{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.952153110047847,
  "eval_steps": 500,
  "global_step": 1040,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.009569377990430622,
      "grad_norm": 452.0,
      "learning_rate": 1.9230769230769234e-06,
      "loss": 52.8047,
      "step": 1
    },
    {
      "epoch": 0.04784688995215311,
      "grad_norm": 348.0,
      "learning_rate": 9.615384615384616e-06,
      "loss": 50.9309,
      "step": 5
    },
    {
      "epoch": 0.09569377990430622,
      "grad_norm": 151.0,
      "learning_rate": 1.923076923076923e-05,
      "loss": 41.6311,
      "step": 10
    },
    {
      "epoch": 0.14354066985645933,
      "grad_norm": 38.5,
      "learning_rate": 2.8846153846153845e-05,
      "loss": 30.0661,
      "step": 15
    },
    {
      "epoch": 0.19138755980861244,
      "grad_norm": 21.625,
      "learning_rate": 3.846153846153846e-05,
      "loss": 25.0223,
      "step": 20
    },
    {
      "epoch": 0.23923444976076555,
      "grad_norm": 9.5,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 22.4207,
      "step": 25
    },
    {
      "epoch": 0.28708133971291866,
      "grad_norm": 5.21875,
      "learning_rate": 5.769230769230769e-05,
      "loss": 20.4665,
      "step": 30
    },
    {
      "epoch": 0.3349282296650718,
      "grad_norm": 5.25,
      "learning_rate": 6.730769230769232e-05,
      "loss": 19.3811,
      "step": 35
    },
    {
      "epoch": 0.3827751196172249,
      "grad_norm": 7.28125,
      "learning_rate": 7.692307692307693e-05,
      "loss": 18.3511,
      "step": 40
    },
    {
      "epoch": 0.430622009569378,
      "grad_norm": 14.6875,
      "learning_rate": 8.653846153846155e-05,
      "loss": 16.8341,
      "step": 45
    },
    {
      "epoch": 0.4784688995215311,
      "grad_norm": 28.25,
      "learning_rate": 9.615384615384617e-05,
      "loss": 13.7818,
      "step": 50
    },
    {
      "epoch": 0.5263157894736842,
      "grad_norm": 28.375,
      "learning_rate": 0.00010576923076923077,
      "loss": 7.7465,
      "step": 55
    },
    {
      "epoch": 0.5741626794258373,
      "grad_norm": 8.25,
      "learning_rate": 0.00011538461538461538,
      "loss": 3.0182,
      "step": 60
    },
    {
      "epoch": 0.6220095693779905,
      "grad_norm": 6.90625,
      "learning_rate": 0.000125,
      "loss": 2.2631,
      "step": 65
    },
    {
      "epoch": 0.6698564593301436,
      "grad_norm": 16.125,
      "learning_rate": 0.00013461538461538464,
      "loss": 1.8651,
      "step": 70
    },
    {
      "epoch": 0.7177033492822966,
      "grad_norm": 0.83203125,
      "learning_rate": 0.00014423076923076924,
      "loss": 1.6482,
      "step": 75
    },
    {
      "epoch": 0.7655502392344498,
      "grad_norm": 1.3125,
      "learning_rate": 0.00015384615384615385,
      "loss": 1.4966,
      "step": 80
    },
    {
      "epoch": 0.8133971291866029,
      "grad_norm": 1.09375,
      "learning_rate": 0.00016346153846153846,
      "loss": 1.4308,
      "step": 85
    },
    {
      "epoch": 0.861244019138756,
      "grad_norm": 1.0625,
      "learning_rate": 0.0001730769230769231,
      "loss": 1.3404,
      "step": 90
    },
    {
      "epoch": 0.9090909090909091,
      "grad_norm": 1.828125,
      "learning_rate": 0.0001826923076923077,
      "loss": 1.2829,
      "step": 95
    },
    {
      "epoch": 0.9569377990430622,
      "grad_norm": 0.72265625,
      "learning_rate": 0.00019230769230769233,
      "loss": 1.2495,
      "step": 100
    },
    {
      "epoch": 0.9952153110047847,
      "eval_loss": 2.6542556285858154,
      "eval_runtime": 0.2797,
      "eval_samples_per_second": 35.754,
      "eval_steps_per_second": 3.575,
      "step": 104
    },
    {
      "epoch": 1.0047846889952152,
      "grad_norm": 0.90234375,
      "learning_rate": 0.0001999994367286727,
      "loss": 1.2336,
      "step": 105
    },
    {
      "epoch": 1.0526315789473684,
      "grad_norm": 2.046875,
      "learning_rate": 0.00019997972289848503,
      "loss": 1.1928,
      "step": 110
    },
    {
      "epoch": 1.1004784688995215,
      "grad_norm": 9.25,
      "learning_rate": 0.00019993185184710165,
      "loss": 1.1702,
      "step": 115
    },
    {
      "epoch": 1.1483253588516746,
      "grad_norm": 1.1015625,
      "learning_rate": 0.00019985583705641418,
      "loss": 1.1684,
      "step": 120
    },
    {
      "epoch": 1.1961722488038278,
      "grad_norm": 2.859375,
      "learning_rate": 0.00019975169993441627,
      "loss": 1.125,
      "step": 125
    },
    {
      "epoch": 1.244019138755981,
      "grad_norm": 0.76171875,
      "learning_rate": 0.00019961946980917456,
      "loss": 1.1078,
      "step": 130
    },
    {
      "epoch": 1.291866028708134,
      "grad_norm": 1.546875,
      "learning_rate": 0.0001994591839205691,
      "loss": 1.1369,
      "step": 135
    },
    {
      "epoch": 1.339712918660287,
      "grad_norm": 3.328125,
      "learning_rate": 0.0001992708874098054,
      "loss": 1.144,
      "step": 140
    },
    {
      "epoch": 1.38755980861244,
      "grad_norm": 2.53125,
      "learning_rate": 0.00019905463330670143,
      "loss": 1.1301,
      "step": 145
    },
    {
      "epoch": 1.4354066985645932,
      "grad_norm": 2.890625,
      "learning_rate": 0.0001988104825147528,
      "loss": 1.146,
      "step": 150
    },
    {
      "epoch": 1.4832535885167464,
      "grad_norm": 2.453125,
      "learning_rate": 0.0001985385037939806,
      "loss": 1.0953,
      "step": 155
    },
    {
      "epoch": 1.5311004784688995,
      "grad_norm": 2.1875,
      "learning_rate": 0.00019823877374156647,
      "loss": 1.0699,
      "step": 160
    },
    {
      "epoch": 1.5789473684210527,
      "grad_norm": 1.984375,
      "learning_rate": 0.00019791137677028082,
      "loss": 1.0699,
      "step": 165
    },
    {
      "epoch": 1.6267942583732058,
      "grad_norm": 2.296875,
      "learning_rate": 0.00019755640508470942,
      "loss": 1.0545,
      "step": 170
    },
    {
      "epoch": 1.674641148325359,
      "grad_norm": 1.8828125,
      "learning_rate": 0.00019717395865528602,
      "loss": 1.0404,
      "step": 175
    },
    {
      "epoch": 1.722488038277512,
      "grad_norm": 2.015625,
      "learning_rate": 0.00019676414519013781,
      "loss": 1.0512,
      "step": 180
    },
    {
      "epoch": 1.7703349282296652,
      "grad_norm": 6.75,
      "learning_rate": 0.00019632708010475165,
      "loss": 1.0382,
      "step": 185
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 0.8046875,
      "learning_rate": 0.00019586288648946947,
      "loss": 1.0383,
      "step": 190
    },
    {
      "epoch": 1.8660287081339713,
      "grad_norm": 0.875,
      "learning_rate": 0.0001953716950748227,
      "loss": 1.0356,
      "step": 195
    },
    {
      "epoch": 1.9138755980861244,
      "grad_norm": 6.03125,
      "learning_rate": 0.00019485364419471454,
      "loss": 1.0354,
      "step": 200
    },
    {
      "epoch": 1.9617224880382775,
      "grad_norm": 1.640625,
      "learning_rate": 0.0001943088797474612,
      "loss": 1.0258,
      "step": 205
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.5231521129608154,
      "eval_runtime": 0.2369,
      "eval_samples_per_second": 42.205,
      "eval_steps_per_second": 4.22,
      "step": 209
    },
    {
      "epoch": 2.0095693779904304,
      "grad_norm": 1.578125,
      "learning_rate": 0.00019373755515470254,
      "loss": 1.0034,
      "step": 210
    },
    {
      "epoch": 2.0574162679425836,
      "grad_norm": 1.5703125,
      "learning_rate": 0.00019313983131819407,
      "loss": 0.9855,
      "step": 215
    },
    {
      "epoch": 2.1052631578947367,
      "grad_norm": 4.4375,
      "learning_rate": 0.00019251587657449236,
      "loss": 0.9769,
      "step": 220
    },
    {
      "epoch": 2.15311004784689,
      "grad_norm": 1.8359375,
      "learning_rate": 0.0001918658666475465,
      "loss": 0.9829,
      "step": 225
    },
    {
      "epoch": 2.200956937799043,
      "grad_norm": 0.7421875,
      "learning_rate": 0.00019118998459920902,
      "loss": 0.9843,
      "step": 230
    },
    {
      "epoch": 2.248803827751196,
      "grad_norm": 2.578125,
      "learning_rate": 0.0001904884207776804,
      "loss": 0.9876,
      "step": 235
    },
    {
      "epoch": 2.2966507177033493,
      "grad_norm": 1.5390625,
      "learning_rate": 0.0001897613727639014,
      "loss": 0.9713,
      "step": 240
    },
    {
      "epoch": 2.3444976076555024,
      "grad_norm": 1.9609375,
      "learning_rate": 0.00018900904531590846,
      "loss": 0.9706,
      "step": 245
    },
    {
      "epoch": 2.3923444976076556,
      "grad_norm": 2.453125,
      "learning_rate": 0.0001882316503111678,
      "loss": 0.9635,
      "step": 250
    },
    {
      "epoch": 2.4401913875598087,
      "grad_norm": 1.6484375,
      "learning_rate": 0.00018742940668690464,
      "loss": 0.9608,
      "step": 255
    },
    {
      "epoch": 2.488038277511962,
      "grad_norm": 1.46875,
      "learning_rate": 0.00018660254037844388,
      "loss": 0.9498,
      "step": 260
    },
    {
      "epoch": 2.535885167464115,
      "grad_norm": 4.34375,
      "learning_rate": 0.00018575128425558023,
      "loss": 0.9539,
      "step": 265
    },
    {
      "epoch": 2.583732057416268,
      "grad_norm": 2.90625,
      "learning_rate": 0.00018487587805699526,
      "loss": 0.96,
      "step": 270
    },
    {
      "epoch": 2.6315789473684212,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0001839765683227398,
      "loss": 0.9431,
      "step": 275
    },
    {
      "epoch": 2.679425837320574,
      "grad_norm": 4.375,
      "learning_rate": 0.00018305360832480117,
      "loss": 0.9457,
      "step": 280
    },
    {
      "epoch": 2.7272727272727275,
      "grad_norm": 11.1875,
      "learning_rate": 0.00018210725799577439,
      "loss": 0.9594,
      "step": 285
    },
    {
      "epoch": 2.77511961722488,
      "grad_norm": 1.078125,
      "learning_rate": 0.00018113778385565733,
      "loss": 0.9542,
      "step": 290
    },
    {
      "epoch": 2.8229665071770333,
      "grad_norm": 3.734375,
      "learning_rate": 0.00018014545893679115,
      "loss": 0.9407,
      "step": 295
    },
    {
      "epoch": 2.8708133971291865,
      "grad_norm": 1.9765625,
      "learning_rate": 0.0001791305627069662,
      "loss": 0.9418,
      "step": 300
    },
    {
      "epoch": 2.9186602870813396,
      "grad_norm": 0.88671875,
      "learning_rate": 0.00017809338099071577,
      "loss": 0.9357,
      "step": 305
    },
    {
      "epoch": 2.9665071770334928,
      "grad_norm": 1.046875,
      "learning_rate": 0.00017703420588881946,
      "loss": 0.9351,
      "step": 310
    },
    {
      "epoch": 2.9952153110047846,
      "eval_loss": 2.466839551925659,
      "eval_runtime": 0.257,
      "eval_samples_per_second": 38.912,
      "eval_steps_per_second": 3.891,
      "step": 313
    },
    {
      "epoch": 3.014354066985646,
      "grad_norm": 1.7109375,
      "learning_rate": 0.0001759533356960391,
      "loss": 0.9283,
      "step": 315
    },
    {
      "epoch": 3.062200956937799,
      "grad_norm": 2.0625,
      "learning_rate": 0.00017485107481711012,
      "loss": 0.9187,
      "step": 320
    },
    {
      "epoch": 3.110047846889952,
      "grad_norm": 1.8125,
      "learning_rate": 0.0001737277336810124,
      "loss": 0.9121,
      "step": 325
    },
    {
      "epoch": 3.1578947368421053,
      "grad_norm": 3.6875,
      "learning_rate": 0.00017258362865354426,
      "loss": 0.9481,
      "step": 330
    },
    {
      "epoch": 3.2057416267942584,
      "grad_norm": 3.84375,
      "learning_rate": 0.00017141908194822446,
      "loss": 0.9215,
      "step": 335
    },
    {
      "epoch": 3.2535885167464116,
      "grad_norm": 8.625,
      "learning_rate": 0.00017023442153554777,
      "loss": 0.9082,
      "step": 340
    },
    {
      "epoch": 3.3014354066985647,
      "grad_norm": 0.94140625,
      "learning_rate": 0.00016902998105061844,
      "loss": 0.9035,
      "step": 345
    },
    {
      "epoch": 3.349282296650718,
      "grad_norm": 0.9921875,
      "learning_rate": 0.0001678060996991891,
      "loss": 0.9084,
      "step": 350
    },
    {
      "epoch": 3.397129186602871,
      "grad_norm": 1.1484375,
      "learning_rate": 0.00016656312216213034,
      "loss": 0.8981,
      "step": 355
    },
    {
      "epoch": 3.444976076555024,
      "grad_norm": 1.0859375,
      "learning_rate": 0.0001653013984983585,
      "loss": 0.9024,
      "step": 360
    },
    {
      "epoch": 3.492822966507177,
      "grad_norm": 0.80078125,
      "learning_rate": 0.00016402128404624882,
      "loss": 0.9027,
      "step": 365
    },
    {
      "epoch": 3.5406698564593304,
      "grad_norm": 0.8125,
      "learning_rate": 0.00016272313932356162,
      "loss": 0.8924,
      "step": 370
    },
    {
      "epoch": 3.588516746411483,
      "grad_norm": 0.99609375,
      "learning_rate": 0.0001614073299259101,
      "loss": 0.8943,
      "step": 375
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 1.6796875,
      "learning_rate": 0.0001600742264237979,
      "loss": 0.9152,
      "step": 380
    },
    {
      "epoch": 3.6842105263157894,
      "grad_norm": 1.9375,
      "learning_rate": 0.0001587242042582554,
      "loss": 0.9076,
      "step": 385
    },
    {
      "epoch": 3.7320574162679425,
      "grad_norm": 1.859375,
      "learning_rate": 0.0001573576436351046,
      "loss": 0.8935,
      "step": 390
    },
    {
      "epoch": 3.7799043062200957,
      "grad_norm": 1.1328125,
      "learning_rate": 0.00015597492941788222,
      "loss": 0.8895,
      "step": 395
    },
    {
      "epoch": 3.827751196172249,
      "grad_norm": 0.89453125,
      "learning_rate": 0.00015457645101945046,
      "loss": 0.8767,
      "step": 400
    },
    {
      "epoch": 3.875598086124402,
      "grad_norm": 3.609375,
      "learning_rate": 0.00015316260229232727,
      "loss": 0.8795,
      "step": 405
    },
    {
      "epoch": 3.923444976076555,
      "grad_norm": 2.40625,
      "learning_rate": 0.00015173378141776568,
      "loss": 0.8847,
      "step": 410
    },
    {
      "epoch": 3.971291866028708,
      "grad_norm": 4.0,
      "learning_rate": 0.00015029039079361448,
      "loss": 0.8914,
      "step": 415
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.4789059162139893,
      "eval_runtime": 0.2403,
      "eval_samples_per_second": 41.62,
      "eval_steps_per_second": 4.162,
      "step": 418
    },
    {
      "epoch": 4.019138755980861,
      "grad_norm": 2.28125,
      "learning_rate": 0.00014883283692099112,
      "loss": 0.8795,
      "step": 420
    },
    {
      "epoch": 4.0669856459330145,
      "grad_norm": 1.890625,
      "learning_rate": 0.00014736153028979893,
      "loss": 0.8598,
      "step": 425
    },
    {
      "epoch": 4.114832535885167,
      "grad_norm": 2.984375,
      "learning_rate": 0.00014587688526312143,
      "loss": 0.8533,
      "step": 430
    },
    {
      "epoch": 4.162679425837321,
      "grad_norm": 2.453125,
      "learning_rate": 0.00014437931996052518,
      "loss": 0.8544,
      "step": 435
    },
    {
      "epoch": 4.2105263157894735,
      "grad_norm": 8.4375,
      "learning_rate": 0.00014286925614030542,
      "loss": 0.8635,
      "step": 440
    },
    {
      "epoch": 4.258373205741627,
      "grad_norm": 4.34375,
      "learning_rate": 0.00014134711908070631,
      "loss": 0.8664,
      "step": 445
    },
    {
      "epoch": 4.30622009569378,
      "grad_norm": 0.83984375,
      "learning_rate": 0.0001398133374601501,
      "loss": 0.8567,
      "step": 450
    },
    {
      "epoch": 4.354066985645933,
      "grad_norm": 0.73828125,
      "learning_rate": 0.000138268343236509,
      "loss": 0.8579,
      "step": 455
    },
    {
      "epoch": 4.401913875598086,
      "grad_norm": 1.3203125,
      "learning_rate": 0.00013671257152545277,
      "loss": 0.8562,
      "step": 460
    },
    {
      "epoch": 4.44976076555024,
      "grad_norm": 1.140625,
      "learning_rate": 0.00013514646047790775,
      "loss": 0.8527,
      "step": 465
    },
    {
      "epoch": 4.497607655502392,
      "grad_norm": 0.80078125,
      "learning_rate": 0.0001335704511566605,
      "loss": 0.8474,
      "step": 470
    },
    {
      "epoch": 4.545454545454545,
      "grad_norm": 2.625,
      "learning_rate": 0.00013198498741214166,
      "loss": 0.856,
      "step": 475
    },
    {
      "epoch": 4.5933014354066986,
      "grad_norm": 0.91015625,
      "learning_rate": 0.0001303905157574247,
      "loss": 0.8586,
      "step": 480
    },
    {
      "epoch": 4.641148325358852,
      "grad_norm": 1.28125,
      "learning_rate": 0.00012878748524247462,
      "loss": 0.8519,
      "step": 485
    },
    {
      "epoch": 4.688995215311005,
      "grad_norm": 1.609375,
      "learning_rate": 0.00012717634732768243,
      "loss": 0.8483,
      "step": 490
    },
    {
      "epoch": 4.7368421052631575,
      "grad_norm": 1.6171875,
      "learning_rate": 0.0001255575557567207,
      "loss": 0.8663,
      "step": 495
    },
    {
      "epoch": 4.784688995215311,
      "grad_norm": 0.890625,
      "learning_rate": 0.0001239315664287558,
      "loss": 0.8684,
      "step": 500
    },
    {
      "epoch": 4.832535885167464,
      "grad_norm": 1.3828125,
      "learning_rate": 0.00012229883727005365,
      "loss": 0.8391,
      "step": 505
    },
    {
      "epoch": 4.880382775119617,
      "grad_norm": 1.21875,
      "learning_rate": 0.00012065982810501404,
      "loss": 0.8491,
      "step": 510
    },
    {
      "epoch": 4.92822966507177,
      "grad_norm": 3.765625,
      "learning_rate": 0.00011901500052667068,
      "loss": 0.8435,
      "step": 515
    },
    {
      "epoch": 4.976076555023924,
      "grad_norm": 6.75,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.8487,
      "step": 520
    },
    {
      "epoch": 4.9952153110047846,
      "eval_loss": 2.475181818008423,
      "eval_runtime": 0.2635,
      "eval_samples_per_second": 37.955,
      "eval_steps_per_second": 3.796,
      "step": 522
    },
    {
      "epoch": 5.023923444976076,
      "grad_norm": 6.28125,
      "learning_rate": 0.00011570974456492678,
      "loss": 0.845,
      "step": 525
    },
    {
      "epoch": 5.07177033492823,
      "grad_norm": 12.1875,
      "learning_rate": 0.00011405024703850929,
      "loss": 0.8426,
      "step": 530
    },
    {
      "epoch": 5.119617224880383,
      "grad_norm": 7.71875,
      "learning_rate": 0.00011238679255059752,
      "loss": 0.8348,
      "step": 535
    },
    {
      "epoch": 5.167464114832536,
      "grad_norm": 6.3125,
      "learning_rate": 0.00011071984957874479,
      "loss": 0.829,
      "step": 540
    },
    {
      "epoch": 5.215311004784689,
      "grad_norm": 0.578125,
      "learning_rate": 0.0001090498875829638,
      "loss": 0.8233,
      "step": 545
    },
    {
      "epoch": 5.2631578947368425,
      "grad_norm": 2.03125,
      "learning_rate": 0.00010737737687351284,
      "loss": 0.8409,
      "step": 550
    },
    {
      "epoch": 5.311004784688995,
      "grad_norm": 0.984375,
      "learning_rate": 0.00010570278847844275,
      "loss": 0.8217,
      "step": 555
    },
    {
      "epoch": 5.358851674641148,
      "grad_norm": 1.4609375,
      "learning_rate": 0.00010402659401094152,
      "loss": 0.8108,
      "step": 560
    },
    {
      "epoch": 5.4066985645933014,
      "grad_norm": 1.2421875,
      "learning_rate": 0.00010234926553651422,
      "loss": 0.8351,
      "step": 565
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 1.140625,
      "learning_rate": 0.00010067127544003563,
      "loss": 0.8194,
      "step": 570
    },
    {
      "epoch": 5.502392344497608,
      "grad_norm": 0.98046875,
      "learning_rate": 9.899309629271246e-05,
      "loss": 0.8161,
      "step": 575
    },
    {
      "epoch": 5.55023923444976,
      "grad_norm": 1.1875,
      "learning_rate": 9.73152007189939e-05,
      "loss": 0.8156,
      "step": 580
    },
    {
      "epoch": 5.598086124401914,
      "grad_norm": 1.4375,
      "learning_rate": 9.563806126346642e-05,
      "loss": 0.8183,
      "step": 585
    },
    {
      "epoch": 5.645933014354067,
      "grad_norm": 0.8359375,
      "learning_rate": 9.396215025777139e-05,
      "loss": 0.8079,
      "step": 590
    },
    {
      "epoch": 5.69377990430622,
      "grad_norm": 1.234375,
      "learning_rate": 9.22879396875828e-05,
      "loss": 0.821,
      "step": 595
    },
    {
      "epoch": 5.741626794258373,
      "grad_norm": 0.85546875,
      "learning_rate": 9.061590105968208e-05,
      "loss": 0.8078,
      "step": 600
    },
    {
      "epoch": 5.7894736842105265,
      "grad_norm": 1.25,
      "learning_rate": 8.894650526916803e-05,
      "loss": 0.8042,
      "step": 605
    },
    {
      "epoch": 5.837320574162679,
      "grad_norm": 1.046875,
      "learning_rate": 8.728022246683894e-05,
      "loss": 0.8298,
      "step": 610
    },
    {
      "epoch": 5.885167464114833,
      "grad_norm": 2.34375,
      "learning_rate": 8.561752192678443e-05,
      "loss": 0.8122,
      "step": 615
    },
    {
      "epoch": 5.9330143540669855,
      "grad_norm": 12.75,
      "learning_rate": 8.395887191422397e-05,
      "loss": 0.819,
      "step": 620
    },
    {
      "epoch": 5.980861244019139,
      "grad_norm": 1.5703125,
      "learning_rate": 8.23047395536298e-05,
      "loss": 0.8222,
      "step": 625
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.4928200244903564,
      "eval_runtime": 0.2377,
      "eval_samples_per_second": 42.076,
      "eval_steps_per_second": 4.208,
      "step": 627
    },
    {
      "epoch": 6.028708133971292,
      "grad_norm": 2.8125,
      "learning_rate": 8.065559069717088e-05,
      "loss": 0.8044,
      "step": 630
    },
    {
      "epoch": 6.076555023923445,
      "grad_norm": 12.0,
      "learning_rate": 7.901188979351526e-05,
      "loss": 0.8055,
      "step": 635
    },
    {
      "epoch": 6.124401913875598,
      "grad_norm": 1.46875,
      "learning_rate": 7.73740997570278e-05,
      "loss": 0.7972,
      "step": 640
    },
    {
      "epoch": 6.172248803827751,
      "grad_norm": 5.34375,
      "learning_rate": 7.574268183739989e-05,
      "loss": 0.8078,
      "step": 645
    },
    {
      "epoch": 6.220095693779904,
      "grad_norm": 4.5,
      "learning_rate": 7.411809548974792e-05,
      "loss": 0.7981,
      "step": 650
    },
    {
      "epoch": 6.267942583732057,
      "grad_norm": 1.1328125,
      "learning_rate": 7.250079824521743e-05,
      "loss": 0.8073,
      "step": 655
    },
    {
      "epoch": 6.315789473684211,
      "grad_norm": 5.21875,
      "learning_rate": 7.089124558212871e-05,
      "loss": 0.8028,
      "step": 660
    },
    {
      "epoch": 6.363636363636363,
      "grad_norm": 6.84375,
      "learning_rate": 6.928989079770094e-05,
      "loss": 0.7954,
      "step": 665
    },
    {
      "epoch": 6.411483253588517,
      "grad_norm": 1.1640625,
      "learning_rate": 6.769718488039023e-05,
      "loss": 0.7899,
      "step": 670
    },
    {
      "epoch": 6.45933014354067,
      "grad_norm": 7.21875,
      "learning_rate": 6.611357638287823e-05,
      "loss": 0.8025,
      "step": 675
    },
    {
      "epoch": 6.507177033492823,
      "grad_norm": 5.1875,
      "learning_rate": 6.453951129574644e-05,
      "loss": 0.7917,
      "step": 680
    },
    {
      "epoch": 6.555023923444976,
      "grad_norm": 0.7421875,
      "learning_rate": 6.297543292187215e-05,
      "loss": 0.7903,
      "step": 685
    },
    {
      "epoch": 6.6028708133971294,
      "grad_norm": 2.390625,
      "learning_rate": 6.142178175158149e-05,
      "loss": 0.7894,
      "step": 690
    },
    {
      "epoch": 6.650717703349282,
      "grad_norm": 2.46875,
      "learning_rate": 5.9878995338594224e-05,
      "loss": 0.7887,
      "step": 695
    },
    {
      "epoch": 6.698564593301436,
      "grad_norm": 0.52734375,
      "learning_rate": 5.834750817679606e-05,
      "loss": 0.7866,
      "step": 700
    },
    {
      "epoch": 6.746411483253588,
      "grad_norm": 1.140625,
      "learning_rate": 5.682775157787213e-05,
      "loss": 0.7845,
      "step": 705
    },
    {
      "epoch": 6.794258373205742,
      "grad_norm": 0.9296875,
      "learning_rate": 5.5320153549837415e-05,
      "loss": 0.7874,
      "step": 710
    },
    {
      "epoch": 6.842105263157895,
      "grad_norm": 0.98046875,
      "learning_rate": 5.382513867649663e-05,
      "loss": 0.7865,
      "step": 715
    },
    {
      "epoch": 6.889952153110048,
      "grad_norm": 0.8515625,
      "learning_rate": 5.234312799786921e-05,
      "loss": 0.7866,
      "step": 720
    },
    {
      "epoch": 6.937799043062201,
      "grad_norm": 3.40625,
      "learning_rate": 5.087453889161229e-05,
      "loss": 0.7742,
      "step": 725
    },
    {
      "epoch": 6.985645933014354,
      "grad_norm": 0.546875,
      "learning_rate": 4.9419784955474524e-05,
      "loss": 0.7746,
      "step": 730
    },
    {
      "epoch": 6.9952153110047846,
      "eval_loss": 2.4925007820129395,
      "eval_runtime": 0.2584,
      "eval_samples_per_second": 38.704,
      "eval_steps_per_second": 3.87,
      "step": 731
    },
    {
      "epoch": 7.033492822966507,
      "grad_norm": 0.95703125,
      "learning_rate": 4.797927589081509e-05,
      "loss": 0.7674,
      "step": 735
    },
    {
      "epoch": 7.08133971291866,
      "grad_norm": 0.5,
      "learning_rate": 4.6553417387219886e-05,
      "loss": 0.7691,
      "step": 740
    },
    {
      "epoch": 7.1291866028708135,
      "grad_norm": 0.5234375,
      "learning_rate": 4.514261100824709e-05,
      "loss": 0.7602,
      "step": 745
    },
    {
      "epoch": 7.177033492822966,
      "grad_norm": 0.41015625,
      "learning_rate": 4.374725407833532e-05,
      "loss": 0.7656,
      "step": 750
    },
    {
      "epoch": 7.22488038277512,
      "grad_norm": 0.6171875,
      "learning_rate": 4.236773957090548e-05,
      "loss": 0.7673,
      "step": 755
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 0.53515625,
      "learning_rate": 4.100445599768774e-05,
      "loss": 0.7712,
      "step": 760
    },
    {
      "epoch": 7.320574162679426,
      "grad_norm": 0.546875,
      "learning_rate": 3.96577872993053e-05,
      "loss": 0.7658,
      "step": 765
    },
    {
      "epoch": 7.368421052631579,
      "grad_norm": 0.671875,
      "learning_rate": 3.832811273714569e-05,
      "loss": 0.7612,
      "step": 770
    },
    {
      "epoch": 7.416267942583732,
      "grad_norm": 0.90234375,
      "learning_rate": 3.701580678654925e-05,
      "loss": 0.7695,
      "step": 775
    },
    {
      "epoch": 7.464114832535885,
      "grad_norm": 0.54296875,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.7702,
      "step": 780
    },
    {
      "epoch": 7.511961722488039,
      "grad_norm": 0.494140625,
      "learning_rate": 3.4444774059770536e-05,
      "loss": 0.7654,
      "step": 785
    },
    {
      "epoch": 7.559808612440191,
      "grad_norm": 0.61328125,
      "learning_rate": 3.318677136178228e-05,
      "loss": 0.763,
      "step": 790
    },
    {
      "epoch": 7.607655502392344,
      "grad_norm": 0.7265625,
      "learning_rate": 3.1947585227823394e-05,
      "loss": 0.772,
      "step": 795
    },
    {
      "epoch": 7.655502392344498,
      "grad_norm": 0.396484375,
      "learning_rate": 3.072756464904006e-05,
      "loss": 0.7763,
      "step": 800
    },
    {
      "epoch": 7.703349282296651,
      "grad_norm": 0.400390625,
      "learning_rate": 2.9527053218996037e-05,
      "loss": 0.7663,
      "step": 805
    },
    {
      "epoch": 7.751196172248804,
      "grad_norm": 0.427734375,
      "learning_rate": 2.8346389036906828e-05,
      "loss": 0.7635,
      "step": 810
    },
    {
      "epoch": 7.7990430622009566,
      "grad_norm": 0.466796875,
      "learning_rate": 2.7185904612421176e-05,
      "loss": 0.7679,
      "step": 815
    },
    {
      "epoch": 7.84688995215311,
      "grad_norm": 0.61328125,
      "learning_rate": 2.6045926771976303e-05,
      "loss": 0.7752,
      "step": 820
    },
    {
      "epoch": 7.894736842105263,
      "grad_norm": 0.462890625,
      "learning_rate": 2.492677656675414e-05,
      "loss": 0.7733,
      "step": 825
    },
    {
      "epoch": 7.942583732057416,
      "grad_norm": 0.392578125,
      "learning_rate": 2.382876918226409e-05,
      "loss": 0.7713,
      "step": 830
    },
    {
      "epoch": 7.990430622009569,
      "grad_norm": 0.5078125,
      "learning_rate": 2.2752213849577188e-05,
      "loss": 0.7644,
      "step": 835
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.5051019191741943,
      "eval_runtime": 0.2382,
      "eval_samples_per_second": 41.986,
      "eval_steps_per_second": 4.199,
      "step": 836
    },
    {
      "epoch": 8.038277511961722,
      "grad_norm": 0.400390625,
      "learning_rate": 2.1697413758237784e-05,
      "loss": 0.7573,
      "step": 840
    },
    {
      "epoch": 8.086124401913876,
      "grad_norm": 0.451171875,
      "learning_rate": 2.0664665970876496e-05,
      "loss": 0.7533,
      "step": 845
    },
    {
      "epoch": 8.133971291866029,
      "grad_norm": 0.494140625,
      "learning_rate": 1.965426133954854e-05,
      "loss": 0.7638,
      "step": 850
    },
    {
      "epoch": 8.181818181818182,
      "grad_norm": 0.55078125,
      "learning_rate": 1.8666484423821373e-05,
      "loss": 0.7594,
      "step": 855
    },
    {
      "epoch": 8.229665071770334,
      "grad_norm": 0.6015625,
      "learning_rate": 1.7701613410634365e-05,
      "loss": 0.758,
      "step": 860
    },
    {
      "epoch": 8.277511961722489,
      "grad_norm": 0.55859375,
      "learning_rate": 1.6759920035953093e-05,
      "loss": 0.7619,
      "step": 865
    },
    {
      "epoch": 8.325358851674642,
      "grad_norm": 0.3828125,
      "learning_rate": 1.584166950824061e-05,
      "loss": 0.7578,
      "step": 870
    },
    {
      "epoch": 8.373205741626794,
      "grad_norm": 0.380859375,
      "learning_rate": 1.4947120433767047e-05,
      "loss": 0.7544,
      "step": 875
    },
    {
      "epoch": 8.421052631578947,
      "grad_norm": 0.388671875,
      "learning_rate": 1.4076524743778319e-05,
      "loss": 0.7606,
      "step": 880
    },
    {
      "epoch": 8.4688995215311,
      "grad_norm": 0.39453125,
      "learning_rate": 1.3230127623545064e-05,
      "loss": 0.7512,
      "step": 885
    },
    {
      "epoch": 8.516746411483254,
      "grad_norm": 0.439453125,
      "learning_rate": 1.2408167443311214e-05,
      "loss": 0.7481,
      "step": 890
    },
    {
      "epoch": 8.564593301435407,
      "grad_norm": 0.53125,
      "learning_rate": 1.1610875691161915e-05,
      "loss": 0.7625,
      "step": 895
    },
    {
      "epoch": 8.61244019138756,
      "grad_norm": 0.396484375,
      "learning_rate": 1.083847690782972e-05,
      "loss": 0.7542,
      "step": 900
    },
    {
      "epoch": 8.660287081339712,
      "grad_norm": 0.388671875,
      "learning_rate": 1.0091188623457415e-05,
      "loss": 0.7658,
      "step": 905
    },
    {
      "epoch": 8.708133971291867,
      "grad_norm": 0.451171875,
      "learning_rate": 9.369221296335006e-06,
      "loss": 0.7581,
      "step": 910
    },
    {
      "epoch": 8.75598086124402,
      "grad_norm": 0.412109375,
      "learning_rate": 8.672778253628621e-06,
      "loss": 0.7485,
      "step": 915
    },
    {
      "epoch": 8.803827751196172,
      "grad_norm": 0.400390625,
      "learning_rate": 8.002055634117578e-06,
      "loss": 0.7544,
      "step": 920
    },
    {
      "epoch": 8.851674641148325,
      "grad_norm": 0.373046875,
      "learning_rate": 7.357242332955916e-06,
      "loss": 0.76,
      "step": 925
    },
    {
      "epoch": 8.89952153110048,
      "grad_norm": 0.498046875,
      "learning_rate": 6.738519948473976e-06,
      "loss": 0.7572,
      "step": 930
    },
    {
      "epoch": 8.947368421052632,
      "grad_norm": 0.388671875,
      "learning_rate": 6.146062731035129e-06,
      "loss": 0.7559,
      "step": 935
    },
    {
      "epoch": 8.995215311004785,
      "grad_norm": 0.373046875,
      "learning_rate": 5.580037533961546e-06,
      "loss": 0.7578,
      "step": 940
    },
    {
      "epoch": 8.995215311004785,
      "eval_loss": 2.52113676071167,
      "eval_runtime": 0.2561,
      "eval_samples_per_second": 39.042,
      "eval_steps_per_second": 3.904,
      "step": 940
    },
    {
      "epoch": 9.043062200956937,
      "grad_norm": 0.416015625,
      "learning_rate": 5.040603766543594e-06,
      "loss": 0.7518,
      "step": 945
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 0.375,
      "learning_rate": 4.527913349145441e-06,
      "loss": 0.7595,
      "step": 950
    },
    {
      "epoch": 9.138755980861244,
      "grad_norm": 0.435546875,
      "learning_rate": 4.042110670419763e-06,
      "loss": 0.7471,
      "step": 955
    },
    {
      "epoch": 9.186602870813397,
      "grad_norm": 0.3984375,
      "learning_rate": 3.5833325466437694e-06,
      "loss": 0.7538,
      "step": 960
    },
    {
      "epoch": 9.23444976076555,
      "grad_norm": 0.412109375,
      "learning_rate": 3.1517081831876737e-06,
      "loss": 0.75,
      "step": 965
    },
    {
      "epoch": 9.282296650717702,
      "grad_norm": 0.37109375,
      "learning_rate": 2.7473591381266708e-06,
      "loss": 0.7547,
      "step": 970
    },
    {
      "epoch": 9.330143540669857,
      "grad_norm": 0.384765625,
      "learning_rate": 2.3703992880066638e-06,
      "loss": 0.764,
      "step": 975
    },
    {
      "epoch": 9.37799043062201,
      "grad_norm": 0.38671875,
      "learning_rate": 2.0209347957732328e-06,
      "loss": 0.7584,
      "step": 980
    },
    {
      "epoch": 9.425837320574162,
      "grad_norm": 0.390625,
      "learning_rate": 1.6990640808730696e-06,
      "loss": 0.7524,
      "step": 985
    },
    {
      "epoch": 9.473684210526315,
      "grad_norm": 0.412109375,
      "learning_rate": 1.404877791536141e-06,
      "loss": 0.7543,
      "step": 990
    },
    {
      "epoch": 9.52153110047847,
      "grad_norm": 0.50390625,
      "learning_rate": 1.1384587792465872e-06,
      "loss": 0.7579,
      "step": 995
    },
    {
      "epoch": 9.569377990430622,
      "grad_norm": 0.380859375,
      "learning_rate": 8.998820754091531e-07,
      "loss": 0.7534,
      "step": 1000
    },
    {
      "epoch": 9.617224880382775,
      "grad_norm": 0.3828125,
      "learning_rate": 6.892148702183133e-07,
      "loss": 0.7522,
      "step": 1005
    },
    {
      "epoch": 9.665071770334928,
      "grad_norm": 0.462890625,
      "learning_rate": 5.065164937354428e-07,
      "loss": 0.7598,
      "step": 1010
    },
    {
      "epoch": 9.712918660287082,
      "grad_norm": 0.37109375,
      "learning_rate": 3.5183839917972697e-07,
      "loss": 0.7556,
      "step": 1015
    },
    {
      "epoch": 9.760765550239235,
      "grad_norm": 0.376953125,
      "learning_rate": 2.2522414843748618e-07,
      "loss": 0.7503,
      "step": 1020
    },
    {
      "epoch": 9.808612440191387,
      "grad_norm": 0.375,
      "learning_rate": 1.2670939979384512e-07,
      "loss": 0.7479,
      "step": 1025
    },
    {
      "epoch": 9.85645933014354,
      "grad_norm": 0.41015625,
      "learning_rate": 5.632189789027687e-08,
      "loss": 0.7581,
      "step": 1030
    },
    {
      "epoch": 9.904306220095695,
      "grad_norm": 0.376953125,
      "learning_rate": 1.4081465910975588e-08,
      "loss": 0.7526,
      "step": 1035
    },
    {
      "epoch": 9.952153110047847,
      "grad_norm": 0.4453125,
      "learning_rate": 0.0,
      "loss": 0.7589,
      "step": 1040
    },
    {
      "epoch": 9.952153110047847,
      "eval_loss": 2.515638828277588,
      "eval_runtime": 0.236,
      "eval_samples_per_second": 42.374,
      "eval_steps_per_second": 4.237,
      "step": 1040
    },
    {
      "epoch": 9.952153110047847,
      "step": 1040,
      "total_flos": 3.1750846888049377e+18,
      "train_loss": 2.1342261617000284,
      "train_runtime": 2535.4321,
      "train_samples_per_second": 26.287,
      "train_steps_per_second": 0.41
    }
  ],
  "logging_steps": 5,
  "max_steps": 1040,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 3.1750846888049377e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}