{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 25.641025641025642,
  "eval_steps": 100,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 92.9238510131836,
      "learning_rate": 9.9907e-06,
      "loss": 2.4016,
      "step": 100
    },
    {
      "epoch": 0.5128205128205128,
      "eval_loss": 2.1255013942718506,
      "eval_runtime": 34.4318,
      "eval_samples_per_second": 11.414,
      "eval_steps_per_second": 1.452,
      "step": 100
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 47.019439697265625,
      "learning_rate": 9.980700000000001e-06,
      "loss": 2.0696,
      "step": 200
    },
    {
      "epoch": 1.0256410256410255,
      "eval_loss": 1.9405728578567505,
      "eval_runtime": 34.0713,
      "eval_samples_per_second": 11.535,
      "eval_steps_per_second": 1.468,
      "step": 200
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 52.230751037597656,
      "learning_rate": 9.970700000000001e-06,
      "loss": 1.9983,
      "step": 300
    },
    {
      "epoch": 1.5384615384615383,
      "eval_loss": 1.900020718574524,
      "eval_runtime": 34.1694,
      "eval_samples_per_second": 11.502,
      "eval_steps_per_second": 1.463,
      "step": 300
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 39.62374496459961,
      "learning_rate": 9.960800000000001e-06,
      "loss": 1.8888,
      "step": 400
    },
    {
      "epoch": 2.051282051282051,
      "eval_loss": 1.8164124488830566,
      "eval_runtime": 34.1213,
      "eval_samples_per_second": 11.518,
      "eval_steps_per_second": 1.465,
      "step": 400
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 39.697731018066406,
      "learning_rate": 9.9508e-06,
      "loss": 1.8456,
      "step": 500
    },
    {
      "epoch": 2.564102564102564,
      "eval_loss": 1.7753618955612183,
      "eval_runtime": 34.149,
      "eval_samples_per_second": 11.508,
      "eval_steps_per_second": 1.464,
      "step": 500
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 52.823184967041016,
      "learning_rate": 9.9408e-06,
      "loss": 1.7839,
      "step": 600
    },
    {
      "epoch": 3.076923076923077,
      "eval_loss": 1.7344136238098145,
      "eval_runtime": 34.121,
      "eval_samples_per_second": 11.518,
      "eval_steps_per_second": 1.465,
      "step": 600
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 107.05223083496094,
      "learning_rate": 9.9308e-06,
      "loss": 1.7544,
      "step": 700
    },
    {
      "epoch": 3.58974358974359,
      "eval_loss": 1.6865615844726562,
      "eval_runtime": 34.0842,
      "eval_samples_per_second": 11.53,
      "eval_steps_per_second": 1.467,
      "step": 700
    },
    {
      "epoch": 4.102564102564102,
      "grad_norm": 53.641353607177734,
      "learning_rate": 9.9208e-06,
      "loss": 1.6812,
      "step": 800
    },
    {
      "epoch": 4.102564102564102,
      "eval_loss": 1.6568766832351685,
      "eval_runtime": 34.432,
      "eval_samples_per_second": 11.414,
      "eval_steps_per_second": 1.452,
      "step": 800
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 40.92328643798828,
      "learning_rate": 9.9109e-06,
      "loss": 1.6501,
      "step": 900
    },
    {
      "epoch": 4.615384615384615,
      "eval_loss": 1.6080188751220703,
      "eval_runtime": 34.1751,
      "eval_samples_per_second": 11.5,
      "eval_steps_per_second": 1.463,
      "step": 900
    },
    {
      "epoch": 5.128205128205128,
      "grad_norm": 28.52039909362793,
      "learning_rate": 9.9009e-06,
      "loss": 1.6579,
      "step": 1000
    },
    {
      "epoch": 5.128205128205128,
      "eval_loss": 1.6059809923171997,
      "eval_runtime": 34.1634,
      "eval_samples_per_second": 11.504,
      "eval_steps_per_second": 1.464,
      "step": 1000
    },
    {
      "epoch": 5.641025641025641,
      "grad_norm": 73.21099090576172,
      "learning_rate": 9.8909e-06,
      "loss": 1.6286,
      "step": 1100
    },
    {
      "epoch": 5.641025641025641,
      "eval_loss": 1.5779187679290771,
      "eval_runtime": 34.1489,
      "eval_samples_per_second": 11.508,
      "eval_steps_per_second": 1.464,
      "step": 1100
    },
    {
      "epoch": 6.153846153846154,
      "grad_norm": 36.768428802490234,
      "learning_rate": 9.8809e-06,
      "loss": 1.5871,
      "step": 1200
    },
    {
      "epoch": 6.153846153846154,
      "eval_loss": 1.5641562938690186,
      "eval_runtime": 34.1081,
      "eval_samples_per_second": 11.522,
      "eval_steps_per_second": 1.466,
      "step": 1200
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 28.098352432250977,
      "learning_rate": 9.8709e-06,
      "loss": 1.6231,
      "step": 1300
    },
    {
      "epoch": 6.666666666666667,
      "eval_loss": 1.530659556388855,
      "eval_runtime": 34.0717,
      "eval_samples_per_second": 11.534,
      "eval_steps_per_second": 1.467,
      "step": 1300
    },
    {
      "epoch": 7.17948717948718,
      "grad_norm": 48.131195068359375,
      "learning_rate": 9.8609e-06,
      "loss": 1.5178,
      "step": 1400
    },
    {
      "epoch": 7.17948717948718,
      "eval_loss": 1.5207794904708862,
      "eval_runtime": 34.0224,
      "eval_samples_per_second": 11.551,
      "eval_steps_per_second": 1.47,
      "step": 1400
    },
    {
      "epoch": 7.6923076923076925,
      "grad_norm": 15.9362211227417,
      "learning_rate": 9.8509e-06,
      "loss": 1.5434,
      "step": 1500
    },
    {
      "epoch": 7.6923076923076925,
      "eval_loss": 1.4978805780410767,
      "eval_runtime": 34.0269,
      "eval_samples_per_second": 11.55,
      "eval_steps_per_second": 1.469,
      "step": 1500
    },
    {
      "epoch": 8.205128205128204,
      "grad_norm": 21.21210479736328,
      "learning_rate": 9.840900000000001e-06,
      "loss": 1.5368,
      "step": 1600
    },
    {
      "epoch": 8.205128205128204,
      "eval_loss": 1.4791795015335083,
      "eval_runtime": 34.5402,
      "eval_samples_per_second": 11.378,
      "eval_steps_per_second": 1.448,
      "step": 1600
    },
    {
      "epoch": 8.717948717948717,
      "grad_norm": 39.53807830810547,
      "learning_rate": 9.830900000000001e-06,
      "loss": 1.5163,
      "step": 1700
    },
    {
      "epoch": 8.717948717948717,
      "eval_loss": 1.4630368947982788,
      "eval_runtime": 34.1482,
      "eval_samples_per_second": 11.509,
      "eval_steps_per_second": 1.464,
      "step": 1700
    },
    {
      "epoch": 9.23076923076923,
      "grad_norm": 42.564842224121094,
      "learning_rate": 9.820900000000001e-06,
      "loss": 1.483,
      "step": 1800
    },
    {
      "epoch": 9.23076923076923,
      "eval_loss": 1.4529582262039185,
      "eval_runtime": 34.2374,
      "eval_samples_per_second": 11.479,
      "eval_steps_per_second": 1.46,
      "step": 1800
    },
    {
      "epoch": 9.743589743589745,
      "grad_norm": 55.969017028808594,
      "learning_rate": 9.810900000000001e-06,
      "loss": 1.4795,
      "step": 1900
    },
    {
      "epoch": 9.743589743589745,
      "eval_loss": 1.4357120990753174,
      "eval_runtime": 34.2522,
      "eval_samples_per_second": 11.474,
      "eval_steps_per_second": 1.46,
      "step": 1900
    },
    {
      "epoch": 10.256410256410255,
      "grad_norm": 25.308517456054688,
      "learning_rate": 9.800900000000001e-06,
      "loss": 1.4151,
      "step": 2000
    },
    {
      "epoch": 10.256410256410255,
      "eval_loss": 1.432659387588501,
      "eval_runtime": 34.2016,
      "eval_samples_per_second": 11.491,
      "eval_steps_per_second": 1.462,
      "step": 2000
    },
    {
      "epoch": 10.76923076923077,
      "grad_norm": 54.97860336303711,
      "learning_rate": 9.790900000000001e-06,
      "loss": 1.4665,
      "step": 2100
    },
    {
      "epoch": 10.76923076923077,
      "eval_loss": 1.4322772026062012,
      "eval_runtime": 34.2513,
      "eval_samples_per_second": 11.474,
      "eval_steps_per_second": 1.46,
      "step": 2100
    },
    {
      "epoch": 11.282051282051283,
      "grad_norm": 26.281103134155273,
      "learning_rate": 9.780900000000002e-06,
      "loss": 1.4336,
      "step": 2200
    },
    {
      "epoch": 11.282051282051283,
      "eval_loss": 1.41593337059021,
      "eval_runtime": 33.815,
      "eval_samples_per_second": 11.622,
      "eval_steps_per_second": 1.479,
      "step": 2200
    },
    {
      "epoch": 11.794871794871796,
      "grad_norm": 37.469024658203125,
      "learning_rate": 9.770900000000002e-06,
      "loss": 1.4149,
      "step": 2300
    },
    {
      "epoch": 11.794871794871796,
      "eval_loss": 1.4032217264175415,
      "eval_runtime": 33.8113,
      "eval_samples_per_second": 11.623,
      "eval_steps_per_second": 1.479,
      "step": 2300
    },
    {
      "epoch": 12.307692307692308,
      "grad_norm": 17.248014450073242,
      "learning_rate": 9.760900000000002e-06,
      "loss": 1.3918,
      "step": 2400
    },
    {
      "epoch": 12.307692307692308,
      "eval_loss": 1.3832417726516724,
      "eval_runtime": 34.1456,
      "eval_samples_per_second": 11.51,
      "eval_steps_per_second": 1.464,
      "step": 2400
    },
    {
      "epoch": 12.820512820512821,
      "grad_norm": 36.33113479614258,
      "learning_rate": 9.7509e-06,
      "loss": 1.4319,
      "step": 2500
    },
    {
      "epoch": 12.820512820512821,
      "eval_loss": 1.384070634841919,
      "eval_runtime": 33.7658,
      "eval_samples_per_second": 11.639,
      "eval_steps_per_second": 1.481,
      "step": 2500
    },
    {
      "epoch": 13.333333333333334,
      "grad_norm": 28.724761962890625,
      "learning_rate": 9.7409e-06,
      "loss": 1.3936,
      "step": 2600
    },
    {
      "epoch": 13.333333333333334,
      "eval_loss": 1.3891230821609497,
      "eval_runtime": 33.7884,
      "eval_samples_per_second": 11.631,
      "eval_steps_per_second": 1.48,
      "step": 2600
    },
    {
      "epoch": 13.846153846153847,
      "grad_norm": 25.300365447998047,
      "learning_rate": 9.7309e-06,
      "loss": 1.386,
      "step": 2700
    },
    {
      "epoch": 13.846153846153847,
      "eval_loss": 1.373278260231018,
      "eval_runtime": 33.8109,
      "eval_samples_per_second": 11.623,
      "eval_steps_per_second": 1.479,
      "step": 2700
    },
    {
      "epoch": 14.35897435897436,
      "grad_norm": 165.74114990234375,
      "learning_rate": 9.7209e-06,
      "loss": 1.3514,
      "step": 2800
    },
    {
      "epoch": 14.35897435897436,
      "eval_loss": 1.376756191253662,
      "eval_runtime": 33.8844,
      "eval_samples_per_second": 11.598,
      "eval_steps_per_second": 1.476,
      "step": 2800
    },
    {
      "epoch": 14.871794871794872,
      "grad_norm": 37.75138854980469,
      "learning_rate": 9.7109e-06,
      "loss": 1.3542,
      "step": 2900
    },
    {
      "epoch": 14.871794871794872,
      "eval_loss": 1.3550430536270142,
      "eval_runtime": 34.0038,
      "eval_samples_per_second": 11.558,
      "eval_steps_per_second": 1.47,
      "step": 2900
    },
    {
      "epoch": 15.384615384615385,
      "grad_norm": 41.38801956176758,
      "learning_rate": 9.7009e-06,
      "loss": 1.3653,
      "step": 3000
    },
    {
      "epoch": 15.384615384615385,
      "eval_loss": 1.3493261337280273,
      "eval_runtime": 33.7831,
      "eval_samples_per_second": 11.633,
      "eval_steps_per_second": 1.48,
      "step": 3000
    },
    {
      "epoch": 15.897435897435898,
      "grad_norm": 54.01289749145508,
      "learning_rate": 9.6909e-06,
      "loss": 1.3451,
      "step": 3100
    },
    {
      "epoch": 15.897435897435898,
      "eval_loss": 1.3563708066940308,
      "eval_runtime": 33.7662,
      "eval_samples_per_second": 11.639,
      "eval_steps_per_second": 1.481,
      "step": 3100
    },
    {
      "epoch": 16.41025641025641,
      "grad_norm": 47.22911834716797,
      "learning_rate": 9.6809e-06,
      "loss": 1.323,
      "step": 3200
    },
    {
      "epoch": 16.41025641025641,
      "eval_loss": 1.343112587928772,
      "eval_runtime": 34.3174,
      "eval_samples_per_second": 11.452,
      "eval_steps_per_second": 1.457,
      "step": 3200
    },
    {
      "epoch": 16.923076923076923,
      "grad_norm": 254.61305236816406,
      "learning_rate": 9.670900000000001e-06,
      "loss": 1.3348,
      "step": 3300
    },
    {
      "epoch": 16.923076923076923,
      "eval_loss": 1.3341063261032104,
      "eval_runtime": 33.9905,
      "eval_samples_per_second": 11.562,
      "eval_steps_per_second": 1.471,
      "step": 3300
    },
    {
      "epoch": 17.435897435897434,
      "grad_norm": 25.823589324951172,
      "learning_rate": 9.660900000000001e-06,
      "loss": 1.3185,
      "step": 3400
    },
    {
      "epoch": 17.435897435897434,
      "eval_loss": 1.3256839513778687,
      "eval_runtime": 33.9185,
      "eval_samples_per_second": 11.587,
      "eval_steps_per_second": 1.474,
      "step": 3400
    },
    {
      "epoch": 17.94871794871795,
      "grad_norm": 44.86510467529297,
      "learning_rate": 9.650900000000001e-06,
      "loss": 1.328,
      "step": 3500
    },
    {
      "epoch": 17.94871794871795,
      "eval_loss": 1.3411694765090942,
      "eval_runtime": 33.8769,
      "eval_samples_per_second": 11.601,
      "eval_steps_per_second": 1.476,
      "step": 3500
    },
    {
      "epoch": 18.46153846153846,
      "grad_norm": 20.39423942565918,
      "learning_rate": 9.640900000000001e-06,
      "loss": 1.3306,
      "step": 3600
    },
    {
      "epoch": 18.46153846153846,
      "eval_loss": 1.318002462387085,
      "eval_runtime": 33.7275,
      "eval_samples_per_second": 11.652,
      "eval_steps_per_second": 1.482,
      "step": 3600
    },
    {
      "epoch": 18.974358974358974,
      "grad_norm": 42.16006088256836,
      "learning_rate": 9.630900000000001e-06,
      "loss": 1.301,
      "step": 3700
    },
    {
      "epoch": 18.974358974358974,
      "eval_loss": 1.3143377304077148,
      "eval_runtime": 33.8038,
      "eval_samples_per_second": 11.626,
      "eval_steps_per_second": 1.479,
      "step": 3700
    },
    {
      "epoch": 19.487179487179485,
      "grad_norm": 36.292999267578125,
      "learning_rate": 9.620900000000001e-06,
      "loss": 1.3036,
      "step": 3800
    },
    {
      "epoch": 19.487179487179485,
      "eval_loss": 1.3157984018325806,
      "eval_runtime": 33.7447,
      "eval_samples_per_second": 11.646,
      "eval_steps_per_second": 1.482,
      "step": 3800
    },
    {
      "epoch": 20.0,
      "grad_norm": 64.44886016845703,
      "learning_rate": 9.610900000000001e-06,
      "loss": 1.3077,
      "step": 3900
    },
    {
      "epoch": 20.0,
      "eval_loss": 1.31955885887146,
      "eval_runtime": 33.8296,
      "eval_samples_per_second": 11.617,
      "eval_steps_per_second": 1.478,
      "step": 3900
    },
    {
      "epoch": 20.51282051282051,
      "grad_norm": 37.88364028930664,
      "learning_rate": 9.600900000000002e-06,
      "loss": 1.3092,
      "step": 4000
    },
    {
      "epoch": 20.51282051282051,
      "eval_loss": 1.3180863857269287,
      "eval_runtime": 34.1642,
      "eval_samples_per_second": 11.503,
      "eval_steps_per_second": 1.464,
      "step": 4000
    },
    {
      "epoch": 21.025641025641026,
      "grad_norm": 27.300247192382812,
      "learning_rate": 9.5909e-06,
      "loss": 1.27,
      "step": 4100
    },
    {
      "epoch": 21.025641025641026,
      "eval_loss": 1.3031089305877686,
      "eval_runtime": 34.0151,
      "eval_samples_per_second": 11.554,
      "eval_steps_per_second": 1.47,
      "step": 4100
    },
    {
      "epoch": 21.53846153846154,
      "grad_norm": 27.94341278076172,
      "learning_rate": 9.5809e-06,
      "loss": 1.2983,
      "step": 4200
    },
    {
      "epoch": 21.53846153846154,
      "eval_loss": 1.3018323183059692,
      "eval_runtime": 34.1294,
      "eval_samples_per_second": 11.515,
      "eval_steps_per_second": 1.465,
      "step": 4200
    },
    {
      "epoch": 22.05128205128205,
      "grad_norm": 35.468135833740234,
      "learning_rate": 9.5709e-06,
      "loss": 1.2568,
      "step": 4300
    },
    {
      "epoch": 22.05128205128205,
      "eval_loss": 1.3130985498428345,
      "eval_runtime": 33.9979,
      "eval_samples_per_second": 11.56,
      "eval_steps_per_second": 1.471,
      "step": 4300
    },
    {
      "epoch": 22.564102564102566,
      "grad_norm": 21.431076049804688,
      "learning_rate": 9.5609e-06,
      "loss": 1.2812,
      "step": 4400
    },
    {
      "epoch": 22.564102564102566,
      "eval_loss": 1.293820858001709,
      "eval_runtime": 33.9036,
      "eval_samples_per_second": 11.592,
      "eval_steps_per_second": 1.475,
      "step": 4400
    },
    {
      "epoch": 23.076923076923077,
      "grad_norm": 59.38164520263672,
      "learning_rate": 9.5509e-06,
      "loss": 1.2776,
      "step": 4500
    },
    {
      "epoch": 23.076923076923077,
      "eval_loss": 1.2872756719589233,
      "eval_runtime": 33.8513,
      "eval_samples_per_second": 11.61,
      "eval_steps_per_second": 1.477,
      "step": 4500
    },
    {
      "epoch": 23.58974358974359,
      "grad_norm": 48.05171585083008,
      "learning_rate": 9.5409e-06,
      "loss": 1.2488,
      "step": 4600
    },
    {
      "epoch": 23.58974358974359,
      "eval_loss": 1.2994896173477173,
      "eval_runtime": 33.9738,
      "eval_samples_per_second": 11.568,
      "eval_steps_per_second": 1.472,
      "step": 4600
    },
    {
      "epoch": 24.102564102564102,
      "grad_norm": 32.86454391479492,
      "learning_rate": 9.5309e-06,
      "loss": 1.3085,
      "step": 4700
    },
    {
      "epoch": 24.102564102564102,
      "eval_loss": 1.2995264530181885,
      "eval_runtime": 33.8598,
      "eval_samples_per_second": 11.607,
      "eval_steps_per_second": 1.477,
      "step": 4700
    },
    {
      "epoch": 24.615384615384617,
      "grad_norm": 36.235042572021484,
      "learning_rate": 9.5209e-06,
      "loss": 1.2439,
      "step": 4800
    },
    {
      "epoch": 24.615384615384617,
      "eval_loss": 1.285250186920166,
      "eval_runtime": 34.2054,
      "eval_samples_per_second": 11.489,
      "eval_steps_per_second": 1.462,
      "step": 4800
    },
    {
      "epoch": 25.128205128205128,
      "grad_norm": 17.993911743164062,
      "learning_rate": 9.5109e-06,
      "loss": 1.2427,
      "step": 4900
    },
    {
      "epoch": 25.128205128205128,
      "eval_loss": 1.2752227783203125,
      "eval_runtime": 33.8959,
      "eval_samples_per_second": 11.594,
      "eval_steps_per_second": 1.475,
      "step": 4900
    },
    {
      "epoch": 25.641025641025642,
      "grad_norm": 55.71548080444336,
      "learning_rate": 9.501000000000001e-06,
      "loss": 1.2774,
      "step": 5000
    },
    {
      "epoch": 25.641025641025642,
      "eval_loss": 1.28258216381073,
      "eval_runtime": 33.7988,
      "eval_samples_per_second": 11.628,
      "eval_steps_per_second": 1.479,
      "step": 5000
    }
  ],
  "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": 1.91121583104e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}