fpadovani/isl-latn-10mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
2df575cab3670f026a6e91dd5259c4bb2bd511dc
isl-latn-10mb-100mb_seed3407/checkpoint-12000/trainer_state.json
ModelHub XC 2df575cab3 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/isl-latn-10mb-100mb_seed3407
Source: Original Platform
2026-08-19 02:59:19 +08:00

24079 lines
662 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.1862396204033214,
"eval_steps": 3000,
"global_step": 12000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.742559432983398,
"epoch": 0.0004942665085013839,
"grad_norm": 5.34375,
"learning_rate": 2e-06,
"loss": 10.7918,
"mean_token_accuracy": 0.0,
"num_tokens": 11289.0,
"step": 5
},
{
"entropy": 10.742565155029297,
"epoch": 0.0009885330170027679,
"grad_norm": 5.21875,
"learning_rate": 4.5e-06,
"loss": 10.7677,
"mean_token_accuracy": 0.00010660981060937048,
"num_tokens": 21541.0,
"step": 10
},
{
"entropy": 10.74260196685791,
"epoch": 0.0014827995255041518,
"grad_norm": 5.28125,
"learning_rate": 7e-06,
"loss": 10.7469,
"mean_token_accuracy": 0.0,
"num_tokens": 32257.0,
"step": 15
},
{
"entropy": 10.742596530914307,
"epoch": 0.0019770660340055358,
"grad_norm": 5.09375,
"learning_rate": 9.5e-06,
"loss": 10.7082,
"mean_token_accuracy": 0.00043735416838899257,
"num_tokens": 44176.0,
"step": 20
},
{
"entropy": 10.742553424835204,
"epoch": 0.0024713325425069197,
"grad_norm": 4.4375,
"learning_rate": 1.2e-05,
"loss": 10.6063,
"mean_token_accuracy": 0.005003847437910736,
"num_tokens": 56115.0,
"step": 25
},
{
"entropy": 10.742230701446534,
"epoch": 0.0029655990510083037,
"grad_norm": 4.0,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.4933,
"mean_token_accuracy": 0.02219230718910694,
"num_tokens": 68828.0,
"step": 30
},
{
"entropy": 10.74111728668213,
"epoch": 0.0034598655595096876,
"grad_norm": 3.390625,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.3232,
"mean_token_accuracy": 0.037596833519637586,
"num_tokens": 80097.0,
"step": 35
},
{
"entropy": 10.73831787109375,
"epoch": 0.0039541320680110716,
"grad_norm": 2.546875,
"learning_rate": 1.95e-05,
"loss": 10.1706,
"mean_token_accuracy": 0.042161832191050054,
"num_tokens": 91510.0,
"step": 40
},
{
"entropy": 10.734669494628907,
"epoch": 0.004448398576512456,
"grad_norm": 2.25,
"learning_rate": 2.2e-05,
"loss": 10.0532,
"mean_token_accuracy": 0.04492984525859356,
"num_tokens": 103443.0,
"step": 45
},
{
"entropy": 10.731356620788574,
"epoch": 0.0049426650850138395,
"grad_norm": 2.171875,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.9944,
"mean_token_accuracy": 0.04110746793448925,
"num_tokens": 114629.0,
"step": 50
},
{
"entropy": 10.72984676361084,
"epoch": 0.005436931593515224,
"grad_norm": 1.984375,
"learning_rate": 2.7e-05,
"loss": 9.9354,
"mean_token_accuracy": 0.0402316827327013,
"num_tokens": 125012.0,
"step": 55
},
{
"entropy": 10.72827968597412,
"epoch": 0.005931198102016607,
"grad_norm": 1.96875,
"learning_rate": 2.95e-05,
"loss": 9.8745,
"mean_token_accuracy": 0.0415823869407177,
"num_tokens": 135834.0,
"step": 60
},
{
"entropy": 10.726681423187255,
"epoch": 0.006425464610517992,
"grad_norm": 1.90625,
"learning_rate": 3.2e-05,
"loss": 9.8094,
"mean_token_accuracy": 0.0378471277654171,
"num_tokens": 147413.0,
"step": 65
},
{
"entropy": 10.725395107269287,
"epoch": 0.006919731119019375,
"grad_norm": 2.0,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.7766,
"mean_token_accuracy": 0.04534405171871185,
"num_tokens": 158046.0,
"step": 70
},
{
"entropy": 10.722041702270507,
"epoch": 0.00741399762752076,
"grad_norm": 2.0,
"learning_rate": 3.7e-05,
"loss": 9.7003,
"mean_token_accuracy": 0.04153196755796671,
"num_tokens": 168564.0,
"step": 75
},
{
"entropy": 10.717565536499023,
"epoch": 0.007908264136022143,
"grad_norm": 2.0,
"learning_rate": 3.95e-05,
"loss": 9.6649,
"mean_token_accuracy": 0.040395910106599334,
"num_tokens": 181168.0,
"step": 80
},
{
"entropy": 10.712557983398437,
"epoch": 0.008402530644523527,
"grad_norm": 1.984375,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.5927,
"mean_token_accuracy": 0.040117064118385316,
"num_tokens": 192666.0,
"step": 85
},
{
"entropy": 10.706313228607177,
"epoch": 0.008896797153024912,
"grad_norm": 1.9609375,
"learning_rate": 4.45e-05,
"loss": 9.5324,
"mean_token_accuracy": 0.04307716451585293,
"num_tokens": 203688.0,
"step": 90
},
{
"entropy": 10.696890449523925,
"epoch": 0.009391063661526295,
"grad_norm": 1.9609375,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.4292,
"mean_token_accuracy": 0.04244718514382839,
"num_tokens": 214399.0,
"step": 95
},
{
"entropy": 10.68424196243286,
"epoch": 0.009885330170027679,
"grad_norm": 1.8046875,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.4064,
"mean_token_accuracy": 0.03980784080922604,
"num_tokens": 227549.0,
"step": 100
},
{
"entropy": 10.669907379150391,
"epoch": 0.010379596678529062,
"grad_norm": 2.09375,
"learning_rate": 5.2e-05,
"loss": 9.2501,
"mean_token_accuracy": 0.05166649520397186,
"num_tokens": 239612.0,
"step": 105
},
{
"entropy": 10.647725677490234,
"epoch": 0.010873863187030448,
"grad_norm": 1.890625,
"learning_rate": 5.45e-05,
"loss": 9.1891,
"mean_token_accuracy": 0.05595755912363529,
"num_tokens": 251163.0,
"step": 110
},
{
"entropy": 10.614391994476318,
"epoch": 0.011368129695531831,
"grad_norm": 1.984375,
"learning_rate": 5.7e-05,
"loss": 9.0675,
"mean_token_accuracy": 0.0543642945587635,
"num_tokens": 263995.0,
"step": 115
},
{
"entropy": 10.580703353881836,
"epoch": 0.011862396204033215,
"grad_norm": 1.765625,
"learning_rate": 5.9499999999999996e-05,
"loss": 9.0334,
"mean_token_accuracy": 0.05128365270793438,
"num_tokens": 275103.0,
"step": 120
},
{
"entropy": 10.53465690612793,
"epoch": 0.012356662712534598,
"grad_norm": 1.6953125,
"learning_rate": 6.2e-05,
"loss": 8.9026,
"mean_token_accuracy": 0.05212775617837906,
"num_tokens": 286838.0,
"step": 125
},
{
"entropy": 10.470762634277344,
"epoch": 0.012850929221035983,
"grad_norm": 1.6796875,
"learning_rate": 6.450000000000001e-05,
"loss": 8.8066,
"mean_token_accuracy": 0.05225841067731381,
"num_tokens": 298750.0,
"step": 130
},
{
"entropy": 10.38624849319458,
"epoch": 0.013345195729537367,
"grad_norm": 1.5703125,
"learning_rate": 6.7e-05,
"loss": 8.6142,
"mean_token_accuracy": 0.06268712393939495,
"num_tokens": 310451.0,
"step": 135
},
{
"entropy": 10.291066646575928,
"epoch": 0.01383946223803875,
"grad_norm": 1.4921875,
"learning_rate": 6.950000000000001e-05,
"loss": 8.5675,
"mean_token_accuracy": 0.05591914691030979,
"num_tokens": 321232.0,
"step": 140
},
{
"entropy": 10.217416381835937,
"epoch": 0.014333728746540134,
"grad_norm": 1.484375,
"learning_rate": 7.2e-05,
"loss": 8.426,
"mean_token_accuracy": 0.06510277166962623,
"num_tokens": 332531.0,
"step": 145
},
{
"entropy": 10.091275501251221,
"epoch": 0.01482799525504152,
"grad_norm": 1.4296875,
"learning_rate": 7.45e-05,
"loss": 8.3371,
"mean_token_accuracy": 0.060455797240138054,
"num_tokens": 346002.0,
"step": 150
},
{
"entropy": 9.966124248504638,
"epoch": 0.015322261763542903,
"grad_norm": 1.3046875,
"learning_rate": 7.7e-05,
"loss": 8.1965,
"mean_token_accuracy": 0.05812167599797249,
"num_tokens": 356691.0,
"step": 155
},
{
"entropy": 9.774528980255127,
"epoch": 0.015816528272044286,
"grad_norm": 1.1796875,
"learning_rate": 7.950000000000001e-05,
"loss": 8.1137,
"mean_token_accuracy": 0.06779846139252185,
"num_tokens": 368749.0,
"step": 160
},
{
"entropy": 9.564791774749756,
"epoch": 0.01631079478054567,
"grad_norm": 1.1171875,
"learning_rate": 8.2e-05,
"loss": 7.9752,
"mean_token_accuracy": 0.06878971830010414,
"num_tokens": 379484.0,
"step": 165
},
{
"entropy": 9.371025466918946,
"epoch": 0.016805061289047053,
"grad_norm": 0.953125,
"learning_rate": 8.450000000000001e-05,
"loss": 7.9252,
"mean_token_accuracy": 0.065206453576684,
"num_tokens": 389751.0,
"step": 170
},
{
"entropy": 9.126683902740478,
"epoch": 0.017299327797548437,
"grad_norm": 0.91015625,
"learning_rate": 8.7e-05,
"loss": 7.8734,
"mean_token_accuracy": 0.06619854085147381,
"num_tokens": 400771.0,
"step": 175
},
{
"entropy": 8.916962146759033,
"epoch": 0.017793594306049824,
"grad_norm": 0.7734375,
"learning_rate": 8.95e-05,
"loss": 7.8145,
"mean_token_accuracy": 0.07182282060384751,
"num_tokens": 413575.0,
"step": 180
},
{
"entropy": 8.716337394714355,
"epoch": 0.018287860814551207,
"grad_norm": 1.0390625,
"learning_rate": 9.2e-05,
"loss": 7.7284,
"mean_token_accuracy": 0.07276704236865043,
"num_tokens": 424825.0,
"step": 185
},
{
"entropy": 8.58464298248291,
"epoch": 0.01878212732305259,
"grad_norm": 0.85546875,
"learning_rate": 9.45e-05,
"loss": 7.7373,
"mean_token_accuracy": 0.07814876064658165,
"num_tokens": 435610.0,
"step": 190
},
{
"entropy": 8.413296604156494,
"epoch": 0.019276393831553974,
"grad_norm": 0.84375,
"learning_rate": 9.7e-05,
"loss": 7.6992,
"mean_token_accuracy": 0.07458564341068268,
"num_tokens": 445855.0,
"step": 195
},
{
"entropy": 8.39066514968872,
"epoch": 0.019770660340055358,
"grad_norm": 0.7578125,
"learning_rate": 9.95e-05,
"loss": 7.5975,
"mean_token_accuracy": 0.07509246952831745,
"num_tokens": 456895.0,
"step": 200
},
{
"entropy": 8.318842697143555,
"epoch": 0.02026492684855674,
"grad_norm": 1.0546875,
"learning_rate": 0.000102,
"loss": 7.6975,
"mean_token_accuracy": 0.07416085563600064,
"num_tokens": 467446.0,
"step": 205
},
{
"entropy": 8.206272411346436,
"epoch": 0.020759193357058125,
"grad_norm": 0.765625,
"learning_rate": 0.00010449999999999999,
"loss": 7.6073,
"mean_token_accuracy": 0.07490024492144584,
"num_tokens": 478988.0,
"step": 210
},
{
"entropy": 8.203931045532226,
"epoch": 0.02125345986555951,
"grad_norm": 0.87109375,
"learning_rate": 0.000107,
"loss": 7.6028,
"mean_token_accuracy": 0.0804927557706833,
"num_tokens": 491053.0,
"step": 215
},
{
"entropy": 8.194980716705322,
"epoch": 0.021747726374060895,
"grad_norm": 1.0,
"learning_rate": 0.0001095,
"loss": 7.6693,
"mean_token_accuracy": 0.07337137088179588,
"num_tokens": 502279.0,
"step": 220
},
{
"entropy": 8.176432991027832,
"epoch": 0.02224199288256228,
"grad_norm": 0.87109375,
"learning_rate": 0.000112,
"loss": 7.5236,
"mean_token_accuracy": 0.08306471854448319,
"num_tokens": 513470.0,
"step": 225
},
{
"entropy": 8.073592710494996,
"epoch": 0.022736259391063662,
"grad_norm": 0.88671875,
"learning_rate": 0.0001145,
"loss": 7.5779,
"mean_token_accuracy": 0.07856447473168374,
"num_tokens": 524380.0,
"step": 230
},
{
"entropy": 8.152504539489746,
"epoch": 0.023230525899565046,
"grad_norm": 1.0078125,
"learning_rate": 0.00011700000000000001,
"loss": 7.5837,
"mean_token_accuracy": 0.08571179807186127,
"num_tokens": 535798.0,
"step": 235
},
{
"entropy": 8.032267951965332,
"epoch": 0.02372479240806643,
"grad_norm": 0.96875,
"learning_rate": 0.00011949999999999999,
"loss": 7.4937,
"mean_token_accuracy": 0.08659510985016823,
"num_tokens": 545720.0,
"step": 240
},
{
"entropy": 8.007299280166626,
"epoch": 0.024219058916567813,
"grad_norm": 1.109375,
"learning_rate": 0.000122,
"loss": 7.5277,
"mean_token_accuracy": 0.0827151209115982,
"num_tokens": 557110.0,
"step": 245
},
{
"entropy": 8.001366901397706,
"epoch": 0.024713325425069196,
"grad_norm": 1.0703125,
"learning_rate": 0.0001245,
"loss": 7.4975,
"mean_token_accuracy": 0.08423097357153893,
"num_tokens": 568966.0,
"step": 250
},
{
"entropy": 8.03173599243164,
"epoch": 0.02520759193357058,
"grad_norm": 0.82421875,
"learning_rate": 0.000127,
"loss": 7.4969,
"mean_token_accuracy": 0.08669283166527748,
"num_tokens": 580441.0,
"step": 255
},
{
"entropy": 8.057562732696534,
"epoch": 0.025701858442071967,
"grad_norm": 0.95703125,
"learning_rate": 0.0001295,
"loss": 7.4982,
"mean_token_accuracy": 0.08434496074914932,
"num_tokens": 592698.0,
"step": 260
},
{
"entropy": 7.953376913070679,
"epoch": 0.02619612495057335,
"grad_norm": 0.79296875,
"learning_rate": 0.000132,
"loss": 7.4669,
"mean_token_accuracy": 0.08737746998667717,
"num_tokens": 604787.0,
"step": 265
},
{
"entropy": 7.998782920837402,
"epoch": 0.026690391459074734,
"grad_norm": 0.90625,
"learning_rate": 0.00013450000000000002,
"loss": 7.4532,
"mean_token_accuracy": 0.08598756566643714,
"num_tokens": 616445.0,
"step": 270
},
{
"entropy": 7.9657186508178714,
"epoch": 0.027184657967576117,
"grad_norm": 0.84375,
"learning_rate": 0.00013700000000000002,
"loss": 7.4802,
"mean_token_accuracy": 0.08283287510275841,
"num_tokens": 627723.0,
"step": 275
},
{
"entropy": 8.00689115524292,
"epoch": 0.0276789244760775,
"grad_norm": 0.96484375,
"learning_rate": 0.0001395,
"loss": 7.5189,
"mean_token_accuracy": 0.08700487017631531,
"num_tokens": 638987.0,
"step": 280
},
{
"entropy": 7.936768245697022,
"epoch": 0.028173190984578884,
"grad_norm": 1.1484375,
"learning_rate": 0.00014199999999999998,
"loss": 7.334,
"mean_token_accuracy": 0.09142324700951576,
"num_tokens": 650626.0,
"step": 285
},
{
"entropy": 7.860474920272827,
"epoch": 0.028667457493080268,
"grad_norm": 0.95703125,
"learning_rate": 0.0001445,
"loss": 7.3572,
"mean_token_accuracy": 0.09691204503178596,
"num_tokens": 661911.0,
"step": 290
},
{
"entropy": 7.903319931030273,
"epoch": 0.02916172400158165,
"grad_norm": 0.89453125,
"learning_rate": 0.000147,
"loss": 7.4067,
"mean_token_accuracy": 0.09089745208621025,
"num_tokens": 674298.0,
"step": 295
},
{
"entropy": 7.882914972305298,
"epoch": 0.02965599051008304,
"grad_norm": 1.0625,
"learning_rate": 0.0001495,
"loss": 7.3443,
"mean_token_accuracy": 0.0963382288813591,
"num_tokens": 686000.0,
"step": 300
},
{
"entropy": 7.80755648612976,
"epoch": 0.030150257018584422,
"grad_norm": 1.1171875,
"learning_rate": 0.000152,
"loss": 7.362,
"mean_token_accuracy": 0.0944715328514576,
"num_tokens": 698464.0,
"step": 305
},
{
"entropy": 7.860896110534668,
"epoch": 0.030644523527085805,
"grad_norm": 0.828125,
"learning_rate": 0.00015450000000000001,
"loss": 7.379,
"mean_token_accuracy": 0.09046986699104309,
"num_tokens": 709702.0,
"step": 310
},
{
"entropy": 7.968799591064453,
"epoch": 0.03113879003558719,
"grad_norm": 2.203125,
"learning_rate": 0.000157,
"loss": 7.4846,
"mean_token_accuracy": 0.0890492059290409,
"num_tokens": 722234.0,
"step": 315
},
{
"entropy": 7.776518249511719,
"epoch": 0.03163305654408857,
"grad_norm": 0.87890625,
"learning_rate": 0.0001595,
"loss": 7.4068,
"mean_token_accuracy": 0.0861303061246872,
"num_tokens": 733001.0,
"step": 320
},
{
"entropy": 7.8235656261444095,
"epoch": 0.03212732305258996,
"grad_norm": 0.890625,
"learning_rate": 0.000162,
"loss": 7.3732,
"mean_token_accuracy": 0.09485758617520332,
"num_tokens": 743315.0,
"step": 325
},
{
"entropy": 7.8506021976470945,
"epoch": 0.03262158956109134,
"grad_norm": 0.921875,
"learning_rate": 0.00016450000000000001,
"loss": 7.3153,
"mean_token_accuracy": 0.09164077565073966,
"num_tokens": 754386.0,
"step": 330
},
{
"entropy": 7.832219028472901,
"epoch": 0.033115856069592726,
"grad_norm": 0.8671875,
"learning_rate": 0.00016700000000000002,
"loss": 7.4069,
"mean_token_accuracy": 0.09743229635059833,
"num_tokens": 766504.0,
"step": 335
},
{
"entropy": 7.739070034027099,
"epoch": 0.033610122578094107,
"grad_norm": 0.96484375,
"learning_rate": 0.00016950000000000003,
"loss": 7.2425,
"mean_token_accuracy": 0.10150521248579025,
"num_tokens": 777039.0,
"step": 340
},
{
"entropy": 7.80741925239563,
"epoch": 0.034104389086595494,
"grad_norm": 0.96484375,
"learning_rate": 0.00017199999999999998,
"loss": 7.3291,
"mean_token_accuracy": 0.09152510985732079,
"num_tokens": 787890.0,
"step": 345
},
{
"entropy": 7.745412302017212,
"epoch": 0.034598655595096874,
"grad_norm": 1.0625,
"learning_rate": 0.00017449999999999999,
"loss": 7.2991,
"mean_token_accuracy": 0.09431157857179642,
"num_tokens": 800129.0,
"step": 350
},
{
"entropy": 7.7933450698852536,
"epoch": 0.03509292210359826,
"grad_norm": 0.94921875,
"learning_rate": 0.000177,
"loss": 7.3748,
"mean_token_accuracy": 0.08783866837620735,
"num_tokens": 812517.0,
"step": 355
},
{
"entropy": 7.6894533157348635,
"epoch": 0.03558718861209965,
"grad_norm": 0.8203125,
"learning_rate": 0.0001795,
"loss": 7.2982,
"mean_token_accuracy": 0.09622510373592377,
"num_tokens": 824570.0,
"step": 360
},
{
"entropy": 7.741975736618042,
"epoch": 0.03608145512060103,
"grad_norm": 0.8125,
"learning_rate": 0.000182,
"loss": 7.2824,
"mean_token_accuracy": 0.09666576460003853,
"num_tokens": 836506.0,
"step": 365
},
{
"entropy": 7.75258502960205,
"epoch": 0.036575721629102415,
"grad_norm": 0.875,
"learning_rate": 0.0001845,
"loss": 7.3016,
"mean_token_accuracy": 0.09653399288654327,
"num_tokens": 846841.0,
"step": 370
},
{
"entropy": 7.799630975723266,
"epoch": 0.037069988137603795,
"grad_norm": 1.0390625,
"learning_rate": 0.000187,
"loss": 7.3364,
"mean_token_accuracy": 0.10044656544923783,
"num_tokens": 858770.0,
"step": 375
},
{
"entropy": 7.680880880355835,
"epoch": 0.03756425464610518,
"grad_norm": 1.0703125,
"learning_rate": 0.0001895,
"loss": 7.1812,
"mean_token_accuracy": 0.09969076812267304,
"num_tokens": 870184.0,
"step": 380
},
{
"entropy": 7.6216026782989506,
"epoch": 0.03805852115460656,
"grad_norm": 0.93359375,
"learning_rate": 0.000192,
"loss": 7.2248,
"mean_token_accuracy": 0.10294909402728081,
"num_tokens": 881843.0,
"step": 385
},
{
"entropy": 7.7437238693237305,
"epoch": 0.03855278766310795,
"grad_norm": 0.87109375,
"learning_rate": 0.0001945,
"loss": 7.2613,
"mean_token_accuracy": 0.09593532085418702,
"num_tokens": 893453.0,
"step": 390
},
{
"entropy": 7.66199746131897,
"epoch": 0.03904705417160933,
"grad_norm": 0.90625,
"learning_rate": 0.00019700000000000002,
"loss": 7.3009,
"mean_token_accuracy": 0.08956974968314171,
"num_tokens": 904417.0,
"step": 395
},
{
"entropy": 7.71559739112854,
"epoch": 0.039541320680110716,
"grad_norm": 1.015625,
"learning_rate": 0.00019950000000000002,
"loss": 7.223,
"mean_token_accuracy": 0.09316879063844681,
"num_tokens": 915134.0,
"step": 400
},
{
"entropy": 7.676138925552368,
"epoch": 0.0400355871886121,
"grad_norm": 0.92578125,
"learning_rate": 0.000202,
"loss": 7.2417,
"mean_token_accuracy": 0.09755214676260948,
"num_tokens": 926361.0,
"step": 405
},
{
"entropy": 7.596137046813965,
"epoch": 0.04052985369711348,
"grad_norm": 1.0078125,
"learning_rate": 0.00020449999999999998,
"loss": 7.1619,
"mean_token_accuracy": 0.09669081121683121,
"num_tokens": 937474.0,
"step": 410
},
{
"entropy": 7.70306715965271,
"epoch": 0.04102412020561487,
"grad_norm": 1.0078125,
"learning_rate": 0.000207,
"loss": 7.2242,
"mean_token_accuracy": 0.10178504958748817,
"num_tokens": 948897.0,
"step": 415
},
{
"entropy": 7.578281736373901,
"epoch": 0.04151838671411625,
"grad_norm": 1.125,
"learning_rate": 0.0002095,
"loss": 7.2445,
"mean_token_accuracy": 0.09385722428560257,
"num_tokens": 960669.0,
"step": 420
},
{
"entropy": 7.606976747512817,
"epoch": 0.04201265322261764,
"grad_norm": 1.046875,
"learning_rate": 0.000212,
"loss": 7.097,
"mean_token_accuracy": 0.10118152871727944,
"num_tokens": 973001.0,
"step": 425
},
{
"entropy": 7.5266501903533936,
"epoch": 0.04250691973111902,
"grad_norm": 1.21875,
"learning_rate": 0.0002145,
"loss": 7.1573,
"mean_token_accuracy": 0.1002947136759758,
"num_tokens": 986214.0,
"step": 430
},
{
"entropy": 7.537716960906982,
"epoch": 0.043001186239620404,
"grad_norm": 0.97265625,
"learning_rate": 0.00021700000000000002,
"loss": 7.1675,
"mean_token_accuracy": 0.09322678521275521,
"num_tokens": 997363.0,
"step": 435
},
{
"entropy": 7.530834531784057,
"epoch": 0.04349545274812179,
"grad_norm": 1.140625,
"learning_rate": 0.0002195,
"loss": 7.092,
"mean_token_accuracy": 0.10840804129838943,
"num_tokens": 1009543.0,
"step": 440
},
{
"entropy": 7.578037929534912,
"epoch": 0.04398971925662317,
"grad_norm": 0.9765625,
"learning_rate": 0.000222,
"loss": 7.1736,
"mean_token_accuracy": 0.0948575533926487,
"num_tokens": 1020561.0,
"step": 445
},
{
"entropy": 7.589271831512451,
"epoch": 0.04448398576512456,
"grad_norm": 0.8359375,
"learning_rate": 0.0002245,
"loss": 7.1256,
"mean_token_accuracy": 0.09639807492494583,
"num_tokens": 1032587.0,
"step": 450
},
{
"entropy": 7.5890984535217285,
"epoch": 0.04497825227362594,
"grad_norm": 0.8671875,
"learning_rate": 0.00022700000000000002,
"loss": 7.1359,
"mean_token_accuracy": 0.09905373975634575,
"num_tokens": 1044249.0,
"step": 455
},
{
"entropy": 7.547998476028442,
"epoch": 0.045472518782127325,
"grad_norm": 1.0859375,
"learning_rate": 0.00022950000000000002,
"loss": 7.1744,
"mean_token_accuracy": 0.09762973859906196,
"num_tokens": 1055415.0,
"step": 460
},
{
"entropy": 7.5688958168029785,
"epoch": 0.045966785290628705,
"grad_norm": 0.9609375,
"learning_rate": 0.00023200000000000003,
"loss": 7.1466,
"mean_token_accuracy": 0.09966595321893693,
"num_tokens": 1066593.0,
"step": 465
},
{
"entropy": 7.529252672195435,
"epoch": 0.04646105179913009,
"grad_norm": 1.0703125,
"learning_rate": 0.00023449999999999998,
"loss": 7.1369,
"mean_token_accuracy": 0.10283891633152961,
"num_tokens": 1076966.0,
"step": 470
},
{
"entropy": 7.504359388351441,
"epoch": 0.04695531830763147,
"grad_norm": 0.875,
"learning_rate": 0.000237,
"loss": 7.1676,
"mean_token_accuracy": 0.09871028363704681,
"num_tokens": 1089927.0,
"step": 475
},
{
"entropy": 7.551082754135132,
"epoch": 0.04744958481613286,
"grad_norm": 1.0,
"learning_rate": 0.0002395,
"loss": 7.1255,
"mean_token_accuracy": 0.09758824929594993,
"num_tokens": 1101920.0,
"step": 480
},
{
"entropy": 7.489306211471558,
"epoch": 0.047943851324634246,
"grad_norm": 1.078125,
"learning_rate": 0.000242,
"loss": 7.1816,
"mean_token_accuracy": 0.10394024401903153,
"num_tokens": 1115078.0,
"step": 485
},
{
"entropy": 7.6370282649993895,
"epoch": 0.048438117833135626,
"grad_norm": 1.1015625,
"learning_rate": 0.0002445,
"loss": 7.2542,
"mean_token_accuracy": 0.09503929167985917,
"num_tokens": 1127138.0,
"step": 490
},
{
"entropy": 7.449513244628906,
"epoch": 0.04893238434163701,
"grad_norm": 1.125,
"learning_rate": 0.000247,
"loss": 7.0705,
"mean_token_accuracy": 0.0948334887623787,
"num_tokens": 1138390.0,
"step": 495
},
{
"entropy": 7.407353210449219,
"epoch": 0.04942665085013839,
"grad_norm": 0.94921875,
"learning_rate": 0.0002495,
"loss": 7.0009,
"mean_token_accuracy": 0.10578849464654923,
"num_tokens": 1149124.0,
"step": 500
},
{
"entropy": 7.487055921554566,
"epoch": 0.04992091735863978,
"grad_norm": 1.0546875,
"learning_rate": 0.000252,
"loss": 7.0787,
"mean_token_accuracy": 0.10215417668223381,
"num_tokens": 1161091.0,
"step": 505
},
{
"entropy": 7.527550268173218,
"epoch": 0.05041518386714116,
"grad_norm": 0.91796875,
"learning_rate": 0.0002545,
"loss": 7.062,
"mean_token_accuracy": 0.10265694484114647,
"num_tokens": 1171919.0,
"step": 510
},
{
"entropy": 7.452037620544433,
"epoch": 0.05090945037564255,
"grad_norm": 0.97265625,
"learning_rate": 0.000257,
"loss": 7.0832,
"mean_token_accuracy": 0.10110483095049858,
"num_tokens": 1183688.0,
"step": 515
},
{
"entropy": 7.405346298217774,
"epoch": 0.051403716884143934,
"grad_norm": 0.94921875,
"learning_rate": 0.0002595,
"loss": 7.012,
"mean_token_accuracy": 0.0998148925602436,
"num_tokens": 1195323.0,
"step": 520
},
{
"entropy": 7.492317199707031,
"epoch": 0.051897983392645314,
"grad_norm": 0.921875,
"learning_rate": 0.000262,
"loss": 7.1082,
"mean_token_accuracy": 0.09472725465893746,
"num_tokens": 1206782.0,
"step": 525
},
{
"entropy": 7.3935158252716064,
"epoch": 0.0523922499011467,
"grad_norm": 1.078125,
"learning_rate": 0.00026450000000000003,
"loss": 7.0087,
"mean_token_accuracy": 0.10457788184285163,
"num_tokens": 1218102.0,
"step": 530
},
{
"entropy": 7.366737365722656,
"epoch": 0.05288651640964808,
"grad_norm": 1.0390625,
"learning_rate": 0.00026700000000000004,
"loss": 6.979,
"mean_token_accuracy": 0.10683351457118988,
"num_tokens": 1230199.0,
"step": 535
},
{
"entropy": 7.456813716888428,
"epoch": 0.05338078291814947,
"grad_norm": 0.9921875,
"learning_rate": 0.00026950000000000005,
"loss": 7.1109,
"mean_token_accuracy": 0.09833906963467598,
"num_tokens": 1242127.0,
"step": 540
},
{
"entropy": 7.489393091201782,
"epoch": 0.05387504942665085,
"grad_norm": 1.28125,
"learning_rate": 0.00027200000000000005,
"loss": 6.9608,
"mean_token_accuracy": 0.10372178331017494,
"num_tokens": 1252329.0,
"step": 545
},
{
"entropy": 7.225747203826904,
"epoch": 0.054369315935152235,
"grad_norm": 0.9765625,
"learning_rate": 0.0002745,
"loss": 6.9648,
"mean_token_accuracy": 0.10731954500079155,
"num_tokens": 1264832.0,
"step": 550
},
{
"entropy": 7.370330858230591,
"epoch": 0.054863582443653615,
"grad_norm": 1.109375,
"learning_rate": 0.000277,
"loss": 7.041,
"mean_token_accuracy": 0.10336996018886566,
"num_tokens": 1276708.0,
"step": 555
},
{
"entropy": 7.453816318511963,
"epoch": 0.055357848952155,
"grad_norm": 1.203125,
"learning_rate": 0.0002795,
"loss": 6.9314,
"mean_token_accuracy": 0.11433103457093238,
"num_tokens": 1288089.0,
"step": 560
},
{
"entropy": 7.364833116531372,
"epoch": 0.05585211546065639,
"grad_norm": 1.09375,
"learning_rate": 0.00028199999999999997,
"loss": 7.0866,
"mean_token_accuracy": 0.09704122841358184,
"num_tokens": 1299411.0,
"step": 565
},
{
"entropy": 7.38861722946167,
"epoch": 0.05634638196915777,
"grad_norm": 0.8359375,
"learning_rate": 0.0002845,
"loss": 7.0718,
"mean_token_accuracy": 0.09769185036420822,
"num_tokens": 1311368.0,
"step": 570
},
{
"entropy": 7.383206033706665,
"epoch": 0.056840648477659156,
"grad_norm": 1.1171875,
"learning_rate": 0.000287,
"loss": 6.9885,
"mean_token_accuracy": 0.10403251945972443,
"num_tokens": 1321962.0,
"step": 575
},
{
"entropy": 7.367150115966797,
"epoch": 0.057334914986160536,
"grad_norm": 0.984375,
"learning_rate": 0.0002895,
"loss": 6.9718,
"mean_token_accuracy": 0.10916787758469582,
"num_tokens": 1335150.0,
"step": 580
},
{
"entropy": 7.376563596725464,
"epoch": 0.05782918149466192,
"grad_norm": 0.9375,
"learning_rate": 0.000292,
"loss": 7.0122,
"mean_token_accuracy": 0.10337089449167251,
"num_tokens": 1346986.0,
"step": 585
},
{
"entropy": 7.361315155029297,
"epoch": 0.0583234480031633,
"grad_norm": 1.2109375,
"learning_rate": 0.0002945,
"loss": 6.951,
"mean_token_accuracy": 0.10719594955444336,
"num_tokens": 1357820.0,
"step": 590
},
{
"entropy": 7.284826040267944,
"epoch": 0.05881771451166469,
"grad_norm": 1.2421875,
"learning_rate": 0.000297,
"loss": 6.929,
"mean_token_accuracy": 0.11087205111980439,
"num_tokens": 1368269.0,
"step": 595
},
{
"entropy": 7.249915885925293,
"epoch": 0.05931198102016608,
"grad_norm": 1.0703125,
"learning_rate": 0.0002995,
"loss": 6.9547,
"mean_token_accuracy": 0.10180922672152519,
"num_tokens": 1380364.0,
"step": 600
},
{
"entropy": 7.395902109146118,
"epoch": 0.05980624752866746,
"grad_norm": 1.1640625,
"learning_rate": 0.000302,
"loss": 6.9082,
"mean_token_accuracy": 0.11050013303756714,
"num_tokens": 1390667.0,
"step": 605
},
{
"entropy": 7.20961561203003,
"epoch": 0.060300514037168844,
"grad_norm": 1.0234375,
"learning_rate": 0.0003045,
"loss": 6.8605,
"mean_token_accuracy": 0.10693677067756653,
"num_tokens": 1401452.0,
"step": 610
},
{
"entropy": 7.270598030090332,
"epoch": 0.060794780545670224,
"grad_norm": 1.15625,
"learning_rate": 0.000307,
"loss": 6.8652,
"mean_token_accuracy": 0.10748695507645607,
"num_tokens": 1412947.0,
"step": 615
},
{
"entropy": 7.267968559265137,
"epoch": 0.06128904705417161,
"grad_norm": 0.94921875,
"learning_rate": 0.0003095,
"loss": 6.923,
"mean_token_accuracy": 0.10039002001285553,
"num_tokens": 1425123.0,
"step": 620
},
{
"entropy": 7.293630933761596,
"epoch": 0.06178331356267299,
"grad_norm": 0.9921875,
"learning_rate": 0.000312,
"loss": 6.8859,
"mean_token_accuracy": 0.10513018518686294,
"num_tokens": 1436538.0,
"step": 625
},
{
"entropy": 7.242138385772705,
"epoch": 0.06227758007117438,
"grad_norm": 0.921875,
"learning_rate": 0.0003145,
"loss": 6.9292,
"mean_token_accuracy": 0.1024887353181839,
"num_tokens": 1448269.0,
"step": 630
},
{
"entropy": 7.26104211807251,
"epoch": 0.06277184657967576,
"grad_norm": 0.953125,
"learning_rate": 0.000317,
"loss": 6.904,
"mean_token_accuracy": 0.11182300597429276,
"num_tokens": 1459729.0,
"step": 635
},
{
"entropy": 7.269317722320556,
"epoch": 0.06326611308817714,
"grad_norm": 0.9609375,
"learning_rate": 0.0003195,
"loss": 6.9178,
"mean_token_accuracy": 0.10927679911255836,
"num_tokens": 1470684.0,
"step": 640
},
{
"entropy": 7.160611963272094,
"epoch": 0.06376037959667853,
"grad_norm": 0.91015625,
"learning_rate": 0.000322,
"loss": 6.8722,
"mean_token_accuracy": 0.10590490102767944,
"num_tokens": 1483571.0,
"step": 645
},
{
"entropy": 7.248231744766235,
"epoch": 0.06425464610517992,
"grad_norm": 1.0078125,
"learning_rate": 0.00032450000000000003,
"loss": 6.8625,
"mean_token_accuracy": 0.113711516559124,
"num_tokens": 1494671.0,
"step": 650
},
{
"entropy": 7.255923461914063,
"epoch": 0.06474891261368129,
"grad_norm": 1.1015625,
"learning_rate": 0.00032700000000000003,
"loss": 6.8742,
"mean_token_accuracy": 0.10942275002598763,
"num_tokens": 1505435.0,
"step": 655
},
{
"entropy": 7.311585712432861,
"epoch": 0.06524317912218268,
"grad_norm": 0.9453125,
"learning_rate": 0.00032950000000000004,
"loss": 6.9542,
"mean_token_accuracy": 0.10462261438369751,
"num_tokens": 1518510.0,
"step": 660
},
{
"entropy": 7.2363609790802,
"epoch": 0.06573744563068407,
"grad_norm": 1.015625,
"learning_rate": 0.00033200000000000005,
"loss": 6.8723,
"mean_token_accuracy": 0.11201808229088783,
"num_tokens": 1530218.0,
"step": 665
},
{
"entropy": 7.211708068847656,
"epoch": 0.06623171213918545,
"grad_norm": 1.15625,
"learning_rate": 0.00033450000000000005,
"loss": 6.9092,
"mean_token_accuracy": 0.10970637276768684,
"num_tokens": 1542486.0,
"step": 670
},
{
"entropy": 7.21020245552063,
"epoch": 0.06672597864768683,
"grad_norm": 0.9453125,
"learning_rate": 0.000337,
"loss": 6.9146,
"mean_token_accuracy": 0.10766631141304969,
"num_tokens": 1553873.0,
"step": 675
},
{
"entropy": 7.245796585083008,
"epoch": 0.06722024515618821,
"grad_norm": 1.0859375,
"learning_rate": 0.0003395,
"loss": 6.8705,
"mean_token_accuracy": 0.11166855990886689,
"num_tokens": 1565343.0,
"step": 680
},
{
"entropy": 7.190037250518799,
"epoch": 0.0677145116646896,
"grad_norm": 0.9609375,
"learning_rate": 0.000342,
"loss": 6.9103,
"mean_token_accuracy": 0.10737355202436447,
"num_tokens": 1577061.0,
"step": 685
},
{
"entropy": 7.27009220123291,
"epoch": 0.06820877817319099,
"grad_norm": 1.0546875,
"learning_rate": 0.00034449999999999997,
"loss": 6.9156,
"mean_token_accuracy": 0.10903515294194221,
"num_tokens": 1588083.0,
"step": 690
},
{
"entropy": 7.203428411483765,
"epoch": 0.06870304468169237,
"grad_norm": 0.98828125,
"learning_rate": 0.000347,
"loss": 6.856,
"mean_token_accuracy": 0.10634591579437255,
"num_tokens": 1599796.0,
"step": 695
},
{
"entropy": 7.246427440643311,
"epoch": 0.06919731119019375,
"grad_norm": 1.0546875,
"learning_rate": 0.0003495,
"loss": 6.9223,
"mean_token_accuracy": 0.10735742151737213,
"num_tokens": 1611994.0,
"step": 700
},
{
"entropy": 7.121251058578491,
"epoch": 0.06969157769869513,
"grad_norm": 0.97265625,
"learning_rate": 0.000352,
"loss": 6.8412,
"mean_token_accuracy": 0.10536888688802719,
"num_tokens": 1623280.0,
"step": 705
},
{
"entropy": 7.183412694931031,
"epoch": 0.07018584420719652,
"grad_norm": 1.1328125,
"learning_rate": 0.0003545,
"loss": 6.9013,
"mean_token_accuracy": 0.10780779123306275,
"num_tokens": 1634989.0,
"step": 710
},
{
"entropy": 7.263358879089355,
"epoch": 0.07068011071569791,
"grad_norm": 0.93359375,
"learning_rate": 0.000357,
"loss": 6.9441,
"mean_token_accuracy": 0.10218141227960587,
"num_tokens": 1647042.0,
"step": 715
},
{
"entropy": 7.1384925365448,
"epoch": 0.0711743772241993,
"grad_norm": 0.96484375,
"learning_rate": 0.0003595,
"loss": 6.8193,
"mean_token_accuracy": 0.11825829073786735,
"num_tokens": 1659245.0,
"step": 720
},
{
"entropy": 7.110760354995728,
"epoch": 0.07166864373270067,
"grad_norm": 1.0390625,
"learning_rate": 0.000362,
"loss": 6.7993,
"mean_token_accuracy": 0.1118646964430809,
"num_tokens": 1670271.0,
"step": 725
},
{
"entropy": 7.166925048828125,
"epoch": 0.07216291024120206,
"grad_norm": 0.94921875,
"learning_rate": 0.0003645,
"loss": 6.8701,
"mean_token_accuracy": 0.10942979305982589,
"num_tokens": 1682648.0,
"step": 730
},
{
"entropy": 7.17819299697876,
"epoch": 0.07265717674970344,
"grad_norm": 1.0,
"learning_rate": 0.000367,
"loss": 6.8884,
"mean_token_accuracy": 0.10658517628908157,
"num_tokens": 1694542.0,
"step": 735
},
{
"entropy": 7.1897674083709715,
"epoch": 0.07315144325820483,
"grad_norm": 0.93359375,
"learning_rate": 0.0003695,
"loss": 6.7792,
"mean_token_accuracy": 0.1051691822707653,
"num_tokens": 1705737.0,
"step": 740
},
{
"entropy": 7.1092109203338625,
"epoch": 0.0736457097667062,
"grad_norm": 0.88671875,
"learning_rate": 0.000372,
"loss": 6.8445,
"mean_token_accuracy": 0.10925951823592187,
"num_tokens": 1718474.0,
"step": 745
},
{
"entropy": 7.171104812622071,
"epoch": 0.07413997627520759,
"grad_norm": 0.9921875,
"learning_rate": 0.0003745,
"loss": 6.8287,
"mean_token_accuracy": 0.1083482749760151,
"num_tokens": 1729360.0,
"step": 750
},
{
"entropy": 7.082499885559082,
"epoch": 0.07463424278370898,
"grad_norm": 0.98828125,
"learning_rate": 0.000377,
"loss": 6.8145,
"mean_token_accuracy": 0.1140282928943634,
"num_tokens": 1739654.0,
"step": 755
},
{
"entropy": 7.164748859405518,
"epoch": 0.07512850929221036,
"grad_norm": 1.1953125,
"learning_rate": 0.0003795,
"loss": 6.8316,
"mean_token_accuracy": 0.1102795235812664,
"num_tokens": 1752158.0,
"step": 760
},
{
"entropy": 7.129274892807007,
"epoch": 0.07562277580071175,
"grad_norm": 1.0703125,
"learning_rate": 0.000382,
"loss": 6.804,
"mean_token_accuracy": 0.11611431241035461,
"num_tokens": 1764545.0,
"step": 765
},
{
"entropy": 7.115315723419189,
"epoch": 0.07611704230921312,
"grad_norm": 1.1796875,
"learning_rate": 0.0003845,
"loss": 6.8045,
"mean_token_accuracy": 0.1117110326886177,
"num_tokens": 1775422.0,
"step": 770
},
{
"entropy": 7.086942100524903,
"epoch": 0.07661130881771451,
"grad_norm": 0.8046875,
"learning_rate": 0.00038700000000000003,
"loss": 6.8269,
"mean_token_accuracy": 0.10424647778272629,
"num_tokens": 1787833.0,
"step": 775
},
{
"entropy": 7.1043376445770265,
"epoch": 0.0771055753262159,
"grad_norm": 1.046875,
"learning_rate": 0.00038950000000000003,
"loss": 6.6876,
"mean_token_accuracy": 0.12525334507226943,
"num_tokens": 1798508.0,
"step": 780
},
{
"entropy": 7.1196941375732425,
"epoch": 0.07759984183471728,
"grad_norm": 1.0859375,
"learning_rate": 0.00039200000000000004,
"loss": 6.7504,
"mean_token_accuracy": 0.11390295699238777,
"num_tokens": 1809873.0,
"step": 785
},
{
"entropy": 7.046653699874878,
"epoch": 0.07809410834321866,
"grad_norm": 1.4609375,
"learning_rate": 0.00039450000000000005,
"loss": 6.8219,
"mean_token_accuracy": 0.10968291312456131,
"num_tokens": 1820876.0,
"step": 790
},
{
"entropy": 7.074091911315918,
"epoch": 0.07858837485172004,
"grad_norm": 0.9609375,
"learning_rate": 0.00039700000000000005,
"loss": 6.7882,
"mean_token_accuracy": 0.10766826942563057,
"num_tokens": 1833233.0,
"step": 795
},
{
"entropy": 7.204562473297119,
"epoch": 0.07908264136022143,
"grad_norm": 1.1328125,
"learning_rate": 0.0003995,
"loss": 6.8631,
"mean_token_accuracy": 0.1043733924627304,
"num_tokens": 1844520.0,
"step": 800
},
{
"entropy": 6.992415571212769,
"epoch": 0.07957690786872282,
"grad_norm": 0.984375,
"learning_rate": 0.000402,
"loss": 6.7529,
"mean_token_accuracy": 0.11640457883477211,
"num_tokens": 1856814.0,
"step": 805
},
{
"entropy": 7.083125209808349,
"epoch": 0.0800711743772242,
"grad_norm": 0.8671875,
"learning_rate": 0.0004045,
"loss": 6.7458,
"mean_token_accuracy": 0.11934995800256729,
"num_tokens": 1867581.0,
"step": 810
},
{
"entropy": 7.114955091476441,
"epoch": 0.08056544088572558,
"grad_norm": 1.21875,
"learning_rate": 0.00040699999999999997,
"loss": 6.793,
"mean_token_accuracy": 0.10545687675476074,
"num_tokens": 1879302.0,
"step": 815
},
{
"entropy": 7.047843027114868,
"epoch": 0.08105970739422697,
"grad_norm": 1.0703125,
"learning_rate": 0.0004095,
"loss": 6.7703,
"mean_token_accuracy": 0.10801751762628556,
"num_tokens": 1891697.0,
"step": 820
},
{
"entropy": 7.011522197723389,
"epoch": 0.08155397390272835,
"grad_norm": 1.0625,
"learning_rate": 0.000412,
"loss": 6.6972,
"mean_token_accuracy": 0.113658956438303,
"num_tokens": 1903544.0,
"step": 825
},
{
"entropy": 7.099178838729858,
"epoch": 0.08204824041122974,
"grad_norm": 1.203125,
"learning_rate": 0.0004145,
"loss": 6.7716,
"mean_token_accuracy": 0.11163810566067696,
"num_tokens": 1915286.0,
"step": 830
},
{
"entropy": 6.956659364700317,
"epoch": 0.08254250691973111,
"grad_norm": 1.109375,
"learning_rate": 0.000417,
"loss": 6.6435,
"mean_token_accuracy": 0.1189879409968853,
"num_tokens": 1926361.0,
"step": 835
},
{
"entropy": 7.089762735366821,
"epoch": 0.0830367734282325,
"grad_norm": 0.921875,
"learning_rate": 0.0004195,
"loss": 6.7388,
"mean_token_accuracy": 0.11271768435835838,
"num_tokens": 1938540.0,
"step": 840
},
{
"entropy": 7.055161714553833,
"epoch": 0.08353103993673389,
"grad_norm": 0.98828125,
"learning_rate": 0.000422,
"loss": 6.7549,
"mean_token_accuracy": 0.11166792660951615,
"num_tokens": 1950532.0,
"step": 845
},
{
"entropy": 7.115563917160034,
"epoch": 0.08402530644523527,
"grad_norm": 1.0546875,
"learning_rate": 0.0004245,
"loss": 6.7771,
"mean_token_accuracy": 0.11082297042012215,
"num_tokens": 1960899.0,
"step": 850
},
{
"entropy": 6.951827669143677,
"epoch": 0.08451957295373666,
"grad_norm": 0.90625,
"learning_rate": 0.000427,
"loss": 6.7352,
"mean_token_accuracy": 0.11721526905894279,
"num_tokens": 1972278.0,
"step": 855
},
{
"entropy": 7.075967121124267,
"epoch": 0.08501383946223803,
"grad_norm": 0.8984375,
"learning_rate": 0.0004295,
"loss": 6.753,
"mean_token_accuracy": 0.11036830991506577,
"num_tokens": 1985079.0,
"step": 860
},
{
"entropy": 6.958264875411987,
"epoch": 0.08550810597073942,
"grad_norm": 1.046875,
"learning_rate": 0.000432,
"loss": 6.7833,
"mean_token_accuracy": 0.10577104613184929,
"num_tokens": 1996467.0,
"step": 865
},
{
"entropy": 7.0288783550262455,
"epoch": 0.08600237247924081,
"grad_norm": 1.0,
"learning_rate": 0.0004345,
"loss": 6.6732,
"mean_token_accuracy": 0.11622288599610328,
"num_tokens": 2007339.0,
"step": 870
},
{
"entropy": 7.031373453140259,
"epoch": 0.0864966389877422,
"grad_norm": 1.125,
"learning_rate": 0.000437,
"loss": 6.6974,
"mean_token_accuracy": 0.11922055259346961,
"num_tokens": 2019962.0,
"step": 875
},
{
"entropy": 7.015153598785401,
"epoch": 0.08699090549624358,
"grad_norm": 1.125,
"learning_rate": 0.0004395,
"loss": 6.6811,
"mean_token_accuracy": 0.1218427762389183,
"num_tokens": 2031251.0,
"step": 880
},
{
"entropy": 6.9253716468811035,
"epoch": 0.08748517200474495,
"grad_norm": 1.0234375,
"learning_rate": 0.000442,
"loss": 6.6908,
"mean_token_accuracy": 0.11059122309088706,
"num_tokens": 2042412.0,
"step": 885
},
{
"entropy": 6.9423582553863525,
"epoch": 0.08797943851324634,
"grad_norm": 0.9921875,
"learning_rate": 0.0004445,
"loss": 6.7154,
"mean_token_accuracy": 0.11710638850927353,
"num_tokens": 2054423.0,
"step": 890
},
{
"entropy": 6.971608352661133,
"epoch": 0.08847370502174773,
"grad_norm": 1.0703125,
"learning_rate": 0.000447,
"loss": 6.6994,
"mean_token_accuracy": 0.11089581474661828,
"num_tokens": 2065657.0,
"step": 895
},
{
"entropy": 7.033730363845825,
"epoch": 0.08896797153024912,
"grad_norm": 0.95703125,
"learning_rate": 0.00044950000000000003,
"loss": 6.6673,
"mean_token_accuracy": 0.11892515495419502,
"num_tokens": 2076305.0,
"step": 900
},
{
"entropy": 6.9093526840209964,
"epoch": 0.08946223803875049,
"grad_norm": 1.03125,
"learning_rate": 0.00045200000000000004,
"loss": 6.6538,
"mean_token_accuracy": 0.11571658477187156,
"num_tokens": 2087734.0,
"step": 905
},
{
"entropy": 6.8397496223449705,
"epoch": 0.08995650454725188,
"grad_norm": 0.95703125,
"learning_rate": 0.00045450000000000004,
"loss": 6.5604,
"mean_token_accuracy": 0.12399489283561707,
"num_tokens": 2098016.0,
"step": 910
},
{
"entropy": 6.953526353836059,
"epoch": 0.09045077105575326,
"grad_norm": 0.953125,
"learning_rate": 0.00045700000000000005,
"loss": 6.7456,
"mean_token_accuracy": 0.11191561967134475,
"num_tokens": 2110293.0,
"step": 915
},
{
"entropy": 6.92013955116272,
"epoch": 0.09094503756425465,
"grad_norm": 0.9453125,
"learning_rate": 0.00045950000000000006,
"loss": 6.5414,
"mean_token_accuracy": 0.12161664664745331,
"num_tokens": 2120441.0,
"step": 920
},
{
"entropy": 6.89482569694519,
"epoch": 0.09143930407275604,
"grad_norm": 0.97265625,
"learning_rate": 0.000462,
"loss": 6.6421,
"mean_token_accuracy": 0.12238077148795128,
"num_tokens": 2131520.0,
"step": 925
},
{
"entropy": 6.9351709365844725,
"epoch": 0.09193357058125741,
"grad_norm": 0.79296875,
"learning_rate": 0.0004645,
"loss": 6.6623,
"mean_token_accuracy": 0.11628105267882347,
"num_tokens": 2145005.0,
"step": 930
},
{
"entropy": 6.952813577651978,
"epoch": 0.0924278370897588,
"grad_norm": 0.97265625,
"learning_rate": 0.000467,
"loss": 6.6147,
"mean_token_accuracy": 0.12266910672187806,
"num_tokens": 2156608.0,
"step": 935
},
{
"entropy": 6.844784832000732,
"epoch": 0.09292210359826018,
"grad_norm": 1.0078125,
"learning_rate": 0.0004695,
"loss": 6.6121,
"mean_token_accuracy": 0.12510473579168319,
"num_tokens": 2168341.0,
"step": 940
},
{
"entropy": 6.867721843719482,
"epoch": 0.09341637010676157,
"grad_norm": 0.9921875,
"learning_rate": 0.000472,
"loss": 6.622,
"mean_token_accuracy": 0.11190108060836793,
"num_tokens": 2178704.0,
"step": 945
},
{
"entropy": 7.011946821212769,
"epoch": 0.09391063661526294,
"grad_norm": 0.99609375,
"learning_rate": 0.0004745,
"loss": 6.7787,
"mean_token_accuracy": 0.1083400510251522,
"num_tokens": 2190884.0,
"step": 950
},
{
"entropy": 6.949655151367187,
"epoch": 0.09440490312376433,
"grad_norm": 1.03125,
"learning_rate": 0.000477,
"loss": 6.6793,
"mean_token_accuracy": 0.11808931306004525,
"num_tokens": 2201822.0,
"step": 955
},
{
"entropy": 6.885033130645752,
"epoch": 0.09489916963226572,
"grad_norm": 1.0625,
"learning_rate": 0.0004795,
"loss": 6.6691,
"mean_token_accuracy": 0.1151963710784912,
"num_tokens": 2213362.0,
"step": 960
},
{
"entropy": 6.988611841201783,
"epoch": 0.0953934361407671,
"grad_norm": 0.9375,
"learning_rate": 0.000482,
"loss": 6.6834,
"mean_token_accuracy": 0.11506200209259987,
"num_tokens": 2224366.0,
"step": 965
},
{
"entropy": 6.796273279190063,
"epoch": 0.09588770264926849,
"grad_norm": 1.078125,
"learning_rate": 0.0004845,
"loss": 6.5308,
"mean_token_accuracy": 0.11869918927550316,
"num_tokens": 2234645.0,
"step": 970
},
{
"entropy": 6.887414264678955,
"epoch": 0.09638196915776986,
"grad_norm": 1.0703125,
"learning_rate": 0.000487,
"loss": 6.6156,
"mean_token_accuracy": 0.12557216882705688,
"num_tokens": 2245904.0,
"step": 975
},
{
"entropy": 6.760451459884644,
"epoch": 0.09687623566627125,
"grad_norm": 0.97265625,
"learning_rate": 0.0004895,
"loss": 6.5608,
"mean_token_accuracy": 0.11967702507972718,
"num_tokens": 2257753.0,
"step": 980
},
{
"entropy": 6.836030530929565,
"epoch": 0.09737050217477264,
"grad_norm": 1.015625,
"learning_rate": 0.000492,
"loss": 6.5764,
"mean_token_accuracy": 0.1251159615814686,
"num_tokens": 2269009.0,
"step": 985
},
{
"entropy": 6.922193813323974,
"epoch": 0.09786476868327403,
"grad_norm": 0.97265625,
"learning_rate": 0.0004945,
"loss": 6.616,
"mean_token_accuracy": 0.11580619663000107,
"num_tokens": 2279840.0,
"step": 990
},
{
"entropy": 6.835604095458985,
"epoch": 0.0983590351917754,
"grad_norm": 1.03125,
"learning_rate": 0.000497,
"loss": 6.5989,
"mean_token_accuracy": 0.11514367312192916,
"num_tokens": 2290577.0,
"step": 995
},
{
"entropy": 6.809481859207153,
"epoch": 0.09885330170027679,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995,
"loss": 6.5868,
"mean_token_accuracy": 0.12064301818609238,
"num_tokens": 2301341.0,
"step": 1000
},
{
"entropy": 6.813262176513672,
"epoch": 0.09934756820877817,
"grad_norm": 0.96875,
"learning_rate": 0.0004999999982287888,
"loss": 6.6368,
"mean_token_accuracy": 0.11775564104318618,
"num_tokens": 2312505.0,
"step": 1005
},
{
"entropy": 6.9637086391448975,
"epoch": 0.09984183471727956,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999999910332434,
"loss": 6.5746,
"mean_token_accuracy": 0.11561667025089264,
"num_tokens": 2323443.0,
"step": 1010
},
{
"entropy": 6.791029787063598,
"epoch": 0.10033610122578095,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999783026635,
"loss": 6.5993,
"mean_token_accuracy": 0.11769351288676262,
"num_tokens": 2335686.0,
"step": 1015
},
{
"entropy": 6.783609867095947,
"epoch": 0.10083036773428232,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999999600370491,
"loss": 6.5922,
"mean_token_accuracy": 0.11608155444264412,
"num_tokens": 2347273.0,
"step": 1020
},
{
"entropy": 6.764084768295288,
"epoch": 0.1013246342427837,
"grad_norm": 1.125,
"learning_rate": 0.0004999999362364008,
"loss": 6.5001,
"mean_token_accuracy": 0.13010589852929116,
"num_tokens": 2358011.0,
"step": 1025
},
{
"entropy": 6.786945724487305,
"epoch": 0.1018189007512851,
"grad_norm": 0.984375,
"learning_rate": 0.0004999999069007191,
"loss": 6.6375,
"mean_token_accuracy": 0.1150654211640358,
"num_tokens": 2371390.0,
"step": 1030
},
{
"entropy": 6.861456680297851,
"epoch": 0.10231316725978648,
"grad_norm": 1.03125,
"learning_rate": 0.0004999998720300048,
"loss": 6.5951,
"mean_token_accuracy": 0.12022169157862664,
"num_tokens": 2381887.0,
"step": 1035
},
{
"entropy": 6.88414945602417,
"epoch": 0.10280743376828787,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999998316242589,
"loss": 6.535,
"mean_token_accuracy": 0.12180361151695251,
"num_tokens": 2393632.0,
"step": 1040
},
{
"entropy": 6.735341882705688,
"epoch": 0.10330170027678924,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999997856834821,
"loss": 6.5544,
"mean_token_accuracy": 0.12011176198720933,
"num_tokens": 2405686.0,
"step": 1045
},
{
"entropy": 6.720200252532959,
"epoch": 0.10379596678529063,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999997342076757,
"loss": 6.5159,
"mean_token_accuracy": 0.11903640404343604,
"num_tokens": 2417459.0,
"step": 1050
},
{
"entropy": 6.890789365768432,
"epoch": 0.10429023329379201,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999996771968408,
"loss": 6.5692,
"mean_token_accuracy": 0.11979348957538605,
"num_tokens": 2428689.0,
"step": 1055
},
{
"entropy": 6.733220434188842,
"epoch": 0.1047844998022934,
"grad_norm": 1.015625,
"learning_rate": 0.0004999996146509791,
"loss": 6.5962,
"mean_token_accuracy": 0.11043051034212112,
"num_tokens": 2439437.0,
"step": 1060
},
{
"entropy": 6.875022935867309,
"epoch": 0.10527876631079477,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999995465700919,
"loss": 6.5727,
"mean_token_accuracy": 0.11816314235329628,
"num_tokens": 2450576.0,
"step": 1065
},
{
"entropy": 6.892560863494873,
"epoch": 0.10577303281929616,
"grad_norm": 1.015625,
"learning_rate": 0.0004999994729541809,
"loss": 6.6961,
"mean_token_accuracy": 0.11508011296391488,
"num_tokens": 2462684.0,
"step": 1070
},
{
"entropy": 6.847589826583862,
"epoch": 0.10626729932779755,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999993938032481,
"loss": 6.5992,
"mean_token_accuracy": 0.12335227355360985,
"num_tokens": 2474193.0,
"step": 1075
},
{
"entropy": 6.71039776802063,
"epoch": 0.10676156583629894,
"grad_norm": 0.921875,
"learning_rate": 0.0004999993091172952,
"loss": 6.5155,
"mean_token_accuracy": 0.118268733471632,
"num_tokens": 2486264.0,
"step": 1080
},
{
"entropy": 6.779750728607178,
"epoch": 0.10725583234480032,
"grad_norm": 0.984375,
"learning_rate": 0.0004999992188963245,
"loss": 6.5201,
"mean_token_accuracy": 0.12872469574213027,
"num_tokens": 2497668.0,
"step": 1085
},
{
"entropy": 6.727497005462647,
"epoch": 0.1077500988533017,
"grad_norm": 0.94921875,
"learning_rate": 0.000499999123140338,
"loss": 6.5714,
"mean_token_accuracy": 0.11793504282832146,
"num_tokens": 2509635.0,
"step": 1090
},
{
"entropy": 6.829061508178711,
"epoch": 0.10824436536180308,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999990218493382,
"loss": 6.5694,
"mean_token_accuracy": 0.11895573884248734,
"num_tokens": 2521980.0,
"step": 1095
},
{
"entropy": 6.751840114593506,
"epoch": 0.10873863187030447,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999989150233276,
"loss": 6.525,
"mean_token_accuracy": 0.12123857289552689,
"num_tokens": 2533750.0,
"step": 1100
},
{
"entropy": 6.78528938293457,
"epoch": 0.10923289837880586,
"grad_norm": 0.953125,
"learning_rate": 0.0004999988026623088,
"loss": 6.622,
"mean_token_accuracy": 0.11473045870661736,
"num_tokens": 2545620.0,
"step": 1105
},
{
"entropy": 6.696118307113648,
"epoch": 0.10972716488730723,
"grad_norm": 1.015625,
"learning_rate": 0.0004999986847662843,
"loss": 6.3993,
"mean_token_accuracy": 0.1327408328652382,
"num_tokens": 2556463.0,
"step": 1110
},
{
"entropy": 6.784327554702759,
"epoch": 0.11022143139580862,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999985613352576,
"loss": 6.5923,
"mean_token_accuracy": 0.11819555461406708,
"num_tokens": 2567396.0,
"step": 1115
},
{
"entropy": 6.7881745338439945,
"epoch": 0.11071569790431,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999984323692311,
"loss": 6.5291,
"mean_token_accuracy": 0.12449823319911957,
"num_tokens": 2579208.0,
"step": 1120
},
{
"entropy": 6.713969278335571,
"epoch": 0.11120996441281139,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999982978682085,
"loss": 6.4819,
"mean_token_accuracy": 0.12787408530712127,
"num_tokens": 2590940.0,
"step": 1125
},
{
"entropy": 6.781643867492676,
"epoch": 0.11170423092131278,
"grad_norm": 0.890625,
"learning_rate": 0.0004999981578321927,
"loss": 6.5599,
"mean_token_accuracy": 0.12885867655277253,
"num_tokens": 2602491.0,
"step": 1130
},
{
"entropy": 6.6775916576385494,
"epoch": 0.11219849742981415,
"grad_norm": 1.25,
"learning_rate": 0.0004999980122611874,
"loss": 6.3952,
"mean_token_accuracy": 0.13354044929146766,
"num_tokens": 2612139.0,
"step": 1135
},
{
"entropy": 6.752026987075806,
"epoch": 0.11269276393831554,
"grad_norm": 0.953125,
"learning_rate": 0.000499997861155196,
"loss": 6.5509,
"mean_token_accuracy": 0.12556230872869492,
"num_tokens": 2624906.0,
"step": 1140
},
{
"entropy": 6.673003530502319,
"epoch": 0.11318703044681692,
"grad_norm": 1.015625,
"learning_rate": 0.0004999977045142223,
"loss": 6.4808,
"mean_token_accuracy": 0.1187933087348938,
"num_tokens": 2636046.0,
"step": 1145
},
{
"entropy": 6.753292274475098,
"epoch": 0.11368129695531831,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999975423382703,
"loss": 6.4753,
"mean_token_accuracy": 0.12741708233952523,
"num_tokens": 2645617.0,
"step": 1150
},
{
"entropy": 6.790705585479737,
"epoch": 0.11417556346381968,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999973746273438,
"loss": 6.5141,
"mean_token_accuracy": 0.122653928399086,
"num_tokens": 2656780.0,
"step": 1155
},
{
"entropy": 6.733772039413452,
"epoch": 0.11466982997232107,
"grad_norm": 1.0078125,
"learning_rate": 0.000499997201381447,
"loss": 6.5012,
"mean_token_accuracy": 0.12425200417637824,
"num_tokens": 2668078.0,
"step": 1160
},
{
"entropy": 6.741680097579956,
"epoch": 0.11516409648082246,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999970226005841,
"loss": 6.5639,
"mean_token_accuracy": 0.11674718782305718,
"num_tokens": 2679049.0,
"step": 1165
},
{
"entropy": 6.803271245956421,
"epoch": 0.11565836298932385,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999968382847595,
"loss": 6.5366,
"mean_token_accuracy": 0.12377994433045388,
"num_tokens": 2690165.0,
"step": 1170
},
{
"entropy": 6.783146047592163,
"epoch": 0.11615262949782523,
"grad_norm": 1.015625,
"learning_rate": 0.0004999966484339778,
"loss": 6.5556,
"mean_token_accuracy": 0.12106125354766846,
"num_tokens": 2701965.0,
"step": 1175
},
{
"entropy": 6.6661967754364015,
"epoch": 0.1166468960063266,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999964530482437,
"loss": 6.4637,
"mean_token_accuracy": 0.12662513479590415,
"num_tokens": 2713479.0,
"step": 1180
},
{
"entropy": 6.716840696334839,
"epoch": 0.11714116251482799,
"grad_norm": 1.03125,
"learning_rate": 0.0004999962521275619,
"loss": 6.4722,
"mean_token_accuracy": 0.1229412630200386,
"num_tokens": 2725629.0,
"step": 1185
},
{
"entropy": 6.520010709762573,
"epoch": 0.11763542902332938,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999960456719374,
"loss": 6.3949,
"mean_token_accuracy": 0.12263003587722779,
"num_tokens": 2736620.0,
"step": 1190
},
{
"entropy": 6.698487615585327,
"epoch": 0.11812969553183077,
"grad_norm": 1.125,
"learning_rate": 0.0004999958336813753,
"loss": 6.3743,
"mean_token_accuracy": 0.13159746527671815,
"num_tokens": 2747118.0,
"step": 1195
},
{
"entropy": 6.5850576877594,
"epoch": 0.11862396204033215,
"grad_norm": 1.078125,
"learning_rate": 0.0004999956161558807,
"loss": 6.4496,
"mean_token_accuracy": 0.12530086562037468,
"num_tokens": 2758371.0,
"step": 1200
},
{
"entropy": 6.637788963317871,
"epoch": 0.11911822854883353,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999953930954591,
"loss": 6.4311,
"mean_token_accuracy": 0.12403330877423287,
"num_tokens": 2770575.0,
"step": 1205
},
{
"entropy": 6.686907768249512,
"epoch": 0.11961249505733491,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999951645001159,
"loss": 6.3959,
"mean_token_accuracy": 0.12846756651997565,
"num_tokens": 2783103.0,
"step": 1210
},
{
"entropy": 6.6527008533477785,
"epoch": 0.1201067615658363,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999949303698568,
"loss": 6.516,
"mean_token_accuracy": 0.12293659523129463,
"num_tokens": 2796636.0,
"step": 1215
},
{
"entropy": 6.695577192306518,
"epoch": 0.12060102807433769,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999946907046875,
"loss": 6.3694,
"mean_token_accuracy": 0.13485616445541382,
"num_tokens": 2807798.0,
"step": 1220
},
{
"entropy": 6.6003742694854735,
"epoch": 0.12109529458283906,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999944455046139,
"loss": 6.487,
"mean_token_accuracy": 0.12794808968901633,
"num_tokens": 2819255.0,
"step": 1225
},
{
"entropy": 6.676394367218018,
"epoch": 0.12158956109134045,
"grad_norm": 1.015625,
"learning_rate": 0.000499994194769642,
"loss": 6.4363,
"mean_token_accuracy": 0.12319996431469918,
"num_tokens": 2829952.0,
"step": 1230
},
{
"entropy": 6.776423740386963,
"epoch": 0.12208382759984183,
"grad_norm": 1.0078125,
"learning_rate": 0.000499993938499778,
"loss": 6.5218,
"mean_token_accuracy": 0.1313498340547085,
"num_tokens": 2841056.0,
"step": 1235
},
{
"entropy": 6.614255428314209,
"epoch": 0.12257809410834322,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999936766950283,
"loss": 6.3736,
"mean_token_accuracy": 0.13403896167874335,
"num_tokens": 2852045.0,
"step": 1240
},
{
"entropy": 6.6999880313873295,
"epoch": 0.12307236061684461,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999934093553992,
"loss": 6.4759,
"mean_token_accuracy": 0.1261597417294979,
"num_tokens": 2863440.0,
"step": 1245
},
{
"entropy": 6.664808034896851,
"epoch": 0.12356662712534598,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999931364808972,
"loss": 6.4951,
"mean_token_accuracy": 0.12566187903285025,
"num_tokens": 2874017.0,
"step": 1250
},
{
"entropy": 6.6710155487060545,
"epoch": 0.12406089363384737,
"grad_norm": 1.03125,
"learning_rate": 0.0004999928580715293,
"loss": 6.4019,
"mean_token_accuracy": 0.1276327282190323,
"num_tokens": 2885677.0,
"step": 1255
},
{
"entropy": 6.632281970977783,
"epoch": 0.12455516014234876,
"grad_norm": 0.9375,
"learning_rate": 0.000499992574127302,
"loss": 6.5095,
"mean_token_accuracy": 0.12587906420230865,
"num_tokens": 2897825.0,
"step": 1260
},
{
"entropy": 6.72322564125061,
"epoch": 0.12504942665085014,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999922846482226,
"loss": 6.558,
"mean_token_accuracy": 0.11891639679670334,
"num_tokens": 2909304.0,
"step": 1265
},
{
"entropy": 6.680528593063355,
"epoch": 0.12554369315935152,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999919896342981,
"loss": 6.494,
"mean_token_accuracy": 0.11919789239764214,
"num_tokens": 2920918.0,
"step": 1270
},
{
"entropy": 6.711538314819336,
"epoch": 0.12603795966785292,
"grad_norm": 0.9375,
"learning_rate": 0.0004999916890855358,
"loss": 6.4429,
"mean_token_accuracy": 0.12809634804725648,
"num_tokens": 2932294.0,
"step": 1275
},
{
"entropy": 6.614521741867065,
"epoch": 0.1265322261763543,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999913830019431,
"loss": 6.3589,
"mean_token_accuracy": 0.13480542078614235,
"num_tokens": 2943372.0,
"step": 1280
},
{
"entropy": 6.5310547828674315,
"epoch": 0.12702649268485566,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999910713835272,
"loss": 6.3364,
"mean_token_accuracy": 0.12778233140707015,
"num_tokens": 2953807.0,
"step": 1285
},
{
"entropy": 6.607894229888916,
"epoch": 0.12752075919335706,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999907542302962,
"loss": 6.2839,
"mean_token_accuracy": 0.13762608990073205,
"num_tokens": 2964372.0,
"step": 1290
},
{
"entropy": 6.600684118270874,
"epoch": 0.12801502570185844,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999904315422577,
"loss": 6.3847,
"mean_token_accuracy": 0.13021055534482,
"num_tokens": 2975537.0,
"step": 1295
},
{
"entropy": 6.563079786300659,
"epoch": 0.12850929221035984,
"grad_norm": 1.03125,
"learning_rate": 0.0004999901033194198,
"loss": 6.3563,
"mean_token_accuracy": 0.12701485753059388,
"num_tokens": 2986094.0,
"step": 1300
},
{
"entropy": 6.685067892074585,
"epoch": 0.1290035587188612,
"grad_norm": 1.0625,
"learning_rate": 0.0004999897695617904,
"loss": 6.4705,
"mean_token_accuracy": 0.12639627903699874,
"num_tokens": 2996023.0,
"step": 1305
},
{
"entropy": 6.640681552886963,
"epoch": 0.12949782522736258,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999894302693778,
"loss": 6.4669,
"mean_token_accuracy": 0.12245377749204636,
"num_tokens": 3007584.0,
"step": 1310
},
{
"entropy": 6.764276647567749,
"epoch": 0.12999209173586398,
"grad_norm": 1.15625,
"learning_rate": 0.0004999890854421902,
"loss": 6.4915,
"mean_token_accuracy": 0.12133662179112434,
"num_tokens": 3018995.0,
"step": 1315
},
{
"entropy": 6.470447969436646,
"epoch": 0.13048635824436536,
"grad_norm": 0.9375,
"learning_rate": 0.0004999887350802362,
"loss": 6.3006,
"mean_token_accuracy": 0.13123086243867874,
"num_tokens": 3030576.0,
"step": 1320
},
{
"entropy": 6.557611608505249,
"epoch": 0.13098062475286676,
"grad_norm": 1.046875,
"learning_rate": 0.0004999883791835246,
"loss": 6.3631,
"mean_token_accuracy": 0.12982181310653687,
"num_tokens": 3040765.0,
"step": 1325
},
{
"entropy": 6.625185441970825,
"epoch": 0.13147489126136813,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999880177520637,
"loss": 6.4162,
"mean_token_accuracy": 0.12189083769917489,
"num_tokens": 3053912.0,
"step": 1330
},
{
"entropy": 6.515785074234008,
"epoch": 0.1319691577698695,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999876507858628,
"loss": 6.3911,
"mean_token_accuracy": 0.12720990628004075,
"num_tokens": 3065921.0,
"step": 1335
},
{
"entropy": 6.7070118427276615,
"epoch": 0.1324634242783709,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999872782849308,
"loss": 6.4407,
"mean_token_accuracy": 0.1257806733250618,
"num_tokens": 3078643.0,
"step": 1340
},
{
"entropy": 6.503523445129394,
"epoch": 0.13295769078687228,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999869002492768,
"loss": 6.3656,
"mean_token_accuracy": 0.12997100427746772,
"num_tokens": 3090300.0,
"step": 1345
},
{
"entropy": 6.547778797149658,
"epoch": 0.13345195729537365,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999865166789102,
"loss": 6.2962,
"mean_token_accuracy": 0.13570954278111458,
"num_tokens": 3102307.0,
"step": 1350
},
{
"entropy": 6.671496915817261,
"epoch": 0.13394622380387505,
"grad_norm": 1.0,
"learning_rate": 0.0004999861275738404,
"loss": 6.438,
"mean_token_accuracy": 0.122230763733387,
"num_tokens": 3113817.0,
"step": 1355
},
{
"entropy": 6.534291315078735,
"epoch": 0.13444049031237643,
"grad_norm": 1.0625,
"learning_rate": 0.0004999857329340769,
"loss": 6.2984,
"mean_token_accuracy": 0.1316831812262535,
"num_tokens": 3124719.0,
"step": 1360
},
{
"entropy": 6.49489164352417,
"epoch": 0.13493475682087783,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999853327596295,
"loss": 6.3292,
"mean_token_accuracy": 0.12992627099156379,
"num_tokens": 3136136.0,
"step": 1365
},
{
"entropy": 6.607687616348267,
"epoch": 0.1354290233293792,
"grad_norm": 0.89453125,
"learning_rate": 0.000499984927050508,
"loss": 6.3386,
"mean_token_accuracy": 0.13101759776473046,
"num_tokens": 3148977.0,
"step": 1370
},
{
"entropy": 6.594228982925415,
"epoch": 0.13592328983788057,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999845158067223,
"loss": 6.3974,
"mean_token_accuracy": 0.12740953639149666,
"num_tokens": 3160520.0,
"step": 1375
},
{
"entropy": 6.579832983016968,
"epoch": 0.13641755634638197,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999840990282828,
"loss": 6.3672,
"mean_token_accuracy": 0.1273616522550583,
"num_tokens": 3172895.0,
"step": 1380
},
{
"entropy": 6.562347316741944,
"epoch": 0.13691182285488335,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999836767151994,
"loss": 6.3811,
"mean_token_accuracy": 0.1285211369395256,
"num_tokens": 3183960.0,
"step": 1385
},
{
"entropy": 6.513716793060302,
"epoch": 0.13740608936338475,
"grad_norm": 1.03125,
"learning_rate": 0.0004999832488674827,
"loss": 6.4157,
"mean_token_accuracy": 0.12590486258268357,
"num_tokens": 3194276.0,
"step": 1390
},
{
"entropy": 6.599906539916992,
"epoch": 0.13790035587188612,
"grad_norm": 0.984375,
"learning_rate": 0.0004999828154851432,
"loss": 6.2909,
"mean_token_accuracy": 0.1359431803226471,
"num_tokens": 3206084.0,
"step": 1395
},
{
"entropy": 6.534882593154907,
"epoch": 0.1383946223803875,
"grad_norm": 1.015625,
"learning_rate": 0.0004999823765681915,
"loss": 6.3305,
"mean_token_accuracy": 0.1354965254664421,
"num_tokens": 3217941.0,
"step": 1400
},
{
"entropy": 6.533507204055786,
"epoch": 0.1388888888888889,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999819321166385,
"loss": 6.3004,
"mean_token_accuracy": 0.13300728499889375,
"num_tokens": 3229088.0,
"step": 1405
},
{
"entropy": 6.476707124710083,
"epoch": 0.13938315539739027,
"grad_norm": 0.9921875,
"learning_rate": 0.000499981482130495,
"loss": 6.2574,
"mean_token_accuracy": 0.13597408458590507,
"num_tokens": 3240708.0,
"step": 1410
},
{
"entropy": 6.52806544303894,
"epoch": 0.13987742190589167,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999810266097721,
"loss": 6.3747,
"mean_token_accuracy": 0.12898992374539375,
"num_tokens": 3251040.0,
"step": 1415
},
{
"entropy": 6.505051183700561,
"epoch": 0.14037168841439304,
"grad_norm": 1.0625,
"learning_rate": 0.0004999805655544811,
"loss": 6.3206,
"mean_token_accuracy": 0.13407422080636025,
"num_tokens": 3262655.0,
"step": 1420
},
{
"entropy": 6.514699459075928,
"epoch": 0.14086595492289442,
"grad_norm": 1.046875,
"learning_rate": 0.0004999800989646333,
"loss": 6.2762,
"mean_token_accuracy": 0.1369866333901882,
"num_tokens": 3273160.0,
"step": 1425
},
{
"entropy": 6.538471126556397,
"epoch": 0.14136022143139582,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999796268402401,
"loss": 6.3008,
"mean_token_accuracy": 0.129919171333313,
"num_tokens": 3285932.0,
"step": 1430
},
{
"entropy": 6.4485719203948975,
"epoch": 0.1418544879398972,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999791491813132,
"loss": 6.3325,
"mean_token_accuracy": 0.12975751757621765,
"num_tokens": 3298834.0,
"step": 1435
},
{
"entropy": 6.628458499908447,
"epoch": 0.1423487544483986,
"grad_norm": 1.03125,
"learning_rate": 0.0004999786659878643,
"loss": 6.3121,
"mean_token_accuracy": 0.13101852536201478,
"num_tokens": 3310473.0,
"step": 1440
},
{
"entropy": 6.355752944946289,
"epoch": 0.14284302095689996,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999781772599055,
"loss": 6.2774,
"mean_token_accuracy": 0.1346455655992031,
"num_tokens": 3322662.0,
"step": 1445
},
{
"entropy": 6.666700553894043,
"epoch": 0.14333728746540134,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999776829974484,
"loss": 6.377,
"mean_token_accuracy": 0.13617630526423455,
"num_tokens": 3334321.0,
"step": 1450
},
{
"entropy": 6.4790514469146725,
"epoch": 0.14383155397390274,
"grad_norm": 1.03125,
"learning_rate": 0.0004999771832005055,
"loss": 6.2639,
"mean_token_accuracy": 0.12925363183021546,
"num_tokens": 3345130.0,
"step": 1455
},
{
"entropy": 6.453184318542481,
"epoch": 0.1443258204824041,
"grad_norm": 1.0078125,
"learning_rate": 0.000499976677869089,
"loss": 6.2703,
"mean_token_accuracy": 0.13220407143235208,
"num_tokens": 3355601.0,
"step": 1460
},
{
"entropy": 6.477377080917359,
"epoch": 0.14482008699090548,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999761670032115,
"loss": 6.2541,
"mean_token_accuracy": 0.13375383466482163,
"num_tokens": 3367075.0,
"step": 1465
},
{
"entropy": 6.511325311660767,
"epoch": 0.14531435349940688,
"grad_norm": 0.90625,
"learning_rate": 0.0004999756506028851,
"loss": 6.2408,
"mean_token_accuracy": 0.13651637956500054,
"num_tokens": 3378888.0,
"step": 1470
},
{
"entropy": 6.501529884338379,
"epoch": 0.14580862000790826,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999751286681229,
"loss": 6.2783,
"mean_token_accuracy": 0.13225537091493605,
"num_tokens": 3390705.0,
"step": 1475
},
{
"entropy": 6.47571210861206,
"epoch": 0.14630288651640966,
"grad_norm": 0.90625,
"learning_rate": 0.0004999746011989376,
"loss": 6.2827,
"mean_token_accuracy": 0.1402031034231186,
"num_tokens": 3401710.0,
"step": 1480
},
{
"entropy": 6.556623744964599,
"epoch": 0.14679715302491103,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999740681953422,
"loss": 6.3976,
"mean_token_accuracy": 0.1264283038675785,
"num_tokens": 3413467.0,
"step": 1485
},
{
"entropy": 6.510135364532471,
"epoch": 0.1472914195334124,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999735296573499,
"loss": 6.3265,
"mean_token_accuracy": 0.1363067075610161,
"num_tokens": 3425885.0,
"step": 1490
},
{
"entropy": 6.520489978790283,
"epoch": 0.1477856860419138,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999729855849738,
"loss": 6.3524,
"mean_token_accuracy": 0.1277867116034031,
"num_tokens": 3437952.0,
"step": 1495
},
{
"entropy": 6.465743494033814,
"epoch": 0.14827995255041518,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999724359782274,
"loss": 6.2184,
"mean_token_accuracy": 0.14817326366901398,
"num_tokens": 3449680.0,
"step": 1500
},
{
"entropy": 6.495955276489258,
"epoch": 0.14877421905891658,
"grad_norm": 1.015625,
"learning_rate": 0.0004999718808371241,
"loss": 6.3814,
"mean_token_accuracy": 0.1254202604293823,
"num_tokens": 3460699.0,
"step": 1505
},
{
"entropy": 6.596548366546631,
"epoch": 0.14926848556741795,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999713201616776,
"loss": 6.2471,
"mean_token_accuracy": 0.13369383811950683,
"num_tokens": 3471919.0,
"step": 1510
},
{
"entropy": 6.353741025924682,
"epoch": 0.14976275207591933,
"grad_norm": 1.0234375,
"learning_rate": 0.000499970753951902,
"loss": 6.356,
"mean_token_accuracy": 0.12923398613929749,
"num_tokens": 3484110.0,
"step": 1515
},
{
"entropy": 6.576723194122314,
"epoch": 0.15025701858442073,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999701822078107,
"loss": 6.265,
"mean_token_accuracy": 0.1321140117943287,
"num_tokens": 3496103.0,
"step": 1520
},
{
"entropy": 6.493707942962646,
"epoch": 0.1507512850929221,
"grad_norm": 1.0703125,
"learning_rate": 0.000499969604929418,
"loss": 6.3409,
"mean_token_accuracy": 0.13155030012130736,
"num_tokens": 3507291.0,
"step": 1525
},
{
"entropy": 6.532199859619141,
"epoch": 0.1512455516014235,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999690221167382,
"loss": 6.3266,
"mean_token_accuracy": 0.1285949692130089,
"num_tokens": 3519564.0,
"step": 1530
},
{
"entropy": 6.396489763259888,
"epoch": 0.15173981810992487,
"grad_norm": 0.96875,
"learning_rate": 0.0004999684337697855,
"loss": 6.1979,
