fpadovani/rus-cyrl-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
b5e94cb78b06e4e52afca337a8c672ea99200454
rus-cyrl-10mb-after-ppt-shu…/checkpoint-3000/trainer_state.json
ModelHub XC b5e94cb78b 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/rus-cyrl-10mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
Source: Original Platform
2026-07-30 04:08:22 +08:00

6046 lines
164 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.2621690116228262,
"eval_steps": 3000,
"global_step": 3000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.896485424041748,
"epoch": 0.0004369483527047103,
"grad_norm": 19.125,
"learning_rate": 2e-06,
"loss": 14.1403,
"mean_token_accuracy": 0.0004707288753706962,
"num_tokens": 10478.0,
"step": 5
},
{
"entropy": 4.8918811798095705,
"epoch": 0.0008738967054094206,
"grad_norm": 19.125,
"learning_rate": 4.5e-06,
"loss": 14.1677,
"mean_token_accuracy": 0.0009102159005124122,
"num_tokens": 21152.0,
"step": 10
},
{
"entropy": 4.9260358810424805,
"epoch": 0.0013108450581141308,
"grad_norm": 21.25,
"learning_rate": 7e-06,
"loss": 14.0135,
"mean_token_accuracy": 0.00043343613506294785,
"num_tokens": 31246.0,
"step": 15
},
{
"entropy": 5.002536630630493,
"epoch": 0.0017477934108188413,
"grad_norm": 27.75,
"learning_rate": 9.5e-06,
"loss": 13.7803,
"mean_token_accuracy": 0.00043386173201724886,
"num_tokens": 41934.0,
"step": 20
},
{
"entropy": 5.2952552318573,
"epoch": 0.0021847417635235517,
"grad_norm": 33.5,
"learning_rate": 1.2e-05,
"loss": 13.2447,
"mean_token_accuracy": 0.00043462926405481996,
"num_tokens": 52830.0,
"step": 25
},
{
"entropy": 6.374371004104614,
"epoch": 0.0026216901162282617,
"grad_norm": 37.75,
"learning_rate": 1.4500000000000002e-05,
"loss": 12.2103,
"mean_token_accuracy": 0.0001860347867477685,
"num_tokens": 62706.0,
"step": 30
},
{
"entropy": 9.60359296798706,
"epoch": 0.003058638468932972,
"grad_norm": 3.5625,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.9691,
"mean_token_accuracy": 0.0,
"num_tokens": 72619.0,
"step": 35
},
{
"entropy": 10.6598690032959,
"epoch": 0.0034955868216376825,
"grad_norm": 3.984375,
"learning_rate": 1.95e-05,
"loss": 10.6826,
"mean_token_accuracy": 0.0027851671096868814,
"num_tokens": 82903.0,
"step": 40
},
{
"entropy": 10.72455415725708,
"epoch": 0.003932535174342393,
"grad_norm": 3.09375,
"learning_rate": 2.2e-05,
"loss": 10.5546,
"mean_token_accuracy": 0.011984702153131365,
"num_tokens": 94243.0,
"step": 45
},
{
"entropy": 10.698091506958008,
"epoch": 0.004369483527047103,
"grad_norm": 2.515625,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.3469,
"mean_token_accuracy": 0.017781304474920035,
"num_tokens": 103618.0,
"step": 50
},
{
"entropy": 10.646620655059815,
"epoch": 0.004806431879751813,
"grad_norm": 2.09375,
"learning_rate": 2.7e-05,
"loss": 10.1899,
"mean_token_accuracy": 0.01909907590597868,
"num_tokens": 112180.0,
"step": 55
},
{
"entropy": 10.682601261138917,
"epoch": 0.005243380232456523,
"grad_norm": 2.296875,
"learning_rate": 2.95e-05,
"loss": 10.0898,
"mean_token_accuracy": 0.01639490285888314,
"num_tokens": 122462.0,
"step": 60
},
{
"entropy": 10.703626728057861,
"epoch": 0.005680328585161234,
"grad_norm": 1.890625,
"learning_rate": 3.2e-05,
"loss": 9.9926,
"mean_token_accuracy": 0.029707580246031283,
"num_tokens": 132741.0,
"step": 65
},
{
"entropy": 10.696713733673096,
"epoch": 0.006117276937865944,
"grad_norm": 2.0625,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.8566,
"mean_token_accuracy": 0.05769682787358761,
"num_tokens": 142973.0,
"step": 70
},
{
"entropy": 10.690417957305907,
"epoch": 0.006554225290570655,
"grad_norm": 1.796875,
"learning_rate": 3.7e-05,
"loss": 9.8804,
"mean_token_accuracy": 0.05056782606989145,
"num_tokens": 153199.0,
"step": 75
},
{
"entropy": 10.685645198822021,
"epoch": 0.006991173643275365,
"grad_norm": 1.8359375,
"learning_rate": 3.95e-05,
"loss": 9.768,
"mean_token_accuracy": 0.05694139041006565,
"num_tokens": 163079.0,
"step": 80
},
{
"entropy": 10.66970682144165,
"epoch": 0.0074281219959800755,
"grad_norm": 1.765625,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.71,
"mean_token_accuracy": 0.06373655460774899,
"num_tokens": 173308.0,
"step": 85
},
{
"entropy": 10.657975387573241,
"epoch": 0.007865070348684786,
"grad_norm": 1.6640625,
"learning_rate": 4.45e-05,
"loss": 9.6122,
"mean_token_accuracy": 0.0695813860744238,
"num_tokens": 183856.0,
"step": 90
},
{
"entropy": 10.634114265441895,
"epoch": 0.008302018701389496,
"grad_norm": 1.796875,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.5337,
"mean_token_accuracy": 0.060128339752554896,
"num_tokens": 193368.0,
"step": 95
},
{
"entropy": 10.613899898529052,
"epoch": 0.008738967054094207,
"grad_norm": 1.75,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.4681,
"mean_token_accuracy": 0.06204190067946911,
"num_tokens": 202686.0,
"step": 100
},
{
"entropy": 10.591835975646973,
"epoch": 0.009175915406798917,
"grad_norm": 1.7265625,
"learning_rate": 5.2e-05,
"loss": 9.3641,
"mean_token_accuracy": 0.06793517135083675,
"num_tokens": 213237.0,
"step": 105
},
{
"entropy": 10.56372594833374,
"epoch": 0.009612863759503626,
"grad_norm": 1.59375,
"learning_rate": 5.45e-05,
"loss": 9.3006,
"mean_token_accuracy": 0.06861048303544522,
"num_tokens": 222406.0,
"step": 110
},
{
"entropy": 10.523597049713135,
"epoch": 0.010049812112208336,
"grad_norm": 1.6640625,
"learning_rate": 5.7e-05,
"loss": 9.2349,
"mean_token_accuracy": 0.06383556276559829,
"num_tokens": 232139.0,
"step": 115
},
{
"entropy": 10.474488925933837,
"epoch": 0.010486760464913047,
"grad_norm": 1.40625,
"learning_rate": 5.9499999999999996e-05,
"loss": 9.1603,
"mean_token_accuracy": 0.06192168295383453,
"num_tokens": 242438.0,
"step": 120
},
{
"entropy": 10.428980922698974,
"epoch": 0.010923708817617757,
"grad_norm": 1.484375,
"learning_rate": 6.2e-05,
"loss": 9.0417,
"mean_token_accuracy": 0.06691903881728649,
"num_tokens": 253082.0,
"step": 125
},
{
"entropy": 10.344741821289062,
"epoch": 0.011360657170322468,
"grad_norm": 1.515625,
"learning_rate": 6.450000000000001e-05,
"loss": 8.9769,
"mean_token_accuracy": 0.07010326012969018,
"num_tokens": 264067.0,
"step": 130
},
{
"entropy": 10.246838283538818,
"epoch": 0.011797605523027178,
"grad_norm": 1.4453125,
"learning_rate": 6.7e-05,
"loss": 8.8954,
"mean_token_accuracy": 0.061506147310137746,
"num_tokens": 274571.0,
"step": 135
},
{
"entropy": 10.181257820129394,
"epoch": 0.012234553875731888,
"grad_norm": 1.4921875,
"learning_rate": 6.950000000000001e-05,
"loss": 8.7923,
"mean_token_accuracy": 0.06518892385065556,
"num_tokens": 284561.0,
"step": 140
},
{
"entropy": 10.101185703277588,
"epoch": 0.012671502228436599,
"grad_norm": 1.2421875,
"learning_rate": 7.2e-05,
"loss": 8.653,
"mean_token_accuracy": 0.06580178663134575,
"num_tokens": 295366.0,
"step": 145
},
{
"entropy": 10.015565013885498,
"epoch": 0.01310845058114131,
"grad_norm": 1.0625,
"learning_rate": 7.45e-05,
"loss": 8.6783,
"mean_token_accuracy": 0.06620895192027092,
"num_tokens": 305423.0,
"step": 150
},
{
"entropy": 9.899238300323486,
"epoch": 0.01354539893384602,
"grad_norm": 1.109375,
"learning_rate": 7.7e-05,
"loss": 8.5289,
"mean_token_accuracy": 0.06539785154163838,
"num_tokens": 315182.0,
"step": 155
},
{
"entropy": 9.770205783843995,
"epoch": 0.01398234728655073,
"grad_norm": 1.1171875,
"learning_rate": 7.950000000000001e-05,
"loss": 8.4638,
"mean_token_accuracy": 0.0697988323867321,
"num_tokens": 324780.0,
"step": 160
},
{
"entropy": 9.595983505249023,
"epoch": 0.01441929563925544,
"grad_norm": 0.83984375,
"learning_rate": 8.2e-05,
"loss": 8.4388,
"mean_token_accuracy": 0.06649183593690396,
"num_tokens": 335104.0,
"step": 165
},
{
"entropy": 9.459788513183593,
"epoch": 0.014856243991960151,
"grad_norm": 0.9453125,
"learning_rate": 8.450000000000001e-05,
"loss": 8.361,
"mean_token_accuracy": 0.06445354036986828,
"num_tokens": 345073.0,
"step": 170
},
{
"entropy": 9.294847583770752,
"epoch": 0.015293192344664861,
"grad_norm": 0.8984375,
"learning_rate": 8.7e-05,
"loss": 8.32,
"mean_token_accuracy": 0.057173581421375276,
"num_tokens": 354342.0,
"step": 175
},
{
"entropy": 9.100028991699219,
"epoch": 0.015730140697369572,
"grad_norm": 0.75390625,
"learning_rate": 8.95e-05,
"loss": 8.21,
"mean_token_accuracy": 0.06767738796770573,
"num_tokens": 364441.0,
"step": 180
},
{
"entropy": 8.956842231750489,
"epoch": 0.01616708905007428,
"grad_norm": 1.140625,
"learning_rate": 9.2e-05,
"loss": 8.2102,
"mean_token_accuracy": 0.06345368176698685,
"num_tokens": 373918.0,
"step": 185
},
{
"entropy": 8.87299690246582,
"epoch": 0.016604037402778993,
"grad_norm": 0.75390625,
"learning_rate": 9.45e-05,
"loss": 8.2134,
"mean_token_accuracy": 0.061258262023329735,
"num_tokens": 383273.0,
"step": 190
},
{
"entropy": 8.69699010848999,
"epoch": 0.0170409857554837,
"grad_norm": 0.8671875,
"learning_rate": 9.7e-05,
"loss": 8.0918,
"mean_token_accuracy": 0.07180375047028065,
"num_tokens": 393262.0,
"step": 195
},
{
"entropy": 8.610151290893555,
"epoch": 0.017477934108188414,
"grad_norm": 0.66796875,
"learning_rate": 9.95e-05,
"loss": 8.194,
"mean_token_accuracy": 0.061346839740872385,
"num_tokens": 402389.0,
"step": 200
},
{
"entropy": 8.630728244781494,
"epoch": 0.017914882460893122,
"grad_norm": 0.74609375,
"learning_rate": 0.000102,
"loss": 8.1296,
"mean_token_accuracy": 0.06509425193071365,
"num_tokens": 412997.0,
"step": 205
},
{
"entropy": 8.581557655334473,
"epoch": 0.018351830813597834,
"grad_norm": 0.6875,
"learning_rate": 0.00010449999999999999,
"loss": 8.1323,
"mean_token_accuracy": 0.06776191368699074,
"num_tokens": 422222.0,
"step": 210
},
{
"entropy": 8.459476566314697,
"epoch": 0.018788779166302543,
"grad_norm": 0.66015625,
"learning_rate": 0.000107,
"loss": 8.0772,
"mean_token_accuracy": 0.06964709646999837,
"num_tokens": 432539.0,
"step": 215
},
{
"entropy": 8.583655071258544,
"epoch": 0.019225727519007252,
"grad_norm": 0.6796875,
"learning_rate": 0.0001095,
"loss": 8.0763,
"mean_token_accuracy": 0.07036846987903118,
"num_tokens": 442311.0,
"step": 220
},
{
"entropy": 8.413945198059082,
"epoch": 0.019662675871711964,
"grad_norm": 0.70703125,
"learning_rate": 0.000112,
"loss": 8.0586,
"mean_token_accuracy": 0.06845476403832436,
"num_tokens": 452727.0,
"step": 225
},
{
"entropy": 8.500650024414062,
"epoch": 0.020099624224416673,
"grad_norm": 0.88671875,
"learning_rate": 0.0001145,
"loss": 8.0851,
"mean_token_accuracy": 0.07086162865161896,
"num_tokens": 462461.0,
"step": 230
},
{
"entropy": 8.524639797210693,
"epoch": 0.020536572577121385,
"grad_norm": 1.0234375,
"learning_rate": 0.00011700000000000001,
"loss": 8.0572,
"mean_token_accuracy": 0.07282712981104851,
"num_tokens": 472362.0,
"step": 235
},
{
"entropy": 8.431649208068848,
"epoch": 0.020973520929826094,
"grad_norm": 0.76953125,
"learning_rate": 0.00011949999999999999,
"loss": 8.0476,
"mean_token_accuracy": 0.07426296398043633,
"num_tokens": 482718.0,
"step": 240
},
{
"entropy": 8.454423236846925,
"epoch": 0.021410469282530806,
"grad_norm": 0.703125,
"learning_rate": 0.000122,
"loss": 8.0537,
"mean_token_accuracy": 0.07146380059421062,
"num_tokens": 492947.0,
"step": 245
},
{
"entropy": 8.405343437194825,
"epoch": 0.021847417635235514,
"grad_norm": 0.92578125,
"learning_rate": 0.0001245,
"loss": 8.0819,
"mean_token_accuracy": 0.06912912614643574,
"num_tokens": 502957.0,
"step": 250
},
{
"entropy": 8.408589458465576,
"epoch": 0.022284365987940227,
"grad_norm": 0.83984375,
"learning_rate": 0.000127,
"loss": 8.037,
"mean_token_accuracy": 0.07069992832839489,
"num_tokens": 513220.0,
"step": 255
},
{
"entropy": 8.416099166870117,
"epoch": 0.022721314340644935,
"grad_norm": 0.74609375,
"learning_rate": 0.0001295,
"loss": 7.9798,
"mean_token_accuracy": 0.07073883488774299,
"num_tokens": 523217.0,
"step": 260
},
{
"entropy": 8.420017719268799,
"epoch": 0.023158262693349647,
"grad_norm": 0.78515625,
"learning_rate": 0.000132,
"loss": 8.0086,
"mean_token_accuracy": 0.06777435280382633,
"num_tokens": 533789.0,
"step": 265
},
{
"entropy": 8.379833126068116,
"epoch": 0.023595211046054356,
"grad_norm": 0.75390625,
"learning_rate": 0.00013450000000000002,
"loss": 7.9839,
"mean_token_accuracy": 0.07827216312289238,
"num_tokens": 544146.0,
"step": 270
},
{
"entropy": 8.337002563476563,
"epoch": 0.02403215939875907,
"grad_norm": 0.7265625,
"learning_rate": 0.00013700000000000002,
"loss": 8.0224,
"mean_token_accuracy": 0.06755217313766479,
"num_tokens": 554215.0,
"step": 275
},
{
"entropy": 8.356577014923095,
"epoch": 0.024469107751463777,
"grad_norm": 0.8671875,
"learning_rate": 0.0001395,
"loss": 7.9919,
"mean_token_accuracy": 0.0684168841689825,
"num_tokens": 564587.0,
"step": 280
},
{
"entropy": 8.399262714385987,
"epoch": 0.024906056104168486,
"grad_norm": 0.8125,
"learning_rate": 0.00014199999999999998,
"loss": 8.0404,
"mean_token_accuracy": 0.07138842865824699,
"num_tokens": 574819.0,
"step": 285
},
{
"entropy": 8.372067642211913,
"epoch": 0.025343004456873198,
"grad_norm": 0.8125,
"learning_rate": 0.0001445,
"loss": 7.9538,
"mean_token_accuracy": 0.07292748279869557,
"num_tokens": 585178.0,
"step": 290
},
{
"entropy": 8.315776348114014,
"epoch": 0.025779952809577907,
"grad_norm": 0.859375,
"learning_rate": 0.000147,
"loss": 7.9432,
"mean_token_accuracy": 0.07378942035138607,
"num_tokens": 595086.0,
"step": 295
},
{
"entropy": 8.330533981323242,
"epoch": 0.02621690116228262,
"grad_norm": 0.7265625,
"learning_rate": 0.0001495,
"loss": 7.9452,
"mean_token_accuracy": 0.07910944558680058,
"num_tokens": 604942.0,
"step": 300
},
{
"entropy": 8.304435920715331,
"epoch": 0.026653849514987327,
"grad_norm": 0.91015625,
"learning_rate": 0.000152,
"loss": 7.9291,
"mean_token_accuracy": 0.07603786624968052,
"num_tokens": 616082.0,
"step": 305
},
{
"entropy": 8.352412796020507,
"epoch": 0.02709079786769204,
"grad_norm": 0.921875,
"learning_rate": 0.00015450000000000001,
"loss": 7.9481,
"mean_token_accuracy": 0.07412574216723442,
"num_tokens": 626375.0,
"step": 310
},
{
"entropy": 8.253109741210938,
"epoch": 0.027527746220396748,
"grad_norm": 0.84765625,
"learning_rate": 0.000157,
"loss": 7.9133,
"mean_token_accuracy": 0.07124261260032654,
"num_tokens": 637563.0,
"step": 315
},
{
"entropy": 8.222493076324463,
"epoch": 0.02796469457310146,
"grad_norm": 0.859375,
"learning_rate": 0.0001595,
"loss": 7.9205,
"mean_token_accuracy": 0.07289120145142078,
"num_tokens": 647256.0,
"step": 320
},
{
"entropy": 8.32628002166748,
"epoch": 0.02840164292580617,
"grad_norm": 0.78125,
"learning_rate": 0.000162,
"loss": 7.9747,
"mean_token_accuracy": 0.07336639389395713,
"num_tokens": 657247.0,
"step": 325
},
{
"entropy": 8.336363315582275,
"epoch": 0.02883859127851088,
"grad_norm": 1.125,
"learning_rate": 0.00016450000000000001,
"loss": 7.8957,
"mean_token_accuracy": 0.07113592401146888,
"num_tokens": 667613.0,
"step": 330
},
{
"entropy": 8.263669490814209,
"epoch": 0.02927553963121559,
"grad_norm": 0.8203125,
"learning_rate": 0.00016700000000000002,
"loss": 7.9176,
"mean_token_accuracy": 0.07490575201809406,
"num_tokens": 678048.0,
"step": 335
},
{
"entropy": 8.235548973083496,
"epoch": 0.029712487983920302,
"grad_norm": 0.9296875,
"learning_rate": 0.00016950000000000003,
"loss": 7.9202,
"mean_token_accuracy": 0.07333524562418461,
"num_tokens": 688468.0,
"step": 340
},
{
"entropy": 8.21406421661377,
"epoch": 0.03014943633662501,
"grad_norm": 0.7734375,
"learning_rate": 0.00017199999999999998,
"loss": 7.885,
"mean_token_accuracy": 0.078083186596632,
"num_tokens": 699809.0,
"step": 345
},
{
"entropy": 8.228328514099122,
"epoch": 0.030586384689329723,
"grad_norm": 0.78125,
"learning_rate": 0.00017449999999999999,
"loss": 7.8454,
"mean_token_accuracy": 0.07926662042737007,
"num_tokens": 709227.0,
"step": 350
},
{
"entropy": 8.25043773651123,
"epoch": 0.03102333304203443,
"grad_norm": 1.0,
"learning_rate": 0.000177,
"loss": 7.9612,
"mean_token_accuracy": 0.07075673900544643,
"num_tokens": 720178.0,
"step": 355
},
{
"entropy": 8.20638427734375,
"epoch": 0.031460281394739144,
"grad_norm": 0.75390625,
"learning_rate": 0.0001795,
"loss": 7.8735,
"mean_token_accuracy": 0.07839432805776596,
"num_tokens": 730514.0,
"step": 360
},
{
"entropy": 8.303699684143066,
"epoch": 0.03189722974744385,
"grad_norm": 0.79296875,
"learning_rate": 0.000182,
"loss": 7.9139,
"mean_token_accuracy": 0.07378935068845749,
"num_tokens": 741345.0,
"step": 365
},
{
"entropy": 8.170354652404786,
"epoch": 0.03233417810014856,
"grad_norm": 0.890625,
"learning_rate": 0.0001845,
"loss": 7.8564,
"mean_token_accuracy": 0.07958053424954414,
"num_tokens": 751163.0,
"step": 370
},
{
"entropy": 8.218745231628418,
"epoch": 0.03277112645285327,
"grad_norm": 1.21875,
"learning_rate": 0.000187,
"loss": 7.8577,
"mean_token_accuracy": 0.07420928552746772,
"num_tokens": 762090.0,
"step": 375
},
{
"entropy": 8.191785526275634,
"epoch": 0.033208074805557986,
"grad_norm": 1.21875,
"learning_rate": 0.0001895,
"loss": 7.831,
"mean_token_accuracy": 0.07720481157302857,
"num_tokens": 772049.0,
"step": 380
},
{
"entropy": 8.150869464874267,
"epoch": 0.03364502315826269,
"grad_norm": 1.0625,
"learning_rate": 0.000192,
"loss": 7.7882,
"mean_token_accuracy": 0.08157530575990676,
"num_tokens": 781280.0,
"step": 385
},
{
"entropy": 8.15115146636963,
"epoch": 0.0340819715109674,
"grad_norm": 1.0546875,
"learning_rate": 0.0001945,
"loss": 7.8737,
"mean_token_accuracy": 0.07709059566259384,
"num_tokens": 791642.0,
"step": 390
},
{
"entropy": 8.20585823059082,
"epoch": 0.034518919863672115,
"grad_norm": 0.83984375,
"learning_rate": 0.00019700000000000002,
"loss": 7.8441,
"mean_token_accuracy": 0.07621582746505737,
"num_tokens": 801224.0,
"step": 395
},
{
"entropy": 8.199099254608154,
"epoch": 0.03495586821637683,
"grad_norm": 0.91796875,
"learning_rate": 0.00019950000000000002,
"loss": 7.7916,
"mean_token_accuracy": 0.07633770778775215,
"num_tokens": 810727.0,
"step": 400
},
{
"entropy": 8.255587577819824,
"epoch": 0.03539281656908153,
"grad_norm": 1.046875,
"learning_rate": 0.000202,
"loss": 7.8705,
"mean_token_accuracy": 0.07444445230066776,
"num_tokens": 819943.0,
"step": 405
},
{
"entropy": 8.048962116241455,
"epoch": 0.035829764921786245,
"grad_norm": 0.86328125,
"learning_rate": 0.00020449999999999998,
"loss": 7.7977,
"mean_token_accuracy": 0.07640674635767937,
"num_tokens": 829474.0,
"step": 410
},
{
"entropy": 8.137162065505981,
"epoch": 0.03626671327449096,
"grad_norm": 1.015625,
"learning_rate": 0.000207,
"loss": 7.7691,
"mean_token_accuracy": 0.0750108852982521,
"num_tokens": 839288.0,
"step": 415
},
{
"entropy": 8.195742893218995,
"epoch": 0.03670366162719567,
"grad_norm": 0.84765625,
"learning_rate": 0.0002095,
"loss": 7.793,
"mean_token_accuracy": 0.07822757884860039,
"num_tokens": 850064.0,
"step": 420
},
{
"entropy": 8.037298011779786,
"epoch": 0.037140609979900374,
"grad_norm": 0.88671875,
"learning_rate": 0.000212,
"loss": 7.7651,
"mean_token_accuracy": 0.08010210767388344,
"num_tokens": 860732.0,
"step": 425
},
{
"entropy": 8.097493267059326,
"epoch": 0.037577558332605086,
"grad_norm": 0.875,
"learning_rate": 0.0002145,
"loss": 7.7724,
"mean_token_accuracy": 0.0774189442396164,
"num_tokens": 870076.0,
"step": 430
},
{
"entropy": 8.05912742614746,
"epoch": 0.0380145066853098,
"grad_norm": 0.90625,
"learning_rate": 0.00021700000000000002,
"loss": 7.6706,
"mean_token_accuracy": 0.08113809823989868,
"num_tokens": 880201.0,
"step": 435
},
{
"entropy": 8.00949149131775,
"epoch": 0.038451455038014504,
"grad_norm": 0.7265625,
"learning_rate": 0.0002195,
"loss": 7.7294,
"mean_token_accuracy": 0.08016021810472011,
"num_tokens": 890344.0,
"step": 440
},
{
"entropy": 7.997669124603272,
"epoch": 0.038888403390719216,
"grad_norm": 0.9609375,
"learning_rate": 0.000222,
"loss": 7.7322,
"mean_token_accuracy": 0.0784931242465973,
"num_tokens": 900443.0,
"step": 445
},
{
"entropy": 8.134346008300781,
"epoch": 0.03932535174342393,
"grad_norm": 0.7734375,
"learning_rate": 0.0002245,
"loss": 7.7529,
"mean_token_accuracy": 0.07578973248600959,
"num_tokens": 910284.0,
"step": 450
},
{
"entropy": 8.158567905426025,
"epoch": 0.03976230009612864,
"grad_norm": 1.0234375,
"learning_rate": 0.00022700000000000002,
"loss": 7.8343,
"mean_token_accuracy": 0.07556049451231957,
"num_tokens": 920986.0,
"step": 455
},
{
"entropy": 7.912718629837036,
"epoch": 0.040199248448833345,
"grad_norm": 0.875,
"learning_rate": 0.00022950000000000002,
"loss": 7.7393,
"mean_token_accuracy": 0.08137748837471008,
"num_tokens": 930481.0,
"step": 460
},
{
"entropy": 8.1414954662323,
"epoch": 0.04063619680153806,
"grad_norm": 0.875,
"learning_rate": 0.00023200000000000003,
"loss": 7.764,
"mean_token_accuracy": 0.08360744118690491,
"num_tokens": 940554.0,
"step": 465
},
{
"entropy": 8.014440441131592,
"epoch": 0.04107314515424277,
"grad_norm": 0.97265625,
"learning_rate": 0.00023449999999999998,
"loss": 7.6943,
"mean_token_accuracy": 0.08052223920822144,
"num_tokens": 950005.0,
"step": 470
},
{
"entropy": 8.032914018630981,
"epoch": 0.04151009350694748,
"grad_norm": 0.7265625,
"learning_rate": 0.000237,
"loss": 7.6964,
"mean_token_accuracy": 0.0742336355149746,
"num_tokens": 959414.0,
"step": 475
},
{
"entropy": 8.050236415863036,
"epoch": 0.04194704185965219,
"grad_norm": 0.83203125,
"learning_rate": 0.0002395,
"loss": 7.7366,
"mean_token_accuracy": 0.07452398836612702,
"num_tokens": 969352.0,
"step": 480
},
{
"entropy": 8.132762765884399,
"epoch": 0.0423839902123569,
"grad_norm": 0.82421875,
"learning_rate": 0.000242,
"loss": 7.7524,
"mean_token_accuracy": 0.07607904821634293,
"num_tokens": 980294.0,
"step": 485
},
{
"entropy": 7.920047473907471,
"epoch": 0.04282093856506161,
"grad_norm": 0.98046875,
"learning_rate": 0.0002445,
"loss": 7.6372,
"mean_token_accuracy": 0.0846704125404358,
"num_tokens": 990453.0,
"step": 490
},
{
"entropy": 8.083022117614746,
"epoch": 0.04325788691776632,
"grad_norm": 1.1640625,
"learning_rate": 0.000247,
"loss": 7.6619,
"mean_token_accuracy": 0.0859624132514,
"num_tokens": 1000509.0,
"step": 495
},
{
"entropy": 8.04277811050415,
"epoch": 0.04369483527047103,
"grad_norm": 0.8984375,
"learning_rate": 0.0002495,
"loss": 7.7908,
"mean_token_accuracy": 0.07289138622581959,
"num_tokens": 1011747.0,
"step": 500
},
{
"entropy": 8.003987789154053,
"epoch": 0.04413178362317574,
"grad_norm": 1.109375,
"learning_rate": 0.000252,
"loss": 7.7292,
"mean_token_accuracy": 0.07683630660176277,
"num_tokens": 1021511.0,
"step": 505
},
{
"entropy": 8.018040466308594,
"epoch": 0.04456873197588045,
"grad_norm": 1.140625,
"learning_rate": 0.0002545,
"loss": 7.7397,
"mean_token_accuracy": 0.07697788439691067,
"num_tokens": 1032116.0,
"step": 510
},
{
"entropy": 7.920564413070679,
"epoch": 0.04500568032858516,
"grad_norm": 0.98828125,
"learning_rate": 0.000257,
"loss": 7.6298,
"mean_token_accuracy": 0.08404637426137924,
"num_tokens": 1041356.0,
"step": 515
},
{
"entropy": 7.942330694198608,
"epoch": 0.04544262868128987,
"grad_norm": 1.0859375,
"learning_rate": 0.0002595,
"loss": 7.7028,
"mean_token_accuracy": 0.07776187285780907,
"num_tokens": 1051314.0,
"step": 520
},
{
"entropy": 8.040958166122437,
"epoch": 0.04587957703399458,
"grad_norm": 0.89453125,
"learning_rate": 0.000262,
"loss": 7.6852,
"mean_token_accuracy": 0.08220702111721039,
"num_tokens": 1060857.0,
"step": 525
},
{
"entropy": 8.020647382736206,
"epoch": 0.046316525386699295,
"grad_norm": 0.984375,
"learning_rate": 0.00026450000000000003,
"loss": 7.6951,
"mean_token_accuracy": 0.07845601513981819,
"num_tokens": 1071744.0,
"step": 530
},
{
"entropy": 7.9700311660766605,
"epoch": 0.046753473739404,
"grad_norm": 0.8671875,
"learning_rate": 0.00026700000000000004,
"loss": 7.6381,
"mean_token_accuracy": 0.08291278034448624,
"num_tokens": 1080777.0,
"step": 535
},
{
"entropy": 7.986575984954834,
"epoch": 0.04719042209210871,
"grad_norm": 1.0234375,
"learning_rate": 0.00026950000000000005,
"loss": 7.6373,
"mean_token_accuracy": 0.08483078628778458,
"num_tokens": 1090951.0,
"step": 540
},
{
"entropy": 7.9460187435150145,
"epoch": 0.047627370444813424,
"grad_norm": 1.1171875,
"learning_rate": 0.00027200000000000005,
"loss": 7.6187,
"mean_token_accuracy": 0.07862685099244118,
"num_tokens": 1101667.0,
"step": 545
},
{
"entropy": 7.9729594707489015,
"epoch": 0.04806431879751814,
"grad_norm": 1.21875,
"learning_rate": 0.0002745,
"loss": 7.7162,
"mean_token_accuracy": 0.07553440034389496,
"num_tokens": 1112652.0,
"step": 550
},
{
"entropy": 7.875039529800415,
"epoch": 0.04850126715022284,
"grad_norm": 0.9140625,
"learning_rate": 0.000277,
"loss": 7.6821,
"mean_token_accuracy": 0.08174617812037469,
"num_tokens": 1122518.0,
"step": 555
},
{
"entropy": 8.032792949676514,
"epoch": 0.048938215502927554,
"grad_norm": 1.078125,
"learning_rate": 0.0002795,
"loss": 7.6805,
"mean_token_accuracy": 0.08198884315788746,
"num_tokens": 1132867.0,
"step": 560
},
{
"entropy": 7.95278844833374,
"epoch": 0.049375163855632266,
"grad_norm": 0.9609375,
"learning_rate": 0.00028199999999999997,
"loss": 7.7381,
"mean_token_accuracy": 0.07585004493594169,
"num_tokens": 1142524.0,
"step": 565
},
{
"entropy": 8.021573352813721,
"epoch": 0.04981211220833697,
"grad_norm": 1.140625,
"learning_rate": 0.0002845,
"loss": 7.6283,
"mean_token_accuracy": 0.08681625351309777,
"num_tokens": 1152983.0,
"step": 570
},
{
"entropy": 7.975176620483398,
"epoch": 0.050249060561041684,
"grad_norm": 0.94140625,
"learning_rate": 0.000287,
"loss": 7.6872,
"mean_token_accuracy": 0.08164099827408791,
"num_tokens": 1164271.0,
"step": 575
},
{
"entropy": 7.9183494567871096,
"epoch": 0.050686008913746396,
"grad_norm": 0.953125,
"learning_rate": 0.0002895,
"loss": 7.541,
"mean_token_accuracy": 0.08640064373612404,
"num_tokens": 1173567.0,
"step": 580
},
{
"entropy": 7.924293804168701,
"epoch": 0.05112295726645111,
"grad_norm": 1.0859375,
"learning_rate": 0.000292,
"loss": 7.5932,
"mean_token_accuracy": 0.08321595788002015,
"num_tokens": 1183619.0,
"step": 585
},
{
"entropy": 7.873846387863159,
"epoch": 0.05155990561915581,
"grad_norm": 0.9609375,
"learning_rate": 0.0002945,
"loss": 7.5453,
"mean_token_accuracy": 0.0821442298591137,
"num_tokens": 1193922.0,
"step": 590
},
{
"entropy": 7.94070348739624,
"epoch": 0.051996853971860525,
"grad_norm": 1.171875,
"learning_rate": 0.000297,
"loss": 7.6536,
"mean_token_accuracy": 0.0783975936472416,
"num_tokens": 1203665.0,
"step": 595
},
{
"entropy": 7.832107400894165,
"epoch": 0.05243380232456524,
"grad_norm": 0.9375,
"learning_rate": 0.0002995,
"loss": 7.6423,
"mean_token_accuracy": 0.0841694563627243,
"num_tokens": 1214119.0,
"step": 600
},
{
"entropy": 7.960584402084351,
"epoch": 0.05287075067726995,
"grad_norm": 1.2265625,
"learning_rate": 0.000302,
"loss": 7.6685,
"mean_token_accuracy": 0.07720493376255036,
"num_tokens": 1225275.0,
"step": 605
},
{
"entropy": 7.838187599182129,
"epoch": 0.053307699029974655,
"grad_norm": 1.078125,
"learning_rate": 0.0003045,
"loss": 7.5356,
"mean_token_accuracy": 0.08203468546271324,
"num_tokens": 1235199.0,
"step": 610
},
{
"entropy": 7.830732774734497,
"epoch": 0.05374464738267937,
"grad_norm": 0.84375,
"learning_rate": 0.000307,
"loss": 7.5585,
"mean_token_accuracy": 0.08500108271837234,
"num_tokens": 1245648.0,
"step": 615
},
{
"entropy": 7.852081918716431,
"epoch": 0.05418159573538408,
"grad_norm": 0.9453125,
"learning_rate": 0.0003095,
"loss": 7.4995,
"mean_token_accuracy": 0.08836419135332108,
"num_tokens": 1254793.0,
"step": 620
},
{
"entropy": 7.985955429077149,
"epoch": 0.05461854408808879,
"grad_norm": 0.96875,
"learning_rate": 0.000312,
"loss": 7.5573,
"mean_token_accuracy": 0.0857199877500534,
"num_tokens": 1264429.0,
"step": 625
},
{
"entropy": 7.8001152038574215,
"epoch": 0.055055492440793496,
"grad_norm": 1.1484375,
"learning_rate": 0.0003145,
"loss": 7.5317,
"mean_token_accuracy": 0.08314449116587638,
"num_tokens": 1274449.0,
"step": 630
},
{
"entropy": 7.813184595108032,
"epoch": 0.05549244079349821,
"grad_norm": 1.1484375,
"learning_rate": 0.000317,
"loss": 7.6222,
"mean_token_accuracy": 0.07996979467570782,
"num_tokens": 1284987.0,
"step": 635
},
{
"entropy": 7.950613355636596,
"epoch": 0.05592938914620292,
"grad_norm": 0.94140625,
"learning_rate": 0.0003195,
"loss": 7.6343,
"mean_token_accuracy": 0.07661416307091713,
"num_tokens": 1295984.0,
"step": 640
},
{
"entropy": 7.814332628250122,
"epoch": 0.056366337498907626,
"grad_norm": 1.0078125,
"learning_rate": 0.000322,
"loss": 7.5214,
"mean_token_accuracy": 0.08348236829042435,
"num_tokens": 1305738.0,
"step": 645
},
{
"entropy": 7.804314756393433,
"epoch": 0.05680328585161234,
"grad_norm": 1.0390625,
"learning_rate": 0.00032450000000000003,
"loss": 7.4795,
"mean_token_accuracy": 0.07995264530181885,
"num_tokens": 1315691.0,
"step": 650
},
{
"entropy": 7.752275705337524,
"epoch": 0.05724023420431705,
"grad_norm": 0.84765625,
"learning_rate": 0.00032700000000000003,
"loss": 7.4882,
"mean_token_accuracy": 0.09246278628706932,
"num_tokens": 1326195.0,
"step": 655
},
{
"entropy": 7.765911626815796,
"epoch": 0.05767718255702176,
"grad_norm": 1.1328125,
"learning_rate": 0.00032950000000000004,
"loss": 7.5164,
"mean_token_accuracy": 0.08554120138287544,
"num_tokens": 1336916.0,
"step": 660
},
{
"entropy": 7.800547170639038,
"epoch": 0.05811413090972647,
"grad_norm": 1.0546875,
"learning_rate": 0.00033200000000000005,
"loss": 7.5327,
"mean_token_accuracy": 0.08561304733157157,
"num_tokens": 1347386.0,
"step": 665
},
{
"entropy": 7.849845457077026,
"epoch": 0.05855107926243118,
"grad_norm": 1.1015625,
"learning_rate": 0.00033450000000000005,
"loss": 7.5301,
"mean_token_accuracy": 0.08292968645691871,
"num_tokens": 1357959.0,
"step": 670
},
{
"entropy": 7.800776195526123,
"epoch": 0.05898802761513589,
"grad_norm": 1.0234375,
"learning_rate": 0.000337,
"loss": 7.4917,
"mean_token_accuracy": 0.08568408265709877,
"num_tokens": 1368135.0,
"step": 675
},
{
"entropy": 7.8050871849060055,
"epoch": 0.059424975967840604,
"grad_norm": 0.90234375,
"learning_rate": 0.0003395,
"loss": 7.5337,
"mean_token_accuracy": 0.08384500592947006,
"num_tokens": 1378702.0,
"step": 680
},
{
"entropy": 7.720953702926636,
"epoch": 0.05986192432054531,
"grad_norm": 1.1640625,
"learning_rate": 0.000342,
"loss": 7.3887,
"mean_token_accuracy": 0.08918830305337906,
"num_tokens": 1387902.0,
"step": 685
},
{
"entropy": 7.801330518722534,
"epoch": 0.06029887267325002,
"grad_norm": 1.015625,
"learning_rate": 0.00034449999999999997,
"loss": 7.4145,
"mean_token_accuracy": 0.09191453829407692,
"num_tokens": 1397288.0,
"step": 690
},
{
"entropy": 7.620101451873779,
"epoch": 0.060735821025954734,
"grad_norm": 0.96875,
"learning_rate": 0.000347,
"loss": 7.4673,
"mean_token_accuracy": 0.08888877630233764,
"num_tokens": 1407416.0,
"step": 695
},
{
"entropy": 7.842352247238159,
"epoch": 0.061172769378659446,
"grad_norm": 0.95703125,
"learning_rate": 0.0003495,
"loss": 7.5187,
"mean_token_accuracy": 0.08510556891560554,
"num_tokens": 1417733.0,
"step": 700
},
{
"entropy": 7.77461576461792,
"epoch": 0.06160971773136415,
"grad_norm": 1.6015625,
"learning_rate": 0.000352,
"loss": 7.4219,
"mean_token_accuracy": 0.0847130298614502,
"num_tokens": 1426723.0,
"step": 705
},
{
"entropy": 7.718229103088379,
"epoch": 0.06204666608406886,
"grad_norm": 1.1484375,
"learning_rate": 0.0003545,
"loss": 7.4886,
"mean_token_accuracy": 0.08168335855007172,
"num_tokens": 1437290.0,
"step": 710
},
{
"entropy": 7.7265349388122555,
"epoch": 0.062483614436773575,
"grad_norm": 1.1015625,
"learning_rate": 0.000357,
"loss": 7.4101,
"mean_token_accuracy": 0.08986271098256111,
"num_tokens": 1446756.0,
"step": 715
},
{
"entropy": 7.732719802856446,
"epoch": 0.06292056278947829,
"grad_norm": 1.1328125,
"learning_rate": 0.0003595,
"loss": 7.4766,
"mean_token_accuracy": 0.08280604630708695,
"num_tokens": 1457074.0,
"step": 720
},
{
"entropy": 7.706831645965576,
"epoch": 0.06335751114218299,
"grad_norm": 1.109375,
"learning_rate": 0.000362,
"loss": 7.5222,
"mean_token_accuracy": 0.08136801272630692,
"num_tokens": 1467785.0,
"step": 725
},
{
"entropy": 7.702896404266357,
"epoch": 0.0637944594948877,
"grad_norm": 0.9140625,
"learning_rate": 0.0003645,
"loss": 7.4124,
"mean_token_accuracy": 0.08612973466515542,
"num_tokens": 1477532.0,
"step": 730
},
{
"entropy": 7.657266235351562,
"epoch": 0.06423140784759242,
"grad_norm": 0.96484375,
"learning_rate": 0.000367,
"loss": 7.354,
"mean_token_accuracy": 0.08939684703946113,
"num_tokens": 1487102.0,
"step": 735
},
{
"entropy": 7.676237678527832,
"epoch": 0.06466835620029712,
"grad_norm": 1.015625,
"learning_rate": 0.0003695,
"loss": 7.4367,
"mean_token_accuracy": 0.08506104946136475,
"num_tokens": 1497459.0,
"step": 740
},
{
"entropy": 7.6919153213500975,
"epoch": 0.06510530455300184,
"grad_norm": 0.8671875,
"learning_rate": 0.000372,
"loss": 7.4116,
"mean_token_accuracy": 0.08696235418319702,
"num_tokens": 1508756.0,
"step": 745
},
{
"entropy": 7.655456924438477,
"epoch": 0.06554225290570655,
"grad_norm": 0.82421875,
"learning_rate": 0.0003745,
"loss": 7.334,
"mean_token_accuracy": 0.09340226426720619,
"num_tokens": 1518735.0,
"step": 750
},
{
"entropy": 7.632836103439331,
"epoch": 0.06597920125841125,
"grad_norm": 1.09375,
"learning_rate": 0.000377,
"loss": 7.373,
"mean_token_accuracy": 0.0900121010839939,
"num_tokens": 1528632.0,
"step": 755
},
{
"entropy": 7.59655385017395,
"epoch": 0.06641614961111597,
"grad_norm": 0.9609375,
"learning_rate": 0.0003795,
"loss": 7.3537,
"mean_token_accuracy": 0.09101043492555619,
"num_tokens": 1538180.0,
"step": 760
},
{
"entropy": 7.675789928436279,
"epoch": 0.06685309796382068,
"grad_norm": 1.0546875,
"learning_rate": 0.000382,
"loss": 7.3641,
"mean_token_accuracy": 0.09065457805991173,
"num_tokens": 1547788.0,
"step": 765
},
{
"entropy": 7.655756425857544,
"epoch": 0.06729004631652538,
"grad_norm": 1.046875,
"learning_rate": 0.0003845,
"loss": 7.3488,
"mean_token_accuracy": 0.09222541823983192,
"num_tokens": 1558008.0,
"step": 770
},
{
"entropy": 7.611498165130615,
"epoch": 0.0677269946692301,
"grad_norm": 1.1875,
"learning_rate": 0.00038700000000000003,
"loss": 7.3996,
"mean_token_accuracy": 0.08548890501260757,
"num_tokens": 1567728.0,
"step": 775
},
{
"entropy": 7.6110608100891115,
"epoch": 0.0681639430219348,
"grad_norm": 1.1015625,
"learning_rate": 0.00038950000000000003,
"loss": 7.3381,
"mean_token_accuracy": 0.08824502006173134,
"num_tokens": 1578488.0,
"step": 780
},
{
"entropy": 7.646148300170898,
"epoch": 0.06860089137463951,
"grad_norm": 1.0859375,
"learning_rate": 0.00039200000000000004,
"loss": 7.3251,
"mean_token_accuracy": 0.09305397123098373,
"num_tokens": 1588070.0,
"step": 785
},
{
"entropy": 7.577291822433471,
"epoch": 0.06903783972734423,
"grad_norm": 0.88671875,
"learning_rate": 0.00039450000000000005,
"loss": 7.4018,
"mean_token_accuracy": 0.08525708541274071,
"num_tokens": 1598572.0,
"step": 790
},
{
"entropy": 7.725144910812378,
"epoch": 0.06947478808004894,
"grad_norm": 1.171875,
"learning_rate": 0.00039700000000000005,
"loss": 7.3944,
"mean_token_accuracy": 0.0888714350759983,
"num_tokens": 1607982.0,
"step": 795
},
{
"entropy": 7.575393867492676,
"epoch": 0.06991173643275365,
"grad_norm": 1.234375,
"learning_rate": 0.0003995,
"loss": 7.4,
"mean_token_accuracy": 0.09165657237172127,
"num_tokens": 1618785.0,
"step": 800
},
{
"entropy": 7.640562582015991,
"epoch": 0.07034868478545836,
"grad_norm": 1.15625,
"learning_rate": 0.000402,
"loss": 7.3264,
"mean_token_accuracy": 0.09215063080191613,
"num_tokens": 1627988.0,
"step": 805
},
{
"entropy": 7.606148958206177,
"epoch": 0.07078563313816306,
"grad_norm": 1.28125,
"learning_rate": 0.0004045,
"loss": 7.4088,
"mean_token_accuracy": 0.08241811096668243,
"num_tokens": 1638024.0,
"step": 810
},
{
"entropy": 7.635509967803955,
"epoch": 0.07122258149086778,
"grad_norm": 1.1328125,
"learning_rate": 0.00040699999999999997,
"loss": 7.3984,
"mean_token_accuracy": 0.08276447355747223,
"num_tokens": 1649215.0,
"step": 815
},
{
"entropy": 7.5351354598999025,
"epoch": 0.07165952984357249,
"grad_norm": 0.8828125,
"learning_rate": 0.0004095,
"loss": 7.2742,
"mean_token_accuracy": 0.09341849088668823,
"num_tokens": 1659198.0,
"step": 820
},
{
"entropy": 7.591096258163452,
"epoch": 0.0720964781962772,
"grad_norm": 1.0703125,
"learning_rate": 0.000412,
"loss": 7.3213,
"mean_token_accuracy": 0.08587326556444168,
"num_tokens": 1668972.0,
"step": 825
},
{
"entropy": 7.505528831481934,
"epoch": 0.07253342654898191,
"grad_norm": 1.046875,
"learning_rate": 0.0004145,
"loss": 7.3155,
"mean_token_accuracy": 0.09277390204370022,
"num_tokens": 1679807.0,
"step": 830
},
{
"entropy": 7.6023828983306885,
"epoch": 0.07297037490168662,
"grad_norm": 1.0625,
"learning_rate": 0.000417,
"loss": 7.3551,
"mean_token_accuracy": 0.08633830323815346,
"num_tokens": 1690567.0,
"step": 835
},
{
"entropy": 7.440132284164429,
"epoch": 0.07340732325439134,
"grad_norm": 0.97265625,
"learning_rate": 0.0004195,
"loss": 7.2531,
"mean_token_accuracy": 0.09126588180661202,
"num_tokens": 1700939.0,
"step": 840
},
{
"entropy": 7.51141676902771,
"epoch": 0.07384427160709604,
"grad_norm": 0.92578125,
"learning_rate": 0.000422,
"loss": 7.2943,
"mean_token_accuracy": 0.09781363755464553,
"num_tokens": 1711181.0,
"step": 845
},
{
"entropy": 7.651186609268189,
"epoch": 0.07428121995980075,
"grad_norm": 1.0546875,
"learning_rate": 0.0004245,
"loss": 7.3272,
"mean_token_accuracy": 0.08526805341243744,
"num_tokens": 1721392.0,
"step": 850
},
{
"entropy": 7.507527637481689,
"epoch": 0.07471816831250547,
"grad_norm": 0.91796875,
"learning_rate": 0.000427,
"loss": 7.2344,
"mean_token_accuracy": 0.09549994468688965,
"num_tokens": 1731942.0,
"step": 855
},
{
"entropy": 7.518540573120117,
"epoch": 0.07515511666521017,
"grad_norm": 0.78515625,
"learning_rate": 0.0004295,
"loss": 7.3444,
"mean_token_accuracy": 0.08721731156110764,
"num_tokens": 1743409.0,
"step": 860
},
{
"entropy": 7.473425197601318,
"epoch": 0.07559206501791488,
"grad_norm": 0.8984375,
"learning_rate": 0.000432,
"loss": 7.2332,
"mean_token_accuracy": 0.09332804158329963,
"num_tokens": 1753425.0,
"step": 865
},
{
"entropy": 7.484533405303955,
"epoch": 0.0760290133706196,
"grad_norm": 0.90625,
"learning_rate": 0.0004345,
"loss": 7.263,
"mean_token_accuracy": 0.09156408160924911,
"num_tokens": 1762722.0,
"step": 870
},
{
"entropy": 7.63799090385437,
"epoch": 0.0764659617233243,
"grad_norm": 0.97265625,
"learning_rate": 0.000437,
"loss": 7.3416,
"mean_token_accuracy": 0.08278580531477928,
"num_tokens": 1771895.0,
"step": 875
},
{
"entropy": 7.450203514099121,
"epoch": 0.07690291007602901,
"grad_norm": 0.90625,
"learning_rate": 0.0004395,
"loss": 7.2387,
"mean_token_accuracy": 0.09428637102246284,
"num_tokens": 1782175.0,
"step": 880
},
{
"entropy": 7.548505687713623,
"epoch": 0.07733985842873373,
"grad_norm": 1.0390625,
"learning_rate": 0.000442,
"loss": 7.3003,
"mean_token_accuracy": 0.09130567386746406,
"num_tokens": 1792409.0,
"step": 885
},
{
"entropy": 7.5660957336425785,
"epoch": 0.07777680678143843,
"grad_norm": 1.0625,
"learning_rate": 0.0004445,
"loss": 7.342,
"mean_token_accuracy": 0.08967537879943847,
"num_tokens": 1802721.0,
"step": 890
},
{
"entropy": 7.403235387802124,
"epoch": 0.07821375513414315,
"grad_norm": 0.9765625,
"learning_rate": 0.000447,
"loss": 7.1835,
"mean_token_accuracy": 0.09422190114855766,
"num_tokens": 1812064.0,
"step": 895
},
{
"entropy": 7.539788246154785,
"epoch": 0.07865070348684786,
"grad_norm": 1.0,
"learning_rate": 0.00044950000000000003,
"loss": 7.2802,
"mean_token_accuracy": 0.0889531396329403,
"num_tokens": 1822284.0,
"step": 900
},
{
"entropy": 7.418588542938233,
"epoch": 0.07908765183955256,
"grad_norm": 1.03125,
"learning_rate": 0.00045200000000000004,
"loss": 7.1734,
"mean_token_accuracy": 0.10003943592309952,
"num_tokens": 1832150.0,
"step": 905
},
{
"entropy": 7.502434968948364,
"epoch": 0.07952460019225728,
"grad_norm": 1.0703125,
"learning_rate": 0.00045450000000000004,
"loss": 7.3111,
"mean_token_accuracy": 0.08902120292186737,
"num_tokens": 1842870.0,
"step": 910
},
{
"entropy": 7.440501403808594,
"epoch": 0.07996154854496199,
"grad_norm": 0.94921875,
"learning_rate": 0.00045700000000000005,
"loss": 7.1386,
"mean_token_accuracy": 0.10121746063232422,
"num_tokens": 1852771.0,
"step": 915
},
{
"entropy": 7.499157047271728,
"epoch": 0.08039849689766669,
"grad_norm": 0.9453125,
"learning_rate": 0.00045950000000000006,
"loss": 7.2641,
"mean_token_accuracy": 0.08988009616732598,
"num_tokens": 1863095.0,
"step": 920
},
{
"entropy": 7.418407726287842,
"epoch": 0.08083544525037141,
"grad_norm": 0.96875,
"learning_rate": 0.000462,
"loss": 7.2466,
"mean_token_accuracy": 0.09539340287446976,
"num_tokens": 1873060.0,
"step": 925
},
{
"entropy": 7.4301539897918705,
"epoch": 0.08127239360307612,
"grad_norm": 0.89453125,
"learning_rate": 0.0004645,
"loss": 7.1842,
"mean_token_accuracy": 0.09798025488853454,
"num_tokens": 1882574.0,
"step": 930
},
{
"entropy": 7.432036447525024,
"epoch": 0.08170934195578082,
"grad_norm": 0.91015625,
"learning_rate": 0.000467,
"loss": 7.2438,
"mean_token_accuracy": 0.09283970594406128,
"num_tokens": 1892227.0,
"step": 935
},
{
"entropy": 7.447924375534058,
"epoch": 0.08214629030848554,
"grad_norm": 1.03125,
"learning_rate": 0.0004695,
"loss": 7.2762,
"mean_token_accuracy": 0.0921746976673603,
"num_tokens": 1902868.0,
"step": 940
},
{
"entropy": 7.454946565628052,
"epoch": 0.08258323866119024,
"grad_norm": 0.86328125,
"learning_rate": 0.000472,
"loss": 7.2095,
"mean_token_accuracy": 0.091494170576334,
"num_tokens": 1912781.0,
"step": 945
},
{
"entropy": 7.3900199890136715,
"epoch": 0.08302018701389496,
"grad_norm": 0.83984375,
"learning_rate": 0.0004745,
"loss": 7.2071,
"mean_token_accuracy": 0.09347877204418183,
"num_tokens": 1923638.0,
"step": 950
},
{
"entropy": 7.410589408874512,
"epoch": 0.08345713536659967,
"grad_norm": 0.9609375,
"learning_rate": 0.000477,
"loss": 7.1738,
"mean_token_accuracy": 0.0997669093310833,
"num_tokens": 1933498.0,
"step": 955
},
{
"entropy": 7.345382642745972,
"epoch": 0.08389408371930437,
"grad_norm": 0.94921875,
"learning_rate": 0.0004795,
"loss": 7.1795,
"mean_token_accuracy": 0.09802865833044053,
"num_tokens": 1943798.0,
"step": 960
},
{
"entropy": 7.41419472694397,
"epoch": 0.0843310320720091,
"grad_norm": 1.0390625,
"learning_rate": 0.000482,
"loss": 7.2289,
"mean_token_accuracy": 0.08998453319072723,
"num_tokens": 1954235.0,
"step": 965
},
{
"entropy": 7.420694160461426,
"epoch": 0.0847679804247138,
"grad_norm": 0.875,
"learning_rate": 0.0004845,
"loss": 7.1975,
"mean_token_accuracy": 0.08924680277705192,
"num_tokens": 1964421.0,
"step": 970
},
{
"entropy": 7.348307037353516,
"epoch": 0.0852049287774185,
"grad_norm": 0.9296875,
"learning_rate": 0.000487,
"loss": 7.1912,
"mean_token_accuracy": 0.09613144919276237,
"num_tokens": 1975862.0,
"step": 975
},
{
"entropy": 7.43187837600708,
"epoch": 0.08564187713012322,
"grad_norm": 0.96875,
"learning_rate": 0.0004895,
"loss": 7.1916,
"mean_token_accuracy": 0.09355136156082153,
"num_tokens": 1985758.0,
"step": 980
},
{
"entropy": 7.304799795150757,
"epoch": 0.08607882548282793,
"grad_norm": 0.984375,
"learning_rate": 0.000492,
"loss": 7.1507,
"mean_token_accuracy": 0.09453744739294052,
"num_tokens": 1996904.0,
"step": 985
},
{
"entropy": 7.544142961502075,
"epoch": 0.08651577383553263,
"grad_norm": 0.94140625,
"learning_rate": 0.0004945,
"loss": 7.2387,
"mean_token_accuracy": 0.09486401230096816,
"num_tokens": 2006114.0,
"step": 990
},
{
"entropy": 7.3271314144134525,
"epoch": 0.08695272218823735,
"grad_norm": 0.9765625,
"learning_rate": 0.000497,
"loss": 7.187,
"mean_token_accuracy": 0.09526639729738236,
"num_tokens": 2016349.0,
"step": 995
},
{
"entropy": 7.326001262664795,
"epoch": 0.08738967054094206,
"grad_norm": 0.9921875,
"learning_rate": 0.0004995,
"loss": 7.1294,
"mean_token_accuracy": 0.09752104729413986,
"num_tokens": 2026701.0,
"step": 1000
},
{
"entropy": 7.363681697845459,
"epoch": 0.08782661889364678,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999999986190023,
"loss": 7.1313,
"mean_token_accuracy": 0.09394256100058555,
"num_tokens": 2036950.0,
"step": 1005
},
{
"entropy": 7.389071273803711,
"epoch": 0.08826356724635148,
"grad_norm": 0.984375,
"learning_rate": 0.0004999999930086991,
"loss": 7.1271,
"mean_token_accuracy": 0.09964141845703126,
"num_tokens": 2047655.0,
"step": 1010
},
{
"entropy": 7.2618661403656,
"epoch": 0.08870051559905619,
"grad_norm": 0.99609375,
"learning_rate": 0.000499999983082778,
"loss": 7.0867,
"mean_token_accuracy": 0.09885328710079193,
"num_tokens": 2056800.0,
"step": 1015
},
{
"entropy": 7.379116010665894,
"epoch": 0.0891374639517609,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999688412394,
"loss": 7.1357,
"mean_token_accuracy": 0.09616564437747002,
"num_tokens": 2066862.0,
"step": 1020
},
{
"entropy": 7.303624439239502,
"epoch": 0.08957441230446561,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999999502840836,
"loss": 7.1615,
"mean_token_accuracy": 0.09557585939764976,
"num_tokens": 2076585.0,
"step": 1025
},
{
"entropy": 7.345246982574463,
"epoch": 0.09001136065717032,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999999274113107,
"loss": 7.2304,
"mean_token_accuracy": 0.09007581695914268,
"num_tokens": 2087768.0,
"step": 1030
},
{
"entropy": 7.375523328781128,
"epoch": 0.09044830900987504,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999999002229214,
"loss": 7.1331,
"mean_token_accuracy": 0.09556613713502884,
"num_tokens": 2097027.0,
"step": 1035
},
{
"entropy": 7.343056964874267,
"epoch": 0.09088525736257974,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999998687189161,
"loss": 7.0493,
"mean_token_accuracy": 0.10475894957780837,
"num_tokens": 2107055.0,
"step": 1040
},
{
"entropy": 7.232951641082764,
"epoch": 0.09132220571528446,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999998328992954,
"loss": 6.9891,
"mean_token_accuracy": 0.09808351546525955,
"num_tokens": 2116771.0,
"step": 1045
},
{
"entropy": 7.251108074188233,
"epoch": 0.09175915406798917,
"grad_norm": 1.0546875,
"learning_rate": 0.00049999979276406,
"loss": 7.1275,
"mean_token_accuracy": 0.09823357462882995,
"num_tokens": 2126881.0,
"step": 1050
},
{
"entropy": 7.334365940093994,
"epoch": 0.09219610242069387,
"grad_norm": 1.09375,
"learning_rate": 0.0004999997483132108,
"loss": 7.1095,
"mean_token_accuracy": 0.09823260232806205,
"num_tokens": 2136964.0,
"step": 1055
},
{
"entropy": 7.285873460769653,
"epoch": 0.09263305077339859,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999996995467483,
"loss": 7.1465,
"mean_token_accuracy": 0.09619581401348114,
"num_tokens": 2147263.0,
"step": 1060
},
{
"entropy": 7.269978046417236,
"epoch": 0.0930699991261033,
"grad_norm": 0.90625,
"learning_rate": 0.0004999996464646738,
"loss": 7.1421,
"mean_token_accuracy": 0.09753268361091613,
"num_tokens": 2157474.0,
"step": 1065
},
{
"entropy": 7.301066493988037,
"epoch": 0.093506947478808,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999995890669882,
"loss": 7.0887,
"mean_token_accuracy": 0.09752897918224335,
"num_tokens": 2167701.0,
"step": 1070
},
{
"entropy": 7.1919232368469235,
"epoch": 0.09394389583151272,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999995273536925,
"loss": 7.0485,
"mean_token_accuracy": 0.10060491785407066,
"num_tokens": 2178581.0,
"step": 1075
},
{
"entropy": 7.291404294967651,
"epoch": 0.09438084418421742,
"grad_norm": 0.85546875,
"learning_rate": 0.000499999461324788,
"loss": 7.0571,
"mean_token_accuracy": 0.0973673515021801,
"num_tokens": 2189529.0,
"step": 1080
},
{
"entropy": 7.293972587585449,
"epoch": 0.09481779253692213,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999993909802759,
"loss": 7.1293,
"mean_token_accuracy": 0.09355574771761895,
"num_tokens": 2199805.0,
"step": 1085
},
{
"entropy": 7.219992399215698,
"epoch": 0.09525474088962685,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999993163201577,
"loss": 7.0882,
"mean_token_accuracy": 0.09490575790405273,
"num_tokens": 2210108.0,
"step": 1090
},
{
"entropy": 7.3503416061401365,
"epoch": 0.09569168924233155,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999992373444347,
"loss": 7.0748,
"mean_token_accuracy": 0.09794154390692711,
"num_tokens": 2220325.0,
"step": 1095
},
{
"entropy": 7.212787294387818,
"epoch": 0.09612863759503627,
"grad_norm": 0.921875,
"learning_rate": 0.0004999991540531084,
"loss": 7.078,
"mean_token_accuracy": 0.09457284510135651,
"num_tokens": 2230521.0,
"step": 1100
},
{
"entropy": 7.296592044830322,
"epoch": 0.09656558594774098,
"grad_norm": 1.125,
"learning_rate": 0.0004999990664461804,
"loss": 7.1008,
"mean_token_accuracy": 0.0982304073870182,
"num_tokens": 2240372.0,
"step": 1105
},
{
"entropy": 7.249142646789551,
"epoch": 0.09700253430044568,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999989745236524,
"loss": 7.0654,
"mean_token_accuracy": 0.10029410496354103,
"num_tokens": 2250522.0,
"step": 1110
},
{
"entropy": 7.172446870803833,
"epoch": 0.0974394826531504,
"grad_norm": 0.984375,
"learning_rate": 0.0004999988782855261,
"loss": 7.0139,
"mean_token_accuracy": 0.0998832106590271,
"num_tokens": 2260142.0,
"step": 1115
},
{
"entropy": 7.262675666809082,
"epoch": 0.09787643100585511,
"grad_norm": 0.96875,
"learning_rate": 0.0004999987777318035,
"loss": 7.0449,
"mean_token_accuracy": 0.0934498131275177,
"num_tokens": 2270372.0,
"step": 1120
},
{
"entropy": 7.145792722702026,
"epoch": 0.09831337935855981,
"grad_norm": 1.015625,
"learning_rate": 0.0004999986728624865,
"loss": 6.9603,
"mean_token_accuracy": 0.1059371255338192,
"num_tokens": 2280517.0,
"step": 1125
},
{
"entropy": 7.340591526031494,
"epoch": 0.09875032771126453,
"grad_norm": 0.98828125,
"learning_rate": 0.000499998563677577,
"loss": 7.0665,
"mean_token_accuracy": 0.097840715944767,
"num_tokens": 2290778.0,
"step": 1130
},
{
"entropy": 7.222999715805054,
"epoch": 0.09918727606396924,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999984501770771,
"loss": 7.1046,
"mean_token_accuracy": 0.09563738405704499,
"num_tokens": 2301692.0,
"step": 1135
},
{
"entropy": 7.17807970046997,
"epoch": 0.09962422441667394,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999983323609891,
"loss": 6.9737,
"mean_token_accuracy": 0.10515563935041428,
"num_tokens": 2312759.0,
"step": 1140
},
{
"entropy": 7.259163951873779,
"epoch": 0.10006117276937866,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999982102293153,
"loss": 6.9729,
"mean_token_accuracy": 0.10644988417625427,
"num_tokens": 2322016.0,
"step": 1145
},
{
"entropy": 7.230524587631225,
"epoch": 0.10049812112208337,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999980837820578,
"loss": 6.9835,
"mean_token_accuracy": 0.10811527222394943,
"num_tokens": 2332476.0,
"step": 1150
},
{
"entropy": 7.183452081680298,
"epoch": 0.10093506947478809,
"grad_norm": 0.984375,
"learning_rate": 0.0004999979530192192,
"loss": 7.0065,
"mean_token_accuracy": 0.0977212980389595,
"num_tokens": 2342914.0,
"step": 1155
},
{
"entropy": 7.134286832809448,
"epoch": 0.10137201782749279,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999978179408019,
"loss": 7.0586,
"mean_token_accuracy": 0.09377534911036492,
"num_tokens": 2353381.0,
"step": 1160
},
{
"entropy": 7.30219407081604,
"epoch": 0.1018089661801975,
"grad_norm": 0.859375,
"learning_rate": 0.0004999976785468085,
"loss": 7.0465,
"mean_token_accuracy": 0.10484937131404877,
"num_tokens": 2364207.0,
"step": 1165
},
{
"entropy": 7.110436010360718,
"epoch": 0.10224591453290222,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999975348372418,
"loss": 6.9636,
"mean_token_accuracy": 0.10252279341220856,
"num_tokens": 2374985.0,
"step": 1170
},
{
"entropy": 7.178512048721314,
"epoch": 0.10268286288560692,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999973868121044,
"loss": 7.0417,
"mean_token_accuracy": 0.10154772102832794,
"num_tokens": 2385684.0,
"step": 1175
},
{
"entropy": 7.308528852462769,
"epoch": 0.10311981123831163,
"grad_norm": 0.921875,
"learning_rate": 0.0004999972344713994,
"loss": 6.9786,
"mean_token_accuracy": 0.10362603068351746,
"num_tokens": 2395177.0,
"step": 1180
},
{
"entropy": 7.079153251647949,
"epoch": 0.10355675959101635,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999970778151295,
"loss": 6.9996,
"mean_token_accuracy": 0.1028733141720295,
"num_tokens": 2405154.0,
"step": 1185
},
{
"entropy": 7.193738412857056,
"epoch": 0.10399370794372105,
"grad_norm": 0.921875,
"learning_rate": 0.0004999969168432976,
"loss": 6.9717,
"mean_token_accuracy": 0.10190057530999183,
"num_tokens": 2415818.0,
"step": 1190
},
{
"entropy": 7.1862547397613525,
"epoch": 0.10443065629642576,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999967515559071,
"loss": 6.948,
"mean_token_accuracy": 0.10600498467683792,
"num_tokens": 2425072.0,
"step": 1195
},
{
"entropy": 7.127433347702026,
"epoch": 0.10486760464913047,
"grad_norm": 0.96875,
"learning_rate": 0.000499996581952961,
"loss": 7.0112,
"mean_token_accuracy": 0.10471701920032501,
"num_tokens": 2435055.0,
"step": 1200
},
{
"entropy": 7.183917045593262,
"epoch": 0.10530455300183518,
"grad_norm": 1.03125,
"learning_rate": 0.0004999964080344625,
"loss": 6.9105,
"mean_token_accuracy": 0.10828652903437615,
"num_tokens": 2444434.0,
"step": 1205
},
{
"entropy": 7.123368644714356,
"epoch": 0.1057415013545399,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999962298004151,
"loss": 6.9342,
"mean_token_accuracy": 0.10267825126647949,
"num_tokens": 2453860.0,
"step": 1210
},
{
"entropy": 7.156991624832154,
"epoch": 0.1061784497072446,
"grad_norm": 1.046875,
"learning_rate": 0.0004999960472508219,
"loss": 7.0422,
"mean_token_accuracy": 0.09686045348644257,
"num_tokens": 2464235.0,
"step": 1215
},
{
"entropy": 7.170539712905883,
"epoch": 0.10661539805994931,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999958603856869,
"loss": 6.9567,
"mean_token_accuracy": 0.1066193401813507,
"num_tokens": 2474561.0,
"step": 1220
},
{
"entropy": 7.07088451385498,
"epoch": 0.10705234641265403,
"grad_norm": 0.875,
"learning_rate": 0.0004999956692050132,
"loss": 6.9471,
"mean_token_accuracy": 0.10514729768037796,
"num_tokens": 2484574.0,
"step": 1225
},
{
"entropy": 7.207800436019897,
"epoch": 0.10748929476535873,
"grad_norm": 1.015625,
"learning_rate": 0.0004999954737088048,
"loss": 6.9074,
"mean_token_accuracy": 0.11164658814668656,
"num_tokens": 2494146.0,
"step": 1230
},
{
"entropy": 7.073509597778321,
"epoch": 0.10792624311806344,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999952738970652,
"loss": 6.9986,
"mean_token_accuracy": 0.10540383160114289,
"num_tokens": 2504157.0,
"step": 1235
},
{
"entropy": 7.107709217071533,
"epoch": 0.10836319147076816,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999950697697983,
"loss": 6.9445,
"mean_token_accuracy": 0.10252791419625282,
"num_tokens": 2512979.0,
"step": 1240
},
{
"entropy": 7.164508867263794,
"epoch": 0.10880013982347286,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999948613270081,
"loss": 6.9823,
"mean_token_accuracy": 0.09834418892860412,
"num_tokens": 2523202.0,
"step": 1245
},
{
"entropy": 7.125106716156006,
"epoch": 0.10923708817617758,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999946485686987,
"loss": 6.9302,
"mean_token_accuracy": 0.1016211062669754,
"num_tokens": 2532790.0,
"step": 1250
},
{
"entropy": 7.114097547531128,
"epoch": 0.10967403652888229,
"grad_norm": 0.984375,
"learning_rate": 0.000499994431494874,
"loss": 6.9191,
"mean_token_accuracy": 0.10513179227709771,
"num_tokens": 2542337.0,
"step": 1255
},
{
"entropy": 7.029775667190552,
"epoch": 0.11011098488158699,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999942101055381,
"loss": 6.8665,
"mean_token_accuracy": 0.11162453517317772,
"num_tokens": 2552762.0,
"step": 1260
},
{
"entropy": 7.129621982574463,
"epoch": 0.11054793323429171,
"grad_norm": 1.0,
"learning_rate": 0.0004999939844006955,
"loss": 6.9116,
"mean_token_accuracy": 0.10495460107922554,
"num_tokens": 2563261.0,
"step": 1265
},
{
"entropy": 7.162447929382324,
"epoch": 0.11098488158699642,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999937543803503,
"loss": 6.8825,
"mean_token_accuracy": 0.10844132751226425,
"num_tokens": 2573089.0,
"step": 1270
},
{
"entropy": 7.0640336036682125,
"epoch": 0.11142182993970112,
"grad_norm": 0.921875,
"learning_rate": 0.000499993520044507,
"loss": 6.9735,
"mean_token_accuracy": 0.10243528857827186,
"num_tokens": 2584358.0,
"step": 1275
},
{
"entropy": 7.16273775100708,
"epoch": 0.11185877829240584,
"grad_norm": 1.109375,
"learning_rate": 0.0004999932813931702,
"loss": 6.9331,
"mean_token_accuracy": 0.1072621412575245,
"num_tokens": 2594981.0,
"step": 1280
},
{
"entropy": 7.042669677734375,
"epoch": 0.11229572664511055,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999930384263441,
"loss": 6.938,
"mean_token_accuracy": 0.10397336781024932,
"num_tokens": 2603508.0,
"step": 1285
},
{
"entropy": 7.144971084594727,
"epoch": 0.11273267499781525,
"grad_norm": 1.0078125,
"learning_rate": 0.000499992791144034,
"loss": 6.9346,
"mean_token_accuracy": 0.10637803673744202,
"num_tokens": 2613666.0,
"step": 1290
},
{
"entropy": 7.241996955871582,
"epoch": 0.11316962335051997,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999925395462439,
"loss": 6.9627,
"mean_token_accuracy": 0.09948229417204857,
"num_tokens": 2624441.0,
"step": 1295
},
{
"entropy": 7.0352034091949465,
"epoch": 0.11360657170322468,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999922836329793,
"loss": 6.923,
"mean_token_accuracy": 0.10289243534207344,
"num_tokens": 2635261.0,
"step": 1300
},
{
"entropy": 7.011289453506469,
"epoch": 0.1140435200559294,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999920234042446,
"loss": 6.8854,
"mean_token_accuracy": 0.11003832370042801,
"num_tokens": 2645493.0,
"step": 1305
},
{
"entropy": 7.1208052158355715,
"epoch": 0.1144804684086341,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999917588600451,
"loss": 6.852,
"mean_token_accuracy": 0.11447785347700119,
"num_tokens": 2655430.0,
"step": 1310
},
{
"entropy": 7.098808240890503,
"epoch": 0.1149174167613388,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999914900003857,
"loss": 6.9128,
"mean_token_accuracy": 0.10278257057070732,
"num_tokens": 2665675.0,
"step": 1315
},
{
"entropy": 7.085298490524292,
"epoch": 0.11535436511404352,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999912168252717,
"loss": 6.8695,
"mean_token_accuracy": 0.10640941336750984,
"num_tokens": 2675962.0,
"step": 1320
},
{
"entropy": 7.03753924369812,
"epoch": 0.11579131346674823,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999909393347083,
"loss": 6.9073,
"mean_token_accuracy": 0.10800022035837173,
"num_tokens": 2687015.0,
"step": 1325
},
{
"entropy": 7.064095830917358,
"epoch": 0.11622826181945294,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999906575287006,
"loss": 6.9613,
"mean_token_accuracy": 0.10142027139663697,
"num_tokens": 2698065.0,
"step": 1330
},
{
"entropy": 7.07314624786377,
"epoch": 0.11666521017215765,
"grad_norm": 0.90625,
"learning_rate": 0.0004999903714072544,
"loss": 6.847,
"mean_token_accuracy": 0.11568356677889824,
"num_tokens": 2709046.0,
"step": 1335
},
{
"entropy": 7.1079401016235355,
"epoch": 0.11710215852486236,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999900809703749,
"loss": 6.9191,
"mean_token_accuracy": 0.10752326548099518,
"num_tokens": 2719705.0,
"step": 1340
},
{
"entropy": 7.064958286285401,
"epoch": 0.11753910687756706,
"grad_norm": 1.3984375,
"learning_rate": 0.0004999897862180678,
"loss": 6.8041,
"mean_token_accuracy": 0.11670546159148217,
"num_tokens": 2729556.0,
"step": 1345
},
{
"entropy": 6.963649272918701,
"epoch": 0.11797605523027178,
"grad_norm": 0.90625,
"learning_rate": 0.0004999894871503386,
"loss": 6.9142,
"mean_token_accuracy": 0.10688972845673561,
"num_tokens": 2741144.0,
"step": 1350
},
{
"entropy": 7.1297893047332765,
"epoch": 0.11841300358297649,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999891837671932,
"loss": 6.8313,
"mean_token_accuracy": 0.11042980030179024,
"num_tokens": 2751384.0,
"step": 1355
},
{
"entropy": 7.002114248275757,
"epoch": 0.11884995193568121,
"grad_norm": 1.046875,
"learning_rate": 0.0004999888760686375,
"loss": 6.8575,
"mean_token_accuracy": 0.10566901117563247,
"num_tokens": 2761044.0,
"step": 1360
},
{
"entropy": 6.988346195220947,
"epoch": 0.11928690028838591,
"grad_norm": 0.87109375,
"learning_rate": 0.000499988564054677,
"loss": 6.8086,
"mean_token_accuracy": 0.11115672811865807,
"num_tokens": 2770946.0,
"step": 1365
},
{
"entropy": 7.10092191696167,
"epoch": 0.11972384864109062,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999882477253182,
"loss": 6.8658,
"mean_token_accuracy": 0.10557007491588592,
"num_tokens": 2781247.0,
"step": 1370
},
{
"entropy": 6.943843507766724,
"epoch": 0.12016079699379534,
"grad_norm": 0.859375,
"learning_rate": 0.0004999879270805667,
"loss": 6.7182,
"mean_token_accuracy": 0.11588958874344826,
"num_tokens": 2792078.0,
"step": 1375
},
{
"entropy": 7.026294994354248,
"epoch": 0.12059774534650004,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999876021204291,
"loss": 6.8013,
"mean_token_accuracy": 0.10884896218776703,
"num_tokens": 2802458.0,
"step": 1380
},
{
"entropy": 6.9799354553222654,
"epoch": 0.12103469369920475,
"grad_norm": 0.984375,
"learning_rate": 0.0004999872728449112,
"loss": 6.9463,
"mean_token_accuracy": 0.10275301113724708,
"num_tokens": 2813352.0,
"step": 1385
},
{
"entropy": 7.118425130844116,
"epoch": 0.12147164205190947,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999869392540195,
"loss": 6.8378,
"mean_token_accuracy": 0.10523320958018303,
"num_tokens": 2822954.0,
"step": 1390
},
{
"entropy": 7.004513549804687,
"epoch": 0.12190859040461417,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999866013477605,
"loss": 6.8669,
"mean_token_accuracy": 0.10665129348635674,
"num_tokens": 2833375.0,
"step": 1395
},
{
"entropy": 7.019338464736938,
"epoch": 0.12234553875731889,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999862591261407,
"loss": 6.7998,
"mean_token_accuracy": 0.10706322714686393,
"num_tokens": 2843103.0,
"step": 1400
},
{
"entropy": 7.052399921417236,
"epoch": 0.1227824871100236,
"grad_norm": 0.890625,
"learning_rate": 0.0004999859125891664,
"loss": 6.8627,
"mean_token_accuracy": 0.10739990398287773,
"num_tokens": 2853801.0,
"step": 1405
},
{
"entropy": 7.02172212600708,
"epoch": 0.1232194354627283,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999855617368445,
"loss": 6.8554,
"mean_token_accuracy": 0.10320612639188767,
"num_tokens": 2863851.0,
"step": 1410
},
{
"entropy": 6.942949008941651,
"epoch": 0.12365638381543302,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999852065691816,
"loss": 6.7966,
"mean_token_accuracy": 0.11276845708489418,
"num_tokens": 2875073.0,
"step": 1415
},
{
"entropy": 7.042190361022949,
"epoch": 0.12409333216813773,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999848470861846,
"loss": 6.8247,
"mean_token_accuracy": 0.11047491133213043,
"num_tokens": 2885676.0,
"step": 1420
},
{
"entropy": 7.036431312561035,
"epoch": 0.12453028052084243,
"grad_norm": 1.09375,
"learning_rate": 0.0004999844832878602,
"loss": 6.835,
"mean_token_accuracy": 0.11439931541681289,
"num_tokens": 2896663.0,
"step": 1425
},
{
"entropy": 6.9360435009002686,
"epoch": 0.12496722887354715,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999841151742156,
"loss": 6.8149,
"mean_token_accuracy": 0.1112043671309948,
"num_tokens": 2907555.0,
"step": 1430
},
{
"entropy": 7.092131185531616,
"epoch": 0.12540417722625186,
"grad_norm": 1.0,
"learning_rate": 0.0004999837427452578,
"loss": 6.8264,
"mean_token_accuracy": 0.11113670319318772,
"num_tokens": 2917173.0,
"step": 1435
},
{
"entropy": 6.94360384941101,
"epoch": 0.12584112557895658,
"grad_norm": 0.9453125,
"learning_rate": 0.000499983366000994,
"loss": 6.8074,
"mean_token_accuracy": 0.10906619876623154,
"num_tokens": 2928268.0,
"step": 1440
},
{
"entropy": 7.049355125427246,
"epoch": 0.12627807393166127,
"grad_norm": 1.015625,
"learning_rate": 0.0004999829849414312,
"loss": 6.7696,
"mean_token_accuracy": 0.11551543027162552,
"num_tokens": 2938203.0,
"step": 1445
},
{
"entropy": 6.926217317581177,
"epoch": 0.12671502228436599,
"grad_norm": 0.9765625,
"learning_rate": 0.000499982599566577,
"loss": 6.8056,
"mean_token_accuracy": 0.11572478860616683,
"num_tokens": 2948094.0,
"step": 1450
},
{
"entropy": 7.031111526489258,
"epoch": 0.1271519706370707,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999822098764385,
"loss": 6.8032,
"mean_token_accuracy": 0.10698436871170998,
"num_tokens": 2957865.0,
"step": 1455
},
{
"entropy": 6.903406238555908,
"epoch": 0.1275889189897754,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999818158710234,
"loss": 6.7279,
"mean_token_accuracy": 0.11594445258378983,
"num_tokens": 2967880.0,
"step": 1460
},
{
"entropy": 6.977146100997925,
"epoch": 0.12802586734248012,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999814175503392,
"loss": 6.7898,
"mean_token_accuracy": 0.11242105588316917,
"num_tokens": 2979537.0,
"step": 1465
},
{
"entropy": 6.962574195861817,
"epoch": 0.12846281569518483,
"grad_norm": 1.015625,
"learning_rate": 0.0004999810149143935,
"loss": 6.7715,
"mean_token_accuracy": 0.10995072573423385,
"num_tokens": 2989700.0,
"step": 1470
},
{
"entropy": 6.879771089553833,
"epoch": 0.12889976404788953,
"grad_norm": 0.9765625,
"learning_rate": 0.000499980607963194,
"loss": 6.7658,
"mean_token_accuracy": 0.11039941981434823,
"num_tokens": 2999141.0,
"step": 1475
},
{
"entropy": 6.992912673950196,
"epoch": 0.12933671240059424,
"grad_norm": 1.046875,
"learning_rate": 0.0004999801966967486,
"loss": 6.7351,
"mean_token_accuracy": 0.11658515930175781,
"num_tokens": 3008259.0,
"step": 1480
},
{
"entropy": 7.007751703262329,
"epoch": 0.12977366075329896,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999797811150652,
"loss": 6.771,
"mean_token_accuracy": 0.10813285559415817,
"num_tokens": 3018281.0,
"step": 1485
},
{
"entropy": 6.9875898361206055,
"epoch": 0.13021060910600368,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999793612181516,
"loss": 6.7811,
"mean_token_accuracy": 0.11372531205415726,
"num_tokens": 3029062.0,
"step": 1490
},
{
"entropy": 6.8786805152893065,
"epoch": 0.13064755745870837,
"grad_norm": 1.046875,
"learning_rate": 0.0004999789370060161,
"loss": 6.7488,
"mean_token_accuracy": 0.11443367153406143,
"num_tokens": 3038142.0,
"step": 1495
},
{
"entropy": 7.021120643615722,
"epoch": 0.1310845058114131,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999785084786667,
"loss": 6.845,
"mean_token_accuracy": 0.1017142303287983,
"num_tokens": 3049123.0,
"step": 1500
},
{
"entropy": 6.958976125717163,
"epoch": 0.1315214541641178,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999780756361114,
"loss": 6.7553,
"mean_token_accuracy": 0.10809942036867141,
"num_tokens": 3059031.0,
"step": 1505
},
{
"entropy": 6.929660081863403,
"epoch": 0.1319584025168225,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999776384783589,
"loss": 6.787,
"mean_token_accuracy": 0.11125006899237633,
"num_tokens": 3068830.0,
"step": 1510
},
{
"entropy": 6.878285837173462,
"epoch": 0.13239535086952722,
"grad_norm": 0.7578125,
"learning_rate": 0.0004999771970054174,
"loss": 6.763,
"mean_token_accuracy": 0.11348566561937332,
"num_tokens": 3080124.0,
"step": 1515
},
{
"entropy": 7.005362796783447,
"epoch": 0.13283229922223194,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999767512172953,
"loss": 6.771,
"mean_token_accuracy": 0.10827827528119087,
"num_tokens": 3089607.0,
"step": 1520
},
{
"entropy": 6.95645546913147,
"epoch": 0.13326924757493663,
"grad_norm": 0.890625,
"learning_rate": 0.0004999763011140013,
"loss": 6.8462,
"mean_token_accuracy": 0.10264672636985779,
"num_tokens": 3099363.0,
"step": 1525
},
{
"entropy": 7.003346729278564,
"epoch": 0.13370619592764135,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999758466955439,
"loss": 6.7483,
"mean_token_accuracy": 0.11257248520851135,
"num_tokens": 3109552.0,
"step": 1530
},
{
"entropy": 6.890552520751953,
"epoch": 0.13414314428034607,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999753879619319,
"loss": 6.7293,
"mean_token_accuracy": 0.11494478285312652,
"num_tokens": 3119536.0,
"step": 1535
},
{
"entropy": 6.977446985244751,
"epoch": 0.13458009263305076,
"grad_norm": 0.76953125,
"learning_rate": 0.0004999749249131741,
"loss": 6.7719,
"mean_token_accuracy": 0.11656006649136544,
"num_tokens": 3130552.0,
"step": 1540
},
{
"entropy": 6.943555307388306,
"epoch": 0.13501704098575548,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999744575492793,
"loss": 6.663,
"mean_token_accuracy": 0.11371774896979332,
"num_tokens": 3140723.0,
"step": 1545
},
{
"entropy": 6.942564630508423,
"epoch": 0.1354539893384602,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999739858702567,
"loss": 6.7176,
"mean_token_accuracy": 0.12080485373735428,
"num_tokens": 3151031.0,
"step": 1550
},
{
"entropy": 6.856326866149902,
"epoch": 0.1358909376911649,
"grad_norm": 0.87890625,
"learning_rate": 0.000499973509876115,
"loss": 6.7141,
"mean_token_accuracy": 0.11423594728112221,
"num_tokens": 3161373.0,
"step": 1555
},
{
"entropy": 6.9513795375823975,
"epoch": 0.1363278860438696,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999730295668634,
"loss": 6.7394,
"mean_token_accuracy": 0.11483867838978767,
"num_tokens": 3171094.0,
"step": 1560
},
{
"entropy": 6.879623651504517,
"epoch": 0.13676483439657433,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999725449425114,
"loss": 6.6879,
"mean_token_accuracy": 0.11481415182352066,
"num_tokens": 3181646.0,
"step": 1565
},
{
"entropy": 6.871405601501465,
"epoch": 0.13720178274927902,
"grad_norm": 0.9765625,
"learning_rate": 0.000499972056003068,
"loss": 6.6551,
"mean_token_accuracy": 0.11951215118169785,
"num_tokens": 3190727.0,
"step": 1570
},
{
"entropy": 6.925275754928589,
"epoch": 0.13763873110198374,
"grad_norm": 0.96875,
"learning_rate": 0.0004999715627485426,
"loss": 6.761,
"mean_token_accuracy": 0.10644741803407669,
"num_tokens": 3200316.0,
"step": 1575
},
{
"entropy": 6.952586555480957,
"epoch": 0.13807567945468846,
"grad_norm": 0.90625,
"learning_rate": 0.000499971065178945,
"loss": 6.7275,
"mean_token_accuracy": 0.10815547704696656,
"num_tokens": 3211830.0,
"step": 1580
},
{
"entropy": 6.832159805297851,
"epoch": 0.13851262780739318,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999705632942843,
"loss": 6.739,
"mean_token_accuracy": 0.11103120297193528,
"num_tokens": 3222711.0,
"step": 1585
},
{
"entropy": 6.977508449554444,
"epoch": 0.13894957616009787,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999700570945704,
"loss": 6.657,
"mean_token_accuracy": 0.12132718712091446,
"num_tokens": 3232810.0,
"step": 1590
},
{
"entropy": 6.748742961883545,
"epoch": 0.1393865245128026,
"grad_norm": 0.984375,
"learning_rate": 0.0004999695465798128,
"loss": 6.722,
"mean_token_accuracy": 0.11645966917276382,
"num_tokens": 3243366.0,
"step": 1595
},
{
"entropy": 7.011563968658447,
"epoch": 0.1398234728655073,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999690317500215,
"loss": 6.7095,
"mean_token_accuracy": 0.11738921105861663,
"num_tokens": 3254000.0,
"step": 1600
},
{
"entropy": 6.944188117980957,
"epoch": 0.140260421218212,
"grad_norm": 1.03125,
"learning_rate": 0.0004999685126052063,
"loss": 6.7427,
"mean_token_accuracy": 0.11331361308693885,
"num_tokens": 3263601.0,
"step": 1605
},
{
"entropy": 6.870913124084472,
"epoch": 0.14069736957091672,
"grad_norm": 0.921875,
"learning_rate": 0.0004999679891453772,
"loss": 6.7751,
"mean_token_accuracy": 0.10785396099090576,
"num_tokens": 3273243.0,
"step": 1610
},
{
"entropy": 6.945408201217651,
"epoch": 0.14113431792362144,
"grad_norm": 0.80078125,
"learning_rate": 0.000499967461370544,
"loss": 6.7204,
"mean_token_accuracy": 0.11248369067907334,
"num_tokens": 3284208.0,
"step": 1615
},
{
"entropy": 6.906504917144775,
"epoch": 0.14157126627632613,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999669292807172,
"loss": 6.6823,
"mean_token_accuracy": 0.11680415868759156,
"num_tokens": 3294216.0,
"step": 1620
},
{
"entropy": 6.915546798706055,
"epoch": 0.14200821462903085,
"grad_norm": 0.953125,
"learning_rate": 0.0004999663928759066,
"loss": 6.7472,
"mean_token_accuracy": 0.11192744225263596,
"num_tokens": 3304576.0,
"step": 1625
},
{
"entropy": 6.897689294815064,
"epoch": 0.14244516298173557,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999658521561229,
"loss": 6.7145,
"mean_token_accuracy": 0.10784238651394844,
"num_tokens": 3314214.0,
"step": 1630
},
{
"entropy": 6.929960298538208,
"epoch": 0.14288211133444026,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999653071213763,
"loss": 6.7304,
"mean_token_accuracy": 0.10703980401158333,
"num_tokens": 3325527.0,
"step": 1635
},
{
"entropy": 6.783003330230713,
"epoch": 0.14331905968714498,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999647577716771,
"loss": 6.6364,
"mean_token_accuracy": 0.11789060086011886,
"num_tokens": 3334511.0,
"step": 1640
},
{
"entropy": 6.935398960113526,
"epoch": 0.1437560080398497,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999642041070361,
"loss": 6.6406,
"mean_token_accuracy": 0.12125914841890335,
"num_tokens": 3343922.0,
"step": 1645
},
{
"entropy": 6.825817012786866,
"epoch": 0.1441929563925544,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999636461274637,
"loss": 6.6609,
"mean_token_accuracy": 0.11183852404356003,
"num_tokens": 3353895.0,
"step": 1650
},
{
"entropy": 6.892046880722046,
"epoch": 0.1446299047452591,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999630838329706,
"loss": 6.712,
"mean_token_accuracy": 0.11429799050092697,
"num_tokens": 3364637.0,
"step": 1655
},
{
"entropy": 6.898443794250488,
"epoch": 0.14506685309796383,
"grad_norm": 0.984375,
"learning_rate": 0.0004999625172235679,
"loss": 6.6721,
"mean_token_accuracy": 0.11531461402773857,
"num_tokens": 3374308.0,
"step": 1660
},
{
"entropy": 6.860412073135376,
"epoch": 0.14550380145066852,
"grad_norm": 0.91796875,
"learning_rate": 0.000499961946299266,
"loss": 6.6627,
"mean_token_accuracy": 0.11255936473608016,
"num_tokens": 3383637.0,
"step": 1665
},
{
"entropy": 6.8619140625,
"epoch": 0.14594074980337324,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999613710600763,
"loss": 6.7309,
"mean_token_accuracy": 0.10881676152348518,
"num_tokens": 3394150.0,
"step": 1670
},
{
"entropy": 6.886914920806885,
"epoch": 0.14637769815607796,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999607915060094,
"loss": 6.6871,
"mean_token_accuracy": 0.11321992725133896,
"num_tokens": 3404445.0,
"step": 1675
},
{
"entropy": 6.786718940734863,
"epoch": 0.14681464650878268,
"grad_norm": 0.796875,
"learning_rate": 0.0004999602076370768,
"loss": 6.6002,
"mean_token_accuracy": 0.12198357656598091,
"num_tokens": 3415183.0,
"step": 1680
},
{
"entropy": 6.8183753490448,
"epoch": 0.14725159486148737,
"grad_norm": 0.9375,
"learning_rate": 0.0004999596194532895,
"loss": 6.5695,
"mean_token_accuracy": 0.1226851724088192,
"num_tokens": 3424824.0,
"step": 1685
},
{
"entropy": 6.875542020797729,
"epoch": 0.1476885432141921,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999590269546588,
"loss": 6.6749,
"mean_token_accuracy": 0.11421856135129929,
"num_tokens": 3434789.0,
"step": 1690
},
{
"entropy": 6.737638378143311,
"epoch": 0.1481254915668968,
"grad_norm": 0.875,
"learning_rate": 0.0004999584301411961,
"loss": 6.5652,
"mean_token_accuracy": 0.1369031101465225,
"num_tokens": 3445810.0,
"step": 1695
},
{
"entropy": 6.955615425109864,
"epoch": 0.1485624399196015,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999578290129129,
"loss": 6.7108,
"mean_token_accuracy": 0.11079858317971229,
"num_tokens": 3455510.0,
"step": 1700
},
{
"entropy": 6.795374155044556,
"epoch": 0.14899938827230622,
"grad_norm": 0.859375,
"learning_rate": 0.0004999572235698206,
"loss": 6.6809,
"mean_token_accuracy": 0.11354952603578568,
"num_tokens": 3465785.0,
"step": 1705
},
{
"entropy": 6.860705327987671,
"epoch": 0.14943633662501093,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999566138119308,
"loss": 6.7346,
"mean_token_accuracy": 0.10918433368206024,
"num_tokens": 3477568.0,
"step": 1710
},
{
"entropy": 6.796298360824585,
"epoch": 0.14987328497771563,
"grad_norm": 0.9375,
"learning_rate": 0.0004999559997392554,
"loss": 6.5705,
"mean_token_accuracy": 0.13033040314912797,
"num_tokens": 3487703.0,
"step": 1715
},
{
"entropy": 6.803538990020752,
"epoch": 0.15031023333042035,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999553813518061,
"loss": 6.6467,
"mean_token_accuracy": 0.11564805582165719,
"num_tokens": 3497757.0,
"step": 1720
},
{
"entropy": 6.815519523620606,
"epoch": 0.15074718168312506,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999547586495946,
"loss": 6.6123,
"mean_token_accuracy": 0.11909747421741486,
"num_tokens": 3507594.0,
"step": 1725
},
{
"entropy": 6.858670568466186,
"epoch": 0.15118413003582976,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999541316326329,
"loss": 6.6982,
"mean_token_accuracy": 0.11421534195542335,
"num_tokens": 3517889.0,
"step": 1730
},
{
"entropy": 6.789118671417237,
"epoch": 0.15162107838853447,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999535003009332,
"loss": 6.6842,
"mean_token_accuracy": 0.11443898230791091,
"num_tokens": 3528158.0,
"step": 1735
},
{
"entropy": 6.85327639579773,
"epoch": 0.1520580267412392,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999528646545074,
"loss": 6.6617,
"mean_token_accuracy": 0.11684061735868453,
"num_tokens": 3539099.0,
"step": 1740
},
{
"entropy": 6.868143224716187,
"epoch": 0.15249497509394389,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999522246933678,
"loss": 6.6744,
"mean_token_accuracy": 0.11591163650155067,
"num_tokens": 3549819.0,
"step": 1745
},
{
"entropy": 6.924664878845215,
"epoch": 0.1529319234466486,
"grad_norm": 1.03125,
"learning_rate": 0.0004999515804175268,
"loss": 6.6708,
"mean_token_accuracy": 0.1160687893629074,
"num_tokens": 3560022.0,
"step": 1750
},
{
"entropy": 6.796440887451172,
"epoch": 0.15336887179935332,
"grad_norm": 0.921875,
"learning_rate": 0.0004999509318269964,
"loss": 6.6229,
"mean_token_accuracy": 0.11483302041888237,
"num_tokens": 3569681.0,
"step": 1755
},
{
"entropy": 6.931779527664185,
"epoch": 0.15380582015205801,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999502789217894,
"loss": 6.7187,
"mean_token_accuracy": 0.11464278921484947,
"num_tokens": 3580731.0,
"step": 1760
},
{
"entropy": 6.735879182815552,
"epoch": 0.15424276850476273,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999496217019181,
"loss": 6.6771,
"mean_token_accuracy": 0.12017816901206971,
"num_tokens": 3590482.0,
"step": 1765
},
{
"entropy": 6.846047735214233,
"epoch": 0.15467971685746745,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999489601673952,
"loss": 6.6588,
"mean_token_accuracy": 0.11184544190764427,
"num_tokens": 3600988.0,
"step": 1770
},
{
"entropy": 6.966546869277954,
"epoch": 0.15511666521017214,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999482943182334,
"loss": 6.7207,
"mean_token_accuracy": 0.11034572273492813,
"num_tokens": 3611226.0,
"step": 1775
},
{
"entropy": 6.7466967582702635,
"epoch": 0.15555361356287686,
"grad_norm": 0.890625,
"learning_rate": 0.0004999476241544454,
"loss": 6.6406,
"mean_token_accuracy": 0.11693034246563912,
"num_tokens": 3621324.0,
"step": 1780
},
{
"entropy": 6.859796237945557,
"epoch": 0.15599056191558158,
"grad_norm": 0.96875,
"learning_rate": 0.0004999469496760441,
"loss": 6.5745,
"mean_token_accuracy": 0.11757055521011353,
"num_tokens": 3631650.0,
"step": 1785
},
{
"entropy": 6.793846988677979,
"epoch": 0.1564275102682863,
"grad_norm": 0.921875,
"learning_rate": 0.0004999462708830425,
"loss": 6.5964,
"mean_token_accuracy": 0.13014772459864615,
"num_tokens": 3641173.0,
"step": 1790
},
{
"entropy": 6.7543669700622555,
"epoch": 0.156864458620991,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999455877754535,
"loss": 6.6099,
"mean_token_accuracy": 0.11783762127161027,
"num_tokens": 3651846.0,
"step": 1795
},
{
"entropy": 6.839160680770874,
"epoch": 0.1573014069736957,
"grad_norm": 0.921875,
"learning_rate": 0.0004999449003532903,
"loss": 6.6181,
"mean_token_accuracy": 0.11612390056252479,
"num_tokens": 3661174.0,
"step": 1800
},
{
"entropy": 6.767947387695313,
"epoch": 0.15773835532640043,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999442086165661,
"loss": 6.5785,
"mean_token_accuracy": 0.11563654839992524,
"num_tokens": 3671042.0,
"step": 1805
},
{
"entropy": 6.836207246780395,
"epoch": 0.15817530367910512,
"grad_norm": 0.89453125,
"learning_rate": 0.000499943512565294,
"loss": 6.5404,
"mean_token_accuracy": 0.11653607487678527,
"num_tokens": 3681206.0,
"step": 1810
},
{
"entropy": 6.777206325531006,
"epoch": 0.15861225203180984,
"grad_norm": 0.921875,
"learning_rate": 0.0004999428121994874,
"loss": 6.6427,
"mean_token_accuracy": 0.11898047327995301,
"num_tokens": 3691748.0,
"step": 1815
},
{
"entropy": 6.702393054962158,
"epoch": 0.15904920038451456,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999421075191599,
"loss": 6.5483,
"mean_token_accuracy": 0.12115112468600273,
"num_tokens": 3701730.0,
"step": 1820
},
{
"entropy": 6.887295866012574,
"epoch": 0.15948614873721925,
"grad_norm": 1.125,
"learning_rate": 0.0004999413985243249,
"loss": 6.6747,
"mean_token_accuracy": 0.1129259705543518,
"num_tokens": 3712525.0,
"step": 1825
},
{
"entropy": 6.863398599624634,
"epoch": 0.15992309708992397,
"grad_norm": 0.95703125,
"learning_rate": 0.000499940685214996,
"loss": 6.6086,
"mean_token_accuracy": 0.11575344502925873,
"num_tokens": 3722016.0,
"step": 1830
},
{
"entropy": 6.761053848266601,
"epoch": 0.1603600454426287,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999399675911869,
"loss": 6.5318,
"mean_token_accuracy": 0.12198757007718086,
"num_tokens": 3731299.0,
"step": 1835
},
{
"entropy": 6.743390274047852,
"epoch": 0.16079699379533338,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999392456529113,
"loss": 6.5806,
"mean_token_accuracy": 0.12458991408348083,
"num_tokens": 3740922.0,
"step": 1840
},
{
"entropy": 6.758446550369262,
"epoch": 0.1612339421480381,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999385194001831,
"loss": 6.5937,
"mean_token_accuracy": 0.11810147613286973,
"num_tokens": 3750339.0,
"step": 1845
},
{
"entropy": 6.787387943267822,
"epoch": 0.16167089050074282,
"grad_norm": 0.953125,
"learning_rate": 0.0004999377888330164,
"loss": 6.5467,
"mean_token_accuracy": 0.12232641279697418,
"num_tokens": 3761379.0,
"step": 1850
},
{
"entropy": 6.749542856216431,
"epoch": 0.1621078388534475,
"grad_norm": 0.90625,
"learning_rate": 0.0004999370539514249,
"loss": 6.6294,
"mean_token_accuracy": 0.11457357108592987,
"num_tokens": 3771649.0,
"step": 1855
},
{
"entropy": 6.757665014266967,
"epoch": 0.16254478720615223,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999363147554229,
"loss": 6.5497,
"mean_token_accuracy": 0.12577750012278557,
"num_tokens": 3782393.0,
"step": 1860
},
{
"entropy": 6.820600986480713,
"epoch": 0.16298173555885695,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999355712450244,
"loss": 6.5986,
"mean_token_accuracy": 0.12362995967268944,
"num_tokens": 3792849.0,
"step": 1865
},
{
"entropy": 6.747550010681152,
"epoch": 0.16341868391156164,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999348234202438,
"loss": 6.5515,
"mean_token_accuracy": 0.12130374014377594,
"num_tokens": 3803848.0,
"step": 1870
},
{
"entropy": 6.627656269073486,
"epoch": 0.16385563226426636,
"grad_norm": 0.90625,
"learning_rate": 0.0004999340712810956,
"loss": 6.4677,
"mean_token_accuracy": 0.13533140793442727,
"num_tokens": 3814704.0,
"step": 1875
},
{
"entropy": 6.747098636627197,
"epoch": 0.16429258061697108,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999333148275939,
"loss": 6.6244,
"mean_token_accuracy": 0.11492595672607422,
"num_tokens": 3824955.0,
"step": 1880
},
{
"entropy": 6.810025024414062,
"epoch": 0.1647295289696758,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999325540597535,
"loss": 6.5992,
"mean_token_accuracy": 0.12268827185034752,
"num_tokens": 3836540.0,
"step": 1885
},
{
"entropy": 6.680448246002197,
"epoch": 0.1651664773223805,
"grad_norm": 0.921875,
"learning_rate": 0.0004999317889775887,
"loss": 6.4789,
"mean_token_accuracy": 0.1307223454117775,
"num_tokens": 3846472.0,
"step": 1890
},
{
"entropy": 6.797214794158935,
"epoch": 0.1656034256750852,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999310195811143,
"loss": 6.6013,
"mean_token_accuracy": 0.11527118012309075,
"num_tokens": 3857187.0,
"step": 1895
},
{
"entropy": 6.796212434768677,
"epoch": 0.16604037402778993,
"grad_norm": 0.9375,
"learning_rate": 0.0004999302458703451,
"loss": 6.5513,
"mean_token_accuracy": 0.12751033753156663,
"num_tokens": 3866259.0,
"step": 1900
},
{
"entropy": 6.679151391983032,
"epoch": 0.16647732238049462,
"grad_norm": 0.953125,
"learning_rate": 0.000499929467845296,
"loss": 6.6161,
"mean_token_accuracy": 0.11295140087604523,
"num_tokens": 3876869.0,
"step": 1905
},
{
"entropy": 6.8329041481018065,
"epoch": 0.16691427073319934,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999286855059818,
"loss": 6.59,
"mean_token_accuracy": 0.12256261631846428,
"num_tokens": 3888278.0,
"step": 1910
},
{
"entropy": 6.806984758377075,
"epoch": 0.16735121908590406,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999278988524177,
"loss": 6.6406,
"mean_token_accuracy": 0.11997235789895058,
"num_tokens": 3898648.0,
"step": 1915
},
{
"entropy": 6.776583909988403,
"epoch": 0.16778816743860875,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999271078846184,
"loss": 6.573,
"mean_token_accuracy": 0.11964118704199792,
"num_tokens": 3908282.0,
"step": 1920
},
{
"entropy": 6.679054212570191,
"epoch": 0.16822511579131347,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999263126025994,
"loss": 6.5397,
"mean_token_accuracy": 0.12275912463665009,
"num_tokens": 3918333.0,
"step": 1925
},
{
"entropy": 6.806920671463013,
"epoch": 0.1686620641440182,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999255130063759,
"loss": 6.5461,
"mean_token_accuracy": 0.12174615561962128,
"num_tokens": 3927403.0,
"step": 1930
},
{
"entropy": 6.764342546463013,
"epoch": 0.16909901249672288,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999247090959633,
"loss": 6.5812,
"mean_token_accuracy": 0.12118082121014595,
"num_tokens": 3937352.0,
"step": 1935
},
{
"entropy": 6.679003667831421,
"epoch": 0.1695359608494276,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999239008713767,
"loss": 6.4088,
"mean_token_accuracy": 0.12937364503741264,
"num_tokens": 3947164.0,
"step": 1940
},
{
"entropy": 6.770146179199219,
"epoch": 0.16997290920213232,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999230883326319,
"loss": 6.5737,
"mean_token_accuracy": 0.1233607918024063,
"num_tokens": 3956887.0,
"step": 1945
},
{
"entropy": 6.735566186904907,
"epoch": 0.170409857554837,
"grad_norm": 0.875,
"learning_rate": 0.0004999222714797445,
"loss": 6.56,
"mean_token_accuracy": 0.11969926655292511,
"num_tokens": 3966542.0,
"step": 1950
},
{
"entropy": 6.785258102416992,
"epoch": 0.17084680590754173,
"grad_norm": 0.90625,
"learning_rate": 0.00049992145031273,
"loss": 6.5065,
"mean_token_accuracy": 0.1332083135843277,
"num_tokens": 3976127.0,
"step": 1955
},
{
"entropy": 6.6167535305023195,
"epoch": 0.17128375426024645,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999206248316043,
"loss": 6.4685,
"mean_token_accuracy": 0.12284221649169921,
"num_tokens": 3987371.0,
"step": 1960
},
{
"entropy": 6.667543745040893,
"epoch": 0.17172070261295114,
"grad_norm": 0.921875,
"learning_rate": 0.000499919795036383,
"loss": 6.4859,
"mean_token_accuracy": 0.1272851638495922,
"num_tokens": 3997924.0,
"step": 1965
},
{
"entropy": 6.740949630737305,
"epoch": 0.17215765096565586,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999189609270823,
"loss": 6.5751,
"mean_token_accuracy": 0.12322761192917824,
"num_tokens": 4009150.0,
"step": 1970
},
{
"entropy": 6.729563045501709,
"epoch": 0.17259459931836058,
"grad_norm": 0.8671875,
"learning_rate": 0.000499918122503718,
"loss": 6.5836,
"mean_token_accuracy": 0.12200360670685768,
"num_tokens": 4018899.0,
"step": 1975
},
{
"entropy": 6.730152893066406,
"epoch": 0.17303154767106527,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999172797663064,
"loss": 6.5475,
"mean_token_accuracy": 0.12035959288477897,
"num_tokens": 4028652.0,
"step": 1980
},
{
"entropy": 6.713821458816528,
"epoch": 0.17346849602376999,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999164327148634,
"loss": 6.5693,
"mean_token_accuracy": 0.12123406901955605,
"num_tokens": 4039819.0,
"step": 1985
},
{
"entropy": 6.741577911376953,
"epoch": 0.1739054443764747,
"grad_norm": 0.921875,
"learning_rate": 0.0004999155813494053,
"loss": 6.4755,
"mean_token_accuracy": 0.1275830537080765,
"num_tokens": 4050029.0,
"step": 1990
},
{
"entropy": 6.6802998065948485,
"epoch": 0.17434239272917942,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999147256699487,
"loss": 6.5211,
"mean_token_accuracy": 0.12550380676984788,
"num_tokens": 4060497.0,
"step": 1995
},
{
"entropy": 6.783202648162842,
"epoch": 0.17477934108188412,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999138656765096,
"loss": 6.4997,
"mean_token_accuracy": 0.12641414254903793,
"num_tokens": 4071477.0,
"step": 2000
},
{
"entropy": 6.732838535308838,
"epoch": 0.17521628943458883,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999130013691049,
"loss": 6.584,
"mean_token_accuracy": 0.11925829127430916,
"num_tokens": 4081869.0,
"step": 2005
},
{
"entropy": 6.760075616836548,
"epoch": 0.17565323778729355,
"grad_norm": 0.890625,
"learning_rate": 0.0004999121327477509,
"loss": 6.5618,
"mean_token_accuracy": 0.1242613822221756,
"num_tokens": 4092797.0,
"step": 2010
},
{
"entropy": 6.757962799072265,
"epoch": 0.17609018613999824,
"grad_norm": 1.015625,
"learning_rate": 0.0004999112598124642,
"loss": 6.5178,
"mean_token_accuracy": 0.12427477762103081,
"num_tokens": 4103688.0,
"step": 2015
},
{
"entropy": 6.619799852371216,
"epoch": 0.17652713449270296,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999103825632618,
"loss": 6.4814,
"mean_token_accuracy": 0.12500133290886878,
"num_tokens": 4114037.0,
"step": 2020
},
{
"entropy": 6.713038921356201,
"epoch": 0.17696408284540768,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999095010001605,
"loss": 6.4755,
"mean_token_accuracy": 0.11943754628300667,
"num_tokens": 4124164.0,
"step": 2025
},
{
"entropy": 6.727624893188477,
"epoch": 0.17740103119811237,
"grad_norm": 0.8984375,
"learning_rate": 0.000499908615123177,
"loss": 6.5808,
"mean_token_accuracy": 0.1142855279147625,
"num_tokens": 4134225.0,
"step": 2030
},
{
"entropy": 6.74792366027832,
"epoch": 0.1778379795508171,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999077249323284,
"loss": 6.6281,
"mean_token_accuracy": 0.12126999646425247,
"num_tokens": 4145962.0,
"step": 2035
},
{
"entropy": 6.74753704071045,
"epoch": 0.1782749279035218,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999068304276319,
"loss": 6.5234,
"mean_token_accuracy": 0.12188527584075928,
"num_tokens": 4156572.0,
"step": 2040
},
{
"entropy": 6.6771139144897464,
"epoch": 0.1787118762562265,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999059316091045,
"loss": 6.5804,
"mean_token_accuracy": 0.11935446932911872,
"num_tokens": 4166960.0,
"step": 2045
},
{
"entropy": 6.767083215713501,
"epoch": 0.17914882460893122,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999050284767634,
"loss": 6.5431,
"mean_token_accuracy": 0.12220514565706253,
"num_tokens": 4177833.0,
"step": 2050
},
{
"entropy": 6.569900465011597,
"epoch": 0.17958577296163594,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999041210306261,
"loss": 6.4581,
"mean_token_accuracy": 0.12861283570528032,
"num_tokens": 4187858.0,
"step": 2055
},
{
"entropy": 6.775664949417115,
"epoch": 0.18002272131434063,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999032092707098,
"loss": 6.5951,
"mean_token_accuracy": 0.11695825308561325,
"num_tokens": 4198207.0,
"step": 2060
},
{
"entropy": 6.746979188919068,
"epoch": 0.18045966966704535,
"grad_norm": 1.0,
"learning_rate": 0.0004999022931970323,
"loss": 6.4872,
"mean_token_accuracy": 0.12345504090189933,
"num_tokens": 4207594.0,
"step": 2065
},
{
"entropy": 6.626251983642578,
"epoch": 0.18089661801975007,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999013728096107,
"loss": 6.4911,
"mean_token_accuracy": 0.12829106748104097,
"num_tokens": 4217334.0,
"step": 2070
},
{
"entropy": 6.721229362487793,
"epoch": 0.18133356637245476,
"grad_norm": 0.921875,
"learning_rate": 0.000499900448108463,
"loss": 6.4728,
"mean_token_accuracy": 0.1287317655980587,
"num_tokens": 4226407.0,
"step": 2075
},
{
"entropy": 6.631599426269531,
"epoch": 0.18177051472515948,
"grad_norm": 0.91796875,
"learning_rate": 0.000499899519093607,
"loss": 6.4147,
"mean_token_accuracy": 0.1277851089835167,
"num_tokens": 4237818.0,
"step": 2080
},
{
"entropy": 6.699046468734741,
"epoch": 0.1822074630778642,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998985857650603,
"loss": 6.5392,
"mean_token_accuracy": 0.11802871823310852,
"num_tokens": 4247783.0,
"step": 2085
},
{
"entropy": 6.62796630859375,
"epoch": 0.18264441143056892,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998976481228409,
"loss": 6.4955,
"mean_token_accuracy": 0.12168094217777252,
"num_tokens": 4258049.0,
"step": 2090
},
{
"entropy": 6.694241905212403,
"epoch": 0.1830813597832736,
"grad_norm": 0.82421875,
"learning_rate": 0.0004998967061669668,
"loss": 6.4824,
"mean_token_accuracy": 0.12094468101859093,
"num_tokens": 4269357.0,
"step": 2095
},
{
"entropy": 6.694693088531494,
"epoch": 0.18351830813597833,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998957598974559,
"loss": 6.4142,
"mean_token_accuracy": 0.13116534650325776,
"num_tokens": 4279576.0,
"step": 2100
},
{
"entropy": 6.660217428207398,
"epoch": 0.18395525648868305,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998948093143266,
"loss": 6.4518,
"mean_token_accuracy": 0.13070998936891556,
"num_tokens": 4289761.0,
"step": 2105
},
{
"entropy": 6.606316137313843,
"epoch": 0.18439220484138774,
"grad_norm": 0.92578125,
"learning_rate": 0.000499893854417597,
"loss": 6.3993,
"mean_token_accuracy": 0.13042399287223816,
"num_tokens": 4299231.0,
"step": 2110
},
{
"entropy": 6.696398591995239,
"epoch": 0.18482915319409246,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998928952072854,
"loss": 6.5671,
"mean_token_accuracy": 0.12065399885177612,
"num_tokens": 4308687.0,
"step": 2115
},
{
"entropy": 6.724217700958252,
"epoch": 0.18526610154679718,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998919316834102,
"loss": 6.5186,
"mean_token_accuracy": 0.1235981173813343,
"num_tokens": 4318183.0,
"step": 2120
},
{
"entropy": 6.707517290115357,
"epoch": 0.18570304989950187,
"grad_norm": 0.9453125,
"learning_rate": 0.00049989096384599,
"loss": 6.544,
"mean_token_accuracy": 0.11985958814620971,
"num_tokens": 4328482.0,
"step": 2125
},
{
"entropy": 6.697282981872559,
"epoch": 0.1861399982522066,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998899916950431,
"loss": 6.4329,
"mean_token_accuracy": 0.1220651850104332,
"num_tokens": 4339313.0,
"step": 2130
},
{
"entropy": 6.647680282592773,
"epoch": 0.1865769466049113,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998890152305885,
"loss": 6.5417,
"mean_token_accuracy": 0.12274084091186524,
"num_tokens": 4350299.0,
"step": 2135
},
{
"entropy": 6.823352575302124,
"epoch": 0.187013894957616,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998880344526447,
"loss": 6.4685,
"mean_token_accuracy": 0.12892222180962562,
"num_tokens": 4361424.0,
"step": 2140
},
{
"entropy": 6.548090219497681,
"epoch": 0.18745084331032072,
"grad_norm": 0.875,
"learning_rate": 0.0004998870493612306,
"loss": 6.4544,
"mean_token_accuracy": 0.12490695714950562,
"num_tokens": 4371480.0,
"step": 2145
},
{
"entropy": 6.629548788070679,
"epoch": 0.18788779166302544,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998860599563649,
"loss": 6.4753,
"mean_token_accuracy": 0.12593501806259155,
"num_tokens": 4381951.0,
"step": 2150
},
{
"entropy": 6.677478265762329,
"epoch": 0.18832474001573013,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998850662380668,
"loss": 6.5327,
"mean_token_accuracy": 0.11975304633378983,
"num_tokens": 4392290.0,
"step": 2155
},
{
"entropy": 6.600126171112061,
"epoch": 0.18876168836843485,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998840682063553,
"loss": 6.281,
"mean_token_accuracy": 0.1401298873126507,
"num_tokens": 4403350.0,
"step": 2160
},
{
"entropy": 6.602562713623047,
"epoch": 0.18919863672113957,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998830658612496,
"loss": 6.4852,
"mean_token_accuracy": 0.12300979048013687,
"num_tokens": 4413171.0,
"step": 2165
},
{
"entropy": 6.6313416957855225,
"epoch": 0.18963558507384426,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998820592027688,
"loss": 6.4376,
"mean_token_accuracy": 0.12936783209443092,
"num_tokens": 4424371.0,
"step": 2170
},
{
"entropy": 6.684165811538696,
"epoch": 0.19007253342654898,
"grad_norm": 0.90625,
"learning_rate": 0.0004998810482309322,
"loss": 6.4014,
"mean_token_accuracy": 0.12163952067494392,
"num_tokens": 4434040.0,
"step": 2175
},
{
"entropy": 6.727186775207519,
"epoch": 0.1905094817792537,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998800329457595,
"loss": 6.5153,
"mean_token_accuracy": 0.11811432018876075,
"num_tokens": 4444663.0,
"step": 2180
},
{
"entropy": 6.65099720954895,
"epoch": 0.1909464301319584,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998790133472697,
"loss": 6.4614,
"mean_token_accuracy": 0.12238967642188073,
"num_tokens": 4454733.0,
"step": 2185
},
{
"entropy": 6.613533592224121,
"epoch": 0.1913833784846631,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998779894354827,
"loss": 6.4482,
"mean_token_accuracy": 0.12639879956841468,
"num_tokens": 4464286.0,
"step": 2190
},
{
"entropy": 6.706346750259399,
"epoch": 0.19182032683736783,
"grad_norm": 0.86328125,
"learning_rate": 0.000499876961210418,
"loss": 6.4161,
"mean_token_accuracy": 0.13247158452868463,
"num_tokens": 4475005.0,
"step": 2195
},
{
"entropy": 6.5794055461883545,
"epoch": 0.19225727519007255,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998759286720953,
"loss": 6.3851,
"mean_token_accuracy": 0.12563437446951867,
"num_tokens": 4484606.0,
"step": 2200
},
{
"entropy": 6.607835817337036,
"epoch": 0.19269422354277724,
"grad_norm": 0.9375,
"learning_rate": 0.0004998748918205345,
"loss": 6.4861,
"mean_token_accuracy": 0.12292944341897964,
"num_tokens": 4494930.0,
"step": 2205
},
{
"entropy": 6.6335704803466795,
"epoch": 0.19313117189548196,
"grad_norm": 0.921875,
"learning_rate": 0.0004998738506557554,
"loss": 6.4963,
"mean_token_accuracy": 0.12030340284109116,
"num_tokens": 4505304.0,
"step": 2210
},
{
"entropy": 6.737761306762695,
"epoch": 0.19356812024818668,
"grad_norm": 0.921875,
"learning_rate": 0.000499872805177778,
"loss": 6.4433,
"mean_token_accuracy": 0.1285339929163456,
"num_tokens": 4514333.0,
"step": 2215
},
{
"entropy": 6.530314922332764,
"epoch": 0.19400506860089137,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998717553866225,
"loss": 6.4858,
"mean_token_accuracy": 0.13358827754855157,
"num_tokens": 4525175.0,
"step": 2220
},
{
"entropy": 6.710753393173218,
"epoch": 0.1944420169535961,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998707012823087,
"loss": 6.3632,
"mean_token_accuracy": 0.12810946255922318,
"num_tokens": 4534212.0,
"step": 2225
},
{
"entropy": 6.524042177200317,
"epoch": 0.1948789653063008,
"grad_norm": 0.82421875,
"learning_rate": 0.0004998696428648573,
"loss": 6.5338,
"mean_token_accuracy": 0.12357461825013161,
"num_tokens": 4545141.0,
"step": 2230
},
{
"entropy": 6.690764856338501,
"epoch": 0.1953159136590055,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998685801342881,
"loss": 6.404,
"mean_token_accuracy": 0.13337788805365564,
"num_tokens": 4554410.0,
"step": 2235
},
{
"entropy": 6.552084350585938,
"epoch": 0.19575286201171022,
"grad_norm": 0.921875,
"learning_rate": 0.0004998675130906217,
"loss": 6.4012,
"mean_token_accuracy": 0.13402877897024154,
"num_tokens": 4564397.0,
"step": 2240
},
{
"entropy": 6.539749431610107,
"epoch": 0.19618981036441493,
"grad_norm": 0.875,
"learning_rate": 0.0004998664417338787,
"loss": 6.3971,
"mean_token_accuracy": 0.13362254872918128,
"num_tokens": 4573986.0,
"step": 2245
},
{
"entropy": 6.70291690826416,
"epoch": 0.19662675871711963,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998653660640794,
"loss": 6.4073,
"mean_token_accuracy": 0.1318839006125927,
"num_tokens": 4583449.0,
"step": 2250
},
{
"entropy": 6.56984543800354,
"epoch": 0.19706370706982435,
"grad_norm": 1.046875,
"learning_rate": 0.0004998642860812447,
"loss": 6.4369,
"mean_token_accuracy": 0.12867740020155907,
"num_tokens": 4593637.0,
"step": 2255
},
{
"entropy": 6.647236490249634,
"epoch": 0.19750065542252906,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998632017853951,
"loss": 6.4067,
"mean_token_accuracy": 0.13120522499084472,
"num_tokens": 4603865.0,
"step": 2260
},
{
"entropy": 6.672588539123535,
"epoch": 0.19793760377523376,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998621131765514,
"loss": 6.4538,
"mean_token_accuracy": 0.13059509620070459,
"num_tokens": 4613502.0,
"step": 2265
},
{
"entropy": 6.535258674621582,
"epoch": 0.19837455212793847,
"grad_norm": 0.953125,
"learning_rate": 0.0004998610202547345,
"loss": 6.4143,
"mean_token_accuracy": 0.13172344267368316,
"num_tokens": 4623163.0,
"step": 2270
},
{
"entropy": 6.695293951034546,
"epoch": 0.1988115004806432,
"grad_norm": 0.875,
"learning_rate": 0.0004998599230199654,
"loss": 6.4193,
"mean_token_accuracy": 0.13490767627954484,
"num_tokens": 4632850.0,
"step": 2275
},
{
"entropy": 6.589443254470825,
"epoch": 0.19924844883334789,
"grad_norm": 0.953125,
"learning_rate": 0.0004998588214722653,
"loss": 6.4396,
"mean_token_accuracy": 0.1286391481757164,
"num_tokens": 4642858.0,
"step": 2280
},
{
"entropy": 6.698593091964722,
"epoch": 0.1996853971860526,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998577156116551,
"loss": 6.4627,
"mean_token_accuracy": 0.11916870698332786,
"num_tokens": 4653083.0,
"step": 2285
},
{
"entropy": 6.624360275268555,
"epoch": 0.20012234553875732,
"grad_norm": 0.921875,
"learning_rate": 0.0004998566054381561,
"loss": 6.4798,
"mean_token_accuracy": 0.1257111832499504,
"num_tokens": 4663353.0,
"step": 2290
},
{
"entropy": 6.623066425323486,
"epoch": 0.20055929389146204,
"grad_norm": 1.21875,
"learning_rate": 0.0004998554909517895,
"loss": 6.3213,
"mean_token_accuracy": 0.13321887254714965,
"num_tokens": 4673878.0,
"step": 2295
},
{
"entropy": 6.537002611160278,
"epoch": 0.20099624224416673,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998543721525769,
"loss": 6.4024,
"mean_token_accuracy": 0.12497134804725647,
"num_tokens": 4683661.0,
"step": 2300
},
{
"entropy": 6.5673808574676515,
"epoch": 0.20143319059687145,
"grad_norm": 0.953125,
"learning_rate": 0.0004998532490405397,
"loss": 6.4518,
"mean_token_accuracy": 0.12829202711582183,
"num_tokens": 4693817.0,
"step": 2305
},
{
"entropy": 6.611567735671997,
"epoch": 0.20187013894957617,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998521216156991,
"loss": 6.3772,
"mean_token_accuracy": 0.12686420530080794,
"num_tokens": 4703550.0,
"step": 2310
},
{
"entropy": 6.595733690261841,
"epoch": 0.20230708730228086,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998509898780771,
"loss": 6.4934,
"mean_token_accuracy": 0.12352922856807709,
"num_tokens": 4714872.0,
"step": 2315
},
{
"entropy": 6.627568149566651,
"epoch": 0.20274403565498558,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998498538276954,
"loss": 6.3898,
"mean_token_accuracy": 0.13363091200590133,
"num_tokens": 4725745.0,
"step": 2320
},
{
"entropy": 6.600788736343384,
"epoch": 0.2031809840076903,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998487134645755,
"loss": 6.4799,
"mean_token_accuracy": 0.1253775343298912,
"num_tokens": 4735560.0,
"step": 2325
},
{
"entropy": 6.63616189956665,
"epoch": 0.203617932360395,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998475687887396,
"loss": 6.4533,
"mean_token_accuracy": 0.12529229298233985,
"num_tokens": 4746304.0,
"step": 2330
},
{
"entropy": 6.672122478485107,
"epoch": 0.2040548807130997,
"grad_norm": 0.921875,
"learning_rate": 0.0004998464198002095,
"loss": 6.4285,
"mean_token_accuracy": 0.12777206525206566,
"num_tokens": 4756200.0,
"step": 2335
},
{
"entropy": 6.560856103897095,
"epoch": 0.20449182906580443,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998452664990072,
"loss": 6.436,
"mean_token_accuracy": 0.12261992320418358,
"num_tokens": 4766249.0,
"step": 2340
},
{
"entropy": 6.714810609817505,
"epoch": 0.20492877741850912,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998441088851548,
"loss": 6.4579,
"mean_token_accuracy": 0.12660123854875566,
"num_tokens": 4776510.0,
"step": 2345
},
{
"entropy": 6.561935997009277,
"epoch": 0.20536572577121384,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998429469586748,
"loss": 6.4211,
"mean_token_accuracy": 0.1272496648132801,
"num_tokens": 4786798.0,
"step": 2350
},
{
"entropy": 6.608212327957153,
"epoch": 0.20580267412391856,
"grad_norm": 0.921875,
"learning_rate": 0.000499841780719589,
"loss": 6.3616,
"mean_token_accuracy": 0.12745508700609207,
"num_tokens": 4796001.0,
"step": 2355
},
{
"entropy": 6.519495153427124,
"epoch": 0.20623962247662325,
"grad_norm": 1.125,
"learning_rate": 0.0004998406101679202,
"loss": 6.3445,
"mean_token_accuracy": 0.13589456528425217,
"num_tokens": 4807634.0,
"step": 2360
},
{
"entropy": 6.578785228729248,
"epoch": 0.20667657082932797,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998394353036906,
"loss": 6.3473,
"mean_token_accuracy": 0.13589533045887947,
"num_tokens": 4817840.0,
"step": 2365
},
{
"entropy": 6.551007890701294,
"epoch": 0.2071135191820327,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998382561269229,
"loss": 6.3326,
"mean_token_accuracy": 0.1385917641222477,
"num_tokens": 4827186.0,
"step": 2370
},
{
"entropy": 6.533224439620971,
"epoch": 0.20755046753473738,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998370726376395,
"loss": 6.3894,
"mean_token_accuracy": 0.12424067407846451,
"num_tokens": 4837140.0,
"step": 2375
},
{
"entropy": 6.589614820480347,
"epoch": 0.2079874158874421,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998358848358634,
"loss": 6.4017,
"mean_token_accuracy": 0.12754351869225503,
"num_tokens": 4846907.0,
"step": 2380
},
{
"entropy": 6.561453104019165,
"epoch": 0.20842436424014682,
"grad_norm": 0.87890625,
"learning_rate": 0.000499834692721617,
"loss": 6.421,
"mean_token_accuracy": 0.13229633197188378,
"num_tokens": 4856165.0,
"step": 2385
},
{
"entropy": 6.578060579299927,
"epoch": 0.2088613125928515,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998334962949237,
"loss": 6.2911,
"mean_token_accuracy": 0.14010017663240432,
"num_tokens": 4865308.0,
"step": 2390
},
{
"entropy": 6.493227291107178,
"epoch": 0.20929826094555623,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998322955558059,
"loss": 6.358,
"mean_token_accuracy": 0.1417185001075268,
"num_tokens": 4876286.0,
"step": 2395
},
{
"entropy": 6.591450119018555,
"epoch": 0.20973520929826095,
"grad_norm": 0.81640625,
"learning_rate": 0.0004998310905042869,
"loss": 6.384,
"mean_token_accuracy": 0.13348621502518654,
"num_tokens": 4887227.0,
"step": 2400
},
{
"entropy": 6.593507528305054,
"epoch": 0.21017215765096567,
"grad_norm": 0.8125,
"learning_rate": 0.0004998298811403898,
"loss": 6.3979,
"mean_token_accuracy": 0.1288406252861023,
"num_tokens": 4896994.0,
"step": 2405
},
{
"entropy": 6.533228969573974,
"epoch": 0.21060910600367036,
"grad_norm": 1.015625,
"learning_rate": 0.0004998286674641378,
"loss": 6.4045,
"mean_token_accuracy": 0.12903448343276977,
"num_tokens": 4907623.0,
"step": 2410
},
{
"entropy": 6.558895254135132,
"epoch": 0.21104605435637508,
"grad_norm": 0.8984375,
"learning_rate": 0.000499827449475554,
"loss": 6.3204,
"mean_token_accuracy": 0.12823840528726577,
"num_tokens": 4917040.0,
"step": 2415
},
{
"entropy": 6.683346271514893,
"epoch": 0.2114830027090798,
"grad_norm": 0.8984375,
"learning_rate": 0.000499826227174662,
"loss": 6.436,
"mean_token_accuracy": 0.12896525338292122,
"num_tokens": 4928390.0,
"step": 2420
},
{
"entropy": 6.50466251373291,
"epoch": 0.2119199510617845,
"grad_norm": 0.890625,
"learning_rate": 0.0004998250005614852,
"loss": 6.3981,
"mean_token_accuracy": 0.12751310169696808,
"num_tokens": 4939040.0,
"step": 2425
},
{
"entropy": 6.558785009384155,
"epoch": 0.2123568994144892,
"grad_norm": 0.8671875,
"learning_rate": 0.000499823769636047,
"loss": 6.4371,
"mean_token_accuracy": 0.12819619551301004,
"num_tokens": 4949915.0,
"step": 2430
},
{
"entropy": 6.607331132888794,
"epoch": 0.21279384776719393,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998225343983712,
"loss": 6.3391,
"mean_token_accuracy": 0.1329527035355568,
"num_tokens": 4960127.0,
"step": 2435
},
{
"entropy": 6.574900817871094,
"epoch": 0.21323079611989862,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998212948484814,
"loss": 6.3847,
"mean_token_accuracy": 0.126432553678751,
"num_tokens": 4970152.0,
"step": 2440
},
{
"entropy": 6.571488666534424,
"epoch": 0.21366774447260334,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998200509864012,
"loss": 6.3577,
"mean_token_accuracy": 0.1341804251074791,
"num_tokens": 4979745.0,
"step": 2445
},
{
"entropy": 6.594966459274292,
"epoch": 0.21410469282530806,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998188028121547,
"loss": 6.3687,
"mean_token_accuracy": 0.1393337517976761,
"num_tokens": 4989924.0,
"step": 2450
},
{
"entropy": 6.498985242843628,
"epoch": 0.21454164117801275,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998175503257658,
"loss": 6.316,
"mean_token_accuracy": 0.14141854122281075,
"num_tokens": 4999698.0,
"step": 2455
},
{
"entropy": 6.539078855514527,
"epoch": 0.21497858953071747,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998162935272585,
"loss": 6.4223,
"mean_token_accuracy": 0.13204047605395317,
"num_tokens": 5010625.0,
"step": 2460
},
{
"entropy": 6.532508611679077,
"epoch": 0.2154155378834222,
"grad_norm": 0.9375,
"learning_rate": 0.0004998150324166568,
"loss": 6.3072,
"mean_token_accuracy": 0.13948291540145874,
"num_tokens": 5019740.0,
"step": 2465
},
{
"entropy": 6.578542041778564,
"epoch": 0.21585248623612688,
"grad_norm": 0.78125,
"learning_rate": 0.000499813766993985,
"loss": 6.4354,
"mean_token_accuracy": 0.12458334863185883,
"num_tokens": 5030601.0,
"step": 2470
},
{
"entropy": 6.645995283126831,
"epoch": 0.2162894345888316,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998124972592674,
"loss": 6.3675,
"mean_token_accuracy": 0.13052909597754478,
"num_tokens": 5040567.0,
"step": 2475
},
{
"entropy": 6.499893093109131,
"epoch": 0.21672638294153632,
"grad_norm": 1.046875,
"learning_rate": 0.0004998112232125283,
"loss": 6.2648,
"mean_token_accuracy": 0.13649773374199867,
"num_tokens": 5049512.0,
"step": 2480
},
{
"entropy": 6.491473197937012,
"epoch": 0.217163331294241,
"grad_norm": 0.82421875,
"learning_rate": 0.000499809944853792,
"loss": 6.4316,
"mean_token_accuracy": 0.1274581141769886,
"num_tokens": 5060986.0,
"step": 2485
},
{
"entropy": 6.577273893356323,
"epoch": 0.21760027964694573,
"grad_norm": 0.9375,
"learning_rate": 0.0004998086621830833,
"loss": 6.3573,
"mean_token_accuracy": 0.13598684519529342,
"num_tokens": 5071492.0,
"step": 2490
},
{
"entropy": 6.544683933258057,
"epoch": 0.21803722799965045,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998073752004267,
"loss": 6.3905,
"mean_token_accuracy": 0.12805117219686507,
"num_tokens": 5081722.0,
"step": 2495
},
{
"entropy": 6.5953404903411865,
"epoch": 0.21847417635235516,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998060839058467,
"loss": 6.3545,
"mean_token_accuracy": 0.1329400770366192,
"num_tokens": 5091906.0,
"step": 2500
},
{
"entropy": 6.5407976627349855,
"epoch": 0.21891112470505986,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998047882993683,
"loss": 6.3426,
"mean_token_accuracy": 0.13389598056674004,
"num_tokens": 5102025.0,
"step": 2505
},
{
"entropy": 6.509171485900879,
"epoch": 0.21934807305776458,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998034883810162,
"loss": 6.3367,
"mean_token_accuracy": 0.13235569521784782,
"num_tokens": 5112354.0,
"step": 2510
},
{
"entropy": 6.513673925399781,
"epoch": 0.2197850214104693,
"grad_norm": 0.875,
"learning_rate": 0.0004998021841508155,
"loss": 6.2825,
"mean_token_accuracy": 0.13621082603931428,
"num_tokens": 5121695.0,
"step": 2515
},
{
"entropy": 6.564906120300293,
"epoch": 0.22022196976317399,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998008756087911,
"loss": 6.3891,
"mean_token_accuracy": 0.12776936888694762,
"num_tokens": 5132561.0,
"step": 2520
},
{
"entropy": 6.589311695098877,
"epoch": 0.2206589181158787,
"grad_norm": 0.88671875,
"learning_rate": 0.000499799562754968,
"loss": 6.4257,
"mean_token_accuracy": 0.12993937209248543,
"num_tokens": 5141981.0,
"step": 2525
},
{
"entropy": 6.544518756866455,
"epoch": 0.22109586646858342,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997982455893716,
"loss": 6.2883,
"mean_token_accuracy": 0.1357527531683445,
"num_tokens": 5151550.0,
"step": 2530
},
{
"entropy": 6.574628019332886,
"epoch": 0.22153281482128812,
"grad_norm": 0.984375,
"learning_rate": 0.0004997969241120271,
"loss": 6.3807,
"mean_token_accuracy": 0.1260584406554699,
"num_tokens": 5161679.0,
"step": 2535
},
{
"entropy": 6.491645574569702,
"epoch": 0.22196976317399283,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997955983229597,
"loss": 6.4026,
"mean_token_accuracy": 0.1277308262884617,
"num_tokens": 5171575.0,
"step": 2540
},
{
"entropy": 6.547885131835938,
"epoch": 0.22240671152669755,
"grad_norm": 0.98828125,
"learning_rate": 0.000499794268222195,
"loss": 6.2719,
"mean_token_accuracy": 0.13648097291588784,
"num_tokens": 5180152.0,
"step": 2545
},
{
"entropy": 6.491997766494751,
"epoch": 0.22284365987940224,
"grad_norm": 0.96875,
"learning_rate": 0.0004997929338097583,
"loss": 6.3554,
"mean_token_accuracy": 0.13152262344956397,
"num_tokens": 5190518.0,
"step": 2550
},
{
"entropy": 6.607607460021972,
"epoch": 0.22328060823210696,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997915950856755,
"loss": 6.3933,
"mean_token_accuracy": 0.12704384401440622,
"num_tokens": 5201729.0,
"step": 2555
},
{
"entropy": 6.559991312026978,
"epoch": 0.22371755658481168,
"grad_norm": 0.921875,
"learning_rate": 0.0004997902520499721,
"loss": 6.3744,
"mean_token_accuracy": 0.1310870312154293,
"num_tokens": 5211940.0,
"step": 2560
},
{
"entropy": 6.471191549301148,
"epoch": 0.22415450493751637,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997889047026738,
"loss": 6.3049,
"mean_token_accuracy": 0.13113719150424002,
"num_tokens": 5221028.0,
"step": 2565
},
{
"entropy": 6.549423122406006,
"epoch": 0.2245914532902211,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997875530438066,
"loss": 6.287,
"mean_token_accuracy": 0.1366332732141018,
"num_tokens": 5230498.0,
"step": 2570
},
{
"entropy": 6.511614179611206,
"epoch": 0.2250284016429258,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997861970733963,
"loss": 6.3353,
"mean_token_accuracy": 0.12960499599575998,
"num_tokens": 5240112.0,
"step": 2575
},
{
"entropy": 6.5579187870025635,
"epoch": 0.2254653499956305,
"grad_norm": 0.91796875,
"learning_rate": 0.000499784836791469,
"loss": 6.3078,
"mean_token_accuracy": 0.13300390243530275,
"num_tokens": 5249201.0,
"step": 2580
},
{
"entropy": 6.493053865432739,
"epoch": 0.22590229834833522,
"grad_norm": 0.87890625,
"learning_rate": 0.0004997834721980507,
"loss": 6.305,
"mean_token_accuracy": 0.13774593099951743,
"num_tokens": 5258541.0,
"step": 2585
},
{
"entropy": 6.616273880004883,
"epoch": 0.22633924670103994,
"grad_norm": 1.1796875,
"learning_rate": 0.0004997821032931676,
"loss": 6.4101,
"mean_token_accuracy": 0.1332601450383663,
"num_tokens": 5268524.0,
"step": 2590
},
{
"entropy": 6.509102439880371,
"epoch": 0.22677619505374463,
"grad_norm": 0.8203125,
"learning_rate": 0.000499780730076846,
"loss": 6.3307,
"mean_token_accuracy": 0.1330648258328438,
"num_tokens": 5280702.0,
"step": 2595
},
{
"entropy": 6.528602409362793,
"epoch": 0.22721314340644935,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997793525491122,
"loss": 6.31,
"mean_token_accuracy": 0.1283218316733837,
"num_tokens": 5291214.0,
"step": 2600
},
{
"entropy": 6.432008266448975,
"epoch": 0.22765009175915407,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997779707099926,
"loss": 6.2639,
"mean_token_accuracy": 0.14060572162270546,
"num_tokens": 5300729.0,
"step": 2605
},
{
"entropy": 6.540947341918946,
"epoch": 0.2280870401118588,
"grad_norm": 0.875,
"learning_rate": 0.0004997765845595138,
"loss": 6.4009,
"mean_token_accuracy": 0.1288476750254631,
"num_tokens": 5310551.0,
"step": 2610
},
{
"entropy": 6.5121190547943115,
"epoch": 0.22852398846456348,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997751940977024,
"loss": 6.2763,
"mean_token_accuracy": 0.13367155566811562,
"num_tokens": 5320038.0,
"step": 2615
},
{
"entropy": 6.493434906005859,
"epoch": 0.2289609368172682,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997737993245848,
"loss": 6.3191,
"mean_token_accuracy": 0.1320577569305897,
"num_tokens": 5329753.0,
"step": 2620
},
{
"entropy": 6.4705726146698,
"epoch": 0.22939788516997292,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997724002401881,
"loss": 6.304,
"mean_token_accuracy": 0.13512743636965752,
"num_tokens": 5340546.0,
"step": 2625
},
{
"entropy": 6.446058702468872,
"epoch": 0.2298348335226776,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997709968445391,
"loss": 6.3795,
"mean_token_accuracy": 0.12804465666413306,
"num_tokens": 5352097.0,
"step": 2630
},
{
"entropy": 6.65869174003601,
"epoch": 0.23027178187538233,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997695891376644,
"loss": 6.3208,
"mean_token_accuracy": 0.13751014918088914,
"num_tokens": 5361646.0,
"step": 2635
},
{
"entropy": 6.395161771774292,
"epoch": 0.23070873022808705,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997681771195912,
"loss": 6.3041,
"mean_token_accuracy": 0.13794874846935273,
"num_tokens": 5371695.0,
"step": 2640
},
{
"entropy": 6.512343025207519,
"epoch": 0.23114567858079174,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997667607903466,
"loss": 6.3805,
"mean_token_accuracy": 0.13010330498218536,
"num_tokens": 5381978.0,
"step": 2645
},
{
"entropy": 6.660280466079712,
"epoch": 0.23158262693349646,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997653401499578,
"loss": 6.3697,
"mean_token_accuracy": 0.13411956354975701,
"num_tokens": 5391891.0,
"step": 2650
},
{
"entropy": 6.493831157684326,
"epoch": 0.23201957528620118,
"grad_norm": 0.8671875,
"learning_rate": 0.000499763915198452,
"loss": 6.327,
"mean_token_accuracy": 0.13065533563494683,
"num_tokens": 5403172.0,
"step": 2655
},
{
"entropy": 6.456011819839477,
"epoch": 0.23245652363890587,
"grad_norm": 0.875,
"learning_rate": 0.0004997624859358565,
"loss": 6.3345,
"mean_token_accuracy": 0.13543095886707307,
"num_tokens": 5413130.0,
"step": 2660
},
{
"entropy": 6.504063415527344,
"epoch": 0.2328934719916106,
"grad_norm": 0.890625,
"learning_rate": 0.0004997610523621988,
"loss": 6.3657,
"mean_token_accuracy": 0.13021062016487123,
"num_tokens": 5423510.0,
"step": 2665
},
{
"entropy": 6.5001222610473635,
"epoch": 0.2333304203443153,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997596144775063,
"loss": 6.2,
"mean_token_accuracy": 0.13856842517852783,
"num_tokens": 5433707.0,
"step": 2670
},
{
"entropy": 6.316236448287964,
"epoch": 0.23376736869702,
"grad_norm": 0.8515625,
"learning_rate": 0.0004997581722818067,
"loss": 6.3145,
"mean_token_accuracy": 0.13552152067422868,
"num_tokens": 5444381.0,
"step": 2675
},
{
"entropy": 6.673515939712525,
"epoch": 0.23420431704972472,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997567257751275,
"loss": 6.264,
"mean_token_accuracy": 0.13442860767245293,
"num_tokens": 5453905.0,
"step": 2680
},
{
"entropy": 6.449249124526977,
"epoch": 0.23464126540242944,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997552749574965,
"loss": 6.2845,
"mean_token_accuracy": 0.13393480256199836,
"num_tokens": 5464839.0,
"step": 2685
},
{
"entropy": 6.488210773468017,
"epoch": 0.23507821375513413,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997538198289416,
"loss": 6.3229,
"mean_token_accuracy": 0.12909892126917838,
"num_tokens": 5475459.0,
"step": 2690
},
{
"entropy": 6.478167104721069,
"epoch": 0.23551516210783885,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997523603894907,
"loss": 6.3397,
"mean_token_accuracy": 0.13458095341920853,
"num_tokens": 5484886.0,
"step": 2695
},
{
"entropy": 6.4847033500671385,
"epoch": 0.23595211046054357,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997508966391717,
"loss": 6.318,
"mean_token_accuracy": 0.13543656319379807,
"num_tokens": 5494803.0,
"step": 2700
},
{
"entropy": 6.465766191482544,
"epoch": 0.2363890588132483,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997494285780129,
"loss": 6.2689,
"mean_token_accuracy": 0.13447927311062813,
"num_tokens": 5504997.0,
"step": 2705
},
{
"entropy": 6.446899366378784,
"epoch": 0.23682600716595298,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997479562060421,
"loss": 6.2569,
"mean_token_accuracy": 0.14331548362970353,
"num_tokens": 5514492.0,
"step": 2710
},
{
"entropy": 6.511632680892944,
"epoch": 0.2372629555186577,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997464795232877,
"loss": 6.3097,
"mean_token_accuracy": 0.13123842626810073,
"num_tokens": 5524419.0,
"step": 2715
},
{
"entropy": 6.525136137008667,
"epoch": 0.23769990387136242,
"grad_norm": 0.859375,
"learning_rate": 0.0004997449985297782,
"loss": 6.323,
"mean_token_accuracy": 0.1331531934440136,
"num_tokens": 5534289.0,
"step": 2720
},
{
"entropy": 6.446746015548706,
"epoch": 0.2381368522240671,
"grad_norm": 1.046875,
"learning_rate": 0.0004997435132255418,
"loss": 6.2524,
"mean_token_accuracy": 0.13549713790416718,
"num_tokens": 5543034.0,
"step": 2725
},
{
"entropy": 6.511647033691406,
"epoch": 0.23857380057677183,
"grad_norm": 0.875,
"learning_rate": 0.0004997420236106071,
"loss": 6.3347,
"mean_token_accuracy": 0.12996043935418128,
"num_tokens": 5553291.0,
"step": 2730
},
{
"entropy": 6.52672061920166,
"epoch": 0.23901074892947655,
"grad_norm": 0.8515625,
"learning_rate": 0.0004997405296850026,
"loss": 6.3147,
"mean_token_accuracy": 0.1387110285460949,
"num_tokens": 5563772.0,
"step": 2735
},
{
"entropy": 6.469681453704834,
"epoch": 0.23944769728218124,
"grad_norm": 0.96875,
"learning_rate": 0.0004997390314487569,
"loss": 6.352,
"mean_token_accuracy": 0.13162039518356322,
"num_tokens": 5574348.0,
"step": 2740
},
{
"entropy": 6.486959552764892,
"epoch": 0.23988464563488596,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997375289018989,
"loss": 6.2096,
"mean_token_accuracy": 0.13282159790396691,
"num_tokens": 5583491.0,
"step": 2745
},
{
"entropy": 6.399618673324585,
"epoch": 0.24032159398759068,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997360220444574,
"loss": 6.3508,
"mean_token_accuracy": 0.12875445038080216,
"num_tokens": 5594273.0,
"step": 2750
},
{
"entropy": 6.516872072219849,
"epoch": 0.24075854234029537,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997345108764611,
"loss": 6.2636,
"mean_token_accuracy": 0.13464151471853256,
"num_tokens": 5604529.0,
"step": 2755
},
{
"entropy": 6.463331794738769,
"epoch": 0.2411954906930001,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997329953979391,
"loss": 6.2527,
"mean_token_accuracy": 0.14141659960150718,
"num_tokens": 5614036.0,
"step": 2760
},
{
"entropy": 6.510278177261353,
"epoch": 0.2416324390457048,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997314756089206,
"loss": 6.2963,
"mean_token_accuracy": 0.13726957514882088,
"num_tokens": 5624036.0,
"step": 2765
},
{
"entropy": 6.450707244873047,
"epoch": 0.2420693873984095,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997299515094347,
"loss": 6.289,
"mean_token_accuracy": 0.13998497128486634,
"num_tokens": 5633987.0,
"step": 2770
},
{
"entropy": 6.427713871002197,
"epoch": 0.24250633575111422,
"grad_norm": 0.90625,
"learning_rate": 0.0004997284230995105,
"loss": 6.166,
"mean_token_accuracy": 0.14919004142284392,
"num_tokens": 5643905.0,
"step": 2775
},
{
"entropy": 6.4683431625366214,
"epoch": 0.24294328410381894,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997268903791773,
"loss": 6.3625,
"mean_token_accuracy": 0.13415857255458832,
"num_tokens": 5654922.0,
"step": 2780
},
{
"entropy": 6.475963926315307,
"epoch": 0.24338023245652363,
"grad_norm": 0.82421875,
"learning_rate": 0.0004997253533484647,
"loss": 6.1973,
"mean_token_accuracy": 0.1422443114221096,
"num_tokens": 5664407.0,
"step": 2785
},
{
"entropy": 6.472312259674072,
"epoch": 0.24381718080922835,
"grad_norm": 0.9375,
"learning_rate": 0.0004997238120074021,
"loss": 6.2675,
"mean_token_accuracy": 0.13551050946116447,
"num_tokens": 5674524.0,
"step": 2790
},
{
"entropy": 6.439700555801392,
"epoch": 0.24425412916193306,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997222663560188,
"loss": 6.3007,
"mean_token_accuracy": 0.1321689762175083,
"num_tokens": 5684691.0,
"step": 2795
},
{
"entropy": 6.4623698711395265,
"epoch": 0.24469107751463778,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997207163943449,
"loss": 6.2353,
"mean_token_accuracy": 0.13899511322379113,
"num_tokens": 5693683.0,
"step": 2800
},
{
"entropy": 6.442118167877197,
"epoch": 0.24512802586734248,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997191621224098,
"loss": 6.2998,
"mean_token_accuracy": 0.13590859174728392,
"num_tokens": 5704394.0,
"step": 2805
},
{
"entropy": 6.445099401473999,
"epoch": 0.2455649742200472,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997176035402435,
"loss": 6.2111,
"mean_token_accuracy": 0.1418813318014145,
"num_tokens": 5715480.0,
"step": 2810
},
{
"entropy": 6.484379482269287,
"epoch": 0.2460019225727519,
"grad_norm": 0.8671875,
"learning_rate": 0.0004997160406478758,
"loss": 6.2227,
"mean_token_accuracy": 0.14038358852267266,
"num_tokens": 5725973.0,
"step": 2815
},
{
"entropy": 6.332956981658936,
"epoch": 0.2464388709254566,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997144734453367,
"loss": 6.2387,
"mean_token_accuracy": 0.14358900636434554,
"num_tokens": 5735798.0,
"step": 2820
},
{
"entropy": 6.579839372634888,
"epoch": 0.24687581927816132,
"grad_norm": 0.859375,
"learning_rate": 0.0004997129019326562,
"loss": 6.3218,
"mean_token_accuracy": 0.12896375134587287,
"num_tokens": 5746623.0,
"step": 2825
},
{
"entropy": 6.478277635574341,
"epoch": 0.24731276763086604,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997113261098645,
"loss": 6.2624,
"mean_token_accuracy": 0.14029808342456818,
"num_tokens": 5756526.0,
"step": 2830
},
{
"entropy": 6.400695037841797,
"epoch": 0.24774971598357073,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997097459769919,
"loss": 6.2642,
"mean_token_accuracy": 0.13547210171818733,
"num_tokens": 5766670.0,
"step": 2835
},
{
"entropy": 6.524441814422607,
"epoch": 0.24818666433627545,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997081615340685,
"loss": 6.3055,
"mean_token_accuracy": 0.1356332018971443,
"num_tokens": 5776473.0,
"step": 2840
},
{
"entropy": 6.552545595169067,
"epoch": 0.24862361268898017,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997065727811249,
"loss": 6.3397,
"mean_token_accuracy": 0.13625893667340278,
"num_tokens": 5787059.0,
"step": 2845
},
{
"entropy": 6.441537141799927,
"epoch": 0.24906056104168486,
"grad_norm": 0.84765625,
"learning_rate": 0.0004997049797181915,
"loss": 6.3197,
"mean_token_accuracy": 0.12882847487926483,
"num_tokens": 5797367.0,
"step": 2850
},
{
"entropy": 6.392263746261596,
"epoch": 0.24949750939438958,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997033823452988,
"loss": 6.2121,
"mean_token_accuracy": 0.14500029981136323,
"num_tokens": 5806704.0,
"step": 2855
},
{
"entropy": 6.455396413803101,
"epoch": 0.2499344577470943,
"grad_norm": 0.80859375,
"learning_rate": 0.0004997017806624774,
"loss": 6.2725,
"mean_token_accuracy": 0.1365084804594517,
"num_tokens": 5817797.0,
"step": 2860
},
{
"entropy": 6.506603479385376,
"epoch": 0.250371406099799,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997001746697582,
"loss": 6.3181,
"mean_token_accuracy": 0.13766707554459573,
"num_tokens": 5829270.0,
"step": 2865
},
{
"entropy": 6.450180387496948,
"epoch": 0.2508083544525037,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996985643671719,
"loss": 6.2536,
"mean_token_accuracy": 0.1362561121582985,
"num_tokens": 5839214.0,
"step": 2870
},
{
"entropy": 6.388732814788819,
"epoch": 0.2512453028052084,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996969497547495,
"loss": 6.2499,
"mean_token_accuracy": 0.13788965120911598,
"num_tokens": 5848171.0,
"step": 2875
},
{
"entropy": 6.440700387954712,
"epoch": 0.25168225115791315,
"grad_norm": 0.8359375,
"learning_rate": 0.0004996953308325217,
"loss": 6.1604,
"mean_token_accuracy": 0.14541597664356232,
"num_tokens": 5858399.0,
"step": 2880
},
{
"entropy": 6.306634092330933,
"epoch": 0.25211919951061784,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996937076005198,
"loss": 6.1752,
"mean_token_accuracy": 0.148306243121624,
"num_tokens": 5868212.0,
"step": 2885
},
{
"entropy": 6.465786075592041,
"epoch": 0.25255614786332253,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996920800587749,
"loss": 6.2557,
"mean_token_accuracy": 0.1431330159306526,
"num_tokens": 5877883.0,
"step": 2890
},
{
"entropy": 6.490456247329712,
"epoch": 0.2529930962160273,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996904482073181,
"loss": 6.2612,
"mean_token_accuracy": 0.13619642183184624,
"num_tokens": 5887090.0,
"step": 2895
},
{
"entropy": 6.379231071472168,
"epoch": 0.25343004456873197,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996888120461808,
"loss": 6.299,
"mean_token_accuracy": 0.1262368731200695,
"num_tokens": 5897491.0,
"step": 2900
},
{
"entropy": 6.50172791481018,
"epoch": 0.25386699292143666,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996871715753943,
"loss": 6.3195,
"mean_token_accuracy": 0.13194667026400567,
"num_tokens": 5908494.0,
"step": 2905
},
{
"entropy": 6.527822208404541,
"epoch": 0.2543039412741414,
"grad_norm": 0.8828125,
"learning_rate": 0.00049968552679499,
"loss": 6.3587,
"mean_token_accuracy": 0.12604505568742752,
"num_tokens": 5919708.0,
"step": 2910
},
{
"entropy": 6.4339385509490965,
"epoch": 0.2547408896268461,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996838777049998,
"loss": 6.2224,
"mean_token_accuracy": 0.13939420580863954,
"num_tokens": 5930122.0,
"step": 2915
},
{
"entropy": 6.35483832359314,
"epoch": 0.2551778379795508,
"grad_norm": 0.921875,
"learning_rate": 0.000499682224305455,
"loss": 6.2036,
"mean_token_accuracy": 0.1342392973601818,
"num_tokens": 5940033.0,
"step": 2920
},
{
"entropy": 6.388920021057129,
"epoch": 0.25561478633225554,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996805665963874,
"loss": 6.2204,
"mean_token_accuracy": 0.13428228572010995,
"num_tokens": 5949335.0,
"step": 2925
},
{
"entropy": 6.490639734268188,
"epoch": 0.25605173468496023,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996789045778287,
"loss": 6.264,
"mean_token_accuracy": 0.13692333027720452,
"num_tokens": 5959299.0,
"step": 2930
},
{
"entropy": 6.447759532928467,
"epoch": 0.2564886830376649,
"grad_norm": 0.828125,
"learning_rate": 0.0004996772382498111,
"loss": 6.2174,
"mean_token_accuracy": 0.13530389219522476,
"num_tokens": 5970114.0,
"step": 2935
},
{
"entropy": 6.415350341796875,
"epoch": 0.25692563139036967,
"grad_norm": 0.859375,
"learning_rate": 0.0004996755676123662,
"loss": 6.3127,
"mean_token_accuracy": 0.134555646777153,
"num_tokens": 5979630.0,
"step": 2940
},
{
"entropy": 6.529413795471191,
"epoch": 0.25736257974307436,
"grad_norm": 0.96875,
"learning_rate": 0.0004996738926655262,
"loss": 6.289,
"mean_token_accuracy": 0.13147507831454278,
"num_tokens": 5990190.0,
"step": 2945
},
{
"entropy": 6.40431022644043,
"epoch": 0.25779952809577905,
"grad_norm": 0.87109375,
"learning_rate": 0.0004996722134093231,
"loss": 6.3126,
"mean_token_accuracy": 0.13044267743825913,
"num_tokens": 6000706.0,
"step": 2950
},
{
"entropy": 6.401627731323242,
"epoch": 0.2582364764484838,
"grad_norm": 0.9375,
"learning_rate": 0.0004996705298437894,
"loss": 6.2655,
"mean_token_accuracy": 0.1423163965344429,
"num_tokens": 6011274.0,
"step": 2955
},
{
"entropy": 6.5021644115448,
"epoch": 0.2586734248011885,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996688419689572,
"loss": 6.1833,
"mean_token_accuracy": 0.1421648696064949,
"num_tokens": 6021481.0,
"step": 2960
},
{
"entropy": 6.357604551315307,
"epoch": 0.25911037315389324,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996671497848589,
"loss": 6.2652,
"mean_token_accuracy": 0.1401491306722164,
"num_tokens": 6032091.0,
"step": 2965
},
{
"entropy": 6.467347812652588,
"epoch": 0.25954732150659793,
"grad_norm": 0.890625,
"learning_rate": 0.0004996654532915269,
"loss": 6.2658,
"mean_token_accuracy": 0.1289597101509571,
"num_tokens": 6043371.0,
"step": 2970
},
{
"entropy": 6.476518726348877,
"epoch": 0.2599842698593026,
"grad_norm": 0.9375,
"learning_rate": 0.0004996637524889937,
"loss": 6.2193,
"mean_token_accuracy": 0.1412220261991024,
"num_tokens": 6052812.0,
"step": 2975
},
{
"entropy": 6.3784160137176515,
"epoch": 0.26042121821200737,
"grad_norm": 0.84765625,
"learning_rate": 0.0004996620473772921,
"loss": 6.2135,
"mean_token_accuracy": 0.13825001195073128,
"num_tokens": 6064033.0,
"step": 2980
},
{
"entropy": 6.435154962539673,
"epoch": 0.26085816656471206,
"grad_norm": 0.85546875,
"learning_rate": 0.0004996603379564546,
"loss": 6.3394,
"mean_token_accuracy": 0.13179437443614006,
"num_tokens": 6075837.0,
"step": 2985
},
{
"entropy": 6.565837717056274,
"epoch": 0.26129511491741675,
"grad_norm": 0.83984375,
"learning_rate": 0.0004996586242265142,
"loss": 6.244,
"mean_token_accuracy": 0.14351730942726135,
"num_tokens": 6085514.0,
"step": 2990
},
{
"entropy": 6.4207676410675045,
"epoch": 0.2617320632701215,
"grad_norm": 0.9375,
"learning_rate": 0.0004996569061875036,
"loss": 6.1518,
"mean_token_accuracy": 0.14098117351531983,
"num_tokens": 6094992.0,
"step": 2995
},
{
"entropy": 6.362293386459351,
"epoch": 0.2621690116228262,
"grad_norm": 0.8671875,
"learning_rate": 0.0004996551838394559,
"loss": 6.2304,
"mean_token_accuracy": 0.13752028718590736,
"num_tokens": 6104841.0,
"step": 3000
},
{
"epoch": 0.2621690116228262,
"eval_entropy": 6.155421564639298,
"eval_loss": 6.250469207763672,
"eval_mean_token_accuracy": 0.14463501781151133,
"eval_num_tokens": 6104841.0,
"eval_runtime": 21.3728,
"eval_samples_per_second": 1731.83,
"eval_steps_per_second": 216.49,
"step": 3000
}
],
"logging_steps": 5,
"max_steps": 114420,
"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": 1356222751211520.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink