{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 28.205128205128204,
  "eval_steps": 100,
  "global_step": 5500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 81.8341293334961,
      "learning_rate": 9.990900000000001e-06,
      "loss": 3.5606,
      "step": 100
    },
    {
      "epoch": 0.5128205128205128,
      "eval_loss": 2.2667126655578613,
      "eval_runtime": 33.7385,
      "eval_samples_per_second": 11.648,
      "eval_steps_per_second": 1.482,
      "step": 100
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 30.16681671142578,
      "learning_rate": 9.980900000000001e-06,
      "loss": 2.1841,
      "step": 200
    },
    {
      "epoch": 1.0256410256410255,
      "eval_loss": 2.059685230255127,
      "eval_runtime": 33.6897,
      "eval_samples_per_second": 11.665,
      "eval_steps_per_second": 1.484,
      "step": 200
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 67.16844177246094,
      "learning_rate": 9.970900000000001e-06,
      "loss": 2.0949,
      "step": 300
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 1.9054155349731445,
      "eval_runtime": 33.6472,
      "eval_samples_per_second": 11.68,
      "eval_steps_per_second": 1.486,
      "step": 300
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 46.190330505371094,
      "learning_rate": 9.960900000000001e-06,
      "loss": 1.9236,
      "step": 400
    },
    {
      "epoch": 2.051282051282051,
      "eval_loss": 1.8464823961257935,
      "eval_runtime": 33.8143,
      "eval_samples_per_second": 11.622,
      "eval_steps_per_second": 1.479,
      "step": 400
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 42.82326889038086,
      "learning_rate": 9.950900000000002e-06,
      "loss": 1.937,
      "step": 500
    },
    {
      "epoch": 2.564102564102564,
      "eval_loss": 1.8186373710632324,
      "eval_runtime": 33.9112,
      "eval_samples_per_second": 11.589,
      "eval_steps_per_second": 1.474,
      "step": 500
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 68.72016906738281,
      "learning_rate": 9.940900000000002e-06,
      "loss": 1.8648,
      "step": 600
    },
    {
      "epoch": 3.076923076923077,
      "eval_loss": 1.7807129621505737,
      "eval_runtime": 33.4957,
      "eval_samples_per_second": 11.733,
      "eval_steps_per_second": 1.493,
      "step": 600
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 51.390071868896484,
      "learning_rate": 9.930900000000002e-06,
      "loss": 1.8162,
      "step": 700
    },
    {
      "epoch": 3.58974358974359,
      "eval_loss": 1.7275961637496948,
      "eval_runtime": 33.6095,
      "eval_samples_per_second": 11.693,
      "eval_steps_per_second": 1.488,
      "step": 700
    },
    {
      "epoch": 4.102564102564102,
      "grad_norm": 33.017948150634766,
      "learning_rate": 9.920900000000002e-06,
      "loss": 1.7331,
      "step": 800
    },
    {
      "epoch": 4.102564102564102,
      "eval_loss": 1.701725959777832,
      "eval_runtime": 33.9602,
      "eval_samples_per_second": 11.572,
      "eval_steps_per_second": 1.472,
      "step": 800
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 43.55423355102539,
      "learning_rate": 9.9109e-06,
      "loss": 1.7091,
      "step": 900
    },
    {
      "epoch": 4.615384615384615,
      "eval_loss": 1.6944857835769653,
      "eval_runtime": 33.6785,
      "eval_samples_per_second": 11.669,
      "eval_steps_per_second": 1.485,
      "step": 900
    },
    {
      "epoch": 5.128205128205128,
      "grad_norm": 67.6618881225586,
      "learning_rate": 9.9009e-06,
      "loss": 1.7258,
      "step": 1000
    },
    {
      "epoch": 5.128205128205128,
      "eval_loss": 1.65646493434906,
      "eval_runtime": 33.8316,
      "eval_samples_per_second": 11.616,
      "eval_steps_per_second": 1.478,
      "step": 1000
    },
    {
      "epoch": 5.641025641025641,
      "grad_norm": 35.863487243652344,
      "learning_rate": 9.8909e-06,
      "loss": 1.661,
      "step": 1100
    },
    {
      "epoch": 5.641025641025641,
      "eval_loss": 1.6175495386123657,
      "eval_runtime": 33.57,
      "eval_samples_per_second": 11.707,
      "eval_steps_per_second": 1.489,
      "step": 1100
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 56.98338317871094,
      "learning_rate": 9.8809e-06,
      "loss": 1.6368,
      "step": 1200
    },
    {
      "epoch": 6.153846153846154,
      "eval_loss": 1.6185818910598755,
      "eval_runtime": 33.6954,
      "eval_samples_per_second": 11.663,
      "eval_steps_per_second": 1.484,
      "step": 1200
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 90.9850845336914,
      "learning_rate": 9.8709e-06,
      "loss": 1.6771,
      "step": 1300
    },
    {
      "epoch": 6.666666666666667,
      "eval_loss": 1.597514033317566,
      "eval_runtime": 33.5635,
      "eval_samples_per_second": 11.709,
      "eval_steps_per_second": 1.49,
      "step": 1300
    },
    {
      "epoch": 7.17948717948718,
      "grad_norm": 83.6468276977539,
      "learning_rate": 9.8609e-06,
      "loss": 1.5796,
      "step": 1400
    },
    {
      "epoch": 7.17948717948718,
      "eval_loss": 1.57249116897583,
      "eval_runtime": 33.6053,
      "eval_samples_per_second": 11.695,
      "eval_steps_per_second": 1.488,
      "step": 1400
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 42.52065658569336,
      "learning_rate": 9.8509e-06,
      "loss": 1.6012,
      "step": 1500
    },
    {
      "epoch": 7.6923076923076925,
      "eval_loss": 1.5940505266189575,
      "eval_runtime": 33.6416,
      "eval_samples_per_second": 11.682,
      "eval_steps_per_second": 1.486,
      "step": 1500
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 64.46149444580078,
      "learning_rate": 9.840900000000001e-06,
      "loss": 1.6107,
      "step": 1600
    },
    {
      "epoch": 8.205128205128204,
      "eval_loss": 1.57711923122406,
      "eval_runtime": 33.8072,
      "eval_samples_per_second": 11.625,
      "eval_steps_per_second": 1.479,
      "step": 1600
    },
    {
      "epoch": 8.717948717948717,
      "grad_norm": 47.13843536376953,
      "learning_rate": 9.830900000000001e-06,
      "loss": 1.6094,
      "step": 1700
    },
    {
      "epoch": 8.717948717948717,
      "eval_loss": 1.5567021369934082,
      "eval_runtime": 33.6256,
      "eval_samples_per_second": 11.688,
      "eval_steps_per_second": 1.487,
      "step": 1700
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 60.96649932861328,
      "learning_rate": 9.820900000000001e-06,
      "loss": 1.5679,
      "step": 1800
    },
    {
      "epoch": 9.23076923076923,
      "eval_loss": 1.5404459238052368,
      "eval_runtime": 33.5651,
      "eval_samples_per_second": 11.709,
      "eval_steps_per_second": 1.49,
      "step": 1800
    },
    {
      "epoch": 9.743589743589745,
      "grad_norm": 48.0488395690918,
      "learning_rate": 9.810900000000001e-06,
      "loss": 1.541,
      "step": 1900
    },
    {
      "epoch": 9.743589743589745,
      "eval_loss": 1.5000860691070557,
      "eval_runtime": 33.729,
      "eval_samples_per_second": 11.652,
      "eval_steps_per_second": 1.482,
      "step": 1900
    },
    {
      "epoch": 10.256410256410255,
      "grad_norm": 21.072124481201172,
      "learning_rate": 9.800900000000001e-06,
      "loss": 1.4747,
      "step": 2000
    },
    {
      "epoch": 10.256410256410255,
      "eval_loss": 1.4845048189163208,
      "eval_runtime": 33.6789,
      "eval_samples_per_second": 11.669,
      "eval_steps_per_second": 1.485,
      "step": 2000
    },
    {
      "epoch": 10.76923076923077,
      "grad_norm": 60.55208969116211,
      "learning_rate": 9.790900000000001e-06,
      "loss": 1.5105,
      "step": 2100
    },
    {
      "epoch": 10.76923076923077,
      "eval_loss": 1.4587421417236328,
      "eval_runtime": 33.4313,
      "eval_samples_per_second": 11.755,
      "eval_steps_per_second": 1.496,
      "step": 2100
    },
    {
      "epoch": 11.282051282051283,
      "grad_norm": 26.457704544067383,
      "learning_rate": 9.780900000000002e-06,
      "loss": 1.4769,
      "step": 2200
    },
    {
      "epoch": 11.282051282051283,
      "eval_loss": 1.4588648080825806,
      "eval_runtime": 33.5978,
      "eval_samples_per_second": 11.697,
      "eval_steps_per_second": 1.488,
      "step": 2200
    },
    {
      "epoch": 11.794871794871796,
      "grad_norm": 23.801916122436523,
      "learning_rate": 9.770900000000002e-06,
      "loss": 1.4555,
      "step": 2300
    },
    {
      "epoch": 11.794871794871796,
      "eval_loss": 1.4805899858474731,
      "eval_runtime": 33.4947,
      "eval_samples_per_second": 11.733,
      "eval_steps_per_second": 1.493,
      "step": 2300
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 24.780275344848633,
      "learning_rate": 9.760900000000002e-06,
      "loss": 1.4537,
      "step": 2400
    },
    {
      "epoch": 12.307692307692308,
      "eval_loss": 1.4648451805114746,
      "eval_runtime": 33.6877,
      "eval_samples_per_second": 11.666,
      "eval_steps_per_second": 1.484,
      "step": 2400
    },
    {
      "epoch": 12.820512820512821,
      "grad_norm": 65.28894805908203,
      "learning_rate": 9.7509e-06,
      "loss": 1.4788,
      "step": 2500
    },
    {
      "epoch": 12.820512820512821,
      "eval_loss": 1.4453215599060059,
      "eval_runtime": 33.972,
      "eval_samples_per_second": 11.568,
      "eval_steps_per_second": 1.472,
      "step": 2500
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 23.36334800720215,
      "learning_rate": 9.7409e-06,
      "loss": 1.4397,
      "step": 2600
    },
    {
      "epoch": 13.333333333333334,
      "eval_loss": 1.4110159873962402,
      "eval_runtime": 33.3905,
      "eval_samples_per_second": 11.77,
      "eval_steps_per_second": 1.497,
      "step": 2600
    },
    {
      "epoch": 13.846153846153847,
      "grad_norm": 24.13091468811035,
      "learning_rate": 9.7309e-06,
      "loss": 1.42,
      "step": 2700
    },
    {
      "epoch": 13.846153846153847,
      "eval_loss": 1.3875733613967896,
      "eval_runtime": 33.5662,
      "eval_samples_per_second": 11.708,
      "eval_steps_per_second": 1.49,
      "step": 2700
    },
    {
      "epoch": 14.35897435897436,
      "grad_norm": 41.02390670776367,
      "learning_rate": 9.7209e-06,
      "loss": 1.3833,
      "step": 2800
    },
    {
      "epoch": 14.35897435897436,
      "eval_loss": 1.3958121538162231,
      "eval_runtime": 33.6111,
      "eval_samples_per_second": 11.693,
      "eval_steps_per_second": 1.488,
      "step": 2800
    },
    {
      "epoch": 14.871794871794872,
      "grad_norm": 25.11317253112793,
      "learning_rate": 9.7109e-06,
      "loss": 1.3839,
      "step": 2900
    },
    {
      "epoch": 14.871794871794872,
      "eval_loss": 1.4024689197540283,
      "eval_runtime": 33.5271,
      "eval_samples_per_second": 11.722,
      "eval_steps_per_second": 1.491,
      "step": 2900
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 32.921974182128906,
      "learning_rate": 9.7009e-06,
      "loss": 1.3998,
      "step": 3000
    },
    {
      "epoch": 15.384615384615385,
      "eval_loss": 1.3834290504455566,
      "eval_runtime": 33.4449,
      "eval_samples_per_second": 11.751,
      "eval_steps_per_second": 1.495,
      "step": 3000
    },
    {
      "epoch": 15.897435897435898,
      "grad_norm": 71.14076232910156,
      "learning_rate": 9.6909e-06,
      "loss": 1.3753,
      "step": 3100
    },
    {
      "epoch": 15.897435897435898,
      "eval_loss": 1.371915578842163,
      "eval_runtime": 33.5787,
      "eval_samples_per_second": 11.704,
      "eval_steps_per_second": 1.489,
      "step": 3100
    },
    {
      "epoch": 16.41025641025641,
      "grad_norm": 85.71231842041016,
      "learning_rate": 9.6809e-06,
      "loss": 1.3639,
      "step": 3200
    },
    {
      "epoch": 16.41025641025641,
      "eval_loss": 1.3810721635818481,
      "eval_runtime": 33.5802,
      "eval_samples_per_second": 11.703,
      "eval_steps_per_second": 1.489,
      "step": 3200
    },
    {
      "epoch": 16.923076923076923,
      "grad_norm": 39.7967529296875,
      "learning_rate": 9.670900000000001e-06,
      "loss": 1.3609,
      "step": 3300
    },
    {
      "epoch": 16.923076923076923,
      "eval_loss": 1.3853365182876587,
      "eval_runtime": 33.6581,
      "eval_samples_per_second": 11.676,
      "eval_steps_per_second": 1.486,
      "step": 3300
    },
    {
      "epoch": 17.435897435897434,
      "grad_norm": 19.84697723388672,
      "learning_rate": 9.660900000000001e-06,
      "loss": 1.3314,
      "step": 3400
    },
    {
      "epoch": 17.435897435897434,
      "eval_loss": 1.3581265211105347,
      "eval_runtime": 33.4833,
      "eval_samples_per_second": 11.737,
      "eval_steps_per_second": 1.493,
      "step": 3400
    },
    {
      "epoch": 17.94871794871795,
      "grad_norm": 43.280426025390625,
      "learning_rate": 9.650900000000001e-06,
      "loss": 1.3408,
      "step": 3500
    },
    {
      "epoch": 17.94871794871795,
      "eval_loss": 1.3400689363479614,
      "eval_runtime": 33.4394,
      "eval_samples_per_second": 11.753,
      "eval_steps_per_second": 1.495,
      "step": 3500
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 35.393314361572266,
      "learning_rate": 9.640900000000001e-06,
      "loss": 1.3471,
      "step": 3600
    },
    {
      "epoch": 18.46153846153846,
      "eval_loss": 1.3405698537826538,
      "eval_runtime": 33.6347,
      "eval_samples_per_second": 11.684,
      "eval_steps_per_second": 1.487,
      "step": 3600
    },
    {
      "epoch": 18.974358974358974,
      "grad_norm": 32.50804138183594,
      "learning_rate": 9.630900000000001e-06,
      "loss": 1.3213,
      "step": 3700
    },
    {
      "epoch": 18.974358974358974,
      "eval_loss": 1.3345987796783447,
      "eval_runtime": 33.5369,
      "eval_samples_per_second": 11.718,
      "eval_steps_per_second": 1.491,
      "step": 3700
    },
    {
      "epoch": 19.487179487179485,
      "grad_norm": 28.999061584472656,
      "learning_rate": 9.620900000000001e-06,
      "loss": 1.3251,
      "step": 3800
    },
    {
      "epoch": 19.487179487179485,
      "eval_loss": 1.3227241039276123,
      "eval_runtime": 33.5001,
      "eval_samples_per_second": 11.731,
      "eval_steps_per_second": 1.493,
      "step": 3800
    },
    {
      "epoch": 20.0,
      "grad_norm": 41.793636322021484,
      "learning_rate": 9.610900000000001e-06,
      "loss": 1.3297,
      "step": 3900
    },
    {
      "epoch": 20.0,
      "eval_loss": 1.3213008642196655,
      "eval_runtime": 33.561,
      "eval_samples_per_second": 11.71,
      "eval_steps_per_second": 1.49,
      "step": 3900
    },
    {
      "epoch": 20.51282051282051,
      "grad_norm": 53.55360794067383,
      "learning_rate": 9.600900000000002e-06,
      "loss": 1.331,
      "step": 4000
    },
    {
      "epoch": 20.51282051282051,
      "eval_loss": 1.3293949365615845,
      "eval_runtime": 33.5542,
      "eval_samples_per_second": 11.712,
      "eval_steps_per_second": 1.49,
      "step": 4000
    },
    {
      "epoch": 21.025641025641026,
      "grad_norm": 73.3389892578125,
      "learning_rate": 9.5909e-06,
      "loss": 1.2856,
      "step": 4100
    },
    {
      "epoch": 21.025641025641026,
      "eval_loss": 1.316487193107605,
      "eval_runtime": 33.5602,
      "eval_samples_per_second": 11.71,
      "eval_steps_per_second": 1.49,
      "step": 4100
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 62.724456787109375,
      "learning_rate": 9.5809e-06,
      "loss": 1.3186,
      "step": 4200
    },
    {
      "epoch": 21.53846153846154,
      "eval_loss": 1.3179290294647217,
      "eval_runtime": 33.621,
      "eval_samples_per_second": 11.689,
      "eval_steps_per_second": 1.487,
      "step": 4200
    },
    {
      "epoch": 22.05128205128205,
      "grad_norm": 30.242937088012695,
      "learning_rate": 9.5709e-06,
      "loss": 1.2623,
      "step": 4300
    },
    {
      "epoch": 22.05128205128205,
      "eval_loss": 1.2920558452606201,
      "eval_runtime": 33.6166,
      "eval_samples_per_second": 11.691,
      "eval_steps_per_second": 1.487,
      "step": 4300
    },
    {
      "epoch": 22.564102564102566,
      "grad_norm": 17.423118591308594,
      "learning_rate": 9.5609e-06,
      "loss": 1.3045,
      "step": 4400
    },
    {
      "epoch": 22.564102564102566,
      "eval_loss": 1.302635669708252,
      "eval_runtime": 33.7322,
      "eval_samples_per_second": 11.651,
      "eval_steps_per_second": 1.482,
      "step": 4400
    },
    {
      "epoch": 23.076923076923077,
      "grad_norm": 115.05293273925781,
      "learning_rate": 9.5509e-06,
      "loss": 1.2997,
      "step": 4500
    },
    {
      "epoch": 23.076923076923077,
      "eval_loss": 1.2834452390670776,
      "eval_runtime": 33.5349,
      "eval_samples_per_second": 11.719,
      "eval_steps_per_second": 1.491,
      "step": 4500
    },
    {
      "epoch": 23.58974358974359,
      "grad_norm": 35.13385009765625,
      "learning_rate": 9.5409e-06,
      "loss": 1.2591,
      "step": 4600
    },
    {
      "epoch": 23.58974358974359,
      "eval_loss": 1.2677439451217651,
      "eval_runtime": 33.5991,
      "eval_samples_per_second": 11.697,
      "eval_steps_per_second": 1.488,
      "step": 4600
    },
    {
      "epoch": 24.102564102564102,
      "grad_norm": 22.429704666137695,
      "learning_rate": 9.5309e-06,
      "loss": 1.3057,
      "step": 4700
    },
    {
      "epoch": 24.102564102564102,
      "eval_loss": 1.2759162187576294,
      "eval_runtime": 33.8336,
      "eval_samples_per_second": 11.616,
      "eval_steps_per_second": 1.478,
      "step": 4700
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 19.258209228515625,
      "learning_rate": 9.5209e-06,
      "loss": 1.2526,
      "step": 4800
    },
    {
      "epoch": 24.615384615384617,
      "eval_loss": 1.2712483406066895,
      "eval_runtime": 33.585,
      "eval_samples_per_second": 11.702,
      "eval_steps_per_second": 1.489,
      "step": 4800
    },
    {
      "epoch": 25.128205128205128,
      "grad_norm": 18.289043426513672,
      "learning_rate": 9.5109e-06,
      "loss": 1.2436,
      "step": 4900
    },
    {
      "epoch": 25.128205128205128,
      "eval_loss": 1.277999758720398,
      "eval_runtime": 33.6639,
      "eval_samples_per_second": 11.674,
      "eval_steps_per_second": 1.485,
      "step": 4900
    },
    {
      "epoch": 25.641025641025642,
      "grad_norm": 53.41452407836914,
      "learning_rate": 9.501000000000001e-06,
      "loss": 1.2768,
      "step": 5000
    },
    {
      "epoch": 25.641025641025642,
      "eval_loss": 1.2602609395980835,
      "eval_runtime": 33.6751,
      "eval_samples_per_second": 11.67,
      "eval_steps_per_second": 1.485,
      "step": 5000
    },
    {
      "epoch": 26.153846153846153,
      "grad_norm": 28.17922019958496,
      "learning_rate": 9.491000000000001e-06,
      "loss": 1.2221,
      "step": 5100
    },
    {
      "epoch": 26.153846153846153,
      "eval_loss": 1.28999662399292,
      "eval_runtime": 33.8104,
      "eval_samples_per_second": 11.624,
      "eval_steps_per_second": 1.479,
      "step": 5100
    },
    {
      "epoch": 26.666666666666668,
      "grad_norm": 28.119773864746094,
      "learning_rate": 9.481000000000001e-06,
      "loss": 1.233,
      "step": 5200
    },
    {
      "epoch": 26.666666666666668,
      "eval_loss": 1.2724733352661133,
      "eval_runtime": 33.9553,
      "eval_samples_per_second": 11.574,
      "eval_steps_per_second": 1.473,
      "step": 5200
    },
    {
      "epoch": 27.17948717948718,
      "grad_norm": 102.97128295898438,
      "learning_rate": 9.471000000000001e-06,
      "loss": 1.2667,
      "step": 5300
    },
    {
      "epoch": 27.17948717948718,
      "eval_loss": 1.265009880065918,
      "eval_runtime": 33.7879,
      "eval_samples_per_second": 11.631,
      "eval_steps_per_second": 1.48,
      "step": 5300
    },
    {
      "epoch": 27.692307692307693,
      "grad_norm": 23.439664840698242,
      "learning_rate": 9.461000000000001e-06,
      "loss": 1.2255,
      "step": 5400
    },
    {
      "epoch": 27.692307692307693,
      "eval_loss": 1.2421387434005737,
      "eval_runtime": 33.867,
      "eval_samples_per_second": 11.604,
      "eval_steps_per_second": 1.476,
      "step": 5400
    },
    {
      "epoch": 28.205128205128204,
      "grad_norm": 34.7141227722168,
      "learning_rate": 9.451000000000002e-06,
      "loss": 1.2557,
      "step": 5500
    },
    {
      "epoch": 28.205128205128204,
      "eval_loss": 1.2613080739974976,
      "eval_runtime": 33.9538,
      "eval_samples_per_second": 11.575,
      "eval_steps_per_second": 1.473,
      "step": 5500
    }
  ],
  "logging_steps": 100,
  "max_steps": 100000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 513,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.102337414144e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}