"mean_token_accuracy": 0.14119874984025954,
"num_tokens": 3530679.0,
"step": 1535
},
{
"entropy": 6.489102125167847,
"epoch": 0.15223408461842625,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999678398885746,
"loss": 6.2593,
"mean_token_accuracy": 0.13449762463569642,
"num_tokens": 3542662.0,
"step": 1540
},
{
"entropy": 6.484030818939209,
"epoch": 0.15272835112692765,
"grad_norm": 0.875,
"learning_rate": 0.0004999672404731197,
"loss": 6.3726,
"mean_token_accuracy": 0.12962616756558418,
"num_tokens": 3554760.0,
"step": 1545
},
{
"entropy": 6.430119800567627,
"epoch": 0.15322261763542902,
"grad_norm": 1.046875,
"learning_rate": 0.000499966635523436,
"loss": 6.184,
"mean_token_accuracy": 0.1408859372138977,
"num_tokens": 3565604.0,
"step": 1550
},
{
"entropy": 6.50899305343628,
"epoch": 0.1537168841439304,
"grad_norm": 1.09375,
"learning_rate": 0.0004999660250395381,
"loss": 6.3469,
"mean_token_accuracy": 0.12966826558113098,
"num_tokens": 3577535.0,
"step": 1555
},
{
"entropy": 6.513133192062378,
"epoch": 0.1542111506524318,
"grad_norm": 0.99609375,
"learning_rate": 0.000499965409021441,
"loss": 6.3448,
"mean_token_accuracy": 0.12810506224632262,
"num_tokens": 3590458.0,
"step": 1560
},
{
"entropy": 6.503912401199341,
"epoch": 0.15470541716093317,
"grad_norm": 1.0078125,
"learning_rate": 0.00049996478746916,
"loss": 6.3018,
"mean_token_accuracy": 0.13346393704414367,
"num_tokens": 3600916.0,
"step": 1565
},
{
"entropy": 6.413891839981079,
"epoch": 0.15519968366943457,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999641603827104,
"loss": 6.3234,
"mean_token_accuracy": 0.13152189254760743,
"num_tokens": 3612794.0,
"step": 1570
},
{
"entropy": 6.491737747192383,
"epoch": 0.15569395017793594,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999635277621073,
"loss": 6.2369,
"mean_token_accuracy": 0.1417726531624794,
"num_tokens": 3624186.0,
"step": 1575
},
{
"entropy": 6.421962738037109,
"epoch": 0.15618821668643731,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999628896073668,
"loss": 6.2647,
"mean_token_accuracy": 0.13585594519972802,
"num_tokens": 3636041.0,
"step": 1580
},
{
"entropy": 6.4938102722167965,
"epoch": 0.15668248319493872,
"grad_norm": 1.125,
"learning_rate": 0.0004999622459185042,
"loss": 6.3248,
"mean_token_accuracy": 0.1295533798635006,
"num_tokens": 3646859.0,
"step": 1585
},
{
"entropy": 6.432663345336914,
"epoch": 0.1571767497034401,
"grad_norm": 0.9375,
"learning_rate": 0.0004999615966955354,
"loss": 6.1457,
"mean_token_accuracy": 0.14173236414790152,
"num_tokens": 3657853.0,
"step": 1590
},
{
"entropy": 6.514078617095947,
"epoch": 0.1576710162119415,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999609419384764,
"loss": 6.2523,
"mean_token_accuracy": 0.1346098266541958,
"num_tokens": 3668341.0,
"step": 1595
},
{
"entropy": 6.446404361724854,
"epoch": 0.15816528272044286,
"grad_norm": 1.15625,
"learning_rate": 0.0004999602816473434,
"loss": 6.2758,
"mean_token_accuracy": 0.13876884281635285,
"num_tokens": 3679020.0,
"step": 1600
},
{
"entropy": 6.419503450393677,
"epoch": 0.15865954922894424,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999596158221526,
"loss": 6.1899,
"mean_token_accuracy": 0.14101243168115615,
"num_tokens": 3691121.0,
"step": 1605
},
{
"entropy": 6.448402786254883,
"epoch": 0.15915381573744564,
"grad_norm": 0.953125,
"learning_rate": 0.0004999589444629203,
"loss": 6.2489,
"mean_token_accuracy": 0.13868038430809976,
"num_tokens": 3702584.0,
"step": 1610
},
{
"entropy": 6.412480592727661,
"epoch": 0.159648082245947,
"grad_norm": 1.03125,
"learning_rate": 0.0004999582675696631,
"loss": 6.1982,
"mean_token_accuracy": 0.14014509692788124,
"num_tokens": 3713085.0,
"step": 1615
},
{
"entropy": 6.4948835372924805,
"epoch": 0.1601423487544484,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999575851423976,
"loss": 6.322,
"mean_token_accuracy": 0.1277170367538929,
"num_tokens": 3724318.0,
"step": 1620
},
{
"entropy": 6.488896036148072,
"epoch": 0.16063661526294978,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999568971811407,
"loss": 6.2169,
"mean_token_accuracy": 0.14014409109950066,
"num_tokens": 3734288.0,
"step": 1625
},
{
"entropy": 6.383061933517456,
"epoch": 0.16113088177145116,
"grad_norm": 1.015625,
"learning_rate": 0.000499956203685909,
"loss": 6.2253,
"mean_token_accuracy": 0.1334229089319706,
"num_tokens": 3744645.0,
"step": 1630
},
{
"entropy": 6.456633996963501,
"epoch": 0.16162514827995256,
"grad_norm": 1.1015625,
"learning_rate": 0.00049995550465672,
"loss": 6.219,
"mean_token_accuracy": 0.1377868391573429,
"num_tokens": 3755660.0,
"step": 1635
},
{
"entropy": 6.372065544128418,
"epoch": 0.16211941478845393,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999548000935906,
"loss": 6.1684,
"mean_token_accuracy": 0.1421402059495449,
"num_tokens": 3767614.0,
"step": 1640
},
{
"entropy": 6.441373538970947,
"epoch": 0.16261368129695533,
"grad_norm": 0.96875,
"learning_rate": 0.0004999540899965383,
"loss": 6.3184,
"mean_token_accuracy": 0.1364623002707958,
"num_tokens": 3780750.0,
"step": 1645
},
{
"entropy": 6.428185367584229,
"epoch": 0.1631079478054567,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999533743655803,
"loss": 6.3161,
"mean_token_accuracy": 0.1284065067768097,
"num_tokens": 3793195.0,
"step": 1650
},
{
"entropy": 6.457364511489868,
"epoch": 0.16360221431395808,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999526532007346,
"loss": 6.2676,
"mean_token_accuracy": 0.13861094564199447,
"num_tokens": 3804256.0,
"step": 1655
},
{
"entropy": 6.42730746269226,
"epoch": 0.16409648082245948,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999519265020187,
"loss": 6.2388,
"mean_token_accuracy": 0.13844978511333467,
"num_tokens": 3815403.0,
"step": 1660
},
{
"entropy": 6.320225238800049,
"epoch": 0.16459074733096085,
"grad_norm": 1.0,
"learning_rate": 0.0004999511942694504,
"loss": 6.1467,
"mean_token_accuracy": 0.14466449469327927,
"num_tokens": 3826682.0,
"step": 1665
},
{
"entropy": 6.43041729927063,
"epoch": 0.16508501383946222,
"grad_norm": 0.9375,
"learning_rate": 0.0004999504565030479,
"loss": 6.1678,
"mean_token_accuracy": 0.13993233293294907,
"num_tokens": 3837780.0,
"step": 1670
},
{
"entropy": 6.429787445068359,
"epoch": 0.16557928034796363,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999497132028293,
"loss": 6.2239,
"mean_token_accuracy": 0.1318391665816307,
"num_tokens": 3848782.0,
"step": 1675
},
{
"entropy": 6.428545904159546,
"epoch": 0.166073546856465,
"grad_norm": 1.0,
"learning_rate": 0.0004999489643688127,
"loss": 6.1696,
"mean_token_accuracy": 0.13959652185440063,
"num_tokens": 3859220.0,
"step": 1680
},
{
"entropy": 6.30494270324707,
"epoch": 0.1665678133649664,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999482100010169,
"loss": 6.1018,
"mean_token_accuracy": 0.1421866834163666,
"num_tokens": 3869794.0,
"step": 1685
},
{
"entropy": 6.35781421661377,
"epoch": 0.16706207987346777,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999474500994601,
"loss": 6.2396,
"mean_token_accuracy": 0.13421173691749572,
"num_tokens": 3881133.0,
"step": 1690
},
{
"entropy": 6.413172864913941,
"epoch": 0.16755634638196915,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999466846641611,
"loss": 6.1624,
"mean_token_accuracy": 0.13713917806744574,
"num_tokens": 3891600.0,
"step": 1695
},
{
"entropy": 6.333802795410156,
"epoch": 0.16805061289047055,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999459136951389,
"loss": 6.2631,
"mean_token_accuracy": 0.13335662260651587,
"num_tokens": 3903287.0,
"step": 1700
},
{
"entropy": 6.439137983322143,
"epoch": 0.16854487939897192,
"grad_norm": 1.015625,
"learning_rate": 0.0004999451371924123,
"loss": 6.1806,
"mean_token_accuracy": 0.13854067176580429,
"num_tokens": 3914497.0,
"step": 1705
},
{
"entropy": 6.338579225540161,
"epoch": 0.16903914590747332,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999443551560003,
"loss": 6.2093,
"mean_token_accuracy": 0.13935932666063308,
"num_tokens": 3925536.0,
"step": 1710
},
{
"entropy": 6.43532075881958,
"epoch": 0.1695334124159747,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999435675859224,
"loss": 6.1608,
"mean_token_accuracy": 0.14295403137803078,
"num_tokens": 3936705.0,
"step": 1715
},
{
"entropy": 6.352819776535034,
"epoch": 0.17002767892447607,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999427744821977,
"loss": 6.0594,
"mean_token_accuracy": 0.1423388808965683,
"num_tokens": 3948422.0,
"step": 1720
},
{
"entropy": 6.321838235855102,
"epoch": 0.17052194543297747,
"grad_norm": 1.0234375,
"learning_rate": 0.000499941975844846,
"loss": 6.1211,
"mean_token_accuracy": 0.14932032078504562,
"num_tokens": 3958697.0,
"step": 1725
},
{
"entropy": 6.3387940406799315,
"epoch": 0.17101621194147884,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999411716738866,
"loss": 6.1359,
"mean_token_accuracy": 0.13918534070253372,
"num_tokens": 3970255.0,
"step": 1730
},
{
"entropy": 6.376590156555176,
"epoch": 0.17151047844998024,
"grad_norm": 1.15625,
"learning_rate": 0.0004999403619693397,
"loss": 6.2221,
"mean_token_accuracy": 0.1375705450773239,
"num_tokens": 3980892.0,
"step": 1735
},
{
"entropy": 6.290405988693237,
"epoch": 0.17200474495848161,
"grad_norm": 1.015625,
"learning_rate": 0.0004999395467312249,
"loss": 6.159,
"mean_token_accuracy": 0.13679868206381798,
"num_tokens": 3991375.0,
"step": 1740
},
{
"entropy": 6.445168209075928,
"epoch": 0.172499011466983,
"grad_norm": 0.9375,
"learning_rate": 0.0004999387259595623,
"loss": 6.2982,
"mean_token_accuracy": 0.13123658969998359,
"num_tokens": 4003352.0,
"step": 1745
},
{
"entropy": 6.349807596206665,
"epoch": 0.1729932779754844,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999378996543723,
"loss": 6.1895,
"mean_token_accuracy": 0.13942418098449708,
"num_tokens": 4014429.0,
"step": 1750
},
{
"entropy": 6.37531623840332,
"epoch": 0.17348754448398576,
"grad_norm": 1.03125,
"learning_rate": 0.0004999370678156749,
"loss": 6.1121,
"mean_token_accuracy": 0.1444903664290905,
"num_tokens": 4025213.0,
"step": 1755
},
{
"entropy": 6.307505178451538,
"epoch": 0.17398181099248716,
"grad_norm": 1.09375,
"learning_rate": 0.0004999362304434909,
"loss": 6.1071,
"mean_token_accuracy": 0.1425596944987774,
"num_tokens": 4036244.0,
"step": 1760
},
{
"entropy": 6.413739204406738,
"epoch": 0.17447607750098854,
"grad_norm": 1.046875,
"learning_rate": 0.0004999353875378406,
"loss": 6.1546,
"mean_token_accuracy": 0.1389375738799572,
"num_tokens": 4047944.0,
"step": 1765
},
{
"entropy": 6.28089075088501,
"epoch": 0.1749703440094899,
"grad_norm": 0.9140625,
"learning_rate": 0.000499934539098745,
"loss": 6.1216,
"mean_token_accuracy": 0.14509597718715667,
"num_tokens": 4058865.0,
"step": 1770
},
{
"entropy": 6.326784420013428,
"epoch": 0.1754646105179913,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999336851262247,
"loss": 6.1735,
"mean_token_accuracy": 0.13881123885512353,
"num_tokens": 4070627.0,
"step": 1775
},
{
"entropy": 6.431045770645142,
"epoch": 0.17595887702649268,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999328256203009,
"loss": 6.221,
"mean_token_accuracy": 0.13756831139326095,
"num_tokens": 4081406.0,
"step": 1780
},
{
"entropy": 6.324437522888184,
"epoch": 0.17645314353499406,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999319605809947,
"loss": 6.1871,
"mean_token_accuracy": 0.12888796105980874,
"num_tokens": 4092957.0,
"step": 1785
},
{
"entropy": 6.411680078506469,
"epoch": 0.17694741004349546,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999310900083274,
"loss": 6.231,
"mean_token_accuracy": 0.1384389355778694,
"num_tokens": 4104381.0,
"step": 1790
},
{
"entropy": 6.3523437023162845,
"epoch": 0.17744167655199683,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999302139023204,
"loss": 6.1525,
"mean_token_accuracy": 0.14232124909758567,
"num_tokens": 4117190.0,
"step": 1795
},
{
"entropy": 6.305995988845825,
"epoch": 0.17793594306049823,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999293322629951,
"loss": 6.0964,
"mean_token_accuracy": 0.14318788796663284,
"num_tokens": 4127870.0,
"step": 1800
},
{
"entropy": 6.298418235778809,
"epoch": 0.1784302095689996,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999284450903736,
"loss": 6.1353,
"mean_token_accuracy": 0.14193641096353532,
"num_tokens": 4139240.0,
"step": 1805
},
{
"entropy": 6.373674917221069,
"epoch": 0.17892447607750098,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999275523844772,
"loss": 6.1209,
"mean_token_accuracy": 0.14574467837810517,
"num_tokens": 4149885.0,
"step": 1810
},
{
"entropy": 6.44624924659729,
"epoch": 0.17941874258600238,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999266541453283,
"loss": 6.2778,
"mean_token_accuracy": 0.13498160094022751,
"num_tokens": 4161824.0,
"step": 1815
},
{
"entropy": 6.335614776611328,
"epoch": 0.17991300909450375,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999257503729487,
"loss": 6.2461,
"mean_token_accuracy": 0.13315987810492516,
"num_tokens": 4173581.0,
"step": 1820
},
{
"entropy": 6.3903834342956545,
"epoch": 0.18040727560300515,
"grad_norm": 1.015625,
"learning_rate": 0.0004999248410673607,
"loss": 6.1128,
"mean_token_accuracy": 0.14748075008392333,
"num_tokens": 4184843.0,
"step": 1825
},
{
"entropy": 6.380693054199218,
"epoch": 0.18090154211150652,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999239262285869,
"loss": 6.2069,
"mean_token_accuracy": 0.1320282258093357,
"num_tokens": 4195016.0,
"step": 1830
},
{
"entropy": 6.386849069595337,
"epoch": 0.1813958086200079,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999230058566495,
"loss": 6.1467,
"mean_token_accuracy": 0.14474730044603348,
"num_tokens": 4206798.0,
"step": 1835
},
{
"entropy": 6.370251035690307,
"epoch": 0.1818900751285093,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999220799515713,
"loss": 6.131,
"mean_token_accuracy": 0.13890684843063356,
"num_tokens": 4218784.0,
"step": 1840
},
{
"entropy": 6.337589168548584,
"epoch": 0.18238434163701067,
"grad_norm": 0.92578125,
"learning_rate": 0.000499921148513375,
"loss": 6.157,
"mean_token_accuracy": 0.13808465152978897,
"num_tokens": 4229875.0,
"step": 1845
},
{
"entropy": 6.241511249542237,
"epoch": 0.18287860814551207,
"grad_norm": 1.015625,
"learning_rate": 0.0004999202115420836,
"loss": 6.1431,
"mean_token_accuracy": 0.1402451813220978,
"num_tokens": 4241801.0,
"step": 1850
},
{
"entropy": 6.385967588424682,
"epoch": 0.18337287465401345,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999192690377201,
"loss": 6.1187,
"mean_token_accuracy": 0.1410462327301502,
"num_tokens": 4253168.0,
"step": 1855
},
{
"entropy": 6.331370687484741,
"epoch": 0.18386714116251482,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999183210003076,
"loss": 6.1497,
"mean_token_accuracy": 0.13985029086470604,
"num_tokens": 4264791.0,
"step": 1860
},
{
"entropy": 6.294579124450683,
"epoch": 0.18436140767101622,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999173674298696,
"loss": 6.0931,
"mean_token_accuracy": 0.14586635753512384,
"num_tokens": 4275816.0,
"step": 1865
},
{
"entropy": 6.413454818725586,
"epoch": 0.1848556741795176,
"grad_norm": 1.0,
"learning_rate": 0.0004999164083264294,
"loss": 6.1696,
"mean_token_accuracy": 0.13626813441514968,
"num_tokens": 4287806.0,
"step": 1870
},
{
"entropy": 6.289838790893555,
"epoch": 0.185349940688019,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999154436900108,
"loss": 6.2006,
"mean_token_accuracy": 0.14326421394944192,
"num_tokens": 4300319.0,
"step": 1875
},
{
"entropy": 6.346887874603271,
"epoch": 0.18584420719652037,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999144735206373,
"loss": 6.2017,
"mean_token_accuracy": 0.13601267710328102,
"num_tokens": 4310987.0,
"step": 1880
},
{
"entropy": 6.371985960006714,
"epoch": 0.18633847370502174,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999134978183328,
"loss": 6.1595,
"mean_token_accuracy": 0.14411325454711915,
"num_tokens": 4322438.0,
"step": 1885
},
{
"entropy": 6.385655212402344,
"epoch": 0.18683274021352314,
"grad_norm": 1.140625,
"learning_rate": 0.0004999125165831215,
"loss": 6.1898,
"mean_token_accuracy": 0.1465175122022629,
"num_tokens": 4333242.0,
"step": 1890
},
{
"entropy": 6.255407238006592,
"epoch": 0.1873270067220245,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999115298150273,
"loss": 6.0597,
"mean_token_accuracy": 0.14152660295367242,
"num_tokens": 4343660.0,
"step": 1895
},
{
"entropy": 6.372375345230102,
"epoch": 0.1878212732305259,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999105375140746,
"loss": 6.1347,
"mean_token_accuracy": 0.14555889666080474,
"num_tokens": 4353382.0,
"step": 1900
},
{
"entropy": 6.268469953536988,
"epoch": 0.1883155397390273,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999095396802878,
"loss": 6.1092,
"mean_token_accuracy": 0.13386452049016953,
"num_tokens": 4364614.0,
"step": 1905
},
{
"entropy": 6.378190708160401,
"epoch": 0.18880980624752866,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999085363136913,
"loss": 6.1705,
"mean_token_accuracy": 0.13516660034656525,
"num_tokens": 4376001.0,
"step": 1910
},
{
"entropy": 6.323903131484985,
"epoch": 0.18930407275603006,
"grad_norm": 1.0,
"learning_rate": 0.00049990752741431,
"loss": 6.1771,
"mean_token_accuracy": 0.13419201746582984,
"num_tokens": 4387083.0,
"step": 1915
},
{
"entropy": 6.297012424468994,
"epoch": 0.18979833926453143,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999065129821687,
"loss": 6.1337,
"mean_token_accuracy": 0.141158889234066,
"num_tokens": 4399977.0,
"step": 1920
},
{
"entropy": 6.267847394943237,
"epoch": 0.1902926057730328,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999054930172923,
"loss": 6.066,
"mean_token_accuracy": 0.14234692901372908,
"num_tokens": 4411609.0,
"step": 1925
},
{
"entropy": 6.402806425094605,
"epoch": 0.1907868722815342,
"grad_norm": 1.078125,
"learning_rate": 0.0004999044675197058,
"loss": 6.1835,
"mean_token_accuracy": 0.14248162806034087,
"num_tokens": 4423567.0,
"step": 1930
},
{
"entropy": 6.22545223236084,
"epoch": 0.19128113879003558,
"grad_norm": 1.078125,
"learning_rate": 0.0004999034364894345,
"loss": 6.141,
"mean_token_accuracy": 0.1415740117430687,
"num_tokens": 4435410.0,
"step": 1935
},
{
"entropy": 6.422749614715576,
"epoch": 0.19177540529853698,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999023999265039,
"loss": 6.1708,
"mean_token_accuracy": 0.13967961594462394,
"num_tokens": 4445279.0,
"step": 1940
},
{
"entropy": 6.232316732406616,
"epoch": 0.19226967180703836,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999013578309392,
"loss": 6.0375,
"mean_token_accuracy": 0.14595175608992578,
"num_tokens": 4456017.0,
"step": 1945
},
{
"entropy": 6.297853326797485,
"epoch": 0.19276393831553973,
"grad_norm": 1.03125,
"learning_rate": 0.0004999003102027663,
"loss": 6.0303,
"mean_token_accuracy": 0.15286524295806886,
"num_tokens": 4466759.0,
"step": 1950
},
{
"entropy": 6.268013620376587,
"epoch": 0.19325820482404113,
"grad_norm": 1.015625,
"learning_rate": 0.000499899257042011,
"loss": 6.1067,
"mean_token_accuracy": 0.14023480713367462,
"num_tokens": 4478316.0,
"step": 1955
},
{
"entropy": 6.377132701873779,
"epoch": 0.1937524713325425,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998981983486988,
"loss": 6.1813,
"mean_token_accuracy": 0.13629069328308105,
"num_tokens": 4490775.0,
"step": 1960
},
{
"entropy": 6.301769161224366,
"epoch": 0.1942467378410439,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998971341228563,
"loss": 6.03,
"mean_token_accuracy": 0.14630968272686004,
"num_tokens": 4501622.0,
"step": 1965
},
{
"entropy": 6.303157186508178,
"epoch": 0.19474100434954528,
"grad_norm": 0.921875,
"learning_rate": 0.0004998960643645093,
"loss": 6.1343,
"mean_token_accuracy": 0.1427514761686325,
"num_tokens": 4512279.0,
"step": 1970
},
{
"entropy": 6.305225229263305,
"epoch": 0.19523527085804665,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998949890736841,
"loss": 6.121,
"mean_token_accuracy": 0.1444181725382805,
"num_tokens": 4524087.0,
"step": 1975
},
{
"entropy": 6.310341024398804,
"epoch": 0.19572953736654805,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998939082504075,
"loss": 6.1019,
"mean_token_accuracy": 0.14459830969572068,
"num_tokens": 4534816.0,
"step": 1980
},
{
"entropy": 6.280261421203614,
"epoch": 0.19622380387504942,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998928218947057,
"loss": 6.074,
"mean_token_accuracy": 0.1454021230340004,
"num_tokens": 4545725.0,
"step": 1985
},
{
"entropy": 6.377605772018432,
"epoch": 0.1967180703835508,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998917300066056,
"loss": 6.1899,
"mean_token_accuracy": 0.1378871500492096,
"num_tokens": 4557003.0,
"step": 1990
},
{
"entropy": 6.222435426712036,
"epoch": 0.1972123368920522,
"grad_norm": 0.90625,
"learning_rate": 0.0004998906325861342,
"loss": 6.101,
"mean_token_accuracy": 0.14017595276236533,
"num_tokens": 4568768.0,
"step": 1995
},
{
"entropy": 6.241748237609864,
"epoch": 0.19770660340055357,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998895296333183,
"loss": 6.0654,
"mean_token_accuracy": 0.14961813986301423,
"num_tokens": 4581116.0,
"step": 2000
},
{
"entropy": 6.334244871139527,
"epoch": 0.19820086990905497,
"grad_norm": 1.0625,
"learning_rate": 0.000499888421148185,
"loss": 6.145,
"mean_token_accuracy": 0.14341819882392884,
"num_tokens": 4592660.0,
"step": 2005
},
{
"entropy": 6.315072679519654,
"epoch": 0.19869513641755635,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998873071307617,
"loss": 6.115,
"mean_token_accuracy": 0.1400900736451149,
"num_tokens": 4603057.0,
"step": 2010
},
{
"entropy": 6.237052297592163,
"epoch": 0.19918940292605772,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998861875810758,
"loss": 6.0078,
"mean_token_accuracy": 0.15424127280712127,
"num_tokens": 4613546.0,
"step": 2015
},
{
"entropy": 6.281518363952637,
"epoch": 0.19968366943455912,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998850624991548,
"loss": 6.0869,
"mean_token_accuracy": 0.14481984600424766,
"num_tokens": 4624613.0,
"step": 2020
},
{
"entropy": 6.22469654083252,
"epoch": 0.2001779359430605,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998839318850263,
"loss": 6.0521,
"mean_token_accuracy": 0.14788607954978944,
"num_tokens": 4636004.0,
"step": 2025
},
{
"entropy": 6.255212354660034,
"epoch": 0.2006722024515619,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998827957387182,
"loss": 6.1074,
"mean_token_accuracy": 0.14652306362986564,
"num_tokens": 4646450.0,
"step": 2030
},
{
"entropy": 6.283095216751098,
"epoch": 0.20116646896006327,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998816540602584,
"loss": 6.094,
"mean_token_accuracy": 0.1446727216243744,
"num_tokens": 4658928.0,
"step": 2035
},
{
"entropy": 6.248164463043213,
"epoch": 0.20166073546856464,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998805068496752,
"loss": 6.1172,
"mean_token_accuracy": 0.14286085069179535,
"num_tokens": 4670392.0,
"step": 2040
},
{
"entropy": 6.35498571395874,
"epoch": 0.20215500197706604,
"grad_norm": 1.09375,
"learning_rate": 0.0004998793541069964,
"loss": 6.0934,
"mean_token_accuracy": 0.1423612579703331,
"num_tokens": 4680605.0,
"step": 2045
},
{
"entropy": 6.207654523849487,
"epoch": 0.2026492684855674,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998781958322507,
"loss": 6.0135,
"mean_token_accuracy": 0.14874968454241752,
"num_tokens": 4694250.0,
"step": 2050
},
{
"entropy": 6.302755451202392,
"epoch": 0.20314353499406881,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998770320254666,
"loss": 6.1632,
"mean_token_accuracy": 0.1457796052098274,
"num_tokens": 4705754.0,
"step": 2055
},
{
"entropy": 6.281993579864502,
"epoch": 0.2036378015025702,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998758626866725,
"loss": 6.0476,
"mean_token_accuracy": 0.1467272847890854,
"num_tokens": 4716393.0,
"step": 2060
},
{
"entropy": 6.196063613891601,
"epoch": 0.20413206801107156,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998746878158973,
"loss": 6.0481,
"mean_token_accuracy": 0.14491303488612176,
"num_tokens": 4726894.0,
"step": 2065
},
{
"entropy": 6.244138574600219,
"epoch": 0.20462633451957296,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998735074131699,
"loss": 6.0629,
"mean_token_accuracy": 0.14195707365870475,
"num_tokens": 4737488.0,
"step": 2070
},
{
"entropy": 6.279796409606933,
"epoch": 0.20512060102807433,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998723214785194,
"loss": 6.1494,
"mean_token_accuracy": 0.1365630991756916,
"num_tokens": 4749131.0,
"step": 2075
},
{
"entropy": 6.2584418773651125,
"epoch": 0.20561486753657574,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998711300119749,
"loss": 6.0195,
"mean_token_accuracy": 0.1438528060913086,
"num_tokens": 4760503.0,
"step": 2080
},
{
"entropy": 6.319554138183594,
"epoch": 0.2061091340450771,
"grad_norm": 0.875,
"learning_rate": 0.0004998699330135656,
"loss": 6.1319,
"mean_token_accuracy": 0.14361703097820283,
"num_tokens": 4771965.0,
"step": 2085
},
{
"entropy": 6.23747992515564,
"epoch": 0.20660340055357848,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998687304833212,
"loss": 6.1247,
"mean_token_accuracy": 0.14026476591825485,
"num_tokens": 4784215.0,
"step": 2090
},
{
"entropy": 6.243879270553589,
"epoch": 0.20709766706207988,
"grad_norm": 0.9765625,
"learning_rate": 0.000499867522421271,
"loss": 6.0339,
"mean_token_accuracy": 0.1493344783782959,
"num_tokens": 4796410.0,
"step": 2095
},
{
"entropy": 6.184952163696289,
"epoch": 0.20759193357058126,
"grad_norm": 1.0703125,
"learning_rate": 0.000499866308827445,
"loss": 6.0151,
"mean_token_accuracy": 0.15359587073326111,
"num_tokens": 4806990.0,
"step": 2100
},
{
"entropy": 6.302822113037109,
"epoch": 0.20808620007908263,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998650897018729,
"loss": 6.0734,
"mean_token_accuracy": 0.1439187079668045,
"num_tokens": 4818903.0,
"step": 2105
},
{
"entropy": 6.244305086135864,
"epoch": 0.20858046658758403,
"grad_norm": 1.03125,
"learning_rate": 0.0004998638650445847,
"loss": 6.0081,
"mean_token_accuracy": 0.1505424477159977,
"num_tokens": 4829130.0,
"step": 2110
},
{
"entropy": 6.217615032196045,
"epoch": 0.2090747330960854,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998626348556105,
"loss": 6.0298,
"mean_token_accuracy": 0.15414684265851974,
"num_tokens": 4840424.0,
"step": 2115
},
{
"entropy": 6.263989543914795,
"epoch": 0.2095689996045868,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998613991349807,
"loss": 6.0376,
"mean_token_accuracy": 0.14594117775559426,
"num_tokens": 4851674.0,
"step": 2120
},
{
"entropy": 6.096542119979858,
"epoch": 0.21006326611308818,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998601578827255,
"loss": 5.9601,
"mean_token_accuracy": 0.15645498782396317,
"num_tokens": 4863884.0,
"step": 2125
},
{
"entropy": 6.19989767074585,
"epoch": 0.21055753262158955,
"grad_norm": 0.921875,
"learning_rate": 0.0004998589110988755,
"loss": 6.0415,
"mean_token_accuracy": 0.15132073909044266,
"num_tokens": 4875582.0,
"step": 2130
},
{
"entropy": 6.274530458450317,
"epoch": 0.21105179913009095,
"grad_norm": 0.921875,
"learning_rate": 0.0004998576587834614,
"loss": 6.0528,
"mean_token_accuracy": 0.14546677693724633,
"num_tokens": 4888325.0,
"step": 2135
},
{
"entropy": 6.172981452941895,
"epoch": 0.21154606563859232,
"grad_norm": 1.0390625,
"learning_rate": 0.000499856400936514,
"loss": 6.0524,
"mean_token_accuracy": 0.14551248848438264,
"num_tokens": 4899631.0,
"step": 2140
},
{
"entropy": 6.245505332946777,
"epoch": 0.21204033214709372,
"grad_norm": 0.96875,
"learning_rate": 0.0004998551375580643,
"loss": 6.0628,
"mean_token_accuracy": 0.14277451261878013,
"num_tokens": 4910954.0,
"step": 2145
},
{
"entropy": 6.237064695358276,
"epoch": 0.2125345986555951,
"grad_norm": 0.96875,
"learning_rate": 0.0004998538686481435,
"loss": 6.0531,
"mean_token_accuracy": 0.14683079421520234,
"num_tokens": 4921952.0,
"step": 2150
},
{
"entropy": 6.203374338150025,
"epoch": 0.21302886516409647,
"grad_norm": 0.90625,
"learning_rate": 0.0004998525942067824,
"loss": 5.997,
"mean_token_accuracy": 0.15139593183994293,
"num_tokens": 4933984.0,
"step": 2155
},
{
"entropy": 6.283549213409424,
"epoch": 0.21352313167259787,
"grad_norm": 0.890625,
"learning_rate": 0.0004998513142340128,
"loss": 6.1433,
"mean_token_accuracy": 0.14827414751052856,
"num_tokens": 4945926.0,
"step": 2160
},
{
"entropy": 6.204256439208985,
"epoch": 0.21401739818109924,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998500287298657,
"loss": 6.0549,
"mean_token_accuracy": 0.15452619791030883,
"num_tokens": 4958123.0,
"step": 2165
},
{
"entropy": 6.319782829284668,
"epoch": 0.21451166468960065,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998487376943733,
"loss": 6.1369,
"mean_token_accuracy": 0.1432581551373005,
"num_tokens": 4970409.0,
"step": 2170
},
{
"entropy": 6.231132888793946,
"epoch": 0.21500593119810202,
"grad_norm": 1.0859375,
"learning_rate": 0.000499847441127567,
"loss": 6.0345,
"mean_token_accuracy": 0.1420576423406601,
"num_tokens": 4980429.0,
"step": 2175
},
{
"entropy": 6.213157796859742,
"epoch": 0.2155001977066034,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998461390294787,
"loss": 6.0472,
"mean_token_accuracy": 0.14993677660822868,
"num_tokens": 4991790.0,
"step": 2180
},
{
"entropy": 6.261713266372681,
"epoch": 0.2159944642151048,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998448314001405,
"loss": 6.069,
"mean_token_accuracy": 0.1435249201953411,
"num_tokens": 5003611.0,
"step": 2185
},
{
"entropy": 6.302479600906372,
"epoch": 0.21648873072360617,
"grad_norm": 1.0625,
"learning_rate": 0.0004998435182395845,
"loss": 6.1232,
"mean_token_accuracy": 0.13771253526210786,
"num_tokens": 5014824.0,
"step": 2190
},
{
"entropy": 6.258434915542603,
"epoch": 0.21698299723210757,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998421995478432,
"loss": 6.0768,
"mean_token_accuracy": 0.14596307873725892,
"num_tokens": 5026937.0,
"step": 2195
},
{
"entropy": 6.298146438598633,
"epoch": 0.21747726374060894,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998408753249489,
"loss": 6.0927,
"mean_token_accuracy": 0.14147113561630248,
"num_tokens": 5037996.0,
"step": 2200
},
{
"entropy": 6.2773766040802,
"epoch": 0.2179715302491103,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998395455709342,
"loss": 6.0422,
"mean_token_accuracy": 0.15115779191255568,
"num_tokens": 5050944.0,
"step": 2205
},
{
"entropy": 6.2505774974823,
"epoch": 0.2184657967576117,
"grad_norm": 1.0,
"learning_rate": 0.0004998382102858317,
"loss": 6.0871,
"mean_token_accuracy": 0.13831509053707122,
"num_tokens": 5062129.0,
"step": 2210
},
{
"entropy": 6.262403440475464,
"epoch": 0.2189600632661131,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998368694696743,
"loss": 6.0806,
"mean_token_accuracy": 0.1436243861913681,
"num_tokens": 5073318.0,
"step": 2215
},
{
"entropy": 6.185823059082031,
"epoch": 0.21945432977461446,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998355231224952,
"loss": 6.1047,
"mean_token_accuracy": 0.14238689020276069,
"num_tokens": 5084607.0,
"step": 2220
},
{
"entropy": 6.298590517044067,
"epoch": 0.21994859628311586,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998341712443273,
"loss": 6.0071,
"mean_token_accuracy": 0.1467602863907814,
"num_tokens": 5096004.0,
"step": 2225
},
{
"entropy": 6.199511909484864,
"epoch": 0.22044286279161723,
"grad_norm": 1.0078125,
"learning_rate": 0.000499832813835204,
"loss": 6.0052,
"mean_token_accuracy": 0.14269362017512321,
"num_tokens": 5107171.0,
"step": 2230
},
{
"entropy": 6.184848213195801,
"epoch": 0.22093712930011863,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998314508951584,
"loss": 5.9611,
"mean_token_accuracy": 0.15424952656030655,
"num_tokens": 5119249.0,
"step": 2235
},
{
"entropy": 6.14829421043396,
"epoch": 0.22143139580862,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998300824242244,
"loss": 5.976,
"mean_token_accuracy": 0.14679246842861177,
"num_tokens": 5130798.0,
"step": 2240
},
{
"entropy": 6.275025987625122,
"epoch": 0.22192566231712138,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998287084224353,
"loss": 6.1437,
"mean_token_accuracy": 0.13429683446884155,
"num_tokens": 5141376.0,
"step": 2245
},
{
"entropy": 6.154032039642334,
"epoch": 0.22241992882562278,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998273288898252,
"loss": 5.9357,
"mean_token_accuracy": 0.15065955817699433,
"num_tokens": 5152888.0,
"step": 2250
},
{
"entropy": 6.156353521347046,
"epoch": 0.22291419533412415,
"grad_norm": 0.91015625,
"learning_rate": 0.000499825943826428,
"loss": 6.0392,
"mean_token_accuracy": 0.14408660531044007,
"num_tokens": 5165720.0,
"step": 2255
},
{
"entropy": 6.161768770217895,
"epoch": 0.22340846184262556,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998245532322776,
"loss": 5.9084,
"mean_token_accuracy": 0.15456337854266167,
"num_tokens": 5177345.0,
"step": 2260
},
{
"entropy": 6.202211046218872,
"epoch": 0.22390272835112693,
"grad_norm": 1.046875,
"learning_rate": 0.0004998231571074083,
"loss": 5.9329,
"mean_token_accuracy": 0.1534419000148773,
"num_tokens": 5187385.0,
"step": 2265
},
{
"entropy": 6.170833683013916,
"epoch": 0.2243969948596283,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998217554518545,
"loss": 6.0403,
"mean_token_accuracy": 0.14258744418621064,
"num_tokens": 5199552.0,
"step": 2270
},
{
"entropy": 6.200662231445312,
"epoch": 0.2248912613681297,
"grad_norm": 0.96875,
"learning_rate": 0.0004998203482656507,
"loss": 6.007,
"mean_token_accuracy": 0.15123232156038285,
"num_tokens": 5210316.0,
"step": 2275
},
{
"entropy": 6.234782314300537,
"epoch": 0.22538552787663108,
"grad_norm": 1.046875,
"learning_rate": 0.0004998189355488314,
"loss": 6.0744,
"mean_token_accuracy": 0.14858508333563805,
"num_tokens": 5221595.0,
"step": 2280
},
{
"entropy": 6.227956914901734,
"epoch": 0.22587979438513248,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998175173014313,
"loss": 6.0504,
"mean_token_accuracy": 0.14020669236779212,
"num_tokens": 5232882.0,
"step": 2285
},
{
"entropy": 6.238587951660156,
"epoch": 0.22637406089363385,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998160935234857,
"loss": 6.1441,
"mean_token_accuracy": 0.14422304853796958,
"num_tokens": 5244841.0,
"step": 2290
},
{
"entropy": 6.146130466461182,
"epoch": 0.22686832740213522,
"grad_norm": 1.0390625,
"learning_rate": 0.000499814664215029,
"loss": 5.9581,
"mean_token_accuracy": 0.152134919911623,
"num_tokens": 5256162.0,
"step": 2295
},
{
"entropy": 6.190051794052124,
"epoch": 0.22736259391063662,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998132293760968,
"loss": 5.8762,
"mean_token_accuracy": 0.15613968968391417,
"num_tokens": 5267130.0,
"step": 2300
},
{
"entropy": 6.0447431087493895,
"epoch": 0.227856860419138,
"grad_norm": 0.875,
"learning_rate": 0.0004998117890067243,
"loss": 6.0008,
"mean_token_accuracy": 0.1474076971411705,
"num_tokens": 5279394.0,
"step": 2305
},
{
"entropy": 6.172190380096436,
"epoch": 0.22835112692763937,
"grad_norm": 0.96875,
"learning_rate": 0.0004998103431069469,
"loss": 5.9787,
"mean_token_accuracy": 0.14816260784864427,
"num_tokens": 5291068.0,
"step": 2310
},
{
"entropy": 6.2173741340637205,
"epoch": 0.22884539343614077,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998088916768002,
"loss": 5.9391,
"mean_token_accuracy": 0.15463171005249024,
"num_tokens": 5300924.0,
"step": 2315
},
{
"entropy": 6.124597406387329,
"epoch": 0.22933965994464214,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998074347163199,
"loss": 5.9419,
"mean_token_accuracy": 0.15559005588293076,
"num_tokens": 5311998.0,
"step": 2320
},
{
"entropy": 6.211737871170044,
"epoch": 0.22983392645314354,
"grad_norm": 0.921875,
"learning_rate": 0.0004998059722255418,
"loss": 5.9836,
"mean_token_accuracy": 0.14932580888271332,
"num_tokens": 5322163.0,
"step": 2325
},
{
"entropy": 6.127090263366699,
"epoch": 0.23032819296164492,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998045042045019,
"loss": 5.981,
"mean_token_accuracy": 0.15460044890642166,
"num_tokens": 5331638.0,
"step": 2330
},
{
"entropy": 6.200595951080322,
"epoch": 0.2308224594701463,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998030306532363,
"loss": 5.9814,
"mean_token_accuracy": 0.15659119933843613,
"num_tokens": 5342399.0,
"step": 2335
},
{
"entropy": 6.172941446304321,
"epoch": 0.2313167259786477,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998015515717813,
"loss": 6.028,
"mean_token_accuracy": 0.14542844295501708,
"num_tokens": 5353481.0,
"step": 2340
},
{
"entropy": 6.180023050308227,
"epoch": 0.23181099248714906,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998000669601733,
"loss": 5.9503,
"mean_token_accuracy": 0.1496099278330803,
"num_tokens": 5365167.0,
"step": 2345
},
{
"entropy": 6.114148902893066,
"epoch": 0.23230525899565047,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997985768184488,
"loss": 5.9658,
"mean_token_accuracy": 0.14984080642461778,
"num_tokens": 5375809.0,
"step": 2350
},
{
"entropy": 6.076729488372803,
"epoch": 0.23279952550415184,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997970811466443,
"loss": 5.8811,
"mean_token_accuracy": 0.15644187480211258,
"num_tokens": 5387372.0,
"step": 2355
},
{
"entropy": 6.341212034225464,
"epoch": 0.2332937920126532,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997955799447969,
"loss": 6.0607,
"mean_token_accuracy": 0.14222865626215936,
"num_tokens": 5399210.0,
"step": 2360
},
{
"entropy": 6.121503829956055,
"epoch": 0.2337880585211546,
"grad_norm": 0.96875,
"learning_rate": 0.0004997940732129432,
"loss": 5.9777,
"mean_token_accuracy": 0.14460844695568084,
"num_tokens": 5410969.0,
"step": 2365
},
{
"entropy": 6.139842796325683,
"epoch": 0.23428232502965599,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997925609511205,
"loss": 5.9347,
"mean_token_accuracy": 0.15759174525737762,
"num_tokens": 5422569.0,
"step": 2370
},
{
"entropy": 6.168006563186646,
"epoch": 0.2347765915381574,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997910431593659,
"loss": 5.9449,
"mean_token_accuracy": 0.15459632575511933,
"num_tokens": 5434592.0,
"step": 2375
},
{
"entropy": 6.1388246536254885,
"epoch": 0.23527085804665876,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997895198377168,
"loss": 6.0155,
"mean_token_accuracy": 0.15070839375257492,
"num_tokens": 5446525.0,
"step": 2380
},
{
"entropy": 6.147398662567139,
"epoch": 0.23576512455516013,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997879909862105,
"loss": 6.0029,
"mean_token_accuracy": 0.14832279682159424,
"num_tokens": 5457327.0,
"step": 2385
},
{
"entropy": 6.24454779624939,
"epoch": 0.23625939106366153,
"grad_norm": 0.97265625,
"learning_rate": 0.000499786456604885,
"loss": 6.0312,
"mean_token_accuracy": 0.14286630377173423,
"num_tokens": 5468533.0,
"step": 2390
},
{
"entropy": 6.117534255981445,
"epoch": 0.2367536575721629,
"grad_norm": 0.82421875,
"learning_rate": 0.0004997849166937776,
"loss": 5.9146,
"mean_token_accuracy": 0.14934935867786409,
"num_tokens": 5480652.0,
"step": 2395
},
{
"entropy": 6.200796270370484,
"epoch": 0.2372479240806643,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997833712529265,
"loss": 5.9595,
"mean_token_accuracy": 0.14828556329011916,
"num_tokens": 5491446.0,
"step": 2400
},
{
"entropy": 6.11882472038269,
"epoch": 0.23774219058916568,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997818202823695,
"loss": 5.9543,
"mean_token_accuracy": 0.1541999250650406,
"num_tokens": 5502886.0,
"step": 2405
},
{
"entropy": 6.157952308654785,
"epoch": 0.23823645709766705,
"grad_norm": 1.03125,
"learning_rate": 0.0004997802637821449,
"loss": 5.9933,
"mean_token_accuracy": 0.14664816558361055,
"num_tokens": 5514305.0,
"step": 2410
},
{
"entropy": 6.153297376632691,
"epoch": 0.23873072360616845,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997787017522909,
"loss": 5.9195,
"mean_token_accuracy": 0.16202860027551652,
"num_tokens": 5525913.0,
"step": 2415
},
{
"entropy": 6.031432867050171,
"epoch": 0.23922499011466983,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997771341928459,
"loss": 5.9028,
"mean_token_accuracy": 0.15993187874555587,
"num_tokens": 5537069.0,
"step": 2420
},
{
"entropy": 6.132914733886719,
"epoch": 0.2397192566231712,
"grad_norm": 1.09375,
"learning_rate": 0.0004997755611038487,
"loss": 6.0155,
"mean_token_accuracy": 0.1475693963468075,
"num_tokens": 5547988.0,
"step": 2425
},
{
"entropy": 6.177093124389648,
"epoch": 0.2402135231316726,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997739824853378,
"loss": 6.0784,
"mean_token_accuracy": 0.14507206082344054,
"num_tokens": 5560400.0,
"step": 2430
},
{
"entropy": 6.125483798980713,
"epoch": 0.24070778964017397,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997723983373519,
"loss": 5.9408,
"mean_token_accuracy": 0.1547750160098076,
"num_tokens": 5572291.0,
"step": 2435
},
{
"entropy": 6.23362283706665,
"epoch": 0.24120205614867538,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997708086599303,
"loss": 5.9657,
"mean_token_accuracy": 0.14544104561209678,
"num_tokens": 5583583.0,
"step": 2440
},
{
"entropy": 6.20045838356018,
"epoch": 0.24169632265717675,
"grad_norm": 0.84765625,
"learning_rate": 0.0004997692134531119,
"loss": 5.9888,
"mean_token_accuracy": 0.1482198178768158,
"num_tokens": 5595499.0,
"step": 2445
},
{
"entropy": 6.0919482707977295,
"epoch": 0.24219058916567812,
"grad_norm": 0.984375,
"learning_rate": 0.000499767612716936,
"loss": 5.9138,
"mean_token_accuracy": 0.1550483450293541,
"num_tokens": 5606745.0,
"step": 2450
},
{
"entropy": 6.092108058929443,
"epoch": 0.24268485567417952,
"grad_norm": 1.0,
"learning_rate": 0.0004997660064514419,
"loss": 5.9935,
"mean_token_accuracy": 0.15098343044519424,
"num_tokens": 5618226.0,
"step": 2455
},
{
"entropy": 6.259077262878418,
"epoch": 0.2431791221826809,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997643946566691,
"loss": 6.0146,
"mean_token_accuracy": 0.14846471548080445,
"num_tokens": 5628941.0,
"step": 2460
},
{
"entropy": 6.092586326599121,
"epoch": 0.2436733886911823,
"grad_norm": 0.984375,
"learning_rate": 0.0004997627773326574,
"loss": 5.8118,
"mean_token_accuracy": 0.1619025096297264,
"num_tokens": 5641077.0,
"step": 2465
},
{
"entropy": 6.1263285160064695,
"epoch": 0.24416765519968367,
"grad_norm": 1.109375,
"learning_rate": 0.0004997611544794465,
"loss": 6.1202,
"mean_token_accuracy": 0.14339257180690765,
"num_tokens": 5652156.0,
"step": 2470
},
{
"entropy": 6.257670974731445,
"epoch": 0.24466192170818504,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997595260970764,
"loss": 6.0358,
"mean_token_accuracy": 0.14745921045541763,
"num_tokens": 5663988.0,
"step": 2475
},
{
"entropy": 6.131538438796997,
"epoch": 0.24515618821668644,
"grad_norm": 0.86328125,
"learning_rate": 0.000499757892185587,
"loss": 5.9627,
"mean_token_accuracy": 0.14847193360328675,
"num_tokens": 5676302.0,
"step": 2480
},
{
"entropy": 6.159386014938354,
"epoch": 0.24565045472518782,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997562527450187,
"loss": 5.9976,
"mean_token_accuracy": 0.15123492628335952,
"num_tokens": 5686686.0,
"step": 2485
},
{
"entropy": 6.16356930732727,
"epoch": 0.24614472123368922,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997546077754116,
"loss": 5.9342,
"mean_token_accuracy": 0.16507403403520585,
"num_tokens": 5698379.0,
"step": 2490
},
{
"entropy": 6.121092033386231,
"epoch": 0.2466389877421906,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997529572768063,
"loss": 5.9167,
"mean_token_accuracy": 0.154787477850914,
"num_tokens": 5708891.0,
"step": 2495
},
{
"entropy": 6.0729203701019285,
"epoch": 0.24713325425069196,
"grad_norm": 1.03125,
"learning_rate": 0.0004997513012492434,
"loss": 5.9648,
"mean_token_accuracy": 0.1536966696381569,
"num_tokens": 5720326.0,
"step": 2500
},
{
"entropy": 6.153505516052246,
"epoch": 0.24762752075919336,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997496396927636,
"loss": 5.8959,
"mean_token_accuracy": 0.15583585798740388,
"num_tokens": 5731121.0,
"step": 2505
},
{
"entropy": 6.155553531646729,
"epoch": 0.24812178726769474,
"grad_norm": 1.078125,
"learning_rate": 0.0004997479726074077,
"loss": 5.93,
"mean_token_accuracy": 0.15216445177793503,
"num_tokens": 5741637.0,
"step": 2510
},
{
"entropy": 6.1915891647338865,
"epoch": 0.24861605377619614,
"grad_norm": 0.96875,
"learning_rate": 0.000499746299993217,
"loss": 5.9735,
"mean_token_accuracy": 0.15114908963441848,
"num_tokens": 5752659.0,
"step": 2515
},
{
"entropy": 6.094469976425171,
"epoch": 0.2491103202846975,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997446218502324,
"loss": 5.9525,
"mean_token_accuracy": 0.15155530497431755,
"num_tokens": 5763829.0,
"step": 2520
},
{
"entropy": 6.132277250289917,
"epoch": 0.24960458679319888,
"grad_norm": 0.953125,
"learning_rate": 0.0004997429381784952,
"loss": 6.022,
"mean_token_accuracy": 0.14910133332014083,
"num_tokens": 5775848.0,
"step": 2525
},
{
"entropy": 6.120943069458008,
"epoch": 0.2500988533017003,
"grad_norm": 1.046875,
"learning_rate": 0.0004997412489780469,
"loss": 5.9612,
"mean_token_accuracy": 0.15950672328472137,
"num_tokens": 5786595.0,
"step": 2530
},
{
"entropy": 6.140192317962646,
"epoch": 0.25059311981020166,
"grad_norm": 1.0390625,
"learning_rate": 0.000499739554248929,
"loss": 5.8703,
"mean_token_accuracy": 0.16149618923664094,
"num_tokens": 5797841.0,
"step": 2535
},
{
"entropy": 6.0338222026824955,
"epoch": 0.25108738631870303,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997378539911832,
"loss": 5.8765,
"mean_token_accuracy": 0.15623045712709427,
"num_tokens": 5809106.0,
"step": 2540
},
{
"entropy": 6.107133483886718,
"epoch": 0.2515816528272044,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997361482048513,
"loss": 5.8885,
"mean_token_accuracy": 0.15565961226820946,
"num_tokens": 5820316.0,
"step": 2545
},
{
"entropy": 6.121990442276001,
"epoch": 0.25207591933570583,
"grad_norm": 0.921875,
"learning_rate": 0.0004997344368899753,
"loss": 6.02,
"mean_token_accuracy": 0.14788583666086197,
"num_tokens": 5833137.0,
"step": 2550
},
{
"entropy": 6.175726079940796,
"epoch": 0.2525701858442072,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997327200465972,
"loss": 5.9304,
"mean_token_accuracy": 0.15318210422992706,
"num_tokens": 5844088.0,
"step": 2555
},
{
"entropy": 6.133507585525512,
"epoch": 0.2530644523527086,
"grad_norm": 0.88671875,
"learning_rate": 0.0004997309976747594,
"loss": 5.9265,
"mean_token_accuracy": 0.15268272608518602,
"num_tokens": 5854546.0,
"step": 2560
},
{
"entropy": 6.149356508255005,
"epoch": 0.25355871886120995,
"grad_norm": 1.109375,
"learning_rate": 0.0004997292697745042,
"loss": 5.9255,
"mean_token_accuracy": 0.1569249227643013,
"num_tokens": 5865326.0,
"step": 2565
},
{
"entropy": 6.136588287353516,
"epoch": 0.2540529853697113,
"grad_norm": 0.9375,
"learning_rate": 0.0004997275363458741,
"loss": 5.8293,
"mean_token_accuracy": 0.15652241110801696,
"num_tokens": 5876487.0,
"step": 2570
},
{
"entropy": 6.0281952857971195,
"epoch": 0.25454725187821275,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997257973889118,
"loss": 5.9355,
"mean_token_accuracy": 0.14910998791456223,
"num_tokens": 5887846.0,
"step": 2575
},
{
"entropy": 6.157446002960205,
"epoch": 0.25504151838671413,
"grad_norm": 1.03125,
"learning_rate": 0.0004997240529036599,
"loss": 5.8919,
"mean_token_accuracy": 0.15892206728458405,
"num_tokens": 5898572.0,
"step": 2580
},
{
"entropy": 6.071728754043579,
"epoch": 0.2555357848952155,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997223028901615,
"loss": 5.9449,
"mean_token_accuracy": 0.15322234779596328,
"num_tokens": 5908927.0,
"step": 2585
},
{
"entropy": 6.124873542785645,
"epoch": 0.2560300514037169,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997205473484596,
"loss": 5.9037,
"mean_token_accuracy": 0.1580432742834091,
"num_tokens": 5919807.0,
"step": 2590
},
{
"entropy": 6.10838508605957,
"epoch": 0.25652431791221825,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997187862785974,
"loss": 5.9259,
"mean_token_accuracy": 0.15537415742874144,
"num_tokens": 5931160.0,
"step": 2595
},
{
"entropy": 6.13609881401062,
"epoch": 0.2570185844207197,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997170196806181,
"loss": 5.9987,
"mean_token_accuracy": 0.14711610227823257,
"num_tokens": 5943694.0,
"step": 2600
},
{
"entropy": 6.120221710205078,
"epoch": 0.25751285092922105,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997152475545652,
"loss": 5.8863,
"mean_token_accuracy": 0.15604913979768753,
"num_tokens": 5954543.0,
"step": 2605
},
{
"entropy": 6.046738481521606,
"epoch": 0.2580071174377224,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997134699004825,
"loss": 5.9358,
"mean_token_accuracy": 0.15628221184015273,
"num_tokens": 5965080.0,
"step": 2610
},
{
"entropy": 6.095473432540894,
"epoch": 0.2585013839462238,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997116867184136,
"loss": 5.937,
"mean_token_accuracy": 0.14930862933397293,
"num_tokens": 5975314.0,
"step": 2615
},
{
"entropy": 6.180641603469849,
"epoch": 0.25899565045472517,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997098980084022,
"loss": 5.9354,
"mean_token_accuracy": 0.15294832438230516,
"num_tokens": 5985927.0,
"step": 2620
},
{
"entropy": 6.087760400772095,
"epoch": 0.2594899169632266,
"grad_norm": 1.0625,
"learning_rate": 0.0004997081037704926,
"loss": 5.9517,
"mean_token_accuracy": 0.15042607337236405,
"num_tokens": 5997311.0,
"step": 2625
},
{
"entropy": 6.086427164077759,
"epoch": 0.25998418347172797,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997063040047288,
"loss": 5.9843,
"mean_token_accuracy": 0.1468744695186615,
"num_tokens": 6008595.0,
"step": 2630
},
{
"entropy": 6.178718137741089,
"epoch": 0.26047844998022934,
"grad_norm": 1.109375,
"learning_rate": 0.0004997044987111548,
"loss": 5.9078,
"mean_token_accuracy": 0.1546178936958313,
"num_tokens": 6019055.0,
"step": 2635
},
{
"entropy": 6.0701744079589846,
"epoch": 0.2609727164887307,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997026878898155,
"loss": 6.0342,
"mean_token_accuracy": 0.14548355862498283,
"num_tokens": 6030029.0,
"step": 2640
},
{
"entropy": 6.153164005279541,
"epoch": 0.2614669829972321,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997008715407552,
"loss": 5.9902,
"mean_token_accuracy": 0.1496541254222393,
"num_tokens": 6041495.0,
"step": 2645
},
{
"entropy": 6.143697834014892,
"epoch": 0.2619612495057335,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996990496640185,
"loss": 5.8818,
"mean_token_accuracy": 0.15100248157978058,
"num_tokens": 6052307.0,
"step": 2650
},
{
"entropy": 6.028233671188355,
"epoch": 0.2624555160142349,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996972222596505,
"loss": 5.8423,
"mean_token_accuracy": 0.15583381801843643,
"num_tokens": 6062550.0,
"step": 2655
},
{
"entropy": 6.0477863311767575,
"epoch": 0.26294978252273626,
"grad_norm": 0.953125,
"learning_rate": 0.0004996953893276958,
"loss": 5.8781,
"mean_token_accuracy": 0.14674556404352188,
"num_tokens": 6074450.0,
"step": 2660
},
{
"entropy": 6.078070211410522,
"epoch": 0.26344404903123764,
"grad_norm": 0.83984375,
"learning_rate": 0.0004996935508681997,
"loss": 5.9265,
"mean_token_accuracy": 0.1591618239879608,
"num_tokens": 6085749.0,
"step": 2665
},
{
"entropy": 6.09918999671936,
"epoch": 0.263938315539739,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996917068812075,
"loss": 5.8488,
"mean_token_accuracy": 0.15992971658706664,
"num_tokens": 6096438.0,
"step": 2670
},
{
"entropy": 6.059639120101929,
"epoch": 0.26443258204824044,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996898573667643,
"loss": 5.9086,
"mean_token_accuracy": 0.14664011746644973,
"num_tokens": 6108386.0,
"step": 2675
},
{
"entropy": 6.083099842071533,
"epoch": 0.2649268485567418,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996880023249158,
"loss": 5.8096,
"mean_token_accuracy": 0.16396086513996125,
"num_tokens": 6118771.0,
"step": 2680
},
{
"entropy": 6.0931103229522705,
"epoch": 0.2654211150652432,
"grad_norm": 0.9375,
"learning_rate": 0.0004996861417557076,
"loss": 5.9009,
"mean_token_accuracy": 0.15202439576387405,
"num_tokens": 6130718.0,
"step": 2685
},
{
"entropy": 6.067891788482666,
"epoch": 0.26591538157374456,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996842756591855,
"loss": 5.9283,
"mean_token_accuracy": 0.15080306231975554,
"num_tokens": 6142086.0,
"step": 2690
},
{
"entropy": 6.042020416259765,
"epoch": 0.26640964808224593,
"grad_norm": 0.8671875,
"learning_rate": 0.0004996824040353953,
"loss": 5.904,
"mean_token_accuracy": 0.1560404196381569,
"num_tokens": 6153857.0,
"step": 2695
},
{
"entropy": 6.241527795791626,
"epoch": 0.2669039145907473,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996805268843832,
"loss": 6.0421,
"mean_token_accuracy": 0.14912690445780755,
"num_tokens": 6166796.0,
"step": 2700
},
{
"entropy": 6.072644186019898,
"epoch": 0.26739818109924873,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996786442061952,
"loss": 5.8597,
"mean_token_accuracy": 0.15616288185119628,
"num_tokens": 6178046.0,
"step": 2705
},
{
"entropy": 6.0524310111999515,
"epoch": 0.2678924476077501,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996767560008777,
"loss": 5.9115,
"mean_token_accuracy": 0.1550481289625168,
"num_tokens": 6190455.0,
"step": 2710
},
{
"entropy": 6.06898980140686,
"epoch": 0.2683867141162515,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996748622684771,
"loss": 5.8107,
"mean_token_accuracy": 0.1622350811958313,
"num_tokens": 6201659.0,
"step": 2715
},
{
"entropy": 5.995411539077759,
"epoch": 0.26888098062475285,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996729630090401,
"loss": 5.9163,
"mean_token_accuracy": 0.14895097762346268,
"num_tokens": 6213870.0,
"step": 2720
},
{
"entropy": 6.167082738876343,
"epoch": 0.2693752471332542,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996710582226133,
"loss": 5.8605,
"mean_token_accuracy": 0.1600447803735733,
"num_tokens": 6224237.0,
"step": 2725
},
{
"entropy": 6.082718801498413,
"epoch": 0.26986951364175565,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996691479092437,
"loss": 5.9085,
"mean_token_accuracy": 0.15194753110408782,
"num_tokens": 6235868.0,
"step": 2730
},
{
"entropy": 6.102431201934815,
"epoch": 0.270363780150257,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996672320689782,
"loss": 5.9515,
"mean_token_accuracy": 0.14958198070526124,
"num_tokens": 6247994.0,
"step": 2735
},
{
"entropy": 5.958840847015381,
"epoch": 0.2708580466587584,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996653107018638,
"loss": 5.8104,
"mean_token_accuracy": 0.16383833289146424,
"num_tokens": 6259654.0,
"step": 2740
},
{
"entropy": 6.046971559524536,
"epoch": 0.2713523131672598,
"grad_norm": 0.90625,
"learning_rate": 0.000499663383807948,
"loss": 5.8601,
"mean_token_accuracy": 0.1581733912229538,
"num_tokens": 6270582.0,
"step": 2745
},
{
"entropy": 6.1495075702667235,
"epoch": 0.27184657967576115,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996614513872781,
"loss": 5.8536,
"mean_token_accuracy": 0.1616063266992569,
"num_tokens": 6281088.0,
"step": 2750
},
{
"entropy": 6.113773393630981,
"epoch": 0.2723408461842626,
"grad_norm": 1.015625,
"learning_rate": 0.0004996595134399017,
"loss": 6.0061,
"mean_token_accuracy": 0.1472685419023037,
"num_tokens": 6292920.0,
"step": 2755
},
{
"entropy": 6.09270396232605,
"epoch": 0.27283511269276395,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996575699658664,
"loss": 5.8642,
"mean_token_accuracy": 0.1566832423210144,
"num_tokens": 6304980.0,
"step": 2760
},
{
"entropy": 6.006596946716309,
"epoch": 0.2733293792012653,
"grad_norm": 0.9375,
"learning_rate": 0.0004996556209652199,
"loss": 5.8635,
"mean_token_accuracy": 0.1554800748825073,
"num_tokens": 6316175.0,
"step": 2765
},
{
"entropy": 6.153335475921631,
"epoch": 0.2738236457097667,
"grad_norm": 0.9375,
"learning_rate": 0.0004996536664380104,
"loss": 5.9069,
"mean_token_accuracy": 0.1567239671945572,
"num_tokens": 6327590.0,
"step": 2770
},
{
"entropy": 6.017695617675781,
"epoch": 0.27431791221826807,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996517063842859,
"loss": 5.9336,
"mean_token_accuracy": 0.15392989069223403,
"num_tokens": 6339066.0,
"step": 2775
},
{
"entropy": 6.086386060714721,
"epoch": 0.2748121787267695,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996497408040944,
"loss": 5.8555,
"mean_token_accuracy": 0.15775285810232162,
"num_tokens": 6349485.0,
"step": 2780
},
{
"entropy": 5.981375408172608,
"epoch": 0.27530644523527087,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996477696974845,
"loss": 5.8113,
"mean_token_accuracy": 0.15953710824251174,
"num_tokens": 6360775.0,
"step": 2785
},
{
"entropy": 6.034738063812256,
"epoch": 0.27580071174377224,
"grad_norm": 0.96875,
"learning_rate": 0.0004996457930645046,
"loss": 5.8281,
"mean_token_accuracy": 0.15122711658477783,
"num_tokens": 6371645.0,
"step": 2790
},
{
"entropy": 6.013451719284058,
"epoch": 0.2762949782522736,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996438109052034,
"loss": 5.8573,
"mean_token_accuracy": 0.15728358179330826,
"num_tokens": 6383877.0,
"step": 2795
},
{
"entropy": 6.08137354850769,
"epoch": 0.276789244760775,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996418232196296,
"loss": 5.9411,
"mean_token_accuracy": 0.1478685200214386,
"num_tokens": 6394488.0,
"step": 2800
},
{
"entropy": 6.082438802719116,
"epoch": 0.2772835112692764,
"grad_norm": 0.921875,
"learning_rate": 0.0004996398300078321,
"loss": 5.8731,
"mean_token_accuracy": 0.1562285304069519,
"num_tokens": 6404948.0,
"step": 2805
},
{
"entropy": 6.112468290328979,
"epoch": 0.2777777777777778,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996378312698598,
"loss": 5.8919,
"mean_token_accuracy": 0.15620481222867966,
"num_tokens": 6416273.0,
"step": 2810
},
{
"entropy": 6.089010572433471,
"epoch": 0.27827204428627916,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996358270057622,
"loss": 5.9114,
"mean_token_accuracy": 0.15275675654411316,
"num_tokens": 6427091.0,
"step": 2815
},
{
"entropy": 6.05656418800354,
"epoch": 0.27876631079478054,
"grad_norm": 0.96875,
"learning_rate": 0.0004996338172155883,
"loss": 5.8618,
"mean_token_accuracy": 0.15593931525945665,
"num_tokens": 6438254.0,
"step": 2820
},
{
"entropy": 6.018289232254029,
"epoch": 0.2792605773032819,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996318018993876,
"loss": 5.7876,
"mean_token_accuracy": 0.15474870353937148,
"num_tokens": 6450618.0,
"step": 2825
},
{
"entropy": 6.027504634857178,
"epoch": 0.27975484381178334,
"grad_norm": 0.875,
"learning_rate": 0.0004996297810572098,
"loss": 5.8482,
"mean_token_accuracy": 0.1601539134979248,
"num_tokens": 6462467.0,
"step": 2830
},
{
"entropy": 6.017673635482788,
"epoch": 0.2802491103202847,
"grad_norm": 0.859375,
"learning_rate": 0.0004996277546891045,
"loss": 5.819,
"mean_token_accuracy": 0.1652624100446701,
"num_tokens": 6474206.0,
"step": 2835
},
{
"entropy": 6.056855010986328,
"epoch": 0.2807433768287861,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996257227951216,
"loss": 5.9084,
"mean_token_accuracy": 0.15035663545131683,
"num_tokens": 6486405.0,
"step": 2840
},
{
"entropy": 6.039477729797364,
"epoch": 0.28123764333728746,
"grad_norm": 1.0078125,
"learning_rate": 0.000499623685375311,
"loss": 5.8714,
"mean_token_accuracy": 0.15469448566436766,
"num_tokens": 6496949.0,
"step": 2845
},
{
"entropy": 6.027335119247437,
"epoch": 0.28173190984578883,
"grad_norm": 0.94140625,
"learning_rate": 0.000499621642429723,
"loss": 5.812,
"mean_token_accuracy": 0.16200096756219864,
"num_tokens": 6508913.0,
"step": 2850
},
{
"entropy": 6.040955018997193,
"epoch": 0.28222617635429026,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996195939584078,
"loss": 5.8639,
"mean_token_accuracy": 0.1561881922185421,
"num_tokens": 6519837.0,
"step": 2855
},
{
"entropy": 6.054210805892945,
"epoch": 0.28272044286279163,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996175399614156,
"loss": 5.8126,
"mean_token_accuracy": 0.1579636812210083,
"num_tokens": 6531409.0,
"step": 2860
},
{
"entropy": 5.995057392120361,
"epoch": 0.283214709371293,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996154804387972,
"loss": 5.8556,
"mean_token_accuracy": 0.15890285670757293,
"num_tokens": 6543433.0,
"step": 2865
},
{
"entropy": 6.053872203826904,
"epoch": 0.2837089758797944,
"grad_norm": 1.0,
"learning_rate": 0.000499613415390603,
"loss": 5.9406,
"mean_token_accuracy": 0.151950404047966,
"num_tokens": 6554511.0,
"step": 2870
},
{
"entropy": 6.0932821273803714,
"epoch": 0.28420324238829575,
"grad_norm": 1.0,
"learning_rate": 0.000499611344816884,
"loss": 5.8527,
"mean_token_accuracy": 0.1596566081047058,
"num_tokens": 6564591.0,
"step": 2875
},
{
"entropy": 6.049549293518067,
"epoch": 0.2846975088967972,
"grad_norm": 0.94140625,
"learning_rate": 0.000499609268717691,
"loss": 5.8486,
"mean_token_accuracy": 0.15844885110855103,
"num_tokens": 6575684.0,
"step": 2880
},
{
"entropy": 6.073301649093628,
"epoch": 0.28519177540529855,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996071870930753,
"loss": 5.9619,
"mean_token_accuracy": 0.15451340526342391,
"num_tokens": 6588504.0,
"step": 2885
},
{
"entropy": 5.963142967224121,
"epoch": 0.2856860419137999,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996050999430878,
"loss": 5.7836,
"mean_token_accuracy": 0.15867630690336226,
"num_tokens": 6599969.0,
"step": 2890
},
{
"entropy": 6.050960683822632,
"epoch": 0.2861803084223013,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996030072677802,
"loss": 5.8486,
"mean_token_accuracy": 0.163725782930851,
"num_tokens": 6611430.0,
"step": 2895
},
{
"entropy": 6.036673974990845,
"epoch": 0.2866745749308027,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996009090672037,
"loss": 5.8447,
"mean_token_accuracy": 0.15628752261400222,
"num_tokens": 6623247.0,
"step": 2900
},
{
"entropy": 6.034439992904663,
"epoch": 0.28716884143930405,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995988053414098,
"loss": 5.9669,
"mean_token_accuracy": 0.15123033076524733,
"num_tokens": 6635508.0,
"step": 2905
},
{
"entropy": 6.039534950256348,
"epoch": 0.2876631079478055,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995966960904506,
"loss": 5.8779,
"mean_token_accuracy": 0.15228607207536698,
"num_tokens": 6646020.0,
"step": 2910
},
{
"entropy": 6.089818906784058,
"epoch": 0.28815737445630685,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995945813143778,
"loss": 5.9051,
"mean_token_accuracy": 0.14688372910022734,
"num_tokens": 6658194.0,
"step": 2915
},
{
"entropy": 6.052178907394409,
"epoch": 0.2886516409648082,
"grad_norm": 0.9453125,
"learning_rate": 0.0004995924610132435,
"loss": 5.8738,
"mean_token_accuracy": 0.15363554805517196,
"num_tokens": 6670582.0,
"step": 2920
},
{
"entropy": 6.036479187011719,
"epoch": 0.2891459074733096,
"grad_norm": 1.03125,
"learning_rate": 0.0004995903351870998,
"loss": 5.8224,
"mean_token_accuracy": 0.1569667100906372,
"num_tokens": 6680836.0,
"step": 2925
},
{
"entropy": 6.112790536880493,
"epoch": 0.28964017398181097,
"grad_norm": 0.95703125,
"learning_rate": 0.000499588203835999,
"loss": 5.9134,
"mean_token_accuracy": 0.14992318898439408,
"num_tokens": 6693045.0,
"step": 2930
},
{
"entropy": 6.067016649246216,
"epoch": 0.2901344404903124,
"grad_norm": 1.046875,
"learning_rate": 0.0004995860669599935,
"loss": 5.8637,
"mean_token_accuracy": 0.1490854561328888,
"num_tokens": 6704006.0,
"step": 2935
},
{
"entropy": 5.941438627243042,
"epoch": 0.29062870699881377,
"grad_norm": 0.89453125,
"learning_rate": 0.000499583924559136,
"loss": 5.8258,
"mean_token_accuracy": 0.15538810789585114,
"num_tokens": 6717027.0,
"step": 2940
},
{
"entropy": 6.075238943099976,
"epoch": 0.29112297350731514,
"grad_norm": 0.9765625,
"learning_rate": 0.000499581776633479,
"loss": 5.8174,
"mean_token_accuracy": 0.1584622010588646,
"num_tokens": 6728107.0,
"step": 2945
},
{
"entropy": 5.981333208084107,
"epoch": 0.2916172400158165,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995796231830756,
"loss": 5.8421,
"mean_token_accuracy": 0.15783749520778656,
"num_tokens": 6738628.0,
"step": 2950
},
{
"entropy": 5.984621667861939,
"epoch": 0.2921115065243179,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995774642079786,
"loss": 5.7884,
"mean_token_accuracy": 0.1646315023303032,
"num_tokens": 6749416.0,
"step": 2955
},
{
"entropy": 5.985525369644165,
"epoch": 0.2926057730328193,
"grad_norm": 0.83984375,
"learning_rate": 0.000499575299708241,
"loss": 5.8929,
"mean_token_accuracy": 0.1488107793033123,
"num_tokens": 6761663.0,
"step": 2960
},
{
"entropy": 6.097906303405762,
"epoch": 0.2931000395413207,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995731296839163,
"loss": 5.8395,
"mean_token_accuracy": 0.15889886170625686,
"num_tokens": 6773557.0,
"step": 2965
},
{
"entropy": 5.906320524215698,
"epoch": 0.29359430604982206,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995709541350578,
"loss": 5.7589,
"mean_token_accuracy": 0.16536967307329178,
"num_tokens": 6782832.0,
"step": 2970
},
{
"entropy": 6.064802503585815,
"epoch": 0.29408857255832344,
"grad_norm": 0.9375,
"learning_rate": 0.000499568773061719,
"loss": 5.882,
"mean_token_accuracy": 0.15133094489574433,
"num_tokens": 6795165.0,
"step": 2975
},
{
"entropy": 6.07623519897461,
"epoch": 0.2945828390668248,
"grad_norm": 0.921875,
"learning_rate": 0.0004995665864639534,
"loss": 5.9023,
"mean_token_accuracy": 0.15480878055095673,
"num_tokens": 6807054.0,
"step": 2980
},
{
"entropy": 5.915165376663208,
"epoch": 0.29507710557532624,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995643943418149,
"loss": 5.6952,
"mean_token_accuracy": 0.16479281187057496,
"num_tokens": 6817681.0,
"step": 2985
},
{
"entropy": 6.030046367645264,
"epoch": 0.2955713720838276,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995621966953576,
"loss": 5.8628,
"mean_token_accuracy": 0.15873999148607254,
"num_tokens": 6829387.0,
"step": 2990
},
{
"entropy": 5.994547605514526,
"epoch": 0.296065638592329,
"grad_norm": 0.8046875,
"learning_rate": 0.0004995599935246355,
"loss": 5.8458,
"mean_token_accuracy": 0.15944462567567824,
"num_tokens": 6842023.0,
"step": 2995
},
{
"entropy": 6.006680917739868,
"epoch": 0.29655990510083036,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995577848297026,
"loss": 5.8521,
"mean_token_accuracy": 0.1528011903166771,
"num_tokens": 6853189.0,
"step": 3000
},
{
"epoch": 0.29655990510083036,
"eval_entropy": 5.838898362913988,
"eval_loss": 5.886573314666748,
"eval_mean_token_accuracy": 0.16203675489657332,
"eval_num_tokens": 6853189.0,
"eval_runtime": 20.7287,
"eval_samples_per_second": 1568.503,
"eval_steps_per_second": 196.105,
"step": 3000
},
{
"entropy": 6.084212350845337,
"epoch": 0.29705417160933173,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995555706106134,
"loss": 5.8966,
"mean_token_accuracy": 0.1488155320286751,
"num_tokens": 6864338.0,
"step": 3005
},
{
"entropy": 5.997887516021729,
"epoch": 0.29754843811783316,
"grad_norm": 1.0625,
"learning_rate": 0.0004995533508674223,
"loss": 5.784,
"mean_token_accuracy": 0.16026814579963683,
"num_tokens": 6875409.0,
"step": 3010
},
{
"entropy": 6.001093864440918,
"epoch": 0.29804270462633453,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995511256001839,
"loss": 5.8172,
"mean_token_accuracy": 0.16170514672994613,
"num_tokens": 6885757.0,
"step": 3015
},
{
"entropy": 6.015048027038574,
"epoch": 0.2985369711348359,
"grad_norm": 1.0078125,
"learning_rate": 0.000499548894808953,
"loss": 5.8598,
"mean_token_accuracy": 0.1570695921778679,
"num_tokens": 6896440.0,
"step": 3020
},
{
"entropy": 6.0726926803588865,
"epoch": 0.2990312376433373,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995466584937846,
"loss": 5.876,
"mean_token_accuracy": 0.1561363235116005,
"num_tokens": 6907319.0,
"step": 3025
},
{
"entropy": 6.034220218658447,
"epoch": 0.29952550415183865,
"grad_norm": 0.96875,
"learning_rate": 0.0004995444166547333,
"loss": 5.8326,
"mean_token_accuracy": 0.15464108437299728,
"num_tokens": 6918683.0,
"step": 3030
},
{
"entropy": 5.9634346008300785,
"epoch": 0.3000197706603401,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995421692918547,
"loss": 5.8091,
"mean_token_accuracy": 0.15337479412555693,
"num_tokens": 6929974.0,
"step": 3035
},
{
"entropy": 6.0475657939910885,
"epoch": 0.30051403716884145,
"grad_norm": 0.9140625,
"learning_rate": 0.0004995399164052038,
"loss": 5.8107,
"mean_token_accuracy": 0.15774545818567276,
"num_tokens": 6941213.0,
"step": 3040
},
{
"entropy": 6.0742419242858885,
"epoch": 0.3010083036773428,
"grad_norm": 1.0,
"learning_rate": 0.0004995376579948362,
"loss": 5.8655,
"mean_token_accuracy": 0.15712235271930694,
"num_tokens": 6952880.0,
"step": 3045
},
{
"entropy": 6.066303396224976,
"epoch": 0.3015025701858442,
"grad_norm": 0.875,
"learning_rate": 0.0004995353940608074,
"loss": 5.8468,
"mean_token_accuracy": 0.15944461822509765,
"num_tokens": 6964344.0,
"step": 3050
},
{
"entropy": 5.925079202651977,
"epoch": 0.30199683669434557,
"grad_norm": 0.9140625,
"learning_rate": 0.000499533124603173,
"loss": 5.8007,
"mean_token_accuracy": 0.1586202546954155,
"num_tokens": 6976042.0,
"step": 3055
},
{
"entropy": 6.026428174972534,
"epoch": 0.302491103202847,
"grad_norm": 0.90234375,
"learning_rate": 0.000499530849621989,
"loss": 5.7648,
"mean_token_accuracy": 0.16500493586063386,
"num_tokens": 6988137.0,
"step": 3060
},
{
"entropy": 5.921867370605469,
"epoch": 0.3029853697113484,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995285691173111,
"loss": 5.799,
"mean_token_accuracy": 0.15778796151280403,
"num_tokens": 6999858.0,
"step": 3065
},
{
"entropy": 6.067946720123291,
"epoch": 0.30347963621984975,
"grad_norm": 0.84765625,
"learning_rate": 0.0004995262830891957,
"loss": 5.8567,
"mean_token_accuracy": 0.1585906222462654,
"num_tokens": 7012756.0,
"step": 3070
},
{
"entropy": 6.023791694641114,
"epoch": 0.3039739027283511,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995239915376988,
"loss": 5.854,
"mean_token_accuracy": 0.1543585553765297,
"num_tokens": 7024769.0,
"step": 3075
},
{
"entropy": 6.019925165176391,
"epoch": 0.3044681692368525,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995216944628769,
"loss": 5.8561,
"mean_token_accuracy": 0.14953593611717225,
"num_tokens": 7035422.0,
"step": 3080
},
{
"entropy": 5.966122579574585,
"epoch": 0.3049624357453539,
"grad_norm": 0.9375,
"learning_rate": 0.0004995193918647865,
"loss": 5.7801,
"mean_token_accuracy": 0.16542816460132598,
"num_tokens": 7046923.0,
"step": 3085
},
{
"entropy": 6.064773893356323,
"epoch": 0.3054567022538553,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995170837434844,
"loss": 5.835,
"mean_token_accuracy": 0.15419175177812577,
"num_tokens": 7058241.0,
"step": 3090
},
{
"entropy": 5.946009635925293,
"epoch": 0.30595096876235667,
"grad_norm": 0.953125,
"learning_rate": 0.0004995147700990269,
"loss": 5.8536,
"mean_token_accuracy": 0.15725975334644318,
"num_tokens": 7070028.0,
"step": 3095
},
{
"entropy": 6.053697729110718,
"epoch": 0.30644523527085804,
"grad_norm": 0.984375,
"learning_rate": 0.0004995124509314714,
"loss": 5.843,
"mean_token_accuracy": 0.1565923348069191,
"num_tokens": 7082641.0,
"step": 3100
},
{
"entropy": 6.01456732749939,
"epoch": 0.3069395017793594,
"grad_norm": 1.0,
"learning_rate": 0.0004995101262408748,
"loss": 5.8442,
"mean_token_accuracy": 0.15755832195281982,
"num_tokens": 7094462.0,
"step": 3105
},
{
"entropy": 5.96401424407959,
"epoch": 0.3074337682878608,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995077960272943,
"loss": 5.8022,
"mean_token_accuracy": 0.15858004838228226,
"num_tokens": 7106598.0,
"step": 3110
},
{
"entropy": 5.959226036071778,
"epoch": 0.3079280347963622,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995054602907871,
"loss": 5.7316,
"mean_token_accuracy": 0.15864774882793425,
"num_tokens": 7117006.0,
"step": 3115
},
{
"entropy": 5.91728892326355,
"epoch": 0.3084223013048636,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995031190314107,
"loss": 5.814,
"mean_token_accuracy": 0.1534339338541031,
"num_tokens": 7128071.0,
"step": 3120
},
{
"entropy": 6.038468742370606,
"epoch": 0.30891656781336496,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995007722492228,
"loss": 5.8174,
"mean_token_accuracy": 0.15981683135032654,
"num_tokens": 7139896.0,
"step": 3125
},
{
"entropy": 5.941322183609008,
"epoch": 0.30941083432186633,
"grad_norm": 0.921875,
"learning_rate": 0.0004994984199442811,
"loss": 5.7765,
"mean_token_accuracy": 0.16602752953767777,
"num_tokens": 7151479.0,
"step": 3130
},
{
"entropy": 6.075007772445678,
"epoch": 0.3099051008303677,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994960621166434,
"loss": 5.8136,
"mean_token_accuracy": 0.1607074946165085,
"num_tokens": 7162942.0,
"step": 3135
},
{
"entropy": 5.900337553024292,
"epoch": 0.31039936733886914,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994936987663678,
"loss": 5.7909,
"mean_token_accuracy": 0.17076555639505386,
"num_tokens": 7174572.0,
"step": 3140
},
{
"entropy": 6.006263065338135,
"epoch": 0.3108936338473705,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994913298935124,
"loss": 5.8289,
"mean_token_accuracy": 0.15694652497768402,
"num_tokens": 7186352.0,
"step": 3145
},
{
"entropy": 6.027054071426392,
"epoch": 0.3113879003558719,
"grad_norm": 0.890625,
"learning_rate": 0.0004994889554981353,
"loss": 5.8046,
"mean_token_accuracy": 0.15830225646495819,
"num_tokens": 7197849.0,
"step": 3150
},
{
"entropy": 5.930756330490112,
"epoch": 0.31188216686437326,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994865755802953,
"loss": 5.7638,
"mean_token_accuracy": 0.16295457482337952,
"num_tokens": 7207897.0,
"step": 3155
},
{
"entropy": 6.050958681106567,
"epoch": 0.31237643337287463,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994841901400505,
"loss": 5.8943,
"mean_token_accuracy": 0.15915813744068147,
"num_tokens": 7219148.0,
"step": 3160
},
{
"entropy": 6.136546277999878,
"epoch": 0.31287069988137606,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994817991774599,
"loss": 5.9668,
"mean_token_accuracy": 0.1445558950304985,
"num_tokens": 7230690.0,
"step": 3165
},
{
"entropy": 6.0489771366119385,
"epoch": 0.31336496638987743,
"grad_norm": 0.953125,
"learning_rate": 0.0004994794026925822,
"loss": 5.8088,
"mean_token_accuracy": 0.1555367588996887,
"num_tokens": 7240935.0,
"step": 3170
},
{
"entropy": 5.9755167961120605,
"epoch": 0.3138592328983788,
"grad_norm": 0.859375,
"learning_rate": 0.0004994770006854763,
"loss": 5.8562,
"mean_token_accuracy": 0.1574193097651005,
"num_tokens": 7253345.0,
"step": 3175
},
{
"entropy": 6.117976140975952,
"epoch": 0.3143534994068802,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994745931562015,
"loss": 5.8705,
"mean_token_accuracy": 0.1560059294104576,
"num_tokens": 7265355.0,
"step": 3180
},
{
"entropy": 5.841166162490845,
"epoch": 0.31484776591538155,
"grad_norm": 0.9140625,
"learning_rate": 0.0004994721801048168,
"loss": 5.766,
"mean_token_accuracy": 0.16469242423772812,
"num_tokens": 7277355.0,
"step": 3185
},
{
"entropy": 6.000030755996704,
"epoch": 0.315342032423883,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994697615313817,
"loss": 5.8185,
"mean_token_accuracy": 0.15941714942455293,
"num_tokens": 7288754.0,
"step": 3190
},
{
"entropy": 6.082979965209961,
"epoch": 0.31583629893238435,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994673374359556,
"loss": 5.8814,
"mean_token_accuracy": 0.15342381596565247,
"num_tokens": 7301158.0,
"step": 3195
},
{
"entropy": 5.9652362823486325,
"epoch": 0.3163305654408857,
"grad_norm": 0.9296875,
"learning_rate": 0.0004994649078185982,
"loss": 5.8598,
"mean_token_accuracy": 0.1522575005888939,
"num_tokens": 7314420.0,
"step": 3200
},
{
"entropy": 5.985113191604614,
"epoch": 0.3168248319493871,
"grad_norm": 0.9375,
"learning_rate": 0.0004994624726793693,
"loss": 5.8004,
"mean_token_accuracy": 0.15395023226737975,
"num_tokens": 7325459.0,
"step": 3205
},
{
"entropy": 6.024287176132202,
"epoch": 0.31731909845788847,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994600320183286,
"loss": 5.7606,
"mean_token_accuracy": 0.1591726526618004,
"num_tokens": 7336016.0,
"step": 3210
},
{
"entropy": 5.913967657089233,
"epoch": 0.3178133649663899,
"grad_norm": 0.9921875,
"learning_rate": 0.0004994575858355364,
"loss": 5.7501,
"mean_token_accuracy": 0.163642817735672,
"num_tokens": 7346716.0,
"step": 3215
},
{
"entropy": 6.081499481201172,
"epoch": 0.3183076314748913,
"grad_norm": 1.0625,
"learning_rate": 0.0004994551341310526,
"loss": 5.8756,
"mean_token_accuracy": 0.15766478031873704,
"num_tokens": 7357246.0,
"step": 3220
},
{
"entropy": 5.9896539688110355,
"epoch": 0.31880189798339265,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994526769049379,
"loss": 5.8908,
"mean_token_accuracy": 0.14943311512470245,
"num_tokens": 7369173.0,
"step": 3225
},
{
"entropy": 6.064849996566773,
"epoch": 0.319296164491894,
"grad_norm": 1.03125,
"learning_rate": 0.0004994502141572524,
"loss": 5.8934,
"mean_token_accuracy": 0.14667481258511544,
"num_tokens": 7381118.0,
"step": 3230
},
{
"entropy": 5.995628929138183,
"epoch": 0.3197904310003954,
"grad_norm": 0.9140625,
"learning_rate": 0.0004994477458880568,
"loss": 5.7268,
"mean_token_accuracy": 0.16845426261425017,
"num_tokens": 7391874.0,
"step": 3235
},
{
"entropy": 5.9674286365509035,
"epoch": 0.3202846975088968,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994452720974119,
"loss": 5.8193,
"mean_token_accuracy": 0.16485921144485474,
"num_tokens": 7403474.0,
"step": 3240
},
{
"entropy": 5.999954891204834,
"epoch": 0.3207789640173982,
"grad_norm": 0.890625,
"learning_rate": 0.0004994427927853785,
"loss": 5.7287,
"mean_token_accuracy": 0.16761064678430557,
"num_tokens": 7415173.0,
"step": 3245
},
{
"entropy": 5.933607673645019,
"epoch": 0.32127323052589957,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994403079520175,
"loss": 5.7487,
"mean_token_accuracy": 0.16161210387945174,
"num_tokens": 7426965.0,
"step": 3250
},
{
"entropy": 5.983630275726318,
"epoch": 0.32176749703440094,
"grad_norm": 0.9921875,
"learning_rate": 0.0004994378175973902,
"loss": 5.7164,
"mean_token_accuracy": 0.1643822655081749,
"num_tokens": 7438162.0,
"step": 3255
},
{
"entropy": 5.964900636672974,
"epoch": 0.3222617635429023,
"grad_norm": 0.95703125,
"learning_rate": 0.0004994353217215577,
"loss": 5.815,
"mean_token_accuracy": 0.1570591390132904,
"num_tokens": 7450241.0,
"step": 3260
},
{
"entropy": 6.0549173831939695,
"epoch": 0.32275603005140374,
"grad_norm": 0.984375,
"learning_rate": 0.0004994328203245814,
"loss": 5.9238,
"mean_token_accuracy": 0.15149068981409072,
"num_tokens": 7462236.0,
"step": 3265
},
{
"entropy": 5.991492128372192,
"epoch": 0.3232502965599051,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994303134065231,
"loss": 5.8109,
"mean_token_accuracy": 0.15657777860760688,
"num_tokens": 7472811.0,
"step": 3270
},
{
"entropy": 5.953076362609863,
"epoch": 0.3237445630684065,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994278009674442,
"loss": 5.7056,
"mean_token_accuracy": 0.17091321498155593,
"num_tokens": 7484837.0,
"step": 3275
},
{
"entropy": 5.976023769378662,
"epoch": 0.32423882957690786,
"grad_norm": 0.93359375,
"learning_rate": 0.0004994252830074064,
"loss": 5.822,
"mean_token_accuracy": 0.15214902013540268,
"num_tokens": 7496712.0,
"step": 3280
},
{
"entropy": 5.9532591819763185,
"epoch": 0.32473309608540923,
"grad_norm": 0.88671875,
"learning_rate": 0.000499422759526472,
"loss": 5.8042,
"mean_token_accuracy": 0.15640552788972856,
"num_tokens": 7508014.0,
"step": 3285
},
{
"entropy": 6.006920337677002,
"epoch": 0.32522736259391066,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994202305247029,
"loss": 5.8459,
"mean_token_accuracy": 0.15756394267082213,
"num_tokens": 7519466.0,
"step": 3290
},
{
"entropy": 6.092730522155762,
"epoch": 0.32572162910241204,
"grad_norm": 0.82421875,
"learning_rate": 0.0004994176960021613,
"loss": 5.9052,
"mean_token_accuracy": 0.15043263658881187,
"num_tokens": 7531661.0,
"step": 3295
},
{
"entropy": 6.011136770248413,
"epoch": 0.3262158956109134,
"grad_norm": 0.84375,
"learning_rate": 0.0004994151559589095,
"loss": 5.8126,
"mean_token_accuracy": 0.15739361941814423,
"num_tokens": 7544096.0,
"step": 3300
},
{
"entropy": 5.966448974609375,
"epoch": 0.3267101621194148,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994126103950101,
"loss": 5.7808,
"mean_token_accuracy": 0.15891228914260863,
"num_tokens": 7556022.0,
"step": 3305
},
{
"entropy": 6.001350831985474,
"epoch": 0.32720442862791616,
"grad_norm": 0.921875,
"learning_rate": 0.0004994100593105257,
"loss": 5.7715,
"mean_token_accuracy": 0.16757655888795853,
"num_tokens": 7567399.0,
"step": 3310
},
{
"entropy": 5.9018275260925295,
"epoch": 0.3276986951364176,
"grad_norm": 0.8671875,
"learning_rate": 0.0004994075027055189,
"loss": 5.8052,
"mean_token_accuracy": 0.16109628975391388,
"num_tokens": 7580673.0,
"step": 3315
},
{
"entropy": 6.0015599727630615,
"epoch": 0.32819296164491896,
"grad_norm": 0.984375,
"learning_rate": 0.0004994049405800529,
"loss": 5.7783,
"mean_token_accuracy": 0.15706816017627717,
"num_tokens": 7591799.0,
"step": 3320
},
{
"entropy": 6.0001458644866945,
"epoch": 0.32868722815342033,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994023729341904,
"loss": 5.8133,
"mean_token_accuracy": 0.15964777916669845,
"num_tokens": 7603663.0,
"step": 3325
},
{
"entropy": 5.934050226211548,
"epoch": 0.3291814946619217,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993997997679949,
"loss": 5.7619,
"mean_token_accuracy": 0.1650229200720787,
"num_tokens": 7615651.0,
"step": 3330
},
{
"entropy": 5.850442028045654,
"epoch": 0.3296757611704231,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993972210815292,
"loss": 5.7733,
"mean_token_accuracy": 0.16448295265436172,
"num_tokens": 7628033.0,
"step": 3335
},
{
"entropy": 6.106792783737182,
"epoch": 0.33017002767892445,
"grad_norm": 0.8671875,
"learning_rate": 0.0004993946368748573,
"loss": 5.8938,
"mean_token_accuracy": 0.15707466006278992,
"num_tokens": 7640227.0,
"step": 3340
},
{
"entropy": 5.990322303771973,
"epoch": 0.3306642941874259,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993920471480424,
"loss": 5.7874,
"mean_token_accuracy": 0.1599864438176155,
"num_tokens": 7651684.0,
"step": 3345
},
{
"entropy": 5.992852735519409,
"epoch": 0.33115856069592725,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993894519011484,
"loss": 5.8139,
"mean_token_accuracy": 0.1593676432967186,
"num_tokens": 7662721.0,
"step": 3350
},
{
"entropy": 5.983349990844727,
"epoch": 0.3316528272044286,
"grad_norm": 0.84765625,
"learning_rate": 0.0004993868511342391,
"loss": 5.7699,
"mean_token_accuracy": 0.16285967379808425,
"num_tokens": 7674060.0,
"step": 3355
},
{
"entropy": 5.922141075134277,
"epoch": 0.33214709371293,
"grad_norm": 1.09375,
"learning_rate": 0.0004993842448473783,
"loss": 5.712,
"mean_token_accuracy": 0.16841071248054504,
"num_tokens": 7685054.0,
"step": 3360
},
{
"entropy": 5.9582678318023685,
"epoch": 0.33264136022143137,
"grad_norm": 0.921875,
"learning_rate": 0.0004993816330406304,
"loss": 5.8457,
"mean_token_accuracy": 0.15618079006671906,
"num_tokens": 7696554.0,
"step": 3365
},
{
"entropy": 5.990711975097656,
"epoch": 0.3331356267299328,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993790157140595,
"loss": 5.7073,
"mean_token_accuracy": 0.15978904366493224,
"num_tokens": 7707000.0,
"step": 3370
},
{
"entropy": 6.0205179214477536,
"epoch": 0.33362989323843417,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993763928677299,
"loss": 5.7961,
"mean_token_accuracy": 0.16219552308321,
"num_tokens": 7718898.0,
"step": 3375
},
{
"entropy": 5.975991106033325,
"epoch": 0.33412415974693555,
"grad_norm": 0.984375,
"learning_rate": 0.0004993737645017064,
"loss": 5.838,
"mean_token_accuracy": 0.15772839039564132,
"num_tokens": 7730493.0,
"step": 3380
},
{
"entropy": 5.923499536514282,
"epoch": 0.3346184262554369,
"grad_norm": 0.953125,
"learning_rate": 0.0004993711306160534,
"loss": 5.7392,
"mean_token_accuracy": 0.1578788787126541,
"num_tokens": 7742177.0,
"step": 3385
},
{
"entropy": 5.978937292098999,
"epoch": 0.3351126927639383,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993684912108358,
"loss": 5.815,
"mean_token_accuracy": 0.16342742890119552,
"num_tokens": 7753386.0,
"step": 3390
},
{
"entropy": 5.9103271484375,
"epoch": 0.3356069592724397,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993658462861184,
"loss": 5.7282,
"mean_token_accuracy": 0.16263284534215927,
"num_tokens": 7764259.0,
"step": 3395
},
{
"entropy": 5.9659192085266115,
"epoch": 0.3361012257809411,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993631958419666,
"loss": 5.7061,
"mean_token_accuracy": 0.1682667076587677,
"num_tokens": 7775853.0,
"step": 3400
},
{
"entropy": 5.945500326156616,
"epoch": 0.33659549228944247,
"grad_norm": 0.91015625,
"learning_rate": 0.0004993605398784451,
"loss": 5.8374,
"mean_token_accuracy": 0.15530660152435302,
"num_tokens": 7787116.0,
"step": 3405
},
{
"entropy": 6.046633911132813,
"epoch": 0.33708975879794384,
"grad_norm": 0.87890625,
"learning_rate": 0.0004993578783956198,
"loss": 5.8617,
"mean_token_accuracy": 0.15285939276218413,
"num_tokens": 7799666.0,
"step": 3410
},
{
"entropy": 5.977567195892334,
"epoch": 0.3375840253064452,
"grad_norm": 0.921875,
"learning_rate": 0.0004993552113935557,
"loss": 5.7654,
"mean_token_accuracy": 0.16652155816555023,
"num_tokens": 7812454.0,
"step": 3415
},
{
"entropy": 5.960903215408325,
"epoch": 0.33807829181494664,
"grad_norm": 0.9375,
"learning_rate": 0.0004993525388723186,
"loss": 5.7933,
"mean_token_accuracy": 0.15966083854436874,
"num_tokens": 7824427.0,
"step": 3420
},
{
"entropy": 5.984457540512085,
"epoch": 0.338572558323448,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993498608319742,
"loss": 5.7961,
"mean_token_accuracy": 0.16384893357753755,
"num_tokens": 7835382.0,
"step": 3425
},
{
"entropy": 5.932029294967651,
"epoch": 0.3390668248319494,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993471772725886,
"loss": 5.7082,
"mean_token_accuracy": 0.17235663235187532,
"num_tokens": 7846618.0,
"step": 3430
},
{
"entropy": 5.924665880203247,
"epoch": 0.33956109134045076,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993444881942274,
"loss": 5.7513,
"mean_token_accuracy": 0.16130691021680832,
"num_tokens": 7859160.0,
"step": 3435
},
{
"entropy": 5.8683253765106205,
"epoch": 0.34005535784895213,
"grad_norm": 0.8671875,
"learning_rate": 0.0004993417935969572,
"loss": 5.771,
"mean_token_accuracy": 0.1602069690823555,
"num_tokens": 7870628.0,
"step": 3440
},
{
"entropy": 6.023205184936524,
"epoch": 0.34054962435745356,
"grad_norm": 0.8984375,
"learning_rate": 0.000499339093480844,
"loss": 5.805,
"mean_token_accuracy": 0.1631205275654793,
"num_tokens": 7883016.0,
"step": 3445
},
{
"entropy": 5.954569149017334,
"epoch": 0.34104389086595494,
"grad_norm": 0.92578125,
"learning_rate": 0.0004993363878459542,
"loss": 5.7992,
"mean_token_accuracy": 0.15368073135614396,
"num_tokens": 7895940.0,
"step": 3450
},
{
"entropy": 5.957338905334472,
"epoch": 0.3415381573744563,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993336766923546,
"loss": 5.811,
"mean_token_accuracy": 0.15807777494192124,
"num_tokens": 7907920.0,
"step": 3455
},
{
"entropy": 5.994265747070313,
"epoch": 0.3420324238829577,
"grad_norm": 0.96875,
"learning_rate": 0.0004993309600201117,
"loss": 5.8035,
"mean_token_accuracy": 0.16190102994441985,
"num_tokens": 7920181.0,
"step": 3460
},
{
"entropy": 5.9474118709564205,
"epoch": 0.34252669039145905,
"grad_norm": 0.9921875,
"learning_rate": 0.0004993282378292923,
"loss": 5.7755,
"mean_token_accuracy": 0.1690393105149269,
"num_tokens": 7931135.0,
"step": 3465
},
{
"entropy": 5.96035213470459,
"epoch": 0.3430209568999605,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993255101199635,
"loss": 5.7411,
"mean_token_accuracy": 0.16467749774456025,
"num_tokens": 7941002.0,
"step": 3470
},
{
"entropy": 5.978878927230835,
"epoch": 0.34351522340846186,
"grad_norm": 1.03125,
"learning_rate": 0.0004993227768921924,
"loss": 5.717,
"mean_token_accuracy": 0.1669140040874481,
"num_tokens": 7950393.0,
"step": 3475
},
{
"entropy": 5.915142822265625,
"epoch": 0.34400948991696323,
"grad_norm": 0.83984375,
"learning_rate": 0.0004993200381460461,
"loss": 5.7424,
"mean_token_accuracy": 0.16538604199886323,
"num_tokens": 7961468.0,
"step": 3480
},
{
"entropy": 5.8853638648986815,
"epoch": 0.3445037564254646,
"grad_norm": 0.8359375,
"learning_rate": 0.0004993172938815922,
"loss": 5.5819,
"mean_token_accuracy": 0.1773862972855568,
"num_tokens": 7973783.0,
"step": 3485
},
{
"entropy": 5.884685754776001,
"epoch": 0.344998022933966,
"grad_norm": 0.96875,
"learning_rate": 0.0004993145440988979,
"loss": 5.6582,
"mean_token_accuracy": 0.16831784844398498,
"num_tokens": 7984099.0,
"step": 3490
},
{
"entropy": 6.006422710418701,
"epoch": 0.3454922894424674,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993117887980311,
"loss": 5.8026,
"mean_token_accuracy": 0.1570899724960327,
"num_tokens": 7995115.0,
"step": 3495
},
{
"entropy": 5.938472509384155,
"epoch": 0.3459865559509688,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993090279790596,
"loss": 5.7549,
"mean_token_accuracy": 0.16229426264762878,
"num_tokens": 8006103.0,
"step": 3500
},
{
"entropy": 5.8403284549713135,
"epoch": 0.34648082245947015,
"grad_norm": 0.9453125,
"learning_rate": 0.000499306261642051,
"loss": 5.712,
"mean_token_accuracy": 0.15982834249734879,
"num_tokens": 8018228.0,
"step": 3505
},
{
"entropy": 5.949107980728149,
"epoch": 0.3469750889679715,
"grad_norm": 0.921875,
"learning_rate": 0.0004993034897870737,
"loss": 5.6888,
"mean_token_accuracy": 0.16040400564670562,
"num_tokens": 8029808.0,
"step": 3510
},
{
"entropy": 5.904809379577637,
"epoch": 0.3474693554764729,
"grad_norm": 0.859375,
"learning_rate": 0.0004993007124141958,
"loss": 5.769,
"mean_token_accuracy": 0.15806748270988463,
"num_tokens": 8041136.0,
"step": 3515
},
{
"entropy": 5.961758375167847,
"epoch": 0.3479636219849743,
"grad_norm": 1.0,
"learning_rate": 0.0004992979295234854,
"loss": 5.8112,
"mean_token_accuracy": 0.15807072520256044,
"num_tokens": 8053013.0,
"step": 3520
},
{
"entropy": 5.986888790130616,
"epoch": 0.3484578884934757,
"grad_norm": 0.890625,
"learning_rate": 0.0004992951411150112,
"loss": 5.7934,
"mean_token_accuracy": 0.16370038390159608,
"num_tokens": 8065139.0,
"step": 3525
},
{
"entropy": 5.906422662734985,
"epoch": 0.34895215500197707,
"grad_norm": 1.0,
"learning_rate": 0.0004992923471888418,
"loss": 5.6435,
"mean_token_accuracy": 0.1711960718035698,
"num_tokens": 8075719.0,
"step": 3530
},
{
"entropy": 5.937537574768067,
"epoch": 0.34944642151047844,
"grad_norm": 0.90625,
"learning_rate": 0.0004992895477450459,
"loss": 5.7729,
"mean_token_accuracy": 0.16310322284698486,
"num_tokens": 8087821.0,
"step": 3535
},
{
"entropy": 5.891085863113403,
"epoch": 0.3499406880189798,
"grad_norm": 1.015625,
"learning_rate": 0.0004992867427836923,
"loss": 5.6526,
"mean_token_accuracy": 0.16491053253412247,
"num_tokens": 8099861.0,
"step": 3540
},
{
"entropy": 6.05497784614563,
"epoch": 0.3504349545274812,
"grad_norm": 0.984375,
"learning_rate": 0.0004992839323048501,
"loss": 5.9043,
"mean_token_accuracy": 0.1527191035449505,
"num_tokens": 8111671.0,
"step": 3545
},
{
"entropy": 6.090972948074341,
"epoch": 0.3509292210359826,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992811163085882,
"loss": 5.852,
"mean_token_accuracy": 0.15342505127191544,
"num_tokens": 8123569.0,
"step": 3550
},
{
"entropy": 5.987070226669312,
"epoch": 0.351423487544484,
"grad_norm": 0.90625,
"learning_rate": 0.0004992782947949761,
"loss": 5.8515,
"mean_token_accuracy": 0.15805700719356536,
"num_tokens": 8135621.0,
"step": 3555
},
{
"entropy": 5.903195428848266,
"epoch": 0.35191775405298537,
"grad_norm": 0.98828125,
"learning_rate": 0.0004992754677640831,
"loss": 5.7147,
"mean_token_accuracy": 0.16510182321071626,
"num_tokens": 8147012.0,
"step": 3560
},
{
"entropy": 5.971565055847168,
"epoch": 0.35241202056148674,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992726352159789,
"loss": 5.7032,
"mean_token_accuracy": 0.16711681187152863,
"num_tokens": 8158686.0,
"step": 3565
},
{
"entropy": 5.921085596084595,
"epoch": 0.3529062870699881,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992697971507331,
"loss": 5.6887,
"mean_token_accuracy": 0.168816876411438,
"num_tokens": 8170188.0,
"step": 3570
},
{
"entropy": 5.903818559646607,
"epoch": 0.35340055357848954,
"grad_norm": 0.8828125,
"learning_rate": 0.0004992669535684155,
"loss": 5.779,
"mean_token_accuracy": 0.1621542379260063,
"num_tokens": 8182467.0,
"step": 3575
},
{
"entropy": 6.023322820663452,
"epoch": 0.3538948200869909,
"grad_norm": 0.90625,
"learning_rate": 0.0004992641044690959,
"loss": 5.8818,
"mean_token_accuracy": 0.1577792763710022,
"num_tokens": 8193433.0,
"step": 3580
},
{
"entropy": 5.934952306747436,
"epoch": 0.3543890865954923,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992612498528447,
"loss": 5.6643,
"mean_token_accuracy": 0.1658499911427498,
"num_tokens": 8204220.0,
"step": 3585
},
{
"entropy": 5.828398132324219,
"epoch": 0.35488335310399366,
"grad_norm": 0.984375,
"learning_rate": 0.0004992583897197319,
"loss": 5.6841,
"mean_token_accuracy": 0.17446459233760833,
"num_tokens": 8214203.0,
"step": 3590
},
{
"entropy": 6.018448543548584,
"epoch": 0.35537761961249503,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992555240698279,
"loss": 5.8262,
"mean_token_accuracy": 0.1579638361930847,
"num_tokens": 8225143.0,
"step": 3595
},
{
"entropy": 5.982320880889892,
"epoch": 0.35587188612099646,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992526529032032,
"loss": 5.6817,
"mean_token_accuracy": 0.16971834599971772,
"num_tokens": 8237715.0,
"step": 3600
},
{
"entropy": 5.860608386993408,
"epoch": 0.35636615262949783,
"grad_norm": 0.8359375,
"learning_rate": 0.0004992497762199285,
"loss": 5.7367,
"mean_token_accuracy": 0.16622899174690248,
"num_tokens": 8249645.0,
"step": 3605
},
{
"entropy": 5.949036455154419,
"epoch": 0.3568604191379992,
"grad_norm": 0.96484375,
"learning_rate": 0.0004992468940200745,
"loss": 5.6976,
"mean_token_accuracy": 0.1648137539625168,
"num_tokens": 8260624.0,
"step": 3610
},
{
"entropy": 5.970369958877564,
"epoch": 0.3573546856465006,
"grad_norm": 0.921875,
"learning_rate": 0.0004992440063037121,
"loss": 5.8078,
"mean_token_accuracy": 0.16165586411952973,
"num_tokens": 8272526.0,
"step": 3615
},
{
"entropy": 5.85641655921936,
"epoch": 0.35784895215500195,
"grad_norm": 0.9375,
"learning_rate": 0.0004992411130709124,
"loss": 5.7017,
"mean_token_accuracy": 0.16460638791322707,
"num_tokens": 8284270.0,
"step": 3620
},
{
"entropy": 6.034624099731445,
"epoch": 0.3583432186635034,
"grad_norm": 0.91796875,
"learning_rate": 0.0004992382143217466,
"loss": 5.8199,
"mean_token_accuracy": 0.16384400129318238,
"num_tokens": 8296407.0,
"step": 3625
},
{
"entropy": 5.897780036926269,
"epoch": 0.35883748517200476,
"grad_norm": 0.91796875,
"learning_rate": 0.0004992353100562858,
"loss": 5.7127,
"mean_token_accuracy": 0.17074024528265,
"num_tokens": 8307695.0,
"step": 3630
},
{
"entropy": 5.851276445388794,
"epoch": 0.35933175168050613,
"grad_norm": 0.93359375,
"learning_rate": 0.0004992324002746016,
"loss": 5.6991,
"mean_token_accuracy": 0.16381751298904418,
"num_tokens": 8319458.0,
"step": 3635
},
{
"entropy": 5.8534314155578615,
"epoch": 0.3598260181890075,
"grad_norm": 0.90625,
"learning_rate": 0.0004992294849767656,
"loss": 5.6067,
"mean_token_accuracy": 0.1727339133620262,
"num_tokens": 8330284.0,
"step": 3640
},
{
"entropy": 5.955981206893921,
"epoch": 0.3603202846975089,
"grad_norm": 0.91015625,
"learning_rate": 0.0004992265641628495,
"loss": 5.7445,
"mean_token_accuracy": 0.16239794641733168,
"num_tokens": 8343622.0,
"step": 3645
},
{
"entropy": 5.90520133972168,
"epoch": 0.3608145512060103,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992236378329252,
"loss": 5.6977,
"mean_token_accuracy": 0.16424720287322997,
"num_tokens": 8354079.0,
"step": 3650
},
{
"entropy": 5.9196117401123045,
"epoch": 0.3613088177145117,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992207059870645,
"loss": 5.7422,
"mean_token_accuracy": 0.1665163055062294,
"num_tokens": 8364876.0,
"step": 3655
},
{
"entropy": 5.96867995262146,
"epoch": 0.36180308422301305,
"grad_norm": 0.92578125,
"learning_rate": 0.0004992177686253396,
"loss": 5.6805,
"mean_token_accuracy": 0.17012346237897874,
"num_tokens": 8375095.0,
"step": 3660
},
{
"entropy": 5.867181015014649,
"epoch": 0.3622973507315144,
"grad_norm": 0.9765625,
"learning_rate": 0.000499214825747823,
"loss": 5.6612,
"mean_token_accuracy": 0.16863478869199752,
"num_tokens": 8385934.0,
"step": 3665
},
{
"entropy": 5.895959329605103,
"epoch": 0.3627916172400158,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992118773545868,
"loss": 5.6241,
"mean_token_accuracy": 0.1754315674304962,
"num_tokens": 8397018.0,
"step": 3670
},
{
"entropy": 5.8759214878082275,
"epoch": 0.3632858837485172,
"grad_norm": 0.97265625,
"learning_rate": 0.0004992089234457036,
"loss": 5.7069,
"mean_token_accuracy": 0.17084360122680664,
"num_tokens": 8406668.0,
"step": 3675
},
{
"entropy": 5.957876110076905,
"epoch": 0.3637801502570186,
"grad_norm": 0.98828125,
"learning_rate": 0.0004992059640212461,
"loss": 5.7349,
"mean_token_accuracy": 0.16381780356168746,
"num_tokens": 8417471.0,
"step": 3680
},
{
"entropy": 5.9014252662658695,
"epoch": 0.36427441676551997,
"grad_norm": 0.8515625,
"learning_rate": 0.0004992029990812872,
"loss": 5.8016,
"mean_token_accuracy": 0.15887220278382302,
"num_tokens": 8430073.0,
"step": 3685
},
{
"entropy": 5.959695863723755,
"epoch": 0.36476868327402134,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992000286258997,
"loss": 5.7188,
"mean_token_accuracy": 0.15926548838615417,
"num_tokens": 8442573.0,
"step": 3690
},
{
"entropy": 5.989988708496094,
"epoch": 0.3652629497825227,
"grad_norm": 0.921875,
"learning_rate": 0.0004991970526551567,
"loss": 5.7659,
"mean_token_accuracy": 0.16171564161777496,
"num_tokens": 8453920.0,
"step": 3695
},
{
"entropy": 5.919428443908691,
"epoch": 0.36575721629102415,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991940711691314,
"loss": 5.7961,
"mean_token_accuracy": 0.1626737430691719,
"num_tokens": 8466652.0,
"step": 3700
},
{
"entropy": 5.965742731094361,
"epoch": 0.3662514827995255,
"grad_norm": 0.87109375,
"learning_rate": 0.0004991910841678972,
"loss": 5.8075,
"mean_token_accuracy": 0.16027045547962188,
"num_tokens": 8480203.0,
"step": 3705
},
{
"entropy": 5.894874143600464,
"epoch": 0.3667457493080269,
"grad_norm": 0.91796875,
"learning_rate": 0.0004991880916515275,
"loss": 5.718,
"mean_token_accuracy": 0.15868055671453477,
"num_tokens": 8491396.0,
"step": 3710
},
{
"entropy": 5.984350538253784,
"epoch": 0.36724001581652826,
"grad_norm": 1.0234375,
"learning_rate": 0.000499185093620096,
"loss": 5.6837,
"mean_token_accuracy": 0.1644155576825142,
"num_tokens": 8501838.0,
"step": 3715
},
{
"entropy": 5.835333776473999,
"epoch": 0.36773428232502964,
"grad_norm": 1.046875,
"learning_rate": 0.0004991820900736765,
"loss": 5.5741,
"mean_token_accuracy": 0.1708022579550743,
"num_tokens": 8512113.0,
"step": 3720
},
{
"entropy": 5.903925180435181,
"epoch": 0.36822854883353107,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991790810123427,
"loss": 5.6721,
"mean_token_accuracy": 0.16471227556467055,
"num_tokens": 8523888.0,
"step": 3725
},
{
"entropy": 5.873724317550659,
"epoch": 0.36872281534203244,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991760664361688,
"loss": 5.5899,
"mean_token_accuracy": 0.1763386145234108,
"num_tokens": 8536078.0,
"step": 3730
},
{
"entropy": 5.917610549926758,
"epoch": 0.3692170818505338,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991730463452288,
"loss": 5.7365,
"mean_token_accuracy": 0.15684023201465608,
"num_tokens": 8547131.0,
"step": 3735
},
{
"entropy": 5.8698803901672365,
"epoch": 0.3697113483590352,
"grad_norm": 0.9609375,
"learning_rate": 0.0004991700207395971,
"loss": 5.6005,
"mean_token_accuracy": 0.1780976265668869,
"num_tokens": 8557269.0,
"step": 3740
},
{
"entropy": 5.956096744537353,
"epoch": 0.37020561486753656,
"grad_norm": 0.96875,
"learning_rate": 0.0004991669896193482,
"loss": 5.7906,
"mean_token_accuracy": 0.15648694485425949,
"num_tokens": 8569193.0,
"step": 3745
},
{
"entropy": 5.880059432983399,
"epoch": 0.370699881376038,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991639529845565,
"loss": 5.6622,
"mean_token_accuracy": 0.16808689683675765,
"num_tokens": 8580888.0,
"step": 3750
},
{
"entropy": 5.89161524772644,
"epoch": 0.37119414788453936,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991609108352968,
"loss": 5.7295,
"mean_token_accuracy": 0.16513347923755645,
"num_tokens": 8591868.0,
"step": 3755
},
{
"entropy": 5.909579420089722,
"epoch": 0.37168841439304073,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991578631716438,
"loss": 5.7067,
"mean_token_accuracy": 0.16595157831907273,
"num_tokens": 8602338.0,
"step": 3760
},
{
"entropy": 5.941970682144165,
"epoch": 0.3721826809015421,
"grad_norm": 0.921875,
"learning_rate": 0.0004991548099936727,
"loss": 5.6346,
"mean_token_accuracy": 0.173343163728714,
"num_tokens": 8611698.0,
"step": 3765
},
{
"entropy": 5.856791067123413,
"epoch": 0.3726769474100435,
"grad_norm": 0.98046875,
"learning_rate": 0.0004991517513014585,
"loss": 5.6387,
"mean_token_accuracy": 0.1660086154937744,
"num_tokens": 8623537.0,
"step": 3770
},
{
"entropy": 5.841311836242676,
"epoch": 0.37317121391854485,
"grad_norm": 0.953125,
"learning_rate": 0.0004991486870950765,
"loss": 5.7014,
"mean_token_accuracy": 0.1621934913098812,
"num_tokens": 8634678.0,
"step": 3775
},
{
"entropy": 5.882809162139893,
"epoch": 0.3736654804270463,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991456173746021,
"loss": 5.687,
"mean_token_accuracy": 0.16786129027605057,
"num_tokens": 8646207.0,
"step": 3780
},
{
"entropy": 6.001916694641113,
"epoch": 0.37415974693554765,
"grad_norm": 0.8984375,
"learning_rate": 0.0004991425421401105,
"loss": 5.7475,
"mean_token_accuracy": 0.1641547426581383,
"num_tokens": 8657069.0,
"step": 3785
},
{
"entropy": 5.807131099700928,
"epoch": 0.374654013444049,
"grad_norm": 0.87890625,
"learning_rate": 0.0004991394613916777,
"loss": 5.6186,
"mean_token_accuracy": 0.1740735113620758,
"num_tokens": 8668296.0,
"step": 3790
},
{
"entropy": 5.855770206451416,
"epoch": 0.3751482799525504,
"grad_norm": 0.88671875,
"learning_rate": 0.0004991363751293795,
"loss": 5.7671,
"mean_token_accuracy": 0.16410524398088455,
"num_tokens": 8680626.0,
"step": 3795
},
{
"entropy": 5.886316013336182,
"epoch": 0.3756425464610518,
"grad_norm": 0.8671875,
"learning_rate": 0.0004991332833532915,
"loss": 5.6633,
"mean_token_accuracy": 0.17040918320417403,
"num_tokens": 8691571.0,
"step": 3800
},
{
"entropy": 5.851473236083985,
"epoch": 0.3761368129695532,
"grad_norm": 1.0078125,
"learning_rate": 0.00049913018606349,
"loss": 5.7018,
"mean_token_accuracy": 0.17158605754375458,
"num_tokens": 8703262.0,
"step": 3805
},
{
"entropy": 5.95896167755127,
"epoch": 0.3766310794780546,
"grad_norm": 0.875,
"learning_rate": 0.0004991270832600512,
"loss": 5.7957,
"mean_token_accuracy": 0.15546323359012604,
"num_tokens": 8714799.0,
"step": 3810
},
{
"entropy": 5.970942783355713,
"epoch": 0.37712534598655595,
"grad_norm": 0.984375,
"learning_rate": 0.0004991239749430514,
"loss": 5.7931,
"mean_token_accuracy": 0.16242330521345139,
"num_tokens": 8726417.0,
"step": 3815
},
{
"entropy": 5.813585662841797,
"epoch": 0.3776196124950573,
"grad_norm": 0.9296875,
"learning_rate": 0.000499120861112567,
"loss": 5.6455,
"mean_token_accuracy": 0.17116913944482803,
"num_tokens": 8738572.0,
"step": 3820
},
{
"entropy": 6.012410116195679,
"epoch": 0.3781138790035587,
"grad_norm": 0.89453125,
"learning_rate": 0.0004991177417686747,
"loss": 5.8212,
"mean_token_accuracy": 0.15966544151306153,
"num_tokens": 8750557.0,
"step": 3825
},
{
"entropy": 5.920189714431762,
"epoch": 0.3786081455120601,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991146169114511,
"loss": 5.6061,
"mean_token_accuracy": 0.16634851098060607,
"num_tokens": 8761921.0,
"step": 3830
},
{
"entropy": 5.795895385742187,
"epoch": 0.3791024120205615,
"grad_norm": 0.96875,
"learning_rate": 0.0004991114865409732,
"loss": 5.6987,
"mean_token_accuracy": 0.16617852002382277,
"num_tokens": 8772705.0,
"step": 3835
},
{
"entropy": 5.9335143089294435,
"epoch": 0.37959667852906287,
"grad_norm": 0.94921875,
"learning_rate": 0.000499108350657318,
"loss": 5.7191,
"mean_token_accuracy": 0.16396770179271697,
"num_tokens": 8783563.0,
"step": 3840
},
{
"entropy": 5.827181816101074,
"epoch": 0.38009094503756424,
"grad_norm": 0.96875,
"learning_rate": 0.0004991052092605627,
"loss": 5.5953,
"mean_token_accuracy": 0.17345058768987656,
"num_tokens": 8794171.0,
"step": 3845
},
{
"entropy": 5.871563816070557,
"epoch": 0.3805852115460656,
"grad_norm": 1.015625,
"learning_rate": 0.0004991020623507842,
"loss": 5.7123,
"mean_token_accuracy": 0.16282366067171097,
"num_tokens": 8806487.0,
"step": 3850
},
{
"entropy": 5.900928640365601,
"epoch": 0.38107947805456704,
"grad_norm": 0.8828125,
"learning_rate": 0.0004990989099280604,
"loss": 5.7145,
"mean_token_accuracy": 0.16446596533060073,
"num_tokens": 8817819.0,
"step": 3855
},
{
"entropy": 5.923869180679321,
"epoch": 0.3815737445630684,
"grad_norm": 0.96484375,
"learning_rate": 0.0004990957519924686,
"loss": 5.6631,
"mean_token_accuracy": 0.16552813053131105,
"num_tokens": 8828485.0,
"step": 3860
},
{
"entropy": 5.869322299957275,
"epoch": 0.3820680110715698,
"grad_norm": 0.890625,
"learning_rate": 0.0004990925885440865,
"loss": 5.7374,
"mean_token_accuracy": 0.1641053080558777,
"num_tokens": 8840179.0,
"step": 3865
},
{
"entropy": 5.883348655700684,
"epoch": 0.38256227758007116,
"grad_norm": 0.94921875,
"learning_rate": 0.0004990894195829921,
"loss": 5.7299,
"mean_token_accuracy": 0.17041839957237243,
"num_tokens": 8851813.0,
"step": 3870
},
{
"entropy": 5.874509334564209,
"epoch": 0.38305654408857254,
"grad_norm": 0.953125,
"learning_rate": 0.0004990862451092631,
"loss": 5.673,
"mean_token_accuracy": 0.16271054595708848,
"num_tokens": 8863072.0,
"step": 3875
},
{
"entropy": 5.908667039871216,
"epoch": 0.38355081059707397,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990830651229776,
"loss": 5.6787,
"mean_token_accuracy": 0.16601400971412658,
"num_tokens": 8873510.0,
"step": 3880
},
{
"entropy": 5.83857684135437,
"epoch": 0.38404507710557534,
"grad_norm": 0.9375,
"learning_rate": 0.0004990798796242142,
"loss": 5.624,
"mean_token_accuracy": 0.16807291507720948,
"num_tokens": 8884296.0,
"step": 3885
},
{
"entropy": 5.888775205612182,
"epoch": 0.3845393436140767,
"grad_norm": 0.91015625,
"learning_rate": 0.0004990766886130508,
"loss": 5.6805,
"mean_token_accuracy": 0.17192094773054123,
"num_tokens": 8894208.0,
"step": 3890
},
{
"entropy": 6.020419263839722,
"epoch": 0.3850336101225781,
"grad_norm": 0.98828125,
"learning_rate": 0.0004990734920895662,
"loss": 5.8482,
"mean_token_accuracy": 0.15748704373836517,
"num_tokens": 8905703.0,
"step": 3895
},
{
"entropy": 5.877511692047119,
"epoch": 0.38552787663107946,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990702900538388,
"loss": 5.6553,
"mean_token_accuracy": 0.16876436471939088,
"num_tokens": 8918077.0,
"step": 3900
},
{
"entropy": 5.776277256011963,
"epoch": 0.3860221431395809,
"grad_norm": 0.81640625,
"learning_rate": 0.0004990670825059477,
"loss": 5.6144,
"mean_token_accuracy": 0.165885029733181,
"num_tokens": 8930240.0,
"step": 3905
},
{
"entropy": 5.9259237289428714,
"epoch": 0.38651640964808226,
"grad_norm": 0.91015625,
"learning_rate": 0.0004990638694459715,
"loss": 5.6923,
"mean_token_accuracy": 0.17438057214021682,
"num_tokens": 8941421.0,
"step": 3910
},
{
"entropy": 5.886525201797485,
"epoch": 0.38701067615658363,
"grad_norm": 0.94921875,
"learning_rate": 0.0004990606508739893,
"loss": 5.7135,
"mean_token_accuracy": 0.1625737339258194,
"num_tokens": 8952296.0,
"step": 3915
},
{
"entropy": 5.887601089477539,
"epoch": 0.387504942665085,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990574267900804,
"loss": 5.7245,
"mean_token_accuracy": 0.16023652553558348,
"num_tokens": 8963167.0,
"step": 3920
},
{
"entropy": 5.860787200927734,
"epoch": 0.3879992091735864,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990541971943241,
"loss": 5.6944,
"mean_token_accuracy": 0.16610553562641145,
"num_tokens": 8974405.0,
"step": 3925
},
{
"entropy": 5.813625812530518,
"epoch": 0.3884934756820878,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990509620867997,
"loss": 5.6097,
"mean_token_accuracy": 0.17485980838537216,
"num_tokens": 8986012.0,
"step": 3930
},
{
"entropy": 5.794552993774414,
"epoch": 0.3889877421905892,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990477214675868,
"loss": 5.5903,
"mean_token_accuracy": 0.1692924752831459,
"num_tokens": 8996759.0,
"step": 3935
},
{
"entropy": 5.887366104125976,
"epoch": 0.38948200869909055,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990444753367653,
"loss": 5.7137,
"mean_token_accuracy": 0.16604164093732834,
"num_tokens": 9007215.0,
"step": 3940
},
{
"entropy": 5.832546091079712,
"epoch": 0.3899762752075919,
"grad_norm": 0.90625,
"learning_rate": 0.0004990412236944149,
"loss": 5.6156,
"mean_token_accuracy": 0.18009421676397325,
"num_tokens": 9018281.0,
"step": 3945
},
{
"entropy": 5.932539606094361,
"epoch": 0.3904705417160933,
"grad_norm": 0.85546875,
"learning_rate": 0.0004990379665406157,
"loss": 5.7766,
"mean_token_accuracy": 0.16505084782838822,
"num_tokens": 9029312.0,
"step": 3950
},
{
"entropy": 5.949569416046143,
"epoch": 0.39096480822459473,
"grad_norm": 1.015625,
"learning_rate": 0.0004990347038754477,
"loss": 5.7419,
"mean_token_accuracy": 0.1618492618203163,
"num_tokens": 9040273.0,
"step": 3955
},
{
"entropy": 5.916225481033325,
"epoch": 0.3914590747330961,
"grad_norm": 0.796875,
"learning_rate": 0.0004990314356989912,
"loss": 5.6798,
"mean_token_accuracy": 0.16216366291046141,
"num_tokens": 9052571.0,
"step": 3960
},
{
"entropy": 5.783906412124634,
"epoch": 0.3919533412415975,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990281620113268,
"loss": 5.6394,
"mean_token_accuracy": 0.16567644774913787,
"num_tokens": 9063531.0,
"step": 3965
},
{
"entropy": 5.918298578262329,
"epoch": 0.39244760775009885,
"grad_norm": 0.984375,
"learning_rate": 0.0004990248828125347,
"loss": 5.6828,
"mean_token_accuracy": 0.1612454503774643,
"num_tokens": 9075404.0,
"step": 3970
},
{
"entropy": 5.828043699264526,
"epoch": 0.3929418742586002,
"grad_norm": 0.9140625,
"learning_rate": 0.0004990215981026958,
"loss": 5.7003,
"mean_token_accuracy": 0.16850079447031022,
"num_tokens": 9086570.0,
"step": 3975
},
{
"entropy": 5.859482908248902,
"epoch": 0.3934361407671016,
"grad_norm": 0.87109375,
"learning_rate": 0.0004990183078818908,
"loss": 5.588,
"mean_token_accuracy": 0.17108299285173417,
"num_tokens": 9097993.0,
"step": 3980
},
{
"entropy": 5.892006063461304,
"epoch": 0.393930407275603,
"grad_norm": 0.9375,
"learning_rate": 0.0004990150121502007,
"loss": 5.6534,
"mean_token_accuracy": 0.1727806583046913,
"num_tokens": 9108783.0,
"step": 3985
},
{
"entropy": 5.804200029373169,
"epoch": 0.3944246737841044,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990117109077066,
"loss": 5.7105,
"mean_token_accuracy": 0.1658673793077469,
"num_tokens": 9120039.0,
"step": 3990
},
{
"entropy": 5.9291599750518795,
"epoch": 0.39491894029260577,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990084041544895,
"loss": 5.6662,
"mean_token_accuracy": 0.17073300033807753,
"num_tokens": 9131543.0,
"step": 3995
},
{
"entropy": 5.854316997528076,
"epoch": 0.39541320680110714,
"grad_norm": 0.8515625,
"learning_rate": 0.000499005091890631,
"loss": 5.6764,
"mean_token_accuracy": 0.16651399731636046,
"num_tokens": 9142968.0,
"step": 4000
},
{
"entropy": 5.838335704803467,
"epoch": 0.3959074733096085,
"grad_norm": 0.83984375,
"learning_rate": 0.0004990017741162125,
"loss": 5.7632,
"mean_token_accuracy": 0.16216625273227692,
"num_tokens": 9155705.0,
"step": 4005
},
{
"entropy": 5.960814714431763,
"epoch": 0.39640173981810994,
"grad_norm": 0.88671875,
"learning_rate": 0.0004989984508313156,
"loss": 5.7528,
"mean_token_accuracy": 0.15642520487308503,
"num_tokens": 9167575.0,
"step": 4010
},
{
"entropy": 5.833995151519775,
"epoch": 0.3968960063266113,
"grad_norm": 0.89453125,
"learning_rate": 0.000498995122036022,
"loss": 5.7127,
"mean_token_accuracy": 0.1676729664206505,
"num_tokens": 9178493.0,
"step": 4015
},
{
"entropy": 5.956012725830078,
"epoch": 0.3973902728351127,
"grad_norm": 0.94921875,
"learning_rate": 0.0004989917877304138,
"loss": 5.7754,
"mean_token_accuracy": 0.1605817899107933,
"num_tokens": 9189677.0,
"step": 4020
},
{
"entropy": 5.868053340911866,
"epoch": 0.39788453934361406,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989884479145727,
"loss": 5.6071,
"mean_token_accuracy": 0.17071904242038727,
"num_tokens": 9199179.0,
"step": 4025
},
{
"entropy": 5.88495979309082,
"epoch": 0.39837880585211544,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989851025885812,
"loss": 5.677,
"mean_token_accuracy": 0.16467752456665039,
"num_tokens": 9210689.0,
"step": 4030
},
{
"entropy": 5.871729373931885,
"epoch": 0.39887307236061686,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989817517525212,
"loss": 5.6344,
"mean_token_accuracy": 0.1702873781323433,
"num_tokens": 9221710.0,
"step": 4035
},
{
"entropy": 5.912847375869751,
"epoch": 0.39936733886911824,
"grad_norm": 0.91796875,
"learning_rate": 0.0004989783954064755,
"loss": 5.7375,
"mean_token_accuracy": 0.16328861117362975,
"num_tokens": 9234079.0,
"step": 4040
},
{
"entropy": 5.854098844528198,
"epoch": 0.3998616053776196,
"grad_norm": 0.90234375,
"learning_rate": 0.0004989750335505265,
"loss": 5.6728,
"mean_token_accuracy": 0.16908831596374513,
"num_tokens": 9245050.0,
"step": 4045
},
{
"entropy": 5.98025598526001,
"epoch": 0.400355871886121,
"grad_norm": 1.0,
"learning_rate": 0.0004989716661847571,
"loss": 5.7485,
"mean_token_accuracy": 0.16286372840404512,
"num_tokens": 9256455.0,
"step": 4050
},
{
"entropy": 5.82414436340332,
"epoch": 0.40085013839462236,
"grad_norm": 0.9453125,
"learning_rate": 0.0004989682933092497,
"loss": 5.641,
"mean_token_accuracy": 0.1701007068157196,
"num_tokens": 9267722.0,
"step": 4055
},
{
"entropy": 5.808041715621949,
"epoch": 0.4013444049031238,
"grad_norm": 1.0546875,
"learning_rate": 0.0004989649149240877,
"loss": 5.6701,
"mean_token_accuracy": 0.16336237341165544,
"num_tokens": 9277533.0,
"step": 4060
},
{
"entropy": 5.926660394668579,
"epoch": 0.40183867141162516,
"grad_norm": 0.875,
"learning_rate": 0.000498961531029354,
"loss": 5.7165,
"mean_token_accuracy": 0.1723726972937584,
"num_tokens": 9290052.0,
"step": 4065
},
{
"entropy": 5.907356834411621,
"epoch": 0.40233293792012653,
"grad_norm": 0.91796875,
"learning_rate": 0.000498958141625132,
"loss": 5.7054,
"mean_token_accuracy": 0.16776670813560485,
"num_tokens": 9302201.0,
"step": 4070
},
{
"entropy": 5.887616729736328,
"epoch": 0.4028272044286279,
"grad_norm": 1.03125,
"learning_rate": 0.0004989547467115047,
"loss": 5.7843,
"mean_token_accuracy": 0.15944056063890458,
"num_tokens": 9313215.0,
"step": 4075
},
{
"entropy": 5.9510393142700195,
"epoch": 0.4033214709371293,
"grad_norm": 0.859375,
"learning_rate": 0.0004989513462885561,
"loss": 5.8032,
"mean_token_accuracy": 0.155883726477623,
"num_tokens": 9324098.0,
"step": 4080
},
{
"entropy": 5.978319501876831,
"epoch": 0.4038157374456307,
"grad_norm": 0.8203125,
"learning_rate": 0.0004989479403563696,
"loss": 5.6837,
"mean_token_accuracy": 0.1646607756614685,
"num_tokens": 9336048.0,
"step": 4085
},
{
"entropy": 5.8422708988189695,
"epoch": 0.4043100039541321,
"grad_norm": 0.8671875,
"learning_rate": 0.0004989445289150289,
"loss": 5.6499,
"mean_token_accuracy": 0.16654919385910033,
"num_tokens": 9347723.0,
"step": 4090
},
{
"entropy": 5.829070520401001,
"epoch": 0.40480427046263345,
"grad_norm": 0.859375,
"learning_rate": 0.0004989411119646183,
"loss": 5.7007,
"mean_token_accuracy": 0.16050561815500258,
"num_tokens": 9360588.0,
"step": 4095
},
{
"entropy": 5.922001028060913,
"epoch": 0.4052985369711348,
"grad_norm": 0.98046875,
"learning_rate": 0.0004989376895052214,
"loss": 5.6551,
"mean_token_accuracy": 0.16239948570728302,
"num_tokens": 9371798.0,
"step": 4100
},
{
"entropy": 5.878776121139526,
"epoch": 0.4057928034796362,
"grad_norm": 0.94921875,
"learning_rate": 0.0004989342615369226,
"loss": 5.6709,
"mean_token_accuracy": 0.1724824145436287,
"num_tokens": 9382421.0,
"step": 4105
},
{
"entropy": 5.878492259979248,
"epoch": 0.40628706998813763,
"grad_norm": 1.0,
"learning_rate": 0.0004989308280598063,
"loss": 5.5976,
"mean_token_accuracy": 0.1707899510860443,
"num_tokens": 9392832.0,
"step": 4110
},
{
"entropy": 5.794064044952393,
"epoch": 0.406781336496639,
"grad_norm": 0.86328125,
"learning_rate": 0.0004989273890739569,
"loss": 5.6036,
"mean_token_accuracy": 0.16852443814277648,
"num_tokens": 9405309.0,
"step": 4115
},
{
"entropy": 5.773290777206421,
"epoch": 0.4072756030051404,
"grad_norm": 0.90625,
"learning_rate": 0.000498923944579459,
"loss": 5.6052,
"mean_token_accuracy": 0.1808168187737465,
"num_tokens": 9416707.0,
"step": 4120
},
{
"entropy": 5.88574185371399,
"epoch": 0.40776986951364175,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989204945763972,
"loss": 5.7362,
"mean_token_accuracy": 0.1645051822066307,
"num_tokens": 9427684.0,
"step": 4125
},
{
"entropy": 5.875238704681396,
"epoch": 0.4082641360221431,
"grad_norm": 0.88671875,
"learning_rate": 0.0004989170390648567,
"loss": 5.685,
"mean_token_accuracy": 0.1686164915561676,
"num_tokens": 9439073.0,
"step": 4130
},
{
"entropy": 5.833017063140869,
"epoch": 0.40875840253064455,
"grad_norm": 0.84375,
"learning_rate": 0.0004989135780449222,
"loss": 5.6132,
"mean_token_accuracy": 0.1716807648539543,
"num_tokens": 9450266.0,
"step": 4135
},
{
"entropy": 5.776205968856812,
"epoch": 0.4092526690391459,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989101115166789,
"loss": 5.6083,
"mean_token_accuracy": 0.17508271783590318,
"num_tokens": 9462662.0,
"step": 4140
},
{
"entropy": 5.8494264602661135,
"epoch": 0.4097469355476473,
"grad_norm": 0.8828125,
"learning_rate": 0.0004989066394802122,
"loss": 5.6922,
"mean_token_accuracy": 0.1669348806142807,
"num_tokens": 9474338.0,
"step": 4145
},
{
"entropy": 5.889415454864502,
"epoch": 0.41024120205614867,
"grad_norm": 0.90625,
"learning_rate": 0.0004989031619356074,
"loss": 5.6803,
"mean_token_accuracy": 0.17034045308828355,
"num_tokens": 9486353.0,
"step": 4150
},
{
"entropy": 5.844692850112915,
"epoch": 0.41073546856465004,
"grad_norm": 0.8515625,
"learning_rate": 0.00049889967888295,
"loss": 5.643,
"mean_token_accuracy": 0.16728283166885377,
"num_tokens": 9497464.0,
"step": 4155
},
{
"entropy": 5.854997682571411,
"epoch": 0.41122973507315147,
"grad_norm": 0.921875,
"learning_rate": 0.0004988961903223258,
"loss": 5.6116,
"mean_token_accuracy": 0.17531556785106658,
"num_tokens": 9508037.0,
"step": 4160
},
{
"entropy": 5.7995391368865965,
"epoch": 0.41172400158165284,
"grad_norm": 0.96875,
"learning_rate": 0.0004988926962538207,
"loss": 5.649,
"mean_token_accuracy": 0.17245213091373443,
"num_tokens": 9518653.0,
"step": 4165
},
{
"entropy": 5.82932858467102,
"epoch": 0.4122182680901542,
"grad_norm": 0.9296875,
"learning_rate": 0.0004988891966775204,
"loss": 5.649,
"mean_token_accuracy": 0.17256172150373458,
"num_tokens": 9529225.0,
"step": 4170
},
{
"entropy": 5.87958402633667,
"epoch": 0.4127125345986556,
"grad_norm": 1.0,
"learning_rate": 0.0004988856915935112,
"loss": 5.6121,
"mean_token_accuracy": 0.16461459249258042,
"num_tokens": 9541010.0,
"step": 4175
},
{
"entropy": 5.78000111579895,
"epoch": 0.41320680110715696,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988821810018791,
"loss": 5.7102,
"mean_token_accuracy": 0.16528748869895935,
"num_tokens": 9552061.0,
"step": 4180
},
{
"entropy": 5.9790253162384035,
"epoch": 0.41370106761565834,
"grad_norm": 0.94921875,
"learning_rate": 0.0004988786649027108,
"loss": 5.7987,
"mean_token_accuracy": 0.15660196989774705,
"num_tokens": 9563970.0,
"step": 4185
},
{
"entropy": 5.864908266067505,
"epoch": 0.41419533412415976,
"grad_norm": 0.875,
"learning_rate": 0.0004988751432960926,
"loss": 5.5784,
"mean_token_accuracy": 0.17485955953598023,
"num_tokens": 9575381.0,
"step": 4190
},
{
"entropy": 5.86647777557373,
"epoch": 0.41468960063266114,
"grad_norm": 0.91796875,
"learning_rate": 0.0004988716161821111,
"loss": 5.6286,
"mean_token_accuracy": 0.17472680807113647,
"num_tokens": 9586821.0,
"step": 4195
},
{
"entropy": 5.876406335830689,
"epoch": 0.4151838671411625,
"grad_norm": 0.890625,
"learning_rate": 0.0004988680835608531,
"loss": 5.6971,
"mean_token_accuracy": 0.16150547564029694,
"num_tokens": 9599580.0,
"step": 4200
},
{
"entropy": 5.784275007247925,
"epoch": 0.4156781336496639,
"grad_norm": 0.87109375,
"learning_rate": 0.0004988645454324055,
"loss": 5.687,
"mean_token_accuracy": 0.16584365963935851,
"num_tokens": 9611495.0,
"step": 4205
},
{
"entropy": 5.936398077011108,
"epoch": 0.41617240015816526,
"grad_norm": 0.8671875,
"learning_rate": 0.0004988610017968554,
"loss": 5.6553,
"mean_token_accuracy": 0.17303829044103622,
"num_tokens": 9623713.0,
"step": 4210
},
{
"entropy": 5.818154144287109,
"epoch": 0.4166666666666667,
"grad_norm": 1.015625,
"learning_rate": 0.0004988574526542899,
"loss": 5.6512,
"mean_token_accuracy": 0.1765165865421295,
"num_tokens": 9633922.0,
"step": 4215
},
{
"entropy": 5.824317121505738,
"epoch": 0.41716093317516806,
"grad_norm": 0.93359375,
"learning_rate": 0.0004988538980047963,
"loss": 5.7404,
"mean_token_accuracy": 0.16559439152479172,
"num_tokens": 9645641.0,
"step": 4220
},
{
"entropy": 5.841555404663086,
"epoch": 0.41765519968366943,
"grad_norm": 0.92578125,
"learning_rate": 0.0004988503378484622,
"loss": 5.6391,
"mean_token_accuracy": 0.17398539781570435,
"num_tokens": 9656927.0,
"step": 4225
},
{
"entropy": 5.814691019058228,
"epoch": 0.4181494661921708,
"grad_norm": 0.9140625,
"learning_rate": 0.000498846772185375,
"loss": 5.6168,
"mean_token_accuracy": 0.1729745402932167,
"num_tokens": 9669148.0,
"step": 4230
},
{
"entropy": 5.893791246414184,
"epoch": 0.4186437327006722,
"grad_norm": 0.875,
"learning_rate": 0.0004988432010156224,
"loss": 5.7396,
"mean_token_accuracy": 0.15868683755397797,
"num_tokens": 9680620.0,
"step": 4235
},
{
"entropy": 5.925040054321289,
"epoch": 0.4191379992091736,
"grad_norm": 1.0078125,
"learning_rate": 0.0004988396243392924,
"loss": 5.6849,
"mean_token_accuracy": 0.16457130312919616,
"num_tokens": 9690984.0,
"step": 4240
},
{
"entropy": 5.835089540481567,
"epoch": 0.419632265717675,
"grad_norm": 0.90625,
"learning_rate": 0.000498836042156473,
"loss": 5.6622,
"mean_token_accuracy": 0.17159282118082048,
"num_tokens": 9701561.0,
"step": 4245
},
{
"entropy": 5.884567213058472,
"epoch": 0.42012653222617635,
"grad_norm": 0.91796875,
"learning_rate": 0.0004988324544672521,
"loss": 5.6556,
"mean_token_accuracy": 0.1672545090317726,
"num_tokens": 9712780.0,
"step": 4250
},
{
"entropy": 5.868431282043457,
"epoch": 0.4206207987346777,
"grad_norm": 0.984375,
"learning_rate": 0.0004988288612717182,
"loss": 5.6469,
"mean_token_accuracy": 0.16327356547117233,
"num_tokens": 9724547.0,
"step": 4255
},
{
"entropy": 5.844528484344482,
"epoch": 0.4211150652431791,
"grad_norm": 0.90625,
"learning_rate": 0.0004988252625699594,
"loss": 5.6823,
"mean_token_accuracy": 0.16297900527715684,
"num_tokens": 9736897.0,
"step": 4260
},
{
"entropy": 5.867266893386841,
"epoch": 0.4216093317516805,
"grad_norm": 0.88671875,
"learning_rate": 0.0004988216583620647,
"loss": 5.6562,
"mean_token_accuracy": 0.17023655623197556,
"num_tokens": 9748214.0,
"step": 4265
},
{
"entropy": 5.869999837875366,
"epoch": 0.4221035982601819,
"grad_norm": 0.96484375,
"learning_rate": 0.0004988180486481223,
"loss": 5.6769,
"mean_token_accuracy": 0.17015081495046616,
"num_tokens": 9759602.0,
"step": 4270
},
{
"entropy": 5.8174549579620365,
"epoch": 0.4225978647686833,
"grad_norm": 0.97265625,
"learning_rate": 0.0004988144334282211,
"loss": 5.6345,
"mean_token_accuracy": 0.17117137610912322,
"num_tokens": 9771645.0,
"step": 4275
},
{
"entropy": 5.859453535079956,
"epoch": 0.42309213127718465,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988108127024503,
"loss": 5.6351,
"mean_token_accuracy": 0.16911172568798066,
"num_tokens": 9783504.0,
"step": 4280
},
{
"entropy": 5.893519067764283,
"epoch": 0.423586397785686,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988071864708987,
"loss": 5.7568,
"mean_token_accuracy": 0.16314952075481415,
"num_tokens": 9795572.0,
"step": 4285
},
{
"entropy": 5.836259841918945,
"epoch": 0.42408066429418745,
"grad_norm": 0.921875,
"learning_rate": 0.0004988035547336555,
"loss": 5.6251,
"mean_token_accuracy": 0.1675659477710724,
"num_tokens": 9806398.0,
"step": 4290
},
{
"entropy": 5.779581737518311,
"epoch": 0.4245749308026888,
"grad_norm": 0.9453125,
"learning_rate": 0.0004987999174908104,
"loss": 5.6195,
"mean_token_accuracy": 0.1663268983364105,
"num_tokens": 9817171.0,
"step": 4295
},
{
"entropy": 5.922465467453003,
"epoch": 0.4250691973111902,
"grad_norm": 1.09375,
"learning_rate": 0.0004987962747424524,
"loss": 5.6613,
"mean_token_accuracy": 0.16931165456771852,
"num_tokens": 9827535.0,
"step": 4300
},
{
"entropy": 5.815274333953857,
"epoch": 0.42556346381969157,
"grad_norm": 0.890625,
"learning_rate": 0.0004987926264886713,
"loss": 5.6386,
"mean_token_accuracy": 0.17092007547616958,
"num_tokens": 9839619.0,
"step": 4305
},
{
"entropy": 5.857655954360962,
"epoch": 0.42605773032819294,
"grad_norm": 1.0,
"learning_rate": 0.000498788972729557,
"loss": 5.6392,
"mean_token_accuracy": 0.16871783286333084,
"num_tokens": 9850100.0,
"step": 4310
},
{
"entropy": 5.928799676895141,
"epoch": 0.42655199683669437,
"grad_norm": 0.97265625,
"learning_rate": 0.0004987853134651992,
"loss": 5.7204,
"mean_token_accuracy": 0.15432474240660668,
"num_tokens": 9862228.0,
"step": 4315
},
{
"entropy": 5.837745666503906,
"epoch": 0.42704626334519574,
"grad_norm": 0.953125,
"learning_rate": 0.000498781648695688,
"loss": 5.6028,
"mean_token_accuracy": 0.17573820501565934,
"num_tokens": 9873080.0,
"step": 4320
},
{
"entropy": 5.808217191696167,
"epoch": 0.4275405298536971,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987779784211135,
"loss": 5.6142,
"mean_token_accuracy": 0.16612853109836578,
"num_tokens": 9884735.0,
"step": 4325
},
{
"entropy": 5.865272378921508,
"epoch": 0.4280347963621985,
"grad_norm": 0.96875,
"learning_rate": 0.000498774302641566,
"loss": 5.6466,
"mean_token_accuracy": 0.16833435297012328,
"num_tokens": 9895018.0,
"step": 4330
},
{
"entropy": 5.842509174346924,
"epoch": 0.42852906287069986,
"grad_norm": 0.94921875,
"learning_rate": 0.0004987706213571359,
"loss": 5.6089,
"mean_token_accuracy": 0.17211945354938507,
"num_tokens": 9906838.0,
"step": 4335
},
{
"entropy": 5.889233684539795,
"epoch": 0.4290233293792013,
"grad_norm": 0.9453125,
"learning_rate": 0.0004987669345679139,
"loss": 5.6678,
"mean_token_accuracy": 0.16429460048675537,
"num_tokens": 9919556.0,
"step": 4340
},
{
"entropy": 5.859847974777222,
"epoch": 0.42951759588770266,
"grad_norm": 1.0234375,
"learning_rate": 0.0004987632422739904,
"loss": 5.5661,
"mean_token_accuracy": 0.16938609331846238,
"num_tokens": 9929783.0,
"step": 4345
},
{
"entropy": 5.787507438659668,
"epoch": 0.43001186239620404,
"grad_norm": 0.87890625,
"learning_rate": 0.0004987595444754567,
"loss": 5.621,
"mean_token_accuracy": 0.17016697227954863,
"num_tokens": 9940894.0,
"step": 4350
},
{
"entropy": 5.81076602935791,
"epoch": 0.4305061289047054,
"grad_norm": 0.9765625,
"learning_rate": 0.0004987558411724032,
"loss": 5.6017,
"mean_token_accuracy": 0.1788248375058174,
"num_tokens": 9951973.0,
"step": 4355
},
{
"entropy": 5.868822050094605,
"epoch": 0.4310003954132068,
"grad_norm": 0.90625,
"learning_rate": 0.0004987521323649215,
"loss": 5.6667,
"mean_token_accuracy": 0.16563129425048828,
"num_tokens": 9963304.0,
"step": 4360
},
{
"entropy": 5.841874361038208,
"epoch": 0.4314946619217082,
"grad_norm": 0.99609375,
"learning_rate": 0.0004987484180531026,
"loss": 5.6748,
"mean_token_accuracy": 0.17026820480823518,
"num_tokens": 9973981.0,
"step": 4365
},
{
"entropy": 5.889091730117798,
"epoch": 0.4319889284302096,
"grad_norm": 0.91015625,
"learning_rate": 0.0004987446982370379,
"loss": 5.7015,
"mean_token_accuracy": 0.16372227519750596,
"num_tokens": 9986461.0,
"step": 4370
},
{
"entropy": 5.871285724639892,
"epoch": 0.43248319493871096,
"grad_norm": 0.890625,
"learning_rate": 0.000498740972916819,
"loss": 5.6439,
"mean_token_accuracy": 0.16812467724084854,
"num_tokens": 9997973.0,
"step": 4375
},
{
"entropy": 5.843570899963379,
"epoch": 0.43297746144721233,
"grad_norm": 0.93359375,
"learning_rate": 0.0004987372420925373,
"loss": 5.6721,
"mean_token_accuracy": 0.16844396740198136,
"num_tokens": 10009572.0,
"step": 4380
},
{
"entropy": 5.805202341079712,
"epoch": 0.4334717279557137,
"grad_norm": 1.0234375,
"learning_rate": 0.0004987335057642847,
"loss": 5.5492,
"mean_token_accuracy": 0.1798780858516693,
"num_tokens": 10020437.0,
"step": 4385
},
{
"entropy": 5.822914123535156,
"epoch": 0.43396599446421513,
"grad_norm": 0.82421875,
"learning_rate": 0.0004987297639321533,
"loss": 5.6587,
"mean_token_accuracy": 0.1684048891067505,
"num_tokens": 10031827.0,
"step": 4390
},
{
"entropy": 5.822333097457886,
"epoch": 0.4344602609727165,
"grad_norm": 0.96875,
"learning_rate": 0.0004987260165962349,
"loss": 5.6485,
"mean_token_accuracy": 0.16710660308599473,
"num_tokens": 10043833.0,
"step": 4395
},
{
"entropy": 5.837571001052856,
"epoch": 0.4349545274812179,
"grad_norm": 0.91015625,
"learning_rate": 0.0004987222637566218,
"loss": 5.5826,
"mean_token_accuracy": 0.17188626825809478,
"num_tokens": 10054645.0,
"step": 4400
},
{
"entropy": 5.769827222824096,
"epoch": 0.43544879398971925,
"grad_norm": 0.90234375,
"learning_rate": 0.0004987185054134064,
"loss": 5.5614,
"mean_token_accuracy": 0.16493925750255584,
"num_tokens": 10065178.0,
"step": 4405
},
{
"entropy": 5.826924991607666,
"epoch": 0.4359430604982206,
"grad_norm": 0.94140625,
"learning_rate": 0.000498714741566681,
"loss": 5.584,
"mean_token_accuracy": 0.17811701595783233,
"num_tokens": 10076701.0,
"step": 4410
},
{
"entropy": 5.780032110214234,
"epoch": 0.436437327006722,
"grad_norm": 0.9609375,
"learning_rate": 0.0004987109722165383,
"loss": 5.59,
"mean_token_accuracy": 0.1700105130672455,
"num_tokens": 10087959.0,
"step": 4415
},
{
"entropy": 5.880767917633056,
"epoch": 0.4369315935152234,
"grad_norm": 1.03125,
"learning_rate": 0.0004987071973630708,
"loss": 5.5943,
"mean_token_accuracy": 0.16818706542253495,
"num_tokens": 10099463.0,
"step": 4420
},
{
"entropy": 5.803549814224243,
"epoch": 0.4374258600237248,
"grad_norm": 0.95703125,
"learning_rate": 0.0004987034170063717,
"loss": 5.6834,
"mean_token_accuracy": 0.1695466473698616,
"num_tokens": 10111384.0,
"step": 4425
},
{
"entropy": 5.801593875885009,
"epoch": 0.4379201265322262,
"grad_norm": 0.9140625,
"learning_rate": 0.0004986996311465338,
"loss": 5.6118,
"mean_token_accuracy": 0.17462965697050095,
"num_tokens": 10121838.0,
"step": 4430
},
{
"entropy": 5.847334098815918,
"epoch": 0.43841439304072755,
"grad_norm": 0.859375,
"learning_rate": 0.0004986958397836503,
"loss": 5.5513,
"mean_token_accuracy": 0.1757386162877083,
"num_tokens": 10132225.0,
"step": 4435
},
{
"entropy": 5.764697647094726,
"epoch": 0.4389086595492289,
"grad_norm": 0.8984375,
"learning_rate": 0.0004986920429178144,
"loss": 5.5531,
"mean_token_accuracy": 0.16869834214448928,
"num_tokens": 10143114.0,
"step": 4440
},
{
"entropy": 5.858071994781494,
"epoch": 0.43940292605773035,
"grad_norm": 0.86328125,
"learning_rate": 0.0004986882405491195,
"loss": 5.7597,
"mean_token_accuracy": 0.15645925253629683,
"num_tokens": 10155057.0,
"step": 4445
},
{
"entropy": 5.825079345703125,
"epoch": 0.4398971925662317,
"grad_norm": 0.85546875,
"learning_rate": 0.0004986844326776593,
"loss": 5.5233,
"mean_token_accuracy": 0.1803316131234169,
"num_tokens": 10166816.0,
"step": 4450
},
{
"entropy": 5.74500150680542,
"epoch": 0.4403914590747331,
"grad_norm": 0.83984375,
"learning_rate": 0.0004986806193035272,
"loss": 5.5681,
"mean_token_accuracy": 0.18103527575731276,
"num_tokens": 10179433.0,
"step": 4455
},
{
"entropy": 5.826761674880982,
"epoch": 0.44088572558323447,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986768004268173,
"loss": 5.5772,
"mean_token_accuracy": 0.17316691279411317,
"num_tokens": 10190784.0,
"step": 4460
},
{
"entropy": 5.801952075958252,
"epoch": 0.44137999209173584,
"grad_norm": 0.98046875,
"learning_rate": 0.0004986729760476234,
"loss": 5.6205,
"mean_token_accuracy": 0.16593633741140365,
"num_tokens": 10201941.0,
"step": 4465
},
{
"entropy": 5.82014422416687,
"epoch": 0.44187425860023727,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986691461660394,
"loss": 5.5956,
"mean_token_accuracy": 0.1733821839094162,
"num_tokens": 10214029.0,
"step": 4470
},
{
"entropy": 5.777339315414428,
"epoch": 0.44236852510873864,
"grad_norm": 0.9609375,
"learning_rate": 0.0004986653107821599,
"loss": 5.6277,
"mean_token_accuracy": 0.1750914230942726,
"num_tokens": 10225585.0,
"step": 4475
},
{
"entropy": 5.764943027496338,
"epoch": 0.44286279161724,
"grad_norm": 0.84765625,
"learning_rate": 0.0004986614698960789,
"loss": 5.5763,
"mean_token_accuracy": 0.17808845043182372,
"num_tokens": 10237753.0,
"step": 4480
},
{
"entropy": 5.887031602859497,
"epoch": 0.4433570581257414,
"grad_norm": 0.9375,
"learning_rate": 0.0004986576235078913,
"loss": 5.8251,
"mean_token_accuracy": 0.1615597724914551,
"num_tokens": 10251522.0,
"step": 4485
},
{
"entropy": 5.887323236465454,
"epoch": 0.44385132463424276,
"grad_norm": 0.99609375,
"learning_rate": 0.0004986537716176912,
"loss": 5.5271,
"mean_token_accuracy": 0.17930984050035476,
"num_tokens": 10263319.0,
"step": 4490
},
{
"entropy": 5.712326526641846,
"epoch": 0.4443455911427442,
"grad_norm": 0.953125,
"learning_rate": 0.0004986499142255738,
"loss": 5.5755,
"mean_token_accuracy": 0.16989955455064773,
"num_tokens": 10274899.0,
"step": 4495
},
{
"entropy": 5.838864421844482,
"epoch": 0.44483985765124556,
"grad_norm": 1.0859375,
"learning_rate": 0.0004986460513316338,
"loss": 5.5666,
"mean_token_accuracy": 0.17787883430719376,
"num_tokens": 10286110.0,
"step": 4500
},
{
"entropy": 5.809325695037842,
"epoch": 0.44533412415974694,
"grad_norm": 0.93359375,
"learning_rate": 0.0004986421829359662,
"loss": 5.6995,
"mean_token_accuracy": 0.16746099144220353,
"num_tokens": 10298870.0,
"step": 4505
},
{
"entropy": 5.820289134979248,
"epoch": 0.4458283906682483,
"grad_norm": 1.0,
"learning_rate": 0.0004986383090386661,
"loss": 5.6204,
"mean_token_accuracy": 0.16782621592283248,
"num_tokens": 10310326.0,
"step": 4510
},
{
"entropy": 5.8581438064575195,
"epoch": 0.4463226571767497,
"grad_norm": 0.9609375,
"learning_rate": 0.000498634429639829,
"loss": 5.5542,
"mean_token_accuracy": 0.17202099561691284,
"num_tokens": 10320899.0,
"step": 4515
},
{
"entropy": 5.825332593917847,
"epoch": 0.4468169236852511,
"grad_norm": 1.015625,
"learning_rate": 0.0004986305447395502,
"loss": 5.6123,
"mean_token_accuracy": 0.17592785209417344,
"num_tokens": 10331669.0,
"step": 4520
},
{
"entropy": 5.816992092132568,
"epoch": 0.4473111901937525,
"grad_norm": 0.9921875,
"learning_rate": 0.0004986266543379254,
"loss": 5.5885,
"mean_token_accuracy": 0.17180951982736586,
"num_tokens": 10343120.0,
"step": 4525
},
{
"entropy": 5.797379636764527,
"epoch": 0.44780545670225386,
"grad_norm": 0.9375,
"learning_rate": 0.0004986227584350501,
"loss": 5.5948,
"mean_token_accuracy": 0.17498999536037446,
"num_tokens": 10354806.0,
"step": 4530
},
{
"entropy": 5.858649969100952,
"epoch": 0.44829972321075523,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986188570310203,
"loss": 5.5961,
"mean_token_accuracy": 0.17178542762994767,
"num_tokens": 10366453.0,
"step": 4535
},
{
"entropy": 5.873257064819336,
"epoch": 0.4487939897192566,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986149501259319,
"loss": 5.703,
"mean_token_accuracy": 0.16281179934740067,
"num_tokens": 10378095.0,
"step": 4540
},
{
"entropy": 5.739948797225952,
"epoch": 0.44928825622775803,
"grad_norm": 0.921875,
"learning_rate": 0.0004986110377198812,
"loss": 5.5707,
"mean_token_accuracy": 0.17470115125179292,
"num_tokens": 10389620.0,
"step": 4545
},
{
"entropy": 5.882530689239502,
"epoch": 0.4497825227362594,
"grad_norm": 0.9921875,
"learning_rate": 0.000498607119812964,
"loss": 5.6976,
"mean_token_accuracy": 0.16568560302257537,
"num_tokens": 10400166.0,
"step": 4550
},
{
"entropy": 5.857796144485474,
"epoch": 0.4502767892447608,
"grad_norm": 1.0234375,
"learning_rate": 0.0004986031964052772,
"loss": 5.5965,
"mean_token_accuracy": 0.1668781518936157,
"num_tokens": 10411116.0,
"step": 4555
},
{
"entropy": 5.778992223739624,
"epoch": 0.45077105575326215,
"grad_norm": 0.93359375,
"learning_rate": 0.0004985992674969169,
"loss": 5.6415,
"mean_token_accuracy": 0.16282664388418197,
"num_tokens": 10423484.0,
"step": 4560
},
{
"entropy": 5.844914436340332,
"epoch": 0.4512653222617635,
"grad_norm": 0.92578125,
"learning_rate": 0.0004985953330879801,
"loss": 5.6905,
"mean_token_accuracy": 0.1586763396859169,
"num_tokens": 10435636.0,
"step": 4565
},
{
"entropy": 5.848702239990234,
"epoch": 0.45175958877026495,
"grad_norm": 0.90625,
"learning_rate": 0.0004985913931785633,
"loss": 5.6128,
"mean_token_accuracy": 0.17407121509313583,
"num_tokens": 10447586.0,
"step": 4570
},
{
"entropy": 5.762767887115478,
"epoch": 0.4522538552787663,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985874477687636,
"loss": 5.6658,
"mean_token_accuracy": 0.16698419600725173,
"num_tokens": 10460678.0,
"step": 4575
},
{
"entropy": 5.831792402267456,
"epoch": 0.4527481217872677,
"grad_norm": 0.9765625,
"learning_rate": 0.0004985834968586779,
"loss": 5.6264,
"mean_token_accuracy": 0.1700241297483444,
"num_tokens": 10472738.0,
"step": 4580
},
{
"entropy": 5.835088634490967,
"epoch": 0.45324238829576907,
"grad_norm": 0.89453125,
"learning_rate": 0.0004985795404484037,
"loss": 5.5226,
"mean_token_accuracy": 0.17453843653202056,
"num_tokens": 10483716.0,
"step": 4585
},
{
"entropy": 5.850312614440918,
"epoch": 0.45373665480427045,
"grad_norm": 1.125,
"learning_rate": 0.0004985755785380379,
"loss": 5.6386,
"mean_token_accuracy": 0.16620276123285294,
"num_tokens": 10495287.0,
"step": 4590
},
{
"entropy": 5.794805860519409,
"epoch": 0.4542309213127719,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985716111276781,
"loss": 5.6008,
"mean_token_accuracy": 0.1721417114138603,
"num_tokens": 10506332.0,
"step": 4595
},
{
"entropy": 5.857477140426636,
"epoch": 0.45472518782127325,
"grad_norm": 0.92578125,
"learning_rate": 0.0004985676382174223,
"loss": 5.6368,
"mean_token_accuracy": 0.16736275106668472,
"num_tokens": 10517574.0,
"step": 4600
},
{
"entropy": 5.8186877250671385,
"epoch": 0.4552194543297746,
"grad_norm": 0.921875,
"learning_rate": 0.0004985636598073677,
"loss": 5.5636,
"mean_token_accuracy": 0.17843152284622193,
"num_tokens": 10528998.0,
"step": 4605
},
{
"entropy": 5.789574527740479,
"epoch": 0.455713720838276,
"grad_norm": 0.87109375,
"learning_rate": 0.0004985596758976125,
"loss": 5.6197,
"mean_token_accuracy": 0.15976811945438385,
"num_tokens": 10539669.0,
"step": 4610
},
{
"entropy": 5.796797466278076,
"epoch": 0.45620798734677737,
"grad_norm": 0.96875,
"learning_rate": 0.0004985556864882545,
"loss": 5.6042,
"mean_token_accuracy": 0.17045630514621735,
"num_tokens": 10550005.0,
"step": 4615
},
{
"entropy": 5.758793020248413,
"epoch": 0.45670225385527874,
"grad_norm": 0.98046875,
"learning_rate": 0.000498551691579392,
"loss": 5.5708,
"mean_token_accuracy": 0.1709605410695076,
"num_tokens": 10561122.0,
"step": 4620
},
{
"entropy": 5.835131645202637,
"epoch": 0.45719652036378017,
"grad_norm": 0.984375,
"learning_rate": 0.0004985476911711232,
"loss": 5.5515,
"mean_token_accuracy": 0.1706647038459778,
"num_tokens": 10572081.0,
"step": 4625
},
{
"entropy": 5.716209506988525,
"epoch": 0.45769078687228154,
"grad_norm": 1.078125,
"learning_rate": 0.0004985436852635465,
"loss": 5.5451,
"mean_token_accuracy": 0.18031772524118422,
"num_tokens": 10582297.0,
"step": 4630
},
{
"entropy": 5.866713285446167,
"epoch": 0.4581850533807829,
"grad_norm": 0.98046875,
"learning_rate": 0.0004985396738567604,
"loss": 5.6473,
"mean_token_accuracy": 0.16463931947946547,
"num_tokens": 10593161.0,
"step": 4635
},
{
"entropy": 5.830714082717895,
"epoch": 0.4586793198892843,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985356569508638,
"loss": 5.6491,
"mean_token_accuracy": 0.17743647396564483,
"num_tokens": 10603900.0,
"step": 4640
},
{
"entropy": 5.7970226287841795,
"epoch": 0.45917358639778566,
"grad_norm": 0.87890625,
"learning_rate": 0.0004985316345459551,
"loss": 5.6324,
"mean_token_accuracy": 0.1742269366979599,
"num_tokens": 10614946.0,
"step": 4645
},
{
"entropy": 5.833553075790405,
"epoch": 0.4596678529062871,
"grad_norm": 0.984375,
"learning_rate": 0.0004985276066421338,
"loss": 5.6298,
"mean_token_accuracy": 0.16616080701351166,
"num_tokens": 10625073.0,
"step": 4650
},
{
"entropy": 5.859928894042969,
"epoch": 0.46016211941478846,
"grad_norm": 0.9375,
"learning_rate": 0.0004985235732394986,
"loss": 5.7275,
"mean_token_accuracy": 0.16246279031038285,
"num_tokens": 10636942.0,
"step": 4655
},
{
"entropy": 5.88986988067627,
"epoch": 0.46065638592328984,
"grad_norm": 1.03125,
"learning_rate": 0.0004985195343381487,
"loss": 5.6639,
"mean_token_accuracy": 0.16346882283687592,
"num_tokens": 10647608.0,
"step": 4660
},
{
"entropy": 5.906115961074829,
"epoch": 0.4611506524317912,
"grad_norm": 0.88671875,
"learning_rate": 0.0004985154899381837,
"loss": 5.7196,
"mean_token_accuracy": 0.16399595737457276,
"num_tokens": 10659532.0,
"step": 4665
},
{
"entropy": 5.830447673797607,
"epoch": 0.4616449189402926,
"grad_norm": 0.921875,
"learning_rate": 0.0004985114400397029,
"loss": 5.6144,
"mean_token_accuracy": 0.17186745554208754,
"num_tokens": 10670342.0,
"step": 4670
},
{
"entropy": 5.846805810928345,
"epoch": 0.462139185448794,
"grad_norm": 0.94921875,
"learning_rate": 0.0004985073846428061,
"loss": 5.6684,
"mean_token_accuracy": 0.16922882795333863,
"num_tokens": 10682984.0,
"step": 4675
},
{
"entropy": 5.81272177696228,
"epoch": 0.4626334519572954,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985033237475929,
"loss": 5.6571,
"mean_token_accuracy": 0.16456034183502197,
"num_tokens": 10694235.0,
"step": 4680
},
{
"entropy": 5.869851016998291,
"epoch": 0.46312771846579676,
"grad_norm": 0.99609375,
"learning_rate": 0.0004984992573541633,
"loss": 5.6675,
"mean_token_accuracy": 0.16956137716770173,
"num_tokens": 10704508.0,
"step": 4685
},
{
"entropy": 5.717702198028564,
"epoch": 0.46362198497429813,
"grad_norm": 0.8828125,
"learning_rate": 0.0004984951854626173,
"loss": 5.5306,
"mean_token_accuracy": 0.17722870260477067,
"num_tokens": 10716927.0,
"step": 4690
},
{
"entropy": 5.770853900909424,
"epoch": 0.4641162514827995,
"grad_norm": 0.91015625,
"learning_rate": 0.0004984911080730551,
"loss": 5.5797,
"mean_token_accuracy": 0.17649805545806885,
"num_tokens": 10728396.0,
"step": 4695
},
{
"entropy": 5.8329957008361815,
"epoch": 0.46461051799130093,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984870251855769,
"loss": 5.6504,
"mean_token_accuracy": 0.17403971701860427,
"num_tokens": 10739001.0,
"step": 4700
},
{
"entropy": 5.846718072891235,
"epoch": 0.4651047844998023,
"grad_norm": 1.015625,
"learning_rate": 0.0004984829368002831,
"loss": 5.5917,
"mean_token_accuracy": 0.17142556011676788,
"num_tokens": 10748768.0,
"step": 4705
},
{
"entropy": 5.830345964431762,
"epoch": 0.4655990510083037,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984788429172745,
"loss": 5.6498,
"mean_token_accuracy": 0.1713384658098221,
"num_tokens": 10760582.0,
"step": 4710
},
{
"entropy": 5.788267183303833,
"epoch": 0.46609331751680505,
"grad_norm": 0.88671875,
"learning_rate": 0.0004984747435366517,
"loss": 5.5594,
"mean_token_accuracy": 0.1783460333943367,
"num_tokens": 10772537.0,
"step": 4715
},
{
"entropy": 5.8207924365997314,
"epoch": 0.4665875840253064,
"grad_norm": 0.87890625,
"learning_rate": 0.0004984706386585156,
"loss": 5.4813,
"mean_token_accuracy": 0.1830333173274994,
"num_tokens": 10783740.0,
"step": 4720
},
{
"entropy": 5.793240213394165,
"epoch": 0.46708185053380785,
"grad_norm": 0.8828125,
"learning_rate": 0.000498466528282967,
"loss": 5.6381,
"mean_token_accuracy": 0.17064726650714873,
"num_tokens": 10794793.0,
"step": 4725
},
{
"entropy": 5.866763877868652,
"epoch": 0.4675761170423092,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984624124101072,
"loss": 5.6207,
"mean_token_accuracy": 0.17204025834798814,
"num_tokens": 10805944.0,
"step": 4730
},
{
"entropy": 5.824394655227661,
"epoch": 0.4680703835508106,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984582910400374,
"loss": 5.6338,
"mean_token_accuracy": 0.16792816668748856,
"num_tokens": 10818362.0,
"step": 4735
},
{
"entropy": 5.7967041015625,
"epoch": 0.46856465005931197,
"grad_norm": 0.84765625,
"learning_rate": 0.000498454164172859,
"loss": 5.6194,
"mean_token_accuracy": 0.17585522532463074,
"num_tokens": 10831306.0,
"step": 4740
},
{
"entropy": 5.778382873535156,
"epoch": 0.46905891656781334,
"grad_norm": 0.953125,
"learning_rate": 0.0004984500318086733,
"loss": 5.5824,
"mean_token_accuracy": 0.170336352288723,
"num_tokens": 10841671.0,
"step": 4745
},
{
"entropy": 5.7844038009643555,
"epoch": 0.4695531830763148,
"grad_norm": 0.96875,
"learning_rate": 0.0004984458939475824,
"loss": 5.5449,
"mean_token_accuracy": 0.17546041458845138,
"num_tokens": 10852754.0,
"step": 4750
},
{
"entropy": 5.811524963378906,
"epoch": 0.47004744958481615,
"grad_norm": 0.9296875,
"learning_rate": 0.0004984417505896877,
"loss": 5.6214,
"mean_token_accuracy": 0.17046928852796556,
"num_tokens": 10864812.0,
"step": 4755
},
{
"entropy": 5.803496170043945,
"epoch": 0.4705417160933175,
"grad_norm": 0.94140625,
"learning_rate": 0.0004984376017350913,
"loss": 5.6295,
"mean_token_accuracy": 0.17111281156539918,
"num_tokens": 10875967.0,
"step": 4760
},
{
"entropy": 5.790186262130737,
"epoch": 0.4710359826018189,
"grad_norm": 0.8828125,
"learning_rate": 0.0004984334473838953,
"loss": 5.5832,
"mean_token_accuracy": 0.17373200058937072,
"num_tokens": 10889575.0,
"step": 4765
},
{
"entropy": 5.820612001419067,
"epoch": 0.47153024911032027,
"grad_norm": 1.0,
"learning_rate": 0.0004984292875362018,
"loss": 5.6894,
"mean_token_accuracy": 0.1711822271347046,
"num_tokens": 10900751.0,
"step": 4770
},
{
"entropy": 5.854453992843628,
"epoch": 0.4720245156188217,
"grad_norm": 0.84765625,
"learning_rate": 0.0004984251221921131,
"loss": 5.6226,
"mean_token_accuracy": 0.17048616111278533,
"num_tokens": 10912618.0,
"step": 4775
},
{
"entropy": 5.797359228134155,
"epoch": 0.47251878212732307,
"grad_norm": 1.0,
"learning_rate": 0.0004984209513517317,
"loss": 5.5921,
"mean_token_accuracy": 0.16904376298189164,
"num_tokens": 10925286.0,
"step": 4780
},
{
"entropy": 5.824411010742187,
"epoch": 0.47301304863582444,
"grad_norm": 0.94140625,
"learning_rate": 0.0004984167750151603,
"loss": 5.5834,
"mean_token_accuracy": 0.1733615979552269,
"num_tokens": 10936033.0,
"step": 4785
},
{
"entropy": 5.806764793395996,
"epoch": 0.4735073151443258,
"grad_norm": 0.95703125,
"learning_rate": 0.0004984125931825018,
"loss": 5.5338,
"mean_token_accuracy": 0.1778370752930641,
"num_tokens": 10947055.0,
"step": 4790
},
{
"entropy": 5.810790061950684,
"epoch": 0.4740015816528272,
"grad_norm": 1.0,
"learning_rate": 0.0004984084058538585,
"loss": 5.5821,
"mean_token_accuracy": 0.16869567036628724,
"num_tokens": 10957468.0,
"step": 4795
},
{
"entropy": 5.83496503829956,
"epoch": 0.4744958481613286,
"grad_norm": 0.859375,
"learning_rate": 0.0004984042130293339,
"loss": 5.6913,
"mean_token_accuracy": 0.17087534070014954,
"num_tokens": 10970273.0,
"step": 4800
},
{
"entropy": 5.864707565307617,
"epoch": 0.47499011466983,
"grad_norm": 0.88671875,
"learning_rate": 0.0004984000147090311,
"loss": 5.6628,
"mean_token_accuracy": 0.1655959367752075,
"num_tokens": 10981842.0,
"step": 4805
},
{
"entropy": 5.832684135437011,
"epoch": 0.47548438117833136,
"grad_norm": 0.93359375,
"learning_rate": 0.0004983958108930531,
"loss": 5.5685,
"mean_token_accuracy": 0.1743028372526169,
"num_tokens": 10993615.0,
"step": 4810
},
{
"entropy": 5.792374134063721,
"epoch": 0.47597864768683273,
"grad_norm": 0.921875,
"learning_rate": 0.0004983916015815036,
"loss": 5.6573,
"mean_token_accuracy": 0.16958209425210952,
"num_tokens": 11004451.0,
"step": 4815
},
{
"entropy": 5.859997797012329,
"epoch": 0.4764729141953341,
"grad_norm": 0.93359375,
"learning_rate": 0.000498387386774486,
"loss": 5.5995,
"mean_token_accuracy": 0.16668929606676103,
"num_tokens": 11014834.0,
"step": 4820
},
{
"entropy": 5.74824743270874,
"epoch": 0.4769671807038355,
"grad_norm": 0.87890625,
"learning_rate": 0.000498383166472104,
"loss": 5.5521,
"mean_token_accuracy": 0.17305494248867034,
"num_tokens": 11026116.0,
"step": 4825
},
{
"entropy": 5.7953392505645756,
"epoch": 0.4774614472123369,
"grad_norm": 0.97265625,
"learning_rate": 0.0004983789406744615,
"loss": 5.6088,
"mean_token_accuracy": 0.17153684049844742,
"num_tokens": 11038780.0,
"step": 4830
},
{
"entropy": 5.897364330291748,
"epoch": 0.4779557137208383,
"grad_norm": 0.96875,
"learning_rate": 0.0004983747093816625,
"loss": 5.6589,
"mean_token_accuracy": 0.17048557847738266,
"num_tokens": 11049824.0,
"step": 4835
},
{
"entropy": 5.761606454849243,
"epoch": 0.47844998022933966,
"grad_norm": 0.9375,
"learning_rate": 0.0004983704725938108,
"loss": 5.5351,
"mean_token_accuracy": 0.1786967322230339,
"num_tokens": 11060863.0,
"step": 4840
},
{
"entropy": 5.732560920715332,
"epoch": 0.47894424673784103,
"grad_norm": 0.89453125,
"learning_rate": 0.000498366230311011,
"loss": 5.5561,
"mean_token_accuracy": 0.17488910108804703,
"num_tokens": 11072533.0,
"step": 4845
},
{
"entropy": 5.747886800765992,
"epoch": 0.4794385132463424,
"grad_norm": 0.96484375,
"learning_rate": 0.0004983619825333671,
"loss": 5.4862,
"mean_token_accuracy": 0.17575133293867112,
"num_tokens": 11083627.0,
"step": 4850
},
{
"entropy": 5.860506629943847,
"epoch": 0.47993277975484383,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983577292609839,
"loss": 5.7019,
"mean_token_accuracy": 0.1633409082889557,
"num_tokens": 11096421.0,
"step": 4855
},
{
"entropy": 5.758524656295776,
"epoch": 0.4804270462633452,
"grad_norm": 0.91015625,
"learning_rate": 0.0004983534704939659,
"loss": 5.5438,
"mean_token_accuracy": 0.16786952167749405,
"num_tokens": 11109053.0,
"step": 4860
},
{
"entropy": 5.763889265060425,
"epoch": 0.4809213127718466,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983492062324179,
"loss": 5.481,
"mean_token_accuracy": 0.1796043336391449,
"num_tokens": 11120431.0,
"step": 4865
},
{
"entropy": 5.772010707855225,
"epoch": 0.48141557928034795,
"grad_norm": 0.92578125,
"learning_rate": 0.0004983449364764447,
"loss": 5.5848,
"mean_token_accuracy": 0.17402093708515168,
"num_tokens": 11132767.0,
"step": 4870
},
{
"entropy": 5.7384312629699705,
"epoch": 0.4819098457888493,
"grad_norm": 0.96875,
"learning_rate": 0.0004983406612261514,
"loss": 5.5766,
"mean_token_accuracy": 0.17281497567892073,
"num_tokens": 11144508.0,
"step": 4875
},
{
"entropy": 5.754504060745239,
"epoch": 0.48240411229735075,
"grad_norm": 0.8671875,
"learning_rate": 0.0004983363804816433,
"loss": 5.5509,
"mean_token_accuracy": 0.17433478832244872,
"num_tokens": 11156905.0,
"step": 4880
},
{
"entropy": 5.790156269073487,
"epoch": 0.4828983788058521,
"grad_norm": 0.82421875,
"learning_rate": 0.0004983320942430256,
"loss": 5.5513,
"mean_token_accuracy": 0.1798376262187958,
"num_tokens": 11169383.0,
"step": 4885
},
{
"entropy": 5.767265319824219,
"epoch": 0.4833926453143535,
"grad_norm": 0.9609375,
"learning_rate": 0.0004983278025104036,
"loss": 5.4902,
"mean_token_accuracy": 0.1851058930158615,
"num_tokens": 11179795.0,
"step": 4890
},
{
"entropy": 5.776008224487304,
"epoch": 0.48388691182285487,
"grad_norm": 0.9296875,
"learning_rate": 0.000498323505283883,
"loss": 5.6159,
"mean_token_accuracy": 0.166591776907444,
"num_tokens": 11190783.0,
"step": 4895
},
{
"entropy": 5.878120708465576,
"epoch": 0.48438117833135624,
"grad_norm": 0.875,
"learning_rate": 0.0004983192025635698,
"loss": 5.5672,
"mean_token_accuracy": 0.16896193027496337,
"num_tokens": 11203000.0,
"step": 4900
},
{
"entropy": 5.807617330551148,
"epoch": 0.48487544483985767,
"grad_norm": 0.984375,
"learning_rate": 0.0004983148943495693,
"loss": 5.5857,
"mean_token_accuracy": 0.16904626190662383,
"num_tokens": 11213721.0,
"step": 4905
},
{
"entropy": 5.807488822937012,
"epoch": 0.48536971134835905,
"grad_norm": 0.9609375,
"learning_rate": 0.0004983105806419878,
"loss": 5.6642,
"mean_token_accuracy": 0.1656417280435562,
"num_tokens": 11224490.0,
"step": 4910
},
{
"entropy": 5.816106748580933,
"epoch": 0.4858639778568604,
"grad_norm": 0.890625,
"learning_rate": 0.0004983062614409314,
"loss": 5.5457,
"mean_token_accuracy": 0.1777344286441803,
"num_tokens": 11235022.0,
"step": 4915
},
{
"entropy": 5.793050861358642,
"epoch": 0.4863582443653618,
"grad_norm": 0.86328125,
"learning_rate": 0.0004983019367465064,
"loss": 5.5869,
"mean_token_accuracy": 0.16975762099027633,
"num_tokens": 11247485.0,
"step": 4920
},
{
"entropy": 5.753307914733886,
"epoch": 0.48685251087386316,
"grad_norm": 1.0078125,
"learning_rate": 0.000498297606558819,
"loss": 5.5015,
"mean_token_accuracy": 0.18103411495685579,
"num_tokens": 11258488.0,
"step": 4925
},
{
"entropy": 5.752778434753418,
"epoch": 0.4873467773823646,
"grad_norm": 0.85546875,
"learning_rate": 0.0004982932708779759,
"loss": 5.5695,
"mean_token_accuracy": 0.17985471934080124,
"num_tokens": 11270786.0,
"step": 4930
},
{
"entropy": 5.886114311218262,
"epoch": 0.48784104389086597,
"grad_norm": 1.0234375,
"learning_rate": 0.0004982889297040836,
"loss": 5.5715,
"mean_token_accuracy": 0.1702333375811577,
"num_tokens": 11282434.0,
"step": 4935
},
{
"entropy": 5.779879140853882,
"epoch": 0.48833531039936734,
"grad_norm": 0.94921875,
"learning_rate": 0.0004982845830372491,
"loss": 5.5569,
"mean_token_accuracy": 0.17203909456729888,
"num_tokens": 11292406.0,
"step": 4940
},
{
"entropy": 5.893464851379394,
"epoch": 0.4888295769078687,
"grad_norm": 0.90625,
"learning_rate": 0.0004982802308775791,
"loss": 5.6919,
"mean_token_accuracy": 0.16397708654403687,
"num_tokens": 11304491.0,
"step": 4945
},
{
"entropy": 5.8376823425292965,
"epoch": 0.4893238434163701,
"grad_norm": 0.9609375,
"learning_rate": 0.000498275873225181,
"loss": 5.6223,
"mean_token_accuracy": 0.16995202153921127,
"num_tokens": 11317038.0,
"step": 4950
},
{
"entropy": 5.7879523754119875,
"epoch": 0.4898181099248715,
"grad_norm": 0.83984375,
"learning_rate": 0.0004982715100801616,
"loss": 5.6255,
"mean_token_accuracy": 0.17164080440998078,
"num_tokens": 11329354.0,
"step": 4955
},
{
"entropy": 5.823068714141845,
"epoch": 0.4903123764333729,
"grad_norm": 0.92578125,
"learning_rate": 0.0004982671414426285,
"loss": 5.6131,
"mean_token_accuracy": 0.17055580019950867,
"num_tokens": 11340508.0,
"step": 4960
},
{
"entropy": 5.75929479598999,
"epoch": 0.49080664294187426,
"grad_norm": 0.9296875,
"learning_rate": 0.0004982627673126889,
"loss": 5.5779,
"mean_token_accuracy": 0.17654407620429993,
"num_tokens": 11352608.0,
"step": 4965
},
{
"entropy": 5.774344491958618,
"epoch": 0.49130090945037563,
"grad_norm": 1.0390625,
"learning_rate": 0.0004982583876904508,
"loss": 5.5617,
"mean_token_accuracy": 0.174379463493824,
"num_tokens": 11364089.0,
"step": 4970
},
{
"entropy": 5.7411417961120605,
"epoch": 0.491795175958877,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982540025760217,
"loss": 5.5062,
"mean_token_accuracy": 0.17841892540454865,
"num_tokens": 11375521.0,
"step": 4975
},
{
"entropy": 5.75401611328125,
"epoch": 0.49228944246737844,
"grad_norm": 0.91796875,
"learning_rate": 0.0004982496119695094,
"loss": 5.608,
"mean_token_accuracy": 0.17747943848371506,
"num_tokens": 11387721.0,
"step": 4980
},
{
"entropy": 5.8155148983001705,
"epoch": 0.4927837089758798,
"grad_norm": 1.0859375,
"learning_rate": 0.000498245215871022,
"loss": 5.6533,
"mean_token_accuracy": 0.17238590121269226,
"num_tokens": 11398688.0,
"step": 4985
},
{
"entropy": 5.775302314758301,
"epoch": 0.4932779754843812,
"grad_norm": 0.91015625,
"learning_rate": 0.0004982408142806678,
"loss": 5.5392,
"mean_token_accuracy": 0.1715422198176384,
"num_tokens": 11410420.0,
"step": 4990
},
{
"entropy": 5.730592966079712,
"epoch": 0.49377224199288255,
"grad_norm": 0.94921875,
"learning_rate": 0.0004982364071985549,
"loss": 5.5171,
"mean_token_accuracy": 0.1779486283659935,
"num_tokens": 11420521.0,
"step": 4995
},
{
"entropy": 5.839930009841919,
"epoch": 0.4942665085013839,
"grad_norm": 0.90625,
"learning_rate": 0.0004982319946247917,
"loss": 5.6067,
"mean_token_accuracy": 0.16767350286245347,
"num_tokens": 11432550.0,
"step": 5000
},
{
"entropy": 5.845134305953979,
"epoch": 0.49476077500988536,
"grad_norm": 0.90234375,
"learning_rate": 0.0004982275765594869,
"loss": 5.5543,
"mean_token_accuracy": 0.16987244784832,
"num_tokens": 11444602.0,
"step": 5005
},
{
"entropy": 5.698256492614746,
"epoch": 0.49525504151838673,
"grad_norm": 0.9375,
"learning_rate": 0.0004982231530027489,
"loss": 5.4255,
"mean_token_accuracy": 0.1847676157951355,
"num_tokens": 11456187.0,
"step": 5010
},
{
"entropy": 5.73066554069519,
"epoch": 0.4957493080268881,
"grad_norm": 0.84375,
"learning_rate": 0.0004982187239546868,
"loss": 5.6288,
"mean_token_accuracy": 0.17178757935762407,
"num_tokens": 11468615.0,
"step": 5015
},
{
"entropy": 5.82112340927124,
"epoch": 0.4962435745353895,
"grad_norm": 0.96875,
"learning_rate": 0.0004982142894154095,
"loss": 5.5439,
"mean_token_accuracy": 0.17167694121599197,
"num_tokens": 11480865.0,
"step": 5020
},
{
"entropy": 5.835798645019532,
"epoch": 0.49673784104389085,
"grad_norm": 0.953125,
"learning_rate": 0.000498209849385026,
"loss": 5.6069,
"mean_token_accuracy": 0.16441622376441956,
"num_tokens": 11492590.0,
"step": 5025
},
{
"entropy": 5.828628826141357,
"epoch": 0.4972321075523923,
"grad_norm": 0.953125,
"learning_rate": 0.0004982054038636456,
"loss": 5.5994,
"mean_token_accuracy": 0.1767118379473686,
"num_tokens": 11503481.0,
"step": 5030
},
{
"entropy": 5.75640377998352,
"epoch": 0.49772637406089365,
"grad_norm": 0.8359375,
"learning_rate": 0.0004982009528513776,
"loss": 5.5205,
"mean_token_accuracy": 0.1742475911974907,
"num_tokens": 11515083.0,
"step": 5035
},
{
"entropy": 5.744735908508301,
"epoch": 0.498220640569395,
"grad_norm": 0.8671875,
"learning_rate": 0.0004981964963483316,
"loss": 5.5997,
"mean_token_accuracy": 0.16762128323316575,
"num_tokens": 11527923.0,
"step": 5040
},
{
"entropy": 5.806128787994385,
"epoch": 0.4987149070778964,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981920343546171,
"loss": 5.6242,
"mean_token_accuracy": 0.17232828587293625,
"num_tokens": 11539456.0,
"step": 5045
},
{
"entropy": 5.8579912185668945,
"epoch": 0.49920917358639777,
"grad_norm": 1.046875,
"learning_rate": 0.000498187566870344,
"loss": 5.5302,
"mean_token_accuracy": 0.17905319184064866,
"num_tokens": 11550809.0,
"step": 5050
},
{
"entropy": 5.844864320755005,
"epoch": 0.49970344009489914,
"grad_norm": 1.140625,
"learning_rate": 0.0004981830938956221,
"loss": 5.6347,
"mean_token_accuracy": 0.1694789245724678,
"num_tokens": 11562545.0,
"step": 5055
},
{
"entropy": 5.684816217422485,
"epoch": 0.5001977066034006,
"grad_norm": 0.90234375,
"learning_rate": 0.0004981786154305615,
"loss": 5.5496,
"mean_token_accuracy": 0.1772656634449959,
"num_tokens": 11574479.0,
"step": 5060
},
{
"entropy": 5.833620500564575,
"epoch": 0.500691973111902,
"grad_norm": 1.109375,
"learning_rate": 0.0004981741314752722,
"loss": 5.4864,
"mean_token_accuracy": 0.1810210108757019,
"num_tokens": 11583942.0,
"step": 5065
},
{
"entropy": 5.795402908325196,
"epoch": 0.5011862396204033,
"grad_norm": 0.89453125,
"learning_rate": 0.0004981696420298647,
"loss": 5.5861,
"mean_token_accuracy": 0.1765076547861099,
"num_tokens": 11596745.0,
"step": 5070
},
{
"entropy": 5.678934144973755,
"epoch": 0.5016805061289047,
"grad_norm": 0.86328125,
"learning_rate": 0.0004981651470944495,
"loss": 5.4664,
"mean_token_accuracy": 0.18235170245170593,
"num_tokens": 11608950.0,
"step": 5075
},
{
"entropy": 5.705907011032105,
"epoch": 0.5021747726374061,
"grad_norm": 0.88671875,
"learning_rate": 0.000498160646669137,
"loss": 5.6211,
"mean_token_accuracy": 0.1725287452340126,
"num_tokens": 11619996.0,
"step": 5080
},
{
"entropy": 5.7708536148071286,
"epoch": 0.5026690391459074,
"grad_norm": 0.85546875,
"learning_rate": 0.0004981561407540379,
"loss": 5.517,
"mean_token_accuracy": 0.179334656894207,
"num_tokens": 11632167.0,
"step": 5085
},
{
"entropy": 5.74412727355957,
"epoch": 0.5031633056544088,
"grad_norm": 0.890625,
"learning_rate": 0.0004981516293492632,
"loss": 5.5268,
"mean_token_accuracy": 0.17766835987567903,
"num_tokens": 11643722.0,
"step": 5090
},
{
"entropy": 5.741613960266113,
"epoch": 0.5036575721629103,
"grad_norm": 0.9140625,
"learning_rate": 0.0004981471124549237,
"loss": 5.5173,
"mean_token_accuracy": 0.18306994438171387,
"num_tokens": 11654890.0,
"step": 5095
},
{
"entropy": 5.780077838897705,
"epoch": 0.5041518386714117,
"grad_norm": 0.984375,
"learning_rate": 0.0004981425900711308,
"loss": 5.5707,
"mean_token_accuracy": 0.1708131328225136,
"num_tokens": 11665485.0,
"step": 5100
},
{
"entropy": 5.757856941223144,
"epoch": 0.504646105179913,
"grad_norm": 0.91015625,
"learning_rate": 0.0004981380621979954,
"loss": 5.4781,
"mean_token_accuracy": 0.1807607591152191,
"num_tokens": 11676145.0,
"step": 5105
},
{
"entropy": 5.746283102035522,
"epoch": 0.5051403716884144,
"grad_norm": 0.9375,
"learning_rate": 0.0004981335288356291,
"loss": 5.5084,
"mean_token_accuracy": 0.17166837453842163,
"num_tokens": 11687777.0,
"step": 5110
},
{
"entropy": 5.780774974822998,
"epoch": 0.5056346381969158,
"grad_norm": 0.921875,
"learning_rate": 0.0004981289899841434,
"loss": 5.5404,
"mean_token_accuracy": 0.17905764877796174,
"num_tokens": 11699703.0,
"step": 5115
},
{
"entropy": 5.799895524978638,
"epoch": 0.5061289047054172,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981244456436499,
"loss": 5.6415,
"mean_token_accuracy": 0.16529887765645981,
"num_tokens": 11711632.0,
"step": 5120
},
{
"entropy": 5.838253498077393,
"epoch": 0.5066231712139185,
"grad_norm": 0.8046875,
"learning_rate": 0.0004981198958142604,
"loss": 5.6343,
"mean_token_accuracy": 0.17375699877738954,
"num_tokens": 11723763.0,
"step": 5125
},
{
"entropy": 5.781112957000732,
"epoch": 0.5071174377224199,
"grad_norm": 0.81640625,
"learning_rate": 0.0004981153404960869,
"loss": 5.6195,
"mean_token_accuracy": 0.169736647605896,
"num_tokens": 11736061.0,
"step": 5130
},
{
"entropy": 5.7520177364349365,
"epoch": 0.5076117042309213,
"grad_norm": 0.8984375,
"learning_rate": 0.0004981107796892414,
"loss": 5.6394,
"mean_token_accuracy": 0.17034420669078826,
"num_tokens": 11749342.0,
"step": 5135
},
{
"entropy": 5.83673939704895,
"epoch": 0.5081059707394227,
"grad_norm": 0.87109375,
"learning_rate": 0.0004981062133938361,
"loss": 5.5484,
"mean_token_accuracy": 0.17624873220920562,
"num_tokens": 11760441.0,
"step": 5140
},
{
"entropy": 5.691654300689697,
"epoch": 0.508600237247924,
"grad_norm": 0.90234375,
"learning_rate": 0.0004981016416099834,
"loss": 5.4753,
"mean_token_accuracy": 0.1785828486084938,
"num_tokens": 11772191.0,
"step": 5145
},
{
"entropy": 5.839021492004394,
"epoch": 0.5090945037564255,
"grad_norm": 0.859375,
"learning_rate": 0.0004980970643377956,
"loss": 5.7409,
"mean_token_accuracy": 0.16388301253318788,
"num_tokens": 11786347.0,
"step": 5150
},
{
"entropy": 5.824480104446411,
"epoch": 0.5095887702649269,
"grad_norm": 0.8984375,
"learning_rate": 0.0004980924815773855,
"loss": 5.4852,
"mean_token_accuracy": 0.17841615229845048,
"num_tokens": 11797741.0,
"step": 5155
},
{
"entropy": 5.7572705268859865,
"epoch": 0.5100830367734283,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980878933288657,
"loss": 5.6131,
"mean_token_accuracy": 0.16382424384355546,
"num_tokens": 11809728.0,
"step": 5160
},
{
"entropy": 5.790679597854615,
"epoch": 0.5105773032819296,
"grad_norm": 0.88671875,
"learning_rate": 0.000498083299592349,
"loss": 5.5648,
"mean_token_accuracy": 0.16833973973989486,
"num_tokens": 11822000.0,
"step": 5165
},
{
"entropy": 5.793793964385986,
"epoch": 0.511071569790431,
"grad_norm": 0.9453125,
"learning_rate": 0.0004980787003679486,
"loss": 5.5415,
"mean_token_accuracy": 0.17131773084402085,
"num_tokens": 11834059.0,
"step": 5170
},
{
"entropy": 5.663825464248657,
"epoch": 0.5115658362989324,
"grad_norm": 0.984375,
"learning_rate": 0.0004980740956557776,
"loss": 5.4398,
"mean_token_accuracy": 0.18564523011446,
"num_tokens": 11844001.0,
"step": 5175
},
{
"entropy": 5.726988124847412,
"epoch": 0.5120601028074337,
"grad_norm": 0.84375,
"learning_rate": 0.0004980694854559493,
"loss": 5.613,
"mean_token_accuracy": 0.17274706214666366,
"num_tokens": 11855991.0,
"step": 5180
},
{
"entropy": 5.838094997406006,
"epoch": 0.5125543693159351,
"grad_norm": 0.89453125,
"learning_rate": 0.000498064869768577,
"loss": 5.5835,
"mean_token_accuracy": 0.18166578263044358,
"num_tokens": 11867975.0,
"step": 5185
},
{
"entropy": 5.738130235671997,
"epoch": 0.5130486358244365,
"grad_norm": 0.89453125,
"learning_rate": 0.0004980602485937743,
"loss": 5.5429,
"mean_token_accuracy": 0.17443735152482986,
"num_tokens": 11880034.0,
"step": 5190
},
{
"entropy": 5.822977781295776,
"epoch": 0.5135429023329379,
"grad_norm": 0.91796875,
"learning_rate": 0.0004980556219316549,
"loss": 5.6098,
"mean_token_accuracy": 0.17538013756275178,
"num_tokens": 11891490.0,
"step": 5195
},
{
"entropy": 5.8056519508361815,
"epoch": 0.5140371688414394,
"grad_norm": 0.9609375,
"learning_rate": 0.0004980509897823326,
"loss": 5.5061,
"mean_token_accuracy": 0.1805155858397484,
"num_tokens": 11902988.0,
"step": 5200
},
{
"entropy": 5.771095228195191,
"epoch": 0.5145314353499407,
"grad_norm": 0.890625,
"learning_rate": 0.0004980463521459212,
"loss": 5.5572,
"mean_token_accuracy": 0.17159489244222642,
"num_tokens": 11915088.0,
"step": 5205
},
{
"entropy": 5.744380903244019,
"epoch": 0.5150257018584421,
"grad_norm": 0.97265625,
"learning_rate": 0.0004980417090225351,
"loss": 5.5689,
"mean_token_accuracy": 0.1802521362900734,
"num_tokens": 11926789.0,
"step": 5210
},
{
"entropy": 5.84690670967102,
"epoch": 0.5155199683669435,
"grad_norm": 0.8671875,
"learning_rate": 0.0004980370604122883,
"loss": 5.6127,
"mean_token_accuracy": 0.17123763561248778,
"num_tokens": 11938876.0,
"step": 5215
},
{
"entropy": 5.80683331489563,
"epoch": 0.5160142348754448,
"grad_norm": 0.9609375,
"learning_rate": 0.0004980324063152952,
"loss": 5.5803,
"mean_token_accuracy": 0.17346588969230653,
"num_tokens": 11950099.0,
"step": 5220
},
{
"entropy": 5.814678430557251,
"epoch": 0.5165085013839462,
"grad_norm": 0.890625,
"learning_rate": 0.0004980277467316704,
"loss": 5.618,
"mean_token_accuracy": 0.1707424610853195,
"num_tokens": 11962585.0,
"step": 5225
},
{
"entropy": 5.7683158874511715,
"epoch": 0.5170027678924476,
"grad_norm": 0.984375,
"learning_rate": 0.0004980230816615282,
"loss": 5.5503,
"mean_token_accuracy": 0.17995088100433348,
"num_tokens": 11972579.0,
"step": 5230
},
{
"entropy": 5.750294351577759,
"epoch": 0.517497034400949,
"grad_norm": 0.9609375,
"learning_rate": 0.0004980184111049837,
"loss": 5.4854,
"mean_token_accuracy": 0.1822764351963997,
"num_tokens": 11983337.0,
"step": 5235
},
{
"entropy": 5.768231058120728,
"epoch": 0.5179913009094503,
"grad_norm": 0.8671875,
"learning_rate": 0.0004980137350621518,
"loss": 5.5966,
"mean_token_accuracy": 0.17121624797582627,
"num_tokens": 11995086.0,
"step": 5240
},
{
"entropy": 5.806021642684937,
"epoch": 0.5184855674179517,
"grad_norm": 0.8125,
"learning_rate": 0.0004980090535331473,
"loss": 5.6109,
"mean_token_accuracy": 0.17547132670879365,
"num_tokens": 12007814.0,
"step": 5245
},
{
"entropy": 5.683626890182495,
"epoch": 0.5189798339264532,
"grad_norm": 0.8203125,
"learning_rate": 0.0004980043665180857,
"loss": 5.4646,
"mean_token_accuracy": 0.1842152625322342,
"num_tokens": 12020714.0,
"step": 5250
},
{
"entropy": 5.758127641677857,
"epoch": 0.5194741004349546,
"grad_norm": 0.91796875,
"learning_rate": 0.0004979996740170819,
"loss": 5.4788,
"mean_token_accuracy": 0.17521584779024124,
"num_tokens": 12031994.0,
"step": 5255
},
{
"entropy": 5.808197927474976,
"epoch": 0.5199683669434559,
"grad_norm": 0.84765625,
"learning_rate": 0.0004979949760302515,
"loss": 5.5182,
"mean_token_accuracy": 0.17272645235061646,
"num_tokens": 12042887.0,
"step": 5260
},
{
"entropy": 5.782262372970581,
"epoch": 0.5204626334519573,
"grad_norm": 0.8671875,
"learning_rate": 0.0004979902725577101,
"loss": 5.5875,
"mean_token_accuracy": 0.177044178545475,
"num_tokens": 12054426.0,
"step": 5265
},
{
"entropy": 5.756087017059326,
"epoch": 0.5209568999604587,
"grad_norm": 0.921875,
"learning_rate": 0.0004979855635995735,
"loss": 5.5722,
"mean_token_accuracy": 0.17331088930368424,
"num_tokens": 12066360.0,
"step": 5270
},
{
"entropy": 5.85052261352539,
"epoch": 0.5214511664689601,
"grad_norm": 0.921875,
"learning_rate": 0.0004979808491559573,
"loss": 5.62,
"mean_token_accuracy": 0.16670458763837814,
"num_tokens": 12076651.0,
"step": 5275
},
{
"entropy": 5.771056127548218,
"epoch": 0.5219454329774614,
"grad_norm": 0.90234375,
"learning_rate": 0.0004979761292269778,
"loss": 5.4991,
"mean_token_accuracy": 0.17414287924766542,
"num_tokens": 12087522.0,
"step": 5280
},
{
"entropy": 5.761127662658692,
"epoch": 0.5224396994859628,
"grad_norm": 0.9140625,
"learning_rate": 0.0004979714038127508,
"loss": 5.5456,
"mean_token_accuracy": 0.17948328703641891,
"num_tokens": 12097999.0,
"step": 5285
},
{
"entropy": 5.718210601806641,
"epoch": 0.5229339659944642,
"grad_norm": 1.0625,
"learning_rate": 0.0004979666729133927,
"loss": 5.4987,
"mean_token_accuracy": 0.1711346134543419,
"num_tokens": 12108080.0,
"step": 5290
},
{
"entropy": 5.881346368789673,
"epoch": 0.5234282325029656,
"grad_norm": 0.84765625,
"learning_rate": 0.0004979619365290199,
"loss": 5.7301,
"mean_token_accuracy": 0.15981903672218323,
"num_tokens": 12122705.0,
"step": 5295
},
{
"entropy": 5.837782144546509,
"epoch": 0.523922499011467,
"grad_norm": 0.97265625,
"learning_rate": 0.0004979571946597488,
"loss": 5.5599,
"mean_token_accuracy": 0.17717295587062837,
"num_tokens": 12134000.0,
"step": 5300
},
{
"entropy": 5.725484991073609,
"epoch": 0.5244167655199684,
"grad_norm": 0.97265625,
"learning_rate": 0.0004979524473056961,
"loss": 5.565,
"mean_token_accuracy": 0.17184950709342955,
"num_tokens": 12144825.0,
"step": 5305
},
{
"entropy": 5.820409774780273,
"epoch": 0.5249110320284698,
"grad_norm": 1.0234375,
"learning_rate": 0.0004979476944669787,
"loss": 5.6598,
"mean_token_accuracy": 0.1701199322938919,
"num_tokens": 12156590.0,
"step": 5310
},
{
"entropy": 5.804708576202392,
"epoch": 0.5254052985369712,
"grad_norm": 0.953125,
"learning_rate": 0.0004979429361437134,
"loss": 5.4794,
"mean_token_accuracy": 0.18055262863636018,
"num_tokens": 12167478.0,
"step": 5315
},
{
"entropy": 5.747417736053467,
"epoch": 0.5258995650454725,
"grad_norm": 0.8984375,
"learning_rate": 0.0004979381723360173,
"loss": 5.4959,
"mean_token_accuracy": 0.17926698178052902,
"num_tokens": 12178728.0,
"step": 5320
},
{
"entropy": 5.686006307601929,
"epoch": 0.5263938315539739,
"grad_norm": 0.953125,
"learning_rate": 0.0004979334030440076,
"loss": 5.4654,
"mean_token_accuracy": 0.17048746943473816,
"num_tokens": 12190061.0,
"step": 5325
},
{
"entropy": 5.760270595550537,
"epoch": 0.5268880980624753,
"grad_norm": 0.96484375,
"learning_rate": 0.0004979286282678015,
"loss": 5.5806,
"mean_token_accuracy": 0.17671838849782945,
"num_tokens": 12200867.0,
"step": 5330
},
{
"entropy": 5.807965469360352,
"epoch": 0.5273823645709766,
"grad_norm": 0.796875,
"learning_rate": 0.0004979238480075166,
"loss": 5.5301,
"mean_token_accuracy": 0.17594483196735383,
"num_tokens": 12212681.0,
"step": 5335
},
{
"entropy": 5.717815256118774,
"epoch": 0.527876631079478,
"grad_norm": 0.97265625,
"learning_rate": 0.0004979190622632705,
"loss": 5.5786,
"mean_token_accuracy": 0.1718001186847687,
"num_tokens": 12224279.0,
"step": 5340
},
{
"entropy": 5.644581842422485,
"epoch": 0.5283708975879794,
"grad_norm": 0.8359375,
"learning_rate": 0.0004979142710351809,
"loss": 5.41,
"mean_token_accuracy": 0.19358813017606735,
"num_tokens": 12235754.0,
"step": 5345
},
{
"entropy": 5.827294206619262,
"epoch": 0.5288651640964809,
"grad_norm": 0.96484375,
"learning_rate": 0.0004979094743233656,
"loss": 5.5685,
"mean_token_accuracy": 0.17078588604927064,
"num_tokens": 12246014.0,
"step": 5350
},
{
"entropy": 5.72751259803772,
"epoch": 0.5293594306049823,
"grad_norm": 0.9765625,
"learning_rate": 0.0004979046721279426,
"loss": 5.4972,
"mean_token_accuracy": 0.1794961079955101,
"num_tokens": 12257519.0,
"step": 5355
},
{
"entropy": 5.747145128250122,
"epoch": 0.5298536971134836,
"grad_norm": 0.859375,
"learning_rate": 0.0004978998644490302,
"loss": 5.5663,
"mean_token_accuracy": 0.17280252277851105,
"num_tokens": 12269806.0,
"step": 5360
},
{
"entropy": 5.7658251285552975,
"epoch": 0.530347963621985,
"grad_norm": 0.89453125,
"learning_rate": 0.0004978950512867465,
"loss": 5.5617,
"mean_token_accuracy": 0.17104875147342682,
"num_tokens": 12281612.0,
"step": 5365
},
{
"entropy": 5.753850746154785,
"epoch": 0.5308422301304864,
"grad_norm": 1.015625,
"learning_rate": 0.0004978902326412099,
"loss": 5.5293,
"mean_token_accuracy": 0.17531534731388093,
"num_tokens": 12293022.0,
"step": 5370
},
{
"entropy": 5.773637151718139,
"epoch": 0.5313364966389877,
"grad_norm": 0.94921875,
"learning_rate": 0.000497885408512539,
"loss": 5.526,
"mean_token_accuracy": 0.17462725639343263,
"num_tokens": 12304414.0,
"step": 5375
},
{
"entropy": 5.884887838363648,
"epoch": 0.5318307631474891,
"grad_norm": 0.98046875,
"learning_rate": 0.0004978805789008526,
"loss": 5.6538,
"mean_token_accuracy": 0.16161142140626908,
"num_tokens": 12315618.0,
"step": 5380
},
{
"entropy": 5.784448099136353,
"epoch": 0.5323250296559905,
"grad_norm": 0.953125,
"learning_rate": 0.0004978757438062692,
"loss": 5.486,
"mean_token_accuracy": 0.18568605333566665,
"num_tokens": 12328026.0,
"step": 5385
},
{
"entropy": 5.714285135269165,
"epoch": 0.5328192961644919,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978709032289081,
"loss": 5.5588,
"mean_token_accuracy": 0.17499915808439254,
"num_tokens": 12339441.0,
"step": 5390
},
{
"entropy": 5.8693047046661375,
"epoch": 0.5333135626729932,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978660571688881,
"loss": 5.5846,
"mean_token_accuracy": 0.17316054701805114,
"num_tokens": 12351005.0,
"step": 5395
},
{
"entropy": 5.7560142993927,
"epoch": 0.5338078291814946,
"grad_norm": 0.9140625,
"learning_rate": 0.0004978612056263286,
"loss": 5.5878,
"mean_token_accuracy": 0.1801098331809044,
"num_tokens": 12363147.0,
"step": 5400
},
{
"entropy": 5.758735609054566,
"epoch": 0.5343020956899961,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978563486013488,
"loss": 5.5374,
"mean_token_accuracy": 0.17649147361516954,
"num_tokens": 12373446.0,
"step": 5405
},
{
"entropy": 5.721365737915039,
"epoch": 0.5347963621984975,
"grad_norm": 0.97265625,
"learning_rate": 0.0004978514860940683,
"loss": 5.504,
"mean_token_accuracy": 0.1789020672440529,
"num_tokens": 12384591.0,
"step": 5410
},
{
"entropy": 5.714781665802002,
"epoch": 0.5352906287069988,
"grad_norm": 0.9609375,
"learning_rate": 0.0004978466181046067,
"loss": 5.4599,
"mean_token_accuracy": 0.17836556285619737,
"num_tokens": 12396117.0,
"step": 5415
},
{
"entropy": 5.737146425247192,
"epoch": 0.5357848952155002,
"grad_norm": 0.9140625,
"learning_rate": 0.0004978417446330837,
"loss": 5.5468,
"mean_token_accuracy": 0.18124449849128724,
"num_tokens": 12407400.0,
"step": 5420
},
{
"entropy": 5.733011627197266,
"epoch": 0.5362791617240016,
"grad_norm": 0.94140625,
"learning_rate": 0.0004978368656796191,
"loss": 5.4497,
"mean_token_accuracy": 0.18468766957521437,
"num_tokens": 12417629.0,
"step": 5425
},
{
"entropy": 5.679784297943115,
"epoch": 0.536773428232503,
"grad_norm": 0.96484375,
"learning_rate": 0.0004978319812443331,
"loss": 5.4841,
"mean_token_accuracy": 0.18626576960086821,
"num_tokens": 12428522.0,
"step": 5430
},
{
"entropy": 5.778678750991821,
"epoch": 0.5372676947410043,
"grad_norm": 1.03125,
"learning_rate": 0.0004978270913273458,
"loss": 5.58,
"mean_token_accuracy": 0.1696370154619217,
"num_tokens": 12440390.0,
"step": 5435
},
{
"entropy": 5.722135162353515,
"epoch": 0.5377619612495057,
"grad_norm": 1.015625,
"learning_rate": 0.0004978221959287775,
"loss": 5.4891,
"mean_token_accuracy": 0.1793165847659111,
"num_tokens": 12451846.0,
"step": 5440
},
{
"entropy": 5.736186552047729,
"epoch": 0.5382562277580071,
"grad_norm": 0.99609375,
"learning_rate": 0.0004978172950487486,
"loss": 5.5114,
"mean_token_accuracy": 0.18181947022676467,
"num_tokens": 12462051.0,
"step": 5445
},
{
"entropy": 5.72767391204834,
"epoch": 0.5387504942665085,
"grad_norm": 0.93359375,
"learning_rate": 0.0004978123886873797,
"loss": 5.4468,
"mean_token_accuracy": 0.18565606623888015,
"num_tokens": 12472730.0,
"step": 5450
},
{
"entropy": 5.714382457733154,
"epoch": 0.5392447607750099,
"grad_norm": 0.9453125,
"learning_rate": 0.0004978074768447915,
"loss": 5.4673,
"mean_token_accuracy": 0.18636564016342164,
"num_tokens": 12482719.0,
"step": 5455
},
{
"entropy": 5.759675645828247,
"epoch": 0.5397390272835113,
"grad_norm": 0.921875,
"learning_rate": 0.0004978025595211048,
"loss": 5.4913,
"mean_token_accuracy": 0.178179569542408,
"num_tokens": 12492659.0,
"step": 5460
},
{
"entropy": 5.803636932373047,
"epoch": 0.5402332937920127,
"grad_norm": 0.88671875,
"learning_rate": 0.0004977976367164404,
"loss": 5.5553,
"mean_token_accuracy": 0.18235886394977568,
"num_tokens": 12504381.0,
"step": 5465
},
{
"entropy": 5.713931465148926,
"epoch": 0.540727560300514,
"grad_norm": 0.9296875,
"learning_rate": 0.0004977927084309198,
"loss": 5.3969,
"mean_token_accuracy": 0.19028392285108567,
"num_tokens": 12515041.0,
"step": 5470
},
{
"entropy": 5.778231859207153,
"epoch": 0.5412218268090154,
"grad_norm": 0.9375,
"learning_rate": 0.0004977877746646639,
"loss": 5.557,
"mean_token_accuracy": 0.17734709233045579,
"num_tokens": 12526313.0,
"step": 5475
},
{
"entropy": 5.78344693183899,
"epoch": 0.5417160933175168,
"grad_norm": 0.9609375,
"learning_rate": 0.0004977828354177941,
"loss": 5.6263,
"mean_token_accuracy": 0.16407859474420547,
"num_tokens": 12538268.0,
"step": 5480
},
{
"entropy": 5.80599570274353,
"epoch": 0.5422103598260182,
"grad_norm": 0.94140625,
"learning_rate": 0.0004977778906904321,
"loss": 5.5419,
"mean_token_accuracy": 0.17868427336215972,
"num_tokens": 12548365.0,
"step": 5485
},
{
"entropy": 5.8335600852966305,
"epoch": 0.5427046263345195,
"grad_norm": 0.859375,
"learning_rate": 0.0004977729404826995,
"loss": 5.6557,
"mean_token_accuracy": 0.16520809680223464,
"num_tokens": 12560652.0,
"step": 5490
},
{
"entropy": 5.7249761581420895,
"epoch": 0.5431988928430209,
"grad_norm": 0.94921875,
"learning_rate": 0.0004977679847947179,
"loss": 5.4995,
"mean_token_accuracy": 0.17777654379606248,
"num_tokens": 12572079.0,
"step": 5495
},
{
"entropy": 5.692812252044678,
"epoch": 0.5436931593515223,
"grad_norm": 0.89453125,
"learning_rate": 0.0004977630236266093,
"loss": 5.4277,
"mean_token_accuracy": 0.18401208221912385,
"num_tokens": 12582601.0,
"step": 5500
},
{
"entropy": 5.757671642303467,
"epoch": 0.5441874258600238,
"grad_norm": 0.921875,
"learning_rate": 0.0004977580569784958,
"loss": 5.5651,
"mean_token_accuracy": 0.17192913889884948,
"num_tokens": 12595176.0,
"step": 5505
},
{
"entropy": 5.774952411651611,
"epoch": 0.5446816923685251,
"grad_norm": 1.0,
"learning_rate": 0.0004977530848504994,
"loss": 5.5494,
"mean_token_accuracy": 0.17247545272111892,
"num_tokens": 12608161.0,
"step": 5510
},
{
"entropy": 5.733337450027466,
"epoch": 0.5451759588770265,
"grad_norm": 0.890625,
"learning_rate": 0.0004977481072427427,
"loss": 5.5682,
"mean_token_accuracy": 0.17189727872610092,
"num_tokens": 12620109.0,
"step": 5515
},
{
"entropy": 5.800567626953125,
"epoch": 0.5456702253855279,
"grad_norm": 0.9296875,
"learning_rate": 0.000497743124155348,
"loss": 5.5808,
"mean_token_accuracy": 0.17410012185573578,
"num_tokens": 12630339.0,
"step": 5520
},
{
"entropy": 5.743124771118164,
"epoch": 0.5461644918940293,
"grad_norm": 1.015625,
"learning_rate": 0.0004977381355884379,
"loss": 5.6153,
"mean_token_accuracy": 0.17426336109638213,
"num_tokens": 12642101.0,
"step": 5525
},
{
"entropy": 5.830995035171509,
"epoch": 0.5466587584025306,
"grad_norm": 0.8046875,
"learning_rate": 0.0004977331415421351,
"loss": 5.6604,
"mean_token_accuracy": 0.16365431249141693,
"num_tokens": 12655436.0,
"step": 5530
},
{
"entropy": 5.704841995239258,
"epoch": 0.547153024911032,
"grad_norm": 0.9609375,
"learning_rate": 0.0004977281420165624,
"loss": 5.3519,
"mean_token_accuracy": 0.18484551161527635,
"num_tokens": 12667100.0,
"step": 5535
},
{
"entropy": 5.728894424438477,
"epoch": 0.5476472914195334,
"grad_norm": 0.9375,
"learning_rate": 0.000497723137011843,
"loss": 5.6237,
"mean_token_accuracy": 0.1720960780978203,
"num_tokens": 12678788.0,
"step": 5540
},
{
"entropy": 5.749471235275268,
"epoch": 0.5481415579280348,
"grad_norm": 0.8359375,
"learning_rate": 0.0004977181265280997,
"loss": 5.5161,
"mean_token_accuracy": 0.18557162582874298,
"num_tokens": 12690980.0,
"step": 5545
},
{
"entropy": 5.806805658340454,
"epoch": 0.5486358244365361,
"grad_norm": 0.99609375,
"learning_rate": 0.000497713110565456,
"loss": 5.6351,
"mean_token_accuracy": 0.1691083773970604,
"num_tokens": 12702642.0,
"step": 5550
},
{
"entropy": 5.682715511322021,
"epoch": 0.5491300909450376,
"grad_norm": 0.9296875,
"learning_rate": 0.0004977080891240352,
"loss": 5.4202,
"mean_token_accuracy": 0.1857164591550827,
"num_tokens": 12712750.0,
"step": 5555
},
{
"entropy": 5.676763010025025,
"epoch": 0.549624357453539,
"grad_norm": 0.9375,
"learning_rate": 0.000497703062203961,
"loss": 5.4229,
"mean_token_accuracy": 0.1835971713066101,
"num_tokens": 12724464.0,
"step": 5560
},
{
"entropy": 5.707128715515137,
"epoch": 0.5501186239620404,
"grad_norm": 0.91796875,
"learning_rate": 0.0004976980298053567,
"loss": 5.5219,
"mean_token_accuracy": 0.17907305359840392,
"num_tokens": 12735419.0,
"step": 5565
},
{
"entropy": 5.786734914779663,
"epoch": 0.5506128904705417,
"grad_norm": 0.96875,
"learning_rate": 0.0004976929919283465,
"loss": 5.5137,
"mean_token_accuracy": 0.17892698645591737,
"num_tokens": 12746279.0,
"step": 5570
},
{
"entropy": 5.759409761428833,
"epoch": 0.5511071569790431,
"grad_norm": 0.8984375,
"learning_rate": 0.0004976879485730541,
"loss": 5.6097,
"mean_token_accuracy": 0.17511751651763915,
"num_tokens": 12758107.0,
"step": 5575
},
{
"entropy": 5.74146876335144,
"epoch": 0.5516014234875445,
"grad_norm": 0.96484375,
"learning_rate": 0.0004976828997396036,
"loss": 5.509,
"mean_token_accuracy": 0.18297296315431594,
"num_tokens": 12769354.0,
"step": 5580
},
{
"entropy": 5.758933639526367,
"epoch": 0.5520956899960459,
"grad_norm": 1.0078125,
"learning_rate": 0.0004976778454281193,
"loss": 5.4941,
"mean_token_accuracy": 0.17354206144809722,
"num_tokens": 12780926.0,
"step": 5585
},
{
"entropy": 5.787692737579346,
"epoch": 0.5525899565045472,
"grad_norm": 0.9375,
"learning_rate": 0.0004976727856387253,
"loss": 5.4781,
"mean_token_accuracy": 0.17749321311712266,
"num_tokens": 12792423.0,
"step": 5590
},
{
"entropy": 5.695298957824707,
"epoch": 0.5530842230130486,
"grad_norm": 0.84375,
"learning_rate": 0.0004976677203715464,
"loss": 5.4262,
"mean_token_accuracy": 0.1868865191936493,
"num_tokens": 12803688.0,
"step": 5595
},
{
"entropy": 5.6502316951751705,
"epoch": 0.55357848952155,
"grad_norm": 1.0234375,
"learning_rate": 0.000497662649626707,
"loss": 5.4244,
"mean_token_accuracy": 0.18135617524385453,
"num_tokens": 12814617.0,
"step": 5600
},
{
"entropy": 5.69397349357605,
"epoch": 0.5540727560300513,
"grad_norm": 0.8671875,
"learning_rate": 0.0004976575734043318,
"loss": 5.4429,
"mean_token_accuracy": 0.17544914335012435,
"num_tokens": 12826027.0,
"step": 5605
},
{
"entropy": 5.719434452056885,
"epoch": 0.5545670225385528,
"grad_norm": 0.90625,
"learning_rate": 0.0004976524917045457,
"loss": 5.5307,
"mean_token_accuracy": 0.1705397292971611,
"num_tokens": 12837454.0,
"step": 5610
},
{
"entropy": 5.689467716217041,
"epoch": 0.5550612890470542,
"grad_norm": 0.9921875,
"learning_rate": 0.000497647404527474,
"loss": 5.4336,
"mean_token_accuracy": 0.17773413360118867,
"num_tokens": 12848550.0,
"step": 5615
},
{
"entropy": 5.722694396972656,
"epoch": 0.5555555555555556,
"grad_norm": 0.94140625,
"learning_rate": 0.0004976423118732413,
"loss": 5.4954,
"mean_token_accuracy": 0.1816477209329605,
"num_tokens": 12858270.0,
"step": 5620
},
{
"entropy": 5.654467916488647,
"epoch": 0.556049822064057,
"grad_norm": 0.9375,
"learning_rate": 0.0004976372137419733,
"loss": 5.4541,
"mean_token_accuracy": 0.1788386359810829,
"num_tokens": 12869507.0,
"step": 5625
},
{
"entropy": 5.737163782119751,
"epoch": 0.5565440885725583,
"grad_norm": 0.9140625,
"learning_rate": 0.0004976321101337954,
"loss": 5.4368,
"mean_token_accuracy": 0.18255448937416077,
"num_tokens": 12880097.0,
"step": 5630
},
{
"entropy": 5.667933368682862,
"epoch": 0.5570383550810597,
"grad_norm": 0.875,
"learning_rate": 0.000497627001048833,
"loss": 5.4304,
"mean_token_accuracy": 0.1818331375718117,
"num_tokens": 12891302.0,
"step": 5635
},
{
"entropy": 5.656579065322876,
"epoch": 0.5575326215895611,
"grad_norm": 0.84765625,
"learning_rate": 0.0004976218864872118,
"loss": 5.4346,
"mean_token_accuracy": 0.1905026078224182,
"num_tokens": 12903950.0,
"step": 5640
},
{
"entropy": 5.692523384094239,
"epoch": 0.5580268880980624,
"grad_norm": 0.9296875,
"learning_rate": 0.0004976167664490576,
"loss": 5.4797,
"mean_token_accuracy": 0.17595995217561722,
"num_tokens": 12915713.0,
"step": 5645
},
{
"entropy": 5.690003156661987,
"epoch": 0.5585211546065638,
"grad_norm": 0.953125,
"learning_rate": 0.0004976116409344966,
"loss": 5.4424,
"mean_token_accuracy": 0.1829339236021042,
"num_tokens": 12926655.0,
"step": 5650
},
{
"entropy": 5.695260763168335,
"epoch": 0.5590154211150652,
"grad_norm": 0.83984375,
"learning_rate": 0.0004976065099436546,
"loss": 5.4364,
"mean_token_accuracy": 0.18093121349811553,
"num_tokens": 12938144.0,
"step": 5655
},
{
"entropy": 5.7835081100463865,
"epoch": 0.5595096876235667,
"grad_norm": 0.89453125,
"learning_rate": 0.0004976013734766579,
"loss": 5.6066,
"mean_token_accuracy": 0.17341161221265794,
"num_tokens": 12950486.0,
"step": 5660
},
{
"entropy": 5.676397037506104,
"epoch": 0.560003954132068,
"grad_norm": 0.9296875,
"learning_rate": 0.000497596231533633,
"loss": 5.3482,
"mean_token_accuracy": 0.19318064153194428,
"num_tokens": 12960966.0,
"step": 5665
},
{
"entropy": 5.62633786201477,
"epoch": 0.5604982206405694,
"grad_norm": 0.94140625,
"learning_rate": 0.000497591084114706,
"loss": 5.47,
"mean_token_accuracy": 0.17677472084760665,
"num_tokens": 12971960.0,
"step": 5670
},
{
"entropy": 5.742179298400879,
"epoch": 0.5609924871490708,
"grad_norm": 0.93359375,
"learning_rate": 0.000497585931220004,
"loss": 5.5464,
"mean_token_accuracy": 0.17461706101894378,
"num_tokens": 12984186.0,
"step": 5675
},
{
"entropy": 5.698014545440674,
"epoch": 0.5614867536575722,
"grad_norm": 0.90234375,
"learning_rate": 0.0004975807728496535,
"loss": 5.3807,
"mean_token_accuracy": 0.19021368771791458,
"num_tokens": 12994205.0,
"step": 5680
},
{
"entropy": 5.76334228515625,
"epoch": 0.5619810201660735,
"grad_norm": 0.921875,
"learning_rate": 0.0004975756090037815,
"loss": 5.5238,
"mean_token_accuracy": 0.17858278900384902,
"num_tokens": 13006143.0,
"step": 5685
},
{
"entropy": 5.787828063964843,
"epoch": 0.5624752866745749,
"grad_norm": 0.921875,
"learning_rate": 0.000497570439682515,
"loss": 5.5582,
"mean_token_accuracy": 0.1784708470106125,
"num_tokens": 13017849.0,
"step": 5690
},
{
"entropy": 5.556822872161865,
"epoch": 0.5629695531830763,
"grad_norm": 0.9296875,
"learning_rate": 0.000497565264885981,
"loss": 5.4408,
"mean_token_accuracy": 0.18541548997163773,
"num_tokens": 13030301.0,
"step": 5695
},
{
"entropy": 5.736711740493774,
"epoch": 0.5634638196915777,
"grad_norm": 0.9921875,
"learning_rate": 0.0004975600846143072,
"loss": 5.4915,
"mean_token_accuracy": 0.18181895613670349,
"num_tokens": 13040229.0,
"step": 5700
},
{
"entropy": 5.723695230484009,
"epoch": 0.563958086200079,
"grad_norm": 0.85546875,
"learning_rate": 0.0004975548988676206,
"loss": 5.4309,
"mean_token_accuracy": 0.184915491938591,
"num_tokens": 13050752.0,
"step": 5705
},
{
"entropy": 5.724451875686645,
"epoch": 0.5644523527085805,
"grad_norm": 0.89453125,
"learning_rate": 0.0004975497076460491,
"loss": 5.6137,
"mean_token_accuracy": 0.16811935752630233,
"num_tokens": 13062003.0,
"step": 5710
},
{
"entropy": 5.731402015686035,
"epoch": 0.5649466192170819,
"grad_norm": 0.9609375,
"learning_rate": 0.0004975445109497202,
"loss": 5.5375,
"mean_token_accuracy": 0.16912083178758622,
"num_tokens": 13073746.0,
"step": 5715
},
{
"entropy": 5.8177405834198,
"epoch": 0.5654408857255833,
"grad_norm": 1.0625,
"learning_rate": 0.0004975393087787617,
"loss": 5.6268,
"mean_token_accuracy": 0.16762364655733109,
"num_tokens": 13086068.0,
"step": 5720
},
{
"entropy": 5.747670316696167,
"epoch": 0.5659351522340846,
"grad_norm": 0.87109375,
"learning_rate": 0.0004975341011333019,
"loss": 5.5234,
"mean_token_accuracy": 0.17420125305652617,
"num_tokens": 13097078.0,
"step": 5725
},
{
"entropy": 5.71416392326355,
"epoch": 0.566429418742586,
"grad_norm": 0.984375,
"learning_rate": 0.0004975288880134685,
"loss": 5.4861,
"mean_token_accuracy": 0.17938829362392425,
"num_tokens": 13108450.0,
"step": 5730
},
{
"entropy": 5.715784931182862,
"epoch": 0.5669236852510874,
"grad_norm": 0.9609375,
"learning_rate": 0.0004975236694193901,
"loss": 5.5348,
"mean_token_accuracy": 0.17660179734230042,
"num_tokens": 13120216.0,
"step": 5735
},
{
"entropy": 5.715113353729248,
"epoch": 0.5674179517595888,
"grad_norm": 0.8671875,
"learning_rate": 0.0004975184453511948,
"loss": 5.5093,
"mean_token_accuracy": 0.17144474685192107,
"num_tokens": 13131214.0,
"step": 5740
},
{
"entropy": 5.777291917800904,
"epoch": 0.5679122182680901,
"grad_norm": 0.95703125,
"learning_rate": 0.0004975132158090112,
"loss": 5.5711,
"mean_token_accuracy": 0.1721394255757332,
"num_tokens": 13143792.0,
"step": 5745
},
{
"entropy": 5.686273527145386,
"epoch": 0.5684064847765915,
"grad_norm": 0.98828125,
"learning_rate": 0.000497507980792968,
"loss": 5.5344,
"mean_token_accuracy": 0.17619652599096297,
"num_tokens": 13156022.0,
"step": 5750
},
{
"entropy": 5.732636404037476,
"epoch": 0.5689007512850929,
"grad_norm": 0.89453125,
"learning_rate": 0.000497502740303194,
"loss": 5.4812,
"mean_token_accuracy": 0.1848985359072685,
"num_tokens": 13167374.0,
"step": 5755
},
{
"entropy": 5.702525520324707,
"epoch": 0.5693950177935944,
"grad_norm": 0.88671875,
"learning_rate": 0.000497497494339818,
"loss": 5.4395,
"mean_token_accuracy": 0.18368844240903853,
"num_tokens": 13177662.0,
"step": 5760
},
{
"entropy": 5.71856722831726,
"epoch": 0.5698892843020957,
"grad_norm": 0.86328125,
"learning_rate": 0.0004974922429029692,
"loss": 5.5105,
"mean_token_accuracy": 0.1769362986087799,
"num_tokens": 13188375.0,
"step": 5765
},
{
"entropy": 5.725184345245362,
"epoch": 0.5703835508105971,
"grad_norm": 0.984375,
"learning_rate": 0.0004974869859927767,
"loss": 5.4092,
"mean_token_accuracy": 0.1829233556985855,
"num_tokens": 13198988.0,
"step": 5770
},
{
"entropy": 5.75066499710083,
"epoch": 0.5708778173190985,
"grad_norm": 0.96484375,
"learning_rate": 0.0004974817236093697,
"loss": 5.475,
"mean_token_accuracy": 0.1693466141819954,
"num_tokens": 13210699.0,
"step": 5775
},
{
"entropy": 5.63893141746521,
"epoch": 0.5713720838275999,
"grad_norm": 0.88671875,
"learning_rate": 0.0004974764557528779,
"loss": 5.4139,
"mean_token_accuracy": 0.18506574779748916,
"num_tokens": 13221925.0,
"step": 5780
},
{
"entropy": 5.719260883331299,
"epoch": 0.5718663503361012,
"grad_norm": 0.96875,
"learning_rate": 0.0004974711824234308,
"loss": 5.5393,
"mean_token_accuracy": 0.17589188367128372,
"num_tokens": 13233278.0,
"step": 5785
},
{
"entropy": 5.768571138381958,
"epoch": 0.5723606168446026,
"grad_norm": 0.89453125,
"learning_rate": 0.000497465903621158,
"loss": 5.4851,
"mean_token_accuracy": 0.17983630001544954,
"num_tokens": 13245423.0,
"step": 5790
},
{
"entropy": 5.738625240325928,
"epoch": 0.572854883353104,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974606193461895,
"loss": 5.5418,
"mean_token_accuracy": 0.18063973784446716,
"num_tokens": 13256312.0,
"step": 5795
},
{
"entropy": 5.735036182403564,
"epoch": 0.5733491498616053,
"grad_norm": 0.9296875,
"learning_rate": 0.0004974553295986553,
"loss": 5.5002,
"mean_token_accuracy": 0.17640191316604614,
"num_tokens": 13268158.0,
"step": 5800
},
{
"entropy": 5.740165710449219,
"epoch": 0.5738434163701067,
"grad_norm": 0.90234375,
"learning_rate": 0.0004974500343786854,
"loss": 5.4731,
"mean_token_accuracy": 0.18072212785482406,
"num_tokens": 13279625.0,
"step": 5805
},
{
"entropy": 5.749358749389648,
"epoch": 0.5743376828786081,
"grad_norm": 0.9296875,
"learning_rate": 0.0004974447336864102,
"loss": 5.4985,
"mean_token_accuracy": 0.17794628590345382,
"num_tokens": 13291967.0,
"step": 5810
},
{
"entropy": 5.602593231201172,
"epoch": 0.5748319493871096,
"grad_norm": 0.890625,
"learning_rate": 0.00049743942752196,
"loss": 5.3886,
"mean_token_accuracy": 0.19058725386857986,
"num_tokens": 13303641.0,
"step": 5815
},
{
"entropy": 5.683610582351685,
"epoch": 0.575326215895611,
"grad_norm": 0.94140625,
"learning_rate": 0.0004974341158854654,
"loss": 5.4409,
"mean_token_accuracy": 0.1757155492901802,
"num_tokens": 13314754.0,
"step": 5820
},
{
"entropy": 5.693660020828247,
"epoch": 0.5758204824041123,
"grad_norm": 0.9609375,
"learning_rate": 0.0004974287987770571,
"loss": 5.5574,
"mean_token_accuracy": 0.17383753061294555,
"num_tokens": 13325574.0,
"step": 5825
},
{
"entropy": 5.7115068435668945,
"epoch": 0.5763147489126137,
"grad_norm": 0.9609375,
"learning_rate": 0.0004974234761968658,
"loss": 5.454,
"mean_token_accuracy": 0.18795131742954255,
"num_tokens": 13335946.0,
"step": 5830
},
{
"entropy": 5.728109359741211,
"epoch": 0.5768090154211151,
"grad_norm": 0.86328125,
"learning_rate": 0.0004974181481450224,
"loss": 5.4645,
"mean_token_accuracy": 0.18762592375278472,
"num_tokens": 13347812.0,
"step": 5835
},
{
"entropy": 5.65715970993042,
"epoch": 0.5773032819296164,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974128146216581,
"loss": 5.4205,
"mean_token_accuracy": 0.179990316927433,
"num_tokens": 13358492.0,
"step": 5840
},
{
"entropy": 5.792056941986084,
"epoch": 0.5777975484381178,
"grad_norm": 0.8671875,
"learning_rate": 0.0004974074756269042,
"loss": 5.5754,
"mean_token_accuracy": 0.17083921730518342,
"num_tokens": 13370465.0,
"step": 5845
},
{
"entropy": 5.757367944717407,
"epoch": 0.5782918149466192,
"grad_norm": 0.85546875,
"learning_rate": 0.0004974021311608917,
"loss": 5.4521,
"mean_token_accuracy": 0.18890322297811507,
"num_tokens": 13382328.0,
"step": 5850
},
{
"entropy": 5.696540975570679,
"epoch": 0.5787860814551206,
"grad_norm": 0.92578125,
"learning_rate": 0.0004973967812237523,
"loss": 5.5435,
"mean_token_accuracy": 0.17575131952762604,
"num_tokens": 13393439.0,
"step": 5855
},
{
"entropy": 5.700885200500489,
"epoch": 0.5792803479636219,
"grad_norm": 0.9296875,
"learning_rate": 0.0004973914258156177,
"loss": 5.4936,
"mean_token_accuracy": 0.17841797769069673,
"num_tokens": 13404778.0,
"step": 5860
},
{
"entropy": 5.7527289390563965,
"epoch": 0.5797746144721234,
"grad_norm": 0.89453125,
"learning_rate": 0.0004973860649366195,
"loss": 5.4728,
"mean_token_accuracy": 0.18216743618249892,
"num_tokens": 13417031.0,
"step": 5865
},
{
"entropy": 5.719571590423584,
"epoch": 0.5802688809806248,
"grad_norm": 0.97265625,
"learning_rate": 0.0004973806985868895,
"loss": 5.4719,
"mean_token_accuracy": 0.18248441964387893,
"num_tokens": 13428030.0,
"step": 5870
},
{
"entropy": 5.7111420154571535,
"epoch": 0.5807631474891262,
"grad_norm": 0.98828125,
"learning_rate": 0.0004973753267665599,
"loss": 5.4766,
"mean_token_accuracy": 0.18524419367313386,
"num_tokens": 13438934.0,
"step": 5875
},
{
"entropy": 5.714941215515137,
"epoch": 0.5812574139976275,
"grad_norm": 0.95703125,
"learning_rate": 0.0004973699494757627,
"loss": 5.4494,
"mean_token_accuracy": 0.17472849637269974,
"num_tokens": 13450542.0,
"step": 5880
},
{
"entropy": 5.673115491867065,
"epoch": 0.5817516805061289,
"grad_norm": 0.90234375,
"learning_rate": 0.0004973645667146302,
"loss": 5.4959,
"mean_token_accuracy": 0.18238535523414612,
"num_tokens": 13462382.0,
"step": 5885
},
{
"entropy": 5.741902160644531,
"epoch": 0.5822459470146303,
"grad_norm": 0.92578125,
"learning_rate": 0.0004973591784832949,
"loss": 5.5083,
"mean_token_accuracy": 0.17877876311540603,
"num_tokens": 13473852.0,
"step": 5890
},
{
"entropy": 5.752352094650268,
"epoch": 0.5827402135231317,
"grad_norm": 0.87890625,
"learning_rate": 0.0004973537847818894,
"loss": 5.5224,
"mean_token_accuracy": 0.176455520093441,
"num_tokens": 13485763.0,
"step": 5895
},
{
"entropy": 5.6837280750274655,
"epoch": 0.583234480031633,
"grad_norm": 0.8359375,
"learning_rate": 0.0004973483856105462,
"loss": 5.4936,
"mean_token_accuracy": 0.17691107243299484,
"num_tokens": 13498154.0,
"step": 5900
},
{
"entropy": 5.812195873260498,
"epoch": 0.5837287465401344,
"grad_norm": 0.84765625,
"learning_rate": 0.0004973429809693983,
"loss": 5.523,
"mean_token_accuracy": 0.17449950724840163,
"num_tokens": 13510563.0,
"step": 5905
},
{
"entropy": 5.851145839691162,
"epoch": 0.5842230130486358,
"grad_norm": 0.96484375,
"learning_rate": 0.0004973375708585785,
"loss": 5.5482,
"mean_token_accuracy": 0.1754000350832939,
"num_tokens": 13521987.0,
"step": 5910
},
{
"entropy": 5.73386697769165,
"epoch": 0.5847172795571373,
"grad_norm": 0.94921875,
"learning_rate": 0.0004973321552782201,
"loss": 5.5107,
"mean_token_accuracy": 0.17520942091941832,
"num_tokens": 13534075.0,
"step": 5915
},
{
"entropy": 5.754039716720581,
"epoch": 0.5852115460656386,
"grad_norm": 0.8984375,
"learning_rate": 0.000497326734228456,
"loss": 5.6506,
"mean_token_accuracy": 0.16962009221315383,
"num_tokens": 13546201.0,
"step": 5920
},
{
"entropy": 5.766037034988403,
"epoch": 0.58570581257414,
"grad_norm": 0.88671875,
"learning_rate": 0.0004973213077094199,
"loss": 5.4652,
"mean_token_accuracy": 0.18465032577514648,
"num_tokens": 13556985.0,
"step": 5925
},
{
"entropy": 5.782201862335205,
"epoch": 0.5862000790826414,
"grad_norm": 0.9140625,
"learning_rate": 0.0004973158757212451,
"loss": 5.5107,
"mean_token_accuracy": 0.18101984709501268,
"num_tokens": 13568617.0,
"step": 5930
},
{
"entropy": 5.699190378189087,
"epoch": 0.5866943455911428,
"grad_norm": 0.90625,
"learning_rate": 0.0004973104382640652,
"loss": 5.4462,
"mean_token_accuracy": 0.18187535405158997,
"num_tokens": 13581326.0,
"step": 5935
},
{
"entropy": 5.667532730102539,
"epoch": 0.5871886120996441,
"grad_norm": 0.91796875,
"learning_rate": 0.0004973049953380141,
"loss": 5.4649,
"mean_token_accuracy": 0.17698167711496354,
"num_tokens": 13591776.0,
"step": 5940
},
{
"entropy": 5.793080759048462,
"epoch": 0.5876828786081455,
"grad_norm": 0.96875,
"learning_rate": 0.0004972995469432256,
"loss": 5.5935,
"mean_token_accuracy": 0.17576271593570708,
"num_tokens": 13603475.0,
"step": 5945
},
{
"entropy": 5.731459617614746,
"epoch": 0.5881771451166469,
"grad_norm": 0.8671875,
"learning_rate": 0.0004972940930798338,
"loss": 5.5109,
"mean_token_accuracy": 0.17928755432367324,
"num_tokens": 13615348.0,
"step": 5950
},
{
"entropy": 5.633429050445557,
"epoch": 0.5886714116251482,
"grad_norm": 0.96875,
"learning_rate": 0.0004972886337479727,
"loss": 5.3704,
"mean_token_accuracy": 0.18748244941234588,
"num_tokens": 13625982.0,
"step": 5955
},
{
"entropy": 5.74733533859253,
"epoch": 0.5891656781336496,
"grad_norm": 0.8671875,
"learning_rate": 0.0004972831689477768,
"loss": 5.5126,
"mean_token_accuracy": 0.17600917518138887,
"num_tokens": 13637203.0,
"step": 5960
},
{
"entropy": 5.699123191833496,
"epoch": 0.5896599446421511,
"grad_norm": 0.87109375,
"learning_rate": 0.0004972776986793806,
"loss": 5.4367,
"mean_token_accuracy": 0.1801018312573433,
"num_tokens": 13648880.0,
"step": 5965
},
{
"entropy": 5.731894826889038,
"epoch": 0.5901542111506525,
"grad_norm": 0.91796875,
"learning_rate": 0.0004972722229429184,
"loss": 5.5227,
"mean_token_accuracy": 0.1764528974890709,
"num_tokens": 13660266.0,
"step": 5970
},
{
"entropy": 5.8591515064239506,
"epoch": 0.5906484776591538,
"grad_norm": 0.91796875,
"learning_rate": 0.000497266741738525,
"loss": 5.5845,
"mean_token_accuracy": 0.1685237005352974,
"num_tokens": 13672063.0,
"step": 5975
},
{
"entropy": 5.797526502609253,
"epoch": 0.5911427441676552,
"grad_norm": 0.9921875,
"learning_rate": 0.0004972612550663353,
"loss": 5.4851,
"mean_token_accuracy": 0.18678943663835526,
"num_tokens": 13683379.0,
"step": 5980
},
{
"entropy": 5.670707130432129,
"epoch": 0.5916370106761566,
"grad_norm": 0.859375,
"learning_rate": 0.0004972557629264842,
"loss": 5.4184,
"mean_token_accuracy": 0.1874980717897415,
"num_tokens": 13694512.0,
"step": 5985
},
{
"entropy": 5.649473810195923,
"epoch": 0.592131277184658,
"grad_norm": 0.93359375,
"learning_rate": 0.0004972502653191069,
"loss": 5.4619,
"mean_token_accuracy": 0.1757025897502899,
"num_tokens": 13705642.0,
"step": 5990
},
{
"entropy": 5.607302522659301,
"epoch": 0.5926255436931593,
"grad_norm": 0.9375,
"learning_rate": 0.0004972447622443387,
"loss": 5.3474,
"mean_token_accuracy": 0.19193902164697646,
"num_tokens": 13717367.0,
"step": 5995
},
{
"entropy": 5.68580436706543,
"epoch": 0.5931198102016607,
"grad_norm": 1.015625,
"learning_rate": 0.0004972392537023147,
"loss": 5.4572,
"mean_token_accuracy": 0.18161719143390656,
"num_tokens": 13728577.0,
"step": 6000
},
{
"epoch": 0.5931198102016607,
"eval_entropy": 5.543885540229075,
"eval_loss": 5.5077314376831055,
"eval_mean_token_accuracy": 0.18510187076356577,
"eval_num_tokens": 13728577.0,
"eval_runtime": 20.3407,
"eval_samples_per_second": 1598.419,
"eval_steps_per_second": 199.845,
"step": 6000
},
{
"entropy": 5.693338632583618,
"epoch": 0.5936140767101621,
"grad_norm": 1.09375,
"learning_rate": 0.0004972337396931707,
"loss": 5.4504,
"mean_token_accuracy": 0.17748550027608873,
"num_tokens": 13739016.0,
"step": 6005
},
{
"entropy": 5.71715931892395,
"epoch": 0.5941083432186635,
"grad_norm": 0.97265625,
"learning_rate": 0.0004972282202170421,
"loss": 5.4866,
"mean_token_accuracy": 0.17308816090226173,
"num_tokens": 13749971.0,
"step": 6010
},
{
"entropy": 5.751609849929809,
"epoch": 0.5946026097271648,
"grad_norm": 0.8828125,
"learning_rate": 0.000497222695274065,
"loss": 5.5131,
"mean_token_accuracy": 0.17564282119274138,
"num_tokens": 13760927.0,
"step": 6015
},
{
"entropy": 5.713732385635376,
"epoch": 0.5950968762356663,
"grad_norm": 0.91796875,
"learning_rate": 0.0004972171648643749,
"loss": 5.4866,
"mean_token_accuracy": 0.17753133326768875,
"num_tokens": 13772725.0,
"step": 6020
},
{
"entropy": 5.656687498092651,
"epoch": 0.5955911427441677,
"grad_norm": 0.8515625,
"learning_rate": 0.0004972116289881082,
"loss": 5.4709,
"mean_token_accuracy": 0.18235348165035248,
"num_tokens": 13783427.0,
"step": 6025
},
{
"entropy": 5.675543308258057,
"epoch": 0.5960854092526691,
"grad_norm": 1.0,
"learning_rate": 0.000497206087645401,
"loss": 5.3855,
"mean_token_accuracy": 0.1879747837781906,
"num_tokens": 13794045.0,
"step": 6030
},
{
"entropy": 5.708596134185791,
"epoch": 0.5965796757611704,
"grad_norm": 0.93359375,
"learning_rate": 0.0004972005408363894,
"loss": 5.4546,
"mean_token_accuracy": 0.18292929530143737,
"num_tokens": 13805629.0,
"step": 6035
},
{
"entropy": 5.68604793548584,
"epoch": 0.5970739422696718,
"grad_norm": 0.9453125,
"learning_rate": 0.0004971949885612102,
"loss": 5.5259,
"mean_token_accuracy": 0.17549283504486085,
"num_tokens": 13817526.0,
"step": 6040
},
{
"entropy": 5.755584383010865,
"epoch": 0.5975682087781732,
"grad_norm": 0.921875,
"learning_rate": 0.0004971894308199997,
"loss": 5.4575,
"mean_token_accuracy": 0.18744254857301712,
"num_tokens": 13828553.0,
"step": 6045
},
{
"entropy": 5.711791324615478,
"epoch": 0.5980624752866746,
"grad_norm": 1.0,
"learning_rate": 0.0004971838676128947,
"loss": 5.4763,
"mean_token_accuracy": 0.18320010602474213,
"num_tokens": 13840323.0,
"step": 6050
},
{
"entropy": 5.6459356307983395,
"epoch": 0.5985567417951759,
"grad_norm": 0.77734375,
"learning_rate": 0.0004971782989400323,
"loss": 5.4525,
"mean_token_accuracy": 0.18255051523447036,
"num_tokens": 13853524.0,
"step": 6055
},
{
"entropy": 5.678977108001709,
"epoch": 0.5990510083036773,
"grad_norm": 0.88671875,
"learning_rate": 0.0004971727248015491,
"loss": 5.4296,
"mean_token_accuracy": 0.17558186650276184,
"num_tokens": 13864462.0,
"step": 6060
},
{
"entropy": 5.720080327987671,
"epoch": 0.5995452748121787,
"grad_norm": 0.96875,
"learning_rate": 0.0004971671451975823,
"loss": 5.4622,
"mean_token_accuracy": 0.1817041128873825,
"num_tokens": 13876615.0,
"step": 6065
},
{
"entropy": 5.739533758163452,
"epoch": 0.6000395413206802,
"grad_norm": 0.90625,
"learning_rate": 0.0004971615601282695,
"loss": 5.5643,
"mean_token_accuracy": 0.17829847931861878,
"num_tokens": 13887652.0,
"step": 6070
},
{
"entropy": 5.691545486450195,
"epoch": 0.6005338078291815,
"grad_norm": 0.859375,
"learning_rate": 0.0004971559695937477,
"loss": 5.4386,
"mean_token_accuracy": 0.18535303324460983,
"num_tokens": 13898961.0,
"step": 6075
},
{
"entropy": 5.772193145751953,
"epoch": 0.6010280743376829,
"grad_norm": 0.953125,
"learning_rate": 0.0004971503735941547,
"loss": 5.503,
"mean_token_accuracy": 0.18379769772291182,
"num_tokens": 13909904.0,
"step": 6080
},
{
"entropy": 5.781509637832642,
"epoch": 0.6015223408461843,
"grad_norm": 0.9375,
"learning_rate": 0.000497144772129628,
"loss": 5.513,
"mean_token_accuracy": 0.1743398889899254,
"num_tokens": 13921376.0,
"step": 6085
},
{
"entropy": 5.669444561004639,
"epoch": 0.6020166073546857,
"grad_norm": 0.90234375,
"learning_rate": 0.0004971391652003054,
"loss": 5.4698,
"mean_token_accuracy": 0.18568859994411469,
"num_tokens": 13932925.0,
"step": 6090
},
{
"entropy": 5.6786703109741214,
"epoch": 0.602510873863187,
"grad_norm": 0.93359375,
"learning_rate": 0.000497133552806325,
"loss": 5.4451,
"mean_token_accuracy": 0.18331165611743927,
"num_tokens": 13945010.0,
"step": 6095
},
{
"entropy": 5.706253242492676,
"epoch": 0.6030051403716884,
"grad_norm": 0.9375,
"learning_rate": 0.0004971279349478245,
"loss": 5.4771,
"mean_token_accuracy": 0.16978354305028914,
"num_tokens": 13957710.0,
"step": 6100
},
{
"entropy": 5.6717077732086185,
"epoch": 0.6034994068801898,
"grad_norm": 0.9921875,
"learning_rate": 0.0004971223116249424,
"loss": 5.3586,
"mean_token_accuracy": 0.19802093803882598,
"num_tokens": 13967819.0,
"step": 6105
},
{
"entropy": 5.81233229637146,
"epoch": 0.6039936733886911,
"grad_norm": 0.8828125,
"learning_rate": 0.0004971166828378172,
"loss": 5.64,
"mean_token_accuracy": 0.16502135545015334,
"num_tokens": 13980342.0,
"step": 6110
},
{
"entropy": 5.7933876514434814,
"epoch": 0.6044879398971925,
"grad_norm": 0.8671875,
"learning_rate": 0.000497111048586587,
"loss": 5.5155,
"mean_token_accuracy": 0.17665462493896483,
"num_tokens": 13990370.0,
"step": 6115
},
{
"entropy": 5.74040117263794,
"epoch": 0.604982206405694,
"grad_norm": 1.078125,
"learning_rate": 0.0004971054088713905,
"loss": 5.5152,
"mean_token_accuracy": 0.18113837242126465,
"num_tokens": 14001761.0,
"step": 6120
},
{
"entropy": 5.712833499908447,
"epoch": 0.6054764729141954,
"grad_norm": 0.96484375,
"learning_rate": 0.0004970997636923665,
"loss": 5.5184,
"mean_token_accuracy": 0.17608647644519806,
"num_tokens": 14012980.0,
"step": 6125
},
{
"entropy": 5.735250759124756,
"epoch": 0.6059707394226967,
"grad_norm": 0.8671875,
"learning_rate": 0.000497094113049654,
"loss": 5.39,
"mean_token_accuracy": 0.1825260877609253,
"num_tokens": 14023609.0,
"step": 6130
},
{
"entropy": 5.718038272857666,
"epoch": 0.6064650059311981,
"grad_norm": 0.95703125,
"learning_rate": 0.0004970884569433917,
"loss": 5.5628,
"mean_token_accuracy": 0.17623211890459062,
"num_tokens": 14036052.0,
"step": 6135
},
{
"entropy": 5.63355860710144,
"epoch": 0.6069592724396995,
"grad_norm": 0.83203125,
"learning_rate": 0.000497082795373719,
"loss": 5.3994,
"mean_token_accuracy": 0.18857624679803847,
"num_tokens": 14047932.0,
"step": 6140
},
{
"entropy": 5.746037721633911,
"epoch": 0.6074535389482009,
"grad_norm": 0.89453125,
"learning_rate": 0.0004970771283407749,
"loss": 5.4876,
"mean_token_accuracy": 0.1831844449043274,
"num_tokens": 14059674.0,
"step": 6145
},
{
"entropy": 5.774702501296997,
"epoch": 0.6079478054567022,
"grad_norm": 0.8515625,
"learning_rate": 0.0004970714558446992,
"loss": 5.519,
"mean_token_accuracy": 0.17914710342884063,
"num_tokens": 14070753.0,
"step": 6150
},
{
"entropy": 5.706883811950684,
"epoch": 0.6084420719652036,
"grad_norm": 0.9453125,
"learning_rate": 0.0004970657778856312,
"loss": 5.4845,
"mean_token_accuracy": 0.17967574745416642,
"num_tokens": 14083565.0,
"step": 6155
},
{
"entropy": 5.806309461593628,
"epoch": 0.608936338473705,
"grad_norm": 0.890625,
"learning_rate": 0.0004970600944637106,
"loss": 5.5687,
"mean_token_accuracy": 0.1718636840581894,
"num_tokens": 14095008.0,
"step": 6160
},
{
"entropy": 5.6775635242462155,
"epoch": 0.6094306049822064,
"grad_norm": 0.9765625,
"learning_rate": 0.0004970544055790772,
"loss": 5.3448,
"mean_token_accuracy": 0.18482159674167634,
"num_tokens": 14106586.0,
"step": 6165
},
{
"entropy": 5.585059976577758,
"epoch": 0.6099248714907078,
"grad_norm": 0.93359375,
"learning_rate": 0.000497048711231871,
"loss": 5.3589,
"mean_token_accuracy": 0.18798351436853408,
"num_tokens": 14117778.0,
"step": 6170
},
{
"entropy": 5.679389190673828,
"epoch": 0.6104191379992092,
"grad_norm": 0.97265625,
"learning_rate": 0.0004970430114222319,
"loss": 5.4259,
"mean_token_accuracy": 0.18183008581399918,
"num_tokens": 14129644.0,
"step": 6175
},
{
"entropy": 5.634195613861084,
"epoch": 0.6109134045077106,
"grad_norm": 1.0,
"learning_rate": 0.0004970373061503005,
"loss": 5.3953,
"mean_token_accuracy": 0.18722597807645797,
"num_tokens": 14141230.0,
"step": 6180
},
{
"entropy": 5.713296222686767,
"epoch": 0.611407671016212,
"grad_norm": 0.98828125,
"learning_rate": 0.0004970315954162169,
"loss": 5.5456,
"mean_token_accuracy": 0.1784057155251503,
"num_tokens": 14151622.0,
"step": 6185
},
{
"entropy": 5.786626958847046,
"epoch": 0.6119019375247133,
"grad_norm": 1.015625,
"learning_rate": 0.0004970258792201214,
"loss": 5.4653,
"mean_token_accuracy": 0.18192349523305892,
"num_tokens": 14163641.0,
"step": 6190
},
{
"entropy": 5.71040678024292,
"epoch": 0.6123962040332147,
"grad_norm": 0.9765625,
"learning_rate": 0.000497020157562155,
"loss": 5.4291,
"mean_token_accuracy": 0.1838184729218483,
"num_tokens": 14175154.0,
"step": 6195
},
{
"entropy": 5.720538473129272,
"epoch": 0.6128904705417161,
"grad_norm": 1.015625,
"learning_rate": 0.0004970144304424582,
"loss": 5.4499,
"mean_token_accuracy": 0.18366681635379792,
"num_tokens": 14185472.0,
"step": 6200
},
{
"entropy": 5.685149717330932,
"epoch": 0.6133847370502175,
"grad_norm": 0.92578125,
"learning_rate": 0.000497008697861172,
"loss": 5.4343,
"mean_token_accuracy": 0.1842101126909256,
"num_tokens": 14196992.0,
"step": 6205
},
{
"entropy": 5.6781305313110355,
"epoch": 0.6138790035587188,
"grad_norm": 0.9453125,
"learning_rate": 0.0004970029598184375,
"loss": 5.4137,
"mean_token_accuracy": 0.17972283512353898,
"num_tokens": 14207732.0,
"step": 6210
},
{
"entropy": 5.831124305725098,
"epoch": 0.6143732700672202,
"grad_norm": 0.9453125,
"learning_rate": 0.0004969972163143956,
"loss": 5.5932,
"mean_token_accuracy": 0.16833030581474304,
"num_tokens": 14219164.0,
"step": 6215
},
{
"entropy": 5.7736307144165036,
"epoch": 0.6148675365757216,
"grad_norm": 0.91015625,
"learning_rate": 0.0004969914673491878,
"loss": 5.5368,
"mean_token_accuracy": 0.1713523432612419,
"num_tokens": 14231687.0,
"step": 6220
},
{
"entropy": 5.731515979766845,
"epoch": 0.6153618030842231,
"grad_norm": 1.0078125,
"learning_rate": 0.0004969857129229555,
"loss": 5.3903,
"mean_token_accuracy": 0.1838352233171463,
"num_tokens": 14243099.0,
"step": 6225
},
{
"entropy": 5.7796424388885494,
"epoch": 0.6158560695927244,
"grad_norm": 0.9765625,
"learning_rate": 0.0004969799530358402,
"loss": 5.5754,
"mean_token_accuracy": 0.17443785965442657,
"num_tokens": 14255835.0,
"step": 6230
},
{
"entropy": 5.639222145080566,
"epoch": 0.6163503361012258,
"grad_norm": 0.98046875,
"learning_rate": 0.0004969741876879835,
"loss": 5.414,
"mean_token_accuracy": 0.17747138887643815,
"num_tokens": 14266661.0,
"step": 6235
},
{
"entropy": 5.714282703399658,
"epoch": 0.6168446026097272,
"grad_norm": 1.015625,
"learning_rate": 0.0004969684168795275,
"loss": 5.4768,
"mean_token_accuracy": 0.18221446871757507,
"num_tokens": 14278474.0,
"step": 6240
},
{
"entropy": 5.743070602416992,
"epoch": 0.6173388691182286,
"grad_norm": 0.99609375,
"learning_rate": 0.0004969626406106139,
"loss": 5.447,
"mean_token_accuracy": 0.17949046045541764,
"num_tokens": 14289516.0,
"step": 6245
},
{
"entropy": 5.6380690574646,
"epoch": 0.6178331356267299,
"grad_norm": 0.953125,
"learning_rate": 0.0004969568588813851,
"loss": 5.4481,
"mean_token_accuracy": 0.18283882290124892,
"num_tokens": 14301312.0,
"step": 6250
},
{
"entropy": 5.769012975692749,
"epoch": 0.6183274021352313,
"grad_norm": 0.8515625,
"learning_rate": 0.0004969510716919829,
"loss": 5.4372,
"mean_token_accuracy": 0.18169225454330445,
"num_tokens": 14312009.0,
"step": 6255
},
{
"entropy": 5.668433618545532,
"epoch": 0.6188216686437327,
"grad_norm": 1.0078125,
"learning_rate": 0.00049694527904255,
"loss": 5.4908,
"mean_token_accuracy": 0.18407995700836183,
"num_tokens": 14322754.0,
"step": 6260
},
{
"entropy": 5.659542465209961,
"epoch": 0.619315935152234,
"grad_norm": 0.99609375,
"learning_rate": 0.0004969394809332287,
"loss": 5.3487,
"mean_token_accuracy": 0.18018449544906617,
"num_tokens": 14333939.0,
"step": 6265
},
{
"entropy": 5.687064218521118,
"epoch": 0.6198102016607354,
"grad_norm": 0.96484375,
"learning_rate": 0.0004969336773641619,
"loss": 5.4644,
"mean_token_accuracy": 0.18374679684638978,
"num_tokens": 14345987.0,
"step": 6270
},
{
"entropy": 5.765881681442261,
"epoch": 0.6203044681692369,
"grad_norm": 0.94140625,
"learning_rate": 0.0004969278683354921,
"loss": 5.4386,
"mean_token_accuracy": 0.1718706965446472,
"num_tokens": 14356974.0,
"step": 6275
},
{
"entropy": 5.717262840270996,
"epoch": 0.6207987346777383,
"grad_norm": 0.90234375,
"learning_rate": 0.0004969220538473625,
"loss": 5.4749,
"mean_token_accuracy": 0.1765012577176094,
"num_tokens": 14368175.0,
"step": 6280
},
{
"entropy": 5.707417917251587,
"epoch": 0.6212930011862396,
"grad_norm": 1.015625,
"learning_rate": 0.0004969162338999157,
"loss": 5.5571,
"mean_token_accuracy": 0.17327929586172103,
"num_tokens": 14379067.0,
"step": 6285
},
{
"entropy": 5.778901672363281,
"epoch": 0.621787267694741,
"grad_norm": 0.94140625,
"learning_rate": 0.0004969104084932953,
"loss": 5.5,
"mean_token_accuracy": 0.17797669619321824,
"num_tokens": 14390483.0,
"step": 6290
},
{
"entropy": 5.703991556167603,
"epoch": 0.6222815342032424,
"grad_norm": 0.9453125,
"learning_rate": 0.0004969045776276443,
"loss": 5.5158,
"mean_token_accuracy": 0.1764296755194664,
"num_tokens": 14402757.0,
"step": 6295
},
{
"entropy": 5.662689924240112,
"epoch": 0.6227758007117438,
"grad_norm": 0.92578125,
"learning_rate": 0.0004968987413031063,
"loss": 5.4505,
"mean_token_accuracy": 0.18288048207759858,
"num_tokens": 14413829.0,
"step": 6300
},
{
"entropy": 5.7220807552337645,
"epoch": 0.6232700672202451,
"grad_norm": 0.87109375,
"learning_rate": 0.0004968928995198249,
"loss": 5.5488,
"mean_token_accuracy": 0.17076856046915054,
"num_tokens": 14425165.0,
"step": 6305
},
{
"entropy": 5.812552976608276,
"epoch": 0.6237643337287465,
"grad_norm": 0.9296875,
"learning_rate": 0.0004968870522779437,
"loss": 5.4965,
"mean_token_accuracy": 0.1804996207356453,
"num_tokens": 14436264.0,
"step": 6310
},
{
"entropy": 5.6658447265625,
"epoch": 0.6242586002372479,
"grad_norm": 0.984375,
"learning_rate": 0.0004968811995776066,
"loss": 5.4433,
"mean_token_accuracy": 0.1808133065700531,
"num_tokens": 14447384.0,
"step": 6315
},
{
"entropy": 5.754222202301025,
"epoch": 0.6247528667457493,
"grad_norm": 0.9453125,
"learning_rate": 0.0004968753414189576,
"loss": 5.5226,
"mean_token_accuracy": 0.17646001577377318,
"num_tokens": 14460013.0,
"step": 6320
},
{
"entropy": 5.7590904712677,
"epoch": 0.6252471332542507,
"grad_norm": 0.94140625,
"learning_rate": 0.0004968694778021408,
"loss": 5.4199,
"mean_token_accuracy": 0.18161038011312486,
"num_tokens": 14472452.0,
"step": 6325
},
{
"entropy": 5.69494857788086,
"epoch": 0.6257413997627521,
"grad_norm": 0.81640625,
"learning_rate": 0.0004968636087273003,
"loss": 5.4603,
"mean_token_accuracy": 0.17987588793039322,
"num_tokens": 14484487.0,
"step": 6330
},
{
"entropy": 5.7048008918762205,
"epoch": 0.6262356662712535,
"grad_norm": 0.92578125,
"learning_rate": 0.0004968577341945806,
"loss": 5.3924,
"mean_token_accuracy": 0.1903366833925247,
"num_tokens": 14496043.0,
"step": 6335
},
{
"entropy": 5.676010608673096,
"epoch": 0.6267299327797549,
"grad_norm": 0.9375,
"learning_rate": 0.0004968518542041264,
"loss": 5.5043,
"mean_token_accuracy": 0.17935250848531722,
"num_tokens": 14508089.0,
"step": 6340
},
{
"entropy": 5.800727367401123,
"epoch": 0.6272241992882562,
"grad_norm": 1.015625,
"learning_rate": 0.000496845968756082,
"loss": 5.5601,
"mean_token_accuracy": 0.17207197695970536,
"num_tokens": 14519480.0,
"step": 6345
},
{
"entropy": 5.7255082607269285,
"epoch": 0.6277184657967576,
"grad_norm": 0.84375,
"learning_rate": 0.0004968400778505925,
"loss": 5.4791,
"mean_token_accuracy": 0.1843769818544388,
"num_tokens": 14531214.0,
"step": 6350
},
{
"entropy": 5.761374855041504,
"epoch": 0.628212732305259,
"grad_norm": 0.77734375,
"learning_rate": 0.0004968341814878025,
"loss": 5.5315,
"mean_token_accuracy": 0.18138425350189208,
"num_tokens": 14544118.0,
"step": 6355
},
{
"entropy": 5.763683319091797,
"epoch": 0.6287069988137604,
"grad_norm": 0.87109375,
"learning_rate": 0.0004968282796678573,
"loss": 5.4467,
"mean_token_accuracy": 0.17990829646587372,
"num_tokens": 14555545.0,
"step": 6360
},
{
"entropy": 5.653489446640014,
"epoch": 0.6292012653222617,
"grad_norm": 0.91796875,
"learning_rate": 0.000496822372390902,
"loss": 5.4591,
"mean_token_accuracy": 0.18334181904792785,
"num_tokens": 14567157.0,
"step": 6365
},
{
"entropy": 5.627438592910766,
"epoch": 0.6296955318307631,
"grad_norm": 0.9609375,
"learning_rate": 0.000496816459657082,
"loss": 5.3948,
"mean_token_accuracy": 0.18774764686822892,
"num_tokens": 14577074.0,
"step": 6370
},
{
"entropy": 5.747024059295654,
"epoch": 0.6301897983392646,
"grad_norm": 0.98046875,
"learning_rate": 0.0004968105414665426,
"loss": 5.4767,
"mean_token_accuracy": 0.18496745675802231,
"num_tokens": 14589133.0,
"step": 6375
},
{
"entropy": 5.678619337081909,
"epoch": 0.630684064847766,
"grad_norm": 0.91015625,
"learning_rate": 0.0004968046178194295,
"loss": 5.4328,
"mean_token_accuracy": 0.18086182624101638,
"num_tokens": 14600985.0,
"step": 6380
},
{
"entropy": 5.662907648086548,
"epoch": 0.6311783313562673,
"grad_norm": 0.953125,
"learning_rate": 0.0004967986887158884,
"loss": 5.4258,
"mean_token_accuracy": 0.18445221483707427,
"num_tokens": 14612163.0,
"step": 6385
},
{
"entropy": 5.732915449142456,
"epoch": 0.6316725978647687,
"grad_norm": 0.8359375,
"learning_rate": 0.0004967927541560651,
"loss": 5.5358,
"mean_token_accuracy": 0.18310444355010985,
"num_tokens": 14623045.0,
"step": 6390
},
{
"entropy": 5.775830936431885,
"epoch": 0.6321668643732701,
"grad_norm": 0.96484375,
"learning_rate": 0.0004967868141401056,
"loss": 5.4923,
"mean_token_accuracy": 0.1803148493170738,
"num_tokens": 14634696.0,
"step": 6395
},
{
"entropy": 5.78247389793396,
"epoch": 0.6326611308817714,
"grad_norm": 0.96875,
"learning_rate": 0.0004967808686681562,
"loss": 5.4994,
"mean_token_accuracy": 0.17738671898841857,
"num_tokens": 14645589.0,
"step": 6400
},
{
"entropy": 5.73599271774292,
"epoch": 0.6331553973902728,
"grad_norm": 0.91796875,
"learning_rate": 0.0004967749177403629,
"loss": 5.4137,
"mean_token_accuracy": 0.18212483525276185,
"num_tokens": 14657199.0,
"step": 6405
},
{
"entropy": 5.7224438190460205,
"epoch": 0.6336496638987742,
"grad_norm": 0.94140625,
"learning_rate": 0.0004967689613568723,
"loss": 5.4872,
"mean_token_accuracy": 0.18070882707834243,
"num_tokens": 14668767.0,
"step": 6410
},
{
"entropy": 5.653053569793701,
"epoch": 0.6341439304072756,
"grad_norm": 0.85546875,
"learning_rate": 0.0004967629995178309,
"loss": 5.3948,
"mean_token_accuracy": 0.1870437741279602,
"num_tokens": 14680860.0,
"step": 6415
},
{
"entropy": 5.60442328453064,
"epoch": 0.6346381969157769,
"grad_norm": 0.8984375,
"learning_rate": 0.0004967570322233852,
"loss": 5.3503,
"mean_token_accuracy": 0.1859385684132576,
"num_tokens": 14692134.0,
"step": 6420
},
{
"entropy": 5.683021068572998,
"epoch": 0.6351324634242784,
"grad_norm": 0.88671875,
"learning_rate": 0.0004967510594736822,
"loss": 5.4587,
"mean_token_accuracy": 0.1802861988544464,
"num_tokens": 14703323.0,
"step": 6425
},
{
"entropy": 5.679267406463623,
"epoch": 0.6356267299327798,
"grad_norm": 0.94140625,
"learning_rate": 0.0004967450812688686,
"loss": 5.3853,
"mean_token_accuracy": 0.18792552798986434,
"num_tokens": 14714453.0,
"step": 6430
},
{
"entropy": 5.706775379180908,
"epoch": 0.6361209964412812,
"grad_norm": 0.91796875,
"learning_rate": 0.0004967390976090917,
"loss": 5.4809,
"mean_token_accuracy": 0.1811143323779106,
"num_tokens": 14726562.0,
"step": 6435
},
{
"entropy": 5.642064619064331,
"epoch": 0.6366152629497825,
"grad_norm": 1.0,
"learning_rate": 0.0004967331084944986,
"loss": 5.322,
"mean_token_accuracy": 0.19146147221326829,
"num_tokens": 14738054.0,
"step": 6440
},
{
"entropy": 5.668184995651245,
"epoch": 0.6371095294582839,
"grad_norm": 0.91796875,
"learning_rate": 0.0004967271139252365,
"loss": 5.4312,
"mean_token_accuracy": 0.18423206657171248,
"num_tokens": 14749037.0,
"step": 6445
},
{
"entropy": 5.684755325317383,
"epoch": 0.6376037959667853,
"grad_norm": 0.99609375,
"learning_rate": 0.0004967211139014531,
"loss": 5.4114,
"mean_token_accuracy": 0.18072779774665831,
"num_tokens": 14759667.0,
"step": 6450
},
{
"entropy": 5.665920972824097,
"epoch": 0.6380980624752867,
"grad_norm": 1.0,
"learning_rate": 0.000496715108423296,
"loss": 5.4239,
"mean_token_accuracy": 0.18640340119600296,
"num_tokens": 14770866.0,
"step": 6455
},
{
"entropy": 5.716802549362183,
"epoch": 0.638592328983788,
"grad_norm": 0.97265625,
"learning_rate": 0.0004967090974909127,
"loss": 5.4431,
"mean_token_accuracy": 0.1808260887861252,
"num_tokens": 14783095.0,
"step": 6460
},
{
"entropy": 5.71856918334961,
"epoch": 0.6390865954922894,
"grad_norm": 0.921875,
"learning_rate": 0.0004967030811044512,
"loss": 5.4953,
"mean_token_accuracy": 0.1751694679260254,
"num_tokens": 14794046.0,
"step": 6465
},
{
"entropy": 5.6662344455719,
"epoch": 0.6395808620007908,
"grad_norm": 0.90234375,
"learning_rate": 0.0004966970592640596,
"loss": 5.4917,
"mean_token_accuracy": 0.1810481920838356,
"num_tokens": 14807295.0,
"step": 6470
},
{
"entropy": 5.67359619140625,
"epoch": 0.6400751285092922,
"grad_norm": 0.921875,
"learning_rate": 0.0004966910319698859,
"loss": 5.4362,
"mean_token_accuracy": 0.1912796750664711,
"num_tokens": 14819109.0,
"step": 6475
},
{
"entropy": 5.696849727630616,
"epoch": 0.6405693950177936,
"grad_norm": 0.90234375,
"learning_rate": 0.0004966849992220784,
"loss": 5.3967,
"mean_token_accuracy": 0.18482088297605515,
"num_tokens": 14830719.0,
"step": 6480
},
{
"entropy": 5.688433980941772,
"epoch": 0.641063661526295,
"grad_norm": 1.0234375,
"learning_rate": 0.0004966789610207856,
"loss": 5.5039,
"mean_token_accuracy": 0.17849323749542237,
"num_tokens": 14843392.0,
"step": 6485
},
{
"entropy": 5.730852794647217,
"epoch": 0.6415579280347964,
"grad_norm": 0.91796875,
"learning_rate": 0.0004966729173661559,
"loss": 5.3745,
"mean_token_accuracy": 0.18784392029047012,
"num_tokens": 14855745.0,
"step": 6490
},
{
"entropy": 5.741331386566162,
"epoch": 0.6420521945432978,
"grad_norm": 0.92578125,
"learning_rate": 0.000496666868258338,
"loss": 5.5029,
"mean_token_accuracy": 0.17939966768026352,
"num_tokens": 14868053.0,
"step": 6495
},
{
"entropy": 5.6654657363891605,
"epoch": 0.6425464610517991,
"grad_norm": 0.96875,
"learning_rate": 0.0004966608136974809,
"loss": 5.4241,
"mean_token_accuracy": 0.18736762702465057,
"num_tokens": 14880095.0,
"step": 6500
},
{
"entropy": 5.7884503364562985,
"epoch": 0.6430407275603005,
"grad_norm": 0.90234375,
"learning_rate": 0.0004966547536837333,
"loss": 5.4956,
"mean_token_accuracy": 0.1772681400179863,
"num_tokens": 14891013.0,
"step": 6505
},
{
"entropy": 5.722235155105591,
"epoch": 0.6435349940688019,
"grad_norm": 0.94921875,
"learning_rate": 0.0004966486882172444,
"loss": 5.4058,
"mean_token_accuracy": 0.18554937839508057,
"num_tokens": 14901875.0,
"step": 6510
},
{
"entropy": 5.681920337677002,
"epoch": 0.6440292605773033,
"grad_norm": 1.0078125,
"learning_rate": 0.0004966426172981634,
"loss": 5.453,
"mean_token_accuracy": 0.18451722115278243,
"num_tokens": 14912620.0,
"step": 6515
},
{
"entropy": 5.644273138046264,
"epoch": 0.6445235270858046,
"grad_norm": 1.171875,
"learning_rate": 0.0004966365409266396,
"loss": 5.4135,
"mean_token_accuracy": 0.1952839434146881,
"num_tokens": 14923128.0,
"step": 6520
},
{
"entropy": 5.694646263122559,
"epoch": 0.645017793594306,
"grad_norm": 0.921875,
"learning_rate": 0.0004966304591028225,
"loss": 5.4072,
"mean_token_accuracy": 0.18826288729906082,
"num_tokens": 14933621.0,
"step": 6525
},
{
"entropy": 5.612667226791382,
"epoch": 0.6455120601028075,
"grad_norm": 0.9609375,
"learning_rate": 0.0004966243718268617,
"loss": 5.3909,
"mean_token_accuracy": 0.18682371973991393,
"num_tokens": 14943292.0,
"step": 6530
},
{
"entropy": 5.661767244338989,
"epoch": 0.6460063266113089,
"grad_norm": 0.9296875,
"learning_rate": 0.000496618279098907,
"loss": 5.4411,
"mean_token_accuracy": 0.17924872636795045,
"num_tokens": 14953998.0,
"step": 6535
},
{
"entropy": 5.72789740562439,
"epoch": 0.6465005931198102,
"grad_norm": 0.89453125,
"learning_rate": 0.0004966121809191083,
"loss": 5.4782,
"mean_token_accuracy": 0.17934831082820893,
"num_tokens": 14965430.0,
"step": 6540
},
{
"entropy": 5.682815170288086,
"epoch": 0.6469948596283116,
"grad_norm": 0.8828125,
"learning_rate": 0.0004966060772876156,
"loss": 5.3539,
"mean_token_accuracy": 0.18645986020565034,
"num_tokens": 14977124.0,
"step": 6545
},
{
"entropy": 5.679669809341431,
"epoch": 0.647489126136813,
"grad_norm": 0.99609375,
"learning_rate": 0.0004965999682045788,
"loss": 5.3632,
"mean_token_accuracy": 0.18724352568387986,
"num_tokens": 14987069.0,
"step": 6550
},
{
"entropy": 5.549817800521851,
"epoch": 0.6479833926453143,
"grad_norm": 1.0234375,
"learning_rate": 0.0004965938536701486,
"loss": 5.4502,
"mean_token_accuracy": 0.19603352546691893,
"num_tokens": 14997919.0,
"step": 6555
},
{
"entropy": 5.698160219192505,
"epoch": 0.6484776591538157,
"grad_norm": 0.92578125,
"learning_rate": 0.0004965877336844752,
"loss": 5.4844,
"mean_token_accuracy": 0.18569322377443315,
"num_tokens": 15010042.0,
"step": 6560
},
{
"entropy": 5.714714622497558,
"epoch": 0.6489719256623171,
"grad_norm": 0.94921875,
"learning_rate": 0.0004965816082477091,
"loss": 5.3888,
"mean_token_accuracy": 0.1896207496523857,
"num_tokens": 15021166.0,
"step": 6565
},
{
"entropy": 5.712548923492432,
"epoch": 0.6494661921708185,
"grad_norm": 0.98828125,
"learning_rate": 0.0004965754773600011,
"loss": 5.4551,
"mean_token_accuracy": 0.18120106011629106,
"num_tokens": 15032182.0,
"step": 6570
},
{
"entropy": 5.693567943572998,
"epoch": 0.6499604586793198,
"grad_norm": 1.015625,
"learning_rate": 0.000496569341021502,
"loss": 5.4072,
"mean_token_accuracy": 0.1808716431260109,
"num_tokens": 15043054.0,
"step": 6575
},
{
"entropy": 5.583023166656494,
"epoch": 0.6504547251878213,
"grad_norm": 0.90234375,
"learning_rate": 0.0004965631992323628,
"loss": 5.3431,
"mean_token_accuracy": 0.1918557107448578,
"num_tokens": 15055462.0,
"step": 6580
},
{
"entropy": 5.669038772583008,
"epoch": 0.6509489916963227,
"grad_norm": 1.0234375,
"learning_rate": 0.0004965570519927345,
"loss": 5.458,
"mean_token_accuracy": 0.1815560758113861,
"num_tokens": 15066849.0,
"step": 6585
},
{
"entropy": 5.655196714401245,
"epoch": 0.6514432582048241,
"grad_norm": 0.92578125,
"learning_rate": 0.0004965508993027682,
"loss": 5.4097,
"mean_token_accuracy": 0.18373749703168868,
"num_tokens": 15078388.0,
"step": 6590
},
{
"entropy": 5.70443549156189,
"epoch": 0.6519375247133254,
"grad_norm": 0.90234375,
"learning_rate": 0.0004965447411626156,
"loss": 5.4525,
"mean_token_accuracy": 0.1772635281085968,
"num_tokens": 15090474.0,
"step": 6595
},
{
"entropy": 5.754064655303955,
"epoch": 0.6524317912218268,
"grad_norm": 0.92578125,
"learning_rate": 0.0004965385775724279,
"loss": 5.4529,
"mean_token_accuracy": 0.18001952916383743,
"num_tokens": 15101104.0,
"step": 6600
},
{
"entropy": 5.692561197280884,
"epoch": 0.6529260577303282,
"grad_norm": 0.95703125,
"learning_rate": 0.0004965324085323569,
"loss": 5.4468,
"mean_token_accuracy": 0.1817217543721199,
"num_tokens": 15112036.0,
"step": 6605
},
{
"entropy": 5.763051605224609,
"epoch": 0.6534203242388296,
"grad_norm": 0.9296875,
"learning_rate": 0.0004965262340425541,
"loss": 5.5776,
"mean_token_accuracy": 0.1677619531750679,
"num_tokens": 15124913.0,
"step": 6610
},
{
"entropy": 5.76483302116394,
"epoch": 0.6539145907473309,
"grad_norm": 0.96875,
"learning_rate": 0.0004965200541031717,
"loss": 5.4783,
"mean_token_accuracy": 0.183321450650692,
"num_tokens": 15135555.0,
"step": 6615
},
{
"entropy": 5.733474779129028,
"epoch": 0.6544088572558323,
"grad_norm": 1.0546875,
"learning_rate": 0.0004965138687143616,
"loss": 5.547,
"mean_token_accuracy": 0.1743663564324379,
"num_tokens": 15146646.0,
"step": 6620
},
{
"entropy": 5.72074818611145,
"epoch": 0.6549031237643337,
"grad_norm": 0.9296875,
"learning_rate": 0.0004965076778762759,
"loss": 5.4446,
"mean_token_accuracy": 0.1807381883263588,
"num_tokens": 15157101.0,
"step": 6625
},
{
"entropy": 5.7268517971038815,
"epoch": 0.6553973902728352,
"grad_norm": 0.8671875,
"learning_rate": 0.0004965014815890671,
"loss": 5.4269,
"mean_token_accuracy": 0.18116676807403564,
"num_tokens": 15169645.0,
"step": 6630
},
{
"entropy": 5.694109916687012,
"epoch": 0.6558916567813365,
"grad_norm": 0.94921875,
"learning_rate": 0.0004964952798528874,
"loss": 5.3433,
"mean_token_accuracy": 0.1830903097987175,
"num_tokens": 15181018.0,
"step": 6635
},
{
"entropy": 5.657157897949219,
"epoch": 0.6563859232898379,
"grad_norm": 0.95703125,
"learning_rate": 0.0004964890726678894,
"loss": 5.4185,
"mean_token_accuracy": 0.1775732085108757,
"num_tokens": 15191880.0,
"step": 6640
},
{
"entropy": 5.610965061187744,
"epoch": 0.6568801897983393,
"grad_norm": 0.9296875,
"learning_rate": 0.0004964828600342259,
"loss": 5.3963,
"mean_token_accuracy": 0.18650533705949784,
"num_tokens": 15203886.0,
"step": 6645
},
{
"entropy": 5.650919437408447,
"epoch": 0.6573744563068407,
"grad_norm": 0.85546875,
"learning_rate": 0.0004964766419520497,
"loss": 5.3708,
"mean_token_accuracy": 0.1918613001704216,
"num_tokens": 15215837.0,
"step": 6650
},
{
"entropy": 5.725353145599366,
"epoch": 0.657868722815342,
"grad_norm": 0.89453125,
"learning_rate": 0.0004964704184215138,
"loss": 5.4839,
"mean_token_accuracy": 0.17489349395036696,
"num_tokens": 15226833.0,
"step": 6655
},
{
"entropy": 5.705512523651123,
"epoch": 0.6583629893238434,
"grad_norm": 0.91796875,
"learning_rate": 0.0004964641894427712,
"loss": 5.5236,
"mean_token_accuracy": 0.17734523564577104,
"num_tokens": 15239312.0,
"step": 6660
},
{
"entropy": 5.673008966445923,
"epoch": 0.6588572558323448,
"grad_norm": 1.0,
"learning_rate": 0.0004964579550159752,
"loss": 5.3939,
"mean_token_accuracy": 0.1783429428935051,
"num_tokens": 15250348.0,
"step": 6665
},
{
"entropy": 5.684113836288452,
"epoch": 0.6593515223408462,
"grad_norm": 0.98828125,
"learning_rate": 0.0004964517151412791,
"loss": 5.4686,
"mean_token_accuracy": 0.18892684876918792,
"num_tokens": 15261445.0,
"step": 6670
},
{
"entropy": 5.699908685684204,
"epoch": 0.6598457888493475,
"grad_norm": 0.9921875,
"learning_rate": 0.0004964454698188364,
"loss": 5.3853,
"mean_token_accuracy": 0.18405035734176636,
"num_tokens": 15271606.0,
"step": 6675
},
{
"entropy": 5.6878753185272215,
"epoch": 0.6603400553578489,
"grad_norm": 0.9296875,
"learning_rate": 0.000496439219048801,
"loss": 5.4544,
"mean_token_accuracy": 0.1803918570280075,
"num_tokens": 15283690.0,
"step": 6680
},
{
"entropy": 5.638463973999023,
"epoch": 0.6608343218663504,
"grad_norm": 0.92578125,
"learning_rate": 0.0004964329628313263,
"loss": 5.4325,
"mean_token_accuracy": 0.1880164697766304,
"num_tokens": 15294792.0,
"step": 6685
},
{
"entropy": 5.757958173751831,
"epoch": 0.6613285883748518,
"grad_norm": 0.98046875,
"learning_rate": 0.0004964267011665664,
"loss": 5.4569,
"mean_token_accuracy": 0.17849235981702805,
"num_tokens": 15305636.0,
"step": 6690
},
{
"entropy": 5.705075693130493,
"epoch": 0.6618228548833531,
"grad_norm": 1.03125,
"learning_rate": 0.0004964204340546753,
"loss": 5.4368,
"mean_token_accuracy": 0.18019956052303315,
"num_tokens": 15316495.0,
"step": 6695
},
{
"entropy": 5.686946821212769,
"epoch": 0.6623171213918545,
"grad_norm": 0.890625,
"learning_rate": 0.0004964141614958071,
"loss": 5.4207,
"mean_token_accuracy": 0.1857387065887451,
"num_tokens": 15328246.0,
"step": 6700
},
{
"entropy": 5.73296275138855,
"epoch": 0.6628113879003559,
"grad_norm": 0.92578125,
"learning_rate": 0.0004964078834901163,
"loss": 5.3829,
"mean_token_accuracy": 0.18510466516017915,
"num_tokens": 15339229.0,
"step": 6705
},
{
"entropy": 5.669465160369873,
"epoch": 0.6633056544088572,
"grad_norm": 0.95703125,
"learning_rate": 0.000496401600037757,
"loss": 5.4322,
"mean_token_accuracy": 0.1813725098967552,
"num_tokens": 15350591.0,
"step": 6710
},
{
"entropy": 5.705862951278687,
"epoch": 0.6637999209173586,
"grad_norm": 0.81640625,
"learning_rate": 0.0004963953111388842,
"loss": 5.5181,
"mean_token_accuracy": 0.17732539474964143,
"num_tokens": 15363726.0,
"step": 6715
},
{
"entropy": 5.677817535400391,
"epoch": 0.66429418742586,
"grad_norm": 0.8359375,
"learning_rate": 0.0004963890167936524,
"loss": 5.3976,
"mean_token_accuracy": 0.1843595862388611,
"num_tokens": 15375569.0,
"step": 6720
},
{
"entropy": 5.709108495712281,
"epoch": 0.6647884539343614,
"grad_norm": 0.87890625,
"learning_rate": 0.0004963827170022164,
"loss": 5.4842,
"mean_token_accuracy": 0.16931309849023818,
"num_tokens": 15387323.0,
"step": 6725
},
{
"entropy": 5.669873046875,
"epoch": 0.6652827204428627,
"grad_norm": 1.0390625,
"learning_rate": 0.0004963764117647311,
"loss": 5.3659,
"mean_token_accuracy": 0.18145694583654404,
"num_tokens": 15397912.0,
"step": 6730
},
{
"entropy": 5.589202356338501,
"epoch": 0.6657769869513642,
"grad_norm": 0.94921875,
"learning_rate": 0.0004963701010813519,
"loss": 5.2732,
"mean_token_accuracy": 0.20091821402311325,
"num_tokens": 15409469.0,
"step": 6735
},
{
"entropy": 5.6283924102783205,
"epoch": 0.6662712534598656,
"grad_norm": 1.0234375,
"learning_rate": 0.0004963637849522338,
"loss": 5.3415,
"mean_token_accuracy": 0.18584844768047332,
"num_tokens": 15419681.0,
"step": 6740
},
{
"entropy": 5.706773376464843,
"epoch": 0.666765519968367,
"grad_norm": 0.83203125,
"learning_rate": 0.0004963574633775323,
"loss": 5.4978,
"mean_token_accuracy": 0.1782970607280731,
"num_tokens": 15431237.0,
"step": 6745
},
{
"entropy": 5.719655752182007,
"epoch": 0.6672597864768683,
"grad_norm": 1.0234375,
"learning_rate": 0.0004963511363574027,
"loss": 5.4412,
"mean_token_accuracy": 0.17875704169273376,
"num_tokens": 15442366.0,
"step": 6750
},
{
"entropy": 5.709488773345948,
"epoch": 0.6677540529853697,
"grad_norm": 0.890625,
"learning_rate": 0.0004963448038920009,
"loss": 5.4114,
"mean_token_accuracy": 0.1814780816435814,
"num_tokens": 15453121.0,
"step": 6755
},
{
"entropy": 5.659934949874878,
"epoch": 0.6682483194938711,
"grad_norm": 0.9765625,
"learning_rate": 0.0004963384659814827,
"loss": 5.3875,
"mean_token_accuracy": 0.18677566051483155,
"num_tokens": 15463489.0,
"step": 6760
},
{
"entropy": 5.6461833953857425,
"epoch": 0.6687425860023725,
"grad_norm": 0.87890625,
"learning_rate": 0.0004963321226260038,
"loss": 5.4453,
"mean_token_accuracy": 0.17960599809885025,
"num_tokens": 15475217.0,
"step": 6765
},
{
"entropy": 5.665601205825806,
"epoch": 0.6692368525108738,
"grad_norm": 0.97265625,
"learning_rate": 0.0004963257738257203,
"loss": 5.4315,
"mean_token_accuracy": 0.1814910888671875,
"num_tokens": 15485844.0,
"step": 6770
},
{
"entropy": 5.667984580993652,
"epoch": 0.6697311190193752,
"grad_norm": 0.875,
"learning_rate": 0.0004963194195807886,
"loss": 5.416,
"mean_token_accuracy": 0.18200096637010574,
"num_tokens": 15496835.0,
"step": 6775
},
{
"entropy": 5.6734765529632565,
"epoch": 0.6702253855278766,
"grad_norm": 0.95703125,
"learning_rate": 0.0004963130598913646,
"loss": 5.4453,
"mean_token_accuracy": 0.1851516157388687,
"num_tokens": 15508087.0,
"step": 6780
},
{
"entropy": 5.764274644851684,
"epoch": 0.6707196520363781,
"grad_norm": 0.9375,
"learning_rate": 0.0004963066947576051,
"loss": 5.4538,
"mean_token_accuracy": 0.1808246150612831,
"num_tokens": 15519596.0,
"step": 6785
},
{
"entropy": 5.690254163742066,
"epoch": 0.6712139185448794,
"grad_norm": 0.91796875,
"learning_rate": 0.0004963003241796666,
"loss": 5.4539,
"mean_token_accuracy": 0.17773212790489196,
"num_tokens": 15530830.0,
"step": 6790
},
{
"entropy": 5.654678916931152,
"epoch": 0.6717081850533808,
"grad_norm": 0.875,
"learning_rate": 0.0004962939481577059,
"loss": 5.3579,
"mean_token_accuracy": 0.1938735544681549,
"num_tokens": 15543195.0,
"step": 6795
},
{
"entropy": 5.7092162609100345,
"epoch": 0.6722024515618822,
"grad_norm": 0.890625,
"learning_rate": 0.0004962875666918796,
"loss": 5.4617,
"mean_token_accuracy": 0.18274641782045364,
"num_tokens": 15554465.0,
"step": 6800
},
{
"entropy": 5.616643476486206,
"epoch": 0.6726967180703836,
"grad_norm": 0.9375,
"learning_rate": 0.0004962811797823448,
"loss": 5.4044,
"mean_token_accuracy": 0.1812501907348633,
"num_tokens": 15566016.0,
"step": 6805
},
{
"entropy": 5.737135314941407,
"epoch": 0.6731909845788849,
"grad_norm": 0.91015625,
"learning_rate": 0.0004962747874292588,
"loss": 5.5203,
"mean_token_accuracy": 0.17539905160665512,
"num_tokens": 15576843.0,
"step": 6810
},
{
"entropy": 5.699502229690552,
"epoch": 0.6736852510873863,
"grad_norm": 0.890625,
"learning_rate": 0.0004962683896327786,
"loss": 5.4387,
"mean_token_accuracy": 0.18047475665807725,
"num_tokens": 15588390.0,
"step": 6815
},
{
"entropy": 5.728416013717651,
"epoch": 0.6741795175958877,
"grad_norm": 0.9921875,
"learning_rate": 0.0004962619863930616,
"loss": 5.5092,
"mean_token_accuracy": 0.1804381787776947,
"num_tokens": 15600270.0,
"step": 6820
},
{
"entropy": 5.728389406204224,
"epoch": 0.674673784104389,
"grad_norm": 0.9765625,
"learning_rate": 0.0004962555777102656,
"loss": 5.3748,
"mean_token_accuracy": 0.18523539155721663,
"num_tokens": 15611617.0,
"step": 6825
},
{
"entropy": 5.642977905273438,
"epoch": 0.6751680506128904,
"grad_norm": 0.953125,
"learning_rate": 0.000496249163584548,
"loss": 5.4547,
"mean_token_accuracy": 0.1856302410364151,
"num_tokens": 15622128.0,
"step": 6830
},
{
"entropy": 5.7459893226623535,
"epoch": 0.6756623171213919,
"grad_norm": 0.96875,
"learning_rate": 0.0004962427440160665,
"loss": 5.452,
"mean_token_accuracy": 0.17632827013731003,
"num_tokens": 15635250.0,
"step": 6835
},
{
"entropy": 5.799910354614258,
"epoch": 0.6761565836298933,
"grad_norm": 0.94140625,
"learning_rate": 0.0004962363190049794,
"loss": 5.529,
"mean_token_accuracy": 0.17796835601329802,
"num_tokens": 15646535.0,
"step": 6840
},
{
"entropy": 5.697930431365966,
"epoch": 0.6766508501383947,
"grad_norm": 1.0078125,
"learning_rate": 0.0004962298885514444,
"loss": 5.4785,
"mean_token_accuracy": 0.18419041037559508,
"num_tokens": 15658399.0,
"step": 6845
},
{
"entropy": 5.688305950164795,
"epoch": 0.677145116646896,
"grad_norm": 0.96484375,
"learning_rate": 0.0004962234526556199,
"loss": 5.4303,
"mean_token_accuracy": 0.18858383893966674,
"num_tokens": 15669321.0,
"step": 6850
},
{
"entropy": 5.68391375541687,
"epoch": 0.6776393831553974,
"grad_norm": 0.94140625,
"learning_rate": 0.000496217011317664,
"loss": 5.4069,
"mean_token_accuracy": 0.18412790149450303,
"num_tokens": 15681666.0,
"step": 6855
},
{
"entropy": 5.601411294937134,
"epoch": 0.6781336496638988,
"grad_norm": 0.90625,
"learning_rate": 0.0004962105645377354,
"loss": 5.3855,
"mean_token_accuracy": 0.18707387149333954,
"num_tokens": 15693432.0,
"step": 6860
},
{
"entropy": 5.693903207778931,
"epoch": 0.6786279161724001,
"grad_norm": 0.875,
"learning_rate": 0.0004962041123159926,
"loss": 5.3564,
"mean_token_accuracy": 0.1900697097182274,
"num_tokens": 15705151.0,
"step": 6865
},
{
"entropy": 5.663706636428833,
"epoch": 0.6791221826809015,
"grad_norm": 0.93359375,
"learning_rate": 0.0004961976546525943,
"loss": 5.3544,
"mean_token_accuracy": 0.18783595263957978,
"num_tokens": 15716549.0,
"step": 6870
},
{
"entropy": 5.701740694046021,
"epoch": 0.6796164491894029,
"grad_norm": 0.9296875,
"learning_rate": 0.0004961911915476994,
"loss": 5.4964,
"mean_token_accuracy": 0.18121766448020935,
"num_tokens": 15727499.0,
"step": 6875
},
{
"entropy": 5.651185655593872,
"epoch": 0.6801107156979043,
"grad_norm": 0.91015625,
"learning_rate": 0.0004961847230014668,
"loss": 5.4082,
"mean_token_accuracy": 0.1849137842655182,
"num_tokens": 15739196.0,
"step": 6880
},
{
"entropy": 5.671923065185547,
"epoch": 0.6806049822064056,
"grad_norm": 0.87109375,
"learning_rate": 0.0004961782490140558,
"loss": 5.3964,
"mean_token_accuracy": 0.18903211653232574,
"num_tokens": 15750890.0,
"step": 6885
},
{
"entropy": 5.662531900405884,
"epoch": 0.6810992487149071,
"grad_norm": 0.88671875,
"learning_rate": 0.0004961717695856256,
"loss": 5.4241,
"mean_token_accuracy": 0.1925715312361717,
"num_tokens": 15762366.0,
"step": 6890
},
{
"entropy": 5.617862510681152,
"epoch": 0.6815935152234085,
"grad_norm": 0.92578125,
"learning_rate": 0.0004961652847163354,
"loss": 5.4179,
"mean_token_accuracy": 0.18369491398334503,
"num_tokens": 15773962.0,
"step": 6895
},
{
"entropy": 5.641139554977417,
"epoch": 0.6820877817319099,
"grad_norm": 0.921875,
"learning_rate": 0.000496158794406345,
"loss": 5.2761,
"mean_token_accuracy": 0.1957280844449997,
"num_tokens": 15784563.0,
"step": 6900
},
{
"entropy": 5.650210762023926,
"epoch": 0.6825820482404112,
"grad_norm": 0.88671875,
"learning_rate": 0.0004961522986558139,
"loss": 5.4008,
"mean_token_accuracy": 0.1822483003139496,
"num_tokens": 15796461.0,
"step": 6905
},
{
"entropy": 5.6847318649292,
"epoch": 0.6830763147489126,
"grad_norm": 0.90234375,
"learning_rate": 0.000496145797464902,
"loss": 5.4189,
"mean_token_accuracy": 0.17998380810022355,
"num_tokens": 15807303.0,
"step": 6910
},
{
"entropy": 5.626545524597168,
"epoch": 0.683570581257414,
"grad_norm": 1.0234375,
"learning_rate": 0.0004961392908337691,
"loss": 5.34,
"mean_token_accuracy": 0.19580137580633164,
"num_tokens": 15818964.0,
"step": 6915
},
{
"entropy": 5.659780597686767,
"epoch": 0.6840648477659154,
"grad_norm": 0.875,
"learning_rate": 0.0004961327787625754,
"loss": 5.5063,
"mean_token_accuracy": 0.1796172648668289,
"num_tokens": 15831181.0,
"step": 6920
},
{
"entropy": 5.649292850494385,
"epoch": 0.6845591142744167,
"grad_norm": 0.890625,
"learning_rate": 0.000496126261251481,
"loss": 5.3687,
"mean_token_accuracy": 0.18648644983768464,
"num_tokens": 15842442.0,
"step": 6925
},
{
"entropy": 5.708519363403321,
"epoch": 0.6850533807829181,
"grad_norm": 1.0,
"learning_rate": 0.0004961197383006463,
"loss": 5.4656,
"mean_token_accuracy": 0.18111221343278885,
"num_tokens": 15854755.0,
"step": 6930
},
{
"entropy": 5.565923118591309,
"epoch": 0.6855476472914195,
"grad_norm": 0.9765625,
"learning_rate": 0.0004961132099102316,
"loss": 5.3365,
"mean_token_accuracy": 0.19061403423547746,
"num_tokens": 15866300.0,
"step": 6935
},
{
"entropy": 5.643864345550537,
"epoch": 0.686041913799921,
"grad_norm": 0.95703125,
"learning_rate": 0.0004961066760803978,
"loss": 5.3409,
"mean_token_accuracy": 0.19204534143209456,
"num_tokens": 15877165.0,
"step": 6940
},
{
"entropy": 5.722785520553589,
"epoch": 0.6865361803084223,
"grad_norm": 0.94140625,
"learning_rate": 0.0004961001368113054,
"loss": 5.4101,
"mean_token_accuracy": 0.17978667616844177,
"num_tokens": 15888893.0,
"step": 6945
},
{
"entropy": 5.6792816638946535,
"epoch": 0.6870304468169237,
"grad_norm": 0.9296875,
"learning_rate": 0.0004960935921031153,
"loss": 5.4407,
"mean_token_accuracy": 0.18164823651313783,
"num_tokens": 15899994.0,
"step": 6950
},
{
"entropy": 5.578463697433472,
"epoch": 0.6875247133254251,
"grad_norm": 0.94140625,
"learning_rate": 0.0004960870419559887,
"loss": 5.361,
"mean_token_accuracy": 0.1927930772304535,
"num_tokens": 15911330.0,
"step": 6955
},
{
"entropy": 5.645543813705444,
"epoch": 0.6880189798339265,
"grad_norm": 1.0078125,
"learning_rate": 0.0004960804863700864,
"loss": 5.4243,
"mean_token_accuracy": 0.19140509217977525,
"num_tokens": 15922397.0,
"step": 6960
},
{
"entropy": 5.668138217926026,
"epoch": 0.6885132463424278,
"grad_norm": 0.91796875,
"learning_rate": 0.0004960739253455699,
"loss": 5.4045,
"mean_token_accuracy": 0.18389109820127486,
"num_tokens": 15934628.0,
"step": 6965
},
{
"entropy": 5.627578592300415,
"epoch": 0.6890075128509292,
"grad_norm": 0.93359375,
"learning_rate": 0.0004960673588826005,
"loss": 5.3867,
"mean_token_accuracy": 0.18917313814163209,
"num_tokens": 15946599.0,
"step": 6970
},
{
"entropy": 5.694380283355713,
"epoch": 0.6895017793594306,
"grad_norm": 0.97265625,
"learning_rate": 0.0004960607869813398,
"loss": 5.4006,
"mean_token_accuracy": 0.18457768559455873,
"num_tokens": 15957402.0,
"step": 6975
},
{
"entropy": 5.700537586212159,
"epoch": 0.689996045867932,
"grad_norm": 1.0390625,
"learning_rate": 0.0004960542096419494,
"loss": 5.4569,
"mean_token_accuracy": 0.18641959130764008,
"num_tokens": 15968091.0,
"step": 6980
},
{
"entropy": 5.6515697002410885,
"epoch": 0.6904903123764333,
"grad_norm": 0.953125,
"learning_rate": 0.0004960476268645912,
"loss": 5.4422,
"mean_token_accuracy": 0.18348170667886735,
"num_tokens": 15980809.0,
"step": 6985
},
{
"entropy": 5.681520938873291,
"epoch": 0.6909845788849348,
"grad_norm": 0.87890625,
"learning_rate": 0.0004960410386494271,
"loss": 5.4311,
"mean_token_accuracy": 0.1831069767475128,
"num_tokens": 15992701.0,
"step": 6990
},
{
"entropy": 5.77148928642273,
"epoch": 0.6914788453934362,
"grad_norm": 0.89453125,
"learning_rate": 0.0004960344449966191,
"loss": 5.5076,
"mean_token_accuracy": 0.17651577144861222,
"num_tokens": 16005871.0,
"step": 6995
},
{
"entropy": 5.763817024230957,
"epoch": 0.6919731119019376,
"grad_norm": 0.8515625,
"learning_rate": 0.0004960278459063296,
"loss": 5.496,
"mean_token_accuracy": 0.17383982986211777,
"num_tokens": 16018126.0,
"step": 7000
},
{
"entropy": 5.660416507720948,
"epoch": 0.6924673784104389,
"grad_norm": 0.9140625,
"learning_rate": 0.0004960212413787206,
"loss": 5.3537,
"mean_token_accuracy": 0.18668118119239807,
"num_tokens": 16028694.0,
"step": 7005
},
{
"entropy": 5.609909439086914,
"epoch": 0.6929616449189403,
"grad_norm": 0.96875,
"learning_rate": 0.000496014631413955,
"loss": 5.4636,
"mean_token_accuracy": 0.1778472289443016,
"num_tokens": 16040354.0,
"step": 7010
},
{
"entropy": 5.6946430683135985,
"epoch": 0.6934559114274417,
"grad_norm": 0.84765625,
"learning_rate": 0.0004960080160121951,
"loss": 5.3964,
"mean_token_accuracy": 0.18865488320589066,
"num_tokens": 16051713.0,
"step": 7015
},
{
"entropy": 5.695664024353027,
"epoch": 0.693950177935943,
"grad_norm": 0.96484375,
"learning_rate": 0.0004960013951736036,
"loss": 5.4765,
"mean_token_accuracy": 0.17734843641519546,
"num_tokens": 16064737.0,
"step": 7020
},
{
"entropy": 5.65210599899292,
"epoch": 0.6944444444444444,
"grad_norm": 0.94140625,
"learning_rate": 0.0004959947688983436,
"loss": 5.348,
"mean_token_accuracy": 0.18901321440935134,
"num_tokens": 16075515.0,
"step": 7025
},
{
"entropy": 5.674917221069336,
"epoch": 0.6949387109529458,
"grad_norm": 0.79296875,
"learning_rate": 0.0004959881371865779,
"loss": 5.4124,
"mean_token_accuracy": 0.18163476139307022,
"num_tokens": 16088702.0,
"step": 7030
},
{
"entropy": 5.660607814788818,
"epoch": 0.6954329774614472,
"grad_norm": 0.96875,
"learning_rate": 0.0004959815000384699,
"loss": 5.347,
"mean_token_accuracy": 0.18652654588222503,
"num_tokens": 16101122.0,
"step": 7035
},
{
"entropy": 5.707927942276001,
"epoch": 0.6959272439699487,
"grad_norm": 0.84375,
"learning_rate": 0.0004959748574541826,
"loss": 5.4268,
"mean_token_accuracy": 0.18145518451929094,
"num_tokens": 16112817.0,
"step": 7040
},
{
"entropy": 5.687083339691162,
"epoch": 0.69642151047845,
"grad_norm": 0.85546875,
"learning_rate": 0.0004959682094338795,
"loss": 5.5114,
"mean_token_accuracy": 0.1754821315407753,
"num_tokens": 16124827.0,
"step": 7045
},
{
"entropy": 5.652220296859741,
"epoch": 0.6969157769869514,
"grad_norm": 0.92578125,
"learning_rate": 0.0004959615559777243,
"loss": 5.3851,
"mean_token_accuracy": 0.18845241516828537,
"num_tokens": 16136940.0,
"step": 7050
},
{
"entropy": 5.713035345077515,
"epoch": 0.6974100434954528,
"grad_norm": 1.0,
"learning_rate": 0.0004959548970858805,
"loss": 5.3998,
"mean_token_accuracy": 0.18737876713275908,
"num_tokens": 16147541.0,
"step": 7055
},
{
"entropy": 5.636923933029175,
"epoch": 0.6979043100039541,
"grad_norm": 0.90625,
"learning_rate": 0.0004959482327585119,
"loss": 5.4234,
"mean_token_accuracy": 0.17895619124174117,
"num_tokens": 16157626.0,
"step": 7060
},
{
"entropy": 5.645445966720581,
"epoch": 0.6983985765124555,
"grad_norm": 0.8203125,
"learning_rate": 0.0004959415629957824,
"loss": 5.3751,
"mean_token_accuracy": 0.1890865683555603,
"num_tokens": 16168494.0,
"step": 7065
},
{
"entropy": 5.677611780166626,
"epoch": 0.6988928430209569,
"grad_norm": 0.8671875,
"learning_rate": 0.0004959348877978563,
"loss": 5.476,
"mean_token_accuracy": 0.18335221856832504,
"num_tokens": 16180749.0,
"step": 7070
},
{
"entropy": 5.677210569381714,
"epoch": 0.6993871095294583,
"grad_norm": 0.91015625,
"learning_rate": 0.0004959282071648977,
"loss": 5.4346,
"mean_token_accuracy": 0.18475838452577592,
"num_tokens": 16193204.0,
"step": 7075
},
{
"entropy": 5.656859493255615,
"epoch": 0.6998813760379596,
"grad_norm": 0.95703125,
"learning_rate": 0.000495921521097071,
"loss": 5.3917,
"mean_token_accuracy": 0.1933037355542183,
"num_tokens": 16204303.0,
"step": 7080
},
{
"entropy": 5.654446697235107,
"epoch": 0.700375642546461,
"grad_norm": 0.9375,
"learning_rate": 0.0004959148295945405,
"loss": 5.343,
"mean_token_accuracy": 0.19118872731924058,
"num_tokens": 16215261.0,
"step": 7085
},
{
"entropy": 5.654616737365723,
"epoch": 0.7008699090549624,
"grad_norm": 0.83984375,
"learning_rate": 0.0004959081326574708,
"loss": 5.3642,
"mean_token_accuracy": 0.1869739294052124,
"num_tokens": 16226973.0,
"step": 7090
},
{
"entropy": 5.537435150146484,
"epoch": 0.7013641755634639,
"grad_norm": 0.921875,
"learning_rate": 0.0004959014302860271,
"loss": 5.2474,
"mean_token_accuracy": 0.2000797674059868,
"num_tokens": 16239152.0,
"step": 7095
},
{
"entropy": 5.671125364303589,
"epoch": 0.7018584420719652,
"grad_norm": 0.81640625,
"learning_rate": 0.0004958947224803737,
"loss": 5.3728,
"mean_token_accuracy": 0.18841296285390854,
"num_tokens": 16251884.0,
"step": 7100
},
{
"entropy": 5.682699298858642,
"epoch": 0.7023527085804666,
"grad_norm": 0.92578125,
"learning_rate": 0.0004958880092406759,
"loss": 5.4536,
"mean_token_accuracy": 0.17482409477233887,
"num_tokens": 16262497.0,
"step": 7105
},
{
"entropy": 5.6305183410644535,
"epoch": 0.702846975088968,
"grad_norm": 0.94140625,
"learning_rate": 0.0004958812905670988,
"loss": 5.3894,
"mean_token_accuracy": 0.1892905205488205,
"num_tokens": 16274624.0,
"step": 7110
},
{
"entropy": 5.746497869491577,
"epoch": 0.7033412415974694,
"grad_norm": 1.0234375,
"learning_rate": 0.0004958745664598077,
"loss": 5.4087,
"mean_token_accuracy": 0.1858081564307213,
"num_tokens": 16286096.0,
"step": 7115
},
{
"entropy": 5.634428548812866,
"epoch": 0.7038355081059707,
"grad_norm": 0.88671875,
"learning_rate": 0.0004958678369189679,
"loss": 5.3311,
"mean_token_accuracy": 0.19181613475084305,
"num_tokens": 16297853.0,
"step": 7120
},
{
"entropy": 5.649743986129761,
"epoch": 0.7043297746144721,
"grad_norm": 0.90625,
"learning_rate": 0.0004958611019447452,
"loss": 5.4854,
"mean_token_accuracy": 0.18137550055980683,
"num_tokens": 16310708.0,
"step": 7125
},
{
"entropy": 5.701823949813843,
"epoch": 0.7048240411229735,
"grad_norm": 1.125,
"learning_rate": 0.0004958543615373051,
"loss": 5.4069,
"mean_token_accuracy": 0.1839786797761917,
"num_tokens": 16321216.0,
"step": 7130
},
{
"entropy": 5.64765739440918,
"epoch": 0.7053183076314749,
"grad_norm": 1.078125,
"learning_rate": 0.0004958476156968134,
"loss": 5.3623,
"mean_token_accuracy": 0.19156715869903565,
"num_tokens": 16331982.0,
"step": 7135
},
{
"entropy": 5.653167867660523,
"epoch": 0.7058125741399762,
"grad_norm": 0.984375,
"learning_rate": 0.000495840864423436,
"loss": 5.4532,
"mean_token_accuracy": 0.18317919969558716,
"num_tokens": 16343001.0,
"step": 7140
},
{
"entropy": 5.750587844848633,
"epoch": 0.7063068406484777,
"grad_norm": 0.88671875,
"learning_rate": 0.0004958341077173392,
"loss": 5.387,
"mean_token_accuracy": 0.1854384422302246,
"num_tokens": 16355519.0,
"step": 7145
},
{
"entropy": 5.678952312469482,
"epoch": 0.7068011071569791,
"grad_norm": 0.9609375,
"learning_rate": 0.0004958273455786892,
"loss": 5.3761,
"mean_token_accuracy": 0.18318860083818436,
"num_tokens": 16366342.0,
"step": 7150
},
{
"entropy": 5.694130849838257,
"epoch": 0.7072953736654805,
"grad_norm": 0.875,
"learning_rate": 0.0004958205780076521,
"loss": 5.4616,
"mean_token_accuracy": 0.17927791625261308,
"num_tokens": 16378395.0,
"step": 7155
},
{
"entropy": 5.663529253005981,
"epoch": 0.7077896401739818,
"grad_norm": 0.90234375,
"learning_rate": 0.0004958138050043945,
"loss": 5.3573,
"mean_token_accuracy": 0.18564121425151825,
"num_tokens": 16388946.0,
"step": 7160
},
{
"entropy": 5.578586578369141,
"epoch": 0.7082839066824832,
"grad_norm": 0.99609375,
"learning_rate": 0.0004958070265690831,
"loss": 5.4235,
"mean_token_accuracy": 0.18518381416797638,
"num_tokens": 16399824.0,
"step": 7165
},
{
"entropy": 5.747214651107788,
"epoch": 0.7087781731909846,
"grad_norm": 1.0078125,
"learning_rate": 0.0004958002427018845,
"loss": 5.4132,
"mean_token_accuracy": 0.17862845510244368,
"num_tokens": 16411932.0,
"step": 7170
},
{
"entropy": 5.6991194725036625,
"epoch": 0.709272439699486,
"grad_norm": 0.9296875,
"learning_rate": 0.0004957934534029658,
"loss": 5.4006,
"mean_token_accuracy": 0.18126825988292694,
"num_tokens": 16423634.0,
"step": 7175
},
{
"entropy": 5.715042686462402,
"epoch": 0.7097667062079873,
"grad_norm": 0.93359375,
"learning_rate": 0.0004957866586724938,
"loss": 5.4698,
"mean_token_accuracy": 0.18193342089653014,
"num_tokens": 16434700.0,
"step": 7180
},
{
"entropy": 5.74013319015503,
"epoch": 0.7102609727164887,
"grad_norm": 0.8984375,
"learning_rate": 0.0004957798585106359,
"loss": 5.4258,
"mean_token_accuracy": 0.18293837904930116,
"num_tokens": 16445829.0,
"step": 7185
},
{
"entropy": 5.5745384216308596,
"epoch": 0.7107552392249901,
"grad_norm": 0.91796875,
"learning_rate": 0.000495773052917559,
"loss": 5.3114,
"mean_token_accuracy": 0.19844848811626434,
"num_tokens": 16456352.0,
"step": 7190
},
{
"entropy": 5.635031843185425,
"epoch": 0.7112495057334915,
"grad_norm": 0.9921875,
"learning_rate": 0.0004957662418934309,
"loss": 5.3496,
"mean_token_accuracy": 0.18821563422679902,
"num_tokens": 16467542.0,
"step": 7195
},
{
"entropy": 5.602718019485474,
"epoch": 0.7117437722419929,
"grad_norm": 0.8828125,
"learning_rate": 0.0004957594254384189,
"loss": 5.2913,
"mean_token_accuracy": 0.20144389420747758,
"num_tokens": 16478614.0,
"step": 7200
},
{
"entropy": 5.653006887435913,
"epoch": 0.7122380387504943,
"grad_norm": 0.93359375,
"learning_rate": 0.0004957526035526908,
"loss": 5.3614,
"mean_token_accuracy": 0.18989845663309096,
"num_tokens": 16489929.0,
"step": 7205
},
{
"entropy": 5.755248165130615,
"epoch": 0.7127323052589957,
"grad_norm": 0.88671875,
"learning_rate": 0.0004957457762364146,
"loss": 5.4849,
"mean_token_accuracy": 0.17849240005016326,
"num_tokens": 16502532.0,
"step": 7210
},
{
"entropy": 5.733708286285401,
"epoch": 0.713226571767497,
"grad_norm": 0.91015625,
"learning_rate": 0.0004957389434897578,
"loss": 5.5021,
"mean_token_accuracy": 0.18215182721614837,
"num_tokens": 16513805.0,
"step": 7215
},
{
"entropy": 5.694751405715943,
"epoch": 0.7137208382759984,
"grad_norm": 0.96484375,
"learning_rate": 0.0004957321053128889,
"loss": 5.4867,
"mean_token_accuracy": 0.1830164983868599,
"num_tokens": 16525189.0,
"step": 7220
},
{
"entropy": 5.696434402465821,
"epoch": 0.7142151047844998,
"grad_norm": 0.95703125,
"learning_rate": 0.0004957252617059759,
"loss": 5.471,
"mean_token_accuracy": 0.1788410171866417,
"num_tokens": 16538246.0,
"step": 7225
},
{
"entropy": 5.734514427185059,
"epoch": 0.7147093712930012,
"grad_norm": 0.921875,
"learning_rate": 0.0004957184126691873,
"loss": 5.4521,
"mean_token_accuracy": 0.1796474426984787,
"num_tokens": 16550036.0,
"step": 7230
},
{
"entropy": 5.660345840454101,
"epoch": 0.7152036378015025,
"grad_norm": 0.91015625,
"learning_rate": 0.0004957115582026915,
"loss": 5.3906,
"mean_token_accuracy": 0.19144424200057983,
"num_tokens": 16562262.0,
"step": 7235
},
{
"entropy": 5.669035863876343,
"epoch": 0.7156979043100039,
"grad_norm": 0.859375,
"learning_rate": 0.000495704698306657,
"loss": 5.3661,
"mean_token_accuracy": 0.19261749237775802,
"num_tokens": 16574213.0,
"step": 7240
},
{
"entropy": 5.683400344848633,
"epoch": 0.7161921708185054,
"grad_norm": 0.93359375,
"learning_rate": 0.0004956978329812528,
"loss": 5.4441,
"mean_token_accuracy": 0.18806204646825792,
"num_tokens": 16586063.0,
"step": 7245
},
{
"entropy": 5.686618185043335,
"epoch": 0.7166864373270068,
"grad_norm": 0.890625,
"learning_rate": 0.0004956909622266476,
"loss": 5.4542,
"mean_token_accuracy": 0.17914970964193344,
"num_tokens": 16598014.0,
"step": 7250
},
{
"entropy": 5.707431745529175,
"epoch": 0.7171807038355081,
"grad_norm": 0.9140625,
"learning_rate": 0.0004956840860430104,
"loss": 5.4215,
"mean_token_accuracy": 0.19171763509511947,
"num_tokens": 16609128.0,
"step": 7255
},
{
"entropy": 5.623238468170166,
"epoch": 0.7176749703440095,
"grad_norm": 0.91796875,
"learning_rate": 0.0004956772044305106,
"loss": 5.3419,
"mean_token_accuracy": 0.19043963104486467,
"num_tokens": 16620095.0,
"step": 7260
},
{
"entropy": 5.6551319599151615,
"epoch": 0.7181692368525109,
"grad_norm": 0.9375,
"learning_rate": 0.0004956703173893173,
"loss": 5.3592,
"mean_token_accuracy": 0.18289692848920822,
"num_tokens": 16630919.0,
"step": 7265
},
{
"entropy": 5.679590797424316,
"epoch": 0.7186635033610123,
"grad_norm": 1.0234375,
"learning_rate": 0.0004956634249196001,
"loss": 5.4544,
"mean_token_accuracy": 0.1806553363800049,
"num_tokens": 16642111.0,
"step": 7270
},
{
"entropy": 5.74385437965393,
"epoch": 0.7191577698695136,
"grad_norm": 0.87109375,
"learning_rate": 0.0004956565270215282,
"loss": 5.4515,
"mean_token_accuracy": 0.17933779656887056,
"num_tokens": 16654805.0,
"step": 7275
},
{
"entropy": 5.736494064331055,
"epoch": 0.719652036378015,
"grad_norm": 1.0078125,
"learning_rate": 0.0004956496236952716,
"loss": 5.4844,
"mean_token_accuracy": 0.17921999394893645,
"num_tokens": 16664735.0,
"step": 7280
},
{
"entropy": 5.67899374961853,
"epoch": 0.7201463028865164,
"grad_norm": 1.0078125,
"learning_rate": 0.000495642714941,
"loss": 5.3381,
"mean_token_accuracy": 0.1852945253252983,
"num_tokens": 16676163.0,
"step": 7285
},
{
"entropy": 5.774364233016968,
"epoch": 0.7206405693950177,
"grad_norm": 0.93359375,
"learning_rate": 0.0004956358007588834,
"loss": 5.5386,
"mean_token_accuracy": 0.17558194547891617,
"num_tokens": 16689683.0,
"step": 7290
},
{
"entropy": 5.756363868713379,
"epoch": 0.7211348359035191,
"grad_norm": 0.8515625,
"learning_rate": 0.0004956288811490919,
"loss": 5.4768,
"mean_token_accuracy": 0.17679381519556045,
"num_tokens": 16702193.0,
"step": 7295
},
{
"entropy": 5.596215677261353,
"epoch": 0.7216291024120206,
"grad_norm": 0.85546875,
"learning_rate": 0.0004956219561117955,
"loss": 5.2711,
"mean_token_accuracy": 0.20026921778917312,
"num_tokens": 16713489.0,
"step": 7300
},
{
"entropy": 5.609003162384033,
"epoch": 0.722123368920522,
"grad_norm": 0.9296875,
"learning_rate": 0.000495615025647165,
"loss": 5.3997,
"mean_token_accuracy": 0.18067056834697723,
"num_tokens": 16725103.0,
"step": 7305
},
{
"entropy": 5.646150684356689,
"epoch": 0.7226176354290234,
"grad_norm": 0.85546875,
"learning_rate": 0.0004956080897553705,
"loss": 5.389,
"mean_token_accuracy": 0.18291271924972535,
"num_tokens": 16737056.0,
"step": 7310
},
{
"entropy": 5.66698989868164,
"epoch": 0.7231119019375247,
"grad_norm": 0.91015625,
"learning_rate": 0.0004956011484365829,
"loss": 5.3352,
"mean_token_accuracy": 0.19922607094049455,
"num_tokens": 16747347.0,
"step": 7315
},
{
"entropy": 5.63881483078003,
"epoch": 0.7236061684460261,
"grad_norm": 0.96484375,
"learning_rate": 0.0004955942016909728,
"loss": 5.3472,
"mean_token_accuracy": 0.19549122154712678,
"num_tokens": 16758738.0,
"step": 7320
},
{
"entropy": 5.659709072113037,
"epoch": 0.7241004349545275,
"grad_norm": 0.97265625,
"learning_rate": 0.000495587249518711,
"loss": 5.403,
"mean_token_accuracy": 0.1806495487689972,
"num_tokens": 16770867.0,
"step": 7325
},
{
"entropy": 5.689478921890259,
"epoch": 0.7245947014630288,
"grad_norm": 0.95703125,
"learning_rate": 0.0004955802919199687,
"loss": 5.3488,
"mean_token_accuracy": 0.1905299812555313,
"num_tokens": 16781537.0,
"step": 7330
},
{
"entropy": 5.677388429641724,
"epoch": 0.7250889679715302,
"grad_norm": 0.90234375,
"learning_rate": 0.000495573328894917,
"loss": 5.3889,
"mean_token_accuracy": 0.184035524725914,
"num_tokens": 16792250.0,
"step": 7335
},
{
"entropy": 5.620421934127807,
"epoch": 0.7255832344800316,
"grad_norm": 0.91015625,
"learning_rate": 0.0004955663604437272,
"loss": 5.3585,
"mean_token_accuracy": 0.1843078002333641,
"num_tokens": 16803673.0,
"step": 7340
},
{
"entropy": 5.6805750846862795,
"epoch": 0.726077500988533,
"grad_norm": 0.8828125,
"learning_rate": 0.0004955593865665707,
"loss": 5.4527,
"mean_token_accuracy": 0.18508365750312805,
"num_tokens": 16815484.0,
"step": 7345
},
{
"entropy": 5.691528081893921,
"epoch": 0.7265717674970344,
"grad_norm": 0.8671875,
"learning_rate": 0.000495552407263619,
"loss": 5.4137,
"mean_token_accuracy": 0.18326476812362671,
"num_tokens": 16827097.0,
"step": 7350
},
{
"entropy": 5.721248006820678,
"epoch": 0.7270660340055358,
"grad_norm": 0.90234375,
"learning_rate": 0.0004955454225350441,
"loss": 5.3772,
"mean_token_accuracy": 0.1852470189332962,
"num_tokens": 16838235.0,
"step": 7355
},
{
"entropy": 5.6892084121704105,
"epoch": 0.7275603005140372,
"grad_norm": 1.0,
"learning_rate": 0.0004955384323810174,
"loss": 5.3945,
"mean_token_accuracy": 0.18810428380966188,
"num_tokens": 16847831.0,
"step": 7360
},
{
"entropy": 5.702212142944336,
"epoch": 0.7280545670225386,
"grad_norm": 0.9453125,
"learning_rate": 0.000495531436801711,
"loss": 5.3961,
"mean_token_accuracy": 0.18493885099887847,
"num_tokens": 16859607.0,
"step": 7365
},
{
"entropy": 5.753056144714355,
"epoch": 0.7285488335310399,
"grad_norm": 1.0,
"learning_rate": 0.0004955244357972972,
"loss": 5.4574,
"mean_token_accuracy": 0.17905111014842987,
"num_tokens": 16869787.0,
"step": 7370
},
{
"entropy": 5.626278877258301,
"epoch": 0.7290431000395413,
"grad_norm": 0.98828125,
"learning_rate": 0.000495517429367948,
"loss": 5.322,
"mean_token_accuracy": 0.19048488289117813,
"num_tokens": 16880819.0,
"step": 7375
},
{
"entropy": 5.71539158821106,
"epoch": 0.7295373665480427,
"grad_norm": 1.0,
"learning_rate": 0.0004955104175138358,
"loss": 5.4505,
"mean_token_accuracy": 0.18184176683425904,
"num_tokens": 16893247.0,
"step": 7380
},
{
"entropy": 5.689558982849121,
"epoch": 0.7300316330565441,
"grad_norm": 0.96484375,
"learning_rate": 0.0004955034002351332,
"loss": 5.3792,
"mean_token_accuracy": 0.18559390902519227,
"num_tokens": 16904469.0,
"step": 7385
},
{
"entropy": 5.6558443069458,
"epoch": 0.7305258995650454,
"grad_norm": 0.84765625,
"learning_rate": 0.0004954963775320127,
"loss": 5.5246,
"mean_token_accuracy": 0.17719829380512236,
"num_tokens": 16918433.0,
"step": 7390
},
{
"entropy": 5.694541168212891,
"epoch": 0.7310201660735468,
"grad_norm": 0.90234375,
"learning_rate": 0.000495489349404647,
"loss": 5.4364,
"mean_token_accuracy": 0.17820830643177032,
"num_tokens": 16930905.0,
"step": 7395
},
{
"entropy": 5.734336233139038,
"epoch": 0.7315144325820483,
"grad_norm": 0.88671875,
"learning_rate": 0.0004954823158532093,
"loss": 5.4629,
"mean_token_accuracy": 0.17735820710659028,
"num_tokens": 16943461.0,
"step": 7400
},
{
"entropy": 5.63409914970398,
"epoch": 0.7320086990905497,
"grad_norm": 0.90234375,
"learning_rate": 0.0004954752768778723,
"loss": 5.4166,
"mean_token_accuracy": 0.18278694599866868,
"num_tokens": 16955539.0,
"step": 7405
},
{
"entropy": 5.670486307144165,
"epoch": 0.732502965599051,
"grad_norm": 0.8984375,
"learning_rate": 0.0004954682324788093,
"loss": 5.3925,
"mean_token_accuracy": 0.17866645008325577,
"num_tokens": 16967508.0,
"step": 7410
},
{
"entropy": 5.652829694747925,
"epoch": 0.7329972321075524,
"grad_norm": 0.9296875,
"learning_rate": 0.0004954611826561935,
"loss": 5.3545,
"mean_token_accuracy": 0.18973411619663239,
"num_tokens": 16979033.0,
"step": 7415
},
{
"entropy": 5.6304679870605465,
"epoch": 0.7334914986160538,
"grad_norm": 0.80859375,
"learning_rate": 0.0004954541274101985,
"loss": 5.3362,
"mean_token_accuracy": 0.19321956485509872,
"num_tokens": 16991874.0,
"step": 7420
},
{
"entropy": 5.73438401222229,
"epoch": 0.7339857651245552,
"grad_norm": 0.97265625,
"learning_rate": 0.0004954470667409978,
"loss": 5.5023,
"mean_token_accuracy": 0.18036859035491942,
"num_tokens": 17003160.0,
"step": 7425
},
{
"entropy": 5.654939031600952,
"epoch": 0.7344800316330565,
"grad_norm": 0.890625,
"learning_rate": 0.000495440000648765,
"loss": 5.3705,
"mean_token_accuracy": 0.18651044368743896,
"num_tokens": 17014043.0,
"step": 7430
},
{
"entropy": 5.679742813110352,
"epoch": 0.7349742981415579,
"grad_norm": 0.91015625,
"learning_rate": 0.0004954329291336739,
"loss": 5.3422,
"mean_token_accuracy": 0.1880825787782669,
"num_tokens": 17025117.0,
"step": 7435
},
{
"entropy": 5.660619735717773,
"epoch": 0.7354685646500593,
"grad_norm": 0.94140625,
"learning_rate": 0.0004954258521958988,
"loss": 5.3809,
"mean_token_accuracy": 0.18671616166830063,
"num_tokens": 17037366.0,
"step": 7440
},
{
"entropy": 5.70739483833313,
"epoch": 0.7359628311585606,
"grad_norm": 0.83984375,
"learning_rate": 0.0004954187698356133,
"loss": 5.4256,
"mean_token_accuracy": 0.18276374340057372,
"num_tokens": 17050683.0,
"step": 7445
},
{
"entropy": 5.5868124008178714,
"epoch": 0.7364570976670621,
"grad_norm": 0.859375,
"learning_rate": 0.0004954116820529921,
"loss": 5.3172,
"mean_token_accuracy": 0.19585615694522857,
"num_tokens": 17062403.0,
"step": 7450
},
{
"entropy": 5.65182785987854,
"epoch": 0.7369513641755635,
"grad_norm": 0.94140625,
"learning_rate": 0.0004954045888482091,
"loss": 5.3348,
"mean_token_accuracy": 0.19099604934453965,
"num_tokens": 17072688.0,
"step": 7455
},
{
"entropy": 5.6631995677948,
"epoch": 0.7374456306840649,
"grad_norm": 0.98828125,
"learning_rate": 0.0004953974902214391,
"loss": 5.3828,
"mean_token_accuracy": 0.1912774920463562,
"num_tokens": 17084118.0,
"step": 7460
},
{
"entropy": 5.664176034927368,
"epoch": 0.7379398971925663,
"grad_norm": 0.94921875,
"learning_rate": 0.0004953903861728567,
"loss": 5.4256,
"mean_token_accuracy": 0.18769233971834182,
"num_tokens": 17095600.0,
"step": 7465
},
{
"entropy": 5.803557062149048,
"epoch": 0.7384341637010676,
"grad_norm": 0.89453125,
"learning_rate": 0.0004953832767026365,
"loss": 5.5317,
"mean_token_accuracy": 0.18064996600151062,
"num_tokens": 17107934.0,
"step": 7470
},
{
"entropy": 5.742346906661988,
"epoch": 0.738928430209569,
"grad_norm": 0.98046875,
"learning_rate": 0.0004953761618109535,
"loss": 5.45,
"mean_token_accuracy": 0.18403366953134537,
"num_tokens": 17119578.0,
"step": 7475
},
{
"entropy": 5.659234714508057,
"epoch": 0.7394226967180704,
"grad_norm": 1.03125,
"learning_rate": 0.0004953690414979828,
"loss": 5.4486,
"mean_token_accuracy": 0.17846433073282242,
"num_tokens": 17129928.0,
"step": 7480
},
{
"entropy": 5.692962598800659,
"epoch": 0.7399169632265717,
"grad_norm": 0.9375,
"learning_rate": 0.0004953619157638995,
"loss": 5.4277,
"mean_token_accuracy": 0.18141815960407257,
"num_tokens": 17141149.0,
"step": 7485
},
{
"entropy": 5.634675598144531,
"epoch": 0.7404112297350731,
"grad_norm": 0.9765625,
"learning_rate": 0.0004953547846088789,
"loss": 5.3486,
"mean_token_accuracy": 0.18541015982627868,
"num_tokens": 17152577.0,
"step": 7490
},
{
"entropy": 5.642905759811401,
"epoch": 0.7409054962435745,
"grad_norm": 0.95703125,
"learning_rate": 0.0004953476480330964,
"loss": 5.3476,
"mean_token_accuracy": 0.18487662076950073,
"num_tokens": 17163997.0,
"step": 7495
},
{
"entropy": 5.635993385314942,
"epoch": 0.741399762752076,
"grad_norm": 0.83984375,
"learning_rate": 0.0004953405060367276,
"loss": 5.3151,
"mean_token_accuracy": 0.1913245663046837,
"num_tokens": 17175317.0,
"step": 7500
},
{
"entropy": 5.558791065216065,
"epoch": 0.7418940292605773,
"grad_norm": 0.92578125,
"learning_rate": 0.0004953333586199481,
"loss": 5.3976,
"mean_token_accuracy": 0.1898777738213539,
"num_tokens": 17185615.0,
"step": 7505
},
{
"entropy": 5.712562322616577,
"epoch": 0.7423882957690787,
"grad_norm": 0.90234375,
"learning_rate": 0.0004953262057829338,
"loss": 5.4027,
"mean_token_accuracy": 0.17779784202575682,
"num_tokens": 17198160.0,
"step": 7510
},
{
"entropy": 5.71424126625061,
"epoch": 0.7428825622775801,
"grad_norm": 0.9765625,
"learning_rate": 0.0004953190475258607,
"loss": 5.3778,
"mean_token_accuracy": 0.1871979668736458,
"num_tokens": 17209652.0,
"step": 7515
},
{
"entropy": 5.681139039993286,
"epoch": 0.7433768287860815,
"grad_norm": 0.9921875,
"learning_rate": 0.0004953118838489049,
"loss": 5.3829,
"mean_token_accuracy": 0.183564230799675,
"num_tokens": 17220832.0,
"step": 7520
},
{
"entropy": 5.644816732406616,
"epoch": 0.7438710952945828,
"grad_norm": 0.9765625,
"learning_rate": 0.0004953047147522424,
"loss": 5.3348,
"mean_token_accuracy": 0.1867297887802124,
"num_tokens": 17231952.0,
"step": 7525
},
{
"entropy": 5.5916526317596436,
"epoch": 0.7443653618030842,
"grad_norm": 0.96484375,
"learning_rate": 0.0004952975402360499,
"loss": 5.3325,
"mean_token_accuracy": 0.19042591601610184,
"num_tokens": 17242705.0,
"step": 7530
},
{
"entropy": 5.7398519039154055,
"epoch": 0.7448596283115856,
"grad_norm": 0.97265625,
"learning_rate": 0.0004952903603005037,
"loss": 5.4221,
"mean_token_accuracy": 0.17742991149425508,
"num_tokens": 17253946.0,
"step": 7535
},
{
"entropy": 5.707596158981323,
"epoch": 0.745353894820087,
"grad_norm": 0.94921875,
"learning_rate": 0.0004952831749457804,
"loss": 5.452,
"mean_token_accuracy": 0.19042180627584457,
"num_tokens": 17265732.0,
"step": 7540
},
{
"entropy": 5.643588733673096,
"epoch": 0.7458481613285883,
"grad_norm": 0.9375,
"learning_rate": 0.0004952759841720569,
"loss": 5.3643,
"mean_token_accuracy": 0.19260530769824982,
"num_tokens": 17277154.0,
"step": 7545
},
{
"entropy": 5.69439206123352,
"epoch": 0.7463424278370897,
"grad_norm": 0.98828125,
"learning_rate": 0.0004952687879795101,
"loss": 5.4191,
"mean_token_accuracy": 0.1851677566766739,
"num_tokens": 17288179.0,
"step": 7550
},
{
"entropy": 5.6562920093536375,
"epoch": 0.7468366943455912,
"grad_norm": 0.9375,
"learning_rate": 0.0004952615863683167,
"loss": 5.3885,
"mean_token_accuracy": 0.189335697889328,
"num_tokens": 17299435.0,
"step": 7555
},
{
"entropy": 5.525950908660889,
"epoch": 0.7473309608540926,
"grad_norm": 0.94140625,
"learning_rate": 0.0004952543793386543,
"loss": 5.224,
"mean_token_accuracy": 0.2058229401707649,
"num_tokens": 17309442.0,
"step": 7560
},
{
"entropy": 5.715339136123657,
"epoch": 0.7478252273625939,
"grad_norm": 0.95703125,
"learning_rate": 0.0004952471668906999,
"loss": 5.4208,
"mean_token_accuracy": 0.1828790307044983,
"num_tokens": 17322072.0,
"step": 7565
},
{
"entropy": 5.651614856719971,
"epoch": 0.7483194938710953,
"grad_norm": 0.921875,
"learning_rate": 0.0004952399490246311,
"loss": 5.3501,
"mean_token_accuracy": 0.18749946355819702,
"num_tokens": 17333458.0,
"step": 7570
},
{
"entropy": 5.626357316970825,
"epoch": 0.7488137603795967,
"grad_norm": 0.8671875,
"learning_rate": 0.0004952327257406253,
"loss": 5.3151,
"mean_token_accuracy": 0.18851567655801774,
"num_tokens": 17344319.0,
"step": 7575
},
{
"entropy": 5.7104216575622555,
"epoch": 0.749308026888098,
"grad_norm": 0.87109375,
"learning_rate": 0.0004952254970388603,
"loss": 5.4255,
"mean_token_accuracy": 0.18660738319158554,
"num_tokens": 17357016.0,
"step": 7580
},
{
"entropy": 5.590426397323609,
"epoch": 0.7498022933965994,
"grad_norm": 0.87890625,
"learning_rate": 0.000495218262919514,
"loss": 5.3171,
"mean_token_accuracy": 0.19457150101661683,
"num_tokens": 17368574.0,
"step": 7585
},
{
"entropy": 5.589792156219483,
"epoch": 0.7502965599051008,
"grad_norm": 0.9453125,
"learning_rate": 0.0004952110233827642,
"loss": 5.2671,
"mean_token_accuracy": 0.19795868545770645,
"num_tokens": 17379858.0,
"step": 7590
},
{
"entropy": 5.693734884262085,
"epoch": 0.7507908264136022,
"grad_norm": 0.87890625,
"learning_rate": 0.0004952037784287891,
"loss": 5.4276,
"mean_token_accuracy": 0.18673142343759536,
"num_tokens": 17391543.0,
"step": 7595
},
{
"entropy": 5.684725666046143,
"epoch": 0.7512850929221035,
"grad_norm": 0.91015625,
"learning_rate": 0.0004951965280577668,
"loss": 5.3647,
"mean_token_accuracy": 0.18937657475471498,
"num_tokens": 17403193.0,
"step": 7600
},
{
"entropy": 5.697777366638183,
"epoch": 0.751779359430605,
"grad_norm": 0.9296875,
"learning_rate": 0.0004951892722698759,
"loss": 5.4876,
"mean_token_accuracy": 0.1847672641277313,
"num_tokens": 17415866.0,
"step": 7605
},
{
"entropy": 5.613073396682739,
"epoch": 0.7522736259391064,
"grad_norm": 0.89453125,
"learning_rate": 0.0004951820110652946,
"loss": 5.3315,
"mean_token_accuracy": 0.18801401406526566,
"num_tokens": 17427880.0,
"step": 7610
},
{
"entropy": 5.650252342224121,
"epoch": 0.7527678924476078,
"grad_norm": 0.93359375,
"learning_rate": 0.0004951747444442017,
"loss": 5.3471,
"mean_token_accuracy": 0.18859250843524933,
"num_tokens": 17440295.0,
"step": 7615
},
{
"entropy": 5.654036188125611,
"epoch": 0.7532621589561092,
"grad_norm": 0.9375,
"learning_rate": 0.0004951674724067759,
"loss": 5.3953,
"mean_token_accuracy": 0.18445403277873992,
"num_tokens": 17451636.0,
"step": 7620
},
{
"entropy": 5.690109729766846,
"epoch": 0.7537564254646105,
"grad_norm": 1.0078125,
"learning_rate": 0.0004951601949531962,
"loss": 5.4466,
"mean_token_accuracy": 0.18781063109636306,
"num_tokens": 17464376.0,
"step": 7625
},
{
"entropy": 5.661364555358887,
"epoch": 0.7542506919731119,
"grad_norm": 0.890625,
"learning_rate": 0.0004951529120836416,
"loss": 5.3055,
"mean_token_accuracy": 0.1877681091427803,
"num_tokens": 17475098.0,
"step": 7630
},
{
"entropy": 5.559615898132324,
"epoch": 0.7547449584816133,
"grad_norm": 0.89453125,
"learning_rate": 0.0004951456237982911,
"loss": 5.3135,
"mean_token_accuracy": 0.1875964432954788,
"num_tokens": 17486176.0,
"step": 7635
},
{
"entropy": 5.614330673217774,
"epoch": 0.7552392249901146,
"grad_norm": 0.90625,
"learning_rate": 0.0004951383300973242,
"loss": 5.3689,
"mean_token_accuracy": 0.19070686846971513,
"num_tokens": 17498386.0,
"step": 7640
},
{
"entropy": 5.6752747058868405,
"epoch": 0.755733491498616,
"grad_norm": 0.98828125,
"learning_rate": 0.0004951310309809201,
"loss": 5.4282,
"mean_token_accuracy": 0.18466150015592575,
"num_tokens": 17510409.0,
"step": 7645
},
{
"entropy": 5.647143459320068,
"epoch": 0.7562277580071174,
"grad_norm": 1.03125,
"learning_rate": 0.0004951237264492585,
"loss": 5.3467,
"mean_token_accuracy": 0.18571873009204865,
"num_tokens": 17521803.0,
"step": 7650
},
{
"entropy": 5.6209554195404055,
"epoch": 0.7567220245156189,
"grad_norm": 0.953125,
"learning_rate": 0.0004951164165025192,
"loss": 5.4158,
"mean_token_accuracy": 0.18765482753515245,
"num_tokens": 17532974.0,
"step": 7655
},
{
"entropy": 5.684657382965088,
"epoch": 0.7572162910241202,
"grad_norm": 0.94140625,
"learning_rate": 0.0004951091011408818,
"loss": 5.3798,
"mean_token_accuracy": 0.1874500870704651,
"num_tokens": 17544309.0,
"step": 7660
},
{
"entropy": 5.641847133636475,
"epoch": 0.7577105575326216,
"grad_norm": 0.91796875,
"learning_rate": 0.0004951017803645264,
"loss": 5.3099,
"mean_token_accuracy": 0.19300690591335296,
"num_tokens": 17555395.0,
"step": 7665
},
{
"entropy": 5.623783922195434,
"epoch": 0.758204824041123,
"grad_norm": 1.0234375,
"learning_rate": 0.000495094454173633,
"loss": 5.2942,
"mean_token_accuracy": 0.18821680992841722,
"num_tokens": 17567434.0,
"step": 7670
},
{
"entropy": 5.705420303344726,
"epoch": 0.7586990905496244,
"grad_norm": 0.9765625,
"learning_rate": 0.0004950871225683819,
"loss": 5.4313,
"mean_token_accuracy": 0.18236252963542937,
"num_tokens": 17578381.0,
"step": 7675
},
{
"entropy": 5.634431171417236,
"epoch": 0.7591933570581257,
"grad_norm": 0.84765625,
"learning_rate": 0.0004950797855489535,
"loss": 5.3801,
"mean_token_accuracy": 0.18885243833065032,
"num_tokens": 17589732.0,
"step": 7680
},
{
"entropy": 5.689999294281006,
"epoch": 0.7596876235666271,
"grad_norm": 0.9609375,
"learning_rate": 0.0004950724431155283,
"loss": 5.3242,
"mean_token_accuracy": 0.18844882100820542,
"num_tokens": 17601092.0,
"step": 7685
},
{
"entropy": 5.6386909008026125,
"epoch": 0.7601818900751285,
"grad_norm": 0.8984375,
"learning_rate": 0.0004950650952682867,
"loss": 5.4365,
"mean_token_accuracy": 0.18836773484945296,
"num_tokens": 17613268.0,
"step": 7690
},
{
"entropy": 5.663677787780761,
"epoch": 0.7606761565836299,
"grad_norm": 0.94140625,
"learning_rate": 0.0004950577420074098,
"loss": 5.3098,
"mean_token_accuracy": 0.18444576859474182,
"num_tokens": 17623500.0,
"step": 7695
},
{
"entropy": 5.686369800567627,
"epoch": 0.7611704230921312,
"grad_norm": 0.92578125,
"learning_rate": 0.0004950503833330781,
"loss": 5.5344,
"mean_token_accuracy": 0.17894016355276107,
"num_tokens": 17636151.0,
"step": 7700
},
{
"entropy": 5.675409030914307,
"epoch": 0.7616646896006327,
"grad_norm": 0.94140625,
"learning_rate": 0.0004950430192454729,
"loss": 5.4064,
"mean_token_accuracy": 0.17720431983470916,
"num_tokens": 17648392.0,
"step": 7705
},
{
"entropy": 5.672790813446045,
"epoch": 0.7621589561091341,
"grad_norm": 0.9140625,
"learning_rate": 0.0004950356497447754,
"loss": 5.3426,
"mean_token_accuracy": 0.18612737208604813,
"num_tokens": 17659777.0,
"step": 7710
},
{
"entropy": 5.694863367080688,
"epoch": 0.7626532226176355,
"grad_norm": 0.90234375,
"learning_rate": 0.0004950282748311666,
"loss": 5.4494,
"mean_token_accuracy": 0.18029696643352508,
"num_tokens": 17671986.0,
"step": 7715
},
{
"entropy": 5.656952524185181,
"epoch": 0.7631474891261368,
"grad_norm": 0.96484375,
"learning_rate": 0.0004950208945048283,
"loss": 5.3634,
"mean_token_accuracy": 0.19005147516727447,
"num_tokens": 17683740.0,
"step": 7720
},
{
"entropy": 5.652459192276001,
"epoch": 0.7636417556346382,
"grad_norm": 0.85546875,
"learning_rate": 0.0004950135087659417,
"loss": 5.4335,
"mean_token_accuracy": 0.18499477207660675,
"num_tokens": 17696034.0,
"step": 7725
},
{
"entropy": 5.661531734466553,
"epoch": 0.7641360221431396,
"grad_norm": 0.87890625,
"learning_rate": 0.0004950061176146887,
"loss": 5.3694,
"mean_token_accuracy": 0.1886142760515213,
"num_tokens": 17707783.0,
"step": 7730
},
{
"entropy": 5.632096576690674,
"epoch": 0.764630288651641,
"grad_norm": 1.0078125,
"learning_rate": 0.0004949987210512511,
"loss": 5.3312,
"mean_token_accuracy": 0.18884276300668718,
"num_tokens": 17718127.0,
"step": 7735
},
{
"entropy": 5.677348661422729,
"epoch": 0.7651245551601423,
"grad_norm": 0.9296875,
"learning_rate": 0.0004949913190758109,
"loss": 5.4147,
"mean_token_accuracy": 0.1795927718281746,
"num_tokens": 17729798.0,
"step": 7740
},
{
"entropy": 5.6840489387512205,
"epoch": 0.7656188216686437,
"grad_norm": 0.93359375,
"learning_rate": 0.0004949839116885499,
"loss": 5.3635,
"mean_token_accuracy": 0.18357911854982376,
"num_tokens": 17741989.0,
"step": 7745
},
{
"entropy": 5.648376083374023,
"epoch": 0.7661130881771451,
"grad_norm": 0.94140625,
"learning_rate": 0.0004949764988896508,
"loss": 5.4056,
"mean_token_accuracy": 0.18787432014942168,
"num_tokens": 17753737.0,
"step": 7750
},
{
"entropy": 5.596786546707153,
"epoch": 0.7666073546856464,
"grad_norm": 1.015625,
"learning_rate": 0.0004949690806792955,
"loss": 5.3047,
"mean_token_accuracy": 0.1968316003680229,
"num_tokens": 17764158.0,
"step": 7755
},
{
"entropy": 5.680300617218018,
"epoch": 0.7671016211941479,
"grad_norm": 0.9140625,
"learning_rate": 0.0004949616570576668,
"loss": 5.4165,
"mean_token_accuracy": 0.1885527640581131,
"num_tokens": 17775878.0,
"step": 7760
},
{
"entropy": 5.650046873092651,
"epoch": 0.7675958877026493,
"grad_norm": 1.0078125,
"learning_rate": 0.0004949542280249473,
"loss": 5.3231,
"mean_token_accuracy": 0.1919610396027565,
"num_tokens": 17786107.0,
"step": 7765
},
{
"entropy": 5.618096351623535,
"epoch": 0.7680901542111507,
"grad_norm": 0.90234375,
"learning_rate": 0.0004949467935813196,
"loss": 5.3806,
"mean_token_accuracy": 0.1877329871058464,
"num_tokens": 17798470.0,
"step": 7770
},
{
"entropy": 5.6560972213745115,
"epoch": 0.768584420719652,
"grad_norm": 0.80859375,
"learning_rate": 0.0004949393537269666,
"loss": 5.3893,
"mean_token_accuracy": 0.1861312359571457,
"num_tokens": 17811125.0,
"step": 7775
},
{
"entropy": 5.6340124130249025,
"epoch": 0.7690786872281534,
"grad_norm": 0.984375,
"learning_rate": 0.0004949319084620714,
"loss": 5.259,
"mean_token_accuracy": 0.19185607135295868,
"num_tokens": 17821535.0,
"step": 7780
},
{
"entropy": 5.710686779022216,
"epoch": 0.7695729537366548,
"grad_norm": 1.015625,
"learning_rate": 0.000494924457786817,
"loss": 5.503,
"mean_token_accuracy": 0.18006813228130342,
"num_tokens": 17833493.0,
"step": 7785
},
{
"entropy": 5.75515193939209,
"epoch": 0.7700672202451562,
"grad_norm": 0.91796875,
"learning_rate": 0.0004949170017013871,
"loss": 5.4566,
"mean_token_accuracy": 0.18286254703998567,
"num_tokens": 17844705.0,
"step": 7790
},
{
"entropy": 5.700857877731323,
"epoch": 0.7705614867536575,
"grad_norm": 0.9765625,
"learning_rate": 0.0004949095402059646,
"loss": 5.4522,
"mean_token_accuracy": 0.18440821766853333,
"num_tokens": 17856088.0,
"step": 7795
},
{
"entropy": 5.734563112258911,
"epoch": 0.7710557532621589,
"grad_norm": 0.953125,
"learning_rate": 0.0004949020733007334,
"loss": 5.4219,
"mean_token_accuracy": 0.1825122967362404,
"num_tokens": 17867602.0,
"step": 7800
},
{
"entropy": 5.735884046554565,
"epoch": 0.7715500197706603,
"grad_norm": 0.9140625,
"learning_rate": 0.000494894600985877,
"loss": 5.4767,
"mean_token_accuracy": 0.1801721751689911,
"num_tokens": 17880487.0,
"step": 7805
},
{
"entropy": 5.687494564056396,
"epoch": 0.7720442862791618,
"grad_norm": 0.90234375,
"learning_rate": 0.0004948871232615794,
"loss": 5.4005,
"mean_token_accuracy": 0.18796298503875733,
"num_tokens": 17891690.0,
"step": 7810
},
{
"entropy": 5.725475549697876,
"epoch": 0.7725385527876631,
"grad_norm": 1.0703125,
"learning_rate": 0.0004948796401280244,
"loss": 5.4484,
"mean_token_accuracy": 0.17923469990491867,
"num_tokens": 17903165.0,
"step": 7815
},
{
"entropy": 5.676923418045044,
"epoch": 0.7730328192961645,
"grad_norm": 0.96484375,
"learning_rate": 0.0004948721515853962,
"loss": 5.2961,
"mean_token_accuracy": 0.1925194516777992,
"num_tokens": 17914492.0,
"step": 7820
},
{
"entropy": 5.690169191360473,
"epoch": 0.7735270858046659,
"grad_norm": 0.953125,
"learning_rate": 0.0004948646576338789,
"loss": 5.3857,
"mean_token_accuracy": 0.17486454397439957,
"num_tokens": 17926231.0,
"step": 7825
},
{
"entropy": 5.604264497756958,
"epoch": 0.7740213523131673,
"grad_norm": 0.97265625,
"learning_rate": 0.000494857158273657,
"loss": 5.3394,
"mean_token_accuracy": 0.1902058318257332,
"num_tokens": 17937167.0,
"step": 7830
},
{
"entropy": 5.644855165481568,
"epoch": 0.7745156188216686,
"grad_norm": 0.8828125,
"learning_rate": 0.0004948496535049148,
"loss": 5.4276,
"mean_token_accuracy": 0.18402523845434188,
"num_tokens": 17948939.0,
"step": 7835
},
{
"entropy": 5.670464086532593,
"epoch": 0.77500988533017,
"grad_norm": 1.0078125,
"learning_rate": 0.000494842143327837,
"loss": 5.2197,
"mean_token_accuracy": 0.20334178060293198,
"num_tokens": 17959250.0,
"step": 7840
},
{
"entropy": 5.6042379379272464,
"epoch": 0.7755041518386714,
"grad_norm": 0.91796875,
"learning_rate": 0.0004948346277426084,
"loss": 5.4515,
"mean_token_accuracy": 0.18002065718173982,
"num_tokens": 17970191.0,
"step": 7845
},
{
"entropy": 5.698566579818726,
"epoch": 0.7759984183471728,
"grad_norm": 0.921875,
"learning_rate": 0.000494827106749414,
"loss": 5.3589,
"mean_token_accuracy": 0.18793852031230926,
"num_tokens": 17982114.0,
"step": 7850
},
{
"entropy": 5.799722337722779,
"epoch": 0.7764926848556741,
"grad_norm": 0.94140625,
"learning_rate": 0.0004948195803484385,
"loss": 5.4494,
"mean_token_accuracy": 0.18278124928474426,
"num_tokens": 17994066.0,
"step": 7855
},
{
"entropy": 5.620365715026855,
"epoch": 0.7769869513641756,
"grad_norm": 0.95703125,
"learning_rate": 0.0004948120485398673,
"loss": 5.3882,
"mean_token_accuracy": 0.1864301308989525,
"num_tokens": 18005406.0,
"step": 7860
},
{
"entropy": 5.635767650604248,
"epoch": 0.777481217872677,
"grad_norm": 1.0625,
"learning_rate": 0.0004948045113238856,
"loss": 5.3816,
"mean_token_accuracy": 0.1857351467013359,
"num_tokens": 18016773.0,
"step": 7865
},
{
"entropy": 5.705584764480591,
"epoch": 0.7779754843811784,
"grad_norm": 0.87109375,
"learning_rate": 0.0004947969687006787,
"loss": 5.4335,
"mean_token_accuracy": 0.18227917850017547,
"num_tokens": 18028682.0,
"step": 7870
},
{
"entropy": 5.6915607929229735,
"epoch": 0.7784697508896797,
"grad_norm": 0.90234375,
"learning_rate": 0.0004947894206704324,
"loss": 5.3974,
"mean_token_accuracy": 0.18589052706956863,
"num_tokens": 18039525.0,
"step": 7875
},
{
"entropy": 5.652793121337891,
"epoch": 0.7789640173981811,
"grad_norm": 0.9375,
"learning_rate": 0.0004947818672333323,
"loss": 5.3746,
"mean_token_accuracy": 0.18564853221178054,
"num_tokens": 18049918.0,
"step": 7880
},
{
"entropy": 5.734889125823974,
"epoch": 0.7794582839066825,
"grad_norm": 0.953125,
"learning_rate": 0.000494774308389564,
"loss": 5.4218,
"mean_token_accuracy": 0.17997895628213884,
"num_tokens": 18061019.0,
"step": 7885
},
{
"entropy": 5.667730093002319,
"epoch": 0.7799525504151839,
"grad_norm": 0.98046875,
"learning_rate": 0.0004947667441393138,
"loss": 5.3682,
"mean_token_accuracy": 0.190121528506279,
"num_tokens": 18072123.0,
"step": 7890
},
{
"entropy": 5.631020641326904,
"epoch": 0.7804468169236852,
"grad_norm": 0.9375,
"learning_rate": 0.0004947591744827674,
"loss": 5.3596,
"mean_token_accuracy": 0.18434544503688813,
"num_tokens": 18083354.0,
"step": 7895
},
{
"entropy": 5.758954620361328,
"epoch": 0.7809410834321866,
"grad_norm": 1.0234375,
"learning_rate": 0.0004947515994201113,
"loss": 5.4785,
"mean_token_accuracy": 0.1783450126647949,
"num_tokens": 18094188.0,
"step": 7900
},
{
"entropy": 5.68081693649292,
"epoch": 0.781435349940688,
"grad_norm": 1.1171875,
"learning_rate": 0.0004947440189515318,
"loss": 5.3775,
"mean_token_accuracy": 0.18587932586669922,
"num_tokens": 18106198.0,
"step": 7905
},
{
"entropy": 5.668081760406494,
"epoch": 0.7819296164491895,
"grad_norm": 0.96875,
"learning_rate": 0.0004947364330772154,
"loss": 5.3331,
"mean_token_accuracy": 0.1908632695674896,
"num_tokens": 18117137.0,
"step": 7910
},
{
"entropy": 5.6121697425842285,
"epoch": 0.7824238829576908,
"grad_norm": 1.0703125,
"learning_rate": 0.0004947288417973486,
"loss": 5.4147,
"mean_token_accuracy": 0.1813672661781311,
"num_tokens": 18127341.0,
"step": 7915
},
{
"entropy": 5.680874872207641,
"epoch": 0.7829181494661922,
"grad_norm": 1.0078125,
"learning_rate": 0.0004947212451121182,
"loss": 5.339,
"mean_token_accuracy": 0.18717844635248185,
"num_tokens": 18138333.0,
"step": 7920
},
{
"entropy": 5.736912250518799,
"epoch": 0.7834124159746936,
"grad_norm": 0.9140625,
"learning_rate": 0.000494713643021711,
"loss": 5.6047,
"mean_token_accuracy": 0.1727058097720146,
"num_tokens": 18150961.0,
"step": 7925
},
{
"entropy": 5.747749853134155,
"epoch": 0.783906682483195,
"grad_norm": 0.82421875,
"learning_rate": 0.0004947060355263142,
"loss": 5.3928,
"mean_token_accuracy": 0.1867763638496399,
"num_tokens": 18162604.0,
"step": 7930
},
{
"entropy": 5.650860834121704,
"epoch": 0.7844009489916963,
"grad_norm": 0.90234375,
"learning_rate": 0.0004946984226261149,
"loss": 5.4235,
"mean_token_accuracy": 0.186197167634964,
"num_tokens": 18174202.0,
"step": 7935
},
{
"entropy": 5.709315061569214,
"epoch": 0.7848952155001977,
"grad_norm": 1.0234375,
"learning_rate": 0.0004946908043213001,
"loss": 5.424,
"mean_token_accuracy": 0.18151017129421235,
"num_tokens": 18185942.0,
"step": 7940
},
{
"entropy": 5.680483102798462,
"epoch": 0.7853894820086991,
"grad_norm": 0.94140625,
"learning_rate": 0.0004946831806120576,
"loss": 5.3387,
"mean_token_accuracy": 0.1878525972366333,
"num_tokens": 18198897.0,
"step": 7945
},
{
"entropy": 5.636187314987183,
"epoch": 0.7858837485172004,
"grad_norm": 0.921875,
"learning_rate": 0.0004946755514985748,
"loss": 5.2777,
"mean_token_accuracy": 0.19495179057121276,
"num_tokens": 18211439.0,
"step": 7950
},
{
"entropy": 5.574185180664062,
"epoch": 0.7863780150257018,
"grad_norm": 0.9765625,
"learning_rate": 0.0004946679169810392,
"loss": 5.313,
"mean_token_accuracy": 0.19399549663066865,
"num_tokens": 18221864.0,
"step": 7955
},
{
"entropy": 5.661316442489624,
"epoch": 0.7868722815342032,
"grad_norm": 0.921875,
"learning_rate": 0.000494660277059639,
"loss": 5.4436,
"mean_token_accuracy": 0.1834837704896927,
"num_tokens": 18234005.0,
"step": 7960
},
{
"entropy": 5.641396713256836,
"epoch": 0.7873665480427047,
"grad_norm": 1.0078125,
"learning_rate": 0.0004946526317345617,
"loss": 5.3167,
"mean_token_accuracy": 0.191530941426754,
"num_tokens": 18246804.0,
"step": 7965
},
{
"entropy": 5.644958782196045,
"epoch": 0.787860814551206,
"grad_norm": 0.875,
"learning_rate": 0.0004946449810059957,
"loss": 5.3434,
"mean_token_accuracy": 0.19076861441135406,
"num_tokens": 18259192.0,
"step": 7970
},
{
"entropy": 5.679888200759888,
"epoch": 0.7883550810597074,
"grad_norm": 0.81640625,
"learning_rate": 0.0004946373248741289,
"loss": 5.3747,
"mean_token_accuracy": 0.18994068652391433,
"num_tokens": 18271073.0,
"step": 7975
},
{
"entropy": 5.651530361175537,
"epoch": 0.7888493475682088,
"grad_norm": 1.0234375,
"learning_rate": 0.00049462966333915,
"loss": 5.3575,
"mean_token_accuracy": 0.1886462688446045,
"num_tokens": 18281650.0,
"step": 7980
},
{
"entropy": 5.665188026428223,
"epoch": 0.7893436140767102,
"grad_norm": 0.98828125,
"learning_rate": 0.0004946219964012473,
"loss": 5.3348,
"mean_token_accuracy": 0.18682026714086533,
"num_tokens": 18292217.0,
"step": 7985
},
{
"entropy": 5.645196723937988,
"epoch": 0.7898378805852115,
"grad_norm": 0.82421875,
"learning_rate": 0.0004946143240606095,
"loss": 5.3259,
"mean_token_accuracy": 0.19166507869958876,
"num_tokens": 18305210.0,
"step": 7990
},
{
"entropy": 5.6403951168060305,
"epoch": 0.7903321470937129,
"grad_norm": 0.94140625,
"learning_rate": 0.0004946066463174251,
"loss": 5.2925,
"mean_token_accuracy": 0.19262539297342302,
"num_tokens": 18316418.0,
"step": 7995
},
{
"entropy": 5.558445596694947,
"epoch": 0.7908264136022143,
"grad_norm": 0.93359375,
"learning_rate": 0.0004945989631718831,
"loss": 5.361,
"mean_token_accuracy": 0.19246632903814315,
"num_tokens": 18328637.0,
"step": 8000
},
{
"entropy": 5.647032308578491,
"epoch": 0.7913206801107157,
"grad_norm": 0.83984375,
"learning_rate": 0.0004945912746241727,
"loss": 5.2746,
"mean_token_accuracy": 0.1939465433359146,
"num_tokens": 18340292.0,
"step": 8005
},
{
"entropy": 5.6033656120300295,
"epoch": 0.791814946619217,
"grad_norm": 0.95703125,
"learning_rate": 0.0004945835806744828,
"loss": 5.3439,
"mean_token_accuracy": 0.19372644424438476,
"num_tokens": 18350784.0,
"step": 8010
},
{
"entropy": 5.702573823928833,
"epoch": 0.7923092131277185,
"grad_norm": 0.9140625,
"learning_rate": 0.0004945758813230028,
"loss": 5.4158,
"mean_token_accuracy": 0.182921838760376,
"num_tokens": 18363147.0,
"step": 8015
},
{
"entropy": 5.706137704849243,
"epoch": 0.7928034796362199,
"grad_norm": 0.9375,
"learning_rate": 0.000494568176569922,
"loss": 5.3792,
"mean_token_accuracy": 0.18668553978204727,
"num_tokens": 18374152.0,
"step": 8020
},
{
"entropy": 5.62548828125,
"epoch": 0.7932977461447213,
"grad_norm": 0.91015625,
"learning_rate": 0.00049456046641543,
"loss": 5.2765,
"mean_token_accuracy": 0.19460718631744384,
"num_tokens": 18385321.0,
"step": 8025
},
{
"entropy": 5.665828895568848,
"epoch": 0.7937920126532226,
"grad_norm": 0.9609375,
"learning_rate": 0.0004945527508597165,
"loss": 5.3654,
"mean_token_accuracy": 0.18383132070302963,
"num_tokens": 18397072.0,
"step": 8030
},
{
"entropy": 5.6188959121704105,
"epoch": 0.794286279161724,
"grad_norm": 0.93359375,
"learning_rate": 0.0004945450299029713,
"loss": 5.3134,
"mean_token_accuracy": 0.18940794765949248,
"num_tokens": 18408172.0,
"step": 8035
},
{
"entropy": 5.651398515701294,
"epoch": 0.7947805456702254,
"grad_norm": 0.86328125,
"learning_rate": 0.0004945373035453842,
"loss": 5.3146,
"mean_token_accuracy": 0.19606102257966995,
"num_tokens": 18419350.0,
"step": 8040
},
{
"entropy": 5.631956243515015,
"epoch": 0.7952748121787268,
"grad_norm": 0.94921875,
"learning_rate": 0.0004945295717871454,
"loss": 5.3611,
"mean_token_accuracy": 0.18574828803539276,
"num_tokens": 18430907.0,
"step": 8045
},
{
"entropy": 5.672289752960205,
"epoch": 0.7957690786872281,
"grad_norm": 1.0234375,
"learning_rate": 0.0004945218346284451,
"loss": 5.3493,
"mean_token_accuracy": 0.1913642480969429,
"num_tokens": 18441167.0,
"step": 8050
},
{
"entropy": 5.6557718276977536,
"epoch": 0.7962633451957295,
"grad_norm": 1.0703125,
"learning_rate": 0.0004945140920694736,
"loss": 5.2694,
"mean_token_accuracy": 0.19807880222797394,
"num_tokens": 18452025.0,
"step": 8055
},
{
"entropy": 5.67685546875,
"epoch": 0.7967576117042309,
"grad_norm": 0.94140625,
"learning_rate": 0.0004945063441104214,
"loss": 5.5101,
"mean_token_accuracy": 0.17122341692447662,
"num_tokens": 18464305.0,
"step": 8060
},
{
"entropy": 5.654890537261963,
"epoch": 0.7972518782127324,
"grad_norm": 0.9375,
"learning_rate": 0.0004944985907514791,
"loss": 5.4066,
"mean_token_accuracy": 0.18807248324155806,
"num_tokens": 18476217.0,
"step": 8065
},
{
"entropy": 5.736039495468139,
"epoch": 0.7977461447212337,
"grad_norm": 0.9609375,
"learning_rate": 0.0004944908319928372,
"loss": 5.374,
"mean_token_accuracy": 0.1839275300502777,
"num_tokens": 18486967.0,
"step": 8070
},
{
"entropy": 5.67571382522583,
"epoch": 0.7982404112297351,
"grad_norm": 1.0234375,
"learning_rate": 0.0004944830678346871,
"loss": 5.3449,
"mean_token_accuracy": 0.18242959231138228,
"num_tokens": 18498918.0,
"step": 8075
},
{
"entropy": 5.63468108177185,
"epoch": 0.7987346777382365,
"grad_norm": 1.0078125,
"learning_rate": 0.0004944752982772194,
"loss": 5.3226,
"mean_token_accuracy": 0.19296954274177552,
"num_tokens": 18509424.0,
"step": 8080
},
{
"entropy": 5.581369543075562,
"epoch": 0.7992289442467378,
"grad_norm": 0.9765625,
"learning_rate": 0.0004944675233206252,
"loss": 5.2736,
"mean_token_accuracy": 0.19389283806085586,
"num_tokens": 18520444.0,
"step": 8085
},
{
"entropy": 5.606768465042114,
"epoch": 0.7997232107552392,
"grad_norm": 0.90625,
"learning_rate": 0.000494459742965096,
"loss": 5.3487,
"mean_token_accuracy": 0.19006512463092803,
"num_tokens": 18531874.0,
"step": 8090
},
{
"entropy": 5.740726375579834,
"epoch": 0.8002174772637406,
"grad_norm": 0.98828125,
"learning_rate": 0.000494451957210823,
"loss": 5.4245,
"mean_token_accuracy": 0.1826380893588066,
"num_tokens": 18542492.0,
"step": 8095
},
{
"entropy": 5.749100589752198,
"epoch": 0.800711743772242,
"grad_norm": 1.21875,
"learning_rate": 0.0004944441660579979,
"loss": 5.4596,
"mean_token_accuracy": 0.177557510137558,
"num_tokens": 18554477.0,
"step": 8100
},
{
"entropy": 5.633746767044068,
"epoch": 0.8012060102807433,
"grad_norm": 0.98046875,
"learning_rate": 0.0004944363695068121,
"loss": 5.3282,
"mean_token_accuracy": 0.1938049763441086,
"num_tokens": 18565416.0,
"step": 8105
},
{
"entropy": 5.649519681930542,
"epoch": 0.8017002767892447,
"grad_norm": 0.98046875,
"learning_rate": 0.0004944285675574576,
"loss": 5.2946,
"mean_token_accuracy": 0.19163744747638703,
"num_tokens": 18575949.0,
"step": 8110
},
{
"entropy": 5.626943635940552,
"epoch": 0.8021945432977462,
"grad_norm": 0.921875,
"learning_rate": 0.0004944207602101265,
"loss": 5.3986,
"mean_token_accuracy": 0.1850971981883049,
"num_tokens": 18588177.0,
"step": 8115
},
{
"entropy": 5.642237997055053,
"epoch": 0.8026888098062476,
"grad_norm": 0.9453125,
"learning_rate": 0.0004944129474650104,
"loss": 5.2672,
"mean_token_accuracy": 0.18568051755428314,
"num_tokens": 18599112.0,
"step": 8120
},
{
"entropy": 5.5302375793457035,
"epoch": 0.803183076314749,
"grad_norm": 0.94921875,
"learning_rate": 0.000494405129322302,
"loss": 5.2122,
"mean_token_accuracy": 0.19962379783391954,
"num_tokens": 18610432.0,
"step": 8125
},
{
"entropy": 5.663758420944214,
"epoch": 0.8036773428232503,
"grad_norm": 0.9921875,
"learning_rate": 0.0004943973057821931,
"loss": 5.3273,
"mean_token_accuracy": 0.1833953693509102,
"num_tokens": 18621739.0,
"step": 8130
},
{
"entropy": 5.680219650268555,
"epoch": 0.8041716093317517,
"grad_norm": 0.96484375,
"learning_rate": 0.0004943894768448768,
"loss": 5.3166,
"mean_token_accuracy": 0.19260417222976683,
"num_tokens": 18631983.0,
"step": 8135
},
{
"entropy": 5.6862352848052975,
"epoch": 0.8046658758402531,
"grad_norm": 0.94140625,
"learning_rate": 0.000494381642510545,
"loss": 5.4256,
"mean_token_accuracy": 0.17927457243204117,
"num_tokens": 18643507.0,
"step": 8140
},
{
"entropy": 5.7087085247039795,
"epoch": 0.8051601423487544,
"grad_norm": 0.9765625,
"learning_rate": 0.0004943738027793909,
"loss": 5.4466,
"mean_token_accuracy": 0.18695662617683412,
"num_tokens": 18654516.0,
"step": 8145
},
{
"entropy": 5.69820761680603,
"epoch": 0.8056544088572558,
"grad_norm": 0.8984375,
"learning_rate": 0.0004943659576516072,
"loss": 5.4419,
"mean_token_accuracy": 0.18055962175130844,
"num_tokens": 18666855.0,
"step": 8150
},
{
"entropy": 5.671946907043457,
"epoch": 0.8061486753657572,
"grad_norm": 0.875,
"learning_rate": 0.0004943581071273869,
"loss": 5.3883,
"mean_token_accuracy": 0.1805286005139351,
"num_tokens": 18679105.0,
"step": 8155
},
{
"entropy": 5.698519659042359,
"epoch": 0.8066429418742586,
"grad_norm": 0.8671875,
"learning_rate": 0.0004943502512069232,
"loss": 5.4727,
"mean_token_accuracy": 0.1906839370727539,
"num_tokens": 18691349.0,
"step": 8160
},
{
"entropy": 5.750455522537232,
"epoch": 0.8071372083827599,
"grad_norm": 0.96484375,
"learning_rate": 0.0004943423898904092,
"loss": 5.4842,
"mean_token_accuracy": 0.18375960737466812,
"num_tokens": 18703773.0,
"step": 8165
},
{
"entropy": 5.706942796707153,
"epoch": 0.8076314748912614,
"grad_norm": 0.828125,
"learning_rate": 0.0004943345231780385,
"loss": 5.3441,
"mean_token_accuracy": 0.18648601621389388,
"num_tokens": 18715281.0,
"step": 8170
},
{
"entropy": 5.662128543853759,
"epoch": 0.8081257413997628,
"grad_norm": 1.0078125,
"learning_rate": 0.0004943266510700043,
"loss": 5.348,
"mean_token_accuracy": 0.19051504135131836,
"num_tokens": 18726446.0,
"step": 8175
},
{
"entropy": 5.574031162261963,
"epoch": 0.8086200079082642,
"grad_norm": 1.0078125,
"learning_rate": 0.0004943187735665007,
"loss": 5.2039,
"mean_token_accuracy": 0.2042323023080826,
"num_tokens": 18737348.0,
"step": 8180
},
{
"entropy": 5.6237263679504395,
"epoch": 0.8091142744167655,
"grad_norm": 0.9296875,
"learning_rate": 0.0004943108906677211,
"loss": 5.3182,
"mean_token_accuracy": 0.19175722897052766,
"num_tokens": 18748921.0,
"step": 8185
},
{
"entropy": 5.64940390586853,
"epoch": 0.8096085409252669,
"grad_norm": 0.99609375,
"learning_rate": 0.0004943030023738595,
"loss": 5.31,
"mean_token_accuracy": 0.19513626247644425,
"num_tokens": 18761252.0,
"step": 8190
},
{
"entropy": 5.643872547149658,
"epoch": 0.8101028074337683,
"grad_norm": 0.90234375,
"learning_rate": 0.0004942951086851102,
"loss": 5.3006,
"mean_token_accuracy": 0.18811838924884797,
"num_tokens": 18772915.0,
"step": 8195
},
{
"entropy": 5.632459449768066,
"epoch": 0.8105970739422697,
"grad_norm": 1.0234375,
"learning_rate": 0.0004942872096016671,
"loss": 5.321,
"mean_token_accuracy": 0.1934368595480919,
"num_tokens": 18783533.0,
"step": 8200
},
{
"entropy": 5.623964548110962,
"epoch": 0.811091340450771,
"grad_norm": 0.9296875,
"learning_rate": 0.0004942793051237247,
"loss": 5.3061,
"mean_token_accuracy": 0.18667526841163634,
"num_tokens": 18795068.0,
"step": 8205
},
{
"entropy": 5.704107761383057,
"epoch": 0.8115856069592724,
"grad_norm": 1.0078125,
"learning_rate": 0.0004942713952514773,
"loss": 5.4691,
"mean_token_accuracy": 0.17669492065906525,
"num_tokens": 18808105.0,
"step": 8210
},
{
"entropy": 5.671244192123413,
"epoch": 0.8120798734677738,
"grad_norm": 0.96484375,
"learning_rate": 0.0004942634799851196,
"loss": 5.3418,
"mean_token_accuracy": 0.18254307061433792,
"num_tokens": 18819275.0,
"step": 8215
},
{
"entropy": 5.6933129787445065,
"epoch": 0.8125741399762753,
"grad_norm": 1.0390625,
"learning_rate": 0.0004942555593248463,
"loss": 5.3079,
"mean_token_accuracy": 0.19309946447610854,
"num_tokens": 18828535.0,
"step": 8220
},
{
"entropy": 5.63012318611145,
"epoch": 0.8130684064847766,
"grad_norm": 0.95703125,
"learning_rate": 0.0004942476332708522,
"loss": 5.3201,
"mean_token_accuracy": 0.1880142495036125,
"num_tokens": 18839983.0,
"step": 8225
},
{
"entropy": 5.619532632827759,
"epoch": 0.813562672993278,
"grad_norm": 0.99609375,
"learning_rate": 0.0004942397018233322,
"loss": 5.3667,
"mean_token_accuracy": 0.18770268112421035,
"num_tokens": 18851875.0,
"step": 8230
},
{
"entropy": 5.621423578262329,
"epoch": 0.8140569395017794,
"grad_norm": 0.828125,
"learning_rate": 0.0004942317649824816,
"loss": 5.3782,
"mean_token_accuracy": 0.18783365488052367,
"num_tokens": 18863629.0,
"step": 8235
},
{
"entropy": 5.647859239578247,
"epoch": 0.8145512060102807,
"grad_norm": 0.9609375,
"learning_rate": 0.0004942238227484956,
"loss": 5.2779,
"mean_token_accuracy": 0.19434317499399184,
"num_tokens": 18875660.0,
"step": 8240
},
{
"entropy": 5.656901836395264,
"epoch": 0.8150454725187821,
"grad_norm": 0.9296875,
"learning_rate": 0.0004942158751215696,
"loss": 5.4063,
"mean_token_accuracy": 0.18389780670404435,
"num_tokens": 18887140.0,
"step": 8245
},
{
"entropy": 5.61653413772583,
"epoch": 0.8155397390272835,
"grad_norm": 0.8828125,
"learning_rate": 0.0004942079221018989,
"loss": 5.3902,
"mean_token_accuracy": 0.18758510798215866,
"num_tokens": 18899386.0,
"step": 8250
},
{
"entropy": 5.730333232879639,
"epoch": 0.8160340055357849,
"grad_norm": 0.96484375,
"learning_rate": 0.0004941999636896795,
"loss": 5.4458,
"mean_token_accuracy": 0.17575519382953644,
"num_tokens": 18911187.0,
"step": 8255
},
{
"entropy": 5.662143564224243,
"epoch": 0.8165282720442862,
"grad_norm": 0.9921875,
"learning_rate": 0.0004941919998851068,
"loss": 5.318,
"mean_token_accuracy": 0.1871063530445099,
"num_tokens": 18922522.0,
"step": 8260
},
{
"entropy": 5.538512277603149,
"epoch": 0.8170225385527876,
"grad_norm": 0.98046875,
"learning_rate": 0.0004941840306883771,
"loss": 5.2362,
"mean_token_accuracy": 0.19285360276699065,
"num_tokens": 18932689.0,
"step": 8265
},
{
"entropy": 5.706580305099488,
"epoch": 0.8175168050612891,
"grad_norm": 0.98828125,
"learning_rate": 0.0004941760560996861,
"loss": 5.4758,
"mean_token_accuracy": 0.1789873644709587,
"num_tokens": 18944296.0,
"step": 8270
},
{
"entropy": 5.7018084049224855,
"epoch": 0.8180110715697905,
"grad_norm": 0.92578125,
"learning_rate": 0.0004941680761192302,
"loss": 5.4325,
"mean_token_accuracy": 0.18427854031324387,
"num_tokens": 18955449.0,
"step": 8275
},
{
"entropy": 5.7311506271362305,
"epoch": 0.8185053380782918,
"grad_norm": 1.0546875,
"learning_rate": 0.0004941600907472055,
"loss": 5.4084,
"mean_token_accuracy": 0.18190620988607406,
"num_tokens": 18969065.0,
"step": 8280
},
{
"entropy": 5.679945564270019,
"epoch": 0.8189996045867932,
"grad_norm": 1.0078125,
"learning_rate": 0.0004941520999838088,
"loss": 5.4114,
"mean_token_accuracy": 0.18917620927095413,
"num_tokens": 18979801.0,
"step": 8285
},
{
"entropy": 5.686408424377442,
"epoch": 0.8194938710952946,
"grad_norm": 0.9765625,
"learning_rate": 0.0004941441038292363,
"loss": 5.4472,
"mean_token_accuracy": 0.18298430889844894,
"num_tokens": 18990795.0,
"step": 8290
},
{
"entropy": 5.710878992080689,
"epoch": 0.819988137603796,
"grad_norm": 0.921875,
"learning_rate": 0.0004941361022836847,
"loss": 5.4142,
"mean_token_accuracy": 0.18471910208463668,
"num_tokens": 19002318.0,
"step": 8295
},
{
"entropy": 5.650199365615845,
"epoch": 0.8204824041122973,
"grad_norm": 0.91015625,
"learning_rate": 0.0004941280953473513,
"loss": 5.4144,
"mean_token_accuracy": 0.19438256621360778,
"num_tokens": 19014977.0,
"step": 8300
},
{
"entropy": 5.697762298583984,
"epoch": 0.8209766706207987,
"grad_norm": 0.9453125,
"learning_rate": 0.0004941200830204325,
"loss": 5.4228,
"mean_token_accuracy": 0.18394842445850373,
"num_tokens": 19026081.0,
"step": 8305
},
{
"entropy": 5.6803977489471436,
"epoch": 0.8214709371293001,
"grad_norm": 1.03125,
"learning_rate": 0.0004941120653031256,
"loss": 5.3963,
"mean_token_accuracy": 0.1822434276342392,
"num_tokens": 19037694.0,
"step": 8310
},
{
"entropy": 5.619025754928589,
"epoch": 0.8219652036378015,
"grad_norm": 1.0546875,
"learning_rate": 0.0004941040421956282,
"loss": 5.3146,
"mean_token_accuracy": 0.19455636739730836,
"num_tokens": 19048344.0,
"step": 8315
},
{
"entropy": 5.722106456756592,
"epoch": 0.8224594701463029,
"grad_norm": 0.9375,
"learning_rate": 0.0004940960136981372,
"loss": 5.3764,
"mean_token_accuracy": 0.1871915876865387,
"num_tokens": 19060213.0,
"step": 8320
},
{
"entropy": 5.606636047363281,
"epoch": 0.8229537366548043,
"grad_norm": 0.96875,
"learning_rate": 0.0004940879798108503,
"loss": 5.2476,
"mean_token_accuracy": 0.19717976301908494,
"num_tokens": 19070881.0,
"step": 8325
},
{
"entropy": 5.5907165050506595,
"epoch": 0.8234480031633057,
"grad_norm": 0.90625,
"learning_rate": 0.0004940799405339651,
"loss": 5.3104,
"mean_token_accuracy": 0.1917427361011505,
"num_tokens": 19082206.0,
"step": 8330
},
{
"entropy": 5.587018775939941,
"epoch": 0.8239422696718071,
"grad_norm": 0.859375,
"learning_rate": 0.0004940718958676794,
"loss": 5.3315,
"mean_token_accuracy": 0.193899068236351,
"num_tokens": 19094014.0,
"step": 8335
},
{
"entropy": 5.690460920333862,
"epoch": 0.8244365361803084,
"grad_norm": 0.90234375,
"learning_rate": 0.000494063845812191,
"loss": 5.3088,
"mean_token_accuracy": 0.1972723424434662,
"num_tokens": 19106859.0,
"step": 8340
},
{
"entropy": 5.587282657623291,
"epoch": 0.8249308026888098,
"grad_norm": 0.96484375,
"learning_rate": 0.0004940557903676981,
"loss": 5.3731,
"mean_token_accuracy": 0.18980784267187117,
"num_tokens": 19118296.0,
"step": 8345
},
{
"entropy": 5.666278123855591,
"epoch": 0.8254250691973112,
"grad_norm": 0.90234375,
"learning_rate": 0.0004940477295343988,
"loss": 5.301,
"mean_token_accuracy": 0.1907990410923958,
"num_tokens": 19130212.0,
"step": 8350
},
{
"entropy": 5.605793571472168,
"epoch": 0.8259193357058126,
"grad_norm": 0.9609375,
"learning_rate": 0.0004940396633124912,
"loss": 5.2785,
"mean_token_accuracy": 0.20178446620702745,
"num_tokens": 19140798.0,
"step": 8355
},
{
"entropy": 5.691914510726929,
"epoch": 0.8264136022143139,
"grad_norm": 0.9453125,
"learning_rate": 0.0004940315917021742,
"loss": 5.4188,
"mean_token_accuracy": 0.18535835891962052,
"num_tokens": 19151875.0,
"step": 8360
},
{
"entropy": 5.742526245117188,
"epoch": 0.8269078687228153,
"grad_norm": 0.859375,
"learning_rate": 0.0004940235147036457,
"loss": 5.4629,
"mean_token_accuracy": 0.17392984479665757,
"num_tokens": 19164532.0,
"step": 8365
},
{
"entropy": 5.665090608596802,
"epoch": 0.8274021352313167,
"grad_norm": 0.88671875,
"learning_rate": 0.000494015432317105,
"loss": 5.3767,
"mean_token_accuracy": 0.18460462540388106,
"num_tokens": 19176828.0,
"step": 8370
},
{
"entropy": 5.572837066650391,
"epoch": 0.8278964017398182,
"grad_norm": 0.94140625,
"learning_rate": 0.0004940073445427506,
"loss": 5.2443,
"mean_token_accuracy": 0.2020925760269165,
"num_tokens": 19188118.0,
"step": 8375
},
{
"entropy": 5.62059473991394,
"epoch": 0.8283906682483195,
"grad_norm": 0.98828125,
"learning_rate": 0.0004939992513807816,
"loss": 5.422,
"mean_token_accuracy": 0.18436657041311263,
"num_tokens": 19201009.0,
"step": 8380
},
{
"entropy": 5.63350772857666,
"epoch": 0.8288849347568209,
"grad_norm": 0.84765625,
"learning_rate": 0.0004939911528313971,
"loss": 5.3125,
"mean_token_accuracy": 0.1860504150390625,
"num_tokens": 19212747.0,
"step": 8385
},
{
"entropy": 5.690491580963135,
"epoch": 0.8293792012653223,
"grad_norm": 0.92578125,
"learning_rate": 0.000493983048894796,
"loss": 5.3526,
"mean_token_accuracy": 0.18318575620651245,
"num_tokens": 19224043.0,
"step": 8390
},
{
"entropy": 5.652051591873169,
"epoch": 0.8298734677738236,
"grad_norm": 0.98046875,
"learning_rate": 0.0004939749395711782,
"loss": 5.2991,
"mean_token_accuracy": 0.19014801383018493,
"num_tokens": 19235475.0,
"step": 8395
},
{
"entropy": 5.719591093063355,
"epoch": 0.830367734282325,
"grad_norm": 0.94140625,
"learning_rate": 0.0004939668248607428,
"loss": 5.3841,
"mean_token_accuracy": 0.18801588863134383,
"num_tokens": 19247208.0,
"step": 8400
},
{
"entropy": 5.619096803665161,
"epoch": 0.8308620007908264,
"grad_norm": 0.859375,
"learning_rate": 0.0004939587047636896,
"loss": 5.3513,
"mean_token_accuracy": 0.18398840725421906,
"num_tokens": 19260172.0,
"step": 8405
},
{
"entropy": 5.644792604446411,
"epoch": 0.8313562672993278,
"grad_norm": 0.92578125,
"learning_rate": 0.0004939505792802182,
"loss": 5.392,
"mean_token_accuracy": 0.18869961947202682,
"num_tokens": 19271173.0,
"step": 8410
},
{
"entropy": 5.674806785583496,
"epoch": 0.8318505338078291,
"grad_norm": 1.0390625,
"learning_rate": 0.0004939424484105287,
"loss": 5.3652,
"mean_token_accuracy": 0.19194461554288864,
"num_tokens": 19282693.0,
"step": 8415
},
{
"entropy": 5.679836702346802,
"epoch": 0.8323448003163305,
"grad_norm": 0.9140625,
"learning_rate": 0.0004939343121548209,
"loss": 5.3848,
"mean_token_accuracy": 0.17997315227985383,
"num_tokens": 19293883.0,
"step": 8420
},
{
"entropy": 5.564936923980713,
"epoch": 0.832839066824832,
"grad_norm": 0.9765625,
"learning_rate": 0.0004939261705132951,
"loss": 5.2542,
"mean_token_accuracy": 0.1949543133378029,
"num_tokens": 19304530.0,
"step": 8425
},
{
"entropy": 5.5415472984313965,
"epoch": 0.8333333333333334,
"grad_norm": 1.0,
"learning_rate": 0.0004939180234861516,
"loss": 5.3094,
"mean_token_accuracy": 0.19091598093509674,
"num_tokens": 19315765.0,
"step": 8430
},
{
"entropy": 5.666932535171509,
"epoch": 0.8338275998418347,
"grad_norm": 0.98046875,
"learning_rate": 0.0004939098710735906,
"loss": 5.3454,
"mean_token_accuracy": 0.1927936092019081,
"num_tokens": 19328484.0,
"step": 8435
},
{
"entropy": 5.615507936477661,
"epoch": 0.8343218663503361,
"grad_norm": 0.9375,
"learning_rate": 0.0004939017132758129,
"loss": 5.3052,
"mean_token_accuracy": 0.1889285370707512,
"num_tokens": 19340594.0,
"step": 8440
},
{
"entropy": 5.724935531616211,
"epoch": 0.8348161328588375,
"grad_norm": 1.0,
"learning_rate": 0.0004938935500930192,
"loss": 5.4184,
"mean_token_accuracy": 0.18532176017761232,
"num_tokens": 19351105.0,
"step": 8445
},
{
"entropy": 5.6668126583099365,
"epoch": 0.8353103993673389,
"grad_norm": 0.97265625,
"learning_rate": 0.0004938853815254101,
"loss": 5.2869,
"mean_token_accuracy": 0.19699563682079316,
"num_tokens": 19361619.0,
"step": 8450
},
{
"entropy": 5.6400025367736815,
"epoch": 0.8358046658758402,
"grad_norm": 0.84375,
"learning_rate": 0.0004938772075731867,
"loss": 5.4693,
"mean_token_accuracy": 0.18332935124635696,
"num_tokens": 19374251.0,
"step": 8455
},
{
"entropy": 5.7364896774292,
"epoch": 0.8362989323843416,
"grad_norm": 1.28125,
"learning_rate": 0.0004938690282365499,
"loss": 5.3948,
"mean_token_accuracy": 0.1773327350616455,
"num_tokens": 19386300.0,
"step": 8460
},
{
"entropy": 5.659038972854614,
"epoch": 0.836793198892843,
"grad_norm": 0.89453125,
"learning_rate": 0.0004938608435157012,
"loss": 5.3183,
"mean_token_accuracy": 0.1898784428834915,
"num_tokens": 19397082.0,
"step": 8465
},
{
"entropy": 5.610478067398072,
"epoch": 0.8372874654013444,
"grad_norm": 0.85546875,
"learning_rate": 0.0004938526534108417,
"loss": 5.4074,
"mean_token_accuracy": 0.18632503896951674,
"num_tokens": 19409123.0,
"step": 8470
},
{
"entropy": 5.542996788024903,
"epoch": 0.8377817319098458,
"grad_norm": 0.94140625,
"learning_rate": 0.0004938444579221732,
"loss": 5.1655,
"mean_token_accuracy": 0.2029905751347542,
"num_tokens": 19419854.0,
"step": 8475
},
{
"entropy": 5.636118745803833,
"epoch": 0.8382759984183472,
"grad_norm": 0.94140625,
"learning_rate": 0.0004938362570498969,
"loss": 5.3572,
"mean_token_accuracy": 0.18787182122468948,
"num_tokens": 19430825.0,
"step": 8480
},
{
"entropy": 5.623866319656372,
"epoch": 0.8387702649268486,
"grad_norm": 0.92578125,
"learning_rate": 0.0004938280507942146,
"loss": 5.3849,
"mean_token_accuracy": 0.1818978264927864,
"num_tokens": 19442617.0,
"step": 8485
},
{
"entropy": 5.721683168411255,
"epoch": 0.83926453143535,
"grad_norm": 1.03125,
"learning_rate": 0.0004938198391553285,
"loss": 5.4034,
"mean_token_accuracy": 0.18268522769212722,
"num_tokens": 19453911.0,
"step": 8490
},
{
"entropy": 5.6255655765533445,
"epoch": 0.8397587979438513,
"grad_norm": 0.9921875,
"learning_rate": 0.0004938116221334404,
"loss": 5.3002,
"mean_token_accuracy": 0.18966425955295563,
"num_tokens": 19464837.0,
"step": 8495
},
{
"entropy": 5.574881935119629,
"epoch": 0.8402530644523527,
"grad_norm": 1.015625,
"learning_rate": 0.0004938033997287524,
"loss": 5.2507,
"mean_token_accuracy": 0.19560590088367463,
"num_tokens": 19477067.0,
"step": 8500
},
{
"entropy": 5.5759570598602295,
"epoch": 0.8407473309608541,
"grad_norm": 0.92578125,
"learning_rate": 0.0004937951719414668,
"loss": 5.3182,
"mean_token_accuracy": 0.1896255925297737,
"num_tokens": 19489442.0,
"step": 8505
},
{
"entropy": 5.618827962875367,
"epoch": 0.8412415974693555,
"grad_norm": 0.94921875,
"learning_rate": 0.0004937869387717861,
"loss": 5.2782,
"mean_token_accuracy": 0.18787578344345093,
"num_tokens": 19501197.0,
"step": 8510
},
{
"entropy": 5.6528948783874515,
"epoch": 0.8417358639778568,
"grad_norm": 0.84765625,
"learning_rate": 0.0004937787002199127,
"loss": 5.3044,
"mean_token_accuracy": 0.19116035848855972,
"num_tokens": 19513311.0,
"step": 8515
},
{
"entropy": 5.735564565658569,
"epoch": 0.8422301304863582,
"grad_norm": 0.8984375,
"learning_rate": 0.0004937704562860493,
"loss": 5.4412,
"mean_token_accuracy": 0.17906060963869094,
"num_tokens": 19525643.0,
"step": 8520
},
{
"entropy": 5.566340589523316,
"epoch": 0.8427243969948597,
"grad_norm": 0.96484375,
"learning_rate": 0.0004937622069703988,
"loss": 5.2506,
"mean_token_accuracy": 0.1952924534678459,
"num_tokens": 19537478.0,
"step": 8525
},
{
"entropy": 5.652986860275268,
"epoch": 0.843218663503361,
"grad_norm": 0.87890625,
"learning_rate": 0.0004937539522731641,
"loss": 5.3283,
"mean_token_accuracy": 0.19262114465236663,
"num_tokens": 19549370.0,
"step": 8530
},
{
"entropy": 5.677238416671753,
"epoch": 0.8437129300118624,
"grad_norm": 0.92578125,
"learning_rate": 0.0004937456921945482,
"loss": 5.2597,
"mean_token_accuracy": 0.19591226130723954,
"num_tokens": 19561793.0,
"step": 8535
},
{
"entropy": 5.5420262813568115,
"epoch": 0.8442071965203638,
"grad_norm": 0.88671875,
"learning_rate": 0.0004937374267347543,
"loss": 5.1966,
"mean_token_accuracy": 0.20097958892583848,
"num_tokens": 19573603.0,
"step": 8540
},
{
"entropy": 5.625102758407593,
"epoch": 0.8447014630288652,
"grad_norm": 0.92578125,
"learning_rate": 0.0004937291558939858,
"loss": 5.3634,
"mean_token_accuracy": 0.18257204294204712,
"num_tokens": 19584848.0,
"step": 8545
},
{
"entropy": 5.616266870498658,
"epoch": 0.8451957295373665,
"grad_norm": 0.94921875,
"learning_rate": 0.0004937208796724461,
"loss": 5.3316,
"mean_token_accuracy": 0.18870454728603364,
"num_tokens": 19594965.0,
"step": 8550
},
{
"entropy": 5.628319978713989,
"epoch": 0.8456899960458679,
"grad_norm": 0.98828125,
"learning_rate": 0.0004937125980703389,
"loss": 5.2184,
"mean_token_accuracy": 0.20184896290302276,
"num_tokens": 19607347.0,
"step": 8555
},
{
"entropy": 5.6301665782928465,
"epoch": 0.8461842625543693,
"grad_norm": 0.9453125,
"learning_rate": 0.0004937043110878678,
"loss": 5.3009,
"mean_token_accuracy": 0.19272432029247283,
"num_tokens": 19618614.0,
"step": 8560
},
{
"entropy": 5.630640888214112,
"epoch": 0.8466785290628707,
"grad_norm": 0.99609375,
"learning_rate": 0.0004936960187252367,
"loss": 5.3383,
"mean_token_accuracy": 0.1898093119263649,
"num_tokens": 19630409.0,
"step": 8565
},
{
"entropy": 5.614495325088501,
"epoch": 0.847172795571372,
"grad_norm": 0.8984375,
"learning_rate": 0.0004936877209826497,
"loss": 5.3644,
"mean_token_accuracy": 0.1860443025827408,
"num_tokens": 19642183.0,
"step": 8570
},
{
"entropy": 5.610469961166382,
"epoch": 0.8476670620798734,
"grad_norm": 0.96484375,
"learning_rate": 0.0004936794178603108,
"loss": 5.344,
"mean_token_accuracy": 0.19837487488985062,
"num_tokens": 19653846.0,
"step": 8575
},
{
"entropy": 5.647045516967774,
"epoch": 0.8481613285883749,
"grad_norm": 1.0,
"learning_rate": 0.0004936711093584242,
"loss": 5.3309,
"mean_token_accuracy": 0.19066232144832612,
"num_tokens": 19664183.0,
"step": 8580
},
{
"entropy": 5.651234722137451,
"epoch": 0.8486555950968763,
"grad_norm": 0.89453125,
"learning_rate": 0.0004936627954771945,
"loss": 5.3182,
"mean_token_accuracy": 0.18965909034013748,
"num_tokens": 19675863.0,
"step": 8585
},
{
"entropy": 5.608486032485962,
"epoch": 0.8491498616053776,
"grad_norm": 0.97265625,
"learning_rate": 0.0004936544762168261,
"loss": 5.3249,
"mean_token_accuracy": 0.19289756417274476,
"num_tokens": 19688410.0,
"step": 8590
},
{
"entropy": 5.639685392379761,
"epoch": 0.849644128113879,
"grad_norm": 1.03125,
"learning_rate": 0.0004936461515775237,
"loss": 5.3136,
"mean_token_accuracy": 0.18706392198801042,
"num_tokens": 19698849.0,
"step": 8595
},
{
"entropy": 5.712615442276001,
"epoch": 0.8501383946223804,
"grad_norm": 0.875,
"learning_rate": 0.000493637821559492,
"loss": 5.474,
"mean_token_accuracy": 0.1762271672487259,
"num_tokens": 19710761.0,
"step": 8600
},
{
"entropy": 5.630744504928589,
"epoch": 0.8506326611308818,
"grad_norm": 1.0078125,
"learning_rate": 0.0004936294861629358,
"loss": 5.3504,
"mean_token_accuracy": 0.18898882120847701,
"num_tokens": 19722530.0,
"step": 8605
},
{
"entropy": 5.624499034881592,
"epoch": 0.8511269276393831,
"grad_norm": 0.9609375,
"learning_rate": 0.0004936211453880606,
"loss": 5.3031,
"mean_token_accuracy": 0.1893622949719429,
"num_tokens": 19733352.0,
"step": 8610
},
{
"entropy": 5.662234544754028,
"epoch": 0.8516211941478845,
"grad_norm": 0.98828125,
"learning_rate": 0.0004936127992350712,
"loss": 5.348,
"mean_token_accuracy": 0.19108893424272538,
"num_tokens": 19745071.0,
"step": 8615
},
{
"entropy": 5.621189308166504,
"epoch": 0.8521154606563859,
"grad_norm": 0.96484375,
"learning_rate": 0.0004936044477041729,
"loss": 5.2738,
"mean_token_accuracy": 0.1932872161269188,
"num_tokens": 19756039.0,
"step": 8620
},
{
"entropy": 5.670988082885742,
"epoch": 0.8526097271648873,
"grad_norm": 0.8984375,
"learning_rate": 0.0004935960907955715,
"loss": 5.3977,
"mean_token_accuracy": 0.17891822755336761,
"num_tokens": 19768728.0,
"step": 8625
},
{
"entropy": 5.61225266456604,
"epoch": 0.8531039936733887,
"grad_norm": 0.9609375,
"learning_rate": 0.0004935877285094722,
"loss": 5.3714,
"mean_token_accuracy": 0.18423692137002945,
"num_tokens": 19780010.0,
"step": 8630
},
{
"entropy": 5.65412220954895,
"epoch": 0.8535982601818901,
"grad_norm": 1.0390625,
"learning_rate": 0.000493579360846081,
"loss": 5.3139,
"mean_token_accuracy": 0.19768353402614594,
"num_tokens": 19790082.0,
"step": 8635
},
{
"entropy": 5.6661571025848385,
"epoch": 0.8540925266903915,
"grad_norm": 0.98046875,
"learning_rate": 0.0004935709878056034,
"loss": 5.3849,
"mean_token_accuracy": 0.1892878919839859,
"num_tokens": 19801702.0,
"step": 8640
},
{
"entropy": 5.661633682250977,
"epoch": 0.8545867931988929,
"grad_norm": 0.98828125,
"learning_rate": 0.0004935626093882457,
"loss": 5.4278,
"mean_token_accuracy": 0.18570428043603898,
"num_tokens": 19812173.0,
"step": 8645
},
{
"entropy": 5.629301738739014,
"epoch": 0.8550810597073942,
"grad_norm": 0.9375,
"learning_rate": 0.000493554225594214,
"loss": 5.3146,
"mean_token_accuracy": 0.19773084670305252,
"num_tokens": 19823286.0,
"step": 8650
},
{
"entropy": 5.693342828750611,
"epoch": 0.8555753262158956,
"grad_norm": 1.046875,
"learning_rate": 0.0004935458364237143,
"loss": 5.326,
"mean_token_accuracy": 0.190741828083992,
"num_tokens": 19834472.0,
"step": 8655
},
{
"entropy": 5.665266036987305,
"epoch": 0.856069592724397,
"grad_norm": 0.9140625,
"learning_rate": 0.0004935374418769532,
"loss": 5.35,
"mean_token_accuracy": 0.1872141808271408,
"num_tokens": 19845073.0,
"step": 8660
},
{
"entropy": 5.66392297744751,
"epoch": 0.8565638592328984,
"grad_norm": 0.94921875,
"learning_rate": 0.0004935290419541371,
"loss": 5.3737,
"mean_token_accuracy": 0.1858500674366951,
"num_tokens": 19856762.0,
"step": 8665
},
{
"entropy": 5.625785779953003,
"epoch": 0.8570581257413997,
"grad_norm": 0.9765625,
"learning_rate": 0.0004935206366554726,
"loss": 5.3188,
"mean_token_accuracy": 0.19144676774740219,
"num_tokens": 19867358.0,
"step": 8670
},
{
"entropy": 5.615385293960571,
"epoch": 0.8575523922499011,
"grad_norm": 0.98046875,
"learning_rate": 0.0004935122259811667,
"loss": 5.3561,
"mean_token_accuracy": 0.19447528421878815,
"num_tokens": 19878527.0,
"step": 8675
},
{
"entropy": 5.697225570678711,
"epoch": 0.8580466587584026,
"grad_norm": 0.921875,
"learning_rate": 0.000493503809931426,
"loss": 5.3896,
"mean_token_accuracy": 0.18498689234256743,
"num_tokens": 19889680.0,
"step": 8680
},
{
"entropy": 5.655039262771607,
"epoch": 0.858540925266904,
"grad_norm": 0.8828125,
"learning_rate": 0.0004934953885064577,
"loss": 5.2848,
"mean_token_accuracy": 0.19361486732959748,
"num_tokens": 19900569.0,
"step": 8685
},
{
"entropy": 5.588249015808105,
"epoch": 0.8590351917754053,
"grad_norm": 0.875,
"learning_rate": 0.000493486961706469,
"loss": 5.2611,
"mean_token_accuracy": 0.19664759039878846,
"num_tokens": 19912537.0,
"step": 8690
},
{
"entropy": 5.6815883159637455,
"epoch": 0.8595294582839067,
"grad_norm": 1.015625,
"learning_rate": 0.0004934785295316672,
"loss": 5.318,
"mean_token_accuracy": 0.1965631663799286,
"num_tokens": 19923931.0,
"step": 8695
},
{
"entropy": 5.705501413345337,
"epoch": 0.8600237247924081,
"grad_norm": 1.046875,
"learning_rate": 0.0004934700919822596,
"loss": 5.3899,
"mean_token_accuracy": 0.1823534056544304,
"num_tokens": 19933956.0,
"step": 8700
},
{
"entropy": 5.559091186523437,
"epoch": 0.8605179913009094,
"grad_norm": 0.9453125,
"learning_rate": 0.0004934616490584538,
"loss": 5.2973,
"mean_token_accuracy": 0.19276793003082277,
"num_tokens": 19945603.0,
"step": 8705
},
{
"entropy": 5.677602958679199,
"epoch": 0.8610122578094108,
"grad_norm": 0.953125,
"learning_rate": 0.0004934532007604577,
"loss": 5.3612,
"mean_token_accuracy": 0.19237906038761138,
"num_tokens": 19956358.0,
"step": 8710
},
{
"entropy": 5.61974778175354,
"epoch": 0.8615065243179122,
"grad_norm": 0.90625,
"learning_rate": 0.0004934447470884788,
"loss": 5.291,
"mean_token_accuracy": 0.19488293677568436,
"num_tokens": 19967856.0,
"step": 8715
},
{
"entropy": 5.60012845993042,
"epoch": 0.8620007908264136,
"grad_norm": 1.46875,
"learning_rate": 0.0004934362880427253,
"loss": 5.3462,
"mean_token_accuracy": 0.19039387553930281,
"num_tokens": 19979323.0,
"step": 8720
},
{
"entropy": 5.699257898330688,
"epoch": 0.8624950573349149,
"grad_norm": 0.921875,
"learning_rate": 0.0004934278236234052,
"loss": 5.3749,
"mean_token_accuracy": 0.18056602478027345,
"num_tokens": 19991543.0,
"step": 8725
},
{
"entropy": 5.683034086227417,
"epoch": 0.8629893238434164,
"grad_norm": 0.94921875,
"learning_rate": 0.0004934193538307268,
"loss": 5.3486,
"mean_token_accuracy": 0.18849651515483856,
"num_tokens": 20003621.0,
"step": 8730
},
{
"entropy": 5.568985176086426,
"epoch": 0.8634835903519178,
"grad_norm": 0.9375,
"learning_rate": 0.0004934108786648983,
"loss": 5.3618,
"mean_token_accuracy": 0.183358895778656,
"num_tokens": 20015183.0,
"step": 8735
},
{
"entropy": 5.660060453414917,
"epoch": 0.8639778568604192,
"grad_norm": 0.84375,
"learning_rate": 0.0004934023981261285,
"loss": 5.2955,
"mean_token_accuracy": 0.18640766888856888,
"num_tokens": 20026735.0,
"step": 8740
},
{
"entropy": 5.624735355377197,
"epoch": 0.8644721233689205,
"grad_norm": 0.89453125,
"learning_rate": 0.0004933939122146256,
"loss": 5.2601,
"mean_token_accuracy": 0.19271255731582643,
"num_tokens": 20037410.0,
"step": 8745
},
{
"entropy": 5.548708438873291,
"epoch": 0.8649663898774219,
"grad_norm": 0.94921875,
"learning_rate": 0.0004933854209305987,
"loss": 5.2634,
"mean_token_accuracy": 0.1923980250954628,
"num_tokens": 20048972.0,
"step": 8750
},
{
"entropy": 5.699293947219848,
"epoch": 0.8654606563859233,
"grad_norm": 1.109375,
"learning_rate": 0.0004933769242742566,
"loss": 5.299,
"mean_token_accuracy": 0.1967056706547737,
"num_tokens": 20059667.0,
"step": 8755
},
{
"entropy": 5.625824689865112,
"epoch": 0.8659549228944247,
"grad_norm": 0.8671875,
"learning_rate": 0.0004933684222458082,
"loss": 5.2486,
"mean_token_accuracy": 0.19832002818584443,
"num_tokens": 20070612.0,
"step": 8760
},
{
"entropy": 5.567928647994995,
"epoch": 0.866449189402926,
"grad_norm": 1.015625,
"learning_rate": 0.0004933599148454628,
"loss": 5.2725,
"mean_token_accuracy": 0.1937839835882187,
"num_tokens": 20081445.0,
"step": 8765
},
{
"entropy": 5.684582853317261,
"epoch": 0.8669434559114274,
"grad_norm": 0.91015625,
"learning_rate": 0.0004933514020734296,
"loss": 5.4135,
"mean_token_accuracy": 0.18456608355045317,
"num_tokens": 20094414.0,
"step": 8770
},
{
"entropy": 5.60057487487793,
"epoch": 0.8674377224199288,
"grad_norm": 0.90625,
"learning_rate": 0.0004933428839299181,
"loss": 5.2231,
"mean_token_accuracy": 0.2031479448080063,
"num_tokens": 20106793.0,
"step": 8775
},
{
"entropy": 5.58985481262207,
"epoch": 0.8679319889284303,
"grad_norm": 0.89453125,
"learning_rate": 0.0004933343604151376,
"loss": 5.3166,
"mean_token_accuracy": 0.1928417980670929,
"num_tokens": 20118582.0,
"step": 8780
},
{
"entropy": 5.627998208999633,
"epoch": 0.8684262554369316,
"grad_norm": 1.0859375,
"learning_rate": 0.0004933258315292982,
"loss": 5.2613,
"mean_token_accuracy": 0.19237957745790482,
"num_tokens": 20130548.0,
"step": 8785
},
{
"entropy": 5.654617643356323,
"epoch": 0.868920521945433,
"grad_norm": 1.0625,
"learning_rate": 0.0004933172972726094,
"loss": 5.4095,
"mean_token_accuracy": 0.18313205391168594,
"num_tokens": 20141821.0,
"step": 8790
},
{
"entropy": 5.6661763191223145,
"epoch": 0.8694147884539344,
"grad_norm": 1.0546875,
"learning_rate": 0.0004933087576452811,
"loss": 5.3354,
"mean_token_accuracy": 0.19099952578544616,
"num_tokens": 20152071.0,
"step": 8795
},
{
"entropy": 5.650398826599121,
"epoch": 0.8699090549624358,
"grad_norm": 0.95703125,
"learning_rate": 0.0004933002126475235,
"loss": 5.353,
"mean_token_accuracy": 0.19017913490533828,
"num_tokens": 20163780.0,
"step": 8800
},
{
"entropy": 5.583788204193115,
"epoch": 0.8704033214709371,
"grad_norm": 0.99609375,
"learning_rate": 0.0004932916622795468,
"loss": 5.3633,
"mean_token_accuracy": 0.1903591126203537,
"num_tokens": 20175477.0,
"step": 8805
},
{
"entropy": 5.574006128311157,
"epoch": 0.8708975879794385,
"grad_norm": 0.80859375,
"learning_rate": 0.0004932831065415614,
"loss": 5.2625,
"mean_token_accuracy": 0.1951898828148842,
"num_tokens": 20189452.0,
"step": 8810
},
{
"entropy": 5.6863140106201175,
"epoch": 0.8713918544879399,
"grad_norm": 0.92578125,
"learning_rate": 0.0004932745454337778,
"loss": 5.348,
"mean_token_accuracy": 0.19187350273132325,
"num_tokens": 20201250.0,
"step": 8815
},
{
"entropy": 5.577649307250977,
"epoch": 0.8718861209964412,
"grad_norm": 0.921875,
"learning_rate": 0.0004932659789564065,
"loss": 5.2832,
"mean_token_accuracy": 0.19492629766464234,
"num_tokens": 20212394.0,
"step": 8820
},
{
"entropy": 5.632228708267212,
"epoch": 0.8723803875049426,
"grad_norm": 0.921875,
"learning_rate": 0.0004932574071096581,
"loss": 5.2799,
"mean_token_accuracy": 0.19227567315101624,
"num_tokens": 20223262.0,
"step": 8825
},
{
"entropy": 5.622003507614136,
"epoch": 0.872874654013444,
"grad_norm": 0.98046875,
"learning_rate": 0.0004932488298937437,
"loss": 5.3573,
"mean_token_accuracy": 0.19351560920476912,
"num_tokens": 20234088.0,
"step": 8830
},
{
"entropy": 5.559382677078247,
"epoch": 0.8733689205219455,
"grad_norm": 0.90234375,
"learning_rate": 0.0004932402473088743,
"loss": 5.2598,
"mean_token_accuracy": 0.19630859941244125,
"num_tokens": 20245373.0,
"step": 8835
},
{
"entropy": 5.683034801483155,
"epoch": 0.8738631870304469,
"grad_norm": 0.984375,
"learning_rate": 0.0004932316593552609,
"loss": 5.3565,
"mean_token_accuracy": 0.19051172584295273,
"num_tokens": 20255824.0,
"step": 8840
},
{
"entropy": 5.575156927108765,
"epoch": 0.8743574535389482,
"grad_norm": 1.0546875,
"learning_rate": 0.0004932230660331148,
"loss": 5.2157,
"mean_token_accuracy": 0.20189533531665801,
"num_tokens": 20266576.0,
"step": 8845
},
{
"entropy": 5.6922413349151615,
"epoch": 0.8748517200474496,
"grad_norm": 1.015625,
"learning_rate": 0.0004932144673426475,
"loss": 5.3528,
"mean_token_accuracy": 0.1854723498225212,
"num_tokens": 20277874.0,
"step": 8850
},
{
"entropy": 5.6022850513458256,
"epoch": 0.875345986555951,
"grad_norm": 1.0,
"learning_rate": 0.0004932058632840705,
"loss": 5.2741,
"mean_token_accuracy": 0.19467467963695526,
"num_tokens": 20288403.0,
"step": 8855
},
{
"entropy": 5.716291666030884,
"epoch": 0.8758402530644523,
"grad_norm": 0.98046875,
"learning_rate": 0.0004931972538575952,
"loss": 5.3317,
"mean_token_accuracy": 0.18976085782051086,
"num_tokens": 20299564.0,
"step": 8860
},
{
"entropy": 5.626631450653076,
"epoch": 0.8763345195729537,
"grad_norm": 0.89453125,
"learning_rate": 0.0004931886390634338,
"loss": 5.2502,
"mean_token_accuracy": 0.19307836890220642,
"num_tokens": 20312208.0,
"step": 8865
},
{
"entropy": 5.586365795135498,
"epoch": 0.8768287860814551,
"grad_norm": 0.890625,
"learning_rate": 0.000493180018901798,
"loss": 5.3529,
"mean_token_accuracy": 0.19199634343385696,
"num_tokens": 20323486.0,
"step": 8870
},
{
"entropy": 5.6254486560821535,
"epoch": 0.8773230525899565,
"grad_norm": 0.984375,
"learning_rate": 0.0004931713933728997,
"loss": 5.3708,
"mean_token_accuracy": 0.18812766075134277,
"num_tokens": 20335806.0,
"step": 8875
},
{
"entropy": 5.627936792373657,
"epoch": 0.8778173190984578,
"grad_norm": 0.91796875,
"learning_rate": 0.0004931627624769515,
"loss": 5.2649,
"mean_token_accuracy": 0.19559958428144456,
"num_tokens": 20347216.0,
"step": 8880
},
{
"entropy": 5.668179130554199,
"epoch": 0.8783115856069593,
"grad_norm": 0.91796875,
"learning_rate": 0.0004931541262141653,
"loss": 5.3406,
"mean_token_accuracy": 0.18359874337911605,
"num_tokens": 20358938.0,
"step": 8885
},
{
"entropy": 5.646789169311523,
"epoch": 0.8788058521154607,
"grad_norm": 0.93359375,
"learning_rate": 0.0004931454845847539,
"loss": 5.3641,
"mean_token_accuracy": 0.1863119438290596,
"num_tokens": 20370742.0,
"step": 8890
},
{
"entropy": 5.533660078048706,
"epoch": 0.8793001186239621,
"grad_norm": 1.0546875,
"learning_rate": 0.0004931368375889298,
"loss": 5.2581,
"mean_token_accuracy": 0.19917842149734497,
"num_tokens": 20382100.0,
"step": 8895
},
{
"entropy": 5.696718072891235,
"epoch": 0.8797943851324634,
"grad_norm": 0.99609375,
"learning_rate": 0.0004931281852269054,
"loss": 5.4047,
"mean_token_accuracy": 0.1852152332663536,
"num_tokens": 20393348.0,
"step": 8900
},
{
"entropy": 5.6462465763092045,
"epoch": 0.8802886516409648,
"grad_norm": 1.015625,
"learning_rate": 0.000493119527498894,
"loss": 5.2435,
"mean_token_accuracy": 0.20135423094034194,
"num_tokens": 20403471.0,
"step": 8905
},
{
"entropy": 5.575044918060303,
"epoch": 0.8807829181494662,
"grad_norm": 0.9140625,
"learning_rate": 0.0004931108644051084,
"loss": 5.2979,
"mean_token_accuracy": 0.20270061641931533,
"num_tokens": 20415146.0,
"step": 8910
},
{
"entropy": 5.665388345718384,
"epoch": 0.8812771846579676,
"grad_norm": 0.94921875,
"learning_rate": 0.0004931021959457617,
"loss": 5.3077,
"mean_token_accuracy": 0.18896603733301162,
"num_tokens": 20425857.0,
"step": 8915
},
{
"entropy": 5.627101469039917,
"epoch": 0.8817714511664689,
"grad_norm": 1.0234375,
"learning_rate": 0.000493093522121067,
"loss": 5.4197,
"mean_token_accuracy": 0.1786689355969429,
"num_tokens": 20437987.0,
"step": 8920
},
{
"entropy": 5.605519247055054,
"epoch": 0.8822657176749703,
"grad_norm": 0.953125,
"learning_rate": 0.0004930848429312379,
"loss": 5.2688,
"mean_token_accuracy": 0.19142666608095169,
"num_tokens": 20449079.0,
"step": 8925
},
{
"entropy": 5.672420835494995,
"epoch": 0.8827599841834717,
"grad_norm": 0.9765625,
"learning_rate": 0.0004930761583764879,
"loss": 5.382,
"mean_token_accuracy": 0.18534404784440994,
"num_tokens": 20461032.0,
"step": 8930
},
{
"entropy": 5.713896703720093,
"epoch": 0.8832542506919732,
"grad_norm": 1.0390625,
"learning_rate": 0.0004930674684570306,
"loss": 5.3968,
"mean_token_accuracy": 0.19122811704874038,
"num_tokens": 20471825.0,
"step": 8935
},
{
"entropy": 5.683028793334961,
"epoch": 0.8837485172004745,
"grad_norm": 1.0078125,
"learning_rate": 0.0004930587731730796,
"loss": 5.3285,
"mean_token_accuracy": 0.1945595845580101,
"num_tokens": 20483329.0,
"step": 8940
},
{
"entropy": 5.636679983139038,
"epoch": 0.8842427837089759,
"grad_norm": 0.9375,
"learning_rate": 0.000493050072524849,
"loss": 5.2825,
"mean_token_accuracy": 0.18361458033323289,
"num_tokens": 20494537.0,
"step": 8945
},
{
"entropy": 5.578615140914917,
"epoch": 0.8847370502174773,
"grad_norm": 0.99609375,
"learning_rate": 0.0004930413665125528,
"loss": 5.2811,
"mean_token_accuracy": 0.19713531881570817,
"num_tokens": 20506205.0,
"step": 8950
},
{
"entropy": 5.645243167877197,
"epoch": 0.8852313167259787,
"grad_norm": 0.921875,
"learning_rate": 0.0004930326551364052,
"loss": 5.3054,
"mean_token_accuracy": 0.1907733276486397,
"num_tokens": 20517447.0,
"step": 8955
},
{
"entropy": 5.6713865280151365,
"epoch": 0.88572558323448,
"grad_norm": 0.890625,
"learning_rate": 0.0004930239383966204,
"loss": 5.3418,
"mean_token_accuracy": 0.1947559803724289,
"num_tokens": 20529435.0,
"step": 8960
},
{
"entropy": 5.6489564895629885,
"epoch": 0.8862198497429814,
"grad_norm": 0.93359375,
"learning_rate": 0.0004930152162934128,
"loss": 5.2065,
"mean_token_accuracy": 0.20294218361377717,
"num_tokens": 20540151.0,
"step": 8965
},
{
"entropy": 5.6032298564910885,
"epoch": 0.8867141162514828,
"grad_norm": 0.90234375,
"learning_rate": 0.0004930064888269973,
"loss": 5.3414,
"mean_token_accuracy": 0.18443470895290376,
"num_tokens": 20552355.0,
"step": 8970
},
{
"entropy": 5.689976978302002,
"epoch": 0.8872083827599841,
"grad_norm": 0.984375,
"learning_rate": 0.0004929977559975881,
"loss": 5.3484,
"mean_token_accuracy": 0.1876038894057274,
"num_tokens": 20562129.0,
"step": 8975
},
{
"entropy": 5.640780973434448,
"epoch": 0.8877026492684855,
"grad_norm": 0.94140625,
"learning_rate": 0.0004929890178054004,
"loss": 5.2937,
"mean_token_accuracy": 0.19223759919404984,
"num_tokens": 20573232.0,
"step": 8980
},
{
"entropy": 5.617765665054321,
"epoch": 0.888196915776987,
"grad_norm": 1.0546875,
"learning_rate": 0.0004929802742506491,
"loss": 5.2305,
"mean_token_accuracy": 0.19902599602937698,
"num_tokens": 20582986.0,
"step": 8985
},
{
"entropy": 5.674885368347168,
"epoch": 0.8886911822854884,
"grad_norm": 1.046875,
"learning_rate": 0.0004929715253335492,
"loss": 5.3303,
"mean_token_accuracy": 0.19323651790618895,
"num_tokens": 20593619.0,
"step": 8990
},
{
"entropy": 5.583137464523316,
"epoch": 0.8891854487939898,
"grad_norm": 0.953125,
"learning_rate": 0.0004929627710543159,
"loss": 5.2682,
"mean_token_accuracy": 0.1936362475156784,
"num_tokens": 20604748.0,
"step": 8995
},
{
"entropy": 5.6882861137390135,
"epoch": 0.8896797153024911,
"grad_norm": 1.015625,
"learning_rate": 0.0004929540114131647,
"loss": 5.3851,
"mean_token_accuracy": 0.1791296437382698,
"num_tokens": 20615915.0,
"step": 9000
},
{
"epoch": 0.8896797153024911,
"eval_entropy": 5.47890327331618,
"eval_loss": 5.328938007354736,
"eval_mean_token_accuracy": 0.1972782794913096,
"eval_num_tokens": 20615915.0,
"eval_runtime": 20.5309,
"eval_samples_per_second": 1583.616,
"eval_steps_per_second": 197.995,
"step": 9000
},
{
"entropy": 5.608039999008179,
"epoch": 0.8901739818109925,
"grad_norm": 0.99609375,
"learning_rate": 0.0004929452464103109,
"loss": 5.2326,
"mean_token_accuracy": 0.1942211478948593,
"num_tokens": 20627127.0,
"step": 9005
},
{
"entropy": 5.661704063415527,
"epoch": 0.8906682483194939,
"grad_norm": 0.984375,
"learning_rate": 0.0004929364760459704,
"loss": 5.3156,
"mean_token_accuracy": 0.18934856355190277,
"num_tokens": 20638568.0,
"step": 9010
},
{
"entropy": 5.576007986068726,
"epoch": 0.8911625148279952,
"grad_norm": 1.21875,
"learning_rate": 0.0004929277003203586,
"loss": 5.2466,
"mean_token_accuracy": 0.19965030401945114,
"num_tokens": 20650803.0,
"step": 9015
},
{
"entropy": 5.635476875305176,
"epoch": 0.8916567813364966,
"grad_norm": 0.94921875,
"learning_rate": 0.0004929189192336916,
"loss": 5.4154,
"mean_token_accuracy": 0.1840985432267189,
"num_tokens": 20663021.0,
"step": 9020
},
{
"entropy": 5.609073114395142,
"epoch": 0.892151047844998,
"grad_norm": 0.9609375,
"learning_rate": 0.0004929101327861854,
"loss": 5.2979,
"mean_token_accuracy": 0.1931273505091667,
"num_tokens": 20675144.0,
"step": 9025
},
{
"entropy": 5.641273498535156,
"epoch": 0.8926453143534994,
"grad_norm": 0.96484375,
"learning_rate": 0.0004929013409780562,
"loss": 5.2808,
"mean_token_accuracy": 0.19299022108316422,
"num_tokens": 20685941.0,
"step": 9030
},
{
"entropy": 5.580771636962891,
"epoch": 0.8931395808620007,
"grad_norm": 0.97265625,
"learning_rate": 0.0004928925438095202,
"loss": 5.3152,
"mean_token_accuracy": 0.18719339221715928,
"num_tokens": 20696996.0,
"step": 9035
},
{
"entropy": 5.625216817855835,
"epoch": 0.8936338473705022,
"grad_norm": 1.078125,
"learning_rate": 0.0004928837412807938,
"loss": 5.2869,
"mean_token_accuracy": 0.19271886795759202,
"num_tokens": 20708571.0,
"step": 9040
},
{
"entropy": 5.645078134536743,
"epoch": 0.8941281138790036,
"grad_norm": 0.92578125,
"learning_rate": 0.0004928749333920935,
"loss": 5.3032,
"mean_token_accuracy": 0.19714017659425737,
"num_tokens": 20719816.0,
"step": 9045
},
{
"entropy": 5.629637002944946,
"epoch": 0.894622380387505,
"grad_norm": 0.875,
"learning_rate": 0.000492866120143636,
"loss": 5.2825,
"mean_token_accuracy": 0.19270482361316682,
"num_tokens": 20731540.0,
"step": 9050
},
{
"entropy": 5.6221473693847654,
"epoch": 0.8951166468960063,
"grad_norm": 0.98046875,
"learning_rate": 0.0004928573015356382,
"loss": 5.3084,
"mean_token_accuracy": 0.1948941543698311,
"num_tokens": 20742949.0,
"step": 9055
},
{
"entropy": 5.633759689331055,
"epoch": 0.8956109134045077,
"grad_norm": 0.97265625,
"learning_rate": 0.0004928484775683171,
"loss": 5.3408,
"mean_token_accuracy": 0.1925402268767357,
"num_tokens": 20752840.0,
"step": 9060
},
{
"entropy": 5.667916584014892,
"epoch": 0.8961051799130091,
"grad_norm": 0.875,
"learning_rate": 0.0004928396482418895,
"loss": 5.3021,
"mean_token_accuracy": 0.18978545665740967,
"num_tokens": 20763772.0,
"step": 9065
},
{
"entropy": 5.62776551246643,
"epoch": 0.8965994464215105,
"grad_norm": 0.9765625,
"learning_rate": 0.0004928308135565728,
"loss": 5.2931,
"mean_token_accuracy": 0.19508009254932404,
"num_tokens": 20775815.0,
"step": 9070
},
{
"entropy": 5.6546248435974125,
"epoch": 0.8970937129300118,
"grad_norm": 0.91015625,
"learning_rate": 0.0004928219735125843,
"loss": 5.3315,
"mean_token_accuracy": 0.1867268294095993,
"num_tokens": 20787307.0,
"step": 9075
},
{
"entropy": 5.662793922424316,
"epoch": 0.8975879794385132,
"grad_norm": 0.9453125,
"learning_rate": 0.0004928131281101415,
"loss": 5.4063,
"mean_token_accuracy": 0.18566095530986787,
"num_tokens": 20798910.0,
"step": 9080
},
{
"entropy": 5.719676733016968,
"epoch": 0.8980822459470146,
"grad_norm": 1.046875,
"learning_rate": 0.000492804277349462,
"loss": 5.3311,
"mean_token_accuracy": 0.18229461908340455,
"num_tokens": 20808628.0,
"step": 9085
},
{
"entropy": 5.623229169845581,
"epoch": 0.8985765124555161,
"grad_norm": 0.98046875,
"learning_rate": 0.0004927954212307634,
"loss": 5.2484,
"mean_token_accuracy": 0.19233248829841615,
"num_tokens": 20820117.0,
"step": 9090
},
{
"entropy": 5.579427194595337,
"epoch": 0.8990707789640174,
"grad_norm": 0.9140625,
"learning_rate": 0.0004927865597542637,
"loss": 5.2329,
"mean_token_accuracy": 0.20122673958539963,
"num_tokens": 20830859.0,
"step": 9095
},
{
"entropy": 5.6382408618927,
"epoch": 0.8995650454725188,
"grad_norm": 0.984375,
"learning_rate": 0.0004927776929201807,
"loss": 5.2479,
"mean_token_accuracy": 0.1934790790081024,
"num_tokens": 20843016.0,
"step": 9100
},
{
"entropy": 5.603344392776489,
"epoch": 0.9000593119810202,
"grad_norm": 0.96875,
"learning_rate": 0.0004927688207287328,
"loss": 5.2467,
"mean_token_accuracy": 0.19735045433044435,
"num_tokens": 20854873.0,
"step": 9105
},
{
"entropy": 5.689843273162841,
"epoch": 0.9005535784895216,
"grad_norm": 0.9609375,
"learning_rate": 0.0004927599431801381,
"loss": 5.4503,
"mean_token_accuracy": 0.19032826274633408,
"num_tokens": 20866756.0,
"step": 9110
},
{
"entropy": 5.617052984237671,
"epoch": 0.9010478449980229,
"grad_norm": 0.9765625,
"learning_rate": 0.000492751060274615,
"loss": 5.2211,
"mean_token_accuracy": 0.2041651874780655,
"num_tokens": 20878859.0,
"step": 9115
},
{
"entropy": 5.6377472400665285,
"epoch": 0.9015421115065243,
"grad_norm": 1.0625,
"learning_rate": 0.0004927421720123821,
"loss": 5.2938,
"mean_token_accuracy": 0.19372290670871734,
"num_tokens": 20889188.0,
"step": 9120
},
{
"entropy": 5.5876377582550045,
"epoch": 0.9020363780150257,
"grad_norm": 1.0078125,
"learning_rate": 0.0004927332783936579,
"loss": 5.2567,
"mean_token_accuracy": 0.19304563254117965,
"num_tokens": 20900318.0,
"step": 9125
},
{
"entropy": 5.596222019195556,
"epoch": 0.902530644523527,
"grad_norm": 1.0,
"learning_rate": 0.0004927243794186614,
"loss": 5.3344,
"mean_token_accuracy": 0.18640169203281404,
"num_tokens": 20912384.0,
"step": 9130
},
{
"entropy": 5.637099075317383,
"epoch": 0.9030249110320284,
"grad_norm": 0.96875,
"learning_rate": 0.0004927154750876112,
"loss": 5.2788,
"mean_token_accuracy": 0.1945919066667557,
"num_tokens": 20923002.0,
"step": 9135
},
{
"entropy": 5.692681837081909,
"epoch": 0.9035191775405299,
"grad_norm": 0.921875,
"learning_rate": 0.0004927065654007268,
"loss": 5.4231,
"mean_token_accuracy": 0.18299901187419892,
"num_tokens": 20934235.0,
"step": 9140
},
{
"entropy": 5.5883252143859865,
"epoch": 0.9040134440490313,
"grad_norm": 0.9609375,
"learning_rate": 0.0004926976503582269,
"loss": 5.2587,
"mean_token_accuracy": 0.18882970362901688,
"num_tokens": 20946788.0,
"step": 9145
},
{
"entropy": 5.636935806274414,
"epoch": 0.9045077105575327,
"grad_norm": 0.91796875,
"learning_rate": 0.000492688729960331,
"loss": 5.2913,
"mean_token_accuracy": 0.19535788744688035,
"num_tokens": 20958963.0,
"step": 9150
},
{
"entropy": 5.6738831996917725,
"epoch": 0.905001977066034,
"grad_norm": 0.90234375,
"learning_rate": 0.0004926798042072587,
"loss": 5.3002,
"mean_token_accuracy": 0.1906489074230194,
"num_tokens": 20969630.0,
"step": 9155
},
{
"entropy": 5.59778790473938,
"epoch": 0.9054962435745354,
"grad_norm": 0.984375,
"learning_rate": 0.0004926708730992294,
"loss": 5.256,
"mean_token_accuracy": 0.19406097680330275,
"num_tokens": 20980489.0,
"step": 9160
},
{
"entropy": 5.64363317489624,
"epoch": 0.9059905100830368,
"grad_norm": 0.97265625,
"learning_rate": 0.0004926619366364629,
"loss": 5.3141,
"mean_token_accuracy": 0.1919349417090416,
"num_tokens": 20992199.0,
"step": 9165
},
{
"entropy": 5.676075553894043,
"epoch": 0.9064847765915381,
"grad_norm": 1.0390625,
"learning_rate": 0.000492652994819179,
"loss": 5.326,
"mean_token_accuracy": 0.1892781972885132,
"num_tokens": 21003680.0,
"step": 9170
},
{
"entropy": 5.608012962341308,
"epoch": 0.9069790431000395,
"grad_norm": 1.0625,
"learning_rate": 0.0004926440476475975,
"loss": 5.2328,
"mean_token_accuracy": 0.20041443407535553,
"num_tokens": 21014191.0,
"step": 9175
},
{
"entropy": 5.5886283874511715,
"epoch": 0.9074733096085409,
"grad_norm": 1.0546875,
"learning_rate": 0.0004926350951219388,
"loss": 5.2465,
"mean_token_accuracy": 0.19715968370437623,
"num_tokens": 21025692.0,
"step": 9180
},
{
"entropy": 5.615425872802734,
"epoch": 0.9079675761170423,
"grad_norm": 0.99609375,
"learning_rate": 0.000492626137242423,
"loss": 5.2478,
"mean_token_accuracy": 0.19887588173151016,
"num_tokens": 21036643.0,
"step": 9185
},
{
"entropy": 5.651152896881103,
"epoch": 0.9084618426255437,
"grad_norm": 0.953125,
"learning_rate": 0.0004926171740092704,
"loss": 5.327,
"mean_token_accuracy": 0.18680619597434997,
"num_tokens": 21047509.0,
"step": 9190
},
{
"entropy": 5.610640954971314,
"epoch": 0.9089561091340451,
"grad_norm": 0.98828125,
"learning_rate": 0.0004926082054227016,
"loss": 5.3698,
"mean_token_accuracy": 0.19079017788171768,
"num_tokens": 21059854.0,
"step": 9195
},
{
"entropy": 5.628097820281982,
"epoch": 0.9094503756425465,
"grad_norm": 1.109375,
"learning_rate": 0.0004925992314829372,
"loss": 5.2149,
"mean_token_accuracy": 0.19790041148662568,
"num_tokens": 21069926.0,
"step": 9200
},
{
"entropy": 5.684758615493775,
"epoch": 0.9099446421510479,
"grad_norm": 0.9140625,
"learning_rate": 0.000492590252190198,
"loss": 5.3227,
"mean_token_accuracy": 0.18842111229896547,
"num_tokens": 21081580.0,
"step": 9205
},
{
"entropy": 5.592061471939087,
"epoch": 0.9104389086595492,
"grad_norm": 1.078125,
"learning_rate": 0.0004925812675447047,
"loss": 5.2156,
"mean_token_accuracy": 0.19864101707935333,
"num_tokens": 21092510.0,
"step": 9210
},
{
"entropy": 5.593926811218262,
"epoch": 0.9109331751680506,
"grad_norm": 0.96875,
"learning_rate": 0.0004925722775466785,
"loss": 5.25,
"mean_token_accuracy": 0.19563381373882294,
"num_tokens": 21103995.0,
"step": 9215
},
{
"entropy": 5.585638093948364,
"epoch": 0.911427441676552,
"grad_norm": 0.97265625,
"learning_rate": 0.0004925632821963406,
"loss": 5.3134,
"mean_token_accuracy": 0.1955977663397789,
"num_tokens": 21114949.0,
"step": 9220
},
{
"entropy": 5.577049303054809,
"epoch": 0.9119217081850534,
"grad_norm": 0.98828125,
"learning_rate": 0.0004925542814939121,
"loss": 5.2239,
"mean_token_accuracy": 0.2003214254975319,
"num_tokens": 21125810.0,
"step": 9225
},
{
"entropy": 5.6128781795501705,
"epoch": 0.9124159746935547,
"grad_norm": 0.8984375,
"learning_rate": 0.0004925452754396146,
"loss": 5.3112,
"mean_token_accuracy": 0.1855644091963768,
"num_tokens": 21137728.0,
"step": 9230
},
{
"entropy": 5.598874044418335,
"epoch": 0.9129102412020561,
"grad_norm": 0.953125,
"learning_rate": 0.0004925362640336696,
"loss": 5.2841,
"mean_token_accuracy": 0.19423135221004487,
"num_tokens": 21149249.0,
"step": 9235
},
{
"entropy": 5.603559255599976,
"epoch": 0.9134045077105575,
"grad_norm": 0.8828125,
"learning_rate": 0.0004925272472762987,
"loss": 5.3666,
"mean_token_accuracy": 0.18429987132549286,
"num_tokens": 21162338.0,
"step": 9240
},
{
"entropy": 5.558680629730224,
"epoch": 0.913898774219059,
"grad_norm": 0.9609375,
"learning_rate": 0.0004925182251677238,
"loss": 5.1388,
"mean_token_accuracy": 0.2053635001182556,
"num_tokens": 21172953.0,
"step": 9245
},
{
"entropy": 5.584990072250366,
"epoch": 0.9143930407275603,
"grad_norm": 0.9921875,
"learning_rate": 0.0004925091977081668,
"loss": 5.2733,
"mean_token_accuracy": 0.19499597698450089,
"num_tokens": 21183835.0,
"step": 9250
},
{
"entropy": 5.551011085510254,
"epoch": 0.9148873072360617,
"grad_norm": 1.0703125,
"learning_rate": 0.0004925001648978498,
"loss": 5.2605,
"mean_token_accuracy": 0.19924670606851577,
"num_tokens": 21195392.0,
"step": 9255
},
{
"entropy": 5.737870788574218,
"epoch": 0.9153815737445631,
"grad_norm": 1.0234375,
"learning_rate": 0.000492491126736995,
"loss": 5.414,
"mean_token_accuracy": 0.18362648338079451,
"num_tokens": 21207213.0,
"step": 9260
},
{
"entropy": 5.68385534286499,
"epoch": 0.9158758402530645,
"grad_norm": 0.953125,
"learning_rate": 0.0004924820832258248,
"loss": 5.3559,
"mean_token_accuracy": 0.1934768319129944,
"num_tokens": 21218843.0,
"step": 9265
},
{
"entropy": 5.57321605682373,
"epoch": 0.9163701067615658,
"grad_norm": 1.046875,
"learning_rate": 0.0004924730343645614,
"loss": 5.3401,
"mean_token_accuracy": 0.1911016121506691,
"num_tokens": 21229596.0,
"step": 9270
},
{
"entropy": 5.667065525054932,
"epoch": 0.9168643732700672,
"grad_norm": 0.95703125,
"learning_rate": 0.0004924639801534278,
"loss": 5.26,
"mean_token_accuracy": 0.1913779318332672,
"num_tokens": 21240040.0,
"step": 9275
},
{
"entropy": 5.679235315322876,
"epoch": 0.9173586397785686,
"grad_norm": 0.890625,
"learning_rate": 0.0004924549205926465,
"loss": 5.3974,
"mean_token_accuracy": 0.18306160867214202,
"num_tokens": 21252046.0,
"step": 9280
},
{
"entropy": 5.6912134170532225,
"epoch": 0.91785290628707,
"grad_norm": 0.8984375,
"learning_rate": 0.0004924458556824405,
"loss": 5.3186,
"mean_token_accuracy": 0.1818564310669899,
"num_tokens": 21263720.0,
"step": 9285
},
{
"entropy": 5.66510329246521,
"epoch": 0.9183471727955713,
"grad_norm": 0.9140625,
"learning_rate": 0.0004924367854230326,
"loss": 5.3333,
"mean_token_accuracy": 0.18773467242717742,
"num_tokens": 21275751.0,
"step": 9290
},
{
"entropy": 5.60714054107666,
"epoch": 0.9188414393040728,
"grad_norm": 0.9140625,
"learning_rate": 0.0004924277098146461,
"loss": 5.3171,
"mean_token_accuracy": 0.19376610368490219,
"num_tokens": 21287081.0,
"step": 9295
},
{
"entropy": 5.591529417037964,
"epoch": 0.9193357058125742,
"grad_norm": 0.9765625,
"learning_rate": 0.0004924186288575043,
"loss": 5.2673,
"mean_token_accuracy": 0.1956242397427559,
"num_tokens": 21298007.0,
"step": 9300
},
{
"entropy": 5.689355087280274,
"epoch": 0.9198299723210756,
"grad_norm": 0.90625,
"learning_rate": 0.0004924095425518304,
"loss": 5.3088,
"mean_token_accuracy": 0.18748197108507156,
"num_tokens": 21309416.0,
"step": 9305
},
{
"entropy": 5.5772706985473635,
"epoch": 0.9203242388295769,
"grad_norm": 0.921875,
"learning_rate": 0.000492400450897848,
"loss": 5.2791,
"mean_token_accuracy": 0.19434358030557633,
"num_tokens": 21321129.0,
"step": 9310
},
{
"entropy": 5.605843257904053,
"epoch": 0.9208185053380783,
"grad_norm": 1.0390625,
"learning_rate": 0.0004923913538957809,
"loss": 5.2471,
"mean_token_accuracy": 0.19848465770483018,
"num_tokens": 21331706.0,
"step": 9315
},
{
"entropy": 5.628567028045654,
"epoch": 0.9213127718465797,
"grad_norm": 0.89453125,
"learning_rate": 0.0004923822515458526,
"loss": 5.3024,
"mean_token_accuracy": 0.19570659399032592,
"num_tokens": 21343738.0,
"step": 9320
},
{
"entropy": 5.602812242507935,
"epoch": 0.921807038355081,
"grad_norm": 0.96484375,
"learning_rate": 0.0004923731438482873,
"loss": 5.2893,
"mean_token_accuracy": 0.19304077476263046,
"num_tokens": 21354714.0,
"step": 9325
},
{
"entropy": 5.672587156295776,
"epoch": 0.9223013048635824,
"grad_norm": 0.96484375,
"learning_rate": 0.0004923640308033088,
"loss": 5.3346,
"mean_token_accuracy": 0.1851639434695244,
"num_tokens": 21365947.0,
"step": 9330
},
{
"entropy": 5.729375123977661,
"epoch": 0.9227955713720838,
"grad_norm": 1.0234375,
"learning_rate": 0.0004923549124111416,
"loss": 5.4585,
"mean_token_accuracy": 0.18300598561763765,
"num_tokens": 21376729.0,
"step": 9335
},
{
"entropy": 5.612264728546142,
"epoch": 0.9232898378805852,
"grad_norm": 1.046875,
"learning_rate": 0.0004923457886720098,
"loss": 5.2604,
"mean_token_accuracy": 0.20023103058338165,
"num_tokens": 21388017.0,
"step": 9340
},
{
"entropy": 5.570937442779541,
"epoch": 0.9237841043890866,
"grad_norm": 0.921875,
"learning_rate": 0.0004923366595861378,
"loss": 5.2105,
"mean_token_accuracy": 0.20207480043172837,
"num_tokens": 21399566.0,
"step": 9345
},
{
"entropy": 5.644844961166382,
"epoch": 0.924278370897588,
"grad_norm": 0.98828125,
"learning_rate": 0.0004923275251537503,
"loss": 5.341,
"mean_token_accuracy": 0.1912105306982994,
"num_tokens": 21410276.0,
"step": 9350
},
{
"entropy": 5.602874231338501,
"epoch": 0.9247726374060894,
"grad_norm": 0.953125,
"learning_rate": 0.000492318385375072,
"loss": 5.2731,
"mean_token_accuracy": 0.19977347403764725,
"num_tokens": 21421090.0,
"step": 9355
},
{
"entropy": 5.591992282867432,
"epoch": 0.9252669039145908,
"grad_norm": 1.0234375,
"learning_rate": 0.0004923092402503277,
"loss": 5.2685,
"mean_token_accuracy": 0.19554599225521088,
"num_tokens": 21431057.0,
"step": 9360
},
{
"entropy": 5.597142171859741,
"epoch": 0.9257611704230921,
"grad_norm": 0.89453125,
"learning_rate": 0.0004923000897797423,
"loss": 5.3303,
"mean_token_accuracy": 0.1854785606265068,
"num_tokens": 21443849.0,
"step": 9365
},
{
"entropy": 5.704565191268921,
"epoch": 0.9262554369315935,
"grad_norm": 1.09375,
"learning_rate": 0.0004922909339635411,
"loss": 5.374,
"mean_token_accuracy": 0.18549126982688904,
"num_tokens": 21454824.0,
"step": 9370
},
{
"entropy": 5.6667567729949955,
"epoch": 0.9267497034400949,
"grad_norm": 0.95703125,
"learning_rate": 0.0004922817728019492,
"loss": 5.2946,
"mean_token_accuracy": 0.19091322273015976,
"num_tokens": 21466082.0,
"step": 9375
},
{
"entropy": 5.577264308929443,
"epoch": 0.9272439699485963,
"grad_norm": 1.0390625,
"learning_rate": 0.0004922726062951918,
"loss": 5.2733,
"mean_token_accuracy": 0.19171712994575502,
"num_tokens": 21477805.0,
"step": 9380
},
{
"entropy": 5.6575517654418945,
"epoch": 0.9277382364570976,
"grad_norm": 0.98828125,
"learning_rate": 0.0004922634344434948,
"loss": 5.2732,
"mean_token_accuracy": 0.18642699718475342,
"num_tokens": 21490014.0,
"step": 9385
},
{
"entropy": 5.615246772766113,
"epoch": 0.928232502965599,
"grad_norm": 0.91015625,
"learning_rate": 0.0004922542572470835,
"loss": 5.3287,
"mean_token_accuracy": 0.1938742831349373,
"num_tokens": 21501713.0,
"step": 9390
},
{
"entropy": 5.536674118041992,
"epoch": 0.9287267694741005,
"grad_norm": 0.9765625,
"learning_rate": 0.0004922450747061837,
"loss": 5.2028,
"mean_token_accuracy": 0.19749983549118041,
"num_tokens": 21513258.0,
"step": 9395
},
{
"entropy": 5.609413528442383,
"epoch": 0.9292210359826019,
"grad_norm": 0.921875,
"learning_rate": 0.0004922358868210214,
"loss": 5.2843,
"mean_token_accuracy": 0.1925557628273964,
"num_tokens": 21526259.0,
"step": 9400
},
{
"entropy": 5.584788465499878,
"epoch": 0.9297153024911032,
"grad_norm": 0.9375,
"learning_rate": 0.0004922266935918227,
"loss": 5.2545,
"mean_token_accuracy": 0.19477527886629104,
"num_tokens": 21539144.0,
"step": 9405
},
{
"entropy": 5.673647069931031,
"epoch": 0.9302095689996046,
"grad_norm": 0.953125,
"learning_rate": 0.0004922174950188136,
"loss": 5.3461,
"mean_token_accuracy": 0.18334164917469026,
"num_tokens": 21550550.0,
"step": 9410
},
{
"entropy": 5.604119157791137,
"epoch": 0.930703835508106,
"grad_norm": 0.9140625,
"learning_rate": 0.0004922082911022204,
"loss": 5.2551,
"mean_token_accuracy": 0.19387674778699876,
"num_tokens": 21562491.0,
"step": 9415
},
{
"entropy": 5.57192096710205,
"epoch": 0.9311981020166074,
"grad_norm": 0.92578125,
"learning_rate": 0.0004921990818422694,
"loss": 5.3138,
"mean_token_accuracy": 0.18991569876670839,
"num_tokens": 21573679.0,
"step": 9420
},
{
"entropy": 5.642074489593506,
"epoch": 0.9316923685251087,
"grad_norm": 0.8671875,
"learning_rate": 0.0004921898672391874,
"loss": 5.2387,
"mean_token_accuracy": 0.19252065271139146,
"num_tokens": 21585491.0,
"step": 9425
},
{
"entropy": 5.643537187576294,
"epoch": 0.9321866350336101,
"grad_norm": 0.875,
"learning_rate": 0.0004921806472932011,
"loss": 5.2986,
"mean_token_accuracy": 0.19361097663640975,
"num_tokens": 21598302.0,
"step": 9430
},
{
"entropy": 5.548214673995972,
"epoch": 0.9326809015421115,
"grad_norm": 0.98046875,
"learning_rate": 0.0004921714220045371,
"loss": 5.3107,
"mean_token_accuracy": 0.19482027143239974,
"num_tokens": 21608442.0,
"step": 9435
},
{
"entropy": 5.64955849647522,
"epoch": 0.9331751680506128,
"grad_norm": 0.98828125,
"learning_rate": 0.0004921621913734223,
"loss": 5.3248,
"mean_token_accuracy": 0.19053839892148972,
"num_tokens": 21619594.0,
"step": 9440
},
{
"entropy": 5.69784140586853,
"epoch": 0.9336694345591142,
"grad_norm": 1.015625,
"learning_rate": 0.0004921529554000841,
"loss": 5.3212,
"mean_token_accuracy": 0.1914744883775711,
"num_tokens": 21631065.0,
"step": 9445
},
{
"entropy": 5.561979198455811,
"epoch": 0.9341637010676157,
"grad_norm": 1.0,
"learning_rate": 0.0004921437140847493,
"loss": 5.2509,
"mean_token_accuracy": 0.19402305036783218,
"num_tokens": 21640949.0,
"step": 9450
},
{
"entropy": 5.760262727737427,
"epoch": 0.9346579675761171,
"grad_norm": 0.96875,
"learning_rate": 0.0004921344674276456,
"loss": 5.373,
"mean_token_accuracy": 0.18913189619779586,
"num_tokens": 21651189.0,
"step": 9455
},
{
"entropy": 5.7121833801269535,
"epoch": 0.9351522340846185,
"grad_norm": 0.890625,
"learning_rate": 0.0004921252154290003,
"loss": 5.3728,
"mean_token_accuracy": 0.18299551159143448,
"num_tokens": 21662790.0,
"step": 9460
},
{
"entropy": 5.626903104782104,
"epoch": 0.9356465005931198,
"grad_norm": 0.921875,
"learning_rate": 0.000492115958089041,
"loss": 5.3888,
"mean_token_accuracy": 0.1882074698805809,
"num_tokens": 21674980.0,
"step": 9465
},
{
"entropy": 5.652727651596069,
"epoch": 0.9361407671016212,
"grad_norm": 0.96875,
"learning_rate": 0.0004921066954079955,
"loss": 5.2535,
"mean_token_accuracy": 0.1957204207777977,
"num_tokens": 21687721.0,
"step": 9470
},
{
"entropy": 5.662063074111939,
"epoch": 0.9366350336101226,
"grad_norm": 1.0390625,
"learning_rate": 0.0004920974273860915,
"loss": 5.4296,
"mean_token_accuracy": 0.18004003465175628,
"num_tokens": 21699522.0,
"step": 9475
},
{
"entropy": 5.616191768646241,
"epoch": 0.9371293001186239,
"grad_norm": 0.984375,
"learning_rate": 0.0004920881540235572,
"loss": 5.1816,
"mean_token_accuracy": 0.19853858798742294,
"num_tokens": 21709972.0,
"step": 9480
},
{
"entropy": 5.553194618225097,
"epoch": 0.9376235666271253,
"grad_norm": 0.9609375,
"learning_rate": 0.0004920788753206205,
"loss": 5.2943,
"mean_token_accuracy": 0.19493181705474855,
"num_tokens": 21722172.0,
"step": 9485
},
{
"entropy": 5.596704435348511,
"epoch": 0.9381178331356267,
"grad_norm": 1.0078125,
"learning_rate": 0.00049206959127751,
"loss": 5.2917,
"mean_token_accuracy": 0.19186157286167144,
"num_tokens": 21734314.0,
"step": 9490
},
{
"entropy": 5.63874249458313,
"epoch": 0.9386120996441281,
"grad_norm": 1.015625,
"learning_rate": 0.0004920603018944537,
"loss": 5.2707,
"mean_token_accuracy": 0.1962213173508644,
"num_tokens": 21744031.0,
"step": 9495
},
{
"entropy": 5.630405330657959,
"epoch": 0.9391063661526295,
"grad_norm": 1.0,
"learning_rate": 0.0004920510071716803,
"loss": 5.2482,
"mean_token_accuracy": 0.19808519929647445,
"num_tokens": 21755653.0,
"step": 9500
},
{
"entropy": 5.674383592605591,
"epoch": 0.9396006326611309,
"grad_norm": 0.85546875,
"learning_rate": 0.0004920417071094186,
"loss": 5.3348,
"mean_token_accuracy": 0.18645153492689132,
"num_tokens": 21768513.0,
"step": 9505
},
{
"entropy": 5.6544403553009035,
"epoch": 0.9400948991696323,
"grad_norm": 0.9140625,
"learning_rate": 0.0004920324017078971,
"loss": 5.2768,
"mean_token_accuracy": 0.18442377597093582,
"num_tokens": 21780857.0,
"step": 9510
},
{
"entropy": 5.579956436157227,
"epoch": 0.9405891656781337,
"grad_norm": 0.921875,
"learning_rate": 0.000492023090967345,
"loss": 5.2651,
"mean_token_accuracy": 0.20063927620649338,
"num_tokens": 21792514.0,
"step": 9515
},
{
"entropy": 5.630829238891602,
"epoch": 0.941083432186635,
"grad_norm": 1.03125,
"learning_rate": 0.000492013774887991,
"loss": 5.3212,
"mean_token_accuracy": 0.19309911578893663,
"num_tokens": 21803154.0,
"step": 9520
},
{
"entropy": 5.634845209121704,
"epoch": 0.9415776986951364,
"grad_norm": 0.89453125,
"learning_rate": 0.0004920044534700645,
"loss": 5.2831,
"mean_token_accuracy": 0.18889188021421432,
"num_tokens": 21814439.0,
"step": 9525
},
{
"entropy": 5.629956912994385,
"epoch": 0.9420719652036378,
"grad_norm": 0.90234375,
"learning_rate": 0.0004919951267137949,
"loss": 5.245,
"mean_token_accuracy": 0.19965579360723495,
"num_tokens": 21826870.0,
"step": 9530
},
{
"entropy": 5.6650889873504635,
"epoch": 0.9425662317121392,
"grad_norm": 0.9609375,
"learning_rate": 0.0004919857946194114,
"loss": 5.2282,
"mean_token_accuracy": 0.19790223240852356,
"num_tokens": 21837627.0,
"step": 9535
},
{
"entropy": 5.607153797149659,
"epoch": 0.9430604982206405,
"grad_norm": 1.03125,
"learning_rate": 0.0004919764571871438,
"loss": 5.367,
"mean_token_accuracy": 0.19002410322427749,
"num_tokens": 21848891.0,
"step": 9540
},
{
"entropy": 5.676448154449463,
"epoch": 0.9435547647291419,
"grad_norm": 0.92578125,
"learning_rate": 0.0004919671144172216,
"loss": 5.2871,
"mean_token_accuracy": 0.1898767203092575,
"num_tokens": 21860094.0,
"step": 9545
},
{
"entropy": 5.58936619758606,
"epoch": 0.9440490312376434,
"grad_norm": 0.98828125,
"learning_rate": 0.0004919577663098748,
"loss": 5.239,
"mean_token_accuracy": 0.194608274102211,
"num_tokens": 21870881.0,
"step": 9550
},
{
"entropy": 5.560912370681763,
"epoch": 0.9445432977461448,
"grad_norm": 0.89453125,
"learning_rate": 0.0004919484128653333,
"loss": 5.2871,
"mean_token_accuracy": 0.19403542578220367,
"num_tokens": 21882825.0,
"step": 9555
},
{
"entropy": 5.662732362747192,
"epoch": 0.9450375642546461,
"grad_norm": 0.8671875,
"learning_rate": 0.0004919390540838271,
"loss": 5.3591,
"mean_token_accuracy": 0.18788367360830308,
"num_tokens": 21894952.0,
"step": 9560
},
{
"entropy": 5.651932430267334,
"epoch": 0.9455318307631475,
"grad_norm": 0.94921875,
"learning_rate": 0.0004919296899655866,
"loss": 5.2923,
"mean_token_accuracy": 0.1855970874428749,
"num_tokens": 21906893.0,
"step": 9565
},
{
"entropy": 5.643131875991822,
"epoch": 0.9460260972716489,
"grad_norm": 0.9765625,
"learning_rate": 0.000491920320510842,
"loss": 5.3582,
"mean_token_accuracy": 0.19035353362560273,
"num_tokens": 21918571.0,
"step": 9570
},
{
"entropy": 5.601896953582764,
"epoch": 0.9465203637801503,
"grad_norm": 0.96484375,
"learning_rate": 0.0004919109457198239,
"loss": 5.2361,
"mean_token_accuracy": 0.1949240520596504,
"num_tokens": 21929488.0,
"step": 9575
},
{
"entropy": 5.5939172267913815,
"epoch": 0.9470146302886516,
"grad_norm": 0.921875,
"learning_rate": 0.0004919015655927629,
"loss": 5.2393,
"mean_token_accuracy": 0.1913357272744179,
"num_tokens": 21940368.0,
"step": 9580
},
{
"entropy": 5.636135387420654,
"epoch": 0.947508896797153,
"grad_norm": 0.9765625,
"learning_rate": 0.0004918921801298897,
"loss": 5.343,
"mean_token_accuracy": 0.19131146520376205,
"num_tokens": 21951171.0,
"step": 9585
},
{
"entropy": 5.673929071426391,
"epoch": 0.9480031633056544,
"grad_norm": 1.0859375,
"learning_rate": 0.0004918827893314353,
"loss": 5.3048,
"mean_token_accuracy": 0.19988007247447967,
"num_tokens": 21962312.0,
"step": 9590
},
{
"entropy": 5.668216514587402,
"epoch": 0.9484974298141557,
"grad_norm": 1.0,
"learning_rate": 0.0004918733931976305,
"loss": 5.2508,
"mean_token_accuracy": 0.19467929303646087,
"num_tokens": 21971994.0,
"step": 9595
},
{
"entropy": 5.680405378341675,
"epoch": 0.9489916963226572,
"grad_norm": 1.0390625,
"learning_rate": 0.0004918639917287067,
"loss": 5.3505,
"mean_token_accuracy": 0.18573708683252335,
"num_tokens": 21983084.0,
"step": 9600
},
{
"entropy": 5.67577052116394,
"epoch": 0.9494859628311586,
"grad_norm": 0.9453125,
"learning_rate": 0.0004918545849248951,
"loss": 5.3628,
"mean_token_accuracy": 0.18384079337120057,
"num_tokens": 21995005.0,
"step": 9605
},
{
"entropy": 5.561293649673462,
"epoch": 0.94998022933966,
"grad_norm": 0.9453125,
"learning_rate": 0.000491845172786427,
"loss": 5.2932,
"mean_token_accuracy": 0.1960711047053337,
"num_tokens": 22006377.0,
"step": 9610
},
{
"entropy": 5.612299299240112,
"epoch": 0.9504744958481613,
"grad_norm": 0.89453125,
"learning_rate": 0.0004918357553135339,
"loss": 5.3708,
"mean_token_accuracy": 0.18998344093561173,
"num_tokens": 22019663.0,
"step": 9615
},
{
"entropy": 5.755678939819336,
"epoch": 0.9509687623566627,
"grad_norm": 1.0,
"learning_rate": 0.0004918263325064476,
"loss": 5.3665,
"mean_token_accuracy": 0.18709334582090378,
"num_tokens": 22030646.0,
"step": 9620
},
{
"entropy": 5.647117233276367,
"epoch": 0.9514630288651641,
"grad_norm": 0.85546875,
"learning_rate": 0.0004918169043654,
"loss": 5.2688,
"mean_token_accuracy": 0.19839103370904923,
"num_tokens": 22043563.0,
"step": 9625
},
{
"entropy": 5.597343301773071,
"epoch": 0.9519572953736655,
"grad_norm": 0.921875,
"learning_rate": 0.0004918074708906229,
"loss": 5.2581,
"mean_token_accuracy": 0.19743698388338088,
"num_tokens": 22055224.0,
"step": 9630
},
{
"entropy": 5.621808052062988,
"epoch": 0.9524515618821668,
"grad_norm": 0.9296875,
"learning_rate": 0.0004917980320823483,
"loss": 5.3463,
"mean_token_accuracy": 0.18741225749254226,
"num_tokens": 22067445.0,
"step": 9635
},
{
"entropy": 5.627112531661988,
"epoch": 0.9529458283906682,
"grad_norm": 0.93359375,
"learning_rate": 0.0004917885879408085,
"loss": 5.2642,
"mean_token_accuracy": 0.1959032729268074,
"num_tokens": 22078565.0,
"step": 9640
},
{
"entropy": 5.5303792476654055,
"epoch": 0.9534400948991696,
"grad_norm": 0.9765625,
"learning_rate": 0.0004917791384662359,
"loss": 5.1679,
"mean_token_accuracy": 0.2047170639038086,
"num_tokens": 22088658.0,
"step": 9645
},
{
"entropy": 5.581570816040039,
"epoch": 0.953934361407671,
"grad_norm": 0.98046875,
"learning_rate": 0.0004917696836588628,
"loss": 5.2869,
"mean_token_accuracy": 0.1871662840247154,
"num_tokens": 22100747.0,
"step": 9650
},
{
"entropy": 5.56132493019104,
"epoch": 0.9544286279161724,
"grad_norm": 1.0078125,
"learning_rate": 0.0004917602235189218,
"loss": 5.3,
"mean_token_accuracy": 0.19224132001399993,
"num_tokens": 22113559.0,
"step": 9655
},
{
"entropy": 5.719173574447632,
"epoch": 0.9549228944246738,
"grad_norm": 0.890625,
"learning_rate": 0.0004917507580466457,
"loss": 5.3076,
"mean_token_accuracy": 0.1880786120891571,
"num_tokens": 22124678.0,
"step": 9660
},
{
"entropy": 5.628768110275269,
"epoch": 0.9554171609331752,
"grad_norm": 0.890625,
"learning_rate": 0.0004917412872422674,
"loss": 5.2901,
"mean_token_accuracy": 0.19654264450073242,
"num_tokens": 22136462.0,
"step": 9665
},
{
"entropy": 5.594167852401734,
"epoch": 0.9559114274416766,
"grad_norm": 0.90234375,
"learning_rate": 0.0004917318111060198,
"loss": 5.3162,
"mean_token_accuracy": 0.1990877091884613,
"num_tokens": 22149101.0,
"step": 9670
},
{
"entropy": 5.610759210586548,
"epoch": 0.9564056939501779,
"grad_norm": 1.015625,
"learning_rate": 0.000491722329638136,
"loss": 5.2591,
"mean_token_accuracy": 0.1942766711115837,
"num_tokens": 22159574.0,
"step": 9675
},
{
"entropy": 5.644517946243286,
"epoch": 0.9568999604586793,
"grad_norm": 0.9375,
"learning_rate": 0.0004917128428388492,
"loss": 5.3283,
"mean_token_accuracy": 0.19246622622013093,
"num_tokens": 22172308.0,
"step": 9680
},
{
"entropy": 5.613810300827026,
"epoch": 0.9573942269671807,
"grad_norm": 0.953125,
"learning_rate": 0.0004917033507083929,
"loss": 5.3023,
"mean_token_accuracy": 0.18665868043899536,
"num_tokens": 22183714.0,
"step": 9685
},
{
"entropy": 5.5523786544799805,
"epoch": 0.9578884934756821,
"grad_norm": 0.97265625,
"learning_rate": 0.0004916938532470006,
"loss": 5.1484,
"mean_token_accuracy": 0.1981526121497154,
"num_tokens": 22194785.0,
"step": 9690
},
{
"entropy": 5.635073041915893,
"epoch": 0.9583827599841834,
"grad_norm": 0.92578125,
"learning_rate": 0.0004916843504549058,
"loss": 5.3288,
"mean_token_accuracy": 0.19318872094154357,
"num_tokens": 22205998.0,
"step": 9695
},
{
"entropy": 5.644573450088501,
"epoch": 0.9588770264926848,
"grad_norm": 0.953125,
"learning_rate": 0.0004916748423323424,
"loss": 5.2805,
"mean_token_accuracy": 0.19198890924453735,
"num_tokens": 22216114.0,
"step": 9700
},
{
"entropy": 5.603600311279297,
"epoch": 0.9593712930011863,
"grad_norm": 0.95703125,
"learning_rate": 0.0004916653288795443,
"loss": 5.2359,
"mean_token_accuracy": 0.20045812726020812,
"num_tokens": 22226793.0,
"step": 9705
},
{
"entropy": 5.645380735397339,
"epoch": 0.9598655595096877,
"grad_norm": 0.953125,
"learning_rate": 0.0004916558100967455,
"loss": 5.3387,
"mean_token_accuracy": 0.1936730772256851,
"num_tokens": 22237444.0,
"step": 9710
},
{
"entropy": 5.610105514526367,
"epoch": 0.960359826018189,
"grad_norm": 0.87890625,
"learning_rate": 0.0004916462859841801,
"loss": 5.2961,
"mean_token_accuracy": 0.19395934790372849,
"num_tokens": 22250704.0,
"step": 9715
},
{
"entropy": 5.620257472991943,
"epoch": 0.9608540925266904,
"grad_norm": 0.859375,
"learning_rate": 0.0004916367565420826,
"loss": 5.3199,
"mean_token_accuracy": 0.1955851584672928,
"num_tokens": 22264533.0,
"step": 9720
},
{
"entropy": 5.698529243469238,
"epoch": 0.9613483590351918,
"grad_norm": 0.93359375,
"learning_rate": 0.0004916272217706871,
"loss": 5.322,
"mean_token_accuracy": 0.1913663163781166,
"num_tokens": 22276510.0,
"step": 9725
},
{
"entropy": 5.689091062545776,
"epoch": 0.9618426255436932,
"grad_norm": 0.9375,
"learning_rate": 0.0004916176816702284,
"loss": 5.3527,
"mean_token_accuracy": 0.18925074338912964,
"num_tokens": 22288634.0,
"step": 9730
},
{
"entropy": 5.693050241470337,
"epoch": 0.9623368920521945,
"grad_norm": 1.0234375,
"learning_rate": 0.0004916081362409411,
"loss": 5.3378,
"mean_token_accuracy": 0.18890800178050995,
"num_tokens": 22299781.0,
"step": 9735
},
{
"entropy": 5.568251609802246,
"epoch": 0.9628311585606959,
"grad_norm": 1.1171875,
"learning_rate": 0.00049159858548306,
"loss": 5.1688,
"mean_token_accuracy": 0.20119997560977937,
"num_tokens": 22310666.0,
"step": 9740
},
{
"entropy": 5.659449911117553,
"epoch": 0.9633254250691973,
"grad_norm": 1.0078125,
"learning_rate": 0.0004915890293968202,
"loss": 5.3624,
"mean_token_accuracy": 0.18602459132671356,
"num_tokens": 22322811.0,
"step": 9745
},
{
"entropy": 5.66870346069336,
"epoch": 0.9638196915776986,
"grad_norm": 1.0234375,
"learning_rate": 0.0004915794679824567,
"loss": 5.3119,
"mean_token_accuracy": 0.1833906427025795,
"num_tokens": 22335017.0,
"step": 9750
},
{
"entropy": 5.587030363082886,
"epoch": 0.9643139580862001,
"grad_norm": 0.9375,
"learning_rate": 0.0004915699012402047,
"loss": 5.2489,
"mean_token_accuracy": 0.19727257192134856,
"num_tokens": 22346250.0,
"step": 9755
},
{
"entropy": 5.632735729217529,
"epoch": 0.9648082245947015,
"grad_norm": 1.0234375,
"learning_rate": 0.0004915603291702995,
"loss": 5.2771,
"mean_token_accuracy": 0.1907932013273239,
"num_tokens": 22357436.0,
"step": 9760
},
{
"entropy": 5.617962598800659,
"epoch": 0.9653024911032029,
"grad_norm": 1.09375,
"learning_rate": 0.0004915507517729765,
"loss": 5.2443,
"mean_token_accuracy": 0.19969998449087142,
"num_tokens": 22368312.0,
"step": 9765
},
{
"entropy": 5.644787311553955,
"epoch": 0.9657967576117042,
"grad_norm": 0.9609375,
"learning_rate": 0.0004915411690484715,
"loss": 5.2818,
"mean_token_accuracy": 0.19088573455810548,
"num_tokens": 22378659.0,
"step": 9770
},
{
"entropy": 5.621055364608765,
"epoch": 0.9662910241202056,
"grad_norm": 1.0078125,
"learning_rate": 0.0004915315809970202,
"loss": 5.2681,
"mean_token_accuracy": 0.19066745191812515,
"num_tokens": 22390642.0,
"step": 9775
},
{
"entropy": 5.682119226455688,
"epoch": 0.966785290628707,
"grad_norm": 0.96484375,
"learning_rate": 0.0004915219876188583,
"loss": 5.3648,
"mean_token_accuracy": 0.1840744897723198,
"num_tokens": 22402484.0,
"step": 9780
},
{
"entropy": 5.640901899337768,
"epoch": 0.9672795571372084,
"grad_norm": 0.9609375,
"learning_rate": 0.0004915123889142221,
"loss": 5.2584,
"mean_token_accuracy": 0.19494434148073198,
"num_tokens": 22413315.0,
"step": 9785
},
{
"entropy": 5.623419761657715,
"epoch": 0.9677738236457097,
"grad_norm": 0.9375,
"learning_rate": 0.0004915027848833472,
"loss": 5.3216,
"mean_token_accuracy": 0.18624964654445647,
"num_tokens": 22425123.0,
"step": 9790
},
{
"entropy": 5.6225732326507565,
"epoch": 0.9682680901542111,
"grad_norm": 1.078125,
"learning_rate": 0.0004914931755264705,
"loss": 5.2282,
"mean_token_accuracy": 0.19590737521648408,
"num_tokens": 22436437.0,
"step": 9795
},
{
"entropy": 5.639229583740234,
"epoch": 0.9687623566627125,
"grad_norm": 1.0703125,
"learning_rate": 0.000491483560843828,
"loss": 5.205,
"mean_token_accuracy": 0.2007865935564041,
"num_tokens": 22446677.0,
"step": 9800
},
{
"entropy": 5.560932874679565,
"epoch": 0.969256623171214,
"grad_norm": 0.97265625,
"learning_rate": 0.0004914739408356563,
"loss": 5.2763,
"mean_token_accuracy": 0.19748343378305436,
"num_tokens": 22457597.0,
"step": 9805
},
{
"entropy": 5.670977163314819,
"epoch": 0.9697508896797153,
"grad_norm": 1.0625,
"learning_rate": 0.0004914643155021921,
"loss": 5.3573,
"mean_token_accuracy": 0.19744355827569962,
"num_tokens": 22469715.0,
"step": 9810
},
{
"entropy": 5.6480104446411135,
"epoch": 0.9702451561882167,
"grad_norm": 1.0078125,
"learning_rate": 0.0004914546848436722,
"loss": 5.311,
"mean_token_accuracy": 0.1916728511452675,
"num_tokens": 22479840.0,
"step": 9815
},
{
"entropy": 5.5793132305145265,
"epoch": 0.9707394226967181,
"grad_norm": 0.95703125,
"learning_rate": 0.0004914450488603334,
"loss": 5.2289,
"mean_token_accuracy": 0.1986646518111229,
"num_tokens": 22491017.0,
"step": 9820
},
{
"entropy": 5.636435270309448,
"epoch": 0.9712336892052195,
"grad_norm": 0.984375,
"learning_rate": 0.0004914354075524129,
"loss": 5.2268,
"mean_token_accuracy": 0.19888969659805297,
"num_tokens": 22502308.0,
"step": 9825
},
{
"entropy": 5.500595664978027,
"epoch": 0.9717279557137208,
"grad_norm": 1.0234375,
"learning_rate": 0.0004914257609201477,
"loss": 5.1699,
"mean_token_accuracy": 0.20340583026409148,
"num_tokens": 22513090.0,
"step": 9830
},
{
"entropy": 5.593112897872925,
"epoch": 0.9722222222222222,
"grad_norm": 1.0703125,
"learning_rate": 0.0004914161089637752,
"loss": 5.2338,
"mean_token_accuracy": 0.19923784583806992,
"num_tokens": 22523391.0,
"step": 9835
},
{
"entropy": 5.61227388381958,
"epoch": 0.9727164887307236,
"grad_norm": 0.8984375,
"learning_rate": 0.0004914064516835329,
"loss": 5.2315,
"mean_token_accuracy": 0.203101547062397,
"num_tokens": 22536038.0,
"step": 9840
},
{
"entropy": 5.61104736328125,
"epoch": 0.973210755239225,
"grad_norm": 0.8984375,
"learning_rate": 0.0004913967890796583,
"loss": 5.2725,
"mean_token_accuracy": 0.1981385678052902,
"num_tokens": 22546897.0,
"step": 9845
},
{
"entropy": 5.599612712860107,
"epoch": 0.9737050217477263,
"grad_norm": 0.93359375,
"learning_rate": 0.0004913871211523891,
"loss": 5.3188,
"mean_token_accuracy": 0.19291321337223052,
"num_tokens": 22558240.0,
"step": 9850
},
{
"entropy": 5.602342844009399,
"epoch": 0.9741992882562278,
"grad_norm": 0.98046875,
"learning_rate": 0.0004913774479019633,
"loss": 5.2684,
"mean_token_accuracy": 0.1954142212867737,
"num_tokens": 22570540.0,
"step": 9855
},
{
"entropy": 5.670391416549682,
"epoch": 0.9746935547647292,
"grad_norm": 0.9140625,
"learning_rate": 0.0004913677693286185,
"loss": 5.3601,
"mean_token_accuracy": 0.1852197229862213,
"num_tokens": 22582657.0,
"step": 9860
},
{
"entropy": 5.7341156005859375,
"epoch": 0.9751878212732306,
"grad_norm": 1.0859375,
"learning_rate": 0.0004913580854325932,
"loss": 5.3929,
"mean_token_accuracy": 0.19256402552127838,
"num_tokens": 22593445.0,
"step": 9865
},
{
"entropy": 5.620249462127686,
"epoch": 0.9756820877817319,
"grad_norm": 0.98046875,
"learning_rate": 0.0004913483962141253,
"loss": 5.246,
"mean_token_accuracy": 0.1965990960597992,
"num_tokens": 22604631.0,
"step": 9870
},
{
"entropy": 5.5677050113677975,
"epoch": 0.9761763542902333,
"grad_norm": 1.0625,
"learning_rate": 0.0004913387016734533,
"loss": 5.2528,
"mean_token_accuracy": 0.19343892782926558,
"num_tokens": 22615130.0,
"step": 9875
},
{
"entropy": 5.589069271087647,
"epoch": 0.9766706207987347,
"grad_norm": 0.90625,
"learning_rate": 0.0004913290018108157,
"loss": 5.2618,
"mean_token_accuracy": 0.19410673528909683,
"num_tokens": 22627372.0,
"step": 9880
},
{
"entropy": 5.683375072479248,
"epoch": 0.977164887307236,
"grad_norm": 0.99609375,
"learning_rate": 0.0004913192966264511,
"loss": 5.3412,
"mean_token_accuracy": 0.1924360066652298,
"num_tokens": 22640354.0,
"step": 9885
},
{
"entropy": 5.677717018127441,
"epoch": 0.9776591538157374,
"grad_norm": 1.015625,
"learning_rate": 0.0004913095861205982,
"loss": 5.3407,
"mean_token_accuracy": 0.18949446380138396,
"num_tokens": 22651717.0,
"step": 9890
},
{
"entropy": 5.691273498535156,
"epoch": 0.9781534203242388,
"grad_norm": 0.91796875,
"learning_rate": 0.0004912998702934961,
"loss": 5.3908,
"mean_token_accuracy": 0.18768298327922822,
"num_tokens": 22663334.0,
"step": 9895
},
{
"entropy": 5.662432718276977,
"epoch": 0.9786476868327402,
"grad_norm": 0.89453125,
"learning_rate": 0.0004912901491453835,
"loss": 5.2677,
"mean_token_accuracy": 0.19687706381082534,
"num_tokens": 22675894.0,
"step": 9900
},
{
"entropy": 5.522920179367065,
"epoch": 0.9791419533412415,
"grad_norm": 0.8984375,
"learning_rate": 0.0004912804226764997,
"loss": 5.2091,
"mean_token_accuracy": 0.205456380546093,
"num_tokens": 22687861.0,
"step": 9905
},
{
"entropy": 5.636275720596314,
"epoch": 0.979636219849743,
"grad_norm": 0.92578125,
"learning_rate": 0.0004912706908870838,
"loss": 5.24,
"mean_token_accuracy": 0.19504302740097046,
"num_tokens": 22698346.0,
"step": 9910
},
{
"entropy": 5.6292554378509525,
"epoch": 0.9801304863582444,
"grad_norm": 0.9609375,
"learning_rate": 0.0004912609537773755,
"loss": 5.3343,
"mean_token_accuracy": 0.18624177128076552,
"num_tokens": 22710665.0,
"step": 9915
},
{
"entropy": 5.641499662399292,
"epoch": 0.9806247528667458,
"grad_norm": 1.0390625,
"learning_rate": 0.0004912512113476142,
"loss": 5.2392,
"mean_token_accuracy": 0.2029682219028473,
"num_tokens": 22720989.0,
"step": 9920
},
{
"entropy": 5.585738229751587,
"epoch": 0.9811190193752471,
"grad_norm": 0.98046875,
"learning_rate": 0.0004912414635980395,
"loss": 5.2625,
"mean_token_accuracy": 0.19947191178798676,
"num_tokens": 22732265.0,
"step": 9925
},
{
"entropy": 5.591073656082154,
"epoch": 0.9816132858837485,
"grad_norm": 0.921875,
"learning_rate": 0.0004912317105288912,
"loss": 5.2405,
"mean_token_accuracy": 0.19825671017169952,
"num_tokens": 22742435.0,
"step": 9930
},
{
"entropy": 5.551272344589234,
"epoch": 0.9821075523922499,
"grad_norm": 0.94140625,
"learning_rate": 0.0004912219521404094,
"loss": 5.2382,
"mean_token_accuracy": 0.1958669289946556,
"num_tokens": 22753750.0,
"step": 9935
},
{
"entropy": 5.653229665756226,
"epoch": 0.9826018189007513,
"grad_norm": 0.90625,
"learning_rate": 0.000491212188432834,
"loss": 5.2873,
"mean_token_accuracy": 0.19283485859632493,
"num_tokens": 22765549.0,
"step": 9940
},
{
"entropy": 5.6413239479064945,
"epoch": 0.9830960854092526,
"grad_norm": 0.9375,
"learning_rate": 0.0004912024194064052,
"loss": 5.3359,
"mean_token_accuracy": 0.18766363263130187,
"num_tokens": 22777882.0,
"step": 9945
},
{
"entropy": 5.700936794281006,
"epoch": 0.983590351917754,
"grad_norm": 0.94140625,
"learning_rate": 0.0004911926450613633,
"loss": 5.3722,
"mean_token_accuracy": 0.18520065844058992,
"num_tokens": 22789671.0,
"step": 9950
},
{
"entropy": 5.63926887512207,
"epoch": 0.9840846184262554,
"grad_norm": 1.0625,
"learning_rate": 0.0004911828653979488,
"loss": 5.3233,
"mean_token_accuracy": 0.18694156110286714,
"num_tokens": 22801330.0,
"step": 9955
},
{
"entropy": 5.620159912109375,
"epoch": 0.9845788849347569,
"grad_norm": 0.93359375,
"learning_rate": 0.0004911730804164024,
"loss": 5.3538,
"mean_token_accuracy": 0.18854619711637496,
"num_tokens": 22813998.0,
"step": 9960
},
{
"entropy": 5.630801439285278,
"epoch": 0.9850731514432582,
"grad_norm": 1.03125,
"learning_rate": 0.0004911632901169645,
"loss": 5.2735,
"mean_token_accuracy": 0.19400959312915803,
"num_tokens": 22825048.0,
"step": 9965
},
{
"entropy": 5.540067911148071,
"epoch": 0.9855674179517596,
"grad_norm": 1.109375,
"learning_rate": 0.0004911534944998762,
"loss": 5.2013,
"mean_token_accuracy": 0.19872192144393921,
"num_tokens": 22835450.0,
"step": 9970
},
{
"entropy": 5.588812398910522,
"epoch": 0.986061684460261,
"grad_norm": 0.828125,
"learning_rate": 0.0004911436935653785,
"loss": 5.2288,
"mean_token_accuracy": 0.19831462055444718,
"num_tokens": 22847320.0,
"step": 9975
},
{
"entropy": 5.633582639694214,
"epoch": 0.9865559509687624,
"grad_norm": 0.97265625,
"learning_rate": 0.0004911338873137122,
"loss": 5.2059,
"mean_token_accuracy": 0.20011822432279586,
"num_tokens": 22857676.0,
"step": 9980
},
{
"entropy": 5.538009405136108,
"epoch": 0.9870502174772637,
"grad_norm": 0.87890625,
"learning_rate": 0.000491124075745119,
"loss": 5.2197,
"mean_token_accuracy": 0.1993396669626236,
"num_tokens": 22869767.0,
"step": 9985
},
{
"entropy": 5.63607702255249,
"epoch": 0.9875444839857651,
"grad_norm": 0.99609375,
"learning_rate": 0.0004911142588598399,
"loss": 5.322,
"mean_token_accuracy": 0.1940407007932663,
"num_tokens": 22880096.0,
"step": 9990
},
{
"entropy": 5.679059743881226,
"epoch": 0.9880387504942665,
"grad_norm": 0.921875,
"learning_rate": 0.0004911044366581165,
"loss": 5.2406,
"mean_token_accuracy": 0.19356839060783387,
"num_tokens": 22892436.0,
"step": 9995
},
{
"entropy": 5.643694686889648,
"epoch": 0.9885330170027679,
"grad_norm": 1.0390625,
"learning_rate": 0.0004910946091401903,
"loss": 5.2682,
"mean_token_accuracy": 0.19104681611061097,
"num_tokens": 22904641.0,
"step": 10000
},
{
"entropy": 5.603832197189331,
"epoch": 0.9890272835112692,
"grad_norm": 0.95703125,
"learning_rate": 0.0004910847763063033,
"loss": 5.1422,
"mean_token_accuracy": 0.2031159982085228,
"num_tokens": 22915372.0,
"step": 10005
},
{
"entropy": 5.66643443107605,
"epoch": 0.9895215500197707,
"grad_norm": 0.94921875,
"learning_rate": 0.0004910749381566973,
"loss": 5.3603,
"mean_token_accuracy": 0.18446232229471207,
"num_tokens": 22926583.0,
"step": 10010
},
{
"entropy": 5.631104850769043,
"epoch": 0.9900158165282721,
"grad_norm": 0.9296875,
"learning_rate": 0.0004910650946916143,
"loss": 5.2241,
"mean_token_accuracy": 0.19322517961263658,
"num_tokens": 22938103.0,
"step": 10015
},
{
"entropy": 5.585657596588135,
"epoch": 0.9905100830367735,
"grad_norm": 0.953125,
"learning_rate": 0.0004910552459112964,
"loss": 5.2766,
"mean_token_accuracy": 0.1936303660273552,
"num_tokens": 22949475.0,
"step": 10020
},
{
"entropy": 5.638175821304321,
"epoch": 0.9910043495452748,
"grad_norm": 0.87890625,
"learning_rate": 0.0004910453918159861,
"loss": 5.298,
"mean_token_accuracy": 0.19244821071624757,
"num_tokens": 22961404.0,
"step": 10025
},
{
"entropy": 5.607744884490967,
"epoch": 0.9914986160537762,
"grad_norm": 0.94140625,
"learning_rate": 0.0004910355324059255,
"loss": 5.2826,
"mean_token_accuracy": 0.19262792319059371,
"num_tokens": 22973009.0,
"step": 10030
},
{
"entropy": 5.633600568771362,
"epoch": 0.9919928825622776,
"grad_norm": 1.0859375,
"learning_rate": 0.0004910256676813572,
"loss": 5.307,
"mean_token_accuracy": 0.1948724642395973,
"num_tokens": 22985104.0,
"step": 10035
},
{
"entropy": 5.616547441482544,
"epoch": 0.992487149070779,
"grad_norm": 0.9375,
"learning_rate": 0.0004910157976425241,
"loss": 5.3107,
"mean_token_accuracy": 0.19048827290534973,
"num_tokens": 22997133.0,
"step": 10040
},
{
"entropy": 5.591722106933593,
"epoch": 0.9929814155792803,
"grad_norm": 1.0703125,
"learning_rate": 0.0004910059222896689,
"loss": 5.1533,
"mean_token_accuracy": 0.19974204003810883,
"num_tokens": 23007621.0,
"step": 10045
},
{
"entropy": 5.668904113769531,
"epoch": 0.9934756820877817,
"grad_norm": 1.0,
"learning_rate": 0.0004909960416230344,
"loss": 5.3152,
"mean_token_accuracy": 0.18902715295553207,
"num_tokens": 23018254.0,
"step": 10050
},
{
"entropy": 5.634682703018188,
"epoch": 0.9939699485962831,
"grad_norm": 1.0078125,
"learning_rate": 0.0004909861556428638,
"loss": 5.2231,
"mean_token_accuracy": 0.19833352118730546,
"num_tokens": 23030733.0,
"step": 10055
},
{
"entropy": 5.55250244140625,
"epoch": 0.9944642151047846,
"grad_norm": 0.921875,
"learning_rate": 0.0004909762643494003,
"loss": 5.2274,
"mean_token_accuracy": 0.19251906424760817,
"num_tokens": 23043086.0,
"step": 10060
},
{
"entropy": 5.711132621765136,
"epoch": 0.9949584816132859,
"grad_norm": 0.97265625,
"learning_rate": 0.0004909663677428872,
"loss": 5.4082,
"mean_token_accuracy": 0.18608732968568803,
"num_tokens": 23055281.0,
"step": 10065
},
{
"entropy": 5.595179414749145,
"epoch": 0.9954527481217873,
"grad_norm": 0.9765625,
"learning_rate": 0.0004909564658235678,
"loss": 5.2787,
"mean_token_accuracy": 0.19430323392152787,
"num_tokens": 23067332.0,
"step": 10070
},
{
"entropy": 5.593054342269897,
"epoch": 0.9959470146302887,
"grad_norm": 0.921875,
"learning_rate": 0.000490946558591686,
"loss": 5.2406,
"mean_token_accuracy": 0.19264515042304992,
"num_tokens": 23079278.0,
"step": 10075
},
{
"entropy": 5.626236009597778,
"epoch": 0.99644128113879,
"grad_norm": 0.97265625,
"learning_rate": 0.0004909366460474852,
"loss": 5.2717,
"mean_token_accuracy": 0.19360380321741105,
"num_tokens": 23092177.0,
"step": 10080
},
{
"entropy": 5.534682416915894,
"epoch": 0.9969355476472914,
"grad_norm": 0.875,
"learning_rate": 0.0004909267281912095,
"loss": 5.2485,
"mean_token_accuracy": 0.19558467268943786,
"num_tokens": 23103249.0,
"step": 10085
},
{
"entropy": 5.642769050598145,
"epoch": 0.9974298141557928,
"grad_norm": 0.95703125,
"learning_rate": 0.0004909168050231029,
"loss": 5.3038,
"mean_token_accuracy": 0.20087829679250718,
"num_tokens": 23113821.0,
"step": 10090
},
{
"entropy": 5.660932922363282,
"epoch": 0.9979240806642942,
"grad_norm": 0.96484375,
"learning_rate": 0.0004909068765434093,
"loss": 5.3064,
"mean_token_accuracy": 0.19209800511598588,
"num_tokens": 23124819.0,
"step": 10095
},
{
"entropy": 5.599535608291626,
"epoch": 0.9984183471727955,
"grad_norm": 1.0078125,
"learning_rate": 0.000490896942752373,
"loss": 5.3086,
"mean_token_accuracy": 0.19286416471004486,
"num_tokens": 23135590.0,
"step": 10100
},
{
"entropy": 5.655676651000976,
"epoch": 0.9989126136812969,
"grad_norm": 0.90234375,
"learning_rate": 0.0004908870036502384,
"loss": 5.2696,
"mean_token_accuracy": 0.19640028327703477,
"num_tokens": 23147290.0,
"step": 10105
},
{
"entropy": 5.683337879180908,
"epoch": 0.9994068801897983,
"grad_norm": 1.09375,
"learning_rate": 0.0004908770592372501,
"loss": 5.2639,
"mean_token_accuracy": 0.1957457184791565,
"num_tokens": 23158276.0,
"step": 10110
},
{
"entropy": 5.601388072967529,
"epoch": 0.9999011466982998,
"grad_norm": 0.9921875,
"learning_rate": 0.0004908671095136526,
"loss": 5.2377,
"mean_token_accuracy": 0.19549649953842163,
"num_tokens": 23169174.0,
"step": 10115
},
{
"entropy": 5.6100077629089355,
"epoch": 1.0003954132068011,
"grad_norm": 0.890625,
"learning_rate": 0.0004908571544796908,
"loss": 5.2574,
"mean_token_accuracy": 0.19631705582141876,
"num_tokens": 23180377.0,
"step": 10120
},
{
"entropy": 5.632832956314087,
"epoch": 1.0008896797153024,
"grad_norm": 1.0,
"learning_rate": 0.0004908471941356094,
"loss": 5.1885,
"mean_token_accuracy": 0.2003858655691147,
"num_tokens": 23191158.0,
"step": 10125
},
{
"entropy": 5.654829835891723,
"epoch": 1.001383946223804,
"grad_norm": 0.9921875,
"learning_rate": 0.0004908372284816537,
"loss": 5.2172,
"mean_token_accuracy": 0.20165897458791732,
"num_tokens": 23201956.0,
"step": 10130
},
{
"entropy": 5.594053077697754,
"epoch": 1.0018782127323052,
"grad_norm": 0.91015625,
"learning_rate": 0.0004908272575180685,
"loss": 5.1738,
"mean_token_accuracy": 0.19583955705165862,
"num_tokens": 23213449.0,
"step": 10135
},
{
"entropy": 5.6795391082763675,
"epoch": 1.0023724792408066,
"grad_norm": 0.9296875,
"learning_rate": 0.0004908172812450993,
"loss": 5.2656,
"mean_token_accuracy": 0.1900208815932274,
"num_tokens": 23226514.0,
"step": 10140
},
{
"entropy": 5.6067225456237795,
"epoch": 1.0028667457493081,
"grad_norm": 0.9921875,
"learning_rate": 0.0004908072996629917,
"loss": 5.1692,
"mean_token_accuracy": 0.19555890709161758,
"num_tokens": 23239322.0,
"step": 10145
},
{
"entropy": 5.627415370941162,
"epoch": 1.0033610122578094,
"grad_norm": 0.97265625,
"learning_rate": 0.0004907973127719909,
"loss": 5.1992,
"mean_token_accuracy": 0.19606511145830155,
"num_tokens": 23250990.0,
"step": 10150
},
{
"entropy": 5.61159496307373,
"epoch": 1.0038552787663109,
"grad_norm": 1.015625,
"learning_rate": 0.0004907873205723427,
"loss": 5.2537,
"mean_token_accuracy": 0.188396155834198,
"num_tokens": 23261630.0,
"step": 10155
},
{
"entropy": 5.646563959121704,
"epoch": 1.0043495452748121,
"grad_norm": 0.91015625,
"learning_rate": 0.000490777323064293,
"loss": 5.2122,
"mean_token_accuracy": 0.19345839321613312,
"num_tokens": 23272823.0,
"step": 10160
},
{
"entropy": 5.514635515213013,
"epoch": 1.0048438117833136,
"grad_norm": 0.98046875,
"learning_rate": 0.0004907673202480876,
"loss": 5.0301,
"mean_token_accuracy": 0.22162040919065476,
"num_tokens": 23282827.0,
"step": 10165
},
{
"entropy": 5.577254724502564,
"epoch": 1.0053380782918149,
"grad_norm": 0.96875,
"learning_rate": 0.0004907573121239727,
"loss": 5.206,
"mean_token_accuracy": 0.19773707687854766,
"num_tokens": 23295359.0,
"step": 10170
},
{
"entropy": 5.637523794174195,
"epoch": 1.0058323448003164,
"grad_norm": 0.9765625,
"learning_rate": 0.0004907472986921945,
"loss": 5.1784,
"mean_token_accuracy": 0.2024553045630455,
"num_tokens": 23305932.0,
"step": 10175
},
{
"entropy": 5.5606465339660645,
"epoch": 1.0063266113088176,
"grad_norm": 0.9140625,
"learning_rate": 0.0004907372799529992,
"loss": 5.174,
"mean_token_accuracy": 0.19294115006923676,
"num_tokens": 23317856.0,
"step": 10180
},
{
"entropy": 5.588295841217041,
"epoch": 1.006820877817319,
"grad_norm": 1.078125,
"learning_rate": 0.0004907272559066334,
"loss": 5.0734,
"mean_token_accuracy": 0.20991285741329194,
"num_tokens": 23328071.0,
"step": 10185
},
{
"entropy": 5.58222074508667,
"epoch": 1.0073151443258206,
"grad_norm": 0.9609375,
"learning_rate": 0.0004907172265533436,
"loss": 5.2413,
"mean_token_accuracy": 0.19638827741146087,
"num_tokens": 23339601.0,
"step": 10190
},
{
"entropy": 5.644859504699707,
"epoch": 1.0078094108343219,
"grad_norm": 0.96484375,
"learning_rate": 0.0004907071918933766,
"loss": 5.2003,
"mean_token_accuracy": 0.1973765879869461,
"num_tokens": 23351672.0,
"step": 10195
},
{
"entropy": 5.668191289901733,
"epoch": 1.0083036773428233,
"grad_norm": 1.0078125,
"learning_rate": 0.0004906971519269793,
"loss": 5.2436,
"mean_token_accuracy": 0.19607312530279158,
"num_tokens": 23361684.0,
"step": 10200
},
{
"entropy": 5.531037521362305,
"epoch": 1.0087979438513246,
"grad_norm": 0.91015625,
"learning_rate": 0.0004906871066543984,
"loss": 5.2131,
"mean_token_accuracy": 0.1975875973701477,
"num_tokens": 23373969.0,
"step": 10205
},
{
"entropy": 5.675751686096191,
"epoch": 1.009292210359826,
"grad_norm": 0.91015625,
"learning_rate": 0.0004906770560758813,
"loss": 5.2363,
"mean_token_accuracy": 0.18354198038578035,
"num_tokens": 23384747.0,
"step": 10210
},
{
"entropy": 5.6887959957122805,
"epoch": 1.0097864768683273,
"grad_norm": 0.94140625,
"learning_rate": 0.0004906670001916752,
"loss": 5.2056,
"mean_token_accuracy": 0.19636847078800201,
"num_tokens": 23395900.0,
"step": 10215
},
{
"entropy": 5.619274044036866,
"epoch": 1.0102807433768288,
"grad_norm": 0.859375,
"learning_rate": 0.0004906569390020274,
"loss": 5.1977,
"mean_token_accuracy": 0.19952213168144226,
"num_tokens": 23408912.0,
"step": 10220
},
{
"entropy": 5.606116819381714,
"epoch": 1.01077500988533,
"grad_norm": 0.98828125,
"learning_rate": 0.0004906468725071855,
"loss": 5.1554,
"mean_token_accuracy": 0.2012458026409149,
"num_tokens": 23419042.0,
"step": 10225
},
{
"entropy": 5.6188249588012695,
"epoch": 1.0112692763938316,
"grad_norm": 0.90234375,
"learning_rate": 0.000490636800707397,
"loss": 5.191,
"mean_token_accuracy": 0.19930345565080643,
"num_tokens": 23430861.0,
"step": 10230
},
{
"entropy": 5.701479864120484,
"epoch": 1.0117635429023328,
"grad_norm": 0.921875,
"learning_rate": 0.0004906267236029099,
"loss": 5.3018,
"mean_token_accuracy": 0.18746026307344438,
"num_tokens": 23443489.0,
"step": 10235
},
{
"entropy": 5.602195072174072,
"epoch": 1.0122578094108343,
"grad_norm": 0.96875,
"learning_rate": 0.0004906166411939718,
"loss": 5.2266,
"mean_token_accuracy": 0.19687227606773378,
"num_tokens": 23454327.0,
"step": 10240
},
{
"entropy": 5.554819965362549,
"epoch": 1.0127520759193358,
"grad_norm": 0.88671875,
"learning_rate": 0.0004906065534808309,
"loss": 5.0982,
"mean_token_accuracy": 0.2068428263068199,
"num_tokens": 23466684.0,
"step": 10245
},
{
"entropy": 5.595309782028198,
"epoch": 1.013246342427837,
"grad_norm": 0.984375,
"learning_rate": 0.0004905964604637353,
"loss": 5.2181,
"mean_token_accuracy": 0.20004372745752336,
"num_tokens": 23478347.0,
"step": 10250
},
{
"entropy": 5.593182468414307,
"epoch": 1.0137406089363385,
"grad_norm": 0.91796875,
"learning_rate": 0.0004905863621429334,
"loss": 5.0971,
"mean_token_accuracy": 0.20330906361341478,
"num_tokens": 23489588.0,
"step": 10255
},
{
"entropy": 5.550411128997803,
"epoch": 1.0142348754448398,
"grad_norm": 0.94921875,
"learning_rate": 0.0004905762585186735,
"loss": 5.1665,
"mean_token_accuracy": 0.20081121176481248,
"num_tokens": 23501347.0,
"step": 10260
},
{
"entropy": 5.660239887237549,
"epoch": 1.0147291419533413,
"grad_norm": 0.91015625,
"learning_rate": 0.0004905661495912042,
"loss": 5.263,
"mean_token_accuracy": 0.1940949246287346,
"num_tokens": 23513036.0,
"step": 10265
},
{
"entropy": 5.575650691986084,
"epoch": 1.0152234084618426,
"grad_norm": 0.99609375,
"learning_rate": 0.0004905560353607742,
"loss": 5.1321,
"mean_token_accuracy": 0.20131876319646835,
"num_tokens": 23524166.0,
"step": 10270
},
{
"entropy": 5.595976495742798,
"epoch": 1.015717674970344,
"grad_norm": 0.94140625,
"learning_rate": 0.0004905459158276323,
"loss": 5.1915,
"mean_token_accuracy": 0.19659801125526427,
"num_tokens": 23536638.0,
"step": 10275
},
{
"entropy": 5.615958309173584,
"epoch": 1.0162119414788453,
"grad_norm": 0.98828125,
"learning_rate": 0.0004905357909920275,
"loss": 5.2279,
"mean_token_accuracy": 0.19983913749456406,
"num_tokens": 23548567.0,
"step": 10280
},
{
"entropy": 5.605298328399658,
"epoch": 1.0167062079873468,
"grad_norm": 1.03125,
"learning_rate": 0.0004905256608542086,
"loss": 5.15,
"mean_token_accuracy": 0.20143326818943025,
"num_tokens": 23560108.0,
"step": 10285
},
{
"entropy": 5.591742372512817,
"epoch": 1.0172004744958483,
"grad_norm": 0.95703125,
"learning_rate": 0.0004905155254144251,
"loss": 5.1603,
"mean_token_accuracy": 0.203885780274868,
"num_tokens": 23571542.0,
"step": 10290
},
{
"entropy": 5.534454488754273,
"epoch": 1.0176947410043495,
"grad_norm": 0.95703125,
"learning_rate": 0.0004905053846729263,
"loss": 5.1261,
"mean_token_accuracy": 0.20065390169620514,
"num_tokens": 23582200.0,
"step": 10295
},
{
"entropy": 5.639520168304443,
"epoch": 1.018189007512851,
"grad_norm": 0.95703125,
"learning_rate": 0.0004904952386299616,
"loss": 5.2329,
"mean_token_accuracy": 0.19528298825025558,
"num_tokens": 23594530.0,
"step": 10300
},
{
"entropy": 5.588271617889404,
"epoch": 1.0186832740213523,
"grad_norm": 0.93359375,
"learning_rate": 0.0004904850872857805,
"loss": 5.0556,
"mean_token_accuracy": 0.2131606474518776,
"num_tokens": 23605697.0,
"step": 10305
},
{
"entropy": 5.618853902816772,
"epoch": 1.0191775405298538,
"grad_norm": 0.98828125,
"learning_rate": 0.000490474930640633,
"loss": 5.2399,
"mean_token_accuracy": 0.1965971902012825,
"num_tokens": 23615730.0,
"step": 10310
},
{
"entropy": 5.566145706176758,
"epoch": 1.019671807038355,
"grad_norm": 0.93359375,
"learning_rate": 0.0004904647686947687,
"loss": 5.1453,
"mean_token_accuracy": 0.20490238815546036,
"num_tokens": 23627242.0,
"step": 10315
},
{
"entropy": 5.5099633693695065,
"epoch": 1.0201660735468565,
"grad_norm": 0.98046875,
"learning_rate": 0.0004904546014484376,
"loss": 5.0786,
"mean_token_accuracy": 0.20741736739873887,
"num_tokens": 23639450.0,
"step": 10320
},
{
"entropy": 5.589915180206299,
"epoch": 1.0206603400553578,
"grad_norm": 0.8984375,
"learning_rate": 0.0004904444289018898,
"loss": 5.1968,
"mean_token_accuracy": 0.2016470402479172,
"num_tokens": 23651154.0,
"step": 10325
},
{
"entropy": 5.63874831199646,
"epoch": 1.0211546065638593,
"grad_norm": 0.80859375,
"learning_rate": 0.0004904342510553758,
"loss": 5.2611,
"mean_token_accuracy": 0.1933414399623871,
"num_tokens": 23663996.0,
"step": 10330
},
{
"entropy": 5.625178718566895,
"epoch": 1.0216488730723605,
"grad_norm": 0.97265625,
"learning_rate": 0.0004904240679091458,
"loss": 5.1713,
"mean_token_accuracy": 0.19972514361143112,
"num_tokens": 23676270.0,
"step": 10335
},
{
"entropy": 5.663323974609375,
"epoch": 1.022143139580862,
"grad_norm": 0.91796875,
"learning_rate": 0.0004904138794634502,
"loss": 5.1702,
"mean_token_accuracy": 0.1995392084121704,
"num_tokens": 23687483.0,
"step": 10340
},
{
"entropy": 5.643520069122315,
"epoch": 1.0226374060893635,
"grad_norm": 1.0859375,
"learning_rate": 0.0004904036857185398,
"loss": 5.2559,
"mean_token_accuracy": 0.195101498067379,
"num_tokens": 23697956.0,
"step": 10345
},
{
"entropy": 5.597309541702271,
"epoch": 1.0231316725978647,
"grad_norm": 1.0078125,
"learning_rate": 0.0004903934866746653,
"loss": 5.1864,
"mean_token_accuracy": 0.2016719549894333,
"num_tokens": 23709029.0,
"step": 10350
},
{
"entropy": 5.599404954910279,
"epoch": 1.0236259391063662,
"grad_norm": 0.9609375,
"learning_rate": 0.0004903832823320777,
"loss": 5.1392,
"mean_token_accuracy": 0.2061514213681221,
"num_tokens": 23719792.0,
"step": 10355
},
{
"entropy": 5.598851919174194,
"epoch": 1.0241202056148675,
"grad_norm": 0.984375,
"learning_rate": 0.0004903730726910278,
"loss": 5.2021,
"mean_token_accuracy": 0.1986728236079216,
"num_tokens": 23730201.0,
"step": 10360
},
{
"entropy": 5.559210872650146,
"epoch": 1.024614472123369,
"grad_norm": 0.89453125,
"learning_rate": 0.000490362857751767,
"loss": 5.1417,
"mean_token_accuracy": 0.2023387923836708,
"num_tokens": 23741743.0,
"step": 10365
},
{
"entropy": 5.614955711364746,
"epoch": 1.0251087386318702,
"grad_norm": 0.8671875,
"learning_rate": 0.0004903526375145465,
"loss": 5.1081,
"mean_token_accuracy": 0.20339439064264297,
"num_tokens": 23753879.0,
"step": 10370
},
{
"entropy": 5.597322940826416,
"epoch": 1.0256030051403717,
"grad_norm": 1.015625,
"learning_rate": 0.0004903424119796176,
"loss": 5.2633,
"mean_token_accuracy": 0.1919170618057251,
"num_tokens": 23765062.0,
"step": 10375
},
{
"entropy": 5.607200002670288,
"epoch": 1.026097271648873,
"grad_norm": 1.0078125,
"learning_rate": 0.0004903321811472321,
"loss": 5.2295,
"mean_token_accuracy": 0.20121922791004182,
"num_tokens": 23777261.0,
"step": 10380
},
{
"entropy": 5.636020612716675,
"epoch": 1.0265915381573745,
"grad_norm": 0.953125,
"learning_rate": 0.0004903219450176413,
"loss": 5.2171,
"mean_token_accuracy": 0.19775684773921967,
"num_tokens": 23789247.0,
"step": 10385
},
{
"entropy": 5.583631610870361,
"epoch": 1.0270858046658757,
"grad_norm": 1.046875,
"learning_rate": 0.0004903117035910975,
"loss": 5.1406,
"mean_token_accuracy": 0.2014937534928322,
"num_tokens": 23798981.0,
"step": 10390
},
{
"entropy": 5.619787216186523,
"epoch": 1.0275800711743772,
"grad_norm": 0.89453125,
"learning_rate": 0.0004903014568678523,
"loss": 5.2304,
"mean_token_accuracy": 0.19623916298151017,
"num_tokens": 23810856.0,
"step": 10395
},
{
"entropy": 5.6564538955688475,
"epoch": 1.0280743376828787,
"grad_norm": 0.93359375,
"learning_rate": 0.0004902912048481579,
"loss": 5.2302,
"mean_token_accuracy": 0.19382828772068023,
"num_tokens": 23821544.0,
"step": 10400
},
{
"entropy": 5.598402595520019,
"epoch": 1.02856860419138,
"grad_norm": 1.0625,
"learning_rate": 0.0004902809475322664,
"loss": 5.1475,
"mean_token_accuracy": 0.19486696869134904,
"num_tokens": 23831972.0,
"step": 10405
},
{
"entropy": 5.6284221649169925,
"epoch": 1.0290628706998814,
"grad_norm": 0.91015625,
"learning_rate": 0.0004902706849204302,
"loss": 5.1658,
"mean_token_accuracy": 0.20296854823827742,
"num_tokens": 23843451.0,
"step": 10410
},
{
"entropy": 5.6014739036560055,
"epoch": 1.0295571372083827,
"grad_norm": 1.0625,
"learning_rate": 0.0004902604170129017,
"loss": 5.185,
"mean_token_accuracy": 0.19674879163503647,
"num_tokens": 23854060.0,
"step": 10415
},
{
"entropy": 5.678252506256103,
"epoch": 1.0300514037168842,
"grad_norm": 0.96484375,
"learning_rate": 0.0004902501438099337,
"loss": 5.2992,
"mean_token_accuracy": 0.19259535819292067,
"num_tokens": 23865722.0,
"step": 10420
},
{
"entropy": 5.655801010131836,
"epoch": 1.0305456702253855,
"grad_norm": 1.0546875,
"learning_rate": 0.0004902398653117787,
"loss": 5.2278,
"mean_token_accuracy": 0.19363904297351836,
"num_tokens": 23875888.0,
"step": 10425
},
{
"entropy": 5.618904638290405,
"epoch": 1.031039936733887,
"grad_norm": 0.91796875,
"learning_rate": 0.0004902295815186896,
"loss": 5.2453,
"mean_token_accuracy": 0.19405853599309922,
"num_tokens": 23887816.0,
"step": 10430
},
{
"entropy": 5.642099857330322,
"epoch": 1.0315342032423882,
"grad_norm": 0.87109375,
"learning_rate": 0.0004902192924309193,
"loss": 5.1695,
"mean_token_accuracy": 0.1990288257598877,
"num_tokens": 23900468.0,
"step": 10435
},
{
"entropy": 5.6218345165252686,
"epoch": 1.0320284697508897,
"grad_norm": 0.921875,
"learning_rate": 0.0004902089980487211,
"loss": 5.1589,
"mean_token_accuracy": 0.19757075756788253,
"num_tokens": 23912086.0,
"step": 10440
},
{
"entropy": 5.593123865127564,
"epoch": 1.032522736259391,
"grad_norm": 1.078125,
"learning_rate": 0.0004901986983723483,
"loss": 5.1937,
"mean_token_accuracy": 0.19984557926654817,
"num_tokens": 23923262.0,
"step": 10445
},
{
"entropy": 5.5935694694519045,
"epoch": 1.0330170027678924,
"grad_norm": 0.9375,
"learning_rate": 0.000490188393402054,
"loss": 5.124,
"mean_token_accuracy": 0.20233773440122604,
"num_tokens": 23934660.0,
"step": 10450
},
{
"entropy": 5.571173286437988,
"epoch": 1.033511269276394,
"grad_norm": 0.9765625,
"learning_rate": 0.0004901780831380919,
"loss": 5.145,
"mean_token_accuracy": 0.20368053168058395,
"num_tokens": 23945822.0,
"step": 10455
},
{
"entropy": 5.578071260452271,
"epoch": 1.0340055357848952,
"grad_norm": 0.88671875,
"learning_rate": 0.0004901677675807155,
"loss": 5.2182,
"mean_token_accuracy": 0.19720763415098191,
"num_tokens": 23957312.0,
"step": 10460
},
{
"entropy": 5.647917366027832,
"epoch": 1.0344998022933967,
"grad_norm": 0.94140625,
"learning_rate": 0.0004901574467301786,
"loss": 5.2055,
"mean_token_accuracy": 0.19453432112932206,
"num_tokens": 23968477.0,
"step": 10465
},
{
"entropy": 5.566391468048096,
"epoch": 1.034994068801898,
"grad_norm": 1.03125,
"learning_rate": 0.0004901471205867353,
"loss": 5.1163,
"mean_token_accuracy": 0.20114147365093232,
"num_tokens": 23979041.0,
"step": 10470
},
{
"entropy": 5.614834356307983,
"epoch": 1.0354883353103994,
"grad_norm": 0.97265625,
"learning_rate": 0.0004901367891506393,
"loss": 5.2199,
"mean_token_accuracy": 0.19947662353515624,
"num_tokens": 23990947.0,
"step": 10475
},
{
"entropy": 5.624517583847046,
"epoch": 1.0359826018189007,
"grad_norm": 0.9296875,
"learning_rate": 0.000490126452422145,
"loss": 5.1461,
"mean_token_accuracy": 0.20279881954193116,
"num_tokens": 24001652.0,
"step": 10480
},
{
"entropy": 5.666518020629883,
"epoch": 1.0364768683274022,
"grad_norm": 0.92578125,
"learning_rate": 0.0004901161104015066,
"loss": 5.2303,
"mean_token_accuracy": 0.19167366176843642,
"num_tokens": 24012707.0,
"step": 10485
},
{
"entropy": 5.5180035591125485,
"epoch": 1.0369711348359034,
"grad_norm": 1.1328125,
"learning_rate": 0.0004901057630889784,
"loss": 5.0543,
"mean_token_accuracy": 0.20800597071647645,
"num_tokens": 24024136.0,
"step": 10490
},
{
"entropy": 5.6054497241973875,
"epoch": 1.037465401344405,
"grad_norm": 0.9765625,
"learning_rate": 0.0004900954104848152,
"loss": 5.192,
"mean_token_accuracy": 0.20203883051872254,
"num_tokens": 24035586.0,
"step": 10495
},
{
"entropy": 5.648937511444092,
"epoch": 1.0379596678529064,
"grad_norm": 1.046875,
"learning_rate": 0.0004900850525892714,
"loss": 5.1294,
"mean_token_accuracy": 0.203294974565506,
"num_tokens": 24045338.0,
"step": 10500
},
{
"entropy": 5.643433666229248,
"epoch": 1.0384539343614076,
"grad_norm": 1.046875,
"learning_rate": 0.000490074689402602,
"loss": 5.2501,
"mean_token_accuracy": 0.18898199796676635,
"num_tokens": 24056898.0,
"step": 10505
},
{
"entropy": 5.512799406051636,
"epoch": 1.0389482008699091,
"grad_norm": 0.9296875,
"learning_rate": 0.0004900643209250619,
"loss": 5.0978,
"mean_token_accuracy": 0.20767101049423217,
"num_tokens": 24068998.0,
"step": 10510
},
{
"entropy": 5.568370723724366,
"epoch": 1.0394424673784104,
"grad_norm": 1.0078125,
"learning_rate": 0.0004900539471569061,
"loss": 5.1767,
"mean_token_accuracy": 0.1978091925382614,
"num_tokens": 24080301.0,
"step": 10515
},
{
"entropy": 5.634547996520996,
"epoch": 1.0399367338869119,
"grad_norm": 0.8984375,
"learning_rate": 0.00049004356809839,
"loss": 5.2053,
"mean_token_accuracy": 0.19516054093837737,
"num_tokens": 24092601.0,
"step": 10520
},
{
"entropy": 5.682157135009765,
"epoch": 1.0404310003954131,
"grad_norm": 0.9296875,
"learning_rate": 0.0004900331837497686,
"loss": 5.3022,
"mean_token_accuracy": 0.18680672645568847,
"num_tokens": 24105621.0,
"step": 10525
},
{
"entropy": 5.598519897460937,
"epoch": 1.0409252669039146,
"grad_norm": 0.8828125,
"learning_rate": 0.0004900227941112976,
"loss": 5.1541,
"mean_token_accuracy": 0.2127085879445076,
"num_tokens": 24117926.0,
"step": 10530
},
{
"entropy": 5.597533416748047,
"epoch": 1.0414195334124159,
"grad_norm": 0.96875,
"learning_rate": 0.0004900123991832323,
"loss": 5.1335,
"mean_token_accuracy": 0.195573154091835,
"num_tokens": 24129119.0,
"step": 10535
},
{
"entropy": 5.526741075515747,
"epoch": 1.0419137999209174,
"grad_norm": 1.1015625,
"learning_rate": 0.0004900019989658288,
"loss": 5.1993,
"mean_token_accuracy": 0.19209674149751663,
"num_tokens": 24141726.0,
"step": 10540
},
{
"entropy": 5.708407402038574,
"epoch": 1.0424080664294189,
"grad_norm": 0.87890625,
"learning_rate": 0.0004899915934593427,
"loss": 5.205,
"mean_token_accuracy": 0.19950339049100876,
"num_tokens": 24152912.0,
"step": 10545
},
{
"entropy": 5.690528154373169,
"epoch": 1.0429023329379201,
"grad_norm": 1.015625,
"learning_rate": 0.0004899811826640302,
"loss": 5.2395,
"mean_token_accuracy": 0.19273905754089354,
"num_tokens": 24163602.0,
"step": 10550
},
{
"entropy": 5.604873085021973,
"epoch": 1.0433965994464216,
"grad_norm": 1.03125,
"learning_rate": 0.0004899707665801471,
"loss": 5.2345,
"mean_token_accuracy": 0.1992226794362068,
"num_tokens": 24175904.0,
"step": 10555
},
{
"entropy": 5.567919158935547,
"epoch": 1.0438908659549229,
"grad_norm": 0.97265625,
"learning_rate": 0.0004899603452079499,
"loss": 5.1516,
"mean_token_accuracy": 0.19423959255218506,
"num_tokens": 24186438.0,
"step": 10560
},
{
"entropy": 5.589591121673584,
"epoch": 1.0443851324634243,
"grad_norm": 1.0703125,
"learning_rate": 0.0004899499185476948,
"loss": 5.2666,
"mean_token_accuracy": 0.19069302678108216,
"num_tokens": 24198357.0,
"step": 10565
},
{
"entropy": 5.729695463180542,
"epoch": 1.0448793989719256,
"grad_norm": 0.92578125,
"learning_rate": 0.0004899394865996383,
"loss": 5.3063,
"mean_token_accuracy": 0.191837902367115,
"num_tokens": 24211257.0,
"step": 10570
},
{
"entropy": 5.7230571746826175,
"epoch": 1.045373665480427,
"grad_norm": 0.99609375,
"learning_rate": 0.0004899290493640371,
"loss": 5.2878,
"mean_token_accuracy": 0.18740485459566117,
"num_tokens": 24223189.0,
"step": 10575
},
{
"entropy": 5.581752443313599,
"epoch": 1.0458679319889284,
"grad_norm": 0.984375,
"learning_rate": 0.0004899186068411481,
"loss": 5.189,
"mean_token_accuracy": 0.19531538039445878,
"num_tokens": 24234701.0,
"step": 10580
},
{
"entropy": 5.6230936527252195,
"epoch": 1.0463621984974298,
"grad_norm": 0.96875,
"learning_rate": 0.0004899081590312278,
"loss": 5.1793,
"mean_token_accuracy": 0.19982028901576995,
"num_tokens": 24245562.0,
"step": 10585
},
{
"entropy": 5.646186780929566,
"epoch": 1.046856465005931,
"grad_norm": 0.9921875,
"learning_rate": 0.0004898977059345336,
"loss": 5.1803,
"mean_token_accuracy": 0.19387844651937486,
"num_tokens": 24256479.0,
"step": 10590
},
{
"entropy": 5.632402229309082,
"epoch": 1.0473507315144326,
"grad_norm": 1.0,
"learning_rate": 0.0004898872475513225,
"loss": 5.2801,
"mean_token_accuracy": 0.18917266875505448,
"num_tokens": 24266900.0,
"step": 10595
},
{
"entropy": 5.673963165283203,
"epoch": 1.047844998022934,
"grad_norm": 0.9296875,
"learning_rate": 0.0004898767838818517,
"loss": 5.2648,
"mean_token_accuracy": 0.18993811160326005,
"num_tokens": 24278915.0,
"step": 10600
},
{
"entropy": 5.592510175704956,
"epoch": 1.0483392645314353,
"grad_norm": 0.99609375,
"learning_rate": 0.0004898663149263787,
"loss": 5.1102,
"mean_token_accuracy": 0.20620992332696914,
"num_tokens": 24289584.0,
"step": 10605
},
{
"entropy": 5.5979828357696535,
"epoch": 1.0488335310399368,
"grad_norm": 1.0078125,
"learning_rate": 0.0004898558406851611,
"loss": 5.135,
"mean_token_accuracy": 0.20248962789773942,
"num_tokens": 24300922.0,
"step": 10610
},
{
"entropy": 5.584086132049561,
"epoch": 1.049327797548438,
"grad_norm": 0.9453125,
"learning_rate": 0.0004898453611584564,
"loss": 5.153,
"mean_token_accuracy": 0.19852738231420516,
"num_tokens": 24311238.0,
"step": 10615
},
{
"entropy": 5.566949558258057,
"epoch": 1.0498220640569396,
"grad_norm": 1.0078125,
"learning_rate": 0.0004898348763465226,
"loss": 5.1416,
"mean_token_accuracy": 0.204348823428154,
"num_tokens": 24322417.0,
"step": 10620
},
{
"entropy": 5.592318773269653,
"epoch": 1.0503163305654408,
"grad_norm": 1.15625,
"learning_rate": 0.0004898243862496173,
"loss": 5.1488,
"mean_token_accuracy": 0.2061009556055069,
"num_tokens": 24332751.0,
"step": 10625
},
{
"entropy": 5.66185736656189,
"epoch": 1.0508105970739423,
"grad_norm": 1.015625,
"learning_rate": 0.0004898138908679989,
"loss": 5.2736,
"mean_token_accuracy": 0.18801855593919753,
"num_tokens": 24345082.0,
"step": 10630
},
{
"entropy": 5.633440017700195,
"epoch": 1.0513048635824436,
"grad_norm": 0.91796875,
"learning_rate": 0.0004898033902019255,
"loss": 5.2042,
"mean_token_accuracy": 0.19885159134864808,
"num_tokens": 24355863.0,
"step": 10635
},
{
"entropy": 5.718380928039551,
"epoch": 1.051799130090945,
"grad_norm": 0.96484375,
"learning_rate": 0.0004897928842516553,
"loss": 5.2603,
"mean_token_accuracy": 0.18917883187532425,
"num_tokens": 24367763.0,
"step": 10640
},
{
"entropy": 5.612067222595215,
"epoch": 1.0522933965994463,
"grad_norm": 0.98046875,
"learning_rate": 0.0004897823730174468,
"loss": 5.2277,
"mean_token_accuracy": 0.1921558752655983,
"num_tokens": 24378943.0,
"step": 10645
},
{
"entropy": 5.6442042827606205,
"epoch": 1.0527876631079478,
"grad_norm": 0.99609375,
"learning_rate": 0.0004897718564995587,
"loss": 5.1799,
"mean_token_accuracy": 0.197923643887043,
"num_tokens": 24389508.0,
"step": 10650
},
{
"entropy": 5.574830436706543,
"epoch": 1.0532819296164493,
"grad_norm": 0.984375,
"learning_rate": 0.0004897613346982495,
"loss": 5.1084,
"mean_token_accuracy": 0.20271708816289902,
"num_tokens": 24400710.0,
"step": 10655
},
{
"entropy": 5.528800773620605,
"epoch": 1.0537761961249505,
"grad_norm": 0.95703125,
"learning_rate": 0.0004897508076137783,
"loss": 5.1922,
"mean_token_accuracy": 0.1976326122879982,
"num_tokens": 24412854.0,
"step": 10660
},
{
"entropy": 5.623622035980224,
"epoch": 1.054270462633452,
"grad_norm": 0.97265625,
"learning_rate": 0.0004897402752464037,
"loss": 5.2277,
"mean_token_accuracy": 0.19712448120117188,
"num_tokens": 24423212.0,
"step": 10665
},
{
"entropy": 5.6668150424957275,
"epoch": 1.0547647291419533,
"grad_norm": 1.0546875,
"learning_rate": 0.0004897297375963852,
"loss": 5.2194,
"mean_token_accuracy": 0.20134411454200746,
"num_tokens": 24436091.0,
"step": 10670
},
{
"entropy": 5.582789087295533,
"epoch": 1.0552589956504548,
"grad_norm": 1.0625,
"learning_rate": 0.0004897191946639819,
"loss": 5.1213,
"mean_token_accuracy": 0.20405897200107576,
"num_tokens": 24448499.0,
"step": 10675
},
{
"entropy": 5.653404092788696,
"epoch": 1.055753262158956,
"grad_norm": 0.9765625,
"learning_rate": 0.0004897086464494529,
"loss": 5.3251,
"mean_token_accuracy": 0.18825308680534364,
"num_tokens": 24460732.0,
"step": 10680
},
{
"entropy": 5.597353219985962,
"epoch": 1.0562475286674575,
"grad_norm": 0.98046875,
"learning_rate": 0.0004896980929530579,
"loss": 5.1953,
"mean_token_accuracy": 0.20154686272144318,
"num_tokens": 24472460.0,
"step": 10685
},
{
"entropy": 5.61302809715271,
"epoch": 1.0567417951759588,
"grad_norm": 0.9765625,
"learning_rate": 0.0004896875341750567,
"loss": 5.1823,
"mean_token_accuracy": 0.20296552032232285,
"num_tokens": 24483820.0,
"step": 10690
},
{
"entropy": 5.565239286422729,
"epoch": 1.0572360616844603,
"grad_norm": 0.96875,
"learning_rate": 0.0004896769701157087,
"loss": 5.0864,
"mean_token_accuracy": 0.20751295536756514,
"num_tokens": 24495775.0,
"step": 10695
},
{
"entropy": 5.553544902801514,
"epoch": 1.0577303281929615,
"grad_norm": 0.96875,
"learning_rate": 0.000489666400775274,
"loss": 5.1228,
"mean_token_accuracy": 0.2052976295351982,
"num_tokens": 24505589.0,
"step": 10700
},
{
"entropy": 5.579530334472656,
"epoch": 1.058224594701463,
"grad_norm": 0.96484375,
"learning_rate": 0.0004896558261540125,
"loss": 5.1686,
"mean_token_accuracy": 0.20085039883852004,
"num_tokens": 24515988.0,
"step": 10705
},
{
"entropy": 5.537688064575195,
"epoch": 1.0587188612099645,
"grad_norm": 0.97265625,
"learning_rate": 0.0004896452462521844,
"loss": 5.126,
"mean_token_accuracy": 0.20575835704803466,
"num_tokens": 24528273.0,
"step": 10710
},
{
"entropy": 5.684996509552002,
"epoch": 1.0592131277184658,
"grad_norm": 1.09375,
"learning_rate": 0.0004896346610700498,
"loss": 5.1973,
"mean_token_accuracy": 0.19887850880622865,
"num_tokens": 24539014.0,
"step": 10715
},
{
"entropy": 5.5847328186035154,
"epoch": 1.0597073942269672,
"grad_norm": 0.921875,
"learning_rate": 0.0004896240706078694,
"loss": 5.1561,
"mean_token_accuracy": 0.20032164603471755,
"num_tokens": 24550210.0,
"step": 10720
},
{
"entropy": 5.550443267822265,
"epoch": 1.0602016607354685,
"grad_norm": 0.90625,
"learning_rate": 0.0004896134748659036,
"loss": 5.1335,
"mean_token_accuracy": 0.20644821971654892,
"num_tokens": 24560934.0,
"step": 10725
},
{
"entropy": 5.596108627319336,
"epoch": 1.06069592724397,
"grad_norm": 0.95703125,
"learning_rate": 0.000489602873844413,
"loss": 5.2042,
"mean_token_accuracy": 0.19550950676202775,
"num_tokens": 24572274.0,
"step": 10730
},
{
"entropy": 5.613519334793091,
"epoch": 1.0611901937524713,
"grad_norm": 0.90625,
"learning_rate": 0.0004895922675436583,
"loss": 5.2124,
"mean_token_accuracy": 0.19638487547636033,
"num_tokens": 24583950.0,
"step": 10735
},
{
"entropy": 5.583965873718261,
"epoch": 1.0616844602609727,
"grad_norm": 0.8125,
"learning_rate": 0.0004895816559639006,
"loss": 5.1483,
"mean_token_accuracy": 0.2041704386472702,
"num_tokens": 24597058.0,
"step": 10740
},
{
"entropy": 5.493352699279785,
"epoch": 1.062178726769474,
"grad_norm": 0.99609375,
"learning_rate": 0.0004895710391054008,
"loss": 5.0219,
"mean_token_accuracy": 0.2144222930073738,
"num_tokens": 24608604.0,
"step": 10745
},
{
"entropy": 5.619201612472534,
"epoch": 1.0626729932779755,
"grad_norm": 0.9765625,
"learning_rate": 0.0004895604169684203,
"loss": 5.247,
"mean_token_accuracy": 0.19471661895513534,
"num_tokens": 24619707.0,
"step": 10750
},
{
"entropy": 5.667275333404541,
"epoch": 1.063167259786477,
"grad_norm": 1.0625,
"learning_rate": 0.0004895497895532202,
"loss": 5.1749,
"mean_token_accuracy": 0.19414396584033966,
"num_tokens": 24629579.0,
"step": 10755
},
{
"entropy": 5.5603422164917,
"epoch": 1.0636615262949782,
"grad_norm": 0.9921875,
"learning_rate": 0.0004895391568600619,
"loss": 5.1818,
"mean_token_accuracy": 0.18896328657865524,
"num_tokens": 24641358.0,
"step": 10760
},
{
"entropy": 5.569506645202637,
"epoch": 1.0641557928034797,
"grad_norm": 1.09375,
"learning_rate": 0.0004895285188892071,
"loss": 5.1548,
"mean_token_accuracy": 0.20016748160123826,
"num_tokens": 24652293.0,
"step": 10765
},
{
"entropy": 5.6454822540283205,
"epoch": 1.064650059311981,
"grad_norm": 1.0078125,
"learning_rate": 0.0004895178756409175,
"loss": 5.2895,
"mean_token_accuracy": 0.19783279448747634,
"num_tokens": 24663370.0,
"step": 10770
},
{
"entropy": 5.649098014831543,
"epoch": 1.0651443258204825,
"grad_norm": 0.98828125,
"learning_rate": 0.0004895072271154548,
"loss": 5.2447,
"mean_token_accuracy": 0.19361262172460555,
"num_tokens": 24675037.0,
"step": 10775
},
{
"entropy": 5.616843271255493,
"epoch": 1.0656385923289837,
"grad_norm": 0.97265625,
"learning_rate": 0.0004894965733130812,
"loss": 5.16,
"mean_token_accuracy": 0.19610688835382462,
"num_tokens": 24685680.0,
"step": 10780
},
{
"entropy": 5.52163667678833,
"epoch": 1.0661328588374852,
"grad_norm": 0.9296875,
"learning_rate": 0.0004894859142340585,
"loss": 5.1401,
"mean_token_accuracy": 0.20336120426654816,
"num_tokens": 24697831.0,
"step": 10785
},
{
"entropy": 5.6424743175506595,
"epoch": 1.0666271253459865,
"grad_norm": 0.9453125,
"learning_rate": 0.0004894752498786491,
"loss": 5.1491,
"mean_token_accuracy": 0.19906843602657318,
"num_tokens": 24710226.0,
"step": 10790
},
{
"entropy": 5.641970443725586,
"epoch": 1.067121391854488,
"grad_norm": 0.93359375,
"learning_rate": 0.0004894645802471152,
"loss": 5.1787,
"mean_token_accuracy": 0.20488873422145842,
"num_tokens": 24721743.0,
"step": 10795
},
{
"entropy": 5.576434469223022,
"epoch": 1.0676156583629894,
"grad_norm": 0.91796875,
"learning_rate": 0.0004894539053397194,
"loss": 5.1801,
"mean_token_accuracy": 0.19224703460931777,
"num_tokens": 24733804.0,
"step": 10800
},
{
"entropy": 5.59564471244812,
"epoch": 1.0681099248714907,
"grad_norm": 1.0390625,
"learning_rate": 0.0004894432251567242,
"loss": 5.1246,
"mean_token_accuracy": 0.21002655029296874,
"num_tokens": 24745285.0,
"step": 10805
},
{
"entropy": 5.622626399993896,
"epoch": 1.0686041913799922,
"grad_norm": 0.9375,
"learning_rate": 0.0004894325396983925,
"loss": 5.1899,
"mean_token_accuracy": 0.19623009562492372,
"num_tokens": 24756069.0,
"step": 10810
},
{
"entropy": 5.544876050949097,
"epoch": 1.0690984578884934,
"grad_norm": 1.046875,
"learning_rate": 0.0004894218489649869,
"loss": 5.1393,
"mean_token_accuracy": 0.2041807159781456,
"num_tokens": 24766935.0,
"step": 10815
},
{
"entropy": 5.612660980224609,
"epoch": 1.069592724396995,
"grad_norm": 1.0390625,
"learning_rate": 0.0004894111529567707,
"loss": 5.1738,
"mean_token_accuracy": 0.1997630402445793,
"num_tokens": 24777642.0,
"step": 10820
},
{
"entropy": 5.594942378997803,
"epoch": 1.0700869909054962,
"grad_norm": 0.98046875,
"learning_rate": 0.0004894004516740069,
"loss": 5.1172,
"mean_token_accuracy": 0.20537441372871398,
"num_tokens": 24788110.0,
"step": 10825
},
{
"entropy": 5.660078048706055,
"epoch": 1.0705812574139977,
"grad_norm": 1.0078125,
"learning_rate": 0.0004893897451169586,
"loss": 5.2406,
"mean_token_accuracy": 0.19300525337457658,
"num_tokens": 24799976.0,
"step": 10830
},
{
"entropy": 5.653074645996094,
"epoch": 1.071075523922499,
"grad_norm": 1.015625,
"learning_rate": 0.0004893790332858893,
"loss": 5.207,
"mean_token_accuracy": 0.1971150279045105,
"num_tokens": 24811131.0,
"step": 10835
},
{
"entropy": 5.625109720230102,
"epoch": 1.0715697904310004,
"grad_norm": 0.93359375,
"learning_rate": 0.0004893683161810626,
"loss": 5.1768,
"mean_token_accuracy": 0.19663408994674683,
"num_tokens": 24822137.0,
"step": 10840
},
{
"entropy": 5.536339139938354,
"epoch": 1.0720640569395017,
"grad_norm": 1.015625,
"learning_rate": 0.0004893575938027421,
"loss": 5.1655,
"mean_token_accuracy": 0.20709992498159407,
"num_tokens": 24832369.0,
"step": 10845
},
{
"entropy": 5.609955501556397,
"epoch": 1.0725583234480032,
"grad_norm": 0.94140625,
"learning_rate": 0.0004893468661511914,
"loss": 5.2044,
"mean_token_accuracy": 0.20163075774908065,
"num_tokens": 24843351.0,
"step": 10850
},
{
"entropy": 5.733151578903199,
"epoch": 1.0730525899565047,
"grad_norm": 0.953125,
"learning_rate": 0.0004893361332266746,
"loss": 5.3056,
"mean_token_accuracy": 0.19099195152521134,
"num_tokens": 24855019.0,
"step": 10855
},
{
"entropy": 5.561675786972046,
"epoch": 1.073546856465006,
"grad_norm": 0.9375,
"learning_rate": 0.0004893253950294557,
"loss": 5.1964,
"mean_token_accuracy": 0.196122644841671,
"num_tokens": 24866224.0,
"step": 10860
},
{
"entropy": 5.589101648330688,
"epoch": 1.0740411229735074,
"grad_norm": 1.078125,
"learning_rate": 0.0004893146515597989,
"loss": 5.1185,
"mean_token_accuracy": 0.2096235290169716,
"num_tokens": 24876995.0,
"step": 10865
},
{
"entropy": 5.638453674316406,
"epoch": 1.0745353894820087,
"grad_norm": 0.95703125,
"learning_rate": 0.0004893039028179683,
"loss": 5.206,
"mean_token_accuracy": 0.1939111828804016,
"num_tokens": 24887846.0,
"step": 10870
},
{
"entropy": 5.63774528503418,
"epoch": 1.0750296559905101,
"grad_norm": 0.9375,
"learning_rate": 0.0004892931488042285,
"loss": 5.3321,
"mean_token_accuracy": 0.18875517547130585,
"num_tokens": 24899221.0,
"step": 10875
},
{
"entropy": 5.657603216171265,
"epoch": 1.0755239224990114,
"grad_norm": 0.87890625,
"learning_rate": 0.000489282389518844,
"loss": 5.2688,
"mean_token_accuracy": 0.1954587608575821,
"num_tokens": 24912103.0,
"step": 10880
},
{
"entropy": 5.694338798522949,
"epoch": 1.076018189007513,
"grad_norm": 1.0,
"learning_rate": 0.0004892716249620795,
"loss": 5.2519,
"mean_token_accuracy": 0.19037206023931502,
"num_tokens": 24925465.0,
"step": 10885
},
{
"entropy": 5.635915899276734,
"epoch": 1.0765124555160142,
"grad_norm": 1.046875,
"learning_rate": 0.0004892608551341997,
"loss": 5.0611,
"mean_token_accuracy": 0.20088418275117875,
"num_tokens": 24935136.0,
"step": 10890
},
{
"entropy": 5.5566919326782225,
"epoch": 1.0770067220245156,
"grad_norm": 0.9140625,
"learning_rate": 0.0004892500800354697,
"loss": 5.2048,
"mean_token_accuracy": 0.1929467126727104,
"num_tokens": 24946845.0,
"step": 10895
},
{
"entropy": 5.54016146659851,
"epoch": 1.077500988533017,
"grad_norm": 1.0390625,
"learning_rate": 0.0004892392996661544,
"loss": 5.1314,
"mean_token_accuracy": 0.19988464266061784,
"num_tokens": 24958337.0,
"step": 10900
},
{
"entropy": 5.6197408676147464,
"epoch": 1.0779952550415184,
"grad_norm": 1.0234375,
"learning_rate": 0.0004892285140265192,
"loss": 5.2508,
"mean_token_accuracy": 0.1911612018942833,
"num_tokens": 24969505.0,
"step": 10905
},
{
"entropy": 5.666508960723877,
"epoch": 1.0784895215500199,
"grad_norm": 0.93359375,
"learning_rate": 0.0004892177231168292,
"loss": 5.1895,
"mean_token_accuracy": 0.19795316755771636,
"num_tokens": 24981577.0,
"step": 10910
},
{
"entropy": 5.659940147399903,
"epoch": 1.0789837880585211,
"grad_norm": 1.015625,
"learning_rate": 0.0004892069269373502,
"loss": 5.2372,
"mean_token_accuracy": 0.19884584844112396,
"num_tokens": 24992758.0,
"step": 10915
},
{
"entropy": 5.621998834609985,
"epoch": 1.0794780545670226,
"grad_norm": 1.109375,
"learning_rate": 0.0004891961254883474,
"loss": 5.2006,
"mean_token_accuracy": 0.20227978527545928,
"num_tokens": 25003057.0,
"step": 10920
},
{
"entropy": 5.5721417427062985,
"epoch": 1.0799723210755239,
"grad_norm": 0.98828125,
"learning_rate": 0.0004891853187700867,
"loss": 5.2115,
"mean_token_accuracy": 0.19908509403467178,
"num_tokens": 25014900.0,
"step": 10925
},
{
"entropy": 5.565425062179566,
"epoch": 1.0804665875840254,
"grad_norm": 0.94921875,
"learning_rate": 0.0004891745067828339,
"loss": 5.1102,
"mean_token_accuracy": 0.2113812357187271,
"num_tokens": 25026313.0,
"step": 10930
},
{
"entropy": 5.644254541397094,
"epoch": 1.0809608540925266,
"grad_norm": 0.9921875,
"learning_rate": 0.0004891636895268551,
"loss": 5.1929,
"mean_token_accuracy": 0.1985420987010002,
"num_tokens": 25037014.0,
"step": 10935
},
{
"entropy": 5.6650491714477536,
"epoch": 1.081455120601028,
"grad_norm": 0.953125,
"learning_rate": 0.0004891528670024162,
"loss": 5.2551,
"mean_token_accuracy": 0.2007567435503006,
"num_tokens": 25048051.0,
"step": 10940
},
{
"entropy": 5.55631799697876,
"epoch": 1.0819493871095294,
"grad_norm": 1.1171875,
"learning_rate": 0.0004891420392097837,
"loss": 5.1048,
"mean_token_accuracy": 0.2099177733063698,
"num_tokens": 25057825.0,
"step": 10945
},
{
"entropy": 5.556065225601197,
"epoch": 1.0824436536180309,
"grad_norm": 0.95703125,
"learning_rate": 0.0004891312061492236,
"loss": 5.1715,
"mean_token_accuracy": 0.2023647114634514,
"num_tokens": 25067702.0,
"step": 10950
},
{
"entropy": 5.605633068084717,
"epoch": 1.0829379201265321,
"grad_norm": 1.0859375,
"learning_rate": 0.0004891203678210027,
"loss": 5.1468,
"mean_token_accuracy": 0.20347417891025543,
"num_tokens": 25078246.0,
"step": 10955
},
{
"entropy": 5.550473403930664,
"epoch": 1.0834321866350336,
"grad_norm": 0.97265625,
"learning_rate": 0.0004891095242253876,
"loss": 5.1661,
"mean_token_accuracy": 0.19707193374633789,
"num_tokens": 25089267.0,
"step": 10960
},
{
"entropy": 5.669507026672363,
"epoch": 1.083926453143535,
"grad_norm": 0.984375,
"learning_rate": 0.0004890986753626449,
"loss": 5.2526,
"mean_token_accuracy": 0.19486370980739592,
"num_tokens": 25101392.0,
"step": 10965
},
{
"entropy": 5.640515422821045,
"epoch": 1.0844207196520363,
"grad_norm": 0.8828125,
"learning_rate": 0.0004890878212330418,
"loss": 5.253,
"mean_token_accuracy": 0.19567573070526123,
"num_tokens": 25113309.0,
"step": 10970
},
{
"entropy": 5.523593378067017,
"epoch": 1.0849149861605378,
"grad_norm": 0.9375,
"learning_rate": 0.0004890769618368448,
"loss": 5.0631,
"mean_token_accuracy": 0.21330609172582626,
"num_tokens": 25124803.0,
"step": 10975
},
{
"entropy": 5.589742422103882,
"epoch": 1.085409252669039,
"grad_norm": 1.0234375,
"learning_rate": 0.0004890660971743215,
"loss": 5.1768,
"mean_token_accuracy": 0.1990187555551529,
"num_tokens": 25136443.0,
"step": 10980
},
{
"entropy": 5.6681482791900635,
"epoch": 1.0859035191775406,
"grad_norm": 1.078125,
"learning_rate": 0.0004890552272457389,
"loss": 5.1919,
"mean_token_accuracy": 0.1939637392759323,
"num_tokens": 25147128.0,
"step": 10985
},
{
"entropy": 5.576626396179199,
"epoch": 1.0863977856860418,
"grad_norm": 0.8671875,
"learning_rate": 0.0004890443520513645,
"loss": 5.2109,
"mean_token_accuracy": 0.19955306351184846,
"num_tokens": 25159420.0,
"step": 10990
},
{
"entropy": 5.5714414596557615,
"epoch": 1.0868920521945433,
"grad_norm": 0.8828125,
"learning_rate": 0.0004890334715914659,
"loss": 5.154,
"mean_token_accuracy": 0.20374392122030258,
"num_tokens": 25171624.0,
"step": 10995
},
{
"entropy": 5.583699083328247,
"epoch": 1.0873863187030446,
"grad_norm": 1.0078125,
"learning_rate": 0.0004890225858663106,
"loss": 5.0867,
"mean_token_accuracy": 0.20949340611696243,
"num_tokens": 25181610.0,
"step": 11000
},
{
"entropy": 5.588444662094116,
"epoch": 1.087880585211546,
"grad_norm": 1.0390625,
"learning_rate": 0.0004890116948761666,
"loss": 5.0904,
"mean_token_accuracy": 0.20562160164117813,
"num_tokens": 25191852.0,
"step": 11005
},
{
"entropy": 5.651496553421021,
"epoch": 1.0883748517200476,
"grad_norm": 0.96875,
"learning_rate": 0.0004890007986213018,
"loss": 5.2812,
"mean_token_accuracy": 0.18907185196876525,
"num_tokens": 25204240.0,
"step": 11010
},
{
"entropy": 5.525005674362182,
"epoch": 1.0888691182285488,
"grad_norm": 1.0859375,
"learning_rate": 0.0004889898971019838,
"loss": 5.1277,
"mean_token_accuracy": 0.20219947695732116,
"num_tokens": 25215512.0,
"step": 11015
},
{
"entropy": 5.596926307678222,
"epoch": 1.0893633847370503,
"grad_norm": 0.9609375,
"learning_rate": 0.0004889789903184815,
"loss": 5.216,
"mean_token_accuracy": 0.19566643387079238,
"num_tokens": 25227903.0,
"step": 11020
},
{
"entropy": 5.643597555160523,
"epoch": 1.0898576512455516,
"grad_norm": 0.9375,
"learning_rate": 0.0004889680782710626,
"loss": 5.1346,
"mean_token_accuracy": 0.20368800014257432,
"num_tokens": 25240204.0,
"step": 11025
},
{
"entropy": 5.56477952003479,
"epoch": 1.090351917754053,
"grad_norm": 0.92578125,
"learning_rate": 0.0004889571609599958,
"loss": 5.1314,
"mean_token_accuracy": 0.2046202763915062,
"num_tokens": 25251978.0,
"step": 11030
},
{
"entropy": 5.586012840270996,
"epoch": 1.0908461842625543,
"grad_norm": 0.953125,
"learning_rate": 0.0004889462383855498,
"loss": 5.1531,
"mean_token_accuracy": 0.20550966411828994,
"num_tokens": 25262496.0,
"step": 11035
},
{
"entropy": 5.60785756111145,
"epoch": 1.0913404507710558,
"grad_norm": 0.98828125,
"learning_rate": 0.000488935310547993,
"loss": 5.2084,
"mean_token_accuracy": 0.2000531330704689,
"num_tokens": 25272814.0,
"step": 11040
},
{
"entropy": 5.61498122215271,
"epoch": 1.091834717279557,
"grad_norm": 1.0,
"learning_rate": 0.0004889243774475945,
"loss": 5.1314,
"mean_token_accuracy": 0.19977964758872985,
"num_tokens": 25283749.0,
"step": 11045
},
{
"entropy": 5.540952396392822,
"epoch": 1.0923289837880585,
"grad_norm": 0.984375,
"learning_rate": 0.000488913439084623,
"loss": 5.1883,
"mean_token_accuracy": 0.2034799799323082,
"num_tokens": 25295993.0,
"step": 11050
},
{
"entropy": 5.58284330368042,
"epoch": 1.09282325029656,
"grad_norm": 0.96875,
"learning_rate": 0.0004889024954593477,
"loss": 5.1959,
"mean_token_accuracy": 0.19914669245481492,
"num_tokens": 25309011.0,
"step": 11055
},
{
"entropy": 5.620150566101074,
"epoch": 1.0933175168050613,
"grad_norm": 0.89453125,
"learning_rate": 0.000488891546572038,
"loss": 5.1481,
"mean_token_accuracy": 0.20043754130601882,
"num_tokens": 25320901.0,
"step": 11060
},
{
"entropy": 5.595513916015625,
"epoch": 1.0938117833135628,
"grad_norm": 0.91015625,
"learning_rate": 0.000488880592422963,
"loss": 5.1147,
"mean_token_accuracy": 0.2071356639266014,
"num_tokens": 25333042.0,
"step": 11065
},
{
"entropy": 5.571773481369019,
"epoch": 1.094306049822064,
"grad_norm": 0.921875,
"learning_rate": 0.0004888696330123921,
"loss": 5.1912,
"mean_token_accuracy": 0.20200786888599395,
"num_tokens": 25346413.0,
"step": 11070
},
{
"entropy": 5.611135101318359,
"epoch": 1.0948003163305655,
"grad_norm": 0.90234375,
"learning_rate": 0.0004888586683405952,
"loss": 5.2033,
"mean_token_accuracy": 0.19944217354059218,
"num_tokens": 25358104.0,
"step": 11075
},
{
"entropy": 5.596868515014648,
"epoch": 1.0952945828390668,
"grad_norm": 1.0234375,
"learning_rate": 0.0004888476984078419,
"loss": 5.1782,
"mean_token_accuracy": 0.19834440648555757,
"num_tokens": 25369485.0,
"step": 11080
},
{
"entropy": 5.655099105834961,
"epoch": 1.0957888493475683,
"grad_norm": 0.9375,
"learning_rate": 0.000488836723214402,
"loss": 5.2512,
"mean_token_accuracy": 0.19397983700037003,
"num_tokens": 25380349.0,
"step": 11085
},
{
"entropy": 5.622129201889038,
"epoch": 1.0962831158560695,
"grad_norm": 0.91796875,
"learning_rate": 0.0004888257427605456,
"loss": 5.2458,
"mean_token_accuracy": 0.19255951195955276,
"num_tokens": 25393042.0,
"step": 11090
},
{
"entropy": 5.708660697937011,
"epoch": 1.096777382364571,
"grad_norm": 1.078125,
"learning_rate": 0.0004888147570465426,
"loss": 5.2793,
"mean_token_accuracy": 0.1964405283331871,
"num_tokens": 25404470.0,
"step": 11095
},
{
"entropy": 5.565913677215576,
"epoch": 1.0972716488730723,
"grad_norm": 0.91796875,
"learning_rate": 0.0004888037660726635,
"loss": 5.116,
"mean_token_accuracy": 0.20731250196695328,
"num_tokens": 25415397.0,
"step": 11100
},
{
"entropy": 5.645141553878784,
"epoch": 1.0977659153815738,
"grad_norm": 0.9375,
"learning_rate": 0.0004887927698391786,
"loss": 5.2793,
"mean_token_accuracy": 0.18996219336986542,
"num_tokens": 25427409.0,
"step": 11105
},
{
"entropy": 5.610153436660767,
"epoch": 1.0982601818900752,
"grad_norm": 1.0234375,
"learning_rate": 0.0004887817683463583,
"loss": 5.1322,
"mean_token_accuracy": 0.20068458020687102,
"num_tokens": 25438063.0,
"step": 11110
},
{
"entropy": 5.58707709312439,
"epoch": 1.0987544483985765,
"grad_norm": 0.96484375,
"learning_rate": 0.0004887707615944734,
"loss": 5.0954,
"mean_token_accuracy": 0.20992284417152404,
"num_tokens": 25449242.0,
"step": 11115
},
{
"entropy": 5.638928365707398,
"epoch": 1.099248714907078,
"grad_norm": 0.96875,
"learning_rate": 0.0004887597495837946,
"loss": 5.2115,
"mean_token_accuracy": 0.20158472955226897,
"num_tokens": 25461434.0,
"step": 11120
},
{
"entropy": 5.558602237701416,
"epoch": 1.0997429814155792,
"grad_norm": 1.0703125,
"learning_rate": 0.0004887487323145929,
"loss": 5.1628,
"mean_token_accuracy": 0.20621996223926545,
"num_tokens": 25472035.0,
"step": 11125
},
{
"entropy": 5.640863275527954,
"epoch": 1.1002372479240807,
"grad_norm": 0.95703125,
"learning_rate": 0.000488737709787139,
"loss": 5.2064,
"mean_token_accuracy": 0.19716196954250337,
"num_tokens": 25483143.0,
"step": 11130
},
{
"entropy": 5.639927196502685,
"epoch": 1.100731514432582,
"grad_norm": 1.125,
"learning_rate": 0.0004887266820017044,
"loss": 5.3395,
"mean_token_accuracy": 0.18816036432981492,
"num_tokens": 25496873.0,
"step": 11135
},
{
"entropy": 5.60790319442749,
"epoch": 1.1012257809410835,
"grad_norm": 0.95703125,
"learning_rate": 0.0004887156489585603,
"loss": 5.1402,
"mean_token_accuracy": 0.20172854959964753,
"num_tokens": 25508661.0,
"step": 11140
},
{
"entropy": 5.692026042938233,
"epoch": 1.1017200474495847,
"grad_norm": 0.98828125,
"learning_rate": 0.000488704610657978,
"loss": 5.2829,
"mean_token_accuracy": 0.19025227129459382,
"num_tokens": 25520789.0,
"step": 11145
},
{
"entropy": 5.545534276962281,
"epoch": 1.1022143139580862,
"grad_norm": 0.99609375,
"learning_rate": 0.0004886935671002293,
"loss": 5.1295,
"mean_token_accuracy": 0.21081394404172898,
"num_tokens": 25531131.0,
"step": 11150
},
{
"entropy": 5.647500371932983,
"epoch": 1.1027085804665875,
"grad_norm": 0.8515625,
"learning_rate": 0.0004886825182855855,
"loss": 5.2167,
"mean_token_accuracy": 0.191609226167202,
"num_tokens": 25543803.0,
"step": 11155
},
{
"entropy": 5.652146434783935,
"epoch": 1.103202846975089,
"grad_norm": 1.0625,
"learning_rate": 0.0004886714642143188,
"loss": 5.1346,
"mean_token_accuracy": 0.20513637810945512,
"num_tokens": 25553588.0,
"step": 11160
},
{
"entropy": 5.593723440170288,
"epoch": 1.1036971134835905,
"grad_norm": 0.96484375,
"learning_rate": 0.0004886604048867007,
"loss": 5.1907,
"mean_token_accuracy": 0.20126118510961533,
"num_tokens": 25566175.0,
"step": 11165
},
{
"entropy": 5.639891624450684,
"epoch": 1.1041913799920917,
"grad_norm": 1.1015625,
"learning_rate": 0.0004886493403030035,
"loss": 5.2765,
"mean_token_accuracy": 0.19249668717384338,
"num_tokens": 25578033.0,
"step": 11170
},
{
"entropy": 5.58329586982727,
"epoch": 1.1046856465005932,
"grad_norm": 0.94921875,
"learning_rate": 0.0004886382704634996,
"loss": 5.147,
"mean_token_accuracy": 0.20333057194948195,
"num_tokens": 25588759.0,
"step": 11175
},
{
"entropy": 5.545940637588501,
"epoch": 1.1051799130090945,
"grad_norm": 1.0546875,
"learning_rate": 0.0004886271953684609,
"loss": 5.0587,
"mean_token_accuracy": 0.205940081179142,
"num_tokens": 25599200.0,
"step": 11180
},
{
"entropy": 5.570791435241699,
"epoch": 1.105674179517596,
"grad_norm": 1.0,
"learning_rate": 0.0004886161150181602,
"loss": 5.1735,
"mean_token_accuracy": 0.20087509453296662,
"num_tokens": 25610279.0,
"step": 11185
},
{
"entropy": 5.554993772506714,
"epoch": 1.1061684460260972,
"grad_norm": 0.91015625,
"learning_rate": 0.0004886050294128698,
"loss": 5.1272,
"mean_token_accuracy": 0.20424916297197343,
"num_tokens": 25622062.0,
"step": 11190
},
{
"entropy": 5.6129734992980955,
"epoch": 1.1066627125345987,
"grad_norm": 1.0859375,
"learning_rate": 0.0004885939385528628,
"loss": 5.1749,
"mean_token_accuracy": 0.19783098250627518,
"num_tokens": 25633451.0,
"step": 11195
},
{
"entropy": 5.544128036499023,
"epoch": 1.1071569790431,
"grad_norm": 0.9609375,
"learning_rate": 0.0004885828424384114,
"loss": 5.1133,
"mean_token_accuracy": 0.20746660530567168,
"num_tokens": 25644537.0,
"step": 11200
},
{
"entropy": 5.57102255821228,
"epoch": 1.1076512455516014,
"grad_norm": 1.0078125,
"learning_rate": 0.0004885717410697892,
"loss": 5.1676,
"mean_token_accuracy": 0.19861021041870117,
"num_tokens": 25655570.0,
"step": 11205
},
{
"entropy": 5.707389211654663,
"epoch": 1.1081455120601027,
"grad_norm": 0.875,
"learning_rate": 0.000488560634447269,
"loss": 5.2963,
"mean_token_accuracy": 0.18938837051391602,
"num_tokens": 25669470.0,
"step": 11210
},
{
"entropy": 5.630181837081909,
"epoch": 1.1086397785686042,
"grad_norm": 0.9609375,
"learning_rate": 0.000488549522571124,
"loss": 5.2087,
"mean_token_accuracy": 0.19894858151674272,
"num_tokens": 25682347.0,
"step": 11215
},
{
"entropy": 5.6691876411437985,
"epoch": 1.1091340450771057,
"grad_norm": 1.140625,
"learning_rate": 0.0004885384054416276,
"loss": 5.2769,
"mean_token_accuracy": 0.19145484268665314,
"num_tokens": 25694038.0,
"step": 11220
},
{
"entropy": 5.593998432159424,
"epoch": 1.109628311585607,
"grad_norm": 0.92578125,
"learning_rate": 0.0004885272830590534,
"loss": 5.0791,
"mean_token_accuracy": 0.20883855670690538,
"num_tokens": 25705850.0,
"step": 11225
},
{
"entropy": 5.575067806243896,
"epoch": 1.1101225780941084,
"grad_norm": 1.046875,
"learning_rate": 0.0004885161554236748,
"loss": 5.1712,
"mean_token_accuracy": 0.20403964221477508,
"num_tokens": 25716294.0,
"step": 11230
},
{
"entropy": 5.568801021575927,
"epoch": 1.1106168446026097,
"grad_norm": 0.9296875,
"learning_rate": 0.0004885050225357657,
"loss": 5.1627,
"mean_token_accuracy": 0.20878512114286424,
"num_tokens": 25727367.0,
"step": 11235
},
{
"entropy": 5.710361957550049,
"epoch": 1.1111111111111112,
"grad_norm": 1.0390625,
"learning_rate": 0.0004884938843955999,
"loss": 5.28,
"mean_token_accuracy": 0.1868166446685791,
"num_tokens": 25738352.0,
"step": 11240
},
{
"entropy": 5.532423973083496,
"epoch": 1.1116053776196124,
"grad_norm": 0.9765625,
"learning_rate": 0.0004884827410034515,
"loss": 5.0743,
"mean_token_accuracy": 0.21203147172927855,
"num_tokens": 25749940.0,
"step": 11245
},
{
"entropy": 5.602033901214599,
"epoch": 1.112099644128114,
"grad_norm": 1.015625,
"learning_rate": 0.0004884715923595945,
"loss": 5.2046,
"mean_token_accuracy": 0.19205178171396256,
"num_tokens": 25761251.0,
"step": 11250
},
{
"entropy": 5.6186737537384035,
"epoch": 1.1125939106366152,
"grad_norm": 0.9453125,
"learning_rate": 0.0004884604384643032,
"loss": 5.2239,
"mean_token_accuracy": 0.19850173592567444,
"num_tokens": 25772382.0,
"step": 11255
},
{
"entropy": 5.552934026718139,
"epoch": 1.1130881771451167,
"grad_norm": 0.91015625,
"learning_rate": 0.000488449279317852,
"loss": 5.1453,
"mean_token_accuracy": 0.20429897606372832,
"num_tokens": 25784960.0,
"step": 11260
},
{
"entropy": 5.629480266571045,
"epoch": 1.113582443653618,
"grad_norm": 1.0,
"learning_rate": 0.0004884381149205154,
"loss": 5.1649,
"mean_token_accuracy": 0.1951213926076889,
"num_tokens": 25796644.0,
"step": 11265
},
{
"entropy": 5.601596975326538,
"epoch": 1.1140767101621194,
"grad_norm": 0.94921875,
"learning_rate": 0.0004884269452725681,
"loss": 5.2286,
"mean_token_accuracy": 0.20162205398082733,
"num_tokens": 25807712.0,
"step": 11270
},
{
"entropy": 5.64228081703186,
"epoch": 1.1145709766706209,
"grad_norm": 1.0078125,
"learning_rate": 0.0004884157703742848,
"loss": 5.1254,
"mean_token_accuracy": 0.20367183089256286,
"num_tokens": 25817944.0,
"step": 11275
},
{
"entropy": 5.540275144577026,
"epoch": 1.1150652431791221,
"grad_norm": 0.91015625,
"learning_rate": 0.0004884045902259405,
"loss": 5.1498,
"mean_token_accuracy": 0.2066369414329529,
"num_tokens": 25830801.0,
"step": 11280
},
{
"entropy": 5.557745456695557,
"epoch": 1.1155595096876236,
"grad_norm": 1.015625,
"learning_rate": 0.0004883934048278102,
"loss": 5.1637,
"mean_token_accuracy": 0.19736449122428895,
"num_tokens": 25843877.0,
"step": 11285
},
{
"entropy": 5.612047529220581,
"epoch": 1.116053776196125,
"grad_norm": 0.94921875,
"learning_rate": 0.0004883822141801689,
"loss": 5.1135,
"mean_token_accuracy": 0.2047475978732109,
"num_tokens": 25853763.0,
"step": 11290
},
{
"entropy": 5.660339212417602,
"epoch": 1.1165480427046264,
"grad_norm": 0.95703125,
"learning_rate": 0.0004883710182832921,
"loss": 5.2393,
"mean_token_accuracy": 0.1922273263335228,
"num_tokens": 25864759.0,
"step": 11295
},
{
"entropy": 5.690654945373535,
"epoch": 1.1170423092131276,
"grad_norm": 0.9296875,
"learning_rate": 0.0004883598171374552,
"loss": 5.2975,
"mean_token_accuracy": 0.18838510662317276,
"num_tokens": 25877325.0,
"step": 11300
},
{
"entropy": 5.598295593261719,
"epoch": 1.1175365757216291,
"grad_norm": 1.0625,
"learning_rate": 0.0004883486107429337,
"loss": 5.1513,
"mean_token_accuracy": 0.20322973430156707,
"num_tokens": 25889308.0,
"step": 11305
},
{
"entropy": 5.570185136795044,
"epoch": 1.1180308422301304,
"grad_norm": 0.98046875,
"learning_rate": 0.0004883373991000031,
"loss": 5.2,
"mean_token_accuracy": 0.19732243269681932,
"num_tokens": 25901721.0,
"step": 11310
},
{
"entropy": 5.6874839782714846,
"epoch": 1.1185251087386319,
"grad_norm": 1.0390625,
"learning_rate": 0.0004883261822089395,
"loss": 5.1879,
"mean_token_accuracy": 0.1956360951066017,
"num_tokens": 25912399.0,
"step": 11315
},
{
"entropy": 5.571163463592529,
"epoch": 1.1190193752471334,
"grad_norm": 1.0390625,
"learning_rate": 0.0004883149600700188,
"loss": 5.1159,
"mean_token_accuracy": 0.20984721034765244,
"num_tokens": 25923560.0,
"step": 11320
},
{
"entropy": 5.6397301197052006,
"epoch": 1.1195136417556346,
"grad_norm": 1.1640625,
"learning_rate": 0.0004883037326835169,
"loss": 5.1658,
"mean_token_accuracy": 0.2001958519220352,
"num_tokens": 25933408.0,
"step": 11325
},
{
"entropy": 5.6041769027709964,
"epoch": 1.120007908264136,
"grad_norm": 1.0234375,
"learning_rate": 0.0004882925000497101,
"loss": 5.1591,
"mean_token_accuracy": 0.19686995893716813,
"num_tokens": 25944492.0,
"step": 11330
},
{
"entropy": 5.5810133934021,
"epoch": 1.1205021747726374,
"grad_norm": 1.03125,
"learning_rate": 0.0004882812621688747,
"loss": 5.1528,
"mean_token_accuracy": 0.2053610309958458,
"num_tokens": 25956439.0,
"step": 11335
},
{
"entropy": 5.585991334915161,
"epoch": 1.1209964412811388,
"grad_norm": 0.9453125,
"learning_rate": 0.0004882700190412873,
"loss": 5.2095,
"mean_token_accuracy": 0.19325564205646514,
"num_tokens": 25968765.0,
"step": 11340
},
{
"entropy": 5.609309339523316,
"epoch": 1.12149070778964,
"grad_norm": 0.90625,
"learning_rate": 0.0004882587706672242,
"loss": 5.2126,
"mean_token_accuracy": 0.19817647635936736,
"num_tokens": 25981300.0,
"step": 11345
},
{
"entropy": 5.61623477935791,
"epoch": 1.1219849742981416,
"grad_norm": 0.99609375,
"learning_rate": 0.0004882475170469622,
"loss": 5.1469,
"mean_token_accuracy": 0.2071049228310585,
"num_tokens": 25991655.0,
"step": 11350
},
{
"entropy": 5.58967695236206,
"epoch": 1.1224792408066429,
"grad_norm": 0.9765625,
"learning_rate": 0.0004882362581807783,
"loss": 5.1276,
"mean_token_accuracy": 0.20514735728502273,
"num_tokens": 26002989.0,
"step": 11355
},
{
"entropy": 5.550833034515381,
"epoch": 1.1229735073151443,
"grad_norm": 1.0625,
"learning_rate": 0.0004882249940689494,
"loss": 5.1502,
"mean_token_accuracy": 0.2068599134683609,
"num_tokens": 26013807.0,
"step": 11360
},
{
"entropy": 5.695304489135742,
"epoch": 1.1234677738236458,
"grad_norm": 0.87109375,
"learning_rate": 0.00048821372471175245,
"loss": 5.269,
"mean_token_accuracy": 0.19761935323476792,
"num_tokens": 26027099.0,
"step": 11365
},
{
"entropy": 5.652538633346557,
"epoch": 1.123962040332147,
"grad_norm": 0.92578125,
"learning_rate": 0.00048820245010946486,
"loss": 5.1417,
"mean_token_accuracy": 0.19648411124944687,
"num_tokens": 26039046.0,
"step": 11370
},
{
"entropy": 5.527272796630859,
"epoch": 1.1244563068406486,
"grad_norm": 1.015625,
"learning_rate": 0.0004881911702623639,
"loss": 5.0248,
"mean_token_accuracy": 0.21236271560192108,
"num_tokens": 26049763.0,
"step": 11375
},
{
"entropy": 5.525219249725342,
"epoch": 1.1249505733491498,
"grad_norm": 1.390625,
"learning_rate": 0.0004881798851707271,
"loss": 5.1811,
"mean_token_accuracy": 0.2024916708469391,
"num_tokens": 26062227.0,
"step": 11380
},
{
"entropy": 5.5266149044036865,
"epoch": 1.1254448398576513,
"grad_norm": 0.98046875,
"learning_rate": 0.00048816859483483214,
"loss": 5.1467,
"mean_token_accuracy": 0.2034767359495163,
"num_tokens": 26074019.0,
"step": 11385
},
{
"entropy": 5.69959864616394,
"epoch": 1.1259391063661526,
"grad_norm": 1.1015625,
"learning_rate": 0.00048815729925495656,
"loss": 5.2631,
"mean_token_accuracy": 0.19384338557720185,
"num_tokens": 26086169.0,
"step": 11390
},
{
"entropy": 5.603618812561035,
"epoch": 1.126433372874654,
"grad_norm": 0.94140625,
"learning_rate": 0.00048814599843137837,
"loss": 5.1808,
"mean_token_accuracy": 0.19195906668901444,
"num_tokens": 26097677.0,
"step": 11395
},
{
"entropy": 5.499852514266967,
"epoch": 1.1269276393831553,
"grad_norm": 0.93359375,
"learning_rate": 0.0004881346923643755,
"loss": 5.1025,
"mean_token_accuracy": 0.20891562551259996,
"num_tokens": 26108825.0,
"step": 11400
},
{
"entropy": 5.629191637039185,
"epoch": 1.1274219058916568,
"grad_norm": 0.91796875,
"learning_rate": 0.0004881233810542263,
"loss": 5.2401,
"mean_token_accuracy": 0.19421287328004838,
"num_tokens": 26120274.0,
"step": 11405
},
{
"entropy": 5.602230501174927,
"epoch": 1.127916172400158,
"grad_norm": 0.91015625,
"learning_rate": 0.00048811206450120875,
"loss": 5.142,
"mean_token_accuracy": 0.2055575057864189,
"num_tokens": 26131964.0,
"step": 11410
},
{
"entropy": 5.630838680267334,
"epoch": 1.1284104389086596,
"grad_norm": 0.984375,
"learning_rate": 0.0004881007427056013,
"loss": 5.2046,
"mean_token_accuracy": 0.19535789489746094,
"num_tokens": 26143308.0,
"step": 11415
},
{
"entropy": 5.6695550918579105,
"epoch": 1.128904705417161,
"grad_norm": 1.03125,
"learning_rate": 0.0004880894156676826,
"loss": 5.2506,
"mean_token_accuracy": 0.19100353866815567,
"num_tokens": 26153951.0,
"step": 11420
},
{
"entropy": 5.60924973487854,
"epoch": 1.1293989719256623,
"grad_norm": 1.0390625,
"learning_rate": 0.0004880780833877312,
"loss": 5.1099,
"mean_token_accuracy": 0.20742035657167435,
"num_tokens": 26165951.0,
"step": 11425
},
{
"entropy": 5.580791664123535,
"epoch": 1.1298932384341638,
"grad_norm": 0.98828125,
"learning_rate": 0.0004880667458660259,
"loss": 5.1578,
"mean_token_accuracy": 0.1946610689163208,
"num_tokens": 26176751.0,
"step": 11430
},
{
"entropy": 5.629481077194214,
"epoch": 1.130387504942665,
"grad_norm": 0.96875,
"learning_rate": 0.0004880554031028456,
"loss": 5.2578,
"mean_token_accuracy": 0.1927572399377823,
"num_tokens": 26187682.0,
"step": 11435
},
{
"entropy": 5.683337879180908,
"epoch": 1.1308817714511665,
"grad_norm": 0.97265625,
"learning_rate": 0.00048804405509846933,
"loss": 5.2036,
"mean_token_accuracy": 0.19814158976078033,
"num_tokens": 26198960.0,
"step": 11440
},
{
"entropy": 5.553512954711914,
"epoch": 1.1313760379596678,
"grad_norm": 1.0234375,
"learning_rate": 0.0004880327018531762,
"loss": 5.1716,
"mean_token_accuracy": 0.1977383866906166,
"num_tokens": 26210198.0,
"step": 11445
},
{
"entropy": 5.609118843078614,
"epoch": 1.1318703044681693,
"grad_norm": 1.0703125,
"learning_rate": 0.00048802134336724566,
"loss": 5.0991,
"mean_token_accuracy": 0.2031602993607521,
"num_tokens": 26220817.0,
"step": 11450
},
{
"entropy": 5.576944732666016,
"epoch": 1.1323645709766705,
"grad_norm": 0.99609375,
"learning_rate": 0.00048800997964095687,
"loss": 5.1078,
"mean_token_accuracy": 0.20851015746593476,
"num_tokens": 26233657.0,
"step": 11455
},
{
"entropy": 5.580789995193482,
"epoch": 1.132858837485172,
"grad_norm": 0.96875,
"learning_rate": 0.0004879986106745897,
"loss": 5.159,
"mean_token_accuracy": 0.2004464715719223,
"num_tokens": 26245566.0,
"step": 11460
},
{
"entropy": 5.608809328079223,
"epoch": 1.1333531039936733,
"grad_norm": 0.87890625,
"learning_rate": 0.00048798723646842357,
"loss": 5.0915,
"mean_token_accuracy": 0.2103253111243248,
"num_tokens": 26256717.0,
"step": 11465
},
{
"entropy": 5.611515522003174,
"epoch": 1.1338473705021748,
"grad_norm": 0.9765625,
"learning_rate": 0.0004879758570227384,
"loss": 5.2073,
"mean_token_accuracy": 0.19773997217416764,
"num_tokens": 26267926.0,
"step": 11470
},
{
"entropy": 5.582551765441894,
"epoch": 1.1343416370106763,
"grad_norm": 1.015625,
"learning_rate": 0.00048796447233781406,
"loss": 5.1725,
"mean_token_accuracy": 0.19708593040704728,
"num_tokens": 26278666.0,
"step": 11475
},
{
"entropy": 5.60650486946106,
"epoch": 1.1348359035191775,
"grad_norm": 1.0078125,
"learning_rate": 0.0004879530824139308,
"loss": 5.1618,
"mean_token_accuracy": 0.20267072767019273,
"num_tokens": 26289961.0,
"step": 11480
},
{
"entropy": 5.589709043502808,
"epoch": 1.135330170027679,
"grad_norm": 0.95703125,
"learning_rate": 0.00048794168725136854,
"loss": 5.1882,
"mean_token_accuracy": 0.19654595106840134,
"num_tokens": 26301245.0,
"step": 11485
},
{
"entropy": 5.6204798221588135,
"epoch": 1.1358244365361803,
"grad_norm": 1.0625,
"learning_rate": 0.0004879302868504078,
"loss": 5.2082,
"mean_token_accuracy": 0.19338615536689757,
"num_tokens": 26312598.0,
"step": 11490
},
{
"entropy": 5.720840120315552,
"epoch": 1.1363187030446817,
"grad_norm": 0.86328125,
"learning_rate": 0.0004879188812113289,
"loss": 5.2089,
"mean_token_accuracy": 0.20111506730318068,
"num_tokens": 26324300.0,
"step": 11495
},
{
"entropy": 5.570767974853515,
"epoch": 1.136812969553183,
"grad_norm": 0.95703125,
"learning_rate": 0.0004879074703344125,
"loss": 5.1969,
"mean_token_accuracy": 0.19834372997283936,
"num_tokens": 26335789.0,
"step": 11500
},
{
"entropy": 5.5793274402618405,
"epoch": 1.1373072360616845,
"grad_norm": 0.9609375,
"learning_rate": 0.0004878960542199394,
"loss": 5.1869,
"mean_token_accuracy": 0.19730139821767806,
"num_tokens": 26347336.0,
"step": 11505
},
{
"entropy": 5.656808757781983,
"epoch": 1.1378015025701858,
"grad_norm": 0.96875,
"learning_rate": 0.00048788463286819037,
"loss": 5.1564,
"mean_token_accuracy": 0.2024213269352913,
"num_tokens": 26358176.0,
"step": 11510
},
{
"entropy": 5.55414605140686,
"epoch": 1.1382957690786872,
"grad_norm": 1.0078125,
"learning_rate": 0.0004878732062794462,
"loss": 5.1487,
"mean_token_accuracy": 0.19850924015045165,
"num_tokens": 26368915.0,
"step": 11515
},
{
"entropy": 5.594301843643189,
"epoch": 1.1387900355871885,
"grad_norm": 0.96484375,
"learning_rate": 0.00048786177445398816,
"loss": 5.2045,
"mean_token_accuracy": 0.19778390675783158,
"num_tokens": 26381687.0,
"step": 11520
},
{
"entropy": 5.646654272079468,
"epoch": 1.13928430209569,
"grad_norm": 0.875,
"learning_rate": 0.0004878503373920975,
"loss": 5.2644,
"mean_token_accuracy": 0.19569268822669983,
"num_tokens": 26394289.0,
"step": 11525
},
{
"entropy": 5.608573150634766,
"epoch": 1.1397785686041915,
"grad_norm": 0.99609375,
"learning_rate": 0.0004878388950940555,
"loss": 5.1351,
"mean_token_accuracy": 0.20786156356334687,
"num_tokens": 26405743.0,
"step": 11530
},
{
"entropy": 5.6764817237854,
"epoch": 1.1402728351126927,
"grad_norm": 1.03125,
"learning_rate": 0.00048782744756014365,
"loss": 5.2797,
"mean_token_accuracy": 0.18656122982501983,
"num_tokens": 26417255.0,
"step": 11535
},
{
"entropy": 5.682086658477783,
"epoch": 1.1407671016211942,
"grad_norm": 0.9140625,
"learning_rate": 0.0004878159947906436,
"loss": 5.2685,
"mean_token_accuracy": 0.19522271901369095,
"num_tokens": 26428992.0,
"step": 11540
},
{
"entropy": 5.593657445907593,
"epoch": 1.1412613681296955,
"grad_norm": 1.0078125,
"learning_rate": 0.000487804536785837,
"loss": 5.1638,
"mean_token_accuracy": 0.2081797391176224,
"num_tokens": 26440375.0,
"step": 11545
},
{
"entropy": 5.583353424072266,
"epoch": 1.141755634638197,
"grad_norm": 0.859375,
"learning_rate": 0.00048779307354600586,
"loss": 5.2271,
"mean_token_accuracy": 0.19794875085353852,
"num_tokens": 26452253.0,
"step": 11550
},
{
"entropy": 5.6686248779296875,
"epoch": 1.1422499011466982,
"grad_norm": 0.90625,
"learning_rate": 0.00048778160507143204,
"loss": 5.2506,
"mean_token_accuracy": 0.19715603590011596,
"num_tokens": 26463766.0,
"step": 11555
},
{
"entropy": 5.642669153213501,
"epoch": 1.1427441676551997,
"grad_norm": 0.9140625,
"learning_rate": 0.0004877701313623978,
"loss": 5.2079,
"mean_token_accuracy": 0.19412551671266556,
"num_tokens": 26475869.0,
"step": 11560
},
{
"entropy": 5.6445270538330075,
"epoch": 1.1432384341637012,
"grad_norm": 1.0,
"learning_rate": 0.0004877586524191852,
"loss": 5.2022,
"mean_token_accuracy": 0.19723747819662094,
"num_tokens": 26487142.0,
"step": 11565
},
{
"entropy": 5.580124187469482,
"epoch": 1.1437327006722025,
"grad_norm": 0.9140625,
"learning_rate": 0.0004877471682420768,
"loss": 5.0396,
"mean_token_accuracy": 0.2061804324388504,
"num_tokens": 26499019.0,
"step": 11570
},
{
"entropy": 5.569995498657226,
"epoch": 1.1442269671807037,
"grad_norm": 1.1796875,
"learning_rate": 0.0004877356788313551,
"loss": 5.1449,
"mean_token_accuracy": 0.20974014103412628,
"num_tokens": 26510060.0,
"step": 11575
},
{
"entropy": 5.548111152648926,
"epoch": 1.1447212336892052,
"grad_norm": 0.88671875,
"learning_rate": 0.0004877241841873027,
"loss": 5.0773,
"mean_token_accuracy": 0.204905404150486,
"num_tokens": 26520723.0,
"step": 11580
},
{
"entropy": 5.563028955459595,
"epoch": 1.1452155001977067,
"grad_norm": 1.171875,
"learning_rate": 0.0004877126843102023,
"loss": 5.1618,
"mean_token_accuracy": 0.20420341491699218,
"num_tokens": 26532545.0,
"step": 11585
},
{
"entropy": 5.565567207336426,
"epoch": 1.145709766706208,
"grad_norm": 1.046875,
"learning_rate": 0.00048770117920033697,
"loss": 5.1579,
"mean_token_accuracy": 0.2123001843690872,
"num_tokens": 26542993.0,
"step": 11590
},
{
"entropy": 5.564682960510254,
"epoch": 1.1462040332147094,
"grad_norm": 0.98046875,
"learning_rate": 0.0004876896688579895,
"loss": 5.1413,
"mean_token_accuracy": 0.20572238117456437,
"num_tokens": 26554658.0,
"step": 11595
},
{
"entropy": 5.629250621795654,
"epoch": 1.1466982997232107,
"grad_norm": 1.0625,
"learning_rate": 0.0004876781532834433,
"loss": 5.1757,
"mean_token_accuracy": 0.2058918759226799,
"num_tokens": 26565925.0,
"step": 11600
},
{
"entropy": 5.597628784179688,
"epoch": 1.1471925662317122,
"grad_norm": 0.9921875,
"learning_rate": 0.0004876666324769815,
"loss": 5.1431,
"mean_token_accuracy": 0.20127459168434142,
"num_tokens": 26577045.0,
"step": 11605
},
{
"entropy": 5.590678405761719,
"epoch": 1.1476868327402134,
"grad_norm": 1.0234375,
"learning_rate": 0.0004876551064388875,
"loss": 5.2483,
"mean_token_accuracy": 0.19435011148452758,
"num_tokens": 26587773.0,
"step": 11610
},
{
"entropy": 5.661182546615601,
"epoch": 1.148181099248715,
"grad_norm": 0.9609375,
"learning_rate": 0.00048764357516944485,
"loss": 5.1223,
"mean_token_accuracy": 0.2018751546740532,
"num_tokens": 26597861.0,
"step": 11615
},
{
"entropy": 5.6584696769714355,
"epoch": 1.1486753657572164,
"grad_norm": 1.03125,
"learning_rate": 0.00048763203866893735,
"loss": 5.182,
"mean_token_accuracy": 0.19973020106554032,
"num_tokens": 26607695.0,
"step": 11620
},
{
"entropy": 5.625906753540039,
"epoch": 1.1491696322657177,
"grad_norm": 0.98046875,
"learning_rate": 0.00048762049693764866,
"loss": 5.2376,
"mean_token_accuracy": 0.18958619236946106,
"num_tokens": 26619845.0,
"step": 11625
},
{
"entropy": 5.661731863021851,
"epoch": 1.1496638987742192,
"grad_norm": 1.0390625,
"learning_rate": 0.00048760894997586284,
"loss": 5.1348,
"mean_token_accuracy": 0.2008199065923691,
"num_tokens": 26630279.0,
"step": 11630
},
{
"entropy": 5.543630838394165,
"epoch": 1.1501581652827204,
"grad_norm": 0.98828125,
"learning_rate": 0.0004875973977838639,
"loss": 5.1637,
"mean_token_accuracy": 0.20060469955205917,
"num_tokens": 26640996.0,
"step": 11635
},
{
"entropy": 5.525980710983276,
"epoch": 1.150652431791222,
"grad_norm": 1.03125,
"learning_rate": 0.00048758584036193587,
"loss": 5.0949,
"mean_token_accuracy": 0.20681791603565217,
"num_tokens": 26652416.0,
"step": 11640
},
{
"entropy": 5.604940414428711,
"epoch": 1.1511466982997232,
"grad_norm": 0.94140625,
"learning_rate": 0.0004875742777103632,
"loss": 5.1951,
"mean_token_accuracy": 0.1943777620792389,
"num_tokens": 26663649.0,
"step": 11645
},
{
"entropy": 5.642388439178466,
"epoch": 1.1516409648082246,
"grad_norm": 0.9296875,
"learning_rate": 0.00048756270982943035,
"loss": 5.2541,
"mean_token_accuracy": 0.19223594218492507,
"num_tokens": 26676138.0,
"step": 11650
},
{
"entropy": 5.678112506866455,
"epoch": 1.152135231316726,
"grad_norm": 0.96484375,
"learning_rate": 0.0004875511367194219,
"loss": 5.2275,
"mean_token_accuracy": 0.19011210799217224,
"num_tokens": 26687364.0,
"step": 11655
},
{
"entropy": 5.614232921600342,
"epoch": 1.1526294978252274,
"grad_norm": 1.0,
"learning_rate": 0.0004875395583806225,
"loss": 5.176,
"mean_token_accuracy": 0.19905974715948105,
"num_tokens": 26698135.0,
"step": 11660
},
{
"entropy": 5.603475761413574,
"epoch": 1.1531237643337287,
"grad_norm": 1.0703125,
"learning_rate": 0.0004875279748133169,
"loss": 5.1227,
"mean_token_accuracy": 0.2048828646540642,
"num_tokens": 26710495.0,
"step": 11665
},
{
"entropy": 5.675576496124267,
"epoch": 1.1536180308422301,
"grad_norm": 1.0234375,
"learning_rate": 0.0004875163860177902,
"loss": 5.3159,
"mean_token_accuracy": 0.18887696117162706,
"num_tokens": 26722876.0,
"step": 11670
},
{
"entropy": 5.629246473312378,
"epoch": 1.1541122973507316,
"grad_norm": 1.0,
"learning_rate": 0.00048750479199432753,
"loss": 5.1651,
"mean_token_accuracy": 0.19840702563524246,
"num_tokens": 26733548.0,
"step": 11675
},
{
"entropy": 5.558599376678467,
"epoch": 1.1546065638592329,
"grad_norm": 0.9921875,
"learning_rate": 0.0004874931927432139,
"loss": 5.0994,
"mean_token_accuracy": 0.20664387792348862,
"num_tokens": 26744754.0,
"step": 11680
},
{
"entropy": 5.560123586654663,
"epoch": 1.1551008303677344,
"grad_norm": 0.95703125,
"learning_rate": 0.0004874815882647348,
"loss": 5.2337,
"mean_token_accuracy": 0.19843924045562744,
"num_tokens": 26757416.0,
"step": 11685
},
{
"entropy": 5.56107382774353,
"epoch": 1.1555950968762356,
"grad_norm": 0.96484375,
"learning_rate": 0.0004874699785591756,
"loss": 5.1245,
"mean_token_accuracy": 0.20160049051046372,
"num_tokens": 26768572.0,
"step": 11690
},
{
"entropy": 5.566872215270996,
"epoch": 1.1560893633847371,
"grad_norm": 0.97265625,
"learning_rate": 0.00048745836362682203,
"loss": 5.1182,
"mean_token_accuracy": 0.20748162418603897,
"num_tokens": 26779309.0,
"step": 11695
},
{
"entropy": 5.628466987609864,
"epoch": 1.1565836298932384,
"grad_norm": 1.0546875,
"learning_rate": 0.0004874467434679598,
"loss": 5.194,
"mean_token_accuracy": 0.20000311434268953,
"num_tokens": 26791107.0,
"step": 11700
},
{
"entropy": 5.569167947769165,
"epoch": 1.1570778964017399,
"grad_norm": 1.0390625,
"learning_rate": 0.0004874351180828746,
"loss": 5.1023,
"mean_token_accuracy": 0.20910670161247252,
"num_tokens": 26801334.0,
"step": 11705
},
{
"entropy": 5.599823808670044,
"epoch": 1.1575721629102411,
"grad_norm": 0.91796875,
"learning_rate": 0.0004874234874718526,
"loss": 5.1469,
"mean_token_accuracy": 0.19623263776302338,
"num_tokens": 26812972.0,
"step": 11710
},
{
"entropy": 5.625125408172607,
"epoch": 1.1580664294187426,
"grad_norm": 1.09375,
"learning_rate": 0.00048741185163517984,
"loss": 5.2023,
"mean_token_accuracy": 0.19551342129707336,
"num_tokens": 26823683.0,
"step": 11715
},
{
"entropy": 5.651418924331665,
"epoch": 1.1585606959272439,
"grad_norm": 0.96484375,
"learning_rate": 0.00048740021057314267,
"loss": 5.1796,
"mean_token_accuracy": 0.2049514576792717,
"num_tokens": 26835898.0,
"step": 11720
},
{
"entropy": 5.589020490646362,
"epoch": 1.1590549624357454,
"grad_norm": 0.91015625,
"learning_rate": 0.00048738856428602736,
"loss": 5.1285,
"mean_token_accuracy": 0.20191656947135925,
"num_tokens": 26847363.0,
"step": 11725
},
{
"entropy": 5.5678403854370115,
"epoch": 1.1595492289442468,
"grad_norm": 1.0078125,
"learning_rate": 0.0004873769127741203,
"loss": 5.0846,
"mean_token_accuracy": 0.20453623086214065,
"num_tokens": 26859081.0,
"step": 11730
},
{
"entropy": 5.594265413284302,
"epoch": 1.160043495452748,
"grad_norm": 0.9375,
"learning_rate": 0.0004873652560377083,
"loss": 5.1128,
"mean_token_accuracy": 0.208095084130764,
"num_tokens": 26869966.0,
"step": 11735
},
{
"entropy": 5.583975172042846,
"epoch": 1.1605377619612496,
"grad_norm": 0.9921875,
"learning_rate": 0.000487353594077078,
"loss": 5.1239,
"mean_token_accuracy": 0.20198491364717483,
"num_tokens": 26880673.0,
"step": 11740
},
{
"entropy": 5.602110576629639,
"epoch": 1.1610320284697508,
"grad_norm": 1.078125,
"learning_rate": 0.00048734192689251643,
"loss": 5.1591,
"mean_token_accuracy": 0.2037808582186699,
"num_tokens": 26892306.0,
"step": 11745
},
{
"entropy": 5.659937381744385,
"epoch": 1.1615262949782523,
"grad_norm": 0.92578125,
"learning_rate": 0.0004873302544843104,
"loss": 5.2084,
"mean_token_accuracy": 0.19477305114269255,
"num_tokens": 26904655.0,
"step": 11750
},
{
"entropy": 5.607494354248047,
"epoch": 1.1620205614867536,
"grad_norm": 0.98046875,
"learning_rate": 0.0004873185768527473,
"loss": 5.2134,
"mean_token_accuracy": 0.20037308037281037,
"num_tokens": 26916580.0,
"step": 11755
},
{
"entropy": 5.59541597366333,
"epoch": 1.162514827995255,
"grad_norm": 1.03125,
"learning_rate": 0.00048730689399811427,
"loss": 5.1675,
"mean_token_accuracy": 0.19932375848293304,
"num_tokens": 26927033.0,
"step": 11760
},
{
"entropy": 5.5870747566223145,
"epoch": 1.1630090945037563,
"grad_norm": 1.1171875,
"learning_rate": 0.0004872952059206986,
"loss": 5.2267,
"mean_token_accuracy": 0.19576814472675325,
"num_tokens": 26938594.0,
"step": 11765
},
{
"entropy": 5.582853078842163,
"epoch": 1.1635033610122578,
"grad_norm": 1.09375,
"learning_rate": 0.0004872835126207881,
"loss": 5.1244,
"mean_token_accuracy": 0.20771938413381577,
"num_tokens": 26949595.0,
"step": 11770
},
{
"entropy": 5.660822582244873,
"epoch": 1.163997627520759,
"grad_norm": 0.91796875,
"learning_rate": 0.0004872718140986701,
"loss": 5.2394,
"mean_token_accuracy": 0.19752313047647477,
"num_tokens": 26961692.0,
"step": 11775
},
{
"entropy": 5.625831651687622,
"epoch": 1.1644918940292606,
"grad_norm": 0.99609375,
"learning_rate": 0.00048726011035463265,
"loss": 5.089,
"mean_token_accuracy": 0.20097362101078034,
"num_tokens": 26972062.0,
"step": 11780
},
{
"entropy": 5.627157592773438,
"epoch": 1.164986160537762,
"grad_norm": 1.0234375,
"learning_rate": 0.0004872484013889635,
"loss": 5.1518,
"mean_token_accuracy": 0.2023749440908432,
"num_tokens": 26983353.0,
"step": 11785
},
{
"entropy": 5.638576173782349,
"epoch": 1.1654804270462633,
"grad_norm": 0.99609375,
"learning_rate": 0.00048723668720195067,
"loss": 5.2611,
"mean_token_accuracy": 0.19006135314702988,
"num_tokens": 26994828.0,
"step": 11790
},
{
"entropy": 5.5532777309417725,
"epoch": 1.1659746935547648,
"grad_norm": 0.89453125,
"learning_rate": 0.0004872249677938825,
"loss": 5.1634,
"mean_token_accuracy": 0.2054157242178917,
"num_tokens": 27008311.0,
"step": 11795
},
{
"entropy": 5.545829200744629,
"epoch": 1.166468960063266,
"grad_norm": 0.96484375,
"learning_rate": 0.0004872132431650472,
"loss": 5.1356,
"mean_token_accuracy": 0.20460338443517684,
"num_tokens": 27018984.0,
"step": 11800
},
{
"entropy": 5.6383496761322025,
"epoch": 1.1669632265717675,
"grad_norm": 1.046875,
"learning_rate": 0.00048720151331573317,
"loss": 5.2452,
"mean_token_accuracy": 0.19370283484458922,
"num_tokens": 27029183.0,
"step": 11805
},
{
"entropy": 5.531656646728516,
"epoch": 1.1674574930802688,
"grad_norm": 1.046875,
"learning_rate": 0.000487189778246229,
"loss": 5.0905,
"mean_token_accuracy": 0.21045765280723572,
"num_tokens": 27039661.0,
"step": 11810
},
{
"entropy": 5.641313743591309,
"epoch": 1.1679517595887703,
"grad_norm": 0.94921875,
"learning_rate": 0.00048717803795682334,
"loss": 5.2514,
"mean_token_accuracy": 0.19960915595293044,
"num_tokens": 27051685.0,
"step": 11815
},
{
"entropy": 5.677731704711914,
"epoch": 1.1684460260972718,
"grad_norm": 1.0390625,
"learning_rate": 0.0004871662924478051,
"loss": 5.211,
"mean_token_accuracy": 0.2004316419363022,
"num_tokens": 27062096.0,
"step": 11820
},
{
"entropy": 5.566514873504639,
"epoch": 1.168940292605773,
"grad_norm": 0.9765625,
"learning_rate": 0.0004871545417194631,
"loss": 5.1036,
"mean_token_accuracy": 0.202404448390007,
"num_tokens": 27073059.0,
"step": 11825
},
{
"entropy": 5.648821687698364,
"epoch": 1.1694345591142743,
"grad_norm": 1.0234375,
"learning_rate": 0.00048714278577208643,
"loss": 5.212,
"mean_token_accuracy": 0.19836764186620712,
"num_tokens": 27085272.0,
"step": 11830
},
{
"entropy": 5.663919734954834,
"epoch": 1.1699288256227758,
"grad_norm": 0.92578125,
"learning_rate": 0.0004871310246059644,
"loss": 5.1884,
"mean_token_accuracy": 0.19877843856811522,
"num_tokens": 27096651.0,
"step": 11835
},
{
"entropy": 5.591395521163941,
"epoch": 1.1704230921312773,
"grad_norm": 0.96875,
"learning_rate": 0.00048711925822138627,
"loss": 5.1339,
"mean_token_accuracy": 0.20013082921504974,
"num_tokens": 27108180.0,
"step": 11840
},
{
"entropy": 5.65935173034668,
"epoch": 1.1709173586397785,
"grad_norm": 1.03125,
"learning_rate": 0.00048710748661864135,
"loss": 5.2014,
"mean_token_accuracy": 0.1983470931649208,
"num_tokens": 27118902.0,
"step": 11845
},
{
"entropy": 5.627192831039428,
"epoch": 1.17141162514828,
"grad_norm": 0.9765625,
"learning_rate": 0.00048709570979801946,
"loss": 5.1632,
"mean_token_accuracy": 0.20421404987573624,
"num_tokens": 27131279.0,
"step": 11850
},
{
"entropy": 5.55908317565918,
"epoch": 1.1719058916567813,
"grad_norm": 1.0234375,
"learning_rate": 0.00048708392775981014,
"loss": 5.1341,
"mean_token_accuracy": 0.20507726520299913,
"num_tokens": 27142983.0,
"step": 11855
},
{
"entropy": 5.625089979171753,
"epoch": 1.1724001581652828,
"grad_norm": 1.0234375,
"learning_rate": 0.0004870721405043033,
"loss": 5.1184,
"mean_token_accuracy": 0.2035112828016281,
"num_tokens": 27154516.0,
"step": 11860
},
{
"entropy": 5.62056975364685,
"epoch": 1.172894424673784,
"grad_norm": 1.0078125,
"learning_rate": 0.0004870603480317889,
"loss": 5.197,
"mean_token_accuracy": 0.20283437967300416,
"num_tokens": 27164717.0,
"step": 11865
},
{
"entropy": 5.61588191986084,
"epoch": 1.1733886911822855,
"grad_norm": 1.0546875,
"learning_rate": 0.00048704855034255713,
"loss": 5.1914,
"mean_token_accuracy": 0.20290824621915818,
"num_tokens": 27174830.0,
"step": 11870
},
{
"entropy": 5.58607234954834,
"epoch": 1.173882957690787,
"grad_norm": 0.9765625,
"learning_rate": 0.00048703674743689806,
"loss": 5.1683,
"mean_token_accuracy": 0.1991803303360939,
"num_tokens": 27186724.0,
"step": 11875
},
{
"entropy": 5.5404561996459964,
"epoch": 1.1743772241992882,
"grad_norm": 0.9140625,
"learning_rate": 0.0004870249393151021,
"loss": 5.129,
"mean_token_accuracy": 0.20301436632871628,
"num_tokens": 27198416.0,
"step": 11880
},
{
"entropy": 5.570279598236084,
"epoch": 1.1748714907077897,
"grad_norm": 1.046875,
"learning_rate": 0.00048701312597745986,
"loss": 5.0545,
"mean_token_accuracy": 0.2034875750541687,
"num_tokens": 27209347.0,
"step": 11885
},
{
"entropy": 5.626441144943238,
"epoch": 1.175365757216291,
"grad_norm": 0.9765625,
"learning_rate": 0.00048700130742426165,
"loss": 5.1498,
"mean_token_accuracy": 0.19788456857204437,
"num_tokens": 27221061.0,
"step": 11890
},
{
"entropy": 5.538058948516846,
"epoch": 1.1758600237247925,
"grad_norm": 0.875,
"learning_rate": 0.0004869894836557985,
"loss": 5.1099,
"mean_token_accuracy": 0.2129423052072525,
"num_tokens": 27233436.0,
"step": 11895
},
{
"entropy": 5.573163318634033,
"epoch": 1.1763542902332937,
"grad_norm": 0.90234375,
"learning_rate": 0.00048697765467236117,
"loss": 5.1184,
"mean_token_accuracy": 0.21005520075559617,
"num_tokens": 27246308.0,
"step": 11900
},
{
"entropy": 5.593556213378906,
"epoch": 1.1768485567417952,
"grad_norm": 0.93359375,
"learning_rate": 0.00048696582047424053,
"loss": 5.1742,
"mean_token_accuracy": 0.199239319562912,
"num_tokens": 27258187.0,
"step": 11905
},
{
"entropy": 5.5561316967010494,
"epoch": 1.1773428232502965,
"grad_norm": 0.9765625,
"learning_rate": 0.00048695398106172793,
"loss": 5.1424,
"mean_token_accuracy": 0.20541492998600006,
"num_tokens": 27269411.0,
"step": 11910
},
{
"entropy": 5.611875343322754,
"epoch": 1.177837089758798,
"grad_norm": 0.9453125,
"learning_rate": 0.0004869421364351145,
"loss": 5.1586,
"mean_token_accuracy": 0.20315765738487243,
"num_tokens": 27280592.0,
"step": 11915
},
{
"entropy": 5.683233690261841,
"epoch": 1.1783313562672992,
"grad_norm": 0.95703125,
"learning_rate": 0.00048693028659469156,
"loss": 5.2272,
"mean_token_accuracy": 0.1890906199812889,
"num_tokens": 27292151.0,
"step": 11920
},
{
"entropy": 5.610967111587525,
"epoch": 1.1788256227758007,
"grad_norm": 0.9453125,
"learning_rate": 0.00048691843154075075,
"loss": 5.1012,
"mean_token_accuracy": 0.2052637293934822,
"num_tokens": 27305371.0,
"step": 11925
},
{
"entropy": 5.579727268218994,
"epoch": 1.1793198892843022,
"grad_norm": 0.9296875,
"learning_rate": 0.00048690657127358367,
"loss": 5.2039,
"mean_token_accuracy": 0.1920991212129593,
"num_tokens": 27316810.0,
"step": 11930
},
{
"entropy": 5.588397884368897,
"epoch": 1.1798141557928035,
"grad_norm": 0.8828125,
"learning_rate": 0.00048689470579348207,
"loss": 5.1676,
"mean_token_accuracy": 0.20227353870868683,
"num_tokens": 27328412.0,
"step": 11935
},
{
"entropy": 5.6597496509552006,
"epoch": 1.180308422301305,
"grad_norm": 1.0546875,
"learning_rate": 0.00048688283510073775,
"loss": 5.2052,
"mean_token_accuracy": 0.19740669429302216,
"num_tokens": 27340216.0,
"step": 11940
},
{
"entropy": 5.609868097305298,
"epoch": 1.1808026888098062,
"grad_norm": 0.984375,
"learning_rate": 0.0004868709591956429,
"loss": 5.072,
"mean_token_accuracy": 0.203493170440197,
"num_tokens": 27350290.0,
"step": 11945
},
{
"entropy": 5.589484691619873,
"epoch": 1.1812969553183077,
"grad_norm": 1.0234375,
"learning_rate": 0.00048685907807848955,
"loss": 5.2432,
"mean_token_accuracy": 0.19132159352302552,
"num_tokens": 27360639.0,
"step": 11950
},
{
"entropy": 5.528190088272095,
"epoch": 1.181791221826809,
"grad_norm": 0.953125,
"learning_rate": 0.00048684719174957005,
"loss": 5.0878,
"mean_token_accuracy": 0.2117920696735382,
"num_tokens": 27372059.0,
"step": 11955
},
{
"entropy": 5.626075744628906,
"epoch": 1.1822854883353104,
"grad_norm": 0.92578125,
"learning_rate": 0.0004868353002091767,
"loss": 5.2058,
"mean_token_accuracy": 0.19016459286212922,
"num_tokens": 27384308.0,
"step": 11960
},
{
"entropy": 5.563534784317016,
"epoch": 1.1827797548438117,
"grad_norm": 0.96484375,
"learning_rate": 0.0004868234034576021,
"loss": 5.0519,
"mean_token_accuracy": 0.21657481789588928,
"num_tokens": 27395510.0,
"step": 11965
},
{
"entropy": 5.565920400619507,
"epoch": 1.1832740213523132,
"grad_norm": 0.984375,
"learning_rate": 0.000486811501495139,
"loss": 5.1294,
"mean_token_accuracy": 0.19558025002479554,
"num_tokens": 27406697.0,
"step": 11970
},
{
"entropy": 5.576605987548828,
"epoch": 1.1837682878608144,
"grad_norm": 1.0234375,
"learning_rate": 0.0004867995943220801,
"loss": 5.1752,
"mean_token_accuracy": 0.2068494588136673,
"num_tokens": 27418047.0,
"step": 11975
},
{
"entropy": 5.6756151676177975,
"epoch": 1.184262554369316,
"grad_norm": 1.0078125,
"learning_rate": 0.00048678768193871825,
"loss": 5.2634,
"mean_token_accuracy": 0.19117049127817154,
"num_tokens": 27430044.0,
"step": 11980
},
{
"entropy": 5.576125001907348,
"epoch": 1.1847568208778174,
"grad_norm": 0.9765625,
"learning_rate": 0.0004867757643453466,
"loss": 5.1781,
"mean_token_accuracy": 0.19635282158851625,
"num_tokens": 27441392.0,
"step": 11985
},
{
"entropy": 5.6109843254089355,
"epoch": 1.1852510873863187,
"grad_norm": 0.96484375,
"learning_rate": 0.00048676384154225827,
"loss": 5.2971,
"mean_token_accuracy": 0.19204170107841492,
"num_tokens": 27451356.0,
"step": 11990
},
{
"entropy": 5.580730199813843,
"epoch": 1.1857453538948202,
"grad_norm": 1.015625,
"learning_rate": 0.0004867519135297466,
"loss": 5.1137,
"mean_token_accuracy": 0.21453266739845275,
"num_tokens": 27462358.0,
"step": 11995
},
{
"entropy": 5.562673139572143,
"epoch": 1.1862396204033214,
"grad_norm": 0.9609375,
"learning_rate": 0.00048673998030810507,
"loss": 5.0888,
"mean_token_accuracy": 0.2045995220541954,
"num_tokens": 27474144.0,
"step": 12000
},
{
"epoch": 1.1862396204033214,
"eval_entropy": 5.383485007667307,
"eval_loss": 5.234320640563965,
"eval_mean_token_accuracy": 0.20572784973759786,
"eval_num_tokens": 27474144.0,
"eval_runtime": 20.2804,
"eval_samples_per_second": 1603.177,
"eval_steps_per_second": 200.44,
"step": 12000
}
],
"logging_steps": 5,
"max_steps": 101150,
"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": 6382004135086080.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink