{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.998804113848362,
  "eval_steps": 500,
  "global_step": 5220,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09567089213106912,
      "grad_norm": 117.45254516601562,
      "learning_rate": 2.5e-06,
      "loss": 31905.095,
      "step": 50
    },
    {
      "epoch": 0.19134178426213824,
      "grad_norm": 71.91663360595703,
      "learning_rate": 5e-06,
      "loss": 2260638924.8,
      "step": 100
    },
    {
      "epoch": 0.2870126763932074,
      "grad_norm": 56.18342590332031,
      "learning_rate": 7.5e-06,
      "loss": 63.7996,
      "step": 150
    },
    {
      "epoch": 0.3826835685242765,
      "grad_norm": 47.14400100708008,
      "learning_rate": 1e-05,
      "loss": 12.4659,
      "step": 200
    },
    {
      "epoch": 0.4783544606553456,
      "grad_norm": 57.82217025756836,
      "learning_rate": 1.25e-05,
      "loss": 9.1207,
      "step": 250
    },
    {
      "epoch": 0.5740253527864148,
      "grad_norm": 36.61322021484375,
      "learning_rate": 1.5e-05,
      "loss": 8.5922,
      "step": 300
    },
    {
      "epoch": 0.6696962449174838,
      "grad_norm": 30.304691314697266,
      "learning_rate": 1.7500000000000002e-05,
      "loss": 7.5559,
      "step": 350
    },
    {
      "epoch": 0.765367137048553,
      "grad_norm": 24.35045051574707,
      "learning_rate": 2e-05,
      "loss": 7.2955,
      "step": 400
    },
    {
      "epoch": 0.8610380291796221,
      "grad_norm": 24.114551544189453,
      "learning_rate": 2.2499999999999998e-05,
      "loss": 7.2298,
      "step": 450
    },
    {
      "epoch": 0.9567089213106912,
      "grad_norm": 22.68138313293457,
      "learning_rate": 2.5e-05,
      "loss": 7.5636,
      "step": 500
    },
    {
      "epoch": 0.9988041138483617,
      "eval_accuracy": 0.0,
      "eval_loss": 10.570171356201172,
      "eval_normalizer": 94012.0,
      "eval_runtime": 76.7018,
      "eval_samples_per_second": 696.945,
      "eval_steps_per_second": 1.369,
      "step": 522
    },
    {
      "epoch": 1.0535756995933987,
      "grad_norm": 136.1419219970703,
      "learning_rate": 2.75e-05,
      "loss": 244134.34,
      "step": 550
    },
    {
      "epoch": 1.1492465917244679,
      "grad_norm": 83.12769317626953,
      "learning_rate": 3e-05,
      "loss": 422854923386.88,
      "step": 600
    },
    {
      "epoch": 1.244917483855537,
      "grad_norm": 75.21563720703125,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 122855.21,
      "step": 650
    },
    {
      "epoch": 1.3405883759866062,
      "grad_norm": 53.24630355834961,
      "learning_rate": 3.5000000000000004e-05,
      "loss": 8.0066,
      "step": 700
    },
    {
      "epoch": 1.4362592681176751,
      "grad_norm": 78.36145782470703,
      "learning_rate": 3.75e-05,
      "loss": 5.7671,
      "step": 750
    },
    {
      "epoch": 1.5319301602487443,
      "grad_norm": 42.857173919677734,
      "learning_rate": 4e-05,
      "loss": 5.6876,
      "step": 800
    },
    {
      "epoch": 1.6276010523798135,
      "grad_norm": 68.13167572021484,
      "learning_rate": 4.25e-05,
      "loss": 5.532,
      "step": 850
    },
    {
      "epoch": 1.7232719445108826,
      "grad_norm": 84.04206085205078,
      "learning_rate": 4.4999999999999996e-05,
      "loss": 5.8722,
      "step": 900
    },
    {
      "epoch": 1.8189428366419516,
      "grad_norm": 172.8482666015625,
      "learning_rate": 4.75e-05,
      "loss": 6.2774,
      "step": 950
    },
    {
      "epoch": 1.9146137287730207,
      "grad_norm": 71.3455810546875,
      "learning_rate": 5e-05,
      "loss": 6.8607,
      "step": 1000
    },
    {
      "epoch": 1.9988041138483617,
      "eval_accuracy": 0.0,
      "eval_loss": 12.504220962524414,
      "eval_normalizer": 93558.0,
      "eval_runtime": 76.0703,
      "eval_samples_per_second": 702.731,
      "eval_steps_per_second": 1.38,
      "step": 1044
    },
    {
      "epoch": 2.0114805070557282,
      "grad_norm": 0.0,
      "learning_rate": 5.25e-05,
      "loss": 49.308,
      "step": 1050
    },
    {
      "epoch": 2.1071513991867974,
      "grad_norm": 95.2462158203125,
      "learning_rate": 5.5e-05,
      "loss": 15104857.6,
      "step": 1100
    },
    {
      "epoch": 2.2028222913178666,
      "grad_norm": 47.490631103515625,
      "learning_rate": 5.75e-05,
      "loss": 164198594.56,
      "step": 1150
    },
    {
      "epoch": 2.2984931834489357,
      "grad_norm": 36.559452056884766,
      "learning_rate": 6e-05,
      "loss": 19.3007,
      "step": 1200
    },
    {
      "epoch": 2.394164075580005,
      "grad_norm": 71.66349029541016,
      "learning_rate": 6.25e-05,
      "loss": 4.3654,
      "step": 1250
    },
    {
      "epoch": 2.489834967711074,
      "grad_norm": 62.64774703979492,
      "learning_rate": 6.500000000000001e-05,
      "loss": 4.5691,
      "step": 1300
    },
    {
      "epoch": 2.5855058598421428,
      "grad_norm": 82.2427749633789,
      "learning_rate": 6.75e-05,
      "loss": 4.6713,
      "step": 1350
    },
    {
      "epoch": 2.6811767519732124,
      "grad_norm": 28.598220825195312,
      "learning_rate": 7.000000000000001e-05,
      "loss": 4.9054,
      "step": 1400
    },
    {
      "epoch": 2.776847644104281,
      "grad_norm": 55.590911865234375,
      "learning_rate": 7.25e-05,
      "loss": 5.3171,
      "step": 1450
    },
    {
      "epoch": 2.8725185362353503,
      "grad_norm": 52.226806640625,
      "learning_rate": 7.5e-05,
      "loss": 5.8207,
      "step": 1500
    },
    {
      "epoch": 2.9681894283664194,
      "grad_norm": 41.47724151611328,
      "learning_rate": 7.75e-05,
      "loss": 6.6854,
      "step": 1550
    },
    {
      "epoch": 2.9988041138483617,
      "eval_accuracy": 0.0,
      "eval_loss": 13.740432739257812,
      "eval_normalizer": 94216.0,
      "eval_runtime": 75.872,
      "eval_samples_per_second": 704.568,
      "eval_steps_per_second": 1.384,
      "step": 1566
    },
    {
      "epoch": 3.065056206649127,
      "grad_norm": 149.8237762451172,
      "learning_rate": 8e-05,
      "loss": 1349238.08,
      "step": 1600
    },
    {
      "epoch": 3.160727098780196,
      "grad_norm": 55.36467361450195,
      "learning_rate": 8.25e-05,
      "loss": 3558577602.56,
      "step": 1650
    },
    {
      "epoch": 3.2563979909112652,
      "grad_norm": 42.80427932739258,
      "learning_rate": 8.5e-05,
      "loss": 888.002,
      "step": 1700
    },
    {
      "epoch": 3.3520688830423344,
      "grad_norm": 113.55453491210938,
      "learning_rate": 8.75e-05,
      "loss": 4.9854,
      "step": 1750
    },
    {
      "epoch": 3.4477397751734036,
      "grad_norm": 72.55592346191406,
      "learning_rate": 8.999999999999999e-05,
      "loss": 3.7219,
      "step": 1800
    },
    {
      "epoch": 3.5434106673044727,
      "grad_norm": 91.00385284423828,
      "learning_rate": 9.25e-05,
      "loss": 4.0603,
      "step": 1850
    },
    {
      "epoch": 3.639081559435542,
      "grad_norm": 99.32891845703125,
      "learning_rate": 9.5e-05,
      "loss": 4.1079,
      "step": 1900
    },
    {
      "epoch": 3.734752451566611,
      "grad_norm": 61.96468734741211,
      "learning_rate": 9.750000000000001e-05,
      "loss": 4.5347,
      "step": 1950
    },
    {
      "epoch": 3.83042334369768,
      "grad_norm": 20.42554473876953,
      "learning_rate": 0.0001,
      "loss": 5.0272,
      "step": 2000
    },
    {
      "epoch": 3.926094235828749,
      "grad_norm": 81.02217864990234,
      "learning_rate": 0.0001025,
      "loss": 5.6779,
      "step": 2050
    },
    {
      "epoch": 3.9988041138483617,
      "eval_accuracy": 0.0,
      "eval_loss": 14.085135459899902,
      "eval_normalizer": 93665.0,
      "eval_runtime": 95.6992,
      "eval_samples_per_second": 558.594,
      "eval_steps_per_second": 1.097,
      "step": 2088
    },
    {
      "epoch": 4.0229610141114565,
      "grad_norm": 0.0,
      "learning_rate": 0.000105,
      "loss": 376.9074,
      "step": 2100
    },
    {
      "epoch": 4.118631906242526,
      "grad_norm": 52.85683822631836,
      "learning_rate": 0.0001075,
      "loss": 6231821050183.68,
      "step": 2150
    },
    {
      "epoch": 4.214302798373595,
      "grad_norm": 42.850257873535156,
      "learning_rate": 0.00011,
      "loss": 50238305.28,
      "step": 2200
    },
    {
      "epoch": 4.309973690504664,
      "grad_norm": 34.97455978393555,
      "learning_rate": 0.00011250000000000001,
      "loss": 7.5581,
      "step": 2250
    },
    {
      "epoch": 4.405644582635733,
      "grad_norm": 135.6980438232422,
      "learning_rate": 0.000115,
      "loss": 3.3287,
      "step": 2300
    },
    {
      "epoch": 4.501315474766802,
      "grad_norm": 34.392635345458984,
      "learning_rate": 0.0001175,
      "loss": 3.5607,
      "step": 2350
    },
    {
      "epoch": 4.596986366897871,
      "grad_norm": 131.91860961914062,
      "learning_rate": 0.00012,
      "loss": 3.6431,
      "step": 2400
    },
    {
      "epoch": 4.69265725902894,
      "grad_norm": 28.22834587097168,
      "learning_rate": 0.0001225,
      "loss": 4.0196,
      "step": 2450
    },
    {
      "epoch": 4.78832815116001,
      "grad_norm": 29.079853057861328,
      "learning_rate": 0.000125,
      "loss": 4.4548,
      "step": 2500
    },
    {
      "epoch": 4.8839990432910785,
      "grad_norm": 48.18919372558594,
      "learning_rate": 0.0001275,
      "loss": 5.0201,
      "step": 2550
    },
    {
      "epoch": 4.979669935422148,
      "grad_norm": 39.95207214355469,
      "learning_rate": 0.00013000000000000002,
      "loss": 6.0313,
      "step": 2600
    },
    {
      "epoch": 4.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 11.841293334960938,
      "eval_normalizer": 94313.0,
      "eval_runtime": 76.8792,
      "eval_samples_per_second": 695.337,
      "eval_steps_per_second": 1.366,
      "step": 2610
    },
    {
      "epoch": 5.076536713704855,
      "grad_norm": 37.46809005737305,
      "learning_rate": 0.00013250000000000002,
      "loss": 6866609111.04,
      "step": 2650
    },
    {
      "epoch": 5.172207605835925,
      "grad_norm": 74.59645080566406,
      "learning_rate": 0.000135,
      "loss": 106694390579.2,
      "step": 2700
    },
    {
      "epoch": 5.2678784979669935,
      "grad_norm": 31.20979881286621,
      "learning_rate": 0.0001375,
      "loss": 169.9297,
      "step": 2750
    },
    {
      "epoch": 5.363549390098063,
      "grad_norm": 73.1556396484375,
      "learning_rate": 0.00014000000000000001,
      "loss": 5.0305,
      "step": 2800
    },
    {
      "epoch": 5.459220282229132,
      "grad_norm": 54.13188552856445,
      "learning_rate": 0.0001425,
      "loss": 3.153,
      "step": 2850
    },
    {
      "epoch": 5.5548911743602005,
      "grad_norm": 101.93888092041016,
      "learning_rate": 0.000145,
      "loss": 3.4885,
      "step": 2900
    },
    {
      "epoch": 5.65056206649127,
      "grad_norm": 93.46497344970703,
      "learning_rate": 0.0001475,
      "loss": 3.5708,
      "step": 2950
    },
    {
      "epoch": 5.746232958622339,
      "grad_norm": 82.99772644042969,
      "learning_rate": 0.00015,
      "loss": 3.9849,
      "step": 3000
    },
    {
      "epoch": 5.8419038507534085,
      "grad_norm": 18.60589599609375,
      "learning_rate": 0.0001525,
      "loss": 4.5315,
      "step": 3050
    },
    {
      "epoch": 5.937574742884477,
      "grad_norm": 101.55665588378906,
      "learning_rate": 0.000155,
      "loss": 5.2747,
      "step": 3100
    },
    {
      "epoch": 5.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 10.367158889770508,
      "eval_normalizer": 93901.0,
      "eval_runtime": 76.7806,
      "eval_samples_per_second": 696.231,
      "eval_steps_per_second": 1.368,
      "step": 3132
    },
    {
      "epoch": 6.034441521167185,
      "grad_norm": 457.962646484375,
      "learning_rate": 0.0001575,
      "loss": 2423.0687,
      "step": 3150
    },
    {
      "epoch": 6.130112413298254,
      "grad_norm": 63.73720932006836,
      "learning_rate": 0.00016,
      "loss": 5.565151275904506e+17,
      "step": 3200
    },
    {
      "epoch": 6.2257833054293235,
      "grad_norm": 90.64252471923828,
      "learning_rate": 0.00016250000000000002,
      "loss": 83188.88,
      "step": 3250
    },
    {
      "epoch": 6.321454197560392,
      "grad_norm": 44.98623275756836,
      "learning_rate": 0.000165,
      "loss": 5.3464,
      "step": 3300
    },
    {
      "epoch": 6.417125089691462,
      "grad_norm": 143.51800537109375,
      "learning_rate": 0.0001675,
      "loss": 3.2073,
      "step": 3350
    },
    {
      "epoch": 6.5127959818225305,
      "grad_norm": 54.77727127075195,
      "learning_rate": 0.00017,
      "loss": 3.0689,
      "step": 3400
    },
    {
      "epoch": 6.608466873953599,
      "grad_norm": 27.115407943725586,
      "learning_rate": 0.0001725,
      "loss": 3.3064,
      "step": 3450
    },
    {
      "epoch": 6.704137766084669,
      "grad_norm": 34.8734245300293,
      "learning_rate": 0.000175,
      "loss": 3.6558,
      "step": 3500
    },
    {
      "epoch": 6.7998086582157375,
      "grad_norm": 74.51162719726562,
      "learning_rate": 0.0001775,
      "loss": 4.0926,
      "step": 3550
    },
    {
      "epoch": 6.895479550346807,
      "grad_norm": 78.53237915039062,
      "learning_rate": 0.00017999999999999998,
      "loss": 4.7036,
      "step": 3600
    },
    {
      "epoch": 6.991150442477876,
      "grad_norm": 39.22505569458008,
      "learning_rate": 0.0001825,
      "loss": 5.5472,
      "step": 3650
    },
    {
      "epoch": 6.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 11.851235389709473,
      "eval_normalizer": 94462.0,
      "eval_runtime": 75.4648,
      "eval_samples_per_second": 708.37,
      "eval_steps_per_second": 1.391,
      "step": 3654
    },
    {
      "epoch": 7.088017220760584,
      "grad_norm": 159.79464721679688,
      "learning_rate": 0.000185,
      "loss": 206454782951424.0,
      "step": 3700
    },
    {
      "epoch": 7.1836881128916525,
      "grad_norm": 48.4265022277832,
      "learning_rate": 0.0001875,
      "loss": 2275898490.88,
      "step": 3750
    },
    {
      "epoch": 7.279359005022722,
      "grad_norm": 112.71805572509766,
      "learning_rate": 0.00019,
      "loss": 11.4533,
      "step": 3800
    },
    {
      "epoch": 7.375029897153791,
      "grad_norm": 61.47443771362305,
      "learning_rate": 0.00019250000000000002,
      "loss": 3.8804,
      "step": 3850
    },
    {
      "epoch": 7.4707007892848605,
      "grad_norm": 147.22906494140625,
      "learning_rate": 0.00019500000000000002,
      "loss": 3.0733,
      "step": 3900
    },
    {
      "epoch": 7.566371681415929,
      "grad_norm": 91.8338851928711,
      "learning_rate": 0.0001975,
      "loss": 3.2563,
      "step": 3950
    },
    {
      "epoch": 7.662042573546998,
      "grad_norm": 136.98451232910156,
      "learning_rate": 0.0002,
      "loss": 3.3574,
      "step": 4000
    },
    {
      "epoch": 7.7577134656780675,
      "grad_norm": 75.63126373291016,
      "learning_rate": 0.00020250000000000002,
      "loss": 3.8024,
      "step": 4050
    },
    {
      "epoch": 7.853384357809136,
      "grad_norm": 24.159828186035156,
      "learning_rate": 0.000205,
      "loss": 4.2424,
      "step": 4100
    },
    {
      "epoch": 7.949055249940206,
      "grad_norm": 41.869728088378906,
      "learning_rate": 0.0002075,
      "loss": 4.9719,
      "step": 4150
    },
    {
      "epoch": 7.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 7.974574089050293,
      "eval_normalizer": 93833.0,
      "eval_runtime": 111.009,
      "eval_samples_per_second": 481.556,
      "eval_steps_per_second": 0.946,
      "step": 4176
    },
    {
      "epoch": 8.045922028222913,
      "grad_norm": 0.0,
      "learning_rate": 0.00021,
      "loss": 65337.845,
      "step": 4200
    },
    {
      "epoch": 8.141592920353983,
      "grad_norm": 45.79527282714844,
      "learning_rate": 0.0002125,
      "loss": 39094751068.16,
      "step": 4250
    },
    {
      "epoch": 8.237263812485052,
      "grad_norm": 62.57452392578125,
      "learning_rate": 0.000215,
      "loss": 22389.095,
      "step": 4300
    },
    {
      "epoch": 8.33293470461612,
      "grad_norm": 50.88811492919922,
      "learning_rate": 0.0002175,
      "loss": 4.5479,
      "step": 4350
    },
    {
      "epoch": 8.42860559674719,
      "grad_norm": 60.74330520629883,
      "learning_rate": 0.00022,
      "loss": 2.9879,
      "step": 4400
    },
    {
      "epoch": 8.52427648887826,
      "grad_norm": 26.475976943969727,
      "learning_rate": 0.00022250000000000001,
      "loss": 2.9984,
      "step": 4450
    },
    {
      "epoch": 8.619947381009329,
      "grad_norm": 67.60611724853516,
      "learning_rate": 0.00022500000000000002,
      "loss": 3.1279,
      "step": 4500
    },
    {
      "epoch": 8.715618273140397,
      "grad_norm": 35.15587615966797,
      "learning_rate": 0.0002275,
      "loss": 3.4657,
      "step": 4550
    },
    {
      "epoch": 8.811289165271466,
      "grad_norm": 54.12407684326172,
      "learning_rate": 0.00023,
      "loss": 3.8863,
      "step": 4600
    },
    {
      "epoch": 8.906960057402536,
      "grad_norm": 74.55989837646484,
      "learning_rate": 0.0002325,
      "loss": 4.3917,
      "step": 4650
    },
    {
      "epoch": 8.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 7.8672075271606445,
      "eval_normalizer": 93564.0,
      "eval_runtime": 75.1834,
      "eval_samples_per_second": 711.022,
      "eval_steps_per_second": 1.397,
      "step": 4698
    },
    {
      "epoch": 9.003826835685242,
      "grad_norm": 0.0,
      "learning_rate": 0.000235,
      "loss": 9.6328,
      "step": 4700
    },
    {
      "epoch": 9.099497727816312,
      "grad_norm": 73.42091369628906,
      "learning_rate": 0.0002375,
      "loss": 3504.0534,
      "step": 4750
    },
    {
      "epoch": 9.195168619947381,
      "grad_norm": 48.64082717895508,
      "learning_rate": 0.00024,
      "loss": 31552325550.08,
      "step": 4800
    },
    {
      "epoch": 9.29083951207845,
      "grad_norm": 70.7625503540039,
      "learning_rate": 0.00024249999999999999,
      "loss": 15.974,
      "step": 4850
    },
    {
      "epoch": 9.386510404209519,
      "grad_norm": 60.01846694946289,
      "learning_rate": 0.000245,
      "loss": 2.8804,
      "step": 4900
    },
    {
      "epoch": 9.482181296340588,
      "grad_norm": 63.00075912475586,
      "learning_rate": 0.0002475,
      "loss": 2.8277,
      "step": 4950
    },
    {
      "epoch": 9.577852188471658,
      "grad_norm": 58.84248733520508,
      "learning_rate": 0.00025,
      "loss": 2.9276,
      "step": 5000
    },
    {
      "epoch": 9.673523080602727,
      "grad_norm": 62.46418380737305,
      "learning_rate": 0.0002525,
      "loss": 3.2495,
      "step": 5050
    },
    {
      "epoch": 9.769193972733795,
      "grad_norm": 151.15887451171875,
      "learning_rate": 0.000255,
      "loss": 3.6513,
      "step": 5100
    },
    {
      "epoch": 9.864864864864865,
      "grad_norm": 29.21762466430664,
      "learning_rate": 0.0002575,
      "loss": 4.0727,
      "step": 5150
    },
    {
      "epoch": 9.960535756995935,
      "grad_norm": 32.33658981323242,
      "learning_rate": 0.00026000000000000003,
      "loss": 4.7651,
      "step": 5200
    },
    {
      "epoch": 9.998804113848362,
      "eval_accuracy": 0.0,
      "eval_loss": 10.32102108001709,
      "eval_normalizer": 93774.0,
      "eval_runtime": 80.1943,
      "eval_samples_per_second": 666.594,
      "eval_steps_per_second": 1.309,
      "step": 5220
    }
  ],
  "logging_steps": 50,
  "max_steps": 5220,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.18783073690837e+17,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}