fpadovani/heb-hebr-10mb-100mb_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
heb-hebr-10mb-100mb_seed3407/checkpoint-6000/trainer_state.json
ModelHub XC dd339cb2fe 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/heb-hebr-10mb-100mb_seed3407
Source: Original Platform
2026-07-17 23:32:55 +08:00

12057 lines
330 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.6420889293167104,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.742589855194092,
"epoch": 0.0005350741077639254,
"grad_norm": 4.28125,
"learning_rate": 2e-06,
"loss": 10.7541,
"mean_token_accuracy": 0.0,
"num_tokens": 10038.0,
"step": 5
},
{
"entropy": 10.742583560943604,
"epoch": 0.0010701482155278507,
"grad_norm": 4.5625,
"learning_rate": 4.5e-06,
"loss": 10.7779,
"mean_token_accuracy": 0.0001301236217841506,
"num_tokens": 21916.0,
"step": 10
},
{
"entropy": 10.74256067276001,
"epoch": 0.0016052223232917758,
"grad_norm": 4.53125,
"learning_rate": 7e-06,
"loss": 10.7579,
"mean_token_accuracy": 0.0,
"num_tokens": 33434.0,
"step": 15
},
{
"entropy": 10.742603588104249,
"epoch": 0.0021402964310557014,
"grad_norm": 4.4375,
"learning_rate": 9.5e-06,
"loss": 10.7083,
"mean_token_accuracy": 0.0009079659474082291,
"num_tokens": 44216.0,
"step": 20
},
{
"entropy": 10.742497825622559,
"epoch": 0.0026753705388196263,
"grad_norm": 4.1875,
"learning_rate": 1.2e-05,
"loss": 10.6241,
"mean_token_accuracy": 0.011269020987674594,
"num_tokens": 54874.0,
"step": 25
},
{
"entropy": 10.742123413085938,
"epoch": 0.0032104446465835517,
"grad_norm": 3.578125,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.5235,
"mean_token_accuracy": 0.036691234819591044,
"num_tokens": 65085.0,
"step": 30
},
{
"entropy": 10.740595722198487,
"epoch": 0.003745518754347477,
"grad_norm": 3.296875,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.4,
"mean_token_accuracy": 0.03980725221335888,
"num_tokens": 75601.0,
"step": 35
},
{
"entropy": 10.73592758178711,
"epoch": 0.004280592862111403,
"grad_norm": 2.59375,
"learning_rate": 1.95e-05,
"loss": 10.305,
"mean_token_accuracy": 0.04284566193819046,
"num_tokens": 86193.0,
"step": 40
},
{
"entropy": 10.72725419998169,
"epoch": 0.004815666969875327,
"grad_norm": 2.1875,
"learning_rate": 2.2e-05,
"loss": 10.2038,
"mean_token_accuracy": 0.04451024234294891,
"num_tokens": 96957.0,
"step": 45
},
{
"entropy": 10.716819572448731,
"epoch": 0.005350741077639253,
"grad_norm": 1.9921875,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.1336,
"mean_token_accuracy": 0.046419617533683774,
"num_tokens": 108408.0,
"step": 50
},
{
"entropy": 10.70680923461914,
"epoch": 0.005885815185403178,
"grad_norm": 1.5625,
"learning_rate": 2.7e-05,
"loss": 10.0839,
"mean_token_accuracy": 0.04571915306150913,
"num_tokens": 119273.0,
"step": 55
},
{
"entropy": 10.699491024017334,
"epoch": 0.006420889293167103,
"grad_norm": 1.765625,
"learning_rate": 2.95e-05,
"loss": 10.0328,
"mean_token_accuracy": 0.045481448620557786,
"num_tokens": 130817.0,
"step": 60
},
{
"entropy": 10.691621112823487,
"epoch": 0.006955963400931029,
"grad_norm": 1.65625,
"learning_rate": 3.2e-05,
"loss": 9.9967,
"mean_token_accuracy": 0.04715100526809692,
"num_tokens": 142599.0,
"step": 65
},
{
"entropy": 10.683260250091553,
"epoch": 0.007491037508694954,
"grad_norm": 1.734375,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.9367,
"mean_token_accuracy": 0.04639097899198532,
"num_tokens": 153617.0,
"step": 70
},
{
"entropy": 10.678304195404053,
"epoch": 0.00802611161645888,
"grad_norm": 1.546875,
"learning_rate": 3.7e-05,
"loss": 9.9037,
"mean_token_accuracy": 0.046589919924736024,
"num_tokens": 165280.0,
"step": 75
},
{
"entropy": 10.674316120147704,
"epoch": 0.008561185724222806,
"grad_norm": 1.421875,
"learning_rate": 3.95e-05,
"loss": 9.8497,
"mean_token_accuracy": 0.045006733387708664,
"num_tokens": 176896.0,
"step": 80
},
{
"entropy": 10.66965456008911,
"epoch": 0.00909625983198673,
"grad_norm": 1.515625,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.7994,
"mean_token_accuracy": 0.04748752862215042,
"num_tokens": 188000.0,
"step": 85
},
{
"entropy": 10.65854778289795,
"epoch": 0.009631333939750655,
"grad_norm": 1.6640625,
"learning_rate": 4.45e-05,
"loss": 9.7526,
"mean_token_accuracy": 0.04719592090696097,
"num_tokens": 199360.0,
"step": 90
},
{
"entropy": 10.643428230285645,
"epoch": 0.01016640804751458,
"grad_norm": 1.4375,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.719,
"mean_token_accuracy": 0.04324237890541553,
"num_tokens": 210214.0,
"step": 95
},
{
"entropy": 10.625392246246339,
"epoch": 0.010701482155278505,
"grad_norm": 1.59375,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.5936,
"mean_token_accuracy": 0.04677215106785297,
"num_tokens": 220567.0,
"step": 100
},
{
"entropy": 10.598260879516602,
"epoch": 0.011236556263042432,
"grad_norm": 1.4765625,
"learning_rate": 5.2e-05,
"loss": 9.501,
"mean_token_accuracy": 0.05159270279109478,
"num_tokens": 230585.0,
"step": 105
},
{
"entropy": 10.57242784500122,
"epoch": 0.011771630370806356,
"grad_norm": 1.5234375,
"learning_rate": 5.45e-05,
"loss": 9.4414,
"mean_token_accuracy": 0.06068843938410282,
"num_tokens": 241996.0,
"step": 110
},
{
"entropy": 10.52662353515625,
"epoch": 0.012306704478570282,
"grad_norm": 1.4296875,
"learning_rate": 5.7e-05,
"loss": 9.4061,
"mean_token_accuracy": 0.0571962833404541,
"num_tokens": 253348.0,
"step": 115
},
{
"entropy": 10.50029878616333,
"epoch": 0.012841778586334207,
"grad_norm": 1.515625,
"learning_rate": 5.9499999999999996e-05,
"loss": 9.3061,
"mean_token_accuracy": 0.05508606694638729,
"num_tokens": 264239.0,
"step": 120
},
{
"entropy": 10.464248943328858,
"epoch": 0.013376852694098133,
"grad_norm": 1.203125,
"learning_rate": 6.2e-05,
"loss": 9.2082,
"mean_token_accuracy": 0.05605855137109757,
"num_tokens": 274522.0,
"step": 125
},
{
"entropy": 10.388753318786621,
"epoch": 0.013911926801862057,
"grad_norm": 1.5390625,
"learning_rate": 6.450000000000001e-05,
"loss": 9.0975,
"mean_token_accuracy": 0.05725604817271233,
"num_tokens": 284492.0,
"step": 130
},
{
"entropy": 10.300678062438966,
"epoch": 0.014447000909625984,
"grad_norm": 1.2265625,
"learning_rate": 6.7e-05,
"loss": 9.0275,
"mean_token_accuracy": 0.05757311806082725,
"num_tokens": 294796.0,
"step": 135
},
{
"entropy": 10.245001888275146,
"epoch": 0.014982075017389908,
"grad_norm": 1.0625,
"learning_rate": 6.950000000000001e-05,
"loss": 8.9879,
"mean_token_accuracy": 0.054746798053383824,
"num_tokens": 305252.0,
"step": 140
},
{
"entropy": 10.183552169799805,
"epoch": 0.015517149125153834,
"grad_norm": 1.09375,
"learning_rate": 7.2e-05,
"loss": 8.863,
"mean_token_accuracy": 0.0523694459348917,
"num_tokens": 315505.0,
"step": 145
},
{
"entropy": 10.06383113861084,
"epoch": 0.01605222323291776,
"grad_norm": 1.0625,
"learning_rate": 7.45e-05,
"loss": 8.8077,
"mean_token_accuracy": 0.05920022949576378,
"num_tokens": 326997.0,
"step": 150
},
{
"entropy": 9.937602138519287,
"epoch": 0.016587297340681685,
"grad_norm": 0.9921875,
"learning_rate": 7.7e-05,
"loss": 8.7117,
"mean_token_accuracy": 0.05458217263221741,
"num_tokens": 338089.0,
"step": 155
},
{
"entropy": 9.84811716079712,
"epoch": 0.01712237144844561,
"grad_norm": 0.9375,
"learning_rate": 7.950000000000001e-05,
"loss": 8.6529,
"mean_token_accuracy": 0.054475641995668414,
"num_tokens": 348536.0,
"step": 160
},
{
"entropy": 9.70736255645752,
"epoch": 0.017657445556209534,
"grad_norm": 0.984375,
"learning_rate": 8.2e-05,
"loss": 8.6324,
"mean_token_accuracy": 0.056421706452965736,
"num_tokens": 359776.0,
"step": 165
},
{
"entropy": 9.479524993896485,
"epoch": 0.01819251966397346,
"grad_norm": 0.8203125,
"learning_rate": 8.450000000000001e-05,
"loss": 8.526,
"mean_token_accuracy": 0.05786490514874458,
"num_tokens": 370810.0,
"step": 170
},
{
"entropy": 9.431317043304443,
"epoch": 0.018727593771737387,
"grad_norm": 0.8828125,
"learning_rate": 8.7e-05,
"loss": 8.4965,
"mean_token_accuracy": 0.05679596289992332,
"num_tokens": 381788.0,
"step": 175
},
{
"entropy": 9.251888465881347,
"epoch": 0.01926266787950131,
"grad_norm": 0.78125,
"learning_rate": 8.95e-05,
"loss": 8.4589,
"mean_token_accuracy": 0.05575237199664116,
"num_tokens": 391984.0,
"step": 180
},
{
"entropy": 9.08928623199463,
"epoch": 0.019797741987265235,
"grad_norm": 0.78125,
"learning_rate": 9.2e-05,
"loss": 8.4473,
"mean_token_accuracy": 0.055337863974273205,
"num_tokens": 403430.0,
"step": 185
},
{
"entropy": 9.020852279663085,
"epoch": 0.02033281609502916,
"grad_norm": 0.703125,
"learning_rate": 9.45e-05,
"loss": 8.4221,
"mean_token_accuracy": 0.05334745533764362,
"num_tokens": 414467.0,
"step": 190
},
{
"entropy": 8.932736873626709,
"epoch": 0.020867890202793088,
"grad_norm": 0.671875,
"learning_rate": 9.7e-05,
"loss": 8.3858,
"mean_token_accuracy": 0.05892428457736969,
"num_tokens": 425752.0,
"step": 195
},
{
"entropy": 8.8809063911438,
"epoch": 0.02140296431055701,
"grad_norm": 0.66796875,
"learning_rate": 9.95e-05,
"loss": 8.3505,
"mean_token_accuracy": 0.06053958684206009,
"num_tokens": 435908.0,
"step": 200
},
{
"entropy": 8.834921646118165,
"epoch": 0.021938038418320937,
"grad_norm": 0.7421875,
"learning_rate": 0.000102,
"loss": 8.3934,
"mean_token_accuracy": 0.05831273719668388,
"num_tokens": 446424.0,
"step": 205
},
{
"entropy": 8.796385478973388,
"epoch": 0.022473112526084863,
"grad_norm": 0.73046875,
"learning_rate": 0.00010449999999999999,
"loss": 8.3548,
"mean_token_accuracy": 0.06081846654415131,
"num_tokens": 456854.0,
"step": 210
},
{
"entropy": 8.77077407836914,
"epoch": 0.02300818663384879,
"grad_norm": 0.92578125,
"learning_rate": 0.000107,
"loss": 8.3721,
"mean_token_accuracy": 0.06371170245110988,
"num_tokens": 467899.0,
"step": 215
},
{
"entropy": 8.727867889404298,
"epoch": 0.023543260741612712,
"grad_norm": 0.875,
"learning_rate": 0.0001095,
"loss": 8.3147,
"mean_token_accuracy": 0.05818791352212429,
"num_tokens": 478683.0,
"step": 220
},
{
"entropy": 8.737598705291749,
"epoch": 0.02407833484937664,
"grad_norm": 0.7265625,
"learning_rate": 0.000112,
"loss": 8.3471,
"mean_token_accuracy": 0.05725232698023319,
"num_tokens": 488745.0,
"step": 225
},
{
"entropy": 8.692263221740722,
"epoch": 0.024613408957140565,
"grad_norm": 0.69140625,
"learning_rate": 0.0001145,
"loss": 8.4263,
"mean_token_accuracy": 0.05620986633002758,
"num_tokens": 499448.0,
"step": 230
},
{
"entropy": 8.758592700958252,
"epoch": 0.02514848306490449,
"grad_norm": 0.70703125,
"learning_rate": 0.00011700000000000001,
"loss": 8.3538,
"mean_token_accuracy": 0.06146201193332672,
"num_tokens": 510148.0,
"step": 235
},
{
"entropy": 8.709297561645508,
"epoch": 0.025683557172668414,
"grad_norm": 0.80859375,
"learning_rate": 0.00011949999999999999,
"loss": 8.2859,
"mean_token_accuracy": 0.054982627555727956,
"num_tokens": 522093.0,
"step": 240
},
{
"entropy": 8.692859649658203,
"epoch": 0.02621863128043234,
"grad_norm": 0.84375,
"learning_rate": 0.000122,
"loss": 8.3126,
"mean_token_accuracy": 0.061951137706637384,
"num_tokens": 532598.0,
"step": 245
},
{
"entropy": 8.706014442443848,
"epoch": 0.026753705388196266,
"grad_norm": 0.796875,
"learning_rate": 0.0001245,
"loss": 8.2807,
"mean_token_accuracy": 0.05960803255438805,
"num_tokens": 542652.0,
"step": 250
},
{
"entropy": 8.617518043518066,
"epoch": 0.027288779495960192,
"grad_norm": 0.6875,
"learning_rate": 0.000127,
"loss": 8.3078,
"mean_token_accuracy": 0.06188668571412563,
"num_tokens": 552526.0,
"step": 255
},
{
"entropy": 8.682402229309082,
"epoch": 0.027823853603724115,
"grad_norm": 0.73828125,
"learning_rate": 0.0001295,
"loss": 8.3288,
"mean_token_accuracy": 0.06049223951995373,
"num_tokens": 563143.0,
"step": 260
},
{
"entropy": 8.659179401397704,
"epoch": 0.02835892771148804,
"grad_norm": 0.92578125,
"learning_rate": 0.000132,
"loss": 8.2969,
"mean_token_accuracy": 0.060585920512676236,
"num_tokens": 573814.0,
"step": 265
},
{
"entropy": 8.617346477508544,
"epoch": 0.028894001819251967,
"grad_norm": 0.6875,
"learning_rate": 0.00013450000000000002,
"loss": 8.2583,
"mean_token_accuracy": 0.0664129201322794,
"num_tokens": 585367.0,
"step": 270
},
{
"entropy": 8.633615016937256,
"epoch": 0.02942907592701589,
"grad_norm": 2.453125,
"learning_rate": 0.00013700000000000002,
"loss": 8.3346,
"mean_token_accuracy": 0.06159670241177082,
"num_tokens": 595902.0,
"step": 275
},
{
"entropy": 8.68937931060791,
"epoch": 0.029964150034779816,
"grad_norm": 1.0859375,
"learning_rate": 0.0001395,
"loss": 8.229,
"mean_token_accuracy": 0.05921008288860321,
"num_tokens": 607178.0,
"step": 280
},
{
"entropy": 8.568243312835694,
"epoch": 0.030499224142543743,
"grad_norm": 0.8359375,
"learning_rate": 0.00014199999999999998,
"loss": 8.2176,
"mean_token_accuracy": 0.06554121486842632,
"num_tokens": 618604.0,
"step": 285
},
{
"entropy": 8.536308574676514,
"epoch": 0.03103429825030767,
"grad_norm": 0.84375,
"learning_rate": 0.0001445,
"loss": 8.2832,
"mean_token_accuracy": 0.06573930382728577,
"num_tokens": 628111.0,
"step": 290
},
{
"entropy": 8.680528736114502,
"epoch": 0.03156937235807159,
"grad_norm": 0.89453125,
"learning_rate": 0.000147,
"loss": 8.3211,
"mean_token_accuracy": 0.06876323223114014,
"num_tokens": 639358.0,
"step": 295
},
{
"entropy": 8.595854473114013,
"epoch": 0.03210444646583552,
"grad_norm": 0.83203125,
"learning_rate": 0.0001495,
"loss": 8.1939,
"mean_token_accuracy": 0.0666448600590229,
"num_tokens": 649504.0,
"step": 300
},
{
"entropy": 8.553579330444336,
"epoch": 0.032639520573599444,
"grad_norm": 0.96875,
"learning_rate": 0.000152,
"loss": 8.2016,
"mean_token_accuracy": 0.06504374705255031,
"num_tokens": 659924.0,
"step": 305
},
{
"entropy": 8.563442516326905,
"epoch": 0.03317459468136337,
"grad_norm": 0.953125,
"learning_rate": 0.00015450000000000001,
"loss": 8.2419,
"mean_token_accuracy": 0.06233549639582634,
"num_tokens": 671120.0,
"step": 310
},
{
"entropy": 8.586988162994384,
"epoch": 0.033709668789127296,
"grad_norm": 0.87109375,
"learning_rate": 0.000157,
"loss": 8.2807,
"mean_token_accuracy": 0.06455633044242859,
"num_tokens": 681956.0,
"step": 315
},
{
"entropy": 8.551072597503662,
"epoch": 0.03424474289689122,
"grad_norm": 0.74609375,
"learning_rate": 0.0001595,
"loss": 8.2646,
"mean_token_accuracy": 0.0649121094495058,
"num_tokens": 693226.0,
"step": 320
},
{
"entropy": 8.662254047393798,
"epoch": 0.03477981700465514,
"grad_norm": 0.97265625,
"learning_rate": 0.000162,
"loss": 8.2671,
"mean_token_accuracy": 0.06917282193899155,
"num_tokens": 704691.0,
"step": 325
},
{
"entropy": 8.417178916931153,
"epoch": 0.03531489111241907,
"grad_norm": 0.90625,
"learning_rate": 0.00016450000000000001,
"loss": 8.2299,
"mean_token_accuracy": 0.05973314344882965,
"num_tokens": 716000.0,
"step": 330
},
{
"entropy": 8.528615379333496,
"epoch": 0.035849965220182994,
"grad_norm": 0.72265625,
"learning_rate": 0.00016700000000000002,
"loss": 8.1406,
"mean_token_accuracy": 0.06505260914564133,
"num_tokens": 726723.0,
"step": 335
},
{
"entropy": 8.503752326965332,
"epoch": 0.03638503932794692,
"grad_norm": 0.8671875,
"learning_rate": 0.00016950000000000003,
"loss": 8.1503,
"mean_token_accuracy": 0.06767830736935139,
"num_tokens": 736740.0,
"step": 340
},
{
"entropy": 8.526687812805175,
"epoch": 0.03692011343571085,
"grad_norm": 1.6953125,
"learning_rate": 0.00017199999999999998,
"loss": 8.211,
"mean_token_accuracy": 0.06581731289625167,
"num_tokens": 747567.0,
"step": 345
},
{
"entropy": 8.503904342651367,
"epoch": 0.03745518754347477,
"grad_norm": 0.86328125,
"learning_rate": 0.00017449999999999999,
"loss": 8.1534,
"mean_token_accuracy": 0.06844821311533451,
"num_tokens": 757937.0,
"step": 350
},
{
"entropy": 8.430353832244872,
"epoch": 0.0379902616512387,
"grad_norm": 0.86328125,
"learning_rate": 0.000177,
"loss": 8.1759,
"mean_token_accuracy": 0.07079985812306404,
"num_tokens": 768003.0,
"step": 355
},
{
"entropy": 8.572574520111084,
"epoch": 0.03852533575900262,
"grad_norm": 0.796875,
"learning_rate": 0.0001795,
"loss": 8.2185,
"mean_token_accuracy": 0.06302542686462402,
"num_tokens": 779098.0,
"step": 360
},
{
"entropy": 8.457394027709961,
"epoch": 0.039060409866766545,
"grad_norm": 0.859375,
"learning_rate": 0.000182,
"loss": 8.1942,
"mean_token_accuracy": 0.06275034993886948,
"num_tokens": 790002.0,
"step": 365
},
{
"entropy": 8.509715270996093,
"epoch": 0.03959548397453047,
"grad_norm": 0.92578125,
"learning_rate": 0.0001845,
"loss": 8.1721,
"mean_token_accuracy": 0.06606314815580845,
"num_tokens": 800444.0,
"step": 370
},
{
"entropy": 8.493050575256348,
"epoch": 0.0401305580822944,
"grad_norm": 0.75390625,
"learning_rate": 0.000187,
"loss": 8.1172,
"mean_token_accuracy": 0.06942100189626217,
"num_tokens": 810912.0,
"step": 375
},
{
"entropy": 8.490777015686035,
"epoch": 0.04066563219005832,
"grad_norm": 0.75390625,
"learning_rate": 0.0001895,
"loss": 8.2116,
"mean_token_accuracy": 0.0633488953113556,
"num_tokens": 822809.0,
"step": 380
},
{
"entropy": 8.404875659942627,
"epoch": 0.04120070629782225,
"grad_norm": 0.84765625,
"learning_rate": 0.000192,
"loss": 8.1156,
"mean_token_accuracy": 0.06941530928015709,
"num_tokens": 833362.0,
"step": 385
},
{
"entropy": 8.491305160522462,
"epoch": 0.041735780405586176,
"grad_norm": 0.8359375,
"learning_rate": 0.0001945,
"loss": 8.1304,
"mean_token_accuracy": 0.0708375744521618,
"num_tokens": 844239.0,
"step": 390
},
{
"entropy": 8.351639652252198,
"epoch": 0.0422708545133501,
"grad_norm": 0.83203125,
"learning_rate": 0.00019700000000000002,
"loss": 8.1313,
"mean_token_accuracy": 0.06419948525726796,
"num_tokens": 855178.0,
"step": 395
},
{
"entropy": 8.44887342453003,
"epoch": 0.04280592862111402,
"grad_norm": 0.90625,
"learning_rate": 0.00019950000000000002,
"loss": 8.1241,
"mean_token_accuracy": 0.06882764175534248,
"num_tokens": 866120.0,
"step": 400
},
{
"entropy": 8.451717185974122,
"epoch": 0.04334100272887795,
"grad_norm": 0.7890625,
"learning_rate": 0.000202,
"loss": 8.1179,
"mean_token_accuracy": 0.07001868635416031,
"num_tokens": 876569.0,
"step": 405
},
{
"entropy": 8.462645149230957,
"epoch": 0.043876076836641874,
"grad_norm": 0.921875,
"learning_rate": 0.00020449999999999998,
"loss": 8.1782,
"mean_token_accuracy": 0.05914541408419609,
"num_tokens": 887580.0,
"step": 410
},
{
"entropy": 8.390836906433105,
"epoch": 0.0444111509444058,
"grad_norm": 0.8515625,
"learning_rate": 0.000207,
"loss": 8.1122,
"mean_token_accuracy": 0.06334667056798934,
"num_tokens": 897922.0,
"step": 415
},
{
"entropy": 8.458688259124756,
"epoch": 0.044946225052169726,
"grad_norm": 0.7890625,
"learning_rate": 0.0002095,
"loss": 8.0675,
"mean_token_accuracy": 0.06874303668737411,
"num_tokens": 908522.0,
"step": 420
},
{
"entropy": 8.417051124572755,
"epoch": 0.04548129915993365,
"grad_norm": 0.81640625,
"learning_rate": 0.000212,
"loss": 8.0935,
"mean_token_accuracy": 0.06554699912667275,
"num_tokens": 918253.0,
"step": 425
},
{
"entropy": 8.360801124572754,
"epoch": 0.04601637326769758,
"grad_norm": 0.73828125,
"learning_rate": 0.0002145,
"loss": 8.0775,
"mean_token_accuracy": 0.07234423086047173,
"num_tokens": 929150.0,
"step": 430
},
{
"entropy": 8.457652568817139,
"epoch": 0.0465514473754615,
"grad_norm": 0.78125,
"learning_rate": 0.00021700000000000002,
"loss": 8.1172,
"mean_token_accuracy": 0.06926538720726967,
"num_tokens": 941041.0,
"step": 435
},
{
"entropy": 8.334060573577881,
"epoch": 0.047086521483225424,
"grad_norm": 1.484375,
"learning_rate": 0.0002195,
"loss": 8.0663,
"mean_token_accuracy": 0.07295753248035908,
"num_tokens": 952714.0,
"step": 440
},
{
"entropy": 8.448612403869628,
"epoch": 0.04762159559098935,
"grad_norm": 0.9296875,
"learning_rate": 0.000222,
"loss": 8.1236,
"mean_token_accuracy": 0.06220442913472653,
"num_tokens": 964330.0,
"step": 445
},
{
"entropy": 8.33154468536377,
"epoch": 0.04815666969875328,
"grad_norm": 0.94921875,
"learning_rate": 0.0002245,
"loss": 8.0483,
"mean_token_accuracy": 0.06833712309598923,
"num_tokens": 975257.0,
"step": 450
},
{
"entropy": 8.419320106506348,
"epoch": 0.0486917438065172,
"grad_norm": 0.8203125,
"learning_rate": 0.00022700000000000002,
"loss": 8.0769,
"mean_token_accuracy": 0.06652066558599472,
"num_tokens": 986335.0,
"step": 455
},
{
"entropy": 8.301145553588867,
"epoch": 0.04922681791428113,
"grad_norm": 0.75,
"learning_rate": 0.00022950000000000002,
"loss": 7.9671,
"mean_token_accuracy": 0.07098020166158676,
"num_tokens": 997521.0,
"step": 460
},
{
"entropy": 8.32881212234497,
"epoch": 0.049761892022045055,
"grad_norm": 0.8359375,
"learning_rate": 0.00023200000000000003,
"loss": 8.0134,
"mean_token_accuracy": 0.06557687968015671,
"num_tokens": 1008383.0,
"step": 465
},
{
"entropy": 8.357664775848388,
"epoch": 0.05029696612980898,
"grad_norm": 0.7890625,
"learning_rate": 0.00023449999999999998,
"loss": 8.0778,
"mean_token_accuracy": 0.0682313721626997,
"num_tokens": 1020552.0,
"step": 470
},
{
"entropy": 8.34301462173462,
"epoch": 0.0508320402375729,
"grad_norm": 0.78125,
"learning_rate": 0.000237,
"loss": 8.0512,
"mean_token_accuracy": 0.07024195492267608,
"num_tokens": 1031127.0,
"step": 475
},
{
"entropy": 8.236884784698486,
"epoch": 0.05136711434533683,
"grad_norm": 0.74609375,
"learning_rate": 0.0002395,
"loss": 7.9657,
"mean_token_accuracy": 0.07320996709167957,
"num_tokens": 1043209.0,
"step": 480
},
{
"entropy": 8.421635818481445,
"epoch": 0.05190218845310075,
"grad_norm": 0.7890625,
"learning_rate": 0.000242,
"loss": 8.1023,
"mean_token_accuracy": 0.06352589540183544,
"num_tokens": 1054032.0,
"step": 485
},
{
"entropy": 8.233527374267577,
"epoch": 0.05243726256086468,
"grad_norm": 0.7734375,
"learning_rate": 0.0002445,
"loss": 7.9292,
"mean_token_accuracy": 0.07238206267356873,
"num_tokens": 1064565.0,
"step": 490
},
{
"entropy": 8.403174209594727,
"epoch": 0.052972336668628606,
"grad_norm": 0.8203125,
"learning_rate": 0.000247,
"loss": 8.0765,
"mean_token_accuracy": 0.07412495836615562,
"num_tokens": 1076352.0,
"step": 495
},
{
"entropy": 8.31511640548706,
"epoch": 0.05350741077639253,
"grad_norm": 0.9453125,
"learning_rate": 0.0002495,
"loss": 8.0,
"mean_token_accuracy": 0.0721416525542736,
"num_tokens": 1087697.0,
"step": 500
},
{
"entropy": 8.30199728012085,
"epoch": 0.05404248488415646,
"grad_norm": 0.81640625,
"learning_rate": 0.000252,
"loss": 7.9346,
"mean_token_accuracy": 0.07342750616371632,
"num_tokens": 1098056.0,
"step": 505
},
{
"entropy": 8.233750343322754,
"epoch": 0.054577558991920384,
"grad_norm": 0.7578125,
"learning_rate": 0.0002545,
"loss": 7.9036,
"mean_token_accuracy": 0.08062965869903564,
"num_tokens": 1109219.0,
"step": 510
},
{
"entropy": 8.323031330108643,
"epoch": 0.055112633099684304,
"grad_norm": 0.80859375,
"learning_rate": 0.000257,
"loss": 7.987,
"mean_token_accuracy": 0.06857425346970558,
"num_tokens": 1120248.0,
"step": 515
},
{
"entropy": 8.24481725692749,
"epoch": 0.05564770720744823,
"grad_norm": 0.94140625,
"learning_rate": 0.0002595,
"loss": 8.0086,
"mean_token_accuracy": 0.07023427784442901,
"num_tokens": 1131278.0,
"step": 520
},
{
"entropy": 8.313823032379151,
"epoch": 0.056182781315212156,
"grad_norm": 0.95703125,
"learning_rate": 0.000262,
"loss": 8.0227,
"mean_token_accuracy": 0.06274679154157639,
"num_tokens": 1143147.0,
"step": 525
},
{
"entropy": 8.287971687316894,
"epoch": 0.05671785542297608,
"grad_norm": 0.8046875,
"learning_rate": 0.00026450000000000003,
"loss": 7.9215,
"mean_token_accuracy": 0.07099093310534954,
"num_tokens": 1154316.0,
"step": 530
},
{
"entropy": 8.246630001068116,
"epoch": 0.05725292953074001,
"grad_norm": 0.91015625,
"learning_rate": 0.00026700000000000004,
"loss": 7.9763,
"mean_token_accuracy": 0.07050252221524715,
"num_tokens": 1164696.0,
"step": 535
},
{
"entropy": 8.212139987945557,
"epoch": 0.057788003638503935,
"grad_norm": 0.921875,
"learning_rate": 0.00026950000000000005,
"loss": 7.9621,
"mean_token_accuracy": 0.07535928599536419,
"num_tokens": 1176129.0,
"step": 540
},
{
"entropy": 8.316623306274414,
"epoch": 0.05832307774626786,
"grad_norm": 0.93359375,
"learning_rate": 0.00027200000000000005,
"loss": 7.9772,
"mean_token_accuracy": 0.07082564979791642,
"num_tokens": 1186499.0,
"step": 545
},
{
"entropy": 8.1876971244812,
"epoch": 0.05885815185403178,
"grad_norm": 0.9609375,
"learning_rate": 0.0002745,
"loss": 7.9314,
"mean_token_accuracy": 0.07009159363806247,
"num_tokens": 1196841.0,
"step": 550
},
{
"entropy": 8.286906814575195,
"epoch": 0.059393225961795706,
"grad_norm": 0.82421875,
"learning_rate": 0.000277,
"loss": 7.8784,
"mean_token_accuracy": 0.07358456589281559,
"num_tokens": 1207647.0,
"step": 555
},
{
"entropy": 8.175436544418336,
"epoch": 0.05992830006955963,
"grad_norm": 1.0,
"learning_rate": 0.0002795,
"loss": 7.8812,
"mean_token_accuracy": 0.07417330220341682,
"num_tokens": 1217072.0,
"step": 560
},
{
"entropy": 8.201017284393311,
"epoch": 0.06046337417732356,
"grad_norm": 1.0,
"learning_rate": 0.00028199999999999997,
"loss": 7.9853,
"mean_token_accuracy": 0.07028395347297192,
"num_tokens": 1228564.0,
"step": 565
},
{
"entropy": 8.22887988090515,
"epoch": 0.060998448285087485,
"grad_norm": 0.86328125,
"learning_rate": 0.0002845,
"loss": 7.9124,
"mean_token_accuracy": 0.072603465244174,
"num_tokens": 1238755.0,
"step": 570
},
{
"entropy": 8.181446075439453,
"epoch": 0.06153352239285141,
"grad_norm": 0.828125,
"learning_rate": 0.000287,
"loss": 7.8745,
"mean_token_accuracy": 0.07408057898283005,
"num_tokens": 1248968.0,
"step": 575
},
{
"entropy": 8.16532106399536,
"epoch": 0.06206859650061534,
"grad_norm": 0.87109375,
"learning_rate": 0.0002895,
"loss": 7.8461,
"mean_token_accuracy": 0.07231933623552322,
"num_tokens": 1259256.0,
"step": 580
},
{
"entropy": 8.09881591796875,
"epoch": 0.06260367060837926,
"grad_norm": 0.984375,
"learning_rate": 0.000292,
"loss": 7.8866,
"mean_token_accuracy": 0.07566695511341096,
"num_tokens": 1270469.0,
"step": 585
},
{
"entropy": 8.216452407836915,
"epoch": 0.06313874471614318,
"grad_norm": 0.85546875,
"learning_rate": 0.0002945,
"loss": 7.8552,
"mean_token_accuracy": 0.07371087111532688,
"num_tokens": 1280558.0,
"step": 590
},
{
"entropy": 8.129891967773437,
"epoch": 0.06367381882390712,
"grad_norm": 0.9609375,
"learning_rate": 0.000297,
"loss": 7.8999,
"mean_token_accuracy": 0.06837657615542411,
"num_tokens": 1291372.0,
"step": 595
},
{
"entropy": 8.213308429718017,
"epoch": 0.06420889293167104,
"grad_norm": 0.8515625,
"learning_rate": 0.0002995,
"loss": 7.9753,
"mean_token_accuracy": 0.06709825992584229,
"num_tokens": 1302521.0,
"step": 600
},
{
"entropy": 8.23453779220581,
"epoch": 0.06474396703943497,
"grad_norm": 0.9609375,
"learning_rate": 0.000302,
"loss": 7.8081,
"mean_token_accuracy": 0.07309315800666809,
"num_tokens": 1313632.0,
"step": 605
},
{
"entropy": 8.05699348449707,
"epoch": 0.06527904114719889,
"grad_norm": 0.984375,
"learning_rate": 0.0003045,
"loss": 7.8949,
"mean_token_accuracy": 0.07557445764541626,
"num_tokens": 1325247.0,
"step": 610
},
{
"entropy": 8.185565948486328,
"epoch": 0.06581411525496281,
"grad_norm": 0.890625,
"learning_rate": 0.000307,
"loss": 7.8687,
"mean_token_accuracy": 0.07852845974266529,
"num_tokens": 1336026.0,
"step": 615
},
{
"entropy": 8.095359706878662,
"epoch": 0.06634918936272674,
"grad_norm": 0.8671875,
"learning_rate": 0.0003095,
"loss": 7.7947,
"mean_token_accuracy": 0.07415896467864513,
"num_tokens": 1346920.0,
"step": 620
},
{
"entropy": 8.091093444824219,
"epoch": 0.06688426347049066,
"grad_norm": 1.078125,
"learning_rate": 0.000312,
"loss": 7.9316,
"mean_token_accuracy": 0.0739424929022789,
"num_tokens": 1359752.0,
"step": 625
},
{
"entropy": 8.266688632965089,
"epoch": 0.06741933757825459,
"grad_norm": 0.84765625,
"learning_rate": 0.0003145,
"loss": 7.938,
"mean_token_accuracy": 0.07053884975612164,
"num_tokens": 1370874.0,
"step": 630
},
{
"entropy": 8.154129219055175,
"epoch": 0.06795441168601851,
"grad_norm": 0.8671875,
"learning_rate": 0.000317,
"loss": 7.8375,
"mean_token_accuracy": 0.07769538760185242,
"num_tokens": 1382316.0,
"step": 635
},
{
"entropy": 8.086437702178955,
"epoch": 0.06848948579378245,
"grad_norm": 0.8671875,
"learning_rate": 0.0003195,
"loss": 7.8429,
"mean_token_accuracy": 0.0740839570760727,
"num_tokens": 1394656.0,
"step": 640
},
{
"entropy": 8.045439958572388,
"epoch": 0.06902455990154636,
"grad_norm": 1.0859375,
"learning_rate": 0.000322,
"loss": 7.8192,
"mean_token_accuracy": 0.07973718792200088,
"num_tokens": 1405061.0,
"step": 645
},
{
"entropy": 8.166390562057495,
"epoch": 0.06955963400931028,
"grad_norm": 0.87890625,
"learning_rate": 0.00032450000000000003,
"loss": 7.7751,
"mean_token_accuracy": 0.08227450475096702,
"num_tokens": 1415573.0,
"step": 650
},
{
"entropy": 8.016403913497925,
"epoch": 0.07009470811707422,
"grad_norm": 1.0078125,
"learning_rate": 0.00032700000000000003,
"loss": 7.8142,
"mean_token_accuracy": 0.07414669394493104,
"num_tokens": 1426103.0,
"step": 655
},
{
"entropy": 8.006370496749877,
"epoch": 0.07062978222483814,
"grad_norm": 1.1640625,
"learning_rate": 0.00032950000000000004,
"loss": 7.8224,
"mean_token_accuracy": 0.07035662531852722,
"num_tokens": 1437256.0,
"step": 660
},
{
"entropy": 8.223912525177003,
"epoch": 0.07116485633260207,
"grad_norm": 0.9453125,
"learning_rate": 0.00033200000000000005,
"loss": 7.7861,
"mean_token_accuracy": 0.07515646256506443,
"num_tokens": 1446975.0,
"step": 665
},
{
"entropy": 8.001112079620361,
"epoch": 0.07169993044036599,
"grad_norm": 0.9296875,
"learning_rate": 0.00033450000000000005,
"loss": 7.8084,
"mean_token_accuracy": 0.07363461852073669,
"num_tokens": 1457751.0,
"step": 670
},
{
"entropy": 8.052605438232423,
"epoch": 0.07223500454812992,
"grad_norm": 0.9375,
"learning_rate": 0.000337,
"loss": 7.7689,
"mean_token_accuracy": 0.07661298438906669,
"num_tokens": 1468186.0,
"step": 675
},
{
"entropy": 8.094243955612182,
"epoch": 0.07277007865589384,
"grad_norm": 0.83984375,
"learning_rate": 0.0003395,
"loss": 7.8108,
"mean_token_accuracy": 0.07768909595906734,
"num_tokens": 1479780.0,
"step": 680
},
{
"entropy": 8.057337331771851,
"epoch": 0.07330515276365776,
"grad_norm": 0.8984375,
"learning_rate": 0.000342,
"loss": 7.7533,
"mean_token_accuracy": 0.0778307005763054,
"num_tokens": 1490565.0,
"step": 685
},
{
"entropy": 8.113437938690186,
"epoch": 0.0738402268714217,
"grad_norm": 0.8359375,
"learning_rate": 0.00034449999999999997,
"loss": 7.7855,
"mean_token_accuracy": 0.0767733097076416,
"num_tokens": 1501552.0,
"step": 690
},
{
"entropy": 8.02952961921692,
"epoch": 0.07437530097918561,
"grad_norm": 0.96484375,
"learning_rate": 0.000347,
"loss": 7.7837,
"mean_token_accuracy": 0.070881437510252,
"num_tokens": 1513311.0,
"step": 695
},
{
"entropy": 8.041796684265137,
"epoch": 0.07491037508694955,
"grad_norm": 0.93359375,
"learning_rate": 0.0003495,
"loss": 7.8081,
"mean_token_accuracy": 0.07614933587610721,
"num_tokens": 1524142.0,
"step": 700
},
{
"entropy": 8.079885721206665,
"epoch": 0.07544544919471347,
"grad_norm": 0.8984375,
"learning_rate": 0.000352,
"loss": 7.7634,
"mean_token_accuracy": 0.08070720061659813,
"num_tokens": 1535571.0,
"step": 705
},
{
"entropy": 8.080323457717896,
"epoch": 0.0759805233024774,
"grad_norm": 0.8203125,
"learning_rate": 0.0003545,
"loss": 7.7663,
"mean_token_accuracy": 0.07677763924002648,
"num_tokens": 1546528.0,
"step": 710
},
{
"entropy": 7.932172536849976,
"epoch": 0.07651559741024132,
"grad_norm": 0.8203125,
"learning_rate": 0.000357,
"loss": 7.7928,
"mean_token_accuracy": 0.06920730993151665,
"num_tokens": 1557264.0,
"step": 715
},
{
"entropy": 8.126523494720459,
"epoch": 0.07705067151800524,
"grad_norm": 0.98046875,
"learning_rate": 0.0003595,
"loss": 7.7492,
"mean_token_accuracy": 0.07938187122344971,
"num_tokens": 1566773.0,
"step": 720
},
{
"entropy": 7.902943658828735,
"epoch": 0.07758574562576917,
"grad_norm": 0.890625,
"learning_rate": 0.000362,
"loss": 7.7072,
"mean_token_accuracy": 0.0833115503191948,
"num_tokens": 1578173.0,
"step": 725
},
{
"entropy": 8.02185640335083,
"epoch": 0.07812081973353309,
"grad_norm": 0.95703125,
"learning_rate": 0.0003645,
"loss": 7.6767,
"mean_token_accuracy": 0.08211550787091255,
"num_tokens": 1588537.0,
"step": 730
},
{
"entropy": 7.918304061889648,
"epoch": 0.07865589384129702,
"grad_norm": 0.984375,
"learning_rate": 0.000367,
"loss": 7.695,
"mean_token_accuracy": 0.08645991012454032,
"num_tokens": 1599159.0,
"step": 735
},
{
"entropy": 7.982047033309937,
"epoch": 0.07919096794906094,
"grad_norm": 0.94921875,
"learning_rate": 0.0003695,
"loss": 7.7685,
"mean_token_accuracy": 0.07947736382484435,
"num_tokens": 1610378.0,
"step": 740
},
{
"entropy": 7.9871502876281735,
"epoch": 0.07972604205682488,
"grad_norm": 1.03125,
"learning_rate": 0.000372,
"loss": 7.7684,
"mean_token_accuracy": 0.0792300634086132,
"num_tokens": 1621215.0,
"step": 745
},
{
"entropy": 7.980308580398559,
"epoch": 0.0802611161645888,
"grad_norm": 1.046875,
"learning_rate": 0.0003745,
"loss": 7.7133,
"mean_token_accuracy": 0.07562910728156566,
"num_tokens": 1631170.0,
"step": 750
},
{
"entropy": 8.027432918548584,
"epoch": 0.08079619027235273,
"grad_norm": 0.9765625,
"learning_rate": 0.000377,
"loss": 7.7105,
"mean_token_accuracy": 0.0776789553463459,
"num_tokens": 1641897.0,
"step": 755
},
{
"entropy": 7.974108934402466,
"epoch": 0.08133126438011665,
"grad_norm": 0.875,
"learning_rate": 0.0003795,
"loss": 7.7762,
"mean_token_accuracy": 0.07812840193510055,
"num_tokens": 1653592.0,
"step": 760
},
{
"entropy": 7.944252681732178,
"epoch": 0.08186633848788057,
"grad_norm": 0.921875,
"learning_rate": 0.000382,
"loss": 7.7089,
"mean_token_accuracy": 0.0731413435190916,
"num_tokens": 1665760.0,
"step": 765
},
{
"entropy": 7.967342901229858,
"epoch": 0.0824014125956445,
"grad_norm": 0.88671875,
"learning_rate": 0.0003845,
"loss": 7.7065,
"mean_token_accuracy": 0.08256332576274872,
"num_tokens": 1676211.0,
"step": 770
},
{
"entropy": 7.979905366897583,
"epoch": 0.08293648670340842,
"grad_norm": 0.94921875,
"learning_rate": 0.00038700000000000003,
"loss": 7.7411,
"mean_token_accuracy": 0.0778872437775135,
"num_tokens": 1686405.0,
"step": 775
},
{
"entropy": 8.09218454360962,
"epoch": 0.08347156081117235,
"grad_norm": 0.91796875,
"learning_rate": 0.00038950000000000003,
"loss": 7.7171,
"mean_token_accuracy": 0.07889893390238285,
"num_tokens": 1697340.0,
"step": 780
},
{
"entropy": 7.979829883575439,
"epoch": 0.08400663491893627,
"grad_norm": 1.0234375,
"learning_rate": 0.00039200000000000004,
"loss": 7.6184,
"mean_token_accuracy": 0.08763319551944733,
"num_tokens": 1707250.0,
"step": 785
},
{
"entropy": 7.876405382156372,
"epoch": 0.0845417090267002,
"grad_norm": 1.015625,
"learning_rate": 0.00039450000000000005,
"loss": 7.5864,
"mean_token_accuracy": 0.08063115775585175,
"num_tokens": 1718333.0,
"step": 790
},
{
"entropy": 7.940855836868286,
"epoch": 0.08507678313446412,
"grad_norm": 0.9609375,
"learning_rate": 0.00039700000000000005,
"loss": 7.6984,
"mean_token_accuracy": 0.07467846013605595,
"num_tokens": 1730520.0,
"step": 795
},
{
"entropy": 7.936916255950928,
"epoch": 0.08561185724222804,
"grad_norm": 1.015625,
"learning_rate": 0.0003995,
"loss": 7.7522,
"mean_token_accuracy": 0.07317487969994545,
"num_tokens": 1741898.0,
"step": 800
},
{
"entropy": 7.9014006614685055,
"epoch": 0.08614693134999198,
"grad_norm": 0.94140625,
"learning_rate": 0.000402,
"loss": 7.6984,
"mean_token_accuracy": 0.08368097692728042,
"num_tokens": 1751671.0,
"step": 805
},
{
"entropy": 7.829085969924927,
"epoch": 0.0866820054577559,
"grad_norm": 0.91015625,
"learning_rate": 0.0004045,
"loss": 7.7182,
"mean_token_accuracy": 0.08191005513072014,
"num_tokens": 1762076.0,
"step": 810
},
{
"entropy": 8.061158800125122,
"epoch": 0.08721707956551983,
"grad_norm": 0.8515625,
"learning_rate": 0.00040699999999999997,
"loss": 7.685,
"mean_token_accuracy": 0.07685056664049625,
"num_tokens": 1772666.0,
"step": 815
},
{
"entropy": 7.887774229049683,
"epoch": 0.08775215367328375,
"grad_norm": 0.953125,
"learning_rate": 0.0004095,
"loss": 7.6979,
"mean_token_accuracy": 0.08384055867791176,
"num_tokens": 1783621.0,
"step": 820
},
{
"entropy": 7.9562437534332275,
"epoch": 0.08828722778104768,
"grad_norm": 1.390625,
"learning_rate": 0.000412,
"loss": 7.7515,
"mean_token_accuracy": 0.07828780002892018,
"num_tokens": 1794654.0,
"step": 825
},
{
"entropy": 7.825522232055664,
"epoch": 0.0888223018888116,
"grad_norm": 0.86328125,
"learning_rate": 0.0004145,
"loss": 7.6368,
"mean_token_accuracy": 0.08319875784218311,
"num_tokens": 1805853.0,
"step": 830
},
{
"entropy": 7.7365336418151855,
"epoch": 0.08935737599657552,
"grad_norm": 0.86328125,
"learning_rate": 0.000417,
"loss": 7.6227,
"mean_token_accuracy": 0.0816656306385994,
"num_tokens": 1817006.0,
"step": 835
},
{
"entropy": 8.006308937072754,
"epoch": 0.08989245010433945,
"grad_norm": 1.046875,
"learning_rate": 0.0004195,
"loss": 7.623,
"mean_token_accuracy": 0.0778016533702612,
"num_tokens": 1827166.0,
"step": 840
},
{
"entropy": 8.007753992080689,
"epoch": 0.09042752421210337,
"grad_norm": 1.1796875,
"learning_rate": 0.000422,
"loss": 7.7146,
"mean_token_accuracy": 0.08085944950580597,
"num_tokens": 1837182.0,
"step": 845
},
{
"entropy": 7.7709431648254395,
"epoch": 0.0909625983198673,
"grad_norm": 0.8515625,
"learning_rate": 0.0004245,
"loss": 7.6545,
"mean_token_accuracy": 0.08275943025946617,
"num_tokens": 1849034.0,
"step": 850
},
{
"entropy": 7.842226839065551,
"epoch": 0.09149767242763122,
"grad_norm": 0.90625,
"learning_rate": 0.000427,
"loss": 7.5383,
"mean_token_accuracy": 0.08894663602113724,
"num_tokens": 1859033.0,
"step": 855
},
{
"entropy": 7.813801717758179,
"epoch": 0.09203274653539516,
"grad_norm": 0.95703125,
"learning_rate": 0.0004295,
"loss": 7.6562,
"mean_token_accuracy": 0.07978184968233108,
"num_tokens": 1870274.0,
"step": 860
},
{
"entropy": 7.786420011520386,
"epoch": 0.09256782064315908,
"grad_norm": 1.1953125,
"learning_rate": 0.000432,
"loss": 7.5977,
"mean_token_accuracy": 0.085463797301054,
"num_tokens": 1881306.0,
"step": 865
},
{
"entropy": 7.810677576065063,
"epoch": 0.093102894750923,
"grad_norm": 0.87890625,
"learning_rate": 0.0004345,
"loss": 7.5858,
"mean_token_accuracy": 0.07887605726718902,
"num_tokens": 1892464.0,
"step": 870
},
{
"entropy": 7.883417320251465,
"epoch": 0.09363796885868693,
"grad_norm": 0.875,
"learning_rate": 0.000437,
"loss": 7.6601,
"mean_token_accuracy": 0.07597660236060619,
"num_tokens": 1903743.0,
"step": 875
},
{
"entropy": 7.838293743133545,
"epoch": 0.09417304296645085,
"grad_norm": 0.91796875,
"learning_rate": 0.0004395,
"loss": 7.6112,
"mean_token_accuracy": 0.08703320100903511,
"num_tokens": 1915253.0,
"step": 880
},
{
"entropy": 7.873800992965698,
"epoch": 0.09470811707421478,
"grad_norm": 1.15625,
"learning_rate": 0.000442,
"loss": 7.5604,
"mean_token_accuracy": 0.08356410227715969,
"num_tokens": 1926113.0,
"step": 885
},
{
"entropy": 7.816686487197876,
"epoch": 0.0952431911819787,
"grad_norm": 0.85546875,
"learning_rate": 0.0004445,
"loss": 7.6579,
"mean_token_accuracy": 0.08339532688260079,
"num_tokens": 1937718.0,
"step": 890
},
{
"entropy": 7.857083654403686,
"epoch": 0.09577826528974263,
"grad_norm": 0.875,
"learning_rate": 0.000447,
"loss": 7.6419,
"mean_token_accuracy": 0.07871352508664131,
"num_tokens": 1948307.0,
"step": 895
},
{
"entropy": 7.7142290592193605,
"epoch": 0.09631333939750655,
"grad_norm": 0.9140625,
"learning_rate": 0.00044950000000000003,
"loss": 7.4658,
"mean_token_accuracy": 0.09317997246980667,
"num_tokens": 1957861.0,
"step": 900
},
{
"entropy": 7.769102573394775,
"epoch": 0.09684841350527049,
"grad_norm": 0.89453125,
"learning_rate": 0.00045200000000000004,
"loss": 7.5118,
"mean_token_accuracy": 0.09313379377126693,
"num_tokens": 1967984.0,
"step": 905
},
{
"entropy": 7.776102876663208,
"epoch": 0.0973834876130344,
"grad_norm": 0.9140625,
"learning_rate": 0.00045450000000000004,
"loss": 7.531,
"mean_token_accuracy": 0.08627012819051742,
"num_tokens": 1978608.0,
"step": 910
},
{
"entropy": 7.774751234054565,
"epoch": 0.09791856172079832,
"grad_norm": 0.9609375,
"learning_rate": 0.00045700000000000005,
"loss": 7.5571,
"mean_token_accuracy": 0.08207484185695649,
"num_tokens": 1989355.0,
"step": 915
},
{
"entropy": 7.759360122680664,
"epoch": 0.09845363582856226,
"grad_norm": 0.89453125,
"learning_rate": 0.00045950000000000006,
"loss": 7.5519,
"mean_token_accuracy": 0.09099748730659485,
"num_tokens": 1999754.0,
"step": 920
},
{
"entropy": 7.7610972881317135,
"epoch": 0.09898870993632618,
"grad_norm": 0.99609375,
"learning_rate": 0.000462,
"loss": 7.5582,
"mean_token_accuracy": 0.08363260142505169,
"num_tokens": 2010877.0,
"step": 925
},
{
"entropy": 7.894351291656494,
"epoch": 0.09952378404409011,
"grad_norm": 0.85546875,
"learning_rate": 0.0004645,
"loss": 7.6443,
"mean_token_accuracy": 0.08178722113370895,
"num_tokens": 2022301.0,
"step": 930
},
{
"entropy": 7.720238924026489,
"epoch": 0.10005885815185403,
"grad_norm": 0.95703125,
"learning_rate": 0.000467,
"loss": 7.5376,
"mean_token_accuracy": 0.08469678834080696,
"num_tokens": 2033537.0,
"step": 935
},
{
"entropy": 7.747068881988525,
"epoch": 0.10059393225961796,
"grad_norm": 0.91015625,
"learning_rate": 0.0004695,
"loss": 7.4595,
"mean_token_accuracy": 0.08593985140323639,
"num_tokens": 2044360.0,
"step": 940
},
{
"entropy": 7.767273759841919,
"epoch": 0.10112900636738188,
"grad_norm": 0.8671875,
"learning_rate": 0.000472,
"loss": 7.5692,
"mean_token_accuracy": 0.08553824871778488,
"num_tokens": 2055000.0,
"step": 945
},
{
"entropy": 7.675411987304687,
"epoch": 0.1016640804751458,
"grad_norm": 0.8515625,
"learning_rate": 0.0004745,
"loss": 7.5747,
"mean_token_accuracy": 0.08887513428926468,
"num_tokens": 2064807.0,
"step": 950
},
{
"entropy": 7.846766614913941,
"epoch": 0.10219915458290973,
"grad_norm": 0.96484375,
"learning_rate": 0.000477,
"loss": 7.5338,
"mean_token_accuracy": 0.08006390556693077,
"num_tokens": 2075618.0,
"step": 955
},
{
"entropy": 7.665963554382325,
"epoch": 0.10273422869067365,
"grad_norm": 0.88671875,
"learning_rate": 0.0004795,
"loss": 7.5318,
"mean_token_accuracy": 0.08242494091391564,
"num_tokens": 2086226.0,
"step": 960
},
{
"entropy": 7.758722734451294,
"epoch": 0.10326930279843759,
"grad_norm": 0.83984375,
"learning_rate": 0.000482,
"loss": 7.5213,
"mean_token_accuracy": 0.08128740042448043,
"num_tokens": 2097222.0,
"step": 965
},
{
"entropy": 7.718167400360107,
"epoch": 0.1038043769062015,
"grad_norm": 0.8515625,
"learning_rate": 0.0004845,
"loss": 7.566,
"mean_token_accuracy": 0.08480455204844475,
"num_tokens": 2108919.0,
"step": 970
},
{
"entropy": 7.67155704498291,
"epoch": 0.10433945101396544,
"grad_norm": 0.96484375,
"learning_rate": 0.000487,
"loss": 7.5121,
"mean_token_accuracy": 0.08769758641719819,
"num_tokens": 2120215.0,
"step": 975
},
{
"entropy": 7.774061918258667,
"epoch": 0.10487452512172936,
"grad_norm": 0.94140625,
"learning_rate": 0.0004895,
"loss": 7.5377,
"mean_token_accuracy": 0.08079898655414582,
"num_tokens": 2131503.0,
"step": 980
},
{
"entropy": 7.695115470886231,
"epoch": 0.10540959922949328,
"grad_norm": 0.921875,
"learning_rate": 0.000492,
"loss": 7.4754,
"mean_token_accuracy": 0.09231993854045868,
"num_tokens": 2141551.0,
"step": 985
},
{
"entropy": 7.619196653366089,
"epoch": 0.10594467333725721,
"grad_norm": 0.94140625,
"learning_rate": 0.0004945,
"loss": 7.4805,
"mean_token_accuracy": 0.0900322899222374,
"num_tokens": 2151821.0,
"step": 990
},
{
"entropy": 7.718583822250366,
"epoch": 0.10647974744502113,
"grad_norm": 1.0078125,
"learning_rate": 0.000497,
"loss": 7.5276,
"mean_token_accuracy": 0.08750699013471604,
"num_tokens": 2162474.0,
"step": 995
},
{
"entropy": 7.677374029159546,
"epoch": 0.10701482155278506,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995,
"loss": 7.5581,
"mean_token_accuracy": 0.08569178096950054,
"num_tokens": 2173985.0,
"step": 1000
},
{
"entropy": 7.77434720993042,
"epoch": 0.10754989566054898,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999999979210101,
"loss": 7.588,
"mean_token_accuracy": 0.08118958622217179,
"num_tokens": 2186432.0,
"step": 1005
},
{
"entropy": 7.720741939544678,
"epoch": 0.10808496976831292,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999999894751138,
"loss": 7.4482,
"mean_token_accuracy": 0.08467482179403304,
"num_tokens": 2197085.0,
"step": 1010
},
{
"entropy": 7.661213302612305,
"epoch": 0.10862004387607684,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999999745323744,
"loss": 7.4699,
"mean_token_accuracy": 0.09365503117442131,
"num_tokens": 2207162.0,
"step": 1015
},
{
"entropy": 7.673577928543091,
"epoch": 0.10915511798384077,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999999530927923,
"loss": 7.5611,
"mean_token_accuracy": 0.08068385906517506,
"num_tokens": 2218714.0,
"step": 1020
},
{
"entropy": 7.6879801750183105,
"epoch": 0.10969019209160469,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999999251563682,
"loss": 7.48,
"mean_token_accuracy": 0.08622352927923202,
"num_tokens": 2229711.0,
"step": 1025
},
{
"entropy": 7.641462993621826,
"epoch": 0.11022526619936861,
"grad_norm": 0.859375,
"learning_rate": 0.0004999998907231029,
"loss": 7.4825,
"mean_token_accuracy": 0.08603335320949554,
"num_tokens": 2239885.0,
"step": 1030
},
{
"entropy": 7.723711681365967,
"epoch": 0.11076034030713254,
"grad_norm": 0.96875,
"learning_rate": 0.0004999998497929974,
"loss": 7.4942,
"mean_token_accuracy": 0.08984629586338996,
"num_tokens": 2250221.0,
"step": 1035
},
{
"entropy": 7.636352014541626,
"epoch": 0.11129541441489646,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999998023660527,
"loss": 7.5667,
"mean_token_accuracy": 0.07426251731812954,
"num_tokens": 2261059.0,
"step": 1040
},
{
"entropy": 7.627584838867188,
"epoch": 0.11183048852266039,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999997484422705,
"loss": 7.4364,
"mean_token_accuracy": 0.08743720948696136,
"num_tokens": 2271174.0,
"step": 1045
},
{
"entropy": 7.669003200531006,
"epoch": 0.11236556263042431,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999996880216521,
"loss": 7.5139,
"mean_token_accuracy": 0.08555590137839317,
"num_tokens": 2282431.0,
"step": 1050
},
{
"entropy": 7.61585578918457,
"epoch": 0.11290063673818825,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999996211041994,
"loss": 7.4858,
"mean_token_accuracy": 0.0835509292781353,
"num_tokens": 2292891.0,
"step": 1055
},
{
"entropy": 7.677904653549194,
"epoch": 0.11343571084595216,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999995476899141,
"loss": 7.5005,
"mean_token_accuracy": 0.08647905215620995,
"num_tokens": 2303764.0,
"step": 1060
},
{
"entropy": 7.639417934417724,
"epoch": 0.11397078495371608,
"grad_norm": 0.953125,
"learning_rate": 0.0004999994677787987,
"loss": 7.4227,
"mean_token_accuracy": 0.08917138800024986,
"num_tokens": 2314172.0,
"step": 1065
},
{
"entropy": 7.604849863052368,
"epoch": 0.11450585906148002,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999993813708551,
"loss": 7.3983,
"mean_token_accuracy": 0.09371343329548835,
"num_tokens": 2324848.0,
"step": 1070
},
{
"entropy": 7.562433481216431,
"epoch": 0.11504093316924394,
"grad_norm": 0.80078125,
"learning_rate": 0.000499999288466086,
"loss": 7.4262,
"mean_token_accuracy": 0.0858787901699543,
"num_tokens": 2336601.0,
"step": 1075
},
{
"entropy": 7.615593194961548,
"epoch": 0.11557600727700787,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999991890644941,
"loss": 7.4123,
"mean_token_accuracy": 0.08523132391273976,
"num_tokens": 2347281.0,
"step": 1080
},
{
"entropy": 7.5736212730407715,
"epoch": 0.11611108138477179,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999990831660822,
"loss": 7.4363,
"mean_token_accuracy": 0.09456784278154373,
"num_tokens": 2358199.0,
"step": 1085
},
{
"entropy": 7.702199506759643,
"epoch": 0.11664615549253572,
"grad_norm": 0.9375,
"learning_rate": 0.0004999989707708534,
"loss": 7.4808,
"mean_token_accuracy": 0.08690382875502109,
"num_tokens": 2369092.0,
"step": 1090
},
{
"entropy": 7.544908285140991,
"epoch": 0.11718122960029964,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999988518788111,
"loss": 7.4507,
"mean_token_accuracy": 0.08844347447156906,
"num_tokens": 2378849.0,
"step": 1095
},
{
"entropy": 7.622500801086426,
"epoch": 0.11771630370806356,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999987264899583,
"loss": 7.4509,
"mean_token_accuracy": 0.09384607821702957,
"num_tokens": 2389587.0,
"step": 1100
},
{
"entropy": 7.658657217025757,
"epoch": 0.1182513778158275,
"grad_norm": 0.87890625,
"learning_rate": 0.000499998594604299,
"loss": 7.4222,
"mean_token_accuracy": 0.09075777530670166,
"num_tokens": 2400442.0,
"step": 1105
},
{
"entropy": 7.513382863998413,
"epoch": 0.11878645192359141,
"grad_norm": 0.8671875,
"learning_rate": 0.000499998456221837,
"loss": 7.4165,
"mean_token_accuracy": 0.09110527783632279,
"num_tokens": 2410771.0,
"step": 1110
},
{
"entropy": 7.624029874801636,
"epoch": 0.11932152603135535,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999983113425762,
"loss": 7.4209,
"mean_token_accuracy": 0.09688404947519302,
"num_tokens": 2421112.0,
"step": 1115
},
{
"entropy": 7.630374145507813,
"epoch": 0.11985660013911927,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999981599665207,
"loss": 7.4046,
"mean_token_accuracy": 0.08740727975964546,
"num_tokens": 2431891.0,
"step": 1120
},
{
"entropy": 7.452549457550049,
"epoch": 0.1203916742468832,
"grad_norm": 0.921875,
"learning_rate": 0.0004999980020936748,
"loss": 7.3562,
"mean_token_accuracy": 0.0897531658411026,
"num_tokens": 2442987.0,
"step": 1125
},
{
"entropy": 7.609330129623413,
"epoch": 0.12092674835464712,
"grad_norm": 1.0,
"learning_rate": 0.0004999978377240434,
"loss": 7.3982,
"mean_token_accuracy": 0.08765116408467293,
"num_tokens": 2452703.0,
"step": 1130
},
{
"entropy": 7.556497859954834,
"epoch": 0.12146182246241104,
"grad_norm": 0.96484375,
"learning_rate": 0.000499997666857631,
"loss": 7.4509,
"mean_token_accuracy": 0.09336601346731185,
"num_tokens": 2462975.0,
"step": 1135
},
{
"entropy": 7.558685207366944,
"epoch": 0.12199689657017497,
"grad_norm": 0.8046875,
"learning_rate": 0.0004999974894944426,
"loss": 7.3995,
"mean_token_accuracy": 0.08522969335317612,
"num_tokens": 2473853.0,
"step": 1140
},
{
"entropy": 7.575222444534302,
"epoch": 0.12253197067793889,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999973056344832,
"loss": 7.3681,
"mean_token_accuracy": 0.0891667477786541,
"num_tokens": 2485633.0,
"step": 1145
},
{
"entropy": 7.465674591064453,
"epoch": 0.12306704478570282,
"grad_norm": 1.0,
"learning_rate": 0.0004999971152777583,
"loss": 7.3416,
"mean_token_accuracy": 0.08593353554606438,
"num_tokens": 2496922.0,
"step": 1150
},
{
"entropy": 7.701901197433472,
"epoch": 0.12360211889346674,
"grad_norm": 0.8046875,
"learning_rate": 0.0004999969184242733,
"loss": 7.4062,
"mean_token_accuracy": 0.09027116596698762,
"num_tokens": 2507890.0,
"step": 1155
},
{
"entropy": 7.481602525711059,
"epoch": 0.12413719300123068,
"grad_norm": 0.98046875,
"learning_rate": 0.000499996715074034,
"loss": 7.4012,
"mean_token_accuracy": 0.0900260142982006,
"num_tokens": 2519749.0,
"step": 1160
},
{
"entropy": 7.518426036834716,
"epoch": 0.1246722671089946,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999965052270461,
"loss": 7.4146,
"mean_token_accuracy": 0.08937735855579376,
"num_tokens": 2530666.0,
"step": 1165
},
{
"entropy": 7.581199979782104,
"epoch": 0.12520734121675853,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999962888833157,
"loss": 7.419,
"mean_token_accuracy": 0.08899203762412071,
"num_tokens": 2541594.0,
"step": 1170
},
{
"entropy": 7.519051456451416,
"epoch": 0.12574241532452243,
"grad_norm": 0.890625,
"learning_rate": 0.0004999960660428491,
"loss": 7.4116,
"mean_token_accuracy": 0.0924226313829422,
"num_tokens": 2552330.0,
"step": 1175
},
{
"entropy": 7.657570695877075,
"epoch": 0.12627748943228637,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999958367056526,
"loss": 7.4478,
"mean_token_accuracy": 0.08449417613446712,
"num_tokens": 2564601.0,
"step": 1180
},
{
"entropy": 7.51881742477417,
"epoch": 0.1268125635400503,
"grad_norm": 0.97265625,
"learning_rate": 0.000499995600871733,
"loss": 7.4143,
"mean_token_accuracy": 0.09234437346458435,
"num_tokens": 2575539.0,
"step": 1185
},
{
"entropy": 7.498287916183472,
"epoch": 0.12734763764781423,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999953585410971,
"loss": 7.3524,
"mean_token_accuracy": 0.08859748020768166,
"num_tokens": 2586635.0,
"step": 1190
},
{
"entropy": 7.593882083892822,
"epoch": 0.12788271175557814,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999951097137518,
"loss": 7.3818,
"mean_token_accuracy": 0.09308656603097916,
"num_tokens": 2596814.0,
"step": 1195
},
{
"entropy": 7.515226888656616,
"epoch": 0.12841778586334207,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999948543897044,
"loss": 7.3996,
"mean_token_accuracy": 0.08756894245743752,
"num_tokens": 2608087.0,
"step": 1200
},
{
"entropy": 7.600381469726562,
"epoch": 0.128952859971106,
"grad_norm": 0.890625,
"learning_rate": 0.0004999945925689621,
"loss": 7.3998,
"mean_token_accuracy": 0.09394723773002625,
"num_tokens": 2619377.0,
"step": 1205
},
{
"entropy": 7.480240154266357,
"epoch": 0.12948793407886994,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999943242515325,
"loss": 7.2998,
"mean_token_accuracy": 0.0932411678135395,
"num_tokens": 2629089.0,
"step": 1210
},
{
"entropy": 7.522952890396118,
"epoch": 0.13002300818663384,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999940494374236,
"loss": 7.3684,
"mean_token_accuracy": 0.09049427583813667,
"num_tokens": 2639743.0,
"step": 1215
},
{
"entropy": 7.585643148422241,
"epoch": 0.13055808229439778,
"grad_norm": 0.921875,
"learning_rate": 0.000499993768126643,
"loss": 7.397,
"mean_token_accuracy": 0.08678129985928536,
"num_tokens": 2650631.0,
"step": 1220
},
{
"entropy": 7.525626993179321,
"epoch": 0.1310931564021617,
"grad_norm": 0.89453125,
"learning_rate": 0.000499993480319199,
"loss": 7.366,
"mean_token_accuracy": 0.09070794358849525,
"num_tokens": 2660197.0,
"step": 1225
},
{
"entropy": 7.495192050933838,
"epoch": 0.13162823050992561,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999931860150999,
"loss": 7.3461,
"mean_token_accuracy": 0.08914455473423004,
"num_tokens": 2670162.0,
"step": 1230
},
{
"entropy": 7.548467350006104,
"epoch": 0.13216330461768955,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999928852143541,
"loss": 7.3309,
"mean_token_accuracy": 0.0927185446023941,
"num_tokens": 2681233.0,
"step": 1235
},
{
"entropy": 7.398493242263794,
"epoch": 0.13269837872545348,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999925779169703,
"loss": 7.2982,
"mean_token_accuracy": 0.09300850480794906,
"num_tokens": 2692145.0,
"step": 1240
},
{
"entropy": 7.526278877258301,
"epoch": 0.13323345283321741,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999922641229575,
"loss": 7.3431,
"mean_token_accuracy": 0.0936188168823719,
"num_tokens": 2703192.0,
"step": 1245
},
{
"entropy": 7.520468711853027,
"epoch": 0.13376852694098132,
"grad_norm": 0.81640625,
"learning_rate": 0.0004999919438323248,
"loss": 7.3363,
"mean_token_accuracy": 0.09251101464033126,
"num_tokens": 2714804.0,
"step": 1250
},
{
"entropy": 7.4432690143585205,
"epoch": 0.13430360104874525,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999916170450812,
"loss": 7.3068,
"mean_token_accuracy": 0.09502202421426773,
"num_tokens": 2724340.0,
"step": 1255
},
{
"entropy": 7.475179243087768,
"epoch": 0.13483867515650919,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999912837612364,
"loss": 7.3071,
"mean_token_accuracy": 0.09392300024628639,
"num_tokens": 2735767.0,
"step": 1260
},
{
"entropy": 7.569590282440186,
"epoch": 0.1353737492642731,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999909439807997,
"loss": 7.3779,
"mean_token_accuracy": 0.08663228303194045,
"num_tokens": 2747658.0,
"step": 1265
},
{
"entropy": 7.444280576705933,
"epoch": 0.13590882337203702,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999905977037813,
"loss": 7.2372,
"mean_token_accuracy": 0.09776455760002137,
"num_tokens": 2757901.0,
"step": 1270
},
{
"entropy": 7.465328884124756,
"epoch": 0.13644389747980096,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999902449301909,
"loss": 7.2776,
"mean_token_accuracy": 0.09301821067929268,
"num_tokens": 2768945.0,
"step": 1275
},
{
"entropy": 7.314613389968872,
"epoch": 0.1369789715875649,
"grad_norm": 0.7734375,
"learning_rate": 0.0004999898856600387,
"loss": 7.2582,
"mean_token_accuracy": 0.1052941657602787,
"num_tokens": 2780276.0,
"step": 1280
},
{
"entropy": 7.536267852783203,
"epoch": 0.1375140456953288,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999895198933354,
"loss": 7.3323,
"mean_token_accuracy": 0.09816832616925239,
"num_tokens": 2791183.0,
"step": 1285
},
{
"entropy": 7.4270100593566895,
"epoch": 0.13804911980309273,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999891476300911,
"loss": 7.2426,
"mean_token_accuracy": 0.09529703482985497,
"num_tokens": 2801928.0,
"step": 1290
},
{
"entropy": 7.473166513442993,
"epoch": 0.13858419391085666,
"grad_norm": 0.8671875,
"learning_rate": 0.000499988768870317,
"loss": 7.3295,
"mean_token_accuracy": 0.08986801654100418,
"num_tokens": 2813415.0,
"step": 1295
},
{
"entropy": 7.48001651763916,
"epoch": 0.13911926801862057,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999883836140236,
"loss": 7.3364,
"mean_token_accuracy": 0.09272250309586524,
"num_tokens": 2825288.0,
"step": 1300
},
{
"entropy": 7.421426439285279,
"epoch": 0.1396543421263845,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999879918612224,
"loss": 7.2918,
"mean_token_accuracy": 0.096827931702137,
"num_tokens": 2836298.0,
"step": 1305
},
{
"entropy": 7.523803615570069,
"epoch": 0.14018941623414843,
"grad_norm": 0.984375,
"learning_rate": 0.0004999875936119245,
"loss": 7.2778,
"mean_token_accuracy": 0.0943089857697487,
"num_tokens": 2846367.0,
"step": 1310
},
{
"entropy": 7.3315290927886965,
"epoch": 0.14072449034191237,
"grad_norm": 0.921875,
"learning_rate": 0.0004999871888661415,
"loss": 7.2799,
"mean_token_accuracy": 0.09142087027430534,
"num_tokens": 2856832.0,
"step": 1315
},
{
"entropy": 7.507561445236206,
"epoch": 0.14125956444967627,
"grad_norm": 0.84375,
"learning_rate": 0.000499986777623885,
"loss": 7.3088,
"mean_token_accuracy": 0.0900144100189209,
"num_tokens": 2867954.0,
"step": 1320
},
{
"entropy": 7.509339809417725,
"epoch": 0.1417946385574402,
"grad_norm": 0.9375,
"learning_rate": 0.0004999863598851669,
"loss": 7.3334,
"mean_token_accuracy": 0.09359355568885804,
"num_tokens": 2878714.0,
"step": 1325
},
{
"entropy": 7.325873184204101,
"epoch": 0.14232971266520414,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999859356499992,
"loss": 7.28,
"mean_token_accuracy": 0.09667187854647637,
"num_tokens": 2890697.0,
"step": 1330
},
{
"entropy": 7.467297649383545,
"epoch": 0.14286478677296804,
"grad_norm": 1.015625,
"learning_rate": 0.0004999855049183943,
"loss": 7.292,
"mean_token_accuracy": 0.09200986102223396,
"num_tokens": 2901468.0,
"step": 1335
},
{
"entropy": 7.420497417449951,
"epoch": 0.14339986088073198,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999850676903646,
"loss": 7.256,
"mean_token_accuracy": 0.09620093181729317,
"num_tokens": 2913466.0,
"step": 1340
},
{
"entropy": 7.381121587753296,
"epoch": 0.1439349349884959,
"grad_norm": 0.859375,
"learning_rate": 0.0004999846239659226,
"loss": 7.2224,
"mean_token_accuracy": 0.10079324766993522,
"num_tokens": 2923638.0,
"step": 1345
},
{
"entropy": 7.451461362838745,
"epoch": 0.14447000909625984,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999841737450812,
"loss": 7.2969,
"mean_token_accuracy": 0.1001754067838192,
"num_tokens": 2934790.0,
"step": 1350
},
{
"entropy": 7.389082574844361,
"epoch": 0.14500508320402375,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999837170278535,
"loss": 7.248,
"mean_token_accuracy": 0.0918399602174759,
"num_tokens": 2946320.0,
"step": 1355
},
{
"entropy": 7.486172676086426,
"epoch": 0.14554015731178768,
"grad_norm": 0.984375,
"learning_rate": 0.0004999832538142526,
"loss": 7.2238,
"mean_token_accuracy": 0.091259303689003,
"num_tokens": 2957077.0,
"step": 1360
},
{
"entropy": 7.342933654785156,
"epoch": 0.14607523141955162,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999827841042917,
"loss": 7.2296,
"mean_token_accuracy": 0.09233908951282502,
"num_tokens": 2967286.0,
"step": 1365
},
{
"entropy": 7.4447362422943115,
"epoch": 0.14661030552731552,
"grad_norm": 1.203125,
"learning_rate": 0.0004999823078979846,
"loss": 7.2744,
"mean_token_accuracy": 0.09223584830760956,
"num_tokens": 2979516.0,
"step": 1370
},
{
"entropy": 7.485718393325806,
"epoch": 0.14714537963507945,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999818251953449,
"loss": 7.399,
"mean_token_accuracy": 0.08901449739933014,
"num_tokens": 2991594.0,
"step": 1375
},
{
"entropy": 7.445301294326782,
"epoch": 0.1476804537428434,
"grad_norm": 0.96875,
"learning_rate": 0.0004999813359963867,
"loss": 7.286,
"mean_token_accuracy": 0.09267361760139466,
"num_tokens": 3002487.0,
"step": 1380
},
{
"entropy": 7.488003301620483,
"epoch": 0.14821552785060732,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999808403011241,
"loss": 7.2851,
"mean_token_accuracy": 0.09769331067800521,
"num_tokens": 3013197.0,
"step": 1385
},
{
"entropy": 7.3263664722442625,
"epoch": 0.14875060195837123,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999803381095712,
"loss": 7.3105,
"mean_token_accuracy": 0.09277946949005127,
"num_tokens": 3024512.0,
"step": 1390
},
{
"entropy": 7.529593801498413,
"epoch": 0.14928567606613516,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999798294217428,
"loss": 7.2724,
"mean_token_accuracy": 0.09323014095425605,
"num_tokens": 3035621.0,
"step": 1395
},
{
"entropy": 7.383570957183838,
"epoch": 0.1498207501738991,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999793142376533,
"loss": 7.2703,
"mean_token_accuracy": 0.09348074868321418,
"num_tokens": 3046492.0,
"step": 1400
},
{
"entropy": 7.38762936592102,
"epoch": 0.150355824281663,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999787925573177,
"loss": 7.2831,
"mean_token_accuracy": 0.09317544624209403,
"num_tokens": 3057715.0,
"step": 1405
},
{
"entropy": 7.511090230941773,
"epoch": 0.15089089838942693,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999782643807513,
"loss": 7.2558,
"mean_token_accuracy": 0.0898089624941349,
"num_tokens": 3068344.0,
"step": 1410
},
{
"entropy": 7.340856122970581,
"epoch": 0.15142597249719086,
"grad_norm": 0.828125,
"learning_rate": 0.0004999777297079689,
"loss": 7.2191,
"mean_token_accuracy": 0.0933483250439167,
"num_tokens": 3079576.0,
"step": 1415
},
{
"entropy": 7.457035493850708,
"epoch": 0.1519610466049548,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999771885389863,
"loss": 7.2075,
"mean_token_accuracy": 0.09207480028271675,
"num_tokens": 3090201.0,
"step": 1420
},
{
"entropy": 7.333426856994629,
"epoch": 0.1524961207127187,
"grad_norm": 0.984375,
"learning_rate": 0.0004999766408738189,
"loss": 7.172,
"mean_token_accuracy": 0.10389826074242592,
"num_tokens": 3099824.0,
"step": 1425
},
{
"entropy": 7.333724737167358,
"epoch": 0.15303119482048264,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999760867124827,
"loss": 7.2405,
"mean_token_accuracy": 0.09329254552721977,
"num_tokens": 3109948.0,
"step": 1430
},
{
"entropy": 7.4186999797821045,
"epoch": 0.15356626892824657,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999755260549937,
"loss": 7.2413,
"mean_token_accuracy": 0.09903448224067687,
"num_tokens": 3121492.0,
"step": 1435
},
{
"entropy": 7.313020372390747,
"epoch": 0.15410134303601047,
"grad_norm": 1.265625,
"learning_rate": 0.0004999749589013677,
"loss": 7.0092,
"mean_token_accuracy": 0.11508586779236793,
"num_tokens": 3131990.0,
"step": 1440
},
{
"entropy": 7.2625288486480715,
"epoch": 0.1546364171437744,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999743852516217,
"loss": 7.2038,
"mean_token_accuracy": 0.09737641140818595,
"num_tokens": 3142567.0,
"step": 1445
},
{
"entropy": 7.4858119010925295,
"epoch": 0.15517149125153834,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999738051057717,
"loss": 7.2253,
"mean_token_accuracy": 0.09757704138755799,
"num_tokens": 3153083.0,
"step": 1450
},
{
"entropy": 7.336223459243774,
"epoch": 0.15570656535930227,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999732184638347,
"loss": 7.2224,
"mean_token_accuracy": 0.09092248976230621,
"num_tokens": 3163985.0,
"step": 1455
},
{
"entropy": 7.436727666854859,
"epoch": 0.15624163946706618,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999726253258278,
"loss": 7.2157,
"mean_token_accuracy": 0.09606200903654098,
"num_tokens": 3175259.0,
"step": 1460
},
{
"entropy": 7.293716049194336,
"epoch": 0.1567767135748301,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999720256917678,
"loss": 7.1804,
"mean_token_accuracy": 0.09885367155075073,
"num_tokens": 3185487.0,
"step": 1465
},
{
"entropy": 7.31433138847351,
"epoch": 0.15731178768259405,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999714195616723,
"loss": 7.1717,
"mean_token_accuracy": 0.10157929807901382,
"num_tokens": 3196200.0,
"step": 1470
},
{
"entropy": 7.441909456253052,
"epoch": 0.15784686179035795,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999708069355586,
"loss": 7.2733,
"mean_token_accuracy": 0.08783909082412719,
"num_tokens": 3207675.0,
"step": 1475
},
{
"entropy": 7.31417498588562,
"epoch": 0.15838193589812188,
"grad_norm": 0.984375,
"learning_rate": 0.0004999701878134445,
"loss": 7.1471,
"mean_token_accuracy": 0.09791682288050652,
"num_tokens": 3218117.0,
"step": 1480
},
{
"entropy": 7.3960308074951175,
"epoch": 0.15891701000588582,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999695621953477,
"loss": 7.2164,
"mean_token_accuracy": 0.10362501963973045,
"num_tokens": 3229208.0,
"step": 1485
},
{
"entropy": 7.309657001495362,
"epoch": 0.15945208411364975,
"grad_norm": 0.90625,
"learning_rate": 0.0004999689300812866,
"loss": 7.1036,
"mean_token_accuracy": 0.10788998156785964,
"num_tokens": 3239586.0,
"step": 1490
},
{
"entropy": 7.392857885360717,
"epoch": 0.15998715822141366,
"grad_norm": 0.78515625,
"learning_rate": 0.0004999682914712791,
"loss": 7.1983,
"mean_token_accuracy": 0.09886570796370506,
"num_tokens": 3250103.0,
"step": 1495
},
{
"entropy": 7.331405687332153,
"epoch": 0.1605222323291776,
"grad_norm": 0.859375,
"learning_rate": 0.0004999676463653439,
"loss": 7.1785,
"mean_token_accuracy": 0.10397682413458824,
"num_tokens": 3261844.0,
"step": 1500
},
{
"entropy": 7.361588764190674,
"epoch": 0.16105730643694152,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999669947634996,
"loss": 7.1436,
"mean_token_accuracy": 0.0984138660132885,
"num_tokens": 3272073.0,
"step": 1505
},
{
"entropy": 7.343551349639893,
"epoch": 0.16159238054470546,
"grad_norm": 0.90625,
"learning_rate": 0.0004999663366657647,
"loss": 7.1261,
"mean_token_accuracy": 0.1005841538310051,
"num_tokens": 3281725.0,
"step": 1510
},
{
"entropy": 7.326076507568359,
"epoch": 0.16212745465246936,
"grad_norm": 0.921875,
"learning_rate": 0.0004999656720721586,
"loss": 7.2726,
"mean_token_accuracy": 0.0962200105190277,
"num_tokens": 3292708.0,
"step": 1515
},
{
"entropy": 7.3782445907592775,
"epoch": 0.1626625287602333,
"grad_norm": 0.875,
"learning_rate": 0.0004999650009827004,
"loss": 7.2564,
"mean_token_accuracy": 0.09442284628748894,
"num_tokens": 3303548.0,
"step": 1520
},
{
"entropy": 7.326221704483032,
"epoch": 0.16319760286799723,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999643233974092,
"loss": 7.117,
"mean_token_accuracy": 0.10348235741257668,
"num_tokens": 3314261.0,
"step": 1525
},
{
"entropy": 7.2751476764678955,
"epoch": 0.16373267697576113,
"grad_norm": 0.859375,
"learning_rate": 0.0004999636393163049,
"loss": 7.1271,
"mean_token_accuracy": 0.10080837979912757,
"num_tokens": 3324833.0,
"step": 1530
},
{
"entropy": 7.2360046863555905,
"epoch": 0.16426775108352507,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999629487394071,
"loss": 7.036,
"mean_token_accuracy": 0.11096755266189576,
"num_tokens": 3334790.0,
"step": 1535
},
{
"entropy": 7.339627885818482,
"epoch": 0.164802825191289,
"grad_norm": 1.046875,
"learning_rate": 0.0004999622516667358,
"loss": 7.1602,
"mean_token_accuracy": 0.09869524762034416,
"num_tokens": 3345115.0,
"step": 1540
},
{
"entropy": 7.268951654434204,
"epoch": 0.16533789929905293,
"grad_norm": 0.7890625,
"learning_rate": 0.000499961548098311,
"loss": 7.0592,
"mean_token_accuracy": 0.10297729372978211,
"num_tokens": 3356187.0,
"step": 1545
},
{
"entropy": 7.235326194763184,
"epoch": 0.16587297340681684,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999608380341533,
"loss": 7.0744,
"mean_token_accuracy": 0.10658099949359894,
"num_tokens": 3367218.0,
"step": 1550
},
{
"entropy": 7.27059588432312,
"epoch": 0.16640804751458077,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999601214742829,
"loss": 7.2223,
"mean_token_accuracy": 0.09261110052466393,
"num_tokens": 3377557.0,
"step": 1555
},
{
"entropy": 7.427518892288208,
"epoch": 0.1669431216223447,
"grad_norm": 0.90625,
"learning_rate": 0.0004999593984187206,
"loss": 7.2095,
"mean_token_accuracy": 0.10194377526640892,
"num_tokens": 3387402.0,
"step": 1560
},
{
"entropy": 7.248392057418823,
"epoch": 0.1674781957301086,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999586688674872,
"loss": 7.0941,
"mean_token_accuracy": 0.10278604626655578,
"num_tokens": 3397924.0,
"step": 1565
},
{
"entropy": 7.3578520774841305,
"epoch": 0.16801326983787254,
"grad_norm": 0.89453125,
"learning_rate": 0.000499957932820604,
"loss": 7.1743,
"mean_token_accuracy": 0.09474834129214287,
"num_tokens": 3408622.0,
"step": 1570
},
{
"entropy": 7.343464088439942,
"epoch": 0.16854834394563648,
"grad_norm": 0.90625,
"learning_rate": 0.000499957190278092,
"loss": 7.1367,
"mean_token_accuracy": 0.10084123015403748,
"num_tokens": 3419692.0,
"step": 1575
},
{
"entropy": 7.277517509460449,
"epoch": 0.1690834180534004,
"grad_norm": 0.9375,
"learning_rate": 0.0004999564412399728,
"loss": 7.1842,
"mean_token_accuracy": 0.09453171491622925,
"num_tokens": 3430986.0,
"step": 1580
},
{
"entropy": 7.289031267166138,
"epoch": 0.1696184921611643,
"grad_norm": 0.90625,
"learning_rate": 0.000499955685706268,
"loss": 7.1056,
"mean_token_accuracy": 0.10056664869189262,
"num_tokens": 3442211.0,
"step": 1585
},
{
"entropy": 7.181107234954834,
"epoch": 0.17015356626892825,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999549236769993,
"loss": 7.0973,
"mean_token_accuracy": 0.10596697479486465,
"num_tokens": 3453585.0,
"step": 1590
},
{
"entropy": 7.261494874954224,
"epoch": 0.17068864037669218,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999541551521888,
"loss": 7.1281,
"mean_token_accuracy": 0.10282958298921585,
"num_tokens": 3464156.0,
"step": 1595
},
{
"entropy": 7.3783262252807615,
"epoch": 0.17122371448445609,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999533801318587,
"loss": 7.1901,
"mean_token_accuracy": 0.095138331502676,
"num_tokens": 3474853.0,
"step": 1600
},
{
"entropy": 7.3372804641723635,
"epoch": 0.17175878859222002,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999525986160313,
"loss": 7.1634,
"mean_token_accuracy": 0.10265101641416549,
"num_tokens": 3486841.0,
"step": 1605
},
{
"entropy": 7.3480212688446045,
"epoch": 0.17229386269998395,
"grad_norm": 0.81640625,
"learning_rate": 0.0004999518106047293,
"loss": 7.1702,
"mean_token_accuracy": 0.0980787456035614,
"num_tokens": 3498735.0,
"step": 1610
},
{
"entropy": 7.253286647796631,
"epoch": 0.17282893680774788,
"grad_norm": 0.84375,
"learning_rate": 0.0004999510160979754,
"loss": 7.2008,
"mean_token_accuracy": 0.10094587206840515,
"num_tokens": 3508829.0,
"step": 1615
},
{
"entropy": 7.2795453548431395,
"epoch": 0.1733640109155118,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999502150957925,
"loss": 7.1209,
"mean_token_accuracy": 0.10228936895728111,
"num_tokens": 3519465.0,
"step": 1620
},
{
"entropy": 7.299279642105103,
"epoch": 0.17389908502327572,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999494075982037,
"loss": 7.0647,
"mean_token_accuracy": 0.10867529138922691,
"num_tokens": 3530401.0,
"step": 1625
},
{
"entropy": 7.235185194015503,
"epoch": 0.17443415913103966,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999485936052324,
"loss": 7.141,
"mean_token_accuracy": 0.10203565210103989,
"num_tokens": 3541136.0,
"step": 1630
},
{
"entropy": 7.373658609390259,
"epoch": 0.17496923323880356,
"grad_norm": 0.88671875,
"learning_rate": 0.000499947773116902,
"loss": 7.1322,
"mean_token_accuracy": 0.1018851414322853,
"num_tokens": 3552525.0,
"step": 1635
},
{
"entropy": 7.22913932800293,
"epoch": 0.1755043073465675,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999469461332365,
"loss": 7.0903,
"mean_token_accuracy": 0.10703189074993133,
"num_tokens": 3563028.0,
"step": 1640
},
{
"entropy": 7.259391117095947,
"epoch": 0.17603938145433143,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999461126542592,
"loss": 7.0522,
"mean_token_accuracy": 0.10235123336315155,
"num_tokens": 3573354.0,
"step": 1645
},
{
"entropy": 7.158776950836182,
"epoch": 0.17657445556209536,
"grad_norm": 0.8125,
"learning_rate": 0.0004999452726799947,
"loss": 7.0373,
"mean_token_accuracy": 0.10366476848721504,
"num_tokens": 3585576.0,
"step": 1650
},
{
"entropy": 7.320888566970825,
"epoch": 0.17710952966985927,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999444262104671,
"loss": 7.1598,
"mean_token_accuracy": 0.09593332335352897,
"num_tokens": 3596478.0,
"step": 1655
},
{
"entropy": 7.289003372192383,
"epoch": 0.1776446037776232,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999435732457007,
"loss": 7.123,
"mean_token_accuracy": 0.09818345233798027,
"num_tokens": 3608014.0,
"step": 1660
},
{
"entropy": 7.291224193572998,
"epoch": 0.17817967788538713,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999427137857203,
"loss": 7.1694,
"mean_token_accuracy": 0.09814337715506553,
"num_tokens": 3620120.0,
"step": 1665
},
{
"entropy": 7.274828290939331,
"epoch": 0.17871475199315104,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999418478305506,
"loss": 7.0748,
"mean_token_accuracy": 0.09978235512971878,
"num_tokens": 3632578.0,
"step": 1670
},
{
"entropy": 7.185247135162354,
"epoch": 0.17924982610091497,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999409753802167,
"loss": 7.0939,
"mean_token_accuracy": 0.10724005252122878,
"num_tokens": 3643245.0,
"step": 1675
},
{
"entropy": 7.27445502281189,
"epoch": 0.1797849002086789,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999400964347437,
"loss": 7.0967,
"mean_token_accuracy": 0.10385240912437439,
"num_tokens": 3655832.0,
"step": 1680
},
{
"entropy": 7.2620195865631105,
"epoch": 0.18031997431644284,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999392109941571,
"loss": 7.0683,
"mean_token_accuracy": 0.10449369996786118,
"num_tokens": 3666566.0,
"step": 1685
},
{
"entropy": 7.303769207000732,
"epoch": 0.18085504842420674,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999383190584822,
"loss": 7.0852,
"mean_token_accuracy": 0.10249918699264526,
"num_tokens": 3677572.0,
"step": 1690
},
{
"entropy": 7.2038905143737795,
"epoch": 0.18139012253197068,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999374206277451,
"loss": 7.0113,
"mean_token_accuracy": 0.10162734165787697,
"num_tokens": 3687116.0,
"step": 1695
},
{
"entropy": 7.175737571716309,
"epoch": 0.1819251966397346,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999365157019717,
"loss": 7.1189,
"mean_token_accuracy": 0.10182299092411995,
"num_tokens": 3697961.0,
"step": 1700
},
{
"entropy": 7.309969711303711,
"epoch": 0.18246027074749852,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999356042811878,
"loss": 7.0444,
"mean_token_accuracy": 0.10850266963243485,
"num_tokens": 3708834.0,
"step": 1705
},
{
"entropy": 7.182546806335449,
"epoch": 0.18299534485526245,
"grad_norm": 0.84375,
"learning_rate": 0.00049993468636542,
"loss": 7.0991,
"mean_token_accuracy": 0.10176495090126991,
"num_tokens": 3719835.0,
"step": 1710
},
{
"entropy": 7.250284337997437,
"epoch": 0.18353041896302638,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999337619546948,
"loss": 7.0009,
"mean_token_accuracy": 0.10658760368824005,
"num_tokens": 3730408.0,
"step": 1715
},
{
"entropy": 7.171570301055908,
"epoch": 0.18406549307079031,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999328310490387,
"loss": 7.0695,
"mean_token_accuracy": 0.10707555189728737,
"num_tokens": 3741257.0,
"step": 1720
},
{
"entropy": 7.23467698097229,
"epoch": 0.18460056717855422,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999318936484789,
"loss": 7.0647,
"mean_token_accuracy": 0.10648670792579651,
"num_tokens": 3752615.0,
"step": 1725
},
{
"entropy": 7.214528036117554,
"epoch": 0.18513564128631815,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999309497530422,
"loss": 7.0949,
"mean_token_accuracy": 0.09849463924765586,
"num_tokens": 3763656.0,
"step": 1730
},
{
"entropy": 7.150198459625244,
"epoch": 0.1856707153940821,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999299993627558,
"loss": 7.0733,
"mean_token_accuracy": 0.10403963699936866,
"num_tokens": 3775257.0,
"step": 1735
},
{
"entropy": 7.201783800125122,
"epoch": 0.186205789501846,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999290424776475,
"loss": 7.0384,
"mean_token_accuracy": 0.11034373492002487,
"num_tokens": 3786094.0,
"step": 1740
},
{
"entropy": 7.355918598175049,
"epoch": 0.18674086360960993,
"grad_norm": 0.953125,
"learning_rate": 0.0004999280790977445,
"loss": 7.1286,
"mean_token_accuracy": 0.09715545624494552,
"num_tokens": 3796672.0,
"step": 1745
},
{
"entropy": 7.305685806274414,
"epoch": 0.18727593771737386,
"grad_norm": 0.88671875,
"learning_rate": 0.000499927109223075,
"loss": 7.0958,
"mean_token_accuracy": 0.10413563027977943,
"num_tokens": 3807910.0,
"step": 1750
},
{
"entropy": 7.1276421546936035,
"epoch": 0.1878110118251378,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999261328536667,
"loss": 7.0578,
"mean_token_accuracy": 0.10628265142440796,
"num_tokens": 3820792.0,
"step": 1755
},
{
"entropy": 7.157390975952149,
"epoch": 0.1883460859329017,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999251499895479,
"loss": 6.9844,
"mean_token_accuracy": 0.10621865093708038,
"num_tokens": 3831126.0,
"step": 1760
},
{
"entropy": 7.23041090965271,
"epoch": 0.18888116004066563,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999241606307471,
"loss": 7.0942,
"mean_token_accuracy": 0.10165138021111489,
"num_tokens": 3841923.0,
"step": 1765
},
{
"entropy": 7.243416452407837,
"epoch": 0.18941623414842956,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999231647772927,
"loss": 7.0078,
"mean_token_accuracy": 0.10644997358322143,
"num_tokens": 3852504.0,
"step": 1770
},
{
"entropy": 7.0957677364349365,
"epoch": 0.1899513082561935,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999221624292136,
"loss": 7.0087,
"mean_token_accuracy": 0.1103954017162323,
"num_tokens": 3862761.0,
"step": 1775
},
{
"entropy": 7.252035474777221,
"epoch": 0.1904863823639574,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999211535865385,
"loss": 7.0234,
"mean_token_accuracy": 0.1052067756652832,
"num_tokens": 3873662.0,
"step": 1780
},
{
"entropy": 7.154085254669189,
"epoch": 0.19102145647172133,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999201382492969,
"loss": 6.9621,
"mean_token_accuracy": 0.10425644144415855,
"num_tokens": 3884835.0,
"step": 1785
},
{
"entropy": 7.187242269515991,
"epoch": 0.19155653057948527,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999191164175178,
"loss": 7.1166,
"mean_token_accuracy": 0.10480258762836456,
"num_tokens": 3895073.0,
"step": 1790
},
{
"entropy": 7.2901567935943605,
"epoch": 0.19209160468724917,
"grad_norm": 0.890625,
"learning_rate": 0.0004999180880912309,
"loss": 7.0556,
"mean_token_accuracy": 0.10658993124961853,
"num_tokens": 3905440.0,
"step": 1795
},
{
"entropy": 7.176698493957519,
"epoch": 0.1926266787950131,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999170532704657,
"loss": 7.0712,
"mean_token_accuracy": 0.10166745483875275,
"num_tokens": 3917135.0,
"step": 1800
},
{
"entropy": 7.2208339214324955,
"epoch": 0.19316175290277704,
"grad_norm": 0.76953125,
"learning_rate": 0.0004999160119552523,
"loss": 7.0716,
"mean_token_accuracy": 0.10915103554725647,
"num_tokens": 3928247.0,
"step": 1805
},
{
"entropy": 7.223503684997558,
"epoch": 0.19369682701054097,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999149641456206,
"loss": 7.0491,
"mean_token_accuracy": 0.1053008608520031,
"num_tokens": 3938609.0,
"step": 1810
},
{
"entropy": 7.265676212310791,
"epoch": 0.19423190111830488,
"grad_norm": 0.8984375,
"learning_rate": 0.000499913909841601,
"loss": 7.0274,
"mean_token_accuracy": 0.10771157816052437,
"num_tokens": 3949221.0,
"step": 1815
},
{
"entropy": 7.100171852111816,
"epoch": 0.1947669752260688,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999128490432238,
"loss": 7.0149,
"mean_token_accuracy": 0.09949721470475197,
"num_tokens": 3959475.0,
"step": 1820
},
{
"entropy": 7.136339998245239,
"epoch": 0.19530204933383274,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999117817505196,
"loss": 7.0138,
"mean_token_accuracy": 0.11632085666060447,
"num_tokens": 3969638.0,
"step": 1825
},
{
"entropy": 7.178588724136352,
"epoch": 0.19583712344159665,
"grad_norm": 0.859375,
"learning_rate": 0.0004999107079635194,
"loss": 7.0644,
"mean_token_accuracy": 0.10715136379003525,
"num_tokens": 3980543.0,
"step": 1830
},
{
"entropy": 7.336812162399292,
"epoch": 0.19637219754936058,
"grad_norm": 0.86328125,
"learning_rate": 0.000499909627682254,
"loss": 7.1143,
"mean_token_accuracy": 0.0920241579413414,
"num_tokens": 3993518.0,
"step": 1835
},
{
"entropy": 7.216370439529419,
"epoch": 0.19690727165712452,
"grad_norm": 0.96875,
"learning_rate": 0.0004999085409067549,
"loss": 7.0948,
"mean_token_accuracy": 0.10310237035155297,
"num_tokens": 4002940.0,
"step": 1840
},
{
"entropy": 7.141087865829467,
"epoch": 0.19744234576488845,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999074476370531,
"loss": 7.0291,
"mean_token_accuracy": 0.10522487610578538,
"num_tokens": 4013982.0,
"step": 1845
},
{
"entropy": 7.2277580261230465,
"epoch": 0.19797741987265235,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999063478731805,
"loss": 6.9933,
"mean_token_accuracy": 0.11192195788025856,
"num_tokens": 4025607.0,
"step": 1850
},
{
"entropy": 7.117806482315063,
"epoch": 0.1985124939804163,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999052416151687,
"loss": 7.0568,
"mean_token_accuracy": 0.10609380379319192,
"num_tokens": 4037467.0,
"step": 1855
},
{
"entropy": 7.221655082702637,
"epoch": 0.19904756808818022,
"grad_norm": 0.828125,
"learning_rate": 0.0004999041288630496,
"loss": 7.065,
"mean_token_accuracy": 0.10668069496750832,
"num_tokens": 4049146.0,
"step": 1860
},
{
"entropy": 7.2340247631073,
"epoch": 0.19958264219594413,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999030096168553,
"loss": 7.1116,
"mean_token_accuracy": 0.10125580132007599,
"num_tokens": 4060500.0,
"step": 1865
},
{
"entropy": 7.248100090026855,
"epoch": 0.20011771630370806,
"grad_norm": 0.96875,
"learning_rate": 0.0004999018838766182,
"loss": 7.0427,
"mean_token_accuracy": 0.10293570905923843,
"num_tokens": 4070553.0,
"step": 1870
},
{
"entropy": 7.2189253807067875,
"epoch": 0.200652790411472,
"grad_norm": 0.9296875,
"learning_rate": 0.000499900751642371,
"loss": 6.988,
"mean_token_accuracy": 0.1170131392776966,
"num_tokens": 4079653.0,
"step": 1875
},
{
"entropy": 7.100331258773804,
"epoch": 0.20118786451923593,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998996129141461,
"loss": 7.0654,
"mean_token_accuracy": 0.10362578108906746,
"num_tokens": 4089912.0,
"step": 1880
},
{
"entropy": 7.275002813339233,
"epoch": 0.20172293862699983,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998984676919764,
"loss": 7.0342,
"mean_token_accuracy": 0.10268538147211075,
"num_tokens": 4100577.0,
"step": 1885
},
{
"entropy": 7.187543249130249,
"epoch": 0.20225801273476376,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998973159758952,
"loss": 7.0658,
"mean_token_accuracy": 0.10256308168172837,
"num_tokens": 4111752.0,
"step": 1890
},
{
"entropy": 7.10661129951477,
"epoch": 0.2027930868425277,
"grad_norm": 0.9375,
"learning_rate": 0.0004998961577659355,
"loss": 7.015,
"mean_token_accuracy": 0.10711082518100738,
"num_tokens": 4122616.0,
"step": 1895
},
{
"entropy": 7.239635181427002,
"epoch": 0.2033281609502916,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998949930621309,
"loss": 7.1135,
"mean_token_accuracy": 0.10389721244573594,
"num_tokens": 4134275.0,
"step": 1900
},
{
"entropy": 7.159951686859131,
"epoch": 0.20386323505805554,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998938218645149,
"loss": 7.0093,
"mean_token_accuracy": 0.10267628878355026,
"num_tokens": 4146412.0,
"step": 1905
},
{
"entropy": 7.137516021728516,
"epoch": 0.20439830916581947,
"grad_norm": 0.828125,
"learning_rate": 0.0004998926441731215,
"loss": 7.0056,
"mean_token_accuracy": 0.10260871946811675,
"num_tokens": 4157592.0,
"step": 1910
},
{
"entropy": 7.170009613037109,
"epoch": 0.2049333832735834,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998914599879845,
"loss": 6.9763,
"mean_token_accuracy": 0.1078723892569542,
"num_tokens": 4168637.0,
"step": 1915
},
{
"entropy": 7.15439133644104,
"epoch": 0.2054684573813473,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998902693091382,
"loss": 7.0316,
"mean_token_accuracy": 0.10298030972480773,
"num_tokens": 4179951.0,
"step": 1920
},
{
"entropy": 7.146353483200073,
"epoch": 0.20600353148911124,
"grad_norm": 0.8203125,
"learning_rate": 0.0004998890721366171,
"loss": 6.9259,
"mean_token_accuracy": 0.1077031098306179,
"num_tokens": 4189895.0,
"step": 1925
},
{
"entropy": 7.1884785175323485,
"epoch": 0.20653860559687517,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998878684704556,
"loss": 6.9636,
"mean_token_accuracy": 0.11009967923164368,
"num_tokens": 4199796.0,
"step": 1930
},
{
"entropy": 7.076934337615967,
"epoch": 0.20707367970463908,
"grad_norm": 0.875,
"learning_rate": 0.0004998866583106884,
"loss": 7.0095,
"mean_token_accuracy": 0.10127623230218888,
"num_tokens": 4210428.0,
"step": 1935
},
{
"entropy": 7.256487798690796,
"epoch": 0.207608753812403,
"grad_norm": 0.921875,
"learning_rate": 0.0004998854416573506,
"loss": 7.0649,
"mean_token_accuracy": 0.1079666055738926,
"num_tokens": 4221515.0,
"step": 1940
},
{
"entropy": 7.144673681259155,
"epoch": 0.20814382792016695,
"grad_norm": 0.953125,
"learning_rate": 0.0004998842185104772,
"loss": 7.0776,
"mean_token_accuracy": 0.11020539179444314,
"num_tokens": 4232580.0,
"step": 1945
},
{
"entropy": 7.1092174530029295,
"epoch": 0.20867890202793088,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998829888701036,
"loss": 6.9893,
"mean_token_accuracy": 0.11271198317408562,
"num_tokens": 4242871.0,
"step": 1950
},
{
"entropy": 7.213183069229126,
"epoch": 0.20921397613569478,
"grad_norm": 0.90625,
"learning_rate": 0.0004998817527362654,
"loss": 7.0261,
"mean_token_accuracy": 0.10301245972514153,
"num_tokens": 4253419.0,
"step": 1955
},
{
"entropy": 7.153800678253174,
"epoch": 0.20974905024345872,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998805101089982,
"loss": 6.9999,
"mean_token_accuracy": 0.10977082848548889,
"num_tokens": 4264327.0,
"step": 1960
},
{
"entropy": 7.214887237548828,
"epoch": 0.21028412435122265,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998792609883378,
"loss": 6.9766,
"mean_token_accuracy": 0.10818460211157799,
"num_tokens": 4274568.0,
"step": 1965
},
{
"entropy": 7.140443277359009,
"epoch": 0.21081919845898656,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998780053743203,
"loss": 6.9731,
"mean_token_accuracy": 0.11215656697750091,
"num_tokens": 4284895.0,
"step": 1970
},
{
"entropy": 7.1276960372924805,
"epoch": 0.2113542725667505,
"grad_norm": 0.8203125,
"learning_rate": 0.0004998767432669822,
"loss": 7.0104,
"mean_token_accuracy": 0.10507805719971657,
"num_tokens": 4297258.0,
"step": 1975
},
{
"entropy": 7.149711656570434,
"epoch": 0.21188934667451442,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998754746663595,
"loss": 6.9365,
"mean_token_accuracy": 0.10693479850888252,
"num_tokens": 4308307.0,
"step": 1980
},
{
"entropy": 7.124268484115601,
"epoch": 0.21242442078227836,
"grad_norm": 0.81640625,
"learning_rate": 0.0004998741995724892,
"loss": 7.0014,
"mean_token_accuracy": 0.10538375228643418,
"num_tokens": 4318801.0,
"step": 1985
},
{
"entropy": 7.139152193069458,
"epoch": 0.21295949489004226,
"grad_norm": 0.875,
"learning_rate": 0.0004998729179854079,
"loss": 6.9777,
"mean_token_accuracy": 0.10666713044047356,
"num_tokens": 4329531.0,
"step": 1990
},
{
"entropy": 7.1325037479400635,
"epoch": 0.2134945689978062,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998716299051527,
"loss": 6.9611,
"mean_token_accuracy": 0.11172771900892257,
"num_tokens": 4341203.0,
"step": 1995
},
{
"entropy": 7.091319608688354,
"epoch": 0.21402964310557013,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998703353317609,
"loss": 7.0092,
"mean_token_accuracy": 0.10446273684501647,
"num_tokens": 4352426.0,
"step": 2000
},
{
"entropy": 7.161164569854736,
"epoch": 0.21456471721333403,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998690342652697,
"loss": 6.9493,
"mean_token_accuracy": 0.11236951574683189,
"num_tokens": 4363413.0,
"step": 2005
},
{
"entropy": 7.104535961151123,
"epoch": 0.21509979132109797,
"grad_norm": 0.8203125,
"learning_rate": 0.0004998677267057166,
"loss": 6.8913,
"mean_token_accuracy": 0.11243644654750824,
"num_tokens": 4373470.0,
"step": 2010
},
{
"entropy": 7.114715242385865,
"epoch": 0.2156348654288619,
"grad_norm": 0.921875,
"learning_rate": 0.0004998664126531397,
"loss": 6.9596,
"mean_token_accuracy": 0.1114291749894619,
"num_tokens": 4384438.0,
"step": 2015
},
{
"entropy": 7.101155853271484,
"epoch": 0.21616993953662583,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998650921075766,
"loss": 6.9454,
"mean_token_accuracy": 0.10823875814676284,
"num_tokens": 4395100.0,
"step": 2020
},
{
"entropy": 7.117280292510986,
"epoch": 0.21670501364438974,
"grad_norm": 0.828125,
"learning_rate": 0.0004998637650690656,
"loss": 7.0544,
"mean_token_accuracy": 0.10239638835191726,
"num_tokens": 4407131.0,
"step": 2025
},
{
"entropy": 7.1200460433959964,
"epoch": 0.21724008775215367,
"grad_norm": 0.953125,
"learning_rate": 0.000499862431537645,
"loss": 6.8888,
"mean_token_accuracy": 0.10876853093504905,
"num_tokens": 4417701.0,
"step": 2030
},
{
"entropy": 7.097174835205078,
"epoch": 0.2177751618599176,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998610915133533,
"loss": 6.949,
"mean_token_accuracy": 0.11059459373354912,
"num_tokens": 4429309.0,
"step": 2035
},
{
"entropy": 7.08780870437622,
"epoch": 0.21831023596768154,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998597449962292,
"loss": 7.0251,
"mean_token_accuracy": 0.10617428198456764,
"num_tokens": 4440850.0,
"step": 2040
},
{
"entropy": 7.156359386444092,
"epoch": 0.21884531007544544,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998583919863115,
"loss": 7.0009,
"mean_token_accuracy": 0.10219309031963349,
"num_tokens": 4451995.0,
"step": 2045
},
{
"entropy": 6.987979745864868,
"epoch": 0.21938038418320938,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998570324836393,
"loss": 6.9059,
"mean_token_accuracy": 0.11296729370951653,
"num_tokens": 4461916.0,
"step": 2050
},
{
"entropy": 7.093602132797241,
"epoch": 0.2199154582909733,
"grad_norm": 1.03125,
"learning_rate": 0.000499855666488252,
"loss": 6.9953,
"mean_token_accuracy": 0.1073579765856266,
"num_tokens": 4471603.0,
"step": 2055
},
{
"entropy": 7.243840980529785,
"epoch": 0.22045053239873721,
"grad_norm": 0.90625,
"learning_rate": 0.0004998542940001889,
"loss": 7.0352,
"mean_token_accuracy": 0.10500341430306434,
"num_tokens": 4483166.0,
"step": 2060
},
{
"entropy": 7.074850082397461,
"epoch": 0.22098560650650115,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998529150194895,
"loss": 6.9424,
"mean_token_accuracy": 0.1041251115500927,
"num_tokens": 4493824.0,
"step": 2065
},
{
"entropy": 7.116635465621949,
"epoch": 0.22152068061426508,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998515295461939,
"loss": 6.837,
"mean_token_accuracy": 0.1261841081082821,
"num_tokens": 4505674.0,
"step": 2070
},
{
"entropy": 7.0095867156982425,
"epoch": 0.22205575472202901,
"grad_norm": 0.88671875,
"learning_rate": 0.000499850137580342,
"loss": 6.9422,
"mean_token_accuracy": 0.11334002017974854,
"num_tokens": 4516516.0,
"step": 2075
},
{
"entropy": 7.119809246063232,
"epoch": 0.22259082882979292,
"grad_norm": 0.9375,
"learning_rate": 0.0004998487391219739,
"loss": 6.9453,
"mean_token_accuracy": 0.10056448429822921,
"num_tokens": 4527811.0,
"step": 2080
},
{
"entropy": 7.050966739654541,
"epoch": 0.22312590293755685,
"grad_norm": 0.8203125,
"learning_rate": 0.0004998473341711301,
"loss": 6.8806,
"mean_token_accuracy": 0.10502795204520225,
"num_tokens": 4538988.0,
"step": 2085
},
{
"entropy": 7.119756889343262,
"epoch": 0.22366097704532079,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998459227278512,
"loss": 6.8997,
"mean_token_accuracy": 0.11361236721277237,
"num_tokens": 4550049.0,
"step": 2090
},
{
"entropy": 7.177942180633545,
"epoch": 0.2241960511530847,
"grad_norm": 0.96875,
"learning_rate": 0.0004998445047921777,
"loss": 7.0332,
"mean_token_accuracy": 0.10892037376761436,
"num_tokens": 4559784.0,
"step": 2095
},
{
"entropy": 7.094767713546753,
"epoch": 0.22473112526084862,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998430803641509,
"loss": 6.873,
"mean_token_accuracy": 0.11253026425838471,
"num_tokens": 4569982.0,
"step": 2100
},
{
"entropy": 6.99695782661438,
"epoch": 0.22526619936861256,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998416494438118,
"loss": 6.8769,
"mean_token_accuracy": 0.11211913973093032,
"num_tokens": 4581117.0,
"step": 2105
},
{
"entropy": 7.084139013290406,
"epoch": 0.2258012734763765,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998402120312014,
"loss": 6.9462,
"mean_token_accuracy": 0.11116094216704368,
"num_tokens": 4592012.0,
"step": 2110
},
{
"entropy": 7.116655254364014,
"epoch": 0.2263363475841404,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998387681263616,
"loss": 6.9931,
"mean_token_accuracy": 0.10694418773055077,
"num_tokens": 4602591.0,
"step": 2115
},
{
"entropy": 7.098616075515747,
"epoch": 0.22687142169190433,
"grad_norm": 0.8515625,
"learning_rate": 0.000499837317729334,
"loss": 6.9291,
"mean_token_accuracy": 0.10939419046044349,
"num_tokens": 4613418.0,
"step": 2120
},
{
"entropy": 7.091100645065308,
"epoch": 0.22740649579966826,
"grad_norm": 0.796875,
"learning_rate": 0.0004998358608401603,
"loss": 6.8499,
"mean_token_accuracy": 0.11751434430480004,
"num_tokens": 4625156.0,
"step": 2125
},
{
"entropy": 7.1228173732757565,
"epoch": 0.22794156990743217,
"grad_norm": 0.859375,
"learning_rate": 0.000499834397458883,
"loss": 6.8865,
"mean_token_accuracy": 0.11367959231138229,
"num_tokens": 4634523.0,
"step": 2130
},
{
"entropy": 6.990410232543946,
"epoch": 0.2284766440151961,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998329275855438,
"loss": 6.87,
"mean_token_accuracy": 0.11853560954332351,
"num_tokens": 4646292.0,
"step": 2135
},
{
"entropy": 7.091177034378052,
"epoch": 0.22901171812296003,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998314512201855,
"loss": 6.8548,
"mean_token_accuracy": 0.10857467800378799,
"num_tokens": 4656417.0,
"step": 2140
},
{
"entropy": 7.000166225433349,
"epoch": 0.22954679223072397,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998299683628507,
"loss": 6.8661,
"mean_token_accuracy": 0.11203439608216285,
"num_tokens": 4666933.0,
"step": 2145
},
{
"entropy": 7.122287797927856,
"epoch": 0.23008186633848787,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998284790135821,
"loss": 7.0464,
"mean_token_accuracy": 0.10003090128302575,
"num_tokens": 4678482.0,
"step": 2150
},
{
"entropy": 7.111646604537964,
"epoch": 0.2306169404462518,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998269831724228,
"loss": 6.894,
"mean_token_accuracy": 0.11034037619829178,
"num_tokens": 4688454.0,
"step": 2155
},
{
"entropy": 7.016481971740722,
"epoch": 0.23115201455401574,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998254808394159,
"loss": 6.9201,
"mean_token_accuracy": 0.11088593229651451,
"num_tokens": 4699545.0,
"step": 2160
},
{
"entropy": 7.125952577590942,
"epoch": 0.23168708866177964,
"grad_norm": 0.7578125,
"learning_rate": 0.0004998239720146048,
"loss": 6.9256,
"mean_token_accuracy": 0.11005692854523659,
"num_tokens": 4711149.0,
"step": 2165
},
{
"entropy": 6.9970673561096195,
"epoch": 0.23222216276954358,
"grad_norm": 0.83984375,
"learning_rate": 0.0004998224566980332,
"loss": 6.9387,
"mean_token_accuracy": 0.10672764554619789,
"num_tokens": 4722651.0,
"step": 2170
},
{
"entropy": 7.063665342330933,
"epoch": 0.2327572368773075,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998209348897447,
"loss": 6.9388,
"mean_token_accuracy": 0.10191527009010315,
"num_tokens": 4733883.0,
"step": 2175
},
{
"entropy": 7.244954586029053,
"epoch": 0.23329231098507144,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998194065897833,
"loss": 6.9158,
"mean_token_accuracy": 0.10730345025658608,
"num_tokens": 4744517.0,
"step": 2180
},
{
"entropy": 7.010083484649658,
"epoch": 0.23382738509283535,
"grad_norm": 0.9375,
"learning_rate": 0.000499817871798193,
"loss": 6.9234,
"mean_token_accuracy": 0.1180235929787159,
"num_tokens": 4755481.0,
"step": 2185
},
{
"entropy": 7.012146234512329,
"epoch": 0.23436245920059928,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998163305150185,
"loss": 6.8316,
"mean_token_accuracy": 0.11478903293609619,
"num_tokens": 4765535.0,
"step": 2190
},
{
"entropy": 7.069972801208496,
"epoch": 0.23489753330836322,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998147827403038,
"loss": 6.9619,
"mean_token_accuracy": 0.10954375863075257,
"num_tokens": 4777045.0,
"step": 2195
},
{
"entropy": 7.119951629638672,
"epoch": 0.23543260741612712,
"grad_norm": 0.90625,
"learning_rate": 0.000499813228474094,
"loss": 6.8399,
"mean_token_accuracy": 0.1167965590953827,
"num_tokens": 4787958.0,
"step": 2200
},
{
"entropy": 6.953368234634399,
"epoch": 0.23596768152389105,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998116677164338,
"loss": 6.8911,
"mean_token_accuracy": 0.10928420349955559,
"num_tokens": 4798502.0,
"step": 2205
},
{
"entropy": 7.026491498947143,
"epoch": 0.236502755631655,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998101004673682,
"loss": 6.8906,
"mean_token_accuracy": 0.11567377597093582,
"num_tokens": 4809179.0,
"step": 2210
},
{
"entropy": 7.102079772949219,
"epoch": 0.23703782973941892,
"grad_norm": 0.96875,
"learning_rate": 0.0004998085267269426,
"loss": 6.9476,
"mean_token_accuracy": 0.10375262200832366,
"num_tokens": 4819803.0,
"step": 2215
},
{
"entropy": 7.111145877838135,
"epoch": 0.23757290384718283,
"grad_norm": 0.80859375,
"learning_rate": 0.0004998069464952025,
"loss": 6.9275,
"mean_token_accuracy": 0.1055069126188755,
"num_tokens": 4831226.0,
"step": 2220
},
{
"entropy": 7.114426469802856,
"epoch": 0.23810797795494676,
"grad_norm": 0.890625,
"learning_rate": 0.0004998053597721932,
"loss": 6.955,
"mean_token_accuracy": 0.11139222607016563,
"num_tokens": 4842168.0,
"step": 2225
},
{
"entropy": 7.013771915435791,
"epoch": 0.2386430520627107,
"grad_norm": 0.875,
"learning_rate": 0.0004998037665579608,
"loss": 6.9151,
"mean_token_accuracy": 0.11002311557531357,
"num_tokens": 4852655.0,
"step": 2230
},
{
"entropy": 6.965934324264526,
"epoch": 0.2391781261704746,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998021668525513,
"loss": 6.8002,
"mean_token_accuracy": 0.11136664599180221,
"num_tokens": 4862164.0,
"step": 2235
},
{
"entropy": 7.1173601150512695,
"epoch": 0.23971320027823853,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998005606560107,
"loss": 6.9347,
"mean_token_accuracy": 0.11432894393801689,
"num_tokens": 4874233.0,
"step": 2240
},
{
"entropy": 7.07620940208435,
"epoch": 0.24024827438600246,
"grad_norm": 0.828125,
"learning_rate": 0.0004997989479683856,
"loss": 6.924,
"mean_token_accuracy": 0.11140381470322609,
"num_tokens": 4885948.0,
"step": 2245
},
{
"entropy": 7.045759534835815,
"epoch": 0.2407833484937664,
"grad_norm": 0.80859375,
"learning_rate": 0.0004997973287897224,
"loss": 6.9123,
"mean_token_accuracy": 0.11342343240976334,
"num_tokens": 4897128.0,
"step": 2250
},
{
"entropy": 7.03762469291687,
"epoch": 0.2413184226015303,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997957031200681,
"loss": 6.8835,
"mean_token_accuracy": 0.11781226769089699,
"num_tokens": 4907699.0,
"step": 2255
},
{
"entropy": 6.970374584197998,
"epoch": 0.24185349670929424,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997940709594692,
"loss": 6.806,
"mean_token_accuracy": 0.11417917162179947,
"num_tokens": 4918772.0,
"step": 2260
},
{
"entropy": 7.130536031723023,
"epoch": 0.24238857081705817,
"grad_norm": 0.96875,
"learning_rate": 0.0004997924323079733,
"loss": 6.8628,
"mean_token_accuracy": 0.10981507375836372,
"num_tokens": 4930046.0,
"step": 2265
},
{
"entropy": 6.969363880157471,
"epoch": 0.24292364492482207,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997907871656273,
"loss": 6.8944,
"mean_token_accuracy": 0.11376282870769501,
"num_tokens": 4940609.0,
"step": 2270
},
{
"entropy": 7.0445019721984865,
"epoch": 0.243458719032586,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997891355324791,
"loss": 6.9047,
"mean_token_accuracy": 0.11144462153315544,
"num_tokens": 4951339.0,
"step": 2275
},
{
"entropy": 7.013249683380127,
"epoch": 0.24399379314034994,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997874774085761,
"loss": 6.8215,
"mean_token_accuracy": 0.10943349301815034,
"num_tokens": 4961083.0,
"step": 2280
},
{
"entropy": 6.997879314422607,
"epoch": 0.24452886724811387,
"grad_norm": 0.96875,
"learning_rate": 0.0004997858127939662,
"loss": 6.824,
"mean_token_accuracy": 0.1134270079433918,
"num_tokens": 4971399.0,
"step": 2285
},
{
"entropy": 7.0827906131744385,
"epoch": 0.24506394135587778,
"grad_norm": 0.86328125,
"learning_rate": 0.0004997841416886976,
"loss": 6.9026,
"mean_token_accuracy": 0.10839502215385437,
"num_tokens": 4982429.0,
"step": 2290
},
{
"entropy": 7.065428256988525,
"epoch": 0.2455990154636417,
"grad_norm": 0.90625,
"learning_rate": 0.0004997824640928186,
"loss": 6.9601,
"mean_token_accuracy": 0.10927876681089402,
"num_tokens": 4993706.0,
"step": 2295
},
{
"entropy": 7.042297983169556,
"epoch": 0.24613408957140565,
"grad_norm": 0.83984375,
"learning_rate": 0.0004997807800063774,
"loss": 6.8437,
"mean_token_accuracy": 0.11474441438913345,
"num_tokens": 5005411.0,
"step": 2300
},
{
"entropy": 7.061811208724976,
"epoch": 0.24666916367916958,
"grad_norm": 0.93359375,
"learning_rate": 0.000499779089429423,
"loss": 6.963,
"mean_token_accuracy": 0.1087774895131588,
"num_tokens": 5016011.0,
"step": 2305
},
{
"entropy": 7.043573570251465,
"epoch": 0.24720423778693348,
"grad_norm": 0.828125,
"learning_rate": 0.0004997773923620037,
"loss": 6.9238,
"mean_token_accuracy": 0.10597949847579002,
"num_tokens": 5027841.0,
"step": 2310
},
{
"entropy": 7.102751350402832,
"epoch": 0.24773931189469742,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997756888041689,
"loss": 6.9271,
"mean_token_accuracy": 0.10425131991505623,
"num_tokens": 5038640.0,
"step": 2315
},
{
"entropy": 6.972273588180542,
"epoch": 0.24827438600246135,
"grad_norm": 0.87890625,
"learning_rate": 0.0004997739787559677,
"loss": 6.8668,
"mean_token_accuracy": 0.11417348384857177,
"num_tokens": 5050423.0,
"step": 2320
},
{
"entropy": 7.035957622528076,
"epoch": 0.24880946011022526,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997722622174494,
"loss": 6.8771,
"mean_token_accuracy": 0.11295794397592544,
"num_tokens": 5060654.0,
"step": 2325
},
{
"entropy": 6.993176651000977,
"epoch": 0.2493445342179892,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997705391886636,
"loss": 6.799,
"mean_token_accuracy": 0.11571635380387306,
"num_tokens": 5070165.0,
"step": 2330
},
{
"entropy": 6.959088754653931,
"epoch": 0.24987960832575312,
"grad_norm": 0.921875,
"learning_rate": 0.0004997688096696601,
"loss": 6.8572,
"mean_token_accuracy": 0.1051781103014946,
"num_tokens": 5080937.0,
"step": 2335
},
{
"entropy": 6.938587760925293,
"epoch": 0.25041468243351706,
"grad_norm": 0.8671875,
"learning_rate": 0.0004997670736604888,
"loss": 6.808,
"mean_token_accuracy": 0.11930364593863488,
"num_tokens": 5090955.0,
"step": 2340
},
{
"entropy": 7.089204788208008,
"epoch": 0.250949756541281,
"grad_norm": 0.8203125,
"learning_rate": 0.0004997653311611998,
"loss": 6.9384,
"mean_token_accuracy": 0.10624079629778863,
"num_tokens": 5102201.0,
"step": 2345
},
{
"entropy": 7.0677807331085205,
"epoch": 0.25148483064904487,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997635821718435,
"loss": 6.8653,
"mean_token_accuracy": 0.11593251675367355,
"num_tokens": 5112235.0,
"step": 2350
},
{
"entropy": 6.948492336273193,
"epoch": 0.2520199047568088,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997618266924704,
"loss": 6.7418,
"mean_token_accuracy": 0.11052383631467819,
"num_tokens": 5122360.0,
"step": 2355
},
{
"entropy": 7.005466365814209,
"epoch": 0.25255497886457273,
"grad_norm": 0.9453125,
"learning_rate": 0.000499760064723131,
"loss": 6.9121,
"mean_token_accuracy": 0.10666457489132881,
"num_tokens": 5133084.0,
"step": 2360
},
{
"entropy": 6.970629024505615,
"epoch": 0.25309005297233667,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997582962638764,
"loss": 6.8565,
"mean_token_accuracy": 0.1083373948931694,
"num_tokens": 5144495.0,
"step": 2365
},
{
"entropy": 7.105882930755615,
"epoch": 0.2536251270801006,
"grad_norm": 0.890625,
"learning_rate": 0.0004997565213147575,
"loss": 6.865,
"mean_token_accuracy": 0.11402732953429222,
"num_tokens": 5154824.0,
"step": 2370
},
{
"entropy": 6.978338050842285,
"epoch": 0.25416020118786453,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997547398758257,
"loss": 6.8449,
"mean_token_accuracy": 0.10908161997795104,
"num_tokens": 5165061.0,
"step": 2375
},
{
"entropy": 6.9950824737548825,
"epoch": 0.25469527529562846,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997529519471323,
"loss": 6.8105,
"mean_token_accuracy": 0.11739344000816346,
"num_tokens": 5174964.0,
"step": 2380
},
{
"entropy": 6.959466028213501,
"epoch": 0.25523034940339234,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997511575287291,
"loss": 6.8388,
"mean_token_accuracy": 0.11174839809536934,
"num_tokens": 5185458.0,
"step": 2385
},
{
"entropy": 7.014845609664917,
"epoch": 0.2557654235111563,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997493566206677,
"loss": 6.8452,
"mean_token_accuracy": 0.11382765620946884,
"num_tokens": 5196193.0,
"step": 2390
},
{
"entropy": 6.993070125579834,
"epoch": 0.2563004976189202,
"grad_norm": 0.8125,
"learning_rate": 0.0004997475492230003,
"loss": 6.9517,
"mean_token_accuracy": 0.10821824818849564,
"num_tokens": 5208541.0,
"step": 2395
},
{
"entropy": 7.031092548370362,
"epoch": 0.25683557172668414,
"grad_norm": 0.859375,
"learning_rate": 0.000499745735335779,
"loss": 6.7908,
"mean_token_accuracy": 0.11512782797217369,
"num_tokens": 5218567.0,
"step": 2400
},
{
"entropy": 6.993920040130615,
"epoch": 0.2573706458344481,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997439149590561,
"loss": 6.8965,
"mean_token_accuracy": 0.11021987870335578,
"num_tokens": 5228293.0,
"step": 2405
},
{
"entropy": 6.987144422531128,
"epoch": 0.257905719942212,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997420880928843,
"loss": 6.8368,
"mean_token_accuracy": 0.11468688249588013,
"num_tokens": 5238880.0,
"step": 2410
},
{
"entropy": 7.0282874584198,
"epoch": 0.25844079404997594,
"grad_norm": 0.86328125,
"learning_rate": 0.0004997402547373162,
"loss": 6.9248,
"mean_token_accuracy": 0.10779067873954773,
"num_tokens": 5250104.0,
"step": 2415
},
{
"entropy": 7.077765417098999,
"epoch": 0.2589758681577399,
"grad_norm": 0.8828125,
"learning_rate": 0.0004997384148924049,
"loss": 6.9476,
"mean_token_accuracy": 0.10701018646359443,
"num_tokens": 5260355.0,
"step": 2420
},
{
"entropy": 7.013334131240844,
"epoch": 0.25951094226550375,
"grad_norm": 0.921875,
"learning_rate": 0.0004997365685582035,
"loss": 6.8839,
"mean_token_accuracy": 0.11221425458788872,
"num_tokens": 5272117.0,
"step": 2425
},
{
"entropy": 7.025537967681885,
"epoch": 0.2600460163732677,
"grad_norm": 0.7890625,
"learning_rate": 0.0004997347157347651,
"loss": 6.9641,
"mean_token_accuracy": 0.11001334488391876,
"num_tokens": 5284583.0,
"step": 2430
},
{
"entropy": 7.070367336273193,
"epoch": 0.2605810904810316,
"grad_norm": 0.83984375,
"learning_rate": 0.0004997328564221435,
"loss": 6.7554,
"mean_token_accuracy": 0.12274964973330497,
"num_tokens": 5295846.0,
"step": 2435
},
{
"entropy": 6.974102544784546,
"epoch": 0.26111616458879555,
"grad_norm": 0.8515625,
"learning_rate": 0.0004997309906203922,
"loss": 6.9064,
"mean_token_accuracy": 0.10685541406273842,
"num_tokens": 5307802.0,
"step": 2440
},
{
"entropy": 7.062164688110352,
"epoch": 0.2616512386965595,
"grad_norm": 0.87890625,
"learning_rate": 0.0004997291183295652,
"loss": 6.8217,
"mean_token_accuracy": 0.11616088747978211,
"num_tokens": 5318263.0,
"step": 2445
},
{
"entropy": 7.043830013275146,
"epoch": 0.2621863128043234,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997272395497164,
"loss": 6.8821,
"mean_token_accuracy": 0.11346295177936554,
"num_tokens": 5328555.0,
"step": 2450
},
{
"entropy": 6.941327333450317,
"epoch": 0.26272138691208735,
"grad_norm": 0.828125,
"learning_rate": 0.0004997253542809002,
"loss": 6.8354,
"mean_token_accuracy": 0.11198842748999596,
"num_tokens": 5339954.0,
"step": 2455
},
{
"entropy": 7.107675504684448,
"epoch": 0.26325646101985123,
"grad_norm": 0.8359375,
"learning_rate": 0.0004997234625231709,
"loss": 6.8836,
"mean_token_accuracy": 0.11381551921367646,
"num_tokens": 5350919.0,
"step": 2460
},
{
"entropy": 6.949951696395874,
"epoch": 0.26379153512761516,
"grad_norm": 0.8203125,
"learning_rate": 0.0004997215642765832,
"loss": 6.8624,
"mean_token_accuracy": 0.11542123556137085,
"num_tokens": 5361844.0,
"step": 2465
},
{
"entropy": 7.018435907363892,
"epoch": 0.2643266092353791,
"grad_norm": 0.875,
"learning_rate": 0.000499719659541192,
"loss": 6.8726,
"mean_token_accuracy": 0.10965746343135833,
"num_tokens": 5372026.0,
"step": 2470
},
{
"entropy": 7.0428282737731935,
"epoch": 0.26486168334314303,
"grad_norm": 0.8203125,
"learning_rate": 0.0004997177483170521,
"loss": 6.9163,
"mean_token_accuracy": 0.11546080112457276,
"num_tokens": 5382965.0,
"step": 2475
},
{
"entropy": 6.963793182373047,
"epoch": 0.26539675745090696,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997158306042188,
"loss": 6.874,
"mean_token_accuracy": 0.11316567808389663,
"num_tokens": 5393738.0,
"step": 2480
},
{
"entropy": 6.974194860458374,
"epoch": 0.2659318315586709,
"grad_norm": 0.90625,
"learning_rate": 0.0004997139064027476,
"loss": 6.74,
"mean_token_accuracy": 0.12156007960438728,
"num_tokens": 5404059.0,
"step": 2485
},
{
"entropy": 6.964081430435181,
"epoch": 0.26646690566643483,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997119757126938,
"loss": 6.7804,
"mean_token_accuracy": 0.11532729938626289,
"num_tokens": 5414152.0,
"step": 2490
},
{
"entropy": 7.028856039047241,
"epoch": 0.2670019797741987,
"grad_norm": 0.890625,
"learning_rate": 0.0004997100385341133,
"loss": 6.9634,
"mean_token_accuracy": 0.1126515194773674,
"num_tokens": 5425227.0,
"step": 2495
},
{
"entropy": 6.9782177925109865,
"epoch": 0.26753705388196264,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997080948670621,
"loss": 6.789,
"mean_token_accuracy": 0.11695293188095093,
"num_tokens": 5435716.0,
"step": 2500
},
{
"entropy": 6.985808944702148,
"epoch": 0.26807212798972657,
"grad_norm": 0.82421875,
"learning_rate": 0.0004997061447115962,
"loss": 6.8623,
"mean_token_accuracy": 0.10614107474684716,
"num_tokens": 5446519.0,
"step": 2505
},
{
"entropy": 7.017519235610962,
"epoch": 0.2686072020974905,
"grad_norm": 0.87890625,
"learning_rate": 0.0004997041880677719,
"loss": 6.8421,
"mean_token_accuracy": 0.11282496824860573,
"num_tokens": 5457496.0,
"step": 2510
},
{
"entropy": 6.900492763519287,
"epoch": 0.26914227620525444,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997022249356457,
"loss": 6.7495,
"mean_token_accuracy": 0.11760077178478241,
"num_tokens": 5469059.0,
"step": 2515
},
{
"entropy": 7.010392713546753,
"epoch": 0.26967735031301837,
"grad_norm": 1.1328125,
"learning_rate": 0.0004997002553152745,
"loss": 6.8399,
"mean_token_accuracy": 0.1156559094786644,
"num_tokens": 5480324.0,
"step": 2520
},
{
"entropy": 7.005701923370362,
"epoch": 0.2702124244207823,
"grad_norm": 0.875,
"learning_rate": 0.0004996982792067148,
"loss": 6.845,
"mean_token_accuracy": 0.1095408894121647,
"num_tokens": 5491513.0,
"step": 2525
},
{
"entropy": 6.884131574630738,
"epoch": 0.2707474985285462,
"grad_norm": 0.84765625,
"learning_rate": 0.000499696296610024,
"loss": 6.7724,
"mean_token_accuracy": 0.11539606228470803,
"num_tokens": 5502414.0,
"step": 2530
},
{
"entropy": 6.973390340805054,
"epoch": 0.2712825726363101,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996943075252592,
"loss": 6.8968,
"mean_token_accuracy": 0.11106460094451905,
"num_tokens": 5512962.0,
"step": 2535
},
{
"entropy": 6.999091625213623,
"epoch": 0.27181764674407405,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996923119524777,
"loss": 6.7602,
"mean_token_accuracy": 0.11408493369817733,
"num_tokens": 5524648.0,
"step": 2540
},
{
"entropy": 6.898416757583618,
"epoch": 0.272352720851838,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996903098917375,
"loss": 6.8386,
"mean_token_accuracy": 0.10790063142776489,
"num_tokens": 5536158.0,
"step": 2545
},
{
"entropy": 6.9561662673950195,
"epoch": 0.2728877949596019,
"grad_norm": 0.8828125,
"learning_rate": 0.000499688301343096,
"loss": 6.7565,
"mean_token_accuracy": 0.11405683234333992,
"num_tokens": 5547374.0,
"step": 2550
},
{
"entropy": 7.05834732055664,
"epoch": 0.27342286906736585,
"grad_norm": 0.8515625,
"learning_rate": 0.0004996862863066114,
"loss": 6.8386,
"mean_token_accuracy": 0.11390230059623718,
"num_tokens": 5559212.0,
"step": 2555
},
{
"entropy": 6.979931354522705,
"epoch": 0.2739579431751298,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996842647823419,
"loss": 6.8098,
"mean_token_accuracy": 0.11664133742451668,
"num_tokens": 5569302.0,
"step": 2560
},
{
"entropy": 6.945258140563965,
"epoch": 0.27449301728289366,
"grad_norm": 0.87109375,
"learning_rate": 0.0004996822367703458,
"loss": 6.8112,
"mean_token_accuracy": 0.10886863917112351,
"num_tokens": 5581688.0,
"step": 2565
},
{
"entropy": 7.029229307174683,
"epoch": 0.2750280913906576,
"grad_norm": 0.953125,
"learning_rate": 0.0004996802022706817,
"loss": 6.7402,
"mean_token_accuracy": 0.11893118396401406,
"num_tokens": 5591376.0,
"step": 2570
},
{
"entropy": 6.872675561904908,
"epoch": 0.2755631654984215,
"grad_norm": 0.90625,
"learning_rate": 0.0004996781612834083,
"loss": 6.7905,
"mean_token_accuracy": 0.11670346409082413,
"num_tokens": 5601510.0,
"step": 2575
},
{
"entropy": 7.052143335342407,
"epoch": 0.27609823960618546,
"grad_norm": 1.0,
"learning_rate": 0.0004996761138085846,
"loss": 6.868,
"mean_token_accuracy": 0.1093013696372509,
"num_tokens": 5611436.0,
"step": 2580
},
{
"entropy": 6.98383355140686,
"epoch": 0.2766333137139494,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996740598462697,
"loss": 6.8517,
"mean_token_accuracy": 0.11311317384243011,
"num_tokens": 5620470.0,
"step": 2585
},
{
"entropy": 6.922229051589966,
"epoch": 0.2771683878217133,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996719993965229,
"loss": 6.7627,
"mean_token_accuracy": 0.11485590264201165,
"num_tokens": 5630740.0,
"step": 2590
},
{
"entropy": 6.873901319503784,
"epoch": 0.27770346192947726,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996699324594036,
"loss": 6.7591,
"mean_token_accuracy": 0.11999578103423118,
"num_tokens": 5641138.0,
"step": 2595
},
{
"entropy": 7.053933334350586,
"epoch": 0.27823853603724114,
"grad_norm": 0.84765625,
"learning_rate": 0.0004996678590349717,
"loss": 6.89,
"mean_token_accuracy": 0.11600473672151565,
"num_tokens": 5652474.0,
"step": 2600
},
{
"entropy": 6.952626085281372,
"epoch": 0.27877361014500507,
"grad_norm": 0.921875,
"learning_rate": 0.0004996657791232868,
"loss": 6.7958,
"mean_token_accuracy": 0.11903122812509537,
"num_tokens": 5663178.0,
"step": 2605
},
{
"entropy": 6.984945917129517,
"epoch": 0.279308684252769,
"grad_norm": 0.984375,
"learning_rate": 0.0004996636927244092,
"loss": 6.8051,
"mean_token_accuracy": 0.11196058839559556,
"num_tokens": 5673565.0,
"step": 2610
},
{
"entropy": 7.001294422149658,
"epoch": 0.27984375836053293,
"grad_norm": 0.87109375,
"learning_rate": 0.000499661599838399,
"loss": 6.8497,
"mean_token_accuracy": 0.1122498370707035,
"num_tokens": 5684651.0,
"step": 2615
},
{
"entropy": 6.904658508300781,
"epoch": 0.28037883246829687,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996595004653167,
"loss": 6.7432,
"mean_token_accuracy": 0.11533383056521415,
"num_tokens": 5696081.0,
"step": 2620
},
{
"entropy": 6.9526012420654295,
"epoch": 0.2809139065760608,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996573946052229,
"loss": 6.7812,
"mean_token_accuracy": 0.1164487712085247,
"num_tokens": 5706830.0,
"step": 2625
},
{
"entropy": 6.8979075908660885,
"epoch": 0.28144898068382473,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996552822581783,
"loss": 6.7986,
"mean_token_accuracy": 0.11754084527492523,
"num_tokens": 5717261.0,
"step": 2630
},
{
"entropy": 7.005229568481445,
"epoch": 0.2819840547915886,
"grad_norm": 1.0703125,
"learning_rate": 0.000499653163424244,
"loss": 6.7734,
"mean_token_accuracy": 0.11206443011760711,
"num_tokens": 5727280.0,
"step": 2635
},
{
"entropy": 6.993945360183716,
"epoch": 0.28251912889935255,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996510381034814,
"loss": 6.8241,
"mean_token_accuracy": 0.10772419720888138,
"num_tokens": 5737485.0,
"step": 2640
},
{
"entropy": 6.9192140102386475,
"epoch": 0.2830542030071165,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996489062959515,
"loss": 6.7957,
"mean_token_accuracy": 0.11468210741877556,
"num_tokens": 5747683.0,
"step": 2645
},
{
"entropy": 6.915011835098267,
"epoch": 0.2835892771148804,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996467680017159,
"loss": 6.862,
"mean_token_accuracy": 0.10823372080922126,
"num_tokens": 5758148.0,
"step": 2650
},
{
"entropy": 6.931305360794068,
"epoch": 0.28412435122264434,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996446232208365,
"loss": 6.7896,
"mean_token_accuracy": 0.11629763916134835,
"num_tokens": 5769109.0,
"step": 2655
},
{
"entropy": 6.928414344787598,
"epoch": 0.2846594253304083,
"grad_norm": 0.953125,
"learning_rate": 0.0004996424719533753,
"loss": 6.8193,
"mean_token_accuracy": 0.11202871650457383,
"num_tokens": 5780344.0,
"step": 2660
},
{
"entropy": 6.936452388763428,
"epoch": 0.2851944994381722,
"grad_norm": 0.96875,
"learning_rate": 0.0004996403141993941,
"loss": 6.627,
"mean_token_accuracy": 0.12468865811824799,
"num_tokens": 5790167.0,
"step": 2665
},
{
"entropy": 6.922191047668457,
"epoch": 0.2857295735459361,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996381499589555,
"loss": 6.791,
"mean_token_accuracy": 0.11766616478562356,
"num_tokens": 5800076.0,
"step": 2670
},
{
"entropy": 7.001429510116577,
"epoch": 0.2862646476537,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996359792321219,
"loss": 6.7247,
"mean_token_accuracy": 0.12332994639873504,
"num_tokens": 5810547.0,
"step": 2675
},
{
"entropy": 6.842127323150635,
"epoch": 0.28679972176146395,
"grad_norm": 0.94140625,
"learning_rate": 0.000499633802018956,
"loss": 6.7651,
"mean_token_accuracy": 0.11882570460438728,
"num_tokens": 5821210.0,
"step": 2680
},
{
"entropy": 6.933741283416748,
"epoch": 0.2873347958692279,
"grad_norm": 0.90234375,
"learning_rate": 0.0004996316183195204,
"loss": 6.8064,
"mean_token_accuracy": 0.11239668801426887,
"num_tokens": 5832361.0,
"step": 2685
},
{
"entropy": 6.997045278549194,
"epoch": 0.2878698699769918,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996294281338786,
"loss": 6.7965,
"mean_token_accuracy": 0.11312427893280982,
"num_tokens": 5843242.0,
"step": 2690
},
{
"entropy": 6.942799377441406,
"epoch": 0.28840494408475575,
"grad_norm": 0.83984375,
"learning_rate": 0.0004996272314620936,
"loss": 6.8534,
"mean_token_accuracy": 0.11269704699516296,
"num_tokens": 5855163.0,
"step": 2695
},
{
"entropy": 6.908647918701172,
"epoch": 0.2889400181925197,
"grad_norm": 0.87109375,
"learning_rate": 0.0004996250283042288,
"loss": 6.7873,
"mean_token_accuracy": 0.11659785062074661,
"num_tokens": 5866908.0,
"step": 2700
},
{
"entropy": 6.983246564865112,
"epoch": 0.28947509230028357,
"grad_norm": 0.9921875,
"learning_rate": 0.000499622818660348,
"loss": 6.8308,
"mean_token_accuracy": 0.11280549988150597,
"num_tokens": 5877342.0,
"step": 2705
},
{
"entropy": 6.902181339263916,
"epoch": 0.2900101664080475,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996206025305147,
"loss": 6.7809,
"mean_token_accuracy": 0.11292893514037132,
"num_tokens": 5887960.0,
"step": 2710
},
{
"entropy": 6.913876533508301,
"epoch": 0.29054524051581143,
"grad_norm": 0.90625,
"learning_rate": 0.0004996183799147932,
"loss": 6.7618,
"mean_token_accuracy": 0.11444101110100746,
"num_tokens": 5899887.0,
"step": 2715
},
{
"entropy": 6.982015180587768,
"epoch": 0.29108031462357536,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996161508132475,
"loss": 6.7616,
"mean_token_accuracy": 0.10974067524075508,
"num_tokens": 5911022.0,
"step": 2720
},
{
"entropy": 6.94197039604187,
"epoch": 0.2916153887313393,
"grad_norm": 0.88671875,
"learning_rate": 0.000499613915225942,
"loss": 6.8321,
"mean_token_accuracy": 0.11855507642030716,
"num_tokens": 5922780.0,
"step": 2725
},
{
"entropy": 6.964428901672363,
"epoch": 0.29215046283910323,
"grad_norm": 0.984375,
"learning_rate": 0.0004996116731529412,
"loss": 6.8632,
"mean_token_accuracy": 0.11426409929990769,
"num_tokens": 5934112.0,
"step": 2730
},
{
"entropy": 6.972011709213257,
"epoch": 0.29268553694686716,
"grad_norm": 0.953125,
"learning_rate": 0.00049960942459431,
"loss": 6.8045,
"mean_token_accuracy": 0.1121431604027748,
"num_tokens": 5944386.0,
"step": 2735
},
{
"entropy": 6.87826681137085,
"epoch": 0.29322061105463104,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996071695501132,
"loss": 6.7394,
"mean_token_accuracy": 0.1182074099779129,
"num_tokens": 5955552.0,
"step": 2740
},
{
"entropy": 6.928379726409912,
"epoch": 0.293755685162395,
"grad_norm": 0.828125,
"learning_rate": 0.0004996049080204159,
"loss": 6.7346,
"mean_token_accuracy": 0.11500842422246933,
"num_tokens": 5966790.0,
"step": 2745
},
{
"entropy": 6.979641532897949,
"epoch": 0.2942907592701589,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996026400052835,
"loss": 6.7392,
"mean_token_accuracy": 0.11592580303549767,
"num_tokens": 5977704.0,
"step": 2750
},
{
"entropy": 6.8253716945648195,
"epoch": 0.29482583337792284,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996003655047814,
"loss": 6.7484,
"mean_token_accuracy": 0.11899459958076478,
"num_tokens": 5989221.0,
"step": 2755
},
{
"entropy": 6.908860492706299,
"epoch": 0.2953609074856868,
"grad_norm": 0.80078125,
"learning_rate": 0.0004995980845189753,
"loss": 6.8069,
"mean_token_accuracy": 0.11177898719906806,
"num_tokens": 6000906.0,
"step": 2760
},
{
"entropy": 6.967212343215943,
"epoch": 0.2958959815934507,
"grad_norm": 0.796875,
"learning_rate": 0.0004995957970479312,
"loss": 6.8099,
"mean_token_accuracy": 0.11111435741186142,
"num_tokens": 6012039.0,
"step": 2765
},
{
"entropy": 6.8845288276672365,
"epoch": 0.29643105570121464,
"grad_norm": 0.7734375,
"learning_rate": 0.0004995935030917148,
"loss": 6.6981,
"mean_token_accuracy": 0.11647267341613769,
"num_tokens": 6024383.0,
"step": 2770
},
{
"entropy": 6.988625144958496,
"epoch": 0.2969661298089785,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995912026503926,
"loss": 6.7698,
"mean_token_accuracy": 0.11584996357560158,
"num_tokens": 6034904.0,
"step": 2775
},
{
"entropy": 6.852597904205322,
"epoch": 0.29750120391674245,
"grad_norm": 0.9140625,
"learning_rate": 0.000499588895724031,
"loss": 6.7743,
"mean_token_accuracy": 0.111159298568964,
"num_tokens": 6044449.0,
"step": 2780
},
{
"entropy": 6.959954643249512,
"epoch": 0.2980362780245064,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995865823126968,
"loss": 6.7933,
"mean_token_accuracy": 0.11641821041703224,
"num_tokens": 6054465.0,
"step": 2785
},
{
"entropy": 6.91104040145874,
"epoch": 0.2985713521322703,
"grad_norm": 0.90234375,
"learning_rate": 0.0004995842624164564,
"loss": 6.8124,
"mean_token_accuracy": 0.11264995485544205,
"num_tokens": 6065539.0,
"step": 2790
},
{
"entropy": 6.946519184112549,
"epoch": 0.29910642624003425,
"grad_norm": 0.8828125,
"learning_rate": 0.000499581936035377,
"loss": 6.7705,
"mean_token_accuracy": 0.11285096481442451,
"num_tokens": 6076135.0,
"step": 2795
},
{
"entropy": 6.859121942520142,
"epoch": 0.2996415003477982,
"grad_norm": 0.85546875,
"learning_rate": 0.0004995796031695257,
"loss": 6.7652,
"mean_token_accuracy": 0.11385954692959785,
"num_tokens": 6086604.0,
"step": 2800
},
{
"entropy": 6.977310657501221,
"epoch": 0.3001765744555621,
"grad_norm": 0.94140625,
"learning_rate": 0.00049957726381897,
"loss": 6.785,
"mean_token_accuracy": 0.11666215732693672,
"num_tokens": 6095672.0,
"step": 2805
},
{
"entropy": 6.85119218826294,
"epoch": 0.300711648563326,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995749179837773,
"loss": 6.6622,
"mean_token_accuracy": 0.12312485352158546,
"num_tokens": 6105652.0,
"step": 2810
},
{
"entropy": 6.840598678588867,
"epoch": 0.30124672267108993,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995725656640155,
"loss": 6.7056,
"mean_token_accuracy": 0.1178930640220642,
"num_tokens": 6116930.0,
"step": 2815
},
{
"entropy": 6.875258922576904,
"epoch": 0.30178179677885386,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995702068597524,
"loss": 6.7157,
"mean_token_accuracy": 0.12197659760713578,
"num_tokens": 6126659.0,
"step": 2820
},
{
"entropy": 6.899388074874878,
"epoch": 0.3023168708866178,
"grad_norm": 0.890625,
"learning_rate": 0.0004995678415710561,
"loss": 6.7457,
"mean_token_accuracy": 0.115099136531353,
"num_tokens": 6137227.0,
"step": 2825
},
{
"entropy": 6.978420209884644,
"epoch": 0.3028519449943817,
"grad_norm": 0.875,
"learning_rate": 0.000499565469797995,
"loss": 6.7977,
"mean_token_accuracy": 0.1148864470422268,
"num_tokens": 6147800.0,
"step": 2830
},
{
"entropy": 6.862562036514282,
"epoch": 0.30338701910214566,
"grad_norm": 0.8671875,
"learning_rate": 0.0004995630915406374,
"loss": 6.6492,
"mean_token_accuracy": 0.12218020334839821,
"num_tokens": 6157671.0,
"step": 2835
},
{
"entropy": 6.896828746795654,
"epoch": 0.3039220932099096,
"grad_norm": 0.88671875,
"learning_rate": 0.0004995607067990521,
"loss": 6.7933,
"mean_token_accuracy": 0.11226967573165894,
"num_tokens": 6168651.0,
"step": 2840
},
{
"entropy": 6.8445981502532955,
"epoch": 0.30445716731767347,
"grad_norm": 1.2109375,
"learning_rate": 0.000499558315573308,
"loss": 6.7653,
"mean_token_accuracy": 0.11298804879188537,
"num_tokens": 6179925.0,
"step": 2845
},
{
"entropy": 6.870256233215332,
"epoch": 0.3049922414254374,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995559178634741,
"loss": 6.7605,
"mean_token_accuracy": 0.11938253119587898,
"num_tokens": 6190541.0,
"step": 2850
},
{
"entropy": 7.023727226257324,
"epoch": 0.30552731553320134,
"grad_norm": 0.9140625,
"learning_rate": 0.0004995535136696196,
"loss": 6.7961,
"mean_token_accuracy": 0.11052935272455215,
"num_tokens": 6201491.0,
"step": 2855
},
{
"entropy": 6.799321126937866,
"epoch": 0.30606238964096527,
"grad_norm": 0.8125,
"learning_rate": 0.000499551102991814,
"loss": 6.616,
"mean_token_accuracy": 0.12260399758815765,
"num_tokens": 6212443.0,
"step": 2860
},
{
"entropy": 6.903628635406494,
"epoch": 0.3065974637487292,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995486858301268,
"loss": 6.8178,
"mean_token_accuracy": 0.108098603785038,
"num_tokens": 6222623.0,
"step": 2865
},
{
"entropy": 6.90746922492981,
"epoch": 0.30713253785649314,
"grad_norm": 0.828125,
"learning_rate": 0.0004995462621846279,
"loss": 6.7064,
"mean_token_accuracy": 0.11651685237884521,
"num_tokens": 6233712.0,
"step": 2870
},
{
"entropy": 6.9290708065032955,
"epoch": 0.30766761196425707,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995438320553871,
"loss": 6.7715,
"mean_token_accuracy": 0.11282215490937234,
"num_tokens": 6244254.0,
"step": 2875
},
{
"entropy": 6.881653308868408,
"epoch": 0.30820268607202095,
"grad_norm": 0.890625,
"learning_rate": 0.0004995413954424748,
"loss": 6.6712,
"mean_token_accuracy": 0.12289544939994812,
"num_tokens": 6254202.0,
"step": 2880
},
{
"entropy": 6.872118377685547,
"epoch": 0.3087377601797849,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995389523459612,
"loss": 6.7368,
"mean_token_accuracy": 0.11404762491583824,
"num_tokens": 6264496.0,
"step": 2885
},
{
"entropy": 6.84141526222229,
"epoch": 0.3092728342875488,
"grad_norm": 0.90625,
"learning_rate": 0.000499536502765917,
"loss": 6.7556,
"mean_token_accuracy": 0.12146841287612915,
"num_tokens": 6275835.0,
"step": 2890
},
{
"entropy": 6.885162115097046,
"epoch": 0.30980790839531275,
"grad_norm": 1.0625,
"learning_rate": 0.0004995340467024128,
"loss": 6.6989,
"mean_token_accuracy": 0.11953729987144471,
"num_tokens": 6286805.0,
"step": 2895
},
{
"entropy": 6.878981494903565,
"epoch": 0.3103429825030767,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995315841555195,
"loss": 6.7113,
"mean_token_accuracy": 0.11021909192204475,
"num_tokens": 6298444.0,
"step": 2900
},
{
"entropy": 6.813249588012695,
"epoch": 0.3108780566108406,
"grad_norm": 0.96875,
"learning_rate": 0.0004995291151253083,
"loss": 6.6436,
"mean_token_accuracy": 0.12175455540418625,
"num_tokens": 6308259.0,
"step": 2905
},
{
"entropy": 6.857611513137817,
"epoch": 0.31141313071860455,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995266396118505,
"loss": 6.72,
"mean_token_accuracy": 0.11006433069705963,
"num_tokens": 6320058.0,
"step": 2910
},
{
"entropy": 6.8574628829956055,
"epoch": 0.3119482048263684,
"grad_norm": 0.83203125,
"learning_rate": 0.0004995241576152174,
"loss": 6.6938,
"mean_token_accuracy": 0.11197758540511131,
"num_tokens": 6331203.0,
"step": 2915
},
{
"entropy": 6.926673221588135,
"epoch": 0.31248327893413236,
"grad_norm": 0.9375,
"learning_rate": 0.000499521669135481,
"loss": 6.8007,
"mean_token_accuracy": 0.11380135864019394,
"num_tokens": 6342027.0,
"step": 2920
},
{
"entropy": 6.901452732086182,
"epoch": 0.3130183530418963,
"grad_norm": 0.890625,
"learning_rate": 0.0004995191741727128,
"loss": 6.6965,
"mean_token_accuracy": 0.11690429598093033,
"num_tokens": 6352580.0,
"step": 2925
},
{
"entropy": 6.897034406661987,
"epoch": 0.3135534271496602,
"grad_norm": 0.80078125,
"learning_rate": 0.000499516672726985,
"loss": 6.699,
"mean_token_accuracy": 0.12443051934242248,
"num_tokens": 6363997.0,
"step": 2930
},
{
"entropy": 6.855990314483643,
"epoch": 0.31408850125742416,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995141647983699,
"loss": 6.7583,
"mean_token_accuracy": 0.11855747178196907,
"num_tokens": 6375284.0,
"step": 2935
},
{
"entropy": 6.844248676300049,
"epoch": 0.3146235753651881,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995116503869398,
"loss": 6.6672,
"mean_token_accuracy": 0.12428331449627876,
"num_tokens": 6386698.0,
"step": 2940
},
{
"entropy": 6.867465877532959,
"epoch": 0.315158649472952,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995091294927673,
"loss": 6.7408,
"mean_token_accuracy": 0.11660289838910103,
"num_tokens": 6397237.0,
"step": 2945
},
{
"entropy": 6.98031268119812,
"epoch": 0.3156937235807159,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995066021159253,
"loss": 6.8387,
"mean_token_accuracy": 0.11507428362965584,
"num_tokens": 6408266.0,
"step": 2950
},
{
"entropy": 6.814388084411621,
"epoch": 0.31622879768847983,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995040682564867,
"loss": 6.7375,
"mean_token_accuracy": 0.12039956524968147,
"num_tokens": 6419563.0,
"step": 2955
},
{
"entropy": 6.865634298324585,
"epoch": 0.31676387179624377,
"grad_norm": 0.90234375,
"learning_rate": 0.0004995015279145247,
"loss": 6.6945,
"mean_token_accuracy": 0.12165094465017319,
"num_tokens": 6430220.0,
"step": 2960
},
{
"entropy": 6.879283094406128,
"epoch": 0.3172989459040077,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994989810901126,
"loss": 6.7139,
"mean_token_accuracy": 0.1167122632265091,
"num_tokens": 6440157.0,
"step": 2965
},
{
"entropy": 6.871557569503784,
"epoch": 0.31783402001177163,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994964277833239,
"loss": 6.7633,
"mean_token_accuracy": 0.1225232295691967,
"num_tokens": 6449511.0,
"step": 2970
},
{
"entropy": 6.925748014450074,
"epoch": 0.31836909411953557,
"grad_norm": 0.8359375,
"learning_rate": 0.0004994938679942324,
"loss": 6.7496,
"mean_token_accuracy": 0.11134742796421052,
"num_tokens": 6460608.0,
"step": 2975
},
{
"entropy": 6.8563477993011475,
"epoch": 0.3189041682272995,
"grad_norm": 0.875,
"learning_rate": 0.0004994913017229121,
"loss": 6.7304,
"mean_token_accuracy": 0.12039977535605431,
"num_tokens": 6472737.0,
"step": 2980
},
{
"entropy": 6.803793001174927,
"epoch": 0.31943924233506343,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994887289694371,
"loss": 6.6205,
"mean_token_accuracy": 0.12295867651700973,
"num_tokens": 6482993.0,
"step": 2985
},
{
"entropy": 6.9399620532989506,
"epoch": 0.3199743164428273,
"grad_norm": 0.9296875,
"learning_rate": 0.0004994861497338814,
"loss": 6.8293,
"mean_token_accuracy": 0.11786540225148201,
"num_tokens": 6494530.0,
"step": 2990
},
{
"entropy": 6.926778125762939,
"epoch": 0.32050939055059124,
"grad_norm": 0.921875,
"learning_rate": 0.0004994835640163198,
"loss": 6.6588,
"mean_token_accuracy": 0.11456671357154846,
"num_tokens": 6504691.0,
"step": 2995
},
{
"entropy": 6.821675395965576,
"epoch": 0.3210444646583552,
"grad_norm": 0.875,
"learning_rate": 0.0004994809718168267,
"loss": 6.684,
"mean_token_accuracy": 0.11950030028820038,
"num_tokens": 6514924.0,
"step": 3000
},
{
"epoch": 0.3210444646583552,
"eval_entropy": 6.548516875521513,
"eval_loss": 6.794521331787109,
"eval_mean_token_accuracy": 0.12106629688332907,
"eval_num_tokens": 6514924.0,
"eval_runtime": 17.6213,
"eval_samples_per_second": 1684.272,
"eval_steps_per_second": 210.541,
"step": 3000
},
{
"entropy": 6.799951505661011,
"epoch": 0.3215795387661191,
"grad_norm": 0.875,
"learning_rate": 0.0004994783731354771,
"loss": 6.663,
"mean_token_accuracy": 0.121905267983675,
"num_tokens": 6526097.0,
"step": 3005
},
{
"entropy": 6.887381935119629,
"epoch": 0.32211461287388304,
"grad_norm": 0.9296875,
"learning_rate": 0.0004994757679723461,
"loss": 6.7307,
"mean_token_accuracy": 0.11466163247823716,
"num_tokens": 6536384.0,
"step": 3010
},
{
"entropy": 6.9197609424591064,
"epoch": 0.322649686981647,
"grad_norm": 0.8359375,
"learning_rate": 0.0004994731563275088,
"loss": 6.7868,
"mean_token_accuracy": 0.11109080761671067,
"num_tokens": 6547161.0,
"step": 3015
},
{
"entropy": 6.826877498626709,
"epoch": 0.3231847610894109,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994705382010406,
"loss": 6.6947,
"mean_token_accuracy": 0.11839734464883804,
"num_tokens": 6556751.0,
"step": 3020
},
{
"entropy": 6.831856346130371,
"epoch": 0.3237198351971748,
"grad_norm": 0.87890625,
"learning_rate": 0.0004994679135930174,
"loss": 6.7379,
"mean_token_accuracy": 0.12035313099622727,
"num_tokens": 6567655.0,
"step": 3025
},
{
"entropy": 6.9234333515167235,
"epoch": 0.3242549093049387,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994652825035144,
"loss": 6.7726,
"mean_token_accuracy": 0.11987186148762703,
"num_tokens": 6579127.0,
"step": 3030
},
{
"entropy": 6.984951210021973,
"epoch": 0.32478998341270265,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994626449326081,
"loss": 6.7544,
"mean_token_accuracy": 0.11932827979326248,
"num_tokens": 6590779.0,
"step": 3035
},
{
"entropy": 6.786751747131348,
"epoch": 0.3253250575204666,
"grad_norm": 0.91015625,
"learning_rate": 0.0004994600008803745,
"loss": 6.6644,
"mean_token_accuracy": 0.12521166279911994,
"num_tokens": 6600849.0,
"step": 3040
},
{
"entropy": 6.839286756515503,
"epoch": 0.3258601316282305,
"grad_norm": 0.890625,
"learning_rate": 0.0004994573503468899,
"loss": 6.6739,
"mean_token_accuracy": 0.12311565428972245,
"num_tokens": 6610633.0,
"step": 3045
},
{
"entropy": 6.862971591949463,
"epoch": 0.32639520573599445,
"grad_norm": 0.82421875,
"learning_rate": 0.0004994546933322308,
"loss": 6.7091,
"mean_token_accuracy": 0.11418486312031746,
"num_tokens": 6621327.0,
"step": 3050
},
{
"entropy": 6.812316656112671,
"epoch": 0.3269302798437584,
"grad_norm": 0.8984375,
"learning_rate": 0.000499452029836474,
"loss": 6.6481,
"mean_token_accuracy": 0.12416826337575912,
"num_tokens": 6632267.0,
"step": 3055
},
{
"entropy": 6.841871833801269,
"epoch": 0.32746535395152226,
"grad_norm": 0.88671875,
"learning_rate": 0.0004994493598596963,
"loss": 6.6816,
"mean_token_accuracy": 0.11459707394242287,
"num_tokens": 6642714.0,
"step": 3060
},
{
"entropy": 6.869484233856201,
"epoch": 0.3280004280592862,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994466834019749,
"loss": 6.685,
"mean_token_accuracy": 0.12255851849913597,
"num_tokens": 6652353.0,
"step": 3065
},
{
"entropy": 6.835297298431397,
"epoch": 0.32853550216705013,
"grad_norm": 0.90234375,
"learning_rate": 0.0004994440004633871,
"loss": 6.6661,
"mean_token_accuracy": 0.11809565648436546,
"num_tokens": 6663133.0,
"step": 3070
},
{
"entropy": 6.749080991744995,
"epoch": 0.32907057627481406,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994413110440103,
"loss": 6.6284,
"mean_token_accuracy": 0.1260196879506111,
"num_tokens": 6673984.0,
"step": 3075
},
{
"entropy": 6.871647500991822,
"epoch": 0.329605650382578,
"grad_norm": 0.8046875,
"learning_rate": 0.0004994386151439222,
"loss": 6.714,
"mean_token_accuracy": 0.11568559035658836,
"num_tokens": 6686171.0,
"step": 3080
},
{
"entropy": 6.957374811172485,
"epoch": 0.33014072449034193,
"grad_norm": 0.859375,
"learning_rate": 0.0004994359127632006,
"loss": 6.7591,
"mean_token_accuracy": 0.1136956661939621,
"num_tokens": 6696609.0,
"step": 3085
},
{
"entropy": 6.855549001693726,
"epoch": 0.33067579859810586,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994332039019235,
"loss": 6.6983,
"mean_token_accuracy": 0.11846785247325897,
"num_tokens": 6706188.0,
"step": 3090
},
{
"entropy": 6.778521108627319,
"epoch": 0.33121087270586974,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994304885601694,
"loss": 6.6773,
"mean_token_accuracy": 0.11152351349592209,
"num_tokens": 6716276.0,
"step": 3095
},
{
"entropy": 6.760422897338867,
"epoch": 0.3317459468136337,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994277667380162,
"loss": 6.6388,
"mean_token_accuracy": 0.12479276284575462,
"num_tokens": 6727176.0,
"step": 3100
},
{
"entropy": 6.854667711257934,
"epoch": 0.3322810209213976,
"grad_norm": 0.859375,
"learning_rate": 0.0004994250384355429,
"loss": 6.727,
"mean_token_accuracy": 0.11873743012547493,
"num_tokens": 6737768.0,
"step": 3105
},
{
"entropy": 6.849891614913941,
"epoch": 0.33281609502916154,
"grad_norm": 0.91015625,
"learning_rate": 0.0004994223036528282,
"loss": 6.6613,
"mean_token_accuracy": 0.11627993434667587,
"num_tokens": 6747962.0,
"step": 3110
},
{
"entropy": 6.821473026275635,
"epoch": 0.3333511691369255,
"grad_norm": 0.9375,
"learning_rate": 0.000499419562389951,
"loss": 6.6551,
"mean_token_accuracy": 0.12350267693400382,
"num_tokens": 6758400.0,
"step": 3115
},
{
"entropy": 6.813153457641602,
"epoch": 0.3338862432446894,
"grad_norm": 0.890625,
"learning_rate": 0.0004994168146469904,
"loss": 6.6619,
"mean_token_accuracy": 0.12469682693481446,
"num_tokens": 6768855.0,
"step": 3120
},
{
"entropy": 6.861107110977173,
"epoch": 0.33442131735245334,
"grad_norm": 0.890625,
"learning_rate": 0.0004994140604240259,
"loss": 6.6474,
"mean_token_accuracy": 0.11827528327703477,
"num_tokens": 6778656.0,
"step": 3125
},
{
"entropy": 6.844491481781006,
"epoch": 0.3349563914602172,
"grad_norm": 0.8515625,
"learning_rate": 0.0004994112997211369,
"loss": 6.7274,
"mean_token_accuracy": 0.12191558927297592,
"num_tokens": 6789960.0,
"step": 3130
},
{
"entropy": 6.770041465759277,
"epoch": 0.33549146556798115,
"grad_norm": 0.984375,
"learning_rate": 0.0004994085325384031,
"loss": 6.7037,
"mean_token_accuracy": 0.1188412569463253,
"num_tokens": 6800188.0,
"step": 3135
},
{
"entropy": 6.833252429962158,
"epoch": 0.3360265396757451,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994057588759045,
"loss": 6.6543,
"mean_token_accuracy": 0.12379577830433845,
"num_tokens": 6811032.0,
"step": 3140
},
{
"entropy": 6.751126480102539,
"epoch": 0.336561613783509,
"grad_norm": 0.8984375,
"learning_rate": 0.0004994029787337212,
"loss": 6.6709,
"mean_token_accuracy": 0.12116707265377044,
"num_tokens": 6823060.0,
"step": 3145
},
{
"entropy": 6.856843137741089,
"epoch": 0.33709668789127295,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994001921119333,
"loss": 6.7397,
"mean_token_accuracy": 0.12123994454741478,
"num_tokens": 6833559.0,
"step": 3150
},
{
"entropy": 6.888458251953125,
"epoch": 0.3376317619990369,
"grad_norm": 0.93359375,
"learning_rate": 0.0004993973990106215,
"loss": 6.7015,
"mean_token_accuracy": 0.12211815863847733,
"num_tokens": 6843657.0,
"step": 3155
},
{
"entropy": 6.793592262268066,
"epoch": 0.3381668361068008,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993945994298662,
"loss": 6.6387,
"mean_token_accuracy": 0.11275321766734123,
"num_tokens": 6853615.0,
"step": 3160
},
{
"entropy": 6.842955875396728,
"epoch": 0.3387019102145647,
"grad_norm": 0.875,
"learning_rate": 0.0004993917933697484,
"loss": 6.7319,
"mean_token_accuracy": 0.11903002113103867,
"num_tokens": 6865256.0,
"step": 3165
},
{
"entropy": 7.049095106124878,
"epoch": 0.3392369843223286,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993889808303493,
"loss": 6.7589,
"mean_token_accuracy": 0.1182759940624237,
"num_tokens": 6875358.0,
"step": 3170
},
{
"entropy": 6.767399740219116,
"epoch": 0.33977205843009256,
"grad_norm": 0.859375,
"learning_rate": 0.0004993861618117498,
"loss": 6.6782,
"mean_token_accuracy": 0.1162776917219162,
"num_tokens": 6886584.0,
"step": 3175
},
{
"entropy": 6.765003776550293,
"epoch": 0.3403071325378565,
"grad_norm": 0.90625,
"learning_rate": 0.0004993833363140314,
"loss": 6.6082,
"mean_token_accuracy": 0.12602533102035524,
"num_tokens": 6896958.0,
"step": 3180
},
{
"entropy": 6.886105108261108,
"epoch": 0.3408422066456204,
"grad_norm": 0.9375,
"learning_rate": 0.0004993805043372756,
"loss": 6.6784,
"mean_token_accuracy": 0.12520490884780883,
"num_tokens": 6907944.0,
"step": 3185
},
{
"entropy": 6.877585315704346,
"epoch": 0.34137728075338436,
"grad_norm": 0.8359375,
"learning_rate": 0.0004993776658815644,
"loss": 6.7681,
"mean_token_accuracy": 0.11567452028393746,
"num_tokens": 6919486.0,
"step": 3190
},
{
"entropy": 6.879427862167359,
"epoch": 0.3419123548611483,
"grad_norm": 0.92578125,
"learning_rate": 0.0004993748209469796,
"loss": 6.7403,
"mean_token_accuracy": 0.12306578159332275,
"num_tokens": 6930650.0,
"step": 3195
},
{
"entropy": 6.7968353748321535,
"epoch": 0.34244742896891217,
"grad_norm": 0.86328125,
"learning_rate": 0.0004993719695336033,
"loss": 6.6739,
"mean_token_accuracy": 0.11598326042294502,
"num_tokens": 6941519.0,
"step": 3200
},
{
"entropy": 6.843740224838257,
"epoch": 0.3429825030766761,
"grad_norm": 1.03125,
"learning_rate": 0.0004993691116415179,
"loss": 6.7212,
"mean_token_accuracy": 0.11809414103627205,
"num_tokens": 6952952.0,
"step": 3205
},
{
"entropy": 6.8387915134429935,
"epoch": 0.34351757718444004,
"grad_norm": 0.859375,
"learning_rate": 0.000499366247270806,
"loss": 6.6856,
"mean_token_accuracy": 0.12287690415978432,
"num_tokens": 6962867.0,
"step": 3210
},
{
"entropy": 6.79352445602417,
"epoch": 0.34405265129220397,
"grad_norm": 0.87109375,
"learning_rate": 0.0004993633764215502,
"loss": 6.6226,
"mean_token_accuracy": 0.12643782943487167,
"num_tokens": 6973434.0,
"step": 3215
},
{
"entropy": 6.819111680984497,
"epoch": 0.3445877253999679,
"grad_norm": 0.9375,
"learning_rate": 0.0004993604990938336,
"loss": 6.604,
"mean_token_accuracy": 0.13255805000662804,
"num_tokens": 6984161.0,
"step": 3220
},
{
"entropy": 6.868915271759033,
"epoch": 0.34512279950773184,
"grad_norm": 0.90625,
"learning_rate": 0.0004993576152877389,
"loss": 6.6275,
"mean_token_accuracy": 0.1198020301759243,
"num_tokens": 6994700.0,
"step": 3225
},
{
"entropy": 6.828886413574219,
"epoch": 0.34565787361549577,
"grad_norm": 0.89453125,
"learning_rate": 0.0004993547250033496,
"loss": 6.7612,
"mean_token_accuracy": 0.11038903146982193,
"num_tokens": 7006497.0,
"step": 3230
},
{
"entropy": 6.8887570858001705,
"epoch": 0.34619294772325965,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993518282407493,
"loss": 6.6798,
"mean_token_accuracy": 0.1266617774963379,
"num_tokens": 7016415.0,
"step": 3235
},
{
"entropy": 6.814205265045166,
"epoch": 0.3467280218310236,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993489250000213,
"loss": 6.6303,
"mean_token_accuracy": 0.11938185915350914,
"num_tokens": 7026610.0,
"step": 3240
},
{
"entropy": 6.814730072021485,
"epoch": 0.3472630959387875,
"grad_norm": 0.86328125,
"learning_rate": 0.0004993460152812497,
"loss": 6.639,
"mean_token_accuracy": 0.11687793508172035,
"num_tokens": 7038236.0,
"step": 3245
},
{
"entropy": 6.799622106552124,
"epoch": 0.34779817004655145,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993430990845184,
"loss": 6.6873,
"mean_token_accuracy": 0.1299070708453655,
"num_tokens": 7049438.0,
"step": 3250
},
{
"entropy": 6.822757863998413,
"epoch": 0.3483332441543154,
"grad_norm": 0.8515625,
"learning_rate": 0.0004993401764099118,
"loss": 6.6614,
"mean_token_accuracy": 0.11520971059799194,
"num_tokens": 7061001.0,
"step": 3255
},
{
"entropy": 6.826554918289185,
"epoch": 0.3488683182620793,
"grad_norm": 0.87109375,
"learning_rate": 0.0004993372472575139,
"loss": 6.6494,
"mean_token_accuracy": 0.11807417050004006,
"num_tokens": 7071328.0,
"step": 3260
},
{
"entropy": 6.862842416763305,
"epoch": 0.34940339236984325,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993343116274096,
"loss": 6.8306,
"mean_token_accuracy": 0.11384405344724655,
"num_tokens": 7082597.0,
"step": 3265
},
{
"entropy": 6.815863227844238,
"epoch": 0.3499384664776071,
"grad_norm": 0.86328125,
"learning_rate": 0.0004993313695196836,
"loss": 6.6586,
"mean_token_accuracy": 0.11965244635939598,
"num_tokens": 7093463.0,
"step": 3270
},
{
"entropy": 6.786405324935913,
"epoch": 0.35047354058537106,
"grad_norm": 0.83984375,
"learning_rate": 0.0004993284209344208,
"loss": 6.7219,
"mean_token_accuracy": 0.12500518634915353,
"num_tokens": 7104350.0,
"step": 3275
},
{
"entropy": 6.882570743560791,
"epoch": 0.351008614693135,
"grad_norm": 0.984375,
"learning_rate": 0.0004993254658717065,
"loss": 6.661,
"mean_token_accuracy": 0.1239277258515358,
"num_tokens": 7114851.0,
"step": 3280
},
{
"entropy": 6.823478555679321,
"epoch": 0.3515436888008989,
"grad_norm": 0.90234375,
"learning_rate": 0.0004993225043316257,
"loss": 6.7135,
"mean_token_accuracy": 0.11792605891823768,
"num_tokens": 7126440.0,
"step": 3285
},
{
"entropy": 6.880914402008057,
"epoch": 0.35207876290866286,
"grad_norm": 0.9375,
"learning_rate": 0.0004993195363142643,
"loss": 6.7407,
"mean_token_accuracy": 0.11478666737675666,
"num_tokens": 7137884.0,
"step": 3290
},
{
"entropy": 6.892389440536499,
"epoch": 0.3526138370164268,
"grad_norm": 0.9453125,
"learning_rate": 0.0004993165618197076,
"loss": 6.7197,
"mean_token_accuracy": 0.1155787654221058,
"num_tokens": 7147930.0,
"step": 3295
},
{
"entropy": 6.791791582107544,
"epoch": 0.3531489111241907,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993135808480418,
"loss": 6.5753,
"mean_token_accuracy": 0.129153461009264,
"num_tokens": 7159200.0,
"step": 3300
},
{
"entropy": 6.859189891815186,
"epoch": 0.3536839852319546,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993105933993529,
"loss": 6.6683,
"mean_token_accuracy": 0.12259154245257378,
"num_tokens": 7169601.0,
"step": 3305
},
{
"entropy": 6.754944276809693,
"epoch": 0.35421905933971853,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993075994737271,
"loss": 6.6144,
"mean_token_accuracy": 0.12793416306376457,
"num_tokens": 7180465.0,
"step": 3310
},
{
"entropy": 6.840865564346314,
"epoch": 0.35475413344748247,
"grad_norm": 0.94140625,
"learning_rate": 0.0004993045990712508,
"loss": 6.6155,
"mean_token_accuracy": 0.12482116669416428,
"num_tokens": 7190603.0,
"step": 3315
},
{
"entropy": 6.774190092086792,
"epoch": 0.3552892075552464,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993015921920109,
"loss": 6.5848,
"mean_token_accuracy": 0.1270811952650547,
"num_tokens": 7201017.0,
"step": 3320
},
{
"entropy": 6.705060338973999,
"epoch": 0.35582428166301033,
"grad_norm": 0.9609375,
"learning_rate": 0.000499298578836094,
"loss": 6.5837,
"mean_token_accuracy": 0.12208671867847443,
"num_tokens": 7210908.0,
"step": 3325
},
{
"entropy": 6.868721342086792,
"epoch": 0.35635935577077427,
"grad_norm": 0.9921875,
"learning_rate": 0.000499295559003587,
"loss": 6.6211,
"mean_token_accuracy": 0.12420659512281418,
"num_tokens": 7221306.0,
"step": 3330
},
{
"entropy": 6.7805100440979,
"epoch": 0.3568944298785382,
"grad_norm": 0.890625,
"learning_rate": 0.0004992925326945774,
"loss": 6.7391,
"mean_token_accuracy": 0.11606214642524719,
"num_tokens": 7233076.0,
"step": 3335
},
{
"entropy": 6.79472131729126,
"epoch": 0.3574295039863021,
"grad_norm": 0.93359375,
"learning_rate": 0.0004992894999091524,
"loss": 6.7376,
"mean_token_accuracy": 0.11862620785832405,
"num_tokens": 7244703.0,
"step": 3340
},
{
"entropy": 6.858727550506591,
"epoch": 0.357964578094066,
"grad_norm": 0.87109375,
"learning_rate": 0.0004992864606473996,
"loss": 6.7249,
"mean_token_accuracy": 0.11088455319404603,
"num_tokens": 7255438.0,
"step": 3345
},
{
"entropy": 6.8269633769989015,
"epoch": 0.35849965220182994,
"grad_norm": 0.93359375,
"learning_rate": 0.0004992834149094069,
"loss": 6.6592,
"mean_token_accuracy": 0.12365302219986915,
"num_tokens": 7265720.0,
"step": 3350
},
{
"entropy": 6.849551486968994,
"epoch": 0.3590347263095939,
"grad_norm": 0.84375,
"learning_rate": 0.0004992803626952619,
"loss": 6.7043,
"mean_token_accuracy": 0.11888771429657936,
"num_tokens": 7277791.0,
"step": 3355
},
{
"entropy": 6.7473889827728275,
"epoch": 0.3595698004173578,
"grad_norm": 0.9296875,
"learning_rate": 0.000499277304005053,
"loss": 6.5728,
"mean_token_accuracy": 0.1280308946967125,
"num_tokens": 7288521.0,
"step": 3360
},
{
"entropy": 6.810279130935669,
"epoch": 0.36010487452512174,
"grad_norm": 0.87109375,
"learning_rate": 0.0004992742388388686,
"loss": 6.7666,
"mean_token_accuracy": 0.11023237109184265,
"num_tokens": 7299298.0,
"step": 3365
},
{
"entropy": 6.792523527145386,
"epoch": 0.3606399486328857,
"grad_norm": 0.90625,
"learning_rate": 0.0004992711671967968,
"loss": 6.595,
"mean_token_accuracy": 0.11808668822050095,
"num_tokens": 7310271.0,
"step": 3370
},
{
"entropy": 6.823890495300293,
"epoch": 0.36117502274064955,
"grad_norm": 0.90625,
"learning_rate": 0.0004992680890789268,
"loss": 6.6509,
"mean_token_accuracy": 0.12248839884996414,
"num_tokens": 7321112.0,
"step": 3375
},
{
"entropy": 6.706878280639648,
"epoch": 0.3617100968484135,
"grad_norm": 0.890625,
"learning_rate": 0.0004992650044853471,
"loss": 6.6047,
"mean_token_accuracy": 0.12075349166989327,
"num_tokens": 7331976.0,
"step": 3380
},
{
"entropy": 6.8689617156982425,
"epoch": 0.3622451709561774,
"grad_norm": 0.953125,
"learning_rate": 0.000499261913416147,
"loss": 6.7263,
"mean_token_accuracy": 0.11769687235355378,
"num_tokens": 7342800.0,
"step": 3385
},
{
"entropy": 6.7807615280151365,
"epoch": 0.36278024506394135,
"grad_norm": 0.83203125,
"learning_rate": 0.0004992588158714157,
"loss": 6.5487,
"mean_token_accuracy": 0.12072362452745437,
"num_tokens": 7354213.0,
"step": 3390
},
{
"entropy": 6.761278200149536,
"epoch": 0.3633153191717053,
"grad_norm": 0.92578125,
"learning_rate": 0.0004992557118512424,
"loss": 6.6253,
"mean_token_accuracy": 0.12343727201223373,
"num_tokens": 7363250.0,
"step": 3395
},
{
"entropy": 6.743581962585449,
"epoch": 0.3638503932794692,
"grad_norm": 0.9375,
"learning_rate": 0.0004992526013557171,
"loss": 6.6557,
"mean_token_accuracy": 0.1214616447687149,
"num_tokens": 7374185.0,
"step": 3400
},
{
"entropy": 6.7366564750671385,
"epoch": 0.36438546738723315,
"grad_norm": 0.85546875,
"learning_rate": 0.0004992494843849294,
"loss": 6.6305,
"mean_token_accuracy": 0.11691405922174454,
"num_tokens": 7384579.0,
"step": 3405
},
{
"entropy": 6.881455659866333,
"epoch": 0.36492054149499703,
"grad_norm": 0.921875,
"learning_rate": 0.0004992463609389694,
"loss": 6.6406,
"mean_token_accuracy": 0.11755282357335091,
"num_tokens": 7395277.0,
"step": 3410
},
{
"entropy": 6.706863021850586,
"epoch": 0.36545561560276096,
"grad_norm": 0.85546875,
"learning_rate": 0.0004992432310179272,
"loss": 6.6343,
"mean_token_accuracy": 0.11993120685219764,
"num_tokens": 7406688.0,
"step": 3415
},
{
"entropy": 6.736936712265015,
"epoch": 0.3659906897105249,
"grad_norm": 0.84375,
"learning_rate": 0.0004992400946218932,
"loss": 6.6842,
"mean_token_accuracy": 0.12183412909507751,
"num_tokens": 7417378.0,
"step": 3420
},
{
"entropy": 6.972946405410767,
"epoch": 0.36652576381828883,
"grad_norm": 0.83984375,
"learning_rate": 0.0004992369517509581,
"loss": 6.7056,
"mean_token_accuracy": 0.11991995200514793,
"num_tokens": 7427675.0,
"step": 3425
},
{
"entropy": 6.802153635025024,
"epoch": 0.36706083792605276,
"grad_norm": 0.91796875,
"learning_rate": 0.0004992338024052123,
"loss": 6.7424,
"mean_token_accuracy": 0.11765128523111343,
"num_tokens": 7438083.0,
"step": 3430
},
{
"entropy": 6.767482089996338,
"epoch": 0.3675959120338167,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992306465847473,
"loss": 6.5952,
"mean_token_accuracy": 0.12695135846734046,
"num_tokens": 7449051.0,
"step": 3435
},
{
"entropy": 6.7729590892791744,
"epoch": 0.36813098614158063,
"grad_norm": 0.859375,
"learning_rate": 0.0004992274842896536,
"loss": 6.6246,
"mean_token_accuracy": 0.12268055081367493,
"num_tokens": 7459074.0,
"step": 3440
},
{
"entropy": 6.784332609176635,
"epoch": 0.3686660602493445,
"grad_norm": 0.84765625,
"learning_rate": 0.0004992243155200229,
"loss": 6.6902,
"mean_token_accuracy": 0.12212735787034035,
"num_tokens": 7470472.0,
"step": 3445
},
{
"entropy": 6.78359055519104,
"epoch": 0.36920113435710844,
"grad_norm": 0.8828125,
"learning_rate": 0.0004992211402759466,
"loss": 6.6568,
"mean_token_accuracy": 0.12637588679790496,
"num_tokens": 7481477.0,
"step": 3450
},
{
"entropy": 6.786507797241211,
"epoch": 0.3697362084648724,
"grad_norm": 0.88671875,
"learning_rate": 0.0004992179585575164,
"loss": 6.6085,
"mean_token_accuracy": 0.12437106668949127,
"num_tokens": 7491906.0,
"step": 3455
},
{
"entropy": 6.794345140457153,
"epoch": 0.3702712825726363,
"grad_norm": 0.9609375,
"learning_rate": 0.000499214770364824,
"loss": 6.5373,
"mean_token_accuracy": 0.11893177703022957,
"num_tokens": 7502447.0,
"step": 3460
},
{
"entropy": 6.721900892257691,
"epoch": 0.37080635668040024,
"grad_norm": 0.8515625,
"learning_rate": 0.0004992115756979617,
"loss": 6.553,
"mean_token_accuracy": 0.129140467941761,
"num_tokens": 7513505.0,
"step": 3465
},
{
"entropy": 6.754326629638672,
"epoch": 0.3713414307881642,
"grad_norm": 0.83984375,
"learning_rate": 0.0004992083745570218,
"loss": 6.5675,
"mean_token_accuracy": 0.12708702757954599,
"num_tokens": 7524884.0,
"step": 3470
},
{
"entropy": 6.7483885288238525,
"epoch": 0.3718765048959281,
"grad_norm": 0.91015625,
"learning_rate": 0.0004992051669420964,
"loss": 6.6312,
"mean_token_accuracy": 0.12879489958286286,
"num_tokens": 7535942.0,
"step": 3475
},
{
"entropy": 6.7121062755584715,
"epoch": 0.372411579003692,
"grad_norm": 0.87890625,
"learning_rate": 0.0004992019528532785,
"loss": 6.5951,
"mean_token_accuracy": 0.12708142325282096,
"num_tokens": 7545506.0,
"step": 3480
},
{
"entropy": 6.795954084396362,
"epoch": 0.3729466531114559,
"grad_norm": 0.97265625,
"learning_rate": 0.0004991987322906605,
"loss": 6.6893,
"mean_token_accuracy": 0.11757104098796844,
"num_tokens": 7555328.0,
"step": 3485
},
{
"entropy": 6.849932861328125,
"epoch": 0.37348172721921985,
"grad_norm": 0.94140625,
"learning_rate": 0.0004991955052543357,
"loss": 6.5845,
"mean_token_accuracy": 0.12351070642471314,
"num_tokens": 7565762.0,
"step": 3490
},
{
"entropy": 6.776114749908447,
"epoch": 0.3740168013269838,
"grad_norm": 0.87109375,
"learning_rate": 0.0004991922717443972,
"loss": 6.6028,
"mean_token_accuracy": 0.12989014089107515,
"num_tokens": 7575884.0,
"step": 3495
},
{
"entropy": 6.684484958648682,
"epoch": 0.3745518754347477,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991890317609384,
"loss": 6.6374,
"mean_token_accuracy": 0.12035185471177101,
"num_tokens": 7586100.0,
"step": 3500
},
{
"entropy": 6.829225254058838,
"epoch": 0.37508694954251165,
"grad_norm": 0.96484375,
"learning_rate": 0.0004991857853040528,
"loss": 6.6546,
"mean_token_accuracy": 0.12205979079008103,
"num_tokens": 7597137.0,
"step": 3505
},
{
"entropy": 6.783835411071777,
"epoch": 0.3756220236502756,
"grad_norm": 0.875,
"learning_rate": 0.0004991825323738341,
"loss": 6.676,
"mean_token_accuracy": 0.12901032269001006,
"num_tokens": 7607640.0,
"step": 3510
},
{
"entropy": 6.829906606674195,
"epoch": 0.3761570977580395,
"grad_norm": 0.953125,
"learning_rate": 0.0004991792729703763,
"loss": 6.7337,
"mean_token_accuracy": 0.12128165885806083,
"num_tokens": 7618319.0,
"step": 3515
},
{
"entropy": 6.812218856811524,
"epoch": 0.3766921718658034,
"grad_norm": 0.84375,
"learning_rate": 0.0004991760070937734,
"loss": 6.6028,
"mean_token_accuracy": 0.12587715685367584,
"num_tokens": 7628350.0,
"step": 3520
},
{
"entropy": 6.777668380737305,
"epoch": 0.3772272459735673,
"grad_norm": 0.94140625,
"learning_rate": 0.0004991727347441199,
"loss": 6.6396,
"mean_token_accuracy": 0.11658850014209747,
"num_tokens": 7637857.0,
"step": 3525
},
{
"entropy": 6.83516297340393,
"epoch": 0.37776232008133126,
"grad_norm": 0.91796875,
"learning_rate": 0.0004991694559215102,
"loss": 6.6543,
"mean_token_accuracy": 0.12148249298334121,
"num_tokens": 7648242.0,
"step": 3530
},
{
"entropy": 6.782798194885254,
"epoch": 0.3782973941890952,
"grad_norm": 0.734375,
"learning_rate": 0.000499166170626039,
"loss": 6.6481,
"mean_token_accuracy": 0.12124886438250541,
"num_tokens": 7660565.0,
"step": 3535
},
{
"entropy": 6.772379589080811,
"epoch": 0.3788324682968591,
"grad_norm": 1.015625,
"learning_rate": 0.000499162878857801,
"loss": 6.6642,
"mean_token_accuracy": 0.12547138035297395,
"num_tokens": 7671600.0,
"step": 3540
},
{
"entropy": 6.786243486404419,
"epoch": 0.37936754240462306,
"grad_norm": 0.890625,
"learning_rate": 0.0004991595806168915,
"loss": 6.6761,
"mean_token_accuracy": 0.11763999313116073,
"num_tokens": 7682453.0,
"step": 3545
},
{
"entropy": 6.761963129043579,
"epoch": 0.379902616512387,
"grad_norm": 0.8359375,
"learning_rate": 0.0004991562759034055,
"loss": 6.6056,
"mean_token_accuracy": 0.12182423919439316,
"num_tokens": 7693490.0,
"step": 3550
},
{
"entropy": 6.8190428733825685,
"epoch": 0.38043769062015087,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991529647174386,
"loss": 6.5641,
"mean_token_accuracy": 0.12036292180418968,
"num_tokens": 7703478.0,
"step": 3555
},
{
"entropy": 6.775476408004761,
"epoch": 0.3809727647279148,
"grad_norm": 0.90234375,
"learning_rate": 0.0004991496470590864,
"loss": 6.6654,
"mean_token_accuracy": 0.1186753623187542,
"num_tokens": 7714647.0,
"step": 3560
},
{
"entropy": 6.7765895366668705,
"epoch": 0.38150783883567874,
"grad_norm": 0.890625,
"learning_rate": 0.0004991463229284445,
"loss": 6.6069,
"mean_token_accuracy": 0.1255822166800499,
"num_tokens": 7725528.0,
"step": 3565
},
{
"entropy": 6.741648101806641,
"epoch": 0.38204291294344267,
"grad_norm": 0.859375,
"learning_rate": 0.0004991429923256091,
"loss": 6.6474,
"mean_token_accuracy": 0.12289627566933632,
"num_tokens": 7737708.0,
"step": 3570
},
{
"entropy": 6.807142543792724,
"epoch": 0.3825779870512066,
"grad_norm": 0.83203125,
"learning_rate": 0.0004991396552506762,
"loss": 6.642,
"mean_token_accuracy": 0.11627409681677818,
"num_tokens": 7749093.0,
"step": 3575
},
{
"entropy": 6.73176646232605,
"epoch": 0.38311306115897054,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991363117037423,
"loss": 6.5812,
"mean_token_accuracy": 0.12449619993567466,
"num_tokens": 7760024.0,
"step": 3580
},
{
"entropy": 6.7203834533691404,
"epoch": 0.38364813526673447,
"grad_norm": 0.8359375,
"learning_rate": 0.0004991329616849039,
"loss": 6.619,
"mean_token_accuracy": 0.12767036110162736,
"num_tokens": 7770801.0,
"step": 3585
},
{
"entropy": 6.82001256942749,
"epoch": 0.38418320937449835,
"grad_norm": 0.890625,
"learning_rate": 0.0004991296051942578,
"loss": 6.6721,
"mean_token_accuracy": 0.1217954471707344,
"num_tokens": 7782257.0,
"step": 3590
},
{
"entropy": 6.768124771118164,
"epoch": 0.3847182834822623,
"grad_norm": 0.85546875,
"learning_rate": 0.0004991262422319008,
"loss": 6.5965,
"mean_token_accuracy": 0.12531521171331406,
"num_tokens": 7793016.0,
"step": 3595
},
{
"entropy": 6.751244163513183,
"epoch": 0.3852533575900262,
"grad_norm": 0.91015625,
"learning_rate": 0.00049912287279793,
"loss": 6.5442,
"mean_token_accuracy": 0.13346222937107086,
"num_tokens": 7802207.0,
"step": 3600
},
{
"entropy": 6.784510135650635,
"epoch": 0.38578843169779015,
"grad_norm": 0.96875,
"learning_rate": 0.0004991194968924429,
"loss": 6.656,
"mean_token_accuracy": 0.12131711095571518,
"num_tokens": 7813487.0,
"step": 3605
},
{
"entropy": 6.82340726852417,
"epoch": 0.3863235058055541,
"grad_norm": 0.9375,
"learning_rate": 0.0004991161145155367,
"loss": 6.6398,
"mean_token_accuracy": 0.1207100585103035,
"num_tokens": 7824587.0,
"step": 3610
},
{
"entropy": 6.838507223129272,
"epoch": 0.386858579913318,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991127256673091,
"loss": 6.6259,
"mean_token_accuracy": 0.11949132606387139,
"num_tokens": 7836240.0,
"step": 3615
},
{
"entropy": 6.724960899353027,
"epoch": 0.38739365402108195,
"grad_norm": 0.89453125,
"learning_rate": 0.0004991093303478582,
"loss": 6.5721,
"mean_token_accuracy": 0.12099727615714073,
"num_tokens": 7847014.0,
"step": 3620
},
{
"entropy": 6.755022096633911,
"epoch": 0.3879287281288458,
"grad_norm": 0.8203125,
"learning_rate": 0.0004991059285572818,
"loss": 6.6905,
"mean_token_accuracy": 0.12505315169692038,
"num_tokens": 7857899.0,
"step": 3625
},
{
"entropy": 6.805648851394653,
"epoch": 0.38846380223660976,
"grad_norm": 0.87890625,
"learning_rate": 0.0004991025202956781,
"loss": 6.5859,
"mean_token_accuracy": 0.1255096547305584,
"num_tokens": 7868987.0,
"step": 3630
},
{
"entropy": 6.84774341583252,
"epoch": 0.3889988763443737,
"grad_norm": 1.015625,
"learning_rate": 0.0004990991055631457,
"loss": 6.7064,
"mean_token_accuracy": 0.1195052333176136,
"num_tokens": 7879547.0,
"step": 3635
},
{
"entropy": 6.737785577774048,
"epoch": 0.3895339504521376,
"grad_norm": 0.89453125,
"learning_rate": 0.0004990956843597832,
"loss": 6.5782,
"mean_token_accuracy": 0.12665883377194403,
"num_tokens": 7889824.0,
"step": 3640
},
{
"entropy": 6.662966775894165,
"epoch": 0.39006902455990156,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990922566856892,
"loss": 6.5733,
"mean_token_accuracy": 0.11881136298179626,
"num_tokens": 7901156.0,
"step": 3645
},
{
"entropy": 6.7981789112091064,
"epoch": 0.3906040986676655,
"grad_norm": 1.0,
"learning_rate": 0.0004990888225409629,
"loss": 6.592,
"mean_token_accuracy": 0.12195635959506035,
"num_tokens": 7911695.0,
"step": 3650
},
{
"entropy": 6.686615133285523,
"epoch": 0.3911391727754294,
"grad_norm": 0.8203125,
"learning_rate": 0.0004990853819257033,
"loss": 6.5732,
"mean_token_accuracy": 0.13316732421517372,
"num_tokens": 7923559.0,
"step": 3655
},
{
"entropy": 6.842485618591309,
"epoch": 0.3916742468831933,
"grad_norm": 0.953125,
"learning_rate": 0.0004990819348400097,
"loss": 6.6183,
"mean_token_accuracy": 0.12080760896205903,
"num_tokens": 7935792.0,
"step": 3660
},
{
"entropy": 6.804695320129395,
"epoch": 0.39220932099095723,
"grad_norm": 0.98046875,
"learning_rate": 0.0004990784812839818,
"loss": 6.5951,
"mean_token_accuracy": 0.13260335549712182,
"num_tokens": 7946987.0,
"step": 3665
},
{
"entropy": 6.670995664596558,
"epoch": 0.39274439509872117,
"grad_norm": 0.9375,
"learning_rate": 0.0004990750212577192,
"loss": 6.5283,
"mean_token_accuracy": 0.12538674250245094,
"num_tokens": 7957160.0,
"step": 3670
},
{
"entropy": 6.772046279907227,
"epoch": 0.3932794692064851,
"grad_norm": 0.89453125,
"learning_rate": 0.0004990715547613219,
"loss": 6.6304,
"mean_token_accuracy": 0.11679298281669617,
"num_tokens": 7968833.0,
"step": 3675
},
{
"entropy": 6.8003373622894285,
"epoch": 0.39381454331424903,
"grad_norm": 0.8515625,
"learning_rate": 0.0004990680817948898,
"loss": 6.6755,
"mean_token_accuracy": 0.11913765817880631,
"num_tokens": 7980661.0,
"step": 3680
},
{
"entropy": 6.844438648223877,
"epoch": 0.39434961742201297,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990646023585234,
"loss": 6.5909,
"mean_token_accuracy": 0.12314732074737549,
"num_tokens": 7992677.0,
"step": 3685
},
{
"entropy": 6.688145542144776,
"epoch": 0.3948846915297769,
"grad_norm": 0.8828125,
"learning_rate": 0.0004990611164523231,
"loss": 6.57,
"mean_token_accuracy": 0.1285789616405964,
"num_tokens": 8002796.0,
"step": 3690
},
{
"entropy": 6.688539314270019,
"epoch": 0.3954197656375408,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990576240763896,
"loss": 6.5999,
"mean_token_accuracy": 0.1252342328429222,
"num_tokens": 8012779.0,
"step": 3695
},
{
"entropy": 6.76634202003479,
"epoch": 0.3959548397453047,
"grad_norm": 0.80078125,
"learning_rate": 0.0004990541252308237,
"loss": 6.5824,
"mean_token_accuracy": 0.12815626487135887,
"num_tokens": 8023442.0,
"step": 3700
},
{
"entropy": 6.820317983627319,
"epoch": 0.39648991385306864,
"grad_norm": 0.90625,
"learning_rate": 0.0004990506199157263,
"loss": 6.6726,
"mean_token_accuracy": 0.11957266852259636,
"num_tokens": 8035163.0,
"step": 3705
},
{
"entropy": 6.777938270568848,
"epoch": 0.3970249879608326,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990471081311988,
"loss": 6.5926,
"mean_token_accuracy": 0.12503935247659684,
"num_tokens": 8045330.0,
"step": 3710
},
{
"entropy": 6.772491693496704,
"epoch": 0.3975600620685965,
"grad_norm": 0.8515625,
"learning_rate": 0.0004990435898773426,
"loss": 6.6203,
"mean_token_accuracy": 0.12175011858344079,
"num_tokens": 8056154.0,
"step": 3715
},
{
"entropy": 6.784107732772827,
"epoch": 0.39809513617636044,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990400651542592,
"loss": 6.5829,
"mean_token_accuracy": 0.1277529016137123,
"num_tokens": 8066101.0,
"step": 3720
},
{
"entropy": 6.751408624649048,
"epoch": 0.3986302102841244,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990365339620504,
"loss": 6.6561,
"mean_token_accuracy": 0.117050202190876,
"num_tokens": 8077249.0,
"step": 3725
},
{
"entropy": 6.759656381607056,
"epoch": 0.39916528439188825,
"grad_norm": 0.9609375,
"learning_rate": 0.000499032996300818,
"loss": 6.5993,
"mean_token_accuracy": 0.12702321633696556,
"num_tokens": 8088532.0,
"step": 3730
},
{
"entropy": 6.800127220153809,
"epoch": 0.3997003584996522,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990294521706645,
"loss": 6.6469,
"mean_token_accuracy": 0.11475934162735939,
"num_tokens": 8099541.0,
"step": 3735
},
{
"entropy": 6.765775728225708,
"epoch": 0.4002354326074161,
"grad_norm": 0.8671875,
"learning_rate": 0.000499025901571692,
"loss": 6.6743,
"mean_token_accuracy": 0.12347187474370003,
"num_tokens": 8110532.0,
"step": 3740
},
{
"entropy": 6.804666519165039,
"epoch": 0.40077050671518005,
"grad_norm": 0.890625,
"learning_rate": 0.0004990223445040031,
"loss": 6.5906,
"mean_token_accuracy": 0.12874274030327798,
"num_tokens": 8120991.0,
"step": 3745
},
{
"entropy": 6.71906795501709,
"epoch": 0.401305580822944,
"grad_norm": 0.8671875,
"learning_rate": 0.0004990187809677004,
"loss": 6.5306,
"mean_token_accuracy": 0.12286594063043595,
"num_tokens": 8130737.0,
"step": 3750
},
{
"entropy": 6.6623273372650145,
"epoch": 0.4018406549307079,
"grad_norm": 0.91796875,
"learning_rate": 0.0004990152109628869,
"loss": 6.5363,
"mean_token_accuracy": 0.12400009483098984,
"num_tokens": 8142315.0,
"step": 3755
},
{
"entropy": 6.764507102966308,
"epoch": 0.40237572903847185,
"grad_norm": 1.046875,
"learning_rate": 0.0004990116344896657,
"loss": 6.6172,
"mean_token_accuracy": 0.11884014382958412,
"num_tokens": 8153618.0,
"step": 3760
},
{
"entropy": 6.701496410369873,
"epoch": 0.40291080314623573,
"grad_norm": 0.83984375,
"learning_rate": 0.00049900805154814,
"loss": 6.5736,
"mean_token_accuracy": 0.13147677332162858,
"num_tokens": 8164304.0,
"step": 3765
},
{
"entropy": 6.719698762893676,
"epoch": 0.40344587725399966,
"grad_norm": 0.85546875,
"learning_rate": 0.0004990044621384134,
"loss": 6.5691,
"mean_token_accuracy": 0.1204621210694313,
"num_tokens": 8175289.0,
"step": 3770
},
{
"entropy": 6.799706554412841,
"epoch": 0.4039809513617636,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990008662605893,
"loss": 6.618,
"mean_token_accuracy": 0.1192642904818058,
"num_tokens": 8186274.0,
"step": 3775
},
{
"entropy": 6.766622829437256,
"epoch": 0.40451602546952753,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989972639147717,
"loss": 6.5458,
"mean_token_accuracy": 0.12909928560256959,
"num_tokens": 8196380.0,
"step": 3780
},
{
"entropy": 6.709709596633911,
"epoch": 0.40505109957729146,
"grad_norm": 0.8984375,
"learning_rate": 0.0004989936551010646,
"loss": 6.6212,
"mean_token_accuracy": 0.12540413141250611,
"num_tokens": 8207204.0,
"step": 3785
},
{
"entropy": 6.662211322784424,
"epoch": 0.4055861736850554,
"grad_norm": 0.89453125,
"learning_rate": 0.0004989900398195722,
"loss": 6.5364,
"mean_token_accuracy": 0.12927303686738015,
"num_tokens": 8217745.0,
"step": 3790
},
{
"entropy": 6.708821868896484,
"epoch": 0.40612124779281933,
"grad_norm": 0.9375,
"learning_rate": 0.0004989864180703988,
"loss": 6.5265,
"mean_token_accuracy": 0.12706319689750672,
"num_tokens": 8228171.0,
"step": 3795
},
{
"entropy": 6.718810129165649,
"epoch": 0.4066563219005832,
"grad_norm": 0.8828125,
"learning_rate": 0.0004989827898536491,
"loss": 6.5805,
"mean_token_accuracy": 0.11968392059206963,
"num_tokens": 8239220.0,
"step": 3800
},
{
"entropy": 6.75914945602417,
"epoch": 0.40719139600834714,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989791551694279,
"loss": 6.6392,
"mean_token_accuracy": 0.12101104408502579,
"num_tokens": 8250363.0,
"step": 3805
},
{
"entropy": 6.7193773746490475,
"epoch": 0.4077264701161111,
"grad_norm": 0.91796875,
"learning_rate": 0.0004989755140178398,
"loss": 6.5934,
"mean_token_accuracy": 0.12401422187685966,
"num_tokens": 8261207.0,
"step": 3810
},
{
"entropy": 6.687061738967896,
"epoch": 0.408261544223875,
"grad_norm": 0.90234375,
"learning_rate": 0.0004989718663989905,
"loss": 6.4898,
"mean_token_accuracy": 0.12349989488720894,
"num_tokens": 8272343.0,
"step": 3815
},
{
"entropy": 6.696186542510986,
"epoch": 0.40879661833163894,
"grad_norm": 0.859375,
"learning_rate": 0.0004989682123129849,
"loss": 6.5487,
"mean_token_accuracy": 0.12425093948841096,
"num_tokens": 8282677.0,
"step": 3820
},
{
"entropy": 6.667926263809204,
"epoch": 0.40933169243940287,
"grad_norm": 0.88671875,
"learning_rate": 0.0004989645517599286,
"loss": 6.5614,
"mean_token_accuracy": 0.12363334447145462,
"num_tokens": 8292753.0,
"step": 3825
},
{
"entropy": 6.730340671539307,
"epoch": 0.4098667665471668,
"grad_norm": 0.87890625,
"learning_rate": 0.0004989608847399275,
"loss": 6.5546,
"mean_token_accuracy": 0.12061431556940079,
"num_tokens": 8302747.0,
"step": 3830
},
{
"entropy": 6.735944366455078,
"epoch": 0.4104018406549307,
"grad_norm": 0.90625,
"learning_rate": 0.0004989572112530871,
"loss": 6.6175,
"mean_token_accuracy": 0.11923456266522407,
"num_tokens": 8314766.0,
"step": 3835
},
{
"entropy": 6.797013807296753,
"epoch": 0.4109369147626946,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989535312995139,
"loss": 6.655,
"mean_token_accuracy": 0.12089989259839058,
"num_tokens": 8325791.0,
"step": 3840
},
{
"entropy": 6.710721731185913,
"epoch": 0.41147198887045855,
"grad_norm": 0.953125,
"learning_rate": 0.0004989498448793139,
"loss": 6.5712,
"mean_token_accuracy": 0.12769034430384635,
"num_tokens": 8335809.0,
"step": 3845
},
{
"entropy": 6.67577633857727,
"epoch": 0.4120070629782225,
"grad_norm": 0.859375,
"learning_rate": 0.0004989461519925935,
"loss": 6.5827,
"mean_token_accuracy": 0.12097218707203865,
"num_tokens": 8347505.0,
"step": 3850
},
{
"entropy": 6.75542802810669,
"epoch": 0.4125421370859864,
"grad_norm": 0.98046875,
"learning_rate": 0.0004989424526394596,
"loss": 6.5775,
"mean_token_accuracy": 0.12156168222427369,
"num_tokens": 8358363.0,
"step": 3855
},
{
"entropy": 6.7438743114471436,
"epoch": 0.41307721119375035,
"grad_norm": 0.89453125,
"learning_rate": 0.0004989387468200187,
"loss": 6.5988,
"mean_token_accuracy": 0.13023255839943887,
"num_tokens": 8369076.0,
"step": 3860
},
{
"entropy": 6.752788734436035,
"epoch": 0.4136122853015143,
"grad_norm": 0.8515625,
"learning_rate": 0.000498935034534378,
"loss": 6.6291,
"mean_token_accuracy": 0.12350267320871353,
"num_tokens": 8380071.0,
"step": 3865
},
{
"entropy": 6.721376562118531,
"epoch": 0.41414735940927816,
"grad_norm": 0.80859375,
"learning_rate": 0.0004989313157826446,
"loss": 6.5938,
"mean_token_accuracy": 0.12291992232203483,
"num_tokens": 8390574.0,
"step": 3870
},
{
"entropy": 6.705049705505371,
"epoch": 0.4146824335170421,
"grad_norm": 0.8125,
"learning_rate": 0.000498927590564926,
"loss": 6.5749,
"mean_token_accuracy": 0.12339994534850121,
"num_tokens": 8401154.0,
"step": 3875
},
{
"entropy": 6.766706037521362,
"epoch": 0.415217507624806,
"grad_norm": 0.84765625,
"learning_rate": 0.0004989238588813296,
"loss": 6.6672,
"mean_token_accuracy": 0.11639968454837799,
"num_tokens": 8412113.0,
"step": 3880
},
{
"entropy": 6.797072076797486,
"epoch": 0.41575258173256996,
"grad_norm": 0.890625,
"learning_rate": 0.0004989201207319634,
"loss": 6.6385,
"mean_token_accuracy": 0.12364522740244865,
"num_tokens": 8422734.0,
"step": 3885
},
{
"entropy": 6.736127710342407,
"epoch": 0.4162876558403339,
"grad_norm": 0.8203125,
"learning_rate": 0.0004989163761169351,
"loss": 6.5662,
"mean_token_accuracy": 0.12166397422552108,
"num_tokens": 8433647.0,
"step": 3890
},
{
"entropy": 6.717603731155395,
"epoch": 0.4168227299480978,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989126250363529,
"loss": 6.6043,
"mean_token_accuracy": 0.12411727085709572,
"num_tokens": 8444726.0,
"step": 3895
},
{
"entropy": 6.732759284973144,
"epoch": 0.41735780405586176,
"grad_norm": 0.9375,
"learning_rate": 0.0004989088674903251,
"loss": 6.6042,
"mean_token_accuracy": 0.12171476036310196,
"num_tokens": 8455351.0,
"step": 3900
},
{
"entropy": 6.69054012298584,
"epoch": 0.41789287816362564,
"grad_norm": 0.8359375,
"learning_rate": 0.0004989051034789602,
"loss": 6.5945,
"mean_token_accuracy": 0.12778979167342186,
"num_tokens": 8466423.0,
"step": 3905
},
{
"entropy": 6.696830415725708,
"epoch": 0.41842795227138957,
"grad_norm": 0.92578125,
"learning_rate": 0.000498901333002367,
"loss": 6.5204,
"mean_token_accuracy": 0.12430474385619164,
"num_tokens": 8475841.0,
"step": 3910
},
{
"entropy": 6.767647314071655,
"epoch": 0.4189630263791535,
"grad_norm": 0.87109375,
"learning_rate": 0.0004988975560606541,
"loss": 6.6309,
"mean_token_accuracy": 0.12111587822437286,
"num_tokens": 8487656.0,
"step": 3915
},
{
"entropy": 6.737721538543701,
"epoch": 0.41949810048691744,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988937726539308,
"loss": 6.6345,
"mean_token_accuracy": 0.11631876155734062,
"num_tokens": 8497959.0,
"step": 3920
},
{
"entropy": 6.702249479293823,
"epoch": 0.42003317459468137,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988899827823062,
"loss": 6.5598,
"mean_token_accuracy": 0.12661055102944374,
"num_tokens": 8508228.0,
"step": 3925
},
{
"entropy": 6.742639064788818,
"epoch": 0.4205682487024453,
"grad_norm": 0.91796875,
"learning_rate": 0.00049888618644589,
"loss": 6.5927,
"mean_token_accuracy": 0.11702611446380615,
"num_tokens": 8518999.0,
"step": 3930
},
{
"entropy": 6.690499782562256,
"epoch": 0.42110332281020924,
"grad_norm": 0.81640625,
"learning_rate": 0.0004988823836447916,
"loss": 6.6409,
"mean_token_accuracy": 0.12285174503922462,
"num_tokens": 8529583.0,
"step": 3935
},
{
"entropy": 6.824066638946533,
"epoch": 0.4216383969179731,
"grad_norm": 1.0703125,
"learning_rate": 0.0004988785743791207,
"loss": 6.6921,
"mean_token_accuracy": 0.11692366898059844,
"num_tokens": 8540069.0,
"step": 3940
},
{
"entropy": 6.752025747299195,
"epoch": 0.42217347102573705,
"grad_norm": 0.95703125,
"learning_rate": 0.0004988747586489876,
"loss": 6.5824,
"mean_token_accuracy": 0.12357909306883812,
"num_tokens": 8550441.0,
"step": 3945
},
{
"entropy": 6.681799650192261,
"epoch": 0.422708545133501,
"grad_norm": 0.9140625,
"learning_rate": 0.000498870936454502,
"loss": 6.543,
"mean_token_accuracy": 0.13031908869743347,
"num_tokens": 8560257.0,
"step": 3950
},
{
"entropy": 6.709078216552735,
"epoch": 0.4232436192412649,
"grad_norm": 0.875,
"learning_rate": 0.0004988671077957749,
"loss": 6.6124,
"mean_token_accuracy": 0.12646457999944688,
"num_tokens": 8570991.0,
"step": 3955
},
{
"entropy": 6.775239324569702,
"epoch": 0.42377869334902885,
"grad_norm": 0.86328125,
"learning_rate": 0.0004988632726729164,
"loss": 6.6754,
"mean_token_accuracy": 0.1174784742295742,
"num_tokens": 8583668.0,
"step": 3960
},
{
"entropy": 6.746391296386719,
"epoch": 0.4243137674567928,
"grad_norm": 0.8515625,
"learning_rate": 0.0004988594310860374,
"loss": 6.6133,
"mean_token_accuracy": 0.12330208793282509,
"num_tokens": 8593727.0,
"step": 3965
},
{
"entropy": 6.668441295623779,
"epoch": 0.4248488415645567,
"grad_norm": 0.76953125,
"learning_rate": 0.0004988555830352485,
"loss": 6.5888,
"mean_token_accuracy": 0.12464616522192955,
"num_tokens": 8605470.0,
"step": 3970
},
{
"entropy": 6.752479600906372,
"epoch": 0.4253839156723206,
"grad_norm": 0.83203125,
"learning_rate": 0.0004988517285206614,
"loss": 6.5361,
"mean_token_accuracy": 0.12684643566608428,
"num_tokens": 8617176.0,
"step": 3975
},
{
"entropy": 6.705382680892944,
"epoch": 0.4259189897800845,
"grad_norm": 0.80078125,
"learning_rate": 0.0004988478675423871,
"loss": 6.6269,
"mean_token_accuracy": 0.12260573580861092,
"num_tokens": 8627756.0,
"step": 3980
},
{
"entropy": 6.75893235206604,
"epoch": 0.42645406388784846,
"grad_norm": 0.8984375,
"learning_rate": 0.0004988440001005369,
"loss": 6.5524,
"mean_token_accuracy": 0.12638817578554154,
"num_tokens": 8637766.0,
"step": 3985
},
{
"entropy": 6.738568544387817,
"epoch": 0.4269891379956124,
"grad_norm": 0.87890625,
"learning_rate": 0.0004988401261952227,
"loss": 6.5903,
"mean_token_accuracy": 0.12412031963467599,
"num_tokens": 8648562.0,
"step": 3990
},
{
"entropy": 6.768758535385132,
"epoch": 0.4275242121033763,
"grad_norm": 0.984375,
"learning_rate": 0.0004988362458265565,
"loss": 6.5978,
"mean_token_accuracy": 0.1206689141690731,
"num_tokens": 8658393.0,
"step": 3995
},
{
"entropy": 6.72909688949585,
"epoch": 0.42805928621114026,
"grad_norm": 0.93359375,
"learning_rate": 0.0004988323589946499,
"loss": 6.5505,
"mean_token_accuracy": 0.12452557906508446,
"num_tokens": 8668365.0,
"step": 4000
},
{
"entropy": 6.706186580657959,
"epoch": 0.4285943603189042,
"grad_norm": 0.91015625,
"learning_rate": 0.0004988284656996156,
"loss": 6.5742,
"mean_token_accuracy": 0.12101777195930481,
"num_tokens": 8679228.0,
"step": 4005
},
{
"entropy": 6.660093975067139,
"epoch": 0.42912943442666807,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988245659415656,
"loss": 6.4583,
"mean_token_accuracy": 0.132504703104496,
"num_tokens": 8688940.0,
"step": 4010
},
{
"entropy": 6.755603170394897,
"epoch": 0.429664508534432,
"grad_norm": 0.91015625,
"learning_rate": 0.0004988206597206129,
"loss": 6.6261,
"mean_token_accuracy": 0.11976755112409591,
"num_tokens": 8699923.0,
"step": 4015
},
{
"entropy": 6.665601968765259,
"epoch": 0.43019958264219593,
"grad_norm": 0.875,
"learning_rate": 0.00049881674703687,
"loss": 6.4542,
"mean_token_accuracy": 0.12824729904532434,
"num_tokens": 8710445.0,
"step": 4020
},
{
"entropy": 6.662732791900635,
"epoch": 0.43073465674995987,
"grad_norm": 0.92578125,
"learning_rate": 0.0004988128278904499,
"loss": 6.5787,
"mean_token_accuracy": 0.12728943005204202,
"num_tokens": 8720686.0,
"step": 4025
},
{
"entropy": 6.717475557327271,
"epoch": 0.4312697308577238,
"grad_norm": 0.8984375,
"learning_rate": 0.000498808902281466,
"loss": 6.5936,
"mean_token_accuracy": 0.12410746738314629,
"num_tokens": 8732122.0,
"step": 4030
},
{
"entropy": 6.69724588394165,
"epoch": 0.43180480496548773,
"grad_norm": 0.8984375,
"learning_rate": 0.0004988049702100315,
"loss": 6.5563,
"mean_token_accuracy": 0.12644634395837784,
"num_tokens": 8744122.0,
"step": 4035
},
{
"entropy": 6.75086350440979,
"epoch": 0.43233987907325166,
"grad_norm": 0.83984375,
"learning_rate": 0.0004988010316762598,
"loss": 6.5638,
"mean_token_accuracy": 0.1258111245930195,
"num_tokens": 8754758.0,
"step": 4040
},
{
"entropy": 6.65916633605957,
"epoch": 0.4328749531810156,
"grad_norm": 0.87890625,
"learning_rate": 0.0004987970866802649,
"loss": 6.5892,
"mean_token_accuracy": 0.11819620206952094,
"num_tokens": 8766577.0,
"step": 4045
},
{
"entropy": 6.749998998641968,
"epoch": 0.4334100272887795,
"grad_norm": 0.90234375,
"learning_rate": 0.0004987931352221605,
"loss": 6.5695,
"mean_token_accuracy": 0.12269530296325684,
"num_tokens": 8777698.0,
"step": 4050
},
{
"entropy": 6.717717838287354,
"epoch": 0.4339451013965434,
"grad_norm": 0.90234375,
"learning_rate": 0.0004987891773020606,
"loss": 6.5937,
"mean_token_accuracy": 0.12456514909863473,
"num_tokens": 8788637.0,
"step": 4055
},
{
"entropy": 6.6792045593261715,
"epoch": 0.43448017550430734,
"grad_norm": 0.9921875,
"learning_rate": 0.0004987852129200799,
"loss": 6.5714,
"mean_token_accuracy": 0.12965806424617768,
"num_tokens": 8799709.0,
"step": 4060
},
{
"entropy": 6.719041061401367,
"epoch": 0.4350152496120713,
"grad_norm": 0.984375,
"learning_rate": 0.0004987812420763326,
"loss": 6.5766,
"mean_token_accuracy": 0.12700210288167,
"num_tokens": 8810469.0,
"step": 4065
},
{
"entropy": 6.795799016952515,
"epoch": 0.4355503237198352,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987772647709333,
"loss": 6.5356,
"mean_token_accuracy": 0.12159947156906128,
"num_tokens": 8821142.0,
"step": 4070
},
{
"entropy": 6.621897602081299,
"epoch": 0.43608539782759914,
"grad_norm": 0.81640625,
"learning_rate": 0.000498773281003997,
"loss": 6.4949,
"mean_token_accuracy": 0.13529240638017653,
"num_tokens": 8832951.0,
"step": 4075
},
{
"entropy": 6.6723052024841305,
"epoch": 0.4366204719353631,
"grad_norm": 0.90625,
"learning_rate": 0.0004987692907756385,
"loss": 6.6075,
"mean_token_accuracy": 0.11884131506085396,
"num_tokens": 8843579.0,
"step": 4080
},
{
"entropy": 6.653975200653076,
"epoch": 0.43715554604312695,
"grad_norm": 0.8828125,
"learning_rate": 0.0004987652940859733,
"loss": 6.5615,
"mean_token_accuracy": 0.11667362749576568,
"num_tokens": 8854612.0,
"step": 4085
},
{
"entropy": 6.680628633499145,
"epoch": 0.4376906201508909,
"grad_norm": 0.89453125,
"learning_rate": 0.0004987612909351167,
"loss": 6.4945,
"mean_token_accuracy": 0.12707522585988046,
"num_tokens": 8865584.0,
"step": 4090
},
{
"entropy": 6.665642642974854,
"epoch": 0.4382256942586548,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987572813231842,
"loss": 6.5216,
"mean_token_accuracy": 0.13140948340296746,
"num_tokens": 8876662.0,
"step": 4095
},
{
"entropy": 6.685274982452393,
"epoch": 0.43876076836641875,
"grad_norm": 0.8515625,
"learning_rate": 0.0004987532652502917,
"loss": 6.5656,
"mean_token_accuracy": 0.12099597677588463,
"num_tokens": 8887249.0,
"step": 4100
},
{
"entropy": 6.698389673233033,
"epoch": 0.4392958424741827,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987492427165551,
"loss": 6.3869,
"mean_token_accuracy": 0.13985745757818221,
"num_tokens": 8898635.0,
"step": 4105
},
{
"entropy": 6.655579090118408,
"epoch": 0.4398309165819466,
"grad_norm": 0.91015625,
"learning_rate": 0.0004987452137220906,
"loss": 6.5696,
"mean_token_accuracy": 0.12459381148219109,
"num_tokens": 8908794.0,
"step": 4110
},
{
"entropy": 6.805545902252197,
"epoch": 0.44036599068971055,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987411782670144,
"loss": 6.5815,
"mean_token_accuracy": 0.12002512514591217,
"num_tokens": 8919015.0,
"step": 4115
},
{
"entropy": 6.803833293914795,
"epoch": 0.44090106479747443,
"grad_norm": 0.95703125,
"learning_rate": 0.0004987371363514431,
"loss": 6.6185,
"mean_token_accuracy": 0.11796738654375076,
"num_tokens": 8929028.0,
"step": 4120
},
{
"entropy": 6.633229732513428,
"epoch": 0.44143613890523836,
"grad_norm": 0.9296875,
"learning_rate": 0.0004987330879754933,
"loss": 6.5136,
"mean_token_accuracy": 0.12121351808309555,
"num_tokens": 8939789.0,
"step": 4125
},
{
"entropy": 6.659311389923095,
"epoch": 0.4419712130130023,
"grad_norm": 0.84765625,
"learning_rate": 0.0004987290331392823,
"loss": 6.4902,
"mean_token_accuracy": 0.13035049065947532,
"num_tokens": 8950657.0,
"step": 4130
},
{
"entropy": 6.771459245681763,
"epoch": 0.44250628712076623,
"grad_norm": 0.8984375,
"learning_rate": 0.0004987249718429269,
"loss": 6.5966,
"mean_token_accuracy": 0.12920275554060937,
"num_tokens": 8960744.0,
"step": 4135
},
{
"entropy": 6.6667276382446286,
"epoch": 0.44304136122853016,
"grad_norm": 0.875,
"learning_rate": 0.0004987209040865441,
"loss": 6.4554,
"mean_token_accuracy": 0.13071943819522858,
"num_tokens": 8972411.0,
"step": 4140
},
{
"entropy": 6.691996479034424,
"epoch": 0.4435764353362941,
"grad_norm": 0.921875,
"learning_rate": 0.0004987168298702519,
"loss": 6.6141,
"mean_token_accuracy": 0.12496730759739876,
"num_tokens": 8982818.0,
"step": 4145
},
{
"entropy": 6.736303377151489,
"epoch": 0.44411150944405803,
"grad_norm": 0.984375,
"learning_rate": 0.0004987127491941675,
"loss": 6.5332,
"mean_token_accuracy": 0.12681084722280503,
"num_tokens": 8993227.0,
"step": 4150
},
{
"entropy": 6.658869075775146,
"epoch": 0.4446465835518219,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987086620584091,
"loss": 6.49,
"mean_token_accuracy": 0.12491258159279824,
"num_tokens": 9003796.0,
"step": 4155
},
{
"entropy": 6.673899793624878,
"epoch": 0.44518165765958584,
"grad_norm": 0.890625,
"learning_rate": 0.0004987045684630944,
"loss": 6.493,
"mean_token_accuracy": 0.13490606471896172,
"num_tokens": 9014931.0,
"step": 4160
},
{
"entropy": 6.723305988311767,
"epoch": 0.44571673176734977,
"grad_norm": 0.86328125,
"learning_rate": 0.0004987004684083417,
"loss": 6.568,
"mean_token_accuracy": 0.12488022521138191,
"num_tokens": 9025312.0,
"step": 4165
},
{
"entropy": 6.641047430038452,
"epoch": 0.4462518058751137,
"grad_norm": 0.88671875,
"learning_rate": 0.0004986963618942694,
"loss": 6.55,
"mean_token_accuracy": 0.13490004763007163,
"num_tokens": 9035563.0,
"step": 4170
},
{
"entropy": 6.764650487899781,
"epoch": 0.44678687998287764,
"grad_norm": 0.89453125,
"learning_rate": 0.0004986922489209961,
"loss": 6.5782,
"mean_token_accuracy": 0.1268523618578911,
"num_tokens": 9046065.0,
"step": 4175
},
{
"entropy": 6.636542081832886,
"epoch": 0.44732195409064157,
"grad_norm": 0.859375,
"learning_rate": 0.0004986881294886406,
"loss": 6.4771,
"mean_token_accuracy": 0.13082251697778702,
"num_tokens": 9056696.0,
"step": 4180
},
{
"entropy": 6.672894048690796,
"epoch": 0.4478570281984055,
"grad_norm": 0.86328125,
"learning_rate": 0.0004986840035973218,
"loss": 6.5487,
"mean_token_accuracy": 0.12361067533493042,
"num_tokens": 9067035.0,
"step": 4185
},
{
"entropy": 6.693661117553711,
"epoch": 0.4483921023061694,
"grad_norm": 0.82421875,
"learning_rate": 0.0004986798712471588,
"loss": 6.5097,
"mean_token_accuracy": 0.12609453201293946,
"num_tokens": 9078253.0,
"step": 4190
},
{
"entropy": 6.8062968254089355,
"epoch": 0.4489271764139333,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986757324382709,
"loss": 6.5838,
"mean_token_accuracy": 0.11728138923645019,
"num_tokens": 9088184.0,
"step": 4195
},
{
"entropy": 6.673310852050781,
"epoch": 0.44946225052169725,
"grad_norm": 0.90625,
"learning_rate": 0.0004986715871707777,
"loss": 6.6401,
"mean_token_accuracy": 0.11650597229599953,
"num_tokens": 9099321.0,
"step": 4200
},
{
"entropy": 6.752178430557251,
"epoch": 0.4499973246294612,
"grad_norm": 0.859375,
"learning_rate": 0.0004986674354447988,
"loss": 6.6718,
"mean_token_accuracy": 0.12286324501037597,
"num_tokens": 9111410.0,
"step": 4205
},
{
"entropy": 6.754798460006714,
"epoch": 0.4505323987372251,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986632772604543,
"loss": 6.4915,
"mean_token_accuracy": 0.12803962752223014,
"num_tokens": 9121932.0,
"step": 4210
},
{
"entropy": 6.554611539840698,
"epoch": 0.45106747284498905,
"grad_norm": 1.0625,
"learning_rate": 0.0004986591126178639,
"loss": 6.5716,
"mean_token_accuracy": 0.13324237689375878,
"num_tokens": 9132686.0,
"step": 4215
},
{
"entropy": 6.649262714385986,
"epoch": 0.451602546952753,
"grad_norm": 1.015625,
"learning_rate": 0.0004986549415171482,
"loss": 6.501,
"mean_token_accuracy": 0.12619779780507087,
"num_tokens": 9143582.0,
"step": 4220
},
{
"entropy": 6.82189040184021,
"epoch": 0.45213762106051686,
"grad_norm": 0.81640625,
"learning_rate": 0.0004986507639584274,
"loss": 6.5821,
"mean_token_accuracy": 0.1168325275182724,
"num_tokens": 9156173.0,
"step": 4225
},
{
"entropy": 6.66492600440979,
"epoch": 0.4526726951682808,
"grad_norm": 0.8671875,
"learning_rate": 0.0004986465799418223,
"loss": 6.5773,
"mean_token_accuracy": 0.12092015594244003,
"num_tokens": 9166882.0,
"step": 4230
},
{
"entropy": 6.669227886199951,
"epoch": 0.4532077692760447,
"grad_norm": 0.77734375,
"learning_rate": 0.0004986423894674534,
"loss": 6.5669,
"mean_token_accuracy": 0.12278837189078332,
"num_tokens": 9177525.0,
"step": 4235
},
{
"entropy": 6.783564901351928,
"epoch": 0.45374284338380866,
"grad_norm": 0.87890625,
"learning_rate": 0.0004986381925354422,
"loss": 6.515,
"mean_token_accuracy": 0.12535930350422858,
"num_tokens": 9188586.0,
"step": 4240
},
{
"entropy": 6.634851264953613,
"epoch": 0.4542779174915726,
"grad_norm": 0.83984375,
"learning_rate": 0.0004986339891459094,
"loss": 6.4353,
"mean_token_accuracy": 0.13272654935717582,
"num_tokens": 9199920.0,
"step": 4245
},
{
"entropy": 6.579446649551391,
"epoch": 0.4548129915993365,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986297792989768,
"loss": 6.4489,
"mean_token_accuracy": 0.13165228366851806,
"num_tokens": 9210923.0,
"step": 4250
},
{
"entropy": 6.771414089202881,
"epoch": 0.45534806570710046,
"grad_norm": 1.0859375,
"learning_rate": 0.0004986255629947655,
"loss": 6.6181,
"mean_token_accuracy": 0.1227750226855278,
"num_tokens": 9221459.0,
"step": 4255
},
{
"entropy": 6.73143310546875,
"epoch": 0.45588313981486434,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986213402333978,
"loss": 6.5599,
"mean_token_accuracy": 0.12391374334692955,
"num_tokens": 9231768.0,
"step": 4260
},
{
"entropy": 6.687683820724487,
"epoch": 0.45641821392262827,
"grad_norm": 0.8515625,
"learning_rate": 0.0004986171110149951,
"loss": 6.5375,
"mean_token_accuracy": 0.12756052240729332,
"num_tokens": 9242230.0,
"step": 4265
},
{
"entropy": 6.7013014793396,
"epoch": 0.4569532880303922,
"grad_norm": 0.890625,
"learning_rate": 0.0004986128753396798,
"loss": 6.4225,
"mean_token_accuracy": 0.14026538878679276,
"num_tokens": 9252485.0,
"step": 4270
},
{
"entropy": 6.55937352180481,
"epoch": 0.45748836213815613,
"grad_norm": 0.953125,
"learning_rate": 0.0004986086332075742,
"loss": 6.518,
"mean_token_accuracy": 0.13188858777284623,
"num_tokens": 9263014.0,
"step": 4275
},
{
"entropy": 6.67518835067749,
"epoch": 0.45802343624592007,
"grad_norm": 0.890625,
"learning_rate": 0.0004986043846188006,
"loss": 6.549,
"mean_token_accuracy": 0.13570101261138917,
"num_tokens": 9273777.0,
"step": 4280
},
{
"entropy": 6.6095668315887455,
"epoch": 0.458558510353684,
"grad_norm": 1.0,
"learning_rate": 0.000498600129573482,
"loss": 6.4885,
"mean_token_accuracy": 0.12344526872038841,
"num_tokens": 9284738.0,
"step": 4285
},
{
"entropy": 6.706184148788452,
"epoch": 0.45909358446144793,
"grad_norm": 0.91015625,
"learning_rate": 0.000498595868071741,
"loss": 6.5584,
"mean_token_accuracy": 0.12425956949591636,
"num_tokens": 9295921.0,
"step": 4290
},
{
"entropy": 6.699369287490844,
"epoch": 0.4596286585692118,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985916001137006,
"loss": 6.524,
"mean_token_accuracy": 0.1261294700205326,
"num_tokens": 9306568.0,
"step": 4295
},
{
"entropy": 6.693387126922607,
"epoch": 0.46016373267697575,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985873256994843,
"loss": 6.4816,
"mean_token_accuracy": 0.12667348012328147,
"num_tokens": 9316468.0,
"step": 4300
},
{
"entropy": 6.6449432373046875,
"epoch": 0.4606988067847397,
"grad_norm": 1.046875,
"learning_rate": 0.0004985830448292154,
"loss": 6.5646,
"mean_token_accuracy": 0.12484502121806144,
"num_tokens": 9328185.0,
"step": 4305
},
{
"entropy": 6.655733108520508,
"epoch": 0.4612338808925036,
"grad_norm": 0.8671875,
"learning_rate": 0.0004985787575030175,
"loss": 6.5055,
"mean_token_accuracy": 0.12664854377508164,
"num_tokens": 9338591.0,
"step": 4310
},
{
"entropy": 6.71276707649231,
"epoch": 0.46176895500026754,
"grad_norm": 0.84375,
"learning_rate": 0.0004985744637210144,
"loss": 6.5447,
"mean_token_accuracy": 0.12165323868393899,
"num_tokens": 9349286.0,
"step": 4315
},
{
"entropy": 6.670126962661743,
"epoch": 0.4623040291080315,
"grad_norm": 0.98046875,
"learning_rate": 0.00049857016348333,
"loss": 6.5149,
"mean_token_accuracy": 0.1324791319668293,
"num_tokens": 9359752.0,
"step": 4320
},
{
"entropy": 6.737763357162476,
"epoch": 0.4628391032157954,
"grad_norm": 0.8828125,
"learning_rate": 0.0004985658567900886,
"loss": 6.5894,
"mean_token_accuracy": 0.11995674669742584,
"num_tokens": 9371823.0,
"step": 4325
},
{
"entropy": 6.588179349899292,
"epoch": 0.4633741773235593,
"grad_norm": 0.85546875,
"learning_rate": 0.0004985615436414145,
"loss": 6.4261,
"mean_token_accuracy": 0.12827276140451432,
"num_tokens": 9383127.0,
"step": 4330
},
{
"entropy": 6.624917984008789,
"epoch": 0.4639092514313232,
"grad_norm": 0.953125,
"learning_rate": 0.0004985572240374321,
"loss": 6.5481,
"mean_token_accuracy": 0.12790464386343955,
"num_tokens": 9394296.0,
"step": 4335
},
{
"entropy": 6.71473331451416,
"epoch": 0.46444432553908715,
"grad_norm": 0.83984375,
"learning_rate": 0.0004985528979782665,
"loss": 6.5219,
"mean_token_accuracy": 0.12147142961621285,
"num_tokens": 9405988.0,
"step": 4340
},
{
"entropy": 6.728419351577759,
"epoch": 0.4649793996468511,
"grad_norm": 0.85546875,
"learning_rate": 0.0004985485654640423,
"loss": 6.5307,
"mean_token_accuracy": 0.1294780731201172,
"num_tokens": 9417479.0,
"step": 4345
},
{
"entropy": 6.677046489715576,
"epoch": 0.465514473754615,
"grad_norm": 0.953125,
"learning_rate": 0.0004985442264948847,
"loss": 6.4781,
"mean_token_accuracy": 0.13104128241539,
"num_tokens": 9428050.0,
"step": 4350
},
{
"entropy": 6.602668333053589,
"epoch": 0.46604954786237895,
"grad_norm": 0.85546875,
"learning_rate": 0.0004985398810709189,
"loss": 6.4289,
"mean_token_accuracy": 0.13291953578591348,
"num_tokens": 9438229.0,
"step": 4355
},
{
"entropy": 6.621841144561768,
"epoch": 0.4665846219701429,
"grad_norm": 0.859375,
"learning_rate": 0.0004985355291922704,
"loss": 6.5209,
"mean_token_accuracy": 0.12576833069324495,
"num_tokens": 9449611.0,
"step": 4360
},
{
"entropy": 6.626019430160523,
"epoch": 0.46711969607790677,
"grad_norm": 0.984375,
"learning_rate": 0.000498531170859065,
"loss": 6.4803,
"mean_token_accuracy": 0.12464792206883431,
"num_tokens": 9460234.0,
"step": 4365
},
{
"entropy": 6.688359355926513,
"epoch": 0.4676547701856707,
"grad_norm": 0.94921875,
"learning_rate": 0.0004985268060714285,
"loss": 6.5359,
"mean_token_accuracy": 0.128062354773283,
"num_tokens": 9472136.0,
"step": 4370
},
{
"entropy": 6.6489707946777346,
"epoch": 0.46818984429343463,
"grad_norm": 0.9609375,
"learning_rate": 0.0004985224348294868,
"loss": 6.4977,
"mean_token_accuracy": 0.13154730200767517,
"num_tokens": 9482602.0,
"step": 4375
},
{
"entropy": 6.75223126411438,
"epoch": 0.46872491840119856,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985180571333663,
"loss": 6.4894,
"mean_token_accuracy": 0.13144660815596582,
"num_tokens": 9493782.0,
"step": 4380
},
{
"entropy": 6.5902422904968265,
"epoch": 0.4692599925089625,
"grad_norm": 0.8515625,
"learning_rate": 0.0004985136729831932,
"loss": 6.4712,
"mean_token_accuracy": 0.12834268659353257,
"num_tokens": 9504389.0,
"step": 4385
},
{
"entropy": 6.615573263168335,
"epoch": 0.46979506661672643,
"grad_norm": 0.8203125,
"learning_rate": 0.0004985092823790942,
"loss": 6.5912,
"mean_token_accuracy": 0.12514225915074348,
"num_tokens": 9515166.0,
"step": 4390
},
{
"entropy": 6.733069658279419,
"epoch": 0.47033014072449036,
"grad_norm": 0.89453125,
"learning_rate": 0.0004985048853211961,
"loss": 6.5085,
"mean_token_accuracy": 0.12728274688124658,
"num_tokens": 9525566.0,
"step": 4395
},
{
"entropy": 6.700274705886841,
"epoch": 0.47086521483225424,
"grad_norm": 0.90625,
"learning_rate": 0.000498500481809626,
"loss": 6.4194,
"mean_token_accuracy": 0.1366426721215248,
"num_tokens": 9536021.0,
"step": 4400
},
{
"entropy": 6.616627550125122,
"epoch": 0.4714002889400182,
"grad_norm": 0.984375,
"learning_rate": 0.0004984960718445107,
"loss": 6.5091,
"mean_token_accuracy": 0.12142780721187592,
"num_tokens": 9546827.0,
"step": 4405
},
{
"entropy": 6.756778526306152,
"epoch": 0.4719353630477821,
"grad_norm": 0.91796875,
"learning_rate": 0.0004984916554259778,
"loss": 6.6272,
"mean_token_accuracy": 0.11766278594732285,
"num_tokens": 9558104.0,
"step": 4410
},
{
"entropy": 6.780975437164306,
"epoch": 0.47247043715554604,
"grad_norm": 0.92578125,
"learning_rate": 0.0004984872325541547,
"loss": 6.5192,
"mean_token_accuracy": 0.1305229753255844,
"num_tokens": 9568487.0,
"step": 4415
},
{
"entropy": 6.638660430908203,
"epoch": 0.47300551126331,
"grad_norm": 0.921875,
"learning_rate": 0.0004984828032291692,
"loss": 6.4634,
"mean_token_accuracy": 0.13527418449521064,
"num_tokens": 9579509.0,
"step": 4420
},
{
"entropy": 6.663227844238281,
"epoch": 0.4735405853710739,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984783674511492,
"loss": 6.5034,
"mean_token_accuracy": 0.12841146439313889,
"num_tokens": 9589931.0,
"step": 4425
},
{
"entropy": 6.6811174869537355,
"epoch": 0.47407565947883784,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984739252202227,
"loss": 6.5353,
"mean_token_accuracy": 0.12742502838373185,
"num_tokens": 9599764.0,
"step": 4430
},
{
"entropy": 6.678421831130981,
"epoch": 0.4746107335866017,
"grad_norm": 0.90625,
"learning_rate": 0.000498469476536518,
"loss": 6.5053,
"mean_token_accuracy": 0.13000835999846458,
"num_tokens": 9611145.0,
"step": 4435
},
{
"entropy": 6.644073152542115,
"epoch": 0.47514580769436565,
"grad_norm": 0.921875,
"learning_rate": 0.0004984650214001636,
"loss": 6.502,
"mean_token_accuracy": 0.1319687359035015,
"num_tokens": 9621441.0,
"step": 4440
},
{
"entropy": 6.634088039398193,
"epoch": 0.4756808818021296,
"grad_norm": 0.92578125,
"learning_rate": 0.0004984605598112881,
"loss": 6.5429,
"mean_token_accuracy": 0.13188807293772697,
"num_tokens": 9631874.0,
"step": 4445
},
{
"entropy": 6.6144147396087645,
"epoch": 0.4762159559098935,
"grad_norm": 0.84765625,
"learning_rate": 0.0004984560917700204,
"loss": 6.522,
"mean_token_accuracy": 0.11959985718131065,
"num_tokens": 9643006.0,
"step": 4450
},
{
"entropy": 6.652404069900513,
"epoch": 0.47675103001765745,
"grad_norm": 0.8671875,
"learning_rate": 0.0004984516172764894,
"loss": 6.4059,
"mean_token_accuracy": 0.1327241726219654,
"num_tokens": 9653897.0,
"step": 4455
},
{
"entropy": 6.608202123641968,
"epoch": 0.4772861041254214,
"grad_norm": 0.87109375,
"learning_rate": 0.0004984471363308243,
"loss": 6.4968,
"mean_token_accuracy": 0.12554212734103204,
"num_tokens": 9664460.0,
"step": 4460
},
{
"entropy": 6.681749153137207,
"epoch": 0.4778211782331853,
"grad_norm": 0.80859375,
"learning_rate": 0.0004984426489331546,
"loss": 6.494,
"mean_token_accuracy": 0.1332928016781807,
"num_tokens": 9674534.0,
"step": 4465
},
{
"entropy": 6.706970596313477,
"epoch": 0.4783562523409492,
"grad_norm": 0.8984375,
"learning_rate": 0.0004984381550836097,
"loss": 6.5747,
"mean_token_accuracy": 0.12295873761177063,
"num_tokens": 9684828.0,
"step": 4470
},
{
"entropy": 6.735209989547729,
"epoch": 0.47889132644871313,
"grad_norm": 0.87890625,
"learning_rate": 0.0004984336547823196,
"loss": 6.538,
"mean_token_accuracy": 0.13103948533535004,
"num_tokens": 9694536.0,
"step": 4475
},
{
"entropy": 6.634707307815551,
"epoch": 0.47942640055647706,
"grad_norm": 0.84765625,
"learning_rate": 0.000498429148029414,
"loss": 6.4102,
"mean_token_accuracy": 0.13732634782791137,
"num_tokens": 9705715.0,
"step": 4480
},
{
"entropy": 6.622171640396118,
"epoch": 0.479961474664241,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984246348250233,
"loss": 6.4753,
"mean_token_accuracy": 0.13292167633771895,
"num_tokens": 9717505.0,
"step": 4485
},
{
"entropy": 6.633249378204345,
"epoch": 0.4804965487720049,
"grad_norm": 0.9375,
"learning_rate": 0.0004984201151692775,
"loss": 6.3711,
"mean_token_accuracy": 0.1316321723163128,
"num_tokens": 9727910.0,
"step": 4490
},
{
"entropy": 6.62043080329895,
"epoch": 0.48103162287976886,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984155890623074,
"loss": 6.4698,
"mean_token_accuracy": 0.13449255973100663,
"num_tokens": 9738528.0,
"step": 4495
},
{
"entropy": 6.671240425109863,
"epoch": 0.4815666969875328,
"grad_norm": 0.91015625,
"learning_rate": 0.0004984110565042435,
"loss": 6.4875,
"mean_token_accuracy": 0.12762050181627274,
"num_tokens": 9749744.0,
"step": 4500
},
{
"entropy": 6.687379980087281,
"epoch": 0.48210177109529667,
"grad_norm": 0.87109375,
"learning_rate": 0.0004984065174952168,
"loss": 6.5035,
"mean_token_accuracy": 0.12369845062494278,
"num_tokens": 9759968.0,
"step": 4505
},
{
"entropy": 6.60925521850586,
"epoch": 0.4826368452030606,
"grad_norm": 0.93359375,
"learning_rate": 0.0004984019720353582,
"loss": 6.5227,
"mean_token_accuracy": 0.12860732153058052,
"num_tokens": 9769791.0,
"step": 4510
},
{
"entropy": 6.633432817459107,
"epoch": 0.48317191931082454,
"grad_norm": 1.0625,
"learning_rate": 0.0004983974201247991,
"loss": 6.4889,
"mean_token_accuracy": 0.1387922890484333,
"num_tokens": 9780369.0,
"step": 4515
},
{
"entropy": 6.638261842727661,
"epoch": 0.48370699341858847,
"grad_norm": 0.8671875,
"learning_rate": 0.0004983928617636709,
"loss": 6.5307,
"mean_token_accuracy": 0.12319683730602264,
"num_tokens": 9791125.0,
"step": 4520
},
{
"entropy": 6.675742292404175,
"epoch": 0.4842420675263524,
"grad_norm": 0.91796875,
"learning_rate": 0.0004983882969521052,
"loss": 6.526,
"mean_token_accuracy": 0.12952028438448906,
"num_tokens": 9801697.0,
"step": 4525
},
{
"entropy": 6.754634857177734,
"epoch": 0.48477714163411634,
"grad_norm": 0.82421875,
"learning_rate": 0.0004983837256902338,
"loss": 6.5621,
"mean_token_accuracy": 0.1297781951725483,
"num_tokens": 9812171.0,
"step": 4530
},
{
"entropy": 6.68020339012146,
"epoch": 0.48531221574188027,
"grad_norm": 0.89453125,
"learning_rate": 0.0004983791479781887,
"loss": 6.4054,
"mean_token_accuracy": 0.1366664409637451,
"num_tokens": 9821596.0,
"step": 4535
},
{
"entropy": 6.679528045654297,
"epoch": 0.48584728984964415,
"grad_norm": 0.90234375,
"learning_rate": 0.0004983745638161021,
"loss": 6.5296,
"mean_token_accuracy": 0.1205269493162632,
"num_tokens": 9831634.0,
"step": 4540
},
{
"entropy": 6.666993045806885,
"epoch": 0.4863823639574081,
"grad_norm": 0.9609375,
"learning_rate": 0.0004983699732041064,
"loss": 6.4882,
"mean_token_accuracy": 0.13205414190888404,
"num_tokens": 9841662.0,
"step": 4545
},
{
"entropy": 6.670028972625732,
"epoch": 0.486917438065172,
"grad_norm": 0.79296875,
"learning_rate": 0.000498365376142334,
"loss": 6.4637,
"mean_token_accuracy": 0.1234777070581913,
"num_tokens": 9853797.0,
"step": 4550
},
{
"entropy": 6.681587982177734,
"epoch": 0.48745251217293595,
"grad_norm": 0.890625,
"learning_rate": 0.0004983607726309179,
"loss": 6.5233,
"mean_token_accuracy": 0.12715924233198167,
"num_tokens": 9865245.0,
"step": 4555
},
{
"entropy": 6.6820274829864506,
"epoch": 0.4879875862806999,
"grad_norm": 0.9140625,
"learning_rate": 0.0004983561626699908,
"loss": 6.6009,
"mean_token_accuracy": 0.11727365702390671,
"num_tokens": 9876207.0,
"step": 4560
},
{
"entropy": 6.639146041870117,
"epoch": 0.4885226603884638,
"grad_norm": 0.9296875,
"learning_rate": 0.0004983515462596859,
"loss": 6.4204,
"mean_token_accuracy": 0.13496807143092154,
"num_tokens": 9887210.0,
"step": 4565
},
{
"entropy": 6.691262245178223,
"epoch": 0.48905773449622775,
"grad_norm": 0.90625,
"learning_rate": 0.0004983469234001365,
"loss": 6.5983,
"mean_token_accuracy": 0.12019298523664475,
"num_tokens": 9899607.0,
"step": 4570
},
{
"entropy": 6.595501470565796,
"epoch": 0.4895928086039917,
"grad_norm": 0.98046875,
"learning_rate": 0.000498342294091476,
"loss": 6.3773,
"mean_token_accuracy": 0.13430055379867553,
"num_tokens": 9909910.0,
"step": 4575
},
{
"entropy": 6.615827751159668,
"epoch": 0.49012788271175556,
"grad_norm": 0.88671875,
"learning_rate": 0.0004983376583338382,
"loss": 6.5079,
"mean_token_accuracy": 0.1254624530673027,
"num_tokens": 9922137.0,
"step": 4580
},
{
"entropy": 6.638407421112061,
"epoch": 0.4906629568195195,
"grad_norm": 0.92578125,
"learning_rate": 0.0004983330161273569,
"loss": 6.5069,
"mean_token_accuracy": 0.13186536133289337,
"num_tokens": 9932514.0,
"step": 4585
},
{
"entropy": 6.654112815856934,
"epoch": 0.4911980309272834,
"grad_norm": 1.1015625,
"learning_rate": 0.0004983283674721662,
"loss": 6.5293,
"mean_token_accuracy": 0.129039965569973,
"num_tokens": 9943054.0,
"step": 4590
},
{
"entropy": 6.740719270706177,
"epoch": 0.49173310503504736,
"grad_norm": 0.88671875,
"learning_rate": 0.0004983237123684002,
"loss": 6.5094,
"mean_token_accuracy": 0.12170372903347015,
"num_tokens": 9953791.0,
"step": 4595
},
{
"entropy": 6.587079620361328,
"epoch": 0.4922681791428113,
"grad_norm": 0.9140625,
"learning_rate": 0.0004983190508161934,
"loss": 6.4627,
"mean_token_accuracy": 0.1294872209429741,
"num_tokens": 9964782.0,
"step": 4600
},
{
"entropy": 6.6498559474945065,
"epoch": 0.4928032532505752,
"grad_norm": 0.87109375,
"learning_rate": 0.0004983143828156804,
"loss": 6.5266,
"mean_token_accuracy": 0.12838228419423103,
"num_tokens": 9975843.0,
"step": 4605
},
{
"entropy": 6.704116678237915,
"epoch": 0.49333832735833916,
"grad_norm": 0.90625,
"learning_rate": 0.0004983097083669959,
"loss": 6.5007,
"mean_token_accuracy": 0.13298282846808435,
"num_tokens": 9987060.0,
"step": 4610
},
{
"entropy": 6.646341371536255,
"epoch": 0.49387340146610303,
"grad_norm": 0.9453125,
"learning_rate": 0.0004983050274702751,
"loss": 6.465,
"mean_token_accuracy": 0.12663726061582564,
"num_tokens": 9997380.0,
"step": 4615
},
{
"entropy": 6.6445447444915775,
"epoch": 0.49440847557386697,
"grad_norm": 0.8984375,
"learning_rate": 0.0004983003401256529,
"loss": 6.5172,
"mean_token_accuracy": 0.12875118479132652,
"num_tokens": 10007955.0,
"step": 4620
},
{
"entropy": 6.78430962562561,
"epoch": 0.4949435496816309,
"grad_norm": 0.984375,
"learning_rate": 0.0004982956463332647,
"loss": 6.585,
"mean_token_accuracy": 0.12407600358128548,
"num_tokens": 10017645.0,
"step": 4625
},
{
"entropy": 6.629924869537353,
"epoch": 0.49547862378939483,
"grad_norm": 0.953125,
"learning_rate": 0.0004982909460932463,
"loss": 6.5323,
"mean_token_accuracy": 0.1285700298845768,
"num_tokens": 10027619.0,
"step": 4630
},
{
"entropy": 6.580510091781616,
"epoch": 0.49601369789715877,
"grad_norm": 0.89453125,
"learning_rate": 0.000498286239405733,
"loss": 6.3937,
"mean_token_accuracy": 0.13874556571245195,
"num_tokens": 10037551.0,
"step": 4635
},
{
"entropy": 6.605573320388794,
"epoch": 0.4965487720049227,
"grad_norm": 0.86328125,
"learning_rate": 0.000498281526270861,
"loss": 6.4215,
"mean_token_accuracy": 0.13800892382860183,
"num_tokens": 10047771.0,
"step": 4640
},
{
"entropy": 6.736871671676636,
"epoch": 0.49708384611268663,
"grad_norm": 0.96875,
"learning_rate": 0.0004982768066887663,
"loss": 6.5738,
"mean_token_accuracy": 0.1221847340464592,
"num_tokens": 10058771.0,
"step": 4645
},
{
"entropy": 6.605147457122802,
"epoch": 0.4976189202204505,
"grad_norm": 0.96875,
"learning_rate": 0.0004982720806595851,
"loss": 6.4377,
"mean_token_accuracy": 0.13613661155104637,
"num_tokens": 10070318.0,
"step": 4650
},
{
"entropy": 6.566076612472534,
"epoch": 0.49815399432821444,
"grad_norm": 0.87890625,
"learning_rate": 0.0004982673481834541,
"loss": 6.4794,
"mean_token_accuracy": 0.12480002790689468,
"num_tokens": 10081810.0,
"step": 4655
},
{
"entropy": 6.641260051727295,
"epoch": 0.4986890684359784,
"grad_norm": 0.88671875,
"learning_rate": 0.0004982626092605097,
"loss": 6.4814,
"mean_token_accuracy": 0.1259951874613762,
"num_tokens": 10091724.0,
"step": 4660
},
{
"entropy": 6.687929010391235,
"epoch": 0.4992241425437423,
"grad_norm": 0.93359375,
"learning_rate": 0.0004982578638908888,
"loss": 6.4605,
"mean_token_accuracy": 0.13417887166142464,
"num_tokens": 10102510.0,
"step": 4665
},
{
"entropy": 6.617040157318115,
"epoch": 0.49975921665150624,
"grad_norm": 0.890625,
"learning_rate": 0.0004982531120747286,
"loss": 6.5499,
"mean_token_accuracy": 0.1230244629085064,
"num_tokens": 10112636.0,
"step": 4670
},
{
"entropy": 6.66232738494873,
"epoch": 0.5002942907592701,
"grad_norm": 0.84765625,
"learning_rate": 0.000498248353812166,
"loss": 6.5287,
"mean_token_accuracy": 0.131372894346714,
"num_tokens": 10123562.0,
"step": 4675
},
{
"entropy": 6.659600734710693,
"epoch": 0.5008293648670341,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982435891033387,
"loss": 6.4954,
"mean_token_accuracy": 0.12019536346197128,
"num_tokens": 10134731.0,
"step": 4680
},
{
"entropy": 6.658867216110229,
"epoch": 0.501364438974798,
"grad_norm": 0.82421875,
"learning_rate": 0.000498238817948384,
"loss": 6.4867,
"mean_token_accuracy": 0.12581022083759308,
"num_tokens": 10145232.0,
"step": 4685
},
{
"entropy": 6.658832597732544,
"epoch": 0.501899513082562,
"grad_norm": 0.91796875,
"learning_rate": 0.0004982340403474398,
"loss": 6.4544,
"mean_token_accuracy": 0.12711327224969865,
"num_tokens": 10155857.0,
"step": 4690
},
{
"entropy": 6.601750755310059,
"epoch": 0.5024345871903259,
"grad_norm": 0.86328125,
"learning_rate": 0.0004982292563006441,
"loss": 6.4533,
"mean_token_accuracy": 0.13474077358841896,
"num_tokens": 10166145.0,
"step": 4695
},
{
"entropy": 6.6569538593292235,
"epoch": 0.5029696612980897,
"grad_norm": 0.95703125,
"learning_rate": 0.000498224465808135,
"loss": 6.5942,
"mean_token_accuracy": 0.12159125730395318,
"num_tokens": 10177086.0,
"step": 4700
},
{
"entropy": 6.703319120407104,
"epoch": 0.5035047354058537,
"grad_norm": 0.9453125,
"learning_rate": 0.0004982196688700509,
"loss": 6.4545,
"mean_token_accuracy": 0.12969210147857665,
"num_tokens": 10187120.0,
"step": 4705
},
{
"entropy": 6.615395879745483,
"epoch": 0.5040398095136176,
"grad_norm": 0.84375,
"learning_rate": 0.0004982148654865302,
"loss": 6.5139,
"mean_token_accuracy": 0.1280078411102295,
"num_tokens": 10197784.0,
"step": 4710
},
{
"entropy": 6.6520003318786625,
"epoch": 0.5045748836213816,
"grad_norm": 0.91796875,
"learning_rate": 0.0004982100556577115,
"loss": 6.3837,
"mean_token_accuracy": 0.13710010796785355,
"num_tokens": 10209373.0,
"step": 4715
},
{
"entropy": 6.653470230102539,
"epoch": 0.5051099577291455,
"grad_norm": 0.9296875,
"learning_rate": 0.0004982052393837338,
"loss": 6.5106,
"mean_token_accuracy": 0.13445726856589318,
"num_tokens": 10220316.0,
"step": 4720
},
{
"entropy": 6.573184156417847,
"epoch": 0.5056450318369095,
"grad_norm": 0.9921875,
"learning_rate": 0.0004982004166647363,
"loss": 6.4179,
"mean_token_accuracy": 0.13162229359149932,
"num_tokens": 10230478.0,
"step": 4725
},
{
"entropy": 6.572769546508789,
"epoch": 0.5061801059446733,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981955875008582,
"loss": 6.5068,
"mean_token_accuracy": 0.13034035861492158,
"num_tokens": 10240641.0,
"step": 4730
},
{
"entropy": 6.663148355484009,
"epoch": 0.5067151800524372,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981907518922387,
"loss": 6.473,
"mean_token_accuracy": 0.13470017835497855,
"num_tokens": 10250676.0,
"step": 4735
},
{
"entropy": 6.7320473194122314,
"epoch": 0.5072502541602012,
"grad_norm": 1.1484375,
"learning_rate": 0.0004981859098390177,
"loss": 6.5316,
"mean_token_accuracy": 0.1317993402481079,
"num_tokens": 10262016.0,
"step": 4740
},
{
"entropy": 6.629784202575683,
"epoch": 0.5077853282679651,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981810613413348,
"loss": 6.4807,
"mean_token_accuracy": 0.1275458127260208,
"num_tokens": 10271985.0,
"step": 4745
},
{
"entropy": 6.645000886917114,
"epoch": 0.5083204023757291,
"grad_norm": 0.95703125,
"learning_rate": 0.0004981762063993302,
"loss": 6.4468,
"mean_token_accuracy": 0.13643463030457498,
"num_tokens": 10282156.0,
"step": 4750
},
{
"entropy": 6.608729267120362,
"epoch": 0.5088554764834929,
"grad_norm": 0.94140625,
"learning_rate": 0.000498171345013144,
"loss": 6.476,
"mean_token_accuracy": 0.12745949104428292,
"num_tokens": 10292576.0,
"step": 4755
},
{
"entropy": 6.594528150558472,
"epoch": 0.5093905505912569,
"grad_norm": 0.875,
"learning_rate": 0.0004981664771829165,
"loss": 6.4777,
"mean_token_accuracy": 0.1305478774011135,
"num_tokens": 10303648.0,
"step": 4760
},
{
"entropy": 6.6102478981018065,
"epoch": 0.5099256246990208,
"grad_norm": 0.859375,
"learning_rate": 0.0004981616029087884,
"loss": 6.354,
"mean_token_accuracy": 0.1412102237343788,
"num_tokens": 10314275.0,
"step": 4765
},
{
"entropy": 6.6168890476226805,
"epoch": 0.5104606988067847,
"grad_norm": 0.84375,
"learning_rate": 0.0004981567221909004,
"loss": 6.48,
"mean_token_accuracy": 0.12254874557256698,
"num_tokens": 10325386.0,
"step": 4770
},
{
"entropy": 6.62398509979248,
"epoch": 0.5109957729145487,
"grad_norm": 1.015625,
"learning_rate": 0.0004981518350293935,
"loss": 6.4652,
"mean_token_accuracy": 0.12788971215486528,
"num_tokens": 10335041.0,
"step": 4775
},
{
"entropy": 6.60414490699768,
"epoch": 0.5115308470223126,
"grad_norm": 0.97265625,
"learning_rate": 0.0004981469414244086,
"loss": 6.4153,
"mean_token_accuracy": 0.13254307731986045,
"num_tokens": 10345746.0,
"step": 4780
},
{
"entropy": 6.59508867263794,
"epoch": 0.5120659211300765,
"grad_norm": 0.890625,
"learning_rate": 0.000498142041376087,
"loss": 6.5051,
"mean_token_accuracy": 0.12980739027261734,
"num_tokens": 10356995.0,
"step": 4785
},
{
"entropy": 6.638574075698853,
"epoch": 0.5126009952378404,
"grad_norm": 0.87890625,
"learning_rate": 0.0004981371348845705,
"loss": 6.4396,
"mean_token_accuracy": 0.13524362444877625,
"num_tokens": 10368445.0,
"step": 4790
},
{
"entropy": 6.630716753005982,
"epoch": 0.5131360693456044,
"grad_norm": 0.875,
"learning_rate": 0.0004981322219500006,
"loss": 6.3772,
"mean_token_accuracy": 0.1370594806969166,
"num_tokens": 10378256.0,
"step": 4795
},
{
"entropy": 6.631394529342652,
"epoch": 0.5136711434533683,
"grad_norm": 0.8515625,
"learning_rate": 0.000498127302572519,
"loss": 6.4759,
"mean_token_accuracy": 0.12833066135644913,
"num_tokens": 10388698.0,
"step": 4800
},
{
"entropy": 6.581023120880127,
"epoch": 0.5142062175611322,
"grad_norm": 0.96484375,
"learning_rate": 0.0004981223767522679,
"loss": 6.4706,
"mean_token_accuracy": 0.1341756135225296,
"num_tokens": 10399975.0,
"step": 4805
},
{
"entropy": 6.611735057830811,
"epoch": 0.5147412916688962,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981174444893896,
"loss": 6.4831,
"mean_token_accuracy": 0.12776545137166978,
"num_tokens": 10410480.0,
"step": 4810
},
{
"entropy": 6.629581546783447,
"epoch": 0.51527636577666,
"grad_norm": 0.87109375,
"learning_rate": 0.0004981125057840264,
"loss": 6.4404,
"mean_token_accuracy": 0.13301406651735306,
"num_tokens": 10421023.0,
"step": 4815
},
{
"entropy": 6.635268592834473,
"epoch": 0.515811439884424,
"grad_norm": 0.87109375,
"learning_rate": 0.000498107560636321,
"loss": 6.4832,
"mean_token_accuracy": 0.13234892711043358,
"num_tokens": 10431152.0,
"step": 4820
},
{
"entropy": 6.609623670578003,
"epoch": 0.5163465139921879,
"grad_norm": 0.921875,
"learning_rate": 0.0004981026090464161,
"loss": 6.3907,
"mean_token_accuracy": 0.12726799473166467,
"num_tokens": 10441986.0,
"step": 4825
},
{
"entropy": 6.665307092666626,
"epoch": 0.5168815880999519,
"grad_norm": 0.91015625,
"learning_rate": 0.0004980976510144548,
"loss": 6.4697,
"mean_token_accuracy": 0.1312413029372692,
"num_tokens": 10452932.0,
"step": 4830
},
{
"entropy": 6.657898330688477,
"epoch": 0.5174166622077158,
"grad_norm": 0.921875,
"learning_rate": 0.0004980926865405801,
"loss": 6.5334,
"mean_token_accuracy": 0.13006979301571847,
"num_tokens": 10463411.0,
"step": 4835
},
{
"entropy": 6.612054872512817,
"epoch": 0.5179517363154797,
"grad_norm": 0.9140625,
"learning_rate": 0.0004980877156249354,
"loss": 6.4312,
"mean_token_accuracy": 0.12908995747566224,
"num_tokens": 10474820.0,
"step": 4840
},
{
"entropy": 6.605563735961914,
"epoch": 0.5184868104232436,
"grad_norm": 1.109375,
"learning_rate": 0.0004980827382676643,
"loss": 6.4336,
"mean_token_accuracy": 0.13449087291955947,
"num_tokens": 10485701.0,
"step": 4845
},
{
"entropy": 6.629001617431641,
"epoch": 0.5190218845310075,
"grad_norm": 0.91796875,
"learning_rate": 0.0004980777544689105,
"loss": 6.4242,
"mean_token_accuracy": 0.13291734606027603,
"num_tokens": 10495825.0,
"step": 4850
},
{
"entropy": 6.63130373954773,
"epoch": 0.5195569586387715,
"grad_norm": 0.9765625,
"learning_rate": 0.0004980727642288178,
"loss": 6.4235,
"mean_token_accuracy": 0.1373104438185692,
"num_tokens": 10505847.0,
"step": 4855
},
{
"entropy": 6.620811843872071,
"epoch": 0.5200920327465354,
"grad_norm": 0.8671875,
"learning_rate": 0.0004980677675475305,
"loss": 6.4745,
"mean_token_accuracy": 0.12906499356031417,
"num_tokens": 10516596.0,
"step": 4860
},
{
"entropy": 6.615943861007691,
"epoch": 0.5206271068542994,
"grad_norm": 0.8984375,
"learning_rate": 0.0004980627644251926,
"loss": 6.3881,
"mean_token_accuracy": 0.13836821541190147,
"num_tokens": 10528175.0,
"step": 4865
},
{
"entropy": 6.57639708518982,
"epoch": 0.5211621809620632,
"grad_norm": 0.81640625,
"learning_rate": 0.0004980577548619489,
"loss": 6.4849,
"mean_token_accuracy": 0.12840275838971138,
"num_tokens": 10540076.0,
"step": 4870
},
{
"entropy": 6.621571588516235,
"epoch": 0.5216972550698272,
"grad_norm": 0.8984375,
"learning_rate": 0.0004980527388579437,
"loss": 6.5063,
"mean_token_accuracy": 0.12991923615336418,
"num_tokens": 10552025.0,
"step": 4875
},
{
"entropy": 6.643404912948609,
"epoch": 0.5222323291775911,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980477164133221,
"loss": 6.4985,
"mean_token_accuracy": 0.12801033034920692,
"num_tokens": 10562535.0,
"step": 4880
},
{
"entropy": 6.591488313674927,
"epoch": 0.522767403285355,
"grad_norm": 0.8828125,
"learning_rate": 0.000498042687528229,
"loss": 6.433,
"mean_token_accuracy": 0.1368302322924137,
"num_tokens": 10572989.0,
"step": 4885
},
{
"entropy": 6.578029823303223,
"epoch": 0.523302477393119,
"grad_norm": 0.953125,
"learning_rate": 0.0004980376522028098,
"loss": 6.4311,
"mean_token_accuracy": 0.13608148843050002,
"num_tokens": 10583033.0,
"step": 4890
},
{
"entropy": 6.67541823387146,
"epoch": 0.5238375515008828,
"grad_norm": 0.87890625,
"learning_rate": 0.0004980326104372095,
"loss": 6.5134,
"mean_token_accuracy": 0.13113198056817055,
"num_tokens": 10595205.0,
"step": 4895
},
{
"entropy": 6.633107995986938,
"epoch": 0.5243726256086468,
"grad_norm": 0.95703125,
"learning_rate": 0.000498027562231574,
"loss": 6.3799,
"mean_token_accuracy": 0.1369300030171871,
"num_tokens": 10605244.0,
"step": 4900
},
{
"entropy": 6.548080635070801,
"epoch": 0.5249076997164107,
"grad_norm": 0.83203125,
"learning_rate": 0.000498022507586049,
"loss": 6.4421,
"mean_token_accuracy": 0.13095175921916963,
"num_tokens": 10615817.0,
"step": 4905
},
{
"entropy": 6.63610873222351,
"epoch": 0.5254427738241747,
"grad_norm": 0.8671875,
"learning_rate": 0.0004980174465007804,
"loss": 6.4653,
"mean_token_accuracy": 0.1272575758397579,
"num_tokens": 10627160.0,
"step": 4910
},
{
"entropy": 6.613020944595337,
"epoch": 0.5259778479319386,
"grad_norm": 0.921875,
"learning_rate": 0.0004980123789759145,
"loss": 6.4621,
"mean_token_accuracy": 0.13695163801312446,
"num_tokens": 10637552.0,
"step": 4915
},
{
"entropy": 6.633289575576782,
"epoch": 0.5265129220397025,
"grad_norm": 0.96875,
"learning_rate": 0.0004980073050115973,
"loss": 6.4837,
"mean_token_accuracy": 0.13580890074372293,
"num_tokens": 10648773.0,
"step": 4920
},
{
"entropy": 6.6644628047943115,
"epoch": 0.5270479961474664,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980022246079755,
"loss": 6.4893,
"mean_token_accuracy": 0.12900413498282431,
"num_tokens": 10659537.0,
"step": 4925
},
{
"entropy": 6.678333568572998,
"epoch": 0.5275830702552303,
"grad_norm": 0.8671875,
"learning_rate": 0.000497997137765196,
"loss": 6.5279,
"mean_token_accuracy": 0.1248202569782734,
"num_tokens": 10671390.0,
"step": 4930
},
{
"entropy": 6.620291471481323,
"epoch": 0.5281181443629943,
"grad_norm": 0.96484375,
"learning_rate": 0.0004979920444834054,
"loss": 6.3809,
"mean_token_accuracy": 0.14244376122951508,
"num_tokens": 10682389.0,
"step": 4935
},
{
"entropy": 6.521419858932495,
"epoch": 0.5286532184707582,
"grad_norm": 0.87890625,
"learning_rate": 0.0004979869447627508,
"loss": 6.4436,
"mean_token_accuracy": 0.13305451348423958,
"num_tokens": 10692400.0,
"step": 4940
},
{
"entropy": 6.563894271850586,
"epoch": 0.5291882925785222,
"grad_norm": 0.90625,
"learning_rate": 0.0004979818386033795,
"loss": 6.3639,
"mean_token_accuracy": 0.13524384126067163,
"num_tokens": 10702396.0,
"step": 4945
},
{
"entropy": 6.578342390060425,
"epoch": 0.5297233666862861,
"grad_norm": 0.83984375,
"learning_rate": 0.000497976726005439,
"loss": 6.4823,
"mean_token_accuracy": 0.1315547928214073,
"num_tokens": 10712863.0,
"step": 4950
},
{
"entropy": 6.675341939926147,
"epoch": 0.5302584407940499,
"grad_norm": 0.86328125,
"learning_rate": 0.0004979716069690768,
"loss": 6.4897,
"mean_token_accuracy": 0.12950288206338884,
"num_tokens": 10724234.0,
"step": 4955
},
{
"entropy": 6.723803901672364,
"epoch": 0.5307935149018139,
"grad_norm": 1.25,
"learning_rate": 0.0004979664814944409,
"loss": 6.5076,
"mean_token_accuracy": 0.12615261673927308,
"num_tokens": 10734405.0,
"step": 4960
},
{
"entropy": 6.647532176971436,
"epoch": 0.5313285890095778,
"grad_norm": 0.91015625,
"learning_rate": 0.0004979613495816791,
"loss": 6.4501,
"mean_token_accuracy": 0.13496298864483833,
"num_tokens": 10744573.0,
"step": 4965
},
{
"entropy": 6.642793846130371,
"epoch": 0.5318636631173418,
"grad_norm": 0.84765625,
"learning_rate": 0.0004979562112309397,
"loss": 6.5223,
"mean_token_accuracy": 0.13261223658919336,
"num_tokens": 10755555.0,
"step": 4970
},
{
"entropy": 6.597110176086426,
"epoch": 0.5323987372251057,
"grad_norm": 0.8359375,
"learning_rate": 0.000497951066442371,
"loss": 6.53,
"mean_token_accuracy": 0.123849056661129,
"num_tokens": 10766493.0,
"step": 4975
},
{
"entropy": 6.682430124282837,
"epoch": 0.5329338113328697,
"grad_norm": 0.8984375,
"learning_rate": 0.0004979459152161215,
"loss": 6.4762,
"mean_token_accuracy": 0.12446538880467414,
"num_tokens": 10776735.0,
"step": 4980
},
{
"entropy": 6.585795831680298,
"epoch": 0.5334688854406335,
"grad_norm": 0.96875,
"learning_rate": 0.0004979407575523402,
"loss": 6.4788,
"mean_token_accuracy": 0.13038726896047592,
"num_tokens": 10787500.0,
"step": 4985
},
{
"entropy": 6.5987447738647464,
"epoch": 0.5340039595483974,
"grad_norm": 0.9375,
"learning_rate": 0.0004979355934511757,
"loss": 6.4956,
"mean_token_accuracy": 0.13150699660182,
"num_tokens": 10798002.0,
"step": 4990
},
{
"entropy": 6.60825834274292,
"epoch": 0.5345390336561614,
"grad_norm": 0.94140625,
"learning_rate": 0.0004979304229127773,
"loss": 6.44,
"mean_token_accuracy": 0.13437903374433519,
"num_tokens": 10808982.0,
"step": 4995
},
{
"entropy": 6.610709095001221,
"epoch": 0.5350741077639253,
"grad_norm": 0.890625,
"learning_rate": 0.0004979252459372944,
"loss": 6.3928,
"mean_token_accuracy": 0.13411291986703872,
"num_tokens": 10821619.0,
"step": 5000
},
{
"entropy": 6.623658800125122,
"epoch": 0.5356091818716893,
"grad_norm": 0.8359375,
"learning_rate": 0.0004979200625248761,
"loss": 6.4286,
"mean_token_accuracy": 0.13552514240145683,
"num_tokens": 10832344.0,
"step": 5005
},
{
"entropy": 6.642703866958618,
"epoch": 0.5361442559794531,
"grad_norm": 0.98046875,
"learning_rate": 0.0004979148726756725,
"loss": 6.3927,
"mean_token_accuracy": 0.13105069994926452,
"num_tokens": 10843149.0,
"step": 5010
},
{
"entropy": 6.594963884353637,
"epoch": 0.5366793300872171,
"grad_norm": 0.79296875,
"learning_rate": 0.0004979096763898333,
"loss": 6.4409,
"mean_token_accuracy": 0.13659456893801689,
"num_tokens": 10854590.0,
"step": 5015
},
{
"entropy": 6.635641431808471,
"epoch": 0.537214404194981,
"grad_norm": 0.86328125,
"learning_rate": 0.0004979044736675084,
"loss": 6.4088,
"mean_token_accuracy": 0.13173776045441626,
"num_tokens": 10865500.0,
"step": 5020
},
{
"entropy": 6.591180086135864,
"epoch": 0.5377494783027449,
"grad_norm": 1.5234375,
"learning_rate": 0.0004978992645088483,
"loss": 6.4565,
"mean_token_accuracy": 0.13780646920204162,
"num_tokens": 10876866.0,
"step": 5025
},
{
"entropy": 6.486580514907837,
"epoch": 0.5382845524105089,
"grad_norm": 1.0234375,
"learning_rate": 0.000497894048914003,
"loss": 6.4116,
"mean_token_accuracy": 0.13192799538373948,
"num_tokens": 10888627.0,
"step": 5030
},
{
"entropy": 6.687264013290405,
"epoch": 0.5388196265182728,
"grad_norm": 0.84375,
"learning_rate": 0.0004978888268831236,
"loss": 6.5199,
"mean_token_accuracy": 0.12521881535649299,
"num_tokens": 10900992.0,
"step": 5035
},
{
"entropy": 6.708090019226074,
"epoch": 0.5393547006260367,
"grad_norm": 0.9140625,
"learning_rate": 0.0004978835984163606,
"loss": 6.5639,
"mean_token_accuracy": 0.1180639810860157,
"num_tokens": 10912323.0,
"step": 5040
},
{
"entropy": 6.620733070373535,
"epoch": 0.5398897747338006,
"grad_norm": 0.94140625,
"learning_rate": 0.0004978783635138649,
"loss": 6.4856,
"mean_token_accuracy": 0.1326413333415985,
"num_tokens": 10923991.0,
"step": 5045
},
{
"entropy": 6.610896444320678,
"epoch": 0.5404248488415646,
"grad_norm": 0.9140625,
"learning_rate": 0.0004978731221757878,
"loss": 6.3531,
"mean_token_accuracy": 0.13417525365948677,
"num_tokens": 10934081.0,
"step": 5050
},
{
"entropy": 6.5825488567352295,
"epoch": 0.5409599229493285,
"grad_norm": 1.0,
"learning_rate": 0.0004978678744022808,
"loss": 6.4329,
"mean_token_accuracy": 0.135209359228611,
"num_tokens": 10944729.0,
"step": 5055
},
{
"entropy": 6.665228748321534,
"epoch": 0.5414949970570924,
"grad_norm": 0.80859375,
"learning_rate": 0.000497862620193495,
"loss": 6.4855,
"mean_token_accuracy": 0.1289301633834839,
"num_tokens": 10955708.0,
"step": 5060
},
{
"entropy": 6.603465986251831,
"epoch": 0.5420300711648564,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978573595495826,
"loss": 6.4197,
"mean_token_accuracy": 0.12746732607483863,
"num_tokens": 10966427.0,
"step": 5065
},
{
"entropy": 6.606443548202515,
"epoch": 0.5425651452726202,
"grad_norm": 0.9609375,
"learning_rate": 0.000497852092470695,
"loss": 6.4768,
"mean_token_accuracy": 0.13206957876682282,
"num_tokens": 10977741.0,
"step": 5070
},
{
"entropy": 6.642637538909912,
"epoch": 0.5431002193803842,
"grad_norm": 0.85546875,
"learning_rate": 0.0004978468189569847,
"loss": 6.4354,
"mean_token_accuracy": 0.1358490638434887,
"num_tokens": 10989127.0,
"step": 5075
},
{
"entropy": 6.555223178863526,
"epoch": 0.5436352934881481,
"grad_norm": 0.9140625,
"learning_rate": 0.0004978415390086038,
"loss": 6.4441,
"mean_token_accuracy": 0.12885510921478271,
"num_tokens": 10999791.0,
"step": 5080
},
{
"entropy": 6.667173433303833,
"epoch": 0.5441703675959121,
"grad_norm": 0.921875,
"learning_rate": 0.0004978362526257047,
"loss": 6.5002,
"mean_token_accuracy": 0.12577469870448113,
"num_tokens": 11010045.0,
"step": 5085
},
{
"entropy": 6.640448665618896,
"epoch": 0.544705441703676,
"grad_norm": 0.88671875,
"learning_rate": 0.0004978309598084403,
"loss": 6.4332,
"mean_token_accuracy": 0.13017582893371582,
"num_tokens": 11020729.0,
"step": 5090
},
{
"entropy": 6.631437110900879,
"epoch": 0.5452405158114398,
"grad_norm": 0.9375,
"learning_rate": 0.0004978256605569631,
"loss": 6.4522,
"mean_token_accuracy": 0.1322314776480198,
"num_tokens": 11030641.0,
"step": 5095
},
{
"entropy": 6.630730199813843,
"epoch": 0.5457755899192038,
"grad_norm": 0.86328125,
"learning_rate": 0.0004978203548714262,
"loss": 6.4586,
"mean_token_accuracy": 0.13038678020238875,
"num_tokens": 11041533.0,
"step": 5100
},
{
"entropy": 6.595696926116943,
"epoch": 0.5463106640269677,
"grad_norm": 0.79296875,
"learning_rate": 0.000497815042751983,
"loss": 6.4069,
"mean_token_accuracy": 0.12977249100804328,
"num_tokens": 11052170.0,
"step": 5105
},
{
"entropy": 6.540053987503052,
"epoch": 0.5468457381347317,
"grad_norm": 0.8671875,
"learning_rate": 0.0004978097241987868,
"loss": 6.3767,
"mean_token_accuracy": 0.1344059519469738,
"num_tokens": 11062872.0,
"step": 5110
},
{
"entropy": 6.614701986312866,
"epoch": 0.5473808122424956,
"grad_norm": 0.890625,
"learning_rate": 0.000497804399211991,
"loss": 6.4883,
"mean_token_accuracy": 0.1307993248105049,
"num_tokens": 11074199.0,
"step": 5115
},
{
"entropy": 6.635615634918213,
"epoch": 0.5479158863502596,
"grad_norm": 0.95703125,
"learning_rate": 0.0004977990677917495,
"loss": 6.4407,
"mean_token_accuracy": 0.13429288640618325,
"num_tokens": 11083881.0,
"step": 5120
},
{
"entropy": 6.56947956085205,
"epoch": 0.5484509604580234,
"grad_norm": 0.875,
"learning_rate": 0.0004977937299382162,
"loss": 6.4897,
"mean_token_accuracy": 0.12704017013311386,
"num_tokens": 11094374.0,
"step": 5125
},
{
"entropy": 6.591516447067261,
"epoch": 0.5489860345657873,
"grad_norm": 1.25,
"learning_rate": 0.0004977883856515453,
"loss": 6.3252,
"mean_token_accuracy": 0.14053784012794496,
"num_tokens": 11105174.0,
"step": 5130
},
{
"entropy": 6.547975683212281,
"epoch": 0.5495211086735513,
"grad_norm": 0.8671875,
"learning_rate": 0.000497783034931891,
"loss": 6.498,
"mean_token_accuracy": 0.13169048056006433,
"num_tokens": 11117223.0,
"step": 5135
},
{
"entropy": 6.671089124679566,
"epoch": 0.5500561827813152,
"grad_norm": 0.92578125,
"learning_rate": 0.000497777677779408,
"loss": 6.4192,
"mean_token_accuracy": 0.12979228496551515,
"num_tokens": 11127240.0,
"step": 5140
},
{
"entropy": 6.596736049652099,
"epoch": 0.5505912568890792,
"grad_norm": 0.9140625,
"learning_rate": 0.0004977723141942509,
"loss": 6.3724,
"mean_token_accuracy": 0.1445842854678631,
"num_tokens": 11137287.0,
"step": 5145
},
{
"entropy": 6.581566190719604,
"epoch": 0.551126330996843,
"grad_norm": 0.95703125,
"learning_rate": 0.0004977669441765743,
"loss": 6.5663,
"mean_token_accuracy": 0.12571654692292214,
"num_tokens": 11148988.0,
"step": 5150
},
{
"entropy": 6.66612982749939,
"epoch": 0.551661405104607,
"grad_norm": 0.90625,
"learning_rate": 0.0004977615677265336,
"loss": 6.4666,
"mean_token_accuracy": 0.13236356154084206,
"num_tokens": 11159874.0,
"step": 5155
},
{
"entropy": 6.656392335891724,
"epoch": 0.5521964792123709,
"grad_norm": 0.98046875,
"learning_rate": 0.000497756184844284,
"loss": 6.4194,
"mean_token_accuracy": 0.1348666474223137,
"num_tokens": 11170703.0,
"step": 5160
},
{
"entropy": 6.514084815979004,
"epoch": 0.5527315533201348,
"grad_norm": 0.86328125,
"learning_rate": 0.0004977507955299809,
"loss": 6.303,
"mean_token_accuracy": 0.14367334991693498,
"num_tokens": 11182230.0,
"step": 5165
},
{
"entropy": 6.5705219268798825,
"epoch": 0.5532666274278988,
"grad_norm": 0.90625,
"learning_rate": 0.0004977453997837797,
"loss": 6.4047,
"mean_token_accuracy": 0.12478601410984994,
"num_tokens": 11192897.0,
"step": 5170
},
{
"entropy": 6.637944078445434,
"epoch": 0.5538017015356627,
"grad_norm": 0.9609375,
"learning_rate": 0.0004977399976058366,
"loss": 6.4917,
"mean_token_accuracy": 0.12794631570577622,
"num_tokens": 11203942.0,
"step": 5175
},
{
"entropy": 6.657525253295899,
"epoch": 0.5543367756434266,
"grad_norm": 0.90625,
"learning_rate": 0.0004977345889963072,
"loss": 6.5191,
"mean_token_accuracy": 0.12642529159784316,
"num_tokens": 11214361.0,
"step": 5180
},
{
"entropy": 6.554194641113281,
"epoch": 0.5548718497511905,
"grad_norm": 0.90234375,
"learning_rate": 0.0004977291739553479,
"loss": 6.4321,
"mean_token_accuracy": 0.13422846421599388,
"num_tokens": 11225928.0,
"step": 5185
},
{
"entropy": 6.594499778747559,
"epoch": 0.5554069238589545,
"grad_norm": 0.890625,
"learning_rate": 0.0004977237524831149,
"loss": 6.3949,
"mean_token_accuracy": 0.12422026544809342,
"num_tokens": 11236660.0,
"step": 5190
},
{
"entropy": 6.540923309326172,
"epoch": 0.5559419979667184,
"grad_norm": 0.8671875,
"learning_rate": 0.0004977183245797649,
"loss": 6.4001,
"mean_token_accuracy": 0.14284231811761855,
"num_tokens": 11247272.0,
"step": 5195
},
{
"entropy": 6.628225469589234,
"epoch": 0.5564770720744823,
"grad_norm": 0.94921875,
"learning_rate": 0.0004977128902454547,
"loss": 6.5004,
"mean_token_accuracy": 0.12502486854791642,
"num_tokens": 11258472.0,
"step": 5200
},
{
"entropy": 6.670383024215698,
"epoch": 0.5570121461822463,
"grad_norm": 0.90625,
"learning_rate": 0.0004977074494803409,
"loss": 6.392,
"mean_token_accuracy": 0.14113382995128632,
"num_tokens": 11268344.0,
"step": 5205
},
{
"entropy": 6.5701521873474125,
"epoch": 0.5575472202900101,
"grad_norm": 0.87109375,
"learning_rate": 0.0004977020022845809,
"loss": 6.4905,
"mean_token_accuracy": 0.12506455108523368,
"num_tokens": 11279931.0,
"step": 5210
},
{
"entropy": 6.671869421005249,
"epoch": 0.5580822943977741,
"grad_norm": 0.9140625,
"learning_rate": 0.0004976965486583318,
"loss": 6.4459,
"mean_token_accuracy": 0.1281243644654751,
"num_tokens": 11290595.0,
"step": 5215
},
{
"entropy": 6.58588662147522,
"epoch": 0.558617368505538,
"grad_norm": 0.9140625,
"learning_rate": 0.0004976910886017513,
"loss": 6.4646,
"mean_token_accuracy": 0.13403649777173995,
"num_tokens": 11300774.0,
"step": 5220
},
{
"entropy": 6.6014903545379635,
"epoch": 0.559152442613302,
"grad_norm": 0.90234375,
"learning_rate": 0.0004976856221149969,
"loss": 6.3569,
"mean_token_accuracy": 0.14249046072363852,
"num_tokens": 11310288.0,
"step": 5225
},
{
"entropy": 6.588343715667724,
"epoch": 0.5596875167210659,
"grad_norm": 0.875,
"learning_rate": 0.0004976801491982263,
"loss": 6.4484,
"mean_token_accuracy": 0.1302250012755394,
"num_tokens": 11321599.0,
"step": 5230
},
{
"entropy": 6.632518863677978,
"epoch": 0.5602225908288297,
"grad_norm": 0.9765625,
"learning_rate": 0.0004976746698515977,
"loss": 6.4876,
"mean_token_accuracy": 0.13621862158179282,
"num_tokens": 11331383.0,
"step": 5235
},
{
"entropy": 6.606530237197876,
"epoch": 0.5607576649365937,
"grad_norm": 0.86328125,
"learning_rate": 0.0004976691840752694,
"loss": 6.452,
"mean_token_accuracy": 0.1273598402738571,
"num_tokens": 11342085.0,
"step": 5240
},
{
"entropy": 6.596298360824585,
"epoch": 0.5612927390443576,
"grad_norm": 0.79296875,
"learning_rate": 0.0004976636918693996,
"loss": 6.4856,
"mean_token_accuracy": 0.12637143433094025,
"num_tokens": 11354061.0,
"step": 5245
},
{
"entropy": 6.6884543895721436,
"epoch": 0.5618278131521216,
"grad_norm": 0.8984375,
"learning_rate": 0.0004976581932341471,
"loss": 6.4222,
"mean_token_accuracy": 0.13169787973165512,
"num_tokens": 11364448.0,
"step": 5250
},
{
"entropy": 6.689448833465576,
"epoch": 0.5623628872598855,
"grad_norm": 0.921875,
"learning_rate": 0.0004976526881696706,
"loss": 6.3971,
"mean_token_accuracy": 0.13254007026553155,
"num_tokens": 11374838.0,
"step": 5255
},
{
"entropy": 6.508886289596558,
"epoch": 0.5628979613676495,
"grad_norm": 0.921875,
"learning_rate": 0.0004976471766761288,
"loss": 6.4524,
"mean_token_accuracy": 0.13301875442266464,
"num_tokens": 11385310.0,
"step": 5260
},
{
"entropy": 6.587552404403686,
"epoch": 0.5634330354754133,
"grad_norm": 0.91796875,
"learning_rate": 0.0004976416587536812,
"loss": 6.4347,
"mean_token_accuracy": 0.12732182890176774,
"num_tokens": 11396590.0,
"step": 5265
},
{
"entropy": 6.618044662475586,
"epoch": 0.5639681095831772,
"grad_norm": 0.875,
"learning_rate": 0.0004976361344024871,
"loss": 6.3971,
"mean_token_accuracy": 0.13341034501791,
"num_tokens": 11407218.0,
"step": 5270
},
{
"entropy": 6.498771047592163,
"epoch": 0.5645031836909412,
"grad_norm": 0.95703125,
"learning_rate": 0.0004976306036227056,
"loss": 6.4152,
"mean_token_accuracy": 0.1334387518465519,
"num_tokens": 11417716.0,
"step": 5275
},
{
"entropy": 6.6630606174469,
"epoch": 0.5650382577987051,
"grad_norm": 0.90234375,
"learning_rate": 0.000497625066414497,
"loss": 6.4962,
"mean_token_accuracy": 0.12818157002329827,
"num_tokens": 11428295.0,
"step": 5280
},
{
"entropy": 6.655110549926758,
"epoch": 0.5655733319064691,
"grad_norm": 0.84765625,
"learning_rate": 0.0004976195227780207,
"loss": 6.432,
"mean_token_accuracy": 0.12855115234851838,
"num_tokens": 11439044.0,
"step": 5285
},
{
"entropy": 6.605824327468872,
"epoch": 0.566108406014233,
"grad_norm": 0.90625,
"learning_rate": 0.0004976139727134371,
"loss": 6.3837,
"mean_token_accuracy": 0.1281607411801815,
"num_tokens": 11450029.0,
"step": 5290
},
{
"entropy": 6.5651922702789305,
"epoch": 0.566643480121997,
"grad_norm": 0.92578125,
"learning_rate": 0.0004976084162209062,
"loss": 6.5481,
"mean_token_accuracy": 0.12115254476666451,
"num_tokens": 11461190.0,
"step": 5295
},
{
"entropy": 6.587199306488037,
"epoch": 0.5671785542297608,
"grad_norm": 0.79296875,
"learning_rate": 0.0004976028533005886,
"loss": 6.4074,
"mean_token_accuracy": 0.13242645785212517,
"num_tokens": 11472143.0,
"step": 5300
},
{
"entropy": 6.648195028305054,
"epoch": 0.5677136283375247,
"grad_norm": 0.91015625,
"learning_rate": 0.0004975972839526449,
"loss": 6.5453,
"mean_token_accuracy": 0.12057378143072128,
"num_tokens": 11483745.0,
"step": 5305
},
{
"entropy": 6.601594543457031,
"epoch": 0.5682487024452887,
"grad_norm": 0.90234375,
"learning_rate": 0.0004975917081772358,
"loss": 6.3929,
"mean_token_accuracy": 0.13485434129834176,
"num_tokens": 11495695.0,
"step": 5310
},
{
"entropy": 6.634777784347534,
"epoch": 0.5687837765530526,
"grad_norm": 0.94921875,
"learning_rate": 0.0004975861259745225,
"loss": 6.4454,
"mean_token_accuracy": 0.13447388708591462,
"num_tokens": 11505576.0,
"step": 5315
},
{
"entropy": 6.576272439956665,
"epoch": 0.5693188506608166,
"grad_norm": 0.8984375,
"learning_rate": 0.0004975805373446662,
"loss": 6.2933,
"mean_token_accuracy": 0.13931943476200104,
"num_tokens": 11516614.0,
"step": 5320
},
{
"entropy": 6.575680208206177,
"epoch": 0.5698539247685804,
"grad_norm": 0.87109375,
"learning_rate": 0.000497574942287828,
"loss": 6.4265,
"mean_token_accuracy": 0.1257178418338299,
"num_tokens": 11528032.0,
"step": 5325
},
{
"entropy": 6.551600313186645,
"epoch": 0.5703889988763444,
"grad_norm": 0.9453125,
"learning_rate": 0.0004975693408041697,
"loss": 6.4361,
"mean_token_accuracy": 0.12763728350400924,
"num_tokens": 11538822.0,
"step": 5330
},
{
"entropy": 6.6312580585479735,
"epoch": 0.5709240729841083,
"grad_norm": 0.9140625,
"learning_rate": 0.0004975637328938529,
"loss": 6.4506,
"mean_token_accuracy": 0.12766908183693887,
"num_tokens": 11548895.0,
"step": 5335
},
{
"entropy": 6.579705810546875,
"epoch": 0.5714591470918722,
"grad_norm": 0.87109375,
"learning_rate": 0.0004975581185570398,
"loss": 6.3573,
"mean_token_accuracy": 0.128676937520504,
"num_tokens": 11559613.0,
"step": 5340
},
{
"entropy": 6.55549578666687,
"epoch": 0.5719942211996362,
"grad_norm": 0.90234375,
"learning_rate": 0.0004975524977938921,
"loss": 6.3779,
"mean_token_accuracy": 0.1374692440032959,
"num_tokens": 11571092.0,
"step": 5345
},
{
"entropy": 6.607112216949463,
"epoch": 0.5725292953074,
"grad_norm": 0.9296875,
"learning_rate": 0.0004975468706045722,
"loss": 6.4942,
"mean_token_accuracy": 0.1302151046693325,
"num_tokens": 11581448.0,
"step": 5350
},
{
"entropy": 6.614511585235595,
"epoch": 0.573064369415164,
"grad_norm": 0.87890625,
"learning_rate": 0.0004975412369892429,
"loss": 6.4029,
"mean_token_accuracy": 0.13683656677603723,
"num_tokens": 11592273.0,
"step": 5355
},
{
"entropy": 6.586702966690064,
"epoch": 0.5735994435229279,
"grad_norm": 0.9453125,
"learning_rate": 0.0004975355969480665,
"loss": 6.3903,
"mean_token_accuracy": 0.1379762977361679,
"num_tokens": 11602759.0,
"step": 5360
},
{
"entropy": 6.574518823623658,
"epoch": 0.5741345176306919,
"grad_norm": 0.80859375,
"learning_rate": 0.0004975299504812061,
"loss": 6.4258,
"mean_token_accuracy": 0.12733170837163926,
"num_tokens": 11613977.0,
"step": 5365
},
{
"entropy": 6.657155513763428,
"epoch": 0.5746695917384558,
"grad_norm": 0.80078125,
"learning_rate": 0.0004975242975888246,
"loss": 6.4517,
"mean_token_accuracy": 0.13247386440634729,
"num_tokens": 11625425.0,
"step": 5370
},
{
"entropy": 6.603577184677124,
"epoch": 0.5752046658462197,
"grad_norm": 0.88671875,
"learning_rate": 0.0004975186382710854,
"loss": 6.4175,
"mean_token_accuracy": 0.13235969319939614,
"num_tokens": 11636530.0,
"step": 5375
},
{
"entropy": 6.6505959033966064,
"epoch": 0.5757397399539836,
"grad_norm": 0.94140625,
"learning_rate": 0.0004975129725281516,
"loss": 6.418,
"mean_token_accuracy": 0.1344207614660263,
"num_tokens": 11646516.0,
"step": 5380
},
{
"entropy": 6.549999094009399,
"epoch": 0.5762748140617475,
"grad_norm": 0.87109375,
"learning_rate": 0.0004975073003601871,
"loss": 6.3904,
"mean_token_accuracy": 0.13115372210741044,
"num_tokens": 11657424.0,
"step": 5385
},
{
"entropy": 6.580683898925781,
"epoch": 0.5768098881695115,
"grad_norm": 0.8359375,
"learning_rate": 0.0004975016217673556,
"loss": 6.4263,
"mean_token_accuracy": 0.1307824395596981,
"num_tokens": 11669447.0,
"step": 5390
},
{
"entropy": 6.633586883544922,
"epoch": 0.5773449622772754,
"grad_norm": 0.90625,
"learning_rate": 0.000497495936749821,
"loss": 6.3755,
"mean_token_accuracy": 0.14045795276761056,
"num_tokens": 11679612.0,
"step": 5395
},
{
"entropy": 6.564531373977661,
"epoch": 0.5778800363850394,
"grad_norm": 0.91015625,
"learning_rate": 0.0004974902453077474,
"loss": 6.3665,
"mean_token_accuracy": 0.140810589492321,
"num_tokens": 11689527.0,
"step": 5400
},
{
"entropy": 6.607308626174927,
"epoch": 0.5784151104928033,
"grad_norm": 0.81640625,
"learning_rate": 0.0004974845474412993,
"loss": 6.4125,
"mean_token_accuracy": 0.13287024796009064,
"num_tokens": 11700452.0,
"step": 5405
},
{
"entropy": 6.539466714859008,
"epoch": 0.5789501846005671,
"grad_norm": 0.9375,
"learning_rate": 0.0004974788431506412,
"loss": 6.4179,
"mean_token_accuracy": 0.13111766055226326,
"num_tokens": 11710902.0,
"step": 5410
},
{
"entropy": 6.580818510055542,
"epoch": 0.5794852587083311,
"grad_norm": 0.80078125,
"learning_rate": 0.0004974731324359378,
"loss": 6.4516,
"mean_token_accuracy": 0.12524218559265138,
"num_tokens": 11721901.0,
"step": 5415
},
{
"entropy": 6.574299621582031,
"epoch": 0.580020332816095,
"grad_norm": 0.859375,
"learning_rate": 0.0004974674152973538,
"loss": 6.436,
"mean_token_accuracy": 0.12758637517690657,
"num_tokens": 11733622.0,
"step": 5420
},
{
"entropy": 6.645131921768188,
"epoch": 0.580555406923859,
"grad_norm": 0.84375,
"learning_rate": 0.0004974616917350545,
"loss": 6.4384,
"mean_token_accuracy": 0.13184043169021606,
"num_tokens": 11745116.0,
"step": 5425
},
{
"entropy": 6.514034986495972,
"epoch": 0.5810904810316229,
"grad_norm": 0.87109375,
"learning_rate": 0.0004974559617492052,
"loss": 6.3601,
"mean_token_accuracy": 0.14398131519556046,
"num_tokens": 11756203.0,
"step": 5430
},
{
"entropy": 6.586912155151367,
"epoch": 0.5816255551393869,
"grad_norm": 0.93359375,
"learning_rate": 0.0004974502253399712,
"loss": 6.4155,
"mean_token_accuracy": 0.13222005367279052,
"num_tokens": 11766797.0,
"step": 5435
},
{
"entropy": 6.65447473526001,
"epoch": 0.5821606292471507,
"grad_norm": 1.03125,
"learning_rate": 0.0004974444825075182,
"loss": 6.5038,
"mean_token_accuracy": 0.12940247878432273,
"num_tokens": 11778110.0,
"step": 5440
},
{
"entropy": 6.526040124893188,
"epoch": 0.5826957033549146,
"grad_norm": 0.921875,
"learning_rate": 0.000497438733252012,
"loss": 6.3593,
"mean_token_accuracy": 0.13486856520175933,
"num_tokens": 11788396.0,
"step": 5445
},
{
"entropy": 6.630442190170288,
"epoch": 0.5832307774626786,
"grad_norm": 0.859375,
"learning_rate": 0.0004974329775736188,
"loss": 6.4293,
"mean_token_accuracy": 0.1297837421298027,
"num_tokens": 11798691.0,
"step": 5450
},
{
"entropy": 6.666377258300781,
"epoch": 0.5837658515704425,
"grad_norm": 1.015625,
"learning_rate": 0.0004974272154725044,
"loss": 6.496,
"mean_token_accuracy": 0.1329609051346779,
"num_tokens": 11808965.0,
"step": 5455
},
{
"entropy": 6.572327184677124,
"epoch": 0.5843009256782065,
"grad_norm": 0.9453125,
"learning_rate": 0.0004974214469488355,
"loss": 6.3343,
"mean_token_accuracy": 0.13848894461989403,
"num_tokens": 11819932.0,
"step": 5460
},
{
"entropy": 6.626104116439819,
"epoch": 0.5848359997859703,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974156720027786,
"loss": 6.426,
"mean_token_accuracy": 0.1278026707470417,
"num_tokens": 11831332.0,
"step": 5465
},
{
"entropy": 6.579942083358764,
"epoch": 0.5853710738937343,
"grad_norm": 0.859375,
"learning_rate": 0.0004974098906345004,
"loss": 6.4187,
"mean_token_accuracy": 0.1270743988454342,
"num_tokens": 11840947.0,
"step": 5470
},
{
"entropy": 6.554082345962525,
"epoch": 0.5859061480014982,
"grad_norm": 0.8828125,
"learning_rate": 0.0004974041028441679,
"loss": 6.3534,
"mean_token_accuracy": 0.13192323818802834,
"num_tokens": 11851640.0,
"step": 5475
},
{
"entropy": 6.650488948822021,
"epoch": 0.5864412221092621,
"grad_norm": 0.83203125,
"learning_rate": 0.0004973983086319481,
"loss": 6.4799,
"mean_token_accuracy": 0.12282219007611275,
"num_tokens": 11863608.0,
"step": 5480
},
{
"entropy": 6.599178647994995,
"epoch": 0.5869762962170261,
"grad_norm": 0.84765625,
"learning_rate": 0.0004973925079980085,
"loss": 6.3844,
"mean_token_accuracy": 0.1334873415529728,
"num_tokens": 11874707.0,
"step": 5485
},
{
"entropy": 6.564403486251831,
"epoch": 0.58751137032479,
"grad_norm": 0.9140625,
"learning_rate": 0.0004973867009425164,
"loss": 6.4576,
"mean_token_accuracy": 0.13435769751667975,
"num_tokens": 11885118.0,
"step": 5490
},
{
"entropy": 6.636895275115966,
"epoch": 0.5880464444325539,
"grad_norm": 0.93359375,
"learning_rate": 0.0004973808874656396,
"loss": 6.433,
"mean_token_accuracy": 0.1338953197002411,
"num_tokens": 11895859.0,
"step": 5495
},
{
"entropy": 6.667998456954956,
"epoch": 0.5885815185403178,
"grad_norm": 0.890625,
"learning_rate": 0.0004973750675675458,
"loss": 6.4449,
"mean_token_accuracy": 0.1335778370499611,
"num_tokens": 11907019.0,
"step": 5500
},
{
"entropy": 6.66331148147583,
"epoch": 0.5891165926480818,
"grad_norm": 0.89453125,
"learning_rate": 0.0004973692412484034,
"loss": 6.3889,
"mean_token_accuracy": 0.13348127380013466,
"num_tokens": 11918257.0,
"step": 5505
},
{
"entropy": 6.5581518650054935,
"epoch": 0.5896516667558457,
"grad_norm": 0.8828125,
"learning_rate": 0.0004973634085083802,
"loss": 6.3977,
"mean_token_accuracy": 0.14219407886266708,
"num_tokens": 11929003.0,
"step": 5510
},
{
"entropy": 6.622295188903808,
"epoch": 0.5901867408636096,
"grad_norm": 0.9453125,
"learning_rate": 0.000497357569347645,
"loss": 6.5309,
"mean_token_accuracy": 0.1261256143450737,
"num_tokens": 11939786.0,
"step": 5515
},
{
"entropy": 6.6540138721466064,
"epoch": 0.5907218149713735,
"grad_norm": 0.84765625,
"learning_rate": 0.0004973517237663661,
"loss": 6.4607,
"mean_token_accuracy": 0.1294330932199955,
"num_tokens": 11951042.0,
"step": 5520
},
{
"entropy": 6.569264698028564,
"epoch": 0.5912568890791374,
"grad_norm": 0.94140625,
"learning_rate": 0.0004973458717647124,
"loss": 6.3549,
"mean_token_accuracy": 0.13981595113873482,
"num_tokens": 11961982.0,
"step": 5525
},
{
"entropy": 6.532705926895142,
"epoch": 0.5917919631869014,
"grad_norm": 0.90234375,
"learning_rate": 0.000497340013342853,
"loss": 6.4091,
"mean_token_accuracy": 0.13178442418575287,
"num_tokens": 11972078.0,
"step": 5530
},
{
"entropy": 6.568721675872803,
"epoch": 0.5923270372946653,
"grad_norm": 0.921875,
"learning_rate": 0.000497334148500957,
"loss": 6.4584,
"mean_token_accuracy": 0.13278395012021066,
"num_tokens": 11984037.0,
"step": 5535
},
{
"entropy": 6.6697516441345215,
"epoch": 0.5928621114024293,
"grad_norm": 0.90625,
"learning_rate": 0.0004973282772391936,
"loss": 6.394,
"mean_token_accuracy": 0.13035947531461717,
"num_tokens": 11993688.0,
"step": 5540
},
{
"entropy": 6.565344905853271,
"epoch": 0.5933971855101932,
"grad_norm": 0.99609375,
"learning_rate": 0.0004973223995577324,
"loss": 6.3777,
"mean_token_accuracy": 0.13061075434088706,
"num_tokens": 12003759.0,
"step": 5545
},
{
"entropy": 6.531543827056884,
"epoch": 0.593932259617957,
"grad_norm": 0.8671875,
"learning_rate": 0.0004973165154567432,
"loss": 6.3875,
"mean_token_accuracy": 0.13072411194443703,
"num_tokens": 12015030.0,
"step": 5550
},
{
"entropy": 6.590817785263061,
"epoch": 0.594467333725721,
"grad_norm": 0.9140625,
"learning_rate": 0.0004973106249363959,
"loss": 6.4772,
"mean_token_accuracy": 0.13105777204036712,
"num_tokens": 12026181.0,
"step": 5555
},
{
"entropy": 6.541166162490844,
"epoch": 0.5950024078334849,
"grad_norm": 0.90234375,
"learning_rate": 0.0004973047279968606,
"loss": 6.2499,
"mean_token_accuracy": 0.1466530330479145,
"num_tokens": 12037336.0,
"step": 5560
},
{
"entropy": 6.567782735824585,
"epoch": 0.5955374819412489,
"grad_norm": 0.85546875,
"learning_rate": 0.0004972988246383074,
"loss": 6.3037,
"mean_token_accuracy": 0.13920291215181352,
"num_tokens": 12048425.0,
"step": 5565
},
{
"entropy": 6.436773157119751,
"epoch": 0.5960725560490128,
"grad_norm": 0.91796875,
"learning_rate": 0.0004972929148609068,
"loss": 6.2698,
"mean_token_accuracy": 0.15105066001415252,
"num_tokens": 12059814.0,
"step": 5570
},
{
"entropy": 6.525291872024536,
"epoch": 0.5966076301567768,
"grad_norm": 0.93359375,
"learning_rate": 0.0004972869986648296,
"loss": 6.3441,
"mean_token_accuracy": 0.138674733042717,
"num_tokens": 12069377.0,
"step": 5575
},
{
"entropy": 6.570910167694092,
"epoch": 0.5971427042645406,
"grad_norm": 0.94140625,
"learning_rate": 0.0004972810760502467,
"loss": 6.4875,
"mean_token_accuracy": 0.12600740566849708,
"num_tokens": 12080277.0,
"step": 5580
},
{
"entropy": 6.599872684478759,
"epoch": 0.5976777783723045,
"grad_norm": 0.890625,
"learning_rate": 0.0004972751470173287,
"loss": 6.3497,
"mean_token_accuracy": 0.13464435711503028,
"num_tokens": 12091375.0,
"step": 5585
},
{
"entropy": 6.550978755950927,
"epoch": 0.5982128524800685,
"grad_norm": 0.984375,
"learning_rate": 0.0004972692115662473,
"loss": 6.4211,
"mean_token_accuracy": 0.13656509220600127,
"num_tokens": 12103110.0,
"step": 5590
},
{
"entropy": 6.599859666824341,
"epoch": 0.5987479265878324,
"grad_norm": 0.99609375,
"learning_rate": 0.0004972632696971735,
"loss": 6.4603,
"mean_token_accuracy": 0.13045730218291282,
"num_tokens": 12112454.0,
"step": 5595
},
{
"entropy": 6.55696702003479,
"epoch": 0.5992830006955964,
"grad_norm": 0.91796875,
"learning_rate": 0.0004972573214102791,
"loss": 6.3395,
"mean_token_accuracy": 0.1349827118217945,
"num_tokens": 12125140.0,
"step": 5600
},
{
"entropy": 6.5705962657928465,
"epoch": 0.5998180748033602,
"grad_norm": 0.91796875,
"learning_rate": 0.0004972513667057358,
"loss": 6.3656,
"mean_token_accuracy": 0.13562314063310624,
"num_tokens": 12136061.0,
"step": 5605
},
{
"entropy": 6.650278472900391,
"epoch": 0.6003531489111242,
"grad_norm": 0.890625,
"learning_rate": 0.0004972454055837155,
"loss": 6.404,
"mean_token_accuracy": 0.13498041927814483,
"num_tokens": 12146654.0,
"step": 5610
},
{
"entropy": 6.614070034027099,
"epoch": 0.6008882230188881,
"grad_norm": 0.953125,
"learning_rate": 0.0004972394380443903,
"loss": 6.4097,
"mean_token_accuracy": 0.1308725282549858,
"num_tokens": 12157881.0,
"step": 5615
},
{
"entropy": 6.578127241134643,
"epoch": 0.601423297126652,
"grad_norm": 0.90625,
"learning_rate": 0.0004972334640879325,
"loss": 6.4287,
"mean_token_accuracy": 0.13418816700577735,
"num_tokens": 12168534.0,
"step": 5620
},
{
"entropy": 6.554200172424316,
"epoch": 0.601958371234416,
"grad_norm": 0.859375,
"learning_rate": 0.0004972274837145147,
"loss": 6.3578,
"mean_token_accuracy": 0.13448846638202666,
"num_tokens": 12178773.0,
"step": 5625
},
{
"entropy": 6.577337265014648,
"epoch": 0.6024934453421799,
"grad_norm": 0.85546875,
"learning_rate": 0.0004972214969243096,
"loss": 6.3879,
"mean_token_accuracy": 0.13618157282471657,
"num_tokens": 12188819.0,
"step": 5630
},
{
"entropy": 6.515968942642212,
"epoch": 0.6030285194499438,
"grad_norm": 0.86328125,
"learning_rate": 0.00049721550371749,
"loss": 6.375,
"mean_token_accuracy": 0.138833886384964,
"num_tokens": 12199198.0,
"step": 5635
},
{
"entropy": 6.569277572631836,
"epoch": 0.6035635935577077,
"grad_norm": 0.7890625,
"learning_rate": 0.0004972095040942289,
"loss": 6.4012,
"mean_token_accuracy": 0.1318491853773594,
"num_tokens": 12210765.0,
"step": 5640
},
{
"entropy": 6.580835676193237,
"epoch": 0.6040986676654717,
"grad_norm": 0.8828125,
"learning_rate": 0.0004972034980546996,
"loss": 6.4207,
"mean_token_accuracy": 0.128849096596241,
"num_tokens": 12221067.0,
"step": 5645
},
{
"entropy": 6.530771636962891,
"epoch": 0.6046337417732356,
"grad_norm": 0.8984375,
"learning_rate": 0.0004971974855990755,
"loss": 6.4017,
"mean_token_accuracy": 0.13618824183940886,
"num_tokens": 12232994.0,
"step": 5650
},
{
"entropy": 6.593167781829834,
"epoch": 0.6051688158809995,
"grad_norm": 0.8984375,
"learning_rate": 0.0004971914667275302,
"loss": 6.3796,
"mean_token_accuracy": 0.13560357168316842,
"num_tokens": 12243327.0,
"step": 5655
},
{
"entropy": 6.473536396026612,
"epoch": 0.6057038899887635,
"grad_norm": 0.890625,
"learning_rate": 0.0004971854414402377,
"loss": 6.3126,
"mean_token_accuracy": 0.13819175511598586,
"num_tokens": 12254249.0,
"step": 5660
},
{
"entropy": 6.540711259841919,
"epoch": 0.6062389640965273,
"grad_norm": 0.86328125,
"learning_rate": 0.0004971794097373716,
"loss": 6.3412,
"mean_token_accuracy": 0.13511233404278755,
"num_tokens": 12264487.0,
"step": 5665
},
{
"entropy": 6.494623517990112,
"epoch": 0.6067740382042913,
"grad_norm": 0.91796875,
"learning_rate": 0.0004971733716191063,
"loss": 6.347,
"mean_token_accuracy": 0.1392446607351303,
"num_tokens": 12274801.0,
"step": 5670
},
{
"entropy": 6.561354398727417,
"epoch": 0.6073091123120552,
"grad_norm": 0.96484375,
"learning_rate": 0.0004971673270856162,
"loss": 6.3022,
"mean_token_accuracy": 0.13149403035640717,
"num_tokens": 12284647.0,
"step": 5675
},
{
"entropy": 6.544518184661865,
"epoch": 0.6078441864198192,
"grad_norm": 0.9375,
"learning_rate": 0.0004971612761370756,
"loss": 6.4221,
"mean_token_accuracy": 0.13287723660469056,
"num_tokens": 12295215.0,
"step": 5680
},
{
"entropy": 6.517327117919922,
"epoch": 0.6083792605275831,
"grad_norm": 0.90625,
"learning_rate": 0.0004971552187736596,
"loss": 6.339,
"mean_token_accuracy": 0.1339027963578701,
"num_tokens": 12305324.0,
"step": 5685
},
{
"entropy": 6.5198009490966795,
"epoch": 0.6089143346353469,
"grad_norm": 0.921875,
"learning_rate": 0.0004971491549955427,
"loss": 6.3515,
"mean_token_accuracy": 0.1382587730884552,
"num_tokens": 12316399.0,
"step": 5690
},
{
"entropy": 6.480385828018188,
"epoch": 0.6094494087431109,
"grad_norm": 0.93359375,
"learning_rate": 0.0004971430848029002,
"loss": 6.3333,
"mean_token_accuracy": 0.1392750360071659,
"num_tokens": 12328033.0,
"step": 5695
},
{
"entropy": 6.518033885955811,
"epoch": 0.6099844828508748,
"grad_norm": 0.87890625,
"learning_rate": 0.0004971370081959073,
"loss": 6.3701,
"mean_token_accuracy": 0.1361820012331009,
"num_tokens": 12338962.0,
"step": 5700
},
{
"entropy": 6.588574457168579,
"epoch": 0.6105195569586388,
"grad_norm": 0.8203125,
"learning_rate": 0.0004971309251747396,
"loss": 6.3557,
"mean_token_accuracy": 0.1340286947786808,
"num_tokens": 12350216.0,
"step": 5705
},
{
"entropy": 6.601768064498901,
"epoch": 0.6110546310664027,
"grad_norm": 0.92578125,
"learning_rate": 0.0004971248357395726,
"loss": 6.3923,
"mean_token_accuracy": 0.13332446068525314,
"num_tokens": 12361277.0,
"step": 5710
},
{
"entropy": 6.487619066238404,
"epoch": 0.6115897051741667,
"grad_norm": 0.859375,
"learning_rate": 0.0004971187398905821,
"loss": 6.3086,
"mean_token_accuracy": 0.13468469381332399,
"num_tokens": 12373308.0,
"step": 5715
},
{
"entropy": 6.604460716247559,
"epoch": 0.6121247792819305,
"grad_norm": 0.8828125,
"learning_rate": 0.0004971126376279443,
"loss": 6.5461,
"mean_token_accuracy": 0.12380608022212983,
"num_tokens": 12383943.0,
"step": 5720
},
{
"entropy": 6.578613996505737,
"epoch": 0.6126598533896944,
"grad_norm": 0.90234375,
"learning_rate": 0.0004971065289518354,
"loss": 6.3125,
"mean_token_accuracy": 0.13367188200354577,
"num_tokens": 12394575.0,
"step": 5725
},
{
"entropy": 6.603823232650757,
"epoch": 0.6131949274974584,
"grad_norm": 0.82421875,
"learning_rate": 0.0004971004138624315,
"loss": 6.3981,
"mean_token_accuracy": 0.13316768556833267,
"num_tokens": 12405536.0,
"step": 5730
},
{
"entropy": 6.497972536087036,
"epoch": 0.6137300016052223,
"grad_norm": 0.8125,
"learning_rate": 0.0004970942923599094,
"loss": 6.3005,
"mean_token_accuracy": 0.14006469771265984,
"num_tokens": 12416148.0,
"step": 5735
},
{
"entropy": 6.541106081008911,
"epoch": 0.6142650757129863,
"grad_norm": 0.98046875,
"learning_rate": 0.0004970881644444459,
"loss": 6.3076,
"mean_token_accuracy": 0.13991214632987975,
"num_tokens": 12426182.0,
"step": 5740
},
{
"entropy": 6.590402269363404,
"epoch": 0.6148001498207502,
"grad_norm": 0.86328125,
"learning_rate": 0.0004970820301162178,
"loss": 6.4833,
"mean_token_accuracy": 0.1292591094970703,
"num_tokens": 12437548.0,
"step": 5745
},
{
"entropy": 6.592732715606689,
"epoch": 0.6153352239285141,
"grad_norm": 0.83203125,
"learning_rate": 0.0004970758893754022,
"loss": 6.4092,
"mean_token_accuracy": 0.12948114275932313,
"num_tokens": 12449720.0,
"step": 5750
},
{
"entropy": 6.630194807052613,
"epoch": 0.615870298036278,
"grad_norm": 0.85546875,
"learning_rate": 0.0004970697422221767,
"loss": 6.3971,
"mean_token_accuracy": 0.13294799551367759,
"num_tokens": 12460666.0,
"step": 5755
},
{
"entropy": 6.558688449859619,
"epoch": 0.6164053721440419,
"grad_norm": 0.90625,
"learning_rate": 0.0004970635886567185,
"loss": 6.3956,
"mean_token_accuracy": 0.1334315963089466,
"num_tokens": 12471472.0,
"step": 5760
},
{
"entropy": 6.496923494338989,
"epoch": 0.6169404462518059,
"grad_norm": 0.890625,
"learning_rate": 0.0004970574286792054,
"loss": 6.3846,
"mean_token_accuracy": 0.13120403438806533,
"num_tokens": 12482127.0,
"step": 5765
},
{
"entropy": 6.568459987640381,
"epoch": 0.6174755203595698,
"grad_norm": 0.8359375,
"learning_rate": 0.0004970512622898152,
"loss": 6.3757,
"mean_token_accuracy": 0.13721705228090286,
"num_tokens": 12493680.0,
"step": 5770
},
{
"entropy": 6.528800058364868,
"epoch": 0.6180105944673338,
"grad_norm": 0.98828125,
"learning_rate": 0.000497045089488726,
"loss": 6.3354,
"mean_token_accuracy": 0.13322472646832467,
"num_tokens": 12505838.0,
"step": 5775
},
{
"entropy": 6.586766910552979,
"epoch": 0.6185456685750976,
"grad_norm": 0.94921875,
"learning_rate": 0.0004970389102761161,
"loss": 6.3525,
"mean_token_accuracy": 0.13949506655335425,
"num_tokens": 12516334.0,
"step": 5780
},
{
"entropy": 6.524970054626465,
"epoch": 0.6190807426828616,
"grad_norm": 1.0859375,
"learning_rate": 0.0004970327246521638,
"loss": 6.3963,
"mean_token_accuracy": 0.1295120485126972,
"num_tokens": 12526440.0,
"step": 5785
},
{
"entropy": 6.550500583648682,
"epoch": 0.6196158167906255,
"grad_norm": 0.84375,
"learning_rate": 0.0004970265326170478,
"loss": 6.3384,
"mean_token_accuracy": 0.12922370806336403,
"num_tokens": 12538127.0,
"step": 5790
},
{
"entropy": 6.614361572265625,
"epoch": 0.6201508908983894,
"grad_norm": 0.96875,
"learning_rate": 0.0004970203341709469,
"loss": 6.4204,
"mean_token_accuracy": 0.13372067511081695,
"num_tokens": 12548459.0,
"step": 5795
},
{
"entropy": 6.504186201095581,
"epoch": 0.6206859650061534,
"grad_norm": 0.94140625,
"learning_rate": 0.00049701412931404,
"loss": 6.3551,
"mean_token_accuracy": 0.13731214702129363,
"num_tokens": 12558989.0,
"step": 5800
},
{
"entropy": 6.5102873802185055,
"epoch": 0.6212210391139172,
"grad_norm": 0.875,
"learning_rate": 0.0004970079180465064,
"loss": 6.2834,
"mean_token_accuracy": 0.14180205017328262,
"num_tokens": 12569449.0,
"step": 5805
},
{
"entropy": 6.547976398468018,
"epoch": 0.6217561132216812,
"grad_norm": 0.8671875,
"learning_rate": 0.0004970017003685253,
"loss": 6.3543,
"mean_token_accuracy": 0.1336100295186043,
"num_tokens": 12580733.0,
"step": 5810
},
{
"entropy": 6.552909994125367,
"epoch": 0.6222911873294451,
"grad_norm": 0.88671875,
"learning_rate": 0.0004969954762802763,
"loss": 6.3611,
"mean_token_accuracy": 0.13334628865122794,
"num_tokens": 12590869.0,
"step": 5815
},
{
"entropy": 6.600517606735229,
"epoch": 0.6228262614372091,
"grad_norm": 0.9765625,
"learning_rate": 0.000496989245781939,
"loss": 6.4181,
"mean_token_accuracy": 0.13195589184761047,
"num_tokens": 12602109.0,
"step": 5820
},
{
"entropy": 6.524480295181275,
"epoch": 0.623361335544973,
"grad_norm": 0.87890625,
"learning_rate": 0.0004969830088736937,
"loss": 6.3577,
"mean_token_accuracy": 0.13163684085011482,
"num_tokens": 12612234.0,
"step": 5825
},
{
"entropy": 6.640629863739013,
"epoch": 0.6238964096527368,
"grad_norm": 0.890625,
"learning_rate": 0.00049697676555572,
"loss": 6.5171,
"mean_token_accuracy": 0.13004111051559447,
"num_tokens": 12623252.0,
"step": 5830
},
{
"entropy": 6.637633466720581,
"epoch": 0.6244314837605008,
"grad_norm": 0.875,
"learning_rate": 0.0004969705158281985,
"loss": 6.2591,
"mean_token_accuracy": 0.1430523671209812,
"num_tokens": 12632796.0,
"step": 5835
},
{
"entropy": 6.52786078453064,
"epoch": 0.6249665578682647,
"grad_norm": 0.9609375,
"learning_rate": 0.0004969642596913095,
"loss": 6.3471,
"mean_token_accuracy": 0.1345833159983158,
"num_tokens": 12643278.0,
"step": 5840
},
{
"entropy": 6.538140487670899,
"epoch": 0.6255016319760287,
"grad_norm": 0.95703125,
"learning_rate": 0.0004969579971452337,
"loss": 6.4262,
"mean_token_accuracy": 0.13390445560216904,
"num_tokens": 12654446.0,
"step": 5845
},
{
"entropy": 6.564716768264771,
"epoch": 0.6260367060837926,
"grad_norm": 0.91015625,
"learning_rate": 0.0004969517281901519,
"loss": 6.4057,
"mean_token_accuracy": 0.13656433820724487,
"num_tokens": 12664281.0,
"step": 5850
},
{
"entropy": 6.584988307952881,
"epoch": 0.6265717801915566,
"grad_norm": 1.0078125,
"learning_rate": 0.0004969454528262451,
"loss": 6.3973,
"mean_token_accuracy": 0.13601098954677582,
"num_tokens": 12675252.0,
"step": 5855
},
{
"entropy": 6.534916353225708,
"epoch": 0.6271068542993204,
"grad_norm": 0.90234375,
"learning_rate": 0.0004969391710536946,
"loss": 6.3701,
"mean_token_accuracy": 0.13320676386356353,
"num_tokens": 12685764.0,
"step": 5860
},
{
"entropy": 6.454722547531128,
"epoch": 0.6276419284070843,
"grad_norm": 0.98046875,
"learning_rate": 0.0004969328828726817,
"loss": 6.2897,
"mean_token_accuracy": 0.13955534845590592,
"num_tokens": 12696974.0,
"step": 5865
},
{
"entropy": 6.6120130062103275,
"epoch": 0.6281770025148483,
"grad_norm": 0.9609375,
"learning_rate": 0.0004969265882833879,
"loss": 6.452,
"mean_token_accuracy": 0.12457979917526245,
"num_tokens": 12708217.0,
"step": 5870
},
{
"entropy": 6.594373893737793,
"epoch": 0.6287120766226122,
"grad_norm": 0.90625,
"learning_rate": 0.0004969202872859952,
"loss": 6.3037,
"mean_token_accuracy": 0.13776595368981362,
"num_tokens": 12717840.0,
"step": 5875
},
{
"entropy": 6.483970737457275,
"epoch": 0.6292471507303762,
"grad_norm": 0.91796875,
"learning_rate": 0.0004969139798806854,
"loss": 6.3973,
"mean_token_accuracy": 0.1298674687743187,
"num_tokens": 12728390.0,
"step": 5880
},
{
"entropy": 6.540892505645752,
"epoch": 0.6297822248381401,
"grad_norm": 0.97265625,
"learning_rate": 0.0004969076660676405,
"loss": 6.4003,
"mean_token_accuracy": 0.13340216055512427,
"num_tokens": 12738650.0,
"step": 5885
},
{
"entropy": 6.595161819458008,
"epoch": 0.630317298945904,
"grad_norm": 0.97265625,
"learning_rate": 0.000496901345847043,
"loss": 6.4032,
"mean_token_accuracy": 0.13205326199531556,
"num_tokens": 12748989.0,
"step": 5890
},
{
"entropy": 6.5864780902862545,
"epoch": 0.6308523730536679,
"grad_norm": 0.8671875,
"learning_rate": 0.0004968950192190752,
"loss": 6.3649,
"mean_token_accuracy": 0.1320453979074955,
"num_tokens": 12761132.0,
"step": 5895
},
{
"entropy": 6.517518615722656,
"epoch": 0.6313874471614318,
"grad_norm": 0.890625,
"learning_rate": 0.00049688868618392,
"loss": 6.3788,
"mean_token_accuracy": 0.13421476930379866,
"num_tokens": 12771478.0,
"step": 5900
},
{
"entropy": 6.581002044677734,
"epoch": 0.6319225212691958,
"grad_norm": 0.86328125,
"learning_rate": 0.0004968823467417602,
"loss": 6.4344,
"mean_token_accuracy": 0.13052433654665946,
"num_tokens": 12782096.0,
"step": 5905
},
{
"entropy": 6.563936185836792,
"epoch": 0.6324575953769597,
"grad_norm": 0.8828125,
"learning_rate": 0.0004968760008927788,
"loss": 6.3382,
"mean_token_accuracy": 0.13559918701648713,
"num_tokens": 12792467.0,
"step": 5910
},
{
"entropy": 6.576694536209106,
"epoch": 0.6329926694847237,
"grad_norm": 0.890625,
"learning_rate": 0.000496869648637159,
"loss": 6.3267,
"mean_token_accuracy": 0.14201204627752304,
"num_tokens": 12803248.0,
"step": 5915
},
{
"entropy": 6.562338829040527,
"epoch": 0.6335277435924875,
"grad_norm": 0.921875,
"learning_rate": 0.0004968632899750844,
"loss": 6.354,
"mean_token_accuracy": 0.13680187240242958,
"num_tokens": 12813701.0,
"step": 5920
},
{
"entropy": 6.6017234325408936,
"epoch": 0.6340628177002515,
"grad_norm": 0.921875,
"learning_rate": 0.0004968569249067383,
"loss": 6.4335,
"mean_token_accuracy": 0.12925372123718262,
"num_tokens": 12824945.0,
"step": 5925
},
{
"entropy": 6.566445016860962,
"epoch": 0.6345978918080154,
"grad_norm": 0.8203125,
"learning_rate": 0.0004968505534323047,
"loss": 6.3768,
"mean_token_accuracy": 0.13098957762122154,
"num_tokens": 12836699.0,
"step": 5930
},
{
"entropy": 6.5367591381073,
"epoch": 0.6351329659157794,
"grad_norm": 0.8359375,
"learning_rate": 0.0004968441755519677,
"loss": 6.3637,
"mean_token_accuracy": 0.131689964979887,
"num_tokens": 12846905.0,
"step": 5935
},
{
"entropy": 6.582563877105713,
"epoch": 0.6356680400235433,
"grad_norm": 0.9609375,
"learning_rate": 0.0004968377912659112,
"loss": 6.2806,
"mean_token_accuracy": 0.1423302873969078,
"num_tokens": 12856919.0,
"step": 5940
},
{
"entropy": 6.4202080249786375,
"epoch": 0.6362031141313071,
"grad_norm": 0.87890625,
"learning_rate": 0.0004968314005743196,
"loss": 6.272,
"mean_token_accuracy": 0.14061389490962029,
"num_tokens": 12867863.0,
"step": 5945
},
{
"entropy": 6.480769872665405,
"epoch": 0.6367381882390711,
"grad_norm": 0.87109375,
"learning_rate": 0.0004968250034773776,
"loss": 6.346,
"mean_token_accuracy": 0.1376643568277359,
"num_tokens": 12877877.0,
"step": 5950
},
{
"entropy": 6.656388235092163,
"epoch": 0.637273262346835,
"grad_norm": 0.984375,
"learning_rate": 0.0004968185999752697,
"loss": 6.4354,
"mean_token_accuracy": 0.12367877438664436,
"num_tokens": 12887954.0,
"step": 5955
},
{
"entropy": 6.554588699340821,
"epoch": 0.637808336454599,
"grad_norm": 0.98828125,
"learning_rate": 0.0004968121900681809,
"loss": 6.361,
"mean_token_accuracy": 0.13090814650058746,
"num_tokens": 12897744.0,
"step": 5960
},
{
"entropy": 6.55085768699646,
"epoch": 0.6383434105623629,
"grad_norm": 0.8671875,
"learning_rate": 0.0004968057737562964,
"loss": 6.4384,
"mean_token_accuracy": 0.13585126921534538,
"num_tokens": 12909193.0,
"step": 5965
},
{
"entropy": 6.55703182220459,
"epoch": 0.6388784846701269,
"grad_norm": 0.8828125,
"learning_rate": 0.0004967993510398012,
"loss": 6.3074,
"mean_token_accuracy": 0.13894035816192626,
"num_tokens": 12919879.0,
"step": 5970
},
{
"entropy": 6.53627986907959,
"epoch": 0.6394135587778907,
"grad_norm": 0.89453125,
"learning_rate": 0.0004967929219188809,
"loss": 6.367,
"mean_token_accuracy": 0.13364173769950866,
"num_tokens": 12930297.0,
"step": 5975
},
{
"entropy": 6.533883190155029,
"epoch": 0.6399486328856546,
"grad_norm": 0.94140625,
"learning_rate": 0.0004967864863937212,
"loss": 6.3185,
"mean_token_accuracy": 0.1434766449034214,
"num_tokens": 12940467.0,
"step": 5980
},
{
"entropy": 6.5478757381439205,
"epoch": 0.6404837069934186,
"grad_norm": 0.9453125,
"learning_rate": 0.0004967800444645079,
"loss": 6.3365,
"mean_token_accuracy": 0.13498707339167595,
"num_tokens": 12950417.0,
"step": 5985
},
{
"entropy": 6.495828294754029,
"epoch": 0.6410187811011825,
"grad_norm": 0.9453125,
"learning_rate": 0.0004967735961314269,
"loss": 6.29,
"mean_token_accuracy": 0.14207748472690582,
"num_tokens": 12960957.0,
"step": 5990
},
{
"entropy": 6.482624292373657,
"epoch": 0.6415538552089465,
"grad_norm": 0.8984375,
"learning_rate": 0.0004967671413946646,
"loss": 6.3971,
"mean_token_accuracy": 0.13469642549753189,
"num_tokens": 12971194.0,
"step": 5995
},
{
"entropy": 6.558772277832031,
"epoch": 0.6420889293167104,
"grad_norm": 1.0078125,
"learning_rate": 0.0004967606802544071,
"loss": 6.3775,
"mean_token_accuracy": 0.13109284415841102,
"num_tokens": 12982479.0,
"step": 6000
},
{
"epoch": 0.6420889293167104,
"eval_entropy": 6.383020276329267,
"eval_loss": 6.415825366973877,
"eval_mean_token_accuracy": 0.13898081381747504,
"eval_num_tokens": 12982479.0,
"eval_runtime": 17.5694,
"eval_samples_per_second": 1689.24,
"eval_steps_per_second": 211.162,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 93440,
"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": 2818236196454400.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink