{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9962073324905183,
  "eval_steps": 100,
  "global_step": 197,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2.5000000000000004e-07,
      "logits/chosen": -2.867753744125366,
      "logits/rejected": -3.003779411315918,
      "logps/chosen": -502.99774169921875,
      "logps/rejected": -334.49993896484375,
      "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": 2.5e-06,
      "logits/chosen": -2.9846487045288086,
      "logits/rejected": -3.0769829750061035,
      "logps/chosen": -746.551025390625,
      "logps/rejected": -522.1072998046875,
      "loss": 0.6885,
      "rewards/accuracies": 0.7222222089767456,
      "rewards/chosen": 0.011263555847108364,
      "rewards/margins": 0.009397478774189949,
      "rewards/rejected": 0.0018660753266885877,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 5e-06,
      "logits/chosen": -2.9451098442077637,
      "logits/rejected": -3.023451805114746,
      "logps/chosen": -828.2189331054688,
      "logps/rejected": -592.4244384765625,
      "loss": 0.6355,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.11439962685108185,
      "rewards/margins": 0.1205785870552063,
      "rewards/rejected": -0.006178960669785738,
      "step": 20
    },
    {
      "epoch": 0.15,
      "learning_rate": 4.960724403357315e-06,
      "logits/chosen": -2.925408124923706,
      "logits/rejected": -2.9945900440216064,
      "logps/chosen": -789.0283203125,
      "logps/rejected": -608.7799072265625,
      "loss": 0.5514,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.2579498291015625,
      "rewards/margins": 0.31374359130859375,
      "rewards/rejected": -0.055793724954128265,
      "step": 30
    },
    {
      "epoch": 0.2,
      "learning_rate": 4.84413167142257e-06,
      "logits/chosen": -2.9325766563415527,
      "logits/rejected": -2.995227813720703,
      "logps/chosen": -765.6848754882812,
      "logps/rejected": -547.8126220703125,
      "loss": 0.4868,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": 0.3068661689758301,
      "rewards/margins": 0.48913106322288513,
      "rewards/rejected": -0.18226495385169983,
      "step": 40
    },
    {
      "epoch": 0.25,
      "learning_rate": 4.653885203484516e-06,
      "logits/chosen": -2.9359793663024902,
      "logits/rejected": -3.0224156379699707,
      "logps/chosen": -789.9358520507812,
      "logps/rejected": -607.2333984375,
      "loss": 0.4211,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.32125765085220337,
      "rewards/margins": 0.6743311285972595,
      "rewards/rejected": -0.35307347774505615,
      "step": 50
    },
    {
      "epoch": 0.3,
      "learning_rate": 4.3959626343730965e-06,
      "logits/chosen": -2.9203407764434814,
      "logits/rejected": -2.9182143211364746,
      "logps/chosen": -825.3511962890625,
      "logps/rejected": -596.4517211914062,
      "loss": 0.3456,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.39548319578170776,
      "rewards/margins": 0.9407623410224915,
      "rewards/rejected": -0.5452791452407837,
      "step": 60
    },
    {
      "epoch": 0.35,
      "learning_rate": 4.078468014319884e-06,
      "logits/chosen": -2.883965015411377,
      "logits/rejected": -2.8747031688690186,
      "logps/chosen": -732.6103515625,
      "logps/rejected": -626.3192138671875,
      "loss": 0.3458,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.10843221098184586,
      "rewards/margins": 0.9775202870368958,
      "rewards/rejected": -0.8690881729125977,
      "step": 70
    },
    {
      "epoch": 0.4,
      "learning_rate": 3.711377175831626e-06,
      "logits/chosen": -2.9415102005004883,
      "logits/rejected": -2.9699349403381348,
      "logps/chosen": -809.07568359375,
      "logps/rejected": -704.0150756835938,
      "loss": 0.2466,
      "rewards/accuracies": 1.0,
      "rewards/chosen": 0.03153586387634277,
      "rewards/margins": 1.388339877128601,
      "rewards/rejected": -1.3568041324615479,
      "step": 80
    },
    {
      "epoch": 0.46,
      "learning_rate": 3.3062242882712727e-06,
      "logits/chosen": -2.8811752796173096,
      "logits/rejected": -2.9307639598846436,
      "logps/chosen": -868.4812622070312,
      "logps/rejected": -745.4024658203125,
      "loss": 0.2185,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -0.1719144582748413,
      "rewards/margins": 1.6975847482681274,
      "rewards/rejected": -1.8694992065429688,
      "step": 90
    },
    {
      "epoch": 0.51,
      "learning_rate": 2.8757394487511766e-06,
      "logits/chosen": -2.8607354164123535,
      "logits/rejected": -2.861823558807373,
      "logps/chosen": -892.8031005859375,
      "logps/rejected": -857.4615478515625,
      "loss": 0.1963,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -0.9989568591117859,
      "rewards/margins": 1.9792629480361938,
      "rewards/rejected": -2.978219509124756,
      "step": 100
    },
    {
      "epoch": 0.51,
      "eval_logits/chosen": -2.8255772590637207,
      "eval_logits/rejected": -2.8398427963256836,
      "eval_logps/chosen": -906.914306640625,
      "eval_logps/rejected": -915.0069580078125,
      "eval_loss": 0.1327642947435379,
      "eval_rewards/accuracies": 1.0,
      "eval_rewards/chosen": -1.0971715450286865,
      "eval_rewards/margins": 2.45654034614563,
      "eval_rewards/rejected": -3.5537116527557373,
      "eval_runtime": 138.8088,
      "eval_samples_per_second": 1.426,
      "eval_steps_per_second": 1.426,
      "step": 100
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.4334486964055635e-06,
      "logits/chosen": -2.7996840476989746,
      "logits/rejected": -2.765969753265381,
      "logps/chosen": -926.7333984375,
      "logps/rejected": -968.1886596679688,
      "loss": 0.1226,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -1.7008222341537476,
      "rewards/margins": 2.827996015548706,
      "rewards/rejected": -4.528818130493164,
      "step": 110
    },
    {
      "epoch": 0.61,
      "learning_rate": 1.9932490177847664e-06,
      "logits/chosen": -2.7951323986053467,
      "logits/rejected": -2.793102741241455,
      "logps/chosen": -1156.3507080078125,
      "logps/rejected": -1393.317626953125,
      "loss": 0.0599,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -3.1200733184814453,
      "rewards/margins": 4.863899230957031,
      "rewards/rejected": -7.983973026275635,
      "step": 120
    },
    {
      "epoch": 0.66,
      "learning_rate": 1.5689716969045848e-06,
      "logits/chosen": -2.722018003463745,
      "logits/rejected": -2.7178285121917725,
      "logps/chosen": -1150.283203125,
      "logps/rejected": -1416.4925537109375,
      "loss": 0.1767,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -4.09153413772583,
      "rewards/margins": 4.929880142211914,
      "rewards/rejected": -9.021414756774902,
      "step": 130
    },
    {
      "epoch": 0.71,
      "learning_rate": 1.1739477297006442e-06,
      "logits/chosen": -2.7221176624298096,
      "logits/rejected": -2.6924355030059814,
      "logps/chosen": -1250.833984375,
      "logps/rejected": -1684.3460693359375,
      "loss": 0.0761,
      "rewards/accuracies": 0.9750000238418579,
      "rewards/chosen": -4.285839080810547,
      "rewards/margins": 6.7211456298828125,
      "rewards/rejected": -11.00698471069336,
      "step": 140
    },
    {
      "epoch": 0.76,
      "learning_rate": 8.20588957773018e-07,
      "logits/chosen": -2.6774404048919678,
      "logits/rejected": -2.671726942062378,
      "logps/chosen": -1310.638427734375,
      "logps/rejected": -1781.271240234375,
      "loss": 0.0435,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -5.060001850128174,
      "rewards/margins": 7.21289587020874,
      "rewards/rejected": -12.272897720336914,
      "step": 150
    },
    {
      "epoch": 0.81,
      "learning_rate": 5.199980823988158e-07,
      "logits/chosen": -2.670541286468506,
      "logits/rejected": -2.615200996398926,
      "logps/chosen": -1342.9185791015625,
      "logps/rejected": -1865.0504150390625,
      "loss": 0.1071,
      "rewards/accuracies": 0.949999988079071,
      "rewards/chosen": -5.760629653930664,
      "rewards/margins": 7.400970458984375,
      "rewards/rejected": -13.161600112915039,
      "step": 160
    },
    {
      "epoch": 0.86,
      "learning_rate": 2.8161981235857145e-07,
      "logits/chosen": -2.6611950397491455,
      "logits/rejected": -2.6389710903167725,
      "logps/chosen": -1471.225341796875,
      "logps/rejected": -1975.236083984375,
      "loss": 0.0247,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -6.465508460998535,
      "rewards/margins": 7.263296604156494,
      "rewards/rejected": -13.728803634643555,
      "step": 170
    },
    {
      "epoch": 0.91,
      "learning_rate": 1.1294410667827022e-07,
      "logits/chosen": -2.6490066051483154,
      "logits/rejected": -2.6506030559539795,
      "logps/chosen": -1516.554443359375,
      "logps/rejected": -2156.269287109375,
      "loss": 0.024,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -7.031141757965088,
      "rewards/margins": 8.472519874572754,
      "rewards/rejected": -15.503662109375,
      "step": 180
    },
    {
      "epoch": 0.96,
      "learning_rate": 1.9270836541688543e-08,
      "logits/chosen": -2.681256055831909,
      "logits/rejected": -2.657177209854126,
      "logps/chosen": -1404.041015625,
      "logps/rejected": -1850.4908447265625,
      "loss": 0.0397,
      "rewards/accuracies": 1.0,
      "rewards/chosen": -6.127170562744141,
      "rewards/margins": 6.76980447769165,
      "rewards/rejected": -12.896974563598633,
      "step": 190
    },
    {
      "epoch": 1.0,
      "step": 197,
      "total_flos": 0.0,
      "train_loss": 0.24469107510474733,
      "train_runtime": 1322.9236,
      "train_samples_per_second": 0.598,
      "train_steps_per_second": 0.149
    }
  ],
  "logging_steps": 10,
  "max_steps": 197,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}