fpadovani/nld-latn-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
nld-latn-10mb-after-ppt-shu…/checkpoint-12000/trainer_state.json
ModelHub XC 2a56643bcb 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/nld-latn-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
Source: Original Platform
2026-07-20 02:19:13 +08:00

24079 lines
661 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0885341074020318,
"eval_steps": 3000,
"global_step": 12000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.902761459350586,
"epoch": 0.00045355587808418,
"grad_norm": 15.0625,
"learning_rate": 2e-06,
"loss": 14.2424,
"mean_token_accuracy": 6.570302066393197e-05,
"num_tokens": 10827.0,
"step": 5
},
{
"entropy": 4.906017255783081,
"epoch": 0.00090711175616836,
"grad_norm": 15.0625,
"learning_rate": 4.5e-06,
"loss": 14.1679,
"mean_token_accuracy": 0.00036321740481071174,
"num_tokens": 21025.0,
"step": 10
},
{
"entropy": 4.919331836700439,
"epoch": 0.00136066763425254,
"grad_norm": 16.25,
"learning_rate": 7e-06,
"loss": 14.1379,
"mean_token_accuracy": 0.00018726952257566153,
"num_tokens": 31678.0,
"step": 15
},
{
"entropy": 4.969941806793213,
"epoch": 0.00181422351233672,
"grad_norm": 17.75,
"learning_rate": 9.5e-06,
"loss": 13.9183,
"mean_token_accuracy": 0.0,
"num_tokens": 41025.0,
"step": 20
},
{
"entropy": 5.119146680831909,
"epoch": 0.0022677793904208998,
"grad_norm": 21.875,
"learning_rate": 1.2e-05,
"loss": 13.5438,
"mean_token_accuracy": 0.0,
"num_tokens": 51270.0,
"step": 25
},
{
"entropy": 5.423425340652466,
"epoch": 0.00272133526850508,
"grad_norm": 26.625,
"learning_rate": 1.4500000000000002e-05,
"loss": 13.082,
"mean_token_accuracy": 0.0,
"num_tokens": 62142.0,
"step": 30
},
{
"entropy": 6.89132776260376,
"epoch": 0.00317489114658926,
"grad_norm": 25.625,
"learning_rate": 1.7000000000000003e-05,
"loss": 11.989,
"mean_token_accuracy": 0.0,
"num_tokens": 72709.0,
"step": 35
},
{
"entropy": 10.029912281036378,
"epoch": 0.00362844702467344,
"grad_norm": 3.671875,
"learning_rate": 1.95e-05,
"loss": 10.9002,
"mean_token_accuracy": 0.0,
"num_tokens": 82313.0,
"step": 40
},
{
"entropy": 10.710572910308837,
"epoch": 0.00408200290275762,
"grad_norm": 3.59375,
"learning_rate": 2.2e-05,
"loss": 10.6789,
"mean_token_accuracy": 0.001249837997602299,
"num_tokens": 92450.0,
"step": 45
},
{
"entropy": 10.739872074127197,
"epoch": 0.0045355587808417995,
"grad_norm": 3.03125,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.5409,
"mean_token_accuracy": 0.010005939414259046,
"num_tokens": 103978.0,
"step": 50
},
{
"entropy": 10.732357597351074,
"epoch": 0.00498911465892598,
"grad_norm": 2.140625,
"learning_rate": 2.7e-05,
"loss": 10.2901,
"mean_token_accuracy": 0.0165672049857676,
"num_tokens": 114134.0,
"step": 55
},
{
"entropy": 10.688269710540771,
"epoch": 0.00544267053701016,
"grad_norm": 2.296875,
"learning_rate": 2.95e-05,
"loss": 10.1413,
"mean_token_accuracy": 0.015207256283611059,
"num_tokens": 125037.0,
"step": 60
},
{
"entropy": 10.711676120758057,
"epoch": 0.00589622641509434,
"grad_norm": 1.90625,
"learning_rate": 3.2e-05,
"loss": 9.9497,
"mean_token_accuracy": 0.017865559458732604,
"num_tokens": 134721.0,
"step": 65
},
{
"entropy": 10.7262864112854,
"epoch": 0.00634978229317852,
"grad_norm": 1.9765625,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.8497,
"mean_token_accuracy": 0.03912395965307951,
"num_tokens": 144448.0,
"step": 70
},
{
"entropy": 10.72291088104248,
"epoch": 0.006803338171262699,
"grad_norm": 1.9609375,
"learning_rate": 3.7e-05,
"loss": 9.7736,
"mean_token_accuracy": 0.0496896930038929,
"num_tokens": 154343.0,
"step": 75
},
{
"entropy": 10.716559219360352,
"epoch": 0.00725689404934688,
"grad_norm": 1.953125,
"learning_rate": 3.95e-05,
"loss": 9.7443,
"mean_token_accuracy": 0.046880177780985834,
"num_tokens": 165650.0,
"step": 80
},
{
"entropy": 10.707624149322509,
"epoch": 0.007710449927431059,
"grad_norm": 1.7734375,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.711,
"mean_token_accuracy": 0.04450351819396019,
"num_tokens": 177098.0,
"step": 85
},
{
"entropy": 10.704974555969239,
"epoch": 0.00816400580551524,
"grad_norm": 1.96875,
"learning_rate": 4.45e-05,
"loss": 9.4919,
"mean_token_accuracy": 0.05480348989367485,
"num_tokens": 187903.0,
"step": 90
},
{
"entropy": 10.689633083343505,
"epoch": 0.00861756168359942,
"grad_norm": 1.8671875,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.4033,
"mean_token_accuracy": 0.052090702950954436,
"num_tokens": 198077.0,
"step": 95
},
{
"entropy": 10.674059963226318,
"epoch": 0.009071117561683599,
"grad_norm": 1.8046875,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.3565,
"mean_token_accuracy": 0.05150817483663559,
"num_tokens": 208641.0,
"step": 100
},
{
"entropy": 10.649502754211426,
"epoch": 0.009524673439767779,
"grad_norm": 1.84375,
"learning_rate": 5.2e-05,
"loss": 9.1936,
"mean_token_accuracy": 0.05499474816024304,
"num_tokens": 218398.0,
"step": 105
},
{
"entropy": 10.621120071411132,
"epoch": 0.00997822931785196,
"grad_norm": 1.6796875,
"learning_rate": 5.45e-05,
"loss": 9.066,
"mean_token_accuracy": 0.05329935699701309,
"num_tokens": 229059.0,
"step": 110
},
{
"entropy": 10.593021678924561,
"epoch": 0.01043178519593614,
"grad_norm": 1.7421875,
"learning_rate": 5.7e-05,
"loss": 8.9831,
"mean_token_accuracy": 0.05141277946531773,
"num_tokens": 239775.0,
"step": 115
},
{
"entropy": 10.551456928253174,
"epoch": 0.01088534107402032,
"grad_norm": 1.8515625,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.842,
"mean_token_accuracy": 0.05272237621247768,
"num_tokens": 249954.0,
"step": 120
},
{
"entropy": 10.502390956878662,
"epoch": 0.011338896952104499,
"grad_norm": 1.65625,
"learning_rate": 6.2e-05,
"loss": 8.7334,
"mean_token_accuracy": 0.05634862557053566,
"num_tokens": 260056.0,
"step": 125
},
{
"entropy": 10.426200675964356,
"epoch": 0.01179245283018868,
"grad_norm": 1.5078125,
"learning_rate": 6.450000000000001e-05,
"loss": 8.6347,
"mean_token_accuracy": 0.05641283132135868,
"num_tokens": 269904.0,
"step": 130
},
{
"entropy": 10.35210132598877,
"epoch": 0.01224600870827286,
"grad_norm": 1.65625,
"learning_rate": 6.7e-05,
"loss": 8.5138,
"mean_token_accuracy": 0.05234182849526405,
"num_tokens": 280900.0,
"step": 135
},
{
"entropy": 10.270804977416992,
"epoch": 0.01269956458635704,
"grad_norm": 1.5546875,
"learning_rate": 6.950000000000001e-05,
"loss": 8.406,
"mean_token_accuracy": 0.05256233848631382,
"num_tokens": 290844.0,
"step": 140
},
{
"entropy": 10.143964767456055,
"epoch": 0.013153120464441219,
"grad_norm": 1.2578125,
"learning_rate": 7.2e-05,
"loss": 8.2834,
"mean_token_accuracy": 0.056692027300596234,
"num_tokens": 301345.0,
"step": 145
},
{
"entropy": 10.030607223510742,
"epoch": 0.013606676342525399,
"grad_norm": 1.40625,
"learning_rate": 7.45e-05,
"loss": 8.1592,
"mean_token_accuracy": 0.05243662744760513,
"num_tokens": 312681.0,
"step": 150
},
{
"entropy": 9.876211452484132,
"epoch": 0.01406023222060958,
"grad_norm": 1.203125,
"learning_rate": 7.7e-05,
"loss": 8.0357,
"mean_token_accuracy": 0.05974825695157051,
"num_tokens": 323492.0,
"step": 155
},
{
"entropy": 9.703791522979737,
"epoch": 0.01451378809869376,
"grad_norm": 1.1015625,
"learning_rate": 7.950000000000001e-05,
"loss": 7.9778,
"mean_token_accuracy": 0.05336504280567169,
"num_tokens": 334521.0,
"step": 160
},
{
"entropy": 9.493086814880371,
"epoch": 0.014967343976777939,
"grad_norm": 0.99609375,
"learning_rate": 8.2e-05,
"loss": 7.9038,
"mean_token_accuracy": 0.05283032245934009,
"num_tokens": 345468.0,
"step": 165
},
{
"entropy": 9.287709426879882,
"epoch": 0.015420899854862119,
"grad_norm": 1.0625,
"learning_rate": 8.450000000000001e-05,
"loss": 7.7788,
"mean_token_accuracy": 0.055317680910229684,
"num_tokens": 355911.0,
"step": 170
},
{
"entropy": 9.030819416046143,
"epoch": 0.0158744557329463,
"grad_norm": 0.87890625,
"learning_rate": 8.7e-05,
"loss": 7.7522,
"mean_token_accuracy": 0.05364949442446232,
"num_tokens": 367537.0,
"step": 175
},
{
"entropy": 8.806273078918457,
"epoch": 0.01632801161103048,
"grad_norm": 0.890625,
"learning_rate": 8.95e-05,
"loss": 7.8067,
"mean_token_accuracy": 0.05446445494890213,
"num_tokens": 377918.0,
"step": 180
},
{
"entropy": 8.59828290939331,
"epoch": 0.01678156748911466,
"grad_norm": 0.703125,
"learning_rate": 9.2e-05,
"loss": 7.6514,
"mean_token_accuracy": 0.055352770164608954,
"num_tokens": 388461.0,
"step": 185
},
{
"entropy": 8.414749145507812,
"epoch": 0.01723512336719884,
"grad_norm": 0.9140625,
"learning_rate": 9.45e-05,
"loss": 7.6092,
"mean_token_accuracy": 0.06770730689167977,
"num_tokens": 399771.0,
"step": 190
},
{
"entropy": 8.264602565765381,
"epoch": 0.01768867924528302,
"grad_norm": 0.6953125,
"learning_rate": 9.7e-05,
"loss": 7.4766,
"mean_token_accuracy": 0.07282306849956513,
"num_tokens": 410254.0,
"step": 195
},
{
"entropy": 8.190978908538819,
"epoch": 0.018142235123367198,
"grad_norm": 1.25,
"learning_rate": 9.95e-05,
"loss": 7.4886,
"mean_token_accuracy": 0.06817287020385265,
"num_tokens": 419792.0,
"step": 200
},
{
"entropy": 8.113365936279298,
"epoch": 0.018595791001451378,
"grad_norm": 0.81640625,
"learning_rate": 0.000102,
"loss": 7.4782,
"mean_token_accuracy": 0.07438576593995094,
"num_tokens": 429264.0,
"step": 205
},
{
"entropy": 8.046414375305176,
"epoch": 0.019049346879535557,
"grad_norm": 0.69140625,
"learning_rate": 0.00010449999999999999,
"loss": 7.4294,
"mean_token_accuracy": 0.0784988671541214,
"num_tokens": 439624.0,
"step": 210
},
{
"entropy": 8.01142177581787,
"epoch": 0.01950290275761974,
"grad_norm": 0.83203125,
"learning_rate": 0.000107,
"loss": 7.4206,
"mean_token_accuracy": 0.07665222510695457,
"num_tokens": 449556.0,
"step": 215
},
{
"entropy": 7.982991123199463,
"epoch": 0.01995645863570392,
"grad_norm": 1.03125,
"learning_rate": 0.0001095,
"loss": 7.4192,
"mean_token_accuracy": 0.07819079160690308,
"num_tokens": 460243.0,
"step": 220
},
{
"entropy": 7.913471603393555,
"epoch": 0.0204100145137881,
"grad_norm": 1.015625,
"learning_rate": 0.000112,
"loss": 7.4682,
"mean_token_accuracy": 0.07565705217421055,
"num_tokens": 469767.0,
"step": 225
},
{
"entropy": 7.974834537506103,
"epoch": 0.02086357039187228,
"grad_norm": 0.83203125,
"learning_rate": 0.0001145,
"loss": 7.3948,
"mean_token_accuracy": 0.07845633998513221,
"num_tokens": 480329.0,
"step": 230
},
{
"entropy": 7.921621561050415,
"epoch": 0.02131712626995646,
"grad_norm": 0.8671875,
"learning_rate": 0.00011700000000000001,
"loss": 7.3597,
"mean_token_accuracy": 0.07565821185708047,
"num_tokens": 490096.0,
"step": 235
},
{
"entropy": 7.9462236881256105,
"epoch": 0.02177068214804064,
"grad_norm": 1.328125,
"learning_rate": 0.00011949999999999999,
"loss": 7.404,
"mean_token_accuracy": 0.07110006958246232,
"num_tokens": 501231.0,
"step": 240
},
{
"entropy": 7.90762996673584,
"epoch": 0.022224238026124818,
"grad_norm": 1.3359375,
"learning_rate": 0.000122,
"loss": 7.3003,
"mean_token_accuracy": 0.07855217903852463,
"num_tokens": 510722.0,
"step": 245
},
{
"entropy": 7.791062259674073,
"epoch": 0.022677793904208998,
"grad_norm": 0.8359375,
"learning_rate": 0.0001245,
"loss": 7.342,
"mean_token_accuracy": 0.0785796582698822,
"num_tokens": 521452.0,
"step": 250
},
{
"entropy": 7.835426616668701,
"epoch": 0.023131349782293177,
"grad_norm": 1.0,
"learning_rate": 0.000127,
"loss": 7.3275,
"mean_token_accuracy": 0.07848411723971367,
"num_tokens": 532397.0,
"step": 255
},
{
"entropy": 7.834844970703125,
"epoch": 0.02358490566037736,
"grad_norm": 0.9375,
"learning_rate": 0.0001295,
"loss": 7.3126,
"mean_token_accuracy": 0.08216355144977569,
"num_tokens": 542271.0,
"step": 260
},
{
"entropy": 7.783105802536011,
"epoch": 0.02403846153846154,
"grad_norm": 1.046875,
"learning_rate": 0.000132,
"loss": 7.2852,
"mean_token_accuracy": 0.08326507434248924,
"num_tokens": 552411.0,
"step": 265
},
{
"entropy": 7.789214849472046,
"epoch": 0.02449201741654572,
"grad_norm": 0.87890625,
"learning_rate": 0.00013450000000000002,
"loss": 7.2853,
"mean_token_accuracy": 0.07970908731222152,
"num_tokens": 561518.0,
"step": 270
},
{
"entropy": 7.8078306198120115,
"epoch": 0.0249455732946299,
"grad_norm": 0.859375,
"learning_rate": 0.00013700000000000002,
"loss": 7.3048,
"mean_token_accuracy": 0.08076792433857918,
"num_tokens": 571361.0,
"step": 275
},
{
"entropy": 7.704005384445191,
"epoch": 0.02539912917271408,
"grad_norm": 0.87890625,
"learning_rate": 0.0001395,
"loss": 7.2079,
"mean_token_accuracy": 0.08890472128987312,
"num_tokens": 580926.0,
"step": 280
},
{
"entropy": 7.777441549301147,
"epoch": 0.025852685050798258,
"grad_norm": 1.2109375,
"learning_rate": 0.00014199999999999998,
"loss": 7.305,
"mean_token_accuracy": 0.07825151681900025,
"num_tokens": 592282.0,
"step": 285
},
{
"entropy": 7.756225681304931,
"epoch": 0.026306240928882438,
"grad_norm": 1.25,
"learning_rate": 0.0001445,
"loss": 7.3123,
"mean_token_accuracy": 0.08080485239624977,
"num_tokens": 602767.0,
"step": 290
},
{
"entropy": 7.716958808898926,
"epoch": 0.026759796806966617,
"grad_norm": 0.86328125,
"learning_rate": 0.000147,
"loss": 7.2177,
"mean_token_accuracy": 0.08552644699811936,
"num_tokens": 612763.0,
"step": 295
},
{
"entropy": 7.706277990341187,
"epoch": 0.027213352685050797,
"grad_norm": 0.9296875,
"learning_rate": 0.0001495,
"loss": 7.1737,
"mean_token_accuracy": 0.08685687705874442,
"num_tokens": 622796.0,
"step": 300
},
{
"entropy": 7.668240880966186,
"epoch": 0.027666908563134977,
"grad_norm": 1.2265625,
"learning_rate": 0.000152,
"loss": 7.1609,
"mean_token_accuracy": 0.08536986410617828,
"num_tokens": 633136.0,
"step": 305
},
{
"entropy": 7.649664688110351,
"epoch": 0.02812046444121916,
"grad_norm": 0.9921875,
"learning_rate": 0.00015450000000000001,
"loss": 7.143,
"mean_token_accuracy": 0.08583157956600189,
"num_tokens": 643347.0,
"step": 310
},
{
"entropy": 7.697743082046509,
"epoch": 0.02857402031930334,
"grad_norm": 0.859375,
"learning_rate": 0.000157,
"loss": 7.1414,
"mean_token_accuracy": 0.09304612576961517,
"num_tokens": 653096.0,
"step": 315
},
{
"entropy": 7.587596750259399,
"epoch": 0.02902757619738752,
"grad_norm": 1.4296875,
"learning_rate": 0.0001595,
"loss": 7.1706,
"mean_token_accuracy": 0.08879687860608101,
"num_tokens": 663557.0,
"step": 320
},
{
"entropy": 7.612160730361938,
"epoch": 0.0294811320754717,
"grad_norm": 0.96484375,
"learning_rate": 0.000162,
"loss": 7.2254,
"mean_token_accuracy": 0.08831041827797889,
"num_tokens": 674395.0,
"step": 325
},
{
"entropy": 7.695938348770142,
"epoch": 0.029934687953555878,
"grad_norm": 1.0390625,
"learning_rate": 0.00016450000000000001,
"loss": 7.1265,
"mean_token_accuracy": 0.09078574851155281,
"num_tokens": 684776.0,
"step": 330
},
{
"entropy": 7.576247215270996,
"epoch": 0.030388243831640058,
"grad_norm": 1.0703125,
"learning_rate": 0.00016700000000000002,
"loss": 7.109,
"mean_token_accuracy": 0.08843011409044266,
"num_tokens": 694886.0,
"step": 335
},
{
"entropy": 7.556908130645752,
"epoch": 0.030841799709724237,
"grad_norm": 1.2109375,
"learning_rate": 0.00016950000000000003,
"loss": 7.171,
"mean_token_accuracy": 0.09018958061933517,
"num_tokens": 705426.0,
"step": 340
},
{
"entropy": 7.578840160369873,
"epoch": 0.03129535558780842,
"grad_norm": 0.87890625,
"learning_rate": 0.00017199999999999998,
"loss": 7.1594,
"mean_token_accuracy": 0.08782409057021141,
"num_tokens": 716271.0,
"step": 345
},
{
"entropy": 7.638495922088623,
"epoch": 0.0317489114658926,
"grad_norm": 1.0234375,
"learning_rate": 0.00017449999999999999,
"loss": 7.1606,
"mean_token_accuracy": 0.09265835508704186,
"num_tokens": 726875.0,
"step": 350
},
{
"entropy": 7.563690900802612,
"epoch": 0.032202467343976776,
"grad_norm": 0.88671875,
"learning_rate": 0.000177,
"loss": 7.0773,
"mean_token_accuracy": 0.08827165514230728,
"num_tokens": 737825.0,
"step": 355
},
{
"entropy": 7.58997654914856,
"epoch": 0.03265602322206096,
"grad_norm": 1.1171875,
"learning_rate": 0.0001795,
"loss": 7.089,
"mean_token_accuracy": 0.0890809178352356,
"num_tokens": 747424.0,
"step": 360
},
{
"entropy": 7.506178569793701,
"epoch": 0.033109579100145135,
"grad_norm": 1.3671875,
"learning_rate": 0.000182,
"loss": 7.0901,
"mean_token_accuracy": 0.09158140644431115,
"num_tokens": 757649.0,
"step": 365
},
{
"entropy": 7.567615652084351,
"epoch": 0.03356313497822932,
"grad_norm": 1.078125,
"learning_rate": 0.0001845,
"loss": 7.0671,
"mean_token_accuracy": 0.0931968331336975,
"num_tokens": 768556.0,
"step": 370
},
{
"entropy": 7.400697326660156,
"epoch": 0.034016690856313495,
"grad_norm": 1.140625,
"learning_rate": 0.000187,
"loss": 7.0591,
"mean_token_accuracy": 0.09333095103502273,
"num_tokens": 778331.0,
"step": 375
},
{
"entropy": 7.539828157424926,
"epoch": 0.03447024673439768,
"grad_norm": 1.0546875,
"learning_rate": 0.0001895,
"loss": 7.0387,
"mean_token_accuracy": 0.09365462064743042,
"num_tokens": 788065.0,
"step": 380
},
{
"entropy": 7.495446348190308,
"epoch": 0.03492380261248186,
"grad_norm": 1.0,
"learning_rate": 0.000192,
"loss": 7.0671,
"mean_token_accuracy": 0.09628497660160065,
"num_tokens": 798912.0,
"step": 385
},
{
"entropy": 7.502418661117554,
"epoch": 0.03537735849056604,
"grad_norm": 1.1328125,
"learning_rate": 0.0001945,
"loss": 7.0166,
"mean_token_accuracy": 0.09867717772722244,
"num_tokens": 809580.0,
"step": 390
},
{
"entropy": 7.49851360321045,
"epoch": 0.03583091436865022,
"grad_norm": 1.0234375,
"learning_rate": 0.00019700000000000002,
"loss": 7.1108,
"mean_token_accuracy": 0.09084957540035248,
"num_tokens": 820207.0,
"step": 395
},
{
"entropy": 7.348110675811768,
"epoch": 0.036284470246734396,
"grad_norm": 1.015625,
"learning_rate": 0.00019950000000000002,
"loss": 6.9336,
"mean_token_accuracy": 0.08944878205657006,
"num_tokens": 831038.0,
"step": 400
},
{
"entropy": 7.545411586761475,
"epoch": 0.03673802612481858,
"grad_norm": 1.0078125,
"learning_rate": 0.000202,
"loss": 7.0186,
"mean_token_accuracy": 0.09372441694140435,
"num_tokens": 842144.0,
"step": 405
},
{
"entropy": 7.47612156867981,
"epoch": 0.037191582002902755,
"grad_norm": 1.609375,
"learning_rate": 0.00020449999999999998,
"loss": 6.9989,
"mean_token_accuracy": 0.09804348424077033,
"num_tokens": 851315.0,
"step": 410
},
{
"entropy": 7.322836017608642,
"epoch": 0.03764513788098694,
"grad_norm": 1.0625,
"learning_rate": 0.000207,
"loss": 7.0063,
"mean_token_accuracy": 0.09803526178002357,
"num_tokens": 861233.0,
"step": 415
},
{
"entropy": 7.406570482254028,
"epoch": 0.038098693759071114,
"grad_norm": 1.171875,
"learning_rate": 0.0002095,
"loss": 6.9994,
"mean_token_accuracy": 0.0946230135858059,
"num_tokens": 871131.0,
"step": 420
},
{
"entropy": 7.391973686218262,
"epoch": 0.0385522496371553,
"grad_norm": 1.2890625,
"learning_rate": 0.000212,
"loss": 6.975,
"mean_token_accuracy": 0.09175262376666068,
"num_tokens": 882195.0,
"step": 425
},
{
"entropy": 7.452529048919677,
"epoch": 0.03900580551523948,
"grad_norm": 1.1328125,
"learning_rate": 0.0002145,
"loss": 6.9467,
"mean_token_accuracy": 0.09589996412396432,
"num_tokens": 891806.0,
"step": 430
},
{
"entropy": 7.431856489181518,
"epoch": 0.03945936139332366,
"grad_norm": 0.9375,
"learning_rate": 0.00021700000000000002,
"loss": 7.0087,
"mean_token_accuracy": 0.09551806971430779,
"num_tokens": 901350.0,
"step": 435
},
{
"entropy": 7.414364910125732,
"epoch": 0.03991291727140784,
"grad_norm": 1.296875,
"learning_rate": 0.0002195,
"loss": 6.9182,
"mean_token_accuracy": 0.09558806866407395,
"num_tokens": 912061.0,
"step": 440
},
{
"entropy": 7.305682420730591,
"epoch": 0.040366473149492016,
"grad_norm": 1.765625,
"learning_rate": 0.000222,
"loss": 6.89,
"mean_token_accuracy": 0.09567090198397636,
"num_tokens": 922220.0,
"step": 445
},
{
"entropy": 7.367461013793945,
"epoch": 0.0408200290275762,
"grad_norm": 1.0625,
"learning_rate": 0.0002245,
"loss": 6.8992,
"mean_token_accuracy": 0.09762913659214974,
"num_tokens": 932319.0,
"step": 450
},
{
"entropy": 7.292191886901856,
"epoch": 0.041273584905660375,
"grad_norm": 1.0859375,
"learning_rate": 0.00022700000000000002,
"loss": 6.9231,
"mean_token_accuracy": 0.09886682778596878,
"num_tokens": 943254.0,
"step": 455
},
{
"entropy": 7.387064361572266,
"epoch": 0.04172714078374456,
"grad_norm": 0.83984375,
"learning_rate": 0.00022950000000000002,
"loss": 6.939,
"mean_token_accuracy": 0.09336472377181053,
"num_tokens": 953921.0,
"step": 460
},
{
"entropy": 7.296272230148316,
"epoch": 0.042180696661828734,
"grad_norm": 1.015625,
"learning_rate": 0.00023200000000000003,
"loss": 6.8498,
"mean_token_accuracy": 0.10429160594940186,
"num_tokens": 963832.0,
"step": 465
},
{
"entropy": 7.304851102828979,
"epoch": 0.04263425253991292,
"grad_norm": 1.234375,
"learning_rate": 0.00023449999999999998,
"loss": 6.9459,
"mean_token_accuracy": 0.09589096009731293,
"num_tokens": 974066.0,
"step": 470
},
{
"entropy": 7.324310684204102,
"epoch": 0.0430878084179971,
"grad_norm": 1.5234375,
"learning_rate": 0.000237,
"loss": 6.9353,
"mean_token_accuracy": 0.09945596382021904,
"num_tokens": 984402.0,
"step": 475
},
{
"entropy": 7.312732458114624,
"epoch": 0.04354136429608128,
"grad_norm": 1.078125,
"learning_rate": 0.0002395,
"loss": 6.8591,
"mean_token_accuracy": 0.09938167333602906,
"num_tokens": 993555.0,
"step": 480
},
{
"entropy": 7.334518146514893,
"epoch": 0.04399492017416546,
"grad_norm": 1.171875,
"learning_rate": 0.000242,
"loss": 6.8404,
"mean_token_accuracy": 0.10574882701039315,
"num_tokens": 1002920.0,
"step": 485
},
{
"entropy": 7.289060401916504,
"epoch": 0.044448476052249636,
"grad_norm": 1.3828125,
"learning_rate": 0.0002445,
"loss": 6.907,
"mean_token_accuracy": 0.09448245465755463,
"num_tokens": 1013611.0,
"step": 490
},
{
"entropy": 7.166425371170044,
"epoch": 0.04490203193033382,
"grad_norm": 1.2109375,
"learning_rate": 0.000247,
"loss": 6.8012,
"mean_token_accuracy": 0.0957680583000183,
"num_tokens": 1024121.0,
"step": 495
},
{
"entropy": 7.363502073287964,
"epoch": 0.045355587808417995,
"grad_norm": 1.015625,
"learning_rate": 0.0002495,
"loss": 7.032,
"mean_token_accuracy": 0.1009724922478199,
"num_tokens": 1033845.0,
"step": 500
},
{
"entropy": 7.24678144454956,
"epoch": 0.04580914368650218,
"grad_norm": 1.0625,
"learning_rate": 0.000252,
"loss": 6.831,
"mean_token_accuracy": 0.09820592999458314,
"num_tokens": 1044642.0,
"step": 505
},
{
"entropy": 7.280577898025513,
"epoch": 0.046262699564586354,
"grad_norm": 1.1640625,
"learning_rate": 0.0002545,
"loss": 6.8867,
"mean_token_accuracy": 0.09730055406689644,
"num_tokens": 1054728.0,
"step": 510
},
{
"entropy": 7.20595383644104,
"epoch": 0.04671625544267054,
"grad_norm": 1.015625,
"learning_rate": 0.000257,
"loss": 6.8504,
"mean_token_accuracy": 0.0942535676062107,
"num_tokens": 1065743.0,
"step": 515
},
{
"entropy": 7.1944211483001705,
"epoch": 0.04716981132075472,
"grad_norm": 1.2578125,
"learning_rate": 0.0002595,
"loss": 6.7719,
"mean_token_accuracy": 0.09884541779756546,
"num_tokens": 1076798.0,
"step": 520
},
{
"entropy": 7.200849342346191,
"epoch": 0.047623367198838897,
"grad_norm": 1.078125,
"learning_rate": 0.000262,
"loss": 6.8921,
"mean_token_accuracy": 0.09789391681551933,
"num_tokens": 1087445.0,
"step": 525
},
{
"entropy": 7.216012907028198,
"epoch": 0.04807692307692308,
"grad_norm": 1.046875,
"learning_rate": 0.00026450000000000003,
"loss": 6.7985,
"mean_token_accuracy": 0.10271445363759994,
"num_tokens": 1098379.0,
"step": 530
},
{
"entropy": 7.228269386291504,
"epoch": 0.048530478955007256,
"grad_norm": 1.171875,
"learning_rate": 0.00026700000000000004,
"loss": 6.8866,
"mean_token_accuracy": 0.09828322008252144,
"num_tokens": 1108016.0,
"step": 535
},
{
"entropy": 7.184875774383545,
"epoch": 0.04898403483309144,
"grad_norm": 1.140625,
"learning_rate": 0.00026950000000000005,
"loss": 6.7837,
"mean_token_accuracy": 0.09823340401053429,
"num_tokens": 1117887.0,
"step": 540
},
{
"entropy": 7.186046075820923,
"epoch": 0.049437590711175615,
"grad_norm": 1.265625,
"learning_rate": 0.00027200000000000005,
"loss": 6.7773,
"mean_token_accuracy": 0.0981587328016758,
"num_tokens": 1128497.0,
"step": 545
},
{
"entropy": 7.1298119068145756,
"epoch": 0.0498911465892598,
"grad_norm": 0.9453125,
"learning_rate": 0.0002745,
"loss": 6.8017,
"mean_token_accuracy": 0.10637210011482238,
"num_tokens": 1138168.0,
"step": 550
},
{
"entropy": 7.136086797714233,
"epoch": 0.050344702467343974,
"grad_norm": 1.0234375,
"learning_rate": 0.000277,
"loss": 6.7618,
"mean_token_accuracy": 0.10049154385924339,
"num_tokens": 1148357.0,
"step": 555
},
{
"entropy": 7.1673754215240475,
"epoch": 0.05079825834542816,
"grad_norm": 1.265625,
"learning_rate": 0.0002795,
"loss": 6.7907,
"mean_token_accuracy": 0.10017752796411514,
"num_tokens": 1159964.0,
"step": 560
},
{
"entropy": 7.23973536491394,
"epoch": 0.05125181422351233,
"grad_norm": 1.4296875,
"learning_rate": 0.00028199999999999997,
"loss": 6.8508,
"mean_token_accuracy": 0.09828138649463654,
"num_tokens": 1170683.0,
"step": 565
},
{
"entropy": 7.192407703399658,
"epoch": 0.051705370101596516,
"grad_norm": 1.515625,
"learning_rate": 0.0002845,
"loss": 6.8052,
"mean_token_accuracy": 0.1006213940680027,
"num_tokens": 1181680.0,
"step": 570
},
{
"entropy": 7.093324756622314,
"epoch": 0.0521589259796807,
"grad_norm": 1.34375,
"learning_rate": 0.000287,
"loss": 6.6997,
"mean_token_accuracy": 0.10036976039409637,
"num_tokens": 1191619.0,
"step": 575
},
{
"entropy": 7.011592292785645,
"epoch": 0.052612481857764876,
"grad_norm": 1.2421875,
"learning_rate": 0.0002895,
"loss": 6.6916,
"mean_token_accuracy": 0.10292767509818077,
"num_tokens": 1202264.0,
"step": 580
},
{
"entropy": 7.043795013427735,
"epoch": 0.05306603773584906,
"grad_norm": 1.0390625,
"learning_rate": 0.000292,
"loss": 6.6999,
"mean_token_accuracy": 0.1045819990336895,
"num_tokens": 1212119.0,
"step": 585
},
{
"entropy": 7.129189682006836,
"epoch": 0.053519593613933235,
"grad_norm": 1.0546875,
"learning_rate": 0.0002945,
"loss": 6.8212,
"mean_token_accuracy": 0.102519890666008,
"num_tokens": 1222317.0,
"step": 590
},
{
"entropy": 7.168576860427857,
"epoch": 0.05397314949201742,
"grad_norm": 1.2265625,
"learning_rate": 0.000297,
"loss": 6.7716,
"mean_token_accuracy": 0.10512009784579276,
"num_tokens": 1232914.0,
"step": 595
},
{
"entropy": 6.968721580505371,
"epoch": 0.054426705370101594,
"grad_norm": 1.1171875,
"learning_rate": 0.0002995,
"loss": 6.6181,
"mean_token_accuracy": 0.10760264396667481,
"num_tokens": 1244214.0,
"step": 600
},
{
"entropy": 7.105316591262818,
"epoch": 0.05488026124818578,
"grad_norm": 1.1875,
"learning_rate": 0.000302,
"loss": 6.7149,
"mean_token_accuracy": 0.10045301765203477,
"num_tokens": 1254626.0,
"step": 605
},
{
"entropy": 6.98553991317749,
"epoch": 0.05533381712626995,
"grad_norm": 1.3046875,
"learning_rate": 0.0003045,
"loss": 6.8113,
"mean_token_accuracy": 0.10129074826836586,
"num_tokens": 1265461.0,
"step": 610
},
{
"entropy": 7.082133436203003,
"epoch": 0.055787373004354136,
"grad_norm": 1.0625,
"learning_rate": 0.000307,
"loss": 6.6817,
"mean_token_accuracy": 0.11003390699625015,
"num_tokens": 1275810.0,
"step": 615
},
{
"entropy": 6.964151763916016,
"epoch": 0.05624092888243832,
"grad_norm": 1.3203125,
"learning_rate": 0.0003095,
"loss": 6.6878,
"mean_token_accuracy": 0.11000289767980576,
"num_tokens": 1285170.0,
"step": 620
},
{
"entropy": 7.021602058410645,
"epoch": 0.056694484760522496,
"grad_norm": 0.99609375,
"learning_rate": 0.000312,
"loss": 6.6361,
"mean_token_accuracy": 0.10547553822398185,
"num_tokens": 1295460.0,
"step": 625
},
{
"entropy": 7.010881614685059,
"epoch": 0.05714804063860668,
"grad_norm": 1.2734375,
"learning_rate": 0.0003145,
"loss": 6.7523,
"mean_token_accuracy": 0.10284495502710342,
"num_tokens": 1306656.0,
"step": 630
},
{
"entropy": 7.070465517044068,
"epoch": 0.057601596516690855,
"grad_norm": 1.0703125,
"learning_rate": 0.000317,
"loss": 6.7338,
"mean_token_accuracy": 0.1056830108165741,
"num_tokens": 1318019.0,
"step": 635
},
{
"entropy": 6.88353590965271,
"epoch": 0.05805515239477504,
"grad_norm": 1.578125,
"learning_rate": 0.0003195,
"loss": 6.6511,
"mean_token_accuracy": 0.10918820276856422,
"num_tokens": 1328120.0,
"step": 640
},
{
"entropy": 7.09418797492981,
"epoch": 0.058508708272859214,
"grad_norm": 1.1953125,
"learning_rate": 0.000322,
"loss": 6.7483,
"mean_token_accuracy": 0.10312345400452613,
"num_tokens": 1338354.0,
"step": 645
},
{
"entropy": 7.009733867645264,
"epoch": 0.0589622641509434,
"grad_norm": 1.40625,
"learning_rate": 0.00032450000000000003,
"loss": 6.7237,
"mean_token_accuracy": 0.1024971716105938,
"num_tokens": 1348666.0,
"step": 650
},
{
"entropy": 7.047331762313843,
"epoch": 0.05941582002902757,
"grad_norm": 0.9453125,
"learning_rate": 0.00032700000000000003,
"loss": 6.7554,
"mean_token_accuracy": 0.10805487632751465,
"num_tokens": 1359713.0,
"step": 655
},
{
"entropy": 6.9947686195373535,
"epoch": 0.059869375907111756,
"grad_norm": 0.97265625,
"learning_rate": 0.00032950000000000004,
"loss": 6.6124,
"mean_token_accuracy": 0.10892275646328926,
"num_tokens": 1369507.0,
"step": 660
},
{
"entropy": 6.998069620132446,
"epoch": 0.06032293178519594,
"grad_norm": 1.1171875,
"learning_rate": 0.00033200000000000005,
"loss": 6.768,
"mean_token_accuracy": 0.10034214109182357,
"num_tokens": 1380726.0,
"step": 665
},
{
"entropy": 7.112451839447021,
"epoch": 0.060776487663280115,
"grad_norm": 1.171875,
"learning_rate": 0.00033450000000000005,
"loss": 6.6933,
"mean_token_accuracy": 0.10579130649566651,
"num_tokens": 1391423.0,
"step": 670
},
{
"entropy": 7.052730226516724,
"epoch": 0.0612300435413643,
"grad_norm": 1.2890625,
"learning_rate": 0.000337,
"loss": 6.6678,
"mean_token_accuracy": 0.11295373067259788,
"num_tokens": 1402492.0,
"step": 675
},
{
"entropy": 6.922665405273437,
"epoch": 0.061683599419448475,
"grad_norm": 1.0390625,
"learning_rate": 0.0003395,
"loss": 6.7161,
"mean_token_accuracy": 0.10829840004444122,
"num_tokens": 1412497.0,
"step": 680
},
{
"entropy": 7.03723840713501,
"epoch": 0.06213715529753266,
"grad_norm": 1.171875,
"learning_rate": 0.000342,
"loss": 6.6772,
"mean_token_accuracy": 0.11061608865857124,
"num_tokens": 1423214.0,
"step": 685
},
{
"entropy": 6.935256671905518,
"epoch": 0.06259071117561683,
"grad_norm": 1.2109375,
"learning_rate": 0.00034449999999999997,
"loss": 6.6745,
"mean_token_accuracy": 0.10797292739152908,
"num_tokens": 1434044.0,
"step": 690
},
{
"entropy": 6.918494129180909,
"epoch": 0.06304426705370102,
"grad_norm": 1.1640625,
"learning_rate": 0.000347,
"loss": 6.6361,
"mean_token_accuracy": 0.10982377529144287,
"num_tokens": 1444387.0,
"step": 695
},
{
"entropy": 6.957173490524292,
"epoch": 0.0634978229317852,
"grad_norm": 1.015625,
"learning_rate": 0.0003495,
"loss": 6.5862,
"mean_token_accuracy": 0.10744345635175705,
"num_tokens": 1455157.0,
"step": 700
},
{
"entropy": 6.938698959350586,
"epoch": 0.06395137880986937,
"grad_norm": 1.125,
"learning_rate": 0.000352,
"loss": 6.6693,
"mean_token_accuracy": 0.11478404849767684,
"num_tokens": 1466145.0,
"step": 705
},
{
"entropy": 6.962664556503296,
"epoch": 0.06440493468795355,
"grad_norm": 0.9765625,
"learning_rate": 0.0003545,
"loss": 6.667,
"mean_token_accuracy": 0.10709880217909813,
"num_tokens": 1476713.0,
"step": 710
},
{
"entropy": 6.92603645324707,
"epoch": 0.06485849056603774,
"grad_norm": 1.1640625,
"learning_rate": 0.000357,
"loss": 6.671,
"mean_token_accuracy": 0.11409479826688766,
"num_tokens": 1486908.0,
"step": 715
},
{
"entropy": 6.899812459945679,
"epoch": 0.06531204644412192,
"grad_norm": 1.375,
"learning_rate": 0.0003595,
"loss": 6.5941,
"mean_token_accuracy": 0.11084994897246361,
"num_tokens": 1498185.0,
"step": 720
},
{
"entropy": 6.901278781890869,
"epoch": 0.0657656023222061,
"grad_norm": 1.2265625,
"learning_rate": 0.000362,
"loss": 6.6413,
"mean_token_accuracy": 0.1077586367726326,
"num_tokens": 1508376.0,
"step": 725
},
{
"entropy": 6.9933178424835205,
"epoch": 0.06621915820029027,
"grad_norm": 1.0625,
"learning_rate": 0.0003645,
"loss": 6.7199,
"mean_token_accuracy": 0.10835012272000313,
"num_tokens": 1519953.0,
"step": 730
},
{
"entropy": 6.853632211685181,
"epoch": 0.06667271407837445,
"grad_norm": 1.0859375,
"learning_rate": 0.000367,
"loss": 6.592,
"mean_token_accuracy": 0.11243074238300324,
"num_tokens": 1529793.0,
"step": 735
},
{
"entropy": 6.829979610443115,
"epoch": 0.06712626995645864,
"grad_norm": 0.95703125,
"learning_rate": 0.0003695,
"loss": 6.5882,
"mean_token_accuracy": 0.11395004838705063,
"num_tokens": 1540859.0,
"step": 740
},
{
"entropy": 6.908363008499146,
"epoch": 0.06757982583454282,
"grad_norm": 1.0859375,
"learning_rate": 0.000372,
"loss": 6.5104,
"mean_token_accuracy": 0.11262088865041733,
"num_tokens": 1549783.0,
"step": 745
},
{
"entropy": 6.84120545387268,
"epoch": 0.06803338171262699,
"grad_norm": 1.125,
"learning_rate": 0.0003745,
"loss": 6.5803,
"mean_token_accuracy": 0.10933975651860237,
"num_tokens": 1560522.0,
"step": 750
},
{
"entropy": 6.878204107284546,
"epoch": 0.06848693759071117,
"grad_norm": 1.421875,
"learning_rate": 0.000377,
"loss": 6.553,
"mean_token_accuracy": 0.11197188422083855,
"num_tokens": 1570668.0,
"step": 755
},
{
"entropy": 6.856545925140381,
"epoch": 0.06894049346879536,
"grad_norm": 1.0390625,
"learning_rate": 0.0003795,
"loss": 6.669,
"mean_token_accuracy": 0.11744550019502639,
"num_tokens": 1581523.0,
"step": 760
},
{
"entropy": 6.84909029006958,
"epoch": 0.06939404934687954,
"grad_norm": 1.1796875,
"learning_rate": 0.000382,
"loss": 6.5917,
"mean_token_accuracy": 0.1113414227962494,
"num_tokens": 1591356.0,
"step": 765
},
{
"entropy": 6.83960976600647,
"epoch": 0.06984760522496372,
"grad_norm": 1.046875,
"learning_rate": 0.0003845,
"loss": 6.5574,
"mean_token_accuracy": 0.11633669957518578,
"num_tokens": 1601779.0,
"step": 770
},
{
"entropy": 6.888131904602051,
"epoch": 0.07030116110304789,
"grad_norm": 1.0234375,
"learning_rate": 0.00038700000000000003,
"loss": 6.5633,
"mean_token_accuracy": 0.1150720700621605,
"num_tokens": 1611995.0,
"step": 775
},
{
"entropy": 6.856516313552857,
"epoch": 0.07075471698113207,
"grad_norm": 1.1328125,
"learning_rate": 0.00038950000000000003,
"loss": 6.586,
"mean_token_accuracy": 0.11050437688827515,
"num_tokens": 1621834.0,
"step": 780
},
{
"entropy": 6.778967618942261,
"epoch": 0.07120827285921626,
"grad_norm": 1.1953125,
"learning_rate": 0.00039200000000000004,
"loss": 6.6288,
"mean_token_accuracy": 0.10744166076183319,
"num_tokens": 1632881.0,
"step": 785
},
{
"entropy": 6.7607427597045895,
"epoch": 0.07166182873730044,
"grad_norm": 1.0625,
"learning_rate": 0.00039450000000000005,
"loss": 6.6028,
"mean_token_accuracy": 0.10911152362823487,
"num_tokens": 1642870.0,
"step": 790
},
{
"entropy": 6.8341546058654785,
"epoch": 0.07211538461538461,
"grad_norm": 1.2578125,
"learning_rate": 0.00039700000000000005,
"loss": 6.484,
"mean_token_accuracy": 0.11779236048460007,
"num_tokens": 1653390.0,
"step": 795
},
{
"entropy": 6.802682399749756,
"epoch": 0.07256894049346879,
"grad_norm": 1.015625,
"learning_rate": 0.0003995,
"loss": 6.6078,
"mean_token_accuracy": 0.11580143421888352,
"num_tokens": 1663382.0,
"step": 800
},
{
"entropy": 6.889751863479614,
"epoch": 0.07302249637155298,
"grad_norm": 1.03125,
"learning_rate": 0.000402,
"loss": 6.6263,
"mean_token_accuracy": 0.10960384383797646,
"num_tokens": 1673601.0,
"step": 805
},
{
"entropy": 6.744623231887817,
"epoch": 0.07347605224963716,
"grad_norm": 1.234375,
"learning_rate": 0.0004045,
"loss": 6.5301,
"mean_token_accuracy": 0.1178050920367241,
"num_tokens": 1683565.0,
"step": 810
},
{
"entropy": 6.720245599746704,
"epoch": 0.07392960812772134,
"grad_norm": 1.9140625,
"learning_rate": 0.00040699999999999997,
"loss": 6.4975,
"mean_token_accuracy": 0.1157220259308815,
"num_tokens": 1693636.0,
"step": 815
},
{
"entropy": 6.888108444213867,
"epoch": 0.07438316400580551,
"grad_norm": 1.265625,
"learning_rate": 0.0004095,
"loss": 6.6028,
"mean_token_accuracy": 0.11490728184580803,
"num_tokens": 1703415.0,
"step": 820
},
{
"entropy": 6.8595898151397705,
"epoch": 0.0748367198838897,
"grad_norm": 1.5703125,
"learning_rate": 0.000412,
"loss": 6.5759,
"mean_token_accuracy": 0.11677221283316612,
"num_tokens": 1713579.0,
"step": 825
},
{
"entropy": 6.7610352516174315,
"epoch": 0.07529027576197388,
"grad_norm": 1.015625,
"learning_rate": 0.0004145,
"loss": 6.5869,
"mean_token_accuracy": 0.11174456626176835,
"num_tokens": 1724583.0,
"step": 830
},
{
"entropy": 6.793128204345703,
"epoch": 0.07574383164005806,
"grad_norm": 1.3984375,
"learning_rate": 0.000417,
"loss": 6.4998,
"mean_token_accuracy": 0.12149128615856171,
"num_tokens": 1735293.0,
"step": 835
},
{
"entropy": 6.784815979003906,
"epoch": 0.07619738751814223,
"grad_norm": 1.03125,
"learning_rate": 0.0004195,
"loss": 6.5285,
"mean_token_accuracy": 0.11545130759477615,
"num_tokens": 1745623.0,
"step": 840
},
{
"entropy": 6.785441732406616,
"epoch": 0.07665094339622641,
"grad_norm": 0.8984375,
"learning_rate": 0.000422,
"loss": 6.5589,
"mean_token_accuracy": 0.11196033582091332,
"num_tokens": 1755713.0,
"step": 845
},
{
"entropy": 6.844113492965699,
"epoch": 0.0771044992743106,
"grad_norm": 1.2890625,
"learning_rate": 0.0004245,
"loss": 6.5448,
"mean_token_accuracy": 0.11228399202227593,
"num_tokens": 1765264.0,
"step": 850
},
{
"entropy": 6.778715133666992,
"epoch": 0.07755805515239478,
"grad_norm": 1.1640625,
"learning_rate": 0.000427,
"loss": 6.5551,
"mean_token_accuracy": 0.11371968314051628,
"num_tokens": 1775769.0,
"step": 855
},
{
"entropy": 6.709914636611939,
"epoch": 0.07801161103047896,
"grad_norm": 1.1015625,
"learning_rate": 0.0004295,
"loss": 6.4821,
"mean_token_accuracy": 0.11689155176281929,
"num_tokens": 1786245.0,
"step": 860
},
{
"entropy": 6.716907548904419,
"epoch": 0.07846516690856313,
"grad_norm": 1.21875,
"learning_rate": 0.000432,
"loss": 6.5182,
"mean_token_accuracy": 0.12198375314474105,
"num_tokens": 1796040.0,
"step": 865
},
{
"entropy": 6.72428708076477,
"epoch": 0.07891872278664731,
"grad_norm": 1.0078125,
"learning_rate": 0.0004345,
"loss": 6.5268,
"mean_token_accuracy": 0.11597820520401,
"num_tokens": 1806007.0,
"step": 870
},
{
"entropy": 6.862047719955444,
"epoch": 0.0793722786647315,
"grad_norm": 1.109375,
"learning_rate": 0.000437,
"loss": 6.6034,
"mean_token_accuracy": 0.11016635000705718,
"num_tokens": 1816935.0,
"step": 875
},
{
"entropy": 6.7226348400115965,
"epoch": 0.07982583454281568,
"grad_norm": 1.234375,
"learning_rate": 0.0004395,
"loss": 6.4647,
"mean_token_accuracy": 0.11464838534593583,
"num_tokens": 1827419.0,
"step": 880
},
{
"entropy": 6.80533652305603,
"epoch": 0.08027939042089985,
"grad_norm": 1.3515625,
"learning_rate": 0.000442,
"loss": 6.5762,
"mean_token_accuracy": 0.11279575228691101,
"num_tokens": 1837933.0,
"step": 885
},
{
"entropy": 6.692126846313476,
"epoch": 0.08073294629898403,
"grad_norm": 1.0703125,
"learning_rate": 0.0004445,
"loss": 6.6181,
"mean_token_accuracy": 0.11968202516436577,
"num_tokens": 1848551.0,
"step": 890
},
{
"entropy": 6.773914003372193,
"epoch": 0.08118650217706821,
"grad_norm": 1.2578125,
"learning_rate": 0.000447,
"loss": 6.4874,
"mean_token_accuracy": 0.11803188994526863,
"num_tokens": 1858380.0,
"step": 895
},
{
"entropy": 6.64626522064209,
"epoch": 0.0816400580551524,
"grad_norm": 1.0859375,
"learning_rate": 0.00044950000000000003,
"loss": 6.4255,
"mean_token_accuracy": 0.11616697832942009,
"num_tokens": 1868291.0,
"step": 900
},
{
"entropy": 6.739772891998291,
"epoch": 0.08209361393323658,
"grad_norm": 1.234375,
"learning_rate": 0.00045200000000000004,
"loss": 6.5222,
"mean_token_accuracy": 0.11596911773085594,
"num_tokens": 1879219.0,
"step": 905
},
{
"entropy": 6.701965379714966,
"epoch": 0.08254716981132075,
"grad_norm": 0.96875,
"learning_rate": 0.00045450000000000004,
"loss": 6.5064,
"mean_token_accuracy": 0.1169955737888813,
"num_tokens": 1889809.0,
"step": 910
},
{
"entropy": 6.736312532424927,
"epoch": 0.08300072568940493,
"grad_norm": 1.0546875,
"learning_rate": 0.00045700000000000005,
"loss": 6.4667,
"mean_token_accuracy": 0.11598963215947151,
"num_tokens": 1899986.0,
"step": 915
},
{
"entropy": 6.627064943313599,
"epoch": 0.08345428156748912,
"grad_norm": 1.0390625,
"learning_rate": 0.00045950000000000006,
"loss": 6.4631,
"mean_token_accuracy": 0.11359920203685761,
"num_tokens": 1909314.0,
"step": 920
},
{
"entropy": 6.714141511917115,
"epoch": 0.0839078374455733,
"grad_norm": 1.078125,
"learning_rate": 0.000462,
"loss": 6.4954,
"mean_token_accuracy": 0.1206825278699398,
"num_tokens": 1918443.0,
"step": 925
},
{
"entropy": 6.711059856414795,
"epoch": 0.08436139332365747,
"grad_norm": 1.140625,
"learning_rate": 0.0004645,
"loss": 6.4599,
"mean_token_accuracy": 0.11760578602552414,
"num_tokens": 1928683.0,
"step": 930
},
{
"entropy": 6.563105583190918,
"epoch": 0.08481494920174165,
"grad_norm": 1.109375,
"learning_rate": 0.000467,
"loss": 6.3733,
"mean_token_accuracy": 0.11783806160092354,
"num_tokens": 1938293.0,
"step": 935
},
{
"entropy": 6.621684551239014,
"epoch": 0.08526850507982583,
"grad_norm": 1.03125,
"learning_rate": 0.0004695,
"loss": 6.4649,
"mean_token_accuracy": 0.1159488506615162,
"num_tokens": 1948233.0,
"step": 940
},
{
"entropy": 6.703399705886841,
"epoch": 0.08572206095791002,
"grad_norm": 1.0390625,
"learning_rate": 0.000472,
"loss": 6.3795,
"mean_token_accuracy": 0.12225733622908593,
"num_tokens": 1958611.0,
"step": 945
},
{
"entropy": 6.67765588760376,
"epoch": 0.0861756168359942,
"grad_norm": 1.03125,
"learning_rate": 0.0004745,
"loss": 6.5279,
"mean_token_accuracy": 0.12170163914561272,
"num_tokens": 1968957.0,
"step": 950
},
{
"entropy": 6.721297121047973,
"epoch": 0.08662917271407837,
"grad_norm": 1.203125,
"learning_rate": 0.000477,
"loss": 6.4022,
"mean_token_accuracy": 0.12271194607019424,
"num_tokens": 1978503.0,
"step": 955
},
{
"entropy": 6.51479959487915,
"epoch": 0.08708272859216255,
"grad_norm": 1.09375,
"learning_rate": 0.0004795,
"loss": 6.3838,
"mean_token_accuracy": 0.12056222185492516,
"num_tokens": 1988458.0,
"step": 960
},
{
"entropy": 6.663824939727784,
"epoch": 0.08753628447024674,
"grad_norm": 1.0703125,
"learning_rate": 0.000482,
"loss": 6.5235,
"mean_token_accuracy": 0.1133016251027584,
"num_tokens": 1999984.0,
"step": 965
},
{
"entropy": 6.534238243103028,
"epoch": 0.08798984034833092,
"grad_norm": 1.1171875,
"learning_rate": 0.0004845,
"loss": 6.38,
"mean_token_accuracy": 0.12614745199680327,
"num_tokens": 2010645.0,
"step": 970
},
{
"entropy": 6.69314603805542,
"epoch": 0.08844339622641509,
"grad_norm": 1.1953125,
"learning_rate": 0.000487,
"loss": 6.4879,
"mean_token_accuracy": 0.11517181396484374,
"num_tokens": 2020196.0,
"step": 975
},
{
"entropy": 6.586405944824219,
"epoch": 0.08889695210449927,
"grad_norm": 1.078125,
"learning_rate": 0.0004895,
"loss": 6.3749,
"mean_token_accuracy": 0.12108265981078148,
"num_tokens": 2030101.0,
"step": 980
},
{
"entropy": 6.620171546936035,
"epoch": 0.08935050798258345,
"grad_norm": 1.03125,
"learning_rate": 0.000492,
"loss": 6.4594,
"mean_token_accuracy": 0.1197955846786499,
"num_tokens": 2040454.0,
"step": 985
},
{
"entropy": 6.581764078140258,
"epoch": 0.08980406386066764,
"grad_norm": 1.09375,
"learning_rate": 0.0004945,
"loss": 6.402,
"mean_token_accuracy": 0.1192368857562542,
"num_tokens": 2050378.0,
"step": 990
},
{
"entropy": 6.651663303375244,
"epoch": 0.09025761973875182,
"grad_norm": 1.0546875,
"learning_rate": 0.000497,
"loss": 6.4549,
"mean_token_accuracy": 0.11869885623455048,
"num_tokens": 2060282.0,
"step": 995
},
{
"entropy": 6.589950847625732,
"epoch": 0.09071117561683599,
"grad_norm": 1.2109375,
"learning_rate": 0.0004995,
"loss": 6.4433,
"mean_token_accuracy": 0.12011284530162811,
"num_tokens": 2070563.0,
"step": 1000
},
{
"entropy": 6.661682939529419,
"epoch": 0.09116473149492017,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999985110217,
"loss": 6.5128,
"mean_token_accuracy": 0.11984694451093673,
"num_tokens": 2080153.0,
"step": 1005
},
{
"entropy": 6.546395254135132,
"epoch": 0.09161828737300436,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999999924620471,
"loss": 6.3602,
"mean_token_accuracy": 0.12557145729660987,
"num_tokens": 2090590.0,
"step": 1010
},
{
"entropy": 6.653158855438233,
"epoch": 0.09207184325108854,
"grad_norm": 0.953125,
"learning_rate": 0.0004999999817600155,
"loss": 6.4314,
"mean_token_accuracy": 0.11608105525374413,
"num_tokens": 2101945.0,
"step": 1015
},
{
"entropy": 6.635874509811401,
"epoch": 0.09252539912917271,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999999664049268,
"loss": 6.4312,
"mean_token_accuracy": 0.12001314610242844,
"num_tokens": 2112514.0,
"step": 1020
},
{
"entropy": 6.546846055984497,
"epoch": 0.09297895500725689,
"grad_norm": 1.40625,
"learning_rate": 0.0004999999463967816,
"loss": 6.3852,
"mean_token_accuracy": 0.11658050194382667,
"num_tokens": 2122296.0,
"step": 1025
},
{
"entropy": 6.713669490814209,
"epoch": 0.09343251088534107,
"grad_norm": 0.87109375,
"learning_rate": 0.00049999992173558,
"loss": 6.5311,
"mean_token_accuracy": 0.10938856825232506,
"num_tokens": 2134512.0,
"step": 1030
},
{
"entropy": 6.649443817138672,
"epoch": 0.09388606676342526,
"grad_norm": 1.1875,
"learning_rate": 0.0004999998924213228,
"loss": 6.4339,
"mean_token_accuracy": 0.12238849774003029,
"num_tokens": 2143934.0,
"step": 1035
},
{
"entropy": 6.488268947601318,
"epoch": 0.09433962264150944,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999998584540105,
"loss": 6.3351,
"mean_token_accuracy": 0.11762306764721871,
"num_tokens": 2155730.0,
"step": 1040
},
{
"entropy": 6.570074510574341,
"epoch": 0.09479317851959361,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999998198336437,
"loss": 6.3681,
"mean_token_accuracy": 0.11555026769638062,
"num_tokens": 2166422.0,
"step": 1045
},
{
"entropy": 6.646327686309815,
"epoch": 0.09524673439767779,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999997765602233,
"loss": 6.469,
"mean_token_accuracy": 0.11361956149339676,
"num_tokens": 2177842.0,
"step": 1050
},
{
"entropy": 6.617121315002441,
"epoch": 0.09570029027576198,
"grad_norm": 0.984375,
"learning_rate": 0.0004999997286337502,
"loss": 6.5398,
"mean_token_accuracy": 0.11744040250778198,
"num_tokens": 2188614.0,
"step": 1055
},
{
"entropy": 6.604910278320313,
"epoch": 0.09615384615384616,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999996760542254,
"loss": 6.3246,
"mean_token_accuracy": 0.12567146718502045,
"num_tokens": 2197886.0,
"step": 1060
},
{
"entropy": 6.525793075561523,
"epoch": 0.09660740203193033,
"grad_norm": 1.0234375,
"learning_rate": 0.00049999961882165,
"loss": 6.3772,
"mean_token_accuracy": 0.12232817932963372,
"num_tokens": 2209930.0,
"step": 1065
},
{
"entropy": 6.564913749694824,
"epoch": 0.09706095791001451,
"grad_norm": 1.0,
"learning_rate": 0.0004999995569360251,
"loss": 6.3959,
"mean_token_accuracy": 0.11825868040323258,
"num_tokens": 2219767.0,
"step": 1070
},
{
"entropy": 6.555662298202515,
"epoch": 0.0975145137880987,
"grad_norm": 0.94921875,
"learning_rate": 0.000499999490397352,
"loss": 6.3192,
"mean_token_accuracy": 0.121416836977005,
"num_tokens": 2230082.0,
"step": 1075
},
{
"entropy": 6.495458173751831,
"epoch": 0.09796806966618288,
"grad_norm": 1.078125,
"learning_rate": 0.0004999994192056321,
"loss": 6.3673,
"mean_token_accuracy": 0.12748734578490256,
"num_tokens": 2240066.0,
"step": 1080
},
{
"entropy": 6.524483871459961,
"epoch": 0.09842162554426705,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999993433608669,
"loss": 6.3942,
"mean_token_accuracy": 0.11839633882045746,
"num_tokens": 2250732.0,
"step": 1085
},
{
"entropy": 6.593288707733154,
"epoch": 0.09887518142235123,
"grad_norm": 0.96875,
"learning_rate": 0.0004999992628630579,
"loss": 6.2797,
"mean_token_accuracy": 0.12327790930867195,
"num_tokens": 2261452.0,
"step": 1090
},
{
"entropy": 6.485133695602417,
"epoch": 0.09932873730043541,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999991777122069,
"loss": 6.3782,
"mean_token_accuracy": 0.12279095351696015,
"num_tokens": 2271929.0,
"step": 1095
},
{
"entropy": 6.531335401535034,
"epoch": 0.0997822931785196,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999990879083156,
"loss": 6.3604,
"mean_token_accuracy": 0.12319791615009308,
"num_tokens": 2282987.0,
"step": 1100
},
{
"entropy": 6.444648933410645,
"epoch": 0.10023584905660378,
"grad_norm": 1.046875,
"learning_rate": 0.0004999989934513857,
"loss": 6.3295,
"mean_token_accuracy": 0.12565099447965622,
"num_tokens": 2292796.0,
"step": 1105
},
{
"entropy": 6.598123359680176,
"epoch": 0.10068940493468795,
"grad_norm": 1.109375,
"learning_rate": 0.0004999988943414193,
"loss": 6.3766,
"mean_token_accuracy": 0.12076297253370286,
"num_tokens": 2303057.0,
"step": 1110
},
{
"entropy": 6.4050017356872555,
"epoch": 0.10114296081277213,
"grad_norm": 1.109375,
"learning_rate": 0.0004999987905784184,
"loss": 6.3638,
"mean_token_accuracy": 0.12473835423588753,
"num_tokens": 2314168.0,
"step": 1115
},
{
"entropy": 6.548490381240844,
"epoch": 0.10159651669085631,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999986821623853,
"loss": 6.335,
"mean_token_accuracy": 0.12088372111320496,
"num_tokens": 2325241.0,
"step": 1120
},
{
"entropy": 6.579931640625,
"epoch": 0.1020500725689405,
"grad_norm": 1.125,
"learning_rate": 0.0004999985690933219,
"loss": 6.3584,
"mean_token_accuracy": 0.12107386961579322,
"num_tokens": 2334866.0,
"step": 1125
},
{
"entropy": 6.416501808166504,
"epoch": 0.10250362844702467,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999984513712311,
"loss": 6.366,
"mean_token_accuracy": 0.12216197848320007,
"num_tokens": 2345422.0,
"step": 1130
},
{
"entropy": 6.573203802108765,
"epoch": 0.10295718432510885,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999983289961146,
"loss": 6.3667,
"mean_token_accuracy": 0.12462597936391831,
"num_tokens": 2356285.0,
"step": 1135
},
{
"entropy": 6.414275217056274,
"epoch": 0.10341074020319303,
"grad_norm": 1.109375,
"learning_rate": 0.0004999982019679755,
"loss": 6.2962,
"mean_token_accuracy": 0.12918335497379302,
"num_tokens": 2366861.0,
"step": 1140
},
{
"entropy": 6.546162033081055,
"epoch": 0.10386429608127722,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999980702868164,
"loss": 6.3186,
"mean_token_accuracy": 0.12778222039341927,
"num_tokens": 2376851.0,
"step": 1145
},
{
"entropy": 6.468019723892212,
"epoch": 0.1043178519593614,
"grad_norm": 0.96875,
"learning_rate": 0.0004999979339526396,
"loss": 6.3967,
"mean_token_accuracy": 0.12601374089717865,
"num_tokens": 2386857.0,
"step": 1150
},
{
"entropy": 6.512000989913941,
"epoch": 0.10477140783744557,
"grad_norm": 0.953125,
"learning_rate": 0.0004999977929654484,
"loss": 6.3197,
"mean_token_accuracy": 0.12597798183560371,
"num_tokens": 2397035.0,
"step": 1155
},
{
"entropy": 6.556376361846924,
"epoch": 0.10522496371552975,
"grad_norm": 1.3359375,
"learning_rate": 0.0004999976473252453,
"loss": 6.3936,
"mean_token_accuracy": 0.12374778017401696,
"num_tokens": 2408177.0,
"step": 1160
},
{
"entropy": 6.569843435287476,
"epoch": 0.10567851959361393,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999974970320338,
"loss": 6.342,
"mean_token_accuracy": 0.12336181774735451,
"num_tokens": 2418445.0,
"step": 1165
},
{
"entropy": 6.446996784210205,
"epoch": 0.10613207547169812,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999973420858165,
"loss": 6.3375,
"mean_token_accuracy": 0.12320749461650848,
"num_tokens": 2428990.0,
"step": 1170
},
{
"entropy": 6.456551933288575,
"epoch": 0.10658563134978229,
"grad_norm": 1.015625,
"learning_rate": 0.0004999971824865968,
"loss": 6.3706,
"mean_token_accuracy": 0.11951386407017708,
"num_tokens": 2439643.0,
"step": 1175
},
{
"entropy": 6.636844635009766,
"epoch": 0.10703918722786647,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999970182343782,
"loss": 6.4152,
"mean_token_accuracy": 0.12312159314751625,
"num_tokens": 2450691.0,
"step": 1180
},
{
"entropy": 6.513713598251343,
"epoch": 0.10749274310595065,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999968493291638,
"loss": 6.3519,
"mean_token_accuracy": 0.12444976568222046,
"num_tokens": 2460706.0,
"step": 1185
},
{
"entropy": 6.332309865951538,
"epoch": 0.10794629898403484,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999966757709573,
"loss": 6.2426,
"mean_token_accuracy": 0.13491440415382386,
"num_tokens": 2470947.0,
"step": 1190
},
{
"entropy": 6.500934410095215,
"epoch": 0.10839985486211902,
"grad_norm": 1.0625,
"learning_rate": 0.0004999964975597622,
"loss": 6.2077,
"mean_token_accuracy": 0.12923589274287223,
"num_tokens": 2481132.0,
"step": 1195
},
{
"entropy": 6.5095055103302,
"epoch": 0.10885341074020319,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999963146955821,
"loss": 6.2492,
"mean_token_accuracy": 0.12882558181881903,
"num_tokens": 2491701.0,
"step": 1200
},
{
"entropy": 6.308234691619873,
"epoch": 0.10930696661828737,
"grad_norm": 0.92578125,
"learning_rate": 0.000499996127178421,
"loss": 6.3173,
"mean_token_accuracy": 0.1222999058663845,
"num_tokens": 2502360.0,
"step": 1205
},
{
"entropy": 6.526607275009155,
"epoch": 0.10976052249637155,
"grad_norm": 1.03125,
"learning_rate": 0.0004999959350082825,
"loss": 6.2211,
"mean_token_accuracy": 0.1294765926897526,
"num_tokens": 2512140.0,
"step": 1210
},
{
"entropy": 6.4609130859375,
"epoch": 0.11021407837445574,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999957381851709,
"loss": 6.3314,
"mean_token_accuracy": 0.12224320769309997,
"num_tokens": 2522721.0,
"step": 1215
},
{
"entropy": 6.3936292171478275,
"epoch": 0.1106676342525399,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999955367090901,
"loss": 6.1999,
"mean_token_accuracy": 0.1303488314151764,
"num_tokens": 2531786.0,
"step": 1220
},
{
"entropy": 6.481017780303955,
"epoch": 0.11112119013062409,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999953305800441,
"loss": 6.3062,
"mean_token_accuracy": 0.12003768384456634,
"num_tokens": 2542895.0,
"step": 1225
},
{
"entropy": 6.442375755310058,
"epoch": 0.11157474600870827,
"grad_norm": 1.015625,
"learning_rate": 0.0004999951197980374,
"loss": 6.305,
"mean_token_accuracy": 0.12665605992078782,
"num_tokens": 2554120.0,
"step": 1230
},
{
"entropy": 6.489529991149903,
"epoch": 0.11202830188679246,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999949043630744,
"loss": 6.3339,
"mean_token_accuracy": 0.12409112006425857,
"num_tokens": 2563935.0,
"step": 1235
},
{
"entropy": 6.3626786231994625,
"epoch": 0.11248185776487664,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999946842751593,
"loss": 6.2477,
"mean_token_accuracy": 0.1251929335296154,
"num_tokens": 2574465.0,
"step": 1240
},
{
"entropy": 6.431154251098633,
"epoch": 0.11293541364296081,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999944595342968,
"loss": 6.2144,
"mean_token_accuracy": 0.12585682272911072,
"num_tokens": 2584694.0,
"step": 1245
},
{
"entropy": 6.402558374404907,
"epoch": 0.11338896952104499,
"grad_norm": 0.96875,
"learning_rate": 0.0004999942301404916,
"loss": 6.1836,
"mean_token_accuracy": 0.1319870851933956,
"num_tokens": 2594540.0,
"step": 1250
},
{
"entropy": 6.334234380722046,
"epoch": 0.11384252539912917,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999939960937484,
"loss": 6.1343,
"mean_token_accuracy": 0.1355540633201599,
"num_tokens": 2604896.0,
"step": 1255
},
{
"entropy": 6.494117069244385,
"epoch": 0.11429608127721336,
"grad_norm": 1.09375,
"learning_rate": 0.0004999937573940721,
"loss": 6.2886,
"mean_token_accuracy": 0.1345356211066246,
"num_tokens": 2614143.0,
"step": 1260
},
{
"entropy": 6.274349880218506,
"epoch": 0.11474963715529753,
"grad_norm": 1.125,
"learning_rate": 0.0004999935140414674,
"loss": 6.1951,
"mean_token_accuracy": 0.12857473343610765,
"num_tokens": 2623999.0,
"step": 1265
},
{
"entropy": 6.511336469650269,
"epoch": 0.11520319303338171,
"grad_norm": 0.984375,
"learning_rate": 0.0004999932660359395,
"loss": 6.4052,
"mean_token_accuracy": 0.12467710450291633,
"num_tokens": 2636208.0,
"step": 1270
},
{
"entropy": 6.449693775177002,
"epoch": 0.11565674891146589,
"grad_norm": 1.015625,
"learning_rate": 0.0004999930133774936,
"loss": 6.2795,
"mean_token_accuracy": 0.13579537123441696,
"num_tokens": 2645854.0,
"step": 1275
},
{
"entropy": 6.469302797317505,
"epoch": 0.11611030478955008,
"grad_norm": 1.046875,
"learning_rate": 0.0004999927560661348,
"loss": 6.2593,
"mean_token_accuracy": 0.1359974443912506,
"num_tokens": 2655877.0,
"step": 1280
},
{
"entropy": 6.474381637573242,
"epoch": 0.11656386066763426,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999924941018685,
"loss": 6.2797,
"mean_token_accuracy": 0.1305759347975254,
"num_tokens": 2666157.0,
"step": 1285
},
{
"entropy": 6.2998865127563475,
"epoch": 0.11701741654571843,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999922274847,
"loss": 6.1961,
"mean_token_accuracy": 0.13332010209560394,
"num_tokens": 2676647.0,
"step": 1290
},
{
"entropy": 6.388902854919434,
"epoch": 0.11747097242380261,
"grad_norm": 1.046875,
"learning_rate": 0.000499991956214635,
"loss": 6.2308,
"mean_token_accuracy": 0.14106824845075608,
"num_tokens": 2686417.0,
"step": 1295
},
{
"entropy": 6.417198610305786,
"epoch": 0.1179245283018868,
"grad_norm": 0.94140625,
"learning_rate": 0.000499991680291679,
"loss": 6.2281,
"mean_token_accuracy": 0.12554471418261529,
"num_tokens": 2697540.0,
"step": 1300
},
{
"entropy": 6.371242427825928,
"epoch": 0.11837808417997098,
"grad_norm": 0.953125,
"learning_rate": 0.0004999913997158378,
"loss": 6.2766,
"mean_token_accuracy": 0.12081151455640793,
"num_tokens": 2709397.0,
"step": 1305
},
{
"entropy": 6.381704521179199,
"epoch": 0.11883164005805515,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999911144871169,
"loss": 6.2198,
"mean_token_accuracy": 0.12881449088454247,
"num_tokens": 2720430.0,
"step": 1310
},
{
"entropy": 6.3669020652771,
"epoch": 0.11928519593613933,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999908246055226,
"loss": 6.1855,
"mean_token_accuracy": 0.1260046534240246,
"num_tokens": 2730648.0,
"step": 1315
},
{
"entropy": 6.3640461444854735,
"epoch": 0.11973875181422351,
"grad_norm": 1.0,
"learning_rate": 0.0004999905300710607,
"loss": 6.1767,
"mean_token_accuracy": 0.12971787378191948,
"num_tokens": 2739979.0,
"step": 1320
},
{
"entropy": 6.467068099975586,
"epoch": 0.1201923076923077,
"grad_norm": 1.234375,
"learning_rate": 0.0004999902308837373,
"loss": 6.3039,
"mean_token_accuracy": 0.12591902986168862,
"num_tokens": 2750329.0,
"step": 1325
},
{
"entropy": 6.297173976898193,
"epoch": 0.12064586357039188,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999899270435584,
"loss": 6.1752,
"mean_token_accuracy": 0.1377357542514801,
"num_tokens": 2760535.0,
"step": 1330
},
{
"entropy": 6.402283620834351,
"epoch": 0.12109941944847605,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999896185505307,
"loss": 6.2099,
"mean_token_accuracy": 0.1282433770596981,
"num_tokens": 2770470.0,
"step": 1335
},
{
"entropy": 6.357119464874268,
"epoch": 0.12155297532656023,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999893054046603,
"loss": 6.2395,
"mean_token_accuracy": 0.12727272063493728,
"num_tokens": 2781506.0,
"step": 1340
},
{
"entropy": 6.281704378128052,
"epoch": 0.12200653120464441,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999889876059537,
"loss": 6.0945,
"mean_token_accuracy": 0.1367440052330494,
"num_tokens": 2791245.0,
"step": 1345
},
{
"entropy": 6.390601301193238,
"epoch": 0.1224600870827286,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999886651544176,
"loss": 6.1272,
"mean_token_accuracy": 0.13428937569260596,
"num_tokens": 2801680.0,
"step": 1350
},
{
"entropy": 6.298570585250855,
"epoch": 0.12291364296081277,
"grad_norm": 1.078125,
"learning_rate": 0.0004999883380500584,
"loss": 6.2077,
"mean_token_accuracy": 0.13555841743946076,
"num_tokens": 2812805.0,
"step": 1355
},
{
"entropy": 6.452785491943359,
"epoch": 0.12336719883889695,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999880062928831,
"loss": 6.2401,
"mean_token_accuracy": 0.12772336453199387,
"num_tokens": 2822834.0,
"step": 1360
},
{
"entropy": 6.322773456573486,
"epoch": 0.12382075471698113,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999876698828985,
"loss": 6.1829,
"mean_token_accuracy": 0.1308009184896946,
"num_tokens": 2832312.0,
"step": 1365
},
{
"entropy": 6.387462949752807,
"epoch": 0.12427431059506532,
"grad_norm": 0.9375,
"learning_rate": 0.0004999873288201116,
"loss": 6.1993,
"mean_token_accuracy": 0.13682973608374596,
"num_tokens": 2842998.0,
"step": 1370
},
{
"entropy": 6.381214332580567,
"epoch": 0.1247278664731495,
"grad_norm": 1.3203125,
"learning_rate": 0.0004999869831045294,
"loss": 6.2182,
"mean_token_accuracy": 0.1281896151602268,
"num_tokens": 2854218.0,
"step": 1375
},
{
"entropy": 6.274281024932861,
"epoch": 0.12518142235123367,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999866327361589,
"loss": 6.1073,
"mean_token_accuracy": 0.13887619599699974,
"num_tokens": 2864484.0,
"step": 1380
},
{
"entropy": 6.392649173736572,
"epoch": 0.12563497822931785,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999862777150077,
"loss": 6.2187,
"mean_token_accuracy": 0.13294630274176597,
"num_tokens": 2873800.0,
"step": 1385
},
{
"entropy": 6.378524923324585,
"epoch": 0.12608853410740203,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999859180410829,
"loss": 6.2196,
"mean_token_accuracy": 0.12941672652959824,
"num_tokens": 2884247.0,
"step": 1390
},
{
"entropy": 6.3022987842559814,
"epoch": 0.12654208998548622,
"grad_norm": 0.953125,
"learning_rate": 0.0004999855537143919,
"loss": 6.1277,
"mean_token_accuracy": 0.13321435153484346,
"num_tokens": 2894739.0,
"step": 1395
},
{
"entropy": 6.330821084976196,
"epoch": 0.1269956458635704,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999851847349425,
"loss": 6.2018,
"mean_token_accuracy": 0.13309190422296524,
"num_tokens": 2904376.0,
"step": 1400
},
{
"entropy": 6.41134204864502,
"epoch": 0.12744920174165458,
"grad_norm": 0.96875,
"learning_rate": 0.0004999848111027419,
"loss": 6.2438,
"mean_token_accuracy": 0.12967129945755004,
"num_tokens": 2914304.0,
"step": 1405
},
{
"entropy": 6.346340560913086,
"epoch": 0.12790275761973874,
"grad_norm": 1.015625,
"learning_rate": 0.0004999844328177983,
"loss": 6.1142,
"mean_token_accuracy": 0.13302754908800124,
"num_tokens": 2924400.0,
"step": 1410
},
{
"entropy": 6.468166971206665,
"epoch": 0.12835631349782292,
"grad_norm": 1.0,
"learning_rate": 0.0004999840498801191,
"loss": 6.3637,
"mean_token_accuracy": 0.13450959250330924,
"num_tokens": 2935040.0,
"step": 1415
},
{
"entropy": 6.250949144363403,
"epoch": 0.1288098693759071,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999836622897126,
"loss": 6.139,
"mean_token_accuracy": 0.13291264548897744,
"num_tokens": 2945536.0,
"step": 1420
},
{
"entropy": 6.240394401550293,
"epoch": 0.1292634252539913,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999832700465865,
"loss": 6.0631,
"mean_token_accuracy": 0.13600366339087486,
"num_tokens": 2954969.0,
"step": 1425
},
{
"entropy": 6.372559261322022,
"epoch": 0.12971698113207547,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999828731507491,
"loss": 6.1846,
"mean_token_accuracy": 0.13019020035862922,
"num_tokens": 2966218.0,
"step": 1430
},
{
"entropy": 6.348944854736328,
"epoch": 0.13017053701015965,
"grad_norm": 1.03125,
"learning_rate": 0.0004999824716022085,
"loss": 6.2124,
"mean_token_accuracy": 0.12568282783031465,
"num_tokens": 2976640.0,
"step": 1435
},
{
"entropy": 6.317825126647949,
"epoch": 0.13062409288824384,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999820654009731,
"loss": 6.1754,
"mean_token_accuracy": 0.13333930373191832,
"num_tokens": 2985883.0,
"step": 1440
},
{
"entropy": 6.33205885887146,
"epoch": 0.13107764876632802,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999816545470513,
"loss": 6.1738,
"mean_token_accuracy": 0.1286979652941227,
"num_tokens": 2996695.0,
"step": 1445
},
{
"entropy": 6.338008069992066,
"epoch": 0.1315312046444122,
"grad_norm": 0.984375,
"learning_rate": 0.0004999812390404514,
"loss": 6.1786,
"mean_token_accuracy": 0.13488187193870543,
"num_tokens": 3007500.0,
"step": 1450
},
{
"entropy": 6.171080303192139,
"epoch": 0.13198476052249636,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999808188811823,
"loss": 6.0426,
"mean_token_accuracy": 0.13422525823116302,
"num_tokens": 3017299.0,
"step": 1455
},
{
"entropy": 6.42709584236145,
"epoch": 0.13243831640058054,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999803940692524,
"loss": 6.3248,
"mean_token_accuracy": 0.12495478615164757,
"num_tokens": 3028389.0,
"step": 1460
},
{
"entropy": 6.34055233001709,
"epoch": 0.13289187227866472,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999799646046707,
"loss": 6.1313,
"mean_token_accuracy": 0.13551339507102966,
"num_tokens": 3038883.0,
"step": 1465
},
{
"entropy": 6.360403776168823,
"epoch": 0.1333454281567489,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999795304874461,
"loss": 6.2634,
"mean_token_accuracy": 0.13484074100852012,
"num_tokens": 3049904.0,
"step": 1470
},
{
"entropy": 6.292270040512085,
"epoch": 0.1337989840348331,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999790917175873,
"loss": 6.0954,
"mean_token_accuracy": 0.13213692530989646,
"num_tokens": 3059788.0,
"step": 1475
},
{
"entropy": 6.332358312606812,
"epoch": 0.13425253991291727,
"grad_norm": 0.96875,
"learning_rate": 0.0004999786482951036,
"loss": 6.0914,
"mean_token_accuracy": 0.13121994063258172,
"num_tokens": 3070413.0,
"step": 1480
},
{
"entropy": 6.2656394958496096,
"epoch": 0.13470609579100146,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999782002200041,
"loss": 6.2047,
"mean_token_accuracy": 0.1304186299443245,
"num_tokens": 3081093.0,
"step": 1485
},
{
"entropy": 6.334182834625244,
"epoch": 0.13515965166908564,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999777474922982,
"loss": 6.1612,
"mean_token_accuracy": 0.12822958827018738,
"num_tokens": 3090594.0,
"step": 1490
},
{
"entropy": 6.352655935287475,
"epoch": 0.13561320754716982,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999772901119951,
"loss": 6.2409,
"mean_token_accuracy": 0.13285785242915155,
"num_tokens": 3101061.0,
"step": 1495
},
{
"entropy": 6.28689432144165,
"epoch": 0.13606676342525398,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999768280791043,
"loss": 6.1627,
"mean_token_accuracy": 0.13513913974165917,
"num_tokens": 3111715.0,
"step": 1500
},
{
"entropy": 6.282320547103882,
"epoch": 0.13652031930333816,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999763613936352,
"loss": 6.0892,
"mean_token_accuracy": 0.13638533875346184,
"num_tokens": 3121378.0,
"step": 1505
},
{
"entropy": 6.252392864227295,
"epoch": 0.13697387518142234,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999758900555979,
"loss": 6.0183,
"mean_token_accuracy": 0.13862146139144899,
"num_tokens": 3131229.0,
"step": 1510
},
{
"entropy": 6.241233825683594,
"epoch": 0.13742743105950653,
"grad_norm": 1.046875,
"learning_rate": 0.0004999754140650017,
"loss": 6.1025,
"mean_token_accuracy": 0.13484096452593802,
"num_tokens": 3140859.0,
"step": 1515
},
{
"entropy": 6.310092401504517,
"epoch": 0.1378809869375907,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999749334218567,
"loss": 6.1048,
"mean_token_accuracy": 0.13604142144322395,
"num_tokens": 3151692.0,
"step": 1520
},
{
"entropy": 6.2527360916137695,
"epoch": 0.1383345428156749,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999744481261726,
"loss": 6.1801,
"mean_token_accuracy": 0.13370710536837577,
"num_tokens": 3161879.0,
"step": 1525
},
{
"entropy": 6.248764514923096,
"epoch": 0.13878809869375908,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999739581779597,
"loss": 6.0456,
"mean_token_accuracy": 0.1446085900068283,
"num_tokens": 3171282.0,
"step": 1530
},
{
"entropy": 6.270263290405273,
"epoch": 0.13924165457184326,
"grad_norm": 0.98828125,
"learning_rate": 0.000499973463577228,
"loss": 6.1068,
"mean_token_accuracy": 0.1418723337352276,
"num_tokens": 3181398.0,
"step": 1535
},
{
"entropy": 6.2894738674163815,
"epoch": 0.13969521044992744,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999729643239877,
"loss": 6.2273,
"mean_token_accuracy": 0.1322890356183052,
"num_tokens": 3191431.0,
"step": 1540
},
{
"entropy": 6.347840642929077,
"epoch": 0.1401487663280116,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999724604182492,
"loss": 6.2145,
"mean_token_accuracy": 0.12926804050803184,
"num_tokens": 3202147.0,
"step": 1545
},
{
"entropy": 6.323010873794556,
"epoch": 0.14060232220609578,
"grad_norm": 1.015625,
"learning_rate": 0.0004999719518600229,
"loss": 6.0864,
"mean_token_accuracy": 0.13912193179130555,
"num_tokens": 3212428.0,
"step": 1550
},
{
"entropy": 6.246621894836426,
"epoch": 0.14105587808417996,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999714386493192,
"loss": 6.1409,
"mean_token_accuracy": 0.1369505174458027,
"num_tokens": 3222099.0,
"step": 1555
},
{
"entropy": 6.316340351104737,
"epoch": 0.14150943396226415,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999709207861489,
"loss": 6.0828,
"mean_token_accuracy": 0.13749494701623916,
"num_tokens": 3232079.0,
"step": 1560
},
{
"entropy": 6.204212045669555,
"epoch": 0.14196298984034833,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999703982705227,
"loss": 6.005,
"mean_token_accuracy": 0.14854954034090043,
"num_tokens": 3241762.0,
"step": 1565
},
{
"entropy": 6.180720520019531,
"epoch": 0.1424165457184325,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999698711024513,
"loss": 6.0869,
"mean_token_accuracy": 0.13718333020806311,
"num_tokens": 3252571.0,
"step": 1570
},
{
"entropy": 6.293680715560913,
"epoch": 0.1428701015965167,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999693392819456,
"loss": 6.1328,
"mean_token_accuracy": 0.1329519271850586,
"num_tokens": 3263542.0,
"step": 1575
},
{
"entropy": 6.287594938278199,
"epoch": 0.14332365747460088,
"grad_norm": 1.0,
"learning_rate": 0.0004999688028090166,
"loss": 6.1936,
"mean_token_accuracy": 0.13597823828458785,
"num_tokens": 3273295.0,
"step": 1580
},
{
"entropy": 6.440543842315674,
"epoch": 0.14377721335268506,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999682616836755,
"loss": 6.2261,
"mean_token_accuracy": 0.1362229973077774,
"num_tokens": 3283605.0,
"step": 1585
},
{
"entropy": 6.197280740737915,
"epoch": 0.14423076923076922,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999677159059334,
"loss": 6.0532,
"mean_token_accuracy": 0.14048462361097336,
"num_tokens": 3293837.0,
"step": 1590
},
{
"entropy": 6.210299730300903,
"epoch": 0.1446843251088534,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999671654758016,
"loss": 6.0692,
"mean_token_accuracy": 0.14433105364441873,
"num_tokens": 3304131.0,
"step": 1595
},
{
"entropy": 6.238543367385864,
"epoch": 0.14513788098693758,
"grad_norm": 1.140625,
"learning_rate": 0.0004999666103932915,
"loss": 6.1103,
"mean_token_accuracy": 0.13127385154366494,
"num_tokens": 3313543.0,
"step": 1600
},
{
"entropy": 6.23192195892334,
"epoch": 0.14559143686502177,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999660506584145,
"loss": 6.0745,
"mean_token_accuracy": 0.1385864198207855,
"num_tokens": 3324487.0,
"step": 1605
},
{
"entropy": 6.296097135543823,
"epoch": 0.14604499274310595,
"grad_norm": 0.921875,
"learning_rate": 0.0004999654862711823,
"loss": 6.1001,
"mean_token_accuracy": 0.13781747072935105,
"num_tokens": 3335094.0,
"step": 1610
},
{
"entropy": 6.208214139938354,
"epoch": 0.14649854862119013,
"grad_norm": 1.03125,
"learning_rate": 0.0004999649172316065,
"loss": 6.04,
"mean_token_accuracy": 0.13336168527603148,
"num_tokens": 3344252.0,
"step": 1615
},
{
"entropy": 6.273281383514404,
"epoch": 0.14695210449927432,
"grad_norm": 0.9609375,
"learning_rate": 0.000499964343539699,
"loss": 6.1072,
"mean_token_accuracy": 0.1359778583049774,
"num_tokens": 3354265.0,
"step": 1620
},
{
"entropy": 6.312763309478759,
"epoch": 0.1474056603773585,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999637651954714,
"loss": 6.2133,
"mean_token_accuracy": 0.13838177993893624,
"num_tokens": 3364616.0,
"step": 1625
},
{
"entropy": 6.300515222549438,
"epoch": 0.14785921625544268,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999631821989358,
"loss": 6.1601,
"mean_token_accuracy": 0.13766503930091858,
"num_tokens": 3373650.0,
"step": 1630
},
{
"entropy": 6.221302127838134,
"epoch": 0.14831277213352684,
"grad_norm": 1.03125,
"learning_rate": 0.0004999625945501043,
"loss": 6.0573,
"mean_token_accuracy": 0.1438056230545044,
"num_tokens": 3384590.0,
"step": 1635
},
{
"entropy": 6.271721410751343,
"epoch": 0.14876632801161102,
"grad_norm": 1.0546875,
"learning_rate": 0.000499962002248989,
"loss": 6.0793,
"mean_token_accuracy": 0.13807034716010094,
"num_tokens": 3395616.0,
"step": 1640
},
{
"entropy": 6.2336437702178955,
"epoch": 0.1492198838896952,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999614052956024,
"loss": 6.0885,
"mean_token_accuracy": 0.13720295056700707,
"num_tokens": 3405810.0,
"step": 1645
},
{
"entropy": 6.2871723651885985,
"epoch": 0.1496734397677794,
"grad_norm": 0.9375,
"learning_rate": 0.0004999608036899563,
"loss": 6.1248,
"mean_token_accuracy": 0.13411571234464645,
"num_tokens": 3415932.0,
"step": 1650
},
{
"entropy": 6.227649116516114,
"epoch": 0.15012699564586357,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999601974320636,
"loss": 6.0773,
"mean_token_accuracy": 0.13326597362756729,
"num_tokens": 3425626.0,
"step": 1655
},
{
"entropy": 6.198207044601441,
"epoch": 0.15058055152394775,
"grad_norm": 1.078125,
"learning_rate": 0.0004999595865219367,
"loss": 6.098,
"mean_token_accuracy": 0.13101402595639228,
"num_tokens": 3434837.0,
"step": 1660
},
{
"entropy": 6.252737283706665,
"epoch": 0.15103410740203194,
"grad_norm": 1.0,
"learning_rate": 0.0004999589709595881,
"loss": 6.0898,
"mean_token_accuracy": 0.14350827634334565,
"num_tokens": 3445238.0,
"step": 1665
},
{
"entropy": 6.187345504760742,
"epoch": 0.15148766328011612,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999583507450308,
"loss": 6.0692,
"mean_token_accuracy": 0.13529428243637084,
"num_tokens": 3456738.0,
"step": 1670
},
{
"entropy": 6.133850383758545,
"epoch": 0.1519412191582003,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999577258782774,
"loss": 5.9664,
"mean_token_accuracy": 0.13891834691166877,
"num_tokens": 3466296.0,
"step": 1675
},
{
"entropy": 6.328598546981811,
"epoch": 0.15239477503628446,
"grad_norm": 0.95703125,
"learning_rate": 0.000499957096359341,
"loss": 6.1502,
"mean_token_accuracy": 0.13001871183514596,
"num_tokens": 3476989.0,
"step": 1680
},
{
"entropy": 6.21191725730896,
"epoch": 0.15284833091436864,
"grad_norm": 1.078125,
"learning_rate": 0.0004999564621882343,
"loss": 5.9815,
"mean_token_accuracy": 0.14194255620241164,
"num_tokens": 3487050.0,
"step": 1685
},
{
"entropy": 6.256222677230835,
"epoch": 0.15330188679245282,
"grad_norm": 0.9375,
"learning_rate": 0.0004999558233649708,
"loss": 6.2052,
"mean_token_accuracy": 0.1338474340736866,
"num_tokens": 3497777.0,
"step": 1690
},
{
"entropy": 6.257563257217408,
"epoch": 0.153755442670537,
"grad_norm": 0.90625,
"learning_rate": 0.0004999551798895636,
"loss": 6.1012,
"mean_token_accuracy": 0.13192513063549996,
"num_tokens": 3509066.0,
"step": 1695
},
{
"entropy": 6.171849775314331,
"epoch": 0.1542089985486212,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999545317620258,
"loss": 6.0377,
"mean_token_accuracy": 0.13509414941072465,
"num_tokens": 3519567.0,
"step": 1700
},
{
"entropy": 6.180337810516358,
"epoch": 0.15466255442670537,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999538789823711,
"loss": 6.0677,
"mean_token_accuracy": 0.1376472532749176,
"num_tokens": 3529402.0,
"step": 1705
},
{
"entropy": 6.29026083946228,
"epoch": 0.15511611030478956,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999532215506127,
"loss": 6.0768,
"mean_token_accuracy": 0.14445007145404815,
"num_tokens": 3539805.0,
"step": 1710
},
{
"entropy": 6.146165990829468,
"epoch": 0.15556966618287374,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999525594667645,
"loss": 5.9557,
"mean_token_accuracy": 0.1396943673491478,
"num_tokens": 3550477.0,
"step": 1715
},
{
"entropy": 6.103784704208374,
"epoch": 0.15602322206095792,
"grad_norm": 0.96875,
"learning_rate": 0.00049995189273084,
"loss": 6.0572,
"mean_token_accuracy": 0.13744304925203324,
"num_tokens": 3560729.0,
"step": 1720
},
{
"entropy": 6.271568822860718,
"epoch": 0.15647677793904208,
"grad_norm": 0.98046875,
"learning_rate": 0.000499951221342853,
"loss": 6.1321,
"mean_token_accuracy": 0.1289873942732811,
"num_tokens": 3571538.0,
"step": 1725
},
{
"entropy": 6.223130464553833,
"epoch": 0.15693033381712626,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999505453028174,
"loss": 6.1212,
"mean_token_accuracy": 0.1351299576461315,
"num_tokens": 3583128.0,
"step": 1730
},
{
"entropy": 6.263720464706421,
"epoch": 0.15738388969521044,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999498646107472,
"loss": 6.0628,
"mean_token_accuracy": 0.13980068638920784,
"num_tokens": 3594358.0,
"step": 1735
},
{
"entropy": 6.227974891662598,
"epoch": 0.15783744557329463,
"grad_norm": 0.984375,
"learning_rate": 0.0004999491792666566,
"loss": 5.9844,
"mean_token_accuracy": 0.14432255774736405,
"num_tokens": 3604345.0,
"step": 1740
},
{
"entropy": 6.122195100784301,
"epoch": 0.1582910014513788,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999484892705597,
"loss": 6.0287,
"mean_token_accuracy": 0.13807731494307518,
"num_tokens": 3616129.0,
"step": 1745
},
{
"entropy": 6.275023746490478,
"epoch": 0.158744557329463,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999477946224705,
"loss": 6.0836,
"mean_token_accuracy": 0.14040916562080383,
"num_tokens": 3627458.0,
"step": 1750
},
{
"entropy": 6.181599187850952,
"epoch": 0.15919811320754718,
"grad_norm": 0.96875,
"learning_rate": 0.0004999470953224037,
"loss": 5.9586,
"mean_token_accuracy": 0.1387786313891411,
"num_tokens": 3636047.0,
"step": 1755
},
{
"entropy": 6.1431786060333256,
"epoch": 0.15965166908563136,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999463913703735,
"loss": 5.9542,
"mean_token_accuracy": 0.13981650844216348,
"num_tokens": 3646930.0,
"step": 1760
},
{
"entropy": 6.141822052001953,
"epoch": 0.16010522496371554,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999456827663948,
"loss": 6.0495,
"mean_token_accuracy": 0.1432150684297085,
"num_tokens": 3657279.0,
"step": 1765
},
{
"entropy": 6.173039388656616,
"epoch": 0.1605587808417997,
"grad_norm": 0.921875,
"learning_rate": 0.000499944969510482,
"loss": 5.9656,
"mean_token_accuracy": 0.141604046523571,
"num_tokens": 3667070.0,
"step": 1770
},
{
"entropy": 6.148836374282837,
"epoch": 0.16101233671988388,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999442516026498,
"loss": 5.9909,
"mean_token_accuracy": 0.13877027854323387,
"num_tokens": 3677560.0,
"step": 1775
},
{
"entropy": 6.135176277160644,
"epoch": 0.16146589259796806,
"grad_norm": 0.96875,
"learning_rate": 0.0004999435290429133,
"loss": 5.951,
"mean_token_accuracy": 0.14585819393396376,
"num_tokens": 3686701.0,
"step": 1780
},
{
"entropy": 6.194845485687256,
"epoch": 0.16191944847605225,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999428018312872,
"loss": 6.0453,
"mean_token_accuracy": 0.14197150394320487,
"num_tokens": 3696338.0,
"step": 1785
},
{
"entropy": 6.213848018646241,
"epoch": 0.16237300435413643,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999420699677867,
"loss": 6.0178,
"mean_token_accuracy": 0.13948137164115906,
"num_tokens": 3705752.0,
"step": 1790
},
{
"entropy": 6.173350286483765,
"epoch": 0.1628265602322206,
"grad_norm": 1.03125,
"learning_rate": 0.0004999413334524269,
"loss": 6.1086,
"mean_token_accuracy": 0.13366367369890214,
"num_tokens": 3716225.0,
"step": 1795
},
{
"entropy": 6.1766557693481445,
"epoch": 0.1632801161103048,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999405922852229,
"loss": 5.9812,
"mean_token_accuracy": 0.1403820961713791,
"num_tokens": 3726593.0,
"step": 1800
},
{
"entropy": 6.1498106002807615,
"epoch": 0.16373367198838898,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999398464661901,
"loss": 6.0292,
"mean_token_accuracy": 0.13524944633245467,
"num_tokens": 3737563.0,
"step": 1805
},
{
"entropy": 6.278463888168335,
"epoch": 0.16418722786647316,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999390959953441,
"loss": 6.0868,
"mean_token_accuracy": 0.13811569064855575,
"num_tokens": 3747752.0,
"step": 1810
},
{
"entropy": 6.124238109588623,
"epoch": 0.16464078374455732,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999383408727002,
"loss": 6.0109,
"mean_token_accuracy": 0.13542224168777467,
"num_tokens": 3758795.0,
"step": 1815
},
{
"entropy": 6.21006441116333,
"epoch": 0.1650943396226415,
"grad_norm": 1.046875,
"learning_rate": 0.0004999375810982741,
"loss": 6.0051,
"mean_token_accuracy": 0.14309126287698745,
"num_tokens": 3768690.0,
"step": 1820
},
{
"entropy": 6.147620582580567,
"epoch": 0.16554789550072568,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999368166720815,
"loss": 6.0588,
"mean_token_accuracy": 0.137413576990366,
"num_tokens": 3779018.0,
"step": 1825
},
{
"entropy": 6.241029739379883,
"epoch": 0.16600145137880987,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999360475941382,
"loss": 5.9743,
"mean_token_accuracy": 0.14609253406524658,
"num_tokens": 3788985.0,
"step": 1830
},
{
"entropy": 6.181500005722046,
"epoch": 0.16645500725689405,
"grad_norm": 0.890625,
"learning_rate": 0.00049993527386446,
"loss": 6.0916,
"mean_token_accuracy": 0.13334065973758696,
"num_tokens": 3800787.0,
"step": 1835
},
{
"entropy": 6.164537858963013,
"epoch": 0.16690856313497823,
"grad_norm": 1.03125,
"learning_rate": 0.0004999344954830633,
"loss": 5.9801,
"mean_token_accuracy": 0.1378617025911808,
"num_tokens": 3811381.0,
"step": 1840
},
{
"entropy": 6.188645315170288,
"epoch": 0.16736211901306242,
"grad_norm": 1.03125,
"learning_rate": 0.0004999337124499638,
"loss": 6.0844,
"mean_token_accuracy": 0.14289560168981552,
"num_tokens": 3822055.0,
"step": 1845
},
{
"entropy": 6.152506732940674,
"epoch": 0.1678156748911466,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999329247651777,
"loss": 6.0558,
"mean_token_accuracy": 0.13555024489760398,
"num_tokens": 3831900.0,
"step": 1850
},
{
"entropy": 6.184814882278443,
"epoch": 0.16826923076923078,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999321324287213,
"loss": 6.0046,
"mean_token_accuracy": 0.1400987111032009,
"num_tokens": 3842702.0,
"step": 1855
},
{
"entropy": 6.251664876937866,
"epoch": 0.16872278664731494,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999313354406114,
"loss": 5.986,
"mean_token_accuracy": 0.1382150486111641,
"num_tokens": 3852690.0,
"step": 1860
},
{
"entropy": 6.014293146133423,
"epoch": 0.16917634252539912,
"grad_norm": 0.90234375,
"learning_rate": 0.000499930533800864,
"loss": 5.9598,
"mean_token_accuracy": 0.15227705091238022,
"num_tokens": 3863040.0,
"step": 1865
},
{
"entropy": 6.209018230438232,
"epoch": 0.1696298984034833,
"grad_norm": 0.921875,
"learning_rate": 0.0004999297275094958,
"loss": 6.065,
"mean_token_accuracy": 0.13639552444219588,
"num_tokens": 3873613.0,
"step": 1870
},
{
"entropy": 6.046637344360351,
"epoch": 0.1700834542815675,
"grad_norm": 1.015625,
"learning_rate": 0.0004999289165665235,
"loss": 5.8751,
"mean_token_accuracy": 0.14939435720443725,
"num_tokens": 3883936.0,
"step": 1875
},
{
"entropy": 6.090919876098633,
"epoch": 0.17053701015965167,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999281009719639,
"loss": 5.8273,
"mean_token_accuracy": 0.15036647394299507,
"num_tokens": 3893788.0,
"step": 1880
},
{
"entropy": 6.0542162418365475,
"epoch": 0.17099056603773585,
"grad_norm": 1.03125,
"learning_rate": 0.0004999272807258339,
"loss": 5.912,
"mean_token_accuracy": 0.14638229012489318,
"num_tokens": 3903434.0,
"step": 1885
},
{
"entropy": 6.142886686325073,
"epoch": 0.17144412191582004,
"grad_norm": 0.9375,
"learning_rate": 0.0004999264558281504,
"loss": 5.9101,
"mean_token_accuracy": 0.14875417798757554,
"num_tokens": 3913476.0,
"step": 1890
},
{
"entropy": 5.993232488632202,
"epoch": 0.17189767779390422,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999256262789305,
"loss": 5.8601,
"mean_token_accuracy": 0.14939367920160293,
"num_tokens": 3923301.0,
"step": 1895
},
{
"entropy": 6.073164653778076,
"epoch": 0.1723512336719884,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999247920781913,
"loss": 6.0334,
"mean_token_accuracy": 0.13975203409790993,
"num_tokens": 3933335.0,
"step": 1900
},
{
"entropy": 6.233634042739868,
"epoch": 0.17280478955007256,
"grad_norm": 0.890625,
"learning_rate": 0.0004999239532259502,
"loss": 6.037,
"mean_token_accuracy": 0.13394607827067376,
"num_tokens": 3944171.0,
"step": 1905
},
{
"entropy": 6.214393186569214,
"epoch": 0.17325834542815674,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999231097222244,
"loss": 6.1071,
"mean_token_accuracy": 0.14265160858631135,
"num_tokens": 3954160.0,
"step": 1910
},
{
"entropy": 6.173836088180542,
"epoch": 0.17371190130624092,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999222615670312,
"loss": 5.9094,
"mean_token_accuracy": 0.14953405708074569,
"num_tokens": 3964110.0,
"step": 1915
},
{
"entropy": 6.1780534267425535,
"epoch": 0.1741654571843251,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999214087603885,
"loss": 6.0034,
"mean_token_accuracy": 0.14108952656388282,
"num_tokens": 3974293.0,
"step": 1920
},
{
"entropy": 6.1007123470306395,
"epoch": 0.1746190130624093,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999205513023135,
"loss": 5.9491,
"mean_token_accuracy": 0.1505696728825569,
"num_tokens": 3984346.0,
"step": 1925
},
{
"entropy": 6.014599132537842,
"epoch": 0.17507256894049347,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999196891928245,
"loss": 5.8944,
"mean_token_accuracy": 0.1506063535809517,
"num_tokens": 3992897.0,
"step": 1930
},
{
"entropy": 6.1384530544281,
"epoch": 0.17552612481857766,
"grad_norm": 0.96875,
"learning_rate": 0.0004999188224319388,
"loss": 5.929,
"mean_token_accuracy": 0.1495840311050415,
"num_tokens": 4002575.0,
"step": 1935
},
{
"entropy": 6.0449896335601805,
"epoch": 0.17597968069666184,
"grad_norm": 0.96875,
"learning_rate": 0.0004999179510196746,
"loss": 6.0023,
"mean_token_accuracy": 0.1404266342520714,
"num_tokens": 4013752.0,
"step": 1940
},
{
"entropy": 6.181995582580567,
"epoch": 0.17643323657474602,
"grad_norm": 1.03125,
"learning_rate": 0.0004999170749560498,
"loss": 5.8891,
"mean_token_accuracy": 0.14904519468545913,
"num_tokens": 4023718.0,
"step": 1945
},
{
"entropy": 6.080067110061646,
"epoch": 0.17688679245283018,
"grad_norm": 1.0625,
"learning_rate": 0.0004999161942410827,
"loss": 5.9877,
"mean_token_accuracy": 0.13862392008304597,
"num_tokens": 4034210.0,
"step": 1950
},
{
"entropy": 6.194681644439697,
"epoch": 0.17734034833091436,
"grad_norm": 1.015625,
"learning_rate": 0.0004999153088747913,
"loss": 5.9752,
"mean_token_accuracy": 0.14118978679180144,
"num_tokens": 4044868.0,
"step": 1955
},
{
"entropy": 6.018691873550415,
"epoch": 0.17779390420899854,
"grad_norm": 0.99609375,
"learning_rate": 0.000499914418857194,
"loss": 5.9759,
"mean_token_accuracy": 0.14467088282108306,
"num_tokens": 4055049.0,
"step": 1960
},
{
"entropy": 6.195721387863159,
"epoch": 0.17824746008708273,
"grad_norm": 0.953125,
"learning_rate": 0.0004999135241883092,
"loss": 6.0328,
"mean_token_accuracy": 0.13953902348876,
"num_tokens": 4065889.0,
"step": 1965
},
{
"entropy": 6.124079751968384,
"epoch": 0.1787010159651669,
"grad_norm": 1.078125,
"learning_rate": 0.0004999126248681555,
"loss": 6.0438,
"mean_token_accuracy": 0.13967278376221656,
"num_tokens": 4077288.0,
"step": 1970
},
{
"entropy": 6.1740796089172365,
"epoch": 0.1791545718432511,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999117208967513,
"loss": 5.9939,
"mean_token_accuracy": 0.14228827133774757,
"num_tokens": 4087402.0,
"step": 1975
},
{
"entropy": 6.136812305450439,
"epoch": 0.17960812772133528,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999108122741154,
"loss": 6.0412,
"mean_token_accuracy": 0.13958762884140014,
"num_tokens": 4098150.0,
"step": 1980
},
{
"entropy": 6.085280323028565,
"epoch": 0.18006168359941946,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999098990002666,
"loss": 5.9088,
"mean_token_accuracy": 0.14206465110182762,
"num_tokens": 4108134.0,
"step": 1985
},
{
"entropy": 6.158121109008789,
"epoch": 0.18051523947750364,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999089810752237,
"loss": 6.0448,
"mean_token_accuracy": 0.13666467741131783,
"num_tokens": 4118251.0,
"step": 1990
},
{
"entropy": 6.078621101379395,
"epoch": 0.1809687953555878,
"grad_norm": 0.921875,
"learning_rate": 0.000499908058499006,
"loss": 5.8876,
"mean_token_accuracy": 0.14455655217170715,
"num_tokens": 4129228.0,
"step": 1995
},
{
"entropy": 6.047148942947388,
"epoch": 0.18142235123367198,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999071312716321,
"loss": 5.9426,
"mean_token_accuracy": 0.15172967463731765,
"num_tokens": 4139391.0,
"step": 2000
},
{
"entropy": 6.093998432159424,
"epoch": 0.18187590711175616,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999061993931215,
"loss": 5.88,
"mean_token_accuracy": 0.15255183577537537,
"num_tokens": 4148891.0,
"step": 2005
},
{
"entropy": 6.305850887298584,
"epoch": 0.18232946298984035,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999052628634934,
"loss": 6.2083,
"mean_token_accuracy": 0.13769319280982018,
"num_tokens": 4159551.0,
"step": 2010
},
{
"entropy": 6.18409833908081,
"epoch": 0.18278301886792453,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999043216827671,
"loss": 6.0219,
"mean_token_accuracy": 0.1419105350971222,
"num_tokens": 4170055.0,
"step": 2015
},
{
"entropy": 6.019263553619385,
"epoch": 0.1832365747460087,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999033758509621,
"loss": 5.8725,
"mean_token_accuracy": 0.142510387301445,
"num_tokens": 4180767.0,
"step": 2020
},
{
"entropy": 6.1171001434326175,
"epoch": 0.1836901306240929,
"grad_norm": 0.8828125,
"learning_rate": 0.000499902425368098,
"loss": 5.99,
"mean_token_accuracy": 0.1484679900109768,
"num_tokens": 4190654.0,
"step": 2025
},
{
"entropy": 6.084097576141358,
"epoch": 0.18414368650217708,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999014702341944,
"loss": 5.9217,
"mean_token_accuracy": 0.14448753595352173,
"num_tokens": 4200753.0,
"step": 2030
},
{
"entropy": 6.135722398757935,
"epoch": 0.18459724238026126,
"grad_norm": 0.953125,
"learning_rate": 0.0004999005104492712,
"loss": 5.9985,
"mean_token_accuracy": 0.14054833576083184,
"num_tokens": 4210994.0,
"step": 2035
},
{
"entropy": 6.120487928390503,
"epoch": 0.18505079825834542,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998995460133481,
"loss": 5.9412,
"mean_token_accuracy": 0.14108474180102348,
"num_tokens": 4220883.0,
"step": 2040
},
{
"entropy": 6.127297973632812,
"epoch": 0.1855043541364296,
"grad_norm": 0.859375,
"learning_rate": 0.0004998985769264451,
"loss": 5.977,
"mean_token_accuracy": 0.13780092522501947,
"num_tokens": 4231269.0,
"step": 2045
},
{
"entropy": 6.101497316360474,
"epoch": 0.18595791001451378,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998976031885821,
"loss": 5.9776,
"mean_token_accuracy": 0.1418766401708126,
"num_tokens": 4241559.0,
"step": 2050
},
{
"entropy": 6.121247720718384,
"epoch": 0.18641146589259797,
"grad_norm": 1.09375,
"learning_rate": 0.0004998966247997794,
"loss": 5.8733,
"mean_token_accuracy": 0.14405595809221267,
"num_tokens": 4250870.0,
"step": 2055
},
{
"entropy": 6.015826034545898,
"epoch": 0.18686502177068215,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998956417600572,
"loss": 5.8246,
"mean_token_accuracy": 0.15058073103427888,
"num_tokens": 4260848.0,
"step": 2060
},
{
"entropy": 6.08300142288208,
"epoch": 0.18731857764876633,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998946540694358,
"loss": 5.9464,
"mean_token_accuracy": 0.1485067129135132,
"num_tokens": 4270973.0,
"step": 2065
},
{
"entropy": 6.14107756614685,
"epoch": 0.18777213352685052,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998936617279357,
"loss": 6.0361,
"mean_token_accuracy": 0.14370361119508743,
"num_tokens": 4281211.0,
"step": 2070
},
{
"entropy": 6.045505952835083,
"epoch": 0.1882256894049347,
"grad_norm": 1.046875,
"learning_rate": 0.0004998926647355773,
"loss": 5.8954,
"mean_token_accuracy": 0.14735701233148574,
"num_tokens": 4291425.0,
"step": 2075
},
{
"entropy": 6.1746431350708,
"epoch": 0.18867924528301888,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998916630923813,
"loss": 5.9993,
"mean_token_accuracy": 0.14234295561909677,
"num_tokens": 4301079.0,
"step": 2080
},
{
"entropy": 6.0917683124542235,
"epoch": 0.18913280116110304,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998906567983685,
"loss": 5.9599,
"mean_token_accuracy": 0.14310015216469765,
"num_tokens": 4312374.0,
"step": 2085
},
{
"entropy": 6.10637731552124,
"epoch": 0.18958635703918722,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998896458535595,
"loss": 5.8884,
"mean_token_accuracy": 0.15468773394823074,
"num_tokens": 4322810.0,
"step": 2090
},
{
"entropy": 6.0409440994262695,
"epoch": 0.1900399129172714,
"grad_norm": 1.0,
"learning_rate": 0.0004998886302579752,
"loss": 5.888,
"mean_token_accuracy": 0.14332466721534728,
"num_tokens": 4332082.0,
"step": 2095
},
{
"entropy": 6.115694570541382,
"epoch": 0.19049346879535559,
"grad_norm": 0.9609375,
"learning_rate": 0.000499887610011637,
"loss": 5.9771,
"mean_token_accuracy": 0.14411457404494285,
"num_tokens": 4342632.0,
"step": 2100
},
{
"entropy": 6.025098752975464,
"epoch": 0.19094702467343977,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998865851145656,
"loss": 5.9079,
"mean_token_accuracy": 0.14562978297472,
"num_tokens": 4353544.0,
"step": 2105
},
{
"entropy": 6.080524492263794,
"epoch": 0.19140058055152395,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998855555667822,
"loss": 6.0337,
"mean_token_accuracy": 0.13668391555547715,
"num_tokens": 4363636.0,
"step": 2110
},
{
"entropy": 6.174575281143189,
"epoch": 0.19185413642960814,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998845213683084,
"loss": 5.91,
"mean_token_accuracy": 0.14217197224497796,
"num_tokens": 4373672.0,
"step": 2115
},
{
"entropy": 5.952403831481933,
"epoch": 0.19230769230769232,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998834825191652,
"loss": 5.8609,
"mean_token_accuracy": 0.14928537905216216,
"num_tokens": 4383199.0,
"step": 2120
},
{
"entropy": 6.0944489479064945,
"epoch": 0.1927612481857765,
"grad_norm": 0.953125,
"learning_rate": 0.0004998824390193743,
"loss": 5.9295,
"mean_token_accuracy": 0.14216511845588684,
"num_tokens": 4394446.0,
"step": 2125
},
{
"entropy": 6.156260538101196,
"epoch": 0.19321480406386066,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998813908689573,
"loss": 5.9814,
"mean_token_accuracy": 0.14446035400032997,
"num_tokens": 4405882.0,
"step": 2130
},
{
"entropy": 6.021085166931153,
"epoch": 0.19366835994194484,
"grad_norm": 1.03125,
"learning_rate": 0.0004998803380679359,
"loss": 5.8395,
"mean_token_accuracy": 0.15445481687784196,
"num_tokens": 4415280.0,
"step": 2135
},
{
"entropy": 6.026923942565918,
"epoch": 0.19412191582002902,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998792806163317,
"loss": 6.0418,
"mean_token_accuracy": 0.1423596292734146,
"num_tokens": 4425755.0,
"step": 2140
},
{
"entropy": 6.0691611766815186,
"epoch": 0.1945754716981132,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998782185141667,
"loss": 5.899,
"mean_token_accuracy": 0.1436906062066555,
"num_tokens": 4436259.0,
"step": 2145
},
{
"entropy": 6.100642156600952,
"epoch": 0.1950290275761974,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998771517614629,
"loss": 5.9354,
"mean_token_accuracy": 0.14032572209835054,
"num_tokens": 4446423.0,
"step": 2150
},
{
"entropy": 6.079388666152954,
"epoch": 0.19548258345428157,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998760803582422,
"loss": 5.8765,
"mean_token_accuracy": 0.15558486431837082,
"num_tokens": 4456678.0,
"step": 2155
},
{
"entropy": 6.038434076309204,
"epoch": 0.19593613933236576,
"grad_norm": 1.0,
"learning_rate": 0.0004998750043045269,
"loss": 5.88,
"mean_token_accuracy": 0.1474220186471939,
"num_tokens": 4467031.0,
"step": 2160
},
{
"entropy": 6.084740686416626,
"epoch": 0.19638969521044994,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998739236003392,
"loss": 5.9457,
"mean_token_accuracy": 0.13701039925217628,
"num_tokens": 4477771.0,
"step": 2165
},
{
"entropy": 6.058000755310059,
"epoch": 0.1968432510885341,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998728382457014,
"loss": 5.9724,
"mean_token_accuracy": 0.14374302327632904,
"num_tokens": 4488426.0,
"step": 2170
},
{
"entropy": 6.157823610305786,
"epoch": 0.19729680696661828,
"grad_norm": 0.890625,
"learning_rate": 0.0004998717482406362,
"loss": 5.9512,
"mean_token_accuracy": 0.14008950740098952,
"num_tokens": 4499735.0,
"step": 2175
},
{
"entropy": 6.083208417892456,
"epoch": 0.19775036284470246,
"grad_norm": 1.0,
"learning_rate": 0.0004998706535851657,
"loss": 5.9067,
"mean_token_accuracy": 0.1461469426751137,
"num_tokens": 4509500.0,
"step": 2180
},
{
"entropy": 6.037715578079224,
"epoch": 0.19820391872278664,
"grad_norm": 0.890625,
"learning_rate": 0.0004998695542793129,
"loss": 5.9079,
"mean_token_accuracy": 0.14525417983531952,
"num_tokens": 4520431.0,
"step": 2185
},
{
"entropy": 6.084038925170899,
"epoch": 0.19865747460087083,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998684503231003,
"loss": 5.9846,
"mean_token_accuracy": 0.14418850764632224,
"num_tokens": 4532294.0,
"step": 2190
},
{
"entropy": 6.027761554718017,
"epoch": 0.199111030478955,
"grad_norm": 0.93359375,
"learning_rate": 0.000499867341716551,
"loss": 5.8985,
"mean_token_accuracy": 0.1443702645599842,
"num_tokens": 4542488.0,
"step": 2195
},
{
"entropy": 6.043829488754272,
"epoch": 0.1995645863570392,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998662284596878,
"loss": 5.8528,
"mean_token_accuracy": 0.1418704494833946,
"num_tokens": 4552933.0,
"step": 2200
},
{
"entropy": 6.060906505584716,
"epoch": 0.20001814223512338,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998651105525335,
"loss": 5.8336,
"mean_token_accuracy": 0.15473553836345671,
"num_tokens": 4562373.0,
"step": 2205
},
{
"entropy": 5.9967522621154785,
"epoch": 0.20047169811320756,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998639879951115,
"loss": 5.9557,
"mean_token_accuracy": 0.14503294080495835,
"num_tokens": 4572715.0,
"step": 2210
},
{
"entropy": 6.131484937667847,
"epoch": 0.2009252539912917,
"grad_norm": 0.92578125,
"learning_rate": 0.000499862860787445,
"loss": 5.8461,
"mean_token_accuracy": 0.1420556351542473,
"num_tokens": 4582782.0,
"step": 2215
},
{
"entropy": 6.036686944961548,
"epoch": 0.2013788098693759,
"grad_norm": 0.83984375,
"learning_rate": 0.0004998617289295571,
"loss": 5.9439,
"mean_token_accuracy": 0.14234896376729012,
"num_tokens": 4594108.0,
"step": 2220
},
{
"entropy": 6.102933645248413,
"epoch": 0.20183236574746008,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998605924214714,
"loss": 5.9103,
"mean_token_accuracy": 0.14780050218105317,
"num_tokens": 4604732.0,
"step": 2225
},
{
"entropy": 6.049504566192627,
"epoch": 0.20228592162554426,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998594512632114,
"loss": 6.0456,
"mean_token_accuracy": 0.13591379299759865,
"num_tokens": 4614998.0,
"step": 2230
},
{
"entropy": 6.057260751724243,
"epoch": 0.20273947750362845,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998583054548006,
"loss": 5.8648,
"mean_token_accuracy": 0.14422834739089013,
"num_tokens": 4624665.0,
"step": 2235
},
{
"entropy": 6.085105228424072,
"epoch": 0.20319303338171263,
"grad_norm": 1.09375,
"learning_rate": 0.0004998571549962628,
"loss": 5.8612,
"mean_token_accuracy": 0.15319595485925674,
"num_tokens": 4633731.0,
"step": 2240
},
{
"entropy": 6.046474456787109,
"epoch": 0.2036465892597968,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998559998876216,
"loss": 5.8897,
"mean_token_accuracy": 0.14516430646181105,
"num_tokens": 4643421.0,
"step": 2245
},
{
"entropy": 6.105594730377197,
"epoch": 0.204100145137881,
"grad_norm": 1.015625,
"learning_rate": 0.0004998548401289012,
"loss": 5.9729,
"mean_token_accuracy": 0.14359490424394608,
"num_tokens": 4653795.0,
"step": 2250
},
{
"entropy": 6.111778020858765,
"epoch": 0.20455370101596518,
"grad_norm": 0.875,
"learning_rate": 0.0004998536757201253,
"loss": 5.9528,
"mean_token_accuracy": 0.14500485807657243,
"num_tokens": 4664840.0,
"step": 2255
},
{
"entropy": 6.058202457427979,
"epoch": 0.20500725689404933,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998525066613181,
"loss": 5.9537,
"mean_token_accuracy": 0.14941285997629167,
"num_tokens": 4675508.0,
"step": 2260
},
{
"entropy": 6.049920320510864,
"epoch": 0.20546081277213352,
"grad_norm": 1.015625,
"learning_rate": 0.0004998513329525038,
"loss": 5.8734,
"mean_token_accuracy": 0.15032386034727097,
"num_tokens": 4685995.0,
"step": 2265
},
{
"entropy": 6.082958173751831,
"epoch": 0.2059143686502177,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998501545937066,
"loss": 5.9052,
"mean_token_accuracy": 0.14433911442756653,
"num_tokens": 4696659.0,
"step": 2270
},
{
"entropy": 6.0467150688171385,
"epoch": 0.20636792452830188,
"grad_norm": 0.9609375,
"learning_rate": 0.000499848971584951,
"loss": 5.8595,
"mean_token_accuracy": 0.15614939630031585,
"num_tokens": 4706438.0,
"step": 2275
},
{
"entropy": 5.930180215835572,
"epoch": 0.20682148040638607,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998477839262611,
"loss": 5.8678,
"mean_token_accuracy": 0.1469499036669731,
"num_tokens": 4716402.0,
"step": 2280
},
{
"entropy": 6.050282192230225,
"epoch": 0.20727503628447025,
"grad_norm": 0.85546875,
"learning_rate": 0.000499846591617662,
"loss": 5.9457,
"mean_token_accuracy": 0.1466635584831238,
"num_tokens": 4728488.0,
"step": 2285
},
{
"entropy": 6.057934761047363,
"epoch": 0.20772859216255443,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998453946591779,
"loss": 5.976,
"mean_token_accuracy": 0.14741047471761703,
"num_tokens": 4738693.0,
"step": 2290
},
{
"entropy": 6.034505462646484,
"epoch": 0.20818214804063861,
"grad_norm": 0.9375,
"learning_rate": 0.0004998441930508339,
"loss": 5.8617,
"mean_token_accuracy": 0.149900358915329,
"num_tokens": 4748895.0,
"step": 2295
},
{
"entropy": 6.12192120552063,
"epoch": 0.2086357039187228,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998429867926547,
"loss": 5.9731,
"mean_token_accuracy": 0.1417195826768875,
"num_tokens": 4760135.0,
"step": 2300
},
{
"entropy": 6.1114363193511965,
"epoch": 0.20908925979680695,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998417758846651,
"loss": 6.0106,
"mean_token_accuracy": 0.1492314174771309,
"num_tokens": 4769935.0,
"step": 2305
},
{
"entropy": 6.163050842285156,
"epoch": 0.20954281567489114,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998405603268903,
"loss": 5.9454,
"mean_token_accuracy": 0.14845321476459503,
"num_tokens": 4779520.0,
"step": 2310
},
{
"entropy": 5.973990440368652,
"epoch": 0.20999637155297532,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998393401193554,
"loss": 5.8416,
"mean_token_accuracy": 0.15259672403335572,
"num_tokens": 4791256.0,
"step": 2315
},
{
"entropy": 5.966838359832764,
"epoch": 0.2104499274310595,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998381152620857,
"loss": 5.811,
"mean_token_accuracy": 0.15300733894109725,
"num_tokens": 4801334.0,
"step": 2320
},
{
"entropy": 5.96552734375,
"epoch": 0.21090348330914369,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998368857551067,
"loss": 5.7882,
"mean_token_accuracy": 0.15469710677862167,
"num_tokens": 4811361.0,
"step": 2325
},
{
"entropy": 6.016611623764038,
"epoch": 0.21135703918722787,
"grad_norm": 0.9375,
"learning_rate": 0.0004998356515984433,
"loss": 5.85,
"mean_token_accuracy": 0.1501396507024765,
"num_tokens": 4821190.0,
"step": 2330
},
{
"entropy": 6.041166496276856,
"epoch": 0.21181059506531205,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998344127921213,
"loss": 5.8821,
"mean_token_accuracy": 0.1504140943288803,
"num_tokens": 4830694.0,
"step": 2335
},
{
"entropy": 5.997649621963501,
"epoch": 0.21226415094339623,
"grad_norm": 0.8359375,
"learning_rate": 0.0004998331693361666,
"loss": 5.8687,
"mean_token_accuracy": 0.1483940526843071,
"num_tokens": 4841344.0,
"step": 2340
},
{
"entropy": 6.003773546218872,
"epoch": 0.21271770682148042,
"grad_norm": 0.953125,
"learning_rate": 0.0004998319212306046,
"loss": 5.8398,
"mean_token_accuracy": 0.14764055758714675,
"num_tokens": 4851161.0,
"step": 2345
},
{
"entropy": 5.979840612411499,
"epoch": 0.21317126269956457,
"grad_norm": 0.94140625,
"learning_rate": 0.000499830668475461,
"loss": 5.831,
"mean_token_accuracy": 0.1507522940635681,
"num_tokens": 4860740.0,
"step": 2350
},
{
"entropy": 6.029822587966919,
"epoch": 0.21362481857764876,
"grad_norm": 0.921875,
"learning_rate": 0.000499829411070762,
"loss": 5.856,
"mean_token_accuracy": 0.147916804254055,
"num_tokens": 4871733.0,
"step": 2355
},
{
"entropy": 5.999858570098877,
"epoch": 0.21407837445573294,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998281490165335,
"loss": 5.9117,
"mean_token_accuracy": 0.14421676099300385,
"num_tokens": 4883617.0,
"step": 2360
},
{
"entropy": 6.043055486679077,
"epoch": 0.21453193033381712,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998268823128016,
"loss": 5.8607,
"mean_token_accuracy": 0.14836216866970062,
"num_tokens": 4894519.0,
"step": 2365
},
{
"entropy": 6.004064750671387,
"epoch": 0.2149854862119013,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998256109595923,
"loss": 5.8134,
"mean_token_accuracy": 0.14691278859972953,
"num_tokens": 4904965.0,
"step": 2370
},
{
"entropy": 5.955627918243408,
"epoch": 0.2154390420899855,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998243349569323,
"loss": 5.7279,
"mean_token_accuracy": 0.157587631046772,
"num_tokens": 4914618.0,
"step": 2375
},
{
"entropy": 5.92179274559021,
"epoch": 0.21589259796806967,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998230543048477,
"loss": 5.7896,
"mean_token_accuracy": 0.15410214364528657,
"num_tokens": 4924341.0,
"step": 2380
},
{
"entropy": 5.918932485580444,
"epoch": 0.21634615384615385,
"grad_norm": 0.92578125,
"learning_rate": 0.000499821769003365,
"loss": 5.7633,
"mean_token_accuracy": 0.15452138409018518,
"num_tokens": 4934231.0,
"step": 2385
},
{
"entropy": 5.887126874923706,
"epoch": 0.21679970972423804,
"grad_norm": 0.984375,
"learning_rate": 0.0004998204790525109,
"loss": 5.7905,
"mean_token_accuracy": 0.1511450305581093,
"num_tokens": 4944303.0,
"step": 2390
},
{
"entropy": 6.001142930984497,
"epoch": 0.2172532656023222,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998191844523119,
"loss": 5.8693,
"mean_token_accuracy": 0.15268588662147523,
"num_tokens": 4954571.0,
"step": 2395
},
{
"entropy": 5.987411212921143,
"epoch": 0.21770682148040638,
"grad_norm": 0.96875,
"learning_rate": 0.000499817885202795,
"loss": 5.848,
"mean_token_accuracy": 0.14666955098509787,
"num_tokens": 4964418.0,
"step": 2400
},
{
"entropy": 6.03399133682251,
"epoch": 0.21816037735849056,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998165813039869,
"loss": 5.8291,
"mean_token_accuracy": 0.1442165307700634,
"num_tokens": 4975036.0,
"step": 2405
},
{
"entropy": 5.994223356246948,
"epoch": 0.21861393323657474,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998152727559145,
"loss": 5.9302,
"mean_token_accuracy": 0.14429473355412484,
"num_tokens": 4985272.0,
"step": 2410
},
{
"entropy": 6.091732931137085,
"epoch": 0.21906748911465893,
"grad_norm": 0.9140625,
"learning_rate": 0.000499813959558605,
"loss": 5.8548,
"mean_token_accuracy": 0.15049488544464112,
"num_tokens": 4995187.0,
"step": 2415
},
{
"entropy": 6.026802444458008,
"epoch": 0.2195210449927431,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998126417120856,
"loss": 5.9175,
"mean_token_accuracy": 0.15209949463605882,
"num_tokens": 5005013.0,
"step": 2420
},
{
"entropy": 6.094258975982666,
"epoch": 0.2199746008708273,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998113192163835,
"loss": 5.9259,
"mean_token_accuracy": 0.14370185285806655,
"num_tokens": 5016789.0,
"step": 2425
},
{
"entropy": 5.985634708404541,
"epoch": 0.22042815674891147,
"grad_norm": 0.90234375,
"learning_rate": 0.000499809992071526,
"loss": 5.8749,
"mean_token_accuracy": 0.15012458711862564,
"num_tokens": 5026913.0,
"step": 2430
},
{
"entropy": 6.017763757705689,
"epoch": 0.22088171262699566,
"grad_norm": 0.953125,
"learning_rate": 0.0004998086602775406,
"loss": 5.9623,
"mean_token_accuracy": 0.1502877563238144,
"num_tokens": 5037501.0,
"step": 2435
},
{
"entropy": 5.990037059783935,
"epoch": 0.2213352685050798,
"grad_norm": 0.984375,
"learning_rate": 0.0004998073238344547,
"loss": 5.8105,
"mean_token_accuracy": 0.144505974650383,
"num_tokens": 5048478.0,
"step": 2440
},
{
"entropy": 6.0197502136230465,
"epoch": 0.221788824383164,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998059827422962,
"loss": 5.8716,
"mean_token_accuracy": 0.13962874338030815,
"num_tokens": 5059118.0,
"step": 2445
},
{
"entropy": 6.010491847991943,
"epoch": 0.22224238026124818,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998046370010926,
"loss": 5.8321,
"mean_token_accuracy": 0.15523343831300734,
"num_tokens": 5070180.0,
"step": 2450
},
{
"entropy": 6.002876043319702,
"epoch": 0.22269593613933236,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998032866108718,
"loss": 5.8404,
"mean_token_accuracy": 0.15497579723596572,
"num_tokens": 5079912.0,
"step": 2455
},
{
"entropy": 6.074828386306763,
"epoch": 0.22314949201741655,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998019315716618,
"loss": 5.9112,
"mean_token_accuracy": 0.15170105695724487,
"num_tokens": 5090549.0,
"step": 2460
},
{
"entropy": 5.936306381225586,
"epoch": 0.22360304789550073,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998005718834905,
"loss": 5.767,
"mean_token_accuracy": 0.15424081236124038,
"num_tokens": 5100880.0,
"step": 2465
},
{
"entropy": 5.983072996139526,
"epoch": 0.2240566037735849,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997992075463862,
"loss": 5.8865,
"mean_token_accuracy": 0.1511937439441681,
"num_tokens": 5111796.0,
"step": 2470
},
{
"entropy": 6.071778011322022,
"epoch": 0.2245101596516691,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997978385603769,
"loss": 5.8722,
"mean_token_accuracy": 0.14768328070640563,
"num_tokens": 5123468.0,
"step": 2475
},
{
"entropy": 5.959640073776245,
"epoch": 0.22496371552975328,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997964649254911,
"loss": 5.8051,
"mean_token_accuracy": 0.15414749532938005,
"num_tokens": 5132925.0,
"step": 2480
},
{
"entropy": 5.989591789245606,
"epoch": 0.22541727140783743,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997950866417572,
"loss": 5.8777,
"mean_token_accuracy": 0.1485254153609276,
"num_tokens": 5143445.0,
"step": 2485
},
{
"entropy": 5.993931770324707,
"epoch": 0.22587082728592162,
"grad_norm": 1.078125,
"learning_rate": 0.0004997937037092034,
"loss": 5.8058,
"mean_token_accuracy": 0.14833374321460724,
"num_tokens": 5153176.0,
"step": 2490
},
{
"entropy": 6.067162322998047,
"epoch": 0.2263243831640058,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997923161278587,
"loss": 5.8445,
"mean_token_accuracy": 0.14823048263788224,
"num_tokens": 5162765.0,
"step": 2495
},
{
"entropy": 5.973460483551025,
"epoch": 0.22677793904208998,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997909238977514,
"loss": 5.8977,
"mean_token_accuracy": 0.150830414891243,
"num_tokens": 5173390.0,
"step": 2500
},
{
"entropy": 5.978850793838501,
"epoch": 0.22723149492017417,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997895270189108,
"loss": 5.8062,
"mean_token_accuracy": 0.1472632072865963,
"num_tokens": 5184212.0,
"step": 2505
},
{
"entropy": 5.894932794570923,
"epoch": 0.22768505079825835,
"grad_norm": 0.9375,
"learning_rate": 0.0004997881254913653,
"loss": 5.7866,
"mean_token_accuracy": 0.14934516549110413,
"num_tokens": 5194311.0,
"step": 2510
},
{
"entropy": 5.951768064498902,
"epoch": 0.22813860667634253,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997867193151441,
"loss": 5.7758,
"mean_token_accuracy": 0.1520687684416771,
"num_tokens": 5204708.0,
"step": 2515
},
{
"entropy": 5.913540315628052,
"epoch": 0.22859216255442671,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997853084902763,
"loss": 5.7608,
"mean_token_accuracy": 0.15067852139472962,
"num_tokens": 5214566.0,
"step": 2520
},
{
"entropy": 5.952923536300659,
"epoch": 0.2290457184325109,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997838930167911,
"loss": 5.8035,
"mean_token_accuracy": 0.15393685698509216,
"num_tokens": 5224762.0,
"step": 2525
},
{
"entropy": 6.007725048065185,
"epoch": 0.22949927431059505,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997824728947176,
"loss": 5.8629,
"mean_token_accuracy": 0.14823582395911217,
"num_tokens": 5235146.0,
"step": 2530
},
{
"entropy": 6.064925336837769,
"epoch": 0.22995283018867924,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997810481240853,
"loss": 5.8276,
"mean_token_accuracy": 0.1468263030052185,
"num_tokens": 5243867.0,
"step": 2535
},
{
"entropy": 5.916410827636719,
"epoch": 0.23040638606676342,
"grad_norm": 0.875,
"learning_rate": 0.0004997796187049238,
"loss": 5.7346,
"mean_token_accuracy": 0.1514027461409569,
"num_tokens": 5254279.0,
"step": 2540
},
{
"entropy": 6.060354948043823,
"epoch": 0.2308599419448476,
"grad_norm": 0.96875,
"learning_rate": 0.0004997781846372623,
"loss": 5.8574,
"mean_token_accuracy": 0.14669395685195924,
"num_tokens": 5264769.0,
"step": 2545
},
{
"entropy": 5.961148881912232,
"epoch": 0.23131349782293179,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997767459211307,
"loss": 5.8023,
"mean_token_accuracy": 0.14596094489097594,
"num_tokens": 5274291.0,
"step": 2550
},
{
"entropy": 5.973608541488647,
"epoch": 0.23176705370101597,
"grad_norm": 1.015625,
"learning_rate": 0.0004997753025565587,
"loss": 5.751,
"mean_token_accuracy": 0.15511572808027269,
"num_tokens": 5284366.0,
"step": 2555
},
{
"entropy": 5.983015012741089,
"epoch": 0.23222060957910015,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997738545435763,
"loss": 5.8651,
"mean_token_accuracy": 0.14840636104345323,
"num_tokens": 5294494.0,
"step": 2560
},
{
"entropy": 5.955273532867432,
"epoch": 0.23267416545718433,
"grad_norm": 0.87890625,
"learning_rate": 0.0004997724018822133,
"loss": 5.8663,
"mean_token_accuracy": 0.14921268746256827,
"num_tokens": 5305606.0,
"step": 2565
},
{
"entropy": 5.970898771286011,
"epoch": 0.23312772133526852,
"grad_norm": 0.96875,
"learning_rate": 0.0004997709445724996,
"loss": 5.856,
"mean_token_accuracy": 0.1487169645726681,
"num_tokens": 5315677.0,
"step": 2570
},
{
"entropy": 6.037318992614746,
"epoch": 0.23358127721335267,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997694826144656,
"loss": 5.9139,
"mean_token_accuracy": 0.14366162866353988,
"num_tokens": 5326453.0,
"step": 2575
},
{
"entropy": 5.969986534118652,
"epoch": 0.23403483309143686,
"grad_norm": 0.796875,
"learning_rate": 0.0004997680160081414,
"loss": 5.805,
"mean_token_accuracy": 0.15401693433523178,
"num_tokens": 5337591.0,
"step": 2580
},
{
"entropy": 5.923111820220948,
"epoch": 0.23448838896952104,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997665447535573,
"loss": 5.798,
"mean_token_accuracy": 0.15265310406684876,
"num_tokens": 5349400.0,
"step": 2585
},
{
"entropy": 6.048951911926269,
"epoch": 0.23494194484760522,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997650688507438,
"loss": 5.7909,
"mean_token_accuracy": 0.1504625990986824,
"num_tokens": 5359984.0,
"step": 2590
},
{
"entropy": 5.95233383178711,
"epoch": 0.2353955007256894,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997635882997314,
"loss": 5.8509,
"mean_token_accuracy": 0.15640779882669448,
"num_tokens": 5369254.0,
"step": 2595
},
{
"entropy": 5.96616530418396,
"epoch": 0.2358490566037736,
"grad_norm": 0.90625,
"learning_rate": 0.0004997621031005507,
"loss": 5.7868,
"mean_token_accuracy": 0.15061139017343522,
"num_tokens": 5379648.0,
"step": 2600
},
{
"entropy": 5.9050251007080075,
"epoch": 0.23630261248185777,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997606132532325,
"loss": 5.7438,
"mean_token_accuracy": 0.1579679898917675,
"num_tokens": 5390321.0,
"step": 2605
},
{
"entropy": 5.978159952163696,
"epoch": 0.23675616835994195,
"grad_norm": 1.03125,
"learning_rate": 0.0004997591187578075,
"loss": 5.7975,
"mean_token_accuracy": 0.15755213797092438,
"num_tokens": 5401078.0,
"step": 2610
},
{
"entropy": 5.922601938247681,
"epoch": 0.23720972423802614,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997576196143067,
"loss": 5.7705,
"mean_token_accuracy": 0.15626103281974793,
"num_tokens": 5411977.0,
"step": 2615
},
{
"entropy": 5.919599151611328,
"epoch": 0.2376632801161103,
"grad_norm": 0.92578125,
"learning_rate": 0.000499756115822761,
"loss": 5.7429,
"mean_token_accuracy": 0.14914836585521699,
"num_tokens": 5422967.0,
"step": 2620
},
{
"entropy": 6.0454573154449465,
"epoch": 0.23811683599419448,
"grad_norm": 0.9375,
"learning_rate": 0.0004997546073832015,
"loss": 5.7987,
"mean_token_accuracy": 0.16146147549152373,
"num_tokens": 5432260.0,
"step": 2625
},
{
"entropy": 5.874481344223023,
"epoch": 0.23857039187227866,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997530942956595,
"loss": 5.7685,
"mean_token_accuracy": 0.15337550789117813,
"num_tokens": 5441706.0,
"step": 2630
},
{
"entropy": 5.954086065292358,
"epoch": 0.23902394775036284,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997515765601662,
"loss": 5.7734,
"mean_token_accuracy": 0.16176194995641707,
"num_tokens": 5452397.0,
"step": 2635
},
{
"entropy": 5.8905235767364506,
"epoch": 0.23947750362844702,
"grad_norm": 0.796875,
"learning_rate": 0.0004997500541767531,
"loss": 5.732,
"mean_token_accuracy": 0.15274471640586854,
"num_tokens": 5463762.0,
"step": 2640
},
{
"entropy": 5.96301908493042,
"epoch": 0.2399310595065312,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997485271454516,
"loss": 5.8042,
"mean_token_accuracy": 0.14668141603469848,
"num_tokens": 5474622.0,
"step": 2645
},
{
"entropy": 5.9286675453186035,
"epoch": 0.2403846153846154,
"grad_norm": 1.015625,
"learning_rate": 0.0004997469954662933,
"loss": 5.7846,
"mean_token_accuracy": 0.1540173590183258,
"num_tokens": 5484697.0,
"step": 2650
},
{
"entropy": 5.917514657974243,
"epoch": 0.24083817126269957,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997454591393097,
"loss": 5.7995,
"mean_token_accuracy": 0.15156363248825072,
"num_tokens": 5496433.0,
"step": 2655
},
{
"entropy": 6.03740963935852,
"epoch": 0.24129172714078376,
"grad_norm": 0.890625,
"learning_rate": 0.0004997439181645329,
"loss": 5.9087,
"mean_token_accuracy": 0.14503674134612082,
"num_tokens": 5507917.0,
"step": 2660
},
{
"entropy": 6.060143232345581,
"epoch": 0.2417452830188679,
"grad_norm": 0.9375,
"learning_rate": 0.0004997423725419945,
"loss": 5.8745,
"mean_token_accuracy": 0.14898139089345933,
"num_tokens": 5519159.0,
"step": 2665
},
{
"entropy": 5.990707302093506,
"epoch": 0.2421988388969521,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997408222717267,
"loss": 5.8503,
"mean_token_accuracy": 0.15473444908857345,
"num_tokens": 5529180.0,
"step": 2670
},
{
"entropy": 5.912744617462158,
"epoch": 0.24265239477503628,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997392673537612,
"loss": 5.7518,
"mean_token_accuracy": 0.1590703994035721,
"num_tokens": 5539363.0,
"step": 2675
},
{
"entropy": 5.91375207901001,
"epoch": 0.24310595065312046,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997377077881306,
"loss": 5.8534,
"mean_token_accuracy": 0.15084453225135802,
"num_tokens": 5549841.0,
"step": 2680
},
{
"entropy": 5.941425514221192,
"epoch": 0.24355950653120464,
"grad_norm": 0.953125,
"learning_rate": 0.0004997361435748668,
"loss": 5.7527,
"mean_token_accuracy": 0.1558105766773224,
"num_tokens": 5560299.0,
"step": 2685
},
{
"entropy": 5.983996868133545,
"epoch": 0.24401306240928883,
"grad_norm": 0.88671875,
"learning_rate": 0.0004997345747140024,
"loss": 5.818,
"mean_token_accuracy": 0.1520366132259369,
"num_tokens": 5571081.0,
"step": 2690
},
{
"entropy": 5.893987655639648,
"epoch": 0.244466618287373,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997330012055695,
"loss": 5.7623,
"mean_token_accuracy": 0.1579376369714737,
"num_tokens": 5581389.0,
"step": 2695
},
{
"entropy": 5.951496028900147,
"epoch": 0.2449201741654572,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997314230496011,
"loss": 5.7861,
"mean_token_accuracy": 0.14917654246091844,
"num_tokens": 5592347.0,
"step": 2700
},
{
"entropy": 5.928359746932983,
"epoch": 0.24537373004354138,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997298402461296,
"loss": 5.7622,
"mean_token_accuracy": 0.1548761934041977,
"num_tokens": 5603411.0,
"step": 2705
},
{
"entropy": 5.997703504562378,
"epoch": 0.24582728592162553,
"grad_norm": 0.83984375,
"learning_rate": 0.0004997282527951876,
"loss": 5.7484,
"mean_token_accuracy": 0.1618112340569496,
"num_tokens": 5614615.0,
"step": 2710
},
{
"entropy": 5.855966997146607,
"epoch": 0.24628084179970972,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997266606968081,
"loss": 5.7219,
"mean_token_accuracy": 0.1555338978767395,
"num_tokens": 5624582.0,
"step": 2715
},
{
"entropy": 5.882028532028198,
"epoch": 0.2467343976777939,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997250639510241,
"loss": 5.7521,
"mean_token_accuracy": 0.1485395088791847,
"num_tokens": 5634310.0,
"step": 2720
},
{
"entropy": 5.959417104721069,
"epoch": 0.24718795355587808,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997234625578684,
"loss": 5.8284,
"mean_token_accuracy": 0.1493539556860924,
"num_tokens": 5644018.0,
"step": 2725
},
{
"entropy": 5.905500411987305,
"epoch": 0.24764150943396226,
"grad_norm": 0.953125,
"learning_rate": 0.0004997218565173742,
"loss": 5.7685,
"mean_token_accuracy": 0.15543410778045655,
"num_tokens": 5653592.0,
"step": 2730
},
{
"entropy": 5.994117975234985,
"epoch": 0.24809506531204645,
"grad_norm": 0.984375,
"learning_rate": 0.0004997202458295748,
"loss": 5.829,
"mean_token_accuracy": 0.15618719160556793,
"num_tokens": 5662812.0,
"step": 2735
},
{
"entropy": 5.947774648666382,
"epoch": 0.24854862119013063,
"grad_norm": 0.921875,
"learning_rate": 0.0004997186304945034,
"loss": 5.7997,
"mean_token_accuracy": 0.15028751492500306,
"num_tokens": 5672939.0,
"step": 2740
},
{
"entropy": 5.8947399139404295,
"epoch": 0.24900217706821481,
"grad_norm": 0.890625,
"learning_rate": 0.0004997170105121935,
"loss": 5.7789,
"mean_token_accuracy": 0.15528277158737183,
"num_tokens": 5683350.0,
"step": 2745
},
{
"entropy": 6.005437898635864,
"epoch": 0.249455732946299,
"grad_norm": 0.828125,
"learning_rate": 0.0004997153858826786,
"loss": 5.8528,
"mean_token_accuracy": 0.1439196802675724,
"num_tokens": 5694516.0,
"step": 2750
},
{
"entropy": 5.938060188293457,
"epoch": 0.24990928882438315,
"grad_norm": 0.9375,
"learning_rate": 0.0004997137566059922,
"loss": 5.6847,
"mean_token_accuracy": 0.15682510733604432,
"num_tokens": 5705443.0,
"step": 2755
},
{
"entropy": 5.872813892364502,
"epoch": 0.25036284470246734,
"grad_norm": 0.921875,
"learning_rate": 0.0004997121226821681,
"loss": 5.7821,
"mean_token_accuracy": 0.15022583901882172,
"num_tokens": 5715933.0,
"step": 2760
},
{
"entropy": 5.941016101837159,
"epoch": 0.25081640058055155,
"grad_norm": 0.953125,
"learning_rate": 0.00049971048411124,
"loss": 5.8439,
"mean_token_accuracy": 0.14352803230285643,
"num_tokens": 5726630.0,
"step": 2765
},
{
"entropy": 6.003946161270141,
"epoch": 0.2512699564586357,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997088408932418,
"loss": 5.7792,
"mean_token_accuracy": 0.1525881379842758,
"num_tokens": 5736890.0,
"step": 2770
},
{
"entropy": 5.804270029067993,
"epoch": 0.25172351233671986,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997071930282076,
"loss": 5.6826,
"mean_token_accuracy": 0.16186858713626862,
"num_tokens": 5747042.0,
"step": 2775
},
{
"entropy": 5.935064744949341,
"epoch": 0.25217706821480407,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997055405161712,
"loss": 5.7996,
"mean_token_accuracy": 0.1525806337594986,
"num_tokens": 5757367.0,
"step": 2780
},
{
"entropy": 5.97401123046875,
"epoch": 0.2526306240928882,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997038833571671,
"loss": 5.8662,
"mean_token_accuracy": 0.1486934617161751,
"num_tokens": 5769557.0,
"step": 2785
},
{
"entropy": 5.932326173782348,
"epoch": 0.25308417997097243,
"grad_norm": 0.875,
"learning_rate": 0.0004997022215512295,
"loss": 5.7594,
"mean_token_accuracy": 0.1602458581328392,
"num_tokens": 5780615.0,
"step": 2790
},
{
"entropy": 5.959806776046753,
"epoch": 0.2535377358490566,
"grad_norm": 0.86328125,
"learning_rate": 0.0004997005550983925,
"loss": 5.737,
"mean_token_accuracy": 0.15268604904413224,
"num_tokens": 5791271.0,
"step": 2795
},
{
"entropy": 5.863793230056762,
"epoch": 0.2539912917271408,
"grad_norm": 1.046875,
"learning_rate": 0.000499698883998691,
"loss": 5.6886,
"mean_token_accuracy": 0.15530947148799895,
"num_tokens": 5801117.0,
"step": 2800
},
{
"entropy": 5.88543930053711,
"epoch": 0.25444484760522496,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996972082521591,
"loss": 5.7122,
"mean_token_accuracy": 0.1563066676259041,
"num_tokens": 5811696.0,
"step": 2805
},
{
"entropy": 5.911672639846802,
"epoch": 0.25489840348330917,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996955278588319,
"loss": 5.738,
"mean_token_accuracy": 0.16307492852210997,
"num_tokens": 5822207.0,
"step": 2810
},
{
"entropy": 5.838271713256836,
"epoch": 0.2553519593613933,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996938428187437,
"loss": 5.7025,
"mean_token_accuracy": 0.15967562049627304,
"num_tokens": 5831927.0,
"step": 2815
},
{
"entropy": 5.887790107727051,
"epoch": 0.2558055152394775,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996921531319297,
"loss": 5.7578,
"mean_token_accuracy": 0.15641293302178383,
"num_tokens": 5842093.0,
"step": 2820
},
{
"entropy": 5.887936401367187,
"epoch": 0.2562590711175617,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996904587984246,
"loss": 5.704,
"mean_token_accuracy": 0.16297227442264556,
"num_tokens": 5850551.0,
"step": 2825
},
{
"entropy": 5.81576018333435,
"epoch": 0.25671262699564584,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996887598182636,
"loss": 5.6901,
"mean_token_accuracy": 0.15636582523584366,
"num_tokens": 5860651.0,
"step": 2830
},
{
"entropy": 5.827975130081176,
"epoch": 0.25716618287373005,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996870561914819,
"loss": 5.6698,
"mean_token_accuracy": 0.15702708512544633,
"num_tokens": 5871500.0,
"step": 2835
},
{
"entropy": 5.840231275558471,
"epoch": 0.2576197387518142,
"grad_norm": 0.921875,
"learning_rate": 0.0004996853479181145,
"loss": 5.6725,
"mean_token_accuracy": 0.15341195613145828,
"num_tokens": 5881523.0,
"step": 2840
},
{
"entropy": 5.9528480052948,
"epoch": 0.2580732946298984,
"grad_norm": 1.0,
"learning_rate": 0.0004996836349981969,
"loss": 5.8174,
"mean_token_accuracy": 0.15260155200958253,
"num_tokens": 5891162.0,
"step": 2845
},
{
"entropy": 5.902073764801026,
"epoch": 0.2585268505079826,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996819174317645,
"loss": 5.7173,
"mean_token_accuracy": 0.1552115336060524,
"num_tokens": 5901818.0,
"step": 2850
},
{
"entropy": 5.874346494674683,
"epoch": 0.2589804063860668,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996801952188528,
"loss": 5.79,
"mean_token_accuracy": 0.15259438008069992,
"num_tokens": 5913889.0,
"step": 2855
},
{
"entropy": 5.9258890628814695,
"epoch": 0.25943396226415094,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996784683594973,
"loss": 5.7571,
"mean_token_accuracy": 0.156342051923275,
"num_tokens": 5923429.0,
"step": 2860
},
{
"entropy": 5.998136568069458,
"epoch": 0.2598875181422351,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996767368537338,
"loss": 5.7779,
"mean_token_accuracy": 0.15545209050178527,
"num_tokens": 5933847.0,
"step": 2865
},
{
"entropy": 5.854821443557739,
"epoch": 0.2603410740203193,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996750007015983,
"loss": 5.6677,
"mean_token_accuracy": 0.15953482687473297,
"num_tokens": 5943874.0,
"step": 2870
},
{
"entropy": 5.930334520339966,
"epoch": 0.26079462989840346,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996732599031264,
"loss": 5.7357,
"mean_token_accuracy": 0.1538098156452179,
"num_tokens": 5953653.0,
"step": 2875
},
{
"entropy": 5.918991279602051,
"epoch": 0.2612481857764877,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996715144583543,
"loss": 5.6866,
"mean_token_accuracy": 0.15937823802232742,
"num_tokens": 5964039.0,
"step": 2880
},
{
"entropy": 5.940640115737915,
"epoch": 0.26170174165457183,
"grad_norm": 0.98828125,
"learning_rate": 0.000499669764367318,
"loss": 5.8108,
"mean_token_accuracy": 0.15168801695108414,
"num_tokens": 5974083.0,
"step": 2885
},
{
"entropy": 5.8325118064880375,
"epoch": 0.26215529753265604,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996680096300537,
"loss": 5.7435,
"mean_token_accuracy": 0.15367064327001573,
"num_tokens": 5984629.0,
"step": 2890
},
{
"entropy": 5.8481368064880375,
"epoch": 0.2626088534107402,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996662502465978,
"loss": 5.6702,
"mean_token_accuracy": 0.1644247755408287,
"num_tokens": 5994295.0,
"step": 2895
},
{
"entropy": 5.873128223419189,
"epoch": 0.2630624092888244,
"grad_norm": 0.92578125,
"learning_rate": 0.0004996644862169864,
"loss": 5.7046,
"mean_token_accuracy": 0.15299894511699677,
"num_tokens": 6004156.0,
"step": 2900
},
{
"entropy": 5.876518249511719,
"epoch": 0.26351596516690856,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996627175412563,
"loss": 5.7805,
"mean_token_accuracy": 0.15570748448371888,
"num_tokens": 6014188.0,
"step": 2905
},
{
"entropy": 5.96129150390625,
"epoch": 0.2639695210449927,
"grad_norm": 0.84765625,
"learning_rate": 0.000499660944219444,
"loss": 5.8435,
"mean_token_accuracy": 0.1530625268816948,
"num_tokens": 6024879.0,
"step": 2910
},
{
"entropy": 5.983526945114136,
"epoch": 0.2644230769230769,
"grad_norm": 0.8203125,
"learning_rate": 0.0004996591662515861,
"loss": 5.8974,
"mean_token_accuracy": 0.15041209682822226,
"num_tokens": 6036374.0,
"step": 2915
},
{
"entropy": 5.880353546142578,
"epoch": 0.2648766328011611,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996573836377193,
"loss": 5.7655,
"mean_token_accuracy": 0.1547638565301895,
"num_tokens": 6046432.0,
"step": 2920
},
{
"entropy": 5.962761545181275,
"epoch": 0.2653301886792453,
"grad_norm": 1.015625,
"learning_rate": 0.0004996555963778806,
"loss": 5.763,
"mean_token_accuracy": 0.15179126113653182,
"num_tokens": 6056864.0,
"step": 2925
},
{
"entropy": 6.006943416595459,
"epoch": 0.26578374455732945,
"grad_norm": 0.921875,
"learning_rate": 0.000499653804472107,
"loss": 5.7827,
"mean_token_accuracy": 0.15777789205312728,
"num_tokens": 6067294.0,
"step": 2930
},
{
"entropy": 5.932409477233887,
"epoch": 0.26623730043541366,
"grad_norm": 0.921875,
"learning_rate": 0.0004996520079204353,
"loss": 5.7948,
"mean_token_accuracy": 0.15130407512187957,
"num_tokens": 6077587.0,
"step": 2935
},
{
"entropy": 5.947808265686035,
"epoch": 0.2666908563134978,
"grad_norm": 0.84375,
"learning_rate": 0.000499650206722903,
"loss": 5.7315,
"mean_token_accuracy": 0.1543162226676941,
"num_tokens": 6089306.0,
"step": 2940
},
{
"entropy": 5.866966199874878,
"epoch": 0.267144412191582,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996484008795471,
"loss": 5.6891,
"mean_token_accuracy": 0.15265745371580125,
"num_tokens": 6099651.0,
"step": 2945
},
{
"entropy": 5.864347743988037,
"epoch": 0.2675979680696662,
"grad_norm": 0.95703125,
"learning_rate": 0.000499646590390405,
"loss": 5.688,
"mean_token_accuracy": 0.1571669578552246,
"num_tokens": 6110063.0,
"step": 2950
},
{
"entropy": 5.873627424240112,
"epoch": 0.26805152394775034,
"grad_norm": 0.96875,
"learning_rate": 0.0004996447752555142,
"loss": 5.7144,
"mean_token_accuracy": 0.15664682239294053,
"num_tokens": 6120031.0,
"step": 2955
},
{
"entropy": 5.848915719985962,
"epoch": 0.26850507982583455,
"grad_norm": 0.85546875,
"learning_rate": 0.0004996429554749122,
"loss": 5.7616,
"mean_token_accuracy": 0.15996424555778505,
"num_tokens": 6131274.0,
"step": 2960
},
{
"entropy": 5.9580456733703615,
"epoch": 0.2689586357039187,
"grad_norm": 0.921875,
"learning_rate": 0.0004996411310486367,
"loss": 5.8021,
"mean_token_accuracy": 0.1580547109246254,
"num_tokens": 6140878.0,
"step": 2965
},
{
"entropy": 5.900578355789184,
"epoch": 0.2694121915820029,
"grad_norm": 0.84375,
"learning_rate": 0.0004996393019767253,
"loss": 5.7986,
"mean_token_accuracy": 0.15211150646209717,
"num_tokens": 6152229.0,
"step": 2970
},
{
"entropy": 5.871568775177002,
"epoch": 0.26986574746008707,
"grad_norm": 0.82421875,
"learning_rate": 0.0004996374682592158,
"loss": 5.6994,
"mean_token_accuracy": 0.15565146058797835,
"num_tokens": 6162311.0,
"step": 2975
},
{
"entropy": 5.939528369903565,
"epoch": 0.2703193033381713,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996356298961463,
"loss": 5.8077,
"mean_token_accuracy": 0.15412216633558273,
"num_tokens": 6172806.0,
"step": 2980
},
{
"entropy": 5.916967105865479,
"epoch": 0.27077285921625543,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996337868875549,
"loss": 5.774,
"mean_token_accuracy": 0.15645444989204407,
"num_tokens": 6183479.0,
"step": 2985
},
{
"entropy": 5.928597736358642,
"epoch": 0.27122641509433965,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996319392334792,
"loss": 5.7363,
"mean_token_accuracy": 0.1543424740433693,
"num_tokens": 6194201.0,
"step": 2990
},
{
"entropy": 5.853014802932739,
"epoch": 0.2716799709724238,
"grad_norm": 0.95703125,
"learning_rate": 0.000499630086933958,
"loss": 5.6793,
"mean_token_accuracy": 0.15483027845621108,
"num_tokens": 6204774.0,
"step": 2995
},
{
"entropy": 5.876689767837524,
"epoch": 0.27213352685050796,
"grad_norm": 0.890625,
"learning_rate": 0.0004996282299890292,
"loss": 5.6781,
"mean_token_accuracy": 0.15287237316370011,
"num_tokens": 6215056.0,
"step": 3000
},
{
"epoch": 0.27213352685050796,
"eval_entropy": 5.712364828856282,
"eval_loss": 5.759267330169678,
"eval_mean_token_accuracy": 0.1617233954799863,
"eval_num_tokens": 6215056.0,
"eval_runtime": 21.065,
"eval_samples_per_second": 1678.618,
"eval_steps_per_second": 209.827,
"step": 3000
},
{
"entropy": 5.969257974624634,
"epoch": 0.27258708272859217,
"grad_norm": 0.84375,
"learning_rate": 0.0004996263683987316,
"loss": 5.7935,
"mean_token_accuracy": 0.15573310405015944,
"num_tokens": 6226264.0,
"step": 3005
},
{
"entropy": 5.891745710372925,
"epoch": 0.2730406386066763,
"grad_norm": 0.96875,
"learning_rate": 0.0004996245021631033,
"loss": 5.7187,
"mean_token_accuracy": 0.16169505268335344,
"num_tokens": 6235137.0,
"step": 3010
},
{
"entropy": 5.822071361541748,
"epoch": 0.27349419448476053,
"grad_norm": 0.95703125,
"learning_rate": 0.000499622631282183,
"loss": 5.7275,
"mean_token_accuracy": 0.1584876984357834,
"num_tokens": 6245411.0,
"step": 3015
},
{
"entropy": 5.966612482070923,
"epoch": 0.2739477503628447,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996207557560095,
"loss": 5.7953,
"mean_token_accuracy": 0.152059169113636,
"num_tokens": 6255633.0,
"step": 3020
},
{
"entropy": 5.933960390090943,
"epoch": 0.2744013062409289,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996188755846216,
"loss": 5.7423,
"mean_token_accuracy": 0.15434571653604506,
"num_tokens": 6265109.0,
"step": 3025
},
{
"entropy": 5.862824583053589,
"epoch": 0.27485486211901305,
"grad_norm": 0.9609375,
"learning_rate": 0.000499616990768058,
"loss": 5.7933,
"mean_token_accuracy": 0.14787728786468507,
"num_tokens": 6275241.0,
"step": 3030
},
{
"entropy": 5.949290132522583,
"epoch": 0.27530841799709727,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996151013063579,
"loss": 5.8252,
"mean_token_accuracy": 0.14647485911846161,
"num_tokens": 6286603.0,
"step": 3035
},
{
"entropy": 5.912906169891357,
"epoch": 0.2757619738751814,
"grad_norm": 0.90625,
"learning_rate": 0.0004996132071995601,
"loss": 5.6852,
"mean_token_accuracy": 0.16648433357477188,
"num_tokens": 6297219.0,
"step": 3040
},
{
"entropy": 5.883165311813355,
"epoch": 0.2762155297532656,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996113084477042,
"loss": 5.719,
"mean_token_accuracy": 0.16386881619691848,
"num_tokens": 6307694.0,
"step": 3045
},
{
"entropy": 5.807434129714966,
"epoch": 0.2766690856313498,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996094050508291,
"loss": 5.6527,
"mean_token_accuracy": 0.16033376455307008,
"num_tokens": 6317384.0,
"step": 3050
},
{
"entropy": 5.947029972076416,
"epoch": 0.27712264150943394,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996074970089741,
"loss": 5.7779,
"mean_token_accuracy": 0.15559548288583755,
"num_tokens": 6327558.0,
"step": 3055
},
{
"entropy": 5.87425684928894,
"epoch": 0.27757619738751815,
"grad_norm": 1.03125,
"learning_rate": 0.0004996055843221789,
"loss": 5.6409,
"mean_token_accuracy": 0.1627122566103935,
"num_tokens": 6337828.0,
"step": 3060
},
{
"entropy": 5.822861051559448,
"epoch": 0.2780297532656023,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996036669904831,
"loss": 5.7058,
"mean_token_accuracy": 0.16010485142469405,
"num_tokens": 6348678.0,
"step": 3065
},
{
"entropy": 5.901842355728149,
"epoch": 0.2784833091436865,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996017450139261,
"loss": 5.7969,
"mean_token_accuracy": 0.1491397202014923,
"num_tokens": 6359670.0,
"step": 3070
},
{
"entropy": 5.95522518157959,
"epoch": 0.2789368650217707,
"grad_norm": 0.87890625,
"learning_rate": 0.0004995998183925478,
"loss": 5.7541,
"mean_token_accuracy": 0.15521771609783172,
"num_tokens": 6371034.0,
"step": 3075
},
{
"entropy": 5.83247013092041,
"epoch": 0.2793904208998549,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995978871263881,
"loss": 5.7014,
"mean_token_accuracy": 0.15728666633367538,
"num_tokens": 6380672.0,
"step": 3080
},
{
"entropy": 5.874285125732422,
"epoch": 0.27984397677793904,
"grad_norm": 1.046875,
"learning_rate": 0.0004995959512154868,
"loss": 5.8151,
"mean_token_accuracy": 0.14317371249198912,
"num_tokens": 6390799.0,
"step": 3085
},
{
"entropy": 5.8934972286224365,
"epoch": 0.2802975326560232,
"grad_norm": 0.87890625,
"learning_rate": 0.000499594010659884,
"loss": 5.6884,
"mean_token_accuracy": 0.16007015258073806,
"num_tokens": 6400965.0,
"step": 3090
},
{
"entropy": 5.899712657928466,
"epoch": 0.2807510885341074,
"grad_norm": 0.875,
"learning_rate": 0.0004995920654596199,
"loss": 5.7049,
"mean_token_accuracy": 0.1535528853535652,
"num_tokens": 6411567.0,
"step": 3095
},
{
"entropy": 5.8801034450531,
"epoch": 0.28120464441219156,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995901156147345,
"loss": 5.7613,
"mean_token_accuracy": 0.1587476283311844,
"num_tokens": 6421193.0,
"step": 3100
},
{
"entropy": 5.9482683658599855,
"epoch": 0.2816582002902758,
"grad_norm": 1.03125,
"learning_rate": 0.0004995881611252684,
"loss": 5.7637,
"mean_token_accuracy": 0.164001926779747,
"num_tokens": 6430281.0,
"step": 3105
},
{
"entropy": 5.899417304992676,
"epoch": 0.28211175616835993,
"grad_norm": 0.8515625,
"learning_rate": 0.0004995862019912619,
"loss": 5.756,
"mean_token_accuracy": 0.15074933171272278,
"num_tokens": 6441055.0,
"step": 3110
},
{
"entropy": 5.860140228271485,
"epoch": 0.28256531204644414,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995842382127555,
"loss": 5.6706,
"mean_token_accuracy": 0.15401192158460617,
"num_tokens": 6451082.0,
"step": 3115
},
{
"entropy": 5.84559817314148,
"epoch": 0.2830188679245283,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995822697897897,
"loss": 5.7639,
"mean_token_accuracy": 0.1507149502635002,
"num_tokens": 6461617.0,
"step": 3120
},
{
"entropy": 5.967565059661865,
"epoch": 0.2834724238026125,
"grad_norm": 0.890625,
"learning_rate": 0.0004995802967224056,
"loss": 5.7685,
"mean_token_accuracy": 0.15601267218589782,
"num_tokens": 6472780.0,
"step": 3125
},
{
"entropy": 5.77694001197815,
"epoch": 0.28392597968069666,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995783190106434,
"loss": 5.6677,
"mean_token_accuracy": 0.15176742225885392,
"num_tokens": 6484078.0,
"step": 3130
},
{
"entropy": 5.900559663772583,
"epoch": 0.2843795355587808,
"grad_norm": 0.921875,
"learning_rate": 0.0004995763366545445,
"loss": 5.7619,
"mean_token_accuracy": 0.15375943332910538,
"num_tokens": 6494366.0,
"step": 3135
},
{
"entropy": 5.787074327468872,
"epoch": 0.284833091436865,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995743496541497,
"loss": 5.6902,
"mean_token_accuracy": 0.16150298565626145,
"num_tokens": 6505461.0,
"step": 3140
},
{
"entropy": 5.881974554061889,
"epoch": 0.2852866473149492,
"grad_norm": 0.875,
"learning_rate": 0.0004995723580095001,
"loss": 5.6923,
"mean_token_accuracy": 0.15881099700927734,
"num_tokens": 6516141.0,
"step": 3145
},
{
"entropy": 5.84017424583435,
"epoch": 0.2857402031930334,
"grad_norm": 0.94921875,
"learning_rate": 0.000499570361720637,
"loss": 5.7058,
"mean_token_accuracy": 0.1565474107861519,
"num_tokens": 6525536.0,
"step": 3150
},
{
"entropy": 5.8899266719818115,
"epoch": 0.28619375907111755,
"grad_norm": 0.875,
"learning_rate": 0.0004995683607876015,
"loss": 5.7579,
"mean_token_accuracy": 0.15209461599588395,
"num_tokens": 6535852.0,
"step": 3155
},
{
"entropy": 5.924435234069824,
"epoch": 0.28664731494920176,
"grad_norm": 0.921875,
"learning_rate": 0.0004995663552104352,
"loss": 5.7252,
"mean_token_accuracy": 0.16235753893852234,
"num_tokens": 6546353.0,
"step": 3160
},
{
"entropy": 5.779373979568481,
"epoch": 0.2871008708272859,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995643449891793,
"loss": 5.6751,
"mean_token_accuracy": 0.1516367569565773,
"num_tokens": 6557782.0,
"step": 3165
},
{
"entropy": 5.916032218933106,
"epoch": 0.2875544267053701,
"grad_norm": 0.96875,
"learning_rate": 0.0004995623301238757,
"loss": 5.6995,
"mean_token_accuracy": 0.15973981767892836,
"num_tokens": 6568100.0,
"step": 3170
},
{
"entropy": 5.865595197677612,
"epoch": 0.2880079825834543,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995603106145657,
"loss": 5.7505,
"mean_token_accuracy": 0.15267078280448915,
"num_tokens": 6578698.0,
"step": 3175
},
{
"entropy": 5.876104164123535,
"epoch": 0.28846153846153844,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995582864612913,
"loss": 5.7308,
"mean_token_accuracy": 0.1522951140999794,
"num_tokens": 6588908.0,
"step": 3180
},
{
"entropy": 5.780710887908936,
"epoch": 0.28891509433962265,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995562576640943,
"loss": 5.589,
"mean_token_accuracy": 0.15807687640190124,
"num_tokens": 6599445.0,
"step": 3185
},
{
"entropy": 5.823026847839356,
"epoch": 0.2893686502177068,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995542242230167,
"loss": 5.6838,
"mean_token_accuracy": 0.16184857338666916,
"num_tokens": 6610246.0,
"step": 3190
},
{
"entropy": 5.840323448181152,
"epoch": 0.289822206095791,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995521861381006,
"loss": 5.6896,
"mean_token_accuracy": 0.16104330569505693,
"num_tokens": 6620501.0,
"step": 3195
},
{
"entropy": 5.950109958648682,
"epoch": 0.29027576197387517,
"grad_norm": 0.84765625,
"learning_rate": 0.000499550143409388,
"loss": 5.7599,
"mean_token_accuracy": 0.1512766182422638,
"num_tokens": 6631206.0,
"step": 3200
},
{
"entropy": 5.829269504547119,
"epoch": 0.2907293178519594,
"grad_norm": 0.88671875,
"learning_rate": 0.0004995480960369211,
"loss": 5.7188,
"mean_token_accuracy": 0.1548565149307251,
"num_tokens": 6641600.0,
"step": 3205
},
{
"entropy": 5.827369594573975,
"epoch": 0.29118287373004353,
"grad_norm": 0.9375,
"learning_rate": 0.0004995460440207425,
"loss": 5.7759,
"mean_token_accuracy": 0.15423371642827988,
"num_tokens": 6652319.0,
"step": 3210
},
{
"entropy": 5.945125961303711,
"epoch": 0.29163642960812775,
"grad_norm": 0.890625,
"learning_rate": 0.0004995439873608945,
"loss": 5.712,
"mean_token_accuracy": 0.14905817657709122,
"num_tokens": 6663702.0,
"step": 3215
},
{
"entropy": 5.812268924713135,
"epoch": 0.2920899854862119,
"grad_norm": 0.90234375,
"learning_rate": 0.0004995419260574196,
"loss": 5.7027,
"mean_token_accuracy": 0.1491316169500351,
"num_tokens": 6673908.0,
"step": 3220
},
{
"entropy": 5.916238164901733,
"epoch": 0.29254354136429606,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995398601103603,
"loss": 5.727,
"mean_token_accuracy": 0.15841997563838958,
"num_tokens": 6683237.0,
"step": 3225
},
{
"entropy": 5.824212837219238,
"epoch": 0.29299709724238027,
"grad_norm": 0.984375,
"learning_rate": 0.0004995377895197597,
"loss": 5.6527,
"mean_token_accuracy": 0.15979209095239638,
"num_tokens": 6693451.0,
"step": 3230
},
{
"entropy": 5.761458158493042,
"epoch": 0.2934506531204644,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995357142856602,
"loss": 5.6124,
"mean_token_accuracy": 0.15973278135061264,
"num_tokens": 6703479.0,
"step": 3235
},
{
"entropy": 5.817833805084229,
"epoch": 0.29390420899854863,
"grad_norm": 0.83984375,
"learning_rate": 0.000499533634408105,
"loss": 5.619,
"mean_token_accuracy": 0.15864339768886565,
"num_tokens": 6715321.0,
"step": 3240
},
{
"entropy": 5.808075428009033,
"epoch": 0.2943577648766328,
"grad_norm": 0.88671875,
"learning_rate": 0.000499531549887137,
"loss": 5.6835,
"mean_token_accuracy": 0.15339466109871863,
"num_tokens": 6725541.0,
"step": 3245
},
{
"entropy": 5.856081151962281,
"epoch": 0.294811320754717,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995294607227993,
"loss": 5.7241,
"mean_token_accuracy": 0.16161034405231475,
"num_tokens": 6735165.0,
"step": 3250
},
{
"entropy": 5.899437427520752,
"epoch": 0.29526487663280115,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995273669151351,
"loss": 5.7105,
"mean_token_accuracy": 0.16171984821558,
"num_tokens": 6744245.0,
"step": 3255
},
{
"entropy": 5.866788148880005,
"epoch": 0.29571843251088537,
"grad_norm": 1.0234375,
"learning_rate": 0.000499525268464188,
"loss": 5.6832,
"mean_token_accuracy": 0.16346385776996614,
"num_tokens": 6754894.0,
"step": 3260
},
{
"entropy": 5.863985061645508,
"epoch": 0.2961719883889695,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995231653700009,
"loss": 5.7333,
"mean_token_accuracy": 0.15298420935869217,
"num_tokens": 6764442.0,
"step": 3265
},
{
"entropy": 5.820025396347046,
"epoch": 0.2966255442670537,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995210576326176,
"loss": 5.6891,
"mean_token_accuracy": 0.15496449172496796,
"num_tokens": 6775054.0,
"step": 3270
},
{
"entropy": 5.862255716323853,
"epoch": 0.2970791001451379,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995189452520816,
"loss": 5.6861,
"mean_token_accuracy": 0.16984087377786636,
"num_tokens": 6784825.0,
"step": 3275
},
{
"entropy": 5.884560394287109,
"epoch": 0.29753265602322204,
"grad_norm": 0.921875,
"learning_rate": 0.0004995168282284366,
"loss": 5.7215,
"mean_token_accuracy": 0.160723876953125,
"num_tokens": 6795928.0,
"step": 3280
},
{
"entropy": 5.787168312072754,
"epoch": 0.29798621190130625,
"grad_norm": 0.87109375,
"learning_rate": 0.0004995147065617263,
"loss": 5.6523,
"mean_token_accuracy": 0.15839823037385942,
"num_tokens": 6805456.0,
"step": 3285
},
{
"entropy": 5.925667858123779,
"epoch": 0.2984397677793904,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995125802519946,
"loss": 5.7188,
"mean_token_accuracy": 0.15525027811527253,
"num_tokens": 6816215.0,
"step": 3290
},
{
"entropy": 5.892514085769653,
"epoch": 0.2988933236574746,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995104492992857,
"loss": 5.7558,
"mean_token_accuracy": 0.15865902602672577,
"num_tokens": 6826545.0,
"step": 3295
},
{
"entropy": 5.798425722122192,
"epoch": 0.2993468795355588,
"grad_norm": 1.046875,
"learning_rate": 0.0004995083137036434,
"loss": 5.6637,
"mean_token_accuracy": 0.1575020283460617,
"num_tokens": 6836184.0,
"step": 3300
},
{
"entropy": 5.924666786193848,
"epoch": 0.299800435413643,
"grad_norm": 0.890625,
"learning_rate": 0.0004995061734651121,
"loss": 5.7419,
"mean_token_accuracy": 0.15869775861501695,
"num_tokens": 6846646.0,
"step": 3305
},
{
"entropy": 5.840148878097534,
"epoch": 0.30025399129172714,
"grad_norm": 0.8671875,
"learning_rate": 0.0004995040285837358,
"loss": 5.6502,
"mean_token_accuracy": 0.16409194618463516,
"num_tokens": 6857597.0,
"step": 3310
},
{
"entropy": 5.853920841217041,
"epoch": 0.3007075471698113,
"grad_norm": 1.046875,
"learning_rate": 0.0004995018790595589,
"loss": 5.7448,
"mean_token_accuracy": 0.15398296862840652,
"num_tokens": 6867523.0,
"step": 3315
},
{
"entropy": 5.901988887786866,
"epoch": 0.3011611030478955,
"grad_norm": 0.96875,
"learning_rate": 0.0004994997248926261,
"loss": 5.8385,
"mean_token_accuracy": 0.15451575368642806,
"num_tokens": 6878190.0,
"step": 3320
},
{
"entropy": 5.849097967147827,
"epoch": 0.30161465892597966,
"grad_norm": 1.015625,
"learning_rate": 0.0004994975660829817,
"loss": 5.6408,
"mean_token_accuracy": 0.16445966064929962,
"num_tokens": 6888483.0,
"step": 3325
},
{
"entropy": 5.7745849132537845,
"epoch": 0.3020682148040639,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994954026306705,
"loss": 5.6223,
"mean_token_accuracy": 0.15969739556312562,
"num_tokens": 6898337.0,
"step": 3330
},
{
"entropy": 5.876712703704834,
"epoch": 0.30252177068214803,
"grad_norm": 0.90625,
"learning_rate": 0.0004994932345357371,
"loss": 5.6971,
"mean_token_accuracy": 0.15568175464868544,
"num_tokens": 6909131.0,
"step": 3335
},
{
"entropy": 5.822457551956177,
"epoch": 0.30297532656023224,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994910617982264,
"loss": 5.6823,
"mean_token_accuracy": 0.15465617924928665,
"num_tokens": 6919790.0,
"step": 3340
},
{
"entropy": 5.841094827651977,
"epoch": 0.3034288824383164,
"grad_norm": 0.92578125,
"learning_rate": 0.0004994888844181833,
"loss": 5.6997,
"mean_token_accuracy": 0.16144252866506575,
"num_tokens": 6930072.0,
"step": 3345
},
{
"entropy": 5.7801776885986325,
"epoch": 0.3038824383164006,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994867023956529,
"loss": 5.5965,
"mean_token_accuracy": 0.15485866814851762,
"num_tokens": 6940646.0,
"step": 3350
},
{
"entropy": 5.740756988525391,
"epoch": 0.30433599419448476,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994845157306803,
"loss": 5.5767,
"mean_token_accuracy": 0.16163998693227768,
"num_tokens": 6951645.0,
"step": 3355
},
{
"entropy": 5.876171684265136,
"epoch": 0.3047895500725689,
"grad_norm": 0.96875,
"learning_rate": 0.0004994823244233107,
"loss": 5.7416,
"mean_token_accuracy": 0.1575745850801468,
"num_tokens": 6963020.0,
"step": 3360
},
{
"entropy": 5.882272481918335,
"epoch": 0.3052431059506531,
"grad_norm": 1.015625,
"learning_rate": 0.0004994801284735895,
"loss": 5.6823,
"mean_token_accuracy": 0.1615156725049019,
"num_tokens": 6972895.0,
"step": 3365
},
{
"entropy": 5.877938604354858,
"epoch": 0.3056966618287373,
"grad_norm": 0.890625,
"learning_rate": 0.0004994779278815619,
"loss": 5.8195,
"mean_token_accuracy": 0.1476362816989422,
"num_tokens": 6985363.0,
"step": 3370
},
{
"entropy": 5.906639528274536,
"epoch": 0.3061502177068215,
"grad_norm": 0.953125,
"learning_rate": 0.0004994757226472736,
"loss": 5.7638,
"mean_token_accuracy": 0.15788324922323227,
"num_tokens": 6997040.0,
"step": 3375
},
{
"entropy": 5.863248682022094,
"epoch": 0.30660377358490565,
"grad_norm": 0.85546875,
"learning_rate": 0.0004994735127707702,
"loss": 5.6891,
"mean_token_accuracy": 0.15913514643907548,
"num_tokens": 7007591.0,
"step": 3380
},
{
"entropy": 5.887237691879273,
"epoch": 0.30705732946298986,
"grad_norm": 1.03125,
"learning_rate": 0.0004994712982520972,
"loss": 5.6886,
"mean_token_accuracy": 0.1630169630050659,
"num_tokens": 7017858.0,
"step": 3385
},
{
"entropy": 5.760348892211914,
"epoch": 0.307510885341074,
"grad_norm": 0.890625,
"learning_rate": 0.0004994690790913008,
"loss": 5.6526,
"mean_token_accuracy": 0.15853054076433182,
"num_tokens": 7027365.0,
"step": 3390
},
{
"entropy": 5.775932741165161,
"epoch": 0.3079644412191582,
"grad_norm": 0.90234375,
"learning_rate": 0.0004994668552884265,
"loss": 5.6393,
"mean_token_accuracy": 0.16287402510643006,
"num_tokens": 7038728.0,
"step": 3395
},
{
"entropy": 5.94070987701416,
"epoch": 0.3084179970972424,
"grad_norm": 1.078125,
"learning_rate": 0.0004994646268435204,
"loss": 5.5879,
"mean_token_accuracy": 0.16382095515727996,
"num_tokens": 7048736.0,
"step": 3400
},
{
"entropy": 5.77776608467102,
"epoch": 0.30887155297532654,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994623937566286,
"loss": 5.7134,
"mean_token_accuracy": 0.16350821107625962,
"num_tokens": 7059205.0,
"step": 3405
},
{
"entropy": 5.802759027481079,
"epoch": 0.30932510885341075,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994601560277974,
"loss": 5.7095,
"mean_token_accuracy": 0.15878489315509797,
"num_tokens": 7069513.0,
"step": 3410
},
{
"entropy": 5.834039211273193,
"epoch": 0.3097786647314949,
"grad_norm": 0.90234375,
"learning_rate": 0.000499457913657073,
"loss": 5.6357,
"mean_token_accuracy": 0.1570906937122345,
"num_tokens": 7079940.0,
"step": 3415
},
{
"entropy": 5.844655847549438,
"epoch": 0.3102322206095791,
"grad_norm": 0.90625,
"learning_rate": 0.0004994556666445016,
"loss": 5.6323,
"mean_token_accuracy": 0.1617613285779953,
"num_tokens": 7090792.0,
"step": 3420
},
{
"entropy": 5.776029062271118,
"epoch": 0.31068577648766327,
"grad_norm": 0.859375,
"learning_rate": 0.0004994534149901299,
"loss": 5.583,
"mean_token_accuracy": 0.16228142231702805,
"num_tokens": 7101964.0,
"step": 3425
},
{
"entropy": 5.8066362857818605,
"epoch": 0.3111393323657475,
"grad_norm": 0.921875,
"learning_rate": 0.0004994511586940044,
"loss": 5.6499,
"mean_token_accuracy": 0.15664990097284318,
"num_tokens": 7113332.0,
"step": 3430
},
{
"entropy": 5.942155599594116,
"epoch": 0.31159288824383163,
"grad_norm": 1.03125,
"learning_rate": 0.0004994488977561717,
"loss": 5.8081,
"mean_token_accuracy": 0.15569447129964828,
"num_tokens": 7123481.0,
"step": 3435
},
{
"entropy": 5.719643592834473,
"epoch": 0.31204644412191584,
"grad_norm": 0.9296875,
"learning_rate": 0.0004994466321766786,
"loss": 5.5667,
"mean_token_accuracy": 0.16554133594036102,
"num_tokens": 7133694.0,
"step": 3440
},
{
"entropy": 5.836681747436524,
"epoch": 0.3125,
"grad_norm": 0.9140625,
"learning_rate": 0.000499444361955572,
"loss": 5.6901,
"mean_token_accuracy": 0.16068354696035386,
"num_tokens": 7144491.0,
"step": 3445
},
{
"entropy": 5.970473051071167,
"epoch": 0.31295355587808416,
"grad_norm": 0.921875,
"learning_rate": 0.0004994420870928988,
"loss": 5.7727,
"mean_token_accuracy": 0.1537349671125412,
"num_tokens": 7154716.0,
"step": 3450
},
{
"entropy": 5.861110353469849,
"epoch": 0.31340711175616837,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994398075887059,
"loss": 5.7654,
"mean_token_accuracy": 0.15032783523201942,
"num_tokens": 7164258.0,
"step": 3455
},
{
"entropy": 5.91595458984375,
"epoch": 0.3138606676342525,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994375234430407,
"loss": 5.6193,
"mean_token_accuracy": 0.15683988332748414,
"num_tokens": 7175098.0,
"step": 3460
},
{
"entropy": 5.792357158660889,
"epoch": 0.31431422351233673,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994352346559504,
"loss": 5.7077,
"mean_token_accuracy": 0.15706134140491484,
"num_tokens": 7186480.0,
"step": 3465
},
{
"entropy": 5.765750789642334,
"epoch": 0.3147677793904209,
"grad_norm": 0.9140625,
"learning_rate": 0.0004994329412274821,
"loss": 5.6699,
"mean_token_accuracy": 0.16495281904935838,
"num_tokens": 7197056.0,
"step": 3470
},
{
"entropy": 5.9141796112060545,
"epoch": 0.3152213352685051,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994306431576835,
"loss": 5.8035,
"mean_token_accuracy": 0.1533171132206917,
"num_tokens": 7206040.0,
"step": 3475
},
{
"entropy": 5.938131999969483,
"epoch": 0.31567489114658925,
"grad_norm": 0.84765625,
"learning_rate": 0.000499428340446602,
"loss": 5.6422,
"mean_token_accuracy": 0.16303373575210572,
"num_tokens": 7216344.0,
"step": 3480
},
{
"entropy": 5.862201881408692,
"epoch": 0.31612844702467346,
"grad_norm": 0.91015625,
"learning_rate": 0.0004994260330942851,
"loss": 5.7168,
"mean_token_accuracy": 0.1532277211546898,
"num_tokens": 7226729.0,
"step": 3485
},
{
"entropy": 5.746544456481933,
"epoch": 0.3165820029027576,
"grad_norm": 0.9375,
"learning_rate": 0.0004994237211007808,
"loss": 5.6695,
"mean_token_accuracy": 0.15737406313419341,
"num_tokens": 7237616.0,
"step": 3490
},
{
"entropy": 5.85200867652893,
"epoch": 0.3170355587808418,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994214044661367,
"loss": 5.5787,
"mean_token_accuracy": 0.16485872715711594,
"num_tokens": 7247857.0,
"step": 3495
},
{
"entropy": 5.749833250045777,
"epoch": 0.317489114658926,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994190831904007,
"loss": 5.5717,
"mean_token_accuracy": 0.16816714257001877,
"num_tokens": 7257858.0,
"step": 3500
},
{
"entropy": 5.678181076049805,
"epoch": 0.31794267053701014,
"grad_norm": 0.9375,
"learning_rate": 0.000499416757273621,
"loss": 5.588,
"mean_token_accuracy": 0.1626364231109619,
"num_tokens": 7268440.0,
"step": 3505
},
{
"entropy": 5.893909215927124,
"epoch": 0.31839622641509435,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994144267158454,
"loss": 5.6453,
"mean_token_accuracy": 0.1645614832639694,
"num_tokens": 7278560.0,
"step": 3510
},
{
"entropy": 5.826103019714355,
"epoch": 0.3188497822931785,
"grad_norm": 1.0,
"learning_rate": 0.0004994120915171224,
"loss": 5.6564,
"mean_token_accuracy": 0.16230306029319763,
"num_tokens": 7288422.0,
"step": 3515
},
{
"entropy": 5.89535002708435,
"epoch": 0.3193033381712627,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994097516775002,
"loss": 5.7283,
"mean_token_accuracy": 0.15988563597202302,
"num_tokens": 7298613.0,
"step": 3520
},
{
"entropy": 5.8722318649292,
"epoch": 0.3197568940493469,
"grad_norm": 0.87109375,
"learning_rate": 0.0004994074071970271,
"loss": 5.7538,
"mean_token_accuracy": 0.15464695021510125,
"num_tokens": 7309509.0,
"step": 3525
},
{
"entropy": 5.882004261016846,
"epoch": 0.3202104499274311,
"grad_norm": 1.015625,
"learning_rate": 0.0004994050580757515,
"loss": 5.6563,
"mean_token_accuracy": 0.15909888744354247,
"num_tokens": 7319015.0,
"step": 3530
},
{
"entropy": 5.759677457809448,
"epoch": 0.32066400580551524,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994027043137223,
"loss": 5.6356,
"mean_token_accuracy": 0.1678672254085541,
"num_tokens": 7329026.0,
"step": 3535
},
{
"entropy": 5.873507308959961,
"epoch": 0.3211175616835994,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994003459109879,
"loss": 5.7247,
"mean_token_accuracy": 0.1562524437904358,
"num_tokens": 7339638.0,
"step": 3540
},
{
"entropy": 5.856282138824463,
"epoch": 0.3215711175616836,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993979828675972,
"loss": 5.6467,
"mean_token_accuracy": 0.1605513334274292,
"num_tokens": 7349979.0,
"step": 3545
},
{
"entropy": 5.731431865692139,
"epoch": 0.32202467343976776,
"grad_norm": 0.953125,
"learning_rate": 0.000499395615183599,
"loss": 5.5759,
"mean_token_accuracy": 0.1669188141822815,
"num_tokens": 7359398.0,
"step": 3550
},
{
"entropy": 5.755496263504028,
"epoch": 0.32247822931785197,
"grad_norm": 0.94921875,
"learning_rate": 0.0004993932428590422,
"loss": 5.6178,
"mean_token_accuracy": 0.16151758432388305,
"num_tokens": 7370418.0,
"step": 3555
},
{
"entropy": 5.8418971538543705,
"epoch": 0.3229317851959361,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993908658939762,
"loss": 5.643,
"mean_token_accuracy": 0.15817937552928923,
"num_tokens": 7380461.0,
"step": 3560
},
{
"entropy": 5.819117832183838,
"epoch": 0.32338534107402034,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993884842884496,
"loss": 5.656,
"mean_token_accuracy": 0.16232847571372985,
"num_tokens": 7390646.0,
"step": 3565
},
{
"entropy": 5.789481115341187,
"epoch": 0.3238388969521045,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993860980425121,
"loss": 5.7402,
"mean_token_accuracy": 0.16244688481092454,
"num_tokens": 7402106.0,
"step": 3570
},
{
"entropy": 5.829900217056275,
"epoch": 0.3242924528301887,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993837071562131,
"loss": 5.7286,
"mean_token_accuracy": 0.15734863579273223,
"num_tokens": 7412936.0,
"step": 3575
},
{
"entropy": 5.834792232513427,
"epoch": 0.32474600870827286,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993813116296017,
"loss": 5.5932,
"mean_token_accuracy": 0.1646380230784416,
"num_tokens": 7422303.0,
"step": 3580
},
{
"entropy": 5.787908029556275,
"epoch": 0.325199564586357,
"grad_norm": 0.9375,
"learning_rate": 0.0004993789114627275,
"loss": 5.6734,
"mean_token_accuracy": 0.1588895797729492,
"num_tokens": 7432829.0,
"step": 3585
},
{
"entropy": 5.904963636398316,
"epoch": 0.3256531204644412,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993765066556403,
"loss": 5.6859,
"mean_token_accuracy": 0.15745299607515334,
"num_tokens": 7443218.0,
"step": 3590
},
{
"entropy": 5.824350261688233,
"epoch": 0.3261066763425254,
"grad_norm": 0.8671875,
"learning_rate": 0.0004993740972083899,
"loss": 5.6363,
"mean_token_accuracy": 0.16018294245004655,
"num_tokens": 7454531.0,
"step": 3595
},
{
"entropy": 5.835407304763794,
"epoch": 0.3265602322206096,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993716831210259,
"loss": 5.7004,
"mean_token_accuracy": 0.16122788041830063,
"num_tokens": 7464993.0,
"step": 3600
},
{
"entropy": 5.760484457015991,
"epoch": 0.32701378809869375,
"grad_norm": 0.921875,
"learning_rate": 0.0004993692643935983,
"loss": 5.7176,
"mean_token_accuracy": 0.15775671675801278,
"num_tokens": 7475755.0,
"step": 3605
},
{
"entropy": 5.829902124404907,
"epoch": 0.32746734397677796,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993668410261571,
"loss": 5.7147,
"mean_token_accuracy": 0.16331727802753448,
"num_tokens": 7486778.0,
"step": 3610
},
{
"entropy": 5.919011974334717,
"epoch": 0.3279208998548621,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993644130187525,
"loss": 5.6816,
"mean_token_accuracy": 0.16154048442840577,
"num_tokens": 7497439.0,
"step": 3615
},
{
"entropy": 5.813072729110718,
"epoch": 0.3283744557329463,
"grad_norm": 1.0625,
"learning_rate": 0.0004993619803714347,
"loss": 5.7209,
"mean_token_accuracy": 0.15986181199550628,
"num_tokens": 7508274.0,
"step": 3620
},
{
"entropy": 5.799656772613526,
"epoch": 0.3288280116110305,
"grad_norm": 0.890625,
"learning_rate": 0.0004993595430842539,
"loss": 5.6925,
"mean_token_accuracy": 0.15679641515016557,
"num_tokens": 7519315.0,
"step": 3625
},
{
"entropy": 5.926063013076782,
"epoch": 0.32928156748911463,
"grad_norm": 0.921875,
"learning_rate": 0.0004993571011572606,
"loss": 5.6989,
"mean_token_accuracy": 0.16154777109622956,
"num_tokens": 7529910.0,
"step": 3630
},
{
"entropy": 5.83719687461853,
"epoch": 0.32973512336719885,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993546545905053,
"loss": 5.6263,
"mean_token_accuracy": 0.16304632276296616,
"num_tokens": 7538966.0,
"step": 3635
},
{
"entropy": 5.795618629455566,
"epoch": 0.330188679245283,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993522033840386,
"loss": 5.6618,
"mean_token_accuracy": 0.1593761846423149,
"num_tokens": 7549683.0,
"step": 3640
},
{
"entropy": 5.746255350112915,
"epoch": 0.3306422351233672,
"grad_norm": 0.8828125,
"learning_rate": 0.000499349747537911,
"loss": 5.6375,
"mean_token_accuracy": 0.1627708211541176,
"num_tokens": 7560629.0,
"step": 3645
},
{
"entropy": 5.824515104293823,
"epoch": 0.33109579100145137,
"grad_norm": 0.9375,
"learning_rate": 0.0004993472870521736,
"loss": 5.5937,
"mean_token_accuracy": 0.16678274869918824,
"num_tokens": 7570613.0,
"step": 3650
},
{
"entropy": 5.780102157592774,
"epoch": 0.3315493468795356,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993448219268771,
"loss": 5.5616,
"mean_token_accuracy": 0.16153461784124373,
"num_tokens": 7580640.0,
"step": 3655
},
{
"entropy": 5.7738007545471195,
"epoch": 0.33200290275761973,
"grad_norm": 1.0625,
"learning_rate": 0.0004993423521620726,
"loss": 5.6566,
"mean_token_accuracy": 0.16400932371616364,
"num_tokens": 7590164.0,
"step": 3660
},
{
"entropy": 5.878917837142945,
"epoch": 0.33245645863570394,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993398777578109,
"loss": 5.7057,
"mean_token_accuracy": 0.15703917518258095,
"num_tokens": 7600173.0,
"step": 3665
},
{
"entropy": 5.718817949295044,
"epoch": 0.3329100145137881,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993373987141435,
"loss": 5.5171,
"mean_token_accuracy": 0.16777506917715074,
"num_tokens": 7609958.0,
"step": 3670
},
{
"entropy": 5.847012901306153,
"epoch": 0.33336357039187225,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993349150311214,
"loss": 5.6535,
"mean_token_accuracy": 0.15843262523412704,
"num_tokens": 7620237.0,
"step": 3675
},
{
"entropy": 5.759278964996338,
"epoch": 0.33381712626995647,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993324267087962,
"loss": 5.5564,
"mean_token_accuracy": 0.16717470437288284,
"num_tokens": 7629886.0,
"step": 3680
},
{
"entropy": 5.757597494125366,
"epoch": 0.3342706821480406,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993299337472192,
"loss": 5.6311,
"mean_token_accuracy": 0.16425763219594955,
"num_tokens": 7640057.0,
"step": 3685
},
{
"entropy": 5.725239229202271,
"epoch": 0.33472423802612483,
"grad_norm": 0.89453125,
"learning_rate": 0.000499327436146442,
"loss": 5.5767,
"mean_token_accuracy": 0.16163882464170456,
"num_tokens": 7649766.0,
"step": 3690
},
{
"entropy": 5.778099012374878,
"epoch": 0.335177793904209,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993249339065162,
"loss": 5.6323,
"mean_token_accuracy": 0.16390450447797775,
"num_tokens": 7660890.0,
"step": 3695
},
{
"entropy": 5.83032078742981,
"epoch": 0.3356313497822932,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993224270274937,
"loss": 5.6562,
"mean_token_accuracy": 0.16339348554611205,
"num_tokens": 7670737.0,
"step": 3700
},
{
"entropy": 5.714012575149536,
"epoch": 0.33608490566037735,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993199155094263,
"loss": 5.6157,
"mean_token_accuracy": 0.16412198096513747,
"num_tokens": 7681407.0,
"step": 3705
},
{
"entropy": 5.795091533660889,
"epoch": 0.33653846153846156,
"grad_norm": 0.8984375,
"learning_rate": 0.0004993173993523658,
"loss": 5.5506,
"mean_token_accuracy": 0.1710594192147255,
"num_tokens": 7691758.0,
"step": 3710
},
{
"entropy": 5.689993810653687,
"epoch": 0.3369920174165457,
"grad_norm": 0.96875,
"learning_rate": 0.0004993148785563643,
"loss": 5.5826,
"mean_token_accuracy": 0.1621573567390442,
"num_tokens": 7702934.0,
"step": 3715
},
{
"entropy": 5.717300510406494,
"epoch": 0.3374455732946299,
"grad_norm": 0.87890625,
"learning_rate": 0.000499312353121474,
"loss": 5.5771,
"mean_token_accuracy": 0.1630354642868042,
"num_tokens": 7713001.0,
"step": 3720
},
{
"entropy": 5.836460018157959,
"epoch": 0.3378991291727141,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993098230477471,
"loss": 5.6475,
"mean_token_accuracy": 0.15897637307643891,
"num_tokens": 7724005.0,
"step": 3725
},
{
"entropy": 5.8066576480865475,
"epoch": 0.33835268505079824,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993072883352358,
"loss": 5.6038,
"mean_token_accuracy": 0.1632534846663475,
"num_tokens": 7733743.0,
"step": 3730
},
{
"entropy": 5.836714839935302,
"epoch": 0.33880624092888245,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993047489839927,
"loss": 5.7837,
"mean_token_accuracy": 0.15582592636346818,
"num_tokens": 7743762.0,
"step": 3735
},
{
"entropy": 5.886009645462036,
"epoch": 0.3392597968069666,
"grad_norm": 0.875,
"learning_rate": 0.0004993022049940702,
"loss": 5.666,
"mean_token_accuracy": 0.15896076560020447,
"num_tokens": 7753758.0,
"step": 3740
},
{
"entropy": 5.778275203704834,
"epoch": 0.3397133526850508,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992996563655209,
"loss": 5.6211,
"mean_token_accuracy": 0.16299140006303786,
"num_tokens": 7764692.0,
"step": 3745
},
{
"entropy": 5.749185180664062,
"epoch": 0.340166908563135,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992971030983975,
"loss": 5.6453,
"mean_token_accuracy": 0.16236765682697296,
"num_tokens": 7776892.0,
"step": 3750
},
{
"entropy": 5.860137844085694,
"epoch": 0.3406204644412192,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992945451927531,
"loss": 5.7017,
"mean_token_accuracy": 0.15955468267202377,
"num_tokens": 7787143.0,
"step": 3755
},
{
"entropy": 5.734303283691406,
"epoch": 0.34107402031930334,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992919826486402,
"loss": 5.6101,
"mean_token_accuracy": 0.16399556547403335,
"num_tokens": 7798537.0,
"step": 3760
},
{
"entropy": 5.734641885757446,
"epoch": 0.3415275761973875,
"grad_norm": 0.859375,
"learning_rate": 0.0004992894154661118,
"loss": 5.5799,
"mean_token_accuracy": 0.16684457361698152,
"num_tokens": 7808434.0,
"step": 3765
},
{
"entropy": 5.806569623947143,
"epoch": 0.3419811320754717,
"grad_norm": 0.89453125,
"learning_rate": 0.0004992868436452213,
"loss": 5.6955,
"mean_token_accuracy": 0.15859481543302537,
"num_tokens": 7819592.0,
"step": 3770
},
{
"entropy": 5.850445795059204,
"epoch": 0.34243468795355586,
"grad_norm": 1.0,
"learning_rate": 0.0004992842671860217,
"loss": 5.5788,
"mean_token_accuracy": 0.16591217517852783,
"num_tokens": 7830602.0,
"step": 3775
},
{
"entropy": 5.835497093200684,
"epoch": 0.34288824383164007,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992816860885662,
"loss": 5.735,
"mean_token_accuracy": 0.15822505801916123,
"num_tokens": 7840850.0,
"step": 3780
},
{
"entropy": 5.917944717407226,
"epoch": 0.3433417997097242,
"grad_norm": 0.9375,
"learning_rate": 0.0004992791003529084,
"loss": 5.6846,
"mean_token_accuracy": 0.15368564128875734,
"num_tokens": 7852371.0,
"step": 3785
},
{
"entropy": 5.72100830078125,
"epoch": 0.34379535558780844,
"grad_norm": 0.9375,
"learning_rate": 0.0004992765099791017,
"loss": 5.6964,
"mean_token_accuracy": 0.15722172409296037,
"num_tokens": 7863290.0,
"step": 3790
},
{
"entropy": 5.7610937595367435,
"epoch": 0.3442489114658926,
"grad_norm": 0.91796875,
"learning_rate": 0.0004992739149671995,
"loss": 5.5333,
"mean_token_accuracy": 0.1643339917063713,
"num_tokens": 7873927.0,
"step": 3795
},
{
"entropy": 5.7489160060882565,
"epoch": 0.3447024673439768,
"grad_norm": 0.91796875,
"learning_rate": 0.0004992713153172556,
"loss": 5.5691,
"mean_token_accuracy": 0.1632430613040924,
"num_tokens": 7884977.0,
"step": 3800
},
{
"entropy": 5.745196342468262,
"epoch": 0.34515602322206096,
"grad_norm": 0.90625,
"learning_rate": 0.0004992687110293238,
"loss": 5.6531,
"mean_token_accuracy": 0.16766562685370445,
"num_tokens": 7894620.0,
"step": 3805
},
{
"entropy": 5.78018045425415,
"epoch": 0.3456095791001451,
"grad_norm": 0.9375,
"learning_rate": 0.0004992661021034579,
"loss": 5.5848,
"mean_token_accuracy": 0.16409669667482377,
"num_tokens": 7904541.0,
"step": 3810
},
{
"entropy": 5.75669412612915,
"epoch": 0.3460631349782293,
"grad_norm": 0.8203125,
"learning_rate": 0.0004992634885397118,
"loss": 5.6036,
"mean_token_accuracy": 0.1641060397028923,
"num_tokens": 7915435.0,
"step": 3815
},
{
"entropy": 5.751991319656372,
"epoch": 0.3465166908563135,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992608703381396,
"loss": 5.5681,
"mean_token_accuracy": 0.1672964349389076,
"num_tokens": 7925870.0,
"step": 3820
},
{
"entropy": 5.787299251556396,
"epoch": 0.3469702467343977,
"grad_norm": 0.921875,
"learning_rate": 0.0004992582474987955,
"loss": 5.5792,
"mean_token_accuracy": 0.16019600182771682,
"num_tokens": 7937046.0,
"step": 3825
},
{
"entropy": 5.747065591812134,
"epoch": 0.34742380261248185,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992556200217337,
"loss": 5.5366,
"mean_token_accuracy": 0.1621662676334381,
"num_tokens": 7947150.0,
"step": 3830
},
{
"entropy": 5.782345485687256,
"epoch": 0.34787735849056606,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992529879070085,
"loss": 5.6463,
"mean_token_accuracy": 0.16273225396871566,
"num_tokens": 7956815.0,
"step": 3835
},
{
"entropy": 5.83603720664978,
"epoch": 0.3483309143686502,
"grad_norm": 0.85546875,
"learning_rate": 0.0004992503511546745,
"loss": 5.6669,
"mean_token_accuracy": 0.16219530552625655,
"num_tokens": 7968114.0,
"step": 3840
},
{
"entropy": 5.786204385757446,
"epoch": 0.3487844702467344,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992477097647859,
"loss": 5.6001,
"mean_token_accuracy": 0.15802547633647918,
"num_tokens": 7977840.0,
"step": 3845
},
{
"entropy": 5.839146471023559,
"epoch": 0.3492380261248186,
"grad_norm": 0.92578125,
"learning_rate": 0.0004992450637373976,
"loss": 5.7159,
"mean_token_accuracy": 0.16593586057424545,
"num_tokens": 7988663.0,
"step": 3850
},
{
"entropy": 5.7088793277740475,
"epoch": 0.34969158200290273,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992424130725642,
"loss": 5.5105,
"mean_token_accuracy": 0.1633152462542057,
"num_tokens": 7999693.0,
"step": 3855
},
{
"entropy": 5.813177061080933,
"epoch": 0.35014513788098695,
"grad_norm": 0.859375,
"learning_rate": 0.0004992397577703406,
"loss": 5.7066,
"mean_token_accuracy": 0.15328351110219957,
"num_tokens": 8010154.0,
"step": 3860
},
{
"entropy": 5.895718193054199,
"epoch": 0.3505986937590711,
"grad_norm": 1.0,
"learning_rate": 0.0004992370978307816,
"loss": 5.6764,
"mean_token_accuracy": 0.15621304959058763,
"num_tokens": 8020921.0,
"step": 3865
},
{
"entropy": 5.780423545837403,
"epoch": 0.3510522496371553,
"grad_norm": 0.89453125,
"learning_rate": 0.0004992344332539424,
"loss": 5.6006,
"mean_token_accuracy": 0.16626923233270646,
"num_tokens": 8031404.0,
"step": 3870
},
{
"entropy": 5.6891073226928714,
"epoch": 0.35150580551523947,
"grad_norm": 0.90625,
"learning_rate": 0.0004992317640398779,
"loss": 5.6086,
"mean_token_accuracy": 0.16590815931558608,
"num_tokens": 8041310.0,
"step": 3875
},
{
"entropy": 5.7839179039001465,
"epoch": 0.3519593613933237,
"grad_norm": 0.9375,
"learning_rate": 0.0004992290901886434,
"loss": 5.5748,
"mean_token_accuracy": 0.16641812920570373,
"num_tokens": 8051054.0,
"step": 3880
},
{
"entropy": 5.776237964630127,
"epoch": 0.35241291727140783,
"grad_norm": 1.0,
"learning_rate": 0.000499226411700294,
"loss": 5.577,
"mean_token_accuracy": 0.16576359570026397,
"num_tokens": 8061063.0,
"step": 3885
},
{
"entropy": 5.721409654617309,
"epoch": 0.35286647314949204,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992237285748855,
"loss": 5.6016,
"mean_token_accuracy": 0.16601926535367967,
"num_tokens": 8071715.0,
"step": 3890
},
{
"entropy": 5.865892457962036,
"epoch": 0.3533200290275762,
"grad_norm": 0.90625,
"learning_rate": 0.0004992210408124731,
"loss": 5.6575,
"mean_token_accuracy": 0.1572970688343048,
"num_tokens": 8082679.0,
"step": 3895
},
{
"entropy": 5.7754655361175535,
"epoch": 0.35377358490566035,
"grad_norm": 0.84765625,
"learning_rate": 0.0004992183484131123,
"loss": 5.6963,
"mean_token_accuracy": 0.15824397653341293,
"num_tokens": 8093785.0,
"step": 3900
},
{
"entropy": 5.733349323272705,
"epoch": 0.35422714078374457,
"grad_norm": 0.90625,
"learning_rate": 0.0004992156513768591,
"loss": 5.5862,
"mean_token_accuracy": 0.16918893307447433,
"num_tokens": 8105822.0,
"step": 3905
},
{
"entropy": 5.746071147918701,
"epoch": 0.3546806966618287,
"grad_norm": 0.93359375,
"learning_rate": 0.000499212949703769,
"loss": 5.5009,
"mean_token_accuracy": 0.16611845195293426,
"num_tokens": 8115521.0,
"step": 3910
},
{
"entropy": 5.764930868148804,
"epoch": 0.35513425253991293,
"grad_norm": 0.89453125,
"learning_rate": 0.0004992102433938981,
"loss": 5.5949,
"mean_token_accuracy": 0.1668561689555645,
"num_tokens": 8125524.0,
"step": 3915
},
{
"entropy": 5.734392976760864,
"epoch": 0.3555878084179971,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992075324473021,
"loss": 5.5131,
"mean_token_accuracy": 0.17593824863433838,
"num_tokens": 8135552.0,
"step": 3920
},
{
"entropy": 5.831735563278198,
"epoch": 0.3560413642960813,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992048168640371,
"loss": 5.6308,
"mean_token_accuracy": 0.1653963729739189,
"num_tokens": 8145617.0,
"step": 3925
},
{
"entropy": 5.727599000930786,
"epoch": 0.35649492017416545,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992020966441595,
"loss": 5.6406,
"mean_token_accuracy": 0.1605748012661934,
"num_tokens": 8156729.0,
"step": 3930
},
{
"entropy": 5.772192573547363,
"epoch": 0.35694847605224966,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991993717877254,
"loss": 5.6209,
"mean_token_accuracy": 0.1616794392466545,
"num_tokens": 8167637.0,
"step": 3935
},
{
"entropy": 5.763149976730347,
"epoch": 0.3574020319303338,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991966422947911,
"loss": 5.6039,
"mean_token_accuracy": 0.1692925289273262,
"num_tokens": 8177354.0,
"step": 3940
},
{
"entropy": 5.721889448165894,
"epoch": 0.357855587808418,
"grad_norm": 0.8828125,
"learning_rate": 0.0004991939081654132,
"loss": 5.5608,
"mean_token_accuracy": 0.17194263488054276,
"num_tokens": 8187580.0,
"step": 3945
},
{
"entropy": 5.723434352874756,
"epoch": 0.3583091436865022,
"grad_norm": 0.85546875,
"learning_rate": 0.000499191169399648,
"loss": 5.6598,
"mean_token_accuracy": 0.16233123391866683,
"num_tokens": 8199382.0,
"step": 3950
},
{
"entropy": 5.781085586547851,
"epoch": 0.35876269956458634,
"grad_norm": 0.91796875,
"learning_rate": 0.0004991884259975525,
"loss": 5.5557,
"mean_token_accuracy": 0.16617808789014815,
"num_tokens": 8208484.0,
"step": 3955
},
{
"entropy": 5.797117471694946,
"epoch": 0.35921625544267055,
"grad_norm": 0.91015625,
"learning_rate": 0.000499185677959183,
"loss": 5.6052,
"mean_token_accuracy": 0.15923126488924028,
"num_tokens": 8219056.0,
"step": 3960
},
{
"entropy": 5.777331686019897,
"epoch": 0.3596698113207547,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991829252845966,
"loss": 5.6232,
"mean_token_accuracy": 0.1578235447406769,
"num_tokens": 8230149.0,
"step": 3965
},
{
"entropy": 5.772022676467896,
"epoch": 0.3601233671988389,
"grad_norm": 0.90625,
"learning_rate": 0.0004991801679738502,
"loss": 5.6806,
"mean_token_accuracy": 0.15451672971248626,
"num_tokens": 8240803.0,
"step": 3970
},
{
"entropy": 5.766995000839233,
"epoch": 0.3605769230769231,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991774060270008,
"loss": 5.632,
"mean_token_accuracy": 0.1565280094742775,
"num_tokens": 8251563.0,
"step": 3975
},
{
"entropy": 5.758525943756103,
"epoch": 0.3610304789550073,
"grad_norm": 0.875,
"learning_rate": 0.0004991746394441055,
"loss": 5.5818,
"mean_token_accuracy": 0.15869568139314652,
"num_tokens": 8261342.0,
"step": 3980
},
{
"entropy": 5.76960334777832,
"epoch": 0.36148403483309144,
"grad_norm": 0.890625,
"learning_rate": 0.0004991718682252216,
"loss": 5.6749,
"mean_token_accuracy": 0.15879870802164078,
"num_tokens": 8271540.0,
"step": 3985
},
{
"entropy": 5.807124662399292,
"epoch": 0.3619375907111756,
"grad_norm": 0.9453125,
"learning_rate": 0.0004991690923704063,
"loss": 5.6018,
"mean_token_accuracy": 0.16289105713367463,
"num_tokens": 8280593.0,
"step": 3990
},
{
"entropy": 5.7926068782806395,
"epoch": 0.3623911465892598,
"grad_norm": 0.96875,
"learning_rate": 0.0004991663118797171,
"loss": 5.6189,
"mean_token_accuracy": 0.16317871809005738,
"num_tokens": 8290126.0,
"step": 3995
},
{
"entropy": 5.760343790054321,
"epoch": 0.36284470246734396,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991635267532113,
"loss": 5.542,
"mean_token_accuracy": 0.16164259910583495,
"num_tokens": 8301416.0,
"step": 4000
},
{
"entropy": 5.794639825820923,
"epoch": 0.36329825834542817,
"grad_norm": 0.96875,
"learning_rate": 0.0004991607369909468,
"loss": 5.6311,
"mean_token_accuracy": 0.15973753184080125,
"num_tokens": 8311034.0,
"step": 4005
},
{
"entropy": 5.767417287826538,
"epoch": 0.3637518142235123,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991579425929811,
"loss": 5.6506,
"mean_token_accuracy": 0.16089941710233688,
"num_tokens": 8322203.0,
"step": 4010
},
{
"entropy": 5.765074586868286,
"epoch": 0.36420537010159654,
"grad_norm": 1.0390625,
"learning_rate": 0.000499155143559372,
"loss": 5.5372,
"mean_token_accuracy": 0.1657981812953949,
"num_tokens": 8330981.0,
"step": 4015
},
{
"entropy": 5.65937442779541,
"epoch": 0.3646589259796807,
"grad_norm": 0.85546875,
"learning_rate": 0.0004991523398901773,
"loss": 5.5613,
"mean_token_accuracy": 0.15952345728874207,
"num_tokens": 8341294.0,
"step": 4020
},
{
"entropy": 5.869492626190185,
"epoch": 0.3651124818577649,
"grad_norm": 0.95703125,
"learning_rate": 0.0004991495315854553,
"loss": 5.7122,
"mean_token_accuracy": 0.15741421282291412,
"num_tokens": 8351824.0,
"step": 4025
},
{
"entropy": 5.71394624710083,
"epoch": 0.36556603773584906,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991467186452637,
"loss": 5.5716,
"mean_token_accuracy": 0.16435191184282302,
"num_tokens": 8361975.0,
"step": 4030
},
{
"entropy": 5.745710849761963,
"epoch": 0.3660195936139332,
"grad_norm": 0.890625,
"learning_rate": 0.0004991439010696609,
"loss": 5.5203,
"mean_token_accuracy": 0.16883532404899598,
"num_tokens": 8372145.0,
"step": 4035
},
{
"entropy": 5.778649616241455,
"epoch": 0.3664731494920174,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991410788587051,
"loss": 5.6151,
"mean_token_accuracy": 0.159750834107399,
"num_tokens": 8382542.0,
"step": 4040
},
{
"entropy": 5.749556159973144,
"epoch": 0.3669267053701016,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991382520124547,
"loss": 5.6097,
"mean_token_accuracy": 0.16143932342529296,
"num_tokens": 8393450.0,
"step": 4045
},
{
"entropy": 5.771503353118897,
"epoch": 0.3673802612481858,
"grad_norm": 0.8984375,
"learning_rate": 0.0004991354205309681,
"loss": 5.5472,
"mean_token_accuracy": 0.16442203521728516,
"num_tokens": 8404929.0,
"step": 4050
},
{
"entropy": 5.770533847808838,
"epoch": 0.36783381712626995,
"grad_norm": 0.87890625,
"learning_rate": 0.000499132584414304,
"loss": 5.5662,
"mean_token_accuracy": 0.1639400228857994,
"num_tokens": 8415703.0,
"step": 4055
},
{
"entropy": 5.767830038070679,
"epoch": 0.36828737300435416,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991297436625207,
"loss": 5.6591,
"mean_token_accuracy": 0.1587003231048584,
"num_tokens": 8426220.0,
"step": 4060
},
{
"entropy": 5.680564117431641,
"epoch": 0.3687409288824383,
"grad_norm": 1.0,
"learning_rate": 0.0004991268982756774,
"loss": 5.5295,
"mean_token_accuracy": 0.1664980798959732,
"num_tokens": 8436220.0,
"step": 4065
},
{
"entropy": 5.850871753692627,
"epoch": 0.3691944847605225,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991240482538327,
"loss": 5.6776,
"mean_token_accuracy": 0.15412717759609224,
"num_tokens": 8446724.0,
"step": 4070
},
{
"entropy": 5.742097663879394,
"epoch": 0.3696480406386067,
"grad_norm": 0.89453125,
"learning_rate": 0.0004991211935970455,
"loss": 5.5408,
"mean_token_accuracy": 0.16396154165267945,
"num_tokens": 8457270.0,
"step": 4075
},
{
"entropy": 5.729263162612915,
"epoch": 0.37010159651669083,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991183343053748,
"loss": 5.6115,
"mean_token_accuracy": 0.15851412415504457,
"num_tokens": 8468137.0,
"step": 4080
},
{
"entropy": 5.771952438354492,
"epoch": 0.37055515239477504,
"grad_norm": 0.87890625,
"learning_rate": 0.00049911547037888,
"loss": 5.5452,
"mean_token_accuracy": 0.1679990693926811,
"num_tokens": 8477574.0,
"step": 4085
},
{
"entropy": 5.742566680908203,
"epoch": 0.3710087082728592,
"grad_norm": 0.9609375,
"learning_rate": 0.0004991126018176202,
"loss": 5.6413,
"mean_token_accuracy": 0.1597900688648224,
"num_tokens": 8487268.0,
"step": 4090
},
{
"entropy": 5.877599859237671,
"epoch": 0.3714622641509434,
"grad_norm": 0.95703125,
"learning_rate": 0.0004991097286216545,
"loss": 5.7002,
"mean_token_accuracy": 0.1654977947473526,
"num_tokens": 8497785.0,
"step": 4095
},
{
"entropy": 5.80135407447815,
"epoch": 0.37191582002902757,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991068507910427,
"loss": 5.5666,
"mean_token_accuracy": 0.16508724838495253,
"num_tokens": 8508169.0,
"step": 4100
},
{
"entropy": 5.709864664077759,
"epoch": 0.3723693759071118,
"grad_norm": 0.84765625,
"learning_rate": 0.000499103968325844,
"loss": 5.5366,
"mean_token_accuracy": 0.16589654237031937,
"num_tokens": 8519201.0,
"step": 4105
},
{
"entropy": 5.795285272598266,
"epoch": 0.37282293178519593,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991010812261181,
"loss": 5.6332,
"mean_token_accuracy": 0.1585421934723854,
"num_tokens": 8531093.0,
"step": 4110
},
{
"entropy": 5.659151411056518,
"epoch": 0.37327648766328014,
"grad_norm": 0.9140625,
"learning_rate": 0.0004990981894919248,
"loss": 5.5081,
"mean_token_accuracy": 0.16795153617858888,
"num_tokens": 8541730.0,
"step": 4115
},
{
"entropy": 5.7505875587463375,
"epoch": 0.3737300435413643,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990952931233238,
"loss": 5.588,
"mean_token_accuracy": 0.16281794607639313,
"num_tokens": 8552570.0,
"step": 4120
},
{
"entropy": 5.799954271316528,
"epoch": 0.37418359941944845,
"grad_norm": 0.98046875,
"learning_rate": 0.000499092392120375,
"loss": 5.4892,
"mean_token_accuracy": 0.1670244187116623,
"num_tokens": 8562340.0,
"step": 4125
},
{
"entropy": 5.737644052505493,
"epoch": 0.37463715529753266,
"grad_norm": 0.8671875,
"learning_rate": 0.0004990894864831385,
"loss": 5.6899,
"mean_token_accuracy": 0.15905696153640747,
"num_tokens": 8573240.0,
"step": 4130
},
{
"entropy": 5.830232810974121,
"epoch": 0.3750907111756168,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990865762116743,
"loss": 5.5424,
"mean_token_accuracy": 0.16745613217353822,
"num_tokens": 8583194.0,
"step": 4135
},
{
"entropy": 5.733665180206299,
"epoch": 0.37554426705370103,
"grad_norm": 1.0,
"learning_rate": 0.0004990836613060426,
"loss": 5.6071,
"mean_token_accuracy": 0.16095204204320906,
"num_tokens": 8593603.0,
"step": 4140
},
{
"entropy": 5.6964035987854,
"epoch": 0.3759978229317852,
"grad_norm": 0.94921875,
"learning_rate": 0.0004990807417663036,
"loss": 5.6133,
"mean_token_accuracy": 0.1622975468635559,
"num_tokens": 8603176.0,
"step": 4145
},
{
"entropy": 5.8237403392791744,
"epoch": 0.3764513788098694,
"grad_norm": 0.8125,
"learning_rate": 0.0004990778175925179,
"loss": 5.6325,
"mean_token_accuracy": 0.16655509024858475,
"num_tokens": 8614801.0,
"step": 4150
},
{
"entropy": 5.793196535110473,
"epoch": 0.37690493468795355,
"grad_norm": 0.8671875,
"learning_rate": 0.0004990748887847457,
"loss": 5.6557,
"mean_token_accuracy": 0.1599831074476242,
"num_tokens": 8624999.0,
"step": 4155
},
{
"entropy": 5.700607967376709,
"epoch": 0.37735849056603776,
"grad_norm": 0.9609375,
"learning_rate": 0.0004990719553430477,
"loss": 5.507,
"mean_token_accuracy": 0.17496906369924545,
"num_tokens": 8635358.0,
"step": 4160
},
{
"entropy": 5.739441013336181,
"epoch": 0.3778120464441219,
"grad_norm": 0.9609375,
"learning_rate": 0.0004990690172674847,
"loss": 5.5865,
"mean_token_accuracy": 0.16173218041658402,
"num_tokens": 8645430.0,
"step": 4165
},
{
"entropy": 5.801822137832642,
"epoch": 0.3782656023222061,
"grad_norm": 0.90625,
"learning_rate": 0.0004990660745581172,
"loss": 5.5416,
"mean_token_accuracy": 0.1737535923719406,
"num_tokens": 8655115.0,
"step": 4170
},
{
"entropy": 5.693522024154663,
"epoch": 0.3787191582002903,
"grad_norm": 0.87109375,
"learning_rate": 0.0004990631272150062,
"loss": 5.5394,
"mean_token_accuracy": 0.16801448911428452,
"num_tokens": 8665391.0,
"step": 4175
},
{
"entropy": 5.767375373840332,
"epoch": 0.37917271407837444,
"grad_norm": 0.8125,
"learning_rate": 0.0004990601752382127,
"loss": 5.632,
"mean_token_accuracy": 0.1633898064494133,
"num_tokens": 8676819.0,
"step": 4180
},
{
"entropy": 5.8388364791870115,
"epoch": 0.37962626995645865,
"grad_norm": 0.8828125,
"learning_rate": 0.0004990572186277975,
"loss": 5.5607,
"mean_token_accuracy": 0.16711350232362748,
"num_tokens": 8687052.0,
"step": 4185
},
{
"entropy": 5.7258948802948,
"epoch": 0.3800798258345428,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990542573838221,
"loss": 5.6467,
"mean_token_accuracy": 0.1595759868621826,
"num_tokens": 8697756.0,
"step": 4190
},
{
"entropy": 5.6451514720916744,
"epoch": 0.380533381712627,
"grad_norm": 0.8984375,
"learning_rate": 0.0004990512915063475,
"loss": 5.5475,
"mean_token_accuracy": 0.17015636712312698,
"num_tokens": 8708388.0,
"step": 4195
},
{
"entropy": 5.812650442123413,
"epoch": 0.38098693759071117,
"grad_norm": 1.015625,
"learning_rate": 0.0004990483209954352,
"loss": 5.6359,
"mean_token_accuracy": 0.1576849937438965,
"num_tokens": 8719475.0,
"step": 4200
},
{
"entropy": 5.7326311588287355,
"epoch": 0.3814404934687954,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990453458511464,
"loss": 5.4986,
"mean_token_accuracy": 0.17547574192285537,
"num_tokens": 8729395.0,
"step": 4205
},
{
"entropy": 5.758198976516724,
"epoch": 0.38189404934687954,
"grad_norm": 0.9140625,
"learning_rate": 0.0004990423660735428,
"loss": 5.5483,
"mean_token_accuracy": 0.16976142972707747,
"num_tokens": 8740351.0,
"step": 4210
},
{
"entropy": 5.7952666759490965,
"epoch": 0.3823476052249637,
"grad_norm": 0.98046875,
"learning_rate": 0.000499039381662686,
"loss": 5.5672,
"mean_token_accuracy": 0.16848132163286209,
"num_tokens": 8749260.0,
"step": 4215
},
{
"entropy": 5.78771185874939,
"epoch": 0.3828011611030479,
"grad_norm": 0.9140625,
"learning_rate": 0.0004990363926186377,
"loss": 5.666,
"mean_token_accuracy": 0.15561919510364533,
"num_tokens": 8759604.0,
"step": 4220
},
{
"entropy": 5.8073570251464846,
"epoch": 0.38325471698113206,
"grad_norm": 1.03125,
"learning_rate": 0.0004990333989414597,
"loss": 5.6851,
"mean_token_accuracy": 0.155071659386158,
"num_tokens": 8769815.0,
"step": 4225
},
{
"entropy": 5.687266063690186,
"epoch": 0.38370827285921627,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990304006312138,
"loss": 5.6015,
"mean_token_accuracy": 0.16185775846242906,
"num_tokens": 8780557.0,
"step": 4230
},
{
"entropy": 5.777766180038452,
"epoch": 0.3841618287373004,
"grad_norm": 0.94921875,
"learning_rate": 0.0004990273976879622,
"loss": 5.5629,
"mean_token_accuracy": 0.17335893511772155,
"num_tokens": 8790993.0,
"step": 4235
},
{
"entropy": 5.768019533157348,
"epoch": 0.38461538461538464,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990243901117669,
"loss": 5.6234,
"mean_token_accuracy": 0.16034475564956666,
"num_tokens": 8800657.0,
"step": 4240
},
{
"entropy": 5.67515926361084,
"epoch": 0.3850689404934688,
"grad_norm": 0.86328125,
"learning_rate": 0.0004990213779026903,
"loss": 5.6138,
"mean_token_accuracy": 0.15936714857816697,
"num_tokens": 8812983.0,
"step": 4245
},
{
"entropy": 5.800050735473633,
"epoch": 0.385522496371553,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990183610607944,
"loss": 5.5895,
"mean_token_accuracy": 0.16557528376579284,
"num_tokens": 8822980.0,
"step": 4250
},
{
"entropy": 5.733969593048096,
"epoch": 0.38597605224963716,
"grad_norm": 0.91015625,
"learning_rate": 0.0004990153395861415,
"loss": 5.5572,
"mean_token_accuracy": 0.16649172306060792,
"num_tokens": 8833821.0,
"step": 4255
},
{
"entropy": 5.672335815429688,
"epoch": 0.3864296081277213,
"grad_norm": 0.89453125,
"learning_rate": 0.0004990123134787946,
"loss": 5.509,
"mean_token_accuracy": 0.16983023583889006,
"num_tokens": 8843929.0,
"step": 4260
},
{
"entropy": 5.780575275421143,
"epoch": 0.3868831640058055,
"grad_norm": 0.94921875,
"learning_rate": 0.0004990092827388159,
"loss": 5.493,
"mean_token_accuracy": 0.16734486669301987,
"num_tokens": 8854180.0,
"step": 4265
},
{
"entropy": 5.6601338386535645,
"epoch": 0.3873367198838897,
"grad_norm": 0.89453125,
"learning_rate": 0.0004990062473662681,
"loss": 5.5037,
"mean_token_accuracy": 0.17173054963350295,
"num_tokens": 8865018.0,
"step": 4270
},
{
"entropy": 5.781656074523926,
"epoch": 0.3877902757619739,
"grad_norm": 1.03125,
"learning_rate": 0.000499003207361214,
"loss": 5.6868,
"mean_token_accuracy": 0.1598363995552063,
"num_tokens": 8875423.0,
"step": 4275
},
{
"entropy": 5.845520830154419,
"epoch": 0.38824383164005805,
"grad_norm": 0.8984375,
"learning_rate": 0.0004990001627237165,
"loss": 5.6193,
"mean_token_accuracy": 0.1609180137515068,
"num_tokens": 8887850.0,
"step": 4280
},
{
"entropy": 5.6677868366241455,
"epoch": 0.38869738751814226,
"grad_norm": 1.03125,
"learning_rate": 0.0004989971134538386,
"loss": 5.5183,
"mean_token_accuracy": 0.16817747205495834,
"num_tokens": 8897429.0,
"step": 4285
},
{
"entropy": 5.684417486190796,
"epoch": 0.3891509433962264,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989940595516432,
"loss": 5.5406,
"mean_token_accuracy": 0.1706473186612129,
"num_tokens": 8907048.0,
"step": 4290
},
{
"entropy": 5.759358978271484,
"epoch": 0.3896044992743106,
"grad_norm": 0.97265625,
"learning_rate": 0.0004989910010171938,
"loss": 5.6105,
"mean_token_accuracy": 0.16018391102552415,
"num_tokens": 8917987.0,
"step": 4295
},
{
"entropy": 5.771234941482544,
"epoch": 0.3900580551523948,
"grad_norm": 0.8046875,
"learning_rate": 0.0004989879378505532,
"loss": 5.4998,
"mean_token_accuracy": 0.1641297698020935,
"num_tokens": 8929339.0,
"step": 4300
},
{
"entropy": 5.629133081436157,
"epoch": 0.39051161103047893,
"grad_norm": 0.875,
"learning_rate": 0.0004989848700517849,
"loss": 5.5118,
"mean_token_accuracy": 0.1678486481308937,
"num_tokens": 8939473.0,
"step": 4305
},
{
"entropy": 5.693001365661621,
"epoch": 0.39096516690856314,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989817976209526,
"loss": 5.542,
"mean_token_accuracy": 0.16416483521461486,
"num_tokens": 8949173.0,
"step": 4310
},
{
"entropy": 5.759200525283814,
"epoch": 0.3914187227866473,
"grad_norm": 0.91015625,
"learning_rate": 0.0004989787205581196,
"loss": 5.5195,
"mean_token_accuracy": 0.17088344991207122,
"num_tokens": 8959879.0,
"step": 4315
},
{
"entropy": 5.704134798049926,
"epoch": 0.3918722786647315,
"grad_norm": 0.953125,
"learning_rate": 0.0004989756388633496,
"loss": 5.6041,
"mean_token_accuracy": 0.1627824455499649,
"num_tokens": 8970643.0,
"step": 4320
},
{
"entropy": 5.710903739929199,
"epoch": 0.39232583454281567,
"grad_norm": 0.90625,
"learning_rate": 0.0004989725525367063,
"loss": 5.4835,
"mean_token_accuracy": 0.17103770971298218,
"num_tokens": 8979882.0,
"step": 4325
},
{
"entropy": 5.7799781322479244,
"epoch": 0.3927793904208999,
"grad_norm": 0.86328125,
"learning_rate": 0.0004989694615782534,
"loss": 5.5981,
"mean_token_accuracy": 0.15929665714502333,
"num_tokens": 8990925.0,
"step": 4330
},
{
"entropy": 5.762790632247925,
"epoch": 0.39323294629898403,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989663659880552,
"loss": 5.6236,
"mean_token_accuracy": 0.16410700231790543,
"num_tokens": 9001365.0,
"step": 4335
},
{
"entropy": 5.808406639099121,
"epoch": 0.3936865021770682,
"grad_norm": 0.91015625,
"learning_rate": 0.0004989632657661754,
"loss": 5.6151,
"mean_token_accuracy": 0.16846253722906113,
"num_tokens": 9011942.0,
"step": 4340
},
{
"entropy": 5.748916530609131,
"epoch": 0.3941400580551524,
"grad_norm": 0.93359375,
"learning_rate": 0.000498960160912678,
"loss": 5.665,
"mean_token_accuracy": 0.16369327455759047,
"num_tokens": 9022610.0,
"step": 4345
},
{
"entropy": 5.734598922729492,
"epoch": 0.39459361393323655,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989570514276276,
"loss": 5.5523,
"mean_token_accuracy": 0.16294773668050766,
"num_tokens": 9033034.0,
"step": 4350
},
{
"entropy": 5.767314720153808,
"epoch": 0.39504716981132076,
"grad_norm": 0.8828125,
"learning_rate": 0.0004989539373110883,
"loss": 5.5898,
"mean_token_accuracy": 0.160574010014534,
"num_tokens": 9043418.0,
"step": 4355
},
{
"entropy": 5.782724094390869,
"epoch": 0.3955007256894049,
"grad_norm": 0.99609375,
"learning_rate": 0.0004989508185631245,
"loss": 5.6765,
"mean_token_accuracy": 0.15920473337173463,
"num_tokens": 9052708.0,
"step": 4360
},
{
"entropy": 5.673357582092285,
"epoch": 0.39595428156748913,
"grad_norm": 0.96875,
"learning_rate": 0.0004989476951838006,
"loss": 5.5266,
"mean_token_accuracy": 0.1611687034368515,
"num_tokens": 9062883.0,
"step": 4365
},
{
"entropy": 5.750123310089111,
"epoch": 0.3964078374455733,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989445671731814,
"loss": 5.4954,
"mean_token_accuracy": 0.17254894524812697,
"num_tokens": 9072830.0,
"step": 4370
},
{
"entropy": 5.673724603652954,
"epoch": 0.3968613933236575,
"grad_norm": 0.94140625,
"learning_rate": 0.0004989414345313315,
"loss": 5.4632,
"mean_token_accuracy": 0.16166009455919267,
"num_tokens": 9082749.0,
"step": 4375
},
{
"entropy": 5.779737043380737,
"epoch": 0.39731494920174165,
"grad_norm": 0.82421875,
"learning_rate": 0.0004989382972583156,
"loss": 5.5433,
"mean_token_accuracy": 0.1635667696595192,
"num_tokens": 9094188.0,
"step": 4380
},
{
"entropy": 5.701415920257569,
"epoch": 0.3977685050798258,
"grad_norm": 1.03125,
"learning_rate": 0.0004989351553541986,
"loss": 5.5767,
"mean_token_accuracy": 0.15773947685956954,
"num_tokens": 9105023.0,
"step": 4385
},
{
"entropy": 5.63522777557373,
"epoch": 0.39822206095791,
"grad_norm": 0.87109375,
"learning_rate": 0.0004989320088190456,
"loss": 5.5136,
"mean_token_accuracy": 0.16800401508808135,
"num_tokens": 9115563.0,
"step": 4390
},
{
"entropy": 5.710983753204346,
"epoch": 0.3986756168359942,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989288576529216,
"loss": 5.5451,
"mean_token_accuracy": 0.16579101979732513,
"num_tokens": 9125394.0,
"step": 4395
},
{
"entropy": 5.7169290542602536,
"epoch": 0.3991291727140784,
"grad_norm": 0.90625,
"learning_rate": 0.0004989257018558917,
"loss": 5.5305,
"mean_token_accuracy": 0.1630067154765129,
"num_tokens": 9137061.0,
"step": 4400
},
{
"entropy": 5.747994899749756,
"epoch": 0.39958272859216254,
"grad_norm": 0.90625,
"learning_rate": 0.0004989225414280213,
"loss": 5.5093,
"mean_token_accuracy": 0.16748605370521547,
"num_tokens": 9148196.0,
"step": 4405
},
{
"entropy": 5.673751354217529,
"epoch": 0.40003628447024675,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989193763693756,
"loss": 5.5305,
"mean_token_accuracy": 0.17181627005338668,
"num_tokens": 9158013.0,
"step": 4410
},
{
"entropy": 5.692364835739136,
"epoch": 0.4004898403483309,
"grad_norm": 0.8984375,
"learning_rate": 0.0004989162066800202,
"loss": 5.4596,
"mean_token_accuracy": 0.17420485466718674,
"num_tokens": 9167813.0,
"step": 4415
},
{
"entropy": 5.68905611038208,
"epoch": 0.4009433962264151,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989130323600205,
"loss": 5.5125,
"mean_token_accuracy": 0.16890402287244796,
"num_tokens": 9177783.0,
"step": 4420
},
{
"entropy": 5.698638820648194,
"epoch": 0.40139695210449927,
"grad_norm": 0.94140625,
"learning_rate": 0.0004989098534094424,
"loss": 5.5409,
"mean_token_accuracy": 0.16684163510799407,
"num_tokens": 9187556.0,
"step": 4425
},
{
"entropy": 5.702458286285401,
"epoch": 0.4018505079825834,
"grad_norm": 0.88671875,
"learning_rate": 0.0004989066698283512,
"loss": 5.6142,
"mean_token_accuracy": 0.16072579473257065,
"num_tokens": 9198659.0,
"step": 4430
},
{
"entropy": 5.740787506103516,
"epoch": 0.40230406386066764,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989034816168132,
"loss": 5.5916,
"mean_token_accuracy": 0.16459962129592895,
"num_tokens": 9209546.0,
"step": 4435
},
{
"entropy": 5.698708724975586,
"epoch": 0.4027576197387518,
"grad_norm": 0.8984375,
"learning_rate": 0.0004989002887748941,
"loss": 5.5123,
"mean_token_accuracy": 0.16735931783914565,
"num_tokens": 9219339.0,
"step": 4440
},
{
"entropy": 5.7313940048217775,
"epoch": 0.403211175616836,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988970913026601,
"loss": 5.6901,
"mean_token_accuracy": 0.15663133114576339,
"num_tokens": 9230364.0,
"step": 4445
},
{
"entropy": 5.746043014526367,
"epoch": 0.40366473149492016,
"grad_norm": 0.91015625,
"learning_rate": 0.0004988938892001771,
"loss": 5.5391,
"mean_token_accuracy": 0.16293537616729736,
"num_tokens": 9240732.0,
"step": 4450
},
{
"entropy": 5.651700019836426,
"epoch": 0.40411828737300437,
"grad_norm": 0.9296875,
"learning_rate": 0.0004988906824675113,
"loss": 5.4611,
"mean_token_accuracy": 0.16927590519189833,
"num_tokens": 9251049.0,
"step": 4455
},
{
"entropy": 5.702632808685303,
"epoch": 0.4045718432510885,
"grad_norm": 0.90625,
"learning_rate": 0.0004988874711047293,
"loss": 5.5209,
"mean_token_accuracy": 0.1687337875366211,
"num_tokens": 9261660.0,
"step": 4460
},
{
"entropy": 5.723224210739136,
"epoch": 0.40502539912917274,
"grad_norm": 0.890625,
"learning_rate": 0.0004988842551118972,
"loss": 5.5738,
"mean_token_accuracy": 0.1679667592048645,
"num_tokens": 9272114.0,
"step": 4465
},
{
"entropy": 5.7394585609436035,
"epoch": 0.4054789550072569,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988810344890818,
"loss": 5.5537,
"mean_token_accuracy": 0.1705055981874466,
"num_tokens": 9282125.0,
"step": 4470
},
{
"entropy": 5.77500205039978,
"epoch": 0.40593251088534105,
"grad_norm": 0.890625,
"learning_rate": 0.0004988778092363494,
"loss": 5.5773,
"mean_token_accuracy": 0.16275593191385268,
"num_tokens": 9294078.0,
"step": 4475
},
{
"entropy": 5.723077011108399,
"epoch": 0.40638606676342526,
"grad_norm": 0.98828125,
"learning_rate": 0.0004988745793537671,
"loss": 5.5452,
"mean_token_accuracy": 0.16284733265638351,
"num_tokens": 9304026.0,
"step": 4480
},
{
"entropy": 5.714406442642212,
"epoch": 0.4068396226415094,
"grad_norm": 0.9296875,
"learning_rate": 0.0004988713448414011,
"loss": 5.5858,
"mean_token_accuracy": 0.17267868667840958,
"num_tokens": 9313933.0,
"step": 4485
},
{
"entropy": 5.722243356704712,
"epoch": 0.4072931785195936,
"grad_norm": 0.90234375,
"learning_rate": 0.000498868105699319,
"loss": 5.6213,
"mean_token_accuracy": 0.16226827800273896,
"num_tokens": 9323932.0,
"step": 4490
},
{
"entropy": 5.6608726978302,
"epoch": 0.4077467343976778,
"grad_norm": 0.91796875,
"learning_rate": 0.0004988648619275872,
"loss": 5.5272,
"mean_token_accuracy": 0.17585674673318863,
"num_tokens": 9333467.0,
"step": 4495
},
{
"entropy": 5.7471529006958,
"epoch": 0.408200290275762,
"grad_norm": 0.8515625,
"learning_rate": 0.0004988616135262729,
"loss": 5.5802,
"mean_token_accuracy": 0.16007377952337265,
"num_tokens": 9344909.0,
"step": 4500
},
{
"entropy": 5.8212768077850345,
"epoch": 0.40865384615384615,
"grad_norm": 0.82421875,
"learning_rate": 0.0004988583604954436,
"loss": 5.6265,
"mean_token_accuracy": 0.1633604183793068,
"num_tokens": 9355630.0,
"step": 4505
},
{
"entropy": 5.634971952438354,
"epoch": 0.40910740203193036,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988551028351663,
"loss": 5.5539,
"mean_token_accuracy": 0.16576218158006667,
"num_tokens": 9365700.0,
"step": 4510
},
{
"entropy": 5.6949104309082035,
"epoch": 0.4095609579100145,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988518405455084,
"loss": 5.6367,
"mean_token_accuracy": 0.15977054387331008,
"num_tokens": 9377088.0,
"step": 4515
},
{
"entropy": 5.749986362457276,
"epoch": 0.41001451378809867,
"grad_norm": 0.8828125,
"learning_rate": 0.0004988485736265374,
"loss": 5.4339,
"mean_token_accuracy": 0.1753544732928276,
"num_tokens": 9386855.0,
"step": 4520
},
{
"entropy": 5.69621696472168,
"epoch": 0.4104680696661829,
"grad_norm": 0.96875,
"learning_rate": 0.000498845302078321,
"loss": 5.5548,
"mean_token_accuracy": 0.16240098476409912,
"num_tokens": 9398661.0,
"step": 4525
},
{
"entropy": 5.738919639587403,
"epoch": 0.41092162554426703,
"grad_norm": 1.0,
"learning_rate": 0.0004988420259009266,
"loss": 5.5603,
"mean_token_accuracy": 0.17131945639848709,
"num_tokens": 9408670.0,
"step": 4530
},
{
"entropy": 5.6992714405059814,
"epoch": 0.41137518142235124,
"grad_norm": 0.85546875,
"learning_rate": 0.000498838745094422,
"loss": 5.5582,
"mean_token_accuracy": 0.16838853061199188,
"num_tokens": 9418740.0,
"step": 4535
},
{
"entropy": 5.6787707805633545,
"epoch": 0.4118287373004354,
"grad_norm": 0.9453125,
"learning_rate": 0.0004988354596588751,
"loss": 5.5485,
"mean_token_accuracy": 0.16721292436122895,
"num_tokens": 9429339.0,
"step": 4540
},
{
"entropy": 5.7141166687011715,
"epoch": 0.4122822931785196,
"grad_norm": 0.87890625,
"learning_rate": 0.0004988321695943539,
"loss": 5.4913,
"mean_token_accuracy": 0.17288169413805007,
"num_tokens": 9438996.0,
"step": 4545
},
{
"entropy": 5.6817183017730715,
"epoch": 0.41273584905660377,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988288749009264,
"loss": 5.5176,
"mean_token_accuracy": 0.16724181026220322,
"num_tokens": 9448669.0,
"step": 4550
},
{
"entropy": 5.677099323272705,
"epoch": 0.413189404934688,
"grad_norm": 0.93359375,
"learning_rate": 0.0004988255755786608,
"loss": 5.489,
"mean_token_accuracy": 0.17341368198394774,
"num_tokens": 9458857.0,
"step": 4555
},
{
"entropy": 5.6843749523162845,
"epoch": 0.41364296081277213,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988222716276251,
"loss": 5.4927,
"mean_token_accuracy": 0.16991052776575089,
"num_tokens": 9470241.0,
"step": 4560
},
{
"entropy": 5.75704345703125,
"epoch": 0.4140965166908563,
"grad_norm": 0.89453125,
"learning_rate": 0.000498818963047888,
"loss": 5.6476,
"mean_token_accuracy": 0.16511964052915573,
"num_tokens": 9480528.0,
"step": 4565
},
{
"entropy": 5.7459704875946045,
"epoch": 0.4145500725689405,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988156498395176,
"loss": 5.5663,
"mean_token_accuracy": 0.16632131338119507,
"num_tokens": 9491000.0,
"step": 4570
},
{
"entropy": 5.728471374511718,
"epoch": 0.41500362844702465,
"grad_norm": 1.3203125,
"learning_rate": 0.0004988123320025825,
"loss": 5.5504,
"mean_token_accuracy": 0.1686439707875252,
"num_tokens": 9501612.0,
"step": 4575
},
{
"entropy": 5.721394157409668,
"epoch": 0.41545718432510886,
"grad_norm": 0.875,
"learning_rate": 0.0004988090095371514,
"loss": 5.6569,
"mean_token_accuracy": 0.16329480931162835,
"num_tokens": 9513999.0,
"step": 4580
},
{
"entropy": 5.723319435119629,
"epoch": 0.415910740203193,
"grad_norm": 0.93359375,
"learning_rate": 0.0004988056824432928,
"loss": 5.5296,
"mean_token_accuracy": 0.1624149963259697,
"num_tokens": 9524392.0,
"step": 4585
},
{
"entropy": 5.747483777999878,
"epoch": 0.41636429608127723,
"grad_norm": 0.953125,
"learning_rate": 0.0004988023507210758,
"loss": 5.5387,
"mean_token_accuracy": 0.16955992728471755,
"num_tokens": 9534227.0,
"step": 4590
},
{
"entropy": 5.693861675262451,
"epoch": 0.4168178519593614,
"grad_norm": 0.93359375,
"learning_rate": 0.0004987990143705692,
"loss": 5.5163,
"mean_token_accuracy": 0.1732896625995636,
"num_tokens": 9544242.0,
"step": 4595
},
{
"entropy": 5.671651411056518,
"epoch": 0.4172714078374456,
"grad_norm": 0.921875,
"learning_rate": 0.0004987956733918418,
"loss": 5.4526,
"mean_token_accuracy": 0.16845765262842177,
"num_tokens": 9553921.0,
"step": 4600
},
{
"entropy": 5.6189477920532225,
"epoch": 0.41772496371552975,
"grad_norm": 0.83984375,
"learning_rate": 0.0004987923277849629,
"loss": 5.5082,
"mean_token_accuracy": 0.16487912237644195,
"num_tokens": 9564768.0,
"step": 4605
},
{
"entropy": 5.716268110275268,
"epoch": 0.4181785195936139,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987889775500017,
"loss": 5.5388,
"mean_token_accuracy": 0.16002229750156402,
"num_tokens": 9576360.0,
"step": 4610
},
{
"entropy": 5.702955532073974,
"epoch": 0.4186320754716981,
"grad_norm": 0.8203125,
"learning_rate": 0.0004987856226870272,
"loss": 5.5542,
"mean_token_accuracy": 0.16782574057579042,
"num_tokens": 9587623.0,
"step": 4615
},
{
"entropy": 5.687364339828491,
"epoch": 0.4190856313497823,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987822631961092,
"loss": 5.5773,
"mean_token_accuracy": 0.16299049258232118,
"num_tokens": 9597837.0,
"step": 4620
},
{
"entropy": 5.740718221664428,
"epoch": 0.4195391872278665,
"grad_norm": 0.88671875,
"learning_rate": 0.000498778899077317,
"loss": 5.5345,
"mean_token_accuracy": 0.1651468500494957,
"num_tokens": 9609509.0,
"step": 4625
},
{
"entropy": 5.688720798492431,
"epoch": 0.41999274310595064,
"grad_norm": 0.93359375,
"learning_rate": 0.00049877553033072,
"loss": 5.5257,
"mean_token_accuracy": 0.1684991091489792,
"num_tokens": 9619954.0,
"step": 4630
},
{
"entropy": 5.647857522964477,
"epoch": 0.42044629898403485,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987721569563882,
"loss": 5.5167,
"mean_token_accuracy": 0.1662772074341774,
"num_tokens": 9629742.0,
"step": 4635
},
{
"entropy": 5.702399444580078,
"epoch": 0.420899854862119,
"grad_norm": 0.90234375,
"learning_rate": 0.000498768778954391,
"loss": 5.5327,
"mean_token_accuracy": 0.16915884912014006,
"num_tokens": 9641378.0,
"step": 4640
},
{
"entropy": 5.798724412918091,
"epoch": 0.4213534107402032,
"grad_norm": 0.875,
"learning_rate": 0.0004987653963247985,
"loss": 5.5901,
"mean_token_accuracy": 0.16833552569150925,
"num_tokens": 9652279.0,
"step": 4645
},
{
"entropy": 5.7283261775970455,
"epoch": 0.42180696661828737,
"grad_norm": 0.8125,
"learning_rate": 0.0004987620090676805,
"loss": 5.6067,
"mean_token_accuracy": 0.16763487905263902,
"num_tokens": 9663416.0,
"step": 4650
},
{
"entropy": 5.70868330001831,
"epoch": 0.4222605224963715,
"grad_norm": 0.85546875,
"learning_rate": 0.0004987586171831072,
"loss": 5.5433,
"mean_token_accuracy": 0.16505215167999268,
"num_tokens": 9674647.0,
"step": 4655
},
{
"entropy": 5.6503729820251465,
"epoch": 0.42271407837445574,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987552206711487,
"loss": 5.4025,
"mean_token_accuracy": 0.17310979068279267,
"num_tokens": 9685084.0,
"step": 4660
},
{
"entropy": 5.647482967376709,
"epoch": 0.4231676342525399,
"grad_norm": 0.875,
"learning_rate": 0.0004987518195318752,
"loss": 5.4772,
"mean_token_accuracy": 0.16532573699951172,
"num_tokens": 9695661.0,
"step": 4665
},
{
"entropy": 5.748704338073731,
"epoch": 0.4236211901306241,
"grad_norm": 0.98828125,
"learning_rate": 0.000498748413765357,
"loss": 5.5315,
"mean_token_accuracy": 0.1663697361946106,
"num_tokens": 9705773.0,
"step": 4670
},
{
"entropy": 5.718989038467408,
"epoch": 0.42407474600870826,
"grad_norm": 0.98046875,
"learning_rate": 0.0004987450033716646,
"loss": 5.5801,
"mean_token_accuracy": 0.16458225697278978,
"num_tokens": 9715799.0,
"step": 4675
},
{
"entropy": 5.71646819114685,
"epoch": 0.42452830188679247,
"grad_norm": 0.84375,
"learning_rate": 0.0004987415883508686,
"loss": 5.5205,
"mean_token_accuracy": 0.1714474231004715,
"num_tokens": 9727012.0,
"step": 4680
},
{
"entropy": 5.704070425033569,
"epoch": 0.4249818577648766,
"grad_norm": 0.9140625,
"learning_rate": 0.0004987381687030394,
"loss": 5.5332,
"mean_token_accuracy": 0.16558614522218704,
"num_tokens": 9737044.0,
"step": 4685
},
{
"entropy": 5.787160110473633,
"epoch": 0.42543541364296084,
"grad_norm": 0.98046875,
"learning_rate": 0.0004987347444282479,
"loss": 5.6288,
"mean_token_accuracy": 0.157316155731678,
"num_tokens": 9747648.0,
"step": 4690
},
{
"entropy": 5.674534463882447,
"epoch": 0.425888969521045,
"grad_norm": 0.984375,
"learning_rate": 0.0004987313155265649,
"loss": 5.5383,
"mean_token_accuracy": 0.16501632183790207,
"num_tokens": 9757812.0,
"step": 4695
},
{
"entropy": 5.7545582294464115,
"epoch": 0.42634252539912915,
"grad_norm": 0.87109375,
"learning_rate": 0.0004987278819980613,
"loss": 5.5327,
"mean_token_accuracy": 0.17061222344636917,
"num_tokens": 9768337.0,
"step": 4700
},
{
"entropy": 5.7120167255401615,
"epoch": 0.42679608127721336,
"grad_norm": 0.87890625,
"learning_rate": 0.0004987244438428081,
"loss": 5.5817,
"mean_token_accuracy": 0.16203325092792512,
"num_tokens": 9779433.0,
"step": 4705
},
{
"entropy": 5.6578521728515625,
"epoch": 0.4272496371552975,
"grad_norm": 0.8671875,
"learning_rate": 0.0004987210010608764,
"loss": 5.4676,
"mean_token_accuracy": 0.16993781328201293,
"num_tokens": 9790069.0,
"step": 4710
},
{
"entropy": 5.673728132247925,
"epoch": 0.4277031930333817,
"grad_norm": 0.87890625,
"learning_rate": 0.0004987175536523373,
"loss": 5.4468,
"mean_token_accuracy": 0.17233940958976746,
"num_tokens": 9799669.0,
"step": 4715
},
{
"entropy": 5.647679948806763,
"epoch": 0.4281567489114659,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987141016172622,
"loss": 5.49,
"mean_token_accuracy": 0.17301111817359924,
"num_tokens": 9809615.0,
"step": 4720
},
{
"entropy": 5.668295860290527,
"epoch": 0.4286103047895501,
"grad_norm": 1.0078125,
"learning_rate": 0.0004987106449557225,
"loss": 5.4647,
"mean_token_accuracy": 0.17097459584474564,
"num_tokens": 9819666.0,
"step": 4725
},
{
"entropy": 5.747781991958618,
"epoch": 0.42906386066763424,
"grad_norm": 0.953125,
"learning_rate": 0.0004987071836677896,
"loss": 5.6161,
"mean_token_accuracy": 0.16620567291975022,
"num_tokens": 9831015.0,
"step": 4730
},
{
"entropy": 5.681993722915649,
"epoch": 0.42951741654571846,
"grad_norm": 0.984375,
"learning_rate": 0.0004987037177535353,
"loss": 5.4838,
"mean_token_accuracy": 0.16538906544446946,
"num_tokens": 9841354.0,
"step": 4735
},
{
"entropy": 5.61437578201294,
"epoch": 0.4299709724238026,
"grad_norm": 0.9765625,
"learning_rate": 0.0004987002472130309,
"loss": 5.4796,
"mean_token_accuracy": 0.1654519483447075,
"num_tokens": 9850467.0,
"step": 4740
},
{
"entropy": 5.749066305160523,
"epoch": 0.43042452830188677,
"grad_norm": 0.91015625,
"learning_rate": 0.0004986967720463485,
"loss": 5.5535,
"mean_token_accuracy": 0.1634979486465454,
"num_tokens": 9861115.0,
"step": 4745
},
{
"entropy": 5.75732159614563,
"epoch": 0.430878084179971,
"grad_norm": 0.91015625,
"learning_rate": 0.0004986932922535598,
"loss": 5.5587,
"mean_token_accuracy": 0.16462093889713286,
"num_tokens": 9871172.0,
"step": 4750
},
{
"entropy": 5.690928268432617,
"epoch": 0.43133164005805513,
"grad_norm": 0.9375,
"learning_rate": 0.0004986898078347368,
"loss": 5.4576,
"mean_token_accuracy": 0.171144238114357,
"num_tokens": 9881997.0,
"step": 4755
},
{
"entropy": 5.67418532371521,
"epoch": 0.43178519593613934,
"grad_norm": 0.89453125,
"learning_rate": 0.0004986863187899516,
"loss": 5.4497,
"mean_token_accuracy": 0.17098109871149064,
"num_tokens": 9892310.0,
"step": 4760
},
{
"entropy": 5.6178083419799805,
"epoch": 0.4322387518142235,
"grad_norm": 0.91796875,
"learning_rate": 0.0004986828251192764,
"loss": 5.4971,
"mean_token_accuracy": 0.16893473714590074,
"num_tokens": 9902664.0,
"step": 4765
},
{
"entropy": 5.691957664489746,
"epoch": 0.4326923076923077,
"grad_norm": 1.0,
"learning_rate": 0.0004986793268227833,
"loss": 5.4794,
"mean_token_accuracy": 0.17000094205141067,
"num_tokens": 9912491.0,
"step": 4770
},
{
"entropy": 5.6628247737884525,
"epoch": 0.43314586357039186,
"grad_norm": 0.9765625,
"learning_rate": 0.0004986758239005448,
"loss": 5.4546,
"mean_token_accuracy": 0.1711384743452072,
"num_tokens": 9922057.0,
"step": 4775
},
{
"entropy": 5.629938793182373,
"epoch": 0.4335994194484761,
"grad_norm": 0.8203125,
"learning_rate": 0.0004986723163526331,
"loss": 5.4915,
"mean_token_accuracy": 0.16978790760040283,
"num_tokens": 9933661.0,
"step": 4780
},
{
"entropy": 5.644828224182129,
"epoch": 0.43405297532656023,
"grad_norm": 0.82421875,
"learning_rate": 0.0004986688041791209,
"loss": 5.4595,
"mean_token_accuracy": 0.16317514777183534,
"num_tokens": 9945177.0,
"step": 4785
},
{
"entropy": 5.686159992218018,
"epoch": 0.4345065312046444,
"grad_norm": 0.94921875,
"learning_rate": 0.000498665287380081,
"loss": 5.5007,
"mean_token_accuracy": 0.1679864302277565,
"num_tokens": 9956226.0,
"step": 4790
},
{
"entropy": 5.728806781768799,
"epoch": 0.4349600870827286,
"grad_norm": 0.91015625,
"learning_rate": 0.0004986617659555859,
"loss": 5.566,
"mean_token_accuracy": 0.1681407630443573,
"num_tokens": 9965877.0,
"step": 4795
},
{
"entropy": 5.666096782684326,
"epoch": 0.43541364296081275,
"grad_norm": 0.95703125,
"learning_rate": 0.0004986582399057085,
"loss": 5.5077,
"mean_token_accuracy": 0.1807382047176361,
"num_tokens": 9975592.0,
"step": 4800
},
{
"entropy": 5.740148448944092,
"epoch": 0.43586719883889696,
"grad_norm": 0.94140625,
"learning_rate": 0.0004986547092305216,
"loss": 5.5476,
"mean_token_accuracy": 0.16806693226099015,
"num_tokens": 9986094.0,
"step": 4805
},
{
"entropy": 5.738724565505981,
"epoch": 0.4363207547169811,
"grad_norm": 0.9375,
"learning_rate": 0.0004986511739300984,
"loss": 5.5802,
"mean_token_accuracy": 0.16459282785654067,
"num_tokens": 9997241.0,
"step": 4810
},
{
"entropy": 5.62156834602356,
"epoch": 0.43677431059506533,
"grad_norm": 0.859375,
"learning_rate": 0.0004986476340045119,
"loss": 5.4877,
"mean_token_accuracy": 0.1692790299654007,
"num_tokens": 10008091.0,
"step": 4815
},
{
"entropy": 5.640922546386719,
"epoch": 0.4372278664731495,
"grad_norm": 0.91796875,
"learning_rate": 0.0004986440894538354,
"loss": 5.5512,
"mean_token_accuracy": 0.16694139540195466,
"num_tokens": 10018794.0,
"step": 4820
},
{
"entropy": 5.696912527084351,
"epoch": 0.4376814223512337,
"grad_norm": 0.921875,
"learning_rate": 0.0004986405402781421,
"loss": 5.477,
"mean_token_accuracy": 0.17061326652765274,
"num_tokens": 10028414.0,
"step": 4825
},
{
"entropy": 5.734801959991455,
"epoch": 0.43813497822931785,
"grad_norm": 0.86328125,
"learning_rate": 0.0004986369864775054,
"loss": 5.5394,
"mean_token_accuracy": 0.1690880686044693,
"num_tokens": 10038749.0,
"step": 4830
},
{
"entropy": 5.673574686050415,
"epoch": 0.438588534107402,
"grad_norm": 0.90625,
"learning_rate": 0.0004986334280519989,
"loss": 5.4989,
"mean_token_accuracy": 0.16510852426290512,
"num_tokens": 10049866.0,
"step": 4835
},
{
"entropy": 5.658323574066162,
"epoch": 0.4390420899854862,
"grad_norm": 0.921875,
"learning_rate": 0.000498629865001696,
"loss": 5.5213,
"mean_token_accuracy": 0.173689503967762,
"num_tokens": 10060680.0,
"step": 4840
},
{
"entropy": 5.773933458328247,
"epoch": 0.43949564586357037,
"grad_norm": 0.89453125,
"learning_rate": 0.0004986262973266706,
"loss": 5.6142,
"mean_token_accuracy": 0.1640276402235031,
"num_tokens": 10070931.0,
"step": 4845
},
{
"entropy": 5.665406703948975,
"epoch": 0.4399492017416546,
"grad_norm": 0.921875,
"learning_rate": 0.0004986227250269964,
"loss": 5.5471,
"mean_token_accuracy": 0.1679753065109253,
"num_tokens": 10080189.0,
"step": 4850
},
{
"entropy": 5.692187070846558,
"epoch": 0.44040275761973874,
"grad_norm": 1.046875,
"learning_rate": 0.0004986191481027472,
"loss": 5.4578,
"mean_token_accuracy": 0.17034661769866943,
"num_tokens": 10088813.0,
"step": 4855
},
{
"entropy": 5.625132513046265,
"epoch": 0.44085631349782295,
"grad_norm": 0.984375,
"learning_rate": 0.0004986155665539971,
"loss": 5.4795,
"mean_token_accuracy": 0.16770929843187332,
"num_tokens": 10098125.0,
"step": 4860
},
{
"entropy": 5.675574922561646,
"epoch": 0.4413098693759071,
"grad_norm": 0.921875,
"learning_rate": 0.00049861198038082,
"loss": 5.5161,
"mean_token_accuracy": 0.1693737342953682,
"num_tokens": 10109125.0,
"step": 4865
},
{
"entropy": 5.682723617553711,
"epoch": 0.4417634252539913,
"grad_norm": 0.88671875,
"learning_rate": 0.0004986083895832903,
"loss": 5.4948,
"mean_token_accuracy": 0.16549968123435974,
"num_tokens": 10120182.0,
"step": 4870
},
{
"entropy": 5.645736122131348,
"epoch": 0.44221698113207547,
"grad_norm": 0.96484375,
"learning_rate": 0.0004986047941614821,
"loss": 5.561,
"mean_token_accuracy": 0.1642124354839325,
"num_tokens": 10130941.0,
"step": 4875
},
{
"entropy": 5.717188692092895,
"epoch": 0.4426705370101596,
"grad_norm": 1.03125,
"learning_rate": 0.0004986011941154696,
"loss": 5.5279,
"mean_token_accuracy": 0.1694575533270836,
"num_tokens": 10141708.0,
"step": 4880
},
{
"entropy": 5.697762584686279,
"epoch": 0.44312409288824384,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985975894453275,
"loss": 5.4623,
"mean_token_accuracy": 0.1702752724289894,
"num_tokens": 10151816.0,
"step": 4885
},
{
"entropy": 5.738364410400391,
"epoch": 0.443577648766328,
"grad_norm": 0.921875,
"learning_rate": 0.0004985939801511304,
"loss": 5.5813,
"mean_token_accuracy": 0.163410022854805,
"num_tokens": 10162470.0,
"step": 4890
},
{
"entropy": 5.693057489395142,
"epoch": 0.4440312046444122,
"grad_norm": 0.8671875,
"learning_rate": 0.0004985903662329527,
"loss": 5.4715,
"mean_token_accuracy": 0.17365510314702987,
"num_tokens": 10173171.0,
"step": 4895
},
{
"entropy": 5.6850179672241214,
"epoch": 0.44448476052249636,
"grad_norm": 0.87109375,
"learning_rate": 0.0004985867476908693,
"loss": 5.5027,
"mean_token_accuracy": 0.17133045196533203,
"num_tokens": 10184185.0,
"step": 4900
},
{
"entropy": 5.753661489486694,
"epoch": 0.44493831640058057,
"grad_norm": 0.921875,
"learning_rate": 0.0004985831245249549,
"loss": 5.5589,
"mean_token_accuracy": 0.17009343653917314,
"num_tokens": 10195516.0,
"step": 4905
},
{
"entropy": 5.651303815841675,
"epoch": 0.4453918722786647,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985794967352846,
"loss": 5.4905,
"mean_token_accuracy": 0.17255396246910096,
"num_tokens": 10204936.0,
"step": 4910
},
{
"entropy": 5.680718088150025,
"epoch": 0.44584542815674894,
"grad_norm": 0.8359375,
"learning_rate": 0.0004985758643219334,
"loss": 5.5193,
"mean_token_accuracy": 0.1686270222067833,
"num_tokens": 10216218.0,
"step": 4915
},
{
"entropy": 5.769064855575562,
"epoch": 0.4462989840348331,
"grad_norm": 0.96875,
"learning_rate": 0.0004985722272849762,
"loss": 5.5302,
"mean_token_accuracy": 0.1625892087817192,
"num_tokens": 10226370.0,
"step": 4920
},
{
"entropy": 5.608448934555054,
"epoch": 0.44675253991291725,
"grad_norm": 1.0078125,
"learning_rate": 0.0004985685856244884,
"loss": 5.4482,
"mean_token_accuracy": 0.17939605712890624,
"num_tokens": 10236170.0,
"step": 4925
},
{
"entropy": 5.718491077423096,
"epoch": 0.44720609579100146,
"grad_norm": 1.015625,
"learning_rate": 0.0004985649393405453,
"loss": 5.5439,
"mean_token_accuracy": 0.16827276796102525,
"num_tokens": 10246119.0,
"step": 4930
},
{
"entropy": 5.671393918991089,
"epoch": 0.4476596516690856,
"grad_norm": 0.859375,
"learning_rate": 0.0004985612884332223,
"loss": 5.5383,
"mean_token_accuracy": 0.16240749210119249,
"num_tokens": 10256762.0,
"step": 4935
},
{
"entropy": 5.633547115325928,
"epoch": 0.4481132075471698,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985576329025949,
"loss": 5.4727,
"mean_token_accuracy": 0.17245685309171677,
"num_tokens": 10267317.0,
"step": 4940
},
{
"entropy": 5.726488828659058,
"epoch": 0.448566763425254,
"grad_norm": 0.87109375,
"learning_rate": 0.0004985539727487385,
"loss": 5.6217,
"mean_token_accuracy": 0.15205846652388572,
"num_tokens": 10279111.0,
"step": 4945
},
{
"entropy": 5.708870220184326,
"epoch": 0.4490203193033382,
"grad_norm": 0.9375,
"learning_rate": 0.0004985503079717292,
"loss": 5.4791,
"mean_token_accuracy": 0.16987659484148027,
"num_tokens": 10289383.0,
"step": 4950
},
{
"entropy": 5.69928617477417,
"epoch": 0.44947387518142234,
"grad_norm": 0.921875,
"learning_rate": 0.0004985466385716425,
"loss": 5.49,
"mean_token_accuracy": 0.17275461703538894,
"num_tokens": 10299806.0,
"step": 4955
},
{
"entropy": 5.615327405929565,
"epoch": 0.44992743105950656,
"grad_norm": 0.87890625,
"learning_rate": 0.0004985429645485543,
"loss": 5.5166,
"mean_token_accuracy": 0.16595803648233415,
"num_tokens": 10310348.0,
"step": 4960
},
{
"entropy": 5.738679695129394,
"epoch": 0.4503809869375907,
"grad_norm": 0.9921875,
"learning_rate": 0.0004985392859025407,
"loss": 5.5192,
"mean_token_accuracy": 0.1689467638731003,
"num_tokens": 10320354.0,
"step": 4965
},
{
"entropy": 5.647083950042725,
"epoch": 0.45083454281567487,
"grad_norm": 0.8984375,
"learning_rate": 0.0004985356026336776,
"loss": 5.4311,
"mean_token_accuracy": 0.1705056294798851,
"num_tokens": 10330292.0,
"step": 4970
},
{
"entropy": 5.634524917602539,
"epoch": 0.4512880986937591,
"grad_norm": 0.87890625,
"learning_rate": 0.0004985319147420414,
"loss": 5.4075,
"mean_token_accuracy": 0.17841226756572723,
"num_tokens": 10340449.0,
"step": 4975
},
{
"entropy": 5.688153219223023,
"epoch": 0.45174165457184323,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985282222277082,
"loss": 5.4677,
"mean_token_accuracy": 0.1669263258576393,
"num_tokens": 10350637.0,
"step": 4980
},
{
"entropy": 5.597547960281372,
"epoch": 0.45219521044992744,
"grad_norm": 0.90625,
"learning_rate": 0.0004985245250907544,
"loss": 5.4141,
"mean_token_accuracy": 0.17463358342647553,
"num_tokens": 10361548.0,
"step": 4985
},
{
"entropy": 5.60842866897583,
"epoch": 0.4526487663280116,
"grad_norm": 0.92578125,
"learning_rate": 0.0004985208233312564,
"loss": 5.4248,
"mean_token_accuracy": 0.1648386925458908,
"num_tokens": 10371765.0,
"step": 4990
},
{
"entropy": 5.716899347305298,
"epoch": 0.4531023222060958,
"grad_norm": 0.89453125,
"learning_rate": 0.0004985171169492908,
"loss": 5.5395,
"mean_token_accuracy": 0.1704647123813629,
"num_tokens": 10382108.0,
"step": 4995
},
{
"entropy": 5.6541526317596436,
"epoch": 0.45355587808417996,
"grad_norm": 0.90625,
"learning_rate": 0.0004985134059449344,
"loss": 5.4909,
"mean_token_accuracy": 0.1751002311706543,
"num_tokens": 10392390.0,
"step": 5000
},
{
"entropy": 5.6331511497497555,
"epoch": 0.4540094339622642,
"grad_norm": 0.91015625,
"learning_rate": 0.0004985096903182637,
"loss": 5.5349,
"mean_token_accuracy": 0.1757347598671913,
"num_tokens": 10402094.0,
"step": 5005
},
{
"entropy": 5.689352416992188,
"epoch": 0.45446298984034833,
"grad_norm": 0.84765625,
"learning_rate": 0.0004985059700693557,
"loss": 5.4904,
"mean_token_accuracy": 0.17052320688962935,
"num_tokens": 10413350.0,
"step": 5010
},
{
"entropy": 5.726061534881592,
"epoch": 0.4549165457184325,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985022451982872,
"loss": 5.6072,
"mean_token_accuracy": 0.16370004415512085,
"num_tokens": 10424292.0,
"step": 5015
},
{
"entropy": 5.757494688034058,
"epoch": 0.4553701015965167,
"grad_norm": 0.97265625,
"learning_rate": 0.0004984985157051354,
"loss": 5.4711,
"mean_token_accuracy": 0.17044610232114793,
"num_tokens": 10433724.0,
"step": 5020
},
{
"entropy": 5.700105333328247,
"epoch": 0.45582365747460085,
"grad_norm": 0.90234375,
"learning_rate": 0.0004984947815899774,
"loss": 5.5438,
"mean_token_accuracy": 0.1677020400762558,
"num_tokens": 10444724.0,
"step": 5025
},
{
"entropy": 5.677012205123901,
"epoch": 0.45627721335268506,
"grad_norm": 0.9609375,
"learning_rate": 0.0004984910428528902,
"loss": 5.4799,
"mean_token_accuracy": 0.1747812494635582,
"num_tokens": 10455086.0,
"step": 5030
},
{
"entropy": 5.592632293701172,
"epoch": 0.4567307692307692,
"grad_norm": 0.875,
"learning_rate": 0.0004984872994939514,
"loss": 5.4252,
"mean_token_accuracy": 0.16925584375858307,
"num_tokens": 10465076.0,
"step": 5035
},
{
"entropy": 5.68324499130249,
"epoch": 0.45718432510885343,
"grad_norm": 0.9375,
"learning_rate": 0.0004984835515132382,
"loss": 5.4663,
"mean_token_accuracy": 0.17937684208154678,
"num_tokens": 10474817.0,
"step": 5040
},
{
"entropy": 5.662622547149658,
"epoch": 0.4576378809869376,
"grad_norm": 0.85546875,
"learning_rate": 0.0004984797989108282,
"loss": 5.4851,
"mean_token_accuracy": 0.16741085201501846,
"num_tokens": 10485735.0,
"step": 5045
},
{
"entropy": 5.636821985244751,
"epoch": 0.4580914368650218,
"grad_norm": 0.984375,
"learning_rate": 0.0004984760416867991,
"loss": 5.5723,
"mean_token_accuracy": 0.16558835059404373,
"num_tokens": 10496731.0,
"step": 5050
},
{
"entropy": 5.630299758911133,
"epoch": 0.45854499274310595,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984722798412286,
"loss": 5.4242,
"mean_token_accuracy": 0.1714159443974495,
"num_tokens": 10506787.0,
"step": 5055
},
{
"entropy": 5.7004170417785645,
"epoch": 0.4589985486211901,
"grad_norm": 0.953125,
"learning_rate": 0.0004984685133741941,
"loss": 5.5117,
"mean_token_accuracy": 0.16912039518356323,
"num_tokens": 10517170.0,
"step": 5060
},
{
"entropy": 5.6526471138000485,
"epoch": 0.4594521044992743,
"grad_norm": 0.95703125,
"learning_rate": 0.000498464742285774,
"loss": 5.4944,
"mean_token_accuracy": 0.16828947961330415,
"num_tokens": 10527342.0,
"step": 5065
},
{
"entropy": 5.692163467407227,
"epoch": 0.45990566037735847,
"grad_norm": 0.88671875,
"learning_rate": 0.000498460966576046,
"loss": 5.4743,
"mean_token_accuracy": 0.17233156114816667,
"num_tokens": 10537253.0,
"step": 5070
},
{
"entropy": 5.7053141593933105,
"epoch": 0.4603592162554427,
"grad_norm": 0.84765625,
"learning_rate": 0.0004984571862450884,
"loss": 5.4828,
"mean_token_accuracy": 0.1697017401456833,
"num_tokens": 10547249.0,
"step": 5075
},
{
"entropy": 5.753048896789551,
"epoch": 0.46081277213352684,
"grad_norm": 0.95703125,
"learning_rate": 0.0004984534012929791,
"loss": 5.5371,
"mean_token_accuracy": 0.17120666801929474,
"num_tokens": 10558174.0,
"step": 5080
},
{
"entropy": 5.728115892410278,
"epoch": 0.46126632801161105,
"grad_norm": 1.0,
"learning_rate": 0.0004984496117197967,
"loss": 5.4753,
"mean_token_accuracy": 0.17589109838008882,
"num_tokens": 10567610.0,
"step": 5085
},
{
"entropy": 5.601778030395508,
"epoch": 0.4617198838896952,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984458175256193,
"loss": 5.4797,
"mean_token_accuracy": 0.16605047583580018,
"num_tokens": 10578494.0,
"step": 5090
},
{
"entropy": 5.7371217727661135,
"epoch": 0.4621734397677794,
"grad_norm": 0.9140625,
"learning_rate": 0.0004984420187105253,
"loss": 5.586,
"mean_token_accuracy": 0.17094989866018295,
"num_tokens": 10589482.0,
"step": 5095
},
{
"entropy": 5.755274152755737,
"epoch": 0.46262699564586357,
"grad_norm": 0.890625,
"learning_rate": 0.0004984382152745936,
"loss": 5.5401,
"mean_token_accuracy": 0.16257912814617156,
"num_tokens": 10599965.0,
"step": 5100
},
{
"entropy": 5.66287579536438,
"epoch": 0.4630805515239477,
"grad_norm": 0.84375,
"learning_rate": 0.0004984344072179026,
"loss": 5.5603,
"mean_token_accuracy": 0.17168972343206407,
"num_tokens": 10610500.0,
"step": 5105
},
{
"entropy": 5.742537546157837,
"epoch": 0.46353410740203194,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984305945405312,
"loss": 5.5188,
"mean_token_accuracy": 0.17038158923387528,
"num_tokens": 10619843.0,
"step": 5110
},
{
"entropy": 5.658458423614502,
"epoch": 0.4639876632801161,
"grad_norm": 0.8984375,
"learning_rate": 0.000498426777242558,
"loss": 5.4436,
"mean_token_accuracy": 0.17783846333622932,
"num_tokens": 10631026.0,
"step": 5115
},
{
"entropy": 5.538843965530395,
"epoch": 0.4644412191582003,
"grad_norm": 0.953125,
"learning_rate": 0.0004984229553240623,
"loss": 5.4496,
"mean_token_accuracy": 0.17878477275371552,
"num_tokens": 10641130.0,
"step": 5120
},
{
"entropy": 5.746715879440307,
"epoch": 0.46489477503628446,
"grad_norm": 0.921875,
"learning_rate": 0.0004984191287851228,
"loss": 5.5067,
"mean_token_accuracy": 0.17264201045036315,
"num_tokens": 10652305.0,
"step": 5125
},
{
"entropy": 5.73653655052185,
"epoch": 0.46534833091436867,
"grad_norm": 0.8984375,
"learning_rate": 0.0004984152976258188,
"loss": 5.5573,
"mean_token_accuracy": 0.1695479616522789,
"num_tokens": 10663083.0,
"step": 5130
},
{
"entropy": 5.683012914657593,
"epoch": 0.4658018867924528,
"grad_norm": 0.890625,
"learning_rate": 0.0004984114618462296,
"loss": 5.572,
"mean_token_accuracy": 0.1649271011352539,
"num_tokens": 10673602.0,
"step": 5135
},
{
"entropy": 5.6376933574676515,
"epoch": 0.46625544267053703,
"grad_norm": 0.85546875,
"learning_rate": 0.0004984076214464343,
"loss": 5.392,
"mean_token_accuracy": 0.1789440020918846,
"num_tokens": 10684622.0,
"step": 5140
},
{
"entropy": 5.665349912643433,
"epoch": 0.4667089985486212,
"grad_norm": 0.9765625,
"learning_rate": 0.0004984037764265126,
"loss": 5.5034,
"mean_token_accuracy": 0.1705288678407669,
"num_tokens": 10695129.0,
"step": 5145
},
{
"entropy": 5.6029088497161865,
"epoch": 0.46716255442670535,
"grad_norm": 0.9375,
"learning_rate": 0.0004983999267865438,
"loss": 5.3525,
"mean_token_accuracy": 0.18145803660154342,
"num_tokens": 10705236.0,
"step": 5150
},
{
"entropy": 5.638808965682983,
"epoch": 0.46761611030478956,
"grad_norm": 1.0390625,
"learning_rate": 0.0004983960725266078,
"loss": 5.4762,
"mean_token_accuracy": 0.17050447314977646,
"num_tokens": 10715748.0,
"step": 5155
},
{
"entropy": 5.656491661071778,
"epoch": 0.4680696661828737,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983922136467838,
"loss": 5.3904,
"mean_token_accuracy": 0.17664045691490174,
"num_tokens": 10725799.0,
"step": 5160
},
{
"entropy": 5.659603071212769,
"epoch": 0.4685232220609579,
"grad_norm": 0.9453125,
"learning_rate": 0.000498388350147152,
"loss": 5.5361,
"mean_token_accuracy": 0.1669347897171974,
"num_tokens": 10735668.0,
"step": 5165
},
{
"entropy": 5.695495986938477,
"epoch": 0.4689767779390421,
"grad_norm": 0.8828125,
"learning_rate": 0.0004983844820277922,
"loss": 5.6331,
"mean_token_accuracy": 0.16362748593091964,
"num_tokens": 10746466.0,
"step": 5170
},
{
"entropy": 5.791701602935791,
"epoch": 0.4694303338171263,
"grad_norm": 0.91796875,
"learning_rate": 0.0004983806092887843,
"loss": 5.5238,
"mean_token_accuracy": 0.16722518354654312,
"num_tokens": 10756724.0,
"step": 5175
},
{
"entropy": 5.698781108856201,
"epoch": 0.46988388969521044,
"grad_norm": 0.9296875,
"learning_rate": 0.0004983767319302086,
"loss": 5.554,
"mean_token_accuracy": 0.17089424878358841,
"num_tokens": 10766919.0,
"step": 5180
},
{
"entropy": 5.584889268875122,
"epoch": 0.47033744557329465,
"grad_norm": 0.9453125,
"learning_rate": 0.0004983728499521451,
"loss": 5.3443,
"mean_token_accuracy": 0.18152898848056792,
"num_tokens": 10777550.0,
"step": 5185
},
{
"entropy": 5.667958736419678,
"epoch": 0.4707910014513788,
"grad_norm": 0.87109375,
"learning_rate": 0.0004983689633546741,
"loss": 5.4103,
"mean_token_accuracy": 0.16942705661058427,
"num_tokens": 10788456.0,
"step": 5190
},
{
"entropy": 5.64146933555603,
"epoch": 0.47124455732946297,
"grad_norm": 0.87109375,
"learning_rate": 0.0004983650721378761,
"loss": 5.449,
"mean_token_accuracy": 0.17425691485404968,
"num_tokens": 10800049.0,
"step": 5195
},
{
"entropy": 5.6370322704315186,
"epoch": 0.4716981132075472,
"grad_norm": 0.89453125,
"learning_rate": 0.0004983611763018314,
"loss": 5.5038,
"mean_token_accuracy": 0.16732366234064103,
"num_tokens": 10810743.0,
"step": 5200
},
{
"entropy": 5.801402473449707,
"epoch": 0.47215166908563133,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983572758466208,
"loss": 5.6878,
"mean_token_accuracy": 0.16397374272346496,
"num_tokens": 10821897.0,
"step": 5205
},
{
"entropy": 5.7822551250457765,
"epoch": 0.47260522496371554,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983533707723247,
"loss": 5.5451,
"mean_token_accuracy": 0.17132560312747955,
"num_tokens": 10831589.0,
"step": 5210
},
{
"entropy": 5.684686899185181,
"epoch": 0.4730587808417997,
"grad_norm": 0.9296875,
"learning_rate": 0.0004983494610790241,
"loss": 5.4469,
"mean_token_accuracy": 0.16281144618988036,
"num_tokens": 10841404.0,
"step": 5215
},
{
"entropy": 5.6485227108001705,
"epoch": 0.4735123367198839,
"grad_norm": 0.9140625,
"learning_rate": 0.0004983455467667996,
"loss": 5.5001,
"mean_token_accuracy": 0.16492253094911574,
"num_tokens": 10851577.0,
"step": 5220
},
{
"entropy": 5.6523786067962645,
"epoch": 0.47396589259796806,
"grad_norm": 0.97265625,
"learning_rate": 0.0004983416278357322,
"loss": 5.4073,
"mean_token_accuracy": 0.17646127343177795,
"num_tokens": 10861403.0,
"step": 5225
},
{
"entropy": 5.64488115310669,
"epoch": 0.4744194484760523,
"grad_norm": 0.93359375,
"learning_rate": 0.0004983377042859032,
"loss": 5.4546,
"mean_token_accuracy": 0.16768310219049454,
"num_tokens": 10871933.0,
"step": 5230
},
{
"entropy": 5.648560285568237,
"epoch": 0.47487300435413643,
"grad_norm": 0.96484375,
"learning_rate": 0.0004983337761173936,
"loss": 5.5315,
"mean_token_accuracy": 0.16874219328165055,
"num_tokens": 10881800.0,
"step": 5235
},
{
"entropy": 5.737661933898925,
"epoch": 0.4753265602322206,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983298433302843,
"loss": 5.4753,
"mean_token_accuracy": 0.1727318361401558,
"num_tokens": 10891830.0,
"step": 5240
},
{
"entropy": 5.6505979061126705,
"epoch": 0.4757801161103048,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983259059246572,
"loss": 5.403,
"mean_token_accuracy": 0.17735837250947953,
"num_tokens": 10901557.0,
"step": 5245
},
{
"entropy": 5.579076051712036,
"epoch": 0.47623367198838895,
"grad_norm": 0.88671875,
"learning_rate": 0.0004983219639005934,
"loss": 5.5252,
"mean_token_accuracy": 0.16280789524316788,
"num_tokens": 10912155.0,
"step": 5250
},
{
"entropy": 5.642884588241577,
"epoch": 0.47668722786647316,
"grad_norm": 1.0078125,
"learning_rate": 0.0004983180172581745,
"loss": 5.4296,
"mean_token_accuracy": 0.17732747346162797,
"num_tokens": 10922291.0,
"step": 5255
},
{
"entropy": 5.7082366943359375,
"epoch": 0.4771407837445573,
"grad_norm": 0.875,
"learning_rate": 0.0004983140659974819,
"loss": 5.5024,
"mean_token_accuracy": 0.1679125025868416,
"num_tokens": 10933176.0,
"step": 5260
},
{
"entropy": 5.698912668228149,
"epoch": 0.47759433962264153,
"grad_norm": 0.9296875,
"learning_rate": 0.0004983101101185977,
"loss": 5.5828,
"mean_token_accuracy": 0.16323981285095215,
"num_tokens": 10944049.0,
"step": 5265
},
{
"entropy": 5.633695650100708,
"epoch": 0.4780478955007257,
"grad_norm": 0.91015625,
"learning_rate": 0.0004983061496216035,
"loss": 5.4238,
"mean_token_accuracy": 0.17720418572425842,
"num_tokens": 10954347.0,
"step": 5270
},
{
"entropy": 5.6351593971252445,
"epoch": 0.4785014513788099,
"grad_norm": 0.921875,
"learning_rate": 0.0004983021845065812,
"loss": 5.4756,
"mean_token_accuracy": 0.175036196410656,
"num_tokens": 10965123.0,
"step": 5275
},
{
"entropy": 5.597521591186523,
"epoch": 0.47895500725689405,
"grad_norm": 0.98046875,
"learning_rate": 0.0004982982147736128,
"loss": 5.4642,
"mean_token_accuracy": 0.17434579282999038,
"num_tokens": 10976395.0,
"step": 5280
},
{
"entropy": 5.691222858428955,
"epoch": 0.4794085631349782,
"grad_norm": 0.92578125,
"learning_rate": 0.0004982942404227805,
"loss": 5.4586,
"mean_token_accuracy": 0.16569167971611024,
"num_tokens": 10986797.0,
"step": 5285
},
{
"entropy": 5.708698081970215,
"epoch": 0.4798621190130624,
"grad_norm": 0.8671875,
"learning_rate": 0.0004982902614541662,
"loss": 5.5117,
"mean_token_accuracy": 0.17047134935855865,
"num_tokens": 10997229.0,
"step": 5290
},
{
"entropy": 5.614826726913452,
"epoch": 0.48031567489114657,
"grad_norm": 0.87109375,
"learning_rate": 0.0004982862778678526,
"loss": 5.4412,
"mean_token_accuracy": 0.17899638414382935,
"num_tokens": 11008180.0,
"step": 5295
},
{
"entropy": 5.683491897583008,
"epoch": 0.4807692307692308,
"grad_norm": 0.890625,
"learning_rate": 0.0004982822896639218,
"loss": 5.4705,
"mean_token_accuracy": 0.17429982572793962,
"num_tokens": 11019248.0,
"step": 5300
},
{
"entropy": 5.617243051528931,
"epoch": 0.48122278664731494,
"grad_norm": 0.87109375,
"learning_rate": 0.0004982782968424563,
"loss": 5.3761,
"mean_token_accuracy": 0.1775210216641426,
"num_tokens": 11029760.0,
"step": 5305
},
{
"entropy": 5.588413190841675,
"epoch": 0.48167634252539915,
"grad_norm": 1.0,
"learning_rate": 0.0004982742994035388,
"loss": 5.4821,
"mean_token_accuracy": 0.17242670059204102,
"num_tokens": 11039211.0,
"step": 5310
},
{
"entropy": 5.662622356414795,
"epoch": 0.4821298984034833,
"grad_norm": 0.94140625,
"learning_rate": 0.0004982702973472518,
"loss": 5.454,
"mean_token_accuracy": 0.17342988401651382,
"num_tokens": 11049892.0,
"step": 5315
},
{
"entropy": 5.746756601333618,
"epoch": 0.4825834542815675,
"grad_norm": 0.890625,
"learning_rate": 0.0004982662906736783,
"loss": 5.5041,
"mean_token_accuracy": 0.17146852016448974,
"num_tokens": 11060823.0,
"step": 5320
},
{
"entropy": 5.609480905532837,
"epoch": 0.48303701015965167,
"grad_norm": 0.83984375,
"learning_rate": 0.0004982622793829009,
"loss": 5.4768,
"mean_token_accuracy": 0.17211514413356782,
"num_tokens": 11072353.0,
"step": 5325
},
{
"entropy": 5.6528795719146725,
"epoch": 0.4834905660377358,
"grad_norm": 0.95703125,
"learning_rate": 0.0004982582634750027,
"loss": 5.491,
"mean_token_accuracy": 0.1721246212720871,
"num_tokens": 11082655.0,
"step": 5330
},
{
"entropy": 5.676597452163696,
"epoch": 0.48394412191582004,
"grad_norm": 0.8828125,
"learning_rate": 0.0004982542429500667,
"loss": 5.5053,
"mean_token_accuracy": 0.17382133454084397,
"num_tokens": 11093315.0,
"step": 5335
},
{
"entropy": 5.524261379241944,
"epoch": 0.4843976777939042,
"grad_norm": 0.9453125,
"learning_rate": 0.0004982502178081761,
"loss": 5.4042,
"mean_token_accuracy": 0.17310083359479905,
"num_tokens": 11103074.0,
"step": 5340
},
{
"entropy": 5.666402530670166,
"epoch": 0.4848512336719884,
"grad_norm": 0.93359375,
"learning_rate": 0.0004982461880494141,
"loss": 5.4279,
"mean_token_accuracy": 0.17441916912794114,
"num_tokens": 11113636.0,
"step": 5345
},
{
"entropy": 5.656398725509644,
"epoch": 0.48530478955007256,
"grad_norm": 0.92578125,
"learning_rate": 0.000498242153673864,
"loss": 5.4039,
"mean_token_accuracy": 0.17608933597803117,
"num_tokens": 11123509.0,
"step": 5350
},
{
"entropy": 5.613732099533081,
"epoch": 0.48575834542815677,
"grad_norm": 0.921875,
"learning_rate": 0.0004982381146816094,
"loss": 5.4686,
"mean_token_accuracy": 0.16522315591573716,
"num_tokens": 11133405.0,
"step": 5355
},
{
"entropy": 5.73980131149292,
"epoch": 0.4862119013062409,
"grad_norm": 0.84375,
"learning_rate": 0.0004982340710727336,
"loss": 5.533,
"mean_token_accuracy": 0.16312984079122544,
"num_tokens": 11144725.0,
"step": 5360
},
{
"entropy": 5.67049765586853,
"epoch": 0.48666545718432513,
"grad_norm": 0.9609375,
"learning_rate": 0.0004982300228473203,
"loss": 5.4184,
"mean_token_accuracy": 0.16829311102628708,
"num_tokens": 11155446.0,
"step": 5365
},
{
"entropy": 5.636474370956421,
"epoch": 0.4871190130624093,
"grad_norm": 0.9296875,
"learning_rate": 0.0004982259700054533,
"loss": 5.4818,
"mean_token_accuracy": 0.17630883306264877,
"num_tokens": 11165380.0,
"step": 5370
},
{
"entropy": 5.647053050994873,
"epoch": 0.48757256894049344,
"grad_norm": 0.890625,
"learning_rate": 0.0004982219125472163,
"loss": 5.5238,
"mean_token_accuracy": 0.1691724866628647,
"num_tokens": 11175641.0,
"step": 5375
},
{
"entropy": 5.799345302581787,
"epoch": 0.48802612481857766,
"grad_norm": 0.921875,
"learning_rate": 0.0004982178504726933,
"loss": 5.6734,
"mean_token_accuracy": 0.17320263534784316,
"num_tokens": 11186257.0,
"step": 5380
},
{
"entropy": 5.727895784378052,
"epoch": 0.4884796806966618,
"grad_norm": 0.86328125,
"learning_rate": 0.0004982137837819682,
"loss": 5.4966,
"mean_token_accuracy": 0.16289989352226258,
"num_tokens": 11197284.0,
"step": 5385
},
{
"entropy": 5.688380861282349,
"epoch": 0.488933236574746,
"grad_norm": 0.91796875,
"learning_rate": 0.0004982097124751252,
"loss": 5.4741,
"mean_token_accuracy": 0.17410095632076264,
"num_tokens": 11208413.0,
"step": 5390
},
{
"entropy": 5.629041337966919,
"epoch": 0.4893867924528302,
"grad_norm": 0.89453125,
"learning_rate": 0.0004982056365522486,
"loss": 5.4667,
"mean_token_accuracy": 0.17381504625082017,
"num_tokens": 11219016.0,
"step": 5395
},
{
"entropy": 5.625274276733398,
"epoch": 0.4898403483309144,
"grad_norm": 0.890625,
"learning_rate": 0.0004982015560134225,
"loss": 5.4451,
"mean_token_accuracy": 0.1747204378247261,
"num_tokens": 11228934.0,
"step": 5400
},
{
"entropy": 5.639862775802612,
"epoch": 0.49029390420899854,
"grad_norm": 1.0625,
"learning_rate": 0.0004981974708587312,
"loss": 5.3724,
"mean_token_accuracy": 0.17818964570760726,
"num_tokens": 11239596.0,
"step": 5405
},
{
"entropy": 5.603957653045654,
"epoch": 0.49074746008708275,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981933810882594,
"loss": 5.3515,
"mean_token_accuracy": 0.17084526121616364,
"num_tokens": 11249560.0,
"step": 5410
},
{
"entropy": 5.707994842529297,
"epoch": 0.4912010159651669,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981892867020917,
"loss": 5.5617,
"mean_token_accuracy": 0.17124133110046386,
"num_tokens": 11259787.0,
"step": 5415
},
{
"entropy": 5.722805786132812,
"epoch": 0.49165457184325106,
"grad_norm": 0.96875,
"learning_rate": 0.0004981851877003125,
"loss": 5.5011,
"mean_token_accuracy": 0.1731458768248558,
"num_tokens": 11269717.0,
"step": 5420
},
{
"entropy": 5.625773620605469,
"epoch": 0.4921081277213353,
"grad_norm": 1.015625,
"learning_rate": 0.0004981810840830068,
"loss": 5.474,
"mean_token_accuracy": 0.18215908706188202,
"num_tokens": 11279863.0,
"step": 5425
},
{
"entropy": 5.666403102874756,
"epoch": 0.49256168359941943,
"grad_norm": 0.94140625,
"learning_rate": 0.0004981769758502596,
"loss": 5.5915,
"mean_token_accuracy": 0.1605565443634987,
"num_tokens": 11291161.0,
"step": 5430
},
{
"entropy": 5.609889507293701,
"epoch": 0.49301523947750364,
"grad_norm": 0.95703125,
"learning_rate": 0.0004981728630021555,
"loss": 5.4073,
"mean_token_accuracy": 0.17431867867708206,
"num_tokens": 11301151.0,
"step": 5435
},
{
"entropy": 5.583493232727051,
"epoch": 0.4934687953555878,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981687455387797,
"loss": 5.4231,
"mean_token_accuracy": 0.1704528123140335,
"num_tokens": 11311466.0,
"step": 5440
},
{
"entropy": 5.651216983795166,
"epoch": 0.493922351233672,
"grad_norm": 0.921875,
"learning_rate": 0.0004981646234602173,
"loss": 5.3677,
"mean_token_accuracy": 0.17543147057294844,
"num_tokens": 11321522.0,
"step": 5445
},
{
"entropy": 5.70766978263855,
"epoch": 0.49437590711175616,
"grad_norm": 0.8671875,
"learning_rate": 0.0004981604967665539,
"loss": 5.4904,
"mean_token_accuracy": 0.17429745495319365,
"num_tokens": 11332317.0,
"step": 5450
},
{
"entropy": 5.638960981369019,
"epoch": 0.4948294629898404,
"grad_norm": 0.91015625,
"learning_rate": 0.0004981563654578743,
"loss": 5.4383,
"mean_token_accuracy": 0.1693056493997574,
"num_tokens": 11343024.0,
"step": 5455
},
{
"entropy": 5.579675197601318,
"epoch": 0.49528301886792453,
"grad_norm": 0.96484375,
"learning_rate": 0.0004981522295342643,
"loss": 5.392,
"mean_token_accuracy": 0.17219439148902893,
"num_tokens": 11353114.0,
"step": 5460
},
{
"entropy": 5.644372177124024,
"epoch": 0.4957365747460087,
"grad_norm": 0.87109375,
"learning_rate": 0.0004981480889958092,
"loss": 5.4615,
"mean_token_accuracy": 0.17464147955179216,
"num_tokens": 11363446.0,
"step": 5465
},
{
"entropy": 5.612289381027222,
"epoch": 0.4961901306240929,
"grad_norm": 0.90625,
"learning_rate": 0.0004981439438425949,
"loss": 5.4536,
"mean_token_accuracy": 0.1771087482571602,
"num_tokens": 11373413.0,
"step": 5470
},
{
"entropy": 5.613510656356811,
"epoch": 0.49664368650217705,
"grad_norm": 0.9921875,
"learning_rate": 0.000498139794074707,
"loss": 5.3264,
"mean_token_accuracy": 0.17951995581388475,
"num_tokens": 11383142.0,
"step": 5475
},
{
"entropy": 5.666054725646973,
"epoch": 0.49709724238026126,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981356396922313,
"loss": 5.4847,
"mean_token_accuracy": 0.17279811352491378,
"num_tokens": 11394635.0,
"step": 5480
},
{
"entropy": 5.604884958267212,
"epoch": 0.4975507982583454,
"grad_norm": 1.3203125,
"learning_rate": 0.0004981314806952535,
"loss": 5.3892,
"mean_token_accuracy": 0.17838599234819413,
"num_tokens": 11404442.0,
"step": 5485
},
{
"entropy": 5.557573413848877,
"epoch": 0.49800435413642963,
"grad_norm": 0.96875,
"learning_rate": 0.00049812731708386,
"loss": 5.3433,
"mean_token_accuracy": 0.17274410128593445,
"num_tokens": 11414217.0,
"step": 5490
},
{
"entropy": 5.678751802444458,
"epoch": 0.4984579100145138,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981231488581367,
"loss": 5.4886,
"mean_token_accuracy": 0.16931686848402022,
"num_tokens": 11423834.0,
"step": 5495
},
{
"entropy": 5.5513814926147464,
"epoch": 0.498911465892598,
"grad_norm": 0.9140625,
"learning_rate": 0.0004981189760181699,
"loss": 5.358,
"mean_token_accuracy": 0.17370127141475677,
"num_tokens": 11433613.0,
"step": 5500
},
{
"entropy": 5.579940223693848,
"epoch": 0.49936502177068215,
"grad_norm": 0.96875,
"learning_rate": 0.0004981147985640456,
"loss": 5.4257,
"mean_token_accuracy": 0.17723614424467088,
"num_tokens": 11443915.0,
"step": 5505
},
{
"entropy": 5.704259824752808,
"epoch": 0.4998185776487663,
"grad_norm": 0.890625,
"learning_rate": 0.0004981106164958506,
"loss": 5.4561,
"mean_token_accuracy": 0.1709003448486328,
"num_tokens": 11454588.0,
"step": 5510
},
{
"entropy": 5.648460054397583,
"epoch": 0.5002721335268505,
"grad_norm": 0.92578125,
"learning_rate": 0.0004981064298136712,
"loss": 5.5067,
"mean_token_accuracy": 0.16957272291183473,
"num_tokens": 11465533.0,
"step": 5515
},
{
"entropy": 5.622145414352417,
"epoch": 0.5007256894049347,
"grad_norm": 0.94921875,
"learning_rate": 0.0004981022385175939,
"loss": 5.4586,
"mean_token_accuracy": 0.16735394597053527,
"num_tokens": 11475776.0,
"step": 5520
},
{
"entropy": 5.620416879653931,
"epoch": 0.5011792452830188,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980980426077054,
"loss": 5.4385,
"mean_token_accuracy": 0.18021399825811385,
"num_tokens": 11486200.0,
"step": 5525
},
{
"entropy": 5.633817100524903,
"epoch": 0.5016328011611031,
"grad_norm": 0.91015625,
"learning_rate": 0.0004980938420840926,
"loss": 5.4609,
"mean_token_accuracy": 0.16805864125490189,
"num_tokens": 11496768.0,
"step": 5530
},
{
"entropy": 5.64311728477478,
"epoch": 0.5020863570391872,
"grad_norm": 0.93359375,
"learning_rate": 0.0004980896369468423,
"loss": 5.438,
"mean_token_accuracy": 0.17364895343780518,
"num_tokens": 11507096.0,
"step": 5535
},
{
"entropy": 5.617955064773559,
"epoch": 0.5025399129172714,
"grad_norm": 0.90234375,
"learning_rate": 0.0004980854271960415,
"loss": 5.4858,
"mean_token_accuracy": 0.17146076411008834,
"num_tokens": 11517301.0,
"step": 5540
},
{
"entropy": 5.691181278228759,
"epoch": 0.5029934687953556,
"grad_norm": 0.89453125,
"learning_rate": 0.0004980812128317772,
"loss": 5.5381,
"mean_token_accuracy": 0.17311448901891707,
"num_tokens": 11528616.0,
"step": 5545
},
{
"entropy": 5.710403871536255,
"epoch": 0.5034470246734397,
"grad_norm": 0.8671875,
"learning_rate": 0.0004980769938541364,
"loss": 5.4596,
"mean_token_accuracy": 0.17918587774038314,
"num_tokens": 11539344.0,
"step": 5550
},
{
"entropy": 5.611640930175781,
"epoch": 0.503900580551524,
"grad_norm": 0.921875,
"learning_rate": 0.0004980727702632066,
"loss": 5.5158,
"mean_token_accuracy": 0.1702592894434929,
"num_tokens": 11549511.0,
"step": 5555
},
{
"entropy": 5.654125452041626,
"epoch": 0.5043541364296081,
"grad_norm": 0.9375,
"learning_rate": 0.0004980685420590752,
"loss": 5.4425,
"mean_token_accuracy": 0.17124981731176375,
"num_tokens": 11559425.0,
"step": 5560
},
{
"entropy": 5.7445450782775875,
"epoch": 0.5048076923076923,
"grad_norm": 0.90234375,
"learning_rate": 0.0004980643092418294,
"loss": 5.564,
"mean_token_accuracy": 0.16161303967237473,
"num_tokens": 11570406.0,
"step": 5565
},
{
"entropy": 5.652337265014649,
"epoch": 0.5052612481857764,
"grad_norm": 0.890625,
"learning_rate": 0.0004980600718115568,
"loss": 5.4435,
"mean_token_accuracy": 0.17120484560728072,
"num_tokens": 11580509.0,
"step": 5570
},
{
"entropy": 5.657799530029297,
"epoch": 0.5057148040638607,
"grad_norm": 0.87890625,
"learning_rate": 0.0004980558297683452,
"loss": 5.4658,
"mean_token_accuracy": 0.17254962921142578,
"num_tokens": 11590860.0,
"step": 5575
},
{
"entropy": 5.69189043045044,
"epoch": 0.5061683599419449,
"grad_norm": 0.96875,
"learning_rate": 0.0004980515831122822,
"loss": 5.4658,
"mean_token_accuracy": 0.16972111016511918,
"num_tokens": 11601954.0,
"step": 5580
},
{
"entropy": 5.641917419433594,
"epoch": 0.506621915820029,
"grad_norm": 0.83203125,
"learning_rate": 0.0004980473318434555,
"loss": 5.4643,
"mean_token_accuracy": 0.17524617165327072,
"num_tokens": 11612581.0,
"step": 5585
},
{
"entropy": 5.54950852394104,
"epoch": 0.5070754716981132,
"grad_norm": 1.1171875,
"learning_rate": 0.0004980430759619532,
"loss": 5.3893,
"mean_token_accuracy": 0.17424534559249877,
"num_tokens": 11622599.0,
"step": 5590
},
{
"entropy": 5.572847175598144,
"epoch": 0.5075290275761973,
"grad_norm": 0.984375,
"learning_rate": 0.0004980388154678633,
"loss": 5.4041,
"mean_token_accuracy": 0.1734546795487404,
"num_tokens": 11633611.0,
"step": 5595
},
{
"entropy": 5.59787654876709,
"epoch": 0.5079825834542816,
"grad_norm": 0.98828125,
"learning_rate": 0.0004980345503612738,
"loss": 5.3836,
"mean_token_accuracy": 0.1732645958662033,
"num_tokens": 11643159.0,
"step": 5600
},
{
"entropy": 5.615271997451782,
"epoch": 0.5084361393323658,
"grad_norm": 0.91796875,
"learning_rate": 0.0004980302806422729,
"loss": 5.3753,
"mean_token_accuracy": 0.1775027260184288,
"num_tokens": 11653174.0,
"step": 5605
},
{
"entropy": 5.674450540542603,
"epoch": 0.5088896952104499,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980260063109491,
"loss": 5.4227,
"mean_token_accuracy": 0.17737311273813247,
"num_tokens": 11662621.0,
"step": 5610
},
{
"entropy": 5.543775224685669,
"epoch": 0.5093432510885341,
"grad_norm": 0.83203125,
"learning_rate": 0.0004980217273673906,
"loss": 5.3793,
"mean_token_accuracy": 0.17156035602092742,
"num_tokens": 11673538.0,
"step": 5615
},
{
"entropy": 5.653875684738159,
"epoch": 0.5097968069666183,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980174438116859,
"loss": 5.3914,
"mean_token_accuracy": 0.17867197394371032,
"num_tokens": 11683203.0,
"step": 5620
},
{
"entropy": 5.552722072601318,
"epoch": 0.5102503628447025,
"grad_norm": 0.96484375,
"learning_rate": 0.0004980131556439236,
"loss": 5.3829,
"mean_token_accuracy": 0.180565182864666,
"num_tokens": 11693313.0,
"step": 5625
},
{
"entropy": 5.640235567092896,
"epoch": 0.5107039187227866,
"grad_norm": 0.8828125,
"learning_rate": 0.0004980088628641925,
"loss": 5.4783,
"mean_token_accuracy": 0.17274700552225114,
"num_tokens": 11704194.0,
"step": 5630
},
{
"entropy": 5.642412567138672,
"epoch": 0.5111574746008708,
"grad_norm": 0.84375,
"learning_rate": 0.0004980045654725812,
"loss": 5.336,
"mean_token_accuracy": 0.17914541512727739,
"num_tokens": 11715109.0,
"step": 5635
},
{
"entropy": 5.5468549728393555,
"epoch": 0.511611030478955,
"grad_norm": 0.96484375,
"learning_rate": 0.0004980002634691787,
"loss": 5.3796,
"mean_token_accuracy": 0.17843705415725708,
"num_tokens": 11724961.0,
"step": 5640
},
{
"entropy": 5.552511072158813,
"epoch": 0.5120645863570392,
"grad_norm": 0.9765625,
"learning_rate": 0.000497995956854074,
"loss": 5.3914,
"mean_token_accuracy": 0.17831180095672608,
"num_tokens": 11735334.0,
"step": 5645
},
{
"entropy": 5.550053977966309,
"epoch": 0.5125181422351234,
"grad_norm": 0.9140625,
"learning_rate": 0.0004979916456273561,
"loss": 5.3935,
"mean_token_accuracy": 0.1712997004389763,
"num_tokens": 11746858.0,
"step": 5650
},
{
"entropy": 5.689655733108521,
"epoch": 0.5129716981132075,
"grad_norm": 0.89453125,
"learning_rate": 0.0004979873297891141,
"loss": 5.4965,
"mean_token_accuracy": 0.16972004026174545,
"num_tokens": 11757819.0,
"step": 5655
},
{
"entropy": 5.650430011749267,
"epoch": 0.5134252539912917,
"grad_norm": 1.0546875,
"learning_rate": 0.0004979830093394372,
"loss": 5.4349,
"mean_token_accuracy": 0.17406515330076217,
"num_tokens": 11767470.0,
"step": 5660
},
{
"entropy": 5.574281358718872,
"epoch": 0.513878809869376,
"grad_norm": 0.9140625,
"learning_rate": 0.0004979786842784149,
"loss": 5.3549,
"mean_token_accuracy": 0.1825924515724182,
"num_tokens": 11777060.0,
"step": 5665
},
{
"entropy": 5.566286706924439,
"epoch": 0.5143323657474601,
"grad_norm": 0.921875,
"learning_rate": 0.0004979743546061366,
"loss": 5.4107,
"mean_token_accuracy": 0.17651544213294984,
"num_tokens": 11787176.0,
"step": 5670
},
{
"entropy": 5.5616296291351315,
"epoch": 0.5147859216255443,
"grad_norm": 1.015625,
"learning_rate": 0.0004979700203226918,
"loss": 5.2849,
"mean_token_accuracy": 0.17933404445648193,
"num_tokens": 11796419.0,
"step": 5675
},
{
"entropy": 5.682236576080323,
"epoch": 0.5152394775036284,
"grad_norm": 0.9140625,
"learning_rate": 0.0004979656814281702,
"loss": 5.5213,
"mean_token_accuracy": 0.1706225335597992,
"num_tokens": 11807279.0,
"step": 5680
},
{
"entropy": 5.606097984313965,
"epoch": 0.5156930333817126,
"grad_norm": 1.015625,
"learning_rate": 0.0004979613379226615,
"loss": 5.4088,
"mean_token_accuracy": 0.1719408765435219,
"num_tokens": 11817669.0,
"step": 5685
},
{
"entropy": 5.544135332107544,
"epoch": 0.5161465892597968,
"grad_norm": 0.9375,
"learning_rate": 0.0004979569898062554,
"loss": 5.415,
"mean_token_accuracy": 0.179472017288208,
"num_tokens": 11827247.0,
"step": 5690
},
{
"entropy": 5.619077444076538,
"epoch": 0.516600145137881,
"grad_norm": 0.9609375,
"learning_rate": 0.000497952637079042,
"loss": 5.402,
"mean_token_accuracy": 0.17077631801366805,
"num_tokens": 11838201.0,
"step": 5695
},
{
"entropy": 5.608509302139282,
"epoch": 0.5170537010159652,
"grad_norm": 0.9375,
"learning_rate": 0.0004979482797411114,
"loss": 5.3538,
"mean_token_accuracy": 0.18213389366865157,
"num_tokens": 11847561.0,
"step": 5700
},
{
"entropy": 5.63592209815979,
"epoch": 0.5175072568940493,
"grad_norm": 0.96875,
"learning_rate": 0.0004979439177925534,
"loss": 5.4601,
"mean_token_accuracy": 0.17978523373603822,
"num_tokens": 11858272.0,
"step": 5705
},
{
"entropy": 5.664514636993408,
"epoch": 0.5179608127721336,
"grad_norm": 0.88671875,
"learning_rate": 0.0004979395512334584,
"loss": 5.4897,
"mean_token_accuracy": 0.1710379973053932,
"num_tokens": 11869660.0,
"step": 5710
},
{
"entropy": 5.657863283157349,
"epoch": 0.5184143686502177,
"grad_norm": 0.87109375,
"learning_rate": 0.0004979351800639166,
"loss": 5.522,
"mean_token_accuracy": 0.16624149084091186,
"num_tokens": 11880239.0,
"step": 5715
},
{
"entropy": 5.723998975753784,
"epoch": 0.5188679245283019,
"grad_norm": 0.8984375,
"learning_rate": 0.0004979308042840186,
"loss": 5.5019,
"mean_token_accuracy": 0.16436144411563874,
"num_tokens": 11890630.0,
"step": 5720
},
{
"entropy": 5.692220687866211,
"epoch": 0.519321480406386,
"grad_norm": 0.88671875,
"learning_rate": 0.0004979264238938547,
"loss": 5.5387,
"mean_token_accuracy": 0.16605993658304213,
"num_tokens": 11900948.0,
"step": 5725
},
{
"entropy": 5.5566572666168215,
"epoch": 0.5197750362844702,
"grad_norm": 0.89453125,
"learning_rate": 0.0004979220388935155,
"loss": 5.4197,
"mean_token_accuracy": 0.1711593523621559,
"num_tokens": 11912528.0,
"step": 5730
},
{
"entropy": 5.620328950881958,
"epoch": 0.5202285921625545,
"grad_norm": 0.91796875,
"learning_rate": 0.0004979176492830918,
"loss": 5.36,
"mean_token_accuracy": 0.1769677922129631,
"num_tokens": 11922211.0,
"step": 5735
},
{
"entropy": 5.6201256275177,
"epoch": 0.5206821480406386,
"grad_norm": 0.9609375,
"learning_rate": 0.0004979132550626742,
"loss": 5.4102,
"mean_token_accuracy": 0.1742950886487961,
"num_tokens": 11932520.0,
"step": 5740
},
{
"entropy": 5.587960624694825,
"epoch": 0.5211357039187228,
"grad_norm": 0.921875,
"learning_rate": 0.0004979088562323538,
"loss": 5.3434,
"mean_token_accuracy": 0.17956214398145676,
"num_tokens": 11943312.0,
"step": 5745
},
{
"entropy": 5.564781379699707,
"epoch": 0.5215892597968069,
"grad_norm": 1.015625,
"learning_rate": 0.0004979044527922214,
"loss": 5.4926,
"mean_token_accuracy": 0.1721508026123047,
"num_tokens": 11954455.0,
"step": 5750
},
{
"entropy": 5.6627356052398685,
"epoch": 0.5220428156748912,
"grad_norm": 0.9453125,
"learning_rate": 0.0004979000447423681,
"loss": 5.3863,
"mean_token_accuracy": 0.17926096618175508,
"num_tokens": 11964650.0,
"step": 5755
},
{
"entropy": 5.575634002685547,
"epoch": 0.5224963715529753,
"grad_norm": 0.87890625,
"learning_rate": 0.0004978956320828851,
"loss": 5.4195,
"mean_token_accuracy": 0.1762705847620964,
"num_tokens": 11975134.0,
"step": 5760
},
{
"entropy": 5.606097459793091,
"epoch": 0.5229499274310595,
"grad_norm": 0.90234375,
"learning_rate": 0.0004978912148138637,
"loss": 5.4532,
"mean_token_accuracy": 0.17324555963277816,
"num_tokens": 11986486.0,
"step": 5765
},
{
"entropy": 5.678210592269897,
"epoch": 0.5234034833091437,
"grad_norm": 0.98046875,
"learning_rate": 0.000497886792935395,
"loss": 5.4704,
"mean_token_accuracy": 0.1706312969326973,
"num_tokens": 11998045.0,
"step": 5770
},
{
"entropy": 5.717112493515015,
"epoch": 0.5238570391872278,
"grad_norm": 0.875,
"learning_rate": 0.0004978823664475707,
"loss": 5.4668,
"mean_token_accuracy": 0.17400625944137574,
"num_tokens": 12008130.0,
"step": 5775
},
{
"entropy": 5.6386171817779545,
"epoch": 0.5243105950653121,
"grad_norm": 0.94921875,
"learning_rate": 0.0004978779353504824,
"loss": 5.5292,
"mean_token_accuracy": 0.1656391978263855,
"num_tokens": 12018753.0,
"step": 5780
},
{
"entropy": 5.631770372390747,
"epoch": 0.5247641509433962,
"grad_norm": 1.0,
"learning_rate": 0.0004978734996442214,
"loss": 5.4048,
"mean_token_accuracy": 0.17983568012714385,
"num_tokens": 12028202.0,
"step": 5785
},
{
"entropy": 5.612694168090821,
"epoch": 0.5252177068214804,
"grad_norm": 0.90625,
"learning_rate": 0.0004978690593288798,
"loss": 5.4843,
"mean_token_accuracy": 0.17659876495599747,
"num_tokens": 12039652.0,
"step": 5790
},
{
"entropy": 5.678992748260498,
"epoch": 0.5256712626995645,
"grad_norm": 1.0234375,
"learning_rate": 0.0004978646144045491,
"loss": 5.3938,
"mean_token_accuracy": 0.16736219227313995,
"num_tokens": 12051269.0,
"step": 5795
},
{
"entropy": 5.46131854057312,
"epoch": 0.5261248185776488,
"grad_norm": 0.93359375,
"learning_rate": 0.0004978601648713215,
"loss": 5.291,
"mean_token_accuracy": 0.17717837989330293,
"num_tokens": 12060647.0,
"step": 5800
},
{
"entropy": 5.585661935806274,
"epoch": 0.526578374455733,
"grad_norm": 0.9140625,
"learning_rate": 0.0004978557107292889,
"loss": 5.4256,
"mean_token_accuracy": 0.17812251448631286,
"num_tokens": 12071669.0,
"step": 5805
},
{
"entropy": 5.561780834197998,
"epoch": 0.5270319303338171,
"grad_norm": 0.96484375,
"learning_rate": 0.0004978512519785433,
"loss": 5.3141,
"mean_token_accuracy": 0.1775515452027321,
"num_tokens": 12081971.0,
"step": 5810
},
{
"entropy": 5.567921876907349,
"epoch": 0.5274854862119013,
"grad_norm": 0.875,
"learning_rate": 0.000497846788619177,
"loss": 5.4346,
"mean_token_accuracy": 0.18071742057800294,
"num_tokens": 12092465.0,
"step": 5815
},
{
"entropy": 5.65427074432373,
"epoch": 0.5279390420899854,
"grad_norm": 0.89453125,
"learning_rate": 0.0004978423206512825,
"loss": 5.3935,
"mean_token_accuracy": 0.17432788610458375,
"num_tokens": 12101752.0,
"step": 5820
},
{
"entropy": 5.673962211608886,
"epoch": 0.5283925979680697,
"grad_norm": 1.03125,
"learning_rate": 0.000497837848074952,
"loss": 5.4643,
"mean_token_accuracy": 0.17431971430778503,
"num_tokens": 12111805.0,
"step": 5825
},
{
"entropy": 5.694564437866211,
"epoch": 0.5288461538461539,
"grad_norm": 0.9140625,
"learning_rate": 0.000497833370890278,
"loss": 5.4918,
"mean_token_accuracy": 0.17020126730203627,
"num_tokens": 12122511.0,
"step": 5830
},
{
"entropy": 5.607981729507446,
"epoch": 0.529299709724238,
"grad_norm": 0.953125,
"learning_rate": 0.0004978288890973529,
"loss": 5.3752,
"mean_token_accuracy": 0.1757800757884979,
"num_tokens": 12132510.0,
"step": 5835
},
{
"entropy": 5.561420106887818,
"epoch": 0.5297532656023222,
"grad_norm": 0.90234375,
"learning_rate": 0.0004978244026962698,
"loss": 5.3859,
"mean_token_accuracy": 0.17332411408424378,
"num_tokens": 12142419.0,
"step": 5840
},
{
"entropy": 5.6124217987060545,
"epoch": 0.5302068214804064,
"grad_norm": 1.0546875,
"learning_rate": 0.0004978199116871212,
"loss": 5.4348,
"mean_token_accuracy": 0.17327335476875305,
"num_tokens": 12152759.0,
"step": 5845
},
{
"entropy": 5.605248260498047,
"epoch": 0.5306603773584906,
"grad_norm": 0.86328125,
"learning_rate": 0.00049781541607,
"loss": 5.3847,
"mean_token_accuracy": 0.17561568468809127,
"num_tokens": 12162758.0,
"step": 5850
},
{
"entropy": 5.59306902885437,
"epoch": 0.5311139332365747,
"grad_norm": 0.921875,
"learning_rate": 0.0004978109158449993,
"loss": 5.444,
"mean_token_accuracy": 0.17255426645278932,
"num_tokens": 12172664.0,
"step": 5855
},
{
"entropy": 5.647549057006836,
"epoch": 0.5315674891146589,
"grad_norm": 0.8984375,
"learning_rate": 0.0004978064110122121,
"loss": 5.402,
"mean_token_accuracy": 0.17630172818899154,
"num_tokens": 12182028.0,
"step": 5860
},
{
"entropy": 5.641605806350708,
"epoch": 0.532021044992743,
"grad_norm": 0.91015625,
"learning_rate": 0.0004978019015717314,
"loss": 5.4504,
"mean_token_accuracy": 0.17413344085216523,
"num_tokens": 12192168.0,
"step": 5865
},
{
"entropy": 5.551989269256592,
"epoch": 0.5324746008708273,
"grad_norm": 0.92578125,
"learning_rate": 0.0004977973875236508,
"loss": 5.3544,
"mean_token_accuracy": 0.1735835164785385,
"num_tokens": 12202889.0,
"step": 5870
},
{
"entropy": 5.55653338432312,
"epoch": 0.5329281567489115,
"grad_norm": 0.9453125,
"learning_rate": 0.0004977928688680634,
"loss": 5.3953,
"mean_token_accuracy": 0.17994537502527236,
"num_tokens": 12212514.0,
"step": 5875
},
{
"entropy": 5.547123765945434,
"epoch": 0.5333817126269956,
"grad_norm": 0.96484375,
"learning_rate": 0.0004977883456050626,
"loss": 5.2653,
"mean_token_accuracy": 0.18208599984645843,
"num_tokens": 12222353.0,
"step": 5880
},
{
"entropy": 5.6921580791473385,
"epoch": 0.5338352685050798,
"grad_norm": 0.9296875,
"learning_rate": 0.000497783817734742,
"loss": 5.5429,
"mean_token_accuracy": 0.16759776920080185,
"num_tokens": 12232845.0,
"step": 5885
},
{
"entropy": 5.606061315536499,
"epoch": 0.534288824383164,
"grad_norm": 0.96875,
"learning_rate": 0.0004977792852571954,
"loss": 5.4554,
"mean_token_accuracy": 0.16563831120729447,
"num_tokens": 12242950.0,
"step": 5890
},
{
"entropy": 5.595444107055664,
"epoch": 0.5347423802612482,
"grad_norm": 0.8828125,
"learning_rate": 0.0004977747481725164,
"loss": 5.4387,
"mean_token_accuracy": 0.17201048582792283,
"num_tokens": 12253132.0,
"step": 5895
},
{
"entropy": 5.632798957824707,
"epoch": 0.5351959361393324,
"grad_norm": 0.94140625,
"learning_rate": 0.0004977702064807987,
"loss": 5.395,
"mean_token_accuracy": 0.17704534381628037,
"num_tokens": 12263214.0,
"step": 5900
},
{
"entropy": 5.541617584228516,
"epoch": 0.5356494920174165,
"grad_norm": 0.953125,
"learning_rate": 0.0004977656601821366,
"loss": 5.3208,
"mean_token_accuracy": 0.1798873171210289,
"num_tokens": 12273250.0,
"step": 5905
},
{
"entropy": 5.7590607643127445,
"epoch": 0.5361030478955007,
"grad_norm": 0.97265625,
"learning_rate": 0.0004977611092766238,
"loss": 5.5101,
"mean_token_accuracy": 0.1787897080183029,
"num_tokens": 12283767.0,
"step": 5910
},
{
"entropy": 5.534255886077881,
"epoch": 0.5365566037735849,
"grad_norm": 0.8515625,
"learning_rate": 0.0004977565537643544,
"loss": 5.3581,
"mean_token_accuracy": 0.17475240379571916,
"num_tokens": 12294612.0,
"step": 5915
},
{
"entropy": 5.534429550170898,
"epoch": 0.5370101596516691,
"grad_norm": 0.859375,
"learning_rate": 0.0004977519936454227,
"loss": 5.4397,
"mean_token_accuracy": 0.17676630169153212,
"num_tokens": 12305614.0,
"step": 5920
},
{
"entropy": 5.567504644393921,
"epoch": 0.5374637155297532,
"grad_norm": 0.90625,
"learning_rate": 0.0004977474289199231,
"loss": 5.3132,
"mean_token_accuracy": 0.1735585629940033,
"num_tokens": 12316152.0,
"step": 5925
},
{
"entropy": 5.551148366928101,
"epoch": 0.5379172714078374,
"grad_norm": 0.97265625,
"learning_rate": 0.0004977428595879499,
"loss": 5.3905,
"mean_token_accuracy": 0.178743177652359,
"num_tokens": 12325940.0,
"step": 5930
},
{
"entropy": 5.566306829452515,
"epoch": 0.5383708272859217,
"grad_norm": 1.140625,
"learning_rate": 0.0004977382856495975,
"loss": 5.39,
"mean_token_accuracy": 0.17976804822683334,
"num_tokens": 12335482.0,
"step": 5935
},
{
"entropy": 5.655665397644043,
"epoch": 0.5388243831640058,
"grad_norm": 0.98828125,
"learning_rate": 0.0004977337071049606,
"loss": 5.4884,
"mean_token_accuracy": 0.1725362166762352,
"num_tokens": 12346386.0,
"step": 5940
},
{
"entropy": 5.696004199981689,
"epoch": 0.53927793904209,
"grad_norm": 0.96484375,
"learning_rate": 0.000497729123954134,
"loss": 5.4451,
"mean_token_accuracy": 0.17383963018655776,
"num_tokens": 12356282.0,
"step": 5945
},
{
"entropy": 5.601561498641968,
"epoch": 0.5397314949201741,
"grad_norm": 0.90625,
"learning_rate": 0.0004977245361972123,
"loss": 5.4141,
"mean_token_accuracy": 0.17678642570972442,
"num_tokens": 12366994.0,
"step": 5950
},
{
"entropy": 5.545989513397217,
"epoch": 0.5401850507982583,
"grad_norm": 0.91015625,
"learning_rate": 0.0004977199438342904,
"loss": 5.3017,
"mean_token_accuracy": 0.18050941973924636,
"num_tokens": 12377524.0,
"step": 5955
},
{
"entropy": 5.595731210708618,
"epoch": 0.5406386066763426,
"grad_norm": 0.93359375,
"learning_rate": 0.0004977153468654634,
"loss": 5.4655,
"mean_token_accuracy": 0.17503209561109542,
"num_tokens": 12387593.0,
"step": 5960
},
{
"entropy": 5.697256326675415,
"epoch": 0.5410921625544267,
"grad_norm": 0.8671875,
"learning_rate": 0.0004977107452908262,
"loss": 5.4273,
"mean_token_accuracy": 0.1787885919213295,
"num_tokens": 12398732.0,
"step": 5965
},
{
"entropy": 5.61314206123352,
"epoch": 0.5415457184325109,
"grad_norm": 0.984375,
"learning_rate": 0.000497706139110474,
"loss": 5.4252,
"mean_token_accuracy": 0.18124329149723054,
"num_tokens": 12409821.0,
"step": 5970
},
{
"entropy": 5.593740558624267,
"epoch": 0.541999274310595,
"grad_norm": 1.0078125,
"learning_rate": 0.0004977015283245021,
"loss": 5.4482,
"mean_token_accuracy": 0.17267510890960694,
"num_tokens": 12420044.0,
"step": 5975
},
{
"entropy": 5.538191413879394,
"epoch": 0.5424528301886793,
"grad_norm": 0.91015625,
"learning_rate": 0.0004976969129330059,
"loss": 5.3615,
"mean_token_accuracy": 0.17179904878139496,
"num_tokens": 12430980.0,
"step": 5980
},
{
"entropy": 5.598214149475098,
"epoch": 0.5429063860667634,
"grad_norm": 0.8828125,
"learning_rate": 0.0004976922929360808,
"loss": 5.4067,
"mean_token_accuracy": 0.1799796149134636,
"num_tokens": 12441999.0,
"step": 5985
},
{
"entropy": 5.655162572860718,
"epoch": 0.5433599419448476,
"grad_norm": 0.8515625,
"learning_rate": 0.0004976876683338222,
"loss": 5.4457,
"mean_token_accuracy": 0.1742375373840332,
"num_tokens": 12452837.0,
"step": 5990
},
{
"entropy": 5.573116827011108,
"epoch": 0.5438134978229318,
"grad_norm": 0.97265625,
"learning_rate": 0.000497683039126326,
"loss": 5.3746,
"mean_token_accuracy": 0.17719283998012542,
"num_tokens": 12462142.0,
"step": 5995
},
{
"entropy": 5.566186046600341,
"epoch": 0.5442670537010159,
"grad_norm": 0.91015625,
"learning_rate": 0.0004976784053136878,
"loss": 5.4094,
"mean_token_accuracy": 0.1800991877913475,
"num_tokens": 12472756.0,
"step": 6000
},
{
"epoch": 0.5442670537010159,
"eval_entropy": 5.364191224024846,
"eval_loss": 5.434150695800781,
"eval_mean_token_accuracy": 0.18153274842238737,
"eval_num_tokens": 12472756.0,
"eval_runtime": 20.5722,
"eval_samples_per_second": 1718.825,
"eval_steps_per_second": 214.853,
"step": 6000
},
{
"entropy": 5.625031852722168,
"epoch": 0.5447206095791002,
"grad_norm": 0.88671875,
"learning_rate": 0.0004976737668960034,
"loss": 5.5318,
"mean_token_accuracy": 0.1683395102620125,
"num_tokens": 12483115.0,
"step": 6005
},
{
"entropy": 5.665145683288574,
"epoch": 0.5451741654571843,
"grad_norm": 0.9140625,
"learning_rate": 0.0004976691238733688,
"loss": 5.3863,
"mean_token_accuracy": 0.17331260144710542,
"num_tokens": 12493280.0,
"step": 6010
},
{
"entropy": 5.628579378128052,
"epoch": 0.5456277213352685,
"grad_norm": 0.96875,
"learning_rate": 0.00049766447624588,
"loss": 5.5164,
"mean_token_accuracy": 0.17446643561124803,
"num_tokens": 12503665.0,
"step": 6015
},
{
"entropy": 5.585593891143799,
"epoch": 0.5460812772133526,
"grad_norm": 1.0,
"learning_rate": 0.000497659824013633,
"loss": 5.4308,
"mean_token_accuracy": 0.17111265659332275,
"num_tokens": 12513381.0,
"step": 6020
},
{
"entropy": 5.681701898574829,
"epoch": 0.5465348330914369,
"grad_norm": 0.85546875,
"learning_rate": 0.0004976551671767241,
"loss": 5.4617,
"mean_token_accuracy": 0.16975319087505342,
"num_tokens": 12524121.0,
"step": 6025
},
{
"entropy": 5.595137882232666,
"epoch": 0.5469883889695211,
"grad_norm": 0.86328125,
"learning_rate": 0.0004976505057352496,
"loss": 5.4145,
"mean_token_accuracy": 0.17792607247829437,
"num_tokens": 12533972.0,
"step": 6030
},
{
"entropy": 5.633500862121582,
"epoch": 0.5474419448476052,
"grad_norm": 0.93359375,
"learning_rate": 0.0004976458396893059,
"loss": 5.5324,
"mean_token_accuracy": 0.1650451496243477,
"num_tokens": 12544110.0,
"step": 6035
},
{
"entropy": 5.641628408432007,
"epoch": 0.5478955007256894,
"grad_norm": 0.87890625,
"learning_rate": 0.0004976411690389896,
"loss": 5.3603,
"mean_token_accuracy": 0.17129435241222382,
"num_tokens": 12554273.0,
"step": 6040
},
{
"entropy": 5.655753946304321,
"epoch": 0.5483490566037735,
"grad_norm": 0.94140625,
"learning_rate": 0.000497636493784397,
"loss": 5.4287,
"mean_token_accuracy": 0.17427540570497513,
"num_tokens": 12564633.0,
"step": 6045
},
{
"entropy": 5.596440839767456,
"epoch": 0.5488026124818578,
"grad_norm": 0.9609375,
"learning_rate": 0.0004976318139256251,
"loss": 5.3635,
"mean_token_accuracy": 0.1775827169418335,
"num_tokens": 12573790.0,
"step": 6050
},
{
"entropy": 5.63446683883667,
"epoch": 0.549256168359942,
"grad_norm": 0.99609375,
"learning_rate": 0.0004976271294627706,
"loss": 5.441,
"mean_token_accuracy": 0.17123547047376633,
"num_tokens": 12584265.0,
"step": 6055
},
{
"entropy": 5.5846747875213625,
"epoch": 0.5497097242380261,
"grad_norm": 0.875,
"learning_rate": 0.0004976224403959303,
"loss": 5.3174,
"mean_token_accuracy": 0.17618338912725448,
"num_tokens": 12594413.0,
"step": 6060
},
{
"entropy": 5.568248081207275,
"epoch": 0.5501632801161103,
"grad_norm": 0.92578125,
"learning_rate": 0.0004976177467252012,
"loss": 5.3923,
"mean_token_accuracy": 0.1851392775774002,
"num_tokens": 12604637.0,
"step": 6065
},
{
"entropy": 5.624586772918701,
"epoch": 0.5506168359941945,
"grad_norm": 0.859375,
"learning_rate": 0.0004976130484506804,
"loss": 5.4582,
"mean_token_accuracy": 0.17681263536214828,
"num_tokens": 12614873.0,
"step": 6070
},
{
"entropy": 5.648846769332886,
"epoch": 0.5510703918722787,
"grad_norm": 0.87109375,
"learning_rate": 0.0004976083455724649,
"loss": 5.3823,
"mean_token_accuracy": 0.17441451996564866,
"num_tokens": 12626182.0,
"step": 6075
},
{
"entropy": 5.535418891906739,
"epoch": 0.5515239477503628,
"grad_norm": 0.9140625,
"learning_rate": 0.0004976036380906523,
"loss": 5.3149,
"mean_token_accuracy": 0.17663441896438598,
"num_tokens": 12635418.0,
"step": 6080
},
{
"entropy": 5.582258033752441,
"epoch": 0.551977503628447,
"grad_norm": 0.91796875,
"learning_rate": 0.0004975989260053395,
"loss": 5.4765,
"mean_token_accuracy": 0.16567691415548325,
"num_tokens": 12645942.0,
"step": 6085
},
{
"entropy": 5.617289209365845,
"epoch": 0.5524310595065312,
"grad_norm": 0.98046875,
"learning_rate": 0.0004975942093166242,
"loss": 5.3982,
"mean_token_accuracy": 0.17294339835643768,
"num_tokens": 12656460.0,
"step": 6090
},
{
"entropy": 5.636151075363159,
"epoch": 0.5528846153846154,
"grad_norm": 0.90625,
"learning_rate": 0.0004975894880246041,
"loss": 5.3228,
"mean_token_accuracy": 0.1788621500134468,
"num_tokens": 12666916.0,
"step": 6095
},
{
"entropy": 5.526174592971802,
"epoch": 0.5533381712626996,
"grad_norm": 0.953125,
"learning_rate": 0.0004975847621293766,
"loss": 5.3853,
"mean_token_accuracy": 0.18073209524154663,
"num_tokens": 12677562.0,
"step": 6100
},
{
"entropy": 5.550758695602417,
"epoch": 0.5537917271407837,
"grad_norm": 0.89453125,
"learning_rate": 0.0004975800316310394,
"loss": 5.3618,
"mean_token_accuracy": 0.1773673951625824,
"num_tokens": 12687349.0,
"step": 6105
},
{
"entropy": 5.643798589706421,
"epoch": 0.5542452830188679,
"grad_norm": 0.8828125,
"learning_rate": 0.0004975752965296906,
"loss": 5.4098,
"mean_token_accuracy": 0.17631264328956603,
"num_tokens": 12697850.0,
"step": 6110
},
{
"entropy": 5.540865325927735,
"epoch": 0.5546988388969522,
"grad_norm": 0.921875,
"learning_rate": 0.0004975705568254278,
"loss": 5.2774,
"mean_token_accuracy": 0.17912191450595855,
"num_tokens": 12708283.0,
"step": 6115
},
{
"entropy": 5.57166051864624,
"epoch": 0.5551523947750363,
"grad_norm": 0.88671875,
"learning_rate": 0.0004975658125183493,
"loss": 5.3814,
"mean_token_accuracy": 0.17749570310115814,
"num_tokens": 12718201.0,
"step": 6120
},
{
"entropy": 5.551891422271728,
"epoch": 0.5556059506531205,
"grad_norm": 0.9453125,
"learning_rate": 0.000497561063608553,
"loss": 5.3217,
"mean_token_accuracy": 0.18132489323616027,
"num_tokens": 12728860.0,
"step": 6125
},
{
"entropy": 5.574612474441528,
"epoch": 0.5560595065312046,
"grad_norm": 0.81640625,
"learning_rate": 0.0004975563100961372,
"loss": 5.3979,
"mean_token_accuracy": 0.17559077590703964,
"num_tokens": 12739007.0,
"step": 6130
},
{
"entropy": 5.578898191452026,
"epoch": 0.5565130624092888,
"grad_norm": 0.9140625,
"learning_rate": 0.0004975515519812002,
"loss": 5.2461,
"mean_token_accuracy": 0.1819448158144951,
"num_tokens": 12749004.0,
"step": 6135
},
{
"entropy": 5.543379449844361,
"epoch": 0.556966618287373,
"grad_norm": 0.93359375,
"learning_rate": 0.0004975467892638402,
"loss": 5.3959,
"mean_token_accuracy": 0.17813464105129242,
"num_tokens": 12760085.0,
"step": 6140
},
{
"entropy": 5.59183669090271,
"epoch": 0.5574201741654572,
"grad_norm": 0.87890625,
"learning_rate": 0.0004975420219441561,
"loss": 5.3429,
"mean_token_accuracy": 0.18399171233177186,
"num_tokens": 12770303.0,
"step": 6145
},
{
"entropy": 5.58007698059082,
"epoch": 0.5578737300435413,
"grad_norm": 0.9921875,
"learning_rate": 0.0004975372500222462,
"loss": 5.3724,
"mean_token_accuracy": 0.17818562537431717,
"num_tokens": 12780334.0,
"step": 6150
},
{
"entropy": 5.587021446228027,
"epoch": 0.5583272859216255,
"grad_norm": 0.9453125,
"learning_rate": 0.0004975324734982094,
"loss": 5.4218,
"mean_token_accuracy": 0.17108138799667358,
"num_tokens": 12790980.0,
"step": 6155
},
{
"entropy": 5.621369981765747,
"epoch": 0.5587808417997098,
"grad_norm": 1.0078125,
"learning_rate": 0.0004975276923721442,
"loss": 5.4611,
"mean_token_accuracy": 0.17504812330007552,
"num_tokens": 12801713.0,
"step": 6160
},
{
"entropy": 5.644250059127808,
"epoch": 0.5592343976777939,
"grad_norm": 0.87109375,
"learning_rate": 0.0004975229066441496,
"loss": 5.4677,
"mean_token_accuracy": 0.1687561482191086,
"num_tokens": 12813040.0,
"step": 6165
},
{
"entropy": 5.577824926376342,
"epoch": 0.5596879535558781,
"grad_norm": 0.8984375,
"learning_rate": 0.0004975181163143247,
"loss": 5.4127,
"mean_token_accuracy": 0.17071571797132493,
"num_tokens": 12824048.0,
"step": 6170
},
{
"entropy": 5.64132513999939,
"epoch": 0.5601415094339622,
"grad_norm": 0.9453125,
"learning_rate": 0.0004975133213827684,
"loss": 5.3687,
"mean_token_accuracy": 0.17483995854854584,
"num_tokens": 12834297.0,
"step": 6175
},
{
"entropy": 5.66928539276123,
"epoch": 0.5605950653120464,
"grad_norm": 0.9296875,
"learning_rate": 0.0004975085218495798,
"loss": 5.4364,
"mean_token_accuracy": 0.17078946828842162,
"num_tokens": 12844574.0,
"step": 6180
},
{
"entropy": 5.536151313781739,
"epoch": 0.5610486211901307,
"grad_norm": 0.91015625,
"learning_rate": 0.0004975037177148585,
"loss": 5.3661,
"mean_token_accuracy": 0.177015121281147,
"num_tokens": 12854670.0,
"step": 6185
},
{
"entropy": 5.536833953857422,
"epoch": 0.5615021770682148,
"grad_norm": 0.87109375,
"learning_rate": 0.0004974989089787035,
"loss": 5.3828,
"mean_token_accuracy": 0.17939384132623673,
"num_tokens": 12865281.0,
"step": 6190
},
{
"entropy": 5.683811044692993,
"epoch": 0.561955732946299,
"grad_norm": 0.921875,
"learning_rate": 0.0004974940956412143,
"loss": 5.3903,
"mean_token_accuracy": 0.1783251017332077,
"num_tokens": 12875222.0,
"step": 6195
},
{
"entropy": 5.7399190902709964,
"epoch": 0.5624092888243831,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974892777024905,
"loss": 5.4623,
"mean_token_accuracy": 0.16634879112243653,
"num_tokens": 12885079.0,
"step": 6200
},
{
"entropy": 5.523637962341309,
"epoch": 0.5628628447024674,
"grad_norm": 0.859375,
"learning_rate": 0.0004974844551626318,
"loss": 5.3998,
"mean_token_accuracy": 0.1781408593058586,
"num_tokens": 12896402.0,
"step": 6205
},
{
"entropy": 5.585960006713867,
"epoch": 0.5633164005805515,
"grad_norm": 0.92578125,
"learning_rate": 0.0004974796280217379,
"loss": 5.3653,
"mean_token_accuracy": 0.18042189478874207,
"num_tokens": 12906403.0,
"step": 6210
},
{
"entropy": 5.643087482452392,
"epoch": 0.5637699564586357,
"grad_norm": 0.9296875,
"learning_rate": 0.0004974747962799086,
"loss": 5.3895,
"mean_token_accuracy": 0.17042631208896636,
"num_tokens": 12917061.0,
"step": 6215
},
{
"entropy": 5.540905570983886,
"epoch": 0.5642235123367199,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974699599372437,
"loss": 5.427,
"mean_token_accuracy": 0.1783065214753151,
"num_tokens": 12927182.0,
"step": 6220
},
{
"entropy": 5.561410760879516,
"epoch": 0.564677068214804,
"grad_norm": 0.86328125,
"learning_rate": 0.0004974651189938434,
"loss": 5.3857,
"mean_token_accuracy": 0.1792475163936615,
"num_tokens": 12937371.0,
"step": 6225
},
{
"entropy": 5.633272504806518,
"epoch": 0.5651306240928883,
"grad_norm": 0.84375,
"learning_rate": 0.0004974602734498079,
"loss": 5.4352,
"mean_token_accuracy": 0.1712818294763565,
"num_tokens": 12948569.0,
"step": 6230
},
{
"entropy": 5.621938037872314,
"epoch": 0.5655841799709724,
"grad_norm": 0.87109375,
"learning_rate": 0.000497455423305237,
"loss": 5.3768,
"mean_token_accuracy": 0.17150397896766661,
"num_tokens": 12958503.0,
"step": 6235
},
{
"entropy": 5.601767444610596,
"epoch": 0.5660377358490566,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974505685602314,
"loss": 5.3988,
"mean_token_accuracy": 0.17525841742753984,
"num_tokens": 12968714.0,
"step": 6240
},
{
"entropy": 5.590086603164673,
"epoch": 0.5664912917271407,
"grad_norm": 0.9375,
"learning_rate": 0.0004974457092148912,
"loss": 5.4284,
"mean_token_accuracy": 0.1741039887070656,
"num_tokens": 12978857.0,
"step": 6245
},
{
"entropy": 5.720824956893921,
"epoch": 0.566944847605225,
"grad_norm": 0.88671875,
"learning_rate": 0.0004974408452693172,
"loss": 5.5322,
"mean_token_accuracy": 0.16714059710502624,
"num_tokens": 12989863.0,
"step": 6250
},
{
"entropy": 5.586817884445191,
"epoch": 0.5673984034833092,
"grad_norm": 0.93359375,
"learning_rate": 0.0004974359767236099,
"loss": 5.3248,
"mean_token_accuracy": 0.17573193907737733,
"num_tokens": 13000307.0,
"step": 6255
},
{
"entropy": 5.545666122436524,
"epoch": 0.5678519593613933,
"grad_norm": 0.99609375,
"learning_rate": 0.0004974311035778698,
"loss": 5.411,
"mean_token_accuracy": 0.18314283788204194,
"num_tokens": 13009053.0,
"step": 6260
},
{
"entropy": 5.5987457752227785,
"epoch": 0.5683055152394775,
"grad_norm": 0.9296875,
"learning_rate": 0.0004974262258321979,
"loss": 5.4269,
"mean_token_accuracy": 0.1716617226600647,
"num_tokens": 13019136.0,
"step": 6265
},
{
"entropy": 5.655656051635742,
"epoch": 0.5687590711175616,
"grad_norm": 0.859375,
"learning_rate": 0.0004974213434866949,
"loss": 5.4723,
"mean_token_accuracy": 0.16619940251111984,
"num_tokens": 13030356.0,
"step": 6270
},
{
"entropy": 5.637882804870605,
"epoch": 0.5692126269956459,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974164565414617,
"loss": 5.4883,
"mean_token_accuracy": 0.1748768150806427,
"num_tokens": 13040477.0,
"step": 6275
},
{
"entropy": 5.596619510650635,
"epoch": 0.56966618287373,
"grad_norm": 0.8515625,
"learning_rate": 0.0004974115649965998,
"loss": 5.3524,
"mean_token_accuracy": 0.18151768296957016,
"num_tokens": 13051755.0,
"step": 6280
},
{
"entropy": 5.604301977157593,
"epoch": 0.5701197387518142,
"grad_norm": 0.89453125,
"learning_rate": 0.0004974066688522098,
"loss": 5.3913,
"mean_token_accuracy": 0.18713292777538298,
"num_tokens": 13061670.0,
"step": 6285
},
{
"entropy": 5.60211033821106,
"epoch": 0.5705732946298984,
"grad_norm": 1.0,
"learning_rate": 0.0004974017681083934,
"loss": 5.3646,
"mean_token_accuracy": 0.1776226207613945,
"num_tokens": 13070917.0,
"step": 6290
},
{
"entropy": 5.585918712615967,
"epoch": 0.5710268505079826,
"grad_norm": 1.3046875,
"learning_rate": 0.0004973968627652516,
"loss": 5.4146,
"mean_token_accuracy": 0.17641005665063858,
"num_tokens": 13080577.0,
"step": 6295
},
{
"entropy": 5.6024633884429935,
"epoch": 0.5714804063860668,
"grad_norm": 0.9375,
"learning_rate": 0.000497391952822886,
"loss": 5.4371,
"mean_token_accuracy": 0.16923494786024093,
"num_tokens": 13091446.0,
"step": 6300
},
{
"entropy": 5.649172449111939,
"epoch": 0.5719339622641509,
"grad_norm": 0.8828125,
"learning_rate": 0.0004973870382813983,
"loss": 5.4614,
"mean_token_accuracy": 0.17273851484060287,
"num_tokens": 13102752.0,
"step": 6305
},
{
"entropy": 5.702681827545166,
"epoch": 0.5723875181422351,
"grad_norm": 0.890625,
"learning_rate": 0.0004973821191408899,
"loss": 5.4819,
"mean_token_accuracy": 0.17141962498426438,
"num_tokens": 13113944.0,
"step": 6310
},
{
"entropy": 5.622271108627319,
"epoch": 0.5728410740203193,
"grad_norm": 0.91015625,
"learning_rate": 0.0004973771954014626,
"loss": 5.4073,
"mean_token_accuracy": 0.17560603171586991,
"num_tokens": 13124316.0,
"step": 6315
},
{
"entropy": 5.6871038436889645,
"epoch": 0.5732946298984035,
"grad_norm": 0.9921875,
"learning_rate": 0.0004973722670632182,
"loss": 5.5038,
"mean_token_accuracy": 0.16931897699832915,
"num_tokens": 13133567.0,
"step": 6320
},
{
"entropy": 5.574843692779541,
"epoch": 0.5737481857764877,
"grad_norm": 0.90234375,
"learning_rate": 0.0004973673341262587,
"loss": 5.4081,
"mean_token_accuracy": 0.17334946990013123,
"num_tokens": 13144279.0,
"step": 6325
},
{
"entropy": 5.626540994644165,
"epoch": 0.5742017416545718,
"grad_norm": 0.95703125,
"learning_rate": 0.000497362396590686,
"loss": 5.4835,
"mean_token_accuracy": 0.16946994364261628,
"num_tokens": 13154463.0,
"step": 6330
},
{
"entropy": 5.65912938117981,
"epoch": 0.574655297532656,
"grad_norm": 0.828125,
"learning_rate": 0.0004973574544566024,
"loss": 5.4457,
"mean_token_accuracy": 0.16746120303869247,
"num_tokens": 13165565.0,
"step": 6335
},
{
"entropy": 5.559062385559082,
"epoch": 0.5751088534107403,
"grad_norm": 0.92578125,
"learning_rate": 0.0004973525077241098,
"loss": 5.3533,
"mean_token_accuracy": 0.18145330846309662,
"num_tokens": 13175839.0,
"step": 6340
},
{
"entropy": 5.5439225196838375,
"epoch": 0.5755624092888244,
"grad_norm": 0.9140625,
"learning_rate": 0.0004973475563933109,
"loss": 5.3421,
"mean_token_accuracy": 0.17647033035755158,
"num_tokens": 13186372.0,
"step": 6345
},
{
"entropy": 5.629715633392334,
"epoch": 0.5760159651669086,
"grad_norm": 0.93359375,
"learning_rate": 0.0004973426004643077,
"loss": 5.4044,
"mean_token_accuracy": 0.18002848476171493,
"num_tokens": 13196259.0,
"step": 6350
},
{
"entropy": 5.643421077728272,
"epoch": 0.5764695210449927,
"grad_norm": 0.99609375,
"learning_rate": 0.0004973376399372029,
"loss": 5.3786,
"mean_token_accuracy": 0.1787116289138794,
"num_tokens": 13206497.0,
"step": 6355
},
{
"entropy": 5.605941486358643,
"epoch": 0.5769230769230769,
"grad_norm": 0.8984375,
"learning_rate": 0.0004973326748120991,
"loss": 5.4201,
"mean_token_accuracy": 0.17948209047317504,
"num_tokens": 13217188.0,
"step": 6360
},
{
"entropy": 5.575176668167114,
"epoch": 0.5773766328011611,
"grad_norm": 0.921875,
"learning_rate": 0.0004973277050890988,
"loss": 5.4289,
"mean_token_accuracy": 0.17635478228330612,
"num_tokens": 13226544.0,
"step": 6365
},
{
"entropy": 5.662482213973999,
"epoch": 0.5778301886792453,
"grad_norm": 0.98046875,
"learning_rate": 0.0004973227307683049,
"loss": 5.4469,
"mean_token_accuracy": 0.17289083003997802,
"num_tokens": 13236572.0,
"step": 6370
},
{
"entropy": 5.716796112060547,
"epoch": 0.5782837445573294,
"grad_norm": 0.87109375,
"learning_rate": 0.0004973177518498204,
"loss": 5.4467,
"mean_token_accuracy": 0.1772434666752815,
"num_tokens": 13247958.0,
"step": 6375
},
{
"entropy": 5.598628664016724,
"epoch": 0.5787373004354136,
"grad_norm": 0.9375,
"learning_rate": 0.0004973127683337482,
"loss": 5.3668,
"mean_token_accuracy": 0.17798337638378142,
"num_tokens": 13258255.0,
"step": 6380
},
{
"entropy": 5.528913688659668,
"epoch": 0.5791908563134979,
"grad_norm": 0.89453125,
"learning_rate": 0.0004973077802201911,
"loss": 5.3936,
"mean_token_accuracy": 0.17277796119451522,
"num_tokens": 13268416.0,
"step": 6385
},
{
"entropy": 5.572034215927124,
"epoch": 0.579644412191582,
"grad_norm": 1.0078125,
"learning_rate": 0.0004973027875092526,
"loss": 5.3724,
"mean_token_accuracy": 0.17329427152872084,
"num_tokens": 13278087.0,
"step": 6390
},
{
"entropy": 5.559347438812256,
"epoch": 0.5800979680696662,
"grad_norm": 0.9609375,
"learning_rate": 0.0004972977902010357,
"loss": 5.2809,
"mean_token_accuracy": 0.18577326983213424,
"num_tokens": 13288121.0,
"step": 6395
},
{
"entropy": 5.696083450317383,
"epoch": 0.5805515239477503,
"grad_norm": 0.9609375,
"learning_rate": 0.0004972927882956439,
"loss": 5.4854,
"mean_token_accuracy": 0.16661122292280198,
"num_tokens": 13298620.0,
"step": 6400
},
{
"entropy": 5.555794334411621,
"epoch": 0.5810050798258345,
"grad_norm": 0.94921875,
"learning_rate": 0.0004972877817931807,
"loss": 5.3054,
"mean_token_accuracy": 0.17329084128141403,
"num_tokens": 13308413.0,
"step": 6405
},
{
"entropy": 5.545690584182739,
"epoch": 0.5814586357039188,
"grad_norm": 0.93359375,
"learning_rate": 0.0004972827706937493,
"loss": 5.3926,
"mean_token_accuracy": 0.17548721879720688,
"num_tokens": 13319387.0,
"step": 6410
},
{
"entropy": 5.60416431427002,
"epoch": 0.5819121915820029,
"grad_norm": 0.875,
"learning_rate": 0.0004972777549974537,
"loss": 5.3385,
"mean_token_accuracy": 0.17935979068279267,
"num_tokens": 13330494.0,
"step": 6415
},
{
"entropy": 5.593369865417481,
"epoch": 0.5823657474600871,
"grad_norm": 0.94140625,
"learning_rate": 0.0004972727347043975,
"loss": 5.3587,
"mean_token_accuracy": 0.17618109732866288,
"num_tokens": 13341009.0,
"step": 6420
},
{
"entropy": 5.541242694854736,
"epoch": 0.5828193033381712,
"grad_norm": 0.9921875,
"learning_rate": 0.0004972677098146845,
"loss": 5.3054,
"mean_token_accuracy": 0.17940136939287185,
"num_tokens": 13351269.0,
"step": 6425
},
{
"entropy": 5.521207714080811,
"epoch": 0.5832728592162555,
"grad_norm": 0.93359375,
"learning_rate": 0.0004972626803284185,
"loss": 5.4117,
"mean_token_accuracy": 0.17470855563879012,
"num_tokens": 13362160.0,
"step": 6430
},
{
"entropy": 5.640540981292725,
"epoch": 0.5837264150943396,
"grad_norm": 0.98828125,
"learning_rate": 0.0004972576462457038,
"loss": 5.4703,
"mean_token_accuracy": 0.17759566456079484,
"num_tokens": 13372497.0,
"step": 6435
},
{
"entropy": 5.697097206115723,
"epoch": 0.5841799709724238,
"grad_norm": 0.96875,
"learning_rate": 0.0004972526075666442,
"loss": 5.4133,
"mean_token_accuracy": 0.18020133376121522,
"num_tokens": 13381665.0,
"step": 6440
},
{
"entropy": 5.543485450744629,
"epoch": 0.584633526850508,
"grad_norm": 0.9140625,
"learning_rate": 0.0004972475642913442,
"loss": 5.3197,
"mean_token_accuracy": 0.18023221790790558,
"num_tokens": 13392507.0,
"step": 6445
},
{
"entropy": 5.505417203903198,
"epoch": 0.5850870827285921,
"grad_norm": 0.98828125,
"learning_rate": 0.0004972425164199078,
"loss": 5.3031,
"mean_token_accuracy": 0.18267336785793303,
"num_tokens": 13402171.0,
"step": 6450
},
{
"entropy": 5.613723850250244,
"epoch": 0.5855406386066764,
"grad_norm": 0.87109375,
"learning_rate": 0.0004972374639524396,
"loss": 5.525,
"mean_token_accuracy": 0.16901201158761978,
"num_tokens": 13412933.0,
"step": 6455
},
{
"entropy": 5.6370524883270265,
"epoch": 0.5859941944847605,
"grad_norm": 0.82421875,
"learning_rate": 0.000497232406889044,
"loss": 5.3751,
"mean_token_accuracy": 0.16950386315584182,
"num_tokens": 13423563.0,
"step": 6460
},
{
"entropy": 5.6557940483093265,
"epoch": 0.5864477503628447,
"grad_norm": 0.96875,
"learning_rate": 0.0004972273452298255,
"loss": 5.3847,
"mean_token_accuracy": 0.1834123909473419,
"num_tokens": 13433998.0,
"step": 6465
},
{
"entropy": 5.559069585800171,
"epoch": 0.5869013062409288,
"grad_norm": 0.8984375,
"learning_rate": 0.0004972222789748888,
"loss": 5.3315,
"mean_token_accuracy": 0.18007088154554368,
"num_tokens": 13443878.0,
"step": 6470
},
{
"entropy": 5.556202745437622,
"epoch": 0.5873548621190131,
"grad_norm": 0.82421875,
"learning_rate": 0.0004972172081243389,
"loss": 5.4097,
"mean_token_accuracy": 0.1707281395792961,
"num_tokens": 13456051.0,
"step": 6475
},
{
"entropy": 5.606028318405151,
"epoch": 0.5878084179970973,
"grad_norm": 0.84375,
"learning_rate": 0.0004972121326782806,
"loss": 5.3974,
"mean_token_accuracy": 0.17891442328691481,
"num_tokens": 13467276.0,
"step": 6480
},
{
"entropy": 5.6323326587677,
"epoch": 0.5882619738751814,
"grad_norm": 0.9140625,
"learning_rate": 0.0004972070526368186,
"loss": 5.3717,
"mean_token_accuracy": 0.17427799254655837,
"num_tokens": 13478331.0,
"step": 6485
},
{
"entropy": 5.632275676727295,
"epoch": 0.5887155297532656,
"grad_norm": 0.9765625,
"learning_rate": 0.0004972019680000582,
"loss": 5.3205,
"mean_token_accuracy": 0.17633114755153656,
"num_tokens": 13489879.0,
"step": 6490
},
{
"entropy": 5.563889217376709,
"epoch": 0.5891690856313497,
"grad_norm": 0.86328125,
"learning_rate": 0.0004971968787681044,
"loss": 5.4351,
"mean_token_accuracy": 0.18205238431692122,
"num_tokens": 13500879.0,
"step": 6495
},
{
"entropy": 5.674321556091309,
"epoch": 0.589622641509434,
"grad_norm": 0.87109375,
"learning_rate": 0.0004971917849410627,
"loss": 5.4231,
"mean_token_accuracy": 0.16968515366315842,
"num_tokens": 13511719.0,
"step": 6500
},
{
"entropy": 5.5601965427398685,
"epoch": 0.5900761973875182,
"grad_norm": 0.91015625,
"learning_rate": 0.0004971866865190381,
"loss": 5.3502,
"mean_token_accuracy": 0.17365157157182692,
"num_tokens": 13521850.0,
"step": 6505
},
{
"entropy": 5.614747524261475,
"epoch": 0.5905297532656023,
"grad_norm": 0.9296875,
"learning_rate": 0.0004971815835021362,
"loss": 5.3834,
"mean_token_accuracy": 0.18184393644332886,
"num_tokens": 13532095.0,
"step": 6510
},
{
"entropy": 5.575075674057007,
"epoch": 0.5909833091436865,
"grad_norm": 0.89453125,
"learning_rate": 0.0004971764758904626,
"loss": 5.3719,
"mean_token_accuracy": 0.17141948640346527,
"num_tokens": 13543133.0,
"step": 6515
},
{
"entropy": 5.587063646316528,
"epoch": 0.5914368650217707,
"grad_norm": 0.93359375,
"learning_rate": 0.0004971713636841229,
"loss": 5.438,
"mean_token_accuracy": 0.17123900204896927,
"num_tokens": 13553771.0,
"step": 6520
},
{
"entropy": 5.544393587112427,
"epoch": 0.5918904208998549,
"grad_norm": 0.875,
"learning_rate": 0.0004971662468832228,
"loss": 5.3763,
"mean_token_accuracy": 0.1815660074353218,
"num_tokens": 13564548.0,
"step": 6525
},
{
"entropy": 5.5182641506195065,
"epoch": 0.592343976777939,
"grad_norm": 0.94140625,
"learning_rate": 0.000497161125487868,
"loss": 5.3729,
"mean_token_accuracy": 0.17794521152973175,
"num_tokens": 13574567.0,
"step": 6530
},
{
"entropy": 5.551377677917481,
"epoch": 0.5927975326560232,
"grad_norm": 0.88671875,
"learning_rate": 0.0004971559994981645,
"loss": 5.3972,
"mean_token_accuracy": 0.17787606418132781,
"num_tokens": 13585302.0,
"step": 6535
},
{
"entropy": 5.653764629364014,
"epoch": 0.5932510885341074,
"grad_norm": 0.93359375,
"learning_rate": 0.0004971508689142184,
"loss": 5.428,
"mean_token_accuracy": 0.17052206993103028,
"num_tokens": 13594754.0,
"step": 6540
},
{
"entropy": 5.5617927551269535,
"epoch": 0.5937046444121916,
"grad_norm": 0.90625,
"learning_rate": 0.0004971457337361356,
"loss": 5.3676,
"mean_token_accuracy": 0.17397885471582414,
"num_tokens": 13604691.0,
"step": 6545
},
{
"entropy": 5.476242637634277,
"epoch": 0.5941582002902758,
"grad_norm": 0.9375,
"learning_rate": 0.0004971405939640225,
"loss": 5.2587,
"mean_token_accuracy": 0.17924531549215317,
"num_tokens": 13613911.0,
"step": 6550
},
{
"entropy": 5.727000761032104,
"epoch": 0.5946117561683599,
"grad_norm": 0.89453125,
"learning_rate": 0.0004971354495979854,
"loss": 5.4647,
"mean_token_accuracy": 0.17712009847164153,
"num_tokens": 13624285.0,
"step": 6555
},
{
"entropy": 5.584723997116089,
"epoch": 0.5950653120464441,
"grad_norm": 0.91796875,
"learning_rate": 0.0004971303006381305,
"loss": 5.3066,
"mean_token_accuracy": 0.18122798055410386,
"num_tokens": 13635434.0,
"step": 6560
},
{
"entropy": 5.511235284805298,
"epoch": 0.5955188679245284,
"grad_norm": 0.9453125,
"learning_rate": 0.0004971251470845643,
"loss": 5.3757,
"mean_token_accuracy": 0.17572649121284484,
"num_tokens": 13645582.0,
"step": 6565
},
{
"entropy": 5.580762100219727,
"epoch": 0.5959724238026125,
"grad_norm": 0.97265625,
"learning_rate": 0.0004971199889373935,
"loss": 5.3365,
"mean_token_accuracy": 0.179083950817585,
"num_tokens": 13655029.0,
"step": 6570
},
{
"entropy": 5.564962148666382,
"epoch": 0.5964259796806967,
"grad_norm": 0.87890625,
"learning_rate": 0.0004971148261967249,
"loss": 5.3325,
"mean_token_accuracy": 0.17644450217485427,
"num_tokens": 13665204.0,
"step": 6575
},
{
"entropy": 5.571329927444458,
"epoch": 0.5968795355587808,
"grad_norm": 0.90234375,
"learning_rate": 0.0004971096588626649,
"loss": 5.4285,
"mean_token_accuracy": 0.174003441631794,
"num_tokens": 13676281.0,
"step": 6580
},
{
"entropy": 5.577637481689453,
"epoch": 0.597333091436865,
"grad_norm": 0.85546875,
"learning_rate": 0.0004971044869353206,
"loss": 5.3917,
"mean_token_accuracy": 0.18069213032722473,
"num_tokens": 13687144.0,
"step": 6585
},
{
"entropy": 5.600950384140015,
"epoch": 0.5977866473149492,
"grad_norm": 0.9921875,
"learning_rate": 0.000497099310414799,
"loss": 5.3665,
"mean_token_accuracy": 0.1816061720252037,
"num_tokens": 13696371.0,
"step": 6590
},
{
"entropy": 5.622022247314453,
"epoch": 0.5982402031930334,
"grad_norm": 1.0,
"learning_rate": 0.0004970941293012069,
"loss": 5.3139,
"mean_token_accuracy": 0.17685348838567733,
"num_tokens": 13706900.0,
"step": 6595
},
{
"entropy": 5.555879068374634,
"epoch": 0.5986937590711175,
"grad_norm": 0.84765625,
"learning_rate": 0.0004970889435946518,
"loss": 5.3977,
"mean_token_accuracy": 0.18049706071615218,
"num_tokens": 13717874.0,
"step": 6600
},
{
"entropy": 5.588134765625,
"epoch": 0.5991473149492017,
"grad_norm": 0.88671875,
"learning_rate": 0.0004970837532952406,
"loss": 5.3653,
"mean_token_accuracy": 0.1767413467168808,
"num_tokens": 13728020.0,
"step": 6605
},
{
"entropy": 5.60174593925476,
"epoch": 0.599600870827286,
"grad_norm": 0.89453125,
"learning_rate": 0.0004970785584030808,
"loss": 5.3963,
"mean_token_accuracy": 0.1784649208188057,
"num_tokens": 13738007.0,
"step": 6610
},
{
"entropy": 5.601901006698609,
"epoch": 0.6000544267053701,
"grad_norm": 0.90234375,
"learning_rate": 0.0004970733589182799,
"loss": 5.4084,
"mean_token_accuracy": 0.1740790531039238,
"num_tokens": 13749429.0,
"step": 6615
},
{
"entropy": 5.5603643417358395,
"epoch": 0.6005079825834543,
"grad_norm": 1.03125,
"learning_rate": 0.0004970681548409452,
"loss": 5.2764,
"mean_token_accuracy": 0.18648878186941148,
"num_tokens": 13760279.0,
"step": 6620
},
{
"entropy": 5.591698074340821,
"epoch": 0.6009615384615384,
"grad_norm": 0.96484375,
"learning_rate": 0.0004970629461711846,
"loss": 5.4861,
"mean_token_accuracy": 0.1701574757695198,
"num_tokens": 13770575.0,
"step": 6625
},
{
"entropy": 5.536564922332763,
"epoch": 0.6014150943396226,
"grad_norm": 0.875,
"learning_rate": 0.0004970577329091057,
"loss": 5.326,
"mean_token_accuracy": 0.18177662342786788,
"num_tokens": 13781068.0,
"step": 6630
},
{
"entropy": 5.604051685333252,
"epoch": 0.6018686502177069,
"grad_norm": 1.015625,
"learning_rate": 0.0004970525150548162,
"loss": 5.4189,
"mean_token_accuracy": 0.17805569767951965,
"num_tokens": 13790745.0,
"step": 6635
},
{
"entropy": 5.590482139587403,
"epoch": 0.602322206095791,
"grad_norm": 1.0625,
"learning_rate": 0.0004970472926084242,
"loss": 5.3506,
"mean_token_accuracy": 0.17441321313381195,
"num_tokens": 13800196.0,
"step": 6640
},
{
"entropy": 5.5683002948760985,
"epoch": 0.6027757619738752,
"grad_norm": 0.88671875,
"learning_rate": 0.0004970420655700375,
"loss": 5.2585,
"mean_token_accuracy": 0.18444105684757234,
"num_tokens": 13809954.0,
"step": 6645
},
{
"entropy": 5.542122173309326,
"epoch": 0.6032293178519593,
"grad_norm": 0.96875,
"learning_rate": 0.0004970368339397643,
"loss": 5.436,
"mean_token_accuracy": 0.17155808210372925,
"num_tokens": 13820281.0,
"step": 6650
},
{
"entropy": 5.6209453582763675,
"epoch": 0.6036828737300436,
"grad_norm": 0.83203125,
"learning_rate": 0.0004970315977177129,
"loss": 5.3867,
"mean_token_accuracy": 0.178961344063282,
"num_tokens": 13830614.0,
"step": 6655
},
{
"entropy": 5.566452884674073,
"epoch": 0.6041364296081277,
"grad_norm": 0.9453125,
"learning_rate": 0.0004970263569039914,
"loss": 5.3345,
"mean_token_accuracy": 0.1751037433743477,
"num_tokens": 13840551.0,
"step": 6660
},
{
"entropy": 5.64999303817749,
"epoch": 0.6045899854862119,
"grad_norm": 0.87890625,
"learning_rate": 0.0004970211114987083,
"loss": 5.5084,
"mean_token_accuracy": 0.17182745039463043,
"num_tokens": 13851830.0,
"step": 6665
},
{
"entropy": 5.598976469039917,
"epoch": 0.6050435413642961,
"grad_norm": 0.953125,
"learning_rate": 0.0004970158615019719,
"loss": 5.3166,
"mean_token_accuracy": 0.17749295830726625,
"num_tokens": 13862299.0,
"step": 6670
},
{
"entropy": 5.650000429153442,
"epoch": 0.6054970972423802,
"grad_norm": 0.8828125,
"learning_rate": 0.0004970106069138912,
"loss": 5.4553,
"mean_token_accuracy": 0.16745540499687195,
"num_tokens": 13872852.0,
"step": 6675
},
{
"entropy": 5.541075944900513,
"epoch": 0.6059506531204645,
"grad_norm": 1.171875,
"learning_rate": 0.0004970053477345743,
"loss": 5.3897,
"mean_token_accuracy": 0.18472966998815538,
"num_tokens": 13883601.0,
"step": 6680
},
{
"entropy": 5.573489952087402,
"epoch": 0.6064042089985486,
"grad_norm": 0.99609375,
"learning_rate": 0.0004970000839641304,
"loss": 5.3771,
"mean_token_accuracy": 0.1835797294974327,
"num_tokens": 13894246.0,
"step": 6685
},
{
"entropy": 5.5467630386352536,
"epoch": 0.6068577648766328,
"grad_norm": 0.91015625,
"learning_rate": 0.0004969948156026681,
"loss": 5.4114,
"mean_token_accuracy": 0.1743486538529396,
"num_tokens": 13904823.0,
"step": 6690
},
{
"entropy": 5.654760503768921,
"epoch": 0.6073113207547169,
"grad_norm": 0.93359375,
"learning_rate": 0.0004969895426502965,
"loss": 5.4214,
"mean_token_accuracy": 0.17350844889879227,
"num_tokens": 13915338.0,
"step": 6695
},
{
"entropy": 5.578554439544678,
"epoch": 0.6077648766328012,
"grad_norm": 0.9453125,
"learning_rate": 0.0004969842651071246,
"loss": 5.3676,
"mean_token_accuracy": 0.18019723445177077,
"num_tokens": 13925838.0,
"step": 6700
},
{
"entropy": 5.63786473274231,
"epoch": 0.6082184325108854,
"grad_norm": 0.94921875,
"learning_rate": 0.0004969789829732615,
"loss": 5.3876,
"mean_token_accuracy": 0.1755361571907997,
"num_tokens": 13936341.0,
"step": 6705
},
{
"entropy": 5.589788913726807,
"epoch": 0.6086719883889695,
"grad_norm": 1.0,
"learning_rate": 0.0004969736962488164,
"loss": 5.3615,
"mean_token_accuracy": 0.17749459445476531,
"num_tokens": 13945438.0,
"step": 6710
},
{
"entropy": 5.552593803405761,
"epoch": 0.6091255442670537,
"grad_norm": 0.875,
"learning_rate": 0.0004969684049338987,
"loss": 5.27,
"mean_token_accuracy": 0.1792834594845772,
"num_tokens": 13956029.0,
"step": 6715
},
{
"entropy": 5.6268645286560055,
"epoch": 0.6095791001451378,
"grad_norm": 0.9140625,
"learning_rate": 0.000496963109028618,
"loss": 5.3858,
"mean_token_accuracy": 0.1727551519870758,
"num_tokens": 13966498.0,
"step": 6720
},
{
"entropy": 5.575082731246948,
"epoch": 0.6100326560232221,
"grad_norm": 0.90234375,
"learning_rate": 0.0004969578085330834,
"loss": 5.3385,
"mean_token_accuracy": 0.17911534309387206,
"num_tokens": 13976997.0,
"step": 6725
},
{
"entropy": 5.628406524658203,
"epoch": 0.6104862119013063,
"grad_norm": 0.98828125,
"learning_rate": 0.0004969525034474049,
"loss": 5.3492,
"mean_token_accuracy": 0.1798671528697014,
"num_tokens": 13986664.0,
"step": 6730
},
{
"entropy": 5.485960388183594,
"epoch": 0.6109397677793904,
"grad_norm": 0.86328125,
"learning_rate": 0.000496947193771692,
"loss": 5.3458,
"mean_token_accuracy": 0.18172517567873,
"num_tokens": 13997828.0,
"step": 6735
},
{
"entropy": 5.5296248435974125,
"epoch": 0.6113933236574746,
"grad_norm": 1.015625,
"learning_rate": 0.0004969418795060548,
"loss": 5.2938,
"mean_token_accuracy": 0.17550018876791001,
"num_tokens": 14008293.0,
"step": 6740
},
{
"entropy": 5.513893079757691,
"epoch": 0.6118468795355588,
"grad_norm": 0.9609375,
"learning_rate": 0.0004969365606506027,
"loss": 5.2821,
"mean_token_accuracy": 0.1838718071579933,
"num_tokens": 14018031.0,
"step": 6745
},
{
"entropy": 5.516136503219604,
"epoch": 0.612300435413643,
"grad_norm": 0.91796875,
"learning_rate": 0.0004969312372054461,
"loss": 5.2563,
"mean_token_accuracy": 0.18612198829650878,
"num_tokens": 14028013.0,
"step": 6750
},
{
"entropy": 5.543424034118653,
"epoch": 0.6127539912917271,
"grad_norm": 0.87890625,
"learning_rate": 0.0004969259091706948,
"loss": 5.3988,
"mean_token_accuracy": 0.17460741549730302,
"num_tokens": 14039787.0,
"step": 6755
},
{
"entropy": 5.577391767501831,
"epoch": 0.6132075471698113,
"grad_norm": 0.89453125,
"learning_rate": 0.0004969205765464593,
"loss": 5.3011,
"mean_token_accuracy": 0.17705341130495073,
"num_tokens": 14049223.0,
"step": 6760
},
{
"entropy": 5.564587926864624,
"epoch": 0.6136611030478955,
"grad_norm": 0.9453125,
"learning_rate": 0.0004969152393328497,
"loss": 5.3386,
"mean_token_accuracy": 0.1833565875887871,
"num_tokens": 14059004.0,
"step": 6765
},
{
"entropy": 5.589889240264893,
"epoch": 0.6141146589259797,
"grad_norm": 0.91796875,
"learning_rate": 0.0004969098975299764,
"loss": 5.433,
"mean_token_accuracy": 0.17130700796842574,
"num_tokens": 14069615.0,
"step": 6770
},
{
"entropy": 5.584448528289795,
"epoch": 0.6145682148040639,
"grad_norm": 0.92578125,
"learning_rate": 0.0004969045511379499,
"loss": 5.3674,
"mean_token_accuracy": 0.17663028091192245,
"num_tokens": 14078886.0,
"step": 6775
},
{
"entropy": 5.650965642929077,
"epoch": 0.615021770682148,
"grad_norm": 0.921875,
"learning_rate": 0.0004968992001568806,
"loss": 5.4354,
"mean_token_accuracy": 0.17254979014396668,
"num_tokens": 14089021.0,
"step": 6780
},
{
"entropy": 5.535510587692261,
"epoch": 0.6154753265602322,
"grad_norm": 0.9375,
"learning_rate": 0.0004968938445868795,
"loss": 5.2588,
"mean_token_accuracy": 0.18511583507061005,
"num_tokens": 14098316.0,
"step": 6785
},
{
"entropy": 5.4947816848754885,
"epoch": 0.6159288824383164,
"grad_norm": 1.0390625,
"learning_rate": 0.0004968884844280569,
"loss": 5.3341,
"mean_token_accuracy": 0.1732075110077858,
"num_tokens": 14108874.0,
"step": 6790
},
{
"entropy": 5.591892051696777,
"epoch": 0.6163824383164006,
"grad_norm": 0.8984375,
"learning_rate": 0.0004968831196805242,
"loss": 5.3912,
"mean_token_accuracy": 0.17782093733549117,
"num_tokens": 14119613.0,
"step": 6795
},
{
"entropy": 5.60273003578186,
"epoch": 0.6168359941944848,
"grad_norm": 0.8515625,
"learning_rate": 0.0004968777503443918,
"loss": 5.3499,
"mean_token_accuracy": 0.17755615413188935,
"num_tokens": 14129085.0,
"step": 6800
},
{
"entropy": 5.454257392883301,
"epoch": 0.6172895500725689,
"grad_norm": 0.98828125,
"learning_rate": 0.0004968723764197711,
"loss": 5.2412,
"mean_token_accuracy": 0.1916737213730812,
"num_tokens": 14138461.0,
"step": 6805
},
{
"entropy": 5.528428459167481,
"epoch": 0.6177431059506531,
"grad_norm": 0.8828125,
"learning_rate": 0.000496866997906773,
"loss": 5.3161,
"mean_token_accuracy": 0.18484087139368058,
"num_tokens": 14149126.0,
"step": 6810
},
{
"entropy": 5.5417094230651855,
"epoch": 0.6181966618287373,
"grad_norm": 0.84375,
"learning_rate": 0.0004968616148055089,
"loss": 5.3186,
"mean_token_accuracy": 0.18229269832372666,
"num_tokens": 14160068.0,
"step": 6815
},
{
"entropy": 5.5963634014129635,
"epoch": 0.6186502177068215,
"grad_norm": 0.91796875,
"learning_rate": 0.0004968562271160901,
"loss": 5.4037,
"mean_token_accuracy": 0.1814342111349106,
"num_tokens": 14170288.0,
"step": 6820
},
{
"entropy": 5.6165848731994625,
"epoch": 0.6191037735849056,
"grad_norm": 0.86328125,
"learning_rate": 0.000496850834838628,
"loss": 5.4088,
"mean_token_accuracy": 0.1776348978281021,
"num_tokens": 14181231.0,
"step": 6825
},
{
"entropy": 5.5837572574615475,
"epoch": 0.6195573294629898,
"grad_norm": 0.90234375,
"learning_rate": 0.0004968454379732342,
"loss": 5.3227,
"mean_token_accuracy": 0.18313741385936738,
"num_tokens": 14191715.0,
"step": 6830
},
{
"entropy": 5.502680969238281,
"epoch": 0.6200108853410741,
"grad_norm": 0.91015625,
"learning_rate": 0.00049684003652002,
"loss": 5.31,
"mean_token_accuracy": 0.17466335743665695,
"num_tokens": 14203416.0,
"step": 6835
},
{
"entropy": 5.496374702453613,
"epoch": 0.6204644412191582,
"grad_norm": 1.0234375,
"learning_rate": 0.0004968346304790973,
"loss": 5.2663,
"mean_token_accuracy": 0.18976710140705108,
"num_tokens": 14213658.0,
"step": 6840
},
{
"entropy": 5.611422061920166,
"epoch": 0.6209179970972424,
"grad_norm": 0.91796875,
"learning_rate": 0.0004968292198505781,
"loss": 5.449,
"mean_token_accuracy": 0.17389458268880845,
"num_tokens": 14223956.0,
"step": 6845
},
{
"entropy": 5.6339301586151125,
"epoch": 0.6213715529753265,
"grad_norm": 0.81640625,
"learning_rate": 0.000496823804634574,
"loss": 5.4206,
"mean_token_accuracy": 0.17541918456554412,
"num_tokens": 14236801.0,
"step": 6850
},
{
"entropy": 5.559810400009155,
"epoch": 0.6218251088534107,
"grad_norm": 0.9140625,
"learning_rate": 0.0004968183848311972,
"loss": 5.241,
"mean_token_accuracy": 0.1860935151576996,
"num_tokens": 14247382.0,
"step": 6855
},
{
"entropy": 5.591649961471558,
"epoch": 0.622278664731495,
"grad_norm": 0.8828125,
"learning_rate": 0.0004968129604405596,
"loss": 5.3308,
"mean_token_accuracy": 0.18287506848573684,
"num_tokens": 14258627.0,
"step": 6860
},
{
"entropy": 5.489435148239136,
"epoch": 0.6227322206095791,
"grad_norm": 0.94921875,
"learning_rate": 0.0004968075314627734,
"loss": 5.3319,
"mean_token_accuracy": 0.18305183500051497,
"num_tokens": 14269013.0,
"step": 6865
},
{
"entropy": 5.5313599586486815,
"epoch": 0.6231857764876633,
"grad_norm": 0.94921875,
"learning_rate": 0.000496802097897951,
"loss": 5.3212,
"mean_token_accuracy": 0.18085480630397796,
"num_tokens": 14279620.0,
"step": 6870
},
{
"entropy": 5.593470907211303,
"epoch": 0.6236393323657474,
"grad_norm": 0.84765625,
"learning_rate": 0.0004967966597462046,
"loss": 5.3233,
"mean_token_accuracy": 0.18621707558631898,
"num_tokens": 14289448.0,
"step": 6875
},
{
"entropy": 5.573109865188599,
"epoch": 0.6240928882438317,
"grad_norm": 0.875,
"learning_rate": 0.0004967912170076469,
"loss": 5.3223,
"mean_token_accuracy": 0.17920063734054564,
"num_tokens": 14299999.0,
"step": 6880
},
{
"entropy": 5.4895679473876955,
"epoch": 0.6245464441219158,
"grad_norm": 0.88671875,
"learning_rate": 0.0004967857696823901,
"loss": 5.2935,
"mean_token_accuracy": 0.18258029073476792,
"num_tokens": 14310384.0,
"step": 6885
},
{
"entropy": 5.6066499710083,
"epoch": 0.625,
"grad_norm": 0.9296875,
"learning_rate": 0.0004967803177705472,
"loss": 5.421,
"mean_token_accuracy": 0.17535581439733505,
"num_tokens": 14320644.0,
"step": 6890
},
{
"entropy": 5.613523244857788,
"epoch": 0.6254535558780842,
"grad_norm": 1.078125,
"learning_rate": 0.0004967748612722308,
"loss": 5.3771,
"mean_token_accuracy": 0.1768083542585373,
"num_tokens": 14330077.0,
"step": 6895
},
{
"entropy": 5.6285642147064205,
"epoch": 0.6259071117561683,
"grad_norm": 0.875,
"learning_rate": 0.0004967694001875538,
"loss": 5.4657,
"mean_token_accuracy": 0.17162772566080092,
"num_tokens": 14341549.0,
"step": 6900
},
{
"entropy": 5.5867125511169435,
"epoch": 0.6263606676342526,
"grad_norm": 0.94921875,
"learning_rate": 0.000496763934516629,
"loss": 5.3252,
"mean_token_accuracy": 0.18651908189058303,
"num_tokens": 14352146.0,
"step": 6905
},
{
"entropy": 5.49106731414795,
"epoch": 0.6268142235123367,
"grad_norm": 0.8984375,
"learning_rate": 0.0004967584642595696,
"loss": 5.3502,
"mean_token_accuracy": 0.1742198124527931,
"num_tokens": 14362521.0,
"step": 6910
},
{
"entropy": 5.613452339172364,
"epoch": 0.6272677793904209,
"grad_norm": 0.89453125,
"learning_rate": 0.0004967529894164887,
"loss": 5.3842,
"mean_token_accuracy": 0.1770589143037796,
"num_tokens": 14373216.0,
"step": 6915
},
{
"entropy": 5.554586124420166,
"epoch": 0.627721335268505,
"grad_norm": 0.92578125,
"learning_rate": 0.0004967475099874992,
"loss": 5.2975,
"mean_token_accuracy": 0.18457451462745667,
"num_tokens": 14383104.0,
"step": 6920
},
{
"entropy": 5.516722202301025,
"epoch": 0.6281748911465893,
"grad_norm": 0.91015625,
"learning_rate": 0.000496742025972715,
"loss": 5.3284,
"mean_token_accuracy": 0.19065247923135759,
"num_tokens": 14393854.0,
"step": 6925
},
{
"entropy": 5.571130990982056,
"epoch": 0.6286284470246735,
"grad_norm": 1.0625,
"learning_rate": 0.000496736537372249,
"loss": 5.2746,
"mean_token_accuracy": 0.18790257424116136,
"num_tokens": 14403213.0,
"step": 6930
},
{
"entropy": 5.6476743698120115,
"epoch": 0.6290820029027576,
"grad_norm": 1.0,
"learning_rate": 0.0004967310441862151,
"loss": 5.4889,
"mean_token_accuracy": 0.17305789291858673,
"num_tokens": 14412792.0,
"step": 6935
},
{
"entropy": 5.5903888702392575,
"epoch": 0.6295355587808418,
"grad_norm": 0.84375,
"learning_rate": 0.0004967255464147265,
"loss": 5.3434,
"mean_token_accuracy": 0.18606058061122893,
"num_tokens": 14423061.0,
"step": 6940
},
{
"entropy": 5.547424602508545,
"epoch": 0.6299891146589259,
"grad_norm": 0.98046875,
"learning_rate": 0.0004967200440578972,
"loss": 5.2801,
"mean_token_accuracy": 0.1876331478357315,
"num_tokens": 14433338.0,
"step": 6945
},
{
"entropy": 5.512516736984253,
"epoch": 0.6304426705370102,
"grad_norm": 0.9375,
"learning_rate": 0.0004967145371158408,
"loss": 5.282,
"mean_token_accuracy": 0.1900091901421547,
"num_tokens": 14444124.0,
"step": 6950
},
{
"entropy": 5.559875965118408,
"epoch": 0.6308962264150944,
"grad_norm": 0.9921875,
"learning_rate": 0.0004967090255886714,
"loss": 5.4059,
"mean_token_accuracy": 0.17928589284420013,
"num_tokens": 14453868.0,
"step": 6955
},
{
"entropy": 5.547641086578369,
"epoch": 0.6313497822931785,
"grad_norm": 0.85546875,
"learning_rate": 0.0004967035094765028,
"loss": 5.3001,
"mean_token_accuracy": 0.17467721551656723,
"num_tokens": 14465015.0,
"step": 6960
},
{
"entropy": 5.53034987449646,
"epoch": 0.6318033381712627,
"grad_norm": 0.95703125,
"learning_rate": 0.0004966979887794491,
"loss": 5.2563,
"mean_token_accuracy": 0.18848009556531906,
"num_tokens": 14475523.0,
"step": 6965
},
{
"entropy": 5.5402673244476315,
"epoch": 0.6322568940493469,
"grad_norm": 0.95703125,
"learning_rate": 0.0004966924634976245,
"loss": 5.3295,
"mean_token_accuracy": 0.18160206973552703,
"num_tokens": 14485523.0,
"step": 6970
},
{
"entropy": 5.61271300315857,
"epoch": 0.6327104499274311,
"grad_norm": 0.859375,
"learning_rate": 0.0004966869336311433,
"loss": 5.4094,
"mean_token_accuracy": 0.17774551808834077,
"num_tokens": 14495237.0,
"step": 6975
},
{
"entropy": 5.5918501853942875,
"epoch": 0.6331640058055152,
"grad_norm": 0.91015625,
"learning_rate": 0.0004966813991801197,
"loss": 5.5141,
"mean_token_accuracy": 0.17085262537002563,
"num_tokens": 14506431.0,
"step": 6980
},
{
"entropy": 5.62278151512146,
"epoch": 0.6336175616835994,
"grad_norm": 0.90234375,
"learning_rate": 0.0004966758601446684,
"loss": 5.4451,
"mean_token_accuracy": 0.17433076500892639,
"num_tokens": 14517445.0,
"step": 6985
},
{
"entropy": 5.636280870437622,
"epoch": 0.6340711175616836,
"grad_norm": 0.92578125,
"learning_rate": 0.0004966703165249038,
"loss": 5.3707,
"mean_token_accuracy": 0.1714874342083931,
"num_tokens": 14527818.0,
"step": 6990
},
{
"entropy": 5.606985330581665,
"epoch": 0.6345246734397678,
"grad_norm": 0.90234375,
"learning_rate": 0.0004966647683209406,
"loss": 5.4616,
"mean_token_accuracy": 0.16598857343196868,
"num_tokens": 14538235.0,
"step": 6995
},
{
"entropy": 5.69085431098938,
"epoch": 0.634978229317852,
"grad_norm": 0.88671875,
"learning_rate": 0.0004966592155328935,
"loss": 5.4577,
"mean_token_accuracy": 0.17015783339738846,
"num_tokens": 14549784.0,
"step": 7000
},
{
"entropy": 5.577909851074219,
"epoch": 0.6354317851959361,
"grad_norm": 0.8203125,
"learning_rate": 0.0004966536581608771,
"loss": 5.4601,
"mean_token_accuracy": 0.16502557843923568,
"num_tokens": 14561057.0,
"step": 7005
},
{
"entropy": 5.6043314933776855,
"epoch": 0.6358853410740203,
"grad_norm": 0.94921875,
"learning_rate": 0.0004966480962050069,
"loss": 5.3663,
"mean_token_accuracy": 0.17885040044784545,
"num_tokens": 14570859.0,
"step": 7010
},
{
"entropy": 5.596745586395263,
"epoch": 0.6363388969521045,
"grad_norm": 0.953125,
"learning_rate": 0.0004966425296653974,
"loss": 5.3727,
"mean_token_accuracy": 0.18546929955482483,
"num_tokens": 14581066.0,
"step": 7015
},
{
"entropy": 5.507768726348877,
"epoch": 0.6367924528301887,
"grad_norm": 0.8984375,
"learning_rate": 0.000496636958542164,
"loss": 5.3266,
"mean_token_accuracy": 0.17755770236253737,
"num_tokens": 14590709.0,
"step": 7020
},
{
"entropy": 5.553393363952637,
"epoch": 0.6372460087082729,
"grad_norm": 1.015625,
"learning_rate": 0.0004966313828354218,
"loss": 5.3268,
"mean_token_accuracy": 0.184244504570961,
"num_tokens": 14600849.0,
"step": 7025
},
{
"entropy": 5.520700263977051,
"epoch": 0.637699564586357,
"grad_norm": 0.97265625,
"learning_rate": 0.000496625802545286,
"loss": 5.2116,
"mean_token_accuracy": 0.1859947547316551,
"num_tokens": 14611427.0,
"step": 7030
},
{
"entropy": 5.530115747451783,
"epoch": 0.6381531204644412,
"grad_norm": 0.94921875,
"learning_rate": 0.0004966202176718723,
"loss": 5.3635,
"mean_token_accuracy": 0.182247893512249,
"num_tokens": 14623860.0,
"step": 7035
},
{
"entropy": 5.564819049835205,
"epoch": 0.6386066763425254,
"grad_norm": 0.9609375,
"learning_rate": 0.0004966146282152959,
"loss": 5.2998,
"mean_token_accuracy": 0.1783895254135132,
"num_tokens": 14633481.0,
"step": 7040
},
{
"entropy": 5.564633512496949,
"epoch": 0.6390602322206096,
"grad_norm": 0.90234375,
"learning_rate": 0.0004966090341756726,
"loss": 5.2811,
"mean_token_accuracy": 0.18869292736053467,
"num_tokens": 14643519.0,
"step": 7045
},
{
"entropy": 5.450642681121826,
"epoch": 0.6395137880986937,
"grad_norm": 0.94140625,
"learning_rate": 0.000496603435553118,
"loss": 5.2722,
"mean_token_accuracy": 0.18621875643730162,
"num_tokens": 14654570.0,
"step": 7050
},
{
"entropy": 5.501469135284424,
"epoch": 0.6399673439767779,
"grad_norm": 0.94921875,
"learning_rate": 0.0004965978323477478,
"loss": 5.3098,
"mean_token_accuracy": 0.17667924016714096,
"num_tokens": 14665596.0,
"step": 7055
},
{
"entropy": 5.580057382583618,
"epoch": 0.6404208998548622,
"grad_norm": 0.97265625,
"learning_rate": 0.000496592224559678,
"loss": 5.2669,
"mean_token_accuracy": 0.18460346013307571,
"num_tokens": 14675660.0,
"step": 7060
},
{
"entropy": 5.580855798721314,
"epoch": 0.6408744557329463,
"grad_norm": 0.984375,
"learning_rate": 0.0004965866121890245,
"loss": 5.4209,
"mean_token_accuracy": 0.17853408306837082,
"num_tokens": 14685229.0,
"step": 7065
},
{
"entropy": 5.619711685180664,
"epoch": 0.6413280116110305,
"grad_norm": 1.015625,
"learning_rate": 0.0004965809952359034,
"loss": 5.3312,
"mean_token_accuracy": 0.17109032422304155,
"num_tokens": 14695672.0,
"step": 7070
},
{
"entropy": 5.563270711898804,
"epoch": 0.6417815674891146,
"grad_norm": 0.91015625,
"learning_rate": 0.0004965753737004309,
"loss": 5.2857,
"mean_token_accuracy": 0.18523964136838914,
"num_tokens": 14705160.0,
"step": 7075
},
{
"entropy": 5.51725435256958,
"epoch": 0.6422351233671988,
"grad_norm": 0.9609375,
"learning_rate": 0.0004965697475827231,
"loss": 5.3204,
"mean_token_accuracy": 0.1778504505753517,
"num_tokens": 14715487.0,
"step": 7080
},
{
"entropy": 5.580355167388916,
"epoch": 0.6426886792452831,
"grad_norm": 1.0,
"learning_rate": 0.0004965641168828965,
"loss": 5.3743,
"mean_token_accuracy": 0.18160596191883088,
"num_tokens": 14726987.0,
"step": 7085
},
{
"entropy": 5.494480085372925,
"epoch": 0.6431422351233672,
"grad_norm": 0.9296875,
"learning_rate": 0.0004965584816010676,
"loss": 5.3176,
"mean_token_accuracy": 0.1852257654070854,
"num_tokens": 14737810.0,
"step": 7090
},
{
"entropy": 5.610119819641113,
"epoch": 0.6435957910014514,
"grad_norm": 0.89453125,
"learning_rate": 0.0004965528417373528,
"loss": 5.3285,
"mean_token_accuracy": 0.17879804968833923,
"num_tokens": 14748997.0,
"step": 7095
},
{
"entropy": 5.566685581207276,
"epoch": 0.6440493468795355,
"grad_norm": 0.86328125,
"learning_rate": 0.0004965471972918688,
"loss": 5.3404,
"mean_token_accuracy": 0.1756565198302269,
"num_tokens": 14760439.0,
"step": 7100
},
{
"entropy": 5.460477256774903,
"epoch": 0.6445029027576198,
"grad_norm": 0.87109375,
"learning_rate": 0.0004965415482647322,
"loss": 5.2521,
"mean_token_accuracy": 0.18281868398189544,
"num_tokens": 14770828.0,
"step": 7105
},
{
"entropy": 5.489846992492676,
"epoch": 0.6449564586357039,
"grad_norm": 0.9140625,
"learning_rate": 0.00049653589465606,
"loss": 5.3957,
"mean_token_accuracy": 0.18351997137069703,
"num_tokens": 14781753.0,
"step": 7110
},
{
"entropy": 5.6520702838897705,
"epoch": 0.6454100145137881,
"grad_norm": 1.0078125,
"learning_rate": 0.000496530236465969,
"loss": 5.4102,
"mean_token_accuracy": 0.17198631465435027,
"num_tokens": 14792241.0,
"step": 7115
},
{
"entropy": 5.531635856628418,
"epoch": 0.6458635703918723,
"grad_norm": 0.92578125,
"learning_rate": 0.0004965245736945763,
"loss": 5.2285,
"mean_token_accuracy": 0.1889130190014839,
"num_tokens": 14802580.0,
"step": 7120
},
{
"entropy": 5.519139194488526,
"epoch": 0.6463171262699564,
"grad_norm": 0.94140625,
"learning_rate": 0.0004965189063419989,
"loss": 5.442,
"mean_token_accuracy": 0.17884785532951356,
"num_tokens": 14813103.0,
"step": 7125
},
{
"entropy": 5.568569326400757,
"epoch": 0.6467706821480407,
"grad_norm": 1.015625,
"learning_rate": 0.0004965132344083541,
"loss": 5.3276,
"mean_token_accuracy": 0.17959365546703338,
"num_tokens": 14823334.0,
"step": 7130
},
{
"entropy": 5.568736124038696,
"epoch": 0.6472242380261248,
"grad_norm": 0.875,
"learning_rate": 0.0004965075578937593,
"loss": 5.3492,
"mean_token_accuracy": 0.17829880565404893,
"num_tokens": 14834850.0,
"step": 7135
},
{
"entropy": 5.513080644607544,
"epoch": 0.647677793904209,
"grad_norm": 0.99609375,
"learning_rate": 0.0004965018767983316,
"loss": 5.3575,
"mean_token_accuracy": 0.1796330839395523,
"num_tokens": 14844276.0,
"step": 7140
},
{
"entropy": 5.4893872261047365,
"epoch": 0.6481313497822931,
"grad_norm": 0.9375,
"learning_rate": 0.0004964961911221886,
"loss": 5.1964,
"mean_token_accuracy": 0.1915486693382263,
"num_tokens": 14854352.0,
"step": 7145
},
{
"entropy": 5.584615421295166,
"epoch": 0.6485849056603774,
"grad_norm": 0.94921875,
"learning_rate": 0.000496490500865448,
"loss": 5.378,
"mean_token_accuracy": 0.1796365812420845,
"num_tokens": 14864395.0,
"step": 7150
},
{
"entropy": 5.5620684146881105,
"epoch": 0.6490384615384616,
"grad_norm": 0.97265625,
"learning_rate": 0.0004964848060282275,
"loss": 5.2599,
"mean_token_accuracy": 0.18747452944517135,
"num_tokens": 14874373.0,
"step": 7155
},
{
"entropy": 5.53101077079773,
"epoch": 0.6494920174165457,
"grad_norm": 0.9375,
"learning_rate": 0.0004964791066106447,
"loss": 5.2642,
"mean_token_accuracy": 0.18159855753183365,
"num_tokens": 14884932.0,
"step": 7160
},
{
"entropy": 5.5659397602081295,
"epoch": 0.6499455732946299,
"grad_norm": 0.87890625,
"learning_rate": 0.0004964734026128175,
"loss": 5.4253,
"mean_token_accuracy": 0.17503994554281235,
"num_tokens": 14896254.0,
"step": 7165
},
{
"entropy": 5.623956346511841,
"epoch": 0.650399129172714,
"grad_norm": 0.9375,
"learning_rate": 0.0004964676940348638,
"loss": 5.3453,
"mean_token_accuracy": 0.17856290340423583,
"num_tokens": 14907392.0,
"step": 7170
},
{
"entropy": 5.5940752029418945,
"epoch": 0.6508526850507983,
"grad_norm": 0.94140625,
"learning_rate": 0.000496461980876902,
"loss": 5.4614,
"mean_token_accuracy": 0.17526094764471054,
"num_tokens": 14917676.0,
"step": 7175
},
{
"entropy": 5.5578306198120115,
"epoch": 0.6513062409288825,
"grad_norm": 0.92578125,
"learning_rate": 0.0004964562631390498,
"loss": 5.3008,
"mean_token_accuracy": 0.18212647140026092,
"num_tokens": 14928500.0,
"step": 7180
},
{
"entropy": 5.551118040084839,
"epoch": 0.6517597968069666,
"grad_norm": 0.9140625,
"learning_rate": 0.0004964505408214257,
"loss": 5.2703,
"mean_token_accuracy": 0.18634399622678757,
"num_tokens": 14938970.0,
"step": 7185
},
{
"entropy": 5.526061344146728,
"epoch": 0.6522133526850508,
"grad_norm": 0.953125,
"learning_rate": 0.000496444813924148,
"loss": 5.3255,
"mean_token_accuracy": 0.18173780143260956,
"num_tokens": 14949017.0,
"step": 7190
},
{
"entropy": 5.52117953300476,
"epoch": 0.652666908563135,
"grad_norm": 0.94140625,
"learning_rate": 0.0004964390824473351,
"loss": 5.2443,
"mean_token_accuracy": 0.19174836874008178,
"num_tokens": 14958529.0,
"step": 7195
},
{
"entropy": 5.527310991287232,
"epoch": 0.6531204644412192,
"grad_norm": 0.9453125,
"learning_rate": 0.0004964333463911056,
"loss": 5.3604,
"mean_token_accuracy": 0.17524895071983337,
"num_tokens": 14968044.0,
"step": 7200
},
{
"entropy": 5.596441411972046,
"epoch": 0.6535740203193033,
"grad_norm": 0.92578125,
"learning_rate": 0.0004964276057555779,
"loss": 5.3771,
"mean_token_accuracy": 0.17824496030807496,
"num_tokens": 14977867.0,
"step": 7205
},
{
"entropy": 5.6655659675598145,
"epoch": 0.6540275761973875,
"grad_norm": 1.015625,
"learning_rate": 0.000496421860540871,
"loss": 5.3958,
"mean_token_accuracy": 0.1837482750415802,
"num_tokens": 14986936.0,
"step": 7210
},
{
"entropy": 5.506159448623658,
"epoch": 0.6544811320754716,
"grad_norm": 0.9921875,
"learning_rate": 0.0004964161107471036,
"loss": 5.3354,
"mean_token_accuracy": 0.18265933394432068,
"num_tokens": 14996681.0,
"step": 7215
},
{
"entropy": 5.5519446849823,
"epoch": 0.6549346879535559,
"grad_norm": 0.91015625,
"learning_rate": 0.0004964103563743946,
"loss": 5.3364,
"mean_token_accuracy": 0.18156073987483978,
"num_tokens": 15007100.0,
"step": 7220
},
{
"entropy": 5.5976849555969235,
"epoch": 0.6553882438316401,
"grad_norm": 0.984375,
"learning_rate": 0.000496404597422863,
"loss": 5.4102,
"mean_token_accuracy": 0.18126775324344635,
"num_tokens": 15017452.0,
"step": 7225
},
{
"entropy": 5.539013767242432,
"epoch": 0.6558417997097242,
"grad_norm": 0.921875,
"learning_rate": 0.000496398833892628,
"loss": 5.254,
"mean_token_accuracy": 0.18298355340957642,
"num_tokens": 15028362.0,
"step": 7230
},
{
"entropy": 5.589620161056518,
"epoch": 0.6562953555878084,
"grad_norm": 0.96484375,
"learning_rate": 0.0004963930657838086,
"loss": 5.3474,
"mean_token_accuracy": 0.17923036962747574,
"num_tokens": 15038691.0,
"step": 7235
},
{
"entropy": 5.547629928588867,
"epoch": 0.6567489114658926,
"grad_norm": 0.91796875,
"learning_rate": 0.0004963872930965242,
"loss": 5.3283,
"mean_token_accuracy": 0.1816704109311104,
"num_tokens": 15050037.0,
"step": 7240
},
{
"entropy": 5.559830236434936,
"epoch": 0.6572024673439768,
"grad_norm": 0.9921875,
"learning_rate": 0.0004963815158308941,
"loss": 5.3057,
"mean_token_accuracy": 0.18500083684921265,
"num_tokens": 15059793.0,
"step": 7245
},
{
"entropy": 5.537407398223877,
"epoch": 0.657656023222061,
"grad_norm": 0.94921875,
"learning_rate": 0.0004963757339870379,
"loss": 5.3896,
"mean_token_accuracy": 0.17124081701040267,
"num_tokens": 15070500.0,
"step": 7250
},
{
"entropy": 5.612113428115845,
"epoch": 0.6581095791001451,
"grad_norm": 0.90625,
"learning_rate": 0.0004963699475650751,
"loss": 5.4467,
"mean_token_accuracy": 0.17665672302246094,
"num_tokens": 15081109.0,
"step": 7255
},
{
"entropy": 5.5193017482757565,
"epoch": 0.6585631349782293,
"grad_norm": 0.95703125,
"learning_rate": 0.0004963641565651255,
"loss": 5.323,
"mean_token_accuracy": 0.18454677164554595,
"num_tokens": 15092315.0,
"step": 7260
},
{
"entropy": 5.568742418289185,
"epoch": 0.6590166908563135,
"grad_norm": 0.90234375,
"learning_rate": 0.0004963583609873084,
"loss": 5.3849,
"mean_token_accuracy": 0.1800144374370575,
"num_tokens": 15103957.0,
"step": 7265
},
{
"entropy": 5.6056946277618405,
"epoch": 0.6594702467343977,
"grad_norm": 0.88671875,
"learning_rate": 0.0004963525608317443,
"loss": 5.3721,
"mean_token_accuracy": 0.1770993635058403,
"num_tokens": 15114996.0,
"step": 7270
},
{
"entropy": 5.574876546859741,
"epoch": 0.6599238026124818,
"grad_norm": 0.9609375,
"learning_rate": 0.0004963467560985527,
"loss": 5.3184,
"mean_token_accuracy": 0.1811333030462265,
"num_tokens": 15125098.0,
"step": 7275
},
{
"entropy": 5.508492565155029,
"epoch": 0.660377358490566,
"grad_norm": 0.99609375,
"learning_rate": 0.0004963409467878538,
"loss": 5.3165,
"mean_token_accuracy": 0.17881869971752168,
"num_tokens": 15134746.0,
"step": 7280
},
{
"entropy": 5.631923818588257,
"epoch": 0.6608309143686503,
"grad_norm": 1.0234375,
"learning_rate": 0.0004963351328997678,
"loss": 5.3384,
"mean_token_accuracy": 0.17652247250080108,
"num_tokens": 15145698.0,
"step": 7285
},
{
"entropy": 5.570781326293945,
"epoch": 0.6612844702467344,
"grad_norm": 0.96484375,
"learning_rate": 0.0004963293144344148,
"loss": 5.3904,
"mean_token_accuracy": 0.17904452234506607,
"num_tokens": 15156026.0,
"step": 7290
},
{
"entropy": 5.555085945129394,
"epoch": 0.6617380261248186,
"grad_norm": 0.953125,
"learning_rate": 0.0004963234913919151,
"loss": 5.3322,
"mean_token_accuracy": 0.17398147583007811,
"num_tokens": 15165900.0,
"step": 7295
},
{
"entropy": 5.565875434875489,
"epoch": 0.6621915820029027,
"grad_norm": 0.90234375,
"learning_rate": 0.0004963176637723893,
"loss": 5.3794,
"mean_token_accuracy": 0.17941929996013642,
"num_tokens": 15176693.0,
"step": 7300
},
{
"entropy": 5.58468017578125,
"epoch": 0.6626451378809869,
"grad_norm": 0.93359375,
"learning_rate": 0.0004963118315759578,
"loss": 5.4007,
"mean_token_accuracy": 0.16982991248369217,
"num_tokens": 15186535.0,
"step": 7305
},
{
"entropy": 5.587566995620728,
"epoch": 0.6630986937590712,
"grad_norm": 1.015625,
"learning_rate": 0.0004963059948027411,
"loss": 5.3162,
"mean_token_accuracy": 0.1796189233660698,
"num_tokens": 15196019.0,
"step": 7310
},
{
"entropy": 5.465688180923462,
"epoch": 0.6635522496371553,
"grad_norm": 1.015625,
"learning_rate": 0.0004963001534528601,
"loss": 5.1812,
"mean_token_accuracy": 0.19228140413761138,
"num_tokens": 15205273.0,
"step": 7315
},
{
"entropy": 5.560149812698365,
"epoch": 0.6640058055152395,
"grad_norm": 0.921875,
"learning_rate": 0.0004962943075264355,
"loss": 5.3427,
"mean_token_accuracy": 0.1783326655626297,
"num_tokens": 15216716.0,
"step": 7320
},
{
"entropy": 5.571094942092896,
"epoch": 0.6644593613933236,
"grad_norm": 0.890625,
"learning_rate": 0.0004962884570235884,
"loss": 5.3928,
"mean_token_accuracy": 0.17626622021198274,
"num_tokens": 15227306.0,
"step": 7325
},
{
"entropy": 5.510771894454956,
"epoch": 0.6649129172714079,
"grad_norm": 0.91015625,
"learning_rate": 0.0004962826019444394,
"loss": 5.2786,
"mean_token_accuracy": 0.18220029771327972,
"num_tokens": 15237302.0,
"step": 7330
},
{
"entropy": 5.515999460220337,
"epoch": 0.665366473149492,
"grad_norm": 0.90625,
"learning_rate": 0.0004962767422891099,
"loss": 5.2859,
"mean_token_accuracy": 0.1832537442445755,
"num_tokens": 15247064.0,
"step": 7335
},
{
"entropy": 5.559939670562744,
"epoch": 0.6658200290275762,
"grad_norm": 0.890625,
"learning_rate": 0.000496270878057721,
"loss": 5.3565,
"mean_token_accuracy": 0.18223199248313904,
"num_tokens": 15257006.0,
"step": 7340
},
{
"entropy": 5.5652186393737795,
"epoch": 0.6662735849056604,
"grad_norm": 0.96875,
"learning_rate": 0.0004962650092503939,
"loss": 5.3333,
"mean_token_accuracy": 0.1857373908162117,
"num_tokens": 15266720.0,
"step": 7345
},
{
"entropy": 5.540846633911133,
"epoch": 0.6667271407837445,
"grad_norm": 0.93359375,
"learning_rate": 0.0004962591358672502,
"loss": 5.3651,
"mean_token_accuracy": 0.17481739670038224,
"num_tokens": 15277330.0,
"step": 7350
},
{
"entropy": 5.548562622070312,
"epoch": 0.6671806966618288,
"grad_norm": 0.8828125,
"learning_rate": 0.000496253257908411,
"loss": 5.2984,
"mean_token_accuracy": 0.1819293111562729,
"num_tokens": 15287401.0,
"step": 7355
},
{
"entropy": 5.485146522521973,
"epoch": 0.6676342525399129,
"grad_norm": 0.90625,
"learning_rate": 0.0004962473753739981,
"loss": 5.2411,
"mean_token_accuracy": 0.18803683817386627,
"num_tokens": 15297426.0,
"step": 7360
},
{
"entropy": 5.565172481536865,
"epoch": 0.6680878084179971,
"grad_norm": 0.93359375,
"learning_rate": 0.000496241488264133,
"loss": 5.2356,
"mean_token_accuracy": 0.1826209083199501,
"num_tokens": 15308412.0,
"step": 7365
},
{
"entropy": 5.527874851226807,
"epoch": 0.6685413642960812,
"grad_norm": 1.0,
"learning_rate": 0.0004962355965789377,
"loss": 5.2316,
"mean_token_accuracy": 0.17745898216962813,
"num_tokens": 15318821.0,
"step": 7370
},
{
"entropy": 5.4749537944793705,
"epoch": 0.6689949201741655,
"grad_norm": 0.984375,
"learning_rate": 0.0004962297003185338,
"loss": 5.3039,
"mean_token_accuracy": 0.18267119228839873,
"num_tokens": 15328737.0,
"step": 7375
},
{
"entropy": 5.598972654342651,
"epoch": 0.6694484760522497,
"grad_norm": 1.0234375,
"learning_rate": 0.0004962237994830433,
"loss": 5.411,
"mean_token_accuracy": 0.17150809168815612,
"num_tokens": 15339695.0,
"step": 7380
},
{
"entropy": 5.633320331573486,
"epoch": 0.6699020319303338,
"grad_norm": 0.99609375,
"learning_rate": 0.0004962178940725883,
"loss": 5.3667,
"mean_token_accuracy": 0.17928486466407775,
"num_tokens": 15350270.0,
"step": 7385
},
{
"entropy": 5.533000087738037,
"epoch": 0.670355587808418,
"grad_norm": 0.9453125,
"learning_rate": 0.0004962119840872908,
"loss": 5.367,
"mean_token_accuracy": 0.18240504264831542,
"num_tokens": 15360986.0,
"step": 7390
},
{
"entropy": 5.531470489501953,
"epoch": 0.6708091436865021,
"grad_norm": 0.87109375,
"learning_rate": 0.0004962060695272733,
"loss": 5.3501,
"mean_token_accuracy": 0.18166522681713104,
"num_tokens": 15371786.0,
"step": 7395
},
{
"entropy": 5.559743118286133,
"epoch": 0.6712626995645864,
"grad_norm": 0.94921875,
"learning_rate": 0.0004962001503926577,
"loss": 5.309,
"mean_token_accuracy": 0.18303101807832717,
"num_tokens": 15382960.0,
"step": 7400
},
{
"entropy": 5.551979637145996,
"epoch": 0.6717162554426706,
"grad_norm": 0.953125,
"learning_rate": 0.0004961942266835667,
"loss": 5.3077,
"mean_token_accuracy": 0.18894487917423247,
"num_tokens": 15392807.0,
"step": 7405
},
{
"entropy": 5.554458045959473,
"epoch": 0.6721698113207547,
"grad_norm": 0.97265625,
"learning_rate": 0.0004961882984001227,
"loss": 5.3527,
"mean_token_accuracy": 0.1719239681959152,
"num_tokens": 15403712.0,
"step": 7410
},
{
"entropy": 5.548364591598511,
"epoch": 0.6726233671988389,
"grad_norm": 0.9296875,
"learning_rate": 0.0004961823655424484,
"loss": 5.4027,
"mean_token_accuracy": 0.1698218837380409,
"num_tokens": 15414839.0,
"step": 7415
},
{
"entropy": 5.563115072250366,
"epoch": 0.6730769230769231,
"grad_norm": 0.9296875,
"learning_rate": 0.0004961764281106664,
"loss": 5.293,
"mean_token_accuracy": 0.18119843751192094,
"num_tokens": 15425320.0,
"step": 7420
},
{
"entropy": 5.544220876693726,
"epoch": 0.6735304789550073,
"grad_norm": 0.921875,
"learning_rate": 0.0004961704861048995,
"loss": 5.288,
"mean_token_accuracy": 0.1794519156217575,
"num_tokens": 15435476.0,
"step": 7425
},
{
"entropy": 5.5448249816894535,
"epoch": 0.6739840348330914,
"grad_norm": 0.953125,
"learning_rate": 0.0004961645395252705,
"loss": 5.4071,
"mean_token_accuracy": 0.1752161681652069,
"num_tokens": 15445786.0,
"step": 7430
},
{
"entropy": 5.524516010284424,
"epoch": 0.6744375907111756,
"grad_norm": 0.9765625,
"learning_rate": 0.0004961585883719025,
"loss": 5.3816,
"mean_token_accuracy": 0.1858324810862541,
"num_tokens": 15455915.0,
"step": 7435
},
{
"entropy": 5.600856065750122,
"epoch": 0.6748911465892597,
"grad_norm": 0.94921875,
"learning_rate": 0.0004961526326449186,
"loss": 5.3712,
"mean_token_accuracy": 0.17852281779050827,
"num_tokens": 15466523.0,
"step": 7440
},
{
"entropy": 5.600769948959351,
"epoch": 0.675344702467344,
"grad_norm": 0.96875,
"learning_rate": 0.0004961466723444419,
"loss": 5.3662,
"mean_token_accuracy": 0.17874794453382492,
"num_tokens": 15476821.0,
"step": 7445
},
{
"entropy": 5.514429330825806,
"epoch": 0.6757982583454282,
"grad_norm": 0.9921875,
"learning_rate": 0.0004961407074705955,
"loss": 5.2923,
"mean_token_accuracy": 0.17971670627593994,
"num_tokens": 15487182.0,
"step": 7450
},
{
"entropy": 5.549853372573852,
"epoch": 0.6762518142235123,
"grad_norm": 0.90234375,
"learning_rate": 0.0004961347380235031,
"loss": 5.2785,
"mean_token_accuracy": 0.18686819225549697,
"num_tokens": 15497600.0,
"step": 7455
},
{
"entropy": 5.616761255264282,
"epoch": 0.6767053701015965,
"grad_norm": 1.0,
"learning_rate": 0.0004961287640032879,
"loss": 5.4268,
"mean_token_accuracy": 0.17058165371418,
"num_tokens": 15508278.0,
"step": 7460
},
{
"entropy": 5.64322600364685,
"epoch": 0.6771589259796807,
"grad_norm": 1.0078125,
"learning_rate": 0.0004961227854100734,
"loss": 5.4467,
"mean_token_accuracy": 0.17577812671661378,
"num_tokens": 15518458.0,
"step": 7465
},
{
"entropy": 5.549437522888184,
"epoch": 0.6776124818577649,
"grad_norm": 0.9140625,
"learning_rate": 0.0004961168022439835,
"loss": 5.2543,
"mean_token_accuracy": 0.1862138107419014,
"num_tokens": 15528738.0,
"step": 7470
},
{
"entropy": 5.498688793182373,
"epoch": 0.6780660377358491,
"grad_norm": 0.95703125,
"learning_rate": 0.0004961108145051417,
"loss": 5.3016,
"mean_token_accuracy": 0.1857891008257866,
"num_tokens": 15538656.0,
"step": 7475
},
{
"entropy": 5.4875250339508055,
"epoch": 0.6785195936139332,
"grad_norm": 0.921875,
"learning_rate": 0.000496104822193672,
"loss": 5.3009,
"mean_token_accuracy": 0.18368335962295532,
"num_tokens": 15549285.0,
"step": 7480
},
{
"entropy": 5.609712409973144,
"epoch": 0.6789731494920174,
"grad_norm": 0.9453125,
"learning_rate": 0.0004960988253096982,
"loss": 5.3807,
"mean_token_accuracy": 0.1867237627506256,
"num_tokens": 15559766.0,
"step": 7485
},
{
"entropy": 5.533083581924439,
"epoch": 0.6794267053701016,
"grad_norm": 0.8671875,
"learning_rate": 0.0004960928238533443,
"loss": 5.2478,
"mean_token_accuracy": 0.18986229747533798,
"num_tokens": 15570873.0,
"step": 7490
},
{
"entropy": 5.5704052448272705,
"epoch": 0.6798802612481858,
"grad_norm": 1.0,
"learning_rate": 0.0004960868178247344,
"loss": 5.4891,
"mean_token_accuracy": 0.17788162529468537,
"num_tokens": 15581229.0,
"step": 7495
},
{
"entropy": 5.540935468673706,
"epoch": 0.68033381712627,
"grad_norm": 0.953125,
"learning_rate": 0.0004960808072239929,
"loss": 5.3557,
"mean_token_accuracy": 0.17538125514984132,
"num_tokens": 15591686.0,
"step": 7500
},
{
"entropy": 5.620772171020508,
"epoch": 0.6807873730043541,
"grad_norm": 0.9765625,
"learning_rate": 0.0004960747920512439,
"loss": 5.3755,
"mean_token_accuracy": 0.17827560752630234,
"num_tokens": 15601869.0,
"step": 7505
},
{
"entropy": 5.558009386062622,
"epoch": 0.6812409288824384,
"grad_norm": 0.8359375,
"learning_rate": 0.000496068772306612,
"loss": 5.3513,
"mean_token_accuracy": 0.17730835527181626,
"num_tokens": 15612946.0,
"step": 7510
},
{
"entropy": 5.558612203598022,
"epoch": 0.6816944847605225,
"grad_norm": 0.9375,
"learning_rate": 0.0004960627479902214,
"loss": 5.3976,
"mean_token_accuracy": 0.1726248562335968,
"num_tokens": 15623761.0,
"step": 7515
},
{
"entropy": 5.559710741043091,
"epoch": 0.6821480406386067,
"grad_norm": 0.93359375,
"learning_rate": 0.000496056719102197,
"loss": 5.3926,
"mean_token_accuracy": 0.17348738610744477,
"num_tokens": 15634500.0,
"step": 7520
},
{
"entropy": 5.487872171401977,
"epoch": 0.6826015965166908,
"grad_norm": 0.9140625,
"learning_rate": 0.0004960506856426631,
"loss": 5.2369,
"mean_token_accuracy": 0.18374142199754714,
"num_tokens": 15644896.0,
"step": 7525
},
{
"entropy": 5.5576379776000975,
"epoch": 0.683055152394775,
"grad_norm": 0.9296875,
"learning_rate": 0.0004960446476117448,
"loss": 5.3577,
"mean_token_accuracy": 0.1801358714699745,
"num_tokens": 15655008.0,
"step": 7530
},
{
"entropy": 5.5583741664886475,
"epoch": 0.6835087082728593,
"grad_norm": 0.91796875,
"learning_rate": 0.0004960386050095668,
"loss": 5.2713,
"mean_token_accuracy": 0.17511853575706482,
"num_tokens": 15665944.0,
"step": 7535
},
{
"entropy": 5.568644237518311,
"epoch": 0.6839622641509434,
"grad_norm": 0.921875,
"learning_rate": 0.0004960325578362543,
"loss": 5.3761,
"mean_token_accuracy": 0.1809030920267105,
"num_tokens": 15677099.0,
"step": 7540
},
{
"entropy": 5.601432275772095,
"epoch": 0.6844158200290276,
"grad_norm": 0.97265625,
"learning_rate": 0.000496026506091932,
"loss": 5.3376,
"mean_token_accuracy": 0.1795408010482788,
"num_tokens": 15687531.0,
"step": 7545
},
{
"entropy": 5.5705641746521,
"epoch": 0.6848693759071117,
"grad_norm": 1.0,
"learning_rate": 0.0004960204497767253,
"loss": 5.3587,
"mean_token_accuracy": 0.18146915584802628,
"num_tokens": 15697002.0,
"step": 7550
},
{
"entropy": 5.470752716064453,
"epoch": 0.685322931785196,
"grad_norm": 0.9375,
"learning_rate": 0.0004960143888907594,
"loss": 5.2251,
"mean_token_accuracy": 0.1819280982017517,
"num_tokens": 15707249.0,
"step": 7555
},
{
"entropy": 5.5658492088317875,
"epoch": 0.6857764876632801,
"grad_norm": 0.87890625,
"learning_rate": 0.0004960083234341595,
"loss": 5.3005,
"mean_token_accuracy": 0.18364818543195724,
"num_tokens": 15718764.0,
"step": 7560
},
{
"entropy": 5.512791872024536,
"epoch": 0.6862300435413643,
"grad_norm": 0.921875,
"learning_rate": 0.0004960022534070513,
"loss": 5.2503,
"mean_token_accuracy": 0.18830282539129256,
"num_tokens": 15728854.0,
"step": 7565
},
{
"entropy": 5.458533000946045,
"epoch": 0.6866835994194485,
"grad_norm": 0.90234375,
"learning_rate": 0.00049599617880956,
"loss": 5.2675,
"mean_token_accuracy": 0.18406953662633896,
"num_tokens": 15739682.0,
"step": 7570
},
{
"entropy": 5.505916738510132,
"epoch": 0.6871371552975326,
"grad_norm": 0.9765625,
"learning_rate": 0.0004959900996418115,
"loss": 5.2657,
"mean_token_accuracy": 0.1874595806002617,
"num_tokens": 15749628.0,
"step": 7575
},
{
"entropy": 5.530863189697266,
"epoch": 0.6875907111756169,
"grad_norm": 1.03125,
"learning_rate": 0.0004959840159039315,
"loss": 5.3281,
"mean_token_accuracy": 0.1795841246843338,
"num_tokens": 15760137.0,
"step": 7580
},
{
"entropy": 5.592649602890015,
"epoch": 0.688044267053701,
"grad_norm": 1.0078125,
"learning_rate": 0.0004959779275960456,
"loss": 5.2881,
"mean_token_accuracy": 0.18079988062381744,
"num_tokens": 15769889.0,
"step": 7585
},
{
"entropy": 5.537004804611206,
"epoch": 0.6884978229317852,
"grad_norm": 0.8984375,
"learning_rate": 0.00049597183471828,
"loss": 5.3876,
"mean_token_accuracy": 0.17896100878715515,
"num_tokens": 15780912.0,
"step": 7590
},
{
"entropy": 5.553856658935547,
"epoch": 0.6889513788098693,
"grad_norm": 0.8828125,
"learning_rate": 0.0004959657372707604,
"loss": 5.2688,
"mean_token_accuracy": 0.18778923004865647,
"num_tokens": 15791719.0,
"step": 7595
},
{
"entropy": 5.547277927398682,
"epoch": 0.6894049346879536,
"grad_norm": 0.9921875,
"learning_rate": 0.0004959596352536132,
"loss": 5.3197,
"mean_token_accuracy": 0.18113720566034316,
"num_tokens": 15802551.0,
"step": 7600
},
{
"entropy": 5.58339672088623,
"epoch": 0.6898584905660378,
"grad_norm": 0.9765625,
"learning_rate": 0.0004959535286669643,
"loss": 5.3653,
"mean_token_accuracy": 0.17841881066560744,
"num_tokens": 15813551.0,
"step": 7605
},
{
"entropy": 5.554733753204346,
"epoch": 0.6903120464441219,
"grad_norm": 0.90625,
"learning_rate": 0.0004959474175109402,
"loss": 5.3373,
"mean_token_accuracy": 0.18408777713775634,
"num_tokens": 15824262.0,
"step": 7610
},
{
"entropy": 5.616608047485352,
"epoch": 0.6907656023222061,
"grad_norm": 0.96484375,
"learning_rate": 0.0004959413017856672,
"loss": 5.3833,
"mean_token_accuracy": 0.18150201439857483,
"num_tokens": 15835493.0,
"step": 7615
},
{
"entropy": 5.446601104736328,
"epoch": 0.6912191582002902,
"grad_norm": 0.828125,
"learning_rate": 0.0004959351814912716,
"loss": 5.2793,
"mean_token_accuracy": 0.1799856498837471,
"num_tokens": 15846224.0,
"step": 7620
},
{
"entropy": 5.5849120140075685,
"epoch": 0.6916727140783745,
"grad_norm": 1.03125,
"learning_rate": 0.0004959290566278803,
"loss": 5.3063,
"mean_token_accuracy": 0.1804905876517296,
"num_tokens": 15855622.0,
"step": 7625
},
{
"entropy": 5.562844371795654,
"epoch": 0.6921262699564587,
"grad_norm": 0.8984375,
"learning_rate": 0.0004959229271956199,
"loss": 5.3789,
"mean_token_accuracy": 0.17874677032232283,
"num_tokens": 15865835.0,
"step": 7630
},
{
"entropy": 5.54128589630127,
"epoch": 0.6925798258345428,
"grad_norm": 0.95703125,
"learning_rate": 0.0004959167931946169,
"loss": 5.3154,
"mean_token_accuracy": 0.18270861357450485,
"num_tokens": 15875915.0,
"step": 7635
},
{
"entropy": 5.585084104537964,
"epoch": 0.693033381712627,
"grad_norm": 0.94921875,
"learning_rate": 0.0004959106546249984,
"loss": 5.3078,
"mean_token_accuracy": 0.18133632242679595,
"num_tokens": 15887111.0,
"step": 7640
},
{
"entropy": 5.5735945224761965,
"epoch": 0.6934869375907112,
"grad_norm": 0.84375,
"learning_rate": 0.0004959045114868913,
"loss": 5.4066,
"mean_token_accuracy": 0.17114051580429077,
"num_tokens": 15899528.0,
"step": 7645
},
{
"entropy": 5.567831516265869,
"epoch": 0.6939404934687954,
"grad_norm": 0.8515625,
"learning_rate": 0.0004958983637804225,
"loss": 5.3237,
"mean_token_accuracy": 0.18822288066148757,
"num_tokens": 15910881.0,
"step": 7650
},
{
"entropy": 5.634607219696045,
"epoch": 0.6943940493468795,
"grad_norm": 0.94140625,
"learning_rate": 0.0004958922115057194,
"loss": 5.3678,
"mean_token_accuracy": 0.17664799243211746,
"num_tokens": 15921189.0,
"step": 7655
},
{
"entropy": 5.580757188796997,
"epoch": 0.6948476052249637,
"grad_norm": 0.94921875,
"learning_rate": 0.0004958860546629089,
"loss": 5.3637,
"mean_token_accuracy": 0.18410303443670273,
"num_tokens": 15931276.0,
"step": 7660
},
{
"entropy": 5.585839080810547,
"epoch": 0.6953011611030478,
"grad_norm": 0.94140625,
"learning_rate": 0.0004958798932521186,
"loss": 5.442,
"mean_token_accuracy": 0.17055700719356537,
"num_tokens": 15941506.0,
"step": 7665
},
{
"entropy": 5.496153450012207,
"epoch": 0.6957547169811321,
"grad_norm": 1.015625,
"learning_rate": 0.0004958737272734759,
"loss": 5.2914,
"mean_token_accuracy": 0.18633339554071426,
"num_tokens": 15951158.0,
"step": 7670
},
{
"entropy": 5.613697624206543,
"epoch": 0.6962082728592163,
"grad_norm": 0.89453125,
"learning_rate": 0.0004958675567271081,
"loss": 5.3059,
"mean_token_accuracy": 0.18212190866470337,
"num_tokens": 15960852.0,
"step": 7675
},
{
"entropy": 5.579945516586304,
"epoch": 0.6966618287373004,
"grad_norm": 0.90234375,
"learning_rate": 0.000495861381613143,
"loss": 5.3139,
"mean_token_accuracy": 0.1766521543264389,
"num_tokens": 15971291.0,
"step": 7680
},
{
"entropy": 5.51633129119873,
"epoch": 0.6971153846153846,
"grad_norm": 0.98046875,
"learning_rate": 0.0004958552019317082,
"loss": 5.377,
"mean_token_accuracy": 0.17974497824907304,
"num_tokens": 15982180.0,
"step": 7685
},
{
"entropy": 5.4834810256958,
"epoch": 0.6975689404934688,
"grad_norm": 0.93359375,
"learning_rate": 0.0004958490176829316,
"loss": 5.2641,
"mean_token_accuracy": 0.18528165817260742,
"num_tokens": 15992363.0,
"step": 7690
},
{
"entropy": 5.580470180511474,
"epoch": 0.698022496371553,
"grad_norm": 0.92578125,
"learning_rate": 0.000495842828866941,
"loss": 5.3566,
"mean_token_accuracy": 0.17209097146987914,
"num_tokens": 16003477.0,
"step": 7695
},
{
"entropy": 5.598789501190185,
"epoch": 0.6984760522496372,
"grad_norm": 0.8984375,
"learning_rate": 0.0004958366354838645,
"loss": 5.3261,
"mean_token_accuracy": 0.18081470876932143,
"num_tokens": 16013747.0,
"step": 7700
},
{
"entropy": 5.516934061050415,
"epoch": 0.6989296081277213,
"grad_norm": 0.90625,
"learning_rate": 0.00049583043753383,
"loss": 5.3303,
"mean_token_accuracy": 0.18622443526983262,
"num_tokens": 16025662.0,
"step": 7705
},
{
"entropy": 5.5949502944946286,
"epoch": 0.6993831640058055,
"grad_norm": 0.89453125,
"learning_rate": 0.0004958242350169658,
"loss": 5.407,
"mean_token_accuracy": 0.18321597278118135,
"num_tokens": 16037315.0,
"step": 7710
},
{
"entropy": 5.47448148727417,
"epoch": 0.6998367198838897,
"grad_norm": 1.0546875,
"learning_rate": 0.0004958180279334001,
"loss": 5.2332,
"mean_token_accuracy": 0.18107860833406447,
"num_tokens": 16047297.0,
"step": 7715
},
{
"entropy": 5.4877598762512205,
"epoch": 0.7002902757619739,
"grad_norm": 0.90234375,
"learning_rate": 0.0004958118162832614,
"loss": 5.2136,
"mean_token_accuracy": 0.18834895491600037,
"num_tokens": 16057573.0,
"step": 7720
},
{
"entropy": 5.5364970684051515,
"epoch": 0.700743831640058,
"grad_norm": 0.98046875,
"learning_rate": 0.0004958056000666781,
"loss": 5.3848,
"mean_token_accuracy": 0.17313725650310516,
"num_tokens": 16067926.0,
"step": 7725
},
{
"entropy": 5.547545146942139,
"epoch": 0.7011973875181422,
"grad_norm": 0.94140625,
"learning_rate": 0.0004957993792837786,
"loss": 5.3135,
"mean_token_accuracy": 0.18305060714483262,
"num_tokens": 16077938.0,
"step": 7730
},
{
"entropy": 5.518269681930542,
"epoch": 0.7016509433962265,
"grad_norm": 1.0,
"learning_rate": 0.0004957931539346917,
"loss": 5.2126,
"mean_token_accuracy": 0.183458811044693,
"num_tokens": 16088163.0,
"step": 7735
},
{
"entropy": 5.6031519889831545,
"epoch": 0.7021044992743106,
"grad_norm": 0.875,
"learning_rate": 0.0004957869240195462,
"loss": 5.4064,
"mean_token_accuracy": 0.1752445250749588,
"num_tokens": 16100640.0,
"step": 7740
},
{
"entropy": 5.54415545463562,
"epoch": 0.7025580551523948,
"grad_norm": 0.87109375,
"learning_rate": 0.0004957806895384707,
"loss": 5.2409,
"mean_token_accuracy": 0.18716129958629607,
"num_tokens": 16111018.0,
"step": 7745
},
{
"entropy": 5.503624105453492,
"epoch": 0.7030116110304789,
"grad_norm": 0.92578125,
"learning_rate": 0.0004957744504915944,
"loss": 5.2766,
"mean_token_accuracy": 0.17956742346286775,
"num_tokens": 16120873.0,
"step": 7750
},
{
"entropy": 5.564220190048218,
"epoch": 0.7034651669085631,
"grad_norm": 0.953125,
"learning_rate": 0.0004957682068790463,
"loss": 5.2998,
"mean_token_accuracy": 0.17889865934848787,
"num_tokens": 16131178.0,
"step": 7755
},
{
"entropy": 5.516436815261841,
"epoch": 0.7039187227866474,
"grad_norm": 0.921875,
"learning_rate": 0.0004957619587009553,
"loss": 5.3,
"mean_token_accuracy": 0.18320292234420776,
"num_tokens": 16141156.0,
"step": 7760
},
{
"entropy": 5.544557237625122,
"epoch": 0.7043722786647315,
"grad_norm": 0.98046875,
"learning_rate": 0.0004957557059574507,
"loss": 5.405,
"mean_token_accuracy": 0.17417299449443818,
"num_tokens": 16150077.0,
"step": 7765
},
{
"entropy": 5.541102647781372,
"epoch": 0.7048258345428157,
"grad_norm": 0.921875,
"learning_rate": 0.0004957494486486618,
"loss": 5.3246,
"mean_token_accuracy": 0.17782566845417022,
"num_tokens": 16160514.0,
"step": 7770
},
{
"entropy": 5.551746559143067,
"epoch": 0.7052793904208998,
"grad_norm": 0.9609375,
"learning_rate": 0.0004957431867747182,
"loss": 5.3671,
"mean_token_accuracy": 0.18455520123243332,
"num_tokens": 16170811.0,
"step": 7775
},
{
"entropy": 5.478965806961059,
"epoch": 0.7057329462989841,
"grad_norm": 0.93359375,
"learning_rate": 0.0004957369203357492,
"loss": 5.2566,
"mean_token_accuracy": 0.1847606912255287,
"num_tokens": 16181553.0,
"step": 7780
},
{
"entropy": 5.55667085647583,
"epoch": 0.7061865021770682,
"grad_norm": 0.8828125,
"learning_rate": 0.0004957306493318844,
"loss": 5.2794,
"mean_token_accuracy": 0.1900545060634613,
"num_tokens": 16191308.0,
"step": 7785
},
{
"entropy": 5.5251086235046385,
"epoch": 0.7066400580551524,
"grad_norm": 0.9140625,
"learning_rate": 0.0004957243737632535,
"loss": 5.2563,
"mean_token_accuracy": 0.18701473474502564,
"num_tokens": 16201311.0,
"step": 7790
},
{
"entropy": 5.555533313751221,
"epoch": 0.7070936139332366,
"grad_norm": 0.9375,
"learning_rate": 0.0004957180936299863,
"loss": 5.3075,
"mean_token_accuracy": 0.1831939309835434,
"num_tokens": 16211105.0,
"step": 7795
},
{
"entropy": 5.557322025299072,
"epoch": 0.7075471698113207,
"grad_norm": 0.93359375,
"learning_rate": 0.0004957118089322127,
"loss": 5.328,
"mean_token_accuracy": 0.17431456744670867,
"num_tokens": 16220836.0,
"step": 7800
},
{
"entropy": 5.510331630706787,
"epoch": 0.708000725689405,
"grad_norm": 0.8984375,
"learning_rate": 0.0004957055196700628,
"loss": 5.2236,
"mean_token_accuracy": 0.1855901911854744,
"num_tokens": 16231630.0,
"step": 7805
},
{
"entropy": 5.560601472854614,
"epoch": 0.7084542815674891,
"grad_norm": 0.91015625,
"learning_rate": 0.0004956992258436663,
"loss": 5.3236,
"mean_token_accuracy": 0.18146177083253862,
"num_tokens": 16242272.0,
"step": 7810
},
{
"entropy": 5.4510585308074955,
"epoch": 0.7089078374455733,
"grad_norm": 0.91015625,
"learning_rate": 0.0004956929274531536,
"loss": 5.235,
"mean_token_accuracy": 0.1826574593782425,
"num_tokens": 16252555.0,
"step": 7815
},
{
"entropy": 5.519144773483276,
"epoch": 0.7093613933236574,
"grad_norm": 0.9609375,
"learning_rate": 0.000495686624498655,
"loss": 5.3509,
"mean_token_accuracy": 0.18719311952590942,
"num_tokens": 16261964.0,
"step": 7820
},
{
"entropy": 5.61521577835083,
"epoch": 0.7098149492017417,
"grad_norm": 0.83203125,
"learning_rate": 0.0004956803169803007,
"loss": 5.3896,
"mean_token_accuracy": 0.17692660838365554,
"num_tokens": 16274543.0,
"step": 7825
},
{
"entropy": 5.509320163726807,
"epoch": 0.7102685050798259,
"grad_norm": 0.859375,
"learning_rate": 0.0004956740048982212,
"loss": 5.3055,
"mean_token_accuracy": 0.17120635658502578,
"num_tokens": 16285124.0,
"step": 7830
},
{
"entropy": 5.523042774200439,
"epoch": 0.71072206095791,
"grad_norm": 0.984375,
"learning_rate": 0.0004956676882525472,
"loss": 5.3757,
"mean_token_accuracy": 0.17950959950685502,
"num_tokens": 16295727.0,
"step": 7835
},
{
"entropy": 5.622578430175781,
"epoch": 0.7111756168359942,
"grad_norm": 0.90234375,
"learning_rate": 0.000495661367043409,
"loss": 5.2772,
"mean_token_accuracy": 0.19084594547748565,
"num_tokens": 16305496.0,
"step": 7840
},
{
"entropy": 5.515881633758545,
"epoch": 0.7116291727140783,
"grad_norm": 0.97265625,
"learning_rate": 0.0004956550412709376,
"loss": 5.2645,
"mean_token_accuracy": 0.18688923120498657,
"num_tokens": 16315112.0,
"step": 7845
},
{
"entropy": 5.537652444839478,
"epoch": 0.7120827285921626,
"grad_norm": 0.87109375,
"learning_rate": 0.0004956487109352637,
"loss": 5.304,
"mean_token_accuracy": 0.18639636486768724,
"num_tokens": 16325613.0,
"step": 7850
},
{
"entropy": 5.522394752502441,
"epoch": 0.7125362844702468,
"grad_norm": 0.953125,
"learning_rate": 0.0004956423760365181,
"loss": 5.2433,
"mean_token_accuracy": 0.18728332221508026,
"num_tokens": 16335317.0,
"step": 7855
},
{
"entropy": 5.5846155166625975,
"epoch": 0.7129898403483309,
"grad_norm": 1.0390625,
"learning_rate": 0.0004956360365748321,
"loss": 5.3926,
"mean_token_accuracy": 0.17209500968456268,
"num_tokens": 16345003.0,
"step": 7860
},
{
"entropy": 5.571717071533203,
"epoch": 0.7134433962264151,
"grad_norm": 0.83984375,
"learning_rate": 0.0004956296925503367,
"loss": 5.379,
"mean_token_accuracy": 0.17814605087041854,
"num_tokens": 16356108.0,
"step": 7865
},
{
"entropy": 5.506393384933472,
"epoch": 0.7138969521044993,
"grad_norm": 0.98828125,
"learning_rate": 0.0004956233439631628,
"loss": 5.1874,
"mean_token_accuracy": 0.18501874059438705,
"num_tokens": 16364733.0,
"step": 7870
},
{
"entropy": 5.588735723495484,
"epoch": 0.7143505079825835,
"grad_norm": 0.94921875,
"learning_rate": 0.0004956169908134421,
"loss": 5.2879,
"mean_token_accuracy": 0.18495833426713942,
"num_tokens": 16375229.0,
"step": 7875
},
{
"entropy": 5.532235622406006,
"epoch": 0.7148040638606676,
"grad_norm": 0.90625,
"learning_rate": 0.0004956106331013057,
"loss": 5.3668,
"mean_token_accuracy": 0.18426259309053422,
"num_tokens": 16386366.0,
"step": 7880
},
{
"entropy": 5.4934896469116214,
"epoch": 0.7152576197387518,
"grad_norm": 0.90234375,
"learning_rate": 0.0004956042708268853,
"loss": 5.2602,
"mean_token_accuracy": 0.1904539719223976,
"num_tokens": 16397067.0,
"step": 7885
},
{
"entropy": 5.464821481704712,
"epoch": 0.715711175616836,
"grad_norm": 0.99609375,
"learning_rate": 0.0004955979039903123,
"loss": 5.2136,
"mean_token_accuracy": 0.1833643287420273,
"num_tokens": 16407256.0,
"step": 7890
},
{
"entropy": 5.49287257194519,
"epoch": 0.7161647314949202,
"grad_norm": 0.87109375,
"learning_rate": 0.0004955915325917183,
"loss": 5.2733,
"mean_token_accuracy": 0.1792703688144684,
"num_tokens": 16418701.0,
"step": 7895
},
{
"entropy": 5.476460409164429,
"epoch": 0.7166182873730044,
"grad_norm": 0.9296875,
"learning_rate": 0.0004955851566312354,
"loss": 5.1877,
"mean_token_accuracy": 0.1884955197572708,
"num_tokens": 16428019.0,
"step": 7900
},
{
"entropy": 5.452414226531983,
"epoch": 0.7170718432510885,
"grad_norm": 0.91796875,
"learning_rate": 0.0004955787761089952,
"loss": 5.2773,
"mean_token_accuracy": 0.18074552565813065,
"num_tokens": 16440122.0,
"step": 7905
},
{
"entropy": 5.62509765625,
"epoch": 0.7175253991291727,
"grad_norm": 1.015625,
"learning_rate": 0.0004955723910251296,
"loss": 5.3613,
"mean_token_accuracy": 0.17514893114566804,
"num_tokens": 16449801.0,
"step": 7910
},
{
"entropy": 5.495181322097778,
"epoch": 0.717978955007257,
"grad_norm": 0.984375,
"learning_rate": 0.0004955660013797709,
"loss": 5.2102,
"mean_token_accuracy": 0.18785545378923416,
"num_tokens": 16460155.0,
"step": 7915
},
{
"entropy": 5.575676202774048,
"epoch": 0.7184325108853411,
"grad_norm": 0.92578125,
"learning_rate": 0.000495559607173051,
"loss": 5.4192,
"mean_token_accuracy": 0.17114308923482896,
"num_tokens": 16470656.0,
"step": 7920
},
{
"entropy": 5.56002631187439,
"epoch": 0.7188860667634253,
"grad_norm": 0.90625,
"learning_rate": 0.0004955532084051022,
"loss": 5.3067,
"mean_token_accuracy": 0.18146676570177078,
"num_tokens": 16481449.0,
"step": 7925
},
{
"entropy": 5.530605125427246,
"epoch": 0.7193396226415094,
"grad_norm": 0.91796875,
"learning_rate": 0.0004955468050760568,
"loss": 5.3295,
"mean_token_accuracy": 0.1831614077091217,
"num_tokens": 16492032.0,
"step": 7930
},
{
"entropy": 5.593453073501587,
"epoch": 0.7197931785195936,
"grad_norm": 0.9375,
"learning_rate": 0.0004955403971860474,
"loss": 5.2551,
"mean_token_accuracy": 0.18039434105157853,
"num_tokens": 16502221.0,
"step": 7935
},
{
"entropy": 5.539740943908692,
"epoch": 0.7202467343976778,
"grad_norm": 1.015625,
"learning_rate": 0.0004955339847352063,
"loss": 5.2528,
"mean_token_accuracy": 0.18457072079181672,
"num_tokens": 16512021.0,
"step": 7940
},
{
"entropy": 5.499135541915893,
"epoch": 0.720700290275762,
"grad_norm": 0.9765625,
"learning_rate": 0.0004955275677236662,
"loss": 5.3026,
"mean_token_accuracy": 0.1828033670783043,
"num_tokens": 16522060.0,
"step": 7945
},
{
"entropy": 5.528159189224243,
"epoch": 0.7211538461538461,
"grad_norm": 1.0390625,
"learning_rate": 0.0004955211461515599,
"loss": 5.2898,
"mean_token_accuracy": 0.1860054388642311,
"num_tokens": 16532827.0,
"step": 7950
},
{
"entropy": 5.551040935516357,
"epoch": 0.7216074020319303,
"grad_norm": 0.87890625,
"learning_rate": 0.00049551472001902,
"loss": 5.3431,
"mean_token_accuracy": 0.18500829637050628,
"num_tokens": 16542838.0,
"step": 7955
},
{
"entropy": 5.509801578521729,
"epoch": 0.7220609579100146,
"grad_norm": 0.875,
"learning_rate": 0.0004955082893261796,
"loss": 5.2815,
"mean_token_accuracy": 0.18686290830373764,
"num_tokens": 16552615.0,
"step": 7960
},
{
"entropy": 5.583999919891357,
"epoch": 0.7225145137880987,
"grad_norm": 1.046875,
"learning_rate": 0.0004955018540731714,
"loss": 5.3489,
"mean_token_accuracy": 0.18032243102788925,
"num_tokens": 16561777.0,
"step": 7965
},
{
"entropy": 5.553142452239991,
"epoch": 0.7229680696661829,
"grad_norm": 0.890625,
"learning_rate": 0.0004954954142601288,
"loss": 5.3082,
"mean_token_accuracy": 0.18332842886447906,
"num_tokens": 16572111.0,
"step": 7970
},
{
"entropy": 5.5559593677520756,
"epoch": 0.723421625544267,
"grad_norm": 1.0,
"learning_rate": 0.0004954889698871849,
"loss": 5.3549,
"mean_token_accuracy": 0.18638369292020798,
"num_tokens": 16581674.0,
"step": 7975
},
{
"entropy": 5.570011758804322,
"epoch": 0.7238751814223512,
"grad_norm": 0.90234375,
"learning_rate": 0.0004954825209544728,
"loss": 5.335,
"mean_token_accuracy": 0.19101521968841553,
"num_tokens": 16591802.0,
"step": 7980
},
{
"entropy": 5.542544221878051,
"epoch": 0.7243287373004355,
"grad_norm": 0.9296875,
"learning_rate": 0.000495476067462126,
"loss": 5.2403,
"mean_token_accuracy": 0.1896628573536873,
"num_tokens": 16602772.0,
"step": 7985
},
{
"entropy": 5.562042999267578,
"epoch": 0.7247822931785196,
"grad_norm": 0.94921875,
"learning_rate": 0.000495469609410278,
"loss": 5.2866,
"mean_token_accuracy": 0.18497096598148347,
"num_tokens": 16613526.0,
"step": 7990
},
{
"entropy": 5.4583722114562985,
"epoch": 0.7252358490566038,
"grad_norm": 1.0859375,
"learning_rate": 0.0004954631467990622,
"loss": 5.2867,
"mean_token_accuracy": 0.18482371866703035,
"num_tokens": 16623583.0,
"step": 7995
},
{
"entropy": 5.544550371170044,
"epoch": 0.7256894049346879,
"grad_norm": 0.9765625,
"learning_rate": 0.0004954566796286124,
"loss": 5.346,
"mean_token_accuracy": 0.17807176262140273,
"num_tokens": 16635068.0,
"step": 8000
},
{
"entropy": 5.589382553100586,
"epoch": 0.7261429608127722,
"grad_norm": 0.98046875,
"learning_rate": 0.0004954502078990623,
"loss": 5.2463,
"mean_token_accuracy": 0.18150110691785812,
"num_tokens": 16645021.0,
"step": 8005
},
{
"entropy": 5.525119733810425,
"epoch": 0.7265965166908563,
"grad_norm": 0.90234375,
"learning_rate": 0.0004954437316105457,
"loss": 5.333,
"mean_token_accuracy": 0.1788739249110222,
"num_tokens": 16656021.0,
"step": 8010
},
{
"entropy": 5.494053888320923,
"epoch": 0.7270500725689405,
"grad_norm": 0.9765625,
"learning_rate": 0.0004954372507631965,
"loss": 5.2201,
"mean_token_accuracy": 0.19220224618911744,
"num_tokens": 16665236.0,
"step": 8015
},
{
"entropy": 5.558600378036499,
"epoch": 0.7275036284470247,
"grad_norm": 0.92578125,
"learning_rate": 0.0004954307653571489,
"loss": 5.4189,
"mean_token_accuracy": 0.17521561533212662,
"num_tokens": 16676690.0,
"step": 8020
},
{
"entropy": 5.557137393951416,
"epoch": 0.7279571843251088,
"grad_norm": 0.93359375,
"learning_rate": 0.0004954242753925367,
"loss": 5.3067,
"mean_token_accuracy": 0.17874979972839355,
"num_tokens": 16687288.0,
"step": 8025
},
{
"entropy": 5.520791292190552,
"epoch": 0.7284107402031931,
"grad_norm": 0.98046875,
"learning_rate": 0.0004954177808694946,
"loss": 5.3288,
"mean_token_accuracy": 0.17826912701129913,
"num_tokens": 16697873.0,
"step": 8030
},
{
"entropy": 5.5447221279144285,
"epoch": 0.7288642960812772,
"grad_norm": 0.9140625,
"learning_rate": 0.0004954112817881565,
"loss": 5.3312,
"mean_token_accuracy": 0.17438385188579558,
"num_tokens": 16709002.0,
"step": 8035
},
{
"entropy": 5.5151759624481205,
"epoch": 0.7293178519593614,
"grad_norm": 0.9765625,
"learning_rate": 0.0004954047781486568,
"loss": 5.2998,
"mean_token_accuracy": 0.1815633311867714,
"num_tokens": 16718989.0,
"step": 8040
},
{
"entropy": 5.594701957702637,
"epoch": 0.7297714078374455,
"grad_norm": 0.91796875,
"learning_rate": 0.0004953982699511303,
"loss": 5.3723,
"mean_token_accuracy": 0.1834247812628746,
"num_tokens": 16730425.0,
"step": 8045
},
{
"entropy": 5.574563789367676,
"epoch": 0.7302249637155298,
"grad_norm": 1.0390625,
"learning_rate": 0.0004953917571957114,
"loss": 5.2844,
"mean_token_accuracy": 0.18223824948072434,
"num_tokens": 16740216.0,
"step": 8050
},
{
"entropy": 5.513731193542481,
"epoch": 0.730678519593614,
"grad_norm": 0.9453125,
"learning_rate": 0.0004953852398825346,
"loss": 5.2599,
"mean_token_accuracy": 0.1835854694247246,
"num_tokens": 16752436.0,
"step": 8055
},
{
"entropy": 5.4929563999176025,
"epoch": 0.7311320754716981,
"grad_norm": 0.94921875,
"learning_rate": 0.000495378718011735,
"loss": 5.2833,
"mean_token_accuracy": 0.18290273547172547,
"num_tokens": 16762631.0,
"step": 8060
},
{
"entropy": 5.570620965957642,
"epoch": 0.7315856313497823,
"grad_norm": 0.91796875,
"learning_rate": 0.0004953721915834473,
"loss": 5.3062,
"mean_token_accuracy": 0.18442205041646959,
"num_tokens": 16772613.0,
"step": 8065
},
{
"entropy": 5.580399417877198,
"epoch": 0.7320391872278664,
"grad_norm": 0.9296875,
"learning_rate": 0.0004953656605978067,
"loss": 5.3126,
"mean_token_accuracy": 0.17444958239793779,
"num_tokens": 16782963.0,
"step": 8070
},
{
"entropy": 5.502667856216431,
"epoch": 0.7324927431059507,
"grad_norm": 0.875,
"learning_rate": 0.000495359125054948,
"loss": 5.2461,
"mean_token_accuracy": 0.1868554711341858,
"num_tokens": 16793892.0,
"step": 8075
},
{
"entropy": 5.453513860702515,
"epoch": 0.7329462989840348,
"grad_norm": 1.1640625,
"learning_rate": 0.0004953525849550063,
"loss": 5.2373,
"mean_token_accuracy": 0.17641406506299973,
"num_tokens": 16803388.0,
"step": 8080
},
{
"entropy": 5.5092305660247805,
"epoch": 0.733399854862119,
"grad_norm": 0.9375,
"learning_rate": 0.0004953460402981172,
"loss": 5.2064,
"mean_token_accuracy": 0.18928418755531312,
"num_tokens": 16813201.0,
"step": 8085
},
{
"entropy": 5.587248516082764,
"epoch": 0.7338534107402032,
"grad_norm": 0.875,
"learning_rate": 0.0004953394910844157,
"loss": 5.4576,
"mean_token_accuracy": 0.17749661803245545,
"num_tokens": 16823265.0,
"step": 8090
},
{
"entropy": 5.514876079559326,
"epoch": 0.7343069666182874,
"grad_norm": 1.0078125,
"learning_rate": 0.0004953329373140374,
"loss": 5.2368,
"mean_token_accuracy": 0.18455445468425752,
"num_tokens": 16833057.0,
"step": 8095
},
{
"entropy": 5.529699420928955,
"epoch": 0.7347605224963716,
"grad_norm": 0.84375,
"learning_rate": 0.0004953263789871179,
"loss": 5.2033,
"mean_token_accuracy": 0.18069398403167725,
"num_tokens": 16843488.0,
"step": 8100
},
{
"entropy": 5.451847314834595,
"epoch": 0.7352140783744557,
"grad_norm": 1.0390625,
"learning_rate": 0.0004953198161037925,
"loss": 5.2423,
"mean_token_accuracy": 0.1972143605351448,
"num_tokens": 16852448.0,
"step": 8105
},
{
"entropy": 5.398800182342529,
"epoch": 0.7356676342525399,
"grad_norm": 0.93359375,
"learning_rate": 0.0004953132486641975,
"loss": 5.2678,
"mean_token_accuracy": 0.182267963886261,
"num_tokens": 16863944.0,
"step": 8110
},
{
"entropy": 5.561413669586182,
"epoch": 0.736121190130624,
"grad_norm": 1.078125,
"learning_rate": 0.000495306676668468,
"loss": 5.1982,
"mean_token_accuracy": 0.18368888199329375,
"num_tokens": 16873923.0,
"step": 8115
},
{
"entropy": 5.5462486743927,
"epoch": 0.7365747460087083,
"grad_norm": 0.94921875,
"learning_rate": 0.0004953001001167406,
"loss": 5.3103,
"mean_token_accuracy": 0.17963693737983705,
"num_tokens": 16884373.0,
"step": 8120
},
{
"entropy": 5.5512104511260985,
"epoch": 0.7370283018867925,
"grad_norm": 0.96484375,
"learning_rate": 0.0004952935190091508,
"loss": 5.4311,
"mean_token_accuracy": 0.17719766050577163,
"num_tokens": 16895869.0,
"step": 8125
},
{
"entropy": 5.482508754730224,
"epoch": 0.7374818577648766,
"grad_norm": 0.94921875,
"learning_rate": 0.0004952869333458349,
"loss": 5.2713,
"mean_token_accuracy": 0.1831250846385956,
"num_tokens": 16906737.0,
"step": 8130
},
{
"entropy": 5.466742324829101,
"epoch": 0.7379354136429608,
"grad_norm": 0.984375,
"learning_rate": 0.0004952803431269292,
"loss": 5.2061,
"mean_token_accuracy": 0.18994556814432145,
"num_tokens": 16916083.0,
"step": 8135
},
{
"entropy": 5.5881688594818115,
"epoch": 0.738388969521045,
"grad_norm": 0.99609375,
"learning_rate": 0.0004952737483525699,
"loss": 5.3836,
"mean_token_accuracy": 0.182338385283947,
"num_tokens": 16926419.0,
"step": 8140
},
{
"entropy": 5.624546480178833,
"epoch": 0.7388425253991292,
"grad_norm": 0.9375,
"learning_rate": 0.0004952671490228932,
"loss": 5.3094,
"mean_token_accuracy": 0.18363170474767684,
"num_tokens": 16936067.0,
"step": 8145
},
{
"entropy": 5.578293991088867,
"epoch": 0.7392960812772134,
"grad_norm": 0.9296875,
"learning_rate": 0.0004952605451380357,
"loss": 5.253,
"mean_token_accuracy": 0.18858592212200165,
"num_tokens": 16946571.0,
"step": 8150
},
{
"entropy": 5.479947805404663,
"epoch": 0.7397496371552975,
"grad_norm": 0.9140625,
"learning_rate": 0.0004952539366981342,
"loss": 5.3013,
"mean_token_accuracy": 0.1838264137506485,
"num_tokens": 16956764.0,
"step": 8155
},
{
"entropy": 5.496211385726928,
"epoch": 0.7402031930333817,
"grad_norm": 0.9765625,
"learning_rate": 0.000495247323703325,
"loss": 5.2438,
"mean_token_accuracy": 0.192402121424675,
"num_tokens": 16966002.0,
"step": 8160
},
{
"entropy": 5.586853837966919,
"epoch": 0.7406567489114659,
"grad_norm": 0.9921875,
"learning_rate": 0.000495240706153745,
"loss": 5.341,
"mean_token_accuracy": 0.1828116625547409,
"num_tokens": 16975705.0,
"step": 8165
},
{
"entropy": 5.531956195831299,
"epoch": 0.7411103047895501,
"grad_norm": 0.95703125,
"learning_rate": 0.0004952340840495312,
"loss": 5.24,
"mean_token_accuracy": 0.18215106874704362,
"num_tokens": 16985945.0,
"step": 8170
},
{
"entropy": 5.4779589653015135,
"epoch": 0.7415638606676342,
"grad_norm": 1.0703125,
"learning_rate": 0.0004952274573908203,
"loss": 5.2237,
"mean_token_accuracy": 0.1873907685279846,
"num_tokens": 16995475.0,
"step": 8175
},
{
"entropy": 5.511494302749634,
"epoch": 0.7420174165457184,
"grad_norm": 0.9609375,
"learning_rate": 0.0004952208261777496,
"loss": 5.2523,
"mean_token_accuracy": 0.1826030820608139,
"num_tokens": 17005028.0,
"step": 8180
},
{
"entropy": 5.54114727973938,
"epoch": 0.7424709724238027,
"grad_norm": 0.94140625,
"learning_rate": 0.000495214190410456,
"loss": 5.309,
"mean_token_accuracy": 0.18562476485967636,
"num_tokens": 17014786.0,
"step": 8185
},
{
"entropy": 5.539929485321045,
"epoch": 0.7429245283018868,
"grad_norm": 0.96875,
"learning_rate": 0.0004952075500890768,
"loss": 5.3124,
"mean_token_accuracy": 0.18346714675426484,
"num_tokens": 17024431.0,
"step": 8190
},
{
"entropy": 5.508410024642944,
"epoch": 0.743378084179971,
"grad_norm": 0.93359375,
"learning_rate": 0.0004952009052137493,
"loss": 5.2929,
"mean_token_accuracy": 0.18714380860328675,
"num_tokens": 17034368.0,
"step": 8195
},
{
"entropy": 5.503003263473511,
"epoch": 0.7438316400580551,
"grad_norm": 0.94140625,
"learning_rate": 0.0004951942557846112,
"loss": 5.3261,
"mean_token_accuracy": 0.18206837624311448,
"num_tokens": 17043818.0,
"step": 8200
},
{
"entropy": 5.5616518497467045,
"epoch": 0.7442851959361393,
"grad_norm": 0.98828125,
"learning_rate": 0.0004951876018017997,
"loss": 5.3205,
"mean_token_accuracy": 0.18540932089090348,
"num_tokens": 17054843.0,
"step": 8205
},
{
"entropy": 5.535712146759034,
"epoch": 0.7447387518142236,
"grad_norm": 0.8515625,
"learning_rate": 0.0004951809432654524,
"loss": 5.3265,
"mean_token_accuracy": 0.182991661131382,
"num_tokens": 17065539.0,
"step": 8210
},
{
"entropy": 5.500705003738403,
"epoch": 0.7451923076923077,
"grad_norm": 0.96875,
"learning_rate": 0.000495174280175707,
"loss": 5.2422,
"mean_token_accuracy": 0.18335951268672943,
"num_tokens": 17075443.0,
"step": 8215
},
{
"entropy": 5.61527829170227,
"epoch": 0.7456458635703919,
"grad_norm": 0.95703125,
"learning_rate": 0.0004951676125327015,
"loss": 5.3673,
"mean_token_accuracy": 0.1847109615802765,
"num_tokens": 17085490.0,
"step": 8220
},
{
"entropy": 5.5724077224731445,
"epoch": 0.746099419448476,
"grad_norm": 0.96875,
"learning_rate": 0.0004951609403365737,
"loss": 5.3522,
"mean_token_accuracy": 0.1829810321331024,
"num_tokens": 17094803.0,
"step": 8225
},
{
"entropy": 5.525753211975098,
"epoch": 0.7465529753265603,
"grad_norm": 0.93359375,
"learning_rate": 0.0004951542635874615,
"loss": 5.2136,
"mean_token_accuracy": 0.18302233666181564,
"num_tokens": 17104373.0,
"step": 8230
},
{
"entropy": 5.557919216156006,
"epoch": 0.7470065312046444,
"grad_norm": 0.87109375,
"learning_rate": 0.0004951475822855029,
"loss": 5.2884,
"mean_token_accuracy": 0.18372104465961456,
"num_tokens": 17115469.0,
"step": 8235
},
{
"entropy": 5.532179355621338,
"epoch": 0.7474600870827286,
"grad_norm": 0.9765625,
"learning_rate": 0.0004951408964308364,
"loss": 5.2668,
"mean_token_accuracy": 0.18367121368646622,
"num_tokens": 17125443.0,
"step": 8240
},
{
"entropy": 5.463788461685181,
"epoch": 0.7479136429608128,
"grad_norm": 0.9921875,
"learning_rate": 0.0004951342060236,
"loss": 5.2343,
"mean_token_accuracy": 0.18693330585956575,
"num_tokens": 17136253.0,
"step": 8245
},
{
"entropy": 5.573167133331299,
"epoch": 0.7483671988388969,
"grad_norm": 1.0546875,
"learning_rate": 0.0004951275110639321,
"loss": 5.2911,
"mean_token_accuracy": 0.18700899928808212,
"num_tokens": 17145544.0,
"step": 8250
},
{
"entropy": 5.55022964477539,
"epoch": 0.7488207547169812,
"grad_norm": 0.9140625,
"learning_rate": 0.0004951208115519712,
"loss": 5.2952,
"mean_token_accuracy": 0.18260560929775238,
"num_tokens": 17155043.0,
"step": 8255
},
{
"entropy": 5.5254528522491455,
"epoch": 0.7492743105950653,
"grad_norm": 1.046875,
"learning_rate": 0.0004951141074878558,
"loss": 5.3393,
"mean_token_accuracy": 0.1820871725678444,
"num_tokens": 17164885.0,
"step": 8260
},
{
"entropy": 5.631131505966186,
"epoch": 0.7497278664731495,
"grad_norm": 0.9765625,
"learning_rate": 0.0004951073988717246,
"loss": 5.3404,
"mean_token_accuracy": 0.1810414507985115,
"num_tokens": 17175681.0,
"step": 8265
},
{
"entropy": 5.473759317398072,
"epoch": 0.7501814223512336,
"grad_norm": 0.9140625,
"learning_rate": 0.0004951006857037163,
"loss": 5.213,
"mean_token_accuracy": 0.1888574704527855,
"num_tokens": 17186002.0,
"step": 8270
},
{
"entropy": 5.524263381958008,
"epoch": 0.7506349782293179,
"grad_norm": 1.0,
"learning_rate": 0.0004950939679839696,
"loss": 5.2646,
"mean_token_accuracy": 0.18346552699804305,
"num_tokens": 17195499.0,
"step": 8275
},
{
"entropy": 5.533853340148926,
"epoch": 0.7510885341074021,
"grad_norm": 0.96875,
"learning_rate": 0.0004950872457126237,
"loss": 5.2716,
"mean_token_accuracy": 0.18572352528572084,
"num_tokens": 17204684.0,
"step": 8280
},
{
"entropy": 5.538997459411621,
"epoch": 0.7515420899854862,
"grad_norm": 0.9453125,
"learning_rate": 0.0004950805188898175,
"loss": 5.3085,
"mean_token_accuracy": 0.17798569202423095,
"num_tokens": 17213987.0,
"step": 8285
},
{
"entropy": 5.532698488235473,
"epoch": 0.7519956458635704,
"grad_norm": 0.91015625,
"learning_rate": 0.00049507378751569,
"loss": 5.3141,
"mean_token_accuracy": 0.17800866514444352,
"num_tokens": 17224512.0,
"step": 8290
},
{
"entropy": 5.616262674331665,
"epoch": 0.7524492017416545,
"grad_norm": 1.015625,
"learning_rate": 0.0004950670515903805,
"loss": 5.3375,
"mean_token_accuracy": 0.18572027534246444,
"num_tokens": 17234362.0,
"step": 8295
},
{
"entropy": 5.554524564743042,
"epoch": 0.7529027576197388,
"grad_norm": 1.0,
"learning_rate": 0.0004950603111140283,
"loss": 5.3147,
"mean_token_accuracy": 0.18971499800682068,
"num_tokens": 17244241.0,
"step": 8300
},
{
"entropy": 5.4980474472045895,
"epoch": 0.753356313497823,
"grad_norm": 0.93359375,
"learning_rate": 0.0004950535660867728,
"loss": 5.2489,
"mean_token_accuracy": 0.18815081119537352,
"num_tokens": 17254689.0,
"step": 8305
},
{
"entropy": 5.564259672164917,
"epoch": 0.7538098693759071,
"grad_norm": 0.92578125,
"learning_rate": 0.0004950468165087535,
"loss": 5.3434,
"mean_token_accuracy": 0.17959407716989517,
"num_tokens": 17265506.0,
"step": 8310
},
{
"entropy": 5.543921327590942,
"epoch": 0.7542634252539913,
"grad_norm": 0.91015625,
"learning_rate": 0.0004950400623801098,
"loss": 5.4162,
"mean_token_accuracy": 0.18204887360334396,
"num_tokens": 17276376.0,
"step": 8315
},
{
"entropy": 5.59347243309021,
"epoch": 0.7547169811320755,
"grad_norm": 0.85546875,
"learning_rate": 0.0004950333037009818,
"loss": 5.3005,
"mean_token_accuracy": 0.17268142551183702,
"num_tokens": 17287005.0,
"step": 8320
},
{
"entropy": 5.582093858718872,
"epoch": 0.7551705370101597,
"grad_norm": 0.93359375,
"learning_rate": 0.0004950265404715089,
"loss": 5.3053,
"mean_token_accuracy": 0.17970646917819977,
"num_tokens": 17297595.0,
"step": 8325
},
{
"entropy": 5.458274936676025,
"epoch": 0.7556240928882438,
"grad_norm": 0.96484375,
"learning_rate": 0.000495019772691831,
"loss": 5.175,
"mean_token_accuracy": 0.1917903169989586,
"num_tokens": 17307319.0,
"step": 8330
},
{
"entropy": 5.572049331665039,
"epoch": 0.756077648766328,
"grad_norm": 0.94140625,
"learning_rate": 0.0004950130003620882,
"loss": 5.4002,
"mean_token_accuracy": 0.1848435267806053,
"num_tokens": 17317978.0,
"step": 8335
},
{
"entropy": 5.51245903968811,
"epoch": 0.7565312046444121,
"grad_norm": 0.9296875,
"learning_rate": 0.0004950062234824204,
"loss": 5.2928,
"mean_token_accuracy": 0.18260230571031572,
"num_tokens": 17328339.0,
"step": 8340
},
{
"entropy": 5.52795376777649,
"epoch": 0.7569847605224964,
"grad_norm": 0.9375,
"learning_rate": 0.0004949994420529679,
"loss": 5.3443,
"mean_token_accuracy": 0.18249064534902573,
"num_tokens": 17338668.0,
"step": 8345
},
{
"entropy": 5.498517084121704,
"epoch": 0.7574383164005806,
"grad_norm": 0.98828125,
"learning_rate": 0.0004949926560738709,
"loss": 5.2917,
"mean_token_accuracy": 0.177150858938694,
"num_tokens": 17348609.0,
"step": 8350
},
{
"entropy": 5.578651428222656,
"epoch": 0.7578918722786647,
"grad_norm": 0.98046875,
"learning_rate": 0.0004949858655452697,
"loss": 5.2962,
"mean_token_accuracy": 0.18345605581998825,
"num_tokens": 17358014.0,
"step": 8355
},
{
"entropy": 5.5646138191223145,
"epoch": 0.7583454281567489,
"grad_norm": 0.87109375,
"learning_rate": 0.0004949790704673048,
"loss": 5.296,
"mean_token_accuracy": 0.186521714925766,
"num_tokens": 17367975.0,
"step": 8360
},
{
"entropy": 5.550845050811768,
"epoch": 0.7587989840348331,
"grad_norm": 0.91015625,
"learning_rate": 0.0004949722708401166,
"loss": 5.2444,
"mean_token_accuracy": 0.1873399719595909,
"num_tokens": 17377960.0,
"step": 8365
},
{
"entropy": 5.570762348175049,
"epoch": 0.7592525399129173,
"grad_norm": 0.95703125,
"learning_rate": 0.0004949654666638457,
"loss": 5.3189,
"mean_token_accuracy": 0.18686124682426453,
"num_tokens": 17387334.0,
"step": 8370
},
{
"entropy": 5.525769948959351,
"epoch": 0.7597060957910015,
"grad_norm": 0.91796875,
"learning_rate": 0.000494958657938633,
"loss": 5.2811,
"mean_token_accuracy": 0.1862280398607254,
"num_tokens": 17396521.0,
"step": 8375
},
{
"entropy": 5.485896825790405,
"epoch": 0.7601596516690856,
"grad_norm": 0.921875,
"learning_rate": 0.0004949518446646192,
"loss": 5.2506,
"mean_token_accuracy": 0.18615950345993043,
"num_tokens": 17407217.0,
"step": 8380
},
{
"entropy": 5.662817144393921,
"epoch": 0.7606132075471698,
"grad_norm": 0.9453125,
"learning_rate": 0.0004949450268419451,
"loss": 5.3806,
"mean_token_accuracy": 0.17769204825162888,
"num_tokens": 17416982.0,
"step": 8385
},
{
"entropy": 5.483257293701172,
"epoch": 0.761066763425254,
"grad_norm": 0.91796875,
"learning_rate": 0.000494938204470752,
"loss": 5.2216,
"mean_token_accuracy": 0.18211308419704436,
"num_tokens": 17427636.0,
"step": 8390
},
{
"entropy": 5.508321857452392,
"epoch": 0.7615203193033382,
"grad_norm": 0.8984375,
"learning_rate": 0.0004949313775511806,
"loss": 5.3514,
"mean_token_accuracy": 0.17798120081424712,
"num_tokens": 17437488.0,
"step": 8395
},
{
"entropy": 5.528272867202759,
"epoch": 0.7619738751814223,
"grad_norm": 0.87890625,
"learning_rate": 0.0004949245460833723,
"loss": 5.1376,
"mean_token_accuracy": 0.19336626231670379,
"num_tokens": 17447838.0,
"step": 8400
},
{
"entropy": 5.554759931564331,
"epoch": 0.7624274310595065,
"grad_norm": 0.9375,
"learning_rate": 0.0004949177100674684,
"loss": 5.3676,
"mean_token_accuracy": 0.18535151779651643,
"num_tokens": 17457257.0,
"step": 8405
},
{
"entropy": 5.545297145843506,
"epoch": 0.7628809869375908,
"grad_norm": 0.91796875,
"learning_rate": 0.0004949108695036102,
"loss": 5.2635,
"mean_token_accuracy": 0.18526664525270461,
"num_tokens": 17467729.0,
"step": 8410
},
{
"entropy": 5.5293680191040036,
"epoch": 0.7633345428156749,
"grad_norm": 0.92578125,
"learning_rate": 0.0004949040243919392,
"loss": 5.3014,
"mean_token_accuracy": 0.17469221204519272,
"num_tokens": 17477692.0,
"step": 8415
},
{
"entropy": 5.5139758586883545,
"epoch": 0.7637880986937591,
"grad_norm": 0.9609375,
"learning_rate": 0.0004948971747325968,
"loss": 5.2971,
"mean_token_accuracy": 0.18337151855230333,
"num_tokens": 17488730.0,
"step": 8420
},
{
"entropy": 5.5645856857299805,
"epoch": 0.7642416545718432,
"grad_norm": 0.98828125,
"learning_rate": 0.000494890320525725,
"loss": 5.3605,
"mean_token_accuracy": 0.18363076001405715,
"num_tokens": 17498374.0,
"step": 8425
},
{
"entropy": 5.513505125045777,
"epoch": 0.7646952104499274,
"grad_norm": 0.953125,
"learning_rate": 0.0004948834617714652,
"loss": 5.2246,
"mean_token_accuracy": 0.19082147926092147,
"num_tokens": 17508601.0,
"step": 8430
},
{
"entropy": 5.500728511810303,
"epoch": 0.7651487663280117,
"grad_norm": 0.9765625,
"learning_rate": 0.0004948765984699595,
"loss": 5.2605,
"mean_token_accuracy": 0.1817544087767601,
"num_tokens": 17518164.0,
"step": 8435
},
{
"entropy": 5.560157108306885,
"epoch": 0.7656023222060958,
"grad_norm": 1.0390625,
"learning_rate": 0.0004948697306213494,
"loss": 5.2384,
"mean_token_accuracy": 0.19223212003707885,
"num_tokens": 17527244.0,
"step": 8440
},
{
"entropy": 5.5458314418792725,
"epoch": 0.76605587808418,
"grad_norm": 0.98828125,
"learning_rate": 0.0004948628582257775,
"loss": 5.2992,
"mean_token_accuracy": 0.18211422115564346,
"num_tokens": 17538045.0,
"step": 8445
},
{
"entropy": 5.500934314727783,
"epoch": 0.7665094339622641,
"grad_norm": 0.96875,
"learning_rate": 0.0004948559812833857,
"loss": 5.2567,
"mean_token_accuracy": 0.18168097883462905,
"num_tokens": 17547584.0,
"step": 8450
},
{
"entropy": 5.520407390594483,
"epoch": 0.7669629898403484,
"grad_norm": 0.88671875,
"learning_rate": 0.0004948490997943162,
"loss": 5.3063,
"mean_token_accuracy": 0.17883863002061845,
"num_tokens": 17558925.0,
"step": 8455
},
{
"entropy": 5.576011705398559,
"epoch": 0.7674165457184325,
"grad_norm": 1.015625,
"learning_rate": 0.0004948422137587111,
"loss": 5.3824,
"mean_token_accuracy": 0.17435939311981202,
"num_tokens": 17569355.0,
"step": 8460
},
{
"entropy": 5.576209831237793,
"epoch": 0.7678701015965167,
"grad_norm": 0.8828125,
"learning_rate": 0.0004948353231767131,
"loss": 5.2571,
"mean_token_accuracy": 0.1864034727215767,
"num_tokens": 17579782.0,
"step": 8465
},
{
"entropy": 5.518272352218628,
"epoch": 0.7683236574746009,
"grad_norm": 0.9296875,
"learning_rate": 0.0004948284280484645,
"loss": 5.301,
"mean_token_accuracy": 0.19304556250572205,
"num_tokens": 17589772.0,
"step": 8470
},
{
"entropy": 5.47354040145874,
"epoch": 0.768777213352685,
"grad_norm": 1.09375,
"learning_rate": 0.0004948215283741081,
"loss": 5.2239,
"mean_token_accuracy": 0.1867895171046257,
"num_tokens": 17599613.0,
"step": 8475
},
{
"entropy": 5.533515357971192,
"epoch": 0.7692307692307693,
"grad_norm": 0.99609375,
"learning_rate": 0.0004948146241537866,
"loss": 5.3134,
"mean_token_accuracy": 0.17987559139728546,
"num_tokens": 17609056.0,
"step": 8480
},
{
"entropy": 5.526946115493774,
"epoch": 0.7696843251088534,
"grad_norm": 0.9609375,
"learning_rate": 0.0004948077153876425,
"loss": 5.2421,
"mean_token_accuracy": 0.1833679959177971,
"num_tokens": 17618620.0,
"step": 8485
},
{
"entropy": 5.498790073394775,
"epoch": 0.7701378809869376,
"grad_norm": 0.95703125,
"learning_rate": 0.0004948008020758188,
"loss": 5.2163,
"mean_token_accuracy": 0.190741603076458,
"num_tokens": 17627630.0,
"step": 8490
},
{
"entropy": 5.430834245681763,
"epoch": 0.7705914368650217,
"grad_norm": 0.91015625,
"learning_rate": 0.0004947938842184585,
"loss": 5.1991,
"mean_token_accuracy": 0.1896286591887474,
"num_tokens": 17638903.0,
"step": 8495
},
{
"entropy": 5.4956201076507565,
"epoch": 0.771044992743106,
"grad_norm": 0.9375,
"learning_rate": 0.0004947869618157048,
"loss": 5.2584,
"mean_token_accuracy": 0.18483953028917313,
"num_tokens": 17649257.0,
"step": 8500
},
{
"entropy": 5.515977144241333,
"epoch": 0.7714985486211902,
"grad_norm": 0.98046875,
"learning_rate": 0.0004947800348677006,
"loss": 5.2287,
"mean_token_accuracy": 0.189926940202713,
"num_tokens": 17659495.0,
"step": 8505
},
{
"entropy": 5.451489019393921,
"epoch": 0.7719521044992743,
"grad_norm": 1.0078125,
"learning_rate": 0.0004947731033745893,
"loss": 5.1979,
"mean_token_accuracy": 0.1842445269227028,
"num_tokens": 17669671.0,
"step": 8510
},
{
"entropy": 5.574065208435059,
"epoch": 0.7724056603773585,
"grad_norm": 0.90234375,
"learning_rate": 0.0004947661673365142,
"loss": 5.3786,
"mean_token_accuracy": 0.1808546781539917,
"num_tokens": 17680927.0,
"step": 8515
},
{
"entropy": 5.518579769134521,
"epoch": 0.7728592162554426,
"grad_norm": 0.93359375,
"learning_rate": 0.0004947592267536188,
"loss": 5.2422,
"mean_token_accuracy": 0.18695141077041627,
"num_tokens": 17691378.0,
"step": 8520
},
{
"entropy": 5.5204674243927006,
"epoch": 0.7733127721335269,
"grad_norm": 0.9140625,
"learning_rate": 0.0004947522816260466,
"loss": 5.327,
"mean_token_accuracy": 0.18174126297235488,
"num_tokens": 17700942.0,
"step": 8525
},
{
"entropy": 5.538609600067138,
"epoch": 0.773766328011611,
"grad_norm": 1.015625,
"learning_rate": 0.0004947453319539411,
"loss": 5.2441,
"mean_token_accuracy": 0.18784799724817275,
"num_tokens": 17710076.0,
"step": 8530
},
{
"entropy": 5.509739780426026,
"epoch": 0.7742198838896952,
"grad_norm": 0.9140625,
"learning_rate": 0.0004947383777374462,
"loss": 5.2607,
"mean_token_accuracy": 0.18655650168657303,
"num_tokens": 17721009.0,
"step": 8535
},
{
"entropy": 5.4949547290802006,
"epoch": 0.7746734397677794,
"grad_norm": 0.85546875,
"learning_rate": 0.0004947314189767058,
"loss": 5.2825,
"mean_token_accuracy": 0.18507248163223267,
"num_tokens": 17731821.0,
"step": 8540
},
{
"entropy": 5.481418752670288,
"epoch": 0.7751269956458636,
"grad_norm": 0.90625,
"learning_rate": 0.0004947244556718635,
"loss": 5.1695,
"mean_token_accuracy": 0.19244110435247422,
"num_tokens": 17741425.0,
"step": 8545
},
{
"entropy": 5.49365496635437,
"epoch": 0.7755805515239478,
"grad_norm": 0.9296875,
"learning_rate": 0.0004947174878230635,
"loss": 5.2346,
"mean_token_accuracy": 0.18750640451908113,
"num_tokens": 17752381.0,
"step": 8550
},
{
"entropy": 5.427494478225708,
"epoch": 0.7760341074020319,
"grad_norm": 0.9375,
"learning_rate": 0.0004947105154304498,
"loss": 5.2263,
"mean_token_accuracy": 0.18348337858915328,
"num_tokens": 17762768.0,
"step": 8555
},
{
"entropy": 5.489666557312011,
"epoch": 0.7764876632801161,
"grad_norm": 0.9296875,
"learning_rate": 0.0004947035384941667,
"loss": 5.2553,
"mean_token_accuracy": 0.1805388867855072,
"num_tokens": 17774032.0,
"step": 8560
},
{
"entropy": 5.582749938964843,
"epoch": 0.7769412191582002,
"grad_norm": 0.9140625,
"learning_rate": 0.0004946965570143584,
"loss": 5.3812,
"mean_token_accuracy": 0.17563943713903427,
"num_tokens": 17785098.0,
"step": 8565
},
{
"entropy": 5.570175743103027,
"epoch": 0.7773947750362845,
"grad_norm": 0.98828125,
"learning_rate": 0.0004946895709911695,
"loss": 5.2531,
"mean_token_accuracy": 0.1870201513171196,
"num_tokens": 17795487.0,
"step": 8570
},
{
"entropy": 5.5367103099823,
"epoch": 0.7778483309143687,
"grad_norm": 0.98828125,
"learning_rate": 0.0004946825804247442,
"loss": 5.3738,
"mean_token_accuracy": 0.17667229026556014,
"num_tokens": 17806045.0,
"step": 8575
},
{
"entropy": 5.615284633636475,
"epoch": 0.7783018867924528,
"grad_norm": 0.90234375,
"learning_rate": 0.000494675585315227,
"loss": 5.3441,
"mean_token_accuracy": 0.17770788669586182,
"num_tokens": 17816948.0,
"step": 8580
},
{
"entropy": 5.56788387298584,
"epoch": 0.778755442670537,
"grad_norm": 0.93359375,
"learning_rate": 0.0004946685856627628,
"loss": 5.2718,
"mean_token_accuracy": 0.18811210244894028,
"num_tokens": 17826929.0,
"step": 8585
},
{
"entropy": 5.522318029403687,
"epoch": 0.7792089985486212,
"grad_norm": 0.90234375,
"learning_rate": 0.0004946615814674963,
"loss": 5.3625,
"mean_token_accuracy": 0.18369622081518172,
"num_tokens": 17837780.0,
"step": 8590
},
{
"entropy": 5.443318319320679,
"epoch": 0.7796625544267054,
"grad_norm": 1.03125,
"learning_rate": 0.0004946545727295724,
"loss": 5.1325,
"mean_token_accuracy": 0.1949272483587265,
"num_tokens": 17847835.0,
"step": 8595
},
{
"entropy": 5.502077627182007,
"epoch": 0.7801161103047896,
"grad_norm": 0.9609375,
"learning_rate": 0.0004946475594491359,
"loss": 5.2175,
"mean_token_accuracy": 0.19205821305513382,
"num_tokens": 17858324.0,
"step": 8600
},
{
"entropy": 5.500710916519165,
"epoch": 0.7805696661828737,
"grad_norm": 1.0,
"learning_rate": 0.0004946405416263319,
"loss": 5.2842,
"mean_token_accuracy": 0.17803624123334885,
"num_tokens": 17868455.0,
"step": 8605
},
{
"entropy": 5.545377159118653,
"epoch": 0.7810232220609579,
"grad_norm": 0.890625,
"learning_rate": 0.0004946335192613054,
"loss": 5.2662,
"mean_token_accuracy": 0.18817005157470704,
"num_tokens": 17879038.0,
"step": 8610
},
{
"entropy": 5.456969118118286,
"epoch": 0.7814767779390421,
"grad_norm": 0.8671875,
"learning_rate": 0.0004946264923542019,
"loss": 5.1368,
"mean_token_accuracy": 0.19142608642578124,
"num_tokens": 17888153.0,
"step": 8615
},
{
"entropy": 5.476630306243896,
"epoch": 0.7819303338171263,
"grad_norm": 0.96484375,
"learning_rate": 0.0004946194609051664,
"loss": 5.262,
"mean_token_accuracy": 0.18682017773389817,
"num_tokens": 17898289.0,
"step": 8620
},
{
"entropy": 5.496731328964233,
"epoch": 0.7823838896952104,
"grad_norm": 0.859375,
"learning_rate": 0.0004946124249143447,
"loss": 5.2527,
"mean_token_accuracy": 0.18772162050008773,
"num_tokens": 17909192.0,
"step": 8625
},
{
"entropy": 5.486784315109253,
"epoch": 0.7828374455732946,
"grad_norm": 0.88671875,
"learning_rate": 0.000494605384381882,
"loss": 5.2113,
"mean_token_accuracy": 0.18046634793281555,
"num_tokens": 17918937.0,
"step": 8630
},
{
"entropy": 5.44954481124878,
"epoch": 0.7832910014513788,
"grad_norm": 0.92578125,
"learning_rate": 0.0004945983393079239,
"loss": 5.2437,
"mean_token_accuracy": 0.19104671776294707,
"num_tokens": 17929346.0,
"step": 8635
},
{
"entropy": 5.5133326053619385,
"epoch": 0.783744557329463,
"grad_norm": 0.921875,
"learning_rate": 0.0004945912896926163,
"loss": 5.2583,
"mean_token_accuracy": 0.18662541210651398,
"num_tokens": 17939479.0,
"step": 8640
},
{
"entropy": 5.582485771179199,
"epoch": 0.7841981132075472,
"grad_norm": 0.921875,
"learning_rate": 0.0004945842355361049,
"loss": 5.3415,
"mean_token_accuracy": 0.1828889235854149,
"num_tokens": 17949841.0,
"step": 8645
},
{
"entropy": 5.457309579849243,
"epoch": 0.7846516690856313,
"grad_norm": 0.99609375,
"learning_rate": 0.0004945771768385355,
"loss": 5.2602,
"mean_token_accuracy": 0.19243107438087464,
"num_tokens": 17960334.0,
"step": 8650
},
{
"entropy": 5.485492467880249,
"epoch": 0.7851052249637155,
"grad_norm": 0.8984375,
"learning_rate": 0.000494570113600054,
"loss": 5.2071,
"mean_token_accuracy": 0.19041529893875123,
"num_tokens": 17970595.0,
"step": 8655
},
{
"entropy": 5.548058986663818,
"epoch": 0.7855587808417998,
"grad_norm": 0.87890625,
"learning_rate": 0.0004945630458208068,
"loss": 5.3477,
"mean_token_accuracy": 0.17438185960054398,
"num_tokens": 17981301.0,
"step": 8660
},
{
"entropy": 5.470705938339234,
"epoch": 0.7860123367198839,
"grad_norm": 0.921875,
"learning_rate": 0.0004945559735009397,
"loss": 5.1765,
"mean_token_accuracy": 0.19010596424341203,
"num_tokens": 17991262.0,
"step": 8665
},
{
"entropy": 5.533977603912353,
"epoch": 0.7864658925979681,
"grad_norm": 0.90234375,
"learning_rate": 0.0004945488966405993,
"loss": 5.2227,
"mean_token_accuracy": 0.18723469227552414,
"num_tokens": 18001212.0,
"step": 8670
},
{
"entropy": 5.502318048477173,
"epoch": 0.7869194484760522,
"grad_norm": 0.94140625,
"learning_rate": 0.0004945418152399317,
"loss": 5.2378,
"mean_token_accuracy": 0.1913598209619522,
"num_tokens": 18010432.0,
"step": 8675
},
{
"entropy": 5.518057823181152,
"epoch": 0.7873730043541364,
"grad_norm": 0.92578125,
"learning_rate": 0.0004945347292990833,
"loss": 5.2782,
"mean_token_accuracy": 0.18441058844327926,
"num_tokens": 18020902.0,
"step": 8680
},
{
"entropy": 5.491929054260254,
"epoch": 0.7878265602322206,
"grad_norm": 0.96484375,
"learning_rate": 0.0004945276388182008,
"loss": 5.1539,
"mean_token_accuracy": 0.1898062765598297,
"num_tokens": 18031133.0,
"step": 8685
},
{
"entropy": 5.53517780303955,
"epoch": 0.7882801161103048,
"grad_norm": 0.9140625,
"learning_rate": 0.0004945205437974309,
"loss": 5.255,
"mean_token_accuracy": 0.18925394713878632,
"num_tokens": 18042050.0,
"step": 8690
},
{
"entropy": 5.424103832244873,
"epoch": 0.788733671988389,
"grad_norm": 0.97265625,
"learning_rate": 0.0004945134442369201,
"loss": 5.2306,
"mean_token_accuracy": 0.1864193484187126,
"num_tokens": 18051642.0,
"step": 8695
},
{
"entropy": 5.54458270072937,
"epoch": 0.7891872278664731,
"grad_norm": 1.0625,
"learning_rate": 0.0004945063401368154,
"loss": 5.2821,
"mean_token_accuracy": 0.17556448727846147,
"num_tokens": 18062180.0,
"step": 8700
},
{
"entropy": 5.519570016860962,
"epoch": 0.7896407837445574,
"grad_norm": 0.953125,
"learning_rate": 0.0004944992314972637,
"loss": 5.2274,
"mean_token_accuracy": 0.18374626189470292,
"num_tokens": 18072302.0,
"step": 8705
},
{
"entropy": 5.469928026199341,
"epoch": 0.7900943396226415,
"grad_norm": 0.9453125,
"learning_rate": 0.0004944921183184118,
"loss": 5.2358,
"mean_token_accuracy": 0.19111284017562866,
"num_tokens": 18081824.0,
"step": 8710
},
{
"entropy": 5.518294858932495,
"epoch": 0.7905478955007257,
"grad_norm": 0.9453125,
"learning_rate": 0.000494485000600407,
"loss": 5.2924,
"mean_token_accuracy": 0.188755401968956,
"num_tokens": 18092696.0,
"step": 8715
},
{
"entropy": 5.5013707160949705,
"epoch": 0.7910014513788098,
"grad_norm": 1.0,
"learning_rate": 0.0004944778783433965,
"loss": 5.1994,
"mean_token_accuracy": 0.19762906432151794,
"num_tokens": 18101790.0,
"step": 8720
},
{
"entropy": 5.563456010818482,
"epoch": 0.791455007256894,
"grad_norm": 0.8828125,
"learning_rate": 0.0004944707515475276,
"loss": 5.3227,
"mean_token_accuracy": 0.18671682476997375,
"num_tokens": 18111572.0,
"step": 8725
},
{
"entropy": 5.548861169815064,
"epoch": 0.7919085631349783,
"grad_norm": 0.94140625,
"learning_rate": 0.0004944636202129475,
"loss": 5.3125,
"mean_token_accuracy": 0.1888110116124153,
"num_tokens": 18120650.0,
"step": 8730
},
{
"entropy": 5.486426162719726,
"epoch": 0.7923621190130624,
"grad_norm": 0.94140625,
"learning_rate": 0.0004944564843398039,
"loss": 5.1831,
"mean_token_accuracy": 0.1958876892924309,
"num_tokens": 18131055.0,
"step": 8735
},
{
"entropy": 5.5568522930145265,
"epoch": 0.7928156748911466,
"grad_norm": 0.9921875,
"learning_rate": 0.0004944493439282443,
"loss": 5.292,
"mean_token_accuracy": 0.18856184631586076,
"num_tokens": 18141423.0,
"step": 8740
},
{
"entropy": 5.456881761550903,
"epoch": 0.7932692307692307,
"grad_norm": 0.93359375,
"learning_rate": 0.0004944421989784162,
"loss": 5.2098,
"mean_token_accuracy": 0.18828487694263457,
"num_tokens": 18152015.0,
"step": 8745
},
{
"entropy": 5.548474550247192,
"epoch": 0.793722786647315,
"grad_norm": 0.94921875,
"learning_rate": 0.0004944350494904676,
"loss": 5.3002,
"mean_token_accuracy": 0.18480850607156754,
"num_tokens": 18161934.0,
"step": 8750
},
{
"entropy": 5.479960775375366,
"epoch": 0.7941763425253991,
"grad_norm": 0.90625,
"learning_rate": 0.0004944278954645463,
"loss": 5.2243,
"mean_token_accuracy": 0.18462625294923782,
"num_tokens": 18172287.0,
"step": 8755
},
{
"entropy": 5.498562955856324,
"epoch": 0.7946298984034833,
"grad_norm": 1.046875,
"learning_rate": 0.0004944207369008002,
"loss": 5.2699,
"mean_token_accuracy": 0.1924784377217293,
"num_tokens": 18181015.0,
"step": 8760
},
{
"entropy": 5.600859260559082,
"epoch": 0.7950834542815675,
"grad_norm": 0.859375,
"learning_rate": 0.0004944135737993773,
"loss": 5.3619,
"mean_token_accuracy": 0.1822371244430542,
"num_tokens": 18192382.0,
"step": 8765
},
{
"entropy": 5.50996584892273,
"epoch": 0.7955370101596516,
"grad_norm": 0.95703125,
"learning_rate": 0.0004944064061604258,
"loss": 5.2123,
"mean_token_accuracy": 0.19568186849355698,
"num_tokens": 18202597.0,
"step": 8770
},
{
"entropy": 5.5370594501495365,
"epoch": 0.7959905660377359,
"grad_norm": 0.890625,
"learning_rate": 0.0004943992339840939,
"loss": 5.2795,
"mean_token_accuracy": 0.1780181646347046,
"num_tokens": 18213756.0,
"step": 8775
},
{
"entropy": 5.419239091873169,
"epoch": 0.79644412191582,
"grad_norm": 0.8984375,
"learning_rate": 0.00049439205727053,
"loss": 5.2022,
"mean_token_accuracy": 0.18264902830123902,
"num_tokens": 18223669.0,
"step": 8780
},
{
"entropy": 5.498378324508667,
"epoch": 0.7968976777939042,
"grad_norm": 0.8984375,
"learning_rate": 0.0004943848760198823,
"loss": 5.2325,
"mean_token_accuracy": 0.19068809002637863,
"num_tokens": 18234884.0,
"step": 8785
},
{
"entropy": 5.486072826385498,
"epoch": 0.7973512336719883,
"grad_norm": 0.921875,
"learning_rate": 0.0004943776902322995,
"loss": 5.1603,
"mean_token_accuracy": 0.19515904933214187,
"num_tokens": 18245155.0,
"step": 8790
},
{
"entropy": 5.39517731666565,
"epoch": 0.7978047895500726,
"grad_norm": 0.8828125,
"learning_rate": 0.0004943704999079302,
"loss": 5.1019,
"mean_token_accuracy": 0.2020408481359482,
"num_tokens": 18254686.0,
"step": 8795
},
{
"entropy": 5.470184278488159,
"epoch": 0.7982583454281568,
"grad_norm": 0.8671875,
"learning_rate": 0.0004943633050469231,
"loss": 5.1363,
"mean_token_accuracy": 0.1927419751882553,
"num_tokens": 18265473.0,
"step": 8800
},
{
"entropy": 5.523016786575317,
"epoch": 0.7987119013062409,
"grad_norm": 0.96875,
"learning_rate": 0.0004943561056494269,
"loss": 5.3438,
"mean_token_accuracy": 0.18321252465248108,
"num_tokens": 18276024.0,
"step": 8805
},
{
"entropy": 5.513629722595215,
"epoch": 0.7991654571843251,
"grad_norm": 1.0859375,
"learning_rate": 0.0004943489017155906,
"loss": 5.3079,
"mean_token_accuracy": 0.17381245493888856,
"num_tokens": 18287261.0,
"step": 8810
},
{
"entropy": 5.530607414245606,
"epoch": 0.7996190130624092,
"grad_norm": 0.85546875,
"learning_rate": 0.0004943416932455631,
"loss": 5.3088,
"mean_token_accuracy": 0.186485455930233,
"num_tokens": 18299374.0,
"step": 8815
},
{
"entropy": 5.456578254699707,
"epoch": 0.8000725689404935,
"grad_norm": 0.98046875,
"learning_rate": 0.0004943344802394935,
"loss": 5.2843,
"mean_token_accuracy": 0.1819461315870285,
"num_tokens": 18309344.0,
"step": 8820
},
{
"entropy": 5.578178882598877,
"epoch": 0.8005261248185777,
"grad_norm": 0.91015625,
"learning_rate": 0.0004943272626975309,
"loss": 5.2664,
"mean_token_accuracy": 0.18679665625095368,
"num_tokens": 18319778.0,
"step": 8825
},
{
"entropy": 5.534620904922486,
"epoch": 0.8009796806966618,
"grad_norm": 1.0234375,
"learning_rate": 0.0004943200406198246,
"loss": 5.2527,
"mean_token_accuracy": 0.18746354728937148,
"num_tokens": 18329787.0,
"step": 8830
},
{
"entropy": 5.45460352897644,
"epoch": 0.801433236574746,
"grad_norm": 0.92578125,
"learning_rate": 0.000494312814006524,
"loss": 5.2254,
"mean_token_accuracy": 0.19326054751873017,
"num_tokens": 18340670.0,
"step": 8835
},
{
"entropy": 5.486133813858032,
"epoch": 0.8018867924528302,
"grad_norm": 0.8984375,
"learning_rate": 0.0004943055828577784,
"loss": 5.2071,
"mean_token_accuracy": 0.18043867200613023,
"num_tokens": 18352484.0,
"step": 8840
},
{
"entropy": 5.4944739818573,
"epoch": 0.8023403483309144,
"grad_norm": 0.94921875,
"learning_rate": 0.0004942983471737376,
"loss": 5.1898,
"mean_token_accuracy": 0.1878673404455185,
"num_tokens": 18363103.0,
"step": 8845
},
{
"entropy": 5.427543020248413,
"epoch": 0.8027939042089985,
"grad_norm": 0.96875,
"learning_rate": 0.0004942911069545511,
"loss": 5.2345,
"mean_token_accuracy": 0.1885596677660942,
"num_tokens": 18373925.0,
"step": 8850
},
{
"entropy": 5.503770256042481,
"epoch": 0.8032474600870827,
"grad_norm": 0.9453125,
"learning_rate": 0.0004942838622003686,
"loss": 5.3038,
"mean_token_accuracy": 0.18221967965364455,
"num_tokens": 18384081.0,
"step": 8855
},
{
"entropy": 5.5707417011260985,
"epoch": 0.8037010159651669,
"grad_norm": 1.015625,
"learning_rate": 0.0004942766129113399,
"loss": 5.1835,
"mean_token_accuracy": 0.18571112006902696,
"num_tokens": 18392510.0,
"step": 8860
},
{
"entropy": 5.6038836479187015,
"epoch": 0.8041545718432511,
"grad_norm": 0.9921875,
"learning_rate": 0.0004942693590876151,
"loss": 5.3186,
"mean_token_accuracy": 0.18732109665870667,
"num_tokens": 18402809.0,
"step": 8865
},
{
"entropy": 5.571398639678955,
"epoch": 0.8046081277213353,
"grad_norm": 1.0703125,
"learning_rate": 0.000494262100729344,
"loss": 5.3388,
"mean_token_accuracy": 0.18695981055498123,
"num_tokens": 18413342.0,
"step": 8870
},
{
"entropy": 5.530296659469604,
"epoch": 0.8050616835994194,
"grad_norm": 0.859375,
"learning_rate": 0.0004942548378366768,
"loss": 5.3305,
"mean_token_accuracy": 0.18433057516813278,
"num_tokens": 18424735.0,
"step": 8875
},
{
"entropy": 5.518608522415161,
"epoch": 0.8055152394775036,
"grad_norm": 0.87890625,
"learning_rate": 0.0004942475704097635,
"loss": 5.2385,
"mean_token_accuracy": 0.18847633004188538,
"num_tokens": 18434797.0,
"step": 8880
},
{
"entropy": 5.474301147460937,
"epoch": 0.8059687953555879,
"grad_norm": 0.921875,
"learning_rate": 0.0004942402984487548,
"loss": 5.2676,
"mean_token_accuracy": 0.1808571606874466,
"num_tokens": 18445456.0,
"step": 8885
},
{
"entropy": 5.495883464813232,
"epoch": 0.806422351233672,
"grad_norm": 0.97265625,
"learning_rate": 0.0004942330219538008,
"loss": 5.2401,
"mean_token_accuracy": 0.18492464423179628,
"num_tokens": 18456728.0,
"step": 8890
},
{
"entropy": 5.701735019683838,
"epoch": 0.8068759071117562,
"grad_norm": 1.109375,
"learning_rate": 0.000494225740925052,
"loss": 5.3843,
"mean_token_accuracy": 0.18247078955173493,
"num_tokens": 18468813.0,
"step": 8895
},
{
"entropy": 5.576275253295899,
"epoch": 0.8073294629898403,
"grad_norm": 0.98046875,
"learning_rate": 0.0004942184553626591,
"loss": 5.2811,
"mean_token_accuracy": 0.18496292531490327,
"num_tokens": 18477912.0,
"step": 8900
},
{
"entropy": 5.4743218421936035,
"epoch": 0.8077830188679245,
"grad_norm": 0.95703125,
"learning_rate": 0.0004942111652667726,
"loss": 5.2282,
"mean_token_accuracy": 0.19448551237583162,
"num_tokens": 18488115.0,
"step": 8905
},
{
"entropy": 5.5077948570251465,
"epoch": 0.8082365747460087,
"grad_norm": 0.9453125,
"learning_rate": 0.0004942038706375433,
"loss": 5.2667,
"mean_token_accuracy": 0.18090982586145402,
"num_tokens": 18498853.0,
"step": 8910
},
{
"entropy": 5.5216956615448,
"epoch": 0.8086901306240929,
"grad_norm": 0.83984375,
"learning_rate": 0.0004941965714751222,
"loss": 5.3132,
"mean_token_accuracy": 0.18184340447187425,
"num_tokens": 18510047.0,
"step": 8915
},
{
"entropy": 5.518053579330444,
"epoch": 0.809143686502177,
"grad_norm": 1.0,
"learning_rate": 0.0004941892677796601,
"loss": 5.2792,
"mean_token_accuracy": 0.18840029388666152,
"num_tokens": 18519673.0,
"step": 8920
},
{
"entropy": 5.470839023590088,
"epoch": 0.8095972423802612,
"grad_norm": 0.89453125,
"learning_rate": 0.0004941819595513081,
"loss": 5.2919,
"mean_token_accuracy": 0.18582451194524766,
"num_tokens": 18530090.0,
"step": 8925
},
{
"entropy": 5.61807107925415,
"epoch": 0.8100507982583455,
"grad_norm": 0.8359375,
"learning_rate": 0.0004941746467902173,
"loss": 5.3426,
"mean_token_accuracy": 0.18165055215358733,
"num_tokens": 18541862.0,
"step": 8930
},
{
"entropy": 5.5928370475769045,
"epoch": 0.8105043541364296,
"grad_norm": 0.921875,
"learning_rate": 0.000494167329496539,
"loss": 5.336,
"mean_token_accuracy": 0.17665406912565232,
"num_tokens": 18551624.0,
"step": 8935
},
{
"entropy": 5.498425483703613,
"epoch": 0.8109579100145138,
"grad_norm": 1.0234375,
"learning_rate": 0.0004941600076704244,
"loss": 5.2709,
"mean_token_accuracy": 0.18561748266220093,
"num_tokens": 18561784.0,
"step": 8940
},
{
"entropy": 5.549371147155762,
"epoch": 0.8114114658925979,
"grad_norm": 0.984375,
"learning_rate": 0.0004941526813120251,
"loss": 5.3371,
"mean_token_accuracy": 0.18125243335962296,
"num_tokens": 18572110.0,
"step": 8945
},
{
"entropy": 5.547827100753784,
"epoch": 0.8118650217706821,
"grad_norm": 1.015625,
"learning_rate": 0.0004941453504214924,
"loss": 5.2792,
"mean_token_accuracy": 0.17551750242710112,
"num_tokens": 18582012.0,
"step": 8950
},
{
"entropy": 5.443738269805908,
"epoch": 0.8123185776487664,
"grad_norm": 0.9453125,
"learning_rate": 0.0004941380149989781,
"loss": 5.1279,
"mean_token_accuracy": 0.1954687461256981,
"num_tokens": 18592850.0,
"step": 8955
},
{
"entropy": 5.469958543777466,
"epoch": 0.8127721335268505,
"grad_norm": 0.9296875,
"learning_rate": 0.0004941306750446336,
"loss": 5.3026,
"mean_token_accuracy": 0.18596144616603852,
"num_tokens": 18602590.0,
"step": 8960
},
{
"entropy": 5.473424339294434,
"epoch": 0.8132256894049347,
"grad_norm": 1.0,
"learning_rate": 0.000494123330558611,
"loss": 5.1992,
"mean_token_accuracy": 0.1891501232981682,
"num_tokens": 18612607.0,
"step": 8965
},
{
"entropy": 5.497609806060791,
"epoch": 0.8136792452830188,
"grad_norm": 0.90234375,
"learning_rate": 0.0004941159815410622,
"loss": 5.1835,
"mean_token_accuracy": 0.19193311184644699,
"num_tokens": 18622936.0,
"step": 8970
},
{
"entropy": 5.518314170837402,
"epoch": 0.8141328011611031,
"grad_norm": 1.046875,
"learning_rate": 0.000494108627992139,
"loss": 5.3462,
"mean_token_accuracy": 0.17709500193595887,
"num_tokens": 18633997.0,
"step": 8975
},
{
"entropy": 5.47372260093689,
"epoch": 0.8145863570391872,
"grad_norm": 0.94140625,
"learning_rate": 0.0004941012699119936,
"loss": 5.183,
"mean_token_accuracy": 0.19488824605941774,
"num_tokens": 18644347.0,
"step": 8980
},
{
"entropy": 5.532529067993164,
"epoch": 0.8150399129172714,
"grad_norm": 0.9296875,
"learning_rate": 0.0004940939073007779,
"loss": 5.2676,
"mean_token_accuracy": 0.18627604395151137,
"num_tokens": 18654989.0,
"step": 8985
},
{
"entropy": 5.508123207092285,
"epoch": 0.8154934687953556,
"grad_norm": 0.93359375,
"learning_rate": 0.0004940865401586446,
"loss": 5.2336,
"mean_token_accuracy": 0.1885010004043579,
"num_tokens": 18665526.0,
"step": 8990
},
{
"entropy": 5.4809952735900875,
"epoch": 0.8159470246734397,
"grad_norm": 0.9453125,
"learning_rate": 0.0004940791684857458,
"loss": 5.2181,
"mean_token_accuracy": 0.1932695686817169,
"num_tokens": 18676226.0,
"step": 8995
},
{
"entropy": 5.489676904678345,
"epoch": 0.816400580551524,
"grad_norm": 0.9140625,
"learning_rate": 0.0004940717922822339,
"loss": 5.1614,
"mean_token_accuracy": 0.1885027691721916,
"num_tokens": 18686499.0,
"step": 9000
},
{
"epoch": 0.816400580551524,
"eval_entropy": 5.369530144104591,
"eval_loss": 5.262591361999512,
"eval_mean_token_accuracy": 0.19346466002421142,
"eval_num_tokens": 18686499.0,
"eval_runtime": 20.8639,
"eval_samples_per_second": 1694.795,
"eval_steps_per_second": 211.849,
"step": 9000
},
{
"entropy": 5.573095512390137,
"epoch": 0.8168541364296081,
"grad_norm": 0.87109375,
"learning_rate": 0.0004940644115482615,
"loss": 5.2266,
"mean_token_accuracy": 0.18318518102169037,
"num_tokens": 18696543.0,
"step": 9005
},
{
"entropy": 5.455431079864502,
"epoch": 0.8173076923076923,
"grad_norm": 0.953125,
"learning_rate": 0.0004940570262839812,
"loss": 5.2858,
"mean_token_accuracy": 0.18343722075223923,
"num_tokens": 18707029.0,
"step": 9010
},
{
"entropy": 5.486490535736084,
"epoch": 0.8177612481857764,
"grad_norm": 0.9921875,
"learning_rate": 0.0004940496364895457,
"loss": 5.2753,
"mean_token_accuracy": 0.18756048530340194,
"num_tokens": 18716937.0,
"step": 9015
},
{
"entropy": 5.61928539276123,
"epoch": 0.8182148040638607,
"grad_norm": 0.9296875,
"learning_rate": 0.0004940422421651082,
"loss": 5.2611,
"mean_token_accuracy": 0.18611169457435608,
"num_tokens": 18727478.0,
"step": 9020
},
{
"entropy": 5.478516626358032,
"epoch": 0.8186683599419449,
"grad_norm": 1.0234375,
"learning_rate": 0.0004940348433108211,
"loss": 5.177,
"mean_token_accuracy": 0.18611731976270676,
"num_tokens": 18737339.0,
"step": 9025
},
{
"entropy": 5.460969638824463,
"epoch": 0.819121915820029,
"grad_norm": 0.875,
"learning_rate": 0.0004940274399268377,
"loss": 5.3027,
"mean_token_accuracy": 0.18202684968709945,
"num_tokens": 18747810.0,
"step": 9030
},
{
"entropy": 5.50574049949646,
"epoch": 0.8195754716981132,
"grad_norm": 0.98828125,
"learning_rate": 0.0004940200320133109,
"loss": 5.2689,
"mean_token_accuracy": 0.18433540016412736,
"num_tokens": 18757564.0,
"step": 9035
},
{
"entropy": 5.58858289718628,
"epoch": 0.8200290275761973,
"grad_norm": 0.9609375,
"learning_rate": 0.0004940126195703941,
"loss": 5.2907,
"mean_token_accuracy": 0.17852483838796615,
"num_tokens": 18768196.0,
"step": 9040
},
{
"entropy": 5.563227224349975,
"epoch": 0.8204825834542816,
"grad_norm": 0.9140625,
"learning_rate": 0.0004940052025982405,
"loss": 5.2627,
"mean_token_accuracy": 0.19672567546367645,
"num_tokens": 18778350.0,
"step": 9045
},
{
"entropy": 5.480757808685302,
"epoch": 0.8209361393323658,
"grad_norm": 1.046875,
"learning_rate": 0.0004939977810970035,
"loss": 5.2175,
"mean_token_accuracy": 0.18037040233612062,
"num_tokens": 18789825.0,
"step": 9050
},
{
"entropy": 5.474230337142944,
"epoch": 0.8213896952104499,
"grad_norm": 1.015625,
"learning_rate": 0.0004939903550668366,
"loss": 5.2226,
"mean_token_accuracy": 0.18781944513320922,
"num_tokens": 18799793.0,
"step": 9055
},
{
"entropy": 5.498127698898315,
"epoch": 0.8218432510885341,
"grad_norm": 0.89453125,
"learning_rate": 0.0004939829245078934,
"loss": 5.2642,
"mean_token_accuracy": 0.1901239588856697,
"num_tokens": 18809771.0,
"step": 9060
},
{
"entropy": 5.501215839385987,
"epoch": 0.8222968069666183,
"grad_norm": 0.98046875,
"learning_rate": 0.0004939754894203275,
"loss": 5.2714,
"mean_token_accuracy": 0.1833363428711891,
"num_tokens": 18820529.0,
"step": 9065
},
{
"entropy": 5.4365602970123295,
"epoch": 0.8227503628447025,
"grad_norm": 0.984375,
"learning_rate": 0.0004939680498042926,
"loss": 5.1891,
"mean_token_accuracy": 0.19763432294130326,
"num_tokens": 18830797.0,
"step": 9070
},
{
"entropy": 5.490671968460083,
"epoch": 0.8232039187227866,
"grad_norm": 1.0390625,
"learning_rate": 0.0004939606056599428,
"loss": 5.1978,
"mean_token_accuracy": 0.18469800502061845,
"num_tokens": 18840526.0,
"step": 9075
},
{
"entropy": 5.4513916015625,
"epoch": 0.8236574746008708,
"grad_norm": 0.90625,
"learning_rate": 0.0004939531569874316,
"loss": 5.2314,
"mean_token_accuracy": 0.1814821556210518,
"num_tokens": 18851201.0,
"step": 9080
},
{
"entropy": 5.403475999832153,
"epoch": 0.824111030478955,
"grad_norm": 1.0078125,
"learning_rate": 0.0004939457037869135,
"loss": 5.0451,
"mean_token_accuracy": 0.207870976626873,
"num_tokens": 18861676.0,
"step": 9085
},
{
"entropy": 5.449726533889771,
"epoch": 0.8245645863570392,
"grad_norm": 0.9609375,
"learning_rate": 0.0004939382460585424,
"loss": 5.1931,
"mean_token_accuracy": 0.18134815692901612,
"num_tokens": 18872669.0,
"step": 9090
},
{
"entropy": 5.451654291152954,
"epoch": 0.8250181422351234,
"grad_norm": 0.97265625,
"learning_rate": 0.0004939307838024726,
"loss": 5.2454,
"mean_token_accuracy": 0.18319953382015228,
"num_tokens": 18882130.0,
"step": 9095
},
{
"entropy": 5.457068681716919,
"epoch": 0.8254716981132075,
"grad_norm": 0.95703125,
"learning_rate": 0.0004939233170188585,
"loss": 5.2186,
"mean_token_accuracy": 0.18625770807266234,
"num_tokens": 18892457.0,
"step": 9100
},
{
"entropy": 5.556639909744263,
"epoch": 0.8259252539912917,
"grad_norm": 0.95703125,
"learning_rate": 0.0004939158457078543,
"loss": 5.2405,
"mean_token_accuracy": 0.18714095652103424,
"num_tokens": 18903152.0,
"step": 9105
},
{
"entropy": 5.496951627731323,
"epoch": 0.826378809869376,
"grad_norm": 0.9765625,
"learning_rate": 0.0004939083698696146,
"loss": 5.208,
"mean_token_accuracy": 0.18759516775608062,
"num_tokens": 18913740.0,
"step": 9110
},
{
"entropy": 5.478869867324829,
"epoch": 0.8268323657474601,
"grad_norm": 0.98046875,
"learning_rate": 0.000493900889504294,
"loss": 5.2528,
"mean_token_accuracy": 0.1898530900478363,
"num_tokens": 18923501.0,
"step": 9115
},
{
"entropy": 5.482793998718262,
"epoch": 0.8272859216255443,
"grad_norm": 1.0390625,
"learning_rate": 0.0004938934046120473,
"loss": 5.238,
"mean_token_accuracy": 0.18646990209817887,
"num_tokens": 18933255.0,
"step": 9120
},
{
"entropy": 5.451048803329468,
"epoch": 0.8277394775036284,
"grad_norm": 1.046875,
"learning_rate": 0.0004938859151930293,
"loss": 5.2061,
"mean_token_accuracy": 0.18621120303869249,
"num_tokens": 18943649.0,
"step": 9125
},
{
"entropy": 5.4907585144042965,
"epoch": 0.8281930333817126,
"grad_norm": 1.0078125,
"learning_rate": 0.0004938784212473947,
"loss": 5.1236,
"mean_token_accuracy": 0.1988166391849518,
"num_tokens": 18953187.0,
"step": 9130
},
{
"entropy": 5.570322036743164,
"epoch": 0.8286465892597968,
"grad_norm": 0.96875,
"learning_rate": 0.0004938709227752985,
"loss": 5.2262,
"mean_token_accuracy": 0.19010987728834153,
"num_tokens": 18963015.0,
"step": 9135
},
{
"entropy": 5.494019985198975,
"epoch": 0.829100145137881,
"grad_norm": 0.91015625,
"learning_rate": 0.000493863419776896,
"loss": 5.2711,
"mean_token_accuracy": 0.18064223527908324,
"num_tokens": 18974184.0,
"step": 9140
},
{
"entropy": 5.450629234313965,
"epoch": 0.8295537010159652,
"grad_norm": 0.984375,
"learning_rate": 0.0004938559122523421,
"loss": 5.2186,
"mean_token_accuracy": 0.1903683364391327,
"num_tokens": 18985036.0,
"step": 9145
},
{
"entropy": 5.573251056671142,
"epoch": 0.8300072568940493,
"grad_norm": 0.89453125,
"learning_rate": 0.0004938484002017921,
"loss": 5.3258,
"mean_token_accuracy": 0.18096655309200288,
"num_tokens": 18995237.0,
"step": 9150
},
{
"entropy": 5.539033031463623,
"epoch": 0.8304608127721336,
"grad_norm": 0.8828125,
"learning_rate": 0.0004938408836254015,
"loss": 5.3335,
"mean_token_accuracy": 0.17874883860349655,
"num_tokens": 19006272.0,
"step": 9155
},
{
"entropy": 5.459480285644531,
"epoch": 0.8309143686502177,
"grad_norm": 0.93359375,
"learning_rate": 0.0004938333625233258,
"loss": 5.2308,
"mean_token_accuracy": 0.18578039556741716,
"num_tokens": 19016728.0,
"step": 9160
},
{
"entropy": 5.545877456665039,
"epoch": 0.8313679245283019,
"grad_norm": 0.91015625,
"learning_rate": 0.0004938258368957202,
"loss": 5.2922,
"mean_token_accuracy": 0.17642530798912048,
"num_tokens": 19028024.0,
"step": 9165
},
{
"entropy": 5.562115573883057,
"epoch": 0.831821480406386,
"grad_norm": 0.95703125,
"learning_rate": 0.0004938183067427406,
"loss": 5.2373,
"mean_token_accuracy": 0.1918642923235893,
"num_tokens": 19038218.0,
"step": 9170
},
{
"entropy": 5.454094696044922,
"epoch": 0.8322750362844702,
"grad_norm": 1.078125,
"learning_rate": 0.0004938107720645426,
"loss": 5.2595,
"mean_token_accuracy": 0.18231881111860276,
"num_tokens": 19048574.0,
"step": 9175
},
{
"entropy": 5.495149421691894,
"epoch": 0.8327285921625545,
"grad_norm": 0.96484375,
"learning_rate": 0.0004938032328612821,
"loss": 5.2385,
"mean_token_accuracy": 0.1852733686566353,
"num_tokens": 19057721.0,
"step": 9180
},
{
"entropy": 5.491014194488526,
"epoch": 0.8331821480406386,
"grad_norm": 0.984375,
"learning_rate": 0.000493795689133115,
"loss": 5.2183,
"mean_token_accuracy": 0.17995734810829161,
"num_tokens": 19067582.0,
"step": 9185
},
{
"entropy": 5.468677616119384,
"epoch": 0.8336357039187228,
"grad_norm": 0.8671875,
"learning_rate": 0.0004937881408801974,
"loss": 5.1722,
"mean_token_accuracy": 0.1918642297387123,
"num_tokens": 19077896.0,
"step": 9190
},
{
"entropy": 5.4998781204223635,
"epoch": 0.8340892597968069,
"grad_norm": 0.94140625,
"learning_rate": 0.000493780588102685,
"loss": 5.2799,
"mean_token_accuracy": 0.19266057163476943,
"num_tokens": 19088082.0,
"step": 9195
},
{
"entropy": 5.577761936187744,
"epoch": 0.8345428156748912,
"grad_norm": 0.91796875,
"learning_rate": 0.0004937730308007346,
"loss": 5.2969,
"mean_token_accuracy": 0.18470134884119033,
"num_tokens": 19097176.0,
"step": 9200
},
{
"entropy": 5.499684047698975,
"epoch": 0.8349963715529753,
"grad_norm": 0.97265625,
"learning_rate": 0.0004937654689745021,
"loss": 5.1837,
"mean_token_accuracy": 0.19089347273111343,
"num_tokens": 19107835.0,
"step": 9205
},
{
"entropy": 5.430692720413208,
"epoch": 0.8354499274310595,
"grad_norm": 0.97265625,
"learning_rate": 0.0004937579026241439,
"loss": 5.2017,
"mean_token_accuracy": 0.1925992026925087,
"num_tokens": 19118028.0,
"step": 9210
},
{
"entropy": 5.486698341369629,
"epoch": 0.8359034833091437,
"grad_norm": 0.91015625,
"learning_rate": 0.0004937503317498165,
"loss": 5.2276,
"mean_token_accuracy": 0.19192424416542053,
"num_tokens": 19129223.0,
"step": 9215
},
{
"entropy": 5.497008609771728,
"epoch": 0.8363570391872278,
"grad_norm": 0.87109375,
"learning_rate": 0.0004937427563516765,
"loss": 5.3335,
"mean_token_accuracy": 0.18186921924352645,
"num_tokens": 19140048.0,
"step": 9220
},
{
"entropy": 5.568466663360596,
"epoch": 0.8368105950653121,
"grad_norm": 0.91796875,
"learning_rate": 0.0004937351764298805,
"loss": 5.2985,
"mean_token_accuracy": 0.1766539365053177,
"num_tokens": 19150853.0,
"step": 9225
},
{
"entropy": 5.551784181594849,
"epoch": 0.8372641509433962,
"grad_norm": 0.96875,
"learning_rate": 0.0004937275919845855,
"loss": 5.2839,
"mean_token_accuracy": 0.18331123888492584,
"num_tokens": 19161645.0,
"step": 9230
},
{
"entropy": 5.518892812728882,
"epoch": 0.8377177068214804,
"grad_norm": 0.8671875,
"learning_rate": 0.0004937200030159481,
"loss": 5.3715,
"mean_token_accuracy": 0.17670943588018417,
"num_tokens": 19173276.0,
"step": 9235
},
{
"entropy": 5.517861652374267,
"epoch": 0.8381712626995645,
"grad_norm": 0.90625,
"learning_rate": 0.0004937124095241252,
"loss": 5.248,
"mean_token_accuracy": 0.19307708144187927,
"num_tokens": 19182949.0,
"step": 9240
},
{
"entropy": 5.553634691238403,
"epoch": 0.8386248185776488,
"grad_norm": 0.9375,
"learning_rate": 0.0004937048115092739,
"loss": 5.2414,
"mean_token_accuracy": 0.18353976160287858,
"num_tokens": 19193504.0,
"step": 9245
},
{
"entropy": 5.5102275848388675,
"epoch": 0.839078374455733,
"grad_norm": 0.984375,
"learning_rate": 0.0004936972089715515,
"loss": 5.3207,
"mean_token_accuracy": 0.18096430003643035,
"num_tokens": 19203406.0,
"step": 9250
},
{
"entropy": 5.4884428024292,
"epoch": 0.8395319303338171,
"grad_norm": 0.92578125,
"learning_rate": 0.000493689601911115,
"loss": 5.2644,
"mean_token_accuracy": 0.18698281645774842,
"num_tokens": 19214325.0,
"step": 9255
},
{
"entropy": 5.499520349502563,
"epoch": 0.8399854862119013,
"grad_norm": 0.94921875,
"learning_rate": 0.0004936819903281219,
"loss": 5.3032,
"mean_token_accuracy": 0.17399130165576934,
"num_tokens": 19226466.0,
"step": 9260
},
{
"entropy": 5.601823616027832,
"epoch": 0.8404390420899854,
"grad_norm": 0.91796875,
"learning_rate": 0.0004936743742227295,
"loss": 5.3281,
"mean_token_accuracy": 0.1854452520608902,
"num_tokens": 19237032.0,
"step": 9265
},
{
"entropy": 5.55287766456604,
"epoch": 0.8408925979680697,
"grad_norm": 0.94921875,
"learning_rate": 0.0004936667535950952,
"loss": 5.3213,
"mean_token_accuracy": 0.1790757581591606,
"num_tokens": 19248009.0,
"step": 9270
},
{
"entropy": 5.548722505569458,
"epoch": 0.8413461538461539,
"grad_norm": 0.8828125,
"learning_rate": 0.0004936591284453768,
"loss": 5.372,
"mean_token_accuracy": 0.1818295806646347,
"num_tokens": 19259363.0,
"step": 9275
},
{
"entropy": 5.5150541305542,
"epoch": 0.841799709724238,
"grad_norm": 0.953125,
"learning_rate": 0.0004936514987737319,
"loss": 5.2678,
"mean_token_accuracy": 0.18624502122402192,
"num_tokens": 19269191.0,
"step": 9280
},
{
"entropy": 5.576925992965698,
"epoch": 0.8422532656023222,
"grad_norm": 0.93359375,
"learning_rate": 0.0004936438645803182,
"loss": 5.2344,
"mean_token_accuracy": 0.18304483443498612,
"num_tokens": 19280278.0,
"step": 9285
},
{
"entropy": 5.511117076873779,
"epoch": 0.8427068214804064,
"grad_norm": 0.96484375,
"learning_rate": 0.0004936362258652938,
"loss": 5.2203,
"mean_token_accuracy": 0.19137237817049027,
"num_tokens": 19290130.0,
"step": 9290
},
{
"entropy": 5.563754463195801,
"epoch": 0.8431603773584906,
"grad_norm": 0.88671875,
"learning_rate": 0.0004936285826288165,
"loss": 5.3249,
"mean_token_accuracy": 0.18893066346645354,
"num_tokens": 19301669.0,
"step": 9295
},
{
"entropy": 5.46899380683899,
"epoch": 0.8436139332365747,
"grad_norm": 0.84765625,
"learning_rate": 0.0004936209348710444,
"loss": 5.2003,
"mean_token_accuracy": 0.18793248534202575,
"num_tokens": 19312490.0,
"step": 9300
},
{
"entropy": 5.502799987792969,
"epoch": 0.8440674891146589,
"grad_norm": 0.9296875,
"learning_rate": 0.0004936132825921355,
"loss": 5.2355,
"mean_token_accuracy": 0.18560146540403366,
"num_tokens": 19321743.0,
"step": 9305
},
{
"entropy": 5.462376737594605,
"epoch": 0.844521044992743,
"grad_norm": 0.98828125,
"learning_rate": 0.0004936056257922484,
"loss": 5.1537,
"mean_token_accuracy": 0.18894000202417374,
"num_tokens": 19332149.0,
"step": 9310
},
{
"entropy": 5.499111986160278,
"epoch": 0.8449746008708273,
"grad_norm": 1.0,
"learning_rate": 0.0004935979644715412,
"loss": 5.259,
"mean_token_accuracy": 0.1819048583507538,
"num_tokens": 19341655.0,
"step": 9315
},
{
"entropy": 5.5421395778656,
"epoch": 0.8454281567489115,
"grad_norm": 0.984375,
"learning_rate": 0.0004935902986301724,
"loss": 5.2892,
"mean_token_accuracy": 0.17833438664674758,
"num_tokens": 19351404.0,
"step": 9320
},
{
"entropy": 5.5445465564727785,
"epoch": 0.8458817126269956,
"grad_norm": 0.96875,
"learning_rate": 0.0004935826282683005,
"loss": 5.3296,
"mean_token_accuracy": 0.17884413450956343,
"num_tokens": 19361788.0,
"step": 9325
},
{
"entropy": 5.52082576751709,
"epoch": 0.8463352685050798,
"grad_norm": 0.953125,
"learning_rate": 0.0004935749533860842,
"loss": 5.2941,
"mean_token_accuracy": 0.1879544198513031,
"num_tokens": 19372902.0,
"step": 9330
},
{
"entropy": 5.456793355941772,
"epoch": 0.846788824383164,
"grad_norm": 0.91015625,
"learning_rate": 0.000493567273983682,
"loss": 5.164,
"mean_token_accuracy": 0.18452823758125306,
"num_tokens": 19382181.0,
"step": 9335
},
{
"entropy": 5.480916547775268,
"epoch": 0.8472423802612482,
"grad_norm": 0.87109375,
"learning_rate": 0.000493559590061253,
"loss": 5.2346,
"mean_token_accuracy": 0.18342400044202806,
"num_tokens": 19393692.0,
"step": 9340
},
{
"entropy": 5.526205110549927,
"epoch": 0.8476959361393324,
"grad_norm": 0.92578125,
"learning_rate": 0.0004935519016189559,
"loss": 5.2629,
"mean_token_accuracy": 0.19290686994791031,
"num_tokens": 19404587.0,
"step": 9345
},
{
"entropy": 5.486667442321777,
"epoch": 0.8481494920174165,
"grad_norm": 0.921875,
"learning_rate": 0.0004935442086569498,
"loss": 5.235,
"mean_token_accuracy": 0.1877764046192169,
"num_tokens": 19415296.0,
"step": 9350
},
{
"entropy": 5.559003782272339,
"epoch": 0.8486030478955007,
"grad_norm": 0.90234375,
"learning_rate": 0.0004935365111753939,
"loss": 5.3125,
"mean_token_accuracy": 0.1837704747915268,
"num_tokens": 19426119.0,
"step": 9355
},
{
"entropy": 5.567704677581787,
"epoch": 0.8490566037735849,
"grad_norm": 0.9609375,
"learning_rate": 0.0004935288091744472,
"loss": 5.2818,
"mean_token_accuracy": 0.17886171042919158,
"num_tokens": 19436828.0,
"step": 9360
},
{
"entropy": 5.552606248855591,
"epoch": 0.8495101596516691,
"grad_norm": 1.015625,
"learning_rate": 0.000493521102654269,
"loss": 5.2655,
"mean_token_accuracy": 0.18383633941411973,
"num_tokens": 19446583.0,
"step": 9365
},
{
"entropy": 5.441853189468384,
"epoch": 0.8499637155297532,
"grad_norm": 0.95703125,
"learning_rate": 0.0004935133916150187,
"loss": 5.1426,
"mean_token_accuracy": 0.19354938864707946,
"num_tokens": 19456757.0,
"step": 9370
},
{
"entropy": 5.4375901222229,
"epoch": 0.8504172714078374,
"grad_norm": 1.03125,
"learning_rate": 0.0004935056760568558,
"loss": 5.2203,
"mean_token_accuracy": 0.18387676924467086,
"num_tokens": 19467624.0,
"step": 9375
},
{
"entropy": 5.507636260986328,
"epoch": 0.8508708272859217,
"grad_norm": 1.015625,
"learning_rate": 0.0004934979559799399,
"loss": 5.1784,
"mean_token_accuracy": 0.19261115789413452,
"num_tokens": 19477759.0,
"step": 9380
},
{
"entropy": 5.50710015296936,
"epoch": 0.8513243831640058,
"grad_norm": 0.890625,
"learning_rate": 0.0004934902313844306,
"loss": 5.1971,
"mean_token_accuracy": 0.1891529604792595,
"num_tokens": 19488877.0,
"step": 9385
},
{
"entropy": 5.452629899978637,
"epoch": 0.85177793904209,
"grad_norm": 0.8828125,
"learning_rate": 0.0004934825022704875,
"loss": 5.1932,
"mean_token_accuracy": 0.18864367604255677,
"num_tokens": 19499170.0,
"step": 9390
},
{
"entropy": 5.495832824707032,
"epoch": 0.8522314949201741,
"grad_norm": 0.93359375,
"learning_rate": 0.0004934747686382707,
"loss": 5.2215,
"mean_token_accuracy": 0.1803721621632576,
"num_tokens": 19509944.0,
"step": 9395
},
{
"entropy": 5.506520843505859,
"epoch": 0.8526850507982583,
"grad_norm": 0.953125,
"learning_rate": 0.0004934670304879398,
"loss": 5.1906,
"mean_token_accuracy": 0.18765731602907182,
"num_tokens": 19519014.0,
"step": 9400
},
{
"entropy": 5.4330939769744875,
"epoch": 0.8531386066763426,
"grad_norm": 1.015625,
"learning_rate": 0.0004934592878196553,
"loss": 5.2022,
"mean_token_accuracy": 0.18718813359737396,
"num_tokens": 19528586.0,
"step": 9405
},
{
"entropy": 5.527695608139038,
"epoch": 0.8535921625544267,
"grad_norm": 0.95703125,
"learning_rate": 0.000493451540633577,
"loss": 5.3372,
"mean_token_accuracy": 0.18059038519859313,
"num_tokens": 19539912.0,
"step": 9410
},
{
"entropy": 5.507913064956665,
"epoch": 0.8540457184325109,
"grad_norm": 0.98046875,
"learning_rate": 0.0004934437889298651,
"loss": 5.2781,
"mean_token_accuracy": 0.1787128046154976,
"num_tokens": 19549860.0,
"step": 9415
},
{
"entropy": 5.537213087081909,
"epoch": 0.854499274310595,
"grad_norm": 0.9296875,
"learning_rate": 0.0004934360327086799,
"loss": 5.2581,
"mean_token_accuracy": 0.1831861764192581,
"num_tokens": 19559446.0,
"step": 9420
},
{
"entropy": 5.57012186050415,
"epoch": 0.8549528301886793,
"grad_norm": 0.9140625,
"learning_rate": 0.000493428271970182,
"loss": 5.2586,
"mean_token_accuracy": 0.19021736234426498,
"num_tokens": 19570282.0,
"step": 9425
},
{
"entropy": 5.526591825485229,
"epoch": 0.8554063860667634,
"grad_norm": 0.953125,
"learning_rate": 0.0004934205067145318,
"loss": 5.1816,
"mean_token_accuracy": 0.18626246899366378,
"num_tokens": 19581087.0,
"step": 9430
},
{
"entropy": 5.495096015930176,
"epoch": 0.8558599419448476,
"grad_norm": 0.90625,
"learning_rate": 0.0004934127369418898,
"loss": 5.2741,
"mean_token_accuracy": 0.19055798500776291,
"num_tokens": 19592100.0,
"step": 9435
},
{
"entropy": 5.49104151725769,
"epoch": 0.8563134978229318,
"grad_norm": 1.0,
"learning_rate": 0.0004934049626524168,
"loss": 5.2558,
"mean_token_accuracy": 0.18784750998020172,
"num_tokens": 19602557.0,
"step": 9440
},
{
"entropy": 5.539940404891968,
"epoch": 0.8567670537010159,
"grad_norm": 0.94921875,
"learning_rate": 0.0004933971838462734,
"loss": 5.2965,
"mean_token_accuracy": 0.17805867195129393,
"num_tokens": 19613643.0,
"step": 9445
},
{
"entropy": 5.516980266571045,
"epoch": 0.8572206095791002,
"grad_norm": 0.94921875,
"learning_rate": 0.0004933894005236207,
"loss": 5.212,
"mean_token_accuracy": 0.1881819874048233,
"num_tokens": 19624180.0,
"step": 9450
},
{
"entropy": 5.535392236709595,
"epoch": 0.8576741654571843,
"grad_norm": 0.94140625,
"learning_rate": 0.0004933816126846195,
"loss": 5.2281,
"mean_token_accuracy": 0.18857050836086273,
"num_tokens": 19634799.0,
"step": 9455
},
{
"entropy": 5.4610528469085695,
"epoch": 0.8581277213352685,
"grad_norm": 0.95703125,
"learning_rate": 0.0004933738203294309,
"loss": 5.2042,
"mean_token_accuracy": 0.1919635698199272,
"num_tokens": 19644618.0,
"step": 9460
},
{
"entropy": 5.458015203475952,
"epoch": 0.8585812772133526,
"grad_norm": 0.9609375,
"learning_rate": 0.000493366023458216,
"loss": 5.2114,
"mean_token_accuracy": 0.18096943944692612,
"num_tokens": 19655550.0,
"step": 9465
},
{
"entropy": 5.538266515731811,
"epoch": 0.8590348330914369,
"grad_norm": 1.0078125,
"learning_rate": 0.0004933582220711362,
"loss": 5.2769,
"mean_token_accuracy": 0.19205545336008073,
"num_tokens": 19665734.0,
"step": 9470
},
{
"entropy": 5.527783441543579,
"epoch": 0.8594883889695211,
"grad_norm": 0.89453125,
"learning_rate": 0.0004933504161683526,
"loss": 5.2234,
"mean_token_accuracy": 0.1832062631845474,
"num_tokens": 19677130.0,
"step": 9475
},
{
"entropy": 5.452576208114624,
"epoch": 0.8599419448476052,
"grad_norm": 0.92578125,
"learning_rate": 0.0004933426057500269,
"loss": 5.1462,
"mean_token_accuracy": 0.19649119079113006,
"num_tokens": 19687374.0,
"step": 9480
},
{
"entropy": 5.5538092136383055,
"epoch": 0.8603955007256894,
"grad_norm": 0.92578125,
"learning_rate": 0.0004933347908163203,
"loss": 5.2585,
"mean_token_accuracy": 0.1857781171798706,
"num_tokens": 19698182.0,
"step": 9485
},
{
"entropy": 5.515981197357178,
"epoch": 0.8608490566037735,
"grad_norm": 0.99609375,
"learning_rate": 0.0004933269713673947,
"loss": 5.2181,
"mean_token_accuracy": 0.18882352858781815,
"num_tokens": 19707664.0,
"step": 9490
},
{
"entropy": 5.435214948654175,
"epoch": 0.8613026124818578,
"grad_norm": 0.953125,
"learning_rate": 0.0004933191474034116,
"loss": 5.2636,
"mean_token_accuracy": 0.18053202778100969,
"num_tokens": 19717647.0,
"step": 9495
},
{
"entropy": 5.499503469467163,
"epoch": 0.861756168359942,
"grad_norm": 0.9375,
"learning_rate": 0.000493311318924533,
"loss": 5.2917,
"mean_token_accuracy": 0.18424864560365678,
"num_tokens": 19727444.0,
"step": 9500
},
{
"entropy": 5.5348368167877195,
"epoch": 0.8622097242380261,
"grad_norm": 0.95703125,
"learning_rate": 0.0004933034859309208,
"loss": 5.3074,
"mean_token_accuracy": 0.17919572144746782,
"num_tokens": 19738719.0,
"step": 9505
},
{
"entropy": 5.5768194675445555,
"epoch": 0.8626632801161103,
"grad_norm": 0.88671875,
"learning_rate": 0.0004932956484227367,
"loss": 5.2775,
"mean_token_accuracy": 0.19209030717611314,
"num_tokens": 19749786.0,
"step": 9510
},
{
"entropy": 5.507620334625244,
"epoch": 0.8631168359941945,
"grad_norm": 0.953125,
"learning_rate": 0.000493287806400143,
"loss": 5.2268,
"mean_token_accuracy": 0.19589438587427138,
"num_tokens": 19759623.0,
"step": 9515
},
{
"entropy": 5.516129493713379,
"epoch": 0.8635703918722787,
"grad_norm": 0.9375,
"learning_rate": 0.0004932799598633018,
"loss": 5.2349,
"mean_token_accuracy": 0.1868974268436432,
"num_tokens": 19770117.0,
"step": 9520
},
{
"entropy": 5.536385726928711,
"epoch": 0.8640239477503628,
"grad_norm": 1.0234375,
"learning_rate": 0.0004932721088123753,
"loss": 5.3231,
"mean_token_accuracy": 0.18734574615955352,
"num_tokens": 19780748.0,
"step": 9525
},
{
"entropy": 5.513698625564575,
"epoch": 0.864477503628447,
"grad_norm": 0.88671875,
"learning_rate": 0.0004932642532475262,
"loss": 5.2632,
"mean_token_accuracy": 0.18844823986291886,
"num_tokens": 19791816.0,
"step": 9530
},
{
"entropy": 5.5973162174224855,
"epoch": 0.8649310595065312,
"grad_norm": 0.984375,
"learning_rate": 0.0004932563931689168,
"loss": 5.3348,
"mean_token_accuracy": 0.18349990099668503,
"num_tokens": 19801096.0,
"step": 9535
},
{
"entropy": 5.519701194763184,
"epoch": 0.8653846153846154,
"grad_norm": 1.0234375,
"learning_rate": 0.0004932485285767093,
"loss": 5.1903,
"mean_token_accuracy": 0.1904284492135048,
"num_tokens": 19811258.0,
"step": 9540
},
{
"entropy": 5.541377305984497,
"epoch": 0.8658381712626996,
"grad_norm": 1.0,
"learning_rate": 0.0004932406594710667,
"loss": 5.2816,
"mean_token_accuracy": 0.19132487922906877,
"num_tokens": 19821765.0,
"step": 9545
},
{
"entropy": 5.5216514587402346,
"epoch": 0.8662917271407837,
"grad_norm": 0.99609375,
"learning_rate": 0.0004932327858521517,
"loss": 5.1445,
"mean_token_accuracy": 0.19365739524364473,
"num_tokens": 19831537.0,
"step": 9550
},
{
"entropy": 5.463690662384034,
"epoch": 0.8667452830188679,
"grad_norm": 0.96875,
"learning_rate": 0.000493224907720127,
"loss": 5.1567,
"mean_token_accuracy": 0.18910293579101561,
"num_tokens": 19841830.0,
"step": 9555
},
{
"entropy": 5.486150312423706,
"epoch": 0.8671988388969522,
"grad_norm": 0.92578125,
"learning_rate": 0.0004932170250751556,
"loss": 5.2631,
"mean_token_accuracy": 0.18096060454845428,
"num_tokens": 19851896.0,
"step": 9560
},
{
"entropy": 5.550594329833984,
"epoch": 0.8676523947750363,
"grad_norm": 1.0078125,
"learning_rate": 0.0004932091379174006,
"loss": 5.298,
"mean_token_accuracy": 0.17869172990322113,
"num_tokens": 19861735.0,
"step": 9565
},
{
"entropy": 5.523482179641723,
"epoch": 0.8681059506531205,
"grad_norm": 0.91015625,
"learning_rate": 0.000493201246247025,
"loss": 5.1781,
"mean_token_accuracy": 0.19788408875465394,
"num_tokens": 19872333.0,
"step": 9570
},
{
"entropy": 5.48471360206604,
"epoch": 0.8685595065312046,
"grad_norm": 0.890625,
"learning_rate": 0.000493193350064192,
"loss": 5.1956,
"mean_token_accuracy": 0.1868165984749794,
"num_tokens": 19882896.0,
"step": 9575
},
{
"entropy": 5.470310354232788,
"epoch": 0.8690130624092888,
"grad_norm": 0.91015625,
"learning_rate": 0.0004931854493690649,
"loss": 5.1672,
"mean_token_accuracy": 0.18842218518257142,
"num_tokens": 19893767.0,
"step": 9580
},
{
"entropy": 5.4305318832397464,
"epoch": 0.869466618287373,
"grad_norm": 0.91796875,
"learning_rate": 0.0004931775441618071,
"loss": 5.1424,
"mean_token_accuracy": 0.19409455507993698,
"num_tokens": 19904242.0,
"step": 9585
},
{
"entropy": 5.474945592880249,
"epoch": 0.8699201741654572,
"grad_norm": 0.96484375,
"learning_rate": 0.0004931696344425821,
"loss": 5.2879,
"mean_token_accuracy": 0.18741385638713837,
"num_tokens": 19913898.0,
"step": 9590
},
{
"entropy": 5.589535808563232,
"epoch": 0.8703737300435413,
"grad_norm": 0.91015625,
"learning_rate": 0.0004931617202115535,
"loss": 5.2727,
"mean_token_accuracy": 0.18905438631772994,
"num_tokens": 19923756.0,
"step": 9595
},
{
"entropy": 5.507710361480713,
"epoch": 0.8708272859216255,
"grad_norm": 1.015625,
"learning_rate": 0.0004931538014688849,
"loss": 5.1986,
"mean_token_accuracy": 0.19067411869764328,
"num_tokens": 19932720.0,
"step": 9600
},
{
"entropy": 5.565124797821045,
"epoch": 0.8712808417997098,
"grad_norm": 1.109375,
"learning_rate": 0.0004931458782147401,
"loss": 5.4096,
"mean_token_accuracy": 0.17565523087978363,
"num_tokens": 19942962.0,
"step": 9605
},
{
"entropy": 5.584523773193359,
"epoch": 0.8717343976777939,
"grad_norm": 1.0078125,
"learning_rate": 0.000493137950449283,
"loss": 5.2822,
"mean_token_accuracy": 0.19391026049852372,
"num_tokens": 19953735.0,
"step": 9610
},
{
"entropy": 5.485487079620361,
"epoch": 0.8721879535558781,
"grad_norm": 0.921875,
"learning_rate": 0.0004931300181726775,
"loss": 5.2473,
"mean_token_accuracy": 0.18759722262620926,
"num_tokens": 19964416.0,
"step": 9615
},
{
"entropy": 5.493893098831177,
"epoch": 0.8726415094339622,
"grad_norm": 0.99609375,
"learning_rate": 0.0004931220813850875,
"loss": 5.2659,
"mean_token_accuracy": 0.18439658731222153,
"num_tokens": 19974531.0,
"step": 9620
},
{
"entropy": 5.551490545272827,
"epoch": 0.8730950653120464,
"grad_norm": 0.98046875,
"learning_rate": 0.0004931141400866773,
"loss": 5.2484,
"mean_token_accuracy": 0.18889303654432296,
"num_tokens": 19984964.0,
"step": 9625
},
{
"entropy": 5.4911480903625485,
"epoch": 0.8735486211901307,
"grad_norm": 0.8984375,
"learning_rate": 0.0004931061942776112,
"loss": 5.186,
"mean_token_accuracy": 0.19389327317476274,
"num_tokens": 19995568.0,
"step": 9630
},
{
"entropy": 5.588620948791504,
"epoch": 0.8740021770682148,
"grad_norm": 1.0234375,
"learning_rate": 0.0004930982439580534,
"loss": 5.3836,
"mean_token_accuracy": 0.17700846791267394,
"num_tokens": 20005856.0,
"step": 9635
},
{
"entropy": 5.57611494064331,
"epoch": 0.874455732946299,
"grad_norm": 1.1015625,
"learning_rate": 0.0004930902891281683,
"loss": 5.283,
"mean_token_accuracy": 0.18396832197904586,
"num_tokens": 20015683.0,
"step": 9640
},
{
"entropy": 5.41850471496582,
"epoch": 0.8749092888243831,
"grad_norm": 0.953125,
"learning_rate": 0.0004930823297881205,
"loss": 5.1013,
"mean_token_accuracy": 0.19724885672330855,
"num_tokens": 20025998.0,
"step": 9645
},
{
"entropy": 5.539435148239136,
"epoch": 0.8753628447024674,
"grad_norm": 0.9296875,
"learning_rate": 0.0004930743659380745,
"loss": 5.2599,
"mean_token_accuracy": 0.18563385009765626,
"num_tokens": 20036486.0,
"step": 9650
},
{
"entropy": 5.479112577438355,
"epoch": 0.8758164005805515,
"grad_norm": 0.9453125,
"learning_rate": 0.000493066397578195,
"loss": 5.1674,
"mean_token_accuracy": 0.19481005817651748,
"num_tokens": 20046784.0,
"step": 9655
},
{
"entropy": 5.480585336685181,
"epoch": 0.8762699564586357,
"grad_norm": 1.0,
"learning_rate": 0.0004930584247086468,
"loss": 5.2326,
"mean_token_accuracy": 0.1877565011382103,
"num_tokens": 20056999.0,
"step": 9660
},
{
"entropy": 5.465516996383667,
"epoch": 0.8767235123367199,
"grad_norm": 0.9609375,
"learning_rate": 0.000493050447329595,
"loss": 5.122,
"mean_token_accuracy": 0.19252620935440062,
"num_tokens": 20067612.0,
"step": 9665
},
{
"entropy": 5.5493165969848635,
"epoch": 0.877177068214804,
"grad_norm": 0.97265625,
"learning_rate": 0.0004930424654412043,
"loss": 5.275,
"mean_token_accuracy": 0.1878029227256775,
"num_tokens": 20077330.0,
"step": 9670
},
{
"entropy": 5.546081256866455,
"epoch": 0.8776306240928883,
"grad_norm": 1.0390625,
"learning_rate": 0.0004930344790436398,
"loss": 5.2745,
"mean_token_accuracy": 0.17999620139598846,
"num_tokens": 20086801.0,
"step": 9675
},
{
"entropy": 5.52485613822937,
"epoch": 0.8780841799709724,
"grad_norm": 0.94921875,
"learning_rate": 0.0004930264881370666,
"loss": 5.2166,
"mean_token_accuracy": 0.18604013770818711,
"num_tokens": 20097047.0,
"step": 9680
},
{
"entropy": 5.471764802932739,
"epoch": 0.8785377358490566,
"grad_norm": 0.93359375,
"learning_rate": 0.0004930184927216504,
"loss": 5.2546,
"mean_token_accuracy": 0.18681764751672744,
"num_tokens": 20107185.0,
"step": 9685
},
{
"entropy": 5.53357629776001,
"epoch": 0.8789912917271407,
"grad_norm": 0.93359375,
"learning_rate": 0.0004930104927975559,
"loss": 5.1947,
"mean_token_accuracy": 0.19048551768064498,
"num_tokens": 20118493.0,
"step": 9690
},
{
"entropy": 5.484797573089599,
"epoch": 0.879444847605225,
"grad_norm": 0.93359375,
"learning_rate": 0.0004930024883649489,
"loss": 5.2524,
"mean_token_accuracy": 0.18750473260879516,
"num_tokens": 20129294.0,
"step": 9695
},
{
"entropy": 5.442465543746948,
"epoch": 0.8798984034833092,
"grad_norm": 1.0078125,
"learning_rate": 0.000492994479423995,
"loss": 5.2124,
"mean_token_accuracy": 0.19394612908363343,
"num_tokens": 20139194.0,
"step": 9700
},
{
"entropy": 5.467197561264038,
"epoch": 0.8803519593613933,
"grad_norm": 0.94921875,
"learning_rate": 0.0004929864659748597,
"loss": 5.2374,
"mean_token_accuracy": 0.18322389721870422,
"num_tokens": 20149681.0,
"step": 9705
},
{
"entropy": 5.549167013168335,
"epoch": 0.8808055152394775,
"grad_norm": 0.90625,
"learning_rate": 0.0004929784480177087,
"loss": 5.3222,
"mean_token_accuracy": 0.18286841213703156,
"num_tokens": 20160557.0,
"step": 9710
},
{
"entropy": 5.519404029846191,
"epoch": 0.8812590711175616,
"grad_norm": 0.94921875,
"learning_rate": 0.0004929704255527078,
"loss": 5.1887,
"mean_token_accuracy": 0.187822188436985,
"num_tokens": 20171047.0,
"step": 9715
},
{
"entropy": 5.523070621490478,
"epoch": 0.8817126269956459,
"grad_norm": 0.89453125,
"learning_rate": 0.000492962398580023,
"loss": 5.3707,
"mean_token_accuracy": 0.17758570462465287,
"num_tokens": 20182148.0,
"step": 9720
},
{
"entropy": 5.421023654937744,
"epoch": 0.88216618287373,
"grad_norm": 0.92578125,
"learning_rate": 0.0004929543670998203,
"loss": 5.1276,
"mean_token_accuracy": 0.19135259091854095,
"num_tokens": 20192131.0,
"step": 9725
},
{
"entropy": 5.531173181533814,
"epoch": 0.8826197387518142,
"grad_norm": 0.8515625,
"learning_rate": 0.0004929463311122657,
"loss": 5.2426,
"mean_token_accuracy": 0.1861494779586792,
"num_tokens": 20202805.0,
"step": 9730
},
{
"entropy": 5.539679002761841,
"epoch": 0.8830732946298984,
"grad_norm": 0.87890625,
"learning_rate": 0.0004929382906175253,
"loss": 5.1717,
"mean_token_accuracy": 0.19339725226163865,
"num_tokens": 20212702.0,
"step": 9735
},
{
"entropy": 5.504897594451904,
"epoch": 0.8835268505079826,
"grad_norm": 0.94921875,
"learning_rate": 0.0004929302456157657,
"loss": 5.2068,
"mean_token_accuracy": 0.19339656680822373,
"num_tokens": 20222661.0,
"step": 9740
},
{
"entropy": 5.490061807632446,
"epoch": 0.8839804063860668,
"grad_norm": 0.94140625,
"learning_rate": 0.0004929221961071531,
"loss": 5.2559,
"mean_token_accuracy": 0.18547530472278595,
"num_tokens": 20233012.0,
"step": 9745
},
{
"entropy": 5.521569061279297,
"epoch": 0.8844339622641509,
"grad_norm": 1.0390625,
"learning_rate": 0.0004929141420918538,
"loss": 5.2627,
"mean_token_accuracy": 0.19188135862350464,
"num_tokens": 20242918.0,
"step": 9750
},
{
"entropy": 5.4918468475341795,
"epoch": 0.8848875181422351,
"grad_norm": 0.9609375,
"learning_rate": 0.0004929060835700347,
"loss": 5.191,
"mean_token_accuracy": 0.19425636231899263,
"num_tokens": 20253254.0,
"step": 9755
},
{
"entropy": 5.500729370117187,
"epoch": 0.8853410740203193,
"grad_norm": 0.95703125,
"learning_rate": 0.000492898020541862,
"loss": 5.2469,
"mean_token_accuracy": 0.18291963785886764,
"num_tokens": 20263355.0,
"step": 9760
},
{
"entropy": 5.450434637069702,
"epoch": 0.8857946298984035,
"grad_norm": 0.9453125,
"learning_rate": 0.0004928899530075029,
"loss": 5.1715,
"mean_token_accuracy": 0.1866050571203232,
"num_tokens": 20274716.0,
"step": 9765
},
{
"entropy": 5.4292707443237305,
"epoch": 0.8862481857764877,
"grad_norm": 1.0234375,
"learning_rate": 0.0004928818809671241,
"loss": 5.1607,
"mean_token_accuracy": 0.18965840190649033,
"num_tokens": 20285631.0,
"step": 9770
},
{
"entropy": 5.558332729339599,
"epoch": 0.8867017416545718,
"grad_norm": 1.0234375,
"learning_rate": 0.0004928738044208923,
"loss": 5.2411,
"mean_token_accuracy": 0.1915654197335243,
"num_tokens": 20295717.0,
"step": 9775
},
{
"entropy": 5.459675216674805,
"epoch": 0.887155297532656,
"grad_norm": 0.9375,
"learning_rate": 0.0004928657233689748,
"loss": 5.2183,
"mean_token_accuracy": 0.18936942964792253,
"num_tokens": 20306637.0,
"step": 9780
},
{
"entropy": 5.490768623352051,
"epoch": 0.8876088534107403,
"grad_norm": 0.921875,
"learning_rate": 0.0004928576378115387,
"loss": 5.1846,
"mean_token_accuracy": 0.1875806614756584,
"num_tokens": 20316629.0,
"step": 9785
},
{
"entropy": 5.597564649581909,
"epoch": 0.8880624092888244,
"grad_norm": 0.94921875,
"learning_rate": 0.0004928495477487509,
"loss": 5.3573,
"mean_token_accuracy": 0.17805679738521576,
"num_tokens": 20327935.0,
"step": 9790
},
{
"entropy": 5.491835212707519,
"epoch": 0.8885159651669086,
"grad_norm": 0.87890625,
"learning_rate": 0.0004928414531807792,
"loss": 5.1762,
"mean_token_accuracy": 0.1839013561606407,
"num_tokens": 20339142.0,
"step": 9795
},
{
"entropy": 5.464007425308227,
"epoch": 0.8889695210449927,
"grad_norm": 0.9296875,
"learning_rate": 0.0004928333541077906,
"loss": 5.1896,
"mean_token_accuracy": 0.20151338577270508,
"num_tokens": 20348474.0,
"step": 9800
},
{
"entropy": 5.500573539733887,
"epoch": 0.8894230769230769,
"grad_norm": 0.984375,
"learning_rate": 0.0004928252505299527,
"loss": 5.2498,
"mean_token_accuracy": 0.1855317085981369,
"num_tokens": 20359440.0,
"step": 9805
},
{
"entropy": 5.579951906204224,
"epoch": 0.8898766328011611,
"grad_norm": 0.97265625,
"learning_rate": 0.0004928171424474332,
"loss": 5.2963,
"mean_token_accuracy": 0.18381822854280472,
"num_tokens": 20370355.0,
"step": 9810
},
{
"entropy": 5.492842674255371,
"epoch": 0.8903301886792453,
"grad_norm": 0.95703125,
"learning_rate": 0.0004928090298603996,
"loss": 5.1795,
"mean_token_accuracy": 0.18734454959630967,
"num_tokens": 20380214.0,
"step": 9815
},
{
"entropy": 5.527317905426026,
"epoch": 0.8907837445573294,
"grad_norm": 1.0859375,
"learning_rate": 0.0004928009127690199,
"loss": 5.1963,
"mean_token_accuracy": 0.18887149095535277,
"num_tokens": 20389190.0,
"step": 9820
},
{
"entropy": 5.4723756313323975,
"epoch": 0.8912373004354136,
"grad_norm": 0.98046875,
"learning_rate": 0.0004927927911734617,
"loss": 5.238,
"mean_token_accuracy": 0.18437635898590088,
"num_tokens": 20399531.0,
"step": 9825
},
{
"entropy": 5.463613748550415,
"epoch": 0.8916908563134979,
"grad_norm": 0.84375,
"learning_rate": 0.0004927846650738931,
"loss": 5.2246,
"mean_token_accuracy": 0.18858011811971664,
"num_tokens": 20410201.0,
"step": 9830
},
{
"entropy": 5.494712591171265,
"epoch": 0.892144412191582,
"grad_norm": 0.88671875,
"learning_rate": 0.0004927765344704821,
"loss": 5.1865,
"mean_token_accuracy": 0.19300392419099807,
"num_tokens": 20421579.0,
"step": 9835
},
{
"entropy": 5.510917329788208,
"epoch": 0.8925979680696662,
"grad_norm": 0.89453125,
"learning_rate": 0.0004927683993633969,
"loss": 5.2522,
"mean_token_accuracy": 0.18945046067237853,
"num_tokens": 20432048.0,
"step": 9840
},
{
"entropy": 5.559658622741699,
"epoch": 0.8930515239477503,
"grad_norm": 1.1484375,
"learning_rate": 0.0004927602597528057,
"loss": 5.2763,
"mean_token_accuracy": 0.18216387331485748,
"num_tokens": 20441752.0,
"step": 9845
},
{
"entropy": 5.437015390396118,
"epoch": 0.8935050798258345,
"grad_norm": 0.91796875,
"learning_rate": 0.0004927521156388769,
"loss": 5.1443,
"mean_token_accuracy": 0.1886148139834404,
"num_tokens": 20451693.0,
"step": 9850
},
{
"entropy": 5.512848711013794,
"epoch": 0.8939586357039188,
"grad_norm": 1.0078125,
"learning_rate": 0.0004927439670217788,
"loss": 5.2784,
"mean_token_accuracy": 0.18574748635292054,
"num_tokens": 20462312.0,
"step": 9855
},
{
"entropy": 5.51384711265564,
"epoch": 0.8944121915820029,
"grad_norm": 0.90234375,
"learning_rate": 0.00049273581390168,
"loss": 5.1842,
"mean_token_accuracy": 0.18325104117393493,
"num_tokens": 20472546.0,
"step": 9860
},
{
"entropy": 5.482762384414673,
"epoch": 0.8948657474600871,
"grad_norm": 0.875,
"learning_rate": 0.000492727656278749,
"loss": 5.2265,
"mean_token_accuracy": 0.189279143512249,
"num_tokens": 20483266.0,
"step": 9865
},
{
"entropy": 5.43660979270935,
"epoch": 0.8953193033381712,
"grad_norm": 1.0390625,
"learning_rate": 0.0004927194941531547,
"loss": 5.1597,
"mean_token_accuracy": 0.18750689327716827,
"num_tokens": 20492490.0,
"step": 9870
},
{
"entropy": 5.482614469528198,
"epoch": 0.8957728592162555,
"grad_norm": 0.9765625,
"learning_rate": 0.0004927113275250657,
"loss": 5.2002,
"mean_token_accuracy": 0.19175224751234055,
"num_tokens": 20503077.0,
"step": 9875
},
{
"entropy": 5.448323917388916,
"epoch": 0.8962264150943396,
"grad_norm": 0.921875,
"learning_rate": 0.0004927031563946511,
"loss": 5.2231,
"mean_token_accuracy": 0.18685709685087204,
"num_tokens": 20513771.0,
"step": 9880
},
{
"entropy": 5.513888883590698,
"epoch": 0.8966799709724238,
"grad_norm": 0.94140625,
"learning_rate": 0.0004926949807620796,
"loss": 5.3107,
"mean_token_accuracy": 0.1844270795583725,
"num_tokens": 20523616.0,
"step": 9885
},
{
"entropy": 5.525784826278686,
"epoch": 0.897133526850508,
"grad_norm": 0.98046875,
"learning_rate": 0.0004926868006275204,
"loss": 5.0908,
"mean_token_accuracy": 0.19574417024850846,
"num_tokens": 20533613.0,
"step": 9890
},
{
"entropy": 5.473994970321655,
"epoch": 0.8975870827285921,
"grad_norm": 1.0078125,
"learning_rate": 0.0004926786159911427,
"loss": 5.2053,
"mean_token_accuracy": 0.18760427534580232,
"num_tokens": 20543812.0,
"step": 9895
},
{
"entropy": 5.449514675140381,
"epoch": 0.8980406386066764,
"grad_norm": 1.046875,
"learning_rate": 0.000492670426853116,
"loss": 5.1502,
"mean_token_accuracy": 0.19117735475301742,
"num_tokens": 20553785.0,
"step": 9900
},
{
"entropy": 5.578982305526734,
"epoch": 0.8984941944847605,
"grad_norm": 0.9453125,
"learning_rate": 0.0004926622332136092,
"loss": 5.2671,
"mean_token_accuracy": 0.1900632604956627,
"num_tokens": 20563326.0,
"step": 9905
},
{
"entropy": 5.484088993072509,
"epoch": 0.8989477503628447,
"grad_norm": 0.90625,
"learning_rate": 0.0004926540350727918,
"loss": 5.1148,
"mean_token_accuracy": 0.19918729811906816,
"num_tokens": 20573643.0,
"step": 9910
},
{
"entropy": 5.387427949905396,
"epoch": 0.8994013062409288,
"grad_norm": 0.9609375,
"learning_rate": 0.0004926458324308337,
"loss": 5.1496,
"mean_token_accuracy": 0.1918700322508812,
"num_tokens": 20583916.0,
"step": 9915
},
{
"entropy": 5.506546497344971,
"epoch": 0.8998548621190131,
"grad_norm": 0.95703125,
"learning_rate": 0.0004926376252879043,
"loss": 5.3024,
"mean_token_accuracy": 0.18872980624437333,
"num_tokens": 20593984.0,
"step": 9920
},
{
"entropy": 5.531512498855591,
"epoch": 0.9003084179970973,
"grad_norm": 0.90234375,
"learning_rate": 0.0004926294136441733,
"loss": 5.141,
"mean_token_accuracy": 0.19161704182624817,
"num_tokens": 20603927.0,
"step": 9925
},
{
"entropy": 5.491614151000976,
"epoch": 0.9007619738751814,
"grad_norm": 0.96484375,
"learning_rate": 0.0004926211974998106,
"loss": 5.2693,
"mean_token_accuracy": 0.18857511729002,
"num_tokens": 20614147.0,
"step": 9930
},
{
"entropy": 5.524622821807862,
"epoch": 0.9012155297532656,
"grad_norm": 0.9375,
"learning_rate": 0.000492612976854986,
"loss": 5.2403,
"mean_token_accuracy": 0.18621373921632767,
"num_tokens": 20623631.0,
"step": 9935
},
{
"entropy": 5.438816690444947,
"epoch": 0.9016690856313497,
"grad_norm": 1.0,
"learning_rate": 0.0004926047517098697,
"loss": 5.1165,
"mean_token_accuracy": 0.19730497598648072,
"num_tokens": 20632566.0,
"step": 9940
},
{
"entropy": 5.500843334197998,
"epoch": 0.902122641509434,
"grad_norm": 0.953125,
"learning_rate": 0.0004925965220646315,
"loss": 5.233,
"mean_token_accuracy": 0.18363679200410843,
"num_tokens": 20643246.0,
"step": 9945
},
{
"entropy": 5.507595014572144,
"epoch": 0.9025761973875182,
"grad_norm": 1.0,
"learning_rate": 0.0004925882879194419,
"loss": 5.296,
"mean_token_accuracy": 0.18696873188018798,
"num_tokens": 20653372.0,
"step": 9950
},
{
"entropy": 5.515101146697998,
"epoch": 0.9030297532656023,
"grad_norm": 0.9375,
"learning_rate": 0.0004925800492744712,
"loss": 5.2455,
"mean_token_accuracy": 0.18492357879877092,
"num_tokens": 20662687.0,
"step": 9955
},
{
"entropy": 5.465108442306518,
"epoch": 0.9034833091436865,
"grad_norm": 0.90625,
"learning_rate": 0.0004925718061298895,
"loss": 5.1691,
"mean_token_accuracy": 0.19537947028875352,
"num_tokens": 20673168.0,
"step": 9960
},
{
"entropy": 5.555444622039795,
"epoch": 0.9039368650217707,
"grad_norm": 0.9921875,
"learning_rate": 0.0004925635584858675,
"loss": 5.2785,
"mean_token_accuracy": 0.18311350643634797,
"num_tokens": 20683100.0,
"step": 9965
},
{
"entropy": 5.589123487472534,
"epoch": 0.9043904208998549,
"grad_norm": 0.96484375,
"learning_rate": 0.0004925553063425755,
"loss": 5.2853,
"mean_token_accuracy": 0.18646423369646073,
"num_tokens": 20694097.0,
"step": 9970
},
{
"entropy": 5.501810359954834,
"epoch": 0.904843976777939,
"grad_norm": 1.0078125,
"learning_rate": 0.0004925470497001844,
"loss": 5.2467,
"mean_token_accuracy": 0.18978470414876938,
"num_tokens": 20704640.0,
"step": 9975
},
{
"entropy": 5.486598873138428,
"epoch": 0.9052975326560232,
"grad_norm": 0.92578125,
"learning_rate": 0.0004925387885588648,
"loss": 5.2002,
"mean_token_accuracy": 0.19009471386671067,
"num_tokens": 20714419.0,
"step": 9980
},
{
"entropy": 5.574535608291626,
"epoch": 0.9057510885341074,
"grad_norm": 1.015625,
"learning_rate": 0.0004925305229187877,
"loss": 5.2849,
"mean_token_accuracy": 0.1879521057009697,
"num_tokens": 20724705.0,
"step": 9985
},
{
"entropy": 5.561985349655151,
"epoch": 0.9062046444121916,
"grad_norm": 1.0625,
"learning_rate": 0.000492522252780124,
"loss": 5.3439,
"mean_token_accuracy": 0.18506076335906982,
"num_tokens": 20734753.0,
"step": 9990
},
{
"entropy": 5.5091822147369385,
"epoch": 0.9066582002902758,
"grad_norm": 0.85546875,
"learning_rate": 0.0004925139781430446,
"loss": 5.1525,
"mean_token_accuracy": 0.1960643097758293,
"num_tokens": 20745389.0,
"step": 9995
},
{
"entropy": 5.480530834197998,
"epoch": 0.9071117561683599,
"grad_norm": 1.0859375,
"learning_rate": 0.0004925056990077207,
"loss": 5.2167,
"mean_token_accuracy": 0.19059233814477922,
"num_tokens": 20755702.0,
"step": 10000
},
{
"entropy": 5.462615489959717,
"epoch": 0.9075653120464441,
"grad_norm": 0.9375,
"learning_rate": 0.0004924974153743235,
"loss": 5.1057,
"mean_token_accuracy": 0.1962900161743164,
"num_tokens": 20766046.0,
"step": 10005
},
{
"entropy": 5.492237567901611,
"epoch": 0.9080188679245284,
"grad_norm": 0.859375,
"learning_rate": 0.0004924891272430244,
"loss": 5.2233,
"mean_token_accuracy": 0.19067611545324326,
"num_tokens": 20776654.0,
"step": 10010
},
{
"entropy": 5.538366174697876,
"epoch": 0.9084724238026125,
"grad_norm": 0.96484375,
"learning_rate": 0.0004924808346139948,
"loss": 5.2699,
"mean_token_accuracy": 0.19199298322200775,
"num_tokens": 20787066.0,
"step": 10015
},
{
"entropy": 5.483625793457032,
"epoch": 0.9089259796806967,
"grad_norm": 0.9921875,
"learning_rate": 0.0004924725374874059,
"loss": 5.1807,
"mean_token_accuracy": 0.18984902501106263,
"num_tokens": 20796822.0,
"step": 10020
},
{
"entropy": 5.530815505981446,
"epoch": 0.9093795355587808,
"grad_norm": 0.91015625,
"learning_rate": 0.0004924642358634297,
"loss": 5.2346,
"mean_token_accuracy": 0.18641554266214372,
"num_tokens": 20807602.0,
"step": 10025
},
{
"entropy": 5.51546082496643,
"epoch": 0.909833091436865,
"grad_norm": 1.0234375,
"learning_rate": 0.0004924559297422376,
"loss": 5.193,
"mean_token_accuracy": 0.19486915022134782,
"num_tokens": 20819046.0,
"step": 10030
},
{
"entropy": 5.520950222015381,
"epoch": 0.9102866473149492,
"grad_norm": 0.921875,
"learning_rate": 0.0004924476191240014,
"loss": 5.2297,
"mean_token_accuracy": 0.18605352193117142,
"num_tokens": 20828989.0,
"step": 10035
},
{
"entropy": 5.47696099281311,
"epoch": 0.9107402031930334,
"grad_norm": 0.93359375,
"learning_rate": 0.000492439304008893,
"loss": 5.2303,
"mean_token_accuracy": 0.18616906255483628,
"num_tokens": 20839366.0,
"step": 10040
},
{
"entropy": 5.507669305801391,
"epoch": 0.9111937590711175,
"grad_norm": 0.9921875,
"learning_rate": 0.0004924309843970847,
"loss": 5.2253,
"mean_token_accuracy": 0.18729967921972274,
"num_tokens": 20849513.0,
"step": 10045
},
{
"entropy": 5.5649864196777346,
"epoch": 0.9116473149492017,
"grad_norm": 0.86328125,
"learning_rate": 0.0004924226602887479,
"loss": 5.2685,
"mean_token_accuracy": 0.18344479352235793,
"num_tokens": 20861406.0,
"step": 10050
},
{
"entropy": 5.495461368560791,
"epoch": 0.912100870827286,
"grad_norm": 0.89453125,
"learning_rate": 0.0004924143316840552,
"loss": 5.1824,
"mean_token_accuracy": 0.19835592210292816,
"num_tokens": 20871190.0,
"step": 10055
},
{
"entropy": 5.505907154083252,
"epoch": 0.9125544267053701,
"grad_norm": 0.92578125,
"learning_rate": 0.0004924059985831787,
"loss": 5.3313,
"mean_token_accuracy": 0.18035977929830552,
"num_tokens": 20881589.0,
"step": 10060
},
{
"entropy": 5.508808231353759,
"epoch": 0.9130079825834543,
"grad_norm": 1.0234375,
"learning_rate": 0.0004923976609862908,
"loss": 5.26,
"mean_token_accuracy": 0.1787053018808365,
"num_tokens": 20892072.0,
"step": 10065
},
{
"entropy": 5.547299718856811,
"epoch": 0.9134615384615384,
"grad_norm": 0.953125,
"learning_rate": 0.0004923893188935638,
"loss": 5.2435,
"mean_token_accuracy": 0.18379553705453872,
"num_tokens": 20902616.0,
"step": 10070
},
{
"entropy": 5.545641326904297,
"epoch": 0.9139150943396226,
"grad_norm": 0.94140625,
"learning_rate": 0.0004923809723051702,
"loss": 5.2337,
"mean_token_accuracy": 0.18797029107809066,
"num_tokens": 20913832.0,
"step": 10075
},
{
"entropy": 5.4296571731567385,
"epoch": 0.9143686502177069,
"grad_norm": 1.0390625,
"learning_rate": 0.0004923726212212829,
"loss": 5.1311,
"mean_token_accuracy": 0.1946079269051552,
"num_tokens": 20923322.0,
"step": 10080
},
{
"entropy": 5.484694910049439,
"epoch": 0.914822206095791,
"grad_norm": 1.015625,
"learning_rate": 0.0004923642656420742,
"loss": 5.2231,
"mean_token_accuracy": 0.18851592838764192,
"num_tokens": 20933334.0,
"step": 10085
},
{
"entropy": 5.500257778167724,
"epoch": 0.9152757619738752,
"grad_norm": 0.93359375,
"learning_rate": 0.0004923559055677173,
"loss": 5.1922,
"mean_token_accuracy": 0.1872135117650032,
"num_tokens": 20944171.0,
"step": 10090
},
{
"entropy": 5.50403790473938,
"epoch": 0.9157293178519593,
"grad_norm": 1.0078125,
"learning_rate": 0.0004923475409983848,
"loss": 5.2372,
"mean_token_accuracy": 0.1875006526708603,
"num_tokens": 20954200.0,
"step": 10095
},
{
"entropy": 5.483087778091431,
"epoch": 0.9161828737300436,
"grad_norm": 0.8984375,
"learning_rate": 0.0004923391719342497,
"loss": 5.2503,
"mean_token_accuracy": 0.18236380070447922,
"num_tokens": 20964935.0,
"step": 10100
},
{
"entropy": 5.502628421783447,
"epoch": 0.9166364296081277,
"grad_norm": 0.92578125,
"learning_rate": 0.0004923307983754852,
"loss": 5.2312,
"mean_token_accuracy": 0.19101871997117997,
"num_tokens": 20975499.0,
"step": 10105
},
{
"entropy": 5.433144903182983,
"epoch": 0.9170899854862119,
"grad_norm": 0.96875,
"learning_rate": 0.0004923224203222645,
"loss": 5.1678,
"mean_token_accuracy": 0.19008324295282364,
"num_tokens": 20986199.0,
"step": 10110
},
{
"entropy": 5.4952733516693115,
"epoch": 0.9175435413642961,
"grad_norm": 1.0,
"learning_rate": 0.0004923140377747606,
"loss": 5.2464,
"mean_token_accuracy": 0.18770486861467361,
"num_tokens": 20996313.0,
"step": 10115
},
{
"entropy": 5.533897113800049,
"epoch": 0.9179970972423802,
"grad_norm": 0.9375,
"learning_rate": 0.0004923056507331471,
"loss": 5.2228,
"mean_token_accuracy": 0.18642332404851913,
"num_tokens": 21006133.0,
"step": 10120
},
{
"entropy": 5.531605768203735,
"epoch": 0.9184506531204645,
"grad_norm": 0.9296875,
"learning_rate": 0.0004922972591975974,
"loss": 5.192,
"mean_token_accuracy": 0.18831069022417068,
"num_tokens": 21016448.0,
"step": 10125
},
{
"entropy": 5.451457118988037,
"epoch": 0.9189042089985486,
"grad_norm": 0.98828125,
"learning_rate": 0.0004922888631682849,
"loss": 5.1271,
"mean_token_accuracy": 0.18411844819784165,
"num_tokens": 21026458.0,
"step": 10130
},
{
"entropy": 5.449035692214966,
"epoch": 0.9193577648766328,
"grad_norm": 1.0390625,
"learning_rate": 0.0004922804626453834,
"loss": 5.1663,
"mean_token_accuracy": 0.192974516749382,
"num_tokens": 21037111.0,
"step": 10135
},
{
"entropy": 5.457956552505493,
"epoch": 0.9198113207547169,
"grad_norm": 1.0390625,
"learning_rate": 0.0004922720576290665,
"loss": 5.2224,
"mean_token_accuracy": 0.19142512679100038,
"num_tokens": 21048041.0,
"step": 10140
},
{
"entropy": 5.433384132385254,
"epoch": 0.9202648766328012,
"grad_norm": 1.015625,
"learning_rate": 0.0004922636481195082,
"loss": 5.0919,
"mean_token_accuracy": 0.20250018388032914,
"num_tokens": 21058099.0,
"step": 10145
},
{
"entropy": 5.501229667663575,
"epoch": 0.9207184325108854,
"grad_norm": 0.9765625,
"learning_rate": 0.0004922552341168823,
"loss": 5.2367,
"mean_token_accuracy": 0.18169017881155014,
"num_tokens": 21068137.0,
"step": 10150
},
{
"entropy": 5.41462550163269,
"epoch": 0.9211719883889695,
"grad_norm": 0.9453125,
"learning_rate": 0.0004922468156213627,
"loss": 5.1188,
"mean_token_accuracy": 0.19230714738368987,
"num_tokens": 21077556.0,
"step": 10155
},
{
"entropy": 5.499750471115112,
"epoch": 0.9216255442670537,
"grad_norm": 1.0234375,
"learning_rate": 0.0004922383926331236,
"loss": 5.1527,
"mean_token_accuracy": 0.19040969759225845,
"num_tokens": 21087524.0,
"step": 10160
},
{
"entropy": 5.438167142868042,
"epoch": 0.9220791001451378,
"grad_norm": 0.92578125,
"learning_rate": 0.0004922299651523392,
"loss": 5.2295,
"mean_token_accuracy": 0.1860123857855797,
"num_tokens": 21098446.0,
"step": 10165
},
{
"entropy": 5.48976616859436,
"epoch": 0.9225326560232221,
"grad_norm": 0.953125,
"learning_rate": 0.0004922215331791837,
"loss": 5.2352,
"mean_token_accuracy": 0.18725680112838744,
"num_tokens": 21109291.0,
"step": 10170
},
{
"entropy": 5.514573240280152,
"epoch": 0.9229862119013063,
"grad_norm": 1.0078125,
"learning_rate": 0.0004922130967138316,
"loss": 5.1208,
"mean_token_accuracy": 0.1932806611061096,
"num_tokens": 21120043.0,
"step": 10175
},
{
"entropy": 5.518204164505005,
"epoch": 0.9234397677793904,
"grad_norm": 0.97265625,
"learning_rate": 0.0004922046557564574,
"loss": 5.2357,
"mean_token_accuracy": 0.18277695327997207,
"num_tokens": 21130339.0,
"step": 10180
},
{
"entropy": 5.500779438018799,
"epoch": 0.9238933236574746,
"grad_norm": 0.96484375,
"learning_rate": 0.0004921962103072355,
"loss": 5.3023,
"mean_token_accuracy": 0.18717591762542723,
"num_tokens": 21142307.0,
"step": 10185
},
{
"entropy": 5.526569557189942,
"epoch": 0.9243468795355588,
"grad_norm": 0.921875,
"learning_rate": 0.0004921877603663406,
"loss": 5.2036,
"mean_token_accuracy": 0.19003356993198395,
"num_tokens": 21151843.0,
"step": 10190
},
{
"entropy": 5.502495908737183,
"epoch": 0.924800435413643,
"grad_norm": 0.9765625,
"learning_rate": 0.0004921793059339474,
"loss": 5.2119,
"mean_token_accuracy": 0.19041070491075515,
"num_tokens": 21162574.0,
"step": 10195
},
{
"entropy": 5.510891437530518,
"epoch": 0.9252539912917271,
"grad_norm": 0.92578125,
"learning_rate": 0.0004921708470102309,
"loss": 5.16,
"mean_token_accuracy": 0.19265214800834657,
"num_tokens": 21172637.0,
"step": 10200
},
{
"entropy": 5.473992538452149,
"epoch": 0.9257075471698113,
"grad_norm": 0.953125,
"learning_rate": 0.000492162383595366,
"loss": 5.168,
"mean_token_accuracy": 0.19391350150108339,
"num_tokens": 21182272.0,
"step": 10205
},
{
"entropy": 5.540949773788452,
"epoch": 0.9261611030478955,
"grad_norm": 0.89453125,
"learning_rate": 0.0004921539156895275,
"loss": 5.3006,
"mean_token_accuracy": 0.18155533522367479,
"num_tokens": 21193712.0,
"step": 10210
},
{
"entropy": 5.566745138168335,
"epoch": 0.9266146589259797,
"grad_norm": 1.09375,
"learning_rate": 0.0004921454432928908,
"loss": 5.303,
"mean_token_accuracy": 0.18065586537122727,
"num_tokens": 21205104.0,
"step": 10215
},
{
"entropy": 5.536470413208008,
"epoch": 0.9270682148040639,
"grad_norm": 0.94921875,
"learning_rate": 0.000492136966405631,
"loss": 5.1908,
"mean_token_accuracy": 0.18957612067461013,
"num_tokens": 21215495.0,
"step": 10220
},
{
"entropy": 5.49644718170166,
"epoch": 0.927521770682148,
"grad_norm": 1.046875,
"learning_rate": 0.0004921284850279233,
"loss": 5.2023,
"mean_token_accuracy": 0.1892085149884224,
"num_tokens": 21226047.0,
"step": 10225
},
{
"entropy": 5.50606107711792,
"epoch": 0.9279753265602322,
"grad_norm": 1.0078125,
"learning_rate": 0.0004921199991599433,
"loss": 5.2756,
"mean_token_accuracy": 0.18349545150995256,
"num_tokens": 21235784.0,
"step": 10230
},
{
"entropy": 5.548442935943603,
"epoch": 0.9284288824383164,
"grad_norm": 0.98828125,
"learning_rate": 0.0004921115088018665,
"loss": 5.2245,
"mean_token_accuracy": 0.18641836047172547,
"num_tokens": 21245493.0,
"step": 10235
},
{
"entropy": 5.592808103561401,
"epoch": 0.9288824383164006,
"grad_norm": 0.93359375,
"learning_rate": 0.000492103013953868,
"loss": 5.3075,
"mean_token_accuracy": 0.18445847779512406,
"num_tokens": 21255541.0,
"step": 10240
},
{
"entropy": 5.46482138633728,
"epoch": 0.9293359941944848,
"grad_norm": 0.953125,
"learning_rate": 0.0004920945146161242,
"loss": 5.16,
"mean_token_accuracy": 0.19516148865222932,
"num_tokens": 21265427.0,
"step": 10245
},
{
"entropy": 5.470294094085693,
"epoch": 0.9297895500725689,
"grad_norm": 0.9140625,
"learning_rate": 0.0004920860107888105,
"loss": 5.164,
"mean_token_accuracy": 0.19581332057714462,
"num_tokens": 21276597.0,
"step": 10250
},
{
"entropy": 5.450829267501831,
"epoch": 0.9302431059506531,
"grad_norm": 0.97265625,
"learning_rate": 0.0004920775024721026,
"loss": 5.1916,
"mean_token_accuracy": 0.18773079216480254,
"num_tokens": 21286416.0,
"step": 10255
},
{
"entropy": 5.497037076950074,
"epoch": 0.9306966618287373,
"grad_norm": 0.94921875,
"learning_rate": 0.0004920689896661767,
"loss": 5.2054,
"mean_token_accuracy": 0.1940332069993019,
"num_tokens": 21296064.0,
"step": 10260
},
{
"entropy": 5.476668024063111,
"epoch": 0.9311502177068215,
"grad_norm": 1.0078125,
"learning_rate": 0.0004920604723712088,
"loss": 5.214,
"mean_token_accuracy": 0.19466549903154373,
"num_tokens": 21306836.0,
"step": 10265
},
{
"entropy": 5.485483455657959,
"epoch": 0.9316037735849056,
"grad_norm": 0.953125,
"learning_rate": 0.0004920519505873751,
"loss": 5.1735,
"mean_token_accuracy": 0.19273321628570556,
"num_tokens": 21316936.0,
"step": 10270
},
{
"entropy": 5.449879693984985,
"epoch": 0.9320573294629898,
"grad_norm": 0.93359375,
"learning_rate": 0.0004920434243148515,
"loss": 5.1367,
"mean_token_accuracy": 0.19561291486024857,
"num_tokens": 21326892.0,
"step": 10275
},
{
"entropy": 5.469815301895141,
"epoch": 0.9325108853410741,
"grad_norm": 0.98046875,
"learning_rate": 0.0004920348935538147,
"loss": 5.1665,
"mean_token_accuracy": 0.18587524741888045,
"num_tokens": 21337270.0,
"step": 10280
},
{
"entropy": 5.525975322723388,
"epoch": 0.9329644412191582,
"grad_norm": 0.9765625,
"learning_rate": 0.000492026358304441,
"loss": 5.2366,
"mean_token_accuracy": 0.1867948666214943,
"num_tokens": 21347968.0,
"step": 10285
},
{
"entropy": 5.483292770385742,
"epoch": 0.9334179970972424,
"grad_norm": 0.92578125,
"learning_rate": 0.000492017818566907,
"loss": 5.1862,
"mean_token_accuracy": 0.1958588868379593,
"num_tokens": 21358312.0,
"step": 10290
},
{
"entropy": 5.477849340438842,
"epoch": 0.9338715529753265,
"grad_norm": 0.96484375,
"learning_rate": 0.0004920092743413892,
"loss": 5.1513,
"mean_token_accuracy": 0.19156461209058762,
"num_tokens": 21368794.0,
"step": 10295
},
{
"entropy": 5.515837144851685,
"epoch": 0.9343251088534107,
"grad_norm": 0.86328125,
"learning_rate": 0.0004920007256280642,
"loss": 5.2094,
"mean_token_accuracy": 0.19680030047893524,
"num_tokens": 21380080.0,
"step": 10300
},
{
"entropy": 5.507870388031006,
"epoch": 0.934778664731495,
"grad_norm": 0.90234375,
"learning_rate": 0.000491992172427109,
"loss": 5.1906,
"mean_token_accuracy": 0.19244442880153656,
"num_tokens": 21389961.0,
"step": 10305
},
{
"entropy": 5.452802419662476,
"epoch": 0.9352322206095791,
"grad_norm": 0.9453125,
"learning_rate": 0.0004919836147387002,
"loss": 5.2319,
"mean_token_accuracy": 0.18641486167907714,
"num_tokens": 21399869.0,
"step": 10310
},
{
"entropy": 5.499537134170533,
"epoch": 0.9356857764876633,
"grad_norm": 0.9140625,
"learning_rate": 0.0004919750525630151,
"loss": 5.2171,
"mean_token_accuracy": 0.19221128672361373,
"num_tokens": 21410343.0,
"step": 10315
},
{
"entropy": 5.527597951889038,
"epoch": 0.9361393323657474,
"grad_norm": 0.9140625,
"learning_rate": 0.0004919664859002306,
"loss": 5.1953,
"mean_token_accuracy": 0.18969294428825378,
"num_tokens": 21420915.0,
"step": 10320
},
{
"entropy": 5.487546443939209,
"epoch": 0.9365928882438317,
"grad_norm": 0.984375,
"learning_rate": 0.0004919579147505239,
"loss": 5.1638,
"mean_token_accuracy": 0.19285682290792466,
"num_tokens": 21430879.0,
"step": 10325
},
{
"entropy": 5.531203317642212,
"epoch": 0.9370464441219158,
"grad_norm": 0.9453125,
"learning_rate": 0.0004919493391140721,
"loss": 5.2574,
"mean_token_accuracy": 0.18837825804948807,
"num_tokens": 21440843.0,
"step": 10330
},
{
"entropy": 5.592268323898315,
"epoch": 0.9375,
"grad_norm": 0.92578125,
"learning_rate": 0.0004919407589910529,
"loss": 5.4086,
"mean_token_accuracy": 0.1807285413146019,
"num_tokens": 21452132.0,
"step": 10335
},
{
"entropy": 5.493322324752808,
"epoch": 0.9379535558780842,
"grad_norm": 0.98046875,
"learning_rate": 0.0004919321743816434,
"loss": 5.194,
"mean_token_accuracy": 0.18563437461853027,
"num_tokens": 21461861.0,
"step": 10340
},
{
"entropy": 5.499019622802734,
"epoch": 0.9384071117561683,
"grad_norm": 1.0625,
"learning_rate": 0.0004919235852860213,
"loss": 5.1976,
"mean_token_accuracy": 0.18736050128936768,
"num_tokens": 21472244.0,
"step": 10345
},
{
"entropy": 5.545416593551636,
"epoch": 0.9388606676342526,
"grad_norm": 1.015625,
"learning_rate": 0.0004919149917043641,
"loss": 5.2515,
"mean_token_accuracy": 0.1836113914847374,
"num_tokens": 21482394.0,
"step": 10350
},
{
"entropy": 5.498862457275391,
"epoch": 0.9393142235123367,
"grad_norm": 0.88671875,
"learning_rate": 0.0004919063936368496,
"loss": 5.276,
"mean_token_accuracy": 0.18350204527378083,
"num_tokens": 21493334.0,
"step": 10355
},
{
"entropy": 5.553494691848755,
"epoch": 0.9397677793904209,
"grad_norm": 0.93359375,
"learning_rate": 0.0004918977910836557,
"loss": 5.2386,
"mean_token_accuracy": 0.1844620883464813,
"num_tokens": 21504443.0,
"step": 10360
},
{
"entropy": 5.539630651473999,
"epoch": 0.940221335268505,
"grad_norm": 0.94140625,
"learning_rate": 0.0004918891840449601,
"loss": 5.2604,
"mean_token_accuracy": 0.18865265250205993,
"num_tokens": 21513908.0,
"step": 10365
},
{
"entropy": 5.566462993621826,
"epoch": 0.9406748911465893,
"grad_norm": 0.8515625,
"learning_rate": 0.000491880572520941,
"loss": 5.2809,
"mean_token_accuracy": 0.17776426523923874,
"num_tokens": 21524349.0,
"step": 10370
},
{
"entropy": 5.651615619659424,
"epoch": 0.9411284470246735,
"grad_norm": 0.9453125,
"learning_rate": 0.0004918719565117763,
"loss": 5.3383,
"mean_token_accuracy": 0.18980624675750732,
"num_tokens": 21535003.0,
"step": 10375
},
{
"entropy": 5.603833770751953,
"epoch": 0.9415820029027576,
"grad_norm": 1.0234375,
"learning_rate": 0.0004918633360176444,
"loss": 5.3558,
"mean_token_accuracy": 0.17872778922319413,
"num_tokens": 21545045.0,
"step": 10380
},
{
"entropy": 5.461150074005127,
"epoch": 0.9420355587808418,
"grad_norm": 0.9296875,
"learning_rate": 0.0004918547110387234,
"loss": 5.1816,
"mean_token_accuracy": 0.19119360893964768,
"num_tokens": 21554830.0,
"step": 10385
},
{
"entropy": 5.5173086643219,
"epoch": 0.9424891146589259,
"grad_norm": 1.0078125,
"learning_rate": 0.0004918460815751916,
"loss": 5.1546,
"mean_token_accuracy": 0.19252003729343414,
"num_tokens": 21564693.0,
"step": 10390
},
{
"entropy": 5.464719247817993,
"epoch": 0.9429426705370102,
"grad_norm": 0.9140625,
"learning_rate": 0.0004918374476272277,
"loss": 5.1676,
"mean_token_accuracy": 0.19065018594264985,
"num_tokens": 21575475.0,
"step": 10395
},
{
"entropy": 5.483359146118164,
"epoch": 0.9433962264150944,
"grad_norm": 0.890625,
"learning_rate": 0.0004918288091950101,
"loss": 5.1769,
"mean_token_accuracy": 0.188684843480587,
"num_tokens": 21586944.0,
"step": 10400
},
{
"entropy": 5.520866012573242,
"epoch": 0.9438497822931785,
"grad_norm": 0.94140625,
"learning_rate": 0.0004918201662787174,
"loss": 5.2096,
"mean_token_accuracy": 0.18385448306798935,
"num_tokens": 21598170.0,
"step": 10405
},
{
"entropy": 5.47841067314148,
"epoch": 0.9443033381712627,
"grad_norm": 0.8515625,
"learning_rate": 0.0004918115188785285,
"loss": 5.2258,
"mean_token_accuracy": 0.1873992457985878,
"num_tokens": 21609170.0,
"step": 10410
},
{
"entropy": 5.554736661911011,
"epoch": 0.9447568940493469,
"grad_norm": 0.8828125,
"learning_rate": 0.0004918028669946221,
"loss": 5.2796,
"mean_token_accuracy": 0.18662892132997513,
"num_tokens": 21620171.0,
"step": 10415
},
{
"entropy": 5.514125776290894,
"epoch": 0.9452104499274311,
"grad_norm": 1.1171875,
"learning_rate": 0.0004917942106271771,
"loss": 5.2234,
"mean_token_accuracy": 0.19277503490447997,
"num_tokens": 21631072.0,
"step": 10420
},
{
"entropy": 5.437767839431762,
"epoch": 0.9456640058055152,
"grad_norm": 0.94140625,
"learning_rate": 0.0004917855497763727,
"loss": 5.0861,
"mean_token_accuracy": 0.19857796430587768,
"num_tokens": 21641543.0,
"step": 10425
},
{
"entropy": 5.445878791809082,
"epoch": 0.9461175616835994,
"grad_norm": 1.171875,
"learning_rate": 0.0004917768844423878,
"loss": 5.2319,
"mean_token_accuracy": 0.19050681293010713,
"num_tokens": 21651342.0,
"step": 10430
},
{
"entropy": 5.514336013793946,
"epoch": 0.9465711175616836,
"grad_norm": 0.96875,
"learning_rate": 0.0004917682146254017,
"loss": 5.1868,
"mean_token_accuracy": 0.19425029307603836,
"num_tokens": 21661043.0,
"step": 10435
},
{
"entropy": 5.445535373687744,
"epoch": 0.9470246734397678,
"grad_norm": 0.89453125,
"learning_rate": 0.0004917595403255938,
"loss": 5.0752,
"mean_token_accuracy": 0.20240829586982728,
"num_tokens": 21671197.0,
"step": 10440
},
{
"entropy": 5.40161247253418,
"epoch": 0.947478229317852,
"grad_norm": 0.94140625,
"learning_rate": 0.0004917508615431431,
"loss": 5.1289,
"mean_token_accuracy": 0.19438163042068482,
"num_tokens": 21682066.0,
"step": 10445
},
{
"entropy": 5.521121835708618,
"epoch": 0.9479317851959361,
"grad_norm": 0.98828125,
"learning_rate": 0.0004917421782782296,
"loss": 5.193,
"mean_token_accuracy": 0.19061861038208008,
"num_tokens": 21692117.0,
"step": 10450
},
{
"entropy": 5.518373441696167,
"epoch": 0.9483853410740203,
"grad_norm": 0.97265625,
"learning_rate": 0.0004917334905310325,
"loss": 5.302,
"mean_token_accuracy": 0.183126300573349,
"num_tokens": 21703544.0,
"step": 10455
},
{
"entropy": 5.4479773998260494,
"epoch": 0.9488388969521045,
"grad_norm": 1.0859375,
"learning_rate": 0.0004917247983017317,
"loss": 5.1334,
"mean_token_accuracy": 0.19209074079990388,
"num_tokens": 21712850.0,
"step": 10460
},
{
"entropy": 5.4908843517303465,
"epoch": 0.9492924528301887,
"grad_norm": 0.94921875,
"learning_rate": 0.0004917161015905067,
"loss": 5.2731,
"mean_token_accuracy": 0.1897297501564026,
"num_tokens": 21723015.0,
"step": 10465
},
{
"entropy": 5.532942295074463,
"epoch": 0.9497460087082729,
"grad_norm": 1.015625,
"learning_rate": 0.0004917074003975376,
"loss": 5.2384,
"mean_token_accuracy": 0.18643336445093156,
"num_tokens": 21732453.0,
"step": 10470
},
{
"entropy": 5.5696944236755375,
"epoch": 0.950199564586357,
"grad_norm": 0.98828125,
"learning_rate": 0.0004916986947230042,
"loss": 5.3059,
"mean_token_accuracy": 0.18505144864320755,
"num_tokens": 21743251.0,
"step": 10475
},
{
"entropy": 5.516680526733398,
"epoch": 0.9506531204644412,
"grad_norm": 0.98828125,
"learning_rate": 0.0004916899845670866,
"loss": 5.2131,
"mean_token_accuracy": 0.18668761998414993,
"num_tokens": 21752891.0,
"step": 10480
},
{
"entropy": 5.535280227661133,
"epoch": 0.9511066763425254,
"grad_norm": 0.93359375,
"learning_rate": 0.0004916812699299649,
"loss": 5.2144,
"mean_token_accuracy": 0.19159982055425645,
"num_tokens": 21763230.0,
"step": 10485
},
{
"entropy": 5.482826519012451,
"epoch": 0.9515602322206096,
"grad_norm": 0.921875,
"learning_rate": 0.0004916725508118193,
"loss": 5.1622,
"mean_token_accuracy": 0.19009662866592408,
"num_tokens": 21773663.0,
"step": 10490
},
{
"entropy": 5.464633321762085,
"epoch": 0.9520137880986937,
"grad_norm": 0.91796875,
"learning_rate": 0.0004916638272128301,
"loss": 5.2258,
"mean_token_accuracy": 0.1891879215836525,
"num_tokens": 21784265.0,
"step": 10495
},
{
"entropy": 5.4609263896942135,
"epoch": 0.9524673439767779,
"grad_norm": 1.0,
"learning_rate": 0.0004916550991331777,
"loss": 5.1746,
"mean_token_accuracy": 0.19152846932411194,
"num_tokens": 21795056.0,
"step": 10500
},
{
"entropy": 5.498981094360351,
"epoch": 0.9529208998548622,
"grad_norm": 0.88671875,
"learning_rate": 0.0004916463665730427,
"loss": 5.1927,
"mean_token_accuracy": 0.19459545016288757,
"num_tokens": 21806204.0,
"step": 10505
},
{
"entropy": 5.49833664894104,
"epoch": 0.9533744557329463,
"grad_norm": 0.94140625,
"learning_rate": 0.0004916376295326057,
"loss": 5.2152,
"mean_token_accuracy": 0.1886645570397377,
"num_tokens": 21816993.0,
"step": 10510
},
{
"entropy": 5.439201021194458,
"epoch": 0.9538280116110305,
"grad_norm": 0.9375,
"learning_rate": 0.0004916288880120472,
"loss": 5.0649,
"mean_token_accuracy": 0.2021304950118065,
"num_tokens": 21827457.0,
"step": 10515
},
{
"entropy": 5.504736804962159,
"epoch": 0.9542815674891146,
"grad_norm": 0.95703125,
"learning_rate": 0.0004916201420115482,
"loss": 5.1988,
"mean_token_accuracy": 0.18791094422340393,
"num_tokens": 21837555.0,
"step": 10520
},
{
"entropy": 5.517563915252685,
"epoch": 0.9547351233671988,
"grad_norm": 0.9453125,
"learning_rate": 0.0004916113915312895,
"loss": 5.2174,
"mean_token_accuracy": 0.19141177833080292,
"num_tokens": 21847074.0,
"step": 10525
},
{
"entropy": 5.554729223251343,
"epoch": 0.9551886792452831,
"grad_norm": 0.9609375,
"learning_rate": 0.0004916026365714519,
"loss": 5.198,
"mean_token_accuracy": 0.18883113265037538,
"num_tokens": 21858333.0,
"step": 10530
},
{
"entropy": 5.469289255142212,
"epoch": 0.9556422351233672,
"grad_norm": 1.03125,
"learning_rate": 0.0004915938771322165,
"loss": 5.2141,
"mean_token_accuracy": 0.19106341451406478,
"num_tokens": 21869132.0,
"step": 10535
},
{
"entropy": 5.455795764923096,
"epoch": 0.9560957910014514,
"grad_norm": 1.0,
"learning_rate": 0.0004915851132137646,
"loss": 5.2446,
"mean_token_accuracy": 0.1872527241706848,
"num_tokens": 21879153.0,
"step": 10540
},
{
"entropy": 5.502502346038819,
"epoch": 0.9565493468795355,
"grad_norm": 0.87109375,
"learning_rate": 0.0004915763448162774,
"loss": 5.1259,
"mean_token_accuracy": 0.19310393184423447,
"num_tokens": 21889063.0,
"step": 10545
},
{
"entropy": 5.538162755966186,
"epoch": 0.9570029027576198,
"grad_norm": 1.0859375,
"learning_rate": 0.0004915675719399362,
"loss": 5.2476,
"mean_token_accuracy": 0.18670725971460342,
"num_tokens": 21899005.0,
"step": 10550
},
{
"entropy": 5.509344387054443,
"epoch": 0.9574564586357039,
"grad_norm": 1.03125,
"learning_rate": 0.0004915587945849225,
"loss": 5.1119,
"mean_token_accuracy": 0.19825160056352614,
"num_tokens": 21908391.0,
"step": 10555
},
{
"entropy": 5.460941791534424,
"epoch": 0.9579100145137881,
"grad_norm": 0.953125,
"learning_rate": 0.0004915500127514176,
"loss": 5.193,
"mean_token_accuracy": 0.19470654428005219,
"num_tokens": 21917655.0,
"step": 10560
},
{
"entropy": 5.555629873275757,
"epoch": 0.9583635703918723,
"grad_norm": 0.99609375,
"learning_rate": 0.0004915412264396034,
"loss": 5.2493,
"mean_token_accuracy": 0.19042410403490068,
"num_tokens": 21928038.0,
"step": 10565
},
{
"entropy": 5.521947193145752,
"epoch": 0.9588171262699564,
"grad_norm": 1.0,
"learning_rate": 0.0004915324356496613,
"loss": 5.14,
"mean_token_accuracy": 0.19917869120836257,
"num_tokens": 21938106.0,
"step": 10570
},
{
"entropy": 5.487814474105835,
"epoch": 0.9592706821480407,
"grad_norm": 0.96875,
"learning_rate": 0.0004915236403817733,
"loss": 5.1474,
"mean_token_accuracy": 0.19220392405986786,
"num_tokens": 21947484.0,
"step": 10575
},
{
"entropy": 5.473394870758057,
"epoch": 0.9597242380261248,
"grad_norm": 0.9765625,
"learning_rate": 0.0004915148406361213,
"loss": 5.2315,
"mean_token_accuracy": 0.18975483775138854,
"num_tokens": 21957139.0,
"step": 10580
},
{
"entropy": 5.487856483459472,
"epoch": 0.960177793904209,
"grad_norm": 1.046875,
"learning_rate": 0.0004915060364128872,
"loss": 5.2299,
"mean_token_accuracy": 0.185565547645092,
"num_tokens": 21967030.0,
"step": 10585
},
{
"entropy": 5.448148107528686,
"epoch": 0.9606313497822931,
"grad_norm": 1.0234375,
"learning_rate": 0.0004914972277122532,
"loss": 5.169,
"mean_token_accuracy": 0.1888902634382248,
"num_tokens": 21976418.0,
"step": 10590
},
{
"entropy": 5.49601378440857,
"epoch": 0.9610849056603774,
"grad_norm": 0.890625,
"learning_rate": 0.0004914884145344012,
"loss": 5.1779,
"mean_token_accuracy": 0.19263349920511247,
"num_tokens": 21987284.0,
"step": 10595
},
{
"entropy": 5.42640495300293,
"epoch": 0.9615384615384616,
"grad_norm": 0.90625,
"learning_rate": 0.0004914795968795137,
"loss": 5.103,
"mean_token_accuracy": 0.1972813591361046,
"num_tokens": 21997106.0,
"step": 10600
},
{
"entropy": 5.467606544494629,
"epoch": 0.9619920174165457,
"grad_norm": 0.9296875,
"learning_rate": 0.0004914707747477728,
"loss": 5.0823,
"mean_token_accuracy": 0.192961286008358,
"num_tokens": 22007589.0,
"step": 10605
},
{
"entropy": 5.48398494720459,
"epoch": 0.9624455732946299,
"grad_norm": 0.9765625,
"learning_rate": 0.0004914619481393613,
"loss": 5.2221,
"mean_token_accuracy": 0.189901639521122,
"num_tokens": 22017669.0,
"step": 10610
},
{
"entropy": 5.617108011245728,
"epoch": 0.962899129172714,
"grad_norm": 0.90625,
"learning_rate": 0.0004914531170544615,
"loss": 5.3711,
"mean_token_accuracy": 0.18461449742317199,
"num_tokens": 22029330.0,
"step": 10615
},
{
"entropy": 5.473918676376343,
"epoch": 0.9633526850507983,
"grad_norm": 0.9765625,
"learning_rate": 0.0004914442814932561,
"loss": 5.1331,
"mean_token_accuracy": 0.19325099438428878,
"num_tokens": 22039877.0,
"step": 10620
},
{
"entropy": 5.480970811843872,
"epoch": 0.9638062409288825,
"grad_norm": 1.0,
"learning_rate": 0.0004914354414559278,
"loss": 5.2025,
"mean_token_accuracy": 0.19430873394012452,
"num_tokens": 22049579.0,
"step": 10625
},
{
"entropy": 5.378883743286133,
"epoch": 0.9642597968069666,
"grad_norm": 0.99609375,
"learning_rate": 0.0004914265969426594,
"loss": 5.0784,
"mean_token_accuracy": 0.20659508407115937,
"num_tokens": 22059608.0,
"step": 10630
},
{
"entropy": 5.469490385055542,
"epoch": 0.9647133526850508,
"grad_norm": 0.96484375,
"learning_rate": 0.0004914177479536338,
"loss": 5.2372,
"mean_token_accuracy": 0.186066897213459,
"num_tokens": 22070036.0,
"step": 10635
},
{
"entropy": 5.514468145370484,
"epoch": 0.965166908563135,
"grad_norm": 0.9140625,
"learning_rate": 0.000491408894489034,
"loss": 5.2266,
"mean_token_accuracy": 0.18473343849182128,
"num_tokens": 22080679.0,
"step": 10640
},
{
"entropy": 5.485489368438721,
"epoch": 0.9656204644412192,
"grad_norm": 0.875,
"learning_rate": 0.0004914000365490432,
"loss": 5.0439,
"mean_token_accuracy": 0.20180499255657197,
"num_tokens": 22092192.0,
"step": 10645
},
{
"entropy": 5.480154228210449,
"epoch": 0.9660740203193033,
"grad_norm": 0.89453125,
"learning_rate": 0.0004913911741338445,
"loss": 5.2587,
"mean_token_accuracy": 0.18562012463808059,
"num_tokens": 22102722.0,
"step": 10650
},
{
"entropy": 5.419040966033935,
"epoch": 0.9665275761973875,
"grad_norm": 0.94921875,
"learning_rate": 0.0004913823072436211,
"loss": 5.1247,
"mean_token_accuracy": 0.2006438970565796,
"num_tokens": 22113521.0,
"step": 10655
},
{
"entropy": 5.514348745346069,
"epoch": 0.9669811320754716,
"grad_norm": 0.921875,
"learning_rate": 0.0004913734358785565,
"loss": 5.2083,
"mean_token_accuracy": 0.19085652083158494,
"num_tokens": 22124891.0,
"step": 10660
},
{
"entropy": 5.4864380836486815,
"epoch": 0.9674346879535559,
"grad_norm": 0.9375,
"learning_rate": 0.0004913645600388341,
"loss": 5.2233,
"mean_token_accuracy": 0.18323416858911515,
"num_tokens": 22135972.0,
"step": 10665
},
{
"entropy": 5.536486482620239,
"epoch": 0.9678882438316401,
"grad_norm": 0.9453125,
"learning_rate": 0.0004913556797246374,
"loss": 5.2795,
"mean_token_accuracy": 0.17618962824344636,
"num_tokens": 22146623.0,
"step": 10670
},
{
"entropy": 5.481680917739868,
"epoch": 0.9683417997097242,
"grad_norm": 0.90234375,
"learning_rate": 0.0004913467949361503,
"loss": 5.2002,
"mean_token_accuracy": 0.18713169544935226,
"num_tokens": 22157619.0,
"step": 10675
},
{
"entropy": 5.543104553222657,
"epoch": 0.9687953555878084,
"grad_norm": 0.89453125,
"learning_rate": 0.0004913379056735561,
"loss": 5.1873,
"mean_token_accuracy": 0.1843522697687149,
"num_tokens": 22168655.0,
"step": 10680
},
{
"entropy": 5.477829074859619,
"epoch": 0.9692489114658926,
"grad_norm": 0.92578125,
"learning_rate": 0.000491329011937039,
"loss": 5.2341,
"mean_token_accuracy": 0.18528518527746202,
"num_tokens": 22178500.0,
"step": 10685
},
{
"entropy": 5.516538763046265,
"epoch": 0.9697024673439768,
"grad_norm": 0.95703125,
"learning_rate": 0.0004913201137267828,
"loss": 5.2278,
"mean_token_accuracy": 0.18400785326957703,
"num_tokens": 22188873.0,
"step": 10690
},
{
"entropy": 5.441410827636719,
"epoch": 0.970156023222061,
"grad_norm": 0.88671875,
"learning_rate": 0.0004913112110429715,
"loss": 5.1433,
"mean_token_accuracy": 0.19197122305631636,
"num_tokens": 22199743.0,
"step": 10695
},
{
"entropy": 5.446821880340576,
"epoch": 0.9706095791001451,
"grad_norm": 1.0390625,
"learning_rate": 0.0004913023038857892,
"loss": 5.096,
"mean_token_accuracy": 0.1934520870447159,
"num_tokens": 22210301.0,
"step": 10700
},
{
"entropy": 5.454444456100464,
"epoch": 0.9710631349782293,
"grad_norm": 0.91015625,
"learning_rate": 0.0004912933922554203,
"loss": 5.1086,
"mean_token_accuracy": 0.19499189853668214,
"num_tokens": 22220771.0,
"step": 10705
},
{
"entropy": 5.497599124908447,
"epoch": 0.9715166908563135,
"grad_norm": 0.890625,
"learning_rate": 0.0004912844761520487,
"loss": 5.2156,
"mean_token_accuracy": 0.18877754956483841,
"num_tokens": 22231100.0,
"step": 10710
},
{
"entropy": 5.513539361953735,
"epoch": 0.9719702467343977,
"grad_norm": 0.90234375,
"learning_rate": 0.0004912755555758592,
"loss": 5.1916,
"mean_token_accuracy": 0.19085400849580764,
"num_tokens": 22240997.0,
"step": 10715
},
{
"entropy": 5.515014171600342,
"epoch": 0.9724238026124818,
"grad_norm": 0.98828125,
"learning_rate": 0.000491266630527036,
"loss": 5.2256,
"mean_token_accuracy": 0.18837058991193772,
"num_tokens": 22251795.0,
"step": 10720
},
{
"entropy": 5.54486312866211,
"epoch": 0.972877358490566,
"grad_norm": 0.94921875,
"learning_rate": 0.0004912577010057638,
"loss": 5.2592,
"mean_token_accuracy": 0.1850780189037323,
"num_tokens": 22262465.0,
"step": 10725
},
{
"entropy": 5.606755971908569,
"epoch": 0.9733309143686503,
"grad_norm": 0.96875,
"learning_rate": 0.0004912487670122273,
"loss": 5.3558,
"mean_token_accuracy": 0.18126363158226014,
"num_tokens": 22272399.0,
"step": 10730
},
{
"entropy": 5.5661256313323975,
"epoch": 0.9737844702467344,
"grad_norm": 1.046875,
"learning_rate": 0.0004912398285466111,
"loss": 5.2756,
"mean_token_accuracy": 0.18773142844438553,
"num_tokens": 22282545.0,
"step": 10735
},
{
"entropy": 5.54059739112854,
"epoch": 0.9742380261248186,
"grad_norm": 1.015625,
"learning_rate": 0.0004912308856091002,
"loss": 5.2244,
"mean_token_accuracy": 0.19068962037563325,
"num_tokens": 22293154.0,
"step": 10740
},
{
"entropy": 5.5106744289398195,
"epoch": 0.9746915820029027,
"grad_norm": 0.94140625,
"learning_rate": 0.0004912219381998795,
"loss": 5.1466,
"mean_token_accuracy": 0.19592952430248262,
"num_tokens": 22302309.0,
"step": 10745
},
{
"entropy": 5.534269857406616,
"epoch": 0.9751451378809869,
"grad_norm": 0.9609375,
"learning_rate": 0.0004912129863191339,
"loss": 5.2587,
"mean_token_accuracy": 0.17782827466726303,
"num_tokens": 22312316.0,
"step": 10750
},
{
"entropy": 5.525972032546997,
"epoch": 0.9755986937590712,
"grad_norm": 0.9296875,
"learning_rate": 0.0004912040299670487,
"loss": 5.2329,
"mean_token_accuracy": 0.19108207672834396,
"num_tokens": 22322387.0,
"step": 10755
},
{
"entropy": 5.502649593353271,
"epoch": 0.9760522496371553,
"grad_norm": 0.984375,
"learning_rate": 0.0004911950691438091,
"loss": 5.2336,
"mean_token_accuracy": 0.19430004805326462,
"num_tokens": 22332759.0,
"step": 10760
},
{
"entropy": 5.5075767993927,
"epoch": 0.9765058055152395,
"grad_norm": 0.80859375,
"learning_rate": 0.0004911861038496004,
"loss": 5.2656,
"mean_token_accuracy": 0.18402133882045746,
"num_tokens": 22344056.0,
"step": 10765
},
{
"entropy": 5.455955648422242,
"epoch": 0.9769593613933236,
"grad_norm": 0.90625,
"learning_rate": 0.0004911771340846079,
"loss": 5.2084,
"mean_token_accuracy": 0.18565851747989653,
"num_tokens": 22353877.0,
"step": 10770
},
{
"entropy": 5.542308616638183,
"epoch": 0.9774129172714079,
"grad_norm": 0.9921875,
"learning_rate": 0.0004911681598490172,
"loss": 5.2525,
"mean_token_accuracy": 0.1882826343178749,
"num_tokens": 22363549.0,
"step": 10775
},
{
"entropy": 5.567686128616333,
"epoch": 0.977866473149492,
"grad_norm": 0.99609375,
"learning_rate": 0.0004911591811430139,
"loss": 5.2817,
"mean_token_accuracy": 0.1853038787841797,
"num_tokens": 22374040.0,
"step": 10780
},
{
"entropy": 5.568281555175782,
"epoch": 0.9783200290275762,
"grad_norm": 1.03125,
"learning_rate": 0.0004911501979667836,
"loss": 5.3363,
"mean_token_accuracy": 0.18623031675815582,
"num_tokens": 22384819.0,
"step": 10785
},
{
"entropy": 5.527229642868042,
"epoch": 0.9787735849056604,
"grad_norm": 0.95703125,
"learning_rate": 0.0004911412103205122,
"loss": 5.199,
"mean_token_accuracy": 0.19130196273326874,
"num_tokens": 22395130.0,
"step": 10790
},
{
"entropy": 5.504364633560181,
"epoch": 0.9792271407837445,
"grad_norm": 0.9453125,
"learning_rate": 0.0004911322182043855,
"loss": 5.2072,
"mean_token_accuracy": 0.18604805320501328,
"num_tokens": 22405093.0,
"step": 10795
},
{
"entropy": 5.493117141723633,
"epoch": 0.9796806966618288,
"grad_norm": 0.92578125,
"learning_rate": 0.0004911232216185893,
"loss": 5.2306,
"mean_token_accuracy": 0.18620621860027314,
"num_tokens": 22415008.0,
"step": 10800
},
{
"entropy": 5.524082040786743,
"epoch": 0.9801342525399129,
"grad_norm": 0.92578125,
"learning_rate": 0.0004911142205633099,
"loss": 5.2778,
"mean_token_accuracy": 0.1835907757282257,
"num_tokens": 22426486.0,
"step": 10805
},
{
"entropy": 5.492581939697265,
"epoch": 0.9805878084179971,
"grad_norm": 0.94140625,
"learning_rate": 0.0004911052150387334,
"loss": 5.1811,
"mean_token_accuracy": 0.18908338397741317,
"num_tokens": 22436844.0,
"step": 10810
},
{
"entropy": 5.553999710083008,
"epoch": 0.9810413642960812,
"grad_norm": 1.125,
"learning_rate": 0.0004910962050450458,
"loss": 5.3059,
"mean_token_accuracy": 0.18974106162786483,
"num_tokens": 22446630.0,
"step": 10815
},
{
"entropy": 5.542135238647461,
"epoch": 0.9814949201741655,
"grad_norm": 0.9921875,
"learning_rate": 0.0004910871905824338,
"loss": 5.1387,
"mean_token_accuracy": 0.1928548112511635,
"num_tokens": 22456406.0,
"step": 10820
},
{
"entropy": 5.45408296585083,
"epoch": 0.9819484760522497,
"grad_norm": 0.9921875,
"learning_rate": 0.0004910781716510837,
"loss": 5.138,
"mean_token_accuracy": 0.1933705762028694,
"num_tokens": 22466210.0,
"step": 10825
},
{
"entropy": 5.4947069644927975,
"epoch": 0.9824020319303338,
"grad_norm": 0.9609375,
"learning_rate": 0.0004910691482511817,
"loss": 5.2356,
"mean_token_accuracy": 0.1944747418165207,
"num_tokens": 22476978.0,
"step": 10830
},
{
"entropy": 5.59715723991394,
"epoch": 0.982855587808418,
"grad_norm": 0.9453125,
"learning_rate": 0.0004910601203829148,
"loss": 5.194,
"mean_token_accuracy": 0.18800618201494218,
"num_tokens": 22488613.0,
"step": 10835
},
{
"entropy": 5.467583513259887,
"epoch": 0.9833091436865021,
"grad_norm": 0.96484375,
"learning_rate": 0.0004910510880464696,
"loss": 5.1093,
"mean_token_accuracy": 0.20096988826990128,
"num_tokens": 22498312.0,
"step": 10840
},
{
"entropy": 5.482868099212647,
"epoch": 0.9837626995645864,
"grad_norm": 1.0,
"learning_rate": 0.0004910420512420327,
"loss": 5.2111,
"mean_token_accuracy": 0.1898895427584648,
"num_tokens": 22508131.0,
"step": 10845
},
{
"entropy": 5.539039945602417,
"epoch": 0.9842162554426706,
"grad_norm": 0.9765625,
"learning_rate": 0.0004910330099697913,
"loss": 5.2132,
"mean_token_accuracy": 0.1903369903564453,
"num_tokens": 22517842.0,
"step": 10850
},
{
"entropy": 5.527409839630127,
"epoch": 0.9846698113207547,
"grad_norm": 0.94140625,
"learning_rate": 0.0004910239642299321,
"loss": 5.1837,
"mean_token_accuracy": 0.19023292511701584,
"num_tokens": 22528239.0,
"step": 10855
},
{
"entropy": 5.466523361206055,
"epoch": 0.9851233671988389,
"grad_norm": 0.890625,
"learning_rate": 0.0004910149140226422,
"loss": 5.1724,
"mean_token_accuracy": 0.1914309471845627,
"num_tokens": 22539734.0,
"step": 10860
},
{
"entropy": 5.503134536743164,
"epoch": 0.9855769230769231,
"grad_norm": 0.90625,
"learning_rate": 0.0004910058593481089,
"loss": 5.1701,
"mean_token_accuracy": 0.19570856541395187,
"num_tokens": 22549796.0,
"step": 10865
},
{
"entropy": 5.4575072765350345,
"epoch": 0.9860304789550073,
"grad_norm": 0.98828125,
"learning_rate": 0.0004909968002065195,
"loss": 5.158,
"mean_token_accuracy": 0.1910772830247879,
"num_tokens": 22559934.0,
"step": 10870
},
{
"entropy": 5.6052713871002195,
"epoch": 0.9864840348330914,
"grad_norm": 1.0234375,
"learning_rate": 0.000490987736598061,
"loss": 5.2598,
"mean_token_accuracy": 0.1797322601079941,
"num_tokens": 22569864.0,
"step": 10875
},
{
"entropy": 5.504677820205688,
"epoch": 0.9869375907111756,
"grad_norm": 0.890625,
"learning_rate": 0.0004909786685229211,
"loss": 5.1964,
"mean_token_accuracy": 0.19451596289873124,
"num_tokens": 22580687.0,
"step": 10880
},
{
"entropy": 5.538554286956787,
"epoch": 0.9873911465892597,
"grad_norm": 0.9765625,
"learning_rate": 0.0004909695959812875,
"loss": 5.2494,
"mean_token_accuracy": 0.18825656920671463,
"num_tokens": 22590507.0,
"step": 10885
},
{
"entropy": 5.55866584777832,
"epoch": 0.987844702467344,
"grad_norm": 1.0234375,
"learning_rate": 0.0004909605189733474,
"loss": 5.2668,
"mean_token_accuracy": 0.18947904407978058,
"num_tokens": 22601901.0,
"step": 10890
},
{
"entropy": 5.501526641845703,
"epoch": 0.9882982583454282,
"grad_norm": 1.0078125,
"learning_rate": 0.0004909514374992889,
"loss": 5.2155,
"mean_token_accuracy": 0.18769705891609192,
"num_tokens": 22613003.0,
"step": 10895
},
{
"entropy": 5.447288465499878,
"epoch": 0.9887518142235123,
"grad_norm": 0.98828125,
"learning_rate": 0.0004909423515592996,
"loss": 5.0577,
"mean_token_accuracy": 0.1979197680950165,
"num_tokens": 22623401.0,
"step": 10900
},
{
"entropy": 5.519672107696533,
"epoch": 0.9892053701015965,
"grad_norm": 1.0,
"learning_rate": 0.0004909332611535675,
"loss": 5.2108,
"mean_token_accuracy": 0.1883891209959984,
"num_tokens": 22633147.0,
"step": 10905
},
{
"entropy": 5.47981538772583,
"epoch": 0.9896589259796807,
"grad_norm": 0.8984375,
"learning_rate": 0.0004909241662822804,
"loss": 5.0727,
"mean_token_accuracy": 0.20440202653408052,
"num_tokens": 22642990.0,
"step": 10910
},
{
"entropy": 5.50399341583252,
"epoch": 0.9901124818577649,
"grad_norm": 1.0,
"learning_rate": 0.0004909150669456265,
"loss": 5.2827,
"mean_token_accuracy": 0.18473678827285767,
"num_tokens": 22652734.0,
"step": 10915
},
{
"entropy": 5.488545656204224,
"epoch": 0.9905660377358491,
"grad_norm": 0.94921875,
"learning_rate": 0.000490905963143794,
"loss": 5.2849,
"mean_token_accuracy": 0.18099266439676284,
"num_tokens": 22662968.0,
"step": 10920
},
{
"entropy": 5.44248948097229,
"epoch": 0.9910195936139332,
"grad_norm": 0.95703125,
"learning_rate": 0.0004908968548769713,
"loss": 5.1281,
"mean_token_accuracy": 0.20352931469678878,
"num_tokens": 22673157.0,
"step": 10925
},
{
"entropy": 5.515015172958374,
"epoch": 0.9914731494920174,
"grad_norm": 1.0078125,
"learning_rate": 0.0004908877421453466,
"loss": 5.2599,
"mean_token_accuracy": 0.18747966289520263,
"num_tokens": 22683105.0,
"step": 10930
},
{
"entropy": 5.4896814823150635,
"epoch": 0.9919267053701016,
"grad_norm": 0.9453125,
"learning_rate": 0.0004908786249491082,
"loss": 5.2022,
"mean_token_accuracy": 0.1996642142534256,
"num_tokens": 22693793.0,
"step": 10935
},
{
"entropy": 5.618080711364746,
"epoch": 0.9923802612481858,
"grad_norm": 1.0,
"learning_rate": 0.000490869503288445,
"loss": 5.252,
"mean_token_accuracy": 0.18813893795013428,
"num_tokens": 22703884.0,
"step": 10940
},
{
"entropy": 5.548908472061157,
"epoch": 0.99283381712627,
"grad_norm": 0.9921875,
"learning_rate": 0.0004908603771635454,
"loss": 5.2763,
"mean_token_accuracy": 0.18646926432847977,
"num_tokens": 22713699.0,
"step": 10945
},
{
"entropy": 5.472669267654419,
"epoch": 0.9932873730043541,
"grad_norm": 1.0,
"learning_rate": 0.0004908512465745982,
"loss": 5.1256,
"mean_token_accuracy": 0.19451946765184402,
"num_tokens": 22722871.0,
"step": 10950
},
{
"entropy": 5.479227352142334,
"epoch": 0.9937409288824384,
"grad_norm": 0.9765625,
"learning_rate": 0.0004908421115217922,
"loss": 5.1383,
"mean_token_accuracy": 0.1876304790377617,
"num_tokens": 22733314.0,
"step": 10955
},
{
"entropy": 5.432501792907715,
"epoch": 0.9941944847605225,
"grad_norm": 0.953125,
"learning_rate": 0.0004908329720053164,
"loss": 5.1323,
"mean_token_accuracy": 0.19701928347349168,
"num_tokens": 22743255.0,
"step": 10960
},
{
"entropy": 5.412756109237671,
"epoch": 0.9946480406386067,
"grad_norm": 0.9296875,
"learning_rate": 0.0004908238280253597,
"loss": 5.1195,
"mean_token_accuracy": 0.19759708195924758,
"num_tokens": 22753857.0,
"step": 10965
},
{
"entropy": 5.531611490249634,
"epoch": 0.9951015965166908,
"grad_norm": 1.0078125,
"learning_rate": 0.0004908146795821112,
"loss": 5.1134,
"mean_token_accuracy": 0.19179392755031585,
"num_tokens": 22764326.0,
"step": 10970
},
{
"entropy": 5.5185198307037355,
"epoch": 0.995555152394775,
"grad_norm": 0.96484375,
"learning_rate": 0.0004908055266757602,
"loss": 5.1934,
"mean_token_accuracy": 0.19486576914787293,
"num_tokens": 22774958.0,
"step": 10975
},
{
"entropy": 5.425844955444336,
"epoch": 0.9960087082728593,
"grad_norm": 0.99609375,
"learning_rate": 0.0004907963693064958,
"loss": 5.1128,
"mean_token_accuracy": 0.1925733730196953,
"num_tokens": 22785616.0,
"step": 10980
},
{
"entropy": 5.565258312225342,
"epoch": 0.9964622641509434,
"grad_norm": 0.8359375,
"learning_rate": 0.0004907872074745075,
"loss": 5.2837,
"mean_token_accuracy": 0.1845133572816849,
"num_tokens": 22795967.0,
"step": 10985
},
{
"entropy": 5.490162181854248,
"epoch": 0.9969158200290276,
"grad_norm": 0.96484375,
"learning_rate": 0.0004907780411799849,
"loss": 5.2794,
"mean_token_accuracy": 0.18557634502649306,
"num_tokens": 22806335.0,
"step": 10990
},
{
"entropy": 5.548031806945801,
"epoch": 0.9973693759071117,
"grad_norm": 0.96484375,
"learning_rate": 0.0004907688704231173,
"loss": 5.2144,
"mean_token_accuracy": 0.19313637167215347,
"num_tokens": 22816390.0,
"step": 10995
},
{
"entropy": 5.6186151027679445,
"epoch": 0.997822931785196,
"grad_norm": 1.0,
"learning_rate": 0.0004907596952040945,
"loss": 5.2726,
"mean_token_accuracy": 0.18671630173921586,
"num_tokens": 22826395.0,
"step": 11000
},
{
"entropy": 5.547494173049927,
"epoch": 0.9982764876632801,
"grad_norm": 0.9921875,
"learning_rate": 0.0004907505155231063,
"loss": 5.2584,
"mean_token_accuracy": 0.18838359266519547,
"num_tokens": 22836790.0,
"step": 11005
},
{
"entropy": 5.488029050827026,
"epoch": 0.9987300435413643,
"grad_norm": 0.97265625,
"learning_rate": 0.0004907413313803424,
"loss": 5.1407,
"mean_token_accuracy": 0.19004008322954177,
"num_tokens": 22846687.0,
"step": 11010
},
{
"entropy": 5.53559718132019,
"epoch": 0.9991835994194485,
"grad_norm": 0.96875,
"learning_rate": 0.0004907321427759928,
"loss": 5.168,
"mean_token_accuracy": 0.19271109402179717,
"num_tokens": 22857208.0,
"step": 11015
},
{
"entropy": 5.513091611862182,
"epoch": 0.9996371552975326,
"grad_norm": 0.88671875,
"learning_rate": 0.0004907229497102475,
"loss": 5.2714,
"mean_token_accuracy": 0.18417642563581466,
"num_tokens": 22868556.0,
"step": 11020
},
{
"entropy": 5.550210618972779,
"epoch": 1.0000907111756168,
"grad_norm": 0.79296875,
"learning_rate": 0.0004907137521832966,
"loss": 5.2531,
"mean_token_accuracy": 0.18888288289308547,
"num_tokens": 22880081.0,
"step": 11025
},
{
"entropy": 5.494132804870605,
"epoch": 1.000544267053701,
"grad_norm": 0.83984375,
"learning_rate": 0.0004907045501953304,
"loss": 5.0935,
"mean_token_accuracy": 0.19637981653213502,
"num_tokens": 22892141.0,
"step": 11030
},
{
"entropy": 5.521714591979981,
"epoch": 1.0009978229317853,
"grad_norm": 0.92578125,
"learning_rate": 0.000490695343746539,
"loss": 5.1043,
"mean_token_accuracy": 0.19702826887369157,
"num_tokens": 22902688.0,
"step": 11035
},
{
"entropy": 5.4263575077056885,
"epoch": 1.0014513788098693,
"grad_norm": 1.046875,
"learning_rate": 0.0004906861328371131,
"loss": 5.0424,
"mean_token_accuracy": 0.1953244224190712,
"num_tokens": 22912453.0,
"step": 11040
},
{
"entropy": 5.511580848693848,
"epoch": 1.0019049346879536,
"grad_norm": 0.96484375,
"learning_rate": 0.0004906769174672429,
"loss": 5.2007,
"mean_token_accuracy": 0.18563229739665985,
"num_tokens": 22923167.0,
"step": 11045
},
{
"entropy": 5.539322376251221,
"epoch": 1.0023584905660377,
"grad_norm": 0.953125,
"learning_rate": 0.000490667697637119,
"loss": 5.1607,
"mean_token_accuracy": 0.18884927779436111,
"num_tokens": 22934753.0,
"step": 11050
},
{
"entropy": 5.468543291091919,
"epoch": 1.002812046444122,
"grad_norm": 1.015625,
"learning_rate": 0.0004906584733469324,
"loss": 5.0864,
"mean_token_accuracy": 0.20028799921274185,
"num_tokens": 22945277.0,
"step": 11055
},
{
"entropy": 5.522582197189331,
"epoch": 1.0032656023222062,
"grad_norm": 1.0625,
"learning_rate": 0.0004906492445968735,
"loss": 5.153,
"mean_token_accuracy": 0.18677352368831635,
"num_tokens": 22955797.0,
"step": 11060
},
{
"entropy": 5.507099533081055,
"epoch": 1.0037191582002902,
"grad_norm": 0.9609375,
"learning_rate": 0.0004906400113871332,
"loss": 5.1494,
"mean_token_accuracy": 0.19461202323436738,
"num_tokens": 22965969.0,
"step": 11065
},
{
"entropy": 5.490888786315918,
"epoch": 1.0041727140783745,
"grad_norm": 1.0703125,
"learning_rate": 0.0004906307737179026,
"loss": 5.0969,
"mean_token_accuracy": 0.20535642355680467,
"num_tokens": 22975323.0,
"step": 11070
},
{
"entropy": 5.465540266036987,
"epoch": 1.0046262699564585,
"grad_norm": 0.890625,
"learning_rate": 0.0004906215315893726,
"loss": 5.1059,
"mean_token_accuracy": 0.19620248824357986,
"num_tokens": 22985063.0,
"step": 11075
},
{
"entropy": 5.544905090332032,
"epoch": 1.0050798258345428,
"grad_norm": 0.890625,
"learning_rate": 0.0004906122850017344,
"loss": 5.2382,
"mean_token_accuracy": 0.1847388505935669,
"num_tokens": 22996001.0,
"step": 11080
},
{
"entropy": 5.487858152389526,
"epoch": 1.005533381712627,
"grad_norm": 0.97265625,
"learning_rate": 0.0004906030339551792,
"loss": 5.1552,
"mean_token_accuracy": 0.1921267867088318,
"num_tokens": 23006896.0,
"step": 11085
},
{
"entropy": 5.534958028793335,
"epoch": 1.0059869375907111,
"grad_norm": 0.96875,
"learning_rate": 0.0004905937784498983,
"loss": 5.1214,
"mean_token_accuracy": 0.19611595422029496,
"num_tokens": 23017080.0,
"step": 11090
},
{
"entropy": 5.504563283920288,
"epoch": 1.0064404934687954,
"grad_norm": 1.078125,
"learning_rate": 0.0004905845184860831,
"loss": 5.0746,
"mean_token_accuracy": 0.19814439564943315,
"num_tokens": 23027168.0,
"step": 11095
},
{
"entropy": 5.45788164138794,
"epoch": 1.0068940493468794,
"grad_norm": 0.93359375,
"learning_rate": 0.0004905752540639252,
"loss": 5.1253,
"mean_token_accuracy": 0.1916827917098999,
"num_tokens": 23037364.0,
"step": 11100
},
{
"entropy": 5.4299335956573485,
"epoch": 1.0073476052249637,
"grad_norm": 0.984375,
"learning_rate": 0.0004905659851836161,
"loss": 5.0356,
"mean_token_accuracy": 0.19612774699926377,
"num_tokens": 23047609.0,
"step": 11105
},
{
"entropy": 5.502022409439087,
"epoch": 1.007801161103048,
"grad_norm": 0.9296875,
"learning_rate": 0.0004905567118453475,
"loss": 5.1641,
"mean_token_accuracy": 0.19047075808048247,
"num_tokens": 23057942.0,
"step": 11110
},
{
"entropy": 5.470209121704102,
"epoch": 1.008254716981132,
"grad_norm": 0.96484375,
"learning_rate": 0.0004905474340493111,
"loss": 5.0727,
"mean_token_accuracy": 0.19394411146640778,
"num_tokens": 23069119.0,
"step": 11115
},
{
"entropy": 5.443198490142822,
"epoch": 1.0087082728592163,
"grad_norm": 0.9609375,
"learning_rate": 0.0004905381517956988,
"loss": 5.0687,
"mean_token_accuracy": 0.19524529427289963,
"num_tokens": 23080060.0,
"step": 11120
},
{
"entropy": 5.4462425231933596,
"epoch": 1.0091618287373005,
"grad_norm": 0.8984375,
"learning_rate": 0.0004905288650847027,
"loss": 5.0992,
"mean_token_accuracy": 0.1914682611823082,
"num_tokens": 23090645.0,
"step": 11125
},
{
"entropy": 5.510693311691284,
"epoch": 1.0096153846153846,
"grad_norm": 0.984375,
"learning_rate": 0.0004905195739165146,
"loss": 5.1079,
"mean_token_accuracy": 0.19879512786865233,
"num_tokens": 23100449.0,
"step": 11130
},
{
"entropy": 5.496145629882813,
"epoch": 1.0100689404934688,
"grad_norm": 0.99609375,
"learning_rate": 0.000490510278291327,
"loss": 5.1084,
"mean_token_accuracy": 0.19123156666755675,
"num_tokens": 23110415.0,
"step": 11135
},
{
"entropy": 5.447995948791504,
"epoch": 1.010522496371553,
"grad_norm": 0.95703125,
"learning_rate": 0.0004905009782093318,
"loss": 5.097,
"mean_token_accuracy": 0.19483419209718705,
"num_tokens": 23120336.0,
"step": 11140
},
{
"entropy": 5.425385141372681,
"epoch": 1.0109760522496372,
"grad_norm": 0.99609375,
"learning_rate": 0.0004904916736707214,
"loss": 5.1687,
"mean_token_accuracy": 0.1853318989276886,
"num_tokens": 23130568.0,
"step": 11145
},
{
"entropy": 5.5349946975708,
"epoch": 1.0114296081277214,
"grad_norm": 0.953125,
"learning_rate": 0.0004904823646756885,
"loss": 5.1248,
"mean_token_accuracy": 0.19526209384202958,
"num_tokens": 23140080.0,
"step": 11150
},
{
"entropy": 5.4925604343414305,
"epoch": 1.0118831640058055,
"grad_norm": 0.96875,
"learning_rate": 0.0004904730512244252,
"loss": 5.0562,
"mean_token_accuracy": 0.1917285680770874,
"num_tokens": 23150360.0,
"step": 11155
},
{
"entropy": 5.417743730545044,
"epoch": 1.0123367198838897,
"grad_norm": 0.93359375,
"learning_rate": 0.0004904637333171243,
"loss": 5.0548,
"mean_token_accuracy": 0.18818178176879882,
"num_tokens": 23160620.0,
"step": 11160
},
{
"entropy": 5.523185205459595,
"epoch": 1.0127902757619738,
"grad_norm": 0.9453125,
"learning_rate": 0.0004904544109539785,
"loss": 5.1555,
"mean_token_accuracy": 0.19521294981241227,
"num_tokens": 23171920.0,
"step": 11165
},
{
"entropy": 5.492150783538818,
"epoch": 1.013243831640058,
"grad_norm": 0.88671875,
"learning_rate": 0.0004904450841351805,
"loss": 5.1168,
"mean_token_accuracy": 0.19893324375152588,
"num_tokens": 23182929.0,
"step": 11170
},
{
"entropy": 5.490244626998901,
"epoch": 1.0136973875181423,
"grad_norm": 0.89453125,
"learning_rate": 0.0004904357528609236,
"loss": 5.0738,
"mean_token_accuracy": 0.18984648883342742,
"num_tokens": 23193020.0,
"step": 11175
},
{
"entropy": 5.427248907089234,
"epoch": 1.0141509433962264,
"grad_norm": 0.99609375,
"learning_rate": 0.0004904264171314,
"loss": 5.0854,
"mean_token_accuracy": 0.19872664362192155,
"num_tokens": 23204073.0,
"step": 11180
},
{
"entropy": 5.426294994354248,
"epoch": 1.0146044992743106,
"grad_norm": 0.921875,
"learning_rate": 0.0004904170769468035,
"loss": 4.9959,
"mean_token_accuracy": 0.1919043093919754,
"num_tokens": 23214376.0,
"step": 11185
},
{
"entropy": 5.511206245422363,
"epoch": 1.0150580551523947,
"grad_norm": 0.9765625,
"learning_rate": 0.0004904077323073268,
"loss": 5.1612,
"mean_token_accuracy": 0.19419621229171752,
"num_tokens": 23224304.0,
"step": 11190
},
{
"entropy": 5.498494100570679,
"epoch": 1.015511611030479,
"grad_norm": 0.9453125,
"learning_rate": 0.0004903983832131633,
"loss": 5.1779,
"mean_token_accuracy": 0.1909933015704155,
"num_tokens": 23235324.0,
"step": 11195
},
{
"entropy": 5.499328231811523,
"epoch": 1.0159651669085632,
"grad_norm": 1.0078125,
"learning_rate": 0.0004903890296645063,
"loss": 5.1352,
"mean_token_accuracy": 0.19310501515865325,
"num_tokens": 23245774.0,
"step": 11200
},
{
"entropy": 5.526724481582642,
"epoch": 1.0164187227866472,
"grad_norm": 0.90234375,
"learning_rate": 0.0004903796716615495,
"loss": 5.0739,
"mean_token_accuracy": 0.19474828839302064,
"num_tokens": 23256274.0,
"step": 11205
},
{
"entropy": 5.4702582359313965,
"epoch": 1.0168722786647315,
"grad_norm": 1.0234375,
"learning_rate": 0.0004903703092044861,
"loss": 5.107,
"mean_token_accuracy": 0.19234081357717514,
"num_tokens": 23266304.0,
"step": 11210
},
{
"entropy": 5.389174270629883,
"epoch": 1.0173258345428158,
"grad_norm": 1.015625,
"learning_rate": 0.0004903609422935097,
"loss": 5.0099,
"mean_token_accuracy": 0.19811663776636124,
"num_tokens": 23276552.0,
"step": 11215
},
{
"entropy": 5.4639805316925045,
"epoch": 1.0177793904208998,
"grad_norm": 0.8828125,
"learning_rate": 0.0004903515709288143,
"loss": 5.1256,
"mean_token_accuracy": 0.1883719265460968,
"num_tokens": 23287799.0,
"step": 11220
},
{
"entropy": 5.401679563522339,
"epoch": 1.018232946298984,
"grad_norm": 0.98046875,
"learning_rate": 0.0004903421951105933,
"loss": 4.9809,
"mean_token_accuracy": 0.20238814353942872,
"num_tokens": 23297958.0,
"step": 11225
},
{
"entropy": 5.5005217552185055,
"epoch": 1.0186865021770681,
"grad_norm": 1.109375,
"learning_rate": 0.0004903328148390411,
"loss": 5.1624,
"mean_token_accuracy": 0.18677162379026413,
"num_tokens": 23307523.0,
"step": 11230
},
{
"entropy": 5.501028251647949,
"epoch": 1.0191400580551524,
"grad_norm": 0.92578125,
"learning_rate": 0.0004903234301143513,
"loss": 5.1515,
"mean_token_accuracy": 0.19437760561704637,
"num_tokens": 23317970.0,
"step": 11235
},
{
"entropy": 5.460223770141601,
"epoch": 1.0195936139332367,
"grad_norm": 0.98046875,
"learning_rate": 0.0004903140409367181,
"loss": 5.1074,
"mean_token_accuracy": 0.19670169800519943,
"num_tokens": 23328612.0,
"step": 11240
},
{
"entropy": 5.452767467498779,
"epoch": 1.0200471698113207,
"grad_norm": 1.0390625,
"learning_rate": 0.0004903046473063357,
"loss": 5.0766,
"mean_token_accuracy": 0.1902572751045227,
"num_tokens": 23338104.0,
"step": 11245
},
{
"entropy": 5.446892786026001,
"epoch": 1.020500725689405,
"grad_norm": 0.96484375,
"learning_rate": 0.0004902952492233983,
"loss": 5.0129,
"mean_token_accuracy": 0.2074433147907257,
"num_tokens": 23348105.0,
"step": 11250
},
{
"entropy": 5.42758994102478,
"epoch": 1.020954281567489,
"grad_norm": 0.984375,
"learning_rate": 0.0004902858466881002,
"loss": 4.9701,
"mean_token_accuracy": 0.19822793155908586,
"num_tokens": 23357150.0,
"step": 11255
},
{
"entropy": 5.470262479782105,
"epoch": 1.0214078374455733,
"grad_norm": 1.0390625,
"learning_rate": 0.000490276439700636,
"loss": 5.182,
"mean_token_accuracy": 0.18963402211666108,
"num_tokens": 23367223.0,
"step": 11260
},
{
"entropy": 5.476350355148315,
"epoch": 1.0218613933236576,
"grad_norm": 0.96484375,
"learning_rate": 0.0004902670282612002,
"loss": 5.1177,
"mean_token_accuracy": 0.19103692173957826,
"num_tokens": 23377664.0,
"step": 11265
},
{
"entropy": 5.442233085632324,
"epoch": 1.0223149492017416,
"grad_norm": 0.94140625,
"learning_rate": 0.0004902576123699873,
"loss": 5.0096,
"mean_token_accuracy": 0.19532963782548904,
"num_tokens": 23388057.0,
"step": 11270
},
{
"entropy": 5.381910991668701,
"epoch": 1.0227685050798259,
"grad_norm": 1.0,
"learning_rate": 0.0004902481920271922,
"loss": 4.9596,
"mean_token_accuracy": 0.19629869908094405,
"num_tokens": 23397808.0,
"step": 11275
},
{
"entropy": 5.395374584197998,
"epoch": 1.02322206095791,
"grad_norm": 0.96875,
"learning_rate": 0.0004902387672330097,
"loss": 5.0443,
"mean_token_accuracy": 0.19195702075958251,
"num_tokens": 23406979.0,
"step": 11280
},
{
"entropy": 5.450476551055909,
"epoch": 1.0236756168359942,
"grad_norm": 1.0546875,
"learning_rate": 0.0004902293379876345,
"loss": 5.0932,
"mean_token_accuracy": 0.19491908848285674,
"num_tokens": 23416545.0,
"step": 11285
},
{
"entropy": 5.444452714920044,
"epoch": 1.0241291727140784,
"grad_norm": 0.87109375,
"learning_rate": 0.0004902199042912618,
"loss": 5.0985,
"mean_token_accuracy": 0.18906430751085282,
"num_tokens": 23427464.0,
"step": 11290
},
{
"entropy": 5.502170610427856,
"epoch": 1.0245827285921625,
"grad_norm": 1.0,
"learning_rate": 0.0004902104661440868,
"loss": 5.1325,
"mean_token_accuracy": 0.19716407209634781,
"num_tokens": 23436690.0,
"step": 11295
},
{
"entropy": 5.468004512786865,
"epoch": 1.0250362844702468,
"grad_norm": 0.98828125,
"learning_rate": 0.0004902010235463042,
"loss": 5.0506,
"mean_token_accuracy": 0.19812326580286027,
"num_tokens": 23446701.0,
"step": 11300
},
{
"entropy": 5.535991668701172,
"epoch": 1.025489840348331,
"grad_norm": 0.96875,
"learning_rate": 0.0004901915764981098,
"loss": 5.1445,
"mean_token_accuracy": 0.19191909879446029,
"num_tokens": 23456966.0,
"step": 11305
},
{
"entropy": 5.470449829101563,
"epoch": 1.025943396226415,
"grad_norm": 1.09375,
"learning_rate": 0.0004901821249996988,
"loss": 5.1001,
"mean_token_accuracy": 0.1941937953233719,
"num_tokens": 23467300.0,
"step": 11310
},
{
"entropy": 5.5109635353088375,
"epoch": 1.0263969521044993,
"grad_norm": 0.9609375,
"learning_rate": 0.0004901726690512665,
"loss": 5.162,
"mean_token_accuracy": 0.18955274671316147,
"num_tokens": 23478457.0,
"step": 11315
},
{
"entropy": 5.4429868221282955,
"epoch": 1.0268505079825834,
"grad_norm": 0.9921875,
"learning_rate": 0.0004901632086530087,
"loss": 5.044,
"mean_token_accuracy": 0.2002337396144867,
"num_tokens": 23488066.0,
"step": 11320
},
{
"entropy": 5.4729485511779785,
"epoch": 1.0273040638606676,
"grad_norm": 1.0,
"learning_rate": 0.0004901537438051207,
"loss": 5.1232,
"mean_token_accuracy": 0.18899547904729844,
"num_tokens": 23498856.0,
"step": 11325
},
{
"entropy": 5.491955137252807,
"epoch": 1.027757619738752,
"grad_norm": 0.96875,
"learning_rate": 0.0004901442745077986,
"loss": 5.0879,
"mean_token_accuracy": 0.1923009291291237,
"num_tokens": 23509845.0,
"step": 11330
},
{
"entropy": 5.527870321273804,
"epoch": 1.028211175616836,
"grad_norm": 1.046875,
"learning_rate": 0.0004901348007612381,
"loss": 5.2004,
"mean_token_accuracy": 0.1924690917134285,
"num_tokens": 23520322.0,
"step": 11335
},
{
"entropy": 5.480651807785034,
"epoch": 1.0286647314949202,
"grad_norm": 0.9609375,
"learning_rate": 0.0004901253225656351,
"loss": 5.1352,
"mean_token_accuracy": 0.19338575154542922,
"num_tokens": 23530030.0,
"step": 11340
},
{
"entropy": 5.565394401550293,
"epoch": 1.0291182873730043,
"grad_norm": 0.9921875,
"learning_rate": 0.0004901158399211855,
"loss": 5.1925,
"mean_token_accuracy": 0.19459724128246308,
"num_tokens": 23541063.0,
"step": 11345
},
{
"entropy": 5.423709774017334,
"epoch": 1.0295718432510885,
"grad_norm": 1.0234375,
"learning_rate": 0.0004901063528280856,
"loss": 5.0923,
"mean_token_accuracy": 0.19012126177549363,
"num_tokens": 23551693.0,
"step": 11350
},
{
"entropy": 5.477392244338989,
"epoch": 1.0300253991291728,
"grad_norm": 1.0390625,
"learning_rate": 0.0004900968612865314,
"loss": 5.1251,
"mean_token_accuracy": 0.1971671923995018,
"num_tokens": 23561758.0,
"step": 11355
},
{
"entropy": 5.418482398986816,
"epoch": 1.0304789550072568,
"grad_norm": 1.03125,
"learning_rate": 0.0004900873652967195,
"loss": 5.0701,
"mean_token_accuracy": 0.19031274020671846,
"num_tokens": 23571912.0,
"step": 11360
},
{
"entropy": 5.442516422271728,
"epoch": 1.030932510885341,
"grad_norm": 0.9375,
"learning_rate": 0.0004900778648588459,
"loss": 5.035,
"mean_token_accuracy": 0.19667790979146957,
"num_tokens": 23582600.0,
"step": 11365
},
{
"entropy": 5.440606641769409,
"epoch": 1.0313860667634251,
"grad_norm": 1.046875,
"learning_rate": 0.0004900683599731073,
"loss": 5.0044,
"mean_token_accuracy": 0.20124102234840394,
"num_tokens": 23592761.0,
"step": 11370
},
{
"entropy": 5.49815354347229,
"epoch": 1.0318396226415094,
"grad_norm": 1.046875,
"learning_rate": 0.0004900588506397002,
"loss": 5.1829,
"mean_token_accuracy": 0.1900082364678383,
"num_tokens": 23603119.0,
"step": 11375
},
{
"entropy": 5.384500169754029,
"epoch": 1.0322931785195937,
"grad_norm": 0.97265625,
"learning_rate": 0.0004900493368588213,
"loss": 5.0585,
"mean_token_accuracy": 0.20479914397001267,
"num_tokens": 23613916.0,
"step": 11380
},
{
"entropy": 5.526654195785523,
"epoch": 1.0327467343976777,
"grad_norm": 1.0625,
"learning_rate": 0.0004900398186306672,
"loss": 5.2217,
"mean_token_accuracy": 0.1877501666545868,
"num_tokens": 23623956.0,
"step": 11385
},
{
"entropy": 5.456167316436767,
"epoch": 1.033200290275762,
"grad_norm": 0.94140625,
"learning_rate": 0.000490030295955435,
"loss": 5.1032,
"mean_token_accuracy": 0.19367883652448653,
"num_tokens": 23634666.0,
"step": 11390
},
{
"entropy": 5.424622917175293,
"epoch": 1.0336538461538463,
"grad_norm": 1.0234375,
"learning_rate": 0.0004900207688333215,
"loss": 5.0608,
"mean_token_accuracy": 0.18913532048463821,
"num_tokens": 23645601.0,
"step": 11395
},
{
"entropy": 5.510989427566528,
"epoch": 1.0341074020319303,
"grad_norm": 1.015625,
"learning_rate": 0.0004900112372645236,
"loss": 5.1375,
"mean_token_accuracy": 0.19037441164255142,
"num_tokens": 23656339.0,
"step": 11400
},
{
"entropy": 5.461063098907471,
"epoch": 1.0345609579100146,
"grad_norm": 0.89453125,
"learning_rate": 0.0004900017012492386,
"loss": 5.1066,
"mean_token_accuracy": 0.19083172529935838,
"num_tokens": 23667433.0,
"step": 11405
},
{
"entropy": 5.502894353866577,
"epoch": 1.0350145137880986,
"grad_norm": 0.9765625,
"learning_rate": 0.0004899921607876634,
"loss": 5.1234,
"mean_token_accuracy": 0.19392693638801575,
"num_tokens": 23678186.0,
"step": 11410
},
{
"entropy": 5.466448020935059,
"epoch": 1.0354680696661829,
"grad_norm": 0.9609375,
"learning_rate": 0.0004899826158799958,
"loss": 5.0781,
"mean_token_accuracy": 0.19278447926044465,
"num_tokens": 23688317.0,
"step": 11415
},
{
"entropy": 5.432892513275147,
"epoch": 1.0359216255442671,
"grad_norm": 1.0078125,
"learning_rate": 0.0004899730665264328,
"loss": 5.0298,
"mean_token_accuracy": 0.19775377660989762,
"num_tokens": 23697752.0,
"step": 11420
},
{
"entropy": 5.389712142944336,
"epoch": 1.0363751814223512,
"grad_norm": 0.9921875,
"learning_rate": 0.0004899635127271721,
"loss": 5.007,
"mean_token_accuracy": 0.20241805762052537,
"num_tokens": 23707803.0,
"step": 11425
},
{
"entropy": 5.372399187088012,
"epoch": 1.0368287373004355,
"grad_norm": 1.0234375,
"learning_rate": 0.0004899539544824111,
"loss": 5.0122,
"mean_token_accuracy": 0.2028212293982506,
"num_tokens": 23717630.0,
"step": 11430
},
{
"entropy": 5.464516067504883,
"epoch": 1.0372822931785195,
"grad_norm": 1.0,
"learning_rate": 0.0004899443917923476,
"loss": 5.1428,
"mean_token_accuracy": 0.1955379366874695,
"num_tokens": 23728836.0,
"step": 11435
},
{
"entropy": 5.42895884513855,
"epoch": 1.0377358490566038,
"grad_norm": 0.96484375,
"learning_rate": 0.0004899348246571792,
"loss": 5.1186,
"mean_token_accuracy": 0.18906114101409913,
"num_tokens": 23740145.0,
"step": 11440
},
{
"entropy": 5.456284904479981,
"epoch": 1.038189404934688,
"grad_norm": 0.98046875,
"learning_rate": 0.000489925253077104,
"loss": 5.1029,
"mean_token_accuracy": 0.19079043120145797,
"num_tokens": 23751162.0,
"step": 11445
},
{
"entropy": 5.491913223266602,
"epoch": 1.038642960812772,
"grad_norm": 0.96875,
"learning_rate": 0.0004899156770523196,
"loss": 5.1081,
"mean_token_accuracy": 0.19240654408931732,
"num_tokens": 23761634.0,
"step": 11450
},
{
"entropy": 5.470863246917725,
"epoch": 1.0390965166908563,
"grad_norm": 0.921875,
"learning_rate": 0.0004899060965830244,
"loss": 5.0733,
"mean_token_accuracy": 0.198917418718338,
"num_tokens": 23772483.0,
"step": 11455
},
{
"entropy": 5.45157265663147,
"epoch": 1.0395500725689404,
"grad_norm": 1.0625,
"learning_rate": 0.0004898965116694162,
"loss": 5.0458,
"mean_token_accuracy": 0.1927415356040001,
"num_tokens": 23782298.0,
"step": 11460
},
{
"entropy": 5.468653631210327,
"epoch": 1.0400036284470247,
"grad_norm": 0.9921875,
"learning_rate": 0.0004898869223116935,
"loss": 5.1224,
"mean_token_accuracy": 0.19079854786396028,
"num_tokens": 23792492.0,
"step": 11465
},
{
"entropy": 5.510800075531006,
"epoch": 1.040457184325109,
"grad_norm": 0.9375,
"learning_rate": 0.0004898773285100546,
"loss": 5.2672,
"mean_token_accuracy": 0.184591606259346,
"num_tokens": 23802999.0,
"step": 11470
},
{
"entropy": 5.473024654388428,
"epoch": 1.040910740203193,
"grad_norm": 0.96484375,
"learning_rate": 0.0004898677302646975,
"loss": 5.0636,
"mean_token_accuracy": 0.19724741876125335,
"num_tokens": 23812541.0,
"step": 11475
},
{
"entropy": 5.47519736289978,
"epoch": 1.0413642960812772,
"grad_norm": 0.90234375,
"learning_rate": 0.0004898581275758211,
"loss": 5.1705,
"mean_token_accuracy": 0.18411127626895904,
"num_tokens": 23823244.0,
"step": 11480
},
{
"entropy": 5.45018162727356,
"epoch": 1.0418178519593615,
"grad_norm": 1.0078125,
"learning_rate": 0.0004898485204436239,
"loss": 5.0092,
"mean_token_accuracy": 0.2057377353310585,
"num_tokens": 23832346.0,
"step": 11485
},
{
"entropy": 5.522273445129395,
"epoch": 1.0422714078374455,
"grad_norm": 1.0390625,
"learning_rate": 0.0004898389088683047,
"loss": 5.1458,
"mean_token_accuracy": 0.20061220675706865,
"num_tokens": 23842554.0,
"step": 11490
},
{
"entropy": 5.405847120285034,
"epoch": 1.0427249637155298,
"grad_norm": 0.88671875,
"learning_rate": 0.0004898292928500619,
"loss": 4.9989,
"mean_token_accuracy": 0.1994738131761551,
"num_tokens": 23852938.0,
"step": 11495
},
{
"entropy": 5.394134998321533,
"epoch": 1.0431785195936139,
"grad_norm": 0.94140625,
"learning_rate": 0.0004898196723890947,
"loss": 5.0153,
"mean_token_accuracy": 0.1991426929831505,
"num_tokens": 23862367.0,
"step": 11500
},
{
"entropy": 5.448208570480347,
"epoch": 1.0436320754716981,
"grad_norm": 1.0,
"learning_rate": 0.0004898100474856019,
"loss": 5.1276,
"mean_token_accuracy": 0.19397516846656798,
"num_tokens": 23873249.0,
"step": 11505
},
{
"entropy": 5.48231987953186,
"epoch": 1.0440856313497824,
"grad_norm": 0.953125,
"learning_rate": 0.0004898004181397824,
"loss": 5.1428,
"mean_token_accuracy": 0.18477273732423782,
"num_tokens": 23883830.0,
"step": 11510
},
{
"entropy": 5.4982703685760494,
"epoch": 1.0445391872278664,
"grad_norm": 0.98828125,
"learning_rate": 0.0004897907843518358,
"loss": 5.0533,
"mean_token_accuracy": 0.19381560236215592,
"num_tokens": 23894359.0,
"step": 11515
},
{
"entropy": 5.442965316772461,
"epoch": 1.0449927431059507,
"grad_norm": 0.9375,
"learning_rate": 0.0004897811461219609,
"loss": 5.0583,
"mean_token_accuracy": 0.19391554594039917,
"num_tokens": 23903779.0,
"step": 11520
},
{
"entropy": 5.438404083251953,
"epoch": 1.0454462989840347,
"grad_norm": 0.93359375,
"learning_rate": 0.0004897715034503571,
"loss": 5.1322,
"mean_token_accuracy": 0.19373399317264556,
"num_tokens": 23914523.0,
"step": 11525
},
{
"entropy": 5.456286334991455,
"epoch": 1.045899854862119,
"grad_norm": 0.96875,
"learning_rate": 0.000489761856337224,
"loss": 5.0995,
"mean_token_accuracy": 0.19320344924926758,
"num_tokens": 23924996.0,
"step": 11530
},
{
"entropy": 5.477241516113281,
"epoch": 1.0463534107402033,
"grad_norm": 0.93359375,
"learning_rate": 0.0004897522047827609,
"loss": 5.0655,
"mean_token_accuracy": 0.19749463647603988,
"num_tokens": 23935843.0,
"step": 11535
},
{
"entropy": 5.444448661804199,
"epoch": 1.0468069666182873,
"grad_norm": 1.0625,
"learning_rate": 0.0004897425487871676,
"loss": 5.0376,
"mean_token_accuracy": 0.19542952328920365,
"num_tokens": 23945044.0,
"step": 11540
},
{
"entropy": 5.400130081176758,
"epoch": 1.0472605224963716,
"grad_norm": 0.99609375,
"learning_rate": 0.0004897328883506436,
"loss": 5.0997,
"mean_token_accuracy": 0.19553857147693635,
"num_tokens": 23955933.0,
"step": 11545
},
{
"entropy": 5.461530494689941,
"epoch": 1.0477140783744556,
"grad_norm": 1.0234375,
"learning_rate": 0.0004897232234733887,
"loss": 5.1683,
"mean_token_accuracy": 0.18461367785930632,
"num_tokens": 23966320.0,
"step": 11550
},
{
"entropy": 5.461652421951294,
"epoch": 1.04816763425254,
"grad_norm": 1.0078125,
"learning_rate": 0.0004897135541556028,
"loss": 5.061,
"mean_token_accuracy": 0.19855878949165345,
"num_tokens": 23976233.0,
"step": 11555
},
{
"entropy": 5.471328020095825,
"epoch": 1.0486211901306242,
"grad_norm": 0.85546875,
"learning_rate": 0.000489703880397486,
"loss": 5.1227,
"mean_token_accuracy": 0.1904810532927513,
"num_tokens": 23987854.0,
"step": 11560
},
{
"entropy": 5.4512513160705565,
"epoch": 1.0490747460087082,
"grad_norm": 0.96875,
"learning_rate": 0.0004896942021992382,
"loss": 5.1113,
"mean_token_accuracy": 0.1881568431854248,
"num_tokens": 23997973.0,
"step": 11565
},
{
"entropy": 5.459875631332397,
"epoch": 1.0495283018867925,
"grad_norm": 1.0234375,
"learning_rate": 0.0004896845195610595,
"loss": 5.1236,
"mean_token_accuracy": 0.1908821791410446,
"num_tokens": 24007313.0,
"step": 11570
},
{
"entropy": 5.449109792709351,
"epoch": 1.0499818577648767,
"grad_norm": 0.921875,
"learning_rate": 0.0004896748324831504,
"loss": 5.1376,
"mean_token_accuracy": 0.18894692361354828,
"num_tokens": 24018254.0,
"step": 11575
},
{
"entropy": 5.488360786437989,
"epoch": 1.0504354136429608,
"grad_norm": 0.87109375,
"learning_rate": 0.0004896651409657109,
"loss": 5.1272,
"mean_token_accuracy": 0.19029901772737504,
"num_tokens": 24029799.0,
"step": 11580
},
{
"entropy": 5.407639503479004,
"epoch": 1.050888969521045,
"grad_norm": 0.95703125,
"learning_rate": 0.0004896554450089417,
"loss": 5.0307,
"mean_token_accuracy": 0.20030184984207153,
"num_tokens": 24040972.0,
"step": 11585
},
{
"entropy": 5.443329524993897,
"epoch": 1.051342525399129,
"grad_norm": 0.95703125,
"learning_rate": 0.0004896457446130431,
"loss": 4.9667,
"mean_token_accuracy": 0.20220689922571183,
"num_tokens": 24050781.0,
"step": 11590
},
{
"entropy": 5.429394340515136,
"epoch": 1.0517960812772134,
"grad_norm": 1.03125,
"learning_rate": 0.0004896360397782159,
"loss": 5.0991,
"mean_token_accuracy": 0.19227786213159562,
"num_tokens": 24060184.0,
"step": 11595
},
{
"entropy": 5.489581871032715,
"epoch": 1.0522496371552976,
"grad_norm": 1.0625,
"learning_rate": 0.0004896263305046607,
"loss": 5.1549,
"mean_token_accuracy": 0.19510595351457596,
"num_tokens": 24070777.0,
"step": 11600
},
{
"entropy": 5.507211112976075,
"epoch": 1.0527031930333817,
"grad_norm": 1.0234375,
"learning_rate": 0.0004896166167925782,
"loss": 5.1887,
"mean_token_accuracy": 0.1927416890859604,
"num_tokens": 24080878.0,
"step": 11605
},
{
"entropy": 5.492761325836182,
"epoch": 1.053156748911466,
"grad_norm": 0.98828125,
"learning_rate": 0.0004896068986421693,
"loss": 5.137,
"mean_token_accuracy": 0.20071599036455154,
"num_tokens": 24090038.0,
"step": 11610
},
{
"entropy": 5.528735876083374,
"epoch": 1.05361030478955,
"grad_norm": 0.88671875,
"learning_rate": 0.0004895971760536351,
"loss": 5.1714,
"mean_token_accuracy": 0.19025161862373352,
"num_tokens": 24100425.0,
"step": 11615
},
{
"entropy": 5.519040775299072,
"epoch": 1.0540638606676342,
"grad_norm": 0.96875,
"learning_rate": 0.0004895874490271768,
"loss": 5.2288,
"mean_token_accuracy": 0.18544750064611434,
"num_tokens": 24112182.0,
"step": 11620
},
{
"entropy": 5.479070711135864,
"epoch": 1.0545174165457185,
"grad_norm": 0.98046875,
"learning_rate": 0.0004895777175629952,
"loss": 5.0768,
"mean_token_accuracy": 0.19402081668376922,
"num_tokens": 24122182.0,
"step": 11625
},
{
"entropy": 5.512710905075073,
"epoch": 1.0549709724238026,
"grad_norm": 1.046875,
"learning_rate": 0.0004895679816612917,
"loss": 5.0755,
"mean_token_accuracy": 0.1985362336039543,
"num_tokens": 24131930.0,
"step": 11630
},
{
"entropy": 5.499684190750122,
"epoch": 1.0554245283018868,
"grad_norm": 0.98828125,
"learning_rate": 0.0004895582413222678,
"loss": 5.0746,
"mean_token_accuracy": 0.20388585031032563,
"num_tokens": 24141575.0,
"step": 11635
},
{
"entropy": 5.482347249984741,
"epoch": 1.0558780841799709,
"grad_norm": 0.98046875,
"learning_rate": 0.0004895484965461247,
"loss": 5.2114,
"mean_token_accuracy": 0.1858080580830574,
"num_tokens": 24151720.0,
"step": 11640
},
{
"entropy": 5.427149391174316,
"epoch": 1.0563316400580551,
"grad_norm": 1.0390625,
"learning_rate": 0.000489538747333064,
"loss": 5.1075,
"mean_token_accuracy": 0.19254984259605407,
"num_tokens": 24161881.0,
"step": 11645
},
{
"entropy": 5.449149322509766,
"epoch": 1.0567851959361394,
"grad_norm": 1.0546875,
"learning_rate": 0.0004895289936832874,
"loss": 5.0637,
"mean_token_accuracy": 0.20020053237676622,
"num_tokens": 24171474.0,
"step": 11650
},
{
"entropy": 5.552807998657227,
"epoch": 1.0572387518142234,
"grad_norm": 0.90234375,
"learning_rate": 0.0004895192355969966,
"loss": 5.1718,
"mean_token_accuracy": 0.19173015356063844,
"num_tokens": 24182186.0,
"step": 11655
},
{
"entropy": 5.49303035736084,
"epoch": 1.0576923076923077,
"grad_norm": 0.90234375,
"learning_rate": 0.0004895094730743932,
"loss": 5.0972,
"mean_token_accuracy": 0.20089954733848572,
"num_tokens": 24192318.0,
"step": 11660
},
{
"entropy": 5.480742073059082,
"epoch": 1.058145863570392,
"grad_norm": 0.95703125,
"learning_rate": 0.0004894997061156793,
"loss": 5.1583,
"mean_token_accuracy": 0.19043267518281937,
"num_tokens": 24202918.0,
"step": 11665
},
{
"entropy": 5.426549005508423,
"epoch": 1.058599419448476,
"grad_norm": 1.0,
"learning_rate": 0.0004894899347210567,
"loss": 5.0545,
"mean_token_accuracy": 0.19537004977464675,
"num_tokens": 24213218.0,
"step": 11670
},
{
"entropy": 5.448081922531128,
"epoch": 1.0590529753265603,
"grad_norm": 0.91015625,
"learning_rate": 0.0004894801588907278,
"loss": 5.0414,
"mean_token_accuracy": 0.19819533377885817,
"num_tokens": 24223229.0,
"step": 11675
},
{
"entropy": 5.508687114715576,
"epoch": 1.0595065312046443,
"grad_norm": 0.984375,
"learning_rate": 0.0004894703786248944,
"loss": 5.1189,
"mean_token_accuracy": 0.19095162451267242,
"num_tokens": 24233637.0,
"step": 11680
},
{
"entropy": 5.4921238899230955,
"epoch": 1.0599600870827286,
"grad_norm": 1.03125,
"learning_rate": 0.000489460593923759,
"loss": 5.149,
"mean_token_accuracy": 0.19814313352108,
"num_tokens": 24244251.0,
"step": 11685
},
{
"entropy": 5.471730422973633,
"epoch": 1.0604136429608129,
"grad_norm": 0.953125,
"learning_rate": 0.0004894508047875237,
"loss": 5.0565,
"mean_token_accuracy": 0.19785570055246354,
"num_tokens": 24254380.0,
"step": 11690
},
{
"entropy": 5.340200662612915,
"epoch": 1.060867198838897,
"grad_norm": 0.9140625,
"learning_rate": 0.0004894410112163913,
"loss": 4.9937,
"mean_token_accuracy": 0.19913735389709472,
"num_tokens": 24264841.0,
"step": 11695
},
{
"entropy": 5.391559076309204,
"epoch": 1.0613207547169812,
"grad_norm": 1.0390625,
"learning_rate": 0.0004894312132105641,
"loss": 5.0239,
"mean_token_accuracy": 0.19785647094249725,
"num_tokens": 24274383.0,
"step": 11700
},
{
"entropy": 5.534787845611572,
"epoch": 1.0617743105950652,
"grad_norm": 1.0078125,
"learning_rate": 0.0004894214107702447,
"loss": 5.11,
"mean_token_accuracy": 0.19086095243692397,
"num_tokens": 24284140.0,
"step": 11705
},
{
"entropy": 5.560810375213623,
"epoch": 1.0622278664731495,
"grad_norm": 1.0546875,
"learning_rate": 0.000489411603895636,
"loss": 5.1878,
"mean_token_accuracy": 0.1921544551849365,
"num_tokens": 24295044.0,
"step": 11710
},
{
"entropy": 5.451190757751465,
"epoch": 1.0626814223512338,
"grad_norm": 1.015625,
"learning_rate": 0.0004894017925869406,
"loss": 5.0975,
"mean_token_accuracy": 0.19553182721138002,
"num_tokens": 24305080.0,
"step": 11715
},
{
"entropy": 5.466040515899659,
"epoch": 1.0631349782293178,
"grad_norm": 0.9765625,
"learning_rate": 0.0004893919768443615,
"loss": 5.1308,
"mean_token_accuracy": 0.19310673475265502,
"num_tokens": 24316306.0,
"step": 11720
},
{
"entropy": 5.445856237411499,
"epoch": 1.063588534107402,
"grad_norm": 0.9921875,
"learning_rate": 0.0004893821566681017,
"loss": 5.1232,
"mean_token_accuracy": 0.19693950265645982,
"num_tokens": 24326740.0,
"step": 11725
},
{
"entropy": 5.489151525497436,
"epoch": 1.0640420899854863,
"grad_norm": 1.015625,
"learning_rate": 0.0004893723320583644,
"loss": 5.162,
"mean_token_accuracy": 0.188465815782547,
"num_tokens": 24337621.0,
"step": 11730
},
{
"entropy": 5.386200571060181,
"epoch": 1.0644956458635704,
"grad_norm": 0.96484375,
"learning_rate": 0.0004893625030153525,
"loss": 4.9959,
"mean_token_accuracy": 0.20463671237230302,
"num_tokens": 24347732.0,
"step": 11735
},
{
"entropy": 5.40708589553833,
"epoch": 1.0649492017416546,
"grad_norm": 0.9765625,
"learning_rate": 0.0004893526695392694,
"loss": 5.0112,
"mean_token_accuracy": 0.20490798354148865,
"num_tokens": 24359173.0,
"step": 11740
},
{
"entropy": 5.431822299957275,
"epoch": 1.0654027576197387,
"grad_norm": 0.95703125,
"learning_rate": 0.0004893428316303185,
"loss": 5.0139,
"mean_token_accuracy": 0.20014695078134537,
"num_tokens": 24369465.0,
"step": 11745
},
{
"entropy": 5.387581729888916,
"epoch": 1.065856313497823,
"grad_norm": 0.94140625,
"learning_rate": 0.0004893329892887031,
"loss": 5.0756,
"mean_token_accuracy": 0.19761465936899186,
"num_tokens": 24378899.0,
"step": 11750
},
{
"entropy": 5.4371435165405275,
"epoch": 1.0663098693759072,
"grad_norm": 1.0078125,
"learning_rate": 0.0004893231425146271,
"loss": 5.1867,
"mean_token_accuracy": 0.18819210082292556,
"num_tokens": 24388919.0,
"step": 11755
},
{
"entropy": 5.407600069046021,
"epoch": 1.0667634252539913,
"grad_norm": 0.98828125,
"learning_rate": 0.0004893132913082937,
"loss": 5.0453,
"mean_token_accuracy": 0.19777143746614456,
"num_tokens": 24399632.0,
"step": 11760
},
{
"entropy": 5.466056203842163,
"epoch": 1.0672169811320755,
"grad_norm": 0.91015625,
"learning_rate": 0.0004893034356699069,
"loss": 5.0721,
"mean_token_accuracy": 0.19814738631248474,
"num_tokens": 24410817.0,
"step": 11765
},
{
"entropy": 5.454527950286865,
"epoch": 1.0676705370101596,
"grad_norm": 0.98046875,
"learning_rate": 0.0004892935755996705,
"loss": 5.0955,
"mean_token_accuracy": 0.19848891198635102,
"num_tokens": 24421058.0,
"step": 11770
},
{
"entropy": 5.412892913818359,
"epoch": 1.0681240928882438,
"grad_norm": 1.015625,
"learning_rate": 0.0004892837110977883,
"loss": 5.0184,
"mean_token_accuracy": 0.1985347628593445,
"num_tokens": 24431522.0,
"step": 11775
},
{
"entropy": 5.420146083831787,
"epoch": 1.068577648766328,
"grad_norm": 0.89453125,
"learning_rate": 0.0004892738421644644,
"loss": 5.0563,
"mean_token_accuracy": 0.19299584180116652,
"num_tokens": 24441746.0,
"step": 11780
},
{
"entropy": 5.373950052261352,
"epoch": 1.0690312046444121,
"grad_norm": 1.015625,
"learning_rate": 0.0004892639687999027,
"loss": 4.9627,
"mean_token_accuracy": 0.206170853972435,
"num_tokens": 24451916.0,
"step": 11785
},
{
"entropy": 5.455459642410278,
"epoch": 1.0694847605224964,
"grad_norm": 0.99609375,
"learning_rate": 0.0004892540910043076,
"loss": 5.0879,
"mean_token_accuracy": 0.19918798059225082,
"num_tokens": 24461935.0,
"step": 11790
},
{
"entropy": 5.435303163528443,
"epoch": 1.0699383164005805,
"grad_norm": 1.03125,
"learning_rate": 0.0004892442087778833,
"loss": 5.0002,
"mean_token_accuracy": 0.19787367433309555,
"num_tokens": 24472437.0,
"step": 11795
},
{
"entropy": 5.431952667236328,
"epoch": 1.0703918722786647,
"grad_norm": 1.046875,
"learning_rate": 0.0004892343221208342,
"loss": 5.1029,
"mean_token_accuracy": 0.19617521166801452,
"num_tokens": 24481726.0,
"step": 11800
},
{
"entropy": 5.498169612884522,
"epoch": 1.070845428156749,
"grad_norm": 0.93359375,
"learning_rate": 0.0004892244310333646,
"loss": 5.184,
"mean_token_accuracy": 0.19032939225435258,
"num_tokens": 24492340.0,
"step": 11805
},
{
"entropy": 5.476718950271606,
"epoch": 1.071298984034833,
"grad_norm": 1.0,
"learning_rate": 0.0004892145355156793,
"loss": 5.05,
"mean_token_accuracy": 0.19708126336336135,
"num_tokens": 24502138.0,
"step": 11810
},
{
"entropy": 5.4611780643463135,
"epoch": 1.0717525399129173,
"grad_norm": 0.91796875,
"learning_rate": 0.0004892046355679829,
"loss": 5.1045,
"mean_token_accuracy": 0.19246915280818938,
"num_tokens": 24512723.0,
"step": 11815
},
{
"entropy": 5.510536956787109,
"epoch": 1.0722060957910013,
"grad_norm": 0.97265625,
"learning_rate": 0.0004891947311904799,
"loss": 5.1444,
"mean_token_accuracy": 0.1910214379429817,
"num_tokens": 24523766.0,
"step": 11820
},
{
"entropy": 5.4847740650177,
"epoch": 1.0726596516690856,
"grad_norm": 0.97265625,
"learning_rate": 0.0004891848223833753,
"loss": 5.1062,
"mean_token_accuracy": 0.19305452108383178,
"num_tokens": 24534571.0,
"step": 11825
},
{
"entropy": 5.456302833557129,
"epoch": 1.0731132075471699,
"grad_norm": 1.015625,
"learning_rate": 0.000489174909146874,
"loss": 5.1088,
"mean_token_accuracy": 0.19871936440467836,
"num_tokens": 24545265.0,
"step": 11830
},
{
"entropy": 5.524204874038697,
"epoch": 1.073566763425254,
"grad_norm": 1.0625,
"learning_rate": 0.0004891649914811811,
"loss": 5.1979,
"mean_token_accuracy": 0.18889098316431047,
"num_tokens": 24555524.0,
"step": 11835
},
{
"entropy": 5.410844326019287,
"epoch": 1.0740203193033382,
"grad_norm": 0.98046875,
"learning_rate": 0.0004891550693865014,
"loss": 5.052,
"mean_token_accuracy": 0.20847347378730774,
"num_tokens": 24565420.0,
"step": 11840
},
{
"entropy": 5.468236207962036,
"epoch": 1.0744738751814222,
"grad_norm": 0.93359375,
"learning_rate": 0.0004891451428630405,
"loss": 5.0912,
"mean_token_accuracy": 0.19379322081804276,
"num_tokens": 24576794.0,
"step": 11845
},
{
"entropy": 5.4929125785827635,
"epoch": 1.0749274310595065,
"grad_norm": 0.921875,
"learning_rate": 0.0004891352119110033,
"loss": 5.1068,
"mean_token_accuracy": 0.19399791359901428,
"num_tokens": 24586320.0,
"step": 11850
},
{
"entropy": 5.448437881469727,
"epoch": 1.0753809869375908,
"grad_norm": 0.98828125,
"learning_rate": 0.0004891252765305954,
"loss": 5.0875,
"mean_token_accuracy": 0.20086531192064286,
"num_tokens": 24596524.0,
"step": 11855
},
{
"entropy": 5.4746064186096195,
"epoch": 1.0758345428156748,
"grad_norm": 0.9921875,
"learning_rate": 0.0004891153367220222,
"loss": 5.0173,
"mean_token_accuracy": 0.2089245080947876,
"num_tokens": 24605877.0,
"step": 11860
},
{
"entropy": 5.502732610702514,
"epoch": 1.076288098693759,
"grad_norm": 0.93359375,
"learning_rate": 0.0004891053924854894,
"loss": 5.2305,
"mean_token_accuracy": 0.18587556779384612,
"num_tokens": 24616719.0,
"step": 11865
},
{
"entropy": 5.43260178565979,
"epoch": 1.0767416545718433,
"grad_norm": 1.015625,
"learning_rate": 0.0004890954438212024,
"loss": 4.9843,
"mean_token_accuracy": 0.20658824294805528,
"num_tokens": 24626882.0,
"step": 11870
},
{
"entropy": 5.486688423156738,
"epoch": 1.0771952104499274,
"grad_norm": 0.984375,
"learning_rate": 0.0004890854907293671,
"loss": 5.0936,
"mean_token_accuracy": 0.1938313826918602,
"num_tokens": 24637884.0,
"step": 11875
},
{
"entropy": 5.4502464771270756,
"epoch": 1.0776487663280117,
"grad_norm": 0.96875,
"learning_rate": 0.0004890755332101893,
"loss": 5.0895,
"mean_token_accuracy": 0.19480091631412505,
"num_tokens": 24647389.0,
"step": 11880
},
{
"entropy": 5.448260498046875,
"epoch": 1.0781023222060957,
"grad_norm": 0.890625,
"learning_rate": 0.0004890655712638749,
"loss": 5.0976,
"mean_token_accuracy": 0.19201536923646928,
"num_tokens": 24657595.0,
"step": 11885
},
{
"entropy": 5.448187303543091,
"epoch": 1.07855587808418,
"grad_norm": 1.171875,
"learning_rate": 0.00048905560489063,
"loss": 5.015,
"mean_token_accuracy": 0.19760590493679048,
"num_tokens": 24667683.0,
"step": 11890
},
{
"entropy": 5.492201042175293,
"epoch": 1.0790094339622642,
"grad_norm": 0.96484375,
"learning_rate": 0.0004890456340906605,
"loss": 5.1742,
"mean_token_accuracy": 0.1939292684197426,
"num_tokens": 24678041.0,
"step": 11895
},
{
"entropy": 5.450703811645508,
"epoch": 1.0794629898403483,
"grad_norm": 1.0078125,
"learning_rate": 0.0004890356588641729,
"loss": 5.1517,
"mean_token_accuracy": 0.19592384546995162,
"num_tokens": 24688569.0,
"step": 11900
},
{
"entropy": 5.494032478332519,
"epoch": 1.0799165457184325,
"grad_norm": 0.94921875,
"learning_rate": 0.0004890256792113733,
"loss": 5.1069,
"mean_token_accuracy": 0.19837442487478257,
"num_tokens": 24699913.0,
"step": 11905
},
{
"entropy": 5.5294829368591305,
"epoch": 1.0803701015965168,
"grad_norm": 0.890625,
"learning_rate": 0.0004890156951324681,
"loss": 5.1323,
"mean_token_accuracy": 0.1929474338889122,
"num_tokens": 24710682.0,
"step": 11910
},
{
"entropy": 5.526803922653198,
"epoch": 1.0808236574746009,
"grad_norm": 1.0390625,
"learning_rate": 0.0004890057066276636,
"loss": 5.2058,
"mean_token_accuracy": 0.18813426196575164,
"num_tokens": 24721087.0,
"step": 11915
},
{
"entropy": 5.43285641670227,
"epoch": 1.0812772133526851,
"grad_norm": 0.91015625,
"learning_rate": 0.0004889957136971668,
"loss": 5.0328,
"mean_token_accuracy": 0.20273849219083787,
"num_tokens": 24731323.0,
"step": 11920
},
{
"entropy": 5.442595529556274,
"epoch": 1.0817307692307692,
"grad_norm": 0.9765625,
"learning_rate": 0.000488985716341184,
"loss": 5.097,
"mean_token_accuracy": 0.19515804946422577,
"num_tokens": 24742559.0,
"step": 11925
},
{
"entropy": 5.497393083572388,
"epoch": 1.0821843251088534,
"grad_norm": 0.96484375,
"learning_rate": 0.0004889757145599222,
"loss": 5.1055,
"mean_token_accuracy": 0.19816650450229645,
"num_tokens": 24752307.0,
"step": 11930
},
{
"entropy": 5.361609315872192,
"epoch": 1.0826378809869377,
"grad_norm": 0.96484375,
"learning_rate": 0.000488965708353588,
"loss": 5.0406,
"mean_token_accuracy": 0.1924867585301399,
"num_tokens": 24762792.0,
"step": 11935
},
{
"entropy": 5.508419227600098,
"epoch": 1.0830914368650217,
"grad_norm": 0.91796875,
"learning_rate": 0.0004889556977223883,
"loss": 5.1613,
"mean_token_accuracy": 0.19042742848396302,
"num_tokens": 24774377.0,
"step": 11940
},
{
"entropy": 5.484889316558838,
"epoch": 1.083544992743106,
"grad_norm": 0.875,
"learning_rate": 0.0004889456826665304,
"loss": 5.0786,
"mean_token_accuracy": 0.195466311275959,
"num_tokens": 24785919.0,
"step": 11945
},
{
"entropy": 5.465774250030518,
"epoch": 1.08399854862119,
"grad_norm": 0.984375,
"learning_rate": 0.0004889356631862211,
"loss": 5.1187,
"mean_token_accuracy": 0.18686763644218446,
"num_tokens": 24796594.0,
"step": 11950
},
{
"entropy": 5.453439331054687,
"epoch": 1.0844521044992743,
"grad_norm": 0.9375,
"learning_rate": 0.000488925639281668,
"loss": 5.1122,
"mean_token_accuracy": 0.1970463752746582,
"num_tokens": 24807322.0,
"step": 11955
},
{
"entropy": 5.438632583618164,
"epoch": 1.0849056603773586,
"grad_norm": 0.921875,
"learning_rate": 0.0004889156109530781,
"loss": 4.999,
"mean_token_accuracy": 0.1978117287158966,
"num_tokens": 24817593.0,
"step": 11960
},
{
"entropy": 5.442756128311157,
"epoch": 1.0853592162554426,
"grad_norm": 1.0,
"learning_rate": 0.0004889055782006588,
"loss": 5.0073,
"mean_token_accuracy": 0.2037275940179825,
"num_tokens": 24827577.0,
"step": 11965
},
{
"entropy": 5.4089408874511715,
"epoch": 1.085812772133527,
"grad_norm": 1.0546875,
"learning_rate": 0.0004888955410246176,
"loss": 5.0577,
"mean_token_accuracy": 0.19720772951841353,
"num_tokens": 24838148.0,
"step": 11970
},
{
"entropy": 5.425878286361694,
"epoch": 1.086266328011611,
"grad_norm": 0.9375,
"learning_rate": 0.0004888854994251621,
"loss": 5.0117,
"mean_token_accuracy": 0.19913591742515563,
"num_tokens": 24848451.0,
"step": 11975
},
{
"entropy": 5.492869567871094,
"epoch": 1.0867198838896952,
"grad_norm": 0.89453125,
"learning_rate": 0.0004888754534025,
"loss": 5.1374,
"mean_token_accuracy": 0.18826611191034318,
"num_tokens": 24859619.0,
"step": 11980
},
{
"entropy": 5.474278116226197,
"epoch": 1.0871734397677795,
"grad_norm": 0.95703125,
"learning_rate": 0.0004888654029568391,
"loss": 5.1732,
"mean_token_accuracy": 0.18208261132240294,
"num_tokens": 24869863.0,
"step": 11985
},
{
"entropy": 5.450355195999146,
"epoch": 1.0876269956458635,
"grad_norm": 0.94921875,
"learning_rate": 0.0004888553480883871,
"loss": 5.1074,
"mean_token_accuracy": 0.1931120127439499,
"num_tokens": 24880524.0,
"step": 11990
},
{
"entropy": 5.3892419815063475,
"epoch": 1.0880805515239478,
"grad_norm": 0.8984375,
"learning_rate": 0.000488845288797352,
"loss": 5.076,
"mean_token_accuracy": 0.19475484490394593,
"num_tokens": 24891514.0,
"step": 11995
},
{
"entropy": 5.428224277496338,
"epoch": 1.0885341074020318,
"grad_norm": 1.0234375,
"learning_rate": 0.0004888352250839418,
"loss": 5.0443,
"mean_token_accuracy": 0.1945518046617508,
"num_tokens": 24900842.0,
"step": 12000
},
{
"epoch": 1.0885341074020318,
"eval_entropy": 5.255034273674045,
"eval_loss": 5.174998760223389,
"eval_mean_token_accuracy": 0.20042603093608202,
"eval_num_tokens": 24900842.0,
"eval_runtime": 20.8388,
"eval_samples_per_second": 1696.832,
"eval_steps_per_second": 212.104,
"step": 12000
}
],
"logging_steps": 5,
"max_steps": 110230,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 5405053433794560.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink