{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.039255671679837,
  "eval_steps": 500,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02549069589599796,
      "grad_norm": 3183.12646484375,
      "learning_rate": 2.5e-06,
      "loss": 227.1212,
      "step": 100
    },
    {
      "epoch": 0.05098139179199592,
      "grad_norm": 363.3997802734375,
      "learning_rate": 5e-06,
      "loss": 88.4222,
      "step": 200
    },
    {
      "epoch": 0.07647208768799388,
      "grad_norm": 263.5315246582031,
      "learning_rate": 7.500000000000001e-06,
      "loss": 67.8251,
      "step": 300
    },
    {
      "epoch": 0.10196278358399184,
      "grad_norm": 328.4805908203125,
      "learning_rate": 1e-05,
      "loss": 63.1532,
      "step": 400
    },
    {
      "epoch": 0.1274534794799898,
      "grad_norm": 496.2308044433594,
      "learning_rate": 9.995728791936505e-06,
      "loss": 60.2136,
      "step": 500
    },
    {
      "epoch": 0.15294417537598776,
      "grad_norm": 405.20025634765625,
      "learning_rate": 9.98292246503335e-06,
      "loss": 59.0975,
      "step": 600
    },
    {
      "epoch": 0.17843487127198573,
      "grad_norm": 451.58197021484375,
      "learning_rate": 9.961602898685225e-06,
      "loss": 55.2574,
      "step": 700
    },
    {
      "epoch": 0.20392556716798368,
      "grad_norm": 127.22090148925781,
      "learning_rate": 9.931806517013612e-06,
      "loss": 56.7531,
      "step": 800
    },
    {
      "epoch": 0.22941626306398163,
      "grad_norm": 431.8385925292969,
      "learning_rate": 9.893584226636773e-06,
      "loss": 57.5419,
      "step": 900
    },
    {
      "epoch": 0.2549069589599796,
      "grad_norm": 249.0337677001953,
      "learning_rate": 9.847001329696653e-06,
      "loss": 57.2545,
      "step": 1000
    },
    {
      "epoch": 0.28039765485597756,
      "grad_norm": 726.9508056640625,
      "learning_rate": 9.792137412291265e-06,
      "loss": 56.1048,
      "step": 1100
    },
    {
      "epoch": 0.3058883507519755,
      "grad_norm": 811.3701171875,
      "learning_rate": 9.729086208503174e-06,
      "loss": 53.1675,
      "step": 1200
    },
    {
      "epoch": 0.33137904664797346,
      "grad_norm": 754.3577270507812,
      "learning_rate": 9.657955440256396e-06,
      "loss": 54.9404,
      "step": 1300
    },
    {
      "epoch": 0.35686974254397147,
      "grad_norm": 267.7255554199219,
      "learning_rate": 9.578866633275289e-06,
      "loss": 55.7077,
      "step": 1400
    },
    {
      "epoch": 0.3823604384399694,
      "grad_norm": 244.13253784179688,
      "learning_rate": 9.491954909459895e-06,
      "loss": 56.8997,
      "step": 1500
    },
    {
      "epoch": 0.40785113433596737,
      "grad_norm": 302.5712890625,
      "learning_rate": 9.397368756032445e-06,
      "loss": 50.8917,
      "step": 1600
    },
    {
      "epoch": 0.4333418302319653,
      "grad_norm": 660.6898193359375,
      "learning_rate": 9.295269771849426e-06,
      "loss": 49.8636,
      "step": 1700
    },
    {
      "epoch": 0.45883252612796327,
      "grad_norm": 235.12942504882812,
      "learning_rate": 9.185832391312644e-06,
      "loss": 53.0454,
      "step": 1800
    },
    {
      "epoch": 0.4843232220239613,
      "grad_norm": 231.6936492919922,
      "learning_rate": 9.069243586350976e-06,
      "loss": 56.1398,
      "step": 1900
    },
    {
      "epoch": 0.5098139179199592,
      "grad_norm": 460.98486328125,
      "learning_rate": 8.94570254698197e-06,
      "loss": 52.8867,
      "step": 2000
    },
    {
      "epoch": 0.5353046138159572,
      "grad_norm": 567.8645629882812,
      "learning_rate": 8.815420340999034e-06,
      "loss": 56.6194,
      "step": 2100
    },
    {
      "epoch": 0.5607953097119551,
      "grad_norm": 158.99224853515625,
      "learning_rate": 8.67861955336566e-06,
      "loss": 50.311,
      "step": 2200
    },
    {
      "epoch": 0.5862860056079531,
      "grad_norm": 463.69635009765625,
      "learning_rate": 8.535533905932739e-06,
      "loss": 48.126,
      "step": 2300
    },
    {
      "epoch": 0.611776701503951,
      "grad_norm": 346.60333251953125,
      "learning_rate": 8.386407858128707e-06,
      "loss": 51.602,
      "step": 2400
    },
    {
      "epoch": 0.637267397399949,
      "grad_norm": 430.1127624511719,
      "learning_rate": 8.231496189304704e-06,
      "loss": 50.3868,
      "step": 2500
    },
    {
      "epoch": 0.6627580932959469,
      "grad_norm": 430.7168884277344,
      "learning_rate": 8.071063563448341e-06,
      "loss": 49.5458,
      "step": 2600
    },
    {
      "epoch": 0.688248789191945,
      "grad_norm": 409.49114990234375,
      "learning_rate": 7.905384077009693e-06,
      "loss": 51.5311,
      "step": 2700
    },
    {
      "epoch": 0.7137394850879429,
      "grad_norm": 443.4747619628906,
      "learning_rate": 7.734740790612137e-06,
      "loss": 50.3823,
      "step": 2800
    },
    {
      "epoch": 0.7392301809839409,
      "grad_norm": 459.6910705566406,
      "learning_rate": 7.559425245448006e-06,
      "loss": 46.766,
      "step": 2900
    },
    {
      "epoch": 0.7647208768799388,
      "grad_norm": 579.1948852539062,
      "learning_rate": 7.379736965185369e-06,
      "loss": 51.7392,
      "step": 3000
    },
    {
      "epoch": 0.7902115727759368,
      "grad_norm": 329.1972961425781,
      "learning_rate": 7.195982944236853e-06,
      "loss": 51.0259,
      "step": 3100
    },
    {
      "epoch": 0.8157022686719347,
      "grad_norm": 488.2768859863281,
      "learning_rate": 7.008477123264849e-06,
      "loss": 53.4051,
      "step": 3200
    },
    {
      "epoch": 0.8411929645679327,
      "grad_norm": 463.906494140625,
      "learning_rate": 6.817539852819149e-06,
      "loss": 43.2941,
      "step": 3300
    },
    {
      "epoch": 0.8666836604639306,
      "grad_norm": 263.90185546875,
      "learning_rate": 6.6234973460234184e-06,
      "loss": 50.0815,
      "step": 3400
    },
    {
      "epoch": 0.8921743563599286,
      "grad_norm": 186.7496337890625,
      "learning_rate": 6.426681121245527e-06,
      "loss": 49.7881,
      "step": 3500
    },
    {
      "epoch": 0.9176650522559265,
      "grad_norm": 398.3317565917969,
      "learning_rate": 6.227427435703997e-06,
      "loss": 50.899,
      "step": 3600
    },
    {
      "epoch": 0.9431557481519246,
      "grad_norm": 238.25408935546875,
      "learning_rate": 6.026076710978172e-06,
      "loss": 44.1347,
      "step": 3700
    },
    {
      "epoch": 0.9686464440479226,
      "grad_norm": 567.3057861328125,
      "learning_rate": 5.82297295140367e-06,
      "loss": 50.0442,
      "step": 3800
    },
    {
      "epoch": 0.9941371399439205,
      "grad_norm": 226.54409790039062,
      "learning_rate": 5.61846315634674e-06,
      "loss": 41.8183,
      "step": 3900
    },
    {
      "epoch": 1.0,
      "eval_loss": 2458.4501953125,
      "eval_runtime": 28.4173,
      "eval_samples_per_second": 122.707,
      "eval_steps_per_second": 15.343,
      "step": 3923
    },
    {
      "epoch": 1.0196278358399185,
      "grad_norm": 698.6807861328125,
      "learning_rate": 5.412896727361663e-06,
      "loss": 44.033,
      "step": 4000
    },
    {
      "epoch": 1.0451185317359164,
      "grad_norm": 525.1095581054688,
      "learning_rate": 5.206624871244066e-06,
      "loss": 44.6573,
      "step": 4100
    },
    {
      "epoch": 1.0706092276319144,
      "grad_norm": 500.8614501953125,
      "learning_rate": 5e-06,
      "loss": 42.8117,
      "step": 4200
    },
    {
      "epoch": 1.0960999235279123,
      "grad_norm": 487.75689697265625,
      "learning_rate": 4.793375128755934e-06,
      "loss": 43.7467,
      "step": 4300
    },
    {
      "epoch": 1.1215906194239103,
      "grad_norm": 385.0135192871094,
      "learning_rate": 4.587103272638339e-06,
      "loss": 43.3253,
      "step": 4400
    },
    {
      "epoch": 1.1470813153199082,
      "grad_norm": 796.0384521484375,
      "learning_rate": 4.381536843653262e-06,
      "loss": 44.0489,
      "step": 4500
    },
    {
      "epoch": 1.1725720112159062,
      "grad_norm": 428.2261962890625,
      "learning_rate": 4.17702704859633e-06,
      "loss": 48.0535,
      "step": 4600
    },
    {
      "epoch": 1.198062707111904,
      "grad_norm": 385.2317199707031,
      "learning_rate": 3.973923289021829e-06,
      "loss": 43.4284,
      "step": 4700
    },
    {
      "epoch": 1.223553403007902,
      "grad_norm": 159.48098754882812,
      "learning_rate": 3.7725725642960047e-06,
      "loss": 40.4862,
      "step": 4800
    },
    {
      "epoch": 1.2490440989039,
      "grad_norm": 200.54164123535156,
      "learning_rate": 3.573318878754475e-06,
      "loss": 44.7052,
      "step": 4900
    },
    {
      "epoch": 1.274534794799898,
      "grad_norm": 301.8349304199219,
      "learning_rate": 3.3765026539765832e-06,
      "loss": 41.2848,
      "step": 5000
    },
    {
      "epoch": 1.3000254906958961,
      "grad_norm": 335.5228271484375,
      "learning_rate": 3.1824601471808504e-06,
      "loss": 43.5137,
      "step": 5100
    },
    {
      "epoch": 1.3255161865918939,
      "grad_norm": 94.21492767333984,
      "learning_rate": 2.991522876735154e-06,
      "loss": 41.9114,
      "step": 5200
    },
    {
      "epoch": 1.351006882487892,
      "grad_norm": 352.31048583984375,
      "learning_rate": 2.804017055763149e-06,
      "loss": 41.0477,
      "step": 5300
    },
    {
      "epoch": 1.3764975783838898,
      "grad_norm": 228.1314697265625,
      "learning_rate": 2.6202630348146323e-06,
      "loss": 38.6621,
      "step": 5400
    },
    {
      "epoch": 1.401988274279888,
      "grad_norm": 170.14947509765625,
      "learning_rate": 2.4405747545519966e-06,
      "loss": 40.0131,
      "step": 5500
    },
    {
      "epoch": 1.4274789701758859,
      "grad_norm": 166.94281005859375,
      "learning_rate": 2.265259209387867e-06,
      "loss": 42.6173,
      "step": 5600
    },
    {
      "epoch": 1.4529696660718838,
      "grad_norm": 255.24964904785156,
      "learning_rate": 2.094615922990309e-06,
      "loss": 41.4291,
      "step": 5700
    },
    {
      "epoch": 1.4784603619678818,
      "grad_norm": 296.5071716308594,
      "learning_rate": 1.928936436551661e-06,
      "loss": 39.6349,
      "step": 5800
    },
    {
      "epoch": 1.5039510578638797,
      "grad_norm": 227.43650817871094,
      "learning_rate": 1.7685038106952952e-06,
      "loss": 40.9158,
      "step": 5900
    },
    {
      "epoch": 1.5294417537598777,
      "grad_norm": 280.7147216796875,
      "learning_rate": 1.6135921418712959e-06,
      "loss": 45.4827,
      "step": 6000
    },
    {
      "epoch": 1.5549324496558756,
      "grad_norm": 166.27865600585938,
      "learning_rate": 1.4644660940672628e-06,
      "loss": 40.6627,
      "step": 6100
    },
    {
      "epoch": 1.5804231455518736,
      "grad_norm": 389.299560546875,
      "learning_rate": 1.321380446634342e-06,
      "loss": 38.7581,
      "step": 6200
    },
    {
      "epoch": 1.6059138414478715,
      "grad_norm": 170.0509033203125,
      "learning_rate": 1.1845796590009684e-06,
      "loss": 40.3946,
      "step": 6300
    },
    {
      "epoch": 1.6314045373438695,
      "grad_norm": 287.59423828125,
      "learning_rate": 1.0542974530180327e-06,
      "loss": 43.811,
      "step": 6400
    },
    {
      "epoch": 1.6568952332398674,
      "grad_norm": 150.49293518066406,
      "learning_rate": 9.307564136490255e-07,
      "loss": 40.4753,
      "step": 6500
    },
    {
      "epoch": 1.6823859291358654,
      "grad_norm": 264.41253662109375,
      "learning_rate": 8.141676086873574e-07,
      "loss": 43.6258,
      "step": 6600
    },
    {
      "epoch": 1.7078766250318633,
      "grad_norm": 157.13479614257812,
      "learning_rate": 7.047302281505735e-07,
      "loss": 43.768,
      "step": 6700
    },
    {
      "epoch": 1.7333673209278613,
      "grad_norm": 206.51609802246094,
      "learning_rate": 6.026312439675553e-07,
      "loss": 39.8553,
      "step": 6800
    },
    {
      "epoch": 1.7588580168238592,
      "grad_norm": 415.54351806640625,
      "learning_rate": 5.080450905401057e-07,
      "loss": 37.9228,
      "step": 6900
    },
    {
      "epoch": 1.7843487127198574,
      "grad_norm": 189.60276794433594,
      "learning_rate": 4.211333667247125e-07,
      "loss": 41.446,
      "step": 7000
    },
    {
      "epoch": 1.8098394086158551,
      "grad_norm": 612.336181640625,
      "learning_rate": 3.420445597436056e-07,
      "loss": 39.8801,
      "step": 7100
    },
    {
      "epoch": 1.8353301045118533,
      "grad_norm": 338.9154052734375,
      "learning_rate": 2.7091379149682683e-07,
      "loss": 41.9962,
      "step": 7200
    },
    {
      "epoch": 1.860820800407851,
      "grad_norm": 207.06942749023438,
      "learning_rate": 2.0786258770873647e-07,
      "loss": 40.2771,
      "step": 7300
    },
    {
      "epoch": 1.8863114963038492,
      "grad_norm": 585.3238525390625,
      "learning_rate": 1.5299867030334815e-07,
      "loss": 37.347,
      "step": 7400
    },
    {
      "epoch": 1.911802192199847,
      "grad_norm": 224.99607849121094,
      "learning_rate": 1.0641577336322761e-07,
      "loss": 45.2395,
      "step": 7500
    },
    {
      "epoch": 1.937292888095845,
      "grad_norm": 304.6292724609375,
      "learning_rate": 6.819348298638839e-08,
      "loss": 51.0356,
      "step": 7600
    },
    {
      "epoch": 1.9627835839918428,
      "grad_norm": 1007.5321044921875,
      "learning_rate": 3.839710131477492e-08,
      "loss": 44.2402,
      "step": 7700
    },
    {
      "epoch": 1.988274279887841,
      "grad_norm": 436.2712097167969,
      "learning_rate": 1.7077534966650767e-08,
      "loss": 41.5961,
      "step": 7800
    },
    {
      "epoch": 2.0,
      "eval_loss": 2133.657470703125,
      "eval_runtime": 28.5718,
      "eval_samples_per_second": 122.043,
      "eval_steps_per_second": 15.26,
      "step": 7846
    },
    {
      "epoch": 2.0137649757838387,
      "grad_norm": 354.5096740722656,
      "learning_rate": 4.2712080634949024e-09,
      "loss": 41.8052,
      "step": 7900
    },
    {
      "epoch": 2.039255671679837,
      "grad_norm": 220.47647094726562,
      "learning_rate": 0.0,
      "loss": 39.9241,
      "step": 8000
    }
  ],
  "logging_steps": 100,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 2000,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}