fpadovani/dan-latn-10mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
dan-latn-10mb-100mb_seed3407/checkpoint-15000/trainer_state.json
ModelHub XC cc5255ece4 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/dan-latn-10mb-100mb_seed3407
Source: Original Platform
2026-08-03 16:17:23 +08:00

30090 lines
827 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.252923488138991,
"eval_steps": 3000,
"global_step": 15000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.742597389221192,
"epoch": 0.000417641162712997,
"grad_norm": 5.3125,
"learning_rate": 2e-06,
"loss": 10.7497,
"mean_token_accuracy": 0.00011876485077664257,
"num_tokens": 9371.0,
"step": 5
},
{
"entropy": 10.742581462860107,
"epoch": 0.000835282325425994,
"grad_norm": 5.15625,
"learning_rate": 4.5e-06,
"loss": 10.7312,
"mean_token_accuracy": 0.0,
"num_tokens": 18155.0,
"step": 10
},
{
"entropy": 10.742617225646972,
"epoch": 0.001252923488138991,
"grad_norm": 4.96875,
"learning_rate": 7e-06,
"loss": 10.7253,
"mean_token_accuracy": 9.920635493472218e-05,
"num_tokens": 27776.0,
"step": 15
},
{
"entropy": 10.742653942108154,
"epoch": 0.001670564650851988,
"grad_norm": 5.09375,
"learning_rate": 9.5e-06,
"loss": 10.6672,
"mean_token_accuracy": 0.0005802143830806016,
"num_tokens": 37166.0,
"step": 20
},
{
"entropy": 10.74262170791626,
"epoch": 0.002088205813564985,
"grad_norm": 5.09375,
"learning_rate": 1.2e-05,
"loss": 10.5444,
"mean_token_accuracy": 0.010037525719963015,
"num_tokens": 46667.0,
"step": 25
},
{
"entropy": 10.742273426055908,
"epoch": 0.002505846976277982,
"grad_norm": 4.1875,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.4193,
"mean_token_accuracy": 0.0329294515773654,
"num_tokens": 55738.0,
"step": 30
},
{
"entropy": 10.74118251800537,
"epoch": 0.002923488138990979,
"grad_norm": 3.234375,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.3134,
"mean_token_accuracy": 0.04002654887735844,
"num_tokens": 65303.0,
"step": 35
},
{
"entropy": 10.738407230377197,
"epoch": 0.003341129301703976,
"grad_norm": 2.78125,
"learning_rate": 1.95e-05,
"loss": 10.1216,
"mean_token_accuracy": 0.041297851502895354,
"num_tokens": 74876.0,
"step": 40
},
{
"entropy": 10.734228038787842,
"epoch": 0.003758770464416973,
"grad_norm": 2.203125,
"learning_rate": 2.2e-05,
"loss": 10.0124,
"mean_token_accuracy": 0.04665607586503029,
"num_tokens": 84683.0,
"step": 45
},
{
"entropy": 10.730060386657716,
"epoch": 0.00417641162712997,
"grad_norm": 2.25,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.9446,
"mean_token_accuracy": 0.04418331272900104,
"num_tokens": 94511.0,
"step": 50
},
{
"entropy": 10.727671051025391,
"epoch": 0.004594052789842967,
"grad_norm": 2.015625,
"learning_rate": 2.7e-05,
"loss": 9.8593,
"mean_token_accuracy": 0.04506920985877514,
"num_tokens": 103820.0,
"step": 55
},
{
"entropy": 10.72537727355957,
"epoch": 0.005011693952555964,
"grad_norm": 1.984375,
"learning_rate": 2.95e-05,
"loss": 9.8006,
"mean_token_accuracy": 0.04406389705836773,
"num_tokens": 112398.0,
"step": 60
},
{
"entropy": 10.723227405548096,
"epoch": 0.005429335115268961,
"grad_norm": 2.046875,
"learning_rate": 3.2e-05,
"loss": 9.7404,
"mean_token_accuracy": 0.041664719581604004,
"num_tokens": 121167.0,
"step": 65
},
{
"entropy": 10.721342849731446,
"epoch": 0.005846976277981958,
"grad_norm": 1.9765625,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.67,
"mean_token_accuracy": 0.04505793582648039,
"num_tokens": 129399.0,
"step": 70
},
{
"entropy": 10.717985439300538,
"epoch": 0.006264617440694955,
"grad_norm": 2.03125,
"learning_rate": 3.7e-05,
"loss": 9.6274,
"mean_token_accuracy": 0.043670152127742765,
"num_tokens": 138512.0,
"step": 75
},
{
"entropy": 10.714579200744629,
"epoch": 0.006682258603407952,
"grad_norm": 1.84375,
"learning_rate": 3.95e-05,
"loss": 9.5989,
"mean_token_accuracy": 0.0400418970733881,
"num_tokens": 147712.0,
"step": 80
},
{
"entropy": 10.709438896179199,
"epoch": 0.007099899766120949,
"grad_norm": 1.9140625,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.5111,
"mean_token_accuracy": 0.04463256634771824,
"num_tokens": 156759.0,
"step": 85
},
{
"entropy": 10.70057258605957,
"epoch": 0.007517540928833946,
"grad_norm": 2.0,
"learning_rate": 4.45e-05,
"loss": 9.4327,
"mean_token_accuracy": 0.044699374586343765,
"num_tokens": 166395.0,
"step": 90
},
{
"entropy": 10.690051460266114,
"epoch": 0.007935182091546943,
"grad_norm": 1.9609375,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.3248,
"mean_token_accuracy": 0.05507284291088581,
"num_tokens": 174739.0,
"step": 95
},
{
"entropy": 10.674839115142822,
"epoch": 0.00835282325425994,
"grad_norm": 1.890625,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.1931,
"mean_token_accuracy": 0.05993704758584499,
"num_tokens": 183010.0,
"step": 100
},
{
"entropy": 10.653488159179688,
"epoch": 0.008770464416972937,
"grad_norm": 1.8671875,
"learning_rate": 5.2e-05,
"loss": 9.16,
"mean_token_accuracy": 0.053666260465979575,
"num_tokens": 192589.0,
"step": 105
},
{
"entropy": 10.622858619689941,
"epoch": 0.009188105579685933,
"grad_norm": 1.890625,
"learning_rate": 5.45e-05,
"loss": 9.013,
"mean_token_accuracy": 0.06004327610135078,
"num_tokens": 201638.0,
"step": 110
},
{
"entropy": 10.569721031188966,
"epoch": 0.00960574674239893,
"grad_norm": 1.9296875,
"learning_rate": 5.7e-05,
"loss": 8.89,
"mean_token_accuracy": 0.05817726403474808,
"num_tokens": 209862.0,
"step": 115
},
{
"entropy": 10.519128704071045,
"epoch": 0.010023387905111928,
"grad_norm": 1.703125,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.8803,
"mean_token_accuracy": 0.04828422553837299,
"num_tokens": 218897.0,
"step": 120
},
{
"entropy": 10.48040943145752,
"epoch": 0.010441029067824925,
"grad_norm": 1.7109375,
"learning_rate": 6.2e-05,
"loss": 8.7326,
"mean_token_accuracy": 0.057341094315052035,
"num_tokens": 227682.0,
"step": 125
},
{
"entropy": 10.404988002777099,
"epoch": 0.010858670230537921,
"grad_norm": 1.734375,
"learning_rate": 6.450000000000001e-05,
"loss": 8.5531,
"mean_token_accuracy": 0.05646193064749241,
"num_tokens": 236170.0,
"step": 130
},
{
"entropy": 10.337473011016845,
"epoch": 0.011276311393250919,
"grad_norm": 1.53125,
"learning_rate": 6.7e-05,
"loss": 8.5016,
"mean_token_accuracy": 0.054118812829256055,
"num_tokens": 245710.0,
"step": 135
},
{
"entropy": 10.269655227661133,
"epoch": 0.011693952555963916,
"grad_norm": 1.7421875,
"learning_rate": 6.950000000000001e-05,
"loss": 8.3193,
"mean_token_accuracy": 0.06037080101668835,
"num_tokens": 254751.0,
"step": 140
},
{
"entropy": 10.123066329956055,
"epoch": 0.012111593718676913,
"grad_norm": 1.5390625,
"learning_rate": 7.2e-05,
"loss": 8.1922,
"mean_token_accuracy": 0.06263931393623352,
"num_tokens": 264210.0,
"step": 145
},
{
"entropy": 10.026834297180176,
"epoch": 0.01252923488138991,
"grad_norm": 1.5,
"learning_rate": 7.45e-05,
"loss": 8.1098,
"mean_token_accuracy": 0.06494732424616814,
"num_tokens": 272944.0,
"step": 150
},
{
"entropy": 9.923717784881593,
"epoch": 0.012946876044102907,
"grad_norm": 1.5078125,
"learning_rate": 7.7e-05,
"loss": 7.9366,
"mean_token_accuracy": 0.06645521372556687,
"num_tokens": 281311.0,
"step": 155
},
{
"entropy": 9.686262702941894,
"epoch": 0.013364517206815904,
"grad_norm": 1.203125,
"learning_rate": 7.950000000000001e-05,
"loss": 7.8102,
"mean_token_accuracy": 0.06696667969226837,
"num_tokens": 289701.0,
"step": 160
},
{
"entropy": 9.53008050918579,
"epoch": 0.013782158369528901,
"grad_norm": 1.1328125,
"learning_rate": 8.2e-05,
"loss": 7.8033,
"mean_token_accuracy": 0.06480560302734376,
"num_tokens": 299459.0,
"step": 165
},
{
"entropy": 9.334712505340576,
"epoch": 0.014199799532241897,
"grad_norm": 1.125,
"learning_rate": 8.450000000000001e-05,
"loss": 7.7276,
"mean_token_accuracy": 0.06751208230853081,
"num_tokens": 309343.0,
"step": 170
},
{
"entropy": 9.117801856994628,
"epoch": 0.014617440694954895,
"grad_norm": 1.125,
"learning_rate": 8.7e-05,
"loss": 7.6628,
"mean_token_accuracy": 0.06329603157937527,
"num_tokens": 318840.0,
"step": 175
},
{
"entropy": 8.88758544921875,
"epoch": 0.015035081857667892,
"grad_norm": 1.046875,
"learning_rate": 8.95e-05,
"loss": 7.4494,
"mean_token_accuracy": 0.07078598737716675,
"num_tokens": 328689.0,
"step": 180
},
{
"entropy": 8.713505363464355,
"epoch": 0.01545272302038089,
"grad_norm": 1.015625,
"learning_rate": 9.2e-05,
"loss": 7.5296,
"mean_token_accuracy": 0.07133941762149335,
"num_tokens": 338209.0,
"step": 185
},
{
"entropy": 8.533864212036132,
"epoch": 0.015870364183093885,
"grad_norm": 0.875,
"learning_rate": 9.45e-05,
"loss": 7.3715,
"mean_token_accuracy": 0.07491575330495834,
"num_tokens": 347270.0,
"step": 190
},
{
"entropy": 8.331917572021485,
"epoch": 0.016288005345806884,
"grad_norm": 1.21875,
"learning_rate": 9.7e-05,
"loss": 7.3864,
"mean_token_accuracy": 0.07862697467207909,
"num_tokens": 356366.0,
"step": 195
},
{
"entropy": 8.292905139923096,
"epoch": 0.01670564650851988,
"grad_norm": 1.078125,
"learning_rate": 9.95e-05,
"loss": 7.3321,
"mean_token_accuracy": 0.07781913727521897,
"num_tokens": 365980.0,
"step": 200
},
{
"entropy": 8.17343864440918,
"epoch": 0.017123287671232876,
"grad_norm": 0.94140625,
"learning_rate": 0.000102,
"loss": 7.2859,
"mean_token_accuracy": 0.08276912495493889,
"num_tokens": 374591.0,
"step": 205
},
{
"entropy": 8.114312171936035,
"epoch": 0.017540928833945875,
"grad_norm": 0.9609375,
"learning_rate": 0.00010449999999999999,
"loss": 7.2652,
"mean_token_accuracy": 0.08255952596664429,
"num_tokens": 383250.0,
"step": 210
},
{
"entropy": 8.142033052444457,
"epoch": 0.01795856999665887,
"grad_norm": 1.109375,
"learning_rate": 0.000107,
"loss": 7.1955,
"mean_token_accuracy": 0.07966764494776726,
"num_tokens": 391502.0,
"step": 215
},
{
"entropy": 8.033121585845947,
"epoch": 0.018376211159371866,
"grad_norm": 1.140625,
"learning_rate": 0.0001095,
"loss": 7.1094,
"mean_token_accuracy": 0.08625469133257865,
"num_tokens": 400225.0,
"step": 220
},
{
"entropy": 8.037012720108033,
"epoch": 0.018793852322084866,
"grad_norm": 1.0,
"learning_rate": 0.000112,
"loss": 7.3113,
"mean_token_accuracy": 0.07627052702009678,
"num_tokens": 410008.0,
"step": 225
},
{
"entropy": 7.976403427124024,
"epoch": 0.01921149348479786,
"grad_norm": 1.0,
"learning_rate": 0.0001145,
"loss": 7.2122,
"mean_token_accuracy": 0.08434663638472557,
"num_tokens": 418979.0,
"step": 230
},
{
"entropy": 7.954119348526001,
"epoch": 0.01962913464751086,
"grad_norm": 0.953125,
"learning_rate": 0.00011700000000000001,
"loss": 7.1858,
"mean_token_accuracy": 0.08367608934640884,
"num_tokens": 428139.0,
"step": 235
},
{
"entropy": 7.8628496646881105,
"epoch": 0.020046775810223856,
"grad_norm": 1.1171875,
"learning_rate": 0.00011949999999999999,
"loss": 7.1046,
"mean_token_accuracy": 0.08570514917373658,
"num_tokens": 437113.0,
"step": 240
},
{
"entropy": 7.932403039932251,
"epoch": 0.020464416972936852,
"grad_norm": 1.09375,
"learning_rate": 0.000122,
"loss": 7.2954,
"mean_token_accuracy": 0.08083686679601669,
"num_tokens": 446879.0,
"step": 245
},
{
"entropy": 7.807932806015015,
"epoch": 0.02088205813564985,
"grad_norm": 1.03125,
"learning_rate": 0.0001245,
"loss": 7.1784,
"mean_token_accuracy": 0.08404311686754226,
"num_tokens": 456362.0,
"step": 250
},
{
"entropy": 7.829455995559693,
"epoch": 0.021299699298362847,
"grad_norm": 1.25,
"learning_rate": 0.000127,
"loss": 6.9918,
"mean_token_accuracy": 0.09061411544680595,
"num_tokens": 464930.0,
"step": 255
},
{
"entropy": 7.877001619338989,
"epoch": 0.021717340461075842,
"grad_norm": 1.0234375,
"learning_rate": 0.0001295,
"loss": 7.1647,
"mean_token_accuracy": 0.08374429196119308,
"num_tokens": 474303.0,
"step": 260
},
{
"entropy": 7.794501399993896,
"epoch": 0.02213498162378884,
"grad_norm": 0.9921875,
"learning_rate": 0.000132,
"loss": 7.1539,
"mean_token_accuracy": 0.08738574758172035,
"num_tokens": 483750.0,
"step": 265
},
{
"entropy": 7.762907552719116,
"epoch": 0.022552622786501837,
"grad_norm": 1.03125,
"learning_rate": 0.00013450000000000002,
"loss": 7.0529,
"mean_token_accuracy": 0.09093769788742065,
"num_tokens": 493325.0,
"step": 270
},
{
"entropy": 7.741158294677734,
"epoch": 0.022970263949214833,
"grad_norm": 1.078125,
"learning_rate": 0.00013700000000000002,
"loss": 7.0377,
"mean_token_accuracy": 0.08475295156240463,
"num_tokens": 501907.0,
"step": 275
},
{
"entropy": 7.704169273376465,
"epoch": 0.023387905111927832,
"grad_norm": 1.1328125,
"learning_rate": 0.0001395,
"loss": 7.0759,
"mean_token_accuracy": 0.09049696251749992,
"num_tokens": 511522.0,
"step": 280
},
{
"entropy": 7.68439040184021,
"epoch": 0.023805546274640828,
"grad_norm": 0.94921875,
"learning_rate": 0.00014199999999999998,
"loss": 7.0277,
"mean_token_accuracy": 0.08685763627290725,
"num_tokens": 521287.0,
"step": 285
},
{
"entropy": 7.689414691925049,
"epoch": 0.024223187437353827,
"grad_norm": 1.09375,
"learning_rate": 0.0001445,
"loss": 7.1013,
"mean_token_accuracy": 0.08630543872714043,
"num_tokens": 530631.0,
"step": 290
},
{
"entropy": 7.624138498306275,
"epoch": 0.024640828600066823,
"grad_norm": 0.97265625,
"learning_rate": 0.000147,
"loss": 6.9712,
"mean_token_accuracy": 0.09986182302236557,
"num_tokens": 540386.0,
"step": 295
},
{
"entropy": 7.671041774749756,
"epoch": 0.02505846976277982,
"grad_norm": 1.0078125,
"learning_rate": 0.0001495,
"loss": 7.0573,
"mean_token_accuracy": 0.08439475223422051,
"num_tokens": 550121.0,
"step": 300
},
{
"entropy": 7.569111251831055,
"epoch": 0.025476110925492818,
"grad_norm": 0.9921875,
"learning_rate": 0.000152,
"loss": 6.9112,
"mean_token_accuracy": 0.09254205971956253,
"num_tokens": 559460.0,
"step": 305
},
{
"entropy": 7.6026232719421385,
"epoch": 0.025893752088205813,
"grad_norm": 1.0078125,
"learning_rate": 0.00015450000000000001,
"loss": 6.8726,
"mean_token_accuracy": 0.0958409771323204,
"num_tokens": 568392.0,
"step": 310
},
{
"entropy": 7.504022693634033,
"epoch": 0.02631139325091881,
"grad_norm": 0.875,
"learning_rate": 0.000157,
"loss": 6.9599,
"mean_token_accuracy": 0.09224955886602401,
"num_tokens": 578248.0,
"step": 315
},
{
"entropy": 7.5831845760345455,
"epoch": 0.026729034413631808,
"grad_norm": 0.9453125,
"learning_rate": 0.0001595,
"loss": 7.0295,
"mean_token_accuracy": 0.08987967744469642,
"num_tokens": 588164.0,
"step": 320
},
{
"entropy": 7.5552794456481935,
"epoch": 0.027146675576344804,
"grad_norm": 0.8828125,
"learning_rate": 0.000162,
"loss": 6.9487,
"mean_token_accuracy": 0.09430753514170646,
"num_tokens": 598715.0,
"step": 325
},
{
"entropy": 7.5547398090362545,
"epoch": 0.027564316739057803,
"grad_norm": 1.125,
"learning_rate": 0.00016450000000000001,
"loss": 6.8841,
"mean_token_accuracy": 0.09914051964879037,
"num_tokens": 606939.0,
"step": 330
},
{
"entropy": 7.417043685913086,
"epoch": 0.0279819579017708,
"grad_norm": 1.1953125,
"learning_rate": 0.00016700000000000002,
"loss": 6.9714,
"mean_token_accuracy": 0.09765534549951553,
"num_tokens": 615803.0,
"step": 335
},
{
"entropy": 7.536453628540039,
"epoch": 0.028399599064483794,
"grad_norm": 1.203125,
"learning_rate": 0.00016950000000000003,
"loss": 6.8297,
"mean_token_accuracy": 0.09691216871142387,
"num_tokens": 624730.0,
"step": 340
},
{
"entropy": 7.539470243453979,
"epoch": 0.028817240227196794,
"grad_norm": 1.1484375,
"learning_rate": 0.00017199999999999998,
"loss": 6.931,
"mean_token_accuracy": 0.09587549567222595,
"num_tokens": 634082.0,
"step": 345
},
{
"entropy": 7.414178943634033,
"epoch": 0.02923488138990979,
"grad_norm": 0.953125,
"learning_rate": 0.00017449999999999999,
"loss": 6.9083,
"mean_token_accuracy": 0.09485536590218543,
"num_tokens": 643367.0,
"step": 350
},
{
"entropy": 7.387429285049438,
"epoch": 0.029652522552622785,
"grad_norm": 1.015625,
"learning_rate": 0.000177,
"loss": 6.7242,
"mean_token_accuracy": 0.09872433170676231,
"num_tokens": 651861.0,
"step": 355
},
{
"entropy": 7.464349985122681,
"epoch": 0.030070163715335784,
"grad_norm": 0.97265625,
"learning_rate": 0.0001795,
"loss": 6.8556,
"mean_token_accuracy": 0.09326324984431267,
"num_tokens": 661064.0,
"step": 360
},
{
"entropy": 7.392336225509643,
"epoch": 0.03048780487804878,
"grad_norm": 1.0859375,
"learning_rate": 0.000182,
"loss": 6.8352,
"mean_token_accuracy": 0.10462646037340165,
"num_tokens": 669299.0,
"step": 365
},
{
"entropy": 7.248696994781494,
"epoch": 0.03090544604076178,
"grad_norm": 1.109375,
"learning_rate": 0.0001845,
"loss": 6.8988,
"mean_token_accuracy": 0.09198433235287666,
"num_tokens": 678631.0,
"step": 370
},
{
"entropy": 7.445321846008301,
"epoch": 0.03132308720347477,
"grad_norm": 1.046875,
"learning_rate": 0.000187,
"loss": 6.8925,
"mean_token_accuracy": 0.09772585704922676,
"num_tokens": 688511.0,
"step": 375
},
{
"entropy": 7.484406995773315,
"epoch": 0.03174072836618777,
"grad_norm": 1.1640625,
"learning_rate": 0.0001895,
"loss": 6.8465,
"mean_token_accuracy": 0.09334805831313134,
"num_tokens": 697956.0,
"step": 380
},
{
"entropy": 7.3463782787323,
"epoch": 0.03215836952890077,
"grad_norm": 1.140625,
"learning_rate": 0.000192,
"loss": 6.7552,
"mean_token_accuracy": 0.10220230743288994,
"num_tokens": 706779.0,
"step": 385
},
{
"entropy": 7.295664262771607,
"epoch": 0.03257601069161377,
"grad_norm": 1.03125,
"learning_rate": 0.0001945,
"loss": 6.7444,
"mean_token_accuracy": 0.09592411890625954,
"num_tokens": 715749.0,
"step": 390
},
{
"entropy": 7.331249046325683,
"epoch": 0.03299365185432676,
"grad_norm": 1.15625,
"learning_rate": 0.00019700000000000002,
"loss": 6.7177,
"mean_token_accuracy": 0.09796938598155976,
"num_tokens": 724756.0,
"step": 395
},
{
"entropy": 7.257160568237305,
"epoch": 0.03341129301703976,
"grad_norm": 1.1484375,
"learning_rate": 0.00019950000000000002,
"loss": 6.7422,
"mean_token_accuracy": 0.09832950606942177,
"num_tokens": 733288.0,
"step": 400
},
{
"entropy": 7.405258226394653,
"epoch": 0.03382893417975276,
"grad_norm": 1.609375,
"learning_rate": 0.000202,
"loss": 6.8476,
"mean_token_accuracy": 0.0943653255701065,
"num_tokens": 741629.0,
"step": 405
},
{
"entropy": 7.263295269012451,
"epoch": 0.03424657534246575,
"grad_norm": 1.015625,
"learning_rate": 0.00020449999999999998,
"loss": 6.7841,
"mean_token_accuracy": 0.1012955941259861,
"num_tokens": 751336.0,
"step": 410
},
{
"entropy": 7.402045774459839,
"epoch": 0.03466421650517875,
"grad_norm": 1.1796875,
"learning_rate": 0.000207,
"loss": 6.856,
"mean_token_accuracy": 0.09290716797113419,
"num_tokens": 759725.0,
"step": 415
},
{
"entropy": 7.322671508789062,
"epoch": 0.03508185766789175,
"grad_norm": 1.1171875,
"learning_rate": 0.0002095,
"loss": 6.845,
"mean_token_accuracy": 0.09813571646809578,
"num_tokens": 769752.0,
"step": 420
},
{
"entropy": 7.268389272689819,
"epoch": 0.03549949883060474,
"grad_norm": 1.15625,
"learning_rate": 0.000212,
"loss": 6.7365,
"mean_token_accuracy": 0.10029055625200271,
"num_tokens": 779124.0,
"step": 425
},
{
"entropy": 7.208940410614014,
"epoch": 0.03591713999331774,
"grad_norm": 1.2109375,
"learning_rate": 0.0002145,
"loss": 6.7024,
"mean_token_accuracy": 0.10402928516268731,
"num_tokens": 788892.0,
"step": 430
},
{
"entropy": 7.189973258972168,
"epoch": 0.03633478115603074,
"grad_norm": 1.0859375,
"learning_rate": 0.00021700000000000002,
"loss": 6.6662,
"mean_token_accuracy": 0.10119749382138252,
"num_tokens": 798265.0,
"step": 435
},
{
"entropy": 7.194095945358276,
"epoch": 0.03675242231874373,
"grad_norm": 1.1328125,
"learning_rate": 0.0002195,
"loss": 6.6231,
"mean_token_accuracy": 0.10882524475455284,
"num_tokens": 806896.0,
"step": 440
},
{
"entropy": 7.201065158843994,
"epoch": 0.03717006348145673,
"grad_norm": 1.203125,
"learning_rate": 0.000222,
"loss": 6.689,
"mean_token_accuracy": 0.10466736480593682,
"num_tokens": 815540.0,
"step": 445
},
{
"entropy": 7.163028383255005,
"epoch": 0.03758770464416973,
"grad_norm": 1.1484375,
"learning_rate": 0.0002245,
"loss": 6.7443,
"mean_token_accuracy": 0.10173500478267669,
"num_tokens": 824555.0,
"step": 450
},
{
"entropy": 7.269956779479981,
"epoch": 0.03800534580688272,
"grad_norm": 1.171875,
"learning_rate": 0.00022700000000000002,
"loss": 6.7737,
"mean_token_accuracy": 0.10270200073719024,
"num_tokens": 833414.0,
"step": 455
},
{
"entropy": 7.098545503616333,
"epoch": 0.03842298696959572,
"grad_norm": 1.03125,
"learning_rate": 0.00022950000000000002,
"loss": 6.6325,
"mean_token_accuracy": 0.10786343663930893,
"num_tokens": 842644.0,
"step": 460
},
{
"entropy": 7.197147083282471,
"epoch": 0.03884062813230872,
"grad_norm": 1.125,
"learning_rate": 0.00023200000000000003,
"loss": 6.6693,
"mean_token_accuracy": 0.10085522830486297,
"num_tokens": 851786.0,
"step": 465
},
{
"entropy": 7.095899295806885,
"epoch": 0.03925826929502172,
"grad_norm": 0.9921875,
"learning_rate": 0.00023449999999999998,
"loss": 6.7144,
"mean_token_accuracy": 0.1063760556280613,
"num_tokens": 861508.0,
"step": 470
},
{
"entropy": 7.154467391967773,
"epoch": 0.03967591045773471,
"grad_norm": 1.0703125,
"learning_rate": 0.000237,
"loss": 6.5905,
"mean_token_accuracy": 0.10809945836663246,
"num_tokens": 870433.0,
"step": 475
},
{
"entropy": 7.184867238998413,
"epoch": 0.04009355162044771,
"grad_norm": 1.125,
"learning_rate": 0.0002395,
"loss": 6.6928,
"mean_token_accuracy": 0.10560849010944366,
"num_tokens": 879059.0,
"step": 480
},
{
"entropy": 7.1450403213500975,
"epoch": 0.04051119278316071,
"grad_norm": 1.125,
"learning_rate": 0.000242,
"loss": 6.7403,
"mean_token_accuracy": 0.09940413013100624,
"num_tokens": 888009.0,
"step": 485
},
{
"entropy": 7.092233419418335,
"epoch": 0.040928833945873704,
"grad_norm": 1.1953125,
"learning_rate": 0.0002445,
"loss": 6.6089,
"mean_token_accuracy": 0.10285484269261361,
"num_tokens": 897335.0,
"step": 490
},
{
"entropy": 7.264479684829712,
"epoch": 0.0413464751085867,
"grad_norm": 1.0625,
"learning_rate": 0.000247,
"loss": 6.6248,
"mean_token_accuracy": 0.10704769417643548,
"num_tokens": 905802.0,
"step": 495
},
{
"entropy": 7.013559818267822,
"epoch": 0.0417641162712997,
"grad_norm": 1.1484375,
"learning_rate": 0.0002495,
"loss": 6.6091,
"mean_token_accuracy": 0.1084491603076458,
"num_tokens": 914836.0,
"step": 500
},
{
"entropy": 6.988007974624634,
"epoch": 0.042181757434012694,
"grad_norm": 1.09375,
"learning_rate": 0.000252,
"loss": 6.5574,
"mean_token_accuracy": 0.11251411363482475,
"num_tokens": 924254.0,
"step": 505
},
{
"entropy": 7.116152048110962,
"epoch": 0.04259939859672569,
"grad_norm": 1.0546875,
"learning_rate": 0.0002545,
"loss": 6.6555,
"mean_token_accuracy": 0.1060157686471939,
"num_tokens": 933841.0,
"step": 510
},
{
"entropy": 7.129690933227539,
"epoch": 0.04301703975943869,
"grad_norm": 0.99609375,
"learning_rate": 0.000257,
"loss": 6.7157,
"mean_token_accuracy": 0.10084601864218712,
"num_tokens": 943304.0,
"step": 515
},
{
"entropy": 7.06993350982666,
"epoch": 0.043434680922151685,
"grad_norm": 1.21875,
"learning_rate": 0.0002595,
"loss": 6.5385,
"mean_token_accuracy": 0.11060969755053521,
"num_tokens": 952915.0,
"step": 520
},
{
"entropy": 7.076866674423218,
"epoch": 0.043852322084864684,
"grad_norm": 1.0703125,
"learning_rate": 0.000262,
"loss": 6.598,
"mean_token_accuracy": 0.10570925623178482,
"num_tokens": 962752.0,
"step": 525
},
{
"entropy": 7.055672740936279,
"epoch": 0.04426996324757768,
"grad_norm": 1.1875,
"learning_rate": 0.00026450000000000003,
"loss": 6.6226,
"mean_token_accuracy": 0.1023420087993145,
"num_tokens": 971240.0,
"step": 530
},
{
"entropy": 7.108932256698608,
"epoch": 0.044687604410290675,
"grad_norm": 1.0859375,
"learning_rate": 0.00026700000000000004,
"loss": 6.6471,
"mean_token_accuracy": 0.10342941582202911,
"num_tokens": 981391.0,
"step": 535
},
{
"entropy": 7.1047523498535154,
"epoch": 0.045105245573003674,
"grad_norm": 0.96875,
"learning_rate": 0.00026950000000000005,
"loss": 6.6554,
"mean_token_accuracy": 0.10619911253452301,
"num_tokens": 990551.0,
"step": 540
},
{
"entropy": 7.004219102859497,
"epoch": 0.045522886735716674,
"grad_norm": 1.15625,
"learning_rate": 0.00027200000000000005,
"loss": 6.5787,
"mean_token_accuracy": 0.11122526898980141,
"num_tokens": 999593.0,
"step": 545
},
{
"entropy": 7.028263998031616,
"epoch": 0.045940527898429666,
"grad_norm": 1.1796875,
"learning_rate": 0.0002745,
"loss": 6.6418,
"mean_token_accuracy": 0.10322482436895371,
"num_tokens": 1008785.0,
"step": 550
},
{
"entropy": 7.000995635986328,
"epoch": 0.046358169061142665,
"grad_norm": 1.1484375,
"learning_rate": 0.000277,
"loss": 6.632,
"mean_token_accuracy": 0.10688607320189476,
"num_tokens": 1016815.0,
"step": 555
},
{
"entropy": 7.063908100128174,
"epoch": 0.046775810223855664,
"grad_norm": 1.1171875,
"learning_rate": 0.0002795,
"loss": 6.5161,
"mean_token_accuracy": 0.11323921233415604,
"num_tokens": 1026182.0,
"step": 560
},
{
"entropy": 6.97571029663086,
"epoch": 0.04719345138656866,
"grad_norm": 1.2265625,
"learning_rate": 0.00028199999999999997,
"loss": 6.4512,
"mean_token_accuracy": 0.11479591503739357,
"num_tokens": 1034032.0,
"step": 565
},
{
"entropy": 6.954489898681641,
"epoch": 0.047611092549281656,
"grad_norm": 0.98828125,
"learning_rate": 0.0002845,
"loss": 6.6465,
"mean_token_accuracy": 0.10568908601999283,
"num_tokens": 1043493.0,
"step": 570
},
{
"entropy": 7.076480293273926,
"epoch": 0.048028733711994655,
"grad_norm": 1.2265625,
"learning_rate": 0.000287,
"loss": 6.5541,
"mean_token_accuracy": 0.1120546206831932,
"num_tokens": 1052989.0,
"step": 575
},
{
"entropy": 6.9995513439178465,
"epoch": 0.048446374874707654,
"grad_norm": 1.265625,
"learning_rate": 0.0002895,
"loss": 6.5191,
"mean_token_accuracy": 0.10802618637681008,
"num_tokens": 1061803.0,
"step": 580
},
{
"entropy": 6.9974480152130125,
"epoch": 0.048864016037420646,
"grad_norm": 1.0703125,
"learning_rate": 0.000292,
"loss": 6.6256,
"mean_token_accuracy": 0.10830648243427277,
"num_tokens": 1071227.0,
"step": 585
},
{
"entropy": 7.019133281707764,
"epoch": 0.049281657200133645,
"grad_norm": 1.15625,
"learning_rate": 0.0002945,
"loss": 6.6256,
"mean_token_accuracy": 0.1137481078505516,
"num_tokens": 1080082.0,
"step": 590
},
{
"entropy": 6.9460889339447025,
"epoch": 0.049699298362846644,
"grad_norm": 1.2109375,
"learning_rate": 0.000297,
"loss": 6.463,
"mean_token_accuracy": 0.11860100626945495,
"num_tokens": 1089750.0,
"step": 595
},
{
"entropy": 6.8880175113677975,
"epoch": 0.05011693952555964,
"grad_norm": 1.1953125,
"learning_rate": 0.0002995,
"loss": 6.4558,
"mean_token_accuracy": 0.11189346835017204,
"num_tokens": 1098691.0,
"step": 600
},
{
"entropy": 7.010653829574585,
"epoch": 0.050534580688272636,
"grad_norm": 1.1796875,
"learning_rate": 0.000302,
"loss": 6.4992,
"mean_token_accuracy": 0.11314608827233315,
"num_tokens": 1107602.0,
"step": 605
},
{
"entropy": 6.826686143875122,
"epoch": 0.050952221850985635,
"grad_norm": 1.2734375,
"learning_rate": 0.0003045,
"loss": 6.4167,
"mean_token_accuracy": 0.11482465639710426,
"num_tokens": 1116426.0,
"step": 610
},
{
"entropy": 6.978339052200317,
"epoch": 0.05136986301369863,
"grad_norm": 0.94140625,
"learning_rate": 0.000307,
"loss": 6.5929,
"mean_token_accuracy": 0.10989554226398468,
"num_tokens": 1126221.0,
"step": 615
},
{
"entropy": 6.9196573257446286,
"epoch": 0.051787504176411626,
"grad_norm": 1.2734375,
"learning_rate": 0.0003095,
"loss": 6.4681,
"mean_token_accuracy": 0.11944901272654533,
"num_tokens": 1135662.0,
"step": 620
},
{
"entropy": 6.880191802978516,
"epoch": 0.052205145339124626,
"grad_norm": 1.2421875,
"learning_rate": 0.000312,
"loss": 6.5525,
"mean_token_accuracy": 0.11337181329727172,
"num_tokens": 1143963.0,
"step": 625
},
{
"entropy": 6.883874082565308,
"epoch": 0.05262278650183762,
"grad_norm": 1.0546875,
"learning_rate": 0.0003145,
"loss": 6.4554,
"mean_token_accuracy": 0.1215918242931366,
"num_tokens": 1152470.0,
"step": 630
},
{
"entropy": 6.959899091720581,
"epoch": 0.05304042766455062,
"grad_norm": 1.109375,
"learning_rate": 0.000317,
"loss": 6.414,
"mean_token_accuracy": 0.11987600699067116,
"num_tokens": 1161257.0,
"step": 635
},
{
"entropy": 6.86087121963501,
"epoch": 0.053458068827263616,
"grad_norm": 0.99609375,
"learning_rate": 0.0003195,
"loss": 6.533,
"mean_token_accuracy": 0.10690693482756615,
"num_tokens": 1170969.0,
"step": 640
},
{
"entropy": 6.990806341171265,
"epoch": 0.053875709989976615,
"grad_norm": 1.1015625,
"learning_rate": 0.000322,
"loss": 6.6717,
"mean_token_accuracy": 0.11218328699469567,
"num_tokens": 1181127.0,
"step": 645
},
{
"entropy": 6.9073868751525875,
"epoch": 0.05429335115268961,
"grad_norm": 1.015625,
"learning_rate": 0.00032450000000000003,
"loss": 6.4613,
"mean_token_accuracy": 0.11650743409991264,
"num_tokens": 1190260.0,
"step": 650
},
{
"entropy": 6.870417165756225,
"epoch": 0.05471099231540261,
"grad_norm": 1.1328125,
"learning_rate": 0.00032700000000000003,
"loss": 6.5375,
"mean_token_accuracy": 0.11634354665875435,
"num_tokens": 1200013.0,
"step": 655
},
{
"entropy": 6.948531675338745,
"epoch": 0.055128633478115606,
"grad_norm": 1.375,
"learning_rate": 0.00032950000000000004,
"loss": 6.4582,
"mean_token_accuracy": 0.11997871845960617,
"num_tokens": 1209239.0,
"step": 660
},
{
"entropy": 6.796932411193848,
"epoch": 0.0555462746408286,
"grad_norm": 1.1328125,
"learning_rate": 0.00033200000000000005,
"loss": 6.4026,
"mean_token_accuracy": 0.11352865546941757,
"num_tokens": 1218552.0,
"step": 665
},
{
"entropy": 6.817508411407471,
"epoch": 0.0559639158035416,
"grad_norm": 1.109375,
"learning_rate": 0.00033450000000000005,
"loss": 6.4757,
"mean_token_accuracy": 0.11437263563275338,
"num_tokens": 1227632.0,
"step": 670
},
{
"entropy": 6.864944791793823,
"epoch": 0.056381556966254597,
"grad_norm": 1.1484375,
"learning_rate": 0.000337,
"loss": 6.5136,
"mean_token_accuracy": 0.11246693953871727,
"num_tokens": 1236433.0,
"step": 675
},
{
"entropy": 6.802402019500732,
"epoch": 0.05679919812896759,
"grad_norm": 1.2890625,
"learning_rate": 0.0003395,
"loss": 6.5371,
"mean_token_accuracy": 0.1055721789598465,
"num_tokens": 1245700.0,
"step": 680
},
{
"entropy": 6.921284770965576,
"epoch": 0.05721683929168059,
"grad_norm": 1.109375,
"learning_rate": 0.000342,
"loss": 6.3732,
"mean_token_accuracy": 0.11802853867411614,
"num_tokens": 1254206.0,
"step": 685
},
{
"entropy": 6.763827323913574,
"epoch": 0.05763448045439359,
"grad_norm": 1.0390625,
"learning_rate": 0.00034449999999999997,
"loss": 6.4184,
"mean_token_accuracy": 0.1180158130824566,
"num_tokens": 1263302.0,
"step": 690
},
{
"entropy": 6.76256685256958,
"epoch": 0.05805212161710658,
"grad_norm": 1.0078125,
"learning_rate": 0.000347,
"loss": 6.4661,
"mean_token_accuracy": 0.11739929467439651,
"num_tokens": 1272998.0,
"step": 695
},
{
"entropy": 6.8152045726776125,
"epoch": 0.05846976277981958,
"grad_norm": 1.0625,
"learning_rate": 0.0003495,
"loss": 6.4304,
"mean_token_accuracy": 0.1213896095752716,
"num_tokens": 1281955.0,
"step": 700
},
{
"entropy": 6.832838869094848,
"epoch": 0.05888740394253258,
"grad_norm": 1.0546875,
"learning_rate": 0.000352,
"loss": 6.4577,
"mean_token_accuracy": 0.11378672420978546,
"num_tokens": 1291219.0,
"step": 705
},
{
"entropy": 6.797576570510865,
"epoch": 0.05930504510524557,
"grad_norm": 1.0859375,
"learning_rate": 0.0003545,
"loss": 6.4718,
"mean_token_accuracy": 0.11139240860939026,
"num_tokens": 1300963.0,
"step": 710
},
{
"entropy": 6.8802446842193605,
"epoch": 0.05972268626795857,
"grad_norm": 1.109375,
"learning_rate": 0.000357,
"loss": 6.4965,
"mean_token_accuracy": 0.11046546846628189,
"num_tokens": 1310112.0,
"step": 715
},
{
"entropy": 6.775373268127441,
"epoch": 0.06014032743067157,
"grad_norm": 1.046875,
"learning_rate": 0.0003595,
"loss": 6.4278,
"mean_token_accuracy": 0.12265736609697342,
"num_tokens": 1319309.0,
"step": 720
},
{
"entropy": 6.66271562576294,
"epoch": 0.06055796859338457,
"grad_norm": 1.0546875,
"learning_rate": 0.000362,
"loss": 6.3938,
"mean_token_accuracy": 0.11732437312602997,
"num_tokens": 1329049.0,
"step": 725
},
{
"entropy": 6.728451251983643,
"epoch": 0.06097560975609756,
"grad_norm": 1.2578125,
"learning_rate": 0.0003645,
"loss": 6.2447,
"mean_token_accuracy": 0.12135154455900192,
"num_tokens": 1337770.0,
"step": 730
},
{
"entropy": 6.623231744766235,
"epoch": 0.06139325091881056,
"grad_norm": 1.171875,
"learning_rate": 0.000367,
"loss": 6.3446,
"mean_token_accuracy": 0.11998333036899567,
"num_tokens": 1346705.0,
"step": 735
},
{
"entropy": 6.744262742996216,
"epoch": 0.06181089208152356,
"grad_norm": 1.328125,
"learning_rate": 0.0003695,
"loss": 6.405,
"mean_token_accuracy": 0.11962749883532524,
"num_tokens": 1354860.0,
"step": 740
},
{
"entropy": 6.765829467773438,
"epoch": 0.06222853324423655,
"grad_norm": 1.1015625,
"learning_rate": 0.000372,
"loss": 6.3715,
"mean_token_accuracy": 0.11659613028168678,
"num_tokens": 1364140.0,
"step": 745
},
{
"entropy": 6.757311677932739,
"epoch": 0.06264617440694954,
"grad_norm": 1.046875,
"learning_rate": 0.0003745,
"loss": 6.5161,
"mean_token_accuracy": 0.10866649895906448,
"num_tokens": 1373645.0,
"step": 750
},
{
"entropy": 6.791751384735107,
"epoch": 0.06306381556966255,
"grad_norm": 1.0703125,
"learning_rate": 0.000377,
"loss": 6.3843,
"mean_token_accuracy": 0.12191172316670418,
"num_tokens": 1383291.0,
"step": 755
},
{
"entropy": 6.669790124893188,
"epoch": 0.06348145673237554,
"grad_norm": 1.21875,
"learning_rate": 0.0003795,
"loss": 6.3243,
"mean_token_accuracy": 0.12068940699100494,
"num_tokens": 1391172.0,
"step": 760
},
{
"entropy": 6.717700242996216,
"epoch": 0.06389909789508853,
"grad_norm": 1.09375,
"learning_rate": 0.000382,
"loss": 6.3198,
"mean_token_accuracy": 0.12138244062662125,
"num_tokens": 1400864.0,
"step": 765
},
{
"entropy": 6.721026468276977,
"epoch": 0.06431673905780154,
"grad_norm": 1.078125,
"learning_rate": 0.0003845,
"loss": 6.3758,
"mean_token_accuracy": 0.1206473708152771,
"num_tokens": 1410570.0,
"step": 770
},
{
"entropy": 6.632958030700683,
"epoch": 0.06473438022051453,
"grad_norm": 1.09375,
"learning_rate": 0.00038700000000000003,
"loss": 6.35,
"mean_token_accuracy": 0.12267208248376846,
"num_tokens": 1419536.0,
"step": 775
},
{
"entropy": 6.612038612365723,
"epoch": 0.06515202138322754,
"grad_norm": 1.109375,
"learning_rate": 0.00038950000000000003,
"loss": 6.2546,
"mean_token_accuracy": 0.1263652376830578,
"num_tokens": 1428000.0,
"step": 780
},
{
"entropy": 6.690039157867432,
"epoch": 0.06556966254594053,
"grad_norm": 1.1875,
"learning_rate": 0.00039200000000000004,
"loss": 6.4562,
"mean_token_accuracy": 0.12093869373202323,
"num_tokens": 1437789.0,
"step": 785
},
{
"entropy": 6.636522626876831,
"epoch": 0.06598730370865352,
"grad_norm": 1.1875,
"learning_rate": 0.00039450000000000005,
"loss": 6.2563,
"mean_token_accuracy": 0.12446803748607635,
"num_tokens": 1446848.0,
"step": 790
},
{
"entropy": 6.619644260406494,
"epoch": 0.06640494487136653,
"grad_norm": 1.125,
"learning_rate": 0.00039700000000000005,
"loss": 6.3158,
"mean_token_accuracy": 0.12070081159472465,
"num_tokens": 1455749.0,
"step": 795
},
{
"entropy": 6.622030401229859,
"epoch": 0.06682258603407952,
"grad_norm": 1.171875,
"learning_rate": 0.0003995,
"loss": 6.2456,
"mean_token_accuracy": 0.12712132856249808,
"num_tokens": 1464556.0,
"step": 800
},
{
"entropy": 6.697847175598144,
"epoch": 0.06724022719679251,
"grad_norm": 1.140625,
"learning_rate": 0.000402,
"loss": 6.3663,
"mean_token_accuracy": 0.12254888191819191,
"num_tokens": 1473617.0,
"step": 805
},
{
"entropy": 6.679299306869507,
"epoch": 0.06765786835950552,
"grad_norm": 1.0390625,
"learning_rate": 0.0004045,
"loss": 6.3852,
"mean_token_accuracy": 0.12291230112314225,
"num_tokens": 1483006.0,
"step": 810
},
{
"entropy": 6.704338502883911,
"epoch": 0.06807550952221851,
"grad_norm": 1.1953125,
"learning_rate": 0.00040699999999999997,
"loss": 6.4006,
"mean_token_accuracy": 0.12003520801663399,
"num_tokens": 1493009.0,
"step": 815
},
{
"entropy": 6.579654169082642,
"epoch": 0.0684931506849315,
"grad_norm": 1.2421875,
"learning_rate": 0.0004095,
"loss": 6.2912,
"mean_token_accuracy": 0.12424400523304939,
"num_tokens": 1501261.0,
"step": 820
},
{
"entropy": 6.61363673210144,
"epoch": 0.06891079184764451,
"grad_norm": 0.89453125,
"learning_rate": 0.000412,
"loss": 6.394,
"mean_token_accuracy": 0.11899233609437943,
"num_tokens": 1511623.0,
"step": 825
},
{
"entropy": 6.693784141540528,
"epoch": 0.0693284330103575,
"grad_norm": 1.046875,
"learning_rate": 0.0004145,
"loss": 6.3861,
"mean_token_accuracy": 0.11563008949160576,
"num_tokens": 1521741.0,
"step": 830
},
{
"entropy": 6.616790151596069,
"epoch": 0.0697460741730705,
"grad_norm": 0.98828125,
"learning_rate": 0.000417,
"loss": 6.3195,
"mean_token_accuracy": 0.12150524035096169,
"num_tokens": 1530988.0,
"step": 835
},
{
"entropy": 6.543398427963257,
"epoch": 0.0701637153357835,
"grad_norm": 1.1171875,
"learning_rate": 0.0004195,
"loss": 6.158,
"mean_token_accuracy": 0.12872020676732063,
"num_tokens": 1539415.0,
"step": 840
},
{
"entropy": 6.674872541427613,
"epoch": 0.07058135649849649,
"grad_norm": 1.171875,
"learning_rate": 0.000422,
"loss": 6.3627,
"mean_token_accuracy": 0.1207451082766056,
"num_tokens": 1547969.0,
"step": 845
},
{
"entropy": 6.544791841506958,
"epoch": 0.07099899766120948,
"grad_norm": 1.0859375,
"learning_rate": 0.0004245,
"loss": 6.2083,
"mean_token_accuracy": 0.12395204156637192,
"num_tokens": 1557526.0,
"step": 850
},
{
"entropy": 6.56576452255249,
"epoch": 0.07141663882392249,
"grad_norm": 1.1875,
"learning_rate": 0.000427,
"loss": 6.1961,
"mean_token_accuracy": 0.12846603170037268,
"num_tokens": 1565982.0,
"step": 855
},
{
"entropy": 6.504228305816651,
"epoch": 0.07183427998663548,
"grad_norm": 1.0859375,
"learning_rate": 0.0004295,
"loss": 6.2432,
"mean_token_accuracy": 0.13052212223410606,
"num_tokens": 1574492.0,
"step": 860
},
{
"entropy": 6.504011058807373,
"epoch": 0.07225192114934847,
"grad_norm": 1.2421875,
"learning_rate": 0.000432,
"loss": 6.2186,
"mean_token_accuracy": 0.1255444750189781,
"num_tokens": 1583726.0,
"step": 865
},
{
"entropy": 6.531704616546631,
"epoch": 0.07266956231206148,
"grad_norm": 1.234375,
"learning_rate": 0.0004345,
"loss": 6.2342,
"mean_token_accuracy": 0.11796416938304902,
"num_tokens": 1593780.0,
"step": 870
},
{
"entropy": 6.550727891921997,
"epoch": 0.07308720347477447,
"grad_norm": 1.1796875,
"learning_rate": 0.000437,
"loss": 6.2362,
"mean_token_accuracy": 0.1253882236778736,
"num_tokens": 1602746.0,
"step": 875
},
{
"entropy": 6.606376981735229,
"epoch": 0.07350484463748747,
"grad_norm": 1.0234375,
"learning_rate": 0.0004395,
"loss": 6.2635,
"mean_token_accuracy": 0.12688489481806756,
"num_tokens": 1611767.0,
"step": 880
},
{
"entropy": 6.605551767349243,
"epoch": 0.07392248580020047,
"grad_norm": 1.0,
"learning_rate": 0.000442,
"loss": 6.335,
"mean_token_accuracy": 0.12890502512454988,
"num_tokens": 1621568.0,
"step": 885
},
{
"entropy": 6.594180345535278,
"epoch": 0.07434012696291346,
"grad_norm": 1.015625,
"learning_rate": 0.0004445,
"loss": 6.2961,
"mean_token_accuracy": 0.13042845353484153,
"num_tokens": 1630650.0,
"step": 890
},
{
"entropy": 6.582787275314331,
"epoch": 0.07475776812562646,
"grad_norm": 1.15625,
"learning_rate": 0.000447,
"loss": 6.2881,
"mean_token_accuracy": 0.12370246201753617,
"num_tokens": 1640482.0,
"step": 895
},
{
"entropy": 6.651394557952881,
"epoch": 0.07517540928833946,
"grad_norm": 1.203125,
"learning_rate": 0.00044950000000000003,
"loss": 6.3247,
"mean_token_accuracy": 0.12517691105604173,
"num_tokens": 1649344.0,
"step": 900
},
{
"entropy": 6.495271348953247,
"epoch": 0.07559305045105245,
"grad_norm": 1.015625,
"learning_rate": 0.00045200000000000004,
"loss": 6.2618,
"mean_token_accuracy": 0.12706776037812234,
"num_tokens": 1658962.0,
"step": 905
},
{
"entropy": 6.504554367065429,
"epoch": 0.07601069161376545,
"grad_norm": 1.0078125,
"learning_rate": 0.00045450000000000004,
"loss": 6.266,
"mean_token_accuracy": 0.12863868474960327,
"num_tokens": 1668615.0,
"step": 910
},
{
"entropy": 6.574895524978638,
"epoch": 0.07642833277647845,
"grad_norm": 1.1015625,
"learning_rate": 0.00045700000000000005,
"loss": 6.2168,
"mean_token_accuracy": 0.12925311848521231,
"num_tokens": 1679282.0,
"step": 915
},
{
"entropy": 6.466578817367553,
"epoch": 0.07684597393919144,
"grad_norm": 0.99609375,
"learning_rate": 0.00045950000000000006,
"loss": 6.3065,
"mean_token_accuracy": 0.1251588962972164,
"num_tokens": 1688362.0,
"step": 920
},
{
"entropy": 6.553056955337524,
"epoch": 0.07726361510190444,
"grad_norm": 0.890625,
"learning_rate": 0.000462,
"loss": 6.1887,
"mean_token_accuracy": 0.12379047647118568,
"num_tokens": 1697593.0,
"step": 925
},
{
"entropy": 6.501152229309082,
"epoch": 0.07768125626461744,
"grad_norm": 1.125,
"learning_rate": 0.0004645,
"loss": 6.2696,
"mean_token_accuracy": 0.1266642227768898,
"num_tokens": 1706961.0,
"step": 930
},
{
"entropy": 6.591590833663941,
"epoch": 0.07809889742733044,
"grad_norm": 1.09375,
"learning_rate": 0.000467,
"loss": 6.3029,
"mean_token_accuracy": 0.12989042624831199,
"num_tokens": 1716222.0,
"step": 935
},
{
"entropy": 6.465153598785401,
"epoch": 0.07851653859004344,
"grad_norm": 1.109375,
"learning_rate": 0.0004695,
"loss": 6.1681,
"mean_token_accuracy": 0.12083110958337784,
"num_tokens": 1726750.0,
"step": 940
},
{
"entropy": 6.532292890548706,
"epoch": 0.07893417975275643,
"grad_norm": 1.0703125,
"learning_rate": 0.000472,
"loss": 6.281,
"mean_token_accuracy": 0.11997251808643342,
"num_tokens": 1736262.0,
"step": 945
},
{
"entropy": 6.490269708633423,
"epoch": 0.07935182091546943,
"grad_norm": 1.1796875,
"learning_rate": 0.0004745,
"loss": 6.2976,
"mean_token_accuracy": 0.12622648924589158,
"num_tokens": 1745555.0,
"step": 950
},
{
"entropy": 6.346992015838623,
"epoch": 0.07976946207818243,
"grad_norm": 1.3828125,
"learning_rate": 0.000477,
"loss": 6.1624,
"mean_token_accuracy": 0.12778715938329696,
"num_tokens": 1754107.0,
"step": 955
},
{
"entropy": 6.720982551574707,
"epoch": 0.08018710324089542,
"grad_norm": 1.015625,
"learning_rate": 0.0004795,
"loss": 6.3498,
"mean_token_accuracy": 0.12380973473191262,
"num_tokens": 1762982.0,
"step": 960
},
{
"entropy": 6.400809383392334,
"epoch": 0.08060474440360842,
"grad_norm": 1.15625,
"learning_rate": 0.000482,
"loss": 6.2299,
"mean_token_accuracy": 0.12820469886064528,
"num_tokens": 1772545.0,
"step": 965
},
{
"entropy": 6.63186354637146,
"epoch": 0.08102238556632142,
"grad_norm": 1.09375,
"learning_rate": 0.0004845,
"loss": 6.3614,
"mean_token_accuracy": 0.1199867382645607,
"num_tokens": 1782052.0,
"step": 970
},
{
"entropy": 6.447825527191162,
"epoch": 0.08144002672903441,
"grad_norm": 1.0703125,
"learning_rate": 0.000487,
"loss": 6.1631,
"mean_token_accuracy": 0.12513367608189582,
"num_tokens": 1791682.0,
"step": 975
},
{
"entropy": 6.466085052490234,
"epoch": 0.08185766789174741,
"grad_norm": 0.9609375,
"learning_rate": 0.0004895,
"loss": 6.1682,
"mean_token_accuracy": 0.1323828585445881,
"num_tokens": 1800526.0,
"step": 980
},
{
"entropy": 6.538319540023804,
"epoch": 0.08227530905446041,
"grad_norm": 1.078125,
"learning_rate": 0.000492,
"loss": 6.2922,
"mean_token_accuracy": 0.1273337259888649,
"num_tokens": 1810062.0,
"step": 985
},
{
"entropy": 6.485649347305298,
"epoch": 0.0826929502171734,
"grad_norm": 1.1484375,
"learning_rate": 0.0004945,
"loss": 6.2886,
"mean_token_accuracy": 0.12735243514180183,
"num_tokens": 1818909.0,
"step": 990
},
{
"entropy": 6.511818504333496,
"epoch": 0.0831105913798864,
"grad_norm": 1.1875,
"learning_rate": 0.000497,
"loss": 6.2268,
"mean_token_accuracy": 0.1313578575849533,
"num_tokens": 1827978.0,
"step": 995
},
{
"entropy": 6.385158729553223,
"epoch": 0.0835282325425994,
"grad_norm": 1.015625,
"learning_rate": 0.0004995,
"loss": 6.1962,
"mean_token_accuracy": 0.13243770301342012,
"num_tokens": 1837955.0,
"step": 1000
},
{
"entropy": 6.5512267589569095,
"epoch": 0.0839458737053124,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999999987395534,
"loss": 6.2897,
"mean_token_accuracy": 0.1274054653942585,
"num_tokens": 1847699.0,
"step": 1005
},
{
"entropy": 6.498107290267944,
"epoch": 0.08436351486802539,
"grad_norm": 1.078125,
"learning_rate": 0.0004999999936189888,
"loss": 6.2363,
"mean_token_accuracy": 0.12386318743228912,
"num_tokens": 1856490.0,
"step": 1010
},
{
"entropy": 6.442971467971802,
"epoch": 0.0847811560307384,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999999845595285,
"loss": 6.2339,
"mean_token_accuracy": 0.1264282323420048,
"num_tokens": 1865644.0,
"step": 1015
},
{
"entropy": 6.474472713470459,
"epoch": 0.08519879719345139,
"grad_norm": 1.234375,
"learning_rate": 0.0004999999715611727,
"loss": 6.1668,
"mean_token_accuracy": 0.1318289414048195,
"num_tokens": 1874450.0,
"step": 1020
},
{
"entropy": 6.359386444091797,
"epoch": 0.08561643835616438,
"grad_norm": 1.0625,
"learning_rate": 0.0004999999546239215,
"loss": 6.1867,
"mean_token_accuracy": 0.12795957028865815,
"num_tokens": 1883259.0,
"step": 1025
},
{
"entropy": 6.465229320526123,
"epoch": 0.08603407951887738,
"grad_norm": 1.171875,
"learning_rate": 0.0004999999337477753,
"loss": 6.2646,
"mean_token_accuracy": 0.12190999686717988,
"num_tokens": 1891897.0,
"step": 1030
},
{
"entropy": 6.531569147109986,
"epoch": 0.08645172068159038,
"grad_norm": 1.171875,
"learning_rate": 0.0004999999089327344,
"loss": 6.2369,
"mean_token_accuracy": 0.1253485880792141,
"num_tokens": 1901011.0,
"step": 1035
},
{
"entropy": 6.3279060363769535,
"epoch": 0.08686936184430337,
"grad_norm": 1.140625,
"learning_rate": 0.0004999998801787993,
"loss": 6.1475,
"mean_token_accuracy": 0.1359507881104946,
"num_tokens": 1910132.0,
"step": 1040
},
{
"entropy": 6.536592149734497,
"epoch": 0.08728700300701638,
"grad_norm": 1.125,
"learning_rate": 0.0004999998474859704,
"loss": 6.2593,
"mean_token_accuracy": 0.12965355440974236,
"num_tokens": 1919938.0,
"step": 1045
},
{
"entropy": 6.347357130050659,
"epoch": 0.08770464416972937,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999998108542483,
"loss": 6.1286,
"mean_token_accuracy": 0.1264238215982914,
"num_tokens": 1928199.0,
"step": 1050
},
{
"entropy": 6.376512479782105,
"epoch": 0.08812228533244236,
"grad_norm": 1.234375,
"learning_rate": 0.0004999997702836339,
"loss": 6.1175,
"mean_token_accuracy": 0.13022146746516228,
"num_tokens": 1937133.0,
"step": 1055
},
{
"entropy": 6.513174438476563,
"epoch": 0.08853992649515537,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999997257741274,
"loss": 6.2975,
"mean_token_accuracy": 0.1291216015815735,
"num_tokens": 1946580.0,
"step": 1060
},
{
"entropy": 6.365123176574707,
"epoch": 0.08895756765786836,
"grad_norm": 1.125,
"learning_rate": 0.00049999967732573,
"loss": 6.2097,
"mean_token_accuracy": 0.12826794609427453,
"num_tokens": 1955413.0,
"step": 1065
},
{
"entropy": 6.605544090270996,
"epoch": 0.08937520882058135,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999996249384425,
"loss": 6.2843,
"mean_token_accuracy": 0.1293049283325672,
"num_tokens": 1964716.0,
"step": 1070
},
{
"entropy": 6.295011186599732,
"epoch": 0.08979284998329436,
"grad_norm": 1.125,
"learning_rate": 0.0004999995686122657,
"loss": 6.143,
"mean_token_accuracy": 0.1339694581925869,
"num_tokens": 1974081.0,
"step": 1075
},
{
"entropy": 6.302339315414429,
"epoch": 0.09021049114600735,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999995083472006,
"loss": 6.12,
"mean_token_accuracy": 0.1303638242185116,
"num_tokens": 1983035.0,
"step": 1080
},
{
"entropy": 6.41491436958313,
"epoch": 0.09062813230872034,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999994441432481,
"loss": 6.1119,
"mean_token_accuracy": 0.14117980897426605,
"num_tokens": 1992475.0,
"step": 1085
},
{
"entropy": 6.36792368888855,
"epoch": 0.09104577347143335,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999993760004097,
"loss": 6.1813,
"mean_token_accuracy": 0.12746159359812737,
"num_tokens": 2002729.0,
"step": 1090
},
{
"entropy": 6.294451570510864,
"epoch": 0.09146341463414634,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999993039186864,
"loss": 6.0415,
"mean_token_accuracy": 0.13401369303464888,
"num_tokens": 2011319.0,
"step": 1095
},
{
"entropy": 6.397387838363647,
"epoch": 0.09188105579685933,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999992278980794,
"loss": 6.095,
"mean_token_accuracy": 0.1325186774134636,
"num_tokens": 2020456.0,
"step": 1100
},
{
"entropy": 6.282045221328735,
"epoch": 0.09229869695957234,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999991479385901,
"loss": 6.1684,
"mean_token_accuracy": 0.1352491483092308,
"num_tokens": 2029417.0,
"step": 1105
},
{
"entropy": 6.475615835189819,
"epoch": 0.09271633812228533,
"grad_norm": 1.109375,
"learning_rate": 0.0004999990640402198,
"loss": 6.2484,
"mean_token_accuracy": 0.13222716152668,
"num_tokens": 2039290.0,
"step": 1110
},
{
"entropy": 6.398699712753296,
"epoch": 0.09313397928499834,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999989762029701,
"loss": 6.0758,
"mean_token_accuracy": 0.13428714573383332,
"num_tokens": 2048485.0,
"step": 1115
},
{
"entropy": 6.212140274047852,
"epoch": 0.09355162044771133,
"grad_norm": 1.0625,
"learning_rate": 0.0004999988844268425,
"loss": 6.0653,
"mean_token_accuracy": 0.13647278472781182,
"num_tokens": 2057131.0,
"step": 1120
},
{
"entropy": 6.3156208992004395,
"epoch": 0.09396926161042432,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999987887118387,
"loss": 6.0425,
"mean_token_accuracy": 0.13471427783370019,
"num_tokens": 2066295.0,
"step": 1125
},
{
"entropy": 6.2854846000671385,
"epoch": 0.09438690277313733,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999986890579602,
"loss": 6.0529,
"mean_token_accuracy": 0.13778177052736282,
"num_tokens": 2075656.0,
"step": 1130
},
{
"entropy": 6.30756311416626,
"epoch": 0.09480454393585032,
"grad_norm": 0.984375,
"learning_rate": 0.0004999985854652087,
"loss": 6.1868,
"mean_token_accuracy": 0.12809674441814423,
"num_tokens": 2086046.0,
"step": 1135
},
{
"entropy": 6.448493432998657,
"epoch": 0.09522218509856331,
"grad_norm": 1.109375,
"learning_rate": 0.0004999984779335862,
"loss": 6.1931,
"mean_token_accuracy": 0.1333407126367092,
"num_tokens": 2094808.0,
"step": 1140
},
{
"entropy": 6.262913179397583,
"epoch": 0.09563982626127632,
"grad_norm": 1.015625,
"learning_rate": 0.0004999983664630945,
"loss": 6.13,
"mean_token_accuracy": 0.13531602770090104,
"num_tokens": 2104155.0,
"step": 1145
},
{
"entropy": 6.3961437225341795,
"epoch": 0.09605746742398931,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999982510537357,
"loss": 6.031,
"mean_token_accuracy": 0.13656094148755074,
"num_tokens": 2112517.0,
"step": 1150
},
{
"entropy": 6.132482528686523,
"epoch": 0.0964751085867023,
"grad_norm": 1.15625,
"learning_rate": 0.0004999981317055116,
"loss": 5.9808,
"mean_token_accuracy": 0.1375528834760189,
"num_tokens": 2121355.0,
"step": 1155
},
{
"entropy": 6.379729604721069,
"epoch": 0.09689274974941531,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999980084184244,
"loss": 6.1163,
"mean_token_accuracy": 0.1360280379652977,
"num_tokens": 2129278.0,
"step": 1160
},
{
"entropy": 6.3625894546508786,
"epoch": 0.0973103909121283,
"grad_norm": 1.0625,
"learning_rate": 0.0004999978811924762,
"loss": 6.2693,
"mean_token_accuracy": 0.12732002511620522,
"num_tokens": 2139548.0,
"step": 1165
},
{
"entropy": 6.331307220458984,
"epoch": 0.09772803207484129,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999977500276693,
"loss": 6.1653,
"mean_token_accuracy": 0.13200777843594552,
"num_tokens": 2148765.0,
"step": 1170
},
{
"entropy": 6.388595914840698,
"epoch": 0.0981456732375543,
"grad_norm": 1.171875,
"learning_rate": 0.0004999976149240058,
"loss": 6.0868,
"mean_token_accuracy": 0.14182495772838594,
"num_tokens": 2157482.0,
"step": 1175
},
{
"entropy": 6.323437643051148,
"epoch": 0.09856331440026729,
"grad_norm": 1.03125,
"learning_rate": 0.0004999974758814884,
"loss": 6.1771,
"mean_token_accuracy": 0.13111152350902558,
"num_tokens": 2167320.0,
"step": 1180
},
{
"entropy": 6.291384696960449,
"epoch": 0.09898095556298028,
"grad_norm": 1.109375,
"learning_rate": 0.0004999973329001193,
"loss": 6.1273,
"mean_token_accuracy": 0.13208657130599022,
"num_tokens": 2175454.0,
"step": 1185
},
{
"entropy": 6.33533353805542,
"epoch": 0.09939859672569329,
"grad_norm": 1.125,
"learning_rate": 0.0004999971859799009,
"loss": 6.0659,
"mean_token_accuracy": 0.13413866832852364,
"num_tokens": 2184391.0,
"step": 1190
},
{
"entropy": 6.168669271469116,
"epoch": 0.09981623788840628,
"grad_norm": 1.140625,
"learning_rate": 0.0004999970351208361,
"loss": 5.9572,
"mean_token_accuracy": 0.1385273478925228,
"num_tokens": 2193309.0,
"step": 1195
},
{
"entropy": 6.358164358139038,
"epoch": 0.10023387905111927,
"grad_norm": 1.078125,
"learning_rate": 0.0004999968803229275,
"loss": 6.101,
"mean_token_accuracy": 0.12906384244561195,
"num_tokens": 2202057.0,
"step": 1200
},
{
"entropy": 6.238880729675293,
"epoch": 0.10065152021383228,
"grad_norm": 1.109375,
"learning_rate": 0.0004999967215861775,
"loss": 6.1872,
"mean_token_accuracy": 0.12864621803164483,
"num_tokens": 2212384.0,
"step": 1205
},
{
"entropy": 6.409511709213257,
"epoch": 0.10106916137654527,
"grad_norm": 1.0546875,
"learning_rate": 0.000499996558910589,
"loss": 6.1065,
"mean_token_accuracy": 0.13120872378349305,
"num_tokens": 2221451.0,
"step": 1210
},
{
"entropy": 6.193742275238037,
"epoch": 0.10148680253925826,
"grad_norm": 1.15625,
"learning_rate": 0.000499996392296165,
"loss": 5.9994,
"mean_token_accuracy": 0.13083511367440223,
"num_tokens": 2230440.0,
"step": 1215
},
{
"entropy": 6.30219783782959,
"epoch": 0.10190444370197127,
"grad_norm": 1.09375,
"learning_rate": 0.0004999962217429083,
"loss": 6.1816,
"mean_token_accuracy": 0.12673772796988486,
"num_tokens": 2239985.0,
"step": 1220
},
{
"entropy": 6.273440933227539,
"epoch": 0.10232208486468426,
"grad_norm": 1.015625,
"learning_rate": 0.0004999960472508219,
"loss": 6.0561,
"mean_token_accuracy": 0.13513024523854256,
"num_tokens": 2250391.0,
"step": 1225
},
{
"entropy": 6.387974548339844,
"epoch": 0.10273972602739725,
"grad_norm": 1.03125,
"learning_rate": 0.000499995868819909,
"loss": 6.156,
"mean_token_accuracy": 0.13272497951984405,
"num_tokens": 2260620.0,
"step": 1230
},
{
"entropy": 6.299834966659546,
"epoch": 0.10315736719011026,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999956864501725,
"loss": 6.1095,
"mean_token_accuracy": 0.13380565866827965,
"num_tokens": 2271176.0,
"step": 1235
},
{
"entropy": 6.320106744766235,
"epoch": 0.10357500835282325,
"grad_norm": 1.140625,
"learning_rate": 0.0004999955001416156,
"loss": 6.0269,
"mean_token_accuracy": 0.14122135639190675,
"num_tokens": 2280371.0,
"step": 1240
},
{
"entropy": 6.157405948638916,
"epoch": 0.10399264951553625,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999953098942417,
"loss": 5.9988,
"mean_token_accuracy": 0.13865399062633516,
"num_tokens": 2289015.0,
"step": 1245
},
{
"entropy": 6.223317098617554,
"epoch": 0.10441029067824925,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999951157080541,
"loss": 6.0702,
"mean_token_accuracy": 0.13662980124354362,
"num_tokens": 2297409.0,
"step": 1250
},
{
"entropy": 6.225243949890137,
"epoch": 0.10482793184096224,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999949175830561,
"loss": 6.0985,
"mean_token_accuracy": 0.1335413746535778,
"num_tokens": 2307017.0,
"step": 1255
},
{
"entropy": 6.245915412902832,
"epoch": 0.10524557300367524,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999947155192514,
"loss": 5.9299,
"mean_token_accuracy": 0.14725383818149568,
"num_tokens": 2316014.0,
"step": 1260
},
{
"entropy": 6.294537830352783,
"epoch": 0.10566321416638824,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999945095166433,
"loss": 6.0469,
"mean_token_accuracy": 0.13486992120742797,
"num_tokens": 2326326.0,
"step": 1265
},
{
"entropy": 6.358298969268799,
"epoch": 0.10608085532910123,
"grad_norm": 1.1875,
"learning_rate": 0.0004999942995752354,
"loss": 6.1415,
"mean_token_accuracy": 0.13432925343513488,
"num_tokens": 2335108.0,
"step": 1270
},
{
"entropy": 6.230367851257324,
"epoch": 0.10649849649181423,
"grad_norm": 1.125,
"learning_rate": 0.0004999940856950315,
"loss": 6.0451,
"mean_token_accuracy": 0.14503730461001396,
"num_tokens": 2343971.0,
"step": 1275
},
{
"entropy": 6.33094162940979,
"epoch": 0.10691613765452723,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999938678760354,
"loss": 6.1053,
"mean_token_accuracy": 0.13394634500145913,
"num_tokens": 2353742.0,
"step": 1280
},
{
"entropy": 6.234221363067627,
"epoch": 0.10733377881724022,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999936461182507,
"loss": 6.0561,
"mean_token_accuracy": 0.13966468647122382,
"num_tokens": 2363629.0,
"step": 1285
},
{
"entropy": 6.106270885467529,
"epoch": 0.10775141997995323,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999934204216814,
"loss": 5.913,
"mean_token_accuracy": 0.14157691672444345,
"num_tokens": 2372705.0,
"step": 1290
},
{
"entropy": 6.213513946533203,
"epoch": 0.10816906114266622,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999931907863315,
"loss": 6.0353,
"mean_token_accuracy": 0.13727413564920427,
"num_tokens": 2382136.0,
"step": 1295
},
{
"entropy": 6.133555603027344,
"epoch": 0.10858670230537922,
"grad_norm": 1.1484375,
"learning_rate": 0.000499992957212205,
"loss": 6.0034,
"mean_token_accuracy": 0.13713872507214547,
"num_tokens": 2391060.0,
"step": 1300
},
{
"entropy": 6.358161687850952,
"epoch": 0.10900434346809222,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999927196993059,
"loss": 6.1334,
"mean_token_accuracy": 0.13744667619466783,
"num_tokens": 2399885.0,
"step": 1305
},
{
"entropy": 6.278725337982178,
"epoch": 0.10942198463080521,
"grad_norm": 0.96875,
"learning_rate": 0.0004999924782476384,
"loss": 6.0035,
"mean_token_accuracy": 0.14220317751169204,
"num_tokens": 2409654.0,
"step": 1310
},
{
"entropy": 6.199410915374756,
"epoch": 0.1098396257935182,
"grad_norm": 1.140625,
"learning_rate": 0.0004999922328572068,
"loss": 6.0624,
"mean_token_accuracy": 0.13762215226888658,
"num_tokens": 2419342.0,
"step": 1315
},
{
"entropy": 6.141462421417236,
"epoch": 0.11025726695623121,
"grad_norm": 1.0625,
"learning_rate": 0.0004999919835280154,
"loss": 5.9506,
"mean_token_accuracy": 0.1402168832719326,
"num_tokens": 2428845.0,
"step": 1320
},
{
"entropy": 6.255073547363281,
"epoch": 0.1106749081189442,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999917302600684,
"loss": 5.9557,
"mean_token_accuracy": 0.14843129366636276,
"num_tokens": 2437134.0,
"step": 1325
},
{
"entropy": 6.033188390731811,
"epoch": 0.1110925492816572,
"grad_norm": 1.015625,
"learning_rate": 0.0004999914730533704,
"loss": 5.7955,
"mean_token_accuracy": 0.1502596452832222,
"num_tokens": 2445438.0,
"step": 1330
},
{
"entropy": 6.2179426670074465,
"epoch": 0.1115101904443702,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999912119079258,
"loss": 6.1172,
"mean_token_accuracy": 0.13575662672519684,
"num_tokens": 2455039.0,
"step": 1335
},
{
"entropy": 6.263426494598389,
"epoch": 0.1119278316070832,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999909468237393,
"loss": 6.0483,
"mean_token_accuracy": 0.13733413219451904,
"num_tokens": 2465170.0,
"step": 1340
},
{
"entropy": 6.09090895652771,
"epoch": 0.11234547276979619,
"grad_norm": 0.984375,
"learning_rate": 0.0004999906778008153,
"loss": 5.8012,
"mean_token_accuracy": 0.151080884039402,
"num_tokens": 2473688.0,
"step": 1345
},
{
"entropy": 6.165483570098877,
"epoch": 0.11276311393250919,
"grad_norm": 1.078125,
"learning_rate": 0.0004999904048391588,
"loss": 6.0541,
"mean_token_accuracy": 0.14202144667506217,
"num_tokens": 2482335.0,
"step": 1350
},
{
"entropy": 6.1746495246887205,
"epoch": 0.11318075509522219,
"grad_norm": 1.125,
"learning_rate": 0.0004999901279387743,
"loss": 6.0017,
"mean_token_accuracy": 0.14338096156716346,
"num_tokens": 2491808.0,
"step": 1355
},
{
"entropy": 6.3466249942779545,
"epoch": 0.11359839625793518,
"grad_norm": 0.96875,
"learning_rate": 0.0004999898470996669,
"loss": 6.13,
"mean_token_accuracy": 0.13258725851774217,
"num_tokens": 2501422.0,
"step": 1360
},
{
"entropy": 6.159154224395752,
"epoch": 0.11401603742064818,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999895623218415,
"loss": 5.9181,
"mean_token_accuracy": 0.14280597865581512,
"num_tokens": 2511201.0,
"step": 1365
},
{
"entropy": 6.199628210067749,
"epoch": 0.11443367858336118,
"grad_norm": 1.078125,
"learning_rate": 0.0004999892736053027,
"loss": 5.9659,
"mean_token_accuracy": 0.13438230976462365,
"num_tokens": 2519798.0,
"step": 1370
},
{
"entropy": 6.113333225250244,
"epoch": 0.11485131974607417,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999889809500561,
"loss": 5.9528,
"mean_token_accuracy": 0.14371661543846131,
"num_tokens": 2529177.0,
"step": 1375
},
{
"entropy": 6.27652678489685,
"epoch": 0.11526896090878717,
"grad_norm": 1.109375,
"learning_rate": 0.0004999886843561065,
"loss": 5.9321,
"mean_token_accuracy": 0.14615674763917924,
"num_tokens": 2537571.0,
"step": 1380
},
{
"entropy": 6.031244230270386,
"epoch": 0.11568660207150017,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999883838234592,
"loss": 5.8548,
"mean_token_accuracy": 0.14628306329250335,
"num_tokens": 2546258.0,
"step": 1385
},
{
"entropy": 6.165388917922973,
"epoch": 0.11610424323421316,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999880793521194,
"loss": 5.8668,
"mean_token_accuracy": 0.13787785470485686,
"num_tokens": 2555862.0,
"step": 1390
},
{
"entropy": 5.933096837997437,
"epoch": 0.11652188439692616,
"grad_norm": 1.078125,
"learning_rate": 0.0004999877709420924,
"loss": 5.8877,
"mean_token_accuracy": 0.14672486782073973,
"num_tokens": 2564583.0,
"step": 1395
},
{
"entropy": 6.226252937316895,
"epoch": 0.11693952555963916,
"grad_norm": 1.0625,
"learning_rate": 0.0004999874585933837,
"loss": 5.984,
"mean_token_accuracy": 0.1394071787595749,
"num_tokens": 2574217.0,
"step": 1400
},
{
"entropy": 6.08908052444458,
"epoch": 0.11735716672235215,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999871423059987,
"loss": 5.8818,
"mean_token_accuracy": 0.14752811044454575,
"num_tokens": 2582821.0,
"step": 1405
},
{
"entropy": 6.154991292953492,
"epoch": 0.11777480788506516,
"grad_norm": 1.0,
"learning_rate": 0.000499986822079943,
"loss": 6.0467,
"mean_token_accuracy": 0.13607672825455666,
"num_tokens": 2592748.0,
"step": 1410
},
{
"entropy": 6.165617561340332,
"epoch": 0.11819244904777815,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999864979152221,
"loss": 5.9434,
"mean_token_accuracy": 0.14729078263044357,
"num_tokens": 2602053.0,
"step": 1415
},
{
"entropy": 6.102991533279419,
"epoch": 0.11861009021049114,
"grad_norm": 1.0625,
"learning_rate": 0.0004999861698118417,
"loss": 6.0165,
"mean_token_accuracy": 0.14120813012123107,
"num_tokens": 2610557.0,
"step": 1420
},
{
"entropy": 6.1263680934906,
"epoch": 0.11902773137320415,
"grad_norm": 1.015625,
"learning_rate": 0.0004999858377698076,
"loss": 5.9156,
"mean_token_accuracy": 0.13886989206075667,
"num_tokens": 2619628.0,
"step": 1425
},
{
"entropy": 6.326252508163452,
"epoch": 0.11944537253591714,
"grad_norm": 1.109375,
"learning_rate": 0.0004999855017891256,
"loss": 6.083,
"mean_token_accuracy": 0.1396564431488514,
"num_tokens": 2629284.0,
"step": 1430
},
{
"entropy": 6.109390592575073,
"epoch": 0.11986301369863013,
"grad_norm": 1.09375,
"learning_rate": 0.0004999851618698017,
"loss": 5.9736,
"mean_token_accuracy": 0.14895331710577012,
"num_tokens": 2638001.0,
"step": 1435
},
{
"entropy": 6.114994764328003,
"epoch": 0.12028065486134314,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999848180118416,
"loss": 5.9908,
"mean_token_accuracy": 0.14804767146706582,
"num_tokens": 2647071.0,
"step": 1440
},
{
"entropy": 6.187527418136597,
"epoch": 0.12069829602405613,
"grad_norm": 1.03125,
"learning_rate": 0.0004999844702152516,
"loss": 5.9583,
"mean_token_accuracy": 0.1474977269768715,
"num_tokens": 2655826.0,
"step": 1445
},
{
"entropy": 6.12548508644104,
"epoch": 0.12111593718676913,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999841184800374,
"loss": 5.8182,
"mean_token_accuracy": 0.15077890008687972,
"num_tokens": 2664431.0,
"step": 1450
},
{
"entropy": 5.976382827758789,
"epoch": 0.12153357834948213,
"grad_norm": 1.09375,
"learning_rate": 0.0004999837628062055,
"loss": 5.7456,
"mean_token_accuracy": 0.1526419997215271,
"num_tokens": 2673631.0,
"step": 1455
},
{
"entropy": 6.0932858943939205,
"epoch": 0.12195121951219512,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999834031937621,
"loss": 6.0214,
"mean_token_accuracy": 0.1415261410176754,
"num_tokens": 2683363.0,
"step": 1460
},
{
"entropy": 6.140729188919067,
"epoch": 0.12236886067490813,
"grad_norm": 0.984375,
"learning_rate": 0.0004999830396427133,
"loss": 5.9755,
"mean_token_accuracy": 0.13880073204636573,
"num_tokens": 2692533.0,
"step": 1465
},
{
"entropy": 6.100193119049072,
"epoch": 0.12278650183762112,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999826721530657,
"loss": 5.9625,
"mean_token_accuracy": 0.14861070960760117,
"num_tokens": 2701211.0,
"step": 1470
},
{
"entropy": 6.213804006576538,
"epoch": 0.12320414300033411,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999823007248255,
"loss": 5.9654,
"mean_token_accuracy": 0.14508800357580184,
"num_tokens": 2709543.0,
"step": 1475
},
{
"entropy": 6.132544040679932,
"epoch": 0.12362178416304712,
"grad_norm": 1.171875,
"learning_rate": 0.0004999819253579992,
"loss": 5.9665,
"mean_token_accuracy": 0.14124592989683152,
"num_tokens": 2718336.0,
"step": 1480
},
{
"entropy": 6.176914262771606,
"epoch": 0.12403942532576011,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999815460525938,
"loss": 5.9221,
"mean_token_accuracy": 0.14702423065900802,
"num_tokens": 2726504.0,
"step": 1485
},
{
"entropy": 6.065746259689331,
"epoch": 0.1244570664884731,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999811628086155,
"loss": 5.9256,
"mean_token_accuracy": 0.14405364468693732,
"num_tokens": 2736076.0,
"step": 1490
},
{
"entropy": 6.031908082962036,
"epoch": 0.1248747076511861,
"grad_norm": 1.0078125,
"learning_rate": 0.000499980775626071,
"loss": 5.8548,
"mean_token_accuracy": 0.14717004522681237,
"num_tokens": 2745497.0,
"step": 1495
},
{
"entropy": 6.086008548736572,
"epoch": 0.12529234881389908,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999803845049674,
"loss": 6.0051,
"mean_token_accuracy": 0.14198841154575348,
"num_tokens": 2754455.0,
"step": 1500
},
{
"entropy": 6.275413751602173,
"epoch": 0.1257099899766121,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999799894453112,
"loss": 6.0214,
"mean_token_accuracy": 0.14551321789622307,
"num_tokens": 2763603.0,
"step": 1505
},
{
"entropy": 6.12061562538147,
"epoch": 0.1261276311393251,
"grad_norm": 1.140625,
"learning_rate": 0.0004999795904471095,
"loss": 5.8695,
"mean_token_accuracy": 0.14632337167859077,
"num_tokens": 2772152.0,
"step": 1510
},
{
"entropy": 6.125328063964844,
"epoch": 0.12654527230203808,
"grad_norm": 1.03125,
"learning_rate": 0.0004999791875103694,
"loss": 5.9827,
"mean_token_accuracy": 0.14955568239092826,
"num_tokens": 2781459.0,
"step": 1515
},
{
"entropy": 6.160517978668213,
"epoch": 0.12696291346475108,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999787806350977,
"loss": 5.9352,
"mean_token_accuracy": 0.1438853219151497,
"num_tokens": 2791177.0,
"step": 1520
},
{
"entropy": 6.0033468246459964,
"epoch": 0.1273805546274641,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999783698213016,
"loss": 5.926,
"mean_token_accuracy": 0.15096332877874374,
"num_tokens": 2800543.0,
"step": 1525
},
{
"entropy": 6.034644699096679,
"epoch": 0.12779819579017707,
"grad_norm": 1.09375,
"learning_rate": 0.0004999779550689885,
"loss": 5.764,
"mean_token_accuracy": 0.15348498970270158,
"num_tokens": 2809301.0,
"step": 1530
},
{
"entropy": 6.169605159759522,
"epoch": 0.12821583695289007,
"grad_norm": 1.109375,
"learning_rate": 0.0004999775363781654,
"loss": 5.9802,
"mean_token_accuracy": 0.1403758257627487,
"num_tokens": 2818269.0,
"step": 1535
},
{
"entropy": 6.0626365661621096,
"epoch": 0.12863347811560308,
"grad_norm": 1.09375,
"learning_rate": 0.0004999771137488395,
"loss": 5.8292,
"mean_token_accuracy": 0.14809781610965728,
"num_tokens": 2827170.0,
"step": 1540
},
{
"entropy": 6.0085289001464846,
"epoch": 0.12905111927831608,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999766871810188,
"loss": 5.8354,
"mean_token_accuracy": 0.15448207408189774,
"num_tokens": 2835320.0,
"step": 1545
},
{
"entropy": 6.026972770690918,
"epoch": 0.12946876044102906,
"grad_norm": 1.125,
"learning_rate": 0.0004999762566747101,
"loss": 5.9073,
"mean_token_accuracy": 0.14486387148499488,
"num_tokens": 2843755.0,
"step": 1550
},
{
"entropy": 6.129249715805054,
"epoch": 0.12988640160374207,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999758222299214,
"loss": 5.9413,
"mean_token_accuracy": 0.14105947464704513,
"num_tokens": 2853462.0,
"step": 1555
},
{
"entropy": 6.067689990997314,
"epoch": 0.13030404276645507,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999753838466601,
"loss": 5.8489,
"mean_token_accuracy": 0.14596890062093734,
"num_tokens": 2863041.0,
"step": 1560
},
{
"entropy": 6.222100067138672,
"epoch": 0.13072168392916805,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999749415249338,
"loss": 6.0523,
"mean_token_accuracy": 0.14630843549966813,
"num_tokens": 2873019.0,
"step": 1565
},
{
"entropy": 6.015613985061646,
"epoch": 0.13113932509188106,
"grad_norm": 1.046875,
"learning_rate": 0.0004999744952647505,
"loss": 5.8065,
"mean_token_accuracy": 0.1499025270342827,
"num_tokens": 2881467.0,
"step": 1570
},
{
"entropy": 6.0279559135437015,
"epoch": 0.13155696625459407,
"grad_norm": 1.046875,
"learning_rate": 0.0004999740450661176,
"loss": 5.8984,
"mean_token_accuracy": 0.14618548154830932,
"num_tokens": 2890914.0,
"step": 1575
},
{
"entropy": 6.068280839920044,
"epoch": 0.13197460741730704,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999735909290436,
"loss": 5.8159,
"mean_token_accuracy": 0.15112878903746604,
"num_tokens": 2901164.0,
"step": 1580
},
{
"entropy": 6.036407995223999,
"epoch": 0.13239224858002005,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999731328535358,
"loss": 5.9705,
"mean_token_accuracy": 0.14157942086458206,
"num_tokens": 2910560.0,
"step": 1585
},
{
"entropy": 6.0331549644470215,
"epoch": 0.13280988974273306,
"grad_norm": 1.140625,
"learning_rate": 0.0004999726708396026,
"loss": 5.7769,
"mean_token_accuracy": 0.1490756705403328,
"num_tokens": 2918980.0,
"step": 1590
},
{
"entropy": 6.02565450668335,
"epoch": 0.13322753090544603,
"grad_norm": 0.9453125,
"learning_rate": 0.000499972204887252,
"loss": 5.9787,
"mean_token_accuracy": 0.1396894186735153,
"num_tokens": 2929516.0,
"step": 1595
},
{
"entropy": 6.176361989974976,
"epoch": 0.13364517206815904,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999717349964922,
"loss": 6.0518,
"mean_token_accuracy": 0.1366614058613777,
"num_tokens": 2940603.0,
"step": 1600
},
{
"entropy": 6.116162443161011,
"epoch": 0.13406281323087205,
"grad_norm": 1.0,
"learning_rate": 0.0004999712611673313,
"loss": 5.8945,
"mean_token_accuracy": 0.14497055262327194,
"num_tokens": 2950082.0,
"step": 1605
},
{
"entropy": 5.9512824535369875,
"epoch": 0.13448045439358502,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999707833997777,
"loss": 5.7648,
"mean_token_accuracy": 0.15802783519029617,
"num_tokens": 2959098.0,
"step": 1610
},
{
"entropy": 6.146846580505371,
"epoch": 0.13489809555629803,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999703016938398,
"loss": 5.9007,
"mean_token_accuracy": 0.14953137785196305,
"num_tokens": 2968585.0,
"step": 1615
},
{
"entropy": 6.060613203048706,
"epoch": 0.13531573671901104,
"grad_norm": 0.9453125,
"learning_rate": 0.000499969816049526,
"loss": 5.9204,
"mean_token_accuracy": 0.14645206406712533,
"num_tokens": 2978430.0,
"step": 1620
},
{
"entropy": 6.181136131286621,
"epoch": 0.13573337788172402,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999693264668446,
"loss": 5.9891,
"mean_token_accuracy": 0.1384026862680912,
"num_tokens": 2988398.0,
"step": 1625
},
{
"entropy": 6.1024940490722654,
"epoch": 0.13615101904443702,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999688329458045,
"loss": 6.0009,
"mean_token_accuracy": 0.14288765341043472,
"num_tokens": 2999255.0,
"step": 1630
},
{
"entropy": 6.158026123046875,
"epoch": 0.13656866020715003,
"grad_norm": 1.046875,
"learning_rate": 0.0004999683354864141,
"loss": 5.8668,
"mean_token_accuracy": 0.15016328170895576,
"num_tokens": 3008638.0,
"step": 1635
},
{
"entropy": 6.011956262588501,
"epoch": 0.136986301369863,
"grad_norm": 1.046875,
"learning_rate": 0.0004999678340886823,
"loss": 5.8952,
"mean_token_accuracy": 0.14607993140816689,
"num_tokens": 3018070.0,
"step": 1640
},
{
"entropy": 5.984478425979614,
"epoch": 0.137403942532576,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999673287526176,
"loss": 5.862,
"mean_token_accuracy": 0.1520274721086025,
"num_tokens": 3027878.0,
"step": 1645
},
{
"entropy": 6.163523769378662,
"epoch": 0.13782158369528902,
"grad_norm": 1.09375,
"learning_rate": 0.0004999668194782291,
"loss": 6.0371,
"mean_token_accuracy": 0.14080698117613794,
"num_tokens": 3037031.0,
"step": 1650
},
{
"entropy": 6.169953918457031,
"epoch": 0.138239224858002,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999663062655257,
"loss": 5.9905,
"mean_token_accuracy": 0.136997552216053,
"num_tokens": 3046311.0,
"step": 1655
},
{
"entropy": 6.062465858459473,
"epoch": 0.138656866020715,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999657891145162,
"loss": 5.8028,
"mean_token_accuracy": 0.14837437868118286,
"num_tokens": 3055108.0,
"step": 1660
},
{
"entropy": 5.975989055633545,
"epoch": 0.139074507183428,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999652680252099,
"loss": 5.829,
"mean_token_accuracy": 0.15185758844017982,
"num_tokens": 3063492.0,
"step": 1665
},
{
"entropy": 6.1554381370544435,
"epoch": 0.139492148346141,
"grad_norm": 1.015625,
"learning_rate": 0.0004999647429976157,
"loss": 5.8732,
"mean_token_accuracy": 0.152986741065979,
"num_tokens": 3072572.0,
"step": 1670
},
{
"entropy": 6.097223615646362,
"epoch": 0.139909789508854,
"grad_norm": 0.953125,
"learning_rate": 0.0004999642140317429,
"loss": 5.9321,
"mean_token_accuracy": 0.14403018429875375,
"num_tokens": 3082942.0,
"step": 1675
},
{
"entropy": 6.034411334991455,
"epoch": 0.140327430671567,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999636811276007,
"loss": 5.8955,
"mean_token_accuracy": 0.15122605189681054,
"num_tokens": 3091410.0,
"step": 1680
},
{
"entropy": 6.105877351760864,
"epoch": 0.14074507183427998,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999631442851986,
"loss": 5.9586,
"mean_token_accuracy": 0.15086555927991868,
"num_tokens": 3100626.0,
"step": 1685
},
{
"entropy": 6.044166851043701,
"epoch": 0.14116271299699298,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999626035045458,
"loss": 5.7604,
"mean_token_accuracy": 0.15637173503637314,
"num_tokens": 3109339.0,
"step": 1690
},
{
"entropy": 6.087968540191651,
"epoch": 0.141580354159706,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999620587856519,
"loss": 5.879,
"mean_token_accuracy": 0.1442359484732151,
"num_tokens": 3118636.0,
"step": 1695
},
{
"entropy": 5.975955629348755,
"epoch": 0.14199799532241897,
"grad_norm": 1.0,
"learning_rate": 0.0004999615101285262,
"loss": 5.8571,
"mean_token_accuracy": 0.14512152522802352,
"num_tokens": 3128162.0,
"step": 1700
},
{
"entropy": 6.065919256210327,
"epoch": 0.14241563648513197,
"grad_norm": 1.046875,
"learning_rate": 0.0004999609575331786,
"loss": 5.8685,
"mean_token_accuracy": 0.15015122443437576,
"num_tokens": 3136451.0,
"step": 1705
},
{
"entropy": 6.057447052001953,
"epoch": 0.14283327764784498,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999604009996186,
"loss": 5.8614,
"mean_token_accuracy": 0.14777368754148484,
"num_tokens": 3146599.0,
"step": 1710
},
{
"entropy": 6.099114274978637,
"epoch": 0.14325091881055796,
"grad_norm": 1.015625,
"learning_rate": 0.000499959840527856,
"loss": 5.916,
"mean_token_accuracy": 0.14324069395661354,
"num_tokens": 3155941.0,
"step": 1715
},
{
"entropy": 6.0154510021209715,
"epoch": 0.14366855997327097,
"grad_norm": 1.140625,
"learning_rate": 0.0004999592761179007,
"loss": 5.8388,
"mean_token_accuracy": 0.15261219143867494,
"num_tokens": 3164641.0,
"step": 1720
},
{
"entropy": 6.028367567062378,
"epoch": 0.14408620113598397,
"grad_norm": 1.03125,
"learning_rate": 0.0004999587077697624,
"loss": 5.872,
"mean_token_accuracy": 0.15229901447892188,
"num_tokens": 3173701.0,
"step": 1725
},
{
"entropy": 5.903607177734375,
"epoch": 0.14450384229869695,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999581354834512,
"loss": 5.7215,
"mean_token_accuracy": 0.1503341905772686,
"num_tokens": 3182268.0,
"step": 1730
},
{
"entropy": 5.921892786026001,
"epoch": 0.14492148346140996,
"grad_norm": 1.0703125,
"learning_rate": 0.000499957559258977,
"loss": 5.8109,
"mean_token_accuracy": 0.153120157122612,
"num_tokens": 3191513.0,
"step": 1735
},
{
"entropy": 6.026451969146729,
"epoch": 0.14533912462412296,
"grad_norm": 1.0546875,
"learning_rate": 0.00049995697909635,
"loss": 5.7633,
"mean_token_accuracy": 0.15072803422808648,
"num_tokens": 3200580.0,
"step": 1740
},
{
"entropy": 5.919099378585815,
"epoch": 0.14575676578683594,
"grad_norm": 1.078125,
"learning_rate": 0.0004999563949955803,
"loss": 5.8504,
"mean_token_accuracy": 0.15168591886758803,
"num_tokens": 3209053.0,
"step": 1745
},
{
"entropy": 6.004643774032592,
"epoch": 0.14617440694954895,
"grad_norm": 1.0,
"learning_rate": 0.000499955806956678,
"loss": 5.8276,
"mean_token_accuracy": 0.15193051397800444,
"num_tokens": 3218241.0,
"step": 1750
},
{
"entropy": 6.114099073410034,
"epoch": 0.14659204811226195,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999552149796536,
"loss": 5.8861,
"mean_token_accuracy": 0.14796419218182563,
"num_tokens": 3227887.0,
"step": 1755
},
{
"entropy": 5.913031387329101,
"epoch": 0.14700968927497493,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999546190645175,
"loss": 5.7115,
"mean_token_accuracy": 0.15936694741249086,
"num_tokens": 3237281.0,
"step": 1760
},
{
"entropy": 5.936592626571655,
"epoch": 0.14742733043768794,
"grad_norm": 1.015625,
"learning_rate": 0.0004999540192112799,
"loss": 5.7418,
"mean_token_accuracy": 0.1534080371260643,
"num_tokens": 3246907.0,
"step": 1765
},
{
"entropy": 5.903944110870361,
"epoch": 0.14784497160040094,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999534154199514,
"loss": 5.7581,
"mean_token_accuracy": 0.15006715506315232,
"num_tokens": 3256312.0,
"step": 1770
},
{
"entropy": 6.11810793876648,
"epoch": 0.14826261276311392,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999528076905426,
"loss": 5.8535,
"mean_token_accuracy": 0.15088604092597963,
"num_tokens": 3265554.0,
"step": 1775
},
{
"entropy": 5.8378783702850345,
"epoch": 0.14868025392582693,
"grad_norm": 1.078125,
"learning_rate": 0.0004999521960230641,
"loss": 5.6918,
"mean_token_accuracy": 0.1556992784142494,
"num_tokens": 3274731.0,
"step": 1780
},
{
"entropy": 5.91249713897705,
"epoch": 0.14909789508853993,
"grad_norm": 1.078125,
"learning_rate": 0.0004999515804175268,
"loss": 5.8334,
"mean_token_accuracy": 0.14639962017536162,
"num_tokens": 3283518.0,
"step": 1785
},
{
"entropy": 6.047197675704956,
"epoch": 0.1495155362512529,
"grad_norm": 0.93359375,
"learning_rate": 0.000499950960873941,
"loss": 5.8387,
"mean_token_accuracy": 0.15209829434752464,
"num_tokens": 3292244.0,
"step": 1790
},
{
"entropy": 5.9918311595916744,
"epoch": 0.14993317741396592,
"grad_norm": 1.125,
"learning_rate": 0.0004999503373923182,
"loss": 5.8133,
"mean_token_accuracy": 0.1558985084295273,
"num_tokens": 3301198.0,
"step": 1795
},
{
"entropy": 5.999289274215698,
"epoch": 0.15035081857667892,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999497099726687,
"loss": 5.7948,
"mean_token_accuracy": 0.15358459949493408,
"num_tokens": 3309983.0,
"step": 1800
},
{
"entropy": 5.985867929458618,
"epoch": 0.1507684597393919,
"grad_norm": 0.921875,
"learning_rate": 0.0004999490786150039,
"loss": 5.8671,
"mean_token_accuracy": 0.152096726000309,
"num_tokens": 3319313.0,
"step": 1805
},
{
"entropy": 5.97202959060669,
"epoch": 0.1511861009021049,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999484433193348,
"loss": 5.7223,
"mean_token_accuracy": 0.1539360076189041,
"num_tokens": 3327497.0,
"step": 1810
},
{
"entropy": 5.8741857528686525,
"epoch": 0.15160374206481791,
"grad_norm": 1.015625,
"learning_rate": 0.0004999478040856722,
"loss": 5.7321,
"mean_token_accuracy": 0.15889767706394195,
"num_tokens": 3337029.0,
"step": 1815
},
{
"entropy": 5.907070827484131,
"epoch": 0.1520213832275309,
"grad_norm": 1.109375,
"learning_rate": 0.0004999471609140276,
"loss": 5.7154,
"mean_token_accuracy": 0.15107612311840057,
"num_tokens": 3346243.0,
"step": 1820
},
{
"entropy": 6.052509498596192,
"epoch": 0.1524390243902439,
"grad_norm": 1.125,
"learning_rate": 0.0004999465138044122,
"loss": 5.8601,
"mean_token_accuracy": 0.14963715374469758,
"num_tokens": 3354846.0,
"step": 1825
},
{
"entropy": 5.945438528060913,
"epoch": 0.1528566655529569,
"grad_norm": 1.0,
"learning_rate": 0.0004999458627568373,
"loss": 5.8311,
"mean_token_accuracy": 0.14300252199172975,
"num_tokens": 3364683.0,
"step": 1830
},
{
"entropy": 5.997858572006225,
"epoch": 0.15327430671566988,
"grad_norm": 0.9375,
"learning_rate": 0.0004999452077713144,
"loss": 5.7828,
"mean_token_accuracy": 0.1520732045173645,
"num_tokens": 3374652.0,
"step": 1835
},
{
"entropy": 5.954912424087524,
"epoch": 0.1536919478783829,
"grad_norm": 1.03125,
"learning_rate": 0.0004999445488478547,
"loss": 5.7793,
"mean_token_accuracy": 0.1517012432217598,
"num_tokens": 3384040.0,
"step": 1840
},
{
"entropy": 5.916548538208008,
"epoch": 0.1541095890410959,
"grad_norm": 1.1640625,
"learning_rate": 0.00049994388598647,
"loss": 5.8458,
"mean_token_accuracy": 0.15544982850551606,
"num_tokens": 3393185.0,
"step": 1845
},
{
"entropy": 5.950350522994995,
"epoch": 0.15452723020380887,
"grad_norm": 1.0,
"learning_rate": 0.0004999432191871719,
"loss": 5.7796,
"mean_token_accuracy": 0.15568242147564887,
"num_tokens": 3403077.0,
"step": 1850
},
{
"entropy": 5.9900336265563965,
"epoch": 0.15494487136652188,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999425484499718,
"loss": 5.7456,
"mean_token_accuracy": 0.16083690375089646,
"num_tokens": 3412165.0,
"step": 1855
},
{
"entropy": 5.951553058624268,
"epoch": 0.1553625125292349,
"grad_norm": 1.046875,
"learning_rate": 0.0004999418737748817,
"loss": 5.8265,
"mean_token_accuracy": 0.1454428181052208,
"num_tokens": 3421162.0,
"step": 1860
},
{
"entropy": 6.050862407684326,
"epoch": 0.15578015369194786,
"grad_norm": 1.0625,
"learning_rate": 0.0004999411951619134,
"loss": 5.8793,
"mean_token_accuracy": 0.14493936151266099,
"num_tokens": 3430363.0,
"step": 1865
},
{
"entropy": 5.987689018249512,
"epoch": 0.15619779485466087,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999405126110786,
"loss": 5.7929,
"mean_token_accuracy": 0.15352775007486344,
"num_tokens": 3439199.0,
"step": 1870
},
{
"entropy": 6.1002357006073,
"epoch": 0.15661543601737388,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999398261223895,
"loss": 5.9129,
"mean_token_accuracy": 0.14442275017499923,
"num_tokens": 3448720.0,
"step": 1875
},
{
"entropy": 5.99772834777832,
"epoch": 0.15703307718008688,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999391356958579,
"loss": 5.8939,
"mean_token_accuracy": 0.14629118889570236,
"num_tokens": 3458472.0,
"step": 1880
},
{
"entropy": 5.93433518409729,
"epoch": 0.15745071834279986,
"grad_norm": 1.078125,
"learning_rate": 0.000499938441331496,
"loss": 5.8354,
"mean_token_accuracy": 0.1485055461525917,
"num_tokens": 3467140.0,
"step": 1885
},
{
"entropy": 6.078635835647583,
"epoch": 0.15786835950551287,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999377430293159,
"loss": 5.8076,
"mean_token_accuracy": 0.15474040806293488,
"num_tokens": 3476431.0,
"step": 1890
},
{
"entropy": 5.902494716644287,
"epoch": 0.15828600066822587,
"grad_norm": 1.09375,
"learning_rate": 0.0004999370407893299,
"loss": 5.7226,
"mean_token_accuracy": 0.15522852241992952,
"num_tokens": 3485768.0,
"step": 1895
},
{
"entropy": 6.016763544082641,
"epoch": 0.15870364183093885,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999363346115501,
"loss": 5.8796,
"mean_token_accuracy": 0.14450376033782958,
"num_tokens": 3495678.0,
"step": 1900
},
{
"entropy": 6.002452945709228,
"epoch": 0.15912128299365186,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999356244959893,
"loss": 5.8607,
"mean_token_accuracy": 0.14681571200489998,
"num_tokens": 3505168.0,
"step": 1905
},
{
"entropy": 5.89468150138855,
"epoch": 0.15953892415636486,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999349104426594,
"loss": 5.8125,
"mean_token_accuracy": 0.14540216326713562,
"num_tokens": 3514498.0,
"step": 1910
},
{
"entropy": 6.03236985206604,
"epoch": 0.15995656531907784,
"grad_norm": 1.078125,
"learning_rate": 0.0004999341924515732,
"loss": 5.7402,
"mean_token_accuracy": 0.14893507361412048,
"num_tokens": 3524617.0,
"step": 1915
},
{
"entropy": 5.950137090682984,
"epoch": 0.16037420648179085,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999334705227431,
"loss": 5.7038,
"mean_token_accuracy": 0.15793468281626702,
"num_tokens": 3533523.0,
"step": 1920
},
{
"entropy": 5.93976845741272,
"epoch": 0.16079184764450385,
"grad_norm": 0.92578125,
"learning_rate": 0.000499932744656182,
"loss": 5.7803,
"mean_token_accuracy": 0.15535735040903093,
"num_tokens": 3542922.0,
"step": 1925
},
{
"entropy": 5.9716380596160885,
"epoch": 0.16120948880721683,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999320148519024,
"loss": 5.7698,
"mean_token_accuracy": 0.1537807211279869,
"num_tokens": 3551827.0,
"step": 1930
},
{
"entropy": 5.912134122848511,
"epoch": 0.16162712996992984,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999312811099171,
"loss": 5.7424,
"mean_token_accuracy": 0.15665630251169205,
"num_tokens": 3560911.0,
"step": 1935
},
{
"entropy": 5.943596744537354,
"epoch": 0.16204477113264285,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999305434302391,
"loss": 5.774,
"mean_token_accuracy": 0.15078182965517045,
"num_tokens": 3569473.0,
"step": 1940
},
{
"entropy": 5.950336027145386,
"epoch": 0.16246241229535582,
"grad_norm": 1.0546875,
"learning_rate": 0.000499929801812881,
"loss": 5.7669,
"mean_token_accuracy": 0.15755152851343154,
"num_tokens": 3578843.0,
"step": 1945
},
{
"entropy": 5.934217214584351,
"epoch": 0.16288005345806883,
"grad_norm": 1.078125,
"learning_rate": 0.0004999290562578561,
"loss": 5.7788,
"mean_token_accuracy": 0.15204366892576218,
"num_tokens": 3588304.0,
"step": 1950
},
{
"entropy": 5.915329647064209,
"epoch": 0.16329769462078184,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999283067651773,
"loss": 5.8426,
"mean_token_accuracy": 0.15053938925266266,
"num_tokens": 3598354.0,
"step": 1955
},
{
"entropy": 6.05399227142334,
"epoch": 0.16371533578349481,
"grad_norm": 0.984375,
"learning_rate": 0.0004999275533348577,
"loss": 5.8723,
"mean_token_accuracy": 0.15038531348109246,
"num_tokens": 3607503.0,
"step": 1960
},
{
"entropy": 5.811189603805542,
"epoch": 0.16413297694620782,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999267959669106,
"loss": 5.6896,
"mean_token_accuracy": 0.1612261191010475,
"num_tokens": 3617467.0,
"step": 1965
},
{
"entropy": 5.984873485565186,
"epoch": 0.16455061810892083,
"grad_norm": 1.078125,
"learning_rate": 0.0004999260346613491,
"loss": 5.7094,
"mean_token_accuracy": 0.14969066977500917,
"num_tokens": 3627048.0,
"step": 1970
},
{
"entropy": 5.889481449127198,
"epoch": 0.1649682592716338,
"grad_norm": 1.015625,
"learning_rate": 0.0004999252694181867,
"loss": 5.7374,
"mean_token_accuracy": 0.15431074053049088,
"num_tokens": 3636683.0,
"step": 1975
},
{
"entropy": 5.81419415473938,
"epoch": 0.1653859004343468,
"grad_norm": 1.03125,
"learning_rate": 0.0004999245002374367,
"loss": 5.6322,
"mean_token_accuracy": 0.16551566421985625,
"num_tokens": 3645231.0,
"step": 1980
},
{
"entropy": 5.923137855529785,
"epoch": 0.16580354159705982,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999237271191125,
"loss": 5.7184,
"mean_token_accuracy": 0.1556422308087349,
"num_tokens": 3653851.0,
"step": 1985
},
{
"entropy": 5.859198522567749,
"epoch": 0.1662211827597728,
"grad_norm": 1.125,
"learning_rate": 0.0004999229500632279,
"loss": 5.6922,
"mean_token_accuracy": 0.16142508387565613,
"num_tokens": 3662720.0,
"step": 1990
},
{
"entropy": 5.774721765518189,
"epoch": 0.1666388239224858,
"grad_norm": 1.171875,
"learning_rate": 0.0004999221690697961,
"loss": 5.6495,
"mean_token_accuracy": 0.15686668008565902,
"num_tokens": 3671777.0,
"step": 1995
},
{
"entropy": 5.897257137298584,
"epoch": 0.1670564650851988,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999213841388312,
"loss": 5.7188,
"mean_token_accuracy": 0.16273299157619475,
"num_tokens": 3681390.0,
"step": 2000
},
{
"entropy": 5.959281492233276,
"epoch": 0.16747410624791179,
"grad_norm": 1.0625,
"learning_rate": 0.0004999205952703466,
"loss": 5.7871,
"mean_token_accuracy": 0.15210252553224562,
"num_tokens": 3690899.0,
"step": 2005
},
{
"entropy": 6.015440273284912,
"epoch": 0.1678917474106248,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999198024643563,
"loss": 5.8759,
"mean_token_accuracy": 0.14671805500984192,
"num_tokens": 3699891.0,
"step": 2010
},
{
"entropy": 5.827287578582764,
"epoch": 0.1683093885733378,
"grad_norm": 1.015625,
"learning_rate": 0.0004999190057208741,
"loss": 5.6696,
"mean_token_accuracy": 0.1584944322705269,
"num_tokens": 3709860.0,
"step": 2015
},
{
"entropy": 5.938063478469848,
"epoch": 0.16872702973605078,
"grad_norm": 1.0078125,
"learning_rate": 0.000499918205039914,
"loss": 5.803,
"mean_token_accuracy": 0.15343469232320786,
"num_tokens": 3718896.0,
"step": 2020
},
{
"entropy": 5.8418293476104735,
"epoch": 0.16914467089876378,
"grad_norm": 1.0625,
"learning_rate": 0.0004999174004214899,
"loss": 5.6123,
"mean_token_accuracy": 0.1574483186006546,
"num_tokens": 3727355.0,
"step": 2025
},
{
"entropy": 5.841806077957154,
"epoch": 0.1695623120614768,
"grad_norm": 1.03125,
"learning_rate": 0.0004999165918656161,
"loss": 5.7103,
"mean_token_accuracy": 0.15504993572831155,
"num_tokens": 3736191.0,
"step": 2030
},
{
"entropy": 5.958983039855957,
"epoch": 0.16997995322418977,
"grad_norm": 1.0625,
"learning_rate": 0.0004999157793723065,
"loss": 5.7395,
"mean_token_accuracy": 0.1586153984069824,
"num_tokens": 3744832.0,
"step": 2035
},
{
"entropy": 5.84045672416687,
"epoch": 0.17039759438690277,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999149629415755,
"loss": 5.7066,
"mean_token_accuracy": 0.15203840285539627,
"num_tokens": 3753835.0,
"step": 2040
},
{
"entropy": 6.055625247955322,
"epoch": 0.17081523554961578,
"grad_norm": 1.015625,
"learning_rate": 0.0004999141425734374,
"loss": 5.7747,
"mean_token_accuracy": 0.15490478426218032,
"num_tokens": 3762816.0,
"step": 2045
},
{
"entropy": 5.835732316970825,
"epoch": 0.17123287671232876,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999133182679064,
"loss": 5.8155,
"mean_token_accuracy": 0.148703470826149,
"num_tokens": 3773244.0,
"step": 2050
},
{
"entropy": 5.953923606872559,
"epoch": 0.17165051787504176,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999124900249971,
"loss": 5.6763,
"mean_token_accuracy": 0.16193537265062333,
"num_tokens": 3782122.0,
"step": 2055
},
{
"entropy": 5.870869255065918,
"epoch": 0.17206815903775477,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999116578447239,
"loss": 5.6705,
"mean_token_accuracy": 0.1667713090777397,
"num_tokens": 3791303.0,
"step": 2060
},
{
"entropy": 5.876138734817505,
"epoch": 0.17248580020046775,
"grad_norm": 0.9375,
"learning_rate": 0.0004999108217271015,
"loss": 5.7479,
"mean_token_accuracy": 0.15425308793783188,
"num_tokens": 3800991.0,
"step": 2065
},
{
"entropy": 5.857042074203491,
"epoch": 0.17290344136318075,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999099816721444,
"loss": 5.6609,
"mean_token_accuracy": 0.15457299053668977,
"num_tokens": 3811293.0,
"step": 2070
},
{
"entropy": 5.99418363571167,
"epoch": 0.17332108252589376,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999091376798673,
"loss": 5.7568,
"mean_token_accuracy": 0.15130494236946107,
"num_tokens": 3820013.0,
"step": 2075
},
{
"entropy": 5.852330875396729,
"epoch": 0.17373872368860674,
"grad_norm": 0.921875,
"learning_rate": 0.0004999082897502849,
"loss": 5.6497,
"mean_token_accuracy": 0.16134200170636176,
"num_tokens": 3829099.0,
"step": 2080
},
{
"entropy": 5.82087459564209,
"epoch": 0.17415636485131974,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999074378834123,
"loss": 5.7483,
"mean_token_accuracy": 0.15421984121203422,
"num_tokens": 3838308.0,
"step": 2085
},
{
"entropy": 5.92053747177124,
"epoch": 0.17457400601403275,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999065820792644,
"loss": 5.7005,
"mean_token_accuracy": 0.15973791033029555,
"num_tokens": 3847650.0,
"step": 2090
},
{
"entropy": 5.99317421913147,
"epoch": 0.17499164717674573,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999057223378559,
"loss": 5.8778,
"mean_token_accuracy": 0.1442355714738369,
"num_tokens": 3857252.0,
"step": 2095
},
{
"entropy": 5.863450002670288,
"epoch": 0.17540928833945874,
"grad_norm": 1.03125,
"learning_rate": 0.000499904858659202,
"loss": 5.7337,
"mean_token_accuracy": 0.15778396278619766,
"num_tokens": 3865747.0,
"step": 2100
},
{
"entropy": 5.886807632446289,
"epoch": 0.17582692950217174,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999039910433179,
"loss": 5.6976,
"mean_token_accuracy": 0.16511771082878113,
"num_tokens": 3874958.0,
"step": 2105
},
{
"entropy": 5.78854775428772,
"epoch": 0.17624457066488472,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999031194902187,
"loss": 5.6507,
"mean_token_accuracy": 0.16393667757511138,
"num_tokens": 3885216.0,
"step": 2110
},
{
"entropy": 5.815725708007813,
"epoch": 0.17666221182759773,
"grad_norm": 1.046875,
"learning_rate": 0.0004999022439999198,
"loss": 5.6814,
"mean_token_accuracy": 0.16577479541301726,
"num_tokens": 3893973.0,
"step": 2115
},
{
"entropy": 5.971462154388428,
"epoch": 0.17707985299031073,
"grad_norm": 1.078125,
"learning_rate": 0.0004999013645724362,
"loss": 5.7731,
"mean_token_accuracy": 0.1582237407565117,
"num_tokens": 3903532.0,
"step": 2120
},
{
"entropy": 5.9179940700531,
"epoch": 0.1774974941530237,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999004812077836,
"loss": 5.6958,
"mean_token_accuracy": 0.15923276245594026,
"num_tokens": 3913427.0,
"step": 2125
},
{
"entropy": 5.932846403121948,
"epoch": 0.17791513531573672,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998995939059773,
"loss": 5.8206,
"mean_token_accuracy": 0.145328551530838,
"num_tokens": 3923870.0,
"step": 2130
},
{
"entropy": 5.825693130493164,
"epoch": 0.17833277647844972,
"grad_norm": 0.97265625,
"learning_rate": 0.000499898702667033,
"loss": 5.6408,
"mean_token_accuracy": 0.1622411608695984,
"num_tokens": 3933469.0,
"step": 2135
},
{
"entropy": 5.8633805274963375,
"epoch": 0.1787504176411627,
"grad_norm": 1.125,
"learning_rate": 0.0004998978074909661,
"loss": 5.6579,
"mean_token_accuracy": 0.16844791918992996,
"num_tokens": 3942427.0,
"step": 2140
},
{
"entropy": 5.818320178985596,
"epoch": 0.1791680588038757,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998969083777925,
"loss": 5.6925,
"mean_token_accuracy": 0.16658158153295516,
"num_tokens": 3951250.0,
"step": 2145
},
{
"entropy": 5.894375801086426,
"epoch": 0.1795856999665887,
"grad_norm": 1.078125,
"learning_rate": 0.0004998960053275277,
"loss": 5.6894,
"mean_token_accuracy": 0.15957503467798234,
"num_tokens": 3959969.0,
"step": 2150
},
{
"entropy": 5.851364850997925,
"epoch": 0.1800033411293017,
"grad_norm": 1.109375,
"learning_rate": 0.0004998950983401875,
"loss": 5.7232,
"mean_token_accuracy": 0.15560920387506486,
"num_tokens": 3968171.0,
"step": 2155
},
{
"entropy": 5.920967769622803,
"epoch": 0.1804209822920147,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998941874157882,
"loss": 5.6209,
"mean_token_accuracy": 0.16588948369026185,
"num_tokens": 3976686.0,
"step": 2160
},
{
"entropy": 5.743179512023926,
"epoch": 0.1808386234547277,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998932725543453,
"loss": 5.7372,
"mean_token_accuracy": 0.15956501960754393,
"num_tokens": 3986552.0,
"step": 2165
},
{
"entropy": 5.888717126846314,
"epoch": 0.18125626461744068,
"grad_norm": 1.09375,
"learning_rate": 0.0004998923537558749,
"loss": 5.6116,
"mean_token_accuracy": 0.1573790043592453,
"num_tokens": 3995501.0,
"step": 2170
},
{
"entropy": 5.726497554779053,
"epoch": 0.1816739057801537,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998914310203932,
"loss": 5.6075,
"mean_token_accuracy": 0.16367665976285933,
"num_tokens": 4004423.0,
"step": 2175
},
{
"entropy": 5.9079173564910885,
"epoch": 0.1820915469428667,
"grad_norm": 1.03125,
"learning_rate": 0.0004998905043479163,
"loss": 5.796,
"mean_token_accuracy": 0.1520387388765812,
"num_tokens": 4013324.0,
"step": 2180
},
{
"entropy": 5.931830978393554,
"epoch": 0.18250918810557967,
"grad_norm": 1.046875,
"learning_rate": 0.0004998895737384605,
"loss": 5.6804,
"mean_token_accuracy": 0.15867839008569717,
"num_tokens": 4021896.0,
"step": 2185
},
{
"entropy": 5.754932498931884,
"epoch": 0.18292682926829268,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998886391920418,
"loss": 5.66,
"mean_token_accuracy": 0.1559813290834427,
"num_tokens": 4030590.0,
"step": 2190
},
{
"entropy": 5.8287865161895756,
"epoch": 0.18334447043100568,
"grad_norm": 1.0625,
"learning_rate": 0.0004998877007086769,
"loss": 5.5567,
"mean_token_accuracy": 0.17052167057991027,
"num_tokens": 4040041.0,
"step": 2195
},
{
"entropy": 5.837833738327026,
"epoch": 0.18376211159371866,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998867582883821,
"loss": 5.7246,
"mean_token_accuracy": 0.150713437050581,
"num_tokens": 4049119.0,
"step": 2200
},
{
"entropy": 5.864384412765503,
"epoch": 0.18417975275643167,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998858119311739,
"loss": 5.6599,
"mean_token_accuracy": 0.15918856784701346,
"num_tokens": 4059265.0,
"step": 2205
},
{
"entropy": 5.8326281070709225,
"epoch": 0.18459739391914468,
"grad_norm": 0.984375,
"learning_rate": 0.0004998848616370689,
"loss": 5.7001,
"mean_token_accuracy": 0.15809556990861892,
"num_tokens": 4068363.0,
"step": 2210
},
{
"entropy": 5.798278284072876,
"epoch": 0.18501503508185768,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998839074060837,
"loss": 5.5881,
"mean_token_accuracy": 0.16857375651597978,
"num_tokens": 4077390.0,
"step": 2215
},
{
"entropy": 5.79602165222168,
"epoch": 0.18543267624457066,
"grad_norm": 1.0546875,
"learning_rate": 0.000499882949238235,
"loss": 5.6749,
"mean_token_accuracy": 0.15564524084329606,
"num_tokens": 4087331.0,
"step": 2220
},
{
"entropy": 5.975500297546387,
"epoch": 0.18585031740728367,
"grad_norm": 1.015625,
"learning_rate": 0.0004998819871335396,
"loss": 5.8549,
"mean_token_accuracy": 0.15241608321666716,
"num_tokens": 4096425.0,
"step": 2225
},
{
"entropy": 5.81837944984436,
"epoch": 0.18626795856999667,
"grad_norm": 1.015625,
"learning_rate": 0.0004998810210920142,
"loss": 5.6292,
"mean_token_accuracy": 0.15902866423130035,
"num_tokens": 4106056.0,
"step": 2230
},
{
"entropy": 5.7763697624206545,
"epoch": 0.18668559973270965,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998800511136759,
"loss": 5.7116,
"mean_token_accuracy": 0.15947857201099397,
"num_tokens": 4115133.0,
"step": 2235
},
{
"entropy": 5.8516068935394285,
"epoch": 0.18710324089542266,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998790771985417,
"loss": 5.6267,
"mean_token_accuracy": 0.16477923840284348,
"num_tokens": 4123630.0,
"step": 2240
},
{
"entropy": 5.774742317199707,
"epoch": 0.18752088205813566,
"grad_norm": 0.9375,
"learning_rate": 0.0004998780993466284,
"loss": 5.7463,
"mean_token_accuracy": 0.16099598556756972,
"num_tokens": 4133337.0,
"step": 2245
},
{
"entropy": 5.892826843261719,
"epoch": 0.18793852322084864,
"grad_norm": 1.0625,
"learning_rate": 0.0004998771175579534,
"loss": 5.676,
"mean_token_accuracy": 0.1580575130879879,
"num_tokens": 4142240.0,
"step": 2250
},
{
"entropy": 5.782235956192016,
"epoch": 0.18835616438356165,
"grad_norm": 0.984375,
"learning_rate": 0.0004998761318325337,
"loss": 5.6686,
"mean_token_accuracy": 0.156616672873497,
"num_tokens": 4151297.0,
"step": 2255
},
{
"entropy": 5.789702367782593,
"epoch": 0.18877380554627465,
"grad_norm": 1.015625,
"learning_rate": 0.0004998751421703868,
"loss": 5.6215,
"mean_token_accuracy": 0.16046866476535798,
"num_tokens": 4160193.0,
"step": 2260
},
{
"entropy": 5.877978038787842,
"epoch": 0.18919144670898763,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998741485715297,
"loss": 5.7708,
"mean_token_accuracy": 0.15555500686168672,
"num_tokens": 4169948.0,
"step": 2265
},
{
"entropy": 5.985896205902099,
"epoch": 0.18960908787170064,
"grad_norm": 0.9296875,
"learning_rate": 0.00049987315103598,
"loss": 5.7972,
"mean_token_accuracy": 0.15081522166728972,
"num_tokens": 4179981.0,
"step": 2270
},
{
"entropy": 5.81521258354187,
"epoch": 0.19002672903441364,
"grad_norm": 1.03125,
"learning_rate": 0.0004998721495637551,
"loss": 5.7957,
"mean_token_accuracy": 0.15333186388015746,
"num_tokens": 4189172.0,
"step": 2275
},
{
"entropy": 5.865990352630615,
"epoch": 0.19044437019712662,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998711441548726,
"loss": 5.6126,
"mean_token_accuracy": 0.16020057201385499,
"num_tokens": 4197312.0,
"step": 2280
},
{
"entropy": 5.87011022567749,
"epoch": 0.19086201135983963,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998701348093499,
"loss": 5.6706,
"mean_token_accuracy": 0.16028127670288086,
"num_tokens": 4206828.0,
"step": 2285
},
{
"entropy": 5.936899137496948,
"epoch": 0.19127965252255263,
"grad_norm": 1.015625,
"learning_rate": 0.000499869121527205,
"loss": 5.6343,
"mean_token_accuracy": 0.1662514716386795,
"num_tokens": 4214835.0,
"step": 2290
},
{
"entropy": 5.84994478225708,
"epoch": 0.1916972936852656,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998681043084553,
"loss": 5.8139,
"mean_token_accuracy": 0.14824395328760148,
"num_tokens": 4225119.0,
"step": 2295
},
{
"entropy": 5.756076002120972,
"epoch": 0.19211493484797862,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998670831531188,
"loss": 5.6526,
"mean_token_accuracy": 0.15910039842128754,
"num_tokens": 4235257.0,
"step": 2300
},
{
"entropy": 5.82043194770813,
"epoch": 0.19253257601069163,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998660580612135,
"loss": 5.7215,
"mean_token_accuracy": 0.16029210090637208,
"num_tokens": 4244211.0,
"step": 2305
},
{
"entropy": 5.849356603622437,
"epoch": 0.1929502171734046,
"grad_norm": 1.078125,
"learning_rate": 0.0004998650290327569,
"loss": 5.7396,
"mean_token_accuracy": 0.15851273387670517,
"num_tokens": 4253149.0,
"step": 2310
},
{
"entropy": 5.765655517578125,
"epoch": 0.1933678583361176,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998639960677675,
"loss": 5.6319,
"mean_token_accuracy": 0.16373072266578675,
"num_tokens": 4262557.0,
"step": 2315
},
{
"entropy": 5.830093955993652,
"epoch": 0.19378549949883062,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998629591662631,
"loss": 5.6452,
"mean_token_accuracy": 0.15700610876083373,
"num_tokens": 4272881.0,
"step": 2320
},
{
"entropy": 5.771763753890991,
"epoch": 0.1942031406615436,
"grad_norm": 1.1328125,
"learning_rate": 0.000499861918328262,
"loss": 5.643,
"mean_token_accuracy": 0.1647744208574295,
"num_tokens": 4281158.0,
"step": 2325
},
{
"entropy": 5.816524362564087,
"epoch": 0.1946207818242566,
"grad_norm": 1.0,
"learning_rate": 0.0004998608735537822,
"loss": 5.644,
"mean_token_accuracy": 0.15966358929872512,
"num_tokens": 4290319.0,
"step": 2330
},
{
"entropy": 5.839132690429688,
"epoch": 0.1950384229869696,
"grad_norm": 1.0625,
"learning_rate": 0.0004998598248428423,
"loss": 5.7273,
"mean_token_accuracy": 0.16326339393854142,
"num_tokens": 4299276.0,
"step": 2335
},
{
"entropy": 5.8503776550292965,
"epoch": 0.19545606414968258,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998587721954604,
"loss": 5.692,
"mean_token_accuracy": 0.16148559898138046,
"num_tokens": 4308291.0,
"step": 2340
},
{
"entropy": 5.7621392726898195,
"epoch": 0.1958737053123956,
"grad_norm": 1.078125,
"learning_rate": 0.0004998577156116551,
"loss": 5.6517,
"mean_token_accuracy": 0.1572328507900238,
"num_tokens": 4316654.0,
"step": 2345
},
{
"entropy": 5.833258485794067,
"epoch": 0.1962913464751086,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998566550914448,
"loss": 5.5887,
"mean_token_accuracy": 0.15987759977579116,
"num_tokens": 4325911.0,
"step": 2350
},
{
"entropy": 5.780224704742432,
"epoch": 0.19670898763782158,
"grad_norm": 1.015625,
"learning_rate": 0.0004998555906348481,
"loss": 5.77,
"mean_token_accuracy": 0.1557260274887085,
"num_tokens": 4334846.0,
"step": 2355
},
{
"entropy": 5.895426082611084,
"epoch": 0.19712662880053458,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998545222418836,
"loss": 5.6489,
"mean_token_accuracy": 0.16278416961431502,
"num_tokens": 4343728.0,
"step": 2360
},
{
"entropy": 5.813882684707641,
"epoch": 0.1975442699632476,
"grad_norm": 1.03125,
"learning_rate": 0.0004998534499125701,
"loss": 5.6286,
"mean_token_accuracy": 0.1627249151468277,
"num_tokens": 4352774.0,
"step": 2365
},
{
"entropy": 5.867633724212647,
"epoch": 0.19796191112596057,
"grad_norm": 1.078125,
"learning_rate": 0.0004998523736469263,
"loss": 5.7684,
"mean_token_accuracy": 0.15933503061532975,
"num_tokens": 4361663.0,
"step": 2370
},
{
"entropy": 5.789565467834473,
"epoch": 0.19837955228867357,
"grad_norm": 1.046875,
"learning_rate": 0.000499851293444971,
"loss": 5.6116,
"mean_token_accuracy": 0.1717309609055519,
"num_tokens": 4370278.0,
"step": 2375
},
{
"entropy": 5.8761261940002445,
"epoch": 0.19879719345138658,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998502093067231,
"loss": 5.6661,
"mean_token_accuracy": 0.16252981573343278,
"num_tokens": 4379010.0,
"step": 2380
},
{
"entropy": 5.848679161071777,
"epoch": 0.19921483461409956,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998491212322017,
"loss": 5.6362,
"mean_token_accuracy": 0.15828320235013962,
"num_tokens": 4388145.0,
"step": 2385
},
{
"entropy": 5.763821983337403,
"epoch": 0.19963247577681256,
"grad_norm": 1.09375,
"learning_rate": 0.0004998480292214258,
"loss": 5.773,
"mean_token_accuracy": 0.15703087002038957,
"num_tokens": 4396321.0,
"step": 2390
},
{
"entropy": 5.807074356079101,
"epoch": 0.20005011693952557,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998469332744144,
"loss": 5.5447,
"mean_token_accuracy": 0.1663380816578865,
"num_tokens": 4404854.0,
"step": 2395
},
{
"entropy": 5.64944372177124,
"epoch": 0.20046775810223855,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998458333911869,
"loss": 5.5536,
"mean_token_accuracy": 0.16941631883382796,
"num_tokens": 4413890.0,
"step": 2400
},
{
"entropy": 5.781361818313599,
"epoch": 0.20088539926495155,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998447295717624,
"loss": 5.7074,
"mean_token_accuracy": 0.16001133918762206,
"num_tokens": 4423019.0,
"step": 2405
},
{
"entropy": 5.797937536239624,
"epoch": 0.20130304042766456,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998436218161602,
"loss": 5.6408,
"mean_token_accuracy": 0.16225794404745103,
"num_tokens": 4432340.0,
"step": 2410
},
{
"entropy": 5.792998313903809,
"epoch": 0.20172068159037754,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998425101243999,
"loss": 5.6188,
"mean_token_accuracy": 0.16200828105211257,
"num_tokens": 4441834.0,
"step": 2415
},
{
"entropy": 5.766503381729126,
"epoch": 0.20213832275309054,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998413944965007,
"loss": 5.6492,
"mean_token_accuracy": 0.16266501992940902,
"num_tokens": 4450864.0,
"step": 2420
},
{
"entropy": 5.73400182723999,
"epoch": 0.20255596391580355,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998402749324823,
"loss": 5.7124,
"mean_token_accuracy": 0.15515679866075516,
"num_tokens": 4460697.0,
"step": 2425
},
{
"entropy": 5.940784645080567,
"epoch": 0.20297360507851653,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998391514323642,
"loss": 5.7285,
"mean_token_accuracy": 0.15854752361774443,
"num_tokens": 4469926.0,
"step": 2430
},
{
"entropy": 5.7850086212158205,
"epoch": 0.20339124624122953,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998380239961661,
"loss": 5.6548,
"mean_token_accuracy": 0.16090933382511138,
"num_tokens": 4478818.0,
"step": 2435
},
{
"entropy": 5.808289384841919,
"epoch": 0.20380888740394254,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998368926239078,
"loss": 5.5844,
"mean_token_accuracy": 0.16591333746910095,
"num_tokens": 4487451.0,
"step": 2440
},
{
"entropy": 5.759612798690796,
"epoch": 0.20422652856665552,
"grad_norm": 1.0078125,
"learning_rate": 0.000499835757315609,
"loss": 5.6756,
"mean_token_accuracy": 0.15726785063743592,
"num_tokens": 4497674.0,
"step": 2445
},
{
"entropy": 5.7742194652557375,
"epoch": 0.20464416972936852,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998346180712898,
"loss": 5.624,
"mean_token_accuracy": 0.16603823602199555,
"num_tokens": 4507666.0,
"step": 2450
},
{
"entropy": 5.790293836593628,
"epoch": 0.20506181089208153,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998334748909698,
"loss": 5.7295,
"mean_token_accuracy": 0.15739290416240692,
"num_tokens": 4517409.0,
"step": 2455
},
{
"entropy": 5.786230230331421,
"epoch": 0.2054794520547945,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998323277746694,
"loss": 5.5241,
"mean_token_accuracy": 0.16496848613023757,
"num_tokens": 4526495.0,
"step": 2460
},
{
"entropy": 5.7622456550598145,
"epoch": 0.20589709321750752,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998311767224081,
"loss": 5.6131,
"mean_token_accuracy": 0.17294674664735793,
"num_tokens": 4535240.0,
"step": 2465
},
{
"entropy": 5.7548096656799315,
"epoch": 0.20631473438022052,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998300217342067,
"loss": 5.6976,
"mean_token_accuracy": 0.15931780487298966,
"num_tokens": 4544716.0,
"step": 2470
},
{
"entropy": 5.745911979675293,
"epoch": 0.2067323755429335,
"grad_norm": 1.0,
"learning_rate": 0.0004998288628100851,
"loss": 5.6222,
"mean_token_accuracy": 0.16541972160339355,
"num_tokens": 4553867.0,
"step": 2475
},
{
"entropy": 5.8166767120361325,
"epoch": 0.2071500167056465,
"grad_norm": 0.96875,
"learning_rate": 0.0004998276999500636,
"loss": 5.6764,
"mean_token_accuracy": 0.1619221478700638,
"num_tokens": 4562529.0,
"step": 2480
},
{
"entropy": 5.709495973587036,
"epoch": 0.2075676578683595,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998265331541627,
"loss": 5.4687,
"mean_token_accuracy": 0.1788138762116432,
"num_tokens": 4571041.0,
"step": 2485
},
{
"entropy": 5.6762677192687985,
"epoch": 0.2079852990310725,
"grad_norm": 1.078125,
"learning_rate": 0.0004998253624224026,
"loss": 5.5986,
"mean_token_accuracy": 0.1655976355075836,
"num_tokens": 4579914.0,
"step": 2490
},
{
"entropy": 5.769092082977295,
"epoch": 0.2084029401937855,
"grad_norm": 1.2109375,
"learning_rate": 0.000499824187754804,
"loss": 5.6967,
"mean_token_accuracy": 0.1630502760410309,
"num_tokens": 4590277.0,
"step": 2495
},
{
"entropy": 5.757724618911743,
"epoch": 0.2088205813564985,
"grad_norm": 1.046875,
"learning_rate": 0.0004998230091513872,
"loss": 5.6117,
"mean_token_accuracy": 0.16430521458387376,
"num_tokens": 4598719.0,
"step": 2500
},
{
"entropy": 5.7639039039611815,
"epoch": 0.20923822251921148,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998218266121731,
"loss": 5.5588,
"mean_token_accuracy": 0.16492444276809692,
"num_tokens": 4607804.0,
"step": 2505
},
{
"entropy": 5.7059930801391605,
"epoch": 0.2096558636819245,
"grad_norm": 0.96875,
"learning_rate": 0.0004998206401371823,
"loss": 5.5602,
"mean_token_accuracy": 0.16498781740665436,
"num_tokens": 4616529.0,
"step": 2510
},
{
"entropy": 5.782010936737061,
"epoch": 0.2100735048446375,
"grad_norm": 1.078125,
"learning_rate": 0.0004998194497264357,
"loss": 5.6004,
"mean_token_accuracy": 0.16248040944337844,
"num_tokens": 4624545.0,
"step": 2515
},
{
"entropy": 5.7379570484161375,
"epoch": 0.21049114600735047,
"grad_norm": 1.03125,
"learning_rate": 0.0004998182553799538,
"loss": 5.577,
"mean_token_accuracy": 0.16051217615604402,
"num_tokens": 4633316.0,
"step": 2520
},
{
"entropy": 5.7334644317626955,
"epoch": 0.21090878717006348,
"grad_norm": 1.046875,
"learning_rate": 0.0004998170570977579,
"loss": 5.5426,
"mean_token_accuracy": 0.16709231436252595,
"num_tokens": 4643353.0,
"step": 2525
},
{
"entropy": 5.724237775802612,
"epoch": 0.21132642833277648,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998158548798687,
"loss": 5.6328,
"mean_token_accuracy": 0.16201059520244598,
"num_tokens": 4651998.0,
"step": 2530
},
{
"entropy": 5.776553726196289,
"epoch": 0.21174406949548946,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998146487263074,
"loss": 5.6237,
"mean_token_accuracy": 0.16452135294675826,
"num_tokens": 4659937.0,
"step": 2535
},
{
"entropy": 5.748791980743408,
"epoch": 0.21216171065820247,
"grad_norm": 0.9375,
"learning_rate": 0.0004998134386370951,
"loss": 5.6036,
"mean_token_accuracy": 0.16133213490247728,
"num_tokens": 4669396.0,
"step": 2540
},
{
"entropy": 5.822993564605713,
"epoch": 0.21257935182091547,
"grad_norm": 0.9375,
"learning_rate": 0.0004998122246122529,
"loss": 5.675,
"mean_token_accuracy": 0.1632826492190361,
"num_tokens": 4679320.0,
"step": 2545
},
{
"entropy": 5.792739677429199,
"epoch": 0.21299699298362845,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998110066518021,
"loss": 5.5965,
"mean_token_accuracy": 0.16390374004840852,
"num_tokens": 4688807.0,
"step": 2550
},
{
"entropy": 5.798362731933594,
"epoch": 0.21341463414634146,
"grad_norm": 1.0390625,
"learning_rate": 0.000499809784755764,
"loss": 5.6089,
"mean_token_accuracy": 0.17050699889659882,
"num_tokens": 4697819.0,
"step": 2555
},
{
"entropy": 5.6853009223937985,
"epoch": 0.21383227530905446,
"grad_norm": 0.96875,
"learning_rate": 0.00049980855892416,
"loss": 5.5795,
"mean_token_accuracy": 0.16765243858098983,
"num_tokens": 4706931.0,
"step": 2560
},
{
"entropy": 5.7483304977417,
"epoch": 0.21424991647176747,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998073291570116,
"loss": 5.5681,
"mean_token_accuracy": 0.16450838446617128,
"num_tokens": 4716262.0,
"step": 2565
},
{
"entropy": 5.734201574325562,
"epoch": 0.21466755763448045,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998060954543404,
"loss": 5.5834,
"mean_token_accuracy": 0.16493556201457976,
"num_tokens": 4725415.0,
"step": 2570
},
{
"entropy": 5.7109521389007565,
"epoch": 0.21508519879719346,
"grad_norm": 1.0625,
"learning_rate": 0.0004998048578161678,
"loss": 5.6546,
"mean_token_accuracy": 0.16459743529558182,
"num_tokens": 4734580.0,
"step": 2575
},
{
"entropy": 5.665662527084351,
"epoch": 0.21550283995990646,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998036162425155,
"loss": 5.465,
"mean_token_accuracy": 0.17132774442434312,
"num_tokens": 4742966.0,
"step": 2580
},
{
"entropy": 5.726974296569824,
"epoch": 0.21592048112261944,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998023707334055,
"loss": 5.5987,
"mean_token_accuracy": 0.16571207493543624,
"num_tokens": 4751696.0,
"step": 2585
},
{
"entropy": 5.802803802490234,
"epoch": 0.21633812228533245,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998011212888592,
"loss": 5.6628,
"mean_token_accuracy": 0.16626541465520858,
"num_tokens": 4760460.0,
"step": 2590
},
{
"entropy": 5.766793346405029,
"epoch": 0.21675576344804545,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997998679088988,
"loss": 5.5858,
"mean_token_accuracy": 0.165497525036335,
"num_tokens": 4769004.0,
"step": 2595
},
{
"entropy": 5.726294231414795,
"epoch": 0.21717340461075843,
"grad_norm": 0.96875,
"learning_rate": 0.0004997986105935461,
"loss": 5.5707,
"mean_token_accuracy": 0.17018312215805054,
"num_tokens": 4777204.0,
"step": 2600
},
{
"entropy": 5.815764331817627,
"epoch": 0.21759104577347144,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997973493428231,
"loss": 5.7033,
"mean_token_accuracy": 0.16050707995891572,
"num_tokens": 4786766.0,
"step": 2605
},
{
"entropy": 5.666415119171143,
"epoch": 0.21800868693618444,
"grad_norm": 1.03125,
"learning_rate": 0.0004997960841567519,
"loss": 5.5961,
"mean_token_accuracy": 0.16938992440700532,
"num_tokens": 4796676.0,
"step": 2610
},
{
"entropy": 5.901765584945679,
"epoch": 0.21842632809889742,
"grad_norm": 1.078125,
"learning_rate": 0.0004997948150353548,
"loss": 5.5966,
"mean_token_accuracy": 0.1654299795627594,
"num_tokens": 4805514.0,
"step": 2615
},
{
"entropy": 5.714858770370483,
"epoch": 0.21884396926161043,
"grad_norm": 1.0,
"learning_rate": 0.0004997935419786537,
"loss": 5.6752,
"mean_token_accuracy": 0.16269783824682235,
"num_tokens": 4815487.0,
"step": 2620
},
{
"entropy": 5.806606769561768,
"epoch": 0.21926161042432343,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997922649866712,
"loss": 5.6183,
"mean_token_accuracy": 0.16302125453948973,
"num_tokens": 4825044.0,
"step": 2625
},
{
"entropy": 5.651657152175903,
"epoch": 0.2196792515870364,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997909840594294,
"loss": 5.4087,
"mean_token_accuracy": 0.17524338662624359,
"num_tokens": 4833875.0,
"step": 2630
},
{
"entropy": 5.638663101196289,
"epoch": 0.22009689274974942,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997896991969509,
"loss": 5.6418,
"mean_token_accuracy": 0.15811533331871033,
"num_tokens": 4843317.0,
"step": 2635
},
{
"entropy": 5.7495684146881105,
"epoch": 0.22051453391246242,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997884103992581,
"loss": 5.6321,
"mean_token_accuracy": 0.15845716297626494,
"num_tokens": 4852514.0,
"step": 2640
},
{
"entropy": 5.695010948181152,
"epoch": 0.2209321750751754,
"grad_norm": 1.0,
"learning_rate": 0.0004997871176663736,
"loss": 5.5792,
"mean_token_accuracy": 0.16959371864795686,
"num_tokens": 4861260.0,
"step": 2645
},
{
"entropy": 5.708796119689941,
"epoch": 0.2213498162378884,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997858209983201,
"loss": 5.5333,
"mean_token_accuracy": 0.16668965592980384,
"num_tokens": 4869654.0,
"step": 2650
},
{
"entropy": 5.724199867248535,
"epoch": 0.22176745740060141,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997845203951202,
"loss": 5.5937,
"mean_token_accuracy": 0.15722348690032958,
"num_tokens": 4878311.0,
"step": 2655
},
{
"entropy": 5.7917327880859375,
"epoch": 0.2221850985633144,
"grad_norm": 1.078125,
"learning_rate": 0.0004997832158567967,
"loss": 5.6604,
"mean_token_accuracy": 0.15930023267865182,
"num_tokens": 4887251.0,
"step": 2660
},
{
"entropy": 5.709334659576416,
"epoch": 0.2226027397260274,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997819073833724,
"loss": 5.5234,
"mean_token_accuracy": 0.1640756532549858,
"num_tokens": 4896528.0,
"step": 2665
},
{
"entropy": 5.699781513214111,
"epoch": 0.2230203808887404,
"grad_norm": 1.015625,
"learning_rate": 0.0004997805949748702,
"loss": 5.5851,
"mean_token_accuracy": 0.16082951575517654,
"num_tokens": 4904962.0,
"step": 2670
},
{
"entropy": 5.751451635360718,
"epoch": 0.22343802205145338,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997792786313132,
"loss": 5.5949,
"mean_token_accuracy": 0.16221776455640793,
"num_tokens": 4914735.0,
"step": 2675
},
{
"entropy": 5.555906057357788,
"epoch": 0.2238556632141664,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997779583527243,
"loss": 5.4638,
"mean_token_accuracy": 0.1665905386209488,
"num_tokens": 4924271.0,
"step": 2680
},
{
"entropy": 5.8695250988006595,
"epoch": 0.2242733043768794,
"grad_norm": 1.015625,
"learning_rate": 0.0004997766341391268,
"loss": 5.7047,
"mean_token_accuracy": 0.155639585852623,
"num_tokens": 4933928.0,
"step": 2685
},
{
"entropy": 5.705683994293213,
"epoch": 0.22469094553959237,
"grad_norm": 1.046875,
"learning_rate": 0.0004997753059905437,
"loss": 5.4135,
"mean_token_accuracy": 0.1759113550186157,
"num_tokens": 4942714.0,
"step": 2690
},
{
"entropy": 5.6081311225891115,
"epoch": 0.22510858670230538,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997739739069981,
"loss": 5.541,
"mean_token_accuracy": 0.16623980104923247,
"num_tokens": 4952297.0,
"step": 2695
},
{
"entropy": 5.701430177688598,
"epoch": 0.22552622786501839,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997726378885138,
"loss": 5.48,
"mean_token_accuracy": 0.1763035088777542,
"num_tokens": 4960947.0,
"step": 2700
},
{
"entropy": 5.713150548934936,
"epoch": 0.22594386902773136,
"grad_norm": 1.21875,
"learning_rate": 0.0004997712979351137,
"loss": 5.6433,
"mean_token_accuracy": 0.16017448753118516,
"num_tokens": 4969705.0,
"step": 2705
},
{
"entropy": 5.7782330989837645,
"epoch": 0.22636151019044437,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997699540468216,
"loss": 5.6095,
"mean_token_accuracy": 0.16398101300001144,
"num_tokens": 4979112.0,
"step": 2710
},
{
"entropy": 5.707289409637451,
"epoch": 0.22677915135315738,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997686062236609,
"loss": 5.5245,
"mean_token_accuracy": 0.1688684806227684,
"num_tokens": 4987894.0,
"step": 2715
},
{
"entropy": 5.728122043609619,
"epoch": 0.22719679251587035,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997672544656553,
"loss": 5.6685,
"mean_token_accuracy": 0.1611431822180748,
"num_tokens": 4997774.0,
"step": 2720
},
{
"entropy": 5.774860763549805,
"epoch": 0.22761443367858336,
"grad_norm": 1.0,
"learning_rate": 0.0004997658987728283,
"loss": 5.5835,
"mean_token_accuracy": 0.17197922468185425,
"num_tokens": 5006183.0,
"step": 2725
},
{
"entropy": 5.711122512817383,
"epoch": 0.22803207484129637,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997645391452035,
"loss": 5.5795,
"mean_token_accuracy": 0.1666414961218834,
"num_tokens": 5016909.0,
"step": 2730
},
{
"entropy": 5.7371501445770265,
"epoch": 0.22844971600400935,
"grad_norm": 1.0,
"learning_rate": 0.000499763175582805,
"loss": 5.6246,
"mean_token_accuracy": 0.16619712710380555,
"num_tokens": 5027326.0,
"step": 2735
},
{
"entropy": 5.773577308654785,
"epoch": 0.22886735716672235,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997618080856566,
"loss": 5.5787,
"mean_token_accuracy": 0.16747207343578338,
"num_tokens": 5037043.0,
"step": 2740
},
{
"entropy": 5.614604187011719,
"epoch": 0.22928499832943536,
"grad_norm": 0.953125,
"learning_rate": 0.0004997604366537823,
"loss": 5.5621,
"mean_token_accuracy": 0.16502413600683213,
"num_tokens": 5045992.0,
"step": 2745
},
{
"entropy": 5.674044513702393,
"epoch": 0.22970263949214834,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997590612872058,
"loss": 5.5419,
"mean_token_accuracy": 0.16780618876218795,
"num_tokens": 5055568.0,
"step": 2750
},
{
"entropy": 5.761773729324341,
"epoch": 0.23012028065486134,
"grad_norm": 0.8359375,
"learning_rate": 0.0004997576819859515,
"loss": 5.617,
"mean_token_accuracy": 0.1640057623386383,
"num_tokens": 5065512.0,
"step": 2755
},
{
"entropy": 5.6007527828216555,
"epoch": 0.23053792181757435,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997562987500434,
"loss": 5.4683,
"mean_token_accuracy": 0.17772611379623413,
"num_tokens": 5074776.0,
"step": 2760
},
{
"entropy": 5.620506668090821,
"epoch": 0.23095556298028733,
"grad_norm": 0.9375,
"learning_rate": 0.0004997549115795058,
"loss": 5.518,
"mean_token_accuracy": 0.16908519864082336,
"num_tokens": 5084743.0,
"step": 2765
},
{
"entropy": 5.7845690727233885,
"epoch": 0.23137320414300033,
"grad_norm": 0.99609375,
"learning_rate": 0.000499753520474363,
"loss": 5.6789,
"mean_token_accuracy": 0.16238866448402406,
"num_tokens": 5095387.0,
"step": 2770
},
{
"entropy": 5.675830888748169,
"epoch": 0.23179084530571334,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997521254346391,
"loss": 5.5874,
"mean_token_accuracy": 0.16865266785025596,
"num_tokens": 5104628.0,
"step": 2775
},
{
"entropy": 5.757327651977539,
"epoch": 0.23220848646842632,
"grad_norm": 0.953125,
"learning_rate": 0.0004997507264603589,
"loss": 5.5401,
"mean_token_accuracy": 0.16944464445114135,
"num_tokens": 5113169.0,
"step": 2780
},
{
"entropy": 5.695029544830322,
"epoch": 0.23262612763113932,
"grad_norm": 0.96875,
"learning_rate": 0.0004997493235515466,
"loss": 5.601,
"mean_token_accuracy": 0.16851502060890197,
"num_tokens": 5122294.0,
"step": 2785
},
{
"entropy": 5.605187559127808,
"epoch": 0.23304376879385233,
"grad_norm": 0.921875,
"learning_rate": 0.0004997479167082269,
"loss": 5.4361,
"mean_token_accuracy": 0.17638742476701735,
"num_tokens": 5131191.0,
"step": 2790
},
{
"entropy": 5.674936103820801,
"epoch": 0.2334614099565653,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997465059304243,
"loss": 5.6504,
"mean_token_accuracy": 0.167001473903656,
"num_tokens": 5141974.0,
"step": 2795
},
{
"entropy": 5.781553602218628,
"epoch": 0.23387905111927831,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997450912181637,
"loss": 5.5828,
"mean_token_accuracy": 0.16855430603027344,
"num_tokens": 5151412.0,
"step": 2800
},
{
"entropy": 5.66849799156189,
"epoch": 0.23429669228199132,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997436725714696,
"loss": 5.5071,
"mean_token_accuracy": 0.17201763838529588,
"num_tokens": 5160989.0,
"step": 2805
},
{
"entropy": 5.687669134140014,
"epoch": 0.2347143334447043,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997422499903671,
"loss": 5.575,
"mean_token_accuracy": 0.16350428611040116,
"num_tokens": 5170241.0,
"step": 2810
},
{
"entropy": 5.852313756942749,
"epoch": 0.2351319746074173,
"grad_norm": 0.96484375,
"learning_rate": 0.000499740823474881,
"loss": 5.693,
"mean_token_accuracy": 0.15919099301099776,
"num_tokens": 5180247.0,
"step": 2815
},
{
"entropy": 5.683443641662597,
"epoch": 0.2355496157701303,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997393930250363,
"loss": 5.5877,
"mean_token_accuracy": 0.16531823873519896,
"num_tokens": 5189144.0,
"step": 2820
},
{
"entropy": 5.660533905029297,
"epoch": 0.2359672569328433,
"grad_norm": 0.91796875,
"learning_rate": 0.000499737958640858,
"loss": 5.6363,
"mean_token_accuracy": 0.16292943805456161,
"num_tokens": 5199410.0,
"step": 2825
},
{
"entropy": 5.772460985183716,
"epoch": 0.2363848980955563,
"grad_norm": 0.88671875,
"learning_rate": 0.0004997365203223711,
"loss": 5.5163,
"mean_token_accuracy": 0.17269660532474518,
"num_tokens": 5209032.0,
"step": 2830
},
{
"entropy": 5.5716253280639645,
"epoch": 0.2368025392582693,
"grad_norm": 0.98828125,
"learning_rate": 0.000499735078069601,
"loss": 5.4117,
"mean_token_accuracy": 0.17505983710289003,
"num_tokens": 5217787.0,
"step": 2835
},
{
"entropy": 5.610530710220337,
"epoch": 0.23722018042098228,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997336318825728,
"loss": 5.5272,
"mean_token_accuracy": 0.1705465003848076,
"num_tokens": 5226962.0,
"step": 2840
},
{
"entropy": 5.678059244155884,
"epoch": 0.23763782158369529,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997321817613119,
"loss": 5.501,
"mean_token_accuracy": 0.166815347969532,
"num_tokens": 5236111.0,
"step": 2845
},
{
"entropy": 5.748322200775147,
"epoch": 0.2380554627464083,
"grad_norm": 0.875,
"learning_rate": 0.0004997307277058436,
"loss": 5.6462,
"mean_token_accuracy": 0.16232668459415436,
"num_tokens": 5246628.0,
"step": 2850
},
{
"entropy": 5.667980480194092,
"epoch": 0.23847310390912127,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997292697161935,
"loss": 5.5754,
"mean_token_accuracy": 0.16228308975696565,
"num_tokens": 5255543.0,
"step": 2855
},
{
"entropy": 5.61335859298706,
"epoch": 0.23889074507183428,
"grad_norm": 0.9921875,
"learning_rate": 0.000499727807792387,
"loss": 5.4157,
"mean_token_accuracy": 0.1765509083867073,
"num_tokens": 5264559.0,
"step": 2860
},
{
"entropy": 5.593845558166504,
"epoch": 0.23930838623454728,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997263419344495,
"loss": 5.5031,
"mean_token_accuracy": 0.17075306475162505,
"num_tokens": 5274599.0,
"step": 2865
},
{
"entropy": 5.6120987892150875,
"epoch": 0.23972602739726026,
"grad_norm": 1.03125,
"learning_rate": 0.0004997248721424071,
"loss": 5.4684,
"mean_token_accuracy": 0.17607724964618682,
"num_tokens": 5283495.0,
"step": 2870
},
{
"entropy": 5.748496675491333,
"epoch": 0.24014366855997327,
"grad_norm": 1.09375,
"learning_rate": 0.0004997233984162853,
"loss": 5.5206,
"mean_token_accuracy": 0.17262364476919173,
"num_tokens": 5292064.0,
"step": 2875
},
{
"entropy": 5.660027360916137,
"epoch": 0.24056130972268627,
"grad_norm": 0.90625,
"learning_rate": 0.0004997219207561099,
"loss": 5.5373,
"mean_token_accuracy": 0.16632017493247986,
"num_tokens": 5301696.0,
"step": 2880
},
{
"entropy": 5.7199009418487545,
"epoch": 0.24097895088539925,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997204391619068,
"loss": 5.647,
"mean_token_accuracy": 0.15879270881414415,
"num_tokens": 5311427.0,
"step": 2885
},
{
"entropy": 5.826184225082398,
"epoch": 0.24139659204811226,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997189536337018,
"loss": 5.5703,
"mean_token_accuracy": 0.1678813174366951,
"num_tokens": 5321406.0,
"step": 2890
},
{
"entropy": 5.70261926651001,
"epoch": 0.24181423321082526,
"grad_norm": 1.09375,
"learning_rate": 0.0004997174641715211,
"loss": 5.6536,
"mean_token_accuracy": 0.16488752663135528,
"num_tokens": 5332257.0,
"step": 2895
},
{
"entropy": 5.561820459365845,
"epoch": 0.24223187437353827,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997159707753906,
"loss": 5.391,
"mean_token_accuracy": 0.17732247412204744,
"num_tokens": 5341857.0,
"step": 2900
},
{
"entropy": 5.639991235733032,
"epoch": 0.24264951553625125,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997144734453367,
"loss": 5.5403,
"mean_token_accuracy": 0.1736094608902931,
"num_tokens": 5351367.0,
"step": 2905
},
{
"entropy": 5.567778444290161,
"epoch": 0.24306715669896425,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997129721813853,
"loss": 5.3888,
"mean_token_accuracy": 0.17335254251956939,
"num_tokens": 5360627.0,
"step": 2910
},
{
"entropy": 5.627420234680176,
"epoch": 0.24348479786167726,
"grad_norm": 1.015625,
"learning_rate": 0.0004997114669835629,
"loss": 5.4713,
"mean_token_accuracy": 0.17678721398115158,
"num_tokens": 5369703.0,
"step": 2915
},
{
"entropy": 5.658342075347901,
"epoch": 0.24390243902439024,
"grad_norm": 1.109375,
"learning_rate": 0.0004997099578518959,
"loss": 5.5467,
"mean_token_accuracy": 0.16832583546638488,
"num_tokens": 5378432.0,
"step": 2920
},
{
"entropy": 5.573386335372925,
"epoch": 0.24432008018710324,
"grad_norm": 1.125,
"learning_rate": 0.0004997084447864105,
"loss": 5.3981,
"mean_token_accuracy": 0.17398632317781448,
"num_tokens": 5386489.0,
"step": 2925
},
{
"entropy": 5.770393562316895,
"epoch": 0.24473772134981625,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997069277871333,
"loss": 5.6219,
"mean_token_accuracy": 0.16319795697927475,
"num_tokens": 5395908.0,
"step": 2930
},
{
"entropy": 5.763458490371704,
"epoch": 0.24515536251252923,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997054068540908,
"loss": 5.5425,
"mean_token_accuracy": 0.17221789509058,
"num_tokens": 5404822.0,
"step": 2935
},
{
"entropy": 5.60934100151062,
"epoch": 0.24557300367524224,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997038819873098,
"loss": 5.4537,
"mean_token_accuracy": 0.17699288725852966,
"num_tokens": 5413980.0,
"step": 2940
},
{
"entropy": 5.652670669555664,
"epoch": 0.24599064483795524,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997023531868167,
"loss": 5.5227,
"mean_token_accuracy": 0.16705751717090606,
"num_tokens": 5423538.0,
"step": 2945
},
{
"entropy": 5.616042518615723,
"epoch": 0.24640828600066822,
"grad_norm": 0.984375,
"learning_rate": 0.0004997008204526386,
"loss": 5.4503,
"mean_token_accuracy": 0.17180379182100297,
"num_tokens": 5431930.0,
"step": 2950
},
{
"entropy": 5.6258384704589846,
"epoch": 0.24682592716338123,
"grad_norm": 0.90625,
"learning_rate": 0.0004996992837848021,
"loss": 5.4514,
"mean_token_accuracy": 0.17689115405082703,
"num_tokens": 5440721.0,
"step": 2955
},
{
"entropy": 5.611420965194702,
"epoch": 0.24724356832609423,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996977431833342,
"loss": 5.5153,
"mean_token_accuracy": 0.1651511862874031,
"num_tokens": 5449617.0,
"step": 2960
},
{
"entropy": 5.621954154968262,
"epoch": 0.2476612094888072,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996961986482618,
"loss": 5.5521,
"mean_token_accuracy": 0.17019403576850892,
"num_tokens": 5458827.0,
"step": 2965
},
{
"entropy": 5.697111034393311,
"epoch": 0.24807885065152022,
"grad_norm": 1.046875,
"learning_rate": 0.0004996946501796119,
"loss": 5.4353,
"mean_token_accuracy": 0.17367945313453675,
"num_tokens": 5467348.0,
"step": 2970
},
{
"entropy": 5.669052839279175,
"epoch": 0.24849649181423322,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996930977774118,
"loss": 5.505,
"mean_token_accuracy": 0.1617908701300621,
"num_tokens": 5476645.0,
"step": 2975
},
{
"entropy": 5.712230110168457,
"epoch": 0.2489141329769462,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996915414416885,
"loss": 5.5382,
"mean_token_accuracy": 0.16787463724613189,
"num_tokens": 5486120.0,
"step": 2980
},
{
"entropy": 5.698201894760132,
"epoch": 0.2493317741396592,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996899811724693,
"loss": 5.5308,
"mean_token_accuracy": 0.1670474722981453,
"num_tokens": 5495313.0,
"step": 2985
},
{
"entropy": 5.647207641601563,
"epoch": 0.2497494153023722,
"grad_norm": 0.9375,
"learning_rate": 0.0004996884169697816,
"loss": 5.5247,
"mean_token_accuracy": 0.1688663676381111,
"num_tokens": 5504587.0,
"step": 2990
},
{
"entropy": 5.615376949310303,
"epoch": 0.2501670564650852,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996868488336526,
"loss": 5.5488,
"mean_token_accuracy": 0.16672224402427674,
"num_tokens": 5514188.0,
"step": 2995
},
{
"entropy": 5.70684118270874,
"epoch": 0.25058469762779817,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996852767641099,
"loss": 5.5913,
"mean_token_accuracy": 0.15994588136672974,
"num_tokens": 5524075.0,
"step": 3000
},
{
"epoch": 0.25058469762779817,
"eval_entropy": 5.478494798397265,
"eval_loss": 5.498563766479492,
"eval_mean_token_accuracy": 0.17943838510794094,
"eval_num_tokens": 5524075.0,
"eval_runtime": 20.9226,
"eval_samples_per_second": 1804.887,
"eval_steps_per_second": 225.641,
"step": 3000
},
{
"entropy": 5.602593278884887,
"epoch": 0.2510023387905112,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996837007611811,
"loss": 5.3595,
"mean_token_accuracy": 0.17527442872524263,
"num_tokens": 5533323.0,
"step": 3005
},
{
"entropy": 5.5606709003448485,
"epoch": 0.2514199799532242,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996821208248935,
"loss": 5.3825,
"mean_token_accuracy": 0.17830902934074402,
"num_tokens": 5542288.0,
"step": 3010
},
{
"entropy": 5.716786861419678,
"epoch": 0.25183762111593716,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996805369552749,
"loss": 5.5523,
"mean_token_accuracy": 0.16320647895336152,
"num_tokens": 5550869.0,
"step": 3015
},
{
"entropy": 5.655161046981812,
"epoch": 0.2522552622786502,
"grad_norm": 1.0,
"learning_rate": 0.0004996789491523532,
"loss": 5.4633,
"mean_token_accuracy": 0.17710816562175752,
"num_tokens": 5560147.0,
"step": 3020
},
{
"entropy": 5.5786840438842775,
"epoch": 0.2526729034413632,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996773574161559,
"loss": 5.3677,
"mean_token_accuracy": 0.17918083518743516,
"num_tokens": 5568358.0,
"step": 3025
},
{
"entropy": 5.612635326385498,
"epoch": 0.25309054460407615,
"grad_norm": 1.0,
"learning_rate": 0.0004996757617467112,
"loss": 5.4814,
"mean_token_accuracy": 0.16537049114704133,
"num_tokens": 5577755.0,
"step": 3030
},
{
"entropy": 5.682578229904175,
"epoch": 0.2535081857667892,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996741621440466,
"loss": 5.5401,
"mean_token_accuracy": 0.15776935815811158,
"num_tokens": 5587515.0,
"step": 3035
},
{
"entropy": 5.710801124572754,
"epoch": 0.25392582692950216,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996725586081906,
"loss": 5.6639,
"mean_token_accuracy": 0.16025283336639404,
"num_tokens": 5597860.0,
"step": 3040
},
{
"entropy": 5.701569509506226,
"epoch": 0.25434346809221514,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996709511391709,
"loss": 5.5256,
"mean_token_accuracy": 0.17191781997680664,
"num_tokens": 5607302.0,
"step": 3045
},
{
"entropy": 5.640579891204834,
"epoch": 0.2547611092549282,
"grad_norm": 0.96875,
"learning_rate": 0.0004996693397370158,
"loss": 5.5923,
"mean_token_accuracy": 0.16542113721370696,
"num_tokens": 5616345.0,
"step": 3050
},
{
"entropy": 5.631747817993164,
"epoch": 0.25517875041764115,
"grad_norm": 1.078125,
"learning_rate": 0.0004996677244017534,
"loss": 5.3853,
"mean_token_accuracy": 0.173073410987854,
"num_tokens": 5624745.0,
"step": 3055
},
{
"entropy": 5.668503999710083,
"epoch": 0.25559639158035413,
"grad_norm": 1.046875,
"learning_rate": 0.0004996661051334121,
"loss": 5.5228,
"mean_token_accuracy": 0.16704076677560806,
"num_tokens": 5634022.0,
"step": 3060
},
{
"entropy": 5.685352897644043,
"epoch": 0.25601403274306717,
"grad_norm": 0.953125,
"learning_rate": 0.0004996644819320201,
"loss": 5.461,
"mean_token_accuracy": 0.17210687100887298,
"num_tokens": 5643631.0,
"step": 3065
},
{
"entropy": 5.68446683883667,
"epoch": 0.25643167390578014,
"grad_norm": 1.078125,
"learning_rate": 0.0004996628547976061,
"loss": 5.5511,
"mean_token_accuracy": 0.16774857938289642,
"num_tokens": 5653155.0,
"step": 3070
},
{
"entropy": 5.601983499526978,
"epoch": 0.2568493150684932,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996612237301983,
"loss": 5.5434,
"mean_token_accuracy": 0.17165842205286025,
"num_tokens": 5663091.0,
"step": 3075
},
{
"entropy": 5.597293710708618,
"epoch": 0.25726695623120616,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996595887298254,
"loss": 5.448,
"mean_token_accuracy": 0.17855413258075714,
"num_tokens": 5671885.0,
"step": 3080
},
{
"entropy": 5.676712846755981,
"epoch": 0.25768459739391913,
"grad_norm": 1.015625,
"learning_rate": 0.000499657949796516,
"loss": 5.4666,
"mean_token_accuracy": 0.17005476504564285,
"num_tokens": 5680225.0,
"step": 3085
},
{
"entropy": 5.5222419738769535,
"epoch": 0.25810223855663217,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996563069302986,
"loss": 5.3245,
"mean_token_accuracy": 0.18394140303134918,
"num_tokens": 5689456.0,
"step": 3090
},
{
"entropy": 5.574301099777221,
"epoch": 0.25851987971934515,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996546601312023,
"loss": 5.4456,
"mean_token_accuracy": 0.1766171708703041,
"num_tokens": 5698127.0,
"step": 3095
},
{
"entropy": 5.573929214477539,
"epoch": 0.2589375208820581,
"grad_norm": 1.0,
"learning_rate": 0.0004996530093992558,
"loss": 5.4954,
"mean_token_accuracy": 0.173156276345253,
"num_tokens": 5706825.0,
"step": 3100
},
{
"entropy": 5.660216188430786,
"epoch": 0.25935516204477116,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996513547344877,
"loss": 5.4357,
"mean_token_accuracy": 0.17253331542015077,
"num_tokens": 5715677.0,
"step": 3105
},
{
"entropy": 5.5564922332763675,
"epoch": 0.25977280320748414,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996496961369274,
"loss": 5.3672,
"mean_token_accuracy": 0.17406684905290604,
"num_tokens": 5725371.0,
"step": 3110
},
{
"entropy": 5.631770658493042,
"epoch": 0.2601904443701971,
"grad_norm": 1.015625,
"learning_rate": 0.0004996480336066037,
"loss": 5.4725,
"mean_token_accuracy": 0.1753530591726303,
"num_tokens": 5734664.0,
"step": 3115
},
{
"entropy": 5.629898118972778,
"epoch": 0.26060808553291015,
"grad_norm": 1.0,
"learning_rate": 0.0004996463671435458,
"loss": 5.4649,
"mean_token_accuracy": 0.17245597690343856,
"num_tokens": 5743870.0,
"step": 3120
},
{
"entropy": 5.492088651657104,
"epoch": 0.26102572669562313,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996446967477829,
"loss": 5.4726,
"mean_token_accuracy": 0.175869382917881,
"num_tokens": 5752855.0,
"step": 3125
},
{
"entropy": 5.738801383972168,
"epoch": 0.2614433678583361,
"grad_norm": 0.8359375,
"learning_rate": 0.0004996430224193441,
"loss": 5.5367,
"mean_token_accuracy": 0.17029510140419007,
"num_tokens": 5762435.0,
"step": 3130
},
{
"entropy": 5.678168392181396,
"epoch": 0.26186100902104914,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996413441582589,
"loss": 5.5338,
"mean_token_accuracy": 0.17078250348567964,
"num_tokens": 5771463.0,
"step": 3135
},
{
"entropy": 5.6406927585601805,
"epoch": 0.2622786501837621,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996396619645564,
"loss": 5.5302,
"mean_token_accuracy": 0.16546972393989562,
"num_tokens": 5780530.0,
"step": 3140
},
{
"entropy": 5.630699539184571,
"epoch": 0.2626962913464751,
"grad_norm": 0.96875,
"learning_rate": 0.0004996379758382662,
"loss": 5.4006,
"mean_token_accuracy": 0.17883287668228148,
"num_tokens": 5789199.0,
"step": 3145
},
{
"entropy": 5.73064489364624,
"epoch": 0.26311393250918813,
"grad_norm": 0.84765625,
"learning_rate": 0.000499636285779418,
"loss": 5.5918,
"mean_token_accuracy": 0.16522125601768495,
"num_tokens": 5799170.0,
"step": 3150
},
{
"entropy": 5.62532901763916,
"epoch": 0.2635315736719011,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996345917880412,
"loss": 5.5021,
"mean_token_accuracy": 0.16831852197647096,
"num_tokens": 5807662.0,
"step": 3155
},
{
"entropy": 5.549431276321411,
"epoch": 0.2639492148346141,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996328938641654,
"loss": 5.4146,
"mean_token_accuracy": 0.17468473613262175,
"num_tokens": 5815790.0,
"step": 3160
},
{
"entropy": 5.642532205581665,
"epoch": 0.2643668559973271,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996311920078204,
"loss": 5.4266,
"mean_token_accuracy": 0.17523301988840104,
"num_tokens": 5824746.0,
"step": 3165
},
{
"entropy": 5.574083042144776,
"epoch": 0.2647844971600401,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996294862190361,
"loss": 5.4408,
"mean_token_accuracy": 0.1776533082127571,
"num_tokens": 5833678.0,
"step": 3170
},
{
"entropy": 5.630044078826904,
"epoch": 0.2652021383227531,
"grad_norm": 1.0078125,
"learning_rate": 0.000499627776497842,
"loss": 5.4853,
"mean_token_accuracy": 0.1765962466597557,
"num_tokens": 5842218.0,
"step": 3175
},
{
"entropy": 5.796161222457886,
"epoch": 0.2656197794854661,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996260628442686,
"loss": 5.6637,
"mean_token_accuracy": 0.1582587666809559,
"num_tokens": 5853250.0,
"step": 3180
},
{
"entropy": 5.619317722320557,
"epoch": 0.2660374206481791,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996243452583454,
"loss": 5.3935,
"mean_token_accuracy": 0.18060708791017532,
"num_tokens": 5862661.0,
"step": 3185
},
{
"entropy": 5.644404077529908,
"epoch": 0.26645506181089207,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996226237401028,
"loss": 5.5492,
"mean_token_accuracy": 0.16795668601989747,
"num_tokens": 5872327.0,
"step": 3190
},
{
"entropy": 5.516964340209961,
"epoch": 0.2668727029736051,
"grad_norm": 1.0,
"learning_rate": 0.0004996208982895706,
"loss": 5.4026,
"mean_token_accuracy": 0.17274871319532395,
"num_tokens": 5881539.0,
"step": 3195
},
{
"entropy": 5.665464544296265,
"epoch": 0.2672903441363181,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996191689067794,
"loss": 5.511,
"mean_token_accuracy": 0.17336390018463135,
"num_tokens": 5891602.0,
"step": 3200
},
{
"entropy": 5.600575494766235,
"epoch": 0.26770798529903106,
"grad_norm": 1.015625,
"learning_rate": 0.0004996174355917591,
"loss": 5.5014,
"mean_token_accuracy": 0.1688656896352768,
"num_tokens": 5899831.0,
"step": 3205
},
{
"entropy": 5.702366590499878,
"epoch": 0.2681256264617441,
"grad_norm": 0.953125,
"learning_rate": 0.0004996156983445404,
"loss": 5.6011,
"mean_token_accuracy": 0.1662008658051491,
"num_tokens": 5909193.0,
"step": 3210
},
{
"entropy": 5.626371002197265,
"epoch": 0.26854326762445707,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996139571651534,
"loss": 5.5028,
"mean_token_accuracy": 0.1691843181848526,
"num_tokens": 5919353.0,
"step": 3215
},
{
"entropy": 5.608973217010498,
"epoch": 0.26896090878717005,
"grad_norm": 1.09375,
"learning_rate": 0.0004996122120536288,
"loss": 5.4441,
"mean_token_accuracy": 0.17524013072252273,
"num_tokens": 5927853.0,
"step": 3220
},
{
"entropy": 5.546483087539673,
"epoch": 0.2693785499498831,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996104630099969,
"loss": 5.5466,
"mean_token_accuracy": 0.17366677075624465,
"num_tokens": 5937943.0,
"step": 3225
},
{
"entropy": 5.6188586235046385,
"epoch": 0.26979619111259606,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996087100342885,
"loss": 5.4546,
"mean_token_accuracy": 0.17476439625024795,
"num_tokens": 5946213.0,
"step": 3230
},
{
"entropy": 5.604651737213135,
"epoch": 0.27021383227530904,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996069531265345,
"loss": 5.3553,
"mean_token_accuracy": 0.1729743644595146,
"num_tokens": 5954997.0,
"step": 3235
},
{
"entropy": 5.557475423812866,
"epoch": 0.2706314734380221,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996051922867653,
"loss": 5.398,
"mean_token_accuracy": 0.17867136895656585,
"num_tokens": 5963947.0,
"step": 3240
},
{
"entropy": 5.570569610595703,
"epoch": 0.27104911460073505,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996034275150118,
"loss": 5.4299,
"mean_token_accuracy": 0.1718892514705658,
"num_tokens": 5972077.0,
"step": 3245
},
{
"entropy": 5.550575113296508,
"epoch": 0.27146675576344803,
"grad_norm": 0.91796875,
"learning_rate": 0.000499601658811305,
"loss": 5.4417,
"mean_token_accuracy": 0.17238760441541673,
"num_tokens": 5982649.0,
"step": 3250
},
{
"entropy": 5.627004861831665,
"epoch": 0.27188439692616106,
"grad_norm": 0.96875,
"learning_rate": 0.0004995998861756758,
"loss": 5.456,
"mean_token_accuracy": 0.17346239238977432,
"num_tokens": 5991349.0,
"step": 3255
},
{
"entropy": 5.517546892166138,
"epoch": 0.27230203808887404,
"grad_norm": 0.953125,
"learning_rate": 0.0004995981096081552,
"loss": 5.4488,
"mean_token_accuracy": 0.1746513530611992,
"num_tokens": 6000555.0,
"step": 3260
},
{
"entropy": 5.787107801437378,
"epoch": 0.272719679251587,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995963291087744,
"loss": 5.5607,
"mean_token_accuracy": 0.1572660341858864,
"num_tokens": 6009554.0,
"step": 3265
},
{
"entropy": 5.683590507507324,
"epoch": 0.27313732041430006,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995945446775645,
"loss": 5.6408,
"mean_token_accuracy": 0.16543456614017488,
"num_tokens": 6019330.0,
"step": 3270
},
{
"entropy": 5.659921407699585,
"epoch": 0.27355496157701303,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995927563145566,
"loss": 5.595,
"mean_token_accuracy": 0.1674351915717125,
"num_tokens": 6028527.0,
"step": 3275
},
{
"entropy": 5.632570886611939,
"epoch": 0.273972602739726,
"grad_norm": 0.97265625,
"learning_rate": 0.0004995909640197825,
"loss": 5.4556,
"mean_token_accuracy": 0.17271459698677064,
"num_tokens": 6038223.0,
"step": 3280
},
{
"entropy": 5.521641635894776,
"epoch": 0.27439024390243905,
"grad_norm": 0.9375,
"learning_rate": 0.0004995891677932729,
"loss": 5.3925,
"mean_token_accuracy": 0.17719604074954987,
"num_tokens": 6047459.0,
"step": 3285
},
{
"entropy": 5.668535614013672,
"epoch": 0.274807885065152,
"grad_norm": 1.015625,
"learning_rate": 0.0004995873676350597,
"loss": 5.572,
"mean_token_accuracy": 0.1681835636496544,
"num_tokens": 6057322.0,
"step": 3290
},
{
"entropy": 5.71825475692749,
"epoch": 0.275225526227865,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995855635451743,
"loss": 5.6851,
"mean_token_accuracy": 0.16252846866846085,
"num_tokens": 6067520.0,
"step": 3295
},
{
"entropy": 5.674379491806031,
"epoch": 0.27564316739057804,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995837555236484,
"loss": 5.567,
"mean_token_accuracy": 0.1648716986179352,
"num_tokens": 6076729.0,
"step": 3300
},
{
"entropy": 5.628241968154907,
"epoch": 0.276060808553291,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995819435705133,
"loss": 5.5019,
"mean_token_accuracy": 0.16563906222581865,
"num_tokens": 6086529.0,
"step": 3305
},
{
"entropy": 5.659373760223389,
"epoch": 0.276478449716004,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995801276858011,
"loss": 5.5239,
"mean_token_accuracy": 0.1756095215678215,
"num_tokens": 6095784.0,
"step": 3310
},
{
"entropy": 5.733769321441651,
"epoch": 0.276896090878717,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995783078695434,
"loss": 5.5343,
"mean_token_accuracy": 0.17202824652194976,
"num_tokens": 6104626.0,
"step": 3315
},
{
"entropy": 5.577187585830688,
"epoch": 0.27731373204143,
"grad_norm": 0.99609375,
"learning_rate": 0.000499576484121772,
"loss": 5.4585,
"mean_token_accuracy": 0.17361778765916824,
"num_tokens": 6113758.0,
"step": 3320
},
{
"entropy": 5.681585025787354,
"epoch": 0.277731373204143,
"grad_norm": 1.0,
"learning_rate": 0.0004995746564425191,
"loss": 5.4673,
"mean_token_accuracy": 0.176327845454216,
"num_tokens": 6123265.0,
"step": 3325
},
{
"entropy": 5.67174882888794,
"epoch": 0.278149014366856,
"grad_norm": 0.953125,
"learning_rate": 0.0004995728248318163,
"loss": 5.543,
"mean_token_accuracy": 0.16984691321849824,
"num_tokens": 6132855.0,
"step": 3330
},
{
"entropy": 5.5259589672088625,
"epoch": 0.278566655529569,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995709892896961,
"loss": 5.4113,
"mean_token_accuracy": 0.17089553028345109,
"num_tokens": 6142739.0,
"step": 3335
},
{
"entropy": 5.549342250823974,
"epoch": 0.278984296692282,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995691498161903,
"loss": 5.3696,
"mean_token_accuracy": 0.1737748771905899,
"num_tokens": 6152169.0,
"step": 3340
},
{
"entropy": 5.475606870651245,
"epoch": 0.279401937854995,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995673064113313,
"loss": 5.401,
"mean_token_accuracy": 0.1780400887131691,
"num_tokens": 6160235.0,
"step": 3345
},
{
"entropy": 5.560038232803345,
"epoch": 0.279819579017708,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995654590751513,
"loss": 5.3862,
"mean_token_accuracy": 0.17502506971359252,
"num_tokens": 6169440.0,
"step": 3350
},
{
"entropy": 5.5738129138946535,
"epoch": 0.28023722018042097,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995636078076825,
"loss": 5.4137,
"mean_token_accuracy": 0.17904069274663925,
"num_tokens": 6178456.0,
"step": 3355
},
{
"entropy": 5.509097719192505,
"epoch": 0.280654861343134,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995617526089575,
"loss": 5.4846,
"mean_token_accuracy": 0.17317838966846466,
"num_tokens": 6186844.0,
"step": 3360
},
{
"entropy": 5.606524991989136,
"epoch": 0.281072502505847,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995598934790087,
"loss": 5.4069,
"mean_token_accuracy": 0.17825868874788284,
"num_tokens": 6196117.0,
"step": 3365
},
{
"entropy": 5.594628858566284,
"epoch": 0.28149014366855996,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995580304178687,
"loss": 5.3869,
"mean_token_accuracy": 0.18268515318632125,
"num_tokens": 6204636.0,
"step": 3370
},
{
"entropy": 5.516972541809082,
"epoch": 0.281907784831273,
"grad_norm": 0.9453125,
"learning_rate": 0.0004995561634255701,
"loss": 5.4196,
"mean_token_accuracy": 0.17650512456893921,
"num_tokens": 6214537.0,
"step": 3375
},
{
"entropy": 5.558368825912476,
"epoch": 0.28232542599398597,
"grad_norm": 0.9140625,
"learning_rate": 0.0004995542925021455,
"loss": 5.47,
"mean_token_accuracy": 0.1726568192243576,
"num_tokens": 6224753.0,
"step": 3380
},
{
"entropy": 5.807767534255982,
"epoch": 0.28274306715669895,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995524176476277,
"loss": 5.6607,
"mean_token_accuracy": 0.17105791866779327,
"num_tokens": 6233842.0,
"step": 3385
},
{
"entropy": 5.734399938583374,
"epoch": 0.283160708319412,
"grad_norm": 0.96875,
"learning_rate": 0.0004995505388620497,
"loss": 5.4957,
"mean_token_accuracy": 0.17600060850381852,
"num_tokens": 6243467.0,
"step": 3390
},
{
"entropy": 5.481904411315918,
"epoch": 0.28357834948212496,
"grad_norm": 0.98828125,
"learning_rate": 0.000499548656145444,
"loss": 5.5422,
"mean_token_accuracy": 0.1659504532814026,
"num_tokens": 6253263.0,
"step": 3395
},
{
"entropy": 5.611874055862427,
"epoch": 0.28399599064483794,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995467694978439,
"loss": 5.3838,
"mean_token_accuracy": 0.1848123013973236,
"num_tokens": 6262249.0,
"step": 3400
},
{
"entropy": 5.530205106735229,
"epoch": 0.28441363180755097,
"grad_norm": 1.046875,
"learning_rate": 0.0004995448789192823,
"loss": 5.393,
"mean_token_accuracy": 0.17520347088575364,
"num_tokens": 6271154.0,
"step": 3405
},
{
"entropy": 5.531033277511597,
"epoch": 0.28483127297026395,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995429844097922,
"loss": 5.4804,
"mean_token_accuracy": 0.17654867917299272,
"num_tokens": 6280377.0,
"step": 3410
},
{
"entropy": 5.6426879405975345,
"epoch": 0.2852489141329769,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995410859694071,
"loss": 5.4755,
"mean_token_accuracy": 0.17208488285541534,
"num_tokens": 6289941.0,
"step": 3415
},
{
"entropy": 5.630606269836425,
"epoch": 0.28566655529568996,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995391835981599,
"loss": 5.5237,
"mean_token_accuracy": 0.1680618032813072,
"num_tokens": 6299281.0,
"step": 3420
},
{
"entropy": 5.581368398666382,
"epoch": 0.28608419645840294,
"grad_norm": 0.99609375,
"learning_rate": 0.000499537277296084,
"loss": 5.4968,
"mean_token_accuracy": 0.1690795600414276,
"num_tokens": 6309408.0,
"step": 3425
},
{
"entropy": 5.590551519393921,
"epoch": 0.2865018376211159,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995353670632127,
"loss": 5.5126,
"mean_token_accuracy": 0.17588986158370973,
"num_tokens": 6318979.0,
"step": 3430
},
{
"entropy": 5.567683458328247,
"epoch": 0.28691947878382895,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995334528995797,
"loss": 5.3332,
"mean_token_accuracy": 0.1819864720106125,
"num_tokens": 6327616.0,
"step": 3435
},
{
"entropy": 5.509003210067749,
"epoch": 0.28733711994654193,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995315348052182,
"loss": 5.3676,
"mean_token_accuracy": 0.1833025187253952,
"num_tokens": 6336486.0,
"step": 3440
},
{
"entropy": 5.576023435592651,
"epoch": 0.2877547611092549,
"grad_norm": 1.0234375,
"learning_rate": 0.000499529612780162,
"loss": 5.4176,
"mean_token_accuracy": 0.17563057094812393,
"num_tokens": 6345671.0,
"step": 3445
},
{
"entropy": 5.653712558746338,
"epoch": 0.28817240227196794,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995276868244446,
"loss": 5.5246,
"mean_token_accuracy": 0.16986915171146394,
"num_tokens": 6354534.0,
"step": 3450
},
{
"entropy": 5.577525091171265,
"epoch": 0.2885900434346809,
"grad_norm": 0.87890625,
"learning_rate": 0.0004995257569380998,
"loss": 5.3662,
"mean_token_accuracy": 0.17813971638679504,
"num_tokens": 6364052.0,
"step": 3455
},
{
"entropy": 5.6135517120361325,
"epoch": 0.2890076845973939,
"grad_norm": 1.03125,
"learning_rate": 0.0004995238231211614,
"loss": 5.4969,
"mean_token_accuracy": 0.17180268615484237,
"num_tokens": 6373305.0,
"step": 3460
},
{
"entropy": 5.55221152305603,
"epoch": 0.28942532576010693,
"grad_norm": 0.88671875,
"learning_rate": 0.0004995218853736631,
"loss": 5.4182,
"mean_token_accuracy": 0.17001649588346482,
"num_tokens": 6382673.0,
"step": 3465
},
{
"entropy": 5.610888242721558,
"epoch": 0.2898429669228199,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995199436956391,
"loss": 5.5342,
"mean_token_accuracy": 0.16178884655237197,
"num_tokens": 6392122.0,
"step": 3470
},
{
"entropy": 5.645338344573974,
"epoch": 0.2902606080855329,
"grad_norm": 0.953125,
"learning_rate": 0.0004995179980871231,
"loss": 5.5279,
"mean_token_accuracy": 0.16972028613090515,
"num_tokens": 6401130.0,
"step": 3475
},
{
"entropy": 5.632619428634643,
"epoch": 0.2906782492482459,
"grad_norm": 1.03125,
"learning_rate": 0.0004995160485481493,
"loss": 5.4099,
"mean_token_accuracy": 0.16997333168983458,
"num_tokens": 6409471.0,
"step": 3480
},
{
"entropy": 5.522201681137085,
"epoch": 0.2910958904109589,
"grad_norm": 0.86328125,
"learning_rate": 0.000499514095078752,
"loss": 5.412,
"mean_token_accuracy": 0.17568757086992265,
"num_tokens": 6418902.0,
"step": 3485
},
{
"entropy": 5.629268074035645,
"epoch": 0.2915135315736719,
"grad_norm": 0.90234375,
"learning_rate": 0.000499512137678965,
"loss": 5.5067,
"mean_token_accuracy": 0.17458643913269042,
"num_tokens": 6428529.0,
"step": 3490
},
{
"entropy": 5.6039900302886965,
"epoch": 0.2919311727363849,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995101763488229,
"loss": 5.3625,
"mean_token_accuracy": 0.176267609000206,
"num_tokens": 6437460.0,
"step": 3495
},
{
"entropy": 5.541733264923096,
"epoch": 0.2923488138990979,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995082110883599,
"loss": 5.4092,
"mean_token_accuracy": 0.17792665511369704,
"num_tokens": 6446005.0,
"step": 3500
},
{
"entropy": 5.528889799118042,
"epoch": 0.29276645506181087,
"grad_norm": 0.875,
"learning_rate": 0.0004995062418976104,
"loss": 5.4607,
"mean_token_accuracy": 0.1745197594165802,
"num_tokens": 6455241.0,
"step": 3505
},
{
"entropy": 5.619756650924683,
"epoch": 0.2931840962245239,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995042687766089,
"loss": 5.4288,
"mean_token_accuracy": 0.17538372576236724,
"num_tokens": 6464748.0,
"step": 3510
},
{
"entropy": 5.5280955791473385,
"epoch": 0.2936017373872369,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995022917253899,
"loss": 5.4933,
"mean_token_accuracy": 0.18021770864725112,
"num_tokens": 6473834.0,
"step": 3515
},
{
"entropy": 5.5699669361114506,
"epoch": 0.29401937854994986,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995003107439882,
"loss": 5.4346,
"mean_token_accuracy": 0.17709693014621736,
"num_tokens": 6482169.0,
"step": 3520
},
{
"entropy": 5.53997220993042,
"epoch": 0.2944370197126629,
"grad_norm": 0.875,
"learning_rate": 0.0004994983258324383,
"loss": 5.3787,
"mean_token_accuracy": 0.17743246406316757,
"num_tokens": 6491798.0,
"step": 3525
},
{
"entropy": 5.479582118988037,
"epoch": 0.2948546608753759,
"grad_norm": 1.03125,
"learning_rate": 0.0004994963369907748,
"loss": 5.3236,
"mean_token_accuracy": 0.17699272483587264,
"num_tokens": 6500143.0,
"step": 3530
},
{
"entropy": 5.462062168121338,
"epoch": 0.29527230203808885,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994943442190328,
"loss": 5.4531,
"mean_token_accuracy": 0.17053767442703247,
"num_tokens": 6509745.0,
"step": 3535
},
{
"entropy": 5.67592306137085,
"epoch": 0.2956899432008019,
"grad_norm": 0.84375,
"learning_rate": 0.0004994923475172471,
"loss": 5.5242,
"mean_token_accuracy": 0.17168646901845933,
"num_tokens": 6520931.0,
"step": 3540
},
{
"entropy": 5.62648639678955,
"epoch": 0.29610758436351486,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994903468854527,
"loss": 5.4493,
"mean_token_accuracy": 0.17539601027965546,
"num_tokens": 6529774.0,
"step": 3545
},
{
"entropy": 5.461532068252564,
"epoch": 0.29652522552622784,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994883423236843,
"loss": 5.3806,
"mean_token_accuracy": 0.17293752282857894,
"num_tokens": 6538922.0,
"step": 3550
},
{
"entropy": 5.513239145278931,
"epoch": 0.2969428666889409,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994863338319775,
"loss": 5.4286,
"mean_token_accuracy": 0.17735143154859542,
"num_tokens": 6548891.0,
"step": 3555
},
{
"entropy": 5.494573307037354,
"epoch": 0.29736050785165385,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994843214103672,
"loss": 5.2551,
"mean_token_accuracy": 0.1761169984936714,
"num_tokens": 6557025.0,
"step": 3560
},
{
"entropy": 5.584866857528686,
"epoch": 0.29777814901436683,
"grad_norm": 0.95703125,
"learning_rate": 0.0004994823050588886,
"loss": 5.4271,
"mean_token_accuracy": 0.17514524310827256,
"num_tokens": 6567037.0,
"step": 3565
},
{
"entropy": 5.555587720870972,
"epoch": 0.29819579017707987,
"grad_norm": 0.97265625,
"learning_rate": 0.000499480284777577,
"loss": 5.4629,
"mean_token_accuracy": 0.17471245378255845,
"num_tokens": 6576766.0,
"step": 3570
},
{
"entropy": 5.570872449874878,
"epoch": 0.29861343133979285,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994782605664677,
"loss": 5.4362,
"mean_token_accuracy": 0.17735102772712708,
"num_tokens": 6585576.0,
"step": 3575
},
{
"entropy": 5.596178579330444,
"epoch": 0.2990310725025058,
"grad_norm": 1.078125,
"learning_rate": 0.0004994762324255965,
"loss": 5.4857,
"mean_token_accuracy": 0.17427317202091216,
"num_tokens": 6594463.0,
"step": 3580
},
{
"entropy": 5.566938352584839,
"epoch": 0.29944871366521886,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994742003549986,
"loss": 5.3677,
"mean_token_accuracy": 0.17447261214256288,
"num_tokens": 6603651.0,
"step": 3585
},
{
"entropy": 5.614844465255738,
"epoch": 0.29986635482793184,
"grad_norm": 0.90625,
"learning_rate": 0.0004994721643547095,
"loss": 5.4392,
"mean_token_accuracy": 0.1780633106827736,
"num_tokens": 6612950.0,
"step": 3590
},
{
"entropy": 5.601415967941284,
"epoch": 0.3002839959906448,
"grad_norm": 0.9453125,
"learning_rate": 0.000499470124424765,
"loss": 5.3823,
"mean_token_accuracy": 0.17879820615053177,
"num_tokens": 6622013.0,
"step": 3595
},
{
"entropy": 5.552940416336059,
"epoch": 0.30070163715335785,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994680805652008,
"loss": 5.4592,
"mean_token_accuracy": 0.1763960376381874,
"num_tokens": 6632046.0,
"step": 3600
},
{
"entropy": 5.561949300765991,
"epoch": 0.3011192783160708,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994660327760527,
"loss": 5.5006,
"mean_token_accuracy": 0.17833555936813356,
"num_tokens": 6641169.0,
"step": 3605
},
{
"entropy": 5.555380964279175,
"epoch": 0.3015369194787838,
"grad_norm": 0.953125,
"learning_rate": 0.0004994639810573565,
"loss": 5.3107,
"mean_token_accuracy": 0.18665722012519836,
"num_tokens": 6649663.0,
"step": 3610
},
{
"entropy": 5.624287843704224,
"epoch": 0.30195456064149684,
"grad_norm": 1.03125,
"learning_rate": 0.000499461925409148,
"loss": 5.5101,
"mean_token_accuracy": 0.17178054600954057,
"num_tokens": 6659448.0,
"step": 3615
},
{
"entropy": 5.6389318943023685,
"epoch": 0.3023722018042098,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994598658314634,
"loss": 5.4612,
"mean_token_accuracy": 0.1729240134358406,
"num_tokens": 6670314.0,
"step": 3620
},
{
"entropy": 5.569744968414307,
"epoch": 0.3027898429669228,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994578023243386,
"loss": 5.4028,
"mean_token_accuracy": 0.1818447932600975,
"num_tokens": 6679413.0,
"step": 3625
},
{
"entropy": 5.564968490600586,
"epoch": 0.30320748412963583,
"grad_norm": 0.9921875,
"learning_rate": 0.0004994557348878098,
"loss": 5.3927,
"mean_token_accuracy": 0.1806464374065399,
"num_tokens": 6688439.0,
"step": 3630
},
{
"entropy": 5.524420499801636,
"epoch": 0.3036251252923488,
"grad_norm": 0.9296875,
"learning_rate": 0.0004994536635219133,
"loss": 5.3662,
"mean_token_accuracy": 0.1851886883378029,
"num_tokens": 6697471.0,
"step": 3635
},
{
"entropy": 5.583075618743896,
"epoch": 0.3040427664550618,
"grad_norm": 0.984375,
"learning_rate": 0.0004994515882266852,
"loss": 5.4185,
"mean_token_accuracy": 0.17669300734996796,
"num_tokens": 6706698.0,
"step": 3640
},
{
"entropy": 5.51670651435852,
"epoch": 0.3044604076177748,
"grad_norm": 1.03125,
"learning_rate": 0.0004994495090021618,
"loss": 5.3535,
"mean_token_accuracy": 0.18122815787792207,
"num_tokens": 6715676.0,
"step": 3645
},
{
"entropy": 5.459908151626587,
"epoch": 0.3048780487804878,
"grad_norm": 1.0,
"learning_rate": 0.0004994474258483796,
"loss": 5.3835,
"mean_token_accuracy": 0.18140835911035538,
"num_tokens": 6723818.0,
"step": 3650
},
{
"entropy": 5.5700725555419925,
"epoch": 0.3052956899432008,
"grad_norm": 1.015625,
"learning_rate": 0.0004994453387653751,
"loss": 5.4033,
"mean_token_accuracy": 0.17778776735067367,
"num_tokens": 6733450.0,
"step": 3655
},
{
"entropy": 5.564090013504028,
"epoch": 0.3057133311059138,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994432477531847,
"loss": 5.3523,
"mean_token_accuracy": 0.1764783501625061,
"num_tokens": 6743142.0,
"step": 3660
},
{
"entropy": 5.472638511657715,
"epoch": 0.3061309722686268,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994411528118452,
"loss": 5.3028,
"mean_token_accuracy": 0.17945352643728257,
"num_tokens": 6752185.0,
"step": 3665
},
{
"entropy": 5.502836799621582,
"epoch": 0.30654861343133977,
"grad_norm": 0.9375,
"learning_rate": 0.0004994390539413931,
"loss": 5.4086,
"mean_token_accuracy": 0.1835372507572174,
"num_tokens": 6761549.0,
"step": 3670
},
{
"entropy": 5.536188411712646,
"epoch": 0.3069662545940528,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994369511418652,
"loss": 5.4247,
"mean_token_accuracy": 0.17736767381429672,
"num_tokens": 6770395.0,
"step": 3675
},
{
"entropy": 5.645047044754028,
"epoch": 0.3073838957567658,
"grad_norm": 1.03125,
"learning_rate": 0.0004994348444132983,
"loss": 5.4952,
"mean_token_accuracy": 0.17320885211229325,
"num_tokens": 6779154.0,
"step": 3680
},
{
"entropy": 5.490355920791626,
"epoch": 0.30780153691947876,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994327337557293,
"loss": 5.3789,
"mean_token_accuracy": 0.1852697342634201,
"num_tokens": 6789263.0,
"step": 3685
},
{
"entropy": 5.555096244812011,
"epoch": 0.3082191780821918,
"grad_norm": 0.96875,
"learning_rate": 0.0004994306191691952,
"loss": 5.4717,
"mean_token_accuracy": 0.17230480462312697,
"num_tokens": 6797985.0,
"step": 3690
},
{
"entropy": 5.54058609008789,
"epoch": 0.30863681924490477,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994285006537329,
"loss": 5.3673,
"mean_token_accuracy": 0.17717754542827607,
"num_tokens": 6807566.0,
"step": 3695
},
{
"entropy": 5.4808722019195555,
"epoch": 0.30905446040761775,
"grad_norm": 0.9921875,
"learning_rate": 0.0004994263782093796,
"loss": 5.3947,
"mean_token_accuracy": 0.17707190960645675,
"num_tokens": 6815637.0,
"step": 3700
},
{
"entropy": 5.525970315933227,
"epoch": 0.3094721015703308,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994242518361723,
"loss": 5.2836,
"mean_token_accuracy": 0.18263801634311677,
"num_tokens": 6823721.0,
"step": 3705
},
{
"entropy": 5.606603193283081,
"epoch": 0.30988974273304376,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994221215341484,
"loss": 5.458,
"mean_token_accuracy": 0.17419039607048034,
"num_tokens": 6833437.0,
"step": 3710
},
{
"entropy": 5.537316417694091,
"epoch": 0.31030738389575674,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994199873033452,
"loss": 5.5149,
"mean_token_accuracy": 0.17269643843173982,
"num_tokens": 6842287.0,
"step": 3715
},
{
"entropy": 5.642424583435059,
"epoch": 0.3107250250584698,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994178491437999,
"loss": 5.3587,
"mean_token_accuracy": 0.17185903638601302,
"num_tokens": 6851520.0,
"step": 3720
},
{
"entropy": 5.519375514984131,
"epoch": 0.31114266622118275,
"grad_norm": 0.9140625,
"learning_rate": 0.0004994157070555501,
"loss": 5.4082,
"mean_token_accuracy": 0.17950767874717713,
"num_tokens": 6861219.0,
"step": 3725
},
{
"entropy": 5.587158298492431,
"epoch": 0.31156030738389573,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994135610386331,
"loss": 5.5369,
"mean_token_accuracy": 0.1727735862135887,
"num_tokens": 6870525.0,
"step": 3730
},
{
"entropy": 5.62889609336853,
"epoch": 0.31197794854660876,
"grad_norm": 0.875,
"learning_rate": 0.0004994114110930867,
"loss": 5.4523,
"mean_token_accuracy": 0.17412902265787125,
"num_tokens": 6880183.0,
"step": 3735
},
{
"entropy": 5.539847660064697,
"epoch": 0.31239558970932174,
"grad_norm": 0.91015625,
"learning_rate": 0.0004994092572189483,
"loss": 5.5036,
"mean_token_accuracy": 0.1694844126701355,
"num_tokens": 6890442.0,
"step": 3740
},
{
"entropy": 5.562208604812622,
"epoch": 0.3128132308720348,
"grad_norm": 1.015625,
"learning_rate": 0.0004994070994162558,
"loss": 5.4078,
"mean_token_accuracy": 0.17751166224479675,
"num_tokens": 6900091.0,
"step": 3745
},
{
"entropy": 5.536312770843506,
"epoch": 0.31323087203474775,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994049376850471,
"loss": 5.4316,
"mean_token_accuracy": 0.17668322771787642,
"num_tokens": 6910489.0,
"step": 3750
},
{
"entropy": 5.551783609390259,
"epoch": 0.31364851319746073,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994027720253596,
"loss": 5.4223,
"mean_token_accuracy": 0.17174892127513885,
"num_tokens": 6919895.0,
"step": 3755
},
{
"entropy": 5.688577699661255,
"epoch": 0.31406615436017377,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994006024372315,
"loss": 5.5632,
"mean_token_accuracy": 0.16963823288679122,
"num_tokens": 6929511.0,
"step": 3760
},
{
"entropy": 5.503642606735229,
"epoch": 0.31448379552288674,
"grad_norm": 0.9921875,
"learning_rate": 0.0004993984289207007,
"loss": 5.3159,
"mean_token_accuracy": 0.18278757035732268,
"num_tokens": 6938059.0,
"step": 3765
},
{
"entropy": 5.518787288665772,
"epoch": 0.3149014366855997,
"grad_norm": 0.953125,
"learning_rate": 0.0004993962514758054,
"loss": 5.3483,
"mean_token_accuracy": 0.17641351968050004,
"num_tokens": 6946894.0,
"step": 3770
},
{
"entropy": 5.626014232635498,
"epoch": 0.31531907784831276,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993940701025836,
"loss": 5.3605,
"mean_token_accuracy": 0.16906849294900894,
"num_tokens": 6955494.0,
"step": 3775
},
{
"entropy": 5.480310583114624,
"epoch": 0.31573671901102573,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993918848010734,
"loss": 5.4729,
"mean_token_accuracy": 0.1719732776284218,
"num_tokens": 6964347.0,
"step": 3780
},
{
"entropy": 5.579901552200317,
"epoch": 0.3161543601737387,
"grad_norm": 0.9765625,
"learning_rate": 0.0004993896955713133,
"loss": 5.4279,
"mean_token_accuracy": 0.17163396626710892,
"num_tokens": 6974750.0,
"step": 3785
},
{
"entropy": 5.583578824996948,
"epoch": 0.31657200133645175,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993875024133414,
"loss": 5.4291,
"mean_token_accuracy": 0.17758399695158006,
"num_tokens": 6984228.0,
"step": 3790
},
{
"entropy": 5.4850465774536135,
"epoch": 0.3169896424991647,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993853053271962,
"loss": 5.4105,
"mean_token_accuracy": 0.17856047451496124,
"num_tokens": 6993173.0,
"step": 3795
},
{
"entropy": 5.710829591751098,
"epoch": 0.3174072836618777,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993831043129161,
"loss": 5.6112,
"mean_token_accuracy": 0.16447764188051223,
"num_tokens": 7003486.0,
"step": 3800
},
{
"entropy": 5.51308650970459,
"epoch": 0.31782492482459074,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993808993705397,
"loss": 5.3517,
"mean_token_accuracy": 0.18299903422594072,
"num_tokens": 7011985.0,
"step": 3805
},
{
"entropy": 5.504875659942627,
"epoch": 0.3182425659873037,
"grad_norm": 0.9921875,
"learning_rate": 0.0004993786905001055,
"loss": 5.3119,
"mean_token_accuracy": 0.18059561103582383,
"num_tokens": 7021148.0,
"step": 3810
},
{
"entropy": 5.495862054824829,
"epoch": 0.3186602071500167,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993764777016524,
"loss": 5.3649,
"mean_token_accuracy": 0.17798255681991576,
"num_tokens": 7030110.0,
"step": 3815
},
{
"entropy": 5.584918355941772,
"epoch": 0.31907784831272973,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993742609752188,
"loss": 5.4413,
"mean_token_accuracy": 0.18019713163375856,
"num_tokens": 7039428.0,
"step": 3820
},
{
"entropy": 5.676028394699097,
"epoch": 0.3194954894754427,
"grad_norm": 0.93359375,
"learning_rate": 0.0004993720403208438,
"loss": 5.5038,
"mean_token_accuracy": 0.17248591482639314,
"num_tokens": 7048703.0,
"step": 3825
},
{
"entropy": 5.5341957092285154,
"epoch": 0.3199131306381557,
"grad_norm": 1.0078125,
"learning_rate": 0.000499369815738566,
"loss": 5.4036,
"mean_token_accuracy": 0.1700318157672882,
"num_tokens": 7057612.0,
"step": 3830
},
{
"entropy": 5.555621862411499,
"epoch": 0.3203307718008687,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993675872284248,
"loss": 5.3607,
"mean_token_accuracy": 0.1777009814977646,
"num_tokens": 7066660.0,
"step": 3835
},
{
"entropy": 5.485004472732544,
"epoch": 0.3207484129635817,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993653547904586,
"loss": 5.3339,
"mean_token_accuracy": 0.17354484349489213,
"num_tokens": 7075368.0,
"step": 3840
},
{
"entropy": 5.51584267616272,
"epoch": 0.3211660541262947,
"grad_norm": 0.9140625,
"learning_rate": 0.000499363118424707,
"loss": 5.4376,
"mean_token_accuracy": 0.18095694333314896,
"num_tokens": 7085276.0,
"step": 3845
},
{
"entropy": 5.539682483673095,
"epoch": 0.3215836952890077,
"grad_norm": 0.9765625,
"learning_rate": 0.0004993608781312088,
"loss": 5.3994,
"mean_token_accuracy": 0.1757436066865921,
"num_tokens": 7094292.0,
"step": 3850
},
{
"entropy": 5.664052057266235,
"epoch": 0.3220013364517207,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993586339100035,
"loss": 5.6208,
"mean_token_accuracy": 0.16767174154520034,
"num_tokens": 7104712.0,
"step": 3855
},
{
"entropy": 5.6224799156188965,
"epoch": 0.32241897761443367,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993563857611301,
"loss": 5.4174,
"mean_token_accuracy": 0.1841977894306183,
"num_tokens": 7113297.0,
"step": 3860
},
{
"entropy": 5.480200672149659,
"epoch": 0.3228366187771467,
"grad_norm": 0.8984375,
"learning_rate": 0.0004993541336846283,
"loss": 5.4335,
"mean_token_accuracy": 0.16748614832758904,
"num_tokens": 7122706.0,
"step": 3865
},
{
"entropy": 5.542711400985718,
"epoch": 0.3232542599398597,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993518776805372,
"loss": 5.3907,
"mean_token_accuracy": 0.17922770231962204,
"num_tokens": 7131939.0,
"step": 3870
},
{
"entropy": 5.451107168197632,
"epoch": 0.32367190110257266,
"grad_norm": 1.0,
"learning_rate": 0.0004993496177488965,
"loss": 5.2662,
"mean_token_accuracy": 0.18536351174116134,
"num_tokens": 7141016.0,
"step": 3875
},
{
"entropy": 5.463331985473633,
"epoch": 0.3240895422652857,
"grad_norm": 0.94921875,
"learning_rate": 0.0004993473538897457,
"loss": 5.2199,
"mean_token_accuracy": 0.1908682778477669,
"num_tokens": 7149177.0,
"step": 3880
},
{
"entropy": 5.517989921569824,
"epoch": 0.32450718342799867,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993450861031243,
"loss": 5.3559,
"mean_token_accuracy": 0.1834622472524643,
"num_tokens": 7157839.0,
"step": 3885
},
{
"entropy": 5.61125168800354,
"epoch": 0.32492482459071165,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993428143890722,
"loss": 5.4994,
"mean_token_accuracy": 0.17009637653827667,
"num_tokens": 7167759.0,
"step": 3890
},
{
"entropy": 5.6287397861480715,
"epoch": 0.3253424657534247,
"grad_norm": 1.046875,
"learning_rate": 0.0004993405387476291,
"loss": 5.4309,
"mean_token_accuracy": 0.16932940483093262,
"num_tokens": 7177141.0,
"step": 3895
},
{
"entropy": 5.4610847473144535,
"epoch": 0.32576010691613766,
"grad_norm": 1.046875,
"learning_rate": 0.0004993382591788349,
"loss": 5.3484,
"mean_token_accuracy": 0.18480813056230544,
"num_tokens": 7185350.0,
"step": 3900
},
{
"entropy": 5.475749492645264,
"epoch": 0.32617774807885064,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993359756827293,
"loss": 5.2774,
"mean_token_accuracy": 0.18510581254959108,
"num_tokens": 7194073.0,
"step": 3905
},
{
"entropy": 5.570122957229614,
"epoch": 0.32659538924156367,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993336882593524,
"loss": 5.4658,
"mean_token_accuracy": 0.17096367180347444,
"num_tokens": 7203421.0,
"step": 3910
},
{
"entropy": 5.504409551620483,
"epoch": 0.32701303040427665,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993313969087444,
"loss": 5.2325,
"mean_token_accuracy": 0.18367886990308763,
"num_tokens": 7212548.0,
"step": 3915
},
{
"entropy": 5.52297306060791,
"epoch": 0.32743067156698963,
"grad_norm": 1.03125,
"learning_rate": 0.0004993291016309452,
"loss": 5.4708,
"mean_token_accuracy": 0.17318004965782166,
"num_tokens": 7221668.0,
"step": 3920
},
{
"entropy": 5.530381393432617,
"epoch": 0.32784831272970266,
"grad_norm": 0.9609375,
"learning_rate": 0.000499326802425995,
"loss": 5.3636,
"mean_token_accuracy": 0.1767316937446594,
"num_tokens": 7231775.0,
"step": 3925
},
{
"entropy": 5.599203157424927,
"epoch": 0.32826595389241564,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993244992939342,
"loss": 5.4185,
"mean_token_accuracy": 0.1742442861199379,
"num_tokens": 7241264.0,
"step": 3930
},
{
"entropy": 5.563284158706665,
"epoch": 0.3286835950551286,
"grad_norm": 0.9375,
"learning_rate": 0.000499322192234803,
"loss": 5.4255,
"mean_token_accuracy": 0.17522766590118408,
"num_tokens": 7250519.0,
"step": 3935
},
{
"entropy": 5.653376913070678,
"epoch": 0.32910123621784165,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993198812486418,
"loss": 5.4749,
"mean_token_accuracy": 0.17463890612125396,
"num_tokens": 7259533.0,
"step": 3940
},
{
"entropy": 5.504630088806152,
"epoch": 0.32951887738055463,
"grad_norm": 0.91796875,
"learning_rate": 0.000499317566335491,
"loss": 5.3,
"mean_token_accuracy": 0.17634646892547606,
"num_tokens": 7268980.0,
"step": 3945
},
{
"entropy": 5.512438726425171,
"epoch": 0.3299365185432676,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993152474953913,
"loss": 5.4041,
"mean_token_accuracy": 0.18160206824541092,
"num_tokens": 7277126.0,
"step": 3950
},
{
"entropy": 5.56659083366394,
"epoch": 0.33035415970598064,
"grad_norm": 0.9765625,
"learning_rate": 0.0004993129247283831,
"loss": 5.3623,
"mean_token_accuracy": 0.18333873450756072,
"num_tokens": 7286223.0,
"step": 3955
},
{
"entropy": 5.548340320587158,
"epoch": 0.3307718008686936,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993105980345071,
"loss": 5.3752,
"mean_token_accuracy": 0.1801811933517456,
"num_tokens": 7295054.0,
"step": 3960
},
{
"entropy": 5.519936227798462,
"epoch": 0.3311894420314066,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993082674138042,
"loss": 5.4018,
"mean_token_accuracy": 0.17907965779304505,
"num_tokens": 7304502.0,
"step": 3965
},
{
"entropy": 5.548730421066284,
"epoch": 0.33160708319411963,
"grad_norm": 0.87109375,
"learning_rate": 0.000499305932866315,
"loss": 5.3661,
"mean_token_accuracy": 0.17640268206596374,
"num_tokens": 7314093.0,
"step": 3970
},
{
"entropy": 5.584903287887573,
"epoch": 0.3320247243568326,
"grad_norm": 1.0,
"learning_rate": 0.0004993035943920805,
"loss": 5.3959,
"mean_token_accuracy": 0.17537485361099242,
"num_tokens": 7323059.0,
"step": 3975
},
{
"entropy": 5.633094596862793,
"epoch": 0.3324423655195456,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993012519911416,
"loss": 5.5284,
"mean_token_accuracy": 0.17066334784030915,
"num_tokens": 7332596.0,
"step": 3980
},
{
"entropy": 5.542002010345459,
"epoch": 0.3328600066822586,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992989056635392,
"loss": 5.3302,
"mean_token_accuracy": 0.1828303039073944,
"num_tokens": 7341671.0,
"step": 3985
},
{
"entropy": 5.49449462890625,
"epoch": 0.3332776478449716,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992965554093146,
"loss": 5.4482,
"mean_token_accuracy": 0.17449371218681337,
"num_tokens": 7350344.0,
"step": 3990
},
{
"entropy": 5.5837198257446286,
"epoch": 0.3336952890076846,
"grad_norm": 0.953125,
"learning_rate": 0.0004992942012285086,
"loss": 5.4578,
"mean_token_accuracy": 0.1801720902323723,
"num_tokens": 7359844.0,
"step": 3995
},
{
"entropy": 5.56695442199707,
"epoch": 0.3341129301703976,
"grad_norm": 1.015625,
"learning_rate": 0.0004992918431211629,
"loss": 5.3852,
"mean_token_accuracy": 0.17716051787137985,
"num_tokens": 7369377.0,
"step": 4000
},
{
"entropy": 5.542109727859497,
"epoch": 0.3345305713331106,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992894810873183,
"loss": 5.3539,
"mean_token_accuracy": 0.17548734098672866,
"num_tokens": 7378814.0,
"step": 4005
},
{
"entropy": 5.4964337825775145,
"epoch": 0.33494821249582357,
"grad_norm": 1.0234375,
"learning_rate": 0.0004992871151270165,
"loss": 5.3576,
"mean_token_accuracy": 0.18033732622861862,
"num_tokens": 7387333.0,
"step": 4010
},
{
"entropy": 5.4655951976776125,
"epoch": 0.3353658536585366,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992847452402986,
"loss": 5.3957,
"mean_token_accuracy": 0.180501726269722,
"num_tokens": 7396895.0,
"step": 4015
},
{
"entropy": 5.518776273727417,
"epoch": 0.3357834948212496,
"grad_norm": 1.0234375,
"learning_rate": 0.0004992823714272064,
"loss": 5.3837,
"mean_token_accuracy": 0.17895213067531585,
"num_tokens": 7406430.0,
"step": 4020
},
{
"entropy": 5.536636161804199,
"epoch": 0.33620113598396256,
"grad_norm": 0.953125,
"learning_rate": 0.0004992799936877813,
"loss": 5.2343,
"mean_token_accuracy": 0.18902939856052398,
"num_tokens": 7415224.0,
"step": 4025
},
{
"entropy": 5.588719654083252,
"epoch": 0.3366187771466756,
"grad_norm": 1.03125,
"learning_rate": 0.0004992776120220649,
"loss": 5.5551,
"mean_token_accuracy": 0.16299639493227006,
"num_tokens": 7424250.0,
"step": 4030
},
{
"entropy": 5.586339092254638,
"epoch": 0.3370364183093886,
"grad_norm": 1.03125,
"learning_rate": 0.0004992752264300989,
"loss": 5.3824,
"mean_token_accuracy": 0.18631240874528884,
"num_tokens": 7433579.0,
"step": 4035
},
{
"entropy": 5.4758015155792235,
"epoch": 0.33745405947210155,
"grad_norm": 1.0625,
"learning_rate": 0.0004992728369119252,
"loss": 5.3442,
"mean_token_accuracy": 0.18162791579961776,
"num_tokens": 7442621.0,
"step": 4040
},
{
"entropy": 5.571116495132446,
"epoch": 0.3378717006348146,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992704434675855,
"loss": 5.4374,
"mean_token_accuracy": 0.17829461991786957,
"num_tokens": 7452077.0,
"step": 4045
},
{
"entropy": 5.502327346801758,
"epoch": 0.33828934179752757,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992680460971218,
"loss": 5.3601,
"mean_token_accuracy": 0.17589134722948074,
"num_tokens": 7461146.0,
"step": 4050
},
{
"entropy": 5.419596338272095,
"epoch": 0.33870698296024054,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992656448005758,
"loss": 5.3586,
"mean_token_accuracy": 0.1813579112291336,
"num_tokens": 7470335.0,
"step": 4055
},
{
"entropy": 5.539237070083618,
"epoch": 0.3391246241229536,
"grad_norm": 0.9609375,
"learning_rate": 0.0004992632395779899,
"loss": 5.3328,
"mean_token_accuracy": 0.17426132410764694,
"num_tokens": 7479458.0,
"step": 4060
},
{
"entropy": 5.450229549407959,
"epoch": 0.33954226528566656,
"grad_norm": 0.96875,
"learning_rate": 0.0004992608304294061,
"loss": 5.3035,
"mean_token_accuracy": 0.19210376739501953,
"num_tokens": 7488005.0,
"step": 4065
},
{
"entropy": 5.420096445083618,
"epoch": 0.33995990644837953,
"grad_norm": 1.0859375,
"learning_rate": 0.0004992584173548664,
"loss": 5.307,
"mean_token_accuracy": 0.17884953767061235,
"num_tokens": 7496807.0,
"step": 4070
},
{
"entropy": 5.509367942810059,
"epoch": 0.34037754761109257,
"grad_norm": 0.90625,
"learning_rate": 0.0004992560003544131,
"loss": 5.3321,
"mean_token_accuracy": 0.18723736256361007,
"num_tokens": 7505383.0,
"step": 4075
},
{
"entropy": 5.572503232955933,
"epoch": 0.34079518877380555,
"grad_norm": 0.96484375,
"learning_rate": 0.0004992535794280887,
"loss": 5.403,
"mean_token_accuracy": 0.18027678579092027,
"num_tokens": 7515093.0,
"step": 4080
},
{
"entropy": 5.565121698379516,
"epoch": 0.3412128299365185,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992511545759354,
"loss": 5.3847,
"mean_token_accuracy": 0.17668999135494232,
"num_tokens": 7524067.0,
"step": 4085
},
{
"entropy": 5.432730484008789,
"epoch": 0.34163047109923156,
"grad_norm": 0.93359375,
"learning_rate": 0.0004992487257979958,
"loss": 5.3256,
"mean_token_accuracy": 0.17488196790218352,
"num_tokens": 7533197.0,
"step": 4090
},
{
"entropy": 5.5197220802307125,
"epoch": 0.34204811226194454,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992462930943123,
"loss": 5.3318,
"mean_token_accuracy": 0.17547912299633026,
"num_tokens": 7543030.0,
"step": 4095
},
{
"entropy": 5.533311986923218,
"epoch": 0.3424657534246575,
"grad_norm": 0.9375,
"learning_rate": 0.0004992438564649275,
"loss": 5.3787,
"mean_token_accuracy": 0.17519449293613434,
"num_tokens": 7551556.0,
"step": 4100
},
{
"entropy": 5.507522869110107,
"epoch": 0.34288339458737055,
"grad_norm": 1.0,
"learning_rate": 0.0004992414159098841,
"loss": 5.3129,
"mean_token_accuracy": 0.18167788684368133,
"num_tokens": 7559284.0,
"step": 4105
},
{
"entropy": 5.598392963409424,
"epoch": 0.3433010357500835,
"grad_norm": 0.9375,
"learning_rate": 0.0004992389714292247,
"loss": 5.4277,
"mean_token_accuracy": 0.17684440910816193,
"num_tokens": 7568750.0,
"step": 4110
},
{
"entropy": 5.456853771209717,
"epoch": 0.3437186769127965,
"grad_norm": 0.9609375,
"learning_rate": 0.0004992365230229922,
"loss": 5.2972,
"mean_token_accuracy": 0.18591678887605667,
"num_tokens": 7578309.0,
"step": 4115
},
{
"entropy": 5.43913836479187,
"epoch": 0.34413631807550954,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992340706912295,
"loss": 5.305,
"mean_token_accuracy": 0.1826748341321945,
"num_tokens": 7586760.0,
"step": 4120
},
{
"entropy": 5.465929222106934,
"epoch": 0.3445539592382225,
"grad_norm": 0.96484375,
"learning_rate": 0.0004992316144339794,
"loss": 5.2855,
"mean_token_accuracy": 0.18348537087440492,
"num_tokens": 7595472.0,
"step": 4125
},
{
"entropy": 5.519556903839112,
"epoch": 0.3449716004009355,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992291542512852,
"loss": 5.4731,
"mean_token_accuracy": 0.16670606136322022,
"num_tokens": 7605663.0,
"step": 4130
},
{
"entropy": 5.583035278320312,
"epoch": 0.34538924156364853,
"grad_norm": 0.8359375,
"learning_rate": 0.0004992266901431895,
"loss": 5.3647,
"mean_token_accuracy": 0.17722720950841903,
"num_tokens": 7616110.0,
"step": 4135
},
{
"entropy": 5.491539525985718,
"epoch": 0.3458068827263615,
"grad_norm": 0.9375,
"learning_rate": 0.0004992242221097358,
"loss": 5.3302,
"mean_token_accuracy": 0.17768455743789674,
"num_tokens": 7625311.0,
"step": 4140
},
{
"entropy": 5.505361318588257,
"epoch": 0.3462245238890745,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992217501509672,
"loss": 5.3432,
"mean_token_accuracy": 0.18014439195394516,
"num_tokens": 7634562.0,
"step": 4145
},
{
"entropy": 5.481019639968872,
"epoch": 0.3466421650517875,
"grad_norm": 0.9296875,
"learning_rate": 0.000499219274266927,
"loss": 5.2885,
"mean_token_accuracy": 0.1800134226679802,
"num_tokens": 7643231.0,
"step": 4150
},
{
"entropy": 5.529042339324951,
"epoch": 0.3470598062145005,
"grad_norm": 0.97265625,
"learning_rate": 0.0004992167944576584,
"loss": 5.4271,
"mean_token_accuracy": 0.1818411871790886,
"num_tokens": 7653556.0,
"step": 4155
},
{
"entropy": 5.492344093322754,
"epoch": 0.3474774473772135,
"grad_norm": 0.9765625,
"learning_rate": 0.000499214310723205,
"loss": 5.326,
"mean_token_accuracy": 0.18246453553438186,
"num_tokens": 7662295.0,
"step": 4160
},
{
"entropy": 5.448925304412842,
"epoch": 0.3478950885399265,
"grad_norm": 1.0,
"learning_rate": 0.0004992118230636102,
"loss": 5.3571,
"mean_token_accuracy": 0.17766649574041365,
"num_tokens": 7671429.0,
"step": 4165
},
{
"entropy": 5.467991352081299,
"epoch": 0.3483127297026395,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992093314789176,
"loss": 5.2605,
"mean_token_accuracy": 0.17754218429327012,
"num_tokens": 7680589.0,
"step": 4170
},
{
"entropy": 5.481946563720703,
"epoch": 0.34873037086535247,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992068359691707,
"loss": 5.3027,
"mean_token_accuracy": 0.18512812107801438,
"num_tokens": 7689005.0,
"step": 4175
},
{
"entropy": 5.3886792182922365,
"epoch": 0.3491480120280655,
"grad_norm": 0.89453125,
"learning_rate": 0.0004992043365344132,
"loss": 5.2412,
"mean_token_accuracy": 0.18713783770799636,
"num_tokens": 7698231.0,
"step": 4180
},
{
"entropy": 5.5312230587005615,
"epoch": 0.3495656531907785,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992018331746891,
"loss": 5.3579,
"mean_token_accuracy": 0.18098943531513215,
"num_tokens": 7706678.0,
"step": 4185
},
{
"entropy": 5.573501062393189,
"epoch": 0.34998329435349146,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991993258900418,
"loss": 5.4105,
"mean_token_accuracy": 0.17862555086612703,
"num_tokens": 7715506.0,
"step": 4190
},
{
"entropy": 5.580753660202026,
"epoch": 0.3504009355162045,
"grad_norm": 0.86328125,
"learning_rate": 0.0004991968146805156,
"loss": 5.3785,
"mean_token_accuracy": 0.1769071564078331,
"num_tokens": 7726549.0,
"step": 4195
},
{
"entropy": 5.509833240509034,
"epoch": 0.35081857667891747,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991942995461542,
"loss": 5.4589,
"mean_token_accuracy": 0.17415786683559417,
"num_tokens": 7735563.0,
"step": 4200
},
{
"entropy": 5.561418914794922,
"epoch": 0.35123621784163045,
"grad_norm": 0.88671875,
"learning_rate": 0.0004991917804870018,
"loss": 5.39,
"mean_token_accuracy": 0.17825252413749695,
"num_tokens": 7746144.0,
"step": 4205
},
{
"entropy": 5.515225028991699,
"epoch": 0.3516538590043435,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991892575031024,
"loss": 5.4043,
"mean_token_accuracy": 0.17637984901666642,
"num_tokens": 7755394.0,
"step": 4210
},
{
"entropy": 5.475852298736572,
"epoch": 0.35207150016705646,
"grad_norm": 0.890625,
"learning_rate": 0.0004991867305945001,
"loss": 5.3548,
"mean_token_accuracy": 0.18296674489974976,
"num_tokens": 7764855.0,
"step": 4215
},
{
"entropy": 5.428618860244751,
"epoch": 0.35248914132976944,
"grad_norm": 0.95703125,
"learning_rate": 0.0004991841997612394,
"loss": 5.2775,
"mean_token_accuracy": 0.17634298652410507,
"num_tokens": 7773394.0,
"step": 4220
},
{
"entropy": 5.527509641647339,
"epoch": 0.3529067824924825,
"grad_norm": 1.03125,
"learning_rate": 0.0004991816650033643,
"loss": 5.3273,
"mean_token_accuracy": 0.17990745604038239,
"num_tokens": 7783111.0,
"step": 4225
},
{
"entropy": 5.581513357162476,
"epoch": 0.35332442365519545,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991791263209194,
"loss": 5.4648,
"mean_token_accuracy": 0.17874054461717606,
"num_tokens": 7792569.0,
"step": 4230
},
{
"entropy": 5.539682054519654,
"epoch": 0.35374206481790843,
"grad_norm": 1.0625,
"learning_rate": 0.0004991765837139491,
"loss": 5.3638,
"mean_token_accuracy": 0.1848058208823204,
"num_tokens": 7801483.0,
"step": 4235
},
{
"entropy": 5.5473864555358885,
"epoch": 0.35415970598062146,
"grad_norm": 0.85546875,
"learning_rate": 0.0004991740371824977,
"loss": 5.475,
"mean_token_accuracy": 0.17225684374570846,
"num_tokens": 7812274.0,
"step": 4240
},
{
"entropy": 5.444568729400634,
"epoch": 0.35457734714333444,
"grad_norm": 0.9609375,
"learning_rate": 0.00049917148672661,
"loss": 5.308,
"mean_token_accuracy": 0.18026846498250962,
"num_tokens": 7821101.0,
"step": 4245
},
{
"entropy": 5.5830871105194095,
"epoch": 0.3549949883060474,
"grad_norm": 0.9609375,
"learning_rate": 0.0004991689323463306,
"loss": 5.278,
"mean_token_accuracy": 0.18820747584104539,
"num_tokens": 7829805.0,
"step": 4250
},
{
"entropy": 5.527326202392578,
"epoch": 0.35541262946876045,
"grad_norm": 0.89453125,
"learning_rate": 0.0004991663740417042,
"loss": 5.3897,
"mean_token_accuracy": 0.17147670984268187,
"num_tokens": 7839772.0,
"step": 4255
},
{
"entropy": 5.46431040763855,
"epoch": 0.35583027063147343,
"grad_norm": 1.0,
"learning_rate": 0.0004991638118127757,
"loss": 5.4657,
"mean_token_accuracy": 0.16767864674329758,
"num_tokens": 7850290.0,
"step": 4260
},
{
"entropy": 5.605262565612793,
"epoch": 0.3562479117941864,
"grad_norm": 0.9375,
"learning_rate": 0.0004991612456595896,
"loss": 5.4276,
"mean_token_accuracy": 0.17392627596855165,
"num_tokens": 7860180.0,
"step": 4265
},
{
"entropy": 5.400895643234253,
"epoch": 0.35666555295689945,
"grad_norm": 1.015625,
"learning_rate": 0.0004991586755821913,
"loss": 5.2992,
"mean_token_accuracy": 0.1876406565308571,
"num_tokens": 7869625.0,
"step": 4270
},
{
"entropy": 5.378699493408203,
"epoch": 0.3570831941196124,
"grad_norm": 0.96484375,
"learning_rate": 0.0004991561015806254,
"loss": 5.2522,
"mean_token_accuracy": 0.17919157296419144,
"num_tokens": 7878671.0,
"step": 4275
},
{
"entropy": 5.505584573745727,
"epoch": 0.3575008352823254,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991535236549372,
"loss": 5.3708,
"mean_token_accuracy": 0.18536537140607834,
"num_tokens": 7887225.0,
"step": 4280
},
{
"entropy": 5.392125129699707,
"epoch": 0.35791847644503844,
"grad_norm": 0.9453125,
"learning_rate": 0.0004991509418051718,
"loss": 5.2674,
"mean_token_accuracy": 0.18982865065336227,
"num_tokens": 7895504.0,
"step": 4285
},
{
"entropy": 5.564821434020996,
"epoch": 0.3583361176077514,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991483560313743,
"loss": 5.3296,
"mean_token_accuracy": 0.1795837849378586,
"num_tokens": 7904875.0,
"step": 4290
},
{
"entropy": 5.544165992736817,
"epoch": 0.3587537587704644,
"grad_norm": 0.83203125,
"learning_rate": 0.00049914576633359,
"loss": 5.3932,
"mean_token_accuracy": 0.1829620286822319,
"num_tokens": 7914747.0,
"step": 4295
},
{
"entropy": 5.496010780334473,
"epoch": 0.3591713999331774,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991431727118643,
"loss": 5.3625,
"mean_token_accuracy": 0.17891425043344497,
"num_tokens": 7923858.0,
"step": 4300
},
{
"entropy": 5.441442918777466,
"epoch": 0.3595890410958904,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991405751662426,
"loss": 5.3107,
"mean_token_accuracy": 0.17826237678527831,
"num_tokens": 7932396.0,
"step": 4305
},
{
"entropy": 5.509703302383423,
"epoch": 0.3600066822586034,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991379736967703,
"loss": 5.3198,
"mean_token_accuracy": 0.18312637209892274,
"num_tokens": 7940763.0,
"step": 4310
},
{
"entropy": 5.437738275527954,
"epoch": 0.3604243234213164,
"grad_norm": 1.03125,
"learning_rate": 0.0004991353683034929,
"loss": 5.3267,
"mean_token_accuracy": 0.17540014386177064,
"num_tokens": 7949424.0,
"step": 4315
},
{
"entropy": 5.485245037078857,
"epoch": 0.3608419645840294,
"grad_norm": 0.95703125,
"learning_rate": 0.0004991327589864563,
"loss": 5.2993,
"mean_token_accuracy": 0.18449853658676146,
"num_tokens": 7958805.0,
"step": 4320
},
{
"entropy": 5.59886531829834,
"epoch": 0.3612596057467424,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991301457457057,
"loss": 5.4093,
"mean_token_accuracy": 0.18323519378900527,
"num_tokens": 7967859.0,
"step": 4325
},
{
"entropy": 5.609905862808228,
"epoch": 0.3616772469094554,
"grad_norm": 0.91796875,
"learning_rate": 0.0004991275285812873,
"loss": 5.5313,
"mean_token_accuracy": 0.16842426657676696,
"num_tokens": 7978357.0,
"step": 4330
},
{
"entropy": 5.582861852645874,
"epoch": 0.3620948880721684,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991249074932466,
"loss": 5.3713,
"mean_token_accuracy": 0.18289734274148942,
"num_tokens": 7986637.0,
"step": 4335
},
{
"entropy": 5.452457475662231,
"epoch": 0.36251252923488136,
"grad_norm": 0.96875,
"learning_rate": 0.0004991222824816296,
"loss": 5.2992,
"mean_token_accuracy": 0.1832432597875595,
"num_tokens": 7995954.0,
"step": 4340
},
{
"entropy": 5.53001594543457,
"epoch": 0.3629301703975944,
"grad_norm": 0.99609375,
"learning_rate": 0.0004991196535464824,
"loss": 5.3821,
"mean_token_accuracy": 0.1710544377565384,
"num_tokens": 8005193.0,
"step": 4345
},
{
"entropy": 5.633359813690186,
"epoch": 0.3633478115603074,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991170206878508,
"loss": 5.3933,
"mean_token_accuracy": 0.1848797470331192,
"num_tokens": 8014419.0,
"step": 4350
},
{
"entropy": 5.475250482559204,
"epoch": 0.36376545272302035,
"grad_norm": 0.95703125,
"learning_rate": 0.000499114383905781,
"loss": 5.3561,
"mean_token_accuracy": 0.17914826422929764,
"num_tokens": 8024133.0,
"step": 4355
},
{
"entropy": 5.414242506027222,
"epoch": 0.3641830938857334,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991117432003192,
"loss": 5.3087,
"mean_token_accuracy": 0.18435169905424117,
"num_tokens": 8033059.0,
"step": 4360
},
{
"entropy": 5.532291221618652,
"epoch": 0.36460073504844637,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991090985715116,
"loss": 5.3559,
"mean_token_accuracy": 0.18285473436117172,
"num_tokens": 8041866.0,
"step": 4365
},
{
"entropy": 5.484489965438843,
"epoch": 0.36501837621115935,
"grad_norm": 1.0234375,
"learning_rate": 0.0004991064500194043,
"loss": 5.3935,
"mean_token_accuracy": 0.17530420869588853,
"num_tokens": 8051541.0,
"step": 4370
},
{
"entropy": 5.559474945068359,
"epoch": 0.3654360173738724,
"grad_norm": 0.984375,
"learning_rate": 0.0004991037975440439,
"loss": 5.4684,
"mean_token_accuracy": 0.17431111484766007,
"num_tokens": 8061964.0,
"step": 4375
},
{
"entropy": 5.4441728591918945,
"epoch": 0.36585365853658536,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991011411454769,
"loss": 5.2885,
"mean_token_accuracy": 0.17922835648059846,
"num_tokens": 8071006.0,
"step": 4380
},
{
"entropy": 5.4478785514831545,
"epoch": 0.36627129969929834,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990984808237495,
"loss": 5.2853,
"mean_token_accuracy": 0.18344878703355788,
"num_tokens": 8080104.0,
"step": 4385
},
{
"entropy": 5.5162395477294925,
"epoch": 0.36668894086201137,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990958165789086,
"loss": 5.3393,
"mean_token_accuracy": 0.18054348081350327,
"num_tokens": 8089397.0,
"step": 4390
},
{
"entropy": 5.520235204696656,
"epoch": 0.36710658202472435,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990931484110005,
"loss": 5.3855,
"mean_token_accuracy": 0.18152005225419998,
"num_tokens": 8099055.0,
"step": 4395
},
{
"entropy": 5.469604539871216,
"epoch": 0.3675242231874373,
"grad_norm": 0.9140625,
"learning_rate": 0.0004990904763200722,
"loss": 5.312,
"mean_token_accuracy": 0.17921042442321777,
"num_tokens": 8107609.0,
"step": 4400
},
{
"entropy": 5.584817218780517,
"epoch": 0.36794186435015036,
"grad_norm": 0.96484375,
"learning_rate": 0.0004990878003061705,
"loss": 5.336,
"mean_token_accuracy": 0.18642372339963914,
"num_tokens": 8116905.0,
"step": 4405
},
{
"entropy": 5.436828756332398,
"epoch": 0.36835950551286334,
"grad_norm": 0.9765625,
"learning_rate": 0.0004990851203693419,
"loss": 5.3175,
"mean_token_accuracy": 0.18151832073926927,
"num_tokens": 8125580.0,
"step": 4410
},
{
"entropy": 5.453390455245971,
"epoch": 0.3687771466755763,
"grad_norm": 0.9375,
"learning_rate": 0.0004990824365096336,
"loss": 5.2437,
"mean_token_accuracy": 0.17969209402799607,
"num_tokens": 8134568.0,
"step": 4415
},
{
"entropy": 5.420157146453858,
"epoch": 0.36919478783828935,
"grad_norm": 0.98046875,
"learning_rate": 0.0004990797487270926,
"loss": 5.2362,
"mean_token_accuracy": 0.1819555178284645,
"num_tokens": 8143044.0,
"step": 4420
},
{
"entropy": 5.3947912693023685,
"epoch": 0.36961242900100233,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990770570217659,
"loss": 5.3487,
"mean_token_accuracy": 0.17773361653089523,
"num_tokens": 8153701.0,
"step": 4425
},
{
"entropy": 5.4455608367919925,
"epoch": 0.37003007016371536,
"grad_norm": 0.83203125,
"learning_rate": 0.0004990743613937005,
"loss": 5.2437,
"mean_token_accuracy": 0.1832289829850197,
"num_tokens": 8163264.0,
"step": 4430
},
{
"entropy": 5.4252534866333,
"epoch": 0.37044771132642834,
"grad_norm": 0.99609375,
"learning_rate": 0.0004990716618429436,
"loss": 5.3598,
"mean_token_accuracy": 0.1746056079864502,
"num_tokens": 8173332.0,
"step": 4435
},
{
"entropy": 5.527809572219849,
"epoch": 0.3708653524891413,
"grad_norm": 0.96484375,
"learning_rate": 0.0004990689583695428,
"loss": 5.3324,
"mean_token_accuracy": 0.17931413501501084,
"num_tokens": 8181881.0,
"step": 4440
},
{
"entropy": 5.585740804672241,
"epoch": 0.37128299365185435,
"grad_norm": 0.97265625,
"learning_rate": 0.000499066250973545,
"loss": 5.3827,
"mean_token_accuracy": 0.17656126022338867,
"num_tokens": 8190771.0,
"step": 4445
},
{
"entropy": 5.485811710357666,
"epoch": 0.37170063481456733,
"grad_norm": 0.984375,
"learning_rate": 0.0004990635396549978,
"loss": 5.3198,
"mean_token_accuracy": 0.18132144212722778,
"num_tokens": 8200740.0,
"step": 4450
},
{
"entropy": 5.397424268722534,
"epoch": 0.3721182759772803,
"grad_norm": 0.94921875,
"learning_rate": 0.0004990608244139486,
"loss": 5.3184,
"mean_token_accuracy": 0.17959628850221634,
"num_tokens": 8209935.0,
"step": 4455
},
{
"entropy": 5.448706912994385,
"epoch": 0.37253591713999334,
"grad_norm": 0.96875,
"learning_rate": 0.000499058105250445,
"loss": 5.3812,
"mean_token_accuracy": 0.1757279559969902,
"num_tokens": 8219137.0,
"step": 4460
},
{
"entropy": 5.576772546768188,
"epoch": 0.3729535583027063,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990553821645346,
"loss": 5.3446,
"mean_token_accuracy": 0.18512373119592668,
"num_tokens": 8228076.0,
"step": 4465
},
{
"entropy": 5.431705379486084,
"epoch": 0.3733711994654193,
"grad_norm": 0.953125,
"learning_rate": 0.000499052655156265,
"loss": 5.3125,
"mean_token_accuracy": 0.18290172070264815,
"num_tokens": 8237581.0,
"step": 4470
},
{
"entropy": 5.4816020965576175,
"epoch": 0.37378884062813234,
"grad_norm": 0.9921875,
"learning_rate": 0.0004990499242256839,
"loss": 5.3676,
"mean_token_accuracy": 0.17110753655433655,
"num_tokens": 8247384.0,
"step": 4475
},
{
"entropy": 5.565718650817871,
"epoch": 0.3742064817908453,
"grad_norm": 0.953125,
"learning_rate": 0.0004990471893728394,
"loss": 5.4229,
"mean_token_accuracy": 0.18246443718671798,
"num_tokens": 8256383.0,
"step": 4480
},
{
"entropy": 5.496962594985962,
"epoch": 0.3746241229535583,
"grad_norm": 1.0625,
"learning_rate": 0.000499044450597779,
"loss": 5.2883,
"mean_token_accuracy": 0.1851549506187439,
"num_tokens": 8264848.0,
"step": 4485
},
{
"entropy": 5.4337639808654785,
"epoch": 0.3750417641162713,
"grad_norm": 0.953125,
"learning_rate": 0.0004990417079005508,
"loss": 5.3225,
"mean_token_accuracy": 0.17415128499269486,
"num_tokens": 8274403.0,
"step": 4490
},
{
"entropy": 5.449707460403443,
"epoch": 0.3754594052789843,
"grad_norm": 1.015625,
"learning_rate": 0.000499038961281203,
"loss": 5.2412,
"mean_token_accuracy": 0.18816229254007338,
"num_tokens": 8282700.0,
"step": 4495
},
{
"entropy": 5.415894365310669,
"epoch": 0.3758770464416973,
"grad_norm": 1.0625,
"learning_rate": 0.0004990362107397834,
"loss": 5.2604,
"mean_token_accuracy": 0.18811499923467637,
"num_tokens": 8291718.0,
"step": 4500
},
{
"entropy": 5.445155239105224,
"epoch": 0.3762946876044103,
"grad_norm": 0.9609375,
"learning_rate": 0.0004990334562763401,
"loss": 5.3146,
"mean_token_accuracy": 0.18061639815568925,
"num_tokens": 8301113.0,
"step": 4505
},
{
"entropy": 5.5739223003387455,
"epoch": 0.3767123287671233,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990306978909216,
"loss": 5.4249,
"mean_token_accuracy": 0.1783556893467903,
"num_tokens": 8310357.0,
"step": 4510
},
{
"entropy": 5.497480010986328,
"epoch": 0.3771299699298363,
"grad_norm": 0.90625,
"learning_rate": 0.0004990279355835761,
"loss": 5.3499,
"mean_token_accuracy": 0.17866246551275253,
"num_tokens": 8319680.0,
"step": 4515
},
{
"entropy": 5.490698766708374,
"epoch": 0.3775476110925493,
"grad_norm": 1.0,
"learning_rate": 0.0004990251693543519,
"loss": 5.3772,
"mean_token_accuracy": 0.17705728113651276,
"num_tokens": 8329127.0,
"step": 4520
},
{
"entropy": 5.421241283416748,
"epoch": 0.3779652522552623,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990223992032975,
"loss": 5.2109,
"mean_token_accuracy": 0.18183238804340363,
"num_tokens": 8338308.0,
"step": 4525
},
{
"entropy": 5.459861898422242,
"epoch": 0.37838289341797526,
"grad_norm": 1.0,
"learning_rate": 0.0004990196251304612,
"loss": 5.2536,
"mean_token_accuracy": 0.1817749932408333,
"num_tokens": 8347491.0,
"step": 4530
},
{
"entropy": 5.35833101272583,
"epoch": 0.3788005345806883,
"grad_norm": 0.9140625,
"learning_rate": 0.000499016847135892,
"loss": 5.2562,
"mean_token_accuracy": 0.18395913243293763,
"num_tokens": 8357112.0,
"step": 4535
},
{
"entropy": 5.438414525985718,
"epoch": 0.3792181757434013,
"grad_norm": 0.91796875,
"learning_rate": 0.000499014065219638,
"loss": 5.2624,
"mean_token_accuracy": 0.18888045400381087,
"num_tokens": 8366285.0,
"step": 4540
},
{
"entropy": 5.432564544677734,
"epoch": 0.37963581690611425,
"grad_norm": 1.09375,
"learning_rate": 0.0004990112793817481,
"loss": 5.1775,
"mean_token_accuracy": 0.19680214822292327,
"num_tokens": 8374362.0,
"step": 4545
},
{
"entropy": 5.461804533004761,
"epoch": 0.3800534580688273,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990084896222713,
"loss": 5.3976,
"mean_token_accuracy": 0.18095886409282685,
"num_tokens": 8383297.0,
"step": 4550
},
{
"entropy": 5.445401191711426,
"epoch": 0.38047109923154027,
"grad_norm": 0.98828125,
"learning_rate": 0.0004990056959412562,
"loss": 5.2877,
"mean_token_accuracy": 0.18479195684194566,
"num_tokens": 8391926.0,
"step": 4555
},
{
"entropy": 5.5135434627532955,
"epoch": 0.38088874039425324,
"grad_norm": 1.015625,
"learning_rate": 0.0004990028983387517,
"loss": 5.2658,
"mean_token_accuracy": 0.18396686911582946,
"num_tokens": 8401226.0,
"step": 4560
},
{
"entropy": 5.504383659362793,
"epoch": 0.3813063815569663,
"grad_norm": 0.99609375,
"learning_rate": 0.0004990000968148068,
"loss": 5.2947,
"mean_token_accuracy": 0.18325761556625367,
"num_tokens": 8409966.0,
"step": 4565
},
{
"entropy": 5.343247270584106,
"epoch": 0.38172402271967926,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989972913694707,
"loss": 5.137,
"mean_token_accuracy": 0.19336719512939454,
"num_tokens": 8418281.0,
"step": 4570
},
{
"entropy": 5.325301742553711,
"epoch": 0.38214166388239224,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989944820027923,
"loss": 5.2612,
"mean_token_accuracy": 0.18612614125013352,
"num_tokens": 8427436.0,
"step": 4575
},
{
"entropy": 5.455880784988404,
"epoch": 0.38255930504510527,
"grad_norm": 0.97265625,
"learning_rate": 0.0004989916687148209,
"loss": 5.3052,
"mean_token_accuracy": 0.18906501680612564,
"num_tokens": 8436066.0,
"step": 4580
},
{
"entropy": 5.450431489944458,
"epoch": 0.38297694620781825,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989888515056057,
"loss": 5.3032,
"mean_token_accuracy": 0.18049872815608978,
"num_tokens": 8445572.0,
"step": 4585
},
{
"entropy": 5.45234866142273,
"epoch": 0.3833945873705312,
"grad_norm": 0.9609375,
"learning_rate": 0.000498986030375196,
"loss": 5.3413,
"mean_token_accuracy": 0.18092950880527497,
"num_tokens": 8454781.0,
"step": 4590
},
{
"entropy": 5.455178594589233,
"epoch": 0.38381222853324426,
"grad_norm": 1.046875,
"learning_rate": 0.0004989832053236413,
"loss": 5.421,
"mean_token_accuracy": 0.1788090169429779,
"num_tokens": 8463239.0,
"step": 4595
},
{
"entropy": 5.526310825347901,
"epoch": 0.38422986969595724,
"grad_norm": 1.0703125,
"learning_rate": 0.000498980376350991,
"loss": 5.3281,
"mean_token_accuracy": 0.18151934444904327,
"num_tokens": 8472664.0,
"step": 4600
},
{
"entropy": 5.490483808517456,
"epoch": 0.3846475108586702,
"grad_norm": 0.81640625,
"learning_rate": 0.0004989775434572945,
"loss": 5.3892,
"mean_token_accuracy": 0.1770053118467331,
"num_tokens": 8483149.0,
"step": 4605
},
{
"entropy": 5.542194366455078,
"epoch": 0.38506515202138325,
"grad_norm": 1.046875,
"learning_rate": 0.0004989747066426015,
"loss": 5.328,
"mean_token_accuracy": 0.1826325237751007,
"num_tokens": 8492297.0,
"step": 4610
},
{
"entropy": 5.458335304260254,
"epoch": 0.38548279318409623,
"grad_norm": 0.94140625,
"learning_rate": 0.0004989718659069618,
"loss": 5.391,
"mean_token_accuracy": 0.17796944677829743,
"num_tokens": 8502092.0,
"step": 4615
},
{
"entropy": 5.485839176177978,
"epoch": 0.3859004343468092,
"grad_norm": 0.92578125,
"learning_rate": 0.000498969021250425,
"loss": 5.3011,
"mean_token_accuracy": 0.1888509839773178,
"num_tokens": 8510816.0,
"step": 4620
},
{
"entropy": 5.493376874923706,
"epoch": 0.38631807550952224,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989661726730407,
"loss": 5.3257,
"mean_token_accuracy": 0.1872298002243042,
"num_tokens": 8520055.0,
"step": 4625
},
{
"entropy": 5.517747402191162,
"epoch": 0.3867357166722352,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989633201748591,
"loss": 5.3181,
"mean_token_accuracy": 0.18400509506464005,
"num_tokens": 8528271.0,
"step": 4630
},
{
"entropy": 5.408600234985352,
"epoch": 0.3871533578349482,
"grad_norm": 1.0234375,
"learning_rate": 0.0004989604637559299,
"loss": 5.23,
"mean_token_accuracy": 0.19140341132879257,
"num_tokens": 8537360.0,
"step": 4635
},
{
"entropy": 5.343968296051026,
"epoch": 0.38757099899766123,
"grad_norm": 0.94921875,
"learning_rate": 0.0004989576034163033,
"loss": 5.1977,
"mean_token_accuracy": 0.19754567593336106,
"num_tokens": 8546574.0,
"step": 4640
},
{
"entropy": 5.442503499984741,
"epoch": 0.3879886401603742,
"grad_norm": 1.0,
"learning_rate": 0.0004989547391560293,
"loss": 5.2907,
"mean_token_accuracy": 0.17888085693120956,
"num_tokens": 8555581.0,
"step": 4645
},
{
"entropy": 5.4283998012542725,
"epoch": 0.3884062813230872,
"grad_norm": 0.8984375,
"learning_rate": 0.000498951870975158,
"loss": 5.3173,
"mean_token_accuracy": 0.1870442271232605,
"num_tokens": 8564539.0,
"step": 4650
},
{
"entropy": 5.4952027797698975,
"epoch": 0.3888239224858002,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989489988737395,
"loss": 5.3606,
"mean_token_accuracy": 0.1791738510131836,
"num_tokens": 8573804.0,
"step": 4655
},
{
"entropy": 5.497137784957886,
"epoch": 0.3892415636485132,
"grad_norm": 0.94921875,
"learning_rate": 0.0004989461228518243,
"loss": 5.2993,
"mean_token_accuracy": 0.18310890942811966,
"num_tokens": 8583262.0,
"step": 4660
},
{
"entropy": 5.472321319580078,
"epoch": 0.3896592048112262,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989432429094627,
"loss": 5.3529,
"mean_token_accuracy": 0.1810036301612854,
"num_tokens": 8592763.0,
"step": 4665
},
{
"entropy": 5.512941455841064,
"epoch": 0.3900768459739392,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989403590467049,
"loss": 5.2376,
"mean_token_accuracy": 0.18580447435379027,
"num_tokens": 8601884.0,
"step": 4670
},
{
"entropy": 5.389830827713013,
"epoch": 0.3904944871366522,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989374712636017,
"loss": 5.3553,
"mean_token_accuracy": 0.17971681356430053,
"num_tokens": 8611207.0,
"step": 4675
},
{
"entropy": 5.355565977096558,
"epoch": 0.39091212829936517,
"grad_norm": 1.015625,
"learning_rate": 0.0004989345795602036,
"loss": 5.1123,
"mean_token_accuracy": 0.2001250520348549,
"num_tokens": 8620104.0,
"step": 4680
},
{
"entropy": 5.455119991302491,
"epoch": 0.3913297694620782,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989316839365609,
"loss": 5.3249,
"mean_token_accuracy": 0.17869911342859268,
"num_tokens": 8629552.0,
"step": 4685
},
{
"entropy": 5.4407007694244385,
"epoch": 0.3917474106247912,
"grad_norm": 1.125,
"learning_rate": 0.0004989287843927246,
"loss": 5.2481,
"mean_token_accuracy": 0.18548665046691895,
"num_tokens": 8638961.0,
"step": 4690
},
{
"entropy": 5.48536171913147,
"epoch": 0.39216505178750416,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989258809287455,
"loss": 5.2666,
"mean_token_accuracy": 0.1833942025899887,
"num_tokens": 8648051.0,
"step": 4695
},
{
"entropy": 5.468334913253784,
"epoch": 0.3925826929502172,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989229735446743,
"loss": 5.2606,
"mean_token_accuracy": 0.18585173040628433,
"num_tokens": 8656863.0,
"step": 4700
},
{
"entropy": 5.288018035888672,
"epoch": 0.39300033411293017,
"grad_norm": 1.03125,
"learning_rate": 0.0004989200622405619,
"loss": 5.1032,
"mean_token_accuracy": 0.19074878543615342,
"num_tokens": 8665455.0,
"step": 4705
},
{
"entropy": 5.3272233486175535,
"epoch": 0.39341797527564315,
"grad_norm": 0.96875,
"learning_rate": 0.0004989171470164592,
"loss": 5.2113,
"mean_token_accuracy": 0.18916970044374465,
"num_tokens": 8674119.0,
"step": 4710
},
{
"entropy": 5.447572708129883,
"epoch": 0.3938356164383562,
"grad_norm": 0.99609375,
"learning_rate": 0.0004989142278724173,
"loss": 5.3297,
"mean_token_accuracy": 0.18369702249765396,
"num_tokens": 8682979.0,
"step": 4715
},
{
"entropy": 5.475515508651734,
"epoch": 0.39425325760106916,
"grad_norm": 0.98046875,
"learning_rate": 0.0004989113048084874,
"loss": 5.2072,
"mean_token_accuracy": 0.18288736492395402,
"num_tokens": 8692392.0,
"step": 4720
},
{
"entropy": 5.3776037216186525,
"epoch": 0.39467089876378214,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989083778247206,
"loss": 5.2771,
"mean_token_accuracy": 0.1847996950149536,
"num_tokens": 8700932.0,
"step": 4725
},
{
"entropy": 5.558016729354859,
"epoch": 0.3950885399264952,
"grad_norm": 1.03125,
"learning_rate": 0.000498905446921168,
"loss": 5.436,
"mean_token_accuracy": 0.1749883309006691,
"num_tokens": 8710683.0,
"step": 4730
},
{
"entropy": 5.387879371643066,
"epoch": 0.39550618108920815,
"grad_norm": 0.984375,
"learning_rate": 0.0004989025120978812,
"loss": 5.2718,
"mean_token_accuracy": 0.18771557956933976,
"num_tokens": 8719939.0,
"step": 4735
},
{
"entropy": 5.485561561584473,
"epoch": 0.39592382225192113,
"grad_norm": 0.9765625,
"learning_rate": 0.0004988995733549113,
"loss": 5.3359,
"mean_token_accuracy": 0.1857832118868828,
"num_tokens": 8729275.0,
"step": 4740
},
{
"entropy": 5.4087183475494385,
"epoch": 0.39634146341463417,
"grad_norm": 1.015625,
"learning_rate": 0.0004988966306923099,
"loss": 5.2156,
"mean_token_accuracy": 0.18902769386768342,
"num_tokens": 8738048.0,
"step": 4745
},
{
"entropy": 5.4032769203186035,
"epoch": 0.39675910457734714,
"grad_norm": 1.03125,
"learning_rate": 0.0004988936841101284,
"loss": 5.2653,
"mean_token_accuracy": 0.18531980514526367,
"num_tokens": 8746967.0,
"step": 4750
},
{
"entropy": 5.339198780059815,
"epoch": 0.3971767457400601,
"grad_norm": 1.1484375,
"learning_rate": 0.0004988907336084186,
"loss": 5.2224,
"mean_token_accuracy": 0.18994965255260468,
"num_tokens": 8754908.0,
"step": 4755
},
{
"entropy": 5.406519651412964,
"epoch": 0.39759438690277316,
"grad_norm": 0.91015625,
"learning_rate": 0.0004988877791872319,
"loss": 5.2574,
"mean_token_accuracy": 0.19423574358224868,
"num_tokens": 8764489.0,
"step": 4760
},
{
"entropy": 5.5614875793457035,
"epoch": 0.39801202806548613,
"grad_norm": 0.98828125,
"learning_rate": 0.0004988848208466202,
"loss": 5.4313,
"mean_token_accuracy": 0.16883301138877868,
"num_tokens": 8774365.0,
"step": 4765
},
{
"entropy": 5.408624649047852,
"epoch": 0.3984296692281991,
"grad_norm": 1.09375,
"learning_rate": 0.0004988818585866351,
"loss": 5.2257,
"mean_token_accuracy": 0.18605009764432906,
"num_tokens": 8783538.0,
"step": 4770
},
{
"entropy": 5.402309942245483,
"epoch": 0.39884731039091215,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988788924073288,
"loss": 5.3255,
"mean_token_accuracy": 0.1894812449812889,
"num_tokens": 8792171.0,
"step": 4775
},
{
"entropy": 5.459277820587158,
"epoch": 0.3992649515536251,
"grad_norm": 0.98046875,
"learning_rate": 0.0004988759223087528,
"loss": 5.3174,
"mean_token_accuracy": 0.1798668995499611,
"num_tokens": 8801325.0,
"step": 4780
},
{
"entropy": 5.469427108764648,
"epoch": 0.3996825927163381,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988729482909595,
"loss": 5.2532,
"mean_token_accuracy": 0.1852971598505974,
"num_tokens": 8810566.0,
"step": 4785
},
{
"entropy": 5.375425672531128,
"epoch": 0.40010023387905114,
"grad_norm": 0.87109375,
"learning_rate": 0.0004988699703540006,
"loss": 5.2013,
"mean_token_accuracy": 0.18881352692842485,
"num_tokens": 8819735.0,
"step": 4790
},
{
"entropy": 5.40230393409729,
"epoch": 0.4005178750417641,
"grad_norm": 0.9375,
"learning_rate": 0.0004988669884979284,
"loss": 5.3252,
"mean_token_accuracy": 0.18047850131988524,
"num_tokens": 8828820.0,
"step": 4795
},
{
"entropy": 5.418774223327636,
"epoch": 0.4009355162044771,
"grad_norm": 0.95703125,
"learning_rate": 0.0004988640027227951,
"loss": 5.2248,
"mean_token_accuracy": 0.18174052387475967,
"num_tokens": 8837858.0,
"step": 4800
},
{
"entropy": 5.425327205657959,
"epoch": 0.4013531573671901,
"grad_norm": 0.90234375,
"learning_rate": 0.000498861013028653,
"loss": 5.2055,
"mean_token_accuracy": 0.18545352071523666,
"num_tokens": 8847172.0,
"step": 4805
},
{
"entropy": 5.384645223617554,
"epoch": 0.4017707985299031,
"grad_norm": 0.921875,
"learning_rate": 0.0004988580194155544,
"loss": 5.233,
"mean_token_accuracy": 0.18156313002109528,
"num_tokens": 8856700.0,
"step": 4810
},
{
"entropy": 5.275307989120483,
"epoch": 0.4021884396926161,
"grad_norm": 1.03125,
"learning_rate": 0.0004988550218835517,
"loss": 5.1187,
"mean_token_accuracy": 0.1975043758749962,
"num_tokens": 8866108.0,
"step": 4815
},
{
"entropy": 5.405415821075439,
"epoch": 0.4026060808553291,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988520204326975,
"loss": 5.2173,
"mean_token_accuracy": 0.1915104940533638,
"num_tokens": 8874833.0,
"step": 4820
},
{
"entropy": 5.45743842124939,
"epoch": 0.4030237220180421,
"grad_norm": 0.95703125,
"learning_rate": 0.0004988490150630441,
"loss": 5.301,
"mean_token_accuracy": 0.18634421974420548,
"num_tokens": 8884211.0,
"step": 4825
},
{
"entropy": 5.4931145191192625,
"epoch": 0.4034413631807551,
"grad_norm": 0.953125,
"learning_rate": 0.0004988460057746443,
"loss": 5.4088,
"mean_token_accuracy": 0.18082808703184128,
"num_tokens": 8893443.0,
"step": 4830
},
{
"entropy": 5.413536787033081,
"epoch": 0.4038590043434681,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988429925675507,
"loss": 5.2743,
"mean_token_accuracy": 0.18913335651159285,
"num_tokens": 8903544.0,
"step": 4835
},
{
"entropy": 5.441909122467041,
"epoch": 0.4042766455061811,
"grad_norm": 1.015625,
"learning_rate": 0.0004988399754418161,
"loss": 5.2809,
"mean_token_accuracy": 0.1874339371919632,
"num_tokens": 8912270.0,
"step": 4840
},
{
"entropy": 5.531057119369507,
"epoch": 0.40469428666889407,
"grad_norm": 0.9296875,
"learning_rate": 0.0004988369543974932,
"loss": 5.387,
"mean_token_accuracy": 0.17393327578902246,
"num_tokens": 8922155.0,
"step": 4845
},
{
"entropy": 5.510114336013794,
"epoch": 0.4051119278316071,
"grad_norm": 0.94921875,
"learning_rate": 0.0004988339294346351,
"loss": 5.3852,
"mean_token_accuracy": 0.18090247809886933,
"num_tokens": 8931844.0,
"step": 4850
},
{
"entropy": 5.429131507873535,
"epoch": 0.4055295689943201,
"grad_norm": 0.96875,
"learning_rate": 0.0004988309005532946,
"loss": 5.2321,
"mean_token_accuracy": 0.18972109705209733,
"num_tokens": 8940874.0,
"step": 4855
},
{
"entropy": 5.40586667060852,
"epoch": 0.40594721015703306,
"grad_norm": 1.015625,
"learning_rate": 0.0004988278677535247,
"loss": 5.3086,
"mean_token_accuracy": 0.18382637053728104,
"num_tokens": 8950108.0,
"step": 4860
},
{
"entropy": 5.452604484558106,
"epoch": 0.4063648513197461,
"grad_norm": 0.91015625,
"learning_rate": 0.0004988248310353788,
"loss": 5.3177,
"mean_token_accuracy": 0.18005881905555726,
"num_tokens": 8960610.0,
"step": 4865
},
{
"entropy": 5.367214918136597,
"epoch": 0.40678249248245907,
"grad_norm": 1.046875,
"learning_rate": 0.0004988217903989095,
"loss": 5.259,
"mean_token_accuracy": 0.18439610600471495,
"num_tokens": 8969853.0,
"step": 4870
},
{
"entropy": 5.476442050933838,
"epoch": 0.40720013364517205,
"grad_norm": 0.875,
"learning_rate": 0.0004988187458441704,
"loss": 5.2234,
"mean_token_accuracy": 0.18550434559583664,
"num_tokens": 8979800.0,
"step": 4875
},
{
"entropy": 5.430323886871338,
"epoch": 0.4076177748078851,
"grad_norm": 1.09375,
"learning_rate": 0.0004988156973712149,
"loss": 5.2922,
"mean_token_accuracy": 0.1802706629037857,
"num_tokens": 8988667.0,
"step": 4880
},
{
"entropy": 5.333445930480957,
"epoch": 0.40803541597059806,
"grad_norm": 1.0546875,
"learning_rate": 0.0004988126449800961,
"loss": 5.1509,
"mean_token_accuracy": 0.19155243933200836,
"num_tokens": 8997385.0,
"step": 4885
},
{
"entropy": 5.337029981613159,
"epoch": 0.40845305713331104,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988095886708676,
"loss": 5.1794,
"mean_token_accuracy": 0.19407922625541688,
"num_tokens": 9005999.0,
"step": 4890
},
{
"entropy": 5.460983085632324,
"epoch": 0.40887069829602407,
"grad_norm": 1.0,
"learning_rate": 0.0004988065284435828,
"loss": 5.3335,
"mean_token_accuracy": 0.18845028430223465,
"num_tokens": 9014837.0,
"step": 4895
},
{
"entropy": 5.475841856002807,
"epoch": 0.40928833945873705,
"grad_norm": 0.98046875,
"learning_rate": 0.0004988034642982954,
"loss": 5.3624,
"mean_token_accuracy": 0.1865231767296791,
"num_tokens": 9023956.0,
"step": 4900
},
{
"entropy": 5.4807093143463135,
"epoch": 0.40970598062145,
"grad_norm": 0.953125,
"learning_rate": 0.0004988003962350589,
"loss": 5.3288,
"mean_token_accuracy": 0.17951313257217408,
"num_tokens": 9032945.0,
"step": 4905
},
{
"entropy": 5.44377326965332,
"epoch": 0.41012362178416306,
"grad_norm": 0.9765625,
"learning_rate": 0.000498797324253927,
"loss": 5.2381,
"mean_token_accuracy": 0.19179674088954926,
"num_tokens": 9042109.0,
"step": 4910
},
{
"entropy": 5.377729606628418,
"epoch": 0.41054126294687604,
"grad_norm": 0.9453125,
"learning_rate": 0.0004987942483549537,
"loss": 5.255,
"mean_token_accuracy": 0.18214826732873918,
"num_tokens": 9052210.0,
"step": 4915
},
{
"entropy": 5.400795888900757,
"epoch": 0.410958904109589,
"grad_norm": 1.09375,
"learning_rate": 0.0004987911685381926,
"loss": 5.1925,
"mean_token_accuracy": 0.1923542872071266,
"num_tokens": 9060089.0,
"step": 4920
},
{
"entropy": 5.5380675315856935,
"epoch": 0.41137654527230205,
"grad_norm": 0.93359375,
"learning_rate": 0.0004987880848036977,
"loss": 5.3105,
"mean_token_accuracy": 0.1849738046526909,
"num_tokens": 9069235.0,
"step": 4925
},
{
"entropy": 5.398316287994385,
"epoch": 0.41179418643501503,
"grad_norm": 1.0,
"learning_rate": 0.0004987849971515232,
"loss": 5.2532,
"mean_token_accuracy": 0.1917312741279602,
"num_tokens": 9077509.0,
"step": 4930
},
{
"entropy": 5.40391321182251,
"epoch": 0.412211827597728,
"grad_norm": 0.953125,
"learning_rate": 0.0004987819055817227,
"loss": 5.1683,
"mean_token_accuracy": 0.19981021136045457,
"num_tokens": 9085493.0,
"step": 4935
},
{
"entropy": 5.274679708480835,
"epoch": 0.41262946876044104,
"grad_norm": 0.9453125,
"learning_rate": 0.0004987788100943507,
"loss": 5.1485,
"mean_token_accuracy": 0.18995134830474852,
"num_tokens": 9094748.0,
"step": 4940
},
{
"entropy": 5.393630743026733,
"epoch": 0.413047109923154,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987757106894613,
"loss": 5.2245,
"mean_token_accuracy": 0.18995702564716338,
"num_tokens": 9103678.0,
"step": 4945
},
{
"entropy": 5.443596220016479,
"epoch": 0.413464751085867,
"grad_norm": 1.015625,
"learning_rate": 0.0004987726073671086,
"loss": 5.2733,
"mean_token_accuracy": 0.18716512024402618,
"num_tokens": 9112851.0,
"step": 4950
},
{
"entropy": 5.515263319015503,
"epoch": 0.41388239224858003,
"grad_norm": 0.84765625,
"learning_rate": 0.0004987695001273471,
"loss": 5.5052,
"mean_token_accuracy": 0.16875243782997132,
"num_tokens": 9122106.0,
"step": 4955
},
{
"entropy": 5.4486387252807615,
"epoch": 0.414300033411293,
"grad_norm": 0.90234375,
"learning_rate": 0.0004987663889702313,
"loss": 5.2059,
"mean_token_accuracy": 0.1922791540622711,
"num_tokens": 9131580.0,
"step": 4960
},
{
"entropy": 5.443259859085083,
"epoch": 0.414717674574006,
"grad_norm": 0.9921875,
"learning_rate": 0.0004987632738958155,
"loss": 5.3536,
"mean_token_accuracy": 0.1818932443857193,
"num_tokens": 9141642.0,
"step": 4965
},
{
"entropy": 5.526469326019287,
"epoch": 0.415135315736719,
"grad_norm": 0.9140625,
"learning_rate": 0.0004987601549041541,
"loss": 5.3832,
"mean_token_accuracy": 0.1804221525788307,
"num_tokens": 9151192.0,
"step": 4970
},
{
"entropy": 5.455605936050415,
"epoch": 0.415552956899432,
"grad_norm": 0.9765625,
"learning_rate": 0.0004987570319953018,
"loss": 5.2786,
"mean_token_accuracy": 0.18481839895248414,
"num_tokens": 9161316.0,
"step": 4975
},
{
"entropy": 5.454035758972168,
"epoch": 0.415970598062145,
"grad_norm": 0.96484375,
"learning_rate": 0.0004987539051693134,
"loss": 5.2849,
"mean_token_accuracy": 0.18582287281751633,
"num_tokens": 9169535.0,
"step": 4980
},
{
"entropy": 5.412672090530395,
"epoch": 0.416388239224858,
"grad_norm": 1.0625,
"learning_rate": 0.0004987507744262436,
"loss": 5.3111,
"mean_token_accuracy": 0.18136962056159972,
"num_tokens": 9178187.0,
"step": 4985
},
{
"entropy": 5.362005805969238,
"epoch": 0.416805880387571,
"grad_norm": 0.90234375,
"learning_rate": 0.0004987476397661472,
"loss": 5.1985,
"mean_token_accuracy": 0.18647850900888444,
"num_tokens": 9186342.0,
"step": 4990
},
{
"entropy": 5.452874612808228,
"epoch": 0.41722352155028397,
"grad_norm": 0.96484375,
"learning_rate": 0.0004987445011890789,
"loss": 5.334,
"mean_token_accuracy": 0.18385065197944642,
"num_tokens": 9196470.0,
"step": 4995
},
{
"entropy": 5.414176321029663,
"epoch": 0.417641162712997,
"grad_norm": 0.95703125,
"learning_rate": 0.000498741358695094,
"loss": 5.2476,
"mean_token_accuracy": 0.19388433545827866,
"num_tokens": 9206587.0,
"step": 5000
},
{
"entropy": 5.410133504867554,
"epoch": 0.41805880387571,
"grad_norm": 0.91015625,
"learning_rate": 0.000498738212284247,
"loss": 5.242,
"mean_token_accuracy": 0.18349627852439881,
"num_tokens": 9216459.0,
"step": 5005
},
{
"entropy": 5.573675489425659,
"epoch": 0.41847644503842296,
"grad_norm": 0.8828125,
"learning_rate": 0.0004987350619565934,
"loss": 5.3523,
"mean_token_accuracy": 0.18130982518196107,
"num_tokens": 9227301.0,
"step": 5010
},
{
"entropy": 5.4957184314727785,
"epoch": 0.418894086201136,
"grad_norm": 0.875,
"learning_rate": 0.0004987319077121884,
"loss": 5.2809,
"mean_token_accuracy": 0.17829721421003342,
"num_tokens": 9236577.0,
"step": 5015
},
{
"entropy": 5.362134742736816,
"epoch": 0.419311727363849,
"grad_norm": 0.890625,
"learning_rate": 0.0004987287495510869,
"loss": 5.2789,
"mean_token_accuracy": 0.18880894035100937,
"num_tokens": 9246451.0,
"step": 5020
},
{
"entropy": 5.370787763595581,
"epoch": 0.41972936852656195,
"grad_norm": 0.94921875,
"learning_rate": 0.0004987255874733444,
"loss": 5.1613,
"mean_token_accuracy": 0.18581471890211104,
"num_tokens": 9256108.0,
"step": 5025
},
{
"entropy": 5.366769075393677,
"epoch": 0.420147009689275,
"grad_norm": 0.99609375,
"learning_rate": 0.0004987224214790161,
"loss": 5.2355,
"mean_token_accuracy": 0.1906559556722641,
"num_tokens": 9265173.0,
"step": 5030
},
{
"entropy": 5.400710725784302,
"epoch": 0.42056465085198796,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987192515681575,
"loss": 5.3162,
"mean_token_accuracy": 0.1848292425274849,
"num_tokens": 9273979.0,
"step": 5035
},
{
"entropy": 5.341616821289063,
"epoch": 0.42098229201470094,
"grad_norm": 1.0703125,
"learning_rate": 0.0004987160777408243,
"loss": 5.2551,
"mean_token_accuracy": 0.18350231200456618,
"num_tokens": 9281893.0,
"step": 5040
},
{
"entropy": 5.483113384246826,
"epoch": 0.421399933177414,
"grad_norm": 0.9375,
"learning_rate": 0.0004987128999970717,
"loss": 5.3695,
"mean_token_accuracy": 0.1833135813474655,
"num_tokens": 9290799.0,
"step": 5045
},
{
"entropy": 5.514388370513916,
"epoch": 0.42181757434012696,
"grad_norm": 0.8828125,
"learning_rate": 0.0004987097183369555,
"loss": 5.3442,
"mean_token_accuracy": 0.17365078181028365,
"num_tokens": 9300562.0,
"step": 5050
},
{
"entropy": 5.361443138122558,
"epoch": 0.42223521550283993,
"grad_norm": 0.9765625,
"learning_rate": 0.0004987065327605315,
"loss": 5.1097,
"mean_token_accuracy": 0.1997506558895111,
"num_tokens": 9308889.0,
"step": 5055
},
{
"entropy": 5.325141096115113,
"epoch": 0.42265285666555297,
"grad_norm": 0.90625,
"learning_rate": 0.0004987033432678553,
"loss": 5.2073,
"mean_token_accuracy": 0.18749006390571593,
"num_tokens": 9317957.0,
"step": 5060
},
{
"entropy": 5.502013301849365,
"epoch": 0.42307049782826595,
"grad_norm": 0.875,
"learning_rate": 0.0004987001498589828,
"loss": 5.4012,
"mean_token_accuracy": 0.17834450006484986,
"num_tokens": 9327849.0,
"step": 5065
},
{
"entropy": 5.440627956390381,
"epoch": 0.4234881389909789,
"grad_norm": 0.96484375,
"learning_rate": 0.00049869695253397,
"loss": 5.2616,
"mean_token_accuracy": 0.180339252948761,
"num_tokens": 9337165.0,
"step": 5070
},
{
"entropy": 5.352905988693237,
"epoch": 0.42390578015369196,
"grad_norm": 0.95703125,
"learning_rate": 0.0004986937512928727,
"loss": 5.1771,
"mean_token_accuracy": 0.1884523794054985,
"num_tokens": 9346817.0,
"step": 5075
},
{
"entropy": 5.5156840801239015,
"epoch": 0.42432342131640494,
"grad_norm": 1.0,
"learning_rate": 0.000498690546135747,
"loss": 5.2966,
"mean_token_accuracy": 0.1840095728635788,
"num_tokens": 9356420.0,
"step": 5080
},
{
"entropy": 5.410769605636597,
"epoch": 0.4247410624791179,
"grad_norm": 1.03125,
"learning_rate": 0.0004986873370626491,
"loss": 5.2288,
"mean_token_accuracy": 0.18823409378528594,
"num_tokens": 9365580.0,
"step": 5085
},
{
"entropy": 5.3939556121826175,
"epoch": 0.42515870364183095,
"grad_norm": 0.96875,
"learning_rate": 0.000498684124073635,
"loss": 5.2639,
"mean_token_accuracy": 0.18051869720220565,
"num_tokens": 9375116.0,
"step": 5090
},
{
"entropy": 5.493084716796875,
"epoch": 0.4255763448045439,
"grad_norm": 1.0,
"learning_rate": 0.0004986809071687612,
"loss": 5.2404,
"mean_token_accuracy": 0.18489679247140883,
"num_tokens": 9384129.0,
"step": 5095
},
{
"entropy": 5.440520906448365,
"epoch": 0.4259939859672569,
"grad_norm": 1.046875,
"learning_rate": 0.0004986776863480838,
"loss": 5.2057,
"mean_token_accuracy": 0.1855569988489151,
"num_tokens": 9392757.0,
"step": 5100
},
{
"entropy": 5.303371953964233,
"epoch": 0.42641162712996994,
"grad_norm": 0.98828125,
"learning_rate": 0.0004986744616116593,
"loss": 5.1688,
"mean_token_accuracy": 0.18864914178848266,
"num_tokens": 9401753.0,
"step": 5105
},
{
"entropy": 5.410655164718628,
"epoch": 0.4268292682926829,
"grad_norm": 0.96484375,
"learning_rate": 0.000498671232959544,
"loss": 5.376,
"mean_token_accuracy": 0.17809276133775712,
"num_tokens": 9411400.0,
"step": 5110
},
{
"entropy": 5.476285028457641,
"epoch": 0.42724690945539595,
"grad_norm": 1.03125,
"learning_rate": 0.0004986680003917947,
"loss": 5.2376,
"mean_token_accuracy": 0.18526431173086166,
"num_tokens": 9420206.0,
"step": 5115
},
{
"entropy": 5.423188352584839,
"epoch": 0.42766455061810893,
"grad_norm": 0.94921875,
"learning_rate": 0.0004986647639084677,
"loss": 5.224,
"mean_token_accuracy": 0.19024873226881028,
"num_tokens": 9428418.0,
"step": 5120
},
{
"entropy": 5.45528998374939,
"epoch": 0.4280821917808219,
"grad_norm": 1.0234375,
"learning_rate": 0.00049866152350962,
"loss": 5.2696,
"mean_token_accuracy": 0.17983853369951247,
"num_tokens": 9437624.0,
"step": 5125
},
{
"entropy": 5.353516912460327,
"epoch": 0.42849983294353494,
"grad_norm": 1.0390625,
"learning_rate": 0.0004986582791953078,
"loss": 5.17,
"mean_token_accuracy": 0.19529253244400024,
"num_tokens": 9446857.0,
"step": 5130
},
{
"entropy": 5.384277486801148,
"epoch": 0.4289174741062479,
"grad_norm": 0.95703125,
"learning_rate": 0.0004986550309655884,
"loss": 5.2684,
"mean_token_accuracy": 0.18311621993780136,
"num_tokens": 9456024.0,
"step": 5135
},
{
"entropy": 5.407329177856445,
"epoch": 0.4293351152689609,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986517788205184,
"loss": 5.1772,
"mean_token_accuracy": 0.19539380967617034,
"num_tokens": 9464018.0,
"step": 5140
},
{
"entropy": 5.410005283355713,
"epoch": 0.42975275643167393,
"grad_norm": 0.9921875,
"learning_rate": 0.0004986485227601548,
"loss": 5.3089,
"mean_token_accuracy": 0.18760735988616944,
"num_tokens": 9472430.0,
"step": 5145
},
{
"entropy": 5.392274856567383,
"epoch": 0.4301703975943869,
"grad_norm": 1.0859375,
"learning_rate": 0.0004986452627845547,
"loss": 5.2359,
"mean_token_accuracy": 0.1965549811720848,
"num_tokens": 9480592.0,
"step": 5150
},
{
"entropy": 5.335669231414795,
"epoch": 0.4305880387570999,
"grad_norm": 1.0078125,
"learning_rate": 0.000498641998893775,
"loss": 5.1734,
"mean_token_accuracy": 0.19266332387924195,
"num_tokens": 9489162.0,
"step": 5155
},
{
"entropy": 5.404368734359741,
"epoch": 0.4310056799198129,
"grad_norm": 0.921875,
"learning_rate": 0.0004986387310878728,
"loss": 5.243,
"mean_token_accuracy": 0.18955194801092148,
"num_tokens": 9499210.0,
"step": 5160
},
{
"entropy": 5.426062774658203,
"epoch": 0.4314233210825259,
"grad_norm": 0.91796875,
"learning_rate": 0.0004986354593669056,
"loss": 5.2849,
"mean_token_accuracy": 0.18945134431123734,
"num_tokens": 9508062.0,
"step": 5165
},
{
"entropy": 5.460740566253662,
"epoch": 0.4318409622452389,
"grad_norm": 0.953125,
"learning_rate": 0.0004986321837309304,
"loss": 5.2285,
"mean_token_accuracy": 0.19201678782701492,
"num_tokens": 9516756.0,
"step": 5170
},
{
"entropy": 5.3514862060546875,
"epoch": 0.4322586034079519,
"grad_norm": 1.0,
"learning_rate": 0.0004986289041800046,
"loss": 5.217,
"mean_token_accuracy": 0.19218968600034714,
"num_tokens": 9525305.0,
"step": 5175
},
{
"entropy": 5.404515171051026,
"epoch": 0.4326762445706649,
"grad_norm": 1.0546875,
"learning_rate": 0.0004986256207141857,
"loss": 5.1963,
"mean_token_accuracy": 0.1971699297428131,
"num_tokens": 9534310.0,
"step": 5180
},
{
"entropy": 5.482242631912231,
"epoch": 0.43309388573337787,
"grad_norm": 0.9453125,
"learning_rate": 0.0004986223333335311,
"loss": 5.3014,
"mean_token_accuracy": 0.18381469696760178,
"num_tokens": 9543648.0,
"step": 5185
},
{
"entropy": 5.361711120605468,
"epoch": 0.4335115268960909,
"grad_norm": 0.90625,
"learning_rate": 0.0004986190420380985,
"loss": 5.213,
"mean_token_accuracy": 0.19145919680595397,
"num_tokens": 9553081.0,
"step": 5190
},
{
"entropy": 5.431469583511353,
"epoch": 0.4339291680588039,
"grad_norm": 0.9765625,
"learning_rate": 0.0004986157468279453,
"loss": 5.2669,
"mean_token_accuracy": 0.18773092627525328,
"num_tokens": 9563184.0,
"step": 5195
},
{
"entropy": 5.414912986755371,
"epoch": 0.43434680922151686,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986124477031293,
"loss": 5.2228,
"mean_token_accuracy": 0.18737793564796448,
"num_tokens": 9572550.0,
"step": 5200
},
{
"entropy": 5.357417154312134,
"epoch": 0.4347644503842299,
"grad_norm": 0.92578125,
"learning_rate": 0.0004986091446637083,
"loss": 5.273,
"mean_token_accuracy": 0.18699531704187394,
"num_tokens": 9580900.0,
"step": 5205
},
{
"entropy": 5.410246229171753,
"epoch": 0.4351820915469429,
"grad_norm": 1.0546875,
"learning_rate": 0.00049860583770974,
"loss": 5.2595,
"mean_token_accuracy": 0.18967737555503844,
"num_tokens": 9589340.0,
"step": 5210
},
{
"entropy": 5.447885990142822,
"epoch": 0.43559973270965585,
"grad_norm": 0.95703125,
"learning_rate": 0.0004986025268412823,
"loss": 5.3718,
"mean_token_accuracy": 0.18643051236867905,
"num_tokens": 9599316.0,
"step": 5215
},
{
"entropy": 5.4523365020751955,
"epoch": 0.4360173738723689,
"grad_norm": 0.91796875,
"learning_rate": 0.0004985992120583933,
"loss": 5.2616,
"mean_token_accuracy": 0.1854226619005203,
"num_tokens": 9609350.0,
"step": 5220
},
{
"entropy": 5.468072843551636,
"epoch": 0.43643501503508186,
"grad_norm": 0.96875,
"learning_rate": 0.000498595893361131,
"loss": 5.3206,
"mean_token_accuracy": 0.18880557864904404,
"num_tokens": 9618491.0,
"step": 5225
},
{
"entropy": 5.445502424240113,
"epoch": 0.43685265619779484,
"grad_norm": 0.890625,
"learning_rate": 0.0004985925707495533,
"loss": 5.3821,
"mean_token_accuracy": 0.17734698802232743,
"num_tokens": 9629613.0,
"step": 5230
},
{
"entropy": 5.373774290084839,
"epoch": 0.4372702973605079,
"grad_norm": 0.98828125,
"learning_rate": 0.0004985892442237186,
"loss": 5.2525,
"mean_token_accuracy": 0.19087811559438705,
"num_tokens": 9638035.0,
"step": 5235
},
{
"entropy": 5.478862714767456,
"epoch": 0.43768793852322085,
"grad_norm": 0.9453125,
"learning_rate": 0.0004985859137836851,
"loss": 5.319,
"mean_token_accuracy": 0.17996042370796203,
"num_tokens": 9647378.0,
"step": 5240
},
{
"entropy": 5.485831451416016,
"epoch": 0.43810557968593383,
"grad_norm": 0.9921875,
"learning_rate": 0.0004985825794295109,
"loss": 5.2293,
"mean_token_accuracy": 0.1899410903453827,
"num_tokens": 9655728.0,
"step": 5245
},
{
"entropy": 5.378673982620239,
"epoch": 0.43852322084864687,
"grad_norm": 0.91796875,
"learning_rate": 0.0004985792411612546,
"loss": 5.2704,
"mean_token_accuracy": 0.18692253082990645,
"num_tokens": 9665063.0,
"step": 5250
},
{
"entropy": 5.424449920654297,
"epoch": 0.43894086201135984,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985758989789746,
"loss": 5.2715,
"mean_token_accuracy": 0.18209364265203476,
"num_tokens": 9674406.0,
"step": 5255
},
{
"entropy": 5.406913375854492,
"epoch": 0.4393585031740728,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985725528827293,
"loss": 5.2101,
"mean_token_accuracy": 0.18680866211652755,
"num_tokens": 9683303.0,
"step": 5260
},
{
"entropy": 5.406108665466308,
"epoch": 0.43977614433678586,
"grad_norm": 0.97265625,
"learning_rate": 0.0004985692028725774,
"loss": 5.2089,
"mean_token_accuracy": 0.18465999513864517,
"num_tokens": 9692938.0,
"step": 5265
},
{
"entropy": 5.433288526535034,
"epoch": 0.44019378549949884,
"grad_norm": 0.8984375,
"learning_rate": 0.0004985658489485774,
"loss": 5.2823,
"mean_token_accuracy": 0.1838356077671051,
"num_tokens": 9702192.0,
"step": 5270
},
{
"entropy": 5.4437394618988035,
"epoch": 0.4406114266622118,
"grad_norm": 0.88671875,
"learning_rate": 0.0004985624911107882,
"loss": 5.2679,
"mean_token_accuracy": 0.1794508084654808,
"num_tokens": 9711301.0,
"step": 5275
},
{
"entropy": 5.388669729232788,
"epoch": 0.44102906782492485,
"grad_norm": 0.98828125,
"learning_rate": 0.0004985591293592685,
"loss": 5.2627,
"mean_token_accuracy": 0.18056987673044206,
"num_tokens": 9720768.0,
"step": 5280
},
{
"entropy": 5.4140290260314945,
"epoch": 0.4414467089876378,
"grad_norm": 1.03125,
"learning_rate": 0.0004985557636940771,
"loss": 5.2976,
"mean_token_accuracy": 0.18801358640193938,
"num_tokens": 9730007.0,
"step": 5285
},
{
"entropy": 5.492460250854492,
"epoch": 0.4418643501503508,
"grad_norm": 0.97265625,
"learning_rate": 0.000498552394115273,
"loss": 5.3356,
"mean_token_accuracy": 0.18004858791828154,
"num_tokens": 9739546.0,
"step": 5290
},
{
"entropy": 5.385093545913696,
"epoch": 0.44228199131306384,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985490206229151,
"loss": 5.2072,
"mean_token_accuracy": 0.1893276423215866,
"num_tokens": 9749615.0,
"step": 5295
},
{
"entropy": 5.4636486053466795,
"epoch": 0.4426996324757768,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985456432170626,
"loss": 5.3061,
"mean_token_accuracy": 0.18105170279741287,
"num_tokens": 9759285.0,
"step": 5300
},
{
"entropy": 5.374495506286621,
"epoch": 0.4431172736384898,
"grad_norm": 0.89453125,
"learning_rate": 0.0004985422618977744,
"loss": 5.1792,
"mean_token_accuracy": 0.18297002613544464,
"num_tokens": 9769396.0,
"step": 5305
},
{
"entropy": 5.441941595077514,
"epoch": 0.44353491480120283,
"grad_norm": 0.9375,
"learning_rate": 0.00049853887666511,
"loss": 5.2961,
"mean_token_accuracy": 0.18522798120975495,
"num_tokens": 9779632.0,
"step": 5310
},
{
"entropy": 5.456464576721191,
"epoch": 0.4439525559639158,
"grad_norm": 1.015625,
"learning_rate": 0.0004985354875191283,
"loss": 5.2307,
"mean_token_accuracy": 0.1865606963634491,
"num_tokens": 9788090.0,
"step": 5315
},
{
"entropy": 5.2392005443573,
"epoch": 0.4443701971266288,
"grad_norm": 1.0,
"learning_rate": 0.0004985320944598889,
"loss": 5.1363,
"mean_token_accuracy": 0.19887329787015914,
"num_tokens": 9795844.0,
"step": 5320
},
{
"entropy": 5.341764259338379,
"epoch": 0.4447878382893418,
"grad_norm": 0.97265625,
"learning_rate": 0.0004985286974874512,
"loss": 5.1368,
"mean_token_accuracy": 0.19803263247013092,
"num_tokens": 9805119.0,
"step": 5325
},
{
"entropy": 5.4308655738830565,
"epoch": 0.4452054794520548,
"grad_norm": 1.015625,
"learning_rate": 0.0004985252966018745,
"loss": 5.2606,
"mean_token_accuracy": 0.1878809705376625,
"num_tokens": 9813842.0,
"step": 5330
},
{
"entropy": 5.3994889736175535,
"epoch": 0.4456231206147678,
"grad_norm": 1.0,
"learning_rate": 0.0004985218918032184,
"loss": 5.2535,
"mean_token_accuracy": 0.1783266320824623,
"num_tokens": 9822872.0,
"step": 5335
},
{
"entropy": 5.489075231552124,
"epoch": 0.4460407617774808,
"grad_norm": 0.90234375,
"learning_rate": 0.0004985184830915426,
"loss": 5.3732,
"mean_token_accuracy": 0.18024482429027558,
"num_tokens": 9832710.0,
"step": 5340
},
{
"entropy": 5.494144630432129,
"epoch": 0.4464584029401938,
"grad_norm": 1.0078125,
"learning_rate": 0.0004985150704669067,
"loss": 5.2658,
"mean_token_accuracy": 0.18504762649536133,
"num_tokens": 9841047.0,
"step": 5345
},
{
"entropy": 5.448194169998169,
"epoch": 0.44687604410290677,
"grad_norm": 0.94921875,
"learning_rate": 0.0004985116539293704,
"loss": 5.2422,
"mean_token_accuracy": 0.18845410645008087,
"num_tokens": 9850168.0,
"step": 5350
},
{
"entropy": 5.34441065788269,
"epoch": 0.4472936852656198,
"grad_norm": 0.984375,
"learning_rate": 0.0004985082334789936,
"loss": 5.3037,
"mean_token_accuracy": 0.1800173655152321,
"num_tokens": 9859550.0,
"step": 5355
},
{
"entropy": 5.488778877258301,
"epoch": 0.4477113264283328,
"grad_norm": 0.984375,
"learning_rate": 0.0004985048091158361,
"loss": 5.2413,
"mean_token_accuracy": 0.18067245930433273,
"num_tokens": 9868317.0,
"step": 5360
},
{
"entropy": 5.518444919586182,
"epoch": 0.44812896759104576,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985013808399578,
"loss": 5.4682,
"mean_token_accuracy": 0.17326602041721345,
"num_tokens": 9878421.0,
"step": 5365
},
{
"entropy": 5.441004753112793,
"epoch": 0.4485466087537588,
"grad_norm": 1.0625,
"learning_rate": 0.0004984979486514189,
"loss": 5.1946,
"mean_token_accuracy": 0.19337001740932463,
"num_tokens": 9887213.0,
"step": 5370
},
{
"entropy": 5.370984268188477,
"epoch": 0.44896424991647177,
"grad_norm": 0.95703125,
"learning_rate": 0.0004984945125502795,
"loss": 5.3106,
"mean_token_accuracy": 0.1782525047659874,
"num_tokens": 9897274.0,
"step": 5375
},
{
"entropy": 5.324212169647216,
"epoch": 0.44938189107918475,
"grad_norm": 0.9609375,
"learning_rate": 0.0004984910725365994,
"loss": 5.2354,
"mean_token_accuracy": 0.18489155322313308,
"num_tokens": 9906765.0,
"step": 5380
},
{
"entropy": 5.4923529624938965,
"epoch": 0.4497995322418978,
"grad_norm": 0.99609375,
"learning_rate": 0.0004984876286104392,
"loss": 5.2447,
"mean_token_accuracy": 0.1800004541873932,
"num_tokens": 9916149.0,
"step": 5385
},
{
"entropy": 5.422781181335449,
"epoch": 0.45021717340461076,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984841807718591,
"loss": 5.2428,
"mean_token_accuracy": 0.18369154930114745,
"num_tokens": 9925657.0,
"step": 5390
},
{
"entropy": 5.369697713851929,
"epoch": 0.45063481456732374,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984807290209193,
"loss": 5.2541,
"mean_token_accuracy": 0.17732944786548616,
"num_tokens": 9934740.0,
"step": 5395
},
{
"entropy": 5.533233261108398,
"epoch": 0.45105245573003677,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984772733576803,
"loss": 5.3809,
"mean_token_accuracy": 0.17752125561237336,
"num_tokens": 9943516.0,
"step": 5400
},
{
"entropy": 5.430769968032837,
"epoch": 0.45147009689274975,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984738137822028,
"loss": 5.3279,
"mean_token_accuracy": 0.18029251992702483,
"num_tokens": 9952998.0,
"step": 5405
},
{
"entropy": 5.385849094390869,
"epoch": 0.45188773805546273,
"grad_norm": 0.953125,
"learning_rate": 0.000498470350294547,
"loss": 5.1365,
"mean_token_accuracy": 0.1915778934955597,
"num_tokens": 9961698.0,
"step": 5410
},
{
"entropy": 5.405647563934326,
"epoch": 0.45230537921817576,
"grad_norm": 0.953125,
"learning_rate": 0.0004984668828947739,
"loss": 5.2799,
"mean_token_accuracy": 0.1811949774622917,
"num_tokens": 9972015.0,
"step": 5415
},
{
"entropy": 5.482445859909058,
"epoch": 0.45272302038088874,
"grad_norm": 0.9140625,
"learning_rate": 0.0004984634115829439,
"loss": 5.3477,
"mean_token_accuracy": 0.18006271719932557,
"num_tokens": 9981834.0,
"step": 5420
},
{
"entropy": 5.368338108062744,
"epoch": 0.4531406615436017,
"grad_norm": 0.984375,
"learning_rate": 0.000498459936359118,
"loss": 5.2058,
"mean_token_accuracy": 0.19074418842792512,
"num_tokens": 9991113.0,
"step": 5425
},
{
"entropy": 5.409156656265258,
"epoch": 0.45355830270631475,
"grad_norm": 0.95703125,
"learning_rate": 0.0004984564572233568,
"loss": 5.1929,
"mean_token_accuracy": 0.18956764787435532,
"num_tokens": 10000048.0,
"step": 5430
},
{
"entropy": 5.424026346206665,
"epoch": 0.45397594386902773,
"grad_norm": 0.9296875,
"learning_rate": 0.0004984529741757216,
"loss": 5.2397,
"mean_token_accuracy": 0.18849152326583862,
"num_tokens": 10009389.0,
"step": 5435
},
{
"entropy": 5.370808172225952,
"epoch": 0.4543935850317407,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984494872162728,
"loss": 5.0906,
"mean_token_accuracy": 0.1974114239215851,
"num_tokens": 10018342.0,
"step": 5440
},
{
"entropy": 5.4293419361114506,
"epoch": 0.45481122619445374,
"grad_norm": 0.95703125,
"learning_rate": 0.000498445996345072,
"loss": 5.2809,
"mean_token_accuracy": 0.19553306102752685,
"num_tokens": 10027219.0,
"step": 5445
},
{
"entropy": 5.375585412979126,
"epoch": 0.4552288673571667,
"grad_norm": 0.96875,
"learning_rate": 0.0004984425015621799,
"loss": 5.1771,
"mean_token_accuracy": 0.1849798634648323,
"num_tokens": 10036635.0,
"step": 5450
},
{
"entropy": 5.359622621536255,
"epoch": 0.4556465085198797,
"grad_norm": 1.109375,
"learning_rate": 0.000498439002867658,
"loss": 5.2232,
"mean_token_accuracy": 0.18533703237771987,
"num_tokens": 10045655.0,
"step": 5455
},
{
"entropy": 5.4857409477233885,
"epoch": 0.45606414968259273,
"grad_norm": 0.9375,
"learning_rate": 0.0004984355002615673,
"loss": 5.2819,
"mean_token_accuracy": 0.17572323679924012,
"num_tokens": 10054366.0,
"step": 5460
},
{
"entropy": 5.413508558273316,
"epoch": 0.4564817908453057,
"grad_norm": 0.9609375,
"learning_rate": 0.0004984319937439693,
"loss": 5.2297,
"mean_token_accuracy": 0.19253379106521606,
"num_tokens": 10062548.0,
"step": 5465
},
{
"entropy": 5.368886852264405,
"epoch": 0.4568994320080187,
"grad_norm": 0.98828125,
"learning_rate": 0.0004984284833149252,
"loss": 5.2346,
"mean_token_accuracy": 0.1868786096572876,
"num_tokens": 10071333.0,
"step": 5470
},
{
"entropy": 5.3347487449646,
"epoch": 0.4573170731707317,
"grad_norm": 0.97265625,
"learning_rate": 0.0004984249689744966,
"loss": 5.1106,
"mean_token_accuracy": 0.1973042905330658,
"num_tokens": 10079502.0,
"step": 5475
},
{
"entropy": 5.36964430809021,
"epoch": 0.4577347143334447,
"grad_norm": 0.9765625,
"learning_rate": 0.000498421450722745,
"loss": 5.1535,
"mean_token_accuracy": 0.19379387199878692,
"num_tokens": 10088648.0,
"step": 5480
},
{
"entropy": 5.343714666366577,
"epoch": 0.4581523554961577,
"grad_norm": 0.96484375,
"learning_rate": 0.000498417928559732,
"loss": 5.166,
"mean_token_accuracy": 0.19369458258152009,
"num_tokens": 10097742.0,
"step": 5485
},
{
"entropy": 5.424908399581909,
"epoch": 0.4585699966588707,
"grad_norm": 0.92578125,
"learning_rate": 0.0004984144024855191,
"loss": 5.2786,
"mean_token_accuracy": 0.18148799389600753,
"num_tokens": 10107228.0,
"step": 5490
},
{
"entropy": 5.416803455352783,
"epoch": 0.4589876378215837,
"grad_norm": 1.0078125,
"learning_rate": 0.0004984108725001683,
"loss": 5.1667,
"mean_token_accuracy": 0.1910475417971611,
"num_tokens": 10115488.0,
"step": 5495
},
{
"entropy": 5.320183229446411,
"epoch": 0.45940527898429667,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984073386037413,
"loss": 5.1075,
"mean_token_accuracy": 0.19790947884321214,
"num_tokens": 10124235.0,
"step": 5500
},
{
"entropy": 5.32041072845459,
"epoch": 0.4598229201470097,
"grad_norm": 0.90625,
"learning_rate": 0.0004984038007962997,
"loss": 5.207,
"mean_token_accuracy": 0.1984640121459961,
"num_tokens": 10133208.0,
"step": 5505
},
{
"entropy": 5.362421655654908,
"epoch": 0.4602405613097227,
"grad_norm": 0.875,
"learning_rate": 0.0004984002590779059,
"loss": 5.135,
"mean_token_accuracy": 0.19263925552368164,
"num_tokens": 10142754.0,
"step": 5510
},
{
"entropy": 5.436812019348144,
"epoch": 0.46065820247243566,
"grad_norm": 0.99609375,
"learning_rate": 0.0004983967134486215,
"loss": 5.2499,
"mean_token_accuracy": 0.1877599313855171,
"num_tokens": 10152114.0,
"step": 5515
},
{
"entropy": 5.351143932342529,
"epoch": 0.4610758436351487,
"grad_norm": 0.921875,
"learning_rate": 0.0004983931639085088,
"loss": 5.1557,
"mean_token_accuracy": 0.19096697121858597,
"num_tokens": 10161562.0,
"step": 5520
},
{
"entropy": 5.45611047744751,
"epoch": 0.4614934847978617,
"grad_norm": 1.015625,
"learning_rate": 0.0004983896104576298,
"loss": 5.2715,
"mean_token_accuracy": 0.18681198209524155,
"num_tokens": 10170461.0,
"step": 5525
},
{
"entropy": 5.362908458709716,
"epoch": 0.46191112596057465,
"grad_norm": 1.09375,
"learning_rate": 0.000498386053096047,
"loss": 5.274,
"mean_token_accuracy": 0.1872062012553215,
"num_tokens": 10179243.0,
"step": 5530
},
{
"entropy": 5.375006246566772,
"epoch": 0.4623287671232877,
"grad_norm": 1.09375,
"learning_rate": 0.0004983824918238221,
"loss": 5.2075,
"mean_token_accuracy": 0.1902789369225502,
"num_tokens": 10187935.0,
"step": 5535
},
{
"entropy": 5.432381200790405,
"epoch": 0.46274640828600067,
"grad_norm": 1.0,
"learning_rate": 0.000498378926641018,
"loss": 5.2808,
"mean_token_accuracy": 0.18302738815546035,
"num_tokens": 10196185.0,
"step": 5540
},
{
"entropy": 5.3857447624206545,
"epoch": 0.46316404944871364,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983753575476969,
"loss": 5.2658,
"mean_token_accuracy": 0.18335500210523606,
"num_tokens": 10205373.0,
"step": 5545
},
{
"entropy": 5.405973052978515,
"epoch": 0.4635816906114267,
"grad_norm": 0.8828125,
"learning_rate": 0.0004983717845439212,
"loss": 5.243,
"mean_token_accuracy": 0.19455428719520568,
"num_tokens": 10214464.0,
"step": 5550
},
{
"entropy": 5.424784326553345,
"epoch": 0.46399933177413966,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983682076297536,
"loss": 5.3229,
"mean_token_accuracy": 0.18389758318662644,
"num_tokens": 10224462.0,
"step": 5555
},
{
"entropy": 5.459895086288452,
"epoch": 0.46441697293685263,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983646268052567,
"loss": 5.2367,
"mean_token_accuracy": 0.18775293081998826,
"num_tokens": 10233123.0,
"step": 5560
},
{
"entropy": 5.333229207992554,
"epoch": 0.46483461409956567,
"grad_norm": 0.9453125,
"learning_rate": 0.0004983610420704931,
"loss": 5.0822,
"mean_token_accuracy": 0.19521746039390564,
"num_tokens": 10241613.0,
"step": 5565
},
{
"entropy": 5.440250730514526,
"epoch": 0.46525225526227865,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983574534255256,
"loss": 5.4372,
"mean_token_accuracy": 0.17527129352092743,
"num_tokens": 10251809.0,
"step": 5570
},
{
"entropy": 5.352139568328857,
"epoch": 0.4656698964249916,
"grad_norm": 0.9609375,
"learning_rate": 0.000498353860870417,
"loss": 5.1216,
"mean_token_accuracy": 0.19288666397333146,
"num_tokens": 10260198.0,
"step": 5575
},
{
"entropy": 5.390357160568238,
"epoch": 0.46608753758770466,
"grad_norm": 0.89453125,
"learning_rate": 0.0004983502644052303,
"loss": 5.1815,
"mean_token_accuracy": 0.19071156680583953,
"num_tokens": 10268906.0,
"step": 5580
},
{
"entropy": 5.36174168586731,
"epoch": 0.46650517875041764,
"grad_norm": 0.99609375,
"learning_rate": 0.0004983466640300282,
"loss": 5.1131,
"mean_token_accuracy": 0.18751837760210038,
"num_tokens": 10278526.0,
"step": 5585
},
{
"entropy": 5.390116214752197,
"epoch": 0.4669228199131306,
"grad_norm": 1.0390625,
"learning_rate": 0.0004983430597448741,
"loss": 5.2434,
"mean_token_accuracy": 0.18560836166143418,
"num_tokens": 10287428.0,
"step": 5590
},
{
"entropy": 5.506964445114136,
"epoch": 0.46734046107584365,
"grad_norm": 0.96484375,
"learning_rate": 0.0004983394515498308,
"loss": 5.1983,
"mean_token_accuracy": 0.1844811961054802,
"num_tokens": 10297136.0,
"step": 5595
},
{
"entropy": 5.410887098312378,
"epoch": 0.46775810223855663,
"grad_norm": 0.92578125,
"learning_rate": 0.0004983358394449616,
"loss": 5.2765,
"mean_token_accuracy": 0.18754280358552933,
"num_tokens": 10305678.0,
"step": 5600
},
{
"entropy": 5.2654417037963865,
"epoch": 0.4681757434012696,
"grad_norm": 0.984375,
"learning_rate": 0.0004983322234303297,
"loss": 5.1988,
"mean_token_accuracy": 0.18852692395448684,
"num_tokens": 10314690.0,
"step": 5605
},
{
"entropy": 5.348377323150634,
"epoch": 0.46859338456398264,
"grad_norm": 1.015625,
"learning_rate": 0.0004983286035059984,
"loss": 5.1602,
"mean_token_accuracy": 0.19106952250003814,
"num_tokens": 10323521.0,
"step": 5610
},
{
"entropy": 5.4127795696258545,
"epoch": 0.4690110257266956,
"grad_norm": 1.0234375,
"learning_rate": 0.0004983249796720311,
"loss": 5.1905,
"mean_token_accuracy": 0.19329796284437178,
"num_tokens": 10333077.0,
"step": 5615
},
{
"entropy": 5.431782388687134,
"epoch": 0.4694286668894086,
"grad_norm": 1.078125,
"learning_rate": 0.0004983213519284912,
"loss": 5.2205,
"mean_token_accuracy": 0.18967044204473496,
"num_tokens": 10342194.0,
"step": 5620
},
{
"entropy": 5.323764038085938,
"epoch": 0.46984630805212163,
"grad_norm": 0.953125,
"learning_rate": 0.0004983177202754421,
"loss": 5.0621,
"mean_token_accuracy": 0.19803830236196518,
"num_tokens": 10351009.0,
"step": 5625
},
{
"entropy": 5.41808853149414,
"epoch": 0.4702639492148346,
"grad_norm": 1.03125,
"learning_rate": 0.0004983140847129476,
"loss": 5.2783,
"mean_token_accuracy": 0.18249369263648987,
"num_tokens": 10360872.0,
"step": 5630
},
{
"entropy": 5.430299997329712,
"epoch": 0.4706815903775476,
"grad_norm": 0.9765625,
"learning_rate": 0.0004983104452410713,
"loss": 5.1754,
"mean_token_accuracy": 0.1885490745306015,
"num_tokens": 10370159.0,
"step": 5635
},
{
"entropy": 5.383899354934693,
"epoch": 0.4710992315402606,
"grad_norm": 1.0390625,
"learning_rate": 0.0004983068018598768,
"loss": 5.1723,
"mean_token_accuracy": 0.19473619759082794,
"num_tokens": 10379219.0,
"step": 5640
},
{
"entropy": 5.3625024318695065,
"epoch": 0.4715168727029736,
"grad_norm": 0.9296875,
"learning_rate": 0.000498303154569428,
"loss": 5.2137,
"mean_token_accuracy": 0.1863038271665573,
"num_tokens": 10389425.0,
"step": 5645
},
{
"entropy": 5.451694583892822,
"epoch": 0.4719345138656866,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982995033697887,
"loss": 5.2286,
"mean_token_accuracy": 0.19073546230792998,
"num_tokens": 10398023.0,
"step": 5650
},
{
"entropy": 5.4090780258178714,
"epoch": 0.4723521550283996,
"grad_norm": 1.0234375,
"learning_rate": 0.0004982958482610226,
"loss": 5.1718,
"mean_token_accuracy": 0.19427365362644194,
"num_tokens": 10406578.0,
"step": 5655
},
{
"entropy": 5.289528942108154,
"epoch": 0.4727697961911126,
"grad_norm": 0.93359375,
"learning_rate": 0.000498292189243194,
"loss": 5.1769,
"mean_token_accuracy": 0.19172403216362,
"num_tokens": 10416448.0,
"step": 5660
},
{
"entropy": 5.445108842849732,
"epoch": 0.47318743735382557,
"grad_norm": 1.015625,
"learning_rate": 0.000498288526316367,
"loss": 5.1999,
"mean_token_accuracy": 0.18580440580844879,
"num_tokens": 10425414.0,
"step": 5665
},
{
"entropy": 5.403898859024048,
"epoch": 0.4736050785165386,
"grad_norm": 0.9140625,
"learning_rate": 0.0004982848594806054,
"loss": 5.1811,
"mean_token_accuracy": 0.19948876798152923,
"num_tokens": 10433900.0,
"step": 5670
},
{
"entropy": 5.4370215892791744,
"epoch": 0.4740227196792516,
"grad_norm": 1.03125,
"learning_rate": 0.0004982811887359736,
"loss": 5.3001,
"mean_token_accuracy": 0.18371444642543794,
"num_tokens": 10443077.0,
"step": 5675
},
{
"entropy": 5.398310708999634,
"epoch": 0.47444036084196456,
"grad_norm": 0.9375,
"learning_rate": 0.0004982775140825359,
"loss": 5.2305,
"mean_token_accuracy": 0.19003031849861146,
"num_tokens": 10452791.0,
"step": 5680
},
{
"entropy": 5.478526496887207,
"epoch": 0.4748580020046776,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982738355203565,
"loss": 5.3936,
"mean_token_accuracy": 0.18002498000860215,
"num_tokens": 10462087.0,
"step": 5685
},
{
"entropy": 5.5195990085601805,
"epoch": 0.47527564316739057,
"grad_norm": 0.96875,
"learning_rate": 0.0004982701530494998,
"loss": 5.3888,
"mean_token_accuracy": 0.1750195562839508,
"num_tokens": 10472576.0,
"step": 5690
},
{
"entropy": 5.404919528961182,
"epoch": 0.47569328433010355,
"grad_norm": 1.0546875,
"learning_rate": 0.0004982664666700304,
"loss": 5.1181,
"mean_token_accuracy": 0.20003928244113922,
"num_tokens": 10481176.0,
"step": 5695
},
{
"entropy": 5.38076376914978,
"epoch": 0.4761109254928166,
"grad_norm": 1.0859375,
"learning_rate": 0.0004982627763820128,
"loss": 5.2059,
"mean_token_accuracy": 0.18965410739183425,
"num_tokens": 10490558.0,
"step": 5700
},
{
"entropy": 5.408396530151367,
"epoch": 0.47652856665552956,
"grad_norm": 0.87890625,
"learning_rate": 0.0004982590821855115,
"loss": 5.274,
"mean_token_accuracy": 0.19038996398448943,
"num_tokens": 10499543.0,
"step": 5705
},
{
"entropy": 5.362735366821289,
"epoch": 0.47694620781824254,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982553840805912,
"loss": 5.1621,
"mean_token_accuracy": 0.19401407837867737,
"num_tokens": 10508162.0,
"step": 5710
},
{
"entropy": 5.34212818145752,
"epoch": 0.4773638489809556,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982516820673167,
"loss": 5.0836,
"mean_token_accuracy": 0.1903264433145523,
"num_tokens": 10516837.0,
"step": 5715
},
{
"entropy": 5.300382041931153,
"epoch": 0.47778149014366855,
"grad_norm": 0.9375,
"learning_rate": 0.0004982479761457528,
"loss": 5.2244,
"mean_token_accuracy": 0.19203804284334183,
"num_tokens": 10526208.0,
"step": 5720
},
{
"entropy": 5.444773626327515,
"epoch": 0.47819913130638153,
"grad_norm": 1.03125,
"learning_rate": 0.0004982442663159643,
"loss": 5.2555,
"mean_token_accuracy": 0.18101535737514496,
"num_tokens": 10536002.0,
"step": 5725
},
{
"entropy": 5.475929498672485,
"epoch": 0.47861677246909456,
"grad_norm": 0.90625,
"learning_rate": 0.0004982405525780162,
"loss": 5.2441,
"mean_token_accuracy": 0.18552520126104355,
"num_tokens": 10545861.0,
"step": 5730
},
{
"entropy": 5.3596611499786375,
"epoch": 0.47903441363180754,
"grad_norm": 0.98046875,
"learning_rate": 0.0004982368349319736,
"loss": 5.147,
"mean_token_accuracy": 0.19177533984184264,
"num_tokens": 10555357.0,
"step": 5735
},
{
"entropy": 5.350248956680298,
"epoch": 0.4794520547945205,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982331133779014,
"loss": 5.2263,
"mean_token_accuracy": 0.18407194167375565,
"num_tokens": 10564744.0,
"step": 5740
},
{
"entropy": 5.3497522354125975,
"epoch": 0.47986969595723356,
"grad_norm": 0.94921875,
"learning_rate": 0.0004982293879158649,
"loss": 5.1941,
"mean_token_accuracy": 0.19337435066699982,
"num_tokens": 10573736.0,
"step": 5745
},
{
"entropy": 5.380324411392212,
"epoch": 0.48028733711994653,
"grad_norm": 0.92578125,
"learning_rate": 0.0004982256585459293,
"loss": 5.252,
"mean_token_accuracy": 0.18973010331392287,
"num_tokens": 10583238.0,
"step": 5750
},
{
"entropy": 5.468371629714966,
"epoch": 0.4807049782826595,
"grad_norm": 0.98046875,
"learning_rate": 0.0004982219252681597,
"loss": 5.2951,
"mean_token_accuracy": 0.18508350402116774,
"num_tokens": 10592316.0,
"step": 5755
},
{
"entropy": 5.426727151870727,
"epoch": 0.48112261944537255,
"grad_norm": 1.0,
"learning_rate": 0.0004982181880826218,
"loss": 5.3192,
"mean_token_accuracy": 0.19533029198646545,
"num_tokens": 10601418.0,
"step": 5760
},
{
"entropy": 5.3590703964233395,
"epoch": 0.4815402606080855,
"grad_norm": 0.92578125,
"learning_rate": 0.0004982144469893807,
"loss": 5.2041,
"mean_token_accuracy": 0.1928260952234268,
"num_tokens": 10610513.0,
"step": 5765
},
{
"entropy": 5.3929990291595455,
"epoch": 0.4819579017707985,
"grad_norm": 0.96484375,
"learning_rate": 0.000498210701988502,
"loss": 5.1422,
"mean_token_accuracy": 0.19776597917079924,
"num_tokens": 10619529.0,
"step": 5770
},
{
"entropy": 5.492319536209107,
"epoch": 0.48237554293351154,
"grad_norm": 0.953125,
"learning_rate": 0.0004982069530800514,
"loss": 5.2251,
"mean_token_accuracy": 0.20145247578620912,
"num_tokens": 10628263.0,
"step": 5775
},
{
"entropy": 5.414431095123291,
"epoch": 0.4827931840962245,
"grad_norm": 0.984375,
"learning_rate": 0.0004982032002640943,
"loss": 5.2551,
"mean_token_accuracy": 0.19498594105243683,
"num_tokens": 10637605.0,
"step": 5780
},
{
"entropy": 5.4010262966156,
"epoch": 0.48321082525893755,
"grad_norm": 1.046875,
"learning_rate": 0.0004981994435406965,
"loss": 5.3177,
"mean_token_accuracy": 0.18424908220767974,
"num_tokens": 10647113.0,
"step": 5785
},
{
"entropy": 5.342437839508056,
"epoch": 0.4836284664216505,
"grad_norm": 1.0625,
"learning_rate": 0.0004981956829099239,
"loss": 5.0705,
"mean_token_accuracy": 0.19768524020910264,
"num_tokens": 10655464.0,
"step": 5790
},
{
"entropy": 5.4189153671264645,
"epoch": 0.4840461075843635,
"grad_norm": 0.98046875,
"learning_rate": 0.0004981919183718422,
"loss": 5.1929,
"mean_token_accuracy": 0.1941748008131981,
"num_tokens": 10664689.0,
"step": 5795
},
{
"entropy": 5.292117357254028,
"epoch": 0.48446374874707654,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981881499265173,
"loss": 5.0936,
"mean_token_accuracy": 0.19338988065719603,
"num_tokens": 10673440.0,
"step": 5800
},
{
"entropy": 5.337329292297364,
"epoch": 0.4848813899097895,
"grad_norm": 0.921875,
"learning_rate": 0.0004981843775740151,
"loss": 5.2323,
"mean_token_accuracy": 0.19166841208934784,
"num_tokens": 10683075.0,
"step": 5805
},
{
"entropy": 5.395388603210449,
"epoch": 0.4852990310725025,
"grad_norm": 0.96875,
"learning_rate": 0.0004981806013144017,
"loss": 5.1888,
"mean_token_accuracy": 0.19218732565641403,
"num_tokens": 10692663.0,
"step": 5810
},
{
"entropy": 5.365407085418701,
"epoch": 0.48571667223521553,
"grad_norm": 0.94140625,
"learning_rate": 0.0004981768211477434,
"loss": 5.081,
"mean_token_accuracy": 0.19704576283693315,
"num_tokens": 10702068.0,
"step": 5815
},
{
"entropy": 5.358780479431152,
"epoch": 0.4861343133979285,
"grad_norm": 0.9140625,
"learning_rate": 0.000498173037074106,
"loss": 5.2354,
"mean_token_accuracy": 0.18202263861894608,
"num_tokens": 10711230.0,
"step": 5820
},
{
"entropy": 5.319102954864502,
"epoch": 0.4865519545606415,
"grad_norm": 0.98046875,
"learning_rate": 0.0004981692490935561,
"loss": 5.2044,
"mean_token_accuracy": 0.19445421248674394,
"num_tokens": 10719651.0,
"step": 5825
},
{
"entropy": 5.340964365005493,
"epoch": 0.4869695957233545,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981654572061597,
"loss": 5.2151,
"mean_token_accuracy": 0.19404265880584717,
"num_tokens": 10729342.0,
"step": 5830
},
{
"entropy": 5.378287601470947,
"epoch": 0.4873872368860675,
"grad_norm": 0.9921875,
"learning_rate": 0.0004981616614119836,
"loss": 5.2188,
"mean_token_accuracy": 0.19113572984933852,
"num_tokens": 10738617.0,
"step": 5835
},
{
"entropy": 5.4555384635925295,
"epoch": 0.4878048780487805,
"grad_norm": 0.96875,
"learning_rate": 0.0004981578617110938,
"loss": 5.249,
"mean_token_accuracy": 0.19108872264623641,
"num_tokens": 10746881.0,
"step": 5840
},
{
"entropy": 5.3574690341949465,
"epoch": 0.4882225192114935,
"grad_norm": 0.96875,
"learning_rate": 0.0004981540581035571,
"loss": 5.1101,
"mean_token_accuracy": 0.19686870127916337,
"num_tokens": 10755848.0,
"step": 5845
},
{
"entropy": 5.291342878341675,
"epoch": 0.4886401603742065,
"grad_norm": 0.94921875,
"learning_rate": 0.00049815025058944,
"loss": 5.1152,
"mean_token_accuracy": 0.18852638751268386,
"num_tokens": 10765519.0,
"step": 5850
},
{
"entropy": 5.288955354690552,
"epoch": 0.48905780153691947,
"grad_norm": 0.89453125,
"learning_rate": 0.0004981464391688092,
"loss": 5.2264,
"mean_token_accuracy": 0.1883165329694748,
"num_tokens": 10776408.0,
"step": 5855
},
{
"entropy": 5.379974412918091,
"epoch": 0.4894754426996325,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981426238417312,
"loss": 5.2238,
"mean_token_accuracy": 0.1914668157696724,
"num_tokens": 10785499.0,
"step": 5860
},
{
"entropy": 5.373927640914917,
"epoch": 0.4898930838623455,
"grad_norm": 1.0390625,
"learning_rate": 0.0004981388046082731,
"loss": 5.1774,
"mean_token_accuracy": 0.1898777186870575,
"num_tokens": 10794736.0,
"step": 5865
},
{
"entropy": 5.354801654815674,
"epoch": 0.49031072502505846,
"grad_norm": 0.9765625,
"learning_rate": 0.0004981349814685016,
"loss": 5.2583,
"mean_token_accuracy": 0.18822991847991943,
"num_tokens": 10804125.0,
"step": 5870
},
{
"entropy": 5.430789422988892,
"epoch": 0.4907283661877715,
"grad_norm": 0.96484375,
"learning_rate": 0.0004981311544224838,
"loss": 5.307,
"mean_token_accuracy": 0.1887751489877701,
"num_tokens": 10813465.0,
"step": 5875
},
{
"entropy": 5.352855205535889,
"epoch": 0.49114600735048447,
"grad_norm": 0.90625,
"learning_rate": 0.0004981273234702864,
"loss": 5.1316,
"mean_token_accuracy": 0.19545983374118805,
"num_tokens": 10821550.0,
"step": 5880
},
{
"entropy": 5.317874622344971,
"epoch": 0.49156364851319745,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981234886119767,
"loss": 5.1435,
"mean_token_accuracy": 0.19215249121189118,
"num_tokens": 10830683.0,
"step": 5885
},
{
"entropy": 5.393210458755493,
"epoch": 0.4919812896759105,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981196498476217,
"loss": 5.1627,
"mean_token_accuracy": 0.19862073808908462,
"num_tokens": 10839492.0,
"step": 5890
},
{
"entropy": 5.290148782730102,
"epoch": 0.49239893083862346,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981158071772887,
"loss": 5.0956,
"mean_token_accuracy": 0.19917930364608766,
"num_tokens": 10847998.0,
"step": 5895
},
{
"entropy": 5.230513000488282,
"epoch": 0.49281657200133644,
"grad_norm": 1.03125,
"learning_rate": 0.0004981119606010449,
"loss": 5.0791,
"mean_token_accuracy": 0.19867262989282608,
"num_tokens": 10856438.0,
"step": 5900
},
{
"entropy": 5.26281852722168,
"epoch": 0.4932342131640495,
"grad_norm": 0.96875,
"learning_rate": 0.0004981081101189577,
"loss": 5.1077,
"mean_token_accuracy": 0.192747999727726,
"num_tokens": 10865936.0,
"step": 5905
},
{
"entropy": 5.326478052139282,
"epoch": 0.49365185432676245,
"grad_norm": 0.95703125,
"learning_rate": 0.0004981042557310945,
"loss": 5.1064,
"mean_token_accuracy": 0.19702416360378266,
"num_tokens": 10874621.0,
"step": 5910
},
{
"entropy": 5.300664663314819,
"epoch": 0.49406949548947543,
"grad_norm": 0.97265625,
"learning_rate": 0.0004981003974375227,
"loss": 5.0667,
"mean_token_accuracy": 0.19391317218542098,
"num_tokens": 10883516.0,
"step": 5915
},
{
"entropy": 5.364432430267334,
"epoch": 0.49448713665218846,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980965352383099,
"loss": 5.2187,
"mean_token_accuracy": 0.18803293257951736,
"num_tokens": 10892580.0,
"step": 5920
},
{
"entropy": 5.390553092956543,
"epoch": 0.49490477781490144,
"grad_norm": 0.9453125,
"learning_rate": 0.0004980926691335238,
"loss": 5.1395,
"mean_token_accuracy": 0.19519626051187516,
"num_tokens": 10901227.0,
"step": 5925
},
{
"entropy": 5.401871061325073,
"epoch": 0.4953224189776144,
"grad_norm": 0.93359375,
"learning_rate": 0.0004980887991232318,
"loss": 5.203,
"mean_token_accuracy": 0.18491297513246535,
"num_tokens": 10909454.0,
"step": 5930
},
{
"entropy": 5.464483308792114,
"epoch": 0.49574006014032745,
"grad_norm": 0.984375,
"learning_rate": 0.000498084925207502,
"loss": 5.2759,
"mean_token_accuracy": 0.18939323127269744,
"num_tokens": 10919792.0,
"step": 5935
},
{
"entropy": 5.343497657775879,
"epoch": 0.49615770130304043,
"grad_norm": 0.87890625,
"learning_rate": 0.000498081047386402,
"loss": 5.3033,
"mean_token_accuracy": 0.18820173144340516,
"num_tokens": 10930523.0,
"step": 5940
},
{
"entropy": 5.310845947265625,
"epoch": 0.4965753424657534,
"grad_norm": 0.94921875,
"learning_rate": 0.0004980771656599998,
"loss": 5.1732,
"mean_token_accuracy": 0.19256822913885116,
"num_tokens": 10940075.0,
"step": 5945
},
{
"entropy": 5.381923866271973,
"epoch": 0.49699298362846644,
"grad_norm": 1.03125,
"learning_rate": 0.0004980732800283633,
"loss": 5.1638,
"mean_token_accuracy": 0.19154672026634217,
"num_tokens": 10949059.0,
"step": 5950
},
{
"entropy": 5.320337104797363,
"epoch": 0.4974106247911794,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980693904915604,
"loss": 5.1501,
"mean_token_accuracy": 0.19438910484313965,
"num_tokens": 10958313.0,
"step": 5955
},
{
"entropy": 5.336426544189453,
"epoch": 0.4978282659538924,
"grad_norm": 0.94140625,
"learning_rate": 0.0004980654970496595,
"loss": 5.2016,
"mean_token_accuracy": 0.18726001381874086,
"num_tokens": 10967666.0,
"step": 5960
},
{
"entropy": 5.432378578186035,
"epoch": 0.49824590711660544,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980615997027284,
"loss": 5.2318,
"mean_token_accuracy": 0.18868058323860168,
"num_tokens": 10976519.0,
"step": 5965
},
{
"entropy": 5.457491302490235,
"epoch": 0.4986635482793184,
"grad_norm": 0.984375,
"learning_rate": 0.0004980576984508356,
"loss": 5.3376,
"mean_token_accuracy": 0.18108574599027633,
"num_tokens": 10987138.0,
"step": 5970
},
{
"entropy": 5.404650020599365,
"epoch": 0.4990811894420314,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980537932940492,
"loss": 5.2079,
"mean_token_accuracy": 0.19051799178123474,
"num_tokens": 10996517.0,
"step": 5975
},
{
"entropy": 5.357486152648926,
"epoch": 0.4994988306047444,
"grad_norm": 0.9765625,
"learning_rate": 0.0004980498842324377,
"loss": 5.0894,
"mean_token_accuracy": 0.18881412744522094,
"num_tokens": 11006312.0,
"step": 5980
},
{
"entropy": 5.333321905136108,
"epoch": 0.4999164717674574,
"grad_norm": 0.95703125,
"learning_rate": 0.0004980459712660694,
"loss": 5.2052,
"mean_token_accuracy": 0.19695570319890976,
"num_tokens": 11015535.0,
"step": 5985
},
{
"entropy": 5.3520020008087155,
"epoch": 0.5003341129301704,
"grad_norm": 0.953125,
"learning_rate": 0.0004980420543950131,
"loss": 5.1861,
"mean_token_accuracy": 0.19640239030122758,
"num_tokens": 11024887.0,
"step": 5990
},
{
"entropy": 5.249091339111328,
"epoch": 0.5007517540928834,
"grad_norm": 0.95703125,
"learning_rate": 0.0004980381336193371,
"loss": 5.059,
"mean_token_accuracy": 0.20034665763378143,
"num_tokens": 11033147.0,
"step": 5995
},
{
"entropy": 5.33840970993042,
"epoch": 0.5011693952555963,
"grad_norm": 0.8984375,
"learning_rate": 0.0004980342089391099,
"loss": 5.2179,
"mean_token_accuracy": 0.19177668243646623,
"num_tokens": 11042726.0,
"step": 6000
},
{
"epoch": 0.5011693952555963,
"eval_entropy": 5.128798316985059,
"eval_loss": 5.161733627319336,
"eval_mean_token_accuracy": 0.20298875240521067,
"eval_num_tokens": 11042726.0,
"eval_runtime": 20.8409,
"eval_samples_per_second": 1811.965,
"eval_steps_per_second": 226.526,
"step": 6000
},
{
"entropy": 5.392579174041748,
"epoch": 0.5015870364183094,
"grad_norm": 0.87890625,
"learning_rate": 0.0004980302803544006,
"loss": 5.2002,
"mean_token_accuracy": 0.19534078538417815,
"num_tokens": 11052181.0,
"step": 6005
},
{
"entropy": 5.389634037017823,
"epoch": 0.5020046775810224,
"grad_norm": 1.015625,
"learning_rate": 0.0004980263478652778,
"loss": 5.2424,
"mean_token_accuracy": 0.1902779757976532,
"num_tokens": 11061586.0,
"step": 6010
},
{
"entropy": 5.491766691207886,
"epoch": 0.5024223187437353,
"grad_norm": 0.8984375,
"learning_rate": 0.0004980224114718102,
"loss": 5.2533,
"mean_token_accuracy": 0.1943809226155281,
"num_tokens": 11070914.0,
"step": 6015
},
{
"entropy": 5.298935890197754,
"epoch": 0.5028399599064484,
"grad_norm": 0.96484375,
"learning_rate": 0.0004980184711740669,
"loss": 5.0781,
"mean_token_accuracy": 0.20595989376306534,
"num_tokens": 11079836.0,
"step": 6020
},
{
"entropy": 5.2567322731018065,
"epoch": 0.5032576010691614,
"grad_norm": 1.046875,
"learning_rate": 0.0004980145269721168,
"loss": 5.1011,
"mean_token_accuracy": 0.19717675745487212,
"num_tokens": 11088320.0,
"step": 6025
},
{
"entropy": 5.363692426681519,
"epoch": 0.5036752422318743,
"grad_norm": 0.94921875,
"learning_rate": 0.000498010578866029,
"loss": 5.2644,
"mean_token_accuracy": 0.18924812078475953,
"num_tokens": 11097803.0,
"step": 6030
},
{
"entropy": 5.472511672973633,
"epoch": 0.5040928833945874,
"grad_norm": 0.9140625,
"learning_rate": 0.0004980066268558726,
"loss": 5.2869,
"mean_token_accuracy": 0.18986473679542543,
"num_tokens": 11107218.0,
"step": 6035
},
{
"entropy": 5.435642051696777,
"epoch": 0.5045105245573004,
"grad_norm": 0.9765625,
"learning_rate": 0.0004980026709417167,
"loss": 5.1853,
"mean_token_accuracy": 0.18860728293657303,
"num_tokens": 11115599.0,
"step": 6040
},
{
"entropy": 5.268574285507202,
"epoch": 0.5049281657200133,
"grad_norm": 1.0625,
"learning_rate": 0.0004979987111236306,
"loss": 5.0732,
"mean_token_accuracy": 0.1984765887260437,
"num_tokens": 11124312.0,
"step": 6045
},
{
"entropy": 5.257491683959961,
"epoch": 0.5053458068827263,
"grad_norm": 0.94921875,
"learning_rate": 0.0004979947474016837,
"loss": 5.1849,
"mean_token_accuracy": 0.19398614764213562,
"num_tokens": 11133117.0,
"step": 6050
},
{
"entropy": 5.480657625198364,
"epoch": 0.5057634480454394,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979907797759453,
"loss": 5.2499,
"mean_token_accuracy": 0.190121790766716,
"num_tokens": 11141972.0,
"step": 6055
},
{
"entropy": 5.488561344146729,
"epoch": 0.5061810892081523,
"grad_norm": 1.0078125,
"learning_rate": 0.0004979868082464849,
"loss": 5.2525,
"mean_token_accuracy": 0.19349999725818634,
"num_tokens": 11151416.0,
"step": 6060
},
{
"entropy": 5.325757694244385,
"epoch": 0.5065987303708653,
"grad_norm": 0.984375,
"learning_rate": 0.0004979828328133719,
"loss": 5.2035,
"mean_token_accuracy": 0.19287971407175064,
"num_tokens": 11160727.0,
"step": 6065
},
{
"entropy": 5.335221290588379,
"epoch": 0.5070163715335784,
"grad_norm": 0.87890625,
"learning_rate": 0.0004979788534766762,
"loss": 5.1618,
"mean_token_accuracy": 0.18678613156080245,
"num_tokens": 11171350.0,
"step": 6070
},
{
"entropy": 5.3660393238067625,
"epoch": 0.5074340126962913,
"grad_norm": 1.078125,
"learning_rate": 0.0004979748702364672,
"loss": 5.1445,
"mean_token_accuracy": 0.19064127057790756,
"num_tokens": 11180233.0,
"step": 6075
},
{
"entropy": 5.3487941265106205,
"epoch": 0.5078516538590043,
"grad_norm": 1.015625,
"learning_rate": 0.0004979708830928147,
"loss": 5.1259,
"mean_token_accuracy": 0.19202678948640822,
"num_tokens": 11189269.0,
"step": 6080
},
{
"entropy": 5.341992282867432,
"epoch": 0.5082692950217174,
"grad_norm": 0.953125,
"learning_rate": 0.0004979668920457884,
"loss": 5.2077,
"mean_token_accuracy": 0.19293300360441207,
"num_tokens": 11199072.0,
"step": 6085
},
{
"entropy": 5.33283953666687,
"epoch": 0.5086869361844303,
"grad_norm": 0.98828125,
"learning_rate": 0.0004979628970954584,
"loss": 5.1182,
"mean_token_accuracy": 0.20168853402137757,
"num_tokens": 11207164.0,
"step": 6090
},
{
"entropy": 5.337658071517945,
"epoch": 0.5091045773471433,
"grad_norm": 0.88671875,
"learning_rate": 0.0004979588982418944,
"loss": 5.1689,
"mean_token_accuracy": 0.18496512472629548,
"num_tokens": 11216557.0,
"step": 6095
},
{
"entropy": 5.495877647399903,
"epoch": 0.5095222185098564,
"grad_norm": 0.90625,
"learning_rate": 0.0004979548954851666,
"loss": 5.3058,
"mean_token_accuracy": 0.1895919069647789,
"num_tokens": 11227277.0,
"step": 6100
},
{
"entropy": 5.3433582305908205,
"epoch": 0.5099398596725693,
"grad_norm": 0.984375,
"learning_rate": 0.000497950888825345,
"loss": 5.1017,
"mean_token_accuracy": 0.20004121214151382,
"num_tokens": 11236226.0,
"step": 6105
},
{
"entropy": 5.340034532546997,
"epoch": 0.5103575008352823,
"grad_norm": 0.88671875,
"learning_rate": 0.0004979468782624996,
"loss": 5.2086,
"mean_token_accuracy": 0.1939321532845497,
"num_tokens": 11245246.0,
"step": 6110
},
{
"entropy": 5.377697515487671,
"epoch": 0.5107751419979953,
"grad_norm": 1.0078125,
"learning_rate": 0.0004979428637967009,
"loss": 5.2689,
"mean_token_accuracy": 0.19204458743333816,
"num_tokens": 11254437.0,
"step": 6115
},
{
"entropy": 5.376149463653564,
"epoch": 0.5111927831607083,
"grad_norm": 1.0546875,
"learning_rate": 0.0004979388454280189,
"loss": 5.1101,
"mean_token_accuracy": 0.19771129190921782,
"num_tokens": 11262413.0,
"step": 6120
},
{
"entropy": 5.363088989257813,
"epoch": 0.5116104243234213,
"grad_norm": 1.046875,
"learning_rate": 0.000497934823156524,
"loss": 5.2988,
"mean_token_accuracy": 0.18558404445648194,
"num_tokens": 11272061.0,
"step": 6125
},
{
"entropy": 5.37300705909729,
"epoch": 0.5120280654861343,
"grad_norm": 0.97265625,
"learning_rate": 0.0004979307969822869,
"loss": 5.1335,
"mean_token_accuracy": 0.19915029108524324,
"num_tokens": 11281623.0,
"step": 6130
},
{
"entropy": 5.315994024276733,
"epoch": 0.5124457066488474,
"grad_norm": 1.0390625,
"learning_rate": 0.0004979267669053776,
"loss": 5.0069,
"mean_token_accuracy": 0.2073518916964531,
"num_tokens": 11290202.0,
"step": 6135
},
{
"entropy": 5.3649908065795895,
"epoch": 0.5128633478115603,
"grad_norm": 0.921875,
"learning_rate": 0.0004979227329258671,
"loss": 5.2741,
"mean_token_accuracy": 0.19212478250265122,
"num_tokens": 11299040.0,
"step": 6140
},
{
"entropy": 5.389776039123535,
"epoch": 0.5132809889742733,
"grad_norm": 1.0703125,
"learning_rate": 0.0004979186950438256,
"loss": 5.1925,
"mean_token_accuracy": 0.19120557904243468,
"num_tokens": 11307938.0,
"step": 6145
},
{
"entropy": 5.35956335067749,
"epoch": 0.5136986301369864,
"grad_norm": 1.015625,
"learning_rate": 0.0004979146532593243,
"loss": 5.1901,
"mean_token_accuracy": 0.1918809175491333,
"num_tokens": 11316915.0,
"step": 6150
},
{
"entropy": 5.390135335922241,
"epoch": 0.5141162712996993,
"grad_norm": 0.8671875,
"learning_rate": 0.0004979106075724334,
"loss": 5.1488,
"mean_token_accuracy": 0.1926642596721649,
"num_tokens": 11326634.0,
"step": 6155
},
{
"entropy": 5.47369532585144,
"epoch": 0.5145339124624123,
"grad_norm": 1.015625,
"learning_rate": 0.0004979065579832242,
"loss": 5.3053,
"mean_token_accuracy": 0.18129904568195343,
"num_tokens": 11336502.0,
"step": 6160
},
{
"entropy": 5.359502744674683,
"epoch": 0.5149515536251253,
"grad_norm": 1.03125,
"learning_rate": 0.0004979025044917673,
"loss": 5.2211,
"mean_token_accuracy": 0.18862180709838866,
"num_tokens": 11346125.0,
"step": 6165
},
{
"entropy": 5.447209692001342,
"epoch": 0.5153691947878383,
"grad_norm": 0.96484375,
"learning_rate": 0.0004978984470981339,
"loss": 5.3212,
"mean_token_accuracy": 0.1809088706970215,
"num_tokens": 11355713.0,
"step": 6170
},
{
"entropy": 5.377956771850586,
"epoch": 0.5157868359505513,
"grad_norm": 1.1015625,
"learning_rate": 0.0004978943858023946,
"loss": 5.1436,
"mean_token_accuracy": 0.19031920284032822,
"num_tokens": 11364726.0,
"step": 6175
},
{
"entropy": 5.355219125747681,
"epoch": 0.5162044771132643,
"grad_norm": 0.96484375,
"learning_rate": 0.000497890320604621,
"loss": 5.1311,
"mean_token_accuracy": 0.19517320841550828,
"num_tokens": 11373779.0,
"step": 6180
},
{
"entropy": 5.271116304397583,
"epoch": 0.5166221182759773,
"grad_norm": 0.98828125,
"learning_rate": 0.000497886251504884,
"loss": 5.1376,
"mean_token_accuracy": 0.19319186806678773,
"num_tokens": 11382641.0,
"step": 6185
},
{
"entropy": 5.326046276092529,
"epoch": 0.5170397594386903,
"grad_norm": 0.98046875,
"learning_rate": 0.0004978821785032549,
"loss": 5.1059,
"mean_token_accuracy": 0.19505345970392227,
"num_tokens": 11392200.0,
"step": 6190
},
{
"entropy": 5.399134016036987,
"epoch": 0.5174574006014033,
"grad_norm": 1.0078125,
"learning_rate": 0.0004978781015998048,
"loss": 5.2194,
"mean_token_accuracy": 0.18324108570814132,
"num_tokens": 11401884.0,
"step": 6195
},
{
"entropy": 5.401180648803711,
"epoch": 0.5178750417641163,
"grad_norm": 0.9453125,
"learning_rate": 0.0004978740207946056,
"loss": 5.1155,
"mean_token_accuracy": 0.20493513345718384,
"num_tokens": 11410710.0,
"step": 6200
},
{
"entropy": 5.283224391937256,
"epoch": 0.5182926829268293,
"grad_norm": 0.90625,
"learning_rate": 0.0004978699360877281,
"loss": 5.0792,
"mean_token_accuracy": 0.20470337867736815,
"num_tokens": 11419622.0,
"step": 6205
},
{
"entropy": 5.2955403327941895,
"epoch": 0.5187103240895423,
"grad_norm": 0.96484375,
"learning_rate": 0.0004978658474792442,
"loss": 5.0935,
"mean_token_accuracy": 0.20088884085416794,
"num_tokens": 11427631.0,
"step": 6210
},
{
"entropy": 5.344830656051636,
"epoch": 0.5191279652522552,
"grad_norm": 0.9765625,
"learning_rate": 0.0004978617549692253,
"loss": 5.1592,
"mean_token_accuracy": 0.19466407001018524,
"num_tokens": 11437630.0,
"step": 6215
},
{
"entropy": 5.470110177993774,
"epoch": 0.5195456064149683,
"grad_norm": 0.875,
"learning_rate": 0.0004978576585577432,
"loss": 5.3574,
"mean_token_accuracy": 0.17846949845552446,
"num_tokens": 11447962.0,
"step": 6220
},
{
"entropy": 5.456324243545533,
"epoch": 0.5199632475776813,
"grad_norm": 0.90234375,
"learning_rate": 0.0004978535582448695,
"loss": 5.2236,
"mean_token_accuracy": 0.1870883971452713,
"num_tokens": 11458154.0,
"step": 6225
},
{
"entropy": 5.301339769363404,
"epoch": 0.5203808887403942,
"grad_norm": 0.9609375,
"learning_rate": 0.0004978494540306761,
"loss": 5.063,
"mean_token_accuracy": 0.20015534162521362,
"num_tokens": 11467449.0,
"step": 6230
},
{
"entropy": 5.369253540039063,
"epoch": 0.5207985299031073,
"grad_norm": 0.9609375,
"learning_rate": 0.0004978453459152346,
"loss": 5.3532,
"mean_token_accuracy": 0.1816184863448143,
"num_tokens": 11477775.0,
"step": 6235
},
{
"entropy": 5.388203430175781,
"epoch": 0.5212161710658203,
"grad_norm": 0.96875,
"learning_rate": 0.0004978412338986172,
"loss": 5.1043,
"mean_token_accuracy": 0.19701416194438934,
"num_tokens": 11486921.0,
"step": 6240
},
{
"entropy": 5.292043542861938,
"epoch": 0.5216338122285332,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978371179808958,
"loss": 5.1337,
"mean_token_accuracy": 0.19857128262519835,
"num_tokens": 11496562.0,
"step": 6245
},
{
"entropy": 5.339953660964966,
"epoch": 0.5220514533912463,
"grad_norm": 1.046875,
"learning_rate": 0.0004978329981621424,
"loss": 5.0972,
"mean_token_accuracy": 0.19187704026699065,
"num_tokens": 11504906.0,
"step": 6250
},
{
"entropy": 5.2793288230896,
"epoch": 0.5224690945539593,
"grad_norm": 0.94921875,
"learning_rate": 0.000497828874442429,
"loss": 5.1473,
"mean_token_accuracy": 0.1895140916109085,
"num_tokens": 11513757.0,
"step": 6255
},
{
"entropy": 5.297783422470093,
"epoch": 0.5228867357166722,
"grad_norm": 0.90625,
"learning_rate": 0.000497824746821828,
"loss": 5.1575,
"mean_token_accuracy": 0.19044381976127625,
"num_tokens": 11523195.0,
"step": 6260
},
{
"entropy": 5.4000497341156,
"epoch": 0.5233043768793852,
"grad_norm": 0.99609375,
"learning_rate": 0.0004978206153004117,
"loss": 5.2021,
"mean_token_accuracy": 0.1967186450958252,
"num_tokens": 11532348.0,
"step": 6265
},
{
"entropy": 5.387604427337647,
"epoch": 0.5237220180420983,
"grad_norm": 0.99609375,
"learning_rate": 0.0004978164798782523,
"loss": 5.1932,
"mean_token_accuracy": 0.19440121650695802,
"num_tokens": 11542356.0,
"step": 6270
},
{
"entropy": 5.340060567855835,
"epoch": 0.5241396592048112,
"grad_norm": 0.94140625,
"learning_rate": 0.0004978123405554221,
"loss": 5.1859,
"mean_token_accuracy": 0.1988328918814659,
"num_tokens": 11550914.0,
"step": 6275
},
{
"entropy": 5.370529270172119,
"epoch": 0.5245573003675242,
"grad_norm": 1.0,
"learning_rate": 0.0004978081973319938,
"loss": 5.1451,
"mean_token_accuracy": 0.19870750308036805,
"num_tokens": 11559314.0,
"step": 6280
},
{
"entropy": 5.313170957565307,
"epoch": 0.5249749415302373,
"grad_norm": 0.9765625,
"learning_rate": 0.0004978040502080398,
"loss": 5.118,
"mean_token_accuracy": 0.19605566263198854,
"num_tokens": 11567990.0,
"step": 6285
},
{
"entropy": 5.331792831420898,
"epoch": 0.5253925826929502,
"grad_norm": 0.91796875,
"learning_rate": 0.0004977998991836327,
"loss": 5.1423,
"mean_token_accuracy": 0.19811130315065384,
"num_tokens": 11577133.0,
"step": 6290
},
{
"entropy": 5.367706727981568,
"epoch": 0.5258102238556632,
"grad_norm": 1.046875,
"learning_rate": 0.0004977957442588451,
"loss": 5.2204,
"mean_token_accuracy": 0.1907789871096611,
"num_tokens": 11585330.0,
"step": 6295
},
{
"entropy": 5.439990043640137,
"epoch": 0.5262278650183763,
"grad_norm": 0.9921875,
"learning_rate": 0.0004977915854337498,
"loss": 5.1716,
"mean_token_accuracy": 0.1921030357480049,
"num_tokens": 11595005.0,
"step": 6300
},
{
"entropy": 5.291791105270386,
"epoch": 0.5266455061810892,
"grad_norm": 0.953125,
"learning_rate": 0.0004977874227084197,
"loss": 5.068,
"mean_token_accuracy": 0.20003140717744827,
"num_tokens": 11603969.0,
"step": 6305
},
{
"entropy": 5.287006568908692,
"epoch": 0.5270631473438022,
"grad_norm": 0.91796875,
"learning_rate": 0.0004977832560829277,
"loss": 5.1375,
"mean_token_accuracy": 0.19020287543535233,
"num_tokens": 11613118.0,
"step": 6310
},
{
"entropy": 5.351668500900269,
"epoch": 0.5274807885065153,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977790855573464,
"loss": 5.0607,
"mean_token_accuracy": 0.2026047885417938,
"num_tokens": 11621799.0,
"step": 6315
},
{
"entropy": 5.283554363250732,
"epoch": 0.5278984296692282,
"grad_norm": 0.95703125,
"learning_rate": 0.0004977749111317492,
"loss": 5.1484,
"mean_token_accuracy": 0.19796689748764038,
"num_tokens": 11630557.0,
"step": 6320
},
{
"entropy": 5.3494213104248045,
"epoch": 0.5283160708319412,
"grad_norm": 0.98046875,
"learning_rate": 0.000497770732806209,
"loss": 5.1706,
"mean_token_accuracy": 0.198416106402874,
"num_tokens": 11639179.0,
"step": 6325
},
{
"entropy": 5.408968734741211,
"epoch": 0.5287337119946542,
"grad_norm": 1.140625,
"learning_rate": 0.000497766550580799,
"loss": 5.2689,
"mean_token_accuracy": 0.18869854658842086,
"num_tokens": 11648273.0,
"step": 6330
},
{
"entropy": 5.348643159866333,
"epoch": 0.5291513531573672,
"grad_norm": 0.94921875,
"learning_rate": 0.0004977623644555923,
"loss": 5.1808,
"mean_token_accuracy": 0.18637329041957856,
"num_tokens": 11656372.0,
"step": 6335
},
{
"entropy": 5.365545225143433,
"epoch": 0.5295689943200802,
"grad_norm": 0.98828125,
"learning_rate": 0.0004977581744306624,
"loss": 5.1185,
"mean_token_accuracy": 0.20012644678354263,
"num_tokens": 11665007.0,
"step": 6340
},
{
"entropy": 5.370207166671753,
"epoch": 0.5299866354827932,
"grad_norm": 0.91015625,
"learning_rate": 0.0004977539805060824,
"loss": 5.2293,
"mean_token_accuracy": 0.19181876629590988,
"num_tokens": 11674519.0,
"step": 6345
},
{
"entropy": 5.401744079589844,
"epoch": 0.5304042766455062,
"grad_norm": 0.91015625,
"learning_rate": 0.0004977497826819259,
"loss": 5.1526,
"mean_token_accuracy": 0.19267868399620056,
"num_tokens": 11683705.0,
"step": 6350
},
{
"entropy": 5.315210485458374,
"epoch": 0.5308219178082192,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977455809582664,
"loss": 5.1177,
"mean_token_accuracy": 0.19677025228738784,
"num_tokens": 11693152.0,
"step": 6355
},
{
"entropy": 5.332680559158325,
"epoch": 0.5312395589709322,
"grad_norm": 0.9765625,
"learning_rate": 0.0004977413753351772,
"loss": 5.1507,
"mean_token_accuracy": 0.18915897011756896,
"num_tokens": 11702584.0,
"step": 6360
},
{
"entropy": 5.32446551322937,
"epoch": 0.5316572001336451,
"grad_norm": 1.0078125,
"learning_rate": 0.0004977371658127323,
"loss": 5.1913,
"mean_token_accuracy": 0.19093988090753555,
"num_tokens": 11711633.0,
"step": 6365
},
{
"entropy": 5.349480628967285,
"epoch": 0.5320748412963582,
"grad_norm": 0.953125,
"learning_rate": 0.0004977329523910051,
"loss": 5.1371,
"mean_token_accuracy": 0.19278361648321152,
"num_tokens": 11720116.0,
"step": 6370
},
{
"entropy": 5.4168623924255375,
"epoch": 0.5324924824590712,
"grad_norm": 0.9609375,
"learning_rate": 0.0004977287350700695,
"loss": 5.2351,
"mean_token_accuracy": 0.19008343368768693,
"num_tokens": 11729564.0,
"step": 6375
},
{
"entropy": 5.3386941909790036,
"epoch": 0.5329101236217841,
"grad_norm": 1.0390625,
"learning_rate": 0.0004977245138499993,
"loss": 5.095,
"mean_token_accuracy": 0.2036321446299553,
"num_tokens": 11737935.0,
"step": 6380
},
{
"entropy": 5.314829730987549,
"epoch": 0.5333277647844972,
"grad_norm": 0.9609375,
"learning_rate": 0.0004977202887308684,
"loss": 5.071,
"mean_token_accuracy": 0.19733900874853133,
"num_tokens": 11745989.0,
"step": 6385
},
{
"entropy": 5.331277656555176,
"epoch": 0.5337454059472102,
"grad_norm": 1.1328125,
"learning_rate": 0.0004977160597127507,
"loss": 5.2443,
"mean_token_accuracy": 0.1911775141954422,
"num_tokens": 11755530.0,
"step": 6390
},
{
"entropy": 5.357501792907715,
"epoch": 0.5341630471099231,
"grad_norm": 0.9765625,
"learning_rate": 0.0004977118267957204,
"loss": 5.198,
"mean_token_accuracy": 0.18840063512325286,
"num_tokens": 11764546.0,
"step": 6395
},
{
"entropy": 5.450428247451782,
"epoch": 0.5345806882726362,
"grad_norm": 0.97265625,
"learning_rate": 0.0004977075899798513,
"loss": 5.3313,
"mean_token_accuracy": 0.18829478472471237,
"num_tokens": 11774117.0,
"step": 6400
},
{
"entropy": 5.377129030227661,
"epoch": 0.5349983294353492,
"grad_norm": 0.94921875,
"learning_rate": 0.000497703349265218,
"loss": 5.1008,
"mean_token_accuracy": 0.20352084934711456,
"num_tokens": 11782952.0,
"step": 6405
},
{
"entropy": 5.3871293544769285,
"epoch": 0.5354159705980621,
"grad_norm": 1.0625,
"learning_rate": 0.0004976991046518943,
"loss": 5.2743,
"mean_token_accuracy": 0.1843955248594284,
"num_tokens": 11792369.0,
"step": 6410
},
{
"entropy": 5.359866094589234,
"epoch": 0.5358336117607752,
"grad_norm": 0.9375,
"learning_rate": 0.0004976948561399549,
"loss": 5.1126,
"mean_token_accuracy": 0.1998010128736496,
"num_tokens": 11801365.0,
"step": 6415
},
{
"entropy": 5.386413145065307,
"epoch": 0.5362512529234882,
"grad_norm": 0.97265625,
"learning_rate": 0.0004976906037294738,
"loss": 5.1722,
"mean_token_accuracy": 0.19516139179468156,
"num_tokens": 11809785.0,
"step": 6420
},
{
"entropy": 5.317812061309814,
"epoch": 0.5366688940862011,
"grad_norm": 1.0,
"learning_rate": 0.0004976863474205257,
"loss": 5.0701,
"mean_token_accuracy": 0.19765079468488694,
"num_tokens": 11817978.0,
"step": 6425
},
{
"entropy": 5.354518747329712,
"epoch": 0.5370865352489141,
"grad_norm": 1.0234375,
"learning_rate": 0.0004976820872131851,
"loss": 5.1852,
"mean_token_accuracy": 0.1935555264353752,
"num_tokens": 11827796.0,
"step": 6430
},
{
"entropy": 5.261296987533569,
"epoch": 0.5375041764116272,
"grad_norm": 1.046875,
"learning_rate": 0.0004976778231075266,
"loss": 5.0466,
"mean_token_accuracy": 0.19626055508852006,
"num_tokens": 11836439.0,
"step": 6435
},
{
"entropy": 5.366777753829956,
"epoch": 0.5379218175743401,
"grad_norm": 0.9921875,
"learning_rate": 0.0004976735551036245,
"loss": 5.248,
"mean_token_accuracy": 0.1846526250243187,
"num_tokens": 11845268.0,
"step": 6440
},
{
"entropy": 5.414008569717407,
"epoch": 0.5383394587370531,
"grad_norm": 0.89453125,
"learning_rate": 0.000497669283201554,
"loss": 5.1656,
"mean_token_accuracy": 0.19413827061653138,
"num_tokens": 11855185.0,
"step": 6445
},
{
"entropy": 5.306928062438965,
"epoch": 0.5387570998997662,
"grad_norm": 1.0078125,
"learning_rate": 0.0004976650074013895,
"loss": 5.2031,
"mean_token_accuracy": 0.1889431208372116,
"num_tokens": 11865073.0,
"step": 6450
},
{
"entropy": 5.353112697601318,
"epoch": 0.5391747410624791,
"grad_norm": 0.8515625,
"learning_rate": 0.0004976607277032061,
"loss": 5.1885,
"mean_token_accuracy": 0.19165299087762833,
"num_tokens": 11875004.0,
"step": 6455
},
{
"entropy": 5.39293532371521,
"epoch": 0.5395923822251921,
"grad_norm": 0.93359375,
"learning_rate": 0.0004976564441070788,
"loss": 5.1839,
"mean_token_accuracy": 0.18849599063396455,
"num_tokens": 11884905.0,
"step": 6460
},
{
"entropy": 5.385581302642822,
"epoch": 0.5400100233879052,
"grad_norm": 0.93359375,
"learning_rate": 0.0004976521566130823,
"loss": 5.1747,
"mean_token_accuracy": 0.18683718591928483,
"num_tokens": 11894669.0,
"step": 6465
},
{
"entropy": 5.468428373336792,
"epoch": 0.5404276645506181,
"grad_norm": 0.93359375,
"learning_rate": 0.0004976478652212918,
"loss": 5.3486,
"mean_token_accuracy": 0.18105758726596832,
"num_tokens": 11904127.0,
"step": 6470
},
{
"entropy": 5.39544448852539,
"epoch": 0.5408453057133311,
"grad_norm": 0.9921875,
"learning_rate": 0.0004976435699317824,
"loss": 5.1839,
"mean_token_accuracy": 0.1931723266839981,
"num_tokens": 11912031.0,
"step": 6475
},
{
"entropy": 5.286249876022339,
"epoch": 0.5412629468760441,
"grad_norm": 0.99609375,
"learning_rate": 0.0004976392707446293,
"loss": 5.091,
"mean_token_accuracy": 0.20129090398550034,
"num_tokens": 11920961.0,
"step": 6480
},
{
"entropy": 5.430025577545166,
"epoch": 0.5416805880387571,
"grad_norm": 0.984375,
"learning_rate": 0.0004976349676599078,
"loss": 5.2408,
"mean_token_accuracy": 0.19328033179044724,
"num_tokens": 11930454.0,
"step": 6485
},
{
"entropy": 5.287836027145386,
"epoch": 0.5420982292014701,
"grad_norm": 0.984375,
"learning_rate": 0.0004976306606776932,
"loss": 5.0443,
"mean_token_accuracy": 0.19630483388900757,
"num_tokens": 11938610.0,
"step": 6490
},
{
"entropy": 5.32088189125061,
"epoch": 0.5425158703641831,
"grad_norm": 1.0,
"learning_rate": 0.0004976263497980609,
"loss": 5.2076,
"mean_token_accuracy": 0.190990050137043,
"num_tokens": 11947882.0,
"step": 6495
},
{
"entropy": 5.345630741119384,
"epoch": 0.5429335115268961,
"grad_norm": 1.046875,
"learning_rate": 0.0004976220350210864,
"loss": 5.1993,
"mean_token_accuracy": 0.19136472791433334,
"num_tokens": 11957552.0,
"step": 6500
},
{
"entropy": 5.347573947906494,
"epoch": 0.5433511526896091,
"grad_norm": 0.94921875,
"learning_rate": 0.0004976177163468451,
"loss": 5.1188,
"mean_token_accuracy": 0.19805283546447755,
"num_tokens": 11967350.0,
"step": 6505
},
{
"entropy": 5.318794822692871,
"epoch": 0.5437687938523221,
"grad_norm": 0.9375,
"learning_rate": 0.0004976133937754128,
"loss": 5.1315,
"mean_token_accuracy": 0.19716765284538268,
"num_tokens": 11976107.0,
"step": 6510
},
{
"entropy": 5.346179676055908,
"epoch": 0.544186435015035,
"grad_norm": 1.03125,
"learning_rate": 0.0004976090673068651,
"loss": 5.1353,
"mean_token_accuracy": 0.2032754361629486,
"num_tokens": 11985059.0,
"step": 6515
},
{
"entropy": 5.351248455047608,
"epoch": 0.5446040761777481,
"grad_norm": 1.046875,
"learning_rate": 0.0004976047369412778,
"loss": 5.1053,
"mean_token_accuracy": 0.19469528198242186,
"num_tokens": 11993453.0,
"step": 6520
},
{
"entropy": 5.209239292144775,
"epoch": 0.5450217173404611,
"grad_norm": 1.015625,
"learning_rate": 0.0004976004026787266,
"loss": 5.078,
"mean_token_accuracy": 0.1995821103453636,
"num_tokens": 12002524.0,
"step": 6525
},
{
"entropy": 5.27763819694519,
"epoch": 0.545439358503174,
"grad_norm": 1.03125,
"learning_rate": 0.0004975960645192873,
"loss": 5.1339,
"mean_token_accuracy": 0.20042127519845962,
"num_tokens": 12011516.0,
"step": 6530
},
{
"entropy": 5.38931794166565,
"epoch": 0.5458569996658871,
"grad_norm": 0.96484375,
"learning_rate": 0.0004975917224630359,
"loss": 5.1692,
"mean_token_accuracy": 0.1899636819958687,
"num_tokens": 12020058.0,
"step": 6535
},
{
"entropy": 5.391335487365723,
"epoch": 0.5462746408286001,
"grad_norm": 0.91015625,
"learning_rate": 0.0004975873765100487,
"loss": 5.2737,
"mean_token_accuracy": 0.19032340496778488,
"num_tokens": 12029167.0,
"step": 6540
},
{
"entropy": 5.414868116378784,
"epoch": 0.546692281991313,
"grad_norm": 0.921875,
"learning_rate": 0.0004975830266604013,
"loss": 5.2169,
"mean_token_accuracy": 0.19321169704198837,
"num_tokens": 12038298.0,
"step": 6545
},
{
"entropy": 5.3901842594146725,
"epoch": 0.5471099231540261,
"grad_norm": 0.9453125,
"learning_rate": 0.0004975786729141702,
"loss": 5.142,
"mean_token_accuracy": 0.1915355607867241,
"num_tokens": 12047437.0,
"step": 6550
},
{
"entropy": 5.343628978729248,
"epoch": 0.5475275643167391,
"grad_norm": 1.03125,
"learning_rate": 0.0004975743152714314,
"loss": 5.1977,
"mean_token_accuracy": 0.1963498279452324,
"num_tokens": 12055821.0,
"step": 6555
},
{
"entropy": 5.354809141159057,
"epoch": 0.547945205479452,
"grad_norm": 0.98046875,
"learning_rate": 0.0004975699537322616,
"loss": 5.071,
"mean_token_accuracy": 0.20276693105697632,
"num_tokens": 12064378.0,
"step": 6560
},
{
"entropy": 5.294686222076416,
"epoch": 0.5483628466421651,
"grad_norm": 1.0,
"learning_rate": 0.0004975655882967366,
"loss": 5.1261,
"mean_token_accuracy": 0.19267136305570604,
"num_tokens": 12073110.0,
"step": 6565
},
{
"entropy": 5.349828910827637,
"epoch": 0.5487804878048781,
"grad_norm": 0.953125,
"learning_rate": 0.0004975612189649331,
"loss": 5.1567,
"mean_token_accuracy": 0.19478155225515364,
"num_tokens": 12081224.0,
"step": 6570
},
{
"entropy": 5.3753687858581545,
"epoch": 0.549198128967591,
"grad_norm": 0.91796875,
"learning_rate": 0.0004975568457369276,
"loss": 5.2122,
"mean_token_accuracy": 0.19452133178710937,
"num_tokens": 12090279.0,
"step": 6575
},
{
"entropy": 5.371964836120606,
"epoch": 0.549615770130304,
"grad_norm": 1.03125,
"learning_rate": 0.0004975524686127966,
"loss": 5.1921,
"mean_token_accuracy": 0.19338359087705612,
"num_tokens": 12098818.0,
"step": 6580
},
{
"entropy": 5.327710771560669,
"epoch": 0.5500334112930171,
"grad_norm": 1.1484375,
"learning_rate": 0.0004975480875926168,
"loss": 5.1468,
"mean_token_accuracy": 0.20131338834762574,
"num_tokens": 12107358.0,
"step": 6585
},
{
"entropy": 5.347161960601807,
"epoch": 0.55045105245573,
"grad_norm": 0.90625,
"learning_rate": 0.0004975437026764648,
"loss": 5.2072,
"mean_token_accuracy": 0.18581592291593552,
"num_tokens": 12116802.0,
"step": 6590
},
{
"entropy": 5.388274431228638,
"epoch": 0.550868693618443,
"grad_norm": 0.890625,
"learning_rate": 0.0004975393138644175,
"loss": 5.13,
"mean_token_accuracy": 0.19624195247888565,
"num_tokens": 12126347.0,
"step": 6595
},
{
"entropy": 5.266183519363404,
"epoch": 0.5512863347811561,
"grad_norm": 0.9765625,
"learning_rate": 0.0004975349211565515,
"loss": 5.0938,
"mean_token_accuracy": 0.19580602645874023,
"num_tokens": 12136138.0,
"step": 6600
},
{
"entropy": 5.254367828369141,
"epoch": 0.551703975943869,
"grad_norm": 0.91015625,
"learning_rate": 0.000497530524552944,
"loss": 5.1656,
"mean_token_accuracy": 0.19330661594867707,
"num_tokens": 12146436.0,
"step": 6605
},
{
"entropy": 5.370379114151001,
"epoch": 0.552121617106582,
"grad_norm": 0.85546875,
"learning_rate": 0.0004975261240536716,
"loss": 5.178,
"mean_token_accuracy": 0.19154346436262132,
"num_tokens": 12157003.0,
"step": 6610
},
{
"entropy": 5.356114482879638,
"epoch": 0.5525392582692951,
"grad_norm": 0.87890625,
"learning_rate": 0.0004975217196588117,
"loss": 5.108,
"mean_token_accuracy": 0.2029719516634941,
"num_tokens": 12166671.0,
"step": 6615
},
{
"entropy": 5.289737892150879,
"epoch": 0.552956899432008,
"grad_norm": 1.1015625,
"learning_rate": 0.0004975173113684413,
"loss": 5.1529,
"mean_token_accuracy": 0.1971219703555107,
"num_tokens": 12176474.0,
"step": 6620
},
{
"entropy": 5.346948766708374,
"epoch": 0.553374540594721,
"grad_norm": 1.0546875,
"learning_rate": 0.0004975128991826374,
"loss": 5.1435,
"mean_token_accuracy": 0.19717600494623183,
"num_tokens": 12185307.0,
"step": 6625
},
{
"entropy": 5.376376533508301,
"epoch": 0.553792181757434,
"grad_norm": 0.9609375,
"learning_rate": 0.0004975084831014774,
"loss": 5.108,
"mean_token_accuracy": 0.1940117746591568,
"num_tokens": 12195179.0,
"step": 6630
},
{
"entropy": 5.342315626144409,
"epoch": 0.554209822920147,
"grad_norm": 1.03125,
"learning_rate": 0.0004975040631250386,
"loss": 5.1737,
"mean_token_accuracy": 0.20250627696514129,
"num_tokens": 12204083.0,
"step": 6635
},
{
"entropy": 5.2787923336029055,
"epoch": 0.55462746408286,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974996392533983,
"loss": 5.1052,
"mean_token_accuracy": 0.20077962279319764,
"num_tokens": 12213322.0,
"step": 6640
},
{
"entropy": 5.339088201522827,
"epoch": 0.555045105245573,
"grad_norm": 1.03125,
"learning_rate": 0.000497495211486634,
"loss": 5.102,
"mean_token_accuracy": 0.20010723322629928,
"num_tokens": 12222305.0,
"step": 6645
},
{
"entropy": 5.3519268989562985,
"epoch": 0.555462746408286,
"grad_norm": 1.0390625,
"learning_rate": 0.0004974907798248232,
"loss": 5.2078,
"mean_token_accuracy": 0.1935238003730774,
"num_tokens": 12232196.0,
"step": 6650
},
{
"entropy": 5.34990348815918,
"epoch": 0.555880387570999,
"grad_norm": 1.0,
"learning_rate": 0.0004974863442680434,
"loss": 5.1518,
"mean_token_accuracy": 0.19610499292612077,
"num_tokens": 12241914.0,
"step": 6655
},
{
"entropy": 5.208635902404785,
"epoch": 0.556298028733712,
"grad_norm": 0.8671875,
"learning_rate": 0.0004974819048163725,
"loss": 5.0471,
"mean_token_accuracy": 0.20544617623090744,
"num_tokens": 12250185.0,
"step": 6660
},
{
"entropy": 5.3693469047546385,
"epoch": 0.556715669896425,
"grad_norm": 0.97265625,
"learning_rate": 0.000497477461469888,
"loss": 5.2177,
"mean_token_accuracy": 0.18970588743686675,
"num_tokens": 12258954.0,
"step": 6665
},
{
"entropy": 5.409700918197632,
"epoch": 0.557133311059138,
"grad_norm": 0.95703125,
"learning_rate": 0.0004974730142286677,
"loss": 5.2277,
"mean_token_accuracy": 0.18751125782728195,
"num_tokens": 12268320.0,
"step": 6670
},
{
"entropy": 5.318266916275024,
"epoch": 0.557550952221851,
"grad_norm": 1.0390625,
"learning_rate": 0.0004974685630927895,
"loss": 5.1118,
"mean_token_accuracy": 0.19236804842948912,
"num_tokens": 12276804.0,
"step": 6675
},
{
"entropy": 5.417097473144532,
"epoch": 0.557968593384564,
"grad_norm": 0.96875,
"learning_rate": 0.0004974641080623314,
"loss": 5.1564,
"mean_token_accuracy": 0.1895575553178787,
"num_tokens": 12286024.0,
"step": 6680
},
{
"entropy": 5.3917840957641605,
"epoch": 0.558386234547277,
"grad_norm": 0.90234375,
"learning_rate": 0.0004974596491373711,
"loss": 5.1908,
"mean_token_accuracy": 0.19468486905097962,
"num_tokens": 12295154.0,
"step": 6685
},
{
"entropy": 5.379201126098633,
"epoch": 0.55880387570999,
"grad_norm": 0.8671875,
"learning_rate": 0.0004974551863179871,
"loss": 5.2481,
"mean_token_accuracy": 0.18427381217479705,
"num_tokens": 12306349.0,
"step": 6690
},
{
"entropy": 5.379152107238769,
"epoch": 0.5592215168727029,
"grad_norm": 1.0859375,
"learning_rate": 0.0004974507196042572,
"loss": 5.1937,
"mean_token_accuracy": 0.19431248158216477,
"num_tokens": 12315496.0,
"step": 6695
},
{
"entropy": 5.2494100570678714,
"epoch": 0.559639158035416,
"grad_norm": 0.94921875,
"learning_rate": 0.0004974462489962597,
"loss": 5.0365,
"mean_token_accuracy": 0.19687401950359346,
"num_tokens": 12325070.0,
"step": 6700
},
{
"entropy": 5.340914011001587,
"epoch": 0.560056799198129,
"grad_norm": 1.015625,
"learning_rate": 0.0004974417744940729,
"loss": 5.1279,
"mean_token_accuracy": 0.1945287764072418,
"num_tokens": 12333509.0,
"step": 6705
},
{
"entropy": 5.373744249343872,
"epoch": 0.5604744403608419,
"grad_norm": 0.99609375,
"learning_rate": 0.000497437296097775,
"loss": 5.2111,
"mean_token_accuracy": 0.18823633939027787,
"num_tokens": 12342495.0,
"step": 6710
},
{
"entropy": 5.3024890422821045,
"epoch": 0.560892081523555,
"grad_norm": 0.9140625,
"learning_rate": 0.0004974328138074445,
"loss": 5.1165,
"mean_token_accuracy": 0.19562546163797379,
"num_tokens": 12351560.0,
"step": 6715
},
{
"entropy": 5.305520725250244,
"epoch": 0.561309722686268,
"grad_norm": 1.0703125,
"learning_rate": 0.0004974283276231599,
"loss": 5.2097,
"mean_token_accuracy": 0.1911316066980362,
"num_tokens": 12360773.0,
"step": 6720
},
{
"entropy": 5.376407289505005,
"epoch": 0.5617273638489809,
"grad_norm": 0.89453125,
"learning_rate": 0.0004974238375449996,
"loss": 5.2198,
"mean_token_accuracy": 0.18474843502044677,
"num_tokens": 12370519.0,
"step": 6725
},
{
"entropy": 5.372508716583252,
"epoch": 0.562145005011694,
"grad_norm": 0.91015625,
"learning_rate": 0.0004974193435730424,
"loss": 5.21,
"mean_token_accuracy": 0.19076197743415832,
"num_tokens": 12380563.0,
"step": 6730
},
{
"entropy": 5.387130212783814,
"epoch": 0.562562646174407,
"grad_norm": 1.1015625,
"learning_rate": 0.0004974148457073668,
"loss": 5.208,
"mean_token_accuracy": 0.19575879722833633,
"num_tokens": 12389291.0,
"step": 6735
},
{
"entropy": 5.327639245986939,
"epoch": 0.5629802873371199,
"grad_norm": 1.015625,
"learning_rate": 0.0004974103439480517,
"loss": 5.1789,
"mean_token_accuracy": 0.19355615079402924,
"num_tokens": 12399037.0,
"step": 6740
},
{
"entropy": 5.3151123046875,
"epoch": 0.563397928499833,
"grad_norm": 0.98046875,
"learning_rate": 0.0004974058382951757,
"loss": 4.9895,
"mean_token_accuracy": 0.20109709799289704,
"num_tokens": 12407553.0,
"step": 6745
},
{
"entropy": 5.22309947013855,
"epoch": 0.563815569662546,
"grad_norm": 1.046875,
"learning_rate": 0.0004974013287488177,
"loss": 5.073,
"mean_token_accuracy": 0.19609258472919464,
"num_tokens": 12416298.0,
"step": 6750
},
{
"entropy": 5.367431354522705,
"epoch": 0.5642332108252589,
"grad_norm": 0.953125,
"learning_rate": 0.0004973968153090569,
"loss": 5.2276,
"mean_token_accuracy": 0.186616013944149,
"num_tokens": 12425918.0,
"step": 6755
},
{
"entropy": 5.379483604431153,
"epoch": 0.5646508519879719,
"grad_norm": 1.0078125,
"learning_rate": 0.0004973922979759722,
"loss": 5.1491,
"mean_token_accuracy": 0.1927480846643448,
"num_tokens": 12434693.0,
"step": 6760
},
{
"entropy": 5.240261459350586,
"epoch": 0.565068493150685,
"grad_norm": 0.93359375,
"learning_rate": 0.0004973877767496424,
"loss": 5.0175,
"mean_token_accuracy": 0.20052199065685272,
"num_tokens": 12444055.0,
"step": 6765
},
{
"entropy": 5.277566432952881,
"epoch": 0.5654861343133979,
"grad_norm": 0.98046875,
"learning_rate": 0.0004973832516301472,
"loss": 5.0904,
"mean_token_accuracy": 0.20363982915878295,
"num_tokens": 12453012.0,
"step": 6770
},
{
"entropy": 5.274981260299683,
"epoch": 0.5659037754761109,
"grad_norm": 1.0078125,
"learning_rate": 0.0004973787226175652,
"loss": 5.1325,
"mean_token_accuracy": 0.19542792290449143,
"num_tokens": 12462292.0,
"step": 6775
},
{
"entropy": 5.278493928909302,
"epoch": 0.566321416638824,
"grad_norm": 0.94921875,
"learning_rate": 0.0004973741897119761,
"loss": 5.0874,
"mean_token_accuracy": 0.20897582918405533,
"num_tokens": 12471178.0,
"step": 6780
},
{
"entropy": 5.306678676605225,
"epoch": 0.5667390578015369,
"grad_norm": 0.9609375,
"learning_rate": 0.0004973696529134592,
"loss": 5.1201,
"mean_token_accuracy": 0.2028688222169876,
"num_tokens": 12479977.0,
"step": 6785
},
{
"entropy": 5.307732725143433,
"epoch": 0.5671566989642499,
"grad_norm": 1.0078125,
"learning_rate": 0.0004973651122220938,
"loss": 5.1888,
"mean_token_accuracy": 0.19187885969877244,
"num_tokens": 12489004.0,
"step": 6790
},
{
"entropy": 5.419384527206421,
"epoch": 0.567574340126963,
"grad_norm": 0.92578125,
"learning_rate": 0.0004973605676379595,
"loss": 5.1949,
"mean_token_accuracy": 0.1894410789012909,
"num_tokens": 12498979.0,
"step": 6795
},
{
"entropy": 5.382754135131836,
"epoch": 0.5679919812896759,
"grad_norm": 1.0234375,
"learning_rate": 0.0004973560191611357,
"loss": 5.2518,
"mean_token_accuracy": 0.18889162242412566,
"num_tokens": 12508293.0,
"step": 6800
},
{
"entropy": 5.430076217651367,
"epoch": 0.5684096224523889,
"grad_norm": 1.125,
"learning_rate": 0.0004973514667917021,
"loss": 5.2651,
"mean_token_accuracy": 0.190037102997303,
"num_tokens": 12517828.0,
"step": 6805
},
{
"entropy": 5.3846653461456295,
"epoch": 0.5688272636151019,
"grad_norm": 0.953125,
"learning_rate": 0.0004973469105297386,
"loss": 5.1187,
"mean_token_accuracy": 0.1927657827734947,
"num_tokens": 12526569.0,
"step": 6810
},
{
"entropy": 5.353836727142334,
"epoch": 0.5692449047778149,
"grad_norm": 0.90234375,
"learning_rate": 0.0004973423503753247,
"loss": 5.2411,
"mean_token_accuracy": 0.19148507863283157,
"num_tokens": 12535706.0,
"step": 6815
},
{
"entropy": 5.389388847351074,
"epoch": 0.5696625459405279,
"grad_norm": 0.890625,
"learning_rate": 0.0004973377863285403,
"loss": 5.118,
"mean_token_accuracy": 0.18832767903804778,
"num_tokens": 12544689.0,
"step": 6820
},
{
"entropy": 5.286836814880371,
"epoch": 0.5700801871032409,
"grad_norm": 0.921875,
"learning_rate": 0.0004973332183894653,
"loss": 5.1008,
"mean_token_accuracy": 0.1946028009057045,
"num_tokens": 12553318.0,
"step": 6825
},
{
"entropy": 5.24740743637085,
"epoch": 0.5704978282659539,
"grad_norm": 0.99609375,
"learning_rate": 0.0004973286465581798,
"loss": 5.0801,
"mean_token_accuracy": 0.20608279556035997,
"num_tokens": 12562009.0,
"step": 6830
},
{
"entropy": 5.379718351364136,
"epoch": 0.5709154694286669,
"grad_norm": 0.94140625,
"learning_rate": 0.0004973240708347637,
"loss": 5.2263,
"mean_token_accuracy": 0.19497342258691788,
"num_tokens": 12571043.0,
"step": 6835
},
{
"entropy": 5.451855421066284,
"epoch": 0.5713331105913799,
"grad_norm": 0.88671875,
"learning_rate": 0.0004973194912192971,
"loss": 5.1315,
"mean_token_accuracy": 0.194309738278389,
"num_tokens": 12581151.0,
"step": 6840
},
{
"entropy": 5.36621675491333,
"epoch": 0.5717507517540928,
"grad_norm": 0.86328125,
"learning_rate": 0.0004973149077118602,
"loss": 5.1751,
"mean_token_accuracy": 0.19538725912570953,
"num_tokens": 12590671.0,
"step": 6845
},
{
"entropy": 5.27402811050415,
"epoch": 0.5721683929168059,
"grad_norm": 0.94921875,
"learning_rate": 0.0004973103203125333,
"loss": 5.2161,
"mean_token_accuracy": 0.19579629749059677,
"num_tokens": 12599693.0,
"step": 6850
},
{
"entropy": 5.327305889129638,
"epoch": 0.5725860340795189,
"grad_norm": 0.921875,
"learning_rate": 0.0004973057290213965,
"loss": 5.1252,
"mean_token_accuracy": 0.19855290353298188,
"num_tokens": 12609238.0,
"step": 6855
},
{
"entropy": 5.358370113372803,
"epoch": 0.5730036752422318,
"grad_norm": 0.92578125,
"learning_rate": 0.0004973011338385304,
"loss": 5.1266,
"mean_token_accuracy": 0.19303234070539474,
"num_tokens": 12618273.0,
"step": 6860
},
{
"entropy": 5.3325108051300045,
"epoch": 0.5734213164049449,
"grad_norm": 0.96484375,
"learning_rate": 0.0004972965347640154,
"loss": 5.1805,
"mean_token_accuracy": 0.193036849796772,
"num_tokens": 12628229.0,
"step": 6865
},
{
"entropy": 5.366177606582641,
"epoch": 0.5738389575676579,
"grad_norm": 1.0390625,
"learning_rate": 0.000497291931797932,
"loss": 5.095,
"mean_token_accuracy": 0.1982719898223877,
"num_tokens": 12636533.0,
"step": 6870
},
{
"entropy": 5.293973970413208,
"epoch": 0.5742565987303708,
"grad_norm": 1.0078125,
"learning_rate": 0.0004972873249403608,
"loss": 5.1133,
"mean_token_accuracy": 0.19605183601379395,
"num_tokens": 12645503.0,
"step": 6875
},
{
"entropy": 5.338414239883423,
"epoch": 0.5746742398930839,
"grad_norm": 0.953125,
"learning_rate": 0.0004972827141913822,
"loss": 5.1332,
"mean_token_accuracy": 0.18842328786849977,
"num_tokens": 12654355.0,
"step": 6880
},
{
"entropy": 5.332372665405273,
"epoch": 0.5750918810557969,
"grad_norm": 1.0,
"learning_rate": 0.0004972780995510773,
"loss": 5.1072,
"mean_token_accuracy": 0.1970501646399498,
"num_tokens": 12663774.0,
"step": 6885
},
{
"entropy": 5.250775575637817,
"epoch": 0.5755095222185098,
"grad_norm": 0.90234375,
"learning_rate": 0.0004972734810195266,
"loss": 5.1357,
"mean_token_accuracy": 0.1956734225153923,
"num_tokens": 12674436.0,
"step": 6890
},
{
"entropy": 5.357981634140015,
"epoch": 0.5759271633812229,
"grad_norm": 1.0234375,
"learning_rate": 0.0004972688585968112,
"loss": 5.0935,
"mean_token_accuracy": 0.1939712405204773,
"num_tokens": 12682891.0,
"step": 6895
},
{
"entropy": 5.250308465957642,
"epoch": 0.5763448045439359,
"grad_norm": 0.97265625,
"learning_rate": 0.0004972642322830118,
"loss": 5.0049,
"mean_token_accuracy": 0.20868447721004485,
"num_tokens": 12691170.0,
"step": 6900
},
{
"entropy": 5.268120145797729,
"epoch": 0.5767624457066488,
"grad_norm": 1.0625,
"learning_rate": 0.0004972596020782096,
"loss": 5.0892,
"mean_token_accuracy": 0.20168329030275345,
"num_tokens": 12700314.0,
"step": 6905
},
{
"entropy": 5.301498794555664,
"epoch": 0.5771800868693618,
"grad_norm": 1.0,
"learning_rate": 0.0004972549679824853,
"loss": 5.1515,
"mean_token_accuracy": 0.1980636477470398,
"num_tokens": 12709418.0,
"step": 6910
},
{
"entropy": 5.239022207260132,
"epoch": 0.5775977280320749,
"grad_norm": 0.8984375,
"learning_rate": 0.0004972503299959204,
"loss": 5.1328,
"mean_token_accuracy": 0.18934394270181656,
"num_tokens": 12719287.0,
"step": 6915
},
{
"entropy": 5.342194938659668,
"epoch": 0.5780153691947878,
"grad_norm": 0.9453125,
"learning_rate": 0.000497245688118596,
"loss": 5.0579,
"mean_token_accuracy": 0.20184237957000734,
"num_tokens": 12728836.0,
"step": 6920
},
{
"entropy": 5.377489757537842,
"epoch": 0.5784330103575008,
"grad_norm": 0.96875,
"learning_rate": 0.0004972410423505932,
"loss": 5.1802,
"mean_token_accuracy": 0.1930708572268486,
"num_tokens": 12738173.0,
"step": 6925
},
{
"entropy": 5.328310346603393,
"epoch": 0.5788506515202139,
"grad_norm": 0.96484375,
"learning_rate": 0.0004972363926919935,
"loss": 5.1658,
"mean_token_accuracy": 0.20007263422012328,
"num_tokens": 12747428.0,
"step": 6930
},
{
"entropy": 5.356612300872802,
"epoch": 0.5792682926829268,
"grad_norm": 0.97265625,
"learning_rate": 0.0004972317391428782,
"loss": 5.226,
"mean_token_accuracy": 0.18761636763811113,
"num_tokens": 12757148.0,
"step": 6935
},
{
"entropy": 5.403679609298706,
"epoch": 0.5796859338456398,
"grad_norm": 0.9296875,
"learning_rate": 0.0004972270817033288,
"loss": 5.2318,
"mean_token_accuracy": 0.18417499214410782,
"num_tokens": 12766820.0,
"step": 6940
},
{
"entropy": 5.388335800170898,
"epoch": 0.5801035750083529,
"grad_norm": 0.96875,
"learning_rate": 0.0004972224203734269,
"loss": 5.2165,
"mean_token_accuracy": 0.19299497902393342,
"num_tokens": 12776082.0,
"step": 6945
},
{
"entropy": 5.345056629180908,
"epoch": 0.5805212161710658,
"grad_norm": 0.9921875,
"learning_rate": 0.000497217755153254,
"loss": 5.1883,
"mean_token_accuracy": 0.19662619531154632,
"num_tokens": 12785057.0,
"step": 6950
},
{
"entropy": 5.330507612228393,
"epoch": 0.5809388573337788,
"grad_norm": 0.9921875,
"learning_rate": 0.0004972130860428918,
"loss": 5.1729,
"mean_token_accuracy": 0.19887613505125046,
"num_tokens": 12793599.0,
"step": 6955
},
{
"entropy": 5.362395000457764,
"epoch": 0.5813564984964918,
"grad_norm": 0.98828125,
"learning_rate": 0.0004972084130424222,
"loss": 5.1677,
"mean_token_accuracy": 0.19218032956123351,
"num_tokens": 12803775.0,
"step": 6960
},
{
"entropy": 5.3146649360656735,
"epoch": 0.5817741396592048,
"grad_norm": 1.0859375,
"learning_rate": 0.0004972037361519267,
"loss": 5.1115,
"mean_token_accuracy": 0.19620238095521927,
"num_tokens": 12812327.0,
"step": 6965
},
{
"entropy": 5.352201986312866,
"epoch": 0.5821917808219178,
"grad_norm": 0.88671875,
"learning_rate": 0.0004971990553714874,
"loss": 5.0605,
"mean_token_accuracy": 0.20081085860729217,
"num_tokens": 12821218.0,
"step": 6970
},
{
"entropy": 5.356791019439697,
"epoch": 0.5826094219846308,
"grad_norm": 0.9296875,
"learning_rate": 0.0004971943707011861,
"loss": 5.1505,
"mean_token_accuracy": 0.1940755397081375,
"num_tokens": 12831579.0,
"step": 6975
},
{
"entropy": 5.359289741516113,
"epoch": 0.5830270631473438,
"grad_norm": 0.95703125,
"learning_rate": 0.000497189682141105,
"loss": 5.2162,
"mean_token_accuracy": 0.19426080733537673,
"num_tokens": 12841026.0,
"step": 6980
},
{
"entropy": 5.25170931816101,
"epoch": 0.5834447043100568,
"grad_norm": 0.953125,
"learning_rate": 0.0004971849896913259,
"loss": 5.1494,
"mean_token_accuracy": 0.19563014209270477,
"num_tokens": 12850136.0,
"step": 6985
},
{
"entropy": 5.354822397232056,
"epoch": 0.5838623454727698,
"grad_norm": 1.046875,
"learning_rate": 0.0004971802933519312,
"loss": 5.2256,
"mean_token_accuracy": 0.19553592205047607,
"num_tokens": 12860345.0,
"step": 6990
},
{
"entropy": 5.304988527297974,
"epoch": 0.5842799866354828,
"grad_norm": 1.015625,
"learning_rate": 0.0004971755931230031,
"loss": 5.1553,
"mean_token_accuracy": 0.19274741113185884,
"num_tokens": 12868995.0,
"step": 6995
},
{
"entropy": 5.414120864868164,
"epoch": 0.5846976277981958,
"grad_norm": 0.94921875,
"learning_rate": 0.0004971708890046236,
"loss": 5.3066,
"mean_token_accuracy": 0.18866612911224365,
"num_tokens": 12878043.0,
"step": 7000
},
{
"entropy": 5.332791137695312,
"epoch": 0.5851152689609088,
"grad_norm": 0.98046875,
"learning_rate": 0.0004971661809968755,
"loss": 5.1266,
"mean_token_accuracy": 0.19653817415237426,
"num_tokens": 12887090.0,
"step": 7005
},
{
"entropy": 5.29463038444519,
"epoch": 0.5855329101236217,
"grad_norm": 1.0078125,
"learning_rate": 0.0004971614690998409,
"loss": 5.0655,
"mean_token_accuracy": 0.20885752886533737,
"num_tokens": 12896349.0,
"step": 7010
},
{
"entropy": 5.308234930038452,
"epoch": 0.5859505512863348,
"grad_norm": 1.015625,
"learning_rate": 0.0004971567533136024,
"loss": 5.1114,
"mean_token_accuracy": 0.19684289246797562,
"num_tokens": 12905698.0,
"step": 7015
},
{
"entropy": 5.248376989364624,
"epoch": 0.5863681924490478,
"grad_norm": 0.9765625,
"learning_rate": 0.0004971520336382424,
"loss": 5.1162,
"mean_token_accuracy": 0.20329161435365678,
"num_tokens": 12915146.0,
"step": 7020
},
{
"entropy": 5.191306734085083,
"epoch": 0.5867858336117607,
"grad_norm": 0.94921875,
"learning_rate": 0.0004971473100738437,
"loss": 4.9775,
"mean_token_accuracy": 0.20541905164718627,
"num_tokens": 12924933.0,
"step": 7025
},
{
"entropy": 5.363535928726196,
"epoch": 0.5872034747744738,
"grad_norm": 0.90234375,
"learning_rate": 0.000497142582620489,
"loss": 5.2124,
"mean_token_accuracy": 0.18779776245355606,
"num_tokens": 12934959.0,
"step": 7030
},
{
"entropy": 5.307815980911255,
"epoch": 0.5876211159371868,
"grad_norm": 1.03125,
"learning_rate": 0.0004971378512782609,
"loss": 5.0625,
"mean_token_accuracy": 0.20597047507762908,
"num_tokens": 12943470.0,
"step": 7035
},
{
"entropy": 5.2257318019866945,
"epoch": 0.5880387570998997,
"grad_norm": 0.9609375,
"learning_rate": 0.0004971331160472424,
"loss": 5.0695,
"mean_token_accuracy": 0.20476973056793213,
"num_tokens": 12953592.0,
"step": 7040
},
{
"entropy": 5.3533648490905765,
"epoch": 0.5884563982626128,
"grad_norm": 1.0546875,
"learning_rate": 0.0004971283769275163,
"loss": 5.1954,
"mean_token_accuracy": 0.1878223314881325,
"num_tokens": 12962893.0,
"step": 7045
},
{
"entropy": 5.265210437774658,
"epoch": 0.5888740394253258,
"grad_norm": 0.9765625,
"learning_rate": 0.0004971236339191656,
"loss": 5.082,
"mean_token_accuracy": 0.20315028429031373,
"num_tokens": 12971635.0,
"step": 7050
},
{
"entropy": 5.2263952732086185,
"epoch": 0.5892916805880387,
"grad_norm": 0.89453125,
"learning_rate": 0.0004971188870222734,
"loss": 5.0587,
"mean_token_accuracy": 0.20553121119737625,
"num_tokens": 12981040.0,
"step": 7055
},
{
"entropy": 5.2849376678466795,
"epoch": 0.5897093217507517,
"grad_norm": 0.953125,
"learning_rate": 0.0004971141362369226,
"loss": 5.0789,
"mean_token_accuracy": 0.1994106352329254,
"num_tokens": 12989898.0,
"step": 7060
},
{
"entropy": 5.298730897903442,
"epoch": 0.5901269629134648,
"grad_norm": 0.96484375,
"learning_rate": 0.0004971093815631965,
"loss": 5.0723,
"mean_token_accuracy": 0.19711526930332185,
"num_tokens": 12998584.0,
"step": 7065
},
{
"entropy": 5.2975461959838865,
"epoch": 0.5905446040761777,
"grad_norm": 1.0234375,
"learning_rate": 0.0004971046230011783,
"loss": 5.107,
"mean_token_accuracy": 0.19629897624254228,
"num_tokens": 13008007.0,
"step": 7070
},
{
"entropy": 5.342353105545044,
"epoch": 0.5909622452388907,
"grad_norm": 0.8515625,
"learning_rate": 0.0004970998605509514,
"loss": 5.2364,
"mean_token_accuracy": 0.18871966153383254,
"num_tokens": 13017034.0,
"step": 7075
},
{
"entropy": 5.407427501678467,
"epoch": 0.5913798864016038,
"grad_norm": 1.046875,
"learning_rate": 0.000497095094212599,
"loss": 5.2997,
"mean_token_accuracy": 0.1838369131088257,
"num_tokens": 13025916.0,
"step": 7080
},
{
"entropy": 5.391162633895874,
"epoch": 0.5917975275643167,
"grad_norm": 0.99609375,
"learning_rate": 0.0004970903239862047,
"loss": 5.1711,
"mean_token_accuracy": 0.18868904262781144,
"num_tokens": 13035242.0,
"step": 7085
},
{
"entropy": 5.3094288349151615,
"epoch": 0.5922151687270297,
"grad_norm": 1.0546875,
"learning_rate": 0.0004970855498718519,
"loss": 5.1051,
"mean_token_accuracy": 0.19548780769109725,
"num_tokens": 13044123.0,
"step": 7090
},
{
"entropy": 5.2993566513061525,
"epoch": 0.5926328098897428,
"grad_norm": 0.9609375,
"learning_rate": 0.0004970807718696243,
"loss": 5.1624,
"mean_token_accuracy": 0.19107201099395751,
"num_tokens": 13053503.0,
"step": 7095
},
{
"entropy": 5.342015790939331,
"epoch": 0.5930504510524557,
"grad_norm": 0.99609375,
"learning_rate": 0.0004970759899796055,
"loss": 5.1496,
"mean_token_accuracy": 0.19394615441560745,
"num_tokens": 13062106.0,
"step": 7100
},
{
"entropy": 5.3626923084259035,
"epoch": 0.5934680922151687,
"grad_norm": 1.0234375,
"learning_rate": 0.0004970712042018791,
"loss": 5.1425,
"mean_token_accuracy": 0.19150077253580094,
"num_tokens": 13071509.0,
"step": 7105
},
{
"entropy": 5.263015222549439,
"epoch": 0.5938857333778818,
"grad_norm": 0.984375,
"learning_rate": 0.0004970664145365289,
"loss": 5.0702,
"mean_token_accuracy": 0.1999054193496704,
"num_tokens": 13080469.0,
"step": 7110
},
{
"entropy": 5.317249155044555,
"epoch": 0.5943033745405947,
"grad_norm": 0.94921875,
"learning_rate": 0.000497061620983639,
"loss": 5.0394,
"mean_token_accuracy": 0.20194881409406662,
"num_tokens": 13089826.0,
"step": 7115
},
{
"entropy": 5.369497966766358,
"epoch": 0.5947210157033077,
"grad_norm": 1.09375,
"learning_rate": 0.0004970568235432931,
"loss": 5.2196,
"mean_token_accuracy": 0.19345467537641525,
"num_tokens": 13098630.0,
"step": 7120
},
{
"entropy": 5.316605472564698,
"epoch": 0.5951386568660207,
"grad_norm": 1.1640625,
"learning_rate": 0.0004970520222155751,
"loss": 5.1198,
"mean_token_accuracy": 0.19804002493619918,
"num_tokens": 13106819.0,
"step": 7125
},
{
"entropy": 5.367663240432739,
"epoch": 0.5955562980287337,
"grad_norm": 1.015625,
"learning_rate": 0.0004970472170005692,
"loss": 5.1313,
"mean_token_accuracy": 0.19699766784906386,
"num_tokens": 13115471.0,
"step": 7130
},
{
"entropy": 5.2743888854980465,
"epoch": 0.5959739391914467,
"grad_norm": 0.953125,
"learning_rate": 0.0004970424078983596,
"loss": 5.0054,
"mean_token_accuracy": 0.209335957467556,
"num_tokens": 13124537.0,
"step": 7135
},
{
"entropy": 5.292069292068481,
"epoch": 0.5963915803541597,
"grad_norm": 0.81640625,
"learning_rate": 0.0004970375949090302,
"loss": 5.1436,
"mean_token_accuracy": 0.19772982895374297,
"num_tokens": 13135326.0,
"step": 7140
},
{
"entropy": 5.417243432998657,
"epoch": 0.5968092215168727,
"grad_norm": 0.8984375,
"learning_rate": 0.0004970327780326655,
"loss": 5.2354,
"mean_token_accuracy": 0.18968381881713867,
"num_tokens": 13145502.0,
"step": 7145
},
{
"entropy": 5.387653493881226,
"epoch": 0.5972268626795857,
"grad_norm": 0.984375,
"learning_rate": 0.0004970279572693498,
"loss": 5.146,
"mean_token_accuracy": 0.20050851851701737,
"num_tokens": 13154808.0,
"step": 7150
},
{
"entropy": 5.259207057952881,
"epoch": 0.5976445038422987,
"grad_norm": 0.96875,
"learning_rate": 0.0004970231326191674,
"loss": 5.2493,
"mean_token_accuracy": 0.18623579889535904,
"num_tokens": 13163982.0,
"step": 7155
},
{
"entropy": 5.332884931564331,
"epoch": 0.5980621450050116,
"grad_norm": 0.91015625,
"learning_rate": 0.0004970183040822029,
"loss": 5.1401,
"mean_token_accuracy": 0.196553036570549,
"num_tokens": 13172446.0,
"step": 7160
},
{
"entropy": 5.365352439880371,
"epoch": 0.5984797861677247,
"grad_norm": 0.94921875,
"learning_rate": 0.0004970134716585407,
"loss": 5.1137,
"mean_token_accuracy": 0.19348438531160356,
"num_tokens": 13181671.0,
"step": 7165
},
{
"entropy": 5.304043531417847,
"epoch": 0.5988974273304377,
"grad_norm": 1.0546875,
"learning_rate": 0.0004970086353482654,
"loss": 5.0449,
"mean_token_accuracy": 0.19625666588544846,
"num_tokens": 13190207.0,
"step": 7170
},
{
"entropy": 5.273567962646484,
"epoch": 0.5993150684931506,
"grad_norm": 0.98828125,
"learning_rate": 0.0004970037951514617,
"loss": 5.0443,
"mean_token_accuracy": 0.20113227665424346,
"num_tokens": 13198866.0,
"step": 7175
},
{
"entropy": 5.301345729827881,
"epoch": 0.5997327096558637,
"grad_norm": 0.984375,
"learning_rate": 0.0004969989510682143,
"loss": 5.0595,
"mean_token_accuracy": 0.2027854800224304,
"num_tokens": 13207921.0,
"step": 7180
},
{
"entropy": 5.399423408508301,
"epoch": 0.6001503508185767,
"grad_norm": 0.96484375,
"learning_rate": 0.0004969941030986082,
"loss": 5.2385,
"mean_token_accuracy": 0.19331772476434708,
"num_tokens": 13217759.0,
"step": 7185
},
{
"entropy": 5.2552680492401125,
"epoch": 0.6005679919812896,
"grad_norm": 0.92578125,
"learning_rate": 0.000496989251242728,
"loss": 5.0112,
"mean_token_accuracy": 0.20147017538547515,
"num_tokens": 13226592.0,
"step": 7190
},
{
"entropy": 5.282441425323486,
"epoch": 0.6009856331440027,
"grad_norm": 1.0078125,
"learning_rate": 0.0004969843955006588,
"loss": 5.0846,
"mean_token_accuracy": 0.1955637276172638,
"num_tokens": 13236123.0,
"step": 7195
},
{
"entropy": 5.317075157165528,
"epoch": 0.6014032743067157,
"grad_norm": 1.0234375,
"learning_rate": 0.0004969795358724855,
"loss": 5.1894,
"mean_token_accuracy": 0.19167284220457076,
"num_tokens": 13246169.0,
"step": 7200
},
{
"entropy": 5.282715368270874,
"epoch": 0.6018209154694286,
"grad_norm": 0.97265625,
"learning_rate": 0.0004969746723582932,
"loss": 5.068,
"mean_token_accuracy": 0.20560144633054733,
"num_tokens": 13256141.0,
"step": 7205
},
{
"entropy": 5.357795715332031,
"epoch": 0.6022385566321417,
"grad_norm": 0.953125,
"learning_rate": 0.0004969698049581672,
"loss": 5.2254,
"mean_token_accuracy": 0.19470075219869615,
"num_tokens": 13265491.0,
"step": 7210
},
{
"entropy": 5.408079290390015,
"epoch": 0.6026561977948547,
"grad_norm": 0.96875,
"learning_rate": 0.0004969649336721924,
"loss": 5.2272,
"mean_token_accuracy": 0.19919759333133696,
"num_tokens": 13274805.0,
"step": 7215
},
{
"entropy": 5.410361862182617,
"epoch": 0.6030738389575676,
"grad_norm": 1.015625,
"learning_rate": 0.0004969600585004544,
"loss": 5.1821,
"mean_token_accuracy": 0.19047997444868087,
"num_tokens": 13284589.0,
"step": 7220
},
{
"entropy": 5.255741548538208,
"epoch": 0.6034914801202806,
"grad_norm": 1.0234375,
"learning_rate": 0.0004969551794430383,
"loss": 5.0199,
"mean_token_accuracy": 0.20712951868772506,
"num_tokens": 13293370.0,
"step": 7225
},
{
"entropy": 5.398855876922608,
"epoch": 0.6039091212829937,
"grad_norm": 0.953125,
"learning_rate": 0.0004969502965000297,
"loss": 5.2001,
"mean_token_accuracy": 0.19466075748205186,
"num_tokens": 13302555.0,
"step": 7230
},
{
"entropy": 5.287164735794067,
"epoch": 0.6043267624457066,
"grad_norm": 0.94921875,
"learning_rate": 0.000496945409671514,
"loss": 5.1207,
"mean_token_accuracy": 0.20137713700532914,
"num_tokens": 13311632.0,
"step": 7235
},
{
"entropy": 5.319089078903199,
"epoch": 0.6047444036084196,
"grad_norm": 0.90625,
"learning_rate": 0.0004969405189575768,
"loss": 5.0785,
"mean_token_accuracy": 0.2026840031147003,
"num_tokens": 13320593.0,
"step": 7240
},
{
"entropy": 5.294135904312133,
"epoch": 0.6051620447711327,
"grad_norm": 1.015625,
"learning_rate": 0.0004969356243583036,
"loss": 5.0401,
"mean_token_accuracy": 0.20420426577329637,
"num_tokens": 13329095.0,
"step": 7245
},
{
"entropy": 5.234756422042847,
"epoch": 0.6055796859338456,
"grad_norm": 0.94140625,
"learning_rate": 0.0004969307258737802,
"loss": 5.0958,
"mean_token_accuracy": 0.19960410743951798,
"num_tokens": 13338404.0,
"step": 7250
},
{
"entropy": 5.270444250106811,
"epoch": 0.6059973270965586,
"grad_norm": 1.0390625,
"learning_rate": 0.0004969258235040921,
"loss": 5.0669,
"mean_token_accuracy": 0.20691566914319992,
"num_tokens": 13347095.0,
"step": 7255
},
{
"entropy": 5.289247989654541,
"epoch": 0.6064149682592717,
"grad_norm": 1.0,
"learning_rate": 0.0004969209172493255,
"loss": 5.13,
"mean_token_accuracy": 0.19697306752204896,
"num_tokens": 13356606.0,
"step": 7260
},
{
"entropy": 5.319506454467773,
"epoch": 0.6068326094219846,
"grad_norm": 0.8984375,
"learning_rate": 0.0004969160071095661,
"loss": 5.1284,
"mean_token_accuracy": 0.19501465409994126,
"num_tokens": 13366889.0,
"step": 7265
},
{
"entropy": 5.345353364944458,
"epoch": 0.6072502505846976,
"grad_norm": 0.95703125,
"learning_rate": 0.0004969110930848999,
"loss": 5.1381,
"mean_token_accuracy": 0.1954762563109398,
"num_tokens": 13376761.0,
"step": 7270
},
{
"entropy": 5.314459133148193,
"epoch": 0.6076678917474106,
"grad_norm": 0.95703125,
"learning_rate": 0.0004969061751754129,
"loss": 5.0622,
"mean_token_accuracy": 0.20645569264888763,
"num_tokens": 13386226.0,
"step": 7275
},
{
"entropy": 5.266441774368286,
"epoch": 0.6080855329101236,
"grad_norm": 1.0703125,
"learning_rate": 0.0004969012533811911,
"loss": 5.0074,
"mean_token_accuracy": 0.20572338849306107,
"num_tokens": 13394243.0,
"step": 7280
},
{
"entropy": 5.287345218658447,
"epoch": 0.6085031740728366,
"grad_norm": 0.95703125,
"learning_rate": 0.0004968963277023208,
"loss": 5.2853,
"mean_token_accuracy": 0.17963769435882568,
"num_tokens": 13403708.0,
"step": 7285
},
{
"entropy": 5.274597883224487,
"epoch": 0.6089208152355496,
"grad_norm": 0.96484375,
"learning_rate": 0.0004968913981388882,
"loss": 5.1087,
"mean_token_accuracy": 0.20476820915937424,
"num_tokens": 13412911.0,
"step": 7290
},
{
"entropy": 5.380892419815064,
"epoch": 0.6093384563982626,
"grad_norm": 0.92578125,
"learning_rate": 0.0004968864646909797,
"loss": 5.2419,
"mean_token_accuracy": 0.19346756786108016,
"num_tokens": 13422423.0,
"step": 7295
},
{
"entropy": 5.321874284744263,
"epoch": 0.6097560975609756,
"grad_norm": 0.8359375,
"learning_rate": 0.0004968815273586814,
"loss": 5.0971,
"mean_token_accuracy": 0.2003725379705429,
"num_tokens": 13430984.0,
"step": 7300
},
{
"entropy": 5.312367868423462,
"epoch": 0.6101737387236886,
"grad_norm": 0.93359375,
"learning_rate": 0.0004968765861420799,
"loss": 5.1513,
"mean_token_accuracy": 0.1950658842921257,
"num_tokens": 13440492.0,
"step": 7305
},
{
"entropy": 5.3808534145355225,
"epoch": 0.6105913798864016,
"grad_norm": 1.03125,
"learning_rate": 0.0004968716410412617,
"loss": 5.1133,
"mean_token_accuracy": 0.19727630466222762,
"num_tokens": 13449222.0,
"step": 7310
},
{
"entropy": 5.310233306884766,
"epoch": 0.6110090210491146,
"grad_norm": 0.98828125,
"learning_rate": 0.0004968666920563134,
"loss": 5.1067,
"mean_token_accuracy": 0.20144483149051667,
"num_tokens": 13459126.0,
"step": 7315
},
{
"entropy": 5.434738111495972,
"epoch": 0.6114266622118276,
"grad_norm": 0.90234375,
"learning_rate": 0.0004968617391873216,
"loss": 5.2465,
"mean_token_accuracy": 0.1796099439263344,
"num_tokens": 13468811.0,
"step": 7320
},
{
"entropy": 5.33213701248169,
"epoch": 0.6118443033745405,
"grad_norm": 0.90625,
"learning_rate": 0.0004968567824343729,
"loss": 5.04,
"mean_token_accuracy": 0.1932415708899498,
"num_tokens": 13477765.0,
"step": 7325
},
{
"entropy": 5.318994760513306,
"epoch": 0.6122619445372536,
"grad_norm": 1.0234375,
"learning_rate": 0.0004968518217975542,
"loss": 5.1748,
"mean_token_accuracy": 0.18547997921705245,
"num_tokens": 13487184.0,
"step": 7330
},
{
"entropy": 5.3060770511627195,
"epoch": 0.6126795856999666,
"grad_norm": 0.9375,
"learning_rate": 0.0004968468572769524,
"loss": 5.1203,
"mean_token_accuracy": 0.19739150553941726,
"num_tokens": 13496784.0,
"step": 7335
},
{
"entropy": 5.322006368637085,
"epoch": 0.6130972268626795,
"grad_norm": 1.0,
"learning_rate": 0.0004968418888726543,
"loss": 5.1565,
"mean_token_accuracy": 0.19425925463438035,
"num_tokens": 13505824.0,
"step": 7340
},
{
"entropy": 5.299488306045532,
"epoch": 0.6135148680253926,
"grad_norm": 1.125,
"learning_rate": 0.0004968369165847469,
"loss": 5.0041,
"mean_token_accuracy": 0.20222590416669844,
"num_tokens": 13515111.0,
"step": 7345
},
{
"entropy": 5.253536128997803,
"epoch": 0.6139325091881056,
"grad_norm": 0.9140625,
"learning_rate": 0.0004968319404133173,
"loss": 5.0382,
"mean_token_accuracy": 0.20265664011240006,
"num_tokens": 13524163.0,
"step": 7350
},
{
"entropy": 5.273534059524536,
"epoch": 0.6143501503508185,
"grad_norm": 1.0625,
"learning_rate": 0.0004968269603584525,
"loss": 5.0173,
"mean_token_accuracy": 0.2090737923979759,
"num_tokens": 13533368.0,
"step": 7355
},
{
"entropy": 5.33686089515686,
"epoch": 0.6147677915135316,
"grad_norm": 0.953125,
"learning_rate": 0.0004968219764202397,
"loss": 5.2172,
"mean_token_accuracy": 0.19302222728729249,
"num_tokens": 13543708.0,
"step": 7360
},
{
"entropy": 5.311557865142822,
"epoch": 0.6151854326762446,
"grad_norm": 0.96484375,
"learning_rate": 0.0004968169885987663,
"loss": 5.1442,
"mean_token_accuracy": 0.19446005374193193,
"num_tokens": 13552561.0,
"step": 7365
},
{
"entropy": 5.411979484558105,
"epoch": 0.6156030738389575,
"grad_norm": 0.96484375,
"learning_rate": 0.0004968119968941193,
"loss": 5.2422,
"mean_token_accuracy": 0.18597899675369262,
"num_tokens": 13562050.0,
"step": 7370
},
{
"entropy": 5.257673883438111,
"epoch": 0.6160207150016705,
"grad_norm": 0.9453125,
"learning_rate": 0.0004968070013063866,
"loss": 5.0711,
"mean_token_accuracy": 0.19194685220718383,
"num_tokens": 13571356.0,
"step": 7375
},
{
"entropy": 5.2729517936706545,
"epoch": 0.6164383561643836,
"grad_norm": 0.875,
"learning_rate": 0.0004968020018356552,
"loss": 5.0731,
"mean_token_accuracy": 0.19190491139888763,
"num_tokens": 13581357.0,
"step": 7380
},
{
"entropy": 5.29729528427124,
"epoch": 0.6168559973270965,
"grad_norm": 0.953125,
"learning_rate": 0.0004967969984820128,
"loss": 5.0621,
"mean_token_accuracy": 0.1971805661916733,
"num_tokens": 13589392.0,
"step": 7385
},
{
"entropy": 5.296608352661133,
"epoch": 0.6172736384898095,
"grad_norm": 0.96875,
"learning_rate": 0.0004967919912455469,
"loss": 5.032,
"mean_token_accuracy": 0.2077103450894356,
"num_tokens": 13598376.0,
"step": 7390
},
{
"entropy": 5.176263618469238,
"epoch": 0.6176912796525226,
"grad_norm": 0.921875,
"learning_rate": 0.0004967869801263454,
"loss": 4.9436,
"mean_token_accuracy": 0.21241846084594726,
"num_tokens": 13607014.0,
"step": 7395
},
{
"entropy": 5.236470127105713,
"epoch": 0.6181089208152355,
"grad_norm": 0.9375,
"learning_rate": 0.0004967819651244957,
"loss": 5.1246,
"mean_token_accuracy": 0.1940328523516655,
"num_tokens": 13615700.0,
"step": 7400
},
{
"entropy": 5.3532181739807125,
"epoch": 0.6185265619779485,
"grad_norm": 0.9609375,
"learning_rate": 0.0004967769462400857,
"loss": 5.1522,
"mean_token_accuracy": 0.2038279429078102,
"num_tokens": 13625230.0,
"step": 7405
},
{
"entropy": 5.44642186164856,
"epoch": 0.6189442031406616,
"grad_norm": 0.89453125,
"learning_rate": 0.0004967719234732033,
"loss": 5.3097,
"mean_token_accuracy": 0.19283871203660966,
"num_tokens": 13635258.0,
"step": 7410
},
{
"entropy": 5.2723753452301025,
"epoch": 0.6193618443033745,
"grad_norm": 1.09375,
"learning_rate": 0.0004967668968239365,
"loss": 5.0181,
"mean_token_accuracy": 0.2040007844567299,
"num_tokens": 13643688.0,
"step": 7415
},
{
"entropy": 5.283078193664551,
"epoch": 0.6197794854660875,
"grad_norm": 0.98046875,
"learning_rate": 0.0004967618662923732,
"loss": 5.1676,
"mean_token_accuracy": 0.1946879133582115,
"num_tokens": 13652894.0,
"step": 7420
},
{
"entropy": 5.291216707229614,
"epoch": 0.6201971266288006,
"grad_norm": 0.9140625,
"learning_rate": 0.0004967568318786016,
"loss": 5.1051,
"mean_token_accuracy": 0.1991724491119385,
"num_tokens": 13662852.0,
"step": 7425
},
{
"entropy": 5.331347799301147,
"epoch": 0.6206147677915135,
"grad_norm": 0.98046875,
"learning_rate": 0.0004967517935827097,
"loss": 5.0826,
"mean_token_accuracy": 0.19448151290416718,
"num_tokens": 13671457.0,
"step": 7430
},
{
"entropy": 5.307916498184204,
"epoch": 0.6210324089542265,
"grad_norm": 0.93359375,
"learning_rate": 0.0004967467514047856,
"loss": 5.1569,
"mean_token_accuracy": 0.19908891916275023,
"num_tokens": 13681002.0,
"step": 7435
},
{
"entropy": 5.293569469451905,
"epoch": 0.6214500501169395,
"grad_norm": 0.92578125,
"learning_rate": 0.0004967417053449178,
"loss": 5.099,
"mean_token_accuracy": 0.19786427021026612,
"num_tokens": 13690677.0,
"step": 7440
},
{
"entropy": 5.272982358932495,
"epoch": 0.6218676912796525,
"grad_norm": 1.015625,
"learning_rate": 0.0004967366554031945,
"loss": 4.9354,
"mean_token_accuracy": 0.20864570438861846,
"num_tokens": 13699151.0,
"step": 7445
},
{
"entropy": 5.291828203201294,
"epoch": 0.6222853324423655,
"grad_norm": 1.0078125,
"learning_rate": 0.0004967316015797042,
"loss": 5.1329,
"mean_token_accuracy": 0.1979476734995842,
"num_tokens": 13708105.0,
"step": 7450
},
{
"entropy": 5.321125841140747,
"epoch": 0.6227029736050785,
"grad_norm": 1.078125,
"learning_rate": 0.0004967265438745352,
"loss": 5.2784,
"mean_token_accuracy": 0.18992661833763122,
"num_tokens": 13717094.0,
"step": 7455
},
{
"entropy": 5.31083312034607,
"epoch": 0.6231206147677915,
"grad_norm": 0.96875,
"learning_rate": 0.0004967214822877763,
"loss": 5.0979,
"mean_token_accuracy": 0.202111092209816,
"num_tokens": 13726639.0,
"step": 7460
},
{
"entropy": 5.328119993209839,
"epoch": 0.6235382559305045,
"grad_norm": 0.90234375,
"learning_rate": 0.0004967164168195159,
"loss": 5.0626,
"mean_token_accuracy": 0.1964184656739235,
"num_tokens": 13735220.0,
"step": 7465
},
{
"entropy": 5.213564682006836,
"epoch": 0.6239558970932175,
"grad_norm": 1.0078125,
"learning_rate": 0.0004967113474698428,
"loss": 5.038,
"mean_token_accuracy": 0.2117079883813858,
"num_tokens": 13743277.0,
"step": 7470
},
{
"entropy": 5.309726905822754,
"epoch": 0.6243735382559304,
"grad_norm": 0.93359375,
"learning_rate": 0.0004967062742388456,
"loss": 5.0866,
"mean_token_accuracy": 0.19689328372478485,
"num_tokens": 13752946.0,
"step": 7475
},
{
"entropy": 5.353356266021729,
"epoch": 0.6247911794186435,
"grad_norm": 1.0078125,
"learning_rate": 0.0004967011971266132,
"loss": 5.158,
"mean_token_accuracy": 0.19761786758899688,
"num_tokens": 13761297.0,
"step": 7480
},
{
"entropy": 5.300906324386597,
"epoch": 0.6252088205813565,
"grad_norm": 0.9140625,
"learning_rate": 0.0004966961161332346,
"loss": 5.0931,
"mean_token_accuracy": 0.1933014437556267,
"num_tokens": 13771238.0,
"step": 7485
},
{
"entropy": 5.290185832977295,
"epoch": 0.6256264617440696,
"grad_norm": 0.9453125,
"learning_rate": 0.0004966910312587985,
"loss": 5.1196,
"mean_token_accuracy": 0.1934815987944603,
"num_tokens": 13781420.0,
"step": 7490
},
{
"entropy": 5.25551323890686,
"epoch": 0.6260441029067825,
"grad_norm": 1.0390625,
"learning_rate": 0.0004966859425033941,
"loss": 5.0973,
"mean_token_accuracy": 0.19970676600933074,
"num_tokens": 13789833.0,
"step": 7495
},
{
"entropy": 5.3691894054412845,
"epoch": 0.6264617440694955,
"grad_norm": 0.99609375,
"learning_rate": 0.0004966808498671104,
"loss": 5.2065,
"mean_token_accuracy": 0.18750286102294922,
"num_tokens": 13799202.0,
"step": 7500
},
{
"entropy": 5.398705530166626,
"epoch": 0.6268793852322085,
"grad_norm": 0.88671875,
"learning_rate": 0.0004966757533500367,
"loss": 5.1751,
"mean_token_accuracy": 0.19055218547582625,
"num_tokens": 13809097.0,
"step": 7505
},
{
"entropy": 5.289015245437622,
"epoch": 0.6272970263949215,
"grad_norm": 1.0546875,
"learning_rate": 0.000496670652952262,
"loss": 5.0906,
"mean_token_accuracy": 0.20338452458381653,
"num_tokens": 13818437.0,
"step": 7510
},
{
"entropy": 5.234919691085816,
"epoch": 0.6277146675576345,
"grad_norm": 1.015625,
"learning_rate": 0.0004966655486738757,
"loss": 5.0904,
"mean_token_accuracy": 0.20160884261131287,
"num_tokens": 13827083.0,
"step": 7515
},
{
"entropy": 5.2574609279632565,
"epoch": 0.6281323087203475,
"grad_norm": 0.99609375,
"learning_rate": 0.0004966604405149672,
"loss": 5.0658,
"mean_token_accuracy": 0.1986864537000656,
"num_tokens": 13835905.0,
"step": 7520
},
{
"entropy": 5.2765950679779055,
"epoch": 0.6285499498830605,
"grad_norm": 0.9296875,
"learning_rate": 0.0004966553284756257,
"loss": 4.9942,
"mean_token_accuracy": 0.20565129220485687,
"num_tokens": 13845136.0,
"step": 7525
},
{
"entropy": 5.277464914321899,
"epoch": 0.6289675910457735,
"grad_norm": 0.84765625,
"learning_rate": 0.0004966502125559411,
"loss": 5.0883,
"mean_token_accuracy": 0.19836325198411942,
"num_tokens": 13855087.0,
"step": 7530
},
{
"entropy": 5.2940673828125,
"epoch": 0.6293852322084865,
"grad_norm": 0.94921875,
"learning_rate": 0.0004966450927560026,
"loss": 5.1278,
"mean_token_accuracy": 0.1913610130548477,
"num_tokens": 13864077.0,
"step": 7535
},
{
"entropy": 5.340163087844848,
"epoch": 0.6298028733711994,
"grad_norm": 0.94140625,
"learning_rate": 0.0004966399690759,
"loss": 5.1162,
"mean_token_accuracy": 0.1937885284423828,
"num_tokens": 13873404.0,
"step": 7540
},
{
"entropy": 5.366912174224853,
"epoch": 0.6302205145339125,
"grad_norm": 0.90625,
"learning_rate": 0.000496634841515723,
"loss": 5.1862,
"mean_token_accuracy": 0.19290923625230788,
"num_tokens": 13883231.0,
"step": 7545
},
{
"entropy": 5.184475564956665,
"epoch": 0.6306381556966255,
"grad_norm": 0.88671875,
"learning_rate": 0.0004966297100755613,
"loss": 4.951,
"mean_token_accuracy": 0.21052946150302887,
"num_tokens": 13892314.0,
"step": 7550
},
{
"entropy": 5.248685216903686,
"epoch": 0.6310557968593384,
"grad_norm": 0.9609375,
"learning_rate": 0.0004966245747555047,
"loss": 5.0437,
"mean_token_accuracy": 0.20218665599823,
"num_tokens": 13900771.0,
"step": 7555
},
{
"entropy": 5.225005292892456,
"epoch": 0.6314734380220515,
"grad_norm": 0.94140625,
"learning_rate": 0.0004966194355556432,
"loss": 5.0302,
"mean_token_accuracy": 0.20207641571760177,
"num_tokens": 13909615.0,
"step": 7560
},
{
"entropy": 5.287607431411743,
"epoch": 0.6318910791847645,
"grad_norm": 0.96484375,
"learning_rate": 0.0004966142924760666,
"loss": 5.1391,
"mean_token_accuracy": 0.19791716635227202,
"num_tokens": 13918491.0,
"step": 7565
},
{
"entropy": 5.286098003387451,
"epoch": 0.6323087203474774,
"grad_norm": 0.98828125,
"learning_rate": 0.0004966091455168652,
"loss": 5.1458,
"mean_token_accuracy": 0.18603342324495314,
"num_tokens": 13927935.0,
"step": 7570
},
{
"entropy": 5.2674699306488035,
"epoch": 0.6327263615101905,
"grad_norm": 0.96484375,
"learning_rate": 0.0004966039946781289,
"loss": 4.9664,
"mean_token_accuracy": 0.20871231108903884,
"num_tokens": 13936524.0,
"step": 7575
},
{
"entropy": 5.246699714660645,
"epoch": 0.6331440026729035,
"grad_norm": 0.97265625,
"learning_rate": 0.000496598839959948,
"loss": 5.0452,
"mean_token_accuracy": 0.19928555488586425,
"num_tokens": 13945544.0,
"step": 7580
},
{
"entropy": 5.269837665557861,
"epoch": 0.6335616438356164,
"grad_norm": 1.0234375,
"learning_rate": 0.0004965936813624126,
"loss": 5.0751,
"mean_token_accuracy": 0.209821717441082,
"num_tokens": 13955016.0,
"step": 7585
},
{
"entropy": 5.252441787719727,
"epoch": 0.6339792849983295,
"grad_norm": 0.8671875,
"learning_rate": 0.000496588518885613,
"loss": 5.0805,
"mean_token_accuracy": 0.19764155000448227,
"num_tokens": 13964045.0,
"step": 7590
},
{
"entropy": 5.284239625930786,
"epoch": 0.6343969261610425,
"grad_norm": 0.9140625,
"learning_rate": 0.0004965833525296397,
"loss": 5.0816,
"mean_token_accuracy": 0.1962558686733246,
"num_tokens": 13973554.0,
"step": 7595
},
{
"entropy": 5.274340867996216,
"epoch": 0.6348145673237554,
"grad_norm": 1.0078125,
"learning_rate": 0.0004965781822945831,
"loss": 5.0814,
"mean_token_accuracy": 0.19627740234136581,
"num_tokens": 13982669.0,
"step": 7600
},
{
"entropy": 5.392740154266358,
"epoch": 0.6352322084864684,
"grad_norm": 0.8984375,
"learning_rate": 0.0004965730081805338,
"loss": 5.194,
"mean_token_accuracy": 0.19119200706481934,
"num_tokens": 13993392.0,
"step": 7605
},
{
"entropy": 5.368127250671387,
"epoch": 0.6356498496491815,
"grad_norm": 0.9765625,
"learning_rate": 0.0004965678301875821,
"loss": 5.1773,
"mean_token_accuracy": 0.19042346626520157,
"num_tokens": 14002659.0,
"step": 7610
},
{
"entropy": 5.2989928245544435,
"epoch": 0.6360674908118944,
"grad_norm": 0.9765625,
"learning_rate": 0.000496562648315819,
"loss": 5.1331,
"mean_token_accuracy": 0.19522275179624557,
"num_tokens": 14011561.0,
"step": 7615
},
{
"entropy": 5.306436252593994,
"epoch": 0.6364851319746074,
"grad_norm": 1.0078125,
"learning_rate": 0.000496557462565335,
"loss": 5.1113,
"mean_token_accuracy": 0.20070280283689498,
"num_tokens": 14021286.0,
"step": 7620
},
{
"entropy": 5.2809803009033205,
"epoch": 0.6369027731373205,
"grad_norm": 0.91015625,
"learning_rate": 0.0004965522729362209,
"loss": 5.1151,
"mean_token_accuracy": 0.19623906910419464,
"num_tokens": 14031901.0,
"step": 7625
},
{
"entropy": 5.326603126525879,
"epoch": 0.6373204143000334,
"grad_norm": 0.9765625,
"learning_rate": 0.0004965470794285676,
"loss": 5.1552,
"mean_token_accuracy": 0.20101962983608246,
"num_tokens": 14041000.0,
"step": 7630
},
{
"entropy": 5.391379261016846,
"epoch": 0.6377380554627464,
"grad_norm": 1.0078125,
"learning_rate": 0.000496541882042466,
"loss": 5.1994,
"mean_token_accuracy": 0.19229317009449004,
"num_tokens": 14050567.0,
"step": 7635
},
{
"entropy": 5.272978019714356,
"epoch": 0.6381556966254595,
"grad_norm": 0.9296875,
"learning_rate": 0.0004965366807780071,
"loss": 5.0184,
"mean_token_accuracy": 0.2011002406477928,
"num_tokens": 14059990.0,
"step": 7640
},
{
"entropy": 5.318778657913208,
"epoch": 0.6385733377881724,
"grad_norm": 0.9375,
"learning_rate": 0.0004965314756352818,
"loss": 5.1571,
"mean_token_accuracy": 0.19478807598352432,
"num_tokens": 14069317.0,
"step": 7645
},
{
"entropy": 5.306348896026611,
"epoch": 0.6389909789508854,
"grad_norm": 0.94140625,
"learning_rate": 0.0004965262666143816,
"loss": 5.1121,
"mean_token_accuracy": 0.20019335448741912,
"num_tokens": 14078809.0,
"step": 7650
},
{
"entropy": 5.311046504974366,
"epoch": 0.6394086201135984,
"grad_norm": 0.921875,
"learning_rate": 0.0004965210537153973,
"loss": 5.1456,
"mean_token_accuracy": 0.19383479058742523,
"num_tokens": 14087660.0,
"step": 7655
},
{
"entropy": 5.320311880111694,
"epoch": 0.6398262612763114,
"grad_norm": 1.1171875,
"learning_rate": 0.0004965158369384204,
"loss": 5.1305,
"mean_token_accuracy": 0.1954892545938492,
"num_tokens": 14096585.0,
"step": 7660
},
{
"entropy": 5.310543727874756,
"epoch": 0.6402439024390244,
"grad_norm": 0.953125,
"learning_rate": 0.000496510616283542,
"loss": 5.0618,
"mean_token_accuracy": 0.2030956745147705,
"num_tokens": 14106079.0,
"step": 7665
},
{
"entropy": 5.290688514709473,
"epoch": 0.6406615436017374,
"grad_norm": 0.9609375,
"learning_rate": 0.0004965053917508537,
"loss": 5.1609,
"mean_token_accuracy": 0.19612673968076705,
"num_tokens": 14115988.0,
"step": 7670
},
{
"entropy": 5.3077010154724125,
"epoch": 0.6410791847644504,
"grad_norm": 0.84375,
"learning_rate": 0.0004965001633404469,
"loss": 5.2246,
"mean_token_accuracy": 0.18936279714107512,
"num_tokens": 14126980.0,
"step": 7675
},
{
"entropy": 5.361602306365967,
"epoch": 0.6414968259271634,
"grad_norm": 0.98046875,
"learning_rate": 0.0004964949310524131,
"loss": 5.1598,
"mean_token_accuracy": 0.19588953703641893,
"num_tokens": 14135780.0,
"step": 7680
},
{
"entropy": 5.360649919509887,
"epoch": 0.6419144670898764,
"grad_norm": 1.0,
"learning_rate": 0.0004964896948868439,
"loss": 5.1737,
"mean_token_accuracy": 0.18618624955415725,
"num_tokens": 14145655.0,
"step": 7685
},
{
"entropy": 5.308623027801514,
"epoch": 0.6423321082525894,
"grad_norm": 1.015625,
"learning_rate": 0.0004964844548438311,
"loss": 5.062,
"mean_token_accuracy": 0.20873176902532578,
"num_tokens": 14154157.0,
"step": 7690
},
{
"entropy": 5.358855152130127,
"epoch": 0.6427497494153024,
"grad_norm": 0.97265625,
"learning_rate": 0.0004964792109234662,
"loss": 5.1866,
"mean_token_accuracy": 0.1956228569149971,
"num_tokens": 14163749.0,
"step": 7695
},
{
"entropy": 5.32371244430542,
"epoch": 0.6431673905780154,
"grad_norm": 0.85546875,
"learning_rate": 0.0004964739631258411,
"loss": 5.117,
"mean_token_accuracy": 0.19713355153799056,
"num_tokens": 14174170.0,
"step": 7700
},
{
"entropy": 5.288732385635376,
"epoch": 0.6435850317407283,
"grad_norm": 1.03125,
"learning_rate": 0.0004964687114510478,
"loss": 5.1049,
"mean_token_accuracy": 0.19329818785190583,
"num_tokens": 14182625.0,
"step": 7705
},
{
"entropy": 5.318352460861206,
"epoch": 0.6440026729034414,
"grad_norm": 0.98046875,
"learning_rate": 0.000496463455899178,
"loss": 5.1001,
"mean_token_accuracy": 0.1936970353126526,
"num_tokens": 14192351.0,
"step": 7710
},
{
"entropy": 5.256820154190064,
"epoch": 0.6444203140661544,
"grad_norm": 0.98046875,
"learning_rate": 0.0004964581964703239,
"loss": 5.022,
"mean_token_accuracy": 0.2059452250599861,
"num_tokens": 14201107.0,
"step": 7715
},
{
"entropy": 5.344346380233764,
"epoch": 0.6448379552288673,
"grad_norm": 1.015625,
"learning_rate": 0.0004964529331645776,
"loss": 5.0548,
"mean_token_accuracy": 0.19872754961252212,
"num_tokens": 14209540.0,
"step": 7720
},
{
"entropy": 5.24354772567749,
"epoch": 0.6452555963915804,
"grad_norm": 1.0078125,
"learning_rate": 0.000496447665982031,
"loss": 4.9641,
"mean_token_accuracy": 0.2039660707116127,
"num_tokens": 14217944.0,
"step": 7725
},
{
"entropy": 5.240013885498047,
"epoch": 0.6456732375542934,
"grad_norm": 1.0390625,
"learning_rate": 0.0004964423949227765,
"loss": 5.0298,
"mean_token_accuracy": 0.20450728088617326,
"num_tokens": 14226044.0,
"step": 7730
},
{
"entropy": 5.299234867095947,
"epoch": 0.6460908787170063,
"grad_norm": 0.9375,
"learning_rate": 0.0004964371199869062,
"loss": 5.1067,
"mean_token_accuracy": 0.1913927376270294,
"num_tokens": 14235566.0,
"step": 7735
},
{
"entropy": 5.274836826324463,
"epoch": 0.6465085198797194,
"grad_norm": 0.92578125,
"learning_rate": 0.0004964318411745127,
"loss": 5.0307,
"mean_token_accuracy": 0.20680461823940277,
"num_tokens": 14244511.0,
"step": 7740
},
{
"entropy": 5.271277618408203,
"epoch": 0.6469261610424324,
"grad_norm": 0.94921875,
"learning_rate": 0.0004964265584856883,
"loss": 5.1153,
"mean_token_accuracy": 0.19961509108543396,
"num_tokens": 14254299.0,
"step": 7745
},
{
"entropy": 5.1947825908660885,
"epoch": 0.6473438022051453,
"grad_norm": 0.98046875,
"learning_rate": 0.0004964212719205254,
"loss": 5.0526,
"mean_token_accuracy": 0.1965569883584976,
"num_tokens": 14262915.0,
"step": 7750
},
{
"entropy": 5.3114409923553465,
"epoch": 0.6477614433678583,
"grad_norm": 0.9375,
"learning_rate": 0.0004964159814791167,
"loss": 5.0959,
"mean_token_accuracy": 0.195010207593441,
"num_tokens": 14272606.0,
"step": 7755
},
{
"entropy": 5.322947025299072,
"epoch": 0.6481790845305714,
"grad_norm": 1.0390625,
"learning_rate": 0.0004964106871615547,
"loss": 5.1762,
"mean_token_accuracy": 0.18923254311084747,
"num_tokens": 14282117.0,
"step": 7760
},
{
"entropy": 5.252630281448364,
"epoch": 0.6485967256932843,
"grad_norm": 0.9765625,
"learning_rate": 0.000496405388967932,
"loss": 5.0909,
"mean_token_accuracy": 0.20270532816648484,
"num_tokens": 14290421.0,
"step": 7765
},
{
"entropy": 5.364769744873047,
"epoch": 0.6490143668559973,
"grad_norm": 0.9921875,
"learning_rate": 0.0004964000868983416,
"loss": 5.1926,
"mean_token_accuracy": 0.19365463107824327,
"num_tokens": 14299870.0,
"step": 7770
},
{
"entropy": 5.3733344078063965,
"epoch": 0.6494320080187104,
"grad_norm": 0.92578125,
"learning_rate": 0.0004963947809528761,
"loss": 5.1025,
"mean_token_accuracy": 0.19596658945083617,
"num_tokens": 14309916.0,
"step": 7775
},
{
"entropy": 5.2920379638671875,
"epoch": 0.6498496491814233,
"grad_norm": 0.9765625,
"learning_rate": 0.0004963894711316285,
"loss": 5.0849,
"mean_token_accuracy": 0.19842049181461335,
"num_tokens": 14319061.0,
"step": 7780
},
{
"entropy": 5.2580322265625,
"epoch": 0.6502672903441363,
"grad_norm": 1.0546875,
"learning_rate": 0.0004963841574346917,
"loss": 5.0632,
"mean_token_accuracy": 0.20437927395105362,
"num_tokens": 14327241.0,
"step": 7785
},
{
"entropy": 5.328748655319214,
"epoch": 0.6506849315068494,
"grad_norm": 0.98828125,
"learning_rate": 0.0004963788398621587,
"loss": 5.1593,
"mean_token_accuracy": 0.19285826086997987,
"num_tokens": 14336586.0,
"step": 7790
},
{
"entropy": 5.345679473876953,
"epoch": 0.6511025726695623,
"grad_norm": 0.98828125,
"learning_rate": 0.0004963735184141227,
"loss": 5.1277,
"mean_token_accuracy": 0.19097969084978103,
"num_tokens": 14346484.0,
"step": 7795
},
{
"entropy": 5.3919901847839355,
"epoch": 0.6515202138322753,
"grad_norm": 0.9609375,
"learning_rate": 0.0004963681930906768,
"loss": 5.2006,
"mean_token_accuracy": 0.1854186996817589,
"num_tokens": 14356255.0,
"step": 7800
},
{
"entropy": 5.30468521118164,
"epoch": 0.6519378549949884,
"grad_norm": 0.98046875,
"learning_rate": 0.0004963628638919141,
"loss": 5.1371,
"mean_token_accuracy": 0.19927368313074112,
"num_tokens": 14365680.0,
"step": 7805
},
{
"entropy": 5.314786338806153,
"epoch": 0.6523554961577013,
"grad_norm": 0.9453125,
"learning_rate": 0.0004963575308179281,
"loss": 5.121,
"mean_token_accuracy": 0.20120913535356522,
"num_tokens": 14374546.0,
"step": 7810
},
{
"entropy": 5.354027605056762,
"epoch": 0.6527731373204143,
"grad_norm": 0.96484375,
"learning_rate": 0.0004963521938688122,
"loss": 5.1261,
"mean_token_accuracy": 0.19885395616292953,
"num_tokens": 14383624.0,
"step": 7815
},
{
"entropy": 5.336073207855224,
"epoch": 0.6531907784831273,
"grad_norm": 0.98046875,
"learning_rate": 0.0004963468530446596,
"loss": 5.2309,
"mean_token_accuracy": 0.19029023349285126,
"num_tokens": 14392482.0,
"step": 7820
},
{
"entropy": 5.3378218650817875,
"epoch": 0.6536084196458403,
"grad_norm": 0.96484375,
"learning_rate": 0.0004963415083455639,
"loss": 5.1289,
"mean_token_accuracy": 0.19663233011960984,
"num_tokens": 14401940.0,
"step": 7825
},
{
"entropy": 5.351269769668579,
"epoch": 0.6540260608085533,
"grad_norm": 1.0078125,
"learning_rate": 0.0004963361597716186,
"loss": 5.1819,
"mean_token_accuracy": 0.19608547687530517,
"num_tokens": 14411127.0,
"step": 7830
},
{
"entropy": 5.306780052185059,
"epoch": 0.6544437019712663,
"grad_norm": 0.921875,
"learning_rate": 0.0004963308073229175,
"loss": 5.0796,
"mean_token_accuracy": 0.20138038247823714,
"num_tokens": 14421358.0,
"step": 7835
},
{
"entropy": 5.334666681289673,
"epoch": 0.6548613431339793,
"grad_norm": 0.9765625,
"learning_rate": 0.0004963254509995542,
"loss": 5.1106,
"mean_token_accuracy": 0.19146810173988343,
"num_tokens": 14430507.0,
"step": 7840
},
{
"entropy": 5.282521343231201,
"epoch": 0.6552789842966923,
"grad_norm": 0.96875,
"learning_rate": 0.0004963200908016224,
"loss": 5.0872,
"mean_token_accuracy": 0.201054947078228,
"num_tokens": 14439723.0,
"step": 7845
},
{
"entropy": 5.305293369293213,
"epoch": 0.6556966254594053,
"grad_norm": 0.9375,
"learning_rate": 0.000496314726729216,
"loss": 5.0607,
"mean_token_accuracy": 0.20362130254507066,
"num_tokens": 14449011.0,
"step": 7850
},
{
"entropy": 5.2757282733917235,
"epoch": 0.6561142666221182,
"grad_norm": 1.0703125,
"learning_rate": 0.0004963093587824289,
"loss": 5.0535,
"mean_token_accuracy": 0.2070242941379547,
"num_tokens": 14457676.0,
"step": 7855
},
{
"entropy": 5.326574611663818,
"epoch": 0.6565319077848313,
"grad_norm": 0.953125,
"learning_rate": 0.0004963039869613551,
"loss": 5.1288,
"mean_token_accuracy": 0.19888340532779694,
"num_tokens": 14466956.0,
"step": 7860
},
{
"entropy": 5.305872297286987,
"epoch": 0.6569495489475443,
"grad_norm": 1.0234375,
"learning_rate": 0.0004962986112660887,
"loss": 5.0384,
"mean_token_accuracy": 0.20392645746469498,
"num_tokens": 14476083.0,
"step": 7865
},
{
"entropy": 5.366641902923584,
"epoch": 0.6573671901102572,
"grad_norm": 1.0390625,
"learning_rate": 0.0004962932316967237,
"loss": 5.1355,
"mean_token_accuracy": 0.1912634178996086,
"num_tokens": 14484980.0,
"step": 7870
},
{
"entropy": 5.339543724060059,
"epoch": 0.6577848312729703,
"grad_norm": 0.953125,
"learning_rate": 0.0004962878482533544,
"loss": 5.1557,
"mean_token_accuracy": 0.1986849457025528,
"num_tokens": 14493965.0,
"step": 7875
},
{
"entropy": 5.371400976181031,
"epoch": 0.6582024724356833,
"grad_norm": 1.1484375,
"learning_rate": 0.0004962824609360748,
"loss": 5.1378,
"mean_token_accuracy": 0.1929267928004265,
"num_tokens": 14501914.0,
"step": 7880
},
{
"entropy": 5.264023399353027,
"epoch": 0.6586201135983962,
"grad_norm": 0.98828125,
"learning_rate": 0.0004962770697449794,
"loss": 5.0521,
"mean_token_accuracy": 0.2027653858065605,
"num_tokens": 14511447.0,
"step": 7885
},
{
"entropy": 5.2572324752807615,
"epoch": 0.6590377547611093,
"grad_norm": 0.98046875,
"learning_rate": 0.0004962716746801625,
"loss": 5.0567,
"mean_token_accuracy": 0.19711564034223555,
"num_tokens": 14520558.0,
"step": 7890
},
{
"entropy": 5.27288556098938,
"epoch": 0.6594553959238223,
"grad_norm": 0.97265625,
"learning_rate": 0.0004962662757417187,
"loss": 5.0332,
"mean_token_accuracy": 0.19598861187696456,
"num_tokens": 14529995.0,
"step": 7895
},
{
"entropy": 5.281893110275268,
"epoch": 0.6598730370865352,
"grad_norm": 0.890625,
"learning_rate": 0.0004962608729297424,
"loss": 5.1839,
"mean_token_accuracy": 0.18899975717067719,
"num_tokens": 14538972.0,
"step": 7900
},
{
"entropy": 5.34423418045044,
"epoch": 0.6602906782492483,
"grad_norm": 0.90625,
"learning_rate": 0.0004962554662443281,
"loss": 5.1052,
"mean_token_accuracy": 0.1958961859345436,
"num_tokens": 14548255.0,
"step": 7905
},
{
"entropy": 5.350181198120117,
"epoch": 0.6607083194119613,
"grad_norm": 1.0,
"learning_rate": 0.0004962500556855706,
"loss": 5.0602,
"mean_token_accuracy": 0.20099108666181564,
"num_tokens": 14557265.0,
"step": 7910
},
{
"entropy": 5.371044874191284,
"epoch": 0.6611259605746742,
"grad_norm": 0.98046875,
"learning_rate": 0.0004962446412535645,
"loss": 5.14,
"mean_token_accuracy": 0.1972575604915619,
"num_tokens": 14566650.0,
"step": 7915
},
{
"entropy": 5.29285831451416,
"epoch": 0.6615436017373872,
"grad_norm": 0.921875,
"learning_rate": 0.0004962392229484046,
"loss": 5.0137,
"mean_token_accuracy": 0.2096802905201912,
"num_tokens": 14575550.0,
"step": 7920
},
{
"entropy": 5.277502822875976,
"epoch": 0.6619612429001003,
"grad_norm": 0.87890625,
"learning_rate": 0.000496233800770186,
"loss": 5.0468,
"mean_token_accuracy": 0.20320571959018707,
"num_tokens": 14584459.0,
"step": 7925
},
{
"entropy": 5.37594165802002,
"epoch": 0.6623788840628132,
"grad_norm": 0.95703125,
"learning_rate": 0.0004962283747190033,
"loss": 5.1841,
"mean_token_accuracy": 0.19367690682411193,
"num_tokens": 14593709.0,
"step": 7930
},
{
"entropy": 5.281787157058716,
"epoch": 0.6627965252255262,
"grad_norm": 0.9609375,
"learning_rate": 0.0004962229447949515,
"loss": 5.0383,
"mean_token_accuracy": 0.19434078633785248,
"num_tokens": 14602071.0,
"step": 7935
},
{
"entropy": 5.352070426940918,
"epoch": 0.6632141663882393,
"grad_norm": 0.94921875,
"learning_rate": 0.0004962175109981259,
"loss": 5.2248,
"mean_token_accuracy": 0.18860409110784532,
"num_tokens": 14612842.0,
"step": 7940
},
{
"entropy": 5.4165356159210205,
"epoch": 0.6636318075509522,
"grad_norm": 1.0234375,
"learning_rate": 0.0004962120733286215,
"loss": 5.2274,
"mean_token_accuracy": 0.19570623487234115,
"num_tokens": 14621645.0,
"step": 7945
},
{
"entropy": 5.270502996444702,
"epoch": 0.6640494487136652,
"grad_norm": 0.9765625,
"learning_rate": 0.0004962066317865334,
"loss": 5.0369,
"mean_token_accuracy": 0.20039988160133362,
"num_tokens": 14630447.0,
"step": 7950
},
{
"entropy": 5.238588285446167,
"epoch": 0.6644670898763783,
"grad_norm": 1.03125,
"learning_rate": 0.000496201186371957,
"loss": 5.1231,
"mean_token_accuracy": 0.2023925691843033,
"num_tokens": 14639371.0,
"step": 7955
},
{
"entropy": 5.291700267791748,
"epoch": 0.6648847310390912,
"grad_norm": 0.96484375,
"learning_rate": 0.0004961957370849875,
"loss": 5.0374,
"mean_token_accuracy": 0.19844225645065308,
"num_tokens": 14648540.0,
"step": 7960
},
{
"entropy": 5.343546295166016,
"epoch": 0.6653023722018042,
"grad_norm": 0.96875,
"learning_rate": 0.0004961902839257205,
"loss": 5.204,
"mean_token_accuracy": 0.1985940381884575,
"num_tokens": 14658130.0,
"step": 7965
},
{
"entropy": 5.351050758361817,
"epoch": 0.6657200133645172,
"grad_norm": 0.88671875,
"learning_rate": 0.0004961848268942513,
"loss": 5.1287,
"mean_token_accuracy": 0.1935892730951309,
"num_tokens": 14667538.0,
"step": 7970
},
{
"entropy": 5.359605836868286,
"epoch": 0.6661376545272302,
"grad_norm": 0.9921875,
"learning_rate": 0.0004961793659906754,
"loss": 5.1256,
"mean_token_accuracy": 0.19481374472379684,
"num_tokens": 14677115.0,
"step": 7975
},
{
"entropy": 5.382454442977905,
"epoch": 0.6665552956899432,
"grad_norm": 1.0078125,
"learning_rate": 0.0004961739012150885,
"loss": 5.0708,
"mean_token_accuracy": 0.20670475959777831,
"num_tokens": 14686444.0,
"step": 7980
},
{
"entropy": 5.291769075393677,
"epoch": 0.6669729368526562,
"grad_norm": 0.94140625,
"learning_rate": 0.0004961684325675862,
"loss": 5.0532,
"mean_token_accuracy": 0.20149117559194565,
"num_tokens": 14695951.0,
"step": 7985
},
{
"entropy": 5.314738845825195,
"epoch": 0.6673905780153692,
"grad_norm": 0.94140625,
"learning_rate": 0.0004961629600482643,
"loss": 5.1815,
"mean_token_accuracy": 0.19651031494140625,
"num_tokens": 14705415.0,
"step": 7990
},
{
"entropy": 5.418278408050537,
"epoch": 0.6678082191780822,
"grad_norm": 0.91796875,
"learning_rate": 0.0004961574836572186,
"loss": 5.1411,
"mean_token_accuracy": 0.19206517636775972,
"num_tokens": 14714793.0,
"step": 7995
},
{
"entropy": 5.219695138931274,
"epoch": 0.6682258603407952,
"grad_norm": 0.96875,
"learning_rate": 0.0004961520033945448,
"loss": 4.9231,
"mean_token_accuracy": 0.21136569529771804,
"num_tokens": 14723820.0,
"step": 8000
},
{
"entropy": 5.250234985351563,
"epoch": 0.6686435015035082,
"grad_norm": 0.98828125,
"learning_rate": 0.0004961465192603391,
"loss": 5.0428,
"mean_token_accuracy": 0.2078256979584694,
"num_tokens": 14733055.0,
"step": 8005
},
{
"entropy": 5.296591758728027,
"epoch": 0.6690611426662212,
"grad_norm": 0.95703125,
"learning_rate": 0.0004961410312546974,
"loss": 5.0776,
"mean_token_accuracy": 0.19873370826244355,
"num_tokens": 14742416.0,
"step": 8010
},
{
"entropy": 5.26785569190979,
"epoch": 0.6694787838289342,
"grad_norm": 0.9140625,
"learning_rate": 0.0004961355393777158,
"loss": 5.0934,
"mean_token_accuracy": 0.19436735063791274,
"num_tokens": 14751384.0,
"step": 8015
},
{
"entropy": 5.2438493251800535,
"epoch": 0.6698964249916471,
"grad_norm": 1.1484375,
"learning_rate": 0.0004961300436294902,
"loss": 4.9582,
"mean_token_accuracy": 0.2011357992887497,
"num_tokens": 14760525.0,
"step": 8020
},
{
"entropy": 5.290905427932739,
"epoch": 0.6703140661543602,
"grad_norm": 0.94140625,
"learning_rate": 0.0004961245440101171,
"loss": 5.1173,
"mean_token_accuracy": 0.19173007905483247,
"num_tokens": 14769405.0,
"step": 8025
},
{
"entropy": 5.225229740142822,
"epoch": 0.6707317073170732,
"grad_norm": 0.984375,
"learning_rate": 0.0004961190405196928,
"loss": 5.0026,
"mean_token_accuracy": 0.1994576022028923,
"num_tokens": 14778945.0,
"step": 8030
},
{
"entropy": 5.329796171188354,
"epoch": 0.6711493484797861,
"grad_norm": 1.015625,
"learning_rate": 0.0004961135331583135,
"loss": 5.1744,
"mean_token_accuracy": 0.19213393479585647,
"num_tokens": 14788289.0,
"step": 8035
},
{
"entropy": 5.2559342861175535,
"epoch": 0.6715669896424992,
"grad_norm": 0.91015625,
"learning_rate": 0.0004961080219260755,
"loss": 5.1005,
"mean_token_accuracy": 0.19120832532644272,
"num_tokens": 14797761.0,
"step": 8040
},
{
"entropy": 5.363939237594605,
"epoch": 0.6719846308052122,
"grad_norm": 1.0625,
"learning_rate": 0.0004961025068230757,
"loss": 5.1427,
"mean_token_accuracy": 0.2003529742360115,
"num_tokens": 14806424.0,
"step": 8045
},
{
"entropy": 5.286025762557983,
"epoch": 0.6724022719679251,
"grad_norm": 0.94140625,
"learning_rate": 0.0004960969878494103,
"loss": 5.076,
"mean_token_accuracy": 0.19537848681211473,
"num_tokens": 14815816.0,
"step": 8050
},
{
"entropy": 5.212112283706665,
"epoch": 0.6728199131306382,
"grad_norm": 1.046875,
"learning_rate": 0.0004960914650051758,
"loss": 5.0714,
"mean_token_accuracy": 0.19653652161359786,
"num_tokens": 14824877.0,
"step": 8055
},
{
"entropy": 5.171276235580445,
"epoch": 0.6732375542933512,
"grad_norm": 1.0078125,
"learning_rate": 0.0004960859382904691,
"loss": 4.976,
"mean_token_accuracy": 0.21297705471515654,
"num_tokens": 14833095.0,
"step": 8060
},
{
"entropy": 5.331959581375122,
"epoch": 0.6736551954560641,
"grad_norm": 0.90234375,
"learning_rate": 0.000496080407705387,
"loss": 5.121,
"mean_token_accuracy": 0.19362065345048904,
"num_tokens": 14842963.0,
"step": 8065
},
{
"entropy": 5.382730293273926,
"epoch": 0.6740728366187771,
"grad_norm": 0.95703125,
"learning_rate": 0.0004960748732500263,
"loss": 5.1816,
"mean_token_accuracy": 0.19915306568145752,
"num_tokens": 14851552.0,
"step": 8070
},
{
"entropy": 5.3230689525604244,
"epoch": 0.6744904777814902,
"grad_norm": 0.9921875,
"learning_rate": 0.0004960693349244838,
"loss": 5.0883,
"mean_token_accuracy": 0.20031056106090545,
"num_tokens": 14859997.0,
"step": 8075
},
{
"entropy": 5.29914288520813,
"epoch": 0.6749081189442031,
"grad_norm": 1.0703125,
"learning_rate": 0.0004960637927288566,
"loss": 5.0141,
"mean_token_accuracy": 0.20240161716938018,
"num_tokens": 14868050.0,
"step": 8080
},
{
"entropy": 5.309638547897339,
"epoch": 0.6753257601069161,
"grad_norm": 0.875,
"learning_rate": 0.0004960582466632414,
"loss": 5.1257,
"mean_token_accuracy": 0.19337987303733825,
"num_tokens": 14877774.0,
"step": 8085
},
{
"entropy": 5.237140083312989,
"epoch": 0.6757434012696292,
"grad_norm": 0.8515625,
"learning_rate": 0.0004960526967277356,
"loss": 5.063,
"mean_token_accuracy": 0.20347631722688675,
"num_tokens": 14886971.0,
"step": 8090
},
{
"entropy": 5.242138051986695,
"epoch": 0.6761610424323421,
"grad_norm": 0.90625,
"learning_rate": 0.0004960471429224363,
"loss": 5.001,
"mean_token_accuracy": 0.19711489975452423,
"num_tokens": 14895815.0,
"step": 8095
},
{
"entropy": 5.260533094406128,
"epoch": 0.6765786835950551,
"grad_norm": 0.98828125,
"learning_rate": 0.0004960415852474406,
"loss": 5.0181,
"mean_token_accuracy": 0.2097852572798729,
"num_tokens": 14904082.0,
"step": 8100
},
{
"entropy": 5.263446903228759,
"epoch": 0.6769963247577682,
"grad_norm": 0.9609375,
"learning_rate": 0.000496036023702846,
"loss": 5.0556,
"mean_token_accuracy": 0.20446614921092987,
"num_tokens": 14913533.0,
"step": 8105
},
{
"entropy": 5.258649158477783,
"epoch": 0.6774139659204811,
"grad_norm": 0.9921875,
"learning_rate": 0.0004960304582887497,
"loss": 5.0273,
"mean_token_accuracy": 0.20526493936777115,
"num_tokens": 14922063.0,
"step": 8110
},
{
"entropy": 5.302123355865478,
"epoch": 0.6778316070831941,
"grad_norm": 1.0234375,
"learning_rate": 0.0004960248890052493,
"loss": 5.0685,
"mean_token_accuracy": 0.1979711025953293,
"num_tokens": 14930219.0,
"step": 8115
},
{
"entropy": 5.3335975170135494,
"epoch": 0.6782492482459072,
"grad_norm": 1.0859375,
"learning_rate": 0.000496019315852442,
"loss": 5.1635,
"mean_token_accuracy": 0.1981356292963028,
"num_tokens": 14939463.0,
"step": 8120
},
{
"entropy": 5.3838402271270756,
"epoch": 0.6786668894086201,
"grad_norm": 0.9921875,
"learning_rate": 0.0004960137388304255,
"loss": 5.0599,
"mean_token_accuracy": 0.20408004373311997,
"num_tokens": 14948289.0,
"step": 8125
},
{
"entropy": 5.294534635543823,
"epoch": 0.6790845305713331,
"grad_norm": 0.921875,
"learning_rate": 0.0004960081579392975,
"loss": 5.156,
"mean_token_accuracy": 0.19927862137556077,
"num_tokens": 14958172.0,
"step": 8130
},
{
"entropy": 5.263111114501953,
"epoch": 0.6795021717340461,
"grad_norm": 1.0703125,
"learning_rate": 0.0004960025731791558,
"loss": 4.9503,
"mean_token_accuracy": 0.20457581877708436,
"num_tokens": 14967362.0,
"step": 8135
},
{
"entropy": 5.264223575592041,
"epoch": 0.6799198128967591,
"grad_norm": 1.03125,
"learning_rate": 0.0004959969845500979,
"loss": 5.1582,
"mean_token_accuracy": 0.1973573535680771,
"num_tokens": 14975942.0,
"step": 8140
},
{
"entropy": 5.2520099639892575,
"epoch": 0.6803374540594721,
"grad_norm": 1.015625,
"learning_rate": 0.0004959913920522218,
"loss": 5.0856,
"mean_token_accuracy": 0.20315959453582763,
"num_tokens": 14985428.0,
"step": 8145
},
{
"entropy": 5.305872964859009,
"epoch": 0.6807550952221851,
"grad_norm": 0.95703125,
"learning_rate": 0.0004959857956856254,
"loss": 5.0405,
"mean_token_accuracy": 0.19473765045404434,
"num_tokens": 14994558.0,
"step": 8150
},
{
"entropy": 5.276914644241333,
"epoch": 0.6811727363848981,
"grad_norm": 0.9609375,
"learning_rate": 0.0004959801954504067,
"loss": 5.0446,
"mean_token_accuracy": 0.20354918241500855,
"num_tokens": 15003674.0,
"step": 8155
},
{
"entropy": 5.314305591583252,
"epoch": 0.6815903775476111,
"grad_norm": 1.0546875,
"learning_rate": 0.0004959745913466635,
"loss": 5.1038,
"mean_token_accuracy": 0.19496819525957107,
"num_tokens": 15012569.0,
"step": 8160
},
{
"entropy": 5.29022445678711,
"epoch": 0.6820080187103241,
"grad_norm": 0.94921875,
"learning_rate": 0.0004959689833744942,
"loss": 5.1093,
"mean_token_accuracy": 0.20106361657381058,
"num_tokens": 15021863.0,
"step": 8165
},
{
"entropy": 5.3149768829345705,
"epoch": 0.682425659873037,
"grad_norm": 0.98046875,
"learning_rate": 0.0004959633715339969,
"loss": 5.1641,
"mean_token_accuracy": 0.18934969902038573,
"num_tokens": 15032272.0,
"step": 8170
},
{
"entropy": 5.278515291213989,
"epoch": 0.6828433010357501,
"grad_norm": 0.96875,
"learning_rate": 0.0004959577558252698,
"loss": 5.0947,
"mean_token_accuracy": 0.1965336576104164,
"num_tokens": 15041066.0,
"step": 8175
},
{
"entropy": 5.413792943954467,
"epoch": 0.6832609421984631,
"grad_norm": 0.9453125,
"learning_rate": 0.0004959521362484112,
"loss": 5.1795,
"mean_token_accuracy": 0.19215056598186492,
"num_tokens": 15050855.0,
"step": 8180
},
{
"entropy": 5.29156723022461,
"epoch": 0.683678583361176,
"grad_norm": 0.95703125,
"learning_rate": 0.0004959465128035195,
"loss": 5.0583,
"mean_token_accuracy": 0.20529651790857315,
"num_tokens": 15060810.0,
"step": 8185
},
{
"entropy": 5.279203033447265,
"epoch": 0.6840962245238891,
"grad_norm": 1.0,
"learning_rate": 0.0004959408854906932,
"loss": 5.0485,
"mean_token_accuracy": 0.19967632591724396,
"num_tokens": 15069761.0,
"step": 8190
},
{
"entropy": 5.249920797348023,
"epoch": 0.6845138656866021,
"grad_norm": 0.9453125,
"learning_rate": 0.0004959352543100306,
"loss": 5.04,
"mean_token_accuracy": 0.2037558451294899,
"num_tokens": 15078691.0,
"step": 8195
},
{
"entropy": 5.227643918991089,
"epoch": 0.684931506849315,
"grad_norm": 1.03125,
"learning_rate": 0.0004959296192616305,
"loss": 5.0172,
"mean_token_accuracy": 0.20608447790145873,
"num_tokens": 15087068.0,
"step": 8200
},
{
"entropy": 5.2339574813842775,
"epoch": 0.6853491480120281,
"grad_norm": 0.94140625,
"learning_rate": 0.0004959239803455916,
"loss": 4.9284,
"mean_token_accuracy": 0.20378715246915818,
"num_tokens": 15095962.0,
"step": 8205
},
{
"entropy": 5.276957893371582,
"epoch": 0.6857667891747411,
"grad_norm": 1.015625,
"learning_rate": 0.0004959183375620123,
"loss": 5.0839,
"mean_token_accuracy": 0.19588572382926941,
"num_tokens": 15105835.0,
"step": 8210
},
{
"entropy": 5.30806975364685,
"epoch": 0.686184430337454,
"grad_norm": 1.015625,
"learning_rate": 0.0004959126909109918,
"loss": 5.0171,
"mean_token_accuracy": 0.21215804815292358,
"num_tokens": 15114140.0,
"step": 8215
},
{
"entropy": 5.294583511352539,
"epoch": 0.686602071500167,
"grad_norm": 1.015625,
"learning_rate": 0.0004959070403926286,
"loss": 5.1102,
"mean_token_accuracy": 0.1996056392788887,
"num_tokens": 15123253.0,
"step": 8220
},
{
"entropy": 5.359203577041626,
"epoch": 0.6870197126628801,
"grad_norm": 0.93359375,
"learning_rate": 0.0004959013860070217,
"loss": 5.2143,
"mean_token_accuracy": 0.18856290876865386,
"num_tokens": 15133699.0,
"step": 8225
},
{
"entropy": 5.2771283149719235,
"epoch": 0.687437353825593,
"grad_norm": 0.95703125,
"learning_rate": 0.0004958957277542701,
"loss": 4.9794,
"mean_token_accuracy": 0.20531087666749953,
"num_tokens": 15142133.0,
"step": 8230
},
{
"entropy": 5.253825902938843,
"epoch": 0.687854994988306,
"grad_norm": 1.046875,
"learning_rate": 0.0004958900656344731,
"loss": 5.0634,
"mean_token_accuracy": 0.20176597386598588,
"num_tokens": 15151367.0,
"step": 8235
},
{
"entropy": 5.3135082721710205,
"epoch": 0.6882726361510191,
"grad_norm": 0.92578125,
"learning_rate": 0.0004958843996477296,
"loss": 5.0721,
"mean_token_accuracy": 0.1943616524338722,
"num_tokens": 15160938.0,
"step": 8240
},
{
"entropy": 5.3709173679351805,
"epoch": 0.688690277313732,
"grad_norm": 0.90625,
"learning_rate": 0.0004958787297941386,
"loss": 5.1449,
"mean_token_accuracy": 0.18945931047201156,
"num_tokens": 15170507.0,
"step": 8245
},
{
"entropy": 5.346620512008667,
"epoch": 0.689107918476445,
"grad_norm": 0.9453125,
"learning_rate": 0.0004958730560737998,
"loss": 5.1351,
"mean_token_accuracy": 0.19945331215858458,
"num_tokens": 15179822.0,
"step": 8250
},
{
"entropy": 5.25492000579834,
"epoch": 0.6895255596391581,
"grad_norm": 1.0078125,
"learning_rate": 0.0004958673784868121,
"loss": 5.0333,
"mean_token_accuracy": 0.19747120887041092,
"num_tokens": 15189353.0,
"step": 8255
},
{
"entropy": 5.342983198165894,
"epoch": 0.689943200801871,
"grad_norm": 1.0703125,
"learning_rate": 0.0004958616970332752,
"loss": 5.1042,
"mean_token_accuracy": 0.19633375704288483,
"num_tokens": 15197772.0,
"step": 8260
},
{
"entropy": 5.28785138130188,
"epoch": 0.690360841964584,
"grad_norm": 0.87109375,
"learning_rate": 0.0004958560117132884,
"loss": 5.0528,
"mean_token_accuracy": 0.20304337590932847,
"num_tokens": 15207453.0,
"step": 8265
},
{
"entropy": 5.266658735275269,
"epoch": 0.6907784831272971,
"grad_norm": 1.1171875,
"learning_rate": 0.0004958503225269513,
"loss": 4.9883,
"mean_token_accuracy": 0.2039983779191971,
"num_tokens": 15215787.0,
"step": 8270
},
{
"entropy": 5.365818119049072,
"epoch": 0.69119612429001,
"grad_norm": 0.9140625,
"learning_rate": 0.0004958446294743633,
"loss": 5.1081,
"mean_token_accuracy": 0.20084684193134308,
"num_tokens": 15225148.0,
"step": 8275
},
{
"entropy": 5.35207724571228,
"epoch": 0.691613765452723,
"grad_norm": 0.9453125,
"learning_rate": 0.0004958389325556245,
"loss": 5.1335,
"mean_token_accuracy": 0.20135464072227477,
"num_tokens": 15234533.0,
"step": 8280
},
{
"entropy": 5.3103758811950685,
"epoch": 0.692031406615436,
"grad_norm": 1.0,
"learning_rate": 0.0004958332317708341,
"loss": 5.114,
"mean_token_accuracy": 0.1895854130387306,
"num_tokens": 15243805.0,
"step": 8285
},
{
"entropy": 5.337362337112427,
"epoch": 0.692449047778149,
"grad_norm": 0.984375,
"learning_rate": 0.0004958275271200924,
"loss": 5.1491,
"mean_token_accuracy": 0.19990954846143721,
"num_tokens": 15253298.0,
"step": 8290
},
{
"entropy": 5.256933403015137,
"epoch": 0.692866688940862,
"grad_norm": 1.015625,
"learning_rate": 0.0004958218186034988,
"loss": 5.0958,
"mean_token_accuracy": 0.20100583285093307,
"num_tokens": 15261860.0,
"step": 8295
},
{
"entropy": 5.256998109817505,
"epoch": 0.693284330103575,
"grad_norm": 0.9140625,
"learning_rate": 0.0004958161062211537,
"loss": 5.0202,
"mean_token_accuracy": 0.20656843036413192,
"num_tokens": 15271021.0,
"step": 8300
},
{
"entropy": 5.268578004837036,
"epoch": 0.693701971266288,
"grad_norm": 1.0546875,
"learning_rate": 0.0004958103899731566,
"loss": 5.0733,
"mean_token_accuracy": 0.1962606877088547,
"num_tokens": 15280489.0,
"step": 8305
},
{
"entropy": 5.294385290145874,
"epoch": 0.694119612429001,
"grad_norm": 1.0078125,
"learning_rate": 0.0004958046698596081,
"loss": 5.0868,
"mean_token_accuracy": 0.20690073668956757,
"num_tokens": 15289387.0,
"step": 8310
},
{
"entropy": 5.352888441085815,
"epoch": 0.694537253591714,
"grad_norm": 0.9453125,
"learning_rate": 0.0004957989458806079,
"loss": 5.1597,
"mean_token_accuracy": 0.19498630464076996,
"num_tokens": 15299334.0,
"step": 8315
},
{
"entropy": 5.29277024269104,
"epoch": 0.694954894754427,
"grad_norm": 0.96875,
"learning_rate": 0.0004957932180362564,
"loss": 5.1086,
"mean_token_accuracy": 0.1986047223210335,
"num_tokens": 15309174.0,
"step": 8320
},
{
"entropy": 5.349100828170776,
"epoch": 0.69537253591714,
"grad_norm": 0.984375,
"learning_rate": 0.0004957874863266538,
"loss": 5.1795,
"mean_token_accuracy": 0.19804793298244477,
"num_tokens": 15318722.0,
"step": 8325
},
{
"entropy": 5.364262199401855,
"epoch": 0.695790177079853,
"grad_norm": 1.046875,
"learning_rate": 0.0004957817507519006,
"loss": 5.1062,
"mean_token_accuracy": 0.19870518147945404,
"num_tokens": 15327606.0,
"step": 8330
},
{
"entropy": 5.249139070510864,
"epoch": 0.696207818242566,
"grad_norm": 1.1171875,
"learning_rate": 0.0004957760113120971,
"loss": 5.0615,
"mean_token_accuracy": 0.20379491299390792,
"num_tokens": 15336520.0,
"step": 8335
},
{
"entropy": 5.328909063339234,
"epoch": 0.696625459405279,
"grad_norm": 0.984375,
"learning_rate": 0.0004957702680073437,
"loss": 5.1171,
"mean_token_accuracy": 0.1945207491517067,
"num_tokens": 15345437.0,
"step": 8340
},
{
"entropy": 5.375838708877564,
"epoch": 0.697043100567992,
"grad_norm": 0.9921875,
"learning_rate": 0.0004957645208377411,
"loss": 5.0858,
"mean_token_accuracy": 0.20251954346895218,
"num_tokens": 15354660.0,
"step": 8345
},
{
"entropy": 5.3071235656738285,
"epoch": 0.6974607417307049,
"grad_norm": 0.9609375,
"learning_rate": 0.0004957587698033898,
"loss": 5.0843,
"mean_token_accuracy": 0.20400189310312272,
"num_tokens": 15363958.0,
"step": 8350
},
{
"entropy": 5.235809230804444,
"epoch": 0.697878382893418,
"grad_norm": 0.95703125,
"learning_rate": 0.0004957530149043905,
"loss": 5.0626,
"mean_token_accuracy": 0.21092468053102492,
"num_tokens": 15373351.0,
"step": 8355
},
{
"entropy": 5.328260993957519,
"epoch": 0.698296024056131,
"grad_norm": 0.87109375,
"learning_rate": 0.0004957472561408438,
"loss": 5.0741,
"mean_token_accuracy": 0.20314389765262603,
"num_tokens": 15383069.0,
"step": 8360
},
{
"entropy": 5.283101415634155,
"epoch": 0.6987136652188439,
"grad_norm": 1.0078125,
"learning_rate": 0.0004957414935128509,
"loss": 5.0681,
"mean_token_accuracy": 0.1992212116718292,
"num_tokens": 15391394.0,
"step": 8365
},
{
"entropy": 5.312133693695069,
"epoch": 0.699131306381557,
"grad_norm": 0.9609375,
"learning_rate": 0.0004957357270205123,
"loss": 5.0496,
"mean_token_accuracy": 0.1943216875195503,
"num_tokens": 15399993.0,
"step": 8370
},
{
"entropy": 5.270085096359253,
"epoch": 0.69954894754427,
"grad_norm": 0.93359375,
"learning_rate": 0.0004957299566639292,
"loss": 5.0586,
"mean_token_accuracy": 0.20217949002981186,
"num_tokens": 15409886.0,
"step": 8375
},
{
"entropy": 5.348963069915771,
"epoch": 0.6999665887069829,
"grad_norm": 0.91796875,
"learning_rate": 0.0004957241824432025,
"loss": 5.0543,
"mean_token_accuracy": 0.20485760867595673,
"num_tokens": 15419772.0,
"step": 8380
},
{
"entropy": 5.3304625988006595,
"epoch": 0.700384229869696,
"grad_norm": 0.91015625,
"learning_rate": 0.0004957184043584333,
"loss": 5.0603,
"mean_token_accuracy": 0.1997678056359291,
"num_tokens": 15429631.0,
"step": 8385
},
{
"entropy": 5.235268259048462,
"epoch": 0.700801871032409,
"grad_norm": 1.0078125,
"learning_rate": 0.0004957126224097227,
"loss": 4.9991,
"mean_token_accuracy": 0.20829747170209884,
"num_tokens": 15438411.0,
"step": 8390
},
{
"entropy": 5.178777456283569,
"epoch": 0.7012195121951219,
"grad_norm": 0.8671875,
"learning_rate": 0.000495706836597172,
"loss": 5.0422,
"mean_token_accuracy": 0.20752644389867783,
"num_tokens": 15448943.0,
"step": 8395
},
{
"entropy": 5.316345167160034,
"epoch": 0.7016371533578349,
"grad_norm": 0.9296875,
"learning_rate": 0.0004957010469208825,
"loss": 4.9992,
"mean_token_accuracy": 0.20072227865457534,
"num_tokens": 15457908.0,
"step": 8400
},
{
"entropy": 5.3117913722991945,
"epoch": 0.702054794520548,
"grad_norm": 0.921875,
"learning_rate": 0.0004956952533809554,
"loss": 5.0861,
"mean_token_accuracy": 0.20061833560466766,
"num_tokens": 15467390.0,
"step": 8405
},
{
"entropy": 5.299106454849243,
"epoch": 0.7024724356832609,
"grad_norm": 1.0390625,
"learning_rate": 0.0004956894559774923,
"loss": 5.09,
"mean_token_accuracy": 0.19801439493894576,
"num_tokens": 15476419.0,
"step": 8410
},
{
"entropy": 5.217239713668823,
"epoch": 0.7028900768459739,
"grad_norm": 0.99609375,
"learning_rate": 0.0004956836547105947,
"loss": 4.9567,
"mean_token_accuracy": 0.20828939527273177,
"num_tokens": 15485147.0,
"step": 8415
},
{
"entropy": 5.339694166183472,
"epoch": 0.703307718008687,
"grad_norm": 1.0234375,
"learning_rate": 0.0004956778495803641,
"loss": 5.0693,
"mean_token_accuracy": 0.20299905836582183,
"num_tokens": 15493816.0,
"step": 8420
},
{
"entropy": 5.341966819763184,
"epoch": 0.7037253591713999,
"grad_norm": 0.921875,
"learning_rate": 0.000495672040586902,
"loss": 5.1719,
"mean_token_accuracy": 0.198117758333683,
"num_tokens": 15503051.0,
"step": 8425
},
{
"entropy": 5.27577576637268,
"epoch": 0.7041430003341129,
"grad_norm": 1.1171875,
"learning_rate": 0.0004956662277303103,
"loss": 4.9703,
"mean_token_accuracy": 0.20579586923122406,
"num_tokens": 15511854.0,
"step": 8430
},
{
"entropy": 5.2234395980834964,
"epoch": 0.704560641496826,
"grad_norm": 1.0546875,
"learning_rate": 0.0004956604110106905,
"loss": 5.1022,
"mean_token_accuracy": 0.20253927260637283,
"num_tokens": 15521492.0,
"step": 8435
},
{
"entropy": 5.385797119140625,
"epoch": 0.7049782826595389,
"grad_norm": 0.92578125,
"learning_rate": 0.0004956545904281447,
"loss": 5.175,
"mean_token_accuracy": 0.19008471965789794,
"num_tokens": 15530983.0,
"step": 8440
},
{
"entropy": 5.2986513614654545,
"epoch": 0.7053959238222519,
"grad_norm": 1.0390625,
"learning_rate": 0.0004956487659827748,
"loss": 5.0165,
"mean_token_accuracy": 0.2094108760356903,
"num_tokens": 15539722.0,
"step": 8445
},
{
"entropy": 5.226106739044189,
"epoch": 0.705813564984965,
"grad_norm": 0.9921875,
"learning_rate": 0.0004956429376746825,
"loss": 4.9884,
"mean_token_accuracy": 0.20108505189418793,
"num_tokens": 15548738.0,
"step": 8450
},
{
"entropy": 5.218436145782471,
"epoch": 0.7062312061476779,
"grad_norm": 0.875,
"learning_rate": 0.0004956371055039701,
"loss": 5.0367,
"mean_token_accuracy": 0.207009457051754,
"num_tokens": 15557520.0,
"step": 8455
},
{
"entropy": 5.371583795547485,
"epoch": 0.7066488473103909,
"grad_norm": 0.84375,
"learning_rate": 0.0004956312694707395,
"loss": 5.1095,
"mean_token_accuracy": 0.18941996544599532,
"num_tokens": 15567860.0,
"step": 8460
},
{
"entropy": 5.342587947845459,
"epoch": 0.7070664884731039,
"grad_norm": 0.96875,
"learning_rate": 0.000495625429575093,
"loss": 5.1189,
"mean_token_accuracy": 0.1891679659485817,
"num_tokens": 15577174.0,
"step": 8465
},
{
"entropy": 5.208229303359985,
"epoch": 0.7074841296358169,
"grad_norm": 1.0234375,
"learning_rate": 0.0004956195858171327,
"loss": 4.9148,
"mean_token_accuracy": 0.21200518310070038,
"num_tokens": 15585476.0,
"step": 8470
},
{
"entropy": 5.253299045562744,
"epoch": 0.7079017707985299,
"grad_norm": 0.94921875,
"learning_rate": 0.0004956137381969611,
"loss": 5.0509,
"mean_token_accuracy": 0.20225137025117873,
"num_tokens": 15595422.0,
"step": 8475
},
{
"entropy": 5.313663625717163,
"epoch": 0.7083194119612429,
"grad_norm": 0.9375,
"learning_rate": 0.0004956078867146805,
"loss": 5.1754,
"mean_token_accuracy": 0.19252022802829744,
"num_tokens": 15604223.0,
"step": 8480
},
{
"entropy": 5.360964393615722,
"epoch": 0.7087370531239559,
"grad_norm": 1.0,
"learning_rate": 0.0004956020313703933,
"loss": 5.1452,
"mean_token_accuracy": 0.19828291088342667,
"num_tokens": 15613715.0,
"step": 8485
},
{
"entropy": 5.283772993087768,
"epoch": 0.7091546942866689,
"grad_norm": 0.8828125,
"learning_rate": 0.0004955961721642018,
"loss": 4.9514,
"mean_token_accuracy": 0.21511617302894592,
"num_tokens": 15623826.0,
"step": 8490
},
{
"entropy": 5.279223775863647,
"epoch": 0.7095723354493819,
"grad_norm": 0.93359375,
"learning_rate": 0.000495590309096209,
"loss": 5.0742,
"mean_token_accuracy": 0.19097951352596282,
"num_tokens": 15633311.0,
"step": 8495
},
{
"entropy": 5.302570962905884,
"epoch": 0.7099899766120948,
"grad_norm": 0.97265625,
"learning_rate": 0.0004955844421665171,
"loss": 5.0585,
"mean_token_accuracy": 0.2011215418577194,
"num_tokens": 15642785.0,
"step": 8500
},
{
"entropy": 5.4760472774505615,
"epoch": 0.7104076177748079,
"grad_norm": 1.0234375,
"learning_rate": 0.0004955785713752292,
"loss": 5.2412,
"mean_token_accuracy": 0.19061580151319504,
"num_tokens": 15653009.0,
"step": 8505
},
{
"entropy": 5.285912990570068,
"epoch": 0.7108252589375209,
"grad_norm": 1.0546875,
"learning_rate": 0.0004955726967224479,
"loss": 5.0509,
"mean_token_accuracy": 0.20130133628845215,
"num_tokens": 15661672.0,
"step": 8510
},
{
"entropy": 5.246235466003418,
"epoch": 0.7112429001002338,
"grad_norm": 0.96484375,
"learning_rate": 0.000495566818208276,
"loss": 5.0419,
"mean_token_accuracy": 0.19729198962450029,
"num_tokens": 15672186.0,
"step": 8515
},
{
"entropy": 5.253251409530639,
"epoch": 0.7116605412629469,
"grad_norm": 1.078125,
"learning_rate": 0.0004955609358328163,
"loss": 5.062,
"mean_token_accuracy": 0.20900222957134246,
"num_tokens": 15680351.0,
"step": 8520
},
{
"entropy": 5.31048092842102,
"epoch": 0.7120781824256599,
"grad_norm": 0.96484375,
"learning_rate": 0.0004955550495961722,
"loss": 5.0378,
"mean_token_accuracy": 0.20055311769247056,
"num_tokens": 15690140.0,
"step": 8525
},
{
"entropy": 5.326376867294312,
"epoch": 0.7124958235883728,
"grad_norm": 0.9453125,
"learning_rate": 0.0004955491594984462,
"loss": 5.1456,
"mean_token_accuracy": 0.19382750391960143,
"num_tokens": 15699968.0,
"step": 8530
},
{
"entropy": 5.25221996307373,
"epoch": 0.7129134647510859,
"grad_norm": 0.9296875,
"learning_rate": 0.0004955432655397419,
"loss": 4.9932,
"mean_token_accuracy": 0.20656435936689377,
"num_tokens": 15709061.0,
"step": 8535
},
{
"entropy": 5.32095422744751,
"epoch": 0.7133311059137989,
"grad_norm": 1.0390625,
"learning_rate": 0.0004955373677201621,
"loss": 5.129,
"mean_token_accuracy": 0.19353578388690948,
"num_tokens": 15718350.0,
"step": 8540
},
{
"entropy": 5.292722558975219,
"epoch": 0.7137487470765118,
"grad_norm": 0.93359375,
"learning_rate": 0.0004955314660398105,
"loss": 5.0299,
"mean_token_accuracy": 0.2031239926815033,
"num_tokens": 15727120.0,
"step": 8545
},
{
"entropy": 5.286660861968994,
"epoch": 0.7141663882392248,
"grad_norm": 1.0234375,
"learning_rate": 0.0004955255604987899,
"loss": 5.1276,
"mean_token_accuracy": 0.19722677916288375,
"num_tokens": 15735865.0,
"step": 8550
},
{
"entropy": 5.2603583335876465,
"epoch": 0.7145840294019379,
"grad_norm": 0.921875,
"learning_rate": 0.000495519651097204,
"loss": 5.082,
"mean_token_accuracy": 0.19790492951869965,
"num_tokens": 15745239.0,
"step": 8555
},
{
"entropy": 5.266436433792114,
"epoch": 0.7150016705646508,
"grad_norm": 0.90625,
"learning_rate": 0.0004955137378351561,
"loss": 4.9968,
"mean_token_accuracy": 0.20420053005218505,
"num_tokens": 15754418.0,
"step": 8560
},
{
"entropy": 5.297505664825439,
"epoch": 0.7154193117273638,
"grad_norm": 0.921875,
"learning_rate": 0.0004955078207127498,
"loss": 5.0638,
"mean_token_accuracy": 0.20508573800325394,
"num_tokens": 15764175.0,
"step": 8565
},
{
"entropy": 5.310714530944824,
"epoch": 0.7158369528900769,
"grad_norm": 1.015625,
"learning_rate": 0.0004955018997300886,
"loss": 5.0352,
"mean_token_accuracy": 0.20058477520942689,
"num_tokens": 15772766.0,
"step": 8570
},
{
"entropy": 5.214248323440552,
"epoch": 0.7162545940527898,
"grad_norm": 0.99609375,
"learning_rate": 0.0004954959748872764,
"loss": 5.0591,
"mean_token_accuracy": 0.20578003525733948,
"num_tokens": 15781768.0,
"step": 8575
},
{
"entropy": 5.334406471252441,
"epoch": 0.7166722352155028,
"grad_norm": 0.98046875,
"learning_rate": 0.0004954900461844167,
"loss": 5.1135,
"mean_token_accuracy": 0.1969414994120598,
"num_tokens": 15791706.0,
"step": 8580
},
{
"entropy": 5.366378211975098,
"epoch": 0.7170898763782159,
"grad_norm": 0.96484375,
"learning_rate": 0.0004954841136216132,
"loss": 5.1272,
"mean_token_accuracy": 0.19717603623867036,
"num_tokens": 15802015.0,
"step": 8585
},
{
"entropy": 5.320854663848877,
"epoch": 0.7175075175409288,
"grad_norm": 1.015625,
"learning_rate": 0.0004954781771989701,
"loss": 5.0665,
"mean_token_accuracy": 0.2000778377056122,
"num_tokens": 15811116.0,
"step": 8590
},
{
"entropy": 5.351994037628174,
"epoch": 0.7179251587036418,
"grad_norm": 1.0859375,
"learning_rate": 0.000495472236916591,
"loss": 5.0377,
"mean_token_accuracy": 0.1998966559767723,
"num_tokens": 15819240.0,
"step": 8595
},
{
"entropy": 5.286137437820434,
"epoch": 0.7183427998663549,
"grad_norm": 0.94921875,
"learning_rate": 0.0004954662927745799,
"loss": 5.0492,
"mean_token_accuracy": 0.1938687175512314,
"num_tokens": 15827740.0,
"step": 8600
},
{
"entropy": 5.233812236785889,
"epoch": 0.7187604410290678,
"grad_norm": 0.96875,
"learning_rate": 0.0004954603447730411,
"loss": 5.1066,
"mean_token_accuracy": 0.19758808612823486,
"num_tokens": 15836763.0,
"step": 8605
},
{
"entropy": 5.399864768981933,
"epoch": 0.7191780821917808,
"grad_norm": 0.88671875,
"learning_rate": 0.0004954543929120786,
"loss": 5.1756,
"mean_token_accuracy": 0.19459035098552704,
"num_tokens": 15846624.0,
"step": 8610
},
{
"entropy": 5.343923711776734,
"epoch": 0.7195957233544938,
"grad_norm": 1.015625,
"learning_rate": 0.0004954484371917967,
"loss": 5.0874,
"mean_token_accuracy": 0.19594507217407225,
"num_tokens": 15855713.0,
"step": 8615
},
{
"entropy": 5.2160741806030275,
"epoch": 0.7200133645172068,
"grad_norm": 1.078125,
"learning_rate": 0.0004954424776122993,
"loss": 4.9217,
"mean_token_accuracy": 0.2120089665055275,
"num_tokens": 15863577.0,
"step": 8620
},
{
"entropy": 5.294074773788452,
"epoch": 0.7204310056799198,
"grad_norm": 0.984375,
"learning_rate": 0.0004954365141736913,
"loss": 5.0769,
"mean_token_accuracy": 0.20392575412988662,
"num_tokens": 15872479.0,
"step": 8625
},
{
"entropy": 5.355267095565796,
"epoch": 0.7208486468426328,
"grad_norm": 0.9921875,
"learning_rate": 0.0004954305468760765,
"loss": 5.1697,
"mean_token_accuracy": 0.2003491461277008,
"num_tokens": 15882170.0,
"step": 8630
},
{
"entropy": 5.341642427444458,
"epoch": 0.7212662880053458,
"grad_norm": 0.90234375,
"learning_rate": 0.0004954245757195598,
"loss": 5.0649,
"mean_token_accuracy": 0.2069210559129715,
"num_tokens": 15891694.0,
"step": 8635
},
{
"entropy": 5.319044208526611,
"epoch": 0.7216839291680588,
"grad_norm": 0.90234375,
"learning_rate": 0.0004954186007042454,
"loss": 5.1771,
"mean_token_accuracy": 0.18936493545770644,
"num_tokens": 15902232.0,
"step": 8640
},
{
"entropy": 5.281481456756592,
"epoch": 0.7221015703307718,
"grad_norm": 0.96875,
"learning_rate": 0.0004954126218302382,
"loss": 5.0232,
"mean_token_accuracy": 0.20015035271644593,
"num_tokens": 15912218.0,
"step": 8645
},
{
"entropy": 5.269867467880249,
"epoch": 0.7225192114934847,
"grad_norm": 1.015625,
"learning_rate": 0.0004954066390976427,
"loss": 5.0158,
"mean_token_accuracy": 0.20863578766584395,
"num_tokens": 15922011.0,
"step": 8650
},
{
"entropy": 5.1690638065338135,
"epoch": 0.7229368526561978,
"grad_norm": 1.015625,
"learning_rate": 0.0004954006525065639,
"loss": 4.9903,
"mean_token_accuracy": 0.20333651453256607,
"num_tokens": 15930394.0,
"step": 8655
},
{
"entropy": 5.291082191467285,
"epoch": 0.7233544938189108,
"grad_norm": 1.015625,
"learning_rate": 0.0004953946620571061,
"loss": 5.085,
"mean_token_accuracy": 0.197337506711483,
"num_tokens": 15939880.0,
"step": 8660
},
{
"entropy": 5.285974216461182,
"epoch": 0.7237721349816237,
"grad_norm": 0.98828125,
"learning_rate": 0.0004953886677493746,
"loss": 4.9699,
"mean_token_accuracy": 0.20998845845460892,
"num_tokens": 15948521.0,
"step": 8665
},
{
"entropy": 5.29067850112915,
"epoch": 0.7241897761443368,
"grad_norm": 0.984375,
"learning_rate": 0.0004953826695834741,
"loss": 5.0987,
"mean_token_accuracy": 0.19873740524053574,
"num_tokens": 15956811.0,
"step": 8670
},
{
"entropy": 5.2703441143035885,
"epoch": 0.7246074173070498,
"grad_norm": 0.91015625,
"learning_rate": 0.0004953766675595098,
"loss": 5.1187,
"mean_token_accuracy": 0.20053162127733232,
"num_tokens": 15967019.0,
"step": 8675
},
{
"entropy": 5.385362720489502,
"epoch": 0.7250250584697627,
"grad_norm": 1.0,
"learning_rate": 0.0004953706616775866,
"loss": 5.1137,
"mean_token_accuracy": 0.19549519270658494,
"num_tokens": 15976149.0,
"step": 8680
},
{
"entropy": 5.302666473388672,
"epoch": 0.7254426996324758,
"grad_norm": 0.93359375,
"learning_rate": 0.0004953646519378097,
"loss": 4.9218,
"mean_token_accuracy": 0.20791263431310653,
"num_tokens": 15986711.0,
"step": 8685
},
{
"entropy": 5.277779483795166,
"epoch": 0.7258603407951888,
"grad_norm": 1.0078125,
"learning_rate": 0.0004953586383402844,
"loss": 5.091,
"mean_token_accuracy": 0.19872635155916213,
"num_tokens": 15996482.0,
"step": 8690
},
{
"entropy": 5.176142930984497,
"epoch": 0.7262779819579017,
"grad_norm": 0.97265625,
"learning_rate": 0.0004953526208851157,
"loss": 4.9632,
"mean_token_accuracy": 0.2036085918545723,
"num_tokens": 16005801.0,
"step": 8695
},
{
"entropy": 5.199421167373657,
"epoch": 0.7266956231206148,
"grad_norm": 0.94921875,
"learning_rate": 0.0004953465995724092,
"loss": 5.0092,
"mean_token_accuracy": 0.2009180873632431,
"num_tokens": 16014807.0,
"step": 8700
},
{
"entropy": 5.310466480255127,
"epoch": 0.7271132642833278,
"grad_norm": 1.0703125,
"learning_rate": 0.0004953405744022705,
"loss": 5.1466,
"mean_token_accuracy": 0.19814235866069793,
"num_tokens": 16023326.0,
"step": 8705
},
{
"entropy": 5.3451879024505615,
"epoch": 0.7275309054460407,
"grad_norm": 1.0234375,
"learning_rate": 0.0004953345453748045,
"loss": 5.0942,
"mean_token_accuracy": 0.194943343102932,
"num_tokens": 16031648.0,
"step": 8710
},
{
"entropy": 5.271716070175171,
"epoch": 0.7279485466087537,
"grad_norm": 1.0234375,
"learning_rate": 0.0004953285124901171,
"loss": 5.0223,
"mean_token_accuracy": 0.20396826714277266,
"num_tokens": 16040179.0,
"step": 8715
},
{
"entropy": 5.303973627090454,
"epoch": 0.7283661877714668,
"grad_norm": 0.8359375,
"learning_rate": 0.000495322475748314,
"loss": 5.1873,
"mean_token_accuracy": 0.1893076092004776,
"num_tokens": 16050770.0,
"step": 8720
},
{
"entropy": 5.361210536956787,
"epoch": 0.7287838289341797,
"grad_norm": 0.96875,
"learning_rate": 0.0004953164351495007,
"loss": 5.0387,
"mean_token_accuracy": 0.20573963075876237,
"num_tokens": 16060504.0,
"step": 8725
},
{
"entropy": 5.220712089538575,
"epoch": 0.7292014700968927,
"grad_norm": 0.9609375,
"learning_rate": 0.0004953103906937831,
"loss": 4.8926,
"mean_token_accuracy": 0.21957828849554062,
"num_tokens": 16069240.0,
"step": 8730
},
{
"entropy": 5.176104211807251,
"epoch": 0.7296191112596058,
"grad_norm": 1.0078125,
"learning_rate": 0.0004953043423812669,
"loss": 5.1434,
"mean_token_accuracy": 0.19955668449401856,
"num_tokens": 16079098.0,
"step": 8735
},
{
"entropy": 5.2147783756256105,
"epoch": 0.7300367524223187,
"grad_norm": 0.94921875,
"learning_rate": 0.0004952982902120579,
"loss": 4.97,
"mean_token_accuracy": 0.20637630224227904,
"num_tokens": 16088454.0,
"step": 8740
},
{
"entropy": 5.328648042678833,
"epoch": 0.7304543935850317,
"grad_norm": 1.0390625,
"learning_rate": 0.0004952922341862622,
"loss": 5.1058,
"mean_token_accuracy": 0.19543055295944214,
"num_tokens": 16096985.0,
"step": 8745
},
{
"entropy": 5.397544193267822,
"epoch": 0.7308720347477448,
"grad_norm": 1.03125,
"learning_rate": 0.0004952861743039858,
"loss": 5.1971,
"mean_token_accuracy": 0.1901056155562401,
"num_tokens": 16105434.0,
"step": 8750
},
{
"entropy": 5.184029293060303,
"epoch": 0.7312896759104577,
"grad_norm": 1.0,
"learning_rate": 0.0004952801105653347,
"loss": 4.9423,
"mean_token_accuracy": 0.21344213038682938,
"num_tokens": 16114933.0,
"step": 8755
},
{
"entropy": 5.333026695251465,
"epoch": 0.7317073170731707,
"grad_norm": 0.87109375,
"learning_rate": 0.0004952740429704152,
"loss": 5.1163,
"mean_token_accuracy": 0.19112866818904878,
"num_tokens": 16124954.0,
"step": 8760
},
{
"entropy": 5.314546346664429,
"epoch": 0.7321249582358837,
"grad_norm": 0.9765625,
"learning_rate": 0.0004952679715193334,
"loss": 4.931,
"mean_token_accuracy": 0.20849736630916596,
"num_tokens": 16134442.0,
"step": 8765
},
{
"entropy": 5.301894617080689,
"epoch": 0.7325425993985967,
"grad_norm": 1.0625,
"learning_rate": 0.0004952618962121956,
"loss": 5.1039,
"mean_token_accuracy": 0.1969986751675606,
"num_tokens": 16143994.0,
"step": 8770
},
{
"entropy": 5.2608765125274655,
"epoch": 0.7329602405613097,
"grad_norm": 0.984375,
"learning_rate": 0.0004952558170491082,
"loss": 5.0265,
"mean_token_accuracy": 0.20228615403175354,
"num_tokens": 16152931.0,
"step": 8775
},
{
"entropy": 5.302912044525146,
"epoch": 0.7333778817240227,
"grad_norm": 0.96484375,
"learning_rate": 0.0004952497340301776,
"loss": 5.0227,
"mean_token_accuracy": 0.1981823980808258,
"num_tokens": 16162716.0,
"step": 8780
},
{
"entropy": 5.308622217178344,
"epoch": 0.7337955228867357,
"grad_norm": 0.921875,
"learning_rate": 0.0004952436471555103,
"loss": 5.0805,
"mean_token_accuracy": 0.1987333506345749,
"num_tokens": 16172371.0,
"step": 8785
},
{
"entropy": 5.302491998672485,
"epoch": 0.7342131640494487,
"grad_norm": 0.98046875,
"learning_rate": 0.0004952375564252128,
"loss": 5.1532,
"mean_token_accuracy": 0.19661444872617723,
"num_tokens": 16181823.0,
"step": 8790
},
{
"entropy": 5.293154573440551,
"epoch": 0.7346308052121617,
"grad_norm": 0.9921875,
"learning_rate": 0.0004952314618393919,
"loss": 5.0655,
"mean_token_accuracy": 0.20362306535243987,
"num_tokens": 16190720.0,
"step": 8795
},
{
"entropy": 5.27875280380249,
"epoch": 0.7350484463748747,
"grad_norm": 1.0078125,
"learning_rate": 0.000495225363398154,
"loss": 4.9997,
"mean_token_accuracy": 0.20117570012807845,
"num_tokens": 16199589.0,
"step": 8800
},
{
"entropy": 5.353147792816162,
"epoch": 0.7354660875375877,
"grad_norm": 0.96484375,
"learning_rate": 0.0004952192611016061,
"loss": 5.1663,
"mean_token_accuracy": 0.19834163784980774,
"num_tokens": 16208800.0,
"step": 8805
},
{
"entropy": 5.310691833496094,
"epoch": 0.7358837287003007,
"grad_norm": 0.921875,
"learning_rate": 0.000495213154949855,
"loss": 5.1518,
"mean_token_accuracy": 0.1956395149230957,
"num_tokens": 16218835.0,
"step": 8810
},
{
"entropy": 5.350011682510376,
"epoch": 0.7363013698630136,
"grad_norm": 1.0078125,
"learning_rate": 0.0004952070449430075,
"loss": 5.0969,
"mean_token_accuracy": 0.20128366947174073,
"num_tokens": 16228441.0,
"step": 8815
},
{
"entropy": 5.362833547592163,
"epoch": 0.7367190110257267,
"grad_norm": 1.171875,
"learning_rate": 0.0004952009310811706,
"loss": 5.0783,
"mean_token_accuracy": 0.2080063506960869,
"num_tokens": 16237877.0,
"step": 8820
},
{
"entropy": 5.2682441711425785,
"epoch": 0.7371366521884397,
"grad_norm": 0.94921875,
"learning_rate": 0.0004951948133644514,
"loss": 5.013,
"mean_token_accuracy": 0.20266563594341278,
"num_tokens": 16246809.0,
"step": 8825
},
{
"entropy": 5.324000597000122,
"epoch": 0.7375542933511526,
"grad_norm": 1.03125,
"learning_rate": 0.000495188691792957,
"loss": 5.0832,
"mean_token_accuracy": 0.2116958737373352,
"num_tokens": 16255754.0,
"step": 8830
},
{
"entropy": 5.309404754638672,
"epoch": 0.7379719345138657,
"grad_norm": 0.92578125,
"learning_rate": 0.0004951825663667943,
"loss": 5.119,
"mean_token_accuracy": 0.20001896917819978,
"num_tokens": 16264861.0,
"step": 8835
},
{
"entropy": 5.268008613586426,
"epoch": 0.7383895756765787,
"grad_norm": 0.9375,
"learning_rate": 0.000495176437086071,
"loss": 5.0394,
"mean_token_accuracy": 0.20309204012155532,
"num_tokens": 16273628.0,
"step": 8840
},
{
"entropy": 5.287184858322144,
"epoch": 0.7388072168392917,
"grad_norm": 0.94140625,
"learning_rate": 0.000495170303950894,
"loss": 5.0394,
"mean_token_accuracy": 0.20065390467643737,
"num_tokens": 16282958.0,
"step": 8845
},
{
"entropy": 5.342790222167968,
"epoch": 0.7392248580020047,
"grad_norm": 0.953125,
"learning_rate": 0.0004951641669613708,
"loss": 5.081,
"mean_token_accuracy": 0.19833653122186662,
"num_tokens": 16292763.0,
"step": 8850
},
{
"entropy": 5.290163946151734,
"epoch": 0.7396424991647177,
"grad_norm": 1.0,
"learning_rate": 0.0004951580261176088,
"loss": 5.0637,
"mean_token_accuracy": 0.20175071358680724,
"num_tokens": 16302216.0,
"step": 8855
},
{
"entropy": 5.213641691207886,
"epoch": 0.7400601403274307,
"grad_norm": 1.0390625,
"learning_rate": 0.0004951518814197156,
"loss": 4.9844,
"mean_token_accuracy": 0.20687176138162613,
"num_tokens": 16311672.0,
"step": 8860
},
{
"entropy": 5.289334297180176,
"epoch": 0.7404777814901436,
"grad_norm": 0.95703125,
"learning_rate": 0.0004951457328677989,
"loss": 4.9879,
"mean_token_accuracy": 0.20168158859014512,
"num_tokens": 16321027.0,
"step": 8865
},
{
"entropy": 5.330817413330078,
"epoch": 0.7408954226528567,
"grad_norm": 1.0390625,
"learning_rate": 0.0004951395804619659,
"loss": 5.105,
"mean_token_accuracy": 0.19175008237361907,
"num_tokens": 16329471.0,
"step": 8870
},
{
"entropy": 5.239703941345215,
"epoch": 0.7413130638155697,
"grad_norm": 0.9921875,
"learning_rate": 0.0004951334242023247,
"loss": 5.0449,
"mean_token_accuracy": 0.2017623022198677,
"num_tokens": 16339085.0,
"step": 8875
},
{
"entropy": 5.346188020706177,
"epoch": 0.7417307049782826,
"grad_norm": 0.953125,
"learning_rate": 0.0004951272640889828,
"loss": 5.1115,
"mean_token_accuracy": 0.20438565462827682,
"num_tokens": 16347950.0,
"step": 8880
},
{
"entropy": 5.27294602394104,
"epoch": 0.7421483461409957,
"grad_norm": 1.0078125,
"learning_rate": 0.0004951211001220484,
"loss": 4.9026,
"mean_token_accuracy": 0.209273998439312,
"num_tokens": 16357129.0,
"step": 8885
},
{
"entropy": 5.2367175102233885,
"epoch": 0.7425659873037087,
"grad_norm": 1.0234375,
"learning_rate": 0.000495114932301629,
"loss": 5.0269,
"mean_token_accuracy": 0.20274857431650162,
"num_tokens": 16365768.0,
"step": 8890
},
{
"entropy": 5.276386499404907,
"epoch": 0.7429836284664216,
"grad_norm": 0.9296875,
"learning_rate": 0.0004951087606278328,
"loss": 4.9989,
"mean_token_accuracy": 0.2076172113418579,
"num_tokens": 16375548.0,
"step": 8895
},
{
"entropy": 5.308200120925903,
"epoch": 0.7434012696291347,
"grad_norm": 0.8515625,
"learning_rate": 0.0004951025851007678,
"loss": 5.0658,
"mean_token_accuracy": 0.2082823783159256,
"num_tokens": 16385472.0,
"step": 8900
},
{
"entropy": 5.2607502937316895,
"epoch": 0.7438189107918477,
"grad_norm": 0.94140625,
"learning_rate": 0.0004950964057205421,
"loss": 4.9373,
"mean_token_accuracy": 0.20607596933841704,
"num_tokens": 16395048.0,
"step": 8905
},
{
"entropy": 5.358189392089844,
"epoch": 0.7442365519545606,
"grad_norm": 0.890625,
"learning_rate": 0.000495090222487264,
"loss": 5.168,
"mean_token_accuracy": 0.193108494579792,
"num_tokens": 16405378.0,
"step": 8910
},
{
"entropy": 5.2722883224487305,
"epoch": 0.7446541931172737,
"grad_norm": 0.921875,
"learning_rate": 0.0004950840354010415,
"loss": 5.0344,
"mean_token_accuracy": 0.20605543553829192,
"num_tokens": 16415206.0,
"step": 8915
},
{
"entropy": 5.290538692474366,
"epoch": 0.7450718342799867,
"grad_norm": 0.98828125,
"learning_rate": 0.0004950778444619831,
"loss": 5.0532,
"mean_token_accuracy": 0.21444262117147445,
"num_tokens": 16424185.0,
"step": 8920
},
{
"entropy": 5.269124174118042,
"epoch": 0.7454894754426996,
"grad_norm": 0.94140625,
"learning_rate": 0.000495071649670197,
"loss": 5.0342,
"mean_token_accuracy": 0.20239607542753218,
"num_tokens": 16433330.0,
"step": 8925
},
{
"entropy": 5.251119804382324,
"epoch": 0.7459071166054126,
"grad_norm": 0.953125,
"learning_rate": 0.000495065451025792,
"loss": 5.0385,
"mean_token_accuracy": 0.20068047046661378,
"num_tokens": 16442777.0,
"step": 8930
},
{
"entropy": 5.272030162811279,
"epoch": 0.7463247577681257,
"grad_norm": 1.0546875,
"learning_rate": 0.0004950592485288761,
"loss": 5.0313,
"mean_token_accuracy": 0.20843802988529206,
"num_tokens": 16451838.0,
"step": 8935
},
{
"entropy": 5.262492990493774,
"epoch": 0.7467423989308386,
"grad_norm": 1.078125,
"learning_rate": 0.0004950530421795584,
"loss": 5.0034,
"mean_token_accuracy": 0.20206386893987655,
"num_tokens": 16460022.0,
"step": 8940
},
{
"entropy": 5.227031803131103,
"epoch": 0.7471600400935516,
"grad_norm": 1.03125,
"learning_rate": 0.0004950468319779472,
"loss": 4.8881,
"mean_token_accuracy": 0.21292097121477127,
"num_tokens": 16468891.0,
"step": 8945
},
{
"entropy": 5.178910493850708,
"epoch": 0.7475776812562647,
"grad_norm": 0.9609375,
"learning_rate": 0.0004950406179241515,
"loss": 4.9798,
"mean_token_accuracy": 0.21524928212165834,
"num_tokens": 16478020.0,
"step": 8950
},
{
"entropy": 5.294888162612915,
"epoch": 0.7479953224189776,
"grad_norm": 0.890625,
"learning_rate": 0.0004950344000182798,
"loss": 5.1492,
"mean_token_accuracy": 0.19855829030275346,
"num_tokens": 16487536.0,
"step": 8955
},
{
"entropy": 5.33293890953064,
"epoch": 0.7484129635816906,
"grad_norm": 0.94921875,
"learning_rate": 0.0004950281782604411,
"loss": 5.0628,
"mean_token_accuracy": 0.19553593695163726,
"num_tokens": 16495786.0,
"step": 8960
},
{
"entropy": 5.392992734909058,
"epoch": 0.7488306047444037,
"grad_norm": 1.0078125,
"learning_rate": 0.0004950219526507442,
"loss": 5.1313,
"mean_token_accuracy": 0.19654490202665328,
"num_tokens": 16505535.0,
"step": 8965
},
{
"entropy": 5.348032522201538,
"epoch": 0.7492482459071166,
"grad_norm": 0.984375,
"learning_rate": 0.0004950157231892983,
"loss": 5.1381,
"mean_token_accuracy": 0.2017208978533745,
"num_tokens": 16515387.0,
"step": 8970
},
{
"entropy": 5.280672788619995,
"epoch": 0.7496658870698296,
"grad_norm": 0.99609375,
"learning_rate": 0.0004950094898762123,
"loss": 5.0877,
"mean_token_accuracy": 0.2006358951330185,
"num_tokens": 16524421.0,
"step": 8975
},
{
"entropy": 5.351682472229004,
"epoch": 0.7500835282325427,
"grad_norm": 0.8984375,
"learning_rate": 0.0004950032527115954,
"loss": 5.078,
"mean_token_accuracy": 0.1996570646762848,
"num_tokens": 16534533.0,
"step": 8980
},
{
"entropy": 5.312414073944092,
"epoch": 0.7505011693952556,
"grad_norm": 0.95703125,
"learning_rate": 0.0004949970116955566,
"loss": 5.0823,
"mean_token_accuracy": 0.19341331273317336,
"num_tokens": 16543712.0,
"step": 8985
},
{
"entropy": 5.277249717712403,
"epoch": 0.7509188105579686,
"grad_norm": 0.96484375,
"learning_rate": 0.0004949907668282054,
"loss": 5.0713,
"mean_token_accuracy": 0.20276295095682145,
"num_tokens": 16552475.0,
"step": 8990
},
{
"entropy": 5.320142984390259,
"epoch": 0.7513364517206816,
"grad_norm": 1.0,
"learning_rate": 0.000494984518109651,
"loss": 5.0511,
"mean_token_accuracy": 0.20469123274087905,
"num_tokens": 16562237.0,
"step": 8995
},
{
"entropy": 5.322559356689453,
"epoch": 0.7517540928833946,
"grad_norm": 1.0859375,
"learning_rate": 0.0004949782655400028,
"loss": 5.0621,
"mean_token_accuracy": 0.20243751555681228,
"num_tokens": 16570508.0,
"step": 9000
},
{
"epoch": 0.7517540928833946,
"eval_entropy": 4.992306073119494,
"eval_loss": 5.0120134353637695,
"eval_mean_token_accuracy": 0.214023694042507,
"eval_num_tokens": 16570508.0,
"eval_runtime": 20.8829,
"eval_samples_per_second": 1808.324,
"eval_steps_per_second": 226.07,
"step": 9000
},
{
"entropy": 5.1767021179199215,
"epoch": 0.7521717340461076,
"grad_norm": 0.94921875,
"learning_rate": 0.0004949720091193704,
"loss": 4.9614,
"mean_token_accuracy": 0.20973383337259294,
"num_tokens": 16578561.0,
"step": 9005
},
{
"entropy": 5.207372665405273,
"epoch": 0.7525893752088206,
"grad_norm": 0.984375,
"learning_rate": 0.000494965748847863,
"loss": 4.9397,
"mean_token_accuracy": 0.20836803168058396,
"num_tokens": 16587168.0,
"step": 9010
},
{
"entropy": 5.2924432277679445,
"epoch": 0.7530070163715336,
"grad_norm": 0.98828125,
"learning_rate": 0.0004949594847255905,
"loss": 5.0971,
"mean_token_accuracy": 0.19820506721735,
"num_tokens": 16597252.0,
"step": 9015
},
{
"entropy": 5.233920478820801,
"epoch": 0.7534246575342466,
"grad_norm": 0.95703125,
"learning_rate": 0.0004949532167526625,
"loss": 4.9273,
"mean_token_accuracy": 0.21284876763820648,
"num_tokens": 16607558.0,
"step": 9020
},
{
"entropy": 5.30990252494812,
"epoch": 0.7538422986969596,
"grad_norm": 0.96484375,
"learning_rate": 0.0004949469449291886,
"loss": 5.132,
"mean_token_accuracy": 0.1954716607928276,
"num_tokens": 16617019.0,
"step": 9025
},
{
"entropy": 5.280575466156006,
"epoch": 0.7542599398596725,
"grad_norm": 1.0625,
"learning_rate": 0.0004949406692552787,
"loss": 4.9777,
"mean_token_accuracy": 0.20624036192893982,
"num_tokens": 16625815.0,
"step": 9030
},
{
"entropy": 5.332900667190552,
"epoch": 0.7546775810223856,
"grad_norm": 1.09375,
"learning_rate": 0.0004949343897310426,
"loss": 5.1946,
"mean_token_accuracy": 0.19692480862140654,
"num_tokens": 16634876.0,
"step": 9035
},
{
"entropy": 5.277521419525146,
"epoch": 0.7550952221850986,
"grad_norm": 1.0,
"learning_rate": 0.0004949281063565904,
"loss": 5.0679,
"mean_token_accuracy": 0.19392486214637755,
"num_tokens": 16643770.0,
"step": 9040
},
{
"entropy": 5.327452850341797,
"epoch": 0.7555128633478115,
"grad_norm": 1.0859375,
"learning_rate": 0.0004949218191320319,
"loss": 5.0189,
"mean_token_accuracy": 0.19956471771001816,
"num_tokens": 16652845.0,
"step": 9045
},
{
"entropy": 5.228081560134887,
"epoch": 0.7559305045105246,
"grad_norm": 0.92578125,
"learning_rate": 0.0004949155280574772,
"loss": 5.004,
"mean_token_accuracy": 0.2051224708557129,
"num_tokens": 16662224.0,
"step": 9050
},
{
"entropy": 5.259872817993164,
"epoch": 0.7563481456732376,
"grad_norm": 0.8671875,
"learning_rate": 0.0004949092331330365,
"loss": 4.9909,
"mean_token_accuracy": 0.20357671678066253,
"num_tokens": 16671114.0,
"step": 9055
},
{
"entropy": 5.279949378967285,
"epoch": 0.7567657868359505,
"grad_norm": 0.98046875,
"learning_rate": 0.00049490293435882,
"loss": 5.0722,
"mean_token_accuracy": 0.19587585479021072,
"num_tokens": 16680395.0,
"step": 9060
},
{
"entropy": 5.3000411033630375,
"epoch": 0.7571834279986636,
"grad_norm": 0.97265625,
"learning_rate": 0.0004948966317349379,
"loss": 5.0347,
"mean_token_accuracy": 0.20271148532629013,
"num_tokens": 16690251.0,
"step": 9065
},
{
"entropy": 5.227061796188354,
"epoch": 0.7576010691613766,
"grad_norm": 0.98828125,
"learning_rate": 0.0004948903252615006,
"loss": 4.9214,
"mean_token_accuracy": 0.2127152979373932,
"num_tokens": 16698945.0,
"step": 9070
},
{
"entropy": 5.328841876983643,
"epoch": 0.7580187103240895,
"grad_norm": 1.015625,
"learning_rate": 0.0004948840149386184,
"loss": 5.1207,
"mean_token_accuracy": 0.1974414274096489,
"num_tokens": 16708325.0,
"step": 9075
},
{
"entropy": 5.257896661758423,
"epoch": 0.7584363514868026,
"grad_norm": 1.0703125,
"learning_rate": 0.000494877700766402,
"loss": 5.0149,
"mean_token_accuracy": 0.20404049456119538,
"num_tokens": 16716392.0,
"step": 9080
},
{
"entropy": 5.272157335281372,
"epoch": 0.7588539926495156,
"grad_norm": 0.93359375,
"learning_rate": 0.0004948713827449616,
"loss": 5.0401,
"mean_token_accuracy": 0.20416046380996705,
"num_tokens": 16725284.0,
"step": 9085
},
{
"entropy": 5.239689540863037,
"epoch": 0.7592716338122285,
"grad_norm": 0.89453125,
"learning_rate": 0.0004948650608744081,
"loss": 5.0596,
"mean_token_accuracy": 0.20529814809560776,
"num_tokens": 16734658.0,
"step": 9090
},
{
"entropy": 5.380432224273681,
"epoch": 0.7596892749749415,
"grad_norm": 0.94921875,
"learning_rate": 0.000494858735154852,
"loss": 5.16,
"mean_token_accuracy": 0.19234798699617386,
"num_tokens": 16744403.0,
"step": 9095
},
{
"entropy": 5.366984844207764,
"epoch": 0.7601069161376546,
"grad_norm": 0.8671875,
"learning_rate": 0.0004948524055864041,
"loss": 5.133,
"mean_token_accuracy": 0.1997056409716606,
"num_tokens": 16754254.0,
"step": 9100
},
{
"entropy": 5.326011705398559,
"epoch": 0.7605245573003675,
"grad_norm": 0.92578125,
"learning_rate": 0.0004948460721691752,
"loss": 5.1359,
"mean_token_accuracy": 0.19720993041992188,
"num_tokens": 16763941.0,
"step": 9105
},
{
"entropy": 5.295243883132935,
"epoch": 0.7609421984630805,
"grad_norm": 1.0,
"learning_rate": 0.0004948397349032762,
"loss": 5.0422,
"mean_token_accuracy": 0.20102442502975465,
"num_tokens": 16773181.0,
"step": 9110
},
{
"entropy": 5.1932860851287845,
"epoch": 0.7613598396257936,
"grad_norm": 1.0,
"learning_rate": 0.000494833393788818,
"loss": 4.8807,
"mean_token_accuracy": 0.21331206560134888,
"num_tokens": 16782622.0,
"step": 9115
},
{
"entropy": 5.314014720916748,
"epoch": 0.7617774807885065,
"grad_norm": 1.046875,
"learning_rate": 0.0004948270488259117,
"loss": 5.1199,
"mean_token_accuracy": 0.20172844231128692,
"num_tokens": 16792805.0,
"step": 9120
},
{
"entropy": 5.306127071380615,
"epoch": 0.7621951219512195,
"grad_norm": 0.8984375,
"learning_rate": 0.0004948207000146682,
"loss": 5.0881,
"mean_token_accuracy": 0.20799920409917833,
"num_tokens": 16802641.0,
"step": 9125
},
{
"entropy": 5.340350389480591,
"epoch": 0.7626127631139326,
"grad_norm": 0.96484375,
"learning_rate": 0.0004948143473551989,
"loss": 5.0775,
"mean_token_accuracy": 0.19613206535577773,
"num_tokens": 16812037.0,
"step": 9130
},
{
"entropy": 5.274501514434815,
"epoch": 0.7630304042766455,
"grad_norm": 1.03125,
"learning_rate": 0.0004948079908476147,
"loss": 5.0327,
"mean_token_accuracy": 0.21585725247859955,
"num_tokens": 16820279.0,
"step": 9135
},
{
"entropy": 5.269076347351074,
"epoch": 0.7634480454393585,
"grad_norm": 0.97265625,
"learning_rate": 0.0004948016304920271,
"loss": 5.0336,
"mean_token_accuracy": 0.19964986592531203,
"num_tokens": 16829396.0,
"step": 9140
},
{
"entropy": 5.275078868865966,
"epoch": 0.7638656866020715,
"grad_norm": 0.98828125,
"learning_rate": 0.0004947952662885473,
"loss": 5.011,
"mean_token_accuracy": 0.20565113723278045,
"num_tokens": 16838389.0,
"step": 9145
},
{
"entropy": 5.271355295181275,
"epoch": 0.7642833277647845,
"grad_norm": 0.95703125,
"learning_rate": 0.000494788898237287,
"loss": 5.1171,
"mean_token_accuracy": 0.19436787068843842,
"num_tokens": 16848012.0,
"step": 9150
},
{
"entropy": 5.31104211807251,
"epoch": 0.7647009689274975,
"grad_norm": 0.875,
"learning_rate": 0.0004947825263383573,
"loss": 5.0785,
"mean_token_accuracy": 0.20117024183273316,
"num_tokens": 16857591.0,
"step": 9155
},
{
"entropy": 5.35554838180542,
"epoch": 0.7651186100902105,
"grad_norm": 0.91796875,
"learning_rate": 0.00049477615059187,
"loss": 5.0821,
"mean_token_accuracy": 0.1895461291074753,
"num_tokens": 16866792.0,
"step": 9160
},
{
"entropy": 5.259443235397339,
"epoch": 0.7655362512529235,
"grad_norm": 0.95703125,
"learning_rate": 0.0004947697709979367,
"loss": 5.0996,
"mean_token_accuracy": 0.2034372240304947,
"num_tokens": 16876602.0,
"step": 9165
},
{
"entropy": 5.301667547225952,
"epoch": 0.7659538924156365,
"grad_norm": 1.0234375,
"learning_rate": 0.0004947633875566688,
"loss": 5.0758,
"mean_token_accuracy": 0.1992911234498024,
"num_tokens": 16886992.0,
"step": 9170
},
{
"entropy": 5.27959041595459,
"epoch": 0.7663715335783495,
"grad_norm": 0.95703125,
"learning_rate": 0.0004947570002681785,
"loss": 5.0027,
"mean_token_accuracy": 0.2038028672337532,
"num_tokens": 16895716.0,
"step": 9175
},
{
"entropy": 5.241288042068481,
"epoch": 0.7667891747410625,
"grad_norm": 0.97265625,
"learning_rate": 0.0004947506091325772,
"loss": 5.0134,
"mean_token_accuracy": 0.21356237679719925,
"num_tokens": 16904568.0,
"step": 9180
},
{
"entropy": 5.292680788040161,
"epoch": 0.7672068159037755,
"grad_norm": 1.0546875,
"learning_rate": 0.0004947442141499771,
"loss": 5.1022,
"mean_token_accuracy": 0.19897875487804412,
"num_tokens": 16913667.0,
"step": 9185
},
{
"entropy": 5.326617431640625,
"epoch": 0.7676244570664885,
"grad_norm": 1.015625,
"learning_rate": 0.0004947378153204899,
"loss": 5.0406,
"mean_token_accuracy": 0.2004725456237793,
"num_tokens": 16922897.0,
"step": 9190
},
{
"entropy": 5.349248886108398,
"epoch": 0.7680420982292014,
"grad_norm": 0.98828125,
"learning_rate": 0.0004947314126442279,
"loss": 5.0406,
"mean_token_accuracy": 0.20128468424081802,
"num_tokens": 16931559.0,
"step": 9195
},
{
"entropy": 5.26908860206604,
"epoch": 0.7684597393919145,
"grad_norm": 0.97265625,
"learning_rate": 0.000494725006121303,
"loss": 5.0244,
"mean_token_accuracy": 0.20029380321502685,
"num_tokens": 16940191.0,
"step": 9200
},
{
"entropy": 5.308033800125122,
"epoch": 0.7688773805546275,
"grad_norm": 0.98828125,
"learning_rate": 0.0004947185957518273,
"loss": 5.0948,
"mean_token_accuracy": 0.19918156713247298,
"num_tokens": 16948971.0,
"step": 9205
},
{
"entropy": 5.313927936553955,
"epoch": 0.7692950217173404,
"grad_norm": 1.0703125,
"learning_rate": 0.0004947121815359133,
"loss": 5.0277,
"mean_token_accuracy": 0.206220580637455,
"num_tokens": 16957944.0,
"step": 9210
},
{
"entropy": 5.229868936538696,
"epoch": 0.7697126628800535,
"grad_norm": 1.0078125,
"learning_rate": 0.0004947057634736729,
"loss": 4.9983,
"mean_token_accuracy": 0.20346777737140656,
"num_tokens": 16967418.0,
"step": 9215
},
{
"entropy": 5.260604810714722,
"epoch": 0.7701303040427665,
"grad_norm": 0.9609375,
"learning_rate": 0.0004946993415652186,
"loss": 4.9869,
"mean_token_accuracy": 0.20868378430604934,
"num_tokens": 16977233.0,
"step": 9220
},
{
"entropy": 5.261973714828491,
"epoch": 0.7705479452054794,
"grad_norm": 0.99609375,
"learning_rate": 0.000494692915810663,
"loss": 5.0621,
"mean_token_accuracy": 0.20642494261264802,
"num_tokens": 16986023.0,
"step": 9225
},
{
"entropy": 5.281343030929565,
"epoch": 0.7709655863681925,
"grad_norm": 0.92578125,
"learning_rate": 0.0004946864862101185,
"loss": 4.9798,
"mean_token_accuracy": 0.20381157398223876,
"num_tokens": 16994987.0,
"step": 9230
},
{
"entropy": 5.260571432113648,
"epoch": 0.7713832275309055,
"grad_norm": 1.0234375,
"learning_rate": 0.0004946800527636975,
"loss": 5.0105,
"mean_token_accuracy": 0.19908543974161147,
"num_tokens": 17003902.0,
"step": 9235
},
{
"entropy": 5.285253715515137,
"epoch": 0.7718008686936184,
"grad_norm": 0.97265625,
"learning_rate": 0.0004946736154715129,
"loss": 5.1003,
"mean_token_accuracy": 0.19527377486228942,
"num_tokens": 17013258.0,
"step": 9240
},
{
"entropy": 5.367343902587891,
"epoch": 0.7722185098563314,
"grad_norm": 1.0390625,
"learning_rate": 0.0004946671743336771,
"loss": 5.0965,
"mean_token_accuracy": 0.20139744728803635,
"num_tokens": 17022258.0,
"step": 9245
},
{
"entropy": 5.3241571426391605,
"epoch": 0.7726361510190445,
"grad_norm": 1.0,
"learning_rate": 0.0004946607293503031,
"loss": 5.019,
"mean_token_accuracy": 0.2053925946354866,
"num_tokens": 17031485.0,
"step": 9250
},
{
"entropy": 5.171794891357422,
"epoch": 0.7730537921817574,
"grad_norm": 1.0546875,
"learning_rate": 0.0004946542805215036,
"loss": 4.9819,
"mean_token_accuracy": 0.20774541199207305,
"num_tokens": 17039916.0,
"step": 9255
},
{
"entropy": 5.177844619750976,
"epoch": 0.7734714333444704,
"grad_norm": 1.0703125,
"learning_rate": 0.0004946478278473914,
"loss": 5.0161,
"mean_token_accuracy": 0.2033155858516693,
"num_tokens": 17048153.0,
"step": 9260
},
{
"entropy": 5.262477922439575,
"epoch": 0.7738890745071835,
"grad_norm": 0.99609375,
"learning_rate": 0.0004946413713280797,
"loss": 5.0453,
"mean_token_accuracy": 0.20080723911523818,
"num_tokens": 17057473.0,
"step": 9265
},
{
"entropy": 5.338902425765991,
"epoch": 0.7743067156698964,
"grad_norm": 0.92578125,
"learning_rate": 0.0004946349109636814,
"loss": 5.0365,
"mean_token_accuracy": 0.2072010949254036,
"num_tokens": 17066431.0,
"step": 9270
},
{
"entropy": 5.333625078201294,
"epoch": 0.7747243568326094,
"grad_norm": 0.98828125,
"learning_rate": 0.0004946284467543096,
"loss": 5.136,
"mean_token_accuracy": 0.1943159893155098,
"num_tokens": 17076519.0,
"step": 9275
},
{
"entropy": 5.3112376689910885,
"epoch": 0.7751419979953225,
"grad_norm": 0.9609375,
"learning_rate": 0.0004946219787000776,
"loss": 5.0759,
"mean_token_accuracy": 0.20009991973638536,
"num_tokens": 17084651.0,
"step": 9280
},
{
"entropy": 5.333436107635498,
"epoch": 0.7755596391580354,
"grad_norm": 0.8984375,
"learning_rate": 0.0004946155068010984,
"loss": 5.1035,
"mean_token_accuracy": 0.19852675050497054,
"num_tokens": 17094943.0,
"step": 9285
},
{
"entropy": 5.28173794746399,
"epoch": 0.7759772803207484,
"grad_norm": 1.03125,
"learning_rate": 0.0004946090310574854,
"loss": 4.9809,
"mean_token_accuracy": 0.20821188241243363,
"num_tokens": 17103548.0,
"step": 9290
},
{
"entropy": 5.295103359222412,
"epoch": 0.7763949214834615,
"grad_norm": 1.0546875,
"learning_rate": 0.0004946025514693519,
"loss": 4.9481,
"mean_token_accuracy": 0.20553445667028428,
"num_tokens": 17112472.0,
"step": 9295
},
{
"entropy": 5.4018957138061525,
"epoch": 0.7768125626461744,
"grad_norm": 1.0859375,
"learning_rate": 0.0004945960680368115,
"loss": 5.0829,
"mean_token_accuracy": 0.19831764400005342,
"num_tokens": 17121536.0,
"step": 9300
},
{
"entropy": 5.416180658340454,
"epoch": 0.7772302038088874,
"grad_norm": 1.0546875,
"learning_rate": 0.0004945895807599775,
"loss": 5.1467,
"mean_token_accuracy": 0.19245276004076003,
"num_tokens": 17130708.0,
"step": 9305
},
{
"entropy": 5.215561962127685,
"epoch": 0.7776478449716004,
"grad_norm": 1.046875,
"learning_rate": 0.0004945830896389636,
"loss": 4.9334,
"mean_token_accuracy": 0.2095179870724678,
"num_tokens": 17139437.0,
"step": 9310
},
{
"entropy": 5.276777553558349,
"epoch": 0.7780654861343134,
"grad_norm": 0.97265625,
"learning_rate": 0.0004945765946738835,
"loss": 5.083,
"mean_token_accuracy": 0.1986304357647896,
"num_tokens": 17149107.0,
"step": 9315
},
{
"entropy": 5.3130237579345705,
"epoch": 0.7784831272970264,
"grad_norm": 0.98046875,
"learning_rate": 0.0004945700958648507,
"loss": 5.0785,
"mean_token_accuracy": 0.19797585159540176,
"num_tokens": 17158866.0,
"step": 9320
},
{
"entropy": 5.261892318725586,
"epoch": 0.7789007684597394,
"grad_norm": 1.1015625,
"learning_rate": 0.000494563593211979,
"loss": 4.9753,
"mean_token_accuracy": 0.2056022822856903,
"num_tokens": 17167909.0,
"step": 9325
},
{
"entropy": 5.2609766006469725,
"epoch": 0.7793184096224524,
"grad_norm": 0.9765625,
"learning_rate": 0.0004945570867153826,
"loss": 5.0885,
"mean_token_accuracy": 0.19690562188625335,
"num_tokens": 17176746.0,
"step": 9330
},
{
"entropy": 5.304251098632813,
"epoch": 0.7797360507851654,
"grad_norm": 0.9765625,
"learning_rate": 0.0004945505763751749,
"loss": 5.0917,
"mean_token_accuracy": 0.19543183445930482,
"num_tokens": 17186219.0,
"step": 9335
},
{
"entropy": 5.298954486846924,
"epoch": 0.7801536919478784,
"grad_norm": 1.0625,
"learning_rate": 0.0004945440621914702,
"loss": 4.9259,
"mean_token_accuracy": 0.21116489320993423,
"num_tokens": 17195811.0,
"step": 9340
},
{
"entropy": 5.1953223705291744,
"epoch": 0.7805713331105913,
"grad_norm": 1.046875,
"learning_rate": 0.0004945375441643823,
"loss": 5.0755,
"mean_token_accuracy": 0.2034341961145401,
"num_tokens": 17205333.0,
"step": 9345
},
{
"entropy": 5.299469518661499,
"epoch": 0.7809889742733044,
"grad_norm": 1.015625,
"learning_rate": 0.0004945310222940254,
"loss": 5.0366,
"mean_token_accuracy": 0.19907909333705903,
"num_tokens": 17214493.0,
"step": 9350
},
{
"entropy": 5.365374946594239,
"epoch": 0.7814066154360174,
"grad_norm": 0.94921875,
"learning_rate": 0.0004945244965805138,
"loss": 5.0419,
"mean_token_accuracy": 0.20529650002717972,
"num_tokens": 17223602.0,
"step": 9355
},
{
"entropy": 5.335985469818115,
"epoch": 0.7818242565987303,
"grad_norm": 1.03125,
"learning_rate": 0.0004945179670239616,
"loss": 5.0758,
"mean_token_accuracy": 0.19936992079019547,
"num_tokens": 17232915.0,
"step": 9360
},
{
"entropy": 5.320971441268921,
"epoch": 0.7822418977614434,
"grad_norm": 1.15625,
"learning_rate": 0.0004945114336244832,
"loss": 5.0456,
"mean_token_accuracy": 0.1973716601729393,
"num_tokens": 17242489.0,
"step": 9365
},
{
"entropy": 5.188671016693116,
"epoch": 0.7826595389241564,
"grad_norm": 0.96484375,
"learning_rate": 0.000494504896382193,
"loss": 4.8118,
"mean_token_accuracy": 0.2163364425301552,
"num_tokens": 17251496.0,
"step": 9370
},
{
"entropy": 5.214679002761841,
"epoch": 0.7830771800868693,
"grad_norm": 0.9921875,
"learning_rate": 0.0004944983552972053,
"loss": 4.9942,
"mean_token_accuracy": 0.20377425700426102,
"num_tokens": 17260550.0,
"step": 9375
},
{
"entropy": 5.271979284286499,
"epoch": 0.7834948212495824,
"grad_norm": 1.0625,
"learning_rate": 0.0004944918103696347,
"loss": 4.9728,
"mean_token_accuracy": 0.20010923892259597,
"num_tokens": 17269085.0,
"step": 9380
},
{
"entropy": 5.366032314300537,
"epoch": 0.7839124624122954,
"grad_norm": 1.015625,
"learning_rate": 0.0004944852615995956,
"loss": 5.198,
"mean_token_accuracy": 0.20187092423439026,
"num_tokens": 17278766.0,
"step": 9385
},
{
"entropy": 5.406021165847778,
"epoch": 0.7843301035750083,
"grad_norm": 1.046875,
"learning_rate": 0.000494478708987203,
"loss": 5.06,
"mean_token_accuracy": 0.19868176728487014,
"num_tokens": 17288316.0,
"step": 9390
},
{
"entropy": 5.229649305343628,
"epoch": 0.7847477447377214,
"grad_norm": 0.953125,
"learning_rate": 0.0004944721525325713,
"loss": 4.9167,
"mean_token_accuracy": 0.20666508078575135,
"num_tokens": 17297829.0,
"step": 9395
},
{
"entropy": 5.273958492279053,
"epoch": 0.7851653859004344,
"grad_norm": 1.0390625,
"learning_rate": 0.0004944655922358154,
"loss": 5.0016,
"mean_token_accuracy": 0.2018284633755684,
"num_tokens": 17306631.0,
"step": 9400
},
{
"entropy": 5.26501317024231,
"epoch": 0.7855830270631473,
"grad_norm": 0.83203125,
"learning_rate": 0.0004944590280970503,
"loss": 5.0067,
"mean_token_accuracy": 0.20416941344738007,
"num_tokens": 17316829.0,
"step": 9405
},
{
"entropy": 5.283139324188232,
"epoch": 0.7860006682258603,
"grad_norm": 0.875,
"learning_rate": 0.0004944524601163905,
"loss": 5.0208,
"mean_token_accuracy": 0.2009298622608185,
"num_tokens": 17326568.0,
"step": 9410
},
{
"entropy": 5.306204080581665,
"epoch": 0.7864183093885734,
"grad_norm": 0.9609375,
"learning_rate": 0.0004944458882939513,
"loss": 5.0835,
"mean_token_accuracy": 0.20056407451629638,
"num_tokens": 17335524.0,
"step": 9415
},
{
"entropy": 5.39315767288208,
"epoch": 0.7868359505512863,
"grad_norm": 0.96484375,
"learning_rate": 0.0004944393126298478,
"loss": 5.1221,
"mean_token_accuracy": 0.19416031390428543,
"num_tokens": 17346198.0,
"step": 9420
},
{
"entropy": 5.251412725448608,
"epoch": 0.7872535917139993,
"grad_norm": 1.0625,
"learning_rate": 0.000494432733124195,
"loss": 4.9666,
"mean_token_accuracy": 0.21045693457126619,
"num_tokens": 17354696.0,
"step": 9425
},
{
"entropy": 5.196813678741455,
"epoch": 0.7876712328767124,
"grad_norm": 0.91796875,
"learning_rate": 0.000494426149777108,
"loss": 4.944,
"mean_token_accuracy": 0.20504340827465056,
"num_tokens": 17363314.0,
"step": 9430
},
{
"entropy": 5.226676893234253,
"epoch": 0.7880888740394253,
"grad_norm": 0.91796875,
"learning_rate": 0.000494419562588702,
"loss": 4.9947,
"mean_token_accuracy": 0.21016329824924468,
"num_tokens": 17372662.0,
"step": 9435
},
{
"entropy": 5.234115219116211,
"epoch": 0.7885065152021383,
"grad_norm": 0.91796875,
"learning_rate": 0.0004944129715590927,
"loss": 4.9701,
"mean_token_accuracy": 0.20339612364768983,
"num_tokens": 17381931.0,
"step": 9440
},
{
"entropy": 5.330491971969605,
"epoch": 0.7889241563648514,
"grad_norm": 1.0234375,
"learning_rate": 0.000494406376688395,
"loss": 5.0671,
"mean_token_accuracy": 0.19536355137825012,
"num_tokens": 17391210.0,
"step": 9445
},
{
"entropy": 5.225590372085572,
"epoch": 0.7893417975275643,
"grad_norm": 0.93359375,
"learning_rate": 0.0004943997779767248,
"loss": 4.9328,
"mean_token_accuracy": 0.2165209487080574,
"num_tokens": 17401102.0,
"step": 9450
},
{
"entropy": 5.222544002532959,
"epoch": 0.7897594386902773,
"grad_norm": 0.89453125,
"learning_rate": 0.0004943931754241972,
"loss": 4.9798,
"mean_token_accuracy": 0.20373812019824983,
"num_tokens": 17410036.0,
"step": 9455
},
{
"entropy": 5.255673933029175,
"epoch": 0.7901770798529903,
"grad_norm": 0.92578125,
"learning_rate": 0.000494386569030928,
"loss": 5.0323,
"mean_token_accuracy": 0.19926638454198836,
"num_tokens": 17419348.0,
"step": 9460
},
{
"entropy": 5.296061849594116,
"epoch": 0.7905947210157033,
"grad_norm": 1.0859375,
"learning_rate": 0.0004943799587970328,
"loss": 5.0361,
"mean_token_accuracy": 0.20071749538183212,
"num_tokens": 17427937.0,
"step": 9465
},
{
"entropy": 5.224281358718872,
"epoch": 0.7910123621784163,
"grad_norm": 0.96875,
"learning_rate": 0.0004943733447226275,
"loss": 4.9016,
"mean_token_accuracy": 0.21215202659368515,
"num_tokens": 17436259.0,
"step": 9470
},
{
"entropy": 5.342166805267334,
"epoch": 0.7914300033411293,
"grad_norm": 1.0390625,
"learning_rate": 0.0004943667268078277,
"loss": 5.084,
"mean_token_accuracy": 0.20154720097780227,
"num_tokens": 17445202.0,
"step": 9475
},
{
"entropy": 5.254055738449097,
"epoch": 0.7918476445038423,
"grad_norm": 0.93359375,
"learning_rate": 0.0004943601050527493,
"loss": 4.988,
"mean_token_accuracy": 0.20560846477746964,
"num_tokens": 17453928.0,
"step": 9480
},
{
"entropy": 5.271201419830322,
"epoch": 0.7922652856665553,
"grad_norm": 0.9609375,
"learning_rate": 0.0004943534794575081,
"loss": 5.0149,
"mean_token_accuracy": 0.2042636454105377,
"num_tokens": 17462609.0,
"step": 9485
},
{
"entropy": 5.307225608825684,
"epoch": 0.7926829268292683,
"grad_norm": 1.0703125,
"learning_rate": 0.0004943468500222203,
"loss": 5.03,
"mean_token_accuracy": 0.20491330027580262,
"num_tokens": 17471326.0,
"step": 9490
},
{
"entropy": 5.229542350769043,
"epoch": 0.7931005679919813,
"grad_norm": 0.9765625,
"learning_rate": 0.0004943402167470019,
"loss": 4.9964,
"mean_token_accuracy": 0.2107575312256813,
"num_tokens": 17480296.0,
"step": 9495
},
{
"entropy": 5.2251122951507565,
"epoch": 0.7935182091546943,
"grad_norm": 1.015625,
"learning_rate": 0.000494333579631969,
"loss": 4.9284,
"mean_token_accuracy": 0.2042476013302803,
"num_tokens": 17489779.0,
"step": 9500
},
{
"entropy": 5.2869805812835695,
"epoch": 0.7939358503174073,
"grad_norm": 0.96484375,
"learning_rate": 0.0004943269386772376,
"loss": 5.0489,
"mean_token_accuracy": 0.20046651661396026,
"num_tokens": 17499796.0,
"step": 9505
},
{
"entropy": 5.335369062423706,
"epoch": 0.7943534914801202,
"grad_norm": 0.9453125,
"learning_rate": 0.0004943202938829243,
"loss": 5.1443,
"mean_token_accuracy": 0.19563747197389603,
"num_tokens": 17509223.0,
"step": 9510
},
{
"entropy": 5.296935319900513,
"epoch": 0.7947711326428333,
"grad_norm": 1.015625,
"learning_rate": 0.0004943136452491452,
"loss": 5.0388,
"mean_token_accuracy": 0.20008025765419007,
"num_tokens": 17518712.0,
"step": 9515
},
{
"entropy": 5.238116598129272,
"epoch": 0.7951887738055463,
"grad_norm": 0.91796875,
"learning_rate": 0.0004943069927760168,
"loss": 4.9883,
"mean_token_accuracy": 0.20782812386751176,
"num_tokens": 17528648.0,
"step": 9520
},
{
"entropy": 5.161343383789062,
"epoch": 0.7956064149682592,
"grad_norm": 1.0390625,
"learning_rate": 0.0004943003364636554,
"loss": 4.9847,
"mean_token_accuracy": 0.20069538652896882,
"num_tokens": 17537859.0,
"step": 9525
},
{
"entropy": 5.316619110107422,
"epoch": 0.7960240561309723,
"grad_norm": 1.1171875,
"learning_rate": 0.0004942936763121777,
"loss": 5.1395,
"mean_token_accuracy": 0.19529732316732407,
"num_tokens": 17546438.0,
"step": 9530
},
{
"entropy": 5.2499274730682375,
"epoch": 0.7964416972936853,
"grad_norm": 1.125,
"learning_rate": 0.0004942870123217002,
"loss": 4.9569,
"mean_token_accuracy": 0.21053215265274047,
"num_tokens": 17554537.0,
"step": 9535
},
{
"entropy": 5.385514211654663,
"epoch": 0.7968593384563982,
"grad_norm": 1.09375,
"learning_rate": 0.0004942803444923398,
"loss": 5.1864,
"mean_token_accuracy": 0.19254078716039658,
"num_tokens": 17564239.0,
"step": 9540
},
{
"entropy": 5.268808794021607,
"epoch": 0.7972769796191113,
"grad_norm": 0.88671875,
"learning_rate": 0.0004942736728242127,
"loss": 5.0497,
"mean_token_accuracy": 0.20318469703197478,
"num_tokens": 17574437.0,
"step": 9545
},
{
"entropy": 5.250163650512695,
"epoch": 0.7976946207818243,
"grad_norm": 0.9609375,
"learning_rate": 0.0004942669973174363,
"loss": 4.9888,
"mean_token_accuracy": 0.20635224431753157,
"num_tokens": 17583173.0,
"step": 9550
},
{
"entropy": 5.293168735504151,
"epoch": 0.7981122619445372,
"grad_norm": 0.984375,
"learning_rate": 0.000494260317972127,
"loss": 5.0292,
"mean_token_accuracy": 0.1972663074731827,
"num_tokens": 17591731.0,
"step": 9555
},
{
"entropy": 5.30541467666626,
"epoch": 0.7985299031072502,
"grad_norm": 0.86328125,
"learning_rate": 0.0004942536347884019,
"loss": 5.0649,
"mean_token_accuracy": 0.20194061398506163,
"num_tokens": 17600778.0,
"step": 9560
},
{
"entropy": 5.220264863967896,
"epoch": 0.7989475442699633,
"grad_norm": 0.99609375,
"learning_rate": 0.0004942469477663782,
"loss": 5.0326,
"mean_token_accuracy": 0.21178349554538728,
"num_tokens": 17609470.0,
"step": 9565
},
{
"entropy": 5.22754225730896,
"epoch": 0.7993651854326762,
"grad_norm": 0.96484375,
"learning_rate": 0.0004942402569061726,
"loss": 4.9282,
"mean_token_accuracy": 0.20935079008340834,
"num_tokens": 17618434.0,
"step": 9570
},
{
"entropy": 5.216368770599365,
"epoch": 0.7997828265953892,
"grad_norm": 0.984375,
"learning_rate": 0.0004942335622079025,
"loss": 5.0486,
"mean_token_accuracy": 0.19940815418958663,
"num_tokens": 17628170.0,
"step": 9575
},
{
"entropy": 5.2389692783355715,
"epoch": 0.8002004677581023,
"grad_norm": 0.9375,
"learning_rate": 0.0004942268636716848,
"loss": 4.9611,
"mean_token_accuracy": 0.20548074841499328,
"num_tokens": 17637391.0,
"step": 9580
},
{
"entropy": 5.275997018814087,
"epoch": 0.8006181089208152,
"grad_norm": 0.94140625,
"learning_rate": 0.0004942201612976372,
"loss": 5.0586,
"mean_token_accuracy": 0.20423658937215805,
"num_tokens": 17647466.0,
"step": 9585
},
{
"entropy": 5.277049112319946,
"epoch": 0.8010357500835282,
"grad_norm": 1.0078125,
"learning_rate": 0.0004942134550858767,
"loss": 4.9855,
"mean_token_accuracy": 0.209126316010952,
"num_tokens": 17656155.0,
"step": 9590
},
{
"entropy": 5.385107898712159,
"epoch": 0.8014533912462413,
"grad_norm": 1.0078125,
"learning_rate": 0.0004942067450365208,
"loss": 5.0747,
"mean_token_accuracy": 0.20015188902616501,
"num_tokens": 17665242.0,
"step": 9595
},
{
"entropy": 5.283998966217041,
"epoch": 0.8018710324089542,
"grad_norm": 0.94140625,
"learning_rate": 0.0004942000311496871,
"loss": 5.0958,
"mean_token_accuracy": 0.20111240893602372,
"num_tokens": 17674964.0,
"step": 9600
},
{
"entropy": 5.351830673217774,
"epoch": 0.8022886735716672,
"grad_norm": 1.0390625,
"learning_rate": 0.0004941933134254928,
"loss": 5.0927,
"mean_token_accuracy": 0.2037873536348343,
"num_tokens": 17683275.0,
"step": 9605
},
{
"entropy": 5.297152185440064,
"epoch": 0.8027063147343803,
"grad_norm": 1.0859375,
"learning_rate": 0.0004941865918640559,
"loss": 5.048,
"mean_token_accuracy": 0.20206041336059571,
"num_tokens": 17691841.0,
"step": 9610
},
{
"entropy": 5.292528247833252,
"epoch": 0.8031239558970932,
"grad_norm": 1.046875,
"learning_rate": 0.0004941798664654937,
"loss": 4.9972,
"mean_token_accuracy": 0.2060507968068123,
"num_tokens": 17701258.0,
"step": 9615
},
{
"entropy": 5.2792106628417965,
"epoch": 0.8035415970598062,
"grad_norm": 0.91796875,
"learning_rate": 0.0004941731372299241,
"loss": 5.082,
"mean_token_accuracy": 0.19895916879177095,
"num_tokens": 17711028.0,
"step": 9620
},
{
"entropy": 5.31102819442749,
"epoch": 0.8039592382225192,
"grad_norm": 0.9609375,
"learning_rate": 0.0004941664041574651,
"loss": 5.0215,
"mean_token_accuracy": 0.21228455603122712,
"num_tokens": 17720175.0,
"step": 9625
},
{
"entropy": 5.288550138473511,
"epoch": 0.8043768793852322,
"grad_norm": 1.109375,
"learning_rate": 0.0004941596672482343,
"loss": 4.9978,
"mean_token_accuracy": 0.20317634344100952,
"num_tokens": 17730420.0,
"step": 9630
},
{
"entropy": 5.2999168872833256,
"epoch": 0.8047945205479452,
"grad_norm": 1.0234375,
"learning_rate": 0.0004941529265023498,
"loss": 5.0976,
"mean_token_accuracy": 0.19421206712722777,
"num_tokens": 17739802.0,
"step": 9635
},
{
"entropy": 5.3412566661834715,
"epoch": 0.8052121617106582,
"grad_norm": 1.0078125,
"learning_rate": 0.0004941461819199293,
"loss": 4.9901,
"mean_token_accuracy": 0.2086260050535202,
"num_tokens": 17749450.0,
"step": 9640
},
{
"entropy": 5.405754327774048,
"epoch": 0.8056298028733712,
"grad_norm": 0.98046875,
"learning_rate": 0.0004941394335010912,
"loss": 5.1788,
"mean_token_accuracy": 0.19541946202516555,
"num_tokens": 17758733.0,
"step": 9645
},
{
"entropy": 5.2307861328125,
"epoch": 0.8060474440360842,
"grad_norm": 0.96484375,
"learning_rate": 0.0004941326812459535,
"loss": 5.1042,
"mean_token_accuracy": 0.1989505708217621,
"num_tokens": 17769037.0,
"step": 9650
},
{
"entropy": 5.287183332443237,
"epoch": 0.8064650851987972,
"grad_norm": 0.99609375,
"learning_rate": 0.0004941259251546346,
"loss": 5.0461,
"mean_token_accuracy": 0.19478376358747482,
"num_tokens": 17777603.0,
"step": 9655
},
{
"entropy": 5.292373275756836,
"epoch": 0.8068827263615101,
"grad_norm": 0.98828125,
"learning_rate": 0.0004941191652272526,
"loss": 5.014,
"mean_token_accuracy": 0.1983320191502571,
"num_tokens": 17787513.0,
"step": 9660
},
{
"entropy": 5.241600799560547,
"epoch": 0.8073003675242232,
"grad_norm": 0.95703125,
"learning_rate": 0.0004941124014639258,
"loss": 5.019,
"mean_token_accuracy": 0.20129087418317795,
"num_tokens": 17796720.0,
"step": 9665
},
{
"entropy": 5.264516973495484,
"epoch": 0.8077180086869362,
"grad_norm": 1.03125,
"learning_rate": 0.0004941056338647726,
"loss": 4.9849,
"mean_token_accuracy": 0.2036397188901901,
"num_tokens": 17806024.0,
"step": 9670
},
{
"entropy": 5.221023321151733,
"epoch": 0.8081356498496491,
"grad_norm": 0.91015625,
"learning_rate": 0.0004940988624299116,
"loss": 4.976,
"mean_token_accuracy": 0.2046495258808136,
"num_tokens": 17815247.0,
"step": 9675
},
{
"entropy": 5.323753118515015,
"epoch": 0.8085532910123622,
"grad_norm": 1.03125,
"learning_rate": 0.0004940920871594613,
"loss": 5.0366,
"mean_token_accuracy": 0.20503654927015305,
"num_tokens": 17824024.0,
"step": 9680
},
{
"entropy": 5.247210264205933,
"epoch": 0.8089709321750752,
"grad_norm": 1.09375,
"learning_rate": 0.0004940853080535403,
"loss": 5.0058,
"mean_token_accuracy": 0.20110260248184203,
"num_tokens": 17832782.0,
"step": 9685
},
{
"entropy": 5.293553590774536,
"epoch": 0.8093885733377881,
"grad_norm": 1.0078125,
"learning_rate": 0.0004940785251122673,
"loss": 4.976,
"mean_token_accuracy": 0.21062718480825424,
"num_tokens": 17841961.0,
"step": 9690
},
{
"entropy": 5.254729890823365,
"epoch": 0.8098062145005012,
"grad_norm": 1.0078125,
"learning_rate": 0.0004940717383357608,
"loss": 5.0608,
"mean_token_accuracy": 0.21443141251802444,
"num_tokens": 17851364.0,
"step": 9695
},
{
"entropy": 5.387207984924316,
"epoch": 0.8102238556632142,
"grad_norm": 0.97265625,
"learning_rate": 0.0004940649477241401,
"loss": 5.1303,
"mean_token_accuracy": 0.19218126833438873,
"num_tokens": 17860871.0,
"step": 9700
},
{
"entropy": 5.290140295028687,
"epoch": 0.8106414968259271,
"grad_norm": 1.0,
"learning_rate": 0.0004940581532775237,
"loss": 5.0574,
"mean_token_accuracy": 0.20372044295072556,
"num_tokens": 17869687.0,
"step": 9705
},
{
"entropy": 5.257996511459351,
"epoch": 0.8110591379886402,
"grad_norm": 1.0546875,
"learning_rate": 0.0004940513549960306,
"loss": 4.9968,
"mean_token_accuracy": 0.2021973043680191,
"num_tokens": 17878815.0,
"step": 9710
},
{
"entropy": 5.264734983444214,
"epoch": 0.8114767791513532,
"grad_norm": 0.953125,
"learning_rate": 0.0004940445528797798,
"loss": 4.9303,
"mean_token_accuracy": 0.20963813066482545,
"num_tokens": 17888302.0,
"step": 9715
},
{
"entropy": 5.187796926498413,
"epoch": 0.8118944203140661,
"grad_norm": 1.0234375,
"learning_rate": 0.0004940377469288906,
"loss": 4.8931,
"mean_token_accuracy": 0.21290806233882903,
"num_tokens": 17896807.0,
"step": 9720
},
{
"entropy": 5.287715768814087,
"epoch": 0.8123120614767791,
"grad_norm": 0.9296875,
"learning_rate": 0.0004940309371434819,
"loss": 4.9735,
"mean_token_accuracy": 0.20826876759529114,
"num_tokens": 17905988.0,
"step": 9725
},
{
"entropy": 5.351922559738159,
"epoch": 0.8127297026394922,
"grad_norm": 1.0078125,
"learning_rate": 0.0004940241235236731,
"loss": 5.1055,
"mean_token_accuracy": 0.1958531156182289,
"num_tokens": 17915917.0,
"step": 9730
},
{
"entropy": 5.323018980026245,
"epoch": 0.8131473438022051,
"grad_norm": 0.91796875,
"learning_rate": 0.0004940173060695833,
"loss": 5.0811,
"mean_token_accuracy": 0.20148218870162965,
"num_tokens": 17926158.0,
"step": 9735
},
{
"entropy": 5.257668209075928,
"epoch": 0.8135649849649181,
"grad_norm": 1.0078125,
"learning_rate": 0.0004940104847813319,
"loss": 5.0421,
"mean_token_accuracy": 0.20052656084299086,
"num_tokens": 17935488.0,
"step": 9740
},
{
"entropy": 5.275480699539185,
"epoch": 0.8139826261276312,
"grad_norm": 1.0,
"learning_rate": 0.0004940036596590384,
"loss": 4.9568,
"mean_token_accuracy": 0.20216303765773774,
"num_tokens": 17944850.0,
"step": 9745
},
{
"entropy": 5.238033199310303,
"epoch": 0.8144002672903441,
"grad_norm": 1.015625,
"learning_rate": 0.0004939968307028221,
"loss": 5.0058,
"mean_token_accuracy": 0.20468695610761642,
"num_tokens": 17954796.0,
"step": 9750
},
{
"entropy": 5.234244441986084,
"epoch": 0.8148179084530571,
"grad_norm": 0.94140625,
"learning_rate": 0.0004939899979128028,
"loss": 4.9853,
"mean_token_accuracy": 0.2116552397608757,
"num_tokens": 17963875.0,
"step": 9755
},
{
"entropy": 5.266030788421631,
"epoch": 0.8152355496157702,
"grad_norm": 1.03125,
"learning_rate": 0.0004939831612890999,
"loss": 5.0129,
"mean_token_accuracy": 0.2031642124056816,
"num_tokens": 17972789.0,
"step": 9760
},
{
"entropy": 5.276300764083862,
"epoch": 0.8156531907784831,
"grad_norm": 0.9375,
"learning_rate": 0.0004939763208318332,
"loss": 5.0293,
"mean_token_accuracy": 0.20778004229068756,
"num_tokens": 17982355.0,
"step": 9765
},
{
"entropy": 5.331578111648559,
"epoch": 0.8160708319411961,
"grad_norm": 1.1015625,
"learning_rate": 0.0004939694765411225,
"loss": 5.0335,
"mean_token_accuracy": 0.20019334703683853,
"num_tokens": 17992046.0,
"step": 9770
},
{
"entropy": 5.260315847396851,
"epoch": 0.8164884731039092,
"grad_norm": 1.0078125,
"learning_rate": 0.0004939626284170874,
"loss": 4.965,
"mean_token_accuracy": 0.21675824522972106,
"num_tokens": 18001440.0,
"step": 9775
},
{
"entropy": 5.349969720840454,
"epoch": 0.8169061142666221,
"grad_norm": 1.0703125,
"learning_rate": 0.000493955776459848,
"loss": 5.1015,
"mean_token_accuracy": 0.19849081188440323,
"num_tokens": 18010817.0,
"step": 9780
},
{
"entropy": 5.273048734664917,
"epoch": 0.8173237554293351,
"grad_norm": 0.92578125,
"learning_rate": 0.0004939489206695241,
"loss": 5.0359,
"mean_token_accuracy": 0.20319301784038543,
"num_tokens": 18019547.0,
"step": 9785
},
{
"entropy": 5.2291323184967045,
"epoch": 0.8177413965920481,
"grad_norm": 0.98046875,
"learning_rate": 0.0004939420610462359,
"loss": 5.0386,
"mean_token_accuracy": 0.20118792802095414,
"num_tokens": 18029191.0,
"step": 9790
},
{
"entropy": 5.320718145370483,
"epoch": 0.8181590377547611,
"grad_norm": 0.9609375,
"learning_rate": 0.0004939351975901034,
"loss": 5.0701,
"mean_token_accuracy": 0.20290120989084243,
"num_tokens": 18038831.0,
"step": 9795
},
{
"entropy": 5.286432790756225,
"epoch": 0.8185766789174741,
"grad_norm": 0.984375,
"learning_rate": 0.0004939283303012467,
"loss": 4.9779,
"mean_token_accuracy": 0.20698291957378387,
"num_tokens": 18049243.0,
"step": 9800
},
{
"entropy": 5.194411182403565,
"epoch": 0.8189943200801871,
"grad_norm": 0.98046875,
"learning_rate": 0.000493921459179786,
"loss": 4.9542,
"mean_token_accuracy": 0.20811349004507065,
"num_tokens": 18057960.0,
"step": 9805
},
{
"entropy": 5.277448987960815,
"epoch": 0.8194119612429,
"grad_norm": 0.94921875,
"learning_rate": 0.0004939145842258418,
"loss": 5.0815,
"mean_token_accuracy": 0.19748504161834718,
"num_tokens": 18066677.0,
"step": 9810
},
{
"entropy": 5.307441759109497,
"epoch": 0.8198296024056131,
"grad_norm": 0.953125,
"learning_rate": 0.0004939077054395341,
"loss": 5.0349,
"mean_token_accuracy": 0.20028697997331618,
"num_tokens": 18075830.0,
"step": 9815
},
{
"entropy": 5.26435604095459,
"epoch": 0.8202472435683261,
"grad_norm": 1.0078125,
"learning_rate": 0.0004939008228209836,
"loss": 4.9383,
"mean_token_accuracy": 0.20332272052764894,
"num_tokens": 18085335.0,
"step": 9820
},
{
"entropy": 5.267291641235351,
"epoch": 0.820664884731039,
"grad_norm": 0.921875,
"learning_rate": 0.0004938939363703106,
"loss": 5.0258,
"mean_token_accuracy": 0.19696779549121857,
"num_tokens": 18094531.0,
"step": 9825
},
{
"entropy": 5.260501098632813,
"epoch": 0.8210825258937521,
"grad_norm": 0.98046875,
"learning_rate": 0.0004938870460876359,
"loss": 4.973,
"mean_token_accuracy": 0.2061942219734192,
"num_tokens": 18103593.0,
"step": 9830
},
{
"entropy": 5.2618958950042725,
"epoch": 0.8215001670564651,
"grad_norm": 0.9609375,
"learning_rate": 0.00049388015197308,
"loss": 5.0385,
"mean_token_accuracy": 0.20461765974760054,
"num_tokens": 18112826.0,
"step": 9835
},
{
"entropy": 5.297413921356201,
"epoch": 0.821917808219178,
"grad_norm": 1.03125,
"learning_rate": 0.0004938732540267635,
"loss": 5.0901,
"mean_token_accuracy": 0.1960194230079651,
"num_tokens": 18121687.0,
"step": 9840
},
{
"entropy": 5.2823728084564205,
"epoch": 0.8223354493818911,
"grad_norm": 1.03125,
"learning_rate": 0.0004938663522488072,
"loss": 4.9599,
"mean_token_accuracy": 0.2121197000145912,
"num_tokens": 18130071.0,
"step": 9845
},
{
"entropy": 5.328467130661011,
"epoch": 0.8227530905446041,
"grad_norm": 1.0078125,
"learning_rate": 0.0004938594466393321,
"loss": 5.1758,
"mean_token_accuracy": 0.19506560266017914,
"num_tokens": 18139077.0,
"step": 9850
},
{
"entropy": 5.326861095428467,
"epoch": 0.823170731707317,
"grad_norm": 1.03125,
"learning_rate": 0.0004938525371984588,
"loss": 5.0565,
"mean_token_accuracy": 0.20626935511827468,
"num_tokens": 18147173.0,
"step": 9855
},
{
"entropy": 5.275492763519287,
"epoch": 0.8235883728700301,
"grad_norm": 0.9296875,
"learning_rate": 0.0004938456239263085,
"loss": 5.0254,
"mean_token_accuracy": 0.20502824932336808,
"num_tokens": 18157061.0,
"step": 9860
},
{
"entropy": 5.2826619148254395,
"epoch": 0.8240060140327431,
"grad_norm": 0.95703125,
"learning_rate": 0.0004938387068230021,
"loss": 4.9609,
"mean_token_accuracy": 0.2075473576784134,
"num_tokens": 18166115.0,
"step": 9865
},
{
"entropy": 5.22989239692688,
"epoch": 0.824423655195456,
"grad_norm": 0.98828125,
"learning_rate": 0.0004938317858886606,
"loss": 4.9787,
"mean_token_accuracy": 0.20749273002147675,
"num_tokens": 18174978.0,
"step": 9870
},
{
"entropy": 5.2950738906860355,
"epoch": 0.824841296358169,
"grad_norm": 0.98046875,
"learning_rate": 0.0004938248611234055,
"loss": 5.053,
"mean_token_accuracy": 0.2042671799659729,
"num_tokens": 18184948.0,
"step": 9875
},
{
"entropy": 5.315822172164917,
"epoch": 0.8252589375208821,
"grad_norm": 0.953125,
"learning_rate": 0.0004938179325273577,
"loss": 4.9928,
"mean_token_accuracy": 0.20614935010671614,
"num_tokens": 18193443.0,
"step": 9880
},
{
"entropy": 5.318535184860229,
"epoch": 0.825676578683595,
"grad_norm": 0.953125,
"learning_rate": 0.0004938110001006385,
"loss": 5.0064,
"mean_token_accuracy": 0.2060106500983238,
"num_tokens": 18202512.0,
"step": 9885
},
{
"entropy": 5.303167152404785,
"epoch": 0.826094219846308,
"grad_norm": 0.94140625,
"learning_rate": 0.0004938040638433694,
"loss": 5.0177,
"mean_token_accuracy": 0.20338963270187377,
"num_tokens": 18212328.0,
"step": 9890
},
{
"entropy": 5.2908491611480715,
"epoch": 0.8265118610090211,
"grad_norm": 1.0390625,
"learning_rate": 0.0004937971237556719,
"loss": 4.9715,
"mean_token_accuracy": 0.20638042390346528,
"num_tokens": 18221155.0,
"step": 9895
},
{
"entropy": 5.284775590896606,
"epoch": 0.826929502171734,
"grad_norm": 0.91015625,
"learning_rate": 0.0004937901798376672,
"loss": 5.0267,
"mean_token_accuracy": 0.2043670192360878,
"num_tokens": 18230604.0,
"step": 9900
},
{
"entropy": 5.276266813278198,
"epoch": 0.827347143334447,
"grad_norm": 0.99609375,
"learning_rate": 0.0004937832320894772,
"loss": 4.9611,
"mean_token_accuracy": 0.20734262466430664,
"num_tokens": 18239126.0,
"step": 9905
},
{
"entropy": 5.265355157852173,
"epoch": 0.8277647844971601,
"grad_norm": 1.0078125,
"learning_rate": 0.0004937762805112233,
"loss": 5.0022,
"mean_token_accuracy": 0.19618372470140458,
"num_tokens": 18248634.0,
"step": 9910
},
{
"entropy": 5.276412057876587,
"epoch": 0.828182425659873,
"grad_norm": 0.85546875,
"learning_rate": 0.0004937693251030273,
"loss": 5.135,
"mean_token_accuracy": 0.19961361587047577,
"num_tokens": 18258500.0,
"step": 9915
},
{
"entropy": 5.319969749450683,
"epoch": 0.828600066822586,
"grad_norm": 0.94921875,
"learning_rate": 0.000493762365865011,
"loss": 4.9467,
"mean_token_accuracy": 0.20834760516881942,
"num_tokens": 18268374.0,
"step": 9920
},
{
"entropy": 5.279907274246216,
"epoch": 0.8290177079852991,
"grad_norm": 0.96484375,
"learning_rate": 0.0004937554027972961,
"loss": 5.0808,
"mean_token_accuracy": 0.20224204808473586,
"num_tokens": 18277667.0,
"step": 9925
},
{
"entropy": 5.313109397888184,
"epoch": 0.829435349148012,
"grad_norm": 1.015625,
"learning_rate": 0.0004937484359000046,
"loss": 5.0856,
"mean_token_accuracy": 0.19987320601940156,
"num_tokens": 18287516.0,
"step": 9930
},
{
"entropy": 5.250987148284912,
"epoch": 0.829852990310725,
"grad_norm": 0.94921875,
"learning_rate": 0.0004937414651732584,
"loss": 4.931,
"mean_token_accuracy": 0.20905919522047042,
"num_tokens": 18296534.0,
"step": 9935
},
{
"entropy": 5.276131677627563,
"epoch": 0.830270631473438,
"grad_norm": 0.98828125,
"learning_rate": 0.0004937344906171795,
"loss": 5.077,
"mean_token_accuracy": 0.2060442104935646,
"num_tokens": 18305743.0,
"step": 9940
},
{
"entropy": 5.353442621231079,
"epoch": 0.830688272636151,
"grad_norm": 1.0078125,
"learning_rate": 0.0004937275122318902,
"loss": 5.0886,
"mean_token_accuracy": 0.20024637132883072,
"num_tokens": 18315524.0,
"step": 9945
},
{
"entropy": 5.2776796340942385,
"epoch": 0.831105913798864,
"grad_norm": 0.9921875,
"learning_rate": 0.0004937205300175123,
"loss": 4.9499,
"mean_token_accuracy": 0.21121760010719298,
"num_tokens": 18324280.0,
"step": 9950
},
{
"entropy": 5.338615846633911,
"epoch": 0.831523554961577,
"grad_norm": 1.0078125,
"learning_rate": 0.0004937135439741682,
"loss": 5.041,
"mean_token_accuracy": 0.20347232818603517,
"num_tokens": 18332661.0,
"step": 9955
},
{
"entropy": 5.3068865776062015,
"epoch": 0.83194119612429,
"grad_norm": 0.91015625,
"learning_rate": 0.0004937065541019806,
"loss": 5.1453,
"mean_token_accuracy": 0.1889631763100624,
"num_tokens": 18342986.0,
"step": 9960
},
{
"entropy": 5.34487738609314,
"epoch": 0.832358837287003,
"grad_norm": 1.015625,
"learning_rate": 0.0004936995604010711,
"loss": 5.0867,
"mean_token_accuracy": 0.2043716013431549,
"num_tokens": 18351873.0,
"step": 9965
},
{
"entropy": 5.3892107009887695,
"epoch": 0.832776478449716,
"grad_norm": 0.91015625,
"learning_rate": 0.0004936925628715628,
"loss": 5.1559,
"mean_token_accuracy": 0.19914899617433549,
"num_tokens": 18362174.0,
"step": 9970
},
{
"entropy": 5.309047508239746,
"epoch": 0.833194119612429,
"grad_norm": 0.9375,
"learning_rate": 0.0004936855615135779,
"loss": 4.961,
"mean_token_accuracy": 0.21297989785671234,
"num_tokens": 18371272.0,
"step": 9975
},
{
"entropy": 5.325804662704468,
"epoch": 0.833611760775142,
"grad_norm": 1.046875,
"learning_rate": 0.000493678556327239,
"loss": 5.0741,
"mean_token_accuracy": 0.20604854822158813,
"num_tokens": 18380526.0,
"step": 9980
},
{
"entropy": 5.252417516708374,
"epoch": 0.834029401937855,
"grad_norm": 0.92578125,
"learning_rate": 0.0004936715473126688,
"loss": 4.9898,
"mean_token_accuracy": 0.2055755466222763,
"num_tokens": 18390027.0,
"step": 9985
},
{
"entropy": 5.209375238418579,
"epoch": 0.8344470431005679,
"grad_norm": 1.0625,
"learning_rate": 0.00049366453446999,
"loss": 4.9682,
"mean_token_accuracy": 0.20910389274358748,
"num_tokens": 18399012.0,
"step": 9990
},
{
"entropy": 5.208969020843506,
"epoch": 0.834864684263281,
"grad_norm": 0.95703125,
"learning_rate": 0.0004936575177993251,
"loss": 4.9206,
"mean_token_accuracy": 0.20752075165510178,
"num_tokens": 18408891.0,
"step": 9995
},
{
"entropy": 5.263211059570312,
"epoch": 0.835282325425994,
"grad_norm": 1.09375,
"learning_rate": 0.0004936504973007973,
"loss": 4.9364,
"mean_token_accuracy": 0.20904496014118196,
"num_tokens": 18417909.0,
"step": 10000
},
{
"entropy": 5.2176923751831055,
"epoch": 0.8356999665887069,
"grad_norm": 1.0,
"learning_rate": 0.0004936434729745294,
"loss": 5.0183,
"mean_token_accuracy": 0.2021762400865555,
"num_tokens": 18427250.0,
"step": 10005
},
{
"entropy": 5.279464435577393,
"epoch": 0.83611760775142,
"grad_norm": 1.1328125,
"learning_rate": 0.0004936364448206442,
"loss": 4.987,
"mean_token_accuracy": 0.204580220580101,
"num_tokens": 18435851.0,
"step": 10010
},
{
"entropy": 5.234575986862183,
"epoch": 0.836535248914133,
"grad_norm": 1.03125,
"learning_rate": 0.0004936294128392649,
"loss": 4.9918,
"mean_token_accuracy": 0.1984171226620674,
"num_tokens": 18444535.0,
"step": 10015
},
{
"entropy": 5.293031787872314,
"epoch": 0.8369528900768459,
"grad_norm": 0.87109375,
"learning_rate": 0.0004936223770305146,
"loss": 5.1164,
"mean_token_accuracy": 0.20174213945865632,
"num_tokens": 18454453.0,
"step": 10020
},
{
"entropy": 5.389978837966919,
"epoch": 0.837370531239559,
"grad_norm": 0.91015625,
"learning_rate": 0.0004936153373945165,
"loss": 5.1418,
"mean_token_accuracy": 0.19709545522928237,
"num_tokens": 18465033.0,
"step": 10025
},
{
"entropy": 5.3353354930877686,
"epoch": 0.837788172402272,
"grad_norm": 0.91015625,
"learning_rate": 0.0004936082939313936,
"loss": 5.0292,
"mean_token_accuracy": 0.20263568609952926,
"num_tokens": 18473786.0,
"step": 10030
},
{
"entropy": 5.206992769241333,
"epoch": 0.8382058135649849,
"grad_norm": 0.95703125,
"learning_rate": 0.0004936012466412695,
"loss": 4.987,
"mean_token_accuracy": 0.20087777972221374,
"num_tokens": 18483282.0,
"step": 10035
},
{
"entropy": 5.381425237655639,
"epoch": 0.838623454727698,
"grad_norm": 0.94921875,
"learning_rate": 0.0004935941955242674,
"loss": 5.1521,
"mean_token_accuracy": 0.19864945411682128,
"num_tokens": 18492612.0,
"step": 10040
},
{
"entropy": 5.295857954025268,
"epoch": 0.839041095890411,
"grad_norm": 1.0546875,
"learning_rate": 0.0004935871405805109,
"loss": 4.9684,
"mean_token_accuracy": 0.21615595519542694,
"num_tokens": 18501120.0,
"step": 10045
},
{
"entropy": 5.2610431671142575,
"epoch": 0.8394587370531239,
"grad_norm": 0.9609375,
"learning_rate": 0.0004935800818101232,
"loss": 5.0656,
"mean_token_accuracy": 0.1995917871594429,
"num_tokens": 18510613.0,
"step": 10050
},
{
"entropy": 5.260000991821289,
"epoch": 0.8398763782158369,
"grad_norm": 1.078125,
"learning_rate": 0.0004935730192132282,
"loss": 4.9681,
"mean_token_accuracy": 0.20497325658798218,
"num_tokens": 18518854.0,
"step": 10055
},
{
"entropy": 5.325906133651733,
"epoch": 0.84029401937855,
"grad_norm": 0.98046875,
"learning_rate": 0.0004935659527899493,
"loss": 5.0657,
"mean_token_accuracy": 0.2045937806367874,
"num_tokens": 18527592.0,
"step": 10060
},
{
"entropy": 5.237090158462524,
"epoch": 0.8407116605412629,
"grad_norm": 0.953125,
"learning_rate": 0.0004935588825404104,
"loss": 4.9167,
"mean_token_accuracy": 0.21674193441867828,
"num_tokens": 18536396.0,
"step": 10065
},
{
"entropy": 5.212545442581177,
"epoch": 0.8411293017039759,
"grad_norm": 1.046875,
"learning_rate": 0.0004935518084647351,
"loss": 4.8951,
"mean_token_accuracy": 0.21914102435111998,
"num_tokens": 18545201.0,
"step": 10070
},
{
"entropy": 5.270266771316528,
"epoch": 0.841546942866689,
"grad_norm": 0.9765625,
"learning_rate": 0.0004935447305630474,
"loss": 5.0878,
"mean_token_accuracy": 0.20132859349250792,
"num_tokens": 18554944.0,
"step": 10075
},
{
"entropy": 5.270673847198486,
"epoch": 0.8419645840294019,
"grad_norm": 0.99609375,
"learning_rate": 0.0004935376488354711,
"loss": 4.9798,
"mean_token_accuracy": 0.20357297956943513,
"num_tokens": 18564642.0,
"step": 10080
},
{
"entropy": 5.296598768234253,
"epoch": 0.8423822251921149,
"grad_norm": 0.9609375,
"learning_rate": 0.0004935305632821303,
"loss": 4.9898,
"mean_token_accuracy": 0.2068466365337372,
"num_tokens": 18573897.0,
"step": 10085
},
{
"entropy": 5.283039236068726,
"epoch": 0.842799866354828,
"grad_norm": 0.97265625,
"learning_rate": 0.0004935234739031488,
"loss": 5.0299,
"mean_token_accuracy": 0.20671347677707672,
"num_tokens": 18583174.0,
"step": 10090
},
{
"entropy": 5.275842952728271,
"epoch": 0.8432175075175409,
"grad_norm": 0.99609375,
"learning_rate": 0.0004935163806986508,
"loss": 4.9425,
"mean_token_accuracy": 0.20930425077676773,
"num_tokens": 18592540.0,
"step": 10095
},
{
"entropy": 5.234079647064209,
"epoch": 0.8436351486802539,
"grad_norm": 0.8984375,
"learning_rate": 0.0004935092836687607,
"loss": 4.9582,
"mean_token_accuracy": 0.20485835820436477,
"num_tokens": 18602743.0,
"step": 10100
},
{
"entropy": 5.2034423828125,
"epoch": 0.8440527898429669,
"grad_norm": 1.0,
"learning_rate": 0.0004935021828136025,
"loss": 4.9157,
"mean_token_accuracy": 0.2104882314801216,
"num_tokens": 18611544.0,
"step": 10105
},
{
"entropy": 5.343887567520142,
"epoch": 0.8444704310056799,
"grad_norm": 1.0703125,
"learning_rate": 0.0004934950781333005,
"loss": 5.0314,
"mean_token_accuracy": 0.20303449183702468,
"num_tokens": 18620841.0,
"step": 10110
},
{
"entropy": 5.299557781219482,
"epoch": 0.8448880721683929,
"grad_norm": 0.953125,
"learning_rate": 0.0004934879696279793,
"loss": 5.0701,
"mean_token_accuracy": 0.20216573178768157,
"num_tokens": 18630890.0,
"step": 10115
},
{
"entropy": 5.332501173019409,
"epoch": 0.8453057133311059,
"grad_norm": 1.03125,
"learning_rate": 0.0004934808572977632,
"loss": 5.0693,
"mean_token_accuracy": 0.1973210945725441,
"num_tokens": 18639768.0,
"step": 10120
},
{
"entropy": 5.210367345809937,
"epoch": 0.8457233544938189,
"grad_norm": 0.99609375,
"learning_rate": 0.0004934737411427765,
"loss": 4.8895,
"mean_token_accuracy": 0.21083588302135467,
"num_tokens": 18649578.0,
"step": 10125
},
{
"entropy": 5.228143835067749,
"epoch": 0.8461409956565319,
"grad_norm": 0.99609375,
"learning_rate": 0.0004934666211631442,
"loss": 4.9924,
"mean_token_accuracy": 0.2090206637978554,
"num_tokens": 18658470.0,
"step": 10130
},
{
"entropy": 5.260270881652832,
"epoch": 0.8465586368192449,
"grad_norm": 0.90234375,
"learning_rate": 0.0004934594973589906,
"loss": 4.9627,
"mean_token_accuracy": 0.20124966502189637,
"num_tokens": 18668037.0,
"step": 10135
},
{
"entropy": 5.26919322013855,
"epoch": 0.8469762779819578,
"grad_norm": 0.99609375,
"learning_rate": 0.0004934523697304406,
"loss": 5.0782,
"mean_token_accuracy": 0.19919137209653853,
"num_tokens": 18677841.0,
"step": 10140
},
{
"entropy": 5.3400341987609865,
"epoch": 0.8473939191446709,
"grad_norm": 0.984375,
"learning_rate": 0.0004934452382776189,
"loss": 5.0133,
"mean_token_accuracy": 0.199882934987545,
"num_tokens": 18686244.0,
"step": 10145
},
{
"entropy": 5.283691787719727,
"epoch": 0.8478115603073839,
"grad_norm": 0.96484375,
"learning_rate": 0.0004934381030006504,
"loss": 5.0547,
"mean_token_accuracy": 0.20212140381336213,
"num_tokens": 18695129.0,
"step": 10150
},
{
"entropy": 5.172067451477051,
"epoch": 0.8482292014700968,
"grad_norm": 1.15625,
"learning_rate": 0.00049343096389966,
"loss": 4.8854,
"mean_token_accuracy": 0.2176145538687706,
"num_tokens": 18703566.0,
"step": 10155
},
{
"entropy": 5.321353244781494,
"epoch": 0.8486468426328099,
"grad_norm": 0.89453125,
"learning_rate": 0.0004934238209747724,
"loss": 5.08,
"mean_token_accuracy": 0.19820543229579926,
"num_tokens": 18713616.0,
"step": 10160
},
{
"entropy": 5.3626219749450685,
"epoch": 0.8490644837955229,
"grad_norm": 1.03125,
"learning_rate": 0.0004934166742261132,
"loss": 5.01,
"mean_token_accuracy": 0.20752353966236115,
"num_tokens": 18722851.0,
"step": 10165
},
{
"entropy": 5.302944755554199,
"epoch": 0.8494821249582358,
"grad_norm": 0.984375,
"learning_rate": 0.000493409523653807,
"loss": 5.0111,
"mean_token_accuracy": 0.20677843689918518,
"num_tokens": 18731961.0,
"step": 10170
},
{
"entropy": 5.2983826160430905,
"epoch": 0.8498997661209489,
"grad_norm": 1.0390625,
"learning_rate": 0.0004934023692579793,
"loss": 5.1786,
"mean_token_accuracy": 0.20431944280862807,
"num_tokens": 18741267.0,
"step": 10175
},
{
"entropy": 5.335294961929321,
"epoch": 0.8503174072836619,
"grad_norm": 0.953125,
"learning_rate": 0.0004933952110387553,
"loss": 5.0227,
"mean_token_accuracy": 0.20469975769519805,
"num_tokens": 18750266.0,
"step": 10180
},
{
"entropy": 5.268962717056274,
"epoch": 0.8507350484463748,
"grad_norm": 0.98046875,
"learning_rate": 0.0004933880489962602,
"loss": 5.0046,
"mean_token_accuracy": 0.20418863892555236,
"num_tokens": 18759910.0,
"step": 10185
},
{
"entropy": 5.3556318283081055,
"epoch": 0.8511526896090879,
"grad_norm": 0.95703125,
"learning_rate": 0.0004933808831306194,
"loss": 5.1485,
"mean_token_accuracy": 0.194237357378006,
"num_tokens": 18769510.0,
"step": 10190
},
{
"entropy": 5.236069297790527,
"epoch": 0.8515703307718009,
"grad_norm": 0.921875,
"learning_rate": 0.0004933737134419583,
"loss": 4.9556,
"mean_token_accuracy": 0.20543353408575057,
"num_tokens": 18779893.0,
"step": 10195
},
{
"entropy": 5.308035469055175,
"epoch": 0.8519879719345138,
"grad_norm": 1.046875,
"learning_rate": 0.0004933665399304025,
"loss": 5.0148,
"mean_token_accuracy": 0.2130729541182518,
"num_tokens": 18788802.0,
"step": 10200
},
{
"entropy": 5.234163379669189,
"epoch": 0.8524056130972268,
"grad_norm": 1.03125,
"learning_rate": 0.0004933593625960777,
"loss": 4.9118,
"mean_token_accuracy": 0.2109411671757698,
"num_tokens": 18797154.0,
"step": 10205
},
{
"entropy": 5.266857051849366,
"epoch": 0.8528232542599399,
"grad_norm": 0.96875,
"learning_rate": 0.0004933521814391094,
"loss": 5.0417,
"mean_token_accuracy": 0.2085142806172371,
"num_tokens": 18806151.0,
"step": 10210
},
{
"entropy": 5.33585844039917,
"epoch": 0.8532408954226529,
"grad_norm": 0.98046875,
"learning_rate": 0.0004933449964596232,
"loss": 4.9471,
"mean_token_accuracy": 0.21520161032676696,
"num_tokens": 18815162.0,
"step": 10215
},
{
"entropy": 5.205343437194824,
"epoch": 0.8536585365853658,
"grad_norm": 1.0390625,
"learning_rate": 0.0004933378076577451,
"loss": 4.9212,
"mean_token_accuracy": 0.2057916358113289,
"num_tokens": 18823676.0,
"step": 10220
},
{
"entropy": 5.351762294769287,
"epoch": 0.8540761777480789,
"grad_norm": 0.99609375,
"learning_rate": 0.0004933306150336009,
"loss": 5.2249,
"mean_token_accuracy": 0.19383471310138703,
"num_tokens": 18832991.0,
"step": 10225
},
{
"entropy": 5.3964050769805905,
"epoch": 0.8544938189107919,
"grad_norm": 0.9453125,
"learning_rate": 0.0004933234185873164,
"loss": 5.0966,
"mean_token_accuracy": 0.19839687943458556,
"num_tokens": 18842614.0,
"step": 10230
},
{
"entropy": 5.373830318450928,
"epoch": 0.8549114600735048,
"grad_norm": 0.94140625,
"learning_rate": 0.0004933162183190177,
"loss": 5.1404,
"mean_token_accuracy": 0.19562443643808364,
"num_tokens": 18852747.0,
"step": 10235
},
{
"entropy": 5.251061058044433,
"epoch": 0.8553291012362179,
"grad_norm": 0.95703125,
"learning_rate": 0.0004933090142288308,
"loss": 5.0091,
"mean_token_accuracy": 0.20302987694740296,
"num_tokens": 18862005.0,
"step": 10240
},
{
"entropy": 5.295311164855957,
"epoch": 0.8557467423989309,
"grad_norm": 1.03125,
"learning_rate": 0.0004933018063168819,
"loss": 4.9707,
"mean_token_accuracy": 0.2199796348810196,
"num_tokens": 18870172.0,
"step": 10245
},
{
"entropy": 5.31706051826477,
"epoch": 0.8561643835616438,
"grad_norm": 0.9375,
"learning_rate": 0.000493294594583297,
"loss": 5.0506,
"mean_token_accuracy": 0.20717416554689408,
"num_tokens": 18879428.0,
"step": 10250
},
{
"entropy": 5.303390455245972,
"epoch": 0.8565820247243568,
"grad_norm": 0.95703125,
"learning_rate": 0.0004932873790282026,
"loss": 4.9896,
"mean_token_accuracy": 0.2154971867799759,
"num_tokens": 18888694.0,
"step": 10255
},
{
"entropy": 5.333094358444214,
"epoch": 0.8569996658870699,
"grad_norm": 1.03125,
"learning_rate": 0.0004932801596517248,
"loss": 5.0496,
"mean_token_accuracy": 0.1969591572880745,
"num_tokens": 18898317.0,
"step": 10260
},
{
"entropy": 5.237871885299683,
"epoch": 0.8574173070497828,
"grad_norm": 1.015625,
"learning_rate": 0.0004932729364539901,
"loss": 5.0297,
"mean_token_accuracy": 0.2007185474038124,
"num_tokens": 18906749.0,
"step": 10265
},
{
"entropy": 5.24932107925415,
"epoch": 0.8578349482124958,
"grad_norm": 1.0859375,
"learning_rate": 0.0004932657094351248,
"loss": 4.9268,
"mean_token_accuracy": 0.2038540616631508,
"num_tokens": 18915404.0,
"step": 10270
},
{
"entropy": 5.311759185791016,
"epoch": 0.8582525893752089,
"grad_norm": 0.9921875,
"learning_rate": 0.0004932584785952555,
"loss": 5.0262,
"mean_token_accuracy": 0.2053885668516159,
"num_tokens": 18925512.0,
"step": 10275
},
{
"entropy": 5.239424848556519,
"epoch": 0.8586702305379218,
"grad_norm": 1.125,
"learning_rate": 0.000493251243934509,
"loss": 4.9454,
"mean_token_accuracy": 0.2159731775522232,
"num_tokens": 18934278.0,
"step": 10280
},
{
"entropy": 5.263280725479126,
"epoch": 0.8590878717006348,
"grad_norm": 0.98828125,
"learning_rate": 0.0004932440054530116,
"loss": 5.0363,
"mean_token_accuracy": 0.20390205830335617,
"num_tokens": 18943126.0,
"step": 10285
},
{
"entropy": 5.300428056716919,
"epoch": 0.8595055128633479,
"grad_norm": 1.125,
"learning_rate": 0.0004932367631508903,
"loss": 5.0113,
"mean_token_accuracy": 0.2080025628209114,
"num_tokens": 18951716.0,
"step": 10290
},
{
"entropy": 5.291877508163452,
"epoch": 0.8599231540260608,
"grad_norm": 1.015625,
"learning_rate": 0.0004932295170282717,
"loss": 5.1041,
"mean_token_accuracy": 0.1983656495809555,
"num_tokens": 18960692.0,
"step": 10295
},
{
"entropy": 5.2682994365692135,
"epoch": 0.8603407951887738,
"grad_norm": 0.953125,
"learning_rate": 0.0004932222670852828,
"loss": 5.0503,
"mean_token_accuracy": 0.20897567570209502,
"num_tokens": 18969813.0,
"step": 10300
},
{
"entropy": 5.276914930343628,
"epoch": 0.8607584363514869,
"grad_norm": 1.03125,
"learning_rate": 0.0004932150133220504,
"loss": 4.8932,
"mean_token_accuracy": 0.21474039256572724,
"num_tokens": 18978536.0,
"step": 10305
},
{
"entropy": 5.261690282821656,
"epoch": 0.8611760775141998,
"grad_norm": 1.0078125,
"learning_rate": 0.0004932077557387015,
"loss": 4.9917,
"mean_token_accuracy": 0.2117157205939293,
"num_tokens": 18987637.0,
"step": 10310
},
{
"entropy": 5.164496040344238,
"epoch": 0.8615937186769128,
"grad_norm": 0.95703125,
"learning_rate": 0.0004932004943353633,
"loss": 4.8371,
"mean_token_accuracy": 0.2142604261636734,
"num_tokens": 18997010.0,
"step": 10315
},
{
"entropy": 5.237453508377075,
"epoch": 0.8620113598396258,
"grad_norm": 1.0,
"learning_rate": 0.0004931932291121627,
"loss": 4.982,
"mean_token_accuracy": 0.20955340415239335,
"num_tokens": 19005685.0,
"step": 10320
},
{
"entropy": 5.253964138031006,
"epoch": 0.8624290010023388,
"grad_norm": 0.98046875,
"learning_rate": 0.0004931859600692269,
"loss": 5.0142,
"mean_token_accuracy": 0.20397783517837526,
"num_tokens": 19014836.0,
"step": 10325
},
{
"entropy": 5.291058444976807,
"epoch": 0.8628466421650518,
"grad_norm": 0.99609375,
"learning_rate": 0.0004931786872066833,
"loss": 5.0441,
"mean_token_accuracy": 0.20211344510316848,
"num_tokens": 19024442.0,
"step": 10330
},
{
"entropy": 5.311799240112305,
"epoch": 0.8632642833277648,
"grad_norm": 0.98046875,
"learning_rate": 0.0004931714105246592,
"loss": 4.9836,
"mean_token_accuracy": 0.21372086852788924,
"num_tokens": 19033730.0,
"step": 10335
},
{
"entropy": 5.263733768463135,
"epoch": 0.8636819244904778,
"grad_norm": 1.0078125,
"learning_rate": 0.0004931641300232819,
"loss": 4.9367,
"mean_token_accuracy": 0.21760034561157227,
"num_tokens": 19042765.0,
"step": 10340
},
{
"entropy": 5.317936515808105,
"epoch": 0.8640995656531908,
"grad_norm": 0.953125,
"learning_rate": 0.0004931568457026789,
"loss": 4.9392,
"mean_token_accuracy": 0.20653872936964035,
"num_tokens": 19051408.0,
"step": 10345
},
{
"entropy": 5.297897148132324,
"epoch": 0.8645172068159038,
"grad_norm": 0.92578125,
"learning_rate": 0.0004931495575629778,
"loss": 5.0502,
"mean_token_accuracy": 0.20026125013828278,
"num_tokens": 19061145.0,
"step": 10350
},
{
"entropy": 5.254010915756226,
"epoch": 0.8649348479786167,
"grad_norm": 0.9453125,
"learning_rate": 0.000493142265604306,
"loss": 5.042,
"mean_token_accuracy": 0.20199861228466034,
"num_tokens": 19070157.0,
"step": 10355
},
{
"entropy": 5.228733110427856,
"epoch": 0.8653524891413298,
"grad_norm": 1.09375,
"learning_rate": 0.0004931349698267914,
"loss": 4.8609,
"mean_token_accuracy": 0.22068461924791336,
"num_tokens": 19078532.0,
"step": 10360
},
{
"entropy": 5.2406251430511475,
"epoch": 0.8657701303040428,
"grad_norm": 1.0,
"learning_rate": 0.0004931276702305615,
"loss": 4.9067,
"mean_token_accuracy": 0.21284476667642593,
"num_tokens": 19087283.0,
"step": 10365
},
{
"entropy": 5.306699419021607,
"epoch": 0.8661877714667557,
"grad_norm": 1.0546875,
"learning_rate": 0.0004931203668157441,
"loss": 5.0964,
"mean_token_accuracy": 0.2053291141986847,
"num_tokens": 19097568.0,
"step": 10370
},
{
"entropy": 5.362837266921997,
"epoch": 0.8666054126294688,
"grad_norm": 1.046875,
"learning_rate": 0.0004931130595824672,
"loss": 5.0662,
"mean_token_accuracy": 0.20025492906570436,
"num_tokens": 19107218.0,
"step": 10375
},
{
"entropy": 5.296062374114991,
"epoch": 0.8670230537921818,
"grad_norm": 0.91796875,
"learning_rate": 0.0004931057485308587,
"loss": 5.0347,
"mean_token_accuracy": 0.2037366136908531,
"num_tokens": 19116128.0,
"step": 10380
},
{
"entropy": 5.197923278808593,
"epoch": 0.8674406949548947,
"grad_norm": 1.0390625,
"learning_rate": 0.0004930984336610465,
"loss": 4.9866,
"mean_token_accuracy": 0.20974574834108353,
"num_tokens": 19124885.0,
"step": 10385
},
{
"entropy": 5.325475263595581,
"epoch": 0.8678583361176078,
"grad_norm": 0.984375,
"learning_rate": 0.0004930911149731587,
"loss": 5.0609,
"mean_token_accuracy": 0.20704571902751923,
"num_tokens": 19134125.0,
"step": 10390
},
{
"entropy": 5.257022190093994,
"epoch": 0.8682759772803208,
"grad_norm": 1.0546875,
"learning_rate": 0.0004930837924673233,
"loss": 4.9434,
"mean_token_accuracy": 0.20590798705816268,
"num_tokens": 19141925.0,
"step": 10395
},
{
"entropy": 5.285811710357666,
"epoch": 0.8686936184430337,
"grad_norm": 0.94140625,
"learning_rate": 0.0004930764661436688,
"loss": 5.0289,
"mean_token_accuracy": 0.20406344532966614,
"num_tokens": 19151440.0,
"step": 10400
},
{
"entropy": 5.315800333023072,
"epoch": 0.8691112596057468,
"grad_norm": 0.91015625,
"learning_rate": 0.0004930691360023231,
"loss": 5.1078,
"mean_token_accuracy": 0.20519358962774276,
"num_tokens": 19161447.0,
"step": 10405
},
{
"entropy": 5.303775262832642,
"epoch": 0.8695289007684598,
"grad_norm": 1.0390625,
"learning_rate": 0.0004930618020434147,
"loss": 5.0323,
"mean_token_accuracy": 0.2023095890879631,
"num_tokens": 19170804.0,
"step": 10410
},
{
"entropy": 5.393555736541748,
"epoch": 0.8699465419311727,
"grad_norm": 1.0625,
"learning_rate": 0.0004930544642670721,
"loss": 5.1202,
"mean_token_accuracy": 0.19721234887838363,
"num_tokens": 19179898.0,
"step": 10415
},
{
"entropy": 5.232870101928711,
"epoch": 0.8703641830938857,
"grad_norm": 1.078125,
"learning_rate": 0.0004930471226734235,
"loss": 4.9481,
"mean_token_accuracy": 0.21083010882139205,
"num_tokens": 19188919.0,
"step": 10420
},
{
"entropy": 5.277183198928833,
"epoch": 0.8707818242565988,
"grad_norm": 1.046875,
"learning_rate": 0.0004930397772625976,
"loss": 5.0306,
"mean_token_accuracy": 0.2075921803712845,
"num_tokens": 19197745.0,
"step": 10425
},
{
"entropy": 5.277238464355468,
"epoch": 0.8711994654193117,
"grad_norm": 1.1171875,
"learning_rate": 0.000493032428034723,
"loss": 5.0321,
"mean_token_accuracy": 0.20863038897514344,
"num_tokens": 19206681.0,
"step": 10430
},
{
"entropy": 5.380311965942383,
"epoch": 0.8716171065820247,
"grad_norm": 1.0078125,
"learning_rate": 0.0004930250749899282,
"loss": 5.0319,
"mean_token_accuracy": 0.2054059088230133,
"num_tokens": 19215806.0,
"step": 10435
},
{
"entropy": 5.281986284255981,
"epoch": 0.8720347477447378,
"grad_norm": 1.0703125,
"learning_rate": 0.0004930177181283422,
"loss": 5.048,
"mean_token_accuracy": 0.20246667712926864,
"num_tokens": 19224681.0,
"step": 10440
},
{
"entropy": 5.239447927474975,
"epoch": 0.8724523889074507,
"grad_norm": 0.99609375,
"learning_rate": 0.0004930103574500934,
"loss": 5.045,
"mean_token_accuracy": 0.2029770776629448,
"num_tokens": 19232684.0,
"step": 10445
},
{
"entropy": 5.30054292678833,
"epoch": 0.8728700300701637,
"grad_norm": 0.94140625,
"learning_rate": 0.0004930029929553109,
"loss": 4.9527,
"mean_token_accuracy": 0.20128995627164842,
"num_tokens": 19242538.0,
"step": 10450
},
{
"entropy": 5.244245004653931,
"epoch": 0.8732876712328768,
"grad_norm": 1.0390625,
"learning_rate": 0.0004929956246441237,
"loss": 5.0008,
"mean_token_accuracy": 0.20038944780826567,
"num_tokens": 19251146.0,
"step": 10455
},
{
"entropy": 5.371267318725586,
"epoch": 0.8737053123955897,
"grad_norm": 1.0234375,
"learning_rate": 0.0004929882525166608,
"loss": 5.0144,
"mean_token_accuracy": 0.2000021755695343,
"num_tokens": 19259931.0,
"step": 10460
},
{
"entropy": 5.325706195831299,
"epoch": 0.8741229535583027,
"grad_norm": 0.953125,
"learning_rate": 0.0004929808765730509,
"loss": 5.0412,
"mean_token_accuracy": 0.2019277274608612,
"num_tokens": 19268796.0,
"step": 10465
},
{
"entropy": 5.270397043228149,
"epoch": 0.8745405947210158,
"grad_norm": 1.03125,
"learning_rate": 0.0004929734968134235,
"loss": 5.0728,
"mean_token_accuracy": 0.19787393212318422,
"num_tokens": 19278396.0,
"step": 10470
},
{
"entropy": 5.359102725982666,
"epoch": 0.8749582358837287,
"grad_norm": 0.93359375,
"learning_rate": 0.0004929661132379076,
"loss": 5.0749,
"mean_token_accuracy": 0.19762273579835893,
"num_tokens": 19287914.0,
"step": 10475
},
{
"entropy": 5.346466159820556,
"epoch": 0.8753758770464417,
"grad_norm": 0.9296875,
"learning_rate": 0.0004929587258466326,
"loss": 5.0483,
"mean_token_accuracy": 0.19957538694143295,
"num_tokens": 19297304.0,
"step": 10480
},
{
"entropy": 5.329871892929077,
"epoch": 0.8757935182091547,
"grad_norm": 1.1015625,
"learning_rate": 0.0004929513346397278,
"loss": 5.0037,
"mean_token_accuracy": 0.20976925790309905,
"num_tokens": 19306245.0,
"step": 10485
},
{
"entropy": 5.296418523788452,
"epoch": 0.8762111593718677,
"grad_norm": 1.03125,
"learning_rate": 0.0004929439396173223,
"loss": 5.0139,
"mean_token_accuracy": 0.20032329112291336,
"num_tokens": 19316160.0,
"step": 10490
},
{
"entropy": 5.29181399345398,
"epoch": 0.8766288005345807,
"grad_norm": 1.09375,
"learning_rate": 0.000492936540779546,
"loss": 5.0996,
"mean_token_accuracy": 0.1979194015264511,
"num_tokens": 19325411.0,
"step": 10495
},
{
"entropy": 5.359336233139038,
"epoch": 0.8770464416972937,
"grad_norm": 1.0234375,
"learning_rate": 0.0004929291381265281,
"loss": 4.9975,
"mean_token_accuracy": 0.20435776710510253,
"num_tokens": 19334297.0,
"step": 10500
},
{
"entropy": 5.37188024520874,
"epoch": 0.8774640828600067,
"grad_norm": 0.96875,
"learning_rate": 0.0004929217316583985,
"loss": 5.1017,
"mean_token_accuracy": 0.19282938838005065,
"num_tokens": 19344121.0,
"step": 10505
},
{
"entropy": 5.28780460357666,
"epoch": 0.8778817240227197,
"grad_norm": 1.015625,
"learning_rate": 0.0004929143213752865,
"loss": 5.0374,
"mean_token_accuracy": 0.20151905864477157,
"num_tokens": 19353428.0,
"step": 10510
},
{
"entropy": 5.309815120697022,
"epoch": 0.8782993651854327,
"grad_norm": 1.0625,
"learning_rate": 0.000492906907277322,
"loss": 5.0842,
"mean_token_accuracy": 0.2049848034977913,
"num_tokens": 19362661.0,
"step": 10515
},
{
"entropy": 5.215135478973389,
"epoch": 0.8787170063481456,
"grad_norm": 0.97265625,
"learning_rate": 0.000492899489364635,
"loss": 4.8873,
"mean_token_accuracy": 0.21897618621587753,
"num_tokens": 19371246.0,
"step": 10520
},
{
"entropy": 5.2584888458251955,
"epoch": 0.8791346475108587,
"grad_norm": 1.1640625,
"learning_rate": 0.0004928920676373549,
"loss": 4.9327,
"mean_token_accuracy": 0.20861630588769914,
"num_tokens": 19379777.0,
"step": 10525
},
{
"entropy": 5.246981287002564,
"epoch": 0.8795522886735717,
"grad_norm": 0.97265625,
"learning_rate": 0.000492884642095612,
"loss": 4.9912,
"mean_token_accuracy": 0.20688994377851486,
"num_tokens": 19388717.0,
"step": 10530
},
{
"entropy": 5.407645750045776,
"epoch": 0.8799699298362846,
"grad_norm": 1.0234375,
"learning_rate": 0.0004928772127395361,
"loss": 5.1231,
"mean_token_accuracy": 0.1965832844376564,
"num_tokens": 19397936.0,
"step": 10535
},
{
"entropy": 5.4296173572540285,
"epoch": 0.8803875709989977,
"grad_norm": 1.0390625,
"learning_rate": 0.0004928697795692575,
"loss": 5.0834,
"mean_token_accuracy": 0.19767144620418547,
"num_tokens": 19406830.0,
"step": 10540
},
{
"entropy": 5.328914737701416,
"epoch": 0.8808052121617107,
"grad_norm": 0.98828125,
"learning_rate": 0.000492862342584906,
"loss": 5.0852,
"mean_token_accuracy": 0.20326122790575027,
"num_tokens": 19415131.0,
"step": 10545
},
{
"entropy": 5.2831205368042,
"epoch": 0.8812228533244236,
"grad_norm": 1.0546875,
"learning_rate": 0.000492854901786612,
"loss": 4.8774,
"mean_token_accuracy": 0.21782393455505372,
"num_tokens": 19423585.0,
"step": 10550
},
{
"entropy": 5.22721757888794,
"epoch": 0.8816404944871367,
"grad_norm": 1.078125,
"learning_rate": 0.0004928474571745058,
"loss": 4.9054,
"mean_token_accuracy": 0.21980790048837662,
"num_tokens": 19431890.0,
"step": 10555
},
{
"entropy": 5.253408432006836,
"epoch": 0.8820581356498497,
"grad_norm": 1.03125,
"learning_rate": 0.0004928400087487175,
"loss": 5.059,
"mean_token_accuracy": 0.20104600340127946,
"num_tokens": 19441377.0,
"step": 10560
},
{
"entropy": 5.2876804828643795,
"epoch": 0.8824757768125626,
"grad_norm": 0.93359375,
"learning_rate": 0.0004928325565093777,
"loss": 5.0382,
"mean_token_accuracy": 0.19772562384605408,
"num_tokens": 19451003.0,
"step": 10565
},
{
"entropy": 5.2811237335205075,
"epoch": 0.8828934179752757,
"grad_norm": 0.875,
"learning_rate": 0.0004928251004566167,
"loss": 4.9288,
"mean_token_accuracy": 0.20963248759508132,
"num_tokens": 19460166.0,
"step": 10570
},
{
"entropy": 5.246886777877807,
"epoch": 0.8833110591379887,
"grad_norm": 0.9453125,
"learning_rate": 0.0004928176405905653,
"loss": 5.0202,
"mean_token_accuracy": 0.20592687726020814,
"num_tokens": 19469417.0,
"step": 10575
},
{
"entropy": 5.330158948898315,
"epoch": 0.8837287003007016,
"grad_norm": 0.9765625,
"learning_rate": 0.0004928101769113537,
"loss": 5.0923,
"mean_token_accuracy": 0.20658836215734483,
"num_tokens": 19479275.0,
"step": 10580
},
{
"entropy": 5.328742456436157,
"epoch": 0.8841463414634146,
"grad_norm": 0.9609375,
"learning_rate": 0.000492802709419113,
"loss": 4.9815,
"mean_token_accuracy": 0.20454305559396743,
"num_tokens": 19488804.0,
"step": 10585
},
{
"entropy": 5.310842561721802,
"epoch": 0.8845639826261277,
"grad_norm": 1.015625,
"learning_rate": 0.0004927952381139735,
"loss": 4.9589,
"mean_token_accuracy": 0.20788733810186386,
"num_tokens": 19497701.0,
"step": 10590
},
{
"entropy": 5.3216719150543215,
"epoch": 0.8849816237888406,
"grad_norm": 0.9765625,
"learning_rate": 0.0004927877629960663,
"loss": 4.9825,
"mean_token_accuracy": 0.20679265409708023,
"num_tokens": 19508177.0,
"step": 10595
},
{
"entropy": 5.250254487991333,
"epoch": 0.8853992649515536,
"grad_norm": 1.078125,
"learning_rate": 0.0004927802840655221,
"loss": 5.0524,
"mean_token_accuracy": 0.20582439452409745,
"num_tokens": 19516866.0,
"step": 10600
},
{
"entropy": 5.3429969310760494,
"epoch": 0.8858169061142667,
"grad_norm": 1.0390625,
"learning_rate": 0.000492772801322472,
"loss": 5.0374,
"mean_token_accuracy": 0.2049921214580536,
"num_tokens": 19525788.0,
"step": 10605
},
{
"entropy": 5.314622402191162,
"epoch": 0.8862345472769796,
"grad_norm": 0.921875,
"learning_rate": 0.0004927653147670468,
"loss": 4.8862,
"mean_token_accuracy": 0.21266158521175385,
"num_tokens": 19534748.0,
"step": 10610
},
{
"entropy": 5.2158294200897215,
"epoch": 0.8866521884396926,
"grad_norm": 1.015625,
"learning_rate": 0.0004927578243993777,
"loss": 4.95,
"mean_token_accuracy": 0.20928464978933334,
"num_tokens": 19544083.0,
"step": 10615
},
{
"entropy": 5.231951093673706,
"epoch": 0.8870698296024057,
"grad_norm": 1.0078125,
"learning_rate": 0.0004927503302195957,
"loss": 4.9784,
"mean_token_accuracy": 0.20495544672012328,
"num_tokens": 19553978.0,
"step": 10620
},
{
"entropy": 5.345741224288941,
"epoch": 0.8874874707651186,
"grad_norm": 0.9296875,
"learning_rate": 0.000492742832227832,
"loss": 5.0564,
"mean_token_accuracy": 0.20942415297031403,
"num_tokens": 19563265.0,
"step": 10625
},
{
"entropy": 5.359615802764893,
"epoch": 0.8879051119278316,
"grad_norm": 0.91015625,
"learning_rate": 0.000492735330424218,
"loss": 5.0864,
"mean_token_accuracy": 0.19267746061086655,
"num_tokens": 19572904.0,
"step": 10630
},
{
"entropy": 5.356269598007202,
"epoch": 0.8883227530905446,
"grad_norm": 1.0703125,
"learning_rate": 0.000492727824808885,
"loss": 5.0792,
"mean_token_accuracy": 0.20686798989772798,
"num_tokens": 19581680.0,
"step": 10635
},
{
"entropy": 5.2298229217529295,
"epoch": 0.8887403942532576,
"grad_norm": 1.03125,
"learning_rate": 0.0004927203153819643,
"loss": 4.9298,
"mean_token_accuracy": 0.20364491939544677,
"num_tokens": 19590885.0,
"step": 10640
},
{
"entropy": 5.2199465274810795,
"epoch": 0.8891580354159706,
"grad_norm": 1.0546875,
"learning_rate": 0.0004927128021435875,
"loss": 4.9232,
"mean_token_accuracy": 0.21275381445884706,
"num_tokens": 19599901.0,
"step": 10645
},
{
"entropy": 5.3653727054595945,
"epoch": 0.8895756765786836,
"grad_norm": 1.0,
"learning_rate": 0.000492705285093886,
"loss": 5.0768,
"mean_token_accuracy": 0.19855906814336777,
"num_tokens": 19608745.0,
"step": 10650
},
{
"entropy": 5.207261371612549,
"epoch": 0.8899933177413966,
"grad_norm": 0.98046875,
"learning_rate": 0.0004926977642329913,
"loss": 4.8383,
"mean_token_accuracy": 0.20991399735212327,
"num_tokens": 19618418.0,
"step": 10655
},
{
"entropy": 5.179308843612671,
"epoch": 0.8904109589041096,
"grad_norm": 1.0390625,
"learning_rate": 0.0004926902395610353,
"loss": 4.9181,
"mean_token_accuracy": 0.2046077623963356,
"num_tokens": 19627118.0,
"step": 10660
},
{
"entropy": 5.312536096572876,
"epoch": 0.8908286000668226,
"grad_norm": 1.046875,
"learning_rate": 0.0004926827110781496,
"loss": 5.0831,
"mean_token_accuracy": 0.20530200153589248,
"num_tokens": 19636565.0,
"step": 10665
},
{
"entropy": 5.2458093643188475,
"epoch": 0.8912462412295356,
"grad_norm": 1.046875,
"learning_rate": 0.000492675178784466,
"loss": 4.9354,
"mean_token_accuracy": 0.20531355142593383,
"num_tokens": 19645038.0,
"step": 10670
},
{
"entropy": 5.244185972213745,
"epoch": 0.8916638823922486,
"grad_norm": 0.9765625,
"learning_rate": 0.0004926676426801164,
"loss": 4.9783,
"mean_token_accuracy": 0.20586746484041213,
"num_tokens": 19654088.0,
"step": 10675
},
{
"entropy": 5.283861446380615,
"epoch": 0.8920815235549616,
"grad_norm": 1.0,
"learning_rate": 0.0004926601027652328,
"loss": 5.0693,
"mean_token_accuracy": 0.20442845821380615,
"num_tokens": 19662551.0,
"step": 10680
},
{
"entropy": 5.328431940078735,
"epoch": 0.8924991647176745,
"grad_norm": 0.9765625,
"learning_rate": 0.000492652559039947,
"loss": 5.0033,
"mean_token_accuracy": 0.2042037934064865,
"num_tokens": 19672405.0,
"step": 10685
},
{
"entropy": 5.265367698669434,
"epoch": 0.8929168058803876,
"grad_norm": 0.90234375,
"learning_rate": 0.0004926450115043912,
"loss": 4.9948,
"mean_token_accuracy": 0.204755437374115,
"num_tokens": 19681288.0,
"step": 10690
},
{
"entropy": 5.287013149261474,
"epoch": 0.8933344470431006,
"grad_norm": 0.97265625,
"learning_rate": 0.0004926374601586975,
"loss": 4.9243,
"mean_token_accuracy": 0.21029869168996812,
"num_tokens": 19690723.0,
"step": 10695
},
{
"entropy": 5.267658948898315,
"epoch": 0.8937520882058135,
"grad_norm": 0.94921875,
"learning_rate": 0.0004926299050029981,
"loss": 4.9793,
"mean_token_accuracy": 0.20806861519813538,
"num_tokens": 19699642.0,
"step": 10700
},
{
"entropy": 5.243737697601318,
"epoch": 0.8941697293685266,
"grad_norm": 0.99609375,
"learning_rate": 0.0004926223460374251,
"loss": 4.9295,
"mean_token_accuracy": 0.21498110443353652,
"num_tokens": 19707563.0,
"step": 10705
},
{
"entropy": 5.297790002822876,
"epoch": 0.8945873705312396,
"grad_norm": 0.9375,
"learning_rate": 0.0004926147832621111,
"loss": 5.0743,
"mean_token_accuracy": 0.2071554332971573,
"num_tokens": 19717109.0,
"step": 10710
},
{
"entropy": 5.216282796859741,
"epoch": 0.8950050116939525,
"grad_norm": 1.0,
"learning_rate": 0.0004926072166771884,
"loss": 4.8931,
"mean_token_accuracy": 0.22160657048225402,
"num_tokens": 19725585.0,
"step": 10715
},
{
"entropy": 5.171828651428223,
"epoch": 0.8954226528566656,
"grad_norm": 1.0546875,
"learning_rate": 0.0004925996462827895,
"loss": 4.8639,
"mean_token_accuracy": 0.21658680438995362,
"num_tokens": 19733502.0,
"step": 10720
},
{
"entropy": 5.2768614292144775,
"epoch": 0.8958402940193786,
"grad_norm": 0.99609375,
"learning_rate": 0.0004925920720790466,
"loss": 4.9806,
"mean_token_accuracy": 0.20507069379091264,
"num_tokens": 19742926.0,
"step": 10725
},
{
"entropy": 5.205672931671143,
"epoch": 0.8962579351820915,
"grad_norm": 1.0,
"learning_rate": 0.0004925844940660928,
"loss": 4.9623,
"mean_token_accuracy": 0.21342433393001556,
"num_tokens": 19752488.0,
"step": 10730
},
{
"entropy": 5.310868167877198,
"epoch": 0.8966755763448045,
"grad_norm": 0.9296875,
"learning_rate": 0.0004925769122440606,
"loss": 5.0981,
"mean_token_accuracy": 0.19976469725370408,
"num_tokens": 19762705.0,
"step": 10735
},
{
"entropy": 5.170697355270386,
"epoch": 0.8970932175075176,
"grad_norm": 1.1015625,
"learning_rate": 0.0004925693266130824,
"loss": 4.8781,
"mean_token_accuracy": 0.21835984587669371,
"num_tokens": 19771156.0,
"step": 10740
},
{
"entropy": 5.256175756454468,
"epoch": 0.8975108586702305,
"grad_norm": 1.1328125,
"learning_rate": 0.0004925617371732914,
"loss": 4.9358,
"mean_token_accuracy": 0.20618152767419815,
"num_tokens": 19780788.0,
"step": 10745
},
{
"entropy": 5.189788770675659,
"epoch": 0.8979284998329435,
"grad_norm": 1.0078125,
"learning_rate": 0.0004925541439248203,
"loss": 4.9265,
"mean_token_accuracy": 0.2176087647676468,
"num_tokens": 19790169.0,
"step": 10750
},
{
"entropy": 5.373470830917358,
"epoch": 0.8983461409956566,
"grad_norm": 1.015625,
"learning_rate": 0.000492546546867802,
"loss": 5.1422,
"mean_token_accuracy": 0.19242381006479264,
"num_tokens": 19800249.0,
"step": 10755
},
{
"entropy": 5.251127243041992,
"epoch": 0.8987637821583695,
"grad_norm": 0.97265625,
"learning_rate": 0.0004925389460023696,
"loss": 4.9958,
"mean_token_accuracy": 0.2054883360862732,
"num_tokens": 19809797.0,
"step": 10760
},
{
"entropy": 5.196225976943969,
"epoch": 0.8991814233210825,
"grad_norm": 0.9375,
"learning_rate": 0.0004925313413286559,
"loss": 4.8812,
"mean_token_accuracy": 0.2160322204232216,
"num_tokens": 19818826.0,
"step": 10765
},
{
"entropy": 5.269199419021606,
"epoch": 0.8995990644837956,
"grad_norm": 1.015625,
"learning_rate": 0.0004925237328467945,
"loss": 5.0388,
"mean_token_accuracy": 0.2077295333147049,
"num_tokens": 19827947.0,
"step": 10770
},
{
"entropy": 5.355157375335693,
"epoch": 0.9000167056465085,
"grad_norm": 1.1328125,
"learning_rate": 0.0004925161205569182,
"loss": 5.0569,
"mean_token_accuracy": 0.20308083146810532,
"num_tokens": 19835734.0,
"step": 10775
},
{
"entropy": 5.406469345092773,
"epoch": 0.9004343468092215,
"grad_norm": 0.84375,
"learning_rate": 0.0004925085044591604,
"loss": 5.1613,
"mean_token_accuracy": 0.19586416184902192,
"num_tokens": 19845175.0,
"step": 10780
},
{
"entropy": 5.2762151718139645,
"epoch": 0.9008519879719346,
"grad_norm": 1.0078125,
"learning_rate": 0.0004925008845536544,
"loss": 4.9804,
"mean_token_accuracy": 0.20715121924877167,
"num_tokens": 19853684.0,
"step": 10785
},
{
"entropy": 5.295101833343506,
"epoch": 0.9012696291346475,
"grad_norm": 1.03125,
"learning_rate": 0.0004924932608405336,
"loss": 4.982,
"mean_token_accuracy": 0.20433221459388734,
"num_tokens": 19862813.0,
"step": 10790
},
{
"entropy": 5.296811246871949,
"epoch": 0.9016872702973605,
"grad_norm": 1.09375,
"learning_rate": 0.0004924856333199315,
"loss": 5.0199,
"mean_token_accuracy": 0.20576008558273315,
"num_tokens": 19871545.0,
"step": 10795
},
{
"entropy": 5.279951477050782,
"epoch": 0.9021049114600735,
"grad_norm": 1.125,
"learning_rate": 0.0004924780019919816,
"loss": 4.9955,
"mean_token_accuracy": 0.2054675415158272,
"num_tokens": 19880060.0,
"step": 10800
},
{
"entropy": 5.244395399093628,
"epoch": 0.9025225526227865,
"grad_norm": 1.03125,
"learning_rate": 0.0004924703668568174,
"loss": 5.0211,
"mean_token_accuracy": 0.20388415306806565,
"num_tokens": 19889151.0,
"step": 10805
},
{
"entropy": 5.331202793121338,
"epoch": 0.9029401937854995,
"grad_norm": 1.0390625,
"learning_rate": 0.0004924627279145728,
"loss": 4.97,
"mean_token_accuracy": 0.20639394074678422,
"num_tokens": 19898087.0,
"step": 10810
},
{
"entropy": 5.24865951538086,
"epoch": 0.9033578349482125,
"grad_norm": 1.0,
"learning_rate": 0.0004924550851653814,
"loss": 4.965,
"mean_token_accuracy": 0.21553518921136855,
"num_tokens": 19906196.0,
"step": 10815
},
{
"entropy": 5.284704256057739,
"epoch": 0.9037754761109255,
"grad_norm": 0.96484375,
"learning_rate": 0.0004924474386093769,
"loss": 5.0098,
"mean_token_accuracy": 0.20644852817058562,
"num_tokens": 19915837.0,
"step": 10820
},
{
"entropy": 5.2688133239746096,
"epoch": 0.9041931172736385,
"grad_norm": 0.9921875,
"learning_rate": 0.0004924397882466933,
"loss": 5.001,
"mean_token_accuracy": 0.2050878494977951,
"num_tokens": 19926012.0,
"step": 10825
},
{
"entropy": 5.243179082870483,
"epoch": 0.9046107584363515,
"grad_norm": 0.9765625,
"learning_rate": 0.0004924321340774644,
"loss": 4.9628,
"mean_token_accuracy": 0.2105299010872841,
"num_tokens": 19935987.0,
"step": 10830
},
{
"entropy": 5.243394231796264,
"epoch": 0.9050283995990644,
"grad_norm": 0.97265625,
"learning_rate": 0.0004924244761018243,
"loss": 4.9833,
"mean_token_accuracy": 0.204965278506279,
"num_tokens": 19945865.0,
"step": 10835
},
{
"entropy": 5.31999363899231,
"epoch": 0.9054460407617775,
"grad_norm": 1.046875,
"learning_rate": 0.000492416814319907,
"loss": 5.0726,
"mean_token_accuracy": 0.2043199881911278,
"num_tokens": 19954824.0,
"step": 10840
},
{
"entropy": 5.234289121627808,
"epoch": 0.9058636819244905,
"grad_norm": 0.9765625,
"learning_rate": 0.0004924091487318468,
"loss": 4.9692,
"mean_token_accuracy": 0.2030353620648384,
"num_tokens": 19964740.0,
"step": 10845
},
{
"entropy": 5.232080554962158,
"epoch": 0.9062813230872034,
"grad_norm": 0.9296875,
"learning_rate": 0.0004924014793377776,
"loss": 4.9289,
"mean_token_accuracy": 0.21160493344068526,
"num_tokens": 19974310.0,
"step": 10850
},
{
"entropy": 5.222050380706787,
"epoch": 0.9066989642499165,
"grad_norm": 1.078125,
"learning_rate": 0.0004923938061378339,
"loss": 4.8703,
"mean_token_accuracy": 0.21050190329551696,
"num_tokens": 19982943.0,
"step": 10855
},
{
"entropy": 5.250167369842529,
"epoch": 0.9071166054126295,
"grad_norm": 0.99609375,
"learning_rate": 0.0004923861291321499,
"loss": 4.9714,
"mean_token_accuracy": 0.20778299868106842,
"num_tokens": 19991971.0,
"step": 10860
},
{
"entropy": 5.336087942123413,
"epoch": 0.9075342465753424,
"grad_norm": 1.0625,
"learning_rate": 0.0004923784483208601,
"loss": 5.037,
"mean_token_accuracy": 0.19941984713077546,
"num_tokens": 20001491.0,
"step": 10865
},
{
"entropy": 5.242482233047485,
"epoch": 0.9079518877380555,
"grad_norm": 0.96875,
"learning_rate": 0.0004923707637040989,
"loss": 4.8845,
"mean_token_accuracy": 0.21396315693855286,
"num_tokens": 20010605.0,
"step": 10870
},
{
"entropy": 5.121878004074096,
"epoch": 0.9083695289007685,
"grad_norm": 1.03125,
"learning_rate": 0.0004923630752820008,
"loss": 4.7886,
"mean_token_accuracy": 0.21492691487073898,
"num_tokens": 20019291.0,
"step": 10875
},
{
"entropy": 5.222338056564331,
"epoch": 0.9087871700634814,
"grad_norm": 0.9609375,
"learning_rate": 0.0004923553830547004,
"loss": 5.085,
"mean_token_accuracy": 0.195840784907341,
"num_tokens": 20029048.0,
"step": 10880
},
{
"entropy": 5.309308242797852,
"epoch": 0.9092048112261945,
"grad_norm": 1.015625,
"learning_rate": 0.0004923476870223324,
"loss": 5.0307,
"mean_token_accuracy": 0.20123770833015442,
"num_tokens": 20037878.0,
"step": 10885
},
{
"entropy": 5.2229204177856445,
"epoch": 0.9096224523889075,
"grad_norm": 1.109375,
"learning_rate": 0.0004923399871850315,
"loss": 4.9588,
"mean_token_accuracy": 0.2096124067902565,
"num_tokens": 20046662.0,
"step": 10890
},
{
"entropy": 5.18960976600647,
"epoch": 0.9100400935516204,
"grad_norm": 1.0078125,
"learning_rate": 0.0004923322835429326,
"loss": 4.9292,
"mean_token_accuracy": 0.21052929162979125,
"num_tokens": 20055329.0,
"step": 10895
},
{
"entropy": 5.148437118530273,
"epoch": 0.9104577347143334,
"grad_norm": 0.95703125,
"learning_rate": 0.0004923245760961704,
"loss": 4.8406,
"mean_token_accuracy": 0.21717104613780974,
"num_tokens": 20064024.0,
"step": 10900
},
{
"entropy": 5.2552488327026365,
"epoch": 0.9108753758770465,
"grad_norm": 0.99609375,
"learning_rate": 0.0004923168648448799,
"loss": 5.0419,
"mean_token_accuracy": 0.20019593983888626,
"num_tokens": 20073022.0,
"step": 10905
},
{
"entropy": 5.291287231445312,
"epoch": 0.9112930170397594,
"grad_norm": 0.96484375,
"learning_rate": 0.0004923091497891961,
"loss": 5.0044,
"mean_token_accuracy": 0.20060400515794755,
"num_tokens": 20081495.0,
"step": 10910
},
{
"entropy": 5.279441165924072,
"epoch": 0.9117106582024724,
"grad_norm": 0.9609375,
"learning_rate": 0.0004923014309292541,
"loss": 4.9457,
"mean_token_accuracy": 0.21415825486183165,
"num_tokens": 20090545.0,
"step": 10915
},
{
"entropy": 5.255401611328125,
"epoch": 0.9121282993651855,
"grad_norm": 1.0234375,
"learning_rate": 0.000492293708265189,
"loss": 5.0246,
"mean_token_accuracy": 0.20592290461063384,
"num_tokens": 20100236.0,
"step": 10920
},
{
"entropy": 5.190516519546509,
"epoch": 0.9125459405278984,
"grad_norm": 1.0234375,
"learning_rate": 0.0004922859817971359,
"loss": 4.9642,
"mean_token_accuracy": 0.210262168943882,
"num_tokens": 20109522.0,
"step": 10925
},
{
"entropy": 5.235836505889893,
"epoch": 0.9129635816906114,
"grad_norm": 1.109375,
"learning_rate": 0.0004922782515252302,
"loss": 4.8962,
"mean_token_accuracy": 0.21298593133687974,
"num_tokens": 20118248.0,
"step": 10930
},
{
"entropy": 5.200658464431763,
"epoch": 0.9133812228533245,
"grad_norm": 1.0859375,
"learning_rate": 0.0004922705174496072,
"loss": 4.9585,
"mean_token_accuracy": 0.20346722900867462,
"num_tokens": 20127096.0,
"step": 10935
},
{
"entropy": 5.214232921600342,
"epoch": 0.9137988640160374,
"grad_norm": 1.03125,
"learning_rate": 0.0004922627795704022,
"loss": 4.9587,
"mean_token_accuracy": 0.20433655232191086,
"num_tokens": 20136815.0,
"step": 10940
},
{
"entropy": 5.261284971237183,
"epoch": 0.9142165051787504,
"grad_norm": 1.03125,
"learning_rate": 0.0004922550378877507,
"loss": 4.9542,
"mean_token_accuracy": 0.2149657592177391,
"num_tokens": 20145455.0,
"step": 10945
},
{
"entropy": 5.284930849075318,
"epoch": 0.9146341463414634,
"grad_norm": 0.98828125,
"learning_rate": 0.0004922472924017882,
"loss": 5.0599,
"mean_token_accuracy": 0.20481502562761306,
"num_tokens": 20156139.0,
"step": 10950
},
{
"entropy": 5.362521553039551,
"epoch": 0.9150517875041764,
"grad_norm": 0.91015625,
"learning_rate": 0.0004922395431126506,
"loss": 5.0348,
"mean_token_accuracy": 0.20549243092536926,
"num_tokens": 20166266.0,
"step": 10955
},
{
"entropy": 5.416420793533325,
"epoch": 0.9154694286668894,
"grad_norm": 1.015625,
"learning_rate": 0.0004922317900204732,
"loss": 5.0677,
"mean_token_accuracy": 0.20066986232995987,
"num_tokens": 20175005.0,
"step": 10960
},
{
"entropy": 5.168547105789185,
"epoch": 0.9158870698296024,
"grad_norm": 1.046875,
"learning_rate": 0.0004922240331253916,
"loss": 4.8835,
"mean_token_accuracy": 0.21400633454322815,
"num_tokens": 20183488.0,
"step": 10965
},
{
"entropy": 5.2441362857818605,
"epoch": 0.9163047109923154,
"grad_norm": 0.98828125,
"learning_rate": 0.0004922162724275421,
"loss": 4.9734,
"mean_token_accuracy": 0.21193478107452393,
"num_tokens": 20193724.0,
"step": 10970
},
{
"entropy": 5.263073968887329,
"epoch": 0.9167223521550284,
"grad_norm": 0.9921875,
"learning_rate": 0.0004922085079270602,
"loss": 5.0584,
"mean_token_accuracy": 0.20677319765090943,
"num_tokens": 20203733.0,
"step": 10975
},
{
"entropy": 5.3670989036560055,
"epoch": 0.9171399933177414,
"grad_norm": 1.0390625,
"learning_rate": 0.0004922007396240819,
"loss": 5.0425,
"mean_token_accuracy": 0.21014897525310516,
"num_tokens": 20213448.0,
"step": 10980
},
{
"entropy": 5.320993089675904,
"epoch": 0.9175576344804544,
"grad_norm": 0.94140625,
"learning_rate": 0.0004921929675187433,
"loss": 4.9651,
"mean_token_accuracy": 0.20755314379930495,
"num_tokens": 20222837.0,
"step": 10985
},
{
"entropy": 5.351680946350098,
"epoch": 0.9179752756431674,
"grad_norm": 1.0078125,
"learning_rate": 0.0004921851916111802,
"loss": 5.1355,
"mean_token_accuracy": 0.1980899006128311,
"num_tokens": 20232056.0,
"step": 10990
},
{
"entropy": 5.375345993041992,
"epoch": 0.9183929168058804,
"grad_norm": 1.0078125,
"learning_rate": 0.0004921774119015289,
"loss": 5.0507,
"mean_token_accuracy": 0.20475190728902817,
"num_tokens": 20241052.0,
"step": 10995
},
{
"entropy": 5.37616753578186,
"epoch": 0.9188105579685933,
"grad_norm": 0.9609375,
"learning_rate": 0.0004921696283899257,
"loss": 5.1092,
"mean_token_accuracy": 0.1991646721959114,
"num_tokens": 20251015.0,
"step": 11000
},
{
"entropy": 5.245299911499023,
"epoch": 0.9192281991313064,
"grad_norm": 1.03125,
"learning_rate": 0.0004921618410765067,
"loss": 4.9514,
"mean_token_accuracy": 0.20571413040161132,
"num_tokens": 20259388.0,
"step": 11005
},
{
"entropy": 5.294223833084106,
"epoch": 0.9196458402940194,
"grad_norm": 0.953125,
"learning_rate": 0.0004921540499614083,
"loss": 4.9312,
"mean_token_accuracy": 0.20074452906847,
"num_tokens": 20268146.0,
"step": 11010
},
{
"entropy": 5.236671781539917,
"epoch": 0.9200634814567323,
"grad_norm": 1.03125,
"learning_rate": 0.0004921462550447669,
"loss": 4.9158,
"mean_token_accuracy": 0.21233065873384477,
"num_tokens": 20277559.0,
"step": 11015
},
{
"entropy": 5.266069841384888,
"epoch": 0.9204811226194454,
"grad_norm": 1.0234375,
"learning_rate": 0.0004921384563267189,
"loss": 4.9534,
"mean_token_accuracy": 0.20886976569890975,
"num_tokens": 20285661.0,
"step": 11020
},
{
"entropy": 5.243994903564453,
"epoch": 0.9208987637821584,
"grad_norm": 0.9453125,
"learning_rate": 0.0004921306538074008,
"loss": 5.0358,
"mean_token_accuracy": 0.20855261087417604,
"num_tokens": 20294636.0,
"step": 11025
},
{
"entropy": 5.316732931137085,
"epoch": 0.9213164049448713,
"grad_norm": 1.0859375,
"learning_rate": 0.0004921228474869493,
"loss": 4.9321,
"mean_token_accuracy": 0.2154439702630043,
"num_tokens": 20302372.0,
"step": 11030
},
{
"entropy": 5.293783855438233,
"epoch": 0.9217340461075844,
"grad_norm": 1.046875,
"learning_rate": 0.000492115037365501,
"loss": 4.9548,
"mean_token_accuracy": 0.21032945066690445,
"num_tokens": 20311974.0,
"step": 11035
},
{
"entropy": 5.239782476425171,
"epoch": 0.9221516872702974,
"grad_norm": 0.94921875,
"learning_rate": 0.0004921072234431925,
"loss": 4.9744,
"mean_token_accuracy": 0.2089963212609291,
"num_tokens": 20321720.0,
"step": 11040
},
{
"entropy": 5.259523344039917,
"epoch": 0.9225693284330103,
"grad_norm": 0.9609375,
"learning_rate": 0.000492099405720161,
"loss": 5.0338,
"mean_token_accuracy": 0.2014230892062187,
"num_tokens": 20331324.0,
"step": 11045
},
{
"entropy": 5.395723628997803,
"epoch": 0.9229869695957233,
"grad_norm": 0.96875,
"learning_rate": 0.0004920915841965429,
"loss": 5.1289,
"mean_token_accuracy": 0.1988341435790062,
"num_tokens": 20340921.0,
"step": 11050
},
{
"entropy": 5.395040130615234,
"epoch": 0.9234046107584364,
"grad_norm": 0.92578125,
"learning_rate": 0.0004920837588724753,
"loss": 4.9965,
"mean_token_accuracy": 0.20739365220069886,
"num_tokens": 20349871.0,
"step": 11055
},
{
"entropy": 5.168009662628174,
"epoch": 0.9238222519211493,
"grad_norm": 0.9609375,
"learning_rate": 0.0004920759297480953,
"loss": 4.7862,
"mean_token_accuracy": 0.21604011803865433,
"num_tokens": 20359171.0,
"step": 11060
},
{
"entropy": 5.255016708374024,
"epoch": 0.9242398930838623,
"grad_norm": 1.109375,
"learning_rate": 0.0004920680968235397,
"loss": 4.9953,
"mean_token_accuracy": 0.2181284636259079,
"num_tokens": 20367779.0,
"step": 11065
},
{
"entropy": 5.226720952987671,
"epoch": 0.9246575342465754,
"grad_norm": 0.98828125,
"learning_rate": 0.0004920602600989459,
"loss": 4.907,
"mean_token_accuracy": 0.21906644403934478,
"num_tokens": 20377568.0,
"step": 11070
},
{
"entropy": 5.34415636062622,
"epoch": 0.9250751754092883,
"grad_norm": 1.0390625,
"learning_rate": 0.0004920524195744509,
"loss": 5.046,
"mean_token_accuracy": 0.2030154287815094,
"num_tokens": 20385915.0,
"step": 11075
},
{
"entropy": 5.228349399566651,
"epoch": 0.9254928165720013,
"grad_norm": 1.0234375,
"learning_rate": 0.0004920445752501921,
"loss": 4.9184,
"mean_token_accuracy": 0.21054769605398177,
"num_tokens": 20394877.0,
"step": 11080
},
{
"entropy": 5.249534225463867,
"epoch": 0.9259104577347144,
"grad_norm": 1.0390625,
"learning_rate": 0.0004920367271263066,
"loss": 4.9262,
"mean_token_accuracy": 0.2117575228214264,
"num_tokens": 20403741.0,
"step": 11085
},
{
"entropy": 5.280802917480469,
"epoch": 0.9263280988974273,
"grad_norm": 1.1796875,
"learning_rate": 0.0004920288752029321,
"loss": 4.9683,
"mean_token_accuracy": 0.20721979588270187,
"num_tokens": 20412702.0,
"step": 11090
},
{
"entropy": 5.27918553352356,
"epoch": 0.9267457400601403,
"grad_norm": 0.94921875,
"learning_rate": 0.0004920210194802059,
"loss": 4.9889,
"mean_token_accuracy": 0.2074853777885437,
"num_tokens": 20421776.0,
"step": 11095
},
{
"entropy": 5.303801488876343,
"epoch": 0.9271633812228534,
"grad_norm": 0.921875,
"learning_rate": 0.0004920131599582654,
"loss": 5.0949,
"mean_token_accuracy": 0.19518010467290878,
"num_tokens": 20432204.0,
"step": 11100
},
{
"entropy": 5.315025472640992,
"epoch": 0.9275810223855663,
"grad_norm": 0.94921875,
"learning_rate": 0.0004920052966372485,
"loss": 4.9951,
"mean_token_accuracy": 0.19930322617292404,
"num_tokens": 20441344.0,
"step": 11105
},
{
"entropy": 5.247663974761963,
"epoch": 0.9279986635482793,
"grad_norm": 1.0,
"learning_rate": 0.0004919974295172926,
"loss": 4.909,
"mean_token_accuracy": 0.21379464715719224,
"num_tokens": 20449800.0,
"step": 11110
},
{
"entropy": 5.231588459014892,
"epoch": 0.9284163047109923,
"grad_norm": 0.98046875,
"learning_rate": 0.0004919895585985353,
"loss": 4.8505,
"mean_token_accuracy": 0.22388574630022048,
"num_tokens": 20458326.0,
"step": 11115
},
{
"entropy": 5.2991370677948,
"epoch": 0.9288339458737053,
"grad_norm": 1.03125,
"learning_rate": 0.0004919816838811149,
"loss": 5.1211,
"mean_token_accuracy": 0.19915229082107544,
"num_tokens": 20468050.0,
"step": 11120
},
{
"entropy": 5.266911745071411,
"epoch": 0.9292515870364183,
"grad_norm": 0.8828125,
"learning_rate": 0.0004919738053651687,
"loss": 4.8833,
"mean_token_accuracy": 0.21326978355646134,
"num_tokens": 20477046.0,
"step": 11125
},
{
"entropy": 5.272528457641601,
"epoch": 0.9296692281991313,
"grad_norm": 1.0546875,
"learning_rate": 0.000491965923050835,
"loss": 4.9921,
"mean_token_accuracy": 0.20975257754325866,
"num_tokens": 20486228.0,
"step": 11130
},
{
"entropy": 5.300780534744263,
"epoch": 0.9300868693618443,
"grad_norm": 1.1484375,
"learning_rate": 0.0004919580369382516,
"loss": 5.0064,
"mean_token_accuracy": 0.20058976858854294,
"num_tokens": 20494764.0,
"step": 11135
},
{
"entropy": 5.278902912139893,
"epoch": 0.9305045105245573,
"grad_norm": 1.1171875,
"learning_rate": 0.0004919501470275567,
"loss": 4.9044,
"mean_token_accuracy": 0.21128441244363785,
"num_tokens": 20502805.0,
"step": 11140
},
{
"entropy": 5.244833755493164,
"epoch": 0.9309221516872703,
"grad_norm": 0.96484375,
"learning_rate": 0.0004919422533188882,
"loss": 4.9672,
"mean_token_accuracy": 0.20627895295619963,
"num_tokens": 20512322.0,
"step": 11145
},
{
"entropy": 5.2312281131744385,
"epoch": 0.9313397928499833,
"grad_norm": 1.1796875,
"learning_rate": 0.0004919343558123845,
"loss": 4.9584,
"mean_token_accuracy": 0.20781377851963043,
"num_tokens": 20520680.0,
"step": 11150
},
{
"entropy": 5.217152070999146,
"epoch": 0.9317574340126963,
"grad_norm": 0.98046875,
"learning_rate": 0.0004919264545081838,
"loss": 4.8278,
"mean_token_accuracy": 0.22223591953516006,
"num_tokens": 20528864.0,
"step": 11155
},
{
"entropy": 5.281461763381958,
"epoch": 0.9321750751754093,
"grad_norm": 1.0078125,
"learning_rate": 0.0004919185494064244,
"loss": 4.9528,
"mean_token_accuracy": 0.20506721138954162,
"num_tokens": 20538195.0,
"step": 11160
},
{
"entropy": 5.234458160400391,
"epoch": 0.9325927163381222,
"grad_norm": 0.984375,
"learning_rate": 0.0004919106405072446,
"loss": 4.9982,
"mean_token_accuracy": 0.20177790373563767,
"num_tokens": 20547312.0,
"step": 11165
},
{
"entropy": 5.277944707870484,
"epoch": 0.9330103575008353,
"grad_norm": 1.0234375,
"learning_rate": 0.000491902727810783,
"loss": 5.0296,
"mean_token_accuracy": 0.19958221316337585,
"num_tokens": 20557184.0,
"step": 11170
},
{
"entropy": 5.263872766494751,
"epoch": 0.9334279986635483,
"grad_norm": 1.1171875,
"learning_rate": 0.0004918948113171781,
"loss": 4.9558,
"mean_token_accuracy": 0.2094264268875122,
"num_tokens": 20565819.0,
"step": 11175
},
{
"entropy": 5.245273160934448,
"epoch": 0.9338456398262612,
"grad_norm": 0.9921875,
"learning_rate": 0.0004918868910265684,
"loss": 5.0102,
"mean_token_accuracy": 0.21310826241970063,
"num_tokens": 20574761.0,
"step": 11180
},
{
"entropy": 5.353975248336792,
"epoch": 0.9342632809889743,
"grad_norm": 1.0078125,
"learning_rate": 0.0004918789669390927,
"loss": 5.0708,
"mean_token_accuracy": 0.19999473989009858,
"num_tokens": 20584003.0,
"step": 11185
},
{
"entropy": 5.3034553050994875,
"epoch": 0.9346809221516873,
"grad_norm": 1.0234375,
"learning_rate": 0.0004918710390548895,
"loss": 4.9182,
"mean_token_accuracy": 0.20792448669672012,
"num_tokens": 20593123.0,
"step": 11190
},
{
"entropy": 5.260336971282959,
"epoch": 0.9350985633144002,
"grad_norm": 1.046875,
"learning_rate": 0.0004918631073740977,
"loss": 5.0135,
"mean_token_accuracy": 0.2061988204717636,
"num_tokens": 20602582.0,
"step": 11195
},
{
"entropy": 5.245416355133057,
"epoch": 0.9355162044771133,
"grad_norm": 0.984375,
"learning_rate": 0.0004918551718968562,
"loss": 4.9022,
"mean_token_accuracy": 0.2127738282084465,
"num_tokens": 20611681.0,
"step": 11200
},
{
"entropy": 5.222401571273804,
"epoch": 0.9359338456398263,
"grad_norm": 1.0078125,
"learning_rate": 0.000491847232623304,
"loss": 4.973,
"mean_token_accuracy": 0.21008606553077697,
"num_tokens": 20620671.0,
"step": 11205
},
{
"entropy": 5.255436658859253,
"epoch": 0.9363514868025392,
"grad_norm": 0.953125,
"learning_rate": 0.0004918392895535798,
"loss": 4.951,
"mean_token_accuracy": 0.20708223581314086,
"num_tokens": 20629986.0,
"step": 11210
},
{
"entropy": 5.234728813171387,
"epoch": 0.9367691279652522,
"grad_norm": 1.0,
"learning_rate": 0.000491831342687823,
"loss": 4.9395,
"mean_token_accuracy": 0.20146991908550263,
"num_tokens": 20639274.0,
"step": 11215
},
{
"entropy": 5.264416837692261,
"epoch": 0.9371867691279653,
"grad_norm": 0.95703125,
"learning_rate": 0.0004918233920261723,
"loss": 4.9448,
"mean_token_accuracy": 0.2100604459643364,
"num_tokens": 20647753.0,
"step": 11220
},
{
"entropy": 5.1743621826171875,
"epoch": 0.9376044102906782,
"grad_norm": 1.03125,
"learning_rate": 0.0004918154375687673,
"loss": 4.8645,
"mean_token_accuracy": 0.2156805217266083,
"num_tokens": 20656226.0,
"step": 11225
},
{
"entropy": 5.183247852325439,
"epoch": 0.9380220514533912,
"grad_norm": 0.984375,
"learning_rate": 0.0004918074793157472,
"loss": 4.87,
"mean_token_accuracy": 0.2159823328256607,
"num_tokens": 20664682.0,
"step": 11230
},
{
"entropy": 5.283345317840576,
"epoch": 0.9384396926161043,
"grad_norm": 1.046875,
"learning_rate": 0.0004917995172672511,
"loss": 4.9607,
"mean_token_accuracy": 0.2104622170329094,
"num_tokens": 20672909.0,
"step": 11235
},
{
"entropy": 5.155306100845337,
"epoch": 0.9388573337788172,
"grad_norm": 1.078125,
"learning_rate": 0.0004917915514234184,
"loss": 4.8271,
"mean_token_accuracy": 0.21582880169153212,
"num_tokens": 20681316.0,
"step": 11240
},
{
"entropy": 5.27658634185791,
"epoch": 0.9392749749415302,
"grad_norm": 0.984375,
"learning_rate": 0.0004917835817843887,
"loss": 4.9374,
"mean_token_accuracy": 0.20264142453670503,
"num_tokens": 20690360.0,
"step": 11245
},
{
"entropy": 5.293812417984009,
"epoch": 0.9396926161042433,
"grad_norm": 0.91015625,
"learning_rate": 0.0004917756083503014,
"loss": 5.0009,
"mean_token_accuracy": 0.203963802754879,
"num_tokens": 20700398.0,
"step": 11250
},
{
"entropy": 5.320089101791382,
"epoch": 0.9401102572669562,
"grad_norm": 0.96875,
"learning_rate": 0.0004917676311212962,
"loss": 5.073,
"mean_token_accuracy": 0.20118411481380463,
"num_tokens": 20709705.0,
"step": 11255
},
{
"entropy": 5.200478363037109,
"epoch": 0.9405278984296692,
"grad_norm": 1.1875,
"learning_rate": 0.0004917596500975127,
"loss": 4.8916,
"mean_token_accuracy": 0.21248382180929185,
"num_tokens": 20717774.0,
"step": 11260
},
{
"entropy": 5.277456903457642,
"epoch": 0.9409455395923823,
"grad_norm": 1.125,
"learning_rate": 0.0004917516652790907,
"loss": 5.031,
"mean_token_accuracy": 0.20552293211221695,
"num_tokens": 20725966.0,
"step": 11265
},
{
"entropy": 5.307353353500366,
"epoch": 0.9413631807550952,
"grad_norm": 0.95703125,
"learning_rate": 0.0004917436766661697,
"loss": 4.9427,
"mean_token_accuracy": 0.20830944180488586,
"num_tokens": 20735254.0,
"step": 11270
},
{
"entropy": 5.240226364135742,
"epoch": 0.9417808219178082,
"grad_norm": 1.1015625,
"learning_rate": 0.00049173568425889,
"loss": 4.8485,
"mean_token_accuracy": 0.2170790359377861,
"num_tokens": 20743331.0,
"step": 11275
},
{
"entropy": 5.259325122833252,
"epoch": 0.9421984630805212,
"grad_norm": 0.94140625,
"learning_rate": 0.0004917276880573911,
"loss": 4.975,
"mean_token_accuracy": 0.20236670076847077,
"num_tokens": 20753966.0,
"step": 11280
},
{
"entropy": 5.275627517700196,
"epoch": 0.9426161042432342,
"grad_norm": 1.0234375,
"learning_rate": 0.0004917196880618132,
"loss": 4.9168,
"mean_token_accuracy": 0.21448712795972824,
"num_tokens": 20762608.0,
"step": 11285
},
{
"entropy": 5.192218828201294,
"epoch": 0.9430337454059472,
"grad_norm": 0.9921875,
"learning_rate": 0.0004917116842722962,
"loss": 4.8757,
"mean_token_accuracy": 0.21021183729171752,
"num_tokens": 20771194.0,
"step": 11290
},
{
"entropy": 5.259437608718872,
"epoch": 0.9434513865686602,
"grad_norm": 1.0234375,
"learning_rate": 0.0004917036766889804,
"loss": 4.9802,
"mean_token_accuracy": 0.20911021232604982,
"num_tokens": 20779700.0,
"step": 11295
},
{
"entropy": 5.316693210601807,
"epoch": 0.9438690277313732,
"grad_norm": 1.078125,
"learning_rate": 0.0004916956653120058,
"loss": 5.0725,
"mean_token_accuracy": 0.20024692565202712,
"num_tokens": 20788117.0,
"step": 11300
},
{
"entropy": 5.264907312393189,
"epoch": 0.9442866688940862,
"grad_norm": 0.94921875,
"learning_rate": 0.0004916876501415129,
"loss": 5.0558,
"mean_token_accuracy": 0.20022192299365998,
"num_tokens": 20798041.0,
"step": 11305
},
{
"entropy": 5.293879556655884,
"epoch": 0.9447043100567992,
"grad_norm": 0.93359375,
"learning_rate": 0.0004916796311776417,
"loss": 5.0026,
"mean_token_accuracy": 0.19762389063835145,
"num_tokens": 20807423.0,
"step": 11310
},
{
"entropy": 5.250337839126587,
"epoch": 0.9451219512195121,
"grad_norm": 1.078125,
"learning_rate": 0.0004916716084205328,
"loss": 4.8904,
"mean_token_accuracy": 0.2176752507686615,
"num_tokens": 20816727.0,
"step": 11315
},
{
"entropy": 5.270186901092529,
"epoch": 0.9455395923822252,
"grad_norm": 1.0625,
"learning_rate": 0.0004916635818703266,
"loss": 4.9121,
"mean_token_accuracy": 0.20919148176908492,
"num_tokens": 20824855.0,
"step": 11320
},
{
"entropy": 5.197906923294068,
"epoch": 0.9459572335449382,
"grad_norm": 1.0234375,
"learning_rate": 0.0004916555515271635,
"loss": 4.9263,
"mean_token_accuracy": 0.21446057111024858,
"num_tokens": 20833096.0,
"step": 11325
},
{
"entropy": 5.254100799560547,
"epoch": 0.9463748747076511,
"grad_norm": 0.97265625,
"learning_rate": 0.0004916475173911844,
"loss": 4.8777,
"mean_token_accuracy": 0.21171997338533402,
"num_tokens": 20842398.0,
"step": 11330
},
{
"entropy": 5.386715650558472,
"epoch": 0.9467925158703642,
"grad_norm": 0.96875,
"learning_rate": 0.0004916394794625296,
"loss": 5.0932,
"mean_token_accuracy": 0.19939033687114716,
"num_tokens": 20851745.0,
"step": 11335
},
{
"entropy": 5.422706842422485,
"epoch": 0.9472101570330772,
"grad_norm": 0.984375,
"learning_rate": 0.0004916314377413399,
"loss": 5.0041,
"mean_token_accuracy": 0.20473885238170625,
"num_tokens": 20860935.0,
"step": 11340
},
{
"entropy": 5.305401611328125,
"epoch": 0.9476277981957901,
"grad_norm": 0.9375,
"learning_rate": 0.0004916233922277562,
"loss": 5.0025,
"mean_token_accuracy": 0.20276060998439788,
"num_tokens": 20871088.0,
"step": 11345
},
{
"entropy": 5.244873189926148,
"epoch": 0.9480454393585032,
"grad_norm": 1.140625,
"learning_rate": 0.0004916153429219192,
"loss": 5.0159,
"mean_token_accuracy": 0.21051167249679564,
"num_tokens": 20879313.0,
"step": 11350
},
{
"entropy": 5.275855207443238,
"epoch": 0.9484630805212162,
"grad_norm": 1.078125,
"learning_rate": 0.0004916072898239699,
"loss": 4.9565,
"mean_token_accuracy": 0.20832324475049974,
"num_tokens": 20888392.0,
"step": 11355
},
{
"entropy": 5.336981391906738,
"epoch": 0.9488807216839291,
"grad_norm": 1.0234375,
"learning_rate": 0.0004915992329340493,
"loss": 5.0598,
"mean_token_accuracy": 0.19961524307727813,
"num_tokens": 20897456.0,
"step": 11360
},
{
"entropy": 5.292699527740479,
"epoch": 0.9492983628466422,
"grad_norm": 0.95703125,
"learning_rate": 0.0004915911722522984,
"loss": 4.9905,
"mean_token_accuracy": 0.20827175676822662,
"num_tokens": 20906621.0,
"step": 11365
},
{
"entropy": 5.404400825500488,
"epoch": 0.9497160040093552,
"grad_norm": 0.9453125,
"learning_rate": 0.0004915831077788583,
"loss": 5.1709,
"mean_token_accuracy": 0.19984681457281112,
"num_tokens": 20916929.0,
"step": 11370
},
{
"entropy": 5.30355224609375,
"epoch": 0.9501336451720681,
"grad_norm": 0.96484375,
"learning_rate": 0.0004915750395138702,
"loss": 4.9905,
"mean_token_accuracy": 0.20681142508983613,
"num_tokens": 20926028.0,
"step": 11375
},
{
"entropy": 5.324733066558838,
"epoch": 0.9505512863347811,
"grad_norm": 1.0703125,
"learning_rate": 0.0004915669674574754,
"loss": 5.0682,
"mean_token_accuracy": 0.19354701191186904,
"num_tokens": 20935516.0,
"step": 11380
},
{
"entropy": 5.242721176147461,
"epoch": 0.9509689274974942,
"grad_norm": 1.078125,
"learning_rate": 0.0004915588916098151,
"loss": 4.8652,
"mean_token_accuracy": 0.21642664074897766,
"num_tokens": 20944100.0,
"step": 11385
},
{
"entropy": 5.259535837173462,
"epoch": 0.9513865686602071,
"grad_norm": 1.0546875,
"learning_rate": 0.0004915508119710307,
"loss": 5.0107,
"mean_token_accuracy": 0.20819492787122726,
"num_tokens": 20953733.0,
"step": 11390
},
{
"entropy": 5.335413551330566,
"epoch": 0.9518042098229201,
"grad_norm": 1.0390625,
"learning_rate": 0.0004915427285412636,
"loss": 5.0196,
"mean_token_accuracy": 0.20083789974451066,
"num_tokens": 20962413.0,
"step": 11395
},
{
"entropy": 5.212759780883789,
"epoch": 0.9522218509856332,
"grad_norm": 0.96484375,
"learning_rate": 0.0004915346413206555,
"loss": 4.8859,
"mean_token_accuracy": 0.21609130054712294,
"num_tokens": 20971172.0,
"step": 11400
},
{
"entropy": 5.236460828781128,
"epoch": 0.9526394921483461,
"grad_norm": 1.03125,
"learning_rate": 0.0004915265503093478,
"loss": 4.8955,
"mean_token_accuracy": 0.2137727677822113,
"num_tokens": 20979450.0,
"step": 11405
},
{
"entropy": 5.258274126052856,
"epoch": 0.9530571333110591,
"grad_norm": 1.03125,
"learning_rate": 0.0004915184555074824,
"loss": 4.9657,
"mean_token_accuracy": 0.2026229605078697,
"num_tokens": 20988518.0,
"step": 11410
},
{
"entropy": 5.405281448364258,
"epoch": 0.9534747744737722,
"grad_norm": 0.9921875,
"learning_rate": 0.0004915103569152005,
"loss": 5.1327,
"mean_token_accuracy": 0.19898025393486024,
"num_tokens": 20998721.0,
"step": 11415
},
{
"entropy": 5.330988502502441,
"epoch": 0.9538924156364851,
"grad_norm": 0.99609375,
"learning_rate": 0.0004915022545326444,
"loss": 5.0066,
"mean_token_accuracy": 0.21241801530122756,
"num_tokens": 21008007.0,
"step": 11420
},
{
"entropy": 5.276477861404419,
"epoch": 0.9543100567991981,
"grad_norm": 1.0,
"learning_rate": 0.0004914941483599558,
"loss": 5.0958,
"mean_token_accuracy": 0.20037966519594191,
"num_tokens": 21017199.0,
"step": 11425
},
{
"entropy": 5.183076667785644,
"epoch": 0.9547276979619111,
"grad_norm": 0.9765625,
"learning_rate": 0.0004914860383972763,
"loss": 4.8931,
"mean_token_accuracy": 0.2137138217687607,
"num_tokens": 21025571.0,
"step": 11430
},
{
"entropy": 5.289412260055542,
"epoch": 0.9551453391246241,
"grad_norm": 0.9453125,
"learning_rate": 0.0004914779246447483,
"loss": 4.9792,
"mean_token_accuracy": 0.2093399241566658,
"num_tokens": 21035040.0,
"step": 11435
},
{
"entropy": 5.3084413528442385,
"epoch": 0.9555629802873371,
"grad_norm": 1.0078125,
"learning_rate": 0.0004914698071025136,
"loss": 5.0034,
"mean_token_accuracy": 0.2120872989296913,
"num_tokens": 21044018.0,
"step": 11440
},
{
"entropy": 5.21974196434021,
"epoch": 0.9559806214500501,
"grad_norm": 0.98046875,
"learning_rate": 0.0004914616857707143,
"loss": 4.9043,
"mean_token_accuracy": 0.2146401211619377,
"num_tokens": 21052489.0,
"step": 11445
},
{
"entropy": 5.26521406173706,
"epoch": 0.9563982626127631,
"grad_norm": 0.9765625,
"learning_rate": 0.0004914535606494927,
"loss": 4.9808,
"mean_token_accuracy": 0.21472277343273163,
"num_tokens": 21061153.0,
"step": 11450
},
{
"entropy": 5.219671678543091,
"epoch": 0.9568159037754761,
"grad_norm": 1.078125,
"learning_rate": 0.000491445431738991,
"loss": 4.8714,
"mean_token_accuracy": 0.2129037097096443,
"num_tokens": 21069671.0,
"step": 11455
},
{
"entropy": 5.3590563297271725,
"epoch": 0.9572335449381891,
"grad_norm": 0.98828125,
"learning_rate": 0.0004914372990393515,
"loss": 5.165,
"mean_token_accuracy": 0.19806600660085677,
"num_tokens": 21079339.0,
"step": 11460
},
{
"entropy": 5.279627990722656,
"epoch": 0.957651186100902,
"grad_norm": 0.95703125,
"learning_rate": 0.0004914291625507164,
"loss": 4.8672,
"mean_token_accuracy": 0.2156984180212021,
"num_tokens": 21087868.0,
"step": 11465
},
{
"entropy": 5.279353380203247,
"epoch": 0.9580688272636151,
"grad_norm": 1.1875,
"learning_rate": 0.0004914210222732284,
"loss": 4.964,
"mean_token_accuracy": 0.21369007378816604,
"num_tokens": 21096258.0,
"step": 11470
},
{
"entropy": 5.254007291793823,
"epoch": 0.9584864684263281,
"grad_norm": 0.98828125,
"learning_rate": 0.0004914128782070299,
"loss": 5.0058,
"mean_token_accuracy": 0.20108014941215516,
"num_tokens": 21104504.0,
"step": 11475
},
{
"entropy": 5.296658372879028,
"epoch": 0.958904109589041,
"grad_norm": 0.9921875,
"learning_rate": 0.0004914047303522636,
"loss": 4.9295,
"mean_token_accuracy": 0.21082594692707063,
"num_tokens": 21114402.0,
"step": 11480
},
{
"entropy": 5.259076309204102,
"epoch": 0.9593217507517541,
"grad_norm": 0.87890625,
"learning_rate": 0.0004913965787090718,
"loss": 4.9438,
"mean_token_accuracy": 0.20675256103277206,
"num_tokens": 21123572.0,
"step": 11485
},
{
"entropy": 5.332728099822998,
"epoch": 0.9597393919144671,
"grad_norm": 1.09375,
"learning_rate": 0.0004913884232775974,
"loss": 5.158,
"mean_token_accuracy": 0.1938065245747566,
"num_tokens": 21132385.0,
"step": 11490
},
{
"entropy": 5.206088066101074,
"epoch": 0.96015703307718,
"grad_norm": 1.015625,
"learning_rate": 0.0004913802640579833,
"loss": 4.9119,
"mean_token_accuracy": 0.21479231417179107,
"num_tokens": 21140383.0,
"step": 11495
},
{
"entropy": 5.301038932800293,
"epoch": 0.9605746742398931,
"grad_norm": 0.9609375,
"learning_rate": 0.0004913721010503722,
"loss": 5.0072,
"mean_token_accuracy": 0.20521082133054733,
"num_tokens": 21149793.0,
"step": 11500
},
{
"entropy": 5.288903045654297,
"epoch": 0.9609923154026061,
"grad_norm": 1.078125,
"learning_rate": 0.000491363934254907,
"loss": 4.9917,
"mean_token_accuracy": 0.20400882810354232,
"num_tokens": 21158955.0,
"step": 11505
},
{
"entropy": 5.334828090667725,
"epoch": 0.961409956565319,
"grad_norm": 0.953125,
"learning_rate": 0.0004913557636717307,
"loss": 4.9866,
"mean_token_accuracy": 0.20568390786647797,
"num_tokens": 21167900.0,
"step": 11510
},
{
"entropy": 5.243389225006103,
"epoch": 0.9618275977280321,
"grad_norm": 1.0078125,
"learning_rate": 0.0004913475893009864,
"loss": 4.9336,
"mean_token_accuracy": 0.2120940640568733,
"num_tokens": 21177243.0,
"step": 11515
},
{
"entropy": 5.269658517837525,
"epoch": 0.9622452388907451,
"grad_norm": 1.1015625,
"learning_rate": 0.000491339411142817,
"loss": 5.0043,
"mean_token_accuracy": 0.20148182958364486,
"num_tokens": 21186118.0,
"step": 11520
},
{
"entropy": 5.292251396179199,
"epoch": 0.962662880053458,
"grad_norm": 1.0234375,
"learning_rate": 0.0004913312291973659,
"loss": 5.0377,
"mean_token_accuracy": 0.20238784998655318,
"num_tokens": 21195032.0,
"step": 11525
},
{
"entropy": 5.292068815231323,
"epoch": 0.963080521216171,
"grad_norm": 0.9140625,
"learning_rate": 0.0004913230434647762,
"loss": 4.9921,
"mean_token_accuracy": 0.20474038869142533,
"num_tokens": 21205388.0,
"step": 11530
},
{
"entropy": 5.245506715774536,
"epoch": 0.9634981623788841,
"grad_norm": 1.015625,
"learning_rate": 0.0004913148539451912,
"loss": 4.873,
"mean_token_accuracy": 0.21146152317523956,
"num_tokens": 21215181.0,
"step": 11535
},
{
"entropy": 5.200902986526489,
"epoch": 0.963915803541597,
"grad_norm": 0.984375,
"learning_rate": 0.0004913066606387544,
"loss": 4.8179,
"mean_token_accuracy": 0.21586035341024398,
"num_tokens": 21223668.0,
"step": 11540
},
{
"entropy": 5.238115549087524,
"epoch": 0.96433344470431,
"grad_norm": 1.0546875,
"learning_rate": 0.000491298463545609,
"loss": 4.9285,
"mean_token_accuracy": 0.20830414593219757,
"num_tokens": 21232448.0,
"step": 11545
},
{
"entropy": 5.219215250015258,
"epoch": 0.9647510858670231,
"grad_norm": 1.03125,
"learning_rate": 0.0004912902626658988,
"loss": 4.9242,
"mean_token_accuracy": 0.20706440210342408,
"num_tokens": 21240777.0,
"step": 11550
},
{
"entropy": 5.305804061889648,
"epoch": 0.965168727029736,
"grad_norm": 1.0703125,
"learning_rate": 0.0004912820579997672,
"loss": 5.0553,
"mean_token_accuracy": 0.19662414193153382,
"num_tokens": 21248803.0,
"step": 11555
},
{
"entropy": 5.337806940078735,
"epoch": 0.965586368192449,
"grad_norm": 1.1328125,
"learning_rate": 0.0004912738495473577,
"loss": 5.0097,
"mean_token_accuracy": 0.2070974439382553,
"num_tokens": 21257861.0,
"step": 11560
},
{
"entropy": 5.2985375881195065,
"epoch": 0.9660040093551621,
"grad_norm": 1.0859375,
"learning_rate": 0.0004912656373088142,
"loss": 4.9121,
"mean_token_accuracy": 0.21308131963014604,
"num_tokens": 21267140.0,
"step": 11565
},
{
"entropy": 5.291729545593261,
"epoch": 0.9664216505178751,
"grad_norm": 0.984375,
"learning_rate": 0.0004912574212842804,
"loss": 5.0173,
"mean_token_accuracy": 0.20834615528583528,
"num_tokens": 21276313.0,
"step": 11570
},
{
"entropy": 5.207673692703247,
"epoch": 0.966839291680588,
"grad_norm": 0.94140625,
"learning_rate": 0.0004912492014739001,
"loss": 4.9348,
"mean_token_accuracy": 0.20610727965831757,
"num_tokens": 21286156.0,
"step": 11575
},
{
"entropy": 5.325664281845093,
"epoch": 0.967256932843301,
"grad_norm": 0.953125,
"learning_rate": 0.0004912409778778173,
"loss": 5.0178,
"mean_token_accuracy": 0.20817912369966507,
"num_tokens": 21295420.0,
"step": 11580
},
{
"entropy": 5.382785320281982,
"epoch": 0.9676745740060141,
"grad_norm": 0.8828125,
"learning_rate": 0.0004912327504961758,
"loss": 5.1729,
"mean_token_accuracy": 0.19417010396718978,
"num_tokens": 21306292.0,
"step": 11585
},
{
"entropy": 5.293606424331665,
"epoch": 0.968092215168727,
"grad_norm": 0.94921875,
"learning_rate": 0.0004912245193291198,
"loss": 5.0095,
"mean_token_accuracy": 0.20670066475868226,
"num_tokens": 21315165.0,
"step": 11590
},
{
"entropy": 5.323351430892944,
"epoch": 0.96850985633144,
"grad_norm": 0.94921875,
"learning_rate": 0.0004912162843767932,
"loss": 4.9903,
"mean_token_accuracy": 0.2083704724907875,
"num_tokens": 21324608.0,
"step": 11595
},
{
"entropy": 5.305003643035889,
"epoch": 0.9689274974941531,
"grad_norm": 1.0625,
"learning_rate": 0.0004912080456393404,
"loss": 4.9499,
"mean_token_accuracy": 0.20398494601249695,
"num_tokens": 21333609.0,
"step": 11600
},
{
"entropy": 5.356293773651123,
"epoch": 0.969345138656866,
"grad_norm": 0.98046875,
"learning_rate": 0.0004911998031169056,
"loss": 5.0649,
"mean_token_accuracy": 0.19928715378046036,
"num_tokens": 21344397.0,
"step": 11605
},
{
"entropy": 5.334855413436889,
"epoch": 0.969762779819579,
"grad_norm": 0.95703125,
"learning_rate": 0.0004911915568096328,
"loss": 5.0575,
"mean_token_accuracy": 0.2017696589231491,
"num_tokens": 21354111.0,
"step": 11610
},
{
"entropy": 5.180050945281982,
"epoch": 0.9701804209822921,
"grad_norm": 1.1328125,
"learning_rate": 0.0004911833067176666,
"loss": 4.8567,
"mean_token_accuracy": 0.2104986861348152,
"num_tokens": 21362807.0,
"step": 11615
},
{
"entropy": 5.298491382598877,
"epoch": 0.970598062145005,
"grad_norm": 0.9375,
"learning_rate": 0.0004911750528411515,
"loss": 5.0375,
"mean_token_accuracy": 0.20231910347938536,
"num_tokens": 21372609.0,
"step": 11620
},
{
"entropy": 5.23612494468689,
"epoch": 0.971015703307718,
"grad_norm": 1.03125,
"learning_rate": 0.0004911667951802318,
"loss": 4.9565,
"mean_token_accuracy": 0.2098189800977707,
"num_tokens": 21381182.0,
"step": 11625
},
{
"entropy": 5.273194026947022,
"epoch": 0.9714333444704311,
"grad_norm": 1.0078125,
"learning_rate": 0.0004911585337350524,
"loss": 4.9653,
"mean_token_accuracy": 0.20987071692943574,
"num_tokens": 21389930.0,
"step": 11630
},
{
"entropy": 5.290770435333252,
"epoch": 0.971850985633144,
"grad_norm": 0.94140625,
"learning_rate": 0.0004911502685057573,
"loss": 5.0108,
"mean_token_accuracy": 0.20383824706077575,
"num_tokens": 21399315.0,
"step": 11635
},
{
"entropy": 5.298996543884277,
"epoch": 0.972268626795857,
"grad_norm": 0.9609375,
"learning_rate": 0.0004911419994924918,
"loss": 5.0227,
"mean_token_accuracy": 0.2062831625342369,
"num_tokens": 21408857.0,
"step": 11640
},
{
"entropy": 5.345471000671386,
"epoch": 0.97268626795857,
"grad_norm": 1.0078125,
"learning_rate": 0.0004911337266954004,
"loss": 4.9407,
"mean_token_accuracy": 0.2096747025847435,
"num_tokens": 21417452.0,
"step": 11645
},
{
"entropy": 5.268736553192139,
"epoch": 0.973103909121283,
"grad_norm": 0.96875,
"learning_rate": 0.0004911254501146279,
"loss": 4.9797,
"mean_token_accuracy": 0.21023061573505403,
"num_tokens": 21427315.0,
"step": 11650
},
{
"entropy": 5.285159921646118,
"epoch": 0.973521550283996,
"grad_norm": 0.87109375,
"learning_rate": 0.0004911171697503193,
"loss": 5.0232,
"mean_token_accuracy": 0.20462870746850967,
"num_tokens": 21437080.0,
"step": 11655
},
{
"entropy": 5.271046733856201,
"epoch": 0.973939191446709,
"grad_norm": 1.0625,
"learning_rate": 0.0004911088856026194,
"loss": 4.8927,
"mean_token_accuracy": 0.2224911093711853,
"num_tokens": 21445421.0,
"step": 11660
},
{
"entropy": 5.2641698837280275,
"epoch": 0.974356832609422,
"grad_norm": 1.015625,
"learning_rate": 0.0004911005976716734,
"loss": 4.9076,
"mean_token_accuracy": 0.2139680877327919,
"num_tokens": 21454684.0,
"step": 11665
},
{
"entropy": 5.196329975128174,
"epoch": 0.974774473772135,
"grad_norm": 0.88671875,
"learning_rate": 0.0004910923059576264,
"loss": 4.886,
"mean_token_accuracy": 0.21306994408369065,
"num_tokens": 21464287.0,
"step": 11670
},
{
"entropy": 5.26487889289856,
"epoch": 0.975192114934848,
"grad_norm": 1.046875,
"learning_rate": 0.0004910840104606234,
"loss": 5.0016,
"mean_token_accuracy": 0.20760725289583207,
"num_tokens": 21472835.0,
"step": 11675
},
{
"entropy": 5.2392106533050535,
"epoch": 0.975609756097561,
"grad_norm": 1.03125,
"learning_rate": 0.0004910757111808097,
"loss": 4.8963,
"mean_token_accuracy": 0.21464416682720183,
"num_tokens": 21482327.0,
"step": 11680
},
{
"entropy": 5.290248966217041,
"epoch": 0.976027397260274,
"grad_norm": 1.078125,
"learning_rate": 0.0004910674081183305,
"loss": 5.0029,
"mean_token_accuracy": 0.20064066350460052,
"num_tokens": 21492257.0,
"step": 11685
},
{
"entropy": 5.24731297492981,
"epoch": 0.976445038422987,
"grad_norm": 0.93359375,
"learning_rate": 0.0004910591012733313,
"loss": 4.9979,
"mean_token_accuracy": 0.20853362530469893,
"num_tokens": 21501930.0,
"step": 11690
},
{
"entropy": 5.328313112258911,
"epoch": 0.9768626795857,
"grad_norm": 0.91015625,
"learning_rate": 0.0004910507906459575,
"loss": 4.9765,
"mean_token_accuracy": 0.20580680519342423,
"num_tokens": 21511924.0,
"step": 11695
},
{
"entropy": 5.29951753616333,
"epoch": 0.977280320748413,
"grad_norm": 1.015625,
"learning_rate": 0.0004910424762363545,
"loss": 5.0063,
"mean_token_accuracy": 0.2004636839032173,
"num_tokens": 21522279.0,
"step": 11700
},
{
"entropy": 5.235112571716309,
"epoch": 0.977697961911126,
"grad_norm": 0.93359375,
"learning_rate": 0.000491034158044668,
"loss": 4.9937,
"mean_token_accuracy": 0.20750623643398286,
"num_tokens": 21532016.0,
"step": 11705
},
{
"entropy": 5.299756240844727,
"epoch": 0.9781156030738389,
"grad_norm": 1.0,
"learning_rate": 0.0004910258360710434,
"loss": 5.0489,
"mean_token_accuracy": 0.20230844020843505,
"num_tokens": 21541952.0,
"step": 11710
},
{
"entropy": 5.2382955074310305,
"epoch": 0.978533244236552,
"grad_norm": 1.0,
"learning_rate": 0.0004910175103156266,
"loss": 4.8707,
"mean_token_accuracy": 0.2190218135714531,
"num_tokens": 21550185.0,
"step": 11715
},
{
"entropy": 5.278531551361084,
"epoch": 0.978950885399265,
"grad_norm": 1.09375,
"learning_rate": 0.0004910091807785632,
"loss": 5.0203,
"mean_token_accuracy": 0.20894922465085983,
"num_tokens": 21558891.0,
"step": 11720
},
{
"entropy": 5.266693925857544,
"epoch": 0.9793685265619779,
"grad_norm": 1.0546875,
"learning_rate": 0.0004910008474599991,
"loss": 4.9945,
"mean_token_accuracy": 0.20724486112594603,
"num_tokens": 21568808.0,
"step": 11725
},
{
"entropy": 5.346462535858154,
"epoch": 0.979786167724691,
"grad_norm": 0.9765625,
"learning_rate": 0.0004909925103600801,
"loss": 5.0377,
"mean_token_accuracy": 0.2024528354406357,
"num_tokens": 21578838.0,
"step": 11730
},
{
"entropy": 5.189047336578369,
"epoch": 0.980203808887404,
"grad_norm": 0.9296875,
"learning_rate": 0.0004909841694789523,
"loss": 4.9191,
"mean_token_accuracy": 0.20829323679208755,
"num_tokens": 21588008.0,
"step": 11735
},
{
"entropy": 5.276016759872436,
"epoch": 0.9806214500501169,
"grad_norm": 1.046875,
"learning_rate": 0.0004909758248167616,
"loss": 5.0176,
"mean_token_accuracy": 0.20463661551475526,
"num_tokens": 21597036.0,
"step": 11740
},
{
"entropy": 5.3263507843017575,
"epoch": 0.98103909121283,
"grad_norm": 1.0234375,
"learning_rate": 0.0004909674763736542,
"loss": 4.9783,
"mean_token_accuracy": 0.20885300189256667,
"num_tokens": 21606028.0,
"step": 11745
},
{
"entropy": 5.330385637283325,
"epoch": 0.981456732375543,
"grad_norm": 1.0,
"learning_rate": 0.0004909591241497761,
"loss": 5.0478,
"mean_token_accuracy": 0.2137303337454796,
"num_tokens": 21615377.0,
"step": 11750
},
{
"entropy": 5.3428325176239015,
"epoch": 0.9818743735382559,
"grad_norm": 0.98046875,
"learning_rate": 0.0004909507681452736,
"loss": 5.1325,
"mean_token_accuracy": 0.19879976511001587,
"num_tokens": 21625751.0,
"step": 11755
},
{
"entropy": 5.407093620300293,
"epoch": 0.9822920147009689,
"grad_norm": 1.0,
"learning_rate": 0.0004909424083602929,
"loss": 5.0568,
"mean_token_accuracy": 0.20653210580348969,
"num_tokens": 21634825.0,
"step": 11760
},
{
"entropy": 5.3239446640014645,
"epoch": 0.982709655863682,
"grad_norm": 1.0390625,
"learning_rate": 0.0004909340447949805,
"loss": 4.9283,
"mean_token_accuracy": 0.2110578015446663,
"num_tokens": 21644286.0,
"step": 11765
},
{
"entropy": 5.313165473937988,
"epoch": 0.9831272970263949,
"grad_norm": 1.0859375,
"learning_rate": 0.0004909256774494827,
"loss": 5.0278,
"mean_token_accuracy": 0.20806028991937636,
"num_tokens": 21653384.0,
"step": 11770
},
{
"entropy": 5.27933030128479,
"epoch": 0.9835449381891079,
"grad_norm": 0.98828125,
"learning_rate": 0.000490917306323946,
"loss": 4.938,
"mean_token_accuracy": 0.21124618649482726,
"num_tokens": 21662126.0,
"step": 11775
},
{
"entropy": 5.273025989532471,
"epoch": 0.983962579351821,
"grad_norm": 0.9765625,
"learning_rate": 0.0004909089314185169,
"loss": 4.9637,
"mean_token_accuracy": 0.20500903576612473,
"num_tokens": 21671940.0,
"step": 11780
},
{
"entropy": 5.218639612197876,
"epoch": 0.9843802205145339,
"grad_norm": 0.9765625,
"learning_rate": 0.0004909005527333421,
"loss": 4.8681,
"mean_token_accuracy": 0.2209738999605179,
"num_tokens": 21680928.0,
"step": 11785
},
{
"entropy": 5.1978364944458,
"epoch": 0.9847978616772469,
"grad_norm": 1.03125,
"learning_rate": 0.0004908921702685683,
"loss": 4.9696,
"mean_token_accuracy": 0.20935233682394028,
"num_tokens": 21689712.0,
"step": 11790
},
{
"entropy": 5.2638105869293215,
"epoch": 0.98521550283996,
"grad_norm": 1.03125,
"learning_rate": 0.0004908837840243421,
"loss": 4.9273,
"mean_token_accuracy": 0.21205318421125413,
"num_tokens": 21698325.0,
"step": 11795
},
{
"entropy": 5.245266485214233,
"epoch": 0.9856331440026729,
"grad_norm": 1.0625,
"learning_rate": 0.0004908753940008105,
"loss": 4.9007,
"mean_token_accuracy": 0.21847618818283082,
"num_tokens": 21707607.0,
"step": 11800
},
{
"entropy": 5.19320969581604,
"epoch": 0.9860507851653859,
"grad_norm": 1.0625,
"learning_rate": 0.0004908670001981201,
"loss": 4.8877,
"mean_token_accuracy": 0.21420792490243912,
"num_tokens": 21715887.0,
"step": 11805
},
{
"entropy": 5.263513612747192,
"epoch": 0.986468426328099,
"grad_norm": 0.953125,
"learning_rate": 0.0004908586026164182,
"loss": 4.993,
"mean_token_accuracy": 0.21018393486738204,
"num_tokens": 21724992.0,
"step": 11810
},
{
"entropy": 5.308016538619995,
"epoch": 0.9868860674908119,
"grad_norm": 1.1015625,
"learning_rate": 0.0004908502012558516,
"loss": 4.9956,
"mean_token_accuracy": 0.20734051913022994,
"num_tokens": 21734317.0,
"step": 11815
},
{
"entropy": 5.3289838314056395,
"epoch": 0.9873037086535249,
"grad_norm": 1.03125,
"learning_rate": 0.0004908417961165674,
"loss": 5.0592,
"mean_token_accuracy": 0.20009820759296418,
"num_tokens": 21743014.0,
"step": 11820
},
{
"entropy": 5.336205720901489,
"epoch": 0.9877213498162379,
"grad_norm": 1.171875,
"learning_rate": 0.0004908333871987126,
"loss": 4.9579,
"mean_token_accuracy": 0.20861730426549913,
"num_tokens": 21751058.0,
"step": 11825
},
{
"entropy": 5.320491981506348,
"epoch": 0.9881389909789509,
"grad_norm": 0.953125,
"learning_rate": 0.0004908249745024347,
"loss": 5.0336,
"mean_token_accuracy": 0.20887478142976762,
"num_tokens": 21760549.0,
"step": 11830
},
{
"entropy": 5.286648893356324,
"epoch": 0.9885566321416639,
"grad_norm": 0.96875,
"learning_rate": 0.000490816558027881,
"loss": 4.9573,
"mean_token_accuracy": 0.21629133075475693,
"num_tokens": 21769873.0,
"step": 11835
},
{
"entropy": 5.316697978973389,
"epoch": 0.9889742733043769,
"grad_norm": 1.0625,
"learning_rate": 0.0004908081377751985,
"loss": 5.0395,
"mean_token_accuracy": 0.20972861647605895,
"num_tokens": 21779036.0,
"step": 11840
},
{
"entropy": 5.379835557937622,
"epoch": 0.9893919144670899,
"grad_norm": 1.109375,
"learning_rate": 0.0004907997137445347,
"loss": 5.0702,
"mean_token_accuracy": 0.1999921277165413,
"num_tokens": 21787671.0,
"step": 11845
},
{
"entropy": 5.317583274841309,
"epoch": 0.9898095556298029,
"grad_norm": 0.94921875,
"learning_rate": 0.0004907912859360373,
"loss": 5.0138,
"mean_token_accuracy": 0.20185888707637786,
"num_tokens": 21797269.0,
"step": 11850
},
{
"entropy": 5.351606941223144,
"epoch": 0.9902271967925159,
"grad_norm": 0.9765625,
"learning_rate": 0.0004907828543498536,
"loss": 5.071,
"mean_token_accuracy": 0.2047065183520317,
"num_tokens": 21806738.0,
"step": 11855
},
{
"entropy": 5.285957527160645,
"epoch": 0.9906448379552288,
"grad_norm": 1.0078125,
"learning_rate": 0.0004907744189861313,
"loss": 4.9577,
"mean_token_accuracy": 0.20922023952007293,
"num_tokens": 21815612.0,
"step": 11860
},
{
"entropy": 5.310531759262085,
"epoch": 0.9910624791179419,
"grad_norm": 1.03125,
"learning_rate": 0.000490765979845018,
"loss": 4.9871,
"mean_token_accuracy": 0.20937884598970413,
"num_tokens": 21824222.0,
"step": 11865
},
{
"entropy": 5.312347078323365,
"epoch": 0.9914801202806549,
"grad_norm": 1.0390625,
"learning_rate": 0.0004907575369266617,
"loss": 4.9777,
"mean_token_accuracy": 0.2090429350733757,
"num_tokens": 21833929.0,
"step": 11870
},
{
"entropy": 5.261252069473267,
"epoch": 0.9918977614433678,
"grad_norm": 0.9765625,
"learning_rate": 0.0004907490902312099,
"loss": 5.0294,
"mean_token_accuracy": 0.20546115189790726,
"num_tokens": 21843261.0,
"step": 11875
},
{
"entropy": 5.32122278213501,
"epoch": 0.9923154026060809,
"grad_norm": 1.0234375,
"learning_rate": 0.0004907406397588106,
"loss": 5.0764,
"mean_token_accuracy": 0.20094220340251923,
"num_tokens": 21853004.0,
"step": 11880
},
{
"entropy": 5.298301792144775,
"epoch": 0.9927330437687939,
"grad_norm": 0.9765625,
"learning_rate": 0.0004907321855096117,
"loss": 4.9326,
"mean_token_accuracy": 0.2069837212562561,
"num_tokens": 21862628.0,
"step": 11885
},
{
"entropy": 5.24231448173523,
"epoch": 0.9931506849315068,
"grad_norm": 1.03125,
"learning_rate": 0.0004907237274837613,
"loss": 4.9139,
"mean_token_accuracy": 0.21116461753845214,
"num_tokens": 21872041.0,
"step": 11890
},
{
"entropy": 5.284882068634033,
"epoch": 0.9935683260942199,
"grad_norm": 0.9375,
"learning_rate": 0.0004907152656814072,
"loss": 5.0107,
"mean_token_accuracy": 0.19981574565172194,
"num_tokens": 21881834.0,
"step": 11895
},
{
"entropy": 5.378417682647705,
"epoch": 0.9939859672569329,
"grad_norm": 0.984375,
"learning_rate": 0.0004907068001026978,
"loss": 5.1137,
"mean_token_accuracy": 0.20446072220802308,
"num_tokens": 21892349.0,
"step": 11900
},
{
"entropy": 5.320441770553589,
"epoch": 0.9944036084196458,
"grad_norm": 1.0390625,
"learning_rate": 0.0004906983307477813,
"loss": 4.9756,
"mean_token_accuracy": 0.20961364954710007,
"num_tokens": 21901427.0,
"step": 11905
},
{
"entropy": 5.263491535186768,
"epoch": 0.9948212495823588,
"grad_norm": 1.046875,
"learning_rate": 0.0004906898576168057,
"loss": 4.9201,
"mean_token_accuracy": 0.2104760780930519,
"num_tokens": 21910654.0,
"step": 11910
},
{
"entropy": 5.155521631240845,
"epoch": 0.9952388907450719,
"grad_norm": 1.03125,
"learning_rate": 0.0004906813807099197,
"loss": 4.9585,
"mean_token_accuracy": 0.21791013330221176,
"num_tokens": 21919583.0,
"step": 11915
},
{
"entropy": 5.160713815689087,
"epoch": 0.9956565319077848,
"grad_norm": 1.0,
"learning_rate": 0.0004906729000272714,
"loss": 4.8632,
"mean_token_accuracy": 0.21212699860334397,
"num_tokens": 21927843.0,
"step": 11920
},
{
"entropy": 5.27823281288147,
"epoch": 0.9960741730704978,
"grad_norm": 1.109375,
"learning_rate": 0.0004906644155690095,
"loss": 4.9417,
"mean_token_accuracy": 0.20950577557086944,
"num_tokens": 21937269.0,
"step": 11925
},
{
"entropy": 5.4290077686309814,
"epoch": 0.9964918142332109,
"grad_norm": 0.96484375,
"learning_rate": 0.0004906559273352823,
"loss": 5.1075,
"mean_token_accuracy": 0.200825834274292,
"num_tokens": 21946965.0,
"step": 11930
},
{
"entropy": 5.309491443634033,
"epoch": 0.9969094553959238,
"grad_norm": 1.0078125,
"learning_rate": 0.0004906474353262386,
"loss": 4.9608,
"mean_token_accuracy": 0.20751097202301025,
"num_tokens": 21955957.0,
"step": 11935
},
{
"entropy": 5.229637813568115,
"epoch": 0.9973270965586368,
"grad_norm": 1.0390625,
"learning_rate": 0.0004906389395420269,
"loss": 4.8988,
"mean_token_accuracy": 0.21192772388458253,
"num_tokens": 21965643.0,
"step": 11940
},
{
"entropy": 5.292104911804199,
"epoch": 0.9977447377213499,
"grad_norm": 1.078125,
"learning_rate": 0.000490630439982796,
"loss": 4.9723,
"mean_token_accuracy": 0.20876029133796692,
"num_tokens": 21974789.0,
"step": 11945
},
{
"entropy": 5.214302682876587,
"epoch": 0.9981623788840628,
"grad_norm": 0.9765625,
"learning_rate": 0.0004906219366486947,
"loss": 4.8627,
"mean_token_accuracy": 0.2213547259569168,
"num_tokens": 21983411.0,
"step": 11950
},
{
"entropy": 5.336352014541626,
"epoch": 0.9985800200467758,
"grad_norm": 1.0078125,
"learning_rate": 0.0004906134295398719,
"loss": 4.9779,
"mean_token_accuracy": 0.20911235511302947,
"num_tokens": 21992570.0,
"step": 11955
},
{
"entropy": 5.156963920593261,
"epoch": 0.9989976612094889,
"grad_norm": 1.015625,
"learning_rate": 0.0004906049186564764,
"loss": 4.8509,
"mean_token_accuracy": 0.21873552054166795,
"num_tokens": 22001411.0,
"step": 11960
},
{
"entropy": 5.383324527740479,
"epoch": 0.9994153023722018,
"grad_norm": 1.0625,
"learning_rate": 0.0004905964039986574,
"loss": 5.0899,
"mean_token_accuracy": 0.20052825659513474,
"num_tokens": 22010882.0,
"step": 11965
},
{
"entropy": 5.321116828918457,
"epoch": 0.9998329435349148,
"grad_norm": 0.9921875,
"learning_rate": 0.0004905878855665637,
"loss": 5.0811,
"mean_token_accuracy": 0.2005285605788231,
"num_tokens": 22020885.0,
"step": 11970
},
{
"entropy": 5.287244653701782,
"epoch": 1.0002505846976277,
"grad_norm": 0.921875,
"learning_rate": 0.0004905793633603446,
"loss": 4.8271,
"mean_token_accuracy": 0.21476868093013762,
"num_tokens": 22030278.0,
"step": 11975
},
{
"entropy": 5.343204212188721,
"epoch": 1.0006682258603408,
"grad_norm": 1.0078125,
"learning_rate": 0.0004905708373801492,
"loss": 5.0199,
"mean_token_accuracy": 0.20292422026395798,
"num_tokens": 22039601.0,
"step": 11980
},
{
"entropy": 5.232033777236938,
"epoch": 1.0010858670230538,
"grad_norm": 0.9375,
"learning_rate": 0.0004905623076261269,
"loss": 4.8632,
"mean_token_accuracy": 0.21405914127826692,
"num_tokens": 22049158.0,
"step": 11985
},
{
"entropy": 5.302607440948487,
"epoch": 1.0015035081857668,
"grad_norm": 0.94140625,
"learning_rate": 0.000490553774098427,
"loss": 4.9669,
"mean_token_accuracy": 0.20085557401180268,
"num_tokens": 22059278.0,
"step": 11990
},
{
"entropy": 5.347648191452026,
"epoch": 1.0019211493484799,
"grad_norm": 1.03125,
"learning_rate": 0.0004905452367971986,
"loss": 4.9186,
"mean_token_accuracy": 0.21432092487812043,
"num_tokens": 22068415.0,
"step": 11995
},
{
"entropy": 5.32201247215271,
"epoch": 1.0023387905111927,
"grad_norm": 0.9296875,
"learning_rate": 0.0004905366957225916,
"loss": 4.8952,
"mean_token_accuracy": 0.21055176705121995,
"num_tokens": 22077815.0,
"step": 12000
},
{
"epoch": 1.0023387905111927,
"eval_entropy": 4.98001087759794,
"eval_loss": 4.918286323547363,
"eval_mean_token_accuracy": 0.22144670719452467,
"eval_num_tokens": 22077815.0,
"eval_runtime": 20.9503,
"eval_samples_per_second": 1802.505,
"eval_steps_per_second": 225.343,
"step": 12000
},
{
"entropy": 5.33688063621521,
"epoch": 1.0027564316739057,
"grad_norm": 0.984375,
"learning_rate": 0.0004905281508747552,
"loss": 4.9849,
"mean_token_accuracy": 0.20465521961450578,
"num_tokens": 22087606.0,
"step": 12005
},
{
"entropy": 5.2549535751342775,
"epoch": 1.0031740728366187,
"grad_norm": 1.015625,
"learning_rate": 0.0004905196022538391,
"loss": 4.8058,
"mean_token_accuracy": 0.218599371612072,
"num_tokens": 22096300.0,
"step": 12010
},
{
"entropy": 5.340641975402832,
"epoch": 1.0035917139993318,
"grad_norm": 1.03125,
"learning_rate": 0.000490511049859993,
"loss": 5.0029,
"mean_token_accuracy": 0.2097248375415802,
"num_tokens": 22105750.0,
"step": 12015
},
{
"entropy": 5.352898740768433,
"epoch": 1.0040093551620448,
"grad_norm": 1.0078125,
"learning_rate": 0.0004905024936933664,
"loss": 4.9821,
"mean_token_accuracy": 0.21515518873929979,
"num_tokens": 22115163.0,
"step": 12020
},
{
"entropy": 5.341146802902221,
"epoch": 1.0044269963247578,
"grad_norm": 0.890625,
"learning_rate": 0.0004904939337541093,
"loss": 4.9702,
"mean_token_accuracy": 0.2097815454006195,
"num_tokens": 22125580.0,
"step": 12025
},
{
"entropy": 5.207156324386597,
"epoch": 1.0048446374874707,
"grad_norm": 0.98046875,
"learning_rate": 0.0004904853700423716,
"loss": 4.8449,
"mean_token_accuracy": 0.22272006124258042,
"num_tokens": 22134985.0,
"step": 12030
},
{
"entropy": 5.285246992111206,
"epoch": 1.0052622786501837,
"grad_norm": 1.078125,
"learning_rate": 0.0004904768025583031,
"loss": 4.8584,
"mean_token_accuracy": 0.21392470747232437,
"num_tokens": 22142530.0,
"step": 12035
},
{
"entropy": 5.306007385253906,
"epoch": 1.0056799198128967,
"grad_norm": 1.046875,
"learning_rate": 0.0004904682313020537,
"loss": 4.9508,
"mean_token_accuracy": 0.21486891508102418,
"num_tokens": 22151696.0,
"step": 12040
},
{
"entropy": 5.2540358543396,
"epoch": 1.0060975609756098,
"grad_norm": 1.0390625,
"learning_rate": 0.0004904596562737737,
"loss": 4.8826,
"mean_token_accuracy": 0.2089384064078331,
"num_tokens": 22161286.0,
"step": 12045
},
{
"entropy": 5.265077304840088,
"epoch": 1.0065152021383228,
"grad_norm": 1.078125,
"learning_rate": 0.000490451077473613,
"loss": 4.8485,
"mean_token_accuracy": 0.22072249054908752,
"num_tokens": 22169678.0,
"step": 12050
},
{
"entropy": 5.24460129737854,
"epoch": 1.0069328433010358,
"grad_norm": 0.9921875,
"learning_rate": 0.0004904424949017218,
"loss": 4.9037,
"mean_token_accuracy": 0.22490696907043456,
"num_tokens": 22178618.0,
"step": 12055
},
{
"entropy": 5.217234897613525,
"epoch": 1.0073504844637486,
"grad_norm": 1.09375,
"learning_rate": 0.0004904339085582504,
"loss": 4.8627,
"mean_token_accuracy": 0.2142759695649147,
"num_tokens": 22187574.0,
"step": 12060
},
{
"entropy": 5.37125244140625,
"epoch": 1.0077681256264617,
"grad_norm": 0.98046875,
"learning_rate": 0.0004904253184433491,
"loss": 5.0355,
"mean_token_accuracy": 0.20329798012971878,
"num_tokens": 22198840.0,
"step": 12065
},
{
"entropy": 5.2858569622039795,
"epoch": 1.0081857667891747,
"grad_norm": 1.1953125,
"learning_rate": 0.0004904167245571684,
"loss": 4.8379,
"mean_token_accuracy": 0.21725084483623505,
"num_tokens": 22207166.0,
"step": 12070
},
{
"entropy": 5.233847284317017,
"epoch": 1.0086034079518877,
"grad_norm": 0.99609375,
"learning_rate": 0.0004904081268998586,
"loss": 4.8738,
"mean_token_accuracy": 0.2079276442527771,
"num_tokens": 22216232.0,
"step": 12075
},
{
"entropy": 5.305522346496582,
"epoch": 1.0090210491146008,
"grad_norm": 1.0625,
"learning_rate": 0.0004903995254715702,
"loss": 4.8478,
"mean_token_accuracy": 0.2131372556090355,
"num_tokens": 22224853.0,
"step": 12080
},
{
"entropy": 5.223891639709473,
"epoch": 1.0094386902773138,
"grad_norm": 0.96484375,
"learning_rate": 0.0004903909202724539,
"loss": 4.8423,
"mean_token_accuracy": 0.2171231836080551,
"num_tokens": 22233971.0,
"step": 12085
},
{
"entropy": 5.178820848464966,
"epoch": 1.0098563314400266,
"grad_norm": 1.0546875,
"learning_rate": 0.0004903823113026602,
"loss": 4.7528,
"mean_token_accuracy": 0.22094895094633102,
"num_tokens": 22242548.0,
"step": 12090
},
{
"entropy": 5.137172603607178,
"epoch": 1.0102739726027397,
"grad_norm": 1.0390625,
"learning_rate": 0.00049037369856234,
"loss": 4.7399,
"mean_token_accuracy": 0.22437500804662705,
"num_tokens": 22251086.0,
"step": 12095
},
{
"entropy": 5.286847734451294,
"epoch": 1.0106916137654527,
"grad_norm": 1.0234375,
"learning_rate": 0.0004903650820516438,
"loss": 4.9351,
"mean_token_accuracy": 0.20436829030513765,
"num_tokens": 22260035.0,
"step": 12100
},
{
"entropy": 5.328700113296509,
"epoch": 1.0111092549281657,
"grad_norm": 1.03125,
"learning_rate": 0.0004903564617707227,
"loss": 4.8378,
"mean_token_accuracy": 0.22022415101528167,
"num_tokens": 22269134.0,
"step": 12105
},
{
"entropy": 5.278768873214721,
"epoch": 1.0115268960908788,
"grad_norm": 1.0234375,
"learning_rate": 0.0004903478377197275,
"loss": 4.9389,
"mean_token_accuracy": 0.21955397129058837,
"num_tokens": 22278123.0,
"step": 12110
},
{
"entropy": 5.243806982040406,
"epoch": 1.0119445372535918,
"grad_norm": 0.97265625,
"learning_rate": 0.0004903392098988092,
"loss": 4.936,
"mean_token_accuracy": 0.20569716095924379,
"num_tokens": 22287557.0,
"step": 12115
},
{
"entropy": 5.279406213760376,
"epoch": 1.0123621784163046,
"grad_norm": 1.1171875,
"learning_rate": 0.0004903305783081188,
"loss": 4.8687,
"mean_token_accuracy": 0.2175604671239853,
"num_tokens": 22296705.0,
"step": 12120
},
{
"entropy": 5.217913436889648,
"epoch": 1.0127798195790176,
"grad_norm": 0.91796875,
"learning_rate": 0.0004903219429478075,
"loss": 4.8414,
"mean_token_accuracy": 0.21395462304353713,
"num_tokens": 22305904.0,
"step": 12125
},
{
"entropy": 5.254189729690552,
"epoch": 1.0131974607417307,
"grad_norm": 1.0703125,
"learning_rate": 0.0004903133038180263,
"loss": 4.873,
"mean_token_accuracy": 0.21612954884767532,
"num_tokens": 22315051.0,
"step": 12130
},
{
"entropy": 5.238156318664551,
"epoch": 1.0136151019044437,
"grad_norm": 0.99609375,
"learning_rate": 0.0004903046609189266,
"loss": 4.9191,
"mean_token_accuracy": 0.215847447514534,
"num_tokens": 22324469.0,
"step": 12135
},
{
"entropy": 5.3610094547271725,
"epoch": 1.0140327430671567,
"grad_norm": 1.0,
"learning_rate": 0.0004902960142506597,
"loss": 4.9961,
"mean_token_accuracy": 0.20541031509637833,
"num_tokens": 22334407.0,
"step": 12140
},
{
"entropy": 5.361011648178101,
"epoch": 1.0144503842298698,
"grad_norm": 1.1015625,
"learning_rate": 0.0004902873638133768,
"loss": 4.9919,
"mean_token_accuracy": 0.20484239906072615,
"num_tokens": 22342975.0,
"step": 12145
},
{
"entropy": 5.253829050064087,
"epoch": 1.0148680253925826,
"grad_norm": 0.9609375,
"learning_rate": 0.0004902787096072296,
"loss": 4.8121,
"mean_token_accuracy": 0.21201975047588348,
"num_tokens": 22351649.0,
"step": 12150
},
{
"entropy": 5.2843334674835205,
"epoch": 1.0152856665552956,
"grad_norm": 1.015625,
"learning_rate": 0.0004902700516323693,
"loss": 4.9145,
"mean_token_accuracy": 0.20759780704975128,
"num_tokens": 22361956.0,
"step": 12155
},
{
"entropy": 5.29724702835083,
"epoch": 1.0157033077180087,
"grad_norm": 1.0234375,
"learning_rate": 0.0004902613898889477,
"loss": 4.9163,
"mean_token_accuracy": 0.2092845246195793,
"num_tokens": 22372354.0,
"step": 12160
},
{
"entropy": 5.233258056640625,
"epoch": 1.0161209488807217,
"grad_norm": 1.046875,
"learning_rate": 0.0004902527243771164,
"loss": 4.7951,
"mean_token_accuracy": 0.21819414794445038,
"num_tokens": 22381225.0,
"step": 12165
},
{
"entropy": 5.159318256378174,
"epoch": 1.0165385900434347,
"grad_norm": 1.0234375,
"learning_rate": 0.000490244055097027,
"loss": 4.8041,
"mean_token_accuracy": 0.2210355669260025,
"num_tokens": 22390666.0,
"step": 12170
},
{
"entropy": 5.2269035339355465,
"epoch": 1.0169562312061478,
"grad_norm": 0.97265625,
"learning_rate": 0.0004902353820488314,
"loss": 4.7972,
"mean_token_accuracy": 0.22569137811660767,
"num_tokens": 22399528.0,
"step": 12175
},
{
"entropy": 5.255279111862182,
"epoch": 1.0173738723688606,
"grad_norm": 0.9765625,
"learning_rate": 0.0004902267052326812,
"loss": 4.8168,
"mean_token_accuracy": 0.22620024532079697,
"num_tokens": 22408827.0,
"step": 12180
},
{
"entropy": 5.2124138355255125,
"epoch": 1.0177915135315736,
"grad_norm": 1.0546875,
"learning_rate": 0.0004902180246487285,
"loss": 4.8524,
"mean_token_accuracy": 0.2139216512441635,
"num_tokens": 22418098.0,
"step": 12185
},
{
"entropy": 5.167256307601929,
"epoch": 1.0182091546942866,
"grad_norm": 1.03125,
"learning_rate": 0.0004902093402971254,
"loss": 4.7222,
"mean_token_accuracy": 0.22291651219129563,
"num_tokens": 22427417.0,
"step": 12190
},
{
"entropy": 5.211392068862915,
"epoch": 1.0186267958569997,
"grad_norm": 0.953125,
"learning_rate": 0.0004902006521780236,
"loss": 4.9273,
"mean_token_accuracy": 0.20917229503393173,
"num_tokens": 22436938.0,
"step": 12195
},
{
"entropy": 5.242594385147095,
"epoch": 1.0190444370197127,
"grad_norm": 1.0234375,
"learning_rate": 0.0004901919602915754,
"loss": 4.9612,
"mean_token_accuracy": 0.20073759257793428,
"num_tokens": 22447361.0,
"step": 12200
},
{
"entropy": 5.327710866928101,
"epoch": 1.0194620781824257,
"grad_norm": 0.97265625,
"learning_rate": 0.0004901832646379328,
"loss": 4.9858,
"mean_token_accuracy": 0.20182490944862366,
"num_tokens": 22457431.0,
"step": 12205
},
{
"entropy": 5.311217403411865,
"epoch": 1.0198797193451385,
"grad_norm": 1.015625,
"learning_rate": 0.0004901745652172483,
"loss": 4.9586,
"mean_token_accuracy": 0.20915098041296004,
"num_tokens": 22466219.0,
"step": 12210
},
{
"entropy": 5.311522960662842,
"epoch": 1.0202973605078516,
"grad_norm": 1.03125,
"learning_rate": 0.000490165862029674,
"loss": 4.9733,
"mean_token_accuracy": 0.21020882576704025,
"num_tokens": 22476231.0,
"step": 12215
},
{
"entropy": 5.234994506835937,
"epoch": 1.0207150016705646,
"grad_norm": 1.0625,
"learning_rate": 0.0004901571550753622,
"loss": 4.7647,
"mean_token_accuracy": 0.2253715366125107,
"num_tokens": 22484288.0,
"step": 12220
},
{
"entropy": 5.282567262649536,
"epoch": 1.0211326428332776,
"grad_norm": 1.0,
"learning_rate": 0.0004901484443544655,
"loss": 4.9162,
"mean_token_accuracy": 0.2173297956585884,
"num_tokens": 22493619.0,
"step": 12225
},
{
"entropy": 5.218139600753784,
"epoch": 1.0215502839959907,
"grad_norm": 1.0703125,
"learning_rate": 0.0004901397298671362,
"loss": 4.8207,
"mean_token_accuracy": 0.22035807073116304,
"num_tokens": 22501606.0,
"step": 12230
},
{
"entropy": 5.253662729263306,
"epoch": 1.0219679251587037,
"grad_norm": 1.0390625,
"learning_rate": 0.0004901310116135271,
"loss": 4.825,
"mean_token_accuracy": 0.21622170507907867,
"num_tokens": 22510410.0,
"step": 12235
},
{
"entropy": 5.267537355422974,
"epoch": 1.0223855663214167,
"grad_norm": 1.03125,
"learning_rate": 0.0004901222895937905,
"loss": 4.9272,
"mean_token_accuracy": 0.21936265528202056,
"num_tokens": 22518796.0,
"step": 12240
},
{
"entropy": 5.2464197158813475,
"epoch": 1.0228032074841296,
"grad_norm": 1.1015625,
"learning_rate": 0.0004901135638080796,
"loss": 4.7585,
"mean_token_accuracy": 0.21891705989837645,
"num_tokens": 22527532.0,
"step": 12245
},
{
"entropy": 5.230825281143188,
"epoch": 1.0232208486468426,
"grad_norm": 1.0703125,
"learning_rate": 0.0004901048342565465,
"loss": 4.8421,
"mean_token_accuracy": 0.21363466531038283,
"num_tokens": 22536882.0,
"step": 12250
},
{
"entropy": 5.193243932723999,
"epoch": 1.0236384898095556,
"grad_norm": 0.98828125,
"learning_rate": 0.0004900961009393445,
"loss": 4.7729,
"mean_token_accuracy": 0.2157750442624092,
"num_tokens": 22545549.0,
"step": 12255
},
{
"entropy": 5.2878265380859375,
"epoch": 1.0240561309722687,
"grad_norm": 1.0390625,
"learning_rate": 0.0004900873638566263,
"loss": 4.9876,
"mean_token_accuracy": 0.20700083673000336,
"num_tokens": 22554941.0,
"step": 12260
},
{
"entropy": 5.187461900711059,
"epoch": 1.0244737721349817,
"grad_norm": 0.90234375,
"learning_rate": 0.000490078623008545,
"loss": 4.8724,
"mean_token_accuracy": 0.21458237916231154,
"num_tokens": 22565159.0,
"step": 12265
},
{
"entropy": 5.231040287017822,
"epoch": 1.0248914132976945,
"grad_norm": 1.0234375,
"learning_rate": 0.0004900698783952533,
"loss": 4.8213,
"mean_token_accuracy": 0.21847899705171586,
"num_tokens": 22574184.0,
"step": 12270
},
{
"entropy": 5.235248994827271,
"epoch": 1.0253090544604075,
"grad_norm": 1.046875,
"learning_rate": 0.0004900611300169048,
"loss": 4.9362,
"mean_token_accuracy": 0.20880208164453506,
"num_tokens": 22584062.0,
"step": 12275
},
{
"entropy": 5.292111730575561,
"epoch": 1.0257266956231206,
"grad_norm": 1.046875,
"learning_rate": 0.0004900523778736521,
"loss": 4.8711,
"mean_token_accuracy": 0.21688505560159682,
"num_tokens": 22592845.0,
"step": 12280
},
{
"entropy": 5.271919775009155,
"epoch": 1.0261443367858336,
"grad_norm": 0.92578125,
"learning_rate": 0.0004900436219656488,
"loss": 4.8738,
"mean_token_accuracy": 0.20890514254570008,
"num_tokens": 22602750.0,
"step": 12285
},
{
"entropy": 5.246266937255859,
"epoch": 1.0265619779485466,
"grad_norm": 1.0859375,
"learning_rate": 0.000490034862293048,
"loss": 4.9075,
"mean_token_accuracy": 0.21034188121557235,
"num_tokens": 22611788.0,
"step": 12290
},
{
"entropy": 5.218010091781617,
"epoch": 1.0269796191112597,
"grad_norm": 1.0625,
"learning_rate": 0.0004900260988560031,
"loss": 4.8281,
"mean_token_accuracy": 0.21251376420259477,
"num_tokens": 22620234.0,
"step": 12295
},
{
"entropy": 5.1695475578308105,
"epoch": 1.0273972602739727,
"grad_norm": 1.0546875,
"learning_rate": 0.0004900173316546675,
"loss": 4.7675,
"mean_token_accuracy": 0.2241442859172821,
"num_tokens": 22629155.0,
"step": 12300
},
{
"entropy": 5.3278333187103275,
"epoch": 1.0278149014366855,
"grad_norm": 1.0625,
"learning_rate": 0.0004900085606891947,
"loss": 4.976,
"mean_token_accuracy": 0.2019646182656288,
"num_tokens": 22638647.0,
"step": 12305
},
{
"entropy": 5.290404224395752,
"epoch": 1.0282325425993986,
"grad_norm": 0.9296875,
"learning_rate": 0.0004899997859597382,
"loss": 4.9011,
"mean_token_accuracy": 0.2035590663552284,
"num_tokens": 22648295.0,
"step": 12310
},
{
"entropy": 5.219697666168213,
"epoch": 1.0286501837621116,
"grad_norm": 1.09375,
"learning_rate": 0.0004899910074664518,
"loss": 4.8798,
"mean_token_accuracy": 0.21274126172065735,
"num_tokens": 22657342.0,
"step": 12315
},
{
"entropy": 5.22415566444397,
"epoch": 1.0290678249248246,
"grad_norm": 1.1640625,
"learning_rate": 0.0004899822252094889,
"loss": 4.8393,
"mean_token_accuracy": 0.21675885617733,
"num_tokens": 22665230.0,
"step": 12320
},
{
"entropy": 5.205628871917725,
"epoch": 1.0294854660875377,
"grad_norm": 1.03125,
"learning_rate": 0.0004899734391890034,
"loss": 4.8449,
"mean_token_accuracy": 0.21535724699497222,
"num_tokens": 22674385.0,
"step": 12325
},
{
"entropy": 5.136063575744629,
"epoch": 1.0299031072502507,
"grad_norm": 1.03125,
"learning_rate": 0.0004899646494051491,
"loss": 4.7339,
"mean_token_accuracy": 0.22966232448816298,
"num_tokens": 22683493.0,
"step": 12330
},
{
"entropy": 5.259289407730103,
"epoch": 1.0303207484129635,
"grad_norm": 1.0078125,
"learning_rate": 0.0004899558558580798,
"loss": 4.8144,
"mean_token_accuracy": 0.211348095536232,
"num_tokens": 22691878.0,
"step": 12335
},
{
"entropy": 5.246351528167724,
"epoch": 1.0307383895756765,
"grad_norm": 1.0234375,
"learning_rate": 0.0004899470585479495,
"loss": 4.8311,
"mean_token_accuracy": 0.2138791039586067,
"num_tokens": 22700112.0,
"step": 12340
},
{
"entropy": 5.246161460876465,
"epoch": 1.0311560307383896,
"grad_norm": 1.0078125,
"learning_rate": 0.0004899382574749124,
"loss": 4.8595,
"mean_token_accuracy": 0.21825456470251084,
"num_tokens": 22709218.0,
"step": 12345
},
{
"entropy": 5.252669906616211,
"epoch": 1.0315736719011026,
"grad_norm": 0.97265625,
"learning_rate": 0.0004899294526391221,
"loss": 4.8741,
"mean_token_accuracy": 0.2166745990514755,
"num_tokens": 22718423.0,
"step": 12350
},
{
"entropy": 5.234964227676391,
"epoch": 1.0319913130638156,
"grad_norm": 0.99609375,
"learning_rate": 0.0004899206440407332,
"loss": 4.9459,
"mean_token_accuracy": 0.20568607449531556,
"num_tokens": 22727333.0,
"step": 12355
},
{
"entropy": 5.28084397315979,
"epoch": 1.0324089542265287,
"grad_norm": 0.94140625,
"learning_rate": 0.0004899118316798996,
"loss": 4.8744,
"mean_token_accuracy": 0.211395201086998,
"num_tokens": 22737248.0,
"step": 12360
},
{
"entropy": 5.236141347885132,
"epoch": 1.0328265953892415,
"grad_norm": 0.89453125,
"learning_rate": 0.0004899030155567759,
"loss": 4.8934,
"mean_token_accuracy": 0.21013821959495543,
"num_tokens": 22747515.0,
"step": 12365
},
{
"entropy": 5.288139963150025,
"epoch": 1.0332442365519545,
"grad_norm": 1.03125,
"learning_rate": 0.000489894195671516,
"loss": 4.8928,
"mean_token_accuracy": 0.21627574414014816,
"num_tokens": 22756670.0,
"step": 12370
},
{
"entropy": 5.303247690200806,
"epoch": 1.0336618777146676,
"grad_norm": 0.98828125,
"learning_rate": 0.0004898853720242747,
"loss": 5.0158,
"mean_token_accuracy": 0.20239486992359162,
"num_tokens": 22766332.0,
"step": 12375
},
{
"entropy": 5.274710655212402,
"epoch": 1.0340795188773806,
"grad_norm": 1.0546875,
"learning_rate": 0.0004898765446152062,
"loss": 4.8443,
"mean_token_accuracy": 0.21857706159353257,
"num_tokens": 22774999.0,
"step": 12380
},
{
"entropy": 5.243936491012573,
"epoch": 1.0344971600400936,
"grad_norm": 0.90234375,
"learning_rate": 0.0004898677134444652,
"loss": 4.8879,
"mean_token_accuracy": 0.20787277519702912,
"num_tokens": 22784578.0,
"step": 12385
},
{
"entropy": 5.274090147018432,
"epoch": 1.0349148012028067,
"grad_norm": 1.0,
"learning_rate": 0.0004898588785122063,
"loss": 4.8879,
"mean_token_accuracy": 0.22128744572401046,
"num_tokens": 22794508.0,
"step": 12390
},
{
"entropy": 5.297973012924194,
"epoch": 1.0353324423655195,
"grad_norm": 1.0625,
"learning_rate": 0.0004898500398185841,
"loss": 4.9938,
"mean_token_accuracy": 0.21094824522733688,
"num_tokens": 22804399.0,
"step": 12395
},
{
"entropy": 5.314655447006226,
"epoch": 1.0357500835282325,
"grad_norm": 0.9453125,
"learning_rate": 0.0004898411973637532,
"loss": 4.8979,
"mean_token_accuracy": 0.20966852456331253,
"num_tokens": 22814361.0,
"step": 12400
},
{
"entropy": 5.198535823822022,
"epoch": 1.0361677246909455,
"grad_norm": 1.0234375,
"learning_rate": 0.0004898323511478686,
"loss": 4.8201,
"mean_token_accuracy": 0.22227233946323394,
"num_tokens": 22823780.0,
"step": 12405
},
{
"entropy": 5.199152517318725,
"epoch": 1.0365853658536586,
"grad_norm": 1.0859375,
"learning_rate": 0.0004898235011710851,
"loss": 4.8263,
"mean_token_accuracy": 0.20881333351135253,
"num_tokens": 22832575.0,
"step": 12410
},
{
"entropy": 5.141612768173218,
"epoch": 1.0370030070163716,
"grad_norm": 1.03125,
"learning_rate": 0.0004898146474335577,
"loss": 4.694,
"mean_token_accuracy": 0.22464607059955596,
"num_tokens": 22841822.0,
"step": 12415
},
{
"entropy": 5.326311445236206,
"epoch": 1.0374206481790846,
"grad_norm": 0.98046875,
"learning_rate": 0.0004898057899354412,
"loss": 4.9362,
"mean_token_accuracy": 0.20533109158277513,
"num_tokens": 22851835.0,
"step": 12420
},
{
"entropy": 5.260351467132568,
"epoch": 1.0378382893417974,
"grad_norm": 1.0703125,
"learning_rate": 0.000489796928676891,
"loss": 4.8762,
"mean_token_accuracy": 0.2133066847920418,
"num_tokens": 22861025.0,
"step": 12425
},
{
"entropy": 5.222363424301148,
"epoch": 1.0382559305045105,
"grad_norm": 1.1015625,
"learning_rate": 0.0004897880636580618,
"loss": 4.8958,
"mean_token_accuracy": 0.20730860978364946,
"num_tokens": 22869721.0,
"step": 12430
},
{
"entropy": 5.225875616073608,
"epoch": 1.0386735716672235,
"grad_norm": 1.078125,
"learning_rate": 0.000489779194879109,
"loss": 4.7773,
"mean_token_accuracy": 0.22214785516262053,
"num_tokens": 22878225.0,
"step": 12435
},
{
"entropy": 5.303024053573608,
"epoch": 1.0390912128299366,
"grad_norm": 0.953125,
"learning_rate": 0.000489770322340188,
"loss": 4.9939,
"mean_token_accuracy": 0.20397255569696426,
"num_tokens": 22888598.0,
"step": 12440
},
{
"entropy": 5.286189794540405,
"epoch": 1.0395088539926496,
"grad_norm": 1.0703125,
"learning_rate": 0.0004897614460414539,
"loss": 4.9181,
"mean_token_accuracy": 0.20596157163381576,
"num_tokens": 22897326.0,
"step": 12445
},
{
"entropy": 5.272188997268676,
"epoch": 1.0399264951553626,
"grad_norm": 0.95703125,
"learning_rate": 0.0004897525659830623,
"loss": 4.892,
"mean_token_accuracy": 0.21004635244607925,
"num_tokens": 22906372.0,
"step": 12450
},
{
"entropy": 5.1476233959197994,
"epoch": 1.0403441363180754,
"grad_norm": 0.953125,
"learning_rate": 0.0004897436821651683,
"loss": 4.7945,
"mean_token_accuracy": 0.21882567852735518,
"num_tokens": 22915987.0,
"step": 12455
},
{
"entropy": 5.261592817306519,
"epoch": 1.0407617774807885,
"grad_norm": 0.9765625,
"learning_rate": 0.0004897347945879278,
"loss": 4.9057,
"mean_token_accuracy": 0.21492024064064025,
"num_tokens": 22926118.0,
"step": 12460
},
{
"entropy": 5.267887306213379,
"epoch": 1.0411794186435015,
"grad_norm": 1.03125,
"learning_rate": 0.0004897259032514963,
"loss": 4.8897,
"mean_token_accuracy": 0.2147431567311287,
"num_tokens": 22934813.0,
"step": 12465
},
{
"entropy": 5.290745830535888,
"epoch": 1.0415970598062145,
"grad_norm": 0.9140625,
"learning_rate": 0.0004897170081560292,
"loss": 4.9762,
"mean_token_accuracy": 0.20521267205476762,
"num_tokens": 22944579.0,
"step": 12470
},
{
"entropy": 5.1731892108917235,
"epoch": 1.0420147009689276,
"grad_norm": 1.046875,
"learning_rate": 0.0004897081093016827,
"loss": 4.8056,
"mean_token_accuracy": 0.21805884093046188,
"num_tokens": 22953515.0,
"step": 12475
},
{
"entropy": 5.308432102203369,
"epoch": 1.0424323421316406,
"grad_norm": 1.0234375,
"learning_rate": 0.0004896992066886121,
"loss": 4.9452,
"mean_token_accuracy": 0.20521930903196334,
"num_tokens": 22962599.0,
"step": 12480
},
{
"entropy": 5.295512056350708,
"epoch": 1.0428499832943534,
"grad_norm": 1.0625,
"learning_rate": 0.0004896903003169735,
"loss": 4.9213,
"mean_token_accuracy": 0.21884492933750152,
"num_tokens": 22971717.0,
"step": 12485
},
{
"entropy": 5.12877631187439,
"epoch": 1.0432676244570664,
"grad_norm": 1.0390625,
"learning_rate": 0.0004896813901869229,
"loss": 4.6963,
"mean_token_accuracy": 0.22369424402713775,
"num_tokens": 22980587.0,
"step": 12490
},
{
"entropy": 5.258592700958252,
"epoch": 1.0436852656197795,
"grad_norm": 0.92578125,
"learning_rate": 0.0004896724762986159,
"loss": 4.8918,
"mean_token_accuracy": 0.22188285142183303,
"num_tokens": 22990197.0,
"step": 12495
},
{
"entropy": 5.248133134841919,
"epoch": 1.0441029067824925,
"grad_norm": 1.0078125,
"learning_rate": 0.0004896635586522089,
"loss": 4.9721,
"mean_token_accuracy": 0.20889776945114136,
"num_tokens": 22999670.0,
"step": 12500
},
{
"entropy": 5.325403928756714,
"epoch": 1.0445205479452055,
"grad_norm": 0.98828125,
"learning_rate": 0.0004896546372478581,
"loss": 4.8816,
"mean_token_accuracy": 0.21454250514507295,
"num_tokens": 23008125.0,
"step": 12505
},
{
"entropy": 5.204495811462403,
"epoch": 1.0449381891079186,
"grad_norm": 1.0546875,
"learning_rate": 0.0004896457120857192,
"loss": 4.741,
"mean_token_accuracy": 0.22184273302555085,
"num_tokens": 23017469.0,
"step": 12510
},
{
"entropy": 5.154119539260864,
"epoch": 1.0453558302706314,
"grad_norm": 1.125,
"learning_rate": 0.000489636783165949,
"loss": 4.9038,
"mean_token_accuracy": 0.210226671397686,
"num_tokens": 23026468.0,
"step": 12515
},
{
"entropy": 5.30165114402771,
"epoch": 1.0457734714333444,
"grad_norm": 1.109375,
"learning_rate": 0.0004896278504887033,
"loss": 4.919,
"mean_token_accuracy": 0.21622417718172074,
"num_tokens": 23034944.0,
"step": 12520
},
{
"entropy": 5.202324628829956,
"epoch": 1.0461911125960575,
"grad_norm": 0.94921875,
"learning_rate": 0.000489618914054139,
"loss": 4.7567,
"mean_token_accuracy": 0.22307320982217788,
"num_tokens": 23043834.0,
"step": 12525
},
{
"entropy": 5.219703674316406,
"epoch": 1.0466087537587705,
"grad_norm": 0.9453125,
"learning_rate": 0.000489609973862412,
"loss": 4.8593,
"mean_token_accuracy": 0.2170213595032692,
"num_tokens": 23052542.0,
"step": 12530
},
{
"entropy": 5.323325872421265,
"epoch": 1.0470263949214835,
"grad_norm": 1.0078125,
"learning_rate": 0.0004896010299136793,
"loss": 4.9,
"mean_token_accuracy": 0.2150929942727089,
"num_tokens": 23061438.0,
"step": 12535
},
{
"entropy": 5.342754364013672,
"epoch": 1.0474440360841966,
"grad_norm": 0.91796875,
"learning_rate": 0.0004895920822080972,
"loss": 5.007,
"mean_token_accuracy": 0.2067944422364235,
"num_tokens": 23071777.0,
"step": 12540
},
{
"entropy": 5.310712909698486,
"epoch": 1.0478616772469094,
"grad_norm": 1.0546875,
"learning_rate": 0.0004895831307458224,
"loss": 4.8686,
"mean_token_accuracy": 0.21585914641618728,
"num_tokens": 23080889.0,
"step": 12545
},
{
"entropy": 5.240080785751343,
"epoch": 1.0482793184096224,
"grad_norm": 1.09375,
"learning_rate": 0.0004895741755270115,
"loss": 4.7828,
"mean_token_accuracy": 0.21873192191123964,
"num_tokens": 23089737.0,
"step": 12550
},
{
"entropy": 5.218736934661865,
"epoch": 1.0486969595723354,
"grad_norm": 1.1015625,
"learning_rate": 0.0004895652165518215,
"loss": 4.8452,
"mean_token_accuracy": 0.21811958849430085,
"num_tokens": 23098861.0,
"step": 12555
},
{
"entropy": 5.239730262756348,
"epoch": 1.0491146007350485,
"grad_norm": 1.015625,
"learning_rate": 0.0004895562538204091,
"loss": 4.9576,
"mean_token_accuracy": 0.20898807644844056,
"num_tokens": 23107798.0,
"step": 12560
},
{
"entropy": 5.270431709289551,
"epoch": 1.0495322418977615,
"grad_norm": 1.046875,
"learning_rate": 0.0004895472873329311,
"loss": 4.7646,
"mean_token_accuracy": 0.22054098248481752,
"num_tokens": 23116698.0,
"step": 12565
},
{
"entropy": 5.308022546768188,
"epoch": 1.0499498830604745,
"grad_norm": 0.99609375,
"learning_rate": 0.0004895383170895447,
"loss": 4.9468,
"mean_token_accuracy": 0.20543525665998458,
"num_tokens": 23125880.0,
"step": 12570
},
{
"entropy": 5.329310703277588,
"epoch": 1.0503675242231874,
"grad_norm": 0.96875,
"learning_rate": 0.0004895293430904067,
"loss": 4.9872,
"mean_token_accuracy": 0.21102734804153442,
"num_tokens": 23135633.0,
"step": 12575
},
{
"entropy": 5.177391242980957,
"epoch": 1.0507851653859004,
"grad_norm": 0.9140625,
"learning_rate": 0.0004895203653356745,
"loss": 4.8079,
"mean_token_accuracy": 0.21286563575267792,
"num_tokens": 23144952.0,
"step": 12580
},
{
"entropy": 5.206628942489624,
"epoch": 1.0512028065486134,
"grad_norm": 1.0,
"learning_rate": 0.0004895113838255049,
"loss": 4.8386,
"mean_token_accuracy": 0.21769546568393708,
"num_tokens": 23154128.0,
"step": 12585
},
{
"entropy": 5.196801328659058,
"epoch": 1.0516204477113265,
"grad_norm": 1.0703125,
"learning_rate": 0.0004895023985600555,
"loss": 4.7551,
"mean_token_accuracy": 0.2243789404630661,
"num_tokens": 23162579.0,
"step": 12590
},
{
"entropy": 5.236465883255005,
"epoch": 1.0520380888740395,
"grad_norm": 0.96484375,
"learning_rate": 0.0004894934095394832,
"loss": 4.8846,
"mean_token_accuracy": 0.2154398277401924,
"num_tokens": 23171889.0,
"step": 12595
},
{
"entropy": 5.1357017993927006,
"epoch": 1.0524557300367525,
"grad_norm": 1.0078125,
"learning_rate": 0.0004894844167639458,
"loss": 4.7473,
"mean_token_accuracy": 0.22371862083673477,
"num_tokens": 23180876.0,
"step": 12600
},
{
"entropy": 5.200278997421265,
"epoch": 1.0528733711994653,
"grad_norm": 1.0390625,
"learning_rate": 0.0004894754202336004,
"loss": 4.7963,
"mean_token_accuracy": 0.22279708832502365,
"num_tokens": 23189567.0,
"step": 12605
},
{
"entropy": 5.340875959396362,
"epoch": 1.0532910123621784,
"grad_norm": 1.1640625,
"learning_rate": 0.0004894664199486047,
"loss": 4.8088,
"mean_token_accuracy": 0.2191482588648796,
"num_tokens": 23198142.0,
"step": 12610
},
{
"entropy": 5.271886682510376,
"epoch": 1.0537086535248914,
"grad_norm": 1.1171875,
"learning_rate": 0.0004894574159091161,
"loss": 4.9357,
"mean_token_accuracy": 0.205353082716465,
"num_tokens": 23207143.0,
"step": 12615
},
{
"entropy": 5.265033292770386,
"epoch": 1.0541262946876044,
"grad_norm": 1.046875,
"learning_rate": 0.0004894484081152922,
"loss": 4.8257,
"mean_token_accuracy": 0.2127431333065033,
"num_tokens": 23215637.0,
"step": 12620
},
{
"entropy": 5.290119171142578,
"epoch": 1.0545439358503175,
"grad_norm": 1.0078125,
"learning_rate": 0.0004894393965672909,
"loss": 4.8784,
"mean_token_accuracy": 0.20768315345048904,
"num_tokens": 23224402.0,
"step": 12625
},
{
"entropy": 5.412623262405395,
"epoch": 1.0549615770130305,
"grad_norm": 1.1484375,
"learning_rate": 0.0004894303812652698,
"loss": 5.0945,
"mean_token_accuracy": 0.20403465628623962,
"num_tokens": 23232669.0,
"step": 12630
},
{
"entropy": 5.159027528762818,
"epoch": 1.0553792181757433,
"grad_norm": 0.95703125,
"learning_rate": 0.0004894213622093867,
"loss": 4.8335,
"mean_token_accuracy": 0.2119513586163521,
"num_tokens": 23242204.0,
"step": 12635
},
{
"entropy": 5.239485883712769,
"epoch": 1.0557968593384564,
"grad_norm": 0.953125,
"learning_rate": 0.0004894123393997997,
"loss": 4.8846,
"mean_token_accuracy": 0.21236128211021424,
"num_tokens": 23250765.0,
"step": 12640
},
{
"entropy": 5.2181117057800295,
"epoch": 1.0562145005011694,
"grad_norm": 0.9921875,
"learning_rate": 0.0004894033128366665,
"loss": 4.8,
"mean_token_accuracy": 0.213079571723938,
"num_tokens": 23259986.0,
"step": 12645
},
{
"entropy": 5.206310224533081,
"epoch": 1.0566321416638824,
"grad_norm": 0.9375,
"learning_rate": 0.0004893942825201453,
"loss": 4.7459,
"mean_token_accuracy": 0.22099792659282685,
"num_tokens": 23268599.0,
"step": 12650
},
{
"entropy": 5.18853988647461,
"epoch": 1.0570497828265955,
"grad_norm": 1.03125,
"learning_rate": 0.000489385248450394,
"loss": 4.8011,
"mean_token_accuracy": 0.221826933324337,
"num_tokens": 23277509.0,
"step": 12655
},
{
"entropy": 5.238329982757568,
"epoch": 1.0574674239893085,
"grad_norm": 1.015625,
"learning_rate": 0.000489376210627571,
"loss": 4.8245,
"mean_token_accuracy": 0.21568035781383516,
"num_tokens": 23286673.0,
"step": 12660
},
{
"entropy": 5.272193384170532,
"epoch": 1.0578850651520213,
"grad_norm": 1.0234375,
"learning_rate": 0.0004893671690518343,
"loss": 4.8029,
"mean_token_accuracy": 0.22597654163837433,
"num_tokens": 23295775.0,
"step": 12665
},
{
"entropy": 5.1898030757904055,
"epoch": 1.0583027063147343,
"grad_norm": 1.046875,
"learning_rate": 0.0004893581237233423,
"loss": 4.8371,
"mean_token_accuracy": 0.2201521575450897,
"num_tokens": 23305321.0,
"step": 12670
},
{
"entropy": 5.2493531703948975,
"epoch": 1.0587203474774474,
"grad_norm": 1.125,
"learning_rate": 0.0004893490746422533,
"loss": 4.8301,
"mean_token_accuracy": 0.21252137869596482,
"num_tokens": 23313798.0,
"step": 12675
},
{
"entropy": 5.161060190200805,
"epoch": 1.0591379886401604,
"grad_norm": 1.0546875,
"learning_rate": 0.0004893400218087258,
"loss": 4.7136,
"mean_token_accuracy": 0.22940206676721572,
"num_tokens": 23322308.0,
"step": 12680
},
{
"entropy": 5.243034315109253,
"epoch": 1.0595556298028734,
"grad_norm": 1.1015625,
"learning_rate": 0.0004893309652229183,
"loss": 4.9387,
"mean_token_accuracy": 0.20927054435014725,
"num_tokens": 23331311.0,
"step": 12685
},
{
"entropy": 5.332195377349853,
"epoch": 1.0599732709655865,
"grad_norm": 1.0,
"learning_rate": 0.0004893219048849891,
"loss": 4.9899,
"mean_token_accuracy": 0.2072555363178253,
"num_tokens": 23340655.0,
"step": 12690
},
{
"entropy": 5.3745348930358885,
"epoch": 1.0603909121282993,
"grad_norm": 0.9765625,
"learning_rate": 0.0004893128407950971,
"loss": 5.0385,
"mean_token_accuracy": 0.2032894805073738,
"num_tokens": 23350999.0,
"step": 12695
},
{
"entropy": 5.242874717712402,
"epoch": 1.0608085532910123,
"grad_norm": 1.078125,
"learning_rate": 0.0004893037729534007,
"loss": 4.8493,
"mean_token_accuracy": 0.21334214508533478,
"num_tokens": 23359311.0,
"step": 12700
},
{
"entropy": 5.260963010787964,
"epoch": 1.0612261944537253,
"grad_norm": 1.0546875,
"learning_rate": 0.0004892947013600589,
"loss": 4.8875,
"mean_token_accuracy": 0.21278003007173538,
"num_tokens": 23367925.0,
"step": 12705
},
{
"entropy": 5.285361051559448,
"epoch": 1.0616438356164384,
"grad_norm": 1.0859375,
"learning_rate": 0.0004892856260152304,
"loss": 4.9906,
"mean_token_accuracy": 0.20609406679868697,
"num_tokens": 23377159.0,
"step": 12710
},
{
"entropy": 5.433383989334106,
"epoch": 1.0620614767791514,
"grad_norm": 1.0234375,
"learning_rate": 0.000489276546919074,
"loss": 5.0354,
"mean_token_accuracy": 0.203872974216938,
"num_tokens": 23387699.0,
"step": 12715
},
{
"entropy": 5.326310729980468,
"epoch": 1.0624791179418644,
"grad_norm": 1.2421875,
"learning_rate": 0.0004892674640717488,
"loss": 4.9358,
"mean_token_accuracy": 0.20843103975057603,
"num_tokens": 23397520.0,
"step": 12720
},
{
"entropy": 5.197593927383423,
"epoch": 1.0628967591045773,
"grad_norm": 1.046875,
"learning_rate": 0.0004892583774734138,
"loss": 4.8538,
"mean_token_accuracy": 0.21133932173252107,
"num_tokens": 23406853.0,
"step": 12725
},
{
"entropy": 5.201407861709595,
"epoch": 1.0633144002672903,
"grad_norm": 1.046875,
"learning_rate": 0.0004892492871242278,
"loss": 4.8057,
"mean_token_accuracy": 0.21800081729888915,
"num_tokens": 23416235.0,
"step": 12730
},
{
"entropy": 5.266881322860717,
"epoch": 1.0637320414300033,
"grad_norm": 1.03125,
"learning_rate": 0.0004892401930243504,
"loss": 4.8743,
"mean_token_accuracy": 0.21012381017208098,
"num_tokens": 23425347.0,
"step": 12735
},
{
"entropy": 5.3433300971984865,
"epoch": 1.0641496825927164,
"grad_norm": 0.984375,
"learning_rate": 0.0004892310951739404,
"loss": 5.0425,
"mean_token_accuracy": 0.20780341923236847,
"num_tokens": 23434905.0,
"step": 12740
},
{
"entropy": 5.325739431381225,
"epoch": 1.0645673237554294,
"grad_norm": 1.0,
"learning_rate": 0.0004892219935731571,
"loss": 4.8539,
"mean_token_accuracy": 0.21316174417734146,
"num_tokens": 23444495.0,
"step": 12745
},
{
"entropy": 5.234795188903808,
"epoch": 1.0649849649181424,
"grad_norm": 1.109375,
"learning_rate": 0.00048921288822216,
"loss": 4.8296,
"mean_token_accuracy": 0.21455989480018617,
"num_tokens": 23452224.0,
"step": 12750
},
{
"entropy": 5.24503345489502,
"epoch": 1.0654026060808552,
"grad_norm": 1.0390625,
"learning_rate": 0.0004892037791211086,
"loss": 4.9694,
"mean_token_accuracy": 0.2096216395497322,
"num_tokens": 23461987.0,
"step": 12755
},
{
"entropy": 5.12617859840393,
"epoch": 1.0658202472435683,
"grad_norm": 1.078125,
"learning_rate": 0.0004891946662701621,
"loss": 4.8025,
"mean_token_accuracy": 0.22298409044742584,
"num_tokens": 23470645.0,
"step": 12760
},
{
"entropy": 5.283561658859253,
"epoch": 1.0662378884062813,
"grad_norm": 1.0390625,
"learning_rate": 0.0004891855496694801,
"loss": 4.969,
"mean_token_accuracy": 0.20727645605802536,
"num_tokens": 23480169.0,
"step": 12765
},
{
"entropy": 5.3262427806854244,
"epoch": 1.0666555295689943,
"grad_norm": 0.97265625,
"learning_rate": 0.0004891764293192222,
"loss": 4.9041,
"mean_token_accuracy": 0.20949862152338028,
"num_tokens": 23488630.0,
"step": 12770
},
{
"entropy": 5.247120904922485,
"epoch": 1.0670731707317074,
"grad_norm": 1.046875,
"learning_rate": 0.0004891673052195482,
"loss": 4.7783,
"mean_token_accuracy": 0.21573328971862793,
"num_tokens": 23497800.0,
"step": 12775
},
{
"entropy": 5.256953620910645,
"epoch": 1.0674908118944204,
"grad_norm": 0.94140625,
"learning_rate": 0.0004891581773706177,
"loss": 4.9636,
"mean_token_accuracy": 0.20531099736690522,
"num_tokens": 23507400.0,
"step": 12780
},
{
"entropy": 5.323882818222046,
"epoch": 1.0679084530571332,
"grad_norm": 0.9765625,
"learning_rate": 0.0004891490457725906,
"loss": 4.9736,
"mean_token_accuracy": 0.19991584420204161,
"num_tokens": 23516441.0,
"step": 12785
},
{
"entropy": 5.272881507873535,
"epoch": 1.0683260942198463,
"grad_norm": 1.015625,
"learning_rate": 0.0004891399104256266,
"loss": 4.8408,
"mean_token_accuracy": 0.2110961526632309,
"num_tokens": 23525066.0,
"step": 12790
},
{
"entropy": 5.164448976516724,
"epoch": 1.0687437353825593,
"grad_norm": 0.98828125,
"learning_rate": 0.0004891307713298857,
"loss": 4.7682,
"mean_token_accuracy": 0.22332222163677215,
"num_tokens": 23533892.0,
"step": 12795
},
{
"entropy": 5.234396314620971,
"epoch": 1.0691613765452723,
"grad_norm": 1.0234375,
"learning_rate": 0.000489121628485528,
"loss": 4.921,
"mean_token_accuracy": 0.20590850263833999,
"num_tokens": 23542680.0,
"step": 12800
},
{
"entropy": 5.208740949630737,
"epoch": 1.0695790177079854,
"grad_norm": 1.0703125,
"learning_rate": 0.0004891124818927133,
"loss": 4.7439,
"mean_token_accuracy": 0.22733963876962662,
"num_tokens": 23550573.0,
"step": 12805
},
{
"entropy": 5.14765510559082,
"epoch": 1.0699966588706984,
"grad_norm": 0.9765625,
"learning_rate": 0.0004891033315516019,
"loss": 4.871,
"mean_token_accuracy": 0.20837533175945283,
"num_tokens": 23561281.0,
"step": 12810
},
{
"entropy": 5.175348997116089,
"epoch": 1.0704143000334112,
"grad_norm": 0.9296875,
"learning_rate": 0.0004890941774623541,
"loss": 4.7753,
"mean_token_accuracy": 0.22325708270072936,
"num_tokens": 23569846.0,
"step": 12815
},
{
"entropy": 5.246774053573608,
"epoch": 1.0708319411961242,
"grad_norm": 1.1171875,
"learning_rate": 0.0004890850196251299,
"loss": 4.775,
"mean_token_accuracy": 0.22480323761701584,
"num_tokens": 23578145.0,
"step": 12820
},
{
"entropy": 5.244913959503174,
"epoch": 1.0712495823588373,
"grad_norm": 1.0,
"learning_rate": 0.0004890758580400898,
"loss": 4.8657,
"mean_token_accuracy": 0.21745838969945908,
"num_tokens": 23587494.0,
"step": 12825
},
{
"entropy": 5.301851034164429,
"epoch": 1.0716672235215503,
"grad_norm": 0.9609375,
"learning_rate": 0.0004890666927073941,
"loss": 4.9667,
"mean_token_accuracy": 0.20601084977388381,
"num_tokens": 23596599.0,
"step": 12830
},
{
"entropy": 5.231004095077514,
"epoch": 1.0720848646842633,
"grad_norm": 1.015625,
"learning_rate": 0.0004890575236272032,
"loss": 4.8993,
"mean_token_accuracy": 0.21948205232620238,
"num_tokens": 23606170.0,
"step": 12835
},
{
"entropy": 5.178211688995361,
"epoch": 1.0725025058469764,
"grad_norm": 1.015625,
"learning_rate": 0.0004890483507996778,
"loss": 4.7243,
"mean_token_accuracy": 0.22773199379444123,
"num_tokens": 23614558.0,
"step": 12840
},
{
"entropy": 5.276197814941407,
"epoch": 1.0729201470096892,
"grad_norm": 1.0390625,
"learning_rate": 0.0004890391742249783,
"loss": 4.9137,
"mean_token_accuracy": 0.20629681050777435,
"num_tokens": 23624125.0,
"step": 12845
},
{
"entropy": 5.331362676620484,
"epoch": 1.0733377881724022,
"grad_norm": 1.015625,
"learning_rate": 0.0004890299939032654,
"loss": 4.8673,
"mean_token_accuracy": 0.21477816253900528,
"num_tokens": 23633025.0,
"step": 12850
},
{
"entropy": 5.195815515518189,
"epoch": 1.0737554293351153,
"grad_norm": 1.0234375,
"learning_rate": 0.0004890208098346999,
"loss": 4.8393,
"mean_token_accuracy": 0.21111361682415009,
"num_tokens": 23643387.0,
"step": 12855
},
{
"entropy": 5.27276873588562,
"epoch": 1.0741730704978283,
"grad_norm": 0.953125,
"learning_rate": 0.0004890116220194425,
"loss": 4.9075,
"mean_token_accuracy": 0.2139476865530014,
"num_tokens": 23653486.0,
"step": 12860
},
{
"entropy": 5.220895195007325,
"epoch": 1.0745907116605413,
"grad_norm": 0.9921875,
"learning_rate": 0.000489002430457654,
"loss": 4.8141,
"mean_token_accuracy": 0.21765146255493165,
"num_tokens": 23663483.0,
"step": 12865
},
{
"entropy": 5.154907131195069,
"epoch": 1.0750083528232544,
"grad_norm": 1.0859375,
"learning_rate": 0.0004889932351494954,
"loss": 4.8482,
"mean_token_accuracy": 0.22139590978622437,
"num_tokens": 23672453.0,
"step": 12870
},
{
"entropy": 5.225982189178467,
"epoch": 1.0754259939859672,
"grad_norm": 1.0078125,
"learning_rate": 0.0004889840360951277,
"loss": 4.8494,
"mean_token_accuracy": 0.21532514393329621,
"num_tokens": 23681351.0,
"step": 12875
},
{
"entropy": 5.282834005355835,
"epoch": 1.0758436351486802,
"grad_norm": 0.91796875,
"learning_rate": 0.0004889748332947118,
"loss": 4.8396,
"mean_token_accuracy": 0.2157393217086792,
"num_tokens": 23689581.0,
"step": 12880
},
{
"entropy": 5.29621844291687,
"epoch": 1.0762612763113932,
"grad_norm": 0.9765625,
"learning_rate": 0.0004889656267484089,
"loss": 4.9479,
"mean_token_accuracy": 0.2063759222626686,
"num_tokens": 23699415.0,
"step": 12885
},
{
"entropy": 5.263041591644287,
"epoch": 1.0766789174741063,
"grad_norm": 1.0390625,
"learning_rate": 0.00048895641645638,
"loss": 4.8733,
"mean_token_accuracy": 0.22042959928512573,
"num_tokens": 23708933.0,
"step": 12890
},
{
"entropy": 5.2040527820587155,
"epoch": 1.0770965586368193,
"grad_norm": 1.015625,
"learning_rate": 0.0004889472024187867,
"loss": 4.9001,
"mean_token_accuracy": 0.20418200343847276,
"num_tokens": 23718423.0,
"step": 12895
},
{
"entropy": 5.266914939880371,
"epoch": 1.0775141997995323,
"grad_norm": 1.0390625,
"learning_rate": 0.00048893798463579,
"loss": 4.8814,
"mean_token_accuracy": 0.21385739594697953,
"num_tokens": 23728131.0,
"step": 12900
},
{
"entropy": 5.255648756027222,
"epoch": 1.0779318409622451,
"grad_norm": 1.015625,
"learning_rate": 0.0004889287631075513,
"loss": 4.8798,
"mean_token_accuracy": 0.21327996402978897,
"num_tokens": 23737378.0,
"step": 12905
},
{
"entropy": 5.393601131439209,
"epoch": 1.0783494821249582,
"grad_norm": 1.0078125,
"learning_rate": 0.0004889195378342322,
"loss": 4.9631,
"mean_token_accuracy": 0.20549146980047225,
"num_tokens": 23747156.0,
"step": 12910
},
{
"entropy": 5.313196468353271,
"epoch": 1.0787671232876712,
"grad_norm": 0.90234375,
"learning_rate": 0.0004889103088159941,
"loss": 4.9167,
"mean_token_accuracy": 0.20896490216255187,
"num_tokens": 23757090.0,
"step": 12915
},
{
"entropy": 5.211637783050537,
"epoch": 1.0791847644503842,
"grad_norm": 0.9921875,
"learning_rate": 0.0004889010760529986,
"loss": 4.901,
"mean_token_accuracy": 0.21270398050546646,
"num_tokens": 23766679.0,
"step": 12920
},
{
"entropy": 5.214770030975342,
"epoch": 1.0796024056130973,
"grad_norm": 1.15625,
"learning_rate": 0.0004888918395454071,
"loss": 4.7799,
"mean_token_accuracy": 0.23126090615987777,
"num_tokens": 23775109.0,
"step": 12925
},
{
"entropy": 5.235091638565064,
"epoch": 1.0800200467758103,
"grad_norm": 0.953125,
"learning_rate": 0.0004888825992933817,
"loss": 4.8665,
"mean_token_accuracy": 0.22285349369049073,
"num_tokens": 23784698.0,
"step": 12930
},
{
"entropy": 5.162795639038086,
"epoch": 1.0804376879385231,
"grad_norm": 1.0703125,
"learning_rate": 0.0004888733552970839,
"loss": 4.783,
"mean_token_accuracy": 0.21931341737508775,
"num_tokens": 23793590.0,
"step": 12935
},
{
"entropy": 5.234830904006958,
"epoch": 1.0808553291012362,
"grad_norm": 1.0546875,
"learning_rate": 0.0004888641075566755,
"loss": 4.8153,
"mean_token_accuracy": 0.21681467294692994,
"num_tokens": 23802421.0,
"step": 12940
},
{
"entropy": 5.305587816238403,
"epoch": 1.0812729702639492,
"grad_norm": 1.0078125,
"learning_rate": 0.0004888548560723184,
"loss": 4.9384,
"mean_token_accuracy": 0.20567968487739563,
"num_tokens": 23811610.0,
"step": 12945
},
{
"entropy": 5.358579158782959,
"epoch": 1.0816906114266622,
"grad_norm": 0.93359375,
"learning_rate": 0.0004888456008441748,
"loss": 4.9246,
"mean_token_accuracy": 0.21314588636159898,
"num_tokens": 23821197.0,
"step": 12950
},
{
"entropy": 5.305755949020385,
"epoch": 1.0821082525893753,
"grad_norm": 1.0390625,
"learning_rate": 0.0004888363418724064,
"loss": 4.8976,
"mean_token_accuracy": 0.21083300411701203,
"num_tokens": 23829757.0,
"step": 12955
},
{
"entropy": 5.256096267700196,
"epoch": 1.0825258937520883,
"grad_norm": 1.0078125,
"learning_rate": 0.0004888270791571756,
"loss": 4.8789,
"mean_token_accuracy": 0.20821415036916732,
"num_tokens": 23839242.0,
"step": 12960
},
{
"entropy": 5.2185557842254635,
"epoch": 1.082943534914801,
"grad_norm": 1.046875,
"learning_rate": 0.0004888178126986444,
"loss": 4.8062,
"mean_token_accuracy": 0.2222756028175354,
"num_tokens": 23848193.0,
"step": 12965
},
{
"entropy": 5.206289672851563,
"epoch": 1.0833611760775141,
"grad_norm": 1.0,
"learning_rate": 0.0004888085424969748,
"loss": 4.8161,
"mean_token_accuracy": 0.2184004858136177,
"num_tokens": 23857623.0,
"step": 12970
},
{
"entropy": 5.274158000946045,
"epoch": 1.0837788172402272,
"grad_norm": 1.140625,
"learning_rate": 0.0004887992685523295,
"loss": 4.9401,
"mean_token_accuracy": 0.20634313374757768,
"num_tokens": 23866424.0,
"step": 12975
},
{
"entropy": 5.266625165939331,
"epoch": 1.0841964584029402,
"grad_norm": 1.109375,
"learning_rate": 0.0004887899908648706,
"loss": 4.9671,
"mean_token_accuracy": 0.20892710983753204,
"num_tokens": 23874855.0,
"step": 12980
},
{
"entropy": 5.29748010635376,
"epoch": 1.0846140995656532,
"grad_norm": 0.9140625,
"learning_rate": 0.0004887807094347606,
"loss": 4.9139,
"mean_token_accuracy": 0.21877105683088302,
"num_tokens": 23884634.0,
"step": 12985
},
{
"entropy": 5.3081128120422365,
"epoch": 1.0850317407283663,
"grad_norm": 1.1328125,
"learning_rate": 0.0004887714242621619,
"loss": 4.9387,
"mean_token_accuracy": 0.21209346801042556,
"num_tokens": 23894307.0,
"step": 12990
},
{
"entropy": 5.284110593795776,
"epoch": 1.085449381891079,
"grad_norm": 0.96484375,
"learning_rate": 0.0004887621353472371,
"loss": 4.8401,
"mean_token_accuracy": 0.21807993352413177,
"num_tokens": 23903274.0,
"step": 12995
},
{
"entropy": 5.253281974792481,
"epoch": 1.0858670230537921,
"grad_norm": 1.046875,
"learning_rate": 0.0004887528426901488,
"loss": 4.8879,
"mean_token_accuracy": 0.21700274348258972,
"num_tokens": 23912787.0,
"step": 13000
},
{
"entropy": 5.313509750366211,
"epoch": 1.0862846642165052,
"grad_norm": 1.0234375,
"learning_rate": 0.0004887435462910597,
"loss": 4.9246,
"mean_token_accuracy": 0.20294243544340135,
"num_tokens": 23921558.0,
"step": 13005
},
{
"entropy": 5.240981197357177,
"epoch": 1.0867023053792182,
"grad_norm": 1.0703125,
"learning_rate": 0.0004887342461501326,
"loss": 4.833,
"mean_token_accuracy": 0.21029905676841737,
"num_tokens": 23929868.0,
"step": 13010
},
{
"entropy": 5.267163228988648,
"epoch": 1.0871199465419312,
"grad_norm": 1.0078125,
"learning_rate": 0.0004887249422675301,
"loss": 4.9146,
"mean_token_accuracy": 0.2056149274110794,
"num_tokens": 23939206.0,
"step": 13015
},
{
"entropy": 5.295957279205322,
"epoch": 1.0875375877046443,
"grad_norm": 1.140625,
"learning_rate": 0.0004887156346434154,
"loss": 4.8385,
"mean_token_accuracy": 0.21635652482509612,
"num_tokens": 23948213.0,
"step": 13020
},
{
"entropy": 5.2973119735717775,
"epoch": 1.087955228867357,
"grad_norm": 1.0625,
"learning_rate": 0.0004887063232779513,
"loss": 4.8759,
"mean_token_accuracy": 0.20758889764547347,
"num_tokens": 23957269.0,
"step": 13025
},
{
"entropy": 5.319609594345093,
"epoch": 1.08837287003007,
"grad_norm": 1.015625,
"learning_rate": 0.0004886970081713006,
"loss": 4.9792,
"mean_token_accuracy": 0.20867991149425508,
"num_tokens": 23967300.0,
"step": 13030
},
{
"entropy": 5.253187561035157,
"epoch": 1.0887905111927831,
"grad_norm": 1.0078125,
"learning_rate": 0.0004886876893236266,
"loss": 4.8786,
"mean_token_accuracy": 0.21111573725938798,
"num_tokens": 23976934.0,
"step": 13035
},
{
"entropy": 5.2428669929504395,
"epoch": 1.0892081523554962,
"grad_norm": 1.0546875,
"learning_rate": 0.0004886783667350925,
"loss": 4.8707,
"mean_token_accuracy": 0.215206877887249,
"num_tokens": 23986119.0,
"step": 13040
},
{
"entropy": 5.221693515777588,
"epoch": 1.0896257935182092,
"grad_norm": 1.03125,
"learning_rate": 0.0004886690404058613,
"loss": 4.8752,
"mean_token_accuracy": 0.21931690126657485,
"num_tokens": 23996093.0,
"step": 13045
},
{
"entropy": 5.384117126464844,
"epoch": 1.0900434346809222,
"grad_norm": 1.0078125,
"learning_rate": 0.0004886597103360963,
"loss": 5.0152,
"mean_token_accuracy": 0.20356177389621735,
"num_tokens": 24006585.0,
"step": 13050
},
{
"entropy": 5.323095989227295,
"epoch": 1.090461075843635,
"grad_norm": 1.1015625,
"learning_rate": 0.0004886503765259609,
"loss": 4.946,
"mean_token_accuracy": 0.20784655213356018,
"num_tokens": 24016398.0,
"step": 13055
},
{
"entropy": 5.270521593093872,
"epoch": 1.090878717006348,
"grad_norm": 1.0234375,
"learning_rate": 0.0004886410389756186,
"loss": 4.8055,
"mean_token_accuracy": 0.21096160411834716,
"num_tokens": 24025988.0,
"step": 13060
},
{
"entropy": 5.294621849060059,
"epoch": 1.0912963581690611,
"grad_norm": 0.89453125,
"learning_rate": 0.0004886316976852326,
"loss": 4.8889,
"mean_token_accuracy": 0.21662607640028,
"num_tokens": 24036448.0,
"step": 13065
},
{
"entropy": 5.314312362670899,
"epoch": 1.0917139993317742,
"grad_norm": 0.9765625,
"learning_rate": 0.0004886223526549666,
"loss": 5.0186,
"mean_token_accuracy": 0.19912010580301284,
"num_tokens": 24047011.0,
"step": 13070
},
{
"entropy": 5.258068323135376,
"epoch": 1.0921316404944872,
"grad_norm": 1.0078125,
"learning_rate": 0.0004886130038849843,
"loss": 4.7831,
"mean_token_accuracy": 0.21857234984636306,
"num_tokens": 24056332.0,
"step": 13075
},
{
"entropy": 5.2940908908844,
"epoch": 1.0925492816572002,
"grad_norm": 1.1328125,
"learning_rate": 0.0004886036513754491,
"loss": 4.9572,
"mean_token_accuracy": 0.2143129900097847,
"num_tokens": 24065538.0,
"step": 13080
},
{
"entropy": 5.244616365432739,
"epoch": 1.092966922819913,
"grad_norm": 0.97265625,
"learning_rate": 0.0004885942951265251,
"loss": 4.8822,
"mean_token_accuracy": 0.21415428519248964,
"num_tokens": 24074950.0,
"step": 13085
},
{
"entropy": 5.342571878433228,
"epoch": 1.093384563982626,
"grad_norm": 1.0703125,
"learning_rate": 0.0004885849351383756,
"loss": 4.9397,
"mean_token_accuracy": 0.20828969031572342,
"num_tokens": 24084571.0,
"step": 13090
},
{
"entropy": 5.284383773803711,
"epoch": 1.093802205145339,
"grad_norm": 1.03125,
"learning_rate": 0.0004885755714111649,
"loss": 4.9368,
"mean_token_accuracy": 0.21385575979948043,
"num_tokens": 24093693.0,
"step": 13095
},
{
"entropy": 5.254420137405395,
"epoch": 1.0942198463080521,
"grad_norm": 0.95703125,
"learning_rate": 0.0004885662039450565,
"loss": 4.8478,
"mean_token_accuracy": 0.21234063804149628,
"num_tokens": 24104140.0,
"step": 13100
},
{
"entropy": 5.3212908744812015,
"epoch": 1.0946374874707652,
"grad_norm": 1.0234375,
"learning_rate": 0.0004885568327402149,
"loss": 4.8459,
"mean_token_accuracy": 0.21311513632535933,
"num_tokens": 24113107.0,
"step": 13105
},
{
"entropy": 5.3180910587310795,
"epoch": 1.0950551286334782,
"grad_norm": 1.03125,
"learning_rate": 0.0004885474577968038,
"loss": 4.8882,
"mean_token_accuracy": 0.21377133280038835,
"num_tokens": 24122271.0,
"step": 13110
},
{
"entropy": 5.316606426239014,
"epoch": 1.095472769796191,
"grad_norm": 0.96484375,
"learning_rate": 0.0004885380791149874,
"loss": 4.9416,
"mean_token_accuracy": 0.2065939873456955,
"num_tokens": 24132068.0,
"step": 13115
},
{
"entropy": 5.257716321945191,
"epoch": 1.095890410958904,
"grad_norm": 1.0703125,
"learning_rate": 0.0004885286966949298,
"loss": 4.8539,
"mean_token_accuracy": 0.20893142819404603,
"num_tokens": 24140429.0,
"step": 13120
},
{
"entropy": 5.243165111541748,
"epoch": 1.096308052121617,
"grad_norm": 0.9609375,
"learning_rate": 0.0004885193105367955,
"loss": 4.9655,
"mean_token_accuracy": 0.20797942876815795,
"num_tokens": 24150563.0,
"step": 13125
},
{
"entropy": 5.333548974990845,
"epoch": 1.0967256932843301,
"grad_norm": 1.1328125,
"learning_rate": 0.0004885099206407486,
"loss": 4.8702,
"mean_token_accuracy": 0.21010884195566176,
"num_tokens": 24159059.0,
"step": 13130
},
{
"entropy": 5.361559534072876,
"epoch": 1.0971433344470432,
"grad_norm": 1.0625,
"learning_rate": 0.0004885005270069535,
"loss": 5.007,
"mean_token_accuracy": 0.2091026335954666,
"num_tokens": 24168858.0,
"step": 13135
},
{
"entropy": 5.247330236434936,
"epoch": 1.0975609756097562,
"grad_norm": 1.046875,
"learning_rate": 0.0004884911296355746,
"loss": 4.9056,
"mean_token_accuracy": 0.21442289054393768,
"num_tokens": 24177573.0,
"step": 13140
},
{
"entropy": 5.290609979629517,
"epoch": 1.097978616772469,
"grad_norm": 0.9375,
"learning_rate": 0.0004884817285267766,
"loss": 4.835,
"mean_token_accuracy": 0.2175571784377098,
"num_tokens": 24186832.0,
"step": 13145
},
{
"entropy": 5.269421339035034,
"epoch": 1.098396257935182,
"grad_norm": 1.09375,
"learning_rate": 0.0004884723236807239,
"loss": 4.8665,
"mean_token_accuracy": 0.21292252093553543,
"num_tokens": 24195281.0,
"step": 13150
},
{
"entropy": 5.289570999145508,
"epoch": 1.098813899097895,
"grad_norm": 1.078125,
"learning_rate": 0.0004884629150975814,
"loss": 4.8824,
"mean_token_accuracy": 0.2166883170604706,
"num_tokens": 24204499.0,
"step": 13155
},
{
"entropy": 5.217087936401367,
"epoch": 1.099231540260608,
"grad_norm": 1.046875,
"learning_rate": 0.0004884535027775135,
"loss": 4.7527,
"mean_token_accuracy": 0.2241186112165451,
"num_tokens": 24213418.0,
"step": 13160
},
{
"entropy": 5.177638196945191,
"epoch": 1.0996491814233211,
"grad_norm": 0.98828125,
"learning_rate": 0.0004884440867206851,
"loss": 4.7634,
"mean_token_accuracy": 0.22278998792171478,
"num_tokens": 24222834.0,
"step": 13165
},
{
"entropy": 5.22744927406311,
"epoch": 1.1000668225860342,
"grad_norm": 1.1328125,
"learning_rate": 0.000488434666927261,
"loss": 4.8797,
"mean_token_accuracy": 0.20938977748155593,
"num_tokens": 24231821.0,
"step": 13170
},
{
"entropy": 5.330484199523926,
"epoch": 1.100484463748747,
"grad_norm": 1.03125,
"learning_rate": 0.0004884252433974062,
"loss": 4.8752,
"mean_token_accuracy": 0.2122475102543831,
"num_tokens": 24240694.0,
"step": 13175
},
{
"entropy": 5.296904993057251,
"epoch": 1.10090210491146,
"grad_norm": 0.9375,
"learning_rate": 0.0004884158161312856,
"loss": 4.9057,
"mean_token_accuracy": 0.21260667890310286,
"num_tokens": 24249927.0,
"step": 13180
},
{
"entropy": 5.280317640304565,
"epoch": 1.101319746074173,
"grad_norm": 1.0859375,
"learning_rate": 0.0004884063851290642,
"loss": 4.8504,
"mean_token_accuracy": 0.21274223178625107,
"num_tokens": 24259174.0,
"step": 13185
},
{
"entropy": 5.27933702468872,
"epoch": 1.101737387236886,
"grad_norm": 1.0390625,
"learning_rate": 0.0004883969503909073,
"loss": 4.9052,
"mean_token_accuracy": 0.20883871912956237,
"num_tokens": 24268381.0,
"step": 13190
},
{
"entropy": 5.323480939865112,
"epoch": 1.1021550283995991,
"grad_norm": 0.98828125,
"learning_rate": 0.0004883875119169798,
"loss": 4.9408,
"mean_token_accuracy": 0.20760657638311386,
"num_tokens": 24277916.0,
"step": 13195
},
{
"entropy": 5.298346710205078,
"epoch": 1.1025726695623121,
"grad_norm": 1.0078125,
"learning_rate": 0.000488378069707447,
"loss": 4.9606,
"mean_token_accuracy": 0.20857729017734528,
"num_tokens": 24287530.0,
"step": 13200
},
{
"entropy": 5.312605333328247,
"epoch": 1.102990310725025,
"grad_norm": 0.8984375,
"learning_rate": 0.0004883686237624744,
"loss": 5.077,
"mean_token_accuracy": 0.1996668115258217,
"num_tokens": 24298239.0,
"step": 13205
},
{
"entropy": 5.363008069992065,
"epoch": 1.103407951887738,
"grad_norm": 1.0625,
"learning_rate": 0.0004883591740822271,
"loss": 4.88,
"mean_token_accuracy": 0.21105607151985167,
"num_tokens": 24307001.0,
"step": 13210
},
{
"entropy": 5.278348159790039,
"epoch": 1.103825593050451,
"grad_norm": 1.0078125,
"learning_rate": 0.0004883497206668707,
"loss": 4.8469,
"mean_token_accuracy": 0.21925781965255736,
"num_tokens": 24315995.0,
"step": 13215
},
{
"entropy": 5.215962266921997,
"epoch": 1.104243234213164,
"grad_norm": 1.015625,
"learning_rate": 0.0004883402635165705,
"loss": 4.7904,
"mean_token_accuracy": 0.22126712501049042,
"num_tokens": 24324421.0,
"step": 13220
},
{
"entropy": 5.257588195800781,
"epoch": 1.104660875375877,
"grad_norm": 1.0234375,
"learning_rate": 0.0004883308026314923,
"loss": 4.9346,
"mean_token_accuracy": 0.20851405411958696,
"num_tokens": 24333715.0,
"step": 13225
},
{
"entropy": 5.256224346160889,
"epoch": 1.1050785165385901,
"grad_norm": 1.015625,
"learning_rate": 0.0004883213380118016,
"loss": 4.83,
"mean_token_accuracy": 0.22057849317789077,
"num_tokens": 24342085.0,
"step": 13230
},
{
"entropy": 5.3145623207092285,
"epoch": 1.105496157701303,
"grad_norm": 1.0859375,
"learning_rate": 0.0004883118696576642,
"loss": 4.9425,
"mean_token_accuracy": 0.20948053449392318,
"num_tokens": 24350906.0,
"step": 13235
},
{
"entropy": 5.172192621231079,
"epoch": 1.105913798864016,
"grad_norm": 0.9921875,
"learning_rate": 0.0004883023975692457,
"loss": 4.796,
"mean_token_accuracy": 0.2146312952041626,
"num_tokens": 24359948.0,
"step": 13240
},
{
"entropy": 5.146786546707153,
"epoch": 1.106331440026729,
"grad_norm": 1.09375,
"learning_rate": 0.0004882929217467118,
"loss": 4.7891,
"mean_token_accuracy": 0.22188411504030228,
"num_tokens": 24369440.0,
"step": 13245
},
{
"entropy": 5.267632722854614,
"epoch": 1.106749081189442,
"grad_norm": 1.1171875,
"learning_rate": 0.0004882834421902287,
"loss": 4.9023,
"mean_token_accuracy": 0.217054283618927,
"num_tokens": 24378442.0,
"step": 13250
},
{
"entropy": 5.24805588722229,
"epoch": 1.107166722352155,
"grad_norm": 0.98828125,
"learning_rate": 0.00048827395889996226,
"loss": 4.8657,
"mean_token_accuracy": 0.21462470591068267,
"num_tokens": 24387750.0,
"step": 13255
},
{
"entropy": 5.220867919921875,
"epoch": 1.107584363514868,
"grad_norm": 1.0859375,
"learning_rate": 0.0004882644718760784,
"loss": 4.6942,
"mean_token_accuracy": 0.2271432861685753,
"num_tokens": 24396380.0,
"step": 13260
},
{
"entropy": 5.190722513198852,
"epoch": 1.108002004677581,
"grad_norm": 1.015625,
"learning_rate": 0.00048825498111874325,
"loss": 4.7734,
"mean_token_accuracy": 0.22173280864953995,
"num_tokens": 24405135.0,
"step": 13265
},
{
"entropy": 5.221589279174805,
"epoch": 1.108419645840294,
"grad_norm": 1.1015625,
"learning_rate": 0.00048824548662812294,
"loss": 4.8912,
"mean_token_accuracy": 0.21462031900882722,
"num_tokens": 24413517.0,
"step": 13270
},
{
"entropy": 5.300827264785767,
"epoch": 1.108837287003007,
"grad_norm": 1.015625,
"learning_rate": 0.00048823598840438384,
"loss": 4.8781,
"mean_token_accuracy": 0.21618583798408508,
"num_tokens": 24422170.0,
"step": 13275
},
{
"entropy": 5.380620670318604,
"epoch": 1.10925492816572,
"grad_norm": 1.109375,
"learning_rate": 0.00048822648644769194,
"loss": 5.0515,
"mean_token_accuracy": 0.20042121708393096,
"num_tokens": 24431083.0,
"step": 13280
},
{
"entropy": 5.317094182968139,
"epoch": 1.109672569328433,
"grad_norm": 1.0703125,
"learning_rate": 0.0004882169807582138,
"loss": 4.907,
"mean_token_accuracy": 0.21104125678539276,
"num_tokens": 24440582.0,
"step": 13285
},
{
"entropy": 5.258843660354614,
"epoch": 1.110090210491146,
"grad_norm": 1.0078125,
"learning_rate": 0.0004882074713361157,
"loss": 4.8593,
"mean_token_accuracy": 0.21963223367929458,
"num_tokens": 24450047.0,
"step": 13290
},
{
"entropy": 5.187111759185791,
"epoch": 1.110507851653859,
"grad_norm": 1.0625,
"learning_rate": 0.0004881979581815642,
"loss": 4.7506,
"mean_token_accuracy": 0.2230774387717247,
"num_tokens": 24457955.0,
"step": 13295
},
{
"entropy": 5.157816219329834,
"epoch": 1.110925492816572,
"grad_norm": 0.98046875,
"learning_rate": 0.0004881884412947259,
"loss": 4.7375,
"mean_token_accuracy": 0.22246803045272828,
"num_tokens": 24467394.0,
"step": 13300
},
{
"entropy": 5.20101900100708,
"epoch": 1.111343133979285,
"grad_norm": 1.109375,
"learning_rate": 0.00048817892067576726,
"loss": 4.8045,
"mean_token_accuracy": 0.22106245458126067,
"num_tokens": 24476386.0,
"step": 13305
},
{
"entropy": 5.172140645980835,
"epoch": 1.111760775141998,
"grad_norm": 1.015625,
"learning_rate": 0.000488169396324855,
"loss": 4.8518,
"mean_token_accuracy": 0.21915037781000138,
"num_tokens": 24484800.0,
"step": 13310
},
{
"entropy": 5.250047302246093,
"epoch": 1.112178416304711,
"grad_norm": 1.0546875,
"learning_rate": 0.00048815986824215585,
"loss": 4.8246,
"mean_token_accuracy": 0.21919767558574677,
"num_tokens": 24494252.0,
"step": 13315
},
{
"entropy": 5.297379779815674,
"epoch": 1.112596057467424,
"grad_norm": 1.0078125,
"learning_rate": 0.0004881503364278366,
"loss": 4.8766,
"mean_token_accuracy": 0.21669656187295913,
"num_tokens": 24503315.0,
"step": 13320
},
{
"entropy": 5.274254083633423,
"epoch": 1.1130136986301369,
"grad_norm": 1.0,
"learning_rate": 0.0004881408008820642,
"loss": 4.8453,
"mean_token_accuracy": 0.21772542744874954,
"num_tokens": 24512660.0,
"step": 13325
},
{
"entropy": 5.247271299362183,
"epoch": 1.11343133979285,
"grad_norm": 1.03125,
"learning_rate": 0.00048813126160500555,
"loss": 4.8654,
"mean_token_accuracy": 0.21289616972208023,
"num_tokens": 24521712.0,
"step": 13330
},
{
"entropy": 5.1636631965637205,
"epoch": 1.113848980955563,
"grad_norm": 1.015625,
"learning_rate": 0.0004881217185968276,
"loss": 4.8193,
"mean_token_accuracy": 0.2247433453798294,
"num_tokens": 24529851.0,
"step": 13335
},
{
"entropy": 5.279679250717163,
"epoch": 1.114266622118276,
"grad_norm": 1.03125,
"learning_rate": 0.0004881121718576974,
"loss": 4.8629,
"mean_token_accuracy": 0.2155037820339203,
"num_tokens": 24538425.0,
"step": 13340
},
{
"entropy": 5.2198402881622314,
"epoch": 1.114684263280989,
"grad_norm": 1.0859375,
"learning_rate": 0.00048810262138778213,
"loss": 4.9518,
"mean_token_accuracy": 0.21139863282442092,
"num_tokens": 24547477.0,
"step": 13345
},
{
"entropy": 5.281708145141602,
"epoch": 1.115101904443702,
"grad_norm": 1.1015625,
"learning_rate": 0.00048809306718724886,
"loss": 4.9065,
"mean_token_accuracy": 0.20673473477363585,
"num_tokens": 24556230.0,
"step": 13350
},
{
"entropy": 5.260258531570434,
"epoch": 1.115519545606415,
"grad_norm": 1.0078125,
"learning_rate": 0.00048808350925626514,
"loss": 4.8676,
"mean_token_accuracy": 0.2165923684835434,
"num_tokens": 24565021.0,
"step": 13355
},
{
"entropy": 5.2782727718353275,
"epoch": 1.115937186769128,
"grad_norm": 1.03125,
"learning_rate": 0.00048807394759499797,
"loss": 4.8454,
"mean_token_accuracy": 0.21774390190839768,
"num_tokens": 24575012.0,
"step": 13360
},
{
"entropy": 5.34255690574646,
"epoch": 1.116354827931841,
"grad_norm": 1.0703125,
"learning_rate": 0.0004880643822036149,
"loss": 5.0048,
"mean_token_accuracy": 0.2024964228272438,
"num_tokens": 24584267.0,
"step": 13365
},
{
"entropy": 5.300237035751342,
"epoch": 1.116772469094554,
"grad_norm": 1.046875,
"learning_rate": 0.00048805481308228333,
"loss": 4.895,
"mean_token_accuracy": 0.21600398868322374,
"num_tokens": 24594299.0,
"step": 13370
},
{
"entropy": 5.283212757110595,
"epoch": 1.117190110257267,
"grad_norm": 1.0625,
"learning_rate": 0.0004880452402311709,
"loss": 4.9006,
"mean_token_accuracy": 0.21374878883361817,
"num_tokens": 24603717.0,
"step": 13375
},
{
"entropy": 5.295483160018921,
"epoch": 1.11760775141998,
"grad_norm": 1.0390625,
"learning_rate": 0.00048803566365044493,
"loss": 4.9566,
"mean_token_accuracy": 0.20913030654191972,
"num_tokens": 24613141.0,
"step": 13380
},
{
"entropy": 5.2521013736724855,
"epoch": 1.1180253925826928,
"grad_norm": 0.984375,
"learning_rate": 0.0004880260833402734,
"loss": 4.8953,
"mean_token_accuracy": 0.22047676742076874,
"num_tokens": 24621837.0,
"step": 13385
},
{
"entropy": 5.350809621810913,
"epoch": 1.1184430337454059,
"grad_norm": 1.1015625,
"learning_rate": 0.00048801649930082375,
"loss": 4.9209,
"mean_token_accuracy": 0.2087827205657959,
"num_tokens": 24630657.0,
"step": 13390
},
{
"entropy": 5.399842739105225,
"epoch": 1.118860674908119,
"grad_norm": 1.09375,
"learning_rate": 0.000488006911532264,
"loss": 5.0105,
"mean_token_accuracy": 0.19916945397853852,
"num_tokens": 24639929.0,
"step": 13395
},
{
"entropy": 5.305797529220581,
"epoch": 1.119278316070832,
"grad_norm": 1.1328125,
"learning_rate": 0.0004879973200347617,
"loss": 4.9461,
"mean_token_accuracy": 0.21596447825431825,
"num_tokens": 24648537.0,
"step": 13400
},
{
"entropy": 5.23604679107666,
"epoch": 1.119695957233545,
"grad_norm": 1.078125,
"learning_rate": 0.00048798772480848504,
"loss": 4.9029,
"mean_token_accuracy": 0.21981456875801086,
"num_tokens": 24656809.0,
"step": 13405
},
{
"entropy": 5.248571252822876,
"epoch": 1.120113598396258,
"grad_norm": 1.0859375,
"learning_rate": 0.00048797812585360187,
"loss": 4.8878,
"mean_token_accuracy": 0.20701291859149934,
"num_tokens": 24665523.0,
"step": 13410
},
{
"entropy": 5.279518127441406,
"epoch": 1.120531239558971,
"grad_norm": 1.03125,
"learning_rate": 0.00048796852317028024,
"loss": 4.8408,
"mean_token_accuracy": 0.21737050712108613,
"num_tokens": 24675618.0,
"step": 13415
},
{
"entropy": 5.261980962753296,
"epoch": 1.1209488807216839,
"grad_norm": 1.0546875,
"learning_rate": 0.0004879589167586882,
"loss": 4.883,
"mean_token_accuracy": 0.21820248514413834,
"num_tokens": 24684922.0,
"step": 13420
},
{
"entropy": 5.257414960861206,
"epoch": 1.121366521884397,
"grad_norm": 1.0390625,
"learning_rate": 0.00048794930661899414,
"loss": 4.9468,
"mean_token_accuracy": 0.21041895151138307,
"num_tokens": 24693791.0,
"step": 13425
},
{
"entropy": 5.233606576919556,
"epoch": 1.12178416304711,
"grad_norm": 1.0078125,
"learning_rate": 0.00048793969275136593,
"loss": 4.8116,
"mean_token_accuracy": 0.21604859381914138,
"num_tokens": 24702661.0,
"step": 13430
},
{
"entropy": 5.26155834197998,
"epoch": 1.122201804209823,
"grad_norm": 1.0859375,
"learning_rate": 0.0004879300751559722,
"loss": 4.8902,
"mean_token_accuracy": 0.21194994896650315,
"num_tokens": 24711019.0,
"step": 13435
},
{
"entropy": 5.339042854309082,
"epoch": 1.122619445372536,
"grad_norm": 0.96484375,
"learning_rate": 0.0004879204538329812,
"loss": 4.9289,
"mean_token_accuracy": 0.20742554664611818,
"num_tokens": 24720456.0,
"step": 13440
},
{
"entropy": 5.273082828521728,
"epoch": 1.1230370865352488,
"grad_norm": 1.1015625,
"learning_rate": 0.0004879108287825614,
"loss": 4.8714,
"mean_token_accuracy": 0.2195258542895317,
"num_tokens": 24729395.0,
"step": 13445
},
{
"entropy": 5.215933513641358,
"epoch": 1.1234547276979618,
"grad_norm": 1.03125,
"learning_rate": 0.00048790120000488115,
"loss": 4.8661,
"mean_token_accuracy": 0.21700615733861922,
"num_tokens": 24738543.0,
"step": 13450
},
{
"entropy": 5.29384331703186,
"epoch": 1.1238723688606749,
"grad_norm": 0.94921875,
"learning_rate": 0.0004878915675001092,
"loss": 4.8765,
"mean_token_accuracy": 0.2101330578327179,
"num_tokens": 24747860.0,
"step": 13455
},
{
"entropy": 5.185804080963135,
"epoch": 1.124290010023388,
"grad_norm": 1.015625,
"learning_rate": 0.0004878819312684141,
"loss": 4.832,
"mean_token_accuracy": 0.2126847416162491,
"num_tokens": 24757457.0,
"step": 13460
},
{
"entropy": 5.326801443099976,
"epoch": 1.124707651186101,
"grad_norm": 1.0625,
"learning_rate": 0.0004878722913099645,
"loss": 4.8844,
"mean_token_accuracy": 0.21128021329641342,
"num_tokens": 24766844.0,
"step": 13465
},
{
"entropy": 5.305566453933716,
"epoch": 1.125125292348814,
"grad_norm": 1.0390625,
"learning_rate": 0.0004878626476249293,
"loss": 4.9584,
"mean_token_accuracy": 0.20470163524150847,
"num_tokens": 24776041.0,
"step": 13470
},
{
"entropy": 5.269011211395264,
"epoch": 1.125542933511527,
"grad_norm": 1.0390625,
"learning_rate": 0.0004878530002134772,
"loss": 4.8742,
"mean_token_accuracy": 0.2190667510032654,
"num_tokens": 24785515.0,
"step": 13475
},
{
"entropy": 5.363717412948608,
"epoch": 1.1259605746742398,
"grad_norm": 0.93359375,
"learning_rate": 0.0004878433490757771,
"loss": 4.9344,
"mean_token_accuracy": 0.21687734574079515,
"num_tokens": 24795251.0,
"step": 13480
},
{
"entropy": 5.216241884231567,
"epoch": 1.1263782158369529,
"grad_norm": 1.046875,
"learning_rate": 0.0004878336942119979,
"loss": 4.8496,
"mean_token_accuracy": 0.21828902512788773,
"num_tokens": 24804672.0,
"step": 13485
},
{
"entropy": 5.273200654983521,
"epoch": 1.126795856999666,
"grad_norm": 0.984375,
"learning_rate": 0.00048782403562230874,
"loss": 4.8917,
"mean_token_accuracy": 0.21693220138549804,
"num_tokens": 24813520.0,
"step": 13490
},
{
"entropy": 5.254609727859497,
"epoch": 1.127213498162379,
"grad_norm": 1.0703125,
"learning_rate": 0.0004878143733068787,
"loss": 4.7351,
"mean_token_accuracy": 0.22602677941322327,
"num_tokens": 24821368.0,
"step": 13495
},
{
"entropy": 5.22702374458313,
"epoch": 1.127631139325092,
"grad_norm": 1.0703125,
"learning_rate": 0.0004878047072658769,
"loss": 4.8304,
"mean_token_accuracy": 0.21328341513872145,
"num_tokens": 24830071.0,
"step": 13500
},
{
"entropy": 5.175024938583374,
"epoch": 1.1280487804878048,
"grad_norm": 0.92578125,
"learning_rate": 0.0004877950374994725,
"loss": 4.7566,
"mean_token_accuracy": 0.22273661941289902,
"num_tokens": 24838982.0,
"step": 13505
},
{
"entropy": 5.30128059387207,
"epoch": 1.1284664216505178,
"grad_norm": 1.03125,
"learning_rate": 0.0004877853640078349,
"loss": 4.9417,
"mean_token_accuracy": 0.20314504504203795,
"num_tokens": 24848313.0,
"step": 13510
},
{
"entropy": 5.2694062232971195,
"epoch": 1.1288840628132308,
"grad_norm": 1.078125,
"learning_rate": 0.0004877756867911333,
"loss": 4.9035,
"mean_token_accuracy": 0.21516518145799637,
"num_tokens": 24857477.0,
"step": 13515
},
{
"entropy": 5.239373350143433,
"epoch": 1.1293017039759439,
"grad_norm": 1.0078125,
"learning_rate": 0.0004877660058495372,
"loss": 4.8624,
"mean_token_accuracy": 0.21387197226285934,
"num_tokens": 24866561.0,
"step": 13520
},
{
"entropy": 5.27010178565979,
"epoch": 1.129719345138657,
"grad_norm": 1.0078125,
"learning_rate": 0.0004877563211832161,
"loss": 4.9551,
"mean_token_accuracy": 0.20328208655118943,
"num_tokens": 24876113.0,
"step": 13525
},
{
"entropy": 5.38572325706482,
"epoch": 1.13013698630137,
"grad_norm": 0.96875,
"learning_rate": 0.00048774663279233947,
"loss": 4.9954,
"mean_token_accuracy": 0.20648373961448668,
"num_tokens": 24885974.0,
"step": 13530
},
{
"entropy": 5.272332334518433,
"epoch": 1.130554627464083,
"grad_norm": 0.95703125,
"learning_rate": 0.00048773694067707693,
"loss": 4.8405,
"mean_token_accuracy": 0.2193993330001831,
"num_tokens": 24895046.0,
"step": 13535
},
{
"entropy": 5.260845041275024,
"epoch": 1.1309722686267958,
"grad_norm": 1.0625,
"learning_rate": 0.00048772724483759825,
"loss": 4.9914,
"mean_token_accuracy": 0.2100438356399536,
"num_tokens": 24903315.0,
"step": 13540
},
{
"entropy": 5.306494522094726,
"epoch": 1.1313899097895088,
"grad_norm": 1.0,
"learning_rate": 0.000487717545274073,
"loss": 4.9752,
"mean_token_accuracy": 0.20811050534248351,
"num_tokens": 24912556.0,
"step": 13545
},
{
"entropy": 5.360071277618408,
"epoch": 1.1318075509522219,
"grad_norm": 1.09375,
"learning_rate": 0.0004877078419866711,
"loss": 4.9106,
"mean_token_accuracy": 0.21467335820198058,
"num_tokens": 24921854.0,
"step": 13550
},
{
"entropy": 5.241201591491699,
"epoch": 1.1322251921149349,
"grad_norm": 1.0625,
"learning_rate": 0.00048769813497556236,
"loss": 4.8596,
"mean_token_accuracy": 0.21896041333675384,
"num_tokens": 24930691.0,
"step": 13555
},
{
"entropy": 5.255487728118896,
"epoch": 1.132642833277648,
"grad_norm": 0.95703125,
"learning_rate": 0.0004876884242409168,
"loss": 4.9203,
"mean_token_accuracy": 0.2162812978029251,
"num_tokens": 24940573.0,
"step": 13560
},
{
"entropy": 5.249113988876343,
"epoch": 1.1330604744403607,
"grad_norm": 0.94140625,
"learning_rate": 0.00048767870978290434,
"loss": 4.8086,
"mean_token_accuracy": 0.21294808238744736,
"num_tokens": 24950902.0,
"step": 13565
},
{
"entropy": 5.255436754226684,
"epoch": 1.1334781156030738,
"grad_norm": 0.97265625,
"learning_rate": 0.0004876689916016951,
"loss": 4.8406,
"mean_token_accuracy": 0.21964537352323532,
"num_tokens": 24960055.0,
"step": 13570
},
{
"entropy": 5.279968881607056,
"epoch": 1.1338957567657868,
"grad_norm": 1.0078125,
"learning_rate": 0.0004876592696974591,
"loss": 4.8741,
"mean_token_accuracy": 0.21051753014326097,
"num_tokens": 24969896.0,
"step": 13575
},
{
"entropy": 5.299686861038208,
"epoch": 1.1343133979284998,
"grad_norm": 1.046875,
"learning_rate": 0.0004876495440703667,
"loss": 4.834,
"mean_token_accuracy": 0.21472468227148056,
"num_tokens": 24978467.0,
"step": 13580
},
{
"entropy": 5.243856191635132,
"epoch": 1.1347310390912129,
"grad_norm": 1.078125,
"learning_rate": 0.0004876398147205879,
"loss": 4.8839,
"mean_token_accuracy": 0.21529708802700043,
"num_tokens": 24987038.0,
"step": 13585
},
{
"entropy": 5.351840877532959,
"epoch": 1.135148680253926,
"grad_norm": 0.94140625,
"learning_rate": 0.0004876300816482933,
"loss": 4.9363,
"mean_token_accuracy": 0.21152514964342117,
"num_tokens": 24997228.0,
"step": 13590
},
{
"entropy": 5.248982906341553,
"epoch": 1.135566321416639,
"grad_norm": 1.03125,
"learning_rate": 0.0004876203448536532,
"loss": 4.8214,
"mean_token_accuracy": 0.21323238760232927,
"num_tokens": 25005784.0,
"step": 13595
},
{
"entropy": 5.268309211730957,
"epoch": 1.1359839625793517,
"grad_norm": 0.94140625,
"learning_rate": 0.0004876106043368381,
"loss": 4.9229,
"mean_token_accuracy": 0.21076042354106903,
"num_tokens": 25015863.0,
"step": 13600
},
{
"entropy": 5.161695194244385,
"epoch": 1.1364016037420648,
"grad_norm": 1.015625,
"learning_rate": 0.0004876008600980183,
"loss": 4.7531,
"mean_token_accuracy": 0.2266981229186058,
"num_tokens": 25025391.0,
"step": 13605
},
{
"entropy": 5.308411169052124,
"epoch": 1.1368192449047778,
"grad_norm": 0.97265625,
"learning_rate": 0.00048759111213736455,
"loss": 4.9694,
"mean_token_accuracy": 0.20882852375507355,
"num_tokens": 25035387.0,
"step": 13610
},
{
"entropy": 5.191502809524536,
"epoch": 1.1372368860674908,
"grad_norm": 1.0078125,
"learning_rate": 0.0004875813604550476,
"loss": 4.7296,
"mean_token_accuracy": 0.22554003000259398,
"num_tokens": 25044485.0,
"step": 13615
},
{
"entropy": 5.2118431568145756,
"epoch": 1.1376545272302039,
"grad_norm": 1.1484375,
"learning_rate": 0.0004875716050512379,
"loss": 4.848,
"mean_token_accuracy": 0.21343065351247786,
"num_tokens": 25053079.0,
"step": 13620
},
{
"entropy": 5.340713262557983,
"epoch": 1.1380721683929167,
"grad_norm": 1.0,
"learning_rate": 0.0004875618459261065,
"loss": 4.961,
"mean_token_accuracy": 0.20659567564725875,
"num_tokens": 25063262.0,
"step": 13625
},
{
"entropy": 5.293006229400635,
"epoch": 1.1384898095556297,
"grad_norm": 1.09375,
"learning_rate": 0.00048755208307982414,
"loss": 4.8392,
"mean_token_accuracy": 0.22037013173103331,
"num_tokens": 25071720.0,
"step": 13630
},
{
"entropy": 5.204473066329956,
"epoch": 1.1389074507183428,
"grad_norm": 0.94921875,
"learning_rate": 0.0004875423165125616,
"loss": 4.8726,
"mean_token_accuracy": 0.2142190083861351,
"num_tokens": 25080460.0,
"step": 13635
},
{
"entropy": 5.22118730545044,
"epoch": 1.1393250918810558,
"grad_norm": 1.0703125,
"learning_rate": 0.00048753254622449006,
"loss": 4.8764,
"mean_token_accuracy": 0.21225495487451554,
"num_tokens": 25089162.0,
"step": 13640
},
{
"entropy": 5.232584667205811,
"epoch": 1.1397427330437688,
"grad_norm": 1.0625,
"learning_rate": 0.0004875227722157804,
"loss": 4.7631,
"mean_token_accuracy": 0.217538021504879,
"num_tokens": 25098234.0,
"step": 13645
},
{
"entropy": 5.206743955612183,
"epoch": 1.1401603742064819,
"grad_norm": 1.0078125,
"learning_rate": 0.0004875129944866039,
"loss": 4.8166,
"mean_token_accuracy": 0.2171617105603218,
"num_tokens": 25107832.0,
"step": 13650
},
{
"entropy": 5.260278701782227,
"epoch": 1.140578015369195,
"grad_norm": 1.1015625,
"learning_rate": 0.0004875032130371316,
"loss": 4.8443,
"mean_token_accuracy": 0.2167891651391983,
"num_tokens": 25116118.0,
"step": 13655
},
{
"entropy": 5.145340776443481,
"epoch": 1.1409956565319077,
"grad_norm": 0.9453125,
"learning_rate": 0.0004874934278675347,
"loss": 4.6927,
"mean_token_accuracy": 0.22428550571203232,
"num_tokens": 25125025.0,
"step": 13660
},
{
"entropy": 5.282260751724243,
"epoch": 1.1414132976946207,
"grad_norm": 1.0390625,
"learning_rate": 0.0004874836389779846,
"loss": 4.9523,
"mean_token_accuracy": 0.21477939933538437,
"num_tokens": 25135763.0,
"step": 13665
},
{
"entropy": 5.4009339809417725,
"epoch": 1.1418309388573338,
"grad_norm": 1.03125,
"learning_rate": 0.00048747384636865267,
"loss": 4.9792,
"mean_token_accuracy": 0.20728751718997956,
"num_tokens": 25145510.0,
"step": 13670
},
{
"entropy": 5.3031398296356205,
"epoch": 1.1422485800200468,
"grad_norm": 1.0546875,
"learning_rate": 0.0004874640500397103,
"loss": 4.8406,
"mean_token_accuracy": 0.21275755316019057,
"num_tokens": 25153910.0,
"step": 13675
},
{
"entropy": 5.2443804264068605,
"epoch": 1.1426662211827598,
"grad_norm": 0.98828125,
"learning_rate": 0.0004874542499913289,
"loss": 4.8622,
"mean_token_accuracy": 0.21285583674907685,
"num_tokens": 25163357.0,
"step": 13680
},
{
"entropy": 5.256487607955933,
"epoch": 1.1430838623454727,
"grad_norm": 1.0859375,
"learning_rate": 0.00048744444622368015,
"loss": 4.8774,
"mean_token_accuracy": 0.21510632634162902,
"num_tokens": 25172423.0,
"step": 13685
},
{
"entropy": 5.24192247390747,
"epoch": 1.1435015035081857,
"grad_norm": 1.0390625,
"learning_rate": 0.00048743463873693576,
"loss": 4.8221,
"mean_token_accuracy": 0.21188147962093354,
"num_tokens": 25181508.0,
"step": 13690
},
{
"entropy": 5.276925706863404,
"epoch": 1.1439191446708987,
"grad_norm": 1.0546875,
"learning_rate": 0.0004874248275312673,
"loss": 4.9338,
"mean_token_accuracy": 0.21511436849832535,
"num_tokens": 25190741.0,
"step": 13695
},
{
"entropy": 5.337406921386719,
"epoch": 1.1443367858336118,
"grad_norm": 0.96875,
"learning_rate": 0.00048741501260684644,
"loss": 4.9981,
"mean_token_accuracy": 0.2045115828514099,
"num_tokens": 25201288.0,
"step": 13700
},
{
"entropy": 5.280719566345215,
"epoch": 1.1447544269963248,
"grad_norm": 1.171875,
"learning_rate": 0.00048740519396384515,
"loss": 4.8627,
"mean_token_accuracy": 0.2210691899061203,
"num_tokens": 25210111.0,
"step": 13705
},
{
"entropy": 5.28420467376709,
"epoch": 1.1451720681590378,
"grad_norm": 0.984375,
"learning_rate": 0.00048739537160243524,
"loss": 4.9042,
"mean_token_accuracy": 0.20876525044441224,
"num_tokens": 25220061.0,
"step": 13710
},
{
"entropy": 5.2749473571777346,
"epoch": 1.1455897093217509,
"grad_norm": 0.9921875,
"learning_rate": 0.00048738554552278874,
"loss": 4.8727,
"mean_token_accuracy": 0.21307254135608672,
"num_tokens": 25229336.0,
"step": 13715
},
{
"entropy": 5.2678478240966795,
"epoch": 1.1460073504844637,
"grad_norm": 0.98046875,
"learning_rate": 0.00048737571572507753,
"loss": 4.8318,
"mean_token_accuracy": 0.22434635758399962,
"num_tokens": 25238539.0,
"step": 13720
},
{
"entropy": 5.269054269790649,
"epoch": 1.1464249916471767,
"grad_norm": 0.94921875,
"learning_rate": 0.00048736588220947384,
"loss": 4.8858,
"mean_token_accuracy": 0.20907706916332244,
"num_tokens": 25247880.0,
"step": 13725
},
{
"entropy": 5.2703601837158205,
"epoch": 1.1468426328098897,
"grad_norm": 1.140625,
"learning_rate": 0.00048735604497614974,
"loss": 4.901,
"mean_token_accuracy": 0.21622994542121887,
"num_tokens": 25257295.0,
"step": 13730
},
{
"entropy": 5.273104572296143,
"epoch": 1.1472602739726028,
"grad_norm": 0.94921875,
"learning_rate": 0.0004873462040252775,
"loss": 4.8904,
"mean_token_accuracy": 0.21302305907011032,
"num_tokens": 25266784.0,
"step": 13735
},
{
"entropy": 5.337120151519775,
"epoch": 1.1476779151353158,
"grad_norm": 1.109375,
"learning_rate": 0.0004873363593570293,
"loss": 4.9571,
"mean_token_accuracy": 0.20905491709709167,
"num_tokens": 25276910.0,
"step": 13740
},
{
"entropy": 5.324481439590454,
"epoch": 1.1480955562980286,
"grad_norm": 1.0859375,
"learning_rate": 0.0004873265109715776,
"loss": 4.8546,
"mean_token_accuracy": 0.2200960025191307,
"num_tokens": 25285776.0,
"step": 13745
},
{
"entropy": 5.290006256103515,
"epoch": 1.1485131974607417,
"grad_norm": 1.125,
"learning_rate": 0.0004873166588690947,
"loss": 4.8376,
"mean_token_accuracy": 0.21993631571531297,
"num_tokens": 25294319.0,
"step": 13750
},
{
"entropy": 5.249959182739258,
"epoch": 1.1489308386234547,
"grad_norm": 1.1015625,
"learning_rate": 0.0004873068030497532,
"loss": 4.8457,
"mean_token_accuracy": 0.21979946196079253,
"num_tokens": 25304396.0,
"step": 13755
},
{
"entropy": 5.2008191585540775,
"epoch": 1.1493484797861677,
"grad_norm": 0.94140625,
"learning_rate": 0.00048729694351372545,
"loss": 4.7946,
"mean_token_accuracy": 0.22379758059978486,
"num_tokens": 25313095.0,
"step": 13760
},
{
"entropy": 5.292687034606933,
"epoch": 1.1497661209488808,
"grad_norm": 1.03125,
"learning_rate": 0.0004872870802611842,
"loss": 4.9888,
"mean_token_accuracy": 0.21225251853466034,
"num_tokens": 25322948.0,
"step": 13765
},
{
"entropy": 5.230883169174194,
"epoch": 1.1501837621115938,
"grad_norm": 1.0234375,
"learning_rate": 0.0004872772132923021,
"loss": 4.8363,
"mean_token_accuracy": 0.21453835517168046,
"num_tokens": 25332707.0,
"step": 13770
},
{
"entropy": 5.238774824142456,
"epoch": 1.1506014032743068,
"grad_norm": 0.96875,
"learning_rate": 0.00048726734260725186,
"loss": 4.8482,
"mean_token_accuracy": 0.2151096999645233,
"num_tokens": 25342029.0,
"step": 13775
},
{
"entropy": 5.229918098449707,
"epoch": 1.1510190444370196,
"grad_norm": 0.99609375,
"learning_rate": 0.0004872574682062063,
"loss": 4.7607,
"mean_token_accuracy": 0.2243787631392479,
"num_tokens": 25350374.0,
"step": 13780
},
{
"entropy": 5.261537027359009,
"epoch": 1.1514366855997327,
"grad_norm": 1.0546875,
"learning_rate": 0.0004872475900893382,
"loss": 4.9228,
"mean_token_accuracy": 0.2166041538119316,
"num_tokens": 25359370.0,
"step": 13785
},
{
"entropy": 5.243124389648438,
"epoch": 1.1518543267624457,
"grad_norm": 1.125,
"learning_rate": 0.0004872377082568206,
"loss": 4.8316,
"mean_token_accuracy": 0.2200513869524002,
"num_tokens": 25368329.0,
"step": 13790
},
{
"entropy": 5.308872222900391,
"epoch": 1.1522719679251587,
"grad_norm": 1.0234375,
"learning_rate": 0.00048722782270882643,
"loss": 4.9687,
"mean_token_accuracy": 0.20813584923744202,
"num_tokens": 25377795.0,
"step": 13795
},
{
"entropy": 5.329696893692017,
"epoch": 1.1526896090878718,
"grad_norm": 0.9921875,
"learning_rate": 0.0004872179334455288,
"loss": 4.9155,
"mean_token_accuracy": 0.211460117995739,
"num_tokens": 25387134.0,
"step": 13800
},
{
"entropy": 5.355197620391846,
"epoch": 1.1531072502505846,
"grad_norm": 1.0546875,
"learning_rate": 0.00048720804046710074,
"loss": 4.9422,
"mean_token_accuracy": 0.2084246888756752,
"num_tokens": 25396320.0,
"step": 13805
},
{
"entropy": 5.303622055053711,
"epoch": 1.1535248914132976,
"grad_norm": 0.96875,
"learning_rate": 0.00048719814377371553,
"loss": 4.9015,
"mean_token_accuracy": 0.21105479300022126,
"num_tokens": 25404727.0,
"step": 13810
},
{
"entropy": 5.181811952590943,
"epoch": 1.1539425325760106,
"grad_norm": 0.9921875,
"learning_rate": 0.0004871882433655464,
"loss": 4.8278,
"mean_token_accuracy": 0.21476244330406188,
"num_tokens": 25414137.0,
"step": 13815
},
{
"entropy": 5.318673086166382,
"epoch": 1.1543601737387237,
"grad_norm": 0.953125,
"learning_rate": 0.0004871783392427666,
"loss": 4.8895,
"mean_token_accuracy": 0.2155558407306671,
"num_tokens": 25423297.0,
"step": 13820
},
{
"entropy": 5.40968689918518,
"epoch": 1.1547778149014367,
"grad_norm": 0.953125,
"learning_rate": 0.0004871684314055497,
"loss": 4.9799,
"mean_token_accuracy": 0.20710690021514894,
"num_tokens": 25433185.0,
"step": 13825
},
{
"entropy": 5.306927680969238,
"epoch": 1.1551954560641498,
"grad_norm": 1.125,
"learning_rate": 0.0004871585198540689,
"loss": 4.8856,
"mean_token_accuracy": 0.22176383584737777,
"num_tokens": 25441769.0,
"step": 13830
},
{
"entropy": 5.223139095306396,
"epoch": 1.1556130972268628,
"grad_norm": 1.078125,
"learning_rate": 0.00048714860458849794,
"loss": 4.8302,
"mean_token_accuracy": 0.2161799266934395,
"num_tokens": 25450650.0,
"step": 13835
},
{
"entropy": 5.2494409561157225,
"epoch": 1.1560307383895756,
"grad_norm": 1.0859375,
"learning_rate": 0.00048713868560901013,
"loss": 4.8332,
"mean_token_accuracy": 0.22578169256448746,
"num_tokens": 25458721.0,
"step": 13840
},
{
"entropy": 5.300110912322998,
"epoch": 1.1564483795522886,
"grad_norm": 1.0859375,
"learning_rate": 0.0004871287629157794,
"loss": 4.8995,
"mean_token_accuracy": 0.2149494931101799,
"num_tokens": 25468004.0,
"step": 13845
},
{
"entropy": 5.333184099197387,
"epoch": 1.1568660207150017,
"grad_norm": 0.99609375,
"learning_rate": 0.0004871188365089793,
"loss": 5.0121,
"mean_token_accuracy": 0.21131540834903717,
"num_tokens": 25478042.0,
"step": 13850
},
{
"entropy": 5.354527282714844,
"epoch": 1.1572836618777147,
"grad_norm": 0.96875,
"learning_rate": 0.0004871089063887837,
"loss": 4.9053,
"mean_token_accuracy": 0.21250555217266082,
"num_tokens": 25488439.0,
"step": 13855
},
{
"entropy": 5.290424489974976,
"epoch": 1.1577013030404277,
"grad_norm": 1.0546875,
"learning_rate": 0.00048709897255536626,
"loss": 4.8935,
"mean_token_accuracy": 0.2129376769065857,
"num_tokens": 25497928.0,
"step": 13860
},
{
"entropy": 5.316667795181274,
"epoch": 1.1581189442031405,
"grad_norm": 0.98046875,
"learning_rate": 0.0004870890350089011,
"loss": 4.933,
"mean_token_accuracy": 0.212449249625206,
"num_tokens": 25506522.0,
"step": 13865
},
{
"entropy": 5.296707677841186,
"epoch": 1.1585365853658536,
"grad_norm": 1.046875,
"learning_rate": 0.0004870790937495621,
"loss": 4.9023,
"mean_token_accuracy": 0.20084939897060394,
"num_tokens": 25516017.0,
"step": 13870
},
{
"entropy": 5.284860420227051,
"epoch": 1.1589542265285666,
"grad_norm": 1.0703125,
"learning_rate": 0.00048706914877752325,
"loss": 4.8668,
"mean_token_accuracy": 0.2173384189605713,
"num_tokens": 25525262.0,
"step": 13875
},
{
"entropy": 5.179403018951416,
"epoch": 1.1593718676912796,
"grad_norm": 1.0390625,
"learning_rate": 0.00048705920009295867,
"loss": 4.8473,
"mean_token_accuracy": 0.21863864809274675,
"num_tokens": 25534479.0,
"step": 13880
},
{
"entropy": 5.27202582359314,
"epoch": 1.1597895088539927,
"grad_norm": 1.0703125,
"learning_rate": 0.0004870492476960426,
"loss": 4.8789,
"mean_token_accuracy": 0.21805175095796586,
"num_tokens": 25543338.0,
"step": 13885
},
{
"entropy": 5.419294738769532,
"epoch": 1.1602071500167057,
"grad_norm": 1.0234375,
"learning_rate": 0.0004870392915869491,
"loss": 5.0097,
"mean_token_accuracy": 0.2063686102628708,
"num_tokens": 25552286.0,
"step": 13890
},
{
"entropy": 5.342674684524536,
"epoch": 1.1606247911794187,
"grad_norm": 1.046875,
"learning_rate": 0.0004870293317658526,
"loss": 4.9198,
"mean_token_accuracy": 0.20889928340911865,
"num_tokens": 25561978.0,
"step": 13895
},
{
"entropy": 5.236069011688232,
"epoch": 1.1610424323421316,
"grad_norm": 1.109375,
"learning_rate": 0.0004870193682329275,
"loss": 4.826,
"mean_token_accuracy": 0.21892983466386795,
"num_tokens": 25570139.0,
"step": 13900
},
{
"entropy": 5.258753538131714,
"epoch": 1.1614600735048446,
"grad_norm": 1.046875,
"learning_rate": 0.00048700940098834805,
"loss": 4.8537,
"mean_token_accuracy": 0.21106000542640685,
"num_tokens": 25579761.0,
"step": 13905
},
{
"entropy": 5.253506469726562,
"epoch": 1.1618777146675576,
"grad_norm": 1.125,
"learning_rate": 0.0004869994300322889,
"loss": 4.8716,
"mean_token_accuracy": 0.21221585422754288,
"num_tokens": 25588740.0,
"step": 13910
},
{
"entropy": 5.256745052337647,
"epoch": 1.1622953558302707,
"grad_norm": 1.15625,
"learning_rate": 0.00048698945536492454,
"loss": 4.908,
"mean_token_accuracy": 0.21066562533378602,
"num_tokens": 25597963.0,
"step": 13915
},
{
"entropy": 5.230707883834839,
"epoch": 1.1627129969929837,
"grad_norm": 1.046875,
"learning_rate": 0.0004869794769864296,
"loss": 4.873,
"mean_token_accuracy": 0.21316303312778473,
"num_tokens": 25606782.0,
"step": 13920
},
{
"entropy": 5.384146356582642,
"epoch": 1.1631306381556965,
"grad_norm": 1.03125,
"learning_rate": 0.0004869694948969787,
"loss": 4.9702,
"mean_token_accuracy": 0.2007288709282875,
"num_tokens": 25616458.0,
"step": 13925
},
{
"entropy": 5.343670511245728,
"epoch": 1.1635482793184095,
"grad_norm": 1.09375,
"learning_rate": 0.00048695950909674666,
"loss": 4.9283,
"mean_token_accuracy": 0.2056759625673294,
"num_tokens": 25625081.0,
"step": 13930
},
{
"entropy": 5.286416053771973,
"epoch": 1.1639659204811226,
"grad_norm": 1.2421875,
"learning_rate": 0.00048694951958590835,
"loss": 4.9451,
"mean_token_accuracy": 0.20673276782035827,
"num_tokens": 25634000.0,
"step": 13935
},
{
"entropy": 5.3067498207092285,
"epoch": 1.1643835616438356,
"grad_norm": 0.984375,
"learning_rate": 0.00048693952636463854,
"loss": 4.8556,
"mean_token_accuracy": 0.21781117022037505,
"num_tokens": 25642715.0,
"step": 13940
},
{
"entropy": 5.1775353908538815,
"epoch": 1.1648012028065486,
"grad_norm": 1.0390625,
"learning_rate": 0.00048692952943311214,
"loss": 4.6865,
"mean_token_accuracy": 0.229061359167099,
"num_tokens": 25650979.0,
"step": 13945
},
{
"entropy": 5.264878463745117,
"epoch": 1.1652188439692617,
"grad_norm": 0.9921875,
"learning_rate": 0.0004869195287915043,
"loss": 4.9073,
"mean_token_accuracy": 0.21405562609434128,
"num_tokens": 25660860.0,
"step": 13950
},
{
"entropy": 5.109746885299683,
"epoch": 1.1656364851319747,
"grad_norm": 1.0546875,
"learning_rate": 0.00048690952443999,
"loss": 4.7033,
"mean_token_accuracy": 0.22849691212177276,
"num_tokens": 25670331.0,
"step": 13955
},
{
"entropy": 5.30588002204895,
"epoch": 1.1660541262946875,
"grad_norm": 1.046875,
"learning_rate": 0.00048689951637874445,
"loss": 4.9707,
"mean_token_accuracy": 0.21301737129688264,
"num_tokens": 25679276.0,
"step": 13960
},
{
"entropy": 5.317697286605835,
"epoch": 1.1664717674574006,
"grad_norm": 1.0703125,
"learning_rate": 0.0004868895046079428,
"loss": 4.8922,
"mean_token_accuracy": 0.21998478919267656,
"num_tokens": 25688509.0,
"step": 13965
},
{
"entropy": 5.221651983261109,
"epoch": 1.1668894086201136,
"grad_norm": 1.015625,
"learning_rate": 0.0004868794891277602,
"loss": 4.7439,
"mean_token_accuracy": 0.21961688250303268,
"num_tokens": 25697776.0,
"step": 13970
},
{
"entropy": 5.212842988967895,
"epoch": 1.1673070497828266,
"grad_norm": 0.99609375,
"learning_rate": 0.0004868694699383722,
"loss": 4.9242,
"mean_token_accuracy": 0.20620774775743483,
"num_tokens": 25707528.0,
"step": 13975
},
{
"entropy": 5.295796537399292,
"epoch": 1.1677246909455397,
"grad_norm": 0.9609375,
"learning_rate": 0.0004868594470399541,
"loss": 4.8571,
"mean_token_accuracy": 0.21744544357061385,
"num_tokens": 25716568.0,
"step": 13980
},
{
"entropy": 5.307967329025269,
"epoch": 1.1681423321082525,
"grad_norm": 1.0,
"learning_rate": 0.00048684942043268137,
"loss": 4.8975,
"mean_token_accuracy": 0.20828187465667725,
"num_tokens": 25726441.0,
"step": 13985
},
{
"entropy": 5.2155413150787355,
"epoch": 1.1685599732709655,
"grad_norm": 1.09375,
"learning_rate": 0.00048683939011672953,
"loss": 4.8562,
"mean_token_accuracy": 0.21843262165784835,
"num_tokens": 25735386.0,
"step": 13990
},
{
"entropy": 5.293379449844361,
"epoch": 1.1689776144336785,
"grad_norm": 1.1171875,
"learning_rate": 0.00048682935609227414,
"loss": 4.8439,
"mean_token_accuracy": 0.2108369439840317,
"num_tokens": 25744758.0,
"step": 13995
},
{
"entropy": 5.235113859176636,
"epoch": 1.1693952555963916,
"grad_norm": 1.1328125,
"learning_rate": 0.00048681931835949083,
"loss": 4.9004,
"mean_token_accuracy": 0.21151605248451233,
"num_tokens": 25753836.0,
"step": 14000
},
{
"entropy": 5.285754776000976,
"epoch": 1.1698128967591046,
"grad_norm": 0.87109375,
"learning_rate": 0.0004868092769185555,
"loss": 4.9311,
"mean_token_accuracy": 0.2012794241309166,
"num_tokens": 25763302.0,
"step": 14005
},
{
"entropy": 5.2695681095123295,
"epoch": 1.1702305379218176,
"grad_norm": 1.125,
"learning_rate": 0.00048679923176964384,
"loss": 4.7901,
"mean_token_accuracy": 0.22135670632123947,
"num_tokens": 25772204.0,
"step": 14010
},
{
"entropy": 5.264920616149903,
"epoch": 1.1706481790845307,
"grad_norm": 0.94140625,
"learning_rate": 0.0004867891829129316,
"loss": 4.8803,
"mean_token_accuracy": 0.21377789229154587,
"num_tokens": 25781762.0,
"step": 14015
},
{
"entropy": 5.265269565582275,
"epoch": 1.1710658202472435,
"grad_norm": 1.0625,
"learning_rate": 0.0004867791303485948,
"loss": 4.9393,
"mean_token_accuracy": 0.2100481480360031,
"num_tokens": 25791573.0,
"step": 14020
},
{
"entropy": 5.2634077072143555,
"epoch": 1.1714834614099565,
"grad_norm": 1.046875,
"learning_rate": 0.0004867690740768094,
"loss": 4.8807,
"mean_token_accuracy": 0.21862648278474808,
"num_tokens": 25800777.0,
"step": 14025
},
{
"entropy": 5.241231250762939,
"epoch": 1.1719011025726696,
"grad_norm": 1.0390625,
"learning_rate": 0.00048675901409775147,
"loss": 4.8383,
"mean_token_accuracy": 0.21715591251850128,
"num_tokens": 25809145.0,
"step": 14030
},
{
"entropy": 5.18926854133606,
"epoch": 1.1723187437353826,
"grad_norm": 1.0703125,
"learning_rate": 0.00048674895041159714,
"loss": 4.7746,
"mean_token_accuracy": 0.22451866865158082,
"num_tokens": 25817884.0,
"step": 14035
},
{
"entropy": 5.267607164382935,
"epoch": 1.1727363848980956,
"grad_norm": 1.046875,
"learning_rate": 0.00048673888301852247,
"loss": 4.9013,
"mean_token_accuracy": 0.21320959627628328,
"num_tokens": 25826953.0,
"step": 14040
},
{
"entropy": 5.325677919387817,
"epoch": 1.1731540260608084,
"grad_norm": 1.0234375,
"learning_rate": 0.0004867288119187039,
"loss": 4.9352,
"mean_token_accuracy": 0.21173907965421676,
"num_tokens": 25835597.0,
"step": 14045
},
{
"entropy": 5.281623268127442,
"epoch": 1.1735716672235215,
"grad_norm": 1.0859375,
"learning_rate": 0.0004867187371123175,
"loss": 4.9053,
"mean_token_accuracy": 0.20905899703502656,
"num_tokens": 25845072.0,
"step": 14050
},
{
"entropy": 5.266669988632202,
"epoch": 1.1739893083862345,
"grad_norm": 0.95703125,
"learning_rate": 0.00048670865859953984,
"loss": 4.8033,
"mean_token_accuracy": 0.2188041925430298,
"num_tokens": 25854902.0,
"step": 14055
},
{
"entropy": 5.308465433120728,
"epoch": 1.1744069495489475,
"grad_norm": 1.078125,
"learning_rate": 0.0004866985763805472,
"loss": 4.9355,
"mean_token_accuracy": 0.20951370149850845,
"num_tokens": 25863807.0,
"step": 14060
},
{
"entropy": 5.291536140441894,
"epoch": 1.1748245907116606,
"grad_norm": 1.109375,
"learning_rate": 0.0004866884904555162,
"loss": 4.9467,
"mean_token_accuracy": 0.2053670957684517,
"num_tokens": 25873709.0,
"step": 14065
},
{
"entropy": 5.2727251052856445,
"epoch": 1.1752422318743736,
"grad_norm": 0.921875,
"learning_rate": 0.0004866784008246233,
"loss": 4.9113,
"mean_token_accuracy": 0.21408339142799376,
"num_tokens": 25883425.0,
"step": 14070
},
{
"entropy": 5.2046891212463375,
"epoch": 1.1756598730370866,
"grad_norm": 1.0625,
"learning_rate": 0.0004866683074880453,
"loss": 4.7784,
"mean_token_accuracy": 0.22162298411130904,
"num_tokens": 25892317.0,
"step": 14075
},
{
"entropy": 5.247200393676758,
"epoch": 1.1760775141997994,
"grad_norm": 0.96484375,
"learning_rate": 0.00048665821044595884,
"loss": 4.8535,
"mean_token_accuracy": 0.2172433093190193,
"num_tokens": 25901733.0,
"step": 14080
},
{
"entropy": 5.204548740386963,
"epoch": 1.1764951553625125,
"grad_norm": 0.96875,
"learning_rate": 0.0004866481096985405,
"loss": 4.8304,
"mean_token_accuracy": 0.2176731452345848,
"num_tokens": 25911574.0,
"step": 14085
},
{
"entropy": 5.252797412872314,
"epoch": 1.1769127965252255,
"grad_norm": 0.96484375,
"learning_rate": 0.0004866380052459673,
"loss": 4.9261,
"mean_token_accuracy": 0.20866089463233947,
"num_tokens": 25921152.0,
"step": 14090
},
{
"entropy": 5.338256549835205,
"epoch": 1.1773304376879385,
"grad_norm": 1.015625,
"learning_rate": 0.0004866278970884161,
"loss": 4.8768,
"mean_token_accuracy": 0.2088471069931984,
"num_tokens": 25930455.0,
"step": 14095
},
{
"entropy": 5.236100149154663,
"epoch": 1.1777480788506516,
"grad_norm": 1.0703125,
"learning_rate": 0.00048661778522606376,
"loss": 4.8103,
"mean_token_accuracy": 0.21504587084054946,
"num_tokens": 25939298.0,
"step": 14100
},
{
"entropy": 5.218051290512085,
"epoch": 1.1781657200133644,
"grad_norm": 1.0625,
"learning_rate": 0.00048660766965908736,
"loss": 4.8645,
"mean_token_accuracy": 0.21135970205068588,
"num_tokens": 25948426.0,
"step": 14105
},
{
"entropy": 5.277697944641114,
"epoch": 1.1785833611760774,
"grad_norm": 1.0234375,
"learning_rate": 0.00048659755038766406,
"loss": 4.865,
"mean_token_accuracy": 0.21691957712173462,
"num_tokens": 25958469.0,
"step": 14110
},
{
"entropy": 5.249806594848633,
"epoch": 1.1790010023387905,
"grad_norm": 1.015625,
"learning_rate": 0.00048658742741197086,
"loss": 4.7767,
"mean_token_accuracy": 0.23051322400569915,
"num_tokens": 25967173.0,
"step": 14115
},
{
"entropy": 5.242997360229492,
"epoch": 1.1794186435015035,
"grad_norm": 0.9609375,
"learning_rate": 0.00048657730073218517,
"loss": 4.8929,
"mean_token_accuracy": 0.21696035861968993,
"num_tokens": 25976474.0,
"step": 14120
},
{
"entropy": 5.297246885299683,
"epoch": 1.1798362846642165,
"grad_norm": 1.0859375,
"learning_rate": 0.0004865671703484841,
"loss": 4.8732,
"mean_token_accuracy": 0.21733936965465545,
"num_tokens": 25984745.0,
"step": 14125
},
{
"entropy": 5.252665853500366,
"epoch": 1.1802539258269296,
"grad_norm": 1.0390625,
"learning_rate": 0.00048655703626104493,
"loss": 4.8108,
"mean_token_accuracy": 0.22127807438373565,
"num_tokens": 25993265.0,
"step": 14130
},
{
"entropy": 5.2750012397766115,
"epoch": 1.1806715669896426,
"grad_norm": 1.125,
"learning_rate": 0.00048654689847004535,
"loss": 4.846,
"mean_token_accuracy": 0.21467218399047852,
"num_tokens": 26002390.0,
"step": 14135
},
{
"entropy": 5.342982149124145,
"epoch": 1.1810892081523554,
"grad_norm": 1.0546875,
"learning_rate": 0.0004865367569756626,
"loss": 4.9883,
"mean_token_accuracy": 0.2171557754278183,
"num_tokens": 26013086.0,
"step": 14140
},
{
"entropy": 5.335472488403321,
"epoch": 1.1815068493150684,
"grad_norm": 0.99609375,
"learning_rate": 0.00048652661177807426,
"loss": 4.8893,
"mean_token_accuracy": 0.21468782275915146,
"num_tokens": 26022119.0,
"step": 14145
},
{
"entropy": 5.283700847625733,
"epoch": 1.1819244904777815,
"grad_norm": 1.0234375,
"learning_rate": 0.00048651646287745803,
"loss": 4.9287,
"mean_token_accuracy": 0.2100604310631752,
"num_tokens": 26031805.0,
"step": 14150
},
{
"entropy": 5.282706832885742,
"epoch": 1.1823421316404945,
"grad_norm": 0.9609375,
"learning_rate": 0.00048650631027399155,
"loss": 4.9003,
"mean_token_accuracy": 0.21225172132253647,
"num_tokens": 26040965.0,
"step": 14155
},
{
"entropy": 5.286909437179565,
"epoch": 1.1827597728032075,
"grad_norm": 1.0546875,
"learning_rate": 0.00048649615396785246,
"loss": 4.8993,
"mean_token_accuracy": 0.20916713625192643,
"num_tokens": 26049618.0,
"step": 14160
},
{
"entropy": 5.275590085983277,
"epoch": 1.1831774139659206,
"grad_norm": 1.0078125,
"learning_rate": 0.00048648599395921863,
"loss": 4.8472,
"mean_token_accuracy": 0.21483348459005355,
"num_tokens": 26058335.0,
"step": 14165
},
{
"entropy": 5.155168104171753,
"epoch": 1.1835950551286334,
"grad_norm": 1.0234375,
"learning_rate": 0.00048647583024826786,
"loss": 4.7323,
"mean_token_accuracy": 0.22726014703512193,
"num_tokens": 26067361.0,
"step": 14170
},
{
"entropy": 5.304991912841797,
"epoch": 1.1840126962913464,
"grad_norm": 0.9921875,
"learning_rate": 0.00048646566283517824,
"loss": 4.9452,
"mean_token_accuracy": 0.20840476602315902,
"num_tokens": 26077401.0,
"step": 14175
},
{
"entropy": 5.211455249786377,
"epoch": 1.1844303374540595,
"grad_norm": 1.1640625,
"learning_rate": 0.00048645549172012756,
"loss": 4.7743,
"mean_token_accuracy": 0.22259992063045503,
"num_tokens": 26086422.0,
"step": 14180
},
{
"entropy": 5.288153648376465,
"epoch": 1.1848479786167725,
"grad_norm": 0.9765625,
"learning_rate": 0.000486445316903294,
"loss": 4.9379,
"mean_token_accuracy": 0.2093539461493492,
"num_tokens": 26095974.0,
"step": 14185
},
{
"entropy": 5.237947940826416,
"epoch": 1.1852656197794855,
"grad_norm": 1.0078125,
"learning_rate": 0.0004864351383848556,
"loss": 4.8124,
"mean_token_accuracy": 0.22141705602407455,
"num_tokens": 26104782.0,
"step": 14190
},
{
"entropy": 5.28892765045166,
"epoch": 1.1856832609421986,
"grad_norm": 1.046875,
"learning_rate": 0.0004864249561649906,
"loss": 4.8966,
"mean_token_accuracy": 0.21697104275226592,
"num_tokens": 26113769.0,
"step": 14195
},
{
"entropy": 5.290374135971069,
"epoch": 1.1861009021049114,
"grad_norm": 1.0390625,
"learning_rate": 0.0004864147702438773,
"loss": 4.9337,
"mean_token_accuracy": 0.2121882975101471,
"num_tokens": 26123759.0,
"step": 14200
},
{
"entropy": 5.264079236984253,
"epoch": 1.1865185432676244,
"grad_norm": 0.97265625,
"learning_rate": 0.000486404580621694,
"loss": 4.8099,
"mean_token_accuracy": 0.21639186441898345,
"num_tokens": 26133392.0,
"step": 14205
},
{
"entropy": 5.316814517974853,
"epoch": 1.1869361844303374,
"grad_norm": 1.234375,
"learning_rate": 0.000486394387298619,
"loss": 4.9674,
"mean_token_accuracy": 0.2074139565229416,
"num_tokens": 26142102.0,
"step": 14210
},
{
"entropy": 5.231172037124634,
"epoch": 1.1873538255930505,
"grad_norm": 1.015625,
"learning_rate": 0.00048638419027483075,
"loss": 4.8411,
"mean_token_accuracy": 0.21798044592142105,
"num_tokens": 26151504.0,
"step": 14215
},
{
"entropy": 5.297414302825928,
"epoch": 1.1877714667557635,
"grad_norm": 1.03125,
"learning_rate": 0.0004863739895505079,
"loss": 4.9885,
"mean_token_accuracy": 0.20553924292325973,
"num_tokens": 26160717.0,
"step": 14220
},
{
"entropy": 5.229414701461792,
"epoch": 1.1881891079184765,
"grad_norm": 1.0546875,
"learning_rate": 0.00048636378512582884,
"loss": 4.8363,
"mean_token_accuracy": 0.21140021383762359,
"num_tokens": 26169788.0,
"step": 14225
},
{
"entropy": 5.254428577423096,
"epoch": 1.1886067490811894,
"grad_norm": 0.953125,
"learning_rate": 0.00048635357700097236,
"loss": 4.8238,
"mean_token_accuracy": 0.20885974615812303,
"num_tokens": 26178904.0,
"step": 14230
},
{
"entropy": 5.306118440628052,
"epoch": 1.1890243902439024,
"grad_norm": 0.9453125,
"learning_rate": 0.0004863433651761171,
"loss": 4.8986,
"mean_token_accuracy": 0.21250129491090775,
"num_tokens": 26188832.0,
"step": 14235
},
{
"entropy": 5.330101108551025,
"epoch": 1.1894420314066154,
"grad_norm": 1.09375,
"learning_rate": 0.0004863331496514418,
"loss": 4.9652,
"mean_token_accuracy": 0.21003050357103348,
"num_tokens": 26199003.0,
"step": 14240
},
{
"entropy": 5.351752281188965,
"epoch": 1.1898596725693285,
"grad_norm": 0.921875,
"learning_rate": 0.00048632293042712545,
"loss": 4.8394,
"mean_token_accuracy": 0.2104760080575943,
"num_tokens": 26209001.0,
"step": 14245
},
{
"entropy": 5.252216100692749,
"epoch": 1.1902773137320415,
"grad_norm": 1.078125,
"learning_rate": 0.00048631270750334674,
"loss": 4.8675,
"mean_token_accuracy": 0.2157977595925331,
"num_tokens": 26217834.0,
"step": 14250
},
{
"entropy": 5.29575629234314,
"epoch": 1.1906949548947545,
"grad_norm": 1.046875,
"learning_rate": 0.00048630248088028474,
"loss": 4.9222,
"mean_token_accuracy": 0.2151055946946144,
"num_tokens": 26226935.0,
"step": 14255
},
{
"entropy": 5.297951984405517,
"epoch": 1.1911125960574673,
"grad_norm": 1.0546875,
"learning_rate": 0.0004862922505581185,
"loss": 4.911,
"mean_token_accuracy": 0.2155863329768181,
"num_tokens": 26236874.0,
"step": 14260
},
{
"entropy": 5.291000509262085,
"epoch": 1.1915302372201804,
"grad_norm": 1.046875,
"learning_rate": 0.000486282016537027,
"loss": 4.8612,
"mean_token_accuracy": 0.2191222459077835,
"num_tokens": 26246361.0,
"step": 14265
},
{
"entropy": 5.197998094558716,
"epoch": 1.1919478783828934,
"grad_norm": 1.0703125,
"learning_rate": 0.00048627177881718954,
"loss": 4.8108,
"mean_token_accuracy": 0.22168292850255966,
"num_tokens": 26255316.0,
"step": 14270
},
{
"entropy": 5.401647520065308,
"epoch": 1.1923655195456064,
"grad_norm": 1.0234375,
"learning_rate": 0.0004862615373987853,
"loss": 4.9469,
"mean_token_accuracy": 0.20879973322153092,
"num_tokens": 26265250.0,
"step": 14275
},
{
"entropy": 5.207538509368897,
"epoch": 1.1927831607083195,
"grad_norm": 1.0859375,
"learning_rate": 0.0004862512922819935,
"loss": 4.7237,
"mean_token_accuracy": 0.2249722436070442,
"num_tokens": 26273858.0,
"step": 14280
},
{
"entropy": 5.28621015548706,
"epoch": 1.1932008018710325,
"grad_norm": 1.078125,
"learning_rate": 0.0004862410434669936,
"loss": 4.9742,
"mean_token_accuracy": 0.21151680797338485,
"num_tokens": 26283536.0,
"step": 14285
},
{
"entropy": 5.270149993896484,
"epoch": 1.1936184430337453,
"grad_norm": 1.15625,
"learning_rate": 0.00048623079095396496,
"loss": 4.8006,
"mean_token_accuracy": 0.21966674476861953,
"num_tokens": 26291761.0,
"step": 14290
},
{
"entropy": 5.2545617580413815,
"epoch": 1.1940360841964583,
"grad_norm": 0.9921875,
"learning_rate": 0.000486220534743087,
"loss": 4.848,
"mean_token_accuracy": 0.20726076662540435,
"num_tokens": 26301082.0,
"step": 14295
},
{
"entropy": 5.248863792419433,
"epoch": 1.1944537253591714,
"grad_norm": 1.15625,
"learning_rate": 0.0004862102748345394,
"loss": 4.9058,
"mean_token_accuracy": 0.20632270276546477,
"num_tokens": 26310424.0,
"step": 14300
},
{
"entropy": 5.287345743179321,
"epoch": 1.1948713665218844,
"grad_norm": 1.0234375,
"learning_rate": 0.00048620001122850164,
"loss": 4.969,
"mean_token_accuracy": 0.21015441715717315,
"num_tokens": 26319442.0,
"step": 14305
},
{
"entropy": 5.226160764694214,
"epoch": 1.1952890076845974,
"grad_norm": 1.15625,
"learning_rate": 0.0004861897439251535,
"loss": 4.7999,
"mean_token_accuracy": 0.21991005092859267,
"num_tokens": 26327757.0,
"step": 14310
},
{
"entropy": 5.269315147399903,
"epoch": 1.1957066488473105,
"grad_norm": 1.140625,
"learning_rate": 0.00048617947292467473,
"loss": 4.8532,
"mean_token_accuracy": 0.21732535660266877,
"num_tokens": 26337327.0,
"step": 14315
},
{
"entropy": 5.305079460144043,
"epoch": 1.1961242900100233,
"grad_norm": 1.1171875,
"learning_rate": 0.00048616919822724506,
"loss": 4.9271,
"mean_token_accuracy": 0.21783028393983841,
"num_tokens": 26345821.0,
"step": 14320
},
{
"entropy": 5.288415336608887,
"epoch": 1.1965419311727363,
"grad_norm": 1.0,
"learning_rate": 0.00048615891983304436,
"loss": 4.881,
"mean_token_accuracy": 0.21573809832334517,
"num_tokens": 26355423.0,
"step": 14325
},
{
"entropy": 5.236432313919067,
"epoch": 1.1969595723354494,
"grad_norm": 1.1015625,
"learning_rate": 0.0004861486377422526,
"loss": 4.762,
"mean_token_accuracy": 0.21987208873033523,
"num_tokens": 26363770.0,
"step": 14330
},
{
"entropy": 5.220354413986206,
"epoch": 1.1973772134981624,
"grad_norm": 0.9765625,
"learning_rate": 0.0004861383519550498,
"loss": 4.8173,
"mean_token_accuracy": 0.21819413751363753,
"num_tokens": 26373530.0,
"step": 14335
},
{
"entropy": 5.348572683334351,
"epoch": 1.1977948546608754,
"grad_norm": 1.0,
"learning_rate": 0.00048612806247161594,
"loss": 4.9762,
"mean_token_accuracy": 0.20434900969266892,
"num_tokens": 26382960.0,
"step": 14340
},
{
"entropy": 5.22479100227356,
"epoch": 1.1982124958235885,
"grad_norm": 1.046875,
"learning_rate": 0.0004861177692921313,
"loss": 4.7668,
"mean_token_accuracy": 0.22742414921522142,
"num_tokens": 26392105.0,
"step": 14345
},
{
"entropy": 5.240310907363892,
"epoch": 1.1986301369863013,
"grad_norm": 1.1953125,
"learning_rate": 0.00048610747241677595,
"loss": 4.8797,
"mean_token_accuracy": 0.21708065271377563,
"num_tokens": 26401780.0,
"step": 14350
},
{
"entropy": 5.245091819763184,
"epoch": 1.1990477781490143,
"grad_norm": 1.0390625,
"learning_rate": 0.0004860971718457302,
"loss": 4.8752,
"mean_token_accuracy": 0.21635661125183106,
"num_tokens": 26410290.0,
"step": 14355
},
{
"entropy": 5.303373384475708,
"epoch": 1.1994654193117273,
"grad_norm": 0.96484375,
"learning_rate": 0.0004860868675791743,
"loss": 4.8997,
"mean_token_accuracy": 0.21623397916555404,
"num_tokens": 26420591.0,
"step": 14360
},
{
"entropy": 5.341029071807862,
"epoch": 1.1998830604744404,
"grad_norm": 1.15625,
"learning_rate": 0.00048607655961728876,
"loss": 4.9195,
"mean_token_accuracy": 0.20581527650356293,
"num_tokens": 26429453.0,
"step": 14365
},
{
"entropy": 5.3321247577667235,
"epoch": 1.2003007016371534,
"grad_norm": 1.0703125,
"learning_rate": 0.000486066247960254,
"loss": 4.9075,
"mean_token_accuracy": 0.21396533399820328,
"num_tokens": 26439111.0,
"step": 14370
},
{
"entropy": 5.258881998062134,
"epoch": 1.2007183427998664,
"grad_norm": 0.9921875,
"learning_rate": 0.00048605593260825044,
"loss": 4.821,
"mean_token_accuracy": 0.22210421711206435,
"num_tokens": 26448071.0,
"step": 14375
},
{
"entropy": 5.3353794574737545,
"epoch": 1.2011359839625793,
"grad_norm": 1.078125,
"learning_rate": 0.00048604561356145875,
"loss": 4.9918,
"mean_token_accuracy": 0.2131146863102913,
"num_tokens": 26457226.0,
"step": 14380
},
{
"entropy": 5.2923667430877686,
"epoch": 1.2015536251252923,
"grad_norm": 1.078125,
"learning_rate": 0.00048603529082005955,
"loss": 4.872,
"mean_token_accuracy": 0.21626408249139786,
"num_tokens": 26466052.0,
"step": 14385
},
{
"entropy": 5.32365870475769,
"epoch": 1.2019712662880053,
"grad_norm": 1.15625,
"learning_rate": 0.00048602496438423354,
"loss": 4.8826,
"mean_token_accuracy": 0.20888697803020478,
"num_tokens": 26476102.0,
"step": 14390
},
{
"entropy": 5.256476926803589,
"epoch": 1.2023889074507184,
"grad_norm": 1.1328125,
"learning_rate": 0.0004860146342541616,
"loss": 4.8823,
"mean_token_accuracy": 0.21264505684375762,
"num_tokens": 26485775.0,
"step": 14395
},
{
"entropy": 5.254738426208496,
"epoch": 1.2028065486134314,
"grad_norm": 1.0078125,
"learning_rate": 0.0004860043004300244,
"loss": 4.8531,
"mean_token_accuracy": 0.21507709920406343,
"num_tokens": 26495268.0,
"step": 14400
},
{
"entropy": 5.343999624252319,
"epoch": 1.2032241897761444,
"grad_norm": 0.98828125,
"learning_rate": 0.000485993962912003,
"loss": 4.9144,
"mean_token_accuracy": 0.21374761015176774,
"num_tokens": 26505253.0,
"step": 14405
},
{
"entropy": 5.269192314147949,
"epoch": 1.2036418309388572,
"grad_norm": 1.1484375,
"learning_rate": 0.0004859836217002783,
"loss": 4.8622,
"mean_token_accuracy": 0.21690179109573365,
"num_tokens": 26513609.0,
"step": 14410
},
{
"entropy": 5.308850240707398,
"epoch": 1.2040594721015703,
"grad_norm": 0.97265625,
"learning_rate": 0.0004859732767950313,
"loss": 4.9351,
"mean_token_accuracy": 0.2087784841656685,
"num_tokens": 26523565.0,
"step": 14415
},
{
"entropy": 5.275409936904907,
"epoch": 1.2044771132642833,
"grad_norm": 0.93359375,
"learning_rate": 0.00048596292819644315,
"loss": 4.8205,
"mean_token_accuracy": 0.22033887058496476,
"num_tokens": 26532554.0,
"step": 14420
},
{
"entropy": 5.253980016708374,
"epoch": 1.2048947544269963,
"grad_norm": 1.1328125,
"learning_rate": 0.000485952575904695,
"loss": 4.9531,
"mean_token_accuracy": 0.21058446615934373,
"num_tokens": 26541474.0,
"step": 14425
},
{
"entropy": 5.228180980682373,
"epoch": 1.2053123955897094,
"grad_norm": 1.078125,
"learning_rate": 0.00048594221991996814,
"loss": 4.8244,
"mean_token_accuracy": 0.21837536096572877,
"num_tokens": 26551095.0,
"step": 14430
},
{
"entropy": 5.242866849899292,
"epoch": 1.2057300367524224,
"grad_norm": 1.015625,
"learning_rate": 0.00048593186024244373,
"loss": 4.8243,
"mean_token_accuracy": 0.2279058039188385,
"num_tokens": 26560144.0,
"step": 14435
},
{
"entropy": 5.1893613815307615,
"epoch": 1.2061476779151352,
"grad_norm": 1.09375,
"learning_rate": 0.0004859214968723033,
"loss": 4.7593,
"mean_token_accuracy": 0.22170279920101166,
"num_tokens": 26568983.0,
"step": 14440
},
{
"entropy": 5.220506286621093,
"epoch": 1.2065653190778483,
"grad_norm": 1.1484375,
"learning_rate": 0.00048591112980972813,
"loss": 4.8049,
"mean_token_accuracy": 0.2230754479765892,
"num_tokens": 26577229.0,
"step": 14445
},
{
"entropy": 5.30432620048523,
"epoch": 1.2069829602405613,
"grad_norm": 1.1015625,
"learning_rate": 0.0004859007590548997,
"loss": 4.8725,
"mean_token_accuracy": 0.21160688549280166,
"num_tokens": 26585567.0,
"step": 14450
},
{
"entropy": 5.264838552474975,
"epoch": 1.2074006014032743,
"grad_norm": 1.0390625,
"learning_rate": 0.00048589038460799974,
"loss": 4.908,
"mean_token_accuracy": 0.21685008555650712,
"num_tokens": 26594656.0,
"step": 14455
},
{
"entropy": 5.2557610988616945,
"epoch": 1.2078182425659874,
"grad_norm": 1.0703125,
"learning_rate": 0.0004858800064692097,
"loss": 4.881,
"mean_token_accuracy": 0.20496947765350343,
"num_tokens": 26604002.0,
"step": 14460
},
{
"entropy": 5.276022720336914,
"epoch": 1.2082358837287004,
"grad_norm": 0.953125,
"learning_rate": 0.00048586962463871124,
"loss": 4.8622,
"mean_token_accuracy": 0.21516236662864685,
"num_tokens": 26613376.0,
"step": 14465
},
{
"entropy": 5.401705408096314,
"epoch": 1.2086535248914134,
"grad_norm": 1.0,
"learning_rate": 0.00048585923911668625,
"loss": 5.0122,
"mean_token_accuracy": 0.20901001691818238,
"num_tokens": 26623117.0,
"step": 14470
},
{
"entropy": 5.281592750549317,
"epoch": 1.2090711660541262,
"grad_norm": 0.9921875,
"learning_rate": 0.00048584884990331644,
"loss": 4.9096,
"mean_token_accuracy": 0.2111772269010544,
"num_tokens": 26632855.0,
"step": 14475
},
{
"entropy": 5.165176439285278,
"epoch": 1.2094888072168393,
"grad_norm": 0.98828125,
"learning_rate": 0.00048583845699878375,
"loss": 4.6867,
"mean_token_accuracy": 0.23215150088071823,
"num_tokens": 26641753.0,
"step": 14480
},
{
"entropy": 5.1794507026672365,
"epoch": 1.2099064483795523,
"grad_norm": 0.9765625,
"learning_rate": 0.0004858280604032701,
"loss": 4.8279,
"mean_token_accuracy": 0.22068871408700944,
"num_tokens": 26650278.0,
"step": 14485
},
{
"entropy": 5.291289234161377,
"epoch": 1.2103240895422653,
"grad_norm": 1.0234375,
"learning_rate": 0.00048581766011695745,
"loss": 4.893,
"mean_token_accuracy": 0.22035497128963472,
"num_tokens": 26658891.0,
"step": 14490
},
{
"entropy": 5.288300848007202,
"epoch": 1.2107417307049784,
"grad_norm": 0.96875,
"learning_rate": 0.00048580725614002785,
"loss": 4.9197,
"mean_token_accuracy": 0.20947515815496445,
"num_tokens": 26668197.0,
"step": 14495
},
{
"entropy": 5.328019523620606,
"epoch": 1.2111593718676912,
"grad_norm": 1.1015625,
"learning_rate": 0.00048579684847266353,
"loss": 4.948,
"mean_token_accuracy": 0.20997326374053954,
"num_tokens": 26676796.0,
"step": 14500
},
{
"entropy": 5.226969242095947,
"epoch": 1.2115770130304042,
"grad_norm": 1.0859375,
"learning_rate": 0.0004857864371150467,
"loss": 4.8267,
"mean_token_accuracy": 0.21793274134397506,
"num_tokens": 26686237.0,
"step": 14505
},
{
"entropy": 5.285734176635742,
"epoch": 1.2119946541931172,
"grad_norm": 1.0234375,
"learning_rate": 0.00048577602206735943,
"loss": 4.8157,
"mean_token_accuracy": 0.22215263843536376,
"num_tokens": 26695931.0,
"step": 14510
},
{
"entropy": 5.214428901672363,
"epoch": 1.2124122953558303,
"grad_norm": 1.09375,
"learning_rate": 0.0004857656033297843,
"loss": 4.8047,
"mean_token_accuracy": 0.2246819704771042,
"num_tokens": 26704326.0,
"step": 14515
},
{
"entropy": 5.197016668319702,
"epoch": 1.2128299365185433,
"grad_norm": 1.125,
"learning_rate": 0.00048575518090250347,
"loss": 4.7335,
"mean_token_accuracy": 0.2224571079015732,
"num_tokens": 26713000.0,
"step": 14520
},
{
"entropy": 5.296850776672363,
"epoch": 1.2132475776812564,
"grad_norm": 1.078125,
"learning_rate": 0.0004857447547856996,
"loss": 4.8821,
"mean_token_accuracy": 0.2097340404987335,
"num_tokens": 26722014.0,
"step": 14525
},
{
"entropy": 5.281072378158569,
"epoch": 1.2136652188439694,
"grad_norm": 1.0078125,
"learning_rate": 0.0004857343249795552,
"loss": 4.8443,
"mean_token_accuracy": 0.21931581795215607,
"num_tokens": 26730975.0,
"step": 14530
},
{
"entropy": 5.207539176940918,
"epoch": 1.2140828600066822,
"grad_norm": 0.97265625,
"learning_rate": 0.0004857238914842526,
"loss": 4.9063,
"mean_token_accuracy": 0.21440516412258148,
"num_tokens": 26740087.0,
"step": 14535
},
{
"entropy": 5.2557408809661865,
"epoch": 1.2145005011693952,
"grad_norm": 0.9296875,
"learning_rate": 0.00048571345429997473,
"loss": 4.906,
"mean_token_accuracy": 0.20803391933441162,
"num_tokens": 26749820.0,
"step": 14540
},
{
"entropy": 5.33833065032959,
"epoch": 1.2149181423321083,
"grad_norm": 1.09375,
"learning_rate": 0.00048570301342690416,
"loss": 4.9662,
"mean_token_accuracy": 0.21338541060686111,
"num_tokens": 26759850.0,
"step": 14545
},
{
"entropy": 5.213677930831909,
"epoch": 1.2153357834948213,
"grad_norm": 0.984375,
"learning_rate": 0.0004856925688652238,
"loss": 4.7242,
"mean_token_accuracy": 0.22617016583681107,
"num_tokens": 26769519.0,
"step": 14550
},
{
"entropy": 5.297413015365601,
"epoch": 1.2157534246575343,
"grad_norm": 0.96875,
"learning_rate": 0.00048568212061511634,
"loss": 4.9121,
"mean_token_accuracy": 0.20078637450933456,
"num_tokens": 26779999.0,
"step": 14555
},
{
"entropy": 5.3005987167358395,
"epoch": 1.2161710658202471,
"grad_norm": 1.03125,
"learning_rate": 0.00048567166867676475,
"loss": 4.858,
"mean_token_accuracy": 0.21063812226057052,
"num_tokens": 26788903.0,
"step": 14560
},
{
"entropy": 5.3428626537323,
"epoch": 1.2165887069829602,
"grad_norm": 1.03125,
"learning_rate": 0.00048566121305035207,
"loss": 4.939,
"mean_token_accuracy": 0.2097403198480606,
"num_tokens": 26799161.0,
"step": 14565
},
{
"entropy": 5.239137029647827,
"epoch": 1.2170063481456732,
"grad_norm": 1.0078125,
"learning_rate": 0.00048565075373606126,
"loss": 4.811,
"mean_token_accuracy": 0.21708568036556244,
"num_tokens": 26807887.0,
"step": 14570
},
{
"entropy": 5.270034408569336,
"epoch": 1.2174239893083862,
"grad_norm": 1.046875,
"learning_rate": 0.00048564029073407545,
"loss": 4.8104,
"mean_token_accuracy": 0.2199506789445877,
"num_tokens": 26816466.0,
"step": 14575
},
{
"entropy": 5.280757808685303,
"epoch": 1.2178416304710993,
"grad_norm": 1.0859375,
"learning_rate": 0.0004856298240445778,
"loss": 4.9463,
"mean_token_accuracy": 0.210982109606266,
"num_tokens": 26825847.0,
"step": 14580
},
{
"entropy": 5.328578090667724,
"epoch": 1.2182592716338123,
"grad_norm": 0.9609375,
"learning_rate": 0.0004856193536677515,
"loss": 4.9216,
"mean_token_accuracy": 0.2063239708542824,
"num_tokens": 26835610.0,
"step": 14585
},
{
"entropy": 5.318467712402343,
"epoch": 1.2186769127965253,
"grad_norm": 0.9765625,
"learning_rate": 0.00048560887960377995,
"loss": 4.8243,
"mean_token_accuracy": 0.2136611044406891,
"num_tokens": 26844319.0,
"step": 14590
},
{
"entropy": 5.2986814975738525,
"epoch": 1.2190945539592382,
"grad_norm": 1.0,
"learning_rate": 0.00048559840185284647,
"loss": 4.9282,
"mean_token_accuracy": 0.20739062279462814,
"num_tokens": 26853508.0,
"step": 14595
},
{
"entropy": 5.264918565750122,
"epoch": 1.2195121951219512,
"grad_norm": 1.0390625,
"learning_rate": 0.00048558792041513446,
"loss": 4.8765,
"mean_token_accuracy": 0.21719129085540773,
"num_tokens": 26863268.0,
"step": 14600
},
{
"entropy": 5.291875219345092,
"epoch": 1.2199298362846642,
"grad_norm": 1.0546875,
"learning_rate": 0.0004855774352908273,
"loss": 4.8684,
"mean_token_accuracy": 0.21217106878757477,
"num_tokens": 26871879.0,
"step": 14605
},
{
"entropy": 5.270335721969604,
"epoch": 1.2203474774473773,
"grad_norm": 1.1015625,
"learning_rate": 0.0004855669464801088,
"loss": 4.9724,
"mean_token_accuracy": 0.20962318778038025,
"num_tokens": 26880920.0,
"step": 14610
},
{
"entropy": 5.334642839431763,
"epoch": 1.2207651186100903,
"grad_norm": 1.0546875,
"learning_rate": 0.00048555645398316237,
"loss": 4.9428,
"mean_token_accuracy": 0.2118311494588852,
"num_tokens": 26890020.0,
"step": 14615
},
{
"entropy": 5.325588941574097,
"epoch": 1.221182759772803,
"grad_norm": 1.03125,
"learning_rate": 0.0004855459578001718,
"loss": 4.8642,
"mean_token_accuracy": 0.20544182062149047,
"num_tokens": 26899486.0,
"step": 14620
},
{
"entropy": 5.249404764175415,
"epoch": 1.2216004009355161,
"grad_norm": 1.09375,
"learning_rate": 0.0004855354579313208,
"loss": 4.8697,
"mean_token_accuracy": 0.21859646141529082,
"num_tokens": 26908298.0,
"step": 14625
},
{
"entropy": 5.357993173599243,
"epoch": 1.2220180420982292,
"grad_norm": 1.046875,
"learning_rate": 0.0004855249543767932,
"loss": 4.8953,
"mean_token_accuracy": 0.21050601899623872,
"num_tokens": 26917601.0,
"step": 14630
},
{
"entropy": 5.241923189163208,
"epoch": 1.2224356832609422,
"grad_norm": 1.0234375,
"learning_rate": 0.00048551444713677285,
"loss": 4.7451,
"mean_token_accuracy": 0.22693941593170167,
"num_tokens": 26926368.0,
"step": 14635
},
{
"entropy": 5.255865287780762,
"epoch": 1.2228533244236552,
"grad_norm": 1.1953125,
"learning_rate": 0.00048550393621144374,
"loss": 4.8763,
"mean_token_accuracy": 0.2065172925591469,
"num_tokens": 26935424.0,
"step": 14640
},
{
"entropy": 5.239245986938476,
"epoch": 1.2232709655863683,
"grad_norm": 1.015625,
"learning_rate": 0.0004854934216009898,
"loss": 4.8085,
"mean_token_accuracy": 0.2216997116804123,
"num_tokens": 26944906.0,
"step": 14645
},
{
"entropy": 5.242288827896118,
"epoch": 1.2236886067490813,
"grad_norm": 1.0234375,
"learning_rate": 0.0004854829033055951,
"loss": 4.7861,
"mean_token_accuracy": 0.22614532709121704,
"num_tokens": 26953709.0,
"step": 14650
},
{
"entropy": 5.204398965835571,
"epoch": 1.2241062479117941,
"grad_norm": 1.03125,
"learning_rate": 0.0004854723813254439,
"loss": 4.8508,
"mean_token_accuracy": 0.2162538304924965,
"num_tokens": 26962726.0,
"step": 14655
},
{
"entropy": 5.314360761642456,
"epoch": 1.2245238890745072,
"grad_norm": 1.015625,
"learning_rate": 0.0004854618556607203,
"loss": 4.9319,
"mean_token_accuracy": 0.20951235592365264,
"num_tokens": 26972519.0,
"step": 14660
},
{
"entropy": 5.294365882873535,
"epoch": 1.2249415302372202,
"grad_norm": 1.1328125,
"learning_rate": 0.00048545132631160856,
"loss": 4.8268,
"mean_token_accuracy": 0.21148503720760345,
"num_tokens": 26981251.0,
"step": 14665
},
{
"entropy": 5.241532516479492,
"epoch": 1.2253591713999332,
"grad_norm": 1.0078125,
"learning_rate": 0.0004854407932782931,
"loss": 4.8318,
"mean_token_accuracy": 0.21605629622936248,
"num_tokens": 26990707.0,
"step": 14670
},
{
"entropy": 5.240707588195801,
"epoch": 1.2257768125626463,
"grad_norm": 0.98828125,
"learning_rate": 0.0004854302565609582,
"loss": 4.8316,
"mean_token_accuracy": 0.21615566313266754,
"num_tokens": 26999364.0,
"step": 14675
},
{
"entropy": 5.3227087497711185,
"epoch": 1.226194453725359,
"grad_norm": 1.0390625,
"learning_rate": 0.00048541971615978833,
"loss": 4.9077,
"mean_token_accuracy": 0.2074078366160393,
"num_tokens": 27008757.0,
"step": 14680
},
{
"entropy": 5.279827499389649,
"epoch": 1.226612094888072,
"grad_norm": 1.0,
"learning_rate": 0.00048540917207496806,
"loss": 4.8084,
"mean_token_accuracy": 0.22025478035211563,
"num_tokens": 27017244.0,
"step": 14685
},
{
"entropy": 5.2828083515167235,
"epoch": 1.2270297360507851,
"grad_norm": 0.98828125,
"learning_rate": 0.00048539862430668204,
"loss": 4.8978,
"mean_token_accuracy": 0.2113216295838356,
"num_tokens": 27026527.0,
"step": 14690
},
{
"entropy": 5.230763006210327,
"epoch": 1.2274473772134982,
"grad_norm": 1.0390625,
"learning_rate": 0.00048538807285511474,
"loss": 4.8334,
"mean_token_accuracy": 0.22120669782161712,
"num_tokens": 27034890.0,
"step": 14695
},
{
"entropy": 5.236389970779419,
"epoch": 1.2278650183762112,
"grad_norm": 1.09375,
"learning_rate": 0.00048537751772045103,
"loss": 4.8132,
"mean_token_accuracy": 0.2269018843770027,
"num_tokens": 27044122.0,
"step": 14700
},
{
"entropy": 5.288397645950317,
"epoch": 1.2282826595389242,
"grad_norm": 0.9609375,
"learning_rate": 0.0004853669589028757,
"loss": 4.8597,
"mean_token_accuracy": 0.21395805329084397,
"num_tokens": 27053929.0,
"step": 14705
},
{
"entropy": 5.285385274887085,
"epoch": 1.2287003007016373,
"grad_norm": 0.9296875,
"learning_rate": 0.0004853563964025734,
"loss": 4.8304,
"mean_token_accuracy": 0.21364526450634003,
"num_tokens": 27062914.0,
"step": 14710
},
{
"entropy": 5.221350860595703,
"epoch": 1.22911794186435,
"grad_norm": 1.09375,
"learning_rate": 0.00048534583021972926,
"loss": 4.8615,
"mean_token_accuracy": 0.21342750042676925,
"num_tokens": 27072121.0,
"step": 14715
},
{
"entropy": 5.228186130523682,
"epoch": 1.2295355830270631,
"grad_norm": 0.93359375,
"learning_rate": 0.0004853352603545281,
"loss": 4.8203,
"mean_token_accuracy": 0.21485669165849686,
"num_tokens": 27081753.0,
"step": 14720
},
{
"entropy": 5.280810594558716,
"epoch": 1.2299532241897762,
"grad_norm": 1.0859375,
"learning_rate": 0.0004853246868071551,
"loss": 4.8748,
"mean_token_accuracy": 0.21020575016736984,
"num_tokens": 27090942.0,
"step": 14725
},
{
"entropy": 5.261235618591309,
"epoch": 1.2303708653524892,
"grad_norm": 1.03125,
"learning_rate": 0.00048531410957779533,
"loss": 4.9003,
"mean_token_accuracy": 0.2128758653998375,
"num_tokens": 27100620.0,
"step": 14730
},
{
"entropy": 5.363924551010132,
"epoch": 1.2307885065152022,
"grad_norm": 0.97265625,
"learning_rate": 0.0004853035286666338,
"loss": 4.92,
"mean_token_accuracy": 0.21257224828004836,
"num_tokens": 27110370.0,
"step": 14735
},
{
"entropy": 5.269014167785644,
"epoch": 1.231206147677915,
"grad_norm": 1.0390625,
"learning_rate": 0.000485292944073856,
"loss": 4.8138,
"mean_token_accuracy": 0.22502953857183455,
"num_tokens": 27119163.0,
"step": 14740
},
{
"entropy": 5.210812568664551,
"epoch": 1.231623788840628,
"grad_norm": 1.0234375,
"learning_rate": 0.000485282355799647,
"loss": 4.7907,
"mean_token_accuracy": 0.22059214860200882,
"num_tokens": 27127855.0,
"step": 14745
},
{
"entropy": 5.275664520263672,
"epoch": 1.232041430003341,
"grad_norm": 1.0234375,
"learning_rate": 0.00048527176384419217,
"loss": 4.8534,
"mean_token_accuracy": 0.21228248924016951,
"num_tokens": 27136932.0,
"step": 14750
},
{
"entropy": 5.2437896728515625,
"epoch": 1.2324590711660541,
"grad_norm": 0.953125,
"learning_rate": 0.00048526116820767713,
"loss": 4.8403,
"mean_token_accuracy": 0.21661580353975296,
"num_tokens": 27146371.0,
"step": 14755
},
{
"entropy": 5.243443870544434,
"epoch": 1.2328767123287672,
"grad_norm": 1.03125,
"learning_rate": 0.0004852505688902872,
"loss": 4.8527,
"mean_token_accuracy": 0.2159963682293892,
"num_tokens": 27156473.0,
"step": 14760
},
{
"entropy": 5.392106342315674,
"epoch": 1.2332943534914802,
"grad_norm": 0.984375,
"learning_rate": 0.0004852399658922079,
"loss": 4.9785,
"mean_token_accuracy": 0.20926444679498674,
"num_tokens": 27165915.0,
"step": 14765
},
{
"entropy": 5.201574039459229,
"epoch": 1.2337119946541932,
"grad_norm": 1.109375,
"learning_rate": 0.0004852293592136251,
"loss": 4.7785,
"mean_token_accuracy": 0.22138311862945556,
"num_tokens": 27174792.0,
"step": 14770
},
{
"entropy": 5.265687799453735,
"epoch": 1.234129635816906,
"grad_norm": 1.0546875,
"learning_rate": 0.0004852187488547242,
"loss": 4.8399,
"mean_token_accuracy": 0.22009023576974868,
"num_tokens": 27183785.0,
"step": 14775
},
{
"entropy": 5.356082820892334,
"epoch": 1.234547276979619,
"grad_norm": 1.0546875,
"learning_rate": 0.00048520813481569116,
"loss": 5.0137,
"mean_token_accuracy": 0.206134232878685,
"num_tokens": 27193599.0,
"step": 14780
},
{
"entropy": 5.337889528274536,
"epoch": 1.2349649181423321,
"grad_norm": 1.0546875,
"learning_rate": 0.0004851975170967116,
"loss": 4.9312,
"mean_token_accuracy": 0.21179380267858505,
"num_tokens": 27202191.0,
"step": 14785
},
{
"entropy": 5.267582368850708,
"epoch": 1.2353825593050451,
"grad_norm": 1.078125,
"learning_rate": 0.0004851868956979715,
"loss": 4.9045,
"mean_token_accuracy": 0.21012948155403138,
"num_tokens": 27210833.0,
"step": 14790
},
{
"entropy": 5.234970045089722,
"epoch": 1.2358002004677582,
"grad_norm": 1.0859375,
"learning_rate": 0.0004851762706196568,
"loss": 4.8332,
"mean_token_accuracy": 0.20961230099201203,
"num_tokens": 27220053.0,
"step": 14795
},
{
"entropy": 5.330142545700073,
"epoch": 1.236217841630471,
"grad_norm": 1.0390625,
"learning_rate": 0.00048516564186195353,
"loss": 4.9292,
"mean_token_accuracy": 0.214675073325634,
"num_tokens": 27229370.0,
"step": 14800
},
{
"entropy": 5.155725288391113,
"epoch": 1.236635482793184,
"grad_norm": 1.0546875,
"learning_rate": 0.00048515500942504763,
"loss": 4.7821,
"mean_token_accuracy": 0.22688209712505342,
"num_tokens": 27238923.0,
"step": 14805
},
{
"entropy": 5.2777024269104,
"epoch": 1.237053123955897,
"grad_norm": 1.09375,
"learning_rate": 0.00048514437330912544,
"loss": 4.8862,
"mean_token_accuracy": 0.21761640459299086,
"num_tokens": 27247929.0,
"step": 14810
},
{
"entropy": 5.276091384887695,
"epoch": 1.23747076511861,
"grad_norm": 1.046875,
"learning_rate": 0.00048513373351437294,
"loss": 4.841,
"mean_token_accuracy": 0.21699076890945435,
"num_tokens": 27257455.0,
"step": 14815
},
{
"entropy": 5.258977317810059,
"epoch": 1.2378884062813231,
"grad_norm": 0.96875,
"learning_rate": 0.00048512309004097653,
"loss": 4.8951,
"mean_token_accuracy": 0.20860559791326522,
"num_tokens": 27267718.0,
"step": 14820
},
{
"entropy": 5.23208794593811,
"epoch": 1.2383060474440362,
"grad_norm": 1.015625,
"learning_rate": 0.0004851124428891226,
"loss": 4.8066,
"mean_token_accuracy": 0.21790262907743455,
"num_tokens": 27276706.0,
"step": 14825
},
{
"entropy": 5.212944555282593,
"epoch": 1.2387236886067492,
"grad_norm": 1.0078125,
"learning_rate": 0.0004851017920589973,
"loss": 4.7239,
"mean_token_accuracy": 0.22478490322828293,
"num_tokens": 27285389.0,
"step": 14830
},
{
"entropy": 5.291102647781372,
"epoch": 1.239141329769462,
"grad_norm": 1.125,
"learning_rate": 0.0004850911375507873,
"loss": 4.9061,
"mean_token_accuracy": 0.21679034382104873,
"num_tokens": 27295529.0,
"step": 14835
},
{
"entropy": 5.310514068603515,
"epoch": 1.239558970932175,
"grad_norm": 1.0546875,
"learning_rate": 0.00048508047936467907,
"loss": 4.9502,
"mean_token_accuracy": 0.21269596815109254,
"num_tokens": 27305358.0,
"step": 14840
},
{
"entropy": 5.237136363983154,
"epoch": 1.239976612094888,
"grad_norm": 1.0390625,
"learning_rate": 0.00048506981750085914,
"loss": 4.7765,
"mean_token_accuracy": 0.2210938185453415,
"num_tokens": 27314765.0,
"step": 14845
},
{
"entropy": 5.303550910949707,
"epoch": 1.240394253257601,
"grad_norm": 1.0859375,
"learning_rate": 0.00048505915195951423,
"loss": 4.8592,
"mean_token_accuracy": 0.2133936420083046,
"num_tokens": 27324227.0,
"step": 14850
},
{
"entropy": 5.3168079376220705,
"epoch": 1.2408118944203141,
"grad_norm": 1.109375,
"learning_rate": 0.00048504848274083095,
"loss": 4.8733,
"mean_token_accuracy": 0.21224427223205566,
"num_tokens": 27332890.0,
"step": 14855
},
{
"entropy": 5.24479079246521,
"epoch": 1.241229535583027,
"grad_norm": 1.0390625,
"learning_rate": 0.00048503780984499625,
"loss": 4.7945,
"mean_token_accuracy": 0.2262277141213417,
"num_tokens": 27342411.0,
"step": 14860
},
{
"entropy": 5.2461936473846436,
"epoch": 1.24164717674574,
"grad_norm": 1.046875,
"learning_rate": 0.0004850271332721968,
"loss": 4.8441,
"mean_token_accuracy": 0.21478026509284973,
"num_tokens": 27351649.0,
"step": 14865
},
{
"entropy": 5.240424680709839,
"epoch": 1.242064817908453,
"grad_norm": 1.09375,
"learning_rate": 0.0004850164530226196,
"loss": 4.7736,
"mean_token_accuracy": 0.2206332266330719,
"num_tokens": 27360093.0,
"step": 14870
},
{
"entropy": 5.229906320571899,
"epoch": 1.242482459071166,
"grad_norm": 1.015625,
"learning_rate": 0.0004850057690964515,
"loss": 4.795,
"mean_token_accuracy": 0.2174660637974739,
"num_tokens": 27368593.0,
"step": 14875
},
{
"entropy": 5.255605220794678,
"epoch": 1.242900100233879,
"grad_norm": 1.0625,
"learning_rate": 0.00048499508149387973,
"loss": 4.8931,
"mean_token_accuracy": 0.20609840154647827,
"num_tokens": 27378126.0,
"step": 14880
},
{
"entropy": 5.239598178863526,
"epoch": 1.2433177413965921,
"grad_norm": 1.1875,
"learning_rate": 0.00048498439021509124,
"loss": 4.747,
"mean_token_accuracy": 0.21923812180757524,
"num_tokens": 27386225.0,
"step": 14885
},
{
"entropy": 5.294436407089234,
"epoch": 1.2437353825593052,
"grad_norm": 1.1171875,
"learning_rate": 0.0004849736952602732,
"loss": 4.8822,
"mean_token_accuracy": 0.21322987079620362,
"num_tokens": 27395629.0,
"step": 14890
},
{
"entropy": 5.2865002155303955,
"epoch": 1.244153023722018,
"grad_norm": 1.046875,
"learning_rate": 0.000484962996629613,
"loss": 4.8854,
"mean_token_accuracy": 0.22254017740488052,
"num_tokens": 27404431.0,
"step": 14895
},
{
"entropy": 5.259862899780273,
"epoch": 1.244570664884731,
"grad_norm": 1.015625,
"learning_rate": 0.00048495229432329775,
"loss": 4.8813,
"mean_token_accuracy": 0.21834188252687453,
"num_tokens": 27412944.0,
"step": 14900
},
{
"entropy": 5.243988752365112,
"epoch": 1.244988306047444,
"grad_norm": 0.98046875,
"learning_rate": 0.00048494158834151485,
"loss": 4.7938,
"mean_token_accuracy": 0.2166894182562828,
"num_tokens": 27422006.0,
"step": 14905
},
{
"entropy": 5.310099649429321,
"epoch": 1.245405947210157,
"grad_norm": 1.0234375,
"learning_rate": 0.00048493087868445176,
"loss": 4.8866,
"mean_token_accuracy": 0.2199595957994461,
"num_tokens": 27431568.0,
"step": 14910
},
{
"entropy": 5.301585531234741,
"epoch": 1.24582358837287,
"grad_norm": 1.03125,
"learning_rate": 0.00048492016535229594,
"loss": 4.8367,
"mean_token_accuracy": 0.2171729251742363,
"num_tokens": 27440471.0,
"step": 14915
},
{
"entropy": 5.203983783721924,
"epoch": 1.246241229535583,
"grad_norm": 0.96484375,
"learning_rate": 0.0004849094483452349,
"loss": 4.7855,
"mean_token_accuracy": 0.22085929960012435,
"num_tokens": 27450209.0,
"step": 14920
},
{
"entropy": 5.317007064819336,
"epoch": 1.246658870698296,
"grad_norm": 1.109375,
"learning_rate": 0.0004848987276634564,
"loss": 4.9862,
"mean_token_accuracy": 0.20518159568309785,
"num_tokens": 27458876.0,
"step": 14925
},
{
"entropy": 5.336245584487915,
"epoch": 1.247076511861009,
"grad_norm": 1.1171875,
"learning_rate": 0.00048488800330714795,
"loss": 4.9053,
"mean_token_accuracy": 0.21860934644937516,
"num_tokens": 27467526.0,
"step": 14930
},
{
"entropy": 5.256207227706909,
"epoch": 1.247494153023722,
"grad_norm": 0.9140625,
"learning_rate": 0.00048487727527649733,
"loss": 4.7824,
"mean_token_accuracy": 0.21915097832679747,
"num_tokens": 27477063.0,
"step": 14935
},
{
"entropy": 5.260677099227905,
"epoch": 1.247911794186435,
"grad_norm": 1.0546875,
"learning_rate": 0.00048486654357169246,
"loss": 4.8983,
"mean_token_accuracy": 0.2150811016559601,
"num_tokens": 27486243.0,
"step": 14940
},
{
"entropy": 5.286910438537598,
"epoch": 1.248329435349148,
"grad_norm": 1.0546875,
"learning_rate": 0.0004848558081929211,
"loss": 4.8885,
"mean_token_accuracy": 0.21131252646446227,
"num_tokens": 27495491.0,
"step": 14945
},
{
"entropy": 5.26129560470581,
"epoch": 1.2487470765118611,
"grad_norm": 1.0546875,
"learning_rate": 0.0004848450691403712,
"loss": 4.8737,
"mean_token_accuracy": 0.21263331323862075,
"num_tokens": 27505105.0,
"step": 14950
},
{
"entropy": 5.214673757553101,
"epoch": 1.249164717674574,
"grad_norm": 1.0234375,
"learning_rate": 0.0004848343264142307,
"loss": 4.8009,
"mean_token_accuracy": 0.2183808669447899,
"num_tokens": 27514457.0,
"step": 14955
},
{
"entropy": 5.300758361816406,
"epoch": 1.249582358837287,
"grad_norm": 0.95703125,
"learning_rate": 0.0004848235800146879,
"loss": 4.9637,
"mean_token_accuracy": 0.21082535684108733,
"num_tokens": 27524073.0,
"step": 14960
},
{
"entropy": 5.280302095413208,
"epoch": 1.25,
"grad_norm": 1.1328125,
"learning_rate": 0.0004848128299419306,
"loss": 4.8219,
"mean_token_accuracy": 0.21599761694669722,
"num_tokens": 27532840.0,
"step": 14965
},
{
"entropy": 5.308600997924804,
"epoch": 1.250417641162713,
"grad_norm": 1.0546875,
"learning_rate": 0.00048480207619614726,
"loss": 4.9336,
"mean_token_accuracy": 0.21012318581342698,
"num_tokens": 27541775.0,
"step": 14970
},
{
"entropy": 5.282397794723511,
"epoch": 1.250835282325426,
"grad_norm": 1.125,
"learning_rate": 0.000484791318777526,
"loss": 4.8543,
"mean_token_accuracy": 0.21930092424154282,
"num_tokens": 27550668.0,
"step": 14975
},
{
"entropy": 5.289754056930542,
"epoch": 1.2512529234881389,
"grad_norm": 0.9765625,
"learning_rate": 0.00048478055768625517,
"loss": 4.8912,
"mean_token_accuracy": 0.2130553811788559,
"num_tokens": 27560255.0,
"step": 14980
},
{
"entropy": 5.172748279571533,
"epoch": 1.251670564650852,
"grad_norm": 1.015625,
"learning_rate": 0.00048476979292252323,
"loss": 4.8297,
"mean_token_accuracy": 0.21807315349578857,
"num_tokens": 27569028.0,
"step": 14985
},
{
"entropy": 5.2405811786651615,
"epoch": 1.252088205813565,
"grad_norm": 1.078125,
"learning_rate": 0.0004847590244865185,
"loss": 4.7564,
"mean_token_accuracy": 0.2257744014263153,
"num_tokens": 27577396.0,
"step": 14990
},
{
"entropy": 5.28857388496399,
"epoch": 1.252505846976278,
"grad_norm": 1.0625,
"learning_rate": 0.00048474825237842964,
"loss": 4.9182,
"mean_token_accuracy": 0.20821880251169206,
"num_tokens": 27586710.0,
"step": 14995
},
{
"entropy": 5.1709624290466305,
"epoch": 1.252923488138991,
"grad_norm": 1.1953125,
"learning_rate": 0.000484737476598445,
"loss": 4.7593,
"mean_token_accuracy": 0.22051508724689484,
"num_tokens": 27594842.0,
"step": 15000
},
{
"epoch": 1.252923488138991,
"eval_entropy": 4.961925816066284,
"eval_loss": 4.875884532928467,
"eval_mean_token_accuracy": 0.22639884216897543,
"eval_num_tokens": 27594842.0,
"eval_runtime": 20.8467,
"eval_samples_per_second": 1811.461,
"eval_steps_per_second": 226.463,
"step": 15000
}
],
"logging_steps": 5,
"max_steps": 119720,
"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": 6060653295206400.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink