{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 1158,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 1.9137931034482762e-05,
      "logits/chosen": -2.7039783000946045,
      "logits/rejected": -2.699718713760376,
      "logps/chosen": -178.0665283203125,
      "logps/rejected": -202.87210083007812,
      "loss": 0.65,
      "rewards/accuracies": 0.798701286315918,
      "rewards/chosen": -0.09282779693603516,
      "rewards/margins": 0.09910193085670471,
      "rewards/rejected": -0.19192971289157867,
      "step": 77
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.899232245681382e-05,
      "logits/chosen": -2.7582526206970215,
      "logits/rejected": -2.6498587131500244,
      "logps/chosen": -206.910888671875,
      "logps/rejected": -244.54962158203125,
      "loss": 0.3215,
      "rewards/accuracies": 0.9740259647369385,
      "rewards/chosen": -2.6819589138031006,
      "rewards/margins": 1.8207621574401855,
      "rewards/rejected": -4.502721309661865,
      "step": 154
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.6833013435700576e-05,
      "logits/chosen": -3.261249542236328,
      "logits/rejected": -3.26186203956604,
      "logps/chosen": -636.440673828125,
      "logps/rejected": -844.8958740234375,
      "loss": 0.1384,
      "rewards/accuracies": 0.9675324559211731,
      "rewards/chosen": -45.84988021850586,
      "rewards/margins": 18.366718292236328,
      "rewards/rejected": -64.21659088134766,
      "step": 231
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.4644913627639156e-05,
      "logits/chosen": -3.799668788909912,
      "logits/rejected": -3.877474308013916,
      "logps/chosen": -1188.7479248046875,
      "logps/rejected": -1513.479248046875,
      "loss": 0.1428,
      "rewards/accuracies": 0.9675324559211731,
      "rewards/chosen": -100.75252532958984,
      "rewards/margins": 30.6811580657959,
      "rewards/rejected": -131.4336700439453,
      "step": 308
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.2428023032629558e-05,
      "logits/chosen": -3.48297119140625,
      "logits/rejected": -3.503549098968506,
      "logps/chosen": -1211.92041015625,
      "logps/rejected": -1585.6292724609375,
      "loss": 0.3247,
      "rewards/accuracies": 0.9545454382896423,
      "rewards/chosen": -102.89164733886719,
      "rewards/margins": 35.46805953979492,
      "rewards/rejected": -138.3596954345703,
      "step": 385
    },
    {
      "epoch": 1.2,
      "learning_rate": 2.0211132437619963e-05,
      "logits/chosen": -3.9055564403533936,
      "logits/rejected": -3.94558048248291,
      "logps/chosen": -1156.05908203125,
      "logps/rejected": -1524.2059326171875,
      "loss": 0.2889,
      "rewards/accuracies": 0.9545454382896423,
      "rewards/chosen": -97.4389877319336,
      "rewards/margins": 34.911155700683594,
      "rewards/rejected": -132.3501434326172,
      "step": 462
    },
    {
      "epoch": 1.4,
      "learning_rate": 1.7994241842610365e-05,
      "logits/chosen": -4.063717365264893,
      "logits/rejected": -4.160610675811768,
      "logps/chosen": -1216.2169189453125,
      "logps/rejected": -1579.79931640625,
      "loss": 0.0666,
      "rewards/accuracies": 0.9935064911842346,
      "rewards/chosen": -103.59471130371094,
      "rewards/margins": 34.364356994628906,
      "rewards/rejected": -137.95909118652344,
      "step": 539
    },
    {
      "epoch": 1.6,
      "learning_rate": 1.577735124760077e-05,
      "logits/chosen": -3.7622299194335938,
      "logits/rejected": -3.8654086589813232,
      "logps/chosen": -1256.462158203125,
      "logps/rejected": -1670.1668701171875,
      "loss": 0.3068,
      "rewards/accuracies": 0.9545454382896423,
      "rewards/chosen": -107.87626647949219,
      "rewards/margins": 38.972930908203125,
      "rewards/rejected": -146.8491973876953,
      "step": 616
    },
    {
      "epoch": 1.8,
      "learning_rate": 1.3589251439539348e-05,
      "logits/chosen": -3.9780585765838623,
      "logits/rejected": -4.0323262214660645,
      "logps/chosen": -1205.580322265625,
      "logps/rejected": -1609.4249267578125,
      "loss": 0.0732,
      "rewards/accuracies": 0.9675324559211731,
      "rewards/chosen": -102.83528137207031,
      "rewards/margins": 38.15076446533203,
      "rewards/rejected": -140.9860382080078,
      "step": 693
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.137236084452975e-05,
      "logits/chosen": -3.9996836185455322,
      "logits/rejected": -4.038970947265625,
      "logps/chosen": -1479.3536376953125,
      "logps/rejected": -1951.1041259765625,
      "loss": 0.2627,
      "rewards/accuracies": 0.9935064911842346,
      "rewards/chosen": -129.55506896972656,
      "rewards/margins": 45.21900177001953,
      "rewards/rejected": -174.77407836914062,
      "step": 770
    },
    {
      "epoch": 2.19,
      "learning_rate": 9.155470249520153e-06,
      "logits/chosen": -4.09077262878418,
      "logits/rejected": -4.1275434494018555,
      "logps/chosen": -1435.885986328125,
      "logps/rejected": -1983.1124267578125,
      "loss": 0.0968,
      "rewards/accuracies": 0.9870129823684692,
      "rewards/chosen": -125.91879272460938,
      "rewards/margins": 52.251136779785156,
      "rewards/rejected": -178.16995239257812,
      "step": 847
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.938579654510557e-06,
      "logits/chosen": -4.158607482910156,
      "logits/rejected": -4.210606098175049,
      "logps/chosen": -1408.3914794921875,
      "logps/rejected": -1875.4822998046875,
      "loss": 0.0166,
      "rewards/accuracies": 0.9805194735527039,
      "rewards/chosen": -122.99951934814453,
      "rewards/margins": 44.64362335205078,
      "rewards/rejected": -167.64315795898438,
      "step": 924
    },
    {
      "epoch": 2.59,
      "learning_rate": 4.72168905950096e-06,
      "logits/chosen": -4.183628559112549,
      "logits/rejected": -4.219183444976807,
      "logps/chosen": -1309.217529296875,
      "logps/rejected": -1762.13232421875,
      "loss": 0.0602,
      "rewards/accuracies": 0.9870129823684692,
      "rewards/chosen": -113.17806243896484,
      "rewards/margins": 43.657859802246094,
      "rewards/rejected": -156.83592224121094,
      "step": 1001
    },
    {
      "epoch": 2.79,
      "learning_rate": 2.5335892514395392e-06,
      "logits/chosen": -4.1587114334106445,
      "logits/rejected": -4.193912029266357,
      "logps/chosen": -1347.27978515625,
      "logps/rejected": -1834.71728515625,
      "loss": 0.0365,
      "rewards/accuracies": 0.9935064911842346,
      "rewards/chosen": -116.40560150146484,
      "rewards/margins": 46.630714416503906,
      "rewards/rejected": -163.03631591796875,
      "step": 1078
    },
    {
      "epoch": 2.99,
      "learning_rate": 3.166986564299424e-07,
      "logits/chosen": -4.090827465057373,
      "logits/rejected": -4.138672351837158,
      "logps/chosen": -1456.274169921875,
      "logps/rejected": -1961.9942626953125,
      "loss": 0.0055,
      "rewards/accuracies": 0.9935064911842346,
      "rewards/chosen": -127.70254516601562,
      "rewards/margins": 48.34178924560547,
      "rewards/rejected": -176.04434204101562,
      "step": 1155
    }
  ],
  "logging_steps": 77,
  "max_steps": 1158,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}