fpadovani/urd-arab-10mb-100mb_seed10
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
urd-arab-10mb-100mb_seed10/checkpoint-9000/trainer_state.json
ModelHub XC d810e269f2 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/urd-arab-10mb-100mb_seed10
Source: Original Platform
2026-08-20 09:45:19 +08:00

18068 lines
495 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.072066706372841,
"eval_steps": 3000,
"global_step": 9000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.742183494567872,
"epoch": 0.0005955926146515784,
"grad_norm": 4.09375,
"learning_rate": 2e-06,
"loss": 10.8347,
"mean_token_accuracy": 6.570302066393197e-05,
"num_tokens": 14043.0,
"step": 5
},
{
"entropy": 10.742181587219239,
"epoch": 0.0011911852293031567,
"grad_norm": 4.625,
"learning_rate": 4.5e-06,
"loss": 10.8184,
"mean_token_accuracy": 5.6306307669728994e-05,
"num_tokens": 27431.0,
"step": 10
},
{
"entropy": 10.742227840423585,
"epoch": 0.001786777843954735,
"grad_norm": 4.625,
"learning_rate": 7e-06,
"loss": 10.792,
"mean_token_accuracy": 0.00015279132057912647,
"num_tokens": 41546.0,
"step": 15
},
{
"entropy": 10.742228794097901,
"epoch": 0.0023823704586063135,
"grad_norm": 4.15625,
"learning_rate": 9.5e-06,
"loss": 10.7654,
"mean_token_accuracy": 5.235602147877216e-05,
"num_tokens": 55610.0,
"step": 20
},
{
"entropy": 10.742142009735108,
"epoch": 0.0029779630732578916,
"grad_norm": 4.25,
"learning_rate": 1.2e-05,
"loss": 10.6461,
"mean_token_accuracy": 0.0007188191404566169,
"num_tokens": 67719.0,
"step": 25
},
{
"entropy": 10.741945934295654,
"epoch": 0.00357355568790947,
"grad_norm": 3.4375,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.5819,
"mean_token_accuracy": 0.0071188481757417325,
"num_tokens": 82309.0,
"step": 30
},
{
"entropy": 10.741309547424317,
"epoch": 0.004169148302561048,
"grad_norm": 2.921875,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.436,
"mean_token_accuracy": 0.025534892454743385,
"num_tokens": 95752.0,
"step": 35
},
{
"entropy": 10.739977741241455,
"epoch": 0.004764740917212627,
"grad_norm": 2.421875,
"learning_rate": 1.95e-05,
"loss": 10.3101,
"mean_token_accuracy": 0.027164995670318604,
"num_tokens": 109473.0,
"step": 40
},
{
"entropy": 10.73803424835205,
"epoch": 0.005360333531864205,
"grad_norm": 2.03125,
"learning_rate": 2.2e-05,
"loss": 10.2408,
"mean_token_accuracy": 0.03164031654596329,
"num_tokens": 124193.0,
"step": 45
},
{
"entropy": 10.736229801177979,
"epoch": 0.005955926146515783,
"grad_norm": 2.03125,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.1299,
"mean_token_accuracy": 0.033015350252389906,
"num_tokens": 136870.0,
"step": 50
},
{
"entropy": 10.7350643157959,
"epoch": 0.006551518761167362,
"grad_norm": 1.890625,
"learning_rate": 2.7e-05,
"loss": 10.0889,
"mean_token_accuracy": 0.035209237970411776,
"num_tokens": 150613.0,
"step": 55
},
{
"entropy": 10.733927249908447,
"epoch": 0.00714711137581894,
"grad_norm": 1.8125,
"learning_rate": 2.95e-05,
"loss": 10.0142,
"mean_token_accuracy": 0.03481624573469162,
"num_tokens": 164213.0,
"step": 60
},
{
"entropy": 10.732409286499024,
"epoch": 0.007742703990470518,
"grad_norm": 1.8515625,
"learning_rate": 3.2e-05,
"loss": 9.9266,
"mean_token_accuracy": 0.03644208796322346,
"num_tokens": 177408.0,
"step": 65
},
{
"entropy": 10.730414295196534,
"epoch": 0.008338296605122097,
"grad_norm": 1.8125,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.8609,
"mean_token_accuracy": 0.035931960307061674,
"num_tokens": 190273.0,
"step": 70
},
{
"entropy": 10.727591800689698,
"epoch": 0.008933889219773675,
"grad_norm": 1.796875,
"learning_rate": 3.7e-05,
"loss": 9.7942,
"mean_token_accuracy": 0.036602536030113694,
"num_tokens": 203323.0,
"step": 75
},
{
"entropy": 10.723168182373048,
"epoch": 0.009529481834425254,
"grad_norm": 1.734375,
"learning_rate": 3.95e-05,
"loss": 9.7562,
"mean_token_accuracy": 0.035238896124064924,
"num_tokens": 218647.0,
"step": 80
},
{
"entropy": 10.717996406555176,
"epoch": 0.01012507444907683,
"grad_norm": 1.703125,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.6793,
"mean_token_accuracy": 0.03231475837528706,
"num_tokens": 232910.0,
"step": 85
},
{
"entropy": 10.712448978424073,
"epoch": 0.01072066706372841,
"grad_norm": 1.6796875,
"learning_rate": 4.45e-05,
"loss": 9.5914,
"mean_token_accuracy": 0.034204628132283685,
"num_tokens": 246683.0,
"step": 90
},
{
"entropy": 10.703381633758545,
"epoch": 0.011316259678379988,
"grad_norm": 1.65625,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.5074,
"mean_token_accuracy": 0.0348433218896389,
"num_tokens": 260150.0,
"step": 95
},
{
"entropy": 10.688982582092285,
"epoch": 0.011911852293031567,
"grad_norm": 1.515625,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.431,
"mean_token_accuracy": 0.034749519638717176,
"num_tokens": 273788.0,
"step": 100
},
{
"entropy": 10.672309303283692,
"epoch": 0.012507444907683145,
"grad_norm": 1.6328125,
"learning_rate": 5.2e-05,
"loss": 9.3078,
"mean_token_accuracy": 0.03429920971393585,
"num_tokens": 287625.0,
"step": 105
},
{
"entropy": 10.650675106048585,
"epoch": 0.013103037522334724,
"grad_norm": 1.6796875,
"learning_rate": 5.45e-05,
"loss": 9.2221,
"mean_token_accuracy": 0.0347648523747921,
"num_tokens": 302938.0,
"step": 110
},
{
"entropy": 10.61922378540039,
"epoch": 0.0136986301369863,
"grad_norm": 1.578125,
"learning_rate": 5.7e-05,
"loss": 9.0864,
"mean_token_accuracy": 0.03500578422099352,
"num_tokens": 316905.0,
"step": 115
},
{
"entropy": 10.572606658935547,
"epoch": 0.01429422275163788,
"grad_norm": 1.6328125,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.924,
"mean_token_accuracy": 0.03559151701629162,
"num_tokens": 331131.0,
"step": 120
},
{
"entropy": 10.516113567352296,
"epoch": 0.014889815366289458,
"grad_norm": 1.546875,
"learning_rate": 6.2e-05,
"loss": 8.8249,
"mean_token_accuracy": 0.032901805639266965,
"num_tokens": 344995.0,
"step": 125
},
{
"entropy": 10.453802871704102,
"epoch": 0.015485407980941036,
"grad_norm": 1.5546875,
"learning_rate": 6.450000000000001e-05,
"loss": 8.6582,
"mean_token_accuracy": 0.03676526509225368,
"num_tokens": 361057.0,
"step": 130
},
{
"entropy": 10.370664405822755,
"epoch": 0.016081000595592615,
"grad_norm": 1.453125,
"learning_rate": 6.7e-05,
"loss": 8.5228,
"mean_token_accuracy": 0.03370327036827803,
"num_tokens": 375037.0,
"step": 135
},
{
"entropy": 10.266613578796386,
"epoch": 0.016676593210244194,
"grad_norm": 1.3828125,
"learning_rate": 6.950000000000001e-05,
"loss": 8.3958,
"mean_token_accuracy": 0.03396179284900427,
"num_tokens": 389797.0,
"step": 140
},
{
"entropy": 10.141527652740479,
"epoch": 0.017272185824895772,
"grad_norm": 1.234375,
"learning_rate": 7.2e-05,
"loss": 8.2779,
"mean_token_accuracy": 0.03149167206138372,
"num_tokens": 403743.0,
"step": 145
},
{
"entropy": 9.992219924926758,
"epoch": 0.01786777843954735,
"grad_norm": 1.2265625,
"learning_rate": 7.45e-05,
"loss": 8.1645,
"mean_token_accuracy": 0.030641537345945834,
"num_tokens": 417337.0,
"step": 150
},
{
"entropy": 9.80675220489502,
"epoch": 0.01846337105419893,
"grad_norm": 1.1796875,
"learning_rate": 7.7e-05,
"loss": 7.9755,
"mean_token_accuracy": 0.035594462603330615,
"num_tokens": 430106.0,
"step": 155
},
{
"entropy": 9.583033466339112,
"epoch": 0.019058963668850508,
"grad_norm": 1.03125,
"learning_rate": 7.950000000000001e-05,
"loss": 7.9146,
"mean_token_accuracy": 0.034621411561965944,
"num_tokens": 444654.0,
"step": 160
},
{
"entropy": 9.349032878875732,
"epoch": 0.019654556283502083,
"grad_norm": 0.90234375,
"learning_rate": 8.2e-05,
"loss": 7.849,
"mean_token_accuracy": 0.029788750410079955,
"num_tokens": 457837.0,
"step": 165
},
{
"entropy": 9.113014602661133,
"epoch": 0.02025014889815366,
"grad_norm": 1.0078125,
"learning_rate": 8.450000000000001e-05,
"loss": 7.6894,
"mean_token_accuracy": 0.03679767306894064,
"num_tokens": 472006.0,
"step": 170
},
{
"entropy": 8.845180225372314,
"epoch": 0.02084574151280524,
"grad_norm": 0.80859375,
"learning_rate": 8.7e-05,
"loss": 7.6327,
"mean_token_accuracy": 0.03697254229336977,
"num_tokens": 486155.0,
"step": 175
},
{
"entropy": 8.680907440185546,
"epoch": 0.02144133412745682,
"grad_norm": 0.71484375,
"learning_rate": 8.95e-05,
"loss": 7.6678,
"mean_token_accuracy": 0.041565592586994174,
"num_tokens": 501152.0,
"step": 180
},
{
"entropy": 8.531956195831299,
"epoch": 0.022036926742108397,
"grad_norm": 0.80859375,
"learning_rate": 9.2e-05,
"loss": 7.5695,
"mean_token_accuracy": 0.04552752338349819,
"num_tokens": 514856.0,
"step": 185
},
{
"entropy": 8.419017887115478,
"epoch": 0.022632519356759976,
"grad_norm": 0.87109375,
"learning_rate": 9.45e-05,
"loss": 7.5451,
"mean_token_accuracy": 0.04798501394689083,
"num_tokens": 529356.0,
"step": 190
},
{
"entropy": 8.345262050628662,
"epoch": 0.023228111971411555,
"grad_norm": 0.875,
"learning_rate": 9.7e-05,
"loss": 7.4986,
"mean_token_accuracy": 0.04905890300869942,
"num_tokens": 542876.0,
"step": 195
},
{
"entropy": 8.245842933654785,
"epoch": 0.023823704586063133,
"grad_norm": 0.83203125,
"learning_rate": 9.95e-05,
"loss": 7.468,
"mean_token_accuracy": 0.05098759308457375,
"num_tokens": 557245.0,
"step": 200
},
{
"entropy": 8.199878883361816,
"epoch": 0.024419297200714712,
"grad_norm": 0.8671875,
"learning_rate": 0.000102,
"loss": 7.3793,
"mean_token_accuracy": 0.055880747735500336,
"num_tokens": 571458.0,
"step": 205
},
{
"entropy": 8.144281673431397,
"epoch": 0.02501488981536629,
"grad_norm": 0.78515625,
"learning_rate": 0.00010449999999999999,
"loss": 7.3802,
"mean_token_accuracy": 0.054352150112390515,
"num_tokens": 588000.0,
"step": 210
},
{
"entropy": 8.136652183532714,
"epoch": 0.02561048243001787,
"grad_norm": 1.0625,
"learning_rate": 0.000107,
"loss": 7.4053,
"mean_token_accuracy": 0.06009378880262375,
"num_tokens": 602302.0,
"step": 215
},
{
"entropy": 8.131936168670654,
"epoch": 0.026206075044669448,
"grad_norm": 1.4453125,
"learning_rate": 0.0001095,
"loss": 7.4315,
"mean_token_accuracy": 0.05938838943839073,
"num_tokens": 616708.0,
"step": 220
},
{
"entropy": 8.018897008895873,
"epoch": 0.026801667659321026,
"grad_norm": 0.875,
"learning_rate": 0.000112,
"loss": 7.2856,
"mean_token_accuracy": 0.05857427380979061,
"num_tokens": 629236.0,
"step": 225
},
{
"entropy": 8.098274040222169,
"epoch": 0.0273972602739726,
"grad_norm": 0.89453125,
"learning_rate": 0.0001145,
"loss": 7.3467,
"mean_token_accuracy": 0.06362572610378266,
"num_tokens": 642934.0,
"step": 230
},
{
"entropy": 8.013978528976441,
"epoch": 0.02799285288862418,
"grad_norm": 1.109375,
"learning_rate": 0.00011700000000000001,
"loss": 7.2145,
"mean_token_accuracy": 0.06474445015192032,
"num_tokens": 656008.0,
"step": 235
},
{
"entropy": 8.028218317031861,
"epoch": 0.02858844550327576,
"grad_norm": 1.109375,
"learning_rate": 0.00011949999999999999,
"loss": 7.4327,
"mean_token_accuracy": 0.06487233191728592,
"num_tokens": 668843.0,
"step": 240
},
{
"entropy": 8.044657325744629,
"epoch": 0.029184038117927337,
"grad_norm": 0.94921875,
"learning_rate": 0.000122,
"loss": 7.3194,
"mean_token_accuracy": 0.06946417763829231,
"num_tokens": 681883.0,
"step": 245
},
{
"entropy": 8.010096073150635,
"epoch": 0.029779630732578916,
"grad_norm": 1.453125,
"learning_rate": 0.0001245,
"loss": 7.2913,
"mean_token_accuracy": 0.06887354739010335,
"num_tokens": 694282.0,
"step": 250
},
{
"entropy": 7.946925401687622,
"epoch": 0.030375223347230494,
"grad_norm": 0.98046875,
"learning_rate": 0.000127,
"loss": 7.21,
"mean_token_accuracy": 0.07376883625984192,
"num_tokens": 708158.0,
"step": 255
},
{
"entropy": 7.903460550308227,
"epoch": 0.030970815961882073,
"grad_norm": 1.0234375,
"learning_rate": 0.0001295,
"loss": 7.2194,
"mean_token_accuracy": 0.06968091353774071,
"num_tokens": 722083.0,
"step": 260
},
{
"entropy": 7.88522777557373,
"epoch": 0.03156640857653365,
"grad_norm": 1.0234375,
"learning_rate": 0.000132,
"loss": 7.2204,
"mean_token_accuracy": 0.07137699350714684,
"num_tokens": 737806.0,
"step": 265
},
{
"entropy": 7.816258811950684,
"epoch": 0.03216200119118523,
"grad_norm": 0.8828125,
"learning_rate": 0.00013450000000000002,
"loss": 7.0864,
"mean_token_accuracy": 0.07385070994496346,
"num_tokens": 754100.0,
"step": 270
},
{
"entropy": 7.8969801902771,
"epoch": 0.03275759380583681,
"grad_norm": 0.87109375,
"learning_rate": 0.00013700000000000002,
"loss": 7.1976,
"mean_token_accuracy": 0.07308120429515838,
"num_tokens": 769128.0,
"step": 275
},
{
"entropy": 7.8850123405456545,
"epoch": 0.03335318642048839,
"grad_norm": 1.0859375,
"learning_rate": 0.0001395,
"loss": 7.2051,
"mean_token_accuracy": 0.07497966662049294,
"num_tokens": 781259.0,
"step": 280
},
{
"entropy": 7.812098693847656,
"epoch": 0.033948779035139966,
"grad_norm": 0.875,
"learning_rate": 0.00014199999999999998,
"loss": 7.1031,
"mean_token_accuracy": 0.07991473153233528,
"num_tokens": 796152.0,
"step": 285
},
{
"entropy": 7.7976302146911625,
"epoch": 0.034544371649791544,
"grad_norm": 1.03125,
"learning_rate": 0.0001445,
"loss": 7.1463,
"mean_token_accuracy": 0.07945740520954132,
"num_tokens": 809686.0,
"step": 290
},
{
"entropy": 7.7539088249206545,
"epoch": 0.03513996426444312,
"grad_norm": 0.9296875,
"learning_rate": 0.000147,
"loss": 7.1006,
"mean_token_accuracy": 0.08280241340398789,
"num_tokens": 824500.0,
"step": 295
},
{
"entropy": 7.737009906768799,
"epoch": 0.0357355568790947,
"grad_norm": 1.03125,
"learning_rate": 0.0001495,
"loss": 7.0368,
"mean_token_accuracy": 0.0804534487426281,
"num_tokens": 840643.0,
"step": 300
},
{
"entropy": 7.828617763519287,
"epoch": 0.03633114949374628,
"grad_norm": 1.234375,
"learning_rate": 0.000152,
"loss": 7.0918,
"mean_token_accuracy": 0.08342065252363681,
"num_tokens": 853207.0,
"step": 305
},
{
"entropy": 7.701815795898438,
"epoch": 0.03692674210839786,
"grad_norm": 0.93359375,
"learning_rate": 0.00015450000000000001,
"loss": 7.1476,
"mean_token_accuracy": 0.07626047655940056,
"num_tokens": 867735.0,
"step": 310
},
{
"entropy": 7.727453899383545,
"epoch": 0.03752233472304944,
"grad_norm": 1.1484375,
"learning_rate": 0.000157,
"loss": 7.0593,
"mean_token_accuracy": 0.08431698828935623,
"num_tokens": 881237.0,
"step": 315
},
{
"entropy": 7.756283855438232,
"epoch": 0.038117927337701016,
"grad_norm": 1.1484375,
"learning_rate": 0.0001595,
"loss": 7.0628,
"mean_token_accuracy": 0.08603817000985145,
"num_tokens": 894288.0,
"step": 320
},
{
"entropy": 7.698923826217651,
"epoch": 0.03871351995235259,
"grad_norm": 1.015625,
"learning_rate": 0.000162,
"loss": 7.0176,
"mean_token_accuracy": 0.08519366085529327,
"num_tokens": 907492.0,
"step": 325
},
{
"entropy": 7.638896846771241,
"epoch": 0.039309112567004166,
"grad_norm": 0.91015625,
"learning_rate": 0.00016450000000000001,
"loss": 6.9325,
"mean_token_accuracy": 0.08895694315433503,
"num_tokens": 922255.0,
"step": 330
},
{
"entropy": 7.63583927154541,
"epoch": 0.039904705181655745,
"grad_norm": 1.078125,
"learning_rate": 0.00016700000000000002,
"loss": 6.9871,
"mean_token_accuracy": 0.08658753335475922,
"num_tokens": 937055.0,
"step": 335
},
{
"entropy": 7.590744733810425,
"epoch": 0.04050029779630732,
"grad_norm": 1.1953125,
"learning_rate": 0.00016950000000000003,
"loss": 6.8883,
"mean_token_accuracy": 0.0914541356265545,
"num_tokens": 951248.0,
"step": 340
},
{
"entropy": 7.55070309638977,
"epoch": 0.0410958904109589,
"grad_norm": 0.91796875,
"learning_rate": 0.00017199999999999998,
"loss": 6.8828,
"mean_token_accuracy": 0.091338162869215,
"num_tokens": 965211.0,
"step": 345
},
{
"entropy": 7.525717115402221,
"epoch": 0.04169148302561048,
"grad_norm": 1.0625,
"learning_rate": 0.00017449999999999999,
"loss": 6.8702,
"mean_token_accuracy": 0.09259014204144478,
"num_tokens": 979787.0,
"step": 350
},
{
"entropy": 7.50576491355896,
"epoch": 0.04228707564026206,
"grad_norm": 1.1875,
"learning_rate": 0.000177,
"loss": 6.8776,
"mean_token_accuracy": 0.09012275263667106,
"num_tokens": 993412.0,
"step": 355
},
{
"entropy": 7.547108888626099,
"epoch": 0.04288266825491364,
"grad_norm": 1.2734375,
"learning_rate": 0.0001795,
"loss": 6.8638,
"mean_token_accuracy": 0.0968759037554264,
"num_tokens": 1006819.0,
"step": 360
},
{
"entropy": 7.480889987945557,
"epoch": 0.043478260869565216,
"grad_norm": 0.921875,
"learning_rate": 0.000182,
"loss": 6.8627,
"mean_token_accuracy": 0.09363154023885727,
"num_tokens": 1022287.0,
"step": 365
},
{
"entropy": 7.4811498641967775,
"epoch": 0.044073853484216795,
"grad_norm": 1.390625,
"learning_rate": 0.0001845,
"loss": 6.8544,
"mean_token_accuracy": 0.0976115770637989,
"num_tokens": 1034908.0,
"step": 370
},
{
"entropy": 7.450716018676758,
"epoch": 0.04466944609886837,
"grad_norm": 0.953125,
"learning_rate": 0.000187,
"loss": 6.894,
"mean_token_accuracy": 0.09215271174907684,
"num_tokens": 1051506.0,
"step": 375
},
{
"entropy": 7.386427736282348,
"epoch": 0.04526503871351995,
"grad_norm": 1.125,
"learning_rate": 0.0001895,
"loss": 6.7631,
"mean_token_accuracy": 0.0934596985578537,
"num_tokens": 1066382.0,
"step": 380
},
{
"entropy": 7.438938188552856,
"epoch": 0.04586063132817153,
"grad_norm": 0.8984375,
"learning_rate": 0.000192,
"loss": 6.8572,
"mean_token_accuracy": 0.08746511340141297,
"num_tokens": 1082030.0,
"step": 385
},
{
"entropy": 7.481917047500611,
"epoch": 0.04645622394282311,
"grad_norm": 0.9609375,
"learning_rate": 0.0001945,
"loss": 6.8641,
"mean_token_accuracy": 0.09139562249183655,
"num_tokens": 1095926.0,
"step": 390
},
{
"entropy": 7.349998188018799,
"epoch": 0.04705181655747469,
"grad_norm": 1.1171875,
"learning_rate": 0.00019700000000000002,
"loss": 6.7656,
"mean_token_accuracy": 0.0982861079275608,
"num_tokens": 1110579.0,
"step": 395
},
{
"entropy": 7.378432703018189,
"epoch": 0.047647409172126266,
"grad_norm": 1.1015625,
"learning_rate": 0.00019950000000000002,
"loss": 6.7926,
"mean_token_accuracy": 0.09419288635253906,
"num_tokens": 1124001.0,
"step": 400
},
{
"entropy": 7.473501968383789,
"epoch": 0.048243001786777845,
"grad_norm": 1.015625,
"learning_rate": 0.000202,
"loss": 6.8251,
"mean_token_accuracy": 0.09458990395069122,
"num_tokens": 1138316.0,
"step": 405
},
{
"entropy": 7.333215761184692,
"epoch": 0.048838594401429423,
"grad_norm": 1.140625,
"learning_rate": 0.00020449999999999998,
"loss": 6.7328,
"mean_token_accuracy": 0.09874959588050843,
"num_tokens": 1150055.0,
"step": 410
},
{
"entropy": 7.342270565032959,
"epoch": 0.049434187016081,
"grad_norm": 1.109375,
"learning_rate": 0.000207,
"loss": 6.8506,
"mean_token_accuracy": 0.09371097683906555,
"num_tokens": 1163645.0,
"step": 415
},
{
"entropy": 7.368842363357544,
"epoch": 0.05002977963073258,
"grad_norm": 1.0546875,
"learning_rate": 0.0002095,
"loss": 6.6872,
"mean_token_accuracy": 0.10014689564704896,
"num_tokens": 1176660.0,
"step": 420
},
{
"entropy": 7.309666538238526,
"epoch": 0.05062537224538416,
"grad_norm": 0.94921875,
"learning_rate": 0.000212,
"loss": 6.7961,
"mean_token_accuracy": 0.09248664155602455,
"num_tokens": 1193169.0,
"step": 425
},
{
"entropy": 7.347740602493286,
"epoch": 0.05122096486003574,
"grad_norm": 1.2421875,
"learning_rate": 0.0002145,
"loss": 6.74,
"mean_token_accuracy": 0.09651226848363877,
"num_tokens": 1208484.0,
"step": 430
},
{
"entropy": 7.2615007877349855,
"epoch": 0.051816557474687316,
"grad_norm": 1.046875,
"learning_rate": 0.00021700000000000002,
"loss": 6.7644,
"mean_token_accuracy": 0.0988834910094738,
"num_tokens": 1222925.0,
"step": 435
},
{
"entropy": 7.298656797409057,
"epoch": 0.052412150089338895,
"grad_norm": 0.93359375,
"learning_rate": 0.0002195,
"loss": 6.698,
"mean_token_accuracy": 0.09822965413331985,
"num_tokens": 1237949.0,
"step": 440
},
{
"entropy": 7.3722340106964115,
"epoch": 0.053007742703990474,
"grad_norm": 1.1171875,
"learning_rate": 0.000222,
"loss": 6.7266,
"mean_token_accuracy": 0.1032131403684616,
"num_tokens": 1251535.0,
"step": 445
},
{
"entropy": 7.26842737197876,
"epoch": 0.05360333531864205,
"grad_norm": 1.125,
"learning_rate": 0.0002245,
"loss": 6.6426,
"mean_token_accuracy": 0.1049805611371994,
"num_tokens": 1265453.0,
"step": 450
},
{
"entropy": 7.230884599685669,
"epoch": 0.054198927933293624,
"grad_norm": 1.0078125,
"learning_rate": 0.00022700000000000002,
"loss": 6.6627,
"mean_token_accuracy": 0.1022911824285984,
"num_tokens": 1280662.0,
"step": 455
},
{
"entropy": 7.22289514541626,
"epoch": 0.0547945205479452,
"grad_norm": 1.1484375,
"learning_rate": 0.00022950000000000002,
"loss": 6.7295,
"mean_token_accuracy": 0.09482937082648277,
"num_tokens": 1294980.0,
"step": 460
},
{
"entropy": 7.170809268951416,
"epoch": 0.05539011316259678,
"grad_norm": 1.015625,
"learning_rate": 0.00023200000000000003,
"loss": 6.6079,
"mean_token_accuracy": 0.09827929437160492,
"num_tokens": 1309686.0,
"step": 465
},
{
"entropy": 7.160658168792724,
"epoch": 0.05598570577724836,
"grad_norm": 1.046875,
"learning_rate": 0.00023449999999999998,
"loss": 6.5961,
"mean_token_accuracy": 0.10312132835388184,
"num_tokens": 1324815.0,
"step": 470
},
{
"entropy": 7.259875392913818,
"epoch": 0.05658129839189994,
"grad_norm": 1.109375,
"learning_rate": 0.000237,
"loss": 6.6489,
"mean_token_accuracy": 0.10013843029737472,
"num_tokens": 1337829.0,
"step": 475
},
{
"entropy": 7.098458099365234,
"epoch": 0.05717689100655152,
"grad_norm": 1.09375,
"learning_rate": 0.0002395,
"loss": 6.6612,
"mean_token_accuracy": 0.10124345347285271,
"num_tokens": 1352163.0,
"step": 480
},
{
"entropy": 7.205197191238403,
"epoch": 0.057772483621203095,
"grad_norm": 1.140625,
"learning_rate": 0.000242,
"loss": 6.5936,
"mean_token_accuracy": 0.10394150242209435,
"num_tokens": 1364954.0,
"step": 485
},
{
"entropy": 7.174849510192871,
"epoch": 0.058368076235854674,
"grad_norm": 1.03125,
"learning_rate": 0.0002445,
"loss": 6.6166,
"mean_token_accuracy": 0.09912089258432388,
"num_tokens": 1379414.0,
"step": 490
},
{
"entropy": 7.093812608718872,
"epoch": 0.05896366885050625,
"grad_norm": 1.0546875,
"learning_rate": 0.000247,
"loss": 6.5052,
"mean_token_accuracy": 0.10558257475495339,
"num_tokens": 1393515.0,
"step": 495
},
{
"entropy": 7.065914583206177,
"epoch": 0.05955926146515783,
"grad_norm": 1.203125,
"learning_rate": 0.0002495,
"loss": 6.5114,
"mean_token_accuracy": 0.10575130134820938,
"num_tokens": 1408986.0,
"step": 500
},
{
"entropy": 7.068597936630249,
"epoch": 0.06015485407980941,
"grad_norm": 0.984375,
"learning_rate": 0.000252,
"loss": 6.5279,
"mean_token_accuracy": 0.1057064950466156,
"num_tokens": 1423176.0,
"step": 505
},
{
"entropy": 7.180021667480469,
"epoch": 0.06075044669446099,
"grad_norm": 1.0234375,
"learning_rate": 0.0002545,
"loss": 6.6594,
"mean_token_accuracy": 0.101287080347538,
"num_tokens": 1436802.0,
"step": 510
},
{
"entropy": 7.107086324691773,
"epoch": 0.06134603930911257,
"grad_norm": 1.25,
"learning_rate": 0.000257,
"loss": 6.5323,
"mean_token_accuracy": 0.1064387544989586,
"num_tokens": 1449212.0,
"step": 515
},
{
"entropy": 7.067509412765503,
"epoch": 0.061941631923764146,
"grad_norm": 1.078125,
"learning_rate": 0.0002595,
"loss": 6.5942,
"mean_token_accuracy": 0.10407756939530373,
"num_tokens": 1463357.0,
"step": 520
},
{
"entropy": 7.188254976272583,
"epoch": 0.06253722453841572,
"grad_norm": 1.171875,
"learning_rate": 0.000262,
"loss": 6.5773,
"mean_token_accuracy": 0.10477585569024087,
"num_tokens": 1477494.0,
"step": 525
},
{
"entropy": 7.07730622291565,
"epoch": 0.0631328171530673,
"grad_norm": 1.265625,
"learning_rate": 0.00026450000000000003,
"loss": 6.5115,
"mean_token_accuracy": 0.10843904837965965,
"num_tokens": 1491687.0,
"step": 530
},
{
"entropy": 6.918906593322754,
"epoch": 0.06372840976771887,
"grad_norm": 1.0703125,
"learning_rate": 0.00026700000000000004,
"loss": 6.4387,
"mean_token_accuracy": 0.11039169207215309,
"num_tokens": 1506656.0,
"step": 535
},
{
"entropy": 7.066193056106568,
"epoch": 0.06432400238237046,
"grad_norm": 1.203125,
"learning_rate": 0.00026950000000000005,
"loss": 6.5055,
"mean_token_accuracy": 0.10977512747049331,
"num_tokens": 1519336.0,
"step": 540
},
{
"entropy": 7.121071767807007,
"epoch": 0.06491959499702203,
"grad_norm": 1.078125,
"learning_rate": 0.00027200000000000005,
"loss": 6.637,
"mean_token_accuracy": 0.10029823333024979,
"num_tokens": 1535204.0,
"step": 545
},
{
"entropy": 7.026650047302246,
"epoch": 0.06551518761167362,
"grad_norm": 1.046875,
"learning_rate": 0.0002745,
"loss": 6.4435,
"mean_token_accuracy": 0.11559138372540474,
"num_tokens": 1547876.0,
"step": 550
},
{
"entropy": 7.059904098510742,
"epoch": 0.06611078022632519,
"grad_norm": 1.140625,
"learning_rate": 0.000277,
"loss": 6.5566,
"mean_token_accuracy": 0.10418016389012337,
"num_tokens": 1560315.0,
"step": 555
},
{
"entropy": 7.000280284881592,
"epoch": 0.06670637284097677,
"grad_norm": 1.171875,
"learning_rate": 0.0002795,
"loss": 6.6114,
"mean_token_accuracy": 0.10283819288015365,
"num_tokens": 1576379.0,
"step": 560
},
{
"entropy": 7.1294999599456785,
"epoch": 0.06730196545562835,
"grad_norm": 1.3125,
"learning_rate": 0.00028199999999999997,
"loss": 6.4815,
"mean_token_accuracy": 0.1085024543106556,
"num_tokens": 1590480.0,
"step": 565
},
{
"entropy": 6.918955373764038,
"epoch": 0.06789755807027993,
"grad_norm": 0.9921875,
"learning_rate": 0.0002845,
"loss": 6.5464,
"mean_token_accuracy": 0.10271873772144317,
"num_tokens": 1605328.0,
"step": 570
},
{
"entropy": 7.0181163311004635,
"epoch": 0.0684931506849315,
"grad_norm": 1.03125,
"learning_rate": 0.000287,
"loss": 6.4882,
"mean_token_accuracy": 0.10755092278122902,
"num_tokens": 1619814.0,
"step": 575
},
{
"entropy": 6.944701337814331,
"epoch": 0.06908874329958309,
"grad_norm": 1.09375,
"learning_rate": 0.0002895,
"loss": 6.4023,
"mean_token_accuracy": 0.10893059074878693,
"num_tokens": 1632561.0,
"step": 580
},
{
"entropy": 6.987877702713012,
"epoch": 0.06968433591423466,
"grad_norm": 1.0390625,
"learning_rate": 0.000292,
"loss": 6.5072,
"mean_token_accuracy": 0.10601216182112694,
"num_tokens": 1645669.0,
"step": 585
},
{
"entropy": 6.988888740539551,
"epoch": 0.07027992852888625,
"grad_norm": 1.2734375,
"learning_rate": 0.0002945,
"loss": 6.454,
"mean_token_accuracy": 0.10420122295618058,
"num_tokens": 1660479.0,
"step": 590
},
{
"entropy": 6.946523141860962,
"epoch": 0.07087552114353782,
"grad_norm": 1.3125,
"learning_rate": 0.000297,
"loss": 6.4849,
"mean_token_accuracy": 0.1114343523979187,
"num_tokens": 1673966.0,
"step": 595
},
{
"entropy": 6.8098907470703125,
"epoch": 0.0714711137581894,
"grad_norm": 0.98828125,
"learning_rate": 0.0002995,
"loss": 6.3577,
"mean_token_accuracy": 0.11343350037932395,
"num_tokens": 1688866.0,
"step": 600
},
{
"entropy": 6.852885770797729,
"epoch": 0.07206670637284097,
"grad_norm": 1.0390625,
"learning_rate": 0.000302,
"loss": 6.3766,
"mean_token_accuracy": 0.11477320343255996,
"num_tokens": 1703810.0,
"step": 605
},
{
"entropy": 7.044281244277954,
"epoch": 0.07266229898749256,
"grad_norm": 1.0078125,
"learning_rate": 0.0003045,
"loss": 6.5907,
"mean_token_accuracy": 0.10320461988449096,
"num_tokens": 1719308.0,
"step": 610
},
{
"entropy": 6.940822982788086,
"epoch": 0.07325789160214413,
"grad_norm": 1.1640625,
"learning_rate": 0.000307,
"loss": 6.369,
"mean_token_accuracy": 0.1170460619032383,
"num_tokens": 1734508.0,
"step": 615
},
{
"entropy": 6.8789849281311035,
"epoch": 0.07385348421679572,
"grad_norm": 1.15625,
"learning_rate": 0.0003095,
"loss": 6.3345,
"mean_token_accuracy": 0.11990054100751876,
"num_tokens": 1745999.0,
"step": 620
},
{
"entropy": 6.763300085067749,
"epoch": 0.07444907683144729,
"grad_norm": 1.0703125,
"learning_rate": 0.000312,
"loss": 6.3454,
"mean_token_accuracy": 0.11457164734601974,
"num_tokens": 1761083.0,
"step": 625
},
{
"entropy": 6.822115707397461,
"epoch": 0.07504466944609887,
"grad_norm": 1.03125,
"learning_rate": 0.0003145,
"loss": 6.3016,
"mean_token_accuracy": 0.11405027881264687,
"num_tokens": 1774948.0,
"step": 630
},
{
"entropy": 6.937965774536133,
"epoch": 0.07564026206075045,
"grad_norm": 1.2421875,
"learning_rate": 0.000317,
"loss": 6.4233,
"mean_token_accuracy": 0.10521830841898919,
"num_tokens": 1787897.0,
"step": 635
},
{
"entropy": 6.7878087043762205,
"epoch": 0.07623585467540203,
"grad_norm": 1.015625,
"learning_rate": 0.0003195,
"loss": 6.3098,
"mean_token_accuracy": 0.11155571267008782,
"num_tokens": 1801208.0,
"step": 640
},
{
"entropy": 6.7966279029846195,
"epoch": 0.0768314472900536,
"grad_norm": 0.98828125,
"learning_rate": 0.000322,
"loss": 6.3072,
"mean_token_accuracy": 0.11627824008464813,
"num_tokens": 1815439.0,
"step": 645
},
{
"entropy": 6.812145853042603,
"epoch": 0.07742703990470517,
"grad_norm": 1.109375,
"learning_rate": 0.00032450000000000003,
"loss": 6.3388,
"mean_token_accuracy": 0.11540080681443214,
"num_tokens": 1828177.0,
"step": 650
},
{
"entropy": 6.74655556678772,
"epoch": 0.07802263251935676,
"grad_norm": 1.171875,
"learning_rate": 0.00032700000000000003,
"loss": 6.3067,
"mean_token_accuracy": 0.11337784305214882,
"num_tokens": 1844854.0,
"step": 655
},
{
"entropy": 6.742314243316651,
"epoch": 0.07861822513400833,
"grad_norm": 1.046875,
"learning_rate": 0.00032950000000000004,
"loss": 6.318,
"mean_token_accuracy": 0.11590223461389541,
"num_tokens": 1858245.0,
"step": 660
},
{
"entropy": 6.87442831993103,
"epoch": 0.07921381774865992,
"grad_norm": 1.109375,
"learning_rate": 0.00033200000000000005,
"loss": 6.4123,
"mean_token_accuracy": 0.10785512253642082,
"num_tokens": 1872167.0,
"step": 665
},
{
"entropy": 6.815114450454712,
"epoch": 0.07980941036331149,
"grad_norm": 1.390625,
"learning_rate": 0.00033450000000000005,
"loss": 6.3716,
"mean_token_accuracy": 0.11168409660458564,
"num_tokens": 1885376.0,
"step": 670
},
{
"entropy": 6.842722368240357,
"epoch": 0.08040500297796307,
"grad_norm": 1.1015625,
"learning_rate": 0.000337,
"loss": 6.402,
"mean_token_accuracy": 0.10913070514798165,
"num_tokens": 1899318.0,
"step": 675
},
{
"entropy": 6.785850381851196,
"epoch": 0.08100059559261465,
"grad_norm": 1.03125,
"learning_rate": 0.0003395,
"loss": 6.4205,
"mean_token_accuracy": 0.10894930735230446,
"num_tokens": 1914028.0,
"step": 680
},
{
"entropy": 6.8544024467468265,
"epoch": 0.08159618820726623,
"grad_norm": 1.1171875,
"learning_rate": 0.000342,
"loss": 6.3637,
"mean_token_accuracy": 0.11298043429851531,
"num_tokens": 1927653.0,
"step": 685
},
{
"entropy": 6.788467264175415,
"epoch": 0.0821917808219178,
"grad_norm": 0.98046875,
"learning_rate": 0.00034449999999999997,
"loss": 6.3522,
"mean_token_accuracy": 0.10948711559176445,
"num_tokens": 1941397.0,
"step": 690
},
{
"entropy": 6.758484315872193,
"epoch": 0.08278737343656939,
"grad_norm": 0.9765625,
"learning_rate": 0.000347,
"loss": 6.3168,
"mean_token_accuracy": 0.11713797152042389,
"num_tokens": 1955941.0,
"step": 695
},
{
"entropy": 6.817478275299072,
"epoch": 0.08338296605122096,
"grad_norm": 1.0078125,
"learning_rate": 0.0003495,
"loss": 6.3172,
"mean_token_accuracy": 0.11794802024960518,
"num_tokens": 1970440.0,
"step": 700
},
{
"entropy": 6.679837894439697,
"epoch": 0.08397855866587255,
"grad_norm": 1.1484375,
"learning_rate": 0.000352,
"loss": 6.2517,
"mean_token_accuracy": 0.11665802001953125,
"num_tokens": 1985770.0,
"step": 705
},
{
"entropy": 6.776376962661743,
"epoch": 0.08457415128052412,
"grad_norm": 1.1953125,
"learning_rate": 0.0003545,
"loss": 6.3912,
"mean_token_accuracy": 0.10708962306380272,
"num_tokens": 1999314.0,
"step": 710
},
{
"entropy": 6.668450260162354,
"epoch": 0.0851697438951757,
"grad_norm": 1.09375,
"learning_rate": 0.000357,
"loss": 6.2358,
"mean_token_accuracy": 0.12091019451618194,
"num_tokens": 2014578.0,
"step": 715
},
{
"entropy": 6.739320993423462,
"epoch": 0.08576533650982728,
"grad_norm": 1.078125,
"learning_rate": 0.0003595,
"loss": 6.3438,
"mean_token_accuracy": 0.11275363117456436,
"num_tokens": 2027650.0,
"step": 720
},
{
"entropy": 6.798482942581177,
"epoch": 0.08636092912447886,
"grad_norm": 1.2265625,
"learning_rate": 0.000362,
"loss": 6.2793,
"mean_token_accuracy": 0.1181432731449604,
"num_tokens": 2042215.0,
"step": 725
},
{
"entropy": 6.676054430007935,
"epoch": 0.08695652173913043,
"grad_norm": 1.1484375,
"learning_rate": 0.0003645,
"loss": 6.33,
"mean_token_accuracy": 0.11466208919882774,
"num_tokens": 2055818.0,
"step": 730
},
{
"entropy": 6.653960132598877,
"epoch": 0.08755211435378202,
"grad_norm": 1.125,
"learning_rate": 0.000367,
"loss": 6.2509,
"mean_token_accuracy": 0.11222518086433411,
"num_tokens": 2069040.0,
"step": 735
},
{
"entropy": 6.682643699645996,
"epoch": 0.08814770696843359,
"grad_norm": 0.8828125,
"learning_rate": 0.0003695,
"loss": 6.2528,
"mean_token_accuracy": 0.11789485588669776,
"num_tokens": 2085085.0,
"step": 740
},
{
"entropy": 6.727219772338867,
"epoch": 0.08874329958308518,
"grad_norm": 0.984375,
"learning_rate": 0.000372,
"loss": 6.3465,
"mean_token_accuracy": 0.11351984664797783,
"num_tokens": 2100780.0,
"step": 745
},
{
"entropy": 6.862605714797974,
"epoch": 0.08933889219773675,
"grad_norm": 0.94140625,
"learning_rate": 0.0003745,
"loss": 6.3257,
"mean_token_accuracy": 0.112281883507967,
"num_tokens": 2116317.0,
"step": 750
},
{
"entropy": 6.620303583145142,
"epoch": 0.08993448481238833,
"grad_norm": 1.0703125,
"learning_rate": 0.000377,
"loss": 6.2802,
"mean_token_accuracy": 0.11524079740047455,
"num_tokens": 2130865.0,
"step": 755
},
{
"entropy": 6.664459180831909,
"epoch": 0.0905300774270399,
"grad_norm": 0.98828125,
"learning_rate": 0.0003795,
"loss": 6.3047,
"mean_token_accuracy": 0.11343947649002076,
"num_tokens": 2145791.0,
"step": 760
},
{
"entropy": 6.749904727935791,
"epoch": 0.09112567004169149,
"grad_norm": 1.1171875,
"learning_rate": 0.000382,
"loss": 6.2298,
"mean_token_accuracy": 0.12268061265349388,
"num_tokens": 2158750.0,
"step": 765
},
{
"entropy": 6.6750328063964846,
"epoch": 0.09172126265634306,
"grad_norm": 1.109375,
"learning_rate": 0.0003845,
"loss": 6.3118,
"mean_token_accuracy": 0.11220015361905097,
"num_tokens": 2175026.0,
"step": 770
},
{
"entropy": 6.596572732925415,
"epoch": 0.09231685527099463,
"grad_norm": 0.984375,
"learning_rate": 0.00038700000000000003,
"loss": 6.2207,
"mean_token_accuracy": 0.11250628605484962,
"num_tokens": 2189597.0,
"step": 775
},
{
"entropy": 6.779586124420166,
"epoch": 0.09291244788564622,
"grad_norm": 1.1640625,
"learning_rate": 0.00038950000000000003,
"loss": 6.4,
"mean_token_accuracy": 0.11090977415442467,
"num_tokens": 2203663.0,
"step": 780
},
{
"entropy": 6.60975866317749,
"epoch": 0.09350804050029779,
"grad_norm": 0.9375,
"learning_rate": 0.00039200000000000004,
"loss": 6.2014,
"mean_token_accuracy": 0.1156569704413414,
"num_tokens": 2218784.0,
"step": 785
},
{
"entropy": 6.622003269195557,
"epoch": 0.09410363311494938,
"grad_norm": 1.0078125,
"learning_rate": 0.00039450000000000005,
"loss": 6.2237,
"mean_token_accuracy": 0.11353443339467048,
"num_tokens": 2233256.0,
"step": 790
},
{
"entropy": 6.63480863571167,
"epoch": 0.09469922572960095,
"grad_norm": 1.03125,
"learning_rate": 0.00039700000000000005,
"loss": 6.2379,
"mean_token_accuracy": 0.11858787089586258,
"num_tokens": 2247689.0,
"step": 795
},
{
"entropy": 6.557226371765137,
"epoch": 0.09529481834425253,
"grad_norm": 1.171875,
"learning_rate": 0.0003995,
"loss": 6.0949,
"mean_token_accuracy": 0.12291062027215957,
"num_tokens": 2261158.0,
"step": 800
},
{
"entropy": 6.518621158599854,
"epoch": 0.0958904109589041,
"grad_norm": 1.03125,
"learning_rate": 0.000402,
"loss": 6.0526,
"mean_token_accuracy": 0.13050554171204568,
"num_tokens": 2274013.0,
"step": 805
},
{
"entropy": 6.6024250984191895,
"epoch": 0.09648600357355569,
"grad_norm": 1.015625,
"learning_rate": 0.0004045,
"loss": 6.3208,
"mean_token_accuracy": 0.11554502844810485,
"num_tokens": 2287471.0,
"step": 810
},
{
"entropy": 6.5365015983581545,
"epoch": 0.09708159618820726,
"grad_norm": 1.21875,
"learning_rate": 0.00040699999999999997,
"loss": 6.1449,
"mean_token_accuracy": 0.1233399897813797,
"num_tokens": 2302536.0,
"step": 815
},
{
"entropy": 6.640579080581665,
"epoch": 0.09767718880285885,
"grad_norm": 1.078125,
"learning_rate": 0.0004095,
"loss": 6.337,
"mean_token_accuracy": 0.11535090804100037,
"num_tokens": 2315235.0,
"step": 820
},
{
"entropy": 6.747769975662232,
"epoch": 0.09827278141751042,
"grad_norm": 1.109375,
"learning_rate": 0.000412,
"loss": 6.3812,
"mean_token_accuracy": 0.11088532656431198,
"num_tokens": 2329010.0,
"step": 825
},
{
"entropy": 6.591479730606079,
"epoch": 0.098868374032162,
"grad_norm": 0.97265625,
"learning_rate": 0.0004145,
"loss": 6.2125,
"mean_token_accuracy": 0.1245424821972847,
"num_tokens": 2342862.0,
"step": 830
},
{
"entropy": 6.614097547531128,
"epoch": 0.09946396664681358,
"grad_norm": 1.03125,
"learning_rate": 0.000417,
"loss": 6.0647,
"mean_token_accuracy": 0.12870702892541885,
"num_tokens": 2354680.0,
"step": 835
},
{
"entropy": 6.493401050567627,
"epoch": 0.10005955926146516,
"grad_norm": 1.1328125,
"learning_rate": 0.0004195,
"loss": 6.1962,
"mean_token_accuracy": 0.11563577875494957,
"num_tokens": 2369251.0,
"step": 840
},
{
"entropy": 6.510954999923706,
"epoch": 0.10065515187611673,
"grad_norm": 1.015625,
"learning_rate": 0.000422,
"loss": 6.2345,
"mean_token_accuracy": 0.11866097152233124,
"num_tokens": 2382880.0,
"step": 845
},
{
"entropy": 6.579282760620117,
"epoch": 0.10125074449076832,
"grad_norm": 0.89453125,
"learning_rate": 0.0004245,
"loss": 6.1124,
"mean_token_accuracy": 0.1205758735537529,
"num_tokens": 2397627.0,
"step": 850
},
{
"entropy": 6.6060676097869875,
"epoch": 0.10184633710541989,
"grad_norm": 0.9296875,
"learning_rate": 0.000427,
"loss": 6.1782,
"mean_token_accuracy": 0.12276405319571496,
"num_tokens": 2411923.0,
"step": 855
},
{
"entropy": 6.614116859436035,
"epoch": 0.10244192972007148,
"grad_norm": 1.046875,
"learning_rate": 0.0004295,
"loss": 6.2741,
"mean_token_accuracy": 0.11239862740039826,
"num_tokens": 2426253.0,
"step": 860
},
{
"entropy": 6.549399185180664,
"epoch": 0.10303752233472305,
"grad_norm": 0.9765625,
"learning_rate": 0.000432,
"loss": 6.1968,
"mean_token_accuracy": 0.12170327082276344,
"num_tokens": 2440887.0,
"step": 865
},
{
"entropy": 6.552827501296997,
"epoch": 0.10363311494937463,
"grad_norm": 1.09375,
"learning_rate": 0.0004345,
"loss": 6.1521,
"mean_token_accuracy": 0.12400813326239586,
"num_tokens": 2453556.0,
"step": 870
},
{
"entropy": 6.546022272109985,
"epoch": 0.1042287075640262,
"grad_norm": 0.94921875,
"learning_rate": 0.000437,
"loss": 6.2013,
"mean_token_accuracy": 0.12121366187930108,
"num_tokens": 2467374.0,
"step": 875
},
{
"entropy": 6.5050191402435305,
"epoch": 0.10482430017867779,
"grad_norm": 1.0546875,
"learning_rate": 0.0004395,
"loss": 6.182,
"mean_token_accuracy": 0.11917331218719482,
"num_tokens": 2481106.0,
"step": 880
},
{
"entropy": 6.468537712097168,
"epoch": 0.10541989279332936,
"grad_norm": 1.1328125,
"learning_rate": 0.000442,
"loss": 6.0994,
"mean_token_accuracy": 0.1258152939379215,
"num_tokens": 2495431.0,
"step": 885
},
{
"entropy": 6.5304240703582765,
"epoch": 0.10601548540798095,
"grad_norm": 1.125,
"learning_rate": 0.0004445,
"loss": 6.1736,
"mean_token_accuracy": 0.11701044365763665,
"num_tokens": 2511312.0,
"step": 890
},
{
"entropy": 6.55829701423645,
"epoch": 0.10661107802263252,
"grad_norm": 0.92578125,
"learning_rate": 0.000447,
"loss": 6.1349,
"mean_token_accuracy": 0.12205464690923691,
"num_tokens": 2524960.0,
"step": 895
},
{
"entropy": 6.500060653686523,
"epoch": 0.1072066706372841,
"grad_norm": 1.0859375,
"learning_rate": 0.00044950000000000003,
"loss": 6.1945,
"mean_token_accuracy": 0.12649419754743577,
"num_tokens": 2537965.0,
"step": 900
},
{
"entropy": 6.5613710403442385,
"epoch": 0.10780226325193568,
"grad_norm": 1.1171875,
"learning_rate": 0.00045200000000000004,
"loss": 6.2021,
"mean_token_accuracy": 0.12154769822955132,
"num_tokens": 2550802.0,
"step": 905
},
{
"entropy": 6.482674026489258,
"epoch": 0.10839785586658725,
"grad_norm": 0.953125,
"learning_rate": 0.00045450000000000004,
"loss": 6.2029,
"mean_token_accuracy": 0.12224112898111343,
"num_tokens": 2565844.0,
"step": 910
},
{
"entropy": 6.567696285247803,
"epoch": 0.10899344848123883,
"grad_norm": 0.98046875,
"learning_rate": 0.00045700000000000005,
"loss": 6.1645,
"mean_token_accuracy": 0.12145803570747375,
"num_tokens": 2580049.0,
"step": 915
},
{
"entropy": 6.474516153335571,
"epoch": 0.1095890410958904,
"grad_norm": 0.9765625,
"learning_rate": 0.00045950000000000006,
"loss": 6.0639,
"mean_token_accuracy": 0.12551279589533806,
"num_tokens": 2594516.0,
"step": 920
},
{
"entropy": 6.357745122909546,
"epoch": 0.11018463371054199,
"grad_norm": 1.1171875,
"learning_rate": 0.000462,
"loss": 6.1298,
"mean_token_accuracy": 0.11735675930976867,
"num_tokens": 2610528.0,
"step": 925
},
{
"entropy": 6.500650882720947,
"epoch": 0.11078022632519356,
"grad_norm": 1.1484375,
"learning_rate": 0.0004645,
"loss": 6.1154,
"mean_token_accuracy": 0.12276835590600968,
"num_tokens": 2624357.0,
"step": 930
},
{
"entropy": 6.390538358688355,
"epoch": 0.11137581893984515,
"grad_norm": 0.921875,
"learning_rate": 0.000467,
"loss": 6.0055,
"mean_token_accuracy": 0.12950707823038102,
"num_tokens": 2640422.0,
"step": 935
},
{
"entropy": 6.3502476692199705,
"epoch": 0.11197141155449672,
"grad_norm": 0.99609375,
"learning_rate": 0.0004695,
"loss": 6.005,
"mean_token_accuracy": 0.1270424395799637,
"num_tokens": 2653573.0,
"step": 940
},
{
"entropy": 6.450726938247681,
"epoch": 0.1125670041691483,
"grad_norm": 1.0234375,
"learning_rate": 0.000472,
"loss": 6.0285,
"mean_token_accuracy": 0.1256076291203499,
"num_tokens": 2667746.0,
"step": 945
},
{
"entropy": 6.3469545364379885,
"epoch": 0.11316259678379988,
"grad_norm": 1.0703125,
"learning_rate": 0.0004745,
"loss": 6.0467,
"mean_token_accuracy": 0.12623366490006446,
"num_tokens": 2681158.0,
"step": 950
},
{
"entropy": 6.49516339302063,
"epoch": 0.11375818939845146,
"grad_norm": 1.15625,
"learning_rate": 0.000477,
"loss": 6.1388,
"mean_token_accuracy": 0.12307176068425178,
"num_tokens": 2693093.0,
"step": 955
},
{
"entropy": 6.446919965744018,
"epoch": 0.11435378201310303,
"grad_norm": 1.0390625,
"learning_rate": 0.0004795,
"loss": 6.1225,
"mean_token_accuracy": 0.1252012223005295,
"num_tokens": 2704664.0,
"step": 960
},
{
"entropy": 6.3282310962677,
"epoch": 0.11494937462775462,
"grad_norm": 1.0078125,
"learning_rate": 0.000482,
"loss": 5.9846,
"mean_token_accuracy": 0.12134386375546455,
"num_tokens": 2718495.0,
"step": 965
},
{
"entropy": 6.422227954864502,
"epoch": 0.11554496724240619,
"grad_norm": 1.0625,
"learning_rate": 0.0004845,
"loss": 6.0006,
"mean_token_accuracy": 0.1294598400592804,
"num_tokens": 2731529.0,
"step": 970
},
{
"entropy": 6.311568117141723,
"epoch": 0.11614055985705778,
"grad_norm": 0.96484375,
"learning_rate": 0.000487,
"loss": 6.0019,
"mean_token_accuracy": 0.13299588859081268,
"num_tokens": 2745382.0,
"step": 975
},
{
"entropy": 6.468044996261597,
"epoch": 0.11673615247170935,
"grad_norm": 0.921875,
"learning_rate": 0.0004895,
"loss": 6.0841,
"mean_token_accuracy": 0.12702603191137313,
"num_tokens": 2758291.0,
"step": 980
},
{
"entropy": 6.341191387176513,
"epoch": 0.11733174508636093,
"grad_norm": 1.0234375,
"learning_rate": 0.000492,
"loss": 5.9994,
"mean_token_accuracy": 0.13111869767308235,
"num_tokens": 2771179.0,
"step": 985
},
{
"entropy": 6.383563899993897,
"epoch": 0.1179273377010125,
"grad_norm": 1.0859375,
"learning_rate": 0.0004945,
"loss": 6.0669,
"mean_token_accuracy": 0.12788205072283745,
"num_tokens": 2784172.0,
"step": 990
},
{
"entropy": 6.4522544860839846,
"epoch": 0.11852293031566409,
"grad_norm": 1.0703125,
"learning_rate": 0.000497,
"loss": 6.1311,
"mean_token_accuracy": 0.12667660266160966,
"num_tokens": 2798402.0,
"step": 995
},
{
"entropy": 6.364996814727784,
"epoch": 0.11911852293031566,
"grad_norm": 0.97265625,
"learning_rate": 0.0004995,
"loss": 6.0483,
"mean_token_accuracy": 0.12513912171125413,
"num_tokens": 2813371.0,
"step": 1000
},
{
"entropy": 6.334622097015381,
"epoch": 0.11971411554496725,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999999974174771,
"loss": 6.0207,
"mean_token_accuracy": 0.1293714627623558,
"num_tokens": 2825837.0,
"step": 1005
},
{
"entropy": 6.454077196121216,
"epoch": 0.12030970815961882,
"grad_norm": 1.109375,
"learning_rate": 0.0004999999869259778,
"loss": 6.0745,
"mean_token_accuracy": 0.12313304245471954,
"num_tokens": 2837988.0,
"step": 1010
},
{
"entropy": 6.282851886749268,
"epoch": 0.1209053007742704,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999999683640947,
"loss": 5.9817,
"mean_token_accuracy": 0.13360010832548141,
"num_tokens": 2851396.0,
"step": 1015
},
{
"entropy": 6.294692039489746,
"epoch": 0.12150089338892198,
"grad_norm": 1.0,
"learning_rate": 0.0004999999417318286,
"loss": 6.0437,
"mean_token_accuracy": 0.12635651752352714,
"num_tokens": 2863958.0,
"step": 1020
},
{
"entropy": 6.461576080322265,
"epoch": 0.12209648600357356,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999999070291805,
"loss": 6.1302,
"mean_token_accuracy": 0.1211632639169693,
"num_tokens": 2876179.0,
"step": 1025
},
{
"entropy": 6.367428874969482,
"epoch": 0.12269207861822513,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999998642561514,
"loss": 6.1356,
"mean_token_accuracy": 0.12488003447651863,
"num_tokens": 2890023.0,
"step": 1030
},
{
"entropy": 6.351331377029419,
"epoch": 0.1232876712328767,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999998134127432,
"loss": 6.0051,
"mean_token_accuracy": 0.13282816261053085,
"num_tokens": 2902345.0,
"step": 1035
},
{
"entropy": 6.4046485900878904,
"epoch": 0.12388326384752829,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999997544989574,
"loss": 6.1176,
"mean_token_accuracy": 0.12030059173703193,
"num_tokens": 2916310.0,
"step": 1040
},
{
"entropy": 6.236508560180664,
"epoch": 0.12447885646217986,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999996875147963,
"loss": 5.9271,
"mean_token_accuracy": 0.13164182603359223,
"num_tokens": 2928523.0,
"step": 1045
},
{
"entropy": 6.436446762084961,
"epoch": 0.12507444907683143,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999996124602621,
"loss": 6.1333,
"mean_token_accuracy": 0.12325587049126625,
"num_tokens": 2943367.0,
"step": 1050
},
{
"entropy": 6.359068441390991,
"epoch": 0.12567004169148302,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999995293353578,
"loss": 6.0347,
"mean_token_accuracy": 0.12907699570059777,
"num_tokens": 2956995.0,
"step": 1055
},
{
"entropy": 6.286796140670776,
"epoch": 0.1262656343061346,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999994381400862,
"loss": 6.0116,
"mean_token_accuracy": 0.12894436493515968,
"num_tokens": 2971380.0,
"step": 1060
},
{
"entropy": 6.278202724456787,
"epoch": 0.1268612269207862,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999993388744504,
"loss": 6.033,
"mean_token_accuracy": 0.12432756423950195,
"num_tokens": 2986048.0,
"step": 1065
},
{
"entropy": 6.425439739227295,
"epoch": 0.12745681953543775,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999992315384543,
"loss": 6.0363,
"mean_token_accuracy": 0.1325755976140499,
"num_tokens": 2997058.0,
"step": 1070
},
{
"entropy": 6.28617296218872,
"epoch": 0.12805241215008933,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999991161321014,
"loss": 6.0221,
"mean_token_accuracy": 0.12454378306865692,
"num_tokens": 3011844.0,
"step": 1075
},
{
"entropy": 6.313051700592041,
"epoch": 0.12864800476474092,
"grad_norm": 0.984375,
"learning_rate": 0.0004999989926553962,
"loss": 5.9951,
"mean_token_accuracy": 0.13137999251484872,
"num_tokens": 3024648.0,
"step": 1080
},
{
"entropy": 6.2854513168334964,
"epoch": 0.1292435973793925,
"grad_norm": 1.09375,
"learning_rate": 0.000499998861108343,
"loss": 5.9666,
"mean_token_accuracy": 0.12976680919528008,
"num_tokens": 3038533.0,
"step": 1085
},
{
"entropy": 6.290816593170166,
"epoch": 0.12983918999404406,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999987214909464,
"loss": 5.9063,
"mean_token_accuracy": 0.12990044206380844,
"num_tokens": 3052531.0,
"step": 1090
},
{
"entropy": 6.258984231948853,
"epoch": 0.13043478260869565,
"grad_norm": 1.0625,
"learning_rate": 0.0004999985738032115,
"loss": 6.0488,
"mean_token_accuracy": 0.12387203425168991,
"num_tokens": 3066526.0,
"step": 1095
},
{
"entropy": 6.306193065643311,
"epoch": 0.13103037522334723,
"grad_norm": 1.03125,
"learning_rate": 0.0004999984180451436,
"loss": 5.9604,
"mean_token_accuracy": 0.1295250229537487,
"num_tokens": 3080729.0,
"step": 1100
},
{
"entropy": 6.284939384460449,
"epoch": 0.13162596783799882,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999982542167483,
"loss": 6.0088,
"mean_token_accuracy": 0.13003896847367286,
"num_tokens": 3093776.0,
"step": 1105
},
{
"entropy": 6.197756433486939,
"epoch": 0.13222156045265038,
"grad_norm": 1.015625,
"learning_rate": 0.0004999980823180314,
"loss": 5.9306,
"mean_token_accuracy": 0.13206810057163237,
"num_tokens": 3107005.0,
"step": 1110
},
{
"entropy": 6.382076263427734,
"epoch": 0.13281715306730196,
"grad_norm": 0.9296875,
"learning_rate": 0.000499997902348999,
"loss": 6.1495,
"mean_token_accuracy": 0.12192723751068116,
"num_tokens": 3120840.0,
"step": 1115
},
{
"entropy": 6.247245407104492,
"epoch": 0.13341274568195355,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999977143096578,
"loss": 5.9423,
"mean_token_accuracy": 0.13738912492990493,
"num_tokens": 3134894.0,
"step": 1120
},
{
"entropy": 6.403128004074096,
"epoch": 0.13400833829660513,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999975182000143,
"loss": 6.1036,
"mean_token_accuracy": 0.12629491686820984,
"num_tokens": 3146917.0,
"step": 1125
},
{
"entropy": 6.3133220195770265,
"epoch": 0.1346039309112567,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999973140200759,
"loss": 6.0041,
"mean_token_accuracy": 0.12788599953055382,
"num_tokens": 3162161.0,
"step": 1130
},
{
"entropy": 6.192528200149536,
"epoch": 0.13519952352590828,
"grad_norm": 0.921875,
"learning_rate": 0.0004999971017698495,
"loss": 5.9295,
"mean_token_accuracy": 0.13294726982712746,
"num_tokens": 3176627.0,
"step": 1135
},
{
"entropy": 6.281828927993774,
"epoch": 0.13579511614055986,
"grad_norm": 1.078125,
"learning_rate": 0.0004999968814493428,
"loss": 5.9814,
"mean_token_accuracy": 0.12526661157608032,
"num_tokens": 3190139.0,
"step": 1140
},
{
"entropy": 6.122537755966187,
"epoch": 0.13639070875521145,
"grad_norm": 1.015625,
"learning_rate": 0.0004999966530585639,
"loss": 5.7922,
"mean_token_accuracy": 0.13596827089786528,
"num_tokens": 3202312.0,
"step": 1145
},
{
"entropy": 6.275762701034546,
"epoch": 0.136986301369863,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999964165975209,
"loss": 5.9466,
"mean_token_accuracy": 0.1306569755077362,
"num_tokens": 3219052.0,
"step": 1150
},
{
"entropy": 6.147595548629761,
"epoch": 0.1375818939845146,
"grad_norm": 1.015625,
"learning_rate": 0.0004999961720662221,
"loss": 5.8729,
"mean_token_accuracy": 0.13916604742407798,
"num_tokens": 3233706.0,
"step": 1155
},
{
"entropy": 6.3489350318908695,
"epoch": 0.13817748659916618,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999959194646766,
"loss": 6.0701,
"mean_token_accuracy": 0.12903773859143258,
"num_tokens": 3249947.0,
"step": 1160
},
{
"entropy": 6.191387891769409,
"epoch": 0.13877307921381773,
"grad_norm": 1.109375,
"learning_rate": 0.0004999956587928932,
"loss": 5.8834,
"mean_token_accuracy": 0.13957050070166588,
"num_tokens": 3264308.0,
"step": 1165
},
{
"entropy": 6.211583471298217,
"epoch": 0.13936867182846932,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999953900508814,
"loss": 5.8737,
"mean_token_accuracy": 0.14119818210601806,
"num_tokens": 3277171.0,
"step": 1170
},
{
"entropy": 6.219825363159179,
"epoch": 0.1399642644431209,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999951132386507,
"loss": 5.9632,
"mean_token_accuracy": 0.13040691390633583,
"num_tokens": 3290445.0,
"step": 1175
},
{
"entropy": 6.152277088165283,
"epoch": 0.1405598570577725,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999948283562111,
"loss": 5.8781,
"mean_token_accuracy": 0.13456293493509291,
"num_tokens": 3304850.0,
"step": 1180
},
{
"entropy": 6.18782114982605,
"epoch": 0.14115544967242405,
"grad_norm": 0.953125,
"learning_rate": 0.000499994535403573,
"loss": 5.9299,
"mean_token_accuracy": 0.1308791384100914,
"num_tokens": 3318765.0,
"step": 1185
},
{
"entropy": 6.285896110534668,
"epoch": 0.14175104228707563,
"grad_norm": 1.1875,
"learning_rate": 0.0004999942343807465,
"loss": 5.9196,
"mean_token_accuracy": 0.12730536982417107,
"num_tokens": 3330403.0,
"step": 1190
},
{
"entropy": 6.243529176712036,
"epoch": 0.14234663490172722,
"grad_norm": 0.875,
"learning_rate": 0.0004999939252877428,
"loss": 5.9123,
"mean_token_accuracy": 0.1316504694521427,
"num_tokens": 3346420.0,
"step": 1195
},
{
"entropy": 6.200117015838623,
"epoch": 0.1429422275163788,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999936081245727,
"loss": 6.0073,
"mean_token_accuracy": 0.12813543006777764,
"num_tokens": 3361853.0,
"step": 1200
},
{
"entropy": 6.207744121551514,
"epoch": 0.14353782013103036,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999932828912478,
"loss": 5.9446,
"mean_token_accuracy": 0.1302527032792568,
"num_tokens": 3375907.0,
"step": 1205
},
{
"entropy": 6.263318586349487,
"epoch": 0.14413341274568195,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999929495877796,
"loss": 5.9602,
"mean_token_accuracy": 0.1266719475388527,
"num_tokens": 3390151.0,
"step": 1210
},
{
"entropy": 6.158126878738403,
"epoch": 0.14472900536033353,
"grad_norm": 1.0,
"learning_rate": 0.0004999926082141801,
"loss": 5.9592,
"mean_token_accuracy": 0.12761132046580315,
"num_tokens": 3405151.0,
"step": 1215
},
{
"entropy": 6.298018312454223,
"epoch": 0.14532459797498512,
"grad_norm": 0.96875,
"learning_rate": 0.0004999922587704616,
"loss": 5.9751,
"mean_token_accuracy": 0.13020757883787154,
"num_tokens": 3420510.0,
"step": 1220
},
{
"entropy": 6.109394550323486,
"epoch": 0.14592019058963668,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999919012566366,
"loss": 5.8712,
"mean_token_accuracy": 0.13398747742176056,
"num_tokens": 3435162.0,
"step": 1225
},
{
"entropy": 6.261184501647949,
"epoch": 0.14651578320428826,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999915356727179,
"loss": 5.9816,
"mean_token_accuracy": 0.1259373977780342,
"num_tokens": 3450923.0,
"step": 1230
},
{
"entropy": 6.163080024719238,
"epoch": 0.14711137581893985,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999911620187186,
"loss": 5.9163,
"mean_token_accuracy": 0.13199557438492776,
"num_tokens": 3464575.0,
"step": 1235
},
{
"entropy": 6.13615665435791,
"epoch": 0.14770696843359143,
"grad_norm": 1.140625,
"learning_rate": 0.0004999907802946521,
"loss": 5.815,
"mean_token_accuracy": 0.13966626226902007,
"num_tokens": 3477494.0,
"step": 1240
},
{
"entropy": 6.3183629512786865,
"epoch": 0.148302561048243,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999903905005323,
"loss": 6.0528,
"mean_token_accuracy": 0.12290267571806908,
"num_tokens": 3492314.0,
"step": 1245
},
{
"entropy": 6.17548680305481,
"epoch": 0.14889815366289458,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999899926363729,
"loss": 5.8749,
"mean_token_accuracy": 0.13766821697354317,
"num_tokens": 3504913.0,
"step": 1250
},
{
"entropy": 6.311449909210205,
"epoch": 0.14949374627754616,
"grad_norm": 0.96875,
"learning_rate": 0.0004999895867021882,
"loss": 6.0175,
"mean_token_accuracy": 0.12477483078837395,
"num_tokens": 3519758.0,
"step": 1255
},
{
"entropy": 6.139519739151001,
"epoch": 0.15008933889219775,
"grad_norm": 1.140625,
"learning_rate": 0.0004999891726979928,
"loss": 5.8116,
"mean_token_accuracy": 0.14197972118854524,
"num_tokens": 3531268.0,
"step": 1260
},
{
"entropy": 6.2304911613464355,
"epoch": 0.1506849315068493,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999887506238018,
"loss": 5.9804,
"mean_token_accuracy": 0.12250895500183105,
"num_tokens": 3546560.0,
"step": 1265
},
{
"entropy": 6.161889743804932,
"epoch": 0.1512805241215009,
"grad_norm": 1.0078125,
"learning_rate": 0.00049998832047963,
"loss": 5.9035,
"mean_token_accuracy": 0.13964627236127852,
"num_tokens": 3560602.0,
"step": 1270
},
{
"entropy": 6.108182716369629,
"epoch": 0.15187611673615248,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999878822654929,
"loss": 5.8591,
"mean_token_accuracy": 0.12805922999978064,
"num_tokens": 3575611.0,
"step": 1275
},
{
"entropy": 6.155970811843872,
"epoch": 0.15247170935080406,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999874359814064,
"loss": 5.8568,
"mean_token_accuracy": 0.1279609300196171,
"num_tokens": 3589933.0,
"step": 1280
},
{
"entropy": 6.128014087677002,
"epoch": 0.15306730196545562,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999869816273863,
"loss": 5.8852,
"mean_token_accuracy": 0.1352235957980156,
"num_tokens": 3602513.0,
"step": 1285
},
{
"entropy": 6.215742731094361,
"epoch": 0.1536628945801072,
"grad_norm": 1.0625,
"learning_rate": 0.0004999865192034489,
"loss": 5.9359,
"mean_token_accuracy": 0.1263606384396553,
"num_tokens": 3615488.0,
"step": 1290
},
{
"entropy": 6.0860411643981935,
"epoch": 0.1542584871947588,
"grad_norm": 0.94921875,
"learning_rate": 0.000499986048709611,
"loss": 5.8016,
"mean_token_accuracy": 0.13899177685379982,
"num_tokens": 3629292.0,
"step": 1295
},
{
"entropy": 6.102784109115601,
"epoch": 0.15485407980941035,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999855701458893,
"loss": 5.8263,
"mean_token_accuracy": 0.1349021963775158,
"num_tokens": 3647010.0,
"step": 1300
},
{
"entropy": 6.094648599624634,
"epoch": 0.15544967242406194,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999850835123009,
"loss": 5.8472,
"mean_token_accuracy": 0.14222987964749337,
"num_tokens": 3661994.0,
"step": 1305
},
{
"entropy": 6.2444103240966795,
"epoch": 0.15604526503871352,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999845888088635,
"loss": 5.9105,
"mean_token_accuracy": 0.1391028083860874,
"num_tokens": 3676789.0,
"step": 1310
},
{
"entropy": 6.099581623077393,
"epoch": 0.1566408576533651,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999840860355946,
"loss": 5.8413,
"mean_token_accuracy": 0.1327517084777355,
"num_tokens": 3689598.0,
"step": 1315
},
{
"entropy": 6.126536417007446,
"epoch": 0.15723645026801666,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999835751925124,
"loss": 5.9387,
"mean_token_accuracy": 0.13856274411082267,
"num_tokens": 3703597.0,
"step": 1320
},
{
"entropy": 6.202872657775879,
"epoch": 0.15783204288266825,
"grad_norm": 1.0,
"learning_rate": 0.0004999830562796351,
"loss": 5.8886,
"mean_token_accuracy": 0.1350098006427288,
"num_tokens": 3717312.0,
"step": 1325
},
{
"entropy": 6.095597791671753,
"epoch": 0.15842763549731984,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999825292969814,
"loss": 5.8198,
"mean_token_accuracy": 0.13469851538538932,
"num_tokens": 3731677.0,
"step": 1330
},
{
"entropy": 6.175401449203491,
"epoch": 0.15902322811197142,
"grad_norm": 0.890625,
"learning_rate": 0.0004999819942445703,
"loss": 5.9277,
"mean_token_accuracy": 0.13118114992976188,
"num_tokens": 3746648.0,
"step": 1335
},
{
"entropy": 6.046168470382691,
"epoch": 0.15961882072662298,
"grad_norm": 1.15625,
"learning_rate": 0.0004999814511224207,
"loss": 5.8116,
"mean_token_accuracy": 0.13979229629039763,
"num_tokens": 3760859.0,
"step": 1340
},
{
"entropy": 6.1087652206420895,
"epoch": 0.16021441334127456,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999808999305523,
"loss": 5.7699,
"mean_token_accuracy": 0.14346203953027725,
"num_tokens": 3774743.0,
"step": 1345
},
{
"entropy": 6.108399534225464,
"epoch": 0.16081000595592615,
"grad_norm": 0.90625,
"learning_rate": 0.0004999803406689848,
"loss": 5.8496,
"mean_token_accuracy": 0.13659733533859253,
"num_tokens": 3789566.0,
"step": 1350
},
{
"entropy": 6.125910091400146,
"epoch": 0.16140559857057774,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999797733377384,
"loss": 5.8376,
"mean_token_accuracy": 0.13647732883691788,
"num_tokens": 3802875.0,
"step": 1355
},
{
"entropy": 6.075304651260376,
"epoch": 0.1620011911852293,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999791979368332,
"loss": 5.8069,
"mean_token_accuracy": 0.1374050907790661,
"num_tokens": 3816197.0,
"step": 1360
},
{
"entropy": 6.018436431884766,
"epoch": 0.16259678379988088,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999786144662899,
"loss": 5.8232,
"mean_token_accuracy": 0.14218598753213882,
"num_tokens": 3829196.0,
"step": 1365
},
{
"entropy": 6.249544095993042,
"epoch": 0.16319237641453246,
"grad_norm": 0.953125,
"learning_rate": 0.0004999780229261295,
"loss": 5.9389,
"mean_token_accuracy": 0.13307114019989968,
"num_tokens": 3842618.0,
"step": 1370
},
{
"entropy": 6.1407723903656,
"epoch": 0.16378796902918405,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999774233163734,
"loss": 5.9284,
"mean_token_accuracy": 0.1387593887746334,
"num_tokens": 3854833.0,
"step": 1375
},
{
"entropy": 6.195996522903442,
"epoch": 0.1643835616438356,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999768156370428,
"loss": 5.8234,
"mean_token_accuracy": 0.14503343403339386,
"num_tokens": 3867971.0,
"step": 1380
},
{
"entropy": 6.054100847244262,
"epoch": 0.1649791542584872,
"grad_norm": 1.140625,
"learning_rate": 0.0004999761998881596,
"loss": 5.7758,
"mean_token_accuracy": 0.14200711995363235,
"num_tokens": 3881555.0,
"step": 1385
},
{
"entropy": 6.100306701660156,
"epoch": 0.16557474687313878,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999755760697459,
"loss": 5.7914,
"mean_token_accuracy": 0.14153566658496858,
"num_tokens": 3894717.0,
"step": 1390
},
{
"entropy": 6.075147581100464,
"epoch": 0.16617033948779036,
"grad_norm": 0.97265625,
"learning_rate": 0.000499974944181824,
"loss": 5.8292,
"mean_token_accuracy": 0.13094881400465966,
"num_tokens": 3909381.0,
"step": 1395
},
{
"entropy": 6.10245189666748,
"epoch": 0.16676593210244192,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999743042244167,
"loss": 5.8371,
"mean_token_accuracy": 0.1395154632627964,
"num_tokens": 3923828.0,
"step": 1400
},
{
"entropy": 5.967838907241822,
"epoch": 0.1673615247170935,
"grad_norm": 1.125,
"learning_rate": 0.000499973656197547,
"loss": 5.7351,
"mean_token_accuracy": 0.14525343626737594,
"num_tokens": 3936901.0,
"step": 1405
},
{
"entropy": 6.132982015609741,
"epoch": 0.1679571173317451,
"grad_norm": 1.03125,
"learning_rate": 0.0004999730001012379,
"loss": 5.8213,
"mean_token_accuracy": 0.14513242095708848,
"num_tokens": 3949052.0,
"step": 1410
},
{
"entropy": 6.045504951477051,
"epoch": 0.16855270994639668,
"grad_norm": 1.109375,
"learning_rate": 0.0004999723359355131,
"loss": 5.8714,
"mean_token_accuracy": 0.13704032599925994,
"num_tokens": 3963099.0,
"step": 1415
},
{
"entropy": 6.16231861114502,
"epoch": 0.16914830256104824,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999716637003965,
"loss": 5.8014,
"mean_token_accuracy": 0.1322947271168232,
"num_tokens": 3977598.0,
"step": 1420
},
{
"entropy": 6.128900623321533,
"epoch": 0.16974389517569982,
"grad_norm": 1.0390625,
"learning_rate": 0.000499970983395912,
"loss": 5.9456,
"mean_token_accuracy": 0.13798302561044692,
"num_tokens": 3989995.0,
"step": 1425
},
{
"entropy": 6.172013568878174,
"epoch": 0.1703394877903514,
"grad_norm": 0.984375,
"learning_rate": 0.0004999702950220841,
"loss": 5.8439,
"mean_token_accuracy": 0.13327042013406754,
"num_tokens": 4002971.0,
"step": 1430
},
{
"entropy": 6.017853879928589,
"epoch": 0.17093508040500296,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999695985789376,
"loss": 5.7647,
"mean_token_accuracy": 0.13795410469174385,
"num_tokens": 4017380.0,
"step": 1435
},
{
"entropy": 6.001563310623169,
"epoch": 0.17153067301965455,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999688940664972,
"loss": 5.8366,
"mean_token_accuracy": 0.13615191131830215,
"num_tokens": 4031125.0,
"step": 1440
},
{
"entropy": 6.0987037181854244,
"epoch": 0.17212626563430614,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999681814847884,
"loss": 5.8352,
"mean_token_accuracy": 0.1364310458302498,
"num_tokens": 4046416.0,
"step": 1445
},
{
"entropy": 6.109888076782227,
"epoch": 0.17272185824895772,
"grad_norm": 0.96875,
"learning_rate": 0.0004999674608338367,
"loss": 5.8641,
"mean_token_accuracy": 0.1359500080347061,
"num_tokens": 4061024.0,
"step": 1450
},
{
"entropy": 6.057561159133911,
"epoch": 0.17331745086360928,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999667321136681,
"loss": 5.7771,
"mean_token_accuracy": 0.1439843602478504,
"num_tokens": 4074753.0,
"step": 1455
},
{
"entropy": 6.0372072696685795,
"epoch": 0.17391304347826086,
"grad_norm": 0.9375,
"learning_rate": 0.0004999659953243085,
"loss": 5.7668,
"mean_token_accuracy": 0.13691218346357345,
"num_tokens": 4089954.0,
"step": 1460
},
{
"entropy": 6.042539405822754,
"epoch": 0.17450863609291245,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999652504657844,
"loss": 5.7714,
"mean_token_accuracy": 0.14127889573574065,
"num_tokens": 4105220.0,
"step": 1465
},
{
"entropy": 6.025610017776489,
"epoch": 0.17510422870756404,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999644975381226,
"loss": 5.7751,
"mean_token_accuracy": 0.14062938392162322,
"num_tokens": 4117578.0,
"step": 1470
},
{
"entropy": 6.0217842102050785,
"epoch": 0.1756998213222156,
"grad_norm": 1.0625,
"learning_rate": 0.00049996373654135,
"loss": 5.6942,
"mean_token_accuracy": 0.13991036266088486,
"num_tokens": 4131113.0,
"step": 1475
},
{
"entropy": 5.923555183410644,
"epoch": 0.17629541393686718,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999629674754939,
"loss": 5.6696,
"mean_token_accuracy": 0.14742299616336824,
"num_tokens": 4145029.0,
"step": 1480
},
{
"entropy": 6.077828407287598,
"epoch": 0.17689100655151876,
"grad_norm": 1.140625,
"learning_rate": 0.0004999621903405819,
"loss": 5.8136,
"mean_token_accuracy": 0.14328548908233643,
"num_tokens": 4157227.0,
"step": 1485
},
{
"entropy": 5.968223714828492,
"epoch": 0.17748659916617035,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999614051366421,
"loss": 5.7978,
"mean_token_accuracy": 0.14207359179854392,
"num_tokens": 4171172.0,
"step": 1490
},
{
"entropy": 6.060792303085327,
"epoch": 0.1780821917808219,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999606118637022,
"loss": 5.7271,
"mean_token_accuracy": 0.1467897191643715,
"num_tokens": 4184219.0,
"step": 1495
},
{
"entropy": 5.948859405517578,
"epoch": 0.1786777843954735,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999598105217911,
"loss": 5.6832,
"mean_token_accuracy": 0.1476665310561657,
"num_tokens": 4199317.0,
"step": 1500
},
{
"entropy": 6.105476236343383,
"epoch": 0.17927337701012508,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999590011109373,
"loss": 5.8373,
"mean_token_accuracy": 0.13571951538324356,
"num_tokens": 4213905.0,
"step": 1505
},
{
"entropy": 6.026069593429566,
"epoch": 0.17986896962477666,
"grad_norm": 0.9375,
"learning_rate": 0.00049995818363117,
"loss": 5.7593,
"mean_token_accuracy": 0.1381727173924446,
"num_tokens": 4228123.0,
"step": 1510
},
{
"entropy": 6.0289146900177,
"epoch": 0.18046456223942822,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999573580825182,
"loss": 5.7348,
"mean_token_accuracy": 0.14089103564620017,
"num_tokens": 4243024.0,
"step": 1515
},
{
"entropy": 5.97723650932312,
"epoch": 0.1810601548540798,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999565244650119,
"loss": 5.7316,
"mean_token_accuracy": 0.14283815920352935,
"num_tokens": 4259061.0,
"step": 1520
},
{
"entropy": 6.053089904785156,
"epoch": 0.1816557474687314,
"grad_norm": 1.265625,
"learning_rate": 0.0004999556827786807,
"loss": 5.906,
"mean_token_accuracy": 0.13426867201924325,
"num_tokens": 4273010.0,
"step": 1525
},
{
"entropy": 6.069848442077637,
"epoch": 0.18225134008338298,
"grad_norm": 1.0625,
"learning_rate": 0.0004999548330235549,
"loss": 5.705,
"mean_token_accuracy": 0.14787458628416061,
"num_tokens": 4286515.0,
"step": 1530
},
{
"entropy": 5.92649474143982,
"epoch": 0.18284693269803454,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999539751996651,
"loss": 5.6984,
"mean_token_accuracy": 0.14708176404237747,
"num_tokens": 4300125.0,
"step": 1535
},
{
"entropy": 5.960150051116943,
"epoch": 0.18344252531268612,
"grad_norm": 0.9453125,
"learning_rate": 0.000499953109307042,
"loss": 5.7101,
"mean_token_accuracy": 0.1453533485531807,
"num_tokens": 4313962.0,
"step": 1540
},
{
"entropy": 5.9575080394744875,
"epoch": 0.1840381179273377,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999522353457165,
"loss": 5.7541,
"mean_token_accuracy": 0.14509638547897338,
"num_tokens": 4328776.0,
"step": 1545
},
{
"entropy": 6.06169753074646,
"epoch": 0.18463371054198927,
"grad_norm": 1.171875,
"learning_rate": 0.0004999513533157201,
"loss": 5.7823,
"mean_token_accuracy": 0.1447771780192852,
"num_tokens": 4340933.0,
"step": 1550
},
{
"entropy": 6.044741296768189,
"epoch": 0.18522930315664085,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999504632170844,
"loss": 5.74,
"mean_token_accuracy": 0.13476346135139466,
"num_tokens": 4355463.0,
"step": 1555
},
{
"entropy": 6.017773771286011,
"epoch": 0.18582489577129244,
"grad_norm": 1.078125,
"learning_rate": 0.0004999495650498413,
"loss": 5.7778,
"mean_token_accuracy": 0.14227432161569595,
"num_tokens": 4369717.0,
"step": 1560
},
{
"entropy": 6.01103105545044,
"epoch": 0.18642048838594402,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999486588140231,
"loss": 5.7738,
"mean_token_accuracy": 0.14884234070777894,
"num_tokens": 4383713.0,
"step": 1565
},
{
"entropy": 5.982986640930176,
"epoch": 0.18701608100059558,
"grad_norm": 1.015625,
"learning_rate": 0.0004999477445096623,
"loss": 5.7041,
"mean_token_accuracy": 0.14456475526094437,
"num_tokens": 4396487.0,
"step": 1570
},
{
"entropy": 6.072724056243897,
"epoch": 0.18761167361524717,
"grad_norm": 1.125,
"learning_rate": 0.0004999468221367916,
"loss": 5.8018,
"mean_token_accuracy": 0.14579196870326996,
"num_tokens": 4408762.0,
"step": 1575
},
{
"entropy": 5.907863330841065,
"epoch": 0.18820726622989875,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999458916954442,
"loss": 5.7009,
"mean_token_accuracy": 0.14011288806796074,
"num_tokens": 4422871.0,
"step": 1580
},
{
"entropy": 5.988130187988281,
"epoch": 0.18880285884455034,
"grad_norm": 1.0625,
"learning_rate": 0.0004999449531856532,
"loss": 5.7507,
"mean_token_accuracy": 0.14636473283171653,
"num_tokens": 4436399.0,
"step": 1585
},
{
"entropy": 5.935398817062378,
"epoch": 0.1893984514592019,
"grad_norm": 1.078125,
"learning_rate": 0.0004999440066074526,
"loss": 5.6411,
"mean_token_accuracy": 0.14530467391014099,
"num_tokens": 4450799.0,
"step": 1590
},
{
"entropy": 5.8840916633605955,
"epoch": 0.18999404407385348,
"grad_norm": 1.015625,
"learning_rate": 0.0004999430519608762,
"loss": 5.6902,
"mean_token_accuracy": 0.15209779292345046,
"num_tokens": 4464567.0,
"step": 1595
},
{
"entropy": 6.023315811157227,
"epoch": 0.19058963668850507,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999420892459584,
"loss": 5.7492,
"mean_token_accuracy": 0.1440256230533123,
"num_tokens": 4478461.0,
"step": 1600
},
{
"entropy": 5.8960765361785885,
"epoch": 0.19118522930315665,
"grad_norm": 0.96875,
"learning_rate": 0.0004999411184627334,
"loss": 5.6975,
"mean_token_accuracy": 0.1430221252143383,
"num_tokens": 4493252.0,
"step": 1605
},
{
"entropy": 5.975708055496216,
"epoch": 0.1917808219178082,
"grad_norm": 0.90625,
"learning_rate": 0.0004999401396112364,
"loss": 5.7855,
"mean_token_accuracy": 0.1421280398964882,
"num_tokens": 4507716.0,
"step": 1610
},
{
"entropy": 5.9918951988220215,
"epoch": 0.1923764145324598,
"grad_norm": 1.1171875,
"learning_rate": 0.000499939152691502,
"loss": 5.7122,
"mean_token_accuracy": 0.14283654913306237,
"num_tokens": 4520903.0,
"step": 1615
},
{
"entropy": 5.9884819984436035,
"epoch": 0.19297200714711138,
"grad_norm": 1.015625,
"learning_rate": 0.0004999381577035663,
"loss": 5.757,
"mean_token_accuracy": 0.14660940766334535,
"num_tokens": 4534939.0,
"step": 1620
},
{
"entropy": 5.941656112670898,
"epoch": 0.19356759976176297,
"grad_norm": 1.078125,
"learning_rate": 0.0004999371546474643,
"loss": 5.7185,
"mean_token_accuracy": 0.14215019568800927,
"num_tokens": 4547502.0,
"step": 1625
},
{
"entropy": 5.985489463806152,
"epoch": 0.19416319237641452,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999361435232324,
"loss": 5.6273,
"mean_token_accuracy": 0.1540282890200615,
"num_tokens": 4560531.0,
"step": 1630
},
{
"entropy": 5.896367597579956,
"epoch": 0.1947587849910661,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999351243309069,
"loss": 5.7045,
"mean_token_accuracy": 0.14529286175966263,
"num_tokens": 4573678.0,
"step": 1635
},
{
"entropy": 5.953095293045044,
"epoch": 0.1953543776057177,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999340970705241,
"loss": 5.5663,
"mean_token_accuracy": 0.15338113605976106,
"num_tokens": 4586313.0,
"step": 1640
},
{
"entropy": 5.876391077041626,
"epoch": 0.19594997022036928,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999330617421209,
"loss": 5.6918,
"mean_token_accuracy": 0.14725659191608428,
"num_tokens": 4600030.0,
"step": 1645
},
{
"entropy": 5.9935619831085205,
"epoch": 0.19654556283502084,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999320183457346,
"loss": 5.7659,
"mean_token_accuracy": 0.14525030702352523,
"num_tokens": 4613289.0,
"step": 1650
},
{
"entropy": 5.943365240097046,
"epoch": 0.19714115544967242,
"grad_norm": 1.21875,
"learning_rate": 0.0004999309668814024,
"loss": 5.6493,
"mean_token_accuracy": 0.14853100031614302,
"num_tokens": 4626642.0,
"step": 1655
},
{
"entropy": 5.986563062667846,
"epoch": 0.197736748064324,
"grad_norm": 1.015625,
"learning_rate": 0.0004999299073491623,
"loss": 5.7421,
"mean_token_accuracy": 0.14294660985469818,
"num_tokens": 4639750.0,
"step": 1660
},
{
"entropy": 5.8959746837615965,
"epoch": 0.1983323406789756,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999288397490519,
"loss": 5.6408,
"mean_token_accuracy": 0.14631985872983932,
"num_tokens": 4656894.0,
"step": 1665
},
{
"entropy": 5.864677000045776,
"epoch": 0.19892793329362715,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999277640811099,
"loss": 5.5882,
"mean_token_accuracy": 0.15125427693128585,
"num_tokens": 4670229.0,
"step": 1670
},
{
"entropy": 5.917145919799805,
"epoch": 0.19952352590827874,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999266803453747,
"loss": 5.7195,
"mean_token_accuracy": 0.1474950507283211,
"num_tokens": 4683754.0,
"step": 1675
},
{
"entropy": 5.861428308486938,
"epoch": 0.20011911852293032,
"grad_norm": 1.09375,
"learning_rate": 0.0004999255885418852,
"loss": 5.6733,
"mean_token_accuracy": 0.14484547674655915,
"num_tokens": 4697824.0,
"step": 1680
},
{
"entropy": 6.056915712356568,
"epoch": 0.20071471113758188,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999244886706804,
"loss": 5.676,
"mean_token_accuracy": 0.14596075415611268,
"num_tokens": 4711004.0,
"step": 1685
},
{
"entropy": 5.886464977264405,
"epoch": 0.20131030375223347,
"grad_norm": 0.984375,
"learning_rate": 0.0004999233807318001,
"loss": 5.6237,
"mean_token_accuracy": 0.1506524845957756,
"num_tokens": 4724992.0,
"step": 1690
},
{
"entropy": 5.93082103729248,
"epoch": 0.20190589636688505,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999222647252837,
"loss": 5.7457,
"mean_token_accuracy": 0.14679386615753173,
"num_tokens": 4737770.0,
"step": 1695
},
{
"entropy": 5.982219123840332,
"epoch": 0.20250148898153664,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999211406511714,
"loss": 5.6381,
"mean_token_accuracy": 0.1514803260564804,
"num_tokens": 4753807.0,
"step": 1700
},
{
"entropy": 5.760978889465332,
"epoch": 0.2030970815961882,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999200085095035,
"loss": 5.6044,
"mean_token_accuracy": 0.15484319478273392,
"num_tokens": 4766294.0,
"step": 1705
},
{
"entropy": 5.888656377792358,
"epoch": 0.20369267421083978,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999188683003207,
"loss": 5.6105,
"mean_token_accuracy": 0.15029520839452742,
"num_tokens": 4780143.0,
"step": 1710
},
{
"entropy": 5.906369304656982,
"epoch": 0.20428826682549137,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999177200236635,
"loss": 5.6624,
"mean_token_accuracy": 0.14471479430794715,
"num_tokens": 4793717.0,
"step": 1715
},
{
"entropy": 6.035571956634522,
"epoch": 0.20488385944014295,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999165636795736,
"loss": 5.8106,
"mean_token_accuracy": 0.1430202305316925,
"num_tokens": 4808925.0,
"step": 1720
},
{
"entropy": 5.839449691772461,
"epoch": 0.2054794520547945,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999153992680923,
"loss": 5.6407,
"mean_token_accuracy": 0.1487026743590832,
"num_tokens": 4821439.0,
"step": 1725
},
{
"entropy": 5.922779560089111,
"epoch": 0.2060750446694461,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999142267892612,
"loss": 5.5719,
"mean_token_accuracy": 0.15140777006745337,
"num_tokens": 4835511.0,
"step": 1730
},
{
"entropy": 5.826804113388062,
"epoch": 0.20667063728409768,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999130462431224,
"loss": 5.6024,
"mean_token_accuracy": 0.1572335720062256,
"num_tokens": 4849500.0,
"step": 1735
},
{
"entropy": 5.848100852966309,
"epoch": 0.20726622989874927,
"grad_norm": 1.03125,
"learning_rate": 0.0004999118576297185,
"loss": 5.669,
"mean_token_accuracy": 0.1487742006778717,
"num_tokens": 4863372.0,
"step": 1740
},
{
"entropy": 5.968707418441772,
"epoch": 0.20786182251340082,
"grad_norm": 1.0,
"learning_rate": 0.0004999106609490918,
"loss": 5.7082,
"mean_token_accuracy": 0.13946339711546898,
"num_tokens": 4876643.0,
"step": 1745
},
{
"entropy": 5.929504442214966,
"epoch": 0.2084574151280524,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999094562012855,
"loss": 5.6743,
"mean_token_accuracy": 0.14708083271980285,
"num_tokens": 4890274.0,
"step": 1750
},
{
"entropy": 5.810866594314575,
"epoch": 0.209053007742704,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999082433863426,
"loss": 5.5208,
"mean_token_accuracy": 0.15671751648187637,
"num_tokens": 4905643.0,
"step": 1755
},
{
"entropy": 5.809827089309692,
"epoch": 0.20964860035735558,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999070225043068,
"loss": 5.6695,
"mean_token_accuracy": 0.14416072219610215,
"num_tokens": 4920106.0,
"step": 1760
},
{
"entropy": 5.933014345169068,
"epoch": 0.21024419297200714,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999057935552217,
"loss": 5.5906,
"mean_token_accuracy": 0.1499159552156925,
"num_tokens": 4934498.0,
"step": 1765
},
{
"entropy": 5.884340429306031,
"epoch": 0.21083978558665872,
"grad_norm": 1.140625,
"learning_rate": 0.0004999045565391313,
"loss": 5.6878,
"mean_token_accuracy": 0.142864453792572,
"num_tokens": 4947562.0,
"step": 1770
},
{
"entropy": 5.923207092285156,
"epoch": 0.2114353782013103,
"grad_norm": 0.953125,
"learning_rate": 0.0004999033114560803,
"loss": 5.6892,
"mean_token_accuracy": 0.14773986637592315,
"num_tokens": 4962375.0,
"step": 1775
},
{
"entropy": 5.941981983184815,
"epoch": 0.2120309708159619,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999020583061132,
"loss": 5.7048,
"mean_token_accuracy": 0.1469015933573246,
"num_tokens": 4975683.0,
"step": 1780
},
{
"entropy": 5.880807876586914,
"epoch": 0.21262656343061345,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999007970892749,
"loss": 5.7095,
"mean_token_accuracy": 0.14753714501857756,
"num_tokens": 4989662.0,
"step": 1785
},
{
"entropy": 5.906244277954102,
"epoch": 0.21322215604526504,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998995278056107,
"loss": 5.5762,
"mean_token_accuracy": 0.155949142575264,
"num_tokens": 5003778.0,
"step": 1790
},
{
"entropy": 5.878344297409058,
"epoch": 0.21381774865991662,
"grad_norm": 1.046875,
"learning_rate": 0.0004998982504551661,
"loss": 5.69,
"mean_token_accuracy": 0.14796825796365737,
"num_tokens": 5016556.0,
"step": 1795
},
{
"entropy": 5.8604559898376465,
"epoch": 0.2144133412745682,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998969650379869,
"loss": 5.5695,
"mean_token_accuracy": 0.15356758683919908,
"num_tokens": 5029732.0,
"step": 1800
},
{
"entropy": 5.7357566356658936,
"epoch": 0.21500893388921977,
"grad_norm": 1.25,
"learning_rate": 0.0004998956715541192,
"loss": 5.5173,
"mean_token_accuracy": 0.15466021448373796,
"num_tokens": 5042123.0,
"step": 1805
},
{
"entropy": 5.880379962921142,
"epoch": 0.21560452650387135,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998943700036095,
"loss": 5.6246,
"mean_token_accuracy": 0.14878395795822144,
"num_tokens": 5055537.0,
"step": 1810
},
{
"entropy": 5.866318416595459,
"epoch": 0.21620011911852294,
"grad_norm": 1.078125,
"learning_rate": 0.0004998930603865042,
"loss": 5.6587,
"mean_token_accuracy": 0.15060170888900756,
"num_tokens": 5068795.0,
"step": 1815
},
{
"entropy": 5.895635652542114,
"epoch": 0.2167957117331745,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998917427028507,
"loss": 5.6534,
"mean_token_accuracy": 0.1463247373700142,
"num_tokens": 5083149.0,
"step": 1820
},
{
"entropy": 5.872237825393677,
"epoch": 0.21739130434782608,
"grad_norm": 1.0,
"learning_rate": 0.0004998904169526959,
"loss": 5.5776,
"mean_token_accuracy": 0.15161531865596772,
"num_tokens": 5096879.0,
"step": 1825
},
{
"entropy": 5.839275884628296,
"epoch": 0.21798689696247767,
"grad_norm": 0.9375,
"learning_rate": 0.0004998890831360876,
"loss": 5.5538,
"mean_token_accuracy": 0.15384691804647446,
"num_tokens": 5110692.0,
"step": 1830
},
{
"entropy": 5.84959192276001,
"epoch": 0.21858248957712925,
"grad_norm": 1.171875,
"learning_rate": 0.0004998877412530735,
"loss": 5.6607,
"mean_token_accuracy": 0.14749133214354515,
"num_tokens": 5125981.0,
"step": 1835
},
{
"entropy": 5.854062128067016,
"epoch": 0.2191780821917808,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998863913037017,
"loss": 5.6106,
"mean_token_accuracy": 0.15050003081560134,
"num_tokens": 5138755.0,
"step": 1840
},
{
"entropy": 5.971685457229614,
"epoch": 0.2197736748064324,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998850332880208,
"loss": 5.7519,
"mean_token_accuracy": 0.14192944541573524,
"num_tokens": 5154208.0,
"step": 1845
},
{
"entropy": 5.9124468803405765,
"epoch": 0.22036926742108398,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998836672060795,
"loss": 5.7997,
"mean_token_accuracy": 0.1436879187822342,
"num_tokens": 5169633.0,
"step": 1850
},
{
"entropy": 5.902647590637207,
"epoch": 0.22096486003573557,
"grad_norm": 0.90625,
"learning_rate": 0.0004998822930579264,
"loss": 5.6791,
"mean_token_accuracy": 0.1411766916513443,
"num_tokens": 5186213.0,
"step": 1855
},
{
"entropy": 5.956612014770508,
"epoch": 0.22156045265038712,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998809108436112,
"loss": 5.6828,
"mean_token_accuracy": 0.15168982446193696,
"num_tokens": 5201171.0,
"step": 1860
},
{
"entropy": 5.78167986869812,
"epoch": 0.2221560452650387,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998795205631834,
"loss": 5.625,
"mean_token_accuracy": 0.15614613145589828,
"num_tokens": 5216669.0,
"step": 1865
},
{
"entropy": 6.015679883956909,
"epoch": 0.2227516378796903,
"grad_norm": 1.03125,
"learning_rate": 0.0004998781222166929,
"loss": 5.7214,
"mean_token_accuracy": 0.1417037233710289,
"num_tokens": 5229774.0,
"step": 1870
},
{
"entropy": 5.945243072509766,
"epoch": 0.22334723049434188,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998767158041897,
"loss": 5.6911,
"mean_token_accuracy": 0.14587639272212982,
"num_tokens": 5241862.0,
"step": 1875
},
{
"entropy": 5.823768424987793,
"epoch": 0.22394282310899344,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998753013257242,
"loss": 5.631,
"mean_token_accuracy": 0.15290849953889846,
"num_tokens": 5255768.0,
"step": 1880
},
{
"entropy": 5.861085605621338,
"epoch": 0.22453841572364502,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998738787813475,
"loss": 5.5694,
"mean_token_accuracy": 0.15960920006036758,
"num_tokens": 5268472.0,
"step": 1885
},
{
"entropy": 5.774131011962891,
"epoch": 0.2251340083382966,
"grad_norm": 0.890625,
"learning_rate": 0.0004998724481711103,
"loss": 5.5279,
"mean_token_accuracy": 0.15801258087158204,
"num_tokens": 5284309.0,
"step": 1890
},
{
"entropy": 5.877685642242431,
"epoch": 0.2257296009529482,
"grad_norm": 1.0546875,
"learning_rate": 0.000499871009495064,
"loss": 5.6587,
"mean_token_accuracy": 0.1461619719862938,
"num_tokens": 5298559.0,
"step": 1895
},
{
"entropy": 5.879857110977173,
"epoch": 0.22632519356759975,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998695627532601,
"loss": 5.582,
"mean_token_accuracy": 0.1536956936120987,
"num_tokens": 5312024.0,
"step": 1900
},
{
"entropy": 5.802623462677002,
"epoch": 0.22692078618225134,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998681079457507,
"loss": 5.5822,
"mean_token_accuracy": 0.15076420307159424,
"num_tokens": 5326848.0,
"step": 1905
},
{
"entropy": 5.7987607479095455,
"epoch": 0.22751637879690292,
"grad_norm": 1.125,
"learning_rate": 0.000499866645072588,
"loss": 5.5381,
"mean_token_accuracy": 0.15402921885252,
"num_tokens": 5341382.0,
"step": 1910
},
{
"entropy": 5.8349268436431885,
"epoch": 0.2281119714115545,
"grad_norm": 1.09375,
"learning_rate": 0.0004998651741338242,
"loss": 5.7287,
"mean_token_accuracy": 0.14886656552553176,
"num_tokens": 5355477.0,
"step": 1915
},
{
"entropy": 5.873121738433838,
"epoch": 0.22870756402620607,
"grad_norm": 0.96875,
"learning_rate": 0.0004998636951295122,
"loss": 5.5967,
"mean_token_accuracy": 0.1476642370223999,
"num_tokens": 5369271.0,
"step": 1920
},
{
"entropy": 5.8935792446136475,
"epoch": 0.22930315664085765,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998622080597052,
"loss": 5.6216,
"mean_token_accuracy": 0.14571685194969178,
"num_tokens": 5385016.0,
"step": 1925
},
{
"entropy": 5.773671102523804,
"epoch": 0.22989874925550924,
"grad_norm": 0.875,
"learning_rate": 0.0004998607129244563,
"loss": 5.503,
"mean_token_accuracy": 0.15096244364976882,
"num_tokens": 5398838.0,
"step": 1930
},
{
"entropy": 5.834370851516724,
"epoch": 0.23049434187016082,
"grad_norm": 1.03125,
"learning_rate": 0.0004998592097238192,
"loss": 5.6146,
"mean_token_accuracy": 0.15553368628025055,
"num_tokens": 5413297.0,
"step": 1935
},
{
"entropy": 5.883207702636719,
"epoch": 0.23108993448481238,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998576984578479,
"loss": 5.7144,
"mean_token_accuracy": 0.14447153508663177,
"num_tokens": 5428013.0,
"step": 1940
},
{
"entropy": 5.853828191757202,
"epoch": 0.23168552709946397,
"grad_norm": 1.203125,
"learning_rate": 0.0004998561791265966,
"loss": 5.6404,
"mean_token_accuracy": 0.15532859563827514,
"num_tokens": 5442252.0,
"step": 1945
},
{
"entropy": 5.853962182998657,
"epoch": 0.23228111971411555,
"grad_norm": 1.078125,
"learning_rate": 0.0004998546517301196,
"loss": 5.6088,
"mean_token_accuracy": 0.1507514476776123,
"num_tokens": 5455919.0,
"step": 1950
},
{
"entropy": 5.824535846710205,
"epoch": 0.2328767123287671,
"grad_norm": 0.99609375,
"learning_rate": 0.000499853116268472,
"loss": 5.5957,
"mean_token_accuracy": 0.15193421691656112,
"num_tokens": 5469615.0,
"step": 1955
},
{
"entropy": 5.766221046447754,
"epoch": 0.2334723049434187,
"grad_norm": 1.0625,
"learning_rate": 0.0004998515727417086,
"loss": 5.5632,
"mean_token_accuracy": 0.15433768928050995,
"num_tokens": 5483955.0,
"step": 1960
},
{
"entropy": 5.896716499328614,
"epoch": 0.23406789755807028,
"grad_norm": 0.953125,
"learning_rate": 0.0004998500211498848,
"loss": 5.6661,
"mean_token_accuracy": 0.14598566442728042,
"num_tokens": 5498304.0,
"step": 1965
},
{
"entropy": 5.828017234802246,
"epoch": 0.23466349017272187,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998484614930566,
"loss": 5.5255,
"mean_token_accuracy": 0.15394308418035507,
"num_tokens": 5512674.0,
"step": 1970
},
{
"entropy": 5.840755033493042,
"epoch": 0.23525908278737342,
"grad_norm": 1.265625,
"learning_rate": 0.0004998468937712794,
"loss": 5.639,
"mean_token_accuracy": 0.14422297328710557,
"num_tokens": 5527553.0,
"step": 1975
},
{
"entropy": 5.782053375244141,
"epoch": 0.235854675402025,
"grad_norm": 1.078125,
"learning_rate": 0.0004998453179846098,
"loss": 5.5,
"mean_token_accuracy": 0.16121405810117723,
"num_tokens": 5539818.0,
"step": 1980
},
{
"entropy": 5.727403783798218,
"epoch": 0.2364502680166766,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998437341331043,
"loss": 5.5548,
"mean_token_accuracy": 0.14863491505384446,
"num_tokens": 5554619.0,
"step": 1985
},
{
"entropy": 5.792196273803711,
"epoch": 0.23704586063132818,
"grad_norm": 1.015625,
"learning_rate": 0.0004998421422168195,
"loss": 5.5463,
"mean_token_accuracy": 0.15689422488212584,
"num_tokens": 5568539.0,
"step": 1990
},
{
"entropy": 5.7989184856414795,
"epoch": 0.23764145324597974,
"grad_norm": 1.015625,
"learning_rate": 0.0004998405422358126,
"loss": 5.5287,
"mean_token_accuracy": 0.15792766064405442,
"num_tokens": 5581476.0,
"step": 1995
},
{
"entropy": 5.7262804985046385,
"epoch": 0.23823704586063132,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998389341901411,
"loss": 5.518,
"mean_token_accuracy": 0.15621651336550713,
"num_tokens": 5593774.0,
"step": 2000
},
{
"entropy": 5.837261819839478,
"epoch": 0.2388326384752829,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998373180798626,
"loss": 5.6837,
"mean_token_accuracy": 0.1490989312529564,
"num_tokens": 5606425.0,
"step": 2005
},
{
"entropy": 5.868685674667359,
"epoch": 0.2394282310899345,
"grad_norm": 1.03125,
"learning_rate": 0.0004998356939050351,
"loss": 5.5978,
"mean_token_accuracy": 0.1526150196790695,
"num_tokens": 5620472.0,
"step": 2010
},
{
"entropy": 5.746627140045166,
"epoch": 0.24002382370458605,
"grad_norm": 1.015625,
"learning_rate": 0.0004998340616657168,
"loss": 5.564,
"mean_token_accuracy": 0.15095988661050797,
"num_tokens": 5634744.0,
"step": 2015
},
{
"entropy": 5.820336484909058,
"epoch": 0.24061941631923764,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998324213619661,
"loss": 5.5353,
"mean_token_accuracy": 0.15817087292671203,
"num_tokens": 5648555.0,
"step": 2020
},
{
"entropy": 5.654189109802246,
"epoch": 0.24121500893388922,
"grad_norm": 1.03125,
"learning_rate": 0.0004998307729938423,
"loss": 5.4035,
"mean_token_accuracy": 0.16331421136856078,
"num_tokens": 5661780.0,
"step": 2025
},
{
"entropy": 5.810167264938355,
"epoch": 0.2418106015485408,
"grad_norm": 1.0078125,
"learning_rate": 0.000499829116561404,
"loss": 5.6237,
"mean_token_accuracy": 0.14594944566488266,
"num_tokens": 5674713.0,
"step": 2030
},
{
"entropy": 5.8684288501739506,
"epoch": 0.24240619416319237,
"grad_norm": 1.078125,
"learning_rate": 0.0004998274520647109,
"loss": 5.5982,
"mean_token_accuracy": 0.1581734761595726,
"num_tokens": 5689152.0,
"step": 2035
},
{
"entropy": 5.821078968048096,
"epoch": 0.24300178677784395,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998257795038227,
"loss": 5.6043,
"mean_token_accuracy": 0.15297086387872696,
"num_tokens": 5704097.0,
"step": 2040
},
{
"entropy": 5.846594524383545,
"epoch": 0.24359737939249554,
"grad_norm": 1.0,
"learning_rate": 0.0004998240988787992,
"loss": 5.6138,
"mean_token_accuracy": 0.1534001722931862,
"num_tokens": 5716579.0,
"step": 2045
},
{
"entropy": 5.791203594207763,
"epoch": 0.24419297200714712,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998224101897009,
"loss": 5.5933,
"mean_token_accuracy": 0.15501691848039628,
"num_tokens": 5731114.0,
"step": 2050
},
{
"entropy": 5.8727644920349125,
"epoch": 0.24478856462179868,
"grad_norm": 1.015625,
"learning_rate": 0.0004998207134365883,
"loss": 5.5935,
"mean_token_accuracy": 0.14780626967549323,
"num_tokens": 5744843.0,
"step": 2055
},
{
"entropy": 5.789867162704468,
"epoch": 0.24538415723645027,
"grad_norm": 1.046875,
"learning_rate": 0.0004998190086195223,
"loss": 5.638,
"mean_token_accuracy": 0.14191780239343643,
"num_tokens": 5759277.0,
"step": 2060
},
{
"entropy": 5.836066865921021,
"epoch": 0.24597974985110185,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998172957385638,
"loss": 5.6409,
"mean_token_accuracy": 0.1461949788033962,
"num_tokens": 5774754.0,
"step": 2065
},
{
"entropy": 5.824730443954468,
"epoch": 0.2465753424657534,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998155747937746,
"loss": 5.6042,
"mean_token_accuracy": 0.1531083270907402,
"num_tokens": 5788524.0,
"step": 2070
},
{
"entropy": 5.784889554977417,
"epoch": 0.247170935080405,
"grad_norm": 0.96875,
"learning_rate": 0.0004998138457852162,
"loss": 5.4937,
"mean_token_accuracy": 0.15846714973449708,
"num_tokens": 5800617.0,
"step": 2075
},
{
"entropy": 5.729056930541992,
"epoch": 0.24776652769505658,
"grad_norm": 1.0625,
"learning_rate": 0.0004998121087129507,
"loss": 5.4642,
"mean_token_accuracy": 0.16036854833364486,
"num_tokens": 5812433.0,
"step": 2080
},
{
"entropy": 5.788734769821167,
"epoch": 0.24836212030970817,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998103635770404,
"loss": 5.5225,
"mean_token_accuracy": 0.15883542001247405,
"num_tokens": 5825507.0,
"step": 2085
},
{
"entropy": 5.766335344314575,
"epoch": 0.24895771292435973,
"grad_norm": 0.90625,
"learning_rate": 0.0004998086103775478,
"loss": 5.5568,
"mean_token_accuracy": 0.15552578046917914,
"num_tokens": 5841214.0,
"step": 2090
},
{
"entropy": 5.7235047817230225,
"epoch": 0.2495533055390113,
"grad_norm": 1.0,
"learning_rate": 0.0004998068491145359,
"loss": 5.5452,
"mean_token_accuracy": 0.15404129177331924,
"num_tokens": 5854788.0,
"step": 2095
},
{
"entropy": 5.7752213954925535,
"epoch": 0.25014889815366287,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998050797880678,
"loss": 5.5039,
"mean_token_accuracy": 0.15705669671297073,
"num_tokens": 5870241.0,
"step": 2100
},
{
"entropy": 5.706209516525268,
"epoch": 0.25074449076831445,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998033023982071,
"loss": 5.5188,
"mean_token_accuracy": 0.15873096585273744,
"num_tokens": 5883245.0,
"step": 2105
},
{
"entropy": 5.771699523925781,
"epoch": 0.25134008338296604,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998015169450174,
"loss": 5.5159,
"mean_token_accuracy": 0.1555697053670883,
"num_tokens": 5896103.0,
"step": 2110
},
{
"entropy": 5.774951791763305,
"epoch": 0.2519356759976176,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997997234285627,
"loss": 5.5349,
"mean_token_accuracy": 0.15437076538801192,
"num_tokens": 5909633.0,
"step": 2115
},
{
"entropy": 5.741052532196045,
"epoch": 0.2525312686122692,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997979218489076,
"loss": 5.461,
"mean_token_accuracy": 0.16289055198431016,
"num_tokens": 5924211.0,
"step": 2120
},
{
"entropy": 5.732291650772095,
"epoch": 0.2531268612269208,
"grad_norm": 1.0,
"learning_rate": 0.0004997961122061164,
"loss": 5.4904,
"mean_token_accuracy": 0.16250183135271073,
"num_tokens": 5937484.0,
"step": 2125
},
{
"entropy": 5.767012739181519,
"epoch": 0.2537224538415724,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997942945002543,
"loss": 5.5117,
"mean_token_accuracy": 0.1622129425406456,
"num_tokens": 5951224.0,
"step": 2130
},
{
"entropy": 5.752144432067871,
"epoch": 0.25431804645622397,
"grad_norm": 1.0625,
"learning_rate": 0.0004997924687313862,
"loss": 5.5107,
"mean_token_accuracy": 0.158035908639431,
"num_tokens": 5964762.0,
"step": 2135
},
{
"entropy": 5.766765213012695,
"epoch": 0.2549136390708755,
"grad_norm": 1.2578125,
"learning_rate": 0.0004997906348995777,
"loss": 5.6099,
"mean_token_accuracy": 0.1441946767270565,
"num_tokens": 5978215.0,
"step": 2140
},
{
"entropy": 5.777844858169556,
"epoch": 0.2555092316855271,
"grad_norm": 0.984375,
"learning_rate": 0.0004997887930048947,
"loss": 5.5241,
"mean_token_accuracy": 0.15517966747283934,
"num_tokens": 5992951.0,
"step": 2145
},
{
"entropy": 5.805502128601074,
"epoch": 0.25610482430017867,
"grad_norm": 1.03125,
"learning_rate": 0.0004997869430474033,
"loss": 5.5912,
"mean_token_accuracy": 0.14943464547395707,
"num_tokens": 6005853.0,
"step": 2150
},
{
"entropy": 5.670528078079224,
"epoch": 0.25670041691483025,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997850850271696,
"loss": 5.5229,
"mean_token_accuracy": 0.1550400286912918,
"num_tokens": 6021502.0,
"step": 2155
},
{
"entropy": 5.797346067428589,
"epoch": 0.25729600952948184,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997832189442604,
"loss": 5.5552,
"mean_token_accuracy": 0.1466439701616764,
"num_tokens": 6038098.0,
"step": 2160
},
{
"entropy": 5.75105504989624,
"epoch": 0.2578916021441334,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997813447987426,
"loss": 5.6476,
"mean_token_accuracy": 0.14942799955606462,
"num_tokens": 6053957.0,
"step": 2165
},
{
"entropy": 5.843917989730835,
"epoch": 0.258487194758785,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997794625906834,
"loss": 5.59,
"mean_token_accuracy": 0.15020484626293182,
"num_tokens": 6069018.0,
"step": 2170
},
{
"entropy": 5.713776731491089,
"epoch": 0.2590827873734366,
"grad_norm": 0.984375,
"learning_rate": 0.0004997775723201504,
"loss": 5.4722,
"mean_token_accuracy": 0.1559767708182335,
"num_tokens": 6085202.0,
"step": 2175
},
{
"entropy": 5.730500555038452,
"epoch": 0.2596783799880881,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997756739872112,
"loss": 5.5648,
"mean_token_accuracy": 0.15834935307502745,
"num_tokens": 6098351.0,
"step": 2180
},
{
"entropy": 5.7543556690216064,
"epoch": 0.2602739726027397,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997737675919344,
"loss": 5.4807,
"mean_token_accuracy": 0.16555993407964706,
"num_tokens": 6111322.0,
"step": 2185
},
{
"entropy": 5.680195426940918,
"epoch": 0.2608695652173913,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997718531343877,
"loss": 5.5008,
"mean_token_accuracy": 0.15907896757125856,
"num_tokens": 6124438.0,
"step": 2190
},
{
"entropy": 5.830910491943359,
"epoch": 0.2614651578320429,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997699306146402,
"loss": 5.5203,
"mean_token_accuracy": 0.15566118210554122,
"num_tokens": 6139409.0,
"step": 2195
},
{
"entropy": 5.772664499282837,
"epoch": 0.26206075044669447,
"grad_norm": 0.953125,
"learning_rate": 0.0004997680000327607,
"loss": 5.5801,
"mean_token_accuracy": 0.1510126292705536,
"num_tokens": 6156678.0,
"step": 2200
},
{
"entropy": 5.804413557052612,
"epoch": 0.26265634306134605,
"grad_norm": 1.0,
"learning_rate": 0.0004997660613888185,
"loss": 5.6304,
"mean_token_accuracy": 0.1492897093296051,
"num_tokens": 6171149.0,
"step": 2205
},
{
"entropy": 5.7141624927520756,
"epoch": 0.26325193567599764,
"grad_norm": 1.0,
"learning_rate": 0.0004997641146828831,
"loss": 5.4751,
"mean_token_accuracy": 0.1593698337674141,
"num_tokens": 6184179.0,
"step": 2210
},
{
"entropy": 5.700485420227051,
"epoch": 0.26384752829064917,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997621599150244,
"loss": 5.4554,
"mean_token_accuracy": 0.16164242178201677,
"num_tokens": 6197564.0,
"step": 2215
},
{
"entropy": 5.78814730644226,
"epoch": 0.26444312090530075,
"grad_norm": 1.0,
"learning_rate": 0.0004997601970853124,
"loss": 5.5613,
"mean_token_accuracy": 0.1535431906580925,
"num_tokens": 6211018.0,
"step": 2220
},
{
"entropy": 5.642172718048096,
"epoch": 0.26503871351995234,
"grad_norm": 1.015625,
"learning_rate": 0.0004997582261938177,
"loss": 5.4824,
"mean_token_accuracy": 0.165253446996212,
"num_tokens": 6223129.0,
"step": 2225
},
{
"entropy": 5.822677659988403,
"epoch": 0.2656343061346039,
"grad_norm": 1.015625,
"learning_rate": 0.0004997562472406107,
"loss": 5.5354,
"mean_token_accuracy": 0.15645369738340378,
"num_tokens": 6238236.0,
"step": 2230
},
{
"entropy": 5.727661037445069,
"epoch": 0.2662298987492555,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997542602257626,
"loss": 5.4792,
"mean_token_accuracy": 0.15571245178580284,
"num_tokens": 6251285.0,
"step": 2235
},
{
"entropy": 5.695922613143921,
"epoch": 0.2668254913639071,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997522651493447,
"loss": 5.4771,
"mean_token_accuracy": 0.1628317266702652,
"num_tokens": 6266221.0,
"step": 2240
},
{
"entropy": 5.682155513763428,
"epoch": 0.2674210839785587,
"grad_norm": 1.1953125,
"learning_rate": 0.0004997502620114284,
"loss": 5.4851,
"mean_token_accuracy": 0.16071774065494537,
"num_tokens": 6278991.0,
"step": 2245
},
{
"entropy": 5.702351808547974,
"epoch": 0.26801667659321027,
"grad_norm": 1.0,
"learning_rate": 0.0004997482508120856,
"loss": 5.4306,
"mean_token_accuracy": 0.16429808586835862,
"num_tokens": 6293095.0,
"step": 2250
},
{
"entropy": 5.6871805667877195,
"epoch": 0.2686122692078618,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997462315513884,
"loss": 5.4682,
"mean_token_accuracy": 0.16122872084379197,
"num_tokens": 6306162.0,
"step": 2255
},
{
"entropy": 5.756260251998901,
"epoch": 0.2692078618225134,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997442042294095,
"loss": 5.5306,
"mean_token_accuracy": 0.15692074596881866,
"num_tokens": 6318636.0,
"step": 2260
},
{
"entropy": 5.667980194091797,
"epoch": 0.26980345443716497,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997421688462213,
"loss": 5.4875,
"mean_token_accuracy": 0.1623864531517029,
"num_tokens": 6332815.0,
"step": 2265
},
{
"entropy": 5.671984720230102,
"epoch": 0.27039904705181655,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997401254018969,
"loss": 5.3859,
"mean_token_accuracy": 0.16339942216873168,
"num_tokens": 6346887.0,
"step": 2270
},
{
"entropy": 5.691684865951538,
"epoch": 0.27099463966646814,
"grad_norm": 1.0,
"learning_rate": 0.0004997380738965096,
"loss": 5.5025,
"mean_token_accuracy": 0.1576998472213745,
"num_tokens": 6360688.0,
"step": 2275
},
{
"entropy": 5.664248991012573,
"epoch": 0.2715902322811197,
"grad_norm": 0.96484375,
"learning_rate": 0.000499736014330133,
"loss": 5.45,
"mean_token_accuracy": 0.16118588149547577,
"num_tokens": 6376560.0,
"step": 2280
},
{
"entropy": 5.816897964477539,
"epoch": 0.2721858248957713,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997339467028409,
"loss": 5.5707,
"mean_token_accuracy": 0.15335736498236657,
"num_tokens": 6390434.0,
"step": 2285
},
{
"entropy": 5.632402467727661,
"epoch": 0.2727814175104229,
"grad_norm": 1.046875,
"learning_rate": 0.0004997318710147076,
"loss": 5.3763,
"mean_token_accuracy": 0.16879032105207442,
"num_tokens": 6403585.0,
"step": 2290
},
{
"entropy": 5.759318161010742,
"epoch": 0.2733770101250744,
"grad_norm": 1.0625,
"learning_rate": 0.0004997297872658075,
"loss": 5.5492,
"mean_token_accuracy": 0.15672422498464583,
"num_tokens": 6414663.0,
"step": 2295
},
{
"entropy": 5.7544502258300785,
"epoch": 0.273972602739726,
"grad_norm": 0.921875,
"learning_rate": 0.0004997276954562152,
"loss": 5.4669,
"mean_token_accuracy": 0.159070248901844,
"num_tokens": 6428655.0,
"step": 2300
},
{
"entropy": 5.731297779083252,
"epoch": 0.2745681953543776,
"grad_norm": 0.96484375,
"learning_rate": 0.000499725595586006,
"loss": 5.551,
"mean_token_accuracy": 0.1552168533205986,
"num_tokens": 6443494.0,
"step": 2305
},
{
"entropy": 5.734210968017578,
"epoch": 0.2751637879690292,
"grad_norm": 1.1875,
"learning_rate": 0.000499723487655255,
"loss": 5.4873,
"mean_token_accuracy": 0.15551268607378005,
"num_tokens": 6456078.0,
"step": 2310
},
{
"entropy": 5.755726957321167,
"epoch": 0.27575938058368077,
"grad_norm": 1.0,
"learning_rate": 0.0004997213716640378,
"loss": 5.5068,
"mean_token_accuracy": 0.15114115923643112,
"num_tokens": 6469805.0,
"step": 2315
},
{
"entropy": 5.635595083236694,
"epoch": 0.27635497319833235,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997192476124306,
"loss": 5.4483,
"mean_token_accuracy": 0.1549799308180809,
"num_tokens": 6484209.0,
"step": 2320
},
{
"entropy": 5.6775514602661135,
"epoch": 0.27695056581298394,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997171155005091,
"loss": 5.4565,
"mean_token_accuracy": 0.16016292721033096,
"num_tokens": 6498471.0,
"step": 2325
},
{
"entropy": 5.741027402877807,
"epoch": 0.27754615842763547,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997149753283502,
"loss": 5.5533,
"mean_token_accuracy": 0.15452294647693635,
"num_tokens": 6513558.0,
"step": 2330
},
{
"entropy": 5.737386846542359,
"epoch": 0.27814175104228706,
"grad_norm": 1.015625,
"learning_rate": 0.0004997128270960304,
"loss": 5.541,
"mean_token_accuracy": 0.15471736043691636,
"num_tokens": 6527538.0,
"step": 2335
},
{
"entropy": 5.6719869613647464,
"epoch": 0.27873734365693864,
"grad_norm": 0.94140625,
"learning_rate": 0.0004997106708036268,
"loss": 5.3917,
"mean_token_accuracy": 0.1682674691081047,
"num_tokens": 6542188.0,
"step": 2340
},
{
"entropy": 5.707267141342163,
"epoch": 0.2793329362715902,
"grad_norm": 1.140625,
"learning_rate": 0.000499708506451217,
"loss": 5.5256,
"mean_token_accuracy": 0.15894713550806044,
"num_tokens": 6557774.0,
"step": 2345
},
{
"entropy": 5.768304443359375,
"epoch": 0.2799285288862418,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997063340388783,
"loss": 5.4998,
"mean_token_accuracy": 0.15813593566417694,
"num_tokens": 6571868.0,
"step": 2350
},
{
"entropy": 5.61864128112793,
"epoch": 0.2805241215008934,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997041535666888,
"loss": 5.5054,
"mean_token_accuracy": 0.15922000110149384,
"num_tokens": 6585534.0,
"step": 2355
},
{
"entropy": 5.707316255569458,
"epoch": 0.281119714115545,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997019650347266,
"loss": 5.4533,
"mean_token_accuracy": 0.16243680715560913,
"num_tokens": 6599855.0,
"step": 2360
},
{
"entropy": 5.691481971740723,
"epoch": 0.28171530673019657,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996997684430703,
"loss": 5.4491,
"mean_token_accuracy": 0.15631041526794434,
"num_tokens": 6615496.0,
"step": 2365
},
{
"entropy": 5.758702373504638,
"epoch": 0.2823108993448481,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996975637917986,
"loss": 5.5146,
"mean_token_accuracy": 0.15837289094924928,
"num_tokens": 6629993.0,
"step": 2370
},
{
"entropy": 5.624302387237549,
"epoch": 0.2829064919594997,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996953510809906,
"loss": 5.3608,
"mean_token_accuracy": 0.16616827994585037,
"num_tokens": 6644076.0,
"step": 2375
},
{
"entropy": 5.614185571670532,
"epoch": 0.28350208457415127,
"grad_norm": 1.2109375,
"learning_rate": 0.0004996931303107258,
"loss": 5.4172,
"mean_token_accuracy": 0.1681667983531952,
"num_tokens": 6656924.0,
"step": 2380
},
{
"entropy": 5.625910329818725,
"epoch": 0.28409767718880286,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996909014810837,
"loss": 5.4846,
"mean_token_accuracy": 0.15688778311014176,
"num_tokens": 6670350.0,
"step": 2385
},
{
"entropy": 5.731524658203125,
"epoch": 0.28469326980345444,
"grad_norm": 1.1796875,
"learning_rate": 0.0004996886645921442,
"loss": 5.4485,
"mean_token_accuracy": 0.16615445613861085,
"num_tokens": 6683174.0,
"step": 2390
},
{
"entropy": 5.764780712127686,
"epoch": 0.285288862418106,
"grad_norm": 0.96875,
"learning_rate": 0.0004996864196439877,
"loss": 5.5304,
"mean_token_accuracy": 0.15738168507814407,
"num_tokens": 6699648.0,
"step": 2395
},
{
"entropy": 5.643772602081299,
"epoch": 0.2858844550327576,
"grad_norm": 1.0625,
"learning_rate": 0.0004996841666366947,
"loss": 5.4652,
"mean_token_accuracy": 0.1589035451412201,
"num_tokens": 6712174.0,
"step": 2400
},
{
"entropy": 5.632435894012451,
"epoch": 0.2864800476474092,
"grad_norm": 1.03125,
"learning_rate": 0.0004996819055703458,
"loss": 5.4545,
"mean_token_accuracy": 0.16203927099704743,
"num_tokens": 6725376.0,
"step": 2405
},
{
"entropy": 5.524544143676758,
"epoch": 0.2870756402620607,
"grad_norm": 1.078125,
"learning_rate": 0.0004996796364450224,
"loss": 5.2506,
"mean_token_accuracy": 0.17558714896440505,
"num_tokens": 6738177.0,
"step": 2410
},
{
"entropy": 5.681800365447998,
"epoch": 0.2876712328767123,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996773592608057,
"loss": 5.4753,
"mean_token_accuracy": 0.16042183339595795,
"num_tokens": 6751521.0,
"step": 2415
},
{
"entropy": 5.704955530166626,
"epoch": 0.2882668254913639,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996750740177775,
"loss": 5.4962,
"mean_token_accuracy": 0.15936244279146194,
"num_tokens": 6765901.0,
"step": 2420
},
{
"entropy": 5.600447654724121,
"epoch": 0.2888624181060155,
"grad_norm": 1.109375,
"learning_rate": 0.0004996727807160196,
"loss": 5.3418,
"mean_token_accuracy": 0.17559294402599335,
"num_tokens": 6778473.0,
"step": 2425
},
{
"entropy": 5.718362140655517,
"epoch": 0.28945801072066707,
"grad_norm": 1.2421875,
"learning_rate": 0.0004996704793556144,
"loss": 5.5413,
"mean_token_accuracy": 0.15127855092287062,
"num_tokens": 6792067.0,
"step": 2430
},
{
"entropy": 5.664472007751465,
"epoch": 0.29005360333531865,
"grad_norm": 1.015625,
"learning_rate": 0.0004996681699366443,
"loss": 5.4076,
"mean_token_accuracy": 0.16567944288253783,
"num_tokens": 6807534.0,
"step": 2435
},
{
"entropy": 5.632346487045288,
"epoch": 0.29064919594997024,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996658524591924,
"loss": 5.4659,
"mean_token_accuracy": 0.15765434131026268,
"num_tokens": 6821600.0,
"step": 2440
},
{
"entropy": 5.686599826812744,
"epoch": 0.2912447885646218,
"grad_norm": 1.0,
"learning_rate": 0.0004996635269233415,
"loss": 5.4359,
"mean_token_accuracy": 0.16716605424880981,
"num_tokens": 6835044.0,
"step": 2445
},
{
"entropy": 5.648890161514283,
"epoch": 0.29184038117927336,
"grad_norm": 0.984375,
"learning_rate": 0.0004996611933291752,
"loss": 5.5003,
"mean_token_accuracy": 0.1567879304289818,
"num_tokens": 6847745.0,
"step": 2450
},
{
"entropy": 5.643044281005859,
"epoch": 0.29243597379392494,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996588516767772,
"loss": 5.3544,
"mean_token_accuracy": 0.16371610164642333,
"num_tokens": 6861403.0,
"step": 2455
},
{
"entropy": 5.605740261077881,
"epoch": 0.2930315664085765,
"grad_norm": 1.03125,
"learning_rate": 0.0004996565019662315,
"loss": 5.4736,
"mean_token_accuracy": 0.16381412744522095,
"num_tokens": 6874418.0,
"step": 2460
},
{
"entropy": 5.801210021972656,
"epoch": 0.2936271590232281,
"grad_norm": 1.078125,
"learning_rate": 0.0004996541441976223,
"loss": 5.5639,
"mean_token_accuracy": 0.15529614984989165,
"num_tokens": 6890166.0,
"step": 2465
},
{
"entropy": 5.690111064910889,
"epoch": 0.2942227516378797,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996517783710342,
"loss": 5.4776,
"mean_token_accuracy": 0.15613416135311126,
"num_tokens": 6904569.0,
"step": 2470
},
{
"entropy": 5.6721052646636965,
"epoch": 0.2948183442525313,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996494044865521,
"loss": 5.4555,
"mean_token_accuracy": 0.16279612928628923,
"num_tokens": 6918562.0,
"step": 2475
},
{
"entropy": 5.740928030014038,
"epoch": 0.29541393686718287,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996470225442611,
"loss": 5.458,
"mean_token_accuracy": 0.15916131287813187,
"num_tokens": 6934751.0,
"step": 2480
},
{
"entropy": 5.646144247055053,
"epoch": 0.2960095294818344,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996446325442466,
"loss": 5.4563,
"mean_token_accuracy": 0.16374265253543854,
"num_tokens": 6947669.0,
"step": 2485
},
{
"entropy": 5.775876522064209,
"epoch": 0.296605122096486,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996422344865945,
"loss": 5.5286,
"mean_token_accuracy": 0.1609296530485153,
"num_tokens": 6959993.0,
"step": 2490
},
{
"entropy": 5.636061096191407,
"epoch": 0.29720071471113757,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996398283713907,
"loss": 5.4031,
"mean_token_accuracy": 0.1665743112564087,
"num_tokens": 6974637.0,
"step": 2495
},
{
"entropy": 5.56963996887207,
"epoch": 0.29779630732578916,
"grad_norm": 1.046875,
"learning_rate": 0.0004996374141987215,
"loss": 5.3158,
"mean_token_accuracy": 0.167204912006855,
"num_tokens": 6989182.0,
"step": 2500
},
{
"entropy": 5.60927939414978,
"epoch": 0.29839189994044074,
"grad_norm": 0.984375,
"learning_rate": 0.0004996349919686733,
"loss": 5.4237,
"mean_token_accuracy": 0.1643602266907692,
"num_tokens": 7003681.0,
"step": 2505
},
{
"entropy": 5.634929227828979,
"epoch": 0.2989874925550923,
"grad_norm": 1.015625,
"learning_rate": 0.0004996325616813334,
"loss": 5.3678,
"mean_token_accuracy": 0.16805891692638397,
"num_tokens": 7017919.0,
"step": 2510
},
{
"entropy": 5.7103218078613285,
"epoch": 0.2995830851697439,
"grad_norm": 0.9375,
"learning_rate": 0.0004996301233367886,
"loss": 5.4991,
"mean_token_accuracy": 0.15703504979610444,
"num_tokens": 7033405.0,
"step": 2515
},
{
"entropy": 5.591740942001342,
"epoch": 0.3001786777843955,
"grad_norm": 0.96875,
"learning_rate": 0.0004996276769351265,
"loss": 5.4012,
"mean_token_accuracy": 0.16865961104631425,
"num_tokens": 7046938.0,
"step": 2520
},
{
"entropy": 5.639147329330444,
"epoch": 0.30077427039904703,
"grad_norm": 1.1953125,
"learning_rate": 0.0004996252224764349,
"loss": 5.4238,
"mean_token_accuracy": 0.15662651062011718,
"num_tokens": 7059708.0,
"step": 2525
},
{
"entropy": 5.636305665969848,
"epoch": 0.3013698630136986,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996227599608018,
"loss": 5.4238,
"mean_token_accuracy": 0.16775305718183517,
"num_tokens": 7074554.0,
"step": 2530
},
{
"entropy": 5.607331466674805,
"epoch": 0.3019654556283502,
"grad_norm": 0.9375,
"learning_rate": 0.0004996202893883156,
"loss": 5.3777,
"mean_token_accuracy": 0.1624634951353073,
"num_tokens": 7089654.0,
"step": 2535
},
{
"entropy": 5.636829471588134,
"epoch": 0.3025610482430018,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996178107590648,
"loss": 5.451,
"mean_token_accuracy": 0.1580259993672371,
"num_tokens": 7103137.0,
"step": 2540
},
{
"entropy": 5.598562431335449,
"epoch": 0.30315664085765337,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996153240731381,
"loss": 5.3565,
"mean_token_accuracy": 0.17140354141592978,
"num_tokens": 7114857.0,
"step": 2545
},
{
"entropy": 5.650972557067871,
"epoch": 0.30375223347230496,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996128293306251,
"loss": 5.5053,
"mean_token_accuracy": 0.16446297764778137,
"num_tokens": 7128243.0,
"step": 2550
},
{
"entropy": 5.6322792053222654,
"epoch": 0.30434782608695654,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996103265316152,
"loss": 5.4407,
"mean_token_accuracy": 0.15940468907356262,
"num_tokens": 7142505.0,
"step": 2555
},
{
"entropy": 5.5690161228179935,
"epoch": 0.3049434187016081,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996078156761979,
"loss": 5.3898,
"mean_token_accuracy": 0.16011220663785936,
"num_tokens": 7157877.0,
"step": 2560
},
{
"entropy": 5.614592552185059,
"epoch": 0.30553901131625966,
"grad_norm": 1.015625,
"learning_rate": 0.0004996052967644636,
"loss": 5.4461,
"mean_token_accuracy": 0.15924295634031296,
"num_tokens": 7170819.0,
"step": 2565
},
{
"entropy": 5.662329292297363,
"epoch": 0.30613460393091124,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996027697965024,
"loss": 5.4602,
"mean_token_accuracy": 0.16824824064970018,
"num_tokens": 7183013.0,
"step": 2570
},
{
"entropy": 5.577002477645874,
"epoch": 0.3067301965455628,
"grad_norm": 1.0078125,
"learning_rate": 0.000499600234772405,
"loss": 5.491,
"mean_token_accuracy": 0.15459822118282318,
"num_tokens": 7196530.0,
"step": 2575
},
{
"entropy": 5.81734299659729,
"epoch": 0.3073257891602144,
"grad_norm": 0.96875,
"learning_rate": 0.0004995976916922624,
"loss": 5.503,
"mean_token_accuracy": 0.1522914230823517,
"num_tokens": 7210657.0,
"step": 2580
},
{
"entropy": 5.577762269973755,
"epoch": 0.307921381774866,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995951405561658,
"loss": 5.3625,
"mean_token_accuracy": 0.16126917749643327,
"num_tokens": 7224450.0,
"step": 2585
},
{
"entropy": 5.6173646450042725,
"epoch": 0.3085169743895176,
"grad_norm": 1.0,
"learning_rate": 0.0004995925813642068,
"loss": 5.4233,
"mean_token_accuracy": 0.1648561418056488,
"num_tokens": 7238767.0,
"step": 2590
},
{
"entropy": 5.692874526977539,
"epoch": 0.30911256700416917,
"grad_norm": 0.98828125,
"learning_rate": 0.0004995900141164768,
"loss": 5.3155,
"mean_token_accuracy": 0.1685239166021347,
"num_tokens": 7253326.0,
"step": 2595
},
{
"entropy": 5.512430953979492,
"epoch": 0.3097081596188207,
"grad_norm": 0.9140625,
"learning_rate": 0.0004995874388130683,
"loss": 5.3076,
"mean_token_accuracy": 0.16743609309196472,
"num_tokens": 7268256.0,
"step": 2600
},
{
"entropy": 5.612433052062988,
"epoch": 0.3103037522334723,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995848554540734,
"loss": 5.3824,
"mean_token_accuracy": 0.16665096580982208,
"num_tokens": 7281561.0,
"step": 2605
},
{
"entropy": 5.645188856124878,
"epoch": 0.31089934484812387,
"grad_norm": 0.95703125,
"learning_rate": 0.000499582264039585,
"loss": 5.4983,
"mean_token_accuracy": 0.1619505003094673,
"num_tokens": 7294302.0,
"step": 2610
},
{
"entropy": 5.5982451915740965,
"epoch": 0.31149493746277546,
"grad_norm": 1.109375,
"learning_rate": 0.000499579664569696,
"loss": 5.4093,
"mean_token_accuracy": 0.16281116455793382,
"num_tokens": 7309257.0,
"step": 2615
},
{
"entropy": 5.6211920261383055,
"epoch": 0.31209053007742704,
"grad_norm": 1.0625,
"learning_rate": 0.0004995770570444996,
"loss": 5.4261,
"mean_token_accuracy": 0.1656322419643402,
"num_tokens": 7322797.0,
"step": 2620
},
{
"entropy": 5.624097919464111,
"epoch": 0.3126861226920786,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995744414640892,
"loss": 5.3767,
"mean_token_accuracy": 0.17238076776266098,
"num_tokens": 7336913.0,
"step": 2625
},
{
"entropy": 5.646049451828003,
"epoch": 0.3132817153067302,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995718178285587,
"loss": 5.4408,
"mean_token_accuracy": 0.16334831714630127,
"num_tokens": 7350267.0,
"step": 2630
},
{
"entropy": 5.671913194656372,
"epoch": 0.3138773079213818,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995691861380022,
"loss": 5.4281,
"mean_token_accuracy": 0.16798181235790252,
"num_tokens": 7365592.0,
"step": 2635
},
{
"entropy": 5.570678615570069,
"epoch": 0.31447290053603333,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995665463925142,
"loss": 5.3984,
"mean_token_accuracy": 0.16728129982948303,
"num_tokens": 7380045.0,
"step": 2640
},
{
"entropy": 5.649079751968384,
"epoch": 0.3150684931506849,
"grad_norm": 1.21875,
"learning_rate": 0.0004995638985921893,
"loss": 5.4436,
"mean_token_accuracy": 0.16623117178678512,
"num_tokens": 7392487.0,
"step": 2645
},
{
"entropy": 5.6728575229644775,
"epoch": 0.3156640857653365,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995612427371224,
"loss": 5.4844,
"mean_token_accuracy": 0.158479180932045,
"num_tokens": 7405578.0,
"step": 2650
},
{
"entropy": 5.6485189437866214,
"epoch": 0.3162596783799881,
"grad_norm": 1.03125,
"learning_rate": 0.000499558578827409,
"loss": 5.3916,
"mean_token_accuracy": 0.16377977579832076,
"num_tokens": 7418940.0,
"step": 2655
},
{
"entropy": 5.57572226524353,
"epoch": 0.31685527099463967,
"grad_norm": 0.921875,
"learning_rate": 0.0004995559068631444,
"loss": 5.3999,
"mean_token_accuracy": 0.16576361954212188,
"num_tokens": 7432688.0,
"step": 2660
},
{
"entropy": 5.7026293754577635,
"epoch": 0.31745086360929126,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995532268444244,
"loss": 5.5374,
"mean_token_accuracy": 0.1577007867395878,
"num_tokens": 7446238.0,
"step": 2665
},
{
"entropy": 5.701103448867798,
"epoch": 0.31804645622394284,
"grad_norm": 1.15625,
"learning_rate": 0.0004995505387713453,
"loss": 5.5149,
"mean_token_accuracy": 0.15402617454528808,
"num_tokens": 7460781.0,
"step": 2670
},
{
"entropy": 5.595748519897461,
"epoch": 0.3186420488385944,
"grad_norm": 0.96875,
"learning_rate": 0.0004995478426440037,
"loss": 5.529,
"mean_token_accuracy": 0.15443253070116042,
"num_tokens": 7475041.0,
"step": 2675
},
{
"entropy": 5.6993509292602536,
"epoch": 0.31923764145324596,
"grad_norm": 0.875,
"learning_rate": 0.0004995451384624959,
"loss": 5.5181,
"mean_token_accuracy": 0.15144548565149307,
"num_tokens": 7491027.0,
"step": 2680
},
{
"entropy": 5.692638921737671,
"epoch": 0.31983323406789754,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995424262269191,
"loss": 5.4428,
"mean_token_accuracy": 0.16102027893066406,
"num_tokens": 7503070.0,
"step": 2685
},
{
"entropy": 5.569590759277344,
"epoch": 0.32042882668254913,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995397059373704,
"loss": 5.333,
"mean_token_accuracy": 0.1691955015063286,
"num_tokens": 7515553.0,
"step": 2690
},
{
"entropy": 5.538298177719116,
"epoch": 0.3210244192972007,
"grad_norm": 0.9453125,
"learning_rate": 0.0004995369775939478,
"loss": 5.3402,
"mean_token_accuracy": 0.16621797084808348,
"num_tokens": 7529603.0,
"step": 2695
},
{
"entropy": 5.663971900939941,
"epoch": 0.3216200119118523,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995342411967487,
"loss": 5.4818,
"mean_token_accuracy": 0.15859650075435638,
"num_tokens": 7543320.0,
"step": 2700
},
{
"entropy": 5.619390296936035,
"epoch": 0.3222156045265039,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995314967458715,
"loss": 5.4141,
"mean_token_accuracy": 0.16396857649087906,
"num_tokens": 7556413.0,
"step": 2705
},
{
"entropy": 5.6256359100341795,
"epoch": 0.32281119714115547,
"grad_norm": 1.1328125,
"learning_rate": 0.0004995287442414145,
"loss": 5.4322,
"mean_token_accuracy": 0.16423783004283904,
"num_tokens": 7570427.0,
"step": 2710
},
{
"entropy": 5.631426906585693,
"epoch": 0.323406789755807,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995259836834767,
"loss": 5.4863,
"mean_token_accuracy": 0.15992715656757356,
"num_tokens": 7585143.0,
"step": 2715
},
{
"entropy": 5.692942667007446,
"epoch": 0.3240023823704586,
"grad_norm": 1.0625,
"learning_rate": 0.0004995232150721568,
"loss": 5.4042,
"mean_token_accuracy": 0.1656443104147911,
"num_tokens": 7597429.0,
"step": 2720
},
{
"entropy": 5.517695045471191,
"epoch": 0.32459797498511017,
"grad_norm": 1.0625,
"learning_rate": 0.0004995204384075542,
"loss": 5.2723,
"mean_token_accuracy": 0.17762928754091262,
"num_tokens": 7610203.0,
"step": 2725
},
{
"entropy": 5.570860481262207,
"epoch": 0.32519356759976176,
"grad_norm": 1.046875,
"learning_rate": 0.0004995176536897685,
"loss": 5.4666,
"mean_token_accuracy": 0.15536038130521773,
"num_tokens": 7625343.0,
"step": 2730
},
{
"entropy": 5.617734956741333,
"epoch": 0.32578916021441334,
"grad_norm": 1.03125,
"learning_rate": 0.0004995148609188997,
"loss": 5.3592,
"mean_token_accuracy": 0.16593704372644424,
"num_tokens": 7637850.0,
"step": 2735
},
{
"entropy": 5.564725017547607,
"epoch": 0.32638475282906493,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995120600950479,
"loss": 5.3542,
"mean_token_accuracy": 0.1703445553779602,
"num_tokens": 7651877.0,
"step": 2740
},
{
"entropy": 5.695179843902588,
"epoch": 0.3269803454437165,
"grad_norm": 1.046875,
"learning_rate": 0.0004995092512183134,
"loss": 5.466,
"mean_token_accuracy": 0.15940675437450408,
"num_tokens": 7666457.0,
"step": 2745
},
{
"entropy": 5.567157411575318,
"epoch": 0.3275759380583681,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995064342887972,
"loss": 5.3655,
"mean_token_accuracy": 0.1676443710923195,
"num_tokens": 7678979.0,
"step": 2750
},
{
"entropy": 5.558333110809326,
"epoch": 0.32817153067301963,
"grad_norm": 0.984375,
"learning_rate": 0.0004995036093066001,
"loss": 5.3185,
"mean_token_accuracy": 0.16749558895826339,
"num_tokens": 7692357.0,
"step": 2755
},
{
"entropy": 5.686964607238769,
"epoch": 0.3287671232876712,
"grad_norm": 1.109375,
"learning_rate": 0.0004995007762718237,
"loss": 5.4471,
"mean_token_accuracy": 0.15954697579145433,
"num_tokens": 7707537.0,
"step": 2760
},
{
"entropy": 5.519318008422852,
"epoch": 0.3293627159023228,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994979351845693,
"loss": 5.3707,
"mean_token_accuracy": 0.16792828291654588,
"num_tokens": 7719737.0,
"step": 2765
},
{
"entropy": 5.609381437301636,
"epoch": 0.3299583085169744,
"grad_norm": 1.0,
"learning_rate": 0.0004994950860449392,
"loss": 5.4084,
"mean_token_accuracy": 0.16744615286588668,
"num_tokens": 7734612.0,
"step": 2770
},
{
"entropy": 5.634121942520141,
"epoch": 0.33055390113162597,
"grad_norm": 0.88671875,
"learning_rate": 0.0004994922288530353,
"loss": 5.4608,
"mean_token_accuracy": 0.1555524453520775,
"num_tokens": 7749109.0,
"step": 2775
},
{
"entropy": 5.515996646881104,
"epoch": 0.33114949374627756,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994893636089601,
"loss": 5.2915,
"mean_token_accuracy": 0.16647806912660598,
"num_tokens": 7761972.0,
"step": 2780
},
{
"entropy": 5.628909635543823,
"epoch": 0.33174508636092914,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994864903128164,
"loss": 5.4156,
"mean_token_accuracy": 0.1572790265083313,
"num_tokens": 7777769.0,
"step": 2785
},
{
"entropy": 5.622025060653686,
"epoch": 0.33234067897558073,
"grad_norm": 0.921875,
"learning_rate": 0.0004994836089647073,
"loss": 5.3714,
"mean_token_accuracy": 0.16501683443784715,
"num_tokens": 7792576.0,
"step": 2790
},
{
"entropy": 5.502942609786987,
"epoch": 0.33293627159023226,
"grad_norm": 1.0,
"learning_rate": 0.0004994807195647362,
"loss": 5.3779,
"mean_token_accuracy": 0.16594948321580888,
"num_tokens": 7805997.0,
"step": 2795
},
{
"entropy": 5.557144498825073,
"epoch": 0.33353186420488384,
"grad_norm": 1.015625,
"learning_rate": 0.0004994778221130066,
"loss": 5.3605,
"mean_token_accuracy": 0.16462566703557968,
"num_tokens": 7819548.0,
"step": 2800
},
{
"entropy": 5.550113344192505,
"epoch": 0.33412745681953543,
"grad_norm": 0.921875,
"learning_rate": 0.0004994749166096226,
"loss": 5.3618,
"mean_token_accuracy": 0.16538318544626235,
"num_tokens": 7832786.0,
"step": 2805
},
{
"entropy": 5.638032674789429,
"epoch": 0.334723049434187,
"grad_norm": 1.078125,
"learning_rate": 0.0004994720030546882,
"loss": 5.3941,
"mean_token_accuracy": 0.1630851998925209,
"num_tokens": 7847524.0,
"step": 2810
},
{
"entropy": 5.516295433044434,
"epoch": 0.3353186420488386,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994690814483081,
"loss": 5.3043,
"mean_token_accuracy": 0.17548558712005616,
"num_tokens": 7861783.0,
"step": 2815
},
{
"entropy": 5.532091665267944,
"epoch": 0.3359142346634902,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994661517905869,
"loss": 5.3988,
"mean_token_accuracy": 0.1643317773938179,
"num_tokens": 7875671.0,
"step": 2820
},
{
"entropy": 5.648615932464599,
"epoch": 0.33650982727814177,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994632140816299,
"loss": 5.3213,
"mean_token_accuracy": 0.1673572242259979,
"num_tokens": 7890624.0,
"step": 2825
},
{
"entropy": 5.561357641220093,
"epoch": 0.33710541989279336,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994602683215424,
"loss": 5.386,
"mean_token_accuracy": 0.16167960613965987,
"num_tokens": 7904296.0,
"step": 2830
},
{
"entropy": 5.505871915817261,
"epoch": 0.3377010125074449,
"grad_norm": 1.0703125,
"learning_rate": 0.00049945731451043,
"loss": 5.2582,
"mean_token_accuracy": 0.16874135732650758,
"num_tokens": 7918588.0,
"step": 2835
},
{
"entropy": 5.522439622879029,
"epoch": 0.33829660512209647,
"grad_norm": 1.0625,
"learning_rate": 0.0004994543526483985,
"loss": 5.2701,
"mean_token_accuracy": 0.17053997218608857,
"num_tokens": 7933087.0,
"step": 2840
},
{
"entropy": 5.587213230133057,
"epoch": 0.33889219773674806,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994513827355546,
"loss": 5.4103,
"mean_token_accuracy": 0.16668923944234848,
"num_tokens": 7945477.0,
"step": 2845
},
{
"entropy": 5.63070592880249,
"epoch": 0.33948779035139964,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994484047720043,
"loss": 5.4989,
"mean_token_accuracy": 0.1570942744612694,
"num_tokens": 7961488.0,
"step": 2850
},
{
"entropy": 5.6248846530914305,
"epoch": 0.34008338296605123,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994454187578547,
"loss": 5.2955,
"mean_token_accuracy": 0.1757759466767311,
"num_tokens": 7974533.0,
"step": 2855
},
{
"entropy": 5.526473426818848,
"epoch": 0.3406789755807028,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994424246932129,
"loss": 5.33,
"mean_token_accuracy": 0.16862046271562575,
"num_tokens": 7988481.0,
"step": 2860
},
{
"entropy": 5.619885015487671,
"epoch": 0.3412745681953544,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994394225781863,
"loss": 5.3599,
"mean_token_accuracy": 0.16942613273859025,
"num_tokens": 8003589.0,
"step": 2865
},
{
"entropy": 5.52764778137207,
"epoch": 0.34187016081000593,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994364124128825,
"loss": 5.3242,
"mean_token_accuracy": 0.1693471595644951,
"num_tokens": 8017560.0,
"step": 2870
},
{
"entropy": 5.552784109115601,
"epoch": 0.3424657534246575,
"grad_norm": 0.8984375,
"learning_rate": 0.0004994333941974094,
"loss": 5.435,
"mean_token_accuracy": 0.1612047329545021,
"num_tokens": 8032339.0,
"step": 2875
},
{
"entropy": 5.595651721954345,
"epoch": 0.3430613460393091,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994303679318754,
"loss": 5.4255,
"mean_token_accuracy": 0.1592611014842987,
"num_tokens": 8047634.0,
"step": 2880
},
{
"entropy": 5.5863272190094,
"epoch": 0.3436569386539607,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994273336163891,
"loss": 5.3918,
"mean_token_accuracy": 0.1652919813990593,
"num_tokens": 8061267.0,
"step": 2885
},
{
"entropy": 5.543440198898315,
"epoch": 0.34425253126861227,
"grad_norm": 1.125,
"learning_rate": 0.0004994242912510591,
"loss": 5.3174,
"mean_token_accuracy": 0.166269950568676,
"num_tokens": 8076225.0,
"step": 2890
},
{
"entropy": 5.573662137985229,
"epoch": 0.34484812388326386,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994212408359948,
"loss": 5.4046,
"mean_token_accuracy": 0.16605644524097443,
"num_tokens": 8089997.0,
"step": 2895
},
{
"entropy": 5.591129922866822,
"epoch": 0.34544371649791544,
"grad_norm": 1.015625,
"learning_rate": 0.0004994181823713054,
"loss": 5.3152,
"mean_token_accuracy": 0.16995038986206054,
"num_tokens": 8102739.0,
"step": 2900
},
{
"entropy": 5.587178468704224,
"epoch": 0.34603930911256703,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994151158571006,
"loss": 5.4872,
"mean_token_accuracy": 0.1624770373106003,
"num_tokens": 8115435.0,
"step": 2905
},
{
"entropy": 5.598823881149292,
"epoch": 0.34663490172721856,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994120412934906,
"loss": 5.3636,
"mean_token_accuracy": 0.1697479158639908,
"num_tokens": 8129833.0,
"step": 2910
},
{
"entropy": 5.569084024429321,
"epoch": 0.34723049434187014,
"grad_norm": 1.203125,
"learning_rate": 0.0004994089586805856,
"loss": 5.3729,
"mean_token_accuracy": 0.1669548973441124,
"num_tokens": 8144316.0,
"step": 2915
},
{
"entropy": 5.578461694717407,
"epoch": 0.34782608695652173,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994058680184959,
"loss": 5.3739,
"mean_token_accuracy": 0.16917092055082322,
"num_tokens": 8158927.0,
"step": 2920
},
{
"entropy": 5.617590284347534,
"epoch": 0.3484216795711733,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994027693073327,
"loss": 5.3779,
"mean_token_accuracy": 0.16542406380176544,
"num_tokens": 8172297.0,
"step": 2925
},
{
"entropy": 5.613320064544678,
"epoch": 0.3490172721858249,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993996625472071,
"loss": 5.4165,
"mean_token_accuracy": 0.1679888367652893,
"num_tokens": 8184985.0,
"step": 2930
},
{
"entropy": 5.588392210006714,
"epoch": 0.3496128648004765,
"grad_norm": 1.0,
"learning_rate": 0.0004993965477382304,
"loss": 5.369,
"mean_token_accuracy": 0.1678030714392662,
"num_tokens": 8198936.0,
"step": 2935
},
{
"entropy": 5.5983302116394045,
"epoch": 0.35020845741512807,
"grad_norm": 1.0,
"learning_rate": 0.0004993934248805143,
"loss": 5.3856,
"mean_token_accuracy": 0.1687172457575798,
"num_tokens": 8213538.0,
"step": 2940
},
{
"entropy": 5.59259786605835,
"epoch": 0.35080405002977966,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993902939741709,
"loss": 5.4503,
"mean_token_accuracy": 0.161647991836071,
"num_tokens": 8226794.0,
"step": 2945
},
{
"entropy": 5.527505016326904,
"epoch": 0.3513996426444312,
"grad_norm": 1.0,
"learning_rate": 0.0004993871550193126,
"loss": 5.347,
"mean_token_accuracy": 0.16996421962976455,
"num_tokens": 8241671.0,
"step": 2950
},
{
"entropy": 5.5442180156707765,
"epoch": 0.3519952352590828,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993840080160519,
"loss": 5.3727,
"mean_token_accuracy": 0.16769520044326783,
"num_tokens": 8255079.0,
"step": 2955
},
{
"entropy": 5.5615553855896,
"epoch": 0.35259082787373436,
"grad_norm": 0.96875,
"learning_rate": 0.0004993808529645016,
"loss": 5.3474,
"mean_token_accuracy": 0.16994720846414565,
"num_tokens": 8267958.0,
"step": 2960
},
{
"entropy": 5.550642204284668,
"epoch": 0.35318642048838594,
"grad_norm": 1.03125,
"learning_rate": 0.0004993776898647749,
"loss": 5.4058,
"mean_token_accuracy": 0.1602596029639244,
"num_tokens": 8281082.0,
"step": 2965
},
{
"entropy": 5.548654079437256,
"epoch": 0.35378201310303753,
"grad_norm": 1.125,
"learning_rate": 0.0004993745187169852,
"loss": 5.2914,
"mean_token_accuracy": 0.16562802940607071,
"num_tokens": 8292675.0,
"step": 2970
},
{
"entropy": 5.510513496398926,
"epoch": 0.3543776057176891,
"grad_norm": 0.96875,
"learning_rate": 0.0004993713395212464,
"loss": 5.313,
"mean_token_accuracy": 0.17679769545793533,
"num_tokens": 8304481.0,
"step": 2975
},
{
"entropy": 5.553654766082763,
"epoch": 0.3549731983323407,
"grad_norm": 0.953125,
"learning_rate": 0.0004993681522776725,
"loss": 5.4097,
"mean_token_accuracy": 0.16842448115348815,
"num_tokens": 8318529.0,
"step": 2980
},
{
"entropy": 5.5886695861816404,
"epoch": 0.35556879094699223,
"grad_norm": 0.921875,
"learning_rate": 0.0004993649569863778,
"loss": 5.3959,
"mean_token_accuracy": 0.16488241851329805,
"num_tokens": 8333181.0,
"step": 2985
},
{
"entropy": 5.497231817245483,
"epoch": 0.3561643835616438,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993617536474768,
"loss": 5.2714,
"mean_token_accuracy": 0.17946436256170273,
"num_tokens": 8346036.0,
"step": 2990
},
{
"entropy": 5.511769723892212,
"epoch": 0.3567599761762954,
"grad_norm": 0.953125,
"learning_rate": 0.0004993585422610845,
"loss": 5.3213,
"mean_token_accuracy": 0.16782039552927017,
"num_tokens": 8361037.0,
"step": 2995
},
{
"entropy": 5.55103235244751,
"epoch": 0.357355568790947,
"grad_norm": 0.93359375,
"learning_rate": 0.0004993553228273161,
"loss": 5.3502,
"mean_token_accuracy": 0.17024607360363006,
"num_tokens": 8374802.0,
"step": 3000
},
{
"epoch": 0.357355568790947,
"eval_entropy": 5.33094315484026,
"eval_loss": 5.315858364105225,
"eval_mean_token_accuracy": 0.17842997617459036,
"eval_num_tokens": 8374802.0,
"eval_runtime": 18.8093,
"eval_samples_per_second": 1356.401,
"eval_steps_per_second": 169.597,
"step": 3000
},
{
"entropy": 5.53194375038147,
"epoch": 0.3579511614055986,
"grad_norm": 0.95703125,
"learning_rate": 0.000499352095346287,
"loss": 5.2731,
"mean_token_accuracy": 0.17845398634672166,
"num_tokens": 8389127.0,
"step": 3005
},
{
"entropy": 5.645079278945923,
"epoch": 0.35854675402025016,
"grad_norm": 0.92578125,
"learning_rate": 0.0004993488598181131,
"loss": 5.4584,
"mean_token_accuracy": 0.1543111339211464,
"num_tokens": 8403358.0,
"step": 3010
},
{
"entropy": 5.578828430175781,
"epoch": 0.35914234663490174,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993456162429103,
"loss": 5.3537,
"mean_token_accuracy": 0.1660466119647026,
"num_tokens": 8418693.0,
"step": 3015
},
{
"entropy": 5.534197664260864,
"epoch": 0.35973793924955333,
"grad_norm": 0.9765625,
"learning_rate": 0.0004993423646207951,
"loss": 5.3074,
"mean_token_accuracy": 0.17261601388454437,
"num_tokens": 8429968.0,
"step": 3020
},
{
"entropy": 5.526310586929322,
"epoch": 0.36033353186420486,
"grad_norm": 0.98828125,
"learning_rate": 0.000499339104951884,
"loss": 5.3498,
"mean_token_accuracy": 0.16782355755567552,
"num_tokens": 8443700.0,
"step": 3025
},
{
"entropy": 5.569401836395263,
"epoch": 0.36092912447885644,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993358372362939,
"loss": 5.381,
"mean_token_accuracy": 0.16564542949199676,
"num_tokens": 8457919.0,
"step": 3030
},
{
"entropy": 5.60796422958374,
"epoch": 0.36152471709350803,
"grad_norm": 0.890625,
"learning_rate": 0.0004993325614741421,
"loss": 5.3781,
"mean_token_accuracy": 0.16518213897943496,
"num_tokens": 8472787.0,
"step": 3035
},
{
"entropy": 5.532761478424073,
"epoch": 0.3621203097081596,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993292776655461,
"loss": 5.3379,
"mean_token_accuracy": 0.17310683876276017,
"num_tokens": 8484235.0,
"step": 3040
},
{
"entropy": 5.460708284378052,
"epoch": 0.3627159023228112,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993259858106236,
"loss": 5.2628,
"mean_token_accuracy": 0.17374361157417298,
"num_tokens": 8496971.0,
"step": 3045
},
{
"entropy": 5.544519662857056,
"epoch": 0.3633114949374628,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993226859094927,
"loss": 5.3449,
"mean_token_accuracy": 0.16985341310501098,
"num_tokens": 8510531.0,
"step": 3050
},
{
"entropy": 5.555769729614258,
"epoch": 0.36390708755211437,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993193779622719,
"loss": 5.4447,
"mean_token_accuracy": 0.16951690912246703,
"num_tokens": 8522492.0,
"step": 3055
},
{
"entropy": 5.602981996536255,
"epoch": 0.36450268016676596,
"grad_norm": 0.87890625,
"learning_rate": 0.0004993160619690796,
"loss": 5.3362,
"mean_token_accuracy": 0.16971496790647506,
"num_tokens": 8537119.0,
"step": 3060
},
{
"entropy": 5.464743804931641,
"epoch": 0.3650982727814175,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993127379300349,
"loss": 5.2638,
"mean_token_accuracy": 0.17136756777763368,
"num_tokens": 8549899.0,
"step": 3065
},
{
"entropy": 5.5660888195037845,
"epoch": 0.3656938653960691,
"grad_norm": 1.03125,
"learning_rate": 0.000499309405845257,
"loss": 5.4365,
"mean_token_accuracy": 0.1557522490620613,
"num_tokens": 8563604.0,
"step": 3070
},
{
"entropy": 5.648580932617188,
"epoch": 0.36628945801072066,
"grad_norm": 1.0,
"learning_rate": 0.0004993060657148654,
"loss": 5.3494,
"mean_token_accuracy": 0.1666161060333252,
"num_tokens": 8577948.0,
"step": 3075
},
{
"entropy": 5.5360502243042,
"epoch": 0.36688505062537224,
"grad_norm": 0.95703125,
"learning_rate": 0.00049930271753898,
"loss": 5.3273,
"mean_token_accuracy": 0.1703394889831543,
"num_tokens": 8591750.0,
"step": 3080
},
{
"entropy": 5.570601511001587,
"epoch": 0.36748064324002383,
"grad_norm": 1.21875,
"learning_rate": 0.0004992993613177208,
"loss": 5.319,
"mean_token_accuracy": 0.17260025888681413,
"num_tokens": 8604091.0,
"step": 3085
},
{
"entropy": 5.427626848220825,
"epoch": 0.3680762358546754,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992959970512082,
"loss": 5.2785,
"mean_token_accuracy": 0.17656562775373458,
"num_tokens": 8617314.0,
"step": 3090
},
{
"entropy": 5.468297958374023,
"epoch": 0.368671828469327,
"grad_norm": 1.0625,
"learning_rate": 0.0004992926247395627,
"loss": 5.2569,
"mean_token_accuracy": 0.17591919004917145,
"num_tokens": 8630050.0,
"step": 3095
},
{
"entropy": 5.4180926322937015,
"epoch": 0.36926742108397853,
"grad_norm": 0.97265625,
"learning_rate": 0.0004992892443829057,
"loss": 5.3133,
"mean_token_accuracy": 0.17195673286914825,
"num_tokens": 8643462.0,
"step": 3100
},
{
"entropy": 5.47792763710022,
"epoch": 0.3698630136986301,
"grad_norm": 0.91015625,
"learning_rate": 0.000499285855981358,
"loss": 5.2804,
"mean_token_accuracy": 0.17004989087581635,
"num_tokens": 8657207.0,
"step": 3105
},
{
"entropy": 5.579573440551758,
"epoch": 0.3704586063132817,
"grad_norm": 1.09375,
"learning_rate": 0.0004992824595350414,
"loss": 5.3467,
"mean_token_accuracy": 0.17448567301034928,
"num_tokens": 8671129.0,
"step": 3110
},
{
"entropy": 5.5449652671813965,
"epoch": 0.3710541989279333,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992790550440775,
"loss": 5.2924,
"mean_token_accuracy": 0.1721959412097931,
"num_tokens": 8684308.0,
"step": 3115
},
{
"entropy": 5.509975385665894,
"epoch": 0.3716497915425849,
"grad_norm": 1.0625,
"learning_rate": 0.0004992756425085887,
"loss": 5.2355,
"mean_token_accuracy": 0.17383912801742554,
"num_tokens": 8697651.0,
"step": 3120
},
{
"entropy": 5.543778657913208,
"epoch": 0.37224538415723646,
"grad_norm": 0.875,
"learning_rate": 0.0004992722219286972,
"loss": 5.3822,
"mean_token_accuracy": 0.16248745173215867,
"num_tokens": 8714559.0,
"step": 3125
},
{
"entropy": 5.575860786437988,
"epoch": 0.37284097677188804,
"grad_norm": 1.046875,
"learning_rate": 0.0004992687933045258,
"loss": 5.3106,
"mean_token_accuracy": 0.1686939001083374,
"num_tokens": 8729999.0,
"step": 3130
},
{
"entropy": 5.5252625942230225,
"epoch": 0.37343656938653963,
"grad_norm": 1.0703125,
"learning_rate": 0.0004992653566361974,
"loss": 5.3064,
"mean_token_accuracy": 0.17304440289735795,
"num_tokens": 8742503.0,
"step": 3135
},
{
"entropy": 5.50363187789917,
"epoch": 0.37403216200119116,
"grad_norm": 1.046875,
"learning_rate": 0.0004992619119238351,
"loss": 5.3414,
"mean_token_accuracy": 0.16426770389080048,
"num_tokens": 8755034.0,
"step": 3140
},
{
"entropy": 5.49028844833374,
"epoch": 0.37462775461584275,
"grad_norm": 0.8828125,
"learning_rate": 0.000499258459167563,
"loss": 5.3407,
"mean_token_accuracy": 0.1699514016509056,
"num_tokens": 8770410.0,
"step": 3145
},
{
"entropy": 5.5367120742797855,
"epoch": 0.37522334723049433,
"grad_norm": 1.140625,
"learning_rate": 0.0004992549983675043,
"loss": 5.2963,
"mean_token_accuracy": 0.17531256079673768,
"num_tokens": 8784020.0,
"step": 3150
},
{
"entropy": 5.495750761032104,
"epoch": 0.3758189398451459,
"grad_norm": 1.09375,
"learning_rate": 0.0004992515295237835,
"loss": 5.3716,
"mean_token_accuracy": 0.17027590200304984,
"num_tokens": 8797763.0,
"step": 3155
},
{
"entropy": 5.494180345535279,
"epoch": 0.3764145324597975,
"grad_norm": 1.125,
"learning_rate": 0.000499248052636525,
"loss": 5.2253,
"mean_token_accuracy": 0.179366996884346,
"num_tokens": 8810306.0,
"step": 3160
},
{
"entropy": 5.444682168960571,
"epoch": 0.3770101250744491,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992445677058534,
"loss": 5.255,
"mean_token_accuracy": 0.175647334754467,
"num_tokens": 8822345.0,
"step": 3165
},
{
"entropy": 5.488202714920044,
"epoch": 0.3776057176891007,
"grad_norm": 0.9296875,
"learning_rate": 0.0004992410747318938,
"loss": 5.3445,
"mean_token_accuracy": 0.1701683670282364,
"num_tokens": 8836024.0,
"step": 3170
},
{
"entropy": 5.539686298370361,
"epoch": 0.37820131030375226,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992375737147715,
"loss": 5.3752,
"mean_token_accuracy": 0.16921445727348328,
"num_tokens": 8849994.0,
"step": 3175
},
{
"entropy": 5.494520139694214,
"epoch": 0.3787969029184038,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992340646546118,
"loss": 5.2648,
"mean_token_accuracy": 0.17560119926929474,
"num_tokens": 8863867.0,
"step": 3180
},
{
"entropy": 5.459944105148315,
"epoch": 0.3793924955330554,
"grad_norm": 0.94921875,
"learning_rate": 0.000499230547551541,
"loss": 5.3574,
"mean_token_accuracy": 0.17062760442495345,
"num_tokens": 8878077.0,
"step": 3185
},
{
"entropy": 5.497016668319702,
"epoch": 0.37998808814770696,
"grad_norm": 0.98828125,
"learning_rate": 0.0004992270224056849,
"loss": 5.3181,
"mean_token_accuracy": 0.16758826673030852,
"num_tokens": 8892729.0,
"step": 3190
},
{
"entropy": 5.513605403900146,
"epoch": 0.38058368076235854,
"grad_norm": 1.1640625,
"learning_rate": 0.0004992234892171702,
"loss": 5.253,
"mean_token_accuracy": 0.1787588343024254,
"num_tokens": 8906242.0,
"step": 3195
},
{
"entropy": 5.500463771820068,
"epoch": 0.38117927337701013,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992199479861235,
"loss": 5.343,
"mean_token_accuracy": 0.16812289208173753,
"num_tokens": 8919814.0,
"step": 3200
},
{
"entropy": 5.489879703521728,
"epoch": 0.3817748659916617,
"grad_norm": 0.9375,
"learning_rate": 0.0004992163987126717,
"loss": 5.1934,
"mean_token_accuracy": 0.1734540954232216,
"num_tokens": 8934345.0,
"step": 3205
},
{
"entropy": 5.469197797775268,
"epoch": 0.3823704586063133,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992128413969423,
"loss": 5.2822,
"mean_token_accuracy": 0.17496797889471055,
"num_tokens": 8947825.0,
"step": 3210
},
{
"entropy": 5.5221010684967045,
"epoch": 0.3829660512209649,
"grad_norm": 0.96875,
"learning_rate": 0.0004992092760390628,
"loss": 5.425,
"mean_token_accuracy": 0.16763191670179367,
"num_tokens": 8962270.0,
"step": 3215
},
{
"entropy": 5.516406583786011,
"epoch": 0.3835616438356164,
"grad_norm": 1.0703125,
"learning_rate": 0.0004992057026391612,
"loss": 5.2772,
"mean_token_accuracy": 0.17425250262022018,
"num_tokens": 8974453.0,
"step": 3220
},
{
"entropy": 5.545296716690063,
"epoch": 0.384157236450268,
"grad_norm": 0.95703125,
"learning_rate": 0.0004992021211973655,
"loss": 5.3742,
"mean_token_accuracy": 0.1678760230541229,
"num_tokens": 8991689.0,
"step": 3225
},
{
"entropy": 5.508287954330444,
"epoch": 0.3847528290649196,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991985317138043,
"loss": 5.3286,
"mean_token_accuracy": 0.17742977142333985,
"num_tokens": 9005783.0,
"step": 3230
},
{
"entropy": 5.489894723892212,
"epoch": 0.3853484216795712,
"grad_norm": 1.0,
"learning_rate": 0.0004991949341886063,
"loss": 5.3078,
"mean_token_accuracy": 0.16761346012353898,
"num_tokens": 9019733.0,
"step": 3235
},
{
"entropy": 5.433395004272461,
"epoch": 0.38594401429422276,
"grad_norm": 0.921875,
"learning_rate": 0.0004991913286219004,
"loss": 5.3033,
"mean_token_accuracy": 0.16695324182510377,
"num_tokens": 9035131.0,
"step": 3240
},
{
"entropy": 5.57075867652893,
"epoch": 0.38653960690887434,
"grad_norm": 1.171875,
"learning_rate": 0.0004991877150138162,
"loss": 5.2041,
"mean_token_accuracy": 0.17603086531162263,
"num_tokens": 9047954.0,
"step": 3245
},
{
"entropy": 5.458488321304321,
"epoch": 0.38713519952352593,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991840933644831,
"loss": 5.2424,
"mean_token_accuracy": 0.17589410841464997,
"num_tokens": 9059945.0,
"step": 3250
},
{
"entropy": 5.48297004699707,
"epoch": 0.38773079213817746,
"grad_norm": 0.8984375,
"learning_rate": 0.000499180463674031,
"loss": 5.338,
"mean_token_accuracy": 0.17242926508188247,
"num_tokens": 9074128.0,
"step": 3255
},
{
"entropy": 5.503276681900024,
"epoch": 0.38832638475282905,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991768259425903,
"loss": 5.4064,
"mean_token_accuracy": 0.16760232895612717,
"num_tokens": 9086760.0,
"step": 3260
},
{
"entropy": 5.555027437210083,
"epoch": 0.38892197736748063,
"grad_norm": 0.96875,
"learning_rate": 0.0004991731801702914,
"loss": 5.301,
"mean_token_accuracy": 0.16785975694656372,
"num_tokens": 9100632.0,
"step": 3265
},
{
"entropy": 5.485769939422608,
"epoch": 0.3895175699821322,
"grad_norm": 1.046875,
"learning_rate": 0.0004991695263572648,
"loss": 5.4162,
"mean_token_accuracy": 0.16383741348981856,
"num_tokens": 9115092.0,
"step": 3270
},
{
"entropy": 5.450682783126831,
"epoch": 0.3901131625967838,
"grad_norm": 1.203125,
"learning_rate": 0.0004991658645036419,
"loss": 5.1939,
"mean_token_accuracy": 0.181158210337162,
"num_tokens": 9127473.0,
"step": 3275
},
{
"entropy": 5.514653205871582,
"epoch": 0.3907087552114354,
"grad_norm": 1.0,
"learning_rate": 0.0004991621946095539,
"loss": 5.3127,
"mean_token_accuracy": 0.1720754772424698,
"num_tokens": 9141028.0,
"step": 3280
},
{
"entropy": 5.4921064376831055,
"epoch": 0.391304347826087,
"grad_norm": 1.109375,
"learning_rate": 0.0004991585166751325,
"loss": 5.258,
"mean_token_accuracy": 0.17332016676664352,
"num_tokens": 9154526.0,
"step": 3285
},
{
"entropy": 5.555002164840698,
"epoch": 0.39189994044073856,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991548307005094,
"loss": 5.337,
"mean_token_accuracy": 0.1678193524479866,
"num_tokens": 9167220.0,
"step": 3290
},
{
"entropy": 5.562772369384765,
"epoch": 0.3924955330553901,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991511366858172,
"loss": 5.3952,
"mean_token_accuracy": 0.1711056910455227,
"num_tokens": 9181648.0,
"step": 3295
},
{
"entropy": 5.4501426219940186,
"epoch": 0.3930911256700417,
"grad_norm": 1.046875,
"learning_rate": 0.0004991474346311879,
"loss": 5.1889,
"mean_token_accuracy": 0.1786603108048439,
"num_tokens": 9194714.0,
"step": 3300
},
{
"entropy": 5.426304912567138,
"epoch": 0.39368671828469326,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991437245367547,
"loss": 5.274,
"mean_token_accuracy": 0.17161348313093186,
"num_tokens": 9211467.0,
"step": 3305
},
{
"entropy": 5.664749717712402,
"epoch": 0.39428231089934485,
"grad_norm": 1.03125,
"learning_rate": 0.0004991400064026506,
"loss": 5.4901,
"mean_token_accuracy": 0.16128215342760086,
"num_tokens": 9226409.0,
"step": 3310
},
{
"entropy": 5.487881278991699,
"epoch": 0.39487790351399643,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991362802290089,
"loss": 5.2442,
"mean_token_accuracy": 0.17049567103385926,
"num_tokens": 9240476.0,
"step": 3315
},
{
"entropy": 5.454006195068359,
"epoch": 0.395473496128648,
"grad_norm": 0.96875,
"learning_rate": 0.0004991325460159631,
"loss": 5.2754,
"mean_token_accuracy": 0.17328597903251647,
"num_tokens": 9254020.0,
"step": 3320
},
{
"entropy": 5.438784456253051,
"epoch": 0.3960690887432996,
"grad_norm": 0.95703125,
"learning_rate": 0.0004991288037636474,
"loss": 5.1954,
"mean_token_accuracy": 0.17969273924827575,
"num_tokens": 9269150.0,
"step": 3325
},
{
"entropy": 5.572743654251099,
"epoch": 0.3966646813579512,
"grad_norm": 1.1875,
"learning_rate": 0.0004991250534721959,
"loss": 5.4535,
"mean_token_accuracy": 0.16024698466062545,
"num_tokens": 9285184.0,
"step": 3330
},
{
"entropy": 5.513273859024048,
"epoch": 0.3972602739726027,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991212951417431,
"loss": 5.269,
"mean_token_accuracy": 0.1805974245071411,
"num_tokens": 9298622.0,
"step": 3335
},
{
"entropy": 5.5300816059112545,
"epoch": 0.3978558665872543,
"grad_norm": 0.91015625,
"learning_rate": 0.0004991175287724239,
"loss": 5.4262,
"mean_token_accuracy": 0.16618907749652861,
"num_tokens": 9314653.0,
"step": 3340
},
{
"entropy": 5.4713304996490475,
"epoch": 0.3984514592019059,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991137543643733,
"loss": 5.2669,
"mean_token_accuracy": 0.1765909180045128,
"num_tokens": 9328428.0,
"step": 3345
},
{
"entropy": 5.549885082244873,
"epoch": 0.3990470518165575,
"grad_norm": 0.921875,
"learning_rate": 0.0004991099719177268,
"loss": 5.3961,
"mean_token_accuracy": 0.1682168647646904,
"num_tokens": 9341250.0,
"step": 3350
},
{
"entropy": 5.48319182395935,
"epoch": 0.39964264443120906,
"grad_norm": 0.9375,
"learning_rate": 0.0004991061814326198,
"loss": 5.257,
"mean_token_accuracy": 0.1751708373427391,
"num_tokens": 9355706.0,
"step": 3355
},
{
"entropy": 5.614543342590332,
"epoch": 0.40023823704586065,
"grad_norm": 1.0,
"learning_rate": 0.0004991023829091887,
"loss": 5.4244,
"mean_token_accuracy": 0.1599671334028244,
"num_tokens": 9368562.0,
"step": 3360
},
{
"entropy": 5.552252292633057,
"epoch": 0.40083382966051223,
"grad_norm": 1.078125,
"learning_rate": 0.0004990985763475694,
"loss": 5.3421,
"mean_token_accuracy": 0.16700713187456132,
"num_tokens": 9383484.0,
"step": 3365
},
{
"entropy": 5.362362194061279,
"epoch": 0.40142942227516376,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990947617478984,
"loss": 5.1191,
"mean_token_accuracy": 0.17983500510454178,
"num_tokens": 9397364.0,
"step": 3370
},
{
"entropy": 5.38932056427002,
"epoch": 0.40202501488981535,
"grad_norm": 1.015625,
"learning_rate": 0.0004990909391103127,
"loss": 5.2506,
"mean_token_accuracy": 0.17546263486146926,
"num_tokens": 9411738.0,
"step": 3375
},
{
"entropy": 5.443359327316284,
"epoch": 0.40262060750446693,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990871084349494,
"loss": 5.223,
"mean_token_accuracy": 0.17884623557329177,
"num_tokens": 9426660.0,
"step": 3380
},
{
"entropy": 5.484179973602295,
"epoch": 0.4032162001191185,
"grad_norm": 1.015625,
"learning_rate": 0.0004990832697219459,
"loss": 5.2815,
"mean_token_accuracy": 0.1664903923869133,
"num_tokens": 9440878.0,
"step": 3385
},
{
"entropy": 5.4818016529083256,
"epoch": 0.4038117927337701,
"grad_norm": 1.03125,
"learning_rate": 0.0004990794229714398,
"loss": 5.2825,
"mean_token_accuracy": 0.16915637701749803,
"num_tokens": 9455174.0,
"step": 3390
},
{
"entropy": 5.443004894256592,
"epoch": 0.4044073853484217,
"grad_norm": 1.109375,
"learning_rate": 0.0004990755681835693,
"loss": 5.1979,
"mean_token_accuracy": 0.1759191244840622,
"num_tokens": 9467936.0,
"step": 3395
},
{
"entropy": 5.435940599441528,
"epoch": 0.4050029779630733,
"grad_norm": 1.078125,
"learning_rate": 0.0004990717053584723,
"loss": 5.3204,
"mean_token_accuracy": 0.17029665559530258,
"num_tokens": 9482529.0,
"step": 3400
},
{
"entropy": 5.501376104354859,
"epoch": 0.40559857057772486,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990678344962876,
"loss": 5.3383,
"mean_token_accuracy": 0.17074424624443055,
"num_tokens": 9497141.0,
"step": 3405
},
{
"entropy": 5.539723205566406,
"epoch": 0.4061941631923764,
"grad_norm": 1.09375,
"learning_rate": 0.0004990639555971539,
"loss": 5.3466,
"mean_token_accuracy": 0.1649772733449936,
"num_tokens": 9509878.0,
"step": 3410
},
{
"entropy": 5.456761884689331,
"epoch": 0.406789755807028,
"grad_norm": 0.91015625,
"learning_rate": 0.0004990600686612106,
"loss": 5.1912,
"mean_token_accuracy": 0.17851642966270448,
"num_tokens": 9522720.0,
"step": 3415
},
{
"entropy": 5.46306676864624,
"epoch": 0.40738534842167956,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990561736885968,
"loss": 5.3033,
"mean_token_accuracy": 0.17651566714048386,
"num_tokens": 9536418.0,
"step": 3420
},
{
"entropy": 5.485482597351075,
"epoch": 0.40798094103633115,
"grad_norm": 1.046875,
"learning_rate": 0.0004990522706794523,
"loss": 5.2147,
"mean_token_accuracy": 0.18023780584335328,
"num_tokens": 9552336.0,
"step": 3425
},
{
"entropy": 5.477826881408691,
"epoch": 0.40857653365098273,
"grad_norm": 1.046875,
"learning_rate": 0.0004990483596339174,
"loss": 5.2355,
"mean_token_accuracy": 0.17453182190656663,
"num_tokens": 9565676.0,
"step": 3430
},
{
"entropy": 5.376121377944946,
"epoch": 0.4091721262656343,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990444405521318,
"loss": 5.1208,
"mean_token_accuracy": 0.1855195179581642,
"num_tokens": 9579361.0,
"step": 3435
},
{
"entropy": 5.407404041290283,
"epoch": 0.4097677188802859,
"grad_norm": 1.1328125,
"learning_rate": 0.0004990405134342366,
"loss": 5.2085,
"mean_token_accuracy": 0.17755534797906875,
"num_tokens": 9591779.0,
"step": 3440
},
{
"entropy": 5.490596723556519,
"epoch": 0.4103633114949375,
"grad_norm": 0.98046875,
"learning_rate": 0.0004990365782803724,
"loss": 5.267,
"mean_token_accuracy": 0.17143639475107192,
"num_tokens": 9608353.0,
"step": 3445
},
{
"entropy": 5.445445871353149,
"epoch": 0.410958904109589,
"grad_norm": 1.0,
"learning_rate": 0.0004990326350906803,
"loss": 5.2079,
"mean_token_accuracy": 0.17303474694490434,
"num_tokens": 9623236.0,
"step": 3450
},
{
"entropy": 5.431187629699707,
"epoch": 0.4115544967242406,
"grad_norm": 1.046875,
"learning_rate": 0.0004990286838653019,
"loss": 5.2901,
"mean_token_accuracy": 0.17017227411270142,
"num_tokens": 9635746.0,
"step": 3455
},
{
"entropy": 5.423646306991577,
"epoch": 0.4121500893388922,
"grad_norm": 0.94140625,
"learning_rate": 0.0004990247246043789,
"loss": 5.1857,
"mean_token_accuracy": 0.17713408172130585,
"num_tokens": 9650458.0,
"step": 3460
},
{
"entropy": 5.4576317310333256,
"epoch": 0.4127456819535438,
"grad_norm": 0.953125,
"learning_rate": 0.0004990207573080533,
"loss": 5.2845,
"mean_token_accuracy": 0.18094676434993745,
"num_tokens": 9665486.0,
"step": 3465
},
{
"entropy": 5.521482372283936,
"epoch": 0.41334127456819536,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990167819764672,
"loss": 5.3522,
"mean_token_accuracy": 0.16958900392055512,
"num_tokens": 9679977.0,
"step": 3470
},
{
"entropy": 5.488754367828369,
"epoch": 0.41393686718284695,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990127986097636,
"loss": 5.3221,
"mean_token_accuracy": 0.16972343921661376,
"num_tokens": 9695092.0,
"step": 3475
},
{
"entropy": 5.354253053665161,
"epoch": 0.41453245979749853,
"grad_norm": 0.984375,
"learning_rate": 0.000499008807208085,
"loss": 5.1623,
"mean_token_accuracy": 0.18037663996219636,
"num_tokens": 9707836.0,
"step": 3480
},
{
"entropy": 5.510741806030273,
"epoch": 0.4151280524121501,
"grad_norm": 1.203125,
"learning_rate": 0.0004990048077715746,
"loss": 5.3223,
"mean_token_accuracy": 0.16598230898380278,
"num_tokens": 9721936.0,
"step": 3485
},
{
"entropy": 5.432139110565186,
"epoch": 0.41572364502680165,
"grad_norm": 1.015625,
"learning_rate": 0.0004990008003003762,
"loss": 5.2033,
"mean_token_accuracy": 0.17956025302410125,
"num_tokens": 9734876.0,
"step": 3490
},
{
"entropy": 5.455330467224121,
"epoch": 0.41631923764145323,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989967847946331,
"loss": 5.423,
"mean_token_accuracy": 0.16417320519685746,
"num_tokens": 9748870.0,
"step": 3495
},
{
"entropy": 5.569710540771484,
"epoch": 0.4169148302561048,
"grad_norm": 0.8828125,
"learning_rate": 0.0004989927612544895,
"loss": 5.323,
"mean_token_accuracy": 0.1703778013586998,
"num_tokens": 9763605.0,
"step": 3500
},
{
"entropy": 5.530073738098144,
"epoch": 0.4175104228707564,
"grad_norm": 1.0234375,
"learning_rate": 0.0004989887296800899,
"loss": 5.3106,
"mean_token_accuracy": 0.17459381520748138,
"num_tokens": 9779312.0,
"step": 3505
},
{
"entropy": 5.442477798461914,
"epoch": 0.418106015485408,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989846900715786,
"loss": 5.3109,
"mean_token_accuracy": 0.16355077177286148,
"num_tokens": 9791770.0,
"step": 3510
},
{
"entropy": 5.451168346405029,
"epoch": 0.4187016081000596,
"grad_norm": 1.09375,
"learning_rate": 0.0004989806424291008,
"loss": 5.2246,
"mean_token_accuracy": 0.17671115100383758,
"num_tokens": 9807341.0,
"step": 3515
},
{
"entropy": 5.448538446426392,
"epoch": 0.41929720071471116,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989765867528014,
"loss": 5.2285,
"mean_token_accuracy": 0.17702449560165406,
"num_tokens": 9821829.0,
"step": 3520
},
{
"entropy": 5.366397714614868,
"epoch": 0.4198927933293627,
"grad_norm": 1.0546875,
"learning_rate": 0.0004989725230428262,
"loss": 5.0954,
"mean_token_accuracy": 0.18542798310518266,
"num_tokens": 9836554.0,
"step": 3525
},
{
"entropy": 5.446051073074341,
"epoch": 0.4204883859440143,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989684512993205,
"loss": 5.2543,
"mean_token_accuracy": 0.18016983121633529,
"num_tokens": 9849904.0,
"step": 3530
},
{
"entropy": 5.431139087677002,
"epoch": 0.42108397855866586,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989643715224307,
"loss": 5.2283,
"mean_token_accuracy": 0.17674075067043304,
"num_tokens": 9863935.0,
"step": 3535
},
{
"entropy": 5.431934213638305,
"epoch": 0.42167957117331745,
"grad_norm": 0.99609375,
"learning_rate": 0.0004989602837123031,
"loss": 5.2876,
"mean_token_accuracy": 0.17591250240802764,
"num_tokens": 9878230.0,
"step": 3540
},
{
"entropy": 5.410821104049683,
"epoch": 0.42227516378796903,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989561878690842,
"loss": 5.1544,
"mean_token_accuracy": 0.18137411326169967,
"num_tokens": 9891411.0,
"step": 3545
},
{
"entropy": 5.358261919021606,
"epoch": 0.4228707564026206,
"grad_norm": 0.97265625,
"learning_rate": 0.000498952083992921,
"loss": 5.1431,
"mean_token_accuracy": 0.17786334007978438,
"num_tokens": 9904891.0,
"step": 3550
},
{
"entropy": 5.3938836574554445,
"epoch": 0.4234663490172722,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989479720839605,
"loss": 5.2969,
"mean_token_accuracy": 0.17019854635000228,
"num_tokens": 9918920.0,
"step": 3555
},
{
"entropy": 5.451056623458863,
"epoch": 0.4240619416319238,
"grad_norm": 0.921875,
"learning_rate": 0.0004989438521423506,
"loss": 5.189,
"mean_token_accuracy": 0.17660196423530578,
"num_tokens": 9932975.0,
"step": 3560
},
{
"entropy": 5.444226360321045,
"epoch": 0.4246575342465753,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989397241682387,
"loss": 5.2737,
"mean_token_accuracy": 0.17603071331977843,
"num_tokens": 9946704.0,
"step": 3565
},
{
"entropy": 5.416390752792358,
"epoch": 0.4252531268612269,
"grad_norm": 1.0,
"learning_rate": 0.000498935588161773,
"loss": 5.239,
"mean_token_accuracy": 0.173039610683918,
"num_tokens": 9961729.0,
"step": 3570
},
{
"entropy": 5.4772782802581785,
"epoch": 0.4258487194758785,
"grad_norm": 0.9453125,
"learning_rate": 0.0004989314441231019,
"loss": 5.3212,
"mean_token_accuracy": 0.17304217368364333,
"num_tokens": 9976980.0,
"step": 3575
},
{
"entropy": 5.51420111656189,
"epoch": 0.4264443120905301,
"grad_norm": 1.0,
"learning_rate": 0.0004989272920523741,
"loss": 5.2568,
"mean_token_accuracy": 0.17432699352502823,
"num_tokens": 9991963.0,
"step": 3580
},
{
"entropy": 5.472890758514405,
"epoch": 0.42703990470518166,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989231319497382,
"loss": 5.2671,
"mean_token_accuracy": 0.1745639756321907,
"num_tokens": 10005323.0,
"step": 3585
},
{
"entropy": 5.40621337890625,
"epoch": 0.42763549731983325,
"grad_norm": 1.0859375,
"learning_rate": 0.0004989189638153437,
"loss": 5.1972,
"mean_token_accuracy": 0.17891811579465866,
"num_tokens": 10018352.0,
"step": 3590
},
{
"entropy": 5.438979244232177,
"epoch": 0.42823108993448483,
"grad_norm": 0.96484375,
"learning_rate": 0.0004989147876493401,
"loss": 5.2265,
"mean_token_accuracy": 0.17730545848608018,
"num_tokens": 10032371.0,
"step": 3595
},
{
"entropy": 5.524429130554199,
"epoch": 0.4288266825491364,
"grad_norm": 1.09375,
"learning_rate": 0.0004989106034518771,
"loss": 5.34,
"mean_token_accuracy": 0.17437071353197098,
"num_tokens": 10044221.0,
"step": 3600
},
{
"entropy": 5.5083776950836185,
"epoch": 0.42942227516378795,
"grad_norm": 0.96875,
"learning_rate": 0.0004989064112231049,
"loss": 5.2818,
"mean_token_accuracy": 0.17418865263462066,
"num_tokens": 10057475.0,
"step": 3605
},
{
"entropy": 5.439066696166992,
"epoch": 0.43001786777843953,
"grad_norm": 1.0546875,
"learning_rate": 0.0004989022109631736,
"loss": 5.2791,
"mean_token_accuracy": 0.17699487656354904,
"num_tokens": 10072272.0,
"step": 3610
},
{
"entropy": 5.466569089889527,
"epoch": 0.4306134603930911,
"grad_norm": 0.9453125,
"learning_rate": 0.0004988980026722341,
"loss": 5.3048,
"mean_token_accuracy": 0.16676444858312606,
"num_tokens": 10086757.0,
"step": 3615
},
{
"entropy": 5.522718000411987,
"epoch": 0.4312090530077427,
"grad_norm": 0.99609375,
"learning_rate": 0.0004988937863504373,
"loss": 5.3068,
"mean_token_accuracy": 0.1667618930339813,
"num_tokens": 10100471.0,
"step": 3620
},
{
"entropy": 5.505567789077759,
"epoch": 0.4318046456223943,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988895619979344,
"loss": 5.2956,
"mean_token_accuracy": 0.17292843610048295,
"num_tokens": 10112764.0,
"step": 3625
},
{
"entropy": 5.445482921600342,
"epoch": 0.4324002382370459,
"grad_norm": 1.046875,
"learning_rate": 0.000498885329614877,
"loss": 5.2074,
"mean_token_accuracy": 0.1857913389801979,
"num_tokens": 10125830.0,
"step": 3630
},
{
"entropy": 5.316396999359131,
"epoch": 0.43299583085169746,
"grad_norm": 1.0859375,
"learning_rate": 0.0004988810892014168,
"loss": 5.1938,
"mean_token_accuracy": 0.18059258460998534,
"num_tokens": 10140007.0,
"step": 3635
},
{
"entropy": 5.438551616668701,
"epoch": 0.433591423466349,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988768407577058,
"loss": 5.2026,
"mean_token_accuracy": 0.18052507936954498,
"num_tokens": 10154508.0,
"step": 3640
},
{
"entropy": 5.468423318862915,
"epoch": 0.4341870160810006,
"grad_norm": 1.0390625,
"learning_rate": 0.0004988725842838968,
"loss": 5.238,
"mean_token_accuracy": 0.17988714724779128,
"num_tokens": 10167832.0,
"step": 3645
},
{
"entropy": 5.416162157058716,
"epoch": 0.43478260869565216,
"grad_norm": 1.0234375,
"learning_rate": 0.000498868319780142,
"loss": 5.2623,
"mean_token_accuracy": 0.1766108751296997,
"num_tokens": 10182421.0,
"step": 3650
},
{
"entropy": 5.394202375411988,
"epoch": 0.43537820131030375,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988640472465947,
"loss": 5.1765,
"mean_token_accuracy": 0.17946621775627136,
"num_tokens": 10196210.0,
"step": 3655
},
{
"entropy": 5.432455539703369,
"epoch": 0.43597379392495533,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988597666834078,
"loss": 5.2386,
"mean_token_accuracy": 0.17361632287502288,
"num_tokens": 10210930.0,
"step": 3660
},
{
"entropy": 5.4363268375396725,
"epoch": 0.4365693865396069,
"grad_norm": 0.984375,
"learning_rate": 0.0004988554780907353,
"loss": 5.2144,
"mean_token_accuracy": 0.179244963824749,
"num_tokens": 10225758.0,
"step": 3665
},
{
"entropy": 5.476582765579224,
"epoch": 0.4371649791542585,
"grad_norm": 1.0625,
"learning_rate": 0.0004988511814687306,
"loss": 5.3228,
"mean_token_accuracy": 0.17901217341423034,
"num_tokens": 10239924.0,
"step": 3670
},
{
"entropy": 5.456946849822998,
"epoch": 0.4377605717689101,
"grad_norm": 1.015625,
"learning_rate": 0.0004988468768175479,
"loss": 5.1868,
"mean_token_accuracy": 0.17777508199214936,
"num_tokens": 10254713.0,
"step": 3675
},
{
"entropy": 5.456771469116211,
"epoch": 0.4383561643835616,
"grad_norm": 1.0703125,
"learning_rate": 0.0004988425641373418,
"loss": 5.3043,
"mean_token_accuracy": 0.1681886374950409,
"num_tokens": 10270335.0,
"step": 3680
},
{
"entropy": 5.4499236106872555,
"epoch": 0.4389517569982132,
"grad_norm": 0.96875,
"learning_rate": 0.0004988382434282668,
"loss": 5.2428,
"mean_token_accuracy": 0.1767041116952896,
"num_tokens": 10286368.0,
"step": 3685
},
{
"entropy": 5.4764598369598385,
"epoch": 0.4395473496128648,
"grad_norm": 0.9296875,
"learning_rate": 0.000498833914690478,
"loss": 5.2904,
"mean_token_accuracy": 0.17117395251989365,
"num_tokens": 10302073.0,
"step": 3690
},
{
"entropy": 5.372010898590088,
"epoch": 0.4401429422275164,
"grad_norm": 1.0625,
"learning_rate": 0.0004988295779241306,
"loss": 5.1514,
"mean_token_accuracy": 0.1833252027630806,
"num_tokens": 10317741.0,
"step": 3695
},
{
"entropy": 5.432416534423828,
"epoch": 0.44073853484216796,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988252331293802,
"loss": 5.2961,
"mean_token_accuracy": 0.174821637570858,
"num_tokens": 10331439.0,
"step": 3700
},
{
"entropy": 5.539056968688965,
"epoch": 0.44133412745681955,
"grad_norm": 1.0,
"learning_rate": 0.0004988208803063827,
"loss": 5.3845,
"mean_token_accuracy": 0.16558377742767333,
"num_tokens": 10345107.0,
"step": 3705
},
{
"entropy": 5.499377059936523,
"epoch": 0.44192972007147113,
"grad_norm": 0.95703125,
"learning_rate": 0.0004988165194552939,
"loss": 5.2814,
"mean_token_accuracy": 0.17398845851421357,
"num_tokens": 10360078.0,
"step": 3710
},
{
"entropy": 5.386675119400024,
"epoch": 0.4425253126861227,
"grad_norm": 1.0078125,
"learning_rate": 0.0004988121505762706,
"loss": 5.2649,
"mean_token_accuracy": 0.17178381532430648,
"num_tokens": 10373221.0,
"step": 3715
},
{
"entropy": 5.515303659439087,
"epoch": 0.44312090530077425,
"grad_norm": 1.03125,
"learning_rate": 0.0004988077736694693,
"loss": 5.3958,
"mean_token_accuracy": 0.16752853691577912,
"num_tokens": 10386200.0,
"step": 3720
},
{
"entropy": 5.400387859344482,
"epoch": 0.44371649791542583,
"grad_norm": 1.046875,
"learning_rate": 0.0004988033887350471,
"loss": 5.2912,
"mean_token_accuracy": 0.1776488721370697,
"num_tokens": 10398680.0,
"step": 3725
},
{
"entropy": 5.4428901195526125,
"epoch": 0.4443120905300774,
"grad_norm": 0.8984375,
"learning_rate": 0.0004987989957731611,
"loss": 5.2769,
"mean_token_accuracy": 0.1718101069331169,
"num_tokens": 10414483.0,
"step": 3730
},
{
"entropy": 5.497593736648559,
"epoch": 0.444907683144729,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987945947839691,
"loss": 5.2713,
"mean_token_accuracy": 0.1737542986869812,
"num_tokens": 10428962.0,
"step": 3735
},
{
"entropy": 5.584635305404663,
"epoch": 0.4455032757593806,
"grad_norm": 1.078125,
"learning_rate": 0.000498790185767629,
"loss": 5.3663,
"mean_token_accuracy": 0.17216833904385567,
"num_tokens": 10442273.0,
"step": 3740
},
{
"entropy": 5.537905216217041,
"epoch": 0.4460988683740322,
"grad_norm": 0.8671875,
"learning_rate": 0.0004987857687242986,
"loss": 5.2825,
"mean_token_accuracy": 0.17508672028779984,
"num_tokens": 10456610.0,
"step": 3745
},
{
"entropy": 5.387487459182739,
"epoch": 0.44669446098868376,
"grad_norm": 0.953125,
"learning_rate": 0.0004987813436541366,
"loss": 5.1734,
"mean_token_accuracy": 0.17823742032051088,
"num_tokens": 10470144.0,
"step": 3750
},
{
"entropy": 5.4657457828521725,
"epoch": 0.4472900536033353,
"grad_norm": 0.97265625,
"learning_rate": 0.0004987769105573016,
"loss": 5.1818,
"mean_token_accuracy": 0.18805933594703675,
"num_tokens": 10483866.0,
"step": 3755
},
{
"entropy": 5.422224617004394,
"epoch": 0.4478856462179869,
"grad_norm": 1.015625,
"learning_rate": 0.0004987724694339526,
"loss": 5.1855,
"mean_token_accuracy": 0.17784465104341507,
"num_tokens": 10497350.0,
"step": 3760
},
{
"entropy": 5.378564071655274,
"epoch": 0.44848123883263846,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987680202842491,
"loss": 5.279,
"mean_token_accuracy": 0.1758370280265808,
"num_tokens": 10510808.0,
"step": 3765
},
{
"entropy": 5.412915515899658,
"epoch": 0.44907683144729005,
"grad_norm": 1.0234375,
"learning_rate": 0.0004987635631083505,
"loss": 5.1006,
"mean_token_accuracy": 0.18201710134744645,
"num_tokens": 10525965.0,
"step": 3770
},
{
"entropy": 5.467325448989868,
"epoch": 0.44967242406194163,
"grad_norm": 0.984375,
"learning_rate": 0.0004987590979064167,
"loss": 5.2719,
"mean_token_accuracy": 0.17504528164863586,
"num_tokens": 10539900.0,
"step": 3775
},
{
"entropy": 5.454932785034179,
"epoch": 0.4502680166765932,
"grad_norm": 0.9609375,
"learning_rate": 0.0004987546246786078,
"loss": 5.3028,
"mean_token_accuracy": 0.1722474977374077,
"num_tokens": 10553653.0,
"step": 3780
},
{
"entropy": 5.374240446090698,
"epoch": 0.4508636092912448,
"grad_norm": 0.921875,
"learning_rate": 0.0004987501434250844,
"loss": 5.187,
"mean_token_accuracy": 0.18090267479419708,
"num_tokens": 10568202.0,
"step": 3785
},
{
"entropy": 5.339132022857666,
"epoch": 0.4514592019058964,
"grad_norm": 1.09375,
"learning_rate": 0.000498745654146007,
"loss": 5.1498,
"mean_token_accuracy": 0.1799025222659111,
"num_tokens": 10581290.0,
"step": 3790
},
{
"entropy": 5.394349145889282,
"epoch": 0.4520547945205479,
"grad_norm": 1.0078125,
"learning_rate": 0.000498741156841537,
"loss": 5.191,
"mean_token_accuracy": 0.1787856101989746,
"num_tokens": 10596697.0,
"step": 3795
},
{
"entropy": 5.4527936458587645,
"epoch": 0.4526503871351995,
"grad_norm": 0.96484375,
"learning_rate": 0.0004987366515118353,
"loss": 5.2483,
"mean_token_accuracy": 0.17962846159934998,
"num_tokens": 10610052.0,
"step": 3800
},
{
"entropy": 5.407793474197388,
"epoch": 0.4532459797498511,
"grad_norm": 0.99609375,
"learning_rate": 0.0004987321381570638,
"loss": 5.1948,
"mean_token_accuracy": 0.1826455920934677,
"num_tokens": 10624268.0,
"step": 3805
},
{
"entropy": 5.324577283859253,
"epoch": 0.4538415723645027,
"grad_norm": 1.03125,
"learning_rate": 0.0004987276167773842,
"loss": 5.113,
"mean_token_accuracy": 0.18459611386060715,
"num_tokens": 10636463.0,
"step": 3810
},
{
"entropy": 5.426402425765991,
"epoch": 0.45443716497915426,
"grad_norm": 1.0390625,
"learning_rate": 0.0004987230873729588,
"loss": 5.1798,
"mean_token_accuracy": 0.18134627342224122,
"num_tokens": 10649566.0,
"step": 3815
},
{
"entropy": 5.405946397781372,
"epoch": 0.45503275759380585,
"grad_norm": 1.1328125,
"learning_rate": 0.00049871854994395,
"loss": 5.2294,
"mean_token_accuracy": 0.18083360195159912,
"num_tokens": 10661373.0,
"step": 3820
},
{
"entropy": 5.371559953689575,
"epoch": 0.45562835020845743,
"grad_norm": 0.91015625,
"learning_rate": 0.0004987140044905205,
"loss": 5.1454,
"mean_token_accuracy": 0.18358622789382933,
"num_tokens": 10677008.0,
"step": 3825
},
{
"entropy": 5.480168676376342,
"epoch": 0.456223942823109,
"grad_norm": 1.171875,
"learning_rate": 0.0004987094510128335,
"loss": 5.2871,
"mean_token_accuracy": 0.17987342029809952,
"num_tokens": 10690497.0,
"step": 3830
},
{
"entropy": 5.388162517547608,
"epoch": 0.45681953543776055,
"grad_norm": 0.9140625,
"learning_rate": 0.0004987048895110522,
"loss": 5.225,
"mean_token_accuracy": 0.1827029123902321,
"num_tokens": 10705315.0,
"step": 3835
},
{
"entropy": 5.373668193817139,
"epoch": 0.45741512805241213,
"grad_norm": 0.97265625,
"learning_rate": 0.0004987003199853401,
"loss": 5.1678,
"mean_token_accuracy": 0.18545912057161332,
"num_tokens": 10718536.0,
"step": 3840
},
{
"entropy": 5.400412559509277,
"epoch": 0.4580107206670637,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986957424358613,
"loss": 5.2552,
"mean_token_accuracy": 0.17034007608890533,
"num_tokens": 10735402.0,
"step": 3845
},
{
"entropy": 5.486510419845581,
"epoch": 0.4586063132817153,
"grad_norm": 1.0,
"learning_rate": 0.00049869115686278,
"loss": 5.2281,
"mean_token_accuracy": 0.17834403812885286,
"num_tokens": 10748809.0,
"step": 3850
},
{
"entropy": 5.331592845916748,
"epoch": 0.4592019058963669,
"grad_norm": 1.0546875,
"learning_rate": 0.0004986865632662606,
"loss": 5.1254,
"mean_token_accuracy": 0.1848363310098648,
"num_tokens": 10761935.0,
"step": 3855
},
{
"entropy": 5.398069715499878,
"epoch": 0.4597974985110185,
"grad_norm": 0.96484375,
"learning_rate": 0.0004986819616464678,
"loss": 5.2504,
"mean_token_accuracy": 0.1697085589170456,
"num_tokens": 10778252.0,
"step": 3860
},
{
"entropy": 5.540862083435059,
"epoch": 0.46039309112567006,
"grad_norm": 1.046875,
"learning_rate": 0.0004986773520035667,
"loss": 5.3641,
"mean_token_accuracy": 0.17047490328550338,
"num_tokens": 10792606.0,
"step": 3865
},
{
"entropy": 5.447160911560059,
"epoch": 0.46098868374032165,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986727343377227,
"loss": 5.2753,
"mean_token_accuracy": 0.17300592213869095,
"num_tokens": 10805433.0,
"step": 3870
},
{
"entropy": 5.388632154464721,
"epoch": 0.4615842763549732,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986681086491014,
"loss": 5.2504,
"mean_token_accuracy": 0.1762865036725998,
"num_tokens": 10818648.0,
"step": 3875
},
{
"entropy": 5.432145118713379,
"epoch": 0.46217986896962476,
"grad_norm": 0.9765625,
"learning_rate": 0.0004986634749378686,
"loss": 5.1115,
"mean_token_accuracy": 0.17859960943460465,
"num_tokens": 10833089.0,
"step": 3880
},
{
"entropy": 5.358439779281616,
"epoch": 0.46277546158427635,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986588332041908,
"loss": 5.1783,
"mean_token_accuracy": 0.17855484187602996,
"num_tokens": 10847850.0,
"step": 3885
},
{
"entropy": 5.457101726531983,
"epoch": 0.46337105419892793,
"grad_norm": 0.9609375,
"learning_rate": 0.0004986541834482342,
"loss": 5.2361,
"mean_token_accuracy": 0.17978406846523284,
"num_tokens": 10861354.0,
"step": 3890
},
{
"entropy": 5.506772565841675,
"epoch": 0.4639666468135795,
"grad_norm": 0.859375,
"learning_rate": 0.0004986495256701657,
"loss": 5.2582,
"mean_token_accuracy": 0.16890907734632493,
"num_tokens": 10877986.0,
"step": 3895
},
{
"entropy": 5.310773229598999,
"epoch": 0.4645622394282311,
"grad_norm": 0.984375,
"learning_rate": 0.0004986448598701523,
"loss": 5.14,
"mean_token_accuracy": 0.18601099401712418,
"num_tokens": 10891253.0,
"step": 3900
},
{
"entropy": 5.398805141448975,
"epoch": 0.4651578320428827,
"grad_norm": 0.9921875,
"learning_rate": 0.0004986401860483615,
"loss": 5.2598,
"mean_token_accuracy": 0.17864535003900528,
"num_tokens": 10905916.0,
"step": 3905
},
{
"entropy": 5.414415884017944,
"epoch": 0.4657534246575342,
"grad_norm": 1.03125,
"learning_rate": 0.0004986355042049608,
"loss": 5.1195,
"mean_token_accuracy": 0.18294261991977692,
"num_tokens": 10920038.0,
"step": 3910
},
{
"entropy": 5.3523149490356445,
"epoch": 0.4663490172721858,
"grad_norm": 1.0859375,
"learning_rate": 0.0004986308143401181,
"loss": 5.1361,
"mean_token_accuracy": 0.18407817482948302,
"num_tokens": 10933161.0,
"step": 3915
},
{
"entropy": 5.384498882293701,
"epoch": 0.4669446098868374,
"grad_norm": 0.99609375,
"learning_rate": 0.0004986261164540018,
"loss": 5.2238,
"mean_token_accuracy": 0.17449699640274047,
"num_tokens": 10946573.0,
"step": 3920
},
{
"entropy": 5.414169502258301,
"epoch": 0.467540202501489,
"grad_norm": 0.95703125,
"learning_rate": 0.0004986214105467801,
"loss": 5.189,
"mean_token_accuracy": 0.17691306173801422,
"num_tokens": 10960727.0,
"step": 3925
},
{
"entropy": 5.478633069992066,
"epoch": 0.46813579511614056,
"grad_norm": 1.015625,
"learning_rate": 0.0004986166966186223,
"loss": 5.2607,
"mean_token_accuracy": 0.17322907149791716,
"num_tokens": 10973298.0,
"step": 3930
},
{
"entropy": 5.416499090194702,
"epoch": 0.46873138773079215,
"grad_norm": 0.98046875,
"learning_rate": 0.0004986119746696969,
"loss": 5.2629,
"mean_token_accuracy": 0.17739322781562805,
"num_tokens": 10988214.0,
"step": 3935
},
{
"entropy": 5.470095634460449,
"epoch": 0.46932698034544373,
"grad_norm": 1.046875,
"learning_rate": 0.0004986072447001736,
"loss": 5.2589,
"mean_token_accuracy": 0.17839152961969376,
"num_tokens": 11003526.0,
"step": 3940
},
{
"entropy": 5.49106731414795,
"epoch": 0.4699225729600953,
"grad_norm": 0.9765625,
"learning_rate": 0.0004986025067102221,
"loss": 5.3526,
"mean_token_accuracy": 0.16881988793611527,
"num_tokens": 11019600.0,
"step": 3945
},
{
"entropy": 5.332584953308105,
"epoch": 0.47051816557474685,
"grad_norm": 1.0546875,
"learning_rate": 0.0004985977607000121,
"loss": 5.0848,
"mean_token_accuracy": 0.18336566388607026,
"num_tokens": 11032591.0,
"step": 3950
},
{
"entropy": 5.344310665130616,
"epoch": 0.47111375818939843,
"grad_norm": 1.1875,
"learning_rate": 0.0004985930066697141,
"loss": 5.1199,
"mean_token_accuracy": 0.18219054490327835,
"num_tokens": 11045724.0,
"step": 3955
},
{
"entropy": 5.479485273361206,
"epoch": 0.47170935080405,
"grad_norm": 0.96875,
"learning_rate": 0.0004985882446194984,
"loss": 5.2311,
"mean_token_accuracy": 0.17386580258607864,
"num_tokens": 11059149.0,
"step": 3960
},
{
"entropy": 5.420220232009887,
"epoch": 0.4723049434187016,
"grad_norm": 0.98828125,
"learning_rate": 0.000498583474549536,
"loss": 5.246,
"mean_token_accuracy": 0.17556451112031937,
"num_tokens": 11073489.0,
"step": 3965
},
{
"entropy": 5.43749771118164,
"epoch": 0.4729005360333532,
"grad_norm": 1.109375,
"learning_rate": 0.0004985786964599978,
"loss": 5.1907,
"mean_token_accuracy": 0.17946831583976747,
"num_tokens": 11086535.0,
"step": 3970
},
{
"entropy": 5.324299335479736,
"epoch": 0.4734961286480048,
"grad_norm": 0.96484375,
"learning_rate": 0.0004985739103510553,
"loss": 5.0832,
"mean_token_accuracy": 0.18641428798437118,
"num_tokens": 11100142.0,
"step": 3975
},
{
"entropy": 5.294971179962158,
"epoch": 0.47409172126265636,
"grad_norm": 0.93359375,
"learning_rate": 0.0004985691162228802,
"loss": 5.1102,
"mean_token_accuracy": 0.18939329385757447,
"num_tokens": 11113364.0,
"step": 3980
},
{
"entropy": 5.502555322647095,
"epoch": 0.47468731387730795,
"grad_norm": 1.1171875,
"learning_rate": 0.0004985643140756442,
"loss": 5.2677,
"mean_token_accuracy": 0.17518739253282548,
"num_tokens": 11127127.0,
"step": 3985
},
{
"entropy": 5.461540746688843,
"epoch": 0.4752829064919595,
"grad_norm": 1.03125,
"learning_rate": 0.00049855950390952,
"loss": 5.1966,
"mean_token_accuracy": 0.17829690128564835,
"num_tokens": 11139788.0,
"step": 3990
},
{
"entropy": 5.389543247222901,
"epoch": 0.47587849910661106,
"grad_norm": 0.8984375,
"learning_rate": 0.0004985546857246797,
"loss": 5.2325,
"mean_token_accuracy": 0.1797940105199814,
"num_tokens": 11156470.0,
"step": 3995
},
{
"entropy": 5.365475416183472,
"epoch": 0.47647409172126265,
"grad_norm": 1.0078125,
"learning_rate": 0.0004985498595212964,
"loss": 5.1048,
"mean_token_accuracy": 0.18815007209777831,
"num_tokens": 11168923.0,
"step": 4000
},
{
"entropy": 5.386591386795044,
"epoch": 0.47706968433591423,
"grad_norm": 0.99609375,
"learning_rate": 0.0004985450252995431,
"loss": 5.2526,
"mean_token_accuracy": 0.1842184230685234,
"num_tokens": 11181328.0,
"step": 4005
},
{
"entropy": 5.487314748764038,
"epoch": 0.4776652769505658,
"grad_norm": 0.828125,
"learning_rate": 0.0004985401830595932,
"loss": 5.3359,
"mean_token_accuracy": 0.17080096155405045,
"num_tokens": 11197800.0,
"step": 4010
},
{
"entropy": 5.52086763381958,
"epoch": 0.4782608695652174,
"grad_norm": 0.99609375,
"learning_rate": 0.0004985353328016204,
"loss": 5.2103,
"mean_token_accuracy": 0.17595488131046294,
"num_tokens": 11209862.0,
"step": 4015
},
{
"entropy": 5.353274869918823,
"epoch": 0.478856462179869,
"grad_norm": 1.03125,
"learning_rate": 0.0004985304745257986,
"loss": 5.2267,
"mean_token_accuracy": 0.17607441693544387,
"num_tokens": 11223433.0,
"step": 4020
},
{
"entropy": 5.4463588237762455,
"epoch": 0.4794520547945205,
"grad_norm": 1.03125,
"learning_rate": 0.0004985256082323022,
"loss": 5.266,
"mean_token_accuracy": 0.17377653568983079,
"num_tokens": 11236975.0,
"step": 4025
},
{
"entropy": 5.379982042312622,
"epoch": 0.4800476474091721,
"grad_norm": 1.0859375,
"learning_rate": 0.0004985207339213057,
"loss": 5.2173,
"mean_token_accuracy": 0.17730392664670944,
"num_tokens": 11250277.0,
"step": 4030
},
{
"entropy": 5.441099500656128,
"epoch": 0.4806432400238237,
"grad_norm": 1.03125,
"learning_rate": 0.0004985158515929837,
"loss": 5.2343,
"mean_token_accuracy": 0.17686755508184432,
"num_tokens": 11264990.0,
"step": 4035
},
{
"entropy": 5.46997332572937,
"epoch": 0.4812388326384753,
"grad_norm": 0.96875,
"learning_rate": 0.0004985109612475116,
"loss": 5.3059,
"mean_token_accuracy": 0.17195749580860137,
"num_tokens": 11277496.0,
"step": 4040
},
{
"entropy": 5.407310247421265,
"epoch": 0.48183442525312686,
"grad_norm": 0.98828125,
"learning_rate": 0.0004985060628850648,
"loss": 5.2637,
"mean_token_accuracy": 0.17768220603466034,
"num_tokens": 11290263.0,
"step": 4045
},
{
"entropy": 5.399970149993896,
"epoch": 0.48243001786777845,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985011565058189,
"loss": 5.1854,
"mean_token_accuracy": 0.18622722029685973,
"num_tokens": 11305611.0,
"step": 4050
},
{
"entropy": 5.459443044662476,
"epoch": 0.48302561048243003,
"grad_norm": 1.0078125,
"learning_rate": 0.0004984962421099498,
"loss": 5.2414,
"mean_token_accuracy": 0.17523133903741836,
"num_tokens": 11321216.0,
"step": 4055
},
{
"entropy": 5.423429250717163,
"epoch": 0.4836212030970816,
"grad_norm": 1.0234375,
"learning_rate": 0.000498491319697634,
"loss": 5.142,
"mean_token_accuracy": 0.18697123378515243,
"num_tokens": 11335029.0,
"step": 4060
},
{
"entropy": 5.340970468521118,
"epoch": 0.48421679571173315,
"grad_norm": 0.9296875,
"learning_rate": 0.000498486389269048,
"loss": 5.2756,
"mean_token_accuracy": 0.1715432271361351,
"num_tokens": 11349301.0,
"step": 4065
},
{
"entropy": 5.455768823623657,
"epoch": 0.48481238832638474,
"grad_norm": 1.0234375,
"learning_rate": 0.0004984814508243685,
"loss": 5.2109,
"mean_token_accuracy": 0.17929711788892747,
"num_tokens": 11364561.0,
"step": 4070
},
{
"entropy": 5.359412908554077,
"epoch": 0.4854079809410363,
"grad_norm": 1.0390625,
"learning_rate": 0.0004984765043637727,
"loss": 5.1941,
"mean_token_accuracy": 0.1832022950053215,
"num_tokens": 11379470.0,
"step": 4075
},
{
"entropy": 5.407134914398194,
"epoch": 0.4860035735556879,
"grad_norm": 0.9765625,
"learning_rate": 0.000498471549887438,
"loss": 5.1928,
"mean_token_accuracy": 0.1830180376768112,
"num_tokens": 11393433.0,
"step": 4080
},
{
"entropy": 5.333357048034668,
"epoch": 0.4865991661703395,
"grad_norm": 1.109375,
"learning_rate": 0.0004984665873955422,
"loss": 5.1236,
"mean_token_accuracy": 0.18091286718845367,
"num_tokens": 11406673.0,
"step": 4085
},
{
"entropy": 5.341376066207886,
"epoch": 0.4871947587849911,
"grad_norm": 0.8984375,
"learning_rate": 0.0004984616168882632,
"loss": 5.1272,
"mean_token_accuracy": 0.18248946219682693,
"num_tokens": 11420395.0,
"step": 4090
},
{
"entropy": 5.427642154693603,
"epoch": 0.48779035139964266,
"grad_norm": 0.99609375,
"learning_rate": 0.0004984566383657794,
"loss": 5.2246,
"mean_token_accuracy": 0.1790100708603859,
"num_tokens": 11432866.0,
"step": 4095
},
{
"entropy": 5.349800157546997,
"epoch": 0.48838594401429425,
"grad_norm": 1.1953125,
"learning_rate": 0.0004984516518282692,
"loss": 5.2222,
"mean_token_accuracy": 0.17877718955278396,
"num_tokens": 11447062.0,
"step": 4100
},
{
"entropy": 5.344159936904907,
"epoch": 0.4889815366289458,
"grad_norm": 0.88671875,
"learning_rate": 0.0004984466572759117,
"loss": 5.2217,
"mean_token_accuracy": 0.1861804947257042,
"num_tokens": 11461255.0,
"step": 4105
},
{
"entropy": 5.463950109481812,
"epoch": 0.48957712924359736,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984416547088858,
"loss": 5.237,
"mean_token_accuracy": 0.17237258106470107,
"num_tokens": 11475711.0,
"step": 4110
},
{
"entropy": 5.41520471572876,
"epoch": 0.49017272185824895,
"grad_norm": 0.984375,
"learning_rate": 0.0004984366441273711,
"loss": 5.328,
"mean_token_accuracy": 0.17271950989961624,
"num_tokens": 11491199.0,
"step": 4115
},
{
"entropy": 5.448308992385864,
"epoch": 0.49076831447290054,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984316255315472,
"loss": 5.1565,
"mean_token_accuracy": 0.18378346115350724,
"num_tokens": 11504365.0,
"step": 4120
},
{
"entropy": 5.420531940460205,
"epoch": 0.4913639070875521,
"grad_norm": 0.96875,
"learning_rate": 0.0004984265989215942,
"loss": 5.2608,
"mean_token_accuracy": 0.17314012795686723,
"num_tokens": 11517969.0,
"step": 4125
},
{
"entropy": 5.394005966186524,
"epoch": 0.4919594997022037,
"grad_norm": 0.81640625,
"learning_rate": 0.0004984215642976923,
"loss": 5.1953,
"mean_token_accuracy": 0.17764792740345,
"num_tokens": 11533203.0,
"step": 4130
},
{
"entropy": 5.403848791122437,
"epoch": 0.4925550923168553,
"grad_norm": 0.9375,
"learning_rate": 0.0004984165216600224,
"loss": 5.1837,
"mean_token_accuracy": 0.1772610768675804,
"num_tokens": 11547820.0,
"step": 4135
},
{
"entropy": 5.391172599792481,
"epoch": 0.4931506849315068,
"grad_norm": 1.0,
"learning_rate": 0.0004984114710087648,
"loss": 5.1353,
"mean_token_accuracy": 0.18545330613851546,
"num_tokens": 11561591.0,
"step": 4140
},
{
"entropy": 5.346930599212646,
"epoch": 0.4937462775461584,
"grad_norm": 0.92578125,
"learning_rate": 0.0004984064123441011,
"loss": 5.2069,
"mean_token_accuracy": 0.18011981397867202,
"num_tokens": 11576722.0,
"step": 4145
},
{
"entropy": 5.481418132781982,
"epoch": 0.49434187016081,
"grad_norm": 0.99609375,
"learning_rate": 0.0004984013456662127,
"loss": 5.2582,
"mean_token_accuracy": 0.1792983591556549,
"num_tokens": 11590455.0,
"step": 4150
},
{
"entropy": 5.446556568145752,
"epoch": 0.4949374627754616,
"grad_norm": 0.98046875,
"learning_rate": 0.0004983962709752812,
"loss": 5.2152,
"mean_token_accuracy": 0.1737887904047966,
"num_tokens": 11605495.0,
"step": 4155
},
{
"entropy": 5.380440711975098,
"epoch": 0.49553305539011316,
"grad_norm": 1.015625,
"learning_rate": 0.0004983911882714886,
"loss": 5.1823,
"mean_token_accuracy": 0.17735483199357988,
"num_tokens": 11619692.0,
"step": 4160
},
{
"entropy": 5.365343761444092,
"epoch": 0.49612864800476475,
"grad_norm": 0.9453125,
"learning_rate": 0.0004983860975550174,
"loss": 5.2079,
"mean_token_accuracy": 0.18099919259548186,
"num_tokens": 11633806.0,
"step": 4165
},
{
"entropy": 5.377055406570435,
"epoch": 0.49672424061941634,
"grad_norm": 1.0703125,
"learning_rate": 0.0004983809988260501,
"loss": 5.1274,
"mean_token_accuracy": 0.18955023884773253,
"num_tokens": 11649871.0,
"step": 4170
},
{
"entropy": 5.34689154624939,
"epoch": 0.4973198332340679,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983758920847696,
"loss": 5.1735,
"mean_token_accuracy": 0.18082754909992219,
"num_tokens": 11661472.0,
"step": 4175
},
{
"entropy": 5.3021033763885494,
"epoch": 0.49791542584871945,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983707773313589,
"loss": 5.0834,
"mean_token_accuracy": 0.18494856208562852,
"num_tokens": 11674689.0,
"step": 4180
},
{
"entropy": 5.301540803909302,
"epoch": 0.49851101846337104,
"grad_norm": 0.96484375,
"learning_rate": 0.0004983656545660016,
"loss": 5.0266,
"mean_token_accuracy": 0.19291228652000428,
"num_tokens": 11687867.0,
"step": 4185
},
{
"entropy": 5.318424558639526,
"epoch": 0.4991066110780226,
"grad_norm": 0.8515625,
"learning_rate": 0.0004983605237888815,
"loss": 5.0979,
"mean_token_accuracy": 0.18405648320913315,
"num_tokens": 11704611.0,
"step": 4190
},
{
"entropy": 5.418560123443603,
"epoch": 0.4997022036926742,
"grad_norm": 0.921875,
"learning_rate": 0.0004983553850001826,
"loss": 5.2135,
"mean_token_accuracy": 0.17981443852186202,
"num_tokens": 11719448.0,
"step": 4195
},
{
"entropy": 5.359702920913696,
"epoch": 0.5002977963073257,
"grad_norm": 1.1328125,
"learning_rate": 0.0004983502382000892,
"loss": 5.1225,
"mean_token_accuracy": 0.18143897801637648,
"num_tokens": 11731160.0,
"step": 4200
},
{
"entropy": 5.373467493057251,
"epoch": 0.5008933889219773,
"grad_norm": 1.0546875,
"learning_rate": 0.0004983450833887859,
"loss": 5.172,
"mean_token_accuracy": 0.1843373268842697,
"num_tokens": 11744319.0,
"step": 4205
},
{
"entropy": 5.384470272064209,
"epoch": 0.5014889815366289,
"grad_norm": 1.0859375,
"learning_rate": 0.0004983399205664575,
"loss": 5.1865,
"mean_token_accuracy": 0.18225540816783906,
"num_tokens": 11757861.0,
"step": 4210
},
{
"entropy": 5.320976638793946,
"epoch": 0.5020845741512805,
"grad_norm": 1.0234375,
"learning_rate": 0.0004983347497332894,
"loss": 5.0898,
"mean_token_accuracy": 0.18314458876848222,
"num_tokens": 11770707.0,
"step": 4215
},
{
"entropy": 5.383945560455322,
"epoch": 0.5026801667659321,
"grad_norm": 1.046875,
"learning_rate": 0.0004983295708894668,
"loss": 5.1015,
"mean_token_accuracy": 0.18670070469379424,
"num_tokens": 11785155.0,
"step": 4220
},
{
"entropy": 5.373613739013672,
"epoch": 0.5032757593805837,
"grad_norm": 1.0546875,
"learning_rate": 0.0004983243840351757,
"loss": 5.1471,
"mean_token_accuracy": 0.17876102179288864,
"num_tokens": 11798532.0,
"step": 4225
},
{
"entropy": 5.24838604927063,
"epoch": 0.5038713519952353,
"grad_norm": 0.9765625,
"learning_rate": 0.0004983191891706021,
"loss": 5.1752,
"mean_token_accuracy": 0.184522907435894,
"num_tokens": 11812708.0,
"step": 4230
},
{
"entropy": 5.42040786743164,
"epoch": 0.5044669446098868,
"grad_norm": 0.97265625,
"learning_rate": 0.0004983139862959322,
"loss": 5.1605,
"mean_token_accuracy": 0.17628637254238128,
"num_tokens": 11828844.0,
"step": 4235
},
{
"entropy": 5.338803482055664,
"epoch": 0.5050625372245384,
"grad_norm": 0.99609375,
"learning_rate": 0.0004983087754113527,
"loss": 5.123,
"mean_token_accuracy": 0.18194308876991272,
"num_tokens": 11845070.0,
"step": 4240
},
{
"entropy": 5.299670934677124,
"epoch": 0.50565812983919,
"grad_norm": 1.0390625,
"learning_rate": 0.0004983035565170505,
"loss": 5.1489,
"mean_token_accuracy": 0.1859847143292427,
"num_tokens": 11858299.0,
"step": 4245
},
{
"entropy": 5.412613868713379,
"epoch": 0.5062537224538416,
"grad_norm": 1.0859375,
"learning_rate": 0.0004982983296132128,
"loss": 5.1934,
"mean_token_accuracy": 0.18599607646465302,
"num_tokens": 11873805.0,
"step": 4250
},
{
"entropy": 5.316448831558228,
"epoch": 0.5068493150684932,
"grad_norm": 0.9609375,
"learning_rate": 0.0004982930947000271,
"loss": 5.2395,
"mean_token_accuracy": 0.171941277384758,
"num_tokens": 11888304.0,
"step": 4255
},
{
"entropy": 5.3904636859893795,
"epoch": 0.5074449076831448,
"grad_norm": 1.0,
"learning_rate": 0.0004982878517776812,
"loss": 5.1488,
"mean_token_accuracy": 0.18893087059259414,
"num_tokens": 11901914.0,
"step": 4260
},
{
"entropy": 5.376463317871094,
"epoch": 0.5080405002977963,
"grad_norm": 1.1171875,
"learning_rate": 0.000498282600846363,
"loss": 5.1933,
"mean_token_accuracy": 0.184621000289917,
"num_tokens": 11913225.0,
"step": 4265
},
{
"entropy": 5.363157176971436,
"epoch": 0.5086360929124479,
"grad_norm": 1.03125,
"learning_rate": 0.000498277341906261,
"loss": 5.1574,
"mean_token_accuracy": 0.18310361355543137,
"num_tokens": 11927200.0,
"step": 4270
},
{
"entropy": 5.332843542098999,
"epoch": 0.5092316855270994,
"grad_norm": 1.0234375,
"learning_rate": 0.0004982720749575637,
"loss": 5.0928,
"mean_token_accuracy": 0.18536149263381957,
"num_tokens": 11939826.0,
"step": 4275
},
{
"entropy": 5.315101099014282,
"epoch": 0.509827278141751,
"grad_norm": 1.0,
"learning_rate": 0.0004982668000004602,
"loss": 5.1168,
"mean_token_accuracy": 0.18348885625600814,
"num_tokens": 11951983.0,
"step": 4280
},
{
"entropy": 5.416784811019897,
"epoch": 0.5104228707564026,
"grad_norm": 0.98828125,
"learning_rate": 0.0004982615170351395,
"loss": 5.2316,
"mean_token_accuracy": 0.177970851957798,
"num_tokens": 11965642.0,
"step": 4285
},
{
"entropy": 5.407585000991821,
"epoch": 0.5110184633710542,
"grad_norm": 0.98828125,
"learning_rate": 0.0004982562260617912,
"loss": 5.2606,
"mean_token_accuracy": 0.17398284673690795,
"num_tokens": 11981472.0,
"step": 4290
},
{
"entropy": 5.348097896575927,
"epoch": 0.5116140559857058,
"grad_norm": 0.92578125,
"learning_rate": 0.0004982509270806052,
"loss": 5.1193,
"mean_token_accuracy": 0.1831167832016945,
"num_tokens": 11995739.0,
"step": 4295
},
{
"entropy": 5.439723157882691,
"epoch": 0.5122096486003573,
"grad_norm": 1.09375,
"learning_rate": 0.0004982456200917712,
"loss": 5.2298,
"mean_token_accuracy": 0.17539945840835572,
"num_tokens": 12007850.0,
"step": 4300
},
{
"entropy": 5.429744052886963,
"epoch": 0.5128052412150089,
"grad_norm": 0.93359375,
"learning_rate": 0.00049824030509548,
"loss": 5.1746,
"mean_token_accuracy": 0.1897364154458046,
"num_tokens": 12022893.0,
"step": 4305
},
{
"entropy": 5.3463188171386715,
"epoch": 0.5134008338296605,
"grad_norm": 1.171875,
"learning_rate": 0.000498234982091922,
"loss": 5.1676,
"mean_token_accuracy": 0.19170445799827576,
"num_tokens": 12035520.0,
"step": 4310
},
{
"entropy": 5.3189493179321286,
"epoch": 0.5139964264443121,
"grad_norm": 1.0390625,
"learning_rate": 0.0004982296510812879,
"loss": 5.082,
"mean_token_accuracy": 0.17921495586633682,
"num_tokens": 12047938.0,
"step": 4315
},
{
"entropy": 5.3553399562835695,
"epoch": 0.5145920190589637,
"grad_norm": 1.0546875,
"learning_rate": 0.0004982243120637694,
"loss": 5.1278,
"mean_token_accuracy": 0.1810295581817627,
"num_tokens": 12061236.0,
"step": 4320
},
{
"entropy": 5.327606868743897,
"epoch": 0.5151876116736153,
"grad_norm": 1.1171875,
"learning_rate": 0.0004982189650395577,
"loss": 5.1243,
"mean_token_accuracy": 0.1883678764104843,
"num_tokens": 12074427.0,
"step": 4325
},
{
"entropy": 5.333645963668824,
"epoch": 0.5157832042882668,
"grad_norm": 0.90625,
"learning_rate": 0.0004982136100088447,
"loss": 5.1171,
"mean_token_accuracy": 0.18838856220245362,
"num_tokens": 12088128.0,
"step": 4330
},
{
"entropy": 5.436138057708741,
"epoch": 0.5163787969029184,
"grad_norm": 1.0234375,
"learning_rate": 0.0004982082469718224,
"loss": 5.2721,
"mean_token_accuracy": 0.1713164508342743,
"num_tokens": 12100938.0,
"step": 4335
},
{
"entropy": 5.374925661087036,
"epoch": 0.51697438951757,
"grad_norm": 1.0625,
"learning_rate": 0.0004982028759286831,
"loss": 5.1457,
"mean_token_accuracy": 0.18169915080070495,
"num_tokens": 12113464.0,
"step": 4340
},
{
"entropy": 5.333883953094483,
"epoch": 0.5175699821322216,
"grad_norm": 0.87890625,
"learning_rate": 0.0004981974968796197,
"loss": 5.2087,
"mean_token_accuracy": 0.17713046222925186,
"num_tokens": 12129113.0,
"step": 4345
},
{
"entropy": 5.402696943283081,
"epoch": 0.5181655747468732,
"grad_norm": 1.078125,
"learning_rate": 0.0004981921098248249,
"loss": 5.2514,
"mean_token_accuracy": 0.17428978979587556,
"num_tokens": 12142024.0,
"step": 4350
},
{
"entropy": 5.40776424407959,
"epoch": 0.5187611673615247,
"grad_norm": 0.8671875,
"learning_rate": 0.000498186714764492,
"loss": 5.1946,
"mean_token_accuracy": 0.17911534011363983,
"num_tokens": 12157889.0,
"step": 4355
},
{
"entropy": 5.397247409820556,
"epoch": 0.5193567599761763,
"grad_norm": 0.94140625,
"learning_rate": 0.0004981813116988145,
"loss": 5.2487,
"mean_token_accuracy": 0.18019398897886277,
"num_tokens": 12171527.0,
"step": 4360
},
{
"entropy": 5.352613306045532,
"epoch": 0.5199523525908278,
"grad_norm": 1.015625,
"learning_rate": 0.0004981759006279864,
"loss": 5.1952,
"mean_token_accuracy": 0.1757894992828369,
"num_tokens": 12186453.0,
"step": 4365
},
{
"entropy": 5.294702100753784,
"epoch": 0.5205479452054794,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981704815522014,
"loss": 5.0686,
"mean_token_accuracy": 0.1854909524321556,
"num_tokens": 12201401.0,
"step": 4370
},
{
"entropy": 5.316075944900513,
"epoch": 0.521143537820131,
"grad_norm": 0.90234375,
"learning_rate": 0.0004981650544716542,
"loss": 5.162,
"mean_token_accuracy": 0.17699071615934373,
"num_tokens": 12215219.0,
"step": 4375
},
{
"entropy": 5.40913577079773,
"epoch": 0.5217391304347826,
"grad_norm": 1.0,
"learning_rate": 0.0004981596193865393,
"loss": 5.2111,
"mean_token_accuracy": 0.17095923721790313,
"num_tokens": 12229932.0,
"step": 4380
},
{
"entropy": 5.341264867782593,
"epoch": 0.5223347230494342,
"grad_norm": 1.125,
"learning_rate": 0.0004981541762970517,
"loss": 5.0666,
"mean_token_accuracy": 0.1836489751935005,
"num_tokens": 12241718.0,
"step": 4385
},
{
"entropy": 5.404876804351806,
"epoch": 0.5229303156640858,
"grad_norm": 1.0546875,
"learning_rate": 0.0004981487252033866,
"loss": 5.2244,
"mean_token_accuracy": 0.18075872212648392,
"num_tokens": 12256395.0,
"step": 4390
},
{
"entropy": 5.323521566390991,
"epoch": 0.5235259082787374,
"grad_norm": 0.890625,
"learning_rate": 0.0004981432661057396,
"loss": 5.0745,
"mean_token_accuracy": 0.1884683221578598,
"num_tokens": 12269966.0,
"step": 4395
},
{
"entropy": 5.359594917297363,
"epoch": 0.5241215008933889,
"grad_norm": 1.0234375,
"learning_rate": 0.0004981377990043064,
"loss": 5.1932,
"mean_token_accuracy": 0.17640553265810013,
"num_tokens": 12284410.0,
"step": 4400
},
{
"entropy": 5.29043231010437,
"epoch": 0.5247170935080405,
"grad_norm": 1.0234375,
"learning_rate": 0.0004981323238992831,
"loss": 5.0208,
"mean_token_accuracy": 0.19154949784278869,
"num_tokens": 12298931.0,
"step": 4405
},
{
"entropy": 5.306810569763184,
"epoch": 0.5253126861226921,
"grad_norm": 0.953125,
"learning_rate": 0.0004981268407908662,
"loss": 5.1405,
"mean_token_accuracy": 0.184539595246315,
"num_tokens": 12313857.0,
"step": 4410
},
{
"entropy": 5.335316228866577,
"epoch": 0.5259082787373437,
"grad_norm": 0.95703125,
"learning_rate": 0.0004981213496792523,
"loss": 5.0948,
"mean_token_accuracy": 0.18559955060482025,
"num_tokens": 12327732.0,
"step": 4415
},
{
"entropy": 5.354093456268311,
"epoch": 0.5265038713519953,
"grad_norm": 1.0546875,
"learning_rate": 0.0004981158505646383,
"loss": 5.1658,
"mean_token_accuracy": 0.1781379371881485,
"num_tokens": 12341754.0,
"step": 4420
},
{
"entropy": 5.340103435516357,
"epoch": 0.5270994639666469,
"grad_norm": 1.0859375,
"learning_rate": 0.0004981103434472216,
"loss": 5.1129,
"mean_token_accuracy": 0.1820943310856819,
"num_tokens": 12355656.0,
"step": 4425
},
{
"entropy": 5.299054861068726,
"epoch": 0.5276950565812983,
"grad_norm": 1.0859375,
"learning_rate": 0.0004981048283271997,
"loss": 5.0752,
"mean_token_accuracy": 0.19043822437524796,
"num_tokens": 12368346.0,
"step": 4430
},
{
"entropy": 5.320931720733642,
"epoch": 0.5282906491959499,
"grad_norm": 1.046875,
"learning_rate": 0.0004980993052047701,
"loss": 5.1398,
"mean_token_accuracy": 0.18237759321928024,
"num_tokens": 12381844.0,
"step": 4435
},
{
"entropy": 5.412809801101685,
"epoch": 0.5288862418106015,
"grad_norm": 0.94921875,
"learning_rate": 0.0004980937740801313,
"loss": 5.2467,
"mean_token_accuracy": 0.1745242342352867,
"num_tokens": 12395432.0,
"step": 4440
},
{
"entropy": 5.348816967010498,
"epoch": 0.5294818344252531,
"grad_norm": 1.109375,
"learning_rate": 0.0004980882349534816,
"loss": 5.1523,
"mean_token_accuracy": 0.1783488169312477,
"num_tokens": 12408897.0,
"step": 4445
},
{
"entropy": 5.319600820541382,
"epoch": 0.5300774270399047,
"grad_norm": 1.0,
"learning_rate": 0.0004980826878250195,
"loss": 5.1091,
"mean_token_accuracy": 0.18761228770017624,
"num_tokens": 12421480.0,
"step": 4450
},
{
"entropy": 5.38168249130249,
"epoch": 0.5306730196545563,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980771326949442,
"loss": 5.1908,
"mean_token_accuracy": 0.1767524540424347,
"num_tokens": 12436931.0,
"step": 4455
},
{
"entropy": 5.357715749740601,
"epoch": 0.5312686122692079,
"grad_norm": 1.03125,
"learning_rate": 0.0004980715695634549,
"loss": 5.2283,
"mean_token_accuracy": 0.1780093342065811,
"num_tokens": 12451687.0,
"step": 4460
},
{
"entropy": 5.346804809570313,
"epoch": 0.5318642048838594,
"grad_norm": 1.0859375,
"learning_rate": 0.0004980659984307511,
"loss": 5.1157,
"mean_token_accuracy": 0.18669354319572448,
"num_tokens": 12464338.0,
"step": 4465
},
{
"entropy": 5.342980432510376,
"epoch": 0.532459797498511,
"grad_norm": 0.859375,
"learning_rate": 0.0004980604192970325,
"loss": 5.1816,
"mean_token_accuracy": 0.17993794977664948,
"num_tokens": 12482520.0,
"step": 4470
},
{
"entropy": 5.473724365234375,
"epoch": 0.5330553901131626,
"grad_norm": 1.0,
"learning_rate": 0.0004980548321624993,
"loss": 5.2242,
"mean_token_accuracy": 0.1799166440963745,
"num_tokens": 12495791.0,
"step": 4475
},
{
"entropy": 5.355010461807251,
"epoch": 0.5336509827278142,
"grad_norm": 1.0390625,
"learning_rate": 0.000498049237027352,
"loss": 5.1539,
"mean_token_accuracy": 0.17974822223186493,
"num_tokens": 12509963.0,
"step": 4480
},
{
"entropy": 5.381701517105102,
"epoch": 0.5342465753424658,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980436338917914,
"loss": 5.1963,
"mean_token_accuracy": 0.18136336356401445,
"num_tokens": 12522825.0,
"step": 4485
},
{
"entropy": 5.381725168228149,
"epoch": 0.5348421679571174,
"grad_norm": 0.90234375,
"learning_rate": 0.000498038022756018,
"loss": 5.1345,
"mean_token_accuracy": 0.1828573077917099,
"num_tokens": 12536097.0,
"step": 4490
},
{
"entropy": 5.2805726528167725,
"epoch": 0.535437760571769,
"grad_norm": 1.0,
"learning_rate": 0.0004980324036202335,
"loss": 5.1276,
"mean_token_accuracy": 0.1897248238325119,
"num_tokens": 12551441.0,
"step": 4495
},
{
"entropy": 5.322610759735108,
"epoch": 0.5360333531864205,
"grad_norm": 0.92578125,
"learning_rate": 0.0004980267764846393,
"loss": 5.1061,
"mean_token_accuracy": 0.18795953691005707,
"num_tokens": 12566271.0,
"step": 4500
},
{
"entropy": 5.420116519927978,
"epoch": 0.5366289458010721,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980211413494373,
"loss": 5.1883,
"mean_token_accuracy": 0.1835271894931793,
"num_tokens": 12580585.0,
"step": 4505
},
{
"entropy": 5.267950582504272,
"epoch": 0.5372245384157236,
"grad_norm": 0.98046875,
"learning_rate": 0.0004980154982148296,
"loss": 5.0166,
"mean_token_accuracy": 0.1915774330496788,
"num_tokens": 12592389.0,
"step": 4510
},
{
"entropy": 5.192738819122314,
"epoch": 0.5378201310303752,
"grad_norm": 0.9375,
"learning_rate": 0.0004980098470810185,
"loss": 5.0143,
"mean_token_accuracy": 0.19835987836122512,
"num_tokens": 12606487.0,
"step": 4515
},
{
"entropy": 5.2980540752410885,
"epoch": 0.5384157236450268,
"grad_norm": 0.9765625,
"learning_rate": 0.0004980041879482068,
"loss": 5.1387,
"mean_token_accuracy": 0.19084453135728835,
"num_tokens": 12621036.0,
"step": 4520
},
{
"entropy": 5.354099130630493,
"epoch": 0.5390113162596784,
"grad_norm": 0.98828125,
"learning_rate": 0.0004979985208165975,
"loss": 5.1031,
"mean_token_accuracy": 0.1919423371553421,
"num_tokens": 12634911.0,
"step": 4525
},
{
"entropy": 5.275477552413941,
"epoch": 0.5396069088743299,
"grad_norm": 1.046875,
"learning_rate": 0.0004979928456863937,
"loss": 5.143,
"mean_token_accuracy": 0.19001434594392777,
"num_tokens": 12649165.0,
"step": 4530
},
{
"entropy": 5.313499069213867,
"epoch": 0.5402025014889815,
"grad_norm": 0.90625,
"learning_rate": 0.0004979871625577991,
"loss": 5.1087,
"mean_token_accuracy": 0.18552252352237703,
"num_tokens": 12663803.0,
"step": 4535
},
{
"entropy": 5.3516298770904545,
"epoch": 0.5407980941036331,
"grad_norm": 1.0078125,
"learning_rate": 0.0004979814714310177,
"loss": 5.0679,
"mean_token_accuracy": 0.1860592007637024,
"num_tokens": 12676336.0,
"step": 4540
},
{
"entropy": 5.264741277694702,
"epoch": 0.5413936867182847,
"grad_norm": 0.97265625,
"learning_rate": 0.0004979757723062533,
"loss": 5.0473,
"mean_token_accuracy": 0.18942687511444092,
"num_tokens": 12689619.0,
"step": 4545
},
{
"entropy": 5.281706666946411,
"epoch": 0.5419892793329363,
"grad_norm": 1.0390625,
"learning_rate": 0.0004979700651837107,
"loss": 5.1259,
"mean_token_accuracy": 0.18445766866207122,
"num_tokens": 12703964.0,
"step": 4550
},
{
"entropy": 5.4037003993988035,
"epoch": 0.5425848719475879,
"grad_norm": 1.140625,
"learning_rate": 0.0004979643500635942,
"loss": 5.1492,
"mean_token_accuracy": 0.18617296367883682,
"num_tokens": 12716832.0,
"step": 4555
},
{
"entropy": 5.303174543380737,
"epoch": 0.5431804645622395,
"grad_norm": 1.03125,
"learning_rate": 0.0004979586269461089,
"loss": 5.0377,
"mean_token_accuracy": 0.1901222810149193,
"num_tokens": 12729665.0,
"step": 4560
},
{
"entropy": 5.287974309921265,
"epoch": 0.543776057176891,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979528958314604,
"loss": 5.0723,
"mean_token_accuracy": 0.1851668655872345,
"num_tokens": 12743576.0,
"step": 4565
},
{
"entropy": 5.258513402938843,
"epoch": 0.5443716497915426,
"grad_norm": 0.90625,
"learning_rate": 0.0004979471567198539,
"loss": 5.1029,
"mean_token_accuracy": 0.18396760225296022,
"num_tokens": 12758511.0,
"step": 4570
},
{
"entropy": 5.299459743499756,
"epoch": 0.5449672424061942,
"grad_norm": 1.03125,
"learning_rate": 0.0004979414096114953,
"loss": 5.0771,
"mean_token_accuracy": 0.18704642951488495,
"num_tokens": 12771694.0,
"step": 4575
},
{
"entropy": 5.312016153335572,
"epoch": 0.5455628350208458,
"grad_norm": 0.9921875,
"learning_rate": 0.000497935654506591,
"loss": 5.1235,
"mean_token_accuracy": 0.1825407028198242,
"num_tokens": 12786367.0,
"step": 4580
},
{
"entropy": 5.245462083816529,
"epoch": 0.5461584276354973,
"grad_norm": 1.0,
"learning_rate": 0.000497929891405347,
"loss": 4.9933,
"mean_token_accuracy": 0.19003615230321885,
"num_tokens": 12801177.0,
"step": 4585
},
{
"entropy": 5.3865326881408695,
"epoch": 0.5467540202501489,
"grad_norm": 0.96484375,
"learning_rate": 0.0004979241203079705,
"loss": 5.1814,
"mean_token_accuracy": 0.1721171796321869,
"num_tokens": 12817347.0,
"step": 4590
},
{
"entropy": 5.3835690975189205,
"epoch": 0.5473496128648004,
"grad_norm": 1.0234375,
"learning_rate": 0.0004979183412146681,
"loss": 5.202,
"mean_token_accuracy": 0.1815650701522827,
"num_tokens": 12830304.0,
"step": 4595
},
{
"entropy": 5.2819915294647215,
"epoch": 0.547945205479452,
"grad_norm": 0.98828125,
"learning_rate": 0.0004979125541256474,
"loss": 5.0242,
"mean_token_accuracy": 0.19441343247890472,
"num_tokens": 12843813.0,
"step": 4600
},
{
"entropy": 5.354576063156128,
"epoch": 0.5485407980941036,
"grad_norm": 0.9921875,
"learning_rate": 0.0004979067590411157,
"loss": 5.2058,
"mean_token_accuracy": 0.1779221698641777,
"num_tokens": 12857989.0,
"step": 4605
},
{
"entropy": 5.365860271453857,
"epoch": 0.5491363907087552,
"grad_norm": 1.046875,
"learning_rate": 0.000497900955961281,
"loss": 5.1199,
"mean_token_accuracy": 0.18672771453857423,
"num_tokens": 12871529.0,
"step": 4610
},
{
"entropy": 5.318894243240356,
"epoch": 0.5497319833234068,
"grad_norm": 1.0625,
"learning_rate": 0.0004978951448863514,
"loss": 5.1329,
"mean_token_accuracy": 0.18040903359651567,
"num_tokens": 12884907.0,
"step": 4615
},
{
"entropy": 5.2730248928070065,
"epoch": 0.5503275759380584,
"grad_norm": 1.0703125,
"learning_rate": 0.0004978893258165354,
"loss": 5.0952,
"mean_token_accuracy": 0.1864814504981041,
"num_tokens": 12898403.0,
"step": 4620
},
{
"entropy": 5.347129392623901,
"epoch": 0.55092316855271,
"grad_norm": 0.99609375,
"learning_rate": 0.0004978834987520417,
"loss": 5.201,
"mean_token_accuracy": 0.18264321535825728,
"num_tokens": 12913982.0,
"step": 4625
},
{
"entropy": 5.344757986068726,
"epoch": 0.5515187611673615,
"grad_norm": 1.015625,
"learning_rate": 0.0004978776636930792,
"loss": 5.0969,
"mean_token_accuracy": 0.18262473940849305,
"num_tokens": 12928826.0,
"step": 4630
},
{
"entropy": 5.261846733093262,
"epoch": 0.5521143537820131,
"grad_norm": 1.0234375,
"learning_rate": 0.0004978718206398574,
"loss": 5.1196,
"mean_token_accuracy": 0.1848312333226204,
"num_tokens": 12941905.0,
"step": 4635
},
{
"entropy": 5.3465046882629395,
"epoch": 0.5527099463966647,
"grad_norm": 0.95703125,
"learning_rate": 0.0004978659695925856,
"loss": 5.2387,
"mean_token_accuracy": 0.1858798533678055,
"num_tokens": 12955522.0,
"step": 4640
},
{
"entropy": 5.371321058273315,
"epoch": 0.5533055390113163,
"grad_norm": 1.0625,
"learning_rate": 0.0004978601105514739,
"loss": 5.1295,
"mean_token_accuracy": 0.18562092036008834,
"num_tokens": 12968491.0,
"step": 4645
},
{
"entropy": 5.297031497955322,
"epoch": 0.5539011316259679,
"grad_norm": 1.0390625,
"learning_rate": 0.0004978542435167324,
"loss": 5.1615,
"mean_token_accuracy": 0.18115428388118743,
"num_tokens": 12983258.0,
"step": 4650
},
{
"entropy": 5.318244791030883,
"epoch": 0.5544967242406195,
"grad_norm": 1.15625,
"learning_rate": 0.0004978483684885716,
"loss": 5.1539,
"mean_token_accuracy": 0.18428515195846557,
"num_tokens": 12996271.0,
"step": 4655
},
{
"entropy": 5.364977312088013,
"epoch": 0.5550923168552709,
"grad_norm": 0.96875,
"learning_rate": 0.0004978424854672021,
"loss": 5.1732,
"mean_token_accuracy": 0.18524766713380814,
"num_tokens": 13010920.0,
"step": 4660
},
{
"entropy": 5.25998821258545,
"epoch": 0.5556879094699225,
"grad_norm": 1.0703125,
"learning_rate": 0.0004978365944528349,
"loss": 4.999,
"mean_token_accuracy": 0.19228564500808715,
"num_tokens": 13023499.0,
"step": 4665
},
{
"entropy": 5.3263512134552,
"epoch": 0.5562835020845741,
"grad_norm": 0.88671875,
"learning_rate": 0.0004978306954456814,
"loss": 5.197,
"mean_token_accuracy": 0.1809248998761177,
"num_tokens": 13040510.0,
"step": 4670
},
{
"entropy": 5.322596454620362,
"epoch": 0.5568790946992257,
"grad_norm": 1.0234375,
"learning_rate": 0.0004978247884459532,
"loss": 5.053,
"mean_token_accuracy": 0.191793230175972,
"num_tokens": 13053708.0,
"step": 4675
},
{
"entropy": 5.338529348373413,
"epoch": 0.5574746873138773,
"grad_norm": 0.9609375,
"learning_rate": 0.0004978188734538621,
"loss": 5.1755,
"mean_token_accuracy": 0.185623100399971,
"num_tokens": 13068666.0,
"step": 4680
},
{
"entropy": 5.309015130996704,
"epoch": 0.5580702799285289,
"grad_norm": 1.1640625,
"learning_rate": 0.0004978129504696203,
"loss": 5.0854,
"mean_token_accuracy": 0.17966341376304626,
"num_tokens": 13082549.0,
"step": 4685
},
{
"entropy": 5.444301462173462,
"epoch": 0.5586658725431805,
"grad_norm": 1.0546875,
"learning_rate": 0.0004978070194934402,
"loss": 5.2386,
"mean_token_accuracy": 0.17759112864732743,
"num_tokens": 13096685.0,
"step": 4690
},
{
"entropy": 5.332045364379883,
"epoch": 0.559261465157832,
"grad_norm": 0.96484375,
"learning_rate": 0.0004978010805255347,
"loss": 5.0476,
"mean_token_accuracy": 0.18240438550710678,
"num_tokens": 13110187.0,
"step": 4695
},
{
"entropy": 5.253191518783569,
"epoch": 0.5598570577724836,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977951335661165,
"loss": 5.1163,
"mean_token_accuracy": 0.18413471579551696,
"num_tokens": 13124105.0,
"step": 4700
},
{
"entropy": 5.359992599487304,
"epoch": 0.5604526503871352,
"grad_norm": 0.9453125,
"learning_rate": 0.0004977891786153992,
"loss": 5.1147,
"mean_token_accuracy": 0.1873343288898468,
"num_tokens": 13139675.0,
"step": 4705
},
{
"entropy": 5.3295289993286135,
"epoch": 0.5610482430017868,
"grad_norm": 1.171875,
"learning_rate": 0.0004977832156735963,
"loss": 5.0775,
"mean_token_accuracy": 0.18729310780763625,
"num_tokens": 13154105.0,
"step": 4710
},
{
"entropy": 5.251990795135498,
"epoch": 0.5616438356164384,
"grad_norm": 1.0390625,
"learning_rate": 0.0004977772447409217,
"loss": 4.9941,
"mean_token_accuracy": 0.19008799344301225,
"num_tokens": 13168473.0,
"step": 4715
},
{
"entropy": 5.342791795730591,
"epoch": 0.56223942823109,
"grad_norm": 0.97265625,
"learning_rate": 0.0004977712658175894,
"loss": 5.111,
"mean_token_accuracy": 0.18688295781612396,
"num_tokens": 13182873.0,
"step": 4720
},
{
"entropy": 5.288819646835327,
"epoch": 0.5628350208457416,
"grad_norm": 1.0625,
"learning_rate": 0.0004977652789038142,
"loss": 5.1176,
"mean_token_accuracy": 0.18229663521051406,
"num_tokens": 13196894.0,
"step": 4725
},
{
"entropy": 5.283871412277222,
"epoch": 0.5634306134603931,
"grad_norm": 0.98828125,
"learning_rate": 0.0004977592839998105,
"loss": 5.1187,
"mean_token_accuracy": 0.18293171674013137,
"num_tokens": 13212285.0,
"step": 4730
},
{
"entropy": 5.386455345153808,
"epoch": 0.5640262060750447,
"grad_norm": 0.98046875,
"learning_rate": 0.0004977532811057936,
"loss": 5.128,
"mean_token_accuracy": 0.18217893838882446,
"num_tokens": 13227272.0,
"step": 4735
},
{
"entropy": 5.334987211227417,
"epoch": 0.5646217986896962,
"grad_norm": 0.98828125,
"learning_rate": 0.0004977472702219786,
"loss": 5.1204,
"mean_token_accuracy": 0.18561779707670212,
"num_tokens": 13241544.0,
"step": 4740
},
{
"entropy": 5.294053316116333,
"epoch": 0.5652173913043478,
"grad_norm": 0.93359375,
"learning_rate": 0.0004977412513485812,
"loss": 5.123,
"mean_token_accuracy": 0.1810023307800293,
"num_tokens": 13255605.0,
"step": 4745
},
{
"entropy": 5.342393112182617,
"epoch": 0.5658129839189994,
"grad_norm": 1.0078125,
"learning_rate": 0.0004977352244858172,
"loss": 5.1564,
"mean_token_accuracy": 0.1864907681941986,
"num_tokens": 13269143.0,
"step": 4750
},
{
"entropy": 5.300639295578003,
"epoch": 0.566408576533651,
"grad_norm": 0.9140625,
"learning_rate": 0.0004977291896339029,
"loss": 5.0493,
"mean_token_accuracy": 0.19212037175893784,
"num_tokens": 13284316.0,
"step": 4755
},
{
"entropy": 5.246579551696778,
"epoch": 0.5670041691483025,
"grad_norm": 1.0703125,
"learning_rate": 0.0004977231467930546,
"loss": 5.0287,
"mean_token_accuracy": 0.1949671611189842,
"num_tokens": 13297714.0,
"step": 4760
},
{
"entropy": 5.283139991760254,
"epoch": 0.5675997617629541,
"grad_norm": 0.97265625,
"learning_rate": 0.0004977170959634893,
"loss": 5.1099,
"mean_token_accuracy": 0.18398559838533401,
"num_tokens": 13311759.0,
"step": 4765
},
{
"entropy": 5.2232029914855955,
"epoch": 0.5681953543776057,
"grad_norm": 1.0703125,
"learning_rate": 0.0004977110371454239,
"loss": 5.0244,
"mean_token_accuracy": 0.19327093958854674,
"num_tokens": 13325966.0,
"step": 4770
},
{
"entropy": 5.332334375381469,
"epoch": 0.5687909469922573,
"grad_norm": 0.96875,
"learning_rate": 0.0004977049703390755,
"loss": 5.1771,
"mean_token_accuracy": 0.18459965735673906,
"num_tokens": 13340939.0,
"step": 4775
},
{
"entropy": 5.345418310165405,
"epoch": 0.5693865396069089,
"grad_norm": 0.96875,
"learning_rate": 0.0004976988955446622,
"loss": 5.2267,
"mean_token_accuracy": 0.18587302714586257,
"num_tokens": 13354400.0,
"step": 4780
},
{
"entropy": 5.482950258255005,
"epoch": 0.5699821322215605,
"grad_norm": 0.96484375,
"learning_rate": 0.0004976928127624013,
"loss": 5.2282,
"mean_token_accuracy": 0.1716497138142586,
"num_tokens": 13368819.0,
"step": 4785
},
{
"entropy": 5.2797523021698,
"epoch": 0.570577724836212,
"grad_norm": 1.015625,
"learning_rate": 0.0004976867219925116,
"loss": 5.0867,
"mean_token_accuracy": 0.19408287107944489,
"num_tokens": 13381106.0,
"step": 4790
},
{
"entropy": 5.313072156906128,
"epoch": 0.5711733174508636,
"grad_norm": 0.98828125,
"learning_rate": 0.000497680623235211,
"loss": 5.0814,
"mean_token_accuracy": 0.1848406746983528,
"num_tokens": 13394718.0,
"step": 4795
},
{
"entropy": 5.354447746276856,
"epoch": 0.5717689100655152,
"grad_norm": 1.0078125,
"learning_rate": 0.0004976745164907185,
"loss": 5.1756,
"mean_token_accuracy": 0.18064648061990737,
"num_tokens": 13408416.0,
"step": 4800
},
{
"entropy": 5.339391088485717,
"epoch": 0.5723645026801668,
"grad_norm": 1.03125,
"learning_rate": 0.0004976684017592533,
"loss": 5.1746,
"mean_token_accuracy": 0.18296917527914047,
"num_tokens": 13422287.0,
"step": 4805
},
{
"entropy": 5.353500032424927,
"epoch": 0.5729600952948184,
"grad_norm": 0.94140625,
"learning_rate": 0.0004976622790410345,
"loss": 5.1445,
"mean_token_accuracy": 0.18404195755720137,
"num_tokens": 13438080.0,
"step": 4810
},
{
"entropy": 5.326828289031982,
"epoch": 0.5735556879094699,
"grad_norm": 0.87890625,
"learning_rate": 0.0004976561483362818,
"loss": 5.0913,
"mean_token_accuracy": 0.18155461102724074,
"num_tokens": 13454781.0,
"step": 4815
},
{
"entropy": 5.295237302780151,
"epoch": 0.5741512805241215,
"grad_norm": 1.046875,
"learning_rate": 0.000497650009645215,
"loss": 5.1042,
"mean_token_accuracy": 0.179357174038887,
"num_tokens": 13467021.0,
"step": 4820
},
{
"entropy": 5.252260828018189,
"epoch": 0.574746873138773,
"grad_norm": 1.125,
"learning_rate": 0.0004976438629680543,
"loss": 5.0881,
"mean_token_accuracy": 0.18549765646457672,
"num_tokens": 13479125.0,
"step": 4825
},
{
"entropy": 5.308937072753906,
"epoch": 0.5753424657534246,
"grad_norm": 0.984375,
"learning_rate": 0.0004976377083050204,
"loss": 5.0668,
"mean_token_accuracy": 0.19174055457115174,
"num_tokens": 13493444.0,
"step": 4830
},
{
"entropy": 5.319361066818237,
"epoch": 0.5759380583680762,
"grad_norm": 0.9609375,
"learning_rate": 0.0004976315456563339,
"loss": 5.158,
"mean_token_accuracy": 0.18553740084171294,
"num_tokens": 13508261.0,
"step": 4835
},
{
"entropy": 5.362977313995361,
"epoch": 0.5765336509827278,
"grad_norm": 1.0859375,
"learning_rate": 0.0004976253750222157,
"loss": 5.1668,
"mean_token_accuracy": 0.18401493281126022,
"num_tokens": 13521371.0,
"step": 4840
},
{
"entropy": 5.315124940872193,
"epoch": 0.5771292435973794,
"grad_norm": 0.96484375,
"learning_rate": 0.0004976191964028874,
"loss": 5.1608,
"mean_token_accuracy": 0.18406408429145812,
"num_tokens": 13536944.0,
"step": 4845
},
{
"entropy": 5.33033242225647,
"epoch": 0.577724836212031,
"grad_norm": 1.0859375,
"learning_rate": 0.0004976130097985703,
"loss": 5.1431,
"mean_token_accuracy": 0.1895923212170601,
"num_tokens": 13552281.0,
"step": 4850
},
{
"entropy": 5.319507026672364,
"epoch": 0.5783204288266826,
"grad_norm": 0.93359375,
"learning_rate": 0.0004976068152094867,
"loss": 5.1237,
"mean_token_accuracy": 0.182035830616951,
"num_tokens": 13567376.0,
"step": 4855
},
{
"entropy": 5.420705127716064,
"epoch": 0.5789160214413341,
"grad_norm": 1.0078125,
"learning_rate": 0.0004976006126358584,
"loss": 5.2231,
"mean_token_accuracy": 0.17559084743261338,
"num_tokens": 13580688.0,
"step": 4860
},
{
"entropy": 5.331259059906006,
"epoch": 0.5795116140559857,
"grad_norm": 1.015625,
"learning_rate": 0.0004975944020779082,
"loss": 5.0996,
"mean_token_accuracy": 0.1855834275484085,
"num_tokens": 13594823.0,
"step": 4865
},
{
"entropy": 5.27305760383606,
"epoch": 0.5801072066706373,
"grad_norm": 1.03125,
"learning_rate": 0.0004975881835358586,
"loss": 5.1076,
"mean_token_accuracy": 0.18907674700021743,
"num_tokens": 13606702.0,
"step": 4870
},
{
"entropy": 5.334150218963623,
"epoch": 0.5807027992852889,
"grad_norm": 0.9296875,
"learning_rate": 0.0004975819570099328,
"loss": 5.0711,
"mean_token_accuracy": 0.18504173457622528,
"num_tokens": 13620651.0,
"step": 4875
},
{
"entropy": 5.254895496368408,
"epoch": 0.5812983918999405,
"grad_norm": 1.125,
"learning_rate": 0.000497575722500354,
"loss": 5.0417,
"mean_token_accuracy": 0.19002680629491805,
"num_tokens": 13633192.0,
"step": 4880
},
{
"entropy": 5.223383855819702,
"epoch": 0.5818939845145921,
"grad_norm": 1.0078125,
"learning_rate": 0.000497569480007346,
"loss": 5.086,
"mean_token_accuracy": 0.18796343058347703,
"num_tokens": 13647082.0,
"step": 4885
},
{
"entropy": 5.309591293334961,
"epoch": 0.5824895771292437,
"grad_norm": 0.89453125,
"learning_rate": 0.0004975632295311327,
"loss": 5.1161,
"mean_token_accuracy": 0.18823856562376023,
"num_tokens": 13662753.0,
"step": 4890
},
{
"entropy": 5.248241519927978,
"epoch": 0.5830851697438951,
"grad_norm": 0.9921875,
"learning_rate": 0.0004975569710719382,
"loss": 5.0569,
"mean_token_accuracy": 0.18975120633840561,
"num_tokens": 13678548.0,
"step": 4895
},
{
"entropy": 5.326264381408691,
"epoch": 0.5836807623585467,
"grad_norm": 1.140625,
"learning_rate": 0.0004975507046299869,
"loss": 5.1488,
"mean_token_accuracy": 0.1870553523302078,
"num_tokens": 13690969.0,
"step": 4900
},
{
"entropy": 5.325755167007446,
"epoch": 0.5842763549731983,
"grad_norm": 0.94140625,
"learning_rate": 0.0004975444302055038,
"loss": 5.0501,
"mean_token_accuracy": 0.18742039203643798,
"num_tokens": 13704519.0,
"step": 4905
},
{
"entropy": 5.270959997177124,
"epoch": 0.5848719475878499,
"grad_norm": 1.140625,
"learning_rate": 0.0004975381477987136,
"loss": 5.1074,
"mean_token_accuracy": 0.18990254104137422,
"num_tokens": 13719664.0,
"step": 4910
},
{
"entropy": 5.2809583187103275,
"epoch": 0.5854675402025015,
"grad_norm": 0.9453125,
"learning_rate": 0.0004975318574098422,
"loss": 5.0033,
"mean_token_accuracy": 0.18962556570768357,
"num_tokens": 13733527.0,
"step": 4915
},
{
"entropy": 5.363389396667481,
"epoch": 0.586063132817153,
"grad_norm": 1.015625,
"learning_rate": 0.0004975255590391147,
"loss": 5.1479,
"mean_token_accuracy": 0.18534155189990997,
"num_tokens": 13747788.0,
"step": 4920
},
{
"entropy": 5.292184972763062,
"epoch": 0.5866587254318046,
"grad_norm": 1.1640625,
"learning_rate": 0.0004975192526867572,
"loss": 5.1181,
"mean_token_accuracy": 0.1887222185730934,
"num_tokens": 13761389.0,
"step": 4925
},
{
"entropy": 5.2499746799469,
"epoch": 0.5872543180464562,
"grad_norm": 0.9609375,
"learning_rate": 0.0004975129383529959,
"loss": 5.0048,
"mean_token_accuracy": 0.18806459307670592,
"num_tokens": 13774153.0,
"step": 4930
},
{
"entropy": 5.300623989105224,
"epoch": 0.5878499106611078,
"grad_norm": 0.93359375,
"learning_rate": 0.0004975066160380573,
"loss": 5.1648,
"mean_token_accuracy": 0.18778980821371077,
"num_tokens": 13788167.0,
"step": 4935
},
{
"entropy": 5.328409147262573,
"epoch": 0.5884455032757594,
"grad_norm": 0.9140625,
"learning_rate": 0.0004975002857421681,
"loss": 5.06,
"mean_token_accuracy": 0.1956685245037079,
"num_tokens": 13801190.0,
"step": 4940
},
{
"entropy": 5.269784164428711,
"epoch": 0.589041095890411,
"grad_norm": 1.03125,
"learning_rate": 0.0004974939474655554,
"loss": 5.1147,
"mean_token_accuracy": 0.18476307094097139,
"num_tokens": 13815744.0,
"step": 4945
},
{
"entropy": 5.317069387435913,
"epoch": 0.5896366885050626,
"grad_norm": 1.046875,
"learning_rate": 0.0004974876012084468,
"loss": 5.1331,
"mean_token_accuracy": 0.19391795843839646,
"num_tokens": 13829329.0,
"step": 4950
},
{
"entropy": 5.366438722610473,
"epoch": 0.5902322811197142,
"grad_norm": 0.96484375,
"learning_rate": 0.0004974812469710694,
"loss": 5.121,
"mean_token_accuracy": 0.18827635049819946,
"num_tokens": 13842401.0,
"step": 4955
},
{
"entropy": 5.246575784683228,
"epoch": 0.5908278737343657,
"grad_norm": 1.0,
"learning_rate": 0.0004974748847536516,
"loss": 5.0374,
"mean_token_accuracy": 0.18482894450426102,
"num_tokens": 13857156.0,
"step": 4960
},
{
"entropy": 5.2397401332855225,
"epoch": 0.5914234663490173,
"grad_norm": 0.98828125,
"learning_rate": 0.0004974685145564213,
"loss": 5.0298,
"mean_token_accuracy": 0.18295381367206573,
"num_tokens": 13870863.0,
"step": 4965
},
{
"entropy": 5.38400182723999,
"epoch": 0.5920190589636688,
"grad_norm": 0.96875,
"learning_rate": 0.0004974621363796071,
"loss": 5.2233,
"mean_token_accuracy": 0.1797648251056671,
"num_tokens": 13886121.0,
"step": 4970
},
{
"entropy": 5.38569016456604,
"epoch": 0.5926146515783204,
"grad_norm": 1.0703125,
"learning_rate": 0.0004974557502234378,
"loss": 5.1529,
"mean_token_accuracy": 0.18728985637426376,
"num_tokens": 13898928.0,
"step": 4975
},
{
"entropy": 5.324422311782837,
"epoch": 0.593210244192972,
"grad_norm": 1.1796875,
"learning_rate": 0.0004974493560881423,
"loss": 5.1651,
"mean_token_accuracy": 0.18712487518787385,
"num_tokens": 13911648.0,
"step": 4980
},
{
"entropy": 5.31566104888916,
"epoch": 0.5938058368076236,
"grad_norm": 0.98828125,
"learning_rate": 0.0004974429539739502,
"loss": 5.1659,
"mean_token_accuracy": 0.18176639825105667,
"num_tokens": 13924705.0,
"step": 4985
},
{
"entropy": 5.28463306427002,
"epoch": 0.5944014294222751,
"grad_norm": 1.0078125,
"learning_rate": 0.0004974365438810911,
"loss": 5.0808,
"mean_token_accuracy": 0.18167000263929367,
"num_tokens": 13939495.0,
"step": 4990
},
{
"entropy": 5.278111171722412,
"epoch": 0.5949970220369267,
"grad_norm": 1.0546875,
"learning_rate": 0.0004974301258097948,
"loss": 5.0595,
"mean_token_accuracy": 0.1901421144604683,
"num_tokens": 13952236.0,
"step": 4995
},
{
"entropy": 5.373788261413575,
"epoch": 0.5955926146515783,
"grad_norm": 1.0078125,
"learning_rate": 0.0004974236997602915,
"loss": 5.1586,
"mean_token_accuracy": 0.18144380897283555,
"num_tokens": 13967699.0,
"step": 5000
},
{
"entropy": 5.328965330123902,
"epoch": 0.5961882072662299,
"grad_norm": 1.015625,
"learning_rate": 0.0004974172657328116,
"loss": 5.1165,
"mean_token_accuracy": 0.18640072345733644,
"num_tokens": 13980619.0,
"step": 5005
},
{
"entropy": 5.265719652175903,
"epoch": 0.5967837998808815,
"grad_norm": 1.0390625,
"learning_rate": 0.000497410823727586,
"loss": 5.0707,
"mean_token_accuracy": 0.18287436813116073,
"num_tokens": 13995965.0,
"step": 5010
},
{
"entropy": 5.272796201705932,
"epoch": 0.5973793924955331,
"grad_norm": 0.90625,
"learning_rate": 0.000497404373744846,
"loss": 5.105,
"mean_token_accuracy": 0.18683422952890397,
"num_tokens": 14010325.0,
"step": 5015
},
{
"entropy": 5.371990966796875,
"epoch": 0.5979749851101847,
"grad_norm": 1.03125,
"learning_rate": 0.0004973979157848227,
"loss": 5.0904,
"mean_token_accuracy": 0.18376192897558213,
"num_tokens": 14025368.0,
"step": 5020
},
{
"entropy": 5.2132796287536625,
"epoch": 0.5985705777248362,
"grad_norm": 0.8984375,
"learning_rate": 0.0004973914498477475,
"loss": 5.062,
"mean_token_accuracy": 0.18886683881282806,
"num_tokens": 14039609.0,
"step": 5025
},
{
"entropy": 5.362352561950684,
"epoch": 0.5991661703394878,
"grad_norm": 0.984375,
"learning_rate": 0.0004973849759338527,
"loss": 5.183,
"mean_token_accuracy": 0.18064136058092117,
"num_tokens": 14053680.0,
"step": 5030
},
{
"entropy": 5.438577604293823,
"epoch": 0.5997617629541394,
"grad_norm": 0.9609375,
"learning_rate": 0.0004973784940433706,
"loss": 5.1898,
"mean_token_accuracy": 0.17920819520950318,
"num_tokens": 14067262.0,
"step": 5035
},
{
"entropy": 5.2425000190734865,
"epoch": 0.600357355568791,
"grad_norm": 1.0078125,
"learning_rate": 0.0004973720041765332,
"loss": 5.0522,
"mean_token_accuracy": 0.18719157874584197,
"num_tokens": 14080821.0,
"step": 5040
},
{
"entropy": 5.305828285217285,
"epoch": 0.6009529481834425,
"grad_norm": 1.109375,
"learning_rate": 0.0004973655063335736,
"loss": 5.1566,
"mean_token_accuracy": 0.18426071256399154,
"num_tokens": 14093586.0,
"step": 5045
},
{
"entropy": 5.2797142505645756,
"epoch": 0.6015485407980941,
"grad_norm": 0.99609375,
"learning_rate": 0.000497359000514725,
"loss": 4.9846,
"mean_token_accuracy": 0.19537042677402497,
"num_tokens": 14108138.0,
"step": 5050
},
{
"entropy": 5.271422624588013,
"epoch": 0.6021441334127456,
"grad_norm": 0.94140625,
"learning_rate": 0.0004973524867202205,
"loss": 5.0523,
"mean_token_accuracy": 0.19041558653116225,
"num_tokens": 14122495.0,
"step": 5055
},
{
"entropy": 5.212626504898071,
"epoch": 0.6027397260273972,
"grad_norm": 0.953125,
"learning_rate": 0.0004973459649502939,
"loss": 5.0562,
"mean_token_accuracy": 0.19054892063140869,
"num_tokens": 14136838.0,
"step": 5060
},
{
"entropy": 5.2571022510528564,
"epoch": 0.6033353186420488,
"grad_norm": 1.0,
"learning_rate": 0.0004973394352051789,
"loss": 5.0525,
"mean_token_accuracy": 0.19038456231355666,
"num_tokens": 14150910.0,
"step": 5065
},
{
"entropy": 5.347372961044312,
"epoch": 0.6039309112567004,
"grad_norm": 1.3046875,
"learning_rate": 0.00049733289748511,
"loss": 5.1144,
"mean_token_accuracy": 0.1909787118434906,
"num_tokens": 14164116.0,
"step": 5070
},
{
"entropy": 5.306113290786743,
"epoch": 0.604526503871352,
"grad_norm": 1.1171875,
"learning_rate": 0.0004973263517903213,
"loss": 4.9942,
"mean_token_accuracy": 0.19805614799261093,
"num_tokens": 14178549.0,
"step": 5075
},
{
"entropy": 5.233040475845337,
"epoch": 0.6051220964860036,
"grad_norm": 1.0,
"learning_rate": 0.0004973197981210481,
"loss": 5.1002,
"mean_token_accuracy": 0.19163627922534943,
"num_tokens": 14191625.0,
"step": 5080
},
{
"entropy": 5.260637426376343,
"epoch": 0.6057176891006552,
"grad_norm": 1.078125,
"learning_rate": 0.0004973132364775252,
"loss": 5.0719,
"mean_token_accuracy": 0.18975056409835817,
"num_tokens": 14205396.0,
"step": 5085
},
{
"entropy": 5.2180005550384525,
"epoch": 0.6063132817153067,
"grad_norm": 1.0234375,
"learning_rate": 0.0004973066668599879,
"loss": 5.104,
"mean_token_accuracy": 0.18550759553909302,
"num_tokens": 14216957.0,
"step": 5090
},
{
"entropy": 5.344973611831665,
"epoch": 0.6069088743299583,
"grad_norm": 0.9296875,
"learning_rate": 0.0004973000892686719,
"loss": 5.1025,
"mean_token_accuracy": 0.18451027572155,
"num_tokens": 14232891.0,
"step": 5095
},
{
"entropy": 5.31133918762207,
"epoch": 0.6075044669446099,
"grad_norm": 1.015625,
"learning_rate": 0.0004972935037038131,
"loss": 5.0448,
"mean_token_accuracy": 0.19374196380376815,
"num_tokens": 14245628.0,
"step": 5100
},
{
"entropy": 5.25178279876709,
"epoch": 0.6081000595592615,
"grad_norm": 0.9375,
"learning_rate": 0.0004972869101656478,
"loss": 5.0723,
"mean_token_accuracy": 0.1899755120277405,
"num_tokens": 14260164.0,
"step": 5105
},
{
"entropy": 5.364306306838989,
"epoch": 0.6086956521739131,
"grad_norm": 0.97265625,
"learning_rate": 0.0004972803086544125,
"loss": 5.1508,
"mean_token_accuracy": 0.18147098869085312,
"num_tokens": 14274828.0,
"step": 5110
},
{
"entropy": 5.312448883056641,
"epoch": 0.6092912447885647,
"grad_norm": 1.0859375,
"learning_rate": 0.0004972736991703439,
"loss": 5.1167,
"mean_token_accuracy": 0.1832786023616791,
"num_tokens": 14287779.0,
"step": 5115
},
{
"entropy": 5.29574728012085,
"epoch": 0.6098868374032163,
"grad_norm": 1.0,
"learning_rate": 0.0004972670817136791,
"loss": 5.0733,
"mean_token_accuracy": 0.18824758976697922,
"num_tokens": 14300946.0,
"step": 5120
},
{
"entropy": 5.240840768814087,
"epoch": 0.6104824300178677,
"grad_norm": 0.98046875,
"learning_rate": 0.0004972604562846554,
"loss": 4.9995,
"mean_token_accuracy": 0.1948929861187935,
"num_tokens": 14314944.0,
"step": 5125
},
{
"entropy": 5.325879812240601,
"epoch": 0.6110780226325193,
"grad_norm": 0.99609375,
"learning_rate": 0.0004972538228835106,
"loss": 5.12,
"mean_token_accuracy": 0.18817064315080642,
"num_tokens": 14330069.0,
"step": 5130
},
{
"entropy": 5.288411235809326,
"epoch": 0.6116736152471709,
"grad_norm": 1.09375,
"learning_rate": 0.0004972471815104825,
"loss": 5.1108,
"mean_token_accuracy": 0.18363292664289474,
"num_tokens": 14344262.0,
"step": 5135
},
{
"entropy": 5.277113962173462,
"epoch": 0.6122692078618225,
"grad_norm": 1.0078125,
"learning_rate": 0.0004972405321658094,
"loss": 5.1074,
"mean_token_accuracy": 0.19191761761903764,
"num_tokens": 14358358.0,
"step": 5140
},
{
"entropy": 5.2646068096160885,
"epoch": 0.6128648004764741,
"grad_norm": 1.0078125,
"learning_rate": 0.0004972338748497297,
"loss": 5.0015,
"mean_token_accuracy": 0.1958531767129898,
"num_tokens": 14371520.0,
"step": 5145
},
{
"entropy": 5.236413240432739,
"epoch": 0.6134603930911257,
"grad_norm": 1.0390625,
"learning_rate": 0.0004972272095624822,
"loss": 4.9428,
"mean_token_accuracy": 0.19725024849176406,
"num_tokens": 14384901.0,
"step": 5150
},
{
"entropy": 5.1662312030792235,
"epoch": 0.6140559857057772,
"grad_norm": 0.96875,
"learning_rate": 0.000497220536304306,
"loss": 4.9525,
"mean_token_accuracy": 0.2038427323102951,
"num_tokens": 14397740.0,
"step": 5155
},
{
"entropy": 5.281447982788086,
"epoch": 0.6146515783204288,
"grad_norm": 1.0859375,
"learning_rate": 0.0004972138550754405,
"loss": 5.1136,
"mean_token_accuracy": 0.18043904304504393,
"num_tokens": 14412850.0,
"step": 5160
},
{
"entropy": 5.375515794754028,
"epoch": 0.6152471709350804,
"grad_norm": 0.984375,
"learning_rate": 0.0004972071658761254,
"loss": 5.1637,
"mean_token_accuracy": 0.17840202003717423,
"num_tokens": 14426174.0,
"step": 5165
},
{
"entropy": 5.296802282333374,
"epoch": 0.615842763549732,
"grad_norm": 0.953125,
"learning_rate": 0.0004972004687066005,
"loss": 5.1517,
"mean_token_accuracy": 0.18499358743429184,
"num_tokens": 14442736.0,
"step": 5170
},
{
"entropy": 5.3573719501495365,
"epoch": 0.6164383561643836,
"grad_norm": 0.9609375,
"learning_rate": 0.0004971937635671061,
"loss": 5.1169,
"mean_token_accuracy": 0.185704804956913,
"num_tokens": 14459631.0,
"step": 5175
},
{
"entropy": 5.250402545928955,
"epoch": 0.6170339487790352,
"grad_norm": 0.96875,
"learning_rate": 0.0004971870504578825,
"loss": 5.0107,
"mean_token_accuracy": 0.18379927724599837,
"num_tokens": 14472896.0,
"step": 5180
},
{
"entropy": 5.291708612442017,
"epoch": 0.6176295413936868,
"grad_norm": 0.953125,
"learning_rate": 0.0004971803293791708,
"loss": 5.0634,
"mean_token_accuracy": 0.18361558765172958,
"num_tokens": 14487959.0,
"step": 5185
},
{
"entropy": 5.193226337432861,
"epoch": 0.6182251340083383,
"grad_norm": 0.9453125,
"learning_rate": 0.0004971736003312119,
"loss": 4.9504,
"mean_token_accuracy": 0.19897136390209197,
"num_tokens": 14501420.0,
"step": 5190
},
{
"entropy": 5.306984567642212,
"epoch": 0.6188207266229899,
"grad_norm": 0.96484375,
"learning_rate": 0.0004971668633142472,
"loss": 5.0719,
"mean_token_accuracy": 0.19501829594373704,
"num_tokens": 14514800.0,
"step": 5195
},
{
"entropy": 5.333307886123658,
"epoch": 0.6194163192376414,
"grad_norm": 1.015625,
"learning_rate": 0.0004971601183285183,
"loss": 5.2006,
"mean_token_accuracy": 0.17928817719221116,
"num_tokens": 14530254.0,
"step": 5200
},
{
"entropy": 5.341461086273194,
"epoch": 0.620011911852293,
"grad_norm": 1.0546875,
"learning_rate": 0.0004971533653742672,
"loss": 5.174,
"mean_token_accuracy": 0.18351953774690627,
"num_tokens": 14545635.0,
"step": 5205
},
{
"entropy": 5.271201133728027,
"epoch": 0.6206075044669446,
"grad_norm": 0.96484375,
"learning_rate": 0.000497146604451736,
"loss": 5.0372,
"mean_token_accuracy": 0.18357431441545485,
"num_tokens": 14558672.0,
"step": 5210
},
{
"entropy": 5.199121570587158,
"epoch": 0.6212030970815962,
"grad_norm": 1.0078125,
"learning_rate": 0.0004971398355611674,
"loss": 5.0937,
"mean_token_accuracy": 0.1879173994064331,
"num_tokens": 14571706.0,
"step": 5215
},
{
"entropy": 5.277734231948853,
"epoch": 0.6217986896962477,
"grad_norm": 1.1328125,
"learning_rate": 0.000497133058702804,
"loss": 5.0578,
"mean_token_accuracy": 0.18633452504873277,
"num_tokens": 14584732.0,
"step": 5220
},
{
"entropy": 5.367848539352417,
"epoch": 0.6223942823108993,
"grad_norm": 1.1015625,
"learning_rate": 0.000497126273876889,
"loss": 5.0711,
"mean_token_accuracy": 0.1857175350189209,
"num_tokens": 14598883.0,
"step": 5225
},
{
"entropy": 5.27206130027771,
"epoch": 0.6229898749255509,
"grad_norm": 0.921875,
"learning_rate": 0.0004971194810836657,
"loss": 5.0502,
"mean_token_accuracy": 0.18820749670267106,
"num_tokens": 14613202.0,
"step": 5230
},
{
"entropy": 5.373336935043335,
"epoch": 0.6235854675402025,
"grad_norm": 1.0,
"learning_rate": 0.0004971126803233778,
"loss": 5.2086,
"mean_token_accuracy": 0.18176941126585006,
"num_tokens": 14627267.0,
"step": 5235
},
{
"entropy": 5.323449230194091,
"epoch": 0.6241810601548541,
"grad_norm": 1.1171875,
"learning_rate": 0.0004971058715962691,
"loss": 5.0966,
"mean_token_accuracy": 0.18643707036972046,
"num_tokens": 14640300.0,
"step": 5240
},
{
"entropy": 5.272343254089355,
"epoch": 0.6247766527695057,
"grad_norm": 1.0078125,
"learning_rate": 0.000497099054902584,
"loss": 5.1231,
"mean_token_accuracy": 0.18567184656858443,
"num_tokens": 14655029.0,
"step": 5245
},
{
"entropy": 5.353311443328858,
"epoch": 0.6253722453841573,
"grad_norm": 0.9609375,
"learning_rate": 0.0004970922302425669,
"loss": 5.0889,
"mean_token_accuracy": 0.18438043296337128,
"num_tokens": 14669321.0,
"step": 5250
},
{
"entropy": 5.245241308212281,
"epoch": 0.6259678379988088,
"grad_norm": 0.9453125,
"learning_rate": 0.0004970853976164624,
"loss": 4.9814,
"mean_token_accuracy": 0.1966283231973648,
"num_tokens": 14682280.0,
"step": 5255
},
{
"entropy": 5.279285287857055,
"epoch": 0.6265634306134604,
"grad_norm": 1.171875,
"learning_rate": 0.000497078557024516,
"loss": 5.1402,
"mean_token_accuracy": 0.1925666868686676,
"num_tokens": 14693761.0,
"step": 5260
},
{
"entropy": 5.280442380905152,
"epoch": 0.627159023228112,
"grad_norm": 1.171875,
"learning_rate": 0.0004970717084669728,
"loss": 5.0315,
"mean_token_accuracy": 0.18646539449691774,
"num_tokens": 14706024.0,
"step": 5265
},
{
"entropy": 5.276974725723266,
"epoch": 0.6277546158427636,
"grad_norm": 0.94921875,
"learning_rate": 0.0004970648519440785,
"loss": 5.0712,
"mean_token_accuracy": 0.185473270714283,
"num_tokens": 14721273.0,
"step": 5270
},
{
"entropy": 5.251651430130005,
"epoch": 0.6283502084574152,
"grad_norm": 0.94140625,
"learning_rate": 0.0004970579874560789,
"loss": 5.0976,
"mean_token_accuracy": 0.1818856731057167,
"num_tokens": 14737033.0,
"step": 5275
},
{
"entropy": 5.289854955673218,
"epoch": 0.6289458010720667,
"grad_norm": 0.89453125,
"learning_rate": 0.0004970511150032205,
"loss": 5.0384,
"mean_token_accuracy": 0.18851457685232162,
"num_tokens": 14751881.0,
"step": 5280
},
{
"entropy": 5.2588685035705565,
"epoch": 0.6295413936867182,
"grad_norm": 1.0859375,
"learning_rate": 0.0004970442345857495,
"loss": 5.0362,
"mean_token_accuracy": 0.18895655870437622,
"num_tokens": 14764866.0,
"step": 5285
},
{
"entropy": 5.31620454788208,
"epoch": 0.6301369863013698,
"grad_norm": 1.0390625,
"learning_rate": 0.0004970373462039128,
"loss": 5.0966,
"mean_token_accuracy": 0.18956113755702972,
"num_tokens": 14777703.0,
"step": 5290
},
{
"entropy": 5.286112213134766,
"epoch": 0.6307325789160214,
"grad_norm": 1.0546875,
"learning_rate": 0.0004970304498579576,
"loss": 5.0606,
"mean_token_accuracy": 0.19186916798353196,
"num_tokens": 14792819.0,
"step": 5295
},
{
"entropy": 5.162862968444824,
"epoch": 0.631328171530673,
"grad_norm": 1.015625,
"learning_rate": 0.0004970235455481312,
"loss": 4.9949,
"mean_token_accuracy": 0.1927088439464569,
"num_tokens": 14806612.0,
"step": 5300
},
{
"entropy": 5.255834531784058,
"epoch": 0.6319237641453246,
"grad_norm": 1.1015625,
"learning_rate": 0.0004970166332746811,
"loss": 5.0381,
"mean_token_accuracy": 0.1965508595108986,
"num_tokens": 14820313.0,
"step": 5305
},
{
"entropy": 5.3676141738891605,
"epoch": 0.6325193567599762,
"grad_norm": 1.0703125,
"learning_rate": 0.0004970097130378554,
"loss": 5.1266,
"mean_token_accuracy": 0.19088566452264785,
"num_tokens": 14832874.0,
"step": 5310
},
{
"entropy": 5.289055967330933,
"epoch": 0.6331149493746278,
"grad_norm": 1.0546875,
"learning_rate": 0.0004970027848379023,
"loss": 5.1086,
"mean_token_accuracy": 0.1893642857670784,
"num_tokens": 14845021.0,
"step": 5315
},
{
"entropy": 5.354841327667236,
"epoch": 0.6337105419892793,
"grad_norm": 1.0859375,
"learning_rate": 0.0004969958486750702,
"loss": 5.1536,
"mean_token_accuracy": 0.1890455886721611,
"num_tokens": 14860928.0,
"step": 5320
},
{
"entropy": 5.321168708801269,
"epoch": 0.6343061346039309,
"grad_norm": 0.96875,
"learning_rate": 0.000496988904549608,
"loss": 5.0148,
"mean_token_accuracy": 0.19722183495759965,
"num_tokens": 14874709.0,
"step": 5325
},
{
"entropy": 5.305152034759521,
"epoch": 0.6349017272185825,
"grad_norm": 0.9453125,
"learning_rate": 0.0004969819524617647,
"loss": 5.1463,
"mean_token_accuracy": 0.1784098356962204,
"num_tokens": 14888323.0,
"step": 5330
},
{
"entropy": 5.282805728912353,
"epoch": 0.6354973198332341,
"grad_norm": 0.9921875,
"learning_rate": 0.0004969749924117896,
"loss": 5.1131,
"mean_token_accuracy": 0.18615499436855315,
"num_tokens": 14901410.0,
"step": 5335
},
{
"entropy": 5.317856025695801,
"epoch": 0.6360929124478857,
"grad_norm": 0.890625,
"learning_rate": 0.0004969680243999325,
"loss": 5.1042,
"mean_token_accuracy": 0.1873985543847084,
"num_tokens": 14915335.0,
"step": 5340
},
{
"entropy": 5.339862203598022,
"epoch": 0.6366885050625373,
"grad_norm": 0.91015625,
"learning_rate": 0.0004969610484264433,
"loss": 5.048,
"mean_token_accuracy": 0.18458202034235,
"num_tokens": 14931549.0,
"step": 5345
},
{
"entropy": 5.212453603744507,
"epoch": 0.6372840976771889,
"grad_norm": 0.94921875,
"learning_rate": 0.0004969540644915721,
"loss": 5.0519,
"mean_token_accuracy": 0.19202922731637956,
"num_tokens": 14943745.0,
"step": 5350
},
{
"entropy": 5.331994152069091,
"epoch": 0.6378796902918403,
"grad_norm": 0.96484375,
"learning_rate": 0.0004969470725955695,
"loss": 5.1465,
"mean_token_accuracy": 0.18600480407476425,
"num_tokens": 14957662.0,
"step": 5355
},
{
"entropy": 5.329409265518189,
"epoch": 0.6384752829064919,
"grad_norm": 1.0234375,
"learning_rate": 0.0004969400727386864,
"loss": 5.1557,
"mean_token_accuracy": 0.18402229696512223,
"num_tokens": 14970175.0,
"step": 5360
},
{
"entropy": 5.280200862884522,
"epoch": 0.6390708755211435,
"grad_norm": 1.046875,
"learning_rate": 0.0004969330649211737,
"loss": 5.1195,
"mean_token_accuracy": 0.18674208521842955,
"num_tokens": 14983901.0,
"step": 5365
},
{
"entropy": 5.393907308578491,
"epoch": 0.6396664681357951,
"grad_norm": 1.1953125,
"learning_rate": 0.0004969260491432826,
"loss": 5.1547,
"mean_token_accuracy": 0.18259987682104112,
"num_tokens": 14997637.0,
"step": 5370
},
{
"entropy": 5.292523717880249,
"epoch": 0.6402620607504467,
"grad_norm": 0.96875,
"learning_rate": 0.0004969190254052651,
"loss": 5.1064,
"mean_token_accuracy": 0.17948938757181168,
"num_tokens": 15014576.0,
"step": 5375
},
{
"entropy": 5.300879335403442,
"epoch": 0.6408576533650983,
"grad_norm": 0.92578125,
"learning_rate": 0.0004969119937073731,
"loss": 5.0835,
"mean_token_accuracy": 0.1831787794828415,
"num_tokens": 15031049.0,
"step": 5380
},
{
"entropy": 5.242600154876709,
"epoch": 0.6414532459797498,
"grad_norm": 1.0234375,
"learning_rate": 0.0004969049540498585,
"loss": 5.0513,
"mean_token_accuracy": 0.18616707921028136,
"num_tokens": 15044632.0,
"step": 5385
},
{
"entropy": 5.245788955688477,
"epoch": 0.6420488385944014,
"grad_norm": 1.03125,
"learning_rate": 0.0004968979064329741,
"loss": 5.0265,
"mean_token_accuracy": 0.1916586220264435,
"num_tokens": 15057500.0,
"step": 5390
},
{
"entropy": 5.317418622970581,
"epoch": 0.642644431209053,
"grad_norm": 0.9375,
"learning_rate": 0.0004968908508569725,
"loss": 5.0957,
"mean_token_accuracy": 0.18928124904632568,
"num_tokens": 15070994.0,
"step": 5395
},
{
"entropy": 5.212126159667969,
"epoch": 0.6432400238237046,
"grad_norm": 0.91015625,
"learning_rate": 0.0004968837873221069,
"loss": 5.0414,
"mean_token_accuracy": 0.18696680516004563,
"num_tokens": 15085824.0,
"step": 5400
},
{
"entropy": 5.252003908157349,
"epoch": 0.6438356164383562,
"grad_norm": 0.87109375,
"learning_rate": 0.0004968767158286306,
"loss": 5.0948,
"mean_token_accuracy": 0.19138366281986235,
"num_tokens": 15101263.0,
"step": 5405
},
{
"entropy": 5.278714513778686,
"epoch": 0.6444312090530078,
"grad_norm": 1.0703125,
"learning_rate": 0.0004968696363767972,
"loss": 5.0004,
"mean_token_accuracy": 0.1978074014186859,
"num_tokens": 15113830.0,
"step": 5410
},
{
"entropy": 5.297952508926391,
"epoch": 0.6450268016676594,
"grad_norm": 1.0234375,
"learning_rate": 0.0004968625489668607,
"loss": 5.1118,
"mean_token_accuracy": 0.18746108710765838,
"num_tokens": 15128427.0,
"step": 5415
},
{
"entropy": 5.286092758178711,
"epoch": 0.6456223942823109,
"grad_norm": 1.0703125,
"learning_rate": 0.0004968554535990754,
"loss": 5.0446,
"mean_token_accuracy": 0.1954586461186409,
"num_tokens": 15143790.0,
"step": 5420
},
{
"entropy": 5.283107757568359,
"epoch": 0.6462179868969625,
"grad_norm": 0.953125,
"learning_rate": 0.0004968483502736956,
"loss": 5.1023,
"mean_token_accuracy": 0.18880869448184967,
"num_tokens": 15157533.0,
"step": 5425
},
{
"entropy": 5.229476737976074,
"epoch": 0.646813579511614,
"grad_norm": 0.99609375,
"learning_rate": 0.0004968412389909762,
"loss": 4.998,
"mean_token_accuracy": 0.18911298513412475,
"num_tokens": 15172164.0,
"step": 5430
},
{
"entropy": 5.279301691055298,
"epoch": 0.6474091721262656,
"grad_norm": 1.140625,
"learning_rate": 0.0004968341197511722,
"loss": 5.0357,
"mean_token_accuracy": 0.19701821953058243,
"num_tokens": 15184334.0,
"step": 5435
},
{
"entropy": 5.191413354873657,
"epoch": 0.6480047647409172,
"grad_norm": 1.015625,
"learning_rate": 0.0004968269925545391,
"loss": 5.0468,
"mean_token_accuracy": 0.18830501139163972,
"num_tokens": 15198783.0,
"step": 5440
},
{
"entropy": 5.247761631011963,
"epoch": 0.6486003573555688,
"grad_norm": 1.0078125,
"learning_rate": 0.0004968198574013323,
"loss": 5.0401,
"mean_token_accuracy": 0.1913047045469284,
"num_tokens": 15214153.0,
"step": 5445
},
{
"entropy": 5.342239618301392,
"epoch": 0.6491959499702203,
"grad_norm": 1.1171875,
"learning_rate": 0.0004968127142918081,
"loss": 5.134,
"mean_token_accuracy": 0.1817881390452385,
"num_tokens": 15229029.0,
"step": 5450
},
{
"entropy": 5.264386415481567,
"epoch": 0.6497915425848719,
"grad_norm": 0.86328125,
"learning_rate": 0.0004968055632262223,
"loss": 5.0598,
"mean_token_accuracy": 0.18658847063779832,
"num_tokens": 15244724.0,
"step": 5455
},
{
"entropy": 5.273080444335937,
"epoch": 0.6503871351995235,
"grad_norm": 0.9921875,
"learning_rate": 0.0004967984042048317,
"loss": 5.0853,
"mean_token_accuracy": 0.18482004404067992,
"num_tokens": 15258703.0,
"step": 5460
},
{
"entropy": 5.350534248352051,
"epoch": 0.6509827278141751,
"grad_norm": 0.8984375,
"learning_rate": 0.0004967912372278928,
"loss": 5.178,
"mean_token_accuracy": 0.18285827338695526,
"num_tokens": 15272281.0,
"step": 5465
},
{
"entropy": 5.293913269042969,
"epoch": 0.6515783204288267,
"grad_norm": 1.0,
"learning_rate": 0.000496784062295663,
"loss": 5.0942,
"mean_token_accuracy": 0.18722161501646042,
"num_tokens": 15286657.0,
"step": 5470
},
{
"entropy": 5.289205408096313,
"epoch": 0.6521739130434783,
"grad_norm": 0.99609375,
"learning_rate": 0.0004967768794083994,
"loss": 5.0711,
"mean_token_accuracy": 0.1903505340218544,
"num_tokens": 15300416.0,
"step": 5475
},
{
"entropy": 5.249413919448853,
"epoch": 0.6527695056581299,
"grad_norm": 1.125,
"learning_rate": 0.0004967696885663597,
"loss": 5.0116,
"mean_token_accuracy": 0.18952654153108597,
"num_tokens": 15313945.0,
"step": 5480
},
{
"entropy": 5.221739816665649,
"epoch": 0.6533650982727814,
"grad_norm": 0.9609375,
"learning_rate": 0.000496762489769802,
"loss": 5.0153,
"mean_token_accuracy": 0.19326596558094025,
"num_tokens": 15328995.0,
"step": 5485
},
{
"entropy": 5.252273988723755,
"epoch": 0.653960690887433,
"grad_norm": 1.046875,
"learning_rate": 0.0004967552830189842,
"loss": 4.9982,
"mean_token_accuracy": 0.19307326376438141,
"num_tokens": 15342588.0,
"step": 5490
},
{
"entropy": 5.250972890853882,
"epoch": 0.6545562835020846,
"grad_norm": 0.96484375,
"learning_rate": 0.0004967480683141651,
"loss": 5.0268,
"mean_token_accuracy": 0.1962678462266922,
"num_tokens": 15357411.0,
"step": 5495
},
{
"entropy": 5.195329093933106,
"epoch": 0.6551518761167362,
"grad_norm": 0.98828125,
"learning_rate": 0.000496740845655603,
"loss": 5.038,
"mean_token_accuracy": 0.19176963716745377,
"num_tokens": 15371968.0,
"step": 5500
},
{
"entropy": 5.272292041778565,
"epoch": 0.6557474687313878,
"grad_norm": 0.94140625,
"learning_rate": 0.0004967336150435576,
"loss": 5.0479,
"mean_token_accuracy": 0.19267256557941437,
"num_tokens": 15387186.0,
"step": 5505
},
{
"entropy": 5.238138628005982,
"epoch": 0.6563430613460393,
"grad_norm": 1.109375,
"learning_rate": 0.0004967263764782878,
"loss": 5.0041,
"mean_token_accuracy": 0.19592117220163346,
"num_tokens": 15398842.0,
"step": 5510
},
{
"entropy": 5.224260091781616,
"epoch": 0.6569386539606908,
"grad_norm": 1.0,
"learning_rate": 0.0004967191299600535,
"loss": 5.008,
"mean_token_accuracy": 0.1924358233809471,
"num_tokens": 15411687.0,
"step": 5515
},
{
"entropy": 5.175791025161743,
"epoch": 0.6575342465753424,
"grad_norm": 1.0859375,
"learning_rate": 0.0004967118754891144,
"loss": 4.9326,
"mean_token_accuracy": 0.19553456753492354,
"num_tokens": 15425868.0,
"step": 5520
},
{
"entropy": 5.324447584152222,
"epoch": 0.658129839189994,
"grad_norm": 0.95703125,
"learning_rate": 0.0004967046130657309,
"loss": 5.1488,
"mean_token_accuracy": 0.17814426273107528,
"num_tokens": 15439489.0,
"step": 5525
},
{
"entropy": 5.328158569335938,
"epoch": 0.6587254318046456,
"grad_norm": 1.203125,
"learning_rate": 0.0004966973426901634,
"loss": 5.1297,
"mean_token_accuracy": 0.19055284410715104,
"num_tokens": 15453067.0,
"step": 5530
},
{
"entropy": 5.209204244613647,
"epoch": 0.6593210244192972,
"grad_norm": 0.94140625,
"learning_rate": 0.0004966900643626726,
"loss": 4.9596,
"mean_token_accuracy": 0.1963087499141693,
"num_tokens": 15465357.0,
"step": 5535
},
{
"entropy": 5.22384786605835,
"epoch": 0.6599166170339488,
"grad_norm": 1.0625,
"learning_rate": 0.0004966827780835197,
"loss": 5.0893,
"mean_token_accuracy": 0.18884702026844025,
"num_tokens": 15479657.0,
"step": 5540
},
{
"entropy": 5.459466791152954,
"epoch": 0.6605122096486004,
"grad_norm": 1.0703125,
"learning_rate": 0.0004966754838529659,
"loss": 5.3208,
"mean_token_accuracy": 0.17242255955934524,
"num_tokens": 15495245.0,
"step": 5545
},
{
"entropy": 5.3127727031707765,
"epoch": 0.6611078022632519,
"grad_norm": 1.109375,
"learning_rate": 0.000496668181671273,
"loss": 5.0203,
"mean_token_accuracy": 0.1900005280971527,
"num_tokens": 15507788.0,
"step": 5550
},
{
"entropy": 5.225294208526611,
"epoch": 0.6617033948779035,
"grad_norm": 0.91796875,
"learning_rate": 0.0004966608715387029,
"loss": 5.0923,
"mean_token_accuracy": 0.1860947862267494,
"num_tokens": 15521970.0,
"step": 5555
},
{
"entropy": 5.261061525344848,
"epoch": 0.6622989874925551,
"grad_norm": 1.03125,
"learning_rate": 0.0004966535534555175,
"loss": 5.0148,
"mean_token_accuracy": 0.19425233453512192,
"num_tokens": 15534720.0,
"step": 5560
},
{
"entropy": 5.266478824615478,
"epoch": 0.6628945801072067,
"grad_norm": 1.1796875,
"learning_rate": 0.0004966462274219797,
"loss": 5.1024,
"mean_token_accuracy": 0.18198966681957246,
"num_tokens": 15549345.0,
"step": 5565
},
{
"entropy": 5.288567447662354,
"epoch": 0.6634901727218583,
"grad_norm": 0.96484375,
"learning_rate": 0.000496638893438352,
"loss": 5.1286,
"mean_token_accuracy": 0.18770654499530792,
"num_tokens": 15562838.0,
"step": 5570
},
{
"entropy": 5.326519775390625,
"epoch": 0.6640857653365099,
"grad_norm": 1.109375,
"learning_rate": 0.0004966315515048977,
"loss": 5.0696,
"mean_token_accuracy": 0.19171165823936462,
"num_tokens": 15575044.0,
"step": 5575
},
{
"entropy": 5.268227005004883,
"epoch": 0.6646813579511615,
"grad_norm": 0.921875,
"learning_rate": 0.0004966242016218799,
"loss": 5.034,
"mean_token_accuracy": 0.19378841519355774,
"num_tokens": 15591912.0,
"step": 5580
},
{
"entropy": 5.315447473526001,
"epoch": 0.6652769505658129,
"grad_norm": 0.98046875,
"learning_rate": 0.0004966168437895624,
"loss": 5.0718,
"mean_token_accuracy": 0.18524022698402404,
"num_tokens": 15606903.0,
"step": 5585
},
{
"entropy": 5.232182836532592,
"epoch": 0.6658725431804645,
"grad_norm": 1.015625,
"learning_rate": 0.0004966094780082089,
"loss": 4.9889,
"mean_token_accuracy": 0.1953340768814087,
"num_tokens": 15620885.0,
"step": 5590
},
{
"entropy": 5.343390560150146,
"epoch": 0.6664681357951161,
"grad_norm": 0.9765625,
"learning_rate": 0.0004966021042780837,
"loss": 5.0474,
"mean_token_accuracy": 0.18884147852659225,
"num_tokens": 15634927.0,
"step": 5595
},
{
"entropy": 5.2636474609375,
"epoch": 0.6670637284097677,
"grad_norm": 0.9921875,
"learning_rate": 0.0004965947225994516,
"loss": 5.0694,
"mean_token_accuracy": 0.1895744800567627,
"num_tokens": 15649369.0,
"step": 5600
},
{
"entropy": 5.270068407058716,
"epoch": 0.6676593210244193,
"grad_norm": 1.0,
"learning_rate": 0.0004965873329725769,
"loss": 5.0885,
"mean_token_accuracy": 0.18069495558738707,
"num_tokens": 15665288.0,
"step": 5605
},
{
"entropy": 5.296831655502319,
"epoch": 0.6682549136390709,
"grad_norm": 0.98828125,
"learning_rate": 0.0004965799353977248,
"loss": 5.0144,
"mean_token_accuracy": 0.19215119779109954,
"num_tokens": 15679117.0,
"step": 5610
},
{
"entropy": 5.2183443069458,
"epoch": 0.6688505062537224,
"grad_norm": 1.0546875,
"learning_rate": 0.0004965725298751608,
"loss": 4.96,
"mean_token_accuracy": 0.1995480015873909,
"num_tokens": 15692177.0,
"step": 5615
},
{
"entropy": 5.2758708000183105,
"epoch": 0.669446098868374,
"grad_norm": 1.0390625,
"learning_rate": 0.0004965651164051504,
"loss": 5.0027,
"mean_token_accuracy": 0.20209874361753463,
"num_tokens": 15706047.0,
"step": 5620
},
{
"entropy": 5.260066795349121,
"epoch": 0.6700416914830256,
"grad_norm": 0.96875,
"learning_rate": 0.0004965576949879594,
"loss": 5.134,
"mean_token_accuracy": 0.18586853444576262,
"num_tokens": 15721191.0,
"step": 5625
},
{
"entropy": 5.349901247024536,
"epoch": 0.6706372840976772,
"grad_norm": 0.92578125,
"learning_rate": 0.0004965502656238543,
"loss": 5.0849,
"mean_token_accuracy": 0.18478769063949585,
"num_tokens": 15735569.0,
"step": 5630
},
{
"entropy": 5.280852890014648,
"epoch": 0.6712328767123288,
"grad_norm": 1.0703125,
"learning_rate": 0.0004965428283131013,
"loss": 4.9949,
"mean_token_accuracy": 0.1951826587319374,
"num_tokens": 15749125.0,
"step": 5635
},
{
"entropy": 5.205649709701538,
"epoch": 0.6718284693269804,
"grad_norm": 0.921875,
"learning_rate": 0.0004965353830559671,
"loss": 5.041,
"mean_token_accuracy": 0.19599765986204148,
"num_tokens": 15765641.0,
"step": 5640
},
{
"entropy": 5.299585247039795,
"epoch": 0.672424061941632,
"grad_norm": 0.8984375,
"learning_rate": 0.0004965279298527192,
"loss": 5.0938,
"mean_token_accuracy": 0.1844843327999115,
"num_tokens": 15780816.0,
"step": 5645
},
{
"entropy": 5.328836250305176,
"epoch": 0.6730196545562835,
"grad_norm": 0.9609375,
"learning_rate": 0.0004965204687036244,
"loss": 5.1619,
"mean_token_accuracy": 0.1795637398958206,
"num_tokens": 15795123.0,
"step": 5650
},
{
"entropy": 5.243233489990234,
"epoch": 0.6736152471709351,
"grad_norm": 1.0078125,
"learning_rate": 0.0004965129996089507,
"loss": 5.022,
"mean_token_accuracy": 0.197913758456707,
"num_tokens": 15808428.0,
"step": 5655
},
{
"entropy": 5.30166482925415,
"epoch": 0.6742108397855867,
"grad_norm": 0.95703125,
"learning_rate": 0.0004965055225689658,
"loss": 5.1376,
"mean_token_accuracy": 0.1827080264687538,
"num_tokens": 15820618.0,
"step": 5660
},
{
"entropy": 5.2875278949737545,
"epoch": 0.6748064324002382,
"grad_norm": 1.0859375,
"learning_rate": 0.0004964980375839379,
"loss": 4.9812,
"mean_token_accuracy": 0.19748762100934983,
"num_tokens": 15833431.0,
"step": 5665
},
{
"entropy": 5.247407531738281,
"epoch": 0.6754020250148898,
"grad_norm": 0.9921875,
"learning_rate": 0.0004964905446541355,
"loss": 4.9941,
"mean_token_accuracy": 0.1989613026380539,
"num_tokens": 15845629.0,
"step": 5670
},
{
"entropy": 5.328744840621948,
"epoch": 0.6759976176295414,
"grad_norm": 0.9921875,
"learning_rate": 0.0004964830437798276,
"loss": 5.1093,
"mean_token_accuracy": 0.18075974136590958,
"num_tokens": 15859464.0,
"step": 5675
},
{
"entropy": 5.18330717086792,
"epoch": 0.6765932102441929,
"grad_norm": 0.91796875,
"learning_rate": 0.0004964755349612829,
"loss": 4.941,
"mean_token_accuracy": 0.20337564945220948,
"num_tokens": 15873207.0,
"step": 5680
},
{
"entropy": 5.247575473785401,
"epoch": 0.6771888028588445,
"grad_norm": 1.0546875,
"learning_rate": 0.0004964680181987709,
"loss": 5.0495,
"mean_token_accuracy": 0.18798441886901857,
"num_tokens": 15888856.0,
"step": 5685
},
{
"entropy": 5.246871471405029,
"epoch": 0.6777843954734961,
"grad_norm": 1.0078125,
"learning_rate": 0.000496460493492561,
"loss": 4.9827,
"mean_token_accuracy": 0.19902548491954802,
"num_tokens": 15901841.0,
"step": 5690
},
{
"entropy": 5.302401304244995,
"epoch": 0.6783799880881477,
"grad_norm": 1.0625,
"learning_rate": 0.0004964529608429234,
"loss": 5.0647,
"mean_token_accuracy": 0.19169118404388427,
"num_tokens": 15914960.0,
"step": 5695
},
{
"entropy": 5.162592601776123,
"epoch": 0.6789755807027993,
"grad_norm": 0.9140625,
"learning_rate": 0.0004964454202501282,
"loss": 4.9885,
"mean_token_accuracy": 0.19473856687545776,
"num_tokens": 15929466.0,
"step": 5700
},
{
"entropy": 5.321282911300659,
"epoch": 0.6795711733174509,
"grad_norm": 1.0078125,
"learning_rate": 0.0004964378717144458,
"loss": 5.1198,
"mean_token_accuracy": 0.18536283820867538,
"num_tokens": 15943934.0,
"step": 5705
},
{
"entropy": 5.270995187759399,
"epoch": 0.6801667659321025,
"grad_norm": 1.1328125,
"learning_rate": 0.000496430315236147,
"loss": 5.0467,
"mean_token_accuracy": 0.1986171320080757,
"num_tokens": 15956214.0,
"step": 5710
},
{
"entropy": 5.285578203201294,
"epoch": 0.680762358546754,
"grad_norm": 1.0390625,
"learning_rate": 0.0004964227508155027,
"loss": 5.0147,
"mean_token_accuracy": 0.187625452876091,
"num_tokens": 15970943.0,
"step": 5715
},
{
"entropy": 5.2747467994689945,
"epoch": 0.6813579511614056,
"grad_norm": 1.015625,
"learning_rate": 0.0004964151784527844,
"loss": 5.127,
"mean_token_accuracy": 0.18986504673957824,
"num_tokens": 15985539.0,
"step": 5720
},
{
"entropy": 5.24966950416565,
"epoch": 0.6819535437760572,
"grad_norm": 1.1796875,
"learning_rate": 0.0004964075981482637,
"loss": 4.9564,
"mean_token_accuracy": 0.20102044939994812,
"num_tokens": 15998622.0,
"step": 5725
},
{
"entropy": 5.28323278427124,
"epoch": 0.6825491363907088,
"grad_norm": 0.9921875,
"learning_rate": 0.0004964000099022124,
"loss": 5.0657,
"mean_token_accuracy": 0.194697830080986,
"num_tokens": 16013820.0,
"step": 5730
},
{
"entropy": 5.363418102264404,
"epoch": 0.6831447290053604,
"grad_norm": 1.1328125,
"learning_rate": 0.0004963924137149027,
"loss": 5.179,
"mean_token_accuracy": 0.1819551855325699,
"num_tokens": 16026615.0,
"step": 5735
},
{
"entropy": 5.307887935638428,
"epoch": 0.6837403216200119,
"grad_norm": 1.0390625,
"learning_rate": 0.0004963848095866072,
"loss": 5.0889,
"mean_token_accuracy": 0.18798820674419403,
"num_tokens": 16039921.0,
"step": 5740
},
{
"entropy": 5.145106363296509,
"epoch": 0.6843359142346634,
"grad_norm": 0.921875,
"learning_rate": 0.0004963771975175984,
"loss": 4.9377,
"mean_token_accuracy": 0.19763311594724656,
"num_tokens": 16054790.0,
"step": 5745
},
{
"entropy": 5.178565549850464,
"epoch": 0.684931506849315,
"grad_norm": 1.0,
"learning_rate": 0.0004963695775081495,
"loss": 5.0523,
"mean_token_accuracy": 0.18603767454624176,
"num_tokens": 16069778.0,
"step": 5750
},
{
"entropy": 5.21142086982727,
"epoch": 0.6855270994639666,
"grad_norm": 0.953125,
"learning_rate": 0.0004963619495585338,
"loss": 4.9884,
"mean_token_accuracy": 0.19405000805854797,
"num_tokens": 16086472.0,
"step": 5755
},
{
"entropy": 5.291449928283692,
"epoch": 0.6861226920786182,
"grad_norm": 0.94921875,
"learning_rate": 0.0004963543136690247,
"loss": 5.1366,
"mean_token_accuracy": 0.18848897218704225,
"num_tokens": 16100743.0,
"step": 5760
},
{
"entropy": 5.2107891082763675,
"epoch": 0.6867182846932698,
"grad_norm": 0.9765625,
"learning_rate": 0.0004963466698398965,
"loss": 4.9887,
"mean_token_accuracy": 0.19094934314489365,
"num_tokens": 16114894.0,
"step": 5765
},
{
"entropy": 5.2387885570526125,
"epoch": 0.6873138773079214,
"grad_norm": 1.0703125,
"learning_rate": 0.0004963390180714231,
"loss": 5.0278,
"mean_token_accuracy": 0.19607480615377426,
"num_tokens": 16129232.0,
"step": 5770
},
{
"entropy": 5.244196939468384,
"epoch": 0.687909469922573,
"grad_norm": 1.0078125,
"learning_rate": 0.000496331358363879,
"loss": 4.9442,
"mean_token_accuracy": 0.19929671436548232,
"num_tokens": 16142983.0,
"step": 5775
},
{
"entropy": 5.2671925067901615,
"epoch": 0.6885050625372245,
"grad_norm": 1.2734375,
"learning_rate": 0.0004963236907175388,
"loss": 5.0733,
"mean_token_accuracy": 0.19095949828624725,
"num_tokens": 16157195.0,
"step": 5780
},
{
"entropy": 5.230446243286133,
"epoch": 0.6891006551518761,
"grad_norm": 1.046875,
"learning_rate": 0.0004963160151326778,
"loss": 5.0745,
"mean_token_accuracy": 0.18954994082450866,
"num_tokens": 16171240.0,
"step": 5785
},
{
"entropy": 5.352623462677002,
"epoch": 0.6896962477665277,
"grad_norm": 0.9453125,
"learning_rate": 0.0004963083316095712,
"loss": 5.087,
"mean_token_accuracy": 0.18528846502304078,
"num_tokens": 16186313.0,
"step": 5790
},
{
"entropy": 5.271135902404785,
"epoch": 0.6902918403811793,
"grad_norm": 1.078125,
"learning_rate": 0.0004963006401484945,
"loss": 5.0956,
"mean_token_accuracy": 0.18692163974046708,
"num_tokens": 16201394.0,
"step": 5795
},
{
"entropy": 5.194055080413818,
"epoch": 0.6908874329958309,
"grad_norm": 1.0546875,
"learning_rate": 0.0004962929407497236,
"loss": 5.0122,
"mean_token_accuracy": 0.19800458699464799,
"num_tokens": 16214917.0,
"step": 5800
},
{
"entropy": 5.270834112167359,
"epoch": 0.6914830256104825,
"grad_norm": 0.99609375,
"learning_rate": 0.0004962852334135348,
"loss": 4.9915,
"mean_token_accuracy": 0.19284452348947526,
"num_tokens": 16228700.0,
"step": 5805
},
{
"entropy": 5.209543514251709,
"epoch": 0.6920786182251341,
"grad_norm": 1.0234375,
"learning_rate": 0.0004962775181402044,
"loss": 4.9037,
"mean_token_accuracy": 0.2023202210664749,
"num_tokens": 16240262.0,
"step": 5810
},
{
"entropy": 5.180195093154907,
"epoch": 0.6926742108397855,
"grad_norm": 1.0546875,
"learning_rate": 0.0004962697949300093,
"loss": 4.9599,
"mean_token_accuracy": 0.19715639650821687,
"num_tokens": 16254830.0,
"step": 5815
},
{
"entropy": 5.200127077102661,
"epoch": 0.6932698034544371,
"grad_norm": 1.1015625,
"learning_rate": 0.0004962620637832263,
"loss": 4.9084,
"mean_token_accuracy": 0.20332298278808594,
"num_tokens": 16267547.0,
"step": 5820
},
{
"entropy": 5.222266101837159,
"epoch": 0.6938653960690887,
"grad_norm": 1.0859375,
"learning_rate": 0.0004962543247001329,
"loss": 4.9849,
"mean_token_accuracy": 0.1953933835029602,
"num_tokens": 16281223.0,
"step": 5825
},
{
"entropy": 5.3081449508667,
"epoch": 0.6944609886837403,
"grad_norm": 0.984375,
"learning_rate": 0.0004962465776810066,
"loss": 5.1444,
"mean_token_accuracy": 0.18559690117835997,
"num_tokens": 16295366.0,
"step": 5830
},
{
"entropy": 5.317699718475342,
"epoch": 0.6950565812983919,
"grad_norm": 0.96484375,
"learning_rate": 0.0004962388227261253,
"loss": 5.076,
"mean_token_accuracy": 0.19120134711265563,
"num_tokens": 16310081.0,
"step": 5835
},
{
"entropy": 5.31610426902771,
"epoch": 0.6956521739130435,
"grad_norm": 0.96875,
"learning_rate": 0.0004962310598357672,
"loss": 5.1209,
"mean_token_accuracy": 0.18748878240585326,
"num_tokens": 16324544.0,
"step": 5840
},
{
"entropy": 5.346761417388916,
"epoch": 0.696247766527695,
"grad_norm": 1.0078125,
"learning_rate": 0.0004962232890102106,
"loss": 5.1683,
"mean_token_accuracy": 0.19018675535917282,
"num_tokens": 16339128.0,
"step": 5845
},
{
"entropy": 5.2298853397369385,
"epoch": 0.6968433591423466,
"grad_norm": 0.9140625,
"learning_rate": 0.0004962155102497342,
"loss": 4.9999,
"mean_token_accuracy": 0.19616030901670456,
"num_tokens": 16351997.0,
"step": 5850
},
{
"entropy": 5.1909592628479,
"epoch": 0.6974389517569982,
"grad_norm": 1.0546875,
"learning_rate": 0.0004962077235546173,
"loss": 4.9805,
"mean_token_accuracy": 0.19996345341205596,
"num_tokens": 16365075.0,
"step": 5855
},
{
"entropy": 5.191736507415771,
"epoch": 0.6980345443716498,
"grad_norm": 1.140625,
"learning_rate": 0.0004961999289251389,
"loss": 5.0014,
"mean_token_accuracy": 0.2038533791899681,
"num_tokens": 16378635.0,
"step": 5860
},
{
"entropy": 5.250685310363769,
"epoch": 0.6986301369863014,
"grad_norm": 1.0859375,
"learning_rate": 0.0004961921263615788,
"loss": 4.9757,
"mean_token_accuracy": 0.20547545850276946,
"num_tokens": 16392224.0,
"step": 5865
},
{
"entropy": 5.200765657424927,
"epoch": 0.699225729600953,
"grad_norm": 1.0,
"learning_rate": 0.0004961843158642167,
"loss": 4.9706,
"mean_token_accuracy": 0.1950543239712715,
"num_tokens": 16407830.0,
"step": 5870
},
{
"entropy": 5.163029909133911,
"epoch": 0.6998213222156046,
"grad_norm": 0.984375,
"learning_rate": 0.0004961764974333328,
"loss": 5.0445,
"mean_token_accuracy": 0.19691397547721862,
"num_tokens": 16421576.0,
"step": 5875
},
{
"entropy": 5.323457479476929,
"epoch": 0.7004169148302561,
"grad_norm": 0.9921875,
"learning_rate": 0.0004961686710692076,
"loss": 5.0848,
"mean_token_accuracy": 0.19107279777526856,
"num_tokens": 16436548.0,
"step": 5880
},
{
"entropy": 5.290355110168457,
"epoch": 0.7010125074449077,
"grad_norm": 0.9296875,
"learning_rate": 0.0004961608367721217,
"loss": 5.0991,
"mean_token_accuracy": 0.192501936852932,
"num_tokens": 16451435.0,
"step": 5885
},
{
"entropy": 5.271506023406983,
"epoch": 0.7016081000595593,
"grad_norm": 1.0,
"learning_rate": 0.0004961529945423563,
"loss": 5.0899,
"mean_token_accuracy": 0.18329890817403793,
"num_tokens": 16465786.0,
"step": 5890
},
{
"entropy": 5.31269679069519,
"epoch": 0.7022036926742108,
"grad_norm": 1.03125,
"learning_rate": 0.0004961451443801924,
"loss": 5.0765,
"mean_token_accuracy": 0.1890162318944931,
"num_tokens": 16479832.0,
"step": 5895
},
{
"entropy": 5.234739923477173,
"epoch": 0.7027992852888624,
"grad_norm": 1.140625,
"learning_rate": 0.0004961372862859119,
"loss": 4.9691,
"mean_token_accuracy": 0.20115922540426254,
"num_tokens": 16491824.0,
"step": 5900
},
{
"entropy": 5.263653564453125,
"epoch": 0.703394877903514,
"grad_norm": 1.09375,
"learning_rate": 0.0004961294202597962,
"loss": 5.0091,
"mean_token_accuracy": 0.1893574208021164,
"num_tokens": 16507273.0,
"step": 5905
},
{
"entropy": 5.251615190505982,
"epoch": 0.7039904705181655,
"grad_norm": 1.0390625,
"learning_rate": 0.0004961215463021281,
"loss": 5.0372,
"mean_token_accuracy": 0.1916299819946289,
"num_tokens": 16521041.0,
"step": 5910
},
{
"entropy": 5.236955451965332,
"epoch": 0.7045860631328171,
"grad_norm": 0.9765625,
"learning_rate": 0.0004961136644131894,
"loss": 4.9959,
"mean_token_accuracy": 0.1968044713139534,
"num_tokens": 16534837.0,
"step": 5915
},
{
"entropy": 5.301002407073975,
"epoch": 0.7051816557474687,
"grad_norm": 0.93359375,
"learning_rate": 0.0004961057745932633,
"loss": 5.0801,
"mean_token_accuracy": 0.19276404976844788,
"num_tokens": 16549183.0,
"step": 5920
},
{
"entropy": 5.226874732971192,
"epoch": 0.7057772483621203,
"grad_norm": 0.9609375,
"learning_rate": 0.0004960978768426323,
"loss": 5.0268,
"mean_token_accuracy": 0.19167272746562958,
"num_tokens": 16562044.0,
"step": 5925
},
{
"entropy": 5.309398221969604,
"epoch": 0.7063728409767719,
"grad_norm": 0.96875,
"learning_rate": 0.00049608997116158,
"loss": 5.1256,
"mean_token_accuracy": 0.1873836487531662,
"num_tokens": 16577014.0,
"step": 5930
},
{
"entropy": 5.2666035175323485,
"epoch": 0.7069684335914235,
"grad_norm": 1.046875,
"learning_rate": 0.00049608205755039,
"loss": 4.9907,
"mean_token_accuracy": 0.20199701189994812,
"num_tokens": 16588289.0,
"step": 5935
},
{
"entropy": 5.229057741165161,
"epoch": 0.7075640262060751,
"grad_norm": 1.078125,
"learning_rate": 0.0004960741360093459,
"loss": 5.0631,
"mean_token_accuracy": 0.19096744805574417,
"num_tokens": 16602433.0,
"step": 5940
},
{
"entropy": 5.350700569152832,
"epoch": 0.7081596188207266,
"grad_norm": 0.9140625,
"learning_rate": 0.0004960662065387322,
"loss": 5.1309,
"mean_token_accuracy": 0.18654951304197312,
"num_tokens": 16618178.0,
"step": 5945
},
{
"entropy": 5.3043416976928714,
"epoch": 0.7087552114353782,
"grad_norm": 0.953125,
"learning_rate": 0.000496058269138833,
"loss": 5.0332,
"mean_token_accuracy": 0.19613948464393616,
"num_tokens": 16634090.0,
"step": 5950
},
{
"entropy": 5.212548780441284,
"epoch": 0.7093508040500298,
"grad_norm": 1.1015625,
"learning_rate": 0.0004960503238099331,
"loss": 4.9483,
"mean_token_accuracy": 0.20183636397123336,
"num_tokens": 16646713.0,
"step": 5955
},
{
"entropy": 5.23512601852417,
"epoch": 0.7099463966646814,
"grad_norm": 1.1015625,
"learning_rate": 0.0004960423705523175,
"loss": 4.988,
"mean_token_accuracy": 0.19313346296548844,
"num_tokens": 16659172.0,
"step": 5960
},
{
"entropy": 5.263709688186646,
"epoch": 0.710541989279333,
"grad_norm": 1.046875,
"learning_rate": 0.0004960344093662714,
"loss": 5.0892,
"mean_token_accuracy": 0.19360265433788298,
"num_tokens": 16672592.0,
"step": 5965
},
{
"entropy": 5.292951822280884,
"epoch": 0.7111375818939845,
"grad_norm": 0.96875,
"learning_rate": 0.0004960264402520804,
"loss": 5.0538,
"mean_token_accuracy": 0.1871548667550087,
"num_tokens": 16687026.0,
"step": 5970
},
{
"entropy": 5.224291753768921,
"epoch": 0.711733174508636,
"grad_norm": 1.09375,
"learning_rate": 0.0004960184632100304,
"loss": 4.9245,
"mean_token_accuracy": 0.20600415617227555,
"num_tokens": 16699489.0,
"step": 5975
},
{
"entropy": 5.217588567733765,
"epoch": 0.7123287671232876,
"grad_norm": 0.90625,
"learning_rate": 0.0004960104782404075,
"loss": 5.0047,
"mean_token_accuracy": 0.19554176479578017,
"num_tokens": 16713821.0,
"step": 5980
},
{
"entropy": 5.306798267364502,
"epoch": 0.7129243597379392,
"grad_norm": 0.9609375,
"learning_rate": 0.0004960024853434982,
"loss": 5.09,
"mean_token_accuracy": 0.19533285349607468,
"num_tokens": 16727271.0,
"step": 5985
},
{
"entropy": 5.288178586959839,
"epoch": 0.7135199523525908,
"grad_norm": 1.0390625,
"learning_rate": 0.000495994484519589,
"loss": 4.978,
"mean_token_accuracy": 0.19798011630773543,
"num_tokens": 16740387.0,
"step": 5990
},
{
"entropy": 5.191060638427734,
"epoch": 0.7141155449672424,
"grad_norm": 1.015625,
"learning_rate": 0.0004959864757689669,
"loss": 5.0175,
"mean_token_accuracy": 0.1933645561337471,
"num_tokens": 16756131.0,
"step": 5995
},
{
"entropy": 5.212758207321167,
"epoch": 0.714711137581894,
"grad_norm": 1.015625,
"learning_rate": 0.0004959784590919192,
"loss": 4.9917,
"mean_token_accuracy": 0.19731339812278748,
"num_tokens": 16768781.0,
"step": 6000
},
{
"epoch": 0.714711137581894,
"eval_entropy": 5.033751152899572,
"eval_loss": 4.9821343421936035,
"eval_mean_token_accuracy": 0.20422642360146517,
"eval_num_tokens": 16768781.0,
"eval_runtime": 18.5538,
"eval_samples_per_second": 1375.081,
"eval_steps_per_second": 171.932,
"step": 6000
},
{
"entropy": 5.315912389755249,
"epoch": 0.7153067301965456,
"grad_norm": 0.95703125,
"learning_rate": 0.0004959704344887335,
"loss": 5.0582,
"mean_token_accuracy": 0.18988261222839356,
"num_tokens": 16782933.0,
"step": 6005
},
{
"entropy": 5.225536870956421,
"epoch": 0.7159023228111971,
"grad_norm": 1.03125,
"learning_rate": 0.0004959624019596976,
"loss": 4.9996,
"mean_token_accuracy": 0.1970362350344658,
"num_tokens": 16795999.0,
"step": 6010
},
{
"entropy": 5.298355674743652,
"epoch": 0.7164979154258487,
"grad_norm": 1.1484375,
"learning_rate": 0.0004959543615050996,
"loss": 5.0652,
"mean_token_accuracy": 0.18930322378873826,
"num_tokens": 16809042.0,
"step": 6015
},
{
"entropy": 5.256893014907837,
"epoch": 0.7170935080405003,
"grad_norm": 1.1328125,
"learning_rate": 0.0004959463131252278,
"loss": 5.0111,
"mean_token_accuracy": 0.19432156383991242,
"num_tokens": 16820896.0,
"step": 6020
},
{
"entropy": 5.289313220977784,
"epoch": 0.7176891006551519,
"grad_norm": 1.03125,
"learning_rate": 0.0004959382568203711,
"loss": 5.0263,
"mean_token_accuracy": 0.18863192647695542,
"num_tokens": 16834418.0,
"step": 6025
},
{
"entropy": 5.2070306777954105,
"epoch": 0.7182846932698035,
"grad_norm": 0.94921875,
"learning_rate": 0.0004959301925908184,
"loss": 4.9908,
"mean_token_accuracy": 0.19485723674297334,
"num_tokens": 16848818.0,
"step": 6030
},
{
"entropy": 5.304145288467407,
"epoch": 0.7188802858844551,
"grad_norm": 1.015625,
"learning_rate": 0.0004959221204368588,
"loss": 5.071,
"mean_token_accuracy": 0.19293779134750366,
"num_tokens": 16863145.0,
"step": 6035
},
{
"entropy": 5.258409595489502,
"epoch": 0.7194758784991067,
"grad_norm": 0.984375,
"learning_rate": 0.0004959140403587819,
"loss": 5.0523,
"mean_token_accuracy": 0.1896543934941292,
"num_tokens": 16876342.0,
"step": 6040
},
{
"entropy": 5.270143032073975,
"epoch": 0.7200714711137582,
"grad_norm": 0.984375,
"learning_rate": 0.0004959059523568778,
"loss": 5.0725,
"mean_token_accuracy": 0.1852214068174362,
"num_tokens": 16892113.0,
"step": 6045
},
{
"entropy": 5.338851261138916,
"epoch": 0.7206670637284097,
"grad_norm": 1.0078125,
"learning_rate": 0.0004958978564314361,
"loss": 5.0268,
"mean_token_accuracy": 0.18928490281105043,
"num_tokens": 16906103.0,
"step": 6050
},
{
"entropy": 5.220661735534668,
"epoch": 0.7212626563430613,
"grad_norm": 0.88671875,
"learning_rate": 0.0004958897525827476,
"loss": 5.0268,
"mean_token_accuracy": 0.1944548025727272,
"num_tokens": 16921655.0,
"step": 6055
},
{
"entropy": 5.217870807647705,
"epoch": 0.7218582489577129,
"grad_norm": 1.1171875,
"learning_rate": 0.0004958816408111027,
"loss": 5.03,
"mean_token_accuracy": 0.19421614557504654,
"num_tokens": 16936143.0,
"step": 6060
},
{
"entropy": 5.2129405498504635,
"epoch": 0.7224538415723645,
"grad_norm": 1.2265625,
"learning_rate": 0.0004958735211167926,
"loss": 5.009,
"mean_token_accuracy": 0.1947688192129135,
"num_tokens": 16950232.0,
"step": 6065
},
{
"entropy": 5.240213298797608,
"epoch": 0.7230494341870161,
"grad_norm": 1.0390625,
"learning_rate": 0.0004958653935001083,
"loss": 4.9437,
"mean_token_accuracy": 0.197464956343174,
"num_tokens": 16963407.0,
"step": 6070
},
{
"entropy": 5.233933782577514,
"epoch": 0.7236450268016676,
"grad_norm": 1.140625,
"learning_rate": 0.0004958572579613415,
"loss": 5.0941,
"mean_token_accuracy": 0.1956447944045067,
"num_tokens": 16975758.0,
"step": 6075
},
{
"entropy": 5.265483808517456,
"epoch": 0.7242406194163192,
"grad_norm": 1.03125,
"learning_rate": 0.000495849114500784,
"loss": 5.0125,
"mean_token_accuracy": 0.19357312023639678,
"num_tokens": 16989496.0,
"step": 6080
},
{
"entropy": 5.282686519622803,
"epoch": 0.7248362120309708,
"grad_norm": 1.1015625,
"learning_rate": 0.0004958409631187278,
"loss": 5.0884,
"mean_token_accuracy": 0.19001179039478303,
"num_tokens": 17002790.0,
"step": 6085
},
{
"entropy": 5.244546413421631,
"epoch": 0.7254318046456224,
"grad_norm": 0.96484375,
"learning_rate": 0.0004958328038154652,
"loss": 4.9853,
"mean_token_accuracy": 0.19572992473840714,
"num_tokens": 17016384.0,
"step": 6090
},
{
"entropy": 5.256314468383789,
"epoch": 0.726027397260274,
"grad_norm": 0.91796875,
"learning_rate": 0.0004958246365912891,
"loss": 5.0742,
"mean_token_accuracy": 0.20317401438951493,
"num_tokens": 17031331.0,
"step": 6095
},
{
"entropy": 5.258700132369995,
"epoch": 0.7266229898749256,
"grad_norm": 1.015625,
"learning_rate": 0.0004958164614464923,
"loss": 5.0035,
"mean_token_accuracy": 0.1902099147439003,
"num_tokens": 17043820.0,
"step": 6100
},
{
"entropy": 5.188328552246094,
"epoch": 0.7272185824895772,
"grad_norm": 1.0234375,
"learning_rate": 0.000495808278381368,
"loss": 4.9363,
"mean_token_accuracy": 0.20457157343626023,
"num_tokens": 17057826.0,
"step": 6105
},
{
"entropy": 5.232245302200317,
"epoch": 0.7278141751042287,
"grad_norm": 0.9453125,
"learning_rate": 0.0004958000873962099,
"loss": 5.0315,
"mean_token_accuracy": 0.1934985935688019,
"num_tokens": 17073410.0,
"step": 6110
},
{
"entropy": 5.244672727584839,
"epoch": 0.7284097677188803,
"grad_norm": 0.9765625,
"learning_rate": 0.0004957918884913115,
"loss": 4.9794,
"mean_token_accuracy": 0.19663485139608383,
"num_tokens": 17086705.0,
"step": 6115
},
{
"entropy": 5.310059118270874,
"epoch": 0.7290053603335319,
"grad_norm": 0.96875,
"learning_rate": 0.0004957836816669672,
"loss": 5.063,
"mean_token_accuracy": 0.18842167109251023,
"num_tokens": 17100787.0,
"step": 6120
},
{
"entropy": 5.207601833343506,
"epoch": 0.7296009529481834,
"grad_norm": 0.92578125,
"learning_rate": 0.0004957754669234711,
"loss": 5.0211,
"mean_token_accuracy": 0.19228127896785735,
"num_tokens": 17116381.0,
"step": 6125
},
{
"entropy": 5.1536914825439455,
"epoch": 0.730196545562835,
"grad_norm": 1.0390625,
"learning_rate": 0.000495767244261118,
"loss": 4.9465,
"mean_token_accuracy": 0.2040816456079483,
"num_tokens": 17129214.0,
"step": 6130
},
{
"entropy": 5.29478178024292,
"epoch": 0.7307921381774866,
"grad_norm": 1.109375,
"learning_rate": 0.0004957590136802028,
"loss": 5.0599,
"mean_token_accuracy": 0.19626087844371795,
"num_tokens": 17143033.0,
"step": 6135
},
{
"entropy": 5.202221012115478,
"epoch": 0.7313877307921381,
"grad_norm": 1.09375,
"learning_rate": 0.0004957507751810207,
"loss": 4.9552,
"mean_token_accuracy": 0.1929696962237358,
"num_tokens": 17156325.0,
"step": 6140
},
{
"entropy": 5.300933361053467,
"epoch": 0.7319833234067897,
"grad_norm": 0.97265625,
"learning_rate": 0.0004957425287638673,
"loss": 5.0265,
"mean_token_accuracy": 0.19122539609670638,
"num_tokens": 17173031.0,
"step": 6145
},
{
"entropy": 5.267198133468628,
"epoch": 0.7325789160214413,
"grad_norm": 1.015625,
"learning_rate": 0.0004957342744290381,
"loss": 5.0128,
"mean_token_accuracy": 0.1912374824285507,
"num_tokens": 17186579.0,
"step": 6150
},
{
"entropy": 5.143907880783081,
"epoch": 0.7331745086360929,
"grad_norm": 1.125,
"learning_rate": 0.0004957260121768296,
"loss": 4.9321,
"mean_token_accuracy": 0.2015999287366867,
"num_tokens": 17200987.0,
"step": 6155
},
{
"entropy": 5.2405191421508786,
"epoch": 0.7337701012507445,
"grad_norm": 0.90234375,
"learning_rate": 0.0004957177420075377,
"loss": 4.9733,
"mean_token_accuracy": 0.19909586161375045,
"num_tokens": 17215429.0,
"step": 6160
},
{
"entropy": 5.198292350769043,
"epoch": 0.7343656938653961,
"grad_norm": 0.953125,
"learning_rate": 0.0004957094639214594,
"loss": 4.9787,
"mean_token_accuracy": 0.19999172687530517,
"num_tokens": 17229460.0,
"step": 6165
},
{
"entropy": 5.203000259399414,
"epoch": 0.7349612864800477,
"grad_norm": 0.87109375,
"learning_rate": 0.0004957011779188915,
"loss": 5.0173,
"mean_token_accuracy": 0.19541125744581223,
"num_tokens": 17243399.0,
"step": 6170
},
{
"entropy": 5.153001880645752,
"epoch": 0.7355568790946992,
"grad_norm": 0.99609375,
"learning_rate": 0.0004956928840001312,
"loss": 4.8967,
"mean_token_accuracy": 0.20010535269975663,
"num_tokens": 17258028.0,
"step": 6175
},
{
"entropy": 5.175713205337525,
"epoch": 0.7361524717093508,
"grad_norm": 0.98046875,
"learning_rate": 0.0004956845821654758,
"loss": 4.9206,
"mean_token_accuracy": 0.20203646421432495,
"num_tokens": 17271666.0,
"step": 6180
},
{
"entropy": 5.298681735992432,
"epoch": 0.7367480643240024,
"grad_norm": 1.1171875,
"learning_rate": 0.0004956762724152234,
"loss": 5.0615,
"mean_token_accuracy": 0.1909468486905098,
"num_tokens": 17285423.0,
"step": 6185
},
{
"entropy": 5.245342350006103,
"epoch": 0.737343656938654,
"grad_norm": 0.9453125,
"learning_rate": 0.0004956679547496719,
"loss": 5.1255,
"mean_token_accuracy": 0.18809884488582612,
"num_tokens": 17299134.0,
"step": 6190
},
{
"entropy": 5.302482604980469,
"epoch": 0.7379392495533056,
"grad_norm": 1.125,
"learning_rate": 0.0004956596291691197,
"loss": 5.0712,
"mean_token_accuracy": 0.19051984697580338,
"num_tokens": 17311835.0,
"step": 6195
},
{
"entropy": 5.319841814041138,
"epoch": 0.7385348421679571,
"grad_norm": 1.0390625,
"learning_rate": 0.0004956512956738652,
"loss": 5.0506,
"mean_token_accuracy": 0.19398678094148636,
"num_tokens": 17324943.0,
"step": 6200
},
{
"entropy": 5.220056867599487,
"epoch": 0.7391304347826086,
"grad_norm": 1.1015625,
"learning_rate": 0.0004956429542642075,
"loss": 5.0626,
"mean_token_accuracy": 0.18147763162851333,
"num_tokens": 17339710.0,
"step": 6205
},
{
"entropy": 5.304931163787842,
"epoch": 0.7397260273972602,
"grad_norm": 1.046875,
"learning_rate": 0.0004956346049404459,
"loss": 5.035,
"mean_token_accuracy": 0.19851257652044296,
"num_tokens": 17353391.0,
"step": 6210
},
{
"entropy": 5.244458103179932,
"epoch": 0.7403216200119118,
"grad_norm": 1.0234375,
"learning_rate": 0.0004956262477028797,
"loss": 5.0262,
"mean_token_accuracy": 0.18980411440134048,
"num_tokens": 17367811.0,
"step": 6215
},
{
"entropy": 5.220595502853394,
"epoch": 0.7409172126265634,
"grad_norm": 0.94921875,
"learning_rate": 0.0004956178825518087,
"loss": 5.0137,
"mean_token_accuracy": 0.19670378267765046,
"num_tokens": 17380323.0,
"step": 6220
},
{
"entropy": 5.259941673278808,
"epoch": 0.741512805241215,
"grad_norm": 1.0625,
"learning_rate": 0.0004956095094875329,
"loss": 4.983,
"mean_token_accuracy": 0.202933406829834,
"num_tokens": 17394650.0,
"step": 6225
},
{
"entropy": 5.261579990386963,
"epoch": 0.7421083978558666,
"grad_norm": 0.984375,
"learning_rate": 0.0004956011285103528,
"loss": 4.9945,
"mean_token_accuracy": 0.19662576764822007,
"num_tokens": 17408461.0,
"step": 6230
},
{
"entropy": 5.270742702484131,
"epoch": 0.7427039904705182,
"grad_norm": 1.1015625,
"learning_rate": 0.0004955927396205689,
"loss": 5.0489,
"mean_token_accuracy": 0.196359084546566,
"num_tokens": 17422086.0,
"step": 6235
},
{
"entropy": 5.235152959823608,
"epoch": 0.7432995830851697,
"grad_norm": 0.95703125,
"learning_rate": 0.000495584342818482,
"loss": 4.9909,
"mean_token_accuracy": 0.1983842670917511,
"num_tokens": 17436845.0,
"step": 6240
},
{
"entropy": 5.1596198081970215,
"epoch": 0.7438951756998213,
"grad_norm": 1.0390625,
"learning_rate": 0.0004955759381043934,
"loss": 4.9219,
"mean_token_accuracy": 0.19904813915491104,
"num_tokens": 17451783.0,
"step": 6245
},
{
"entropy": 5.36769061088562,
"epoch": 0.7444907683144729,
"grad_norm": 1.0078125,
"learning_rate": 0.0004955675254786045,
"loss": 5.2221,
"mean_token_accuracy": 0.18005552440881728,
"num_tokens": 17466310.0,
"step": 6250
},
{
"entropy": 5.250159025192261,
"epoch": 0.7450863609291245,
"grad_norm": 1.03125,
"learning_rate": 0.0004955591049414171,
"loss": 4.9774,
"mean_token_accuracy": 0.1959860071539879,
"num_tokens": 17481595.0,
"step": 6255
},
{
"entropy": 5.280541944503784,
"epoch": 0.7456819535437761,
"grad_norm": 0.97265625,
"learning_rate": 0.0004955506764931332,
"loss": 5.0454,
"mean_token_accuracy": 0.19590727984905243,
"num_tokens": 17496947.0,
"step": 6260
},
{
"entropy": 5.209000301361084,
"epoch": 0.7462775461584277,
"grad_norm": 1.1171875,
"learning_rate": 0.0004955422401340553,
"loss": 4.9443,
"mean_token_accuracy": 0.19418541043996812,
"num_tokens": 17509772.0,
"step": 6265
},
{
"entropy": 5.177103662490845,
"epoch": 0.7468731387730793,
"grad_norm": 1.046875,
"learning_rate": 0.0004955337958644857,
"loss": 5.0369,
"mean_token_accuracy": 0.19827371686697007,
"num_tokens": 17522819.0,
"step": 6270
},
{
"entropy": 5.29366979598999,
"epoch": 0.7474687313877308,
"grad_norm": 1.046875,
"learning_rate": 0.0004955253436847273,
"loss": 5.0106,
"mean_token_accuracy": 0.19525747001171112,
"num_tokens": 17536524.0,
"step": 6275
},
{
"entropy": 5.2548445701599125,
"epoch": 0.7480643240023823,
"grad_norm": 1.078125,
"learning_rate": 0.0004955168835950835,
"loss": 4.9724,
"mean_token_accuracy": 0.19834137558937073,
"num_tokens": 17549033.0,
"step": 6280
},
{
"entropy": 5.216752243041992,
"epoch": 0.7486599166170339,
"grad_norm": 1.0390625,
"learning_rate": 0.0004955084155958576,
"loss": 5.0029,
"mean_token_accuracy": 0.19402988106012345,
"num_tokens": 17562615.0,
"step": 6285
},
{
"entropy": 5.226556301116943,
"epoch": 0.7492555092316855,
"grad_norm": 1.0234375,
"learning_rate": 0.0004954999396873534,
"loss": 4.9464,
"mean_token_accuracy": 0.19425200074911117,
"num_tokens": 17576388.0,
"step": 6290
},
{
"entropy": 5.217913007736206,
"epoch": 0.7498511018463371,
"grad_norm": 1.0546875,
"learning_rate": 0.0004954914558698748,
"loss": 4.9326,
"mean_token_accuracy": 0.20156165361404418,
"num_tokens": 17588275.0,
"step": 6295
},
{
"entropy": 5.192208862304687,
"epoch": 0.7504466944609887,
"grad_norm": 0.9453125,
"learning_rate": 0.0004954829641437261,
"loss": 5.0209,
"mean_token_accuracy": 0.1970132604241371,
"num_tokens": 17601534.0,
"step": 6300
},
{
"entropy": 5.192778539657593,
"epoch": 0.7510422870756402,
"grad_norm": 1.109375,
"learning_rate": 0.0004954744645092121,
"loss": 4.9691,
"mean_token_accuracy": 0.19475208669900895,
"num_tokens": 17614801.0,
"step": 6305
},
{
"entropy": 5.2234632015228275,
"epoch": 0.7516378796902918,
"grad_norm": 1.1640625,
"learning_rate": 0.0004954659569666375,
"loss": 4.9242,
"mean_token_accuracy": 0.20828134268522264,
"num_tokens": 17628056.0,
"step": 6310
},
{
"entropy": 5.222542238235474,
"epoch": 0.7522334723049434,
"grad_norm": 1.046875,
"learning_rate": 0.0004954574415163075,
"loss": 5.0374,
"mean_token_accuracy": 0.19400094002485274,
"num_tokens": 17642676.0,
"step": 6315
},
{
"entropy": 5.21635422706604,
"epoch": 0.752829064919595,
"grad_norm": 1.015625,
"learning_rate": 0.0004954489181585275,
"loss": 4.9719,
"mean_token_accuracy": 0.19849273264408113,
"num_tokens": 17655911.0,
"step": 6320
},
{
"entropy": 5.2171228408813475,
"epoch": 0.7534246575342466,
"grad_norm": 1.1015625,
"learning_rate": 0.0004954403868936032,
"loss": 5.0592,
"mean_token_accuracy": 0.19450298100709915,
"num_tokens": 17669386.0,
"step": 6325
},
{
"entropy": 5.283917760848999,
"epoch": 0.7540202501488982,
"grad_norm": 1.0,
"learning_rate": 0.0004954318477218406,
"loss": 5.0002,
"mean_token_accuracy": 0.19972282648086548,
"num_tokens": 17683426.0,
"step": 6330
},
{
"entropy": 5.363969945907593,
"epoch": 0.7546158427635498,
"grad_norm": 0.90234375,
"learning_rate": 0.000495423300643546,
"loss": 5.1678,
"mean_token_accuracy": 0.1822700783610344,
"num_tokens": 17697188.0,
"step": 6335
},
{
"entropy": 5.2457301139831545,
"epoch": 0.7552114353782013,
"grad_norm": 1.03125,
"learning_rate": 0.0004954147456590262,
"loss": 4.9968,
"mean_token_accuracy": 0.19821426570415496,
"num_tokens": 17711302.0,
"step": 6340
},
{
"entropy": 5.295173645019531,
"epoch": 0.7558070279928529,
"grad_norm": 0.94140625,
"learning_rate": 0.0004954061827685877,
"loss": 5.071,
"mean_token_accuracy": 0.196294142305851,
"num_tokens": 17725173.0,
"step": 6345
},
{
"entropy": 5.269693326950073,
"epoch": 0.7564026206075045,
"grad_norm": 0.984375,
"learning_rate": 0.0004953976119725378,
"loss": 5.0291,
"mean_token_accuracy": 0.18754708170890808,
"num_tokens": 17740770.0,
"step": 6350
},
{
"entropy": 5.227873134613037,
"epoch": 0.756998213222156,
"grad_norm": 1.0390625,
"learning_rate": 0.0004953890332711839,
"loss": 5.0026,
"mean_token_accuracy": 0.1868187367916107,
"num_tokens": 17754077.0,
"step": 6355
},
{
"entropy": 5.248606967926025,
"epoch": 0.7575938058368076,
"grad_norm": 1.03125,
"learning_rate": 0.0004953804466648337,
"loss": 5.0236,
"mean_token_accuracy": 0.19412286281585694,
"num_tokens": 17768572.0,
"step": 6360
},
{
"entropy": 5.229633378982544,
"epoch": 0.7581893984514592,
"grad_norm": 0.98828125,
"learning_rate": 0.0004953718521537952,
"loss": 5.0415,
"mean_token_accuracy": 0.19131537675857543,
"num_tokens": 17783678.0,
"step": 6365
},
{
"entropy": 5.249992609024048,
"epoch": 0.7587849910661107,
"grad_norm": 1.0703125,
"learning_rate": 0.0004953632497383766,
"loss": 5.0196,
"mean_token_accuracy": 0.1904531165957451,
"num_tokens": 17796453.0,
"step": 6370
},
{
"entropy": 5.217026281356811,
"epoch": 0.7593805836807623,
"grad_norm": 0.96875,
"learning_rate": 0.0004953546394188867,
"loss": 4.989,
"mean_token_accuracy": 0.1982920140028,
"num_tokens": 17810459.0,
"step": 6375
},
{
"entropy": 5.2041398048400875,
"epoch": 0.7599761762954139,
"grad_norm": 0.99609375,
"learning_rate": 0.000495346021195634,
"loss": 4.9461,
"mean_token_accuracy": 0.19724552184343339,
"num_tokens": 17822976.0,
"step": 6380
},
{
"entropy": 5.280095005035401,
"epoch": 0.7605717689100655,
"grad_norm": 1.109375,
"learning_rate": 0.0004953373950689279,
"loss": 5.0373,
"mean_token_accuracy": 0.1962214320898056,
"num_tokens": 17836842.0,
"step": 6385
},
{
"entropy": 5.276013898849487,
"epoch": 0.7611673615247171,
"grad_norm": 1.046875,
"learning_rate": 0.0004953287610390775,
"loss": 5.0637,
"mean_token_accuracy": 0.19559549242258073,
"num_tokens": 17849300.0,
"step": 6390
},
{
"entropy": 5.218065929412842,
"epoch": 0.7617629541393687,
"grad_norm": 0.95703125,
"learning_rate": 0.0004953201191063928,
"loss": 4.9642,
"mean_token_accuracy": 0.2056609496474266,
"num_tokens": 17862526.0,
"step": 6395
},
{
"entropy": 5.239102602005005,
"epoch": 0.7623585467540203,
"grad_norm": 1.0390625,
"learning_rate": 0.0004953114692711837,
"loss": 4.9531,
"mean_token_accuracy": 0.197422556579113,
"num_tokens": 17874895.0,
"step": 6400
},
{
"entropy": 5.365125131607056,
"epoch": 0.7629541393686718,
"grad_norm": 1.015625,
"learning_rate": 0.0004953028115337603,
"loss": 5.1249,
"mean_token_accuracy": 0.1886749267578125,
"num_tokens": 17888898.0,
"step": 6405
},
{
"entropy": 5.254565811157226,
"epoch": 0.7635497319833234,
"grad_norm": 1.0234375,
"learning_rate": 0.0004952941458944333,
"loss": 5.0375,
"mean_token_accuracy": 0.19601858109235765,
"num_tokens": 17902185.0,
"step": 6410
},
{
"entropy": 5.235987329483033,
"epoch": 0.764145324597975,
"grad_norm": 1.03125,
"learning_rate": 0.0004952854723535134,
"loss": 5.0501,
"mean_token_accuracy": 0.1944836124777794,
"num_tokens": 17917309.0,
"step": 6415
},
{
"entropy": 5.274077033996582,
"epoch": 0.7647409172126266,
"grad_norm": 1.1796875,
"learning_rate": 0.0004952767909113118,
"loss": 5.019,
"mean_token_accuracy": 0.18800001442432404,
"num_tokens": 17931147.0,
"step": 6420
},
{
"entropy": 5.2962137222290036,
"epoch": 0.7653365098272782,
"grad_norm": 1.0703125,
"learning_rate": 0.0004952681015681398,
"loss": 5.016,
"mean_token_accuracy": 0.19110841751098634,
"num_tokens": 17944792.0,
"step": 6425
},
{
"entropy": 5.245725011825561,
"epoch": 0.7659321024419298,
"grad_norm": 1.171875,
"learning_rate": 0.0004952594043243092,
"loss": 5.0382,
"mean_token_accuracy": 0.19326163232326507,
"num_tokens": 17958580.0,
"step": 6430
},
{
"entropy": 5.27111234664917,
"epoch": 0.7665276950565812,
"grad_norm": 0.98828125,
"learning_rate": 0.0004952506991801319,
"loss": 5.0152,
"mean_token_accuracy": 0.1957751363515854,
"num_tokens": 17972245.0,
"step": 6435
},
{
"entropy": 5.214343118667602,
"epoch": 0.7671232876712328,
"grad_norm": 0.94921875,
"learning_rate": 0.0004952419861359201,
"loss": 5.0101,
"mean_token_accuracy": 0.19959178566932678,
"num_tokens": 17987082.0,
"step": 6440
},
{
"entropy": 5.265559339523316,
"epoch": 0.7677188802858844,
"grad_norm": 1.0859375,
"learning_rate": 0.0004952332651919864,
"loss": 5.0509,
"mean_token_accuracy": 0.19265156537294387,
"num_tokens": 18001477.0,
"step": 6445
},
{
"entropy": 5.257998943328857,
"epoch": 0.768314472900536,
"grad_norm": 1.0,
"learning_rate": 0.0004952245363486435,
"loss": 4.9935,
"mean_token_accuracy": 0.1947251155972481,
"num_tokens": 18014553.0,
"step": 6450
},
{
"entropy": 5.241938972473145,
"epoch": 0.7689100655151876,
"grad_norm": 1.0625,
"learning_rate": 0.0004952157996062046,
"loss": 4.9518,
"mean_token_accuracy": 0.19388168007135392,
"num_tokens": 18027454.0,
"step": 6455
},
{
"entropy": 5.249928522109985,
"epoch": 0.7695056581298392,
"grad_norm": 0.9453125,
"learning_rate": 0.000495207054964983,
"loss": 5.0345,
"mean_token_accuracy": 0.19968264997005464,
"num_tokens": 18040632.0,
"step": 6460
},
{
"entropy": 5.188938140869141,
"epoch": 0.7701012507444908,
"grad_norm": 1.1171875,
"learning_rate": 0.0004951983024252924,
"loss": 4.9366,
"mean_token_accuracy": 0.20443653017282487,
"num_tokens": 18053859.0,
"step": 6465
},
{
"entropy": 5.197919034957886,
"epoch": 0.7706968433591423,
"grad_norm": 0.9609375,
"learning_rate": 0.0004951895419874466,
"loss": 4.9247,
"mean_token_accuracy": 0.1967886731028557,
"num_tokens": 18067734.0,
"step": 6470
},
{
"entropy": 5.274455881118774,
"epoch": 0.7712924359737939,
"grad_norm": 0.9609375,
"learning_rate": 0.00049518077365176,
"loss": 5.0605,
"mean_token_accuracy": 0.19145707935094833,
"num_tokens": 18083831.0,
"step": 6475
},
{
"entropy": 5.262814092636108,
"epoch": 0.7718880285884455,
"grad_norm": 1.0390625,
"learning_rate": 0.000495171997418547,
"loss": 4.9402,
"mean_token_accuracy": 0.20350784212350845,
"num_tokens": 18097085.0,
"step": 6480
},
{
"entropy": 5.203363466262817,
"epoch": 0.7724836212030971,
"grad_norm": 0.953125,
"learning_rate": 0.0004951632132881225,
"loss": 4.9361,
"mean_token_accuracy": 0.19885888546705247,
"num_tokens": 18111311.0,
"step": 6485
},
{
"entropy": 5.229902601242065,
"epoch": 0.7730792138177487,
"grad_norm": 1.09375,
"learning_rate": 0.0004951544212608015,
"loss": 4.9972,
"mean_token_accuracy": 0.2002006307244301,
"num_tokens": 18126622.0,
"step": 6490
},
{
"entropy": 5.303749895095825,
"epoch": 0.7736748064324003,
"grad_norm": 1.0625,
"learning_rate": 0.0004951456213368994,
"loss": 5.0358,
"mean_token_accuracy": 0.19430384784936905,
"num_tokens": 18138074.0,
"step": 6495
},
{
"entropy": 5.196266508102417,
"epoch": 0.7742703990470519,
"grad_norm": 1.0078125,
"learning_rate": 0.0004951368135167316,
"loss": 4.9861,
"mean_token_accuracy": 0.20177092552185058,
"num_tokens": 18150865.0,
"step": 6500
},
{
"entropy": 5.183703088760376,
"epoch": 0.7748659916617034,
"grad_norm": 1.0625,
"learning_rate": 0.0004951279978006143,
"loss": 4.9076,
"mean_token_accuracy": 0.20593772381544112,
"num_tokens": 18163779.0,
"step": 6505
},
{
"entropy": 5.212162494659424,
"epoch": 0.7754615842763549,
"grad_norm": 1.046875,
"learning_rate": 0.0004951191741888638,
"loss": 4.9995,
"mean_token_accuracy": 0.19343529939651488,
"num_tokens": 18177924.0,
"step": 6510
},
{
"entropy": 5.323496866226196,
"epoch": 0.7760571768910065,
"grad_norm": 0.9765625,
"learning_rate": 0.0004951103426817961,
"loss": 5.1083,
"mean_token_accuracy": 0.18052300363779067,
"num_tokens": 18191818.0,
"step": 6515
},
{
"entropy": 5.266404533386231,
"epoch": 0.7766527695056581,
"grad_norm": 1.0546875,
"learning_rate": 0.0004951015032797286,
"loss": 5.0283,
"mean_token_accuracy": 0.19075798094272614,
"num_tokens": 18205681.0,
"step": 6520
},
{
"entropy": 5.188322257995606,
"epoch": 0.7772483621203097,
"grad_norm": 0.99609375,
"learning_rate": 0.0004950926559829778,
"loss": 5.0443,
"mean_token_accuracy": 0.19074616134166716,
"num_tokens": 18220420.0,
"step": 6525
},
{
"entropy": 5.183560180664062,
"epoch": 0.7778439547349613,
"grad_norm": 1.03125,
"learning_rate": 0.0004950838007918615,
"loss": 4.9249,
"mean_token_accuracy": 0.2013428121805191,
"num_tokens": 18233891.0,
"step": 6530
},
{
"entropy": 5.230009889602661,
"epoch": 0.7784395473496128,
"grad_norm": 1.0390625,
"learning_rate": 0.000495074937706697,
"loss": 4.9786,
"mean_token_accuracy": 0.19963490068912507,
"num_tokens": 18246671.0,
"step": 6535
},
{
"entropy": 5.174095535278321,
"epoch": 0.7790351399642644,
"grad_norm": 1.0390625,
"learning_rate": 0.0004950660667278023,
"loss": 5.0016,
"mean_token_accuracy": 0.19353700280189515,
"num_tokens": 18260351.0,
"step": 6540
},
{
"entropy": 5.203449726104736,
"epoch": 0.779630732578916,
"grad_norm": 1.203125,
"learning_rate": 0.0004950571878554956,
"loss": 4.9356,
"mean_token_accuracy": 0.20444403290748597,
"num_tokens": 18275475.0,
"step": 6545
},
{
"entropy": 5.203335857391357,
"epoch": 0.7802263251935676,
"grad_norm": 1.0546875,
"learning_rate": 0.0004950483010900953,
"loss": 5.0191,
"mean_token_accuracy": 0.19109074026346207,
"num_tokens": 18289598.0,
"step": 6550
},
{
"entropy": 5.176995038986206,
"epoch": 0.7808219178082192,
"grad_norm": 1.0234375,
"learning_rate": 0.0004950394064319203,
"loss": 4.9006,
"mean_token_accuracy": 0.1962534263730049,
"num_tokens": 18302118.0,
"step": 6555
},
{
"entropy": 5.280214309692383,
"epoch": 0.7814175104228708,
"grad_norm": 0.98828125,
"learning_rate": 0.0004950305038812895,
"loss": 5.1165,
"mean_token_accuracy": 0.18959629088640212,
"num_tokens": 18316462.0,
"step": 6560
},
{
"entropy": 5.239339876174927,
"epoch": 0.7820131030375224,
"grad_norm": 1.046875,
"learning_rate": 0.0004950215934385223,
"loss": 4.9487,
"mean_token_accuracy": 0.20173856914043425,
"num_tokens": 18328241.0,
"step": 6565
},
{
"entropy": 5.246262836456299,
"epoch": 0.782608695652174,
"grad_norm": 1.1015625,
"learning_rate": 0.0004950126751039382,
"loss": 5.0016,
"mean_token_accuracy": 0.19621127396821975,
"num_tokens": 18341580.0,
"step": 6570
},
{
"entropy": 5.200206613540649,
"epoch": 0.7832042882668255,
"grad_norm": 1.0390625,
"learning_rate": 0.0004950037488778572,
"loss": 5.0413,
"mean_token_accuracy": 0.18947482705116273,
"num_tokens": 18354996.0,
"step": 6575
},
{
"entropy": 5.244330930709839,
"epoch": 0.7837998808814771,
"grad_norm": 1.1171875,
"learning_rate": 0.0004949948147605996,
"loss": 5.0112,
"mean_token_accuracy": 0.1966759517788887,
"num_tokens": 18369284.0,
"step": 6580
},
{
"entropy": 5.268989944458008,
"epoch": 0.7843954734961287,
"grad_norm": 0.99609375,
"learning_rate": 0.0004949858727524855,
"loss": 5.0835,
"mean_token_accuracy": 0.19031598418951035,
"num_tokens": 18383008.0,
"step": 6585
},
{
"entropy": 5.246993494033814,
"epoch": 0.7849910661107802,
"grad_norm": 0.96484375,
"learning_rate": 0.0004949769228538359,
"loss": 4.9708,
"mean_token_accuracy": 0.1958140179514885,
"num_tokens": 18396676.0,
"step": 6590
},
{
"entropy": 5.166881418228149,
"epoch": 0.7855866587254318,
"grad_norm": 0.9921875,
"learning_rate": 0.0004949679650649717,
"loss": 4.9563,
"mean_token_accuracy": 0.188148532807827,
"num_tokens": 18411155.0,
"step": 6595
},
{
"entropy": 5.1803357124328615,
"epoch": 0.7861822513400833,
"grad_norm": 0.96875,
"learning_rate": 0.0004949589993862143,
"loss": 4.9271,
"mean_token_accuracy": 0.19651424437761306,
"num_tokens": 18425793.0,
"step": 6600
},
{
"entropy": 5.1673651218414305,
"epoch": 0.7867778439547349,
"grad_norm": 1.0546875,
"learning_rate": 0.0004949500258178851,
"loss": 4.9263,
"mean_token_accuracy": 0.20513025373220445,
"num_tokens": 18438920.0,
"step": 6605
},
{
"entropy": 5.206192111968994,
"epoch": 0.7873734365693865,
"grad_norm": 0.94140625,
"learning_rate": 0.0004949410443603062,
"loss": 4.977,
"mean_token_accuracy": 0.19365163445472716,
"num_tokens": 18454440.0,
"step": 6610
},
{
"entropy": 5.293622398376465,
"epoch": 0.7879690291840381,
"grad_norm": 1.0,
"learning_rate": 0.0004949320550137996,
"loss": 5.0002,
"mean_token_accuracy": 0.19985851943492888,
"num_tokens": 18469269.0,
"step": 6615
},
{
"entropy": 5.2897570610046385,
"epoch": 0.7885646217986897,
"grad_norm": 0.984375,
"learning_rate": 0.0004949230577786877,
"loss": 5.0812,
"mean_token_accuracy": 0.18586118370294571,
"num_tokens": 18484071.0,
"step": 6620
},
{
"entropy": 5.231650495529175,
"epoch": 0.7891602144133413,
"grad_norm": 1.0078125,
"learning_rate": 0.0004949140526552934,
"loss": 5.0179,
"mean_token_accuracy": 0.19302808493375778,
"num_tokens": 18498647.0,
"step": 6625
},
{
"entropy": 5.279578304290771,
"epoch": 0.7897558070279929,
"grad_norm": 1.0078125,
"learning_rate": 0.0004949050396439395,
"loss": 5.006,
"mean_token_accuracy": 0.1939067378640175,
"num_tokens": 18513406.0,
"step": 6630
},
{
"entropy": 5.207550287246704,
"epoch": 0.7903513996426444,
"grad_norm": 1.1640625,
"learning_rate": 0.0004948960187449495,
"loss": 4.9994,
"mean_token_accuracy": 0.19381461292505264,
"num_tokens": 18528150.0,
"step": 6635
},
{
"entropy": 5.296985101699829,
"epoch": 0.790946992257296,
"grad_norm": 0.984375,
"learning_rate": 0.0004948869899586467,
"loss": 5.0879,
"mean_token_accuracy": 0.18660400807857513,
"num_tokens": 18542402.0,
"step": 6640
},
{
"entropy": 5.293434000015258,
"epoch": 0.7915425848719476,
"grad_norm": 1.0546875,
"learning_rate": 0.0004948779532853552,
"loss": 5.0606,
"mean_token_accuracy": 0.19054797738790513,
"num_tokens": 18557156.0,
"step": 6645
},
{
"entropy": 5.182132244110107,
"epoch": 0.7921381774865992,
"grad_norm": 1.0625,
"learning_rate": 0.0004948689087253989,
"loss": 4.912,
"mean_token_accuracy": 0.20110061317682265,
"num_tokens": 18570544.0,
"step": 6650
},
{
"entropy": 5.248407697677612,
"epoch": 0.7927337701012508,
"grad_norm": 0.984375,
"learning_rate": 0.0004948598562791025,
"loss": 5.0404,
"mean_token_accuracy": 0.19712057411670686,
"num_tokens": 18583755.0,
"step": 6655
},
{
"entropy": 5.249928951263428,
"epoch": 0.7933293627159024,
"grad_norm": 0.91796875,
"learning_rate": 0.0004948507959467905,
"loss": 5.0433,
"mean_token_accuracy": 0.19544795155525208,
"num_tokens": 18596991.0,
"step": 6660
},
{
"entropy": 5.243222904205322,
"epoch": 0.7939249553305538,
"grad_norm": 1.0625,
"learning_rate": 0.0004948417277287877,
"loss": 4.9931,
"mean_token_accuracy": 0.19455894231796264,
"num_tokens": 18611206.0,
"step": 6665
},
{
"entropy": 5.3063524723052975,
"epoch": 0.7945205479452054,
"grad_norm": 1.109375,
"learning_rate": 0.0004948326516254197,
"loss": 5.1274,
"mean_token_accuracy": 0.18850287944078445,
"num_tokens": 18626567.0,
"step": 6670
},
{
"entropy": 5.360056638717651,
"epoch": 0.795116140559857,
"grad_norm": 0.9453125,
"learning_rate": 0.000494823567637012,
"loss": 5.0198,
"mean_token_accuracy": 0.18900996148586274,
"num_tokens": 18641515.0,
"step": 6675
},
{
"entropy": 5.259357881546021,
"epoch": 0.7957117331745086,
"grad_norm": 0.95703125,
"learning_rate": 0.0004948144757638902,
"loss": 4.9999,
"mean_token_accuracy": 0.19067494720220565,
"num_tokens": 18656622.0,
"step": 6680
},
{
"entropy": 5.238596343994141,
"epoch": 0.7963073257891602,
"grad_norm": 1.0625,
"learning_rate": 0.0004948053760063806,
"loss": 4.9945,
"mean_token_accuracy": 0.19153920859098433,
"num_tokens": 18669919.0,
"step": 6685
},
{
"entropy": 5.202534580230713,
"epoch": 0.7969029184038118,
"grad_norm": 1.0703125,
"learning_rate": 0.0004947962683648095,
"loss": 4.9481,
"mean_token_accuracy": 0.20641686618328095,
"num_tokens": 18683254.0,
"step": 6690
},
{
"entropy": 5.1866843700408936,
"epoch": 0.7974985110184634,
"grad_norm": 1.0,
"learning_rate": 0.0004947871528395038,
"loss": 4.952,
"mean_token_accuracy": 0.1974340036511421,
"num_tokens": 18697611.0,
"step": 6695
},
{
"entropy": 5.2900083541870115,
"epoch": 0.798094103633115,
"grad_norm": 0.953125,
"learning_rate": 0.00049477802943079,
"loss": 5.0175,
"mean_token_accuracy": 0.19403107464313507,
"num_tokens": 18711363.0,
"step": 6700
},
{
"entropy": 5.270847129821777,
"epoch": 0.7986896962477665,
"grad_norm": 0.9296875,
"learning_rate": 0.0004947688981389958,
"loss": 4.9776,
"mean_token_accuracy": 0.19537118971347808,
"num_tokens": 18725641.0,
"step": 6705
},
{
"entropy": 5.188566064834594,
"epoch": 0.7992852888624181,
"grad_norm": 0.94921875,
"learning_rate": 0.0004947597589644484,
"loss": 5.0109,
"mean_token_accuracy": 0.19411744475364684,
"num_tokens": 18739724.0,
"step": 6710
},
{
"entropy": 5.18310170173645,
"epoch": 0.7998808814770697,
"grad_norm": 0.9375,
"learning_rate": 0.0004947506119074759,
"loss": 4.9283,
"mean_token_accuracy": 0.2007715106010437,
"num_tokens": 18755257.0,
"step": 6715
},
{
"entropy": 5.291159725189209,
"epoch": 0.8004764740917213,
"grad_norm": 1.171875,
"learning_rate": 0.0004947414569684062,
"loss": 4.9806,
"mean_token_accuracy": 0.19430087357759476,
"num_tokens": 18770300.0,
"step": 6720
},
{
"entropy": 5.188164329528808,
"epoch": 0.8010720667063729,
"grad_norm": 1.2890625,
"learning_rate": 0.0004947322941475677,
"loss": 4.8882,
"mean_token_accuracy": 0.20109564810991287,
"num_tokens": 18781939.0,
"step": 6725
},
{
"entropy": 5.195001077651978,
"epoch": 0.8016676593210245,
"grad_norm": 1.0,
"learning_rate": 0.0004947231234452891,
"loss": 4.9984,
"mean_token_accuracy": 0.1954596757888794,
"num_tokens": 18794851.0,
"step": 6730
},
{
"entropy": 5.186333894729614,
"epoch": 0.802263251935676,
"grad_norm": 0.984375,
"learning_rate": 0.0004947139448618994,
"loss": 4.9855,
"mean_token_accuracy": 0.19963230639696122,
"num_tokens": 18810776.0,
"step": 6735
},
{
"entropy": 5.196971416473389,
"epoch": 0.8028588445503275,
"grad_norm": 0.9609375,
"learning_rate": 0.0004947047583977276,
"loss": 4.9452,
"mean_token_accuracy": 0.20185763090848924,
"num_tokens": 18825602.0,
"step": 6740
},
{
"entropy": 5.283857345581055,
"epoch": 0.8034544371649791,
"grad_norm": 1.0625,
"learning_rate": 0.0004946955640531033,
"loss": 5.0422,
"mean_token_accuracy": 0.19473544657230377,
"num_tokens": 18840284.0,
"step": 6745
},
{
"entropy": 5.203998613357544,
"epoch": 0.8040500297796307,
"grad_norm": 1.203125,
"learning_rate": 0.0004946863618283565,
"loss": 4.9676,
"mean_token_accuracy": 0.1951712191104889,
"num_tokens": 18853372.0,
"step": 6750
},
{
"entropy": 5.196728897094727,
"epoch": 0.8046456223942823,
"grad_norm": 1.0234375,
"learning_rate": 0.0004946771517238169,
"loss": 5.0043,
"mean_token_accuracy": 0.1962709456682205,
"num_tokens": 18867320.0,
"step": 6755
},
{
"entropy": 5.258518123626709,
"epoch": 0.8052412150089339,
"grad_norm": 0.98046875,
"learning_rate": 0.0004946679337398151,
"loss": 5.0902,
"mean_token_accuracy": 0.19173054546117782,
"num_tokens": 18881441.0,
"step": 6760
},
{
"entropy": 5.222652912139893,
"epoch": 0.8058368076235854,
"grad_norm": 0.9296875,
"learning_rate": 0.0004946587078766818,
"loss": 4.975,
"mean_token_accuracy": 0.19958893060684205,
"num_tokens": 18895225.0,
"step": 6765
},
{
"entropy": 5.286642169952392,
"epoch": 0.806432400238237,
"grad_norm": 0.9453125,
"learning_rate": 0.0004946494741347476,
"loss": 5.0,
"mean_token_accuracy": 0.1967850774526596,
"num_tokens": 18909834.0,
"step": 6770
},
{
"entropy": 5.210959529876709,
"epoch": 0.8070279928528886,
"grad_norm": 0.99609375,
"learning_rate": 0.0004946402325143439,
"loss": 4.9608,
"mean_token_accuracy": 0.2000841110944748,
"num_tokens": 18922304.0,
"step": 6775
},
{
"entropy": 5.241202688217163,
"epoch": 0.8076235854675402,
"grad_norm": 1.0703125,
"learning_rate": 0.0004946309830158024,
"loss": 5.0545,
"mean_token_accuracy": 0.1970576301217079,
"num_tokens": 18935959.0,
"step": 6780
},
{
"entropy": 5.265333604812622,
"epoch": 0.8082191780821918,
"grad_norm": 1.0078125,
"learning_rate": 0.0004946217256394544,
"loss": 4.9644,
"mean_token_accuracy": 0.1960201933979988,
"num_tokens": 18950468.0,
"step": 6785
},
{
"entropy": 5.186208724975586,
"epoch": 0.8088147706968434,
"grad_norm": 1.0859375,
"learning_rate": 0.0004946124603856322,
"loss": 4.9414,
"mean_token_accuracy": 0.20229204148054122,
"num_tokens": 18966015.0,
"step": 6790
},
{
"entropy": 5.206463718414307,
"epoch": 0.809410363311495,
"grad_norm": 1.046875,
"learning_rate": 0.0004946031872546683,
"loss": 4.9918,
"mean_token_accuracy": 0.1964615225791931,
"num_tokens": 18980689.0,
"step": 6795
},
{
"entropy": 5.2235252380371096,
"epoch": 0.8100059559261465,
"grad_norm": 1.0546875,
"learning_rate": 0.000494593906246895,
"loss": 4.9598,
"mean_token_accuracy": 0.1949807733297348,
"num_tokens": 18994111.0,
"step": 6800
},
{
"entropy": 5.2537970542907715,
"epoch": 0.8106015485407981,
"grad_norm": 1.015625,
"learning_rate": 0.0004945846173626453,
"loss": 4.9899,
"mean_token_accuracy": 0.19343099892139434,
"num_tokens": 19006845.0,
"step": 6805
},
{
"entropy": 5.125173425674438,
"epoch": 0.8111971411554497,
"grad_norm": 0.93359375,
"learning_rate": 0.0004945753206022525,
"loss": 4.9051,
"mean_token_accuracy": 0.2023426592350006,
"num_tokens": 19021182.0,
"step": 6810
},
{
"entropy": 5.117804956436157,
"epoch": 0.8117927337701013,
"grad_norm": 1.0390625,
"learning_rate": 0.0004945660159660499,
"loss": 4.8381,
"mean_token_accuracy": 0.20070830285549163,
"num_tokens": 19034850.0,
"step": 6815
},
{
"entropy": 5.182636260986328,
"epoch": 0.8123883263847528,
"grad_norm": 1.0859375,
"learning_rate": 0.0004945567034543712,
"loss": 4.9807,
"mean_token_accuracy": 0.19463106989860535,
"num_tokens": 19047405.0,
"step": 6820
},
{
"entropy": 5.204124498367309,
"epoch": 0.8129839189994044,
"grad_norm": 0.9140625,
"learning_rate": 0.0004945473830675507,
"loss": 5.0161,
"mean_token_accuracy": 0.19837694019079208,
"num_tokens": 19061475.0,
"step": 6825
},
{
"entropy": 5.262389230728149,
"epoch": 0.813579511614056,
"grad_norm": 1.0703125,
"learning_rate": 0.0004945380548059224,
"loss": 4.9846,
"mean_token_accuracy": 0.19712873548269272,
"num_tokens": 19077312.0,
"step": 6830
},
{
"entropy": 5.308856678009033,
"epoch": 0.8141751042287075,
"grad_norm": 1.109375,
"learning_rate": 0.0004945287186698211,
"loss": 5.0278,
"mean_token_accuracy": 0.1955183669924736,
"num_tokens": 19091066.0,
"step": 6835
},
{
"entropy": 5.210989904403687,
"epoch": 0.8147706968433591,
"grad_norm": 1.078125,
"learning_rate": 0.0004945193746595815,
"loss": 5.0218,
"mean_token_accuracy": 0.19577715396881104,
"num_tokens": 19105081.0,
"step": 6840
},
{
"entropy": 5.23146276473999,
"epoch": 0.8153662894580107,
"grad_norm": 0.9921875,
"learning_rate": 0.0004945100227755389,
"loss": 4.9988,
"mean_token_accuracy": 0.19308142960071564,
"num_tokens": 19119223.0,
"step": 6845
},
{
"entropy": 5.220105600357056,
"epoch": 0.8159618820726623,
"grad_norm": 1.1015625,
"learning_rate": 0.0004945006630180288,
"loss": 4.9799,
"mean_token_accuracy": 0.19839766770601272,
"num_tokens": 19132704.0,
"step": 6850
},
{
"entropy": 5.185937356948853,
"epoch": 0.8165574746873139,
"grad_norm": 0.91015625,
"learning_rate": 0.0004944912953873865,
"loss": 4.9476,
"mean_token_accuracy": 0.20155406147241592,
"num_tokens": 19147882.0,
"step": 6855
},
{
"entropy": 5.21461672782898,
"epoch": 0.8171530673019655,
"grad_norm": 1.15625,
"learning_rate": 0.0004944819198839486,
"loss": 4.9621,
"mean_token_accuracy": 0.198540598154068,
"num_tokens": 19160815.0,
"step": 6860
},
{
"entropy": 5.167899703979492,
"epoch": 0.817748659916617,
"grad_norm": 0.96875,
"learning_rate": 0.000494472536508051,
"loss": 4.9423,
"mean_token_accuracy": 0.19462167471647263,
"num_tokens": 19175654.0,
"step": 6865
},
{
"entropy": 5.15754280090332,
"epoch": 0.8183442525312686,
"grad_norm": 1.125,
"learning_rate": 0.0004944631452600303,
"loss": 4.9813,
"mean_token_accuracy": 0.20107701271772385,
"num_tokens": 19189428.0,
"step": 6870
},
{
"entropy": 5.149784660339355,
"epoch": 0.8189398451459202,
"grad_norm": 1.078125,
"learning_rate": 0.0004944537461402234,
"loss": 4.8332,
"mean_token_accuracy": 0.20679444074630737,
"num_tokens": 19203050.0,
"step": 6875
},
{
"entropy": 5.136848592758179,
"epoch": 0.8195354377605718,
"grad_norm": 1.046875,
"learning_rate": 0.0004944443391489676,
"loss": 4.9056,
"mean_token_accuracy": 0.2084156483411789,
"num_tokens": 19216875.0,
"step": 6880
},
{
"entropy": 5.125460481643676,
"epoch": 0.8201310303752234,
"grad_norm": 0.86328125,
"learning_rate": 0.0004944349242865999,
"loss": 4.8928,
"mean_token_accuracy": 0.203537055850029,
"num_tokens": 19231423.0,
"step": 6885
},
{
"entropy": 5.203351879119873,
"epoch": 0.820726622989875,
"grad_norm": 1.015625,
"learning_rate": 0.0004944255015534585,
"loss": 4.9761,
"mean_token_accuracy": 0.19353565126657485,
"num_tokens": 19245360.0,
"step": 6890
},
{
"entropy": 5.195250415802002,
"epoch": 0.8213222156045265,
"grad_norm": 0.96484375,
"learning_rate": 0.000494416070949881,
"loss": 5.0199,
"mean_token_accuracy": 0.1889205515384674,
"num_tokens": 19261233.0,
"step": 6895
},
{
"entropy": 5.146976089477539,
"epoch": 0.821917808219178,
"grad_norm": 1.03125,
"learning_rate": 0.0004944066324762058,
"loss": 4.8479,
"mean_token_accuracy": 0.20101901441812514,
"num_tokens": 19274656.0,
"step": 6900
},
{
"entropy": 5.201786804199219,
"epoch": 0.8225134008338296,
"grad_norm": 0.99609375,
"learning_rate": 0.0004943971861327713,
"loss": 4.9904,
"mean_token_accuracy": 0.1978381723165512,
"num_tokens": 19290243.0,
"step": 6905
},
{
"entropy": 5.189332056045532,
"epoch": 0.8231089934484812,
"grad_norm": 0.9609375,
"learning_rate": 0.0004943877319199167,
"loss": 4.9377,
"mean_token_accuracy": 0.2020708903670311,
"num_tokens": 19303314.0,
"step": 6910
},
{
"entropy": 5.259398603439331,
"epoch": 0.8237045860631328,
"grad_norm": 0.90625,
"learning_rate": 0.0004943782698379807,
"loss": 5.0166,
"mean_token_accuracy": 0.19135457128286362,
"num_tokens": 19317856.0,
"step": 6915
},
{
"entropy": 5.2593162059783936,
"epoch": 0.8243001786777844,
"grad_norm": 1.015625,
"learning_rate": 0.0004943687998873029,
"loss": 4.9671,
"mean_token_accuracy": 0.20057986676692963,
"num_tokens": 19332876.0,
"step": 6920
},
{
"entropy": 5.269564008712768,
"epoch": 0.824895771292436,
"grad_norm": 0.93359375,
"learning_rate": 0.0004943593220682229,
"loss": 4.9687,
"mean_token_accuracy": 0.19782715439796447,
"num_tokens": 19345938.0,
"step": 6925
},
{
"entropy": 5.16669454574585,
"epoch": 0.8254913639070875,
"grad_norm": 0.9765625,
"learning_rate": 0.0004943498363810807,
"loss": 4.9259,
"mean_token_accuracy": 0.1993590772151947,
"num_tokens": 19359439.0,
"step": 6930
},
{
"entropy": 5.185785484313965,
"epoch": 0.8260869565217391,
"grad_norm": 1.046875,
"learning_rate": 0.0004943403428262164,
"loss": 4.9432,
"mean_token_accuracy": 0.20048972368240356,
"num_tokens": 19374440.0,
"step": 6935
},
{
"entropy": 5.220820379257202,
"epoch": 0.8266825491363907,
"grad_norm": 1.0,
"learning_rate": 0.0004943308414039708,
"loss": 4.9921,
"mean_token_accuracy": 0.19697284549474717,
"num_tokens": 19386937.0,
"step": 6940
},
{
"entropy": 5.251041412353516,
"epoch": 0.8272781417510423,
"grad_norm": 0.93359375,
"learning_rate": 0.0004943213321146843,
"loss": 4.97,
"mean_token_accuracy": 0.1977197676897049,
"num_tokens": 19401077.0,
"step": 6945
},
{
"entropy": 5.271163606643677,
"epoch": 0.8278737343656939,
"grad_norm": 0.91015625,
"learning_rate": 0.0004943118149586984,
"loss": 5.0927,
"mean_token_accuracy": 0.19365117102861404,
"num_tokens": 19416093.0,
"step": 6950
},
{
"entropy": 5.265637969970703,
"epoch": 0.8284693269803455,
"grad_norm": 1.0625,
"learning_rate": 0.0004943022899363542,
"loss": 5.0127,
"mean_token_accuracy": 0.19765622466802596,
"num_tokens": 19428961.0,
"step": 6955
},
{
"entropy": 5.228713846206665,
"epoch": 0.8290649195949971,
"grad_norm": 1.0546875,
"learning_rate": 0.0004942927570479934,
"loss": 4.9418,
"mean_token_accuracy": 0.19572076797485352,
"num_tokens": 19442909.0,
"step": 6960
},
{
"entropy": 5.088505029678345,
"epoch": 0.8296605122096486,
"grad_norm": 1.0234375,
"learning_rate": 0.0004942832162939581,
"loss": 4.8457,
"mean_token_accuracy": 0.20685051679611205,
"num_tokens": 19457702.0,
"step": 6965
},
{
"entropy": 5.188463258743286,
"epoch": 0.8302561048243002,
"grad_norm": 0.9765625,
"learning_rate": 0.0004942736676745902,
"loss": 4.9811,
"mean_token_accuracy": 0.1929618775844574,
"num_tokens": 19471662.0,
"step": 6970
},
{
"entropy": 5.207181215286255,
"epoch": 0.8308516974389517,
"grad_norm": 1.0390625,
"learning_rate": 0.0004942641111902324,
"loss": 4.9349,
"mean_token_accuracy": 0.20619865655899047,
"num_tokens": 19486109.0,
"step": 6975
},
{
"entropy": 5.1717713356018065,
"epoch": 0.8314472900536033,
"grad_norm": 0.97265625,
"learning_rate": 0.0004942545468412275,
"loss": 4.975,
"mean_token_accuracy": 0.19500770419836044,
"num_tokens": 19501600.0,
"step": 6980
},
{
"entropy": 5.193951988220215,
"epoch": 0.8320428826682549,
"grad_norm": 0.94140625,
"learning_rate": 0.0004942449746279184,
"loss": 4.9919,
"mean_token_accuracy": 0.19565050154924393,
"num_tokens": 19515721.0,
"step": 6985
},
{
"entropy": 5.276266241073609,
"epoch": 0.8326384752829065,
"grad_norm": 1.03125,
"learning_rate": 0.0004942353945506485,
"loss": 4.988,
"mean_token_accuracy": 0.19881187975406647,
"num_tokens": 19530536.0,
"step": 6990
},
{
"entropy": 5.214112663269043,
"epoch": 0.833234067897558,
"grad_norm": 1.0703125,
"learning_rate": 0.0004942258066097617,
"loss": 4.973,
"mean_token_accuracy": 0.2029110684990883,
"num_tokens": 19542976.0,
"step": 6995
},
{
"entropy": 5.165704584121704,
"epoch": 0.8338296605122096,
"grad_norm": 0.9296875,
"learning_rate": 0.0004942162108056015,
"loss": 4.8911,
"mean_token_accuracy": 0.20454884320497513,
"num_tokens": 19557884.0,
"step": 7000
},
{
"entropy": 5.174490022659302,
"epoch": 0.8344252531268612,
"grad_norm": 0.94921875,
"learning_rate": 0.0004942066071385122,
"loss": 4.9147,
"mean_token_accuracy": 0.2007325679063797,
"num_tokens": 19572438.0,
"step": 7005
},
{
"entropy": 5.256436538696289,
"epoch": 0.8350208457415128,
"grad_norm": 0.97265625,
"learning_rate": 0.0004941969956088383,
"loss": 5.0454,
"mean_token_accuracy": 0.19116324335336685,
"num_tokens": 19586114.0,
"step": 7010
},
{
"entropy": 5.110114574432373,
"epoch": 0.8356164383561644,
"grad_norm": 0.9921875,
"learning_rate": 0.0004941873762169246,
"loss": 4.8748,
"mean_token_accuracy": 0.20300572961568833,
"num_tokens": 19600338.0,
"step": 7015
},
{
"entropy": 5.251081228256226,
"epoch": 0.836212030970816,
"grad_norm": 0.92578125,
"learning_rate": 0.0004941777489631161,
"loss": 5.0427,
"mean_token_accuracy": 0.19273286461830139,
"num_tokens": 19613049.0,
"step": 7020
},
{
"entropy": 5.185570621490479,
"epoch": 0.8368076235854676,
"grad_norm": 1.0625,
"learning_rate": 0.0004941681138477582,
"loss": 4.9107,
"mean_token_accuracy": 0.2034965544939041,
"num_tokens": 19626685.0,
"step": 7025
},
{
"entropy": 5.2014172077178955,
"epoch": 0.8374032162001191,
"grad_norm": 1.03125,
"learning_rate": 0.0004941584708711962,
"loss": 5.0114,
"mean_token_accuracy": 0.20236924439668655,
"num_tokens": 19642114.0,
"step": 7030
},
{
"entropy": 5.2741310596466064,
"epoch": 0.8379988088147707,
"grad_norm": 0.984375,
"learning_rate": 0.0004941488200337763,
"loss": 4.9549,
"mean_token_accuracy": 0.20340208411216737,
"num_tokens": 19655505.0,
"step": 7035
},
{
"entropy": 5.193226099014282,
"epoch": 0.8385944014294223,
"grad_norm": 0.8984375,
"learning_rate": 0.0004941391613358444,
"loss": 4.9534,
"mean_token_accuracy": 0.19352470636367797,
"num_tokens": 19671845.0,
"step": 7040
},
{
"entropy": 5.176735305786133,
"epoch": 0.8391899940440739,
"grad_norm": 1.140625,
"learning_rate": 0.0004941294947777473,
"loss": 4.9154,
"mean_token_accuracy": 0.20528090447187425,
"num_tokens": 19683578.0,
"step": 7045
},
{
"entropy": 5.240229845046997,
"epoch": 0.8397855866587254,
"grad_norm": 1.015625,
"learning_rate": 0.0004941198203598314,
"loss": 4.9745,
"mean_token_accuracy": 0.19976136833429337,
"num_tokens": 19697974.0,
"step": 7050
},
{
"entropy": 5.193073081970215,
"epoch": 0.840381179273377,
"grad_norm": 1.0,
"learning_rate": 0.0004941101380824438,
"loss": 4.923,
"mean_token_accuracy": 0.2063810333609581,
"num_tokens": 19713024.0,
"step": 7055
},
{
"entropy": 5.250869655609131,
"epoch": 0.8409767718880286,
"grad_norm": 0.98046875,
"learning_rate": 0.0004941004479459318,
"loss": 5.0572,
"mean_token_accuracy": 0.18808308839797974,
"num_tokens": 19726562.0,
"step": 7060
},
{
"entropy": 5.196570301055909,
"epoch": 0.8415723645026801,
"grad_norm": 1.015625,
"learning_rate": 0.000494090749950643,
"loss": 4.964,
"mean_token_accuracy": 0.2032913699746132,
"num_tokens": 19741838.0,
"step": 7065
},
{
"entropy": 5.210614109039307,
"epoch": 0.8421679571173317,
"grad_norm": 0.9765625,
"learning_rate": 0.0004940810440969253,
"loss": 4.896,
"mean_token_accuracy": 0.19946436136960982,
"num_tokens": 19754852.0,
"step": 7070
},
{
"entropy": 5.138875532150268,
"epoch": 0.8427635497319833,
"grad_norm": 1.0390625,
"learning_rate": 0.0004940713303851265,
"loss": 4.7758,
"mean_token_accuracy": 0.21203377842903137,
"num_tokens": 19767355.0,
"step": 7075
},
{
"entropy": 5.1182904720306395,
"epoch": 0.8433591423466349,
"grad_norm": 1.1796875,
"learning_rate": 0.0004940616088155954,
"loss": 4.88,
"mean_token_accuracy": 0.20134452283382415,
"num_tokens": 19780861.0,
"step": 7080
},
{
"entropy": 5.1602380752563475,
"epoch": 0.8439547349612865,
"grad_norm": 1.0234375,
"learning_rate": 0.0004940518793886807,
"loss": 4.9618,
"mean_token_accuracy": 0.201132895052433,
"num_tokens": 19793802.0,
"step": 7085
},
{
"entropy": 5.294936943054199,
"epoch": 0.8445503275759381,
"grad_norm": 0.91015625,
"learning_rate": 0.0004940421421047311,
"loss": 5.0221,
"mean_token_accuracy": 0.1864053338766098,
"num_tokens": 19810748.0,
"step": 7090
},
{
"entropy": 5.205626535415649,
"epoch": 0.8451459201905897,
"grad_norm": 0.96875,
"learning_rate": 0.000494032396964096,
"loss": 4.9497,
"mean_token_accuracy": 0.19336152970790862,
"num_tokens": 19826046.0,
"step": 7095
},
{
"entropy": 5.165968990325927,
"epoch": 0.8457415128052412,
"grad_norm": 1.0390625,
"learning_rate": 0.000494022643967125,
"loss": 4.935,
"mean_token_accuracy": 0.1955157309770584,
"num_tokens": 19839591.0,
"step": 7100
},
{
"entropy": 5.222088193893432,
"epoch": 0.8463371054198928,
"grad_norm": 1.0,
"learning_rate": 0.0004940128831141679,
"loss": 4.9429,
"mean_token_accuracy": 0.20535173267126083,
"num_tokens": 19852560.0,
"step": 7105
},
{
"entropy": 5.212052345275879,
"epoch": 0.8469326980345444,
"grad_norm": 1.03125,
"learning_rate": 0.0004940031144055747,
"loss": 5.0162,
"mean_token_accuracy": 0.19861605614423752,
"num_tokens": 19867663.0,
"step": 7110
},
{
"entropy": 5.139697313308716,
"epoch": 0.847528290649196,
"grad_norm": 0.89453125,
"learning_rate": 0.0004939933378416959,
"loss": 4.9712,
"mean_token_accuracy": 0.19581947326660157,
"num_tokens": 19883761.0,
"step": 7115
},
{
"entropy": 5.2352766513824465,
"epoch": 0.8481238832638476,
"grad_norm": 1.015625,
"learning_rate": 0.0004939835534228821,
"loss": 4.9709,
"mean_token_accuracy": 0.1951030895113945,
"num_tokens": 19897205.0,
"step": 7120
},
{
"entropy": 5.301513385772705,
"epoch": 0.848719475878499,
"grad_norm": 1.078125,
"learning_rate": 0.0004939737611494842,
"loss": 5.0611,
"mean_token_accuracy": 0.19240733832120896,
"num_tokens": 19912563.0,
"step": 7125
},
{
"entropy": 5.217139387130738,
"epoch": 0.8493150684931506,
"grad_norm": 0.98046875,
"learning_rate": 0.0004939639610218536,
"loss": 4.9561,
"mean_token_accuracy": 0.19408978074789046,
"num_tokens": 19926308.0,
"step": 7130
},
{
"entropy": 5.181148433685303,
"epoch": 0.8499106611078022,
"grad_norm": 1.0703125,
"learning_rate": 0.0004939541530403419,
"loss": 4.9209,
"mean_token_accuracy": 0.20192811638116837,
"num_tokens": 19939372.0,
"step": 7135
},
{
"entropy": 5.255271673202515,
"epoch": 0.8505062537224538,
"grad_norm": 0.98046875,
"learning_rate": 0.0004939443372053004,
"loss": 5.0097,
"mean_token_accuracy": 0.1948626905679703,
"num_tokens": 19952755.0,
"step": 7140
},
{
"entropy": 5.204237699508667,
"epoch": 0.8511018463371054,
"grad_norm": 1.0234375,
"learning_rate": 0.0004939345135170818,
"loss": 4.9571,
"mean_token_accuracy": 0.20370981395244597,
"num_tokens": 19965678.0,
"step": 7145
},
{
"entropy": 5.196007013320923,
"epoch": 0.851697438951757,
"grad_norm": 0.98046875,
"learning_rate": 0.000493924681976038,
"loss": 4.9686,
"mean_token_accuracy": 0.20286183506250383,
"num_tokens": 19979022.0,
"step": 7150
},
{
"entropy": 5.195898962020874,
"epoch": 0.8522930315664086,
"grad_norm": 1.1953125,
"learning_rate": 0.0004939148425825219,
"loss": 4.8899,
"mean_token_accuracy": 0.2037184089422226,
"num_tokens": 19991861.0,
"step": 7155
},
{
"entropy": 5.174033832550049,
"epoch": 0.8528886241810602,
"grad_norm": 1.0234375,
"learning_rate": 0.0004939049953368863,
"loss": 4.9192,
"mean_token_accuracy": 0.20359836965799333,
"num_tokens": 20004444.0,
"step": 7160
},
{
"entropy": 5.122095441818237,
"epoch": 0.8534842167957117,
"grad_norm": 1.0,
"learning_rate": 0.0004938951402394844,
"loss": 4.8955,
"mean_token_accuracy": 0.2082478627562523,
"num_tokens": 20018302.0,
"step": 7165
},
{
"entropy": 5.166162919998169,
"epoch": 0.8540798094103633,
"grad_norm": 1.0546875,
"learning_rate": 0.0004938852772906697,
"loss": 4.8627,
"mean_token_accuracy": 0.20636003762483596,
"num_tokens": 20031193.0,
"step": 7170
},
{
"entropy": 5.187306261062622,
"epoch": 0.8546754020250149,
"grad_norm": 0.97265625,
"learning_rate": 0.0004938754064907961,
"loss": 4.963,
"mean_token_accuracy": 0.20344823449850083,
"num_tokens": 20044883.0,
"step": 7175
},
{
"entropy": 5.210363483428955,
"epoch": 0.8552709946396665,
"grad_norm": 1.015625,
"learning_rate": 0.0004938655278402175,
"loss": 5.0217,
"mean_token_accuracy": 0.19423196762800216,
"num_tokens": 20058024.0,
"step": 7180
},
{
"entropy": 5.286203908920288,
"epoch": 0.8558665872543181,
"grad_norm": 0.96875,
"learning_rate": 0.0004938556413392882,
"loss": 5.0351,
"mean_token_accuracy": 0.19231879562139512,
"num_tokens": 20072622.0,
"step": 7185
},
{
"entropy": 5.220455121994019,
"epoch": 0.8564621798689697,
"grad_norm": 0.9765625,
"learning_rate": 0.0004938457469883629,
"loss": 4.9736,
"mean_token_accuracy": 0.2021484375,
"num_tokens": 20086774.0,
"step": 7190
},
{
"entropy": 5.23595609664917,
"epoch": 0.8570577724836212,
"grad_norm": 1.046875,
"learning_rate": 0.0004938358447877966,
"loss": 5.0317,
"mean_token_accuracy": 0.1910096973180771,
"num_tokens": 20102435.0,
"step": 7195
},
{
"entropy": 5.262790393829346,
"epoch": 0.8576533650982728,
"grad_norm": 1.0390625,
"learning_rate": 0.0004938259347379442,
"loss": 4.9782,
"mean_token_accuracy": 0.19890339374542237,
"num_tokens": 20115860.0,
"step": 7200
},
{
"entropy": 5.163473892211914,
"epoch": 0.8582489577129243,
"grad_norm": 0.92578125,
"learning_rate": 0.0004938160168391614,
"loss": 4.9283,
"mean_token_accuracy": 0.20654567778110505,
"num_tokens": 20130910.0,
"step": 7205
},
{
"entropy": 5.153239011764526,
"epoch": 0.8588445503275759,
"grad_norm": 0.98046875,
"learning_rate": 0.0004938060910918038,
"loss": 4.8533,
"mean_token_accuracy": 0.2060517743229866,
"num_tokens": 20143093.0,
"step": 7210
},
{
"entropy": 5.166943025588989,
"epoch": 0.8594401429422275,
"grad_norm": 1.046875,
"learning_rate": 0.0004937961574962274,
"loss": 4.9352,
"mean_token_accuracy": 0.19775982499122619,
"num_tokens": 20157637.0,
"step": 7215
},
{
"entropy": 5.244236612319947,
"epoch": 0.8600357355568791,
"grad_norm": 1.1328125,
"learning_rate": 0.0004937862160527887,
"loss": 5.0331,
"mean_token_accuracy": 0.1976795718073845,
"num_tokens": 20170455.0,
"step": 7220
},
{
"entropy": 5.194699239730835,
"epoch": 0.8606313281715307,
"grad_norm": 0.9921875,
"learning_rate": 0.0004937762667618441,
"loss": 4.869,
"mean_token_accuracy": 0.20774535536766053,
"num_tokens": 20186452.0,
"step": 7225
},
{
"entropy": 5.217760324478149,
"epoch": 0.8612269207861822,
"grad_norm": 0.984375,
"learning_rate": 0.0004937663096237505,
"loss": 4.9636,
"mean_token_accuracy": 0.20097594410181047,
"num_tokens": 20199754.0,
"step": 7230
},
{
"entropy": 5.2022617816925045,
"epoch": 0.8618225134008338,
"grad_norm": 0.96484375,
"learning_rate": 0.0004937563446388652,
"loss": 5.018,
"mean_token_accuracy": 0.19501520842313766,
"num_tokens": 20215765.0,
"step": 7235
},
{
"entropy": 5.2486100673675535,
"epoch": 0.8624181060154854,
"grad_norm": 0.93359375,
"learning_rate": 0.0004937463718075453,
"loss": 5.0237,
"mean_token_accuracy": 0.19340973496437072,
"num_tokens": 20229225.0,
"step": 7240
},
{
"entropy": 5.168346500396728,
"epoch": 0.863013698630137,
"grad_norm": 0.96484375,
"learning_rate": 0.0004937363911301488,
"loss": 4.9418,
"mean_token_accuracy": 0.20625072568655015,
"num_tokens": 20242729.0,
"step": 7245
},
{
"entropy": 5.222889566421509,
"epoch": 0.8636092912447886,
"grad_norm": 1.09375,
"learning_rate": 0.0004937264026070336,
"loss": 4.9161,
"mean_token_accuracy": 0.19941461235284805,
"num_tokens": 20252914.0,
"step": 7250
},
{
"entropy": 5.102902412414551,
"epoch": 0.8642048838594402,
"grad_norm": 1.078125,
"learning_rate": 0.000493716406238558,
"loss": 4.8384,
"mean_token_accuracy": 0.2079011857509613,
"num_tokens": 20265611.0,
"step": 7255
},
{
"entropy": 5.235066890716553,
"epoch": 0.8648004764740918,
"grad_norm": 1.046875,
"learning_rate": 0.0004937064020250804,
"loss": 5.0386,
"mean_token_accuracy": 0.19073348343372346,
"num_tokens": 20279808.0,
"step": 7260
},
{
"entropy": 5.162436008453369,
"epoch": 0.8653960690887433,
"grad_norm": 0.91796875,
"learning_rate": 0.0004936963899669599,
"loss": 4.9113,
"mean_token_accuracy": 0.20909116119146348,
"num_tokens": 20295028.0,
"step": 7265
},
{
"entropy": 5.196700143814087,
"epoch": 0.8659916617033949,
"grad_norm": 0.98046875,
"learning_rate": 0.0004936863700645554,
"loss": 4.9062,
"mean_token_accuracy": 0.20599477589130402,
"num_tokens": 20308780.0,
"step": 7270
},
{
"entropy": 5.295598888397217,
"epoch": 0.8665872543180465,
"grad_norm": 1.09375,
"learning_rate": 0.0004936763423182264,
"loss": 5.0662,
"mean_token_accuracy": 0.19535856693983078,
"num_tokens": 20322513.0,
"step": 7275
},
{
"entropy": 5.296857118606567,
"epoch": 0.867182846932698,
"grad_norm": 1.015625,
"learning_rate": 0.0004936663067283325,
"loss": 5.0559,
"mean_token_accuracy": 0.19072242081165314,
"num_tokens": 20336457.0,
"step": 7280
},
{
"entropy": 5.248151874542236,
"epoch": 0.8677784395473496,
"grad_norm": 1.125,
"learning_rate": 0.0004936562632952337,
"loss": 4.9359,
"mean_token_accuracy": 0.20287286937236787,
"num_tokens": 20349063.0,
"step": 7285
},
{
"entropy": 5.242201280593872,
"epoch": 0.8683740321620012,
"grad_norm": 1.03125,
"learning_rate": 0.0004936462120192903,
"loss": 5.0072,
"mean_token_accuracy": 0.1941452980041504,
"num_tokens": 20361832.0,
"step": 7290
},
{
"entropy": 5.095669746398926,
"epoch": 0.8689696247766527,
"grad_norm": 1.0078125,
"learning_rate": 0.0004936361529008627,
"loss": 4.891,
"mean_token_accuracy": 0.202139450609684,
"num_tokens": 20376291.0,
"step": 7295
},
{
"entropy": 5.122959470748901,
"epoch": 0.8695652173913043,
"grad_norm": 1.0078125,
"learning_rate": 0.0004936260859403119,
"loss": 4.8837,
"mean_token_accuracy": 0.20639112442731858,
"num_tokens": 20389654.0,
"step": 7300
},
{
"entropy": 5.277730083465576,
"epoch": 0.8701608100059559,
"grad_norm": 1.0,
"learning_rate": 0.0004936160111379987,
"loss": 5.0127,
"mean_token_accuracy": 0.195535808801651,
"num_tokens": 20403782.0,
"step": 7305
},
{
"entropy": 5.217010927200318,
"epoch": 0.8707564026206075,
"grad_norm": 1.1171875,
"learning_rate": 0.0004936059284942848,
"loss": 4.8706,
"mean_token_accuracy": 0.2092948630452156,
"num_tokens": 20416547.0,
"step": 7310
},
{
"entropy": 5.226112174987793,
"epoch": 0.8713519952352591,
"grad_norm": 1.015625,
"learning_rate": 0.0004935958380095316,
"loss": 5.0595,
"mean_token_accuracy": 0.19023230224847792,
"num_tokens": 20430831.0,
"step": 7315
},
{
"entropy": 5.1732405662536625,
"epoch": 0.8719475878499107,
"grad_norm": 0.94921875,
"learning_rate": 0.000493585739684101,
"loss": 4.9239,
"mean_token_accuracy": 0.20209971815347672,
"num_tokens": 20445905.0,
"step": 7320
},
{
"entropy": 5.2477555751800535,
"epoch": 0.8725431804645623,
"grad_norm": 1.0546875,
"learning_rate": 0.0004935756335183554,
"loss": 5.0575,
"mean_token_accuracy": 0.19383396208286285,
"num_tokens": 20461506.0,
"step": 7325
},
{
"entropy": 5.251984643936157,
"epoch": 0.8731387730792138,
"grad_norm": 1.078125,
"learning_rate": 0.0004935655195126572,
"loss": 4.9945,
"mean_token_accuracy": 0.20426169633865357,
"num_tokens": 20474753.0,
"step": 7330
},
{
"entropy": 5.227786016464234,
"epoch": 0.8737343656938654,
"grad_norm": 0.99609375,
"learning_rate": 0.0004935553976673692,
"loss": 4.9701,
"mean_token_accuracy": 0.20384037494659424,
"num_tokens": 20489877.0,
"step": 7335
},
{
"entropy": 5.145942354202271,
"epoch": 0.874329958308517,
"grad_norm": 1.1484375,
"learning_rate": 0.0004935452679828545,
"loss": 4.8906,
"mean_token_accuracy": 0.20552892088890076,
"num_tokens": 20503693.0,
"step": 7340
},
{
"entropy": 5.1959357261657715,
"epoch": 0.8749255509231686,
"grad_norm": 0.9765625,
"learning_rate": 0.0004935351304594763,
"loss": 4.9652,
"mean_token_accuracy": 0.1953132003545761,
"num_tokens": 20519188.0,
"step": 7345
},
{
"entropy": 5.2345884323120115,
"epoch": 0.8755211435378202,
"grad_norm": 0.91015625,
"learning_rate": 0.0004935249850975983,
"loss": 4.9974,
"mean_token_accuracy": 0.19142743647098542,
"num_tokens": 20535218.0,
"step": 7350
},
{
"entropy": 5.169121217727661,
"epoch": 0.8761167361524718,
"grad_norm": 1.0390625,
"learning_rate": 0.0004935148318975843,
"loss": 4.9181,
"mean_token_accuracy": 0.21117680817842482,
"num_tokens": 20548037.0,
"step": 7355
},
{
"entropy": 5.18068561553955,
"epoch": 0.8767123287671232,
"grad_norm": 0.9609375,
"learning_rate": 0.0004935046708597986,
"loss": 4.968,
"mean_token_accuracy": 0.19743893295526505,
"num_tokens": 20562248.0,
"step": 7360
},
{
"entropy": 5.222414541244507,
"epoch": 0.8773079213817748,
"grad_norm": 0.94140625,
"learning_rate": 0.0004934945019846057,
"loss": 5.0138,
"mean_token_accuracy": 0.1944548338651657,
"num_tokens": 20577796.0,
"step": 7365
},
{
"entropy": 5.199978828430176,
"epoch": 0.8779035139964264,
"grad_norm": 0.98828125,
"learning_rate": 0.0004934843252723702,
"loss": 4.946,
"mean_token_accuracy": 0.20352555364370345,
"num_tokens": 20591413.0,
"step": 7370
},
{
"entropy": 5.189595127105713,
"epoch": 0.878499106611078,
"grad_norm": 0.9375,
"learning_rate": 0.0004934741407234572,
"loss": 4.9603,
"mean_token_accuracy": 0.1987425595521927,
"num_tokens": 20606648.0,
"step": 7375
},
{
"entropy": 5.291537809371948,
"epoch": 0.8790946992257296,
"grad_norm": 1.0390625,
"learning_rate": 0.000493463948338232,
"loss": 5.0659,
"mean_token_accuracy": 0.19581476747989654,
"num_tokens": 20620531.0,
"step": 7380
},
{
"entropy": 5.22481861114502,
"epoch": 0.8796902918403812,
"grad_norm": 1.046875,
"learning_rate": 0.0004934537481170602,
"loss": 4.9598,
"mean_token_accuracy": 0.19594647884368896,
"num_tokens": 20636639.0,
"step": 7385
},
{
"entropy": 5.165452384948731,
"epoch": 0.8802858844550328,
"grad_norm": 1.0625,
"learning_rate": 0.0004934435400603077,
"loss": 4.964,
"mean_token_accuracy": 0.1959933340549469,
"num_tokens": 20648799.0,
"step": 7390
},
{
"entropy": 5.130779838562011,
"epoch": 0.8808814770696843,
"grad_norm": 0.97265625,
"learning_rate": 0.0004934333241683404,
"loss": 4.8914,
"mean_token_accuracy": 0.20729562789201736,
"num_tokens": 20662612.0,
"step": 7395
},
{
"entropy": 5.212938022613526,
"epoch": 0.8814770696843359,
"grad_norm": 1.015625,
"learning_rate": 0.0004934231004415251,
"loss": 4.9427,
"mean_token_accuracy": 0.20575273931026458,
"num_tokens": 20676382.0,
"step": 7400
},
{
"entropy": 5.165149307250976,
"epoch": 0.8820726622989875,
"grad_norm": 0.890625,
"learning_rate": 0.0004934128688802282,
"loss": 4.9308,
"mean_token_accuracy": 0.19968362897634506,
"num_tokens": 20691659.0,
"step": 7405
},
{
"entropy": 5.195574951171875,
"epoch": 0.8826682549136391,
"grad_norm": 1.015625,
"learning_rate": 0.0004934026294848168,
"loss": 4.9536,
"mean_token_accuracy": 0.20279901921749116,
"num_tokens": 20705276.0,
"step": 7410
},
{
"entropy": 5.24741849899292,
"epoch": 0.8832638475282907,
"grad_norm": 1.09375,
"learning_rate": 0.0004933923822556583,
"loss": 4.9093,
"mean_token_accuracy": 0.19772066473960875,
"num_tokens": 20719632.0,
"step": 7415
},
{
"entropy": 5.187991380691528,
"epoch": 0.8838594401429423,
"grad_norm": 1.0625,
"learning_rate": 0.00049338212719312,
"loss": 5.0214,
"mean_token_accuracy": 0.19982338696718216,
"num_tokens": 20733190.0,
"step": 7420
},
{
"entropy": 5.218126058578491,
"epoch": 0.8844550327575939,
"grad_norm": 0.9375,
"learning_rate": 0.00049337186429757,
"loss": 5.074,
"mean_token_accuracy": 0.19157612770795823,
"num_tokens": 20749019.0,
"step": 7425
},
{
"entropy": 5.265118598937988,
"epoch": 0.8850506253722454,
"grad_norm": 1.0234375,
"learning_rate": 0.0004933615935693762,
"loss": 4.908,
"mean_token_accuracy": 0.1962705910205841,
"num_tokens": 20762731.0,
"step": 7430
},
{
"entropy": 5.1789146900177006,
"epoch": 0.8856462179868969,
"grad_norm": 1.0078125,
"learning_rate": 0.0004933513150089069,
"loss": 4.9609,
"mean_token_accuracy": 0.19758737236261367,
"num_tokens": 20775599.0,
"step": 7435
},
{
"entropy": 5.152475214004516,
"epoch": 0.8862418106015485,
"grad_norm": 0.984375,
"learning_rate": 0.0004933410286165311,
"loss": 4.8694,
"mean_token_accuracy": 0.21106995791196823,
"num_tokens": 20788945.0,
"step": 7440
},
{
"entropy": 5.136826848983764,
"epoch": 0.8868374032162001,
"grad_norm": 0.97265625,
"learning_rate": 0.0004933307343926177,
"loss": 4.8434,
"mean_token_accuracy": 0.20298906415700912,
"num_tokens": 20803079.0,
"step": 7445
},
{
"entropy": 5.167130470275879,
"epoch": 0.8874329958308517,
"grad_norm": 0.9609375,
"learning_rate": 0.0004933204323375358,
"loss": 4.8527,
"mean_token_accuracy": 0.20021582990884781,
"num_tokens": 20816321.0,
"step": 7450
},
{
"entropy": 5.127469205856324,
"epoch": 0.8880285884455033,
"grad_norm": 0.98828125,
"learning_rate": 0.000493310122451655,
"loss": 4.8899,
"mean_token_accuracy": 0.20435265749692916,
"num_tokens": 20830095.0,
"step": 7455
},
{
"entropy": 5.181621456146241,
"epoch": 0.8886241810601548,
"grad_norm": 1.0546875,
"learning_rate": 0.0004932998047353449,
"loss": 4.917,
"mean_token_accuracy": 0.2052110731601715,
"num_tokens": 20845022.0,
"step": 7460
},
{
"entropy": 5.201427459716797,
"epoch": 0.8892197736748064,
"grad_norm": 1.109375,
"learning_rate": 0.0004932894791889759,
"loss": 4.9526,
"mean_token_accuracy": 0.20713700503110885,
"num_tokens": 20857851.0,
"step": 7465
},
{
"entropy": 5.2019579887390135,
"epoch": 0.889815366289458,
"grad_norm": 1.1171875,
"learning_rate": 0.0004932791458129181,
"loss": 4.8849,
"mean_token_accuracy": 0.2078794479370117,
"num_tokens": 20870343.0,
"step": 7470
},
{
"entropy": 5.254610681533814,
"epoch": 0.8904109589041096,
"grad_norm": 0.96875,
"learning_rate": 0.0004932688046075422,
"loss": 5.0336,
"mean_token_accuracy": 0.19293030500411987,
"num_tokens": 20883805.0,
"step": 7475
},
{
"entropy": 5.210497713088989,
"epoch": 0.8910065515187612,
"grad_norm": 1.0625,
"learning_rate": 0.0004932584555732192,
"loss": 5.0198,
"mean_token_accuracy": 0.1963283509016037,
"num_tokens": 20898160.0,
"step": 7480
},
{
"entropy": 5.18649525642395,
"epoch": 0.8916021441334128,
"grad_norm": 1.03125,
"learning_rate": 0.0004932480987103204,
"loss": 4.9286,
"mean_token_accuracy": 0.2024695724248886,
"num_tokens": 20912528.0,
"step": 7485
},
{
"entropy": 5.283925771713257,
"epoch": 0.8921977367480644,
"grad_norm": 1.0,
"learning_rate": 0.0004932377340192169,
"loss": 5.0067,
"mean_token_accuracy": 0.19672355353832244,
"num_tokens": 20925603.0,
"step": 7490
},
{
"entropy": 5.133279752731323,
"epoch": 0.8927933293627159,
"grad_norm": 1.046875,
"learning_rate": 0.0004932273615002809,
"loss": 4.8645,
"mean_token_accuracy": 0.20261523723602295,
"num_tokens": 20937319.0,
"step": 7495
},
{
"entropy": 5.103966617584229,
"epoch": 0.8933889219773675,
"grad_norm": 0.9921875,
"learning_rate": 0.0004932169811538843,
"loss": 4.8953,
"mean_token_accuracy": 0.20095157474279404,
"num_tokens": 20952963.0,
"step": 7500
},
{
"entropy": 5.243526983261108,
"epoch": 0.8939845145920191,
"grad_norm": 1.046875,
"learning_rate": 0.0004932065929803993,
"loss": 4.9478,
"mean_token_accuracy": 0.20004228204488755,
"num_tokens": 20966696.0,
"step": 7505
},
{
"entropy": 5.223113584518432,
"epoch": 0.8945801072066706,
"grad_norm": 1.0390625,
"learning_rate": 0.0004931961969801985,
"loss": 4.9369,
"mean_token_accuracy": 0.20149192363023757,
"num_tokens": 20981093.0,
"step": 7510
},
{
"entropy": 5.178506088256836,
"epoch": 0.8951756998213222,
"grad_norm": 0.88671875,
"learning_rate": 0.000493185793153655,
"loss": 5.0071,
"mean_token_accuracy": 0.19638022780418396,
"num_tokens": 20997069.0,
"step": 7515
},
{
"entropy": 5.187963151931763,
"epoch": 0.8957712924359738,
"grad_norm": 1.09375,
"learning_rate": 0.0004931753815011419,
"loss": 4.8652,
"mean_token_accuracy": 0.21015918552875518,
"num_tokens": 21010123.0,
"step": 7520
},
{
"entropy": 5.209871625900268,
"epoch": 0.8963668850506253,
"grad_norm": 1.0078125,
"learning_rate": 0.0004931649620230325,
"loss": 5.0137,
"mean_token_accuracy": 0.19659007489681243,
"num_tokens": 21024774.0,
"step": 7525
},
{
"entropy": 5.195480823516846,
"epoch": 0.8969624776652769,
"grad_norm": 1.0546875,
"learning_rate": 0.0004931545347197006,
"loss": 4.9293,
"mean_token_accuracy": 0.19975898861885072,
"num_tokens": 21039533.0,
"step": 7530
},
{
"entropy": 5.180737495422363,
"epoch": 0.8975580702799285,
"grad_norm": 0.984375,
"learning_rate": 0.0004931440995915203,
"loss": 4.9113,
"mean_token_accuracy": 0.20675092339515685,
"num_tokens": 21053593.0,
"step": 7535
},
{
"entropy": 5.215735816955567,
"epoch": 0.8981536628945801,
"grad_norm": 1.0703125,
"learning_rate": 0.0004931336566388658,
"loss": 4.8896,
"mean_token_accuracy": 0.2051610141992569,
"num_tokens": 21064659.0,
"step": 7540
},
{
"entropy": 5.144244050979614,
"epoch": 0.8987492555092317,
"grad_norm": 0.98046875,
"learning_rate": 0.0004931232058621117,
"loss": 4.8302,
"mean_token_accuracy": 0.21257673650979997,
"num_tokens": 21078815.0,
"step": 7545
},
{
"entropy": 5.096023941040039,
"epoch": 0.8993448481238833,
"grad_norm": 0.97265625,
"learning_rate": 0.0004931127472616328,
"loss": 4.8579,
"mean_token_accuracy": 0.20722172409296036,
"num_tokens": 21093415.0,
"step": 7550
},
{
"entropy": 5.229115009307861,
"epoch": 0.8999404407385349,
"grad_norm": 0.94921875,
"learning_rate": 0.0004931022808378044,
"loss": 4.9558,
"mean_token_accuracy": 0.19836414754390716,
"num_tokens": 21108347.0,
"step": 7555
},
{
"entropy": 5.24768762588501,
"epoch": 0.9005360333531864,
"grad_norm": 0.9375,
"learning_rate": 0.0004930918065910018,
"loss": 5.018,
"mean_token_accuracy": 0.1925779700279236,
"num_tokens": 21122647.0,
"step": 7560
},
{
"entropy": 5.195303964614868,
"epoch": 0.901131625967838,
"grad_norm": 1.0234375,
"learning_rate": 0.0004930813245216005,
"loss": 4.8716,
"mean_token_accuracy": 0.20514909774065018,
"num_tokens": 21136465.0,
"step": 7565
},
{
"entropy": 5.2140923023223875,
"epoch": 0.9017272185824896,
"grad_norm": 0.98828125,
"learning_rate": 0.0004930708346299769,
"loss": 5.0167,
"mean_token_accuracy": 0.19479562491178512,
"num_tokens": 21152047.0,
"step": 7570
},
{
"entropy": 5.1643455028533936,
"epoch": 0.9023228111971412,
"grad_norm": 1.03125,
"learning_rate": 0.0004930603369165069,
"loss": 5.0296,
"mean_token_accuracy": 0.20204702466726304,
"num_tokens": 21166355.0,
"step": 7575
},
{
"entropy": 5.184153413772583,
"epoch": 0.9029184038117928,
"grad_norm": 1.0234375,
"learning_rate": 0.0004930498313815672,
"loss": 4.8674,
"mean_token_accuracy": 0.2051374137401581,
"num_tokens": 21180676.0,
"step": 7580
},
{
"entropy": 5.151522541046143,
"epoch": 0.9035139964264444,
"grad_norm": 0.99609375,
"learning_rate": 0.0004930393180255345,
"loss": 4.9393,
"mean_token_accuracy": 0.20314026921987532,
"num_tokens": 21193091.0,
"step": 7585
},
{
"entropy": 5.204026079177856,
"epoch": 0.9041095890410958,
"grad_norm": 0.94140625,
"learning_rate": 0.000493028796848786,
"loss": 4.9595,
"mean_token_accuracy": 0.19744882136583328,
"num_tokens": 21208072.0,
"step": 7590
},
{
"entropy": 5.168215751647949,
"epoch": 0.9047051816557474,
"grad_norm": 1.1328125,
"learning_rate": 0.0004930182678516992,
"loss": 4.8767,
"mean_token_accuracy": 0.2094111993908882,
"num_tokens": 21220249.0,
"step": 7595
},
{
"entropy": 5.151907587051392,
"epoch": 0.905300774270399,
"grad_norm": 1.125,
"learning_rate": 0.0004930077310346515,
"loss": 4.8607,
"mean_token_accuracy": 0.2092012956738472,
"num_tokens": 21233609.0,
"step": 7600
},
{
"entropy": 5.185947704315185,
"epoch": 0.9058963668850506,
"grad_norm": 0.97265625,
"learning_rate": 0.0004929971863980209,
"loss": 4.9733,
"mean_token_accuracy": 0.19832993149757386,
"num_tokens": 21247083.0,
"step": 7605
},
{
"entropy": 5.166575956344604,
"epoch": 0.9064919594997022,
"grad_norm": 0.9296875,
"learning_rate": 0.0004929866339421857,
"loss": 4.9198,
"mean_token_accuracy": 0.19654588401317596,
"num_tokens": 21260614.0,
"step": 7610
},
{
"entropy": 5.237410354614258,
"epoch": 0.9070875521143538,
"grad_norm": 1.015625,
"learning_rate": 0.0004929760736675243,
"loss": 4.9171,
"mean_token_accuracy": 0.20265579372644424,
"num_tokens": 21275823.0,
"step": 7615
},
{
"entropy": 5.13779296875,
"epoch": 0.9076831447290054,
"grad_norm": 0.984375,
"learning_rate": 0.0004929655055744157,
"loss": 4.8518,
"mean_token_accuracy": 0.2085930123925209,
"num_tokens": 21288787.0,
"step": 7620
},
{
"entropy": 5.149636125564575,
"epoch": 0.9082787373436569,
"grad_norm": 1.046875,
"learning_rate": 0.0004929549296632387,
"loss": 4.9764,
"mean_token_accuracy": 0.19542155414819717,
"num_tokens": 21304060.0,
"step": 7625
},
{
"entropy": 5.162720680236816,
"epoch": 0.9088743299583085,
"grad_norm": 1.015625,
"learning_rate": 0.0004929443459343728,
"loss": 4.865,
"mean_token_accuracy": 0.20526884794235228,
"num_tokens": 21317116.0,
"step": 7630
},
{
"entropy": 5.1623985290527346,
"epoch": 0.9094699225729601,
"grad_norm": 0.95703125,
"learning_rate": 0.0004929337543881975,
"loss": 4.864,
"mean_token_accuracy": 0.2060674712061882,
"num_tokens": 21330894.0,
"step": 7635
},
{
"entropy": 5.185609436035156,
"epoch": 0.9100655151876117,
"grad_norm": 0.9921875,
"learning_rate": 0.0004929231550250927,
"loss": 4.9204,
"mean_token_accuracy": 0.20490434914827346,
"num_tokens": 21344933.0,
"step": 7640
},
{
"entropy": 5.225409746170044,
"epoch": 0.9106611078022633,
"grad_norm": 0.99609375,
"learning_rate": 0.0004929125478454389,
"loss": 4.9774,
"mean_token_accuracy": 0.19510270804166793,
"num_tokens": 21358675.0,
"step": 7645
},
{
"entropy": 5.169189691543579,
"epoch": 0.9112567004169149,
"grad_norm": 0.99609375,
"learning_rate": 0.0004929019328496161,
"loss": 4.9521,
"mean_token_accuracy": 0.19673780649900435,
"num_tokens": 21373611.0,
"step": 7650
},
{
"entropy": 5.164188528060913,
"epoch": 0.9118522930315665,
"grad_norm": 1.046875,
"learning_rate": 0.0004928913100380053,
"loss": 4.8491,
"mean_token_accuracy": 0.21032273918390273,
"num_tokens": 21387292.0,
"step": 7655
},
{
"entropy": 5.2347705364227295,
"epoch": 0.912447885646218,
"grad_norm": 1.03125,
"learning_rate": 0.0004928806794109875,
"loss": 5.0591,
"mean_token_accuracy": 0.1917748123407364,
"num_tokens": 21400950.0,
"step": 7660
},
{
"entropy": 5.2469151496887205,
"epoch": 0.9130434782608695,
"grad_norm": 1.03125,
"learning_rate": 0.000492870040968944,
"loss": 5.0219,
"mean_token_accuracy": 0.19737962484359742,
"num_tokens": 21414625.0,
"step": 7665
},
{
"entropy": 5.131920862197876,
"epoch": 0.9136390708755211,
"grad_norm": 0.98046875,
"learning_rate": 0.0004928593947122564,
"loss": 4.9372,
"mean_token_accuracy": 0.20875568985939025,
"num_tokens": 21430408.0,
"step": 7670
},
{
"entropy": 5.20658540725708,
"epoch": 0.9142346634901727,
"grad_norm": 0.99609375,
"learning_rate": 0.0004928487406413065,
"loss": 4.9176,
"mean_token_accuracy": 0.1993631511926651,
"num_tokens": 21444921.0,
"step": 7675
},
{
"entropy": 5.2130261898040775,
"epoch": 0.9148302561048243,
"grad_norm": 1.0234375,
"learning_rate": 0.0004928380787564764,
"loss": 4.9707,
"mean_token_accuracy": 0.19577376395463944,
"num_tokens": 21458541.0,
"step": 7680
},
{
"entropy": 5.165385675430298,
"epoch": 0.9154258487194759,
"grad_norm": 0.9609375,
"learning_rate": 0.0004928274090581486,
"loss": 4.8991,
"mean_token_accuracy": 0.20316062718629838,
"num_tokens": 21472117.0,
"step": 7685
},
{
"entropy": 5.1834691047668455,
"epoch": 0.9160214413341274,
"grad_norm": 0.96875,
"learning_rate": 0.0004928167315467059,
"loss": 4.8945,
"mean_token_accuracy": 0.20826341807842255,
"num_tokens": 21484983.0,
"step": 7690
},
{
"entropy": 5.165830373764038,
"epoch": 0.916617033948779,
"grad_norm": 1.0234375,
"learning_rate": 0.0004928060462225312,
"loss": 4.8657,
"mean_token_accuracy": 0.20047949105501175,
"num_tokens": 21498786.0,
"step": 7695
},
{
"entropy": 5.17872428894043,
"epoch": 0.9172126265634306,
"grad_norm": 0.8984375,
"learning_rate": 0.0004927953530860077,
"loss": 4.8805,
"mean_token_accuracy": 0.20007957220077516,
"num_tokens": 21513482.0,
"step": 7700
},
{
"entropy": 5.194043922424316,
"epoch": 0.9178082191780822,
"grad_norm": 0.97265625,
"learning_rate": 0.0004927846521375189,
"loss": 4.9929,
"mean_token_accuracy": 0.20238216370344161,
"num_tokens": 21528648.0,
"step": 7705
},
{
"entropy": 5.178226280212402,
"epoch": 0.9184038117927338,
"grad_norm": 0.9453125,
"learning_rate": 0.0004927739433774489,
"loss": 4.8746,
"mean_token_accuracy": 0.2064097374677658,
"num_tokens": 21541490.0,
"step": 7710
},
{
"entropy": 5.137168264389038,
"epoch": 0.9189994044073854,
"grad_norm": 1.0,
"learning_rate": 0.0004927632268061815,
"loss": 4.8828,
"mean_token_accuracy": 0.20421603918075562,
"num_tokens": 21557156.0,
"step": 7715
},
{
"entropy": 5.123203992843628,
"epoch": 0.919594997022037,
"grad_norm": 1.0546875,
"learning_rate": 0.0004927525024241013,
"loss": 4.8182,
"mean_token_accuracy": 0.21074913442134857,
"num_tokens": 21571697.0,
"step": 7720
},
{
"entropy": 5.144837951660156,
"epoch": 0.9201905896366885,
"grad_norm": 1.1015625,
"learning_rate": 0.0004927417702315928,
"loss": 4.8636,
"mean_token_accuracy": 0.20946976840496062,
"num_tokens": 21585936.0,
"step": 7725
},
{
"entropy": 5.180933237075806,
"epoch": 0.9207861822513401,
"grad_norm": 0.94921875,
"learning_rate": 0.0004927310302290411,
"loss": 5.0021,
"mean_token_accuracy": 0.2012785032391548,
"num_tokens": 21601003.0,
"step": 7730
},
{
"entropy": 5.15260500907898,
"epoch": 0.9213817748659917,
"grad_norm": 0.96875,
"learning_rate": 0.0004927202824168313,
"loss": 4.8376,
"mean_token_accuracy": 0.20751550048589706,
"num_tokens": 21616018.0,
"step": 7735
},
{
"entropy": 5.221897459030151,
"epoch": 0.9219773674806433,
"grad_norm": 0.98828125,
"learning_rate": 0.000492709526795349,
"loss": 4.9376,
"mean_token_accuracy": 0.19613521844148635,
"num_tokens": 21628548.0,
"step": 7740
},
{
"entropy": 5.217655181884766,
"epoch": 0.9225729600952948,
"grad_norm": 0.98828125,
"learning_rate": 0.00049269876336498,
"loss": 5.0,
"mean_token_accuracy": 0.19123327136039733,
"num_tokens": 21643637.0,
"step": 7745
},
{
"entropy": 5.1366743564605715,
"epoch": 0.9231685527099464,
"grad_norm": 1.0625,
"learning_rate": 0.0004926879921261101,
"loss": 4.8843,
"mean_token_accuracy": 0.20081025809049607,
"num_tokens": 21658189.0,
"step": 7750
},
{
"entropy": 5.235289716720581,
"epoch": 0.9237641453245979,
"grad_norm": 1.1171875,
"learning_rate": 0.0004926772130791261,
"loss": 4.9813,
"mean_token_accuracy": 0.19532056897878647,
"num_tokens": 21673023.0,
"step": 7755
},
{
"entropy": 5.260555553436279,
"epoch": 0.9243597379392495,
"grad_norm": 0.98828125,
"learning_rate": 0.0004926664262244143,
"loss": 4.9684,
"mean_token_accuracy": 0.19974571317434311,
"num_tokens": 21687417.0,
"step": 7760
},
{
"entropy": 5.16275577545166,
"epoch": 0.9249553305539011,
"grad_norm": 1.0390625,
"learning_rate": 0.0004926556315623617,
"loss": 4.9778,
"mean_token_accuracy": 0.20222274363040924,
"num_tokens": 21700679.0,
"step": 7765
},
{
"entropy": 5.209200048446656,
"epoch": 0.9255509231685527,
"grad_norm": 0.91796875,
"learning_rate": 0.0004926448290933555,
"loss": 4.9343,
"mean_token_accuracy": 0.200205659866333,
"num_tokens": 21715832.0,
"step": 7770
},
{
"entropy": 5.158528518676758,
"epoch": 0.9261465157832043,
"grad_norm": 1.046875,
"learning_rate": 0.0004926340188177831,
"loss": 4.9124,
"mean_token_accuracy": 0.20357139706611632,
"num_tokens": 21730193.0,
"step": 7775
},
{
"entropy": 5.231134986877441,
"epoch": 0.9267421083978559,
"grad_norm": 0.984375,
"learning_rate": 0.0004926232007360324,
"loss": 4.9862,
"mean_token_accuracy": 0.19970730245113372,
"num_tokens": 21743177.0,
"step": 7780
},
{
"entropy": 5.0898134231567385,
"epoch": 0.9273377010125075,
"grad_norm": 1.2109375,
"learning_rate": 0.0004926123748484911,
"loss": 4.7507,
"mean_token_accuracy": 0.21657219529151917,
"num_tokens": 21757379.0,
"step": 7785
},
{
"entropy": 5.1121910572052,
"epoch": 0.927933293627159,
"grad_norm": 1.09375,
"learning_rate": 0.000492601541155548,
"loss": 4.8869,
"mean_token_accuracy": 0.20719768553972245,
"num_tokens": 21769872.0,
"step": 7790
},
{
"entropy": 5.231585454940796,
"epoch": 0.9285288862418106,
"grad_norm": 1.015625,
"learning_rate": 0.0004925906996575912,
"loss": 4.9294,
"mean_token_accuracy": 0.20413232296705247,
"num_tokens": 21784367.0,
"step": 7795
},
{
"entropy": 5.098396492004395,
"epoch": 0.9291244788564622,
"grad_norm": 1.0078125,
"learning_rate": 0.0004925798503550099,
"loss": 4.8469,
"mean_token_accuracy": 0.21068057268857956,
"num_tokens": 21797896.0,
"step": 7800
},
{
"entropy": 5.113478755950927,
"epoch": 0.9297200714711138,
"grad_norm": 1.03125,
"learning_rate": 0.0004925689932481931,
"loss": 4.8101,
"mean_token_accuracy": 0.20613028854131699,
"num_tokens": 21810668.0,
"step": 7805
},
{
"entropy": 5.18878207206726,
"epoch": 0.9303156640857654,
"grad_norm": 0.94140625,
"learning_rate": 0.0004925581283375303,
"loss": 4.9518,
"mean_token_accuracy": 0.1975715458393097,
"num_tokens": 21824923.0,
"step": 7810
},
{
"entropy": 5.18545355796814,
"epoch": 0.930911256700417,
"grad_norm": 0.96875,
"learning_rate": 0.0004925472556234111,
"loss": 4.9397,
"mean_token_accuracy": 0.19598566442728044,
"num_tokens": 21839967.0,
"step": 7815
},
{
"entropy": 5.161255264282227,
"epoch": 0.9315068493150684,
"grad_norm": 1.1015625,
"learning_rate": 0.0004925363751062256,
"loss": 4.8987,
"mean_token_accuracy": 0.21003638803958893,
"num_tokens": 21852254.0,
"step": 7820
},
{
"entropy": 5.158992195129395,
"epoch": 0.93210244192972,
"grad_norm": 1.109375,
"learning_rate": 0.000492525486786364,
"loss": 4.9598,
"mean_token_accuracy": 0.1967307969927788,
"num_tokens": 21865227.0,
"step": 7825
},
{
"entropy": 5.240191793441772,
"epoch": 0.9326980345443716,
"grad_norm": 1.0625,
"learning_rate": 0.0004925145906642169,
"loss": 4.9709,
"mean_token_accuracy": 0.1956178441643715,
"num_tokens": 21880235.0,
"step": 7830
},
{
"entropy": 5.164791584014893,
"epoch": 0.9332936271590232,
"grad_norm": 1.0703125,
"learning_rate": 0.0004925036867401751,
"loss": 4.8783,
"mean_token_accuracy": 0.20806253999471663,
"num_tokens": 21892514.0,
"step": 7835
},
{
"entropy": 5.198020887374878,
"epoch": 0.9338892197736748,
"grad_norm": 1.1171875,
"learning_rate": 0.0004924927750146296,
"loss": 4.9717,
"mean_token_accuracy": 0.19989160299301148,
"num_tokens": 21907212.0,
"step": 7840
},
{
"entropy": 5.134514904022216,
"epoch": 0.9344848123883264,
"grad_norm": 1.125,
"learning_rate": 0.0004924818554879718,
"loss": 4.8159,
"mean_token_accuracy": 0.2098813146352768,
"num_tokens": 21920627.0,
"step": 7845
},
{
"entropy": 5.201883697509766,
"epoch": 0.935080405002978,
"grad_norm": 1.1171875,
"learning_rate": 0.0004924709281605938,
"loss": 4.9465,
"mean_token_accuracy": 0.20151709616184235,
"num_tokens": 21933557.0,
"step": 7850
},
{
"entropy": 5.2099262237548825,
"epoch": 0.9356759976176295,
"grad_norm": 1.09375,
"learning_rate": 0.0004924599930328869,
"loss": 4.9347,
"mean_token_accuracy": 0.2006511077284813,
"num_tokens": 21948122.0,
"step": 7855
},
{
"entropy": 5.150583362579345,
"epoch": 0.9362715902322811,
"grad_norm": 1.0234375,
"learning_rate": 0.0004924490501052437,
"loss": 4.8976,
"mean_token_accuracy": 0.21140341013669967,
"num_tokens": 21961925.0,
"step": 7860
},
{
"entropy": 5.24985146522522,
"epoch": 0.9368671828469327,
"grad_norm": 1.109375,
"learning_rate": 0.0004924380993780566,
"loss": 5.0153,
"mean_token_accuracy": 0.19524954855442048,
"num_tokens": 21976825.0,
"step": 7865
},
{
"entropy": 5.1776855945587155,
"epoch": 0.9374627754615843,
"grad_norm": 1.140625,
"learning_rate": 0.0004924271408517185,
"loss": 4.9144,
"mean_token_accuracy": 0.2031736522912979,
"num_tokens": 21989599.0,
"step": 7870
},
{
"entropy": 5.188190746307373,
"epoch": 0.9380583680762359,
"grad_norm": 0.94921875,
"learning_rate": 0.0004924161745266222,
"loss": 4.9402,
"mean_token_accuracy": 0.20079342573881148,
"num_tokens": 22002921.0,
"step": 7875
},
{
"entropy": 5.137540292739868,
"epoch": 0.9386539606908875,
"grad_norm": 1.0,
"learning_rate": 0.0004924052004031613,
"loss": 4.8432,
"mean_token_accuracy": 0.2069815218448639,
"num_tokens": 22015933.0,
"step": 7880
},
{
"entropy": 5.116696500778199,
"epoch": 0.939249553305539,
"grad_norm": 0.890625,
"learning_rate": 0.0004923942184817293,
"loss": 4.8342,
"mean_token_accuracy": 0.21012721061706544,
"num_tokens": 22029769.0,
"step": 7885
},
{
"entropy": 5.159595108032226,
"epoch": 0.9398451459201906,
"grad_norm": 1.03125,
"learning_rate": 0.0004923832287627202,
"loss": 4.9354,
"mean_token_accuracy": 0.19617464989423752,
"num_tokens": 22044025.0,
"step": 7890
},
{
"entropy": 5.159423398971557,
"epoch": 0.9404407385348421,
"grad_norm": 1.109375,
"learning_rate": 0.0004923722312465281,
"loss": 4.8492,
"mean_token_accuracy": 0.20644398927688598,
"num_tokens": 22057875.0,
"step": 7895
},
{
"entropy": 5.129915237426758,
"epoch": 0.9410363311494937,
"grad_norm": 0.98046875,
"learning_rate": 0.0004923612259335476,
"loss": 4.8245,
"mean_token_accuracy": 0.2051606446504593,
"num_tokens": 22070995.0,
"step": 7900
},
{
"entropy": 5.152171277999878,
"epoch": 0.9416319237641453,
"grad_norm": 0.96484375,
"learning_rate": 0.0004923502128241732,
"loss": 4.9012,
"mean_token_accuracy": 0.20087936669588088,
"num_tokens": 22086429.0,
"step": 7905
},
{
"entropy": 5.092618751525879,
"epoch": 0.9422275163787969,
"grad_norm": 1.0,
"learning_rate": 0.0004923391919188001,
"loss": 4.8036,
"mean_token_accuracy": 0.21061836928129196,
"num_tokens": 22101146.0,
"step": 7910
},
{
"entropy": 5.1590869426727295,
"epoch": 0.9428231089934485,
"grad_norm": 0.98046875,
"learning_rate": 0.0004923281632178236,
"loss": 4.8643,
"mean_token_accuracy": 0.21170777976512908,
"num_tokens": 22113103.0,
"step": 7915
},
{
"entropy": 5.16511492729187,
"epoch": 0.9434187016081,
"grad_norm": 0.95703125,
"learning_rate": 0.0004923171267216392,
"loss": 4.8757,
"mean_token_accuracy": 0.2064968854188919,
"num_tokens": 22126540.0,
"step": 7920
},
{
"entropy": 5.146010541915894,
"epoch": 0.9440142942227516,
"grad_norm": 0.97265625,
"learning_rate": 0.0004923060824306428,
"loss": 4.9057,
"mean_token_accuracy": 0.1991073340177536,
"num_tokens": 22141219.0,
"step": 7925
},
{
"entropy": 5.203396034240723,
"epoch": 0.9446098868374032,
"grad_norm": 0.96875,
"learning_rate": 0.0004922950303452306,
"loss": 4.9769,
"mean_token_accuracy": 0.20016351193189622,
"num_tokens": 22156178.0,
"step": 7930
},
{
"entropy": 5.147933435440064,
"epoch": 0.9452054794520548,
"grad_norm": 1.078125,
"learning_rate": 0.0004922839704657988,
"loss": 4.8287,
"mean_token_accuracy": 0.21361913084983825,
"num_tokens": 22168981.0,
"step": 7935
},
{
"entropy": 5.206509780883789,
"epoch": 0.9458010720667064,
"grad_norm": 1.0078125,
"learning_rate": 0.0004922729027927444,
"loss": 4.8822,
"mean_token_accuracy": 0.2085515394806862,
"num_tokens": 22182288.0,
"step": 7940
},
{
"entropy": 5.221108102798462,
"epoch": 0.946396664681358,
"grad_norm": 0.9921875,
"learning_rate": 0.0004922618273264642,
"loss": 4.9942,
"mean_token_accuracy": 0.20079006105661393,
"num_tokens": 22197564.0,
"step": 7945
},
{
"entropy": 5.190614557266235,
"epoch": 0.9469922572960096,
"grad_norm": 0.90234375,
"learning_rate": 0.0004922507440673555,
"loss": 5.0619,
"mean_token_accuracy": 0.1911136642098427,
"num_tokens": 22212030.0,
"step": 7950
},
{
"entropy": 5.162139749526977,
"epoch": 0.9475878499106611,
"grad_norm": 0.9453125,
"learning_rate": 0.0004922396530158159,
"loss": 4.9119,
"mean_token_accuracy": 0.19273923486471176,
"num_tokens": 22226941.0,
"step": 7955
},
{
"entropy": 5.1744640350341795,
"epoch": 0.9481834425253127,
"grad_norm": 1.015625,
"learning_rate": 0.0004922285541722431,
"loss": 4.9326,
"mean_token_accuracy": 0.20148904770612716,
"num_tokens": 22241129.0,
"step": 7960
},
{
"entropy": 5.1741554737091064,
"epoch": 0.9487790351399643,
"grad_norm": 0.96875,
"learning_rate": 0.0004922174475370352,
"loss": 4.9198,
"mean_token_accuracy": 0.19810419529676437,
"num_tokens": 22256298.0,
"step": 7965
},
{
"entropy": 5.162768745422364,
"epoch": 0.9493746277546159,
"grad_norm": 0.96484375,
"learning_rate": 0.0004922063331105907,
"loss": 4.894,
"mean_token_accuracy": 0.20560975819826127,
"num_tokens": 22270128.0,
"step": 7970
},
{
"entropy": 5.237190389633179,
"epoch": 0.9499702203692674,
"grad_norm": 0.96875,
"learning_rate": 0.0004921952108933083,
"loss": 4.9683,
"mean_token_accuracy": 0.19755639433860778,
"num_tokens": 22284847.0,
"step": 7975
},
{
"entropy": 5.130140495300293,
"epoch": 0.950565812983919,
"grad_norm": 0.94140625,
"learning_rate": 0.0004921840808855865,
"loss": 4.8026,
"mean_token_accuracy": 0.21034963577985763,
"num_tokens": 22300010.0,
"step": 7980
},
{
"entropy": 5.107722187042237,
"epoch": 0.9511614055985705,
"grad_norm": 1.015625,
"learning_rate": 0.0004921729430878252,
"loss": 4.8869,
"mean_token_accuracy": 0.20978888869285583,
"num_tokens": 22313339.0,
"step": 7985
},
{
"entropy": 5.188035774230957,
"epoch": 0.9517569982132221,
"grad_norm": 0.984375,
"learning_rate": 0.0004921617975004233,
"loss": 4.9018,
"mean_token_accuracy": 0.19784007221460342,
"num_tokens": 22327659.0,
"step": 7990
},
{
"entropy": 5.257749366760254,
"epoch": 0.9523525908278737,
"grad_norm": 0.9296875,
"learning_rate": 0.0004921506441237809,
"loss": 4.9958,
"mean_token_accuracy": 0.19972198456525803,
"num_tokens": 22343910.0,
"step": 7995
},
{
"entropy": 5.0519969940185545,
"epoch": 0.9529481834425253,
"grad_norm": 0.9765625,
"learning_rate": 0.0004921394829582979,
"loss": 4.8635,
"mean_token_accuracy": 0.20977049171924592,
"num_tokens": 22359705.0,
"step": 8000
},
{
"entropy": 5.1608421325683596,
"epoch": 0.9535437760571769,
"grad_norm": 0.96875,
"learning_rate": 0.0004921283140043748,
"loss": 4.9456,
"mean_token_accuracy": 0.1973240330815315,
"num_tokens": 22373446.0,
"step": 8005
},
{
"entropy": 5.227081537246704,
"epoch": 0.9541393686718285,
"grad_norm": 0.9453125,
"learning_rate": 0.0004921171372624121,
"loss": 4.9204,
"mean_token_accuracy": 0.20275699496269226,
"num_tokens": 22386628.0,
"step": 8010
},
{
"entropy": 5.210403299331665,
"epoch": 0.95473496128648,
"grad_norm": 1.0546875,
"learning_rate": 0.0004921059527328106,
"loss": 4.9937,
"mean_token_accuracy": 0.20061930567026137,
"num_tokens": 22400084.0,
"step": 8015
},
{
"entropy": 5.168334293365478,
"epoch": 0.9553305539011316,
"grad_norm": 1.1015625,
"learning_rate": 0.0004920947604159716,
"loss": 4.8744,
"mean_token_accuracy": 0.19991486966609956,
"num_tokens": 22414135.0,
"step": 8020
},
{
"entropy": 5.129685974121093,
"epoch": 0.9559261465157832,
"grad_norm": 0.98046875,
"learning_rate": 0.0004920835603122965,
"loss": 4.8526,
"mean_token_accuracy": 0.20220229774713516,
"num_tokens": 22427738.0,
"step": 8025
},
{
"entropy": 5.162521314620972,
"epoch": 0.9565217391304348,
"grad_norm": 0.98828125,
"learning_rate": 0.0004920723524221871,
"loss": 4.9373,
"mean_token_accuracy": 0.20917006880044936,
"num_tokens": 22440024.0,
"step": 8030
},
{
"entropy": 5.204454803466797,
"epoch": 0.9571173317450864,
"grad_norm": 0.8984375,
"learning_rate": 0.0004920611367460453,
"loss": 4.8417,
"mean_token_accuracy": 0.20752593576908113,
"num_tokens": 22453948.0,
"step": 8035
},
{
"entropy": 5.224903106689453,
"epoch": 0.957712924359738,
"grad_norm": 1.0703125,
"learning_rate": 0.0004920499132842734,
"loss": 4.9477,
"mean_token_accuracy": 0.20166724175214767,
"num_tokens": 22465613.0,
"step": 8040
},
{
"entropy": 5.12958779335022,
"epoch": 0.9583085169743896,
"grad_norm": 1.03125,
"learning_rate": 0.0004920386820372741,
"loss": 4.8738,
"mean_token_accuracy": 0.19878884106874467,
"num_tokens": 22478702.0,
"step": 8045
},
{
"entropy": 5.218903255462647,
"epoch": 0.958904109589041,
"grad_norm": 1.125,
"learning_rate": 0.0004920274430054502,
"loss": 4.9623,
"mean_token_accuracy": 0.1980475977063179,
"num_tokens": 22490947.0,
"step": 8050
},
{
"entropy": 5.100000667572021,
"epoch": 0.9594997022036926,
"grad_norm": 1.03125,
"learning_rate": 0.0004920161961892047,
"loss": 4.9077,
"mean_token_accuracy": 0.197797092795372,
"num_tokens": 22506308.0,
"step": 8055
},
{
"entropy": 5.305327653884888,
"epoch": 0.9600952948183442,
"grad_norm": 0.953125,
"learning_rate": 0.0004920049415889411,
"loss": 4.9553,
"mean_token_accuracy": 0.1938668355345726,
"num_tokens": 22522045.0,
"step": 8060
},
{
"entropy": 5.161783170700073,
"epoch": 0.9606908874329958,
"grad_norm": 0.9609375,
"learning_rate": 0.0004919936792050631,
"loss": 4.8703,
"mean_token_accuracy": 0.2016933888196945,
"num_tokens": 22536183.0,
"step": 8065
},
{
"entropy": 5.133864688873291,
"epoch": 0.9612864800476474,
"grad_norm": 1.03125,
"learning_rate": 0.0004919824090379747,
"loss": 4.8697,
"mean_token_accuracy": 0.20367311835289,
"num_tokens": 22548239.0,
"step": 8070
},
{
"entropy": 5.210617876052856,
"epoch": 0.961882072662299,
"grad_norm": 1.015625,
"learning_rate": 0.00049197113108808,
"loss": 4.9271,
"mean_token_accuracy": 0.20728749334812163,
"num_tokens": 22563467.0,
"step": 8075
},
{
"entropy": 5.163472318649292,
"epoch": 0.9624776652769506,
"grad_norm": 0.94140625,
"learning_rate": 0.0004919598453557836,
"loss": 4.9086,
"mean_token_accuracy": 0.2029006913304329,
"num_tokens": 22577446.0,
"step": 8080
},
{
"entropy": 5.164157962799072,
"epoch": 0.9630732578916021,
"grad_norm": 0.984375,
"learning_rate": 0.0004919485518414902,
"loss": 4.864,
"mean_token_accuracy": 0.2043229341506958,
"num_tokens": 22591044.0,
"step": 8085
},
{
"entropy": 5.102521991729736,
"epoch": 0.9636688505062537,
"grad_norm": 1.0,
"learning_rate": 0.0004919372505456051,
"loss": 4.8751,
"mean_token_accuracy": 0.2069119244813919,
"num_tokens": 22604872.0,
"step": 8090
},
{
"entropy": 5.18564772605896,
"epoch": 0.9642644431209053,
"grad_norm": 0.91796875,
"learning_rate": 0.0004919259414685336,
"loss": 4.9769,
"mean_token_accuracy": 0.19262531399726868,
"num_tokens": 22620138.0,
"step": 8095
},
{
"entropy": 5.25115065574646,
"epoch": 0.9648600357355569,
"grad_norm": 1.15625,
"learning_rate": 0.0004919146246106812,
"loss": 4.95,
"mean_token_accuracy": 0.19694037437438966,
"num_tokens": 22635580.0,
"step": 8100
},
{
"entropy": 5.1543457984924315,
"epoch": 0.9654556283502085,
"grad_norm": 0.87890625,
"learning_rate": 0.0004919032999724539,
"loss": 4.8781,
"mean_token_accuracy": 0.20748892426490784,
"num_tokens": 22650223.0,
"step": 8105
},
{
"entropy": 5.090401268005371,
"epoch": 0.9660512209648601,
"grad_norm": 0.93359375,
"learning_rate": 0.0004918919675542579,
"loss": 4.8278,
"mean_token_accuracy": 0.21135323941707612,
"num_tokens": 22664899.0,
"step": 8110
},
{
"entropy": 5.194217300415039,
"epoch": 0.9666468135795117,
"grad_norm": 0.97265625,
"learning_rate": 0.0004918806273564996,
"loss": 4.9122,
"mean_token_accuracy": 0.2026853084564209,
"num_tokens": 22680052.0,
"step": 8115
},
{
"entropy": 5.178218936920166,
"epoch": 0.9672424061941632,
"grad_norm": 1.015625,
"learning_rate": 0.000491869279379586,
"loss": 4.8345,
"mean_token_accuracy": 0.21272536963224412,
"num_tokens": 22693740.0,
"step": 8120
},
{
"entropy": 5.076748514175415,
"epoch": 0.9678379988088148,
"grad_norm": 1.0078125,
"learning_rate": 0.0004918579236239238,
"loss": 4.84,
"mean_token_accuracy": 0.21446588933467864,
"num_tokens": 22707795.0,
"step": 8125
},
{
"entropy": 5.1775915145874025,
"epoch": 0.9684335914234663,
"grad_norm": 1.0703125,
"learning_rate": 0.0004918465600899205,
"loss": 4.9193,
"mean_token_accuracy": 0.20889559090137483,
"num_tokens": 22720296.0,
"step": 8130
},
{
"entropy": 5.19767689704895,
"epoch": 0.9690291840381179,
"grad_norm": 0.97265625,
"learning_rate": 0.0004918351887779837,
"loss": 4.9204,
"mean_token_accuracy": 0.20312157422304153,
"num_tokens": 22733680.0,
"step": 8135
},
{
"entropy": 5.1238484382629395,
"epoch": 0.9696247766527695,
"grad_norm": 1.03125,
"learning_rate": 0.0004918238096885213,
"loss": 4.864,
"mean_token_accuracy": 0.2062617540359497,
"num_tokens": 22746202.0,
"step": 8140
},
{
"entropy": 5.168711948394775,
"epoch": 0.9702203692674211,
"grad_norm": 0.953125,
"learning_rate": 0.0004918124228219412,
"loss": 4.8945,
"mean_token_accuracy": 0.20336710810661315,
"num_tokens": 22759547.0,
"step": 8145
},
{
"entropy": 5.187703657150268,
"epoch": 0.9708159618820726,
"grad_norm": 0.9921875,
"learning_rate": 0.0004918010281786522,
"loss": 4.8583,
"mean_token_accuracy": 0.20362784415483476,
"num_tokens": 22772593.0,
"step": 8150
},
{
"entropy": 5.165614366531372,
"epoch": 0.9714115544967242,
"grad_norm": 0.97265625,
"learning_rate": 0.0004917896257590627,
"loss": 4.9544,
"mean_token_accuracy": 0.19932200759649277,
"num_tokens": 22786482.0,
"step": 8155
},
{
"entropy": 5.161979055404663,
"epoch": 0.9720071471113758,
"grad_norm": 0.9296875,
"learning_rate": 0.0004917782155635817,
"loss": 4.8799,
"mean_token_accuracy": 0.20211428850889207,
"num_tokens": 22800582.0,
"step": 8160
},
{
"entropy": 5.180449390411377,
"epoch": 0.9726027397260274,
"grad_norm": 1.0078125,
"learning_rate": 0.0004917667975926187,
"loss": 4.9193,
"mean_token_accuracy": 0.19984211921691894,
"num_tokens": 22813661.0,
"step": 8165
},
{
"entropy": 5.203529739379883,
"epoch": 0.973198332340679,
"grad_norm": 0.98046875,
"learning_rate": 0.0004917553718465832,
"loss": 4.8898,
"mean_token_accuracy": 0.21034502238035202,
"num_tokens": 22827648.0,
"step": 8170
},
{
"entropy": 5.2240376472473145,
"epoch": 0.9737939249553306,
"grad_norm": 1.0546875,
"learning_rate": 0.0004917439383258848,
"loss": 4.9921,
"mean_token_accuracy": 0.19576196372509003,
"num_tokens": 22843345.0,
"step": 8175
},
{
"entropy": 5.252078247070313,
"epoch": 0.9743895175699822,
"grad_norm": 1.0546875,
"learning_rate": 0.0004917324970309337,
"loss": 4.9857,
"mean_token_accuracy": 0.19702121019363403,
"num_tokens": 22857474.0,
"step": 8180
},
{
"entropy": 5.144397163391114,
"epoch": 0.9749851101846337,
"grad_norm": 0.96484375,
"learning_rate": 0.0004917210479621404,
"loss": 4.8819,
"mean_token_accuracy": 0.20598209798336028,
"num_tokens": 22870233.0,
"step": 8185
},
{
"entropy": 5.177191066741943,
"epoch": 0.9755807027992853,
"grad_norm": 0.87109375,
"learning_rate": 0.0004917095911199155,
"loss": 4.9235,
"mean_token_accuracy": 0.20674002468585967,
"num_tokens": 22885616.0,
"step": 8190
},
{
"entropy": 5.140366172790527,
"epoch": 0.9761762954139369,
"grad_norm": 1.0625,
"learning_rate": 0.00049169812650467,
"loss": 4.8655,
"mean_token_accuracy": 0.2119220957159996,
"num_tokens": 22899553.0,
"step": 8195
},
{
"entropy": 5.13837833404541,
"epoch": 0.9767718880285885,
"grad_norm": 1.0625,
"learning_rate": 0.0004916866541168148,
"loss": 4.8357,
"mean_token_accuracy": 0.20886047333478927,
"num_tokens": 22912671.0,
"step": 8200
},
{
"entropy": 5.20396671295166,
"epoch": 0.97736748064324,
"grad_norm": 1.0546875,
"learning_rate": 0.0004916751739567617,
"loss": 5.0044,
"mean_token_accuracy": 0.19555548578500748,
"num_tokens": 22926112.0,
"step": 8205
},
{
"entropy": 5.164089632034302,
"epoch": 0.9779630732578916,
"grad_norm": 0.83203125,
"learning_rate": 0.0004916636860249225,
"loss": 4.823,
"mean_token_accuracy": 0.2166283071041107,
"num_tokens": 22940348.0,
"step": 8210
},
{
"entropy": 5.148074722290039,
"epoch": 0.9785586658725431,
"grad_norm": 1.046875,
"learning_rate": 0.000491652190321709,
"loss": 4.9517,
"mean_token_accuracy": 0.20277225226163864,
"num_tokens": 22953347.0,
"step": 8215
},
{
"entropy": 5.1890722751617435,
"epoch": 0.9791542584871947,
"grad_norm": 0.98828125,
"learning_rate": 0.0004916406868475338,
"loss": 5.0294,
"mean_token_accuracy": 0.18531037569046022,
"num_tokens": 22967963.0,
"step": 8220
},
{
"entropy": 5.213222408294678,
"epoch": 0.9797498511018463,
"grad_norm": 0.984375,
"learning_rate": 0.0004916291756028094,
"loss": 4.9019,
"mean_token_accuracy": 0.19597947150468825,
"num_tokens": 22985611.0,
"step": 8225
},
{
"entropy": 5.207565975189209,
"epoch": 0.9803454437164979,
"grad_norm": 1.1875,
"learning_rate": 0.0004916176565879487,
"loss": 4.9133,
"mean_token_accuracy": 0.1998802751302719,
"num_tokens": 22999210.0,
"step": 8230
},
{
"entropy": 5.15940670967102,
"epoch": 0.9809410363311495,
"grad_norm": 1.0234375,
"learning_rate": 0.0004916061298033648,
"loss": 4.8663,
"mean_token_accuracy": 0.20396122634410857,
"num_tokens": 23012315.0,
"step": 8235
},
{
"entropy": 5.16358757019043,
"epoch": 0.9815366289458011,
"grad_norm": 1.0234375,
"learning_rate": 0.0004915945952494713,
"loss": 4.9065,
"mean_token_accuracy": 0.20330557823181153,
"num_tokens": 23025886.0,
"step": 8240
},
{
"entropy": 5.163049507141113,
"epoch": 0.9821322215604527,
"grad_norm": 0.94140625,
"learning_rate": 0.0004915830529266818,
"loss": 4.7955,
"mean_token_accuracy": 0.21103596985340117,
"num_tokens": 23040308.0,
"step": 8245
},
{
"entropy": 5.178329706192017,
"epoch": 0.9827278141751042,
"grad_norm": 0.984375,
"learning_rate": 0.0004915715028354102,
"loss": 4.9197,
"mean_token_accuracy": 0.20345995277166368,
"num_tokens": 23053608.0,
"step": 8250
},
{
"entropy": 5.203071975708008,
"epoch": 0.9833234067897558,
"grad_norm": 0.9140625,
"learning_rate": 0.0004915599449760711,
"loss": 4.8933,
"mean_token_accuracy": 0.2109983429312706,
"num_tokens": 23068153.0,
"step": 8255
},
{
"entropy": 5.1310789585113525,
"epoch": 0.9839189994044074,
"grad_norm": 1.015625,
"learning_rate": 0.0004915483793490788,
"loss": 4.9494,
"mean_token_accuracy": 0.19942150712013246,
"num_tokens": 23081334.0,
"step": 8260
},
{
"entropy": 5.193494653701782,
"epoch": 0.984514592019059,
"grad_norm": 1.171875,
"learning_rate": 0.0004915368059548482,
"loss": 4.8989,
"mean_token_accuracy": 0.1990894705057144,
"num_tokens": 23095943.0,
"step": 8265
},
{
"entropy": 5.207312870025635,
"epoch": 0.9851101846337106,
"grad_norm": 0.9375,
"learning_rate": 0.0004915252247937944,
"loss": 4.8635,
"mean_token_accuracy": 0.2022635281085968,
"num_tokens": 23110311.0,
"step": 8270
},
{
"entropy": 5.1716773986816404,
"epoch": 0.9857057772483622,
"grad_norm": 1.0078125,
"learning_rate": 0.000491513635866333,
"loss": 4.9302,
"mean_token_accuracy": 0.20561778843402861,
"num_tokens": 23123783.0,
"step": 8275
},
{
"entropy": 5.225357866287231,
"epoch": 0.9863013698630136,
"grad_norm": 0.97265625,
"learning_rate": 0.0004915020391728792,
"loss": 4.994,
"mean_token_accuracy": 0.2005189388990402,
"num_tokens": 23139147.0,
"step": 8280
},
{
"entropy": 5.167164468765259,
"epoch": 0.9868969624776652,
"grad_norm": 0.9296875,
"learning_rate": 0.0004914904347138495,
"loss": 4.8667,
"mean_token_accuracy": 0.20874838083982467,
"num_tokens": 23153216.0,
"step": 8285
},
{
"entropy": 5.093227291107178,
"epoch": 0.9874925550923168,
"grad_norm": 1.1640625,
"learning_rate": 0.0004914788224896599,
"loss": 4.8145,
"mean_token_accuracy": 0.21401685625314712,
"num_tokens": 23166263.0,
"step": 8290
},
{
"entropy": 5.150248908996582,
"epoch": 0.9880881477069684,
"grad_norm": 1.03125,
"learning_rate": 0.0004914672025007269,
"loss": 4.8568,
"mean_token_accuracy": 0.2012554332613945,
"num_tokens": 23182093.0,
"step": 8295
},
{
"entropy": 5.152685546875,
"epoch": 0.98868374032162,
"grad_norm": 0.9609375,
"learning_rate": 0.0004914555747474673,
"loss": 4.9202,
"mean_token_accuracy": 0.2052308961749077,
"num_tokens": 23195492.0,
"step": 8300
},
{
"entropy": 5.231600618362426,
"epoch": 0.9892793329362716,
"grad_norm": 0.953125,
"learning_rate": 0.0004914439392302981,
"loss": 4.9638,
"mean_token_accuracy": 0.193992979824543,
"num_tokens": 23208333.0,
"step": 8305
},
{
"entropy": 5.1744935512542725,
"epoch": 0.9898749255509232,
"grad_norm": 0.9609375,
"learning_rate": 0.0004914322959496367,
"loss": 4.8752,
"mean_token_accuracy": 0.19933773577213287,
"num_tokens": 23223265.0,
"step": 8310
},
{
"entropy": 5.170458507537842,
"epoch": 0.9904705181655747,
"grad_norm": 1.03125,
"learning_rate": 0.0004914206449059008,
"loss": 4.8264,
"mean_token_accuracy": 0.20593057423830033,
"num_tokens": 23237534.0,
"step": 8315
},
{
"entropy": 5.123005533218384,
"epoch": 0.9910661107802263,
"grad_norm": 1.09375,
"learning_rate": 0.0004914089860995084,
"loss": 4.8877,
"mean_token_accuracy": 0.20614349991083145,
"num_tokens": 23252038.0,
"step": 8320
},
{
"entropy": 5.188803100585938,
"epoch": 0.9916617033948779,
"grad_norm": 0.94921875,
"learning_rate": 0.0004913973195308772,
"loss": 4.9288,
"mean_token_accuracy": 0.19964564144611358,
"num_tokens": 23265728.0,
"step": 8325
},
{
"entropy": 5.199964427947998,
"epoch": 0.9922572960095295,
"grad_norm": 1.078125,
"learning_rate": 0.0004913856452004262,
"loss": 4.8712,
"mean_token_accuracy": 0.20444389879703523,
"num_tokens": 23278282.0,
"step": 8330
},
{
"entropy": 5.1015252590179445,
"epoch": 0.9928528886241811,
"grad_norm": 0.94140625,
"learning_rate": 0.0004913739631085739,
"loss": 4.9088,
"mean_token_accuracy": 0.21046992242336274,
"num_tokens": 23293041.0,
"step": 8335
},
{
"entropy": 5.127294635772705,
"epoch": 0.9934484812388327,
"grad_norm": 1.0078125,
"learning_rate": 0.0004913622732557394,
"loss": 4.8724,
"mean_token_accuracy": 0.2052022010087967,
"num_tokens": 23307307.0,
"step": 8340
},
{
"entropy": 5.238845014572144,
"epoch": 0.9940440738534843,
"grad_norm": 1.1015625,
"learning_rate": 0.0004913505756423418,
"loss": 4.942,
"mean_token_accuracy": 0.19663116484880447,
"num_tokens": 23324064.0,
"step": 8345
},
{
"entropy": 5.144301176071167,
"epoch": 0.9946396664681358,
"grad_norm": 1.0390625,
"learning_rate": 0.0004913388702688009,
"loss": 4.8329,
"mean_token_accuracy": 0.21152608394622802,
"num_tokens": 23338498.0,
"step": 8350
},
{
"entropy": 5.099467658996582,
"epoch": 0.9952352590827874,
"grad_norm": 1.0625,
"learning_rate": 0.0004913271571355365,
"loss": 4.8576,
"mean_token_accuracy": 0.2067858874797821,
"num_tokens": 23352252.0,
"step": 8355
},
{
"entropy": 5.190935230255127,
"epoch": 0.9958308516974389,
"grad_norm": 1.0390625,
"learning_rate": 0.0004913154362429686,
"loss": 4.9761,
"mean_token_accuracy": 0.19361001998186111,
"num_tokens": 23366220.0,
"step": 8360
},
{
"entropy": 5.172525119781494,
"epoch": 0.9964264443120905,
"grad_norm": 0.96875,
"learning_rate": 0.0004913037075915178,
"loss": 4.8472,
"mean_token_accuracy": 0.20368663519620894,
"num_tokens": 23378926.0,
"step": 8365
},
{
"entropy": 5.17773962020874,
"epoch": 0.9970220369267421,
"grad_norm": 1.078125,
"learning_rate": 0.0004912919711816047,
"loss": 4.903,
"mean_token_accuracy": 0.20295153558254242,
"num_tokens": 23391252.0,
"step": 8370
},
{
"entropy": 5.144041967391968,
"epoch": 0.9976176295413937,
"grad_norm": 1.1171875,
"learning_rate": 0.0004912802270136503,
"loss": 4.8929,
"mean_token_accuracy": 0.2024414911866188,
"num_tokens": 23403812.0,
"step": 8375
},
{
"entropy": 5.2071083068847654,
"epoch": 0.9982132221560452,
"grad_norm": 1.0078125,
"learning_rate": 0.0004912684750880757,
"loss": 4.8845,
"mean_token_accuracy": 0.20607912093400954,
"num_tokens": 23417536.0,
"step": 8380
},
{
"entropy": 5.248421478271484,
"epoch": 0.9988088147706968,
"grad_norm": 0.9140625,
"learning_rate": 0.0004912567154053025,
"loss": 4.9482,
"mean_token_accuracy": 0.20182356834411622,
"num_tokens": 23431897.0,
"step": 8385
},
{
"entropy": 5.188880062103271,
"epoch": 0.9994044073853484,
"grad_norm": 0.94140625,
"learning_rate": 0.0004912449479657526,
"loss": 4.9064,
"mean_token_accuracy": 0.19837099313735962,
"num_tokens": 23446561.0,
"step": 8390
},
{
"entropy": 5.094776010513305,
"epoch": 1.0,
"grad_norm": 1.0234375,
"learning_rate": 0.0004912331727698481,
"loss": 4.8664,
"mean_token_accuracy": 0.2057012528181076,
"num_tokens": 23461012.0,
"step": 8395
},
{
"entropy": 5.138588714599609,
"epoch": 1.0005955926146515,
"grad_norm": 1.1953125,
"learning_rate": 0.0004912213898180111,
"loss": 4.8038,
"mean_token_accuracy": 0.21128290444612502,
"num_tokens": 23473208.0,
"step": 8400
},
{
"entropy": 5.217167758941651,
"epoch": 1.0011911852293032,
"grad_norm": 1.03125,
"learning_rate": 0.0004912095991106644,
"loss": 4.8897,
"mean_token_accuracy": 0.20315583497285844,
"num_tokens": 23487450.0,
"step": 8405
},
{
"entropy": 5.201423120498657,
"epoch": 1.0017867778439546,
"grad_norm": 1.046875,
"learning_rate": 0.0004911978006482311,
"loss": 4.9261,
"mean_token_accuracy": 0.20133694857358933,
"num_tokens": 23500630.0,
"step": 8410
},
{
"entropy": 5.209515476226807,
"epoch": 1.0023823704586063,
"grad_norm": 1.0234375,
"learning_rate": 0.0004911859944311341,
"loss": 4.8873,
"mean_token_accuracy": 0.20373353213071824,
"num_tokens": 23513999.0,
"step": 8415
},
{
"entropy": 5.192874383926392,
"epoch": 1.0029779630732578,
"grad_norm": 0.95703125,
"learning_rate": 0.000491174180459797,
"loss": 4.8824,
"mean_token_accuracy": 0.20343583524227143,
"num_tokens": 23529046.0,
"step": 8420
},
{
"entropy": 5.247560644149781,
"epoch": 1.0035735556879095,
"grad_norm": 0.8984375,
"learning_rate": 0.0004911623587346434,
"loss": 4.8921,
"mean_token_accuracy": 0.2000837281346321,
"num_tokens": 23545040.0,
"step": 8425
},
{
"entropy": 5.14494366645813,
"epoch": 1.004169148302561,
"grad_norm": 1.0859375,
"learning_rate": 0.0004911505292560977,
"loss": 4.8442,
"mean_token_accuracy": 0.206904336810112,
"num_tokens": 23558019.0,
"step": 8430
},
{
"entropy": 5.137672567367554,
"epoch": 1.0047647409172127,
"grad_norm": 0.98046875,
"learning_rate": 0.0004911386920245837,
"loss": 4.8369,
"mean_token_accuracy": 0.20792090594768525,
"num_tokens": 23571263.0,
"step": 8435
},
{
"entropy": 5.183880567550659,
"epoch": 1.0053603335318642,
"grad_norm": 1.140625,
"learning_rate": 0.0004911268470405263,
"loss": 4.8757,
"mean_token_accuracy": 0.20961372703313827,
"num_tokens": 23584992.0,
"step": 8440
},
{
"entropy": 5.1023517608642575,
"epoch": 1.0059559261465159,
"grad_norm": 1.03125,
"learning_rate": 0.0004911149943043504,
"loss": 4.6953,
"mean_token_accuracy": 0.22323408275842666,
"num_tokens": 23597751.0,
"step": 8445
},
{
"entropy": 5.099691486358642,
"epoch": 1.0065515187611673,
"grad_norm": 0.890625,
"learning_rate": 0.000491103133816481,
"loss": 4.8495,
"mean_token_accuracy": 0.2088629722595215,
"num_tokens": 23613045.0,
"step": 8450
},
{
"entropy": 5.14667067527771,
"epoch": 1.007147111375819,
"grad_norm": 0.93359375,
"learning_rate": 0.0004910912655773436,
"loss": 4.8185,
"mean_token_accuracy": 0.2120829075574875,
"num_tokens": 23626785.0,
"step": 8455
},
{
"entropy": 5.168072271347046,
"epoch": 1.0077427039904705,
"grad_norm": 1.0078125,
"learning_rate": 0.0004910793895873637,
"loss": 4.7902,
"mean_token_accuracy": 0.2075516790151596,
"num_tokens": 23640496.0,
"step": 8460
},
{
"entropy": 5.19717321395874,
"epoch": 1.0083382966051222,
"grad_norm": 1.046875,
"learning_rate": 0.0004910675058469675,
"loss": 4.8406,
"mean_token_accuracy": 0.2068528339266777,
"num_tokens": 23652604.0,
"step": 8465
},
{
"entropy": 5.149213266372681,
"epoch": 1.0089338892197737,
"grad_norm": 0.93359375,
"learning_rate": 0.0004910556143565812,
"loss": 4.8177,
"mean_token_accuracy": 0.20693489611148835,
"num_tokens": 23666326.0,
"step": 8470
},
{
"entropy": 5.1137748718261715,
"epoch": 1.0095294818344251,
"grad_norm": 0.99609375,
"learning_rate": 0.0004910437151166312,
"loss": 4.7747,
"mean_token_accuracy": 0.20855310559272766,
"num_tokens": 23680728.0,
"step": 8475
},
{
"entropy": 5.099502611160278,
"epoch": 1.0101250744490768,
"grad_norm": 1.03125,
"learning_rate": 0.0004910318081275445,
"loss": 4.7274,
"mean_token_accuracy": 0.21316419243812562,
"num_tokens": 23694326.0,
"step": 8480
},
{
"entropy": 5.177095174789429,
"epoch": 1.0107206670637283,
"grad_norm": 1.078125,
"learning_rate": 0.000491019893389748,
"loss": 4.7771,
"mean_token_accuracy": 0.20100835859775543,
"num_tokens": 23709459.0,
"step": 8485
},
{
"entropy": 5.158708763122559,
"epoch": 1.01131625967838,
"grad_norm": 0.97265625,
"learning_rate": 0.0004910079709036692,
"loss": 4.8752,
"mean_token_accuracy": 0.20225850492715836,
"num_tokens": 23725227.0,
"step": 8490
},
{
"entropy": 5.185327005386353,
"epoch": 1.0119118522930315,
"grad_norm": 0.95703125,
"learning_rate": 0.0004909960406697358,
"loss": 4.9163,
"mean_token_accuracy": 0.20514773428440095,
"num_tokens": 23738733.0,
"step": 8495
},
{
"entropy": 5.174262380599975,
"epoch": 1.0125074449076832,
"grad_norm": 1.0234375,
"learning_rate": 0.0004909841026883755,
"loss": 4.865,
"mean_token_accuracy": 0.19892584681510925,
"num_tokens": 23753189.0,
"step": 8500
},
{
"entropy": 5.215628957748413,
"epoch": 1.0131030375223347,
"grad_norm": 1.1640625,
"learning_rate": 0.0004909721569600166,
"loss": 4.8691,
"mean_token_accuracy": 0.2083604544401169,
"num_tokens": 23766665.0,
"step": 8505
},
{
"entropy": 5.181106615066528,
"epoch": 1.0136986301369864,
"grad_norm": 1.0625,
"learning_rate": 0.0004909602034850875,
"loss": 4.7831,
"mean_token_accuracy": 0.21158466935157777,
"num_tokens": 23780190.0,
"step": 8510
},
{
"entropy": 5.101481294631958,
"epoch": 1.0142942227516378,
"grad_norm": 0.99609375,
"learning_rate": 0.0004909482422640172,
"loss": 4.7877,
"mean_token_accuracy": 0.20905089378356934,
"num_tokens": 23794652.0,
"step": 8515
},
{
"entropy": 5.152735185623169,
"epoch": 1.0148898153662895,
"grad_norm": 1.015625,
"learning_rate": 0.0004909362732972344,
"loss": 4.8177,
"mean_token_accuracy": 0.20504015684127808,
"num_tokens": 23810423.0,
"step": 8520
},
{
"entropy": 5.156676816940307,
"epoch": 1.015485407980941,
"grad_norm": 1.0234375,
"learning_rate": 0.0004909242965851688,
"loss": 4.8163,
"mean_token_accuracy": 0.21547921895980834,
"num_tokens": 23823676.0,
"step": 8525
},
{
"entropy": 5.112121200561523,
"epoch": 1.0160810005955927,
"grad_norm": 1.015625,
"learning_rate": 0.0004909123121282495,
"loss": 4.7713,
"mean_token_accuracy": 0.20795815587043762,
"num_tokens": 23836703.0,
"step": 8530
},
{
"entropy": 5.0185370445251465,
"epoch": 1.0166765932102442,
"grad_norm": 0.98046875,
"learning_rate": 0.0004909003199269068,
"loss": 4.6892,
"mean_token_accuracy": 0.2186439201235771,
"num_tokens": 23850804.0,
"step": 8535
},
{
"entropy": 5.209352302551269,
"epoch": 1.0172721858248959,
"grad_norm": 0.953125,
"learning_rate": 0.0004908883199815705,
"loss": 4.8602,
"mean_token_accuracy": 0.20209619849920274,
"num_tokens": 23864893.0,
"step": 8540
},
{
"entropy": 5.161143112182617,
"epoch": 1.0178677784395473,
"grad_norm": 0.99609375,
"learning_rate": 0.0004908763122926712,
"loss": 4.7735,
"mean_token_accuracy": 0.20437371581792832,
"num_tokens": 23878268.0,
"step": 8545
},
{
"entropy": 5.116016626358032,
"epoch": 1.018463371054199,
"grad_norm": 1.0390625,
"learning_rate": 0.0004908642968606396,
"loss": 4.7869,
"mean_token_accuracy": 0.21039541512727739,
"num_tokens": 23892398.0,
"step": 8550
},
{
"entropy": 5.07320647239685,
"epoch": 1.0190589636688505,
"grad_norm": 1.0546875,
"learning_rate": 0.0004908522736859068,
"loss": 4.7201,
"mean_token_accuracy": 0.21724511235952376,
"num_tokens": 23907268.0,
"step": 8555
},
{
"entropy": 5.171864366531372,
"epoch": 1.019654556283502,
"grad_norm": 0.96484375,
"learning_rate": 0.0004908402427689037,
"loss": 4.7863,
"mean_token_accuracy": 0.2086649715900421,
"num_tokens": 23921484.0,
"step": 8560
},
{
"entropy": 5.033004522323608,
"epoch": 1.0202501488981537,
"grad_norm": 1.0234375,
"learning_rate": 0.0004908282041100621,
"loss": 4.7435,
"mean_token_accuracy": 0.2135300427675247,
"num_tokens": 23934312.0,
"step": 8565
},
{
"entropy": 5.165141344070435,
"epoch": 1.0208457415128052,
"grad_norm": 0.984375,
"learning_rate": 0.0004908161577098137,
"loss": 4.8681,
"mean_token_accuracy": 0.20340251326560974,
"num_tokens": 23946866.0,
"step": 8570
},
{
"entropy": 5.152228260040284,
"epoch": 1.0214413341274569,
"grad_norm": 1.0,
"learning_rate": 0.0004908041035685907,
"loss": 4.8296,
"mean_token_accuracy": 0.2096054285764694,
"num_tokens": 23960582.0,
"step": 8575
},
{
"entropy": 5.162730169296265,
"epoch": 1.0220369267421083,
"grad_norm": 1.0234375,
"learning_rate": 0.0004907920416868253,
"loss": 4.8424,
"mean_token_accuracy": 0.2019661098718643,
"num_tokens": 23976130.0,
"step": 8580
},
{
"entropy": 5.195549058914184,
"epoch": 1.02263251935676,
"grad_norm": 0.93359375,
"learning_rate": 0.0004907799720649502,
"loss": 4.8882,
"mean_token_accuracy": 0.20378868132829667,
"num_tokens": 23989740.0,
"step": 8585
},
{
"entropy": 5.183404636383057,
"epoch": 1.0232281119714115,
"grad_norm": 0.91796875,
"learning_rate": 0.0004907678947033984,
"loss": 4.8461,
"mean_token_accuracy": 0.20923347771167755,
"num_tokens": 24005840.0,
"step": 8590
},
{
"entropy": 5.200975131988526,
"epoch": 1.0238237045860632,
"grad_norm": 1.03125,
"learning_rate": 0.0004907558096026031,
"loss": 4.8796,
"mean_token_accuracy": 0.2001400485634804,
"num_tokens": 24019861.0,
"step": 8595
},
{
"entropy": 5.150867223739624,
"epoch": 1.0244192972007147,
"grad_norm": 0.90625,
"learning_rate": 0.0004907437167629976,
"loss": 4.8043,
"mean_token_accuracy": 0.20841440558433533,
"num_tokens": 24034741.0,
"step": 8600
},
{
"entropy": 5.165017604827881,
"epoch": 1.0250148898153664,
"grad_norm": 0.8984375,
"learning_rate": 0.0004907316161850158,
"loss": 4.8594,
"mean_token_accuracy": 0.20150828063488008,
"num_tokens": 24048905.0,
"step": 8605
},
{
"entropy": 5.09525580406189,
"epoch": 1.0256104824300178,
"grad_norm": 0.88671875,
"learning_rate": 0.0004907195078690916,
"loss": 4.7774,
"mean_token_accuracy": 0.2068053141236305,
"num_tokens": 24064937.0,
"step": 8610
},
{
"entropy": 5.107802677154541,
"epoch": 1.0262060750446695,
"grad_norm": 0.96484375,
"learning_rate": 0.0004907073918156595,
"loss": 4.7826,
"mean_token_accuracy": 0.20779221504926682,
"num_tokens": 24080293.0,
"step": 8615
},
{
"entropy": 5.258295392990112,
"epoch": 1.026801667659321,
"grad_norm": 1.0625,
"learning_rate": 0.0004906952680251539,
"loss": 4.8753,
"mean_token_accuracy": 0.20227858424186707,
"num_tokens": 24092048.0,
"step": 8620
},
{
"entropy": 5.1107954502105715,
"epoch": 1.0273972602739727,
"grad_norm": 0.96875,
"learning_rate": 0.0004906831364980098,
"loss": 4.7743,
"mean_token_accuracy": 0.20843288749456407,
"num_tokens": 24106446.0,
"step": 8625
},
{
"entropy": 5.092709684371949,
"epoch": 1.0279928528886242,
"grad_norm": 0.94921875,
"learning_rate": 0.0004906709972346623,
"loss": 4.8371,
"mean_token_accuracy": 0.2080872967839241,
"num_tokens": 24120867.0,
"step": 8630
},
{
"entropy": 5.153543424606323,
"epoch": 1.0285884455032757,
"grad_norm": 0.95703125,
"learning_rate": 0.0004906588502355467,
"loss": 4.7987,
"mean_token_accuracy": 0.2104442834854126,
"num_tokens": 24135139.0,
"step": 8635
},
{
"entropy": 5.073699283599853,
"epoch": 1.0291840381179274,
"grad_norm": 1.1171875,
"learning_rate": 0.0004906466955010988,
"loss": 4.7165,
"mean_token_accuracy": 0.21477803885936736,
"num_tokens": 24147649.0,
"step": 8640
},
{
"entropy": 5.083175992965698,
"epoch": 1.0297796307325788,
"grad_norm": 1.1171875,
"learning_rate": 0.0004906345330317546,
"loss": 4.744,
"mean_token_accuracy": 0.21306189894676208,
"num_tokens": 24162408.0,
"step": 8645
},
{
"entropy": 5.167844915390015,
"epoch": 1.0303752233472305,
"grad_norm": 1.0078125,
"learning_rate": 0.0004906223628279502,
"loss": 4.7655,
"mean_token_accuracy": 0.21247482299804688,
"num_tokens": 24175011.0,
"step": 8650
},
{
"entropy": 5.153194427490234,
"epoch": 1.030970815961882,
"grad_norm": 1.0625,
"learning_rate": 0.0004906101848901224,
"loss": 4.7833,
"mean_token_accuracy": 0.2079053893685341,
"num_tokens": 24187832.0,
"step": 8655
},
{
"entropy": 5.137058591842651,
"epoch": 1.0315664085765337,
"grad_norm": 0.953125,
"learning_rate": 0.0004905979992187076,
"loss": 4.8294,
"mean_token_accuracy": 0.20494291931390762,
"num_tokens": 24202347.0,
"step": 8660
},
{
"entropy": 5.141408538818359,
"epoch": 1.0321620011911852,
"grad_norm": 0.9921875,
"learning_rate": 0.0004905858058141433,
"loss": 4.8124,
"mean_token_accuracy": 0.21434862166643143,
"num_tokens": 24216988.0,
"step": 8665
},
{
"entropy": 5.168860006332397,
"epoch": 1.0327575938058369,
"grad_norm": 1.140625,
"learning_rate": 0.0004905736046768665,
"loss": 4.8535,
"mean_token_accuracy": 0.20405864864587783,
"num_tokens": 24231091.0,
"step": 8670
},
{
"entropy": 5.12582221031189,
"epoch": 1.0333531864204883,
"grad_norm": 1.03125,
"learning_rate": 0.0004905613958073151,
"loss": 4.7572,
"mean_token_accuracy": 0.2139848694205284,
"num_tokens": 24244684.0,
"step": 8675
},
{
"entropy": 5.044980049133301,
"epoch": 1.03394877903514,
"grad_norm": 1.0078125,
"learning_rate": 0.000490549179205927,
"loss": 4.7519,
"mean_token_accuracy": 0.21736170202493668,
"num_tokens": 24259637.0,
"step": 8680
},
{
"entropy": 5.1405863761901855,
"epoch": 1.0345443716497915,
"grad_norm": 0.99609375,
"learning_rate": 0.0004905369548731402,
"loss": 4.8632,
"mean_token_accuracy": 0.20024951547384262,
"num_tokens": 24274046.0,
"step": 8685
},
{
"entropy": 5.18059720993042,
"epoch": 1.0351399642644432,
"grad_norm": 1.03125,
"learning_rate": 0.0004905247228093933,
"loss": 4.8958,
"mean_token_accuracy": 0.20295093059539795,
"num_tokens": 24288266.0,
"step": 8690
},
{
"entropy": 5.125336313247681,
"epoch": 1.0357355568790947,
"grad_norm": 0.98046875,
"learning_rate": 0.000490512483015125,
"loss": 4.8774,
"mean_token_accuracy": 0.19899014830589296,
"num_tokens": 24303029.0,
"step": 8695
},
{
"entropy": 5.040747928619385,
"epoch": 1.0363311494937464,
"grad_norm": 1.1171875,
"learning_rate": 0.0004905002354907743,
"loss": 4.6577,
"mean_token_accuracy": 0.22837863713502884,
"num_tokens": 24315369.0,
"step": 8700
},
{
"entropy": 5.256495523452759,
"epoch": 1.0369267421083979,
"grad_norm": 1.09375,
"learning_rate": 0.0004904879802367806,
"loss": 4.8536,
"mean_token_accuracy": 0.2029978320002556,
"num_tokens": 24327049.0,
"step": 8705
},
{
"entropy": 5.231810855865478,
"epoch": 1.0375223347230493,
"grad_norm": 1.0078125,
"learning_rate": 0.0004904757172535833,
"loss": 4.8418,
"mean_token_accuracy": 0.20405835062265396,
"num_tokens": 24341797.0,
"step": 8710
},
{
"entropy": 5.106027030944825,
"epoch": 1.038117927337701,
"grad_norm": 1.078125,
"learning_rate": 0.0004904634465416226,
"loss": 4.8676,
"mean_token_accuracy": 0.203257218003273,
"num_tokens": 24357560.0,
"step": 8715
},
{
"entropy": 5.1268998146057125,
"epoch": 1.0387135199523525,
"grad_norm": 1.0390625,
"learning_rate": 0.0004904511681013382,
"loss": 4.7874,
"mean_token_accuracy": 0.20785263031721116,
"num_tokens": 24371856.0,
"step": 8720
},
{
"entropy": 5.188601636886597,
"epoch": 1.0393091125670042,
"grad_norm": 1.0390625,
"learning_rate": 0.0004904388819331707,
"loss": 4.8371,
"mean_token_accuracy": 0.2094535320997238,
"num_tokens": 24384494.0,
"step": 8725
},
{
"entropy": 5.24585599899292,
"epoch": 1.0399047051816557,
"grad_norm": 1.125,
"learning_rate": 0.0004904265880375609,
"loss": 4.8778,
"mean_token_accuracy": 0.2027064800262451,
"num_tokens": 24398901.0,
"step": 8730
},
{
"entropy": 5.12142767906189,
"epoch": 1.0405002977963074,
"grad_norm": 0.97265625,
"learning_rate": 0.0004904142864149496,
"loss": 4.7959,
"mean_token_accuracy": 0.20762286335229874,
"num_tokens": 24413900.0,
"step": 8735
},
{
"entropy": 5.148635578155518,
"epoch": 1.0410958904109588,
"grad_norm": 1.0546875,
"learning_rate": 0.000490401977065778,
"loss": 4.8649,
"mean_token_accuracy": 0.20714886635541915,
"num_tokens": 24427764.0,
"step": 8740
},
{
"entropy": 5.1255059242248535,
"epoch": 1.0416914830256105,
"grad_norm": 1.03125,
"learning_rate": 0.0004903896599904878,
"loss": 4.7831,
"mean_token_accuracy": 0.20894113034009934,
"num_tokens": 24442510.0,
"step": 8745
},
{
"entropy": 5.2562402248382565,
"epoch": 1.042287075640262,
"grad_norm": 1.046875,
"learning_rate": 0.0004903773351895206,
"loss": 4.925,
"mean_token_accuracy": 0.20110782980918884,
"num_tokens": 24457119.0,
"step": 8750
},
{
"entropy": 5.177147769927979,
"epoch": 1.0428826682549137,
"grad_norm": 1.0,
"learning_rate": 0.0004903650026633187,
"loss": 4.8325,
"mean_token_accuracy": 0.20287259817123413,
"num_tokens": 24470232.0,
"step": 8755
},
{
"entropy": 5.102942895889282,
"epoch": 1.0434782608695652,
"grad_norm": 1.0078125,
"learning_rate": 0.0004903526624123243,
"loss": 4.8206,
"mean_token_accuracy": 0.2104717582464218,
"num_tokens": 24484886.0,
"step": 8760
},
{
"entropy": 5.146371698379516,
"epoch": 1.0440738534842169,
"grad_norm": 0.93359375,
"learning_rate": 0.00049034031443698,
"loss": 4.8193,
"mean_token_accuracy": 0.2068724438548088,
"num_tokens": 24499682.0,
"step": 8765
},
{
"entropy": 5.16482572555542,
"epoch": 1.0446694460988684,
"grad_norm": 0.92578125,
"learning_rate": 0.0004903279587377287,
"loss": 4.8406,
"mean_token_accuracy": 0.21020206958055496,
"num_tokens": 24514812.0,
"step": 8770
},
{
"entropy": 5.176180839538574,
"epoch": 1.04526503871352,
"grad_norm": 1.1015625,
"learning_rate": 0.0004903155953150137,
"loss": 4.807,
"mean_token_accuracy": 0.20769522488117217,
"num_tokens": 24527723.0,
"step": 8775
},
{
"entropy": 5.127501392364502,
"epoch": 1.0458606313281715,
"grad_norm": 1.0234375,
"learning_rate": 0.0004903032241692784,
"loss": 4.7819,
"mean_token_accuracy": 0.21579769551753997,
"num_tokens": 24542082.0,
"step": 8780
},
{
"entropy": 5.1282562732696535,
"epoch": 1.046456223942823,
"grad_norm": 1.0078125,
"learning_rate": 0.0004902908453009664,
"loss": 4.7801,
"mean_token_accuracy": 0.20459772795438766,
"num_tokens": 24556854.0,
"step": 8785
},
{
"entropy": 5.04892783164978,
"epoch": 1.0470518165574747,
"grad_norm": 0.94921875,
"learning_rate": 0.0004902784587105218,
"loss": 4.6775,
"mean_token_accuracy": 0.2182971566915512,
"num_tokens": 24570927.0,
"step": 8790
},
{
"entropy": 5.127280330657959,
"epoch": 1.0476474091721262,
"grad_norm": 1.1171875,
"learning_rate": 0.000490266064398389,
"loss": 4.7864,
"mean_token_accuracy": 0.21088114082813264,
"num_tokens": 24584458.0,
"step": 8795
},
{
"entropy": 5.106057977676391,
"epoch": 1.0482430017867779,
"grad_norm": 0.96484375,
"learning_rate": 0.0004902536623650125,
"loss": 4.7339,
"mean_token_accuracy": 0.21336031556129456,
"num_tokens": 24597576.0,
"step": 8800
},
{
"entropy": 5.055494117736816,
"epoch": 1.0488385944014293,
"grad_norm": 0.859375,
"learning_rate": 0.0004902412526108371,
"loss": 4.7419,
"mean_token_accuracy": 0.21380414366722106,
"num_tokens": 24613022.0,
"step": 8805
},
{
"entropy": 5.2035492897033695,
"epoch": 1.049434187016081,
"grad_norm": 0.96484375,
"learning_rate": 0.0004902288351363079,
"loss": 4.8564,
"mean_token_accuracy": 0.20787192434072493,
"num_tokens": 24627730.0,
"step": 8810
},
{
"entropy": 5.075625085830689,
"epoch": 1.0500297796307325,
"grad_norm": 1.0078125,
"learning_rate": 0.0004902164099418703,
"loss": 4.7491,
"mean_token_accuracy": 0.21297594010829926,
"num_tokens": 24642870.0,
"step": 8815
},
{
"entropy": 5.212960577011108,
"epoch": 1.0506253722453842,
"grad_norm": 0.97265625,
"learning_rate": 0.00049020397702797,
"loss": 4.9443,
"mean_token_accuracy": 0.19699231535196304,
"num_tokens": 24658416.0,
"step": 8820
},
{
"entropy": 5.134889125823975,
"epoch": 1.0512209648600357,
"grad_norm": 0.984375,
"learning_rate": 0.000490191536395053,
"loss": 4.8338,
"mean_token_accuracy": 0.20598364919424056,
"num_tokens": 24673294.0,
"step": 8825
},
{
"entropy": 5.028152561187744,
"epoch": 1.0518165574746874,
"grad_norm": 0.98046875,
"learning_rate": 0.0004901790880435655,
"loss": 4.7712,
"mean_token_accuracy": 0.2094384029507637,
"num_tokens": 24687841.0,
"step": 8830
},
{
"entropy": 5.133012533187866,
"epoch": 1.0524121500893389,
"grad_norm": 1.0859375,
"learning_rate": 0.0004901666319739537,
"loss": 4.6939,
"mean_token_accuracy": 0.21305721551179885,
"num_tokens": 24701489.0,
"step": 8835
},
{
"entropy": 5.121408987045288,
"epoch": 1.0530077427039906,
"grad_norm": 1.015625,
"learning_rate": 0.0004901541681866649,
"loss": 4.7682,
"mean_token_accuracy": 0.2109750837087631,
"num_tokens": 24714312.0,
"step": 8840
},
{
"entropy": 5.1196364402771,
"epoch": 1.053603335318642,
"grad_norm": 1.046875,
"learning_rate": 0.0004901416966821459,
"loss": 4.7745,
"mean_token_accuracy": 0.21171149015426635,
"num_tokens": 24728982.0,
"step": 8845
},
{
"entropy": 5.133532953262329,
"epoch": 1.0541989279332937,
"grad_norm": 0.91796875,
"learning_rate": 0.000490129217460844,
"loss": 4.808,
"mean_token_accuracy": 0.21003280133008956,
"num_tokens": 24744374.0,
"step": 8850
},
{
"entropy": 5.210978841781616,
"epoch": 1.0547945205479452,
"grad_norm": 1.0234375,
"learning_rate": 0.0004901167305232068,
"loss": 4.9448,
"mean_token_accuracy": 0.19654947519302368,
"num_tokens": 24759129.0,
"step": 8855
},
{
"entropy": 5.158556175231934,
"epoch": 1.0553901131625967,
"grad_norm": 1.03125,
"learning_rate": 0.0004901042358696823,
"loss": 4.8251,
"mean_token_accuracy": 0.20848578214645386,
"num_tokens": 24772154.0,
"step": 8860
},
{
"entropy": 5.088905096054077,
"epoch": 1.0559857057772484,
"grad_norm": 0.98046875,
"learning_rate": 0.0004900917335007186,
"loss": 4.7257,
"mean_token_accuracy": 0.21268839240074158,
"num_tokens": 24785961.0,
"step": 8865
},
{
"entropy": 5.140409278869629,
"epoch": 1.0565812983918998,
"grad_norm": 1.265625,
"learning_rate": 0.0004900792234167642,
"loss": 4.8192,
"mean_token_accuracy": 0.21458131819963455,
"num_tokens": 24797337.0,
"step": 8870
},
{
"entropy": 5.194036436080933,
"epoch": 1.0571768910065515,
"grad_norm": 1.0234375,
"learning_rate": 0.0004900667056182676,
"loss": 4.9077,
"mean_token_accuracy": 0.20716588348150253,
"num_tokens": 24811023.0,
"step": 8875
},
{
"entropy": 5.186428165435791,
"epoch": 1.057772483621203,
"grad_norm": 0.97265625,
"learning_rate": 0.0004900541801056781,
"loss": 4.8577,
"mean_token_accuracy": 0.2032381683588028,
"num_tokens": 24823951.0,
"step": 8880
},
{
"entropy": 5.097773218154908,
"epoch": 1.0583680762358547,
"grad_norm": 1.0390625,
"learning_rate": 0.0004900416468794446,
"loss": 4.7852,
"mean_token_accuracy": 0.21291813403367996,
"num_tokens": 24838856.0,
"step": 8885
},
{
"entropy": 5.131308841705322,
"epoch": 1.0589636688505062,
"grad_norm": 1.0,
"learning_rate": 0.0004900291059400171,
"loss": 4.8731,
"mean_token_accuracy": 0.2068126916885376,
"num_tokens": 24853539.0,
"step": 8890
},
{
"entropy": 5.194414806365967,
"epoch": 1.0595592614651579,
"grad_norm": 1.0625,
"learning_rate": 0.000490016557287845,
"loss": 4.759,
"mean_token_accuracy": 0.21602594703435898,
"num_tokens": 24866385.0,
"step": 8895
},
{
"entropy": 5.177035284042359,
"epoch": 1.0601548540798094,
"grad_norm": 1.0546875,
"learning_rate": 0.0004900040009233787,
"loss": 4.8421,
"mean_token_accuracy": 0.2012791320681572,
"num_tokens": 24880168.0,
"step": 8900
},
{
"entropy": 5.119781827926635,
"epoch": 1.060750446694461,
"grad_norm": 1.0390625,
"learning_rate": 0.0004899914368470685,
"loss": 4.8111,
"mean_token_accuracy": 0.20698730200529097,
"num_tokens": 24893673.0,
"step": 8905
},
{
"entropy": 5.1871910572052,
"epoch": 1.0613460393091125,
"grad_norm": 1.03125,
"learning_rate": 0.0004899788650593648,
"loss": 4.8483,
"mean_token_accuracy": 0.21000517159700394,
"num_tokens": 24907950.0,
"step": 8910
},
{
"entropy": 5.155257797241211,
"epoch": 1.0619416319237642,
"grad_norm": 0.95703125,
"learning_rate": 0.0004899662855607189,
"loss": 4.7629,
"mean_token_accuracy": 0.21739620119333267,
"num_tokens": 24921443.0,
"step": 8915
},
{
"entropy": 5.120499086380005,
"epoch": 1.0625372245384157,
"grad_norm": 1.046875,
"learning_rate": 0.0004899536983515819,
"loss": 4.807,
"mean_token_accuracy": 0.20788541585206985,
"num_tokens": 24935485.0,
"step": 8920
},
{
"entropy": 5.156124830245972,
"epoch": 1.0631328171530674,
"grad_norm": 0.9765625,
"learning_rate": 0.0004899411034324051,
"loss": 4.8106,
"mean_token_accuracy": 0.20609290450811385,
"num_tokens": 24950372.0,
"step": 8925
},
{
"entropy": 5.083027744293213,
"epoch": 1.0637284097677189,
"grad_norm": 0.95703125,
"learning_rate": 0.0004899285008036406,
"loss": 4.7803,
"mean_token_accuracy": 0.20988123416900634,
"num_tokens": 24964360.0,
"step": 8930
},
{
"entropy": 5.089720582962036,
"epoch": 1.0643240023823703,
"grad_norm": 0.9296875,
"learning_rate": 0.00048991589046574,
"loss": 4.7667,
"mean_token_accuracy": 0.21860623806715013,
"num_tokens": 24979101.0,
"step": 8935
},
{
"entropy": 5.169832038879394,
"epoch": 1.064919594997022,
"grad_norm": 1.109375,
"learning_rate": 0.0004899032724191559,
"loss": 4.8292,
"mean_token_accuracy": 0.20461245477199555,
"num_tokens": 24994350.0,
"step": 8940
},
{
"entropy": 5.0844746112823485,
"epoch": 1.0655151876116735,
"grad_norm": 1.015625,
"learning_rate": 0.0004898906466643409,
"loss": 4.7359,
"mean_token_accuracy": 0.2128108784556389,
"num_tokens": 25007343.0,
"step": 8945
},
{
"entropy": 5.216156244277954,
"epoch": 1.0661107802263252,
"grad_norm": 1.0625,
"learning_rate": 0.0004898780132017478,
"loss": 4.9009,
"mean_token_accuracy": 0.2007828950881958,
"num_tokens": 25020191.0,
"step": 8950
},
{
"entropy": 5.110795736312866,
"epoch": 1.0667063728409767,
"grad_norm": 0.90625,
"learning_rate": 0.0004898653720318296,
"loss": 4.8243,
"mean_token_accuracy": 0.20330047756433486,
"num_tokens": 25034910.0,
"step": 8955
},
{
"entropy": 5.146774768829346,
"epoch": 1.0673019654556284,
"grad_norm": 1.1171875,
"learning_rate": 0.00048985272315504,
"loss": 4.8091,
"mean_token_accuracy": 0.21246378421783446,
"num_tokens": 25048958.0,
"step": 8960
},
{
"entropy": 5.197678852081299,
"epoch": 1.0678975580702799,
"grad_norm": 0.97265625,
"learning_rate": 0.0004898400665718325,
"loss": 4.9048,
"mean_token_accuracy": 0.2010442927479744,
"num_tokens": 25065752.0,
"step": 8965
},
{
"entropy": 5.134333610534668,
"epoch": 1.0684931506849316,
"grad_norm": 1.0625,
"learning_rate": 0.0004898274022826611,
"loss": 4.7413,
"mean_token_accuracy": 0.2209034815430641,
"num_tokens": 25078569.0,
"step": 8970
},
{
"entropy": 5.112339019775391,
"epoch": 1.069088743299583,
"grad_norm": 0.93359375,
"learning_rate": 0.0004898147302879801,
"loss": 4.8181,
"mean_token_accuracy": 0.213752081990242,
"num_tokens": 25093973.0,
"step": 8975
},
{
"entropy": 5.128629684448242,
"epoch": 1.0696843359142347,
"grad_norm": 0.9921875,
"learning_rate": 0.0004898020505882441,
"loss": 4.7317,
"mean_token_accuracy": 0.21151299625635148,
"num_tokens": 25109338.0,
"step": 8980
},
{
"entropy": 5.107893753051758,
"epoch": 1.0702799285288862,
"grad_norm": 1.015625,
"learning_rate": 0.0004897893631839076,
"loss": 4.8149,
"mean_token_accuracy": 0.2082427427172661,
"num_tokens": 25124902.0,
"step": 8985
},
{
"entropy": 5.1293683528900145,
"epoch": 1.070875521143538,
"grad_norm": 0.953125,
"learning_rate": 0.000489776668075426,
"loss": 4.7661,
"mean_token_accuracy": 0.21519884169101716,
"num_tokens": 25138982.0,
"step": 8990
},
{
"entropy": 5.135135841369629,
"epoch": 1.0714711137581894,
"grad_norm": 1.0,
"learning_rate": 0.0004897639652632544,
"loss": 4.7886,
"mean_token_accuracy": 0.2178763598203659,
"num_tokens": 25152684.0,
"step": 8995
},
{
"entropy": 5.124427032470703,
"epoch": 1.072066706372841,
"grad_norm": 1.0078125,
"learning_rate": 0.0004897512547478487,
"loss": 4.7903,
"mean_token_accuracy": 0.21109549850225448,
"num_tokens": 25167275.0,
"step": 9000
},
{
"epoch": 1.072066706372841,
"eval_entropy": 4.913777104739485,
"eval_loss": 4.829866886138916,
"eval_mean_token_accuracy": 0.21852801226686722,
"eval_num_tokens": 25167275.0,
"eval_runtime": 18.5427,
"eval_samples_per_second": 1375.908,
"eval_steps_per_second": 172.036,
"step": 9000
}
],
"logging_steps": 5,
"max_steps": 83940,
"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": 6651974257090560.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink