fpadovani/dan-latn-10mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
cc5255ece472a973fe3a41e6b5899987874faf0f
dan-latn-10mb-100mb_seed3407/checkpoint-6000/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

12057 lines
330 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5011693952555963,
"eval_steps": 3000,
"global_step": 6000,
"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
}
],
"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": 2420004800102400.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink