies": 0.5139999985694885,
      "eval_rewards/chosen": -0.12465862929821014,
      "eval_rewards/margins": 0.00494543369859457,
      "eval_rewards/rejected": -0.12960407137870789,
      "eval_runtime": 383.981,
      "eval_samples_per_second": 5.209,
      "eval_steps_per_second": 0.651,
      "step": 100
    },
    {
      "epoch": 0.7,
      "learning_rate": 1.217751806485235e-06,
      "logits/chosen": -2.651780128479004,
      "logits/rejected": -2.581507921218872,
      "logps/chosen": -194.07461547851562,
      "logps/rejected": -183.43362426757812,
      "loss": 0.6898,
      "rewards/accuracies": 0.4781250059604645,
      "rewards/chosen": -0.11336056143045425,
      "rewards/margins": 0.0036790375597774982,
      "rewards/rejected": -0.11703959852457047,
      "step": 110
    },
    {
      "epoch": 0.77,
      "learning_rate": 7.723433775328385e-07,
      "logits/chosen": -2.6290206909179688,
      "logits/rejected": -2.6049842834472656,
      "logps/chosen": -186.02259826660156,
      "logps/rejected": -193.3793487548828,
      "loss": 0.6902,
      "rewards/accuracies": 0.534375011920929,
      "rewards/chosen": -0.11839202791452408,
      "rewards/margins": 0.004132995847612619,
      "rewards/rejected": -0.12252502143383026,
      "step": 120
    },
    {
      "epoch": 0.83,
      "learning_rate": 4.1356686569674344e-07,
      "logits/chosen": -2.5702383518218994,
      "logits/rejected": -2.5472493171691895,
      "logps/chosen": -169.70834350585938,
      "logps/rejected": -174.14761352539062,
      "loss": 0.6891,
      "rewards/accuracies": 0.5562499761581421,
      "rewards/chosen": -0.09281644970178604,
      "rewards/margins": 0.012840485200285912,
      "rewards/rejected": -0.1056569367647171,
      "step": 130
    },
    {
      "epoch": 0.9,
      "learning_rate": 1.59412823400657e-07,
      "logits/chosen": -2.5276107788085938,
      "logits/rejected": -2.5228161811828613,
      "logps/chosen": -174.78802490234375,
      "logps/rejected": -176.2909698486328,
      "loss": 0.6878,
      "rewards/accuracies": 0.5625,
      "rewards/chosen": -0.05636407807469368,
      "rewards/margins": 0.011297663673758507,
      "rewards/rejected": -0.06766173988580704,
      "step": 140
    },
    {
      "epoch": 0.96,
      "learning_rate": 2.262559558016325e-08,
      "logits/chosen": -2.568333864212036,
      "logits/rejected": -2.586660623550415,
      "logps/chosen": -166.882080078125,
      "logps/rejected": -171.17140197753906,
      "loss": 0.6892,
      "rewards/accuracies": 0.565625011920929,
      "rewards/chosen": -0.061885736882686615,
      "rewards/margins": 0.011388251557946205,
      "rewards/rejected": -0.07327400147914886,
      "step": 150
    },
    {
      "epoch": 1.0,
      "step": 156,
      "total_flos": 0.0,
      "train_loss": 0.6914212252849188,
      "train_runtime": 7065.3773,
      "train_samples_per_second": 2.831,
      "train_steps_per_second": 0.022
    }
  ],
  "logging_steps": 10,
  "max_steps": 156,
  "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
}