{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9978431433840766,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.001000281329123816,
      "grad_norm": 77.5,
      "learning_rate": 6.666666666666668e-08,
      "loss": 2.1336,
      "step": 1
    },
    {
      "epoch": 0.0500140664561908,
      "grad_norm": 9.0625,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 1.2593,
      "step": 50
    },
    {
      "epoch": 0.1000281329123816,
      "grad_norm": 4.1875,
      "learning_rate": 6.666666666666667e-06,
      "loss": 0.6353,
      "step": 100
    },
    {
      "epoch": 0.1500421993685724,
      "grad_norm": 3.4375,
      "learning_rate": 1e-05,
      "loss": 0.5783,
      "step": 150
    },
    {
      "epoch": 0.2000562658247632,
      "grad_norm": 3.09375,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.5614,
      "step": 200
    },
    {
      "epoch": 0.25007033228095404,
      "grad_norm": 3.421875,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.6185,
      "step": 250
    },
    {
      "epoch": 0.3000843987371448,
      "grad_norm": 7.5,
      "learning_rate": 2e-05,
      "loss": 0.6262,
      "step": 300
    },
    {
      "epoch": 0.35009846519333565,
      "grad_norm": 2.421875,
      "learning_rate": 1.9983043934122208e-05,
      "loss": 0.5878,
      "step": 350
    },
    {
      "epoch": 0.4001125316495264,
      "grad_norm": 3.171875,
      "learning_rate": 1.9932233238122834e-05,
      "loss": 0.575,
      "step": 400
    },
    {
      "epoch": 0.45012659810571726,
      "grad_norm": 2.0,
      "learning_rate": 1.984774022190361e-05,
      "loss": 0.5526,
      "step": 450
    },
    {
      "epoch": 0.5001406645619081,
      "grad_norm": 1.8203125,
      "learning_rate": 1.972985141929439e-05,
      "loss": 0.54,
      "step": 500
    },
    {
      "epoch": 0.5501547310180989,
      "grad_norm": 2.03125,
      "learning_rate": 1.9578966616355823e-05,
      "loss": 0.527,
      "step": 550
    },
    {
      "epoch": 0.6001687974742896,
      "grad_norm": 2.046875,
      "learning_rate": 1.9395597495619634e-05,
      "loss": 0.5229,
      "step": 600
    },
    {
      "epoch": 0.6501828639304804,
      "grad_norm": 1.734375,
      "learning_rate": 1.918036590086405e-05,
      "loss": 0.5062,
      "step": 650
    },
    {
      "epoch": 0.7001969303866713,
      "grad_norm": 1.578125,
      "learning_rate": 1.8934001728309003e-05,
      "loss": 0.5055,
      "step": 700
    },
    {
      "epoch": 0.7502109968428621,
      "grad_norm": 1.7265625,
      "learning_rate": 1.865734045138245e-05,
      "loss": 0.4947,
      "step": 750
    },
    {
      "epoch": 0.8002250632990529,
      "grad_norm": 1.5390625,
      "learning_rate": 1.8351320287451865e-05,
      "loss": 0.491,
      "step": 800
    },
    {
      "epoch": 0.8502391297552436,
      "grad_norm": 1.4765625,
      "learning_rate": 1.8016979016129164e-05,
      "loss": 0.4824,
      "step": 850
    },
    {
      "epoch": 0.9002531962114345,
      "grad_norm": 1.53125,
      "learning_rate": 1.7655450459938786e-05,
      "loss": 0.4738,
      "step": 900
    },
    {
      "epoch": 0.9502672626676253,
      "grad_norm": 1.65625,
      "learning_rate": 1.726796063928382e-05,
      "loss": 0.4676,
      "step": 950
    },
    {
      "epoch": 1.0002813291238162,
      "grad_norm": 1.40625,
      "learning_rate": 1.6855823614749474e-05,
      "loss": 0.4657,
      "step": 1000
    },
    {
      "epoch": 1.050295395580007,
      "grad_norm": 1.40625,
      "learning_rate": 1.6420437030843482e-05,
      "loss": 0.3223,
      "step": 1050
    },
    {
      "epoch": 1.1003094620361977,
      "grad_norm": 1.3359375,
      "learning_rate": 1.5963277376285646e-05,
      "loss": 0.3197,
      "step": 1100
    },
    {
      "epoch": 1.1503235284923885,
      "grad_norm": 1.2890625,
      "learning_rate": 1.5485894976919836e-05,
      "loss": 0.3246,
      "step": 1150
    },
    {
      "epoch": 1.2003375949485793,
      "grad_norm": 1.421875,
      "learning_rate": 1.4989908738228567e-05,
      "loss": 0.3167,
      "step": 1200
    },
    {
      "epoch": 1.25035166140477,
      "grad_norm": 1.3828125,
      "learning_rate": 1.4477000655279376e-05,
      "loss": 0.3186,
      "step": 1250
    },
    {
      "epoch": 1.3003657278609608,
      "grad_norm": 1.3515625,
      "learning_rate": 1.394891010872102e-05,
      "loss": 0.3149,
      "step": 1300
    },
    {
      "epoch": 1.3503797943171518,
      "grad_norm": 1.3828125,
      "learning_rate": 1.3407427966172866e-05,
      "loss": 0.3164,
      "step": 1350
    },
    {
      "epoch": 1.4003938607733426,
      "grad_norm": 1.34375,
      "learning_rate": 1.2854390509011061e-05,
      "loss": 0.313,
      "step": 1400
    },
    {
      "epoch": 1.4504079272295334,
      "grad_norm": 1.390625,
      "learning_rate": 1.2291673205146908e-05,
      "loss": 0.3071,
      "step": 1450
    },
    {
      "epoch": 1.5004219936857242,
      "grad_norm": 1.328125,
      "learning_rate": 1.1721184348915384e-05,
      "loss": 0.3063,
      "step": 1500
    },
    {
      "epoch": 1.550436060141915,
      "grad_norm": 1.2890625,
      "learning_rate": 1.1144858589642251e-05,
      "loss": 0.3022,
      "step": 1550
    },
    {
      "epoch": 1.6004501265981057,
      "grad_norm": 1.3359375,
      "learning_rate": 1.0564650370835772e-05,
      "loss": 0.3006,
      "step": 1600
    },
    {
      "epoch": 1.6504641930542965,
      "grad_norm": 1.3359375,
      "learning_rate": 9.982527302252135e-06,
      "loss": 0.2998,
      "step": 1650
    },
    {
      "epoch": 1.7004782595104873,
      "grad_norm": 1.359375,
      "learning_rate": 9.40046348731131e-06,
      "loss": 0.2947,
      "step": 1700
    },
    {
      "epoch": 1.750492325966678,
      "grad_norm": 1.34375,
      "learning_rate": 8.820432828491542e-06,
      "loss": 0.294,
      "step": 1750
    },
    {
      "epoch": 1.8005063924228688,
      "grad_norm": 1.375,
      "learning_rate": 8.244402333405252e-06,
      "loss": 0.2894,
      "step": 1800
    },
    {
      "epoch": 1.8505204588790596,
      "grad_norm": 1.375,
      "learning_rate": 7.674325444256899e-06,
      "loss": 0.2879,
      "step": 1850
    },
    {
      "epoch": 1.9005345253352506,
      "grad_norm": 1.3671875,
      "learning_rate": 7.112135413304042e-06,
      "loss": 0.2842,
      "step": 1900
    },
    {
      "epoch": 1.9505485917914414,
      "grad_norm": 1.3125,
      "learning_rate": 6.55973874678682e-06,
      "loss": 0.2839,
      "step": 1950
    },
    {
      "epoch": 2.0005626582476324,
      "grad_norm": 1.1171875,
      "learning_rate": 6.0190087395588596e-06,
      "loss": 0.2766,
      "step": 2000
    },
    {
      "epoch": 2.050576724703823,
      "grad_norm": 1.3046875,
      "learning_rate": 5.491779122345093e-06,
      "loss": 0.1517,
      "step": 2050
    },
    {
      "epoch": 2.100590791160014,
      "grad_norm": 1.1953125,
      "learning_rate": 4.979837843169959e-06,
      "loss": 0.1492,
      "step": 2100
    },
    {
      "epoch": 2.1506048576162047,
      "grad_norm": 1.203125,
      "learning_rate": 4.484921004044509e-06,
      "loss": 0.1494,
      "step": 2150
    },
    {
      "epoch": 2.2006189240723955,
      "grad_norm": 1.1484375,
      "learning_rate": 4.008706973474391e-06,
      "loss": 0.1498,
      "step": 2200
    },
    {
      "epoch": 2.2506329905285862,
      "grad_norm": 1.1328125,
      "learning_rate": 3.5528106947544626e-06,
      "loss": 0.1482,
      "step": 2250
    },
    {
      "epoch": 2.300647056984777,
      "grad_norm": 1.1796875,
      "learning_rate": 3.118778209351808e-06,
      "loss": 0.1477,
      "step": 2300
    },
    {
      "epoch": 2.350661123440968,
      "grad_norm": 1.234375,
      "learning_rate": 2.7080814139495402e-06,
      "loss": 0.1473,
      "step": 2350
    },
    {
      "epoch": 2.4006751898971586,
      "grad_norm": 1.2421875,
      "learning_rate": 2.322113068931391e-06,
      "loss": 0.147,
      "step": 2400
    },
    {
      "epoch": 2.4506892563533493,
      "grad_norm": 1.1875,
      "learning_rate": 1.9621820752343324e-06,
      "loss": 0.1466,
      "step": 2450
    },
    {
      "epoch": 2.50070332280954,
      "grad_norm": 1.21875,
      "learning_rate": 1.629509035586484e-06,
      "loss": 0.145,
      "step": 2500
    },
    {
      "epoch": 2.550717389265731,
      "grad_norm": 1.4140625,
      "learning_rate": 1.3252221151830513e-06,
      "loss": 0.146,
      "step": 2550
    },
    {
      "epoch": 2.6007314557219217,
      "grad_norm": 1.15625,
      "learning_rate": 1.0503532158376584e-06,
      "loss": 0.1453,
      "step": 2600
    },
    {
      "epoch": 2.6507455221781124,
      "grad_norm": 1.203125,
      "learning_rate": 8.058344765833171e-07,
      "loss": 0.1466,
      "step": 2650
    },
    {
      "epoch": 2.7007595886343037,
      "grad_norm": 1.2734375,
      "learning_rate": 5.924951125902545e-07,
      "loss": 0.1458,
      "step": 2700
    },
    {
      "epoch": 2.7507736550904944,
      "grad_norm": 1.1640625,
      "learning_rate": 4.11058603120511e-07,
      "loss": 0.1445,
      "step": 2750
    },
    {
      "epoch": 2.800787721546685,
      "grad_norm": 1.1875,
      "learning_rate": 2.6214023805552826e-07,
      "loss": 0.1449,
      "step": 2800
    },
    {
      "epoch": 2.850801788002876,
      "grad_norm": 1.390625,
      "learning_rate": 1.462450313169983e-07,
      "loss": 0.1457,
      "step": 2850
    },
    {
      "epoch": 2.9008158544590668,
      "grad_norm": 1.2265625,
      "learning_rate": 6.376600825699463e-08,
      "loss": 0.1444,
      "step": 2900
    },
    {
      "epoch": 2.9508299209152575,
      "grad_norm": 1.171875,
      "learning_rate": 1.49828728252277e-08,
      "loss": 0.1451,
      "step": 2950
    },
    {
      "epoch": 2.9978431433840766,
      "step": 2997,
      "total_flos": 1.9278929080237425e+18,
      "train_loss": 0.3418351112304626,
      "train_runtime": 31635.6273,
      "train_samples_per_second": 6.067,
      "train_steps_per_second": 0.095
    }
  ],
  "logging_steps": 50,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 1.9278929080237425e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}