fpadovani/tur-latn-10mb-100mb_seed10
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
tur-latn-10mb-100mb_seed10/checkpoint-9000/trainer_state.json
ModelHub XC 3f77a82d44 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/tur-latn-10mb-100mb_seed10
Source: Original Platform
2026-09-28 20:59:09 +08:00

18068 lines
495 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.899685110211426,
"eval_steps": 3000,
"global_step": 9000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.742113018035889,
"epoch": 0.00049982506122857,
"grad_norm": 4.875,
"learning_rate": 2e-06,
"loss": 10.7354,
"mean_token_accuracy": 0.0002496711676940322,
"num_tokens": 8821.0,
"step": 5
},
{
"entropy": 10.742119884490966,
"epoch": 0.00099965012245714,
"grad_norm": 5.125,
"learning_rate": 4.5e-06,
"loss": 10.7497,
"mean_token_accuracy": 0.0,
"num_tokens": 18288.0,
"step": 10
},
{
"entropy": 10.742119789123535,
"epoch": 0.00149947518368571,
"grad_norm": 4.65625,
"learning_rate": 7e-06,
"loss": 10.7421,
"mean_token_accuracy": 0.0002510460326448083,
"num_tokens": 28664.0,
"step": 15
},
{
"entropy": 10.742084407806397,
"epoch": 0.00199930024491428,
"grad_norm": 6.0,
"learning_rate": 9.5e-06,
"loss": 10.6882,
"mean_token_accuracy": 0.00013869625981897116,
"num_tokens": 37807.0,
"step": 20
},
{
"entropy": 10.74200439453125,
"epoch": 0.00249912530614285,
"grad_norm": 4.375,
"learning_rate": 1.2e-05,
"loss": 10.6462,
"mean_token_accuracy": 0.004488382575800642,
"num_tokens": 48086.0,
"step": 25
},
{
"entropy": 10.741621685028075,
"epoch": 0.00299895036737142,
"grad_norm": 4.03125,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.5223,
"mean_token_accuracy": 0.030498435720801353,
"num_tokens": 57427.0,
"step": 30
},
{
"entropy": 10.740404415130616,
"epoch": 0.00349877542859999,
"grad_norm": 3.453125,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.4113,
"mean_token_accuracy": 0.04445975758135319,
"num_tokens": 66662.0,
"step": 35
},
{
"entropy": 10.736149501800536,
"epoch": 0.00399860048982856,
"grad_norm": 2.484375,
"learning_rate": 1.95e-05,
"loss": 10.3341,
"mean_token_accuracy": 0.04500314220786095,
"num_tokens": 77388.0,
"step": 40
},
{
"entropy": 10.725442790985108,
"epoch": 0.00449842555105713,
"grad_norm": 2.125,
"learning_rate": 2.2e-05,
"loss": 10.1941,
"mean_token_accuracy": 0.04626995846629143,
"num_tokens": 86093.0,
"step": 45
},
{
"entropy": 10.711077690124512,
"epoch": 0.0049982506122857,
"grad_norm": 1.9140625,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.1926,
"mean_token_accuracy": 0.04310738146305084,
"num_tokens": 96097.0,
"step": 50
},
{
"entropy": 10.69948215484619,
"epoch": 0.00549807567351427,
"grad_norm": 1.7890625,
"learning_rate": 2.7e-05,
"loss": 10.1482,
"mean_token_accuracy": 0.042906814627349375,
"num_tokens": 106328.0,
"step": 55
},
{
"entropy": 10.692364120483399,
"epoch": 0.00599790073474284,
"grad_norm": 1.8046875,
"learning_rate": 2.95e-05,
"loss": 10.0247,
"mean_token_accuracy": 0.04474271964281797,
"num_tokens": 114822.0,
"step": 60
},
{
"entropy": 10.684545135498047,
"epoch": 0.00649772579597141,
"grad_norm": 1.8671875,
"learning_rate": 3.2e-05,
"loss": 10.0297,
"mean_token_accuracy": 0.04712004661560058,
"num_tokens": 124984.0,
"step": 65
},
{
"entropy": 10.676221084594726,
"epoch": 0.00699755085719998,
"grad_norm": 1.6328125,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.9346,
"mean_token_accuracy": 0.04551910758018494,
"num_tokens": 134054.0,
"step": 70
},
{
"entropy": 10.668924236297608,
"epoch": 0.00749737591842855,
"grad_norm": 1.65625,
"learning_rate": 3.7e-05,
"loss": 9.9081,
"mean_token_accuracy": 0.04679625704884529,
"num_tokens": 144561.0,
"step": 75
},
{
"entropy": 10.661227989196778,
"epoch": 0.00799720097965712,
"grad_norm": 1.671875,
"learning_rate": 3.95e-05,
"loss": 9.8997,
"mean_token_accuracy": 0.0436519380658865,
"num_tokens": 153701.0,
"step": 80
},
{
"entropy": 10.653296184539794,
"epoch": 0.00849702604088569,
"grad_norm": 1.6640625,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.7945,
"mean_token_accuracy": 0.05020407922565937,
"num_tokens": 164291.0,
"step": 85
},
{
"entropy": 10.636521530151366,
"epoch": 0.00899685110211426,
"grad_norm": 1.6015625,
"learning_rate": 4.45e-05,
"loss": 9.7719,
"mean_token_accuracy": 0.05209365636110306,
"num_tokens": 174136.0,
"step": 90
},
{
"entropy": 10.617370223999023,
"epoch": 0.00949667616334283,
"grad_norm": 1.4765625,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.708,
"mean_token_accuracy": 0.05729338489472866,
"num_tokens": 184368.0,
"step": 95
},
{
"entropy": 10.591236019134522,
"epoch": 0.0099965012245714,
"grad_norm": 1.46875,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.6914,
"mean_token_accuracy": 0.056233976408839224,
"num_tokens": 194659.0,
"step": 100
},
{
"entropy": 10.571689224243164,
"epoch": 0.01049632628579997,
"grad_norm": 1.546875,
"learning_rate": 5.2e-05,
"loss": 9.5729,
"mean_token_accuracy": 0.057006143406033515,
"num_tokens": 204875.0,
"step": 105
},
{
"entropy": 10.52930965423584,
"epoch": 0.01099615134702854,
"grad_norm": 1.90625,
"learning_rate": 5.45e-05,
"loss": 9.577,
"mean_token_accuracy": 0.05334683395922184,
"num_tokens": 217033.0,
"step": 110
},
{
"entropy": 10.460849475860595,
"epoch": 0.01149597640825711,
"grad_norm": 1.359375,
"learning_rate": 5.7e-05,
"loss": 9.4015,
"mean_token_accuracy": 0.062102947011590005,
"num_tokens": 225946.0,
"step": 115
},
{
"entropy": 10.396935844421387,
"epoch": 0.01199580146948568,
"grad_norm": 1.3515625,
"learning_rate": 5.9499999999999996e-05,
"loss": 9.3632,
"mean_token_accuracy": 0.056867911294102666,
"num_tokens": 235413.0,
"step": 120
},
{
"entropy": 10.36595640182495,
"epoch": 0.01249562653071425,
"grad_norm": 1.1640625,
"learning_rate": 6.2e-05,
"loss": 9.3793,
"mean_token_accuracy": 0.048073610663414,
"num_tokens": 247913.0,
"step": 125
},
{
"entropy": 10.35603609085083,
"epoch": 0.01299545159194282,
"grad_norm": 1.234375,
"learning_rate": 6.450000000000001e-05,
"loss": 9.2476,
"mean_token_accuracy": 0.06040509082376957,
"num_tokens": 258486.0,
"step": 130
},
{
"entropy": 10.29154281616211,
"epoch": 0.01349527665317139,
"grad_norm": 1.1015625,
"learning_rate": 6.7e-05,
"loss": 9.161,
"mean_token_accuracy": 0.05578649267554283,
"num_tokens": 268027.0,
"step": 135
},
{
"entropy": 10.207606983184814,
"epoch": 0.01399510171439996,
"grad_norm": 1.1328125,
"learning_rate": 6.950000000000001e-05,
"loss": 9.0596,
"mean_token_accuracy": 0.06114833988249302,
"num_tokens": 277470.0,
"step": 140
},
{
"entropy": 10.14081039428711,
"epoch": 0.01449492677562853,
"grad_norm": 1.0625,
"learning_rate": 7.2e-05,
"loss": 8.9902,
"mean_token_accuracy": 0.05893770642578602,
"num_tokens": 286891.0,
"step": 145
},
{
"entropy": 10.036300086975098,
"epoch": 0.0149947518368571,
"grad_norm": 1.0703125,
"learning_rate": 7.45e-05,
"loss": 8.8746,
"mean_token_accuracy": 0.06821754761040211,
"num_tokens": 295636.0,
"step": 150
},
{
"entropy": 9.924018096923827,
"epoch": 0.01549457689808567,
"grad_norm": 0.9765625,
"learning_rate": 7.7e-05,
"loss": 8.8147,
"mean_token_accuracy": 0.06259672716259956,
"num_tokens": 306542.0,
"step": 155
},
{
"entropy": 9.83233995437622,
"epoch": 0.01599440195931424,
"grad_norm": 0.96484375,
"learning_rate": 7.950000000000001e-05,
"loss": 8.7427,
"mean_token_accuracy": 0.060157052055001256,
"num_tokens": 315956.0,
"step": 160
},
{
"entropy": 9.716417980194091,
"epoch": 0.01649422702054281,
"grad_norm": 0.90234375,
"learning_rate": 8.2e-05,
"loss": 8.7001,
"mean_token_accuracy": 0.05990239083766937,
"num_tokens": 325500.0,
"step": 165
},
{
"entropy": 9.563173198699952,
"epoch": 0.01699405208177138,
"grad_norm": 0.96484375,
"learning_rate": 8.450000000000001e-05,
"loss": 8.6683,
"mean_token_accuracy": 0.06383000239729882,
"num_tokens": 335194.0,
"step": 170
},
{
"entropy": 9.43425817489624,
"epoch": 0.01749387714299995,
"grad_norm": 0.83203125,
"learning_rate": 8.7e-05,
"loss": 8.6082,
"mean_token_accuracy": 0.061490171030163764,
"num_tokens": 344717.0,
"step": 175
},
{
"entropy": 9.373362159729004,
"epoch": 0.01799370220422852,
"grad_norm": 0.875,
"learning_rate": 8.95e-05,
"loss": 8.5056,
"mean_token_accuracy": 0.06233922764658928,
"num_tokens": 354476.0,
"step": 180
},
{
"entropy": 9.256193733215332,
"epoch": 0.01849352726545709,
"grad_norm": 0.91796875,
"learning_rate": 9.2e-05,
"loss": 8.5486,
"mean_token_accuracy": 0.06391786970198154,
"num_tokens": 363939.0,
"step": 185
},
{
"entropy": 9.170073986053467,
"epoch": 0.01899335232668566,
"grad_norm": 0.87109375,
"learning_rate": 9.45e-05,
"loss": 8.5757,
"mean_token_accuracy": 0.061041193455457686,
"num_tokens": 373187.0,
"step": 190
},
{
"entropy": 9.10431718826294,
"epoch": 0.01949317738791423,
"grad_norm": 0.80859375,
"learning_rate": 9.7e-05,
"loss": 8.4889,
"mean_token_accuracy": 0.06612157672643662,
"num_tokens": 382999.0,
"step": 195
},
{
"entropy": 9.082199668884277,
"epoch": 0.0199930024491428,
"grad_norm": 0.83984375,
"learning_rate": 9.95e-05,
"loss": 8.5374,
"mean_token_accuracy": 0.061392434686422345,
"num_tokens": 392639.0,
"step": 200
},
{
"entropy": 9.000153255462646,
"epoch": 0.02049282751037137,
"grad_norm": 0.80859375,
"learning_rate": 0.000102,
"loss": 8.5164,
"mean_token_accuracy": 0.06042192354798317,
"num_tokens": 403089.0,
"step": 205
},
{
"entropy": 8.976395606994629,
"epoch": 0.02099265257159994,
"grad_norm": 0.80078125,
"learning_rate": 0.00010449999999999999,
"loss": 8.4681,
"mean_token_accuracy": 0.06152507662773132,
"num_tokens": 412321.0,
"step": 210
},
{
"entropy": 8.91478672027588,
"epoch": 0.02149247763282851,
"grad_norm": 0.97265625,
"learning_rate": 0.000107,
"loss": 8.502,
"mean_token_accuracy": 0.06330714970827103,
"num_tokens": 421599.0,
"step": 215
},
{
"entropy": 8.860779762268066,
"epoch": 0.02199230269405708,
"grad_norm": 0.84375,
"learning_rate": 0.0001095,
"loss": 8.4778,
"mean_token_accuracy": 0.06263926923274994,
"num_tokens": 432012.0,
"step": 220
},
{
"entropy": 8.922132301330567,
"epoch": 0.02249212775528565,
"grad_norm": 0.92578125,
"learning_rate": 0.000112,
"loss": 8.453,
"mean_token_accuracy": 0.06589648500084877,
"num_tokens": 441989.0,
"step": 225
},
{
"entropy": 8.896614170074463,
"epoch": 0.02299195281651422,
"grad_norm": 0.9140625,
"learning_rate": 0.0001145,
"loss": 8.4297,
"mean_token_accuracy": 0.06669879406690597,
"num_tokens": 451532.0,
"step": 230
},
{
"entropy": 8.810688591003418,
"epoch": 0.02349177787774279,
"grad_norm": 0.83203125,
"learning_rate": 0.00011700000000000001,
"loss": 8.4408,
"mean_token_accuracy": 0.06297187581658363,
"num_tokens": 461815.0,
"step": 235
},
{
"entropy": 8.825593757629395,
"epoch": 0.02399160293897136,
"grad_norm": 0.96484375,
"learning_rate": 0.00011949999999999999,
"loss": 8.5024,
"mean_token_accuracy": 0.06163628771901131,
"num_tokens": 470860.0,
"step": 240
},
{
"entropy": 8.804152870178223,
"epoch": 0.02449142800019993,
"grad_norm": 0.7265625,
"learning_rate": 0.000122,
"loss": 8.4705,
"mean_token_accuracy": 0.06292311102151871,
"num_tokens": 480649.0,
"step": 245
},
{
"entropy": 8.85675745010376,
"epoch": 0.0249912530614285,
"grad_norm": 0.90625,
"learning_rate": 0.0001245,
"loss": 8.4032,
"mean_token_accuracy": 0.06878387741744518,
"num_tokens": 489969.0,
"step": 250
},
{
"entropy": 8.763758945465089,
"epoch": 0.02549107812265707,
"grad_norm": 0.7421875,
"learning_rate": 0.000127,
"loss": 8.4088,
"mean_token_accuracy": 0.06565447077155113,
"num_tokens": 500000.0,
"step": 255
},
{
"entropy": 8.834418106079102,
"epoch": 0.02599090318388564,
"grad_norm": 0.8515625,
"learning_rate": 0.0001295,
"loss": 8.3918,
"mean_token_accuracy": 0.06619769148528576,
"num_tokens": 510013.0,
"step": 260
},
{
"entropy": 8.717261409759521,
"epoch": 0.026490728245114212,
"grad_norm": 0.8984375,
"learning_rate": 0.000132,
"loss": 8.4183,
"mean_token_accuracy": 0.06544177196919918,
"num_tokens": 519698.0,
"step": 265
},
{
"entropy": 8.820175933837891,
"epoch": 0.02699055330634278,
"grad_norm": 0.81640625,
"learning_rate": 0.00013450000000000002,
"loss": 8.4225,
"mean_token_accuracy": 0.06441575214266777,
"num_tokens": 531631.0,
"step": 270
},
{
"entropy": 8.775634002685546,
"epoch": 0.02749037836757135,
"grad_norm": 0.984375,
"learning_rate": 0.00013700000000000002,
"loss": 8.3921,
"mean_token_accuracy": 0.0714694295078516,
"num_tokens": 542204.0,
"step": 275
},
{
"entropy": 8.750507164001466,
"epoch": 0.02799020342879992,
"grad_norm": 0.90234375,
"learning_rate": 0.0001395,
"loss": 8.313,
"mean_token_accuracy": 0.06778912842273713,
"num_tokens": 552269.0,
"step": 280
},
{
"entropy": 8.71799602508545,
"epoch": 0.02849002849002849,
"grad_norm": 1.0234375,
"learning_rate": 0.00014199999999999998,
"loss": 8.3159,
"mean_token_accuracy": 0.0691824123263359,
"num_tokens": 561907.0,
"step": 285
},
{
"entropy": 8.687597274780273,
"epoch": 0.02898985355125706,
"grad_norm": 0.9375,
"learning_rate": 0.0001445,
"loss": 8.426,
"mean_token_accuracy": 0.06967848278582096,
"num_tokens": 571283.0,
"step": 290
},
{
"entropy": 8.732817840576171,
"epoch": 0.02948967861248563,
"grad_norm": 0.87109375,
"learning_rate": 0.000147,
"loss": 8.2974,
"mean_token_accuracy": 0.07371204271912575,
"num_tokens": 579896.0,
"step": 295
},
{
"entropy": 8.68296022415161,
"epoch": 0.0299895036737142,
"grad_norm": 1.0625,
"learning_rate": 0.0001495,
"loss": 8.2576,
"mean_token_accuracy": 0.08281500861048699,
"num_tokens": 588856.0,
"step": 300
},
{
"entropy": 8.575243282318116,
"epoch": 0.03048932873494277,
"grad_norm": 0.94921875,
"learning_rate": 0.000152,
"loss": 8.3118,
"mean_token_accuracy": 0.07554283365607262,
"num_tokens": 598910.0,
"step": 305
},
{
"entropy": 8.752356338500977,
"epoch": 0.03098915379617134,
"grad_norm": 0.94140625,
"learning_rate": 0.00015450000000000001,
"loss": 8.3723,
"mean_token_accuracy": 0.0699452206492424,
"num_tokens": 609248.0,
"step": 310
},
{
"entropy": 8.696213340759277,
"epoch": 0.03148897885739991,
"grad_norm": 0.8515625,
"learning_rate": 0.000157,
"loss": 8.3617,
"mean_token_accuracy": 0.07190549783408642,
"num_tokens": 619352.0,
"step": 315
},
{
"entropy": 8.718466758728027,
"epoch": 0.03198880391862848,
"grad_norm": 0.96875,
"learning_rate": 0.0001595,
"loss": 8.3708,
"mean_token_accuracy": 0.07149303182959557,
"num_tokens": 628300.0,
"step": 320
},
{
"entropy": 8.677219200134278,
"epoch": 0.03248862897985705,
"grad_norm": 1.1171875,
"learning_rate": 0.000162,
"loss": 8.3068,
"mean_token_accuracy": 0.0743933729827404,
"num_tokens": 637969.0,
"step": 325
},
{
"entropy": 8.681064319610595,
"epoch": 0.03298845404108562,
"grad_norm": 0.8984375,
"learning_rate": 0.00016450000000000001,
"loss": 8.2194,
"mean_token_accuracy": 0.07997227124869824,
"num_tokens": 647518.0,
"step": 330
},
{
"entropy": 8.710070419311524,
"epoch": 0.03348827910231419,
"grad_norm": 2.03125,
"learning_rate": 0.00016700000000000002,
"loss": 8.401,
"mean_token_accuracy": 0.07310340739786625,
"num_tokens": 658292.0,
"step": 335
},
{
"entropy": 8.64979133605957,
"epoch": 0.03398810416354276,
"grad_norm": 1.078125,
"learning_rate": 0.00016950000000000003,
"loss": 8.282,
"mean_token_accuracy": 0.0751060452312231,
"num_tokens": 667872.0,
"step": 340
},
{
"entropy": 8.58129940032959,
"epoch": 0.03448792922477133,
"grad_norm": 0.98046875,
"learning_rate": 0.00017199999999999998,
"loss": 8.195,
"mean_token_accuracy": 0.0765431810170412,
"num_tokens": 676976.0,
"step": 345
},
{
"entropy": 8.645341205596925,
"epoch": 0.0349877542859999,
"grad_norm": 1.4609375,
"learning_rate": 0.00017449999999999999,
"loss": 8.2187,
"mean_token_accuracy": 0.08100063130259513,
"num_tokens": 686149.0,
"step": 350
},
{
"entropy": 8.57761583328247,
"epoch": 0.03548757934722847,
"grad_norm": 0.85546875,
"learning_rate": 0.000177,
"loss": 8.1215,
"mean_token_accuracy": 0.08577202558517456,
"num_tokens": 696439.0,
"step": 355
},
{
"entropy": 8.503854370117187,
"epoch": 0.03598740440845704,
"grad_norm": 0.91796875,
"learning_rate": 0.0001795,
"loss": 8.1526,
"mean_token_accuracy": 0.08154284283518791,
"num_tokens": 705902.0,
"step": 360
},
{
"entropy": 8.575439071655273,
"epoch": 0.03648722946968561,
"grad_norm": 0.90234375,
"learning_rate": 0.000182,
"loss": 8.304,
"mean_token_accuracy": 0.07753718234598636,
"num_tokens": 715973.0,
"step": 365
},
{
"entropy": 8.580668735504151,
"epoch": 0.03698705453091418,
"grad_norm": 0.90234375,
"learning_rate": 0.0001845,
"loss": 8.2083,
"mean_token_accuracy": 0.07895785272121429,
"num_tokens": 725463.0,
"step": 370
},
{
"entropy": 8.613331699371338,
"epoch": 0.03748687959214275,
"grad_norm": 0.96484375,
"learning_rate": 0.000187,
"loss": 8.2239,
"mean_token_accuracy": 0.07793777883052826,
"num_tokens": 734941.0,
"step": 375
},
{
"entropy": 8.614311981201173,
"epoch": 0.03798670465337132,
"grad_norm": 1.1015625,
"learning_rate": 0.0001895,
"loss": 8.2425,
"mean_token_accuracy": 0.07460713312029839,
"num_tokens": 744302.0,
"step": 380
},
{
"entropy": 8.519310474395752,
"epoch": 0.03848652971459989,
"grad_norm": 1.0078125,
"learning_rate": 0.000192,
"loss": 8.1684,
"mean_token_accuracy": 0.08556054756045342,
"num_tokens": 754341.0,
"step": 385
},
{
"entropy": 8.755522155761719,
"epoch": 0.03898635477582846,
"grad_norm": 0.9609375,
"learning_rate": 0.0001945,
"loss": 8.2985,
"mean_token_accuracy": 0.07072081565856933,
"num_tokens": 764087.0,
"step": 390
},
{
"entropy": 8.485608291625976,
"epoch": 0.03948617983705703,
"grad_norm": 0.96875,
"learning_rate": 0.00019700000000000002,
"loss": 8.2273,
"mean_token_accuracy": 0.07447122037410736,
"num_tokens": 773728.0,
"step": 395
},
{
"entropy": 8.685750102996826,
"epoch": 0.0399860048982856,
"grad_norm": 0.890625,
"learning_rate": 0.00019950000000000002,
"loss": 8.2393,
"mean_token_accuracy": 0.07686416134238243,
"num_tokens": 784839.0,
"step": 400
},
{
"entropy": 8.61361484527588,
"epoch": 0.04048582995951417,
"grad_norm": 0.95703125,
"learning_rate": 0.000202,
"loss": 8.2227,
"mean_token_accuracy": 0.08219211772084237,
"num_tokens": 794514.0,
"step": 405
},
{
"entropy": 8.532085800170899,
"epoch": 0.04098565502074274,
"grad_norm": 0.88671875,
"learning_rate": 0.00020449999999999998,
"loss": 8.2007,
"mean_token_accuracy": 0.08283689394593238,
"num_tokens": 805254.0,
"step": 410
},
{
"entropy": 8.567570400238036,
"epoch": 0.04148548008197131,
"grad_norm": 1.03125,
"learning_rate": 0.000207,
"loss": 8.107,
"mean_token_accuracy": 0.08443296402692795,
"num_tokens": 814135.0,
"step": 415
},
{
"entropy": 8.450470733642579,
"epoch": 0.04198530514319988,
"grad_norm": 0.890625,
"learning_rate": 0.0002095,
"loss": 8.1524,
"mean_token_accuracy": 0.08148454427719116,
"num_tokens": 823531.0,
"step": 420
},
{
"entropy": 8.421721839904786,
"epoch": 0.04248513020442845,
"grad_norm": 0.9765625,
"learning_rate": 0.000212,
"loss": 8.0779,
"mean_token_accuracy": 0.08629797920584678,
"num_tokens": 832833.0,
"step": 425
},
{
"entropy": 8.48620538711548,
"epoch": 0.04298495526565702,
"grad_norm": 0.91015625,
"learning_rate": 0.0002145,
"loss": 8.0685,
"mean_token_accuracy": 0.08419421762228012,
"num_tokens": 841541.0,
"step": 430
},
{
"entropy": 8.445399856567382,
"epoch": 0.04348478032688559,
"grad_norm": 1.0234375,
"learning_rate": 0.00021700000000000002,
"loss": 8.0899,
"mean_token_accuracy": 0.08340806141495705,
"num_tokens": 850689.0,
"step": 435
},
{
"entropy": 8.459619426727295,
"epoch": 0.04398460538811416,
"grad_norm": 0.90234375,
"learning_rate": 0.0002195,
"loss": 8.0362,
"mean_token_accuracy": 0.09014891162514686,
"num_tokens": 860573.0,
"step": 440
},
{
"entropy": 8.543528842926026,
"epoch": 0.04448443044934273,
"grad_norm": 1.03125,
"learning_rate": 0.000222,
"loss": 8.1546,
"mean_token_accuracy": 0.07898386307060719,
"num_tokens": 869784.0,
"step": 445
},
{
"entropy": 8.455932331085204,
"epoch": 0.0449842555105713,
"grad_norm": 0.94921875,
"learning_rate": 0.0002245,
"loss": 8.1017,
"mean_token_accuracy": 0.08015754446387291,
"num_tokens": 878716.0,
"step": 450
},
{
"entropy": 8.446991348266602,
"epoch": 0.04548408057179987,
"grad_norm": 0.92578125,
"learning_rate": 0.00022700000000000002,
"loss": 8.175,
"mean_token_accuracy": 0.08237311467528344,
"num_tokens": 888334.0,
"step": 455
},
{
"entropy": 8.467612648010254,
"epoch": 0.04598390563302844,
"grad_norm": 1.171875,
"learning_rate": 0.00022950000000000002,
"loss": 8.1105,
"mean_token_accuracy": 0.07949303276836872,
"num_tokens": 897701.0,
"step": 460
},
{
"entropy": 8.498649311065673,
"epoch": 0.04648373069425701,
"grad_norm": 1.0390625,
"learning_rate": 0.00023200000000000003,
"loss": 8.0756,
"mean_token_accuracy": 0.0814065046608448,
"num_tokens": 907278.0,
"step": 465
},
{
"entropy": 8.423473739624024,
"epoch": 0.04698355575548558,
"grad_norm": 0.8984375,
"learning_rate": 0.00023449999999999998,
"loss": 8.0835,
"mean_token_accuracy": 0.07995141670107841,
"num_tokens": 916194.0,
"step": 470
},
{
"entropy": 8.476182556152343,
"epoch": 0.04748338081671415,
"grad_norm": 0.93359375,
"learning_rate": 0.000237,
"loss": 8.0997,
"mean_token_accuracy": 0.08503060936927795,
"num_tokens": 926743.0,
"step": 475
},
{
"entropy": 8.453644371032714,
"epoch": 0.04798320587794272,
"grad_norm": 0.8828125,
"learning_rate": 0.0002395,
"loss": 8.1388,
"mean_token_accuracy": 0.07848219200968742,
"num_tokens": 936059.0,
"step": 480
},
{
"entropy": 8.39522933959961,
"epoch": 0.04848303093917129,
"grad_norm": 0.9296875,
"learning_rate": 0.000242,
"loss": 8.0484,
"mean_token_accuracy": 0.08674658462405205,
"num_tokens": 945990.0,
"step": 485
},
{
"entropy": 8.503808116912841,
"epoch": 0.04898285600039986,
"grad_norm": 1.3046875,
"learning_rate": 0.0002445,
"loss": 8.1173,
"mean_token_accuracy": 0.08412509486079216,
"num_tokens": 956626.0,
"step": 490
},
{
"entropy": 8.41580810546875,
"epoch": 0.04948268106162843,
"grad_norm": 0.96484375,
"learning_rate": 0.000247,
"loss": 8.0777,
"mean_token_accuracy": 0.08279605060815812,
"num_tokens": 966200.0,
"step": 495
},
{
"entropy": 8.39886407852173,
"epoch": 0.049982506122857,
"grad_norm": 0.90234375,
"learning_rate": 0.0002495,
"loss": 8.0361,
"mean_token_accuracy": 0.08273001238703728,
"num_tokens": 975651.0,
"step": 500
},
{
"entropy": 8.383611392974853,
"epoch": 0.05048233118408557,
"grad_norm": 1.0546875,
"learning_rate": 0.000252,
"loss": 8.0191,
"mean_token_accuracy": 0.08777477145195008,
"num_tokens": 985264.0,
"step": 505
},
{
"entropy": 8.448287868499756,
"epoch": 0.05098215624531414,
"grad_norm": 1.2265625,
"learning_rate": 0.0002545,
"loss": 8.053,
"mean_token_accuracy": 0.08592012897133827,
"num_tokens": 996723.0,
"step": 510
},
{
"entropy": 8.31779079437256,
"epoch": 0.05148198130654271,
"grad_norm": 0.91796875,
"learning_rate": 0.000257,
"loss": 8.0489,
"mean_token_accuracy": 0.08642567917704583,
"num_tokens": 1006098.0,
"step": 515
},
{
"entropy": 8.480357837677001,
"epoch": 0.05198180636777128,
"grad_norm": 0.9453125,
"learning_rate": 0.0002595,
"loss": 8.0707,
"mean_token_accuracy": 0.08443118557333947,
"num_tokens": 1016595.0,
"step": 520
},
{
"entropy": 8.306590175628662,
"epoch": 0.05248163142899985,
"grad_norm": 0.984375,
"learning_rate": 0.000262,
"loss": 7.9342,
"mean_token_accuracy": 0.09165540859103202,
"num_tokens": 1026281.0,
"step": 525
},
{
"entropy": 8.408623600006104,
"epoch": 0.052981456490228424,
"grad_norm": 0.82421875,
"learning_rate": 0.00026450000000000003,
"loss": 8.0266,
"mean_token_accuracy": 0.09572977796196938,
"num_tokens": 1036697.0,
"step": 530
},
{
"entropy": 8.276596641540527,
"epoch": 0.05348128155145699,
"grad_norm": 1.09375,
"learning_rate": 0.00026700000000000004,
"loss": 7.9901,
"mean_token_accuracy": 0.08793360516428947,
"num_tokens": 1047433.0,
"step": 535
},
{
"entropy": 8.411270713806152,
"epoch": 0.05398110661268556,
"grad_norm": 0.984375,
"learning_rate": 0.00026950000000000005,
"loss": 8.0427,
"mean_token_accuracy": 0.08355218432843685,
"num_tokens": 1056774.0,
"step": 540
},
{
"entropy": 8.40280523300171,
"epoch": 0.05448093167391413,
"grad_norm": 1.0078125,
"learning_rate": 0.00027200000000000005,
"loss": 7.9525,
"mean_token_accuracy": 0.08867315500974655,
"num_tokens": 1068056.0,
"step": 545
},
{
"entropy": 8.254864311218261,
"epoch": 0.0549807567351427,
"grad_norm": 1.1328125,
"learning_rate": 0.0002745,
"loss": 7.9596,
"mean_token_accuracy": 0.0920575998723507,
"num_tokens": 1077698.0,
"step": 550
},
{
"entropy": 8.321324825286865,
"epoch": 0.05548058179637127,
"grad_norm": 1.25,
"learning_rate": 0.000277,
"loss": 7.9305,
"mean_token_accuracy": 0.08997995480895042,
"num_tokens": 1086485.0,
"step": 555
},
{
"entropy": 8.285623931884766,
"epoch": 0.05598040685759984,
"grad_norm": 0.984375,
"learning_rate": 0.0002795,
"loss": 8.0035,
"mean_token_accuracy": 0.08188552521169186,
"num_tokens": 1097329.0,
"step": 560
},
{
"entropy": 8.301086711883546,
"epoch": 0.05648023191882841,
"grad_norm": 0.87109375,
"learning_rate": 0.00028199999999999997,
"loss": 7.9439,
"mean_token_accuracy": 0.08813558220863342,
"num_tokens": 1107990.0,
"step": 565
},
{
"entropy": 8.220473194122315,
"epoch": 0.05698005698005698,
"grad_norm": 1.15625,
"learning_rate": 0.0002845,
"loss": 7.8945,
"mean_token_accuracy": 0.09917720332741738,
"num_tokens": 1117798.0,
"step": 570
},
{
"entropy": 8.299683856964112,
"epoch": 0.05747988204128555,
"grad_norm": 0.90234375,
"learning_rate": 0.000287,
"loss": 8.0225,
"mean_token_accuracy": 0.08569787368178368,
"num_tokens": 1126976.0,
"step": 575
},
{
"entropy": 8.327748203277588,
"epoch": 0.05797970710251412,
"grad_norm": 1.2265625,
"learning_rate": 0.0002895,
"loss": 7.9334,
"mean_token_accuracy": 0.09346841722726822,
"num_tokens": 1136410.0,
"step": 580
},
{
"entropy": 8.28176794052124,
"epoch": 0.05847953216374269,
"grad_norm": 1.0703125,
"learning_rate": 0.000292,
"loss": 7.987,
"mean_token_accuracy": 0.090962153673172,
"num_tokens": 1147162.0,
"step": 585
},
{
"entropy": 8.228501224517823,
"epoch": 0.05897935722497126,
"grad_norm": 1.046875,
"learning_rate": 0.0002945,
"loss": 7.9276,
"mean_token_accuracy": 0.09324892237782478,
"num_tokens": 1157578.0,
"step": 590
},
{
"entropy": 8.30532102584839,
"epoch": 0.05947918228619983,
"grad_norm": 0.98828125,
"learning_rate": 0.000297,
"loss": 7.8498,
"mean_token_accuracy": 0.09714705944061279,
"num_tokens": 1166914.0,
"step": 595
},
{
"entropy": 8.3526291847229,
"epoch": 0.0599790073474284,
"grad_norm": 1.1015625,
"learning_rate": 0.0002995,
"loss": 8.0223,
"mean_token_accuracy": 0.07643549218773842,
"num_tokens": 1176740.0,
"step": 600
},
{
"entropy": 8.228054428100586,
"epoch": 0.06047883240865697,
"grad_norm": 1.0703125,
"learning_rate": 0.000302,
"loss": 7.9041,
"mean_token_accuracy": 0.08873437419533729,
"num_tokens": 1185497.0,
"step": 605
},
{
"entropy": 8.404792785644531,
"epoch": 0.06097865746988554,
"grad_norm": 1.1484375,
"learning_rate": 0.0003045,
"loss": 8.0288,
"mean_token_accuracy": 0.08556954115629196,
"num_tokens": 1197390.0,
"step": 610
},
{
"entropy": 8.208370685577393,
"epoch": 0.06147848253111411,
"grad_norm": 0.828125,
"learning_rate": 0.000307,
"loss": 7.8988,
"mean_token_accuracy": 0.09034750014543533,
"num_tokens": 1208293.0,
"step": 615
},
{
"entropy": 8.260112953186034,
"epoch": 0.06197830759234268,
"grad_norm": 1.0,
"learning_rate": 0.0003095,
"loss": 7.8837,
"mean_token_accuracy": 0.09412865191698075,
"num_tokens": 1217860.0,
"step": 620
},
{
"entropy": 8.192390155792236,
"epoch": 0.06247813265357125,
"grad_norm": 0.9765625,
"learning_rate": 0.000312,
"loss": 7.8716,
"mean_token_accuracy": 0.09198905006051064,
"num_tokens": 1227259.0,
"step": 625
},
{
"entropy": 8.16130018234253,
"epoch": 0.06297795771479982,
"grad_norm": 1.1640625,
"learning_rate": 0.0003145,
"loss": 7.7885,
"mean_token_accuracy": 0.10033405348658561,
"num_tokens": 1236775.0,
"step": 630
},
{
"entropy": 8.210112524032592,
"epoch": 0.06347778277602839,
"grad_norm": 1.109375,
"learning_rate": 0.000317,
"loss": 7.9116,
"mean_token_accuracy": 0.09359579831361771,
"num_tokens": 1246307.0,
"step": 635
},
{
"entropy": 8.269742012023926,
"epoch": 0.06397760783725696,
"grad_norm": 1.046875,
"learning_rate": 0.0003195,
"loss": 7.8657,
"mean_token_accuracy": 0.09205928593873977,
"num_tokens": 1255197.0,
"step": 640
},
{
"entropy": 8.189111995697022,
"epoch": 0.06447743289848552,
"grad_norm": 0.98828125,
"learning_rate": 0.000322,
"loss": 7.8667,
"mean_token_accuracy": 0.09405043423175811,
"num_tokens": 1265056.0,
"step": 645
},
{
"entropy": 8.203232288360596,
"epoch": 0.0649772579597141,
"grad_norm": 1.015625,
"learning_rate": 0.00032450000000000003,
"loss": 7.8283,
"mean_token_accuracy": 0.09532282426953316,
"num_tokens": 1274718.0,
"step": 650
},
{
"entropy": 8.204540061950684,
"epoch": 0.06547708302094267,
"grad_norm": 1.1484375,
"learning_rate": 0.00032700000000000003,
"loss": 7.9137,
"mean_token_accuracy": 0.09175522401928901,
"num_tokens": 1284364.0,
"step": 655
},
{
"entropy": 8.217914533615112,
"epoch": 0.06597690808217124,
"grad_norm": 1.0859375,
"learning_rate": 0.00032950000000000004,
"loss": 7.8891,
"mean_token_accuracy": 0.09070554673671723,
"num_tokens": 1294186.0,
"step": 660
},
{
"entropy": 8.16135516166687,
"epoch": 0.06647673314339982,
"grad_norm": 1.0859375,
"learning_rate": 0.00033200000000000005,
"loss": 7.8175,
"mean_token_accuracy": 0.09384845942258835,
"num_tokens": 1303393.0,
"step": 665
},
{
"entropy": 8.106479072570801,
"epoch": 0.06697655820462838,
"grad_norm": 1.3125,
"learning_rate": 0.00033450000000000005,
"loss": 7.817,
"mean_token_accuracy": 0.09376823343336582,
"num_tokens": 1312819.0,
"step": 670
},
{
"entropy": 8.17269515991211,
"epoch": 0.06747638326585695,
"grad_norm": 0.96875,
"learning_rate": 0.000337,
"loss": 7.851,
"mean_token_accuracy": 0.08638178259134292,
"num_tokens": 1323001.0,
"step": 675
},
{
"entropy": 8.182546710968017,
"epoch": 0.06797620832708552,
"grad_norm": 0.9375,
"learning_rate": 0.0003395,
"loss": 7.8232,
"mean_token_accuracy": 0.09359516352415084,
"num_tokens": 1333260.0,
"step": 680
},
{
"entropy": 8.1503267288208,
"epoch": 0.0684760333883141,
"grad_norm": 1.0390625,
"learning_rate": 0.000342,
"loss": 7.8104,
"mean_token_accuracy": 0.09704278111457824,
"num_tokens": 1342696.0,
"step": 685
},
{
"entropy": 8.134940958023071,
"epoch": 0.06897585844954265,
"grad_norm": 1.0,
"learning_rate": 0.00034449999999999997,
"loss": 7.7834,
"mean_token_accuracy": 0.09516805186867713,
"num_tokens": 1353104.0,
"step": 690
},
{
"entropy": 8.105061912536621,
"epoch": 0.06947568351077123,
"grad_norm": 1.0078125,
"learning_rate": 0.000347,
"loss": 7.8471,
"mean_token_accuracy": 0.08959203399717808,
"num_tokens": 1362871.0,
"step": 695
},
{
"entropy": 8.226502609252929,
"epoch": 0.0699755085719998,
"grad_norm": 0.91015625,
"learning_rate": 0.0003495,
"loss": 7.8302,
"mean_token_accuracy": 0.09120392128825187,
"num_tokens": 1373622.0,
"step": 700
},
{
"entropy": 7.967414331436157,
"epoch": 0.07047533363322837,
"grad_norm": 1.1171875,
"learning_rate": 0.000352,
"loss": 7.7096,
"mean_token_accuracy": 0.09465244933962821,
"num_tokens": 1382707.0,
"step": 705
},
{
"entropy": 8.168174076080323,
"epoch": 0.07097515869445695,
"grad_norm": 1.1875,
"learning_rate": 0.0003545,
"loss": 7.794,
"mean_token_accuracy": 0.09188380241394042,
"num_tokens": 1392048.0,
"step": 710
},
{
"entropy": 8.040040349960327,
"epoch": 0.0714749837556855,
"grad_norm": 1.0625,
"learning_rate": 0.000357,
"loss": 7.6498,
"mean_token_accuracy": 0.10075373873114586,
"num_tokens": 1401387.0,
"step": 715
},
{
"entropy": 8.03524203300476,
"epoch": 0.07197480881691408,
"grad_norm": 0.86328125,
"learning_rate": 0.0003595,
"loss": 7.7718,
"mean_token_accuracy": 0.09415309280157089,
"num_tokens": 1411404.0,
"step": 720
},
{
"entropy": 8.118272829055787,
"epoch": 0.07247463387814265,
"grad_norm": 0.90234375,
"learning_rate": 0.000362,
"loss": 7.7716,
"mean_token_accuracy": 0.09523647800087928,
"num_tokens": 1420292.0,
"step": 725
},
{
"entropy": 8.011260223388671,
"epoch": 0.07297445893937123,
"grad_norm": 0.9375,
"learning_rate": 0.0003645,
"loss": 7.7221,
"mean_token_accuracy": 0.09846484884619713,
"num_tokens": 1429990.0,
"step": 730
},
{
"entropy": 8.110036754608155,
"epoch": 0.07347428400059978,
"grad_norm": 1.078125,
"learning_rate": 0.000367,
"loss": 7.7661,
"mean_token_accuracy": 0.0931557647883892,
"num_tokens": 1440148.0,
"step": 735
},
{
"entropy": 8.04368724822998,
"epoch": 0.07397410906182836,
"grad_norm": 1.078125,
"learning_rate": 0.0003695,
"loss": 7.6823,
"mean_token_accuracy": 0.09766408279538155,
"num_tokens": 1450015.0,
"step": 740
},
{
"entropy": 8.074540615081787,
"epoch": 0.07447393412305693,
"grad_norm": 1.0078125,
"learning_rate": 0.000372,
"loss": 7.7128,
"mean_token_accuracy": 0.09001053497195244,
"num_tokens": 1459582.0,
"step": 745
},
{
"entropy": 8.047771739959718,
"epoch": 0.0749737591842855,
"grad_norm": 1.0390625,
"learning_rate": 0.0003745,
"loss": 7.7654,
"mean_token_accuracy": 0.09219523742794991,
"num_tokens": 1468151.0,
"step": 750
},
{
"entropy": 7.934108066558838,
"epoch": 0.07547358424551406,
"grad_norm": 0.9765625,
"learning_rate": 0.000377,
"loss": 7.6629,
"mean_token_accuracy": 0.09758676961064339,
"num_tokens": 1479099.0,
"step": 755
},
{
"entropy": 8.08583436012268,
"epoch": 0.07597340930674264,
"grad_norm": 1.0546875,
"learning_rate": 0.0003795,
"loss": 7.7606,
"mean_token_accuracy": 0.09898279458284379,
"num_tokens": 1488060.0,
"step": 760
},
{
"entropy": 8.037435007095336,
"epoch": 0.07647323436797121,
"grad_norm": 0.99609375,
"learning_rate": 0.000382,
"loss": 7.7476,
"mean_token_accuracy": 0.09002138823270797,
"num_tokens": 1498230.0,
"step": 765
},
{
"entropy": 8.06327486038208,
"epoch": 0.07697305942919978,
"grad_norm": 1.015625,
"learning_rate": 0.0003845,
"loss": 7.7279,
"mean_token_accuracy": 0.09608541578054428,
"num_tokens": 1507962.0,
"step": 770
},
{
"entropy": 8.107155275344848,
"epoch": 0.07747288449042836,
"grad_norm": 1.0546875,
"learning_rate": 0.00038700000000000003,
"loss": 7.8234,
"mean_token_accuracy": 0.09302043840289116,
"num_tokens": 1518607.0,
"step": 775
},
{
"entropy": 8.132760810852051,
"epoch": 0.07797270955165692,
"grad_norm": 1.1015625,
"learning_rate": 0.00038950000000000003,
"loss": 7.6654,
"mean_token_accuracy": 0.09478092044591904,
"num_tokens": 1527166.0,
"step": 780
},
{
"entropy": 7.994145393371582,
"epoch": 0.07847253461288549,
"grad_norm": 1.0625,
"learning_rate": 0.00039200000000000004,
"loss": 7.7769,
"mean_token_accuracy": 0.09216032102704048,
"num_tokens": 1536683.0,
"step": 785
},
{
"entropy": 8.052890396118164,
"epoch": 0.07897235967411406,
"grad_norm": 0.921875,
"learning_rate": 0.00039450000000000005,
"loss": 7.6444,
"mean_token_accuracy": 0.09586679339408874,
"num_tokens": 1546243.0,
"step": 790
},
{
"entropy": 7.990080165863037,
"epoch": 0.07947218473534264,
"grad_norm": 1.1171875,
"learning_rate": 0.00039700000000000005,
"loss": 7.6838,
"mean_token_accuracy": 0.09463807120919228,
"num_tokens": 1555916.0,
"step": 795
},
{
"entropy": 7.892803049087524,
"epoch": 0.0799720097965712,
"grad_norm": 0.98828125,
"learning_rate": 0.0003995,
"loss": 7.5913,
"mean_token_accuracy": 0.10165544375777244,
"num_tokens": 1566617.0,
"step": 800
},
{
"entropy": 7.9309196949005125,
"epoch": 0.08047183485779977,
"grad_norm": 1.09375,
"learning_rate": 0.000402,
"loss": 7.6573,
"mean_token_accuracy": 0.09189758896827697,
"num_tokens": 1575413.0,
"step": 805
},
{
"entropy": 7.9370897769927975,
"epoch": 0.08097165991902834,
"grad_norm": 1.2734375,
"learning_rate": 0.0004045,
"loss": 7.6576,
"mean_token_accuracy": 0.09608973935246468,
"num_tokens": 1586863.0,
"step": 810
},
{
"entropy": 8.001241874694824,
"epoch": 0.08147148498025691,
"grad_norm": 0.9296875,
"learning_rate": 0.00040699999999999997,
"loss": 7.7234,
"mean_token_accuracy": 0.09731132909655571,
"num_tokens": 1597685.0,
"step": 815
},
{
"entropy": 7.99445104598999,
"epoch": 0.08197131004148547,
"grad_norm": 1.0859375,
"learning_rate": 0.0004095,
"loss": 7.6245,
"mean_token_accuracy": 0.09776123687624931,
"num_tokens": 1607103.0,
"step": 820
},
{
"entropy": 7.999743795394897,
"epoch": 0.08247113510271405,
"grad_norm": 1.0078125,
"learning_rate": 0.000412,
"loss": 7.6631,
"mean_token_accuracy": 0.0952106162905693,
"num_tokens": 1617919.0,
"step": 825
},
{
"entropy": 7.934334754943848,
"epoch": 0.08297096016394262,
"grad_norm": 0.8828125,
"learning_rate": 0.0004145,
"loss": 7.7062,
"mean_token_accuracy": 0.09701430797576904,
"num_tokens": 1627293.0,
"step": 830
},
{
"entropy": 7.999687910079956,
"epoch": 0.08347078522517119,
"grad_norm": 0.94921875,
"learning_rate": 0.000417,
"loss": 7.7125,
"mean_token_accuracy": 0.09134984835982322,
"num_tokens": 1638440.0,
"step": 835
},
{
"entropy": 7.880404663085938,
"epoch": 0.08397061028639977,
"grad_norm": 0.9375,
"learning_rate": 0.0004195,
"loss": 7.6324,
"mean_token_accuracy": 0.09711692780256272,
"num_tokens": 1648148.0,
"step": 840
},
{
"entropy": 7.96956696510315,
"epoch": 0.08447043534762833,
"grad_norm": 1.109375,
"learning_rate": 0.000422,
"loss": 7.6519,
"mean_token_accuracy": 0.10145560279488564,
"num_tokens": 1658334.0,
"step": 845
},
{
"entropy": 8.041061925888062,
"epoch": 0.0849702604088569,
"grad_norm": 1.2421875,
"learning_rate": 0.0004245,
"loss": 7.6965,
"mean_token_accuracy": 0.09095216616988182,
"num_tokens": 1667963.0,
"step": 850
},
{
"entropy": 7.87789101600647,
"epoch": 0.08547008547008547,
"grad_norm": 1.0234375,
"learning_rate": 0.000427,
"loss": 7.5978,
"mean_token_accuracy": 0.09792420715093612,
"num_tokens": 1677270.0,
"step": 855
},
{
"entropy": 7.916245174407959,
"epoch": 0.08596991053131404,
"grad_norm": 0.98828125,
"learning_rate": 0.0004295,
"loss": 7.5792,
"mean_token_accuracy": 0.09994457513093949,
"num_tokens": 1686770.0,
"step": 860
},
{
"entropy": 7.882506227493286,
"epoch": 0.0864697355925426,
"grad_norm": 0.9296875,
"learning_rate": 0.000432,
"loss": 7.6509,
"mean_token_accuracy": 0.0953475922346115,
"num_tokens": 1697507.0,
"step": 865
},
{
"entropy": 7.81713080406189,
"epoch": 0.08696956065377118,
"grad_norm": 0.94921875,
"learning_rate": 0.0004345,
"loss": 7.5814,
"mean_token_accuracy": 0.10015851929783821,
"num_tokens": 1708253.0,
"step": 870
},
{
"entropy": 7.997309446334839,
"epoch": 0.08746938571499975,
"grad_norm": 0.98828125,
"learning_rate": 0.000437,
"loss": 7.6658,
"mean_token_accuracy": 0.09474888741970063,
"num_tokens": 1717827.0,
"step": 875
},
{
"entropy": 7.760024833679199,
"epoch": 0.08796921077622832,
"grad_norm": 1.03125,
"learning_rate": 0.0004395,
"loss": 7.4538,
"mean_token_accuracy": 0.10792813673615456,
"num_tokens": 1726678.0,
"step": 880
},
{
"entropy": 7.866535377502442,
"epoch": 0.0884690358374569,
"grad_norm": 0.8671875,
"learning_rate": 0.000442,
"loss": 7.5961,
"mean_token_accuracy": 0.09604379087686539,
"num_tokens": 1737518.0,
"step": 885
},
{
"entropy": 7.879868602752685,
"epoch": 0.08896886089868546,
"grad_norm": 1.1171875,
"learning_rate": 0.0004445,
"loss": 7.4935,
"mean_token_accuracy": 0.10566615983843804,
"num_tokens": 1746338.0,
"step": 890
},
{
"entropy": 7.823741006851196,
"epoch": 0.08946868595991403,
"grad_norm": 1.0,
"learning_rate": 0.000447,
"loss": 7.5613,
"mean_token_accuracy": 0.10079468935728073,
"num_tokens": 1756538.0,
"step": 895
},
{
"entropy": 7.876118326187134,
"epoch": 0.0899685110211426,
"grad_norm": 0.89453125,
"learning_rate": 0.00044950000000000003,
"loss": 7.5836,
"mean_token_accuracy": 0.09676425233483314,
"num_tokens": 1766722.0,
"step": 900
},
{
"entropy": 7.753104639053345,
"epoch": 0.09046833608237118,
"grad_norm": 0.9609375,
"learning_rate": 0.00045200000000000004,
"loss": 7.4967,
"mean_token_accuracy": 0.10072999745607376,
"num_tokens": 1777039.0,
"step": 905
},
{
"entropy": 7.91089015007019,
"epoch": 0.09096816114359974,
"grad_norm": 0.875,
"learning_rate": 0.00045450000000000004,
"loss": 7.6107,
"mean_token_accuracy": 0.0950322225689888,
"num_tokens": 1787621.0,
"step": 910
},
{
"entropy": 7.87430214881897,
"epoch": 0.09146798620482831,
"grad_norm": 1.203125,
"learning_rate": 0.00045700000000000005,
"loss": 7.6136,
"mean_token_accuracy": 0.09905622452497483,
"num_tokens": 1798345.0,
"step": 915
},
{
"entropy": 7.831703472137451,
"epoch": 0.09196781126605688,
"grad_norm": 0.9765625,
"learning_rate": 0.00045950000000000006,
"loss": 7.56,
"mean_token_accuracy": 0.1018152303993702,
"num_tokens": 1807994.0,
"step": 920
},
{
"entropy": 7.771154356002808,
"epoch": 0.09246763632728545,
"grad_norm": 1.0859375,
"learning_rate": 0.000462,
"loss": 7.5007,
"mean_token_accuracy": 0.0945871725678444,
"num_tokens": 1817965.0,
"step": 925
},
{
"entropy": 7.880890274047852,
"epoch": 0.09296746138851401,
"grad_norm": 1.078125,
"learning_rate": 0.0004645,
"loss": 7.5537,
"mean_token_accuracy": 0.09793060868978501,
"num_tokens": 1827094.0,
"step": 930
},
{
"entropy": 7.78585376739502,
"epoch": 0.09346728644974259,
"grad_norm": 1.046875,
"learning_rate": 0.000467,
"loss": 7.5184,
"mean_token_accuracy": 0.09652820006012916,
"num_tokens": 1837754.0,
"step": 935
},
{
"entropy": 7.771854496002197,
"epoch": 0.09396711151097116,
"grad_norm": 1.03125,
"learning_rate": 0.0004695,
"loss": 7.5075,
"mean_token_accuracy": 0.10058806762099266,
"num_tokens": 1847500.0,
"step": 940
},
{
"entropy": 7.791559839248658,
"epoch": 0.09446693657219973,
"grad_norm": 0.98828125,
"learning_rate": 0.000472,
"loss": 7.5732,
"mean_token_accuracy": 0.09383994415402412,
"num_tokens": 1857663.0,
"step": 945
},
{
"entropy": 7.871378183364868,
"epoch": 0.0949667616334283,
"grad_norm": 1.0078125,
"learning_rate": 0.0004745,
"loss": 7.6231,
"mean_token_accuracy": 0.09650576040148735,
"num_tokens": 1867752.0,
"step": 950
},
{
"entropy": 7.807382106781006,
"epoch": 0.09546658669465687,
"grad_norm": 1.0390625,
"learning_rate": 0.000477,
"loss": 7.5452,
"mean_token_accuracy": 0.09623196721076965,
"num_tokens": 1877479.0,
"step": 955
},
{
"entropy": 7.763889932632447,
"epoch": 0.09596641175588544,
"grad_norm": 0.98046875,
"learning_rate": 0.0004795,
"loss": 7.5723,
"mean_token_accuracy": 0.09741483479738236,
"num_tokens": 1887944.0,
"step": 960
},
{
"entropy": 7.879271364212036,
"epoch": 0.09646623681711401,
"grad_norm": 1.046875,
"learning_rate": 0.000482,
"loss": 7.4957,
"mean_token_accuracy": 0.10076584815979003,
"num_tokens": 1896991.0,
"step": 965
},
{
"entropy": 7.705447959899902,
"epoch": 0.09696606187834259,
"grad_norm": 1.578125,
"learning_rate": 0.0004845,
"loss": 7.4952,
"mean_token_accuracy": 0.10603907033801079,
"num_tokens": 1907216.0,
"step": 970
},
{
"entropy": 7.8304407596588135,
"epoch": 0.09746588693957114,
"grad_norm": 1.1328125,
"learning_rate": 0.000487,
"loss": 7.5332,
"mean_token_accuracy": 0.09332873225212097,
"num_tokens": 1917206.0,
"step": 975
},
{
"entropy": 7.699482154846192,
"epoch": 0.09796571200079972,
"grad_norm": 0.90234375,
"learning_rate": 0.0004895,
"loss": 7.5962,
"mean_token_accuracy": 0.09308115392923355,
"num_tokens": 1929669.0,
"step": 980
},
{
"entropy": 7.849009084701538,
"epoch": 0.09846553706202829,
"grad_norm": 1.0234375,
"learning_rate": 0.000492,
"loss": 7.4585,
"mean_token_accuracy": 0.10018454045057297,
"num_tokens": 1939881.0,
"step": 985
},
{
"entropy": 7.612346506118774,
"epoch": 0.09896536212325686,
"grad_norm": 1.1796875,
"learning_rate": 0.0004945,
"loss": 7.4929,
"mean_token_accuracy": 0.09991356357932091,
"num_tokens": 1949423.0,
"step": 990
},
{
"entropy": 7.8288792133331295,
"epoch": 0.09946518718448542,
"grad_norm": 0.96875,
"learning_rate": 0.000497,
"loss": 7.5251,
"mean_token_accuracy": 0.09757097363471985,
"num_tokens": 1959214.0,
"step": 995
},
{
"entropy": 7.802850389480591,
"epoch": 0.099965012245714,
"grad_norm": 1.015625,
"learning_rate": 0.0004995,
"loss": 7.4851,
"mean_token_accuracy": 0.0973809115588665,
"num_tokens": 1968963.0,
"step": 1000
},
{
"entropy": 7.674502229690551,
"epoch": 0.10046483730694257,
"grad_norm": 1.0,
"learning_rate": 0.0004999999981884985,
"loss": 7.4751,
"mean_token_accuracy": 0.09699616283178329,
"num_tokens": 1979551.0,
"step": 1005
},
{
"entropy": 7.707011651992798,
"epoch": 0.10096466236817114,
"grad_norm": 1.234375,
"learning_rate": 0.0004999999908292739,
"loss": 7.4402,
"mean_token_accuracy": 0.10390993803739548,
"num_tokens": 1988432.0,
"step": 1010
},
{
"entropy": 7.738900566101075,
"epoch": 0.10146448742939972,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999999778091072,
"loss": 7.4858,
"mean_token_accuracy": 0.1000777617096901,
"num_tokens": 1997509.0,
"step": 1015
},
{
"entropy": 7.792381763458252,
"epoch": 0.10196431249062828,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999999591279991,
"loss": 7.6484,
"mean_token_accuracy": 0.09373814240098,
"num_tokens": 2007406.0,
"step": 1020
},
{
"entropy": 7.805833625793457,
"epoch": 0.10246413755185685,
"grad_norm": 1.09375,
"learning_rate": 0.00049999993478595,
"loss": 7.4674,
"mean_token_accuracy": 0.09718732684850692,
"num_tokens": 2016760.0,
"step": 1025
},
{
"entropy": 7.635655355453491,
"epoch": 0.10296396261308542,
"grad_norm": 1.0,
"learning_rate": 0.0004999999047829604,
"loss": 7.4949,
"mean_token_accuracy": 0.10022416040301323,
"num_tokens": 2027776.0,
"step": 1030
},
{
"entropy": 7.69910683631897,
"epoch": 0.103463787674314,
"grad_norm": 1.0625,
"learning_rate": 0.000499999869119031,
"loss": 7.4395,
"mean_token_accuracy": 0.10272038653492928,
"num_tokens": 2036955.0,
"step": 1035
},
{
"entropy": 7.6816685676574705,
"epoch": 0.10396361273554255,
"grad_norm": 1.078125,
"learning_rate": 0.0004999998277941629,
"loss": 7.3558,
"mean_token_accuracy": 0.10612019002437592,
"num_tokens": 2046118.0,
"step": 1040
},
{
"entropy": 7.632660675048828,
"epoch": 0.10446343779677113,
"grad_norm": 1.046875,
"learning_rate": 0.0004999997808083569,
"loss": 7.4229,
"mean_token_accuracy": 0.09605257287621498,
"num_tokens": 2056266.0,
"step": 1045
},
{
"entropy": 7.6546745777130125,
"epoch": 0.1049632628579997,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999997281616144,
"loss": 7.4632,
"mean_token_accuracy": 0.10460180416703224,
"num_tokens": 2067416.0,
"step": 1050
},
{
"entropy": 7.745044374465943,
"epoch": 0.10546308791922827,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999996698539366,
"loss": 7.4574,
"mean_token_accuracy": 0.09959002882242203,
"num_tokens": 2076926.0,
"step": 1055
},
{
"entropy": 7.669534826278687,
"epoch": 0.10596291298045685,
"grad_norm": 1.0078125,
"learning_rate": 0.000499999605885325,
"loss": 7.4478,
"mean_token_accuracy": 0.10303856655955315,
"num_tokens": 2088017.0,
"step": 1060
},
{
"entropy": 7.680190181732177,
"epoch": 0.1064627380416854,
"grad_norm": 1.109375,
"learning_rate": 0.0004999995362557813,
"loss": 7.4762,
"mean_token_accuracy": 0.10051115751266479,
"num_tokens": 2097249.0,
"step": 1065
},
{
"entropy": 7.5991401195526125,
"epoch": 0.10696256310291398,
"grad_norm": 1.1484375,
"learning_rate": 0.000499999460965307,
"loss": 7.4297,
"mean_token_accuracy": 0.10126611217856407,
"num_tokens": 2105778.0,
"step": 1070
},
{
"entropy": 7.731232929229736,
"epoch": 0.10746238816414255,
"grad_norm": 1.0,
"learning_rate": 0.0004999993800139043,
"loss": 7.4517,
"mean_token_accuracy": 0.10296982377767563,
"num_tokens": 2114710.0,
"step": 1075
},
{
"entropy": 7.639475536346436,
"epoch": 0.10796221322537113,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999992934015752,
"loss": 7.3907,
"mean_token_accuracy": 0.10413632541894913,
"num_tokens": 2124002.0,
"step": 1080
},
{
"entropy": 7.7447285652160645,
"epoch": 0.10846203828659969,
"grad_norm": 1.140625,
"learning_rate": 0.0004999992011283215,
"loss": 7.4876,
"mean_token_accuracy": 0.09875040203332901,
"num_tokens": 2134592.0,
"step": 1085
},
{
"entropy": 7.503586864471435,
"epoch": 0.10896186334782826,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999991031941459,
"loss": 7.3316,
"mean_token_accuracy": 0.1076856717467308,
"num_tokens": 2143641.0,
"step": 1090
},
{
"entropy": 7.718939399719238,
"epoch": 0.10946168840905683,
"grad_norm": 1.140625,
"learning_rate": 0.0004999989995990509,
"loss": 7.4484,
"mean_token_accuracy": 0.10091273114085197,
"num_tokens": 2154738.0,
"step": 1095
},
{
"entropy": 7.511824655532837,
"epoch": 0.1099615134702854,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999988903430389,
"loss": 7.3804,
"mean_token_accuracy": 0.11026987135410309,
"num_tokens": 2164365.0,
"step": 1100
},
{
"entropy": 7.610298585891724,
"epoch": 0.11046133853151396,
"grad_norm": 1.0625,
"learning_rate": 0.0004999987754261126,
"loss": 7.2694,
"mean_token_accuracy": 0.1102310411632061,
"num_tokens": 2173185.0,
"step": 1105
},
{
"entropy": 7.579056692123413,
"epoch": 0.11096116359274254,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999986548482751,
"loss": 7.382,
"mean_token_accuracy": 0.10696444734930992,
"num_tokens": 2184717.0,
"step": 1110
},
{
"entropy": 7.6235607147216795,
"epoch": 0.11146098865397111,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999985286095293,
"loss": 7.3611,
"mean_token_accuracy": 0.10928932800889016,
"num_tokens": 2193681.0,
"step": 1115
},
{
"entropy": 7.571579217910767,
"epoch": 0.11196081371519968,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999983967098785,
"loss": 7.3752,
"mean_token_accuracy": 0.10050557106733322,
"num_tokens": 2203366.0,
"step": 1120
},
{
"entropy": 7.621008968353271,
"epoch": 0.11246063877642826,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999982591493258,
"loss": 7.3709,
"mean_token_accuracy": 0.10311305895447731,
"num_tokens": 2213982.0,
"step": 1125
},
{
"entropy": 7.596260356903076,
"epoch": 0.11296046383765682,
"grad_norm": 0.953125,
"learning_rate": 0.0004999981159278749,
"loss": 7.4162,
"mean_token_accuracy": 0.09981422498822212,
"num_tokens": 2222897.0,
"step": 1130
},
{
"entropy": 7.580492305755615,
"epoch": 0.11346028889888539,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999979670455292,
"loss": 7.3682,
"mean_token_accuracy": 0.10767317786812783,
"num_tokens": 2232472.0,
"step": 1135
},
{
"entropy": 7.618267774581909,
"epoch": 0.11396011396011396,
"grad_norm": 1.625,
"learning_rate": 0.0004999978125022926,
"loss": 7.408,
"mean_token_accuracy": 0.10514756217598915,
"num_tokens": 2242666.0,
"step": 1140
},
{
"entropy": 7.6292320728302006,
"epoch": 0.11445993902134254,
"grad_norm": 0.9453125,
"learning_rate": 0.000499997652298169,
"loss": 7.4861,
"mean_token_accuracy": 0.09657154306769371,
"num_tokens": 2254586.0,
"step": 1145
},
{
"entropy": 7.628611993789673,
"epoch": 0.1149597640825711,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999974864331624,
"loss": 7.3073,
"mean_token_accuracy": 0.11019685044884682,
"num_tokens": 2264619.0,
"step": 1150
},
{
"entropy": 7.67947096824646,
"epoch": 0.11545958914379967,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999973149072768,
"loss": 7.3913,
"mean_token_accuracy": 0.09589539244771003,
"num_tokens": 2274114.0,
"step": 1155
},
{
"entropy": 7.469546747207642,
"epoch": 0.11595941420502824,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999971377205167,
"loss": 7.303,
"mean_token_accuracy": 0.10781149342656135,
"num_tokens": 2283887.0,
"step": 1160
},
{
"entropy": 7.548113679885864,
"epoch": 0.11645923926625681,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999969548728864,
"loss": 7.3248,
"mean_token_accuracy": 0.10801458954811097,
"num_tokens": 2293090.0,
"step": 1165
},
{
"entropy": 7.564089012145996,
"epoch": 0.11695906432748537,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999967663643908,
"loss": 7.4209,
"mean_token_accuracy": 0.10221518129110337,
"num_tokens": 2302828.0,
"step": 1170
},
{
"entropy": 7.659123516082763,
"epoch": 0.11745888938871395,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999965721950345,
"loss": 7.3104,
"mean_token_accuracy": 0.10870119705796241,
"num_tokens": 2311415.0,
"step": 1175
},
{
"entropy": 7.541514921188354,
"epoch": 0.11795871444994252,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999963723648222,
"loss": 7.3946,
"mean_token_accuracy": 0.10008503422141075,
"num_tokens": 2321512.0,
"step": 1180
},
{
"entropy": 7.579466390609741,
"epoch": 0.1184585395111711,
"grad_norm": 0.98046875,
"learning_rate": 0.000499996166873759,
"loss": 7.2918,
"mean_token_accuracy": 0.1035538524389267,
"num_tokens": 2331398.0,
"step": 1185
},
{
"entropy": 7.571394538879394,
"epoch": 0.11895836457239967,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999959557218502,
"loss": 7.2865,
"mean_token_accuracy": 0.11054573953151703,
"num_tokens": 2341751.0,
"step": 1190
},
{
"entropy": 7.55456509590149,
"epoch": 0.11945818963362823,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999957389091012,
"loss": 7.3187,
"mean_token_accuracy": 0.10006105899810791,
"num_tokens": 2352538.0,
"step": 1195
},
{
"entropy": 7.590258026123047,
"epoch": 0.1199580146948568,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999955164355171,
"loss": 7.3704,
"mean_token_accuracy": 0.09887034147977829,
"num_tokens": 2362433.0,
"step": 1200
},
{
"entropy": 7.561069488525391,
"epoch": 0.12045783975608537,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999952883011038,
"loss": 7.3513,
"mean_token_accuracy": 0.10028525441884995,
"num_tokens": 2372431.0,
"step": 1205
},
{
"entropy": 7.55907883644104,
"epoch": 0.12095766481731395,
"grad_norm": 0.92578125,
"learning_rate": 0.000499995054505867,
"loss": 7.361,
"mean_token_accuracy": 0.10952726975083352,
"num_tokens": 2382483.0,
"step": 1210
},
{
"entropy": 7.617351865768432,
"epoch": 0.1214574898785425,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999948150498124,
"loss": 7.33,
"mean_token_accuracy": 0.0975058175623417,
"num_tokens": 2392835.0,
"step": 1215
},
{
"entropy": 7.545529508590699,
"epoch": 0.12195731493977108,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999945699329462,
"loss": 7.3021,
"mean_token_accuracy": 0.1018272578716278,
"num_tokens": 2402305.0,
"step": 1220
},
{
"entropy": 7.388929176330566,
"epoch": 0.12245714000099965,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999943191552745,
"loss": 7.2427,
"mean_token_accuracy": 0.10890847146511078,
"num_tokens": 2412093.0,
"step": 1225
},
{
"entropy": 7.547239828109741,
"epoch": 0.12295696506222822,
"grad_norm": 1.015625,
"learning_rate": 0.0004999940627168037,
"loss": 7.2174,
"mean_token_accuracy": 0.11171707957983017,
"num_tokens": 2421349.0,
"step": 1230
},
{
"entropy": 7.439934778213501,
"epoch": 0.12345679012345678,
"grad_norm": 1.046875,
"learning_rate": 0.00049999380061754,
"loss": 7.2001,
"mean_token_accuracy": 0.12013261988759041,
"num_tokens": 2430672.0,
"step": 1235
},
{
"entropy": 7.468397808074951,
"epoch": 0.12395661518468536,
"grad_norm": 1.078125,
"learning_rate": 0.0004999935328574903,
"loss": 7.2823,
"mean_token_accuracy": 0.10177658125758171,
"num_tokens": 2439355.0,
"step": 1240
},
{
"entropy": 7.4700652122497555,
"epoch": 0.12445644024591393,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999932594366613,
"loss": 7.2817,
"mean_token_accuracy": 0.11226417720317841,
"num_tokens": 2448860.0,
"step": 1245
},
{
"entropy": 7.478579568862915,
"epoch": 0.1249562653071425,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999929803550597,
"loss": 7.3139,
"mean_token_accuracy": 0.10246847122907639,
"num_tokens": 2458276.0,
"step": 1250
},
{
"entropy": 7.4616804122924805,
"epoch": 0.12545609036837108,
"grad_norm": 1.015625,
"learning_rate": 0.0004999926956126927,
"loss": 7.2455,
"mean_token_accuracy": 0.10908352881669998,
"num_tokens": 2467791.0,
"step": 1255
},
{
"entropy": 7.4985151290893555,
"epoch": 0.12595591542959964,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999924052095672,
"loss": 7.3538,
"mean_token_accuracy": 0.10554418563842774,
"num_tokens": 2477549.0,
"step": 1260
},
{
"entropy": 7.4841855525970455,
"epoch": 0.12645574049082822,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999921091456907,
"loss": 7.267,
"mean_token_accuracy": 0.10368448942899704,
"num_tokens": 2487549.0,
"step": 1265
},
{
"entropy": 7.513351202011108,
"epoch": 0.12695556555205678,
"grad_norm": 1.046875,
"learning_rate": 0.0004999918074210707,
"loss": 7.2834,
"mean_token_accuracy": 0.10522937551140785,
"num_tokens": 2496799.0,
"step": 1270
},
{
"entropy": 7.574540710449218,
"epoch": 0.12745539061328534,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999915000357147,
"loss": 7.3662,
"mean_token_accuracy": 0.10323404893279076,
"num_tokens": 2505795.0,
"step": 1275
},
{
"entropy": 7.454508924484253,
"epoch": 0.12795521567451393,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999911869896305,
"loss": 7.1844,
"mean_token_accuracy": 0.10558920055627823,
"num_tokens": 2515289.0,
"step": 1280
},
{
"entropy": 7.503220081329346,
"epoch": 0.1284550407357425,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999908682828258,
"loss": 7.3608,
"mean_token_accuracy": 0.0966832160949707,
"num_tokens": 2524563.0,
"step": 1285
},
{
"entropy": 7.508555841445923,
"epoch": 0.12895486579697105,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999905439153089,
"loss": 7.2596,
"mean_token_accuracy": 0.10827950909733772,
"num_tokens": 2534057.0,
"step": 1290
},
{
"entropy": 7.381141519546508,
"epoch": 0.12945469085819963,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999902138870878,
"loss": 7.236,
"mean_token_accuracy": 0.11027004197239876,
"num_tokens": 2544646.0,
"step": 1295
},
{
"entropy": 7.5540282249450685,
"epoch": 0.1299545159194282,
"grad_norm": 0.953125,
"learning_rate": 0.0004999898781981707,
"loss": 7.3068,
"mean_token_accuracy": 0.10171382352709771,
"num_tokens": 2553843.0,
"step": 1300
},
{
"entropy": 7.427649259567261,
"epoch": 0.13045434098065678,
"grad_norm": 0.953125,
"learning_rate": 0.0004999895368485662,
"loss": 7.3316,
"mean_token_accuracy": 0.10106317475438117,
"num_tokens": 2564061.0,
"step": 1305
},
{
"entropy": 7.458715009689331,
"epoch": 0.13095416604188534,
"grad_norm": 0.87890625,
"learning_rate": 0.000499989189838283,
"loss": 7.2052,
"mean_token_accuracy": 0.10916719138622284,
"num_tokens": 2574758.0,
"step": 1310
},
{
"entropy": 7.455377197265625,
"epoch": 0.1314539911031139,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999888371673295,
"loss": 7.2648,
"mean_token_accuracy": 0.10652579218149186,
"num_tokens": 2585213.0,
"step": 1315
},
{
"entropy": 7.495338344573975,
"epoch": 0.13195381616434249,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999884788357147,
"loss": 7.3251,
"mean_token_accuracy": 0.1046051949262619,
"num_tokens": 2595511.0,
"step": 1320
},
{
"entropy": 7.45457911491394,
"epoch": 0.13245364122557104,
"grad_norm": 1.046875,
"learning_rate": 0.0004999881148434478,
"loss": 7.2535,
"mean_token_accuracy": 0.1045147307217121,
"num_tokens": 2605930.0,
"step": 1325
},
{
"entropy": 7.457170677185059,
"epoch": 0.13295346628679963,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999877451905378,
"loss": 7.1207,
"mean_token_accuracy": 0.11842303052544594,
"num_tokens": 2614224.0,
"step": 1330
},
{
"entropy": 7.472532939910889,
"epoch": 0.1334532913480282,
"grad_norm": 0.96875,
"learning_rate": 0.0004999873698769941,
"loss": 7.3418,
"mean_token_accuracy": 0.09856024906039237,
"num_tokens": 2624879.0,
"step": 1335
},
{
"entropy": 7.5039904594421385,
"epoch": 0.13395311640925675,
"grad_norm": 0.93359375,
"learning_rate": 0.000499986988902826,
"loss": 7.2125,
"mean_token_accuracy": 0.10768636465072631,
"num_tokens": 2633836.0,
"step": 1340
},
{
"entropy": 7.338539552688599,
"epoch": 0.13445294147048534,
"grad_norm": 0.921875,
"learning_rate": 0.0004999866022680431,
"loss": 7.2666,
"mean_token_accuracy": 0.1048534668982029,
"num_tokens": 2643754.0,
"step": 1345
},
{
"entropy": 7.622525882720947,
"epoch": 0.1349527665317139,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999862099726552,
"loss": 7.3281,
"mean_token_accuracy": 0.1025710292160511,
"num_tokens": 2652708.0,
"step": 1350
},
{
"entropy": 7.392505550384522,
"epoch": 0.13545259159294248,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999858120166721,
"loss": 7.2885,
"mean_token_accuracy": 0.10265006273984909,
"num_tokens": 2662353.0,
"step": 1355
},
{
"entropy": 7.424199056625366,
"epoch": 0.13595241665417104,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999854084001039,
"loss": 7.19,
"mean_token_accuracy": 0.10960527732968331,
"num_tokens": 2672376.0,
"step": 1360
},
{
"entropy": 7.4624227523803714,
"epoch": 0.1364522417153996,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999849991229608,
"loss": 7.2762,
"mean_token_accuracy": 0.10267233923077583,
"num_tokens": 2682115.0,
"step": 1365
},
{
"entropy": 7.424701595306397,
"epoch": 0.1369520667766282,
"grad_norm": 0.91015625,
"learning_rate": 0.000499984584185253,
"loss": 7.2615,
"mean_token_accuracy": 0.1053126648068428,
"num_tokens": 2692429.0,
"step": 1370
},
{
"entropy": 7.444271993637085,
"epoch": 0.13745189183785675,
"grad_norm": 0.96875,
"learning_rate": 0.0004999841635869909,
"loss": 7.2144,
"mean_token_accuracy": 0.10922285541892052,
"num_tokens": 2701373.0,
"step": 1375
},
{
"entropy": 7.525673151016235,
"epoch": 0.1379517168990853,
"grad_norm": 1.21875,
"learning_rate": 0.0004999837373281852,
"loss": 7.3898,
"mean_token_accuracy": 0.10230721682310104,
"num_tokens": 2710717.0,
"step": 1380
},
{
"entropy": 7.4701214790344235,
"epoch": 0.1384515419603139,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999833054088465,
"loss": 7.2443,
"mean_token_accuracy": 0.10736606866121293,
"num_tokens": 2720936.0,
"step": 1385
},
{
"entropy": 7.437333965301514,
"epoch": 0.13895136702154245,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999828678289858,
"loss": 7.1823,
"mean_token_accuracy": 0.11095159128308296,
"num_tokens": 2730504.0,
"step": 1390
},
{
"entropy": 7.421475410461426,
"epoch": 0.13945119208277104,
"grad_norm": 1.0703125,
"learning_rate": 0.000499982424588614,
"loss": 7.2116,
"mean_token_accuracy": 0.10913697853684426,
"num_tokens": 2739069.0,
"step": 1395
},
{
"entropy": 7.433441162109375,
"epoch": 0.1399510171439996,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999819756877422,
"loss": 7.2829,
"mean_token_accuracy": 0.10340380370616913,
"num_tokens": 2748941.0,
"step": 1400
},
{
"entropy": 7.4657440185546875,
"epoch": 0.14045084220522816,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999815211263819,
"loss": 7.2414,
"mean_token_accuracy": 0.10870002508163452,
"num_tokens": 2758915.0,
"step": 1405
},
{
"entropy": 7.471285152435303,
"epoch": 0.14095066726645675,
"grad_norm": 0.984375,
"learning_rate": 0.0004999810609045444,
"loss": 7.2686,
"mean_token_accuracy": 0.10555689707398415,
"num_tokens": 2768972.0,
"step": 1410
},
{
"entropy": 7.414257478713989,
"epoch": 0.1414504923276853,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999805950222412,
"loss": 7.1266,
"mean_token_accuracy": 0.112129794806242,
"num_tokens": 2777989.0,
"step": 1415
},
{
"entropy": 7.3253881454467775,
"epoch": 0.1419503173889139,
"grad_norm": 1.015625,
"learning_rate": 0.0004999801234794843,
"loss": 7.2124,
"mean_token_accuracy": 0.10934107601642609,
"num_tokens": 2788227.0,
"step": 1420
},
{
"entropy": 7.480621862411499,
"epoch": 0.14245014245014245,
"grad_norm": 1.015625,
"learning_rate": 0.0004999796462762853,
"loss": 7.212,
"mean_token_accuracy": 0.1068321980535984,
"num_tokens": 2797729.0,
"step": 1425
},
{
"entropy": 7.441812515258789,
"epoch": 0.142949967511371,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999791634126562,
"loss": 7.1993,
"mean_token_accuracy": 0.10914001166820526,
"num_tokens": 2807215.0,
"step": 1430
},
{
"entropy": 7.380958318710327,
"epoch": 0.1434497925725996,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999786748886094,
"loss": 7.1273,
"mean_token_accuracy": 0.10741899609565735,
"num_tokens": 2816945.0,
"step": 1435
},
{
"entropy": 7.335004377365112,
"epoch": 0.14394961763382816,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999781807041569,
"loss": 7.2064,
"mean_token_accuracy": 0.10659912303090095,
"num_tokens": 2827072.0,
"step": 1440
},
{
"entropy": 7.436053657531739,
"epoch": 0.14444944269505672,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999776808593113,
"loss": 7.1381,
"mean_token_accuracy": 0.10905283689498901,
"num_tokens": 2837928.0,
"step": 1445
},
{
"entropy": 7.265142488479614,
"epoch": 0.1449492677562853,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999771753540852,
"loss": 7.179,
"mean_token_accuracy": 0.10916463062167167,
"num_tokens": 2847861.0,
"step": 1450
},
{
"entropy": 7.440875148773193,
"epoch": 0.14544909281751386,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999766641884912,
"loss": 7.1819,
"mean_token_accuracy": 0.1099862240254879,
"num_tokens": 2857885.0,
"step": 1455
},
{
"entropy": 7.47276120185852,
"epoch": 0.14594891787874245,
"grad_norm": 0.953125,
"learning_rate": 0.0004999761473625422,
"loss": 7.1931,
"mean_token_accuracy": 0.10840637236833572,
"num_tokens": 2867004.0,
"step": 1460
},
{
"entropy": 7.34009747505188,
"epoch": 0.146448742939971,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999756248762513,
"loss": 7.0759,
"mean_token_accuracy": 0.11003444865345954,
"num_tokens": 2876285.0,
"step": 1465
},
{
"entropy": 7.343789863586426,
"epoch": 0.14694856800119957,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999750967296315,
"loss": 7.1943,
"mean_token_accuracy": 0.10783570408821105,
"num_tokens": 2886967.0,
"step": 1470
},
{
"entropy": 7.352244138717651,
"epoch": 0.14744839306242816,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999745629226962,
"loss": 7.1707,
"mean_token_accuracy": 0.10942457839846612,
"num_tokens": 2897137.0,
"step": 1475
},
{
"entropy": 7.343924140930175,
"epoch": 0.14794821812365672,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999740234554589,
"loss": 7.1515,
"mean_token_accuracy": 0.11578499227762222,
"num_tokens": 2907658.0,
"step": 1480
},
{
"entropy": 7.364530467987061,
"epoch": 0.1484480431848853,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999734783279327,
"loss": 7.1334,
"mean_token_accuracy": 0.11196301355957985,
"num_tokens": 2917801.0,
"step": 1485
},
{
"entropy": 7.394926309585571,
"epoch": 0.14894786824611386,
"grad_norm": 1.015625,
"learning_rate": 0.000499972927540132,
"loss": 7.237,
"mean_token_accuracy": 0.10743141695857047,
"num_tokens": 2927188.0,
"step": 1490
},
{
"entropy": 7.377376556396484,
"epoch": 0.14944769330734242,
"grad_norm": 0.953125,
"learning_rate": 0.0004999723710920703,
"loss": 7.1152,
"mean_token_accuracy": 0.10680729150772095,
"num_tokens": 2936781.0,
"step": 1495
},
{
"entropy": 7.2929119110107425,
"epoch": 0.149947518368571,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999718089837616,
"loss": 7.0999,
"mean_token_accuracy": 0.10934963151812553,
"num_tokens": 2946426.0,
"step": 1500
},
{
"entropy": 7.274065399169922,
"epoch": 0.15044734342979957,
"grad_norm": 0.9453125,
"learning_rate": 0.00049997124121522,
"loss": 7.1414,
"mean_token_accuracy": 0.10659637004137039,
"num_tokens": 2956246.0,
"step": 1505
},
{
"entropy": 7.381241321563721,
"epoch": 0.15094716849102813,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999706677864599,
"loss": 7.1364,
"mean_token_accuracy": 0.10887710601091385,
"num_tokens": 2965930.0,
"step": 1510
},
{
"entropy": 7.290860176086426,
"epoch": 0.15144699355225671,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999700886974958,
"loss": 7.0529,
"mean_token_accuracy": 0.11699229776859284,
"num_tokens": 2975975.0,
"step": 1515
},
{
"entropy": 7.316656589508057,
"epoch": 0.15194681861348527,
"grad_norm": 0.92578125,
"learning_rate": 0.000499969503948342,
"loss": 7.082,
"mean_token_accuracy": 0.1087965801358223,
"num_tokens": 2985343.0,
"step": 1520
},
{
"entropy": 7.285784769058227,
"epoch": 0.15244664367471386,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999689135390134,
"loss": 7.1083,
"mean_token_accuracy": 0.11093569695949554,
"num_tokens": 2994420.0,
"step": 1525
},
{
"entropy": 7.318411874771118,
"epoch": 0.15294646873594242,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999683174695249,
"loss": 7.2103,
"mean_token_accuracy": 0.10868243128061295,
"num_tokens": 3004447.0,
"step": 1530
},
{
"entropy": 7.479772996902466,
"epoch": 0.15344629379717098,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999677157398914,
"loss": 7.1,
"mean_token_accuracy": 0.11058899015188217,
"num_tokens": 3014596.0,
"step": 1535
},
{
"entropy": 7.251273250579834,
"epoch": 0.15394611885839957,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999671083501281,
"loss": 7.0292,
"mean_token_accuracy": 0.11444317251443863,
"num_tokens": 3024528.0,
"step": 1540
},
{
"entropy": 7.28997917175293,
"epoch": 0.15444594391962813,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999664953002502,
"loss": 7.1557,
"mean_token_accuracy": 0.10664113461971284,
"num_tokens": 3035233.0,
"step": 1545
},
{
"entropy": 7.3472541809082035,
"epoch": 0.1549457689808567,
"grad_norm": 1.046875,
"learning_rate": 0.000499965876590273,
"loss": 7.1567,
"mean_token_accuracy": 0.11339146867394448,
"num_tokens": 3045890.0,
"step": 1550
},
{
"entropy": 7.251162910461426,
"epoch": 0.15544559404208527,
"grad_norm": 0.984375,
"learning_rate": 0.0004999652522202125,
"loss": 7.0648,
"mean_token_accuracy": 0.11411697939038276,
"num_tokens": 3056640.0,
"step": 1555
},
{
"entropy": 7.315458631515503,
"epoch": 0.15594541910331383,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999646221900839,
"loss": 7.0761,
"mean_token_accuracy": 0.11611308604478836,
"num_tokens": 3066938.0,
"step": 1560
},
{
"entropy": 7.28765344619751,
"epoch": 0.15644524416454242,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999639864999034,
"loss": 7.1933,
"mean_token_accuracy": 0.1076837420463562,
"num_tokens": 3076791.0,
"step": 1565
},
{
"entropy": 7.381633186340332,
"epoch": 0.15694506922577098,
"grad_norm": 0.9765625,
"learning_rate": 0.000499963345149687,
"loss": 7.0536,
"mean_token_accuracy": 0.11385932415723801,
"num_tokens": 3087532.0,
"step": 1570
},
{
"entropy": 7.220392751693725,
"epoch": 0.15744489428699954,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999626981394506,
"loss": 7.1733,
"mean_token_accuracy": 0.11016052812337876,
"num_tokens": 3097465.0,
"step": 1575
},
{
"entropy": 7.435248279571534,
"epoch": 0.15794471934822812,
"grad_norm": 1.03125,
"learning_rate": 0.0004999620454692106,
"loss": 7.0846,
"mean_token_accuracy": 0.1115849569439888,
"num_tokens": 3107162.0,
"step": 1580
},
{
"entropy": 7.236566162109375,
"epoch": 0.15844454440945668,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999613871389836,
"loss": 7.0768,
"mean_token_accuracy": 0.11040661185979843,
"num_tokens": 3117154.0,
"step": 1585
},
{
"entropy": 7.266448545455932,
"epoch": 0.15894436947068527,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999607231487858,
"loss": 7.0859,
"mean_token_accuracy": 0.11145442575216294,
"num_tokens": 3127635.0,
"step": 1590
},
{
"entropy": 7.3993692874908445,
"epoch": 0.15944419453191383,
"grad_norm": 0.91796875,
"learning_rate": 0.000499960053498634,
"loss": 7.1231,
"mean_token_accuracy": 0.10866989269852638,
"num_tokens": 3137806.0,
"step": 1595
},
{
"entropy": 7.321648645401001,
"epoch": 0.1599440195931424,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999593781885454,
"loss": 7.1575,
"mean_token_accuracy": 0.10646053105592727,
"num_tokens": 3147704.0,
"step": 1600
},
{
"entropy": 7.28309473991394,
"epoch": 0.16044384465437098,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999586972185366,
"loss": 7.1107,
"mean_token_accuracy": 0.105899029225111,
"num_tokens": 3158486.0,
"step": 1605
},
{
"entropy": 7.324327421188355,
"epoch": 0.16094366971559954,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999580105886248,
"loss": 7.108,
"mean_token_accuracy": 0.10965174809098244,
"num_tokens": 3168518.0,
"step": 1610
},
{
"entropy": 7.243863439559936,
"epoch": 0.16144349477682812,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999573182988275,
"loss": 7.0088,
"mean_token_accuracy": 0.11485414430499077,
"num_tokens": 3177502.0,
"step": 1615
},
{
"entropy": 7.24871244430542,
"epoch": 0.16194331983805668,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999566203491619,
"loss": 7.0729,
"mean_token_accuracy": 0.11296822354197503,
"num_tokens": 3187408.0,
"step": 1620
},
{
"entropy": 7.314590549468994,
"epoch": 0.16244314489928524,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999559167396455,
"loss": 7.1227,
"mean_token_accuracy": 0.11361486688256264,
"num_tokens": 3196812.0,
"step": 1625
},
{
"entropy": 7.246614170074463,
"epoch": 0.16294296996051383,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999552074702963,
"loss": 6.9737,
"mean_token_accuracy": 0.11634291410446167,
"num_tokens": 3206159.0,
"step": 1630
},
{
"entropy": 7.23604302406311,
"epoch": 0.1634427950217424,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999544925411318,
"loss": 6.9716,
"mean_token_accuracy": 0.11750346943736076,
"num_tokens": 3217329.0,
"step": 1635
},
{
"entropy": 7.23541374206543,
"epoch": 0.16394262008297095,
"grad_norm": 0.875,
"learning_rate": 0.0004999537719521703,
"loss": 7.0904,
"mean_token_accuracy": 0.11265335828065873,
"num_tokens": 3227558.0,
"step": 1640
},
{
"entropy": 7.3270646095275875,
"epoch": 0.16444244514419953,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999530457034297,
"loss": 7.1907,
"mean_token_accuracy": 0.10715283155441284,
"num_tokens": 3237719.0,
"step": 1645
},
{
"entropy": 7.3062530040740965,
"epoch": 0.1649422702054281,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999523137949283,
"loss": 7.0868,
"mean_token_accuracy": 0.11413407698273659,
"num_tokens": 3246483.0,
"step": 1650
},
{
"entropy": 7.198418855667114,
"epoch": 0.16544209526665668,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999515762266846,
"loss": 6.9876,
"mean_token_accuracy": 0.11951323971152306,
"num_tokens": 3256955.0,
"step": 1655
},
{
"entropy": 7.281463003158569,
"epoch": 0.16594192032788524,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999508329987173,
"loss": 7.0615,
"mean_token_accuracy": 0.10993959754705429,
"num_tokens": 3266799.0,
"step": 1660
},
{
"entropy": 7.3243176460266115,
"epoch": 0.1664417453891138,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999500841110447,
"loss": 7.0627,
"mean_token_accuracy": 0.11115454137325287,
"num_tokens": 3276595.0,
"step": 1665
},
{
"entropy": 7.170014476776123,
"epoch": 0.16694157045034239,
"grad_norm": 0.9765625,
"learning_rate": 0.000499949329563686,
"loss": 7.0036,
"mean_token_accuracy": 0.11233359575271606,
"num_tokens": 3286047.0,
"step": 1670
},
{
"entropy": 7.2867292881011965,
"epoch": 0.16744139551157095,
"grad_norm": 0.99609375,
"learning_rate": 0.00049994856935666,
"loss": 7.0971,
"mean_token_accuracy": 0.10945143923163414,
"num_tokens": 3295364.0,
"step": 1675
},
{
"entropy": 7.282499170303344,
"epoch": 0.16794122057279953,
"grad_norm": 0.95703125,
"learning_rate": 0.000499947803489986,
"loss": 6.9968,
"mean_token_accuracy": 0.11751805990934372,
"num_tokens": 3305319.0,
"step": 1680
},
{
"entropy": 7.17340202331543,
"epoch": 0.1684410456340281,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999470319636829,
"loss": 7.0171,
"mean_token_accuracy": 0.11624193787574769,
"num_tokens": 3315349.0,
"step": 1685
},
{
"entropy": 7.2853912830352785,
"epoch": 0.16894087069525665,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999462547777705,
"loss": 7.0887,
"mean_token_accuracy": 0.10680258646607399,
"num_tokens": 3325603.0,
"step": 1690
},
{
"entropy": 7.203047037124634,
"epoch": 0.16944069575648524,
"grad_norm": 0.984375,
"learning_rate": 0.0004999454719322682,
"loss": 6.9998,
"mean_token_accuracy": 0.11204721629619599,
"num_tokens": 3334693.0,
"step": 1695
},
{
"entropy": 7.272320556640625,
"epoch": 0.1699405208177138,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999446834271958,
"loss": 7.0064,
"mean_token_accuracy": 0.10611082762479782,
"num_tokens": 3345259.0,
"step": 1700
},
{
"entropy": 7.210876941680908,
"epoch": 0.17044034587894236,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999438892625728,
"loss": 7.0013,
"mean_token_accuracy": 0.11196986362338066,
"num_tokens": 3354072.0,
"step": 1705
},
{
"entropy": 7.286466646194458,
"epoch": 0.17094017094017094,
"grad_norm": 1.015625,
"learning_rate": 0.0004999430894384198,
"loss": 7.0142,
"mean_token_accuracy": 0.1119252622127533,
"num_tokens": 3364881.0,
"step": 1710
},
{
"entropy": 7.226142120361328,
"epoch": 0.1714399960013995,
"grad_norm": 1.0,
"learning_rate": 0.0004999422839547564,
"loss": 6.9904,
"mean_token_accuracy": 0.11430766358971596,
"num_tokens": 3374030.0,
"step": 1715
},
{
"entropy": 7.220478677749634,
"epoch": 0.1719398210626281,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999414728116029,
"loss": 6.9778,
"mean_token_accuracy": 0.11128946840763092,
"num_tokens": 3384009.0,
"step": 1720
},
{
"entropy": 7.301162242889404,
"epoch": 0.17243964612385665,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999406560089798,
"loss": 7.1074,
"mean_token_accuracy": 0.11010550707578659,
"num_tokens": 3393660.0,
"step": 1725
},
{
"entropy": 7.27868685722351,
"epoch": 0.1729394711850852,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999398335469078,
"loss": 7.1187,
"mean_token_accuracy": 0.10912684947252274,
"num_tokens": 3404364.0,
"step": 1730
},
{
"entropy": 7.307126235961914,
"epoch": 0.1734392962463138,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999390054254074,
"loss": 7.0129,
"mean_token_accuracy": 0.10965274721384048,
"num_tokens": 3412990.0,
"step": 1735
},
{
"entropy": 7.103516006469727,
"epoch": 0.17393912130754235,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999381716444996,
"loss": 6.984,
"mean_token_accuracy": 0.11489634066820145,
"num_tokens": 3422339.0,
"step": 1740
},
{
"entropy": 7.192869377136231,
"epoch": 0.17443894636877094,
"grad_norm": 0.84375,
"learning_rate": 0.0004999373322042052,
"loss": 6.9704,
"mean_token_accuracy": 0.1187138944864273,
"num_tokens": 3433093.0,
"step": 1745
},
{
"entropy": 7.266874742507935,
"epoch": 0.1749387714299995,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999364871045455,
"loss": 7.0579,
"mean_token_accuracy": 0.1107969380915165,
"num_tokens": 3443464.0,
"step": 1750
},
{
"entropy": 7.208977365493775,
"epoch": 0.17543859649122806,
"grad_norm": 0.875,
"learning_rate": 0.0004999356363455415,
"loss": 6.9891,
"mean_token_accuracy": 0.11229778155684471,
"num_tokens": 3453380.0,
"step": 1755
},
{
"entropy": 7.2512282371521,
"epoch": 0.17593842155245665,
"grad_norm": 0.93359375,
"learning_rate": 0.000499934779927215,
"loss": 7.0487,
"mean_token_accuracy": 0.11276780366897583,
"num_tokens": 3463583.0,
"step": 1760
},
{
"entropy": 7.108209085464478,
"epoch": 0.1764382466136852,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999339178495872,
"loss": 6.9567,
"mean_token_accuracy": 0.1195647343993187,
"num_tokens": 3473640.0,
"step": 1765
},
{
"entropy": 7.258290004730225,
"epoch": 0.1769380716749138,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999330501126798,
"loss": 7.0087,
"mean_token_accuracy": 0.11589695587754249,
"num_tokens": 3483827.0,
"step": 1770
},
{
"entropy": 7.162285470962525,
"epoch": 0.17743789673614235,
"grad_norm": 0.953125,
"learning_rate": 0.0004999321767165148,
"loss": 6.9711,
"mean_token_accuracy": 0.12137874588370323,
"num_tokens": 3493763.0,
"step": 1775
},
{
"entropy": 7.261633634567261,
"epoch": 0.1779377217973709,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999312976611142,
"loss": 7.0714,
"mean_token_accuracy": 0.11497013196349144,
"num_tokens": 3503119.0,
"step": 1780
},
{
"entropy": 7.239193248748779,
"epoch": 0.1784375468585995,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999304129465,
"loss": 7.0063,
"mean_token_accuracy": 0.11441105306148529,
"num_tokens": 3513235.0,
"step": 1785
},
{
"entropy": 7.172239589691162,
"epoch": 0.17893737191982806,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999295225726945,
"loss": 7.109,
"mean_token_accuracy": 0.11077995523810387,
"num_tokens": 3523633.0,
"step": 1790
},
{
"entropy": 7.217508792877197,
"epoch": 0.17943719698105662,
"grad_norm": 0.94140625,
"learning_rate": 0.00049992862653972,
"loss": 6.9661,
"mean_token_accuracy": 0.11305666714906693,
"num_tokens": 3532888.0,
"step": 1795
},
{
"entropy": 7.171756887435913,
"epoch": 0.1799370220422852,
"grad_norm": 1.09375,
"learning_rate": 0.0004999277248475992,
"loss": 6.9914,
"mean_token_accuracy": 0.11619005724787712,
"num_tokens": 3541942.0,
"step": 1800
},
{
"entropy": 7.162946081161499,
"epoch": 0.18043684710351376,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999268174963547,
"loss": 6.9532,
"mean_token_accuracy": 0.11453788876533508,
"num_tokens": 3551544.0,
"step": 1805
},
{
"entropy": 7.187183475494384,
"epoch": 0.18093667216474235,
"grad_norm": 1.046875,
"learning_rate": 0.0004999259044860093,
"loss": 6.9411,
"mean_token_accuracy": 0.11845703423023224,
"num_tokens": 3561024.0,
"step": 1810
},
{
"entropy": 7.203239154815674,
"epoch": 0.1814364972259709,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999249858165861,
"loss": 6.9324,
"mean_token_accuracy": 0.12066594958305359,
"num_tokens": 3570962.0,
"step": 1815
},
{
"entropy": 7.026967144012451,
"epoch": 0.18193632228719947,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999240614881082,
"loss": 6.9255,
"mean_token_accuracy": 0.12051551789045334,
"num_tokens": 3580669.0,
"step": 1820
},
{
"entropy": 7.2166660785675045,
"epoch": 0.18243614734842806,
"grad_norm": 0.9375,
"learning_rate": 0.0004999231315005987,
"loss": 6.9524,
"mean_token_accuracy": 0.11456257104873657,
"num_tokens": 3590627.0,
"step": 1825
},
{
"entropy": 7.190853023529053,
"epoch": 0.18293597240965662,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999221958540811,
"loss": 7.015,
"mean_token_accuracy": 0.10618584454059601,
"num_tokens": 3599872.0,
"step": 1830
},
{
"entropy": 7.208268022537231,
"epoch": 0.1834357974708852,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999212545485789,
"loss": 7.0008,
"mean_token_accuracy": 0.11748863831162452,
"num_tokens": 3609809.0,
"step": 1835
},
{
"entropy": 7.2950187683105465,
"epoch": 0.18393562253211376,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999203075841159,
"loss": 7.0256,
"mean_token_accuracy": 0.11501933261752129,
"num_tokens": 3619385.0,
"step": 1840
},
{
"entropy": 7.049505376815796,
"epoch": 0.18443544759334232,
"grad_norm": 0.859375,
"learning_rate": 0.0004999193549607157,
"loss": 6.8527,
"mean_token_accuracy": 0.12542221918702126,
"num_tokens": 3628867.0,
"step": 1845
},
{
"entropy": 7.11543025970459,
"epoch": 0.1849352726545709,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999183966784026,
"loss": 6.8588,
"mean_token_accuracy": 0.12358809560537339,
"num_tokens": 3637528.0,
"step": 1850
},
{
"entropy": 7.202391195297241,
"epoch": 0.18543509771579947,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999174327372004,
"loss": 7.0118,
"mean_token_accuracy": 0.11167162656784058,
"num_tokens": 3648011.0,
"step": 1855
},
{
"entropy": 7.162098455429077,
"epoch": 0.18593492277702803,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999164631371335,
"loss": 6.9861,
"mean_token_accuracy": 0.11887979209423065,
"num_tokens": 3657648.0,
"step": 1860
},
{
"entropy": 7.202825736999512,
"epoch": 0.18643474783825661,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999154878782262,
"loss": 6.9828,
"mean_token_accuracy": 0.11495407447218894,
"num_tokens": 3666965.0,
"step": 1865
},
{
"entropy": 7.200334119796753,
"epoch": 0.18693457289948517,
"grad_norm": 1.015625,
"learning_rate": 0.000499914506960503,
"loss": 6.9956,
"mean_token_accuracy": 0.11006476879119872,
"num_tokens": 3676609.0,
"step": 1870
},
{
"entropy": 7.093168592453003,
"epoch": 0.18743439796071376,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999135203839889,
"loss": 6.8839,
"mean_token_accuracy": 0.11935680136084556,
"num_tokens": 3685884.0,
"step": 1875
},
{
"entropy": 7.126411867141724,
"epoch": 0.18793422302194232,
"grad_norm": 0.984375,
"learning_rate": 0.0004999125281487084,
"loss": 7.0068,
"mean_token_accuracy": 0.1182105928659439,
"num_tokens": 3695436.0,
"step": 1880
},
{
"entropy": 7.144247961044312,
"epoch": 0.18843404808317088,
"grad_norm": 0.96875,
"learning_rate": 0.0004999115302546866,
"loss": 6.8441,
"mean_token_accuracy": 0.12458746284246444,
"num_tokens": 3704806.0,
"step": 1885
},
{
"entropy": 7.070190811157227,
"epoch": 0.18893387314439947,
"grad_norm": 1.5390625,
"learning_rate": 0.0004999105267019486,
"loss": 6.927,
"mean_token_accuracy": 0.11586586683988571,
"num_tokens": 3713927.0,
"step": 1890
},
{
"entropy": 7.240120220184326,
"epoch": 0.18943369820562803,
"grad_norm": 1.171875,
"learning_rate": 0.0004999095174905195,
"loss": 6.9918,
"mean_token_accuracy": 0.11079314202070237,
"num_tokens": 3723685.0,
"step": 1895
},
{
"entropy": 7.096197319030762,
"epoch": 0.1899335232668566,
"grad_norm": 0.984375,
"learning_rate": 0.0004999085026204249,
"loss": 6.949,
"mean_token_accuracy": 0.11694203093647956,
"num_tokens": 3733918.0,
"step": 1900
},
{
"entropy": 7.182708120346069,
"epoch": 0.19043334832808517,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999074820916903,
"loss": 6.9316,
"mean_token_accuracy": 0.11336019709706306,
"num_tokens": 3744466.0,
"step": 1905
},
{
"entropy": 7.1592817306518555,
"epoch": 0.19093317338931373,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999064559043412,
"loss": 6.8797,
"mean_token_accuracy": 0.11740054339170455,
"num_tokens": 3754082.0,
"step": 1910
},
{
"entropy": 7.081003665924072,
"epoch": 0.19143299845054232,
"grad_norm": 1.046875,
"learning_rate": 0.0004999054240584037,
"loss": 6.9088,
"mean_token_accuracy": 0.1203138992190361,
"num_tokens": 3763111.0,
"step": 1915
},
{
"entropy": 7.212647819519043,
"epoch": 0.19193282351177088,
"grad_norm": 1.0625,
"learning_rate": 0.0004999043865539035,
"loss": 6.9635,
"mean_token_accuracy": 0.10895317792892456,
"num_tokens": 3772555.0,
"step": 1920
},
{
"entropy": 7.189334392547607,
"epoch": 0.19243264857299944,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999033433908667,
"loss": 6.998,
"mean_token_accuracy": 0.11409115567803382,
"num_tokens": 3783073.0,
"step": 1925
},
{
"entropy": 7.0710899353027346,
"epoch": 0.19293247363422802,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999022945693198,
"loss": 6.9411,
"mean_token_accuracy": 0.11544915810227394,
"num_tokens": 3792277.0,
"step": 1930
},
{
"entropy": 7.1727738857269285,
"epoch": 0.19343229869545658,
"grad_norm": 0.98828125,
"learning_rate": 0.000499901240089289,
"loss": 6.8969,
"mean_token_accuracy": 0.11691495999693871,
"num_tokens": 3801098.0,
"step": 1935
},
{
"entropy": 7.160139465332032,
"epoch": 0.19393212375668517,
"grad_norm": 0.96875,
"learning_rate": 0.0004999001799508009,
"loss": 6.9196,
"mean_token_accuracy": 0.11668399348855019,
"num_tokens": 3810614.0,
"step": 1940
},
{
"entropy": 7.043404626846313,
"epoch": 0.19443194881791373,
"grad_norm": 0.984375,
"learning_rate": 0.0004998991141538821,
"loss": 6.9305,
"mean_token_accuracy": 0.11645101308822632,
"num_tokens": 3818989.0,
"step": 1945
},
{
"entropy": 7.1233776092529295,
"epoch": 0.1949317738791423,
"grad_norm": 0.953125,
"learning_rate": 0.0004998980426985596,
"loss": 6.9665,
"mean_token_accuracy": 0.11670495495200157,
"num_tokens": 3829319.0,
"step": 1950
},
{
"entropy": 7.134263134002685,
"epoch": 0.19543159894037088,
"grad_norm": 0.87109375,
"learning_rate": 0.00049989696558486,
"loss": 6.9779,
"mean_token_accuracy": 0.11338206753134727,
"num_tokens": 3839545.0,
"step": 1955
},
{
"entropy": 7.1283763408660885,
"epoch": 0.19593142400159944,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998958828128108,
"loss": 6.9192,
"mean_token_accuracy": 0.11778575927019119,
"num_tokens": 3849029.0,
"step": 1960
},
{
"entropy": 7.100479316711426,
"epoch": 0.19643124906282802,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998947943824389,
"loss": 6.9572,
"mean_token_accuracy": 0.11571668013930321,
"num_tokens": 3859463.0,
"step": 1965
},
{
"entropy": 7.251784420013427,
"epoch": 0.19693107412405658,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998937002937719,
"loss": 6.9985,
"mean_token_accuracy": 0.11572480946779251,
"num_tokens": 3869695.0,
"step": 1970
},
{
"entropy": 7.1218890190124515,
"epoch": 0.19743089918528514,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998926005468373,
"loss": 6.9564,
"mean_token_accuracy": 0.11750205978751183,
"num_tokens": 3879608.0,
"step": 1975
},
{
"entropy": 7.101331901550293,
"epoch": 0.19793072424651373,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998914951416627,
"loss": 6.8697,
"mean_token_accuracy": 0.11973301321268082,
"num_tokens": 3888709.0,
"step": 1980
},
{
"entropy": 7.0645873069763185,
"epoch": 0.1984305493077423,
"grad_norm": 0.921875,
"learning_rate": 0.000499890384078276,
"loss": 6.8144,
"mean_token_accuracy": 0.11889860406517982,
"num_tokens": 3898392.0,
"step": 1985
},
{
"entropy": 7.083717393875122,
"epoch": 0.19893037436897085,
"grad_norm": 0.921875,
"learning_rate": 0.0004998892673567051,
"loss": 7.0328,
"mean_token_accuracy": 0.10867243111133576,
"num_tokens": 3909508.0,
"step": 1990
},
{
"entropy": 7.095362138748169,
"epoch": 0.19943019943019943,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998881449769781,
"loss": 6.8772,
"mean_token_accuracy": 0.11684270799160004,
"num_tokens": 3918155.0,
"step": 1995
},
{
"entropy": 7.156151819229126,
"epoch": 0.199930024491428,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998870169391232,
"loss": 6.91,
"mean_token_accuracy": 0.11658932641148567,
"num_tokens": 3928478.0,
"step": 2000
},
{
"entropy": 7.094034481048584,
"epoch": 0.20042984955265658,
"grad_norm": 0.95703125,
"learning_rate": 0.000499885883243169,
"loss": 6.9807,
"mean_token_accuracy": 0.12299856469035149,
"num_tokens": 3938126.0,
"step": 2005
},
{
"entropy": 7.0389543056488035,
"epoch": 0.20092967461388514,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998847438891437,
"loss": 6.7931,
"mean_token_accuracy": 0.12135762944817544,
"num_tokens": 3948176.0,
"step": 2010
},
{
"entropy": 7.130215978622436,
"epoch": 0.2014294996751137,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998835988770761,
"loss": 6.9243,
"mean_token_accuracy": 0.11784106865525246,
"num_tokens": 3958983.0,
"step": 2015
},
{
"entropy": 7.148845577239991,
"epoch": 0.20192932473634229,
"grad_norm": 0.94921875,
"learning_rate": 0.000499882448206995,
"loss": 6.8748,
"mean_token_accuracy": 0.12108960896730422,
"num_tokens": 3968766.0,
"step": 2020
},
{
"entropy": 7.093853235244751,
"epoch": 0.20242914979757085,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998812918789293,
"loss": 6.8862,
"mean_token_accuracy": 0.12036861553788185,
"num_tokens": 3979487.0,
"step": 2025
},
{
"entropy": 7.037854337692261,
"epoch": 0.20292897485879943,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998801298929084,
"loss": 6.8591,
"mean_token_accuracy": 0.11948419883847236,
"num_tokens": 3988450.0,
"step": 2030
},
{
"entropy": 7.083759737014771,
"epoch": 0.203428799920028,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998789622489611,
"loss": 6.8639,
"mean_token_accuracy": 0.1209349848330021,
"num_tokens": 3997492.0,
"step": 2035
},
{
"entropy": 7.120600318908691,
"epoch": 0.20392862498125655,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998777889471171,
"loss": 6.8659,
"mean_token_accuracy": 0.12269705757498742,
"num_tokens": 4007298.0,
"step": 2040
},
{
"entropy": 7.064807891845703,
"epoch": 0.20442845004248514,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998766099874056,
"loss": 6.8977,
"mean_token_accuracy": 0.11901947408914566,
"num_tokens": 4017727.0,
"step": 2045
},
{
"entropy": 7.092329645156861,
"epoch": 0.2049282751037137,
"grad_norm": 1.0,
"learning_rate": 0.0004998754253698566,
"loss": 6.9052,
"mean_token_accuracy": 0.1160533107817173,
"num_tokens": 4027618.0,
"step": 2050
},
{
"entropy": 7.058936882019043,
"epoch": 0.20542810016494226,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998742350944998,
"loss": 6.9305,
"mean_token_accuracy": 0.11532141789793968,
"num_tokens": 4038831.0,
"step": 2055
},
{
"entropy": 7.055427837371826,
"epoch": 0.20592792522617084,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998730391613651,
"loss": 6.8681,
"mean_token_accuracy": 0.12060280740261078,
"num_tokens": 4048736.0,
"step": 2060
},
{
"entropy": 7.093530511856079,
"epoch": 0.2064277502873994,
"grad_norm": 1.03125,
"learning_rate": 0.0004998718375704825,
"loss": 6.805,
"mean_token_accuracy": 0.12220611199736595,
"num_tokens": 4057633.0,
"step": 2065
},
{
"entropy": 7.054113101959229,
"epoch": 0.206927575348628,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998706303218825,
"loss": 6.9191,
"mean_token_accuracy": 0.12357110977172851,
"num_tokens": 4067883.0,
"step": 2070
},
{
"entropy": 7.159766292572021,
"epoch": 0.20742740040985655,
"grad_norm": 1.0,
"learning_rate": 0.0004998694174155952,
"loss": 6.7905,
"mean_token_accuracy": 0.1274022191762924,
"num_tokens": 4078147.0,
"step": 2075
},
{
"entropy": 7.028505516052246,
"epoch": 0.2079272254710851,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998681988516512,
"loss": 6.7782,
"mean_token_accuracy": 0.12301289215683937,
"num_tokens": 4087130.0,
"step": 2080
},
{
"entropy": 6.935732984542847,
"epoch": 0.2084270505323137,
"grad_norm": 0.984375,
"learning_rate": 0.0004998669746300811,
"loss": 6.903,
"mean_token_accuracy": 0.124887103587389,
"num_tokens": 4096840.0,
"step": 2085
},
{
"entropy": 7.1919714450836185,
"epoch": 0.20892687559354226,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998657447509159,
"loss": 6.8965,
"mean_token_accuracy": 0.11713318973779678,
"num_tokens": 4105842.0,
"step": 2090
},
{
"entropy": 6.997682905197143,
"epoch": 0.20942670065477084,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998645092141864,
"loss": 6.8033,
"mean_token_accuracy": 0.12010766267776489,
"num_tokens": 4114865.0,
"step": 2095
},
{
"entropy": 7.080593633651733,
"epoch": 0.2099265257159994,
"grad_norm": 1.078125,
"learning_rate": 0.0004998632680199238,
"loss": 6.9373,
"mean_token_accuracy": 0.11356281042098999,
"num_tokens": 4124954.0,
"step": 2100
},
{
"entropy": 7.025081729888916,
"epoch": 0.21042635077722796,
"grad_norm": 0.921875,
"learning_rate": 0.0004998620211681591,
"loss": 6.8203,
"mean_token_accuracy": 0.12838008180260657,
"num_tokens": 4134091.0,
"step": 2105
},
{
"entropy": 7.1411622047424315,
"epoch": 0.21092617583845655,
"grad_norm": 0.953125,
"learning_rate": 0.0004998607686589239,
"loss": 6.8802,
"mean_token_accuracy": 0.11798109486699104,
"num_tokens": 4143566.0,
"step": 2110
},
{
"entropy": 7.026936674118042,
"epoch": 0.2114260008996851,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998595104922496,
"loss": 6.8579,
"mean_token_accuracy": 0.11757076904177666,
"num_tokens": 4153053.0,
"step": 2115
},
{
"entropy": 7.064190101623535,
"epoch": 0.2119258259609137,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998582466681678,
"loss": 6.8287,
"mean_token_accuracy": 0.12464014589786529,
"num_tokens": 4162925.0,
"step": 2120
},
{
"entropy": 6.9577514171600345,
"epoch": 0.21242565102214225,
"grad_norm": 0.9375,
"learning_rate": 0.0004998569771867105,
"loss": 6.759,
"mean_token_accuracy": 0.1262947879731655,
"num_tokens": 4172077.0,
"step": 2125
},
{
"entropy": 6.999064683914185,
"epoch": 0.2129254760833708,
"grad_norm": 0.96875,
"learning_rate": 0.0004998557020479095,
"loss": 6.8494,
"mean_token_accuracy": 0.12305980324745178,
"num_tokens": 4182134.0,
"step": 2130
},
{
"entropy": 7.176851749420166,
"epoch": 0.2134253011445994,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998544212517968,
"loss": 6.8664,
"mean_token_accuracy": 0.11641905829310417,
"num_tokens": 4191963.0,
"step": 2135
},
{
"entropy": 6.967975282669068,
"epoch": 0.21392512620582796,
"grad_norm": 0.9375,
"learning_rate": 0.0004998531347984047,
"loss": 6.8121,
"mean_token_accuracy": 0.12236538529396057,
"num_tokens": 4202180.0,
"step": 2140
},
{
"entropy": 7.056608772277832,
"epoch": 0.21442495126705652,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998518426877656,
"loss": 6.8498,
"mean_token_accuracy": 0.12232566028833389,
"num_tokens": 4211874.0,
"step": 2145
},
{
"entropy": 7.057928800582886,
"epoch": 0.2149247763282851,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998505449199122,
"loss": 6.7936,
"mean_token_accuracy": 0.1208051361143589,
"num_tokens": 4221610.0,
"step": 2150
},
{
"entropy": 6.943663740158081,
"epoch": 0.21542460138951366,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998492414948768,
"loss": 6.748,
"mean_token_accuracy": 0.133094572275877,
"num_tokens": 4230610.0,
"step": 2155
},
{
"entropy": 6.993412733078003,
"epoch": 0.21592442645074225,
"grad_norm": 0.78515625,
"learning_rate": 0.0004998479324126924,
"loss": 6.8782,
"mean_token_accuracy": 0.11309932991862297,
"num_tokens": 4241389.0,
"step": 2160
},
{
"entropy": 7.034138774871826,
"epoch": 0.2164242515119708,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998466176733918,
"loss": 6.8262,
"mean_token_accuracy": 0.12450739443302154,
"num_tokens": 4250617.0,
"step": 2165
},
{
"entropy": 6.97466025352478,
"epoch": 0.21692407657319937,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998452972770082,
"loss": 6.7973,
"mean_token_accuracy": 0.12435769960284233,
"num_tokens": 4260000.0,
"step": 2170
},
{
"entropy": 6.991814851760864,
"epoch": 0.21742390163442796,
"grad_norm": 0.984375,
"learning_rate": 0.0004998439712235747,
"loss": 6.7444,
"mean_token_accuracy": 0.12331395819783211,
"num_tokens": 4269751.0,
"step": 2175
},
{
"entropy": 7.0533545970916744,
"epoch": 0.21792372669565652,
"grad_norm": 1.015625,
"learning_rate": 0.0004998426395131248,
"loss": 6.7905,
"mean_token_accuracy": 0.12443327084183693,
"num_tokens": 4279210.0,
"step": 2180
},
{
"entropy": 6.989436435699463,
"epoch": 0.2184235517568851,
"grad_norm": 0.984375,
"learning_rate": 0.0004998413021456919,
"loss": 6.7657,
"mean_token_accuracy": 0.12329083904623986,
"num_tokens": 4288573.0,
"step": 2185
},
{
"entropy": 7.008385276794433,
"epoch": 0.21892337681811366,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998399591213098,
"loss": 6.7964,
"mean_token_accuracy": 0.12139227017760276,
"num_tokens": 4298438.0,
"step": 2190
},
{
"entropy": 6.953334188461303,
"epoch": 0.21942320187934222,
"grad_norm": 0.9375,
"learning_rate": 0.0004998386104400122,
"loss": 6.7268,
"mean_token_accuracy": 0.13097667396068574,
"num_tokens": 4307174.0,
"step": 2195
},
{
"entropy": 7.032305383682251,
"epoch": 0.2199230269405708,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998372561018329,
"loss": 6.8946,
"mean_token_accuracy": 0.11677326709032058,
"num_tokens": 4317220.0,
"step": 2200
},
{
"entropy": 7.0936168193817135,
"epoch": 0.22042285200179937,
"grad_norm": 1.0,
"learning_rate": 0.0004998358961068062,
"loss": 6.7989,
"mean_token_accuracy": 0.12612595781683922,
"num_tokens": 4326496.0,
"step": 2205
},
{
"entropy": 7.025656223297119,
"epoch": 0.22092267706302793,
"grad_norm": 0.984375,
"learning_rate": 0.0004998345304549661,
"loss": 6.8081,
"mean_token_accuracy": 0.12592438012361526,
"num_tokens": 4335705.0,
"step": 2210
},
{
"entropy": 7.022608041763306,
"epoch": 0.22142250212425652,
"grad_norm": 0.984375,
"learning_rate": 0.0004998331591463471,
"loss": 6.8211,
"mean_token_accuracy": 0.12835633084177972,
"num_tokens": 4344711.0,
"step": 2215
},
{
"entropy": 7.006923866271973,
"epoch": 0.22192232718548507,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998317821809837,
"loss": 6.8391,
"mean_token_accuracy": 0.1216700553894043,
"num_tokens": 4355448.0,
"step": 2220
},
{
"entropy": 6.954414653778076,
"epoch": 0.22242215224671366,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998303995589104,
"loss": 6.7257,
"mean_token_accuracy": 0.12261889353394509,
"num_tokens": 4365104.0,
"step": 2225
},
{
"entropy": 6.914372348785401,
"epoch": 0.22292197730794222,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998290112801623,
"loss": 6.6823,
"mean_token_accuracy": 0.1286713495850563,
"num_tokens": 4375219.0,
"step": 2230
},
{
"entropy": 7.024904727935791,
"epoch": 0.22342180236917078,
"grad_norm": 1.0234375,
"learning_rate": 0.000499827617344774,
"loss": 6.8458,
"mean_token_accuracy": 0.11705941632390023,
"num_tokens": 4385294.0,
"step": 2235
},
{
"entropy": 7.090032148361206,
"epoch": 0.22392162743039937,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998262177527807,
"loss": 6.7875,
"mean_token_accuracy": 0.11647440493106842,
"num_tokens": 4394305.0,
"step": 2240
},
{
"entropy": 6.9880218505859375,
"epoch": 0.22442145249162793,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998248125042177,
"loss": 6.6909,
"mean_token_accuracy": 0.12566715776920317,
"num_tokens": 4403732.0,
"step": 2245
},
{
"entropy": 6.976096105575562,
"epoch": 0.2249212775528565,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998234015991202,
"loss": 6.8776,
"mean_token_accuracy": 0.11910765767097473,
"num_tokens": 4413730.0,
"step": 2250
},
{
"entropy": 7.104625558853149,
"epoch": 0.22542110261408507,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998219850375238,
"loss": 6.8875,
"mean_token_accuracy": 0.11642134636640548,
"num_tokens": 4423051.0,
"step": 2255
},
{
"entropy": 7.09229040145874,
"epoch": 0.22592092767531363,
"grad_norm": 0.82421875,
"learning_rate": 0.0004998205628194641,
"loss": 6.8914,
"mean_token_accuracy": 0.12018313780426979,
"num_tokens": 4433363.0,
"step": 2260
},
{
"entropy": 7.084233140945434,
"epoch": 0.22642075273654222,
"grad_norm": 0.83984375,
"learning_rate": 0.0004998191349449768,
"loss": 6.8453,
"mean_token_accuracy": 0.11719286441802979,
"num_tokens": 4442984.0,
"step": 2265
},
{
"entropy": 6.962313604354859,
"epoch": 0.22692057779777078,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998177014140981,
"loss": 6.7755,
"mean_token_accuracy": 0.11933308243751525,
"num_tokens": 4452364.0,
"step": 2270
},
{
"entropy": 7.018482780456543,
"epoch": 0.22742040285899934,
"grad_norm": 0.984375,
"learning_rate": 0.0004998162622268637,
"loss": 6.8136,
"mean_token_accuracy": 0.11589978188276291,
"num_tokens": 4461904.0,
"step": 2275
},
{
"entropy": 7.027294683456421,
"epoch": 0.22792022792022792,
"grad_norm": 1.015625,
"learning_rate": 0.0004998148173833101,
"loss": 6.7632,
"mean_token_accuracy": 0.12299803197383881,
"num_tokens": 4470633.0,
"step": 2280
},
{
"entropy": 6.9562835693359375,
"epoch": 0.22842005298145648,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998133668834735,
"loss": 6.7806,
"mean_token_accuracy": 0.12278707697987556,
"num_tokens": 4480673.0,
"step": 2285
},
{
"entropy": 7.003961133956909,
"epoch": 0.22891987804268507,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998119107273905,
"loss": 6.7636,
"mean_token_accuracy": 0.11756833791732788,
"num_tokens": 4489923.0,
"step": 2290
},
{
"entropy": 6.959622764587403,
"epoch": 0.22941970310391363,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998104489150975,
"loss": 6.7422,
"mean_token_accuracy": 0.12293882742524147,
"num_tokens": 4501007.0,
"step": 2295
},
{
"entropy": 6.924298858642578,
"epoch": 0.2299195281651422,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998089814466318,
"loss": 6.757,
"mean_token_accuracy": 0.12394008114933967,
"num_tokens": 4510460.0,
"step": 2300
},
{
"entropy": 7.008149099349976,
"epoch": 0.23041935322637078,
"grad_norm": 0.875,
"learning_rate": 0.0004998075083220295,
"loss": 6.8235,
"mean_token_accuracy": 0.12430360317230224,
"num_tokens": 4521026.0,
"step": 2305
},
{
"entropy": 6.980177974700927,
"epoch": 0.23091917828759934,
"grad_norm": 0.953125,
"learning_rate": 0.0004998060295413286,
"loss": 6.7636,
"mean_token_accuracy": 0.12551465407013893,
"num_tokens": 4530818.0,
"step": 2310
},
{
"entropy": 6.8737717628479,
"epoch": 0.23141900334882792,
"grad_norm": 0.9375,
"learning_rate": 0.0004998045451045656,
"loss": 6.6986,
"mean_token_accuracy": 0.12920840978622436,
"num_tokens": 4539512.0,
"step": 2315
},
{
"entropy": 7.047420930862427,
"epoch": 0.23191882841005648,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998030550117782,
"loss": 6.8763,
"mean_token_accuracy": 0.1164549469947815,
"num_tokens": 4549030.0,
"step": 2320
},
{
"entropy": 6.9811506271362305,
"epoch": 0.23241865347128504,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998015592630037,
"loss": 6.7185,
"mean_token_accuracy": 0.12747346982359886,
"num_tokens": 4559347.0,
"step": 2325
},
{
"entropy": 6.960136795043946,
"epoch": 0.23291847853251363,
"grad_norm": 1.390625,
"learning_rate": 0.0004998000578582798,
"loss": 6.7664,
"mean_token_accuracy": 0.11763800606131554,
"num_tokens": 4568721.0,
"step": 2330
},
{
"entropy": 7.016366481781006,
"epoch": 0.2334183035937422,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997985507976443,
"loss": 6.7841,
"mean_token_accuracy": 0.12403129637241364,
"num_tokens": 4578265.0,
"step": 2335
},
{
"entropy": 6.948030614852906,
"epoch": 0.23391812865497075,
"grad_norm": 1.0,
"learning_rate": 0.0004997970380811351,
"loss": 6.7471,
"mean_token_accuracy": 0.1287473239004612,
"num_tokens": 4588694.0,
"step": 2340
},
{
"entropy": 6.98386058807373,
"epoch": 0.23441795371619933,
"grad_norm": 1.015625,
"learning_rate": 0.0004997955197087904,
"loss": 6.7427,
"mean_token_accuracy": 0.12793392390012742,
"num_tokens": 4597754.0,
"step": 2345
},
{
"entropy": 6.998698759078979,
"epoch": 0.2349177787774279,
"grad_norm": 1.015625,
"learning_rate": 0.0004997939956806481,
"loss": 6.7509,
"mean_token_accuracy": 0.12661289870738984,
"num_tokens": 4606874.0,
"step": 2350
},
{
"entropy": 6.916998815536499,
"epoch": 0.23541760383865648,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997924659967467,
"loss": 6.6337,
"mean_token_accuracy": 0.1290208764374256,
"num_tokens": 4615588.0,
"step": 2355
},
{
"entropy": 6.882282018661499,
"epoch": 0.23591742889988504,
"grad_norm": 0.9375,
"learning_rate": 0.0004997909306571248,
"loss": 6.8043,
"mean_token_accuracy": 0.12569234296679496,
"num_tokens": 4625919.0,
"step": 2360
},
{
"entropy": 6.9637391567230225,
"epoch": 0.2364172539611136,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997893896618209,
"loss": 6.7433,
"mean_token_accuracy": 0.12109558060765266,
"num_tokens": 4635539.0,
"step": 2365
},
{
"entropy": 6.9393453121185305,
"epoch": 0.2369170790223422,
"grad_norm": 0.81640625,
"learning_rate": 0.0004997878430108738,
"loss": 6.7328,
"mean_token_accuracy": 0.1250187486410141,
"num_tokens": 4645833.0,
"step": 2370
},
{
"entropy": 6.908280897140503,
"epoch": 0.23741690408357075,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997862907043224,
"loss": 6.8564,
"mean_token_accuracy": 0.12477396801114082,
"num_tokens": 4655166.0,
"step": 2375
},
{
"entropy": 7.021889305114746,
"epoch": 0.23791672914479933,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997847327422059,
"loss": 6.7268,
"mean_token_accuracy": 0.12371973022818565,
"num_tokens": 4664884.0,
"step": 2380
},
{
"entropy": 6.913659334182739,
"epoch": 0.2384165542060279,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997831691245632,
"loss": 6.821,
"mean_token_accuracy": 0.12349852249026298,
"num_tokens": 4675172.0,
"step": 2385
},
{
"entropy": 7.036015272140503,
"epoch": 0.23891637926725645,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997815998514337,
"loss": 6.7131,
"mean_token_accuracy": 0.12983411476016044,
"num_tokens": 4684588.0,
"step": 2390
},
{
"entropy": 6.936409759521484,
"epoch": 0.23941620432848504,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997800249228572,
"loss": 6.8333,
"mean_token_accuracy": 0.12305981665849686,
"num_tokens": 4694434.0,
"step": 2395
},
{
"entropy": 6.906350755691529,
"epoch": 0.2399160293897136,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997784443388732,
"loss": 6.77,
"mean_token_accuracy": 0.12484433725476266,
"num_tokens": 4705080.0,
"step": 2400
},
{
"entropy": 7.01893367767334,
"epoch": 0.24041585445094216,
"grad_norm": 1.3515625,
"learning_rate": 0.0004997768580995212,
"loss": 6.8441,
"mean_token_accuracy": 0.11970154121518135,
"num_tokens": 4714466.0,
"step": 2405
},
{
"entropy": 6.934593868255615,
"epoch": 0.24091567951217074,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997752662048413,
"loss": 6.7455,
"mean_token_accuracy": 0.12879132330417634,
"num_tokens": 4724047.0,
"step": 2410
},
{
"entropy": 6.970644330978393,
"epoch": 0.2414155045733993,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997736686548736,
"loss": 6.7948,
"mean_token_accuracy": 0.12079165503382683,
"num_tokens": 4733153.0,
"step": 2415
},
{
"entropy": 6.966701793670654,
"epoch": 0.2419153296346279,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997720654496582,
"loss": 6.6605,
"mean_token_accuracy": 0.12834780439734458,
"num_tokens": 4743239.0,
"step": 2420
},
{
"entropy": 6.895601415634156,
"epoch": 0.24241515469585645,
"grad_norm": 0.953125,
"learning_rate": 0.0004997704565892355,
"loss": 6.7198,
"mean_token_accuracy": 0.12349500507116318,
"num_tokens": 4753128.0,
"step": 2425
},
{
"entropy": 7.051494407653808,
"epoch": 0.242914979757085,
"grad_norm": 1.15625,
"learning_rate": 0.0004997688420736459,
"loss": 6.9319,
"mean_token_accuracy": 0.11143164709210396,
"num_tokens": 4764056.0,
"step": 2430
},
{
"entropy": 6.981206607818604,
"epoch": 0.2434148048183136,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997672219029301,
"loss": 6.7644,
"mean_token_accuracy": 0.12327315658330917,
"num_tokens": 4773354.0,
"step": 2435
},
{
"entropy": 6.906646108627319,
"epoch": 0.24391462987954216,
"grad_norm": 0.9375,
"learning_rate": 0.0004997655960771288,
"loss": 6.7012,
"mean_token_accuracy": 0.12343468591570854,
"num_tokens": 4783367.0,
"step": 2440
},
{
"entropy": 6.925524520874023,
"epoch": 0.24441445494077074,
"grad_norm": 1.4921875,
"learning_rate": 0.000499763964596283,
"loss": 6.5979,
"mean_token_accuracy": 0.13321683332324027,
"num_tokens": 4793028.0,
"step": 2445
},
{
"entropy": 6.9859123706817625,
"epoch": 0.2449142800019993,
"grad_norm": 0.90625,
"learning_rate": 0.0004997623274604336,
"loss": 6.7693,
"mean_token_accuracy": 0.12341286316514015,
"num_tokens": 4804032.0,
"step": 2450
},
{
"entropy": 6.952797555923462,
"epoch": 0.24541410506322786,
"grad_norm": 0.90625,
"learning_rate": 0.000499760684669622,
"loss": 6.8046,
"mean_token_accuracy": 0.12445300966501235,
"num_tokens": 4813294.0,
"step": 2455
},
{
"entropy": 6.929312753677368,
"epoch": 0.24591393012445645,
"grad_norm": 0.921875,
"learning_rate": 0.0004997590362238894,
"loss": 6.72,
"mean_token_accuracy": 0.12648141980171204,
"num_tokens": 4823461.0,
"step": 2460
},
{
"entropy": 6.869597625732422,
"epoch": 0.246413755185685,
"grad_norm": 0.9375,
"learning_rate": 0.0004997573821232772,
"loss": 6.7532,
"mean_token_accuracy": 0.12967433482408525,
"num_tokens": 4832464.0,
"step": 2465
},
{
"entropy": 6.955246114730835,
"epoch": 0.24691358024691357,
"grad_norm": 0.90625,
"learning_rate": 0.0004997557223678272,
"loss": 6.7354,
"mean_token_accuracy": 0.12091749757528306,
"num_tokens": 4842697.0,
"step": 2470
},
{
"entropy": 6.9567407131195065,
"epoch": 0.24741340530814215,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997540569575809,
"loss": 6.7918,
"mean_token_accuracy": 0.1257416233420372,
"num_tokens": 4853129.0,
"step": 2475
},
{
"entropy": 6.962767553329468,
"epoch": 0.2479132303693707,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997523858925805,
"loss": 6.8053,
"mean_token_accuracy": 0.11744737774133682,
"num_tokens": 4864048.0,
"step": 2480
},
{
"entropy": 6.851801061630249,
"epoch": 0.2484130554305993,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997507091728678,
"loss": 6.7685,
"mean_token_accuracy": 0.12485729232430458,
"num_tokens": 4874368.0,
"step": 2485
},
{
"entropy": 7.036509084701538,
"epoch": 0.24891288049182786,
"grad_norm": 0.953125,
"learning_rate": 0.0004997490267984853,
"loss": 6.7286,
"mean_token_accuracy": 0.12454072982072831,
"num_tokens": 4884813.0,
"step": 2490
},
{
"entropy": 6.88193793296814,
"epoch": 0.24941270555305642,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997473387694749,
"loss": 6.8044,
"mean_token_accuracy": 0.1220268465578556,
"num_tokens": 4895594.0,
"step": 2495
},
{
"entropy": 6.93081693649292,
"epoch": 0.249912530614285,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997456450858793,
"loss": 6.7087,
"mean_token_accuracy": 0.12661490142345427,
"num_tokens": 4904922.0,
"step": 2500
},
{
"entropy": 6.91534686088562,
"epoch": 0.25041235567551356,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997439457477412,
"loss": 6.6958,
"mean_token_accuracy": 0.12675042897462846,
"num_tokens": 4915009.0,
"step": 2505
},
{
"entropy": 6.967141246795654,
"epoch": 0.25091218073674215,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997422407551032,
"loss": 6.8361,
"mean_token_accuracy": 0.11825927272439003,
"num_tokens": 4924916.0,
"step": 2510
},
{
"entropy": 6.931365871429444,
"epoch": 0.2514120057979707,
"grad_norm": 0.96875,
"learning_rate": 0.0004997405301080082,
"loss": 6.7417,
"mean_token_accuracy": 0.12296365350484847,
"num_tokens": 4935127.0,
"step": 2515
},
{
"entropy": 6.934671115875244,
"epoch": 0.25191183085919927,
"grad_norm": 1.109375,
"learning_rate": 0.0004997388138064994,
"loss": 6.7531,
"mean_token_accuracy": 0.12879148945212365,
"num_tokens": 4944749.0,
"step": 2520
},
{
"entropy": 6.895633125305176,
"epoch": 0.25241165592042786,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997370918506198,
"loss": 6.5853,
"mean_token_accuracy": 0.13921203166246415,
"num_tokens": 4953868.0,
"step": 2525
},
{
"entropy": 6.875580501556397,
"epoch": 0.25291148098165644,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997353642404129,
"loss": 6.7073,
"mean_token_accuracy": 0.12661241069436074,
"num_tokens": 4963282.0,
"step": 2530
},
{
"entropy": 6.783172845840454,
"epoch": 0.253411306042885,
"grad_norm": 0.96875,
"learning_rate": 0.0004997336309759219,
"loss": 6.546,
"mean_token_accuracy": 0.13192983344197273,
"num_tokens": 4973099.0,
"step": 2535
},
{
"entropy": 6.893124055862427,
"epoch": 0.25391113110411356,
"grad_norm": 1.0625,
"learning_rate": 0.0004997318920571907,
"loss": 6.729,
"mean_token_accuracy": 0.12053842768073082,
"num_tokens": 4983771.0,
"step": 2540
},
{
"entropy": 6.787672758102417,
"epoch": 0.25441095616534215,
"grad_norm": 1.0078125,
"learning_rate": 0.000499730147484263,
"loss": 6.634,
"mean_token_accuracy": 0.1280955545604229,
"num_tokens": 4993648.0,
"step": 2545
},
{
"entropy": 6.879558324813843,
"epoch": 0.2549107812265707,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997283972571824,
"loss": 6.7273,
"mean_token_accuracy": 0.12890966087579728,
"num_tokens": 5002821.0,
"step": 2550
},
{
"entropy": 6.9560950756072994,
"epoch": 0.25541060628779927,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997266413759934,
"loss": 6.7505,
"mean_token_accuracy": 0.12259442135691642,
"num_tokens": 5012655.0,
"step": 2555
},
{
"entropy": 6.8323667526245115,
"epoch": 0.25591043134902786,
"grad_norm": 0.84765625,
"learning_rate": 0.0004997248798407397,
"loss": 6.7075,
"mean_token_accuracy": 0.1233678974211216,
"num_tokens": 5023649.0,
"step": 2560
},
{
"entropy": 6.922650480270386,
"epoch": 0.2564102564102564,
"grad_norm": 1.03125,
"learning_rate": 0.0004997231126514659,
"loss": 6.6567,
"mean_token_accuracy": 0.12875650003552436,
"num_tokens": 5031945.0,
"step": 2565
},
{
"entropy": 6.931125116348267,
"epoch": 0.256910081471485,
"grad_norm": 1.7109375,
"learning_rate": 0.0004997213398082165,
"loss": 6.7269,
"mean_token_accuracy": 0.12369013354182243,
"num_tokens": 5042036.0,
"step": 2570
},
{
"entropy": 6.919758987426758,
"epoch": 0.25740990653271356,
"grad_norm": 0.90234375,
"learning_rate": 0.000499719561311036,
"loss": 6.7212,
"mean_token_accuracy": 0.1250573143362999,
"num_tokens": 5050986.0,
"step": 2575
},
{
"entropy": 6.9465330123901365,
"epoch": 0.2579097315939421,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997177771599691,
"loss": 6.7021,
"mean_token_accuracy": 0.1267305240035057,
"num_tokens": 5061408.0,
"step": 2580
},
{
"entropy": 6.826210260391235,
"epoch": 0.2584095566551707,
"grad_norm": 0.984375,
"learning_rate": 0.0004997159873550607,
"loss": 6.6251,
"mean_token_accuracy": 0.12948956787586213,
"num_tokens": 5071646.0,
"step": 2585
},
{
"entropy": 6.825673580169678,
"epoch": 0.25890938171639927,
"grad_norm": 0.7890625,
"learning_rate": 0.0004997141918963561,
"loss": 6.7227,
"mean_token_accuracy": 0.12857370674610138,
"num_tokens": 5083185.0,
"step": 2590
},
{
"entropy": 6.901279354095459,
"epoch": 0.25940920677762785,
"grad_norm": 1.03125,
"learning_rate": 0.0004997123907839001,
"loss": 6.6898,
"mean_token_accuracy": 0.12528114542365074,
"num_tokens": 5092993.0,
"step": 2595
},
{
"entropy": 6.949271011352539,
"epoch": 0.2599090318388564,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997105840177381,
"loss": 6.7435,
"mean_token_accuracy": 0.12175171375274658,
"num_tokens": 5102709.0,
"step": 2600
},
{
"entropy": 6.911493444442749,
"epoch": 0.260408856900085,
"grad_norm": 1.109375,
"learning_rate": 0.0004997087715979157,
"loss": 6.6904,
"mean_token_accuracy": 0.12231686487793922,
"num_tokens": 5112858.0,
"step": 2605
},
{
"entropy": 6.915732145309448,
"epoch": 0.26090868196131356,
"grad_norm": 1.0,
"learning_rate": 0.0004997069535244785,
"loss": 6.6057,
"mean_token_accuracy": 0.13236649036407472,
"num_tokens": 5123573.0,
"step": 2610
},
{
"entropy": 6.77972183227539,
"epoch": 0.2614085070225421,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997051297974722,
"loss": 6.7799,
"mean_token_accuracy": 0.11823756992816925,
"num_tokens": 5132908.0,
"step": 2615
},
{
"entropy": 6.903223943710327,
"epoch": 0.2619083320837707,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997033004169425,
"loss": 6.6634,
"mean_token_accuracy": 0.13124899789690972,
"num_tokens": 5142331.0,
"step": 2620
},
{
"entropy": 6.986439514160156,
"epoch": 0.26240815714499927,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997014653829357,
"loss": 6.7434,
"mean_token_accuracy": 0.1203452467918396,
"num_tokens": 5152517.0,
"step": 2625
},
{
"entropy": 6.8694833278656,
"epoch": 0.2629079822062278,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996996246954977,
"loss": 6.6749,
"mean_token_accuracy": 0.129129771143198,
"num_tokens": 5162418.0,
"step": 2630
},
{
"entropy": 6.847884941101074,
"epoch": 0.2634078072674564,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996977783546752,
"loss": 6.7101,
"mean_token_accuracy": 0.12350467145442963,
"num_tokens": 5171948.0,
"step": 2635
},
{
"entropy": 6.947595262527466,
"epoch": 0.26390763232868497,
"grad_norm": 1.03125,
"learning_rate": 0.0004996959263605143,
"loss": 6.626,
"mean_token_accuracy": 0.1277811646461487,
"num_tokens": 5180963.0,
"step": 2640
},
{
"entropy": 6.835096788406372,
"epoch": 0.2644074573899135,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996940687130617,
"loss": 6.743,
"mean_token_accuracy": 0.12740767300128936,
"num_tokens": 5192305.0,
"step": 2645
},
{
"entropy": 6.8839599132537845,
"epoch": 0.2649072824511421,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996922054123641,
"loss": 6.6193,
"mean_token_accuracy": 0.12462258934974671,
"num_tokens": 5201503.0,
"step": 2650
},
{
"entropy": 6.921473693847656,
"epoch": 0.2654071075123707,
"grad_norm": 1.0625,
"learning_rate": 0.0004996903364584685,
"loss": 6.6076,
"mean_token_accuracy": 0.12627073228359223,
"num_tokens": 5210206.0,
"step": 2655
},
{
"entropy": 6.762358474731445,
"epoch": 0.26590693257359926,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996884618514219,
"loss": 6.6135,
"mean_token_accuracy": 0.1311767116189003,
"num_tokens": 5220256.0,
"step": 2660
},
{
"entropy": 6.803161573410034,
"epoch": 0.2664067576348278,
"grad_norm": 1.046875,
"learning_rate": 0.0004996865815912713,
"loss": 6.637,
"mean_token_accuracy": 0.1329315148293972,
"num_tokens": 5229128.0,
"step": 2665
},
{
"entropy": 6.895367097854614,
"epoch": 0.2669065826960564,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996846956780642,
"loss": 6.5866,
"mean_token_accuracy": 0.12857141569256783,
"num_tokens": 5239468.0,
"step": 2670
},
{
"entropy": 6.834448194503784,
"epoch": 0.26740640775728497,
"grad_norm": 0.96484375,
"learning_rate": 0.000499682804111848,
"loss": 6.6913,
"mean_token_accuracy": 0.12485808730125428,
"num_tokens": 5247957.0,
"step": 2675
},
{
"entropy": 6.868956804275513,
"epoch": 0.2679062328185135,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996809068926702,
"loss": 6.6983,
"mean_token_accuracy": 0.1235993355512619,
"num_tokens": 5257960.0,
"step": 2680
},
{
"entropy": 6.826748180389404,
"epoch": 0.2684060578797421,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996790040205786,
"loss": 6.5729,
"mean_token_accuracy": 0.1311322994530201,
"num_tokens": 5266696.0,
"step": 2685
},
{
"entropy": 6.84700608253479,
"epoch": 0.2689058829409707,
"grad_norm": 1.03125,
"learning_rate": 0.0004996770954956211,
"loss": 6.542,
"mean_token_accuracy": 0.1364964410662651,
"num_tokens": 5275619.0,
"step": 2690
},
{
"entropy": 6.814588451385498,
"epoch": 0.2694057080021992,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996751813178456,
"loss": 6.6823,
"mean_token_accuracy": 0.12757562324404717,
"num_tokens": 5284884.0,
"step": 2695
},
{
"entropy": 6.885785722732544,
"epoch": 0.2699055330634278,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996732614873005,
"loss": 6.5761,
"mean_token_accuracy": 0.13537995815277098,
"num_tokens": 5293909.0,
"step": 2700
},
{
"entropy": 6.8018866062164305,
"epoch": 0.2704053581246564,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996713360040339,
"loss": 6.6667,
"mean_token_accuracy": 0.12470442354679108,
"num_tokens": 5303142.0,
"step": 2705
},
{
"entropy": 6.832062864303589,
"epoch": 0.27090518318588497,
"grad_norm": 1.015625,
"learning_rate": 0.0004996694048680942,
"loss": 6.6007,
"mean_token_accuracy": 0.1250078909099102,
"num_tokens": 5313055.0,
"step": 2710
},
{
"entropy": 6.856688165664673,
"epoch": 0.2714050082471135,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996674680795302,
"loss": 6.7171,
"mean_token_accuracy": 0.11953588724136352,
"num_tokens": 5323773.0,
"step": 2715
},
{
"entropy": 6.701740312576294,
"epoch": 0.2719048333083421,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996655256383905,
"loss": 6.5089,
"mean_token_accuracy": 0.1433877095580101,
"num_tokens": 5334613.0,
"step": 2720
},
{
"entropy": 6.785739469528198,
"epoch": 0.2724046583695707,
"grad_norm": 0.88671875,
"learning_rate": 0.000499663577544724,
"loss": 6.5988,
"mean_token_accuracy": 0.1337459959089756,
"num_tokens": 5344151.0,
"step": 2725
},
{
"entropy": 6.790981101989746,
"epoch": 0.2729044834307992,
"grad_norm": 1.03125,
"learning_rate": 0.0004996616237985796,
"loss": 6.5667,
"mean_token_accuracy": 0.1349438451230526,
"num_tokens": 5352425.0,
"step": 2730
},
{
"entropy": 6.783331489562988,
"epoch": 0.2734043084920278,
"grad_norm": 0.92578125,
"learning_rate": 0.0004996596644000066,
"loss": 6.7637,
"mean_token_accuracy": 0.12367659956216812,
"num_tokens": 5361811.0,
"step": 2735
},
{
"entropy": 6.8914836883544925,
"epoch": 0.2739041335532564,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996576993490543,
"loss": 6.6965,
"mean_token_accuracy": 0.12055673524737358,
"num_tokens": 5371463.0,
"step": 2740
},
{
"entropy": 6.8567499160766605,
"epoch": 0.2744039586144849,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996557286457721,
"loss": 6.5675,
"mean_token_accuracy": 0.13478375598788261,
"num_tokens": 5381464.0,
"step": 2745
},
{
"entropy": 6.737442064285278,
"epoch": 0.2749037836757135,
"grad_norm": 1.015625,
"learning_rate": 0.0004996537522902096,
"loss": 6.643,
"mean_token_accuracy": 0.12358913272619247,
"num_tokens": 5391012.0,
"step": 2750
},
{
"entropy": 6.900225830078125,
"epoch": 0.2754036087369421,
"grad_norm": 0.953125,
"learning_rate": 0.0004996517702824163,
"loss": 6.7441,
"mean_token_accuracy": 0.1277411937713623,
"num_tokens": 5401198.0,
"step": 2755
},
{
"entropy": 6.965499496459961,
"epoch": 0.2759034337981706,
"grad_norm": 0.890625,
"learning_rate": 0.0004996497826224425,
"loss": 6.6872,
"mean_token_accuracy": 0.1266508638858795,
"num_tokens": 5410592.0,
"step": 2760
},
{
"entropy": 6.733990907669067,
"epoch": 0.2764032588593992,
"grad_norm": 0.953125,
"learning_rate": 0.0004996477893103379,
"loss": 6.6455,
"mean_token_accuracy": 0.13018811568617822,
"num_tokens": 5420878.0,
"step": 2765
},
{
"entropy": 6.837689924240112,
"epoch": 0.2769030839206278,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996457903461527,
"loss": 6.6557,
"mean_token_accuracy": 0.12990267798304558,
"num_tokens": 5431199.0,
"step": 2770
},
{
"entropy": 6.825644302368164,
"epoch": 0.2774029089818564,
"grad_norm": 0.87109375,
"learning_rate": 0.0004996437857299373,
"loss": 6.5869,
"mean_token_accuracy": 0.13335442095994948,
"num_tokens": 5441661.0,
"step": 2775
},
{
"entropy": 6.765251636505127,
"epoch": 0.2779027340430849,
"grad_norm": 0.92578125,
"learning_rate": 0.000499641775461742,
"loss": 6.5892,
"mean_token_accuracy": 0.13393662124872208,
"num_tokens": 5451863.0,
"step": 2780
},
{
"entropy": 6.830867719650269,
"epoch": 0.2784025591043135,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996397595416173,
"loss": 6.6382,
"mean_token_accuracy": 0.13025332391262054,
"num_tokens": 5461600.0,
"step": 2785
},
{
"entropy": 6.891679525375366,
"epoch": 0.2789023841655421,
"grad_norm": 0.953125,
"learning_rate": 0.0004996377379696143,
"loss": 6.6689,
"mean_token_accuracy": 0.1225666806101799,
"num_tokens": 5470097.0,
"step": 2790
},
{
"entropy": 6.880445766448974,
"epoch": 0.2794022092267706,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996357107457834,
"loss": 6.7974,
"mean_token_accuracy": 0.1164699487388134,
"num_tokens": 5480906.0,
"step": 2795
},
{
"entropy": 6.941256761550903,
"epoch": 0.2799020342879992,
"grad_norm": 1.0,
"learning_rate": 0.000499633677870176,
"loss": 6.677,
"mean_token_accuracy": 0.12477467060089112,
"num_tokens": 5489732.0,
"step": 2800
},
{
"entropy": 6.70820631980896,
"epoch": 0.2804018593492278,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996316393428429,
"loss": 6.5058,
"mean_token_accuracy": 0.13490227609872818,
"num_tokens": 5498422.0,
"step": 2805
},
{
"entropy": 6.741797018051147,
"epoch": 0.2809016844104563,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996295951638357,
"loss": 6.6417,
"mean_token_accuracy": 0.12436849176883698,
"num_tokens": 5507958.0,
"step": 2810
},
{
"entropy": 6.871691799163818,
"epoch": 0.2814015094716849,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996275453332057,
"loss": 6.6753,
"mean_token_accuracy": 0.12770553976297377,
"num_tokens": 5516752.0,
"step": 2815
},
{
"entropy": 6.804839038848877,
"epoch": 0.2819013345329135,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996254898510043,
"loss": 6.68,
"mean_token_accuracy": 0.12850109338760377,
"num_tokens": 5527211.0,
"step": 2820
},
{
"entropy": 6.820689105987549,
"epoch": 0.282401159594142,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996234287172835,
"loss": 6.5746,
"mean_token_accuracy": 0.12999471724033357,
"num_tokens": 5536925.0,
"step": 2825
},
{
"entropy": 6.883645534515381,
"epoch": 0.2829009846553706,
"grad_norm": 0.91796875,
"learning_rate": 0.000499621361932095,
"loss": 6.7608,
"mean_token_accuracy": 0.12017307057976723,
"num_tokens": 5546800.0,
"step": 2830
},
{
"entropy": 6.764953994750977,
"epoch": 0.2834008097165992,
"grad_norm": 1.09375,
"learning_rate": 0.0004996192894954909,
"loss": 6.6006,
"mean_token_accuracy": 0.12868443727493287,
"num_tokens": 5557266.0,
"step": 2835
},
{
"entropy": 6.817695951461792,
"epoch": 0.2839006347778278,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996172114075232,
"loss": 6.5858,
"mean_token_accuracy": 0.13033762723207473,
"num_tokens": 5566169.0,
"step": 2840
},
{
"entropy": 6.8027301788330075,
"epoch": 0.2844004598390563,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996151276682443,
"loss": 6.6751,
"mean_token_accuracy": 0.1305215150117874,
"num_tokens": 5575566.0,
"step": 2845
},
{
"entropy": 6.791625738143921,
"epoch": 0.2849002849002849,
"grad_norm": 0.921875,
"learning_rate": 0.0004996130382777065,
"loss": 6.5596,
"mean_token_accuracy": 0.1286344051361084,
"num_tokens": 5584804.0,
"step": 2850
},
{
"entropy": 6.78387484550476,
"epoch": 0.2854001099615135,
"grad_norm": 0.9375,
"learning_rate": 0.0004996109432359625,
"loss": 6.6093,
"mean_token_accuracy": 0.12663276568055154,
"num_tokens": 5594361.0,
"step": 2855
},
{
"entropy": 6.868547153472901,
"epoch": 0.285899935022742,
"grad_norm": 1.921875,
"learning_rate": 0.0004996088425430651,
"loss": 6.619,
"mean_token_accuracy": 0.12699493318796157,
"num_tokens": 5605682.0,
"step": 2860
},
{
"entropy": 6.808011245727539,
"epoch": 0.2863997600839706,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996067361990669,
"loss": 6.6979,
"mean_token_accuracy": 0.12324106916785241,
"num_tokens": 5617251.0,
"step": 2865
},
{
"entropy": 6.7848552703857425,
"epoch": 0.2868995851451992,
"grad_norm": 0.85546875,
"learning_rate": 0.0004996046242040209,
"loss": 6.5719,
"mean_token_accuracy": 0.13272615447640418,
"num_tokens": 5627140.0,
"step": 2870
},
{
"entropy": 6.77541823387146,
"epoch": 0.28739941020642773,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996025065579805,
"loss": 6.5212,
"mean_token_accuracy": 0.13640954494476318,
"num_tokens": 5635486.0,
"step": 2875
},
{
"entropy": 6.838054370880127,
"epoch": 0.2878992352676563,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996003832609988,
"loss": 6.6731,
"mean_token_accuracy": 0.1194192998111248,
"num_tokens": 5645621.0,
"step": 2880
},
{
"entropy": 6.928879976272583,
"epoch": 0.2883990603288849,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995982543131292,
"loss": 6.7457,
"mean_token_accuracy": 0.12103333249688149,
"num_tokens": 5655626.0,
"step": 2885
},
{
"entropy": 6.829890489578247,
"epoch": 0.28889888539011344,
"grad_norm": 0.87109375,
"learning_rate": 0.0004995961197144254,
"loss": 6.6795,
"mean_token_accuracy": 0.12712434753775598,
"num_tokens": 5664547.0,
"step": 2890
},
{
"entropy": 6.834103298187256,
"epoch": 0.289398710451342,
"grad_norm": 0.93359375,
"learning_rate": 0.000499593979464941,
"loss": 6.6403,
"mean_token_accuracy": 0.12402470111846924,
"num_tokens": 5674861.0,
"step": 2895
},
{
"entropy": 6.787194728851318,
"epoch": 0.2898985355125706,
"grad_norm": 0.859375,
"learning_rate": 0.0004995918335647298,
"loss": 6.5982,
"mean_token_accuracy": 0.12460636496543884,
"num_tokens": 5684143.0,
"step": 2900
},
{
"entropy": 6.791854190826416,
"epoch": 0.2903983605737992,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995896820138459,
"loss": 6.558,
"mean_token_accuracy": 0.12753776907920839,
"num_tokens": 5693837.0,
"step": 2905
},
{
"entropy": 6.880584764480591,
"epoch": 0.29089818563502773,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995875248123434,
"loss": 6.655,
"mean_token_accuracy": 0.12204006910324097,
"num_tokens": 5704827.0,
"step": 2910
},
{
"entropy": 6.8240148544311525,
"epoch": 0.2913980106962563,
"grad_norm": 0.953125,
"learning_rate": 0.0004995853619602766,
"loss": 6.694,
"mean_token_accuracy": 0.1257927805185318,
"num_tokens": 5714419.0,
"step": 2915
},
{
"entropy": 6.83175892829895,
"epoch": 0.2918978357574849,
"grad_norm": 0.7890625,
"learning_rate": 0.0004995831934576998,
"loss": 6.6069,
"mean_token_accuracy": 0.1266295939683914,
"num_tokens": 5724983.0,
"step": 2920
},
{
"entropy": 6.7419945240020756,
"epoch": 0.29239766081871343,
"grad_norm": 0.87109375,
"learning_rate": 0.0004995810193046677,
"loss": 6.5335,
"mean_token_accuracy": 0.1366861306130886,
"num_tokens": 5734471.0,
"step": 2925
},
{
"entropy": 6.81366982460022,
"epoch": 0.292897485879942,
"grad_norm": 0.9765625,
"learning_rate": 0.000499578839501235,
"loss": 6.6862,
"mean_token_accuracy": 0.12929900363087654,
"num_tokens": 5744801.0,
"step": 2930
},
{
"entropy": 6.82947154045105,
"epoch": 0.2933973109411706,
"grad_norm": 0.97265625,
"learning_rate": 0.0004995766540474564,
"loss": 6.6191,
"mean_token_accuracy": 0.1266006588935852,
"num_tokens": 5754267.0,
"step": 2935
},
{
"entropy": 6.7092602252960205,
"epoch": 0.29389713600239914,
"grad_norm": 0.83203125,
"learning_rate": 0.000499574462943387,
"loss": 6.5378,
"mean_token_accuracy": 0.13515163734555244,
"num_tokens": 5764466.0,
"step": 2940
},
{
"entropy": 6.785275030136108,
"epoch": 0.2943969610636277,
"grad_norm": 0.84765625,
"learning_rate": 0.000499572266189082,
"loss": 6.6676,
"mean_token_accuracy": 0.12391036003828049,
"num_tokens": 5775158.0,
"step": 2945
},
{
"entropy": 6.8545256614685055,
"epoch": 0.2948967861248563,
"grad_norm": 1.03125,
"learning_rate": 0.0004995700637845964,
"loss": 6.4676,
"mean_token_accuracy": 0.13657867833971976,
"num_tokens": 5783320.0,
"step": 2950
},
{
"entropy": 6.7285840034484865,
"epoch": 0.29539661118608485,
"grad_norm": 0.859375,
"learning_rate": 0.0004995678557299859,
"loss": 6.6247,
"mean_token_accuracy": 0.13014830350875856,
"num_tokens": 5792980.0,
"step": 2955
},
{
"entropy": 6.84141960144043,
"epoch": 0.29589643624731343,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995656420253058,
"loss": 6.6712,
"mean_token_accuracy": 0.1261442005634308,
"num_tokens": 5802338.0,
"step": 2960
},
{
"entropy": 6.790224504470825,
"epoch": 0.296396261308542,
"grad_norm": 1.015625,
"learning_rate": 0.000499563422670612,
"loss": 6.5638,
"mean_token_accuracy": 0.13231244459748268,
"num_tokens": 5811388.0,
"step": 2965
},
{
"entropy": 6.8048484325408936,
"epoch": 0.2968960863697706,
"grad_norm": 0.921875,
"learning_rate": 0.0004995611976659604,
"loss": 6.6027,
"mean_token_accuracy": 0.12747790962457656,
"num_tokens": 5821108.0,
"step": 2970
},
{
"entropy": 6.801449108123779,
"epoch": 0.29739591143099914,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995589670114067,
"loss": 6.6456,
"mean_token_accuracy": 0.1197404257953167,
"num_tokens": 5831389.0,
"step": 2975
},
{
"entropy": 6.763365077972412,
"epoch": 0.2978957364922277,
"grad_norm": 0.87109375,
"learning_rate": 0.0004995567307070072,
"loss": 6.5087,
"mean_token_accuracy": 0.13231854736804963,
"num_tokens": 5840392.0,
"step": 2980
},
{
"entropy": 6.758113431930542,
"epoch": 0.2983955615534563,
"grad_norm": 0.90234375,
"learning_rate": 0.0004995544887528182,
"loss": 6.5226,
"mean_token_accuracy": 0.13557453379034995,
"num_tokens": 5850032.0,
"step": 2985
},
{
"entropy": 6.788800477981567,
"epoch": 0.29889538661468484,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995522411488959,
"loss": 6.5933,
"mean_token_accuracy": 0.13116523027420043,
"num_tokens": 5859204.0,
"step": 2990
},
{
"entropy": 6.742308712005615,
"epoch": 0.29939521167591343,
"grad_norm": 0.984375,
"learning_rate": 0.0004995499878952971,
"loss": 6.5951,
"mean_token_accuracy": 0.13226681798696518,
"num_tokens": 5868582.0,
"step": 2995
},
{
"entropy": 6.788286781311035,
"epoch": 0.299895036737142,
"grad_norm": 0.875,
"learning_rate": 0.0004995477289920784,
"loss": 6.5333,
"mean_token_accuracy": 0.13418369889259338,
"num_tokens": 5879601.0,
"step": 3000
},
{
"epoch": 0.299895036737142,
"eval_entropy": 6.630776028631889,
"eval_loss": 6.627320289611816,
"eval_mean_token_accuracy": 0.1339761095548169,
"eval_num_tokens": 5879601.0,
"eval_runtime": 18.4371,
"eval_samples_per_second": 1683.83,
"eval_steps_per_second": 210.499,
"step": 3000
},
{
"entropy": 6.783554172515869,
"epoch": 0.30039486179837055,
"grad_norm": 0.85546875,
"learning_rate": 0.0004995454644392967,
"loss": 6.606,
"mean_token_accuracy": 0.12053637430071831,
"num_tokens": 5890255.0,
"step": 3005
},
{
"entropy": 6.720775699615478,
"epoch": 0.30089468685959914,
"grad_norm": 1.0,
"learning_rate": 0.0004995431942370087,
"loss": 6.6203,
"mean_token_accuracy": 0.13547174111008645,
"num_tokens": 5899965.0,
"step": 3010
},
{
"entropy": 6.76155366897583,
"epoch": 0.3013945119208277,
"grad_norm": 1.03125,
"learning_rate": 0.0004995409183852718,
"loss": 6.5231,
"mean_token_accuracy": 0.13390932902693747,
"num_tokens": 5910082.0,
"step": 3015
},
{
"entropy": 6.7451270580291744,
"epoch": 0.30189433698205626,
"grad_norm": 0.97265625,
"learning_rate": 0.0004995386368841432,
"loss": 6.6386,
"mean_token_accuracy": 0.12414429485797882,
"num_tokens": 5920267.0,
"step": 3020
},
{
"entropy": 6.810834884643555,
"epoch": 0.30239416204328484,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995363497336802,
"loss": 6.6254,
"mean_token_accuracy": 0.12998512089252473,
"num_tokens": 5929275.0,
"step": 3025
},
{
"entropy": 6.804172372817993,
"epoch": 0.30289398710451343,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995340569339405,
"loss": 6.6433,
"mean_token_accuracy": 0.1219985917210579,
"num_tokens": 5939305.0,
"step": 3030
},
{
"entropy": 6.841426134109497,
"epoch": 0.303393812165742,
"grad_norm": 1.515625,
"learning_rate": 0.0004995317584849817,
"loss": 6.6928,
"mean_token_accuracy": 0.12304298430681229,
"num_tokens": 5950816.0,
"step": 3035
},
{
"entropy": 6.70551700592041,
"epoch": 0.30389363722697055,
"grad_norm": 0.890625,
"learning_rate": 0.0004995294543868615,
"loss": 6.4907,
"mean_token_accuracy": 0.1406556949019432,
"num_tokens": 5959824.0,
"step": 3040
},
{
"entropy": 6.794424343109131,
"epoch": 0.30439346228819913,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995271446396382,
"loss": 6.5975,
"mean_token_accuracy": 0.1330313928425312,
"num_tokens": 5968752.0,
"step": 3045
},
{
"entropy": 6.750303077697754,
"epoch": 0.3048932873494277,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995248292433697,
"loss": 6.5396,
"mean_token_accuracy": 0.12796283811330794,
"num_tokens": 5979085.0,
"step": 3050
},
{
"entropy": 6.761692380905151,
"epoch": 0.30539311241065625,
"grad_norm": 0.78515625,
"learning_rate": 0.0004995225081981142,
"loss": 6.5971,
"mean_token_accuracy": 0.12856600284576417,
"num_tokens": 5990378.0,
"step": 3055
},
{
"entropy": 6.748683977127075,
"epoch": 0.30589293747188484,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995201815039301,
"loss": 6.4713,
"mean_token_accuracy": 0.14121382609009742,
"num_tokens": 5999524.0,
"step": 3060
},
{
"entropy": 6.7300026416778564,
"epoch": 0.3063927625331134,
"grad_norm": 0.875,
"learning_rate": 0.0004995178491608762,
"loss": 6.5851,
"mean_token_accuracy": 0.1298016183078289,
"num_tokens": 6009928.0,
"step": 3065
},
{
"entropy": 6.667418193817139,
"epoch": 0.30689258759434196,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995155111690107,
"loss": 6.4623,
"mean_token_accuracy": 0.13725680708885193,
"num_tokens": 6019683.0,
"step": 3070
},
{
"entropy": 6.802922630310059,
"epoch": 0.30739241265557055,
"grad_norm": 0.9921875,
"learning_rate": 0.000499513167528393,
"loss": 6.6029,
"mean_token_accuracy": 0.12657553628087043,
"num_tokens": 6030506.0,
"step": 3075
},
{
"entropy": 6.765255117416382,
"epoch": 0.30789223771679913,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995108182390815,
"loss": 6.5603,
"mean_token_accuracy": 0.130464369058609,
"num_tokens": 6039720.0,
"step": 3080
},
{
"entropy": 6.649257755279541,
"epoch": 0.30839206277802766,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995084633011359,
"loss": 6.4125,
"mean_token_accuracy": 0.1420907124876976,
"num_tokens": 6049336.0,
"step": 3085
},
{
"entropy": 6.6750640869140625,
"epoch": 0.30889188783925625,
"grad_norm": 0.9609375,
"learning_rate": 0.000499506102714615,
"loss": 6.4497,
"mean_token_accuracy": 0.13145486935973166,
"num_tokens": 6059420.0,
"step": 3090
},
{
"entropy": 6.771726655960083,
"epoch": 0.30939171290048484,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995037364795782,
"loss": 6.5325,
"mean_token_accuracy": 0.1350548505783081,
"num_tokens": 6069035.0,
"step": 3095
},
{
"entropy": 6.751552534103394,
"epoch": 0.3098915379617134,
"grad_norm": 0.9140625,
"learning_rate": 0.0004995013645960853,
"loss": 6.5333,
"mean_token_accuracy": 0.13230074495077132,
"num_tokens": 6077863.0,
"step": 3100
},
{
"entropy": 6.707284641265869,
"epoch": 0.31039136302294196,
"grad_norm": 0.8984375,
"learning_rate": 0.0004994989870641958,
"loss": 6.5719,
"mean_token_accuracy": 0.12928358912467958,
"num_tokens": 6088929.0,
"step": 3105
},
{
"entropy": 6.739338779449463,
"epoch": 0.31089118808417054,
"grad_norm": 0.953125,
"learning_rate": 0.0004994966038839695,
"loss": 6.5183,
"mean_token_accuracy": 0.13488781303167344,
"num_tokens": 6098673.0,
"step": 3110
},
{
"entropy": 6.739440679550171,
"epoch": 0.31139101314539913,
"grad_norm": 1.1640625,
"learning_rate": 0.0004994942150554666,
"loss": 6.3982,
"mean_token_accuracy": 0.13903475776314736,
"num_tokens": 6107289.0,
"step": 3115
},
{
"entropy": 6.680595541000367,
"epoch": 0.31189083820662766,
"grad_norm": 0.828125,
"learning_rate": 0.0004994918205787468,
"loss": 6.4573,
"mean_token_accuracy": 0.1357430823147297,
"num_tokens": 6117811.0,
"step": 3120
},
{
"entropy": 6.713736963272095,
"epoch": 0.31239066326785625,
"grad_norm": 0.88671875,
"learning_rate": 0.0004994894204538706,
"loss": 6.5585,
"mean_token_accuracy": 0.12887152135372162,
"num_tokens": 6127740.0,
"step": 3125
},
{
"entropy": 6.769714260101319,
"epoch": 0.31289048832908484,
"grad_norm": 0.8984375,
"learning_rate": 0.0004994870146808984,
"loss": 6.5349,
"mean_token_accuracy": 0.13567256852984427,
"num_tokens": 6137253.0,
"step": 3130
},
{
"entropy": 6.71686201095581,
"epoch": 0.31339031339031337,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994846032598906,
"loss": 6.5743,
"mean_token_accuracy": 0.13248624205589293,
"num_tokens": 6147420.0,
"step": 3135
},
{
"entropy": 6.699933815002441,
"epoch": 0.31389013845154196,
"grad_norm": 0.890625,
"learning_rate": 0.0004994821861909082,
"loss": 6.5169,
"mean_token_accuracy": 0.13643142953515053,
"num_tokens": 6156436.0,
"step": 3140
},
{
"entropy": 6.659816646575928,
"epoch": 0.31438996351277054,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994797634740114,
"loss": 6.4767,
"mean_token_accuracy": 0.12807029634714126,
"num_tokens": 6165526.0,
"step": 3145
},
{
"entropy": 6.840546131134033,
"epoch": 0.3148897885739991,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994773351092618,
"loss": 6.6801,
"mean_token_accuracy": 0.12378201708197593,
"num_tokens": 6176093.0,
"step": 3150
},
{
"entropy": 6.737400197982788,
"epoch": 0.31538961363522766,
"grad_norm": 0.97265625,
"learning_rate": 0.00049947490109672,
"loss": 6.4903,
"mean_token_accuracy": 0.1427481345832348,
"num_tokens": 6185373.0,
"step": 3155
},
{
"entropy": 6.738193511962891,
"epoch": 0.31588943869645625,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994724614364476,
"loss": 6.614,
"mean_token_accuracy": 0.12659135907888414,
"num_tokens": 6196252.0,
"step": 3160
},
{
"entropy": 6.806258583068848,
"epoch": 0.31638926375768484,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994700161285058,
"loss": 6.5388,
"mean_token_accuracy": 0.13199907094240187,
"num_tokens": 6206463.0,
"step": 3165
},
{
"entropy": 6.748357820510864,
"epoch": 0.31688908881891337,
"grad_norm": 0.90234375,
"learning_rate": 0.000499467565172956,
"loss": 6.5052,
"mean_token_accuracy": 0.13882621228694916,
"num_tokens": 6216676.0,
"step": 3170
},
{
"entropy": 6.709172964096069,
"epoch": 0.31738891388014195,
"grad_norm": 0.859375,
"learning_rate": 0.0004994651085698602,
"loss": 6.5471,
"mean_token_accuracy": 0.13136286586523055,
"num_tokens": 6227261.0,
"step": 3175
},
{
"entropy": 6.738838720321655,
"epoch": 0.31788873894137054,
"grad_norm": 0.98828125,
"learning_rate": 0.00049946264631928,
"loss": 6.4719,
"mean_token_accuracy": 0.13089419528841972,
"num_tokens": 6237468.0,
"step": 3180
},
{
"entropy": 6.661934852600098,
"epoch": 0.3183885640025991,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994601784212774,
"loss": 6.4472,
"mean_token_accuracy": 0.1331954061985016,
"num_tokens": 6247164.0,
"step": 3185
},
{
"entropy": 6.6419719696044925,
"epoch": 0.31888838906382766,
"grad_norm": 0.8125,
"learning_rate": 0.0004994577048759144,
"loss": 6.5523,
"mean_token_accuracy": 0.132597766071558,
"num_tokens": 6257635.0,
"step": 3190
},
{
"entropy": 6.771932888031006,
"epoch": 0.31938821412505625,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994552256832533,
"loss": 6.6061,
"mean_token_accuracy": 0.1313496634364128,
"num_tokens": 6266823.0,
"step": 3195
},
{
"entropy": 6.75528392791748,
"epoch": 0.3198880391862848,
"grad_norm": 1.046875,
"learning_rate": 0.0004994527408433566,
"loss": 6.531,
"mean_token_accuracy": 0.13688323944807052,
"num_tokens": 6276527.0,
"step": 3200
},
{
"entropy": 6.73806791305542,
"epoch": 0.32038786424751337,
"grad_norm": 1.015625,
"learning_rate": 0.0004994502503562865,
"loss": 6.5808,
"mean_token_accuracy": 0.12626547291874884,
"num_tokens": 6286499.0,
"step": 3205
},
{
"entropy": 6.7163379192352295,
"epoch": 0.32088768930874195,
"grad_norm": 0.859375,
"learning_rate": 0.000499447754222106,
"loss": 6.512,
"mean_token_accuracy": 0.1329052247107029,
"num_tokens": 6296550.0,
"step": 3210
},
{
"entropy": 6.700007677078247,
"epoch": 0.3213875143699705,
"grad_norm": 0.875,
"learning_rate": 0.0004994452524408777,
"loss": 6.553,
"mean_token_accuracy": 0.12814397141337394,
"num_tokens": 6306951.0,
"step": 3215
},
{
"entropy": 6.822968578338623,
"epoch": 0.32188733943119907,
"grad_norm": 1.109375,
"learning_rate": 0.0004994427450126646,
"loss": 6.5738,
"mean_token_accuracy": 0.1254580870270729,
"num_tokens": 6316026.0,
"step": 3220
},
{
"entropy": 6.688732433319092,
"epoch": 0.32238716449242766,
"grad_norm": 1.09375,
"learning_rate": 0.0004994402319375298,
"loss": 6.5166,
"mean_token_accuracy": 0.13167576789855956,
"num_tokens": 6325064.0,
"step": 3225
},
{
"entropy": 6.757549905776978,
"epoch": 0.32288698955365625,
"grad_norm": 1.09375,
"learning_rate": 0.0004994377132155366,
"loss": 6.5588,
"mean_token_accuracy": 0.1331995628774166,
"num_tokens": 6334443.0,
"step": 3230
},
{
"entropy": 6.740247392654419,
"epoch": 0.3233868146148848,
"grad_norm": 0.90234375,
"learning_rate": 0.0004994351888467482,
"loss": 6.5674,
"mean_token_accuracy": 0.13573309332132338,
"num_tokens": 6346103.0,
"step": 3235
},
{
"entropy": 6.6513190269470215,
"epoch": 0.32388663967611336,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994326588312282,
"loss": 6.532,
"mean_token_accuracy": 0.1347044326364994,
"num_tokens": 6356576.0,
"step": 3240
},
{
"entropy": 6.810565853118897,
"epoch": 0.32438646473734195,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994301231690403,
"loss": 6.53,
"mean_token_accuracy": 0.13375264555215835,
"num_tokens": 6364796.0,
"step": 3245
},
{
"entropy": 6.718886804580689,
"epoch": 0.3248862897985705,
"grad_norm": 0.8984375,
"learning_rate": 0.0004994275818602482,
"loss": 6.611,
"mean_token_accuracy": 0.12929279431700708,
"num_tokens": 6374951.0,
"step": 3250
},
{
"entropy": 6.654961729049683,
"epoch": 0.32538611485979907,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994250349049159,
"loss": 6.4877,
"mean_token_accuracy": 0.1318598873913288,
"num_tokens": 6384230.0,
"step": 3255
},
{
"entropy": 6.6994367122650145,
"epoch": 0.32588593992102766,
"grad_norm": 0.88671875,
"learning_rate": 0.0004994224823031075,
"loss": 6.5009,
"mean_token_accuracy": 0.13323622792959214,
"num_tokens": 6394070.0,
"step": 3260
},
{
"entropy": 6.685267400741577,
"epoch": 0.3263857649822562,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994199240548873,
"loss": 6.5074,
"mean_token_accuracy": 0.13391340300440788,
"num_tokens": 6403968.0,
"step": 3265
},
{
"entropy": 6.800641918182373,
"epoch": 0.3268855900434848,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994173601603194,
"loss": 6.6056,
"mean_token_accuracy": 0.1315523624420166,
"num_tokens": 6413254.0,
"step": 3270
},
{
"entropy": 6.716506385803223,
"epoch": 0.32738541510471336,
"grad_norm": 1.03125,
"learning_rate": 0.0004994147906194684,
"loss": 6.4388,
"mean_token_accuracy": 0.13731389120221138,
"num_tokens": 6421266.0,
"step": 3275
},
{
"entropy": 6.752088499069214,
"epoch": 0.3278852401659419,
"grad_norm": 1.03125,
"learning_rate": 0.0004994122154323992,
"loss": 6.5521,
"mean_token_accuracy": 0.13202136158943176,
"num_tokens": 6429596.0,
"step": 3280
},
{
"entropy": 6.7733715057373045,
"epoch": 0.3283850652271705,
"grad_norm": 0.91015625,
"learning_rate": 0.0004994096345991763,
"loss": 6.5447,
"mean_token_accuracy": 0.12929541245102882,
"num_tokens": 6440825.0,
"step": 3285
},
{
"entropy": 6.628010320663452,
"epoch": 0.32888489028839907,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994070481198649,
"loss": 6.5088,
"mean_token_accuracy": 0.13302811831235886,
"num_tokens": 6450576.0,
"step": 3290
},
{
"entropy": 6.725134086608887,
"epoch": 0.32938471534962765,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994044559945296,
"loss": 6.5108,
"mean_token_accuracy": 0.1391789697110653,
"num_tokens": 6459673.0,
"step": 3295
},
{
"entropy": 6.701463890075684,
"epoch": 0.3298845404108562,
"grad_norm": 0.89453125,
"learning_rate": 0.0004994018582232361,
"loss": 6.4957,
"mean_token_accuracy": 0.13158030807971954,
"num_tokens": 6469353.0,
"step": 3300
},
{
"entropy": 6.686995601654052,
"epoch": 0.3303843654720848,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993992548060497,
"loss": 6.5204,
"mean_token_accuracy": 0.1333637237548828,
"num_tokens": 6479610.0,
"step": 3305
},
{
"entropy": 6.653195333480835,
"epoch": 0.33088419053331336,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993966457430357,
"loss": 6.54,
"mean_token_accuracy": 0.13573898300528525,
"num_tokens": 6488866.0,
"step": 3310
},
{
"entropy": 6.707664632797242,
"epoch": 0.3313840155945419,
"grad_norm": 0.8984375,
"learning_rate": 0.0004993940310342598,
"loss": 6.4089,
"mean_token_accuracy": 0.14174082949757577,
"num_tokens": 6498396.0,
"step": 3315
},
{
"entropy": 6.6880530834198,
"epoch": 0.3318838406557705,
"grad_norm": 0.94921875,
"learning_rate": 0.0004993914106797879,
"loss": 6.462,
"mean_token_accuracy": 0.13719740733504296,
"num_tokens": 6508347.0,
"step": 3320
},
{
"entropy": 6.746197128295899,
"epoch": 0.33238366571699907,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993887846796858,
"loss": 6.55,
"mean_token_accuracy": 0.13088712841272354,
"num_tokens": 6518369.0,
"step": 3325
},
{
"entropy": 6.689602565765381,
"epoch": 0.3328834907782276,
"grad_norm": 0.93359375,
"learning_rate": 0.0004993861530340196,
"loss": 6.5129,
"mean_token_accuracy": 0.13891283497214318,
"num_tokens": 6527477.0,
"step": 3330
},
{
"entropy": 6.6981299877166744,
"epoch": 0.3333833158394562,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993835157428555,
"loss": 6.427,
"mean_token_accuracy": 0.13673900961875915,
"num_tokens": 6536271.0,
"step": 3335
},
{
"entropy": 6.709859848022461,
"epoch": 0.33388314090068477,
"grad_norm": 0.92578125,
"learning_rate": 0.0004993808728062599,
"loss": 6.5468,
"mean_token_accuracy": 0.1339237831532955,
"num_tokens": 6546295.0,
"step": 3340
},
{
"entropy": 6.751101922988892,
"epoch": 0.3343829659619133,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993782242242994,
"loss": 6.4807,
"mean_token_accuracy": 0.1411467172205448,
"num_tokens": 6557338.0,
"step": 3345
},
{
"entropy": 6.62189621925354,
"epoch": 0.3348827910231419,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993755699970404,
"loss": 6.4857,
"mean_token_accuracy": 0.13300028294324875,
"num_tokens": 6567916.0,
"step": 3350
},
{
"entropy": 6.669709730148315,
"epoch": 0.3353826160843705,
"grad_norm": 0.9375,
"learning_rate": 0.0004993729101245498,
"loss": 6.5897,
"mean_token_accuracy": 0.13144382759928702,
"num_tokens": 6577821.0,
"step": 3355
},
{
"entropy": 6.73370041847229,
"epoch": 0.33588244114559906,
"grad_norm": 0.86328125,
"learning_rate": 0.0004993702446068945,
"loss": 6.4813,
"mean_token_accuracy": 0.13519612178206444,
"num_tokens": 6587674.0,
"step": 3360
},
{
"entropy": 6.7071089267730715,
"epoch": 0.3363822662068276,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993675734441416,
"loss": 6.5485,
"mean_token_accuracy": 0.1344440333545208,
"num_tokens": 6597292.0,
"step": 3365
},
{
"entropy": 6.598604822158814,
"epoch": 0.3368820912680562,
"grad_norm": 0.9609375,
"learning_rate": 0.0004993648966363583,
"loss": 6.4849,
"mean_token_accuracy": 0.13681957721710206,
"num_tokens": 6608698.0,
"step": 3370
},
{
"entropy": 6.67167592048645,
"epoch": 0.33738191632928477,
"grad_norm": 0.98046875,
"learning_rate": 0.000499362214183612,
"loss": 6.4849,
"mean_token_accuracy": 0.1337135910987854,
"num_tokens": 6617914.0,
"step": 3375
},
{
"entropy": 6.646800756454468,
"epoch": 0.3378817413905133,
"grad_norm": 0.87109375,
"learning_rate": 0.00049935952608597,
"loss": 6.4374,
"mean_token_accuracy": 0.13931273743510247,
"num_tokens": 6627457.0,
"step": 3380
},
{
"entropy": 6.691873693466187,
"epoch": 0.3383815664517419,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993568323435002,
"loss": 6.4614,
"mean_token_accuracy": 0.13891924247145654,
"num_tokens": 6636511.0,
"step": 3385
},
{
"entropy": 6.6973389148712155,
"epoch": 0.3388813915129705,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993541329562702,
"loss": 6.4567,
"mean_token_accuracy": 0.13489941284060478,
"num_tokens": 6647028.0,
"step": 3390
},
{
"entropy": 6.704586410522461,
"epoch": 0.339381216574199,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993514279243477,
"loss": 6.5207,
"mean_token_accuracy": 0.13709425255656243,
"num_tokens": 6656407.0,
"step": 3395
},
{
"entropy": 6.684051561355591,
"epoch": 0.3398810416354276,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993487172478012,
"loss": 6.4482,
"mean_token_accuracy": 0.1370632082223892,
"num_tokens": 6665554.0,
"step": 3400
},
{
"entropy": 6.693660068511963,
"epoch": 0.3403808666966562,
"grad_norm": 0.8671875,
"learning_rate": 0.0004993460009266988,
"loss": 6.5328,
"mean_token_accuracy": 0.13296378701925277,
"num_tokens": 6675730.0,
"step": 3405
},
{
"entropy": 6.618661642074585,
"epoch": 0.3408806917578847,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993432789611086,
"loss": 6.4407,
"mean_token_accuracy": 0.1402636870741844,
"num_tokens": 6685871.0,
"step": 3410
},
{
"entropy": 6.68305401802063,
"epoch": 0.3413805168191133,
"grad_norm": 0.85546875,
"learning_rate": 0.0004993405513510992,
"loss": 6.6555,
"mean_token_accuracy": 0.12181652784347534,
"num_tokens": 6697503.0,
"step": 3415
},
{
"entropy": 6.764069128036499,
"epoch": 0.3418803418803419,
"grad_norm": 1.0,
"learning_rate": 0.0004993378180967394,
"loss": 6.4533,
"mean_token_accuracy": 0.13776906058192254,
"num_tokens": 6708022.0,
"step": 3420
},
{
"entropy": 6.661357784271241,
"epoch": 0.3423801669415705,
"grad_norm": 0.87890625,
"learning_rate": 0.0004993350791980978,
"loss": 6.5332,
"mean_token_accuracy": 0.13591760620474816,
"num_tokens": 6717510.0,
"step": 3425
},
{
"entropy": 6.690187072753906,
"epoch": 0.342879992002799,
"grad_norm": 1.140625,
"learning_rate": 0.0004993323346552433,
"loss": 6.5449,
"mean_token_accuracy": 0.1342686742544174,
"num_tokens": 6726538.0,
"step": 3430
},
{
"entropy": 6.745046710968017,
"epoch": 0.3433798170640276,
"grad_norm": 0.90625,
"learning_rate": 0.0004993295844682449,
"loss": 6.4522,
"mean_token_accuracy": 0.13962674587965013,
"num_tokens": 6736750.0,
"step": 3435
},
{
"entropy": 6.694317245483399,
"epoch": 0.3438796421252562,
"grad_norm": 0.8671875,
"learning_rate": 0.000499326828637172,
"loss": 6.4945,
"mean_token_accuracy": 0.1315005213022232,
"num_tokens": 6746366.0,
"step": 3440
},
{
"entropy": 6.620949411392212,
"epoch": 0.3443794671864847,
"grad_norm": 0.94921875,
"learning_rate": 0.0004993240671620938,
"loss": 6.506,
"mean_token_accuracy": 0.13108294382691382,
"num_tokens": 6755887.0,
"step": 3445
},
{
"entropy": 6.657417678833008,
"epoch": 0.3448792922477133,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993213000430797,
"loss": 6.5253,
"mean_token_accuracy": 0.1426558807492256,
"num_tokens": 6764434.0,
"step": 3450
},
{
"entropy": 6.746588850021363,
"epoch": 0.3453791173089419,
"grad_norm": 1.0,
"learning_rate": 0.0004993185272801995,
"loss": 6.5931,
"mean_token_accuracy": 0.12617274150252342,
"num_tokens": 6774237.0,
"step": 3455
},
{
"entropy": 6.757678699493408,
"epoch": 0.3458789423701704,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993157488735228,
"loss": 6.4517,
"mean_token_accuracy": 0.13550997152924538,
"num_tokens": 6784537.0,
"step": 3460
},
{
"entropy": 6.620314359664917,
"epoch": 0.346378767431399,
"grad_norm": 0.94921875,
"learning_rate": 0.0004993129648231197,
"loss": 6.4528,
"mean_token_accuracy": 0.13878890573978425,
"num_tokens": 6794223.0,
"step": 3465
},
{
"entropy": 6.578345108032226,
"epoch": 0.3468785924926276,
"grad_norm": 0.9140625,
"learning_rate": 0.00049931017512906,
"loss": 6.4867,
"mean_token_accuracy": 0.13346309959888458,
"num_tokens": 6803680.0,
"step": 3470
},
{
"entropy": 6.795741987228394,
"epoch": 0.3473784175538562,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993073797914141,
"loss": 6.4955,
"mean_token_accuracy": 0.12922861874103547,
"num_tokens": 6812853.0,
"step": 3475
},
{
"entropy": 6.641557931900024,
"epoch": 0.3478782426150847,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993045788102522,
"loss": 6.5486,
"mean_token_accuracy": 0.1343071274459362,
"num_tokens": 6822104.0,
"step": 3480
},
{
"entropy": 6.695757436752319,
"epoch": 0.3483780676763133,
"grad_norm": 0.953125,
"learning_rate": 0.0004993017721856449,
"loss": 6.525,
"mean_token_accuracy": 0.1304760381579399,
"num_tokens": 6832683.0,
"step": 3485
},
{
"entropy": 6.766643762588501,
"epoch": 0.3488778927375419,
"grad_norm": 0.84765625,
"learning_rate": 0.0004992989599176626,
"loss": 6.4937,
"mean_token_accuracy": 0.13055627271533013,
"num_tokens": 6843191.0,
"step": 3490
},
{
"entropy": 6.646753549575806,
"epoch": 0.3493777177987704,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992961420063763,
"loss": 6.488,
"mean_token_accuracy": 0.1350575193762779,
"num_tokens": 6852205.0,
"step": 3495
},
{
"entropy": 6.518329524993897,
"epoch": 0.349877542859999,
"grad_norm": 1.015625,
"learning_rate": 0.0004992933184518567,
"loss": 6.4188,
"mean_token_accuracy": 0.13439195230603218,
"num_tokens": 6863052.0,
"step": 3500
},
{
"entropy": 6.669634056091309,
"epoch": 0.3503773679212276,
"grad_norm": 0.98828125,
"learning_rate": 0.0004992904892541749,
"loss": 6.4826,
"mean_token_accuracy": 0.13792546093463898,
"num_tokens": 6872717.0,
"step": 3505
},
{
"entropy": 6.661230278015137,
"epoch": 0.3508771929824561,
"grad_norm": 1.015625,
"learning_rate": 0.0004992876544134022,
"loss": 6.4042,
"mean_token_accuracy": 0.14081311523914336,
"num_tokens": 6881351.0,
"step": 3510
},
{
"entropy": 6.657224369049072,
"epoch": 0.3513770180436847,
"grad_norm": 0.9609375,
"learning_rate": 0.0004992848139296098,
"loss": 6.4693,
"mean_token_accuracy": 0.1372024357318878,
"num_tokens": 6891353.0,
"step": 3515
},
{
"entropy": 6.670392465591431,
"epoch": 0.3518768431049133,
"grad_norm": 0.91015625,
"learning_rate": 0.0004992819678028692,
"loss": 6.5511,
"mean_token_accuracy": 0.12800328955054283,
"num_tokens": 6902352.0,
"step": 3520
},
{
"entropy": 6.704895782470703,
"epoch": 0.3523766681661418,
"grad_norm": 0.87890625,
"learning_rate": 0.000499279116033252,
"loss": 6.5333,
"mean_token_accuracy": 0.12561603039503097,
"num_tokens": 6911950.0,
"step": 3525
},
{
"entropy": 6.643403100967407,
"epoch": 0.3528764932273704,
"grad_norm": 0.92578125,
"learning_rate": 0.0004992762586208299,
"loss": 6.3963,
"mean_token_accuracy": 0.14974639415740967,
"num_tokens": 6921949.0,
"step": 3530
},
{
"entropy": 6.679204940795898,
"epoch": 0.353376318288599,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992733955656749,
"loss": 6.4509,
"mean_token_accuracy": 0.1364126905798912,
"num_tokens": 6932561.0,
"step": 3535
},
{
"entropy": 6.588069105148316,
"epoch": 0.3538761433498276,
"grad_norm": 1.015625,
"learning_rate": 0.0004992705268678589,
"loss": 6.4587,
"mean_token_accuracy": 0.14087284058332444,
"num_tokens": 6941663.0,
"step": 3540
},
{
"entropy": 6.66983814239502,
"epoch": 0.3543759684110561,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992676525274542,
"loss": 6.4736,
"mean_token_accuracy": 0.13467896357178688,
"num_tokens": 6952373.0,
"step": 3545
},
{
"entropy": 6.582099103927613,
"epoch": 0.3548757934722847,
"grad_norm": 0.9765625,
"learning_rate": 0.0004992647725445331,
"loss": 6.4782,
"mean_token_accuracy": 0.1386075034737587,
"num_tokens": 6961337.0,
"step": 3550
},
{
"entropy": 6.63136854171753,
"epoch": 0.3553756185335133,
"grad_norm": 0.92578125,
"learning_rate": 0.000499261886919168,
"loss": 6.4389,
"mean_token_accuracy": 0.14342176765203477,
"num_tokens": 6971649.0,
"step": 3555
},
{
"entropy": 6.664999771118164,
"epoch": 0.3558754435947418,
"grad_norm": 0.90625,
"learning_rate": 0.0004992589956514316,
"loss": 6.4629,
"mean_token_accuracy": 0.14281074404716493,
"num_tokens": 6982261.0,
"step": 3560
},
{
"entropy": 6.6441254138946535,
"epoch": 0.3563752686559704,
"grad_norm": 0.98828125,
"learning_rate": 0.0004992560987413965,
"loss": 6.4532,
"mean_token_accuracy": 0.1330704264342785,
"num_tokens": 6991324.0,
"step": 3565
},
{
"entropy": 6.598061990737915,
"epoch": 0.356875093717199,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992531961891356,
"loss": 6.4082,
"mean_token_accuracy": 0.13796095177531242,
"num_tokens": 7001396.0,
"step": 3570
},
{
"entropy": 6.738022136688232,
"epoch": 0.35737491877842753,
"grad_norm": 0.95703125,
"learning_rate": 0.0004992502879947222,
"loss": 6.418,
"mean_token_accuracy": 0.13556585609912872,
"num_tokens": 7010354.0,
"step": 3575
},
{
"entropy": 6.683406686782837,
"epoch": 0.3578747438396561,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992473741582291,
"loss": 6.5325,
"mean_token_accuracy": 0.13325171321630477,
"num_tokens": 7019835.0,
"step": 3580
},
{
"entropy": 6.611473226547242,
"epoch": 0.3583745689008847,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992444546797298,
"loss": 6.3968,
"mean_token_accuracy": 0.14161786958575248,
"num_tokens": 7030515.0,
"step": 3585
},
{
"entropy": 6.557840633392334,
"epoch": 0.35887439396211324,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992415295592977,
"loss": 6.4576,
"mean_token_accuracy": 0.1419329449534416,
"num_tokens": 7041248.0,
"step": 3590
},
{
"entropy": 6.674872970581054,
"epoch": 0.3593742190233418,
"grad_norm": 1.125,
"learning_rate": 0.0004992385987970064,
"loss": 6.4892,
"mean_token_accuracy": 0.1341181978583336,
"num_tokens": 7050578.0,
"step": 3595
},
{
"entropy": 6.7030867576599125,
"epoch": 0.3598740440845704,
"grad_norm": 0.953125,
"learning_rate": 0.0004992356623929297,
"loss": 6.5005,
"mean_token_accuracy": 0.13172903209924697,
"num_tokens": 7060270.0,
"step": 3600
},
{
"entropy": 6.659629535675049,
"epoch": 0.360373869145799,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992327203471414,
"loss": 6.5262,
"mean_token_accuracy": 0.13164970204234122,
"num_tokens": 7070259.0,
"step": 3605
},
{
"entropy": 6.609259414672851,
"epoch": 0.36087369420702753,
"grad_norm": 1.0625,
"learning_rate": 0.0004992297726597156,
"loss": 6.4417,
"mean_token_accuracy": 0.1405242405831814,
"num_tokens": 7078784.0,
"step": 3610
},
{
"entropy": 6.668364095687866,
"epoch": 0.3613735192682561,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992268193307264,
"loss": 6.4893,
"mean_token_accuracy": 0.13246863335371017,
"num_tokens": 7088632.0,
"step": 3615
},
{
"entropy": 6.66567554473877,
"epoch": 0.3618733443294847,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992238603602481,
"loss": 6.4853,
"mean_token_accuracy": 0.1330085888504982,
"num_tokens": 7097703.0,
"step": 3620
},
{
"entropy": 6.603204727172852,
"epoch": 0.36237316939071323,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992208957483552,
"loss": 6.4036,
"mean_token_accuracy": 0.14176397770643234,
"num_tokens": 7107183.0,
"step": 3625
},
{
"entropy": 6.723782539367676,
"epoch": 0.3628729944519418,
"grad_norm": 1.0703125,
"learning_rate": 0.0004992179254951223,
"loss": 6.5149,
"mean_token_accuracy": 0.13444778174161912,
"num_tokens": 7118220.0,
"step": 3630
},
{
"entropy": 6.647170352935791,
"epoch": 0.3633728195131704,
"grad_norm": 0.8828125,
"learning_rate": 0.000499214949600624,
"loss": 6.4518,
"mean_token_accuracy": 0.13443228974938393,
"num_tokens": 7128322.0,
"step": 3635
},
{
"entropy": 6.547049283981323,
"epoch": 0.36387264457439894,
"grad_norm": 0.87109375,
"learning_rate": 0.0004992119680649355,
"loss": 6.4497,
"mean_token_accuracy": 0.13709384575486183,
"num_tokens": 7137284.0,
"step": 3640
},
{
"entropy": 6.602659273147583,
"epoch": 0.3643724696356275,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992089808881313,
"loss": 6.3848,
"mean_token_accuracy": 0.14133235216140747,
"num_tokens": 7146598.0,
"step": 3645
},
{
"entropy": 6.784192419052124,
"epoch": 0.3648722946968561,
"grad_norm": 1.0,
"learning_rate": 0.0004992059880702869,
"loss": 6.5476,
"mean_token_accuracy": 0.13983522057533265,
"num_tokens": 7156816.0,
"step": 3650
},
{
"entropy": 6.682523441314697,
"epoch": 0.36537211975808465,
"grad_norm": 1.015625,
"learning_rate": 0.0004992029896114775,
"loss": 6.4747,
"mean_token_accuracy": 0.13191716745495796,
"num_tokens": 7166855.0,
"step": 3655
},
{
"entropy": 6.522796726226806,
"epoch": 0.36587194481931323,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991999855117787,
"loss": 6.3611,
"mean_token_accuracy": 0.14709938764572145,
"num_tokens": 7176477.0,
"step": 3660
},
{
"entropy": 6.633411931991577,
"epoch": 0.3663717698805418,
"grad_norm": 1.1484375,
"learning_rate": 0.0004991969757712659,
"loss": 6.4072,
"mean_token_accuracy": 0.13829985335469247,
"num_tokens": 7186173.0,
"step": 3665
},
{
"entropy": 6.5861372470855715,
"epoch": 0.3668715949417704,
"grad_norm": 0.91796875,
"learning_rate": 0.0004991939603900149,
"loss": 6.3603,
"mean_token_accuracy": 0.14236575812101365,
"num_tokens": 7195403.0,
"step": 3670
},
{
"entropy": 6.645879220962525,
"epoch": 0.36737142000299894,
"grad_norm": 0.9609375,
"learning_rate": 0.0004991909393681015,
"loss": 6.4559,
"mean_token_accuracy": 0.1353384457528591,
"num_tokens": 7204179.0,
"step": 3675
},
{
"entropy": 6.585832214355468,
"epoch": 0.3678712450642275,
"grad_norm": 0.9453125,
"learning_rate": 0.0004991879127056017,
"loss": 6.4186,
"mean_token_accuracy": 0.13716975525021552,
"num_tokens": 7214159.0,
"step": 3680
},
{
"entropy": 6.606347703933716,
"epoch": 0.3683710701254561,
"grad_norm": 0.87109375,
"learning_rate": 0.0004991848804025918,
"loss": 6.3778,
"mean_token_accuracy": 0.1415353387594223,
"num_tokens": 7224410.0,
"step": 3685
},
{
"entropy": 6.625489568710327,
"epoch": 0.36887089518668464,
"grad_norm": 0.86328125,
"learning_rate": 0.000499181842459148,
"loss": 6.4366,
"mean_token_accuracy": 0.1429909348487854,
"num_tokens": 7233740.0,
"step": 3690
},
{
"entropy": 6.654446172714233,
"epoch": 0.36937072024791323,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991787988753467,
"loss": 6.4132,
"mean_token_accuracy": 0.13803667649626733,
"num_tokens": 7242834.0,
"step": 3695
},
{
"entropy": 6.568963098526001,
"epoch": 0.3698705453091418,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991757496512645,
"loss": 6.369,
"mean_token_accuracy": 0.14709216505289077,
"num_tokens": 7252098.0,
"step": 3700
},
{
"entropy": 6.699278259277344,
"epoch": 0.37037037037037035,
"grad_norm": 0.953125,
"learning_rate": 0.0004991726947869782,
"loss": 6.5498,
"mean_token_accuracy": 0.12648288458585738,
"num_tokens": 7261396.0,
"step": 3705
},
{
"entropy": 6.666980504989624,
"epoch": 0.37087019543159894,
"grad_norm": 0.8984375,
"learning_rate": 0.0004991696342825645,
"loss": 6.476,
"mean_token_accuracy": 0.13404362574219703,
"num_tokens": 7271138.0,
"step": 3710
},
{
"entropy": 6.602921342849731,
"epoch": 0.3713700204928275,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991665681381005,
"loss": 6.3853,
"mean_token_accuracy": 0.1445363648235798,
"num_tokens": 7280164.0,
"step": 3715
},
{
"entropy": 6.648625707626342,
"epoch": 0.37186984555405606,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991634963536633,
"loss": 6.4619,
"mean_token_accuracy": 0.13741569444537163,
"num_tokens": 7288830.0,
"step": 3720
},
{
"entropy": 6.619314241409302,
"epoch": 0.37236967061528464,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991604189293303,
"loss": 6.4828,
"mean_token_accuracy": 0.1355450950562954,
"num_tokens": 7298955.0,
"step": 3725
},
{
"entropy": 6.656387805938721,
"epoch": 0.37286949567651323,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991573358651787,
"loss": 6.3975,
"mean_token_accuracy": 0.13694959357380868,
"num_tokens": 7309829.0,
"step": 3730
},
{
"entropy": 6.646700286865235,
"epoch": 0.3733693207377418,
"grad_norm": 0.890625,
"learning_rate": 0.0004991542471612864,
"loss": 6.4689,
"mean_token_accuracy": 0.1364390507340431,
"num_tokens": 7319654.0,
"step": 3735
},
{
"entropy": 6.689327192306519,
"epoch": 0.37386914579897035,
"grad_norm": 0.89453125,
"learning_rate": 0.0004991511528177307,
"loss": 6.4565,
"mean_token_accuracy": 0.13406700342893602,
"num_tokens": 7330415.0,
"step": 3740
},
{
"entropy": 6.686400318145752,
"epoch": 0.37436897086019894,
"grad_norm": 1.03125,
"learning_rate": 0.0004991480528345898,
"loss": 6.487,
"mean_token_accuracy": 0.131680615991354,
"num_tokens": 7339511.0,
"step": 3745
},
{
"entropy": 6.741026449203491,
"epoch": 0.3748687959214275,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991449472119416,
"loss": 6.4707,
"mean_token_accuracy": 0.12732115983963013,
"num_tokens": 7348488.0,
"step": 3750
},
{
"entropy": 6.6078413963317875,
"epoch": 0.37536862098265605,
"grad_norm": 0.875,
"learning_rate": 0.0004991418359498642,
"loss": 6.4016,
"mean_token_accuracy": 0.13857970386743546,
"num_tokens": 7357804.0,
"step": 3755
},
{
"entropy": 6.611809253692627,
"epoch": 0.37586844604388464,
"grad_norm": 0.97265625,
"learning_rate": 0.0004991387190484358,
"loss": 6.4852,
"mean_token_accuracy": 0.13374910578131677,
"num_tokens": 7367644.0,
"step": 3760
},
{
"entropy": 6.676985216140747,
"epoch": 0.37636827110511323,
"grad_norm": 1.1171875,
"learning_rate": 0.000499135596507735,
"loss": 6.334,
"mean_token_accuracy": 0.13773026019334794,
"num_tokens": 7377357.0,
"step": 3765
},
{
"entropy": 6.600540542602539,
"epoch": 0.37686809616634176,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991324683278402,
"loss": 6.5151,
"mean_token_accuracy": 0.1318119928240776,
"num_tokens": 7387306.0,
"step": 3770
},
{
"entropy": 6.620249891281128,
"epoch": 0.37736792122757035,
"grad_norm": 0.890625,
"learning_rate": 0.0004991293345088301,
"loss": 6.4405,
"mean_token_accuracy": 0.13481418415904045,
"num_tokens": 7397472.0,
"step": 3775
},
{
"entropy": 6.617590999603271,
"epoch": 0.37786774628879893,
"grad_norm": 0.85546875,
"learning_rate": 0.0004991261950507837,
"loss": 6.493,
"mean_token_accuracy": 0.13492076247930526,
"num_tokens": 7408786.0,
"step": 3780
},
{
"entropy": 6.633843469619751,
"epoch": 0.37836757135002747,
"grad_norm": 0.9140625,
"learning_rate": 0.00049912304995378,
"loss": 6.4569,
"mean_token_accuracy": 0.13087058067321777,
"num_tokens": 7417857.0,
"step": 3785
},
{
"entropy": 6.580232810974121,
"epoch": 0.37886739641125605,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991198992178979,
"loss": 6.4244,
"mean_token_accuracy": 0.13517454341053964,
"num_tokens": 7427007.0,
"step": 3790
},
{
"entropy": 6.663159704208374,
"epoch": 0.37936722147248464,
"grad_norm": 0.875,
"learning_rate": 0.0004991167428432168,
"loss": 6.3556,
"mean_token_accuracy": 0.1432080551981926,
"num_tokens": 7437090.0,
"step": 3795
},
{
"entropy": 6.653858995437622,
"epoch": 0.3798670465337132,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991135808298162,
"loss": 6.3963,
"mean_token_accuracy": 0.14008912220597267,
"num_tokens": 7446249.0,
"step": 3800
},
{
"entropy": 6.626196384429932,
"epoch": 0.38036687159494176,
"grad_norm": 0.984375,
"learning_rate": 0.0004991104131777755,
"loss": 6.4173,
"mean_token_accuracy": 0.13473987355828285,
"num_tokens": 7455838.0,
"step": 3805
},
{
"entropy": 6.583133172988892,
"epoch": 0.38086669665617034,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991072398871746,
"loss": 6.4987,
"mean_token_accuracy": 0.1326663985848427,
"num_tokens": 7466057.0,
"step": 3810
},
{
"entropy": 6.670406246185303,
"epoch": 0.38136652171739893,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991040609580932,
"loss": 6.4548,
"mean_token_accuracy": 0.13467531204223632,
"num_tokens": 7475174.0,
"step": 3815
},
{
"entropy": 6.622626399993896,
"epoch": 0.38186634677862746,
"grad_norm": 0.91796875,
"learning_rate": 0.0004991008763906112,
"loss": 6.3807,
"mean_token_accuracy": 0.14110917896032332,
"num_tokens": 7484635.0,
"step": 3820
},
{
"entropy": 6.591571807861328,
"epoch": 0.38236617183985605,
"grad_norm": 0.828125,
"learning_rate": 0.0004990976861848087,
"loss": 6.4522,
"mean_token_accuracy": 0.13135599717497826,
"num_tokens": 7496428.0,
"step": 3825
},
{
"entropy": 6.641083669662476,
"epoch": 0.38286599690108464,
"grad_norm": 0.99609375,
"learning_rate": 0.0004990944903407663,
"loss": 6.4786,
"mean_token_accuracy": 0.13468215465545655,
"num_tokens": 7506581.0,
"step": 3830
},
{
"entropy": 6.631896591186523,
"epoch": 0.38336582196231317,
"grad_norm": 0.90625,
"learning_rate": 0.0004990912888585641,
"loss": 6.35,
"mean_token_accuracy": 0.14103600159287452,
"num_tokens": 7515809.0,
"step": 3835
},
{
"entropy": 6.509975290298462,
"epoch": 0.38386564702354176,
"grad_norm": 0.91796875,
"learning_rate": 0.0004990880817382828,
"loss": 6.3881,
"mean_token_accuracy": 0.13978929817676544,
"num_tokens": 7525800.0,
"step": 3840
},
{
"entropy": 6.590206432342529,
"epoch": 0.38436547208477034,
"grad_norm": 0.8203125,
"learning_rate": 0.000499084868980003,
"loss": 6.4761,
"mean_token_accuracy": 0.13252102136611937,
"num_tokens": 7536903.0,
"step": 3845
},
{
"entropy": 6.62482647895813,
"epoch": 0.3848652971459989,
"grad_norm": 0.91796875,
"learning_rate": 0.0004990816505838054,
"loss": 6.4483,
"mean_token_accuracy": 0.13781202882528304,
"num_tokens": 7546004.0,
"step": 3850
},
{
"entropy": 6.642586517333984,
"epoch": 0.38536512220722746,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990784265497713,
"loss": 6.5578,
"mean_token_accuracy": 0.13435211181640624,
"num_tokens": 7555971.0,
"step": 3855
},
{
"entropy": 6.7109456062316895,
"epoch": 0.38586494726845605,
"grad_norm": 0.99609375,
"learning_rate": 0.0004990751968779817,
"loss": 6.4538,
"mean_token_accuracy": 0.13786686584353447,
"num_tokens": 7564569.0,
"step": 3860
},
{
"entropy": 6.602899742126465,
"epoch": 0.38636477232968464,
"grad_norm": 0.99609375,
"learning_rate": 0.0004990719615685176,
"loss": 6.4608,
"mean_token_accuracy": 0.13531645983457566,
"num_tokens": 7574134.0,
"step": 3865
},
{
"entropy": 6.598666667938232,
"epoch": 0.38686459739091317,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990687206214608,
"loss": 6.3216,
"mean_token_accuracy": 0.14182653799653053,
"num_tokens": 7583631.0,
"step": 3870
},
{
"entropy": 6.506135177612305,
"epoch": 0.38736442245214175,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990654740368925,
"loss": 6.294,
"mean_token_accuracy": 0.1420180991292,
"num_tokens": 7593281.0,
"step": 3875
},
{
"entropy": 6.484540557861328,
"epoch": 0.38786424751337034,
"grad_norm": 0.8984375,
"learning_rate": 0.0004990622218148947,
"loss": 6.4371,
"mean_token_accuracy": 0.1398145854473114,
"num_tokens": 7603735.0,
"step": 3880
},
{
"entropy": 6.647881507873535,
"epoch": 0.3883640725745989,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990589639555489,
"loss": 6.399,
"mean_token_accuracy": 0.13606541529297828,
"num_tokens": 7613126.0,
"step": 3885
},
{
"entropy": 6.662182903289795,
"epoch": 0.38886389763582746,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990557004589373,
"loss": 6.4325,
"mean_token_accuracy": 0.13136639297008515,
"num_tokens": 7623191.0,
"step": 3890
},
{
"entropy": 6.603696489334107,
"epoch": 0.38936372269705605,
"grad_norm": 1.03125,
"learning_rate": 0.0004990524313251419,
"loss": 6.4133,
"mean_token_accuracy": 0.14168039411306382,
"num_tokens": 7634118.0,
"step": 3895
},
{
"entropy": 6.528379964828491,
"epoch": 0.3898635477582846,
"grad_norm": 0.94140625,
"learning_rate": 0.0004990491565542449,
"loss": 6.288,
"mean_token_accuracy": 0.15081499218940736,
"num_tokens": 7643228.0,
"step": 3900
},
{
"entropy": 6.561570167541504,
"epoch": 0.39036337281951317,
"grad_norm": 0.984375,
"learning_rate": 0.0004990458761463289,
"loss": 6.4042,
"mean_token_accuracy": 0.13431327641010285,
"num_tokens": 7652578.0,
"step": 3905
},
{
"entropy": 6.5966715812683105,
"epoch": 0.39086319788074175,
"grad_norm": 0.9765625,
"learning_rate": 0.0004990425901014763,
"loss": 6.3664,
"mean_token_accuracy": 0.13791628554463387,
"num_tokens": 7661617.0,
"step": 3910
},
{
"entropy": 6.582742547988891,
"epoch": 0.3913630229419703,
"grad_norm": 0.84765625,
"learning_rate": 0.0004990392984197697,
"loss": 6.3775,
"mean_token_accuracy": 0.14419438168406487,
"num_tokens": 7670864.0,
"step": 3915
},
{
"entropy": 6.57593560218811,
"epoch": 0.39186284800319887,
"grad_norm": 0.953125,
"learning_rate": 0.0004990360011012921,
"loss": 6.4378,
"mean_token_accuracy": 0.1298800878226757,
"num_tokens": 7681528.0,
"step": 3920
},
{
"entropy": 6.612252187728882,
"epoch": 0.39236267306442746,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990326981461264,
"loss": 6.346,
"mean_token_accuracy": 0.14040571600198745,
"num_tokens": 7690398.0,
"step": 3925
},
{
"entropy": 6.536719942092896,
"epoch": 0.39286249812565605,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990293895543555,
"loss": 6.3832,
"mean_token_accuracy": 0.1446215718984604,
"num_tokens": 7699843.0,
"step": 3930
},
{
"entropy": 6.584111309051513,
"epoch": 0.3933623231868846,
"grad_norm": 0.96484375,
"learning_rate": 0.000499026075326063,
"loss": 6.4227,
"mean_token_accuracy": 0.1415751188993454,
"num_tokens": 7710181.0,
"step": 3935
},
{
"entropy": 6.592550754547119,
"epoch": 0.39386214824811316,
"grad_norm": 0.96875,
"learning_rate": 0.000499022755461332,
"loss": 6.3677,
"mean_token_accuracy": 0.1341855473816395,
"num_tokens": 7719148.0,
"step": 3940
},
{
"entropy": 6.473948907852173,
"epoch": 0.39436197330934175,
"grad_norm": 0.93359375,
"learning_rate": 0.000499019429960246,
"loss": 6.2491,
"mean_token_accuracy": 0.14969748854637147,
"num_tokens": 7728551.0,
"step": 3945
},
{
"entropy": 6.605637121200561,
"epoch": 0.3948617983705703,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990160988228888,
"loss": 6.451,
"mean_token_accuracy": 0.1340394489467144,
"num_tokens": 7739083.0,
"step": 3950
},
{
"entropy": 6.63070068359375,
"epoch": 0.39536162343179887,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990127620493444,
"loss": 6.434,
"mean_token_accuracy": 0.13626187443733215,
"num_tokens": 7749684.0,
"step": 3955
},
{
"entropy": 6.54850549697876,
"epoch": 0.39586144849302746,
"grad_norm": 0.98046875,
"learning_rate": 0.0004990094196396964,
"loss": 6.3054,
"mean_token_accuracy": 0.1447399452328682,
"num_tokens": 7759281.0,
"step": 3960
},
{
"entropy": 6.550680875778198,
"epoch": 0.396361273554256,
"grad_norm": 0.9765625,
"learning_rate": 0.000499006071594029,
"loss": 6.3874,
"mean_token_accuracy": 0.1380893476307392,
"num_tokens": 7768612.0,
"step": 3965
},
{
"entropy": 6.532160091400146,
"epoch": 0.3968610986154846,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990027179124266,
"loss": 6.3271,
"mean_token_accuracy": 0.14160313084721565,
"num_tokens": 7778333.0,
"step": 3970
},
{
"entropy": 6.684771490097046,
"epoch": 0.39736092367671316,
"grad_norm": 0.91796875,
"learning_rate": 0.0004989993585949733,
"loss": 6.4557,
"mean_token_accuracy": 0.13541282787919046,
"num_tokens": 7787547.0,
"step": 3975
},
{
"entropy": 6.553624486923217,
"epoch": 0.3978607487379417,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989959936417541,
"loss": 6.3988,
"mean_token_accuracy": 0.13911321386694908,
"num_tokens": 7798407.0,
"step": 3980
},
{
"entropy": 6.549784469604492,
"epoch": 0.3983605737991703,
"grad_norm": 0.83203125,
"learning_rate": 0.000498992623052853,
"loss": 6.3266,
"mean_token_accuracy": 0.14800373762845992,
"num_tokens": 7807879.0,
"step": 3985
},
{
"entropy": 6.575151538848877,
"epoch": 0.39886039886039887,
"grad_norm": 1.0234375,
"learning_rate": 0.0004989892468283553,
"loss": 6.3033,
"mean_token_accuracy": 0.14408419281244278,
"num_tokens": 7817210.0,
"step": 3990
},
{
"entropy": 6.662887382507324,
"epoch": 0.39936022392162746,
"grad_norm": 1.25,
"learning_rate": 0.0004989858649683457,
"loss": 6.5548,
"mean_token_accuracy": 0.12639033570885658,
"num_tokens": 7827845.0,
"step": 3995
},
{
"entropy": 6.706279182434082,
"epoch": 0.399860048982856,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989824774729094,
"loss": 6.4104,
"mean_token_accuracy": 0.14265441671013832,
"num_tokens": 7836994.0,
"step": 4000
},
{
"entropy": 6.557466077804565,
"epoch": 0.4003598740440846,
"grad_norm": 1.0,
"learning_rate": 0.0004989790843421317,
"loss": 6.4394,
"mean_token_accuracy": 0.13453166633844377,
"num_tokens": 7846733.0,
"step": 4005
},
{
"entropy": 6.472391653060913,
"epoch": 0.40085969910531316,
"grad_norm": 1.1953125,
"learning_rate": 0.0004989756855760979,
"loss": 6.3679,
"mean_token_accuracy": 0.13990991711616516,
"num_tokens": 7856393.0,
"step": 4010
},
{
"entropy": 6.608847188949585,
"epoch": 0.4013595241665417,
"grad_norm": 0.8671875,
"learning_rate": 0.0004989722811748934,
"loss": 6.4517,
"mean_token_accuracy": 0.13580357208847998,
"num_tokens": 7866394.0,
"step": 4015
},
{
"entropy": 6.6143959045410154,
"epoch": 0.4018593492277703,
"grad_norm": 0.9140625,
"learning_rate": 0.0004989688711386039,
"loss": 6.4439,
"mean_token_accuracy": 0.134534602612257,
"num_tokens": 7877084.0,
"step": 4020
},
{
"entropy": 6.606017923355102,
"epoch": 0.40235917428899887,
"grad_norm": 0.953125,
"learning_rate": 0.0004989654554673152,
"loss": 6.4577,
"mean_token_accuracy": 0.134698349237442,
"num_tokens": 7887289.0,
"step": 4025
},
{
"entropy": 6.532919216156006,
"epoch": 0.4028589993502274,
"grad_norm": 0.859375,
"learning_rate": 0.0004989620341611133,
"loss": 6.4101,
"mean_token_accuracy": 0.13764130547642708,
"num_tokens": 7897315.0,
"step": 4030
},
{
"entropy": 6.674659585952758,
"epoch": 0.403358824411456,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989586072200843,
"loss": 6.5104,
"mean_token_accuracy": 0.13267643377184868,
"num_tokens": 7905913.0,
"step": 4035
},
{
"entropy": 6.574588871002197,
"epoch": 0.40385864947268457,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989551746443143,
"loss": 6.308,
"mean_token_accuracy": 0.14384820610284804,
"num_tokens": 7914879.0,
"step": 4040
},
{
"entropy": 6.553117942810059,
"epoch": 0.4043584745339131,
"grad_norm": 0.97265625,
"learning_rate": 0.0004989517364338899,
"loss": 6.4574,
"mean_token_accuracy": 0.13905400186777114,
"num_tokens": 7925163.0,
"step": 4045
},
{
"entropy": 6.620127153396607,
"epoch": 0.4048582995951417,
"grad_norm": 0.91015625,
"learning_rate": 0.0004989482925888974,
"loss": 6.3524,
"mean_token_accuracy": 0.14474843591451644,
"num_tokens": 7934380.0,
"step": 4050
},
{
"entropy": 6.6293004035949705,
"epoch": 0.4053581246563703,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989448431094233,
"loss": 6.4243,
"mean_token_accuracy": 0.14015459418296813,
"num_tokens": 7944600.0,
"step": 4055
},
{
"entropy": 6.489453935623169,
"epoch": 0.40585794971759886,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989413879955548,
"loss": 6.3354,
"mean_token_accuracy": 0.14372531324625015,
"num_tokens": 7953545.0,
"step": 4060
},
{
"entropy": 6.560867500305176,
"epoch": 0.4063577747788274,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989379272473785,
"loss": 6.4095,
"mean_token_accuracy": 0.13583200797438622,
"num_tokens": 7962931.0,
"step": 4065
},
{
"entropy": 6.608096885681152,
"epoch": 0.406857599840056,
"grad_norm": 1.40625,
"learning_rate": 0.0004989344608649815,
"loss": 6.3699,
"mean_token_accuracy": 0.1427628867328167,
"num_tokens": 7973090.0,
"step": 4070
},
{
"entropy": 6.61473650932312,
"epoch": 0.40735742490128457,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989309888484511,
"loss": 6.4935,
"mean_token_accuracy": 0.13091094121336938,
"num_tokens": 7982208.0,
"step": 4075
},
{
"entropy": 6.5768373012542725,
"epoch": 0.4078572499625131,
"grad_norm": 0.953125,
"learning_rate": 0.0004989275111978748,
"loss": 6.4575,
"mean_token_accuracy": 0.13970190584659575,
"num_tokens": 7993389.0,
"step": 4080
},
{
"entropy": 6.586777591705323,
"epoch": 0.4083570750237417,
"grad_norm": 0.859375,
"learning_rate": 0.0004989240279133398,
"loss": 6.3408,
"mean_token_accuracy": 0.1417361781001091,
"num_tokens": 8002749.0,
"step": 4085
},
{
"entropy": 6.559087896347046,
"epoch": 0.4088569000849703,
"grad_norm": 0.85546875,
"learning_rate": 0.0004989205389949339,
"loss": 6.3169,
"mean_token_accuracy": 0.14500403106212617,
"num_tokens": 8011552.0,
"step": 4090
},
{
"entropy": 6.474533319473267,
"epoch": 0.4093567251461988,
"grad_norm": 0.984375,
"learning_rate": 0.0004989170444427448,
"loss": 6.3374,
"mean_token_accuracy": 0.14319820553064347,
"num_tokens": 8020622.0,
"step": 4095
},
{
"entropy": 6.530226612091065,
"epoch": 0.4098565502074274,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989135442568606,
"loss": 6.2953,
"mean_token_accuracy": 0.14594606310129166,
"num_tokens": 8030356.0,
"step": 4100
},
{
"entropy": 6.526857328414917,
"epoch": 0.410356375268656,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989100384373693,
"loss": 6.4168,
"mean_token_accuracy": 0.13288672342896463,
"num_tokens": 8039903.0,
"step": 4105
},
{
"entropy": 6.5449991703033445,
"epoch": 0.4108562003298845,
"grad_norm": 0.953125,
"learning_rate": 0.0004989065269843588,
"loss": 6.3355,
"mean_token_accuracy": 0.1405773088335991,
"num_tokens": 8050488.0,
"step": 4110
},
{
"entropy": 6.554579830169677,
"epoch": 0.4113560253911131,
"grad_norm": 0.9453125,
"learning_rate": 0.000498903009897918,
"loss": 6.368,
"mean_token_accuracy": 0.13881764709949493,
"num_tokens": 8060179.0,
"step": 4115
},
{
"entropy": 6.54913649559021,
"epoch": 0.4118558504523417,
"grad_norm": 0.8828125,
"learning_rate": 0.0004988994871781349,
"loss": 6.4159,
"mean_token_accuracy": 0.13827834725379945,
"num_tokens": 8070915.0,
"step": 4120
},
{
"entropy": 6.544614601135254,
"epoch": 0.4123556755135703,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988959588250983,
"loss": 6.3528,
"mean_token_accuracy": 0.14057195484638213,
"num_tokens": 8080203.0,
"step": 4125
},
{
"entropy": 6.590111541748047,
"epoch": 0.4128555005747988,
"grad_norm": 0.921875,
"learning_rate": 0.000498892424838897,
"loss": 6.3763,
"mean_token_accuracy": 0.1408051535487175,
"num_tokens": 8090552.0,
"step": 4130
},
{
"entropy": 6.577900981903076,
"epoch": 0.4133553256360274,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988888852196199,
"loss": 6.3411,
"mean_token_accuracy": 0.14606880396604538,
"num_tokens": 8099947.0,
"step": 4135
},
{
"entropy": 6.520494270324707,
"epoch": 0.413855150697256,
"grad_norm": 0.8359375,
"learning_rate": 0.0004988853399673562,
"loss": 6.4335,
"mean_token_accuracy": 0.13198177739977837,
"num_tokens": 8110502.0,
"step": 4140
},
{
"entropy": 6.601160717010498,
"epoch": 0.4143549757584845,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988817890821948,
"loss": 6.4419,
"mean_token_accuracy": 0.13627717047929763,
"num_tokens": 8118762.0,
"step": 4145
},
{
"entropy": 6.629729175567627,
"epoch": 0.4148548008197131,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988782325642252,
"loss": 6.359,
"mean_token_accuracy": 0.13726264461874962,
"num_tokens": 8129072.0,
"step": 4150
},
{
"entropy": 6.5286548137664795,
"epoch": 0.4153546258809417,
"grad_norm": 0.9765625,
"learning_rate": 0.000498874670413537,
"loss": 6.3832,
"mean_token_accuracy": 0.1353824555873871,
"num_tokens": 8138504.0,
"step": 4155
},
{
"entropy": 6.600216102600098,
"epoch": 0.4158544509421702,
"grad_norm": 0.890625,
"learning_rate": 0.0004988711026302197,
"loss": 6.4528,
"mean_token_accuracy": 0.13507835268974305,
"num_tokens": 8148534.0,
"step": 4160
},
{
"entropy": 6.6404993534088135,
"epoch": 0.4163542760033988,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988675292143629,
"loss": 6.5159,
"mean_token_accuracy": 0.13052211925387383,
"num_tokens": 8160272.0,
"step": 4165
},
{
"entropy": 6.600973844528198,
"epoch": 0.4168541010646274,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988639501660568,
"loss": 6.3619,
"mean_token_accuracy": 0.1418714180588722,
"num_tokens": 8170053.0,
"step": 4170
},
{
"entropy": 6.596292352676391,
"epoch": 0.4173539261258559,
"grad_norm": 0.984375,
"learning_rate": 0.0004988603654853913,
"loss": 6.393,
"mean_token_accuracy": 0.13247931376099586,
"num_tokens": 8179425.0,
"step": 4175
},
{
"entropy": 6.527098798751831,
"epoch": 0.4178537511870845,
"grad_norm": 0.86328125,
"learning_rate": 0.0004988567751724565,
"loss": 6.4046,
"mean_token_accuracy": 0.13386574983596802,
"num_tokens": 8189846.0,
"step": 4180
},
{
"entropy": 6.578232669830323,
"epoch": 0.4183535762483131,
"grad_norm": 0.96484375,
"learning_rate": 0.000498853179227343,
"loss": 6.4186,
"mean_token_accuracy": 0.13852027505636216,
"num_tokens": 8198596.0,
"step": 4185
},
{
"entropy": 6.62082052230835,
"epoch": 0.4188534013095417,
"grad_norm": 0.99609375,
"learning_rate": 0.000498849577650141,
"loss": 6.3176,
"mean_token_accuracy": 0.1402053378522396,
"num_tokens": 8206849.0,
"step": 4190
},
{
"entropy": 6.564785194396973,
"epoch": 0.4193532263707702,
"grad_norm": 0.953125,
"learning_rate": 0.0004988459704409413,
"loss": 6.3569,
"mean_token_accuracy": 0.1440422423183918,
"num_tokens": 8216260.0,
"step": 4195
},
{
"entropy": 6.584865760803223,
"epoch": 0.4198530514319988,
"grad_norm": 0.99609375,
"learning_rate": 0.0004988423575998345,
"loss": 6.3703,
"mean_token_accuracy": 0.14182392060756682,
"num_tokens": 8224650.0,
"step": 4200
},
{
"entropy": 6.464220762252808,
"epoch": 0.4203528764932274,
"grad_norm": 1.5859375,
"learning_rate": 0.0004988387391269117,
"loss": 6.2355,
"mean_token_accuracy": 0.14646336287260056,
"num_tokens": 8235276.0,
"step": 4205
},
{
"entropy": 6.60213942527771,
"epoch": 0.4208527015544559,
"grad_norm": 0.984375,
"learning_rate": 0.0004988351150222638,
"loss": 6.3147,
"mean_token_accuracy": 0.14053323045372962,
"num_tokens": 8244687.0,
"step": 4210
},
{
"entropy": 6.560609292984009,
"epoch": 0.4213525266156845,
"grad_norm": 0.8515625,
"learning_rate": 0.0004988314852859819,
"loss": 6.441,
"mean_token_accuracy": 0.13898905217647553,
"num_tokens": 8255091.0,
"step": 4215
},
{
"entropy": 6.518747138977051,
"epoch": 0.4218523516769131,
"grad_norm": 1.4375,
"learning_rate": 0.0004988278499181575,
"loss": 6.2472,
"mean_token_accuracy": 0.15841120705008507,
"num_tokens": 8265172.0,
"step": 4220
},
{
"entropy": 6.577982473373413,
"epoch": 0.4223521767381416,
"grad_norm": 1.015625,
"learning_rate": 0.000498824208918882,
"loss": 6.3486,
"mean_token_accuracy": 0.13996930196881294,
"num_tokens": 8275625.0,
"step": 4225
},
{
"entropy": 6.514586544036865,
"epoch": 0.4228520017993702,
"grad_norm": 1.0,
"learning_rate": 0.000498820562288247,
"loss": 6.3355,
"mean_token_accuracy": 0.13544740602374078,
"num_tokens": 8284343.0,
"step": 4230
},
{
"entropy": 6.574178791046142,
"epoch": 0.4233518268605988,
"grad_norm": 0.97265625,
"learning_rate": 0.0004988169100263442,
"loss": 6.4196,
"mean_token_accuracy": 0.13436390906572343,
"num_tokens": 8293430.0,
"step": 4235
},
{
"entropy": 6.509334516525269,
"epoch": 0.4238516519218274,
"grad_norm": 0.859375,
"learning_rate": 0.0004988132521332656,
"loss": 6.3945,
"mean_token_accuracy": 0.13619714975357056,
"num_tokens": 8302925.0,
"step": 4240
},
{
"entropy": 6.548348474502563,
"epoch": 0.4243514769830559,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988095886091031,
"loss": 6.3395,
"mean_token_accuracy": 0.1377752937376499,
"num_tokens": 8312134.0,
"step": 4245
},
{
"entropy": 6.588669157028198,
"epoch": 0.4248513020442845,
"grad_norm": 0.92578125,
"learning_rate": 0.0004988059194539489,
"loss": 6.3445,
"mean_token_accuracy": 0.13960151746869087,
"num_tokens": 8320942.0,
"step": 4250
},
{
"entropy": 6.555509424209594,
"epoch": 0.4253511271055131,
"grad_norm": 0.859375,
"learning_rate": 0.0004988022446678954,
"loss": 6.3726,
"mean_token_accuracy": 0.13994579166173934,
"num_tokens": 8331898.0,
"step": 4255
},
{
"entropy": 6.532318449020385,
"epoch": 0.4258509521667416,
"grad_norm": 0.9140625,
"learning_rate": 0.0004987985642510352,
"loss": 6.3511,
"mean_token_accuracy": 0.1447647415101528,
"num_tokens": 8341728.0,
"step": 4260
},
{
"entropy": 6.56305890083313,
"epoch": 0.4263507772279702,
"grad_norm": 0.890625,
"learning_rate": 0.0004987948782034605,
"loss": 6.3038,
"mean_token_accuracy": 0.14301208257675171,
"num_tokens": 8351611.0,
"step": 4265
},
{
"entropy": 6.5076521873474125,
"epoch": 0.4268506022891988,
"grad_norm": 1.0078125,
"learning_rate": 0.0004987911865252644,
"loss": 6.3281,
"mean_token_accuracy": 0.14357817620038987,
"num_tokens": 8360528.0,
"step": 4270
},
{
"entropy": 6.506665182113648,
"epoch": 0.42735042735042733,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987874892165395,
"loss": 6.3834,
"mean_token_accuracy": 0.14152655079960824,
"num_tokens": 8369898.0,
"step": 4275
},
{
"entropy": 6.586957693099976,
"epoch": 0.4278502524116559,
"grad_norm": 0.91796875,
"learning_rate": 0.000498783786277379,
"loss": 6.3355,
"mean_token_accuracy": 0.14503186643123628,
"num_tokens": 8379928.0,
"step": 4280
},
{
"entropy": 6.557859277725219,
"epoch": 0.4283500774728845,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987800777078759,
"loss": 6.3643,
"mean_token_accuracy": 0.1393843911588192,
"num_tokens": 8389208.0,
"step": 4285
},
{
"entropy": 6.531264352798462,
"epoch": 0.42884990253411304,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987763635081238,
"loss": 6.3598,
"mean_token_accuracy": 0.14071909934282303,
"num_tokens": 8398364.0,
"step": 4290
},
{
"entropy": 6.5567357540130615,
"epoch": 0.4293497275953416,
"grad_norm": 0.8828125,
"learning_rate": 0.000498772643678216,
"loss": 6.366,
"mean_token_accuracy": 0.13652568832039833,
"num_tokens": 8408756.0,
"step": 4295
},
{
"entropy": 6.554677152633667,
"epoch": 0.4298495526565702,
"grad_norm": 0.97265625,
"learning_rate": 0.0004987689182182459,
"loss": 6.3251,
"mean_token_accuracy": 0.14305730685591697,
"num_tokens": 8418318.0,
"step": 4300
},
{
"entropy": 6.496006107330322,
"epoch": 0.4303493777177988,
"grad_norm": 0.90234375,
"learning_rate": 0.0004987651871283075,
"loss": 6.358,
"mean_token_accuracy": 0.13657767921686173,
"num_tokens": 8427709.0,
"step": 4305
},
{
"entropy": 6.571256923675537,
"epoch": 0.43084920277902733,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987614504084946,
"loss": 6.374,
"mean_token_accuracy": 0.1362711362540722,
"num_tokens": 8437848.0,
"step": 4310
},
{
"entropy": 6.557821655273438,
"epoch": 0.4313490278402559,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987577080589012,
"loss": 6.3204,
"mean_token_accuracy": 0.13983287513256074,
"num_tokens": 8447466.0,
"step": 4315
},
{
"entropy": 6.476554441452026,
"epoch": 0.4318488529014845,
"grad_norm": 0.99609375,
"learning_rate": 0.0004987539600796213,
"loss": 6.3137,
"mean_token_accuracy": 0.14809551909565927,
"num_tokens": 8456651.0,
"step": 4320
},
{
"entropy": 6.576989698410034,
"epoch": 0.43234867796271303,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987502064707494,
"loss": 6.3649,
"mean_token_accuracy": 0.1401773512363434,
"num_tokens": 8466421.0,
"step": 4325
},
{
"entropy": 6.549679517745972,
"epoch": 0.4328485030239416,
"grad_norm": 0.87890625,
"learning_rate": 0.0004987464472323799,
"loss": 6.3053,
"mean_token_accuracy": 0.1403607189655304,
"num_tokens": 8476419.0,
"step": 4330
},
{
"entropy": 6.514598703384399,
"epoch": 0.4333483280851702,
"grad_norm": 0.86328125,
"learning_rate": 0.0004987426823646074,
"loss": 6.3337,
"mean_token_accuracy": 0.13671851754188538,
"num_tokens": 8486652.0,
"step": 4335
},
{
"entropy": 6.48084568977356,
"epoch": 0.43384815314639874,
"grad_norm": 0.9921875,
"learning_rate": 0.0004987389118675266,
"loss": 6.2866,
"mean_token_accuracy": 0.14497914016246796,
"num_tokens": 8495965.0,
"step": 4340
},
{
"entropy": 6.615687036514283,
"epoch": 0.4343479782076273,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987351357412322,
"loss": 6.3703,
"mean_token_accuracy": 0.1346478521823883,
"num_tokens": 8506018.0,
"step": 4345
},
{
"entropy": 6.516758775711059,
"epoch": 0.4348478032688559,
"grad_norm": 0.984375,
"learning_rate": 0.0004987313539858196,
"loss": 6.3686,
"mean_token_accuracy": 0.14214786589145662,
"num_tokens": 8516178.0,
"step": 4350
},
{
"entropy": 6.59868311882019,
"epoch": 0.43534762833008445,
"grad_norm": 0.9921875,
"learning_rate": 0.0004987275666013836,
"loss": 6.3895,
"mean_token_accuracy": 0.13959346786141397,
"num_tokens": 8525548.0,
"step": 4355
},
{
"entropy": 6.62802791595459,
"epoch": 0.43584745339131303,
"grad_norm": 0.89453125,
"learning_rate": 0.0004987237735880195,
"loss": 6.457,
"mean_token_accuracy": 0.13786746114492415,
"num_tokens": 8535941.0,
"step": 4360
},
{
"entropy": 6.453845882415772,
"epoch": 0.4363472784525416,
"grad_norm": 0.984375,
"learning_rate": 0.0004987199749458229,
"loss": 6.253,
"mean_token_accuracy": 0.14585483819246292,
"num_tokens": 8545381.0,
"step": 4365
},
{
"entropy": 6.556795263290406,
"epoch": 0.4368471035137702,
"grad_norm": 0.9765625,
"learning_rate": 0.0004987161706748893,
"loss": 6.3584,
"mean_token_accuracy": 0.14306560084223746,
"num_tokens": 8555827.0,
"step": 4370
},
{
"entropy": 6.542530155181884,
"epoch": 0.43734692857499874,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987123607753144,
"loss": 6.2519,
"mean_token_accuracy": 0.15178705900907516,
"num_tokens": 8565508.0,
"step": 4375
},
{
"entropy": 6.58016881942749,
"epoch": 0.4378467536362273,
"grad_norm": 0.99609375,
"learning_rate": 0.000498708545247194,
"loss": 6.4514,
"mean_token_accuracy": 0.13461157009005548,
"num_tokens": 8576273.0,
"step": 4380
},
{
"entropy": 6.53968505859375,
"epoch": 0.4383465786974559,
"grad_norm": 0.9921875,
"learning_rate": 0.0004987047240906243,
"loss": 6.4078,
"mean_token_accuracy": 0.13572848811745644,
"num_tokens": 8585488.0,
"step": 4385
},
{
"entropy": 6.536394500732422,
"epoch": 0.43884640375868444,
"grad_norm": 0.89453125,
"learning_rate": 0.0004987008973057012,
"loss": 6.4451,
"mean_token_accuracy": 0.13745556324720382,
"num_tokens": 8595299.0,
"step": 4390
},
{
"entropy": 6.534514331817627,
"epoch": 0.43934622881991303,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986970648925211,
"loss": 6.3101,
"mean_token_accuracy": 0.13881314992904664,
"num_tokens": 8604910.0,
"step": 4395
},
{
"entropy": 6.559590148925781,
"epoch": 0.4398460538811416,
"grad_norm": 0.984375,
"learning_rate": 0.0004986932268511804,
"loss": 6.3358,
"mean_token_accuracy": 0.1425906352698803,
"num_tokens": 8614966.0,
"step": 4400
},
{
"entropy": 6.479949283599853,
"epoch": 0.44034587894237015,
"grad_norm": 1.3671875,
"learning_rate": 0.0004986893831817758,
"loss": 6.3389,
"mean_token_accuracy": 0.13987187445163726,
"num_tokens": 8624539.0,
"step": 4405
},
{
"entropy": 6.448312044143677,
"epoch": 0.44084570400359874,
"grad_norm": 0.8984375,
"learning_rate": 0.0004986855338844038,
"loss": 6.2097,
"mean_token_accuracy": 0.15294349938631058,
"num_tokens": 8633775.0,
"step": 4410
},
{
"entropy": 6.5140574932098385,
"epoch": 0.4413455290648273,
"grad_norm": 0.96875,
"learning_rate": 0.0004986816789591613,
"loss": 6.3497,
"mean_token_accuracy": 0.13973081037402152,
"num_tokens": 8643798.0,
"step": 4415
},
{
"entropy": 6.4340917587280275,
"epoch": 0.44184535412605586,
"grad_norm": 1.0625,
"learning_rate": 0.0004986778184061453,
"loss": 6.2814,
"mean_token_accuracy": 0.14660026282072067,
"num_tokens": 8653695.0,
"step": 4420
},
{
"entropy": 6.5121674060821535,
"epoch": 0.44234517918728444,
"grad_norm": 0.96484375,
"learning_rate": 0.0004986739522254529,
"loss": 6.2931,
"mean_token_accuracy": 0.14262207597494125,
"num_tokens": 8663521.0,
"step": 4425
},
{
"entropy": 6.610686874389648,
"epoch": 0.44284500424851303,
"grad_norm": 0.8515625,
"learning_rate": 0.0004986700804171816,
"loss": 6.4275,
"mean_token_accuracy": 0.13674290776252745,
"num_tokens": 8673108.0,
"step": 4430
},
{
"entropy": 6.578842210769653,
"epoch": 0.4433448293097416,
"grad_norm": 0.93359375,
"learning_rate": 0.0004986662029814285,
"loss": 6.3075,
"mean_token_accuracy": 0.14130824506282808,
"num_tokens": 8682917.0,
"step": 4435
},
{
"entropy": 6.528602027893067,
"epoch": 0.44384465437097015,
"grad_norm": 0.94140625,
"learning_rate": 0.0004986623199182913,
"loss": 6.4007,
"mean_token_accuracy": 0.13404660820960998,
"num_tokens": 8692436.0,
"step": 4440
},
{
"entropy": 6.5239604949951175,
"epoch": 0.44434447943219874,
"grad_norm": 0.81640625,
"learning_rate": 0.0004986584312278678,
"loss": 6.3106,
"mean_token_accuracy": 0.14313255548477172,
"num_tokens": 8702627.0,
"step": 4445
},
{
"entropy": 6.543689584732055,
"epoch": 0.4448443044934273,
"grad_norm": 0.953125,
"learning_rate": 0.0004986545369102556,
"loss": 6.3211,
"mean_token_accuracy": 0.1331249937415123,
"num_tokens": 8711697.0,
"step": 4450
},
{
"entropy": 6.585853910446167,
"epoch": 0.44534412955465585,
"grad_norm": 0.98046875,
"learning_rate": 0.0004986506369655528,
"loss": 6.3508,
"mean_token_accuracy": 0.14003459364175797,
"num_tokens": 8721449.0,
"step": 4455
},
{
"entropy": 6.521495056152344,
"epoch": 0.44584395461588444,
"grad_norm": 0.9140625,
"learning_rate": 0.0004986467313938575,
"loss": 6.3286,
"mean_token_accuracy": 0.13868139386177064,
"num_tokens": 8731829.0,
"step": 4460
},
{
"entropy": 6.488602638244629,
"epoch": 0.44634377967711303,
"grad_norm": 0.83984375,
"learning_rate": 0.000498642820195268,
"loss": 6.1499,
"mean_token_accuracy": 0.15414085388183593,
"num_tokens": 8741768.0,
"step": 4465
},
{
"entropy": 6.543363189697265,
"epoch": 0.44684360473834156,
"grad_norm": 0.984375,
"learning_rate": 0.0004986389033698827,
"loss": 6.4436,
"mean_token_accuracy": 0.1356188327074051,
"num_tokens": 8751473.0,
"step": 4470
},
{
"entropy": 6.566509199142456,
"epoch": 0.44734342979957015,
"grad_norm": 0.8984375,
"learning_rate": 0.0004986349809178002,
"loss": 6.3693,
"mean_token_accuracy": 0.13763006404042244,
"num_tokens": 8761229.0,
"step": 4475
},
{
"entropy": 6.47111701965332,
"epoch": 0.44784325486079873,
"grad_norm": 0.93359375,
"learning_rate": 0.0004986310528391192,
"loss": 6.3158,
"mean_token_accuracy": 0.14091528058052064,
"num_tokens": 8770670.0,
"step": 4480
},
{
"entropy": 6.554780721664429,
"epoch": 0.44834307992202727,
"grad_norm": 0.9921875,
"learning_rate": 0.0004986271191339383,
"loss": 6.3321,
"mean_token_accuracy": 0.14046338498592376,
"num_tokens": 8780816.0,
"step": 4485
},
{
"entropy": 6.643815612792968,
"epoch": 0.44884290498325585,
"grad_norm": 0.8984375,
"learning_rate": 0.0004986231798023566,
"loss": 6.4008,
"mean_token_accuracy": 0.13553248718380928,
"num_tokens": 8790304.0,
"step": 4490
},
{
"entropy": 6.523479557037353,
"epoch": 0.44934273004448444,
"grad_norm": 0.984375,
"learning_rate": 0.0004986192348444734,
"loss": 6.3399,
"mean_token_accuracy": 0.14030980542302132,
"num_tokens": 8799188.0,
"step": 4495
},
{
"entropy": 6.53016505241394,
"epoch": 0.449842555105713,
"grad_norm": 0.90625,
"learning_rate": 0.0004986152842603877,
"loss": 6.395,
"mean_token_accuracy": 0.14093470871448516,
"num_tokens": 8810194.0,
"step": 4500
},
{
"entropy": 6.619841718673706,
"epoch": 0.45034238016694156,
"grad_norm": 0.82421875,
"learning_rate": 0.0004986113280501989,
"loss": 6.4002,
"mean_token_accuracy": 0.13877397179603576,
"num_tokens": 8820394.0,
"step": 4505
},
{
"entropy": 6.525702428817749,
"epoch": 0.45084220522817015,
"grad_norm": 0.859375,
"learning_rate": 0.0004986073662140068,
"loss": 6.3535,
"mean_token_accuracy": 0.14507648348808289,
"num_tokens": 8831308.0,
"step": 4510
},
{
"entropy": 6.482215404510498,
"epoch": 0.45134203028939873,
"grad_norm": 0.91015625,
"learning_rate": 0.0004986033987519109,
"loss": 6.2753,
"mean_token_accuracy": 0.14635651409626008,
"num_tokens": 8841461.0,
"step": 4515
},
{
"entropy": 6.533067750930786,
"epoch": 0.45184185535062726,
"grad_norm": 0.890625,
"learning_rate": 0.0004985994256640108,
"loss": 6.2928,
"mean_token_accuracy": 0.14116144999861718,
"num_tokens": 8850356.0,
"step": 4520
},
{
"entropy": 6.559308958053589,
"epoch": 0.45234168041185585,
"grad_norm": 0.8984375,
"learning_rate": 0.0004985954469504069,
"loss": 6.392,
"mean_token_accuracy": 0.13512633517384529,
"num_tokens": 8859863.0,
"step": 4525
},
{
"entropy": 6.538211917877197,
"epoch": 0.45284150547308444,
"grad_norm": 0.953125,
"learning_rate": 0.0004985914626111991,
"loss": 6.2717,
"mean_token_accuracy": 0.1402684345841408,
"num_tokens": 8869887.0,
"step": 4530
},
{
"entropy": 6.546883058547974,
"epoch": 0.45334133053431297,
"grad_norm": 0.89453125,
"learning_rate": 0.0004985874726464875,
"loss": 6.4243,
"mean_token_accuracy": 0.1382495306432247,
"num_tokens": 8879950.0,
"step": 4535
},
{
"entropy": 6.52933406829834,
"epoch": 0.45384115559554156,
"grad_norm": 0.90625,
"learning_rate": 0.0004985834770563727,
"loss": 6.1893,
"mean_token_accuracy": 0.1510006695985794,
"num_tokens": 8888304.0,
"step": 4540
},
{
"entropy": 6.478248548507691,
"epoch": 0.45434098065677014,
"grad_norm": 0.82421875,
"learning_rate": 0.0004985794758409552,
"loss": 6.2096,
"mean_token_accuracy": 0.15427092462778091,
"num_tokens": 8898642.0,
"step": 4545
},
{
"entropy": 6.4841570377349855,
"epoch": 0.4548408057179987,
"grad_norm": 0.8203125,
"learning_rate": 0.0004985754690003354,
"loss": 6.4212,
"mean_token_accuracy": 0.13655322417616844,
"num_tokens": 8909202.0,
"step": 4550
},
{
"entropy": 6.4617311477661135,
"epoch": 0.45534063077922726,
"grad_norm": 0.984375,
"learning_rate": 0.0004985714565346145,
"loss": 6.25,
"mean_token_accuracy": 0.15060166120529175,
"num_tokens": 8918844.0,
"step": 4555
},
{
"entropy": 6.5481210231781,
"epoch": 0.45584045584045585,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985674384438933,
"loss": 6.3483,
"mean_token_accuracy": 0.13722689300775529,
"num_tokens": 8928706.0,
"step": 4560
},
{
"entropy": 6.530044269561768,
"epoch": 0.45634028090168444,
"grad_norm": 0.91015625,
"learning_rate": 0.0004985634147282729,
"loss": 6.2919,
"mean_token_accuracy": 0.14466736614704132,
"num_tokens": 8939293.0,
"step": 4565
},
{
"entropy": 6.468539047241211,
"epoch": 0.45684010596291297,
"grad_norm": 0.93359375,
"learning_rate": 0.0004985593853878544,
"loss": 6.2609,
"mean_token_accuracy": 0.1399562895298004,
"num_tokens": 8947922.0,
"step": 4570
},
{
"entropy": 6.459963941574097,
"epoch": 0.45733993102414155,
"grad_norm": 0.86328125,
"learning_rate": 0.0004985553504227394,
"loss": 6.2852,
"mean_token_accuracy": 0.14619463086128234,
"num_tokens": 8957745.0,
"step": 4575
},
{
"entropy": 6.572744941711425,
"epoch": 0.45783975608537014,
"grad_norm": 0.97265625,
"learning_rate": 0.0004985513098330293,
"loss": 6.4252,
"mean_token_accuracy": 0.134528299421072,
"num_tokens": 8968331.0,
"step": 4580
},
{
"entropy": 6.494533395767212,
"epoch": 0.4583395811465987,
"grad_norm": 0.98828125,
"learning_rate": 0.0004985472636188257,
"loss": 6.258,
"mean_token_accuracy": 0.14801517724990845,
"num_tokens": 8977752.0,
"step": 4585
},
{
"entropy": 6.491020774841308,
"epoch": 0.45883940620782726,
"grad_norm": 1.046875,
"learning_rate": 0.0004985432117802305,
"loss": 6.3066,
"mean_token_accuracy": 0.13831876888871192,
"num_tokens": 8986701.0,
"step": 4590
},
{
"entropy": 6.481177520751953,
"epoch": 0.45933923126905585,
"grad_norm": 0.9609375,
"learning_rate": 0.0004985391543173456,
"loss": 6.3461,
"mean_token_accuracy": 0.13883562609553338,
"num_tokens": 8996192.0,
"step": 4595
},
{
"entropy": 6.54420051574707,
"epoch": 0.4598390563302844,
"grad_norm": 0.83203125,
"learning_rate": 0.000498535091230273,
"loss": 6.3086,
"mean_token_accuracy": 0.14105446189641951,
"num_tokens": 9007059.0,
"step": 4600
},
{
"entropy": 6.458060646057129,
"epoch": 0.46033888139151297,
"grad_norm": 0.92578125,
"learning_rate": 0.0004985310225191152,
"loss": 6.2587,
"mean_token_accuracy": 0.14425296783447267,
"num_tokens": 9016923.0,
"step": 4605
},
{
"entropy": 6.469832992553711,
"epoch": 0.46083870645274155,
"grad_norm": 0.9375,
"learning_rate": 0.0004985269481839744,
"loss": 6.3767,
"mean_token_accuracy": 0.13866270035505296,
"num_tokens": 9026034.0,
"step": 4610
},
{
"entropy": 6.590351867675781,
"epoch": 0.4613385315139701,
"grad_norm": 0.93359375,
"learning_rate": 0.0004985228682249529,
"loss": 6.3882,
"mean_token_accuracy": 0.1375481203198433,
"num_tokens": 9036773.0,
"step": 4615
},
{
"entropy": 6.531398010253906,
"epoch": 0.46183835657519867,
"grad_norm": 0.80859375,
"learning_rate": 0.0004985187826421538,
"loss": 6.3289,
"mean_token_accuracy": 0.13996466919779776,
"num_tokens": 9047825.0,
"step": 4620
},
{
"entropy": 6.561636114120484,
"epoch": 0.46233818163642726,
"grad_norm": 0.953125,
"learning_rate": 0.0004985146914356795,
"loss": 6.4206,
"mean_token_accuracy": 0.13960269317030907,
"num_tokens": 9056295.0,
"step": 4625
},
{
"entropy": 6.5317614555358885,
"epoch": 0.46283800669765585,
"grad_norm": 0.8984375,
"learning_rate": 0.000498510594605633,
"loss": 6.3632,
"mean_token_accuracy": 0.1384408563375473,
"num_tokens": 9067245.0,
"step": 4630
},
{
"entropy": 6.4532373428344725,
"epoch": 0.4633378317588844,
"grad_norm": 0.84765625,
"learning_rate": 0.0004985064921521177,
"loss": 6.3319,
"mean_token_accuracy": 0.13723526373505593,
"num_tokens": 9077528.0,
"step": 4635
},
{
"entropy": 6.5533863544464115,
"epoch": 0.46383765682011296,
"grad_norm": 0.96484375,
"learning_rate": 0.0004985023840752364,
"loss": 6.2419,
"mean_token_accuracy": 0.1463705249130726,
"num_tokens": 9086015.0,
"step": 4640
},
{
"entropy": 6.536900424957276,
"epoch": 0.46433748188134155,
"grad_norm": 1.0625,
"learning_rate": 0.0004984982703750926,
"loss": 6.3194,
"mean_token_accuracy": 0.13669760674238204,
"num_tokens": 9095963.0,
"step": 4645
},
{
"entropy": 6.3751177310943605,
"epoch": 0.4648373069425701,
"grad_norm": 1.0,
"learning_rate": 0.0004984941510517899,
"loss": 6.1413,
"mean_token_accuracy": 0.15125792622566223,
"num_tokens": 9105819.0,
"step": 4650
},
{
"entropy": 6.464892911911011,
"epoch": 0.46533713200379867,
"grad_norm": 1.046875,
"learning_rate": 0.0004984900261054319,
"loss": 6.1824,
"mean_token_accuracy": 0.14837499260902404,
"num_tokens": 9114574.0,
"step": 4655
},
{
"entropy": 6.515645122528076,
"epoch": 0.46583695706502726,
"grad_norm": 1.0546875,
"learning_rate": 0.0004984858955361224,
"loss": 6.3693,
"mean_token_accuracy": 0.1343770481646061,
"num_tokens": 9125490.0,
"step": 4660
},
{
"entropy": 6.599453639984131,
"epoch": 0.4663367821262558,
"grad_norm": 0.98828125,
"learning_rate": 0.0004984817593439651,
"loss": 6.3226,
"mean_token_accuracy": 0.13833653926849365,
"num_tokens": 9134886.0,
"step": 4665
},
{
"entropy": 6.482732820510864,
"epoch": 0.4668366071874844,
"grad_norm": 0.9296875,
"learning_rate": 0.0004984776175290643,
"loss": 6.274,
"mean_token_accuracy": 0.1443267822265625,
"num_tokens": 9143598.0,
"step": 4670
},
{
"entropy": 6.461186647415161,
"epoch": 0.46733643224871296,
"grad_norm": 1.015625,
"learning_rate": 0.0004984734700915242,
"loss": 6.3832,
"mean_token_accuracy": 0.13372259736061096,
"num_tokens": 9153051.0,
"step": 4675
},
{
"entropy": 6.584674549102783,
"epoch": 0.4678362573099415,
"grad_norm": 0.90625,
"learning_rate": 0.000498469317031449,
"loss": 6.2878,
"mean_token_accuracy": 0.14061242938041688,
"num_tokens": 9162472.0,
"step": 4680
},
{
"entropy": 6.486149549484253,
"epoch": 0.4683360823711701,
"grad_norm": 0.97265625,
"learning_rate": 0.0004984651583489434,
"loss": 6.3395,
"mean_token_accuracy": 0.14511577412486076,
"num_tokens": 9171453.0,
"step": 4685
},
{
"entropy": 6.4301036357879635,
"epoch": 0.46883590743239867,
"grad_norm": 0.87890625,
"learning_rate": 0.0004984609940441119,
"loss": 6.2823,
"mean_token_accuracy": 0.14500125348567963,
"num_tokens": 9181104.0,
"step": 4690
},
{
"entropy": 6.508776473999023,
"epoch": 0.46933573249362726,
"grad_norm": 0.9140625,
"learning_rate": 0.0004984568241170593,
"loss": 6.3576,
"mean_token_accuracy": 0.14792504906654358,
"num_tokens": 9190403.0,
"step": 4695
},
{
"entropy": 6.45200548171997,
"epoch": 0.4698355575548558,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984526485678905,
"loss": 6.2202,
"mean_token_accuracy": 0.14412133619189263,
"num_tokens": 9199132.0,
"step": 4700
},
{
"entropy": 6.5063223361969,
"epoch": 0.4703353826160844,
"grad_norm": 0.97265625,
"learning_rate": 0.0004984484673967105,
"loss": 6.394,
"mean_token_accuracy": 0.13039350882172585,
"num_tokens": 9208274.0,
"step": 4705
},
{
"entropy": 6.47504997253418,
"epoch": 0.47083520767731296,
"grad_norm": 0.87890625,
"learning_rate": 0.0004984442806036246,
"loss": 6.3539,
"mean_token_accuracy": 0.14769015461206436,
"num_tokens": 9218513.0,
"step": 4710
},
{
"entropy": 6.551234483718872,
"epoch": 0.4713350327385415,
"grad_norm": 0.92578125,
"learning_rate": 0.0004984400881887381,
"loss": 6.3541,
"mean_token_accuracy": 0.13651967868208886,
"num_tokens": 9229197.0,
"step": 4715
},
{
"entropy": 6.499589490890503,
"epoch": 0.4718348577997701,
"grad_norm": 0.90625,
"learning_rate": 0.0004984358901521565,
"loss": 6.2107,
"mean_token_accuracy": 0.14268688559532167,
"num_tokens": 9239465.0,
"step": 4720
},
{
"entropy": 6.4695335865020756,
"epoch": 0.47233468286099867,
"grad_norm": 0.87890625,
"learning_rate": 0.0004984316864939853,
"loss": 6.348,
"mean_token_accuracy": 0.1469755545258522,
"num_tokens": 9250112.0,
"step": 4725
},
{
"entropy": 6.508824443817138,
"epoch": 0.4728345079222272,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984274772143304,
"loss": 6.3276,
"mean_token_accuracy": 0.1430274710059166,
"num_tokens": 9259406.0,
"step": 4730
},
{
"entropy": 6.583673524856567,
"epoch": 0.4733343329834558,
"grad_norm": 0.9609375,
"learning_rate": 0.0004984232623132977,
"loss": 6.266,
"mean_token_accuracy": 0.14246689677238464,
"num_tokens": 9269462.0,
"step": 4735
},
{
"entropy": 6.483604621887207,
"epoch": 0.4738341580446844,
"grad_norm": 0.91796875,
"learning_rate": 0.0004984190417909932,
"loss": 6.323,
"mean_token_accuracy": 0.13639454841613768,
"num_tokens": 9280291.0,
"step": 4740
},
{
"entropy": 6.5070860385894775,
"epoch": 0.4743339831059129,
"grad_norm": 0.99609375,
"learning_rate": 0.0004984148156475231,
"loss": 6.2965,
"mean_token_accuracy": 0.13384532183408737,
"num_tokens": 9289680.0,
"step": 4745
},
{
"entropy": 6.543754577636719,
"epoch": 0.4748338081671415,
"grad_norm": 0.9375,
"learning_rate": 0.0004984105838829937,
"loss": 6.3061,
"mean_token_accuracy": 0.13828267604112626,
"num_tokens": 9301043.0,
"step": 4750
},
{
"entropy": 6.415587282180786,
"epoch": 0.4753336332283701,
"grad_norm": 1.0078125,
"learning_rate": 0.0004984063464975116,
"loss": 6.3145,
"mean_token_accuracy": 0.14800099208950995,
"num_tokens": 9311851.0,
"step": 4755
},
{
"entropy": 6.579112482070923,
"epoch": 0.47583345828959867,
"grad_norm": 1.0,
"learning_rate": 0.0004984021034911833,
"loss": 6.2891,
"mean_token_accuracy": 0.14502299576997757,
"num_tokens": 9320886.0,
"step": 4760
},
{
"entropy": 6.488643503189087,
"epoch": 0.4763332833508272,
"grad_norm": 1.0546875,
"learning_rate": 0.0004983978548641154,
"loss": 6.3394,
"mean_token_accuracy": 0.13805975541472434,
"num_tokens": 9330518.0,
"step": 4765
},
{
"entropy": 6.442863607406617,
"epoch": 0.4768331084120558,
"grad_norm": 0.95703125,
"learning_rate": 0.0004983936006164151,
"loss": 6.204,
"mean_token_accuracy": 0.14599989354610443,
"num_tokens": 9339703.0,
"step": 4770
},
{
"entropy": 6.587995529174805,
"epoch": 0.47733293347328437,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983893407481892,
"loss": 6.4342,
"mean_token_accuracy": 0.1312703713774681,
"num_tokens": 9349931.0,
"step": 4775
},
{
"entropy": 6.507864952087402,
"epoch": 0.4778327585345129,
"grad_norm": 0.98046875,
"learning_rate": 0.0004983850752595451,
"loss": 6.3116,
"mean_token_accuracy": 0.14585210010409355,
"num_tokens": 9358984.0,
"step": 4780
},
{
"entropy": 6.5157966136932375,
"epoch": 0.4783325835957415,
"grad_norm": 0.92578125,
"learning_rate": 0.0004983808041505898,
"loss": 6.2416,
"mean_token_accuracy": 0.14548552930355071,
"num_tokens": 9368081.0,
"step": 4785
},
{
"entropy": 6.501175260543823,
"epoch": 0.4788324086569701,
"grad_norm": 0.98828125,
"learning_rate": 0.0004983765274214311,
"loss": 6.3742,
"mean_token_accuracy": 0.136806371062994,
"num_tokens": 9377346.0,
"step": 4790
},
{
"entropy": 6.436359977722168,
"epoch": 0.4793322337181986,
"grad_norm": 0.8671875,
"learning_rate": 0.0004983722450721764,
"loss": 6.3337,
"mean_token_accuracy": 0.14385157749056815,
"num_tokens": 9387493.0,
"step": 4795
},
{
"entropy": 6.520362520217896,
"epoch": 0.4798320587794272,
"grad_norm": 0.98046875,
"learning_rate": 0.0004983679571029336,
"loss": 6.2952,
"mean_token_accuracy": 0.1397965058684349,
"num_tokens": 9397154.0,
"step": 4800
},
{
"entropy": 6.57193775177002,
"epoch": 0.4803318838406558,
"grad_norm": 0.9140625,
"learning_rate": 0.0004983636635138103,
"loss": 6.4053,
"mean_token_accuracy": 0.13497058004140855,
"num_tokens": 9406705.0,
"step": 4805
},
{
"entropy": 6.510138034820557,
"epoch": 0.4808317089018843,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983593643049148,
"loss": 6.2791,
"mean_token_accuracy": 0.148053440451622,
"num_tokens": 9416850.0,
"step": 4810
},
{
"entropy": 6.421587371826172,
"epoch": 0.4813315339631129,
"grad_norm": 1.0625,
"learning_rate": 0.0004983550594763552,
"loss": 6.2334,
"mean_token_accuracy": 0.14565369784832,
"num_tokens": 9425841.0,
"step": 4815
},
{
"entropy": 6.430533838272095,
"epoch": 0.4818313590243415,
"grad_norm": 1.0078125,
"learning_rate": 0.0004983507490282397,
"loss": 6.1829,
"mean_token_accuracy": 0.14602613002061843,
"num_tokens": 9436748.0,
"step": 4820
},
{
"entropy": 6.473450517654419,
"epoch": 0.4823311840855701,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983464329606768,
"loss": 6.3341,
"mean_token_accuracy": 0.14309628009796144,
"num_tokens": 9446043.0,
"step": 4825
},
{
"entropy": 6.497003698348999,
"epoch": 0.4828310091467986,
"grad_norm": 1.015625,
"learning_rate": 0.0004983421112737752,
"loss": 6.3426,
"mean_token_accuracy": 0.14059921652078627,
"num_tokens": 9455480.0,
"step": 4830
},
{
"entropy": 6.52227029800415,
"epoch": 0.4833308342080272,
"grad_norm": 0.9296875,
"learning_rate": 0.0004983377839676435,
"loss": 6.3415,
"mean_token_accuracy": 0.1429636374115944,
"num_tokens": 9464412.0,
"step": 4835
},
{
"entropy": 6.487901496887207,
"epoch": 0.4838306592692558,
"grad_norm": 0.921875,
"learning_rate": 0.0004983334510423906,
"loss": 6.2687,
"mean_token_accuracy": 0.14371415302157403,
"num_tokens": 9475639.0,
"step": 4840
},
{
"entropy": 6.496836519241333,
"epoch": 0.4843304843304843,
"grad_norm": 0.92578125,
"learning_rate": 0.0004983291124981257,
"loss": 6.2324,
"mean_token_accuracy": 0.1428309366106987,
"num_tokens": 9484721.0,
"step": 4845
},
{
"entropy": 6.392262268066406,
"epoch": 0.4848303093917129,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983247683349576,
"loss": 6.1839,
"mean_token_accuracy": 0.15078072547912597,
"num_tokens": 9494594.0,
"step": 4850
},
{
"entropy": 6.379246187210083,
"epoch": 0.4853301344529415,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983204185529959,
"loss": 6.2297,
"mean_token_accuracy": 0.15043352991342546,
"num_tokens": 9503707.0,
"step": 4855
},
{
"entropy": 6.49135274887085,
"epoch": 0.48582995951417,
"grad_norm": 1.0546875,
"learning_rate": 0.0004983160631523499,
"loss": 6.1902,
"mean_token_accuracy": 0.14579902365803718,
"num_tokens": 9512473.0,
"step": 4860
},
{
"entropy": 6.504326486587525,
"epoch": 0.4863297845753986,
"grad_norm": 0.8671875,
"learning_rate": 0.0004983117021331293,
"loss": 6.328,
"mean_token_accuracy": 0.14437096416950226,
"num_tokens": 9523253.0,
"step": 4865
},
{
"entropy": 6.5039753913879395,
"epoch": 0.4868296096366272,
"grad_norm": 1.0546875,
"learning_rate": 0.0004983073354954436,
"loss": 6.2876,
"mean_token_accuracy": 0.14891613274812698,
"num_tokens": 9531938.0,
"step": 4870
},
{
"entropy": 6.523850584030152,
"epoch": 0.4873294346978557,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983029632394029,
"loss": 6.2931,
"mean_token_accuracy": 0.14198393672704696,
"num_tokens": 9541542.0,
"step": 4875
},
{
"entropy": 6.428567218780517,
"epoch": 0.4878292597590843,
"grad_norm": 0.890625,
"learning_rate": 0.000498298585365117,
"loss": 6.2193,
"mean_token_accuracy": 0.14667681604623795,
"num_tokens": 9551521.0,
"step": 4880
},
{
"entropy": 6.466378831863404,
"epoch": 0.4883290848203129,
"grad_norm": 1.078125,
"learning_rate": 0.0004982942018726963,
"loss": 6.3514,
"mean_token_accuracy": 0.14658211767673493,
"num_tokens": 9561605.0,
"step": 4885
},
{
"entropy": 6.514699125289917,
"epoch": 0.4888289098815415,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982898127622508,
"loss": 6.2553,
"mean_token_accuracy": 0.14559754058718682,
"num_tokens": 9571591.0,
"step": 4890
},
{
"entropy": 6.407695722579956,
"epoch": 0.48932873494277,
"grad_norm": 1.0625,
"learning_rate": 0.0004982854180338911,
"loss": 6.2647,
"mean_token_accuracy": 0.1466855026781559,
"num_tokens": 9580994.0,
"step": 4895
},
{
"entropy": 6.5345968246459964,
"epoch": 0.4898285600039986,
"grad_norm": 0.9921875,
"learning_rate": 0.0004982810176877279,
"loss": 6.268,
"mean_token_accuracy": 0.14102225825190545,
"num_tokens": 9589268.0,
"step": 4900
},
{
"entropy": 6.470806026458741,
"epoch": 0.4903283850652272,
"grad_norm": 0.90234375,
"learning_rate": 0.0004982766117238716,
"loss": 6.3216,
"mean_token_accuracy": 0.1400758869946003,
"num_tokens": 9599461.0,
"step": 4905
},
{
"entropy": 6.453066110610962,
"epoch": 0.4908282101264557,
"grad_norm": 1.0859375,
"learning_rate": 0.0004982722001424331,
"loss": 6.2917,
"mean_token_accuracy": 0.14182407408952713,
"num_tokens": 9609629.0,
"step": 4910
},
{
"entropy": 6.418763780593872,
"epoch": 0.4913280351876843,
"grad_norm": 0.953125,
"learning_rate": 0.0004982677829435236,
"loss": 6.2079,
"mean_token_accuracy": 0.15076979696750642,
"num_tokens": 9619618.0,
"step": 4915
},
{
"entropy": 6.472671461105347,
"epoch": 0.4918278602489129,
"grad_norm": 0.984375,
"learning_rate": 0.0004982633601272543,
"loss": 6.2695,
"mean_token_accuracy": 0.14576450288295745,
"num_tokens": 9629042.0,
"step": 4920
},
{
"entropy": 6.462853670120239,
"epoch": 0.4923276853101414,
"grad_norm": 0.9140625,
"learning_rate": 0.0004982589316937361,
"loss": 6.243,
"mean_token_accuracy": 0.14421821683645247,
"num_tokens": 9638596.0,
"step": 4925
},
{
"entropy": 6.503546905517578,
"epoch": 0.49282751037137,
"grad_norm": 0.921875,
"learning_rate": 0.0004982544976430808,
"loss": 6.333,
"mean_token_accuracy": 0.14002783223986626,
"num_tokens": 9648505.0,
"step": 4930
},
{
"entropy": 6.422399234771729,
"epoch": 0.4933273354325986,
"grad_norm": 0.94921875,
"learning_rate": 0.0004982500579753995,
"loss": 6.3015,
"mean_token_accuracy": 0.14369556680321693,
"num_tokens": 9658113.0,
"step": 4935
},
{
"entropy": 6.4771932601928714,
"epoch": 0.49382716049382713,
"grad_norm": 0.9375,
"learning_rate": 0.0004982456126908044,
"loss": 6.328,
"mean_token_accuracy": 0.14563848972320556,
"num_tokens": 9668075.0,
"step": 4940
},
{
"entropy": 6.582532358169556,
"epoch": 0.4943269855550557,
"grad_norm": 0.88671875,
"learning_rate": 0.0004982411617894071,
"loss": 6.2838,
"mean_token_accuracy": 0.1421746701002121,
"num_tokens": 9678139.0,
"step": 4945
},
{
"entropy": 6.458785152435302,
"epoch": 0.4948268106162843,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982367052713196,
"loss": 6.3209,
"mean_token_accuracy": 0.1364033341407776,
"num_tokens": 9687328.0,
"step": 4950
},
{
"entropy": 6.428125238418579,
"epoch": 0.4953266356775129,
"grad_norm": 0.859375,
"learning_rate": 0.0004982322431366541,
"loss": 6.2963,
"mean_token_accuracy": 0.14726474955677987,
"num_tokens": 9697536.0,
"step": 4955
},
{
"entropy": 6.507616567611694,
"epoch": 0.4958264607387414,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982277753855227,
"loss": 6.3239,
"mean_token_accuracy": 0.14271707609295844,
"num_tokens": 9706804.0,
"step": 4960
},
{
"entropy": 6.499647188186645,
"epoch": 0.49632628579997,
"grad_norm": 0.9453125,
"learning_rate": 0.000498223302018038,
"loss": 6.2152,
"mean_token_accuracy": 0.15250640586018563,
"num_tokens": 9716868.0,
"step": 4965
},
{
"entropy": 6.424004459381104,
"epoch": 0.4968261108611986,
"grad_norm": 1.0,
"learning_rate": 0.0004982188230343124,
"loss": 6.2936,
"mean_token_accuracy": 0.14402196258306504,
"num_tokens": 9728790.0,
"step": 4970
},
{
"entropy": 6.521448183059692,
"epoch": 0.49732593592242713,
"grad_norm": 0.9453125,
"learning_rate": 0.0004982143384344587,
"loss": 6.3209,
"mean_token_accuracy": 0.14097120687365533,
"num_tokens": 9739118.0,
"step": 4975
},
{
"entropy": 6.506854438781739,
"epoch": 0.4978257609836557,
"grad_norm": 0.9140625,
"learning_rate": 0.0004982098482185897,
"loss": 6.3063,
"mean_token_accuracy": 0.14251478612422944,
"num_tokens": 9749347.0,
"step": 4980
},
{
"entropy": 6.511482858657837,
"epoch": 0.4983255860448843,
"grad_norm": 0.9375,
"learning_rate": 0.0004982053523868182,
"loss": 6.3376,
"mean_token_accuracy": 0.14065980836749076,
"num_tokens": 9759432.0,
"step": 4985
},
{
"entropy": 6.447167539596558,
"epoch": 0.49882541110611284,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982008509392576,
"loss": 6.2469,
"mean_token_accuracy": 0.14047390222549438,
"num_tokens": 9768666.0,
"step": 4990
},
{
"entropy": 6.373848533630371,
"epoch": 0.4993252361673414,
"grad_norm": 1.0234375,
"learning_rate": 0.000498196343876021,
"loss": 6.2014,
"mean_token_accuracy": 0.1518920511007309,
"num_tokens": 9778699.0,
"step": 4995
},
{
"entropy": 6.458275175094604,
"epoch": 0.49982506122857,
"grad_norm": 0.9296875,
"learning_rate": 0.0004981918311972219,
"loss": 6.2379,
"mean_token_accuracy": 0.14668502882122994,
"num_tokens": 9788156.0,
"step": 5000
},
{
"entropy": 6.414515447616577,
"epoch": 0.5003248862897985,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981873129029737,
"loss": 6.19,
"mean_token_accuracy": 0.15714230835437776,
"num_tokens": 9797293.0,
"step": 5005
},
{
"entropy": 6.414603519439697,
"epoch": 0.5008247113510271,
"grad_norm": 0.953125,
"learning_rate": 0.00049818278899339,
"loss": 6.208,
"mean_token_accuracy": 0.14584677964448928,
"num_tokens": 9806962.0,
"step": 5010
},
{
"entropy": 6.464185571670532,
"epoch": 0.5013245364122557,
"grad_norm": 1.0546875,
"learning_rate": 0.000498178259468585,
"loss": 6.3421,
"mean_token_accuracy": 0.14279545992612838,
"num_tokens": 9816689.0,
"step": 5015
},
{
"entropy": 6.481401824951172,
"epoch": 0.5018243614734843,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981737243286722,
"loss": 6.1872,
"mean_token_accuracy": 0.1465285986661911,
"num_tokens": 9826321.0,
"step": 5020
},
{
"entropy": 6.539868974685669,
"epoch": 0.5023241865347129,
"grad_norm": 0.984375,
"learning_rate": 0.0004981691835737661,
"loss": 6.3484,
"mean_token_accuracy": 0.13538563400506973,
"num_tokens": 9836054.0,
"step": 5025
},
{
"entropy": 6.552703857421875,
"epoch": 0.5028240115959414,
"grad_norm": 0.87109375,
"learning_rate": 0.0004981646372039807,
"loss": 6.3359,
"mean_token_accuracy": 0.1394810661673546,
"num_tokens": 9846516.0,
"step": 5030
},
{
"entropy": 6.45863299369812,
"epoch": 0.50332383665717,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981600852194305,
"loss": 6.1251,
"mean_token_accuracy": 0.15314733237028122,
"num_tokens": 9856599.0,
"step": 5035
},
{
"entropy": 6.457842016220093,
"epoch": 0.5038236617183985,
"grad_norm": 0.890625,
"learning_rate": 0.00049815552762023,
"loss": 6.2501,
"mean_token_accuracy": 0.14722048193216325,
"num_tokens": 9866922.0,
"step": 5040
},
{
"entropy": 6.429340839385986,
"epoch": 0.5043234867796271,
"grad_norm": 0.95703125,
"learning_rate": 0.0004981509644064937,
"loss": 6.2465,
"mean_token_accuracy": 0.14557284638285636,
"num_tokens": 9876174.0,
"step": 5045
},
{
"entropy": 6.437441682815551,
"epoch": 0.5048233118408557,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981463955783367,
"loss": 6.197,
"mean_token_accuracy": 0.14810907393693923,
"num_tokens": 9886208.0,
"step": 5050
},
{
"entropy": 6.48028335571289,
"epoch": 0.5053231369020843,
"grad_norm": 0.9453125,
"learning_rate": 0.0004981418211358736,
"loss": 6.2502,
"mean_token_accuracy": 0.14508378654718398,
"num_tokens": 9896522.0,
"step": 5055
},
{
"entropy": 6.432210636138916,
"epoch": 0.5058229619633129,
"grad_norm": 1.0,
"learning_rate": 0.0004981372410792199,
"loss": 6.2389,
"mean_token_accuracy": 0.1507919669151306,
"num_tokens": 9905753.0,
"step": 5060
},
{
"entropy": 6.496094846725464,
"epoch": 0.5063227870245414,
"grad_norm": 0.91015625,
"learning_rate": 0.0004981326554084906,
"loss": 6.336,
"mean_token_accuracy": 0.14507129937410354,
"num_tokens": 9916566.0,
"step": 5065
},
{
"entropy": 6.500605773925781,
"epoch": 0.50682261208577,
"grad_norm": 0.9453125,
"learning_rate": 0.0004981280641238011,
"loss": 6.3197,
"mean_token_accuracy": 0.13981844410300254,
"num_tokens": 9926126.0,
"step": 5070
},
{
"entropy": 6.484925937652588,
"epoch": 0.5073224371469985,
"grad_norm": 0.92578125,
"learning_rate": 0.0004981234672252669,
"loss": 6.3035,
"mean_token_accuracy": 0.14005504474043845,
"num_tokens": 9937666.0,
"step": 5075
},
{
"entropy": 6.536907005310058,
"epoch": 0.5078222622082271,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981188647130037,
"loss": 6.3467,
"mean_token_accuracy": 0.15003084391355515,
"num_tokens": 9947107.0,
"step": 5080
},
{
"entropy": 6.478869819641114,
"epoch": 0.5083220872694557,
"grad_norm": 0.9453125,
"learning_rate": 0.0004981142565871272,
"loss": 6.1996,
"mean_token_accuracy": 0.14651109427213668,
"num_tokens": 9956484.0,
"step": 5085
},
{
"entropy": 6.357135772705078,
"epoch": 0.5088219123306843,
"grad_norm": 0.93359375,
"learning_rate": 0.0004981096428477535,
"loss": 6.2558,
"mean_token_accuracy": 0.1442209430038929,
"num_tokens": 9966731.0,
"step": 5090
},
{
"entropy": 6.476908922195435,
"epoch": 0.5093217373919128,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981050234949986,
"loss": 6.2411,
"mean_token_accuracy": 0.14739679247140886,
"num_tokens": 9976491.0,
"step": 5095
},
{
"entropy": 6.519944143295288,
"epoch": 0.5098215624531414,
"grad_norm": 0.9765625,
"learning_rate": 0.0004981003985289786,
"loss": 6.1776,
"mean_token_accuracy": 0.149426232278347,
"num_tokens": 9985248.0,
"step": 5100
},
{
"entropy": 6.409350252151489,
"epoch": 0.51032138751437,
"grad_norm": 0.89453125,
"learning_rate": 0.0004980957679498101,
"loss": 6.2726,
"mean_token_accuracy": 0.14669610783457757,
"num_tokens": 9995578.0,
"step": 5105
},
{
"entropy": 6.521508741378784,
"epoch": 0.5108212125755985,
"grad_norm": 0.8203125,
"learning_rate": 0.0004980911317576095,
"loss": 6.3613,
"mean_token_accuracy": 0.13964017778635024,
"num_tokens": 10006078.0,
"step": 5110
},
{
"entropy": 6.545692682266235,
"epoch": 0.5113210376368271,
"grad_norm": 0.98046875,
"learning_rate": 0.0004980864899524933,
"loss": 6.3158,
"mean_token_accuracy": 0.13703773245215417,
"num_tokens": 10015943.0,
"step": 5115
},
{
"entropy": 6.488755893707276,
"epoch": 0.5118208626980557,
"grad_norm": 0.9375,
"learning_rate": 0.0004980818425345786,
"loss": 6.3259,
"mean_token_accuracy": 0.13959860056638718,
"num_tokens": 10025738.0,
"step": 5120
},
{
"entropy": 6.538602304458618,
"epoch": 0.5123206877592843,
"grad_norm": 0.98046875,
"learning_rate": 0.000498077189503982,
"loss": 6.2814,
"mean_token_accuracy": 0.13867102414369584,
"num_tokens": 10035032.0,
"step": 5125
},
{
"entropy": 6.493793916702271,
"epoch": 0.5128205128205128,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980725308608208,
"loss": 6.2353,
"mean_token_accuracy": 0.1468093454837799,
"num_tokens": 10044201.0,
"step": 5130
},
{
"entropy": 6.3738292217254635,
"epoch": 0.5133203378817414,
"grad_norm": 0.96875,
"learning_rate": 0.0004980678666052121,
"loss": 6.2928,
"mean_token_accuracy": 0.14822009056806565,
"num_tokens": 10054163.0,
"step": 5135
},
{
"entropy": 6.444634294509887,
"epoch": 0.51382016294297,
"grad_norm": 0.8515625,
"learning_rate": 0.0004980631967372733,
"loss": 6.2834,
"mean_token_accuracy": 0.14270321428775787,
"num_tokens": 10065225.0,
"step": 5140
},
{
"entropy": 6.511883735656738,
"epoch": 0.5143199880041985,
"grad_norm": 1.0390625,
"learning_rate": 0.0004980585212571218,
"loss": 6.2556,
"mean_token_accuracy": 0.1469488687813282,
"num_tokens": 10074776.0,
"step": 5145
},
{
"entropy": 6.522441959381103,
"epoch": 0.5148198130654271,
"grad_norm": 0.87890625,
"learning_rate": 0.0004980538401648753,
"loss": 6.4009,
"mean_token_accuracy": 0.1429189771413803,
"num_tokens": 10085917.0,
"step": 5150
},
{
"entropy": 6.460656452178955,
"epoch": 0.5153196381266557,
"grad_norm": 0.94140625,
"learning_rate": 0.0004980491534606517,
"loss": 6.2226,
"mean_token_accuracy": 0.14594015330076218,
"num_tokens": 10094727.0,
"step": 5155
},
{
"entropy": 6.460877370834351,
"epoch": 0.5158194631878842,
"grad_norm": 0.88671875,
"learning_rate": 0.0004980444611445686,
"loss": 6.2006,
"mean_token_accuracy": 0.15095902010798454,
"num_tokens": 10104839.0,
"step": 5160
},
{
"entropy": 6.4641259670257565,
"epoch": 0.5163192882491128,
"grad_norm": 0.890625,
"learning_rate": 0.0004980397632167444,
"loss": 6.1788,
"mean_token_accuracy": 0.14686077684164048,
"num_tokens": 10114501.0,
"step": 5165
},
{
"entropy": 6.442966508865356,
"epoch": 0.5168191133103414,
"grad_norm": 0.97265625,
"learning_rate": 0.0004980350596772971,
"loss": 6.2806,
"mean_token_accuracy": 0.14431710094213485,
"num_tokens": 10124597.0,
"step": 5170
},
{
"entropy": 6.368315505981445,
"epoch": 0.51731893837157,
"grad_norm": 1.3671875,
"learning_rate": 0.0004980303505263451,
"loss": 6.1467,
"mean_token_accuracy": 0.14768103063106536,
"num_tokens": 10135695.0,
"step": 5175
},
{
"entropy": 6.395515727996826,
"epoch": 0.5178187634327985,
"grad_norm": 0.9140625,
"learning_rate": 0.0004980256357640069,
"loss": 6.2358,
"mean_token_accuracy": 0.14446025490760803,
"num_tokens": 10146107.0,
"step": 5180
},
{
"entropy": 6.517473602294922,
"epoch": 0.5183185884940271,
"grad_norm": 0.921875,
"learning_rate": 0.0004980209153904011,
"loss": 6.3026,
"mean_token_accuracy": 0.15090568512678146,
"num_tokens": 10155983.0,
"step": 5185
},
{
"entropy": 6.474668788909912,
"epoch": 0.5188184135552557,
"grad_norm": 0.98046875,
"learning_rate": 0.0004980161894056464,
"loss": 6.2742,
"mean_token_accuracy": 0.14347012788057328,
"num_tokens": 10166183.0,
"step": 5190
},
{
"entropy": 6.434601736068726,
"epoch": 0.5193182386164842,
"grad_norm": 1.0703125,
"learning_rate": 0.0004980114578098618,
"loss": 6.1767,
"mean_token_accuracy": 0.1427476190030575,
"num_tokens": 10175524.0,
"step": 5195
},
{
"entropy": 6.493268632888794,
"epoch": 0.5198180636777128,
"grad_norm": 0.84765625,
"learning_rate": 0.0004980067206031663,
"loss": 6.3107,
"mean_token_accuracy": 0.13323971778154373,
"num_tokens": 10186235.0,
"step": 5200
},
{
"entropy": 6.546220159530639,
"epoch": 0.5203178887389414,
"grad_norm": 0.8828125,
"learning_rate": 0.0004980019777856791,
"loss": 6.3085,
"mean_token_accuracy": 0.1422312580049038,
"num_tokens": 10196166.0,
"step": 5205
},
{
"entropy": 6.449694967269897,
"epoch": 0.52081771380017,
"grad_norm": 0.87109375,
"learning_rate": 0.0004979972293575195,
"loss": 6.2382,
"mean_token_accuracy": 0.14611050486564636,
"num_tokens": 10205691.0,
"step": 5210
},
{
"entropy": 6.425623273849487,
"epoch": 0.5213175388613985,
"grad_norm": 0.875,
"learning_rate": 0.0004979924753188069,
"loss": 6.2086,
"mean_token_accuracy": 0.14596198126673698,
"num_tokens": 10216224.0,
"step": 5215
},
{
"entropy": 6.424330139160157,
"epoch": 0.5218173639226271,
"grad_norm": 0.83203125,
"learning_rate": 0.0004979877156696612,
"loss": 6.2295,
"mean_token_accuracy": 0.1465998724102974,
"num_tokens": 10226468.0,
"step": 5220
},
{
"entropy": 6.575912141799927,
"epoch": 0.5223171889838556,
"grad_norm": 0.8828125,
"learning_rate": 0.0004979829504102018,
"loss": 6.2764,
"mean_token_accuracy": 0.14345307424664497,
"num_tokens": 10235932.0,
"step": 5225
},
{
"entropy": 6.516261863708496,
"epoch": 0.5228170140450842,
"grad_norm": 0.8359375,
"learning_rate": 0.0004979781795405488,
"loss": 6.3639,
"mean_token_accuracy": 0.1411351665854454,
"num_tokens": 10246052.0,
"step": 5230
},
{
"entropy": 6.561315822601318,
"epoch": 0.5233168391063128,
"grad_norm": 0.8828125,
"learning_rate": 0.0004979734030608222,
"loss": 6.2724,
"mean_token_accuracy": 0.1478128671646118,
"num_tokens": 10255056.0,
"step": 5235
},
{
"entropy": 6.44011173248291,
"epoch": 0.5238166641675414,
"grad_norm": 0.9453125,
"learning_rate": 0.0004979686209711422,
"loss": 6.2522,
"mean_token_accuracy": 0.14555570483207703,
"num_tokens": 10265403.0,
"step": 5240
},
{
"entropy": 6.352910900115967,
"epoch": 0.5243164892287699,
"grad_norm": 0.88671875,
"learning_rate": 0.000497963833271629,
"loss": 6.2061,
"mean_token_accuracy": 0.15292444676160813,
"num_tokens": 10275148.0,
"step": 5245
},
{
"entropy": 6.492708539962768,
"epoch": 0.5248163142899985,
"grad_norm": 0.984375,
"learning_rate": 0.0004979590399624031,
"loss": 6.2561,
"mean_token_accuracy": 0.14439393132925032,
"num_tokens": 10285083.0,
"step": 5250
},
{
"entropy": 6.423582601547241,
"epoch": 0.5253161393512271,
"grad_norm": 0.97265625,
"learning_rate": 0.0004979542410435852,
"loss": 6.216,
"mean_token_accuracy": 0.1387999877333641,
"num_tokens": 10294466.0,
"step": 5255
},
{
"entropy": 6.508393669128418,
"epoch": 0.5258159644124556,
"grad_norm": 0.94140625,
"learning_rate": 0.0004979494365152959,
"loss": 6.3139,
"mean_token_accuracy": 0.13836781978607177,
"num_tokens": 10304333.0,
"step": 5260
},
{
"entropy": 6.501136064529419,
"epoch": 0.5263157894736842,
"grad_norm": 0.8984375,
"learning_rate": 0.0004979446263776562,
"loss": 6.2463,
"mean_token_accuracy": 0.14248027056455612,
"num_tokens": 10314880.0,
"step": 5265
},
{
"entropy": 6.464469003677368,
"epoch": 0.5268156145349128,
"grad_norm": 1.015625,
"learning_rate": 0.000497939810630787,
"loss": 6.2065,
"mean_token_accuracy": 0.1451038010418415,
"num_tokens": 10324060.0,
"step": 5270
},
{
"entropy": 6.37868218421936,
"epoch": 0.5273154395961414,
"grad_norm": 0.80859375,
"learning_rate": 0.0004979349892748096,
"loss": 6.3192,
"mean_token_accuracy": 0.1428627409040928,
"num_tokens": 10335248.0,
"step": 5275
},
{
"entropy": 6.446943569183349,
"epoch": 0.5278152646573699,
"grad_norm": 1.09375,
"learning_rate": 0.0004979301623098452,
"loss": 6.1332,
"mean_token_accuracy": 0.15440346747636796,
"num_tokens": 10343974.0,
"step": 5280
},
{
"entropy": 6.47099871635437,
"epoch": 0.5283150897185985,
"grad_norm": 0.9921875,
"learning_rate": 0.0004979253297360153,
"loss": 6.2439,
"mean_token_accuracy": 0.1459724336862564,
"num_tokens": 10354037.0,
"step": 5285
},
{
"entropy": 6.445359754562378,
"epoch": 0.528814914779827,
"grad_norm": 0.82421875,
"learning_rate": 0.0004979204915534415,
"loss": 6.2953,
"mean_token_accuracy": 0.14339115396142005,
"num_tokens": 10364714.0,
"step": 5290
},
{
"entropy": 6.44316873550415,
"epoch": 0.5293147398410556,
"grad_norm": 1.015625,
"learning_rate": 0.0004979156477622454,
"loss": 6.2646,
"mean_token_accuracy": 0.1449811987578869,
"num_tokens": 10373972.0,
"step": 5295
},
{
"entropy": 6.415341997146607,
"epoch": 0.5298145649022842,
"grad_norm": 0.84765625,
"learning_rate": 0.0004979107983625491,
"loss": 6.1693,
"mean_token_accuracy": 0.1506952553987503,
"num_tokens": 10382882.0,
"step": 5300
},
{
"entropy": 6.407958555221557,
"epoch": 0.5303143899635128,
"grad_norm": 0.88671875,
"learning_rate": 0.0004979059433544744,
"loss": 6.2219,
"mean_token_accuracy": 0.14825646877288817,
"num_tokens": 10392257.0,
"step": 5305
},
{
"entropy": 6.448230981826782,
"epoch": 0.5308142150247414,
"grad_norm": 0.92578125,
"learning_rate": 0.0004979010827381435,
"loss": 6.2795,
"mean_token_accuracy": 0.14600449651479722,
"num_tokens": 10401055.0,
"step": 5310
},
{
"entropy": 6.492560720443725,
"epoch": 0.5313140400859699,
"grad_norm": 0.83984375,
"learning_rate": 0.0004978962165136787,
"loss": 6.3526,
"mean_token_accuracy": 0.13936564922332764,
"num_tokens": 10411443.0,
"step": 5315
},
{
"entropy": 6.534284257888794,
"epoch": 0.5318138651471985,
"grad_norm": 0.90625,
"learning_rate": 0.0004978913446812025,
"loss": 6.3428,
"mean_token_accuracy": 0.14371119439601898,
"num_tokens": 10422142.0,
"step": 5320
},
{
"entropy": 6.515093898773193,
"epoch": 0.532313690208427,
"grad_norm": 0.9375,
"learning_rate": 0.0004978864672408374,
"loss": 6.2592,
"mean_token_accuracy": 0.14365993440151215,
"num_tokens": 10432454.0,
"step": 5325
},
{
"entropy": 6.421989154815674,
"epoch": 0.5328135152696556,
"grad_norm": 0.8359375,
"learning_rate": 0.0004978815841927061,
"loss": 6.2525,
"mean_token_accuracy": 0.14374485164880751,
"num_tokens": 10443625.0,
"step": 5330
},
{
"entropy": 6.5193106174469,
"epoch": 0.5333133403308842,
"grad_norm": 0.91015625,
"learning_rate": 0.0004978766955369315,
"loss": 6.3038,
"mean_token_accuracy": 0.1444941483438015,
"num_tokens": 10454055.0,
"step": 5335
},
{
"entropy": 6.388335418701172,
"epoch": 0.5338131653921128,
"grad_norm": 1.1171875,
"learning_rate": 0.0004978718012736367,
"loss": 6.1437,
"mean_token_accuracy": 0.15102465599775314,
"num_tokens": 10462683.0,
"step": 5340
},
{
"entropy": 6.3790459632873535,
"epoch": 0.5343129904533414,
"grad_norm": 1.0078125,
"learning_rate": 0.0004978669014029446,
"loss": 6.1853,
"mean_token_accuracy": 0.15175978988409042,
"num_tokens": 10471699.0,
"step": 5345
},
{
"entropy": 6.343296670913697,
"epoch": 0.5348128155145699,
"grad_norm": 0.94140625,
"learning_rate": 0.0004978619959249787,
"loss": 6.1187,
"mean_token_accuracy": 0.15506782457232476,
"num_tokens": 10480790.0,
"step": 5350
},
{
"entropy": 6.406260776519775,
"epoch": 0.5353126405757984,
"grad_norm": 0.9296875,
"learning_rate": 0.0004978570848398623,
"loss": 6.2206,
"mean_token_accuracy": 0.142399550229311,
"num_tokens": 10490202.0,
"step": 5355
},
{
"entropy": 6.423729515075683,
"epoch": 0.535812465637027,
"grad_norm": 0.89453125,
"learning_rate": 0.000497852168147719,
"loss": 6.2603,
"mean_token_accuracy": 0.1435429960489273,
"num_tokens": 10499690.0,
"step": 5360
},
{
"entropy": 6.444197463989258,
"epoch": 0.5363122906982556,
"grad_norm": 0.90234375,
"learning_rate": 0.0004978472458486724,
"loss": 6.2321,
"mean_token_accuracy": 0.14401177167892457,
"num_tokens": 10510120.0,
"step": 5365
},
{
"entropy": 6.504869747161865,
"epoch": 0.5368121157594842,
"grad_norm": 0.921875,
"learning_rate": 0.0004978423179428466,
"loss": 6.3725,
"mean_token_accuracy": 0.1378040425479412,
"num_tokens": 10520521.0,
"step": 5370
},
{
"entropy": 6.377797651290893,
"epoch": 0.5373119408207128,
"grad_norm": 1.0859375,
"learning_rate": 0.0004978373844303653,
"loss": 6.0614,
"mean_token_accuracy": 0.15365611091256143,
"num_tokens": 10530233.0,
"step": 5375
},
{
"entropy": 6.388729000091553,
"epoch": 0.5378117658819414,
"grad_norm": 0.8828125,
"learning_rate": 0.0004978324453113528,
"loss": 6.2021,
"mean_token_accuracy": 0.14688197374343873,
"num_tokens": 10539683.0,
"step": 5380
},
{
"entropy": 6.384079170227051,
"epoch": 0.5383115909431699,
"grad_norm": 0.9765625,
"learning_rate": 0.0004978275005859333,
"loss": 6.2265,
"mean_token_accuracy": 0.14789662063121795,
"num_tokens": 10549483.0,
"step": 5385
},
{
"entropy": 6.47556619644165,
"epoch": 0.5388114160043984,
"grad_norm": 0.85546875,
"learning_rate": 0.0004978225502542313,
"loss": 6.34,
"mean_token_accuracy": 0.13866292610764502,
"num_tokens": 10560464.0,
"step": 5390
},
{
"entropy": 6.476946926116943,
"epoch": 0.539311241065627,
"grad_norm": 0.9921875,
"learning_rate": 0.0004978175943163712,
"loss": 6.2428,
"mean_token_accuracy": 0.1488696701824665,
"num_tokens": 10568660.0,
"step": 5395
},
{
"entropy": 6.466958045959473,
"epoch": 0.5398110661268556,
"grad_norm": 0.9921875,
"learning_rate": 0.0004978126327724777,
"loss": 6.289,
"mean_token_accuracy": 0.14666010886430741,
"num_tokens": 10577304.0,
"step": 5400
},
{
"entropy": 6.428526496887207,
"epoch": 0.5403108911880842,
"grad_norm": 0.90234375,
"learning_rate": 0.0004978076656226758,
"loss": 6.1838,
"mean_token_accuracy": 0.15669340565800666,
"num_tokens": 10587591.0,
"step": 5405
},
{
"entropy": 6.43843789100647,
"epoch": 0.5408107162493128,
"grad_norm": 1.46875,
"learning_rate": 0.0004978026928670902,
"loss": 6.1286,
"mean_token_accuracy": 0.14771923720836638,
"num_tokens": 10598591.0,
"step": 5410
},
{
"entropy": 6.400979900360108,
"epoch": 0.5413105413105413,
"grad_norm": 0.84375,
"learning_rate": 0.0004977977145058463,
"loss": 6.1542,
"mean_token_accuracy": 0.14819953516125678,
"num_tokens": 10608842.0,
"step": 5415
},
{
"entropy": 6.435693693161011,
"epoch": 0.5418103663717699,
"grad_norm": 0.875,
"learning_rate": 0.0004977927305390693,
"loss": 6.2422,
"mean_token_accuracy": 0.15102696046233177,
"num_tokens": 10618812.0,
"step": 5420
},
{
"entropy": 6.4845438480377195,
"epoch": 0.5423101914329984,
"grad_norm": 1.1953125,
"learning_rate": 0.0004977877409668845,
"loss": 6.3625,
"mean_token_accuracy": 0.14045077338814735,
"num_tokens": 10629244.0,
"step": 5425
},
{
"entropy": 6.497104930877685,
"epoch": 0.542810016494227,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977827457894174,
"loss": 6.3105,
"mean_token_accuracy": 0.13931018710136414,
"num_tokens": 10638931.0,
"step": 5430
},
{
"entropy": 6.44123330116272,
"epoch": 0.5433098415554556,
"grad_norm": 1.0078125,
"learning_rate": 0.0004977777450067937,
"loss": 6.1622,
"mean_token_accuracy": 0.1503096863627434,
"num_tokens": 10648754.0,
"step": 5435
},
{
"entropy": 6.368002653121948,
"epoch": 0.5438096666166842,
"grad_norm": 0.9453125,
"learning_rate": 0.0004977727386191393,
"loss": 6.1405,
"mean_token_accuracy": 0.15083228647708893,
"num_tokens": 10657716.0,
"step": 5440
},
{
"entropy": 6.392760848999023,
"epoch": 0.5443094916779128,
"grad_norm": 0.89453125,
"learning_rate": 0.0004977677266265802,
"loss": 6.1741,
"mean_token_accuracy": 0.14957669675350188,
"num_tokens": 10667621.0,
"step": 5445
},
{
"entropy": 6.423801469802856,
"epoch": 0.5448093167391413,
"grad_norm": 0.95703125,
"learning_rate": 0.0004977627090292424,
"loss": 6.1616,
"mean_token_accuracy": 0.1466075710952282,
"num_tokens": 10676640.0,
"step": 5450
},
{
"entropy": 6.386547374725342,
"epoch": 0.5453091418003698,
"grad_norm": 1.015625,
"learning_rate": 0.0004977576858272521,
"loss": 6.2624,
"mean_token_accuracy": 0.14105069115757943,
"num_tokens": 10687034.0,
"step": 5455
},
{
"entropy": 6.424188661575317,
"epoch": 0.5458089668615984,
"grad_norm": 0.93359375,
"learning_rate": 0.0004977526570207358,
"loss": 6.1505,
"mean_token_accuracy": 0.1486357644200325,
"num_tokens": 10696234.0,
"step": 5460
},
{
"entropy": 6.357306289672851,
"epoch": 0.546308791922827,
"grad_norm": 0.9921875,
"learning_rate": 0.0004977476226098199,
"loss": 6.174,
"mean_token_accuracy": 0.15825203508138658,
"num_tokens": 10705733.0,
"step": 5465
},
{
"entropy": 6.452626752853393,
"epoch": 0.5468086169840556,
"grad_norm": 0.90234375,
"learning_rate": 0.0004977425825946312,
"loss": 6.2025,
"mean_token_accuracy": 0.1456859290599823,
"num_tokens": 10714628.0,
"step": 5470
},
{
"entropy": 6.504712104797363,
"epoch": 0.5473084420452842,
"grad_norm": 0.96484375,
"learning_rate": 0.0004977375369752964,
"loss": 6.3574,
"mean_token_accuracy": 0.1356508083641529,
"num_tokens": 10726120.0,
"step": 5475
},
{
"entropy": 6.428026485443115,
"epoch": 0.5478082671065128,
"grad_norm": 0.89453125,
"learning_rate": 0.0004977324857519425,
"loss": 6.2019,
"mean_token_accuracy": 0.1466704823076725,
"num_tokens": 10735976.0,
"step": 5480
},
{
"entropy": 6.4640013694763185,
"epoch": 0.5483080921677413,
"grad_norm": 0.91015625,
"learning_rate": 0.0004977274289246967,
"loss": 6.2781,
"mean_token_accuracy": 0.1415743499994278,
"num_tokens": 10745730.0,
"step": 5485
},
{
"entropy": 6.442827749252319,
"epoch": 0.5488079172289698,
"grad_norm": 0.8203125,
"learning_rate": 0.0004977223664936859,
"loss": 6.2149,
"mean_token_accuracy": 0.1492815963923931,
"num_tokens": 10755435.0,
"step": 5490
},
{
"entropy": 6.371844530105591,
"epoch": 0.5493077422901984,
"grad_norm": 0.90625,
"learning_rate": 0.0004977172984590378,
"loss": 6.1079,
"mean_token_accuracy": 0.1523657664656639,
"num_tokens": 10765483.0,
"step": 5495
},
{
"entropy": 6.380004167556763,
"epoch": 0.549807567351427,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977122248208798,
"loss": 6.1937,
"mean_token_accuracy": 0.15215383172035218,
"num_tokens": 10775034.0,
"step": 5500
},
{
"entropy": 6.3309595584869385,
"epoch": 0.5503073924126556,
"grad_norm": 0.87109375,
"learning_rate": 0.0004977071455793394,
"loss": 6.1585,
"mean_token_accuracy": 0.15155598968267442,
"num_tokens": 10784799.0,
"step": 5505
},
{
"entropy": 6.428274869918823,
"epoch": 0.5508072174738842,
"grad_norm": 0.890625,
"learning_rate": 0.0004977020607345445,
"loss": 6.2542,
"mean_token_accuracy": 0.13817675560712814,
"num_tokens": 10794907.0,
"step": 5510
},
{
"entropy": 6.428401327133178,
"epoch": 0.5513070425351128,
"grad_norm": 0.9140625,
"learning_rate": 0.0004976969702866231,
"loss": 6.1521,
"mean_token_accuracy": 0.15460718870162965,
"num_tokens": 10804374.0,
"step": 5515
},
{
"entropy": 6.422916030883789,
"epoch": 0.5518068675963412,
"grad_norm": 0.99609375,
"learning_rate": 0.0004976918742357033,
"loss": 6.1853,
"mean_token_accuracy": 0.14841345995664595,
"num_tokens": 10814245.0,
"step": 5520
},
{
"entropy": 6.369015789031982,
"epoch": 0.5523066926575698,
"grad_norm": 0.86328125,
"learning_rate": 0.0004976867725819133,
"loss": 6.1965,
"mean_token_accuracy": 0.15178618505597113,
"num_tokens": 10824667.0,
"step": 5525
},
{
"entropy": 6.395843553543091,
"epoch": 0.5528065177187984,
"grad_norm": 0.95703125,
"learning_rate": 0.0004976816653253813,
"loss": 6.172,
"mean_token_accuracy": 0.14885421246290206,
"num_tokens": 10834890.0,
"step": 5530
},
{
"entropy": 6.408791017532349,
"epoch": 0.553306342780027,
"grad_norm": 0.87109375,
"learning_rate": 0.0004976765524662359,
"loss": 6.1477,
"mean_token_accuracy": 0.15529358461499215,
"num_tokens": 10844751.0,
"step": 5535
},
{
"entropy": 6.479263210296631,
"epoch": 0.5538061678412556,
"grad_norm": 1.0,
"learning_rate": 0.0004976714340046058,
"loss": 6.2688,
"mean_token_accuracy": 0.1512459084391594,
"num_tokens": 10854113.0,
"step": 5540
},
{
"entropy": 6.38036961555481,
"epoch": 0.5543059929024842,
"grad_norm": 1.0234375,
"learning_rate": 0.0004976663099406196,
"loss": 6.1083,
"mean_token_accuracy": 0.1507673144340515,
"num_tokens": 10863468.0,
"step": 5545
},
{
"entropy": 6.417061519622803,
"epoch": 0.5548058179637128,
"grad_norm": 1.0234375,
"learning_rate": 0.0004976611802744066,
"loss": 6.2324,
"mean_token_accuracy": 0.1521667256951332,
"num_tokens": 10873381.0,
"step": 5550
},
{
"entropy": 6.4708188533782955,
"epoch": 0.5553056430249412,
"grad_norm": 0.984375,
"learning_rate": 0.0004976560450060954,
"loss": 6.1443,
"mean_token_accuracy": 0.15245285779237747,
"num_tokens": 10883457.0,
"step": 5555
},
{
"entropy": 6.2844969749450685,
"epoch": 0.5558054680861698,
"grad_norm": 0.90625,
"learning_rate": 0.0004976509041358154,
"loss": 6.1316,
"mean_token_accuracy": 0.15327222198247908,
"num_tokens": 10892822.0,
"step": 5560
},
{
"entropy": 6.376227521896363,
"epoch": 0.5563052931473984,
"grad_norm": 1.0234375,
"learning_rate": 0.000497645757663696,
"loss": 6.0839,
"mean_token_accuracy": 0.16154446303844452,
"num_tokens": 10901935.0,
"step": 5565
},
{
"entropy": 6.347236442565918,
"epoch": 0.556805118208627,
"grad_norm": 0.94140625,
"learning_rate": 0.0004976406055898665,
"loss": 6.1838,
"mean_token_accuracy": 0.14250542744994163,
"num_tokens": 10911863.0,
"step": 5570
},
{
"entropy": 6.374010705947876,
"epoch": 0.5573049432698556,
"grad_norm": 0.98828125,
"learning_rate": 0.0004976354479144568,
"loss": 6.1622,
"mean_token_accuracy": 0.14557452499866486,
"num_tokens": 10921196.0,
"step": 5575
},
{
"entropy": 6.435430908203125,
"epoch": 0.5578047683310842,
"grad_norm": 0.9609375,
"learning_rate": 0.0004976302846375965,
"loss": 6.132,
"mean_token_accuracy": 0.14937174171209336,
"num_tokens": 10931374.0,
"step": 5580
},
{
"entropy": 6.522235774993897,
"epoch": 0.5583045933923126,
"grad_norm": 0.91796875,
"learning_rate": 0.0004976251157594155,
"loss": 6.2753,
"mean_token_accuracy": 0.14256571829319,
"num_tokens": 10941002.0,
"step": 5585
},
{
"entropy": 6.445342969894409,
"epoch": 0.5588044184535412,
"grad_norm": 0.97265625,
"learning_rate": 0.0004976199412800437,
"loss": 6.2148,
"mean_token_accuracy": 0.1462072491645813,
"num_tokens": 10950602.0,
"step": 5590
},
{
"entropy": 6.382048845291138,
"epoch": 0.5593042435147698,
"grad_norm": 0.98046875,
"learning_rate": 0.0004976147611996118,
"loss": 6.1817,
"mean_token_accuracy": 0.14778511226177216,
"num_tokens": 10960363.0,
"step": 5595
},
{
"entropy": 6.43527193069458,
"epoch": 0.5598040685759984,
"grad_norm": 0.9453125,
"learning_rate": 0.0004976095755182496,
"loss": 6.2752,
"mean_token_accuracy": 0.14148695692420005,
"num_tokens": 10970179.0,
"step": 5600
},
{
"entropy": 6.3876708984375,
"epoch": 0.560303893637227,
"grad_norm": 0.92578125,
"learning_rate": 0.0004976043842360877,
"loss": 6.1898,
"mean_token_accuracy": 0.14226483702659606,
"num_tokens": 10980446.0,
"step": 5605
},
{
"entropy": 6.412945365905761,
"epoch": 0.5608037186984556,
"grad_norm": 0.91015625,
"learning_rate": 0.000497599187353257,
"loss": 6.2172,
"mean_token_accuracy": 0.143201994150877,
"num_tokens": 10990883.0,
"step": 5610
},
{
"entropy": 6.466598033905029,
"epoch": 0.5613035437596842,
"grad_norm": 0.8515625,
"learning_rate": 0.0004975939848698879,
"loss": 6.1145,
"mean_token_accuracy": 0.15407094955444336,
"num_tokens": 11000956.0,
"step": 5615
},
{
"entropy": 6.383142566680908,
"epoch": 0.5618033688209126,
"grad_norm": 0.88671875,
"learning_rate": 0.0004975887767861113,
"loss": 6.1915,
"mean_token_accuracy": 0.1499265745282173,
"num_tokens": 11010873.0,
"step": 5620
},
{
"entropy": 6.400553226470947,
"epoch": 0.5623031938821412,
"grad_norm": 0.953125,
"learning_rate": 0.0004975835631020585,
"loss": 6.1561,
"mean_token_accuracy": 0.1501181647181511,
"num_tokens": 11020192.0,
"step": 5625
},
{
"entropy": 6.396825790405273,
"epoch": 0.5628030189433698,
"grad_norm": 0.92578125,
"learning_rate": 0.0004975783438178605,
"loss": 6.1791,
"mean_token_accuracy": 0.14332359358668328,
"num_tokens": 11030466.0,
"step": 5630
},
{
"entropy": 6.38042106628418,
"epoch": 0.5633028440045984,
"grad_norm": 0.84765625,
"learning_rate": 0.0004975731189336487,
"loss": 6.1524,
"mean_token_accuracy": 0.15266245305538179,
"num_tokens": 11040309.0,
"step": 5635
},
{
"entropy": 6.402611923217774,
"epoch": 0.563802669065827,
"grad_norm": 0.9140625,
"learning_rate": 0.0004975678884495544,
"loss": 6.2249,
"mean_token_accuracy": 0.14522777497768402,
"num_tokens": 11049560.0,
"step": 5640
},
{
"entropy": 6.489155292510986,
"epoch": 0.5643024941270556,
"grad_norm": 0.95703125,
"learning_rate": 0.0004975626523657093,
"loss": 6.3017,
"mean_token_accuracy": 0.13924765810370446,
"num_tokens": 11059352.0,
"step": 5645
},
{
"entropy": 6.469867610931397,
"epoch": 0.564802319188284,
"grad_norm": 0.84375,
"learning_rate": 0.0004975574106822451,
"loss": 6.2131,
"mean_token_accuracy": 0.15009221136569978,
"num_tokens": 11069159.0,
"step": 5650
},
{
"entropy": 6.420383310317993,
"epoch": 0.5653021442495126,
"grad_norm": 0.93359375,
"learning_rate": 0.0004975521633992938,
"loss": 6.2864,
"mean_token_accuracy": 0.14288507103919984,
"num_tokens": 11078960.0,
"step": 5655
},
{
"entropy": 6.345410537719727,
"epoch": 0.5658019693107412,
"grad_norm": 0.796875,
"learning_rate": 0.0004975469105169873,
"loss": 6.1235,
"mean_token_accuracy": 0.15768041908740998,
"num_tokens": 11089408.0,
"step": 5660
},
{
"entropy": 6.484627294540405,
"epoch": 0.5663017943719698,
"grad_norm": 1.03125,
"learning_rate": 0.0004975416520354579,
"loss": 6.2816,
"mean_token_accuracy": 0.14217357113957405,
"num_tokens": 11099485.0,
"step": 5665
},
{
"entropy": 6.472567987442017,
"epoch": 0.5668016194331984,
"grad_norm": 0.8984375,
"learning_rate": 0.0004975363879548377,
"loss": 6.2813,
"mean_token_accuracy": 0.15160842835903168,
"num_tokens": 11109568.0,
"step": 5670
},
{
"entropy": 6.539796590805054,
"epoch": 0.567301444494427,
"grad_norm": 0.93359375,
"learning_rate": 0.0004975311182752592,
"loss": 6.3548,
"mean_token_accuracy": 0.13486458957195283,
"num_tokens": 11119433.0,
"step": 5675
},
{
"entropy": 6.45374436378479,
"epoch": 0.5678012695556556,
"grad_norm": 0.96484375,
"learning_rate": 0.0004975258429968551,
"loss": 6.2547,
"mean_token_accuracy": 0.1414391130208969,
"num_tokens": 11130060.0,
"step": 5680
},
{
"entropy": 6.499431180953979,
"epoch": 0.568301094616884,
"grad_norm": 0.9765625,
"learning_rate": 0.0004975205621197581,
"loss": 6.322,
"mean_token_accuracy": 0.14436408877372742,
"num_tokens": 11140417.0,
"step": 5685
},
{
"entropy": 6.483169174194336,
"epoch": 0.5688009196781126,
"grad_norm": 0.89453125,
"learning_rate": 0.0004975152756441009,
"loss": 6.264,
"mean_token_accuracy": 0.1493012122809887,
"num_tokens": 11150830.0,
"step": 5690
},
{
"entropy": 6.367444610595703,
"epoch": 0.5693007447393412,
"grad_norm": 0.99609375,
"learning_rate": 0.0004975099835700166,
"loss": 6.0914,
"mean_token_accuracy": 0.15718345940113068,
"num_tokens": 11161057.0,
"step": 5695
},
{
"entropy": 6.362324380874634,
"epoch": 0.5698005698005698,
"grad_norm": 1.015625,
"learning_rate": 0.0004975046858976386,
"loss": 6.1687,
"mean_token_accuracy": 0.14591417461633682,
"num_tokens": 11170182.0,
"step": 5700
},
{
"entropy": 6.3738373756408695,
"epoch": 0.5703003948617984,
"grad_norm": 0.953125,
"learning_rate": 0.0004974993826270997,
"loss": 6.1104,
"mean_token_accuracy": 0.147789067029953,
"num_tokens": 11179962.0,
"step": 5705
},
{
"entropy": 6.406001186370849,
"epoch": 0.570800219923027,
"grad_norm": 1.0703125,
"learning_rate": 0.0004974940737585337,
"loss": 6.0505,
"mean_token_accuracy": 0.16331670731306075,
"num_tokens": 11189341.0,
"step": 5710
},
{
"entropy": 6.34999189376831,
"epoch": 0.5713000449842555,
"grad_norm": 1.0,
"learning_rate": 0.000497488759292074,
"loss": 6.1981,
"mean_token_accuracy": 0.1564978390932083,
"num_tokens": 11199264.0,
"step": 5715
},
{
"entropy": 6.320228910446167,
"epoch": 0.571799870045484,
"grad_norm": 0.8828125,
"learning_rate": 0.0004974834392278545,
"loss": 6.199,
"mean_token_accuracy": 0.15573666393756866,
"num_tokens": 11209257.0,
"step": 5720
},
{
"entropy": 6.453316259384155,
"epoch": 0.5722996951067126,
"grad_norm": 0.86328125,
"learning_rate": 0.0004974781135660088,
"loss": 6.255,
"mean_token_accuracy": 0.14868643283843994,
"num_tokens": 11218294.0,
"step": 5725
},
{
"entropy": 6.35547194480896,
"epoch": 0.5727995201679412,
"grad_norm": 0.9453125,
"learning_rate": 0.0004974727823066709,
"loss": 6.0812,
"mean_token_accuracy": 0.1533886343240738,
"num_tokens": 11227684.0,
"step": 5730
},
{
"entropy": 6.45539698600769,
"epoch": 0.5732993452291698,
"grad_norm": 0.9140625,
"learning_rate": 0.0004974674454499753,
"loss": 6.2986,
"mean_token_accuracy": 0.13858716040849686,
"num_tokens": 11237195.0,
"step": 5735
},
{
"entropy": 6.5260608196258545,
"epoch": 0.5737991702903984,
"grad_norm": 0.8515625,
"learning_rate": 0.0004974621029960559,
"loss": 6.2493,
"mean_token_accuracy": 0.14635457396507262,
"num_tokens": 11246871.0,
"step": 5740
},
{
"entropy": 6.434552001953125,
"epoch": 0.574298995351627,
"grad_norm": 1.078125,
"learning_rate": 0.0004974567549450472,
"loss": 6.1166,
"mean_token_accuracy": 0.15176171213388442,
"num_tokens": 11256298.0,
"step": 5745
},
{
"entropy": 6.430844736099243,
"epoch": 0.5747988204128555,
"grad_norm": 0.953125,
"learning_rate": 0.0004974514012970839,
"loss": 6.2957,
"mean_token_accuracy": 0.14615521505475043,
"num_tokens": 11265857.0,
"step": 5750
},
{
"entropy": 6.362429904937744,
"epoch": 0.575298645474084,
"grad_norm": 0.8515625,
"learning_rate": 0.0004974460420523005,
"loss": 6.0778,
"mean_token_accuracy": 0.15784539580345153,
"num_tokens": 11275406.0,
"step": 5755
},
{
"entropy": 6.41445484161377,
"epoch": 0.5757984705353126,
"grad_norm": 0.93359375,
"learning_rate": 0.0004974406772108319,
"loss": 6.0761,
"mean_token_accuracy": 0.15416193678975104,
"num_tokens": 11283910.0,
"step": 5760
},
{
"entropy": 6.401713132858276,
"epoch": 0.5762982955965412,
"grad_norm": 0.9140625,
"learning_rate": 0.0004974353067728132,
"loss": 6.1497,
"mean_token_accuracy": 0.14307589083909988,
"num_tokens": 11293808.0,
"step": 5765
},
{
"entropy": 6.411474323272705,
"epoch": 0.5767981206577698,
"grad_norm": 1.0078125,
"learning_rate": 0.0004974299307383794,
"loss": 6.1918,
"mean_token_accuracy": 0.1455647498369217,
"num_tokens": 11302519.0,
"step": 5770
},
{
"entropy": 6.461847305297852,
"epoch": 0.5772979457189984,
"grad_norm": 0.94921875,
"learning_rate": 0.0004974245491076656,
"loss": 6.2883,
"mean_token_accuracy": 0.14553347378969192,
"num_tokens": 11312609.0,
"step": 5775
},
{
"entropy": 6.460581731796265,
"epoch": 0.5777977707802269,
"grad_norm": 0.91015625,
"learning_rate": 0.0004974191618808075,
"loss": 6.2231,
"mean_token_accuracy": 0.14999218359589578,
"num_tokens": 11322471.0,
"step": 5780
},
{
"entropy": 6.489849090576172,
"epoch": 0.5782975958414555,
"grad_norm": 1.21875,
"learning_rate": 0.0004974137690579405,
"loss": 6.2189,
"mean_token_accuracy": 0.14874884039163588,
"num_tokens": 11333332.0,
"step": 5785
},
{
"entropy": 6.405481672286987,
"epoch": 0.578797420902684,
"grad_norm": 0.91015625,
"learning_rate": 0.0004974083706392004,
"loss": 6.2115,
"mean_token_accuracy": 0.14520230516791344,
"num_tokens": 11342952.0,
"step": 5790
},
{
"entropy": 6.403951978683471,
"epoch": 0.5792972459639126,
"grad_norm": 0.859375,
"learning_rate": 0.0004974029666247228,
"loss": 6.292,
"mean_token_accuracy": 0.14103597924113273,
"num_tokens": 11353509.0,
"step": 5795
},
{
"entropy": 6.534165239334106,
"epoch": 0.5797970710251412,
"grad_norm": 0.8828125,
"learning_rate": 0.0004973975570146438,
"loss": 6.2846,
"mean_token_accuracy": 0.14150195941329002,
"num_tokens": 11363065.0,
"step": 5800
},
{
"entropy": 6.369716548919678,
"epoch": 0.5802968960863698,
"grad_norm": 1.0390625,
"learning_rate": 0.0004973921418090997,
"loss": 6.1528,
"mean_token_accuracy": 0.1501327022910118,
"num_tokens": 11373720.0,
"step": 5805
},
{
"entropy": 6.3334290981292725,
"epoch": 0.5807967211475984,
"grad_norm": 0.93359375,
"learning_rate": 0.0004973867210082263,
"loss": 6.1558,
"mean_token_accuracy": 0.15207915008068085,
"num_tokens": 11384432.0,
"step": 5810
},
{
"entropy": 6.435679340362549,
"epoch": 0.5812965462088269,
"grad_norm": 0.9296875,
"learning_rate": 0.0004973812946121603,
"loss": 6.3084,
"mean_token_accuracy": 0.14357167333364487,
"num_tokens": 11394100.0,
"step": 5815
},
{
"entropy": 6.392176246643066,
"epoch": 0.5817963712700555,
"grad_norm": 0.97265625,
"learning_rate": 0.0004973758626210382,
"loss": 6.1979,
"mean_token_accuracy": 0.14371618926525115,
"num_tokens": 11403359.0,
"step": 5820
},
{
"entropy": 6.464417791366577,
"epoch": 0.582296196331284,
"grad_norm": 0.95703125,
"learning_rate": 0.0004973704250349965,
"loss": 6.2106,
"mean_token_accuracy": 0.1430139996111393,
"num_tokens": 11412921.0,
"step": 5825
},
{
"entropy": 6.407368803024292,
"epoch": 0.5827960213925126,
"grad_norm": 0.90625,
"learning_rate": 0.0004973649818541722,
"loss": 6.1908,
"mean_token_accuracy": 0.14863360822200775,
"num_tokens": 11422459.0,
"step": 5830
},
{
"entropy": 6.36591362953186,
"epoch": 0.5832958464537412,
"grad_norm": 0.98046875,
"learning_rate": 0.0004973595330787022,
"loss": 6.136,
"mean_token_accuracy": 0.14982437193393708,
"num_tokens": 11431517.0,
"step": 5835
},
{
"entropy": 6.445391035079956,
"epoch": 0.5837956715149698,
"grad_norm": 0.90625,
"learning_rate": 0.0004973540787087235,
"loss": 6.1754,
"mean_token_accuracy": 0.14735695570707322,
"num_tokens": 11440662.0,
"step": 5840
},
{
"entropy": 6.367056179046631,
"epoch": 0.5842954965761983,
"grad_norm": 0.96875,
"learning_rate": 0.0004973486187443734,
"loss": 6.0289,
"mean_token_accuracy": 0.16187193095684052,
"num_tokens": 11449762.0,
"step": 5845
},
{
"entropy": 6.292886066436767,
"epoch": 0.5847953216374269,
"grad_norm": 0.94921875,
"learning_rate": 0.0004973431531857893,
"loss": 6.0315,
"mean_token_accuracy": 0.15769076347351074,
"num_tokens": 11459963.0,
"step": 5850
},
{
"entropy": 6.4106189727783205,
"epoch": 0.5852951466986555,
"grad_norm": 0.9609375,
"learning_rate": 0.0004973376820331087,
"loss": 6.2113,
"mean_token_accuracy": 0.14407769367098808,
"num_tokens": 11469063.0,
"step": 5855
},
{
"entropy": 6.413491582870483,
"epoch": 0.585794971759884,
"grad_norm": 0.9609375,
"learning_rate": 0.0004973322052864692,
"loss": 6.1149,
"mean_token_accuracy": 0.15207016319036484,
"num_tokens": 11477651.0,
"step": 5860
},
{
"entropy": 6.439757919311523,
"epoch": 0.5862947968211126,
"grad_norm": 0.8984375,
"learning_rate": 0.0004973267229460087,
"loss": 6.3233,
"mean_token_accuracy": 0.13958670869469642,
"num_tokens": 11487248.0,
"step": 5865
},
{
"entropy": 6.4254437446594235,
"epoch": 0.5867946218823412,
"grad_norm": 0.9140625,
"learning_rate": 0.000497321235011865,
"loss": 6.2761,
"mean_token_accuracy": 0.1482052683830261,
"num_tokens": 11497066.0,
"step": 5870
},
{
"entropy": 6.509588432312012,
"epoch": 0.5872944469435698,
"grad_norm": 0.83203125,
"learning_rate": 0.0004973157414841763,
"loss": 6.2506,
"mean_token_accuracy": 0.14056255891919137,
"num_tokens": 11508490.0,
"step": 5875
},
{
"entropy": 6.468327188491822,
"epoch": 0.5877942720047983,
"grad_norm": 0.98046875,
"learning_rate": 0.0004973102423630806,
"loss": 6.2349,
"mean_token_accuracy": 0.14740702211856843,
"num_tokens": 11518851.0,
"step": 5880
},
{
"entropy": 6.401625251770019,
"epoch": 0.5882940970660269,
"grad_norm": 1.0234375,
"learning_rate": 0.0004973047376487166,
"loss": 6.1816,
"mean_token_accuracy": 0.150423264503479,
"num_tokens": 11529102.0,
"step": 5885
},
{
"entropy": 6.372255086898804,
"epoch": 0.5887939221272555,
"grad_norm": 0.98046875,
"learning_rate": 0.0004972992273412224,
"loss": 6.2182,
"mean_token_accuracy": 0.14426240026950837,
"num_tokens": 11540563.0,
"step": 5890
},
{
"entropy": 6.3703391551971436,
"epoch": 0.589293747188484,
"grad_norm": 0.9296875,
"learning_rate": 0.0004972937114407369,
"loss": 6.1403,
"mean_token_accuracy": 0.1549333855509758,
"num_tokens": 11549955.0,
"step": 5895
},
{
"entropy": 6.40195665359497,
"epoch": 0.5897935722497126,
"grad_norm": 0.91015625,
"learning_rate": 0.000497288189947399,
"loss": 6.1181,
"mean_token_accuracy": 0.1515316516160965,
"num_tokens": 11558932.0,
"step": 5900
},
{
"entropy": 6.308910465240478,
"epoch": 0.5902933973109412,
"grad_norm": 0.859375,
"learning_rate": 0.0004972826628613472,
"loss": 6.1271,
"mean_token_accuracy": 0.15501016229391099,
"num_tokens": 11568901.0,
"step": 5905
},
{
"entropy": 6.437039089202881,
"epoch": 0.5907932223721697,
"grad_norm": 0.87890625,
"learning_rate": 0.0004972771301827209,
"loss": 6.1912,
"mean_token_accuracy": 0.14727447926998138,
"num_tokens": 11579738.0,
"step": 5910
},
{
"entropy": 6.4059497833251955,
"epoch": 0.5912930474333983,
"grad_norm": 0.96875,
"learning_rate": 0.0004972715919116592,
"loss": 6.1518,
"mean_token_accuracy": 0.1520642265677452,
"num_tokens": 11590293.0,
"step": 5915
},
{
"entropy": 6.389673900604248,
"epoch": 0.5917928724946269,
"grad_norm": 0.8984375,
"learning_rate": 0.0004972660480483015,
"loss": 6.2351,
"mean_token_accuracy": 0.1434284321963787,
"num_tokens": 11599745.0,
"step": 5920
},
{
"entropy": 6.437257289886475,
"epoch": 0.5922926975558555,
"grad_norm": 0.87890625,
"learning_rate": 0.0004972604985927871,
"loss": 6.2199,
"mean_token_accuracy": 0.14112672805786133,
"num_tokens": 11609958.0,
"step": 5925
},
{
"entropy": 6.457437992095947,
"epoch": 0.592792522617084,
"grad_norm": 0.921875,
"learning_rate": 0.0004972549435452558,
"loss": 6.2125,
"mean_token_accuracy": 0.14297294169664382,
"num_tokens": 11619941.0,
"step": 5930
},
{
"entropy": 6.411074447631836,
"epoch": 0.5932923476783126,
"grad_norm": 0.91796875,
"learning_rate": 0.0004972493829058474,
"loss": 6.0894,
"mean_token_accuracy": 0.15717873275279998,
"num_tokens": 11630287.0,
"step": 5935
},
{
"entropy": 6.387608861923217,
"epoch": 0.5937921727395412,
"grad_norm": 0.9453125,
"learning_rate": 0.0004972438166747016,
"loss": 6.1421,
"mean_token_accuracy": 0.14942783191800119,
"num_tokens": 11639806.0,
"step": 5940
},
{
"entropy": 6.374695730209351,
"epoch": 0.5942919978007697,
"grad_norm": 0.87890625,
"learning_rate": 0.0004972382448519586,
"loss": 6.2379,
"mean_token_accuracy": 0.14599600732326506,
"num_tokens": 11650132.0,
"step": 5945
},
{
"entropy": 6.424087762832642,
"epoch": 0.5947918228619983,
"grad_norm": 0.97265625,
"learning_rate": 0.0004972326674377586,
"loss": 6.161,
"mean_token_accuracy": 0.14550729915499688,
"num_tokens": 11659670.0,
"step": 5950
},
{
"entropy": 6.427029752731324,
"epoch": 0.5952916479232269,
"grad_norm": 0.81640625,
"learning_rate": 0.0004972270844322418,
"loss": 6.1978,
"mean_token_accuracy": 0.1531231015920639,
"num_tokens": 11669998.0,
"step": 5955
},
{
"entropy": 6.402192211151123,
"epoch": 0.5957914729844555,
"grad_norm": 0.9609375,
"learning_rate": 0.0004972214958355488,
"loss": 6.2097,
"mean_token_accuracy": 0.14409659951925277,
"num_tokens": 11679052.0,
"step": 5960
},
{
"entropy": 6.398138427734375,
"epoch": 0.596291298045684,
"grad_norm": 1.03125,
"learning_rate": 0.0004972159016478201,
"loss": 6.1735,
"mean_token_accuracy": 0.1523020088672638,
"num_tokens": 11687957.0,
"step": 5965
},
{
"entropy": 6.407905435562133,
"epoch": 0.5967911231069126,
"grad_norm": 0.93359375,
"learning_rate": 0.0004972103018691965,
"loss": 6.1189,
"mean_token_accuracy": 0.15084125101566315,
"num_tokens": 11696922.0,
"step": 5970
},
{
"entropy": 6.357454299926758,
"epoch": 0.5972909481681411,
"grad_norm": 0.953125,
"learning_rate": 0.0004972046964998189,
"loss": 6.0937,
"mean_token_accuracy": 0.1565779283642769,
"num_tokens": 11706600.0,
"step": 5975
},
{
"entropy": 6.358139848709106,
"epoch": 0.5977907732293697,
"grad_norm": 1.0546875,
"learning_rate": 0.0004971990855398283,
"loss": 6.154,
"mean_token_accuracy": 0.1468999430537224,
"num_tokens": 11715933.0,
"step": 5980
},
{
"entropy": 6.39614429473877,
"epoch": 0.5982905982905983,
"grad_norm": 1.09375,
"learning_rate": 0.0004971934689893658,
"loss": 6.0974,
"mean_token_accuracy": 0.1552150011062622,
"num_tokens": 11724807.0,
"step": 5985
},
{
"entropy": 6.429772758483887,
"epoch": 0.5987904233518269,
"grad_norm": 1.1171875,
"learning_rate": 0.0004971878468485729,
"loss": 6.1776,
"mean_token_accuracy": 0.14787413328886032,
"num_tokens": 11734985.0,
"step": 5990
},
{
"entropy": 6.373316287994385,
"epoch": 0.5992902484130554,
"grad_norm": 1.0859375,
"learning_rate": 0.000497182219117591,
"loss": 6.1633,
"mean_token_accuracy": 0.1456233836710453,
"num_tokens": 11743418.0,
"step": 5995
},
{
"entropy": 6.3798729419708256,
"epoch": 0.599790073474284,
"grad_norm": 0.9453125,
"learning_rate": 0.0004971765857965615,
"loss": 6.102,
"mean_token_accuracy": 0.1504545032978058,
"num_tokens": 11752953.0,
"step": 6000
},
{
"epoch": 0.599790073474284,
"eval_entropy": 6.190445968030317,
"eval_loss": 6.201533794403076,
"eval_mean_token_accuracy": 0.15422710121331626,
"eval_num_tokens": 11752953.0,
"eval_runtime": 18.1743,
"eval_samples_per_second": 1708.18,
"eval_steps_per_second": 213.543,
"step": 6000
},
{
"entropy": 6.414371156692505,
"epoch": 0.6002898985355126,
"grad_norm": 0.94921875,
"learning_rate": 0.0004971709468856263,
"loss": 6.2484,
"mean_token_accuracy": 0.1458593115210533,
"num_tokens": 11763260.0,
"step": 6005
},
{
"entropy": 6.304040479660034,
"epoch": 0.6007897235967411,
"grad_norm": 1.171875,
"learning_rate": 0.0004971653023849274,
"loss": 6.0209,
"mean_token_accuracy": 0.16909098103642464,
"num_tokens": 11773255.0,
"step": 6010
},
{
"entropy": 6.384859848022461,
"epoch": 0.6012895486579697,
"grad_norm": 0.8984375,
"learning_rate": 0.0004971596522946065,
"loss": 6.1636,
"mean_token_accuracy": 0.14832017049193383,
"num_tokens": 11783443.0,
"step": 6015
},
{
"entropy": 6.317033958435059,
"epoch": 0.6017893737191983,
"grad_norm": 0.97265625,
"learning_rate": 0.0004971539966148059,
"loss": 6.1204,
"mean_token_accuracy": 0.15199731811881065,
"num_tokens": 11791932.0,
"step": 6020
},
{
"entropy": 6.357777881622314,
"epoch": 0.6022891987804269,
"grad_norm": 0.93359375,
"learning_rate": 0.0004971483353456681,
"loss": 6.0974,
"mean_token_accuracy": 0.15433032661676407,
"num_tokens": 11801420.0,
"step": 6025
},
{
"entropy": 6.431672286987305,
"epoch": 0.6027890238416554,
"grad_norm": 0.8984375,
"learning_rate": 0.0004971426684873352,
"loss": 6.2544,
"mean_token_accuracy": 0.14638083875179292,
"num_tokens": 11810977.0,
"step": 6030
},
{
"entropy": 6.361601877212524,
"epoch": 0.603288848902884,
"grad_norm": 1.0078125,
"learning_rate": 0.0004971369960399499,
"loss": 6.0536,
"mean_token_accuracy": 0.1578028455376625,
"num_tokens": 11821016.0,
"step": 6035
},
{
"entropy": 6.3685469150543215,
"epoch": 0.6037886739641125,
"grad_norm": 0.921875,
"learning_rate": 0.000497131318003655,
"loss": 6.2039,
"mean_token_accuracy": 0.14420628547668457,
"num_tokens": 11831157.0,
"step": 6040
},
{
"entropy": 6.346438884735107,
"epoch": 0.6042884990253411,
"grad_norm": 0.96484375,
"learning_rate": 0.0004971256343785932,
"loss": 6.1101,
"mean_token_accuracy": 0.15514256656169892,
"num_tokens": 11840319.0,
"step": 6045
},
{
"entropy": 6.4193297863006595,
"epoch": 0.6047883240865697,
"grad_norm": 0.91796875,
"learning_rate": 0.0004971199451649078,
"loss": 6.2462,
"mean_token_accuracy": 0.1454211428761482,
"num_tokens": 11850808.0,
"step": 6050
},
{
"entropy": 6.488301753997803,
"epoch": 0.6052881491477983,
"grad_norm": 0.9296875,
"learning_rate": 0.0004971142503627415,
"loss": 6.2771,
"mean_token_accuracy": 0.14777473211288453,
"num_tokens": 11861123.0,
"step": 6055
},
{
"entropy": 6.30822229385376,
"epoch": 0.6057879742090269,
"grad_norm": 0.87109375,
"learning_rate": 0.000497108549972238,
"loss": 6.0222,
"mean_token_accuracy": 0.16141136288642882,
"num_tokens": 11870308.0,
"step": 6060
},
{
"entropy": 6.443945360183716,
"epoch": 0.6062877992702554,
"grad_norm": 0.93359375,
"learning_rate": 0.0004971028439935404,
"loss": 6.1644,
"mean_token_accuracy": 0.1443210080265999,
"num_tokens": 11880254.0,
"step": 6065
},
{
"entropy": 6.410022115707397,
"epoch": 0.606787624331484,
"grad_norm": 0.98828125,
"learning_rate": 0.0004970971324267925,
"loss": 6.1608,
"mean_token_accuracy": 0.15488466769456863,
"num_tokens": 11891272.0,
"step": 6070
},
{
"entropy": 6.310713243484497,
"epoch": 0.6072874493927125,
"grad_norm": 0.9453125,
"learning_rate": 0.0004970914152721378,
"loss": 6.1808,
"mean_token_accuracy": 0.14843776524066926,
"num_tokens": 11901176.0,
"step": 6075
},
{
"entropy": 6.418321180343628,
"epoch": 0.6077872744539411,
"grad_norm": 0.984375,
"learning_rate": 0.0004970856925297204,
"loss": 6.1982,
"mean_token_accuracy": 0.14708398804068565,
"num_tokens": 11910820.0,
"step": 6080
},
{
"entropy": 6.421906614303589,
"epoch": 0.6082870995151697,
"grad_norm": 1.0546875,
"learning_rate": 0.000497079964199684,
"loss": 6.1611,
"mean_token_accuracy": 0.1443112276494503,
"num_tokens": 11920458.0,
"step": 6085
},
{
"entropy": 6.283801746368408,
"epoch": 0.6087869245763983,
"grad_norm": 0.88671875,
"learning_rate": 0.0004970742302821729,
"loss": 5.9987,
"mean_token_accuracy": 0.15824998021125794,
"num_tokens": 11931150.0,
"step": 6090
},
{
"entropy": 6.341992807388306,
"epoch": 0.6092867496376269,
"grad_norm": 0.90625,
"learning_rate": 0.0004970684907773312,
"loss": 6.12,
"mean_token_accuracy": 0.15010675191879272,
"num_tokens": 11939960.0,
"step": 6095
},
{
"entropy": 6.450232076644897,
"epoch": 0.6097865746988554,
"grad_norm": 1.0703125,
"learning_rate": 0.0004970627456853036,
"loss": 6.2644,
"mean_token_accuracy": 0.14708733037114144,
"num_tokens": 11950781.0,
"step": 6100
},
{
"entropy": 6.437028884887695,
"epoch": 0.6102863997600839,
"grad_norm": 1.0078125,
"learning_rate": 0.0004970569950062343,
"loss": 6.1299,
"mean_token_accuracy": 0.152546326816082,
"num_tokens": 11960489.0,
"step": 6105
},
{
"entropy": 6.409123706817627,
"epoch": 0.6107862248213125,
"grad_norm": 0.87109375,
"learning_rate": 0.0004970512387402682,
"loss": 6.1653,
"mean_token_accuracy": 0.15546404644846917,
"num_tokens": 11970651.0,
"step": 6110
},
{
"entropy": 6.289830303192138,
"epoch": 0.6112860498825411,
"grad_norm": 0.90625,
"learning_rate": 0.0004970454768875501,
"loss": 6.1354,
"mean_token_accuracy": 0.15251707658171654,
"num_tokens": 11980349.0,
"step": 6115
},
{
"entropy": 6.353547525405884,
"epoch": 0.6117858749437697,
"grad_norm": 0.953125,
"learning_rate": 0.0004970397094482249,
"loss": 6.1631,
"mean_token_accuracy": 0.15197541415691376,
"num_tokens": 11990076.0,
"step": 6120
},
{
"entropy": 6.347495079040527,
"epoch": 0.6122857000049983,
"grad_norm": 0.9140625,
"learning_rate": 0.0004970339364224378,
"loss": 6.0802,
"mean_token_accuracy": 0.16009331941604615,
"num_tokens": 11999515.0,
"step": 6125
},
{
"entropy": 6.420997190475464,
"epoch": 0.6127855250662269,
"grad_norm": 0.8515625,
"learning_rate": 0.0004970281578103339,
"loss": 6.1992,
"mean_token_accuracy": 0.14599146842956542,
"num_tokens": 12009224.0,
"step": 6130
},
{
"entropy": 6.327088117599487,
"epoch": 0.6132853501274554,
"grad_norm": 0.8671875,
"learning_rate": 0.0004970223736120587,
"loss": 6.0934,
"mean_token_accuracy": 0.15674970299005508,
"num_tokens": 12019295.0,
"step": 6135
},
{
"entropy": 6.397483015060425,
"epoch": 0.6137851751886839,
"grad_norm": 0.88671875,
"learning_rate": 0.0004970165838277578,
"loss": 6.1744,
"mean_token_accuracy": 0.14948792681097983,
"num_tokens": 12029130.0,
"step": 6140
},
{
"entropy": 6.382640838623047,
"epoch": 0.6142850002499125,
"grad_norm": 0.91796875,
"learning_rate": 0.0004970107884575767,
"loss": 6.1415,
"mean_token_accuracy": 0.1544598527252674,
"num_tokens": 12038676.0,
"step": 6145
},
{
"entropy": 6.411570501327515,
"epoch": 0.6147848253111411,
"grad_norm": 0.9609375,
"learning_rate": 0.0004970049875016613,
"loss": 6.0955,
"mean_token_accuracy": 0.16193794459104538,
"num_tokens": 12048089.0,
"step": 6150
},
{
"entropy": 6.379500150680542,
"epoch": 0.6152846503723697,
"grad_norm": 0.890625,
"learning_rate": 0.0004969991809601576,
"loss": 6.1842,
"mean_token_accuracy": 0.1491437926888466,
"num_tokens": 12059531.0,
"step": 6155
},
{
"entropy": 6.412176322937012,
"epoch": 0.6157844754335983,
"grad_norm": 0.8984375,
"learning_rate": 0.0004969933688332115,
"loss": 6.1891,
"mean_token_accuracy": 0.14602637887001038,
"num_tokens": 12070103.0,
"step": 6160
},
{
"entropy": 6.351621532440186,
"epoch": 0.6162843004948269,
"grad_norm": 1.0234375,
"learning_rate": 0.0004969875511209694,
"loss": 6.167,
"mean_token_accuracy": 0.14610393717885017,
"num_tokens": 12078481.0,
"step": 6165
},
{
"entropy": 6.401747369766236,
"epoch": 0.6167841255560553,
"grad_norm": 0.84765625,
"learning_rate": 0.0004969817278235777,
"loss": 6.2561,
"mean_token_accuracy": 0.1471801958978176,
"num_tokens": 12088875.0,
"step": 6170
},
{
"entropy": 6.435576868057251,
"epoch": 0.6172839506172839,
"grad_norm": 0.87890625,
"learning_rate": 0.0004969758989411829,
"loss": 6.1621,
"mean_token_accuracy": 0.14885308817029,
"num_tokens": 12099769.0,
"step": 6175
},
{
"entropy": 6.451336288452149,
"epoch": 0.6177837756785125,
"grad_norm": 0.91796875,
"learning_rate": 0.0004969700644739315,
"loss": 6.1266,
"mean_token_accuracy": 0.15046553462743759,
"num_tokens": 12109334.0,
"step": 6180
},
{
"entropy": 6.3981605052948,
"epoch": 0.6182836007397411,
"grad_norm": 0.875,
"learning_rate": 0.0004969642244219705,
"loss": 6.1376,
"mean_token_accuracy": 0.1497718021273613,
"num_tokens": 12119017.0,
"step": 6185
},
{
"entropy": 6.281883335113525,
"epoch": 0.6187834258009697,
"grad_norm": 0.9453125,
"learning_rate": 0.0004969583787854467,
"loss": 6.0423,
"mean_token_accuracy": 0.16523782461881636,
"num_tokens": 12128002.0,
"step": 6190
},
{
"entropy": 6.386834239959716,
"epoch": 0.6192832508621983,
"grad_norm": 0.921875,
"learning_rate": 0.0004969525275645071,
"loss": 6.1505,
"mean_token_accuracy": 0.14462432786822318,
"num_tokens": 12137802.0,
"step": 6195
},
{
"entropy": 6.300003099441528,
"epoch": 0.6197830759234269,
"grad_norm": 1.0703125,
"learning_rate": 0.0004969466707592991,
"loss": 6.1093,
"mean_token_accuracy": 0.15344394445419313,
"num_tokens": 12147270.0,
"step": 6200
},
{
"entropy": 6.36989016532898,
"epoch": 0.6202829009846553,
"grad_norm": 0.9609375,
"learning_rate": 0.00049694080836997,
"loss": 6.1736,
"mean_token_accuracy": 0.15202322900295256,
"num_tokens": 12157358.0,
"step": 6205
},
{
"entropy": 6.441101694107056,
"epoch": 0.6207827260458839,
"grad_norm": 1.0625,
"learning_rate": 0.0004969349403966672,
"loss": 6.2169,
"mean_token_accuracy": 0.1466719038784504,
"num_tokens": 12166159.0,
"step": 6210
},
{
"entropy": 6.377940511703491,
"epoch": 0.6212825511071125,
"grad_norm": 0.90234375,
"learning_rate": 0.0004969290668395385,
"loss": 6.1817,
"mean_token_accuracy": 0.14901216626167296,
"num_tokens": 12176405.0,
"step": 6215
},
{
"entropy": 6.4397608757019045,
"epoch": 0.6217823761683411,
"grad_norm": 1.0234375,
"learning_rate": 0.0004969231876987315,
"loss": 6.2399,
"mean_token_accuracy": 0.14596698209643363,
"num_tokens": 12186021.0,
"step": 6220
},
{
"entropy": 6.343081760406494,
"epoch": 0.6222822012295697,
"grad_norm": 0.94921875,
"learning_rate": 0.0004969173029743942,
"loss": 6.1053,
"mean_token_accuracy": 0.15684943795204162,
"num_tokens": 12195222.0,
"step": 6225
},
{
"entropy": 6.349526119232178,
"epoch": 0.6227820262907983,
"grad_norm": 0.953125,
"learning_rate": 0.0004969114126666746,
"loss": 6.0835,
"mean_token_accuracy": 0.15733208805322646,
"num_tokens": 12205443.0,
"step": 6230
},
{
"entropy": 6.434093904495239,
"epoch": 0.6232818513520267,
"grad_norm": 1.0,
"learning_rate": 0.0004969055167757211,
"loss": 6.2823,
"mean_token_accuracy": 0.1480371654033661,
"num_tokens": 12215181.0,
"step": 6235
},
{
"entropy": 6.390381383895874,
"epoch": 0.6237816764132553,
"grad_norm": 0.9140625,
"learning_rate": 0.0004968996153016819,
"loss": 6.1581,
"mean_token_accuracy": 0.150498066842556,
"num_tokens": 12224327.0,
"step": 6240
},
{
"entropy": 6.394075202941894,
"epoch": 0.6242815014744839,
"grad_norm": 0.90625,
"learning_rate": 0.0004968937082447054,
"loss": 6.1761,
"mean_token_accuracy": 0.14878209233283995,
"num_tokens": 12234962.0,
"step": 6245
},
{
"entropy": 6.42145185470581,
"epoch": 0.6247813265357125,
"grad_norm": 1.0234375,
"learning_rate": 0.0004968877956049403,
"loss": 6.2465,
"mean_token_accuracy": 0.15092108845710756,
"num_tokens": 12245883.0,
"step": 6250
},
{
"entropy": 6.448516130447388,
"epoch": 0.6252811515969411,
"grad_norm": 1.03125,
"learning_rate": 0.0004968818773825355,
"loss": 6.2182,
"mean_token_accuracy": 0.14223939478397368,
"num_tokens": 12254412.0,
"step": 6255
},
{
"entropy": 6.299964761734008,
"epoch": 0.6257809766581697,
"grad_norm": 0.8359375,
"learning_rate": 0.0004968759535776396,
"loss": 6.0931,
"mean_token_accuracy": 0.15128926932811737,
"num_tokens": 12264894.0,
"step": 6260
},
{
"entropy": 6.374241733551026,
"epoch": 0.6262808017193983,
"grad_norm": 0.90625,
"learning_rate": 0.0004968700241904018,
"loss": 6.2001,
"mean_token_accuracy": 0.1505408175289631,
"num_tokens": 12274557.0,
"step": 6265
},
{
"entropy": 6.40683422088623,
"epoch": 0.6267806267806267,
"grad_norm": 1.0546875,
"learning_rate": 0.0004968640892209714,
"loss": 6.0088,
"mean_token_accuracy": 0.15743694752454757,
"num_tokens": 12283696.0,
"step": 6270
},
{
"entropy": 6.3402036190032955,
"epoch": 0.6272804518418553,
"grad_norm": 0.96875,
"learning_rate": 0.0004968581486694975,
"loss": 6.1202,
"mean_token_accuracy": 0.15782214179635048,
"num_tokens": 12292981.0,
"step": 6275
},
{
"entropy": 6.3552894115448,
"epoch": 0.6277802769030839,
"grad_norm": 0.953125,
"learning_rate": 0.0004968522025361297,
"loss": 5.9951,
"mean_token_accuracy": 0.15552034676074983,
"num_tokens": 12302865.0,
"step": 6280
},
{
"entropy": 6.336669015884399,
"epoch": 0.6282801019643125,
"grad_norm": 1.0,
"learning_rate": 0.0004968462508210176,
"loss": 6.1545,
"mean_token_accuracy": 0.15222094357013702,
"num_tokens": 12313099.0,
"step": 6285
},
{
"entropy": 6.381136894226074,
"epoch": 0.6287799270255411,
"grad_norm": 0.8515625,
"learning_rate": 0.0004968402935243109,
"loss": 6.1941,
"mean_token_accuracy": 0.15020949989557267,
"num_tokens": 12324155.0,
"step": 6290
},
{
"entropy": 6.343426322937011,
"epoch": 0.6292797520867697,
"grad_norm": 0.9921875,
"learning_rate": 0.0004968343306461595,
"loss": 6.0233,
"mean_token_accuracy": 0.1567442983388901,
"num_tokens": 12334082.0,
"step": 6295
},
{
"entropy": 6.3027894496917725,
"epoch": 0.6297795771479981,
"grad_norm": 0.91015625,
"learning_rate": 0.0004968283621867134,
"loss": 6.1438,
"mean_token_accuracy": 0.15131822675466539,
"num_tokens": 12344673.0,
"step": 6300
},
{
"entropy": 6.371920156478882,
"epoch": 0.6302794022092267,
"grad_norm": 0.93359375,
"learning_rate": 0.0004968223881461227,
"loss": 6.1368,
"mean_token_accuracy": 0.14937038272619246,
"num_tokens": 12355149.0,
"step": 6305
},
{
"entropy": 6.373556089401245,
"epoch": 0.6307792272704553,
"grad_norm": 0.98046875,
"learning_rate": 0.000496816408524538,
"loss": 6.0463,
"mean_token_accuracy": 0.15889935195446014,
"num_tokens": 12364043.0,
"step": 6310
},
{
"entropy": 6.367973566055298,
"epoch": 0.6312790523316839,
"grad_norm": 1.0625,
"learning_rate": 0.0004968104233221094,
"loss": 6.2408,
"mean_token_accuracy": 0.14706538245081902,
"num_tokens": 12373591.0,
"step": 6315
},
{
"entropy": 6.360639667510986,
"epoch": 0.6317788773929125,
"grad_norm": 0.8828125,
"learning_rate": 0.0004968044325389877,
"loss": 6.1734,
"mean_token_accuracy": 0.15179983973503114,
"num_tokens": 12383570.0,
"step": 6320
},
{
"entropy": 6.4673093318939205,
"epoch": 0.6322787024541411,
"grad_norm": 0.97265625,
"learning_rate": 0.0004967984361753235,
"loss": 6.1557,
"mean_token_accuracy": 0.15092750489711762,
"num_tokens": 12393682.0,
"step": 6325
},
{
"entropy": 6.418556165695191,
"epoch": 0.6327785275153697,
"grad_norm": 1.1953125,
"learning_rate": 0.0004967924342312677,
"loss": 6.0859,
"mean_token_accuracy": 0.15357855558395386,
"num_tokens": 12403478.0,
"step": 6330
},
{
"entropy": 6.325319099426269,
"epoch": 0.6332783525765981,
"grad_norm": 0.92578125,
"learning_rate": 0.0004967864267069713,
"loss": 6.185,
"mean_token_accuracy": 0.15216561108827592,
"num_tokens": 12412810.0,
"step": 6335
},
{
"entropy": 6.284311437606812,
"epoch": 0.6337781776378267,
"grad_norm": 0.86328125,
"learning_rate": 0.0004967804136025856,
"loss": 6.0813,
"mean_token_accuracy": 0.1570043072104454,
"num_tokens": 12422209.0,
"step": 6340
},
{
"entropy": 6.3864611148834225,
"epoch": 0.6342780026990553,
"grad_norm": 0.94140625,
"learning_rate": 0.0004967743949182617,
"loss": 6.1379,
"mean_token_accuracy": 0.15099855959415437,
"num_tokens": 12431569.0,
"step": 6345
},
{
"entropy": 6.308241939544677,
"epoch": 0.6347778277602839,
"grad_norm": 1.015625,
"learning_rate": 0.0004967683706541511,
"loss": 5.9946,
"mean_token_accuracy": 0.17168646901845933,
"num_tokens": 12441607.0,
"step": 6350
},
{
"entropy": 6.295763063430786,
"epoch": 0.6352776528215125,
"grad_norm": 0.9765625,
"learning_rate": 0.0004967623408104054,
"loss": 6.1351,
"mean_token_accuracy": 0.14483131766319274,
"num_tokens": 12451072.0,
"step": 6355
},
{
"entropy": 6.38209114074707,
"epoch": 0.6357774778827411,
"grad_norm": 0.92578125,
"learning_rate": 0.0004967563053871762,
"loss": 6.1505,
"mean_token_accuracy": 0.14997408762574196,
"num_tokens": 12460798.0,
"step": 6360
},
{
"entropy": 6.31462550163269,
"epoch": 0.6362773029439696,
"grad_norm": 0.98046875,
"learning_rate": 0.0004967502643846155,
"loss": 6.0818,
"mean_token_accuracy": 0.15518784523010254,
"num_tokens": 12471891.0,
"step": 6365
},
{
"entropy": 6.3016307830810545,
"epoch": 0.6367771280051981,
"grad_norm": 0.91015625,
"learning_rate": 0.000496744217802875,
"loss": 6.122,
"mean_token_accuracy": 0.15114265233278273,
"num_tokens": 12482212.0,
"step": 6370
},
{
"entropy": 6.3723132610321045,
"epoch": 0.6372769530664267,
"grad_norm": 0.9609375,
"learning_rate": 0.0004967381656421073,
"loss": 6.1353,
"mean_token_accuracy": 0.15127579718828202,
"num_tokens": 12492517.0,
"step": 6375
},
{
"entropy": 6.35697603225708,
"epoch": 0.6377767781276553,
"grad_norm": 0.828125,
"learning_rate": 0.0004967321079024643,
"loss": 6.0612,
"mean_token_accuracy": 0.15172800570726394,
"num_tokens": 12502521.0,
"step": 6380
},
{
"entropy": 6.369220495223999,
"epoch": 0.6382766031888839,
"grad_norm": 0.92578125,
"learning_rate": 0.0004967260445840985,
"loss": 6.1996,
"mean_token_accuracy": 0.1520134449005127,
"num_tokens": 12513195.0,
"step": 6385
},
{
"entropy": 6.34788179397583,
"epoch": 0.6387764282501125,
"grad_norm": 0.9296875,
"learning_rate": 0.0004967199756871624,
"loss": 6.0109,
"mean_token_accuracy": 0.1613096699118614,
"num_tokens": 12521640.0,
"step": 6390
},
{
"entropy": 6.350938510894776,
"epoch": 0.6392762533113411,
"grad_norm": 1.0390625,
"learning_rate": 0.0004967139012118088,
"loss": 6.1505,
"mean_token_accuracy": 0.1580931320786476,
"num_tokens": 12530807.0,
"step": 6395
},
{
"entropy": 6.383451271057129,
"epoch": 0.6397760783725696,
"grad_norm": 0.86328125,
"learning_rate": 0.0004967078211581907,
"loss": 6.1942,
"mean_token_accuracy": 0.15094359517097472,
"num_tokens": 12542049.0,
"step": 6400
},
{
"entropy": 6.417659139633178,
"epoch": 0.6402759034337981,
"grad_norm": 1.0234375,
"learning_rate": 0.0004967017355264606,
"loss": 6.1472,
"mean_token_accuracy": 0.15765562951564788,
"num_tokens": 12551699.0,
"step": 6405
},
{
"entropy": 6.386373233795166,
"epoch": 0.6407757284950267,
"grad_norm": 0.9765625,
"learning_rate": 0.000496695644316772,
"loss": 6.223,
"mean_token_accuracy": 0.14688841253519058,
"num_tokens": 12561414.0,
"step": 6410
},
{
"entropy": 6.383068656921386,
"epoch": 0.6412755535562553,
"grad_norm": 0.84375,
"learning_rate": 0.0004966895475292781,
"loss": 6.2077,
"mean_token_accuracy": 0.15030010491609574,
"num_tokens": 12571950.0,
"step": 6415
},
{
"entropy": 6.473227834701538,
"epoch": 0.6417753786174839,
"grad_norm": 0.90625,
"learning_rate": 0.0004966834451641322,
"loss": 6.2193,
"mean_token_accuracy": 0.1474399521946907,
"num_tokens": 12581264.0,
"step": 6420
},
{
"entropy": 6.440133810043335,
"epoch": 0.6422752036787125,
"grad_norm": 0.87109375,
"learning_rate": 0.0004966773372214878,
"loss": 6.1878,
"mean_token_accuracy": 0.15022074282169343,
"num_tokens": 12590950.0,
"step": 6425
},
{
"entropy": 6.3599693775177,
"epoch": 0.642775028739941,
"grad_norm": 0.92578125,
"learning_rate": 0.0004966712237014987,
"loss": 6.1044,
"mean_token_accuracy": 0.15040089339017867,
"num_tokens": 12599766.0,
"step": 6430
},
{
"entropy": 6.37359471321106,
"epoch": 0.6432748538011696,
"grad_norm": 0.921875,
"learning_rate": 0.0004966651046043186,
"loss": 6.1352,
"mean_token_accuracy": 0.15839634090662003,
"num_tokens": 12610074.0,
"step": 6435
},
{
"entropy": 6.2677969455719,
"epoch": 0.6437746788623981,
"grad_norm": 1.046875,
"learning_rate": 0.0004966589799301016,
"loss": 6.0262,
"mean_token_accuracy": 0.16640736907720566,
"num_tokens": 12619117.0,
"step": 6440
},
{
"entropy": 6.417395734786988,
"epoch": 0.6442745039236267,
"grad_norm": 1.109375,
"learning_rate": 0.0004966528496790015,
"loss": 6.1514,
"mean_token_accuracy": 0.14867937788367272,
"num_tokens": 12627993.0,
"step": 6445
},
{
"entropy": 6.38639235496521,
"epoch": 0.6447743289848553,
"grad_norm": 0.9609375,
"learning_rate": 0.0004966467138511729,
"loss": 6.125,
"mean_token_accuracy": 0.14879951775074005,
"num_tokens": 12637787.0,
"step": 6450
},
{
"entropy": 6.417224407196045,
"epoch": 0.6452741540460839,
"grad_norm": 0.98828125,
"learning_rate": 0.00049664057244677,
"loss": 6.2622,
"mean_token_accuracy": 0.14699337184429168,
"num_tokens": 12648409.0,
"step": 6455
},
{
"entropy": 6.358612155914306,
"epoch": 0.6457739791073125,
"grad_norm": 0.97265625,
"learning_rate": 0.0004966344254659473,
"loss": 6.0705,
"mean_token_accuracy": 0.15389290004968642,
"num_tokens": 12658391.0,
"step": 6460
},
{
"entropy": 6.329519510269165,
"epoch": 0.646273804168541,
"grad_norm": 0.9375,
"learning_rate": 0.0004966282729088595,
"loss": 6.1155,
"mean_token_accuracy": 0.1458040416240692,
"num_tokens": 12669926.0,
"step": 6465
},
{
"entropy": 6.300841665267944,
"epoch": 0.6467736292297696,
"grad_norm": 1.03125,
"learning_rate": 0.0004966221147756615,
"loss": 6.1863,
"mean_token_accuracy": 0.15250664204359055,
"num_tokens": 12679176.0,
"step": 6470
},
{
"entropy": 6.375632810592651,
"epoch": 0.6472734542909981,
"grad_norm": 0.8828125,
"learning_rate": 0.0004966159510665079,
"loss": 6.0779,
"mean_token_accuracy": 0.15300790071487427,
"num_tokens": 12688954.0,
"step": 6475
},
{
"entropy": 6.3250854969024655,
"epoch": 0.6477732793522267,
"grad_norm": 1.0625,
"learning_rate": 0.0004966097817815542,
"loss": 6.0973,
"mean_token_accuracy": 0.1529160514473915,
"num_tokens": 12697577.0,
"step": 6480
},
{
"entropy": 6.3207316398620605,
"epoch": 0.6482731044134553,
"grad_norm": 0.90234375,
"learning_rate": 0.0004966036069209552,
"loss": 6.2109,
"mean_token_accuracy": 0.14236616268754004,
"num_tokens": 12708153.0,
"step": 6485
},
{
"entropy": 6.369741916656494,
"epoch": 0.6487729294746839,
"grad_norm": 0.90625,
"learning_rate": 0.0004965974264848666,
"loss": 6.1726,
"mean_token_accuracy": 0.15143117308616638,
"num_tokens": 12717329.0,
"step": 6490
},
{
"entropy": 6.426022863388061,
"epoch": 0.6492727545359124,
"grad_norm": 0.875,
"learning_rate": 0.0004965912404734438,
"loss": 6.1534,
"mean_token_accuracy": 0.15484849885106086,
"num_tokens": 12726528.0,
"step": 6495
},
{
"entropy": 6.416892147064209,
"epoch": 0.649772579597141,
"grad_norm": 0.74609375,
"learning_rate": 0.0004965850488868424,
"loss": 6.1758,
"mean_token_accuracy": 0.14054541513323784,
"num_tokens": 12737341.0,
"step": 6500
},
{
"entropy": 6.32563099861145,
"epoch": 0.6502724046583696,
"grad_norm": 1.0,
"learning_rate": 0.0004965788517252181,
"loss": 6.0704,
"mean_token_accuracy": 0.15614385306835174,
"num_tokens": 12746973.0,
"step": 6505
},
{
"entropy": 6.400835609436035,
"epoch": 0.6507722297195981,
"grad_norm": 0.9453125,
"learning_rate": 0.000496572648988727,
"loss": 6.107,
"mean_token_accuracy": 0.15600904524326326,
"num_tokens": 12756176.0,
"step": 6510
},
{
"entropy": 6.372300529479981,
"epoch": 0.6512720547808267,
"grad_norm": 0.9140625,
"learning_rate": 0.000496566440677525,
"loss": 6.1068,
"mean_token_accuracy": 0.1507421314716339,
"num_tokens": 12766349.0,
"step": 6515
},
{
"entropy": 6.280128812789917,
"epoch": 0.6517718798420553,
"grad_norm": 1.0546875,
"learning_rate": 0.0004965602267917685,
"loss": 6.059,
"mean_token_accuracy": 0.15402039885520935,
"num_tokens": 12775857.0,
"step": 6520
},
{
"entropy": 6.325213050842285,
"epoch": 0.6522717049032839,
"grad_norm": 0.859375,
"learning_rate": 0.0004965540073316136,
"loss": 6.1391,
"mean_token_accuracy": 0.1516575925052166,
"num_tokens": 12786376.0,
"step": 6525
},
{
"entropy": 6.344586610794067,
"epoch": 0.6527715299645124,
"grad_norm": 0.98046875,
"learning_rate": 0.0004965477822972169,
"loss": 6.1409,
"mean_token_accuracy": 0.14783751219511032,
"num_tokens": 12795719.0,
"step": 6530
},
{
"entropy": 6.3499805450439455,
"epoch": 0.653271355025741,
"grad_norm": 0.9375,
"learning_rate": 0.0004965415516887349,
"loss": 6.1566,
"mean_token_accuracy": 0.14976436272263527,
"num_tokens": 12806742.0,
"step": 6535
},
{
"entropy": 6.293507242202759,
"epoch": 0.6537711800869695,
"grad_norm": 0.8515625,
"learning_rate": 0.0004965353155063246,
"loss": 5.9848,
"mean_token_accuracy": 0.15153535157442094,
"num_tokens": 12817362.0,
"step": 6540
},
{
"entropy": 6.23037805557251,
"epoch": 0.6542710051481981,
"grad_norm": 0.9609375,
"learning_rate": 0.0004965290737501427,
"loss": 5.9875,
"mean_token_accuracy": 0.16109038293361663,
"num_tokens": 12826516.0,
"step": 6545
},
{
"entropy": 6.3271490097045895,
"epoch": 0.6547708302094267,
"grad_norm": 0.91015625,
"learning_rate": 0.0004965228264203464,
"loss": 6.1655,
"mean_token_accuracy": 0.14252509251236917,
"num_tokens": 12836321.0,
"step": 6550
},
{
"entropy": 6.355516052246093,
"epoch": 0.6552706552706553,
"grad_norm": 0.92578125,
"learning_rate": 0.0004965165735170928,
"loss": 6.0351,
"mean_token_accuracy": 0.16001931726932525,
"num_tokens": 12845955.0,
"step": 6555
},
{
"entropy": 6.3337170600891115,
"epoch": 0.6557704803318838,
"grad_norm": 1.0234375,
"learning_rate": 0.000496510315040539,
"loss": 6.2007,
"mean_token_accuracy": 0.14780328646302224,
"num_tokens": 12854749.0,
"step": 6560
},
{
"entropy": 6.328001642227173,
"epoch": 0.6562703053931124,
"grad_norm": 0.90625,
"learning_rate": 0.0004965040509908429,
"loss": 6.1559,
"mean_token_accuracy": 0.15129912793636321,
"num_tokens": 12865145.0,
"step": 6565
},
{
"entropy": 6.372905540466308,
"epoch": 0.656770130454341,
"grad_norm": 0.84765625,
"learning_rate": 0.0004964977813681618,
"loss": 6.0883,
"mean_token_accuracy": 0.1573726311326027,
"num_tokens": 12876109.0,
"step": 6570
},
{
"entropy": 6.429256916046143,
"epoch": 0.6572699555155695,
"grad_norm": 0.890625,
"learning_rate": 0.0004964915061726534,
"loss": 6.1508,
"mean_token_accuracy": 0.1481800615787506,
"num_tokens": 12886813.0,
"step": 6575
},
{
"entropy": 6.326648187637329,
"epoch": 0.6577697805767981,
"grad_norm": 0.87109375,
"learning_rate": 0.0004964852254044759,
"loss": 6.1247,
"mean_token_accuracy": 0.15234682857990264,
"num_tokens": 12897991.0,
"step": 6580
},
{
"entropy": 6.314520025253296,
"epoch": 0.6582696056380267,
"grad_norm": 1.0234375,
"learning_rate": 0.0004964789390637871,
"loss": 6.2017,
"mean_token_accuracy": 0.14077168628573417,
"num_tokens": 12907354.0,
"step": 6585
},
{
"entropy": 6.394770336151123,
"epoch": 0.6587694306992553,
"grad_norm": 0.859375,
"learning_rate": 0.0004964726471507451,
"loss": 6.1104,
"mean_token_accuracy": 0.1573098734021187,
"num_tokens": 12917187.0,
"step": 6590
},
{
"entropy": 6.32343111038208,
"epoch": 0.6592692557604838,
"grad_norm": 0.9375,
"learning_rate": 0.0004964663496655083,
"loss": 6.0656,
"mean_token_accuracy": 0.15588307976722718,
"num_tokens": 12926470.0,
"step": 6595
},
{
"entropy": 6.319895076751709,
"epoch": 0.6597690808217124,
"grad_norm": 1.0078125,
"learning_rate": 0.0004964600466082351,
"loss": 6.1129,
"mean_token_accuracy": 0.15508380830287932,
"num_tokens": 12936609.0,
"step": 6600
},
{
"entropy": 6.329181909561157,
"epoch": 0.660268905882941,
"grad_norm": 0.92578125,
"learning_rate": 0.0004964537379790842,
"loss": 6.0444,
"mean_token_accuracy": 0.15705326795578003,
"num_tokens": 12946286.0,
"step": 6605
},
{
"entropy": 6.352501058578492,
"epoch": 0.6607687309441695,
"grad_norm": 0.94921875,
"learning_rate": 0.0004964474237782141,
"loss": 6.0614,
"mean_token_accuracy": 0.15459955632686614,
"num_tokens": 12955881.0,
"step": 6610
},
{
"entropy": 6.383995819091797,
"epoch": 0.6612685560053981,
"grad_norm": 1.0625,
"learning_rate": 0.0004964411040057839,
"loss": 6.197,
"mean_token_accuracy": 0.14883046746253967,
"num_tokens": 12966273.0,
"step": 6615
},
{
"entropy": 6.314850854873657,
"epoch": 0.6617683810666267,
"grad_norm": 1.0703125,
"learning_rate": 0.0004964347786619524,
"loss": 6.1615,
"mean_token_accuracy": 0.1448350228369236,
"num_tokens": 12975300.0,
"step": 6620
},
{
"entropy": 6.382529592514038,
"epoch": 0.6622682061278552,
"grad_norm": 0.98828125,
"learning_rate": 0.0004964284477468791,
"loss": 6.1152,
"mean_token_accuracy": 0.1607019305229187,
"num_tokens": 12985291.0,
"step": 6625
},
{
"entropy": 6.393477964401245,
"epoch": 0.6627680311890838,
"grad_norm": 1.09375,
"learning_rate": 0.0004964221112607228,
"loss": 6.0805,
"mean_token_accuracy": 0.15713976621627807,
"num_tokens": 12995935.0,
"step": 6630
},
{
"entropy": 6.304874563217163,
"epoch": 0.6632678562503124,
"grad_norm": 0.921875,
"learning_rate": 0.0004964157692036432,
"loss": 6.0623,
"mean_token_accuracy": 0.15830949395895005,
"num_tokens": 13006916.0,
"step": 6635
},
{
"entropy": 6.393911600112915,
"epoch": 0.663767681311541,
"grad_norm": 0.890625,
"learning_rate": 0.0004964094215757998,
"loss": 6.328,
"mean_token_accuracy": 0.13789430782198905,
"num_tokens": 13017247.0,
"step": 6640
},
{
"entropy": 6.345292568206787,
"epoch": 0.6642675063727695,
"grad_norm": 0.96875,
"learning_rate": 0.0004964030683773525,
"loss": 6.1419,
"mean_token_accuracy": 0.15398175567388533,
"num_tokens": 13026038.0,
"step": 6645
},
{
"entropy": 6.373729228973389,
"epoch": 0.6647673314339981,
"grad_norm": 1.0078125,
"learning_rate": 0.0004963967096084608,
"loss": 6.2139,
"mean_token_accuracy": 0.14222794100642205,
"num_tokens": 13036436.0,
"step": 6650
},
{
"entropy": 6.420047473907471,
"epoch": 0.6652671564952267,
"grad_norm": 0.9765625,
"learning_rate": 0.0004963903452692849,
"loss": 6.1556,
"mean_token_accuracy": 0.14814745411276817,
"num_tokens": 13045594.0,
"step": 6655
},
{
"entropy": 6.406752109527588,
"epoch": 0.6657669815564552,
"grad_norm": 0.9609375,
"learning_rate": 0.0004963839753599848,
"loss": 6.1465,
"mean_token_accuracy": 0.15128230899572373,
"num_tokens": 13056023.0,
"step": 6660
},
{
"entropy": 6.397957801818848,
"epoch": 0.6662668066176838,
"grad_norm": 0.96484375,
"learning_rate": 0.000496377599880721,
"loss": 6.0909,
"mean_token_accuracy": 0.1555365115404129,
"num_tokens": 13064230.0,
"step": 6665
},
{
"entropy": 6.324149799346924,
"epoch": 0.6667666316789124,
"grad_norm": 1.640625,
"learning_rate": 0.0004963712188316537,
"loss": 5.9924,
"mean_token_accuracy": 0.16710476577281952,
"num_tokens": 13073767.0,
"step": 6670
},
{
"entropy": 6.307120752334595,
"epoch": 0.667266456740141,
"grad_norm": 0.9296875,
"learning_rate": 0.0004963648322129435,
"loss": 6.046,
"mean_token_accuracy": 0.15849119648337365,
"num_tokens": 13083026.0,
"step": 6675
},
{
"entropy": 6.337471437454224,
"epoch": 0.6677662818013695,
"grad_norm": 0.8984375,
"learning_rate": 0.000496358440024751,
"loss": 6.0679,
"mean_token_accuracy": 0.15523538291454314,
"num_tokens": 13092029.0,
"step": 6680
},
{
"entropy": 6.432424545288086,
"epoch": 0.6682661068625981,
"grad_norm": 0.89453125,
"learning_rate": 0.0004963520422672372,
"loss": 6.1974,
"mean_token_accuracy": 0.14917913526296617,
"num_tokens": 13101859.0,
"step": 6685
},
{
"entropy": 6.377618026733399,
"epoch": 0.6687659319238266,
"grad_norm": 0.87109375,
"learning_rate": 0.000496345638940563,
"loss": 6.0756,
"mean_token_accuracy": 0.15455148965120316,
"num_tokens": 13111007.0,
"step": 6690
},
{
"entropy": 6.268700122833252,
"epoch": 0.6692657569850552,
"grad_norm": 1.0,
"learning_rate": 0.0004963392300448894,
"loss": 6.0938,
"mean_token_accuracy": 0.15741010010242462,
"num_tokens": 13120662.0,
"step": 6695
},
{
"entropy": 6.315941667556762,
"epoch": 0.6697655820462838,
"grad_norm": 1.015625,
"learning_rate": 0.0004963328155803778,
"loss": 6.0119,
"mean_token_accuracy": 0.1480046570301056,
"num_tokens": 13130263.0,
"step": 6700
},
{
"entropy": 6.391905784606934,
"epoch": 0.6702654071075124,
"grad_norm": 1.0078125,
"learning_rate": 0.0004963263955471895,
"loss": 6.0676,
"mean_token_accuracy": 0.157219298183918,
"num_tokens": 13139821.0,
"step": 6705
},
{
"entropy": 6.291745901107788,
"epoch": 0.670765232168741,
"grad_norm": 0.89453125,
"learning_rate": 0.000496319969945486,
"loss": 6.0684,
"mean_token_accuracy": 0.1545048825442791,
"num_tokens": 13149731.0,
"step": 6710
},
{
"entropy": 6.310548782348633,
"epoch": 0.6712650572299695,
"grad_norm": 0.953125,
"learning_rate": 0.000496313538775429,
"loss": 6.1415,
"mean_token_accuracy": 0.15519914478063584,
"num_tokens": 13160689.0,
"step": 6715
},
{
"entropy": 6.372151947021484,
"epoch": 0.6717648822911981,
"grad_norm": 0.94140625,
"learning_rate": 0.0004963071020371803,
"loss": 6.2493,
"mean_token_accuracy": 0.15224682688713073,
"num_tokens": 13170694.0,
"step": 6720
},
{
"entropy": 6.416892147064209,
"epoch": 0.6722647073524266,
"grad_norm": 0.96875,
"learning_rate": 0.0004963006597309019,
"loss": 6.2142,
"mean_token_accuracy": 0.15051673650741576,
"num_tokens": 13180514.0,
"step": 6725
},
{
"entropy": 6.3184366703033445,
"epoch": 0.6727645324136552,
"grad_norm": 0.9453125,
"learning_rate": 0.0004962942118567559,
"loss": 6.0267,
"mean_token_accuracy": 0.15538074523210527,
"num_tokens": 13190144.0,
"step": 6730
},
{
"entropy": 6.370538854598999,
"epoch": 0.6732643574748838,
"grad_norm": 0.984375,
"learning_rate": 0.0004962877584149043,
"loss": 6.0767,
"mean_token_accuracy": 0.151009464263916,
"num_tokens": 13199640.0,
"step": 6735
},
{
"entropy": 6.365364742279053,
"epoch": 0.6737641825361124,
"grad_norm": 0.98046875,
"learning_rate": 0.0004962812994055098,
"loss": 6.0285,
"mean_token_accuracy": 0.16212347745895386,
"num_tokens": 13208824.0,
"step": 6740
},
{
"entropy": 6.27418098449707,
"epoch": 0.674264007597341,
"grad_norm": 0.93359375,
"learning_rate": 0.0004962748348287347,
"loss": 5.993,
"mean_token_accuracy": 0.17144986763596534,
"num_tokens": 13219305.0,
"step": 6745
},
{
"entropy": 6.328550195693969,
"epoch": 0.6747638326585695,
"grad_norm": 1.0625,
"learning_rate": 0.0004962683646847416,
"loss": 6.0149,
"mean_token_accuracy": 0.16121758967638017,
"num_tokens": 13227727.0,
"step": 6750
},
{
"entropy": 6.331701040267944,
"epoch": 0.675263657719798,
"grad_norm": 0.9765625,
"learning_rate": 0.0004962618889736936,
"loss": 6.0705,
"mean_token_accuracy": 0.1556951217353344,
"num_tokens": 13237522.0,
"step": 6755
},
{
"entropy": 6.298553991317749,
"epoch": 0.6757634827810266,
"grad_norm": 1.0390625,
"learning_rate": 0.0004962554076957533,
"loss": 6.1272,
"mean_token_accuracy": 0.15072486400604249,
"num_tokens": 13247155.0,
"step": 6760
},
{
"entropy": 6.333307409286499,
"epoch": 0.6762633078422552,
"grad_norm": 0.91015625,
"learning_rate": 0.0004962489208510838,
"loss": 6.1542,
"mean_token_accuracy": 0.14725901037454606,
"num_tokens": 13257324.0,
"step": 6765
},
{
"entropy": 6.295897579193115,
"epoch": 0.6767631329034838,
"grad_norm": 0.94140625,
"learning_rate": 0.0004962424284398484,
"loss": 6.1339,
"mean_token_accuracy": 0.1507594883441925,
"num_tokens": 13267025.0,
"step": 6770
},
{
"entropy": 6.34990758895874,
"epoch": 0.6772629579647124,
"grad_norm": 0.9921875,
"learning_rate": 0.0004962359304622105,
"loss": 6.0501,
"mean_token_accuracy": 0.16186515390872955,
"num_tokens": 13276709.0,
"step": 6775
},
{
"entropy": 6.335904741287232,
"epoch": 0.677762783025941,
"grad_norm": 0.953125,
"learning_rate": 0.0004962294269183335,
"loss": 6.0657,
"mean_token_accuracy": 0.14934478104114532,
"num_tokens": 13286029.0,
"step": 6780
},
{
"entropy": 6.357704019546508,
"epoch": 0.6782626080871695,
"grad_norm": 0.98046875,
"learning_rate": 0.000496222917808381,
"loss": 6.2126,
"mean_token_accuracy": 0.14190776348114015,
"num_tokens": 13296367.0,
"step": 6785
},
{
"entropy": 6.422132253646851,
"epoch": 0.678762433148398,
"grad_norm": 0.9375,
"learning_rate": 0.0004962164031325169,
"loss": 6.1451,
"mean_token_accuracy": 0.1498284511268139,
"num_tokens": 13305186.0,
"step": 6790
},
{
"entropy": 6.359393644332886,
"epoch": 0.6792622582096266,
"grad_norm": 1.0078125,
"learning_rate": 0.000496209882890905,
"loss": 6.0906,
"mean_token_accuracy": 0.1526859447360039,
"num_tokens": 13314512.0,
"step": 6795
},
{
"entropy": 6.35632586479187,
"epoch": 0.6797620832708552,
"grad_norm": 0.953125,
"learning_rate": 0.0004962033570837094,
"loss": 6.1499,
"mean_token_accuracy": 0.15353505611419677,
"num_tokens": 13324423.0,
"step": 6800
},
{
"entropy": 6.385927295684814,
"epoch": 0.6802619083320838,
"grad_norm": 0.921875,
"learning_rate": 0.0004961968257110941,
"loss": 6.1834,
"mean_token_accuracy": 0.14684308022260667,
"num_tokens": 13334133.0,
"step": 6805
},
{
"entropy": 6.434768915176392,
"epoch": 0.6807617333933124,
"grad_norm": 0.90625,
"learning_rate": 0.0004961902887732236,
"loss": 6.1664,
"mean_token_accuracy": 0.15219189077615738,
"num_tokens": 13344017.0,
"step": 6810
},
{
"entropy": 6.303944396972656,
"epoch": 0.681261558454541,
"grad_norm": 0.94921875,
"learning_rate": 0.0004961837462702626,
"loss": 5.9352,
"mean_token_accuracy": 0.16335101723670958,
"num_tokens": 13353922.0,
"step": 6815
},
{
"entropy": 6.333379888534546,
"epoch": 0.6817613835157694,
"grad_norm": 0.8828125,
"learning_rate": 0.0004961771982023752,
"loss": 6.1071,
"mean_token_accuracy": 0.1507889896631241,
"num_tokens": 13363890.0,
"step": 6820
},
{
"entropy": 6.3336488723754885,
"epoch": 0.682261208576998,
"grad_norm": 0.8984375,
"learning_rate": 0.0004961706445697265,
"loss": 6.0536,
"mean_token_accuracy": 0.1545298859477043,
"num_tokens": 13373726.0,
"step": 6825
},
{
"entropy": 6.295986080169678,
"epoch": 0.6827610336382266,
"grad_norm": 0.9921875,
"learning_rate": 0.0004961640853724813,
"loss": 5.9732,
"mean_token_accuracy": 0.16524138748645784,
"num_tokens": 13382049.0,
"step": 6830
},
{
"entropy": 6.233406591415405,
"epoch": 0.6832608586994552,
"grad_norm": 1.0234375,
"learning_rate": 0.0004961575206108046,
"loss": 5.9093,
"mean_token_accuracy": 0.16352120339870452,
"num_tokens": 13390977.0,
"step": 6835
},
{
"entropy": 6.394397211074829,
"epoch": 0.6837606837606838,
"grad_norm": 0.94921875,
"learning_rate": 0.0004961509502848616,
"loss": 6.1173,
"mean_token_accuracy": 0.15197082236409187,
"num_tokens": 13400165.0,
"step": 6840
},
{
"entropy": 6.38779845237732,
"epoch": 0.6842605088219124,
"grad_norm": 0.89453125,
"learning_rate": 0.0004961443743948177,
"loss": 6.1312,
"mean_token_accuracy": 0.14822078496217728,
"num_tokens": 13410201.0,
"step": 6845
},
{
"entropy": 6.359711456298828,
"epoch": 0.684760333883141,
"grad_norm": 0.97265625,
"learning_rate": 0.000496137792940838,
"loss": 6.1173,
"mean_token_accuracy": 0.15910622775554656,
"num_tokens": 13421074.0,
"step": 6850
},
{
"entropy": 6.295166540145874,
"epoch": 0.6852601589443694,
"grad_norm": 1.015625,
"learning_rate": 0.0004961312059230885,
"loss": 5.9639,
"mean_token_accuracy": 0.163887619972229,
"num_tokens": 13430387.0,
"step": 6855
},
{
"entropy": 6.290193653106689,
"epoch": 0.685759984005598,
"grad_norm": 0.9296875,
"learning_rate": 0.0004961246133417348,
"loss": 6.0307,
"mean_token_accuracy": 0.15789291858673096,
"num_tokens": 13439497.0,
"step": 6860
},
{
"entropy": 6.362320423126221,
"epoch": 0.6862598090668266,
"grad_norm": 1.078125,
"learning_rate": 0.0004961180151969427,
"loss": 6.167,
"mean_token_accuracy": 0.1485782764852047,
"num_tokens": 13450543.0,
"step": 6865
},
{
"entropy": 6.317654848098755,
"epoch": 0.6867596341280552,
"grad_norm": 0.96875,
"learning_rate": 0.0004961114114888782,
"loss": 6.0955,
"mean_token_accuracy": 0.15495266914367675,
"num_tokens": 13459996.0,
"step": 6870
},
{
"entropy": 6.314934349060058,
"epoch": 0.6872594591892838,
"grad_norm": 1.015625,
"learning_rate": 0.0004961048022177074,
"loss": 6.0379,
"mean_token_accuracy": 0.15928415656089784,
"num_tokens": 13469141.0,
"step": 6875
},
{
"entropy": 6.280258464813232,
"epoch": 0.6877592842505124,
"grad_norm": 0.99609375,
"learning_rate": 0.0004960981873835968,
"loss": 6.0039,
"mean_token_accuracy": 0.16132644712924957,
"num_tokens": 13478349.0,
"step": 6880
},
{
"entropy": 6.2729377269744875,
"epoch": 0.6882591093117408,
"grad_norm": 0.98828125,
"learning_rate": 0.0004960915669867126,
"loss": 5.963,
"mean_token_accuracy": 0.15914051234722137,
"num_tokens": 13487417.0,
"step": 6885
},
{
"entropy": 6.439836692810059,
"epoch": 0.6887589343729694,
"grad_norm": 0.96484375,
"learning_rate": 0.0004960849410272216,
"loss": 6.2185,
"mean_token_accuracy": 0.15174390077590943,
"num_tokens": 13497607.0,
"step": 6890
},
{
"entropy": 6.363502311706543,
"epoch": 0.689258759434198,
"grad_norm": 0.93359375,
"learning_rate": 0.0004960783095052903,
"loss": 6.1838,
"mean_token_accuracy": 0.14652732163667678,
"num_tokens": 13508222.0,
"step": 6895
},
{
"entropy": 6.220916843414306,
"epoch": 0.6897585844954266,
"grad_norm": 0.890625,
"learning_rate": 0.0004960716724210856,
"loss": 5.9654,
"mean_token_accuracy": 0.16771477311849595,
"num_tokens": 13519593.0,
"step": 6900
},
{
"entropy": 6.444327306747437,
"epoch": 0.6902584095566552,
"grad_norm": 0.87109375,
"learning_rate": 0.0004960650297747746,
"loss": 6.1769,
"mean_token_accuracy": 0.1447721801698208,
"num_tokens": 13531089.0,
"step": 6905
},
{
"entropy": 6.406421327590943,
"epoch": 0.6907582346178838,
"grad_norm": 0.99609375,
"learning_rate": 0.0004960583815665242,
"loss": 6.0815,
"mean_token_accuracy": 0.15794920697808265,
"num_tokens": 13539897.0,
"step": 6910
},
{
"entropy": 6.379333543777466,
"epoch": 0.6912580596791124,
"grad_norm": 0.86328125,
"learning_rate": 0.0004960517277965018,
"loss": 6.1673,
"mean_token_accuracy": 0.1449688643217087,
"num_tokens": 13549986.0,
"step": 6915
},
{
"entropy": 6.373147964477539,
"epoch": 0.6917578847403408,
"grad_norm": 0.953125,
"learning_rate": 0.0004960450684648749,
"loss": 6.1677,
"mean_token_accuracy": 0.15427129566669465,
"num_tokens": 13559866.0,
"step": 6920
},
{
"entropy": 6.437018775939942,
"epoch": 0.6922577098015694,
"grad_norm": 1.015625,
"learning_rate": 0.000496038403571811,
"loss": 6.181,
"mean_token_accuracy": 0.14901704341173172,
"num_tokens": 13569789.0,
"step": 6925
},
{
"entropy": 6.391650104522705,
"epoch": 0.692757534862798,
"grad_norm": 0.81640625,
"learning_rate": 0.0004960317331174777,
"loss": 6.1717,
"mean_token_accuracy": 0.14617592915892602,
"num_tokens": 13579121.0,
"step": 6930
},
{
"entropy": 6.293175888061524,
"epoch": 0.6932573599240266,
"grad_norm": 0.96875,
"learning_rate": 0.0004960250571020428,
"loss": 6.0553,
"mean_token_accuracy": 0.1584838956594467,
"num_tokens": 13589281.0,
"step": 6935
},
{
"entropy": 6.421004772186279,
"epoch": 0.6937571849852552,
"grad_norm": 0.8671875,
"learning_rate": 0.0004960183755256744,
"loss": 6.05,
"mean_token_accuracy": 0.1573459357023239,
"num_tokens": 13598312.0,
"step": 6940
},
{
"entropy": 6.325068998336792,
"epoch": 0.6942570100464838,
"grad_norm": 0.96484375,
"learning_rate": 0.0004960116883885406,
"loss": 6.1065,
"mean_token_accuracy": 0.15536510646343232,
"num_tokens": 13608164.0,
"step": 6945
},
{
"entropy": 6.3117225646972654,
"epoch": 0.6947568351077124,
"grad_norm": 0.99609375,
"learning_rate": 0.0004960049956908095,
"loss": 6.136,
"mean_token_accuracy": 0.1498507909476757,
"num_tokens": 13618314.0,
"step": 6950
},
{
"entropy": 6.355185413360596,
"epoch": 0.6952566601689408,
"grad_norm": 0.90234375,
"learning_rate": 0.0004959982974326496,
"loss": 6.1067,
"mean_token_accuracy": 0.15258670076727868,
"num_tokens": 13628846.0,
"step": 6955
},
{
"entropy": 6.299815368652344,
"epoch": 0.6957564852301694,
"grad_norm": 0.96484375,
"learning_rate": 0.0004959915936142294,
"loss": 6.004,
"mean_token_accuracy": 0.16522947400808335,
"num_tokens": 13637941.0,
"step": 6960
},
{
"entropy": 6.364964723587036,
"epoch": 0.696256310291398,
"grad_norm": 1.0,
"learning_rate": 0.0004959848842357175,
"loss": 6.0706,
"mean_token_accuracy": 0.1542408749461174,
"num_tokens": 13647734.0,
"step": 6965
},
{
"entropy": 6.2456310272216795,
"epoch": 0.6967561353526266,
"grad_norm": 1.03125,
"learning_rate": 0.0004959781692972829,
"loss": 6.0409,
"mean_token_accuracy": 0.15711082890629768,
"num_tokens": 13657929.0,
"step": 6970
},
{
"entropy": 6.349732685089111,
"epoch": 0.6972559604138552,
"grad_norm": 0.95703125,
"learning_rate": 0.0004959714487990943,
"loss": 6.1373,
"mean_token_accuracy": 0.15083275437355043,
"num_tokens": 13668258.0,
"step": 6975
},
{
"entropy": 6.377843856811523,
"epoch": 0.6977557854750838,
"grad_norm": 0.98828125,
"learning_rate": 0.000495964722741321,
"loss": 6.0763,
"mean_token_accuracy": 0.15968114882707596,
"num_tokens": 13677515.0,
"step": 6980
},
{
"entropy": 6.364668655395508,
"epoch": 0.6982556105363122,
"grad_norm": 0.89453125,
"learning_rate": 0.0004959579911241322,
"loss": 6.0952,
"mean_token_accuracy": 0.1535758212208748,
"num_tokens": 13686865.0,
"step": 6985
},
{
"entropy": 6.347046661376953,
"epoch": 0.6987554355975408,
"grad_norm": 0.9296875,
"learning_rate": 0.0004959512539476971,
"loss": 6.1689,
"mean_token_accuracy": 0.15347059890627862,
"num_tokens": 13696683.0,
"step": 6990
},
{
"entropy": 6.355521202087402,
"epoch": 0.6992552606587694,
"grad_norm": 0.96484375,
"learning_rate": 0.0004959445112121853,
"loss": 6.1423,
"mean_token_accuracy": 0.14905546456575394,
"num_tokens": 13706452.0,
"step": 6995
},
{
"entropy": 6.295386886596679,
"epoch": 0.699755085719998,
"grad_norm": 0.984375,
"learning_rate": 0.0004959377629177665,
"loss": 6.0201,
"mean_token_accuracy": 0.15940037220716477,
"num_tokens": 13717474.0,
"step": 7000
},
{
"entropy": 6.295186948776245,
"epoch": 0.7002549107812266,
"grad_norm": 0.93359375,
"learning_rate": 0.0004959310090646104,
"loss": 6.0354,
"mean_token_accuracy": 0.1640301376581192,
"num_tokens": 13726723.0,
"step": 7005
},
{
"entropy": 6.403949785232544,
"epoch": 0.7007547358424552,
"grad_norm": 0.93359375,
"learning_rate": 0.0004959242496528869,
"loss": 6.2578,
"mean_token_accuracy": 0.14559866935014726,
"num_tokens": 13736432.0,
"step": 7010
},
{
"entropy": 6.313732671737671,
"epoch": 0.7012545609036838,
"grad_norm": 1.015625,
"learning_rate": 0.0004959174846827663,
"loss": 6.0769,
"mean_token_accuracy": 0.15393958166241645,
"num_tokens": 13745767.0,
"step": 7015
},
{
"entropy": 6.395278596878052,
"epoch": 0.7017543859649122,
"grad_norm": 0.9765625,
"learning_rate": 0.0004959107141544186,
"loss": 6.1002,
"mean_token_accuracy": 0.1565880261361599,
"num_tokens": 13756035.0,
"step": 7020
},
{
"entropy": 6.28444013595581,
"epoch": 0.7022542110261408,
"grad_norm": 0.984375,
"learning_rate": 0.000495903938068014,
"loss": 5.9301,
"mean_token_accuracy": 0.1703647792339325,
"num_tokens": 13765715.0,
"step": 7025
},
{
"entropy": 6.30950961112976,
"epoch": 0.7027540360873694,
"grad_norm": 0.87890625,
"learning_rate": 0.0004958971564237233,
"loss": 6.1109,
"mean_token_accuracy": 0.15119080170989035,
"num_tokens": 13775987.0,
"step": 7030
},
{
"entropy": 6.309575510025025,
"epoch": 0.703253861148598,
"grad_norm": 0.87890625,
"learning_rate": 0.000495890369221717,
"loss": 6.0342,
"mean_token_accuracy": 0.15727882981300353,
"num_tokens": 13785095.0,
"step": 7035
},
{
"entropy": 6.385285139083862,
"epoch": 0.7037536862098266,
"grad_norm": 1.0390625,
"learning_rate": 0.0004958835764621657,
"loss": 6.0876,
"mean_token_accuracy": 0.15080909579992294,
"num_tokens": 13794777.0,
"step": 7040
},
{
"entropy": 6.289635372161865,
"epoch": 0.7042535112710552,
"grad_norm": 0.98046875,
"learning_rate": 0.0004958767781452406,
"loss": 5.9865,
"mean_token_accuracy": 0.15363326221704482,
"num_tokens": 13803534.0,
"step": 7045
},
{
"entropy": 6.2926548480987545,
"epoch": 0.7047533363322837,
"grad_norm": 0.88671875,
"learning_rate": 0.0004958699742711126,
"loss": 6.0717,
"mean_token_accuracy": 0.15242997258901597,
"num_tokens": 13813941.0,
"step": 7050
},
{
"entropy": 6.318150615692138,
"epoch": 0.7052531613935122,
"grad_norm": 0.875,
"learning_rate": 0.0004958631648399528,
"loss": 6.0659,
"mean_token_accuracy": 0.15480947196483613,
"num_tokens": 13823611.0,
"step": 7055
},
{
"entropy": 6.4087670803070065,
"epoch": 0.7057529864547408,
"grad_norm": 0.90625,
"learning_rate": 0.0004958563498519327,
"loss": 6.177,
"mean_token_accuracy": 0.14841430634260178,
"num_tokens": 13833047.0,
"step": 7060
},
{
"entropy": 6.3627767086029055,
"epoch": 0.7062528115159694,
"grad_norm": 1.2265625,
"learning_rate": 0.0004958495293072235,
"loss": 6.0846,
"mean_token_accuracy": 0.15810488313436508,
"num_tokens": 13843524.0,
"step": 7065
},
{
"entropy": 6.311433601379394,
"epoch": 0.706752636577198,
"grad_norm": 0.953125,
"learning_rate": 0.0004958427032059971,
"loss": 6.0049,
"mean_token_accuracy": 0.16070111840963364,
"num_tokens": 13852188.0,
"step": 7070
},
{
"entropy": 6.289581298828125,
"epoch": 0.7072524616384266,
"grad_norm": 0.94921875,
"learning_rate": 0.0004958358715484251,
"loss": 6.113,
"mean_token_accuracy": 0.15342209935188295,
"num_tokens": 13863540.0,
"step": 7075
},
{
"entropy": 6.4064867973327635,
"epoch": 0.7077522866996552,
"grad_norm": 0.98828125,
"learning_rate": 0.0004958290343346795,
"loss": 6.1144,
"mean_token_accuracy": 0.15415723323822023,
"num_tokens": 13871828.0,
"step": 7080
},
{
"entropy": 6.364339208602905,
"epoch": 0.7082521117608837,
"grad_norm": 0.90625,
"learning_rate": 0.000495822191564932,
"loss": 6.1364,
"mean_token_accuracy": 0.15215339213609697,
"num_tokens": 13882222.0,
"step": 7085
},
{
"entropy": 6.308574485778808,
"epoch": 0.7087519368221122,
"grad_norm": 0.98046875,
"learning_rate": 0.0004958153432393553,
"loss": 6.1712,
"mean_token_accuracy": 0.14875813722610473,
"num_tokens": 13891689.0,
"step": 7090
},
{
"entropy": 6.291139602661133,
"epoch": 0.7092517618833408,
"grad_norm": 0.89453125,
"learning_rate": 0.000495808489358121,
"loss": 6.0596,
"mean_token_accuracy": 0.1585016131401062,
"num_tokens": 13901310.0,
"step": 7095
},
{
"entropy": 6.336572980880737,
"epoch": 0.7097515869445694,
"grad_norm": 0.95703125,
"learning_rate": 0.0004958016299214022,
"loss": 6.0459,
"mean_token_accuracy": 0.1555250734090805,
"num_tokens": 13910350.0,
"step": 7100
},
{
"entropy": 6.3348150730133055,
"epoch": 0.710251412005798,
"grad_norm": 0.9375,
"learning_rate": 0.0004957947649293712,
"loss": 6.1006,
"mean_token_accuracy": 0.15273423865437508,
"num_tokens": 13920481.0,
"step": 7105
},
{
"entropy": 6.364114236831665,
"epoch": 0.7107512370670266,
"grad_norm": 0.84375,
"learning_rate": 0.0004957878943822006,
"loss": 6.1107,
"mean_token_accuracy": 0.15007705762982368,
"num_tokens": 13931232.0,
"step": 7110
},
{
"entropy": 6.381584882736206,
"epoch": 0.7112510621282551,
"grad_norm": 0.89453125,
"learning_rate": 0.0004957810182800634,
"loss": 6.1701,
"mean_token_accuracy": 0.14837423115968704,
"num_tokens": 13940717.0,
"step": 7115
},
{
"entropy": 6.352689504623413,
"epoch": 0.7117508871894836,
"grad_norm": 0.99609375,
"learning_rate": 0.0004957741366231327,
"loss": 6.0159,
"mean_token_accuracy": 0.159651917219162,
"num_tokens": 13949918.0,
"step": 7120
},
{
"entropy": 6.336006450653076,
"epoch": 0.7122507122507122,
"grad_norm": 0.87890625,
"learning_rate": 0.0004957672494115815,
"loss": 6.0373,
"mean_token_accuracy": 0.1582901269197464,
"num_tokens": 13959914.0,
"step": 7125
},
{
"entropy": 6.346986103057861,
"epoch": 0.7127505373119408,
"grad_norm": 1.0,
"learning_rate": 0.000495760356645583,
"loss": 6.0555,
"mean_token_accuracy": 0.1581800937652588,
"num_tokens": 13970054.0,
"step": 7130
},
{
"entropy": 6.2381785869598385,
"epoch": 0.7132503623731694,
"grad_norm": 1.2265625,
"learning_rate": 0.0004957534583253108,
"loss": 5.9713,
"mean_token_accuracy": 0.1689275324344635,
"num_tokens": 13980339.0,
"step": 7135
},
{
"entropy": 6.282525110244751,
"epoch": 0.713750187434398,
"grad_norm": 0.89453125,
"learning_rate": 0.0004957465544509384,
"loss": 6.0771,
"mean_token_accuracy": 0.15145995244383811,
"num_tokens": 13990466.0,
"step": 7140
},
{
"entropy": 6.3833839893341064,
"epoch": 0.7142500124956266,
"grad_norm": 0.98046875,
"learning_rate": 0.0004957396450226395,
"loss": 6.1087,
"mean_token_accuracy": 0.15041610598564148,
"num_tokens": 13999822.0,
"step": 7145
},
{
"entropy": 6.423709201812744,
"epoch": 0.7147498375568551,
"grad_norm": 0.8984375,
"learning_rate": 0.0004957327300405879,
"loss": 6.1129,
"mean_token_accuracy": 0.15330879464745523,
"num_tokens": 14008939.0,
"step": 7150
},
{
"entropy": 6.3598388671875,
"epoch": 0.7152496626180836,
"grad_norm": 0.94140625,
"learning_rate": 0.0004957258095049577,
"loss": 6.0744,
"mean_token_accuracy": 0.1473304107785225,
"num_tokens": 14019005.0,
"step": 7155
},
{
"entropy": 6.315775966644287,
"epoch": 0.7157494876793122,
"grad_norm": 1.0078125,
"learning_rate": 0.0004957188834159229,
"loss": 5.9753,
"mean_token_accuracy": 0.16202560365200042,
"num_tokens": 14029390.0,
"step": 7160
},
{
"entropy": 6.313910865783692,
"epoch": 0.7162493127405408,
"grad_norm": 0.93359375,
"learning_rate": 0.0004957119517736577,
"loss": 6.0548,
"mean_token_accuracy": 0.1552247166633606,
"num_tokens": 14038613.0,
"step": 7165
},
{
"entropy": 6.277583837509155,
"epoch": 0.7167491378017694,
"grad_norm": 0.9375,
"learning_rate": 0.0004957050145783366,
"loss": 6.0511,
"mean_token_accuracy": 0.15635842680931092,
"num_tokens": 14048144.0,
"step": 7170
},
{
"entropy": 6.245211029052735,
"epoch": 0.717248962862998,
"grad_norm": 0.97265625,
"learning_rate": 0.0004956980718301342,
"loss": 5.9671,
"mean_token_accuracy": 0.16858417689800262,
"num_tokens": 14056894.0,
"step": 7175
},
{
"entropy": 6.305768537521362,
"epoch": 0.7177487879242265,
"grad_norm": 0.921875,
"learning_rate": 0.0004956911235292251,
"loss": 6.0449,
"mean_token_accuracy": 0.15435709953308105,
"num_tokens": 14066024.0,
"step": 7180
},
{
"entropy": 6.248495817184448,
"epoch": 0.7182486129854551,
"grad_norm": 0.921875,
"learning_rate": 0.0004956841696757841,
"loss": 6.0587,
"mean_token_accuracy": 0.15853175073862075,
"num_tokens": 14075805.0,
"step": 7185
},
{
"entropy": 6.292652034759522,
"epoch": 0.7187484380466836,
"grad_norm": 0.90625,
"learning_rate": 0.0004956772102699862,
"loss": 6.0769,
"mean_token_accuracy": 0.15876784026622773,
"num_tokens": 14085229.0,
"step": 7190
},
{
"entropy": 6.367125034332275,
"epoch": 0.7192482631079122,
"grad_norm": 0.99609375,
"learning_rate": 0.0004956702453120064,
"loss": 6.0491,
"mean_token_accuracy": 0.15297560393810272,
"num_tokens": 14094594.0,
"step": 7195
},
{
"entropy": 6.294458389282227,
"epoch": 0.7197480881691408,
"grad_norm": 0.9765625,
"learning_rate": 0.0004956632748020201,
"loss": 6.0845,
"mean_token_accuracy": 0.15934179052710534,
"num_tokens": 14104373.0,
"step": 7200
},
{
"entropy": 6.253583383560181,
"epoch": 0.7202479132303694,
"grad_norm": 0.92578125,
"learning_rate": 0.0004956562987402026,
"loss": 5.9804,
"mean_token_accuracy": 0.16113699227571487,
"num_tokens": 14113918.0,
"step": 7205
},
{
"entropy": 6.42399754524231,
"epoch": 0.720747738291598,
"grad_norm": 0.93359375,
"learning_rate": 0.0004956493171267293,
"loss": 6.201,
"mean_token_accuracy": 0.14336002692580224,
"num_tokens": 14123711.0,
"step": 7210
},
{
"entropy": 6.38614559173584,
"epoch": 0.7212475633528265,
"grad_norm": 0.95703125,
"learning_rate": 0.000495642329961776,
"loss": 5.9908,
"mean_token_accuracy": 0.1615636467933655,
"num_tokens": 14132994.0,
"step": 7215
},
{
"entropy": 6.280452919006348,
"epoch": 0.7217473884140551,
"grad_norm": 0.890625,
"learning_rate": 0.0004956353372455185,
"loss": 6.032,
"mean_token_accuracy": 0.15625657886266708,
"num_tokens": 14143398.0,
"step": 7220
},
{
"entropy": 6.278959321975708,
"epoch": 0.7222472134752836,
"grad_norm": 1.0390625,
"learning_rate": 0.0004956283389781327,
"loss": 6.0382,
"mean_token_accuracy": 0.1586635857820511,
"num_tokens": 14152374.0,
"step": 7225
},
{
"entropy": 6.346828269958496,
"epoch": 0.7227470385365122,
"grad_norm": 0.98046875,
"learning_rate": 0.0004956213351597946,
"loss": 5.9974,
"mean_token_accuracy": 0.15392591804265976,
"num_tokens": 14162123.0,
"step": 7230
},
{
"entropy": 6.390681219100952,
"epoch": 0.7232468635977408,
"grad_norm": 1.0234375,
"learning_rate": 0.0004956143257906805,
"loss": 6.176,
"mean_token_accuracy": 0.1489823803305626,
"num_tokens": 14172004.0,
"step": 7235
},
{
"entropy": 6.23836088180542,
"epoch": 0.7237466886589694,
"grad_norm": 1.0546875,
"learning_rate": 0.0004956073108709667,
"loss": 6.0475,
"mean_token_accuracy": 0.15373433083295823,
"num_tokens": 14181659.0,
"step": 7240
},
{
"entropy": 6.266837310791016,
"epoch": 0.7242465137201979,
"grad_norm": 0.921875,
"learning_rate": 0.0004956002904008298,
"loss": 6.0348,
"mean_token_accuracy": 0.15708984881639482,
"num_tokens": 14191051.0,
"step": 7245
},
{
"entropy": 6.3565154552459715,
"epoch": 0.7247463387814265,
"grad_norm": 0.91015625,
"learning_rate": 0.0004955932643804463,
"loss": 6.1488,
"mean_token_accuracy": 0.15095992833375932,
"num_tokens": 14201232.0,
"step": 7250
},
{
"entropy": 6.352097702026367,
"epoch": 0.7252461638426551,
"grad_norm": 1.0390625,
"learning_rate": 0.000495586232809993,
"loss": 6.0911,
"mean_token_accuracy": 0.15086081027984619,
"num_tokens": 14210426.0,
"step": 7255
},
{
"entropy": 6.256494808197021,
"epoch": 0.7257459889038836,
"grad_norm": 0.92578125,
"learning_rate": 0.0004955791956896469,
"loss": 5.9819,
"mean_token_accuracy": 0.16083145290613174,
"num_tokens": 14220013.0,
"step": 7260
},
{
"entropy": 6.325906705856323,
"epoch": 0.7262458139651122,
"grad_norm": 1.0546875,
"learning_rate": 0.000495572153019585,
"loss": 6.1054,
"mean_token_accuracy": 0.15201605707406998,
"num_tokens": 14229249.0,
"step": 7265
},
{
"entropy": 6.350022649765014,
"epoch": 0.7267456390263408,
"grad_norm": 0.98046875,
"learning_rate": 0.0004955651047999844,
"loss": 5.9794,
"mean_token_accuracy": 0.16117955297231673,
"num_tokens": 14238370.0,
"step": 7270
},
{
"entropy": 6.236727142333985,
"epoch": 0.7272454640875694,
"grad_norm": 0.875,
"learning_rate": 0.0004955580510310226,
"loss": 6.0078,
"mean_token_accuracy": 0.1697586938738823,
"num_tokens": 14248037.0,
"step": 7275
},
{
"entropy": 6.23610553741455,
"epoch": 0.7277452891487979,
"grad_norm": 1.0625,
"learning_rate": 0.0004955509917128771,
"loss": 6.097,
"mean_token_accuracy": 0.15329890176653863,
"num_tokens": 14257480.0,
"step": 7280
},
{
"entropy": 6.288200092315674,
"epoch": 0.7282451142100265,
"grad_norm": 0.96875,
"learning_rate": 0.0004955439268457252,
"loss": 6.045,
"mean_token_accuracy": 0.15405199602246283,
"num_tokens": 14266114.0,
"step": 7285
},
{
"entropy": 6.28690390586853,
"epoch": 0.728744939271255,
"grad_norm": 0.89453125,
"learning_rate": 0.0004955368564297449,
"loss": 6.0692,
"mean_token_accuracy": 0.15843201130628587,
"num_tokens": 14276048.0,
"step": 7290
},
{
"entropy": 6.310165357589722,
"epoch": 0.7292447643324836,
"grad_norm": 0.94921875,
"learning_rate": 0.0004955297804651142,
"loss": 5.9545,
"mean_token_accuracy": 0.16382774114608764,
"num_tokens": 14285225.0,
"step": 7295
},
{
"entropy": 6.393028450012207,
"epoch": 0.7297445893937122,
"grad_norm": 0.94921875,
"learning_rate": 0.0004955226989520107,
"loss": 6.021,
"mean_token_accuracy": 0.15758152678608894,
"num_tokens": 14295566.0,
"step": 7300
},
{
"entropy": 6.297847747802734,
"epoch": 0.7302444144549408,
"grad_norm": 1.046875,
"learning_rate": 0.000495515611890613,
"loss": 6.0322,
"mean_token_accuracy": 0.15654397904872894,
"num_tokens": 14305072.0,
"step": 7305
},
{
"entropy": 6.228882741928101,
"epoch": 0.7307442395161693,
"grad_norm": 0.9453125,
"learning_rate": 0.0004955085192810993,
"loss": 5.8721,
"mean_token_accuracy": 0.16369707733392716,
"num_tokens": 14315273.0,
"step": 7310
},
{
"entropy": 6.357260704040527,
"epoch": 0.7312440645773979,
"grad_norm": 1.078125,
"learning_rate": 0.0004955014211236479,
"loss": 6.0891,
"mean_token_accuracy": 0.1517564371228218,
"num_tokens": 14324818.0,
"step": 7315
},
{
"entropy": 6.319687032699585,
"epoch": 0.7317438896386265,
"grad_norm": 0.9375,
"learning_rate": 0.0004954943174184375,
"loss": 6.1402,
"mean_token_accuracy": 0.15445371568202973,
"num_tokens": 14335129.0,
"step": 7320
},
{
"entropy": 6.346318769454956,
"epoch": 0.732243714699855,
"grad_norm": 0.9140625,
"learning_rate": 0.0004954872081656469,
"loss": 6.1243,
"mean_token_accuracy": 0.14883216470479965,
"num_tokens": 14344339.0,
"step": 7325
},
{
"entropy": 6.372745370864868,
"epoch": 0.7327435397610836,
"grad_norm": 0.90625,
"learning_rate": 0.0004954800933654548,
"loss": 6.0556,
"mean_token_accuracy": 0.1492668554186821,
"num_tokens": 14353391.0,
"step": 7330
},
{
"entropy": 6.425935125350952,
"epoch": 0.7332433648223122,
"grad_norm": 0.98046875,
"learning_rate": 0.0004954729730180403,
"loss": 6.1897,
"mean_token_accuracy": 0.15134270042181014,
"num_tokens": 14363684.0,
"step": 7335
},
{
"entropy": 6.320706415176391,
"epoch": 0.7337431898835408,
"grad_norm": 1.0,
"learning_rate": 0.0004954658471235825,
"loss": 6.0673,
"mean_token_accuracy": 0.1559371069073677,
"num_tokens": 14372802.0,
"step": 7340
},
{
"entropy": 6.352352333068848,
"epoch": 0.7342430149447693,
"grad_norm": 0.953125,
"learning_rate": 0.0004954587156822607,
"loss": 6.0945,
"mean_token_accuracy": 0.14530139714479445,
"num_tokens": 14382665.0,
"step": 7345
},
{
"entropy": 6.323452091217041,
"epoch": 0.7347428400059979,
"grad_norm": 0.8828125,
"learning_rate": 0.0004954515786942544,
"loss": 6.0524,
"mean_token_accuracy": 0.15289186835289,
"num_tokens": 14393631.0,
"step": 7350
},
{
"entropy": 6.35225191116333,
"epoch": 0.7352426650672265,
"grad_norm": 0.96484375,
"learning_rate": 0.000495444436159743,
"loss": 6.0698,
"mean_token_accuracy": 0.15652450323104858,
"num_tokens": 14402713.0,
"step": 7355
},
{
"entropy": 6.407626104354859,
"epoch": 0.735742490128455,
"grad_norm": 0.9453125,
"learning_rate": 0.0004954372880789064,
"loss": 6.2033,
"mean_token_accuracy": 0.14676266089081763,
"num_tokens": 14412777.0,
"step": 7360
},
{
"entropy": 6.352416038513184,
"epoch": 0.7362423151896836,
"grad_norm": 0.8984375,
"learning_rate": 0.0004954301344519244,
"loss": 6.1302,
"mean_token_accuracy": 0.15103916823863983,
"num_tokens": 14422839.0,
"step": 7365
},
{
"entropy": 6.234886693954468,
"epoch": 0.7367421402509122,
"grad_norm": 0.9609375,
"learning_rate": 0.0004954229752789769,
"loss": 6.0122,
"mean_token_accuracy": 0.15759240835905075,
"num_tokens": 14433239.0,
"step": 7370
},
{
"entropy": 6.348111581802368,
"epoch": 0.7372419653121407,
"grad_norm": 0.9609375,
"learning_rate": 0.0004954158105602441,
"loss": 6.1149,
"mean_token_accuracy": 0.1497027561068535,
"num_tokens": 14442907.0,
"step": 7375
},
{
"entropy": 6.290708065032959,
"epoch": 0.7377417903733693,
"grad_norm": 0.90234375,
"learning_rate": 0.0004954086402959061,
"loss": 5.9551,
"mean_token_accuracy": 0.15731109082698821,
"num_tokens": 14452492.0,
"step": 7380
},
{
"entropy": 6.335931253433228,
"epoch": 0.7382416154345979,
"grad_norm": 1.09375,
"learning_rate": 0.0004954014644861435,
"loss": 6.1077,
"mean_token_accuracy": 0.14976682364940644,
"num_tokens": 14461353.0,
"step": 7385
},
{
"entropy": 6.275920867919922,
"epoch": 0.7387414404958265,
"grad_norm": 0.91796875,
"learning_rate": 0.0004953942831311367,
"loss": 5.9842,
"mean_token_accuracy": 0.16973174065351487,
"num_tokens": 14470579.0,
"step": 7390
},
{
"entropy": 6.271680068969727,
"epoch": 0.739241265557055,
"grad_norm": 0.83984375,
"learning_rate": 0.0004953870962310666,
"loss": 5.9459,
"mean_token_accuracy": 0.16496060639619828,
"num_tokens": 14481608.0,
"step": 7395
},
{
"entropy": 6.312617778778076,
"epoch": 0.7397410906182836,
"grad_norm": 0.9453125,
"learning_rate": 0.0004953799037861138,
"loss": 6.1275,
"mean_token_accuracy": 0.15205059200525284,
"num_tokens": 14493074.0,
"step": 7400
},
{
"entropy": 6.312835025787353,
"epoch": 0.7402409156795122,
"grad_norm": 0.99609375,
"learning_rate": 0.0004953727057964594,
"loss": 5.9817,
"mean_token_accuracy": 0.16150762885808945,
"num_tokens": 14502469.0,
"step": 7405
},
{
"entropy": 6.3073982238769535,
"epoch": 0.7407407407407407,
"grad_norm": 1.0078125,
"learning_rate": 0.0004953655022622843,
"loss": 6.1209,
"mean_token_accuracy": 0.15200251787900926,
"num_tokens": 14512939.0,
"step": 7410
},
{
"entropy": 6.349795389175415,
"epoch": 0.7412405658019693,
"grad_norm": 0.91015625,
"learning_rate": 0.0004953582931837699,
"loss": 6.1168,
"mean_token_accuracy": 0.15250215232372283,
"num_tokens": 14522681.0,
"step": 7415
},
{
"entropy": 6.332542037963867,
"epoch": 0.7417403908631979,
"grad_norm": 0.96484375,
"learning_rate": 0.0004953510785610976,
"loss": 5.9724,
"mean_token_accuracy": 0.1658090427517891,
"num_tokens": 14532115.0,
"step": 7420
},
{
"entropy": 6.289231538772583,
"epoch": 0.7422402159244265,
"grad_norm": 0.98046875,
"learning_rate": 0.0004953438583944487,
"loss": 6.0547,
"mean_token_accuracy": 0.15433966666460036,
"num_tokens": 14541147.0,
"step": 7425
},
{
"entropy": 6.300818347930909,
"epoch": 0.742740040985655,
"grad_norm": 0.91796875,
"learning_rate": 0.0004953366326840052,
"loss": 6.1087,
"mean_token_accuracy": 0.15979793667793274,
"num_tokens": 14550373.0,
"step": 7430
},
{
"entropy": 6.402552652359009,
"epoch": 0.7432398660468836,
"grad_norm": 0.984375,
"learning_rate": 0.0004953294014299486,
"loss": 6.1188,
"mean_token_accuracy": 0.16001539453864097,
"num_tokens": 14559396.0,
"step": 7435
},
{
"entropy": 6.38272271156311,
"epoch": 0.7437396911081121,
"grad_norm": 0.94921875,
"learning_rate": 0.0004953221646324611,
"loss": 6.1602,
"mean_token_accuracy": 0.14881806969642639,
"num_tokens": 14570292.0,
"step": 7440
},
{
"entropy": 6.283556222915649,
"epoch": 0.7442395161693407,
"grad_norm": 0.91796875,
"learning_rate": 0.0004953149222917245,
"loss": 6.026,
"mean_token_accuracy": 0.15976691395044326,
"num_tokens": 14579887.0,
"step": 7445
},
{
"entropy": 6.357969188690186,
"epoch": 0.7447393412305693,
"grad_norm": 0.92578125,
"learning_rate": 0.0004953076744079213,
"loss": 6.1131,
"mean_token_accuracy": 0.15271517336368562,
"num_tokens": 14590260.0,
"step": 7450
},
{
"entropy": 6.363396072387696,
"epoch": 0.7452391662917979,
"grad_norm": 0.953125,
"learning_rate": 0.0004953004209812335,
"loss": 6.0402,
"mean_token_accuracy": 0.15461803376674652,
"num_tokens": 14599812.0,
"step": 7455
},
{
"entropy": 6.293329191207886,
"epoch": 0.7457389913530265,
"grad_norm": 0.98046875,
"learning_rate": 0.0004952931620118439,
"loss": 5.9757,
"mean_token_accuracy": 0.15982502102851867,
"num_tokens": 14609147.0,
"step": 7460
},
{
"entropy": 6.341886711120606,
"epoch": 0.746238816414255,
"grad_norm": 0.8828125,
"learning_rate": 0.0004952858974999351,
"loss": 6.0962,
"mean_token_accuracy": 0.14547627791762352,
"num_tokens": 14618834.0,
"step": 7465
},
{
"entropy": 6.281481456756592,
"epoch": 0.7467386414754836,
"grad_norm": 1.0078125,
"learning_rate": 0.0004952786274456896,
"loss": 5.9894,
"mean_token_accuracy": 0.16308936029672622,
"num_tokens": 14628556.0,
"step": 7470
},
{
"entropy": 6.282746362686157,
"epoch": 0.7472384665367121,
"grad_norm": 0.96484375,
"learning_rate": 0.0004952713518492907,
"loss": 6.0668,
"mean_token_accuracy": 0.15337453484535218,
"num_tokens": 14637486.0,
"step": 7475
},
{
"entropy": 6.285450553894043,
"epoch": 0.7477382915979407,
"grad_norm": 0.94921875,
"learning_rate": 0.0004952640707109213,
"loss": 6.0511,
"mean_token_accuracy": 0.1565631791949272,
"num_tokens": 14647599.0,
"step": 7480
},
{
"entropy": 6.3424841403961185,
"epoch": 0.7482381166591693,
"grad_norm": 0.921875,
"learning_rate": 0.0004952567840307644,
"loss": 5.9472,
"mean_token_accuracy": 0.16097092032432556,
"num_tokens": 14656246.0,
"step": 7485
},
{
"entropy": 6.290265703201294,
"epoch": 0.7487379417203979,
"grad_norm": 0.96875,
"learning_rate": 0.0004952494918090036,
"loss": 6.0838,
"mean_token_accuracy": 0.15815743207931518,
"num_tokens": 14665550.0,
"step": 7490
},
{
"entropy": 6.348566484451294,
"epoch": 0.7492377667816265,
"grad_norm": 0.9296875,
"learning_rate": 0.0004952421940458222,
"loss": 6.0857,
"mean_token_accuracy": 0.1551155313849449,
"num_tokens": 14676462.0,
"step": 7495
},
{
"entropy": 6.268645429611206,
"epoch": 0.749737591842855,
"grad_norm": 0.859375,
"learning_rate": 0.0004952348907414039,
"loss": 5.9432,
"mean_token_accuracy": 0.16161280274391174,
"num_tokens": 14685963.0,
"step": 7500
},
{
"entropy": 6.269847679138183,
"epoch": 0.7502374169040835,
"grad_norm": 0.90625,
"learning_rate": 0.0004952275818959325,
"loss": 6.0696,
"mean_token_accuracy": 0.15773803144693374,
"num_tokens": 14695764.0,
"step": 7505
},
{
"entropy": 6.284567880630493,
"epoch": 0.7507372419653121,
"grad_norm": 1.015625,
"learning_rate": 0.0004952202675095916,
"loss": 6.0188,
"mean_token_accuracy": 0.16041758805513381,
"num_tokens": 14705839.0,
"step": 7510
},
{
"entropy": 6.308699750900269,
"epoch": 0.7512370670265407,
"grad_norm": 0.9296875,
"learning_rate": 0.0004952129475825657,
"loss": 5.999,
"mean_token_accuracy": 0.1592727228999138,
"num_tokens": 14714801.0,
"step": 7515
},
{
"entropy": 6.369070196151734,
"epoch": 0.7517368920877693,
"grad_norm": 1.0390625,
"learning_rate": 0.0004952056221150385,
"loss": 6.1093,
"mean_token_accuracy": 0.15584300011396407,
"num_tokens": 14723803.0,
"step": 7520
},
{
"entropy": 6.301577520370484,
"epoch": 0.7522367171489979,
"grad_norm": 0.9453125,
"learning_rate": 0.0004951982911071945,
"loss": 6.0071,
"mean_token_accuracy": 0.1624963477253914,
"num_tokens": 14732598.0,
"step": 7525
},
{
"entropy": 6.339388418197632,
"epoch": 0.7527365422102265,
"grad_norm": 0.890625,
"learning_rate": 0.0004951909545592183,
"loss": 6.0743,
"mean_token_accuracy": 0.1598440870642662,
"num_tokens": 14742598.0,
"step": 7530
},
{
"entropy": 6.332610321044922,
"epoch": 0.753236367271455,
"grad_norm": 1.0234375,
"learning_rate": 0.0004951836124712942,
"loss": 6.0829,
"mean_token_accuracy": 0.160172039270401,
"num_tokens": 14751434.0,
"step": 7535
},
{
"entropy": 6.3407433986663815,
"epoch": 0.7537361923326835,
"grad_norm": 0.98046875,
"learning_rate": 0.0004951762648436073,
"loss": 6.0855,
"mean_token_accuracy": 0.15404346212744713,
"num_tokens": 14760661.0,
"step": 7540
},
{
"entropy": 6.353408908843994,
"epoch": 0.7542360173939121,
"grad_norm": 0.92578125,
"learning_rate": 0.0004951689116763419,
"loss": 6.0778,
"mean_token_accuracy": 0.1475398376584053,
"num_tokens": 14770431.0,
"step": 7545
},
{
"entropy": 6.3286073207855225,
"epoch": 0.7547358424551407,
"grad_norm": 0.98828125,
"learning_rate": 0.0004951615529696835,
"loss": 6.0835,
"mean_token_accuracy": 0.15042700469493867,
"num_tokens": 14780066.0,
"step": 7550
},
{
"entropy": 6.36154146194458,
"epoch": 0.7552356675163693,
"grad_norm": 0.92578125,
"learning_rate": 0.0004951541887238172,
"loss": 6.0905,
"mean_token_accuracy": 0.16041741371154786,
"num_tokens": 14789531.0,
"step": 7555
},
{
"entropy": 6.329356575012207,
"epoch": 0.7557354925775979,
"grad_norm": 1.0078125,
"learning_rate": 0.0004951468189389281,
"loss": 5.9914,
"mean_token_accuracy": 0.15692398101091384,
"num_tokens": 14799000.0,
"step": 7560
},
{
"entropy": 6.328855562210083,
"epoch": 0.7562353176388265,
"grad_norm": 0.8828125,
"learning_rate": 0.0004951394436152016,
"loss": 6.0338,
"mean_token_accuracy": 0.1653498664498329,
"num_tokens": 14808991.0,
"step": 7565
},
{
"entropy": 6.308737134933471,
"epoch": 0.7567351427000549,
"grad_norm": 0.890625,
"learning_rate": 0.0004951320627528234,
"loss": 6.062,
"mean_token_accuracy": 0.1673951506614685,
"num_tokens": 14819351.0,
"step": 7570
},
{
"entropy": 6.293022918701172,
"epoch": 0.7572349677612835,
"grad_norm": 0.9921875,
"learning_rate": 0.0004951246763519791,
"loss": 6.0977,
"mean_token_accuracy": 0.15020481944084169,
"num_tokens": 14828924.0,
"step": 7575
},
{
"entropy": 6.486008167266846,
"epoch": 0.7577347928225121,
"grad_norm": 0.84765625,
"learning_rate": 0.0004951172844128546,
"loss": 6.2263,
"mean_token_accuracy": 0.13779762610793114,
"num_tokens": 14841077.0,
"step": 7580
},
{
"entropy": 6.404449224472046,
"epoch": 0.7582346178837407,
"grad_norm": 0.8828125,
"learning_rate": 0.0004951098869356358,
"loss": 6.1409,
"mean_token_accuracy": 0.14772530198097228,
"num_tokens": 14851223.0,
"step": 7585
},
{
"entropy": 6.319031763076782,
"epoch": 0.7587344429449693,
"grad_norm": 0.93359375,
"learning_rate": 0.000495102483920509,
"loss": 6.0015,
"mean_token_accuracy": 0.15831595212221145,
"num_tokens": 14861003.0,
"step": 7590
},
{
"entropy": 6.362440824508667,
"epoch": 0.7592342680061979,
"grad_norm": 0.984375,
"learning_rate": 0.0004950950753676603,
"loss": 6.1243,
"mean_token_accuracy": 0.15234379321336747,
"num_tokens": 14870207.0,
"step": 7595
},
{
"entropy": 6.239006471633911,
"epoch": 0.7597340930674265,
"grad_norm": 1.0,
"learning_rate": 0.0004950876612772763,
"loss": 6.0222,
"mean_token_accuracy": 0.1556150048971176,
"num_tokens": 14879886.0,
"step": 7600
},
{
"entropy": 6.249630975723266,
"epoch": 0.7602339181286549,
"grad_norm": 0.921875,
"learning_rate": 0.0004950802416495431,
"loss": 5.998,
"mean_token_accuracy": 0.1585118979215622,
"num_tokens": 14889275.0,
"step": 7605
},
{
"entropy": 6.38168511390686,
"epoch": 0.7607337431898835,
"grad_norm": 0.953125,
"learning_rate": 0.0004950728164846478,
"loss": 6.0967,
"mean_token_accuracy": 0.154704949259758,
"num_tokens": 14898317.0,
"step": 7610
},
{
"entropy": 6.2991108894348145,
"epoch": 0.7612335682511121,
"grad_norm": 0.9296875,
"learning_rate": 0.0004950653857827771,
"loss": 6.0152,
"mean_token_accuracy": 0.1585605561733246,
"num_tokens": 14907183.0,
"step": 7615
},
{
"entropy": 6.289191198348999,
"epoch": 0.7617333933123407,
"grad_norm": 0.99609375,
"learning_rate": 0.0004950579495441177,
"loss": 6.1565,
"mean_token_accuracy": 0.15245688557624817,
"num_tokens": 14917564.0,
"step": 7620
},
{
"entropy": 6.339101552963257,
"epoch": 0.7622332183735693,
"grad_norm": 0.95703125,
"learning_rate": 0.0004950505077688572,
"loss": 5.9979,
"mean_token_accuracy": 0.161814321577549,
"num_tokens": 14927102.0,
"step": 7625
},
{
"entropy": 6.236083555221557,
"epoch": 0.7627330434347979,
"grad_norm": 0.86328125,
"learning_rate": 0.0004950430604571824,
"loss": 5.891,
"mean_token_accuracy": 0.17401071041822433,
"num_tokens": 14937934.0,
"step": 7630
},
{
"entropy": 6.33174147605896,
"epoch": 0.7632328684960263,
"grad_norm": 0.9140625,
"learning_rate": 0.0004950356076092808,
"loss": 6.0419,
"mean_token_accuracy": 0.1568172812461853,
"num_tokens": 14948109.0,
"step": 7635
},
{
"entropy": 6.32998366355896,
"epoch": 0.7637326935572549,
"grad_norm": 1.015625,
"learning_rate": 0.00049502814922534,
"loss": 6.0861,
"mean_token_accuracy": 0.15185921490192414,
"num_tokens": 14956971.0,
"step": 7640
},
{
"entropy": 6.3382265090942385,
"epoch": 0.7642325186184835,
"grad_norm": 2.328125,
"learning_rate": 0.0004950206853055475,
"loss": 6.0427,
"mean_token_accuracy": 0.16388896405696868,
"num_tokens": 14967046.0,
"step": 7645
},
{
"entropy": 6.33615403175354,
"epoch": 0.7647323436797121,
"grad_norm": 0.9921875,
"learning_rate": 0.0004950132158500913,
"loss": 6.03,
"mean_token_accuracy": 0.15015559047460555,
"num_tokens": 14976234.0,
"step": 7650
},
{
"entropy": 6.213335418701172,
"epoch": 0.7652321687409407,
"grad_norm": 1.0,
"learning_rate": 0.0004950057408591591,
"loss": 5.9746,
"mean_token_accuracy": 0.16839360147714616,
"num_tokens": 14985205.0,
"step": 7655
},
{
"entropy": 6.24402346611023,
"epoch": 0.7657319938021693,
"grad_norm": 1.046875,
"learning_rate": 0.0004949982603329391,
"loss": 5.9755,
"mean_token_accuracy": 0.16419958919286728,
"num_tokens": 14994233.0,
"step": 7660
},
{
"entropy": 6.288731050491333,
"epoch": 0.7662318188633979,
"grad_norm": 0.98828125,
"learning_rate": 0.0004949907742716195,
"loss": 6.0018,
"mean_token_accuracy": 0.15989499390125275,
"num_tokens": 15004001.0,
"step": 7665
},
{
"entropy": 6.40303406715393,
"epoch": 0.7667316439246263,
"grad_norm": 0.8828125,
"learning_rate": 0.0004949832826753886,
"loss": 6.0547,
"mean_token_accuracy": 0.1487782247364521,
"num_tokens": 15015046.0,
"step": 7670
},
{
"entropy": 6.278237056732178,
"epoch": 0.7672314689858549,
"grad_norm": 0.93359375,
"learning_rate": 0.000494975785544435,
"loss": 5.995,
"mean_token_accuracy": 0.15508314371109008,
"num_tokens": 15025404.0,
"step": 7675
},
{
"entropy": 6.331465864181519,
"epoch": 0.7677312940470835,
"grad_norm": 0.90234375,
"learning_rate": 0.0004949682828789472,
"loss": 6.0805,
"mean_token_accuracy": 0.14965164363384248,
"num_tokens": 15036160.0,
"step": 7680
},
{
"entropy": 6.277357912063598,
"epoch": 0.7682311191083121,
"grad_norm": 1.0390625,
"learning_rate": 0.000494960774679114,
"loss": 6.0086,
"mean_token_accuracy": 0.16053372919559478,
"num_tokens": 15046356.0,
"step": 7685
},
{
"entropy": 6.292904233932495,
"epoch": 0.7687309441695407,
"grad_norm": 1.1796875,
"learning_rate": 0.0004949532609451242,
"loss": 5.9958,
"mean_token_accuracy": 0.1582586422562599,
"num_tokens": 15056101.0,
"step": 7690
},
{
"entropy": 6.307175350189209,
"epoch": 0.7692307692307693,
"grad_norm": 1.046875,
"learning_rate": 0.0004949457416771671,
"loss": 6.0047,
"mean_token_accuracy": 0.1672866091132164,
"num_tokens": 15065726.0,
"step": 7695
},
{
"entropy": 6.265319299697876,
"epoch": 0.7697305942919977,
"grad_norm": 0.875,
"learning_rate": 0.0004949382168754316,
"loss": 6.0023,
"mean_token_accuracy": 0.16166395097970962,
"num_tokens": 15075211.0,
"step": 7700
},
{
"entropy": 6.253762483596802,
"epoch": 0.7702304193532263,
"grad_norm": 0.953125,
"learning_rate": 0.0004949306865401073,
"loss": 5.9046,
"mean_token_accuracy": 0.16504451781511306,
"num_tokens": 15085525.0,
"step": 7705
},
{
"entropy": 6.340442562103272,
"epoch": 0.7707302444144549,
"grad_norm": 1.2109375,
"learning_rate": 0.0004949231506713833,
"loss": 6.0988,
"mean_token_accuracy": 0.14864008128643036,
"num_tokens": 15093851.0,
"step": 7710
},
{
"entropy": 6.417589330673218,
"epoch": 0.7712300694756835,
"grad_norm": 0.91796875,
"learning_rate": 0.0004949156092694496,
"loss": 6.1718,
"mean_token_accuracy": 0.15148043632507324,
"num_tokens": 15104066.0,
"step": 7715
},
{
"entropy": 6.343349885940552,
"epoch": 0.7717298945369121,
"grad_norm": 0.94140625,
"learning_rate": 0.0004949080623344958,
"loss": 6.0174,
"mean_token_accuracy": 0.16413416117429733,
"num_tokens": 15114047.0,
"step": 7720
},
{
"entropy": 6.291923236846924,
"epoch": 0.7722297195981407,
"grad_norm": 0.859375,
"learning_rate": 0.0004949005098667117,
"loss": 6.0302,
"mean_token_accuracy": 0.1616566374897957,
"num_tokens": 15123719.0,
"step": 7725
},
{
"entropy": 6.273437070846557,
"epoch": 0.7727295446593693,
"grad_norm": 0.95703125,
"learning_rate": 0.0004948929518662873,
"loss": 6.0762,
"mean_token_accuracy": 0.15936142727732658,
"num_tokens": 15133978.0,
"step": 7730
},
{
"entropy": 6.3448058605194095,
"epoch": 0.7732293697205977,
"grad_norm": 0.828125,
"learning_rate": 0.0004948853883334129,
"loss": 6.1028,
"mean_token_accuracy": 0.15688480734825133,
"num_tokens": 15145058.0,
"step": 7735
},
{
"entropy": 6.401759910583496,
"epoch": 0.7737291947818263,
"grad_norm": 1.1015625,
"learning_rate": 0.0004948778192682787,
"loss": 6.0856,
"mean_token_accuracy": 0.15542424619197845,
"num_tokens": 15155400.0,
"step": 7740
},
{
"entropy": 6.276227903366089,
"epoch": 0.7742290198430549,
"grad_norm": 0.984375,
"learning_rate": 0.000494870244671075,
"loss": 5.9725,
"mean_token_accuracy": 0.1656661108136177,
"num_tokens": 15164969.0,
"step": 7745
},
{
"entropy": 6.2992761611938475,
"epoch": 0.7747288449042835,
"grad_norm": 1.015625,
"learning_rate": 0.0004948626645419926,
"loss": 5.9961,
"mean_token_accuracy": 0.16712433993816375,
"num_tokens": 15173840.0,
"step": 7750
},
{
"entropy": 6.325164842605591,
"epoch": 0.7752286699655121,
"grad_norm": 0.8828125,
"learning_rate": 0.0004948550788812222,
"loss": 6.1298,
"mean_token_accuracy": 0.14570028707385063,
"num_tokens": 15184541.0,
"step": 7755
},
{
"entropy": 6.319697713851928,
"epoch": 0.7757284950267407,
"grad_norm": 1.0,
"learning_rate": 0.0004948474876889545,
"loss": 6.0815,
"mean_token_accuracy": 0.156877700984478,
"num_tokens": 15195060.0,
"step": 7760
},
{
"entropy": 6.3745880126953125,
"epoch": 0.7762283200879692,
"grad_norm": 0.87109375,
"learning_rate": 0.0004948398909653806,
"loss": 6.0774,
"mean_token_accuracy": 0.15215039253234863,
"num_tokens": 15206093.0,
"step": 7765
},
{
"entropy": 6.339943838119507,
"epoch": 0.7767281451491977,
"grad_norm": 0.875,
"learning_rate": 0.0004948322887106917,
"loss": 6.0568,
"mean_token_accuracy": 0.15105302929878234,
"num_tokens": 15215359.0,
"step": 7770
},
{
"entropy": 6.287955093383789,
"epoch": 0.7772279702104263,
"grad_norm": 0.94140625,
"learning_rate": 0.0004948246809250788,
"loss": 6.0727,
"mean_token_accuracy": 0.15486107170581817,
"num_tokens": 15225283.0,
"step": 7775
},
{
"entropy": 6.2795709609985355,
"epoch": 0.7777277952716549,
"grad_norm": 1.0,
"learning_rate": 0.0004948170676087338,
"loss": 5.9447,
"mean_token_accuracy": 0.1627866581082344,
"num_tokens": 15234319.0,
"step": 7780
},
{
"entropy": 6.298702716827393,
"epoch": 0.7782276203328835,
"grad_norm": 0.99609375,
"learning_rate": 0.0004948094487618476,
"loss": 5.9503,
"mean_token_accuracy": 0.16435001492500306,
"num_tokens": 15244410.0,
"step": 7785
},
{
"entropy": 6.270851993560791,
"epoch": 0.7787274453941121,
"grad_norm": 0.96484375,
"learning_rate": 0.0004948018243846126,
"loss": 5.998,
"mean_token_accuracy": 0.15858768075704574,
"num_tokens": 15253431.0,
"step": 7790
},
{
"entropy": 6.276344871520996,
"epoch": 0.7792272704553407,
"grad_norm": 0.97265625,
"learning_rate": 0.0004947941944772199,
"loss": 6.0365,
"mean_token_accuracy": 0.15742659866809844,
"num_tokens": 15262566.0,
"step": 7795
},
{
"entropy": 6.425141429901123,
"epoch": 0.7797270955165692,
"grad_norm": 0.90234375,
"learning_rate": 0.000494786559039862,
"loss": 6.1204,
"mean_token_accuracy": 0.15137385129928588,
"num_tokens": 15272122.0,
"step": 7800
},
{
"entropy": 6.359556198120117,
"epoch": 0.7802269205777977,
"grad_norm": 0.9453125,
"learning_rate": 0.0004947789180727308,
"loss": 6.0816,
"mean_token_accuracy": 0.15132945254445077,
"num_tokens": 15281165.0,
"step": 7805
},
{
"entropy": 6.255874872207642,
"epoch": 0.7807267456390263,
"grad_norm": 1.2890625,
"learning_rate": 0.0004947712715760186,
"loss": 6.0336,
"mean_token_accuracy": 0.15571321696043014,
"num_tokens": 15290992.0,
"step": 7810
},
{
"entropy": 6.416817474365234,
"epoch": 0.7812265707002549,
"grad_norm": 0.98046875,
"learning_rate": 0.0004947636195499177,
"loss": 6.0613,
"mean_token_accuracy": 0.15470456779003144,
"num_tokens": 15300454.0,
"step": 7815
},
{
"entropy": 6.330163240432739,
"epoch": 0.7817263957614835,
"grad_norm": 0.9375,
"learning_rate": 0.0004947559619946207,
"loss": 6.009,
"mean_token_accuracy": 0.16466774940490722,
"num_tokens": 15310471.0,
"step": 7820
},
{
"entropy": 6.31559648513794,
"epoch": 0.7822262208227121,
"grad_norm": 0.87890625,
"learning_rate": 0.0004947482989103203,
"loss": 6.0913,
"mean_token_accuracy": 0.1510937914252281,
"num_tokens": 15320554.0,
"step": 7825
},
{
"entropy": 6.292080879211426,
"epoch": 0.7827260458839406,
"grad_norm": 0.8828125,
"learning_rate": 0.0004947406302972092,
"loss": 6.1229,
"mean_token_accuracy": 0.15209597572684289,
"num_tokens": 15331623.0,
"step": 7830
},
{
"entropy": 6.3830437660217285,
"epoch": 0.7832258709451692,
"grad_norm": 1.03125,
"learning_rate": 0.0004947329561554804,
"loss": 5.9978,
"mean_token_accuracy": 0.15783745497465135,
"num_tokens": 15340954.0,
"step": 7835
},
{
"entropy": 6.3216980457305905,
"epoch": 0.7837256960063977,
"grad_norm": 1.0234375,
"learning_rate": 0.0004947252764853269,
"loss": 6.0846,
"mean_token_accuracy": 0.15559569448232652,
"num_tokens": 15350858.0,
"step": 7840
},
{
"entropy": 6.303387069702149,
"epoch": 0.7842255210676263,
"grad_norm": 0.9609375,
"learning_rate": 0.0004947175912869421,
"loss": 5.9876,
"mean_token_accuracy": 0.15498465597629546,
"num_tokens": 15360382.0,
"step": 7845
},
{
"entropy": 6.305787086486816,
"epoch": 0.7847253461288549,
"grad_norm": 0.953125,
"learning_rate": 0.0004947099005605191,
"loss": 6.08,
"mean_token_accuracy": 0.15489040389657022,
"num_tokens": 15370626.0,
"step": 7850
},
{
"entropy": 6.339403247833252,
"epoch": 0.7852251711900835,
"grad_norm": 1.03125,
"learning_rate": 0.0004947022043062517,
"loss": 6.0591,
"mean_token_accuracy": 0.1591949924826622,
"num_tokens": 15379582.0,
"step": 7855
},
{
"entropy": 6.305923986434936,
"epoch": 0.7857249962513121,
"grad_norm": 0.921875,
"learning_rate": 0.0004946945025243333,
"loss": 5.9576,
"mean_token_accuracy": 0.16097877770662308,
"num_tokens": 15390168.0,
"step": 7860
},
{
"entropy": 6.185457229614258,
"epoch": 0.7862248213125406,
"grad_norm": 0.96875,
"learning_rate": 0.0004946867952149577,
"loss": 5.9029,
"mean_token_accuracy": 0.17234977185726166,
"num_tokens": 15400979.0,
"step": 7865
},
{
"entropy": 6.325086879730224,
"epoch": 0.7867246463737692,
"grad_norm": 0.9140625,
"learning_rate": 0.0004946790823783189,
"loss": 6.0043,
"mean_token_accuracy": 0.1578134700655937,
"num_tokens": 15409954.0,
"step": 7870
},
{
"entropy": 6.296181631088257,
"epoch": 0.7872244714349977,
"grad_norm": 0.90625,
"learning_rate": 0.000494671364014611,
"loss": 6.0902,
"mean_token_accuracy": 0.1527877777814865,
"num_tokens": 15419539.0,
"step": 7875
},
{
"entropy": 6.252876901626587,
"epoch": 0.7877242964962263,
"grad_norm": 0.9765625,
"learning_rate": 0.000494663640124028,
"loss": 6.0102,
"mean_token_accuracy": 0.15679827108979225,
"num_tokens": 15428751.0,
"step": 7880
},
{
"entropy": 6.2969481468200685,
"epoch": 0.7882241215574549,
"grad_norm": 1.046875,
"learning_rate": 0.0004946559107067643,
"loss": 6.0493,
"mean_token_accuracy": 0.16218971014022826,
"num_tokens": 15438466.0,
"step": 7885
},
{
"entropy": 6.264706182479858,
"epoch": 0.7887239466186835,
"grad_norm": 1.0,
"learning_rate": 0.0004946481757630146,
"loss": 6.0204,
"mean_token_accuracy": 0.1577471151947975,
"num_tokens": 15448395.0,
"step": 7890
},
{
"entropy": 6.295848035812378,
"epoch": 0.789223771679912,
"grad_norm": 0.953125,
"learning_rate": 0.0004946404352929732,
"loss": 6.054,
"mean_token_accuracy": 0.15462035834789276,
"num_tokens": 15457986.0,
"step": 7895
},
{
"entropy": 6.314553117752075,
"epoch": 0.7897235967411406,
"grad_norm": 0.8671875,
"learning_rate": 0.000494632689296835,
"loss": 6.0428,
"mean_token_accuracy": 0.15846251398324968,
"num_tokens": 15467989.0,
"step": 7900
},
{
"entropy": 6.316967582702636,
"epoch": 0.7902234218023692,
"grad_norm": 0.91796875,
"learning_rate": 0.0004946249377747947,
"loss": 5.9373,
"mean_token_accuracy": 0.1606815442442894,
"num_tokens": 15477537.0,
"step": 7905
},
{
"entropy": 6.322283792495727,
"epoch": 0.7907232468635977,
"grad_norm": 0.96484375,
"learning_rate": 0.0004946171807270476,
"loss": 6.0101,
"mean_token_accuracy": 0.15901870280504227,
"num_tokens": 15488007.0,
"step": 7910
},
{
"entropy": 6.294283628463745,
"epoch": 0.7912230719248263,
"grad_norm": 0.98046875,
"learning_rate": 0.0004946094181537887,
"loss": 6.0399,
"mean_token_accuracy": 0.15930212587118148,
"num_tokens": 15498122.0,
"step": 7915
},
{
"entropy": 6.312228488922119,
"epoch": 0.7917228969860549,
"grad_norm": 0.9609375,
"learning_rate": 0.0004946016500552134,
"loss": 6.0547,
"mean_token_accuracy": 0.15714965015649796,
"num_tokens": 15507585.0,
"step": 7920
},
{
"entropy": 6.3129908561706545,
"epoch": 0.7922227220472835,
"grad_norm": 0.98046875,
"learning_rate": 0.0004945938764315171,
"loss": 5.9702,
"mean_token_accuracy": 0.1596760258078575,
"num_tokens": 15516260.0,
"step": 7925
},
{
"entropy": 6.34543743133545,
"epoch": 0.792722547108512,
"grad_norm": 0.81640625,
"learning_rate": 0.0004945860972828954,
"loss": 6.0749,
"mean_token_accuracy": 0.1604975938796997,
"num_tokens": 15527372.0,
"step": 7930
},
{
"entropy": 6.284914207458496,
"epoch": 0.7932223721697406,
"grad_norm": 0.88671875,
"learning_rate": 0.0004945783126095439,
"loss": 6.0632,
"mean_token_accuracy": 0.15501769483089448,
"num_tokens": 15537244.0,
"step": 7935
},
{
"entropy": 6.399238157272339,
"epoch": 0.7937221972309692,
"grad_norm": 0.95703125,
"learning_rate": 0.0004945705224116586,
"loss": 6.0507,
"mean_token_accuracy": 0.15411500185728072,
"num_tokens": 15546512.0,
"step": 7940
},
{
"entropy": 6.268219089508056,
"epoch": 0.7942220222921977,
"grad_norm": 0.95703125,
"learning_rate": 0.0004945627266894354,
"loss": 6.0151,
"mean_token_accuracy": 0.16168650090694428,
"num_tokens": 15557381.0,
"step": 7945
},
{
"entropy": 6.286045408248901,
"epoch": 0.7947218473534263,
"grad_norm": 0.94921875,
"learning_rate": 0.0004945549254430707,
"loss": 6.1098,
"mean_token_accuracy": 0.15220927447080612,
"num_tokens": 15565806.0,
"step": 7950
},
{
"entropy": 6.289082765579224,
"epoch": 0.7952216724146549,
"grad_norm": 0.921875,
"learning_rate": 0.0004945471186727603,
"loss": 6.0209,
"mean_token_accuracy": 0.1558805376291275,
"num_tokens": 15577059.0,
"step": 7955
},
{
"entropy": 6.390850400924682,
"epoch": 0.7957214974758834,
"grad_norm": 1.046875,
"learning_rate": 0.0004945393063787011,
"loss": 6.1931,
"mean_token_accuracy": 0.14717739894986154,
"num_tokens": 15587827.0,
"step": 7960
},
{
"entropy": 6.419885969161987,
"epoch": 0.796221322537112,
"grad_norm": 0.984375,
"learning_rate": 0.0004945314885610894,
"loss": 5.9992,
"mean_token_accuracy": 0.15139688029885293,
"num_tokens": 15598232.0,
"step": 7965
},
{
"entropy": 6.285418891906739,
"epoch": 0.7967211475983406,
"grad_norm": 0.953125,
"learning_rate": 0.0004945236652201219,
"loss": 5.9543,
"mean_token_accuracy": 0.15880410224199296,
"num_tokens": 15606896.0,
"step": 7970
},
{
"entropy": 6.226030492782593,
"epoch": 0.7972209726595691,
"grad_norm": 1.0078125,
"learning_rate": 0.0004945158363559955,
"loss": 5.9495,
"mean_token_accuracy": 0.16316137760877608,
"num_tokens": 15616868.0,
"step": 7975
},
{
"entropy": 6.290950870513916,
"epoch": 0.7977207977207977,
"grad_norm": 1.0390625,
"learning_rate": 0.0004945080019689071,
"loss": 5.9824,
"mean_token_accuracy": 0.15485470741987228,
"num_tokens": 15626521.0,
"step": 7980
},
{
"entropy": 6.298295593261718,
"epoch": 0.7982206227820263,
"grad_norm": 0.9609375,
"learning_rate": 0.0004945001620590538,
"loss": 5.9729,
"mean_token_accuracy": 0.16681952327489852,
"num_tokens": 15635901.0,
"step": 7985
},
{
"entropy": 6.282628202438355,
"epoch": 0.7987204478432549,
"grad_norm": 0.96875,
"learning_rate": 0.000494492316626633,
"loss": 6.0232,
"mean_token_accuracy": 0.15542822778224946,
"num_tokens": 15645672.0,
"step": 7990
},
{
"entropy": 6.38148603439331,
"epoch": 0.7992202729044834,
"grad_norm": 0.88671875,
"learning_rate": 0.0004944844656718419,
"loss": 6.0653,
"mean_token_accuracy": 0.15439800694584846,
"num_tokens": 15656668.0,
"step": 7995
},
{
"entropy": 6.331287002563476,
"epoch": 0.799720097965712,
"grad_norm": 0.9296875,
"learning_rate": 0.0004944766091948783,
"loss": 6.0094,
"mean_token_accuracy": 0.16176359057426454,
"num_tokens": 15666176.0,
"step": 8000
},
{
"entropy": 6.295291185379028,
"epoch": 0.8002199230269406,
"grad_norm": 0.94921875,
"learning_rate": 0.0004944687471959395,
"loss": 6.0458,
"mean_token_accuracy": 0.1539482966065407,
"num_tokens": 15676162.0,
"step": 8005
},
{
"entropy": 6.330791330337524,
"epoch": 0.8007197480881691,
"grad_norm": 0.97265625,
"learning_rate": 0.0004944608796752237,
"loss": 6.04,
"mean_token_accuracy": 0.16030583530664444,
"num_tokens": 15685393.0,
"step": 8010
},
{
"entropy": 6.287073183059692,
"epoch": 0.8012195731493977,
"grad_norm": 0.94921875,
"learning_rate": 0.0004944530066329284,
"loss": 5.9583,
"mean_token_accuracy": 0.1641255423426628,
"num_tokens": 15694944.0,
"step": 8015
},
{
"entropy": 6.330994033813477,
"epoch": 0.8017193982106263,
"grad_norm": 0.9296875,
"learning_rate": 0.0004944451280692521,
"loss": 6.1077,
"mean_token_accuracy": 0.15276397317647933,
"num_tokens": 15704410.0,
"step": 8020
},
{
"entropy": 6.335663318634033,
"epoch": 0.8022192232718548,
"grad_norm": 1.0703125,
"learning_rate": 0.0004944372439843927,
"loss": 5.9902,
"mean_token_accuracy": 0.16062835603952408,
"num_tokens": 15713663.0,
"step": 8025
},
{
"entropy": 6.310619258880616,
"epoch": 0.8027190483330834,
"grad_norm": 1.0,
"learning_rate": 0.0004944293543785489,
"loss": 5.9494,
"mean_token_accuracy": 0.16796149760484697,
"num_tokens": 15722662.0,
"step": 8030
},
{
"entropy": 6.293089580535889,
"epoch": 0.803218873394312,
"grad_norm": 0.96484375,
"learning_rate": 0.0004944214592519188,
"loss": 6.136,
"mean_token_accuracy": 0.14677956327795982,
"num_tokens": 15732615.0,
"step": 8035
},
{
"entropy": 6.367898321151733,
"epoch": 0.8037186984555406,
"grad_norm": 0.95703125,
"learning_rate": 0.0004944135586047014,
"loss": 6.0613,
"mean_token_accuracy": 0.1663420766592026,
"num_tokens": 15741569.0,
"step": 8040
},
{
"entropy": 6.326604604721069,
"epoch": 0.8042185235167691,
"grad_norm": 1.0,
"learning_rate": 0.0004944056524370953,
"loss": 5.9257,
"mean_token_accuracy": 0.16761356443166733,
"num_tokens": 15750734.0,
"step": 8045
},
{
"entropy": 6.350539255142212,
"epoch": 0.8047183485779977,
"grad_norm": 0.921875,
"learning_rate": 0.0004943977407492994,
"loss": 6.026,
"mean_token_accuracy": 0.1593258261680603,
"num_tokens": 15761664.0,
"step": 8050
},
{
"entropy": 6.268751049041748,
"epoch": 0.8052181736392263,
"grad_norm": 0.97265625,
"learning_rate": 0.0004943898235415128,
"loss": 5.9433,
"mean_token_accuracy": 0.1661013588309288,
"num_tokens": 15771159.0,
"step": 8055
},
{
"entropy": 6.329112386703491,
"epoch": 0.8057179987004548,
"grad_norm": 0.92578125,
"learning_rate": 0.0004943819008139347,
"loss": 6.0625,
"mean_token_accuracy": 0.15712908059358596,
"num_tokens": 15781429.0,
"step": 8060
},
{
"entropy": 6.317018222808838,
"epoch": 0.8062178237616834,
"grad_norm": 0.9609375,
"learning_rate": 0.0004943739725667644,
"loss": 5.9804,
"mean_token_accuracy": 0.1601518914103508,
"num_tokens": 15791361.0,
"step": 8065
},
{
"entropy": 6.332014799118042,
"epoch": 0.806717648822912,
"grad_norm": 0.9296875,
"learning_rate": 0.0004943660388002014,
"loss": 6.1,
"mean_token_accuracy": 0.1526860162615776,
"num_tokens": 15802020.0,
"step": 8070
},
{
"entropy": 6.251399850845337,
"epoch": 0.8072174738841406,
"grad_norm": 0.9765625,
"learning_rate": 0.0004943580995144454,
"loss": 5.9242,
"mean_token_accuracy": 0.16700202226638794,
"num_tokens": 15811319.0,
"step": 8075
},
{
"entropy": 6.355843782424927,
"epoch": 0.8077172989453691,
"grad_norm": 0.921875,
"learning_rate": 0.0004943501547096961,
"loss": 6.0495,
"mean_token_accuracy": 0.15723416358232498,
"num_tokens": 15821337.0,
"step": 8080
},
{
"entropy": 6.3409507274627686,
"epoch": 0.8082171240065977,
"grad_norm": 1.0078125,
"learning_rate": 0.0004943422043861531,
"loss": 6.0507,
"mean_token_accuracy": 0.1606202594935894,
"num_tokens": 15832234.0,
"step": 8085
},
{
"entropy": 6.319289636611939,
"epoch": 0.8087169490678262,
"grad_norm": 0.98828125,
"learning_rate": 0.0004943342485440169,
"loss": 6.015,
"mean_token_accuracy": 0.15711220502853393,
"num_tokens": 15841797.0,
"step": 8090
},
{
"entropy": 6.286961030960083,
"epoch": 0.8092167741290548,
"grad_norm": 1.0859375,
"learning_rate": 0.0004943262871834874,
"loss": 6.0475,
"mean_token_accuracy": 0.1557351380586624,
"num_tokens": 15851031.0,
"step": 8095
},
{
"entropy": 6.319517278671265,
"epoch": 0.8097165991902834,
"grad_norm": 0.98046875,
"learning_rate": 0.0004943183203047649,
"loss": 6.0464,
"mean_token_accuracy": 0.15256616175174714,
"num_tokens": 15861297.0,
"step": 8100
},
{
"entropy": 6.345840072631836,
"epoch": 0.810216424251512,
"grad_norm": 0.91796875,
"learning_rate": 0.0004943103479080499,
"loss": 6.0605,
"mean_token_accuracy": 0.15655364692211152,
"num_tokens": 15871066.0,
"step": 8105
},
{
"entropy": 6.360694885253906,
"epoch": 0.8107162493127406,
"grad_norm": 1.0234375,
"learning_rate": 0.000494302369993543,
"loss": 6.0542,
"mean_token_accuracy": 0.15328925251960754,
"num_tokens": 15880395.0,
"step": 8110
},
{
"entropy": 6.309396028518677,
"epoch": 0.8112160743739691,
"grad_norm": 1.03125,
"learning_rate": 0.0004942943865614449,
"loss": 5.9993,
"mean_token_accuracy": 0.16092857122421264,
"num_tokens": 15889937.0,
"step": 8115
},
{
"entropy": 6.33162522315979,
"epoch": 0.8117158994351977,
"grad_norm": 1.1484375,
"learning_rate": 0.0004942863976119565,
"loss": 5.991,
"mean_token_accuracy": 0.15774793028831482,
"num_tokens": 15899894.0,
"step": 8120
},
{
"entropy": 6.20619764328003,
"epoch": 0.8122157244964262,
"grad_norm": 1.0390625,
"learning_rate": 0.0004942784031452788,
"loss": 5.8821,
"mean_token_accuracy": 0.17743793427944182,
"num_tokens": 15909038.0,
"step": 8125
},
{
"entropy": 6.272015857696533,
"epoch": 0.8127155495576548,
"grad_norm": 0.9765625,
"learning_rate": 0.0004942704031616127,
"loss": 6.0149,
"mean_token_accuracy": 0.1574663609266281,
"num_tokens": 15917891.0,
"step": 8130
},
{
"entropy": 6.3723695278167725,
"epoch": 0.8132153746188834,
"grad_norm": 0.98828125,
"learning_rate": 0.0004942623976611598,
"loss": 6.1394,
"mean_token_accuracy": 0.14506006017327308,
"num_tokens": 15928079.0,
"step": 8135
},
{
"entropy": 6.27050142288208,
"epoch": 0.813715199680112,
"grad_norm": 0.953125,
"learning_rate": 0.0004942543866441215,
"loss": 5.9642,
"mean_token_accuracy": 0.1592273771762848,
"num_tokens": 15938065.0,
"step": 8140
},
{
"entropy": 6.258691740036011,
"epoch": 0.8142150247413406,
"grad_norm": 0.9296875,
"learning_rate": 0.0004942463701106992,
"loss": 5.9992,
"mean_token_accuracy": 0.1551909014582634,
"num_tokens": 15947650.0,
"step": 8145
},
{
"entropy": 6.285524368286133,
"epoch": 0.8147148498025691,
"grad_norm": 0.9765625,
"learning_rate": 0.0004942383480610946,
"loss": 5.9976,
"mean_token_accuracy": 0.159341561794281,
"num_tokens": 15957194.0,
"step": 8150
},
{
"entropy": 6.351723527908325,
"epoch": 0.8152146748637976,
"grad_norm": 1.0,
"learning_rate": 0.0004942303204955096,
"loss": 5.9696,
"mean_token_accuracy": 0.15660293102264405,
"num_tokens": 15966068.0,
"step": 8155
},
{
"entropy": 6.272865247726441,
"epoch": 0.8157144999250262,
"grad_norm": 0.9453125,
"learning_rate": 0.0004942222874141461,
"loss": 5.9908,
"mean_token_accuracy": 0.16238573789596558,
"num_tokens": 15975672.0,
"step": 8160
},
{
"entropy": 6.279624938964844,
"epoch": 0.8162143249862548,
"grad_norm": 0.90625,
"learning_rate": 0.0004942142488172063,
"loss": 6.0054,
"mean_token_accuracy": 0.16126897037029267,
"num_tokens": 15986216.0,
"step": 8165
},
{
"entropy": 6.240652704238892,
"epoch": 0.8167141500474834,
"grad_norm": 1.0703125,
"learning_rate": 0.0004942062047048924,
"loss": 5.9457,
"mean_token_accuracy": 0.1646802917122841,
"num_tokens": 15994673.0,
"step": 8170
},
{
"entropy": 6.258727169036865,
"epoch": 0.817213975108712,
"grad_norm": 0.9765625,
"learning_rate": 0.0004941981550774068,
"loss": 5.9669,
"mean_token_accuracy": 0.1606065884232521,
"num_tokens": 16003576.0,
"step": 8175
},
{
"entropy": 6.213871335983276,
"epoch": 0.8177138001699406,
"grad_norm": 1.1015625,
"learning_rate": 0.0004941900999349521,
"loss": 5.9646,
"mean_token_accuracy": 0.16639024764299393,
"num_tokens": 16013334.0,
"step": 8180
},
{
"entropy": 6.334703540802002,
"epoch": 0.8182136252311691,
"grad_norm": 0.921875,
"learning_rate": 0.0004941820392777307,
"loss": 5.9798,
"mean_token_accuracy": 0.16231327503919601,
"num_tokens": 16023101.0,
"step": 8185
},
{
"entropy": 6.282994222640991,
"epoch": 0.8187134502923976,
"grad_norm": 0.89453125,
"learning_rate": 0.0004941739731059457,
"loss": 6.0535,
"mean_token_accuracy": 0.15936345607042313,
"num_tokens": 16034189.0,
"step": 8190
},
{
"entropy": 6.307040691375732,
"epoch": 0.8192132753536262,
"grad_norm": 0.90625,
"learning_rate": 0.0004941659014198,
"loss": 5.9579,
"mean_token_accuracy": 0.16265781372785568,
"num_tokens": 16044008.0,
"step": 8195
},
{
"entropy": 6.352166604995728,
"epoch": 0.8197131004148548,
"grad_norm": 0.91796875,
"learning_rate": 0.0004941578242194965,
"loss": 6.0538,
"mean_token_accuracy": 0.1634441077709198,
"num_tokens": 16053557.0,
"step": 8200
},
{
"entropy": 6.326949834823608,
"epoch": 0.8202129254760834,
"grad_norm": 0.93359375,
"learning_rate": 0.0004941497415052387,
"loss": 5.9862,
"mean_token_accuracy": 0.15928223729133606,
"num_tokens": 16063191.0,
"step": 8205
},
{
"entropy": 6.288394212722778,
"epoch": 0.820712750537312,
"grad_norm": 1.0234375,
"learning_rate": 0.0004941416532772296,
"loss": 5.9931,
"mean_token_accuracy": 0.15942111611366272,
"num_tokens": 16072482.0,
"step": 8210
},
{
"entropy": 6.222275876998902,
"epoch": 0.8212125755985406,
"grad_norm": 0.93359375,
"learning_rate": 0.000494133559535673,
"loss": 5.9711,
"mean_token_accuracy": 0.16516152769327164,
"num_tokens": 16082848.0,
"step": 8215
},
{
"entropy": 6.365378522872925,
"epoch": 0.821712400659769,
"grad_norm": 1.0390625,
"learning_rate": 0.0004941254602807724,
"loss": 6.126,
"mean_token_accuracy": 0.1533277228474617,
"num_tokens": 16091604.0,
"step": 8220
},
{
"entropy": 6.386179447174072,
"epoch": 0.8222122257209976,
"grad_norm": 0.97265625,
"learning_rate": 0.0004941173555127315,
"loss": 5.9959,
"mean_token_accuracy": 0.15468641519546508,
"num_tokens": 16101427.0,
"step": 8225
},
{
"entropy": 6.266720581054687,
"epoch": 0.8227120507822262,
"grad_norm": 0.94921875,
"learning_rate": 0.0004941092452317546,
"loss": 5.9521,
"mean_token_accuracy": 0.1627202808856964,
"num_tokens": 16112206.0,
"step": 8230
},
{
"entropy": 6.348733854293823,
"epoch": 0.8232118758434548,
"grad_norm": 0.89453125,
"learning_rate": 0.0004941011294380452,
"loss": 6.081,
"mean_token_accuracy": 0.16018219143152237,
"num_tokens": 16122652.0,
"step": 8235
},
{
"entropy": 6.282577753067017,
"epoch": 0.8237117009046834,
"grad_norm": 0.890625,
"learning_rate": 0.0004940930081318079,
"loss": 5.9703,
"mean_token_accuracy": 0.16304928660392762,
"num_tokens": 16132976.0,
"step": 8240
},
{
"entropy": 6.3599306583404545,
"epoch": 0.824211525965912,
"grad_norm": 0.859375,
"learning_rate": 0.0004940848813132469,
"loss": 6.0602,
"mean_token_accuracy": 0.15872150510549546,
"num_tokens": 16143514.0,
"step": 8245
},
{
"entropy": 6.337618827819824,
"epoch": 0.8247113510271405,
"grad_norm": 0.9921875,
"learning_rate": 0.0004940767489825666,
"loss": 6.0347,
"mean_token_accuracy": 0.1554365411400795,
"num_tokens": 16154235.0,
"step": 8250
},
{
"entropy": 6.311557006835938,
"epoch": 0.825211176088369,
"grad_norm": 0.953125,
"learning_rate": 0.0004940686111399716,
"loss": 5.9213,
"mean_token_accuracy": 0.16360237300395966,
"num_tokens": 16164251.0,
"step": 8255
},
{
"entropy": 6.223285436630249,
"epoch": 0.8257110011495976,
"grad_norm": 0.99609375,
"learning_rate": 0.0004940604677856669,
"loss": 5.9097,
"mean_token_accuracy": 0.17255427986383437,
"num_tokens": 16172780.0,
"step": 8260
},
{
"entropy": 6.176182651519776,
"epoch": 0.8262108262108262,
"grad_norm": 0.94140625,
"learning_rate": 0.000494052318919857,
"loss": 5.8797,
"mean_token_accuracy": 0.1700695902109146,
"num_tokens": 16182443.0,
"step": 8265
},
{
"entropy": 6.302811527252198,
"epoch": 0.8267106512720548,
"grad_norm": 0.921875,
"learning_rate": 0.0004940441645427473,
"loss": 5.9613,
"mean_token_accuracy": 0.1650758519768715,
"num_tokens": 16192515.0,
"step": 8270
},
{
"entropy": 6.309085083007813,
"epoch": 0.8272104763332834,
"grad_norm": 0.96484375,
"learning_rate": 0.0004940360046545427,
"loss": 5.9899,
"mean_token_accuracy": 0.1594035267829895,
"num_tokens": 16201615.0,
"step": 8275
},
{
"entropy": 6.289867162704468,
"epoch": 0.827710301394512,
"grad_norm": 1.109375,
"learning_rate": 0.0004940278392554486,
"loss": 6.0208,
"mean_token_accuracy": 0.1563822664320469,
"num_tokens": 16210461.0,
"step": 8280
},
{
"entropy": 6.211748933792114,
"epoch": 0.8282101264557404,
"grad_norm": 0.9453125,
"learning_rate": 0.0004940196683456704,
"loss": 6.0048,
"mean_token_accuracy": 0.15217093974351883,
"num_tokens": 16220293.0,
"step": 8285
},
{
"entropy": 6.247614860534668,
"epoch": 0.828709951516969,
"grad_norm": 1.015625,
"learning_rate": 0.0004940114919254137,
"loss": 5.9201,
"mean_token_accuracy": 0.16677623838186265,
"num_tokens": 16230020.0,
"step": 8290
},
{
"entropy": 6.289156246185303,
"epoch": 0.8292097765781976,
"grad_norm": 1.0625,
"learning_rate": 0.0004940033099948843,
"loss": 5.9484,
"mean_token_accuracy": 0.15729523301124573,
"num_tokens": 16240187.0,
"step": 8295
},
{
"entropy": 6.290634059906006,
"epoch": 0.8297096016394262,
"grad_norm": 0.9375,
"learning_rate": 0.0004939951225542879,
"loss": 5.9497,
"mean_token_accuracy": 0.15783804953098296,
"num_tokens": 16250200.0,
"step": 8300
},
{
"entropy": 6.322738456726074,
"epoch": 0.8302094267006548,
"grad_norm": 0.9609375,
"learning_rate": 0.0004939869296038306,
"loss": 6.0424,
"mean_token_accuracy": 0.1600612223148346,
"num_tokens": 16259396.0,
"step": 8305
},
{
"entropy": 6.34406042098999,
"epoch": 0.8307092517618834,
"grad_norm": 0.9765625,
"learning_rate": 0.0004939787311437185,
"loss": 6.0641,
"mean_token_accuracy": 0.14704681187868118,
"num_tokens": 16270033.0,
"step": 8310
},
{
"entropy": 6.356411790847778,
"epoch": 0.831209076823112,
"grad_norm": 0.9453125,
"learning_rate": 0.0004939705271741579,
"loss": 6.0194,
"mean_token_accuracy": 0.1587398409843445,
"num_tokens": 16281464.0,
"step": 8315
},
{
"entropy": 6.35593547821045,
"epoch": 0.8317089018843404,
"grad_norm": 0.92578125,
"learning_rate": 0.0004939623176953551,
"loss": 6.0765,
"mean_token_accuracy": 0.15352010428905488,
"num_tokens": 16292125.0,
"step": 8320
},
{
"entropy": 6.279843187332153,
"epoch": 0.832208726945569,
"grad_norm": 1.171875,
"learning_rate": 0.0004939541027075168,
"loss": 6.0081,
"mean_token_accuracy": 0.15737985596060752,
"num_tokens": 16302993.0,
"step": 8325
},
{
"entropy": 6.296454381942749,
"epoch": 0.8327085520067976,
"grad_norm": 1.015625,
"learning_rate": 0.0004939458822108496,
"loss": 5.9758,
"mean_token_accuracy": 0.1600040689110756,
"num_tokens": 16311287.0,
"step": 8330
},
{
"entropy": 6.2996152400970455,
"epoch": 0.8332083770680262,
"grad_norm": 1.015625,
"learning_rate": 0.0004939376562055602,
"loss": 5.9734,
"mean_token_accuracy": 0.15882916301488875,
"num_tokens": 16320701.0,
"step": 8335
},
{
"entropy": 6.252297353744507,
"epoch": 0.8337082021292548,
"grad_norm": 0.96875,
"learning_rate": 0.0004939294246918558,
"loss": 5.9976,
"mean_token_accuracy": 0.1609439618885517,
"num_tokens": 16329834.0,
"step": 8340
},
{
"entropy": 6.296989965438843,
"epoch": 0.8342080271904834,
"grad_norm": 0.96875,
"learning_rate": 0.0004939211876699435,
"loss": 5.9992,
"mean_token_accuracy": 0.15677739232778548,
"num_tokens": 16339043.0,
"step": 8345
},
{
"entropy": 6.37321105003357,
"epoch": 0.8347078522517118,
"grad_norm": 0.984375,
"learning_rate": 0.0004939129451400305,
"loss": 6.1096,
"mean_token_accuracy": 0.15234270095825195,
"num_tokens": 16348774.0,
"step": 8350
},
{
"entropy": 6.29345474243164,
"epoch": 0.8352076773129404,
"grad_norm": 1.0390625,
"learning_rate": 0.000493904697102324,
"loss": 5.9733,
"mean_token_accuracy": 0.15583033114671707,
"num_tokens": 16357999.0,
"step": 8355
},
{
"entropy": 6.317603683471679,
"epoch": 0.835707502374169,
"grad_norm": 1.4609375,
"learning_rate": 0.0004938964435570317,
"loss": 5.9919,
"mean_token_accuracy": 0.16160280331969262,
"num_tokens": 16369308.0,
"step": 8360
},
{
"entropy": 6.244389152526855,
"epoch": 0.8362073274353976,
"grad_norm": 0.93359375,
"learning_rate": 0.0004938881845043612,
"loss": 5.9239,
"mean_token_accuracy": 0.16524045318365096,
"num_tokens": 16378001.0,
"step": 8365
},
{
"entropy": 6.399140071868897,
"epoch": 0.8367071524966262,
"grad_norm": 1.0,
"learning_rate": 0.0004938799199445202,
"loss": 6.1011,
"mean_token_accuracy": 0.15495192408561706,
"num_tokens": 16387925.0,
"step": 8370
},
{
"entropy": 6.361681127548218,
"epoch": 0.8372069775578548,
"grad_norm": 0.94921875,
"learning_rate": 0.0004938716498777169,
"loss": 6.0831,
"mean_token_accuracy": 0.1503365769982338,
"num_tokens": 16398200.0,
"step": 8375
},
{
"entropy": 6.203743410110474,
"epoch": 0.8377068026190834,
"grad_norm": 0.953125,
"learning_rate": 0.0004938633743041592,
"loss": 5.8792,
"mean_token_accuracy": 0.1681562125682831,
"num_tokens": 16407874.0,
"step": 8380
},
{
"entropy": 6.315232372283935,
"epoch": 0.8382066276803118,
"grad_norm": 0.95703125,
"learning_rate": 0.0004938550932240552,
"loss": 5.9983,
"mean_token_accuracy": 0.16121616810560227,
"num_tokens": 16417825.0,
"step": 8385
},
{
"entropy": 6.285078287124634,
"epoch": 0.8387064527415404,
"grad_norm": 0.9375,
"learning_rate": 0.0004938468066376134,
"loss": 5.9981,
"mean_token_accuracy": 0.1610456958413124,
"num_tokens": 16427265.0,
"step": 8390
},
{
"entropy": 6.177667474746704,
"epoch": 0.839206277802769,
"grad_norm": 0.99609375,
"learning_rate": 0.0004938385145450423,
"loss": 5.9246,
"mean_token_accuracy": 0.17087895572185516,
"num_tokens": 16436639.0,
"step": 8395
},
{
"entropy": 6.293653726577759,
"epoch": 0.8397061028639976,
"grad_norm": 0.921875,
"learning_rate": 0.0004938302169465505,
"loss": 6.0044,
"mean_token_accuracy": 0.16215359568595886,
"num_tokens": 16446411.0,
"step": 8400
},
{
"entropy": 6.329254531860352,
"epoch": 0.8402059279252262,
"grad_norm": 1.0390625,
"learning_rate": 0.0004938219138423467,
"loss": 6.0529,
"mean_token_accuracy": 0.16097778379917144,
"num_tokens": 16455784.0,
"step": 8405
},
{
"entropy": 6.223066806793213,
"epoch": 0.8407057529864548,
"grad_norm": 1.0078125,
"learning_rate": 0.0004938136052326399,
"loss": 5.8976,
"mean_token_accuracy": 0.17020540684461594,
"num_tokens": 16464411.0,
"step": 8410
},
{
"entropy": 6.280291748046875,
"epoch": 0.8412055780476833,
"grad_norm": 1.03125,
"learning_rate": 0.0004938052911176392,
"loss": 6.0331,
"mean_token_accuracy": 0.15589212775230407,
"num_tokens": 16473299.0,
"step": 8415
},
{
"entropy": 6.209199094772339,
"epoch": 0.8417054031089118,
"grad_norm": 0.90625,
"learning_rate": 0.0004937969714975534,
"loss": 5.8685,
"mean_token_accuracy": 0.1721692994236946,
"num_tokens": 16483040.0,
"step": 8420
},
{
"entropy": 6.289967060089111,
"epoch": 0.8422052281701404,
"grad_norm": 1.0390625,
"learning_rate": 0.0004937886463725923,
"loss": 5.9743,
"mean_token_accuracy": 0.15987657457590104,
"num_tokens": 16492608.0,
"step": 8425
},
{
"entropy": 6.360797452926636,
"epoch": 0.842705053231369,
"grad_norm": 1.09375,
"learning_rate": 0.0004937803157429651,
"loss": 6.0987,
"mean_token_accuracy": 0.15504923164844514,
"num_tokens": 16503139.0,
"step": 8430
},
{
"entropy": 6.248201894760132,
"epoch": 0.8432048782925976,
"grad_norm": 0.984375,
"learning_rate": 0.0004937719796088815,
"loss": 5.9631,
"mean_token_accuracy": 0.1689337372779846,
"num_tokens": 16511584.0,
"step": 8435
},
{
"entropy": 6.235577869415283,
"epoch": 0.8437047033538262,
"grad_norm": 0.9453125,
"learning_rate": 0.0004937636379705511,
"loss": 5.9656,
"mean_token_accuracy": 0.1608283832669258,
"num_tokens": 16522407.0,
"step": 8440
},
{
"entropy": 6.2678779602050785,
"epoch": 0.8442045284150548,
"grad_norm": 0.90625,
"learning_rate": 0.0004937552908281839,
"loss": 5.9888,
"mean_token_accuracy": 0.15717996060848236,
"num_tokens": 16532351.0,
"step": 8445
},
{
"entropy": 6.357618188858032,
"epoch": 0.8447043534762833,
"grad_norm": 0.9765625,
"learning_rate": 0.0004937469381819899,
"loss": 6.0509,
"mean_token_accuracy": 0.15357735753059387,
"num_tokens": 16541673.0,
"step": 8450
},
{
"entropy": 6.269177341461182,
"epoch": 0.8452041785375118,
"grad_norm": 1.046875,
"learning_rate": 0.0004937385800321792,
"loss": 6.0097,
"mean_token_accuracy": 0.1666796922683716,
"num_tokens": 16551622.0,
"step": 8455
},
{
"entropy": 6.296357727050781,
"epoch": 0.8457040035987404,
"grad_norm": 1.078125,
"learning_rate": 0.0004937302163789621,
"loss": 5.9649,
"mean_token_accuracy": 0.16314167827367781,
"num_tokens": 16559074.0,
"step": 8460
},
{
"entropy": 6.256822681427002,
"epoch": 0.846203828659969,
"grad_norm": 0.9609375,
"learning_rate": 0.0004937218472225491,
"loss": 5.9538,
"mean_token_accuracy": 0.172720205783844,
"num_tokens": 16568650.0,
"step": 8465
},
{
"entropy": 6.31704683303833,
"epoch": 0.8467036537211976,
"grad_norm": 0.95703125,
"learning_rate": 0.0004937134725631506,
"loss": 5.9869,
"mean_token_accuracy": 0.16269383132457732,
"num_tokens": 16578696.0,
"step": 8470
},
{
"entropy": 6.341166591644287,
"epoch": 0.8472034787824262,
"grad_norm": 0.90234375,
"learning_rate": 0.0004937050924009775,
"loss": 6.0467,
"mean_token_accuracy": 0.1565137267112732,
"num_tokens": 16589014.0,
"step": 8475
},
{
"entropy": 6.298156881332398,
"epoch": 0.8477033038436548,
"grad_norm": 0.91796875,
"learning_rate": 0.0004936967067362406,
"loss": 5.9196,
"mean_token_accuracy": 0.16253736913204192,
"num_tokens": 16598319.0,
"step": 8480
},
{
"entropy": 6.335547542572021,
"epoch": 0.8482031289048833,
"grad_norm": 0.96484375,
"learning_rate": 0.0004936883155691507,
"loss": 6.0654,
"mean_token_accuracy": 0.15766218304634094,
"num_tokens": 16607941.0,
"step": 8485
},
{
"entropy": 6.266610908508301,
"epoch": 0.8487029539661118,
"grad_norm": 1.171875,
"learning_rate": 0.0004936799188999192,
"loss": 5.9928,
"mean_token_accuracy": 0.1601306438446045,
"num_tokens": 16618398.0,
"step": 8490
},
{
"entropy": 6.320463991165161,
"epoch": 0.8492027790273404,
"grad_norm": 1.0234375,
"learning_rate": 0.0004936715167287571,
"loss": 6.038,
"mean_token_accuracy": 0.150790324062109,
"num_tokens": 16628953.0,
"step": 8495
},
{
"entropy": 6.35287709236145,
"epoch": 0.849702604088569,
"grad_norm": 0.99609375,
"learning_rate": 0.000493663109055876,
"loss": 5.9734,
"mean_token_accuracy": 0.15737926512956618,
"num_tokens": 16638783.0,
"step": 8500
},
{
"entropy": 6.309719896316528,
"epoch": 0.8502024291497976,
"grad_norm": 0.91796875,
"learning_rate": 0.0004936546958814873,
"loss": 6.0213,
"mean_token_accuracy": 0.15978955030441283,
"num_tokens": 16647432.0,
"step": 8505
},
{
"entropy": 6.330782794952393,
"epoch": 0.8507022542110262,
"grad_norm": 1.0234375,
"learning_rate": 0.0004936462772058026,
"loss": 5.9956,
"mean_token_accuracy": 0.15184449702501296,
"num_tokens": 16656839.0,
"step": 8510
},
{
"entropy": 6.3292662620544435,
"epoch": 0.8512020792722547,
"grad_norm": 1.4765625,
"learning_rate": 0.0004936378530290339,
"loss": 6.0523,
"mean_token_accuracy": 0.1559779539704323,
"num_tokens": 16667054.0,
"step": 8515
},
{
"entropy": 6.31857476234436,
"epoch": 0.8517019043334833,
"grad_norm": 1.1484375,
"learning_rate": 0.0004936294233513931,
"loss": 5.9948,
"mean_token_accuracy": 0.15994424596428872,
"num_tokens": 16676851.0,
"step": 8520
},
{
"entropy": 6.363214349746704,
"epoch": 0.8522017293947118,
"grad_norm": 1.03125,
"learning_rate": 0.0004936209881730922,
"loss": 6.0069,
"mean_token_accuracy": 0.15777713060379028,
"num_tokens": 16686433.0,
"step": 8525
},
{
"entropy": 6.291280508041382,
"epoch": 0.8527015544559404,
"grad_norm": 0.97265625,
"learning_rate": 0.0004936125474943436,
"loss": 5.9583,
"mean_token_accuracy": 0.16686154305934905,
"num_tokens": 16695859.0,
"step": 8530
},
{
"entropy": 6.2647522449493405,
"epoch": 0.853201379517169,
"grad_norm": 0.95703125,
"learning_rate": 0.0004936041013153596,
"loss": 5.9846,
"mean_token_accuracy": 0.16125230491161346,
"num_tokens": 16704391.0,
"step": 8535
},
{
"entropy": 6.286400079727173,
"epoch": 0.8537012045783976,
"grad_norm": 0.90625,
"learning_rate": 0.0004935956496363525,
"loss": 6.086,
"mean_token_accuracy": 0.15205976963043213,
"num_tokens": 16715316.0,
"step": 8540
},
{
"entropy": 6.441294002532959,
"epoch": 0.8542010296396262,
"grad_norm": 0.91796875,
"learning_rate": 0.0004935871924575352,
"loss": 6.1031,
"mean_token_accuracy": 0.14834528043866158,
"num_tokens": 16726215.0,
"step": 8545
},
{
"entropy": 6.331367921829224,
"epoch": 0.8547008547008547,
"grad_norm": 0.93359375,
"learning_rate": 0.0004935787297791204,
"loss": 5.9988,
"mean_token_accuracy": 0.1598176211118698,
"num_tokens": 16735768.0,
"step": 8550
},
{
"entropy": 6.236656999588012,
"epoch": 0.8552006797620832,
"grad_norm": 1.0625,
"learning_rate": 0.0004935702616013209,
"loss": 5.8181,
"mean_token_accuracy": 0.17496980130672454,
"num_tokens": 16744265.0,
"step": 8555
},
{
"entropy": 6.229803228378296,
"epoch": 0.8557005048233118,
"grad_norm": 0.97265625,
"learning_rate": 0.00049356178792435,
"loss": 6.0414,
"mean_token_accuracy": 0.16270768642425537,
"num_tokens": 16753713.0,
"step": 8560
},
{
"entropy": 6.32741756439209,
"epoch": 0.8562003298845404,
"grad_norm": 1.015625,
"learning_rate": 0.0004935533087484207,
"loss": 5.9475,
"mean_token_accuracy": 0.1605185091495514,
"num_tokens": 16763670.0,
"step": 8565
},
{
"entropy": 6.283723974227906,
"epoch": 0.856700154945769,
"grad_norm": 1.0703125,
"learning_rate": 0.0004935448240737464,
"loss": 5.9774,
"mean_token_accuracy": 0.16209800243377687,
"num_tokens": 16773504.0,
"step": 8570
},
{
"entropy": 6.247960233688355,
"epoch": 0.8571999800069976,
"grad_norm": 1.078125,
"learning_rate": 0.0004935363339005405,
"loss": 6.049,
"mean_token_accuracy": 0.15743016302585602,
"num_tokens": 16784181.0,
"step": 8575
},
{
"entropy": 6.345371389389038,
"epoch": 0.8576998050682261,
"grad_norm": 0.94140625,
"learning_rate": 0.0004935278382290168,
"loss": 6.026,
"mean_token_accuracy": 0.15428334921598436,
"num_tokens": 16793610.0,
"step": 8580
},
{
"entropy": 6.3127820014953615,
"epoch": 0.8581996301294547,
"grad_norm": 0.8984375,
"learning_rate": 0.0004935193370593887,
"loss": 6.0704,
"mean_token_accuracy": 0.14957648515701294,
"num_tokens": 16804068.0,
"step": 8585
},
{
"entropy": 6.263094568252564,
"epoch": 0.8586994551906832,
"grad_norm": 1.015625,
"learning_rate": 0.0004935108303918705,
"loss": 6.0201,
"mean_token_accuracy": 0.1627806007862091,
"num_tokens": 16813769.0,
"step": 8590
},
{
"entropy": 6.258586692810058,
"epoch": 0.8591992802519118,
"grad_norm": 0.9609375,
"learning_rate": 0.0004935023182266761,
"loss": 5.952,
"mean_token_accuracy": 0.16163772642612456,
"num_tokens": 16822859.0,
"step": 8595
},
{
"entropy": 6.410567712783814,
"epoch": 0.8596991053131404,
"grad_norm": 0.84765625,
"learning_rate": 0.0004934938005640194,
"loss": 6.1838,
"mean_token_accuracy": 0.14456939920783043,
"num_tokens": 16834839.0,
"step": 8600
},
{
"entropy": 6.372849416732788,
"epoch": 0.860198930374369,
"grad_norm": 0.90625,
"learning_rate": 0.000493485277404115,
"loss": 6.0232,
"mean_token_accuracy": 0.15320356637239457,
"num_tokens": 16845205.0,
"step": 8605
},
{
"entropy": 6.325511312484741,
"epoch": 0.8606987554355976,
"grad_norm": 1.0078125,
"learning_rate": 0.0004934767487471772,
"loss": 6.0224,
"mean_token_accuracy": 0.15564613193273544,
"num_tokens": 16854874.0,
"step": 8610
},
{
"entropy": 6.2826642990112305,
"epoch": 0.8611985804968261,
"grad_norm": 0.98828125,
"learning_rate": 0.0004934682145934207,
"loss": 5.9157,
"mean_token_accuracy": 0.16233194023370742,
"num_tokens": 16864771.0,
"step": 8615
},
{
"entropy": 6.30085940361023,
"epoch": 0.8616984055580547,
"grad_norm": 1.0234375,
"learning_rate": 0.00049345967494306,
"loss": 6.0091,
"mean_token_accuracy": 0.15559957921504974,
"num_tokens": 16874227.0,
"step": 8620
},
{
"entropy": 6.271422576904297,
"epoch": 0.8621982306192832,
"grad_norm": 1.0234375,
"learning_rate": 0.0004934511297963102,
"loss": 5.9177,
"mean_token_accuracy": 0.16722774505615234,
"num_tokens": 16883895.0,
"step": 8625
},
{
"entropy": 6.267648506164551,
"epoch": 0.8626980556805118,
"grad_norm": 0.98046875,
"learning_rate": 0.0004934425791533861,
"loss": 5.9322,
"mean_token_accuracy": 0.16592230498790742,
"num_tokens": 16893012.0,
"step": 8630
},
{
"entropy": 6.2932511329650875,
"epoch": 0.8631978807417404,
"grad_norm": 1.375,
"learning_rate": 0.0004934340230145031,
"loss": 6.0014,
"mean_token_accuracy": 0.16674718037247657,
"num_tokens": 16903831.0,
"step": 8635
},
{
"entropy": 6.3498108863830565,
"epoch": 0.863697705802969,
"grad_norm": 1.015625,
"learning_rate": 0.0004934254613798761,
"loss": 6.0543,
"mean_token_accuracy": 0.14921953678131103,
"num_tokens": 16913649.0,
"step": 8640
},
{
"entropy": 6.277431535720825,
"epoch": 0.8641975308641975,
"grad_norm": 1.1875,
"learning_rate": 0.0004934168942497207,
"loss": 5.9924,
"mean_token_accuracy": 0.16252291649580003,
"num_tokens": 16923063.0,
"step": 8645
},
{
"entropy": 6.35346474647522,
"epoch": 0.8646973559254261,
"grad_norm": 0.9921875,
"learning_rate": 0.0004934083216242526,
"loss": 5.9747,
"mean_token_accuracy": 0.15674864649772643,
"num_tokens": 16932901.0,
"step": 8650
},
{
"entropy": 6.3288496971130375,
"epoch": 0.8651971809866547,
"grad_norm": 0.91796875,
"learning_rate": 0.0004933997435036873,
"loss": 6.0915,
"mean_token_accuracy": 0.14949913769960405,
"num_tokens": 16942781.0,
"step": 8655
},
{
"entropy": 6.3220373630523685,
"epoch": 0.8656970060478832,
"grad_norm": 0.953125,
"learning_rate": 0.0004933911598882405,
"loss": 5.9768,
"mean_token_accuracy": 0.1641226142644882,
"num_tokens": 16952064.0,
"step": 8660
},
{
"entropy": 6.229477071762085,
"epoch": 0.8661968311091118,
"grad_norm": 1.125,
"learning_rate": 0.0004933825707781284,
"loss": 5.9256,
"mean_token_accuracy": 0.16205176711082458,
"num_tokens": 16961528.0,
"step": 8665
},
{
"entropy": 6.289839601516723,
"epoch": 0.8666966561703404,
"grad_norm": 1.0,
"learning_rate": 0.0004933739761735671,
"loss": 6.0177,
"mean_token_accuracy": 0.159897018969059,
"num_tokens": 16971066.0,
"step": 8670
},
{
"entropy": 6.328601741790772,
"epoch": 0.867196481231569,
"grad_norm": 0.89453125,
"learning_rate": 0.0004933653760747726,
"loss": 6.0363,
"mean_token_accuracy": 0.15876379013061523,
"num_tokens": 16981438.0,
"step": 8675
},
{
"entropy": 6.296061944961548,
"epoch": 0.8676963062927975,
"grad_norm": 0.921875,
"learning_rate": 0.0004933567704819615,
"loss": 5.9097,
"mean_token_accuracy": 0.17154327780008316,
"num_tokens": 16990836.0,
"step": 8680
},
{
"entropy": 6.251953268051148,
"epoch": 0.8681961313540261,
"grad_norm": 0.88671875,
"learning_rate": 0.0004933481593953502,
"loss": 5.9034,
"mean_token_accuracy": 0.16766417026519775,
"num_tokens": 17000254.0,
"step": 8685
},
{
"entropy": 6.293018245697022,
"epoch": 0.8686959564152547,
"grad_norm": 0.88671875,
"learning_rate": 0.0004933395428151554,
"loss": 5.9693,
"mean_token_accuracy": 0.167994187772274,
"num_tokens": 17010081.0,
"step": 8690
},
{
"entropy": 6.242474794387817,
"epoch": 0.8691957814764832,
"grad_norm": 1.0546875,
"learning_rate": 0.000493330920741594,
"loss": 5.8377,
"mean_token_accuracy": 0.16764838248491287,
"num_tokens": 17018967.0,
"step": 8695
},
{
"entropy": 6.241475629806518,
"epoch": 0.8696956065377118,
"grad_norm": 0.9765625,
"learning_rate": 0.0004933222931748827,
"loss": 5.9817,
"mean_token_accuracy": 0.16205032765865326,
"num_tokens": 17027615.0,
"step": 8700
},
{
"entropy": 6.277568912506103,
"epoch": 0.8701954315989404,
"grad_norm": 0.96875,
"learning_rate": 0.0004933136601152387,
"loss": 5.9657,
"mean_token_accuracy": 0.15991829335689545,
"num_tokens": 17038446.0,
"step": 8705
},
{
"entropy": 6.330829000473022,
"epoch": 0.8706952566601689,
"grad_norm": 0.9453125,
"learning_rate": 0.0004933050215628793,
"loss": 5.9964,
"mean_token_accuracy": 0.15547903925180434,
"num_tokens": 17047787.0,
"step": 8710
},
{
"entropy": 6.290249252319336,
"epoch": 0.8711950817213975,
"grad_norm": 1.0234375,
"learning_rate": 0.0004932963775180216,
"loss": 6.0243,
"mean_token_accuracy": 0.1581503041088581,
"num_tokens": 17056842.0,
"step": 8715
},
{
"entropy": 6.253875207901001,
"epoch": 0.8716949067826261,
"grad_norm": 0.93359375,
"learning_rate": 0.0004932877279808833,
"loss": 6.0439,
"mean_token_accuracy": 0.1574680671095848,
"num_tokens": 17066217.0,
"step": 8720
},
{
"entropy": 6.38287763595581,
"epoch": 0.8721947318438547,
"grad_norm": 0.89453125,
"learning_rate": 0.0004932790729516818,
"loss": 6.0859,
"mean_token_accuracy": 0.14897339791059494,
"num_tokens": 17077760.0,
"step": 8725
},
{
"entropy": 6.287076473236084,
"epoch": 0.8726945569050832,
"grad_norm": 0.8671875,
"learning_rate": 0.0004932704124306352,
"loss": 5.9441,
"mean_token_accuracy": 0.16397355273365974,
"num_tokens": 17088251.0,
"step": 8730
},
{
"entropy": 6.324968576431274,
"epoch": 0.8731943819663118,
"grad_norm": 0.921875,
"learning_rate": 0.000493261746417961,
"loss": 6.0009,
"mean_token_accuracy": 0.1548951581120491,
"num_tokens": 17098192.0,
"step": 8735
},
{
"entropy": 6.320489978790283,
"epoch": 0.8736942070275404,
"grad_norm": 0.9921875,
"learning_rate": 0.0004932530749138775,
"loss": 5.9802,
"mean_token_accuracy": 0.15758218765258789,
"num_tokens": 17107787.0,
"step": 8740
},
{
"entropy": 6.325667953491211,
"epoch": 0.8741940320887689,
"grad_norm": 0.96875,
"learning_rate": 0.0004932443979186028,
"loss": 6.0697,
"mean_token_accuracy": 0.15468001887202262,
"num_tokens": 17118125.0,
"step": 8745
},
{
"entropy": 6.246259164810181,
"epoch": 0.8746938571499975,
"grad_norm": 0.88671875,
"learning_rate": 0.0004932357154323552,
"loss": 5.9382,
"mean_token_accuracy": 0.16619091480970383,
"num_tokens": 17127909.0,
"step": 8750
},
{
"entropy": 6.2894508838653564,
"epoch": 0.8751936822112261,
"grad_norm": 1.015625,
"learning_rate": 0.0004932270274553531,
"loss": 5.9932,
"mean_token_accuracy": 0.1639045424759388,
"num_tokens": 17137672.0,
"step": 8755
},
{
"entropy": 6.253464698791504,
"epoch": 0.8756935072724547,
"grad_norm": 1.015625,
"learning_rate": 0.0004932183339878153,
"loss": 5.9505,
"mean_token_accuracy": 0.16514331996440887,
"num_tokens": 17147151.0,
"step": 8760
},
{
"entropy": 6.269558668136597,
"epoch": 0.8761933323336832,
"grad_norm": 1.0,
"learning_rate": 0.0004932096350299602,
"loss": 5.8312,
"mean_token_accuracy": 0.16723036617040635,
"num_tokens": 17156615.0,
"step": 8765
},
{
"entropy": 6.27840633392334,
"epoch": 0.8766931573949118,
"grad_norm": 0.9765625,
"learning_rate": 0.0004932009305820069,
"loss": 5.9784,
"mean_token_accuracy": 0.16638180166482924,
"num_tokens": 17166089.0,
"step": 8770
},
{
"entropy": 6.316409873962402,
"epoch": 0.8771929824561403,
"grad_norm": 0.9140625,
"learning_rate": 0.0004931922206441743,
"loss": 6.0096,
"mean_token_accuracy": 0.1600225031375885,
"num_tokens": 17176565.0,
"step": 8775
},
{
"entropy": 6.241904830932617,
"epoch": 0.8776928075173689,
"grad_norm": 0.91796875,
"learning_rate": 0.0004931835052166817,
"loss": 5.9332,
"mean_token_accuracy": 0.1653585284948349,
"num_tokens": 17186987.0,
"step": 8780
},
{
"entropy": 6.268025064468384,
"epoch": 0.8781926325785975,
"grad_norm": 0.96875,
"learning_rate": 0.0004931747842997481,
"loss": 5.9406,
"mean_token_accuracy": 0.16368341445922852,
"num_tokens": 17196325.0,
"step": 8785
},
{
"entropy": 6.347244501113892,
"epoch": 0.8786924576398261,
"grad_norm": 1.0,
"learning_rate": 0.0004931660578935931,
"loss": 5.992,
"mean_token_accuracy": 0.16090947836637498,
"num_tokens": 17206210.0,
"step": 8790
},
{
"entropy": 6.278488969802856,
"epoch": 0.8791922827010547,
"grad_norm": 0.92578125,
"learning_rate": 0.0004931573259984363,
"loss": 5.9323,
"mean_token_accuracy": 0.16109316945075988,
"num_tokens": 17216521.0,
"step": 8795
},
{
"entropy": 6.270577621459961,
"epoch": 0.8796921077622832,
"grad_norm": 1.03125,
"learning_rate": 0.0004931485886144972,
"loss": 5.9463,
"mean_token_accuracy": 0.161488676071167,
"num_tokens": 17225125.0,
"step": 8800
},
{
"entropy": 6.2946327209472654,
"epoch": 0.8801919328235118,
"grad_norm": 0.98828125,
"learning_rate": 0.0004931398457419959,
"loss": 5.9592,
"mean_token_accuracy": 0.1619647979736328,
"num_tokens": 17234674.0,
"step": 8805
},
{
"entropy": 6.253011989593506,
"epoch": 0.8806917578847403,
"grad_norm": 0.9296875,
"learning_rate": 0.0004931310973811521,
"loss": 5.9129,
"mean_token_accuracy": 0.1603519558906555,
"num_tokens": 17244362.0,
"step": 8810
},
{
"entropy": 6.249275875091553,
"epoch": 0.8811915829459689,
"grad_norm": 1.0703125,
"learning_rate": 0.000493122343532186,
"loss": 5.8972,
"mean_token_accuracy": 0.16217835247516632,
"num_tokens": 17253541.0,
"step": 8815
},
{
"entropy": 6.262957572937012,
"epoch": 0.8816914080071975,
"grad_norm": 0.890625,
"learning_rate": 0.000493113584195318,
"loss": 5.9096,
"mean_token_accuracy": 0.1663643643260002,
"num_tokens": 17263214.0,
"step": 8820
},
{
"entropy": 6.297371959686279,
"epoch": 0.8821912330684261,
"grad_norm": 0.8828125,
"learning_rate": 0.0004931048193707681,
"loss": 5.981,
"mean_token_accuracy": 0.1590479001402855,
"num_tokens": 17273991.0,
"step": 8825
},
{
"entropy": 6.257483386993409,
"epoch": 0.8826910581296546,
"grad_norm": 0.96484375,
"learning_rate": 0.0004930960490587573,
"loss": 5.8419,
"mean_token_accuracy": 0.1723306208848953,
"num_tokens": 17283221.0,
"step": 8830
},
{
"entropy": 6.220382976531982,
"epoch": 0.8831908831908832,
"grad_norm": 1.0078125,
"learning_rate": 0.000493087273259506,
"loss": 5.9715,
"mean_token_accuracy": 0.16373782455921174,
"num_tokens": 17293520.0,
"step": 8835
},
{
"entropy": 6.228935861587525,
"epoch": 0.8836907082521117,
"grad_norm": 0.9296875,
"learning_rate": 0.0004930784919732349,
"loss": 5.9658,
"mean_token_accuracy": 0.1591949477791786,
"num_tokens": 17303881.0,
"step": 8840
},
{
"entropy": 6.232513046264648,
"epoch": 0.8841905333133403,
"grad_norm": 0.8984375,
"learning_rate": 0.0004930697052001651,
"loss": 5.879,
"mean_token_accuracy": 0.16962621510028839,
"num_tokens": 17314115.0,
"step": 8845
},
{
"entropy": 6.311823225021362,
"epoch": 0.8846903583745689,
"grad_norm": 0.984375,
"learning_rate": 0.0004930609129405177,
"loss": 6.0372,
"mean_token_accuracy": 0.15722524076700212,
"num_tokens": 17324517.0,
"step": 8850
},
{
"entropy": 6.297610759735107,
"epoch": 0.8851901834357975,
"grad_norm": 1.03125,
"learning_rate": 0.0004930521151945138,
"loss": 5.9879,
"mean_token_accuracy": 0.16122807413339615,
"num_tokens": 17334109.0,
"step": 8855
},
{
"entropy": 6.258606815338135,
"epoch": 0.8856900084970261,
"grad_norm": 0.8984375,
"learning_rate": 0.0004930433119623749,
"loss": 5.9415,
"mean_token_accuracy": 0.16207222491502762,
"num_tokens": 17344602.0,
"step": 8860
},
{
"entropy": 6.253335571289062,
"epoch": 0.8861898335582546,
"grad_norm": 0.984375,
"learning_rate": 0.0004930345032443221,
"loss": 5.916,
"mean_token_accuracy": 0.1735669493675232,
"num_tokens": 17354476.0,
"step": 8865
},
{
"entropy": 6.285289144515991,
"epoch": 0.8866896586194832,
"grad_norm": 1.015625,
"learning_rate": 0.0004930256890405777,
"loss": 5.9016,
"mean_token_accuracy": 0.16704263985157014,
"num_tokens": 17364748.0,
"step": 8870
},
{
"entropy": 6.265931892395019,
"epoch": 0.8871894836807117,
"grad_norm": 0.9921875,
"learning_rate": 0.0004930168693513627,
"loss": 5.9724,
"mean_token_accuracy": 0.16169362366199494,
"num_tokens": 17374162.0,
"step": 8875
},
{
"entropy": 6.270309209823608,
"epoch": 0.8876893087419403,
"grad_norm": 0.99609375,
"learning_rate": 0.0004930080441768995,
"loss": 6.0023,
"mean_token_accuracy": 0.1623801603913307,
"num_tokens": 17384042.0,
"step": 8880
},
{
"entropy": 6.265348434448242,
"epoch": 0.8881891338031689,
"grad_norm": 0.96484375,
"learning_rate": 0.00049299921351741,
"loss": 5.9075,
"mean_token_accuracy": 0.1636982500553131,
"num_tokens": 17394181.0,
"step": 8885
},
{
"entropy": 6.309110927581787,
"epoch": 0.8886889588643975,
"grad_norm": 0.99609375,
"learning_rate": 0.0004929903773731164,
"loss": 6.0657,
"mean_token_accuracy": 0.15426195710897445,
"num_tokens": 17402983.0,
"step": 8890
},
{
"entropy": 6.370062732696534,
"epoch": 0.8891887839256261,
"grad_norm": 0.9375,
"learning_rate": 0.0004929815357442409,
"loss": 6.1257,
"mean_token_accuracy": 0.15101911649107932,
"num_tokens": 17413442.0,
"step": 8895
},
{
"entropy": 6.289839220046997,
"epoch": 0.8896886089868546,
"grad_norm": 0.8984375,
"learning_rate": 0.0004929726886310062,
"loss": 5.9618,
"mean_token_accuracy": 0.15983574241399764,
"num_tokens": 17423644.0,
"step": 8900
},
{
"entropy": 6.32444953918457,
"epoch": 0.8901884340480831,
"grad_norm": 0.9453125,
"learning_rate": 0.0004929638360336348,
"loss": 6.001,
"mean_token_accuracy": 0.15785527378320693,
"num_tokens": 17433400.0,
"step": 8905
},
{
"entropy": 6.312784194946289,
"epoch": 0.8906882591093117,
"grad_norm": 0.9296875,
"learning_rate": 0.0004929549779523492,
"loss": 5.9725,
"mean_token_accuracy": 0.1608631819486618,
"num_tokens": 17443146.0,
"step": 8910
},
{
"entropy": 6.181512308120728,
"epoch": 0.8911880841705403,
"grad_norm": 0.9609375,
"learning_rate": 0.0004929461143873725,
"loss": 5.8509,
"mean_token_accuracy": 0.17120044827461242,
"num_tokens": 17452880.0,
"step": 8915
},
{
"entropy": 6.216238307952881,
"epoch": 0.8916879092317689,
"grad_norm": 0.96484375,
"learning_rate": 0.0004929372453389277,
"loss": 5.8919,
"mean_token_accuracy": 0.16418985575437545,
"num_tokens": 17461633.0,
"step": 8920
},
{
"entropy": 6.371498966217041,
"epoch": 0.8921877342929975,
"grad_norm": 0.9609375,
"learning_rate": 0.0004929283708072378,
"loss": 6.0573,
"mean_token_accuracy": 0.16026912182569503,
"num_tokens": 17470892.0,
"step": 8925
},
{
"entropy": 6.312393712997436,
"epoch": 0.8926875593542261,
"grad_norm": 0.9375,
"learning_rate": 0.0004929194907925262,
"loss": 5.9447,
"mean_token_accuracy": 0.15884686559438704,
"num_tokens": 17481457.0,
"step": 8930
},
{
"entropy": 6.2684807777404785,
"epoch": 0.8931873844154546,
"grad_norm": 1.09375,
"learning_rate": 0.0004929106052950163,
"loss": 5.9879,
"mean_token_accuracy": 0.15770442113280297,
"num_tokens": 17490387.0,
"step": 8935
},
{
"entropy": 6.284839916229248,
"epoch": 0.8936872094766831,
"grad_norm": 1.0234375,
"learning_rate": 0.0004929017143149317,
"loss": 5.9446,
"mean_token_accuracy": 0.1668263405561447,
"num_tokens": 17499478.0,
"step": 8940
},
{
"entropy": 6.33533034324646,
"epoch": 0.8941870345379117,
"grad_norm": 1.1171875,
"learning_rate": 0.000492892817852496,
"loss": 6.0577,
"mean_token_accuracy": 0.15721446871757508,
"num_tokens": 17509396.0,
"step": 8945
},
{
"entropy": 6.289205551147461,
"epoch": 0.8946868595991403,
"grad_norm": 0.92578125,
"learning_rate": 0.000492883915907933,
"loss": 6.0016,
"mean_token_accuracy": 0.16132455170154572,
"num_tokens": 17519773.0,
"step": 8950
},
{
"entropy": 6.268876218795777,
"epoch": 0.8951866846603689,
"grad_norm": 0.9453125,
"learning_rate": 0.0004928750084814668,
"loss": 5.9595,
"mean_token_accuracy": 0.16649766266345978,
"num_tokens": 17529873.0,
"step": 8955
},
{
"entropy": 6.207161092758179,
"epoch": 0.8956865097215975,
"grad_norm": 0.94140625,
"learning_rate": 0.0004928660955733214,
"loss": 5.9124,
"mean_token_accuracy": 0.1666805163025856,
"num_tokens": 17538642.0,
"step": 8960
},
{
"entropy": 6.30468258857727,
"epoch": 0.896186334782826,
"grad_norm": 1.046875,
"learning_rate": 0.0004928571771837211,
"loss": 5.9432,
"mean_token_accuracy": 0.15742214620113373,
"num_tokens": 17546892.0,
"step": 8965
},
{
"entropy": 6.354885482788086,
"epoch": 0.8966861598440545,
"grad_norm": 0.99609375,
"learning_rate": 0.0004928482533128902,
"loss": 5.9948,
"mean_token_accuracy": 0.15852722972631456,
"num_tokens": 17556283.0,
"step": 8970
},
{
"entropy": 6.2500628471374515,
"epoch": 0.8971859849052831,
"grad_norm": 0.91796875,
"learning_rate": 0.0004928393239610535,
"loss": 5.9317,
"mean_token_accuracy": 0.15904702469706536,
"num_tokens": 17567021.0,
"step": 8975
},
{
"entropy": 6.286598110198975,
"epoch": 0.8976858099665117,
"grad_norm": 0.95703125,
"learning_rate": 0.0004928303891284354,
"loss": 5.9404,
"mean_token_accuracy": 0.16273575872182847,
"num_tokens": 17575927.0,
"step": 8980
},
{
"entropy": 6.2996362209320065,
"epoch": 0.8981856350277403,
"grad_norm": 0.96484375,
"learning_rate": 0.0004928214488152607,
"loss": 5.9861,
"mean_token_accuracy": 0.16216324120759965,
"num_tokens": 17585882.0,
"step": 8985
},
{
"entropy": 6.197223091125489,
"epoch": 0.8986854600889689,
"grad_norm": 0.8671875,
"learning_rate": 0.0004928125030217545,
"loss": 5.9243,
"mean_token_accuracy": 0.16700368672609328,
"num_tokens": 17595998.0,
"step": 8990
},
{
"entropy": 6.353938865661621,
"epoch": 0.8991852851501975,
"grad_norm": 0.921875,
"learning_rate": 0.0004928035517481416,
"loss": 5.9475,
"mean_token_accuracy": 0.16484413295984268,
"num_tokens": 17605177.0,
"step": 8995
},
{
"entropy": 6.328395652770996,
"epoch": 0.899685110211426,
"grad_norm": 0.9921875,
"learning_rate": 0.0004927945949946476,
"loss": 6.0537,
"mean_token_accuracy": 0.15768050402402878,
"num_tokens": 17615089.0,
"step": 9000
},
{
"epoch": 0.899685110211426,
"eval_entropy": 6.089246969313942,
"eval_loss": 5.979366302490234,
"eval_mean_token_accuracy": 0.16835626312266364,
"eval_num_tokens": 17615089.0,
"eval_runtime": 18.4396,
"eval_samples_per_second": 1683.607,
"eval_steps_per_second": 210.471,
"step": 9000
}
],
"logging_steps": 5,
"max_steps": 100030,
"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": 4147004648816640.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink