fpadovani/nld-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
nld-latn-100mb-after-ppt-Dp…/checkpoint-9000/trainer_state.json
ModelHub XC 4628d1d8eb 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/nld-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed3407
Source: Original Platform
2026-08-12 22:26:33 +08:00

18068 lines
494 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.816400580551524,
"eval_steps": 3000,
"global_step": 9000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 7.095729112625122,
"epoch": 0.00045355587808418,
"grad_norm": 1.0390625,
"learning_rate": 2e-06,
"loss": 6.7265,
"mean_token_accuracy": 0.10285235941410065,
"num_tokens": 10827.0,
"step": 5
},
{
"entropy": 7.09377703666687,
"epoch": 0.00090711175616836,
"grad_norm": 1.2265625,
"learning_rate": 4.5e-06,
"loss": 6.7538,
"mean_token_accuracy": 0.10200931578874588,
"num_tokens": 21025.0,
"step": 10
},
{
"entropy": 7.057293462753296,
"epoch": 0.00136066763425254,
"grad_norm": 1.1953125,
"learning_rate": 7e-06,
"loss": 6.7002,
"mean_token_accuracy": 0.1035932444036007,
"num_tokens": 31678.0,
"step": 15
},
{
"entropy": 6.979911279678345,
"epoch": 0.00181422351233672,
"grad_norm": 1.140625,
"learning_rate": 9.5e-06,
"loss": 6.6235,
"mean_token_accuracy": 0.11412417292594909,
"num_tokens": 41025.0,
"step": 20
},
{
"entropy": 7.066703462600708,
"epoch": 0.0022677793904208998,
"grad_norm": 1.078125,
"learning_rate": 1.2e-05,
"loss": 6.7259,
"mean_token_accuracy": 0.10575319603085517,
"num_tokens": 51270.0,
"step": 25
},
{
"entropy": 6.882899045944214,
"epoch": 0.00272133526850508,
"grad_norm": 1.0859375,
"learning_rate": 1.4500000000000002e-05,
"loss": 6.6281,
"mean_token_accuracy": 0.09819270819425582,
"num_tokens": 62142.0,
"step": 30
},
{
"entropy": 6.884727811813354,
"epoch": 0.00317489114658926,
"grad_norm": 1.015625,
"learning_rate": 1.7000000000000003e-05,
"loss": 6.7714,
"mean_token_accuracy": 0.10656015574932098,
"num_tokens": 72709.0,
"step": 35
},
{
"entropy": 6.849240255355835,
"epoch": 0.00362844702467344,
"grad_norm": 0.98046875,
"learning_rate": 1.95e-05,
"loss": 6.5919,
"mean_token_accuracy": 0.11315458789467811,
"num_tokens": 82313.0,
"step": 40
},
{
"entropy": 6.859521293640137,
"epoch": 0.00408200290275762,
"grad_norm": 1.140625,
"learning_rate": 2.2e-05,
"loss": 6.6149,
"mean_token_accuracy": 0.11464895755052566,
"num_tokens": 92450.0,
"step": 45
},
{
"entropy": 6.880142593383789,
"epoch": 0.0045355587808417995,
"grad_norm": 1.1171875,
"learning_rate": 2.4500000000000003e-05,
"loss": 6.6249,
"mean_token_accuracy": 0.10058464556932449,
"num_tokens": 103978.0,
"step": 50
},
{
"entropy": 6.875945663452148,
"epoch": 0.00498911465892598,
"grad_norm": 0.890625,
"learning_rate": 2.7e-05,
"loss": 6.5967,
"mean_token_accuracy": 0.11548029407858848,
"num_tokens": 114134.0,
"step": 55
},
{
"entropy": 6.909515619277954,
"epoch": 0.00544267053701016,
"grad_norm": 1.0625,
"learning_rate": 2.95e-05,
"loss": 6.803,
"mean_token_accuracy": 0.10447383597493172,
"num_tokens": 125037.0,
"step": 60
},
{
"entropy": 6.890460014343262,
"epoch": 0.00589622641509434,
"grad_norm": 1.125,
"learning_rate": 3.2e-05,
"loss": 6.6158,
"mean_token_accuracy": 0.1143025241792202,
"num_tokens": 134721.0,
"step": 65
},
{
"entropy": 6.870286560058593,
"epoch": 0.00634978229317852,
"grad_norm": 1.015625,
"learning_rate": 3.4500000000000005e-05,
"loss": 6.5605,
"mean_token_accuracy": 0.11281021013855934,
"num_tokens": 144448.0,
"step": 70
},
{
"entropy": 6.873729610443116,
"epoch": 0.006803338171262699,
"grad_norm": 1.046875,
"learning_rate": 3.7e-05,
"loss": 6.6081,
"mean_token_accuracy": 0.11354126334190369,
"num_tokens": 154343.0,
"step": 75
},
{
"entropy": 6.87849268913269,
"epoch": 0.00725689404934688,
"grad_norm": 0.9765625,
"learning_rate": 3.95e-05,
"loss": 6.667,
"mean_token_accuracy": 0.11082403212785721,
"num_tokens": 165650.0,
"step": 80
},
{
"entropy": 7.006619787216186,
"epoch": 0.007710449927431059,
"grad_norm": 1.140625,
"learning_rate": 4.2000000000000004e-05,
"loss": 6.8666,
"mean_token_accuracy": 0.10090332627296447,
"num_tokens": 177098.0,
"step": 85
},
{
"entropy": 6.911909055709839,
"epoch": 0.00816400580551524,
"grad_norm": 1.109375,
"learning_rate": 4.45e-05,
"loss": 6.5864,
"mean_token_accuracy": 0.11723631173372269,
"num_tokens": 187903.0,
"step": 90
},
{
"entropy": 6.861669921875,
"epoch": 0.00861756168359942,
"grad_norm": 1.109375,
"learning_rate": 4.7000000000000004e-05,
"loss": 6.6282,
"mean_token_accuracy": 0.11214306727051734,
"num_tokens": 198077.0,
"step": 95
},
{
"entropy": 6.918485116958618,
"epoch": 0.009071117561683599,
"grad_norm": 1.0703125,
"learning_rate": 4.9500000000000004e-05,
"loss": 6.7492,
"mean_token_accuracy": 0.10802269652485848,
"num_tokens": 208641.0,
"step": 100
},
{
"entropy": 6.897343158721924,
"epoch": 0.009524673439767779,
"grad_norm": 1.0859375,
"learning_rate": 5.2e-05,
"loss": 6.6283,
"mean_token_accuracy": 0.11756790578365325,
"num_tokens": 218398.0,
"step": 105
},
{
"entropy": 6.847284555435181,
"epoch": 0.00997822931785196,
"grad_norm": 1.0703125,
"learning_rate": 5.45e-05,
"loss": 6.5401,
"mean_token_accuracy": 0.11274854093790054,
"num_tokens": 229059.0,
"step": 110
},
{
"entropy": 6.852484703063965,
"epoch": 0.01043178519593614,
"grad_norm": 0.97265625,
"learning_rate": 5.7e-05,
"loss": 6.6082,
"mean_token_accuracy": 0.10856100469827652,
"num_tokens": 239775.0,
"step": 115
},
{
"entropy": 6.885813903808594,
"epoch": 0.01088534107402032,
"grad_norm": 1.015625,
"learning_rate": 5.9499999999999996e-05,
"loss": 6.5006,
"mean_token_accuracy": 0.11381678581237793,
"num_tokens": 249954.0,
"step": 120
},
{
"entropy": 6.860991525650024,
"epoch": 0.011338896952104499,
"grad_norm": 1.015625,
"learning_rate": 6.2e-05,
"loss": 6.6168,
"mean_token_accuracy": 0.10877051502466202,
"num_tokens": 260056.0,
"step": 125
},
{
"entropy": 6.84809103012085,
"epoch": 0.01179245283018868,
"grad_norm": 1.0859375,
"learning_rate": 6.450000000000001e-05,
"loss": 6.5904,
"mean_token_accuracy": 0.11548284813761711,
"num_tokens": 269904.0,
"step": 130
},
{
"entropy": 6.958385229110718,
"epoch": 0.01224600870827286,
"grad_norm": 1.0546875,
"learning_rate": 6.7e-05,
"loss": 6.5946,
"mean_token_accuracy": 0.11288051903247834,
"num_tokens": 280900.0,
"step": 135
},
{
"entropy": 6.819544267654419,
"epoch": 0.01269956458635704,
"grad_norm": 1.078125,
"learning_rate": 6.950000000000001e-05,
"loss": 6.634,
"mean_token_accuracy": 0.11247359365224838,
"num_tokens": 290844.0,
"step": 140
},
{
"entropy": 6.850368165969849,
"epoch": 0.013153120464441219,
"grad_norm": 1.0546875,
"learning_rate": 7.2e-05,
"loss": 6.611,
"mean_token_accuracy": 0.11267527937889099,
"num_tokens": 301345.0,
"step": 145
},
{
"entropy": 6.930296087265015,
"epoch": 0.013606676342525399,
"grad_norm": 1.0,
"learning_rate": 7.45e-05,
"loss": 6.6225,
"mean_token_accuracy": 0.11140098199248313,
"num_tokens": 312681.0,
"step": 150
},
{
"entropy": 6.830138874053955,
"epoch": 0.01406023222060958,
"grad_norm": 1.0234375,
"learning_rate": 7.7e-05,
"loss": 6.571,
"mean_token_accuracy": 0.11029604226350784,
"num_tokens": 323492.0,
"step": 155
},
{
"entropy": 6.854236030578614,
"epoch": 0.01451378809869376,
"grad_norm": 1.140625,
"learning_rate": 7.950000000000001e-05,
"loss": 6.6352,
"mean_token_accuracy": 0.10969627797603607,
"num_tokens": 334521.0,
"step": 160
},
{
"entropy": 6.917273998260498,
"epoch": 0.014967343976777939,
"grad_norm": 1.078125,
"learning_rate": 8.2e-05,
"loss": 6.6505,
"mean_token_accuracy": 0.11680250763893127,
"num_tokens": 345468.0,
"step": 165
},
{
"entropy": 6.939955139160157,
"epoch": 0.015420899854862119,
"grad_norm": 1.2109375,
"learning_rate": 8.450000000000001e-05,
"loss": 6.5748,
"mean_token_accuracy": 0.11744663268327712,
"num_tokens": 355911.0,
"step": 170
},
{
"entropy": 6.8326245784759525,
"epoch": 0.0158744557329463,
"grad_norm": 1.0546875,
"learning_rate": 8.7e-05,
"loss": 6.5974,
"mean_token_accuracy": 0.11093726232647896,
"num_tokens": 367537.0,
"step": 175
},
{
"entropy": 7.006328916549682,
"epoch": 0.01632801161103048,
"grad_norm": 1.234375,
"learning_rate": 8.95e-05,
"loss": 6.785,
"mean_token_accuracy": 0.10842694193124772,
"num_tokens": 377918.0,
"step": 180
},
{
"entropy": 6.904155826568603,
"epoch": 0.01678156748911466,
"grad_norm": 1.046875,
"learning_rate": 9.2e-05,
"loss": 6.6467,
"mean_token_accuracy": 0.11190110594034194,
"num_tokens": 388461.0,
"step": 185
},
{
"entropy": 6.9462827205657955,
"epoch": 0.01723512336719884,
"grad_norm": 1.171875,
"learning_rate": 9.45e-05,
"loss": 6.624,
"mean_token_accuracy": 0.11254355385899543,
"num_tokens": 399771.0,
"step": 190
},
{
"entropy": 6.802505254745483,
"epoch": 0.01768867924528302,
"grad_norm": 1.0703125,
"learning_rate": 9.7e-05,
"loss": 6.5542,
"mean_token_accuracy": 0.11513888537883758,
"num_tokens": 410254.0,
"step": 195
},
{
"entropy": 6.881962490081787,
"epoch": 0.018142235123367198,
"grad_norm": 1.1328125,
"learning_rate": 9.95e-05,
"loss": 6.5364,
"mean_token_accuracy": 0.10974982380867004,
"num_tokens": 419792.0,
"step": 200
},
{
"entropy": 6.786790180206299,
"epoch": 0.018595791001451378,
"grad_norm": 1.140625,
"learning_rate": 0.000102,
"loss": 6.5911,
"mean_token_accuracy": 0.11597937643527985,
"num_tokens": 429264.0,
"step": 205
},
{
"entropy": 6.895502185821533,
"epoch": 0.019049346879535557,
"grad_norm": 1.015625,
"learning_rate": 0.00010449999999999999,
"loss": 6.5448,
"mean_token_accuracy": 0.11790847852826118,
"num_tokens": 439624.0,
"step": 210
},
{
"entropy": 6.8149701118469235,
"epoch": 0.01950290275761974,
"grad_norm": 1.171875,
"learning_rate": 0.000107,
"loss": 6.5652,
"mean_token_accuracy": 0.11152517050504684,
"num_tokens": 449556.0,
"step": 215
},
{
"entropy": 6.898656415939331,
"epoch": 0.01995645863570392,
"grad_norm": 1.4296875,
"learning_rate": 0.0001095,
"loss": 6.6111,
"mean_token_accuracy": 0.1134468026459217,
"num_tokens": 460243.0,
"step": 220
},
{
"entropy": 6.8338030815124515,
"epoch": 0.0204100145137881,
"grad_norm": 1.1875,
"learning_rate": 0.000112,
"loss": 6.626,
"mean_token_accuracy": 0.11214038878679275,
"num_tokens": 469767.0,
"step": 225
},
{
"entropy": 6.919766855239868,
"epoch": 0.02086357039187228,
"grad_norm": 1.140625,
"learning_rate": 0.0001145,
"loss": 6.5885,
"mean_token_accuracy": 0.11247648671269417,
"num_tokens": 480329.0,
"step": 230
},
{
"entropy": 6.827954387664795,
"epoch": 0.02131712626995646,
"grad_norm": 1.1328125,
"learning_rate": 0.00011700000000000001,
"loss": 6.541,
"mean_token_accuracy": 0.11013604179024697,
"num_tokens": 490096.0,
"step": 235
},
{
"entropy": 6.920001554489136,
"epoch": 0.02177068214804064,
"grad_norm": 1.1953125,
"learning_rate": 0.00011949999999999999,
"loss": 6.5329,
"mean_token_accuracy": 0.11023009046912194,
"num_tokens": 501231.0,
"step": 240
},
{
"entropy": 6.743205976486206,
"epoch": 0.022224238026124818,
"grad_norm": 1.25,
"learning_rate": 0.000122,
"loss": 6.5103,
"mean_token_accuracy": 0.12003011107444764,
"num_tokens": 510722.0,
"step": 245
},
{
"entropy": 6.816294765472412,
"epoch": 0.022677793904208998,
"grad_norm": 1.140625,
"learning_rate": 0.0001245,
"loss": 6.575,
"mean_token_accuracy": 0.11273853480815887,
"num_tokens": 521452.0,
"step": 250
},
{
"entropy": 6.891448831558227,
"epoch": 0.023131349782293177,
"grad_norm": 1.15625,
"learning_rate": 0.000127,
"loss": 6.5903,
"mean_token_accuracy": 0.11311487331986428,
"num_tokens": 532397.0,
"step": 255
},
{
"entropy": 6.873995685577393,
"epoch": 0.02358490566037736,
"grad_norm": 1.09375,
"learning_rate": 0.0001295,
"loss": 6.561,
"mean_token_accuracy": 0.11095410361886024,
"num_tokens": 542271.0,
"step": 260
},
{
"entropy": 6.863357591629028,
"epoch": 0.02403846153846154,
"grad_norm": 1.3359375,
"learning_rate": 0.000132,
"loss": 6.5836,
"mean_token_accuracy": 0.114754718542099,
"num_tokens": 552411.0,
"step": 265
},
{
"entropy": 6.8279015064239506,
"epoch": 0.02449201741654572,
"grad_norm": 1.1875,
"learning_rate": 0.00013450000000000002,
"loss": 6.5534,
"mean_token_accuracy": 0.11625841185450554,
"num_tokens": 561518.0,
"step": 270
},
{
"entropy": 6.916412401199341,
"epoch": 0.0249455732946299,
"grad_norm": 1.3125,
"learning_rate": 0.00013700000000000002,
"loss": 6.6294,
"mean_token_accuracy": 0.11257987022399903,
"num_tokens": 571361.0,
"step": 275
},
{
"entropy": 6.762298631668091,
"epoch": 0.02539912917271408,
"grad_norm": 1.1640625,
"learning_rate": 0.0001395,
"loss": 6.5132,
"mean_token_accuracy": 0.11271893158555031,
"num_tokens": 580926.0,
"step": 280
},
{
"entropy": 6.915192937850952,
"epoch": 0.025852685050798258,
"grad_norm": 1.015625,
"learning_rate": 0.00014199999999999998,
"loss": 6.6046,
"mean_token_accuracy": 0.10786937177181244,
"num_tokens": 592282.0,
"step": 285
},
{
"entropy": 6.926600074768066,
"epoch": 0.026306240928882438,
"grad_norm": 1.1875,
"learning_rate": 0.0001445,
"loss": 6.6318,
"mean_token_accuracy": 0.1095262661576271,
"num_tokens": 602767.0,
"step": 290
},
{
"entropy": 6.829335880279541,
"epoch": 0.026759796806966617,
"grad_norm": 1.1953125,
"learning_rate": 0.000147,
"loss": 6.5364,
"mean_token_accuracy": 0.1123177632689476,
"num_tokens": 612763.0,
"step": 295
},
{
"entropy": 6.844592142105102,
"epoch": 0.027213352685050797,
"grad_norm": 1.1953125,
"learning_rate": 0.0001495,
"loss": 6.5237,
"mean_token_accuracy": 0.11495825499296189,
"num_tokens": 622796.0,
"step": 300
},
{
"entropy": 6.8091404914855955,
"epoch": 0.027666908563134977,
"grad_norm": 1.25,
"learning_rate": 0.000152,
"loss": 6.4945,
"mean_token_accuracy": 0.11651193872094154,
"num_tokens": 633136.0,
"step": 305
},
{
"entropy": 6.694456291198731,
"epoch": 0.02812046444121916,
"grad_norm": 1.21875,
"learning_rate": 0.00015450000000000001,
"loss": 6.5303,
"mean_token_accuracy": 0.11837984547019005,
"num_tokens": 643347.0,
"step": 310
},
{
"entropy": 6.884937953948975,
"epoch": 0.02857402031930334,
"grad_norm": 1.2578125,
"learning_rate": 0.000157,
"loss": 6.5096,
"mean_token_accuracy": 0.12059485390782357,
"num_tokens": 653096.0,
"step": 315
},
{
"entropy": 6.679642105102539,
"epoch": 0.02902757619738752,
"grad_norm": 1.2890625,
"learning_rate": 0.0001595,
"loss": 6.5502,
"mean_token_accuracy": 0.11287975832819938,
"num_tokens": 663557.0,
"step": 320
},
{
"entropy": 6.917266988754273,
"epoch": 0.0294811320754717,
"grad_norm": 1.28125,
"learning_rate": 0.000162,
"loss": 6.6358,
"mean_token_accuracy": 0.1109463818371296,
"num_tokens": 674395.0,
"step": 325
},
{
"entropy": 6.825232934951782,
"epoch": 0.029934687953555878,
"grad_norm": 1.2109375,
"learning_rate": 0.00016450000000000001,
"loss": 6.5606,
"mean_token_accuracy": 0.11592193096876144,
"num_tokens": 684776.0,
"step": 330
},
{
"entropy": 6.785255765914917,
"epoch": 0.030388243831640058,
"grad_norm": 1.109375,
"learning_rate": 0.00016700000000000002,
"loss": 6.5179,
"mean_token_accuracy": 0.11499937400221824,
"num_tokens": 694886.0,
"step": 335
},
{
"entropy": 6.853105878829956,
"epoch": 0.030841799709724237,
"grad_norm": 1.1875,
"learning_rate": 0.00016950000000000003,
"loss": 6.5986,
"mean_token_accuracy": 0.10862654820084572,
"num_tokens": 705426.0,
"step": 340
},
{
"entropy": 6.87786340713501,
"epoch": 0.03129535558780842,
"grad_norm": 1.125,
"learning_rate": 0.00017199999999999998,
"loss": 6.6063,
"mean_token_accuracy": 0.11100632250308991,
"num_tokens": 716271.0,
"step": 345
},
{
"entropy": 6.8458168506622314,
"epoch": 0.0317489114658926,
"grad_norm": 1.0859375,
"learning_rate": 0.00017449999999999999,
"loss": 6.5824,
"mean_token_accuracy": 0.11013020351529121,
"num_tokens": 726875.0,
"step": 350
},
{
"entropy": 6.771889114379883,
"epoch": 0.032202467343976776,
"grad_norm": 1.1015625,
"learning_rate": 0.000177,
"loss": 6.5288,
"mean_token_accuracy": 0.11027872115373612,
"num_tokens": 737825.0,
"step": 355
},
{
"entropy": 6.777115345001221,
"epoch": 0.03265602322206096,
"grad_norm": 1.09375,
"learning_rate": 0.0001795,
"loss": 6.4789,
"mean_token_accuracy": 0.11788042932748795,
"num_tokens": 747424.0,
"step": 360
},
{
"entropy": 6.832462644577026,
"epoch": 0.033109579100145135,
"grad_norm": 1.2109375,
"learning_rate": 0.000182,
"loss": 6.5126,
"mean_token_accuracy": 0.11428287327289581,
"num_tokens": 757649.0,
"step": 365
},
{
"entropy": 6.798842334747315,
"epoch": 0.03356313497822932,
"grad_norm": 1.1796875,
"learning_rate": 0.0001845,
"loss": 6.53,
"mean_token_accuracy": 0.11638613939285278,
"num_tokens": 768556.0,
"step": 370
},
{
"entropy": 6.701859331130981,
"epoch": 0.034016690856313495,
"grad_norm": 1.3125,
"learning_rate": 0.000187,
"loss": 6.5256,
"mean_token_accuracy": 0.11531673222780228,
"num_tokens": 778331.0,
"step": 375
},
{
"entropy": 6.794470977783203,
"epoch": 0.03447024673439768,
"grad_norm": 1.3203125,
"learning_rate": 0.0001895,
"loss": 6.5285,
"mean_token_accuracy": 0.11249540448188781,
"num_tokens": 788065.0,
"step": 380
},
{
"entropy": 6.837066221237182,
"epoch": 0.03492380261248186,
"grad_norm": 1.3359375,
"learning_rate": 0.000192,
"loss": 6.5856,
"mean_token_accuracy": 0.11070910617709159,
"num_tokens": 798912.0,
"step": 385
},
{
"entropy": 6.788638496398926,
"epoch": 0.03537735849056604,
"grad_norm": 1.1328125,
"learning_rate": 0.0001945,
"loss": 6.5211,
"mean_token_accuracy": 0.12141896188259124,
"num_tokens": 809580.0,
"step": 390
},
{
"entropy": 6.805673980712891,
"epoch": 0.03583091436865022,
"grad_norm": 1.171875,
"learning_rate": 0.00019700000000000002,
"loss": 6.5986,
"mean_token_accuracy": 0.11551168486475945,
"num_tokens": 820207.0,
"step": 395
},
{
"entropy": 6.665769624710083,
"epoch": 0.036284470246734396,
"grad_norm": 1.125,
"learning_rate": 0.00019950000000000002,
"loss": 6.4231,
"mean_token_accuracy": 0.11134176030755043,
"num_tokens": 831038.0,
"step": 400
},
{
"entropy": 6.818566799163818,
"epoch": 0.03673802612481858,
"grad_norm": 1.0625,
"learning_rate": 0.000202,
"loss": 6.5358,
"mean_token_accuracy": 0.11346253231167794,
"num_tokens": 842144.0,
"step": 405
},
{
"entropy": 6.820947122573853,
"epoch": 0.037191582002902755,
"grad_norm": 1.1640625,
"learning_rate": 0.00020449999999999998,
"loss": 6.5339,
"mean_token_accuracy": 0.11876299902796746,
"num_tokens": 851315.0,
"step": 410
},
{
"entropy": 6.729552555084228,
"epoch": 0.03764513788098694,
"grad_norm": 1.1953125,
"learning_rate": 0.000207,
"loss": 6.563,
"mean_token_accuracy": 0.1145752988755703,
"num_tokens": 861233.0,
"step": 415
},
{
"entropy": 6.747600364685058,
"epoch": 0.038098693759071114,
"grad_norm": 1.171875,
"learning_rate": 0.0002095,
"loss": 6.5175,
"mean_token_accuracy": 0.11273587495088577,
"num_tokens": 871131.0,
"step": 420
},
{
"entropy": 6.765731859207153,
"epoch": 0.0385522496371553,
"grad_norm": 1.0546875,
"learning_rate": 0.000212,
"loss": 6.5096,
"mean_token_accuracy": 0.11347832456231118,
"num_tokens": 882195.0,
"step": 425
},
{
"entropy": 6.763120365142822,
"epoch": 0.03900580551523948,
"grad_norm": 1.2421875,
"learning_rate": 0.0002145,
"loss": 6.493,
"mean_token_accuracy": 0.11830192133784294,
"num_tokens": 891806.0,
"step": 430
},
{
"entropy": 6.7991108894348145,
"epoch": 0.03945936139332366,
"grad_norm": 1.1171875,
"learning_rate": 0.00021700000000000002,
"loss": 6.5534,
"mean_token_accuracy": 0.10942765697836876,
"num_tokens": 901350.0,
"step": 435
},
{
"entropy": 6.767538928985596,
"epoch": 0.03991291727140784,
"grad_norm": 1.265625,
"learning_rate": 0.0002195,
"loss": 6.461,
"mean_token_accuracy": 0.11672212779521943,
"num_tokens": 912061.0,
"step": 440
},
{
"entropy": 6.663478183746338,
"epoch": 0.040366473149492016,
"grad_norm": 1.28125,
"learning_rate": 0.000222,
"loss": 6.4298,
"mean_token_accuracy": 0.1151677392423153,
"num_tokens": 922220.0,
"step": 445
},
{
"entropy": 6.714645051956177,
"epoch": 0.0408200290275762,
"grad_norm": 1.2734375,
"learning_rate": 0.0002245,
"loss": 6.4781,
"mean_token_accuracy": 0.11699502617120743,
"num_tokens": 932319.0,
"step": 450
},
{
"entropy": 6.740534543991089,
"epoch": 0.041273584905660375,
"grad_norm": 1.265625,
"learning_rate": 0.00022700000000000002,
"loss": 6.4927,
"mean_token_accuracy": 0.1162272535264492,
"num_tokens": 943254.0,
"step": 455
},
{
"entropy": 6.749291038513183,
"epoch": 0.04172714078374456,
"grad_norm": 1.1015625,
"learning_rate": 0.00022950000000000002,
"loss": 6.5346,
"mean_token_accuracy": 0.10895546674728393,
"num_tokens": 953921.0,
"step": 460
},
{
"entropy": 6.654842042922974,
"epoch": 0.042180696661828734,
"grad_norm": 1.1796875,
"learning_rate": 0.00023200000000000003,
"loss": 6.4555,
"mean_token_accuracy": 0.11954837366938591,
"num_tokens": 963832.0,
"step": 465
},
{
"entropy": 6.707518577575684,
"epoch": 0.04263425253991292,
"grad_norm": 1.375,
"learning_rate": 0.00023449999999999998,
"loss": 6.5509,
"mean_token_accuracy": 0.1089623972773552,
"num_tokens": 974066.0,
"step": 470
},
{
"entropy": 6.776539993286133,
"epoch": 0.0430878084179971,
"grad_norm": 1.15625,
"learning_rate": 0.000237,
"loss": 6.5395,
"mean_token_accuracy": 0.11641593873500825,
"num_tokens": 984402.0,
"step": 475
},
{
"entropy": 6.677361297607422,
"epoch": 0.04354136429608128,
"grad_norm": 1.109375,
"learning_rate": 0.0002395,
"loss": 6.4249,
"mean_token_accuracy": 0.11875243708491326,
"num_tokens": 993555.0,
"step": 480
},
{
"entropy": 6.7442159175872805,
"epoch": 0.04399492017416546,
"grad_norm": 1.3359375,
"learning_rate": 0.000242,
"loss": 6.4435,
"mean_token_accuracy": 0.12278054803609847,
"num_tokens": 1002920.0,
"step": 485
},
{
"entropy": 6.676013088226318,
"epoch": 0.044448476052249636,
"grad_norm": 1.1875,
"learning_rate": 0.0002445,
"loss": 6.5368,
"mean_token_accuracy": 0.1078170619904995,
"num_tokens": 1013611.0,
"step": 490
},
{
"entropy": 6.653495216369629,
"epoch": 0.04490203193033382,
"grad_norm": 1.1328125,
"learning_rate": 0.000247,
"loss": 6.4169,
"mean_token_accuracy": 0.1161510780453682,
"num_tokens": 1024121.0,
"step": 495
},
{
"entropy": 6.81571741104126,
"epoch": 0.045355587808417995,
"grad_norm": 1.2109375,
"learning_rate": 0.0002495,
"loss": 6.653,
"mean_token_accuracy": 0.11692796349525451,
"num_tokens": 1033845.0,
"step": 500
},
{
"entropy": 6.724951601028442,
"epoch": 0.04580914368650218,
"grad_norm": 1.078125,
"learning_rate": 0.000252,
"loss": 6.4562,
"mean_token_accuracy": 0.1180861733853817,
"num_tokens": 1044642.0,
"step": 505
},
{
"entropy": 6.615443086624145,
"epoch": 0.046262699564586354,
"grad_norm": 1.3046875,
"learning_rate": 0.0002545,
"loss": 6.5297,
"mean_token_accuracy": 0.1160704404115677,
"num_tokens": 1054728.0,
"step": 510
},
{
"entropy": 6.7214963912963865,
"epoch": 0.04671625544267054,
"grad_norm": 1.140625,
"learning_rate": 0.000257,
"loss": 6.4834,
"mean_token_accuracy": 0.10873565226793289,
"num_tokens": 1065743.0,
"step": 515
},
{
"entropy": 6.680114889144898,
"epoch": 0.04716981132075472,
"grad_norm": 1.2265625,
"learning_rate": 0.0002595,
"loss": 6.3986,
"mean_token_accuracy": 0.11717408373951912,
"num_tokens": 1076798.0,
"step": 520
},
{
"entropy": 6.656686067581177,
"epoch": 0.047623367198838897,
"grad_norm": 1.34375,
"learning_rate": 0.000262,
"loss": 6.5431,
"mean_token_accuracy": 0.11418797224760055,
"num_tokens": 1087445.0,
"step": 525
},
{
"entropy": 6.606220769882202,
"epoch": 0.04807692307692308,
"grad_norm": 1.1328125,
"learning_rate": 0.00026450000000000003,
"loss": 6.4584,
"mean_token_accuracy": 0.1161401018500328,
"num_tokens": 1098379.0,
"step": 530
},
{
"entropy": 6.722146511077881,
"epoch": 0.048530478955007256,
"grad_norm": 1.296875,
"learning_rate": 0.00026700000000000004,
"loss": 6.5429,
"mean_token_accuracy": 0.11141678988933564,
"num_tokens": 1108016.0,
"step": 535
},
{
"entropy": 6.6710772037506105,
"epoch": 0.04898403483309144,
"grad_norm": 1.375,
"learning_rate": 0.00026950000000000005,
"loss": 6.4014,
"mean_token_accuracy": 0.11449897214770317,
"num_tokens": 1117887.0,
"step": 540
},
{
"entropy": 6.654284572601318,
"epoch": 0.049437590711175615,
"grad_norm": 1.171875,
"learning_rate": 0.00027200000000000005,
"loss": 6.4097,
"mean_token_accuracy": 0.11381929144263267,
"num_tokens": 1128497.0,
"step": 545
},
{
"entropy": 6.613014507293701,
"epoch": 0.0498911465892598,
"grad_norm": 1.21875,
"learning_rate": 0.0002745,
"loss": 6.4609,
"mean_token_accuracy": 0.11565488874912262,
"num_tokens": 1138168.0,
"step": 550
},
{
"entropy": 6.618120574951172,
"epoch": 0.050344702467343974,
"grad_norm": 1.1796875,
"learning_rate": 0.000277,
"loss": 6.445,
"mean_token_accuracy": 0.11517123132944107,
"num_tokens": 1148357.0,
"step": 555
},
{
"entropy": 6.67120852470398,
"epoch": 0.05079825834542816,
"grad_norm": 1.03125,
"learning_rate": 0.0002795,
"loss": 6.4813,
"mean_token_accuracy": 0.1128874771296978,
"num_tokens": 1159964.0,
"step": 560
},
{
"entropy": 6.685760498046875,
"epoch": 0.05125181422351233,
"grad_norm": 1.28125,
"learning_rate": 0.00028199999999999997,
"loss": 6.5201,
"mean_token_accuracy": 0.1172065868973732,
"num_tokens": 1170683.0,
"step": 565
},
{
"entropy": 6.719839191436767,
"epoch": 0.051705370101596516,
"grad_norm": 1.125,
"learning_rate": 0.0002845,
"loss": 6.4993,
"mean_token_accuracy": 0.11043535247445106,
"num_tokens": 1181680.0,
"step": 570
},
{
"entropy": 6.516751384735107,
"epoch": 0.0521589259796807,
"grad_norm": 1.234375,
"learning_rate": 0.000287,
"loss": 6.3766,
"mean_token_accuracy": 0.11628183797001838,
"num_tokens": 1191619.0,
"step": 575
},
{
"entropy": 6.552532625198364,
"epoch": 0.052612481857764876,
"grad_norm": 1.2421875,
"learning_rate": 0.0002895,
"loss": 6.387,
"mean_token_accuracy": 0.11748170480132103,
"num_tokens": 1202264.0,
"step": 580
},
{
"entropy": 6.598452091217041,
"epoch": 0.05306603773584906,
"grad_norm": 1.2890625,
"learning_rate": 0.000292,
"loss": 6.3936,
"mean_token_accuracy": 0.11468102261424065,
"num_tokens": 1212119.0,
"step": 585
},
{
"entropy": 6.638687658309936,
"epoch": 0.053519593613933235,
"grad_norm": 1.2109375,
"learning_rate": 0.0002945,
"loss": 6.5243,
"mean_token_accuracy": 0.11499153003096581,
"num_tokens": 1222317.0,
"step": 590
},
{
"entropy": 6.6762138366699215,
"epoch": 0.05397314949201742,
"grad_norm": 1.2734375,
"learning_rate": 0.000297,
"loss": 6.4613,
"mean_token_accuracy": 0.11701688691973686,
"num_tokens": 1232914.0,
"step": 595
},
{
"entropy": 6.519153881072998,
"epoch": 0.054426705370101594,
"grad_norm": 1.046875,
"learning_rate": 0.0002995,
"loss": 6.3034,
"mean_token_accuracy": 0.11886714175343513,
"num_tokens": 1244214.0,
"step": 600
},
{
"entropy": 6.619960975646973,
"epoch": 0.05488026124818578,
"grad_norm": 1.375,
"learning_rate": 0.000302,
"loss": 6.4138,
"mean_token_accuracy": 0.11526961103081704,
"num_tokens": 1254626.0,
"step": 605
},
{
"entropy": 6.602964544296265,
"epoch": 0.05533381712626995,
"grad_norm": 1.328125,
"learning_rate": 0.0003045,
"loss": 6.5216,
"mean_token_accuracy": 0.11393038928508759,
"num_tokens": 1265461.0,
"step": 610
},
{
"entropy": 6.613763475418091,
"epoch": 0.055787373004354136,
"grad_norm": 1.28125,
"learning_rate": 0.000307,
"loss": 6.387,
"mean_token_accuracy": 0.11922261342406273,
"num_tokens": 1275810.0,
"step": 615
},
{
"entropy": 6.47516770362854,
"epoch": 0.05624092888243832,
"grad_norm": 1.3359375,
"learning_rate": 0.0003095,
"loss": 6.413,
"mean_token_accuracy": 0.1167520485818386,
"num_tokens": 1285170.0,
"step": 620
},
{
"entropy": 6.6089705467224125,
"epoch": 0.056694484760522496,
"grad_norm": 1.1171875,
"learning_rate": 0.000312,
"loss": 6.3501,
"mean_token_accuracy": 0.11907305344939231,
"num_tokens": 1295460.0,
"step": 625
},
{
"entropy": 6.5755761623382565,
"epoch": 0.05714804063860668,
"grad_norm": 1.2890625,
"learning_rate": 0.0003145,
"loss": 6.4893,
"mean_token_accuracy": 0.11553953886032105,
"num_tokens": 1306656.0,
"step": 630
},
{
"entropy": 6.614667463302612,
"epoch": 0.057601596516690855,
"grad_norm": 1.1640625,
"learning_rate": 0.000317,
"loss": 6.4463,
"mean_token_accuracy": 0.11898912116885185,
"num_tokens": 1318019.0,
"step": 635
},
{
"entropy": 6.514087343215943,
"epoch": 0.05805515239477504,
"grad_norm": 1.1875,
"learning_rate": 0.0003195,
"loss": 6.3713,
"mean_token_accuracy": 0.11803188174962997,
"num_tokens": 1328120.0,
"step": 640
},
{
"entropy": 6.696631908416748,
"epoch": 0.058508708272859214,
"grad_norm": 1.28125,
"learning_rate": 0.000322,
"loss": 6.4671,
"mean_token_accuracy": 0.11322417482733727,
"num_tokens": 1338354.0,
"step": 645
},
{
"entropy": 6.591117477416992,
"epoch": 0.0589622641509434,
"grad_norm": 1.5390625,
"learning_rate": 0.00032450000000000003,
"loss": 6.4288,
"mean_token_accuracy": 0.11614542976021766,
"num_tokens": 1348666.0,
"step": 650
},
{
"entropy": 6.542242813110351,
"epoch": 0.05941582002902757,
"grad_norm": 1.109375,
"learning_rate": 0.00032700000000000003,
"loss": 6.4847,
"mean_token_accuracy": 0.11858619302511215,
"num_tokens": 1359713.0,
"step": 655
},
{
"entropy": 6.600148725509643,
"epoch": 0.059869375907111756,
"grad_norm": 1.1328125,
"learning_rate": 0.00032950000000000004,
"loss": 6.3456,
"mean_token_accuracy": 0.12028133273124694,
"num_tokens": 1369507.0,
"step": 660
},
{
"entropy": 6.558723402023316,
"epoch": 0.06032293178519594,
"grad_norm": 1.265625,
"learning_rate": 0.00033200000000000005,
"loss": 6.5014,
"mean_token_accuracy": 0.11275067254900932,
"num_tokens": 1380726.0,
"step": 665
},
{
"entropy": 6.666363334655761,
"epoch": 0.060776487663280115,
"grad_norm": 1.15625,
"learning_rate": 0.00033450000000000005,
"loss": 6.4262,
"mean_token_accuracy": 0.11828458458185195,
"num_tokens": 1391423.0,
"step": 670
},
{
"entropy": 6.609200382232666,
"epoch": 0.0612300435413643,
"grad_norm": 1.3046875,
"learning_rate": 0.000337,
"loss": 6.4021,
"mean_token_accuracy": 0.1221817396581173,
"num_tokens": 1402492.0,
"step": 675
},
{
"entropy": 6.487628936767578,
"epoch": 0.061683599419448475,
"grad_norm": 1.2421875,
"learning_rate": 0.0003395,
"loss": 6.4558,
"mean_token_accuracy": 0.11870089024305344,
"num_tokens": 1412497.0,
"step": 680
},
{
"entropy": 6.698206186294556,
"epoch": 0.06213715529753266,
"grad_norm": 1.3828125,
"learning_rate": 0.000342,
"loss": 6.4291,
"mean_token_accuracy": 0.12117572948336601,
"num_tokens": 1423214.0,
"step": 685
},
{
"entropy": 6.5064043521881105,
"epoch": 0.06259071117561683,
"grad_norm": 1.3828125,
"learning_rate": 0.00034449999999999997,
"loss": 6.4106,
"mean_token_accuracy": 0.11815783530473709,
"num_tokens": 1434044.0,
"step": 690
},
{
"entropy": 6.522539710998535,
"epoch": 0.06304426705370102,
"grad_norm": 1.140625,
"learning_rate": 0.000347,
"loss": 6.3789,
"mean_token_accuracy": 0.1227622389793396,
"num_tokens": 1444387.0,
"step": 695
},
{
"entropy": 6.471425151824951,
"epoch": 0.0634978229317852,
"grad_norm": 1.1640625,
"learning_rate": 0.0003495,
"loss": 6.3352,
"mean_token_accuracy": 0.1156968854367733,
"num_tokens": 1455157.0,
"step": 700
},
{
"entropy": 6.585581874847412,
"epoch": 0.06395137880986937,
"grad_norm": 1.2734375,
"learning_rate": 0.000352,
"loss": 6.4501,
"mean_token_accuracy": 0.12337471693754196,
"num_tokens": 1466145.0,
"step": 705
},
{
"entropy": 6.528838872909546,
"epoch": 0.06440493468795355,
"grad_norm": 1.0703125,
"learning_rate": 0.0003545,
"loss": 6.4251,
"mean_token_accuracy": 0.11607742905616761,
"num_tokens": 1476713.0,
"step": 710
},
{
"entropy": 6.579174137115478,
"epoch": 0.06485849056603774,
"grad_norm": 1.1875,
"learning_rate": 0.000357,
"loss": 6.4357,
"mean_token_accuracy": 0.12061923295259476,
"num_tokens": 1486908.0,
"step": 715
},
{
"entropy": 6.555825901031494,
"epoch": 0.06531204644412192,
"grad_norm": 1.0390625,
"learning_rate": 0.0003595,
"loss": 6.3554,
"mean_token_accuracy": 0.12144754379987717,
"num_tokens": 1498185.0,
"step": 720
},
{
"entropy": 6.533269929885864,
"epoch": 0.0657656023222061,
"grad_norm": 1.1953125,
"learning_rate": 0.000362,
"loss": 6.3975,
"mean_token_accuracy": 0.11883825734257698,
"num_tokens": 1508376.0,
"step": 725
},
{
"entropy": 6.5100702285766605,
"epoch": 0.06621915820029027,
"grad_norm": 1.1171875,
"learning_rate": 0.0003645,
"loss": 6.4877,
"mean_token_accuracy": 0.11778334230184555,
"num_tokens": 1519953.0,
"step": 730
},
{
"entropy": 6.46563720703125,
"epoch": 0.06667271407837445,
"grad_norm": 1.0859375,
"learning_rate": 0.000367,
"loss": 6.368,
"mean_token_accuracy": 0.1215110793709755,
"num_tokens": 1529793.0,
"step": 735
},
{
"entropy": 6.522617959976197,
"epoch": 0.06712626995645864,
"grad_norm": 1.0078125,
"learning_rate": 0.0003695,
"loss": 6.376,
"mean_token_accuracy": 0.11825870126485824,
"num_tokens": 1540859.0,
"step": 740
},
{
"entropy": 6.524055290222168,
"epoch": 0.06757982583454282,
"grad_norm": 1.1875,
"learning_rate": 0.000372,
"loss": 6.3031,
"mean_token_accuracy": 0.12105328515172005,
"num_tokens": 1549783.0,
"step": 745
},
{
"entropy": 6.425115442276001,
"epoch": 0.06803338171262699,
"grad_norm": 1.1484375,
"learning_rate": 0.0003745,
"loss": 6.3511,
"mean_token_accuracy": 0.12026687189936638,
"num_tokens": 1560522.0,
"step": 750
},
{
"entropy": 6.401650762557983,
"epoch": 0.06848693759071117,
"grad_norm": 1.140625,
"learning_rate": 0.000377,
"loss": 6.3384,
"mean_token_accuracy": 0.11981744021177292,
"num_tokens": 1570668.0,
"step": 755
},
{
"entropy": 6.540406513214111,
"epoch": 0.06894049346879536,
"grad_norm": 1.1015625,
"learning_rate": 0.0003795,
"loss": 6.4551,
"mean_token_accuracy": 0.12221564278006554,
"num_tokens": 1581523.0,
"step": 760
},
{
"entropy": 6.523171138763428,
"epoch": 0.06939404934687954,
"grad_norm": 1.1484375,
"learning_rate": 0.000382,
"loss": 6.3504,
"mean_token_accuracy": 0.12361696362495422,
"num_tokens": 1591356.0,
"step": 765
},
{
"entropy": 6.4687755584716795,
"epoch": 0.06984760522496372,
"grad_norm": 1.2421875,
"learning_rate": 0.0003845,
"loss": 6.3504,
"mean_token_accuracy": 0.12131494581699372,
"num_tokens": 1601779.0,
"step": 770
},
{
"entropy": 6.49934868812561,
"epoch": 0.07030116110304789,
"grad_norm": 1.1328125,
"learning_rate": 0.00038700000000000003,
"loss": 6.3423,
"mean_token_accuracy": 0.12172357738018036,
"num_tokens": 1611995.0,
"step": 775
},
{
"entropy": 6.505820989608765,
"epoch": 0.07075471698113207,
"grad_norm": 1.125,
"learning_rate": 0.00038950000000000003,
"loss": 6.3775,
"mean_token_accuracy": 0.11691758409142494,
"num_tokens": 1621834.0,
"step": 780
},
{
"entropy": 6.44027795791626,
"epoch": 0.07120827285921626,
"grad_norm": 1.2734375,
"learning_rate": 0.00039200000000000004,
"loss": 6.4122,
"mean_token_accuracy": 0.11729778945446015,
"num_tokens": 1632881.0,
"step": 785
},
{
"entropy": 6.436710596084595,
"epoch": 0.07166182873730044,
"grad_norm": 1.234375,
"learning_rate": 0.00039450000000000005,
"loss": 6.4103,
"mean_token_accuracy": 0.11784379482269287,
"num_tokens": 1642870.0,
"step": 790
},
{
"entropy": 6.473271608352661,
"epoch": 0.07211538461538461,
"grad_norm": 1.25,
"learning_rate": 0.00039700000000000005,
"loss": 6.2824,
"mean_token_accuracy": 0.1295071743428707,
"num_tokens": 1653390.0,
"step": 795
},
{
"entropy": 6.4784633159637455,
"epoch": 0.07256894049346879,
"grad_norm": 1.1640625,
"learning_rate": 0.0003995,
"loss": 6.3965,
"mean_token_accuracy": 0.12507268264889718,
"num_tokens": 1663382.0,
"step": 800
},
{
"entropy": 6.5645670890808105,
"epoch": 0.07302249637155298,
"grad_norm": 1.0625,
"learning_rate": 0.000402,
"loss": 6.4115,
"mean_token_accuracy": 0.1164341501891613,
"num_tokens": 1673601.0,
"step": 805
},
{
"entropy": 6.325691747665405,
"epoch": 0.07347605224963716,
"grad_norm": 1.328125,
"learning_rate": 0.0004045,
"loss": 6.312,
"mean_token_accuracy": 0.12586987242102624,
"num_tokens": 1683565.0,
"step": 810
},
{
"entropy": 6.426483488082885,
"epoch": 0.07392960812772134,
"grad_norm": 2.03125,
"learning_rate": 0.00040699999999999997,
"loss": 6.3037,
"mean_token_accuracy": 0.12335021495819092,
"num_tokens": 1693636.0,
"step": 815
},
{
"entropy": 6.604643249511719,
"epoch": 0.07438316400580551,
"grad_norm": 1.34375,
"learning_rate": 0.0004095,
"loss": 6.4244,
"mean_token_accuracy": 0.11771097406744957,
"num_tokens": 1703415.0,
"step": 820
},
{
"entropy": 6.449386930465698,
"epoch": 0.0748367198838897,
"grad_norm": 1.2421875,
"learning_rate": 0.000412,
"loss": 6.3872,
"mean_token_accuracy": 0.12382198274135589,
"num_tokens": 1713579.0,
"step": 825
},
{
"entropy": 6.428066158294678,
"epoch": 0.07529027576197388,
"grad_norm": 1.1171875,
"learning_rate": 0.0004145,
"loss": 6.392,
"mean_token_accuracy": 0.11516611129045487,
"num_tokens": 1724583.0,
"step": 830
},
{
"entropy": 6.460074424743652,
"epoch": 0.07574383164005806,
"grad_norm": 1.203125,
"learning_rate": 0.000417,
"loss": 6.3107,
"mean_token_accuracy": 0.12721009701490402,
"num_tokens": 1735293.0,
"step": 835
},
{
"entropy": 6.448750114440918,
"epoch": 0.07619738751814223,
"grad_norm": 1.078125,
"learning_rate": 0.0004195,
"loss": 6.3412,
"mean_token_accuracy": 0.12112727388739586,
"num_tokens": 1745623.0,
"step": 840
},
{
"entropy": 6.439601993560791,
"epoch": 0.07665094339622641,
"grad_norm": 1.0546875,
"learning_rate": 0.000422,
"loss": 6.3679,
"mean_token_accuracy": 0.12129366248846055,
"num_tokens": 1755713.0,
"step": 845
},
{
"entropy": 6.432407855987549,
"epoch": 0.0771044992743106,
"grad_norm": 1.21875,
"learning_rate": 0.0004245,
"loss": 6.3598,
"mean_token_accuracy": 0.11952197775244713,
"num_tokens": 1765264.0,
"step": 850
},
{
"entropy": 6.459024906158447,
"epoch": 0.07755805515239478,
"grad_norm": 1.2109375,
"learning_rate": 0.000427,
"loss": 6.3829,
"mean_token_accuracy": 0.11767206266522408,
"num_tokens": 1775769.0,
"step": 855
},
{
"entropy": 6.405208253860474,
"epoch": 0.07801161103047896,
"grad_norm": 1.1171875,
"learning_rate": 0.0004295,
"loss": 6.3104,
"mean_token_accuracy": 0.12149815633893013,
"num_tokens": 1786245.0,
"step": 860
},
{
"entropy": 6.358996915817261,
"epoch": 0.07846516690856313,
"grad_norm": 1.1953125,
"learning_rate": 0.000432,
"loss": 6.3452,
"mean_token_accuracy": 0.1247312806546688,
"num_tokens": 1796040.0,
"step": 865
},
{
"entropy": 6.409610795974731,
"epoch": 0.07891872278664731,
"grad_norm": 1.109375,
"learning_rate": 0.0004345,
"loss": 6.3672,
"mean_token_accuracy": 0.1209503524005413,
"num_tokens": 1806007.0,
"step": 870
},
{
"entropy": 6.574124145507812,
"epoch": 0.0793722786647315,
"grad_norm": 1.1953125,
"learning_rate": 0.000437,
"loss": 6.4432,
"mean_token_accuracy": 0.11531289964914322,
"num_tokens": 1816935.0,
"step": 875
},
{
"entropy": 6.328552627563477,
"epoch": 0.07982583454281568,
"grad_norm": 1.1875,
"learning_rate": 0.0004395,
"loss": 6.2861,
"mean_token_accuracy": 0.1238143265247345,
"num_tokens": 1827419.0,
"step": 880
},
{
"entropy": 6.487691879272461,
"epoch": 0.08027939042089985,
"grad_norm": 1.203125,
"learning_rate": 0.000442,
"loss": 6.3912,
"mean_token_accuracy": 0.12026699110865594,
"num_tokens": 1837933.0,
"step": 885
},
{
"entropy": 6.400113248825074,
"epoch": 0.08073294629898403,
"grad_norm": 1.125,
"learning_rate": 0.0004445,
"loss": 6.4463,
"mean_token_accuracy": 0.12280047684907913,
"num_tokens": 1848551.0,
"step": 890
},
{
"entropy": 6.482123517990113,
"epoch": 0.08118650217706821,
"grad_norm": 1.28125,
"learning_rate": 0.000447,
"loss": 6.3217,
"mean_token_accuracy": 0.120026945322752,
"num_tokens": 1858380.0,
"step": 895
},
{
"entropy": 6.3884306907653805,
"epoch": 0.0816400580551524,
"grad_norm": 1.2109375,
"learning_rate": 0.00044950000000000003,
"loss": 6.261,
"mean_token_accuracy": 0.12377997189760208,
"num_tokens": 1868291.0,
"step": 900
},
{
"entropy": 6.3934813976287845,
"epoch": 0.08209361393323658,
"grad_norm": 1.078125,
"learning_rate": 0.00045200000000000004,
"loss": 6.3581,
"mean_token_accuracy": 0.12294313907623292,
"num_tokens": 1879219.0,
"step": 905
},
{
"entropy": 6.447540664672852,
"epoch": 0.08254716981132075,
"grad_norm": 1.171875,
"learning_rate": 0.00045450000000000004,
"loss": 6.3587,
"mean_token_accuracy": 0.12189177200198173,
"num_tokens": 1889809.0,
"step": 910
},
{
"entropy": 6.309494495391846,
"epoch": 0.08300072568940493,
"grad_norm": 1.2421875,
"learning_rate": 0.00045700000000000005,
"loss": 6.3265,
"mean_token_accuracy": 0.11860769167542458,
"num_tokens": 1899986.0,
"step": 915
},
{
"entropy": 6.396956491470337,
"epoch": 0.08345428156748912,
"grad_norm": 1.1640625,
"learning_rate": 0.00045950000000000006,
"loss": 6.2974,
"mean_token_accuracy": 0.12569222077727318,
"num_tokens": 1909314.0,
"step": 920
},
{
"entropy": 6.480527687072754,
"epoch": 0.0839078374455733,
"grad_norm": 1.1640625,
"learning_rate": 0.000462,
"loss": 6.3497,
"mean_token_accuracy": 0.12690884619951248,
"num_tokens": 1918443.0,
"step": 925
},
{
"entropy": 6.319289779663086,
"epoch": 0.08436139332365747,
"grad_norm": 1.265625,
"learning_rate": 0.0004645,
"loss": 6.2885,
"mean_token_accuracy": 0.12226624712347985,
"num_tokens": 1928683.0,
"step": 930
},
{
"entropy": 6.2751952648162845,
"epoch": 0.08481494920174165,
"grad_norm": 1.1640625,
"learning_rate": 0.000467,
"loss": 6.2257,
"mean_token_accuracy": 0.12297250479459762,
"num_tokens": 1938293.0,
"step": 935
},
{
"entropy": 6.350123882293701,
"epoch": 0.08526850507982583,
"grad_norm": 1.1328125,
"learning_rate": 0.0004695,
"loss": 6.3185,
"mean_token_accuracy": 0.12192818373441697,
"num_tokens": 1948233.0,
"step": 940
},
{
"entropy": 6.3410228252410885,
"epoch": 0.08572206095791002,
"grad_norm": 1.171875,
"learning_rate": 0.000472,
"loss": 6.2253,
"mean_token_accuracy": 0.12578235045075417,
"num_tokens": 1958611.0,
"step": 945
},
{
"entropy": 6.3663990020751955,
"epoch": 0.0861756168359942,
"grad_norm": 1.0625,
"learning_rate": 0.0004745,
"loss": 6.3898,
"mean_token_accuracy": 0.1257556840777397,
"num_tokens": 1968957.0,
"step": 950
},
{
"entropy": 6.412576484680176,
"epoch": 0.08662917271407837,
"grad_norm": 1.265625,
"learning_rate": 0.000477,
"loss": 6.2254,
"mean_token_accuracy": 0.129705747961998,
"num_tokens": 1978503.0,
"step": 955
},
{
"entropy": 6.213819122314453,
"epoch": 0.08708272859216255,
"grad_norm": 1.1484375,
"learning_rate": 0.0004795,
"loss": 6.2192,
"mean_token_accuracy": 0.12796008363366126,
"num_tokens": 1988458.0,
"step": 960
},
{
"entropy": 6.385213994979859,
"epoch": 0.08753628447024674,
"grad_norm": 1.0703125,
"learning_rate": 0.000482,
"loss": 6.3616,
"mean_token_accuracy": 0.12156156599521636,
"num_tokens": 1999984.0,
"step": 965
},
{
"entropy": 6.23322491645813,
"epoch": 0.08798984034833092,
"grad_norm": 1.125,
"learning_rate": 0.0004845,
"loss": 6.2223,
"mean_token_accuracy": 0.13277149796485901,
"num_tokens": 2010645.0,
"step": 970
},
{
"entropy": 6.410303211212158,
"epoch": 0.08844339622641509,
"grad_norm": 1.1171875,
"learning_rate": 0.000487,
"loss": 6.3426,
"mean_token_accuracy": 0.12236039787530899,
"num_tokens": 2020196.0,
"step": 975
},
{
"entropy": 6.342306232452392,
"epoch": 0.08889695210449927,
"grad_norm": 1.1796875,
"learning_rate": 0.0004895,
"loss": 6.2375,
"mean_token_accuracy": 0.12410336285829544,
"num_tokens": 2030101.0,
"step": 980
},
{
"entropy": 6.2711248874664305,
"epoch": 0.08935050798258345,
"grad_norm": 1.1953125,
"learning_rate": 0.000492,
"loss": 6.3066,
"mean_token_accuracy": 0.12489919736981392,
"num_tokens": 2040454.0,
"step": 985
},
{
"entropy": 6.2733996391296385,
"epoch": 0.08980406386066764,
"grad_norm": 1.09375,
"learning_rate": 0.0004945,
"loss": 6.2668,
"mean_token_accuracy": 0.12355489358305931,
"num_tokens": 2050378.0,
"step": 990
},
{
"entropy": 6.404911756515503,
"epoch": 0.09025761973875182,
"grad_norm": 1.078125,
"learning_rate": 0.000497,
"loss": 6.3254,
"mean_token_accuracy": 0.1241577684879303,
"num_tokens": 2060282.0,
"step": 995
},
{
"entropy": 6.299366188049317,
"epoch": 0.09071117561683599,
"grad_norm": 1.171875,
"learning_rate": 0.0004995,
"loss": 6.3041,
"mean_token_accuracy": 0.12481492385268211,
"num_tokens": 2070563.0,
"step": 1000
},
{
"entropy": 6.384706211090088,
"epoch": 0.09116473149492017,
"grad_norm": 1.125,
"learning_rate": 0.0004999999985110217,
"loss": 6.3913,
"mean_token_accuracy": 0.12241514995694161,
"num_tokens": 2080153.0,
"step": 1005
},
{
"entropy": 6.240800046920777,
"epoch": 0.09161828737300436,
"grad_norm": 1.109375,
"learning_rate": 0.0004999999924620471,
"loss": 6.2265,
"mean_token_accuracy": 0.1292314775288105,
"num_tokens": 2090590.0,
"step": 1010
},
{
"entropy": 6.359726142883301,
"epoch": 0.09207184325108854,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999817600155,
"loss": 6.308,
"mean_token_accuracy": 0.12286418452858924,
"num_tokens": 2101945.0,
"step": 1015
},
{
"entropy": 6.352720880508423,
"epoch": 0.09252539912917271,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999999664049268,
"loss": 6.2828,
"mean_token_accuracy": 0.12876259461045264,
"num_tokens": 2112514.0,
"step": 1020
},
{
"entropy": 6.185303258895874,
"epoch": 0.09297895500725689,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999999463967816,
"loss": 6.2503,
"mean_token_accuracy": 0.12381088435649872,
"num_tokens": 2122296.0,
"step": 1025
},
{
"entropy": 6.4518931865692135,
"epoch": 0.09343251088534107,
"grad_norm": 0.99609375,
"learning_rate": 0.00049999992173558,
"loss": 6.3985,
"mean_token_accuracy": 0.11333390846848487,
"num_tokens": 2134512.0,
"step": 1030
},
{
"entropy": 6.320862913131714,
"epoch": 0.09388606676342526,
"grad_norm": 1.15625,
"learning_rate": 0.0004999998924213228,
"loss": 6.3194,
"mean_token_accuracy": 0.12776733413338662,
"num_tokens": 2143934.0,
"step": 1035
},
{
"entropy": 6.229617166519165,
"epoch": 0.09433962264150944,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999998584540105,
"loss": 6.2042,
"mean_token_accuracy": 0.1237527720630169,
"num_tokens": 2155730.0,
"step": 1040
},
{
"entropy": 6.362137651443481,
"epoch": 0.09479317851959361,
"grad_norm": 1.03125,
"learning_rate": 0.0004999998198336437,
"loss": 6.2322,
"mean_token_accuracy": 0.11689665541052818,
"num_tokens": 2166422.0,
"step": 1045
},
{
"entropy": 6.291525554656983,
"epoch": 0.09524673439767779,
"grad_norm": 1.046875,
"learning_rate": 0.0004999997765602233,
"loss": 6.3434,
"mean_token_accuracy": 0.12079075872898101,
"num_tokens": 2177842.0,
"step": 1050
},
{
"entropy": 6.393911600112915,
"epoch": 0.09570029027576198,
"grad_norm": 1.09375,
"learning_rate": 0.0004999997286337502,
"loss": 6.4109,
"mean_token_accuracy": 0.12265082895755768,
"num_tokens": 2188614.0,
"step": 1055
},
{
"entropy": 6.289089870452881,
"epoch": 0.09615384615384616,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999996760542254,
"loss": 6.191,
"mean_token_accuracy": 0.12765964791178702,
"num_tokens": 2197886.0,
"step": 1060
},
{
"entropy": 6.177943181991577,
"epoch": 0.09660740203193033,
"grad_norm": 1.1015625,
"learning_rate": 0.00049999961882165,
"loss": 6.2718,
"mean_token_accuracy": 0.12614985033869744,
"num_tokens": 2209930.0,
"step": 1065
},
{
"entropy": 6.398168087005615,
"epoch": 0.09706095791001451,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999995569360251,
"loss": 6.2684,
"mean_token_accuracy": 0.12817343920469285,
"num_tokens": 2219767.0,
"step": 1070
},
{
"entropy": 6.201504325866699,
"epoch": 0.0975145137880987,
"grad_norm": 1.0546875,
"learning_rate": 0.000499999490397352,
"loss": 6.2079,
"mean_token_accuracy": 0.1247169628739357,
"num_tokens": 2230082.0,
"step": 1075
},
{
"entropy": 6.2470996379852295,
"epoch": 0.09796806966618288,
"grad_norm": 1.078125,
"learning_rate": 0.0004999994192056321,
"loss": 6.2232,
"mean_token_accuracy": 0.12979647517204285,
"num_tokens": 2240066.0,
"step": 1080
},
{
"entropy": 6.27187294960022,
"epoch": 0.09842162554426705,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999993433608669,
"loss": 6.2689,
"mean_token_accuracy": 0.12491475120186805,
"num_tokens": 2250732.0,
"step": 1085
},
{
"entropy": 6.230953168869019,
"epoch": 0.09887518142235123,
"grad_norm": 1.09375,
"learning_rate": 0.0004999992628630579,
"loss": 6.1459,
"mean_token_accuracy": 0.1282850295305252,
"num_tokens": 2261452.0,
"step": 1090
},
{
"entropy": 6.253374814987183,
"epoch": 0.09932873730043541,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999991777122069,
"loss": 6.2632,
"mean_token_accuracy": 0.12474793419241906,
"num_tokens": 2271929.0,
"step": 1095
},
{
"entropy": 6.200113439559937,
"epoch": 0.0997822931785196,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999990879083156,
"loss": 6.2292,
"mean_token_accuracy": 0.12713540568947793,
"num_tokens": 2282987.0,
"step": 1100
},
{
"entropy": 6.271349954605102,
"epoch": 0.10023584905660378,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999989934513857,
"loss": 6.2242,
"mean_token_accuracy": 0.13376623317599295,
"num_tokens": 2292796.0,
"step": 1105
},
{
"entropy": 6.309939861297607,
"epoch": 0.10068940493468795,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999988943414193,
"loss": 6.2685,
"mean_token_accuracy": 0.1259679116308689,
"num_tokens": 2303057.0,
"step": 1110
},
{
"entropy": 6.165333366394043,
"epoch": 0.10114296081277213,
"grad_norm": 1.140625,
"learning_rate": 0.0004999987905784184,
"loss": 6.2533,
"mean_token_accuracy": 0.126700322329998,
"num_tokens": 2314168.0,
"step": 1115
},
{
"entropy": 6.271357774734497,
"epoch": 0.10159651669085631,
"grad_norm": 1.09375,
"learning_rate": 0.0004999986821623853,
"loss": 6.2141,
"mean_token_accuracy": 0.12545626759529113,
"num_tokens": 2325241.0,
"step": 1120
},
{
"entropy": 6.313687515258789,
"epoch": 0.1020500725689405,
"grad_norm": 1.265625,
"learning_rate": 0.0004999985690933219,
"loss": 6.2417,
"mean_token_accuracy": 0.12604843005537986,
"num_tokens": 2334866.0,
"step": 1125
},
{
"entropy": 6.125355434417725,
"epoch": 0.10250362844702467,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999984513712311,
"loss": 6.2362,
"mean_token_accuracy": 0.1284227713942528,
"num_tokens": 2345422.0,
"step": 1130
},
{
"entropy": 6.371741151809692,
"epoch": 0.10295718432510885,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999983289961146,
"loss": 6.2714,
"mean_token_accuracy": 0.1241820827126503,
"num_tokens": 2356285.0,
"step": 1135
},
{
"entropy": 6.155414390563965,
"epoch": 0.10341074020319303,
"grad_norm": 1.1875,
"learning_rate": 0.0004999982019679755,
"loss": 6.1867,
"mean_token_accuracy": 0.1323022335767746,
"num_tokens": 2366861.0,
"step": 1140
},
{
"entropy": 6.258955812454223,
"epoch": 0.10386429608127722,
"grad_norm": 1.078125,
"learning_rate": 0.0004999980702868164,
"loss": 6.2102,
"mean_token_accuracy": 0.13267010524868966,
"num_tokens": 2376851.0,
"step": 1145
},
{
"entropy": 6.270642280578613,
"epoch": 0.1043178519593614,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999979339526396,
"loss": 6.2633,
"mean_token_accuracy": 0.13088681548833847,
"num_tokens": 2386857.0,
"step": 1150
},
{
"entropy": 6.23888578414917,
"epoch": 0.10477140783744557,
"grad_norm": 1.03125,
"learning_rate": 0.0004999977929654484,
"loss": 6.1974,
"mean_token_accuracy": 0.13122352808713914,
"num_tokens": 2397035.0,
"step": 1155
},
{
"entropy": 6.273252153396607,
"epoch": 0.10522496371552975,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999976473252453,
"loss": 6.2753,
"mean_token_accuracy": 0.12375183030962944,
"num_tokens": 2408177.0,
"step": 1160
},
{
"entropy": 6.322346830368042,
"epoch": 0.10567851959361393,
"grad_norm": 1.09375,
"learning_rate": 0.0004999974970320338,
"loss": 6.2266,
"mean_token_accuracy": 0.12485374137759209,
"num_tokens": 2418445.0,
"step": 1165
},
{
"entropy": 6.186552906036377,
"epoch": 0.10613207547169812,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999973420858165,
"loss": 6.225,
"mean_token_accuracy": 0.12650659754872323,
"num_tokens": 2428990.0,
"step": 1170
},
{
"entropy": 6.227190589904785,
"epoch": 0.10658563134978229,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999971824865968,
"loss": 6.2697,
"mean_token_accuracy": 0.12284443452954293,
"num_tokens": 2439643.0,
"step": 1175
},
{
"entropy": 6.330835676193237,
"epoch": 0.10703918722786647,
"grad_norm": 1.03125,
"learning_rate": 0.0004999970182343782,
"loss": 6.3127,
"mean_token_accuracy": 0.12625510469079018,
"num_tokens": 2450691.0,
"step": 1180
},
{
"entropy": 6.270051288604736,
"epoch": 0.10749274310595065,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999968493291638,
"loss": 6.2433,
"mean_token_accuracy": 0.12399949803948403,
"num_tokens": 2460706.0,
"step": 1185
},
{
"entropy": 6.072198438644409,
"epoch": 0.10794629898403484,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999966757709573,
"loss": 6.1397,
"mean_token_accuracy": 0.1322955772280693,
"num_tokens": 2470947.0,
"step": 1190
},
{
"entropy": 6.232493448257446,
"epoch": 0.10839985486211902,
"grad_norm": 1.109375,
"learning_rate": 0.0004999964975597622,
"loss": 6.0869,
"mean_token_accuracy": 0.13433996289968492,
"num_tokens": 2481132.0,
"step": 1195
},
{
"entropy": 6.181258201599121,
"epoch": 0.10885341074020319,
"grad_norm": 1.125,
"learning_rate": 0.0004999963146955821,
"loss": 6.1403,
"mean_token_accuracy": 0.133623556047678,
"num_tokens": 2491701.0,
"step": 1200
},
{
"entropy": 6.120849657058716,
"epoch": 0.10930696661828737,
"grad_norm": 1.0078125,
"learning_rate": 0.000499996127178421,
"loss": 6.2375,
"mean_token_accuracy": 0.12445509806275368,
"num_tokens": 2502360.0,
"step": 1205
},
{
"entropy": 6.195035171508789,
"epoch": 0.10976052249637155,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999959350082825,
"loss": 6.097,
"mean_token_accuracy": 0.12750829607248307,
"num_tokens": 2512140.0,
"step": 1210
},
{
"entropy": 6.253734350204468,
"epoch": 0.11021407837445574,
"grad_norm": 1.09375,
"learning_rate": 0.0004999957381851709,
"loss": 6.2466,
"mean_token_accuracy": 0.12660726830363273,
"num_tokens": 2522721.0,
"step": 1215
},
{
"entropy": 6.106527471542359,
"epoch": 0.1106676342525399,
"grad_norm": 1.140625,
"learning_rate": 0.0004999955367090901,
"loss": 6.108,
"mean_token_accuracy": 0.13220437839627266,
"num_tokens": 2531786.0,
"step": 1220
},
{
"entropy": 6.26336636543274,
"epoch": 0.11112119013062409,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999953305800441,
"loss": 6.2201,
"mean_token_accuracy": 0.12029675543308258,
"num_tokens": 2542895.0,
"step": 1225
},
{
"entropy": 6.13406777381897,
"epoch": 0.11157474600870827,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999951197980374,
"loss": 6.2011,
"mean_token_accuracy": 0.12797781825065613,
"num_tokens": 2554120.0,
"step": 1230
},
{
"entropy": 6.287543106079101,
"epoch": 0.11202830188679246,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999949043630744,
"loss": 6.2446,
"mean_token_accuracy": 0.12370867282152176,
"num_tokens": 2563935.0,
"step": 1235
},
{
"entropy": 6.092472743988037,
"epoch": 0.11248185776487664,
"grad_norm": 1.125,
"learning_rate": 0.0004999946842751593,
"loss": 6.1501,
"mean_token_accuracy": 0.12944189086556435,
"num_tokens": 2574465.0,
"step": 1240
},
{
"entropy": 6.177829313278198,
"epoch": 0.11293541364296081,
"grad_norm": 1.046875,
"learning_rate": 0.0004999944595342968,
"loss": 6.0986,
"mean_token_accuracy": 0.13265978991985322,
"num_tokens": 2584694.0,
"step": 1245
},
{
"entropy": 6.156110572814941,
"epoch": 0.11338896952104499,
"grad_norm": 1.0625,
"learning_rate": 0.0004999942301404916,
"loss": 6.0902,
"mean_token_accuracy": 0.13416919559240342,
"num_tokens": 2594540.0,
"step": 1250
},
{
"entropy": 6.102447032928467,
"epoch": 0.11384252539912917,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999939960937484,
"loss": 6.022,
"mean_token_accuracy": 0.13621553182601928,
"num_tokens": 2604896.0,
"step": 1255
},
{
"entropy": 6.183004379272461,
"epoch": 0.11429608127721336,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999937573940721,
"loss": 6.2082,
"mean_token_accuracy": 0.13601385727524756,
"num_tokens": 2614143.0,
"step": 1260
},
{
"entropy": 6.0483887672424315,
"epoch": 0.11474963715529753,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999935140414674,
"loss": 6.0993,
"mean_token_accuracy": 0.12777388021349906,
"num_tokens": 2623999.0,
"step": 1265
},
{
"entropy": 6.307592487335205,
"epoch": 0.11520319303338171,
"grad_norm": 1.0625,
"learning_rate": 0.0004999932660359395,
"loss": 6.3112,
"mean_token_accuracy": 0.12414655461907387,
"num_tokens": 2636208.0,
"step": 1270
},
{
"entropy": 6.134375429153442,
"epoch": 0.11565674891146589,
"grad_norm": 1.109375,
"learning_rate": 0.0004999930133774936,
"loss": 6.1749,
"mean_token_accuracy": 0.13444133400917052,
"num_tokens": 2645854.0,
"step": 1275
},
{
"entropy": 6.208854484558105,
"epoch": 0.11611030478955008,
"grad_norm": 1.125,
"learning_rate": 0.0004999927560661348,
"loss": 6.1606,
"mean_token_accuracy": 0.14128717482089997,
"num_tokens": 2655877.0,
"step": 1280
},
{
"entropy": 6.193210935592651,
"epoch": 0.11656386066763426,
"grad_norm": 1.125,
"learning_rate": 0.0004999924941018685,
"loss": 6.1854,
"mean_token_accuracy": 0.13444199338555335,
"num_tokens": 2666157.0,
"step": 1285
},
{
"entropy": 6.116014003753662,
"epoch": 0.11701741654571843,
"grad_norm": 1.046875,
"learning_rate": 0.0004999922274847,
"loss": 6.1024,
"mean_token_accuracy": 0.13178513795137406,
"num_tokens": 2676647.0,
"step": 1290
},
{
"entropy": 6.093279409408569,
"epoch": 0.11747097242380261,
"grad_norm": 1.15625,
"learning_rate": 0.000499991956214635,
"loss": 6.1301,
"mean_token_accuracy": 0.14098571091890336,
"num_tokens": 2686417.0,
"step": 1295
},
{
"entropy": 6.198765611648559,
"epoch": 0.1179245283018868,
"grad_norm": 1.0234375,
"learning_rate": 0.000499991680291679,
"loss": 6.1374,
"mean_token_accuracy": 0.12717395946383475,
"num_tokens": 2697540.0,
"step": 1300
},
{
"entropy": 6.106028366088867,
"epoch": 0.11837808417997098,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999913997158378,
"loss": 6.1846,
"mean_token_accuracy": 0.12284301593899727,
"num_tokens": 2709397.0,
"step": 1305
},
{
"entropy": 6.217894172668457,
"epoch": 0.11883164005805515,
"grad_norm": 1.09375,
"learning_rate": 0.0004999911144871169,
"loss": 6.1486,
"mean_token_accuracy": 0.12695815786719322,
"num_tokens": 2720430.0,
"step": 1310
},
{
"entropy": 6.11812047958374,
"epoch": 0.11928519593613933,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999908246055226,
"loss": 6.1094,
"mean_token_accuracy": 0.12854591459035875,
"num_tokens": 2730648.0,
"step": 1315
},
{
"entropy": 6.136201524734497,
"epoch": 0.11973875181422351,
"grad_norm": 1.046875,
"learning_rate": 0.0004999905300710607,
"loss": 6.0688,
"mean_token_accuracy": 0.13269162625074388,
"num_tokens": 2739979.0,
"step": 1320
},
{
"entropy": 6.138657140731811,
"epoch": 0.1201923076923077,
"grad_norm": 1.234375,
"learning_rate": 0.0004999902308837373,
"loss": 6.2314,
"mean_token_accuracy": 0.12551036328077317,
"num_tokens": 2750329.0,
"step": 1325
},
{
"entropy": 6.094318199157715,
"epoch": 0.12064586357039188,
"grad_norm": 1.09375,
"learning_rate": 0.0004999899270435584,
"loss": 6.1032,
"mean_token_accuracy": 0.13775131702423096,
"num_tokens": 2760535.0,
"step": 1330
},
{
"entropy": 6.1581787109375,
"epoch": 0.12109941944847605,
"grad_norm": 1.140625,
"learning_rate": 0.0004999896185505307,
"loss": 6.128,
"mean_token_accuracy": 0.12702896371483802,
"num_tokens": 2770470.0,
"step": 1335
},
{
"entropy": 6.087944221496582,
"epoch": 0.12155297532656023,
"grad_norm": 1.03125,
"learning_rate": 0.0004999893054046603,
"loss": 6.1455,
"mean_token_accuracy": 0.12925868555903436,
"num_tokens": 2781506.0,
"step": 1340
},
{
"entropy": 6.0965156078338625,
"epoch": 0.12200653120464441,
"grad_norm": 1.140625,
"learning_rate": 0.0004999889876059537,
"loss": 6.022,
"mean_token_accuracy": 0.1381001003086567,
"num_tokens": 2791245.0,
"step": 1345
},
{
"entropy": 6.076093435287476,
"epoch": 0.1224600870827286,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999886651544176,
"loss": 6.0262,
"mean_token_accuracy": 0.1368500992655754,
"num_tokens": 2801680.0,
"step": 1350
},
{
"entropy": 6.1108323097229,
"epoch": 0.12291364296081277,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999883380500584,
"loss": 6.1129,
"mean_token_accuracy": 0.13711178302764893,
"num_tokens": 2812805.0,
"step": 1355
},
{
"entropy": 6.181576919555664,
"epoch": 0.12336719883889695,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999880062928831,
"loss": 6.1608,
"mean_token_accuracy": 0.129413640499115,
"num_tokens": 2822834.0,
"step": 1360
},
{
"entropy": 6.089189243316651,
"epoch": 0.12382075471698113,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999876698828985,
"loss": 6.1075,
"mean_token_accuracy": 0.13718436509370804,
"num_tokens": 2832312.0,
"step": 1365
},
{
"entropy": 6.152549886703492,
"epoch": 0.12427431059506532,
"grad_norm": 1.015625,
"learning_rate": 0.0004999873288201116,
"loss": 6.1187,
"mean_token_accuracy": 0.14113932773470877,
"num_tokens": 2842998.0,
"step": 1370
},
{
"entropy": 6.136705541610718,
"epoch": 0.1247278664731495,
"grad_norm": 1.578125,
"learning_rate": 0.0004999869831045294,
"loss": 6.1438,
"mean_token_accuracy": 0.13647317588329316,
"num_tokens": 2854218.0,
"step": 1375
},
{
"entropy": 6.104028177261353,
"epoch": 0.12518142235123367,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999866327361589,
"loss": 6.0237,
"mean_token_accuracy": 0.1436094231903553,
"num_tokens": 2864484.0,
"step": 1380
},
{
"entropy": 6.104306650161743,
"epoch": 0.12563497822931785,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999862777150077,
"loss": 6.1447,
"mean_token_accuracy": 0.13356181010603904,
"num_tokens": 2873800.0,
"step": 1385
},
{
"entropy": 6.142058610916138,
"epoch": 0.12608853410740203,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999859180410829,
"loss": 6.1636,
"mean_token_accuracy": 0.12791035771369935,
"num_tokens": 2884247.0,
"step": 1390
},
{
"entropy": 6.0876030921936035,
"epoch": 0.12654208998548622,
"grad_norm": 0.984375,
"learning_rate": 0.0004999855537143919,
"loss": 6.0556,
"mean_token_accuracy": 0.13154122978448868,
"num_tokens": 2894739.0,
"step": 1395
},
{
"entropy": 6.03957347869873,
"epoch": 0.1269956458635704,
"grad_norm": 1.046875,
"learning_rate": 0.0004999851847349425,
"loss": 6.1199,
"mean_token_accuracy": 0.13250315189361572,
"num_tokens": 2904376.0,
"step": 1400
},
{
"entropy": 6.174634170532227,
"epoch": 0.12744920174165458,
"grad_norm": 1.015625,
"learning_rate": 0.0004999848111027419,
"loss": 6.1702,
"mean_token_accuracy": 0.1307752922177315,
"num_tokens": 2914304.0,
"step": 1405
},
{
"entropy": 6.1659255027771,
"epoch": 0.12790275761973874,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999844328177983,
"loss": 6.0538,
"mean_token_accuracy": 0.1304120510816574,
"num_tokens": 2924400.0,
"step": 1410
},
{
"entropy": 6.208686876296997,
"epoch": 0.12835631349782292,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999840498801191,
"loss": 6.2771,
"mean_token_accuracy": 0.13209760636091233,
"num_tokens": 2935040.0,
"step": 1415
},
{
"entropy": 6.04742112159729,
"epoch": 0.1288098693759071,
"grad_norm": 1.046875,
"learning_rate": 0.0004999836622897126,
"loss": 6.0794,
"mean_token_accuracy": 0.13408395498991013,
"num_tokens": 2945536.0,
"step": 1420
},
{
"entropy": 6.038312911987305,
"epoch": 0.1292634252539913,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999832700465865,
"loss": 5.9749,
"mean_token_accuracy": 0.13549989610910415,
"num_tokens": 2954969.0,
"step": 1425
},
{
"entropy": 6.066872215270996,
"epoch": 0.12971698113207547,
"grad_norm": 1.078125,
"learning_rate": 0.0004999828731507491,
"loss": 6.112,
"mean_token_accuracy": 0.12939976528286934,
"num_tokens": 2966218.0,
"step": 1430
},
{
"entropy": 6.094667911529541,
"epoch": 0.13017053701015965,
"grad_norm": 1.125,
"learning_rate": 0.0004999824716022085,
"loss": 6.136,
"mean_token_accuracy": 0.12750567644834518,
"num_tokens": 2976640.0,
"step": 1435
},
{
"entropy": 6.105727338790894,
"epoch": 0.13062409288824384,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999820654009731,
"loss": 6.0905,
"mean_token_accuracy": 0.13567687720060348,
"num_tokens": 2985883.0,
"step": 1440
},
{
"entropy": 6.088870000839234,
"epoch": 0.13107764876632802,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999816545470513,
"loss": 6.1031,
"mean_token_accuracy": 0.12947410196065903,
"num_tokens": 2996695.0,
"step": 1445
},
{
"entropy": 6.171409797668457,
"epoch": 0.1315312046444122,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999812390404514,
"loss": 6.1171,
"mean_token_accuracy": 0.1340610735118389,
"num_tokens": 3007500.0,
"step": 1450
},
{
"entropy": 5.898324060440063,
"epoch": 0.13198476052249636,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999808188811823,
"loss": 5.9616,
"mean_token_accuracy": 0.1348599299788475,
"num_tokens": 3017299.0,
"step": 1455
},
{
"entropy": 6.216505908966065,
"epoch": 0.13243831640058054,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999803940692524,
"loss": 6.2449,
"mean_token_accuracy": 0.12761179730296135,
"num_tokens": 3028389.0,
"step": 1460
},
{
"entropy": 6.154873561859131,
"epoch": 0.13289187227866472,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999799646046707,
"loss": 6.0668,
"mean_token_accuracy": 0.1371331825852394,
"num_tokens": 3038883.0,
"step": 1465
},
{
"entropy": 6.139013624191284,
"epoch": 0.1333454281567489,
"grad_norm": 0.984375,
"learning_rate": 0.0004999795304874461,
"loss": 6.1742,
"mean_token_accuracy": 0.1367236204445362,
"num_tokens": 3049904.0,
"step": 1470
},
{
"entropy": 6.055725383758545,
"epoch": 0.1337989840348331,
"grad_norm": 1.078125,
"learning_rate": 0.0004999790917175873,
"loss": 6.026,
"mean_token_accuracy": 0.133791683614254,
"num_tokens": 3059788.0,
"step": 1475
},
{
"entropy": 6.11804518699646,
"epoch": 0.13425253991291727,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999786482951036,
"loss": 6.0123,
"mean_token_accuracy": 0.13600108176469802,
"num_tokens": 3070413.0,
"step": 1480
},
{
"entropy": 6.0903236865997314,
"epoch": 0.13470609579100146,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999782002200041,
"loss": 6.1348,
"mean_token_accuracy": 0.1337863601744175,
"num_tokens": 3081093.0,
"step": 1485
},
{
"entropy": 6.083447599411011,
"epoch": 0.13515965166908564,
"grad_norm": 1.078125,
"learning_rate": 0.0004999777474922982,
"loss": 6.0731,
"mean_token_accuracy": 0.12955736368894577,
"num_tokens": 3090594.0,
"step": 1490
},
{
"entropy": 6.138653802871704,
"epoch": 0.13561320754716982,
"grad_norm": 1.046875,
"learning_rate": 0.0004999772901119951,
"loss": 6.1789,
"mean_token_accuracy": 0.12780415564775466,
"num_tokens": 3101061.0,
"step": 1495
},
{
"entropy": 6.0858407497406,
"epoch": 0.13606676342525398,
"grad_norm": 1.0,
"learning_rate": 0.0004999768280791043,
"loss": 6.0895,
"mean_token_accuracy": 0.1373533673584461,
"num_tokens": 3111715.0,
"step": 1500
},
{
"entropy": 6.09755220413208,
"epoch": 0.13652031930333816,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999763613936352,
"loss": 5.9993,
"mean_token_accuracy": 0.14067018330097197,
"num_tokens": 3121378.0,
"step": 1505
},
{
"entropy": 5.970461797714234,
"epoch": 0.13697387518142234,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999758900555979,
"loss": 5.9441,
"mean_token_accuracy": 0.14521051645278932,
"num_tokens": 3131229.0,
"step": 1510
},
{
"entropy": 6.036447477340698,
"epoch": 0.13742743105950653,
"grad_norm": 1.1875,
"learning_rate": 0.0004999754140650017,
"loss": 6.0404,
"mean_token_accuracy": 0.1319979339838028,
"num_tokens": 3140859.0,
"step": 1515
},
{
"entropy": 6.097309494018555,
"epoch": 0.1378809869375907,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999749334218567,
"loss": 6.051,
"mean_token_accuracy": 0.1364140808582306,
"num_tokens": 3151692.0,
"step": 1520
},
{
"entropy": 5.99874792098999,
"epoch": 0.1383345428156749,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999744481261726,
"loss": 6.1157,
"mean_token_accuracy": 0.1304463878273964,
"num_tokens": 3161879.0,
"step": 1525
},
{
"entropy": 6.048851442337036,
"epoch": 0.13878809869375908,
"grad_norm": 1.28125,
"learning_rate": 0.0004999739581779597,
"loss": 5.982,
"mean_token_accuracy": 0.14158392399549485,
"num_tokens": 3171282.0,
"step": 1530
},
{
"entropy": 6.069761800765991,
"epoch": 0.13924165457184326,
"grad_norm": 1.0546875,
"learning_rate": 0.000499973463577228,
"loss": 6.0283,
"mean_token_accuracy": 0.1374294213950634,
"num_tokens": 3181398.0,
"step": 1535
},
{
"entropy": 6.063587093353272,
"epoch": 0.13969521044992744,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999729643239877,
"loss": 6.1616,
"mean_token_accuracy": 0.1350327856838703,
"num_tokens": 3191431.0,
"step": 1540
},
{
"entropy": 6.10133204460144,
"epoch": 0.1401487663280116,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999724604182492,
"loss": 6.1621,
"mean_token_accuracy": 0.12566626071929932,
"num_tokens": 3202147.0,
"step": 1545
},
{
"entropy": 6.160943412780762,
"epoch": 0.14060232220609578,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999719518600229,
"loss": 6.0291,
"mean_token_accuracy": 0.1362243190407753,
"num_tokens": 3212428.0,
"step": 1550
},
{
"entropy": 6.014663982391357,
"epoch": 0.14105587808417996,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999714386493192,
"loss": 6.0893,
"mean_token_accuracy": 0.13574016094207764,
"num_tokens": 3222099.0,
"step": 1555
},
{
"entropy": 6.097148656845093,
"epoch": 0.14150943396226415,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999709207861489,
"loss": 6.0313,
"mean_token_accuracy": 0.13692608699202538,
"num_tokens": 3232079.0,
"step": 1560
},
{
"entropy": 5.976779842376709,
"epoch": 0.14196298984034833,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999703982705227,
"loss": 5.9373,
"mean_token_accuracy": 0.1448259696364403,
"num_tokens": 3241762.0,
"step": 1565
},
{
"entropy": 6.014270257949829,
"epoch": 0.1424165457184325,
"grad_norm": 1.15625,
"learning_rate": 0.0004999698711024513,
"loss": 6.0239,
"mean_token_accuracy": 0.13922590911388397,
"num_tokens": 3252571.0,
"step": 1570
},
{
"entropy": 6.057532691955567,
"epoch": 0.1428701015965167,
"grad_norm": 1.03125,
"learning_rate": 0.0004999693392819456,
"loss": 6.0752,
"mean_token_accuracy": 0.13099500313401222,
"num_tokens": 3263542.0,
"step": 1575
},
{
"entropy": 6.098192596435547,
"epoch": 0.14332365747460088,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999688028090166,
"loss": 6.1447,
"mean_token_accuracy": 0.1325105018913746,
"num_tokens": 3273295.0,
"step": 1580
},
{
"entropy": 6.178218173980713,
"epoch": 0.14377721335268506,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999682616836755,
"loss": 6.1644,
"mean_token_accuracy": 0.1343841977417469,
"num_tokens": 3283605.0,
"step": 1585
},
{
"entropy": 6.034059715270996,
"epoch": 0.14423076923076922,
"grad_norm": 1.078125,
"learning_rate": 0.0004999677159059334,
"loss": 6.0018,
"mean_token_accuracy": 0.13836689442396163,
"num_tokens": 3293837.0,
"step": 1590
},
{
"entropy": 5.9601891040802,
"epoch": 0.1446843251088534,
"grad_norm": 1.078125,
"learning_rate": 0.0004999671654758016,
"loss": 6.0074,
"mean_token_accuracy": 0.1401748426258564,
"num_tokens": 3304131.0,
"step": 1595
},
{
"entropy": 6.019832801818848,
"epoch": 0.14513788098693758,
"grad_norm": 1.59375,
"learning_rate": 0.0004999666103932915,
"loss": 6.0457,
"mean_token_accuracy": 0.13282065689563752,
"num_tokens": 3313543.0,
"step": 1600
},
{
"entropy": 5.98559947013855,
"epoch": 0.14559143686502177,
"grad_norm": 1.078125,
"learning_rate": 0.0004999660506584145,
"loss": 6.0292,
"mean_token_accuracy": 0.13611204028129578,
"num_tokens": 3324487.0,
"step": 1605
},
{
"entropy": 6.09367094039917,
"epoch": 0.14604499274310595,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999654862711823,
"loss": 6.0432,
"mean_token_accuracy": 0.13722373321652412,
"num_tokens": 3335094.0,
"step": 1610
},
{
"entropy": 6.020828676223755,
"epoch": 0.14649854862119013,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999649172316065,
"loss": 5.9875,
"mean_token_accuracy": 0.13779204934835435,
"num_tokens": 3344252.0,
"step": 1615
},
{
"entropy": 6.04573392868042,
"epoch": 0.14695210449927432,
"grad_norm": 1.0859375,
"learning_rate": 0.000499964343539699,
"loss": 6.0523,
"mean_token_accuracy": 0.134325809776783,
"num_tokens": 3354265.0,
"step": 1620
},
{
"entropy": 6.130579662322998,
"epoch": 0.1474056603773585,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999637651954714,
"loss": 6.1609,
"mean_token_accuracy": 0.13543225973844528,
"num_tokens": 3364616.0,
"step": 1625
},
{
"entropy": 6.084848880767822,
"epoch": 0.14785921625544268,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999631821989358,
"loss": 6.0887,
"mean_token_accuracy": 0.1368525169789791,
"num_tokens": 3373650.0,
"step": 1630
},
{
"entropy": 6.006167268753051,
"epoch": 0.14831277213352684,
"grad_norm": 1.15625,
"learning_rate": 0.0004999625945501043,
"loss": 5.9953,
"mean_token_accuracy": 0.140744087100029,
"num_tokens": 3384590.0,
"step": 1635
},
{
"entropy": 6.116048669815063,
"epoch": 0.14876632801161102,
"grad_norm": 1.1796875,
"learning_rate": 0.000499962002248989,
"loss": 6.0452,
"mean_token_accuracy": 0.13539330065250396,
"num_tokens": 3395616.0,
"step": 1640
},
{
"entropy": 5.996517515182495,
"epoch": 0.1492198838896952,
"grad_norm": 1.015625,
"learning_rate": 0.0004999614052956024,
"loss": 6.032,
"mean_token_accuracy": 0.1371271312236786,
"num_tokens": 3405810.0,
"step": 1645
},
{
"entropy": 6.041546487808228,
"epoch": 0.1496734397677794,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999608036899563,
"loss": 6.0743,
"mean_token_accuracy": 0.13593365326523782,
"num_tokens": 3415932.0,
"step": 1650
},
{
"entropy": 6.038019609451294,
"epoch": 0.15012699564586357,
"grad_norm": 1.1875,
"learning_rate": 0.0004999601974320636,
"loss": 6.011,
"mean_token_accuracy": 0.13596160858869552,
"num_tokens": 3425626.0,
"step": 1655
},
{
"entropy": 6.005490827560425,
"epoch": 0.15058055152394775,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999595865219367,
"loss": 6.0204,
"mean_token_accuracy": 0.13287046179175377,
"num_tokens": 3434837.0,
"step": 1660
},
{
"entropy": 6.034183692932129,
"epoch": 0.15103410740203194,
"grad_norm": 1.078125,
"learning_rate": 0.0004999589709595881,
"loss": 6.0485,
"mean_token_accuracy": 0.13829124122858047,
"num_tokens": 3445238.0,
"step": 1665
},
{
"entropy": 5.948350954055786,
"epoch": 0.15148766328011612,
"grad_norm": 1.09375,
"learning_rate": 0.0004999583507450308,
"loss": 6.0381,
"mean_token_accuracy": 0.13264935240149497,
"num_tokens": 3456738.0,
"step": 1670
},
{
"entropy": 5.909289693832397,
"epoch": 0.1519412191582003,
"grad_norm": 1.046875,
"learning_rate": 0.0004999577258782774,
"loss": 5.9012,
"mean_token_accuracy": 0.13860071077942848,
"num_tokens": 3466296.0,
"step": 1675
},
{
"entropy": 6.124201059341431,
"epoch": 0.15239477503628446,
"grad_norm": 1.046875,
"learning_rate": 0.000499957096359341,
"loss": 6.1028,
"mean_token_accuracy": 0.13263867199420928,
"num_tokens": 3476989.0,
"step": 1680
},
{
"entropy": 5.982616138458252,
"epoch": 0.15284833091436864,
"grad_norm": 1.203125,
"learning_rate": 0.0004999564621882343,
"loss": 5.9443,
"mean_token_accuracy": 0.13828226998448373,
"num_tokens": 3487050.0,
"step": 1685
},
{
"entropy": 6.07118854522705,
"epoch": 0.15330188679245282,
"grad_norm": 1.078125,
"learning_rate": 0.0004999558233649708,
"loss": 6.1489,
"mean_token_accuracy": 0.1293186977505684,
"num_tokens": 3497777.0,
"step": 1690
},
{
"entropy": 6.066112852096557,
"epoch": 0.153755442670537,
"grad_norm": 1.0625,
"learning_rate": 0.0004999551798895636,
"loss": 6.0597,
"mean_token_accuracy": 0.13373040035367012,
"num_tokens": 3509066.0,
"step": 1695
},
{
"entropy": 6.0372473239898685,
"epoch": 0.1542089985486212,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999545317620258,
"loss": 5.9953,
"mean_token_accuracy": 0.13631840348243712,
"num_tokens": 3519567.0,
"step": 1700
},
{
"entropy": 5.967922782897949,
"epoch": 0.15466255442670537,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999538789823711,
"loss": 6.0394,
"mean_token_accuracy": 0.13856917023658752,
"num_tokens": 3529402.0,
"step": 1705
},
{
"entropy": 6.11129059791565,
"epoch": 0.15511611030478956,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999532215506127,
"loss": 6.0335,
"mean_token_accuracy": 0.14377474635839463,
"num_tokens": 3539805.0,
"step": 1710
},
{
"entropy": 5.909777498245239,
"epoch": 0.15556966618287374,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999525594667645,
"loss": 5.9067,
"mean_token_accuracy": 0.142856465280056,
"num_tokens": 3550477.0,
"step": 1715
},
{
"entropy": 5.982000541687012,
"epoch": 0.15602322206095792,
"grad_norm": 1.0859375,
"learning_rate": 0.00049995189273084,
"loss": 6.0103,
"mean_token_accuracy": 0.13765751868486403,
"num_tokens": 3560729.0,
"step": 1720
},
{
"entropy": 6.05636043548584,
"epoch": 0.15647677793904208,
"grad_norm": 1.1015625,
"learning_rate": 0.000499951221342853,
"loss": 6.0812,
"mean_token_accuracy": 0.1257948972284794,
"num_tokens": 3571538.0,
"step": 1725
},
{
"entropy": 6.04515323638916,
"epoch": 0.15693033381712626,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999505453028174,
"loss": 6.0887,
"mean_token_accuracy": 0.13670211359858514,
"num_tokens": 3583128.0,
"step": 1730
},
{
"entropy": 6.103114271163941,
"epoch": 0.15738388969521044,
"grad_norm": 1.03125,
"learning_rate": 0.0004999498646107472,
"loss": 6.0257,
"mean_token_accuracy": 0.13807276338338853,
"num_tokens": 3594358.0,
"step": 1735
},
{
"entropy": 5.9683116436004635,
"epoch": 0.15783744557329463,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999491792666566,
"loss": 5.9489,
"mean_token_accuracy": 0.14279852136969567,
"num_tokens": 3604345.0,
"step": 1740
},
{
"entropy": 6.023993253707886,
"epoch": 0.1582910014513788,
"grad_norm": 1.109375,
"learning_rate": 0.0004999484892705597,
"loss": 5.9883,
"mean_token_accuracy": 0.1382973462343216,
"num_tokens": 3616129.0,
"step": 1745
},
{
"entropy": 6.001973867416382,
"epoch": 0.158744557329463,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999477946224705,
"loss": 6.0409,
"mean_token_accuracy": 0.13687437549233436,
"num_tokens": 3627458.0,
"step": 1750
},
{
"entropy": 6.058828830718994,
"epoch": 0.15919811320754718,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999470953224037,
"loss": 5.9206,
"mean_token_accuracy": 0.14193053394556046,
"num_tokens": 3636047.0,
"step": 1755
},
{
"entropy": 5.878429508209228,
"epoch": 0.15965166908563136,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999463913703735,
"loss": 5.9243,
"mean_token_accuracy": 0.13841559886932372,
"num_tokens": 3646930.0,
"step": 1760
},
{
"entropy": 6.051437282562256,
"epoch": 0.16010522496371554,
"grad_norm": 1.09375,
"learning_rate": 0.0004999456827663948,
"loss": 6.0061,
"mean_token_accuracy": 0.14167415127158164,
"num_tokens": 3657279.0,
"step": 1765
},
{
"entropy": 5.9463623523712155,
"epoch": 0.1605587808417997,
"grad_norm": 1.1484375,
"learning_rate": 0.000499944969510482,
"loss": 5.911,
"mean_token_accuracy": 0.13895074278116226,
"num_tokens": 3667070.0,
"step": 1770
},
{
"entropy": 5.968365859985352,
"epoch": 0.16101233671988388,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999442516026498,
"loss": 5.9461,
"mean_token_accuracy": 0.1364276871085167,
"num_tokens": 3677560.0,
"step": 1775
},
{
"entropy": 5.948978471755981,
"epoch": 0.16146589259796806,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999435290429133,
"loss": 5.899,
"mean_token_accuracy": 0.14505873918533324,
"num_tokens": 3686701.0,
"step": 1780
},
{
"entropy": 5.980872631072998,
"epoch": 0.16191944847605225,
"grad_norm": 1.1875,
"learning_rate": 0.0004999428018312872,
"loss": 5.9934,
"mean_token_accuracy": 0.1406108967959881,
"num_tokens": 3696338.0,
"step": 1785
},
{
"entropy": 5.991557312011719,
"epoch": 0.16237300435413643,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999420699677867,
"loss": 5.9721,
"mean_token_accuracy": 0.14421124309301375,
"num_tokens": 3705752.0,
"step": 1790
},
{
"entropy": 6.0219776153564455,
"epoch": 0.1628265602322206,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999413334524269,
"loss": 6.0601,
"mean_token_accuracy": 0.13283660635352135,
"num_tokens": 3716225.0,
"step": 1795
},
{
"entropy": 6.030018615722656,
"epoch": 0.1632801161103048,
"grad_norm": 1.265625,
"learning_rate": 0.0004999405922852229,
"loss": 5.9367,
"mean_token_accuracy": 0.1414524123072624,
"num_tokens": 3726593.0,
"step": 1800
},
{
"entropy": 5.9711730003356935,
"epoch": 0.16373367198838898,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999398464661901,
"loss": 6.0046,
"mean_token_accuracy": 0.13162903413176535,
"num_tokens": 3737563.0,
"step": 1805
},
{
"entropy": 6.145048904418945,
"epoch": 0.16418722786647316,
"grad_norm": 1.3125,
"learning_rate": 0.0004999390959953441,
"loss": 6.0538,
"mean_token_accuracy": 0.1345907710492611,
"num_tokens": 3747752.0,
"step": 1810
},
{
"entropy": 5.8741460800170895,
"epoch": 0.16464078374455732,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999383408727002,
"loss": 5.9927,
"mean_token_accuracy": 0.13676512464880944,
"num_tokens": 3758795.0,
"step": 1815
},
{
"entropy": 6.02309947013855,
"epoch": 0.1650943396226415,
"grad_norm": 1.265625,
"learning_rate": 0.0004999375810982741,
"loss": 5.9739,
"mean_token_accuracy": 0.1364384949207306,
"num_tokens": 3768690.0,
"step": 1820
},
{
"entropy": 5.949415493011474,
"epoch": 0.16554789550072568,
"grad_norm": 1.0625,
"learning_rate": 0.0004999368166720815,
"loss": 6.0354,
"mean_token_accuracy": 0.1367568477988243,
"num_tokens": 3779018.0,
"step": 1825
},
{
"entropy": 6.097243118286133,
"epoch": 0.16600145137880987,
"grad_norm": 1.046875,
"learning_rate": 0.0004999360475941382,
"loss": 5.9274,
"mean_token_accuracy": 0.14494050815701484,
"num_tokens": 3788985.0,
"step": 1830
},
{
"entropy": 5.9869836330413815,
"epoch": 0.16645500725689405,
"grad_norm": 1.1015625,
"learning_rate": 0.00049993527386446,
"loss": 6.08,
"mean_token_accuracy": 0.12971132546663283,
"num_tokens": 3800787.0,
"step": 1835
},
{
"entropy": 6.007747030258178,
"epoch": 0.16690856313497823,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999344954830633,
"loss": 5.9478,
"mean_token_accuracy": 0.13803392350673677,
"num_tokens": 3811381.0,
"step": 1840
},
{
"entropy": 6.022607088088989,
"epoch": 0.16736211901306242,
"grad_norm": 1.15625,
"learning_rate": 0.0004999337124499638,
"loss": 6.0713,
"mean_token_accuracy": 0.1421744704246521,
"num_tokens": 3822055.0,
"step": 1845
},
{
"entropy": 5.9824930191040036,
"epoch": 0.1678156748911466,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999329247651777,
"loss": 6.0099,
"mean_token_accuracy": 0.13485550805926322,
"num_tokens": 3831900.0,
"step": 1850
},
{
"entropy": 6.0410867691040036,
"epoch": 0.16826923076923078,
"grad_norm": 1.171875,
"learning_rate": 0.0004999321324287213,
"loss": 5.9797,
"mean_token_accuracy": 0.13857458531856537,
"num_tokens": 3842702.0,
"step": 1855
},
{
"entropy": 6.063360166549683,
"epoch": 0.16872278664731494,
"grad_norm": 1.15625,
"learning_rate": 0.0004999313354406114,
"loss": 5.9382,
"mean_token_accuracy": 0.13797046095132828,
"num_tokens": 3852690.0,
"step": 1860
},
{
"entropy": 5.861346340179443,
"epoch": 0.16917634252539912,
"grad_norm": 1.0703125,
"learning_rate": 0.000499930533800864,
"loss": 5.9436,
"mean_token_accuracy": 0.14458322674036025,
"num_tokens": 3863040.0,
"step": 1865
},
{
"entropy": 5.986399984359741,
"epoch": 0.1696298984034833,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999297275094958,
"loss": 6.0249,
"mean_token_accuracy": 0.13291169926524163,
"num_tokens": 3873613.0,
"step": 1870
},
{
"entropy": 5.904158306121826,
"epoch": 0.1700834542815675,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999289165665235,
"loss": 5.8408,
"mean_token_accuracy": 0.1431124784052372,
"num_tokens": 3883936.0,
"step": 1875
},
{
"entropy": 5.933678531646729,
"epoch": 0.17053701015965167,
"grad_norm": 1.390625,
"learning_rate": 0.0004999281009719639,
"loss": 5.8053,
"mean_token_accuracy": 0.15056236162781717,
"num_tokens": 3893788.0,
"step": 1880
},
{
"entropy": 5.911346578598023,
"epoch": 0.17099056603773585,
"grad_norm": 1.1875,
"learning_rate": 0.0004999272807258339,
"loss": 5.8711,
"mean_token_accuracy": 0.1449036605656147,
"num_tokens": 3903434.0,
"step": 1885
},
{
"entropy": 5.9308332920074465,
"epoch": 0.17144412191582004,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999264558281504,
"loss": 5.8934,
"mean_token_accuracy": 0.14332058131694794,
"num_tokens": 3913476.0,
"step": 1890
},
{
"entropy": 5.821413946151734,
"epoch": 0.17189767779390422,
"grad_norm": 1.140625,
"learning_rate": 0.0004999256262789305,
"loss": 5.8453,
"mean_token_accuracy": 0.14364899098873138,
"num_tokens": 3923301.0,
"step": 1895
},
{
"entropy": 5.960890817642212,
"epoch": 0.1723512336719884,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999247920781913,
"loss": 5.9933,
"mean_token_accuracy": 0.13700442612171174,
"num_tokens": 3933335.0,
"step": 1900
},
{
"entropy": 6.01145076751709,
"epoch": 0.17280478955007256,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999239532259502,
"loss": 6.0224,
"mean_token_accuracy": 0.13089234083890916,
"num_tokens": 3944171.0,
"step": 1905
},
{
"entropy": 6.04728136062622,
"epoch": 0.17325834542815674,
"grad_norm": 1.140625,
"learning_rate": 0.0004999231097222244,
"loss": 6.0882,
"mean_token_accuracy": 0.1425239235162735,
"num_tokens": 3954160.0,
"step": 1910
},
{
"entropy": 6.0153731346130375,
"epoch": 0.17371190130624092,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999222615670312,
"loss": 5.8789,
"mean_token_accuracy": 0.1452776275575161,
"num_tokens": 3964110.0,
"step": 1915
},
{
"entropy": 5.966774988174438,
"epoch": 0.1741654571843251,
"grad_norm": 1.0625,
"learning_rate": 0.0004999214087603885,
"loss": 5.9743,
"mean_token_accuracy": 0.13833973333239555,
"num_tokens": 3974293.0,
"step": 1920
},
{
"entropy": 5.89274525642395,
"epoch": 0.1746190130624093,
"grad_norm": 1.09375,
"learning_rate": 0.0004999205513023135,
"loss": 5.9224,
"mean_token_accuracy": 0.1437511757016182,
"num_tokens": 3984346.0,
"step": 1925
},
{
"entropy": 5.918350028991699,
"epoch": 0.17507256894049347,
"grad_norm": 1.09375,
"learning_rate": 0.0004999196891928245,
"loss": 5.8621,
"mean_token_accuracy": 0.1486036777496338,
"num_tokens": 3992897.0,
"step": 1930
},
{
"entropy": 5.9130213260650635,
"epoch": 0.17552612481857766,
"grad_norm": 1.171875,
"learning_rate": 0.0004999188224319388,
"loss": 5.9059,
"mean_token_accuracy": 0.1498722940683365,
"num_tokens": 4002575.0,
"step": 1935
},
{
"entropy": 5.936802387237549,
"epoch": 0.17597968069666184,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999179510196746,
"loss": 5.9803,
"mean_token_accuracy": 0.13661175593733788,
"num_tokens": 4013752.0,
"step": 1940
},
{
"entropy": 5.9770777225494385,
"epoch": 0.17643323657474602,
"grad_norm": 1.1953125,
"learning_rate": 0.0004999170749560498,
"loss": 5.8615,
"mean_token_accuracy": 0.14331807419657708,
"num_tokens": 4023718.0,
"step": 1945
},
{
"entropy": 5.900900363922119,
"epoch": 0.17688679245283018,
"grad_norm": 1.4375,
"learning_rate": 0.0004999161942410827,
"loss": 5.9739,
"mean_token_accuracy": 0.1353725589811802,
"num_tokens": 4034210.0,
"step": 1950
},
{
"entropy": 6.036285829544068,
"epoch": 0.17734034833091436,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999153088747913,
"loss": 5.951,
"mean_token_accuracy": 0.13893376514315606,
"num_tokens": 4044868.0,
"step": 1955
},
{
"entropy": 5.865960550308228,
"epoch": 0.17779390420899854,
"grad_norm": 1.125,
"learning_rate": 0.000499914418857194,
"loss": 5.9483,
"mean_token_accuracy": 0.1431120105087757,
"num_tokens": 4055049.0,
"step": 1960
},
{
"entropy": 6.078762149810791,
"epoch": 0.17824746008708273,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999135241883092,
"loss": 6.01,
"mean_token_accuracy": 0.1387563019990921,
"num_tokens": 4065889.0,
"step": 1965
},
{
"entropy": 6.000588369369507,
"epoch": 0.1787010159651669,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999126248681555,
"loss": 6.0182,
"mean_token_accuracy": 0.1365288347005844,
"num_tokens": 4077288.0,
"step": 1970
},
{
"entropy": 5.914630365371704,
"epoch": 0.1791545718432511,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999117208967513,
"loss": 5.9669,
"mean_token_accuracy": 0.13962030857801438,
"num_tokens": 4087402.0,
"step": 1975
},
{
"entropy": 6.029739284515381,
"epoch": 0.17960812772133528,
"grad_norm": 1.03125,
"learning_rate": 0.0004999108122741154,
"loss": 6.0221,
"mean_token_accuracy": 0.1369963027536869,
"num_tokens": 4098150.0,
"step": 1980
},
{
"entropy": 5.888871479034424,
"epoch": 0.18006168359941946,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999098990002666,
"loss": 5.8678,
"mean_token_accuracy": 0.14218315333127976,
"num_tokens": 4108134.0,
"step": 1985
},
{
"entropy": 6.019744491577148,
"epoch": 0.18051523947750364,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999089810752237,
"loss": 6.0193,
"mean_token_accuracy": 0.1359066478908062,
"num_tokens": 4118251.0,
"step": 1990
},
{
"entropy": 5.931015396118164,
"epoch": 0.1809687953555878,
"grad_norm": 1.09375,
"learning_rate": 0.000499908058499006,
"loss": 5.8581,
"mean_token_accuracy": 0.145183701813221,
"num_tokens": 4129228.0,
"step": 1995
},
{
"entropy": 5.9268440246582035,
"epoch": 0.18142235123367198,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999071312716321,
"loss": 5.9273,
"mean_token_accuracy": 0.14528655931353568,
"num_tokens": 4139391.0,
"step": 2000
},
{
"entropy": 5.924670076370239,
"epoch": 0.18187590711175616,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999061993931215,
"loss": 5.8668,
"mean_token_accuracy": 0.14713739082217217,
"num_tokens": 4148891.0,
"step": 2005
},
{
"entropy": 6.1402112483978275,
"epoch": 0.18232946298984035,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999052628634934,
"loss": 6.2053,
"mean_token_accuracy": 0.13204461038112641,
"num_tokens": 4159551.0,
"step": 2010
},
{
"entropy": 6.043571424484253,
"epoch": 0.18278301886792453,
"grad_norm": 1.078125,
"learning_rate": 0.0004999043216827671,
"loss": 6.0086,
"mean_token_accuracy": 0.13990638628602028,
"num_tokens": 4170055.0,
"step": 2015
},
{
"entropy": 5.888297462463379,
"epoch": 0.1832365747460087,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999033758509621,
"loss": 5.8599,
"mean_token_accuracy": 0.13923854306340216,
"num_tokens": 4180767.0,
"step": 2020
},
{
"entropy": 5.950031042098999,
"epoch": 0.1836901306240929,
"grad_norm": 1.25,
"learning_rate": 0.000499902425368098,
"loss": 5.9798,
"mean_token_accuracy": 0.14310303330421448,
"num_tokens": 4190654.0,
"step": 2025
},
{
"entropy": 5.984168815612793,
"epoch": 0.18414368650217708,
"grad_norm": 1.21875,
"learning_rate": 0.0004999014702341944,
"loss": 5.9011,
"mean_token_accuracy": 0.14403254836797713,
"num_tokens": 4200753.0,
"step": 2030
},
{
"entropy": 5.928147506713867,
"epoch": 0.18459724238026126,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999005104492712,
"loss": 6.0011,
"mean_token_accuracy": 0.1354036383330822,
"num_tokens": 4210994.0,
"step": 2035
},
{
"entropy": 6.021360778808594,
"epoch": 0.18505079825834542,
"grad_norm": 1.171875,
"learning_rate": 0.0004998995460133481,
"loss": 5.9295,
"mean_token_accuracy": 0.13851736932992936,
"num_tokens": 4220883.0,
"step": 2040
},
{
"entropy": 5.972361421585083,
"epoch": 0.1855043541364296,
"grad_norm": 1.109375,
"learning_rate": 0.0004998985769264451,
"loss": 5.9692,
"mean_token_accuracy": 0.1328580841422081,
"num_tokens": 4231269.0,
"step": 2045
},
{
"entropy": 5.92348313331604,
"epoch": 0.18595791001451378,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998976031885821,
"loss": 5.9583,
"mean_token_accuracy": 0.1385072022676468,
"num_tokens": 4241559.0,
"step": 2050
},
{
"entropy": 5.93124942779541,
"epoch": 0.18641146589259797,
"grad_norm": 1.3359375,
"learning_rate": 0.0004998966247997794,
"loss": 5.8573,
"mean_token_accuracy": 0.14284690618515014,
"num_tokens": 4250870.0,
"step": 2055
},
{
"entropy": 5.912771606445313,
"epoch": 0.18686502177068215,
"grad_norm": 1.171875,
"learning_rate": 0.0004998956417600572,
"loss": 5.8197,
"mean_token_accuracy": 0.1453770861029625,
"num_tokens": 4260848.0,
"step": 2060
},
{
"entropy": 5.8631298542022705,
"epoch": 0.18731857764876633,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998946540694358,
"loss": 5.947,
"mean_token_accuracy": 0.14410295933485032,
"num_tokens": 4270973.0,
"step": 2065
},
{
"entropy": 6.008993816375733,
"epoch": 0.18777213352685052,
"grad_norm": 1.203125,
"learning_rate": 0.0004998936617279357,
"loss": 6.0175,
"mean_token_accuracy": 0.1405512697994709,
"num_tokens": 4281211.0,
"step": 2070
},
{
"entropy": 5.907906436920166,
"epoch": 0.1882256894049347,
"grad_norm": 1.140625,
"learning_rate": 0.0004998926647355773,
"loss": 5.8701,
"mean_token_accuracy": 0.14646850228309632,
"num_tokens": 4291425.0,
"step": 2075
},
{
"entropy": 5.9778430461883545,
"epoch": 0.18867924528301888,
"grad_norm": 1.21875,
"learning_rate": 0.0004998916630923813,
"loss": 5.9779,
"mean_token_accuracy": 0.14027420431375504,
"num_tokens": 4301079.0,
"step": 2080
},
{
"entropy": 6.034561729431152,
"epoch": 0.18913280116110304,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998906567983685,
"loss": 5.9646,
"mean_token_accuracy": 0.1393656000494957,
"num_tokens": 4312374.0,
"step": 2085
},
{
"entropy": 5.909461545944214,
"epoch": 0.18958635703918722,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998896458535595,
"loss": 5.879,
"mean_token_accuracy": 0.14832498878240585,
"num_tokens": 4322810.0,
"step": 2090
},
{
"entropy": 5.9581499099731445,
"epoch": 0.1900399129172714,
"grad_norm": 1.2890625,
"learning_rate": 0.0004998886302579752,
"loss": 5.8683,
"mean_token_accuracy": 0.14442463368177413,
"num_tokens": 4332082.0,
"step": 2095
},
{
"entropy": 5.972524499893188,
"epoch": 0.19049346879535559,
"grad_norm": 1.234375,
"learning_rate": 0.000499887610011637,
"loss": 5.9623,
"mean_token_accuracy": 0.1426968216896057,
"num_tokens": 4342632.0,
"step": 2100
},
{
"entropy": 5.894812726974488,
"epoch": 0.19094702467343977,
"grad_norm": 1.296875,
"learning_rate": 0.0004998865851145656,
"loss": 5.8987,
"mean_token_accuracy": 0.1407205879688263,
"num_tokens": 4353544.0,
"step": 2105
},
{
"entropy": 5.882326269149781,
"epoch": 0.19140058055152395,
"grad_norm": 1.2734375,
"learning_rate": 0.0004998855555667822,
"loss": 6.0315,
"mean_token_accuracy": 0.13898359686136247,
"num_tokens": 4363636.0,
"step": 2110
},
{
"entropy": 6.04261794090271,
"epoch": 0.19185413642960814,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998845213683084,
"loss": 5.8898,
"mean_token_accuracy": 0.13970340117812158,
"num_tokens": 4373672.0,
"step": 2115
},
{
"entropy": 5.864584922790527,
"epoch": 0.19230769230769232,
"grad_norm": 1.3125,
"learning_rate": 0.0004998834825191652,
"loss": 5.8591,
"mean_token_accuracy": 0.1457592159509659,
"num_tokens": 4383199.0,
"step": 2120
},
{
"entropy": 5.9757240295410154,
"epoch": 0.1927612481857765,
"grad_norm": 1.109375,
"learning_rate": 0.0004998824390193743,
"loss": 5.9402,
"mean_token_accuracy": 0.13909565806388854,
"num_tokens": 4394446.0,
"step": 2125
},
{
"entropy": 6.01021409034729,
"epoch": 0.19321480406386066,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998813908689573,
"loss": 5.9763,
"mean_token_accuracy": 0.1380794882774353,
"num_tokens": 4405882.0,
"step": 2130
},
{
"entropy": 5.878481531143189,
"epoch": 0.19366835994194484,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998803380679359,
"loss": 5.8407,
"mean_token_accuracy": 0.15159011334180833,
"num_tokens": 4415280.0,
"step": 2135
},
{
"entropy": 5.953402328491211,
"epoch": 0.19412191582002902,
"grad_norm": 1.2734375,
"learning_rate": 0.0004998792806163317,
"loss": 6.0475,
"mean_token_accuracy": 0.13735857382416725,
"num_tokens": 4425755.0,
"step": 2140
},
{
"entropy": 5.908466815948486,
"epoch": 0.1945754716981132,
"grad_norm": 1.2109375,
"learning_rate": 0.0004998782185141667,
"loss": 5.9127,
"mean_token_accuracy": 0.138707172870636,
"num_tokens": 4436259.0,
"step": 2145
},
{
"entropy": 6.002693319320679,
"epoch": 0.1950290275761974,
"grad_norm": 1.2734375,
"learning_rate": 0.0004998771517614629,
"loss": 5.9482,
"mean_token_accuracy": 0.13556500673294067,
"num_tokens": 4446423.0,
"step": 2150
},
{
"entropy": 5.941006422042847,
"epoch": 0.19548258345428157,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998760803582422,
"loss": 5.8841,
"mean_token_accuracy": 0.15068939626216887,
"num_tokens": 4456678.0,
"step": 2155
},
{
"entropy": 5.921889400482177,
"epoch": 0.19593613933236576,
"grad_norm": 1.3046875,
"learning_rate": 0.0004998750043045269,
"loss": 5.8956,
"mean_token_accuracy": 0.14178509563207625,
"num_tokens": 4467031.0,
"step": 2160
},
{
"entropy": 5.974037790298462,
"epoch": 0.19638969521044994,
"grad_norm": 1.203125,
"learning_rate": 0.0004998739236003392,
"loss": 5.9568,
"mean_token_accuracy": 0.13483025133609772,
"num_tokens": 4477771.0,
"step": 2165
},
{
"entropy": 5.998474931716919,
"epoch": 0.1968432510885341,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998728382457014,
"loss": 5.9763,
"mean_token_accuracy": 0.13997794538736344,
"num_tokens": 4488426.0,
"step": 2170
},
{
"entropy": 6.019248390197754,
"epoch": 0.19729680696661828,
"grad_norm": 1.1953125,
"learning_rate": 0.0004998717482406362,
"loss": 5.9578,
"mean_token_accuracy": 0.13564275726675987,
"num_tokens": 4499735.0,
"step": 2175
},
{
"entropy": 5.934429025650024,
"epoch": 0.19775036284470246,
"grad_norm": 1.421875,
"learning_rate": 0.0004998706535851657,
"loss": 5.88,
"mean_token_accuracy": 0.1441620409488678,
"num_tokens": 4509500.0,
"step": 2180
},
{
"entropy": 5.90686182975769,
"epoch": 0.19820391872278664,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998695542793129,
"loss": 5.906,
"mean_token_accuracy": 0.14358371123671532,
"num_tokens": 4520431.0,
"step": 2185
},
{
"entropy": 5.962169981002807,
"epoch": 0.19865747460087083,
"grad_norm": 1.21875,
"learning_rate": 0.0004998684503231003,
"loss": 5.9944,
"mean_token_accuracy": 0.14260854870080947,
"num_tokens": 4532294.0,
"step": 2190
},
{
"entropy": 5.9562725067138675,
"epoch": 0.199111030478955,
"grad_norm": 1.3203125,
"learning_rate": 0.000499867341716551,
"loss": 5.9081,
"mean_token_accuracy": 0.14024144038558006,
"num_tokens": 4542488.0,
"step": 2195
},
{
"entropy": 5.872625350952148,
"epoch": 0.1995645863570392,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998662284596878,
"loss": 5.8462,
"mean_token_accuracy": 0.136696857213974,
"num_tokens": 4552933.0,
"step": 2200
},
{
"entropy": 5.941536951065063,
"epoch": 0.20001814223512338,
"grad_norm": 1.21875,
"learning_rate": 0.0004998651105525335,
"loss": 5.8378,
"mean_token_accuracy": 0.14871531575918198,
"num_tokens": 4562373.0,
"step": 2205
},
{
"entropy": 5.881709098815918,
"epoch": 0.20047169811320756,
"grad_norm": 1.125,
"learning_rate": 0.0004998639879951115,
"loss": 5.9588,
"mean_token_accuracy": 0.14208371788263321,
"num_tokens": 4572715.0,
"step": 2210
},
{
"entropy": 5.96837158203125,
"epoch": 0.2009252539912917,
"grad_norm": 1.1953125,
"learning_rate": 0.000499862860787445,
"loss": 5.8464,
"mean_token_accuracy": 0.13923058286309242,
"num_tokens": 4582782.0,
"step": 2215
},
{
"entropy": 5.983753299713134,
"epoch": 0.2013788098693759,
"grad_norm": 1.171875,
"learning_rate": 0.0004998617289295571,
"loss": 5.9569,
"mean_token_accuracy": 0.13587422147393227,
"num_tokens": 4594108.0,
"step": 2220
},
{
"entropy": 5.929776239395141,
"epoch": 0.20183236574746008,
"grad_norm": 1.421875,
"learning_rate": 0.0004998605924214714,
"loss": 5.9107,
"mean_token_accuracy": 0.14216437190771103,
"num_tokens": 4604732.0,
"step": 2225
},
{
"entropy": 5.93411808013916,
"epoch": 0.20228592162554426,
"grad_norm": 1.2734375,
"learning_rate": 0.0004998594512632114,
"loss": 6.0441,
"mean_token_accuracy": 0.13005613088607787,
"num_tokens": 4614998.0,
"step": 2230
},
{
"entropy": 5.9276261806488035,
"epoch": 0.20273947750362845,
"grad_norm": 2.421875,
"learning_rate": 0.0004998583054548006,
"loss": 5.8687,
"mean_token_accuracy": 0.14243716672062873,
"num_tokens": 4624665.0,
"step": 2235
},
{
"entropy": 5.920424890518189,
"epoch": 0.20319303338171263,
"grad_norm": 1.515625,
"learning_rate": 0.0004998571549962628,
"loss": 5.8625,
"mean_token_accuracy": 0.14737029150128364,
"num_tokens": 4633731.0,
"step": 2240
},
{
"entropy": 5.960051679611206,
"epoch": 0.2036465892597968,
"grad_norm": 1.1875,
"learning_rate": 0.0004998559998876216,
"loss": 5.8949,
"mean_token_accuracy": 0.13982552140951157,
"num_tokens": 4643421.0,
"step": 2245
},
{
"entropy": 6.00085940361023,
"epoch": 0.204100145137881,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998548401289012,
"loss": 5.9692,
"mean_token_accuracy": 0.13818764984607695,
"num_tokens": 4653795.0,
"step": 2250
},
{
"entropy": 5.941312074661255,
"epoch": 0.20455370101596518,
"grad_norm": 1.1484375,
"learning_rate": 0.0004998536757201253,
"loss": 5.9414,
"mean_token_accuracy": 0.13912858814001083,
"num_tokens": 4664840.0,
"step": 2255
},
{
"entropy": 5.994112920761109,
"epoch": 0.20500725689404933,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998525066613181,
"loss": 5.952,
"mean_token_accuracy": 0.14518292397260665,
"num_tokens": 4675508.0,
"step": 2260
},
{
"entropy": 5.889045238494873,
"epoch": 0.20546081277213352,
"grad_norm": 1.4609375,
"learning_rate": 0.0004998513329525038,
"loss": 5.8761,
"mean_token_accuracy": 0.14128366410732268,
"num_tokens": 4685995.0,
"step": 2265
},
{
"entropy": 6.009013843536377,
"epoch": 0.2059143686502177,
"grad_norm": 1.140625,
"learning_rate": 0.0004998501545937066,
"loss": 5.908,
"mean_token_accuracy": 0.13695503398776054,
"num_tokens": 4696659.0,
"step": 2270
},
{
"entropy": 5.840634775161743,
"epoch": 0.20636792452830188,
"grad_norm": 1.1953125,
"learning_rate": 0.000499848971584951,
"loss": 5.8695,
"mean_token_accuracy": 0.14796618670225142,
"num_tokens": 4706438.0,
"step": 2275
},
{
"entropy": 5.9189924716949465,
"epoch": 0.20682148040638607,
"grad_norm": 1.3671875,
"learning_rate": 0.0004998477839262611,
"loss": 5.8764,
"mean_token_accuracy": 0.14219514206051825,
"num_tokens": 4716402.0,
"step": 2280
},
{
"entropy": 5.96020998954773,
"epoch": 0.20727503628447025,
"grad_norm": 1.125,
"learning_rate": 0.000499846591617662,
"loss": 5.9719,
"mean_token_accuracy": 0.14258323535323142,
"num_tokens": 4728488.0,
"step": 2285
},
{
"entropy": 5.915734434127808,
"epoch": 0.20772859216255443,
"grad_norm": 1.4140625,
"learning_rate": 0.0004998453946591779,
"loss": 5.9794,
"mean_token_accuracy": 0.14094745591282845,
"num_tokens": 4738693.0,
"step": 2290
},
{
"entropy": 5.8744336605072025,
"epoch": 0.20818214804063861,
"grad_norm": 1.359375,
"learning_rate": 0.0004998441930508339,
"loss": 5.8724,
"mean_token_accuracy": 0.1447533279657364,
"num_tokens": 4748895.0,
"step": 2295
},
{
"entropy": 5.995182752609253,
"epoch": 0.2086357039187228,
"grad_norm": 1.3125,
"learning_rate": 0.0004998429867926547,
"loss": 5.9878,
"mean_token_accuracy": 0.1373719684779644,
"num_tokens": 4760135.0,
"step": 2300
},
{
"entropy": 6.019215488433838,
"epoch": 0.20908925979680695,
"grad_norm": 1.5078125,
"learning_rate": 0.0004998417758846651,
"loss": 6.0195,
"mean_token_accuracy": 0.14542822539806366,
"num_tokens": 4769935.0,
"step": 2305
},
{
"entropy": 6.055789279937744,
"epoch": 0.20954281567489114,
"grad_norm": 1.2421875,
"learning_rate": 0.0004998405603268903,
"loss": 5.9553,
"mean_token_accuracy": 0.14619530588388444,
"num_tokens": 4779520.0,
"step": 2310
},
{
"entropy": 5.880806922912598,
"epoch": 0.20999637155297532,
"grad_norm": 1.203125,
"learning_rate": 0.0004998393401193554,
"loss": 5.8646,
"mean_token_accuracy": 0.1471490442752838,
"num_tokens": 4791256.0,
"step": 2315
},
{
"entropy": 5.815768003463745,
"epoch": 0.2104499274310595,
"grad_norm": 1.125,
"learning_rate": 0.0004998381152620857,
"loss": 5.8248,
"mean_token_accuracy": 0.14878984093666076,
"num_tokens": 4801334.0,
"step": 2320
},
{
"entropy": 5.8872205257415775,
"epoch": 0.21090348330914369,
"grad_norm": 1.2890625,
"learning_rate": 0.0004998368857551067,
"loss": 5.8025,
"mean_token_accuracy": 0.14462457224726677,
"num_tokens": 4811361.0,
"step": 2325
},
{
"entropy": 5.880616140365601,
"epoch": 0.21135703918722787,
"grad_norm": 1.15625,
"learning_rate": 0.0004998356515984433,
"loss": 5.8777,
"mean_token_accuracy": 0.14419098272919656,
"num_tokens": 4821190.0,
"step": 2330
},
{
"entropy": 5.971280717849732,
"epoch": 0.21181059506531205,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998344127921213,
"loss": 5.903,
"mean_token_accuracy": 0.14422993659973143,
"num_tokens": 4830694.0,
"step": 2335
},
{
"entropy": 5.846283674240112,
"epoch": 0.21226415094339623,
"grad_norm": 1.140625,
"learning_rate": 0.0004998331693361666,
"loss": 5.874,
"mean_token_accuracy": 0.14026885032653807,
"num_tokens": 4841344.0,
"step": 2340
},
{
"entropy": 5.894744539260865,
"epoch": 0.21271770682148042,
"grad_norm": 1.203125,
"learning_rate": 0.0004998319212306046,
"loss": 5.8561,
"mean_token_accuracy": 0.14421792402863504,
"num_tokens": 4851161.0,
"step": 2345
},
{
"entropy": 5.859916687011719,
"epoch": 0.21317126269956457,
"grad_norm": 1.1015625,
"learning_rate": 0.000499830668475461,
"loss": 5.8403,
"mean_token_accuracy": 0.1468031883239746,
"num_tokens": 4860740.0,
"step": 2350
},
{
"entropy": 5.916117763519287,
"epoch": 0.21362481857764876,
"grad_norm": 1.2109375,
"learning_rate": 0.000499829411070762,
"loss": 5.8851,
"mean_token_accuracy": 0.14431235790252686,
"num_tokens": 4871733.0,
"step": 2355
},
{
"entropy": 5.938611316680908,
"epoch": 0.21407837445573294,
"grad_norm": 1.3359375,
"learning_rate": 0.0004998281490165335,
"loss": 5.9215,
"mean_token_accuracy": 0.13747453391551973,
"num_tokens": 4883617.0,
"step": 2360
},
{
"entropy": 5.9607947826385494,
"epoch": 0.21453193033381712,
"grad_norm": 1.2734375,
"learning_rate": 0.0004998268823128016,
"loss": 5.8793,
"mean_token_accuracy": 0.1396989904344082,
"num_tokens": 4894519.0,
"step": 2365
},
{
"entropy": 5.893149185180664,
"epoch": 0.2149854862119013,
"grad_norm": 1.203125,
"learning_rate": 0.0004998256109595923,
"loss": 5.8394,
"mean_token_accuracy": 0.14341802075505256,
"num_tokens": 4904965.0,
"step": 2370
},
{
"entropy": 5.780011129379273,
"epoch": 0.2154390420899855,
"grad_norm": 1.328125,
"learning_rate": 0.0004998243349569323,
"loss": 5.7528,
"mean_token_accuracy": 0.14949761778116227,
"num_tokens": 4914618.0,
"step": 2375
},
{
"entropy": 5.839861011505127,
"epoch": 0.21589259796806967,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998230543048477,
"loss": 5.808,
"mean_token_accuracy": 0.14917464703321456,
"num_tokens": 4924341.0,
"step": 2380
},
{
"entropy": 5.833130645751953,
"epoch": 0.21634615384615385,
"grad_norm": 1.2109375,
"learning_rate": 0.000499821769003365,
"loss": 5.7863,
"mean_token_accuracy": 0.14926746040582656,
"num_tokens": 4934231.0,
"step": 2385
},
{
"entropy": 5.826928949356079,
"epoch": 0.21679970972423804,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998204790525109,
"loss": 5.8055,
"mean_token_accuracy": 0.1478435292840004,
"num_tokens": 4944303.0,
"step": 2390
},
{
"entropy": 5.877607536315918,
"epoch": 0.2172532656023222,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998191844523119,
"loss": 5.8831,
"mean_token_accuracy": 0.1455991432070732,
"num_tokens": 4954571.0,
"step": 2395
},
{
"entropy": 5.891971683502197,
"epoch": 0.21770682148040638,
"grad_norm": 1.265625,
"learning_rate": 0.000499817885202795,
"loss": 5.8568,
"mean_token_accuracy": 0.14144586399197578,
"num_tokens": 4964418.0,
"step": 2400
},
{
"entropy": 5.854138994216919,
"epoch": 0.21816037735849056,
"grad_norm": 1.625,
"learning_rate": 0.0004998165813039869,
"loss": 5.8641,
"mean_token_accuracy": 0.13880210369825363,
"num_tokens": 4975036.0,
"step": 2405
},
{
"entropy": 5.882414531707764,
"epoch": 0.21861393323657474,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998152727559145,
"loss": 5.9467,
"mean_token_accuracy": 0.1381090022623539,
"num_tokens": 4985272.0,
"step": 2410
},
{
"entropy": 5.958352327346802,
"epoch": 0.21906748911465893,
"grad_norm": 1.4140625,
"learning_rate": 0.000499813959558605,
"loss": 5.8973,
"mean_token_accuracy": 0.1454383596777916,
"num_tokens": 4995187.0,
"step": 2415
},
{
"entropy": 5.983824586868286,
"epoch": 0.2195210449927431,
"grad_norm": 1.203125,
"learning_rate": 0.0004998126417120856,
"loss": 5.9346,
"mean_token_accuracy": 0.14655492603778839,
"num_tokens": 5005013.0,
"step": 2420
},
{
"entropy": 5.981702709197998,
"epoch": 0.2199746008708273,
"grad_norm": 1.234375,
"learning_rate": 0.0004998113192163835,
"loss": 5.9514,
"mean_token_accuracy": 0.13586568385362624,
"num_tokens": 5016789.0,
"step": 2425
},
{
"entropy": 5.891997861862182,
"epoch": 0.22042815674891147,
"grad_norm": 1.1953125,
"learning_rate": 0.000499809992071526,
"loss": 5.9075,
"mean_token_accuracy": 0.14112533405423164,
"num_tokens": 5026913.0,
"step": 2430
},
{
"entropy": 5.908569240570069,
"epoch": 0.22088171262699566,
"grad_norm": 1.21875,
"learning_rate": 0.0004998086602775406,
"loss": 6.0077,
"mean_token_accuracy": 0.13833704516291617,
"num_tokens": 5037501.0,
"step": 2435
},
{
"entropy": 5.948585081100464,
"epoch": 0.2213352685050798,
"grad_norm": 1.3203125,
"learning_rate": 0.0004998073238344547,
"loss": 5.8537,
"mean_token_accuracy": 0.1406433142721653,
"num_tokens": 5048478.0,
"step": 2440
},
{
"entropy": 5.867254590988159,
"epoch": 0.221788824383164,
"grad_norm": 1.15625,
"learning_rate": 0.0004998059827422962,
"loss": 5.9062,
"mean_token_accuracy": 0.13890414834022521,
"num_tokens": 5059118.0,
"step": 2445
},
{
"entropy": 5.970612239837647,
"epoch": 0.22224238026124818,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998046370010926,
"loss": 5.8774,
"mean_token_accuracy": 0.14610251635313035,
"num_tokens": 5070180.0,
"step": 2450
},
{
"entropy": 5.825871181488037,
"epoch": 0.22269593613933236,
"grad_norm": 1.34375,
"learning_rate": 0.0004998032866108718,
"loss": 5.8617,
"mean_token_accuracy": 0.1452118158340454,
"num_tokens": 5079912.0,
"step": 2455
},
{
"entropy": 5.963987159729004,
"epoch": 0.22314949201741655,
"grad_norm": 1.1796875,
"learning_rate": 0.0004998019315716618,
"loss": 5.9397,
"mean_token_accuracy": 0.143439419567585,
"num_tokens": 5090549.0,
"step": 2460
},
{
"entropy": 5.878372097015381,
"epoch": 0.22360304789550073,
"grad_norm": 1.1875,
"learning_rate": 0.0004998005718834905,
"loss": 5.7929,
"mean_token_accuracy": 0.14878275394439697,
"num_tokens": 5100880.0,
"step": 2465
},
{
"entropy": 5.89096622467041,
"epoch": 0.2240566037735849,
"grad_norm": 1.15625,
"learning_rate": 0.0004997992075463862,
"loss": 5.9171,
"mean_token_accuracy": 0.14251485839486122,
"num_tokens": 5111796.0,
"step": 2470
},
{
"entropy": 5.983297872543335,
"epoch": 0.2245101596516691,
"grad_norm": 1.15625,
"learning_rate": 0.0004997978385603769,
"loss": 5.9062,
"mean_token_accuracy": 0.1443158358335495,
"num_tokens": 5123468.0,
"step": 2475
},
{
"entropy": 5.861389255523681,
"epoch": 0.22496371552975328,
"grad_norm": 1.2421875,
"learning_rate": 0.0004997964649254911,
"loss": 5.8225,
"mean_token_accuracy": 0.1483868971467018,
"num_tokens": 5132925.0,
"step": 2480
},
{
"entropy": 5.9410045623779295,
"epoch": 0.22541727140783743,
"grad_norm": 1.2265625,
"learning_rate": 0.0004997950866417572,
"loss": 5.9016,
"mean_token_accuracy": 0.14117321670055388,
"num_tokens": 5143445.0,
"step": 2485
},
{
"entropy": 5.899289989471436,
"epoch": 0.22587082728592162,
"grad_norm": 1.3203125,
"learning_rate": 0.0004997937037092034,
"loss": 5.8287,
"mean_token_accuracy": 0.14134425967931746,
"num_tokens": 5153176.0,
"step": 2490
},
{
"entropy": 5.919216680526733,
"epoch": 0.2263243831640058,
"grad_norm": 1.203125,
"learning_rate": 0.0004997923161278587,
"loss": 5.866,
"mean_token_accuracy": 0.14413893669843675,
"num_tokens": 5162765.0,
"step": 2495
},
{
"entropy": 5.879088354110718,
"epoch": 0.22677793904208998,
"grad_norm": 1.3203125,
"learning_rate": 0.0004997909238977514,
"loss": 5.9173,
"mean_token_accuracy": 0.14777278080582618,
"num_tokens": 5173390.0,
"step": 2500
},
{
"entropy": 5.875536203384399,
"epoch": 0.22723149492017417,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997895270189108,
"loss": 5.836,
"mean_token_accuracy": 0.13986415043473244,
"num_tokens": 5184212.0,
"step": 2505
},
{
"entropy": 5.811423492431641,
"epoch": 0.22768505079825835,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997881254913653,
"loss": 5.8297,
"mean_token_accuracy": 0.13980135917663575,
"num_tokens": 5194311.0,
"step": 2510
},
{
"entropy": 5.82978572845459,
"epoch": 0.22813860667634253,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997867193151441,
"loss": 5.8043,
"mean_token_accuracy": 0.14709037020802498,
"num_tokens": 5204708.0,
"step": 2515
},
{
"entropy": 5.838364934921264,
"epoch": 0.22859216255442671,
"grad_norm": 1.234375,
"learning_rate": 0.0004997853084902763,
"loss": 5.8108,
"mean_token_accuracy": 0.14625583738088607,
"num_tokens": 5214566.0,
"step": 2520
},
{
"entropy": 5.870654010772705,
"epoch": 0.2290457184325109,
"grad_norm": 1.2734375,
"learning_rate": 0.0004997838930167911,
"loss": 5.8253,
"mean_token_accuracy": 0.15037785470485687,
"num_tokens": 5224762.0,
"step": 2525
},
{
"entropy": 5.907916641235351,
"epoch": 0.22949927431059505,
"grad_norm": 1.1640625,
"learning_rate": 0.0004997824728947176,
"loss": 5.8973,
"mean_token_accuracy": 0.14429675713181495,
"num_tokens": 5235146.0,
"step": 2530
},
{
"entropy": 5.949673509597778,
"epoch": 0.22995283018867924,
"grad_norm": 1.65625,
"learning_rate": 0.0004997810481240853,
"loss": 5.8623,
"mean_token_accuracy": 0.1440894216299057,
"num_tokens": 5243867.0,
"step": 2535
},
{
"entropy": 5.856524896621704,
"epoch": 0.23040638606676342,
"grad_norm": 1.25,
"learning_rate": 0.0004997796187049238,
"loss": 5.7751,
"mean_token_accuracy": 0.1456816427409649,
"num_tokens": 5254279.0,
"step": 2540
},
{
"entropy": 5.948967504501343,
"epoch": 0.2308599419448476,
"grad_norm": 1.296875,
"learning_rate": 0.0004997781846372623,
"loss": 5.8973,
"mean_token_accuracy": 0.14449210837483406,
"num_tokens": 5264769.0,
"step": 2545
},
{
"entropy": 5.927372407913208,
"epoch": 0.23131349782293179,
"grad_norm": 1.1953125,
"learning_rate": 0.0004997767459211307,
"loss": 5.8323,
"mean_token_accuracy": 0.14081402122974396,
"num_tokens": 5274291.0,
"step": 2550
},
{
"entropy": 5.867244148254395,
"epoch": 0.23176705370101597,
"grad_norm": 1.3203125,
"learning_rate": 0.0004997753025565587,
"loss": 5.7691,
"mean_token_accuracy": 0.14731070771813393,
"num_tokens": 5284366.0,
"step": 2555
},
{
"entropy": 5.863168334960937,
"epoch": 0.23222060957910015,
"grad_norm": 1.5,
"learning_rate": 0.0004997738545435763,
"loss": 5.8745,
"mean_token_accuracy": 0.14436920285224913,
"num_tokens": 5294494.0,
"step": 2560
},
{
"entropy": 5.8350911140441895,
"epoch": 0.23267416545718433,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997724018822133,
"loss": 5.9145,
"mean_token_accuracy": 0.144419763982296,
"num_tokens": 5305606.0,
"step": 2565
},
{
"entropy": 5.909123945236206,
"epoch": 0.23312772133526852,
"grad_norm": 1.28125,
"learning_rate": 0.0004997709445724996,
"loss": 5.8743,
"mean_token_accuracy": 0.14309675469994546,
"num_tokens": 5315677.0,
"step": 2570
},
{
"entropy": 5.922089576721191,
"epoch": 0.23358127721335267,
"grad_norm": 1.296875,
"learning_rate": 0.0004997694826144656,
"loss": 5.9518,
"mean_token_accuracy": 0.13734646663069724,
"num_tokens": 5326453.0,
"step": 2575
},
{
"entropy": 5.906578445434571,
"epoch": 0.23403483309143686,
"grad_norm": 1.203125,
"learning_rate": 0.0004997680160081414,
"loss": 5.834,
"mean_token_accuracy": 0.14441939890384675,
"num_tokens": 5337591.0,
"step": 2580
},
{
"entropy": 5.937219524383545,
"epoch": 0.23448838896952104,
"grad_norm": 1.25,
"learning_rate": 0.0004997665447535573,
"loss": 5.8398,
"mean_token_accuracy": 0.14593878537416458,
"num_tokens": 5349400.0,
"step": 2585
},
{
"entropy": 5.929783010482788,
"epoch": 0.23494194484760522,
"grad_norm": 1.3125,
"learning_rate": 0.0004997650688507438,
"loss": 5.8162,
"mean_token_accuracy": 0.14761714413762092,
"num_tokens": 5359984.0,
"step": 2590
},
{
"entropy": 5.810983467102051,
"epoch": 0.2353955007256894,
"grad_norm": 1.4609375,
"learning_rate": 0.0004997635882997314,
"loss": 5.8885,
"mean_token_accuracy": 0.1452771820127964,
"num_tokens": 5369254.0,
"step": 2595
},
{
"entropy": 5.926082897186279,
"epoch": 0.2358490566037736,
"grad_norm": 1.328125,
"learning_rate": 0.0004997621031005507,
"loss": 5.8331,
"mean_token_accuracy": 0.14687830135226249,
"num_tokens": 5379648.0,
"step": 2600
},
{
"entropy": 5.818135976791382,
"epoch": 0.23630261248185777,
"grad_norm": 1.4765625,
"learning_rate": 0.0004997606132532325,
"loss": 5.7848,
"mean_token_accuracy": 0.14700028523802758,
"num_tokens": 5390321.0,
"step": 2605
},
{
"entropy": 5.809545612335205,
"epoch": 0.23675616835994195,
"grad_norm": 1.4453125,
"learning_rate": 0.0004997591187578075,
"loss": 5.8338,
"mean_token_accuracy": 0.14946144074201584,
"num_tokens": 5401078.0,
"step": 2610
},
{
"entropy": 5.867106103897095,
"epoch": 0.23720972423802614,
"grad_norm": 1.4140625,
"learning_rate": 0.0004997576196143067,
"loss": 5.7905,
"mean_token_accuracy": 0.1520560935139656,
"num_tokens": 5411977.0,
"step": 2615
},
{
"entropy": 5.854715633392334,
"epoch": 0.2376632801161103,
"grad_norm": 1.3046875,
"learning_rate": 0.000499756115822761,
"loss": 5.8082,
"mean_token_accuracy": 0.14396803379058837,
"num_tokens": 5422967.0,
"step": 2620
},
{
"entropy": 5.910807275772095,
"epoch": 0.23811683599419448,
"grad_norm": 1.359375,
"learning_rate": 0.0004997546073832015,
"loss": 5.833,
"mean_token_accuracy": 0.1544169843196869,
"num_tokens": 5432260.0,
"step": 2625
},
{
"entropy": 5.849782180786133,
"epoch": 0.23857039187227866,
"grad_norm": 1.34375,
"learning_rate": 0.0004997530942956595,
"loss": 5.8033,
"mean_token_accuracy": 0.14687435179948807,
"num_tokens": 5441706.0,
"step": 2630
},
{
"entropy": 5.882516288757325,
"epoch": 0.23902394775036284,
"grad_norm": 1.4296875,
"learning_rate": 0.0004997515765601662,
"loss": 5.8086,
"mean_token_accuracy": 0.15525150001049043,
"num_tokens": 5452397.0,
"step": 2635
},
{
"entropy": 5.856340026855468,
"epoch": 0.23947750362844702,
"grad_norm": 1.21875,
"learning_rate": 0.0004997500541767531,
"loss": 5.779,
"mean_token_accuracy": 0.14695454835891725,
"num_tokens": 5463762.0,
"step": 2640
},
{
"entropy": 5.83820071220398,
"epoch": 0.2399310595065312,
"grad_norm": 1.4921875,
"learning_rate": 0.0004997485271454516,
"loss": 5.8409,
"mean_token_accuracy": 0.1390252538025379,
"num_tokens": 5474622.0,
"step": 2645
},
{
"entropy": 5.850206089019776,
"epoch": 0.2403846153846154,
"grad_norm": 1.328125,
"learning_rate": 0.0004997469954662933,
"loss": 5.8314,
"mean_token_accuracy": 0.14660486727952957,
"num_tokens": 5484697.0,
"step": 2650
},
{
"entropy": 5.859882640838623,
"epoch": 0.24083817126269957,
"grad_norm": 1.2421875,
"learning_rate": 0.0004997454591393097,
"loss": 5.8257,
"mean_token_accuracy": 0.1425713561475277,
"num_tokens": 5496433.0,
"step": 2655
},
{
"entropy": 5.948536205291748,
"epoch": 0.24129172714078376,
"grad_norm": 1.65625,
"learning_rate": 0.0004997439181645329,
"loss": 5.946,
"mean_token_accuracy": 0.14295659139752387,
"num_tokens": 5507917.0,
"step": 2660
},
{
"entropy": 5.989947938919068,
"epoch": 0.2417452830188679,
"grad_norm": 1.578125,
"learning_rate": 0.0004997423725419945,
"loss": 5.9205,
"mean_token_accuracy": 0.13935526087880135,
"num_tokens": 5519159.0,
"step": 2665
},
{
"entropy": 5.88839020729065,
"epoch": 0.2421988388969521,
"grad_norm": 1.3984375,
"learning_rate": 0.0004997408222717267,
"loss": 5.8802,
"mean_token_accuracy": 0.1494501657783985,
"num_tokens": 5529180.0,
"step": 2670
},
{
"entropy": 5.834697914123535,
"epoch": 0.24265239477503628,
"grad_norm": 1.375,
"learning_rate": 0.0004997392673537612,
"loss": 5.7992,
"mean_token_accuracy": 0.1539318174123764,
"num_tokens": 5539363.0,
"step": 2675
},
{
"entropy": 5.8882616519927975,
"epoch": 0.24310595065312046,
"grad_norm": 2.375,
"learning_rate": 0.0004997377077881306,
"loss": 5.9105,
"mean_token_accuracy": 0.14342357888817786,
"num_tokens": 5549841.0,
"step": 2680
},
{
"entropy": 5.897100496292114,
"epoch": 0.24355950653120464,
"grad_norm": 1.4453125,
"learning_rate": 0.0004997361435748668,
"loss": 5.7884,
"mean_token_accuracy": 0.14797737300395966,
"num_tokens": 5560299.0,
"step": 2685
},
{
"entropy": 5.912242698669433,
"epoch": 0.24401306240928883,
"grad_norm": 1.6015625,
"learning_rate": 0.0004997345747140024,
"loss": 5.8527,
"mean_token_accuracy": 0.14805733114480973,
"num_tokens": 5571081.0,
"step": 2690
},
{
"entropy": 5.8140528202056885,
"epoch": 0.244466618287373,
"grad_norm": 1.3828125,
"learning_rate": 0.0004997330012055695,
"loss": 5.8148,
"mean_token_accuracy": 0.15010562539100647,
"num_tokens": 5581389.0,
"step": 2695
},
{
"entropy": 5.859999370574951,
"epoch": 0.2449201741654572,
"grad_norm": 1.6953125,
"learning_rate": 0.0004997314230496011,
"loss": 5.8329,
"mean_token_accuracy": 0.14136491268873214,
"num_tokens": 5592347.0,
"step": 2700
},
{
"entropy": 5.869523763656616,
"epoch": 0.24537373004354138,
"grad_norm": 1.3828125,
"learning_rate": 0.0004997298402461296,
"loss": 5.7951,
"mean_token_accuracy": 0.15069439709186555,
"num_tokens": 5603411.0,
"step": 2705
},
{
"entropy": 5.925837564468384,
"epoch": 0.24582728592162553,
"grad_norm": 1.390625,
"learning_rate": 0.0004997282527951876,
"loss": 5.8089,
"mean_token_accuracy": 0.15103363543748854,
"num_tokens": 5614615.0,
"step": 2710
},
{
"entropy": 5.8883668899536135,
"epoch": 0.24628084179970972,
"grad_norm": 1.2890625,
"learning_rate": 0.0004997266606968081,
"loss": 5.7827,
"mean_token_accuracy": 0.1461820498108864,
"num_tokens": 5624582.0,
"step": 2715
},
{
"entropy": 5.759700155258178,
"epoch": 0.2467343976777939,
"grad_norm": 2.359375,
"learning_rate": 0.0004997250639510241,
"loss": 5.7983,
"mean_token_accuracy": 0.14800044298171997,
"num_tokens": 5634310.0,
"step": 2720
},
{
"entropy": 5.890387344360351,
"epoch": 0.24718795355587808,
"grad_norm": 1.59375,
"learning_rate": 0.0004997234625578684,
"loss": 5.8809,
"mean_token_accuracy": 0.14121756255626677,
"num_tokens": 5644018.0,
"step": 2725
},
{
"entropy": 5.869387245178222,
"epoch": 0.24764150943396226,
"grad_norm": 1.8125,
"learning_rate": 0.0004997218565173742,
"loss": 5.8328,
"mean_token_accuracy": 0.14269990473985672,
"num_tokens": 5653592.0,
"step": 2730
},
{
"entropy": 5.918030786514282,
"epoch": 0.24809506531204645,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997202458295748,
"loss": 5.868,
"mean_token_accuracy": 0.1497960552573204,
"num_tokens": 5662812.0,
"step": 2735
},
{
"entropy": 5.938170909881592,
"epoch": 0.24854862119013063,
"grad_norm": 1.3203125,
"learning_rate": 0.0004997186304945034,
"loss": 5.8345,
"mean_token_accuracy": 0.14251421988010407,
"num_tokens": 5672939.0,
"step": 2740
},
{
"entropy": 5.814105081558227,
"epoch": 0.24900217706821481,
"grad_norm": 1.328125,
"learning_rate": 0.0004997170105121935,
"loss": 5.8216,
"mean_token_accuracy": 0.14442233145236968,
"num_tokens": 5683350.0,
"step": 2745
},
{
"entropy": 5.970436525344849,
"epoch": 0.249455732946299,
"grad_norm": 1.3125,
"learning_rate": 0.0004997153858826786,
"loss": 5.9116,
"mean_token_accuracy": 0.14313025623559952,
"num_tokens": 5694516.0,
"step": 2750
},
{
"entropy": 5.86408429145813,
"epoch": 0.24990928882438315,
"grad_norm": 1.5234375,
"learning_rate": 0.0004997137566059922,
"loss": 5.7443,
"mean_token_accuracy": 0.15340098589658738,
"num_tokens": 5705443.0,
"step": 2755
},
{
"entropy": 5.797891855239868,
"epoch": 0.25036284470246734,
"grad_norm": 1.453125,
"learning_rate": 0.0004997121226821681,
"loss": 5.8459,
"mean_token_accuracy": 0.14177675023674965,
"num_tokens": 5715933.0,
"step": 2760
},
{
"entropy": 5.9241045951843265,
"epoch": 0.25081640058055155,
"grad_norm": 1.328125,
"learning_rate": 0.00049971048411124,
"loss": 5.8902,
"mean_token_accuracy": 0.14111201092600822,
"num_tokens": 5726630.0,
"step": 2765
},
{
"entropy": 5.972933149337768,
"epoch": 0.2512699564586357,
"grad_norm": 1.9296875,
"learning_rate": 0.0004997088408932418,
"loss": 5.8228,
"mean_token_accuracy": 0.14817024320363997,
"num_tokens": 5736890.0,
"step": 2770
},
{
"entropy": 5.754113483428955,
"epoch": 0.25172351233671986,
"grad_norm": 1.4140625,
"learning_rate": 0.0004997071930282076,
"loss": 5.7587,
"mean_token_accuracy": 0.1547361731529236,
"num_tokens": 5747042.0,
"step": 2775
},
{
"entropy": 5.825574254989624,
"epoch": 0.25217706821480407,
"grad_norm": 1.421875,
"learning_rate": 0.0004997055405161712,
"loss": 5.8655,
"mean_token_accuracy": 0.1438675545156002,
"num_tokens": 5757367.0,
"step": 2780
},
{
"entropy": 5.959324026107788,
"epoch": 0.2526306240928882,
"grad_norm": 2.0625,
"learning_rate": 0.0004997038833571671,
"loss": 5.9387,
"mean_token_accuracy": 0.14031409472227097,
"num_tokens": 5769557.0,
"step": 2785
},
{
"entropy": 5.897234869003296,
"epoch": 0.25308417997097243,
"grad_norm": 1.4453125,
"learning_rate": 0.0004997022215512295,
"loss": 5.8122,
"mean_token_accuracy": 0.15133150815963745,
"num_tokens": 5780615.0,
"step": 2790
},
{
"entropy": 5.852604007720947,
"epoch": 0.2535377358490566,
"grad_norm": 1.359375,
"learning_rate": 0.0004997005550983925,
"loss": 5.7948,
"mean_token_accuracy": 0.14928770214319229,
"num_tokens": 5791271.0,
"step": 2795
},
{
"entropy": 5.822273206710816,
"epoch": 0.2539912917271408,
"grad_norm": 1.484375,
"learning_rate": 0.000499698883998691,
"loss": 5.7444,
"mean_token_accuracy": 0.15246350467205047,
"num_tokens": 5801117.0,
"step": 2800
},
{
"entropy": 5.823004579544067,
"epoch": 0.25444484760522496,
"grad_norm": 1.359375,
"learning_rate": 0.0004996972082521591,
"loss": 5.7649,
"mean_token_accuracy": 0.15054019093513488,
"num_tokens": 5811696.0,
"step": 2805
},
{
"entropy": 5.84231653213501,
"epoch": 0.25489840348330917,
"grad_norm": 1.34375,
"learning_rate": 0.0004996955278588319,
"loss": 5.8063,
"mean_token_accuracy": 0.15581155121326445,
"num_tokens": 5822207.0,
"step": 2810
},
{
"entropy": 5.750502443313598,
"epoch": 0.2553519593613933,
"grad_norm": 1.6484375,
"learning_rate": 0.0004996938428187437,
"loss": 5.7558,
"mean_token_accuracy": 0.15488317459821702,
"num_tokens": 5831927.0,
"step": 2815
},
{
"entropy": 5.871583747863769,
"epoch": 0.2558055152394775,
"grad_norm": 1.453125,
"learning_rate": 0.0004996921531319297,
"loss": 5.8142,
"mean_token_accuracy": 0.1495785243809223,
"num_tokens": 5842093.0,
"step": 2820
},
{
"entropy": 5.852832698822022,
"epoch": 0.2562590711175617,
"grad_norm": 1.7421875,
"learning_rate": 0.0004996904587984246,
"loss": 5.7355,
"mean_token_accuracy": 0.15692816227674483,
"num_tokens": 5850551.0,
"step": 2825
},
{
"entropy": 5.794354200363159,
"epoch": 0.25671262699564584,
"grad_norm": 1.3984375,
"learning_rate": 0.0004996887598182636,
"loss": 5.7362,
"mean_token_accuracy": 0.1481923684477806,
"num_tokens": 5860651.0,
"step": 2830
},
{
"entropy": 5.777127456665039,
"epoch": 0.25716618287373005,
"grad_norm": 1.515625,
"learning_rate": 0.0004996870561914819,
"loss": 5.7259,
"mean_token_accuracy": 0.14743045717477798,
"num_tokens": 5871500.0,
"step": 2835
},
{
"entropy": 5.849510860443115,
"epoch": 0.2576197387518142,
"grad_norm": 1.5859375,
"learning_rate": 0.0004996853479181145,
"loss": 5.7273,
"mean_token_accuracy": 0.1445586636662483,
"num_tokens": 5881523.0,
"step": 2840
},
{
"entropy": 5.877736568450928,
"epoch": 0.2580732946298984,
"grad_norm": 1.40625,
"learning_rate": 0.0004996836349981969,
"loss": 5.8806,
"mean_token_accuracy": 0.14388025403022767,
"num_tokens": 5891162.0,
"step": 2845
},
{
"entropy": 5.786015796661377,
"epoch": 0.2585268505079826,
"grad_norm": 1.40625,
"learning_rate": 0.0004996819174317645,
"loss": 5.7689,
"mean_token_accuracy": 0.14543924555182458,
"num_tokens": 5901818.0,
"step": 2850
},
{
"entropy": 5.901238822937012,
"epoch": 0.2589804063860668,
"grad_norm": 1.4296875,
"learning_rate": 0.0004996801952188528,
"loss": 5.8677,
"mean_token_accuracy": 0.14473312646150588,
"num_tokens": 5913889.0,
"step": 2855
},
{
"entropy": 5.870723915100098,
"epoch": 0.25943396226415094,
"grad_norm": 1.3515625,
"learning_rate": 0.0004996784683594973,
"loss": 5.8056,
"mean_token_accuracy": 0.15123703628778457,
"num_tokens": 5923429.0,
"step": 2860
},
{
"entropy": 5.918413352966309,
"epoch": 0.2598875181422351,
"grad_norm": 1.703125,
"learning_rate": 0.0004996767368537338,
"loss": 5.8483,
"mean_token_accuracy": 0.14715803265571595,
"num_tokens": 5933847.0,
"step": 2865
},
{
"entropy": 5.805452156066894,
"epoch": 0.2603410740203193,
"grad_norm": 1.3828125,
"learning_rate": 0.0004996750007015983,
"loss": 5.7376,
"mean_token_accuracy": 0.14590465426445007,
"num_tokens": 5943874.0,
"step": 2870
},
{
"entropy": 5.8460094928741455,
"epoch": 0.26079462989840346,
"grad_norm": 1.546875,
"learning_rate": 0.0004996732599031264,
"loss": 5.7988,
"mean_token_accuracy": 0.14883473068475722,
"num_tokens": 5953653.0,
"step": 2875
},
{
"entropy": 5.866086196899414,
"epoch": 0.2612481857764877,
"grad_norm": 1.390625,
"learning_rate": 0.0004996715144583543,
"loss": 5.7492,
"mean_token_accuracy": 0.15125548392534255,
"num_tokens": 5964039.0,
"step": 2880
},
{
"entropy": 5.892557859420776,
"epoch": 0.26170174165457183,
"grad_norm": 2.265625,
"learning_rate": 0.000499669764367318,
"loss": 5.877,
"mean_token_accuracy": 0.14600060135126114,
"num_tokens": 5974083.0,
"step": 2885
},
{
"entropy": 5.818076229095459,
"epoch": 0.26215529753265604,
"grad_norm": 1.6328125,
"learning_rate": 0.0004996680096300537,
"loss": 5.8153,
"mean_token_accuracy": 0.14445886462926866,
"num_tokens": 5984629.0,
"step": 2890
},
{
"entropy": 5.754880905151367,
"epoch": 0.2626088534107402,
"grad_norm": 1.4609375,
"learning_rate": 0.0004996662502465978,
"loss": 5.7222,
"mean_token_accuracy": 0.15813563466072084,
"num_tokens": 5994295.0,
"step": 2895
},
{
"entropy": 5.811830186843872,
"epoch": 0.2630624092888244,
"grad_norm": 1.6796875,
"learning_rate": 0.0004996644862169864,
"loss": 5.7687,
"mean_token_accuracy": 0.14370385855436324,
"num_tokens": 6004156.0,
"step": 2900
},
{
"entropy": 5.8295127868652346,
"epoch": 0.26351596516690856,
"grad_norm": 1.5390625,
"learning_rate": 0.0004996627175412563,
"loss": 5.8448,
"mean_token_accuracy": 0.14985169619321823,
"num_tokens": 6014188.0,
"step": 2905
},
{
"entropy": 5.921795082092285,
"epoch": 0.2639695210449927,
"grad_norm": 1.5,
"learning_rate": 0.000499660944219444,
"loss": 5.906,
"mean_token_accuracy": 0.14295441657304764,
"num_tokens": 6024879.0,
"step": 2910
},
{
"entropy": 5.9418816566467285,
"epoch": 0.2644230769230769,
"grad_norm": 1.4296875,
"learning_rate": 0.0004996591662515861,
"loss": 5.9623,
"mean_token_accuracy": 0.14250990375876427,
"num_tokens": 6036374.0,
"step": 2915
},
{
"entropy": 5.917361497879028,
"epoch": 0.2648766328011611,
"grad_norm": 1.53125,
"learning_rate": 0.0004996573836377193,
"loss": 5.8081,
"mean_token_accuracy": 0.1467065617442131,
"num_tokens": 6046432.0,
"step": 2920
},
{
"entropy": 5.907295179367066,
"epoch": 0.2653301886792453,
"grad_norm": 1.828125,
"learning_rate": 0.0004996555963778806,
"loss": 5.8192,
"mean_token_accuracy": 0.14340163618326188,
"num_tokens": 6056864.0,
"step": 2925
},
{
"entropy": 5.953917980194092,
"epoch": 0.26578374455732945,
"grad_norm": 1.5234375,
"learning_rate": 0.000499653804472107,
"loss": 5.8473,
"mean_token_accuracy": 0.15042725205421448,
"num_tokens": 6067294.0,
"step": 2930
},
{
"entropy": 5.882991552352905,
"epoch": 0.26623730043541366,
"grad_norm": 1.65625,
"learning_rate": 0.0004996520079204353,
"loss": 5.85,
"mean_token_accuracy": 0.14360335022211074,
"num_tokens": 6077587.0,
"step": 2935
},
{
"entropy": 5.867487525939941,
"epoch": 0.2666908563134978,
"grad_norm": 1.8828125,
"learning_rate": 0.000499650206722903,
"loss": 5.8125,
"mean_token_accuracy": 0.14450383931398392,
"num_tokens": 6089306.0,
"step": 2940
},
{
"entropy": 5.856965208053589,
"epoch": 0.267144412191582,
"grad_norm": 1.6875,
"learning_rate": 0.0004996484008795471,
"loss": 5.7564,
"mean_token_accuracy": 0.14599506705999374,
"num_tokens": 6099651.0,
"step": 2945
},
{
"entropy": 5.834507179260254,
"epoch": 0.2675979680696662,
"grad_norm": 1.8671875,
"learning_rate": 0.000499646590390405,
"loss": 5.7657,
"mean_token_accuracy": 0.14419298321008683,
"num_tokens": 6110063.0,
"step": 2950
},
{
"entropy": 5.830401659011841,
"epoch": 0.26805152394775034,
"grad_norm": 1.4296875,
"learning_rate": 0.0004996447752555142,
"loss": 5.7865,
"mean_token_accuracy": 0.14656304121017455,
"num_tokens": 6120031.0,
"step": 2955
},
{
"entropy": 5.829970359802246,
"epoch": 0.26850507982583455,
"grad_norm": 1.2890625,
"learning_rate": 0.0004996429554749122,
"loss": 5.8324,
"mean_token_accuracy": 0.14829770997166633,
"num_tokens": 6131274.0,
"step": 2960
},
{
"entropy": 5.936783218383789,
"epoch": 0.2689586357039187,
"grad_norm": 1.53125,
"learning_rate": 0.0004996411310486367,
"loss": 5.8788,
"mean_token_accuracy": 0.14855986982584,
"num_tokens": 6140878.0,
"step": 2965
},
{
"entropy": 5.949974441528321,
"epoch": 0.2694121915820029,
"grad_norm": 1.734375,
"learning_rate": 0.0004996393019767253,
"loss": 5.879,
"mean_token_accuracy": 0.13954189866781236,
"num_tokens": 6152229.0,
"step": 2970
},
{
"entropy": 5.824478960037231,
"epoch": 0.26986574746008707,
"grad_norm": 1.2421875,
"learning_rate": 0.0004996374682592158,
"loss": 5.7744,
"mean_token_accuracy": 0.1475573480129242,
"num_tokens": 6162311.0,
"step": 2975
},
{
"entropy": 5.901595211029052,
"epoch": 0.2703193033381713,
"grad_norm": 1.7890625,
"learning_rate": 0.0004996356298961463,
"loss": 5.8833,
"mean_token_accuracy": 0.14554917365312575,
"num_tokens": 6172806.0,
"step": 2980
},
{
"entropy": 5.908715009689331,
"epoch": 0.27077285921625543,
"grad_norm": 1.84375,
"learning_rate": 0.0004996337868875549,
"loss": 5.8448,
"mean_token_accuracy": 0.14733956456184388,
"num_tokens": 6183479.0,
"step": 2985
},
{
"entropy": 5.861931991577149,
"epoch": 0.27122641509433965,
"grad_norm": 1.34375,
"learning_rate": 0.0004996319392334792,
"loss": 5.8176,
"mean_token_accuracy": 0.14184632897377014,
"num_tokens": 6194201.0,
"step": 2990
},
{
"entropy": 5.795678424835205,
"epoch": 0.2716799709724238,
"grad_norm": 1.421875,
"learning_rate": 0.000499630086933958,
"loss": 5.7641,
"mean_token_accuracy": 0.1457250237464905,
"num_tokens": 6204774.0,
"step": 2995
},
{
"entropy": 5.849139070510864,
"epoch": 0.27213352685050796,
"grad_norm": 1.5546875,
"learning_rate": 0.0004996282299890292,
"loss": 5.7696,
"mean_token_accuracy": 0.14415498152375222,
"num_tokens": 6215056.0,
"step": 3000
},
{
"epoch": 0.27213352685050796,
"eval_entropy": 5.725370093816006,
"eval_loss": 5.846706390380859,
"eval_mean_token_accuracy": 0.15213655662465825,
"eval_num_tokens": 6215056.0,
"eval_runtime": 26.8622,
"eval_samples_per_second": 1316.349,
"eval_steps_per_second": 164.544,
"step": 3000
},
{
"entropy": 5.905508375167846,
"epoch": 0.27258708272859217,
"grad_norm": 1.6796875,
"learning_rate": 0.0004996263683987316,
"loss": 5.8742,
"mean_token_accuracy": 0.14993443191051484,
"num_tokens": 6226264.0,
"step": 3005
},
{
"entropy": 5.872657299041748,
"epoch": 0.2730406386066763,
"grad_norm": 1.6015625,
"learning_rate": 0.0004996245021631033,
"loss": 5.7841,
"mean_token_accuracy": 0.14555028602480888,
"num_tokens": 6235137.0,
"step": 3010
},
{
"entropy": 5.804933309555054,
"epoch": 0.27349419448476053,
"grad_norm": 1.796875,
"learning_rate": 0.000499622631282183,
"loss": 5.8043,
"mean_token_accuracy": 0.14663855880498886,
"num_tokens": 6245411.0,
"step": 3015
},
{
"entropy": 5.824642276763916,
"epoch": 0.2739477503628447,
"grad_norm": 1.75,
"learning_rate": 0.0004996207557560095,
"loss": 5.8782,
"mean_token_accuracy": 0.1400340422987938,
"num_tokens": 6255633.0,
"step": 3020
},
{
"entropy": 5.924058246612549,
"epoch": 0.2744013062409289,
"grad_norm": 1.953125,
"learning_rate": 0.0004996188755846216,
"loss": 5.821,
"mean_token_accuracy": 0.14444293826818466,
"num_tokens": 6265109.0,
"step": 3025
},
{
"entropy": 5.900212144851684,
"epoch": 0.27485486211901305,
"grad_norm": 1.5625,
"learning_rate": 0.000499616990768058,
"loss": 5.8635,
"mean_token_accuracy": 0.14477069228887557,
"num_tokens": 6275241.0,
"step": 3030
},
{
"entropy": 5.914565849304199,
"epoch": 0.27530841799709727,
"grad_norm": 1.390625,
"learning_rate": 0.0004996151013063579,
"loss": 5.9067,
"mean_token_accuracy": 0.13996558040380477,
"num_tokens": 6286603.0,
"step": 3035
},
{
"entropy": 5.87188949584961,
"epoch": 0.2757619738751814,
"grad_norm": 1.421875,
"learning_rate": 0.0004996132071995601,
"loss": 5.7703,
"mean_token_accuracy": 0.1570067062973976,
"num_tokens": 6297219.0,
"step": 3040
},
{
"entropy": 5.8587874412536625,
"epoch": 0.2762155297532656,
"grad_norm": 1.765625,
"learning_rate": 0.0004996113084477042,
"loss": 5.8165,
"mean_token_accuracy": 0.15096617713570595,
"num_tokens": 6307694.0,
"step": 3045
},
{
"entropy": 5.857360363006592,
"epoch": 0.2766690856313498,
"grad_norm": 1.390625,
"learning_rate": 0.0004996094050508291,
"loss": 5.7446,
"mean_token_accuracy": 0.1507061928510666,
"num_tokens": 6317384.0,
"step": 3050
},
{
"entropy": 5.88347978591919,
"epoch": 0.27712264150943394,
"grad_norm": 1.890625,
"learning_rate": 0.0004996074970089741,
"loss": 5.8684,
"mean_token_accuracy": 0.1453227236866951,
"num_tokens": 6327558.0,
"step": 3055
},
{
"entropy": 5.903385925292969,
"epoch": 0.27757619738751815,
"grad_norm": 2.8125,
"learning_rate": 0.0004996055843221789,
"loss": 5.733,
"mean_token_accuracy": 0.15211346447467805,
"num_tokens": 6337828.0,
"step": 3060
},
{
"entropy": 5.817547464370728,
"epoch": 0.2780297532656023,
"grad_norm": 1.6328125,
"learning_rate": 0.0004996036669904831,
"loss": 5.796,
"mean_token_accuracy": 0.1491447240114212,
"num_tokens": 6348678.0,
"step": 3065
},
{
"entropy": 5.855010223388672,
"epoch": 0.2784833091436865,
"grad_norm": 1.5859375,
"learning_rate": 0.0004996017450139261,
"loss": 5.8856,
"mean_token_accuracy": 0.13843294233083725,
"num_tokens": 6359670.0,
"step": 3070
},
{
"entropy": 5.939880132675171,
"epoch": 0.2789368650217707,
"grad_norm": 1.5390625,
"learning_rate": 0.0004995998183925478,
"loss": 5.8442,
"mean_token_accuracy": 0.14635261446237563,
"num_tokens": 6371034.0,
"step": 3075
},
{
"entropy": 5.86761474609375,
"epoch": 0.2793904208998549,
"grad_norm": 1.5234375,
"learning_rate": 0.0004995978871263881,
"loss": 5.7823,
"mean_token_accuracy": 0.15000314861536027,
"num_tokens": 6380672.0,
"step": 3080
},
{
"entropy": 5.839673566818237,
"epoch": 0.27984397677793904,
"grad_norm": 1.921875,
"learning_rate": 0.0004995959512154868,
"loss": 5.8882,
"mean_token_accuracy": 0.1362109065055847,
"num_tokens": 6390799.0,
"step": 3085
},
{
"entropy": 5.879818630218506,
"epoch": 0.2802975326560232,
"grad_norm": 1.484375,
"learning_rate": 0.000499594010659884,
"loss": 5.7542,
"mean_token_accuracy": 0.14788413494825364,
"num_tokens": 6400965.0,
"step": 3090
},
{
"entropy": 5.896807670593262,
"epoch": 0.2807510885341074,
"grad_norm": 1.3984375,
"learning_rate": 0.0004995920654596199,
"loss": 5.7929,
"mean_token_accuracy": 0.14485295563936235,
"num_tokens": 6411567.0,
"step": 3095
},
{
"entropy": 5.884419775009155,
"epoch": 0.28120464441219156,
"grad_norm": 1.765625,
"learning_rate": 0.0004995901156147345,
"loss": 5.8379,
"mean_token_accuracy": 0.15128903090953827,
"num_tokens": 6421193.0,
"step": 3100
},
{
"entropy": 5.8871454238891605,
"epoch": 0.2816582002902758,
"grad_norm": 2.34375,
"learning_rate": 0.0004995881611252684,
"loss": 5.8437,
"mean_token_accuracy": 0.14912686347961426,
"num_tokens": 6430281.0,
"step": 3105
},
{
"entropy": 5.873918056488037,
"epoch": 0.28211175616835993,
"grad_norm": 1.6015625,
"learning_rate": 0.0004995862019912619,
"loss": 5.8482,
"mean_token_accuracy": 0.139264590293169,
"num_tokens": 6441055.0,
"step": 3110
},
{
"entropy": 5.8297398567199705,
"epoch": 0.28256531204644414,
"grad_norm": 1.375,
"learning_rate": 0.0004995842382127555,
"loss": 5.7594,
"mean_token_accuracy": 0.14350441247224807,
"num_tokens": 6451082.0,
"step": 3115
},
{
"entropy": 5.855344438552857,
"epoch": 0.2830188679245283,
"grad_norm": 1.5390625,
"learning_rate": 0.0004995822697897897,
"loss": 5.8585,
"mean_token_accuracy": 0.14142778068780898,
"num_tokens": 6461617.0,
"step": 3120
},
{
"entropy": 5.966115283966064,
"epoch": 0.2834724238026125,
"grad_norm": 1.6171875,
"learning_rate": 0.0004995802967224056,
"loss": 5.8475,
"mean_token_accuracy": 0.14636234045028687,
"num_tokens": 6472780.0,
"step": 3125
},
{
"entropy": 5.809420013427735,
"epoch": 0.28392597968069666,
"grad_norm": 2.28125,
"learning_rate": 0.0004995783190106434,
"loss": 5.7486,
"mean_token_accuracy": 0.1427581399679184,
"num_tokens": 6484078.0,
"step": 3130
},
{
"entropy": 5.916213893890381,
"epoch": 0.2843795355587808,
"grad_norm": 1.7578125,
"learning_rate": 0.0004995763366545445,
"loss": 5.8466,
"mean_token_accuracy": 0.14371495842933654,
"num_tokens": 6494366.0,
"step": 3135
},
{
"entropy": 5.791625833511352,
"epoch": 0.284833091436865,
"grad_norm": 1.828125,
"learning_rate": 0.0004995743496541497,
"loss": 5.8035,
"mean_token_accuracy": 0.15003674924373628,
"num_tokens": 6505461.0,
"step": 3140
},
{
"entropy": 5.825983953475952,
"epoch": 0.2852866473149492,
"grad_norm": 1.4765625,
"learning_rate": 0.0004995723580095001,
"loss": 5.7958,
"mean_token_accuracy": 0.15073160082101822,
"num_tokens": 6516141.0,
"step": 3145
},
{
"entropy": 5.8044600009918215,
"epoch": 0.2857402031930334,
"grad_norm": 1.765625,
"learning_rate": 0.000499570361720637,
"loss": 5.7816,
"mean_token_accuracy": 0.14948485642671586,
"num_tokens": 6525536.0,
"step": 3150
},
{
"entropy": 5.8823357105255125,
"epoch": 0.28619375907111755,
"grad_norm": 1.65625,
"learning_rate": 0.0004995683607876015,
"loss": 5.8401,
"mean_token_accuracy": 0.1408316560089588,
"num_tokens": 6535852.0,
"step": 3155
},
{
"entropy": 5.944931507110596,
"epoch": 0.28664731494920176,
"grad_norm": 2.359375,
"learning_rate": 0.0004995663552104352,
"loss": 5.8303,
"mean_token_accuracy": 0.14839636236429216,
"num_tokens": 6546353.0,
"step": 3160
},
{
"entropy": 5.803114652633667,
"epoch": 0.2871008708272859,
"grad_norm": 1.703125,
"learning_rate": 0.0004995643449891793,
"loss": 5.7719,
"mean_token_accuracy": 0.1433407634496689,
"num_tokens": 6557782.0,
"step": 3165
},
{
"entropy": 5.86722617149353,
"epoch": 0.2875544267053701,
"grad_norm": 1.703125,
"learning_rate": 0.0004995623301238757,
"loss": 5.7925,
"mean_token_accuracy": 0.1488977774977684,
"num_tokens": 6568100.0,
"step": 3170
},
{
"entropy": 5.907700634002685,
"epoch": 0.2880079825834543,
"grad_norm": 1.421875,
"learning_rate": 0.0004995603106145657,
"loss": 5.8455,
"mean_token_accuracy": 0.14495818465948104,
"num_tokens": 6578698.0,
"step": 3175
},
{
"entropy": 5.845079708099365,
"epoch": 0.28846153846153844,
"grad_norm": 1.5,
"learning_rate": 0.0004995582864612913,
"loss": 5.829,
"mean_token_accuracy": 0.14225727319717407,
"num_tokens": 6588908.0,
"step": 3180
},
{
"entropy": 5.81000714302063,
"epoch": 0.28891509433962265,
"grad_norm": 1.78125,
"learning_rate": 0.0004995562576640943,
"loss": 5.6816,
"mean_token_accuracy": 0.1479066714644432,
"num_tokens": 6599445.0,
"step": 3185
},
{
"entropy": 5.873440742492676,
"epoch": 0.2893686502177068,
"grad_norm": 1.4921875,
"learning_rate": 0.0004995542242230167,
"loss": 5.7901,
"mean_token_accuracy": 0.14480070844292642,
"num_tokens": 6610246.0,
"step": 3190
},
{
"entropy": 5.8062098026275635,
"epoch": 0.289822206095791,
"grad_norm": 2.84375,
"learning_rate": 0.0004995521861381006,
"loss": 5.7896,
"mean_token_accuracy": 0.1486334338784218,
"num_tokens": 6620501.0,
"step": 3195
},
{
"entropy": 6.001724576950073,
"epoch": 0.29027576197387517,
"grad_norm": 1.546875,
"learning_rate": 0.000499550143409388,
"loss": 5.8694,
"mean_token_accuracy": 0.13826330080628396,
"num_tokens": 6631206.0,
"step": 3200
},
{
"entropy": 5.822874259948731,
"epoch": 0.2907293178519594,
"grad_norm": 1.390625,
"learning_rate": 0.0004995480960369211,
"loss": 5.803,
"mean_token_accuracy": 0.14362403452396394,
"num_tokens": 6641600.0,
"step": 3205
},
{
"entropy": 5.847076559066773,
"epoch": 0.29118287373004353,
"grad_norm": 1.7734375,
"learning_rate": 0.0004995460440207425,
"loss": 5.8789,
"mean_token_accuracy": 0.13984269350767137,
"num_tokens": 6652319.0,
"step": 3210
},
{
"entropy": 5.926595544815063,
"epoch": 0.29163642960812775,
"grad_norm": 1.765625,
"learning_rate": 0.0004995439873608945,
"loss": 5.8209,
"mean_token_accuracy": 0.13916896730661393,
"num_tokens": 6663702.0,
"step": 3215
},
{
"entropy": 5.832406425476075,
"epoch": 0.2920899854862119,
"grad_norm": 1.5078125,
"learning_rate": 0.0004995419260574196,
"loss": 5.7932,
"mean_token_accuracy": 0.14231335520744323,
"num_tokens": 6673908.0,
"step": 3220
},
{
"entropy": 5.786562442779541,
"epoch": 0.29254354136429606,
"grad_norm": 2.375,
"learning_rate": 0.0004995398601103603,
"loss": 5.8235,
"mean_token_accuracy": 0.14949814826250077,
"num_tokens": 6683237.0,
"step": 3225
},
{
"entropy": 5.869636964797974,
"epoch": 0.29299709724238027,
"grad_norm": 1.8359375,
"learning_rate": 0.0004995377895197597,
"loss": 5.7685,
"mean_token_accuracy": 0.1467011883854866,
"num_tokens": 6693451.0,
"step": 3230
},
{
"entropy": 5.8066086769104,
"epoch": 0.2934506531204644,
"grad_norm": 1.5,
"learning_rate": 0.0004995357142856602,
"loss": 5.7103,
"mean_token_accuracy": 0.14754725396633148,
"num_tokens": 6703479.0,
"step": 3235
},
{
"entropy": 5.785593414306641,
"epoch": 0.29390420899854863,
"grad_norm": 1.6015625,
"learning_rate": 0.000499533634408105,
"loss": 5.722,
"mean_token_accuracy": 0.15292183309793472,
"num_tokens": 6715321.0,
"step": 3240
},
{
"entropy": 5.8156593322753904,
"epoch": 0.2943577648766328,
"grad_norm": 1.625,
"learning_rate": 0.000499531549887137,
"loss": 5.7658,
"mean_token_accuracy": 0.14359535425901412,
"num_tokens": 6725541.0,
"step": 3245
},
{
"entropy": 5.82113208770752,
"epoch": 0.294811320754717,
"grad_norm": 2.046875,
"learning_rate": 0.0004995294607227993,
"loss": 5.8205,
"mean_token_accuracy": 0.15142790526151656,
"num_tokens": 6735165.0,
"step": 3250
},
{
"entropy": 5.913179206848144,
"epoch": 0.29526487663280115,
"grad_norm": 1.515625,
"learning_rate": 0.0004995273669151351,
"loss": 5.8263,
"mean_token_accuracy": 0.1507417544722557,
"num_tokens": 6744245.0,
"step": 3255
},
{
"entropy": 5.818991231918335,
"epoch": 0.29571843251088537,
"grad_norm": 1.7265625,
"learning_rate": 0.000499525268464188,
"loss": 5.7806,
"mean_token_accuracy": 0.14803050607442855,
"num_tokens": 6754894.0,
"step": 3260
},
{
"entropy": 5.822706174850464,
"epoch": 0.2961719883889695,
"grad_norm": 1.5078125,
"learning_rate": 0.0004995231653700009,
"loss": 5.8398,
"mean_token_accuracy": 0.13901015520095825,
"num_tokens": 6764442.0,
"step": 3265
},
{
"entropy": 5.904806232452392,
"epoch": 0.2966255442670537,
"grad_norm": 1.78125,
"learning_rate": 0.0004995210576326176,
"loss": 5.8025,
"mean_token_accuracy": 0.14261318892240524,
"num_tokens": 6775054.0,
"step": 3270
},
{
"entropy": 5.842242336273193,
"epoch": 0.2970791001451379,
"grad_norm": 1.7890625,
"learning_rate": 0.0004995189452520816,
"loss": 5.7562,
"mean_token_accuracy": 0.1599966213107109,
"num_tokens": 6784825.0,
"step": 3275
},
{
"entropy": 5.856751728057861,
"epoch": 0.29753265602322204,
"grad_norm": 1.5546875,
"learning_rate": 0.0004995168282284366,
"loss": 5.8218,
"mean_token_accuracy": 0.14728388488292693,
"num_tokens": 6795928.0,
"step": 3280
},
{
"entropy": 5.783289575576783,
"epoch": 0.29798621190130625,
"grad_norm": 1.5234375,
"learning_rate": 0.0004995147065617263,
"loss": 5.7394,
"mean_token_accuracy": 0.14838822185993195,
"num_tokens": 6805456.0,
"step": 3285
},
{
"entropy": 5.89187536239624,
"epoch": 0.2984397677793904,
"grad_norm": 1.4921875,
"learning_rate": 0.0004995125802519946,
"loss": 5.8327,
"mean_token_accuracy": 0.14377516657114028,
"num_tokens": 6816215.0,
"step": 3290
},
{
"entropy": 5.943170499801636,
"epoch": 0.2988933236574746,
"grad_norm": 3.015625,
"learning_rate": 0.0004995104492992857,
"loss": 5.8535,
"mean_token_accuracy": 0.14646795988082886,
"num_tokens": 6826545.0,
"step": 3295
},
{
"entropy": 5.759806346893311,
"epoch": 0.2993468795355588,
"grad_norm": 1.8046875,
"learning_rate": 0.0004995083137036434,
"loss": 5.7702,
"mean_token_accuracy": 0.14964656680822372,
"num_tokens": 6836184.0,
"step": 3300
},
{
"entropy": 5.910308742523194,
"epoch": 0.299800435413643,
"grad_norm": 1.5625,
"learning_rate": 0.0004995061734651121,
"loss": 5.8488,
"mean_token_accuracy": 0.14388236552476882,
"num_tokens": 6846646.0,
"step": 3305
},
{
"entropy": 5.911490964889526,
"epoch": 0.30025399129172714,
"grad_norm": 1.8828125,
"learning_rate": 0.0004995040285837358,
"loss": 5.7612,
"mean_token_accuracy": 0.15329435765743255,
"num_tokens": 6857597.0,
"step": 3310
},
{
"entropy": 5.85549144744873,
"epoch": 0.3007075471698113,
"grad_norm": 1.8203125,
"learning_rate": 0.0004995018790595589,
"loss": 5.854,
"mean_token_accuracy": 0.14282366782426834,
"num_tokens": 6867523.0,
"step": 3315
},
{
"entropy": 5.91640453338623,
"epoch": 0.3011611030478955,
"grad_norm": 1.625,
"learning_rate": 0.0004994997248926261,
"loss": 5.946,
"mean_token_accuracy": 0.14262509644031524,
"num_tokens": 6878190.0,
"step": 3320
},
{
"entropy": 5.903401231765747,
"epoch": 0.30161465892597966,
"grad_norm": 1.9296875,
"learning_rate": 0.0004994975660829817,
"loss": 5.7563,
"mean_token_accuracy": 0.1535111017525196,
"num_tokens": 6888483.0,
"step": 3325
},
{
"entropy": 5.83467206954956,
"epoch": 0.3020682148040639,
"grad_norm": 1.6328125,
"learning_rate": 0.0004994954026306705,
"loss": 5.7374,
"mean_token_accuracy": 0.14736103788018226,
"num_tokens": 6898337.0,
"step": 3330
},
{
"entropy": 5.88395357131958,
"epoch": 0.30252177068214803,
"grad_norm": 1.5078125,
"learning_rate": 0.0004994932345357371,
"loss": 5.8065,
"mean_token_accuracy": 0.14424017667770386,
"num_tokens": 6909131.0,
"step": 3335
},
{
"entropy": 5.859859037399292,
"epoch": 0.30297532656023224,
"grad_norm": 1.46875,
"learning_rate": 0.0004994910617982264,
"loss": 5.7906,
"mean_token_accuracy": 0.14290710389614106,
"num_tokens": 6919790.0,
"step": 3340
},
{
"entropy": 5.829497575759888,
"epoch": 0.3034288824383164,
"grad_norm": 1.5859375,
"learning_rate": 0.0004994888844181833,
"loss": 5.8129,
"mean_token_accuracy": 0.14728963971138,
"num_tokens": 6930072.0,
"step": 3345
},
{
"entropy": 5.788367462158203,
"epoch": 0.3038824383164006,
"grad_norm": 2.296875,
"learning_rate": 0.0004994867023956529,
"loss": 5.707,
"mean_token_accuracy": 0.14217354729771614,
"num_tokens": 6940646.0,
"step": 3350
},
{
"entropy": 5.819201564788818,
"epoch": 0.30433599419448476,
"grad_norm": 1.484375,
"learning_rate": 0.0004994845157306803,
"loss": 5.7089,
"mean_token_accuracy": 0.1530962035059929,
"num_tokens": 6951645.0,
"step": 3355
},
{
"entropy": 5.88507342338562,
"epoch": 0.3047895500725689,
"grad_norm": 1.6796875,
"learning_rate": 0.0004994823244233107,
"loss": 5.8611,
"mean_token_accuracy": 0.14595605432987213,
"num_tokens": 6963020.0,
"step": 3360
},
{
"entropy": 5.900927877426147,
"epoch": 0.3052431059506531,
"grad_norm": 1.5234375,
"learning_rate": 0.0004994801284735895,
"loss": 5.7878,
"mean_token_accuracy": 0.15086106210947037,
"num_tokens": 6972895.0,
"step": 3365
},
{
"entropy": 5.894740104675293,
"epoch": 0.3056966618287373,
"grad_norm": 1.5390625,
"learning_rate": 0.0004994779278815619,
"loss": 5.9297,
"mean_token_accuracy": 0.13789867609739304,
"num_tokens": 6985363.0,
"step": 3370
},
{
"entropy": 5.9384606838226315,
"epoch": 0.3061502177068215,
"grad_norm": 1.6640625,
"learning_rate": 0.0004994757226472736,
"loss": 5.8593,
"mean_token_accuracy": 0.1438434675335884,
"num_tokens": 6997040.0,
"step": 3375
},
{
"entropy": 5.856701517105103,
"epoch": 0.30660377358490565,
"grad_norm": 1.421875,
"learning_rate": 0.0004994735127707702,
"loss": 5.7905,
"mean_token_accuracy": 0.1451782375574112,
"num_tokens": 7007591.0,
"step": 3380
},
{
"entropy": 5.88822865486145,
"epoch": 0.30705732946298986,
"grad_norm": 1.765625,
"learning_rate": 0.0004994712982520972,
"loss": 5.7923,
"mean_token_accuracy": 0.15586419701576232,
"num_tokens": 7017858.0,
"step": 3385
},
{
"entropy": 5.770430278778076,
"epoch": 0.307510885341074,
"grad_norm": 1.3984375,
"learning_rate": 0.0004994690790913008,
"loss": 5.7643,
"mean_token_accuracy": 0.148378449678421,
"num_tokens": 7027365.0,
"step": 3390
},
{
"entropy": 5.80420503616333,
"epoch": 0.3079644412191582,
"grad_norm": 1.7890625,
"learning_rate": 0.0004994668552884265,
"loss": 5.7623,
"mean_token_accuracy": 0.14726244360208512,
"num_tokens": 7038728.0,
"step": 3395
},
{
"entropy": 5.905961561203003,
"epoch": 0.3084179970972424,
"grad_norm": 1.6015625,
"learning_rate": 0.0004994646268435204,
"loss": 5.7059,
"mean_token_accuracy": 0.1534990206360817,
"num_tokens": 7048736.0,
"step": 3400
},
{
"entropy": 5.824774742126465,
"epoch": 0.30887155297532654,
"grad_norm": 1.5625,
"learning_rate": 0.0004994623937566286,
"loss": 5.8291,
"mean_token_accuracy": 0.14722448736429214,
"num_tokens": 7059205.0,
"step": 3405
},
{
"entropy": 5.819243335723877,
"epoch": 0.30932510885341075,
"grad_norm": 1.5078125,
"learning_rate": 0.0004994601560277974,
"loss": 5.839,
"mean_token_accuracy": 0.1430293306708336,
"num_tokens": 7069513.0,
"step": 3410
},
{
"entropy": 5.857053995132446,
"epoch": 0.3097786647314949,
"grad_norm": 1.59375,
"learning_rate": 0.000499457913657073,
"loss": 5.75,
"mean_token_accuracy": 0.14733252078294753,
"num_tokens": 7079940.0,
"step": 3415
},
{
"entropy": 5.85497088432312,
"epoch": 0.3102322206095791,
"grad_norm": 1.4609375,
"learning_rate": 0.0004994556666445016,
"loss": 5.7543,
"mean_token_accuracy": 0.14875707179307937,
"num_tokens": 7090792.0,
"step": 3420
},
{
"entropy": 5.770295095443726,
"epoch": 0.31068577648766327,
"grad_norm": 1.4453125,
"learning_rate": 0.0004994534149901299,
"loss": 5.6985,
"mean_token_accuracy": 0.1492018073797226,
"num_tokens": 7101964.0,
"step": 3425
},
{
"entropy": 5.804609155654907,
"epoch": 0.3111393323657475,
"grad_norm": 1.6640625,
"learning_rate": 0.0004994511586940044,
"loss": 5.774,
"mean_token_accuracy": 0.14447008967399597,
"num_tokens": 7113332.0,
"step": 3430
},
{
"entropy": 5.976766204833984,
"epoch": 0.31159288824383163,
"grad_norm": 1.515625,
"learning_rate": 0.0004994488977561717,
"loss": 5.9305,
"mean_token_accuracy": 0.14386181086301802,
"num_tokens": 7123481.0,
"step": 3435
},
{
"entropy": 5.814102554321289,
"epoch": 0.31204644412191584,
"grad_norm": 4.09375,
"learning_rate": 0.0004994466321766786,
"loss": 5.6934,
"mean_token_accuracy": 0.15259597599506378,
"num_tokens": 7133694.0,
"step": 3440
},
{
"entropy": 5.8040472030639645,
"epoch": 0.3125,
"grad_norm": 1.5859375,
"learning_rate": 0.000499444361955572,
"loss": 5.8344,
"mean_token_accuracy": 0.1501459300518036,
"num_tokens": 7144491.0,
"step": 3445
},
{
"entropy": 6.012900972366333,
"epoch": 0.31295355587808416,
"grad_norm": 1.328125,
"learning_rate": 0.0004994420870928988,
"loss": 5.8826,
"mean_token_accuracy": 0.1436660721898079,
"num_tokens": 7154716.0,
"step": 3450
},
{
"entropy": 5.904228782653808,
"epoch": 0.31340711175616837,
"grad_norm": 1.4296875,
"learning_rate": 0.0004994398075887059,
"loss": 5.8806,
"mean_token_accuracy": 0.14098595678806305,
"num_tokens": 7164258.0,
"step": 3455
},
{
"entropy": 5.848430109024048,
"epoch": 0.3138606676342525,
"grad_norm": 1.2734375,
"learning_rate": 0.0004994375234430407,
"loss": 5.745,
"mean_token_accuracy": 0.14507319629192353,
"num_tokens": 7175098.0,
"step": 3460
},
{
"entropy": 5.868549537658692,
"epoch": 0.31431422351233673,
"grad_norm": 1.578125,
"learning_rate": 0.0004994352346559504,
"loss": 5.8168,
"mean_token_accuracy": 0.1449630543589592,
"num_tokens": 7186480.0,
"step": 3465
},
{
"entropy": 5.788954830169677,
"epoch": 0.3147677793904209,
"grad_norm": 1.71875,
"learning_rate": 0.0004994329412274821,
"loss": 5.7886,
"mean_token_accuracy": 0.1490990549325943,
"num_tokens": 7197056.0,
"step": 3470
},
{
"entropy": 5.904733800888062,
"epoch": 0.3152213352685051,
"grad_norm": 1.515625,
"learning_rate": 0.0004994306431576835,
"loss": 5.9056,
"mean_token_accuracy": 0.1437280908226967,
"num_tokens": 7206040.0,
"step": 3475
},
{
"entropy": 5.937067651748658,
"epoch": 0.31567489114658925,
"grad_norm": 1.3203125,
"learning_rate": 0.000499428340446602,
"loss": 5.7542,
"mean_token_accuracy": 0.1492527186870575,
"num_tokens": 7216344.0,
"step": 3480
},
{
"entropy": 5.88311448097229,
"epoch": 0.31612844702467346,
"grad_norm": 1.65625,
"learning_rate": 0.0004994260330942851,
"loss": 5.8348,
"mean_token_accuracy": 0.14428199976682662,
"num_tokens": 7226729.0,
"step": 3485
},
{
"entropy": 5.773128843307495,
"epoch": 0.3165820029027576,
"grad_norm": 1.3828125,
"learning_rate": 0.0004994237211007808,
"loss": 5.7887,
"mean_token_accuracy": 0.14559608250856398,
"num_tokens": 7237616.0,
"step": 3490
},
{
"entropy": 5.857798624038696,
"epoch": 0.3170355587808418,
"grad_norm": 1.5625,
"learning_rate": 0.0004994214044661367,
"loss": 5.7252,
"mean_token_accuracy": 0.1498093731701374,
"num_tokens": 7247857.0,
"step": 3495
},
{
"entropy": 5.719632959365844,
"epoch": 0.317489114658926,
"grad_norm": 1.578125,
"learning_rate": 0.0004994190831904007,
"loss": 5.688,
"mean_token_accuracy": 0.15702230781316756,
"num_tokens": 7257858.0,
"step": 3500
},
{
"entropy": 5.65029010772705,
"epoch": 0.31794267053701014,
"grad_norm": 1.5625,
"learning_rate": 0.000499416757273621,
"loss": 5.6898,
"mean_token_accuracy": 0.15247652679681778,
"num_tokens": 7268440.0,
"step": 3505
},
{
"entropy": 5.920102977752686,
"epoch": 0.31839622641509435,
"grad_norm": 1.7109375,
"learning_rate": 0.0004994144267158454,
"loss": 5.7827,
"mean_token_accuracy": 0.1534908801317215,
"num_tokens": 7278560.0,
"step": 3510
},
{
"entropy": 5.888502550125122,
"epoch": 0.3188497822931785,
"grad_norm": 2.296875,
"learning_rate": 0.0004994120915171224,
"loss": 5.7946,
"mean_token_accuracy": 0.14598636478185653,
"num_tokens": 7288422.0,
"step": 3515
},
{
"entropy": 5.87982177734375,
"epoch": 0.3193033381712627,
"grad_norm": 1.4609375,
"learning_rate": 0.0004994097516775002,
"loss": 5.8287,
"mean_token_accuracy": 0.14544341266155242,
"num_tokens": 7298613.0,
"step": 3520
},
{
"entropy": 5.874675273895264,
"epoch": 0.3197568940493469,
"grad_norm": 1.640625,
"learning_rate": 0.0004994074071970271,
"loss": 5.8742,
"mean_token_accuracy": 0.1415143169462681,
"num_tokens": 7309509.0,
"step": 3525
},
{
"entropy": 5.883598661422729,
"epoch": 0.3202104499274311,
"grad_norm": 2.078125,
"learning_rate": 0.0004994050580757515,
"loss": 5.7807,
"mean_token_accuracy": 0.14771121591329575,
"num_tokens": 7319015.0,
"step": 3530
},
{
"entropy": 5.7930444240570065,
"epoch": 0.32066400580551524,
"grad_norm": 3.65625,
"learning_rate": 0.0004994027043137223,
"loss": 5.7605,
"mean_token_accuracy": 0.15096884369850158,
"num_tokens": 7329026.0,
"step": 3535
},
{
"entropy": 5.928781509399414,
"epoch": 0.3211175616835994,
"grad_norm": 1.671875,
"learning_rate": 0.0004994003459109879,
"loss": 5.8383,
"mean_token_accuracy": 0.14945371747016906,
"num_tokens": 7339638.0,
"step": 3540
},
{
"entropy": 5.856835460662841,
"epoch": 0.3215711175616836,
"grad_norm": 1.5703125,
"learning_rate": 0.0004993979828675972,
"loss": 5.7468,
"mean_token_accuracy": 0.15235433876514434,
"num_tokens": 7349979.0,
"step": 3545
},
{
"entropy": 5.739511013031006,
"epoch": 0.32202467343976776,
"grad_norm": 1.6484375,
"learning_rate": 0.000499395615183599,
"loss": 5.6826,
"mean_token_accuracy": 0.1585940971970558,
"num_tokens": 7359398.0,
"step": 3550
},
{
"entropy": 5.767411756515503,
"epoch": 0.32247822931785197,
"grad_norm": 1.859375,
"learning_rate": 0.0004993932428590422,
"loss": 5.7347,
"mean_token_accuracy": 0.14846241772174834,
"num_tokens": 7370418.0,
"step": 3555
},
{
"entropy": 5.858153820037842,
"epoch": 0.3229317851959361,
"grad_norm": 1.7265625,
"learning_rate": 0.0004993908658939762,
"loss": 5.7541,
"mean_token_accuracy": 0.1416775479912758,
"num_tokens": 7380461.0,
"step": 3560
},
{
"entropy": 5.848527336120606,
"epoch": 0.32338534107402034,
"grad_norm": 1.453125,
"learning_rate": 0.0004993884842884496,
"loss": 5.7846,
"mean_token_accuracy": 0.14853670224547386,
"num_tokens": 7390646.0,
"step": 3565
},
{
"entropy": 5.840066242218017,
"epoch": 0.3238388969521045,
"grad_norm": 1.5078125,
"learning_rate": 0.0004993860980425121,
"loss": 5.8707,
"mean_token_accuracy": 0.14578727930784224,
"num_tokens": 7402106.0,
"step": 3570
},
{
"entropy": 5.859009647369385,
"epoch": 0.3242924528301887,
"grad_norm": 1.2890625,
"learning_rate": 0.0004993837071562131,
"loss": 5.86,
"mean_token_accuracy": 0.1438901849091053,
"num_tokens": 7412936.0,
"step": 3575
},
{
"entropy": 5.876029348373413,
"epoch": 0.32474600870827286,
"grad_norm": 1.5546875,
"learning_rate": 0.0004993813116296017,
"loss": 5.7174,
"mean_token_accuracy": 0.15487630516290665,
"num_tokens": 7422303.0,
"step": 3580
},
{
"entropy": 5.815586090087891,
"epoch": 0.325199564586357,
"grad_norm": 1.59375,
"learning_rate": 0.0004993789114627275,
"loss": 5.7829,
"mean_token_accuracy": 0.14683734625577927,
"num_tokens": 7432829.0,
"step": 3585
},
{
"entropy": 5.8798154354095455,
"epoch": 0.3256531204644412,
"grad_norm": 1.3671875,
"learning_rate": 0.0004993765066556403,
"loss": 5.8139,
"mean_token_accuracy": 0.1471313163638115,
"num_tokens": 7443218.0,
"step": 3590
},
{
"entropy": 5.878670930862427,
"epoch": 0.3261066763425254,
"grad_norm": 1.53125,
"learning_rate": 0.0004993740972083899,
"loss": 5.7585,
"mean_token_accuracy": 0.14762550145387648,
"num_tokens": 7454531.0,
"step": 3595
},
{
"entropy": 5.842435550689697,
"epoch": 0.3265602322206096,
"grad_norm": 1.4140625,
"learning_rate": 0.0004993716831210259,
"loss": 5.819,
"mean_token_accuracy": 0.14858943670988084,
"num_tokens": 7464993.0,
"step": 3600
},
{
"entropy": 5.836248922348022,
"epoch": 0.32701378809869375,
"grad_norm": 1.3359375,
"learning_rate": 0.0004993692643935983,
"loss": 5.8484,
"mean_token_accuracy": 0.14623567909002305,
"num_tokens": 7475755.0,
"step": 3605
},
{
"entropy": 5.869288301467895,
"epoch": 0.32746734397677796,
"grad_norm": 1.828125,
"learning_rate": 0.0004993668410261571,
"loss": 5.8342,
"mean_token_accuracy": 0.14813216850161554,
"num_tokens": 7486778.0,
"step": 3610
},
{
"entropy": 5.928134870529175,
"epoch": 0.3279208998548621,
"grad_norm": 1.609375,
"learning_rate": 0.0004993644130187525,
"loss": 5.8186,
"mean_token_accuracy": 0.14781841933727263,
"num_tokens": 7497439.0,
"step": 3615
},
{
"entropy": 5.862684679031372,
"epoch": 0.3283744557329463,
"grad_norm": 1.5859375,
"learning_rate": 0.0004993619803714347,
"loss": 5.8521,
"mean_token_accuracy": 0.14881017282605172,
"num_tokens": 7508274.0,
"step": 3620
},
{
"entropy": 5.91369595527649,
"epoch": 0.3288280116110305,
"grad_norm": 1.59375,
"learning_rate": 0.0004993595430842539,
"loss": 5.8215,
"mean_token_accuracy": 0.14630042612552643,
"num_tokens": 7519315.0,
"step": 3625
},
{
"entropy": 5.955614137649536,
"epoch": 0.32928156748911463,
"grad_norm": 1.484375,
"learning_rate": 0.0004993571011572606,
"loss": 5.8233,
"mean_token_accuracy": 0.14899712055921555,
"num_tokens": 7529910.0,
"step": 3630
},
{
"entropy": 5.791502332687378,
"epoch": 0.32973512336719885,
"grad_norm": 1.7734375,
"learning_rate": 0.0004993546545905053,
"loss": 5.7414,
"mean_token_accuracy": 0.15022289454936982,
"num_tokens": 7538966.0,
"step": 3635
},
{
"entropy": 5.87908525466919,
"epoch": 0.330188679245283,
"grad_norm": 1.4453125,
"learning_rate": 0.0004993522033840386,
"loss": 5.7835,
"mean_token_accuracy": 0.14871595650911332,
"num_tokens": 7549683.0,
"step": 3640
},
{
"entropy": 5.7719274997711185,
"epoch": 0.3306422351233672,
"grad_norm": 1.265625,
"learning_rate": 0.000499349747537911,
"loss": 5.7657,
"mean_token_accuracy": 0.1503421112895012,
"num_tokens": 7560629.0,
"step": 3645
},
{
"entropy": 5.821841669082642,
"epoch": 0.33109579100145137,
"grad_norm": 1.375,
"learning_rate": 0.0004993472870521736,
"loss": 5.7092,
"mean_token_accuracy": 0.15626542568206786,
"num_tokens": 7570613.0,
"step": 3650
},
{
"entropy": 5.8016807556152346,
"epoch": 0.3315493468795356,
"grad_norm": 1.46875,
"learning_rate": 0.0004993448219268771,
"loss": 5.6923,
"mean_token_accuracy": 0.14716383442282677,
"num_tokens": 7580640.0,
"step": 3655
},
{
"entropy": 5.802517318725586,
"epoch": 0.33200290275761973,
"grad_norm": 1.65625,
"learning_rate": 0.0004993423521620726,
"loss": 5.7793,
"mean_token_accuracy": 0.14909568727016448,
"num_tokens": 7590164.0,
"step": 3660
},
{
"entropy": 5.9077465534210205,
"epoch": 0.33245645863570394,
"grad_norm": 1.625,
"learning_rate": 0.0004993398777578109,
"loss": 5.8267,
"mean_token_accuracy": 0.14484120905399323,
"num_tokens": 7600173.0,
"step": 3665
},
{
"entropy": 5.806748008728027,
"epoch": 0.3329100145137881,
"grad_norm": 1.5390625,
"learning_rate": 0.0004993373987141435,
"loss": 5.6382,
"mean_token_accuracy": 0.1568427190184593,
"num_tokens": 7609958.0,
"step": 3670
},
{
"entropy": 5.874347925186157,
"epoch": 0.33336357039187225,
"grad_norm": 1.484375,
"learning_rate": 0.0004993349150311214,
"loss": 5.7765,
"mean_token_accuracy": 0.14990526288747788,
"num_tokens": 7620237.0,
"step": 3675
},
{
"entropy": 5.73986234664917,
"epoch": 0.33381712626995647,
"grad_norm": 1.359375,
"learning_rate": 0.0004993324267087962,
"loss": 5.6779,
"mean_token_accuracy": 0.15874082297086717,
"num_tokens": 7629886.0,
"step": 3680
},
{
"entropy": 5.78931999206543,
"epoch": 0.3342706821480406,
"grad_norm": 1.5,
"learning_rate": 0.0004993299337472192,
"loss": 5.7604,
"mean_token_accuracy": 0.15158218294382095,
"num_tokens": 7640057.0,
"step": 3685
},
{
"entropy": 5.79057207107544,
"epoch": 0.33472423802612483,
"grad_norm": 1.8125,
"learning_rate": 0.000499327436146442,
"loss": 5.6948,
"mean_token_accuracy": 0.14951762855052947,
"num_tokens": 7649766.0,
"step": 3690
},
{
"entropy": 5.817282056808471,
"epoch": 0.335177793904209,
"grad_norm": 2.546875,
"learning_rate": 0.0004993249339065162,
"loss": 5.7712,
"mean_token_accuracy": 0.15388251468539238,
"num_tokens": 7660890.0,
"step": 3695
},
{
"entropy": 5.87283844947815,
"epoch": 0.3356313497822932,
"grad_norm": 1.671875,
"learning_rate": 0.0004993224270274937,
"loss": 5.7946,
"mean_token_accuracy": 0.14841417819261551,
"num_tokens": 7670737.0,
"step": 3700
},
{
"entropy": 5.744346475601196,
"epoch": 0.33608490566037735,
"grad_norm": 1.625,
"learning_rate": 0.0004993199155094263,
"loss": 5.7533,
"mean_token_accuracy": 0.14859601408243178,
"num_tokens": 7681407.0,
"step": 3705
},
{
"entropy": 5.775976657867432,
"epoch": 0.33653846153846156,
"grad_norm": 1.4296875,
"learning_rate": 0.0004993173993523658,
"loss": 5.6882,
"mean_token_accuracy": 0.15778839737176895,
"num_tokens": 7691758.0,
"step": 3710
},
{
"entropy": 5.817459774017334,
"epoch": 0.3369920174165457,
"grad_norm": 1.3203125,
"learning_rate": 0.0004993148785563643,
"loss": 5.7122,
"mean_token_accuracy": 0.153327476978302,
"num_tokens": 7702934.0,
"step": 3715
},
{
"entropy": 5.7213376522064205,
"epoch": 0.3374455732946299,
"grad_norm": 1.359375,
"learning_rate": 0.000499312353121474,
"loss": 5.7154,
"mean_token_accuracy": 0.14955039769411088,
"num_tokens": 7713001.0,
"step": 3720
},
{
"entropy": 5.843008708953858,
"epoch": 0.3378991291727141,
"grad_norm": 1.640625,
"learning_rate": 0.0004993098230477471,
"loss": 5.7729,
"mean_token_accuracy": 0.14066959023475648,
"num_tokens": 7724005.0,
"step": 3725
},
{
"entropy": 5.86332745552063,
"epoch": 0.33835268505079824,
"grad_norm": 1.859375,
"learning_rate": 0.0004993072883352358,
"loss": 5.7307,
"mean_token_accuracy": 0.14685281068086625,
"num_tokens": 7733743.0,
"step": 3730
},
{
"entropy": 5.855350208282471,
"epoch": 0.33880624092888245,
"grad_norm": 1.65625,
"learning_rate": 0.0004993047489839927,
"loss": 5.9077,
"mean_token_accuracy": 0.1421743795275688,
"num_tokens": 7743762.0,
"step": 3735
},
{
"entropy": 5.921744632720947,
"epoch": 0.3392597968069666,
"grad_norm": 1.375,
"learning_rate": 0.0004993022049940702,
"loss": 5.8021,
"mean_token_accuracy": 0.15184042006731033,
"num_tokens": 7753758.0,
"step": 3740
},
{
"entropy": 5.794305658340454,
"epoch": 0.3397133526850508,
"grad_norm": 1.4453125,
"learning_rate": 0.0004992996563655209,
"loss": 5.7383,
"mean_token_accuracy": 0.14835357964038848,
"num_tokens": 7764692.0,
"step": 3745
},
{
"entropy": 5.811636686325073,
"epoch": 0.340166908563135,
"grad_norm": 1.5546875,
"learning_rate": 0.0004992971030983975,
"loss": 5.7867,
"mean_token_accuracy": 0.147117081284523,
"num_tokens": 7776892.0,
"step": 3750
},
{
"entropy": 5.941185331344604,
"epoch": 0.3406204644412192,
"grad_norm": 1.2734375,
"learning_rate": 0.0004992945451927531,
"loss": 5.8453,
"mean_token_accuracy": 0.1420529529452324,
"num_tokens": 7787143.0,
"step": 3755
},
{
"entropy": 5.799036169052124,
"epoch": 0.34107402031930334,
"grad_norm": 1.7734375,
"learning_rate": 0.0004992919826486402,
"loss": 5.7395,
"mean_token_accuracy": 0.14718473628163337,
"num_tokens": 7798537.0,
"step": 3760
},
{
"entropy": 5.7314859390258786,
"epoch": 0.3415275761973875,
"grad_norm": 1.59375,
"learning_rate": 0.0004992894154661118,
"loss": 5.7224,
"mean_token_accuracy": 0.15050877258181572,
"num_tokens": 7808434.0,
"step": 3765
},
{
"entropy": 5.784837865829468,
"epoch": 0.3419811320754717,
"grad_norm": 1.46875,
"learning_rate": 0.0004992868436452213,
"loss": 5.8353,
"mean_token_accuracy": 0.14429779648780822,
"num_tokens": 7819592.0,
"step": 3770
},
{
"entropy": 5.885675477981567,
"epoch": 0.34243468795355586,
"grad_norm": 1.5390625,
"learning_rate": 0.0004992842671860217,
"loss": 5.7091,
"mean_token_accuracy": 0.15716975629329683,
"num_tokens": 7830602.0,
"step": 3775
},
{
"entropy": 5.8570620059967045,
"epoch": 0.34288824383164007,
"grad_norm": 1.6640625,
"learning_rate": 0.0004992816860885662,
"loss": 5.8617,
"mean_token_accuracy": 0.14727574586868286,
"num_tokens": 7840850.0,
"step": 3780
},
{
"entropy": 5.939270067214966,
"epoch": 0.3433417997097242,
"grad_norm": 1.3359375,
"learning_rate": 0.0004992791003529084,
"loss": 5.8267,
"mean_token_accuracy": 0.1436672791838646,
"num_tokens": 7852371.0,
"step": 3785
},
{
"entropy": 5.835111427307129,
"epoch": 0.34379535558780844,
"grad_norm": 1.6484375,
"learning_rate": 0.0004992765099791017,
"loss": 5.8251,
"mean_token_accuracy": 0.14049971401691436,
"num_tokens": 7863290.0,
"step": 3790
},
{
"entropy": 5.752039957046509,
"epoch": 0.3442489114658926,
"grad_norm": 1.6328125,
"learning_rate": 0.0004992739149671995,
"loss": 5.6612,
"mean_token_accuracy": 0.15343843773007393,
"num_tokens": 7873927.0,
"step": 3795
},
{
"entropy": 5.78824667930603,
"epoch": 0.3447024673439768,
"grad_norm": 2.34375,
"learning_rate": 0.0004992713153172556,
"loss": 5.7247,
"mean_token_accuracy": 0.1494988791644573,
"num_tokens": 7884977.0,
"step": 3800
},
{
"entropy": 5.708825397491455,
"epoch": 0.34515602322206096,
"grad_norm": 1.609375,
"learning_rate": 0.0004992687110293238,
"loss": 5.7795,
"mean_token_accuracy": 0.1571035400032997,
"num_tokens": 7894620.0,
"step": 3805
},
{
"entropy": 5.795299863815307,
"epoch": 0.3456095791001451,
"grad_norm": 1.4921875,
"learning_rate": 0.0004992661021034579,
"loss": 5.721,
"mean_token_accuracy": 0.15556821525096892,
"num_tokens": 7904541.0,
"step": 3810
},
{
"entropy": 5.81929988861084,
"epoch": 0.3460631349782293,
"grad_norm": 1.3359375,
"learning_rate": 0.0004992634885397118,
"loss": 5.7389,
"mean_token_accuracy": 0.15238524973392487,
"num_tokens": 7915435.0,
"step": 3815
},
{
"entropy": 5.8006524562835695,
"epoch": 0.3465166908563135,
"grad_norm": 1.3359375,
"learning_rate": 0.0004992608703381396,
"loss": 5.722,
"mean_token_accuracy": 0.15279221534729004,
"num_tokens": 7925870.0,
"step": 3820
},
{
"entropy": 5.812300109863282,
"epoch": 0.3469702467343977,
"grad_norm": 2.078125,
"learning_rate": 0.0004992582474987955,
"loss": 5.7162,
"mean_token_accuracy": 0.1484467327594757,
"num_tokens": 7937046.0,
"step": 3825
},
{
"entropy": 5.721387004852295,
"epoch": 0.34742380261248185,
"grad_norm": 1.484375,
"learning_rate": 0.0004992556200217337,
"loss": 5.6744,
"mean_token_accuracy": 0.14899658560752868,
"num_tokens": 7947150.0,
"step": 3830
},
{
"entropy": 5.799996328353882,
"epoch": 0.34787735849056606,
"grad_norm": 1.46875,
"learning_rate": 0.0004992529879070085,
"loss": 5.7708,
"mean_token_accuracy": 0.15342669039964676,
"num_tokens": 7956815.0,
"step": 3835
},
{
"entropy": 5.85635929107666,
"epoch": 0.3483309143686502,
"grad_norm": 1.5,
"learning_rate": 0.0004992503511546745,
"loss": 5.7991,
"mean_token_accuracy": 0.1478790670633316,
"num_tokens": 7968114.0,
"step": 3840
},
{
"entropy": 5.853620481491089,
"epoch": 0.3487844702467344,
"grad_norm": 1.5390625,
"learning_rate": 0.0004992477097647859,
"loss": 5.7294,
"mean_token_accuracy": 0.14952304363250732,
"num_tokens": 7977840.0,
"step": 3845
},
{
"entropy": 5.873941707611084,
"epoch": 0.3492380261248186,
"grad_norm": 1.4921875,
"learning_rate": 0.0004992450637373976,
"loss": 5.8523,
"mean_token_accuracy": 0.15121273100376129,
"num_tokens": 7988663.0,
"step": 3850
},
{
"entropy": 5.7464821338653564,
"epoch": 0.34969158200290273,
"grad_norm": 1.8828125,
"learning_rate": 0.0004992424130725642,
"loss": 5.6591,
"mean_token_accuracy": 0.1479647561907768,
"num_tokens": 7999693.0,
"step": 3855
},
{
"entropy": 5.874104356765747,
"epoch": 0.35014513788098695,
"grad_norm": 1.6484375,
"learning_rate": 0.0004992397577703406,
"loss": 5.8291,
"mean_token_accuracy": 0.14254632443189622,
"num_tokens": 8010154.0,
"step": 3860
},
{
"entropy": 5.892876815795899,
"epoch": 0.3505986937590711,
"grad_norm": 1.8125,
"learning_rate": 0.0004992370978307816,
"loss": 5.8091,
"mean_token_accuracy": 0.14753764420747756,
"num_tokens": 8020921.0,
"step": 3865
},
{
"entropy": 5.79165678024292,
"epoch": 0.3510522496371553,
"grad_norm": 1.5234375,
"learning_rate": 0.0004992344332539424,
"loss": 5.7381,
"mean_token_accuracy": 0.15288853794336318,
"num_tokens": 8031404.0,
"step": 3870
},
{
"entropy": 5.772537279129028,
"epoch": 0.35150580551523947,
"grad_norm": 1.4921875,
"learning_rate": 0.0004992317640398779,
"loss": 5.7453,
"mean_token_accuracy": 0.15194570422172546,
"num_tokens": 8041310.0,
"step": 3875
},
{
"entropy": 5.790036058425903,
"epoch": 0.3519593613933237,
"grad_norm": 1.7265625,
"learning_rate": 0.0004992290901886434,
"loss": 5.7096,
"mean_token_accuracy": 0.15084798410534858,
"num_tokens": 8051054.0,
"step": 3880
},
{
"entropy": 5.789835023880005,
"epoch": 0.35241291727140783,
"grad_norm": 1.6328125,
"learning_rate": 0.000499226411700294,
"loss": 5.7134,
"mean_token_accuracy": 0.15393220782279968,
"num_tokens": 8061063.0,
"step": 3885
},
{
"entropy": 5.797818565368653,
"epoch": 0.35286647314949204,
"grad_norm": 1.8359375,
"learning_rate": 0.0004992237285748855,
"loss": 5.7305,
"mean_token_accuracy": 0.15079212784767151,
"num_tokens": 8071715.0,
"step": 3890
},
{
"entropy": 5.9086751461029055,
"epoch": 0.3533200290275762,
"grad_norm": 1.6875,
"learning_rate": 0.0004992210408124731,
"loss": 5.806,
"mean_token_accuracy": 0.14620591923594475,
"num_tokens": 8082679.0,
"step": 3895
},
{
"entropy": 5.804943513870239,
"epoch": 0.35377358490566035,
"grad_norm": 2.171875,
"learning_rate": 0.0004992183484131123,
"loss": 5.8262,
"mean_token_accuracy": 0.14822732135653496,
"num_tokens": 8093785.0,
"step": 3900
},
{
"entropy": 5.772074508666992,
"epoch": 0.35422714078374457,
"grad_norm": 2.015625,
"learning_rate": 0.0004992156513768591,
"loss": 5.7145,
"mean_token_accuracy": 0.15389340221881867,
"num_tokens": 8105822.0,
"step": 3905
},
{
"entropy": 5.795384550094605,
"epoch": 0.3546806966618287,
"grad_norm": 1.5234375,
"learning_rate": 0.000499212949703769,
"loss": 5.6463,
"mean_token_accuracy": 0.15382388085126877,
"num_tokens": 8115521.0,
"step": 3910
},
{
"entropy": 5.807135534286499,
"epoch": 0.35513425253991293,
"grad_norm": 1.78125,
"learning_rate": 0.0004992102433938981,
"loss": 5.7466,
"mean_token_accuracy": 0.15025942102074624,
"num_tokens": 8125524.0,
"step": 3915
},
{
"entropy": 5.7783197402954105,
"epoch": 0.3555878084179971,
"grad_norm": 1.59375,
"learning_rate": 0.0004992075324473021,
"loss": 5.6313,
"mean_token_accuracy": 0.16174003332853318,
"num_tokens": 8135552.0,
"step": 3920
},
{
"entropy": 5.7959658145904545,
"epoch": 0.3560413642960813,
"grad_norm": 1.671875,
"learning_rate": 0.0004992048168640371,
"loss": 5.7601,
"mean_token_accuracy": 0.15194501876831054,
"num_tokens": 8145617.0,
"step": 3925
},
{
"entropy": 5.798158788681031,
"epoch": 0.35649492017416545,
"grad_norm": 1.3359375,
"learning_rate": 0.0004992020966441595,
"loss": 5.7931,
"mean_token_accuracy": 0.14742157012224197,
"num_tokens": 8156729.0,
"step": 3930
},
{
"entropy": 5.819314575195312,
"epoch": 0.35694847605224966,
"grad_norm": 1.375,
"learning_rate": 0.0004991993717877254,
"loss": 5.7603,
"mean_token_accuracy": 0.14829237759113312,
"num_tokens": 8167637.0,
"step": 3935
},
{
"entropy": 5.80439453125,
"epoch": 0.3574020319303338,
"grad_norm": 1.4140625,
"learning_rate": 0.0004991966422947911,
"loss": 5.7588,
"mean_token_accuracy": 0.1592489078640938,
"num_tokens": 8177354.0,
"step": 3940
},
{
"entropy": 5.746254396438599,
"epoch": 0.357855587808418,
"grad_norm": 1.7578125,
"learning_rate": 0.0004991939081654132,
"loss": 5.6822,
"mean_token_accuracy": 0.16436415165662766,
"num_tokens": 8187580.0,
"step": 3945
},
{
"entropy": 5.760440444946289,
"epoch": 0.3583091436865022,
"grad_norm": 1.4453125,
"learning_rate": 0.000499191169399648,
"loss": 5.796,
"mean_token_accuracy": 0.14408852458000182,
"num_tokens": 8199382.0,
"step": 3950
},
{
"entropy": 5.797565078735351,
"epoch": 0.35876269956458634,
"grad_norm": 1.4765625,
"learning_rate": 0.0004991884259975525,
"loss": 5.693,
"mean_token_accuracy": 0.15424915179610252,
"num_tokens": 8208484.0,
"step": 3955
},
{
"entropy": 5.83342866897583,
"epoch": 0.35921625544267055,
"grad_norm": 1.484375,
"learning_rate": 0.000499185677959183,
"loss": 5.7418,
"mean_token_accuracy": 0.14727935642004014,
"num_tokens": 8219056.0,
"step": 3960
},
{
"entropy": 5.818744325637818,
"epoch": 0.3596698113207547,
"grad_norm": 1.421875,
"learning_rate": 0.0004991829252845966,
"loss": 5.752,
"mean_token_accuracy": 0.14865149557590485,
"num_tokens": 8230149.0,
"step": 3965
},
{
"entropy": 5.77530198097229,
"epoch": 0.3601233671988389,
"grad_norm": 1.453125,
"learning_rate": 0.0004991801679738502,
"loss": 5.827,
"mean_token_accuracy": 0.14391214549541473,
"num_tokens": 8240803.0,
"step": 3970
},
{
"entropy": 5.857344913482666,
"epoch": 0.3605769230769231,
"grad_norm": 2.03125,
"learning_rate": 0.0004991774060270008,
"loss": 5.7787,
"mean_token_accuracy": 0.1442903220653534,
"num_tokens": 8251563.0,
"step": 3975
},
{
"entropy": 5.802289152145386,
"epoch": 0.3610304789550073,
"grad_norm": 1.2734375,
"learning_rate": 0.0004991746394441055,
"loss": 5.7145,
"mean_token_accuracy": 0.1495348870754242,
"num_tokens": 8261342.0,
"step": 3980
},
{
"entropy": 5.781592130661011,
"epoch": 0.36148403483309144,
"grad_norm": 1.4765625,
"learning_rate": 0.0004991718682252216,
"loss": 5.8175,
"mean_token_accuracy": 0.14287600964307784,
"num_tokens": 8271540.0,
"step": 3985
},
{
"entropy": 5.85759482383728,
"epoch": 0.3619375907111756,
"grad_norm": 1.4765625,
"learning_rate": 0.0004991690923704063,
"loss": 5.7354,
"mean_token_accuracy": 0.15011506527662277,
"num_tokens": 8280593.0,
"step": 3990
},
{
"entropy": 5.822527742385864,
"epoch": 0.3623911465892598,
"grad_norm": 1.921875,
"learning_rate": 0.0004991663118797171,
"loss": 5.7578,
"mean_token_accuracy": 0.15425925850868225,
"num_tokens": 8290126.0,
"step": 3995
},
{
"entropy": 5.822154951095581,
"epoch": 0.36284470246734396,
"grad_norm": 1.3671875,
"learning_rate": 0.0004991635267532113,
"loss": 5.6961,
"mean_token_accuracy": 0.14956869333982467,
"num_tokens": 8301416.0,
"step": 4000
},
{
"entropy": 5.832658195495606,
"epoch": 0.36329825834542817,
"grad_norm": 1.3671875,
"learning_rate": 0.0004991607369909468,
"loss": 5.7754,
"mean_token_accuracy": 0.15036688297986983,
"num_tokens": 8311034.0,
"step": 4005
},
{
"entropy": 5.8005434513092045,
"epoch": 0.3637518142235123,
"grad_norm": 1.7109375,
"learning_rate": 0.0004991579425929811,
"loss": 5.7885,
"mean_token_accuracy": 0.14681355208158492,
"num_tokens": 8322203.0,
"step": 4010
},
{
"entropy": 5.76598744392395,
"epoch": 0.36420537010159654,
"grad_norm": 1.828125,
"learning_rate": 0.000499155143559372,
"loss": 5.6611,
"mean_token_accuracy": 0.1537575125694275,
"num_tokens": 8330981.0,
"step": 4015
},
{
"entropy": 5.722754430770874,
"epoch": 0.3646589259796807,
"grad_norm": 1.5390625,
"learning_rate": 0.0004991523398901773,
"loss": 5.7148,
"mean_token_accuracy": 0.15234556347131728,
"num_tokens": 8341294.0,
"step": 4020
},
{
"entropy": 5.931970548629761,
"epoch": 0.3651124818577649,
"grad_norm": 2.03125,
"learning_rate": 0.0004991495315854553,
"loss": 5.8497,
"mean_token_accuracy": 0.14425820037722586,
"num_tokens": 8351824.0,
"step": 4025
},
{
"entropy": 5.785876369476318,
"epoch": 0.36556603773584906,
"grad_norm": 1.9921875,
"learning_rate": 0.0004991467186452637,
"loss": 5.6981,
"mean_token_accuracy": 0.150992052257061,
"num_tokens": 8361975.0,
"step": 4030
},
{
"entropy": 5.698846769332886,
"epoch": 0.3660195936139332,
"grad_norm": 2.40625,
"learning_rate": 0.0004991439010696609,
"loss": 5.6746,
"mean_token_accuracy": 0.1555260330438614,
"num_tokens": 8372145.0,
"step": 4035
},
{
"entropy": 5.84372878074646,
"epoch": 0.3664731494920174,
"grad_norm": 2.5625,
"learning_rate": 0.0004991410788587051,
"loss": 5.7538,
"mean_token_accuracy": 0.1453930489718914,
"num_tokens": 8382542.0,
"step": 4040
},
{
"entropy": 5.848459148406983,
"epoch": 0.3669267053701016,
"grad_norm": 1.390625,
"learning_rate": 0.0004991382520124547,
"loss": 5.7821,
"mean_token_accuracy": 0.14706437662243843,
"num_tokens": 8393450.0,
"step": 4045
},
{
"entropy": 5.844983816146851,
"epoch": 0.3673802612481858,
"grad_norm": 1.4375,
"learning_rate": 0.0004991354205309681,
"loss": 5.7058,
"mean_token_accuracy": 0.14525790363550187,
"num_tokens": 8404929.0,
"step": 4050
},
{
"entropy": 5.838060188293457,
"epoch": 0.36783381712626995,
"grad_norm": 2.0625,
"learning_rate": 0.000499132584414304,
"loss": 5.7298,
"mean_token_accuracy": 0.1509381502866745,
"num_tokens": 8415703.0,
"step": 4055
},
{
"entropy": 5.8065282821655275,
"epoch": 0.36828737300435416,
"grad_norm": 1.8046875,
"learning_rate": 0.0004991297436625207,
"loss": 5.8075,
"mean_token_accuracy": 0.15024188235402108,
"num_tokens": 8426220.0,
"step": 4060
},
{
"entropy": 5.717031574249267,
"epoch": 0.3687409288824383,
"grad_norm": 1.6015625,
"learning_rate": 0.0004991268982756774,
"loss": 5.6718,
"mean_token_accuracy": 0.15563272684812546,
"num_tokens": 8436220.0,
"step": 4065
},
{
"entropy": 5.872826862335205,
"epoch": 0.3691944847605225,
"grad_norm": 1.6328125,
"learning_rate": 0.0004991240482538327,
"loss": 5.8326,
"mean_token_accuracy": 0.14214779883623124,
"num_tokens": 8446724.0,
"step": 4070
},
{
"entropy": 5.8477081775665285,
"epoch": 0.3696480406386067,
"grad_norm": 1.8203125,
"learning_rate": 0.0004991211935970455,
"loss": 5.6876,
"mean_token_accuracy": 0.14904116541147233,
"num_tokens": 8457270.0,
"step": 4075
},
{
"entropy": 5.828044128417969,
"epoch": 0.37010159651669083,
"grad_norm": 1.734375,
"learning_rate": 0.0004991183343053748,
"loss": 5.7618,
"mean_token_accuracy": 0.14451126977801323,
"num_tokens": 8468137.0,
"step": 4080
},
{
"entropy": 5.8057512760162355,
"epoch": 0.37055515239477504,
"grad_norm": 1.6171875,
"learning_rate": 0.00049911547037888,
"loss": 5.6947,
"mean_token_accuracy": 0.15399888902902603,
"num_tokens": 8477574.0,
"step": 4085
},
{
"entropy": 5.769902849197388,
"epoch": 0.3710087082728592,
"grad_norm": 1.390625,
"learning_rate": 0.0004991126018176202,
"loss": 5.7932,
"mean_token_accuracy": 0.1436054840683937,
"num_tokens": 8487268.0,
"step": 4090
},
{
"entropy": 5.928350782394409,
"epoch": 0.3714622641509434,
"grad_norm": 2.0,
"learning_rate": 0.0004991097286216545,
"loss": 5.8514,
"mean_token_accuracy": 0.14995737373828888,
"num_tokens": 8497785.0,
"step": 4095
},
{
"entropy": 5.887904167175293,
"epoch": 0.37191582002902757,
"grad_norm": 2.046875,
"learning_rate": 0.0004991068507910427,
"loss": 5.7223,
"mean_token_accuracy": 0.14823060035705565,
"num_tokens": 8508169.0,
"step": 4100
},
{
"entropy": 5.738240814208984,
"epoch": 0.3723693759071118,
"grad_norm": 2.078125,
"learning_rate": 0.000499103968325844,
"loss": 5.7058,
"mean_token_accuracy": 0.14781549870967864,
"num_tokens": 8519201.0,
"step": 4105
},
{
"entropy": 5.888035488128662,
"epoch": 0.37282293178519593,
"grad_norm": 1.90625,
"learning_rate": 0.0004991010812261181,
"loss": 5.8075,
"mean_token_accuracy": 0.14320318549871444,
"num_tokens": 8531093.0,
"step": 4110
},
{
"entropy": 5.799619388580322,
"epoch": 0.37327648766328014,
"grad_norm": 1.8125,
"learning_rate": 0.0004990981894919248,
"loss": 5.6616,
"mean_token_accuracy": 0.1559358224272728,
"num_tokens": 8541730.0,
"step": 4115
},
{
"entropy": 5.728639888763428,
"epoch": 0.3737300435413643,
"grad_norm": 1.9375,
"learning_rate": 0.0004990952931233238,
"loss": 5.7516,
"mean_token_accuracy": 0.1526630237698555,
"num_tokens": 8552570.0,
"step": 4120
},
{
"entropy": 5.801388740539551,
"epoch": 0.37418359941944845,
"grad_norm": 1.90625,
"learning_rate": 0.000499092392120375,
"loss": 5.6416,
"mean_token_accuracy": 0.15281095802783967,
"num_tokens": 8562340.0,
"step": 4125
},
{
"entropy": 5.86045470237732,
"epoch": 0.37463715529753266,
"grad_norm": 1.7734375,
"learning_rate": 0.0004990894864831385,
"loss": 5.8447,
"mean_token_accuracy": 0.14465546905994414,
"num_tokens": 8573240.0,
"step": 4130
},
{
"entropy": 5.83771858215332,
"epoch": 0.3750907111756168,
"grad_norm": 1.921875,
"learning_rate": 0.0004990865762116743,
"loss": 5.7,
"mean_token_accuracy": 0.1569647617638111,
"num_tokens": 8583194.0,
"step": 4135
},
{
"entropy": 5.840100049972534,
"epoch": 0.37554426705370103,
"grad_norm": 2.546875,
"learning_rate": 0.0004990836613060426,
"loss": 5.7629,
"mean_token_accuracy": 0.1451824389398098,
"num_tokens": 8593603.0,
"step": 4140
},
{
"entropy": 5.742400503158569,
"epoch": 0.3759978229317852,
"grad_norm": 1.9609375,
"learning_rate": 0.0004990807417663036,
"loss": 5.7677,
"mean_token_accuracy": 0.15051018446683884,
"num_tokens": 8603176.0,
"step": 4145
},
{
"entropy": 5.8087677478790285,
"epoch": 0.3764513788098694,
"grad_norm": 3.90625,
"learning_rate": 0.0004990778175925179,
"loss": 5.7931,
"mean_token_accuracy": 0.15137793719768525,
"num_tokens": 8614801.0,
"step": 4150
},
{
"entropy": 5.911026477813721,
"epoch": 0.37690493468795355,
"grad_norm": 1.765625,
"learning_rate": 0.0004990748887847457,
"loss": 5.8055,
"mean_token_accuracy": 0.14926701039075851,
"num_tokens": 8624999.0,
"step": 4155
},
{
"entropy": 5.786199283599854,
"epoch": 0.37735849056603776,
"grad_norm": 1.828125,
"learning_rate": 0.0004990719553430477,
"loss": 5.6852,
"mean_token_accuracy": 0.16073830276727677,
"num_tokens": 8635358.0,
"step": 4160
},
{
"entropy": 5.812259101867676,
"epoch": 0.3778120464441219,
"grad_norm": 1.515625,
"learning_rate": 0.0004990690172674847,
"loss": 5.7376,
"mean_token_accuracy": 0.15018846094608307,
"num_tokens": 8645430.0,
"step": 4165
},
{
"entropy": 5.876455307006836,
"epoch": 0.3782656023222061,
"grad_norm": 1.71875,
"learning_rate": 0.0004990660745581172,
"loss": 5.6993,
"mean_token_accuracy": 0.15587360560894012,
"num_tokens": 8655115.0,
"step": 4170
},
{
"entropy": 5.731990957260132,
"epoch": 0.3787191582002903,
"grad_norm": 1.8828125,
"learning_rate": 0.0004990631272150062,
"loss": 5.6902,
"mean_token_accuracy": 0.14820799827575684,
"num_tokens": 8665391.0,
"step": 4175
},
{
"entropy": 5.804074239730835,
"epoch": 0.37917271407837444,
"grad_norm": 1.5,
"learning_rate": 0.0004990601752382127,
"loss": 5.7963,
"mean_token_accuracy": 0.1466014303267002,
"num_tokens": 8676819.0,
"step": 4180
},
{
"entropy": 5.87374176979065,
"epoch": 0.37962626995645865,
"grad_norm": 1.671875,
"learning_rate": 0.0004990572186277975,
"loss": 5.7317,
"mean_token_accuracy": 0.15189690440893172,
"num_tokens": 8687052.0,
"step": 4185
},
{
"entropy": 5.8404378414154055,
"epoch": 0.3800798258345428,
"grad_norm": 1.6875,
"learning_rate": 0.0004990542573838221,
"loss": 5.7927,
"mean_token_accuracy": 0.14445817843079567,
"num_tokens": 8697756.0,
"step": 4190
},
{
"entropy": 5.761879825592041,
"epoch": 0.380533381712627,
"grad_norm": 1.75,
"learning_rate": 0.0004990512915063475,
"loss": 5.6999,
"mean_token_accuracy": 0.15357010215520858,
"num_tokens": 8708388.0,
"step": 4195
},
{
"entropy": 5.891747236251831,
"epoch": 0.38098693759071117,
"grad_norm": 2.421875,
"learning_rate": 0.0004990483209954352,
"loss": 5.8001,
"mean_token_accuracy": 0.14270555898547171,
"num_tokens": 8719475.0,
"step": 4200
},
{
"entropy": 5.826990032196045,
"epoch": 0.3814404934687954,
"grad_norm": 1.7890625,
"learning_rate": 0.0004990453458511464,
"loss": 5.6709,
"mean_token_accuracy": 0.15746982842683793,
"num_tokens": 8729395.0,
"step": 4205
},
{
"entropy": 5.862640428543091,
"epoch": 0.38189404934687954,
"grad_norm": 2.375,
"learning_rate": 0.0004990423660735428,
"loss": 5.7227,
"mean_token_accuracy": 0.15197302550077438,
"num_tokens": 8740351.0,
"step": 4210
},
{
"entropy": 5.838958644866944,
"epoch": 0.3823476052249637,
"grad_norm": 2.5625,
"learning_rate": 0.000499039381662686,
"loss": 5.7375,
"mean_token_accuracy": 0.15115175843238832,
"num_tokens": 8749260.0,
"step": 4215
},
{
"entropy": 5.813646125793457,
"epoch": 0.3828011611030479,
"grad_norm": 1.59375,
"learning_rate": 0.0004990363926186377,
"loss": 5.8251,
"mean_token_accuracy": 0.14027270674705505,
"num_tokens": 8759604.0,
"step": 4220
},
{
"entropy": 5.858145141601563,
"epoch": 0.38325471698113206,
"grad_norm": 1.9296875,
"learning_rate": 0.0004990333989414597,
"loss": 5.826,
"mean_token_accuracy": 0.14018016085028648,
"num_tokens": 8769815.0,
"step": 4225
},
{
"entropy": 5.801722669601441,
"epoch": 0.38370827285921627,
"grad_norm": 1.640625,
"learning_rate": 0.0004990304006312138,
"loss": 5.77,
"mean_token_accuracy": 0.1485055848956108,
"num_tokens": 8780557.0,
"step": 4230
},
{
"entropy": 5.830351734161377,
"epoch": 0.3841618287373004,
"grad_norm": 1.9609375,
"learning_rate": 0.0004990273976879622,
"loss": 5.7202,
"mean_token_accuracy": 0.15922572910785676,
"num_tokens": 8790993.0,
"step": 4235
},
{
"entropy": 5.873066186904907,
"epoch": 0.38461538461538464,
"grad_norm": 1.6484375,
"learning_rate": 0.0004990243901117669,
"loss": 5.7568,
"mean_token_accuracy": 0.14900755137205124,
"num_tokens": 8800657.0,
"step": 4240
},
{
"entropy": 5.7463071823120115,
"epoch": 0.3850689404934688,
"grad_norm": 2.640625,
"learning_rate": 0.0004990213779026903,
"loss": 5.7749,
"mean_token_accuracy": 0.14669715017080306,
"num_tokens": 8812983.0,
"step": 4245
},
{
"entropy": 5.889048004150391,
"epoch": 0.385522496371553,
"grad_norm": 3.34375,
"learning_rate": 0.0004990183610607944,
"loss": 5.7358,
"mean_token_accuracy": 0.150221536308527,
"num_tokens": 8822980.0,
"step": 4250
},
{
"entropy": 5.827584266662598,
"epoch": 0.38597605224963716,
"grad_norm": 1.6015625,
"learning_rate": 0.0004990153395861415,
"loss": 5.7337,
"mean_token_accuracy": 0.15071167424321175,
"num_tokens": 8833821.0,
"step": 4255
},
{
"entropy": 5.789939260482788,
"epoch": 0.3864296081277213,
"grad_norm": 1.796875,
"learning_rate": 0.0004990123134787946,
"loss": 5.6854,
"mean_token_accuracy": 0.15111219882965088,
"num_tokens": 8843929.0,
"step": 4260
},
{
"entropy": 5.774974298477173,
"epoch": 0.3868831640058055,
"grad_norm": 1.609375,
"learning_rate": 0.0004990092827388159,
"loss": 5.6597,
"mean_token_accuracy": 0.15415194928646087,
"num_tokens": 8854180.0,
"step": 4265
},
{
"entropy": 5.727933740615844,
"epoch": 0.3873367198838897,
"grad_norm": 2.8125,
"learning_rate": 0.0004990062473662681,
"loss": 5.6851,
"mean_token_accuracy": 0.15864559412002563,
"num_tokens": 8865018.0,
"step": 4270
},
{
"entropy": 5.857227516174317,
"epoch": 0.3877902757619739,
"grad_norm": 2.15625,
"learning_rate": 0.000499003207361214,
"loss": 5.864,
"mean_token_accuracy": 0.14215147495269775,
"num_tokens": 8875423.0,
"step": 4275
},
{
"entropy": 5.942317485809326,
"epoch": 0.38824383164005805,
"grad_norm": 1.6484375,
"learning_rate": 0.0004990001627237165,
"loss": 5.8048,
"mean_token_accuracy": 0.14681227505207062,
"num_tokens": 8887850.0,
"step": 4280
},
{
"entropy": 5.756715774536133,
"epoch": 0.38869738751814226,
"grad_norm": 1.9453125,
"learning_rate": 0.0004989971134538386,
"loss": 5.6856,
"mean_token_accuracy": 0.15011156648397445,
"num_tokens": 8897429.0,
"step": 4285
},
{
"entropy": 5.8178752899169925,
"epoch": 0.3891509433962264,
"grad_norm": 1.6015625,
"learning_rate": 0.0004989940595516432,
"loss": 5.7018,
"mean_token_accuracy": 0.15288467705249786,
"num_tokens": 8907048.0,
"step": 4290
},
{
"entropy": 5.800828266143799,
"epoch": 0.3896044992743106,
"grad_norm": 1.5703125,
"learning_rate": 0.0004989910010171938,
"loss": 5.7545,
"mean_token_accuracy": 0.1477030858397484,
"num_tokens": 8917987.0,
"step": 4295
},
{
"entropy": 5.897664308547974,
"epoch": 0.3900580551523948,
"grad_norm": 1.6640625,
"learning_rate": 0.0004989879378505532,
"loss": 5.6798,
"mean_token_accuracy": 0.14783306419849396,
"num_tokens": 8929339.0,
"step": 4300
},
{
"entropy": 5.746922588348388,
"epoch": 0.39051161103047893,
"grad_norm": 1.9296875,
"learning_rate": 0.0004989848700517849,
"loss": 5.6842,
"mean_token_accuracy": 0.14888181313872337,
"num_tokens": 8939473.0,
"step": 4305
},
{
"entropy": 5.758429908752442,
"epoch": 0.39096516690856314,
"grad_norm": 2.046875,
"learning_rate": 0.0004989817976209526,
"loss": 5.7207,
"mean_token_accuracy": 0.15242671966552734,
"num_tokens": 8949173.0,
"step": 4310
},
{
"entropy": 5.7987030982971195,
"epoch": 0.3914187227866473,
"grad_norm": 1.921875,
"learning_rate": 0.0004989787205581196,
"loss": 5.6937,
"mean_token_accuracy": 0.15537856966257096,
"num_tokens": 8959879.0,
"step": 4315
},
{
"entropy": 5.801929140090943,
"epoch": 0.3918722786647315,
"grad_norm": 3.265625,
"learning_rate": 0.0004989756388633496,
"loss": 5.7669,
"mean_token_accuracy": 0.1492094039916992,
"num_tokens": 8970643.0,
"step": 4320
},
{
"entropy": 5.826022291183472,
"epoch": 0.39232583454281567,
"grad_norm": 2.8125,
"learning_rate": 0.0004989725525367063,
"loss": 5.6519,
"mean_token_accuracy": 0.15020777881145478,
"num_tokens": 8979882.0,
"step": 4325
},
{
"entropy": 5.819568634033203,
"epoch": 0.3927793904208999,
"grad_norm": 1.75,
"learning_rate": 0.0004989694615782534,
"loss": 5.792,
"mean_token_accuracy": 0.14471497684717177,
"num_tokens": 8990925.0,
"step": 4330
},
{
"entropy": 5.835562753677368,
"epoch": 0.39323294629898403,
"grad_norm": 1.53125,
"learning_rate": 0.0004989663659880552,
"loss": 5.8056,
"mean_token_accuracy": 0.14450787603855134,
"num_tokens": 9001365.0,
"step": 4335
},
{
"entropy": 5.854931926727295,
"epoch": 0.3936865021770682,
"grad_norm": 1.6640625,
"learning_rate": 0.0004989632657661754,
"loss": 5.8011,
"mean_token_accuracy": 0.15069715231657027,
"num_tokens": 9011942.0,
"step": 4340
},
{
"entropy": 5.9251405715942385,
"epoch": 0.3941400580551524,
"grad_norm": 1.6328125,
"learning_rate": 0.000498960160912678,
"loss": 5.8149,
"mean_token_accuracy": 0.14857330471277236,
"num_tokens": 9022610.0,
"step": 4345
},
{
"entropy": 5.802004814147949,
"epoch": 0.39459361393323655,
"grad_norm": 1.7421875,
"learning_rate": 0.0004989570514276276,
"loss": 5.7119,
"mean_token_accuracy": 0.15157701522111894,
"num_tokens": 9033034.0,
"step": 4350
},
{
"entropy": 5.798203468322754,
"epoch": 0.39504716981132076,
"grad_norm": 1.4140625,
"learning_rate": 0.0004989539373110883,
"loss": 5.7609,
"mean_token_accuracy": 0.14471814632415772,
"num_tokens": 9043418.0,
"step": 4355
},
{
"entropy": 5.870092725753784,
"epoch": 0.3955007256894049,
"grad_norm": 1.875,
"learning_rate": 0.0004989508185631245,
"loss": 5.8328,
"mean_token_accuracy": 0.14952152520418166,
"num_tokens": 9052708.0,
"step": 4360
},
{
"entropy": 5.786429405212402,
"epoch": 0.39595428156748913,
"grad_norm": 1.765625,
"learning_rate": 0.0004989476951838006,
"loss": 5.7219,
"mean_token_accuracy": 0.14567517340183259,
"num_tokens": 9062883.0,
"step": 4365
},
{
"entropy": 5.7873271942138675,
"epoch": 0.3964078374455733,
"grad_norm": 1.578125,
"learning_rate": 0.0004989445671731814,
"loss": 5.6701,
"mean_token_accuracy": 0.1534933939576149,
"num_tokens": 9072830.0,
"step": 4370
},
{
"entropy": 5.761323547363281,
"epoch": 0.3968613933236575,
"grad_norm": 1.7109375,
"learning_rate": 0.0004989414345313315,
"loss": 5.6374,
"mean_token_accuracy": 0.14723165333271027,
"num_tokens": 9082749.0,
"step": 4375
},
{
"entropy": 5.85912070274353,
"epoch": 0.39731494920174165,
"grad_norm": 1.59375,
"learning_rate": 0.0004989382972583156,
"loss": 5.7277,
"mean_token_accuracy": 0.14558447450399398,
"num_tokens": 9094188.0,
"step": 4380
},
{
"entropy": 5.806872129440308,
"epoch": 0.3977685050798258,
"grad_norm": 1.703125,
"learning_rate": 0.0004989351553541986,
"loss": 5.7496,
"mean_token_accuracy": 0.14632164612412452,
"num_tokens": 9105023.0,
"step": 4385
},
{
"entropy": 5.782722663879395,
"epoch": 0.39822206095791,
"grad_norm": 2.0,
"learning_rate": 0.0004989320088190456,
"loss": 5.6934,
"mean_token_accuracy": 0.1507030040025711,
"num_tokens": 9115563.0,
"step": 4390
},
{
"entropy": 5.769656801223755,
"epoch": 0.3986756168359942,
"grad_norm": 3.46875,
"learning_rate": 0.0004989288576529216,
"loss": 5.7101,
"mean_token_accuracy": 0.14767857640981674,
"num_tokens": 9125394.0,
"step": 4395
},
{
"entropy": 5.779007911682129,
"epoch": 0.3991291727140784,
"grad_norm": 1.5546875,
"learning_rate": 0.0004989257018558917,
"loss": 5.7158,
"mean_token_accuracy": 0.1508276343345642,
"num_tokens": 9137061.0,
"step": 4400
},
{
"entropy": 5.856002616882324,
"epoch": 0.39958272859216254,
"grad_norm": 2.03125,
"learning_rate": 0.0004989225414280213,
"loss": 5.6813,
"mean_token_accuracy": 0.1500905178487301,
"num_tokens": 9148196.0,
"step": 4405
},
{
"entropy": 5.78254542350769,
"epoch": 0.40003628447024675,
"grad_norm": 1.875,
"learning_rate": 0.0004989193763693756,
"loss": 5.7107,
"mean_token_accuracy": 0.15331541001796722,
"num_tokens": 9158013.0,
"step": 4410
},
{
"entropy": 5.738414621353149,
"epoch": 0.4004898403483309,
"grad_norm": 1.890625,
"learning_rate": 0.0004989162066800202,
"loss": 5.6106,
"mean_token_accuracy": 0.15551046580076217,
"num_tokens": 9167813.0,
"step": 4415
},
{
"entropy": 5.764182472229004,
"epoch": 0.4009433962264151,
"grad_norm": 2.453125,
"learning_rate": 0.0004989130323600205,
"loss": 5.6889,
"mean_token_accuracy": 0.1522517278790474,
"num_tokens": 9177783.0,
"step": 4420
},
{
"entropy": 5.800490760803223,
"epoch": 0.40139695210449927,
"grad_norm": 1.625,
"learning_rate": 0.0004989098534094424,
"loss": 5.7324,
"mean_token_accuracy": 0.15464526265859604,
"num_tokens": 9187556.0,
"step": 4425
},
{
"entropy": 5.805605030059814,
"epoch": 0.4018505079825834,
"grad_norm": 1.8125,
"learning_rate": 0.0004989066698283512,
"loss": 5.7733,
"mean_token_accuracy": 0.14691560864448547,
"num_tokens": 9198659.0,
"step": 4430
},
{
"entropy": 5.811403799057007,
"epoch": 0.40230406386066764,
"grad_norm": 2.109375,
"learning_rate": 0.0004989034816168132,
"loss": 5.758,
"mean_token_accuracy": 0.1504255309700966,
"num_tokens": 9209546.0,
"step": 4435
},
{
"entropy": 5.78139705657959,
"epoch": 0.4027576197387518,
"grad_norm": 2.828125,
"learning_rate": 0.0004989002887748941,
"loss": 5.6752,
"mean_token_accuracy": 0.15406596213579177,
"num_tokens": 9219339.0,
"step": 4440
},
{
"entropy": 5.84251184463501,
"epoch": 0.403211175616836,
"grad_norm": 1.8203125,
"learning_rate": 0.0004988970913026601,
"loss": 5.8354,
"mean_token_accuracy": 0.14621326699852943,
"num_tokens": 9230364.0,
"step": 4445
},
{
"entropy": 5.873086214065552,
"epoch": 0.40366473149492016,
"grad_norm": 2.21875,
"learning_rate": 0.0004988938892001771,
"loss": 5.7136,
"mean_token_accuracy": 0.1498823121190071,
"num_tokens": 9240732.0,
"step": 4450
},
{
"entropy": 5.731912326812744,
"epoch": 0.40411828737300437,
"grad_norm": 1.7421875,
"learning_rate": 0.0004988906824675113,
"loss": 5.6363,
"mean_token_accuracy": 0.15650977939367294,
"num_tokens": 9251049.0,
"step": 4455
},
{
"entropy": 5.779788398742676,
"epoch": 0.4045718432510885,
"grad_norm": 2.703125,
"learning_rate": 0.0004988874711047293,
"loss": 5.6905,
"mean_token_accuracy": 0.1514260247349739,
"num_tokens": 9261660.0,
"step": 4460
},
{
"entropy": 5.810968637466431,
"epoch": 0.40502539912917274,
"grad_norm": 1.7109375,
"learning_rate": 0.0004988842551118972,
"loss": 5.7739,
"mean_token_accuracy": 0.14920282363891602,
"num_tokens": 9272114.0,
"step": 4465
},
{
"entropy": 5.797091484069824,
"epoch": 0.4054789550072569,
"grad_norm": 1.734375,
"learning_rate": 0.0004988810344890818,
"loss": 5.7297,
"mean_token_accuracy": 0.15285758376121522,
"num_tokens": 9282125.0,
"step": 4470
},
{
"entropy": 5.8813393115997314,
"epoch": 0.40593251088534105,
"grad_norm": 2.3125,
"learning_rate": 0.0004988778092363494,
"loss": 5.7761,
"mean_token_accuracy": 0.1445347175002098,
"num_tokens": 9294078.0,
"step": 4475
},
{
"entropy": 5.831450319290161,
"epoch": 0.40638606676342526,
"grad_norm": 1.8671875,
"learning_rate": 0.0004988745793537671,
"loss": 5.6933,
"mean_token_accuracy": 0.14883248656988143,
"num_tokens": 9304026.0,
"step": 4480
},
{
"entropy": 5.85241436958313,
"epoch": 0.4068396226415094,
"grad_norm": 2.359375,
"learning_rate": 0.0004988713448414011,
"loss": 5.7471,
"mean_token_accuracy": 0.15584234297275543,
"num_tokens": 9313933.0,
"step": 4485
},
{
"entropy": 5.8357024669647215,
"epoch": 0.4072931785195936,
"grad_norm": 2.453125,
"learning_rate": 0.000498868105699319,
"loss": 5.7767,
"mean_token_accuracy": 0.14843285381793975,
"num_tokens": 9323932.0,
"step": 4490
},
{
"entropy": 5.723560190200805,
"epoch": 0.4077467343976778,
"grad_norm": 1.71875,
"learning_rate": 0.0004988648619275872,
"loss": 5.6851,
"mean_token_accuracy": 0.1634763866662979,
"num_tokens": 9333467.0,
"step": 4495
},
{
"entropy": 5.817450141906738,
"epoch": 0.408200290275762,
"grad_norm": 1.4375,
"learning_rate": 0.0004988616135262729,
"loss": 5.7577,
"mean_token_accuracy": 0.14326895773410797,
"num_tokens": 9344909.0,
"step": 4500
},
{
"entropy": 5.897395086288452,
"epoch": 0.40865384615384615,
"grad_norm": 1.734375,
"learning_rate": 0.0004988583604954436,
"loss": 5.7977,
"mean_token_accuracy": 0.14532231390476227,
"num_tokens": 9355630.0,
"step": 4505
},
{
"entropy": 5.8017322540283205,
"epoch": 0.40910740203193036,
"grad_norm": 1.6328125,
"learning_rate": 0.0004988551028351663,
"loss": 5.7392,
"mean_token_accuracy": 0.147707736492157,
"num_tokens": 9365700.0,
"step": 4510
},
{
"entropy": 5.831019258499145,
"epoch": 0.4095609579100145,
"grad_norm": 1.6171875,
"learning_rate": 0.0004988518405455084,
"loss": 5.8225,
"mean_token_accuracy": 0.14114691615104674,
"num_tokens": 9377088.0,
"step": 4515
},
{
"entropy": 5.8371169090271,
"epoch": 0.41001451378809867,
"grad_norm": 1.9453125,
"learning_rate": 0.0004988485736265374,
"loss": 5.6158,
"mean_token_accuracy": 0.15830224454402925,
"num_tokens": 9386855.0,
"step": 4520
},
{
"entropy": 5.791143941879272,
"epoch": 0.4104680696661829,
"grad_norm": 1.7890625,
"learning_rate": 0.000498845302078321,
"loss": 5.7444,
"mean_token_accuracy": 0.14412623643875122,
"num_tokens": 9398661.0,
"step": 4525
},
{
"entropy": 5.826167106628418,
"epoch": 0.41092162554426703,
"grad_norm": 1.5703125,
"learning_rate": 0.0004988420259009266,
"loss": 5.7428,
"mean_token_accuracy": 0.1512528121471405,
"num_tokens": 9408670.0,
"step": 4530
},
{
"entropy": 5.844571304321289,
"epoch": 0.41137518142235124,
"grad_norm": 2.65625,
"learning_rate": 0.000498838745094422,
"loss": 5.7423,
"mean_token_accuracy": 0.15003526955842972,
"num_tokens": 9418740.0,
"step": 4535
},
{
"entropy": 5.7954200267791744,
"epoch": 0.4118287373004354,
"grad_norm": 1.59375,
"learning_rate": 0.0004988354596588751,
"loss": 5.7326,
"mean_token_accuracy": 0.15099671930074693,
"num_tokens": 9429339.0,
"step": 4540
},
{
"entropy": 5.78496036529541,
"epoch": 0.4122822931785196,
"grad_norm": 2.78125,
"learning_rate": 0.0004988321695943539,
"loss": 5.6709,
"mean_token_accuracy": 0.15879771262407302,
"num_tokens": 9438996.0,
"step": 4545
},
{
"entropy": 5.82200813293457,
"epoch": 0.41273584905660377,
"grad_norm": 2.421875,
"learning_rate": 0.0004988288749009264,
"loss": 5.6959,
"mean_token_accuracy": 0.15078579634428024,
"num_tokens": 9448669.0,
"step": 4550
},
{
"entropy": 5.776082324981689,
"epoch": 0.413189404934688,
"grad_norm": 1.9375,
"learning_rate": 0.0004988255755786608,
"loss": 5.6834,
"mean_token_accuracy": 0.15365493297576904,
"num_tokens": 9458857.0,
"step": 4555
},
{
"entropy": 5.7420238018035885,
"epoch": 0.41364296081277213,
"grad_norm": 2.3125,
"learning_rate": 0.0004988222716276251,
"loss": 5.6845,
"mean_token_accuracy": 0.15307349115610122,
"num_tokens": 9470241.0,
"step": 4560
},
{
"entropy": 5.889754772186279,
"epoch": 0.4140965166908563,
"grad_norm": 2.03125,
"learning_rate": 0.000498818963047888,
"loss": 5.8581,
"mean_token_accuracy": 0.1453462913632393,
"num_tokens": 9480528.0,
"step": 4565
},
{
"entropy": 5.914613819122314,
"epoch": 0.4145500725689405,
"grad_norm": 1.765625,
"learning_rate": 0.0004988156498395176,
"loss": 5.7684,
"mean_token_accuracy": 0.14872041046619416,
"num_tokens": 9491000.0,
"step": 4570
},
{
"entropy": 5.829993867874146,
"epoch": 0.41500362844702465,
"grad_norm": 2.125,
"learning_rate": 0.0004988123320025825,
"loss": 5.756,
"mean_token_accuracy": 0.15005763098597527,
"num_tokens": 9501612.0,
"step": 4575
},
{
"entropy": 5.857288932800293,
"epoch": 0.41545718432510886,
"grad_norm": 1.9765625,
"learning_rate": 0.0004988090095371514,
"loss": 5.8461,
"mean_token_accuracy": 0.145644311606884,
"num_tokens": 9513999.0,
"step": 4580
},
{
"entropy": 5.859341239929199,
"epoch": 0.415910740203193,
"grad_norm": 1.875,
"learning_rate": 0.0004988056824432928,
"loss": 5.7257,
"mean_token_accuracy": 0.1431219108402729,
"num_tokens": 9524392.0,
"step": 4585
},
{
"entropy": 5.816585493087769,
"epoch": 0.41636429608127723,
"grad_norm": 2.265625,
"learning_rate": 0.0004988023507210758,
"loss": 5.7304,
"mean_token_accuracy": 0.15149542987346648,
"num_tokens": 9534227.0,
"step": 4590
},
{
"entropy": 5.802891588211059,
"epoch": 0.4168178519593614,
"grad_norm": 1.640625,
"learning_rate": 0.0004987990143705692,
"loss": 5.7088,
"mean_token_accuracy": 0.155563186109066,
"num_tokens": 9544242.0,
"step": 4595
},
{
"entropy": 5.840281343460083,
"epoch": 0.4172714078374456,
"grad_norm": 1.6484375,
"learning_rate": 0.0004987956733918418,
"loss": 5.6398,
"mean_token_accuracy": 0.1510664239525795,
"num_tokens": 9553921.0,
"step": 4600
},
{
"entropy": 5.723497200012207,
"epoch": 0.41772496371552975,
"grad_norm": 2.078125,
"learning_rate": 0.0004987923277849629,
"loss": 5.6785,
"mean_token_accuracy": 0.14915904104709626,
"num_tokens": 9564768.0,
"step": 4605
},
{
"entropy": 5.789427089691162,
"epoch": 0.4181785195936139,
"grad_norm": 2.296875,
"learning_rate": 0.0004987889775500017,
"loss": 5.7736,
"mean_token_accuracy": 0.14345343932509422,
"num_tokens": 9576360.0,
"step": 4610
},
{
"entropy": 5.862417459487915,
"epoch": 0.4186320754716981,
"grad_norm": 2.421875,
"learning_rate": 0.0004987856226870272,
"loss": 5.7484,
"mean_token_accuracy": 0.1491814523935318,
"num_tokens": 9587623.0,
"step": 4615
},
{
"entropy": 5.795092535018921,
"epoch": 0.4190856313497823,
"grad_norm": 2.15625,
"learning_rate": 0.0004987822631961092,
"loss": 5.7795,
"mean_token_accuracy": 0.1469440668821335,
"num_tokens": 9597837.0,
"step": 4620
},
{
"entropy": 5.825919055938721,
"epoch": 0.4195391872278665,
"grad_norm": 1.5546875,
"learning_rate": 0.000498778899077317,
"loss": 5.7333,
"mean_token_accuracy": 0.14518298730254173,
"num_tokens": 9609509.0,
"step": 4625
},
{
"entropy": 5.799678754806519,
"epoch": 0.41999274310595064,
"grad_norm": 1.6484375,
"learning_rate": 0.00049877553033072,
"loss": 5.703,
"mean_token_accuracy": 0.15073038637638092,
"num_tokens": 9619954.0,
"step": 4630
},
{
"entropy": 5.785829067230225,
"epoch": 0.42044629898403485,
"grad_norm": 1.5625,
"learning_rate": 0.0004987721569563882,
"loss": 5.7136,
"mean_token_accuracy": 0.14777721464633942,
"num_tokens": 9629742.0,
"step": 4635
},
{
"entropy": 5.8123650550842285,
"epoch": 0.420899854862119,
"grad_norm": 2.21875,
"learning_rate": 0.000498768778954391,
"loss": 5.7296,
"mean_token_accuracy": 0.15000276267528534,
"num_tokens": 9641378.0,
"step": 4640
},
{
"entropy": 5.854241609573364,
"epoch": 0.4213534107402032,
"grad_norm": 1.6015625,
"learning_rate": 0.0004987653963247985,
"loss": 5.7931,
"mean_token_accuracy": 0.15243044048547744,
"num_tokens": 9652279.0,
"step": 4645
},
{
"entropy": 5.831078624725341,
"epoch": 0.42180696661828737,
"grad_norm": 1.9609375,
"learning_rate": 0.0004987620090676805,
"loss": 5.8085,
"mean_token_accuracy": 0.14800476655364037,
"num_tokens": 9663416.0,
"step": 4650
},
{
"entropy": 5.810733652114868,
"epoch": 0.4222605224963715,
"grad_norm": 2.0625,
"learning_rate": 0.0004987586171831072,
"loss": 5.7501,
"mean_token_accuracy": 0.14851362258195877,
"num_tokens": 9674647.0,
"step": 4655
},
{
"entropy": 5.792376756668091,
"epoch": 0.42271407837445574,
"grad_norm": 2.1875,
"learning_rate": 0.0004987552206711487,
"loss": 5.6,
"mean_token_accuracy": 0.15128281936049462,
"num_tokens": 9685084.0,
"step": 4660
},
{
"entropy": 5.711585760116577,
"epoch": 0.4231676342525399,
"grad_norm": 1.9453125,
"learning_rate": 0.0004987518195318752,
"loss": 5.6618,
"mean_token_accuracy": 0.1485517218708992,
"num_tokens": 9695661.0,
"step": 4665
},
{
"entropy": 5.7958086967468265,
"epoch": 0.4236211901306241,
"grad_norm": 1.875,
"learning_rate": 0.000498748413765357,
"loss": 5.7194,
"mean_token_accuracy": 0.14617605209350587,
"num_tokens": 9705773.0,
"step": 4670
},
{
"entropy": 5.870110130310058,
"epoch": 0.42407474600870826,
"grad_norm": 1.578125,
"learning_rate": 0.0004987450033716646,
"loss": 5.7716,
"mean_token_accuracy": 0.1500180780887604,
"num_tokens": 9715799.0,
"step": 4675
},
{
"entropy": 5.825811100006104,
"epoch": 0.42452830188679247,
"grad_norm": 1.890625,
"learning_rate": 0.0004987415883508686,
"loss": 5.7232,
"mean_token_accuracy": 0.1532239183783531,
"num_tokens": 9727012.0,
"step": 4680
},
{
"entropy": 5.829875516891479,
"epoch": 0.4249818577648766,
"grad_norm": 2.40625,
"learning_rate": 0.0004987381687030394,
"loss": 5.711,
"mean_token_accuracy": 0.15218381732702255,
"num_tokens": 9737044.0,
"step": 4685
},
{
"entropy": 5.85615348815918,
"epoch": 0.42543541364296084,
"grad_norm": 1.6171875,
"learning_rate": 0.0004987347444282479,
"loss": 5.8218,
"mean_token_accuracy": 0.1412054069340229,
"num_tokens": 9747648.0,
"step": 4690
},
{
"entropy": 5.794318532943725,
"epoch": 0.425888969521045,
"grad_norm": 2.5,
"learning_rate": 0.0004987313155265649,
"loss": 5.745,
"mean_token_accuracy": 0.1462424784898758,
"num_tokens": 9757812.0,
"step": 4695
},
{
"entropy": 5.915569925308228,
"epoch": 0.42634252539912915,
"grad_norm": 1.9453125,
"learning_rate": 0.0004987278819980613,
"loss": 5.7262,
"mean_token_accuracy": 0.15311951488256453,
"num_tokens": 9768337.0,
"step": 4700
},
{
"entropy": 5.859813261032104,
"epoch": 0.42679608127721336,
"grad_norm": 2.71875,
"learning_rate": 0.0004987244438428081,
"loss": 5.7565,
"mean_token_accuracy": 0.14580674096941948,
"num_tokens": 9779433.0,
"step": 4705
},
{
"entropy": 5.738063478469849,
"epoch": 0.4272496371552975,
"grad_norm": 1.6875,
"learning_rate": 0.0004987210010608764,
"loss": 5.6671,
"mean_token_accuracy": 0.15135972946882248,
"num_tokens": 9790069.0,
"step": 4710
},
{
"entropy": 5.79045934677124,
"epoch": 0.4277031930333817,
"grad_norm": 2.03125,
"learning_rate": 0.0004987175536523373,
"loss": 5.6582,
"mean_token_accuracy": 0.15610816180706025,
"num_tokens": 9799669.0,
"step": 4715
},
{
"entropy": 5.724021482467651,
"epoch": 0.4281567489114659,
"grad_norm": 1.90625,
"learning_rate": 0.0004987141016172622,
"loss": 5.6897,
"mean_token_accuracy": 0.15267266184091569,
"num_tokens": 9809615.0,
"step": 4720
},
{
"entropy": 5.697861099243164,
"epoch": 0.4286103047895501,
"grad_norm": 1.8828125,
"learning_rate": 0.0004987106449557225,
"loss": 5.6572,
"mean_token_accuracy": 0.1498062126338482,
"num_tokens": 9819666.0,
"step": 4725
},
{
"entropy": 5.8632340908050535,
"epoch": 0.42906386066763424,
"grad_norm": 1.8203125,
"learning_rate": 0.0004987071836677896,
"loss": 5.8091,
"mean_token_accuracy": 0.15095954835414888,
"num_tokens": 9831015.0,
"step": 4730
},
{
"entropy": 5.816402864456177,
"epoch": 0.42951741654571846,
"grad_norm": 1.7265625,
"learning_rate": 0.0004987037177535353,
"loss": 5.6802,
"mean_token_accuracy": 0.15113823860883713,
"num_tokens": 9841354.0,
"step": 4735
},
{
"entropy": 5.754711484909057,
"epoch": 0.4299709724238026,
"grad_norm": 2.21875,
"learning_rate": 0.0004987002472130309,
"loss": 5.6849,
"mean_token_accuracy": 0.1426093779504299,
"num_tokens": 9850467.0,
"step": 4740
},
{
"entropy": 5.866177654266357,
"epoch": 0.43042452830188677,
"grad_norm": 1.828125,
"learning_rate": 0.0004986967720463485,
"loss": 5.7402,
"mean_token_accuracy": 0.1482555314898491,
"num_tokens": 9861115.0,
"step": 4745
},
{
"entropy": 5.90579195022583,
"epoch": 0.430878084179971,
"grad_norm": 1.765625,
"learning_rate": 0.0004986932922535598,
"loss": 5.7527,
"mean_token_accuracy": 0.1503136560320854,
"num_tokens": 9871172.0,
"step": 4750
},
{
"entropy": 5.7877655029296875,
"epoch": 0.43133164005805513,
"grad_norm": 1.8359375,
"learning_rate": 0.0004986898078347368,
"loss": 5.649,
"mean_token_accuracy": 0.1556771621108055,
"num_tokens": 9881997.0,
"step": 4755
},
{
"entropy": 5.760748863220215,
"epoch": 0.43178519593613934,
"grad_norm": 1.5234375,
"learning_rate": 0.0004986863187899516,
"loss": 5.6544,
"mean_token_accuracy": 0.15546627789735795,
"num_tokens": 9892310.0,
"step": 4760
},
{
"entropy": 5.752875518798828,
"epoch": 0.4322387518142235,
"grad_norm": 2.296875,
"learning_rate": 0.0004986828251192764,
"loss": 5.7035,
"mean_token_accuracy": 0.1481189787387848,
"num_tokens": 9902664.0,
"step": 4765
},
{
"entropy": 5.767500114440918,
"epoch": 0.4326923076923077,
"grad_norm": 1.7578125,
"learning_rate": 0.0004986793268227833,
"loss": 5.6954,
"mean_token_accuracy": 0.15243975818157196,
"num_tokens": 9912491.0,
"step": 4770
},
{
"entropy": 5.762394523620605,
"epoch": 0.43314586357039186,
"grad_norm": 2.578125,
"learning_rate": 0.0004986758239005448,
"loss": 5.6399,
"mean_token_accuracy": 0.15671616941690444,
"num_tokens": 9922057.0,
"step": 4775
},
{
"entropy": 5.789015674591065,
"epoch": 0.4335994194484761,
"grad_norm": 1.6953125,
"learning_rate": 0.0004986723163526331,
"loss": 5.6908,
"mean_token_accuracy": 0.1488301694393158,
"num_tokens": 9933661.0,
"step": 4780
},
{
"entropy": 5.713762474060059,
"epoch": 0.43405297532656023,
"grad_norm": 2.015625,
"learning_rate": 0.0004986688041791209,
"loss": 5.6666,
"mean_token_accuracy": 0.1482269987463951,
"num_tokens": 9945177.0,
"step": 4785
},
{
"entropy": 5.790286445617676,
"epoch": 0.4345065312046444,
"grad_norm": 2.46875,
"learning_rate": 0.000498665287380081,
"loss": 5.7042,
"mean_token_accuracy": 0.1474942907691002,
"num_tokens": 9956226.0,
"step": 4790
},
{
"entropy": 5.862675619125366,
"epoch": 0.4349600870827286,
"grad_norm": 1.6953125,
"learning_rate": 0.0004986617659555859,
"loss": 5.7606,
"mean_token_accuracy": 0.15089563727378846,
"num_tokens": 9965877.0,
"step": 4795
},
{
"entropy": 5.725464248657227,
"epoch": 0.43541364296081275,
"grad_norm": 3.109375,
"learning_rate": 0.0004986582399057085,
"loss": 5.7076,
"mean_token_accuracy": 0.16087467074394227,
"num_tokens": 9975592.0,
"step": 4800
},
{
"entropy": 5.848356771469116,
"epoch": 0.43586719883889696,
"grad_norm": 3.515625,
"learning_rate": 0.0004986547092305216,
"loss": 5.7627,
"mean_token_accuracy": 0.14793257415294647,
"num_tokens": 9986094.0,
"step": 4805
},
{
"entropy": 5.916541671752929,
"epoch": 0.4363207547169811,
"grad_norm": 2.140625,
"learning_rate": 0.0004986511739300984,
"loss": 5.7923,
"mean_token_accuracy": 0.1465185172855854,
"num_tokens": 9997241.0,
"step": 4810
},
{
"entropy": 5.787732934951782,
"epoch": 0.43677431059506533,
"grad_norm": 2.046875,
"learning_rate": 0.0004986476340045119,
"loss": 5.7003,
"mean_token_accuracy": 0.15020343214273452,
"num_tokens": 10008091.0,
"step": 4815
},
{
"entropy": 5.809946966171265,
"epoch": 0.4372278664731495,
"grad_norm": 1.7890625,
"learning_rate": 0.0004986440894538354,
"loss": 5.7684,
"mean_token_accuracy": 0.14748568460345268,
"num_tokens": 10018794.0,
"step": 4820
},
{
"entropy": 5.7670200824737545,
"epoch": 0.4376814223512337,
"grad_norm": 3.0625,
"learning_rate": 0.0004986405402781421,
"loss": 5.656,
"mean_token_accuracy": 0.15169040709733964,
"num_tokens": 10028414.0,
"step": 4825
},
{
"entropy": 5.7817178726196286,
"epoch": 0.43813497822931785,
"grad_norm": 1.796875,
"learning_rate": 0.0004986369864775054,
"loss": 5.7511,
"mean_token_accuracy": 0.14933410361409188,
"num_tokens": 10038749.0,
"step": 4830
},
{
"entropy": 5.8474963188171385,
"epoch": 0.438588534107402,
"grad_norm": 2.09375,
"learning_rate": 0.0004986334280519989,
"loss": 5.6953,
"mean_token_accuracy": 0.15171761214733123,
"num_tokens": 10049866.0,
"step": 4835
},
{
"entropy": 5.808549976348877,
"epoch": 0.4390420899854862,
"grad_norm": 2.453125,
"learning_rate": 0.000498629865001696,
"loss": 5.7227,
"mean_token_accuracy": 0.15592157542705537,
"num_tokens": 10060680.0,
"step": 4840
},
{
"entropy": 5.869030904769898,
"epoch": 0.43949564586357037,
"grad_norm": 2.140625,
"learning_rate": 0.0004986262973266706,
"loss": 5.8215,
"mean_token_accuracy": 0.14849047362804413,
"num_tokens": 10070931.0,
"step": 4845
},
{
"entropy": 5.759446382522583,
"epoch": 0.4399492017416546,
"grad_norm": 2.765625,
"learning_rate": 0.0004986227250269964,
"loss": 5.7429,
"mean_token_accuracy": 0.15099930614233018,
"num_tokens": 10080189.0,
"step": 4850
},
{
"entropy": 5.7940160751342775,
"epoch": 0.44040275761973874,
"grad_norm": 1.9140625,
"learning_rate": 0.0004986191481027472,
"loss": 5.6391,
"mean_token_accuracy": 0.15560751110315324,
"num_tokens": 10088813.0,
"step": 4855
},
{
"entropy": 5.818889951705932,
"epoch": 0.44085631349782295,
"grad_norm": 2.34375,
"learning_rate": 0.0004986155665539971,
"loss": 5.6825,
"mean_token_accuracy": 0.15352422297000884,
"num_tokens": 10098125.0,
"step": 4860
},
{
"entropy": 5.838857555389405,
"epoch": 0.4413098693759071,
"grad_norm": 1.765625,
"learning_rate": 0.00049861198038082,
"loss": 5.7218,
"mean_token_accuracy": 0.14922336339950562,
"num_tokens": 10109125.0,
"step": 4865
},
{
"entropy": 5.759595823287964,
"epoch": 0.4417634252539913,
"grad_norm": 1.78125,
"learning_rate": 0.0004986083895832903,
"loss": 5.6983,
"mean_token_accuracy": 0.15355753526091576,
"num_tokens": 10120182.0,
"step": 4870
},
{
"entropy": 5.748368501663208,
"epoch": 0.44221698113207547,
"grad_norm": 2.625,
"learning_rate": 0.0004986047941614821,
"loss": 5.7589,
"mean_token_accuracy": 0.1470588982105255,
"num_tokens": 10130941.0,
"step": 4875
},
{
"entropy": 5.866014432907105,
"epoch": 0.4426705370101596,
"grad_norm": 2.328125,
"learning_rate": 0.0004986011941154696,
"loss": 5.7273,
"mean_token_accuracy": 0.14755107015371322,
"num_tokens": 10141708.0,
"step": 4880
},
{
"entropy": 5.807976961135864,
"epoch": 0.44312409288824384,
"grad_norm": 2.28125,
"learning_rate": 0.0004985975894453275,
"loss": 5.6554,
"mean_token_accuracy": 0.15046119764447213,
"num_tokens": 10151816.0,
"step": 4885
},
{
"entropy": 5.810320949554443,
"epoch": 0.443577648766328,
"grad_norm": 2.125,
"learning_rate": 0.0004985939801511304,
"loss": 5.7727,
"mean_token_accuracy": 0.14902833551168443,
"num_tokens": 10162470.0,
"step": 4890
},
{
"entropy": 5.759214115142822,
"epoch": 0.4440312046444122,
"grad_norm": 1.5703125,
"learning_rate": 0.0004985903662329527,
"loss": 5.6688,
"mean_token_accuracy": 0.15527875274419783,
"num_tokens": 10173171.0,
"step": 4895
},
{
"entropy": 5.795230531692505,
"epoch": 0.44448476052249636,
"grad_norm": 1.828125,
"learning_rate": 0.0004985867476908693,
"loss": 5.7134,
"mean_token_accuracy": 0.15014212280511857,
"num_tokens": 10184185.0,
"step": 4900
},
{
"entropy": 5.781923294067383,
"epoch": 0.44493831640058057,
"grad_norm": 1.6875,
"learning_rate": 0.0004985831245249549,
"loss": 5.764,
"mean_token_accuracy": 0.1502835273742676,
"num_tokens": 10195516.0,
"step": 4905
},
{
"entropy": 5.752807474136352,
"epoch": 0.4453918722786647,
"grad_norm": 2.09375,
"learning_rate": 0.0004985794967352846,
"loss": 5.6884,
"mean_token_accuracy": 0.15242686122655869,
"num_tokens": 10204936.0,
"step": 4910
},
{
"entropy": 5.839371204376221,
"epoch": 0.44584542815674894,
"grad_norm": 1.6484375,
"learning_rate": 0.0004985758643219334,
"loss": 5.7183,
"mean_token_accuracy": 0.15398919135332106,
"num_tokens": 10216218.0,
"step": 4915
},
{
"entropy": 5.87683072090149,
"epoch": 0.4462989840348331,
"grad_norm": 1.6796875,
"learning_rate": 0.0004985722272849762,
"loss": 5.7421,
"mean_token_accuracy": 0.14640209078788757,
"num_tokens": 10226370.0,
"step": 4920
},
{
"entropy": 5.758306503295898,
"epoch": 0.44675253991291725,
"grad_norm": 1.7421875,
"learning_rate": 0.0004985685856244884,
"loss": 5.6542,
"mean_token_accuracy": 0.15684788078069686,
"num_tokens": 10236170.0,
"step": 4925
},
{
"entropy": 5.79812445640564,
"epoch": 0.44720609579100146,
"grad_norm": 1.5546875,
"learning_rate": 0.0004985649393405453,
"loss": 5.7329,
"mean_token_accuracy": 0.1511545680463314,
"num_tokens": 10246119.0,
"step": 4930
},
{
"entropy": 5.804480504989624,
"epoch": 0.4476596516690856,
"grad_norm": 1.71875,
"learning_rate": 0.0004985612884332223,
"loss": 5.7306,
"mean_token_accuracy": 0.14677015766501428,
"num_tokens": 10256762.0,
"step": 4935
},
{
"entropy": 5.74443507194519,
"epoch": 0.4481132075471698,
"grad_norm": 1.609375,
"learning_rate": 0.0004985576329025949,
"loss": 5.6832,
"mean_token_accuracy": 0.1538843147456646,
"num_tokens": 10267317.0,
"step": 4940
},
{
"entropy": 5.833537530899048,
"epoch": 0.448566763425254,
"grad_norm": 2.0,
"learning_rate": 0.0004985539727487385,
"loss": 5.8213,
"mean_token_accuracy": 0.13688837736845016,
"num_tokens": 10279111.0,
"step": 4945
},
{
"entropy": 5.81101803779602,
"epoch": 0.4490203193033382,
"grad_norm": 2.421875,
"learning_rate": 0.0004985503079717292,
"loss": 5.6847,
"mean_token_accuracy": 0.152994741499424,
"num_tokens": 10289383.0,
"step": 4950
},
{
"entropy": 5.826288223266602,
"epoch": 0.44947387518142234,
"grad_norm": 2.703125,
"learning_rate": 0.0004985466385716425,
"loss": 5.7095,
"mean_token_accuracy": 0.15056138932704927,
"num_tokens": 10299806.0,
"step": 4955
},
{
"entropy": 5.772299671173096,
"epoch": 0.44992743105950656,
"grad_norm": 1.6953125,
"learning_rate": 0.0004985429645485543,
"loss": 5.7137,
"mean_token_accuracy": 0.14749846756458282,
"num_tokens": 10310348.0,
"step": 4960
},
{
"entropy": 5.797397184371948,
"epoch": 0.4503809869375907,
"grad_norm": 2.609375,
"learning_rate": 0.0004985392859025407,
"loss": 5.7179,
"mean_token_accuracy": 0.1548716500401497,
"num_tokens": 10320354.0,
"step": 4965
},
{
"entropy": 5.755021715164185,
"epoch": 0.45083454281567487,
"grad_norm": 2.125,
"learning_rate": 0.0004985356026336776,
"loss": 5.6464,
"mean_token_accuracy": 0.15377871841192245,
"num_tokens": 10330292.0,
"step": 4970
},
{
"entropy": 5.813851833343506,
"epoch": 0.4512880986937591,
"grad_norm": 2.03125,
"learning_rate": 0.0004985319147420414,
"loss": 5.6087,
"mean_token_accuracy": 0.15954625755548477,
"num_tokens": 10340449.0,
"step": 4975
},
{
"entropy": 5.773745012283325,
"epoch": 0.45174165457184323,
"grad_norm": 1.9375,
"learning_rate": 0.0004985282222277082,
"loss": 5.6679,
"mean_token_accuracy": 0.15184469521045685,
"num_tokens": 10350637.0,
"step": 4980
},
{
"entropy": 5.7085271835327145,
"epoch": 0.45219521044992744,
"grad_norm": 2.015625,
"learning_rate": 0.0004985245250907544,
"loss": 5.6161,
"mean_token_accuracy": 0.149700116366148,
"num_tokens": 10361548.0,
"step": 4985
},
{
"entropy": 5.759440231323242,
"epoch": 0.4526487663280116,
"grad_norm": 1.7265625,
"learning_rate": 0.0004985208233312564,
"loss": 5.6254,
"mean_token_accuracy": 0.14682499915361405,
"num_tokens": 10371765.0,
"step": 4990
},
{
"entropy": 5.842993068695068,
"epoch": 0.4531023222060958,
"grad_norm": 2.09375,
"learning_rate": 0.0004985171169492908,
"loss": 5.7484,
"mean_token_accuracy": 0.15154104083776473,
"num_tokens": 10382108.0,
"step": 4995
},
{
"entropy": 5.793890333175659,
"epoch": 0.45355587808417996,
"grad_norm": 3.21875,
"learning_rate": 0.0004985134059449344,
"loss": 5.6946,
"mean_token_accuracy": 0.152512563765049,
"num_tokens": 10392390.0,
"step": 5000
},
{
"entropy": 5.75899076461792,
"epoch": 0.4540094339622642,
"grad_norm": 3.234375,
"learning_rate": 0.0004985096903182637,
"loss": 5.7425,
"mean_token_accuracy": 0.15878509283065795,
"num_tokens": 10402094.0,
"step": 5005
},
{
"entropy": 5.810795450210572,
"epoch": 0.45446298984034833,
"grad_norm": 2.328125,
"learning_rate": 0.0004985059700693557,
"loss": 5.7089,
"mean_token_accuracy": 0.15410526245832443,
"num_tokens": 10413350.0,
"step": 5010
},
{
"entropy": 5.805242109298706,
"epoch": 0.4549165457184325,
"grad_norm": 2.375,
"learning_rate": 0.0004985022451982872,
"loss": 5.8358,
"mean_token_accuracy": 0.14019331857562065,
"num_tokens": 10424292.0,
"step": 5015
},
{
"entropy": 5.866724157333374,
"epoch": 0.4553701015965167,
"grad_norm": 1.71875,
"learning_rate": 0.0004984985157051354,
"loss": 5.6723,
"mean_token_accuracy": 0.15092671811580657,
"num_tokens": 10433724.0,
"step": 5020
},
{
"entropy": 5.848974752426147,
"epoch": 0.45582365747460085,
"grad_norm": 2.046875,
"learning_rate": 0.0004984947815899774,
"loss": 5.7603,
"mean_token_accuracy": 0.14992742538452147,
"num_tokens": 10444724.0,
"step": 5025
},
{
"entropy": 5.833653450012207,
"epoch": 0.45627721335268506,
"grad_norm": 1.71875,
"learning_rate": 0.0004984910428528902,
"loss": 5.6826,
"mean_token_accuracy": 0.1587917149066925,
"num_tokens": 10455086.0,
"step": 5030
},
{
"entropy": 5.742634630203247,
"epoch": 0.4567307692307692,
"grad_norm": 2.296875,
"learning_rate": 0.0004984872994939514,
"loss": 5.65,
"mean_token_accuracy": 0.14961964190006255,
"num_tokens": 10465076.0,
"step": 5035
},
{
"entropy": 5.792289400100708,
"epoch": 0.45718432510885343,
"grad_norm": 3.171875,
"learning_rate": 0.0004984835515132382,
"loss": 5.6824,
"mean_token_accuracy": 0.1561535358428955,
"num_tokens": 10474817.0,
"step": 5040
},
{
"entropy": 5.819376420974732,
"epoch": 0.4576378809869376,
"grad_norm": 1.6953125,
"learning_rate": 0.0004984797989108282,
"loss": 5.6929,
"mean_token_accuracy": 0.14523695558309554,
"num_tokens": 10485735.0,
"step": 5045
},
{
"entropy": 5.760282325744629,
"epoch": 0.4580914368650218,
"grad_norm": 5.40625,
"learning_rate": 0.0004984760416867991,
"loss": 5.7899,
"mean_token_accuracy": 0.14944545552134514,
"num_tokens": 10496731.0,
"step": 5050
},
{
"entropy": 5.723328304290772,
"epoch": 0.45854499274310595,
"grad_norm": 2.046875,
"learning_rate": 0.0004984722798412286,
"loss": 5.6457,
"mean_token_accuracy": 0.15127401053905487,
"num_tokens": 10506787.0,
"step": 5055
},
{
"entropy": 5.821808862686157,
"epoch": 0.4589985486211901,
"grad_norm": 2.40625,
"learning_rate": 0.0004984685133741941,
"loss": 5.7131,
"mean_token_accuracy": 0.15030114501714706,
"num_tokens": 10517170.0,
"step": 5060
},
{
"entropy": 5.830482482910156,
"epoch": 0.4594521044992743,
"grad_norm": 1.8046875,
"learning_rate": 0.000498464742285774,
"loss": 5.7078,
"mean_token_accuracy": 0.1527443915605545,
"num_tokens": 10527342.0,
"step": 5065
},
{
"entropy": 5.818163919448852,
"epoch": 0.45990566037735847,
"grad_norm": 1.859375,
"learning_rate": 0.000498460966576046,
"loss": 5.6866,
"mean_token_accuracy": 0.15609674155712128,
"num_tokens": 10537253.0,
"step": 5070
},
{
"entropy": 5.819856405258179,
"epoch": 0.4603592162554427,
"grad_norm": 1.96875,
"learning_rate": 0.0004984571862450884,
"loss": 5.6989,
"mean_token_accuracy": 0.15164147466421127,
"num_tokens": 10547249.0,
"step": 5075
},
{
"entropy": 5.820227432250976,
"epoch": 0.46081277213352684,
"grad_norm": 1.703125,
"learning_rate": 0.0004984534012929791,
"loss": 5.7285,
"mean_token_accuracy": 0.15311849117279053,
"num_tokens": 10558174.0,
"step": 5080
},
{
"entropy": 5.867569541931152,
"epoch": 0.46126632801161105,
"grad_norm": 1.7109375,
"learning_rate": 0.0004984496117197967,
"loss": 5.6984,
"mean_token_accuracy": 0.15882893800735473,
"num_tokens": 10567610.0,
"step": 5085
},
{
"entropy": 5.770347452163696,
"epoch": 0.4617198838896952,
"grad_norm": 1.984375,
"learning_rate": 0.0004984458175256193,
"loss": 5.7127,
"mean_token_accuracy": 0.1474909946322441,
"num_tokens": 10578494.0,
"step": 5090
},
{
"entropy": 5.84233832359314,
"epoch": 0.4621734397677794,
"grad_norm": 2.140625,
"learning_rate": 0.0004984420187105253,
"loss": 5.8042,
"mean_token_accuracy": 0.15005899965763092,
"num_tokens": 10589482.0,
"step": 5095
},
{
"entropy": 5.844469022750855,
"epoch": 0.46262699564586357,
"grad_norm": 1.5703125,
"learning_rate": 0.0004984382152745936,
"loss": 5.7548,
"mean_token_accuracy": 0.14896771609783171,
"num_tokens": 10599965.0,
"step": 5100
},
{
"entropy": 5.837211656570434,
"epoch": 0.4630805515239477,
"grad_norm": 3.15625,
"learning_rate": 0.0004984344072179026,
"loss": 5.7981,
"mean_token_accuracy": 0.15154444575309753,
"num_tokens": 10610500.0,
"step": 5105
},
{
"entropy": 5.847231006622314,
"epoch": 0.46353410740203194,
"grad_norm": 2.03125,
"learning_rate": 0.0004984305945405312,
"loss": 5.7322,
"mean_token_accuracy": 0.15490227639675141,
"num_tokens": 10619843.0,
"step": 5110
},
{
"entropy": 5.841805505752563,
"epoch": 0.4639876632801161,
"grad_norm": 1.890625,
"learning_rate": 0.000498426777242558,
"loss": 5.6631,
"mean_token_accuracy": 0.15643091574311258,
"num_tokens": 10631026.0,
"step": 5115
},
{
"entropy": 5.654076147079468,
"epoch": 0.4644412191582003,
"grad_norm": 1.921875,
"learning_rate": 0.0004984229553240623,
"loss": 5.6675,
"mean_token_accuracy": 0.15852237939834596,
"num_tokens": 10641130.0,
"step": 5120
},
{
"entropy": 5.7756763935089115,
"epoch": 0.46489477503628446,
"grad_norm": 1.875,
"learning_rate": 0.0004984191287851228,
"loss": 5.725,
"mean_token_accuracy": 0.15592035204172133,
"num_tokens": 10652305.0,
"step": 5125
},
{
"entropy": 5.886051511764526,
"epoch": 0.46534833091436867,
"grad_norm": 1.8828125,
"learning_rate": 0.0004984152976258188,
"loss": 5.7613,
"mean_token_accuracy": 0.1512484297156334,
"num_tokens": 10663083.0,
"step": 5130
},
{
"entropy": 5.774004697799683,
"epoch": 0.4658018867924528,
"grad_norm": 1.6953125,
"learning_rate": 0.0004984114618462296,
"loss": 5.7842,
"mean_token_accuracy": 0.14426478892564773,
"num_tokens": 10673602.0,
"step": 5135
},
{
"entropy": 5.794326400756836,
"epoch": 0.46625544267053703,
"grad_norm": 2.140625,
"learning_rate": 0.0004984076214464343,
"loss": 5.6242,
"mean_token_accuracy": 0.15901431143283845,
"num_tokens": 10684622.0,
"step": 5140
},
{
"entropy": 5.887795543670654,
"epoch": 0.4667089985486212,
"grad_norm": 1.9140625,
"learning_rate": 0.0004984037764265126,
"loss": 5.7383,
"mean_token_accuracy": 0.15451854169368745,
"num_tokens": 10695129.0,
"step": 5145
},
{
"entropy": 5.766786050796509,
"epoch": 0.46716255442670535,
"grad_norm": 2.296875,
"learning_rate": 0.0004983999267865438,
"loss": 5.5757,
"mean_token_accuracy": 0.1634918212890625,
"num_tokens": 10705236.0,
"step": 5150
},
{
"entropy": 5.73983416557312,
"epoch": 0.46761611030478956,
"grad_norm": 2.203125,
"learning_rate": 0.0004983960725266078,
"loss": 5.6921,
"mean_token_accuracy": 0.1524804301559925,
"num_tokens": 10715748.0,
"step": 5155
},
{
"entropy": 5.793629693984985,
"epoch": 0.4680696661828737,
"grad_norm": 1.65625,
"learning_rate": 0.0004983922136467838,
"loss": 5.6295,
"mean_token_accuracy": 0.15277455896139144,
"num_tokens": 10725799.0,
"step": 5160
},
{
"entropy": 5.825521945953369,
"epoch": 0.4685232220609579,
"grad_norm": 1.96875,
"learning_rate": 0.000498388350147152,
"loss": 5.7382,
"mean_token_accuracy": 0.15017426609992982,
"num_tokens": 10735668.0,
"step": 5165
},
{
"entropy": 5.887571239471436,
"epoch": 0.4689767779390421,
"grad_norm": 1.953125,
"learning_rate": 0.0004983844820277922,
"loss": 5.8401,
"mean_token_accuracy": 0.14530294463038446,
"num_tokens": 10746466.0,
"step": 5170
},
{
"entropy": 5.886936998367309,
"epoch": 0.4694303338171263,
"grad_norm": 2.34375,
"learning_rate": 0.0004983806092887843,
"loss": 5.7568,
"mean_token_accuracy": 0.1505974516272545,
"num_tokens": 10756724.0,
"step": 5175
},
{
"entropy": 5.797704792022705,
"epoch": 0.46988388969521044,
"grad_norm": 1.6484375,
"learning_rate": 0.0004983767319302086,
"loss": 5.7777,
"mean_token_accuracy": 0.152411787211895,
"num_tokens": 10766919.0,
"step": 5180
},
{
"entropy": 5.723730754852295,
"epoch": 0.47033744557329465,
"grad_norm": 2.296875,
"learning_rate": 0.0004983728499521451,
"loss": 5.5886,
"mean_token_accuracy": 0.1582037776708603,
"num_tokens": 10777550.0,
"step": 5185
},
{
"entropy": 5.76413779258728,
"epoch": 0.4707910014513788,
"grad_norm": 1.765625,
"learning_rate": 0.0004983689633546741,
"loss": 5.6539,
"mean_token_accuracy": 0.14905960708856583,
"num_tokens": 10788456.0,
"step": 5190
},
{
"entropy": 5.785346984863281,
"epoch": 0.47124455732946297,
"grad_norm": 2.28125,
"learning_rate": 0.0004983650721378761,
"loss": 5.7027,
"mean_token_accuracy": 0.150688636302948,
"num_tokens": 10800049.0,
"step": 5195
},
{
"entropy": 5.839448738098144,
"epoch": 0.4716981132075472,
"grad_norm": 2.859375,
"learning_rate": 0.0004983611763018314,
"loss": 5.7305,
"mean_token_accuracy": 0.14705110639333724,
"num_tokens": 10810743.0,
"step": 5200
},
{
"entropy": 5.903285980224609,
"epoch": 0.47215166908563133,
"grad_norm": 1.5859375,
"learning_rate": 0.0004983572758466208,
"loss": 5.9008,
"mean_token_accuracy": 0.14664009138941764,
"num_tokens": 10821897.0,
"step": 5205
},
{
"entropy": 5.864152622222901,
"epoch": 0.47260522496371554,
"grad_norm": 1.6796875,
"learning_rate": 0.0004983533707723247,
"loss": 5.7536,
"mean_token_accuracy": 0.1474456422030926,
"num_tokens": 10831589.0,
"step": 5210
},
{
"entropy": 5.839668083190918,
"epoch": 0.4730587808417997,
"grad_norm": 3.15625,
"learning_rate": 0.0004983494610790241,
"loss": 5.6822,
"mean_token_accuracy": 0.14612138271331787,
"num_tokens": 10841404.0,
"step": 5215
},
{
"entropy": 5.779344511032105,
"epoch": 0.4735123367198839,
"grad_norm": 2.1875,
"learning_rate": 0.0004983455467667996,
"loss": 5.7107,
"mean_token_accuracy": 0.14737479761242867,
"num_tokens": 10851577.0,
"step": 5220
},
{
"entropy": 5.737367773056031,
"epoch": 0.47396589259796806,
"grad_norm": 1.515625,
"learning_rate": 0.0004983416278357322,
"loss": 5.6337,
"mean_token_accuracy": 0.15840625911951065,
"num_tokens": 10861403.0,
"step": 5225
},
{
"entropy": 5.810776615142823,
"epoch": 0.4744194484760523,
"grad_norm": 2.296875,
"learning_rate": 0.0004983377042859032,
"loss": 5.683,
"mean_token_accuracy": 0.15175287276506425,
"num_tokens": 10871933.0,
"step": 5230
},
{
"entropy": 5.828769016265869,
"epoch": 0.47487300435413643,
"grad_norm": 2.125,
"learning_rate": 0.0004983337761173936,
"loss": 5.7531,
"mean_token_accuracy": 0.15067264437675476,
"num_tokens": 10881800.0,
"step": 5235
},
{
"entropy": 5.891356468200684,
"epoch": 0.4753265602322206,
"grad_norm": 2.859375,
"learning_rate": 0.0004983298433302843,
"loss": 5.7103,
"mean_token_accuracy": 0.148338982462883,
"num_tokens": 10891830.0,
"step": 5240
},
{
"entropy": 5.783552646636963,
"epoch": 0.4757801161103048,
"grad_norm": 1.6015625,
"learning_rate": 0.0004983259059246572,
"loss": 5.6244,
"mean_token_accuracy": 0.15633412152528764,
"num_tokens": 10901557.0,
"step": 5245
},
{
"entropy": 5.69632363319397,
"epoch": 0.47623367198838895,
"grad_norm": 4.9375,
"learning_rate": 0.0004983219639005934,
"loss": 5.7349,
"mean_token_accuracy": 0.14440841749310493,
"num_tokens": 10912155.0,
"step": 5250
},
{
"entropy": 5.76448540687561,
"epoch": 0.47668722786647316,
"grad_norm": 2.40625,
"learning_rate": 0.0004983180172581745,
"loss": 5.6588,
"mean_token_accuracy": 0.15334019362926482,
"num_tokens": 10922291.0,
"step": 5255
},
{
"entropy": 5.8298650741577145,
"epoch": 0.4771407837445573,
"grad_norm": 2.1875,
"learning_rate": 0.0004983140659974819,
"loss": 5.7061,
"mean_token_accuracy": 0.1492989644408226,
"num_tokens": 10933176.0,
"step": 5260
},
{
"entropy": 5.863755178451538,
"epoch": 0.47759433962264153,
"grad_norm": 2.46875,
"learning_rate": 0.0004983101101185977,
"loss": 5.7933,
"mean_token_accuracy": 0.14786036014556886,
"num_tokens": 10944049.0,
"step": 5265
},
{
"entropy": 5.7591461658477785,
"epoch": 0.4780478955007257,
"grad_norm": 1.765625,
"learning_rate": 0.0004983061496216035,
"loss": 5.6509,
"mean_token_accuracy": 0.15584611147642136,
"num_tokens": 10954347.0,
"step": 5270
},
{
"entropy": 5.797364330291748,
"epoch": 0.4785014513788099,
"grad_norm": 2.4375,
"learning_rate": 0.0004983021845065812,
"loss": 5.7072,
"mean_token_accuracy": 0.14649278968572615,
"num_tokens": 10965123.0,
"step": 5275
},
{
"entropy": 5.786959505081176,
"epoch": 0.47895500725689405,
"grad_norm": 2.1875,
"learning_rate": 0.0004982982147736128,
"loss": 5.6946,
"mean_token_accuracy": 0.15307522416114808,
"num_tokens": 10976395.0,
"step": 5280
},
{
"entropy": 5.78529634475708,
"epoch": 0.4794085631349782,
"grad_norm": 2.421875,
"learning_rate": 0.0004982942404227805,
"loss": 5.6782,
"mean_token_accuracy": 0.15125040709972382,
"num_tokens": 10986797.0,
"step": 5285
},
{
"entropy": 5.859493732452393,
"epoch": 0.4798621190130624,
"grad_norm": 2.859375,
"learning_rate": 0.0004982902614541662,
"loss": 5.7408,
"mean_token_accuracy": 0.1478838101029396,
"num_tokens": 10997229.0,
"step": 5290
},
{
"entropy": 5.80925989151001,
"epoch": 0.48031567489114657,
"grad_norm": 2.203125,
"learning_rate": 0.0004982862778678526,
"loss": 5.6847,
"mean_token_accuracy": 0.15592680871486664,
"num_tokens": 11008180.0,
"step": 5295
},
{
"entropy": 5.8700981616973875,
"epoch": 0.4807692307692308,
"grad_norm": 2.9375,
"learning_rate": 0.0004982822896639218,
"loss": 5.7015,
"mean_token_accuracy": 0.1546702854335308,
"num_tokens": 11019248.0,
"step": 5300
},
{
"entropy": 5.742985486984253,
"epoch": 0.48122278664731494,
"grad_norm": 2.296875,
"learning_rate": 0.0004982782968424563,
"loss": 5.627,
"mean_token_accuracy": 0.1553596355021,
"num_tokens": 11029760.0,
"step": 5305
},
{
"entropy": 5.761133146286011,
"epoch": 0.48167634252539915,
"grad_norm": 2.140625,
"learning_rate": 0.0004982742994035388,
"loss": 5.6959,
"mean_token_accuracy": 0.15285762995481492,
"num_tokens": 11039211.0,
"step": 5310
},
{
"entropy": 5.807599353790283,
"epoch": 0.4821298984034833,
"grad_norm": 2.546875,
"learning_rate": 0.0004982702973472518,
"loss": 5.68,
"mean_token_accuracy": 0.14845879822969438,
"num_tokens": 11049892.0,
"step": 5315
},
{
"entropy": 5.902496957778931,
"epoch": 0.4825834542815675,
"grad_norm": 2.046875,
"learning_rate": 0.0004982662906736783,
"loss": 5.7432,
"mean_token_accuracy": 0.1499881386756897,
"num_tokens": 11060823.0,
"step": 5320
},
{
"entropy": 5.758925247192383,
"epoch": 0.48303701015965167,
"grad_norm": 2.59375,
"learning_rate": 0.0004982622793829009,
"loss": 5.7125,
"mean_token_accuracy": 0.15161036998033522,
"num_tokens": 11072353.0,
"step": 5325
},
{
"entropy": 5.734290838241577,
"epoch": 0.4834905660377358,
"grad_norm": 2.140625,
"learning_rate": 0.0004982582634750027,
"loss": 5.7187,
"mean_token_accuracy": 0.1546842187643051,
"num_tokens": 11082655.0,
"step": 5330
},
{
"entropy": 5.862860250473022,
"epoch": 0.48394412191582004,
"grad_norm": 2.46875,
"learning_rate": 0.0004982542429500667,
"loss": 5.7398,
"mean_token_accuracy": 0.15047571957111358,
"num_tokens": 11093315.0,
"step": 5335
},
{
"entropy": 5.680254030227661,
"epoch": 0.4843976777939042,
"grad_norm": 2.0625,
"learning_rate": 0.0004982502178081761,
"loss": 5.6287,
"mean_token_accuracy": 0.15498099178075792,
"num_tokens": 11103074.0,
"step": 5340
},
{
"entropy": 5.764597225189209,
"epoch": 0.4848512336719884,
"grad_norm": 2.296875,
"learning_rate": 0.0004982461880494141,
"loss": 5.685,
"mean_token_accuracy": 0.15466154515743255,
"num_tokens": 11113636.0,
"step": 5345
},
{
"entropy": 5.837761116027832,
"epoch": 0.48530478955007256,
"grad_norm": 1.875,
"learning_rate": 0.000498242153673864,
"loss": 5.6285,
"mean_token_accuracy": 0.15595686733722686,
"num_tokens": 11123509.0,
"step": 5350
},
{
"entropy": 5.773257827758789,
"epoch": 0.48575834542815677,
"grad_norm": 5.40625,
"learning_rate": 0.0004982381146816094,
"loss": 5.6869,
"mean_token_accuracy": 0.14768464267253875,
"num_tokens": 11133405.0,
"step": 5355
},
{
"entropy": 5.839870834350586,
"epoch": 0.4862119013062409,
"grad_norm": 1.75,
"learning_rate": 0.0004982340710727336,
"loss": 5.7506,
"mean_token_accuracy": 0.14249877035617828,
"num_tokens": 11144725.0,
"step": 5360
},
{
"entropy": 5.817921447753906,
"epoch": 0.48666545718432513,
"grad_norm": 2.1875,
"learning_rate": 0.0004982300228473203,
"loss": 5.6389,
"mean_token_accuracy": 0.15178342014551163,
"num_tokens": 11155446.0,
"step": 5365
},
{
"entropy": 5.790373659133911,
"epoch": 0.4871190130624093,
"grad_norm": 2.265625,
"learning_rate": 0.0004982259700054533,
"loss": 5.7171,
"mean_token_accuracy": 0.15540633648633956,
"num_tokens": 11165380.0,
"step": 5370
},
{
"entropy": 5.789110803604126,
"epoch": 0.48757256894049344,
"grad_norm": 1.8671875,
"learning_rate": 0.0004982219125472163,
"loss": 5.7436,
"mean_token_accuracy": 0.15281054824590684,
"num_tokens": 11175641.0,
"step": 5375
},
{
"entropy": 5.936801338195801,
"epoch": 0.48802612481857766,
"grad_norm": 2.546875,
"learning_rate": 0.0004982178504726933,
"loss": 5.9078,
"mean_token_accuracy": 0.14865122362971306,
"num_tokens": 11186257.0,
"step": 5380
},
{
"entropy": 5.878651189804077,
"epoch": 0.4884796806966618,
"grad_norm": 1.96875,
"learning_rate": 0.0004982137837819682,
"loss": 5.7208,
"mean_token_accuracy": 0.14411981850862504,
"num_tokens": 11197284.0,
"step": 5385
},
{
"entropy": 5.793639707565307,
"epoch": 0.488933236574746,
"grad_norm": 2.265625,
"learning_rate": 0.0004982097124751252,
"loss": 5.7082,
"mean_token_accuracy": 0.15416025072336198,
"num_tokens": 11208413.0,
"step": 5390
},
{
"entropy": 5.765456247329712,
"epoch": 0.4893867924528302,
"grad_norm": 2.140625,
"learning_rate": 0.0004982056365522486,
"loss": 5.7144,
"mean_token_accuracy": 0.15092865526676177,
"num_tokens": 11219016.0,
"step": 5395
},
{
"entropy": 5.786841249465942,
"epoch": 0.4898403483309144,
"grad_norm": 1.578125,
"learning_rate": 0.0004982015560134225,
"loss": 5.6889,
"mean_token_accuracy": 0.15537251830101012,
"num_tokens": 11228934.0,
"step": 5400
},
{
"entropy": 5.782479906082154,
"epoch": 0.49029390420899854,
"grad_norm": 1.9921875,
"learning_rate": 0.0004981974708587312,
"loss": 5.6246,
"mean_token_accuracy": 0.1535349190235138,
"num_tokens": 11239596.0,
"step": 5405
},
{
"entropy": 5.733927965164185,
"epoch": 0.49074746008708275,
"grad_norm": 2.390625,
"learning_rate": 0.0004981933810882594,
"loss": 5.6093,
"mean_token_accuracy": 0.1489216908812523,
"num_tokens": 11249560.0,
"step": 5410
},
{
"entropy": 5.848641347885132,
"epoch": 0.4912010159651669,
"grad_norm": 2.5,
"learning_rate": 0.0004981892867020917,
"loss": 5.8021,
"mean_token_accuracy": 0.1476719044148922,
"num_tokens": 11259787.0,
"step": 5415
},
{
"entropy": 5.838620996475219,
"epoch": 0.49165457184325106,
"grad_norm": 1.96875,
"learning_rate": 0.0004981851877003125,
"loss": 5.7201,
"mean_token_accuracy": 0.15290852189064025,
"num_tokens": 11269717.0,
"step": 5420
},
{
"entropy": 5.760362434387207,
"epoch": 0.4921081277213353,
"grad_norm": 1.921875,
"learning_rate": 0.0004981810840830068,
"loss": 5.7136,
"mean_token_accuracy": 0.15744489878416063,
"num_tokens": 11279863.0,
"step": 5425
},
{
"entropy": 5.87311692237854,
"epoch": 0.49256168359941943,
"grad_norm": 1.4140625,
"learning_rate": 0.0004981769758502596,
"loss": 5.8277,
"mean_token_accuracy": 0.1397744357585907,
"num_tokens": 11291161.0,
"step": 5430
},
{
"entropy": 5.768136358261108,
"epoch": 0.49301523947750364,
"grad_norm": 2.671875,
"learning_rate": 0.0004981728630021555,
"loss": 5.6521,
"mean_token_accuracy": 0.1497293785214424,
"num_tokens": 11301151.0,
"step": 5435
},
{
"entropy": 5.687364625930786,
"epoch": 0.4934687953555878,
"grad_norm": 1.8203125,
"learning_rate": 0.0004981687455387797,
"loss": 5.6485,
"mean_token_accuracy": 0.15389353632926941,
"num_tokens": 11311466.0,
"step": 5440
},
{
"entropy": 5.780608606338501,
"epoch": 0.493922351233672,
"grad_norm": 1.7421875,
"learning_rate": 0.0004981646234602173,
"loss": 5.6273,
"mean_token_accuracy": 0.15507608875632287,
"num_tokens": 11321522.0,
"step": 5445
},
{
"entropy": 5.889606618881226,
"epoch": 0.49437590711175616,
"grad_norm": 2.171875,
"learning_rate": 0.0004981604967665539,
"loss": 5.7239,
"mean_token_accuracy": 0.15444112122058867,
"num_tokens": 11332317.0,
"step": 5450
},
{
"entropy": 5.775820875167847,
"epoch": 0.4948294629898404,
"grad_norm": 2.359375,
"learning_rate": 0.0004981563654578743,
"loss": 5.6535,
"mean_token_accuracy": 0.1533982649445534,
"num_tokens": 11343024.0,
"step": 5455
},
{
"entropy": 5.723749732971191,
"epoch": 0.49528301886792453,
"grad_norm": 2.015625,
"learning_rate": 0.0004981522295342643,
"loss": 5.6269,
"mean_token_accuracy": 0.15509956181049347,
"num_tokens": 11353114.0,
"step": 5460
},
{
"entropy": 5.742253732681275,
"epoch": 0.4957365747460087,
"grad_norm": 1.9921875,
"learning_rate": 0.0004981480889958092,
"loss": 5.6881,
"mean_token_accuracy": 0.15197990238666534,
"num_tokens": 11363446.0,
"step": 5465
},
{
"entropy": 5.742916679382324,
"epoch": 0.4961901306240929,
"grad_norm": 1.6640625,
"learning_rate": 0.0004981439438425949,
"loss": 5.6744,
"mean_token_accuracy": 0.1593330755829811,
"num_tokens": 11373413.0,
"step": 5470
},
{
"entropy": 5.726079893112183,
"epoch": 0.49664368650217705,
"grad_norm": 2.390625,
"learning_rate": 0.000498139794074707,
"loss": 5.5631,
"mean_token_accuracy": 0.1633937731385231,
"num_tokens": 11383142.0,
"step": 5475
},
{
"entropy": 5.801775550842285,
"epoch": 0.49709724238026126,
"grad_norm": 1.9921875,
"learning_rate": 0.0004981356396922313,
"loss": 5.725,
"mean_token_accuracy": 0.1524003639817238,
"num_tokens": 11394635.0,
"step": 5480
},
{
"entropy": 5.723491621017456,
"epoch": 0.4975507982583454,
"grad_norm": 2.0625,
"learning_rate": 0.0004981314806952535,
"loss": 5.605,
"mean_token_accuracy": 0.15672532916069032,
"num_tokens": 11404442.0,
"step": 5485
},
{
"entropy": 5.684481000900268,
"epoch": 0.49800435413642963,
"grad_norm": 2.015625,
"learning_rate": 0.00049812731708386,
"loss": 5.5663,
"mean_token_accuracy": 0.15349014550447465,
"num_tokens": 11414217.0,
"step": 5490
},
{
"entropy": 5.811336708068848,
"epoch": 0.4984579100145138,
"grad_norm": 1.8671875,
"learning_rate": 0.0004981231488581367,
"loss": 5.7135,
"mean_token_accuracy": 0.14815668091177941,
"num_tokens": 11423834.0,
"step": 5495
},
{
"entropy": 5.693011617660522,
"epoch": 0.498911465892598,
"grad_norm": 1.7734375,
"learning_rate": 0.0004981189760181699,
"loss": 5.5814,
"mean_token_accuracy": 0.15378208011388778,
"num_tokens": 11433613.0,
"step": 5500
},
{
"entropy": 5.705790615081787,
"epoch": 0.49936502177068215,
"grad_norm": 2.234375,
"learning_rate": 0.0004981147985640456,
"loss": 5.6571,
"mean_token_accuracy": 0.15362219214439393,
"num_tokens": 11443915.0,
"step": 5505
},
{
"entropy": 5.848033380508423,
"epoch": 0.4998185776487663,
"grad_norm": 1.78125,
"learning_rate": 0.0004981106164958506,
"loss": 5.6773,
"mean_token_accuracy": 0.14845840334892274,
"num_tokens": 11454588.0,
"step": 5510
},
{
"entropy": 5.781160163879394,
"epoch": 0.5002721335268505,
"grad_norm": 3.078125,
"learning_rate": 0.0004981064298136712,
"loss": 5.7431,
"mean_token_accuracy": 0.15234173238277435,
"num_tokens": 11465533.0,
"step": 5515
},
{
"entropy": 5.716646528244018,
"epoch": 0.5007256894049347,
"grad_norm": 1.7421875,
"learning_rate": 0.0004981022385175939,
"loss": 5.6984,
"mean_token_accuracy": 0.14256466552615166,
"num_tokens": 11475776.0,
"step": 5520
},
{
"entropy": 5.807872915267945,
"epoch": 0.5011792452830188,
"grad_norm": 2.203125,
"learning_rate": 0.0004980980426077054,
"loss": 5.6711,
"mean_token_accuracy": 0.16105100214481355,
"num_tokens": 11486200.0,
"step": 5525
},
{
"entropy": 5.822145271301269,
"epoch": 0.5016328011611031,
"grad_norm": 1.6875,
"learning_rate": 0.0004980938420840926,
"loss": 5.6859,
"mean_token_accuracy": 0.15076019763946533,
"num_tokens": 11496768.0,
"step": 5530
},
{
"entropy": 5.805276823043823,
"epoch": 0.5020863570391872,
"grad_norm": 2.046875,
"learning_rate": 0.0004980896369468423,
"loss": 5.6786,
"mean_token_accuracy": 0.1519462376832962,
"num_tokens": 11507096.0,
"step": 5535
},
{
"entropy": 5.724911499023437,
"epoch": 0.5025399129172714,
"grad_norm": 1.7265625,
"learning_rate": 0.0004980854271960415,
"loss": 5.7198,
"mean_token_accuracy": 0.14966288208961487,
"num_tokens": 11517301.0,
"step": 5540
},
{
"entropy": 5.842235517501831,
"epoch": 0.5029934687953556,
"grad_norm": 3.734375,
"learning_rate": 0.0004980812128317772,
"loss": 5.7666,
"mean_token_accuracy": 0.15556451082229614,
"num_tokens": 11528616.0,
"step": 5545
},
{
"entropy": 5.861762285232544,
"epoch": 0.5034470246734397,
"grad_norm": 2.078125,
"learning_rate": 0.0004980769938541364,
"loss": 5.6969,
"mean_token_accuracy": 0.15842931866645812,
"num_tokens": 11539344.0,
"step": 5550
},
{
"entropy": 5.806245136260986,
"epoch": 0.503900580551524,
"grad_norm": 1.8984375,
"learning_rate": 0.0004980727702632066,
"loss": 5.7601,
"mean_token_accuracy": 0.15045208632946014,
"num_tokens": 11549511.0,
"step": 5555
},
{
"entropy": 5.816326236724853,
"epoch": 0.5043541364296081,
"grad_norm": 1.71875,
"learning_rate": 0.0004980685420590752,
"loss": 5.6754,
"mean_token_accuracy": 0.1513192594051361,
"num_tokens": 11559425.0,
"step": 5560
},
{
"entropy": 5.87260160446167,
"epoch": 0.5048076923076923,
"grad_norm": 1.9375,
"learning_rate": 0.0004980643092418294,
"loss": 5.7977,
"mean_token_accuracy": 0.14498255625367165,
"num_tokens": 11570406.0,
"step": 5565
},
{
"entropy": 5.791260766983032,
"epoch": 0.5052612481857764,
"grad_norm": 1.8671875,
"learning_rate": 0.0004980600718115568,
"loss": 5.6713,
"mean_token_accuracy": 0.15540144369006156,
"num_tokens": 11580509.0,
"step": 5570
},
{
"entropy": 5.755546998977661,
"epoch": 0.5057148040638607,
"grad_norm": 1.6015625,
"learning_rate": 0.0004980558297683452,
"loss": 5.6906,
"mean_token_accuracy": 0.15466409027576447,
"num_tokens": 11590860.0,
"step": 5575
},
{
"entropy": 5.817662000656128,
"epoch": 0.5061683599419449,
"grad_norm": 2.015625,
"learning_rate": 0.0004980515831122822,
"loss": 5.698,
"mean_token_accuracy": 0.15118605718016626,
"num_tokens": 11601954.0,
"step": 5580
},
{
"entropy": 5.831102275848389,
"epoch": 0.506621915820029,
"grad_norm": 2.09375,
"learning_rate": 0.0004980473318434555,
"loss": 5.7026,
"mean_token_accuracy": 0.15453651696443557,
"num_tokens": 11612581.0,
"step": 5585
},
{
"entropy": 5.728653049468994,
"epoch": 0.5070754716981132,
"grad_norm": 1.890625,
"learning_rate": 0.0004980430759619532,
"loss": 5.6293,
"mean_token_accuracy": 0.15604719668626785,
"num_tokens": 11622599.0,
"step": 5590
},
{
"entropy": 5.730202388763428,
"epoch": 0.5075290275761973,
"grad_norm": 1.859375,
"learning_rate": 0.0004980388154678633,
"loss": 5.6586,
"mean_token_accuracy": 0.1495005428791046,
"num_tokens": 11633611.0,
"step": 5595
},
{
"entropy": 5.6968724727630615,
"epoch": 0.5079825834542816,
"grad_norm": 2.296875,
"learning_rate": 0.0004980345503612738,
"loss": 5.6107,
"mean_token_accuracy": 0.1538061574101448,
"num_tokens": 11643159.0,
"step": 5600
},
{
"entropy": 5.753897953033447,
"epoch": 0.5084361393323658,
"grad_norm": 2.1875,
"learning_rate": 0.0004980302806422729,
"loss": 5.6204,
"mean_token_accuracy": 0.15607504695653915,
"num_tokens": 11653174.0,
"step": 5605
},
{
"entropy": 5.755003118515015,
"epoch": 0.5088896952104499,
"grad_norm": 1.7109375,
"learning_rate": 0.0004980260063109491,
"loss": 5.6672,
"mean_token_accuracy": 0.1535380646586418,
"num_tokens": 11662621.0,
"step": 5610
},
{
"entropy": 5.701317262649536,
"epoch": 0.5093432510885341,
"grad_norm": 2.171875,
"learning_rate": 0.0004980217273673906,
"loss": 5.6218,
"mean_token_accuracy": 0.15659913867712022,
"num_tokens": 11673538.0,
"step": 5615
},
{
"entropy": 5.802594709396362,
"epoch": 0.5097968069666183,
"grad_norm": 1.7265625,
"learning_rate": 0.0004980174438116859,
"loss": 5.6228,
"mean_token_accuracy": 0.16106417030096054,
"num_tokens": 11683203.0,
"step": 5620
},
{
"entropy": 5.706654167175293,
"epoch": 0.5102503628447025,
"grad_norm": 2.359375,
"learning_rate": 0.0004980131556439236,
"loss": 5.6376,
"mean_token_accuracy": 0.15619875639677047,
"num_tokens": 11693313.0,
"step": 5625
},
{
"entropy": 5.769161176681519,
"epoch": 0.5107039187227866,
"grad_norm": 1.9921875,
"learning_rate": 0.0004980088628641925,
"loss": 5.7332,
"mean_token_accuracy": 0.15326391980051995,
"num_tokens": 11704194.0,
"step": 5630
},
{
"entropy": 5.775632333755493,
"epoch": 0.5111574746008708,
"grad_norm": 2.328125,
"learning_rate": 0.0004980045654725812,
"loss": 5.5708,
"mean_token_accuracy": 0.15834364145994187,
"num_tokens": 11715109.0,
"step": 5635
},
{
"entropy": 5.664262056350708,
"epoch": 0.511611030478955,
"grad_norm": 1.59375,
"learning_rate": 0.0004980002634691787,
"loss": 5.6186,
"mean_token_accuracy": 0.16121489852666854,
"num_tokens": 11724961.0,
"step": 5640
},
{
"entropy": 5.703507709503174,
"epoch": 0.5120645863570392,
"grad_norm": 2.296875,
"learning_rate": 0.000497995956854074,
"loss": 5.6411,
"mean_token_accuracy": 0.15814631804823875,
"num_tokens": 11735334.0,
"step": 5645
},
{
"entropy": 5.810642290115356,
"epoch": 0.5125181422351234,
"grad_norm": 1.7265625,
"learning_rate": 0.0004979916456273561,
"loss": 5.6458,
"mean_token_accuracy": 0.14870579093694686,
"num_tokens": 11746858.0,
"step": 5650
},
{
"entropy": 5.807532930374146,
"epoch": 0.5129716981132075,
"grad_norm": 1.59375,
"learning_rate": 0.0004979873297891141,
"loss": 5.7475,
"mean_token_accuracy": 0.14519907236099244,
"num_tokens": 11757819.0,
"step": 5655
},
{
"entropy": 5.808407640457153,
"epoch": 0.5134252539912917,
"grad_norm": 1.6328125,
"learning_rate": 0.0004979830093394372,
"loss": 5.6646,
"mean_token_accuracy": 0.15376524329185487,
"num_tokens": 11767470.0,
"step": 5660
},
{
"entropy": 5.742450141906739,
"epoch": 0.513878809869376,
"grad_norm": 2.109375,
"learning_rate": 0.0004979786842784149,
"loss": 5.6069,
"mean_token_accuracy": 0.1618952751159668,
"num_tokens": 11777060.0,
"step": 5665
},
{
"entropy": 5.6815831661224365,
"epoch": 0.5143323657474601,
"grad_norm": 2.03125,
"learning_rate": 0.0004979743546061366,
"loss": 5.6476,
"mean_token_accuracy": 0.15582273453474044,
"num_tokens": 11787176.0,
"step": 5670
},
{
"entropy": 5.681001758575439,
"epoch": 0.5147859216255443,
"grad_norm": 1.7734375,
"learning_rate": 0.0004979700203226918,
"loss": 5.5463,
"mean_token_accuracy": 0.15773169249296187,
"num_tokens": 11796419.0,
"step": 5675
},
{
"entropy": 5.835689210891724,
"epoch": 0.5152394775036284,
"grad_norm": 2.921875,
"learning_rate": 0.0004979656814281702,
"loss": 5.752,
"mean_token_accuracy": 0.1510649248957634,
"num_tokens": 11807279.0,
"step": 5680
},
{
"entropy": 5.757753849029541,
"epoch": 0.5156930333817126,
"grad_norm": 1.7265625,
"learning_rate": 0.0004979613379226615,
"loss": 5.6502,
"mean_token_accuracy": 0.1518499568104744,
"num_tokens": 11817669.0,
"step": 5685
},
{
"entropy": 5.725679922103882,
"epoch": 0.5161465892597968,
"grad_norm": 1.7109375,
"learning_rate": 0.0004979569898062554,
"loss": 5.6616,
"mean_token_accuracy": 0.15862690210342406,
"num_tokens": 11827247.0,
"step": 5690
},
{
"entropy": 5.789619445800781,
"epoch": 0.516600145137881,
"grad_norm": 1.8671875,
"learning_rate": 0.000497952637079042,
"loss": 5.6487,
"mean_token_accuracy": 0.14972210973501204,
"num_tokens": 11838201.0,
"step": 5695
},
{
"entropy": 5.730072736740112,
"epoch": 0.5170537010159652,
"grad_norm": 1.859375,
"learning_rate": 0.0004979482797411114,
"loss": 5.593,
"mean_token_accuracy": 0.16091672480106353,
"num_tokens": 11847561.0,
"step": 5700
},
{
"entropy": 5.738336229324341,
"epoch": 0.5175072568940493,
"grad_norm": 1.75,
"learning_rate": 0.0004979439177925534,
"loss": 5.714,
"mean_token_accuracy": 0.15794804096221923,
"num_tokens": 11858272.0,
"step": 5705
},
{
"entropy": 5.827757883071899,
"epoch": 0.5179608127721336,
"grad_norm": 2.96875,
"learning_rate": 0.0004979395512334584,
"loss": 5.7507,
"mean_token_accuracy": 0.15172273367643357,
"num_tokens": 11869660.0,
"step": 5710
},
{
"entropy": 5.86238169670105,
"epoch": 0.5184143686502177,
"grad_norm": 2.015625,
"learning_rate": 0.0004979351800639166,
"loss": 5.7652,
"mean_token_accuracy": 0.14836279600858687,
"num_tokens": 11880239.0,
"step": 5715
},
{
"entropy": 5.888429594039917,
"epoch": 0.5188679245283019,
"grad_norm": 1.9375,
"learning_rate": 0.0004979308042840186,
"loss": 5.7581,
"mean_token_accuracy": 0.14716654121875763,
"num_tokens": 11890630.0,
"step": 5720
},
{
"entropy": 5.802192068099975,
"epoch": 0.519321480406386,
"grad_norm": 1.8359375,
"learning_rate": 0.0004979264238938547,
"loss": 5.7882,
"mean_token_accuracy": 0.14498534202575683,
"num_tokens": 11900948.0,
"step": 5725
},
{
"entropy": 5.752587032318115,
"epoch": 0.5197750362844702,
"grad_norm": 1.9453125,
"learning_rate": 0.0004979220388935155,
"loss": 5.6749,
"mean_token_accuracy": 0.15167703554034234,
"num_tokens": 11912528.0,
"step": 5730
},
{
"entropy": 5.756484889984131,
"epoch": 0.5202285921625545,
"grad_norm": 2.5,
"learning_rate": 0.0004979176492830918,
"loss": 5.6087,
"mean_token_accuracy": 0.15607684403657912,
"num_tokens": 11922211.0,
"step": 5735
},
{
"entropy": 5.700596475601197,
"epoch": 0.5206821480406386,
"grad_norm": 2.421875,
"learning_rate": 0.0004979132550626742,
"loss": 5.6598,
"mean_token_accuracy": 0.14853055775165558,
"num_tokens": 11932520.0,
"step": 5740
},
{
"entropy": 5.7969828128814695,
"epoch": 0.5211357039187228,
"grad_norm": 2.34375,
"learning_rate": 0.0004979088562323538,
"loss": 5.5931,
"mean_token_accuracy": 0.15993748605251312,
"num_tokens": 11943312.0,
"step": 5745
},
{
"entropy": 5.753064203262329,
"epoch": 0.5215892597968069,
"grad_norm": 2.296875,
"learning_rate": 0.0004979044527922214,
"loss": 5.7227,
"mean_token_accuracy": 0.1540391340851784,
"num_tokens": 11954455.0,
"step": 5750
},
{
"entropy": 5.7552355289459225,
"epoch": 0.5220428156748912,
"grad_norm": 2.0625,
"learning_rate": 0.0004979000447423681,
"loss": 5.6357,
"mean_token_accuracy": 0.15711707174777984,
"num_tokens": 11964650.0,
"step": 5755
},
{
"entropy": 5.757721376419068,
"epoch": 0.5224963715529753,
"grad_norm": 1.9609375,
"learning_rate": 0.0004978956320828851,
"loss": 5.658,
"mean_token_accuracy": 0.15847069025039673,
"num_tokens": 11975134.0,
"step": 5760
},
{
"entropy": 5.783113622665406,
"epoch": 0.5229499274310595,
"grad_norm": 1.5078125,
"learning_rate": 0.0004978912148138637,
"loss": 5.699,
"mean_token_accuracy": 0.1517071783542633,
"num_tokens": 11986486.0,
"step": 5765
},
{
"entropy": 5.8543494701385494,
"epoch": 0.5234034833091437,
"grad_norm": 1.6015625,
"learning_rate": 0.000497886792935395,
"loss": 5.735,
"mean_token_accuracy": 0.15076159983873366,
"num_tokens": 11998045.0,
"step": 5770
},
{
"entropy": 5.86229944229126,
"epoch": 0.5238570391872278,
"grad_norm": 1.671875,
"learning_rate": 0.0004978823664475707,
"loss": 5.7001,
"mean_token_accuracy": 0.15565317273139953,
"num_tokens": 12008130.0,
"step": 5775
},
{
"entropy": 5.790041208267212,
"epoch": 0.5243105950653121,
"grad_norm": 2.0625,
"learning_rate": 0.0004978779353504824,
"loss": 5.7661,
"mean_token_accuracy": 0.14593174383044244,
"num_tokens": 12018753.0,
"step": 5780
},
{
"entropy": 5.795868539810181,
"epoch": 0.5247641509433962,
"grad_norm": 2.234375,
"learning_rate": 0.0004978734996442214,
"loss": 5.6456,
"mean_token_accuracy": 0.16076426655054094,
"num_tokens": 12028202.0,
"step": 5785
},
{
"entropy": 5.814069175720215,
"epoch": 0.5252177068214804,
"grad_norm": 1.796875,
"learning_rate": 0.0004978690593288798,
"loss": 5.7247,
"mean_token_accuracy": 0.15733310282230378,
"num_tokens": 12039652.0,
"step": 5790
},
{
"entropy": 5.789728403091431,
"epoch": 0.5256712626995645,
"grad_norm": 17.0,
"learning_rate": 0.0004978646144045491,
"loss": 5.6497,
"mean_token_accuracy": 0.15042980760335922,
"num_tokens": 12051269.0,
"step": 5795
},
{
"entropy": 5.644839191436768,
"epoch": 0.5261248185776488,
"grad_norm": 2.296875,
"learning_rate": 0.0004978601648713215,
"loss": 5.5269,
"mean_token_accuracy": 0.15518265813589097,
"num_tokens": 12060647.0,
"step": 5800
},
{
"entropy": 5.72273964881897,
"epoch": 0.526578374455733,
"grad_norm": 2.421875,
"learning_rate": 0.0004978557107292889,
"loss": 5.691,
"mean_token_accuracy": 0.15798260569572448,
"num_tokens": 12071669.0,
"step": 5805
},
{
"entropy": 5.710134744644165,
"epoch": 0.5270319303338171,
"grad_norm": 1.671875,
"learning_rate": 0.0004978512519785433,
"loss": 5.5538,
"mean_token_accuracy": 0.1603996381163597,
"num_tokens": 12081971.0,
"step": 5810
},
{
"entropy": 5.761967611312866,
"epoch": 0.5274854862119013,
"grad_norm": 2.203125,
"learning_rate": 0.000497846788619177,
"loss": 5.693,
"mean_token_accuracy": 0.14980525076389312,
"num_tokens": 12092465.0,
"step": 5815
},
{
"entropy": 5.783017539978028,
"epoch": 0.5279390420899854,
"grad_norm": 2.359375,
"learning_rate": 0.0004978423206512825,
"loss": 5.6323,
"mean_token_accuracy": 0.1552136331796646,
"num_tokens": 12101752.0,
"step": 5820
},
{
"entropy": 5.815354442596435,
"epoch": 0.5283925979680697,
"grad_norm": 2.796875,
"learning_rate": 0.000497837848074952,
"loss": 5.7277,
"mean_token_accuracy": 0.15265906900167464,
"num_tokens": 12111805.0,
"step": 5825
},
{
"entropy": 5.818612146377563,
"epoch": 0.5288461538461539,
"grad_norm": 3.046875,
"learning_rate": 0.000497833370890278,
"loss": 5.7416,
"mean_token_accuracy": 0.1517587274312973,
"num_tokens": 12122511.0,
"step": 5830
},
{
"entropy": 5.808302402496338,
"epoch": 0.529299709724238,
"grad_norm": 3.328125,
"learning_rate": 0.0004978288890973529,
"loss": 5.6312,
"mean_token_accuracy": 0.1537729099392891,
"num_tokens": 12132510.0,
"step": 5835
},
{
"entropy": 5.739053058624267,
"epoch": 0.5297532656023222,
"grad_norm": 2.4375,
"learning_rate": 0.0004978244026962698,
"loss": 5.6233,
"mean_token_accuracy": 0.15461270958185197,
"num_tokens": 12142419.0,
"step": 5840
},
{
"entropy": 5.757953739166259,
"epoch": 0.5302068214804064,
"grad_norm": 5.9375,
"learning_rate": 0.0004978199116871212,
"loss": 5.6917,
"mean_token_accuracy": 0.1523043468594551,
"num_tokens": 12152759.0,
"step": 5845
},
{
"entropy": 5.741458559036255,
"epoch": 0.5306603773584906,
"grad_norm": 2.703125,
"learning_rate": 0.00049781541607,
"loss": 5.6286,
"mean_token_accuracy": 0.15267193913459778,
"num_tokens": 12162758.0,
"step": 5850
},
{
"entropy": 5.783251619338989,
"epoch": 0.5311139332365747,
"grad_norm": 2.40625,
"learning_rate": 0.0004978109158449993,
"loss": 5.6838,
"mean_token_accuracy": 0.15076133981347084,
"num_tokens": 12172664.0,
"step": 5855
},
{
"entropy": 5.809034013748169,
"epoch": 0.5315674891146589,
"grad_norm": 2.421875,
"learning_rate": 0.0004978064110122121,
"loss": 5.65,
"mean_token_accuracy": 0.1549011915922165,
"num_tokens": 12182028.0,
"step": 5860
},
{
"entropy": 5.824132633209229,
"epoch": 0.532021044992743,
"grad_norm": 1.9609375,
"learning_rate": 0.0004978019015717314,
"loss": 5.6972,
"mean_token_accuracy": 0.15745327472686768,
"num_tokens": 12192168.0,
"step": 5865
},
{
"entropy": 5.67472128868103,
"epoch": 0.5324746008708273,
"grad_norm": 2.1875,
"learning_rate": 0.0004977973875236508,
"loss": 5.5932,
"mean_token_accuracy": 0.15128588378429414,
"num_tokens": 12202889.0,
"step": 5870
},
{
"entropy": 5.71777195930481,
"epoch": 0.5329281567489115,
"grad_norm": 2.15625,
"learning_rate": 0.0004977928688680634,
"loss": 5.6614,
"mean_token_accuracy": 0.1567210003733635,
"num_tokens": 12212514.0,
"step": 5875
},
{
"entropy": 5.755164003372192,
"epoch": 0.5333817126269956,
"grad_norm": 2.109375,
"learning_rate": 0.0004977883456050626,
"loss": 5.5106,
"mean_token_accuracy": 0.1608267605304718,
"num_tokens": 12222353.0,
"step": 5880
},
{
"entropy": 5.838532447814941,
"epoch": 0.5338352685050798,
"grad_norm": 1.9921875,
"learning_rate": 0.000497783817734742,
"loss": 5.7678,
"mean_token_accuracy": 0.1494266524910927,
"num_tokens": 12232845.0,
"step": 5885
},
{
"entropy": 5.804184675216675,
"epoch": 0.534288824383164,
"grad_norm": 3.0625,
"learning_rate": 0.0004977792852571954,
"loss": 5.726,
"mean_token_accuracy": 0.1480010524392128,
"num_tokens": 12242950.0,
"step": 5890
},
{
"entropy": 5.798137092590332,
"epoch": 0.5347423802612482,
"grad_norm": 2.109375,
"learning_rate": 0.0004977747481725164,
"loss": 5.702,
"mean_token_accuracy": 0.14956545680761338,
"num_tokens": 12253132.0,
"step": 5895
},
{
"entropy": 5.74386191368103,
"epoch": 0.5351959361393324,
"grad_norm": 2.515625,
"learning_rate": 0.0004977702064807987,
"loss": 5.6457,
"mean_token_accuracy": 0.15413923114538192,
"num_tokens": 12263214.0,
"step": 5900
},
{
"entropy": 5.707874059677124,
"epoch": 0.5356494920174165,
"grad_norm": 1.9375,
"learning_rate": 0.0004977656601821366,
"loss": 5.5845,
"mean_token_accuracy": 0.16530820578336716,
"num_tokens": 12273250.0,
"step": 5905
},
{
"entropy": 5.919317960739136,
"epoch": 0.5361030478955007,
"grad_norm": 2.359375,
"learning_rate": 0.0004977611092766238,
"loss": 5.7643,
"mean_token_accuracy": 0.15915993154048919,
"num_tokens": 12283767.0,
"step": 5910
},
{
"entropy": 5.754854917526245,
"epoch": 0.5365566037735849,
"grad_norm": 2.546875,
"learning_rate": 0.0004977565537643544,
"loss": 5.619,
"mean_token_accuracy": 0.1566011890769005,
"num_tokens": 12294612.0,
"step": 5915
},
{
"entropy": 5.735382413864135,
"epoch": 0.5370101596516691,
"grad_norm": 2.546875,
"learning_rate": 0.0004977519936454227,
"loss": 5.7207,
"mean_token_accuracy": 0.1573534607887268,
"num_tokens": 12305614.0,
"step": 5920
},
{
"entropy": 5.734095716476441,
"epoch": 0.5374637155297532,
"grad_norm": 2.671875,
"learning_rate": 0.0004977474289199231,
"loss": 5.5688,
"mean_token_accuracy": 0.15029709711670874,
"num_tokens": 12316152.0,
"step": 5925
},
{
"entropy": 5.705564880371094,
"epoch": 0.5379172714078374,
"grad_norm": 2.640625,
"learning_rate": 0.0004977428595879499,
"loss": 5.6581,
"mean_token_accuracy": 0.15709878951311113,
"num_tokens": 12325940.0,
"step": 5930
},
{
"entropy": 5.736659002304077,
"epoch": 0.5383708272859217,
"grad_norm": 2.9375,
"learning_rate": 0.0004977382856495975,
"loss": 5.6538,
"mean_token_accuracy": 0.1553759306669235,
"num_tokens": 12335482.0,
"step": 5935
},
{
"entropy": 5.8129387378692625,
"epoch": 0.5388243831640058,
"grad_norm": 2.328125,
"learning_rate": 0.0004977337071049606,
"loss": 5.7538,
"mean_token_accuracy": 0.1464443251490593,
"num_tokens": 12346386.0,
"step": 5940
},
{
"entropy": 5.854721212387085,
"epoch": 0.53927793904209,
"grad_norm": 2.375,
"learning_rate": 0.000497729123954134,
"loss": 5.6683,
"mean_token_accuracy": 0.15644141435623168,
"num_tokens": 12356282.0,
"step": 5945
},
{
"entropy": 5.735949468612671,
"epoch": 0.5397314949201741,
"grad_norm": 2.265625,
"learning_rate": 0.0004977245361972123,
"loss": 5.6765,
"mean_token_accuracy": 0.1545313149690628,
"num_tokens": 12366994.0,
"step": 5950
},
{
"entropy": 5.7212425708770756,
"epoch": 0.5401850507982583,
"grad_norm": 1.8984375,
"learning_rate": 0.0004977199438342904,
"loss": 5.5716,
"mean_token_accuracy": 0.15779082775115966,
"num_tokens": 12377524.0,
"step": 5955
},
{
"entropy": 5.761970043182373,
"epoch": 0.5406386066763426,
"grad_norm": 1.7578125,
"learning_rate": 0.0004977153468654634,
"loss": 5.7211,
"mean_token_accuracy": 0.15492843165993692,
"num_tokens": 12387593.0,
"step": 5960
},
{
"entropy": 5.80047345161438,
"epoch": 0.5410921625544267,
"grad_norm": 2.609375,
"learning_rate": 0.0004977107452908262,
"loss": 5.6901,
"mean_token_accuracy": 0.15060260742902756,
"num_tokens": 12398732.0,
"step": 5965
},
{
"entropy": 5.821221828460693,
"epoch": 0.5415457184325109,
"grad_norm": 1.8828125,
"learning_rate": 0.000497706139110474,
"loss": 5.6681,
"mean_token_accuracy": 0.16170208156108856,
"num_tokens": 12409821.0,
"step": 5970
},
{
"entropy": 5.776604270935058,
"epoch": 0.541999274310595,
"grad_norm": 1.9140625,
"learning_rate": 0.0004977015283245021,
"loss": 5.6963,
"mean_token_accuracy": 0.15199582576751708,
"num_tokens": 12420044.0,
"step": 5975
},
{
"entropy": 5.716190004348755,
"epoch": 0.5424528301886793,
"grad_norm": 5.3125,
"learning_rate": 0.0004976969129330059,
"loss": 5.6405,
"mean_token_accuracy": 0.15238496512174607,
"num_tokens": 12430980.0,
"step": 5980
},
{
"entropy": 5.807556629180908,
"epoch": 0.5429063860667634,
"grad_norm": 2.046875,
"learning_rate": 0.0004976922929360808,
"loss": 5.6766,
"mean_token_accuracy": 0.15749867260456085,
"num_tokens": 12441999.0,
"step": 5985
},
{
"entropy": 5.795617580413818,
"epoch": 0.5433599419448476,
"grad_norm": 2.34375,
"learning_rate": 0.0004976876683338222,
"loss": 5.7151,
"mean_token_accuracy": 0.15440924018621444,
"num_tokens": 12452837.0,
"step": 5990
},
{
"entropy": 5.781675815582275,
"epoch": 0.5438134978229318,
"grad_norm": 2.125,
"learning_rate": 0.000497683039126326,
"loss": 5.6243,
"mean_token_accuracy": 0.16212155669927597,
"num_tokens": 12462142.0,
"step": 5995
},
{
"entropy": 5.757653427124024,
"epoch": 0.5442670537010159,
"grad_norm": 4.21875,
"learning_rate": 0.0004976784053136878,
"loss": 5.6862,
"mean_token_accuracy": 0.15880632698535918,
"num_tokens": 12472756.0,
"step": 6000
},
{
"epoch": 0.5442670537010159,
"eval_entropy": 5.571439638612497,
"eval_loss": 5.7100067138671875,
"eval_mean_token_accuracy": 0.15796612128615378,
"eval_num_tokens": 12472756.0,
"eval_runtime": 26.5304,
"eval_samples_per_second": 1332.812,
"eval_steps_per_second": 166.602,
"step": 6000
},
{
"entropy": 5.8166303634643555,
"epoch": 0.5447206095791002,
"grad_norm": 2.90625,
"learning_rate": 0.0004976737668960034,
"loss": 5.7832,
"mean_token_accuracy": 0.14682883471250535,
"num_tokens": 12483115.0,
"step": 6005
},
{
"entropy": 5.832898235321045,
"epoch": 0.5451741654571843,
"grad_norm": 3.28125,
"learning_rate": 0.0004976691238733688,
"loss": 5.6337,
"mean_token_accuracy": 0.15088710635900499,
"num_tokens": 12493280.0,
"step": 6010
},
{
"entropy": 5.788230085372925,
"epoch": 0.5456277213352685,
"grad_norm": 1.8984375,
"learning_rate": 0.00049766447624588,
"loss": 5.7572,
"mean_token_accuracy": 0.1538712576031685,
"num_tokens": 12503665.0,
"step": 6015
},
{
"entropy": 5.701073551177979,
"epoch": 0.5460812772133526,
"grad_norm": 2.15625,
"learning_rate": 0.000497659824013633,
"loss": 5.688,
"mean_token_accuracy": 0.14975677132606507,
"num_tokens": 12513381.0,
"step": 6020
},
{
"entropy": 5.85684871673584,
"epoch": 0.5465348330914369,
"grad_norm": 2.890625,
"learning_rate": 0.0004976551671767241,
"loss": 5.7085,
"mean_token_accuracy": 0.15038748383522033,
"num_tokens": 12524121.0,
"step": 6025
},
{
"entropy": 5.8416108131408695,
"epoch": 0.5469883889695211,
"grad_norm": 2.296875,
"learning_rate": 0.0004976505057352496,
"loss": 5.6631,
"mean_token_accuracy": 0.15823794305324554,
"num_tokens": 12533972.0,
"step": 6030
},
{
"entropy": 5.784581518173217,
"epoch": 0.5474419448476052,
"grad_norm": 2.296875,
"learning_rate": 0.0004976458396893059,
"loss": 5.7891,
"mean_token_accuracy": 0.1468980848789215,
"num_tokens": 12544110.0,
"step": 6035
},
{
"entropy": 5.758417320251465,
"epoch": 0.5478955007256894,
"grad_norm": 2.09375,
"learning_rate": 0.0004976411690389896,
"loss": 5.6342,
"mean_token_accuracy": 0.15048055052757264,
"num_tokens": 12554273.0,
"step": 6040
},
{
"entropy": 5.893464517593384,
"epoch": 0.5483490566037735,
"grad_norm": 2.78125,
"learning_rate": 0.000497636493784397,
"loss": 5.7141,
"mean_token_accuracy": 0.15111250206828117,
"num_tokens": 12564633.0,
"step": 6045
},
{
"entropy": 5.845950984954834,
"epoch": 0.5488026124818578,
"grad_norm": 1.6640625,
"learning_rate": 0.0004976318139256251,
"loss": 5.6268,
"mean_token_accuracy": 0.1532691612839699,
"num_tokens": 12573790.0,
"step": 6050
},
{
"entropy": 5.727607202529907,
"epoch": 0.549256168359942,
"grad_norm": 2.796875,
"learning_rate": 0.0004976271294627706,
"loss": 5.702,
"mean_token_accuracy": 0.15061550587415695,
"num_tokens": 12584265.0,
"step": 6055
},
{
"entropy": 5.711427831649781,
"epoch": 0.5497097242380261,
"grad_norm": 2.703125,
"learning_rate": 0.0004976224403959303,
"loss": 5.5784,
"mean_token_accuracy": 0.15742302387952806,
"num_tokens": 12594413.0,
"step": 6060
},
{
"entropy": 5.788397359848022,
"epoch": 0.5501632801161103,
"grad_norm": 2.6875,
"learning_rate": 0.0004976177467252012,
"loss": 5.6608,
"mean_token_accuracy": 0.15904350876808165,
"num_tokens": 12604637.0,
"step": 6065
},
{
"entropy": 5.771824026107788,
"epoch": 0.5506168359941945,
"grad_norm": 2.75,
"learning_rate": 0.0004976130484506804,
"loss": 5.7192,
"mean_token_accuracy": 0.15422330498695375,
"num_tokens": 12614873.0,
"step": 6070
},
{
"entropy": 5.745659923553466,
"epoch": 0.5510703918722787,
"grad_norm": 2.046875,
"learning_rate": 0.0004976083455724649,
"loss": 5.6623,
"mean_token_accuracy": 0.15023809373378755,
"num_tokens": 12626182.0,
"step": 6075
},
{
"entropy": 5.757641315460205,
"epoch": 0.5515239477503628,
"grad_norm": 2.015625,
"learning_rate": 0.0004976036380906523,
"loss": 5.5818,
"mean_token_accuracy": 0.15435850918292998,
"num_tokens": 12635418.0,
"step": 6080
},
{
"entropy": 5.838926458358765,
"epoch": 0.551977503628447,
"grad_norm": 1.859375,
"learning_rate": 0.0004975989260053395,
"loss": 5.7269,
"mean_token_accuracy": 0.1468985289335251,
"num_tokens": 12645942.0,
"step": 6085
},
{
"entropy": 5.728414392471313,
"epoch": 0.5524310595065312,
"grad_norm": 2.1875,
"learning_rate": 0.0004975942093166242,
"loss": 5.6659,
"mean_token_accuracy": 0.15283730030059814,
"num_tokens": 12656460.0,
"step": 6090
},
{
"entropy": 5.809977293014526,
"epoch": 0.5528846153846154,
"grad_norm": 3.1875,
"learning_rate": 0.0004975894880246041,
"loss": 5.6028,
"mean_token_accuracy": 0.15169214606285095,
"num_tokens": 12666916.0,
"step": 6095
},
{
"entropy": 5.775780582427979,
"epoch": 0.5533381712626996,
"grad_norm": 1.828125,
"learning_rate": 0.0004975847621293766,
"loss": 5.6709,
"mean_token_accuracy": 0.16105171889066697,
"num_tokens": 12677562.0,
"step": 6100
},
{
"entropy": 5.711263704299927,
"epoch": 0.5537917271407837,
"grad_norm": 2.203125,
"learning_rate": 0.0004975800316310394,
"loss": 5.6235,
"mean_token_accuracy": 0.15581511408090593,
"num_tokens": 12687349.0,
"step": 6105
},
{
"entropy": 5.806702566146851,
"epoch": 0.5542452830188679,
"grad_norm": 4.25,
"learning_rate": 0.0004975752965296906,
"loss": 5.6815,
"mean_token_accuracy": 0.1545094668865204,
"num_tokens": 12697850.0,
"step": 6110
},
{
"entropy": 5.724259376525879,
"epoch": 0.5546988388969522,
"grad_norm": 6.9375,
"learning_rate": 0.0004975705568254278,
"loss": 5.5467,
"mean_token_accuracy": 0.15480877012014388,
"num_tokens": 12708283.0,
"step": 6115
},
{
"entropy": 5.708226060867309,
"epoch": 0.5551523947750363,
"grad_norm": 2.703125,
"learning_rate": 0.0004975658125183493,
"loss": 5.6221,
"mean_token_accuracy": 0.1571838840842247,
"num_tokens": 12718201.0,
"step": 6120
},
{
"entropy": 5.745707035064697,
"epoch": 0.5556059506531205,
"grad_norm": 2.15625,
"learning_rate": 0.000497561063608553,
"loss": 5.6058,
"mean_token_accuracy": 0.15703478157520295,
"num_tokens": 12728860.0,
"step": 6125
},
{
"entropy": 5.779846572875977,
"epoch": 0.5560595065312046,
"grad_norm": 3.046875,
"learning_rate": 0.0004975563100961372,
"loss": 5.6533,
"mean_token_accuracy": 0.15242056548595428,
"num_tokens": 12739007.0,
"step": 6130
},
{
"entropy": 5.750319910049439,
"epoch": 0.5565130624092888,
"grad_norm": 2.53125,
"learning_rate": 0.0004975515519812002,
"loss": 5.5186,
"mean_token_accuracy": 0.16321926414966584,
"num_tokens": 12749004.0,
"step": 6135
},
{
"entropy": 5.690269613265992,
"epoch": 0.556966618287373,
"grad_norm": 2.140625,
"learning_rate": 0.0004975467892638402,
"loss": 5.6845,
"mean_token_accuracy": 0.1539756253361702,
"num_tokens": 12760085.0,
"step": 6140
},
{
"entropy": 5.728693866729737,
"epoch": 0.5574201741654572,
"grad_norm": 2.03125,
"learning_rate": 0.0004975420219441561,
"loss": 5.6269,
"mean_token_accuracy": 0.16305172443389893,
"num_tokens": 12770303.0,
"step": 6145
},
{
"entropy": 5.800379467010498,
"epoch": 0.5578737300435413,
"grad_norm": 2.203125,
"learning_rate": 0.0004975372500222462,
"loss": 5.638,
"mean_token_accuracy": 0.15569627434015273,
"num_tokens": 12780334.0,
"step": 6150
},
{
"entropy": 5.805584144592285,
"epoch": 0.5583272859216255,
"grad_norm": 2.9375,
"learning_rate": 0.0004975324734982094,
"loss": 5.6992,
"mean_token_accuracy": 0.1512184336781502,
"num_tokens": 12790980.0,
"step": 6155
},
{
"entropy": 5.7633744239807125,
"epoch": 0.5587808417997098,
"grad_norm": 2.03125,
"learning_rate": 0.0004975276923721442,
"loss": 5.7373,
"mean_token_accuracy": 0.15086752325296401,
"num_tokens": 12801713.0,
"step": 6160
},
{
"entropy": 5.746546363830566,
"epoch": 0.5592343976777939,
"grad_norm": 2.34375,
"learning_rate": 0.0004975229066441496,
"loss": 5.7356,
"mean_token_accuracy": 0.14781113117933273,
"num_tokens": 12813040.0,
"step": 6165
},
{
"entropy": 5.787732267379761,
"epoch": 0.5596879535558781,
"grad_norm": 2.296875,
"learning_rate": 0.0004975181163143247,
"loss": 5.6784,
"mean_token_accuracy": 0.1479372873902321,
"num_tokens": 12824048.0,
"step": 6170
},
{
"entropy": 5.804344987869262,
"epoch": 0.5601415094339622,
"grad_norm": 3.625,
"learning_rate": 0.0004975133213827684,
"loss": 5.6157,
"mean_token_accuracy": 0.1567558988928795,
"num_tokens": 12834297.0,
"step": 6175
},
{
"entropy": 5.892291069030762,
"epoch": 0.5605950653120464,
"grad_norm": 2.234375,
"learning_rate": 0.0004975085218495798,
"loss": 5.7061,
"mean_token_accuracy": 0.14884814321994783,
"num_tokens": 12844574.0,
"step": 6180
},
{
"entropy": 5.689108371734619,
"epoch": 0.5610486211901307,
"grad_norm": 2.140625,
"learning_rate": 0.0004975037177148585,
"loss": 5.6468,
"mean_token_accuracy": 0.15237224772572516,
"num_tokens": 12854670.0,
"step": 6185
},
{
"entropy": 5.645991802215576,
"epoch": 0.5615021770682148,
"grad_norm": 3.34375,
"learning_rate": 0.0004974989089787035,
"loss": 5.6568,
"mean_token_accuracy": 0.15477271229028702,
"num_tokens": 12865281.0,
"step": 6190
},
{
"entropy": 5.790045928955078,
"epoch": 0.561955732946299,
"grad_norm": 2.109375,
"learning_rate": 0.0004974940956412143,
"loss": 5.6569,
"mean_token_accuracy": 0.15305998921394348,
"num_tokens": 12875222.0,
"step": 6195
},
{
"entropy": 5.919342136383056,
"epoch": 0.5624092888243831,
"grad_norm": 9.1875,
"learning_rate": 0.0004974892777024905,
"loss": 5.7241,
"mean_token_accuracy": 0.14940885305404664,
"num_tokens": 12885079.0,
"step": 6200
},
{
"entropy": 5.777828359603882,
"epoch": 0.5628628447024674,
"grad_norm": 3.171875,
"learning_rate": 0.0004974844551626318,
"loss": 5.687,
"mean_token_accuracy": 0.1558162033557892,
"num_tokens": 12896402.0,
"step": 6205
},
{
"entropy": 5.709645462036133,
"epoch": 0.5633164005805515,
"grad_norm": 1.921875,
"learning_rate": 0.0004974796280217379,
"loss": 5.6453,
"mean_token_accuracy": 0.16267076134681702,
"num_tokens": 12906403.0,
"step": 6210
},
{
"entropy": 5.809755325317383,
"epoch": 0.5637699564586357,
"grad_norm": 1.96875,
"learning_rate": 0.0004974747962799086,
"loss": 5.6645,
"mean_token_accuracy": 0.1476689323782921,
"num_tokens": 12917061.0,
"step": 6215
},
{
"entropy": 5.737009000778198,
"epoch": 0.5642235123367199,
"grad_norm": 2.078125,
"learning_rate": 0.0004974699599372437,
"loss": 5.6883,
"mean_token_accuracy": 0.1575671598315239,
"num_tokens": 12927182.0,
"step": 6220
},
{
"entropy": 5.730798959732056,
"epoch": 0.564677068214804,
"grad_norm": 2.265625,
"learning_rate": 0.0004974651189938434,
"loss": 5.6492,
"mean_token_accuracy": 0.15385740101337433,
"num_tokens": 12937371.0,
"step": 6225
},
{
"entropy": 5.819814729690552,
"epoch": 0.5651306240928883,
"grad_norm": 1.9375,
"learning_rate": 0.0004974602734498079,
"loss": 5.6974,
"mean_token_accuracy": 0.147367987036705,
"num_tokens": 12948569.0,
"step": 6230
},
{
"entropy": 5.849042892456055,
"epoch": 0.5655841799709724,
"grad_norm": 2.90625,
"learning_rate": 0.000497455423305237,
"loss": 5.6517,
"mean_token_accuracy": 0.14883127957582473,
"num_tokens": 12958503.0,
"step": 6235
},
{
"entropy": 5.816850376129151,
"epoch": 0.5660377358490566,
"grad_norm": 2.109375,
"learning_rate": 0.0004974505685602314,
"loss": 5.6696,
"mean_token_accuracy": 0.15917163342237473,
"num_tokens": 12968714.0,
"step": 6240
},
{
"entropy": 5.738234329223633,
"epoch": 0.5664912917271407,
"grad_norm": 1.890625,
"learning_rate": 0.0004974457092148912,
"loss": 5.7003,
"mean_token_accuracy": 0.15269872844219207,
"num_tokens": 12978857.0,
"step": 6245
},
{
"entropy": 5.845728397369385,
"epoch": 0.566944847605225,
"grad_norm": 2.109375,
"learning_rate": 0.0004974408452693172,
"loss": 5.8093,
"mean_token_accuracy": 0.14675376266241075,
"num_tokens": 12989863.0,
"step": 6250
},
{
"entropy": 5.77975435256958,
"epoch": 0.5673984034833092,
"grad_norm": 1.875,
"learning_rate": 0.0004974359767236099,
"loss": 5.6093,
"mean_token_accuracy": 0.15659026503562928,
"num_tokens": 13000307.0,
"step": 6255
},
{
"entropy": 5.753757667541504,
"epoch": 0.5678519593613933,
"grad_norm": 2.78125,
"learning_rate": 0.0004974311035778698,
"loss": 5.6585,
"mean_token_accuracy": 0.1654161751270294,
"num_tokens": 13009053.0,
"step": 6260
},
{
"entropy": 5.768190145492554,
"epoch": 0.5683055152394775,
"grad_norm": 3.78125,
"learning_rate": 0.0004974262258321979,
"loss": 5.7077,
"mean_token_accuracy": 0.14887018352746964,
"num_tokens": 13019136.0,
"step": 6265
},
{
"entropy": 5.851829242706299,
"epoch": 0.5687590711175616,
"grad_norm": 1.984375,
"learning_rate": 0.0004974213434866949,
"loss": 5.7188,
"mean_token_accuracy": 0.1501341938972473,
"num_tokens": 13030356.0,
"step": 6270
},
{
"entropy": 5.811538934707642,
"epoch": 0.5692126269956459,
"grad_norm": 2.234375,
"learning_rate": 0.0004974164565414617,
"loss": 5.7585,
"mean_token_accuracy": 0.14968807846307755,
"num_tokens": 13040477.0,
"step": 6275
},
{
"entropy": 5.80972990989685,
"epoch": 0.56966618287373,
"grad_norm": 3.46875,
"learning_rate": 0.0004974115649965998,
"loss": 5.6423,
"mean_token_accuracy": 0.16031724512577056,
"num_tokens": 13051755.0,
"step": 6280
},
{
"entropy": 5.850274562835693,
"epoch": 0.5701197387518142,
"grad_norm": 2.296875,
"learning_rate": 0.0004974066688522098,
"loss": 5.6499,
"mean_token_accuracy": 0.16230104714632035,
"num_tokens": 13061670.0,
"step": 6285
},
{
"entropy": 5.754382801055908,
"epoch": 0.5705732946298984,
"grad_norm": 1.765625,
"learning_rate": 0.0004974017681083934,
"loss": 5.6138,
"mean_token_accuracy": 0.15561977475881578,
"num_tokens": 13070917.0,
"step": 6290
},
{
"entropy": 5.703136301040649,
"epoch": 0.5710268505079826,
"grad_norm": 3.046875,
"learning_rate": 0.0004973968627652516,
"loss": 5.6785,
"mean_token_accuracy": 0.15372648537158967,
"num_tokens": 13080577.0,
"step": 6295
},
{
"entropy": 5.778132009506225,
"epoch": 0.5714804063860668,
"grad_norm": 2.09375,
"learning_rate": 0.000497391952822886,
"loss": 5.726,
"mean_token_accuracy": 0.15329584032297133,
"num_tokens": 13091446.0,
"step": 6300
},
{
"entropy": 5.842317533493042,
"epoch": 0.5719339622641509,
"grad_norm": 2.484375,
"learning_rate": 0.0004973870382813983,
"loss": 5.7267,
"mean_token_accuracy": 0.15676882565021516,
"num_tokens": 13102752.0,
"step": 6305
},
{
"entropy": 5.913141870498658,
"epoch": 0.5723875181422351,
"grad_norm": 2.125,
"learning_rate": 0.0004973821191408899,
"loss": 5.7531,
"mean_token_accuracy": 0.15161021798849106,
"num_tokens": 13113944.0,
"step": 6310
},
{
"entropy": 5.81167368888855,
"epoch": 0.5728410740203193,
"grad_norm": 2.0625,
"learning_rate": 0.0004973771954014626,
"loss": 5.6856,
"mean_token_accuracy": 0.1475789338350296,
"num_tokens": 13124316.0,
"step": 6315
},
{
"entropy": 5.883531665802002,
"epoch": 0.5732946298984035,
"grad_norm": 2.875,
"learning_rate": 0.0004973722670632182,
"loss": 5.784,
"mean_token_accuracy": 0.14491094052791595,
"num_tokens": 13133567.0,
"step": 6320
},
{
"entropy": 5.793781757354736,
"epoch": 0.5737481857764877,
"grad_norm": 3.609375,
"learning_rate": 0.0004973673341262587,
"loss": 5.7042,
"mean_token_accuracy": 0.15035074204206467,
"num_tokens": 13144279.0,
"step": 6325
},
{
"entropy": 5.812184858322143,
"epoch": 0.5742017416545718,
"grad_norm": 2.015625,
"learning_rate": 0.000497362396590686,
"loss": 5.7121,
"mean_token_accuracy": 0.15618311017751693,
"num_tokens": 13154463.0,
"step": 6330
},
{
"entropy": 5.828989505767822,
"epoch": 0.574655297532656,
"grad_norm": 2.765625,
"learning_rate": 0.0004973574544566024,
"loss": 5.7131,
"mean_token_accuracy": 0.1466642439365387,
"num_tokens": 13165565.0,
"step": 6335
},
{
"entropy": 5.76364049911499,
"epoch": 0.5751088534107403,
"grad_norm": 2.78125,
"learning_rate": 0.0004973525077241098,
"loss": 5.6335,
"mean_token_accuracy": 0.15842198878526687,
"num_tokens": 13175839.0,
"step": 6340
},
{
"entropy": 5.741176509857178,
"epoch": 0.5755624092888244,
"grad_norm": 3.921875,
"learning_rate": 0.0004973475563933109,
"loss": 5.6211,
"mean_token_accuracy": 0.15324287861585617,
"num_tokens": 13186372.0,
"step": 6345
},
{
"entropy": 5.764437246322632,
"epoch": 0.5760159651669086,
"grad_norm": 2.03125,
"learning_rate": 0.0004973426004643077,
"loss": 5.6682,
"mean_token_accuracy": 0.15345639437437059,
"num_tokens": 13196259.0,
"step": 6350
},
{
"entropy": 5.8232909679412845,
"epoch": 0.5764695210449927,
"grad_norm": 2.390625,
"learning_rate": 0.0004973376399372029,
"loss": 5.6659,
"mean_token_accuracy": 0.1532199814915657,
"num_tokens": 13206497.0,
"step": 6355
},
{
"entropy": 5.839055967330933,
"epoch": 0.5769230769230769,
"grad_norm": 2.828125,
"learning_rate": 0.0004973326748120991,
"loss": 5.7096,
"mean_token_accuracy": 0.15685370415449143,
"num_tokens": 13217188.0,
"step": 6360
},
{
"entropy": 5.755173206329346,
"epoch": 0.5773766328011611,
"grad_norm": 2.796875,
"learning_rate": 0.0004973277050890988,
"loss": 5.6947,
"mean_token_accuracy": 0.149724081158638,
"num_tokens": 13226544.0,
"step": 6365
},
{
"entropy": 5.835665464401245,
"epoch": 0.5778301886792453,
"grad_norm": 2.8125,
"learning_rate": 0.0004973227307683049,
"loss": 5.7452,
"mean_token_accuracy": 0.15042823553085327,
"num_tokens": 13236572.0,
"step": 6370
},
{
"entropy": 5.967599248886108,
"epoch": 0.5782837445573294,
"grad_norm": 1.953125,
"learning_rate": 0.0004973177518498204,
"loss": 5.7337,
"mean_token_accuracy": 0.14907948523759842,
"num_tokens": 13247958.0,
"step": 6375
},
{
"entropy": 5.800376892089844,
"epoch": 0.5787373004354136,
"grad_norm": 2.0625,
"learning_rate": 0.0004973127683337482,
"loss": 5.6412,
"mean_token_accuracy": 0.1538746476173401,
"num_tokens": 13258255.0,
"step": 6380
},
{
"entropy": 5.734482097625732,
"epoch": 0.5791908563134979,
"grad_norm": 2.1875,
"learning_rate": 0.0004973077802201911,
"loss": 5.6489,
"mean_token_accuracy": 0.15466178506612777,
"num_tokens": 13268416.0,
"step": 6385
},
{
"entropy": 5.756858682632446,
"epoch": 0.579644412191582,
"grad_norm": 2.53125,
"learning_rate": 0.0004973027875092526,
"loss": 5.6512,
"mean_token_accuracy": 0.15202094316482545,
"num_tokens": 13278087.0,
"step": 6390
},
{
"entropy": 5.738783597946167,
"epoch": 0.5800979680696662,
"grad_norm": 1.59375,
"learning_rate": 0.0004972977902010357,
"loss": 5.5609,
"mean_token_accuracy": 0.16106198728084564,
"num_tokens": 13288121.0,
"step": 6395
},
{
"entropy": 5.840384197235108,
"epoch": 0.5805515239477503,
"grad_norm": 2.171875,
"learning_rate": 0.0004972927882956439,
"loss": 5.7589,
"mean_token_accuracy": 0.1446023926138878,
"num_tokens": 13298620.0,
"step": 6400
},
{
"entropy": 5.742712450027466,
"epoch": 0.5810050798258345,
"grad_norm": 2.15625,
"learning_rate": 0.0004972877817931807,
"loss": 5.5853,
"mean_token_accuracy": 0.15124613642692566,
"num_tokens": 13308413.0,
"step": 6405
},
{
"entropy": 5.773351430892944,
"epoch": 0.5814586357039188,
"grad_norm": 2.640625,
"learning_rate": 0.0004972827706937493,
"loss": 5.6795,
"mean_token_accuracy": 0.15599592328071593,
"num_tokens": 13319387.0,
"step": 6410
},
{
"entropy": 5.802117109298706,
"epoch": 0.5819121915820029,
"grad_norm": 2.1875,
"learning_rate": 0.0004972777549974537,
"loss": 5.6197,
"mean_token_accuracy": 0.1567063510417938,
"num_tokens": 13330494.0,
"step": 6415
},
{
"entropy": 5.739154100418091,
"epoch": 0.5823657474600871,
"grad_norm": 2.28125,
"learning_rate": 0.0004972727347043975,
"loss": 5.6486,
"mean_token_accuracy": 0.1529611423611641,
"num_tokens": 13341009.0,
"step": 6420
},
{
"entropy": 5.731442546844482,
"epoch": 0.5828193033381712,
"grad_norm": 2.09375,
"learning_rate": 0.0004972677098146845,
"loss": 5.5792,
"mean_token_accuracy": 0.15856897830963135,
"num_tokens": 13351269.0,
"step": 6425
},
{
"entropy": 5.72405776977539,
"epoch": 0.5832728592162555,
"grad_norm": 4.625,
"learning_rate": 0.0004972626803284185,
"loss": 5.7119,
"mean_token_accuracy": 0.1523909494280815,
"num_tokens": 13362160.0,
"step": 6430
},
{
"entropy": 5.816617250442505,
"epoch": 0.5837264150943396,
"grad_norm": 1.8125,
"learning_rate": 0.0004972576462457038,
"loss": 5.763,
"mean_token_accuracy": 0.14996671974658965,
"num_tokens": 13372497.0,
"step": 6435
},
{
"entropy": 5.914173889160156,
"epoch": 0.5841799709724238,
"grad_norm": 2.203125,
"learning_rate": 0.0004972526075666442,
"loss": 5.6892,
"mean_token_accuracy": 0.1595553070306778,
"num_tokens": 13381665.0,
"step": 6440
},
{
"entropy": 5.676478433609009,
"epoch": 0.584633526850508,
"grad_norm": 2.59375,
"learning_rate": 0.0004972475642913442,
"loss": 5.5968,
"mean_token_accuracy": 0.15799911618232726,
"num_tokens": 13392507.0,
"step": 6445
},
{
"entropy": 5.675682163238525,
"epoch": 0.5850870827285921,
"grad_norm": 2.578125,
"learning_rate": 0.0004972425164199078,
"loss": 5.6109,
"mean_token_accuracy": 0.1597040683031082,
"num_tokens": 13402171.0,
"step": 6450
},
{
"entropy": 5.843981122970581,
"epoch": 0.5855406386066764,
"grad_norm": 2.234375,
"learning_rate": 0.0004972374639524396,
"loss": 5.7999,
"mean_token_accuracy": 0.14906672090291978,
"num_tokens": 13412933.0,
"step": 6455
},
{
"entropy": 5.806398105621338,
"epoch": 0.5859941944847605,
"grad_norm": 2.65625,
"learning_rate": 0.000497232406889044,
"loss": 5.6419,
"mean_token_accuracy": 0.1513127401471138,
"num_tokens": 13423563.0,
"step": 6460
},
{
"entropy": 5.852730798721313,
"epoch": 0.5864477503628447,
"grad_norm": 2.421875,
"learning_rate": 0.0004972273452298255,
"loss": 5.667,
"mean_token_accuracy": 0.15766609236598014,
"num_tokens": 13433998.0,
"step": 6465
},
{
"entropy": 5.733954334259034,
"epoch": 0.5869013062409288,
"grad_norm": 2.875,
"learning_rate": 0.0004972222789748888,
"loss": 5.6163,
"mean_token_accuracy": 0.1577115073800087,
"num_tokens": 13443878.0,
"step": 6470
},
{
"entropy": 5.718626022338867,
"epoch": 0.5873548621190131,
"grad_norm": 2.828125,
"learning_rate": 0.0004972172081243389,
"loss": 5.709,
"mean_token_accuracy": 0.14359332993626595,
"num_tokens": 13456051.0,
"step": 6475
},
{
"entropy": 5.832206058502197,
"epoch": 0.5878084179970973,
"grad_norm": 3.359375,
"learning_rate": 0.0004972121326782806,
"loss": 5.7064,
"mean_token_accuracy": 0.15390692949295043,
"num_tokens": 13467276.0,
"step": 6480
},
{
"entropy": 5.777996397018432,
"epoch": 0.5882619738751814,
"grad_norm": 2.53125,
"learning_rate": 0.0004972070526368186,
"loss": 5.6924,
"mean_token_accuracy": 0.1549999363720417,
"num_tokens": 13478331.0,
"step": 6485
},
{
"entropy": 5.788837480545044,
"epoch": 0.5887155297532656,
"grad_norm": 1.5234375,
"learning_rate": 0.0004972019680000582,
"loss": 5.6127,
"mean_token_accuracy": 0.15205416679382325,
"num_tokens": 13489879.0,
"step": 6490
},
{
"entropy": 5.781062746047974,
"epoch": 0.5891690856313497,
"grad_norm": 2.921875,
"learning_rate": 0.0004971968787681044,
"loss": 5.7449,
"mean_token_accuracy": 0.15462507754564286,
"num_tokens": 13500879.0,
"step": 6495
},
{
"entropy": 5.815672445297241,
"epoch": 0.589622641509434,
"grad_norm": 2.546875,
"learning_rate": 0.0004971917849410627,
"loss": 5.7127,
"mean_token_accuracy": 0.15203070044517517,
"num_tokens": 13511719.0,
"step": 6500
},
{
"entropy": 5.77512640953064,
"epoch": 0.5900761973875182,
"grad_norm": 2.796875,
"learning_rate": 0.0004971866865190381,
"loss": 5.635,
"mean_token_accuracy": 0.15535284876823424,
"num_tokens": 13521850.0,
"step": 6505
},
{
"entropy": 5.83427414894104,
"epoch": 0.5905297532656023,
"grad_norm": 2.46875,
"learning_rate": 0.0004971815835021362,
"loss": 5.6695,
"mean_token_accuracy": 0.15581846982240677,
"num_tokens": 13532095.0,
"step": 6510
},
{
"entropy": 5.7484681606292725,
"epoch": 0.5909833091436865,
"grad_norm": 3.09375,
"learning_rate": 0.0004971764758904626,
"loss": 5.6417,
"mean_token_accuracy": 0.14961459189653398,
"num_tokens": 13543133.0,
"step": 6515
},
{
"entropy": 5.751182270050049,
"epoch": 0.5914368650217707,
"grad_norm": 2.703125,
"learning_rate": 0.0004971713636841229,
"loss": 5.7191,
"mean_token_accuracy": 0.14607228338718414,
"num_tokens": 13553771.0,
"step": 6520
},
{
"entropy": 5.724217700958252,
"epoch": 0.5918904208998549,
"grad_norm": 2.34375,
"learning_rate": 0.0004971662468832228,
"loss": 5.6471,
"mean_token_accuracy": 0.15626089870929719,
"num_tokens": 13564548.0,
"step": 6525
},
{
"entropy": 5.78565034866333,
"epoch": 0.592343976777939,
"grad_norm": 1.890625,
"learning_rate": 0.000497161125487868,
"loss": 5.6305,
"mean_token_accuracy": 0.1553097426891327,
"num_tokens": 13574567.0,
"step": 6530
},
{
"entropy": 5.737082815170288,
"epoch": 0.5927975326560232,
"grad_norm": 2.21875,
"learning_rate": 0.0004971559994981645,
"loss": 5.6748,
"mean_token_accuracy": 0.1555297389626503,
"num_tokens": 13585302.0,
"step": 6535
},
{
"entropy": 5.789227914810181,
"epoch": 0.5932510885341074,
"grad_norm": 2.171875,
"learning_rate": 0.0004971508689142184,
"loss": 5.7061,
"mean_token_accuracy": 0.1495283529162407,
"num_tokens": 13594754.0,
"step": 6540
},
{
"entropy": 5.746747541427612,
"epoch": 0.5937046444121916,
"grad_norm": 5.1875,
"learning_rate": 0.0004971457337361356,
"loss": 5.6402,
"mean_token_accuracy": 0.15646791607141494,
"num_tokens": 13604691.0,
"step": 6545
},
{
"entropy": 5.648950910568237,
"epoch": 0.5941582002902758,
"grad_norm": 3.0,
"learning_rate": 0.0004971405939640225,
"loss": 5.5617,
"mean_token_accuracy": 0.16001858487725257,
"num_tokens": 13613911.0,
"step": 6550
},
{
"entropy": 5.920597696304322,
"epoch": 0.5946117561683599,
"grad_norm": 2.375,
"learning_rate": 0.0004971354495979854,
"loss": 5.768,
"mean_token_accuracy": 0.152364581823349,
"num_tokens": 13624285.0,
"step": 6555
},
{
"entropy": 5.822053241729736,
"epoch": 0.5950653120464441,
"grad_norm": 1.8046875,
"learning_rate": 0.0004971303006381305,
"loss": 5.6054,
"mean_token_accuracy": 0.1551180973649025,
"num_tokens": 13635434.0,
"step": 6560
},
{
"entropy": 5.768273448944091,
"epoch": 0.5955188679245284,
"grad_norm": 2.21875,
"learning_rate": 0.0004971251470845643,
"loss": 5.6781,
"mean_token_accuracy": 0.1518924981355667,
"num_tokens": 13645582.0,
"step": 6565
},
{
"entropy": 5.699561405181885,
"epoch": 0.5959724238026125,
"grad_norm": 2.28125,
"learning_rate": 0.0004971199889373935,
"loss": 5.6228,
"mean_token_accuracy": 0.15436237156391144,
"num_tokens": 13655029.0,
"step": 6570
},
{
"entropy": 5.711526155471802,
"epoch": 0.5964259796806967,
"grad_norm": 3.328125,
"learning_rate": 0.0004971148261967249,
"loss": 5.6159,
"mean_token_accuracy": 0.15454744100570678,
"num_tokens": 13665204.0,
"step": 6575
},
{
"entropy": 5.790972518920898,
"epoch": 0.5968795355587808,
"grad_norm": 4.0,
"learning_rate": 0.0004971096588626649,
"loss": 5.7243,
"mean_token_accuracy": 0.15180352926254273,
"num_tokens": 13676281.0,
"step": 6580
},
{
"entropy": 5.758441495895386,
"epoch": 0.597333091436865,
"grad_norm": 2.5625,
"learning_rate": 0.0004971044869353206,
"loss": 5.7068,
"mean_token_accuracy": 0.16017238646745682,
"num_tokens": 13687144.0,
"step": 6585
},
{
"entropy": 5.8007988929748535,
"epoch": 0.5977866473149492,
"grad_norm": 3.453125,
"learning_rate": 0.000497099310414799,
"loss": 5.6572,
"mean_token_accuracy": 0.14968026280403138,
"num_tokens": 13696371.0,
"step": 6590
},
{
"entropy": 5.84863429069519,
"epoch": 0.5982402031930334,
"grad_norm": 3.078125,
"learning_rate": 0.0004970941293012069,
"loss": 5.6164,
"mean_token_accuracy": 0.15415885299444199,
"num_tokens": 13706900.0,
"step": 6595
},
{
"entropy": 5.796767044067383,
"epoch": 0.5986937590711175,
"grad_norm": 2.390625,
"learning_rate": 0.0004970889435946518,
"loss": 5.6944,
"mean_token_accuracy": 0.15410789847373962,
"num_tokens": 13717874.0,
"step": 6600
},
{
"entropy": 5.756249475479126,
"epoch": 0.5991473149492017,
"grad_norm": 2.359375,
"learning_rate": 0.0004970837532952406,
"loss": 5.6485,
"mean_token_accuracy": 0.1519013985991478,
"num_tokens": 13728020.0,
"step": 6605
},
{
"entropy": 5.794941711425781,
"epoch": 0.599600870827286,
"grad_norm": 2.78125,
"learning_rate": 0.0004970785584030808,
"loss": 5.6978,
"mean_token_accuracy": 0.15266939252614975,
"num_tokens": 13738007.0,
"step": 6610
},
{
"entropy": 5.823725843429566,
"epoch": 0.6000544267053701,
"grad_norm": 2.015625,
"learning_rate": 0.0004970733589182799,
"loss": 5.7034,
"mean_token_accuracy": 0.15135803669691086,
"num_tokens": 13749429.0,
"step": 6615
},
{
"entropy": 5.7685328960418705,
"epoch": 0.6005079825834543,
"grad_norm": 2.59375,
"learning_rate": 0.0004970681548409452,
"loss": 5.5744,
"mean_token_accuracy": 0.16029660254716874,
"num_tokens": 13760279.0,
"step": 6620
},
{
"entropy": 5.792378854751587,
"epoch": 0.6009615384615384,
"grad_norm": 2.421875,
"learning_rate": 0.0004970629461711846,
"loss": 5.7863,
"mean_token_accuracy": 0.1477381929755211,
"num_tokens": 13770575.0,
"step": 6625
},
{
"entropy": 5.754233694076538,
"epoch": 0.6014150943396226,
"grad_norm": 2.140625,
"learning_rate": 0.0004970577329091057,
"loss": 5.6294,
"mean_token_accuracy": 0.15650067627429962,
"num_tokens": 13781068.0,
"step": 6630
},
{
"entropy": 5.857264184951783,
"epoch": 0.6018686502177069,
"grad_norm": 2.015625,
"learning_rate": 0.0004970525150548162,
"loss": 5.7015,
"mean_token_accuracy": 0.15616442114114762,
"num_tokens": 13790745.0,
"step": 6635
},
{
"entropy": 5.786818790435791,
"epoch": 0.602322206095791,
"grad_norm": 2.03125,
"learning_rate": 0.0004970472926084242,
"loss": 5.627,
"mean_token_accuracy": 0.1557512104511261,
"num_tokens": 13800196.0,
"step": 6640
},
{
"entropy": 5.717859745025635,
"epoch": 0.6027757619738752,
"grad_norm": 1.953125,
"learning_rate": 0.0004970420655700375,
"loss": 5.5438,
"mean_token_accuracy": 0.15962371081113816,
"num_tokens": 13809954.0,
"step": 6645
},
{
"entropy": 5.706457853317261,
"epoch": 0.6032293178519593,
"grad_norm": 9.3125,
"learning_rate": 0.0004970368339397643,
"loss": 5.6972,
"mean_token_accuracy": 0.15097275376319885,
"num_tokens": 13820281.0,
"step": 6650
},
{
"entropy": 5.847550535202027,
"epoch": 0.6036828737300436,
"grad_norm": 2.53125,
"learning_rate": 0.0004970315977177129,
"loss": 5.6611,
"mean_token_accuracy": 0.15552590638399125,
"num_tokens": 13830614.0,
"step": 6655
},
{
"entropy": 5.773291969299317,
"epoch": 0.6041364296081277,
"grad_norm": 2.375,
"learning_rate": 0.0004970263569039914,
"loss": 5.6072,
"mean_token_accuracy": 0.1574172392487526,
"num_tokens": 13840551.0,
"step": 6660
},
{
"entropy": 5.838003158569336,
"epoch": 0.6045899854862119,
"grad_norm": 2.25,
"learning_rate": 0.0004970211114987083,
"loss": 5.7941,
"mean_token_accuracy": 0.14930742010474204,
"num_tokens": 13851830.0,
"step": 6665
},
{
"entropy": 5.783898496627808,
"epoch": 0.6050435413642961,
"grad_norm": 1.9765625,
"learning_rate": 0.0004970158615019719,
"loss": 5.597,
"mean_token_accuracy": 0.1558218017220497,
"num_tokens": 13862299.0,
"step": 6670
},
{
"entropy": 5.888981723785401,
"epoch": 0.6054970972423802,
"grad_norm": 2.28125,
"learning_rate": 0.0004970106069138912,
"loss": 5.7321,
"mean_token_accuracy": 0.14841488599777222,
"num_tokens": 13872852.0,
"step": 6675
},
{
"entropy": 5.780848693847656,
"epoch": 0.6059506531204645,
"grad_norm": 2.078125,
"learning_rate": 0.0004970053477345743,
"loss": 5.679,
"mean_token_accuracy": 0.16062871217727662,
"num_tokens": 13883601.0,
"step": 6680
},
{
"entropy": 5.772215461730957,
"epoch": 0.6064042089985486,
"grad_norm": 2.0625,
"learning_rate": 0.0004970000839641304,
"loss": 5.6774,
"mean_token_accuracy": 0.1588228940963745,
"num_tokens": 13894246.0,
"step": 6685
},
{
"entropy": 5.728923892974853,
"epoch": 0.6068577648766328,
"grad_norm": 2.171875,
"learning_rate": 0.0004969948156026681,
"loss": 5.6967,
"mean_token_accuracy": 0.1531152456998825,
"num_tokens": 13904823.0,
"step": 6690
},
{
"entropy": 5.882969808578491,
"epoch": 0.6073113207547169,
"grad_norm": 2.515625,
"learning_rate": 0.0004969895426502965,
"loss": 5.7097,
"mean_token_accuracy": 0.14678643122315407,
"num_tokens": 13915338.0,
"step": 6695
},
{
"entropy": 5.754983043670654,
"epoch": 0.6077648766328012,
"grad_norm": 2.453125,
"learning_rate": 0.0004969842651071246,
"loss": 5.6622,
"mean_token_accuracy": 0.15358965694904328,
"num_tokens": 13925838.0,
"step": 6700
},
{
"entropy": 5.784938812255859,
"epoch": 0.6082184325108854,
"grad_norm": 2.53125,
"learning_rate": 0.0004969789829732615,
"loss": 5.6533,
"mean_token_accuracy": 0.15099577605724335,
"num_tokens": 13936341.0,
"step": 6705
},
{
"entropy": 5.796441841125488,
"epoch": 0.6086719883889695,
"grad_norm": 5.4375,
"learning_rate": 0.0004969736962488164,
"loss": 5.6789,
"mean_token_accuracy": 0.15407094806432725,
"num_tokens": 13945438.0,
"step": 6710
},
{
"entropy": 5.788770055770874,
"epoch": 0.6091255442670537,
"grad_norm": 2.609375,
"learning_rate": 0.0004969684049338987,
"loss": 5.5788,
"mean_token_accuracy": 0.14951442778110505,
"num_tokens": 13956029.0,
"step": 6715
},
{
"entropy": 5.802248239517212,
"epoch": 0.6095791001451378,
"grad_norm": 1.9921875,
"learning_rate": 0.000496963109028618,
"loss": 5.6715,
"mean_token_accuracy": 0.14977479204535485,
"num_tokens": 13966498.0,
"step": 6720
},
{
"entropy": 5.763838195800782,
"epoch": 0.6100326560232221,
"grad_norm": 2.765625,
"learning_rate": 0.0004969578085330834,
"loss": 5.6424,
"mean_token_accuracy": 0.1522256314754486,
"num_tokens": 13976997.0,
"step": 6725
},
{
"entropy": 5.8410039901733395,
"epoch": 0.6104862119013063,
"grad_norm": 1.9765625,
"learning_rate": 0.0004969525034474049,
"loss": 5.6349,
"mean_token_accuracy": 0.1578125163912773,
"num_tokens": 13986664.0,
"step": 6730
},
{
"entropy": 5.728191089630127,
"epoch": 0.6109397677793904,
"grad_norm": 2.296875,
"learning_rate": 0.000496947193771692,
"loss": 5.6585,
"mean_token_accuracy": 0.15620095580816268,
"num_tokens": 13997828.0,
"step": 6735
},
{
"entropy": 5.7189115524292,
"epoch": 0.6113933236574746,
"grad_norm": 1.9921875,
"learning_rate": 0.0004969418795060548,
"loss": 5.5975,
"mean_token_accuracy": 0.15450024902820586,
"num_tokens": 14008293.0,
"step": 6740
},
{
"entropy": 5.708045387268067,
"epoch": 0.6118468795355588,
"grad_norm": 2.09375,
"learning_rate": 0.0004969365606506027,
"loss": 5.5776,
"mean_token_accuracy": 0.1588921070098877,
"num_tokens": 14018031.0,
"step": 6745
},
{
"entropy": 5.698815870285034,
"epoch": 0.612300435413643,
"grad_norm": 1.90625,
"learning_rate": 0.0004969312372054461,
"loss": 5.5546,
"mean_token_accuracy": 0.1623407393693924,
"num_tokens": 14028013.0,
"step": 6750
},
{
"entropy": 5.706755018234253,
"epoch": 0.6127539912917271,
"grad_norm": 1.7890625,
"learning_rate": 0.0004969259091706948,
"loss": 5.681,
"mean_token_accuracy": 0.15510369837284088,
"num_tokens": 14039787.0,
"step": 6755
},
{
"entropy": 5.749244928359985,
"epoch": 0.6132075471698113,
"grad_norm": 3.109375,
"learning_rate": 0.0004969205765464593,
"loss": 5.5586,
"mean_token_accuracy": 0.156866218149662,
"num_tokens": 14049223.0,
"step": 6760
},
{
"entropy": 5.757191324234009,
"epoch": 0.6136611030478955,
"grad_norm": 2.359375,
"learning_rate": 0.0004969152393328497,
"loss": 5.6249,
"mean_token_accuracy": 0.16180411130189895,
"num_tokens": 14059004.0,
"step": 6765
},
{
"entropy": 5.812931060791016,
"epoch": 0.6141146589259797,
"grad_norm": 1.9765625,
"learning_rate": 0.0004969098975299764,
"loss": 5.7093,
"mean_token_accuracy": 0.14600952416658403,
"num_tokens": 14069615.0,
"step": 6770
},
{
"entropy": 5.7707301616668705,
"epoch": 0.6145682148040639,
"grad_norm": 2.203125,
"learning_rate": 0.0004969045511379499,
"loss": 5.6474,
"mean_token_accuracy": 0.15415120422840117,
"num_tokens": 14078886.0,
"step": 6775
},
{
"entropy": 5.795013856887818,
"epoch": 0.615021770682148,
"grad_norm": 1.6640625,
"learning_rate": 0.0004968992001568806,
"loss": 5.7203,
"mean_token_accuracy": 0.14738596454262734,
"num_tokens": 14089021.0,
"step": 6780
},
{
"entropy": 5.7487105369567875,
"epoch": 0.6154753265602322,
"grad_norm": 2.390625,
"learning_rate": 0.0004968938445868795,
"loss": 5.5626,
"mean_token_accuracy": 0.16147104948759078,
"num_tokens": 14098316.0,
"step": 6785
},
{
"entropy": 5.750917434692383,
"epoch": 0.6159288824383164,
"grad_norm": 2.5,
"learning_rate": 0.0004968884844280569,
"loss": 5.6483,
"mean_token_accuracy": 0.14891300946474076,
"num_tokens": 14108874.0,
"step": 6790
},
{
"entropy": 5.798082160949707,
"epoch": 0.6163824383164006,
"grad_norm": 2.75,
"learning_rate": 0.0004968831196805242,
"loss": 5.7037,
"mean_token_accuracy": 0.15190935805439948,
"num_tokens": 14119613.0,
"step": 6795
},
{
"entropy": 5.810618591308594,
"epoch": 0.6168359941944848,
"grad_norm": 3.8125,
"learning_rate": 0.0004968777503443918,
"loss": 5.6416,
"mean_token_accuracy": 0.1516499012708664,
"num_tokens": 14129085.0,
"step": 6800
},
{
"entropy": 5.648369979858399,
"epoch": 0.6172895500725689,
"grad_norm": 2.625,
"learning_rate": 0.0004968723764197711,
"loss": 5.5289,
"mean_token_accuracy": 0.16768998205661773,
"num_tokens": 14138461.0,
"step": 6805
},
{
"entropy": 5.750721549987793,
"epoch": 0.6177431059506531,
"grad_norm": 2.34375,
"learning_rate": 0.000496866997906773,
"loss": 5.5976,
"mean_token_accuracy": 0.15950866043567657,
"num_tokens": 14149126.0,
"step": 6810
},
{
"entropy": 5.729853963851928,
"epoch": 0.6181966618287373,
"grad_norm": 2.234375,
"learning_rate": 0.0004968616148055089,
"loss": 5.6058,
"mean_token_accuracy": 0.16136523485183715,
"num_tokens": 14160068.0,
"step": 6815
},
{
"entropy": 5.7663956642150875,
"epoch": 0.6186502177068215,
"grad_norm": 2.640625,
"learning_rate": 0.0004968562271160901,
"loss": 5.7195,
"mean_token_accuracy": 0.15396957993507385,
"num_tokens": 14170288.0,
"step": 6820
},
{
"entropy": 5.772651624679566,
"epoch": 0.6191037735849056,
"grad_norm": 2.203125,
"learning_rate": 0.000496850834838628,
"loss": 5.6992,
"mean_token_accuracy": 0.15586561411619188,
"num_tokens": 14181231.0,
"step": 6825
},
{
"entropy": 5.778563213348389,
"epoch": 0.6195573294629898,
"grad_norm": 2.03125,
"learning_rate": 0.0004968454379732342,
"loss": 5.6078,
"mean_token_accuracy": 0.1582838088274002,
"num_tokens": 14191715.0,
"step": 6830
},
{
"entropy": 5.735866069793701,
"epoch": 0.6200108853410741,
"grad_norm": 2.078125,
"learning_rate": 0.00049684003652002,
"loss": 5.5997,
"mean_token_accuracy": 0.15433213114738464,
"num_tokens": 14203416.0,
"step": 6835
},
{
"entropy": 5.695660734176636,
"epoch": 0.6204644412191582,
"grad_norm": 2.75,
"learning_rate": 0.0004968346304790973,
"loss": 5.5682,
"mean_token_accuracy": 0.1618633046746254,
"num_tokens": 14213658.0,
"step": 6840
},
{
"entropy": 5.8019201278686525,
"epoch": 0.6209179970972424,
"grad_norm": 1.9296875,
"learning_rate": 0.0004968292198505781,
"loss": 5.732,
"mean_token_accuracy": 0.14817453622817994,
"num_tokens": 14223956.0,
"step": 6845
},
{
"entropy": 5.831279468536377,
"epoch": 0.6213715529753265,
"grad_norm": 3.984375,
"learning_rate": 0.000496823804634574,
"loss": 5.73,
"mean_token_accuracy": 0.15022608637809753,
"num_tokens": 14236801.0,
"step": 6850
},
{
"entropy": 5.737774848937988,
"epoch": 0.6218251088534107,
"grad_norm": 2.5,
"learning_rate": 0.0004968183848311972,
"loss": 5.5545,
"mean_token_accuracy": 0.1597556382417679,
"num_tokens": 14247382.0,
"step": 6855
},
{
"entropy": 5.777302408218384,
"epoch": 0.622278664731495,
"grad_norm": 2.5,
"learning_rate": 0.0004968129604405596,
"loss": 5.6332,
"mean_token_accuracy": 0.15609384775161744,
"num_tokens": 14258627.0,
"step": 6860
},
{
"entropy": 5.725931549072266,
"epoch": 0.6227322206095791,
"grad_norm": 3.546875,
"learning_rate": 0.0004968075314627734,
"loss": 5.622,
"mean_token_accuracy": 0.15463528782129288,
"num_tokens": 14269013.0,
"step": 6865
},
{
"entropy": 5.804495143890381,
"epoch": 0.6231857764876633,
"grad_norm": 2.109375,
"learning_rate": 0.000496802097897951,
"loss": 5.6109,
"mean_token_accuracy": 0.16023218929767608,
"num_tokens": 14279620.0,
"step": 6870
},
{
"entropy": 5.734587478637695,
"epoch": 0.6236393323657474,
"grad_norm": 1.8125,
"learning_rate": 0.0004967966597462046,
"loss": 5.6276,
"mean_token_accuracy": 0.16197851449251174,
"num_tokens": 14289448.0,
"step": 6875
},
{
"entropy": 5.7310723781585695,
"epoch": 0.6240928882438317,
"grad_norm": 1.7734375,
"learning_rate": 0.0004967912170076469,
"loss": 5.6409,
"mean_token_accuracy": 0.153085844963789,
"num_tokens": 14299999.0,
"step": 6880
},
{
"entropy": 5.724435091018677,
"epoch": 0.6245464441219158,
"grad_norm": 3.8125,
"learning_rate": 0.0004967857696823901,
"loss": 5.6104,
"mean_token_accuracy": 0.15244664996862411,
"num_tokens": 14310384.0,
"step": 6885
},
{
"entropy": 5.800471925735474,
"epoch": 0.625,
"grad_norm": 1.859375,
"learning_rate": 0.0004967803177705472,
"loss": 5.705,
"mean_token_accuracy": 0.15142763704061507,
"num_tokens": 14320644.0,
"step": 6890
},
{
"entropy": 5.820406770706176,
"epoch": 0.6254535558780842,
"grad_norm": 1.671875,
"learning_rate": 0.0004967748612722308,
"loss": 5.6746,
"mean_token_accuracy": 0.150961971282959,
"num_tokens": 14330077.0,
"step": 6895
},
{
"entropy": 5.784531354904175,
"epoch": 0.6259071117561683,
"grad_norm": 2.0,
"learning_rate": 0.0004967694001875538,
"loss": 5.7679,
"mean_token_accuracy": 0.14856128245592118,
"num_tokens": 14341549.0,
"step": 6900
},
{
"entropy": 5.771959972381592,
"epoch": 0.6263606676342526,
"grad_norm": 2.046875,
"learning_rate": 0.000496763934516629,
"loss": 5.6458,
"mean_token_accuracy": 0.1570087343454361,
"num_tokens": 14352146.0,
"step": 6905
},
{
"entropy": 5.73153977394104,
"epoch": 0.6268142235123367,
"grad_norm": 1.6640625,
"learning_rate": 0.0004967584642595696,
"loss": 5.6305,
"mean_token_accuracy": 0.15252010822296141,
"num_tokens": 14362521.0,
"step": 6910
},
{
"entropy": 5.779860877990723,
"epoch": 0.6272677793904209,
"grad_norm": 2.78125,
"learning_rate": 0.0004967529894164887,
"loss": 5.6846,
"mean_token_accuracy": 0.1533489242196083,
"num_tokens": 14373216.0,
"step": 6915
},
{
"entropy": 5.749915695190429,
"epoch": 0.627721335268505,
"grad_norm": 2.84375,
"learning_rate": 0.0004967475099874992,
"loss": 5.5904,
"mean_token_accuracy": 0.1652483731508255,
"num_tokens": 14383104.0,
"step": 6920
},
{
"entropy": 5.7152907371521,
"epoch": 0.6281748911465893,
"grad_norm": 1.8828125,
"learning_rate": 0.000496742025972715,
"loss": 5.6335,
"mean_token_accuracy": 0.16333371847867967,
"num_tokens": 14393854.0,
"step": 6925
},
{
"entropy": 5.775513935089111,
"epoch": 0.6286284470246735,
"grad_norm": 1.78125,
"learning_rate": 0.000496736537372249,
"loss": 5.5885,
"mean_token_accuracy": 0.1643975183367729,
"num_tokens": 14403213.0,
"step": 6930
},
{
"entropy": 5.840988874435425,
"epoch": 0.6290820029027576,
"grad_norm": 2.171875,
"learning_rate": 0.0004967310441862151,
"loss": 5.7764,
"mean_token_accuracy": 0.14523438438773156,
"num_tokens": 14412792.0,
"step": 6935
},
{
"entropy": 5.806761837005615,
"epoch": 0.6295355587808418,
"grad_norm": 2.375,
"learning_rate": 0.0004967255464147265,
"loss": 5.6461,
"mean_token_accuracy": 0.15509824603796005,
"num_tokens": 14423061.0,
"step": 6940
},
{
"entropy": 5.78497633934021,
"epoch": 0.6299891146589259,
"grad_norm": 5.8125,
"learning_rate": 0.0004967200440578972,
"loss": 5.5931,
"mean_token_accuracy": 0.15885335355997085,
"num_tokens": 14433338.0,
"step": 6945
},
{
"entropy": 5.7435633659362795,
"epoch": 0.6304426705370102,
"grad_norm": 2.046875,
"learning_rate": 0.0004967145371158408,
"loss": 5.5782,
"mean_token_accuracy": 0.1651199445128441,
"num_tokens": 14444124.0,
"step": 6950
},
{
"entropy": 5.7816649913787845,
"epoch": 0.6308962264150944,
"grad_norm": 2.296875,
"learning_rate": 0.0004967090255886714,
"loss": 5.6981,
"mean_token_accuracy": 0.15614887624979018,
"num_tokens": 14453868.0,
"step": 6955
},
{
"entropy": 5.772961521148682,
"epoch": 0.6313497822931785,
"grad_norm": 2.234375,
"learning_rate": 0.0004967035094765028,
"loss": 5.6012,
"mean_token_accuracy": 0.15304825007915496,
"num_tokens": 14465015.0,
"step": 6960
},
{
"entropy": 5.720942449569702,
"epoch": 0.6318033381712627,
"grad_norm": 3.984375,
"learning_rate": 0.0004966979887794491,
"loss": 5.5757,
"mean_token_accuracy": 0.1620814934372902,
"num_tokens": 14475523.0,
"step": 6965
},
{
"entropy": 5.7405132293701175,
"epoch": 0.6322568940493469,
"grad_norm": 1.7421875,
"learning_rate": 0.0004966924634976245,
"loss": 5.6347,
"mean_token_accuracy": 0.15292872935533525,
"num_tokens": 14485523.0,
"step": 6970
},
{
"entropy": 5.809803819656372,
"epoch": 0.6327104499274311,
"grad_norm": 3.015625,
"learning_rate": 0.0004966869336311433,
"loss": 5.6984,
"mean_token_accuracy": 0.16131531298160554,
"num_tokens": 14495237.0,
"step": 6975
},
{
"entropy": 5.816791963577271,
"epoch": 0.6331640058055152,
"grad_norm": 2.453125,
"learning_rate": 0.0004966813991801197,
"loss": 5.8113,
"mean_token_accuracy": 0.1457821674644947,
"num_tokens": 14506431.0,
"step": 6980
},
{
"entropy": 5.85917067527771,
"epoch": 0.6336175616835994,
"grad_norm": 2.359375,
"learning_rate": 0.0004966758601446684,
"loss": 5.7372,
"mean_token_accuracy": 0.14926016479730606,
"num_tokens": 14517445.0,
"step": 6985
},
{
"entropy": 5.7707703590393065,
"epoch": 0.6340711175616836,
"grad_norm": 3.0,
"learning_rate": 0.0004966703165249038,
"loss": 5.6704,
"mean_token_accuracy": 0.14917180985212325,
"num_tokens": 14527818.0,
"step": 6990
},
{
"entropy": 5.7728606224060055,
"epoch": 0.6345246734397678,
"grad_norm": 1.9296875,
"learning_rate": 0.0004966647683209406,
"loss": 5.7489,
"mean_token_accuracy": 0.14479509368538857,
"num_tokens": 14538235.0,
"step": 6995
},
{
"entropy": 5.981025266647339,
"epoch": 0.634978229317852,
"grad_norm": 3.125,
"learning_rate": 0.0004966592155328935,
"loss": 5.7754,
"mean_token_accuracy": 0.14579777866601945,
"num_tokens": 14549784.0,
"step": 7000
},
{
"entropy": 5.875027751922607,
"epoch": 0.6354317851959361,
"grad_norm": 2.171875,
"learning_rate": 0.0004966536581608771,
"loss": 5.7679,
"mean_token_accuracy": 0.14473922178149223,
"num_tokens": 14561057.0,
"step": 7005
},
{
"entropy": 5.778143692016601,
"epoch": 0.6358853410740203,
"grad_norm": 1.9921875,
"learning_rate": 0.0004966480962050069,
"loss": 5.675,
"mean_token_accuracy": 0.15713430494070052,
"num_tokens": 14570859.0,
"step": 7010
},
{
"entropy": 5.8072175025939945,
"epoch": 0.6363388969521045,
"grad_norm": 2.46875,
"learning_rate": 0.0004966425296653974,
"loss": 5.6699,
"mean_token_accuracy": 0.16011741906404495,
"num_tokens": 14581066.0,
"step": 7015
},
{
"entropy": 5.772267436981201,
"epoch": 0.6367924528301887,
"grad_norm": 2.5,
"learning_rate": 0.000496636958542164,
"loss": 5.6578,
"mean_token_accuracy": 0.15138954669237137,
"num_tokens": 14590709.0,
"step": 7020
},
{
"entropy": 5.733920240402222,
"epoch": 0.6372460087082729,
"grad_norm": 2.1875,
"learning_rate": 0.0004966313828354218,
"loss": 5.634,
"mean_token_accuracy": 0.1559073120355606,
"num_tokens": 14600849.0,
"step": 7025
},
{
"entropy": 5.766506862640381,
"epoch": 0.637699564586357,
"grad_norm": 2.375,
"learning_rate": 0.000496625802545286,
"loss": 5.5168,
"mean_token_accuracy": 0.1601102754473686,
"num_tokens": 14611427.0,
"step": 7030
},
{
"entropy": 5.8376936435699465,
"epoch": 0.6381531204644412,
"grad_norm": 2.59375,
"learning_rate": 0.0004966202176718723,
"loss": 5.6952,
"mean_token_accuracy": 0.15300938487052917,
"num_tokens": 14623860.0,
"step": 7035
},
{
"entropy": 5.736017656326294,
"epoch": 0.6386066763425254,
"grad_norm": 2.5,
"learning_rate": 0.0004966146282152959,
"loss": 5.5947,
"mean_token_accuracy": 0.15307328402996062,
"num_tokens": 14633481.0,
"step": 7040
},
{
"entropy": 5.720715570449829,
"epoch": 0.6390602322206096,
"grad_norm": 3.125,
"learning_rate": 0.0004966090341756726,
"loss": 5.5762,
"mean_token_accuracy": 0.16294495910406112,
"num_tokens": 14643519.0,
"step": 7045
},
{
"entropy": 5.712062215805053,
"epoch": 0.6395137880986937,
"grad_norm": 2.90625,
"learning_rate": 0.000496603435553118,
"loss": 5.6037,
"mean_token_accuracy": 0.16128794848918915,
"num_tokens": 14654570.0,
"step": 7050
},
{
"entropy": 5.797862768173218,
"epoch": 0.6399673439767779,
"grad_norm": 2.203125,
"learning_rate": 0.0004965978323477478,
"loss": 5.6411,
"mean_token_accuracy": 0.14805451184511184,
"num_tokens": 14665596.0,
"step": 7055
},
{
"entropy": 5.791617488861084,
"epoch": 0.6404208998548622,
"grad_norm": 2.09375,
"learning_rate": 0.000496592224559678,
"loss": 5.575,
"mean_token_accuracy": 0.1581504911184311,
"num_tokens": 14675660.0,
"step": 7060
},
{
"entropy": 5.788625335693359,
"epoch": 0.6408744557329463,
"grad_norm": 2.03125,
"learning_rate": 0.0004965866121890245,
"loss": 5.7199,
"mean_token_accuracy": 0.1488729402422905,
"num_tokens": 14685229.0,
"step": 7065
},
{
"entropy": 5.798623323440552,
"epoch": 0.6413280116110305,
"grad_norm": 2.1875,
"learning_rate": 0.0004965809952359034,
"loss": 5.6428,
"mean_token_accuracy": 0.15143994837999344,
"num_tokens": 14695672.0,
"step": 7070
},
{
"entropy": 5.774364948272705,
"epoch": 0.6417815674891146,
"grad_norm": 3.078125,
"learning_rate": 0.0004965753737004309,
"loss": 5.5628,
"mean_token_accuracy": 0.16613354682922363,
"num_tokens": 14705160.0,
"step": 7075
},
{
"entropy": 5.745255422592163,
"epoch": 0.6422351233671988,
"grad_norm": 3.6875,
"learning_rate": 0.0004965697475827231,
"loss": 5.6138,
"mean_token_accuracy": 0.1568262130022049,
"num_tokens": 14715487.0,
"step": 7080
},
{
"entropy": 5.786352157592773,
"epoch": 0.6426886792452831,
"grad_norm": 2.140625,
"learning_rate": 0.0004965641168828965,
"loss": 5.6693,
"mean_token_accuracy": 0.15273340791463852,
"num_tokens": 14726987.0,
"step": 7085
},
{
"entropy": 5.734464120864868,
"epoch": 0.6431422351233672,
"grad_norm": 2.53125,
"learning_rate": 0.0004965584816010676,
"loss": 5.6277,
"mean_token_accuracy": 0.16173090487718583,
"num_tokens": 14737810.0,
"step": 7090
},
{
"entropy": 5.791029024124145,
"epoch": 0.6435957910014514,
"grad_norm": 2.234375,
"learning_rate": 0.0004965528417373528,
"loss": 5.6363,
"mean_token_accuracy": 0.1521812468767166,
"num_tokens": 14748997.0,
"step": 7095
},
{
"entropy": 5.811992931365967,
"epoch": 0.6440493468795355,
"grad_norm": 4.25,
"learning_rate": 0.0004965471972918688,
"loss": 5.6477,
"mean_token_accuracy": 0.1499110609292984,
"num_tokens": 14760439.0,
"step": 7100
},
{
"entropy": 5.744043684005737,
"epoch": 0.6445029027576198,
"grad_norm": 3.3125,
"learning_rate": 0.0004965415482647322,
"loss": 5.5695,
"mean_token_accuracy": 0.15322099328041078,
"num_tokens": 14770828.0,
"step": 7105
},
{
"entropy": 5.692976188659668,
"epoch": 0.6449564586357039,
"grad_norm": 3.40625,
"learning_rate": 0.00049653589465606,
"loss": 5.7056,
"mean_token_accuracy": 0.15300424844026567,
"num_tokens": 14781753.0,
"step": 7110
},
{
"entropy": 5.834149742126465,
"epoch": 0.6454100145137881,
"grad_norm": 2.640625,
"learning_rate": 0.000496530236465969,
"loss": 5.7219,
"mean_token_accuracy": 0.14937618225812913,
"num_tokens": 14792241.0,
"step": 7115
},
{
"entropy": 5.750567626953125,
"epoch": 0.6458635703918723,
"grad_norm": 2.859375,
"learning_rate": 0.0004965245736945763,
"loss": 5.5366,
"mean_token_accuracy": 0.15907809436321257,
"num_tokens": 14802580.0,
"step": 7120
},
{
"entropy": 5.771503925323486,
"epoch": 0.6463171262699564,
"grad_norm": 2.5,
"learning_rate": 0.0004965189063419989,
"loss": 5.7252,
"mean_token_accuracy": 0.15481867045164108,
"num_tokens": 14813103.0,
"step": 7125
},
{
"entropy": 5.722277021408081,
"epoch": 0.6467706821480407,
"grad_norm": 2.453125,
"learning_rate": 0.0004965132344083541,
"loss": 5.6078,
"mean_token_accuracy": 0.15793556421995164,
"num_tokens": 14823334.0,
"step": 7130
},
{
"entropy": 5.79091854095459,
"epoch": 0.6472242380261248,
"grad_norm": 4.65625,
"learning_rate": 0.0004965075578937593,
"loss": 5.6483,
"mean_token_accuracy": 0.15331077128648757,
"num_tokens": 14834850.0,
"step": 7135
},
{
"entropy": 5.806456232070923,
"epoch": 0.647677793904209,
"grad_norm": 2.765625,
"learning_rate": 0.0004965018767983316,
"loss": 5.6496,
"mean_token_accuracy": 0.15626297444105147,
"num_tokens": 14844276.0,
"step": 7140
},
{
"entropy": 5.720260286331177,
"epoch": 0.6481313497822931,
"grad_norm": 2.71875,
"learning_rate": 0.0004964961911221886,
"loss": 5.5226,
"mean_token_accuracy": 0.16832068115472792,
"num_tokens": 14854352.0,
"step": 7145
},
{
"entropy": 5.799531936645508,
"epoch": 0.6485849056603774,
"grad_norm": 2.5625,
"learning_rate": 0.000496490500865448,
"loss": 5.6711,
"mean_token_accuracy": 0.15777702629566193,
"num_tokens": 14864395.0,
"step": 7150
},
{
"entropy": 5.791047620773315,
"epoch": 0.6490384615384616,
"grad_norm": 1.8046875,
"learning_rate": 0.0004964848060282275,
"loss": 5.5851,
"mean_token_accuracy": 0.1580512434244156,
"num_tokens": 14874373.0,
"step": 7155
},
{
"entropy": 5.729786777496338,
"epoch": 0.6494920174165457,
"grad_norm": 2.359375,
"learning_rate": 0.0004964791066106447,
"loss": 5.5609,
"mean_token_accuracy": 0.15438223034143447,
"num_tokens": 14884932.0,
"step": 7160
},
{
"entropy": 5.746135950088501,
"epoch": 0.6499455732946299,
"grad_norm": 4.21875,
"learning_rate": 0.0004964734026128175,
"loss": 5.7285,
"mean_token_accuracy": 0.1496640130877495,
"num_tokens": 14896254.0,
"step": 7165
},
{
"entropy": 5.825554847717285,
"epoch": 0.650399129172714,
"grad_norm": 2.953125,
"learning_rate": 0.0004964676940348638,
"loss": 5.6592,
"mean_token_accuracy": 0.151274374127388,
"num_tokens": 14907392.0,
"step": 7170
},
{
"entropy": 5.8277100086212155,
"epoch": 0.6508526850507983,
"grad_norm": 2.546875,
"learning_rate": 0.000496461980876902,
"loss": 5.7629,
"mean_token_accuracy": 0.1563155859708786,
"num_tokens": 14917676.0,
"step": 7175
},
{
"entropy": 5.813205575942993,
"epoch": 0.6513062409288825,
"grad_norm": 2.9375,
"learning_rate": 0.0004964562631390498,
"loss": 5.6093,
"mean_token_accuracy": 0.15569815784692764,
"num_tokens": 14928500.0,
"step": 7180
},
{
"entropy": 5.776795959472656,
"epoch": 0.6517597968069666,
"grad_norm": 2.046875,
"learning_rate": 0.0004964505408214257,
"loss": 5.5776,
"mean_token_accuracy": 0.1599561631679535,
"num_tokens": 14938970.0,
"step": 7185
},
{
"entropy": 5.759690141677856,
"epoch": 0.6522133526850508,
"grad_norm": 2.703125,
"learning_rate": 0.000496444813924148,
"loss": 5.6413,
"mean_token_accuracy": 0.15742683559656143,
"num_tokens": 14949017.0,
"step": 7190
},
{
"entropy": 5.756869697570801,
"epoch": 0.652666908563135,
"grad_norm": 2.453125,
"learning_rate": 0.0004964390824473351,
"loss": 5.5752,
"mean_token_accuracy": 0.16529665887355804,
"num_tokens": 14958529.0,
"step": 7195
},
{
"entropy": 5.754773473739624,
"epoch": 0.6531204644412192,
"grad_norm": 3.046875,
"learning_rate": 0.0004964333463911056,
"loss": 5.6542,
"mean_token_accuracy": 0.15476091653108598,
"num_tokens": 14968044.0,
"step": 7200
},
{
"entropy": 5.767954730987549,
"epoch": 0.6535740203193033,
"grad_norm": 4.59375,
"learning_rate": 0.0004964276057555779,
"loss": 5.6874,
"mean_token_accuracy": 0.1521929308772087,
"num_tokens": 14977867.0,
"step": 7205
},
{
"entropy": 5.851280736923218,
"epoch": 0.6540275761973875,
"grad_norm": 2.421875,
"learning_rate": 0.000496421860540871,
"loss": 5.6771,
"mean_token_accuracy": 0.15839732587337493,
"num_tokens": 14986936.0,
"step": 7210
},
{
"entropy": 5.7605140686035154,
"epoch": 0.6544811320754716,
"grad_norm": 3.015625,
"learning_rate": 0.0004964161107471036,
"loss": 5.6302,
"mean_token_accuracy": 0.16003877222537993,
"num_tokens": 14996681.0,
"step": 7215
},
{
"entropy": 5.780618619918823,
"epoch": 0.6549346879535559,
"grad_norm": 2.1875,
"learning_rate": 0.0004964103563743946,
"loss": 5.6416,
"mean_token_accuracy": 0.1552106834948063,
"num_tokens": 15007100.0,
"step": 7220
},
{
"entropy": 5.831463623046875,
"epoch": 0.6553882438316401,
"grad_norm": 2.1875,
"learning_rate": 0.000496404597422863,
"loss": 5.7213,
"mean_token_accuracy": 0.15619465559720994,
"num_tokens": 15017452.0,
"step": 7225
},
{
"entropy": 5.757536935806274,
"epoch": 0.6558417997097242,
"grad_norm": 2.578125,
"learning_rate": 0.000496398833892628,
"loss": 5.579,
"mean_token_accuracy": 0.15511720776557922,
"num_tokens": 15028362.0,
"step": 7230
},
{
"entropy": 5.809765815734863,
"epoch": 0.6562953555878084,
"grad_norm": 7.5,
"learning_rate": 0.0004963930657838086,
"loss": 5.663,
"mean_token_accuracy": 0.1546101078391075,
"num_tokens": 15038691.0,
"step": 7235
},
{
"entropy": 5.753928279876709,
"epoch": 0.6567489114658926,
"grad_norm": 2.09375,
"learning_rate": 0.0004963872930965242,
"loss": 5.6387,
"mean_token_accuracy": 0.15518327057361603,
"num_tokens": 15050037.0,
"step": 7240
},
{
"entropy": 5.745072507858277,
"epoch": 0.6572024673439768,
"grad_norm": 4.625,
"learning_rate": 0.0004963815158308941,
"loss": 5.6003,
"mean_token_accuracy": 0.1518049106001854,
"num_tokens": 15059793.0,
"step": 7245
},
{
"entropy": 5.815020132064819,
"epoch": 0.657656023222061,
"grad_norm": 3.421875,
"learning_rate": 0.0004963757339870379,
"loss": 5.7025,
"mean_token_accuracy": 0.14955889880657197,
"num_tokens": 15070500.0,
"step": 7250
},
{
"entropy": 5.865824222564697,
"epoch": 0.6581095791001451,
"grad_norm": 2.6875,
"learning_rate": 0.0004963699475650751,
"loss": 5.7522,
"mean_token_accuracy": 0.15184831321239473,
"num_tokens": 15081109.0,
"step": 7255
},
{
"entropy": 5.785812902450561,
"epoch": 0.6585631349782293,
"grad_norm": 2.078125,
"learning_rate": 0.0004963641565651255,
"loss": 5.6338,
"mean_token_accuracy": 0.16198570430278778,
"num_tokens": 15092315.0,
"step": 7260
},
{
"entropy": 5.797928714752198,
"epoch": 0.6590166908563135,
"grad_norm": 2.203125,
"learning_rate": 0.0004963583609873084,
"loss": 5.7092,
"mean_token_accuracy": 0.15403601601719857,
"num_tokens": 15103957.0,
"step": 7265
},
{
"entropy": 5.831661939620972,
"epoch": 0.6594702467343977,
"grad_norm": 2.671875,
"learning_rate": 0.0004963525608317443,
"loss": 5.6641,
"mean_token_accuracy": 0.1534239813685417,
"num_tokens": 15114996.0,
"step": 7270
},
{
"entropy": 5.83020806312561,
"epoch": 0.6599238026124818,
"grad_norm": 2.46875,
"learning_rate": 0.0004963467560985527,
"loss": 5.6124,
"mean_token_accuracy": 0.15501478910446168,
"num_tokens": 15125098.0,
"step": 7275
},
{
"entropy": 5.744346618652344,
"epoch": 0.660377358490566,
"grad_norm": 1.96875,
"learning_rate": 0.0004963409467878538,
"loss": 5.606,
"mean_token_accuracy": 0.15144749432802201,
"num_tokens": 15134746.0,
"step": 7280
},
{
"entropy": 5.809712219238281,
"epoch": 0.6608309143686503,
"grad_norm": 2.71875,
"learning_rate": 0.0004963351328997678,
"loss": 5.6385,
"mean_token_accuracy": 0.15188922137022018,
"num_tokens": 15145698.0,
"step": 7285
},
{
"entropy": 5.797967863082886,
"epoch": 0.6612844702467344,
"grad_norm": 2.078125,
"learning_rate": 0.0004963293144344148,
"loss": 5.6861,
"mean_token_accuracy": 0.15129384547472,
"num_tokens": 15156026.0,
"step": 7290
},
{
"entropy": 5.770978927612305,
"epoch": 0.6617380261248186,
"grad_norm": 1.953125,
"learning_rate": 0.0004963234913919151,
"loss": 5.6272,
"mean_token_accuracy": 0.15003437101840972,
"num_tokens": 15165900.0,
"step": 7295
},
{
"entropy": 5.750339603424072,
"epoch": 0.6621915820029027,
"grad_norm": 5.25,
"learning_rate": 0.0004963176637723893,
"loss": 5.7027,
"mean_token_accuracy": 0.15393707156181335,
"num_tokens": 15176693.0,
"step": 7300
},
{
"entropy": 5.836119508743286,
"epoch": 0.6626451378809869,
"grad_norm": 2.46875,
"learning_rate": 0.0004963118315759578,
"loss": 5.7345,
"mean_token_accuracy": 0.14329422190785407,
"num_tokens": 15186535.0,
"step": 7305
},
{
"entropy": 5.836709785461426,
"epoch": 0.6630986937590712,
"grad_norm": 3.0625,
"learning_rate": 0.0004963059948027411,
"loss": 5.6336,
"mean_token_accuracy": 0.15115350857377052,
"num_tokens": 15196019.0,
"step": 7310
},
{
"entropy": 5.738832950592041,
"epoch": 0.6635522496371553,
"grad_norm": 2.453125,
"learning_rate": 0.0004963001534528601,
"loss": 5.4937,
"mean_token_accuracy": 0.16718341410160065,
"num_tokens": 15205273.0,
"step": 7315
},
{
"entropy": 5.790439748764038,
"epoch": 0.6640058055152395,
"grad_norm": 3.765625,
"learning_rate": 0.0004962943075264355,
"loss": 5.6465,
"mean_token_accuracy": 0.1561621308326721,
"num_tokens": 15216716.0,
"step": 7320
},
{
"entropy": 5.835460090637207,
"epoch": 0.6644593613933236,
"grad_norm": 2.90625,
"learning_rate": 0.0004962884570235884,
"loss": 5.6903,
"mean_token_accuracy": 0.15185065120458602,
"num_tokens": 15227306.0,
"step": 7325
},
{
"entropy": 5.701790475845337,
"epoch": 0.6649129172714079,
"grad_norm": 3.0,
"learning_rate": 0.0004962826019444394,
"loss": 5.5503,
"mean_token_accuracy": 0.1638151526451111,
"num_tokens": 15237302.0,
"step": 7330
},
{
"entropy": 5.720064258575439,
"epoch": 0.665366473149492,
"grad_norm": 2.859375,
"learning_rate": 0.0004962767422891099,
"loss": 5.6099,
"mean_token_accuracy": 0.15619055330753326,
"num_tokens": 15247064.0,
"step": 7335
},
{
"entropy": 5.794697284698486,
"epoch": 0.6658200290275762,
"grad_norm": 2.578125,
"learning_rate": 0.000496270878057721,
"loss": 5.6748,
"mean_token_accuracy": 0.15739299803972245,
"num_tokens": 15257006.0,
"step": 7340
},
{
"entropy": 5.824474954605103,
"epoch": 0.6662735849056604,
"grad_norm": 2.109375,
"learning_rate": 0.0004962650092503939,
"loss": 5.6343,
"mean_token_accuracy": 0.16214921474456787,
"num_tokens": 15266720.0,
"step": 7345
},
{
"entropy": 5.731961631774903,
"epoch": 0.6667271407837445,
"grad_norm": 2.28125,
"learning_rate": 0.0004962591358672502,
"loss": 5.6749,
"mean_token_accuracy": 0.15213515236973763,
"num_tokens": 15277330.0,
"step": 7350
},
{
"entropy": 5.809416246414185,
"epoch": 0.6671806966618288,
"grad_norm": 3.234375,
"learning_rate": 0.000496253257908411,
"loss": 5.6045,
"mean_token_accuracy": 0.15907943099737168,
"num_tokens": 15287401.0,
"step": 7355
},
{
"entropy": 5.687846374511719,
"epoch": 0.6676342525399129,
"grad_norm": 2.0,
"learning_rate": 0.0004962473753739981,
"loss": 5.5649,
"mean_token_accuracy": 0.1594688728451729,
"num_tokens": 15297426.0,
"step": 7360
},
{
"entropy": 5.793103313446045,
"epoch": 0.6680878084179971,
"grad_norm": 2.28125,
"learning_rate": 0.000496241488264133,
"loss": 5.5589,
"mean_token_accuracy": 0.15536411553621293,
"num_tokens": 15308412.0,
"step": 7365
},
{
"entropy": 5.794301748275757,
"epoch": 0.6685413642960812,
"grad_norm": 2.640625,
"learning_rate": 0.0004962355965789377,
"loss": 5.5857,
"mean_token_accuracy": 0.15055083632469177,
"num_tokens": 15318821.0,
"step": 7370
},
{
"entropy": 5.731050777435303,
"epoch": 0.6689949201741655,
"grad_norm": 3.84375,
"learning_rate": 0.0004962297003185338,
"loss": 5.6394,
"mean_token_accuracy": 0.15725694745779037,
"num_tokens": 15328737.0,
"step": 7375
},
{
"entropy": 5.831857347488404,
"epoch": 0.6694484760522497,
"grad_norm": 3.0625,
"learning_rate": 0.0004962237994830433,
"loss": 5.7373,
"mean_token_accuracy": 0.1487860843539238,
"num_tokens": 15339695.0,
"step": 7380
},
{
"entropy": 5.857686281204224,
"epoch": 0.6699020319303338,
"grad_norm": 2.0,
"learning_rate": 0.0004962178940725883,
"loss": 5.6736,
"mean_token_accuracy": 0.15191865265369414,
"num_tokens": 15350270.0,
"step": 7385
},
{
"entropy": 5.78417935371399,
"epoch": 0.670355587808418,
"grad_norm": 3.09375,
"learning_rate": 0.0004962119840872908,
"loss": 5.6725,
"mean_token_accuracy": 0.1553597405552864,
"num_tokens": 15360986.0,
"step": 7390
},
{
"entropy": 5.780443906784058,
"epoch": 0.6708091436865021,
"grad_norm": 2.03125,
"learning_rate": 0.0004962060695272733,
"loss": 5.6863,
"mean_token_accuracy": 0.15490424484014512,
"num_tokens": 15371786.0,
"step": 7395
},
{
"entropy": 5.773851728439331,
"epoch": 0.6712626995645864,
"grad_norm": 4.1875,
"learning_rate": 0.0004962001503926577,
"loss": 5.6349,
"mean_token_accuracy": 0.15896140784025192,
"num_tokens": 15382960.0,
"step": 7400
},
{
"entropy": 5.738351774215698,
"epoch": 0.6717162554426706,
"grad_norm": 2.15625,
"learning_rate": 0.0004961942266835667,
"loss": 5.6209,
"mean_token_accuracy": 0.1595415860414505,
"num_tokens": 15392807.0,
"step": 7405
},
{
"entropy": 5.794447135925293,
"epoch": 0.6721698113207547,
"grad_norm": 2.171875,
"learning_rate": 0.0004961882984001227,
"loss": 5.6749,
"mean_token_accuracy": 0.15016180127859116,
"num_tokens": 15403712.0,
"step": 7410
},
{
"entropy": 5.822537660598755,
"epoch": 0.6726233671988389,
"grad_norm": 1.9765625,
"learning_rate": 0.0004961823655424484,
"loss": 5.7309,
"mean_token_accuracy": 0.14892438873648645,
"num_tokens": 15414839.0,
"step": 7415
},
{
"entropy": 5.800685882568359,
"epoch": 0.6730769230769231,
"grad_norm": 2.171875,
"learning_rate": 0.0004961764281106664,
"loss": 5.6032,
"mean_token_accuracy": 0.16113678514957427,
"num_tokens": 15425320.0,
"step": 7420
},
{
"entropy": 5.759464979171753,
"epoch": 0.6735304789550073,
"grad_norm": 2.328125,
"learning_rate": 0.0004961704861048995,
"loss": 5.6071,
"mean_token_accuracy": 0.15447303056716918,
"num_tokens": 15435476.0,
"step": 7425
},
{
"entropy": 5.791233015060425,
"epoch": 0.6739840348330914,
"grad_norm": 2.34375,
"learning_rate": 0.0004961645395252705,
"loss": 5.6996,
"mean_token_accuracy": 0.15540242716670036,
"num_tokens": 15445786.0,
"step": 7430
},
{
"entropy": 5.772368955612182,
"epoch": 0.6744375907111756,
"grad_norm": 2.453125,
"learning_rate": 0.0004961585883719025,
"loss": 5.7077,
"mean_token_accuracy": 0.15462125688791276,
"num_tokens": 15455915.0,
"step": 7435
},
{
"entropy": 5.828447437286377,
"epoch": 0.6748911465892597,
"grad_norm": 2.40625,
"learning_rate": 0.0004961526326449186,
"loss": 5.675,
"mean_token_accuracy": 0.15044990181922913,
"num_tokens": 15466523.0,
"step": 7440
},
{
"entropy": 5.856289434432983,
"epoch": 0.675344702467344,
"grad_norm": 3.46875,
"learning_rate": 0.0004961466723444419,
"loss": 5.6557,
"mean_token_accuracy": 0.15675067156553268,
"num_tokens": 15476821.0,
"step": 7445
},
{
"entropy": 5.733017301559448,
"epoch": 0.6757982583454282,
"grad_norm": 1.9609375,
"learning_rate": 0.0004961407074705955,
"loss": 5.6031,
"mean_token_accuracy": 0.15133341252803803,
"num_tokens": 15487182.0,
"step": 7450
},
{
"entropy": 5.766598081588745,
"epoch": 0.6762518142235123,
"grad_norm": 2.40625,
"learning_rate": 0.0004961347380235031,
"loss": 5.584,
"mean_token_accuracy": 0.15954807996749878,
"num_tokens": 15497600.0,
"step": 7455
},
{
"entropy": 5.799952602386474,
"epoch": 0.6767053701015965,
"grad_norm": 2.21875,
"learning_rate": 0.0004961287640032879,
"loss": 5.7363,
"mean_token_accuracy": 0.14449917227029802,
"num_tokens": 15508278.0,
"step": 7460
},
{
"entropy": 5.854372262954712,
"epoch": 0.6771589259796807,
"grad_norm": 3.84375,
"learning_rate": 0.0004961227854100734,
"loss": 5.744,
"mean_token_accuracy": 0.15246044844388962,
"num_tokens": 15518458.0,
"step": 7465
},
{
"entropy": 5.7955224990844725,
"epoch": 0.6776124818577649,
"grad_norm": 2.0,
"learning_rate": 0.0004961168022439835,
"loss": 5.5636,
"mean_token_accuracy": 0.1587345838546753,
"num_tokens": 15528738.0,
"step": 7470
},
{
"entropy": 5.739163064956665,
"epoch": 0.6780660377358491,
"grad_norm": 1.7890625,
"learning_rate": 0.0004961108145051417,
"loss": 5.6104,
"mean_token_accuracy": 0.16207170635461807,
"num_tokens": 15538656.0,
"step": 7475
},
{
"entropy": 5.684577178955078,
"epoch": 0.6785195936139332,
"grad_norm": 2.21875,
"learning_rate": 0.000496104822193672,
"loss": 5.6305,
"mean_token_accuracy": 0.15659867376089096,
"num_tokens": 15549285.0,
"step": 7480
},
{
"entropy": 5.828121995925903,
"epoch": 0.6789731494920174,
"grad_norm": 2.421875,
"learning_rate": 0.0004960988253096982,
"loss": 5.6925,
"mean_token_accuracy": 0.16133199483156205,
"num_tokens": 15559766.0,
"step": 7485
},
{
"entropy": 5.793510580062867,
"epoch": 0.6794267053701016,
"grad_norm": 2.25,
"learning_rate": 0.0004960928238533443,
"loss": 5.5625,
"mean_token_accuracy": 0.16435614079236985,
"num_tokens": 15570873.0,
"step": 7490
},
{
"entropy": 5.804307985305786,
"epoch": 0.6798802612481858,
"grad_norm": 3.015625,
"learning_rate": 0.0004960868178247344,
"loss": 5.8058,
"mean_token_accuracy": 0.15049900710582734,
"num_tokens": 15581229.0,
"step": 7495
},
{
"entropy": 5.682524824142456,
"epoch": 0.68033381712627,
"grad_norm": 2.328125,
"learning_rate": 0.0004960808072239929,
"loss": 5.6645,
"mean_token_accuracy": 0.150663161277771,
"num_tokens": 15591686.0,
"step": 7500
},
{
"entropy": 5.781607913970947,
"epoch": 0.6807873730043541,
"grad_norm": 1.8359375,
"learning_rate": 0.0004960747920512439,
"loss": 5.6798,
"mean_token_accuracy": 0.15253570526838303,
"num_tokens": 15601869.0,
"step": 7505
},
{
"entropy": 5.774205732345581,
"epoch": 0.6812409288824384,
"grad_norm": 1.7265625,
"learning_rate": 0.000496068772306612,
"loss": 5.6678,
"mean_token_accuracy": 0.1541148066520691,
"num_tokens": 15612946.0,
"step": 7510
},
{
"entropy": 5.832863903045654,
"epoch": 0.6816944847605225,
"grad_norm": 2.0625,
"learning_rate": 0.0004960627479902214,
"loss": 5.7158,
"mean_token_accuracy": 0.14823995307087898,
"num_tokens": 15623761.0,
"step": 7515
},
{
"entropy": 5.809633827209472,
"epoch": 0.6821480406386067,
"grad_norm": 1.9921875,
"learning_rate": 0.000496056719102197,
"loss": 5.6913,
"mean_token_accuracy": 0.14876921027898787,
"num_tokens": 15634500.0,
"step": 7520
},
{
"entropy": 5.738462495803833,
"epoch": 0.6826015965166908,
"grad_norm": 1.8125,
"learning_rate": 0.0004960506856426631,
"loss": 5.576,
"mean_token_accuracy": 0.15111364424228668,
"num_tokens": 15644896.0,
"step": 7525
},
{
"entropy": 5.787351369857788,
"epoch": 0.683055152394775,
"grad_norm": 2.296875,
"learning_rate": 0.0004960446476117448,
"loss": 5.6822,
"mean_token_accuracy": 0.16053580194711686,
"num_tokens": 15655008.0,
"step": 7530
},
{
"entropy": 5.746401119232178,
"epoch": 0.6835087082728593,
"grad_norm": 4.46875,
"learning_rate": 0.0004960386050095668,
"loss": 5.6004,
"mean_token_accuracy": 0.14865829050540924,
"num_tokens": 15665944.0,
"step": 7535
},
{
"entropy": 5.801738309860229,
"epoch": 0.6839622641509434,
"grad_norm": 2.421875,
"learning_rate": 0.0004960325578362543,
"loss": 5.6797,
"mean_token_accuracy": 0.15832635313272475,
"num_tokens": 15677099.0,
"step": 7540
},
{
"entropy": 5.798512172698975,
"epoch": 0.6844158200290276,
"grad_norm": 2.6875,
"learning_rate": 0.000496026506091932,
"loss": 5.6344,
"mean_token_accuracy": 0.15838538110256195,
"num_tokens": 15687531.0,
"step": 7545
},
{
"entropy": 5.790757894515991,
"epoch": 0.6848693759071117,
"grad_norm": 2.296875,
"learning_rate": 0.0004960204497767253,
"loss": 5.6583,
"mean_token_accuracy": 0.1546577602624893,
"num_tokens": 15697002.0,
"step": 7550
},
{
"entropy": 5.724558973312378,
"epoch": 0.685322931785196,
"grad_norm": 4.28125,
"learning_rate": 0.0004960143888907594,
"loss": 5.5426,
"mean_token_accuracy": 0.15625495165586473,
"num_tokens": 15707249.0,
"step": 7555
},
{
"entropy": 5.743619441986084,
"epoch": 0.6857764876632801,
"grad_norm": 2.421875,
"learning_rate": 0.0004960083234341595,
"loss": 5.6223,
"mean_token_accuracy": 0.15563429296016693,
"num_tokens": 15718764.0,
"step": 7560
},
{
"entropy": 5.727084493637085,
"epoch": 0.6862300435413643,
"grad_norm": 2.5625,
"learning_rate": 0.0004960022534070513,
"loss": 5.5709,
"mean_token_accuracy": 0.16300940364599228,
"num_tokens": 15728854.0,
"step": 7565
},
{
"entropy": 5.648557186126709,
"epoch": 0.6866835994194485,
"grad_norm": 4.34375,
"learning_rate": 0.00049599617880956,
"loss": 5.5759,
"mean_token_accuracy": 0.1575843796133995,
"num_tokens": 15739682.0,
"step": 7570
},
{
"entropy": 5.703317785263062,
"epoch": 0.6871371552975326,
"grad_norm": 2.09375,
"learning_rate": 0.0004959900996418115,
"loss": 5.5863,
"mean_token_accuracy": 0.16021239310503005,
"num_tokens": 15749628.0,
"step": 7575
},
{
"entropy": 5.804837465286255,
"epoch": 0.6875907111756169,
"grad_norm": 3.359375,
"learning_rate": 0.0004959840159039315,
"loss": 5.6567,
"mean_token_accuracy": 0.1535927399992943,
"num_tokens": 15760137.0,
"step": 7580
},
{
"entropy": 5.777477645874024,
"epoch": 0.688044267053701,
"grad_norm": 2.109375,
"learning_rate": 0.0004959779275960456,
"loss": 5.6112,
"mean_token_accuracy": 0.1556934803724289,
"num_tokens": 15769889.0,
"step": 7585
},
{
"entropy": 5.738125896453857,
"epoch": 0.6884978229317852,
"grad_norm": 1.984375,
"learning_rate": 0.00049597183471828,
"loss": 5.7256,
"mean_token_accuracy": 0.15178040713071822,
"num_tokens": 15780912.0,
"step": 7590
},
{
"entropy": 5.850151681900025,
"epoch": 0.6889513788098693,
"grad_norm": 2.078125,
"learning_rate": 0.0004959657372707604,
"loss": 5.5925,
"mean_token_accuracy": 0.16462052166461943,
"num_tokens": 15791719.0,
"step": 7595
},
{
"entropy": 5.8561851501464846,
"epoch": 0.6894049346879536,
"grad_norm": 2.109375,
"learning_rate": 0.0004959596352536132,
"loss": 5.6261,
"mean_token_accuracy": 0.15682563334703445,
"num_tokens": 15802551.0,
"step": 7600
},
{
"entropy": 5.782931661605835,
"epoch": 0.6898584905660378,
"grad_norm": 1.8515625,
"learning_rate": 0.0004959535286669643,
"loss": 5.6941,
"mean_token_accuracy": 0.15105319619178773,
"num_tokens": 15813551.0,
"step": 7605
},
{
"entropy": 5.715517950057984,
"epoch": 0.6903120464441219,
"grad_norm": 2.78125,
"learning_rate": 0.0004959474175109402,
"loss": 5.6681,
"mean_token_accuracy": 0.15595430433750151,
"num_tokens": 15824262.0,
"step": 7610
},
{
"entropy": 5.818006896972657,
"epoch": 0.6907656023222061,
"grad_norm": 1.984375,
"learning_rate": 0.0004959413017856672,
"loss": 5.7191,
"mean_token_accuracy": 0.15502288639545442,
"num_tokens": 15835493.0,
"step": 7615
},
{
"entropy": 5.697161531448364,
"epoch": 0.6912191582002902,
"grad_norm": 2.359375,
"learning_rate": 0.0004959351814912716,
"loss": 5.5857,
"mean_token_accuracy": 0.15691763609647752,
"num_tokens": 15846224.0,
"step": 7620
},
{
"entropy": 5.754197263717652,
"epoch": 0.6916727140783745,
"grad_norm": 2.203125,
"learning_rate": 0.0004959290566278803,
"loss": 5.6271,
"mean_token_accuracy": 0.15909348279237748,
"num_tokens": 15855622.0,
"step": 7625
},
{
"entropy": 5.813883924484253,
"epoch": 0.6921262699564587,
"grad_norm": 1.9765625,
"learning_rate": 0.0004959229271956199,
"loss": 5.6805,
"mean_token_accuracy": 0.15607887357473374,
"num_tokens": 15865835.0,
"step": 7630
},
{
"entropy": 5.830732297897339,
"epoch": 0.6925798258345428,
"grad_norm": 2.484375,
"learning_rate": 0.0004959167931946169,
"loss": 5.6201,
"mean_token_accuracy": 0.15640912801027299,
"num_tokens": 15875915.0,
"step": 7635
},
{
"entropy": 5.825406503677368,
"epoch": 0.693033381712627,
"grad_norm": 1.859375,
"learning_rate": 0.0004959106546249984,
"loss": 5.6337,
"mean_token_accuracy": 0.15450009256601333,
"num_tokens": 15887111.0,
"step": 7640
},
{
"entropy": 5.8027575492858885,
"epoch": 0.6934869375907112,
"grad_norm": 2.25,
"learning_rate": 0.0004959045114868913,
"loss": 5.7488,
"mean_token_accuracy": 0.14544340074062348,
"num_tokens": 15899528.0,
"step": 7645
},
{
"entropy": 5.752656936645508,
"epoch": 0.6939404934687954,
"grad_norm": 1.7109375,
"learning_rate": 0.0004958983637804225,
"loss": 5.6388,
"mean_token_accuracy": 0.16282033026218415,
"num_tokens": 15910881.0,
"step": 7650
},
{
"entropy": 5.816255569458008,
"epoch": 0.6943940493468795,
"grad_norm": 2.015625,
"learning_rate": 0.0004958922115057194,
"loss": 5.6955,
"mean_token_accuracy": 0.15611306130886077,
"num_tokens": 15921189.0,
"step": 7655
},
{
"entropy": 5.8625730037689205,
"epoch": 0.6948476052249637,
"grad_norm": 1.875,
"learning_rate": 0.0004958860546629089,
"loss": 5.6923,
"mean_token_accuracy": 0.15917486399412156,
"num_tokens": 15931276.0,
"step": 7660
},
{
"entropy": 5.844645547866821,
"epoch": 0.6953011611030478,
"grad_norm": 3.328125,
"learning_rate": 0.0004958798932521186,
"loss": 5.7599,
"mean_token_accuracy": 0.1465900495648384,
"num_tokens": 15941506.0,
"step": 7665
},
{
"entropy": 5.710515451431275,
"epoch": 0.6957547169811321,
"grad_norm": 2.265625,
"learning_rate": 0.0004958737272734759,
"loss": 5.5973,
"mean_token_accuracy": 0.1594126984477043,
"num_tokens": 15951158.0,
"step": 7670
},
{
"entropy": 5.798179674148559,
"epoch": 0.6962082728592163,
"grad_norm": 2.171875,
"learning_rate": 0.0004958675567271081,
"loss": 5.6198,
"mean_token_accuracy": 0.16269323378801345,
"num_tokens": 15960852.0,
"step": 7675
},
{
"entropy": 5.81680121421814,
"epoch": 0.6966618287373004,
"grad_norm": 2.46875,
"learning_rate": 0.000495861381613143,
"loss": 5.6278,
"mean_token_accuracy": 0.1538613721728325,
"num_tokens": 15971291.0,
"step": 7680
},
{
"entropy": 5.788067293167114,
"epoch": 0.6971153846153846,
"grad_norm": 2.5625,
"learning_rate": 0.0004958552019317082,
"loss": 5.6985,
"mean_token_accuracy": 0.15725067257881165,
"num_tokens": 15982180.0,
"step": 7685
},
{
"entropy": 5.736803913116455,
"epoch": 0.6975689404934688,
"grad_norm": 2.75,
"learning_rate": 0.0004958490176829316,
"loss": 5.5857,
"mean_token_accuracy": 0.15838269367814065,
"num_tokens": 15992363.0,
"step": 7690
},
{
"entropy": 5.7980883598327635,
"epoch": 0.698022496371553,
"grad_norm": 10.5625,
"learning_rate": 0.000495842828866941,
"loss": 5.6489,
"mean_token_accuracy": 0.15137666910886766,
"num_tokens": 16003477.0,
"step": 7695
},
{
"entropy": 5.805958795547485,
"epoch": 0.6984760522496372,
"grad_norm": 2.796875,
"learning_rate": 0.0004958366354838645,
"loss": 5.6463,
"mean_token_accuracy": 0.15117594450712205,
"num_tokens": 16013747.0,
"step": 7700
},
{
"entropy": 5.774502849578857,
"epoch": 0.6989296081277213,
"grad_norm": 2.484375,
"learning_rate": 0.00049583043753383,
"loss": 5.6676,
"mean_token_accuracy": 0.1588193356990814,
"num_tokens": 16025662.0,
"step": 7705
},
{
"entropy": 5.7774632453918455,
"epoch": 0.6993831640058055,
"grad_norm": 2.5625,
"learning_rate": 0.0004958242350169658,
"loss": 5.7169,
"mean_token_accuracy": 0.1599838688969612,
"num_tokens": 16037315.0,
"step": 7710
},
{
"entropy": 5.695142984390259,
"epoch": 0.6998367198838897,
"grad_norm": 2.375,
"learning_rate": 0.0004958180279334001,
"loss": 5.5541,
"mean_token_accuracy": 0.15912531763315202,
"num_tokens": 16047297.0,
"step": 7715
},
{
"entropy": 5.766730165481567,
"epoch": 0.7002902757619739,
"grad_norm": 5.4375,
"learning_rate": 0.0004958118162832614,
"loss": 5.5364,
"mean_token_accuracy": 0.1601413071155548,
"num_tokens": 16057573.0,
"step": 7720
},
{
"entropy": 5.7812422752380375,
"epoch": 0.700743831640058,
"grad_norm": 1.7734375,
"learning_rate": 0.0004958056000666781,
"loss": 5.6859,
"mean_token_accuracy": 0.15329635739326478,
"num_tokens": 16067926.0,
"step": 7725
},
{
"entropy": 5.773109102249146,
"epoch": 0.7011973875181422,
"grad_norm": 2.03125,
"learning_rate": 0.0004957993792837786,
"loss": 5.626,
"mean_token_accuracy": 0.1544777363538742,
"num_tokens": 16077938.0,
"step": 7730
},
{
"entropy": 5.742027997970581,
"epoch": 0.7016509433962265,
"grad_norm": 1.7578125,
"learning_rate": 0.0004957931539346917,
"loss": 5.5164,
"mean_token_accuracy": 0.15980168730020522,
"num_tokens": 16088163.0,
"step": 7735
},
{
"entropy": 5.800528526306152,
"epoch": 0.7021044992743106,
"grad_norm": 2.484375,
"learning_rate": 0.0004957869240195462,
"loss": 5.7459,
"mean_token_accuracy": 0.1448234349489212,
"num_tokens": 16100640.0,
"step": 7740
},
{
"entropy": 5.73759388923645,
"epoch": 0.7025580551523948,
"grad_norm": 1.9765625,
"learning_rate": 0.0004957806895384707,
"loss": 5.5875,
"mean_token_accuracy": 0.15452245771884918,
"num_tokens": 16111018.0,
"step": 7745
},
{
"entropy": 5.731401681900024,
"epoch": 0.7030116110304789,
"grad_norm": 1.8203125,
"learning_rate": 0.0004957744504915944,
"loss": 5.5878,
"mean_token_accuracy": 0.15764452815055846,
"num_tokens": 16120873.0,
"step": 7750
},
{
"entropy": 5.75859522819519,
"epoch": 0.7034651669085631,
"grad_norm": 2.296875,
"learning_rate": 0.0004957682068790463,
"loss": 5.6298,
"mean_token_accuracy": 0.15953876823186874,
"num_tokens": 16131178.0,
"step": 7755
},
{
"entropy": 5.724505662918091,
"epoch": 0.7039187227866474,
"grad_norm": 2.59375,
"learning_rate": 0.0004957619587009553,
"loss": 5.6187,
"mean_token_accuracy": 0.15462536960840226,
"num_tokens": 16141156.0,
"step": 7760
},
{
"entropy": 5.779624652862549,
"epoch": 0.7043722786647315,
"grad_norm": 1.90625,
"learning_rate": 0.0004957557059574507,
"loss": 5.7157,
"mean_token_accuracy": 0.1444093957543373,
"num_tokens": 16150077.0,
"step": 7765
},
{
"entropy": 5.778850507736206,
"epoch": 0.7048258345428157,
"grad_norm": 2.0,
"learning_rate": 0.0004957494486486618,
"loss": 5.6362,
"mean_token_accuracy": 0.1569339767098427,
"num_tokens": 16160514.0,
"step": 7770
},
{
"entropy": 5.798329210281372,
"epoch": 0.7052793904208998,
"grad_norm": 1.7890625,
"learning_rate": 0.0004957431867747182,
"loss": 5.668,
"mean_token_accuracy": 0.16042965203523635,
"num_tokens": 16170811.0,
"step": 7775
},
{
"entropy": 5.755690336227417,
"epoch": 0.7057329462989841,
"grad_norm": 2.890625,
"learning_rate": 0.0004957369203357492,
"loss": 5.5824,
"mean_token_accuracy": 0.15862363576889038,
"num_tokens": 16181553.0,
"step": 7780
},
{
"entropy": 5.736315679550171,
"epoch": 0.7061865021770682,
"grad_norm": 2.734375,
"learning_rate": 0.0004957306493318844,
"loss": 5.6339,
"mean_token_accuracy": 0.16354555636644363,
"num_tokens": 16191308.0,
"step": 7785
},
{
"entropy": 5.707941198348999,
"epoch": 0.7066400580551524,
"grad_norm": 2.78125,
"learning_rate": 0.0004957243737632535,
"loss": 5.5795,
"mean_token_accuracy": 0.1632107749581337,
"num_tokens": 16201311.0,
"step": 7790
},
{
"entropy": 5.817873144149781,
"epoch": 0.7070936139332366,
"grad_norm": 2.4375,
"learning_rate": 0.0004957180936299863,
"loss": 5.6108,
"mean_token_accuracy": 0.1581616669893265,
"num_tokens": 16211105.0,
"step": 7795
},
{
"entropy": 5.818648386001587,
"epoch": 0.7075471698113207,
"grad_norm": 3.375,
"learning_rate": 0.0004957118089322127,
"loss": 5.6237,
"mean_token_accuracy": 0.15291189551353454,
"num_tokens": 16220836.0,
"step": 7800
},
{
"entropy": 5.740155744552612,
"epoch": 0.708000725689405,
"grad_norm": 3.765625,
"learning_rate": 0.0004957055196700628,
"loss": 5.5702,
"mean_token_accuracy": 0.16200145781040193,
"num_tokens": 16231630.0,
"step": 7805
},
{
"entropy": 5.792011737823486,
"epoch": 0.7084542815674891,
"grad_norm": 3.421875,
"learning_rate": 0.0004956992258436663,
"loss": 5.6402,
"mean_token_accuracy": 0.1585893839597702,
"num_tokens": 16242272.0,
"step": 7810
},
{
"entropy": 5.714456701278687,
"epoch": 0.7089078374455733,
"grad_norm": 3.875,
"learning_rate": 0.0004956929274531536,
"loss": 5.5431,
"mean_token_accuracy": 0.15601417273283005,
"num_tokens": 16252555.0,
"step": 7815
},
{
"entropy": 5.744070816040039,
"epoch": 0.7093613933236574,
"grad_norm": 3.1875,
"learning_rate": 0.000495686624498655,
"loss": 5.6631,
"mean_token_accuracy": 0.16189786046743393,
"num_tokens": 16261964.0,
"step": 7820
},
{
"entropy": 5.846288108825684,
"epoch": 0.7098149492017417,
"grad_norm": 2.609375,
"learning_rate": 0.0004956803169803007,
"loss": 5.728,
"mean_token_accuracy": 0.14986019879579543,
"num_tokens": 16274543.0,
"step": 7825
},
{
"entropy": 5.80157732963562,
"epoch": 0.7102685050798259,
"grad_norm": 2.703125,
"learning_rate": 0.0004956740048982212,
"loss": 5.6358,
"mean_token_accuracy": 0.14794817417860032,
"num_tokens": 16285124.0,
"step": 7830
},
{
"entropy": 5.753595542907715,
"epoch": 0.71072206095791,
"grad_norm": 2.25,
"learning_rate": 0.0004956676882525472,
"loss": 5.697,
"mean_token_accuracy": 0.15353426784276963,
"num_tokens": 16295727.0,
"step": 7835
},
{
"entropy": 5.766923284530639,
"epoch": 0.7111756168359942,
"grad_norm": 2.015625,
"learning_rate": 0.000495661367043409,
"loss": 5.6115,
"mean_token_accuracy": 0.16063830107450486,
"num_tokens": 16305496.0,
"step": 7840
},
{
"entropy": 5.7579789638519285,
"epoch": 0.7116291727140783,
"grad_norm": 2.203125,
"learning_rate": 0.0004956550412709376,
"loss": 5.582,
"mean_token_accuracy": 0.16114101260900499,
"num_tokens": 16315112.0,
"step": 7845
},
{
"entropy": 5.813846206665039,
"epoch": 0.7120827285921626,
"grad_norm": 3.171875,
"learning_rate": 0.0004956487109352637,
"loss": 5.6442,
"mean_token_accuracy": 0.16229847073554993,
"num_tokens": 16325613.0,
"step": 7850
},
{
"entropy": 5.773796272277832,
"epoch": 0.7125362844702468,
"grad_norm": 3.828125,
"learning_rate": 0.0004956423760365181,
"loss": 5.567,
"mean_token_accuracy": 0.15991624295711518,
"num_tokens": 16335317.0,
"step": 7855
},
{
"entropy": 5.803212738037109,
"epoch": 0.7129898403483309,
"grad_norm": 2.8125,
"learning_rate": 0.0004956360365748321,
"loss": 5.7073,
"mean_token_accuracy": 0.1481119990348816,
"num_tokens": 16345003.0,
"step": 7860
},
{
"entropy": 5.822187519073486,
"epoch": 0.7134433962264151,
"grad_norm": 3.921875,
"learning_rate": 0.0004956296925503367,
"loss": 5.6876,
"mean_token_accuracy": 0.1523458927869797,
"num_tokens": 16356108.0,
"step": 7865
},
{
"entropy": 5.7450230598449705,
"epoch": 0.7138969521044993,
"grad_norm": 4.8125,
"learning_rate": 0.0004956233439631628,
"loss": 5.5004,
"mean_token_accuracy": 0.16043572574853898,
"num_tokens": 16364733.0,
"step": 7870
},
{
"entropy": 5.791822290420532,
"epoch": 0.7143505079825835,
"grad_norm": 2.140625,
"learning_rate": 0.0004956169908134421,
"loss": 5.6154,
"mean_token_accuracy": 0.1572842299938202,
"num_tokens": 16375229.0,
"step": 7875
},
{
"entropy": 5.792284727096558,
"epoch": 0.7148040638606676,
"grad_norm": 2.765625,
"learning_rate": 0.0004956106331013057,
"loss": 5.7067,
"mean_token_accuracy": 0.15212783664464952,
"num_tokens": 16386366.0,
"step": 7880
},
{
"entropy": 5.711308336257934,
"epoch": 0.7152576197387518,
"grad_norm": 2.640625,
"learning_rate": 0.0004956042708268853,
"loss": 5.5746,
"mean_token_accuracy": 0.16024879962205887,
"num_tokens": 16397067.0,
"step": 7885
},
{
"entropy": 5.715039968490601,
"epoch": 0.715711175616836,
"grad_norm": 4.375,
"learning_rate": 0.0004955979039903123,
"loss": 5.5332,
"mean_token_accuracy": 0.15609787553548812,
"num_tokens": 16407256.0,
"step": 7890
},
{
"entropy": 5.693595457077026,
"epoch": 0.7161647314949202,
"grad_norm": 2.828125,
"learning_rate": 0.0004955915325917183,
"loss": 5.6008,
"mean_token_accuracy": 0.1544367030262947,
"num_tokens": 16418701.0,
"step": 7895
},
{
"entropy": 5.702099895477295,
"epoch": 0.7166182873730044,
"grad_norm": 8.4375,
"learning_rate": 0.0004955851566312354,
"loss": 5.5426,
"mean_token_accuracy": 0.1610220715403557,
"num_tokens": 16428019.0,
"step": 7900
},
{
"entropy": 5.7384286403656,
"epoch": 0.7170718432510885,
"grad_norm": 3.265625,
"learning_rate": 0.0004955787761089952,
"loss": 5.6022,
"mean_token_accuracy": 0.15697715431451797,
"num_tokens": 16440122.0,
"step": 7905
},
{
"entropy": 5.8590857028961185,
"epoch": 0.7175253991291727,
"grad_norm": 2.390625,
"learning_rate": 0.0004955723910251296,
"loss": 5.662,
"mean_token_accuracy": 0.15431325137615204,
"num_tokens": 16449801.0,
"step": 7910
},
{
"entropy": 5.763668775558472,
"epoch": 0.717978955007257,
"grad_norm": 2.703125,
"learning_rate": 0.0004955660013797709,
"loss": 5.5431,
"mean_token_accuracy": 0.15996514409780502,
"num_tokens": 16460155.0,
"step": 7915
},
{
"entropy": 5.810144853591919,
"epoch": 0.7184325108853411,
"grad_norm": 2.578125,
"learning_rate": 0.000495559607173051,
"loss": 5.7181,
"mean_token_accuracy": 0.14558496475219726,
"num_tokens": 16470656.0,
"step": 7920
},
{
"entropy": 5.769259643554688,
"epoch": 0.7188860667634253,
"grad_norm": 2.15625,
"learning_rate": 0.0004955532084051022,
"loss": 5.6338,
"mean_token_accuracy": 0.1549476057291031,
"num_tokens": 16481449.0,
"step": 7925
},
{
"entropy": 5.808447551727295,
"epoch": 0.7193396226415094,
"grad_norm": 2.890625,
"learning_rate": 0.0004955468050760568,
"loss": 5.6547,
"mean_token_accuracy": 0.15426983535289765,
"num_tokens": 16492032.0,
"step": 7930
},
{
"entropy": 5.813133907318115,
"epoch": 0.7197931785195936,
"grad_norm": 2.09375,
"learning_rate": 0.0004955403971860474,
"loss": 5.5981,
"mean_token_accuracy": 0.15324619114398957,
"num_tokens": 16502221.0,
"step": 7935
},
{
"entropy": 5.813241052627563,
"epoch": 0.7202467343976778,
"grad_norm": 2.3125,
"learning_rate": 0.0004955339847352063,
"loss": 5.5784,
"mean_token_accuracy": 0.158725568652153,
"num_tokens": 16512021.0,
"step": 7940
},
{
"entropy": 5.748729562759399,
"epoch": 0.720700290275762,
"grad_norm": 3.25,
"learning_rate": 0.0004955275677236662,
"loss": 5.6247,
"mean_token_accuracy": 0.15192925482988356,
"num_tokens": 16522060.0,
"step": 7945
},
{
"entropy": 5.764633464813232,
"epoch": 0.7211538461538461,
"grad_norm": 2.734375,
"learning_rate": 0.0004955211461515599,
"loss": 5.6256,
"mean_token_accuracy": 0.16055750697851182,
"num_tokens": 16532827.0,
"step": 7950
},
{
"entropy": 5.77035083770752,
"epoch": 0.7216074020319303,
"grad_norm": 2.5,
"learning_rate": 0.00049551472001902,
"loss": 5.6584,
"mean_token_accuracy": 0.16088539361953735,
"num_tokens": 16542838.0,
"step": 7955
},
{
"entropy": 5.747999858856201,
"epoch": 0.7220609579100146,
"grad_norm": 2.5625,
"learning_rate": 0.0004955082893261796,
"loss": 5.5968,
"mean_token_accuracy": 0.16128583997488022,
"num_tokens": 16552615.0,
"step": 7960
},
{
"entropy": 5.824002838134765,
"epoch": 0.7225145137880987,
"grad_norm": 3.15625,
"learning_rate": 0.0004955018540731714,
"loss": 5.6658,
"mean_token_accuracy": 0.1531428575515747,
"num_tokens": 16561777.0,
"step": 7965
},
{
"entropy": 5.844862222671509,
"epoch": 0.7229680696661829,
"grad_norm": 2.390625,
"learning_rate": 0.0004954954142601288,
"loss": 5.6485,
"mean_token_accuracy": 0.15446362793445587,
"num_tokens": 16572111.0,
"step": 7970
},
{
"entropy": 5.804814147949219,
"epoch": 0.723421625544267,
"grad_norm": 1.890625,
"learning_rate": 0.0004954889698871849,
"loss": 5.6817,
"mean_token_accuracy": 0.15889163613319396,
"num_tokens": 16581674.0,
"step": 7975
},
{
"entropy": 5.81389856338501,
"epoch": 0.7238751814223512,
"grad_norm": 3.078125,
"learning_rate": 0.0004954825209544728,
"loss": 5.6737,
"mean_token_accuracy": 0.16090579181909562,
"num_tokens": 16591802.0,
"step": 7980
},
{
"entropy": 5.730545282363892,
"epoch": 0.7243287373004355,
"grad_norm": 3.515625,
"learning_rate": 0.000495476067462126,
"loss": 5.5695,
"mean_token_accuracy": 0.16237852573394776,
"num_tokens": 16602772.0,
"step": 7985
},
{
"entropy": 5.777552795410156,
"epoch": 0.7247822931785196,
"grad_norm": 3.796875,
"learning_rate": 0.000495469609410278,
"loss": 5.6104,
"mean_token_accuracy": 0.1624607339501381,
"num_tokens": 16613526.0,
"step": 7990
},
{
"entropy": 5.762968063354492,
"epoch": 0.7252358490566038,
"grad_norm": 3.484375,
"learning_rate": 0.0004954631467990622,
"loss": 5.6337,
"mean_token_accuracy": 0.15944494158029557,
"num_tokens": 16623583.0,
"step": 7995
},
{
"entropy": 5.799592447280884,
"epoch": 0.7256894049346879,
"grad_norm": 2.5625,
"learning_rate": 0.0004954566796286124,
"loss": 5.6835,
"mean_token_accuracy": 0.15078045651316643,
"num_tokens": 16635068.0,
"step": 8000
},
{
"entropy": 5.77793869972229,
"epoch": 0.7261429608127722,
"grad_norm": 2.375,
"learning_rate": 0.0004954502078990623,
"loss": 5.597,
"mean_token_accuracy": 0.1602660283446312,
"num_tokens": 16645021.0,
"step": 8005
},
{
"entropy": 5.779656219482422,
"epoch": 0.7265965166908563,
"grad_norm": 3.046875,
"learning_rate": 0.0004954437316105457,
"loss": 5.6564,
"mean_token_accuracy": 0.15226438343524934,
"num_tokens": 16656021.0,
"step": 8010
},
{
"entropy": 5.736810302734375,
"epoch": 0.7270500725689405,
"grad_norm": 2.296875,
"learning_rate": 0.0004954372507631965,
"loss": 5.5506,
"mean_token_accuracy": 0.1654258355498314,
"num_tokens": 16665236.0,
"step": 8015
},
{
"entropy": 5.738938236236573,
"epoch": 0.7275036284470247,
"grad_norm": 3.15625,
"learning_rate": 0.0004954307653571489,
"loss": 5.7423,
"mean_token_accuracy": 0.1501321703195572,
"num_tokens": 16676690.0,
"step": 8020
},
{
"entropy": 5.804196786880493,
"epoch": 0.7279571843251088,
"grad_norm": 2.125,
"learning_rate": 0.0004954242753925367,
"loss": 5.6363,
"mean_token_accuracy": 0.15009679719805719,
"num_tokens": 16687288.0,
"step": 8025
},
{
"entropy": 5.809215927124024,
"epoch": 0.7284107402031931,
"grad_norm": 3.28125,
"learning_rate": 0.0004954177808694946,
"loss": 5.6449,
"mean_token_accuracy": 0.15161658823490143,
"num_tokens": 16697873.0,
"step": 8030
},
{
"entropy": 5.766271352767944,
"epoch": 0.7288642960812772,
"grad_norm": 2.703125,
"learning_rate": 0.0004954112817881565,
"loss": 5.6496,
"mean_token_accuracy": 0.15360791981220245,
"num_tokens": 16709002.0,
"step": 8035
},
{
"entropy": 5.761889362335205,
"epoch": 0.7293178519593614,
"grad_norm": 2.453125,
"learning_rate": 0.0004954047781486568,
"loss": 5.6205,
"mean_token_accuracy": 0.16080274432897568,
"num_tokens": 16718989.0,
"step": 8040
},
{
"entropy": 5.9060508728027346,
"epoch": 0.7297714078374455,
"grad_norm": 1.859375,
"learning_rate": 0.0004953982699511303,
"loss": 5.7199,
"mean_token_accuracy": 0.15385592058300973,
"num_tokens": 16730425.0,
"step": 8045
},
{
"entropy": 5.826246070861816,
"epoch": 0.7302249637155298,
"grad_norm": 2.59375,
"learning_rate": 0.0004953917571957114,
"loss": 5.6016,
"mean_token_accuracy": 0.15725287199020385,
"num_tokens": 16740216.0,
"step": 8050
},
{
"entropy": 5.725392770767212,
"epoch": 0.730678519593614,
"grad_norm": 2.46875,
"learning_rate": 0.0004953852398825346,
"loss": 5.6029,
"mean_token_accuracy": 0.15493546724319457,
"num_tokens": 16752436.0,
"step": 8055
},
{
"entropy": 5.739631319046021,
"epoch": 0.7311320754716981,
"grad_norm": 2.15625,
"learning_rate": 0.000495378718011735,
"loss": 5.6102,
"mean_token_accuracy": 0.15602830350399016,
"num_tokens": 16762631.0,
"step": 8060
},
{
"entropy": 5.8749692916870115,
"epoch": 0.7315856313497823,
"grad_norm": 1.859375,
"learning_rate": 0.0004953721915834473,
"loss": 5.6495,
"mean_token_accuracy": 0.1531766250729561,
"num_tokens": 16772613.0,
"step": 8065
},
{
"entropy": 5.821211862564087,
"epoch": 0.7320391872278664,
"grad_norm": 3.03125,
"learning_rate": 0.0004953656605978067,
"loss": 5.6449,
"mean_token_accuracy": 0.1528734877705574,
"num_tokens": 16782963.0,
"step": 8070
},
{
"entropy": 5.712953901290893,
"epoch": 0.7324927431059507,
"grad_norm": 2.203125,
"learning_rate": 0.000495359125054948,
"loss": 5.5783,
"mean_token_accuracy": 0.15503397285938264,
"num_tokens": 16793892.0,
"step": 8075
},
{
"entropy": 5.654757499694824,
"epoch": 0.7329462989840348,
"grad_norm": 1.7578125,
"learning_rate": 0.0004953525849550063,
"loss": 5.5538,
"mean_token_accuracy": 0.15403005629777908,
"num_tokens": 16803388.0,
"step": 8080
},
{
"entropy": 5.736060619354248,
"epoch": 0.733399854862119,
"grad_norm": 2.4375,
"learning_rate": 0.0004953460402981172,
"loss": 5.5299,
"mean_token_accuracy": 0.1605105310678482,
"num_tokens": 16813201.0,
"step": 8085
},
{
"entropy": 5.83330774307251,
"epoch": 0.7338534107402032,
"grad_norm": 1.9609375,
"learning_rate": 0.0004953394910844157,
"loss": 5.7471,
"mean_token_accuracy": 0.15453502386808396,
"num_tokens": 16823265.0,
"step": 8090
},
{
"entropy": 5.816150331497193,
"epoch": 0.7343069666182874,
"grad_norm": 1.9140625,
"learning_rate": 0.0004953329373140374,
"loss": 5.5551,
"mean_token_accuracy": 0.15653111785650253,
"num_tokens": 16833057.0,
"step": 8095
},
{
"entropy": 5.7323627948760985,
"epoch": 0.7347605224963716,
"grad_norm": 2.171875,
"learning_rate": 0.0004953263789871179,
"loss": 5.5185,
"mean_token_accuracy": 0.1549549698829651,
"num_tokens": 16843488.0,
"step": 8100
},
{
"entropy": 5.68474645614624,
"epoch": 0.7352140783744557,
"grad_norm": 2.484375,
"learning_rate": 0.0004953198161037925,
"loss": 5.5595,
"mean_token_accuracy": 0.16272432655096053,
"num_tokens": 16852448.0,
"step": 8105
},
{
"entropy": 5.706146478652954,
"epoch": 0.7356676342525399,
"grad_norm": 2.765625,
"learning_rate": 0.0004953132486641975,
"loss": 5.6299,
"mean_token_accuracy": 0.15620479583740235,
"num_tokens": 16863944.0,
"step": 8110
},
{
"entropy": 5.779846954345703,
"epoch": 0.736121190130624,
"grad_norm": 1.8046875,
"learning_rate": 0.000495306676668468,
"loss": 5.5523,
"mean_token_accuracy": 0.1597314640879631,
"num_tokens": 16873923.0,
"step": 8115
},
{
"entropy": 5.7534888744354244,
"epoch": 0.7365747460087083,
"grad_norm": 2.015625,
"learning_rate": 0.0004953001001167406,
"loss": 5.6338,
"mean_token_accuracy": 0.15616717636585237,
"num_tokens": 16884373.0,
"step": 8120
},
{
"entropy": 5.797355031967163,
"epoch": 0.7370283018867925,
"grad_norm": 1.75,
"learning_rate": 0.0004952935190091508,
"loss": 5.751,
"mean_token_accuracy": 0.14971877932548522,
"num_tokens": 16895869.0,
"step": 8125
},
{
"entropy": 5.675172948837281,
"epoch": 0.7374818577648766,
"grad_norm": 2.5,
"learning_rate": 0.0004952869333458349,
"loss": 5.6137,
"mean_token_accuracy": 0.15598379522562028,
"num_tokens": 16906737.0,
"step": 8130
},
{
"entropy": 5.729729843139649,
"epoch": 0.7379354136429608,
"grad_norm": 2.40625,
"learning_rate": 0.0004952803431269292,
"loss": 5.5513,
"mean_token_accuracy": 0.16393394023180008,
"num_tokens": 16916083.0,
"step": 8135
},
{
"entropy": 5.9084995746612545,
"epoch": 0.738388969521045,
"grad_norm": 2.203125,
"learning_rate": 0.0004952737483525699,
"loss": 5.6972,
"mean_token_accuracy": 0.1537786453962326,
"num_tokens": 16926419.0,
"step": 8140
},
{
"entropy": 5.845476150512695,
"epoch": 0.7388425253991292,
"grad_norm": 3.9375,
"learning_rate": 0.0004952671490228932,
"loss": 5.6348,
"mean_token_accuracy": 0.15125631541013718,
"num_tokens": 16936067.0,
"step": 8145
},
{
"entropy": 5.785321569442749,
"epoch": 0.7392960812772134,
"grad_norm": 2.765625,
"learning_rate": 0.0004952605451380357,
"loss": 5.5902,
"mean_token_accuracy": 0.15724589079618453,
"num_tokens": 16946571.0,
"step": 8150
},
{
"entropy": 5.69130859375,
"epoch": 0.7397496371552975,
"grad_norm": 2.828125,
"learning_rate": 0.0004952539366981342,
"loss": 5.6351,
"mean_token_accuracy": 0.15460131987929343,
"num_tokens": 16956764.0,
"step": 8155
},
{
"entropy": 5.730801725387574,
"epoch": 0.7402031930333817,
"grad_norm": 2.390625,
"learning_rate": 0.000495247323703325,
"loss": 5.5779,
"mean_token_accuracy": 0.16185448169708253,
"num_tokens": 16966002.0,
"step": 8160
},
{
"entropy": 5.810474014282226,
"epoch": 0.7406567489114659,
"grad_norm": 1.5625,
"learning_rate": 0.000495240706153745,
"loss": 5.672,
"mean_token_accuracy": 0.16184382885694504,
"num_tokens": 16975705.0,
"step": 8165
},
{
"entropy": 5.792193794250489,
"epoch": 0.7411103047895501,
"grad_norm": 1.7890625,
"learning_rate": 0.0004952340840495312,
"loss": 5.5654,
"mean_token_accuracy": 0.1592943087220192,
"num_tokens": 16985945.0,
"step": 8170
},
{
"entropy": 5.7322290420532225,
"epoch": 0.7415638606676342,
"grad_norm": 3.21875,
"learning_rate": 0.0004952274573908203,
"loss": 5.5333,
"mean_token_accuracy": 0.15994324982166291,
"num_tokens": 16995475.0,
"step": 8175
},
{
"entropy": 5.720385551452637,
"epoch": 0.7420174165457184,
"grad_norm": 1.8125,
"learning_rate": 0.0004952208261777496,
"loss": 5.5863,
"mean_token_accuracy": 0.1577586606144905,
"num_tokens": 17005028.0,
"step": 8180
},
{
"entropy": 5.70277590751648,
"epoch": 0.7424709724238027,
"grad_norm": 1.65625,
"learning_rate": 0.000495214190410456,
"loss": 5.6413,
"mean_token_accuracy": 0.16015808135271073,
"num_tokens": 17014786.0,
"step": 8185
},
{
"entropy": 5.782215023040772,
"epoch": 0.7429245283018868,
"grad_norm": 1.7734375,
"learning_rate": 0.0004952075500890768,
"loss": 5.6375,
"mean_token_accuracy": 0.15373314470052718,
"num_tokens": 17024431.0,
"step": 8190
},
{
"entropy": 5.7832722663879395,
"epoch": 0.743378084179971,
"grad_norm": 2.5625,
"learning_rate": 0.0004952009052137493,
"loss": 5.6131,
"mean_token_accuracy": 0.15820928514003754,
"num_tokens": 17034368.0,
"step": 8195
},
{
"entropy": 5.704320812225342,
"epoch": 0.7438316400580551,
"grad_norm": 2.0625,
"learning_rate": 0.0004951942557846112,
"loss": 5.6477,
"mean_token_accuracy": 0.15668941885232926,
"num_tokens": 17043818.0,
"step": 8200
},
{
"entropy": 5.762405443191528,
"epoch": 0.7442851959361393,
"grad_norm": 2.359375,
"learning_rate": 0.0004951876018017997,
"loss": 5.6449,
"mean_token_accuracy": 0.16201995462179183,
"num_tokens": 17054843.0,
"step": 8205
},
{
"entropy": 5.81440749168396,
"epoch": 0.7447387518142236,
"grad_norm": 2.296875,
"learning_rate": 0.0004951809432654524,
"loss": 5.6456,
"mean_token_accuracy": 0.15417395681142806,
"num_tokens": 17065539.0,
"step": 8210
},
{
"entropy": 5.7607855796813965,
"epoch": 0.7451923076923077,
"grad_norm": 2.03125,
"learning_rate": 0.000495174280175707,
"loss": 5.5399,
"mean_token_accuracy": 0.16154730021953584,
"num_tokens": 17075443.0,
"step": 8215
},
{
"entropy": 5.8358794212341305,
"epoch": 0.7456458635703919,
"grad_norm": 3.15625,
"learning_rate": 0.0004951676125327015,
"loss": 5.6909,
"mean_token_accuracy": 0.15732420980930328,
"num_tokens": 17085490.0,
"step": 8220
},
{
"entropy": 5.773124504089355,
"epoch": 0.746099419448476,
"grad_norm": 4.21875,
"learning_rate": 0.0004951609403365737,
"loss": 5.6691,
"mean_token_accuracy": 0.16350300014019012,
"num_tokens": 17094803.0,
"step": 8225
},
{
"entropy": 5.741847515106201,
"epoch": 0.7465529753265603,
"grad_norm": 2.234375,
"learning_rate": 0.0004951542635874615,
"loss": 5.5236,
"mean_token_accuracy": 0.15575825721025466,
"num_tokens": 17104373.0,
"step": 8230
},
{
"entropy": 5.7692542552948,
"epoch": 0.7470065312046444,
"grad_norm": 4.78125,
"learning_rate": 0.0004951475822855029,
"loss": 5.6208,
"mean_token_accuracy": 0.1559954658150673,
"num_tokens": 17115469.0,
"step": 8235
},
{
"entropy": 5.7214123725891115,
"epoch": 0.7474600870827286,
"grad_norm": 2.640625,
"learning_rate": 0.0004951408964308364,
"loss": 5.5945,
"mean_token_accuracy": 0.15529240369796754,
"num_tokens": 17125443.0,
"step": 8240
},
{
"entropy": 5.701405763626099,
"epoch": 0.7479136429608128,
"grad_norm": 1.9453125,
"learning_rate": 0.0004951342060236,
"loss": 5.5529,
"mean_token_accuracy": 0.15890992432832718,
"num_tokens": 17136253.0,
"step": 8245
},
{
"entropy": 5.851835012435913,
"epoch": 0.7483671988388969,
"grad_norm": 2.390625,
"learning_rate": 0.0004951275110639321,
"loss": 5.6167,
"mean_token_accuracy": 0.160022309422493,
"num_tokens": 17145544.0,
"step": 8250
},
{
"entropy": 5.778053617477417,
"epoch": 0.7488207547169812,
"grad_norm": 4.28125,
"learning_rate": 0.0004951208115519712,
"loss": 5.6155,
"mean_token_accuracy": 0.156676721572876,
"num_tokens": 17155043.0,
"step": 8255
},
{
"entropy": 5.664043331146241,
"epoch": 0.7492743105950653,
"grad_norm": 2.71875,
"learning_rate": 0.0004951141074878558,
"loss": 5.6681,
"mean_token_accuracy": 0.15661310255527497,
"num_tokens": 17164885.0,
"step": 8260
},
{
"entropy": 5.803770303726196,
"epoch": 0.7497278664731495,
"grad_norm": 2.734375,
"learning_rate": 0.0004951073988717246,
"loss": 5.6546,
"mean_token_accuracy": 0.15547531992197036,
"num_tokens": 17175681.0,
"step": 8265
},
{
"entropy": 5.721224546432495,
"epoch": 0.7501814223512336,
"grad_norm": 2.328125,
"learning_rate": 0.0004951006857037163,
"loss": 5.5184,
"mean_token_accuracy": 0.160471174120903,
"num_tokens": 17186002.0,
"step": 8270
},
{
"entropy": 5.759907197952271,
"epoch": 0.7506349782293179,
"grad_norm": 2.390625,
"learning_rate": 0.0004950939679839696,
"loss": 5.5906,
"mean_token_accuracy": 0.15814287662506105,
"num_tokens": 17195499.0,
"step": 8275
},
{
"entropy": 5.763993167877198,
"epoch": 0.7510885341074021,
"grad_norm": 2.265625,
"learning_rate": 0.0004950872457126237,
"loss": 5.5893,
"mean_token_accuracy": 0.1606327325105667,
"num_tokens": 17204684.0,
"step": 8280
},
{
"entropy": 5.736480617523194,
"epoch": 0.7515420899854862,
"grad_norm": 2.6875,
"learning_rate": 0.0004950805188898175,
"loss": 5.6116,
"mean_token_accuracy": 0.15062818825244903,
"num_tokens": 17213987.0,
"step": 8285
},
{
"entropy": 5.706430864334107,
"epoch": 0.7519956458635704,
"grad_norm": 2.09375,
"learning_rate": 0.00049507378751569,
"loss": 5.6653,
"mean_token_accuracy": 0.15044132471084595,
"num_tokens": 17224512.0,
"step": 8290
},
{
"entropy": 5.813805341720581,
"epoch": 0.7524492017416545,
"grad_norm": 2.421875,
"learning_rate": 0.0004950670515903805,
"loss": 5.6677,
"mean_token_accuracy": 0.156072798371315,
"num_tokens": 17234362.0,
"step": 8295
},
{
"entropy": 5.849670457839966,
"epoch": 0.7529027576197388,
"grad_norm": 2.5,
"learning_rate": 0.0004950603111140283,
"loss": 5.6485,
"mean_token_accuracy": 0.16128491908311843,
"num_tokens": 17244241.0,
"step": 8300
},
{
"entropy": 5.746774530410766,
"epoch": 0.753356313497823,
"grad_norm": 2.28125,
"learning_rate": 0.0004950535660867728,
"loss": 5.5874,
"mean_token_accuracy": 0.1597753271460533,
"num_tokens": 17254689.0,
"step": 8305
},
{
"entropy": 5.774064588546753,
"epoch": 0.7538098693759071,
"grad_norm": 2.5,
"learning_rate": 0.0004950468165087535,
"loss": 5.6731,
"mean_token_accuracy": 0.15421361476182938,
"num_tokens": 17265506.0,
"step": 8310
},
{
"entropy": 5.829351091384888,
"epoch": 0.7542634252539913,
"grad_norm": 2.53125,
"learning_rate": 0.0004950400623801098,
"loss": 5.7373,
"mean_token_accuracy": 0.15709088742733002,
"num_tokens": 17276376.0,
"step": 8315
},
{
"entropy": 5.817118406295776,
"epoch": 0.7547169811320755,
"grad_norm": 3.09375,
"learning_rate": 0.0004950333037009818,
"loss": 5.6416,
"mean_token_accuracy": 0.14860429465770722,
"num_tokens": 17287005.0,
"step": 8320
},
{
"entropy": 5.800573635101318,
"epoch": 0.7551705370101597,
"grad_norm": 2.953125,
"learning_rate": 0.0004950265404715089,
"loss": 5.6514,
"mean_token_accuracy": 0.15255427807569505,
"num_tokens": 17297595.0,
"step": 8325
},
{
"entropy": 5.712475156784057,
"epoch": 0.7556240928882438,
"grad_norm": 2.375,
"learning_rate": 0.000495019772691831,
"loss": 5.5188,
"mean_token_accuracy": 0.16102788150310515,
"num_tokens": 17307319.0,
"step": 8330
},
{
"entropy": 5.753348875045776,
"epoch": 0.756077648766328,
"grad_norm": 4.15625,
"learning_rate": 0.0004950130003620882,
"loss": 5.7276,
"mean_token_accuracy": 0.15592837631702422,
"num_tokens": 17317978.0,
"step": 8335
},
{
"entropy": 5.79032154083252,
"epoch": 0.7565312046444121,
"grad_norm": 4.21875,
"learning_rate": 0.0004950062234824204,
"loss": 5.619,
"mean_token_accuracy": 0.1543349266052246,
"num_tokens": 17328339.0,
"step": 8340
},
{
"entropy": 5.8189697265625,
"epoch": 0.7569847605224964,
"grad_norm": 3.421875,
"learning_rate": 0.0004949994420529679,
"loss": 5.6732,
"mean_token_accuracy": 0.15629276037216186,
"num_tokens": 17338668.0,
"step": 8345
},
{
"entropy": 5.722092628479004,
"epoch": 0.7574383164005806,
"grad_norm": 3.296875,
"learning_rate": 0.0004949926560738709,
"loss": 5.61,
"mean_token_accuracy": 0.1548818901181221,
"num_tokens": 17348609.0,
"step": 8350
},
{
"entropy": 5.778551483154297,
"epoch": 0.7578918722786647,
"grad_norm": 2.296875,
"learning_rate": 0.0004949858655452697,
"loss": 5.6182,
"mean_token_accuracy": 0.1580677330493927,
"num_tokens": 17358014.0,
"step": 8355
},
{
"entropy": 5.8439408302307125,
"epoch": 0.7583454281567489,
"grad_norm": 2.34375,
"learning_rate": 0.0004949790704673048,
"loss": 5.6406,
"mean_token_accuracy": 0.15462183952331543,
"num_tokens": 17367975.0,
"step": 8360
},
{
"entropy": 5.79978575706482,
"epoch": 0.7587989840348331,
"grad_norm": 2.4375,
"learning_rate": 0.0004949722708401166,
"loss": 5.5922,
"mean_token_accuracy": 0.16122720241546631,
"num_tokens": 17377960.0,
"step": 8365
},
{
"entropy": 5.787926769256591,
"epoch": 0.7592525399129173,
"grad_norm": 2.21875,
"learning_rate": 0.0004949654666638457,
"loss": 5.6539,
"mean_token_accuracy": 0.1638580560684204,
"num_tokens": 17387334.0,
"step": 8370
},
{
"entropy": 5.751470899581909,
"epoch": 0.7597060957910015,
"grad_norm": 2.734375,
"learning_rate": 0.000494958657938633,
"loss": 5.6273,
"mean_token_accuracy": 0.15456898808479308,
"num_tokens": 17396521.0,
"step": 8375
},
{
"entropy": 5.730402326583862,
"epoch": 0.7601596516690856,
"grad_norm": 2.46875,
"learning_rate": 0.0004949518446646192,
"loss": 5.6011,
"mean_token_accuracy": 0.15843016803264617,
"num_tokens": 17407217.0,
"step": 8380
},
{
"entropy": 5.833608818054199,
"epoch": 0.7606132075471698,
"grad_norm": 3.46875,
"learning_rate": 0.0004949450268419451,
"loss": 5.6841,
"mean_token_accuracy": 0.1537593498826027,
"num_tokens": 17416982.0,
"step": 8385
},
{
"entropy": 5.7752313137054445,
"epoch": 0.761066763425254,
"grad_norm": 3.15625,
"learning_rate": 0.000494938204470752,
"loss": 5.5581,
"mean_token_accuracy": 0.15870605111122132,
"num_tokens": 17427636.0,
"step": 8390
},
{
"entropy": 5.752287817001343,
"epoch": 0.7615203193033382,
"grad_norm": 1.9921875,
"learning_rate": 0.0004949313775511806,
"loss": 5.6584,
"mean_token_accuracy": 0.15613557100296022,
"num_tokens": 17437488.0,
"step": 8395
},
{
"entropy": 5.766221952438355,
"epoch": 0.7619738751814223,
"grad_norm": 3.109375,
"learning_rate": 0.0004949245460833723,
"loss": 5.4833,
"mean_token_accuracy": 0.15987526923418044,
"num_tokens": 17447838.0,
"step": 8400
},
{
"entropy": 5.833551645278931,
"epoch": 0.7624274310595065,
"grad_norm": 3.765625,
"learning_rate": 0.0004949177100674684,
"loss": 5.6733,
"mean_token_accuracy": 0.16033527255058289,
"num_tokens": 17457257.0,
"step": 8405
},
{
"entropy": 5.769413089752197,
"epoch": 0.7628809869375908,
"grad_norm": 2.09375,
"learning_rate": 0.0004949108695036102,
"loss": 5.5967,
"mean_token_accuracy": 0.1562741219997406,
"num_tokens": 17467729.0,
"step": 8410
},
{
"entropy": 5.738751697540283,
"epoch": 0.7633345428156749,
"grad_norm": 2.890625,
"learning_rate": 0.0004949040243919392,
"loss": 5.6211,
"mean_token_accuracy": 0.14854976236820222,
"num_tokens": 17477692.0,
"step": 8415
},
{
"entropy": 5.765266466140747,
"epoch": 0.7637880986937591,
"grad_norm": 2.984375,
"learning_rate": 0.0004948971747325968,
"loss": 5.6265,
"mean_token_accuracy": 0.15609366446733475,
"num_tokens": 17488730.0,
"step": 8420
},
{
"entropy": 5.79982271194458,
"epoch": 0.7642416545718432,
"grad_norm": 2.546875,
"learning_rate": 0.000494890320525725,
"loss": 5.6998,
"mean_token_accuracy": 0.15643664747476577,
"num_tokens": 17498374.0,
"step": 8425
},
{
"entropy": 5.818951892852783,
"epoch": 0.7646952104499274,
"grad_norm": 2.5,
"learning_rate": 0.0004948834617714652,
"loss": 5.5518,
"mean_token_accuracy": 0.16498626470565797,
"num_tokens": 17508601.0,
"step": 8430
},
{
"entropy": 5.724190378189087,
"epoch": 0.7651487663280117,
"grad_norm": 2.375,
"learning_rate": 0.0004948765984699595,
"loss": 5.5833,
"mean_token_accuracy": 0.1610970064997673,
"num_tokens": 17518164.0,
"step": 8435
},
{
"entropy": 5.705046081542969,
"epoch": 0.7656023222060958,
"grad_norm": 2.265625,
"learning_rate": 0.0004948697306213494,
"loss": 5.5474,
"mean_token_accuracy": 0.16527391970157623,
"num_tokens": 17527244.0,
"step": 8440
},
{
"entropy": 5.8157226085662845,
"epoch": 0.76605587808418,
"grad_norm": 2.4375,
"learning_rate": 0.0004948628582257775,
"loss": 5.6383,
"mean_token_accuracy": 0.15614416748285292,
"num_tokens": 17538045.0,
"step": 8445
},
{
"entropy": 5.772269582748413,
"epoch": 0.7665094339622641,
"grad_norm": 2.09375,
"learning_rate": 0.0004948559812833857,
"loss": 5.597,
"mean_token_accuracy": 0.15702936947345733,
"num_tokens": 17547584.0,
"step": 8450
},
{
"entropy": 5.776641750335694,
"epoch": 0.7669629898403484,
"grad_norm": 3.5625,
"learning_rate": 0.0004948490997943162,
"loss": 5.6249,
"mean_token_accuracy": 0.1593453913927078,
"num_tokens": 17558925.0,
"step": 8455
},
{
"entropy": 5.762310457229614,
"epoch": 0.7674165457184325,
"grad_norm": 5.9375,
"learning_rate": 0.0004948422137587111,
"loss": 5.724,
"mean_token_accuracy": 0.1505131408572197,
"num_tokens": 17569355.0,
"step": 8460
},
{
"entropy": 5.784338808059692,
"epoch": 0.7678701015965167,
"grad_norm": 2.8125,
"learning_rate": 0.0004948353231767131,
"loss": 5.5881,
"mean_token_accuracy": 0.15870064198970796,
"num_tokens": 17579782.0,
"step": 8465
},
{
"entropy": 5.826953411102295,
"epoch": 0.7683236574746009,
"grad_norm": 2.828125,
"learning_rate": 0.0004948284280484645,
"loss": 5.6428,
"mean_token_accuracy": 0.16042931079864503,
"num_tokens": 17589772.0,
"step": 8470
},
{
"entropy": 5.7036386013031,
"epoch": 0.768777213352685,
"grad_norm": 2.796875,
"learning_rate": 0.0004948215283741081,
"loss": 5.5711,
"mean_token_accuracy": 0.1547003537416458,
"num_tokens": 17599613.0,
"step": 8475
},
{
"entropy": 5.738386487960815,
"epoch": 0.7692307692307693,
"grad_norm": 2.40625,
"learning_rate": 0.0004948146241537866,
"loss": 5.6615,
"mean_token_accuracy": 0.1533209577202797,
"num_tokens": 17609056.0,
"step": 8480
},
{
"entropy": 5.764581918716431,
"epoch": 0.7696843251088534,
"grad_norm": 2.4375,
"learning_rate": 0.0004948077153876425,
"loss": 5.5759,
"mean_token_accuracy": 0.1585683375597,
"num_tokens": 17618620.0,
"step": 8485
},
{
"entropy": 5.7266357898712155,
"epoch": 0.7701378809869376,
"grad_norm": 1.953125,
"learning_rate": 0.0004948008020758188,
"loss": 5.5342,
"mean_token_accuracy": 0.16150321513414384,
"num_tokens": 17627630.0,
"step": 8490
},
{
"entropy": 5.6926289081573485,
"epoch": 0.7705914368650217,
"grad_norm": 2.75,
"learning_rate": 0.0004947938842184585,
"loss": 5.5501,
"mean_token_accuracy": 0.1610981583595276,
"num_tokens": 17638903.0,
"step": 8495
},
{
"entropy": 5.751527643203735,
"epoch": 0.771044992743106,
"grad_norm": 2.03125,
"learning_rate": 0.0004947869618157048,
"loss": 5.5903,
"mean_token_accuracy": 0.15254145711660386,
"num_tokens": 17649257.0,
"step": 8500
},
{
"entropy": 5.730644702911377,
"epoch": 0.7714985486211902,
"grad_norm": 3.59375,
"learning_rate": 0.0004947800348677006,
"loss": 5.5627,
"mean_token_accuracy": 0.1666986346244812,
"num_tokens": 17659495.0,
"step": 8505
},
{
"entropy": 5.650215673446655,
"epoch": 0.7719521044992743,
"grad_norm": 2.296875,
"learning_rate": 0.0004947731033745893,
"loss": 5.554,
"mean_token_accuracy": 0.16249235570430756,
"num_tokens": 17669671.0,
"step": 8510
},
{
"entropy": 5.83202748298645,
"epoch": 0.7724056603773585,
"grad_norm": 2.765625,
"learning_rate": 0.0004947661673365142,
"loss": 5.721,
"mean_token_accuracy": 0.15400478094816208,
"num_tokens": 17680927.0,
"step": 8515
},
{
"entropy": 5.78148136138916,
"epoch": 0.7728592162554426,
"grad_norm": 2.90625,
"learning_rate": 0.0004947592267536188,
"loss": 5.595,
"mean_token_accuracy": 0.1603729546070099,
"num_tokens": 17691378.0,
"step": 8520
},
{
"entropy": 5.750188398361206,
"epoch": 0.7733127721335269,
"grad_norm": 8.5,
"learning_rate": 0.0004947522816260466,
"loss": 5.6546,
"mean_token_accuracy": 0.15359988436102867,
"num_tokens": 17700942.0,
"step": 8525
},
{
"entropy": 5.755286645889282,
"epoch": 0.773766328011611,
"grad_norm": 2.546875,
"learning_rate": 0.0004947453319539411,
"loss": 5.5675,
"mean_token_accuracy": 0.16288118213415145,
"num_tokens": 17710076.0,
"step": 8530
},
{
"entropy": 5.735703372955323,
"epoch": 0.7742198838896952,
"grad_norm": 4.3125,
"learning_rate": 0.0004947383777374462,
"loss": 5.5998,
"mean_token_accuracy": 0.1575216144323349,
"num_tokens": 17721009.0,
"step": 8535
},
{
"entropy": 5.730885696411133,
"epoch": 0.7746734397677794,
"grad_norm": 3.5,
"learning_rate": 0.0004947314189767058,
"loss": 5.6231,
"mean_token_accuracy": 0.16047685146331786,
"num_tokens": 17731821.0,
"step": 8540
},
{
"entropy": 5.712780427932739,
"epoch": 0.7751269956458636,
"grad_norm": 4.28125,
"learning_rate": 0.0004947244556718635,
"loss": 5.5151,
"mean_token_accuracy": 0.1676603823900223,
"num_tokens": 17741425.0,
"step": 8545
},
{
"entropy": 5.759090185165405,
"epoch": 0.7755805515239478,
"grad_norm": 2.21875,
"learning_rate": 0.0004947174878230635,
"loss": 5.5761,
"mean_token_accuracy": 0.15907540917396545,
"num_tokens": 17752381.0,
"step": 8550
},
{
"entropy": 5.757405757904053,
"epoch": 0.7760341074020319,
"grad_norm": 2.40625,
"learning_rate": 0.0004947105154304498,
"loss": 5.5515,
"mean_token_accuracy": 0.15621833950281144,
"num_tokens": 17762768.0,
"step": 8555
},
{
"entropy": 5.747535753250122,
"epoch": 0.7764876632801161,
"grad_norm": 2.28125,
"learning_rate": 0.0004947035384941667,
"loss": 5.5911,
"mean_token_accuracy": 0.15688183903694153,
"num_tokens": 17774032.0,
"step": 8560
},
{
"entropy": 5.787305736541748,
"epoch": 0.7769412191582002,
"grad_norm": 3.046875,
"learning_rate": 0.0004946965570143584,
"loss": 5.7105,
"mean_token_accuracy": 0.14983301609754562,
"num_tokens": 17785098.0,
"step": 8565
},
{
"entropy": 5.8069006443023685,
"epoch": 0.7773947750362845,
"grad_norm": 2.28125,
"learning_rate": 0.0004946895709911695,
"loss": 5.6007,
"mean_token_accuracy": 0.15720168501138687,
"num_tokens": 17795487.0,
"step": 8570
},
{
"entropy": 5.76495361328125,
"epoch": 0.7778483309143687,
"grad_norm": 1.9296875,
"learning_rate": 0.0004946825804247442,
"loss": 5.7118,
"mean_token_accuracy": 0.14942110031843187,
"num_tokens": 17806045.0,
"step": 8575
},
{
"entropy": 5.878367519378662,
"epoch": 0.7783018867924528,
"grad_norm": 2.703125,
"learning_rate": 0.000494675585315227,
"loss": 5.6808,
"mean_token_accuracy": 0.15481762886047362,
"num_tokens": 17816948.0,
"step": 8580
},
{
"entropy": 5.813451194763184,
"epoch": 0.778755442670537,
"grad_norm": 2.796875,
"learning_rate": 0.0004946685856627628,
"loss": 5.5921,
"mean_token_accuracy": 0.16048079282045363,
"num_tokens": 17826929.0,
"step": 8585
},
{
"entropy": 5.775233936309815,
"epoch": 0.7792089985486212,
"grad_norm": 1.984375,
"learning_rate": 0.0004946615814674963,
"loss": 5.711,
"mean_token_accuracy": 0.15520803034305572,
"num_tokens": 17837780.0,
"step": 8590
},
{
"entropy": 5.691840076446534,
"epoch": 0.7796625544267054,
"grad_norm": 3.125,
"learning_rate": 0.0004946545727295724,
"loss": 5.464,
"mean_token_accuracy": 0.1652519851922989,
"num_tokens": 17847835.0,
"step": 8595
},
{
"entropy": 5.729661703109741,
"epoch": 0.7801161103047896,
"grad_norm": 1.7109375,
"learning_rate": 0.0004946475594491359,
"loss": 5.559,
"mean_token_accuracy": 0.16285682022571563,
"num_tokens": 17858324.0,
"step": 8600
},
{
"entropy": 5.791449785232544,
"epoch": 0.7805696661828737,
"grad_norm": 2.3125,
"learning_rate": 0.0004946405416263319,
"loss": 5.6036,
"mean_token_accuracy": 0.15357813686132432,
"num_tokens": 17868455.0,
"step": 8605
},
{
"entropy": 5.839656734466553,
"epoch": 0.7810232220609579,
"grad_norm": 2.328125,
"learning_rate": 0.0004946335192613054,
"loss": 5.6171,
"mean_token_accuracy": 0.15866965800523758,
"num_tokens": 17879038.0,
"step": 8610
},
{
"entropy": 5.6663342952728275,
"epoch": 0.7814767779390421,
"grad_norm": 2.578125,
"learning_rate": 0.0004946264923542019,
"loss": 5.4686,
"mean_token_accuracy": 0.16128309816122055,
"num_tokens": 17888153.0,
"step": 8615
},
{
"entropy": 5.67993278503418,
"epoch": 0.7819303338171263,
"grad_norm": 3.046875,
"learning_rate": 0.0004946194609051664,
"loss": 5.6109,
"mean_token_accuracy": 0.16191176623106002,
"num_tokens": 17898289.0,
"step": 8620
},
{
"entropy": 5.754681921005249,
"epoch": 0.7823838896952104,
"grad_norm": 2.265625,
"learning_rate": 0.0004946124249143447,
"loss": 5.6054,
"mean_token_accuracy": 0.15954133123159409,
"num_tokens": 17909192.0,
"step": 8625
},
{
"entropy": 5.712142086029052,
"epoch": 0.7828374455732946,
"grad_norm": 3.140625,
"learning_rate": 0.000494605384381882,
"loss": 5.5466,
"mean_token_accuracy": 0.15608295798301697,
"num_tokens": 17918937.0,
"step": 8630
},
{
"entropy": 5.738121271133423,
"epoch": 0.7832910014513788,
"grad_norm": 2.921875,
"learning_rate": 0.0004945983393079239,
"loss": 5.6076,
"mean_token_accuracy": 0.16086831092834472,
"num_tokens": 17929346.0,
"step": 8635
},
{
"entropy": 5.758172702789307,
"epoch": 0.783744557329463,
"grad_norm": 1.7265625,
"learning_rate": 0.0004945912896926163,
"loss": 5.5852,
"mean_token_accuracy": 0.16034913659095765,
"num_tokens": 17939479.0,
"step": 8640
},
{
"entropy": 5.853363275527954,
"epoch": 0.7841981132075472,
"grad_norm": 2.265625,
"learning_rate": 0.0004945842355361049,
"loss": 5.6648,
"mean_token_accuracy": 0.15613145902752876,
"num_tokens": 17949841.0,
"step": 8645
},
{
"entropy": 5.746683597564697,
"epoch": 0.7846516690856313,
"grad_norm": 2.03125,
"learning_rate": 0.0004945771768385355,
"loss": 5.6014,
"mean_token_accuracy": 0.1604134052991867,
"num_tokens": 17960334.0,
"step": 8650
},
{
"entropy": 5.675119495391845,
"epoch": 0.7851052249637155,
"grad_norm": 2.453125,
"learning_rate": 0.000494570113600054,
"loss": 5.5415,
"mean_token_accuracy": 0.16287485212087632,
"num_tokens": 17970595.0,
"step": 8655
},
{
"entropy": 5.7526007175445555,
"epoch": 0.7855587808417998,
"grad_norm": 3.84375,
"learning_rate": 0.0004945630458208068,
"loss": 5.6828,
"mean_token_accuracy": 0.15230964571237565,
"num_tokens": 17981301.0,
"step": 8660
},
{
"entropy": 5.6740015029907225,
"epoch": 0.7860123367198839,
"grad_norm": 4.5,
"learning_rate": 0.0004945559735009397,
"loss": 5.5156,
"mean_token_accuracy": 0.16261018067598343,
"num_tokens": 17991262.0,
"step": 8665
},
{
"entropy": 5.749355220794678,
"epoch": 0.7864658925979681,
"grad_norm": 2.359375,
"learning_rate": 0.0004945488966405993,
"loss": 5.566,
"mean_token_accuracy": 0.16028408706188202,
"num_tokens": 18001212.0,
"step": 8670
},
{
"entropy": 5.75854902267456,
"epoch": 0.7869194484760522,
"grad_norm": 3.171875,
"learning_rate": 0.0004945418152399317,
"loss": 5.5743,
"mean_token_accuracy": 0.1625185176730156,
"num_tokens": 18010432.0,
"step": 8675
},
{
"entropy": 5.762905740737915,
"epoch": 0.7873730043541364,
"grad_norm": 2.1875,
"learning_rate": 0.0004945347292990833,
"loss": 5.6108,
"mean_token_accuracy": 0.15499364733695983,
"num_tokens": 18020902.0,
"step": 8680
},
{
"entropy": 5.739315366744995,
"epoch": 0.7878265602322206,
"grad_norm": 2.171875,
"learning_rate": 0.0004945276388182008,
"loss": 5.4799,
"mean_token_accuracy": 0.1620177999138832,
"num_tokens": 18031133.0,
"step": 8685
},
{
"entropy": 5.8286998748779295,
"epoch": 0.7882801161103048,
"grad_norm": 2.375,
"learning_rate": 0.0004945205437974309,
"loss": 5.6143,
"mean_token_accuracy": 0.15827255249023436,
"num_tokens": 18042050.0,
"step": 8690
},
{
"entropy": 5.670088768005371,
"epoch": 0.788733671988389,
"grad_norm": 2.78125,
"learning_rate": 0.0004945134442369201,
"loss": 5.5594,
"mean_token_accuracy": 0.16206253170967103,
"num_tokens": 18051642.0,
"step": 8695
},
{
"entropy": 5.752692031860351,
"epoch": 0.7891872278664731,
"grad_norm": 2.625,
"learning_rate": 0.0004945063401368154,
"loss": 5.6145,
"mean_token_accuracy": 0.15136099308729173,
"num_tokens": 18062180.0,
"step": 8700
},
{
"entropy": 5.785160160064697,
"epoch": 0.7896407837445574,
"grad_norm": 3.3125,
"learning_rate": 0.0004944992314972637,
"loss": 5.5671,
"mean_token_accuracy": 0.15702632069587708,
"num_tokens": 18072302.0,
"step": 8705
},
{
"entropy": 5.765988302230835,
"epoch": 0.7900943396226415,
"grad_norm": 2.8125,
"learning_rate": 0.0004944921183184118,
"loss": 5.5781,
"mean_token_accuracy": 0.1638057678937912,
"num_tokens": 18081824.0,
"step": 8710
},
{
"entropy": 5.75767707824707,
"epoch": 0.7905478955007257,
"grad_norm": 3.84375,
"learning_rate": 0.000494485000600407,
"loss": 5.6276,
"mean_token_accuracy": 0.15722796022892,
"num_tokens": 18092696.0,
"step": 8715
},
{
"entropy": 5.683173036575317,
"epoch": 0.7910014513788098,
"grad_norm": 2.265625,
"learning_rate": 0.0004944778783433965,
"loss": 5.5345,
"mean_token_accuracy": 0.1658234864473343,
"num_tokens": 18101790.0,
"step": 8720
},
{
"entropy": 5.7710024356842045,
"epoch": 0.791455007256894,
"grad_norm": 2.546875,
"learning_rate": 0.0004944707515475276,
"loss": 5.6583,
"mean_token_accuracy": 0.15452334880828858,
"num_tokens": 18111572.0,
"step": 8725
},
{
"entropy": 5.827447557449341,
"epoch": 0.7919085631349783,
"grad_norm": 3.890625,
"learning_rate": 0.0004944636202129475,
"loss": 5.6427,
"mean_token_accuracy": 0.16203664243221283,
"num_tokens": 18120650.0,
"step": 8730
},
{
"entropy": 5.746698474884033,
"epoch": 0.7923621190130624,
"grad_norm": 2.15625,
"learning_rate": 0.0004944564843398039,
"loss": 5.5385,
"mean_token_accuracy": 0.16859735995531083,
"num_tokens": 18131055.0,
"step": 8735
},
{
"entropy": 5.779311656951904,
"epoch": 0.7928156748911466,
"grad_norm": 2.078125,
"learning_rate": 0.0004944493439282443,
"loss": 5.6266,
"mean_token_accuracy": 0.1563514158129692,
"num_tokens": 18141423.0,
"step": 8740
},
{
"entropy": 5.705952596664429,
"epoch": 0.7932692307692307,
"grad_norm": 3.375,
"learning_rate": 0.0004944421989784162,
"loss": 5.5581,
"mean_token_accuracy": 0.15892547070980073,
"num_tokens": 18152015.0,
"step": 8745
},
{
"entropy": 5.773093366622925,
"epoch": 0.793722786647315,
"grad_norm": 2.359375,
"learning_rate": 0.0004944350494904676,
"loss": 5.6469,
"mean_token_accuracy": 0.15265853106975555,
"num_tokens": 18161934.0,
"step": 8750
},
{
"entropy": 5.790240049362183,
"epoch": 0.7941763425253991,
"grad_norm": 2.90625,
"learning_rate": 0.0004944278954645463,
"loss": 5.5568,
"mean_token_accuracy": 0.16019119322299957,
"num_tokens": 18172287.0,
"step": 8755
},
{
"entropy": 5.752348232269287,
"epoch": 0.7946298984034833,
"grad_norm": 2.296875,
"learning_rate": 0.0004944207369008002,
"loss": 5.6221,
"mean_token_accuracy": 0.1672727346420288,
"num_tokens": 18181015.0,
"step": 8760
},
{
"entropy": 5.778642749786377,
"epoch": 0.7950834542815675,
"grad_norm": 2.453125,
"learning_rate": 0.0004944135737993773,
"loss": 5.6999,
"mean_token_accuracy": 0.15213402807712556,
"num_tokens": 18192382.0,
"step": 8765
},
{
"entropy": 5.7769043922424315,
"epoch": 0.7955370101596516,
"grad_norm": 2.671875,
"learning_rate": 0.0004944064061604258,
"loss": 5.5714,
"mean_token_accuracy": 0.1669528529047966,
"num_tokens": 18202597.0,
"step": 8770
},
{
"entropy": 5.782745456695556,
"epoch": 0.7959905660377359,
"grad_norm": 1.875,
"learning_rate": 0.0004943992339840939,
"loss": 5.6143,
"mean_token_accuracy": 0.15483993738889695,
"num_tokens": 18213756.0,
"step": 8775
},
{
"entropy": 5.652510166168213,
"epoch": 0.79644412191582,
"grad_norm": 2.046875,
"learning_rate": 0.00049439205727053,
"loss": 5.5573,
"mean_token_accuracy": 0.15681147426366807,
"num_tokens": 18223669.0,
"step": 8780
},
{
"entropy": 5.733938455581665,
"epoch": 0.7968976777939042,
"grad_norm": 1.9765625,
"learning_rate": 0.0004943848760198823,
"loss": 5.5877,
"mean_token_accuracy": 0.16213256418704985,
"num_tokens": 18234884.0,
"step": 8785
},
{
"entropy": 5.734215450286865,
"epoch": 0.7973512336719883,
"grad_norm": 1.921875,
"learning_rate": 0.0004943776902322995,
"loss": 5.5166,
"mean_token_accuracy": 0.16281318217515944,
"num_tokens": 18245155.0,
"step": 8790
},
{
"entropy": 5.6022714138031,
"epoch": 0.7978047895500726,
"grad_norm": 1.7890625,
"learning_rate": 0.0004943704999079302,
"loss": 5.4641,
"mean_token_accuracy": 0.1630011886358261,
"num_tokens": 18254686.0,
"step": 8795
},
{
"entropy": 5.687679624557495,
"epoch": 0.7982583454281568,
"grad_norm": 2.96875,
"learning_rate": 0.0004943633050469231,
"loss": 5.4844,
"mean_token_accuracy": 0.1627473920583725,
"num_tokens": 18265473.0,
"step": 8800
},
{
"entropy": 5.779775762557984,
"epoch": 0.7987119013062409,
"grad_norm": 2.65625,
"learning_rate": 0.0004943561056494269,
"loss": 5.6915,
"mean_token_accuracy": 0.1531972423195839,
"num_tokens": 18276024.0,
"step": 8805
},
{
"entropy": 5.760975074768067,
"epoch": 0.7991654571843251,
"grad_norm": 1.7109375,
"learning_rate": 0.0004943489017155906,
"loss": 5.6456,
"mean_token_accuracy": 0.14553324580192567,
"num_tokens": 18287261.0,
"step": 8810
},
{
"entropy": 5.757473945617676,
"epoch": 0.7996190130624092,
"grad_norm": 1.671875,
"learning_rate": 0.0004943416932455631,
"loss": 5.6599,
"mean_token_accuracy": 0.15629383623600007,
"num_tokens": 18299374.0,
"step": 8815
},
{
"entropy": 5.683649587631225,
"epoch": 0.8000725689404935,
"grad_norm": 2.171875,
"learning_rate": 0.0004943344802394935,
"loss": 5.612,
"mean_token_accuracy": 0.1526947721838951,
"num_tokens": 18309344.0,
"step": 8820
},
{
"entropy": 5.776987028121948,
"epoch": 0.8005261248185777,
"grad_norm": 1.8359375,
"learning_rate": 0.0004943272626975309,
"loss": 5.6144,
"mean_token_accuracy": 0.15544928908348082,
"num_tokens": 18319778.0,
"step": 8825
},
{
"entropy": 5.779241418838501,
"epoch": 0.8009796806966618,
"grad_norm": 2.234375,
"learning_rate": 0.0004943200406198246,
"loss": 5.61,
"mean_token_accuracy": 0.1576782390475273,
"num_tokens": 18329787.0,
"step": 8830
},
{
"entropy": 5.666056108474732,
"epoch": 0.801433236574746,
"grad_norm": 3.953125,
"learning_rate": 0.000494312814006524,
"loss": 5.5551,
"mean_token_accuracy": 0.1649927094578743,
"num_tokens": 18340670.0,
"step": 8835
},
{
"entropy": 5.717959022521972,
"epoch": 0.8018867924528302,
"grad_norm": 2.34375,
"learning_rate": 0.0004943055828577784,
"loss": 5.5826,
"mean_token_accuracy": 0.1525124877691269,
"num_tokens": 18352484.0,
"step": 8840
},
{
"entropy": 5.7433929443359375,
"epoch": 0.8023403483309144,
"grad_norm": 3.03125,
"learning_rate": 0.0004942983471737376,
"loss": 5.5262,
"mean_token_accuracy": 0.1591146096587181,
"num_tokens": 18363103.0,
"step": 8845
},
{
"entropy": 5.667786931991577,
"epoch": 0.8027939042089985,
"grad_norm": 2.375,
"learning_rate": 0.0004942911069545511,
"loss": 5.5843,
"mean_token_accuracy": 0.16140155047178267,
"num_tokens": 18373925.0,
"step": 8850
},
{
"entropy": 5.766671943664551,
"epoch": 0.8032474600870827,
"grad_norm": 2.625,
"learning_rate": 0.0004942838622003686,
"loss": 5.6506,
"mean_token_accuracy": 0.1573023408651352,
"num_tokens": 18384081.0,
"step": 8855
},
{
"entropy": 5.812975883483887,
"epoch": 0.8037010159651669,
"grad_norm": 1.6953125,
"learning_rate": 0.0004942766129113399,
"loss": 5.5441,
"mean_token_accuracy": 0.1619538903236389,
"num_tokens": 18392510.0,
"step": 8860
},
{
"entropy": 5.8084618091583256,
"epoch": 0.8041545718432511,
"grad_norm": 2.109375,
"learning_rate": 0.0004942693590876151,
"loss": 5.6509,
"mean_token_accuracy": 0.15825066938996316,
"num_tokens": 18402809.0,
"step": 8865
},
{
"entropy": 5.805789184570313,
"epoch": 0.8046081277213353,
"grad_norm": 2.40625,
"learning_rate": 0.000494262100729344,
"loss": 5.6909,
"mean_token_accuracy": 0.1581580124795437,
"num_tokens": 18413342.0,
"step": 8870
},
{
"entropy": 5.7798045635223385,
"epoch": 0.8050616835994194,
"grad_norm": 3.234375,
"learning_rate": 0.0004942548378366768,
"loss": 5.6826,
"mean_token_accuracy": 0.1586776688694954,
"num_tokens": 18424735.0,
"step": 8875
},
{
"entropy": 5.720227336883545,
"epoch": 0.8055152394775036,
"grad_norm": 1.8359375,
"learning_rate": 0.0004942475704097635,
"loss": 5.5634,
"mean_token_accuracy": 0.16137248426675796,
"num_tokens": 18434797.0,
"step": 8880
},
{
"entropy": 5.74770712852478,
"epoch": 0.8059687953555879,
"grad_norm": 2.734375,
"learning_rate": 0.0004942402984487548,
"loss": 5.6327,
"mean_token_accuracy": 0.15215983539819716,
"num_tokens": 18445456.0,
"step": 8885
},
{
"entropy": 5.698119592666626,
"epoch": 0.806422351233672,
"grad_norm": 2.59375,
"learning_rate": 0.0004942330219538008,
"loss": 5.5914,
"mean_token_accuracy": 0.1551514595746994,
"num_tokens": 18456728.0,
"step": 8890
},
{
"entropy": 5.965833568572998,
"epoch": 0.8068759071117562,
"grad_norm": 2.515625,
"learning_rate": 0.000494225740925052,
"loss": 5.7198,
"mean_token_accuracy": 0.15536733493208885,
"num_tokens": 18468813.0,
"step": 8895
},
{
"entropy": 5.853075170516968,
"epoch": 0.8073294629898403,
"grad_norm": 2.609375,
"learning_rate": 0.0004942184553626591,
"loss": 5.6342,
"mean_token_accuracy": 0.1581724092364311,
"num_tokens": 18477912.0,
"step": 8900
},
{
"entropy": 5.71473879814148,
"epoch": 0.8077830188679245,
"grad_norm": 1.890625,
"learning_rate": 0.0004942111652667726,
"loss": 5.566,
"mean_token_accuracy": 0.16377193331718445,
"num_tokens": 18488115.0,
"step": 8905
},
{
"entropy": 5.696315670013428,
"epoch": 0.8082365747460087,
"grad_norm": 2.578125,
"learning_rate": 0.0004942038706375433,
"loss": 5.6123,
"mean_token_accuracy": 0.15618089884519576,
"num_tokens": 18498853.0,
"step": 8910
},
{
"entropy": 5.74398775100708,
"epoch": 0.8086901306240929,
"grad_norm": 2.234375,
"learning_rate": 0.0004941965714751222,
"loss": 5.6499,
"mean_token_accuracy": 0.15113084837794305,
"num_tokens": 18510047.0,
"step": 8915
},
{
"entropy": 5.7405627250671385,
"epoch": 0.809143686502177,
"grad_norm": 2.28125,
"learning_rate": 0.0004941892677796601,
"loss": 5.6142,
"mean_token_accuracy": 0.16218534708023072,
"num_tokens": 18519673.0,
"step": 8920
},
{
"entropy": 5.759426736831665,
"epoch": 0.8095972423802612,
"grad_norm": 2.015625,
"learning_rate": 0.0004941819595513081,
"loss": 5.6437,
"mean_token_accuracy": 0.1601275697350502,
"num_tokens": 18530090.0,
"step": 8925
},
{
"entropy": 5.853196859359741,
"epoch": 0.8100507982583455,
"grad_norm": 2.4375,
"learning_rate": 0.0004941746467902173,
"loss": 5.6738,
"mean_token_accuracy": 0.1514594793319702,
"num_tokens": 18541862.0,
"step": 8930
},
{
"entropy": 5.817650413513183,
"epoch": 0.8105043541364296,
"grad_norm": 2.53125,
"learning_rate": 0.000494167329496539,
"loss": 5.6518,
"mean_token_accuracy": 0.147237129509449,
"num_tokens": 18551624.0,
"step": 8935
},
{
"entropy": 5.733151960372925,
"epoch": 0.8109579100145138,
"grad_norm": 1.75,
"learning_rate": 0.0004941600076704244,
"loss": 5.6118,
"mean_token_accuracy": 0.1600864425301552,
"num_tokens": 18561784.0,
"step": 8940
},
{
"entropy": 5.7939455032348635,
"epoch": 0.8114114658925979,
"grad_norm": 6.4375,
"learning_rate": 0.0004941526813120251,
"loss": 5.6836,
"mean_token_accuracy": 0.15367743968963624,
"num_tokens": 18572110.0,
"step": 8945
},
{
"entropy": 5.769635391235352,
"epoch": 0.8118650217706821,
"grad_norm": 2.6875,
"learning_rate": 0.0004941453504214924,
"loss": 5.6259,
"mean_token_accuracy": 0.1491383895277977,
"num_tokens": 18582012.0,
"step": 8950
},
{
"entropy": 5.6899253845214846,
"epoch": 0.8123185776487664,
"grad_norm": 1.953125,
"learning_rate": 0.0004941380149989781,
"loss": 5.4788,
"mean_token_accuracy": 0.16632459461688995,
"num_tokens": 18592850.0,
"step": 8955
},
{
"entropy": 5.748365592956543,
"epoch": 0.8127721335268505,
"grad_norm": 4.21875,
"learning_rate": 0.0004941306750446336,
"loss": 5.645,
"mean_token_accuracy": 0.1560724064707756,
"num_tokens": 18602590.0,
"step": 8960
},
{
"entropy": 5.822504663467408,
"epoch": 0.8132256894049347,
"grad_norm": 2.515625,
"learning_rate": 0.000494123330558611,
"loss": 5.5522,
"mean_token_accuracy": 0.15976229161024094,
"num_tokens": 18612607.0,
"step": 8965
},
{
"entropy": 5.706759023666382,
"epoch": 0.8136792452830188,
"grad_norm": 1.7734375,
"learning_rate": 0.0004941159815410622,
"loss": 5.5347,
"mean_token_accuracy": 0.1639419287443161,
"num_tokens": 18622936.0,
"step": 8970
},
{
"entropy": 5.783347320556641,
"epoch": 0.8141328011611031,
"grad_norm": 2.6875,
"learning_rate": 0.000494108627992139,
"loss": 5.7089,
"mean_token_accuracy": 0.1504053518176079,
"num_tokens": 18633997.0,
"step": 8975
},
{
"entropy": 5.779018068313599,
"epoch": 0.8145863570391872,
"grad_norm": 2.484375,
"learning_rate": 0.0004941012699119936,
"loss": 5.5268,
"mean_token_accuracy": 0.16512532979249955,
"num_tokens": 18644347.0,
"step": 8980
},
{
"entropy": 5.722207021713257,
"epoch": 0.8150399129172714,
"grad_norm": 2.109375,
"learning_rate": 0.0004940939073007779,
"loss": 5.608,
"mean_token_accuracy": 0.1625707656145096,
"num_tokens": 18654989.0,
"step": 8985
},
{
"entropy": 5.705081415176392,
"epoch": 0.8154934687953556,
"grad_norm": 1.8125,
"learning_rate": 0.0004940865401586446,
"loss": 5.5902,
"mean_token_accuracy": 0.15785395950078965,
"num_tokens": 18665526.0,
"step": 8990
},
{
"entropy": 5.767810583114624,
"epoch": 0.8159470246734397,
"grad_norm": 2.109375,
"learning_rate": 0.0004940791684857458,
"loss": 5.591,
"mean_token_accuracy": 0.16355077922344208,
"num_tokens": 18676226.0,
"step": 8995
},
{
"entropy": 5.763008880615234,
"epoch": 0.816400580551524,
"grad_norm": 2.8125,
"learning_rate": 0.0004940717922822339,
"loss": 5.5201,
"mean_token_accuracy": 0.16101572811603546,
"num_tokens": 18686499.0,
"step": 9000
},
{
"epoch": 0.816400580551524,
"eval_entropy": 5.6369146356755255,
"eval_loss": 5.623342037200928,
"eval_mean_token_accuracy": 0.16392953284590492,
"eval_num_tokens": 18686499.0,
"eval_runtime": 26.5654,
"eval_samples_per_second": 1331.054,
"eval_steps_per_second": 166.382,
"step": 9000
}
],
"logging_steps": 5,
"max_steps": 110230,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2.7328623446016e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink