fpadovani/jpn-jpan-10mb-100mb_seed10
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
jpn-jpan-10mb-100mb_seed10/checkpoint-9000/trainer_state.json
ModelHub XC 9975275e05 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/jpn-jpan-10mb-100mb_seed10
Source: Original Platform
2026-09-28 08:11:19 +08:00

18068 lines
496 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.03866128101558,
"eval_steps": 3000,
"global_step": 9000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 10.74217529296875,
"epoch": 0.0005770340450086555,
"grad_norm": 4.34375,
"learning_rate": 2e-06,
"loss": 10.8326,
"mean_token_accuracy": 0.0,
"num_tokens": 14330.0,
"step": 5
},
{
"entropy": 10.742151832580566,
"epoch": 0.001154068090017311,
"grad_norm": 4.65625,
"learning_rate": 4.5e-06,
"loss": 10.8057,
"mean_token_accuracy": 8.11688310932368e-05,
"num_tokens": 26284.0,
"step": 10
},
{
"entropy": 10.742152214050293,
"epoch": 0.0017311021350259665,
"grad_norm": 5.125,
"learning_rate": 7e-06,
"loss": 10.7851,
"mean_token_accuracy": 0.00030720371869392694,
"num_tokens": 38394.0,
"step": 15
},
{
"entropy": 10.742230319976807,
"epoch": 0.002308136180034622,
"grad_norm": 4.8125,
"learning_rate": 9.5e-06,
"loss": 10.7433,
"mean_token_accuracy": 0.0,
"num_tokens": 51341.0,
"step": 20
},
{
"entropy": 10.742213916778564,
"epoch": 0.0028851702250432777,
"grad_norm": 4.4375,
"learning_rate": 1.2e-05,
"loss": 10.64,
"mean_token_accuracy": 0.003926298324950039,
"num_tokens": 64559.0,
"step": 25
},
{
"entropy": 10.741898727416991,
"epoch": 0.003462204270051933,
"grad_norm": 3.921875,
"learning_rate": 1.4500000000000002e-05,
"loss": 10.5216,
"mean_token_accuracy": 0.03795875590294599,
"num_tokens": 76688.0,
"step": 30
},
{
"entropy": 10.740620136260986,
"epoch": 0.004039238315060588,
"grad_norm": 3.46875,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.3693,
"mean_token_accuracy": 0.05394260324537754,
"num_tokens": 88392.0,
"step": 35
},
{
"entropy": 10.736398124694825,
"epoch": 0.004616272360069244,
"grad_norm": 2.578125,
"learning_rate": 1.95e-05,
"loss": 10.278,
"mean_token_accuracy": 0.052969465777277944,
"num_tokens": 100542.0,
"step": 40
},
{
"entropy": 10.728193473815917,
"epoch": 0.0051933064050779,
"grad_norm": 2.34375,
"learning_rate": 2.2e-05,
"loss": 10.0828,
"mean_token_accuracy": 0.05751506164669991,
"num_tokens": 113190.0,
"step": 45
},
{
"entropy": 10.716622734069825,
"epoch": 0.005770340450086555,
"grad_norm": 2.109375,
"learning_rate": 2.4500000000000003e-05,
"loss": 10.0239,
"mean_token_accuracy": 0.05997682772576809,
"num_tokens": 125554.0,
"step": 50
},
{
"entropy": 10.704752922058105,
"epoch": 0.006347374495095211,
"grad_norm": 1.9765625,
"learning_rate": 2.7e-05,
"loss": 9.9661,
"mean_token_accuracy": 0.06483750529587269,
"num_tokens": 138527.0,
"step": 55
},
{
"entropy": 10.696050262451172,
"epoch": 0.006924408540103866,
"grad_norm": 1.9375,
"learning_rate": 2.95e-05,
"loss": 10.0065,
"mean_token_accuracy": 0.0495222195982933,
"num_tokens": 150404.0,
"step": 60
},
{
"entropy": 10.69492769241333,
"epoch": 0.0075014425851125215,
"grad_norm": 1.859375,
"learning_rate": 3.2e-05,
"loss": 9.9359,
"mean_token_accuracy": 0.055068480968475345,
"num_tokens": 163743.0,
"step": 65
},
{
"entropy": 10.692989349365234,
"epoch": 0.008078476630121177,
"grad_norm": 1.8671875,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.9027,
"mean_token_accuracy": 0.052650950476527215,
"num_tokens": 175657.0,
"step": 70
},
{
"entropy": 10.689637851715087,
"epoch": 0.008655510675129832,
"grad_norm": 1.828125,
"learning_rate": 3.7e-05,
"loss": 9.8572,
"mean_token_accuracy": 0.053944576904177664,
"num_tokens": 188756.0,
"step": 75
},
{
"entropy": 10.678643798828125,
"epoch": 0.009232544720138488,
"grad_norm": 1.8984375,
"learning_rate": 3.95e-05,
"loss": 9.801,
"mean_token_accuracy": 0.053839181736111644,
"num_tokens": 200548.0,
"step": 80
},
{
"entropy": 10.661964607238769,
"epoch": 0.009809578765147143,
"grad_norm": 1.859375,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.7136,
"mean_token_accuracy": 0.06665473207831382,
"num_tokens": 213307.0,
"step": 85
},
{
"entropy": 10.63293104171753,
"epoch": 0.0103866128101558,
"grad_norm": 1.7890625,
"learning_rate": 4.45e-05,
"loss": 9.6765,
"mean_token_accuracy": 0.06476713679730892,
"num_tokens": 226863.0,
"step": 90
},
{
"entropy": 10.615675067901611,
"epoch": 0.010963646855164455,
"grad_norm": 1.6015625,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.632,
"mean_token_accuracy": 0.06121704056859016,
"num_tokens": 238655.0,
"step": 95
},
{
"entropy": 10.613676834106446,
"epoch": 0.01154068090017311,
"grad_norm": 1.734375,
"learning_rate": 4.9500000000000004e-05,
"loss": 9.5655,
"mean_token_accuracy": 0.0704349122941494,
"num_tokens": 251118.0,
"step": 100
},
{
"entropy": 10.56223793029785,
"epoch": 0.012117714945181766,
"grad_norm": 1.953125,
"learning_rate": 5.2e-05,
"loss": 9.4708,
"mean_token_accuracy": 0.07793616428971291,
"num_tokens": 263665.0,
"step": 105
},
{
"entropy": 10.50258274078369,
"epoch": 0.012694748990190421,
"grad_norm": 1.6796875,
"learning_rate": 5.45e-05,
"loss": 9.4156,
"mean_token_accuracy": 0.07520635947585105,
"num_tokens": 276531.0,
"step": 110
},
{
"entropy": 10.506224632263184,
"epoch": 0.013271783035199077,
"grad_norm": 1.5625,
"learning_rate": 5.7e-05,
"loss": 9.367,
"mean_token_accuracy": 0.0757908646017313,
"num_tokens": 289878.0,
"step": 115
},
{
"entropy": 10.472508811950684,
"epoch": 0.013848817080207732,
"grad_norm": 1.5546875,
"learning_rate": 5.9499999999999996e-05,
"loss": 9.2817,
"mean_token_accuracy": 0.07052403129637241,
"num_tokens": 302705.0,
"step": 120
},
{
"entropy": 10.418921184539794,
"epoch": 0.014425851125216388,
"grad_norm": 1.59375,
"learning_rate": 6.2e-05,
"loss": 9.1572,
"mean_token_accuracy": 0.08047313243150711,
"num_tokens": 314775.0,
"step": 125
},
{
"entropy": 10.330512046813965,
"epoch": 0.015002885170225043,
"grad_norm": 1.2890625,
"learning_rate": 6.450000000000001e-05,
"loss": 9.1596,
"mean_token_accuracy": 0.07958029508590699,
"num_tokens": 328563.0,
"step": 130
},
{
"entropy": 10.242788600921632,
"epoch": 0.015579919215233698,
"grad_norm": 1.3984375,
"learning_rate": 6.7e-05,
"loss": 9.0052,
"mean_token_accuracy": 0.08549408987164497,
"num_tokens": 341820.0,
"step": 135
},
{
"entropy": 10.257353591918946,
"epoch": 0.016156953260242354,
"grad_norm": 1.34375,
"learning_rate": 6.950000000000001e-05,
"loss": 8.9633,
"mean_token_accuracy": 0.08436901345849038,
"num_tokens": 354994.0,
"step": 140
},
{
"entropy": 10.1291428565979,
"epoch": 0.01673398730525101,
"grad_norm": 1.203125,
"learning_rate": 7.2e-05,
"loss": 8.8768,
"mean_token_accuracy": 0.08217294588685035,
"num_tokens": 365852.0,
"step": 145
},
{
"entropy": 10.04762372970581,
"epoch": 0.017311021350259664,
"grad_norm": 1.4609375,
"learning_rate": 7.45e-05,
"loss": 8.8657,
"mean_token_accuracy": 0.07920254692435265,
"num_tokens": 379147.0,
"step": 150
},
{
"entropy": 10.023890781402589,
"epoch": 0.01788805539526832,
"grad_norm": 1.3359375,
"learning_rate": 7.7e-05,
"loss": 8.8181,
"mean_token_accuracy": 0.0735938459634781,
"num_tokens": 392257.0,
"step": 155
},
{
"entropy": 9.896601295471191,
"epoch": 0.018465089440276975,
"grad_norm": 1.25,
"learning_rate": 7.950000000000001e-05,
"loss": 8.7072,
"mean_token_accuracy": 0.07832617908716202,
"num_tokens": 405249.0,
"step": 160
},
{
"entropy": 9.709205722808838,
"epoch": 0.019042123485285632,
"grad_norm": 0.9765625,
"learning_rate": 8.2e-05,
"loss": 8.6516,
"mean_token_accuracy": 0.08040346056222916,
"num_tokens": 417819.0,
"step": 165
},
{
"entropy": 9.60182809829712,
"epoch": 0.019619157530294286,
"grad_norm": 0.98828125,
"learning_rate": 8.450000000000001e-05,
"loss": 8.5351,
"mean_token_accuracy": 0.08392796665430069,
"num_tokens": 429364.0,
"step": 170
},
{
"entropy": 9.431674385070801,
"epoch": 0.020196191575302943,
"grad_norm": 1.140625,
"learning_rate": 8.7e-05,
"loss": 8.5151,
"mean_token_accuracy": 0.08911501616239548,
"num_tokens": 441804.0,
"step": 175
},
{
"entropy": 9.291469860076905,
"epoch": 0.0207732256203116,
"grad_norm": 1.140625,
"learning_rate": 8.95e-05,
"loss": 8.4571,
"mean_token_accuracy": 0.08912580311298371,
"num_tokens": 453886.0,
"step": 180
},
{
"entropy": 9.148678588867188,
"epoch": 0.021350259665320254,
"grad_norm": 0.89453125,
"learning_rate": 9.2e-05,
"loss": 8.4362,
"mean_token_accuracy": 0.08946132510900498,
"num_tokens": 465828.0,
"step": 185
},
{
"entropy": 9.080477714538574,
"epoch": 0.02192729371032891,
"grad_norm": 0.796875,
"learning_rate": 9.45e-05,
"loss": 8.3788,
"mean_token_accuracy": 0.09273218810558319,
"num_tokens": 478783.0,
"step": 190
},
{
"entropy": 8.97773780822754,
"epoch": 0.022504327755337564,
"grad_norm": 1.0625,
"learning_rate": 9.7e-05,
"loss": 8.4168,
"mean_token_accuracy": 0.08846393302083015,
"num_tokens": 491167.0,
"step": 195
},
{
"entropy": 8.89993076324463,
"epoch": 0.02308136180034622,
"grad_norm": 1.3359375,
"learning_rate": 9.95e-05,
"loss": 8.3844,
"mean_token_accuracy": 0.09320372641086579,
"num_tokens": 503511.0,
"step": 200
},
{
"entropy": 8.787849521636963,
"epoch": 0.023658395845354875,
"grad_norm": 0.89453125,
"learning_rate": 0.000102,
"loss": 8.2705,
"mean_token_accuracy": 0.0978940948843956,
"num_tokens": 515816.0,
"step": 205
},
{
"entropy": 8.848995685577393,
"epoch": 0.024235429890363532,
"grad_norm": 0.921875,
"learning_rate": 0.00010449999999999999,
"loss": 8.3781,
"mean_token_accuracy": 0.08895341083407401,
"num_tokens": 528919.0,
"step": 210
},
{
"entropy": 8.744617557525634,
"epoch": 0.024812463935372186,
"grad_norm": 0.8828125,
"learning_rate": 0.000107,
"loss": 8.2875,
"mean_token_accuracy": 0.10044435858726501,
"num_tokens": 540973.0,
"step": 215
},
{
"entropy": 8.707606983184814,
"epoch": 0.025389497980380843,
"grad_norm": 0.90234375,
"learning_rate": 0.0001095,
"loss": 8.2458,
"mean_token_accuracy": 0.10567026063799859,
"num_tokens": 553265.0,
"step": 220
},
{
"entropy": 8.753429126739501,
"epoch": 0.025966532025389497,
"grad_norm": 0.83203125,
"learning_rate": 0.000112,
"loss": 8.3474,
"mean_token_accuracy": 0.0927657037973404,
"num_tokens": 567534.0,
"step": 225
},
{
"entropy": 8.750615215301513,
"epoch": 0.026543566070398154,
"grad_norm": 1.1015625,
"learning_rate": 0.0001145,
"loss": 8.2643,
"mean_token_accuracy": 0.09682516902685165,
"num_tokens": 579297.0,
"step": 230
},
{
"entropy": 8.695602512359619,
"epoch": 0.027120600115406807,
"grad_norm": 0.875,
"learning_rate": 0.00011700000000000001,
"loss": 8.2384,
"mean_token_accuracy": 0.09817592948675155,
"num_tokens": 592554.0,
"step": 235
},
{
"entropy": 8.732758808135987,
"epoch": 0.027697634160415464,
"grad_norm": 0.921875,
"learning_rate": 0.00011949999999999999,
"loss": 8.2291,
"mean_token_accuracy": 0.09817634895443916,
"num_tokens": 605038.0,
"step": 240
},
{
"entropy": 8.617175579071045,
"epoch": 0.02827466820542412,
"grad_norm": 1.1640625,
"learning_rate": 0.000122,
"loss": 8.2451,
"mean_token_accuracy": 0.09733725637197495,
"num_tokens": 617172.0,
"step": 245
},
{
"entropy": 8.57886381149292,
"epoch": 0.028851702250432775,
"grad_norm": 1.0703125,
"learning_rate": 0.0001245,
"loss": 8.1868,
"mean_token_accuracy": 0.09926503598690033,
"num_tokens": 629243.0,
"step": 250
},
{
"entropy": 8.597729873657226,
"epoch": 0.029428736295441432,
"grad_norm": 0.921875,
"learning_rate": 0.000127,
"loss": 8.2453,
"mean_token_accuracy": 0.09898478463292122,
"num_tokens": 641487.0,
"step": 255
},
{
"entropy": 8.657223415374755,
"epoch": 0.030005770340450086,
"grad_norm": 1.0078125,
"learning_rate": 0.0001295,
"loss": 8.2382,
"mean_token_accuracy": 0.08884634375572205,
"num_tokens": 653214.0,
"step": 260
},
{
"entropy": 8.59217348098755,
"epoch": 0.030582804385458743,
"grad_norm": 1.1796875,
"learning_rate": 0.000132,
"loss": 8.1923,
"mean_token_accuracy": 0.11060753613710403,
"num_tokens": 666549.0,
"step": 265
},
{
"entropy": 8.513946914672852,
"epoch": 0.031159838430467397,
"grad_norm": 0.828125,
"learning_rate": 0.00013450000000000002,
"loss": 8.2137,
"mean_token_accuracy": 0.10257397592067719,
"num_tokens": 679443.0,
"step": 270
},
{
"entropy": 8.499358367919921,
"epoch": 0.031736872475476054,
"grad_norm": 1.015625,
"learning_rate": 0.00013700000000000002,
"loss": 8.2136,
"mean_token_accuracy": 0.09787259846925736,
"num_tokens": 692929.0,
"step": 275
},
{
"entropy": 8.5993070602417,
"epoch": 0.03231390652048471,
"grad_norm": 1.078125,
"learning_rate": 0.0001395,
"loss": 8.133,
"mean_token_accuracy": 0.1011851780116558,
"num_tokens": 705605.0,
"step": 280
},
{
"entropy": 8.563784217834472,
"epoch": 0.03289094056549336,
"grad_norm": 0.7109375,
"learning_rate": 0.00014199999999999998,
"loss": 8.2181,
"mean_token_accuracy": 0.09577601477503776,
"num_tokens": 719064.0,
"step": 285
},
{
"entropy": 8.528156089782716,
"epoch": 0.03346797461050202,
"grad_norm": 0.86328125,
"learning_rate": 0.0001445,
"loss": 8.148,
"mean_token_accuracy": 0.10154027119278908,
"num_tokens": 732572.0,
"step": 290
},
{
"entropy": 8.501526927947998,
"epoch": 0.034045008655510675,
"grad_norm": 0.94921875,
"learning_rate": 0.000147,
"loss": 8.1452,
"mean_token_accuracy": 0.09947796389460564,
"num_tokens": 745161.0,
"step": 295
},
{
"entropy": 8.492893695831299,
"epoch": 0.03462204270051933,
"grad_norm": 0.9296875,
"learning_rate": 0.0001495,
"loss": 8.1182,
"mean_token_accuracy": 0.1088537909090519,
"num_tokens": 756180.0,
"step": 300
},
{
"entropy": 8.508162879943848,
"epoch": 0.03519907674552799,
"grad_norm": 1.2109375,
"learning_rate": 0.000152,
"loss": 8.142,
"mean_token_accuracy": 0.10471795424818993,
"num_tokens": 768568.0,
"step": 305
},
{
"entropy": 8.514046001434327,
"epoch": 0.03577611079053664,
"grad_norm": 1.078125,
"learning_rate": 0.00015450000000000001,
"loss": 8.1756,
"mean_token_accuracy": 0.10089959502220154,
"num_tokens": 781580.0,
"step": 310
},
{
"entropy": 8.390854835510254,
"epoch": 0.0363531448355453,
"grad_norm": 0.9375,
"learning_rate": 0.000157,
"loss": 8.0541,
"mean_token_accuracy": 0.11660909354686737,
"num_tokens": 793486.0,
"step": 315
},
{
"entropy": 8.494402980804443,
"epoch": 0.03693017888055395,
"grad_norm": 1.0078125,
"learning_rate": 0.0001595,
"loss": 8.1249,
"mean_token_accuracy": 0.10236181169748307,
"num_tokens": 805997.0,
"step": 320
},
{
"entropy": 8.422712516784667,
"epoch": 0.03750721292556261,
"grad_norm": 1.125,
"learning_rate": 0.000162,
"loss": 8.0829,
"mean_token_accuracy": 0.10648757740855216,
"num_tokens": 817681.0,
"step": 325
},
{
"entropy": 8.429091262817384,
"epoch": 0.038084246970571264,
"grad_norm": 0.87890625,
"learning_rate": 0.00016450000000000001,
"loss": 8.0916,
"mean_token_accuracy": 0.11145856082439423,
"num_tokens": 830027.0,
"step": 330
},
{
"entropy": 8.464876174926758,
"epoch": 0.03866128101557992,
"grad_norm": 0.78125,
"learning_rate": 0.00016700000000000002,
"loss": 8.1081,
"mean_token_accuracy": 0.10330765470862388,
"num_tokens": 842862.0,
"step": 335
},
{
"entropy": 8.401940250396729,
"epoch": 0.03923831506058857,
"grad_norm": 1.015625,
"learning_rate": 0.00016950000000000003,
"loss": 8.0324,
"mean_token_accuracy": 0.10951320230960845,
"num_tokens": 856548.0,
"step": 340
},
{
"entropy": 8.391893863677979,
"epoch": 0.03981534910559723,
"grad_norm": 1.0078125,
"learning_rate": 0.00017199999999999998,
"loss": 8.0417,
"mean_token_accuracy": 0.10960106775164605,
"num_tokens": 869651.0,
"step": 345
},
{
"entropy": 8.451942253112794,
"epoch": 0.040392383150605886,
"grad_norm": 1.1015625,
"learning_rate": 0.00017449999999999999,
"loss": 8.0689,
"mean_token_accuracy": 0.1082707829773426,
"num_tokens": 880717.0,
"step": 350
},
{
"entropy": 8.369545364379883,
"epoch": 0.04096941719561454,
"grad_norm": 0.921875,
"learning_rate": 0.000177,
"loss": 8.0521,
"mean_token_accuracy": 0.11259332075715064,
"num_tokens": 894066.0,
"step": 355
},
{
"entropy": 8.282257270812988,
"epoch": 0.0415464512406232,
"grad_norm": 1.046875,
"learning_rate": 0.0001795,
"loss": 7.9567,
"mean_token_accuracy": 0.11248702630400657,
"num_tokens": 905287.0,
"step": 360
},
{
"entropy": 8.459172248840332,
"epoch": 0.042123485285631854,
"grad_norm": 1.0078125,
"learning_rate": 0.000182,
"loss": 8.0392,
"mean_token_accuracy": 0.11402102708816528,
"num_tokens": 917512.0,
"step": 365
},
{
"entropy": 8.318458461761475,
"epoch": 0.04270051933064051,
"grad_norm": 1.265625,
"learning_rate": 0.0001845,
"loss": 7.9899,
"mean_token_accuracy": 0.10831499472260475,
"num_tokens": 930263.0,
"step": 370
},
{
"entropy": 8.339896297454834,
"epoch": 0.04327755337564916,
"grad_norm": 0.90625,
"learning_rate": 0.000187,
"loss": 8.0425,
"mean_token_accuracy": 0.11203662231564522,
"num_tokens": 942330.0,
"step": 375
},
{
"entropy": 8.366154861450195,
"epoch": 0.04385458742065782,
"grad_norm": 1.03125,
"learning_rate": 0.0001895,
"loss": 8.0316,
"mean_token_accuracy": 0.10789357051253319,
"num_tokens": 954005.0,
"step": 380
},
{
"entropy": 8.355004215240479,
"epoch": 0.044431621465666475,
"grad_norm": 0.91796875,
"learning_rate": 0.000192,
"loss": 8.026,
"mean_token_accuracy": 0.1113132230937481,
"num_tokens": 966725.0,
"step": 385
},
{
"entropy": 8.353032684326172,
"epoch": 0.04500865551067513,
"grad_norm": 1.0390625,
"learning_rate": 0.0001945,
"loss": 8.0044,
"mean_token_accuracy": 0.1139283761382103,
"num_tokens": 978021.0,
"step": 390
},
{
"entropy": 8.344230270385742,
"epoch": 0.04558568955568378,
"grad_norm": 0.98828125,
"learning_rate": 0.00019700000000000002,
"loss": 7.9905,
"mean_token_accuracy": 0.11658623144030571,
"num_tokens": 991321.0,
"step": 395
},
{
"entropy": 8.339315605163574,
"epoch": 0.04616272360069244,
"grad_norm": 0.92578125,
"learning_rate": 0.00019950000000000002,
"loss": 7.9818,
"mean_token_accuracy": 0.11692097708582878,
"num_tokens": 1004172.0,
"step": 400
},
{
"entropy": 8.253484535217286,
"epoch": 0.0467397576457011,
"grad_norm": 1.015625,
"learning_rate": 0.000202,
"loss": 7.936,
"mean_token_accuracy": 0.11915154382586479,
"num_tokens": 1016513.0,
"step": 405
},
{
"entropy": 8.23137845993042,
"epoch": 0.04731679169070975,
"grad_norm": 1.265625,
"learning_rate": 0.00020449999999999998,
"loss": 7.9323,
"mean_token_accuracy": 0.11277970448136329,
"num_tokens": 1028536.0,
"step": 410
},
{
"entropy": 8.298001146316528,
"epoch": 0.04789382573571841,
"grad_norm": 1.421875,
"learning_rate": 0.000207,
"loss": 7.8757,
"mean_token_accuracy": 0.12688373401761055,
"num_tokens": 1040534.0,
"step": 415
},
{
"entropy": 8.242660903930664,
"epoch": 0.048470859780727064,
"grad_norm": 1.03125,
"learning_rate": 0.0002095,
"loss": 7.9649,
"mean_token_accuracy": 0.11573690325021743,
"num_tokens": 1054995.0,
"step": 420
},
{
"entropy": 8.27421875,
"epoch": 0.04904789382573572,
"grad_norm": 0.98046875,
"learning_rate": 0.000212,
"loss": 7.9873,
"mean_token_accuracy": 0.11399574726819992,
"num_tokens": 1067722.0,
"step": 425
},
{
"entropy": 8.373875331878661,
"epoch": 0.04962492787074437,
"grad_norm": 1.0234375,
"learning_rate": 0.0002145,
"loss": 7.9637,
"mean_token_accuracy": 0.10732617229223251,
"num_tokens": 1079788.0,
"step": 430
},
{
"entropy": 8.326363945007325,
"epoch": 0.05020196191575303,
"grad_norm": 1.0546875,
"learning_rate": 0.00021700000000000002,
"loss": 7.9539,
"mean_token_accuracy": 0.1163248635828495,
"num_tokens": 1092016.0,
"step": 435
},
{
"entropy": 8.17879877090454,
"epoch": 0.050778995960761686,
"grad_norm": 1.1328125,
"learning_rate": 0.0002195,
"loss": 7.9654,
"mean_token_accuracy": 0.11178191602230073,
"num_tokens": 1104522.0,
"step": 440
},
{
"entropy": 8.363206481933593,
"epoch": 0.05135603000577034,
"grad_norm": 0.91015625,
"learning_rate": 0.000222,
"loss": 7.979,
"mean_token_accuracy": 0.12337485775351524,
"num_tokens": 1117239.0,
"step": 445
},
{
"entropy": 8.268813133239746,
"epoch": 0.05193306405077899,
"grad_norm": 0.96875,
"learning_rate": 0.0002245,
"loss": 7.9035,
"mean_token_accuracy": 0.11866443902254105,
"num_tokens": 1129061.0,
"step": 450
},
{
"entropy": 8.266489696502685,
"epoch": 0.052510098095787654,
"grad_norm": 0.95703125,
"learning_rate": 0.00022700000000000002,
"loss": 7.8976,
"mean_token_accuracy": 0.11053306236863136,
"num_tokens": 1141409.0,
"step": 455
},
{
"entropy": 8.20300588607788,
"epoch": 0.05308713214079631,
"grad_norm": 0.9609375,
"learning_rate": 0.00022950000000000002,
"loss": 7.9036,
"mean_token_accuracy": 0.11711592748761177,
"num_tokens": 1153623.0,
"step": 460
},
{
"entropy": 8.255742359161378,
"epoch": 0.05366416618580496,
"grad_norm": 1.3359375,
"learning_rate": 0.00023200000000000003,
"loss": 7.9518,
"mean_token_accuracy": 0.11555556952953339,
"num_tokens": 1166877.0,
"step": 465
},
{
"entropy": 8.26829776763916,
"epoch": 0.054241200230813615,
"grad_norm": 0.7734375,
"learning_rate": 0.00023449999999999998,
"loss": 7.9228,
"mean_token_accuracy": 0.11642130017280579,
"num_tokens": 1179625.0,
"step": 470
},
{
"entropy": 8.19015064239502,
"epoch": 0.054818234275822275,
"grad_norm": 0.91796875,
"learning_rate": 0.000237,
"loss": 7.9401,
"mean_token_accuracy": 0.11644659489393235,
"num_tokens": 1193586.0,
"step": 475
},
{
"entropy": 8.207570362091065,
"epoch": 0.05539526832083093,
"grad_norm": 0.90625,
"learning_rate": 0.0002395,
"loss": 7.8841,
"mean_token_accuracy": 0.12152690812945366,
"num_tokens": 1207086.0,
"step": 480
},
{
"entropy": 8.263338947296143,
"epoch": 0.05597230236583958,
"grad_norm": 0.92578125,
"learning_rate": 0.000242,
"loss": 7.8895,
"mean_token_accuracy": 0.11943561136722565,
"num_tokens": 1219389.0,
"step": 485
},
{
"entropy": 8.21270751953125,
"epoch": 0.05654933641084824,
"grad_norm": 1.1171875,
"learning_rate": 0.0002445,
"loss": 7.825,
"mean_token_accuracy": 0.12387257441878319,
"num_tokens": 1232080.0,
"step": 490
},
{
"entropy": 8.094493389129639,
"epoch": 0.0571263704558569,
"grad_norm": 1.0390625,
"learning_rate": 0.000247,
"loss": 7.874,
"mean_token_accuracy": 0.12492263168096543,
"num_tokens": 1246775.0,
"step": 495
},
{
"entropy": 8.204738426208497,
"epoch": 0.05770340450086555,
"grad_norm": 0.87890625,
"learning_rate": 0.0002495,
"loss": 7.9068,
"mean_token_accuracy": 0.114556435495615,
"num_tokens": 1258839.0,
"step": 500
},
{
"entropy": 8.169677925109863,
"epoch": 0.058280438545874204,
"grad_norm": 0.9296875,
"learning_rate": 0.000252,
"loss": 7.8279,
"mean_token_accuracy": 0.12553127557039262,
"num_tokens": 1270710.0,
"step": 505
},
{
"entropy": 8.218509674072266,
"epoch": 0.058857472590882864,
"grad_norm": 1.0625,
"learning_rate": 0.0002545,
"loss": 7.786,
"mean_token_accuracy": 0.12104848474264145,
"num_tokens": 1281930.0,
"step": 510
},
{
"entropy": 8.104724454879761,
"epoch": 0.05943450663589152,
"grad_norm": 0.9765625,
"learning_rate": 0.000257,
"loss": 7.8534,
"mean_token_accuracy": 0.11715041622519493,
"num_tokens": 1294548.0,
"step": 515
},
{
"entropy": 8.166370916366578,
"epoch": 0.06001154068090017,
"grad_norm": 1.0546875,
"learning_rate": 0.0002595,
"loss": 7.8064,
"mean_token_accuracy": 0.11803867742419243,
"num_tokens": 1307588.0,
"step": 520
},
{
"entropy": 8.132765865325927,
"epoch": 0.060588574725908825,
"grad_norm": 1.0078125,
"learning_rate": 0.000262,
"loss": 7.7863,
"mean_token_accuracy": 0.12690332680940627,
"num_tokens": 1319478.0,
"step": 525
},
{
"entropy": 8.216028451919556,
"epoch": 0.061165608770917486,
"grad_norm": 0.98828125,
"learning_rate": 0.00026450000000000003,
"loss": 7.8708,
"mean_token_accuracy": 0.11611150279641151,
"num_tokens": 1332950.0,
"step": 530
},
{
"entropy": 8.090120029449462,
"epoch": 0.06174264281592614,
"grad_norm": 1.1953125,
"learning_rate": 0.00026700000000000004,
"loss": 7.7615,
"mean_token_accuracy": 0.12739897295832633,
"num_tokens": 1345510.0,
"step": 535
},
{
"entropy": 8.01339612007141,
"epoch": 0.06231967686093479,
"grad_norm": 1.1875,
"learning_rate": 0.00026950000000000005,
"loss": 7.8042,
"mean_token_accuracy": 0.12997828647494317,
"num_tokens": 1359388.0,
"step": 540
},
{
"entropy": 8.145500087738037,
"epoch": 0.06289671090594345,
"grad_norm": 1.0546875,
"learning_rate": 0.00027200000000000005,
"loss": 7.8149,
"mean_token_accuracy": 0.12175369709730148,
"num_tokens": 1373067.0,
"step": 545
},
{
"entropy": 8.04567756652832,
"epoch": 0.06347374495095211,
"grad_norm": 0.90234375,
"learning_rate": 0.0002745,
"loss": 7.7835,
"mean_token_accuracy": 0.12414814531803131,
"num_tokens": 1385718.0,
"step": 550
},
{
"entropy": 8.235342979431152,
"epoch": 0.06405077899596076,
"grad_norm": 1.171875,
"learning_rate": 0.000277,
"loss": 7.7898,
"mean_token_accuracy": 0.12540573999285698,
"num_tokens": 1398788.0,
"step": 555
},
{
"entropy": 8.071519756317139,
"epoch": 0.06462781304096941,
"grad_norm": 1.0703125,
"learning_rate": 0.0002795,
"loss": 7.7882,
"mean_token_accuracy": 0.12845603749155998,
"num_tokens": 1411752.0,
"step": 560
},
{
"entropy": 8.075310802459716,
"epoch": 0.06520484708597807,
"grad_norm": 1.015625,
"learning_rate": 0.00028199999999999997,
"loss": 7.8443,
"mean_token_accuracy": 0.12352012917399406,
"num_tokens": 1423979.0,
"step": 565
},
{
"entropy": 8.11637830734253,
"epoch": 0.06578188113098672,
"grad_norm": 0.88671875,
"learning_rate": 0.0002845,
"loss": 7.8721,
"mean_token_accuracy": 0.11852672100067138,
"num_tokens": 1437009.0,
"step": 570
},
{
"entropy": 8.124029541015625,
"epoch": 0.06635891517599539,
"grad_norm": 1.2109375,
"learning_rate": 0.000287,
"loss": 7.7174,
"mean_token_accuracy": 0.12357776835560799,
"num_tokens": 1448220.0,
"step": 575
},
{
"entropy": 8.00622100830078,
"epoch": 0.06693594922100404,
"grad_norm": 1.234375,
"learning_rate": 0.0002895,
"loss": 7.7566,
"mean_token_accuracy": 0.12290028035640717,
"num_tokens": 1460024.0,
"step": 580
},
{
"entropy": 8.137111854553222,
"epoch": 0.0675129832660127,
"grad_norm": 0.984375,
"learning_rate": 0.000292,
"loss": 7.7451,
"mean_token_accuracy": 0.1224621057510376,
"num_tokens": 1472283.0,
"step": 585
},
{
"entropy": 8.003714799880981,
"epoch": 0.06809001731102135,
"grad_norm": 0.84765625,
"learning_rate": 0.0002945,
"loss": 7.7908,
"mean_token_accuracy": 0.12675876170396805,
"num_tokens": 1486565.0,
"step": 590
},
{
"entropy": 7.9941528797149655,
"epoch": 0.06866705135603,
"grad_norm": 1.03125,
"learning_rate": 0.000297,
"loss": 7.7704,
"mean_token_accuracy": 0.1255685530602932,
"num_tokens": 1498708.0,
"step": 595
},
{
"entropy": 8.076406526565552,
"epoch": 0.06924408540103866,
"grad_norm": 1.0546875,
"learning_rate": 0.0002995,
"loss": 7.7121,
"mean_token_accuracy": 0.1268436163663864,
"num_tokens": 1511318.0,
"step": 600
},
{
"entropy": 7.977707815170288,
"epoch": 0.06982111944604731,
"grad_norm": 0.8828125,
"learning_rate": 0.000302,
"loss": 7.695,
"mean_token_accuracy": 0.13601174131035804,
"num_tokens": 1525415.0,
"step": 605
},
{
"entropy": 8.098061847686768,
"epoch": 0.07039815349105598,
"grad_norm": 0.96875,
"learning_rate": 0.0003045,
"loss": 7.7433,
"mean_token_accuracy": 0.12430487498641014,
"num_tokens": 1538757.0,
"step": 610
},
{
"entropy": 7.966400146484375,
"epoch": 0.07097518753606463,
"grad_norm": 1.140625,
"learning_rate": 0.000307,
"loss": 7.6977,
"mean_token_accuracy": 0.12538538724184037,
"num_tokens": 1550619.0,
"step": 615
},
{
"entropy": 8.005772590637207,
"epoch": 0.07155222158107329,
"grad_norm": 1.0625,
"learning_rate": 0.0003095,
"loss": 7.7496,
"mean_token_accuracy": 0.12077321037650109,
"num_tokens": 1562588.0,
"step": 620
},
{
"entropy": 7.989212322235107,
"epoch": 0.07212925562608194,
"grad_norm": 1.0390625,
"learning_rate": 0.000312,
"loss": 7.6787,
"mean_token_accuracy": 0.12967532649636268,
"num_tokens": 1574892.0,
"step": 625
},
{
"entropy": 7.974766254425049,
"epoch": 0.0727062896710906,
"grad_norm": 1.1171875,
"learning_rate": 0.0003145,
"loss": 7.7117,
"mean_token_accuracy": 0.12752822190523147,
"num_tokens": 1587525.0,
"step": 630
},
{
"entropy": 8.038787698745727,
"epoch": 0.07328332371609925,
"grad_norm": 1.265625,
"learning_rate": 0.000317,
"loss": 7.6921,
"mean_token_accuracy": 0.12393600195646286,
"num_tokens": 1600350.0,
"step": 635
},
{
"entropy": 7.9325361251831055,
"epoch": 0.0738603577611079,
"grad_norm": 0.8984375,
"learning_rate": 0.0003195,
"loss": 7.6823,
"mean_token_accuracy": 0.12833620086312295,
"num_tokens": 1613318.0,
"step": 640
},
{
"entropy": 8.086976289749146,
"epoch": 0.07443739180611655,
"grad_norm": 1.0546875,
"learning_rate": 0.000322,
"loss": 7.8105,
"mean_token_accuracy": 0.12034560292959214,
"num_tokens": 1626757.0,
"step": 645
},
{
"entropy": 7.982501125335693,
"epoch": 0.07501442585112522,
"grad_norm": 0.984375,
"learning_rate": 0.00032450000000000003,
"loss": 7.6224,
"mean_token_accuracy": 0.13317669555544853,
"num_tokens": 1639185.0,
"step": 650
},
{
"entropy": 7.921093416213989,
"epoch": 0.07559145989613388,
"grad_norm": 0.97265625,
"learning_rate": 0.00032700000000000003,
"loss": 7.727,
"mean_token_accuracy": 0.12360085397958756,
"num_tokens": 1652009.0,
"step": 655
},
{
"entropy": 7.985281324386596,
"epoch": 0.07616849394114253,
"grad_norm": 1.0859375,
"learning_rate": 0.00032950000000000004,
"loss": 7.7099,
"mean_token_accuracy": 0.12278636991977691,
"num_tokens": 1665604.0,
"step": 660
},
{
"entropy": 7.997657918930054,
"epoch": 0.07674552798615118,
"grad_norm": 0.953125,
"learning_rate": 0.00033200000000000005,
"loss": 7.6893,
"mean_token_accuracy": 0.12272793427109718,
"num_tokens": 1679036.0,
"step": 665
},
{
"entropy": 7.8749652862548825,
"epoch": 0.07732256203115984,
"grad_norm": 1.0625,
"learning_rate": 0.00033450000000000005,
"loss": 7.664,
"mean_token_accuracy": 0.13161094784736632,
"num_tokens": 1691512.0,
"step": 670
},
{
"entropy": 8.132879447937011,
"epoch": 0.07789959607616849,
"grad_norm": 1.0078125,
"learning_rate": 0.000337,
"loss": 7.7292,
"mean_token_accuracy": 0.12437204793095588,
"num_tokens": 1704589.0,
"step": 675
},
{
"entropy": 7.819326114654541,
"epoch": 0.07847663012117714,
"grad_norm": 0.94140625,
"learning_rate": 0.0003395,
"loss": 7.7067,
"mean_token_accuracy": 0.12104201540350915,
"num_tokens": 1718612.0,
"step": 680
},
{
"entropy": 7.861719846725464,
"epoch": 0.07905366416618581,
"grad_norm": 0.90234375,
"learning_rate": 0.000342,
"loss": 7.6127,
"mean_token_accuracy": 0.12460441142320633,
"num_tokens": 1730472.0,
"step": 685
},
{
"entropy": 8.00782494544983,
"epoch": 0.07963069821119446,
"grad_norm": 0.91796875,
"learning_rate": 0.00034449999999999997,
"loss": 7.7467,
"mean_token_accuracy": 0.12077393978834153,
"num_tokens": 1742535.0,
"step": 690
},
{
"entropy": 8.006076240539551,
"epoch": 0.08020773225620312,
"grad_norm": 1.03125,
"learning_rate": 0.000347,
"loss": 7.7141,
"mean_token_accuracy": 0.12221984565258026,
"num_tokens": 1753877.0,
"step": 695
},
{
"entropy": 7.926956653594971,
"epoch": 0.08078476630121177,
"grad_norm": 0.8671875,
"learning_rate": 0.0003495,
"loss": 7.6293,
"mean_token_accuracy": 0.12848915457725524,
"num_tokens": 1766413.0,
"step": 700
},
{
"entropy": 7.813401508331299,
"epoch": 0.08136180034622043,
"grad_norm": 1.0234375,
"learning_rate": 0.000352,
"loss": 7.4754,
"mean_token_accuracy": 0.13737191781401634,
"num_tokens": 1779719.0,
"step": 705
},
{
"entropy": 7.915401935577393,
"epoch": 0.08193883439122908,
"grad_norm": 1.1328125,
"learning_rate": 0.0003545,
"loss": 7.6389,
"mean_token_accuracy": 0.13027774319052696,
"num_tokens": 1791448.0,
"step": 710
},
{
"entropy": 7.866633939743042,
"epoch": 0.08251586843623773,
"grad_norm": 0.90234375,
"learning_rate": 0.000357,
"loss": 7.6429,
"mean_token_accuracy": 0.12651973143219947,
"num_tokens": 1803802.0,
"step": 715
},
{
"entropy": 7.941481637954712,
"epoch": 0.0830929024812464,
"grad_norm": 0.90234375,
"learning_rate": 0.0003595,
"loss": 7.6641,
"mean_token_accuracy": 0.12703499272465707,
"num_tokens": 1817545.0,
"step": 720
},
{
"entropy": 7.91947979927063,
"epoch": 0.08366993652625505,
"grad_norm": 0.8671875,
"learning_rate": 0.000362,
"loss": 7.6464,
"mean_token_accuracy": 0.12531809359788895,
"num_tokens": 1830115.0,
"step": 725
},
{
"entropy": 7.799381828308105,
"epoch": 0.08424697057126371,
"grad_norm": 1.0234375,
"learning_rate": 0.0003645,
"loss": 7.5711,
"mean_token_accuracy": 0.12959943637251853,
"num_tokens": 1844707.0,
"step": 730
},
{
"entropy": 7.925750207901001,
"epoch": 0.08482400461627236,
"grad_norm": 0.9453125,
"learning_rate": 0.000367,
"loss": 7.6557,
"mean_token_accuracy": 0.12370599582791328,
"num_tokens": 1858719.0,
"step": 735
},
{
"entropy": 7.862723875045776,
"epoch": 0.08540103866128101,
"grad_norm": 1.0390625,
"learning_rate": 0.0003695,
"loss": 7.5893,
"mean_token_accuracy": 0.12989407479763032,
"num_tokens": 1870421.0,
"step": 740
},
{
"entropy": 7.863739395141602,
"epoch": 0.08597807270628967,
"grad_norm": 0.890625,
"learning_rate": 0.000372,
"loss": 7.6572,
"mean_token_accuracy": 0.1180440478026867,
"num_tokens": 1883792.0,
"step": 745
},
{
"entropy": 7.822341585159302,
"epoch": 0.08655510675129832,
"grad_norm": 1.046875,
"learning_rate": 0.0003745,
"loss": 7.6349,
"mean_token_accuracy": 0.12786092832684517,
"num_tokens": 1896209.0,
"step": 750
},
{
"entropy": 7.870438051223755,
"epoch": 0.08713214079630698,
"grad_norm": 1.046875,
"learning_rate": 0.000377,
"loss": 7.5787,
"mean_token_accuracy": 0.1304923750460148,
"num_tokens": 1908520.0,
"step": 755
},
{
"entropy": 7.85968918800354,
"epoch": 0.08770917484131564,
"grad_norm": 1.0234375,
"learning_rate": 0.0003795,
"loss": 7.5898,
"mean_token_accuracy": 0.12856273204088212,
"num_tokens": 1922489.0,
"step": 760
},
{
"entropy": 7.790146064758301,
"epoch": 0.0882862088863243,
"grad_norm": 1.015625,
"learning_rate": 0.000382,
"loss": 7.5841,
"mean_token_accuracy": 0.1291157804429531,
"num_tokens": 1935206.0,
"step": 765
},
{
"entropy": 7.834140348434448,
"epoch": 0.08886324293133295,
"grad_norm": 1.0625,
"learning_rate": 0.0003845,
"loss": 7.5396,
"mean_token_accuracy": 0.13339237496256828,
"num_tokens": 1948798.0,
"step": 770
},
{
"entropy": 7.79986310005188,
"epoch": 0.0894402769763416,
"grad_norm": 1.0390625,
"learning_rate": 0.00038700000000000003,
"loss": 7.537,
"mean_token_accuracy": 0.13368105366826058,
"num_tokens": 1961095.0,
"step": 775
},
{
"entropy": 7.886825084686279,
"epoch": 0.09001731102135026,
"grad_norm": 1.0703125,
"learning_rate": 0.00038950000000000003,
"loss": 7.5479,
"mean_token_accuracy": 0.12701786160469056,
"num_tokens": 1974229.0,
"step": 780
},
{
"entropy": 7.805331611633301,
"epoch": 0.09059434506635891,
"grad_norm": 0.92578125,
"learning_rate": 0.00039200000000000004,
"loss": 7.6401,
"mean_token_accuracy": 0.12609824389219285,
"num_tokens": 1986334.0,
"step": 785
},
{
"entropy": 7.807611322402954,
"epoch": 0.09117137911136756,
"grad_norm": 0.828125,
"learning_rate": 0.00039450000000000005,
"loss": 7.5884,
"mean_token_accuracy": 0.12830588668584825,
"num_tokens": 2000337.0,
"step": 790
},
{
"entropy": 7.795743417739868,
"epoch": 0.09174841315637623,
"grad_norm": 0.93359375,
"learning_rate": 0.00039700000000000005,
"loss": 7.51,
"mean_token_accuracy": 0.13368136137723924,
"num_tokens": 2012048.0,
"step": 795
},
{
"entropy": 7.784616470336914,
"epoch": 0.09232544720138489,
"grad_norm": 1.046875,
"learning_rate": 0.0003995,
"loss": 7.529,
"mean_token_accuracy": 0.13036411479115487,
"num_tokens": 2025244.0,
"step": 800
},
{
"entropy": 7.810312700271607,
"epoch": 0.09290248124639354,
"grad_norm": 0.9765625,
"learning_rate": 0.000402,
"loss": 7.5725,
"mean_token_accuracy": 0.13225525096058846,
"num_tokens": 2036542.0,
"step": 805
},
{
"entropy": 7.813270044326782,
"epoch": 0.0934795152914022,
"grad_norm": 0.859375,
"learning_rate": 0.0004045,
"loss": 7.5115,
"mean_token_accuracy": 0.13005766347050668,
"num_tokens": 2049358.0,
"step": 810
},
{
"entropy": 7.700181245803833,
"epoch": 0.09405654933641085,
"grad_norm": 0.953125,
"learning_rate": 0.00040699999999999997,
"loss": 7.5409,
"mean_token_accuracy": 0.12904200404882432,
"num_tokens": 2062114.0,
"step": 815
},
{
"entropy": 7.861858272552491,
"epoch": 0.0946335833814195,
"grad_norm": 1.0625,
"learning_rate": 0.0004095,
"loss": 7.5139,
"mean_token_accuracy": 0.13364754766225814,
"num_tokens": 2074236.0,
"step": 820
},
{
"entropy": 7.705525732040405,
"epoch": 0.09521061742642815,
"grad_norm": 0.89453125,
"learning_rate": 0.000412,
"loss": 7.4266,
"mean_token_accuracy": 0.13591486886143683,
"num_tokens": 2086320.0,
"step": 825
},
{
"entropy": 7.766620826721192,
"epoch": 0.09578765147143682,
"grad_norm": 0.8984375,
"learning_rate": 0.0004145,
"loss": 7.4885,
"mean_token_accuracy": 0.13510627821087837,
"num_tokens": 2099119.0,
"step": 830
},
{
"entropy": 7.711123943328857,
"epoch": 0.09636468551644548,
"grad_norm": 0.9921875,
"learning_rate": 0.000417,
"loss": 7.5148,
"mean_token_accuracy": 0.1323984071612358,
"num_tokens": 2111795.0,
"step": 835
},
{
"entropy": 7.660836982727051,
"epoch": 0.09694171956145413,
"grad_norm": 1.2734375,
"learning_rate": 0.0004195,
"loss": 7.4957,
"mean_token_accuracy": 0.11923368275165558,
"num_tokens": 2126621.0,
"step": 840
},
{
"entropy": 7.764021539688111,
"epoch": 0.09751875360646278,
"grad_norm": 0.9375,
"learning_rate": 0.000422,
"loss": 7.3964,
"mean_token_accuracy": 0.1446671850979328,
"num_tokens": 2139770.0,
"step": 845
},
{
"entropy": 7.73313717842102,
"epoch": 0.09809578765147144,
"grad_norm": 0.99609375,
"learning_rate": 0.0004245,
"loss": 7.4817,
"mean_token_accuracy": 0.13560690730810165,
"num_tokens": 2151683.0,
"step": 850
},
{
"entropy": 7.648449945449829,
"epoch": 0.09867282169648009,
"grad_norm": 0.9453125,
"learning_rate": 0.000427,
"loss": 7.4604,
"mean_token_accuracy": 0.1294977843761444,
"num_tokens": 2163178.0,
"step": 855
},
{
"entropy": 7.755085849761963,
"epoch": 0.09924985574148874,
"grad_norm": 0.91796875,
"learning_rate": 0.0004295,
"loss": 7.5485,
"mean_token_accuracy": 0.12189057022333145,
"num_tokens": 2176312.0,
"step": 860
},
{
"entropy": 7.789842224121093,
"epoch": 0.0998268897864974,
"grad_norm": 1.390625,
"learning_rate": 0.000432,
"loss": 7.48,
"mean_token_accuracy": 0.12981949299573897,
"num_tokens": 2189045.0,
"step": 865
},
{
"entropy": 7.571734762191772,
"epoch": 0.10040392383150606,
"grad_norm": 1.03125,
"learning_rate": 0.0004345,
"loss": 7.4754,
"mean_token_accuracy": 0.12462260350584983,
"num_tokens": 2202083.0,
"step": 870
},
{
"entropy": 7.715145778656006,
"epoch": 0.10098095787651472,
"grad_norm": 1.1015625,
"learning_rate": 0.000437,
"loss": 7.5248,
"mean_token_accuracy": 0.1264673389494419,
"num_tokens": 2215775.0,
"step": 875
},
{
"entropy": 7.77250919342041,
"epoch": 0.10155799192152337,
"grad_norm": 1.0,
"learning_rate": 0.0004395,
"loss": 7.4774,
"mean_token_accuracy": 0.12318376153707504,
"num_tokens": 2228387.0,
"step": 880
},
{
"entropy": 7.6760975360870365,
"epoch": 0.10213502596653203,
"grad_norm": 0.98046875,
"learning_rate": 0.000442,
"loss": 7.4827,
"mean_token_accuracy": 0.13410628139972686,
"num_tokens": 2241491.0,
"step": 885
},
{
"entropy": 7.5755048274993895,
"epoch": 0.10271206001154068,
"grad_norm": 0.9921875,
"learning_rate": 0.0004445,
"loss": 7.4057,
"mean_token_accuracy": 0.13969002217054366,
"num_tokens": 2255282.0,
"step": 890
},
{
"entropy": 7.67344913482666,
"epoch": 0.10328909405654933,
"grad_norm": 1.0859375,
"learning_rate": 0.000447,
"loss": 7.4078,
"mean_token_accuracy": 0.13210881277918815,
"num_tokens": 2268081.0,
"step": 895
},
{
"entropy": 7.641108894348145,
"epoch": 0.10386612810155799,
"grad_norm": 0.953125,
"learning_rate": 0.00044950000000000003,
"loss": 7.3767,
"mean_token_accuracy": 0.1385306254029274,
"num_tokens": 2281069.0,
"step": 900
},
{
"entropy": 7.662174987792969,
"epoch": 0.10444316214656665,
"grad_norm": 0.92578125,
"learning_rate": 0.00045200000000000004,
"loss": 7.4474,
"mean_token_accuracy": 0.12799655497074128,
"num_tokens": 2293606.0,
"step": 905
},
{
"entropy": 7.689595937728882,
"epoch": 0.10502019619157531,
"grad_norm": 1.078125,
"learning_rate": 0.00045450000000000004,
"loss": 7.4478,
"mean_token_accuracy": 0.12830137014389037,
"num_tokens": 2304973.0,
"step": 910
},
{
"entropy": 7.686489248275757,
"epoch": 0.10559723023658396,
"grad_norm": 0.90234375,
"learning_rate": 0.00045700000000000005,
"loss": 7.4863,
"mean_token_accuracy": 0.12266820147633553,
"num_tokens": 2319016.0,
"step": 915
},
{
"entropy": 7.6014503002166744,
"epoch": 0.10617426428159261,
"grad_norm": 1.0,
"learning_rate": 0.00045950000000000006,
"loss": 7.4463,
"mean_token_accuracy": 0.1375369183719158,
"num_tokens": 2330599.0,
"step": 920
},
{
"entropy": 7.717125272750854,
"epoch": 0.10675129832660127,
"grad_norm": 1.09375,
"learning_rate": 0.000462,
"loss": 7.4444,
"mean_token_accuracy": 0.12856984063982962,
"num_tokens": 2343153.0,
"step": 925
},
{
"entropy": 7.697896146774292,
"epoch": 0.10732833237160992,
"grad_norm": 0.87890625,
"learning_rate": 0.0004645,
"loss": 7.4695,
"mean_token_accuracy": 0.12443050369620323,
"num_tokens": 2356762.0,
"step": 930
},
{
"entropy": 7.53573899269104,
"epoch": 0.10790536641661858,
"grad_norm": 1.0546875,
"learning_rate": 0.000467,
"loss": 7.4916,
"mean_token_accuracy": 0.13005805611610413,
"num_tokens": 2369581.0,
"step": 935
},
{
"entropy": 7.7132821559906,
"epoch": 0.10848240046162723,
"grad_norm": 1.03125,
"learning_rate": 0.0004695,
"loss": 7.4025,
"mean_token_accuracy": 0.1298571690917015,
"num_tokens": 2381419.0,
"step": 940
},
{
"entropy": 7.599806404113769,
"epoch": 0.1090594345066359,
"grad_norm": 0.94921875,
"learning_rate": 0.000472,
"loss": 7.404,
"mean_token_accuracy": 0.1387295603752136,
"num_tokens": 2393586.0,
"step": 945
},
{
"entropy": 7.5154454708099365,
"epoch": 0.10963646855164455,
"grad_norm": 0.98828125,
"learning_rate": 0.0004745,
"loss": 7.3841,
"mean_token_accuracy": 0.13711024671792985,
"num_tokens": 2407720.0,
"step": 950
},
{
"entropy": 7.691217041015625,
"epoch": 0.1102135025966532,
"grad_norm": 0.88671875,
"learning_rate": 0.000477,
"loss": 7.4341,
"mean_token_accuracy": 0.12648762315511702,
"num_tokens": 2421089.0,
"step": 955
},
{
"entropy": 7.593528985977173,
"epoch": 0.11079053664166186,
"grad_norm": 0.94921875,
"learning_rate": 0.0004795,
"loss": 7.3719,
"mean_token_accuracy": 0.13786724731326103,
"num_tokens": 2432859.0,
"step": 960
},
{
"entropy": 7.563003826141357,
"epoch": 0.11136757068667051,
"grad_norm": 0.96875,
"learning_rate": 0.000482,
"loss": 7.3823,
"mean_token_accuracy": 0.1313467375934124,
"num_tokens": 2445072.0,
"step": 965
},
{
"entropy": 7.626921939849853,
"epoch": 0.11194460473167916,
"grad_norm": 0.9609375,
"learning_rate": 0.0004845,
"loss": 7.3802,
"mean_token_accuracy": 0.13656689301133157,
"num_tokens": 2458724.0,
"step": 970
},
{
"entropy": 7.551381254196167,
"epoch": 0.11252163877668782,
"grad_norm": 0.953125,
"learning_rate": 0.000487,
"loss": 7.4385,
"mean_token_accuracy": 0.13167190551757812,
"num_tokens": 2472976.0,
"step": 975
},
{
"entropy": 7.675592613220215,
"epoch": 0.11309867282169649,
"grad_norm": 0.91015625,
"learning_rate": 0.0004895,
"loss": 7.4166,
"mean_token_accuracy": 0.12921208068728446,
"num_tokens": 2486164.0,
"step": 980
},
{
"entropy": 7.476428079605102,
"epoch": 0.11367570686670514,
"grad_norm": 1.2578125,
"learning_rate": 0.000492,
"loss": 7.3646,
"mean_token_accuracy": 0.13289769142866134,
"num_tokens": 2499395.0,
"step": 985
},
{
"entropy": 7.640331506729126,
"epoch": 0.1142527409117138,
"grad_norm": 0.984375,
"learning_rate": 0.0004945,
"loss": 7.3659,
"mean_token_accuracy": 0.13304105922579765,
"num_tokens": 2512163.0,
"step": 990
},
{
"entropy": 7.560707950592041,
"epoch": 0.11482977495672245,
"grad_norm": 0.89453125,
"learning_rate": 0.000497,
"loss": 7.3714,
"mean_token_accuracy": 0.12891542091965674,
"num_tokens": 2523634.0,
"step": 995
},
{
"entropy": 7.603393411636352,
"epoch": 0.1154068090017311,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995,
"loss": 7.4008,
"mean_token_accuracy": 0.1331055149435997,
"num_tokens": 2536237.0,
"step": 1000
},
{
"entropy": 7.625240325927734,
"epoch": 0.11598384304673975,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999999975783157,
"loss": 7.3994,
"mean_token_accuracy": 0.11986244916915893,
"num_tokens": 2548035.0,
"step": 1005
},
{
"entropy": 7.594330739974976,
"epoch": 0.11656087709174841,
"grad_norm": 0.8125,
"learning_rate": 0.0004999999877402236,
"loss": 7.3344,
"mean_token_accuracy": 0.13319490626454353,
"num_tokens": 2561670.0,
"step": 1010
},
{
"entropy": 7.533303833007812,
"epoch": 0.11713791113675708,
"grad_norm": 0.9375,
"learning_rate": 0.0004999999703343686,
"loss": 7.3709,
"mean_token_accuracy": 0.1304928630590439,
"num_tokens": 2573471.0,
"step": 1015
},
{
"entropy": 7.531730508804321,
"epoch": 0.11771494518176573,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999999453607515,
"loss": 7.3175,
"mean_token_accuracy": 0.13719066530466079,
"num_tokens": 2586534.0,
"step": 1020
},
{
"entropy": 7.636745643615723,
"epoch": 0.11829197922677438,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999999128193729,
"loss": 7.3719,
"mean_token_accuracy": 0.12794509902596474,
"num_tokens": 2597970.0,
"step": 1025
},
{
"entropy": 7.373585796356201,
"epoch": 0.11886901327178304,
"grad_norm": 0.88671875,
"learning_rate": 0.000499999872710234,
"loss": 7.308,
"mean_token_accuracy": 0.13802397325634957,
"num_tokens": 2610924.0,
"step": 1030
},
{
"entropy": 7.593305397033691,
"epoch": 0.11944604731679169,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999998250333361,
"loss": 7.3737,
"mean_token_accuracy": 0.13383316844701768,
"num_tokens": 2623571.0,
"step": 1035
},
{
"entropy": 7.540350818634034,
"epoch": 0.12002308136180034,
"grad_norm": 1.0,
"learning_rate": 0.000499999769788681,
"loss": 7.3823,
"mean_token_accuracy": 0.1317036457359791,
"num_tokens": 2636063.0,
"step": 1040
},
{
"entropy": 7.56674108505249,
"epoch": 0.120600115406809,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999997069762703,
"loss": 7.3334,
"mean_token_accuracy": 0.1320666678249836,
"num_tokens": 2648621.0,
"step": 1045
},
{
"entropy": 7.478792190551758,
"epoch": 0.12117714945181765,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999996365961062,
"loss": 7.2548,
"mean_token_accuracy": 0.13802519217133521,
"num_tokens": 2661794.0,
"step": 1050
},
{
"entropy": 7.471138048171997,
"epoch": 0.12175418349682632,
"grad_norm": 1.015625,
"learning_rate": 0.0004999995586481912,
"loss": 7.3311,
"mean_token_accuracy": 0.1260454349219799,
"num_tokens": 2674048.0,
"step": 1055
},
{
"entropy": 7.464170122146607,
"epoch": 0.12233121754183497,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999994731325276,
"loss": 7.2485,
"mean_token_accuracy": 0.13324502855539322,
"num_tokens": 2685510.0,
"step": 1060
},
{
"entropy": 7.5221740245819095,
"epoch": 0.12290825158684363,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999993800491187,
"loss": 7.3328,
"mean_token_accuracy": 0.13226443603634835,
"num_tokens": 2698414.0,
"step": 1065
},
{
"entropy": 7.50278525352478,
"epoch": 0.12348528563185228,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999992793979672,
"loss": 7.3332,
"mean_token_accuracy": 0.1338995724916458,
"num_tokens": 2711343.0,
"step": 1070
},
{
"entropy": 7.443923377990723,
"epoch": 0.12406231967686093,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999991711790769,
"loss": 7.2217,
"mean_token_accuracy": 0.14058572947978973,
"num_tokens": 2723123.0,
"step": 1075
},
{
"entropy": 7.387674856185913,
"epoch": 0.12463935372186959,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999990553924511,
"loss": 7.2621,
"mean_token_accuracy": 0.13784897848963737,
"num_tokens": 2736015.0,
"step": 1080
},
{
"entropy": 7.483056497573853,
"epoch": 0.12521638776687824,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999989320380939,
"loss": 7.2327,
"mean_token_accuracy": 0.1457214944064617,
"num_tokens": 2747561.0,
"step": 1085
},
{
"entropy": 7.438428640365601,
"epoch": 0.1257934218118869,
"grad_norm": 0.90625,
"learning_rate": 0.0004999988011160094,
"loss": 7.2835,
"mean_token_accuracy": 0.13277626112103463,
"num_tokens": 2758992.0,
"step": 1090
},
{
"entropy": 7.489367580413818,
"epoch": 0.12637045585689555,
"grad_norm": 1.0234375,
"learning_rate": 0.000499998662626202,
"loss": 7.2773,
"mean_token_accuracy": 0.13354492634534837,
"num_tokens": 2771164.0,
"step": 1095
},
{
"entropy": 7.357281446456909,
"epoch": 0.12694748990190421,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999985165686764,
"loss": 7.2575,
"mean_token_accuracy": 0.1360307179391384,
"num_tokens": 2784400.0,
"step": 1100
},
{
"entropy": 7.427149534225464,
"epoch": 0.12752452394691285,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999983629434373,
"loss": 7.223,
"mean_token_accuracy": 0.13857290968298913,
"num_tokens": 2795729.0,
"step": 1105
},
{
"entropy": 7.424337244033813,
"epoch": 0.12810155799192152,
"grad_norm": 1.03125,
"learning_rate": 0.0004999982017504901,
"loss": 7.307,
"mean_token_accuracy": 0.13230945765972138,
"num_tokens": 2808118.0,
"step": 1110
},
{
"entropy": 7.445949077606201,
"epoch": 0.1286785920369302,
"grad_norm": 0.96875,
"learning_rate": 0.0004999980329898401,
"loss": 7.236,
"mean_token_accuracy": 0.1337020590901375,
"num_tokens": 2821118.0,
"step": 1115
},
{
"entropy": 7.446945095062256,
"epoch": 0.12925562608193883,
"grad_norm": 1.1953125,
"learning_rate": 0.000499997856661493,
"loss": 7.233,
"mean_token_accuracy": 0.13448369577527047,
"num_tokens": 2833636.0,
"step": 1120
},
{
"entropy": 7.35188136100769,
"epoch": 0.1298326601269475,
"grad_norm": 0.99609375,
"learning_rate": 0.000499997672765455,
"loss": 7.2509,
"mean_token_accuracy": 0.1398313045501709,
"num_tokens": 2846617.0,
"step": 1125
},
{
"entropy": 7.367458200454712,
"epoch": 0.13040969417195614,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999974813017318,
"loss": 7.2097,
"mean_token_accuracy": 0.14283193722367288,
"num_tokens": 2859191.0,
"step": 1130
},
{
"entropy": 7.45753436088562,
"epoch": 0.1309867282169648,
"grad_norm": 0.875,
"learning_rate": 0.0004999972822703301,
"loss": 7.2303,
"mean_token_accuracy": 0.13343884348869323,
"num_tokens": 2872592.0,
"step": 1135
},
{
"entropy": 7.310500621795654,
"epoch": 0.13156376226197344,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999970756712567,
"loss": 7.2068,
"mean_token_accuracy": 0.14036742150783538,
"num_tokens": 2884094.0,
"step": 1140
},
{
"entropy": 7.466288280487061,
"epoch": 0.1321407963069821,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999968615045183,
"loss": 7.265,
"mean_token_accuracy": 0.1353508099913597,
"num_tokens": 2896947.0,
"step": 1145
},
{
"entropy": 7.352451229095459,
"epoch": 0.13271783035199078,
"grad_norm": 0.921875,
"learning_rate": 0.0004999966397701222,
"loss": 7.2677,
"mean_token_accuracy": 0.13296897783875467,
"num_tokens": 2909029.0,
"step": 1150
},
{
"entropy": 7.461985635757446,
"epoch": 0.13329486439699942,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999964104680759,
"loss": 7.2432,
"mean_token_accuracy": 0.13714048266410828,
"num_tokens": 2920778.0,
"step": 1155
},
{
"entropy": 7.354297304153443,
"epoch": 0.13387189844200809,
"grad_norm": 0.921875,
"learning_rate": 0.0004999961735983871,
"loss": 7.1539,
"mean_token_accuracy": 0.14073749631643295,
"num_tokens": 2933754.0,
"step": 1160
},
{
"entropy": 7.476262760162354,
"epoch": 0.13444893248701673,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999959291610639,
"loss": 7.2608,
"mean_token_accuracy": 0.13915467262268066,
"num_tokens": 2946019.0,
"step": 1165
},
{
"entropy": 7.351610231399536,
"epoch": 0.1350259665320254,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999956771561143,
"loss": 7.2192,
"mean_token_accuracy": 0.13505645617842674,
"num_tokens": 2959688.0,
"step": 1170
},
{
"entropy": 7.36425518989563,
"epoch": 0.13560300057703403,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999954175835469,
"loss": 7.1361,
"mean_token_accuracy": 0.14523073583841323,
"num_tokens": 2971211.0,
"step": 1175
},
{
"entropy": 7.323524904251099,
"epoch": 0.1361800346220427,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999951504433703,
"loss": 7.2293,
"mean_token_accuracy": 0.13175321519374847,
"num_tokens": 2983559.0,
"step": 1180
},
{
"entropy": 7.404617214202881,
"epoch": 0.13675706866705137,
"grad_norm": 1.0625,
"learning_rate": 0.0004999948757355935,
"loss": 7.2025,
"mean_token_accuracy": 0.1324743650853634,
"num_tokens": 2995053.0,
"step": 1185
},
{
"entropy": 7.352153348922729,
"epoch": 0.13733410271206,
"grad_norm": 1.03125,
"learning_rate": 0.000499994593460226,
"loss": 7.1457,
"mean_token_accuracy": 0.13915842920541763,
"num_tokens": 3006143.0,
"step": 1190
},
{
"entropy": 7.277075576782226,
"epoch": 0.13791113675706868,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999943036172771,
"loss": 7.147,
"mean_token_accuracy": 0.14147392511367798,
"num_tokens": 3018793.0,
"step": 1195
},
{
"entropy": 7.392578411102295,
"epoch": 0.13848817080207732,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999940062067565,
"loss": 7.1464,
"mean_token_accuracy": 0.13925720006227493,
"num_tokens": 3031547.0,
"step": 1200
},
{
"entropy": 7.269504356384277,
"epoch": 0.13906520484708598,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999937012286742,
"loss": 7.1604,
"mean_token_accuracy": 0.13719142302870752,
"num_tokens": 3044035.0,
"step": 1205
},
{
"entropy": 7.297099208831787,
"epoch": 0.13964223889209462,
"grad_norm": 0.890625,
"learning_rate": 0.0004999933886830405,
"loss": 7.1862,
"mean_token_accuracy": 0.14078905582427978,
"num_tokens": 3056238.0,
"step": 1210
},
{
"entropy": 7.321411800384522,
"epoch": 0.1402192729371033,
"grad_norm": 0.90234375,
"learning_rate": 0.000499993068569866,
"loss": 7.1833,
"mean_token_accuracy": 0.14090409278869628,
"num_tokens": 3069250.0,
"step": 1215
},
{
"entropy": 7.357383728027344,
"epoch": 0.14079630698211196,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999927408891614,
"loss": 7.1557,
"mean_token_accuracy": 0.13715188950300217,
"num_tokens": 3082048.0,
"step": 1220
},
{
"entropy": 7.301809167861938,
"epoch": 0.1413733410271206,
"grad_norm": 0.9375,
"learning_rate": 0.0004999924056409378,
"loss": 7.2215,
"mean_token_accuracy": 0.13659722208976746,
"num_tokens": 3094588.0,
"step": 1225
},
{
"entropy": 7.427079439163208,
"epoch": 0.14195037507212926,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999920628252063,
"loss": 7.2441,
"mean_token_accuracy": 0.128413824737072,
"num_tokens": 3108067.0,
"step": 1230
},
{
"entropy": 7.301552772521973,
"epoch": 0.1425274091171379,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999917124419785,
"loss": 7.1715,
"mean_token_accuracy": 0.1430498704314232,
"num_tokens": 3120524.0,
"step": 1235
},
{
"entropy": 7.375691175460815,
"epoch": 0.14310444316214657,
"grad_norm": 0.96875,
"learning_rate": 0.0004999913544912664,
"loss": 7.2377,
"mean_token_accuracy": 0.13174993246793748,
"num_tokens": 3133533.0,
"step": 1240
},
{
"entropy": 7.397565317153931,
"epoch": 0.1436814772071552,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999909889730817,
"loss": 7.2182,
"mean_token_accuracy": 0.137331123650074,
"num_tokens": 3146183.0,
"step": 1245
},
{
"entropy": 7.296873569488525,
"epoch": 0.14425851125216388,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999906158874368,
"loss": 7.1694,
"mean_token_accuracy": 0.13684176057577133,
"num_tokens": 3158436.0,
"step": 1250
},
{
"entropy": 7.299059915542602,
"epoch": 0.14483554529717255,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999902352343444,
"loss": 7.1194,
"mean_token_accuracy": 0.1443310186266899,
"num_tokens": 3170426.0,
"step": 1255
},
{
"entropy": 7.344976425170898,
"epoch": 0.1454125793421812,
"grad_norm": 0.90625,
"learning_rate": 0.0004999898470138171,
"loss": 7.1264,
"mean_token_accuracy": 0.13932598382234573,
"num_tokens": 3182865.0,
"step": 1260
},
{
"entropy": 7.26788363456726,
"epoch": 0.14598961338718985,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999894512258681,
"loss": 7.1381,
"mean_token_accuracy": 0.13955464959144592,
"num_tokens": 3195674.0,
"step": 1265
},
{
"entropy": 7.33721137046814,
"epoch": 0.1465666474321985,
"grad_norm": 0.84375,
"learning_rate": 0.0004999890478705107,
"loss": 7.114,
"mean_token_accuracy": 0.14049444273114203,
"num_tokens": 3208848.0,
"step": 1270
},
{
"entropy": 7.3063184261322025,
"epoch": 0.14714368147720716,
"grad_norm": 0.921875,
"learning_rate": 0.0004999886369477585,
"loss": 7.2017,
"mean_token_accuracy": 0.1356468491256237,
"num_tokens": 3221179.0,
"step": 1275
},
{
"entropy": 7.336249351501465,
"epoch": 0.1477207155222158,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999882184576251,
"loss": 7.2815,
"mean_token_accuracy": 0.13257396966218948,
"num_tokens": 3233671.0,
"step": 1280
},
{
"entropy": 7.405120134353638,
"epoch": 0.14829774956722447,
"grad_norm": 0.84375,
"learning_rate": 0.0004999877924001248,
"loss": 7.1979,
"mean_token_accuracy": 0.13243449851870537,
"num_tokens": 3247219.0,
"step": 1285
},
{
"entropy": 7.313788509368896,
"epoch": 0.1488747836122331,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999873587752718,
"loss": 7.1673,
"mean_token_accuracy": 0.13630512207746506,
"num_tokens": 3259444.0,
"step": 1290
},
{
"entropy": 7.180786943435669,
"epoch": 0.14945181765724178,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999869175830808,
"loss": 7.019,
"mean_token_accuracy": 0.14473430812358856,
"num_tokens": 3272092.0,
"step": 1295
},
{
"entropy": 7.377786588668823,
"epoch": 0.15002885170225044,
"grad_norm": 0.875,
"learning_rate": 0.0004999864688235666,
"loss": 7.1995,
"mean_token_accuracy": 0.12725819498300553,
"num_tokens": 3284260.0,
"step": 1300
},
{
"entropy": 7.249001407623291,
"epoch": 0.15060588574725908,
"grad_norm": 0.90625,
"learning_rate": 0.0004999860124967442,
"loss": 7.1613,
"mean_token_accuracy": 0.13158362433314325,
"num_tokens": 3296662.0,
"step": 1305
},
{
"entropy": 7.297910118103028,
"epoch": 0.15118291979226775,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999855486026291,
"loss": 7.1213,
"mean_token_accuracy": 0.14045775979757308,
"num_tokens": 3307782.0,
"step": 1310
},
{
"entropy": 7.283909845352173,
"epoch": 0.1517599538372764,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999850771412369,
"loss": 7.1728,
"mean_token_accuracy": 0.13270583972334862,
"num_tokens": 3320519.0,
"step": 1315
},
{
"entropy": 7.31526288986206,
"epoch": 0.15233698788228506,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999845981125832,
"loss": 7.0413,
"mean_token_accuracy": 0.14500000029802323,
"num_tokens": 3331722.0,
"step": 1320
},
{
"entropy": 7.212509202957153,
"epoch": 0.1529140219272937,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999841115166844,
"loss": 7.1385,
"mean_token_accuracy": 0.13171057179570198,
"num_tokens": 3345143.0,
"step": 1325
},
{
"entropy": 7.255769968032837,
"epoch": 0.15349105597230236,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999836173535568,
"loss": 7.0707,
"mean_token_accuracy": 0.14495590403676034,
"num_tokens": 3358946.0,
"step": 1330
},
{
"entropy": 7.313462114334106,
"epoch": 0.15406809001731103,
"grad_norm": 0.90625,
"learning_rate": 0.0004999831156232169,
"loss": 7.1788,
"mean_token_accuracy": 0.13734060302376747,
"num_tokens": 3372465.0,
"step": 1335
},
{
"entropy": 7.221960115432739,
"epoch": 0.15464512406231967,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999826063256818,
"loss": 7.0636,
"mean_token_accuracy": 0.1419673442840576,
"num_tokens": 3384392.0,
"step": 1340
},
{
"entropy": 7.322636604309082,
"epoch": 0.15522215810732834,
"grad_norm": 0.80859375,
"learning_rate": 0.0004999820894609683,
"loss": 7.1346,
"mean_token_accuracy": 0.12882967665791512,
"num_tokens": 3398137.0,
"step": 1345
},
{
"entropy": 7.281394720077515,
"epoch": 0.15579919215233698,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999815650290941,
"loss": 7.1003,
"mean_token_accuracy": 0.13911654129624368,
"num_tokens": 3410353.0,
"step": 1350
},
{
"entropy": 7.242393922805786,
"epoch": 0.15637622619734565,
"grad_norm": 1.015625,
"learning_rate": 0.0004999810330300766,
"loss": 7.0515,
"mean_token_accuracy": 0.13970422446727754,
"num_tokens": 3423206.0,
"step": 1355
},
{
"entropy": 7.206804037094116,
"epoch": 0.1569532602423543,
"grad_norm": 0.828125,
"learning_rate": 0.000499980493463934,
"loss": 7.1134,
"mean_token_accuracy": 0.1346282660961151,
"num_tokens": 3436780.0,
"step": 1360
},
{
"entropy": 7.3222535133361815,
"epoch": 0.15753029428736295,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999799463306841,
"loss": 7.1791,
"mean_token_accuracy": 0.13330562338232993,
"num_tokens": 3450285.0,
"step": 1365
},
{
"entropy": 7.289401626586914,
"epoch": 0.15810732833237162,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999793916303455,
"loss": 7.0678,
"mean_token_accuracy": 0.14209839105606079,
"num_tokens": 3462157.0,
"step": 1370
},
{
"entropy": 7.195488595962525,
"epoch": 0.15868436237738026,
"grad_norm": 0.9375,
"learning_rate": 0.0004999788293629368,
"loss": 7.0299,
"mean_token_accuracy": 0.14806710481643676,
"num_tokens": 3475027.0,
"step": 1375
},
{
"entropy": 7.218692922592163,
"epoch": 0.15926139642238893,
"grad_norm": 0.8984375,
"learning_rate": 0.000499978259528477,
"loss": 7.1068,
"mean_token_accuracy": 0.14241924211382867,
"num_tokens": 3489111.0,
"step": 1380
},
{
"entropy": 7.240436649322509,
"epoch": 0.15983843046739757,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999776821269852,
"loss": 7.0521,
"mean_token_accuracy": 0.13724546879529953,
"num_tokens": 3501673.0,
"step": 1385
},
{
"entropy": 7.110876035690308,
"epoch": 0.16041546451240624,
"grad_norm": 0.90625,
"learning_rate": 0.0004999770971584807,
"loss": 6.9797,
"mean_token_accuracy": 0.1431804358959198,
"num_tokens": 3513511.0,
"step": 1390
},
{
"entropy": 7.26352219581604,
"epoch": 0.16099249855741488,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999765046229834,
"loss": 7.1193,
"mean_token_accuracy": 0.13453444615006446,
"num_tokens": 3525218.0,
"step": 1395
},
{
"entropy": 7.1966602325439455,
"epoch": 0.16156953260242354,
"grad_norm": 0.9765625,
"learning_rate": 0.000499975904520513,
"loss": 6.9928,
"mean_token_accuracy": 0.1506774343550205,
"num_tokens": 3537006.0,
"step": 1400
},
{
"entropy": 7.221553230285645,
"epoch": 0.1621465666474322,
"grad_norm": 0.93359375,
"learning_rate": 0.00049997529685109,
"loss": 7.1169,
"mean_token_accuracy": 0.13538158535957337,
"num_tokens": 3549873.0,
"step": 1405
},
{
"entropy": 7.317210006713867,
"epoch": 0.16272360069244085,
"grad_norm": 1.0,
"learning_rate": 0.0004999746816147344,
"loss": 7.1801,
"mean_token_accuracy": 0.1276666961610317,
"num_tokens": 3563542.0,
"step": 1410
},
{
"entropy": 7.172958326339722,
"epoch": 0.16330063473744952,
"grad_norm": 0.80078125,
"learning_rate": 0.0004999740588114673,
"loss": 7.0726,
"mean_token_accuracy": 0.13991388827562332,
"num_tokens": 3575679.0,
"step": 1415
},
{
"entropy": 7.290738105773926,
"epoch": 0.16387766878245816,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999734284413095,
"loss": 7.0954,
"mean_token_accuracy": 0.14116299375891686,
"num_tokens": 3588605.0,
"step": 1420
},
{
"entropy": 7.13003716468811,
"epoch": 0.16445470282746683,
"grad_norm": 0.9453125,
"learning_rate": 0.000499972790504282,
"loss": 7.0764,
"mean_token_accuracy": 0.1506669543683529,
"num_tokens": 3601894.0,
"step": 1425
},
{
"entropy": 7.193125915527344,
"epoch": 0.16503173687247547,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999721450004067,
"loss": 7.0103,
"mean_token_accuracy": 0.14832852333784102,
"num_tokens": 3614436.0,
"step": 1430
},
{
"entropy": 7.113531923294067,
"epoch": 0.16560877091748413,
"grad_norm": 0.890625,
"learning_rate": 0.0004999714919297049,
"loss": 6.9698,
"mean_token_accuracy": 0.14278605505824088,
"num_tokens": 3627590.0,
"step": 1435
},
{
"entropy": 7.303047561645508,
"epoch": 0.1661858049624928,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999708312921987,
"loss": 7.1398,
"mean_token_accuracy": 0.1305020570755005,
"num_tokens": 3640834.0,
"step": 1440
},
{
"entropy": 7.112206935882568,
"epoch": 0.16676283900750144,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999701630879103,
"loss": 6.9822,
"mean_token_accuracy": 0.14259293228387832,
"num_tokens": 3653024.0,
"step": 1445
},
{
"entropy": 7.1947370052337645,
"epoch": 0.1673398730525101,
"grad_norm": 1.015625,
"learning_rate": 0.0004999694873168623,
"loss": 7.0366,
"mean_token_accuracy": 0.1446026474237442,
"num_tokens": 3665311.0,
"step": 1450
},
{
"entropy": 7.208151340484619,
"epoch": 0.16791690709751875,
"grad_norm": 0.80078125,
"learning_rate": 0.0004999688039790773,
"loss": 7.0803,
"mean_token_accuracy": 0.1384113185107708,
"num_tokens": 3679102.0,
"step": 1455
},
{
"entropy": 7.164588069915771,
"epoch": 0.16849394114252741,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999681130745784,
"loss": 6.9906,
"mean_token_accuracy": 0.14679074734449388,
"num_tokens": 3690046.0,
"step": 1460
},
{
"entropy": 7.1119578838348385,
"epoch": 0.16907097518753605,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999674146033888,
"loss": 7.0146,
"mean_token_accuracy": 0.1386751838028431,
"num_tokens": 3703668.0,
"step": 1465
},
{
"entropy": 7.215666484832764,
"epoch": 0.16964800923254472,
"grad_norm": 0.8125,
"learning_rate": 0.0004999667085655318,
"loss": 7.0317,
"mean_token_accuracy": 0.142906092107296,
"num_tokens": 3715852.0,
"step": 1470
},
{
"entropy": 7.163801479339599,
"epoch": 0.1702250432775534,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999659949610313,
"loss": 6.9624,
"mean_token_accuracy": 0.1409413382411003,
"num_tokens": 3728310.0,
"step": 1475
},
{
"entropy": 7.225057554244995,
"epoch": 0.17080207732256203,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999652737899114,
"loss": 7.0428,
"mean_token_accuracy": 0.13530816361308098,
"num_tokens": 3741323.0,
"step": 1480
},
{
"entropy": 7.164874649047851,
"epoch": 0.1713791113675707,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999645450521963,
"loss": 7.0429,
"mean_token_accuracy": 0.13799375295639038,
"num_tokens": 3752667.0,
"step": 1485
},
{
"entropy": 7.169693231582642,
"epoch": 0.17195614541257934,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999638087479104,
"loss": 7.0422,
"mean_token_accuracy": 0.13898203894495964,
"num_tokens": 3764282.0,
"step": 1490
},
{
"entropy": 7.190469121932983,
"epoch": 0.172533179457588,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999630648770786,
"loss": 7.0687,
"mean_token_accuracy": 0.13370918706059456,
"num_tokens": 3778280.0,
"step": 1495
},
{
"entropy": 7.144445991516113,
"epoch": 0.17311021350259664,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999623134397257,
"loss": 7.0879,
"mean_token_accuracy": 0.14445693269371987,
"num_tokens": 3791010.0,
"step": 1500
},
{
"entropy": 7.24975266456604,
"epoch": 0.1736872475476053,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999615544358774,
"loss": 7.0116,
"mean_token_accuracy": 0.13824615478515626,
"num_tokens": 3802738.0,
"step": 1505
},
{
"entropy": 7.121045970916748,
"epoch": 0.17426428159261395,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999607878655587,
"loss": 7.0366,
"mean_token_accuracy": 0.14151245802640916,
"num_tokens": 3815448.0,
"step": 1510
},
{
"entropy": 7.1481671810150145,
"epoch": 0.17484131563762262,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999600137287958,
"loss": 7.0332,
"mean_token_accuracy": 0.1367475040256977,
"num_tokens": 3829833.0,
"step": 1515
},
{
"entropy": 7.121943235397339,
"epoch": 0.17541834968263129,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999592320256146,
"loss": 7.0023,
"mean_token_accuracy": 0.1391633428633213,
"num_tokens": 3841544.0,
"step": 1520
},
{
"entropy": 7.198780822753906,
"epoch": 0.17599538372763993,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999584427560412,
"loss": 6.9958,
"mean_token_accuracy": 0.13729065656661987,
"num_tokens": 3853886.0,
"step": 1525
},
{
"entropy": 7.095717239379883,
"epoch": 0.1765724177726486,
"grad_norm": 0.83984375,
"learning_rate": 0.0004999576459201024,
"loss": 7.0104,
"mean_token_accuracy": 0.1445713609457016,
"num_tokens": 3867942.0,
"step": 1530
},
{
"entropy": 7.133176279067993,
"epoch": 0.17714945181765723,
"grad_norm": 0.96484375,
"learning_rate": 0.000499956841517825,
"loss": 7.0257,
"mean_token_accuracy": 0.14167340844869614,
"num_tokens": 3881764.0,
"step": 1535
},
{
"entropy": 7.127892589569091,
"epoch": 0.1777264858626659,
"grad_norm": 0.875,
"learning_rate": 0.000499956029549236,
"loss": 6.9289,
"mean_token_accuracy": 0.1465323746204376,
"num_tokens": 3893979.0,
"step": 1540
},
{
"entropy": 7.17246675491333,
"epoch": 0.17830351990767454,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999552100143625,
"loss": 7.0449,
"mean_token_accuracy": 0.13901807218790055,
"num_tokens": 3905934.0,
"step": 1545
},
{
"entropy": 7.145467090606689,
"epoch": 0.1788805539526832,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999543829132324,
"loss": 6.9948,
"mean_token_accuracy": 0.14433147236704827,
"num_tokens": 3918348.0,
"step": 1550
},
{
"entropy": 7.112019491195679,
"epoch": 0.17945758799769188,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999535482458734,
"loss": 6.9814,
"mean_token_accuracy": 0.13943516984581947,
"num_tokens": 3931863.0,
"step": 1555
},
{
"entropy": 7.105517101287842,
"epoch": 0.18003462204270052,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999527060123135,
"loss": 6.9699,
"mean_token_accuracy": 0.1462463617324829,
"num_tokens": 3945108.0,
"step": 1560
},
{
"entropy": 7.1699309825897215,
"epoch": 0.18061165608770918,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999518562125811,
"loss": 6.9963,
"mean_token_accuracy": 0.13957771062850952,
"num_tokens": 3956695.0,
"step": 1565
},
{
"entropy": 7.086351537704468,
"epoch": 0.18118869013271782,
"grad_norm": 0.76171875,
"learning_rate": 0.0004999509988467047,
"loss": 6.9873,
"mean_token_accuracy": 0.1385016493499279,
"num_tokens": 3970458.0,
"step": 1570
},
{
"entropy": 7.148052930831909,
"epoch": 0.1817657241777265,
"grad_norm": 1.0,
"learning_rate": 0.0004999501339147132,
"loss": 6.9739,
"mean_token_accuracy": 0.13998864516615867,
"num_tokens": 3981962.0,
"step": 1575
},
{
"entropy": 7.1447553634643555,
"epoch": 0.18234275822273513,
"grad_norm": 0.890625,
"learning_rate": 0.0004999492614166357,
"loss": 6.982,
"mean_token_accuracy": 0.13720383271574974,
"num_tokens": 3993512.0,
"step": 1580
},
{
"entropy": 7.046021413803101,
"epoch": 0.1829197922677438,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999483813525014,
"loss": 6.8679,
"mean_token_accuracy": 0.15238967761397362,
"num_tokens": 4005826.0,
"step": 1585
},
{
"entropy": 7.077240610122681,
"epoch": 0.18349682631275246,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999474937223403,
"loss": 6.9461,
"mean_token_accuracy": 0.14126977398991586,
"num_tokens": 4018927.0,
"step": 1590
},
{
"entropy": 7.139278173446655,
"epoch": 0.1840738603577611,
"grad_norm": 0.796875,
"learning_rate": 0.0004999465985261818,
"loss": 6.9951,
"mean_token_accuracy": 0.13713297992944717,
"num_tokens": 4030970.0,
"step": 1595
},
{
"entropy": 7.041838884353638,
"epoch": 0.18465089440276977,
"grad_norm": 0.84375,
"learning_rate": 0.0004999456957640562,
"loss": 6.9665,
"mean_token_accuracy": 0.14336296245455743,
"num_tokens": 4043396.0,
"step": 1600
},
{
"entropy": 7.167615985870361,
"epoch": 0.1852279284477784,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999447854359939,
"loss": 6.9946,
"mean_token_accuracy": 0.13789970949292182,
"num_tokens": 4055588.0,
"step": 1605
},
{
"entropy": 7.082289409637451,
"epoch": 0.18580496249278708,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999438675420255,
"loss": 6.9775,
"mean_token_accuracy": 0.1350480891764164,
"num_tokens": 4068036.0,
"step": 1610
},
{
"entropy": 7.123587226867675,
"epoch": 0.18638199653779572,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999429420821818,
"loss": 7.015,
"mean_token_accuracy": 0.1355690099298954,
"num_tokens": 4080531.0,
"step": 1615
},
{
"entropy": 7.1394895076751705,
"epoch": 0.1869590305828044,
"grad_norm": 0.80078125,
"learning_rate": 0.000499942009056494,
"loss": 6.9259,
"mean_token_accuracy": 0.14576323702931404,
"num_tokens": 4093614.0,
"step": 1620
},
{
"entropy": 7.157642984390259,
"epoch": 0.18753606462781305,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999410684649935,
"loss": 7.0282,
"mean_token_accuracy": 0.13709268644452094,
"num_tokens": 4106836.0,
"step": 1625
},
{
"entropy": 7.0767253875732425,
"epoch": 0.1881130986728217,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999401203077119,
"loss": 6.9509,
"mean_token_accuracy": 0.14356981962919235,
"num_tokens": 4118739.0,
"step": 1630
},
{
"entropy": 7.107544803619385,
"epoch": 0.18869013271783036,
"grad_norm": 0.859375,
"learning_rate": 0.000499939164584681,
"loss": 6.9214,
"mean_token_accuracy": 0.14475391134619714,
"num_tokens": 4131174.0,
"step": 1635
},
{
"entropy": 7.091096496582031,
"epoch": 0.189267166762839,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999382012959331,
"loss": 6.9713,
"mean_token_accuracy": 0.13550762236118316,
"num_tokens": 4143459.0,
"step": 1640
},
{
"entropy": 7.121346950531006,
"epoch": 0.18984420080784767,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999372304415005,
"loss": 7.0043,
"mean_token_accuracy": 0.13536836579442024,
"num_tokens": 4155864.0,
"step": 1645
},
{
"entropy": 6.936057138442993,
"epoch": 0.1904212348528563,
"grad_norm": 0.890625,
"learning_rate": 0.0004999362520214159,
"loss": 6.8156,
"mean_token_accuracy": 0.14860668629407883,
"num_tokens": 4168433.0,
"step": 1650
},
{
"entropy": 7.137879848480225,
"epoch": 0.19099826889786498,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999352660357122,
"loss": 6.9852,
"mean_token_accuracy": 0.13815735951066016,
"num_tokens": 4180852.0,
"step": 1655
},
{
"entropy": 7.092313289642334,
"epoch": 0.19157530294287364,
"grad_norm": 0.953125,
"learning_rate": 0.0004999342724844226,
"loss": 6.9307,
"mean_token_accuracy": 0.14323657527565956,
"num_tokens": 4194712.0,
"step": 1660
},
{
"entropy": 7.107403707504273,
"epoch": 0.19215233698788228,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999332713675804,
"loss": 6.9964,
"mean_token_accuracy": 0.14108359590172767,
"num_tokens": 4207970.0,
"step": 1665
},
{
"entropy": 7.045479440689087,
"epoch": 0.19272937103289095,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999322626852193,
"loss": 6.9161,
"mean_token_accuracy": 0.1442576117813587,
"num_tokens": 4220893.0,
"step": 1670
},
{
"entropy": 7.002764892578125,
"epoch": 0.1933064050778996,
"grad_norm": 0.890625,
"learning_rate": 0.0004999312464373734,
"loss": 6.8981,
"mean_token_accuracy": 0.13980508148670195,
"num_tokens": 4232463.0,
"step": 1675
},
{
"entropy": 7.067962741851806,
"epoch": 0.19388343912290826,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999302226240767,
"loss": 6.8634,
"mean_token_accuracy": 0.1389753997325897,
"num_tokens": 4244604.0,
"step": 1680
},
{
"entropy": 7.014446783065796,
"epoch": 0.1944604731679169,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999291912453637,
"loss": 6.9458,
"mean_token_accuracy": 0.14495371729135514,
"num_tokens": 4257260.0,
"step": 1685
},
{
"entropy": 7.177657508850098,
"epoch": 0.19503750721292556,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999281523012691,
"loss": 6.9024,
"mean_token_accuracy": 0.13922334164381028,
"num_tokens": 4269885.0,
"step": 1690
},
{
"entropy": 7.044678115844727,
"epoch": 0.1956145412579342,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999271057918278,
"loss": 7.0126,
"mean_token_accuracy": 0.13741706758737565,
"num_tokens": 4281560.0,
"step": 1695
},
{
"entropy": 7.127787733078003,
"epoch": 0.19619157530294287,
"grad_norm": 1.046875,
"learning_rate": 0.0004999260517170751,
"loss": 6.8977,
"mean_token_accuracy": 0.14665021896362304,
"num_tokens": 4293746.0,
"step": 1700
},
{
"entropy": 7.0871453285217285,
"epoch": 0.19676860934795154,
"grad_norm": 0.921875,
"learning_rate": 0.0004999249900770463,
"loss": 6.9355,
"mean_token_accuracy": 0.1397123083472252,
"num_tokens": 4306595.0,
"step": 1705
},
{
"entropy": 7.123568153381347,
"epoch": 0.19734564339296018,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999239208717772,
"loss": 6.8848,
"mean_token_accuracy": 0.14523950070142747,
"num_tokens": 4318705.0,
"step": 1710
},
{
"entropy": 6.925599145889282,
"epoch": 0.19792267743796885,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999228441013037,
"loss": 6.9293,
"mean_token_accuracy": 0.14079967737197877,
"num_tokens": 4331321.0,
"step": 1715
},
{
"entropy": 7.063972091674804,
"epoch": 0.1984997114829775,
"grad_norm": 0.93359375,
"learning_rate": 0.000499921759765662,
"loss": 6.8839,
"mean_token_accuracy": 0.14801763668656348,
"num_tokens": 4343517.0,
"step": 1720
},
{
"entropy": 7.0327746868133545,
"epoch": 0.19907674552798615,
"grad_norm": 0.8125,
"learning_rate": 0.0004999206678648888,
"loss": 6.9032,
"mean_token_accuracy": 0.14797452092170715,
"num_tokens": 4356342.0,
"step": 1725
},
{
"entropy": 6.983541250228882,
"epoch": 0.1996537795729948,
"grad_norm": 0.984375,
"learning_rate": 0.0004999195683990206,
"loss": 6.8887,
"mean_token_accuracy": 0.1441090725362301,
"num_tokens": 4369078.0,
"step": 1730
},
{
"entropy": 7.019094848632813,
"epoch": 0.20023081361800346,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999184613680945,
"loss": 6.8854,
"mean_token_accuracy": 0.13662683367729186,
"num_tokens": 4382091.0,
"step": 1735
},
{
"entropy": 7.010761690139771,
"epoch": 0.20080784766301213,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999173467721476,
"loss": 6.9208,
"mean_token_accuracy": 0.1393657937645912,
"num_tokens": 4393919.0,
"step": 1740
},
{
"entropy": 7.006711435317993,
"epoch": 0.20138488170802077,
"grad_norm": 0.80078125,
"learning_rate": 0.0004999162246112175,
"loss": 6.8789,
"mean_token_accuracy": 0.1454889804124832,
"num_tokens": 4407054.0,
"step": 1745
},
{
"entropy": 7.003733777999878,
"epoch": 0.20196191575302944,
"grad_norm": 1.375,
"learning_rate": 0.0004999150948853419,
"loss": 6.94,
"mean_token_accuracy": 0.137217216193676,
"num_tokens": 4420180.0,
"step": 1750
},
{
"entropy": 7.14780068397522,
"epoch": 0.20253894979803808,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999139575945587,
"loss": 6.9242,
"mean_token_accuracy": 0.150433312356472,
"num_tokens": 4433248.0,
"step": 1755
},
{
"entropy": 6.930108451843262,
"epoch": 0.20311598384304674,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999128127389065,
"loss": 6.9106,
"mean_token_accuracy": 0.14306253790855408,
"num_tokens": 4446340.0,
"step": 1760
},
{
"entropy": 7.059494829177856,
"epoch": 0.20369301788805538,
"grad_norm": 0.88671875,
"learning_rate": 0.0004999116603184233,
"loss": 6.9249,
"mean_token_accuracy": 0.140160171687603,
"num_tokens": 4459918.0,
"step": 1765
},
{
"entropy": 6.981036758422851,
"epoch": 0.20427005193306405,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999105003331482,
"loss": 6.8797,
"mean_token_accuracy": 0.1448831871151924,
"num_tokens": 4472020.0,
"step": 1770
},
{
"entropy": 7.015055799484253,
"epoch": 0.20484708597807272,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999093327831202,
"loss": 6.8445,
"mean_token_accuracy": 0.14826457872986792,
"num_tokens": 4484269.0,
"step": 1775
},
{
"entropy": 6.998699283599853,
"epoch": 0.20542412002308136,
"grad_norm": 0.890625,
"learning_rate": 0.0004999081576683784,
"loss": 6.9411,
"mean_token_accuracy": 0.13955088704824448,
"num_tokens": 4498040.0,
"step": 1780
},
{
"entropy": 7.111977243423462,
"epoch": 0.20600115406809003,
"grad_norm": 1.078125,
"learning_rate": 0.0004999069749889626,
"loss": 6.8938,
"mean_token_accuracy": 0.14736564159393312,
"num_tokens": 4509701.0,
"step": 1785
},
{
"entropy": 6.967676115036011,
"epoch": 0.20657818811309867,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999057847449123,
"loss": 6.9249,
"mean_token_accuracy": 0.13947466611862183,
"num_tokens": 4521840.0,
"step": 1790
},
{
"entropy": 7.0544905185699465,
"epoch": 0.20715522215810733,
"grad_norm": 0.9375,
"learning_rate": 0.0004999045869362678,
"loss": 6.9169,
"mean_token_accuracy": 0.1435271330177784,
"num_tokens": 4534241.0,
"step": 1795
},
{
"entropy": 7.035863065719605,
"epoch": 0.20773225620311597,
"grad_norm": 0.96484375,
"learning_rate": 0.000499903381563069,
"loss": 6.8669,
"mean_token_accuracy": 0.1447395406663418,
"num_tokens": 4547255.0,
"step": 1800
},
{
"entropy": 7.061683273315429,
"epoch": 0.20830929024812464,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999021686253568,
"loss": 6.902,
"mean_token_accuracy": 0.14312518313527106,
"num_tokens": 4559538.0,
"step": 1805
},
{
"entropy": 6.928671979904175,
"epoch": 0.2088863242931333,
"grad_norm": 0.89453125,
"learning_rate": 0.0004999009481231719,
"loss": 6.8802,
"mean_token_accuracy": 0.14481844902038574,
"num_tokens": 4571444.0,
"step": 1810
},
{
"entropy": 6.93266978263855,
"epoch": 0.20946335833814195,
"grad_norm": 1.2890625,
"learning_rate": 0.0004998997200565553,
"loss": 6.7182,
"mean_token_accuracy": 0.15741085708141328,
"num_tokens": 4586294.0,
"step": 1815
},
{
"entropy": 7.03538408279419,
"epoch": 0.21004039238315061,
"grad_norm": 0.83203125,
"learning_rate": 0.0004998984844255483,
"loss": 6.8886,
"mean_token_accuracy": 0.14464563131332397,
"num_tokens": 4599661.0,
"step": 1820
},
{
"entropy": 6.98320894241333,
"epoch": 0.21061742642815925,
"grad_norm": 0.83203125,
"learning_rate": 0.0004998972412301925,
"loss": 6.8964,
"mean_token_accuracy": 0.14106034338474274,
"num_tokens": 4611305.0,
"step": 1825
},
{
"entropy": 7.015894365310669,
"epoch": 0.21119446047316792,
"grad_norm": 0.890625,
"learning_rate": 0.0004998959904705297,
"loss": 6.9004,
"mean_token_accuracy": 0.13522190302610398,
"num_tokens": 4623992.0,
"step": 1830
},
{
"entropy": 7.014666414260864,
"epoch": 0.21177149451817656,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998947321466021,
"loss": 6.8942,
"mean_token_accuracy": 0.1404628872871399,
"num_tokens": 4636495.0,
"step": 1835
},
{
"entropy": 6.996430206298828,
"epoch": 0.21234852856318523,
"grad_norm": 0.859375,
"learning_rate": 0.0004998934662584518,
"loss": 6.8858,
"mean_token_accuracy": 0.14315241277217866,
"num_tokens": 4650266.0,
"step": 1840
},
{
"entropy": 7.023065519332886,
"epoch": 0.2129255626081939,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998921928061214,
"loss": 6.8519,
"mean_token_accuracy": 0.14534742385149002,
"num_tokens": 4663238.0,
"step": 1845
},
{
"entropy": 6.998432397842407,
"epoch": 0.21350259665320254,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998909117896539,
"loss": 6.8968,
"mean_token_accuracy": 0.14261196106672286,
"num_tokens": 4675501.0,
"step": 1850
},
{
"entropy": 6.991396522521972,
"epoch": 0.2140796306982112,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998896232090922,
"loss": 6.8619,
"mean_token_accuracy": 0.13769274204969406,
"num_tokens": 4688996.0,
"step": 1855
},
{
"entropy": 7.000917768478393,
"epoch": 0.21465666474321984,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998883270644798,
"loss": 6.9589,
"mean_token_accuracy": 0.138343658298254,
"num_tokens": 4703226.0,
"step": 1860
},
{
"entropy": 7.035671234130859,
"epoch": 0.2152336987882285,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998870233558602,
"loss": 6.8455,
"mean_token_accuracy": 0.14217291474342347,
"num_tokens": 4715825.0,
"step": 1865
},
{
"entropy": 6.937395477294922,
"epoch": 0.21581073283323715,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998857120832774,
"loss": 6.842,
"mean_token_accuracy": 0.14556335806846618,
"num_tokens": 4727819.0,
"step": 1870
},
{
"entropy": 7.042612791061401,
"epoch": 0.21638776687824582,
"grad_norm": 0.765625,
"learning_rate": 0.0004998843932467751,
"loss": 6.875,
"mean_token_accuracy": 0.14139556288719177,
"num_tokens": 4742890.0,
"step": 1875
},
{
"entropy": 6.9743971824646,
"epoch": 0.21696480092325446,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998830668463982,
"loss": 6.8641,
"mean_token_accuracy": 0.1403570555150509,
"num_tokens": 4755609.0,
"step": 1880
},
{
"entropy": 6.923630332946777,
"epoch": 0.21754183496826313,
"grad_norm": 0.859375,
"learning_rate": 0.0004998817328821909,
"loss": 6.842,
"mean_token_accuracy": 0.14689554423093795,
"num_tokens": 4766674.0,
"step": 1885
},
{
"entropy": 6.97991976737976,
"epoch": 0.2181188690132718,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998803913541983,
"loss": 6.8786,
"mean_token_accuracy": 0.14042821601033212,
"num_tokens": 4779639.0,
"step": 1890
},
{
"entropy": 7.112171745300293,
"epoch": 0.21869590305828043,
"grad_norm": 1.09375,
"learning_rate": 0.0004998790422624654,
"loss": 6.9663,
"mean_token_accuracy": 0.13750633150339125,
"num_tokens": 4791702.0,
"step": 1895
},
{
"entropy": 6.947701549530029,
"epoch": 0.2192729371032891,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998776856070376,
"loss": 6.7836,
"mean_token_accuracy": 0.15090153217315674,
"num_tokens": 4805756.0,
"step": 1900
},
{
"entropy": 6.948662805557251,
"epoch": 0.21984997114829774,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998763213879606,
"loss": 6.843,
"mean_token_accuracy": 0.145341220498085,
"num_tokens": 4817941.0,
"step": 1905
},
{
"entropy": 6.965862226486206,
"epoch": 0.2204270051933064,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998749496052803,
"loss": 6.7269,
"mean_token_accuracy": 0.14902952387928964,
"num_tokens": 4832298.0,
"step": 1910
},
{
"entropy": 6.848330736160278,
"epoch": 0.22100403923831505,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998735702590426,
"loss": 6.8195,
"mean_token_accuracy": 0.1508714735507965,
"num_tokens": 4844814.0,
"step": 1915
},
{
"entropy": 7.058192014694214,
"epoch": 0.22158107328332372,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998721833492942,
"loss": 6.8856,
"mean_token_accuracy": 0.1431492656469345,
"num_tokens": 4857888.0,
"step": 1920
},
{
"entropy": 6.893492078781128,
"epoch": 0.22215810732833238,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998707888760816,
"loss": 6.7919,
"mean_token_accuracy": 0.14359032064676286,
"num_tokens": 4870479.0,
"step": 1925
},
{
"entropy": 6.971369791030884,
"epoch": 0.22273514137334102,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998693868394516,
"loss": 6.7938,
"mean_token_accuracy": 0.1480065330862999,
"num_tokens": 4883294.0,
"step": 1930
},
{
"entropy": 7.05402398109436,
"epoch": 0.2233121754183497,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998679772394516,
"loss": 6.9319,
"mean_token_accuracy": 0.13698863312602044,
"num_tokens": 4896331.0,
"step": 1935
},
{
"entropy": 6.9416258335113525,
"epoch": 0.22388920946335833,
"grad_norm": 0.859375,
"learning_rate": 0.0004998665600761289,
"loss": 6.8675,
"mean_token_accuracy": 0.14291590824723244,
"num_tokens": 4908857.0,
"step": 1940
},
{
"entropy": 6.996976089477539,
"epoch": 0.224466243508367,
"grad_norm": 0.94140625,
"learning_rate": 0.000499865135349531,
"loss": 6.8545,
"mean_token_accuracy": 0.14505003690719603,
"num_tokens": 4921564.0,
"step": 1945
},
{
"entropy": 6.967352771759034,
"epoch": 0.22504327755337564,
"grad_norm": 0.80859375,
"learning_rate": 0.0004998637030597061,
"loss": 6.8762,
"mean_token_accuracy": 0.14787980914115906,
"num_tokens": 4934960.0,
"step": 1950
},
{
"entropy": 6.854097366333008,
"epoch": 0.2256203115983843,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998622632067022,
"loss": 6.7272,
"mean_token_accuracy": 0.15127141177654266,
"num_tokens": 4946942.0,
"step": 1955
},
{
"entropy": 6.964528608322143,
"epoch": 0.22619734564339297,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998608157905678,
"loss": 6.8454,
"mean_token_accuracy": 0.13881651386618615,
"num_tokens": 4960751.0,
"step": 1960
},
{
"entropy": 6.9480382919311525,
"epoch": 0.2267743796884016,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998593608113515,
"loss": 6.8417,
"mean_token_accuracy": 0.14182508513331413,
"num_tokens": 4973551.0,
"step": 1965
},
{
"entropy": 6.950195169448852,
"epoch": 0.22735141373341028,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998578982691024,
"loss": 6.87,
"mean_token_accuracy": 0.14090688973665239,
"num_tokens": 4986224.0,
"step": 1970
},
{
"entropy": 6.947819328308105,
"epoch": 0.22792844777841892,
"grad_norm": 0.8359375,
"learning_rate": 0.0004998564281638694,
"loss": 6.8368,
"mean_token_accuracy": 0.14499269127845765,
"num_tokens": 4999874.0,
"step": 1975
},
{
"entropy": 6.946592903137207,
"epoch": 0.2285054818234276,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998549504957023,
"loss": 6.8382,
"mean_token_accuracy": 0.1452639691531658,
"num_tokens": 5013106.0,
"step": 1980
},
{
"entropy": 6.935692501068115,
"epoch": 0.22908251586843623,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998534652646506,
"loss": 6.8023,
"mean_token_accuracy": 0.14065282493829728,
"num_tokens": 5026145.0,
"step": 1985
},
{
"entropy": 6.941158723831177,
"epoch": 0.2296595499134449,
"grad_norm": 0.828125,
"learning_rate": 0.0004998519724707642,
"loss": 6.8182,
"mean_token_accuracy": 0.14938395619392394,
"num_tokens": 5039216.0,
"step": 1990
},
{
"entropy": 6.966537952423096,
"epoch": 0.23023658395845356,
"grad_norm": 0.80859375,
"learning_rate": 0.0004998504721140934,
"loss": 6.795,
"mean_token_accuracy": 0.14359883442521096,
"num_tokens": 5051956.0,
"step": 1995
},
{
"entropy": 6.932427787780762,
"epoch": 0.2308136180034622,
"grad_norm": 0.77734375,
"learning_rate": 0.0004998489641946887,
"loss": 6.8764,
"mean_token_accuracy": 0.14441612362861633,
"num_tokens": 5065325.0,
"step": 2000
},
{
"entropy": 6.94489541053772,
"epoch": 0.23139065204847087,
"grad_norm": 0.89453125,
"learning_rate": 0.0004998474487126009,
"loss": 6.875,
"mean_token_accuracy": 0.14884231835603715,
"num_tokens": 5077776.0,
"step": 2005
},
{
"entropy": 6.97218074798584,
"epoch": 0.2319676860934795,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998459256678806,
"loss": 6.8698,
"mean_token_accuracy": 0.14031682312488555,
"num_tokens": 5091167.0,
"step": 2010
},
{
"entropy": 7.01816759109497,
"epoch": 0.23254472013848818,
"grad_norm": 0.84765625,
"learning_rate": 0.0004998443950605796,
"loss": 6.8068,
"mean_token_accuracy": 0.14413101896643638,
"num_tokens": 5103321.0,
"step": 2015
},
{
"entropy": 6.925113582611084,
"epoch": 0.23312175418349682,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998428568907488,
"loss": 6.8077,
"mean_token_accuracy": 0.14342234432697296,
"num_tokens": 5116999.0,
"step": 2020
},
{
"entropy": 6.899875020980835,
"epoch": 0.23369878822850548,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998413111584403,
"loss": 6.803,
"mean_token_accuracy": 0.14782628118991853,
"num_tokens": 5130429.0,
"step": 2025
},
{
"entropy": 6.895891237258911,
"epoch": 0.23427582227351415,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998397578637059,
"loss": 6.7544,
"mean_token_accuracy": 0.14964505434036254,
"num_tokens": 5142728.0,
"step": 2030
},
{
"entropy": 6.994881963729858,
"epoch": 0.2348528563185228,
"grad_norm": 0.859375,
"learning_rate": 0.000499838197006598,
"loss": 6.854,
"mean_token_accuracy": 0.14518777132034302,
"num_tokens": 5155105.0,
"step": 2035
},
{
"entropy": 6.9891077995300295,
"epoch": 0.23542989036353146,
"grad_norm": 0.8515625,
"learning_rate": 0.000499836628587169,
"loss": 6.8814,
"mean_token_accuracy": 0.13513262197375298,
"num_tokens": 5169206.0,
"step": 2040
},
{
"entropy": 6.860302734375,
"epoch": 0.2360069244085401,
"grad_norm": 0.76953125,
"learning_rate": 0.0004998350526054716,
"loss": 6.7535,
"mean_token_accuracy": 0.1476128190755844,
"num_tokens": 5182007.0,
"step": 2045
},
{
"entropy": 6.977161026000976,
"epoch": 0.23658395845354876,
"grad_norm": 0.85546875,
"learning_rate": 0.0004998334690615591,
"loss": 6.7814,
"mean_token_accuracy": 0.14299210011959076,
"num_tokens": 5195137.0,
"step": 2050
},
{
"entropy": 6.875006675720215,
"epoch": 0.2371609924985574,
"grad_norm": 0.84375,
"learning_rate": 0.0004998318779554844,
"loss": 6.7999,
"mean_token_accuracy": 0.14403216391801835,
"num_tokens": 5207713.0,
"step": 2055
},
{
"entropy": 6.91954345703125,
"epoch": 0.23773802654356607,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998302792873012,
"loss": 6.7539,
"mean_token_accuracy": 0.1439061000943184,
"num_tokens": 5220125.0,
"step": 2060
},
{
"entropy": 6.926095628738404,
"epoch": 0.23831506058857474,
"grad_norm": 0.82421875,
"learning_rate": 0.0004998286730570631,
"loss": 6.8071,
"mean_token_accuracy": 0.1458573669195175,
"num_tokens": 5232142.0,
"step": 2065
},
{
"entropy": 6.897360849380493,
"epoch": 0.23889209463358338,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998270592648245,
"loss": 6.7912,
"mean_token_accuracy": 0.15026629641652106,
"num_tokens": 5245517.0,
"step": 2070
},
{
"entropy": 6.937630748748779,
"epoch": 0.23946912867859205,
"grad_norm": 0.859375,
"learning_rate": 0.0004998254379106394,
"loss": 6.7863,
"mean_token_accuracy": 0.14699850231409073,
"num_tokens": 5257862.0,
"step": 2075
},
{
"entropy": 6.865565061569214,
"epoch": 0.2400461627236007,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998238089945622,
"loss": 6.7348,
"mean_token_accuracy": 0.15117276906967164,
"num_tokens": 5270141.0,
"step": 2080
},
{
"entropy": 6.878887939453125,
"epoch": 0.24062319676860935,
"grad_norm": 0.81640625,
"learning_rate": 0.0004998221725166479,
"loss": 6.7593,
"mean_token_accuracy": 0.15452364832162857,
"num_tokens": 5283143.0,
"step": 2085
},
{
"entropy": 6.871646356582642,
"epoch": 0.241200230813618,
"grad_norm": 0.82421875,
"learning_rate": 0.0004998205284769516,
"loss": 6.8235,
"mean_token_accuracy": 0.13959436044096946,
"num_tokens": 5295340.0,
"step": 2090
},
{
"entropy": 6.964116764068604,
"epoch": 0.24177726485862666,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998188768755285,
"loss": 6.7683,
"mean_token_accuracy": 0.14472750276327134,
"num_tokens": 5306821.0,
"step": 2095
},
{
"entropy": 6.834314155578613,
"epoch": 0.2423542989036353,
"grad_norm": 0.82421875,
"learning_rate": 0.0004998172177124341,
"loss": 6.7865,
"mean_token_accuracy": 0.14368025809526444,
"num_tokens": 5319428.0,
"step": 2100
},
{
"entropy": 7.012720727920533,
"epoch": 0.24293133294864397,
"grad_norm": 0.765625,
"learning_rate": 0.0004998155509877242,
"loss": 6.8145,
"mean_token_accuracy": 0.13865589946508408,
"num_tokens": 5331777.0,
"step": 2105
},
{
"entropy": 6.713190269470215,
"epoch": 0.24350836699365264,
"grad_norm": 0.92578125,
"learning_rate": 0.000499813876701455,
"loss": 6.7742,
"mean_token_accuracy": 0.14583554863929749,
"num_tokens": 5344533.0,
"step": 2110
},
{
"entropy": 6.8935747146606445,
"epoch": 0.24408540103866128,
"grad_norm": 1.046875,
"learning_rate": 0.0004998121948536828,
"loss": 6.7137,
"mean_token_accuracy": 0.15403580144047738,
"num_tokens": 5357123.0,
"step": 2115
},
{
"entropy": 6.924336910247803,
"epoch": 0.24466243508366994,
"grad_norm": 0.890625,
"learning_rate": 0.0004998105054444639,
"loss": 6.8329,
"mean_token_accuracy": 0.13986173048615455,
"num_tokens": 5370123.0,
"step": 2120
},
{
"entropy": 6.897236251831055,
"epoch": 0.24523946912867858,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998088084738555,
"loss": 6.7763,
"mean_token_accuracy": 0.15366744846105576,
"num_tokens": 5382870.0,
"step": 2125
},
{
"entropy": 6.823340272903442,
"epoch": 0.24581650317368725,
"grad_norm": 0.875,
"learning_rate": 0.0004998071039419144,
"loss": 6.7582,
"mean_token_accuracy": 0.14249610602855683,
"num_tokens": 5395052.0,
"step": 2130
},
{
"entropy": 6.9604456424713135,
"epoch": 0.2463935372186959,
"grad_norm": 0.8359375,
"learning_rate": 0.000499805391848698,
"loss": 6.7837,
"mean_token_accuracy": 0.1490890696644783,
"num_tokens": 5406720.0,
"step": 2135
},
{
"entropy": 6.8235283374786375,
"epoch": 0.24697057126370456,
"grad_norm": 0.828125,
"learning_rate": 0.000499803672194264,
"loss": 6.7838,
"mean_token_accuracy": 0.14721129089593887,
"num_tokens": 5419465.0,
"step": 2140
},
{
"entropy": 6.8882369041442875,
"epoch": 0.24754760530871323,
"grad_norm": 0.7890625,
"learning_rate": 0.0004998019449786701,
"loss": 6.7894,
"mean_token_accuracy": 0.14069402888417243,
"num_tokens": 5432434.0,
"step": 2145
},
{
"entropy": 6.860111331939697,
"epoch": 0.24812463935372187,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998002102019744,
"loss": 6.7522,
"mean_token_accuracy": 0.15137931406497956,
"num_tokens": 5445249.0,
"step": 2150
},
{
"entropy": 6.983528804779053,
"epoch": 0.24870167339873053,
"grad_norm": 0.76171875,
"learning_rate": 0.0004997984678642354,
"loss": 6.7916,
"mean_token_accuracy": 0.1442440040409565,
"num_tokens": 5458996.0,
"step": 2155
},
{
"entropy": 6.853938817977905,
"epoch": 0.24927870744373917,
"grad_norm": 1.3125,
"learning_rate": 0.0004997967179655115,
"loss": 6.7744,
"mean_token_accuracy": 0.14349499493837356,
"num_tokens": 5472174.0,
"step": 2160
},
{
"entropy": 6.808877658843994,
"epoch": 0.24985574148874784,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997949605058617,
"loss": 6.7069,
"mean_token_accuracy": 0.15367118269205093,
"num_tokens": 5484464.0,
"step": 2165
},
{
"entropy": 6.934421110153198,
"epoch": 0.2504327755337565,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997931954853451,
"loss": 6.8061,
"mean_token_accuracy": 0.14632220715284347,
"num_tokens": 5497278.0,
"step": 2170
},
{
"entropy": 6.86837854385376,
"epoch": 0.2510098095787651,
"grad_norm": 0.890625,
"learning_rate": 0.000499791422904021,
"loss": 6.8287,
"mean_token_accuracy": 0.1460747793316841,
"num_tokens": 5510847.0,
"step": 2175
},
{
"entropy": 6.842457818984985,
"epoch": 0.2515868436237738,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997896427619491,
"loss": 6.7482,
"mean_token_accuracy": 0.14972207546234131,
"num_tokens": 5522543.0,
"step": 2180
},
{
"entropy": 6.892645740509034,
"epoch": 0.25216387766878245,
"grad_norm": 0.81640625,
"learning_rate": 0.0004997878550591892,
"loss": 6.7552,
"mean_token_accuracy": 0.1436435416340828,
"num_tokens": 5534856.0,
"step": 2185
},
{
"entropy": 6.831497383117676,
"epoch": 0.2527409117137911,
"grad_norm": 0.8359375,
"learning_rate": 0.0004997860597958013,
"loss": 6.6717,
"mean_token_accuracy": 0.1602167546749115,
"num_tokens": 5546881.0,
"step": 2190
},
{
"entropy": 6.895463514328003,
"epoch": 0.2533179457587998,
"grad_norm": 0.84765625,
"learning_rate": 0.0004997842569718461,
"loss": 6.7887,
"mean_token_accuracy": 0.14391066506505013,
"num_tokens": 5559711.0,
"step": 2195
},
{
"entropy": 6.865831708908081,
"epoch": 0.25389497980380843,
"grad_norm": 0.8671875,
"learning_rate": 0.000499782446587384,
"loss": 6.7843,
"mean_token_accuracy": 0.1506132774055004,
"num_tokens": 5573553.0,
"step": 2200
},
{
"entropy": 6.900459480285645,
"epoch": 0.25447201384881707,
"grad_norm": 0.88671875,
"learning_rate": 0.000499780628642476,
"loss": 6.7603,
"mean_token_accuracy": 0.13870777040719987,
"num_tokens": 5586046.0,
"step": 2205
},
{
"entropy": 6.854037666320801,
"epoch": 0.2550490478938257,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997788031371834,
"loss": 6.7603,
"mean_token_accuracy": 0.14891895428299903,
"num_tokens": 5597876.0,
"step": 2210
},
{
"entropy": 6.933327293395996,
"epoch": 0.2556260819388344,
"grad_norm": 0.90625,
"learning_rate": 0.0004997769700715672,
"loss": 6.6716,
"mean_token_accuracy": 0.1543987885117531,
"num_tokens": 5611264.0,
"step": 2215
},
{
"entropy": 6.804627752304077,
"epoch": 0.25620311598384304,
"grad_norm": 0.828125,
"learning_rate": 0.0004997751294456892,
"loss": 6.7679,
"mean_token_accuracy": 0.1464955747127533,
"num_tokens": 5625093.0,
"step": 2220
},
{
"entropy": 6.852876806259156,
"epoch": 0.2567801500288517,
"grad_norm": 0.875,
"learning_rate": 0.0004997732812596114,
"loss": 6.7744,
"mean_token_accuracy": 0.14379966408014297,
"num_tokens": 5637206.0,
"step": 2225
},
{
"entropy": 6.857150936126709,
"epoch": 0.2573571840738604,
"grad_norm": 0.859375,
"learning_rate": 0.0004997714255133961,
"loss": 6.6563,
"mean_token_accuracy": 0.15202945172786714,
"num_tokens": 5649136.0,
"step": 2230
},
{
"entropy": 6.837795639038086,
"epoch": 0.257934218118869,
"grad_norm": 1.015625,
"learning_rate": 0.0004997695622071054,
"loss": 6.6956,
"mean_token_accuracy": 0.15131831020116807,
"num_tokens": 5660714.0,
"step": 2235
},
{
"entropy": 6.87720136642456,
"epoch": 0.25851125216387766,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997676913408021,
"loss": 6.7654,
"mean_token_accuracy": 0.14454615712165833,
"num_tokens": 5673295.0,
"step": 2240
},
{
"entropy": 6.8868872165679935,
"epoch": 0.2590882862088863,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997658129145493,
"loss": 6.777,
"mean_token_accuracy": 0.14726119190454484,
"num_tokens": 5686822.0,
"step": 2245
},
{
"entropy": 6.793786144256591,
"epoch": 0.259665320253895,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997639269284099,
"loss": 6.8295,
"mean_token_accuracy": 0.1436148166656494,
"num_tokens": 5700621.0,
"step": 2250
},
{
"entropy": 6.9439198017120365,
"epoch": 0.26024235429890363,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997620333824476,
"loss": 6.7379,
"mean_token_accuracy": 0.15637929141521453,
"num_tokens": 5713565.0,
"step": 2255
},
{
"entropy": 6.800307655334473,
"epoch": 0.2608193883439123,
"grad_norm": 0.83984375,
"learning_rate": 0.0004997601322767258,
"loss": 6.783,
"mean_token_accuracy": 0.1416866213083267,
"num_tokens": 5726487.0,
"step": 2260
},
{
"entropy": 6.966365957260132,
"epoch": 0.26139642238892097,
"grad_norm": 0.8515625,
"learning_rate": 0.0004997582236113087,
"loss": 6.8581,
"mean_token_accuracy": 0.1411590591073036,
"num_tokens": 5739934.0,
"step": 2265
},
{
"entropy": 6.865359163284301,
"epoch": 0.2619734564339296,
"grad_norm": 0.8046875,
"learning_rate": 0.0004997563073862603,
"loss": 6.8048,
"mean_token_accuracy": 0.13894407004117965,
"num_tokens": 5752187.0,
"step": 2270
},
{
"entropy": 6.836117839813232,
"epoch": 0.26255049047893825,
"grad_norm": 0.8515625,
"learning_rate": 0.0004997543836016453,
"loss": 6.7345,
"mean_token_accuracy": 0.14709190279245377,
"num_tokens": 5764850.0,
"step": 2275
},
{
"entropy": 6.9021507740020756,
"epoch": 0.2631275245239469,
"grad_norm": 0.71484375,
"learning_rate": 0.0004997524522575281,
"loss": 6.8019,
"mean_token_accuracy": 0.14260414987802505,
"num_tokens": 5778577.0,
"step": 2280
},
{
"entropy": 6.944571781158447,
"epoch": 0.2637045585689556,
"grad_norm": 0.8203125,
"learning_rate": 0.0004997505133539738,
"loss": 6.7487,
"mean_token_accuracy": 0.1511955589056015,
"num_tokens": 5791699.0,
"step": 2285
},
{
"entropy": 6.796835660934448,
"epoch": 0.2642815926139642,
"grad_norm": 0.828125,
"learning_rate": 0.0004997485668910476,
"loss": 6.7555,
"mean_token_accuracy": 0.14254892766475677,
"num_tokens": 5804553.0,
"step": 2290
},
{
"entropy": 6.993517017364502,
"epoch": 0.26485862665897286,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997466128688151,
"loss": 6.7579,
"mean_token_accuracy": 0.14862642660737038,
"num_tokens": 5816758.0,
"step": 2295
},
{
"entropy": 6.853436231613159,
"epoch": 0.26543566070398156,
"grad_norm": 0.828125,
"learning_rate": 0.000499744651287342,
"loss": 6.7812,
"mean_token_accuracy": 0.13678565695881845,
"num_tokens": 5831048.0,
"step": 2300
},
{
"entropy": 6.919308710098266,
"epoch": 0.2660126947489902,
"grad_norm": 0.8828125,
"learning_rate": 0.000499742682146694,
"loss": 6.7909,
"mean_token_accuracy": 0.13784368932247162,
"num_tokens": 5843159.0,
"step": 2305
},
{
"entropy": 6.807005167007446,
"epoch": 0.26658972879399884,
"grad_norm": 0.80078125,
"learning_rate": 0.0004997407054469377,
"loss": 6.6791,
"mean_token_accuracy": 0.15264758318662644,
"num_tokens": 5856078.0,
"step": 2310
},
{
"entropy": 6.854742860794067,
"epoch": 0.2671667628390075,
"grad_norm": 1.3046875,
"learning_rate": 0.0004997387211881392,
"loss": 6.706,
"mean_token_accuracy": 0.14707566574215888,
"num_tokens": 5867835.0,
"step": 2315
},
{
"entropy": 6.8351672172546385,
"epoch": 0.26774379688401617,
"grad_norm": 0.83984375,
"learning_rate": 0.0004997367293703656,
"loss": 6.7444,
"mean_token_accuracy": 0.13954102098941804,
"num_tokens": 5881179.0,
"step": 2320
},
{
"entropy": 6.878024482727051,
"epoch": 0.2683208309290248,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997347299936837,
"loss": 6.7195,
"mean_token_accuracy": 0.14241515398025512,
"num_tokens": 5892774.0,
"step": 2325
},
{
"entropy": 6.870539665222168,
"epoch": 0.26889786497403345,
"grad_norm": 0.83203125,
"learning_rate": 0.0004997327230581608,
"loss": 6.756,
"mean_token_accuracy": 0.1463689960539341,
"num_tokens": 5905857.0,
"step": 2330
},
{
"entropy": 6.805995798110962,
"epoch": 0.26947489901904215,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997307085638644,
"loss": 6.7166,
"mean_token_accuracy": 0.14379769265651704,
"num_tokens": 5918128.0,
"step": 2335
},
{
"entropy": 6.918767642974854,
"epoch": 0.2700519330640508,
"grad_norm": 0.78515625,
"learning_rate": 0.0004997286865108621,
"loss": 6.8232,
"mean_token_accuracy": 0.14087387174367905,
"num_tokens": 5930606.0,
"step": 2340
},
{
"entropy": 6.841729688644409,
"epoch": 0.2706289671090594,
"grad_norm": 0.88671875,
"learning_rate": 0.0004997266568992222,
"loss": 6.7696,
"mean_token_accuracy": 0.13774231299757958,
"num_tokens": 5944139.0,
"step": 2345
},
{
"entropy": 6.887184953689575,
"epoch": 0.27120600115406807,
"grad_norm": 0.82421875,
"learning_rate": 0.0004997246197290128,
"loss": 6.7646,
"mean_token_accuracy": 0.14507462382316588,
"num_tokens": 5956779.0,
"step": 2350
},
{
"entropy": 6.837090969085693,
"epoch": 0.27178303519907676,
"grad_norm": 0.921875,
"learning_rate": 0.0004997225750003024,
"loss": 6.7083,
"mean_token_accuracy": 0.15230057314038276,
"num_tokens": 5971510.0,
"step": 2355
},
{
"entropy": 6.695552968978882,
"epoch": 0.2723600692440854,
"grad_norm": 0.84375,
"learning_rate": 0.0004997205227131599,
"loss": 6.7371,
"mean_token_accuracy": 0.1434318020939827,
"num_tokens": 5983877.0,
"step": 2360
},
{
"entropy": 6.974739694595337,
"epoch": 0.27293710328909404,
"grad_norm": 0.828125,
"learning_rate": 0.0004997184628676542,
"loss": 6.7617,
"mean_token_accuracy": 0.14533916413784026,
"num_tokens": 5996739.0,
"step": 2365
},
{
"entropy": 6.768356561660767,
"epoch": 0.27351413733410274,
"grad_norm": 0.859375,
"learning_rate": 0.0004997163954638546,
"loss": 6.7255,
"mean_token_accuracy": 0.15101923793554306,
"num_tokens": 6009289.0,
"step": 2370
},
{
"entropy": 6.87177038192749,
"epoch": 0.2740911713791114,
"grad_norm": 0.96875,
"learning_rate": 0.0004997143205018306,
"loss": 6.794,
"mean_token_accuracy": 0.14188004285097122,
"num_tokens": 6021375.0,
"step": 2375
},
{
"entropy": 6.898721408843994,
"epoch": 0.27466820542412,
"grad_norm": 0.84375,
"learning_rate": 0.0004997122379816523,
"loss": 6.7271,
"mean_token_accuracy": 0.14577853456139564,
"num_tokens": 6034182.0,
"step": 2380
},
{
"entropy": 6.774568223953247,
"epoch": 0.27524523946912866,
"grad_norm": 0.86328125,
"learning_rate": 0.0004997101479033893,
"loss": 6.7104,
"mean_token_accuracy": 0.1492339551448822,
"num_tokens": 6046486.0,
"step": 2385
},
{
"entropy": 6.845656013488769,
"epoch": 0.27582227351413735,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997080502671122,
"loss": 6.7644,
"mean_token_accuracy": 0.14610921069979668,
"num_tokens": 6060867.0,
"step": 2390
},
{
"entropy": 6.7973473072052,
"epoch": 0.276399307559146,
"grad_norm": 0.83203125,
"learning_rate": 0.0004997059450728913,
"loss": 6.6264,
"mean_token_accuracy": 0.15345287024974824,
"num_tokens": 6073000.0,
"step": 2395
},
{
"entropy": 6.76253228187561,
"epoch": 0.27697634160415463,
"grad_norm": 0.796875,
"learning_rate": 0.0004997038323207977,
"loss": 6.6906,
"mean_token_accuracy": 0.14774474650621414,
"num_tokens": 6085884.0,
"step": 2400
},
{
"entropy": 6.787846565246582,
"epoch": 0.2775533756491633,
"grad_norm": 0.87109375,
"learning_rate": 0.0004997017120109022,
"loss": 6.6146,
"mean_token_accuracy": 0.15057797580957413,
"num_tokens": 6098230.0,
"step": 2405
},
{
"entropy": 6.782491827011109,
"epoch": 0.27813040969417197,
"grad_norm": 0.84765625,
"learning_rate": 0.0004996995841432762,
"loss": 6.6701,
"mean_token_accuracy": 0.15104399472475052,
"num_tokens": 6111568.0,
"step": 2410
},
{
"entropy": 6.779537677764893,
"epoch": 0.2787074437391806,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996974487179915,
"loss": 6.646,
"mean_token_accuracy": 0.14741345196962358,
"num_tokens": 6123992.0,
"step": 2415
},
{
"entropy": 6.796364402770996,
"epoch": 0.27928447778418924,
"grad_norm": 0.83203125,
"learning_rate": 0.0004996953057351195,
"loss": 6.686,
"mean_token_accuracy": 0.1465798109769821,
"num_tokens": 6135452.0,
"step": 2420
},
{
"entropy": 6.774402904510498,
"epoch": 0.27986151182919794,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996931551947327,
"loss": 6.5866,
"mean_token_accuracy": 0.15778433084487914,
"num_tokens": 6150144.0,
"step": 2425
},
{
"entropy": 6.837385368347168,
"epoch": 0.2804385458742066,
"grad_norm": 0.8125,
"learning_rate": 0.0004996909970969031,
"loss": 6.7329,
"mean_token_accuracy": 0.14868736639618874,
"num_tokens": 6163397.0,
"step": 2430
},
{
"entropy": 6.832604503631591,
"epoch": 0.2810155799192152,
"grad_norm": 0.8671875,
"learning_rate": 0.0004996888314417034,
"loss": 6.7233,
"mean_token_accuracy": 0.14309920221567154,
"num_tokens": 6175579.0,
"step": 2435
},
{
"entropy": 6.790614795684815,
"epoch": 0.2815926139642239,
"grad_norm": 0.81640625,
"learning_rate": 0.0004996866582292067,
"loss": 6.7063,
"mean_token_accuracy": 0.15070078819990157,
"num_tokens": 6189229.0,
"step": 2440
},
{
"entropy": 6.777169132232666,
"epoch": 0.28216964800923255,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996844774594856,
"loss": 6.6934,
"mean_token_accuracy": 0.1502875879406929,
"num_tokens": 6203817.0,
"step": 2445
},
{
"entropy": 6.82915768623352,
"epoch": 0.2827466820542412,
"grad_norm": 0.859375,
"learning_rate": 0.0004996822891326138,
"loss": 6.7206,
"mean_token_accuracy": 0.14171916842460633,
"num_tokens": 6216098.0,
"step": 2450
},
{
"entropy": 6.812518548965454,
"epoch": 0.28332371609924983,
"grad_norm": 0.80859375,
"learning_rate": 0.0004996800932486648,
"loss": 6.7166,
"mean_token_accuracy": 0.14603719413280486,
"num_tokens": 6228987.0,
"step": 2455
},
{
"entropy": 6.742992639541626,
"epoch": 0.28390075014425853,
"grad_norm": 0.8671875,
"learning_rate": 0.0004996778898077126,
"loss": 6.6685,
"mean_token_accuracy": 0.15222399532794953,
"num_tokens": 6241676.0,
"step": 2460
},
{
"entropy": 6.889522314071655,
"epoch": 0.28447778418926717,
"grad_norm": 0.8046875,
"learning_rate": 0.0004996756788098309,
"loss": 6.702,
"mean_token_accuracy": 0.14694569408893585,
"num_tokens": 6254744.0,
"step": 2465
},
{
"entropy": 6.758044672012329,
"epoch": 0.2850548182342758,
"grad_norm": 0.76953125,
"learning_rate": 0.0004996734602550945,
"loss": 6.6596,
"mean_token_accuracy": 0.1465069644153118,
"num_tokens": 6268242.0,
"step": 2470
},
{
"entropy": 6.747747182846069,
"epoch": 0.2856318522792845,
"grad_norm": 0.8359375,
"learning_rate": 0.0004996712341435779,
"loss": 6.5867,
"mean_token_accuracy": 0.1545226328074932,
"num_tokens": 6279904.0,
"step": 2475
},
{
"entropy": 6.745885562896729,
"epoch": 0.28620888632429314,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996690004753559,
"loss": 6.7018,
"mean_token_accuracy": 0.14820884317159652,
"num_tokens": 6292885.0,
"step": 2480
},
{
"entropy": 6.753550338745117,
"epoch": 0.2867859203693018,
"grad_norm": 0.765625,
"learning_rate": 0.0004996667592505037,
"loss": 6.6887,
"mean_token_accuracy": 0.15229137539863585,
"num_tokens": 6306278.0,
"step": 2485
},
{
"entropy": 6.884258556365967,
"epoch": 0.2873629544143104,
"grad_norm": 0.859375,
"learning_rate": 0.0004996645104690967,
"loss": 6.7179,
"mean_token_accuracy": 0.13899706602096557,
"num_tokens": 6320360.0,
"step": 2490
},
{
"entropy": 6.700476408004761,
"epoch": 0.2879399884593191,
"grad_norm": 0.82421875,
"learning_rate": 0.0004996622541312103,
"loss": 6.7035,
"mean_token_accuracy": 0.144860952347517,
"num_tokens": 6332778.0,
"step": 2495
},
{
"entropy": 6.817806053161621,
"epoch": 0.28851702250432776,
"grad_norm": 0.81640625,
"learning_rate": 0.0004996599902369207,
"loss": 6.6808,
"mean_token_accuracy": 0.14956472143530847,
"num_tokens": 6345028.0,
"step": 2500
},
{
"entropy": 6.786530876159668,
"epoch": 0.2890940565493364,
"grad_norm": 0.875,
"learning_rate": 0.0004996577187863039,
"loss": 6.6425,
"mean_token_accuracy": 0.15257902592420577,
"num_tokens": 6356931.0,
"step": 2505
},
{
"entropy": 6.698649883270264,
"epoch": 0.2896710905943451,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996554397794364,
"loss": 6.5883,
"mean_token_accuracy": 0.15760106444358826,
"num_tokens": 6370247.0,
"step": 2510
},
{
"entropy": 6.7704665184021,
"epoch": 0.29024812463935373,
"grad_norm": 0.91015625,
"learning_rate": 0.0004996531532163947,
"loss": 6.6603,
"mean_token_accuracy": 0.14782440811395645,
"num_tokens": 6382352.0,
"step": 2515
},
{
"entropy": 6.754695749282837,
"epoch": 0.2908251586843624,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996508590972558,
"loss": 6.6397,
"mean_token_accuracy": 0.14761101007461547,
"num_tokens": 6395949.0,
"step": 2520
},
{
"entropy": 6.754686546325684,
"epoch": 0.291402192729371,
"grad_norm": 0.81640625,
"learning_rate": 0.0004996485574220969,
"loss": 6.6647,
"mean_token_accuracy": 0.15044545978307725,
"num_tokens": 6408469.0,
"step": 2525
},
{
"entropy": 6.768384695053101,
"epoch": 0.2919792267743797,
"grad_norm": 0.828125,
"learning_rate": 0.0004996462481909953,
"loss": 6.6579,
"mean_token_accuracy": 0.1553287237882614,
"num_tokens": 6419953.0,
"step": 2530
},
{
"entropy": 6.790286922454834,
"epoch": 0.29255626081938835,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996439314040287,
"loss": 6.7094,
"mean_token_accuracy": 0.143075567483902,
"num_tokens": 6432567.0,
"step": 2535
},
{
"entropy": 6.826392459869385,
"epoch": 0.293133294864397,
"grad_norm": 1.015625,
"learning_rate": 0.000499641607061275,
"loss": 6.6549,
"mean_token_accuracy": 0.1453825443983078,
"num_tokens": 6445951.0,
"step": 2540
},
{
"entropy": 6.742476511001587,
"epoch": 0.2937103289094057,
"grad_norm": 0.890625,
"learning_rate": 0.0004996392751628126,
"loss": 6.7032,
"mean_token_accuracy": 0.15503485202789308,
"num_tokens": 6458799.0,
"step": 2545
},
{
"entropy": 6.815782499313355,
"epoch": 0.2942873629544143,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996369357087196,
"loss": 6.7065,
"mean_token_accuracy": 0.1430730052292347,
"num_tokens": 6471423.0,
"step": 2550
},
{
"entropy": 6.799359512329102,
"epoch": 0.29486439699942296,
"grad_norm": 0.921875,
"learning_rate": 0.000499634588699075,
"loss": 6.7392,
"mean_token_accuracy": 0.14322351813316345,
"num_tokens": 6484248.0,
"step": 2555
},
{
"entropy": 6.833653163909912,
"epoch": 0.2954414310444316,
"grad_norm": 0.859375,
"learning_rate": 0.0004996322341339575,
"loss": 6.6637,
"mean_token_accuracy": 0.14406396448612213,
"num_tokens": 6497048.0,
"step": 2560
},
{
"entropy": 6.775566720962525,
"epoch": 0.2960184650894403,
"grad_norm": 0.84765625,
"learning_rate": 0.0004996298720134465,
"loss": 6.6848,
"mean_token_accuracy": 0.14442287534475326,
"num_tokens": 6508840.0,
"step": 2565
},
{
"entropy": 6.7831347465515135,
"epoch": 0.29659549913444894,
"grad_norm": 0.8515625,
"learning_rate": 0.0004996275023376213,
"loss": 6.6583,
"mean_token_accuracy": 0.1537433847784996,
"num_tokens": 6520976.0,
"step": 2570
},
{
"entropy": 6.741522121429443,
"epoch": 0.2971725331794576,
"grad_norm": 0.82421875,
"learning_rate": 0.0004996251251065615,
"loss": 6.673,
"mean_token_accuracy": 0.15090716034173965,
"num_tokens": 6533009.0,
"step": 2575
},
{
"entropy": 6.822439098358155,
"epoch": 0.2977495672244662,
"grad_norm": 0.890625,
"learning_rate": 0.0004996227403203473,
"loss": 6.6662,
"mean_token_accuracy": 0.14721700698137283,
"num_tokens": 6545196.0,
"step": 2580
},
{
"entropy": 6.812913274765014,
"epoch": 0.2983266012694749,
"grad_norm": 0.859375,
"learning_rate": 0.0004996203479790589,
"loss": 6.7017,
"mean_token_accuracy": 0.1475054517388344,
"num_tokens": 6557479.0,
"step": 2585
},
{
"entropy": 6.819405889511108,
"epoch": 0.29890363531448355,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996179480827764,
"loss": 6.7114,
"mean_token_accuracy": 0.14923257529735565,
"num_tokens": 6569832.0,
"step": 2590
},
{
"entropy": 6.738608884811401,
"epoch": 0.2994806693594922,
"grad_norm": 0.80859375,
"learning_rate": 0.000499615540631581,
"loss": 6.6737,
"mean_token_accuracy": 0.14806562215089797,
"num_tokens": 6581651.0,
"step": 2595
},
{
"entropy": 6.774818849563599,
"epoch": 0.3000577034045009,
"grad_norm": 0.765625,
"learning_rate": 0.0004996131256255534,
"loss": 6.7423,
"mean_token_accuracy": 0.14710890799760817,
"num_tokens": 6596331.0,
"step": 2600
},
{
"entropy": 6.817371892929077,
"epoch": 0.3006347374495095,
"grad_norm": 0.8046875,
"learning_rate": 0.0004996107030647749,
"loss": 6.6685,
"mean_token_accuracy": 0.14770287573337554,
"num_tokens": 6609712.0,
"step": 2605
},
{
"entropy": 6.759057664871216,
"epoch": 0.30121177149451817,
"grad_norm": 0.85546875,
"learning_rate": 0.0004996082729493269,
"loss": 6.6962,
"mean_token_accuracy": 0.1502513587474823,
"num_tokens": 6622780.0,
"step": 2610
},
{
"entropy": 6.784829711914062,
"epoch": 0.3017888055395268,
"grad_norm": 0.9609375,
"learning_rate": 0.0004996058352792913,
"loss": 6.7561,
"mean_token_accuracy": 0.1468655101954937,
"num_tokens": 6636288.0,
"step": 2615
},
{
"entropy": 6.858647060394287,
"epoch": 0.3023658395845355,
"grad_norm": 0.91796875,
"learning_rate": 0.00049960339005475,
"loss": 6.7313,
"mean_token_accuracy": 0.1382056713104248,
"num_tokens": 6649883.0,
"step": 2620
},
{
"entropy": 6.787996578216553,
"epoch": 0.30294287362954414,
"grad_norm": 0.78125,
"learning_rate": 0.0004996009372757852,
"loss": 6.6803,
"mean_token_accuracy": 0.14866930097341538,
"num_tokens": 6662241.0,
"step": 2625
},
{
"entropy": 6.705225419998169,
"epoch": 0.3035199076745528,
"grad_norm": 0.78515625,
"learning_rate": 0.0004995984769424795,
"loss": 6.6389,
"mean_token_accuracy": 0.14856984168291093,
"num_tokens": 6674790.0,
"step": 2630
},
{
"entropy": 6.879620838165283,
"epoch": 0.3040969417195615,
"grad_norm": 0.83984375,
"learning_rate": 0.0004995960090549155,
"loss": 6.6451,
"mean_token_accuracy": 0.14790016412734985,
"num_tokens": 6687713.0,
"step": 2635
},
{
"entropy": 6.629987955093384,
"epoch": 0.3046739757645701,
"grad_norm": 0.8984375,
"learning_rate": 0.0004995935336131764,
"loss": 6.6574,
"mean_token_accuracy": 0.1474847599864006,
"num_tokens": 6701123.0,
"step": 2640
},
{
"entropy": 6.796099090576172,
"epoch": 0.30525100980957875,
"grad_norm": 0.90625,
"learning_rate": 0.0004995910506173452,
"loss": 6.6029,
"mean_token_accuracy": 0.15525631308555604,
"num_tokens": 6712674.0,
"step": 2645
},
{
"entropy": 6.721958589553833,
"epoch": 0.3058280438545874,
"grad_norm": 0.875,
"learning_rate": 0.0004995885600675059,
"loss": 6.601,
"mean_token_accuracy": 0.15212517231702805,
"num_tokens": 6724840.0,
"step": 2650
},
{
"entropy": 6.760172510147095,
"epoch": 0.3064050778995961,
"grad_norm": 0.85546875,
"learning_rate": 0.0004995860619637415,
"loss": 6.7425,
"mean_token_accuracy": 0.13689624965190889,
"num_tokens": 6738746.0,
"step": 2655
},
{
"entropy": 6.875043773651123,
"epoch": 0.30698211194460473,
"grad_norm": 0.87109375,
"learning_rate": 0.0004995835563061367,
"loss": 6.6436,
"mean_token_accuracy": 0.1482662469148636,
"num_tokens": 6751620.0,
"step": 2660
},
{
"entropy": 6.727645397186279,
"epoch": 0.30755914598961337,
"grad_norm": 0.84375,
"learning_rate": 0.0004995810430947756,
"loss": 6.6337,
"mean_token_accuracy": 0.1513223595917225,
"num_tokens": 6763957.0,
"step": 2665
},
{
"entropy": 6.559813404083252,
"epoch": 0.30813618003462206,
"grad_norm": 1.3359375,
"learning_rate": 0.0004995785223297426,
"loss": 6.5468,
"mean_token_accuracy": 0.1547765851020813,
"num_tokens": 6776211.0,
"step": 2670
},
{
"entropy": 6.862458086013794,
"epoch": 0.3087132140796307,
"grad_norm": 0.80859375,
"learning_rate": 0.0004995759940111225,
"loss": 6.786,
"mean_token_accuracy": 0.14168819040060043,
"num_tokens": 6789745.0,
"step": 2675
},
{
"entropy": 6.805125570297241,
"epoch": 0.30929024812463934,
"grad_norm": 0.8671875,
"learning_rate": 0.0004995734581390005,
"loss": 6.7054,
"mean_token_accuracy": 0.14849800616502762,
"num_tokens": 6801771.0,
"step": 2680
},
{
"entropy": 6.708049964904785,
"epoch": 0.309867282169648,
"grad_norm": 0.79296875,
"learning_rate": 0.0004995709147134617,
"loss": 6.6365,
"mean_token_accuracy": 0.14683517664670945,
"num_tokens": 6813993.0,
"step": 2685
},
{
"entropy": 6.818436479568481,
"epoch": 0.3104443162146567,
"grad_norm": 0.8125,
"learning_rate": 0.0004995683637345919,
"loss": 6.7144,
"mean_token_accuracy": 0.14922749251127243,
"num_tokens": 6826833.0,
"step": 2690
},
{
"entropy": 6.754158878326416,
"epoch": 0.3110213502596653,
"grad_norm": 0.84765625,
"learning_rate": 0.0004995658052024765,
"loss": 6.6478,
"mean_token_accuracy": 0.1457843691110611,
"num_tokens": 6838947.0,
"step": 2695
},
{
"entropy": 6.80776834487915,
"epoch": 0.31159838430467396,
"grad_norm": 0.828125,
"learning_rate": 0.0004995632391172019,
"loss": 6.7147,
"mean_token_accuracy": 0.14534788727760314,
"num_tokens": 6851781.0,
"step": 2700
},
{
"entropy": 6.8032268524169925,
"epoch": 0.31217541834968265,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995606654788543,
"loss": 6.601,
"mean_token_accuracy": 0.1524400979280472,
"num_tokens": 6863479.0,
"step": 2705
},
{
"entropy": 6.714714527130127,
"epoch": 0.3127524523946913,
"grad_norm": 0.921875,
"learning_rate": 0.0004995580842875202,
"loss": 6.593,
"mean_token_accuracy": 0.15386273488402366,
"num_tokens": 6877000.0,
"step": 2710
},
{
"entropy": 6.8048097610473635,
"epoch": 0.31332948643969993,
"grad_norm": 0.91796875,
"learning_rate": 0.0004995554955432866,
"loss": 6.6981,
"mean_token_accuracy": 0.14284373223781585,
"num_tokens": 6888878.0,
"step": 2715
},
{
"entropy": 6.769490671157837,
"epoch": 0.3139065204847086,
"grad_norm": 0.8203125,
"learning_rate": 0.0004995528992462404,
"loss": 6.6196,
"mean_token_accuracy": 0.14831875115633011,
"num_tokens": 6902147.0,
"step": 2720
},
{
"entropy": 6.750906562805175,
"epoch": 0.31448355452971727,
"grad_norm": 0.83984375,
"learning_rate": 0.000499550295396469,
"loss": 6.6661,
"mean_token_accuracy": 0.15283936485648156,
"num_tokens": 6914927.0,
"step": 2725
},
{
"entropy": 6.761933422088623,
"epoch": 0.3150605885747259,
"grad_norm": 0.8359375,
"learning_rate": 0.0004995476839940598,
"loss": 6.6683,
"mean_token_accuracy": 0.1505434513092041,
"num_tokens": 6927757.0,
"step": 2730
},
{
"entropy": 6.762182760238647,
"epoch": 0.31563762261973455,
"grad_norm": 0.79296875,
"learning_rate": 0.000499545065039101,
"loss": 6.6725,
"mean_token_accuracy": 0.1466390758752823,
"num_tokens": 6941039.0,
"step": 2735
},
{
"entropy": 6.781767749786377,
"epoch": 0.31621465666474324,
"grad_norm": 1.140625,
"learning_rate": 0.0004995424385316803,
"loss": 6.7293,
"mean_token_accuracy": 0.14698702692985535,
"num_tokens": 6954876.0,
"step": 2740
},
{
"entropy": 6.725474309921265,
"epoch": 0.3167916907097519,
"grad_norm": 0.80859375,
"learning_rate": 0.0004995398044718863,
"loss": 6.5811,
"mean_token_accuracy": 0.1516340285539627,
"num_tokens": 6966889.0,
"step": 2745
},
{
"entropy": 6.753961610794067,
"epoch": 0.3173687247547605,
"grad_norm": 0.875,
"learning_rate": 0.0004995371628598074,
"loss": 6.6151,
"mean_token_accuracy": 0.1514896869659424,
"num_tokens": 6979894.0,
"step": 2750
},
{
"entropy": 6.722421312332154,
"epoch": 0.31794575879976916,
"grad_norm": 0.8515625,
"learning_rate": 0.0004995345136955328,
"loss": 6.6356,
"mean_token_accuracy": 0.1484585776925087,
"num_tokens": 6992734.0,
"step": 2755
},
{
"entropy": 6.755990600585937,
"epoch": 0.31852279284477786,
"grad_norm": 1.15625,
"learning_rate": 0.0004995318569791512,
"loss": 6.663,
"mean_token_accuracy": 0.14797065705060958,
"num_tokens": 7006504.0,
"step": 2760
},
{
"entropy": 6.7291755199432375,
"epoch": 0.3190998268897865,
"grad_norm": 0.79296875,
"learning_rate": 0.0004995291927107522,
"loss": 6.5971,
"mean_token_accuracy": 0.15441082864999772,
"num_tokens": 7018978.0,
"step": 2765
},
{
"entropy": 6.710022735595703,
"epoch": 0.31967686093479514,
"grad_norm": 0.7578125,
"learning_rate": 0.0004995265208904252,
"loss": 6.6868,
"mean_token_accuracy": 0.1457875519990921,
"num_tokens": 7031294.0,
"step": 2770
},
{
"entropy": 6.80343861579895,
"epoch": 0.32025389497980383,
"grad_norm": 0.87890625,
"learning_rate": 0.0004995238415182603,
"loss": 6.6386,
"mean_token_accuracy": 0.1497604802250862,
"num_tokens": 7044139.0,
"step": 2775
},
{
"entropy": 6.643225526809692,
"epoch": 0.3208309290248125,
"grad_norm": 0.875,
"learning_rate": 0.0004995211545943476,
"loss": 6.5431,
"mean_token_accuracy": 0.15468174815177918,
"num_tokens": 7056232.0,
"step": 2780
},
{
"entropy": 6.770214605331421,
"epoch": 0.3214079630698211,
"grad_norm": 0.7890625,
"learning_rate": 0.0004995184601187772,
"loss": 6.6978,
"mean_token_accuracy": 0.1458260402083397,
"num_tokens": 7070125.0,
"step": 2785
},
{
"entropy": 6.735332918167114,
"epoch": 0.32198499711482975,
"grad_norm": 0.8125,
"learning_rate": 0.00049951575809164,
"loss": 6.5678,
"mean_token_accuracy": 0.15374045222997665,
"num_tokens": 7083535.0,
"step": 2790
},
{
"entropy": 6.679257535934449,
"epoch": 0.32256203115983845,
"grad_norm": 0.85546875,
"learning_rate": 0.0004995130485130269,
"loss": 6.537,
"mean_token_accuracy": 0.15215112119913102,
"num_tokens": 7095081.0,
"step": 2795
},
{
"entropy": 6.756899642944336,
"epoch": 0.3231390652048471,
"grad_norm": 0.8203125,
"learning_rate": 0.0004995103313830289,
"loss": 6.6676,
"mean_token_accuracy": 0.14167121201753616,
"num_tokens": 7107764.0,
"step": 2800
},
{
"entropy": 6.7260784149169925,
"epoch": 0.3237160992498557,
"grad_norm": 0.8125,
"learning_rate": 0.0004995076067017373,
"loss": 6.6009,
"mean_token_accuracy": 0.1534109964966774,
"num_tokens": 7119754.0,
"step": 2805
},
{
"entropy": 6.6373598098754885,
"epoch": 0.3242931332948644,
"grad_norm": 0.8046875,
"learning_rate": 0.0004995048744692439,
"loss": 6.6299,
"mean_token_accuracy": 0.15635569840669633,
"num_tokens": 7132087.0,
"step": 2810
},
{
"entropy": 6.691185474395752,
"epoch": 0.32487016733987306,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995021346856405,
"loss": 6.6053,
"mean_token_accuracy": 0.14959825202822685,
"num_tokens": 7143325.0,
"step": 2815
},
{
"entropy": 6.7681121826171875,
"epoch": 0.3254472013848817,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994993873510196,
"loss": 6.5619,
"mean_token_accuracy": 0.15113845020532607,
"num_tokens": 7154771.0,
"step": 2820
},
{
"entropy": 6.670676565170288,
"epoch": 0.32602423542989034,
"grad_norm": 0.828125,
"learning_rate": 0.0004994966324654731,
"loss": 6.6585,
"mean_token_accuracy": 0.14671303182840348,
"num_tokens": 7168352.0,
"step": 2825
},
{
"entropy": 6.619832849502563,
"epoch": 0.32660126947489904,
"grad_norm": 0.82421875,
"learning_rate": 0.000499493870029094,
"loss": 6.4456,
"mean_token_accuracy": 0.16549365520477294,
"num_tokens": 7180618.0,
"step": 2830
},
{
"entropy": 6.737293481826782,
"epoch": 0.3271783035199077,
"grad_norm": 0.8125,
"learning_rate": 0.0004994911000419749,
"loss": 6.6031,
"mean_token_accuracy": 0.14624983444809914,
"num_tokens": 7192634.0,
"step": 2835
},
{
"entropy": 6.660342025756836,
"epoch": 0.3277553375649163,
"grad_norm": 0.84375,
"learning_rate": 0.0004994883225042093,
"loss": 6.6186,
"mean_token_accuracy": 0.15046972185373306,
"num_tokens": 7205839.0,
"step": 2840
},
{
"entropy": 6.736796808242798,
"epoch": 0.328332371609925,
"grad_norm": 0.73828125,
"learning_rate": 0.0004994855374158905,
"loss": 6.627,
"mean_token_accuracy": 0.15292316824197769,
"num_tokens": 7218371.0,
"step": 2845
},
{
"entropy": 6.6943048477172855,
"epoch": 0.32890940565493365,
"grad_norm": 1.4609375,
"learning_rate": 0.000499482744777112,
"loss": 6.4322,
"mean_token_accuracy": 0.1693370446562767,
"num_tokens": 7231506.0,
"step": 2850
},
{
"entropy": 6.649769687652588,
"epoch": 0.3294864396999423,
"grad_norm": 0.8125,
"learning_rate": 0.000499479944587968,
"loss": 6.5888,
"mean_token_accuracy": 0.1511973798274994,
"num_tokens": 7243088.0,
"step": 2855
},
{
"entropy": 6.694813299179077,
"epoch": 0.33006347374495093,
"grad_norm": 0.828125,
"learning_rate": 0.0004994771368485526,
"loss": 6.5163,
"mean_token_accuracy": 0.1597664922475815,
"num_tokens": 7255383.0,
"step": 2860
},
{
"entropy": 6.647224283218383,
"epoch": 0.3306405077899596,
"grad_norm": 0.80078125,
"learning_rate": 0.0004994743215589602,
"loss": 6.608,
"mean_token_accuracy": 0.15329548567533494,
"num_tokens": 7268270.0,
"step": 2865
},
{
"entropy": 6.673927307128906,
"epoch": 0.33121754183496827,
"grad_norm": 0.8515625,
"learning_rate": 0.0004994714987192854,
"loss": 6.5362,
"mean_token_accuracy": 0.1510409414768219,
"num_tokens": 7281108.0,
"step": 2870
},
{
"entropy": 6.702598619461059,
"epoch": 0.3317945758799769,
"grad_norm": 0.76171875,
"learning_rate": 0.0004994686683296234,
"loss": 6.5981,
"mean_token_accuracy": 0.14916437938809396,
"num_tokens": 7293961.0,
"step": 2875
},
{
"entropy": 6.701837539672852,
"epoch": 0.3323716099249856,
"grad_norm": 0.8046875,
"learning_rate": 0.0004994658303900691,
"loss": 6.6155,
"mean_token_accuracy": 0.15184523910284042,
"num_tokens": 7306574.0,
"step": 2880
},
{
"entropy": 6.728485918045044,
"epoch": 0.33294864396999424,
"grad_norm": 0.78515625,
"learning_rate": 0.0004994629849007182,
"loss": 6.6466,
"mean_token_accuracy": 0.14797799810767173,
"num_tokens": 7319046.0,
"step": 2885
},
{
"entropy": 6.79762864112854,
"epoch": 0.3335256780150029,
"grad_norm": 0.90625,
"learning_rate": 0.0004994601318616663,
"loss": 6.64,
"mean_token_accuracy": 0.14297626465559005,
"num_tokens": 7332033.0,
"step": 2890
},
{
"entropy": 6.6803630828857425,
"epoch": 0.3341027120600115,
"grad_norm": 0.8359375,
"learning_rate": 0.0004994572712730092,
"loss": 6.6512,
"mean_token_accuracy": 0.14246861189603804,
"num_tokens": 7344658.0,
"step": 2895
},
{
"entropy": 6.816131925582885,
"epoch": 0.3346797461050202,
"grad_norm": 0.82421875,
"learning_rate": 0.0004994544031348434,
"loss": 6.5279,
"mean_token_accuracy": 0.15453597754240037,
"num_tokens": 7357252.0,
"step": 2900
},
{
"entropy": 6.603551912307739,
"epoch": 0.33525678015002885,
"grad_norm": 0.78125,
"learning_rate": 0.0004994515274472653,
"loss": 6.4869,
"mean_token_accuracy": 0.1610173389315605,
"num_tokens": 7369280.0,
"step": 2905
},
{
"entropy": 6.595618057250976,
"epoch": 0.3358338141950375,
"grad_norm": 0.90625,
"learning_rate": 0.0004994486442103715,
"loss": 6.4916,
"mean_token_accuracy": 0.1597081333398819,
"num_tokens": 7383017.0,
"step": 2910
},
{
"entropy": 6.698597812652588,
"epoch": 0.3364108482400462,
"grad_norm": 0.9296875,
"learning_rate": 0.000499445753424259,
"loss": 6.6324,
"mean_token_accuracy": 0.1377485029399395,
"num_tokens": 7395634.0,
"step": 2915
},
{
"entropy": 6.708563899993896,
"epoch": 0.33698788228505483,
"grad_norm": 0.8125,
"learning_rate": 0.0004994428550890251,
"loss": 6.601,
"mean_token_accuracy": 0.15198962539434432,
"num_tokens": 7408590.0,
"step": 2920
},
{
"entropy": 6.720327949523925,
"epoch": 0.33756491633006347,
"grad_norm": 0.84375,
"learning_rate": 0.0004994399492047672,
"loss": 6.5912,
"mean_token_accuracy": 0.1499352663755417,
"num_tokens": 7420202.0,
"step": 2925
},
{
"entropy": 6.670965766906738,
"epoch": 0.3381419503750721,
"grad_norm": 0.84765625,
"learning_rate": 0.0004994370357715832,
"loss": 6.5504,
"mean_token_accuracy": 0.16089099049568176,
"num_tokens": 7432664.0,
"step": 2930
},
{
"entropy": 6.661357879638672,
"epoch": 0.3387189844200808,
"grad_norm": 0.75390625,
"learning_rate": 0.000499434114789571,
"loss": 6.7051,
"mean_token_accuracy": 0.1464277185499668,
"num_tokens": 7446084.0,
"step": 2935
},
{
"entropy": 6.714499425888062,
"epoch": 0.33929601846508944,
"grad_norm": 0.83984375,
"learning_rate": 0.0004994311862588286,
"loss": 6.5508,
"mean_token_accuracy": 0.15420828610658646,
"num_tokens": 7458314.0,
"step": 2940
},
{
"entropy": 6.676367473602295,
"epoch": 0.3398730525100981,
"grad_norm": 0.87890625,
"learning_rate": 0.0004994282501794551,
"loss": 6.5995,
"mean_token_accuracy": 0.15544719249010086,
"num_tokens": 7471995.0,
"step": 2945
},
{
"entropy": 6.703731489181519,
"epoch": 0.3404500865551068,
"grad_norm": 0.7734375,
"learning_rate": 0.0004994253065515486,
"loss": 6.622,
"mean_token_accuracy": 0.1454193964600563,
"num_tokens": 7485676.0,
"step": 2950
},
{
"entropy": 6.740993881225586,
"epoch": 0.3410271206001154,
"grad_norm": 0.81640625,
"learning_rate": 0.0004994223553752084,
"loss": 6.5061,
"mean_token_accuracy": 0.15575395077466964,
"num_tokens": 7497057.0,
"step": 2955
},
{
"entropy": 6.6203021049499515,
"epoch": 0.34160415464512406,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994193966505339,
"loss": 6.5331,
"mean_token_accuracy": 0.15374020338058472,
"num_tokens": 7510625.0,
"step": 2960
},
{
"entropy": 6.623828363418579,
"epoch": 0.3421811886901327,
"grad_norm": 0.83984375,
"learning_rate": 0.0004994164303776244,
"loss": 6.5414,
"mean_token_accuracy": 0.15313206017017364,
"num_tokens": 7522677.0,
"step": 2965
},
{
"entropy": 6.613079309463501,
"epoch": 0.3427582227351414,
"grad_norm": 0.78125,
"learning_rate": 0.0004994134565565797,
"loss": 6.4729,
"mean_token_accuracy": 0.15777821689844132,
"num_tokens": 7535406.0,
"step": 2970
},
{
"entropy": 6.725028848648071,
"epoch": 0.34333525678015003,
"grad_norm": 0.82421875,
"learning_rate": 0.0004994104751875,
"loss": 6.6259,
"mean_token_accuracy": 0.1496363550424576,
"num_tokens": 7547577.0,
"step": 2975
},
{
"entropy": 6.7461694240570065,
"epoch": 0.3439122908251587,
"grad_norm": 0.80859375,
"learning_rate": 0.0004994074862704854,
"loss": 6.6151,
"mean_token_accuracy": 0.15172602161765097,
"num_tokens": 7561734.0,
"step": 2980
},
{
"entropy": 6.712504768371582,
"epoch": 0.3444893248701673,
"grad_norm": 0.890625,
"learning_rate": 0.0004994044898056363,
"loss": 6.6621,
"mean_token_accuracy": 0.14886820018291474,
"num_tokens": 7574256.0,
"step": 2985
},
{
"entropy": 6.752849674224853,
"epoch": 0.345066358915176,
"grad_norm": 0.8125,
"learning_rate": 0.0004994014857930538,
"loss": 6.5801,
"mean_token_accuracy": 0.15615265667438508,
"num_tokens": 7587149.0,
"step": 2990
},
{
"entropy": 6.642135143280029,
"epoch": 0.34564339296018465,
"grad_norm": 0.87109375,
"learning_rate": 0.0004993984742328387,
"loss": 6.5502,
"mean_token_accuracy": 0.15114353597164154,
"num_tokens": 7599124.0,
"step": 2995
},
{
"entropy": 6.641741514205933,
"epoch": 0.3462204270051933,
"grad_norm": 0.82421875,
"learning_rate": 0.0004993954551250925,
"loss": 6.5556,
"mean_token_accuracy": 0.1523570418357849,
"num_tokens": 7612070.0,
"step": 3000
},
{
"epoch": 0.3462204270051933,
"eval_entropy": 6.5966810582204065,
"eval_loss": 6.6427154541015625,
"eval_mean_token_accuracy": 0.1525940384301639,
"eval_num_tokens": 7612070.0,
"eval_runtime": 17.8197,
"eval_samples_per_second": 1578.922,
"eval_steps_per_second": 197.365,
"step": 3000
},
{
"entropy": 6.603645992279053,
"epoch": 0.346797461050202,
"grad_norm": 0.8046875,
"learning_rate": 0.0004993924284699165,
"loss": 6.4531,
"mean_token_accuracy": 0.15813793241977692,
"num_tokens": 7626096.0,
"step": 3005
},
{
"entropy": 6.70902099609375,
"epoch": 0.3473744950952106,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993893942674127,
"loss": 6.5844,
"mean_token_accuracy": 0.15069538950920106,
"num_tokens": 7639666.0,
"step": 3010
},
{
"entropy": 6.644587516784668,
"epoch": 0.34795152914021926,
"grad_norm": 1.0234375,
"learning_rate": 0.000499386352517683,
"loss": 6.5407,
"mean_token_accuracy": 0.15207042545080185,
"num_tokens": 7652904.0,
"step": 3015
},
{
"entropy": 6.669216442108154,
"epoch": 0.3485285631852279,
"grad_norm": 0.82421875,
"learning_rate": 0.00049938330322083,
"loss": 6.5692,
"mean_token_accuracy": 0.15089073702692984,
"num_tokens": 7665523.0,
"step": 3020
},
{
"entropy": 6.743306303024292,
"epoch": 0.3491055972302366,
"grad_norm": 0.8515625,
"learning_rate": 0.0004993802463769558,
"loss": 6.6148,
"mean_token_accuracy": 0.14982035011053085,
"num_tokens": 7677063.0,
"step": 3025
},
{
"entropy": 6.61863112449646,
"epoch": 0.34968263127524524,
"grad_norm": 0.83203125,
"learning_rate": 0.0004993771819861636,
"loss": 6.5418,
"mean_token_accuracy": 0.15196796357631684,
"num_tokens": 7688980.0,
"step": 3030
},
{
"entropy": 6.773717403411865,
"epoch": 0.3502596653202539,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993741100485564,
"loss": 6.664,
"mean_token_accuracy": 0.15363381579518318,
"num_tokens": 7702604.0,
"step": 3035
},
{
"entropy": 6.7194211959838865,
"epoch": 0.35083669936526257,
"grad_norm": 0.81640625,
"learning_rate": 0.0004993710305642374,
"loss": 6.6045,
"mean_token_accuracy": 0.148685784637928,
"num_tokens": 7715324.0,
"step": 3040
},
{
"entropy": 6.778428649902343,
"epoch": 0.3514137334102712,
"grad_norm": 1.1796875,
"learning_rate": 0.0004993679435333102,
"loss": 6.6311,
"mean_token_accuracy": 0.15059239715337752,
"num_tokens": 7729098.0,
"step": 3045
},
{
"entropy": 6.679904794692993,
"epoch": 0.35199076745527985,
"grad_norm": 0.8359375,
"learning_rate": 0.0004993648489558788,
"loss": 6.5728,
"mean_token_accuracy": 0.14844775199890137,
"num_tokens": 7740717.0,
"step": 3050
},
{
"entropy": 6.585766887664795,
"epoch": 0.3525678015002885,
"grad_norm": 0.953125,
"learning_rate": 0.0004993617468320471,
"loss": 6.5,
"mean_token_accuracy": 0.1497762158513069,
"num_tokens": 7753928.0,
"step": 3055
},
{
"entropy": 6.741098833084107,
"epoch": 0.3531448355452972,
"grad_norm": 0.87890625,
"learning_rate": 0.0004993586371619193,
"loss": 6.5951,
"mean_token_accuracy": 0.14959297627210616,
"num_tokens": 7765728.0,
"step": 3060
},
{
"entropy": 6.595219230651855,
"epoch": 0.3537218695903058,
"grad_norm": 0.828125,
"learning_rate": 0.0004993555199456004,
"loss": 6.54,
"mean_token_accuracy": 0.15218116194009781,
"num_tokens": 7778199.0,
"step": 3065
},
{
"entropy": 6.61001672744751,
"epoch": 0.35429890363531447,
"grad_norm": 0.82421875,
"learning_rate": 0.0004993523951831949,
"loss": 6.5172,
"mean_token_accuracy": 0.16153395771980286,
"num_tokens": 7791026.0,
"step": 3070
},
{
"entropy": 6.656833124160767,
"epoch": 0.35487593768032316,
"grad_norm": 0.84375,
"learning_rate": 0.0004993492628748081,
"loss": 6.5961,
"mean_token_accuracy": 0.1556033119559288,
"num_tokens": 7804333.0,
"step": 3075
},
{
"entropy": 6.677346849441529,
"epoch": 0.3554529717253318,
"grad_norm": 0.73046875,
"learning_rate": 0.0004993461230205453,
"loss": 6.597,
"mean_token_accuracy": 0.1483662024140358,
"num_tokens": 7818339.0,
"step": 3080
},
{
"entropy": 6.623977136611939,
"epoch": 0.35603000577034044,
"grad_norm": 0.85546875,
"learning_rate": 0.0004993429756205121,
"loss": 6.5718,
"mean_token_accuracy": 0.15360086113214494,
"num_tokens": 7831537.0,
"step": 3085
},
{
"entropy": 6.702095079421997,
"epoch": 0.3566070398153491,
"grad_norm": 0.8828125,
"learning_rate": 0.0004993398206748142,
"loss": 6.576,
"mean_token_accuracy": 0.15210341215133666,
"num_tokens": 7843705.0,
"step": 3090
},
{
"entropy": 6.664427471160889,
"epoch": 0.3571840738603578,
"grad_norm": 0.79296875,
"learning_rate": 0.0004993366581835581,
"loss": 6.5747,
"mean_token_accuracy": 0.15079948231577872,
"num_tokens": 7856481.0,
"step": 3095
},
{
"entropy": 6.682329511642456,
"epoch": 0.3577611079053664,
"grad_norm": 0.921875,
"learning_rate": 0.0004993334881468498,
"loss": 6.5722,
"mean_token_accuracy": 0.1513454347848892,
"num_tokens": 7868732.0,
"step": 3100
},
{
"entropy": 6.700321531295776,
"epoch": 0.35833814195037506,
"grad_norm": 0.8046875,
"learning_rate": 0.0004993303105647961,
"loss": 6.5369,
"mean_token_accuracy": 0.1532818630337715,
"num_tokens": 7881470.0,
"step": 3105
},
{
"entropy": 6.685085391998291,
"epoch": 0.35891517599538375,
"grad_norm": 0.80859375,
"learning_rate": 0.0004993271254375038,
"loss": 6.6055,
"mean_token_accuracy": 0.14828379899263383,
"num_tokens": 7895074.0,
"step": 3110
},
{
"entropy": 6.695715379714966,
"epoch": 0.3594922100403924,
"grad_norm": 0.80078125,
"learning_rate": 0.00049932393276508,
"loss": 6.5743,
"mean_token_accuracy": 0.15514225661754608,
"num_tokens": 7909140.0,
"step": 3115
},
{
"entropy": 6.622471666336059,
"epoch": 0.36006924408540103,
"grad_norm": 0.9375,
"learning_rate": 0.0004993207325476323,
"loss": 6.5438,
"mean_token_accuracy": 0.15456567257642745,
"num_tokens": 7922689.0,
"step": 3120
},
{
"entropy": 6.6618585109710695,
"epoch": 0.36064627813040967,
"grad_norm": 0.796875,
"learning_rate": 0.0004993175247852681,
"loss": 6.5498,
"mean_token_accuracy": 0.14956910461187362,
"num_tokens": 7935207.0,
"step": 3125
},
{
"entropy": 6.619128084182739,
"epoch": 0.36122331217541837,
"grad_norm": 0.90625,
"learning_rate": 0.0004993143094780954,
"loss": 6.4961,
"mean_token_accuracy": 0.15582271665334702,
"num_tokens": 7947491.0,
"step": 3130
},
{
"entropy": 6.628898763656617,
"epoch": 0.361800346220427,
"grad_norm": 0.83984375,
"learning_rate": 0.0004993110866262224,
"loss": 6.502,
"mean_token_accuracy": 0.15252822190523146,
"num_tokens": 7959329.0,
"step": 3135
},
{
"entropy": 6.6644287109375,
"epoch": 0.36237738026543564,
"grad_norm": 0.81640625,
"learning_rate": 0.0004993078562297573,
"loss": 6.5191,
"mean_token_accuracy": 0.15571288764476776,
"num_tokens": 7971794.0,
"step": 3140
},
{
"entropy": 6.633200788497925,
"epoch": 0.36295441431044434,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993046182888089,
"loss": 6.658,
"mean_token_accuracy": 0.15406894385814668,
"num_tokens": 7983942.0,
"step": 3145
},
{
"entropy": 6.721062707901001,
"epoch": 0.363531448355453,
"grad_norm": 0.828125,
"learning_rate": 0.0004993013728034861,
"loss": 6.527,
"mean_token_accuracy": 0.15496130287647247,
"num_tokens": 7996266.0,
"step": 3150
},
{
"entropy": 6.672237873077393,
"epoch": 0.3641084824004616,
"grad_norm": 0.87109375,
"learning_rate": 0.000499298119773898,
"loss": 6.5901,
"mean_token_accuracy": 0.14736275672912597,
"num_tokens": 8009318.0,
"step": 3155
},
{
"entropy": 6.651838397979736,
"epoch": 0.36468551644547026,
"grad_norm": 0.8515625,
"learning_rate": 0.0004992948592001541,
"loss": 6.5862,
"mean_token_accuracy": 0.15489773005247115,
"num_tokens": 8021213.0,
"step": 3160
},
{
"entropy": 6.635162830352783,
"epoch": 0.36526255049047895,
"grad_norm": 0.86328125,
"learning_rate": 0.000499291591082364,
"loss": 6.5594,
"mean_token_accuracy": 0.14926373437047005,
"num_tokens": 8033927.0,
"step": 3165
},
{
"entropy": 6.569822597503662,
"epoch": 0.3658395845354876,
"grad_norm": 0.7890625,
"learning_rate": 0.0004992883154206377,
"loss": 6.4402,
"mean_token_accuracy": 0.15944022238254546,
"num_tokens": 8046709.0,
"step": 3170
},
{
"entropy": 6.684438991546631,
"epoch": 0.36641661858049623,
"grad_norm": 0.7734375,
"learning_rate": 0.0004992850322150853,
"loss": 6.5909,
"mean_token_accuracy": 0.14909223467111588,
"num_tokens": 8060127.0,
"step": 3175
},
{
"entropy": 6.648997974395752,
"epoch": 0.36699365262550493,
"grad_norm": 0.8359375,
"learning_rate": 0.0004992817414658172,
"loss": 6.5246,
"mean_token_accuracy": 0.15754903107881546,
"num_tokens": 8072927.0,
"step": 3180
},
{
"entropy": 6.563595008850098,
"epoch": 0.36757068667051357,
"grad_norm": 0.828125,
"learning_rate": 0.0004992784431729442,
"loss": 6.4497,
"mean_token_accuracy": 0.15927306860685347,
"num_tokens": 8083973.0,
"step": 3185
},
{
"entropy": 6.6668370246887205,
"epoch": 0.3681477207155222,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992751373365771,
"loss": 6.5226,
"mean_token_accuracy": 0.15365545451641083,
"num_tokens": 8096801.0,
"step": 3190
},
{
"entropy": 6.642384719848633,
"epoch": 0.36872475476053085,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992718239568273,
"loss": 6.5395,
"mean_token_accuracy": 0.15474597811698915,
"num_tokens": 8109099.0,
"step": 3195
},
{
"entropy": 6.654610347747803,
"epoch": 0.36930178880553954,
"grad_norm": 0.8125,
"learning_rate": 0.000499268503033806,
"loss": 6.4807,
"mean_token_accuracy": 0.15879638940095903,
"num_tokens": 8120712.0,
"step": 3200
},
{
"entropy": 6.599427938461304,
"epoch": 0.3698788228505482,
"grad_norm": 0.81640625,
"learning_rate": 0.0004992651745676249,
"loss": 6.5206,
"mean_token_accuracy": 0.15537554174661636,
"num_tokens": 8132228.0,
"step": 3205
},
{
"entropy": 6.6891693592071535,
"epoch": 0.3704558568955568,
"grad_norm": 0.86328125,
"learning_rate": 0.0004992618385583962,
"loss": 6.531,
"mean_token_accuracy": 0.1528848886489868,
"num_tokens": 8145015.0,
"step": 3210
},
{
"entropy": 6.579451227188111,
"epoch": 0.3710328909405655,
"grad_norm": 0.890625,
"learning_rate": 0.0004992584950062319,
"loss": 6.4827,
"mean_token_accuracy": 0.1555587813258171,
"num_tokens": 8156985.0,
"step": 3215
},
{
"entropy": 6.705874490737915,
"epoch": 0.37160992498557416,
"grad_norm": 1.3359375,
"learning_rate": 0.0004992551439112446,
"loss": 6.6359,
"mean_token_accuracy": 0.14975506514310838,
"num_tokens": 8170633.0,
"step": 3220
},
{
"entropy": 6.702548408508301,
"epoch": 0.3721869590305828,
"grad_norm": 0.79296875,
"learning_rate": 0.0004992517852735469,
"loss": 6.5145,
"mean_token_accuracy": 0.15716839730739593,
"num_tokens": 8182096.0,
"step": 3225
},
{
"entropy": 6.643147325515747,
"epoch": 0.37276399307559144,
"grad_norm": 0.80078125,
"learning_rate": 0.0004992484190932517,
"loss": 6.5591,
"mean_token_accuracy": 0.1519121289253235,
"num_tokens": 8193768.0,
"step": 3230
},
{
"entropy": 6.594755172729492,
"epoch": 0.37334102712060013,
"grad_norm": 0.87109375,
"learning_rate": 0.0004992450453704723,
"loss": 6.4174,
"mean_token_accuracy": 0.1655030742287636,
"num_tokens": 8206237.0,
"step": 3235
},
{
"entropy": 6.629844379425049,
"epoch": 0.3739180611656088,
"grad_norm": 0.84765625,
"learning_rate": 0.0004992416641053223,
"loss": 6.5086,
"mean_token_accuracy": 0.15595395714044571,
"num_tokens": 8218066.0,
"step": 3240
},
{
"entropy": 6.635203266143799,
"epoch": 0.3744950952106174,
"grad_norm": 0.84765625,
"learning_rate": 0.0004992382752979153,
"loss": 6.4969,
"mean_token_accuracy": 0.1581435188651085,
"num_tokens": 8230976.0,
"step": 3245
},
{
"entropy": 6.595430040359497,
"epoch": 0.3750721292556261,
"grad_norm": 0.87890625,
"learning_rate": 0.0004992348789483651,
"loss": 6.4836,
"mean_token_accuracy": 0.16031598895788193,
"num_tokens": 8242713.0,
"step": 3250
},
{
"entropy": 6.556427621841431,
"epoch": 0.37564916330063475,
"grad_norm": 0.8359375,
"learning_rate": 0.0004992314750567864,
"loss": 6.4254,
"mean_token_accuracy": 0.1674373909831047,
"num_tokens": 8255698.0,
"step": 3255
},
{
"entropy": 6.6029973983764645,
"epoch": 0.3762261973456434,
"grad_norm": 0.82421875,
"learning_rate": 0.0004992280636232933,
"loss": 6.5007,
"mean_token_accuracy": 0.158811055123806,
"num_tokens": 8267138.0,
"step": 3260
},
{
"entropy": 6.55572772026062,
"epoch": 0.376803231390652,
"grad_norm": 0.7890625,
"learning_rate": 0.0004992246446480007,
"loss": 6.5147,
"mean_token_accuracy": 0.1620226725935936,
"num_tokens": 8280902.0,
"step": 3265
},
{
"entropy": 6.678423690795898,
"epoch": 0.3773802654356607,
"grad_norm": 0.87109375,
"learning_rate": 0.0004992212181310235,
"loss": 6.5346,
"mean_token_accuracy": 0.1485668234527111,
"num_tokens": 8293027.0,
"step": 3270
},
{
"entropy": 6.711181497573852,
"epoch": 0.37795729948066936,
"grad_norm": 0.828125,
"learning_rate": 0.000499217784072477,
"loss": 6.5666,
"mean_token_accuracy": 0.1553458794951439,
"num_tokens": 8306676.0,
"step": 3275
},
{
"entropy": 6.473035192489624,
"epoch": 0.378534333525678,
"grad_norm": 0.8203125,
"learning_rate": 0.0004992143424724768,
"loss": 6.5013,
"mean_token_accuracy": 0.16312037110328675,
"num_tokens": 8319134.0,
"step": 3280
},
{
"entropy": 6.6510356903076175,
"epoch": 0.3791113675706867,
"grad_norm": 0.8203125,
"learning_rate": 0.0004992108933311385,
"loss": 6.471,
"mean_token_accuracy": 0.15111429542303084,
"num_tokens": 8333010.0,
"step": 3285
},
{
"entropy": 6.661249351501465,
"epoch": 0.37968840161569534,
"grad_norm": 0.84375,
"learning_rate": 0.0004992074366485782,
"loss": 6.5438,
"mean_token_accuracy": 0.15512855350971222,
"num_tokens": 8345576.0,
"step": 3290
},
{
"entropy": 6.524981737136841,
"epoch": 0.380265435660704,
"grad_norm": 0.85546875,
"learning_rate": 0.000499203972424912,
"loss": 6.5412,
"mean_token_accuracy": 0.1497179873287678,
"num_tokens": 8359278.0,
"step": 3295
},
{
"entropy": 6.744177532196045,
"epoch": 0.3808424697057126,
"grad_norm": 0.8203125,
"learning_rate": 0.0004992005006602567,
"loss": 6.5289,
"mean_token_accuracy": 0.1496485650539398,
"num_tokens": 8372159.0,
"step": 3300
},
{
"entropy": 6.55145263671875,
"epoch": 0.3814195037507213,
"grad_norm": 0.76953125,
"learning_rate": 0.0004991970213547289,
"loss": 6.5775,
"mean_token_accuracy": 0.14963392093777655,
"num_tokens": 8385735.0,
"step": 3305
},
{
"entropy": 6.705130910873413,
"epoch": 0.38199653779572995,
"grad_norm": 0.8515625,
"learning_rate": 0.0004991935345084457,
"loss": 6.5483,
"mean_token_accuracy": 0.14926708191633226,
"num_tokens": 8397714.0,
"step": 3310
},
{
"entropy": 6.710420894622803,
"epoch": 0.3825735718407386,
"grad_norm": 0.8671875,
"learning_rate": 0.0004991900401215243,
"loss": 6.5794,
"mean_token_accuracy": 0.14717367589473723,
"num_tokens": 8410812.0,
"step": 3315
},
{
"entropy": 6.6759387969970705,
"epoch": 0.3831506058857473,
"grad_norm": 0.828125,
"learning_rate": 0.0004991865381940822,
"loss": 6.5506,
"mean_token_accuracy": 0.15251373499631882,
"num_tokens": 8422400.0,
"step": 3320
},
{
"entropy": 6.678918695449829,
"epoch": 0.3837276399307559,
"grad_norm": 0.82421875,
"learning_rate": 0.0004991830287262374,
"loss": 6.5629,
"mean_token_accuracy": 0.1532184287905693,
"num_tokens": 8435080.0,
"step": 3325
},
{
"entropy": 6.6434855461120605,
"epoch": 0.38430467397576457,
"grad_norm": 0.76953125,
"learning_rate": 0.0004991795117181077,
"loss": 6.4836,
"mean_token_accuracy": 0.16186368614435195,
"num_tokens": 8447866.0,
"step": 3330
},
{
"entropy": 6.498814868927002,
"epoch": 0.3848817080207732,
"grad_norm": 0.859375,
"learning_rate": 0.0004991759871698113,
"loss": 6.4477,
"mean_token_accuracy": 0.16394371688365936,
"num_tokens": 8461447.0,
"step": 3335
},
{
"entropy": 6.702191495895386,
"epoch": 0.3854587420657819,
"grad_norm": 0.7734375,
"learning_rate": 0.0004991724550814671,
"loss": 6.6218,
"mean_token_accuracy": 0.15160019621253013,
"num_tokens": 8475389.0,
"step": 3340
},
{
"entropy": 6.667648410797119,
"epoch": 0.38603577611079054,
"grad_norm": 0.8046875,
"learning_rate": 0.0004991689154531937,
"loss": 6.5244,
"mean_token_accuracy": 0.15132362991571427,
"num_tokens": 8487271.0,
"step": 3345
},
{
"entropy": 6.618583345413208,
"epoch": 0.3866128101557992,
"grad_norm": 0.81640625,
"learning_rate": 0.0004991653682851102,
"loss": 6.5175,
"mean_token_accuracy": 0.15368050336837769,
"num_tokens": 8498914.0,
"step": 3350
},
{
"entropy": 6.5969428539276125,
"epoch": 0.3871898442008078,
"grad_norm": 0.9140625,
"learning_rate": 0.000499161813577336,
"loss": 6.5066,
"mean_token_accuracy": 0.15542698353528978,
"num_tokens": 8510254.0,
"step": 3355
},
{
"entropy": 6.588715410232544,
"epoch": 0.3877668782458165,
"grad_norm": 0.87109375,
"learning_rate": 0.0004991582513299903,
"loss": 6.5463,
"mean_token_accuracy": 0.15643467903137206,
"num_tokens": 8522612.0,
"step": 3360
},
{
"entropy": 6.603983736038208,
"epoch": 0.38834391229082516,
"grad_norm": 0.84765625,
"learning_rate": 0.0004991546815431932,
"loss": 6.4701,
"mean_token_accuracy": 0.15623842924833298,
"num_tokens": 8534952.0,
"step": 3365
},
{
"entropy": 6.585997819900513,
"epoch": 0.3889209463358338,
"grad_norm": 0.8125,
"learning_rate": 0.000499151104217065,
"loss": 6.5975,
"mean_token_accuracy": 0.15297566056251527,
"num_tokens": 8548388.0,
"step": 3370
},
{
"entropy": 6.616267681121826,
"epoch": 0.3894979803808425,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991475193517254,
"loss": 6.4104,
"mean_token_accuracy": 0.16712062805891037,
"num_tokens": 8561576.0,
"step": 3375
},
{
"entropy": 6.5804126262664795,
"epoch": 0.39007501442585113,
"grad_norm": 0.8359375,
"learning_rate": 0.0004991439269472956,
"loss": 6.5575,
"mean_token_accuracy": 0.15445181280374526,
"num_tokens": 8574253.0,
"step": 3380
},
{
"entropy": 6.6331945896148685,
"epoch": 0.39065204847085977,
"grad_norm": 0.84375,
"learning_rate": 0.0004991403270038961,
"loss": 6.458,
"mean_token_accuracy": 0.16088307946920394,
"num_tokens": 8585708.0,
"step": 3385
},
{
"entropy": 6.60706205368042,
"epoch": 0.3912290825158684,
"grad_norm": 0.8359375,
"learning_rate": 0.000499136719521648,
"loss": 6.5642,
"mean_token_accuracy": 0.14254746586084366,
"num_tokens": 8598327.0,
"step": 3390
},
{
"entropy": 6.649341821670532,
"epoch": 0.3918061165608771,
"grad_norm": 0.890625,
"learning_rate": 0.0004991331045006726,
"loss": 6.569,
"mean_token_accuracy": 0.1483719229698181,
"num_tokens": 8610606.0,
"step": 3395
},
{
"entropy": 6.717316722869873,
"epoch": 0.39238315060588574,
"grad_norm": 0.8125,
"learning_rate": 0.0004991294819410916,
"loss": 6.5855,
"mean_token_accuracy": 0.14726121872663497,
"num_tokens": 8625201.0,
"step": 3400
},
{
"entropy": 6.582614088058472,
"epoch": 0.3929601846508944,
"grad_norm": 0.82421875,
"learning_rate": 0.0004991258518430268,
"loss": 6.5456,
"mean_token_accuracy": 0.16018152683973313,
"num_tokens": 8638288.0,
"step": 3405
},
{
"entropy": 6.658870792388916,
"epoch": 0.3935372186959031,
"grad_norm": 0.86328125,
"learning_rate": 0.0004991222142066003,
"loss": 6.4757,
"mean_token_accuracy": 0.15923092365264893,
"num_tokens": 8649550.0,
"step": 3410
},
{
"entropy": 6.565553760528564,
"epoch": 0.3941142527409117,
"grad_norm": 0.875,
"learning_rate": 0.0004991185690319345,
"loss": 6.4775,
"mean_token_accuracy": 0.1563425600528717,
"num_tokens": 8660956.0,
"step": 3415
},
{
"entropy": 6.676780319213867,
"epoch": 0.39469128678592036,
"grad_norm": 0.89453125,
"learning_rate": 0.000499114916319152,
"loss": 6.5616,
"mean_token_accuracy": 0.14661440700292588,
"num_tokens": 8675130.0,
"step": 3420
},
{
"entropy": 6.595038652420044,
"epoch": 0.395268320830929,
"grad_norm": 0.8203125,
"learning_rate": 0.0004991112560683755,
"loss": 6.4548,
"mean_token_accuracy": 0.1577983319759369,
"num_tokens": 8687482.0,
"step": 3425
},
{
"entropy": 6.5211607933044435,
"epoch": 0.3958453548759377,
"grad_norm": 0.85546875,
"learning_rate": 0.0004991075882797283,
"loss": 6.4495,
"mean_token_accuracy": 0.16318398714065552,
"num_tokens": 8700202.0,
"step": 3430
},
{
"entropy": 6.686310291290283,
"epoch": 0.39642238892094633,
"grad_norm": 0.7890625,
"learning_rate": 0.0004991039129533337,
"loss": 6.6103,
"mean_token_accuracy": 0.1472760871052742,
"num_tokens": 8713602.0,
"step": 3435
},
{
"entropy": 6.605076885223388,
"epoch": 0.396999422965955,
"grad_norm": 0.81640625,
"learning_rate": 0.0004991002300893152,
"loss": 6.5444,
"mean_token_accuracy": 0.15414105653762816,
"num_tokens": 8727318.0,
"step": 3440
},
{
"entropy": 6.642222690582275,
"epoch": 0.39757645701096367,
"grad_norm": 0.8125,
"learning_rate": 0.0004990965396877969,
"loss": 6.4035,
"mean_token_accuracy": 0.16198018193244934,
"num_tokens": 8739941.0,
"step": 3445
},
{
"entropy": 6.616573715209961,
"epoch": 0.3981534910559723,
"grad_norm": 0.8046875,
"learning_rate": 0.0004990928417489027,
"loss": 6.5685,
"mean_token_accuracy": 0.1492309018969536,
"num_tokens": 8753012.0,
"step": 3450
},
{
"entropy": 6.5814603805542,
"epoch": 0.39873052510098095,
"grad_norm": 0.87109375,
"learning_rate": 0.0004990891362727572,
"loss": 6.4986,
"mean_token_accuracy": 0.15747803896665574,
"num_tokens": 8766998.0,
"step": 3455
},
{
"entropy": 6.569999074935913,
"epoch": 0.3993075591459896,
"grad_norm": 0.8828125,
"learning_rate": 0.0004990854232594848,
"loss": 6.4675,
"mean_token_accuracy": 0.15911515578627586,
"num_tokens": 8778993.0,
"step": 3460
},
{
"entropy": 6.586950969696045,
"epoch": 0.3998845931909983,
"grad_norm": 0.859375,
"learning_rate": 0.0004990817027092106,
"loss": 6.46,
"mean_token_accuracy": 0.15577831119298935,
"num_tokens": 8790612.0,
"step": 3465
},
{
"entropy": 6.649136638641357,
"epoch": 0.4004616272360069,
"grad_norm": 0.828125,
"learning_rate": 0.0004990779746220595,
"loss": 6.5891,
"mean_token_accuracy": 0.15244996547698975,
"num_tokens": 8803040.0,
"step": 3470
},
{
"entropy": 6.64963927268982,
"epoch": 0.40103866128101556,
"grad_norm": 0.80078125,
"learning_rate": 0.0004990742389981572,
"loss": 6.5433,
"mean_token_accuracy": 0.1518448457121849,
"num_tokens": 8816179.0,
"step": 3475
},
{
"entropy": 6.681894111633301,
"epoch": 0.40161569532602426,
"grad_norm": 0.8046875,
"learning_rate": 0.000499070495837629,
"loss": 6.5148,
"mean_token_accuracy": 0.1452247053384781,
"num_tokens": 8828779.0,
"step": 3480
},
{
"entropy": 6.678278493881225,
"epoch": 0.4021927293710329,
"grad_norm": 0.7265625,
"learning_rate": 0.0004990667451406012,
"loss": 6.5099,
"mean_token_accuracy": 0.1595836415886879,
"num_tokens": 8842354.0,
"step": 3485
},
{
"entropy": 6.577237606048584,
"epoch": 0.40276976341604154,
"grad_norm": 0.74609375,
"learning_rate": 0.0004990629869071995,
"loss": 6.5268,
"mean_token_accuracy": 0.14757629632949829,
"num_tokens": 8855462.0,
"step": 3490
},
{
"entropy": 6.64400954246521,
"epoch": 0.4033467974610502,
"grad_norm": 0.87109375,
"learning_rate": 0.0004990592211375506,
"loss": 6.5004,
"mean_token_accuracy": 0.15390561521053314,
"num_tokens": 8867882.0,
"step": 3495
},
{
"entropy": 6.5557056903839115,
"epoch": 0.4039238315060589,
"grad_norm": 0.87890625,
"learning_rate": 0.0004990554478317811,
"loss": 6.4443,
"mean_token_accuracy": 0.1584362655878067,
"num_tokens": 8880906.0,
"step": 3500
},
{
"entropy": 6.489110326766967,
"epoch": 0.4045008655510675,
"grad_norm": 0.8671875,
"learning_rate": 0.0004990516669900179,
"loss": 6.3377,
"mean_token_accuracy": 0.16366536319255828,
"num_tokens": 8893005.0,
"step": 3505
},
{
"entropy": 6.614466857910156,
"epoch": 0.40507789959607615,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990478786123882,
"loss": 6.5371,
"mean_token_accuracy": 0.14857055991888046,
"num_tokens": 8905611.0,
"step": 3510
},
{
"entropy": 6.641386365890503,
"epoch": 0.40565493364108485,
"grad_norm": 0.98046875,
"learning_rate": 0.0004990440826990193,
"loss": 6.468,
"mean_token_accuracy": 0.15724545121192932,
"num_tokens": 8918454.0,
"step": 3515
},
{
"entropy": 6.577886247634888,
"epoch": 0.4062319676860935,
"grad_norm": 0.8671875,
"learning_rate": 0.0004990402792500392,
"loss": 6.4571,
"mean_token_accuracy": 0.1559569999575615,
"num_tokens": 8931871.0,
"step": 3520
},
{
"entropy": 6.567701816558838,
"epoch": 0.4068090017311021,
"grad_norm": 0.83203125,
"learning_rate": 0.0004990364682655754,
"loss": 6.5097,
"mean_token_accuracy": 0.15251438468694686,
"num_tokens": 8944813.0,
"step": 3525
},
{
"entropy": 6.611566114425659,
"epoch": 0.40738603577611077,
"grad_norm": 0.796875,
"learning_rate": 0.0004990326497457564,
"loss": 6.5267,
"mean_token_accuracy": 0.15673490911722182,
"num_tokens": 8958902.0,
"step": 3530
},
{
"entropy": 6.5758881092071535,
"epoch": 0.40796306982111946,
"grad_norm": 0.828125,
"learning_rate": 0.0004990288236907104,
"loss": 6.5406,
"mean_token_accuracy": 0.15299722254276277,
"num_tokens": 8971390.0,
"step": 3535
},
{
"entropy": 6.597666025161743,
"epoch": 0.4085401038661281,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990249901005661,
"loss": 6.4701,
"mean_token_accuracy": 0.15784490257501602,
"num_tokens": 8983510.0,
"step": 3540
},
{
"entropy": 6.558063411712647,
"epoch": 0.40911713791113674,
"grad_norm": 0.84375,
"learning_rate": 0.0004990211489754527,
"loss": 6.4978,
"mean_token_accuracy": 0.1489548921585083,
"num_tokens": 8997039.0,
"step": 3545
},
{
"entropy": 6.638218259811401,
"epoch": 0.40969417195614544,
"grad_norm": 0.80078125,
"learning_rate": 0.0004990173003154993,
"loss": 6.5936,
"mean_token_accuracy": 0.14932667464017868,
"num_tokens": 9009672.0,
"step": 3550
},
{
"entropy": 6.660990905761719,
"epoch": 0.4102712060011541,
"grad_norm": 0.76953125,
"learning_rate": 0.0004990134441208351,
"loss": 6.5222,
"mean_token_accuracy": 0.15500398725271225,
"num_tokens": 9023477.0,
"step": 3555
},
{
"entropy": 6.542504739761353,
"epoch": 0.4108482400461627,
"grad_norm": 1.03125,
"learning_rate": 0.0004990095803915901,
"loss": 6.3898,
"mean_token_accuracy": 0.16572054326534272,
"num_tokens": 9035914.0,
"step": 3560
},
{
"entropy": 6.596242094039917,
"epoch": 0.41142527409117136,
"grad_norm": 0.8671875,
"learning_rate": 0.0004990057091278942,
"loss": 6.4774,
"mean_token_accuracy": 0.15172698199748993,
"num_tokens": 9048205.0,
"step": 3565
},
{
"entropy": 6.598004865646362,
"epoch": 0.41200230813618005,
"grad_norm": 0.765625,
"learning_rate": 0.0004990018303298773,
"loss": 6.5425,
"mean_token_accuracy": 0.15234472006559371,
"num_tokens": 9061467.0,
"step": 3570
},
{
"entropy": 6.677182149887085,
"epoch": 0.4125793421811887,
"grad_norm": 0.80859375,
"learning_rate": 0.0004989979439976703,
"loss": 6.504,
"mean_token_accuracy": 0.15467257797718048,
"num_tokens": 9074434.0,
"step": 3575
},
{
"entropy": 6.4798554420471195,
"epoch": 0.41315637622619733,
"grad_norm": 0.8125,
"learning_rate": 0.0004989940501314037,
"loss": 6.3941,
"mean_token_accuracy": 0.16334292888641358,
"num_tokens": 9087745.0,
"step": 3580
},
{
"entropy": 6.609991073608398,
"epoch": 0.413733410271206,
"grad_norm": 0.796875,
"learning_rate": 0.0004989901487312084,
"loss": 6.4618,
"mean_token_accuracy": 0.15822864919900895,
"num_tokens": 9099424.0,
"step": 3585
},
{
"entropy": 6.562083673477173,
"epoch": 0.41431044431621467,
"grad_norm": 0.85546875,
"learning_rate": 0.0004989862397972157,
"loss": 6.4246,
"mean_token_accuracy": 0.1650937482714653,
"num_tokens": 9111056.0,
"step": 3590
},
{
"entropy": 6.623899793624878,
"epoch": 0.4148874783612233,
"grad_norm": 1.046875,
"learning_rate": 0.0004989823233295572,
"loss": 6.5662,
"mean_token_accuracy": 0.15530386716127395,
"num_tokens": 9124539.0,
"step": 3595
},
{
"entropy": 6.6407468795776365,
"epoch": 0.41546451240623195,
"grad_norm": 0.78125,
"learning_rate": 0.0004989783993283644,
"loss": 6.4541,
"mean_token_accuracy": 0.15695587545633316,
"num_tokens": 9136580.0,
"step": 3600
},
{
"entropy": 6.544411754608154,
"epoch": 0.41604154645124064,
"grad_norm": 0.859375,
"learning_rate": 0.0004989744677937694,
"loss": 6.5157,
"mean_token_accuracy": 0.15283992290496826,
"num_tokens": 9148411.0,
"step": 3605
},
{
"entropy": 6.5882384300231935,
"epoch": 0.4166185804962493,
"grad_norm": 0.85546875,
"learning_rate": 0.0004989705287259046,
"loss": 6.4407,
"mean_token_accuracy": 0.1591333732008934,
"num_tokens": 9160778.0,
"step": 3610
},
{
"entropy": 6.622359991073608,
"epoch": 0.4171956145412579,
"grad_norm": 0.8515625,
"learning_rate": 0.0004989665821249021,
"loss": 6.4965,
"mean_token_accuracy": 0.15284003764390947,
"num_tokens": 9174174.0,
"step": 3615
},
{
"entropy": 6.63104214668274,
"epoch": 0.4177726485862666,
"grad_norm": 0.82421875,
"learning_rate": 0.000498962627990895,
"loss": 6.473,
"mean_token_accuracy": 0.1580877423286438,
"num_tokens": 9186465.0,
"step": 3620
},
{
"entropy": 6.522603988647461,
"epoch": 0.41834968263127525,
"grad_norm": 0.88671875,
"learning_rate": 0.0004989586663240161,
"loss": 6.4506,
"mean_token_accuracy": 0.15930631905794143,
"num_tokens": 9198900.0,
"step": 3625
},
{
"entropy": 6.6117743968963625,
"epoch": 0.4189267166762839,
"grad_norm": 0.82421875,
"learning_rate": 0.0004989546971243988,
"loss": 6.4617,
"mean_token_accuracy": 0.15446389317512513,
"num_tokens": 9211148.0,
"step": 3630
},
{
"entropy": 6.6376965045928955,
"epoch": 0.41950375072129253,
"grad_norm": 0.8203125,
"learning_rate": 0.0004989507203921763,
"loss": 6.5696,
"mean_token_accuracy": 0.1492958277463913,
"num_tokens": 9226216.0,
"step": 3635
},
{
"entropy": 6.662663078308105,
"epoch": 0.42008078476630123,
"grad_norm": 0.8046875,
"learning_rate": 0.0004989467361274828,
"loss": 6.5074,
"mean_token_accuracy": 0.15405069887638093,
"num_tokens": 9238361.0,
"step": 3640
},
{
"entropy": 6.5690850734710695,
"epoch": 0.42065781881130987,
"grad_norm": 0.8828125,
"learning_rate": 0.0004989427443304519,
"loss": 6.428,
"mean_token_accuracy": 0.1591155469417572,
"num_tokens": 9251828.0,
"step": 3645
},
{
"entropy": 6.554191303253174,
"epoch": 0.4212348528563185,
"grad_norm": 0.8671875,
"learning_rate": 0.000498938745001218,
"loss": 6.475,
"mean_token_accuracy": 0.1533695265650749,
"num_tokens": 9264433.0,
"step": 3650
},
{
"entropy": 6.662342739105225,
"epoch": 0.4218118869013272,
"grad_norm": 0.77734375,
"learning_rate": 0.0004989347381399158,
"loss": 6.506,
"mean_token_accuracy": 0.15566451102495193,
"num_tokens": 9277859.0,
"step": 3655
},
{
"entropy": 6.549519300460815,
"epoch": 0.42238892094633584,
"grad_norm": 0.7578125,
"learning_rate": 0.0004989307237466799,
"loss": 6.524,
"mean_token_accuracy": 0.1491675093770027,
"num_tokens": 9290585.0,
"step": 3660
},
{
"entropy": 6.6417014598846436,
"epoch": 0.4229659549913445,
"grad_norm": 0.828125,
"learning_rate": 0.0004989267018216453,
"loss": 6.5246,
"mean_token_accuracy": 0.1502677731215954,
"num_tokens": 9302633.0,
"step": 3665
},
{
"entropy": 6.578337478637695,
"epoch": 0.4235429890363531,
"grad_norm": 0.82421875,
"learning_rate": 0.0004989226723649473,
"loss": 6.5606,
"mean_token_accuracy": 0.14945575445890427,
"num_tokens": 9314577.0,
"step": 3670
},
{
"entropy": 6.560885238647461,
"epoch": 0.4241200230813618,
"grad_norm": 0.953125,
"learning_rate": 0.0004989186353767214,
"loss": 6.421,
"mean_token_accuracy": 0.16296332329511642,
"num_tokens": 9327260.0,
"step": 3675
},
{
"entropy": 6.561640357971191,
"epoch": 0.42469705712637046,
"grad_norm": 0.80078125,
"learning_rate": 0.0004989145908571035,
"loss": 6.485,
"mean_token_accuracy": 0.16140415668487548,
"num_tokens": 9339684.0,
"step": 3680
},
{
"entropy": 6.605708312988281,
"epoch": 0.4252740911713791,
"grad_norm": 0.875,
"learning_rate": 0.0004989105388062295,
"loss": 6.4979,
"mean_token_accuracy": 0.1537920966744423,
"num_tokens": 9352532.0,
"step": 3685
},
{
"entropy": 6.655845975875854,
"epoch": 0.4258511252163878,
"grad_norm": 0.875,
"learning_rate": 0.0004989064792242358,
"loss": 6.5319,
"mean_token_accuracy": 0.1520333930850029,
"num_tokens": 9364680.0,
"step": 3690
},
{
"entropy": 6.6132406234741214,
"epoch": 0.42642815926139643,
"grad_norm": 0.80859375,
"learning_rate": 0.0004989024121112589,
"loss": 6.4949,
"mean_token_accuracy": 0.1559430792927742,
"num_tokens": 9377472.0,
"step": 3695
},
{
"entropy": 6.559667921066284,
"epoch": 0.4270051933064051,
"grad_norm": 0.76953125,
"learning_rate": 0.0004988983374674356,
"loss": 6.5138,
"mean_token_accuracy": 0.15148606151342392,
"num_tokens": 9391646.0,
"step": 3700
},
{
"entropy": 6.607914209365845,
"epoch": 0.4275822273514137,
"grad_norm": 0.890625,
"learning_rate": 0.0004988942552929029,
"loss": 6.461,
"mean_token_accuracy": 0.1514444500207901,
"num_tokens": 9402864.0,
"step": 3705
},
{
"entropy": 6.598295068740844,
"epoch": 0.4281592613964224,
"grad_norm": 0.828125,
"learning_rate": 0.0004988901655877981,
"loss": 6.4723,
"mean_token_accuracy": 0.1514604777097702,
"num_tokens": 9414498.0,
"step": 3710
},
{
"entropy": 6.556909513473511,
"epoch": 0.42873629544143105,
"grad_norm": 0.8828125,
"learning_rate": 0.0004988860683522589,
"loss": 6.5078,
"mean_token_accuracy": 0.15649012476205826,
"num_tokens": 9427903.0,
"step": 3715
},
{
"entropy": 6.611214685440063,
"epoch": 0.4293133294864397,
"grad_norm": 0.7734375,
"learning_rate": 0.000498881963586423,
"loss": 6.4587,
"mean_token_accuracy": 0.16037748903036117,
"num_tokens": 9441566.0,
"step": 3720
},
{
"entropy": 6.466671895980835,
"epoch": 0.4298903635314484,
"grad_norm": 0.85546875,
"learning_rate": 0.0004988778512904282,
"loss": 6.3517,
"mean_token_accuracy": 0.1674039363861084,
"num_tokens": 9453614.0,
"step": 3725
},
{
"entropy": 6.64181957244873,
"epoch": 0.430467397576457,
"grad_norm": 0.84765625,
"learning_rate": 0.0004988737314644134,
"loss": 6.5038,
"mean_token_accuracy": 0.15030203759670258,
"num_tokens": 9468046.0,
"step": 3730
},
{
"entropy": 6.4948383331298825,
"epoch": 0.43104443162146566,
"grad_norm": 0.77734375,
"learning_rate": 0.0004988696041085168,
"loss": 6.3915,
"mean_token_accuracy": 0.16536259204149245,
"num_tokens": 9481469.0,
"step": 3735
},
{
"entropy": 6.563466310501099,
"epoch": 0.4316214656664743,
"grad_norm": 0.81640625,
"learning_rate": 0.0004988654692228772,
"loss": 6.4936,
"mean_token_accuracy": 0.15875237435102463,
"num_tokens": 9492833.0,
"step": 3740
},
{
"entropy": 6.588851833343506,
"epoch": 0.432198499711483,
"grad_norm": 0.83984375,
"learning_rate": 0.0004988613268076335,
"loss": 6.4178,
"mean_token_accuracy": 0.15342947393655776,
"num_tokens": 9504955.0,
"step": 3745
},
{
"entropy": 6.5497125625610355,
"epoch": 0.43277553375649164,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988571768629257,
"loss": 6.425,
"mean_token_accuracy": 0.16287132054567338,
"num_tokens": 9516830.0,
"step": 3750
},
{
"entropy": 6.517823219299316,
"epoch": 0.4333525678015003,
"grad_norm": 0.8515625,
"learning_rate": 0.0004988530193888927,
"loss": 6.4592,
"mean_token_accuracy": 0.15513070076704025,
"num_tokens": 9528965.0,
"step": 3755
},
{
"entropy": 6.682994794845581,
"epoch": 0.4339296018465089,
"grad_norm": 0.8046875,
"learning_rate": 0.0004988488543856747,
"loss": 6.5369,
"mean_token_accuracy": 0.1505722388625145,
"num_tokens": 9541838.0,
"step": 3760
},
{
"entropy": 6.553580951690674,
"epoch": 0.4345066358915176,
"grad_norm": 0.7578125,
"learning_rate": 0.0004988446818534115,
"loss": 6.4714,
"mean_token_accuracy": 0.15720604211091996,
"num_tokens": 9554502.0,
"step": 3765
},
{
"entropy": 6.539259767532348,
"epoch": 0.43508366993652625,
"grad_norm": 0.95703125,
"learning_rate": 0.0004988405017922438,
"loss": 6.3194,
"mean_token_accuracy": 0.16168797761201859,
"num_tokens": 9567028.0,
"step": 3770
},
{
"entropy": 6.590677881240845,
"epoch": 0.4356607039815349,
"grad_norm": 0.79296875,
"learning_rate": 0.000498836314202312,
"loss": 6.4877,
"mean_token_accuracy": 0.14764939695596696,
"num_tokens": 9580336.0,
"step": 3775
},
{
"entropy": 6.573682975769043,
"epoch": 0.4362377380265436,
"grad_norm": 0.7890625,
"learning_rate": 0.0004988321190837568,
"loss": 6.4161,
"mean_token_accuracy": 0.1552751526236534,
"num_tokens": 9592410.0,
"step": 3780
},
{
"entropy": 6.57819676399231,
"epoch": 0.4368147720715522,
"grad_norm": 0.86328125,
"learning_rate": 0.0004988279164367196,
"loss": 6.4278,
"mean_token_accuracy": 0.15180183351039886,
"num_tokens": 9605693.0,
"step": 3785
},
{
"entropy": 6.623029613494873,
"epoch": 0.43739180611656087,
"grad_norm": 0.78125,
"learning_rate": 0.0004988237062613415,
"loss": 6.4494,
"mean_token_accuracy": 0.1581479698419571,
"num_tokens": 9618555.0,
"step": 3790
},
{
"entropy": 6.549624633789063,
"epoch": 0.4379688401615695,
"grad_norm": 0.87109375,
"learning_rate": 0.0004988194885577644,
"loss": 6.4702,
"mean_token_accuracy": 0.15138823837041854,
"num_tokens": 9630330.0,
"step": 3795
},
{
"entropy": 6.601396465301514,
"epoch": 0.4385458742065782,
"grad_norm": 0.87890625,
"learning_rate": 0.0004988152633261299,
"loss": 6.4799,
"mean_token_accuracy": 0.15994462221860886,
"num_tokens": 9642516.0,
"step": 3800
},
{
"entropy": 6.569391345977783,
"epoch": 0.43912290825158684,
"grad_norm": 0.875,
"learning_rate": 0.00049881103056658,
"loss": 6.4483,
"mean_token_accuracy": 0.15572853684425353,
"num_tokens": 9654886.0,
"step": 3805
},
{
"entropy": 6.535688161849976,
"epoch": 0.4396999422965955,
"grad_norm": 0.8515625,
"learning_rate": 0.0004988067902792575,
"loss": 6.4739,
"mean_token_accuracy": 0.1606386885046959,
"num_tokens": 9666346.0,
"step": 3810
},
{
"entropy": 6.550254011154175,
"epoch": 0.4402769763416042,
"grad_norm": 0.81640625,
"learning_rate": 0.0004988025424643047,
"loss": 6.4505,
"mean_token_accuracy": 0.15357128232717515,
"num_tokens": 9679971.0,
"step": 3815
},
{
"entropy": 6.616861248016358,
"epoch": 0.4408540103866128,
"grad_norm": 0.80859375,
"learning_rate": 0.0004987982871218645,
"loss": 6.5232,
"mean_token_accuracy": 0.15241505280137063,
"num_tokens": 9693032.0,
"step": 3820
},
{
"entropy": 6.587024784088134,
"epoch": 0.44143104443162146,
"grad_norm": 0.7421875,
"learning_rate": 0.0004987940242520802,
"loss": 6.4342,
"mean_token_accuracy": 0.16022422015666962,
"num_tokens": 9705706.0,
"step": 3825
},
{
"entropy": 6.461678075790405,
"epoch": 0.4420080784766301,
"grad_norm": 0.796875,
"learning_rate": 0.000498789753855095,
"loss": 6.3687,
"mean_token_accuracy": 0.16021044701337814,
"num_tokens": 9718235.0,
"step": 3830
},
{
"entropy": 6.554477596282959,
"epoch": 0.4425851125216388,
"grad_norm": 0.8203125,
"learning_rate": 0.0004987854759310526,
"loss": 6.5244,
"mean_token_accuracy": 0.1447369821369648,
"num_tokens": 9730758.0,
"step": 3835
},
{
"entropy": 6.565464019775391,
"epoch": 0.44316214656664743,
"grad_norm": 0.82421875,
"learning_rate": 0.0004987811904800968,
"loss": 6.3592,
"mean_token_accuracy": 0.16637056618928908,
"num_tokens": 9743131.0,
"step": 3840
},
{
"entropy": 6.500174474716187,
"epoch": 0.44373918061165607,
"grad_norm": 0.84375,
"learning_rate": 0.0004987768975023719,
"loss": 6.4459,
"mean_token_accuracy": 0.15130365639925003,
"num_tokens": 9755965.0,
"step": 3845
},
{
"entropy": 6.591799068450928,
"epoch": 0.44431621465666477,
"grad_norm": 0.8203125,
"learning_rate": 0.0004987725969980222,
"loss": 6.3492,
"mean_token_accuracy": 0.16706654876470567,
"num_tokens": 9768460.0,
"step": 3850
},
{
"entropy": 6.561370229721069,
"epoch": 0.4448932487016734,
"grad_norm": 0.81640625,
"learning_rate": 0.0004987682889671923,
"loss": 6.5234,
"mean_token_accuracy": 0.15095077604055404,
"num_tokens": 9781653.0,
"step": 3855
},
{
"entropy": 6.50119457244873,
"epoch": 0.44547028274668204,
"grad_norm": 0.81640625,
"learning_rate": 0.0004987639734100272,
"loss": 6.3379,
"mean_token_accuracy": 0.1660802647471428,
"num_tokens": 9793814.0,
"step": 3860
},
{
"entropy": 6.53929648399353,
"epoch": 0.4460473167916907,
"grad_norm": 0.87109375,
"learning_rate": 0.0004987596503266721,
"loss": 6.4071,
"mean_token_accuracy": 0.159231798350811,
"num_tokens": 9806635.0,
"step": 3865
},
{
"entropy": 6.531664180755615,
"epoch": 0.4466243508366994,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987553197172722,
"loss": 6.3851,
"mean_token_accuracy": 0.15614811927080155,
"num_tokens": 9819213.0,
"step": 3870
},
{
"entropy": 6.460581731796265,
"epoch": 0.447201384881708,
"grad_norm": 0.81640625,
"learning_rate": 0.0004987509815819732,
"loss": 6.3818,
"mean_token_accuracy": 0.15675631165504456,
"num_tokens": 9831823.0,
"step": 3875
},
{
"entropy": 6.535300016403198,
"epoch": 0.44777841892671666,
"grad_norm": 1.0703125,
"learning_rate": 0.0004987466359209213,
"loss": 6.3739,
"mean_token_accuracy": 0.15925684422254563,
"num_tokens": 9844993.0,
"step": 3880
},
{
"entropy": 6.474202013015747,
"epoch": 0.44835545297172535,
"grad_norm": 0.83984375,
"learning_rate": 0.0004987422827342622,
"loss": 6.2131,
"mean_token_accuracy": 0.17272613048553467,
"num_tokens": 9857853.0,
"step": 3885
},
{
"entropy": 6.5748802661895756,
"epoch": 0.448932487016734,
"grad_norm": 0.7734375,
"learning_rate": 0.0004987379220221426,
"loss": 6.4607,
"mean_token_accuracy": 0.15414051860570907,
"num_tokens": 9871465.0,
"step": 3890
},
{
"entropy": 6.5424511432647705,
"epoch": 0.44950952106174263,
"grad_norm": 0.7890625,
"learning_rate": 0.0004987335537847092,
"loss": 6.443,
"mean_token_accuracy": 0.16424099653959273,
"num_tokens": 9884973.0,
"step": 3895
},
{
"entropy": 6.639307117462158,
"epoch": 0.4500865551067513,
"grad_norm": 0.78515625,
"learning_rate": 0.0004987291780221088,
"loss": 6.5656,
"mean_token_accuracy": 0.14660622775554658,
"num_tokens": 9898143.0,
"step": 3900
},
{
"entropy": 6.629957818984986,
"epoch": 0.45066358915175997,
"grad_norm": 0.84765625,
"learning_rate": 0.0004987247947344887,
"loss": 6.4866,
"mean_token_accuracy": 0.154781673848629,
"num_tokens": 9910584.0,
"step": 3905
},
{
"entropy": 6.60298523902893,
"epoch": 0.4512406231967686,
"grad_norm": 0.81640625,
"learning_rate": 0.0004987204039219962,
"loss": 6.5319,
"mean_token_accuracy": 0.14886820018291474,
"num_tokens": 9922901.0,
"step": 3910
},
{
"entropy": 6.651839780807495,
"epoch": 0.45181765724177725,
"grad_norm": 0.76953125,
"learning_rate": 0.0004987160055847791,
"loss": 6.523,
"mean_token_accuracy": 0.14895009398460388,
"num_tokens": 9936131.0,
"step": 3915
},
{
"entropy": 6.493554782867432,
"epoch": 0.45239469128678594,
"grad_norm": 0.8359375,
"learning_rate": 0.0004987115997229852,
"loss": 6.2838,
"mean_token_accuracy": 0.1695529192686081,
"num_tokens": 9950292.0,
"step": 3920
},
{
"entropy": 6.512192010879517,
"epoch": 0.4529717253317946,
"grad_norm": 0.82421875,
"learning_rate": 0.0004987071863367629,
"loss": 6.4452,
"mean_token_accuracy": 0.15502430945634843,
"num_tokens": 9963765.0,
"step": 3925
},
{
"entropy": 6.560297203063965,
"epoch": 0.4535487593768032,
"grad_norm": 0.8359375,
"learning_rate": 0.0004987027654262604,
"loss": 6.4376,
"mean_token_accuracy": 0.1608374685049057,
"num_tokens": 9975981.0,
"step": 3930
},
{
"entropy": 6.54636697769165,
"epoch": 0.45412579342181186,
"grad_norm": 0.84765625,
"learning_rate": 0.0004986983369916264,
"loss": 6.3945,
"mean_token_accuracy": 0.15865204036235808,
"num_tokens": 9988763.0,
"step": 3935
},
{
"entropy": 6.480308389663696,
"epoch": 0.45470282746682056,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986939010330102,
"loss": 6.4012,
"mean_token_accuracy": 0.1632219061255455,
"num_tokens": 10000663.0,
"step": 3940
},
{
"entropy": 6.530785274505615,
"epoch": 0.4552798615118292,
"grad_norm": 0.86328125,
"learning_rate": 0.0004986894575505606,
"loss": 6.4096,
"mean_token_accuracy": 0.15825921148061753,
"num_tokens": 10012673.0,
"step": 3945
},
{
"entropy": 6.528098106384277,
"epoch": 0.45585689555683784,
"grad_norm": 0.77734375,
"learning_rate": 0.0004986850065444272,
"loss": 6.4407,
"mean_token_accuracy": 0.15066046863794327,
"num_tokens": 10024410.0,
"step": 3950
},
{
"entropy": 6.55545563697815,
"epoch": 0.45643392960184653,
"grad_norm": 0.88671875,
"learning_rate": 0.0004986805480147598,
"loss": 6.3386,
"mean_token_accuracy": 0.16893032789230347,
"num_tokens": 10036782.0,
"step": 3955
},
{
"entropy": 6.51139645576477,
"epoch": 0.4570109636468552,
"grad_norm": 0.82421875,
"learning_rate": 0.0004986760819617082,
"loss": 6.4205,
"mean_token_accuracy": 0.15875447541475296,
"num_tokens": 10049234.0,
"step": 3960
},
{
"entropy": 6.578258800506592,
"epoch": 0.4575879976918638,
"grad_norm": 0.8125,
"learning_rate": 0.0004986716083854228,
"loss": 6.4511,
"mean_token_accuracy": 0.15605315566062927,
"num_tokens": 10061853.0,
"step": 3965
},
{
"entropy": 6.626540565490723,
"epoch": 0.45816503173687245,
"grad_norm": 0.87890625,
"learning_rate": 0.0004986671272860538,
"loss": 6.47,
"mean_token_accuracy": 0.15049075484275817,
"num_tokens": 10074890.0,
"step": 3970
},
{
"entropy": 6.581723546981811,
"epoch": 0.45874206578188115,
"grad_norm": 0.7421875,
"learning_rate": 0.0004986626386637521,
"loss": 6.5189,
"mean_token_accuracy": 0.1504073806107044,
"num_tokens": 10090226.0,
"step": 3975
},
{
"entropy": 6.5390383243560795,
"epoch": 0.4593190998268898,
"grad_norm": 0.83203125,
"learning_rate": 0.0004986581425186688,
"loss": 6.47,
"mean_token_accuracy": 0.15238555520772934,
"num_tokens": 10103716.0,
"step": 3980
},
{
"entropy": 6.602221393585205,
"epoch": 0.4598961338718984,
"grad_norm": 0.80078125,
"learning_rate": 0.0004986536388509548,
"loss": 6.5884,
"mean_token_accuracy": 0.14533913433551787,
"num_tokens": 10117553.0,
"step": 3985
},
{
"entropy": 6.543276596069336,
"epoch": 0.4604731679169071,
"grad_norm": 0.875,
"learning_rate": 0.0004986491276607618,
"loss": 6.4153,
"mean_token_accuracy": 0.15889854580163956,
"num_tokens": 10129897.0,
"step": 3990
},
{
"entropy": 6.616291904449463,
"epoch": 0.46105020196191576,
"grad_norm": 0.82421875,
"learning_rate": 0.0004986446089482416,
"loss": 6.5394,
"mean_token_accuracy": 0.1505708009004593,
"num_tokens": 10141974.0,
"step": 3995
},
{
"entropy": 6.616165256500244,
"epoch": 0.4616272360069244,
"grad_norm": 0.8359375,
"learning_rate": 0.0004986400827135459,
"loss": 6.4595,
"mean_token_accuracy": 0.14896104633808135,
"num_tokens": 10155471.0,
"step": 4000
},
{
"entropy": 6.565793371200561,
"epoch": 0.46220427005193304,
"grad_norm": 0.8828125,
"learning_rate": 0.0004986355489568272,
"loss": 6.4254,
"mean_token_accuracy": 0.15408960729837418,
"num_tokens": 10167362.0,
"step": 4005
},
{
"entropy": 6.527460670471191,
"epoch": 0.46278130409694174,
"grad_norm": 0.8671875,
"learning_rate": 0.0004986310076782379,
"loss": 6.4344,
"mean_token_accuracy": 0.15517805516719818,
"num_tokens": 10180706.0,
"step": 4010
},
{
"entropy": 6.633301687240601,
"epoch": 0.4633583381419504,
"grad_norm": 0.77734375,
"learning_rate": 0.0004986264588779307,
"loss": 6.4721,
"mean_token_accuracy": 0.15134465247392653,
"num_tokens": 10193642.0,
"step": 4015
},
{
"entropy": 6.590562105178833,
"epoch": 0.463935372186959,
"grad_norm": 0.859375,
"learning_rate": 0.0004986219025560586,
"loss": 6.4913,
"mean_token_accuracy": 0.15406545549631118,
"num_tokens": 10206552.0,
"step": 4020
},
{
"entropy": 6.54869499206543,
"epoch": 0.4645124062319677,
"grad_norm": 0.81640625,
"learning_rate": 0.000498617338712775,
"loss": 6.4934,
"mean_token_accuracy": 0.157432359457016,
"num_tokens": 10219576.0,
"step": 4025
},
{
"entropy": 6.630867338180542,
"epoch": 0.46508944027697635,
"grad_norm": 0.80859375,
"learning_rate": 0.0004986127673482332,
"loss": 6.3994,
"mean_token_accuracy": 0.15101128816604614,
"num_tokens": 10231816.0,
"step": 4030
},
{
"entropy": 6.429687786102295,
"epoch": 0.465666474321985,
"grad_norm": 0.8125,
"learning_rate": 0.0004986081884625871,
"loss": 6.2776,
"mean_token_accuracy": 0.16077178120613098,
"num_tokens": 10244329.0,
"step": 4035
},
{
"entropy": 6.522308301925659,
"epoch": 0.46624350836699363,
"grad_norm": 0.81640625,
"learning_rate": 0.0004986036020559906,
"loss": 6.4183,
"mean_token_accuracy": 0.1520432010293007,
"num_tokens": 10256521.0,
"step": 4040
},
{
"entropy": 6.558711004257202,
"epoch": 0.4668205424120023,
"grad_norm": 0.765625,
"learning_rate": 0.000498599008128598,
"loss": 6.4135,
"mean_token_accuracy": 0.15548562109470368,
"num_tokens": 10269023.0,
"step": 4045
},
{
"entropy": 6.542877626419068,
"epoch": 0.46739757645701097,
"grad_norm": 0.8203125,
"learning_rate": 0.0004985944066805639,
"loss": 6.4606,
"mean_token_accuracy": 0.1481800675392151,
"num_tokens": 10281476.0,
"step": 4050
},
{
"entropy": 6.557100534439087,
"epoch": 0.4679746105020196,
"grad_norm": 0.921875,
"learning_rate": 0.000498589797712043,
"loss": 6.3751,
"mean_token_accuracy": 0.15909120291471482,
"num_tokens": 10293749.0,
"step": 4055
},
{
"entropy": 6.567505502700806,
"epoch": 0.4685516445470283,
"grad_norm": 0.8125,
"learning_rate": 0.0004985851812231903,
"loss": 6.3867,
"mean_token_accuracy": 0.16700287014245987,
"num_tokens": 10306554.0,
"step": 4060
},
{
"entropy": 6.560662698745728,
"epoch": 0.46912867859203694,
"grad_norm": 0.828125,
"learning_rate": 0.0004985805572141611,
"loss": 6.4409,
"mean_token_accuracy": 0.15463736951351165,
"num_tokens": 10319374.0,
"step": 4065
},
{
"entropy": 6.51935601234436,
"epoch": 0.4697057126370456,
"grad_norm": 0.8203125,
"learning_rate": 0.0004985759256851108,
"loss": 6.4818,
"mean_token_accuracy": 0.15578267723321915,
"num_tokens": 10333358.0,
"step": 4070
},
{
"entropy": 6.555917501449585,
"epoch": 0.4702827466820542,
"grad_norm": 0.86328125,
"learning_rate": 0.0004985712866361954,
"loss": 6.4144,
"mean_token_accuracy": 0.16267106384038926,
"num_tokens": 10344830.0,
"step": 4075
},
{
"entropy": 6.5785972595214846,
"epoch": 0.4708597807270629,
"grad_norm": 0.87109375,
"learning_rate": 0.0004985666400675709,
"loss": 6.4821,
"mean_token_accuracy": 0.15228304713964463,
"num_tokens": 10357813.0,
"step": 4080
},
{
"entropy": 6.615689039230347,
"epoch": 0.47143681477207156,
"grad_norm": 0.79296875,
"learning_rate": 0.0004985619859793934,
"loss": 6.5128,
"mean_token_accuracy": 0.15323988795280458,
"num_tokens": 10370925.0,
"step": 4085
},
{
"entropy": 6.606573104858398,
"epoch": 0.4720138488170802,
"grad_norm": 0.75390625,
"learning_rate": 0.0004985573243718195,
"loss": 6.4639,
"mean_token_accuracy": 0.1552817478775978,
"num_tokens": 10384172.0,
"step": 4090
},
{
"entropy": 6.58382625579834,
"epoch": 0.4725908828620889,
"grad_norm": 0.80859375,
"learning_rate": 0.0004985526552450061,
"loss": 6.406,
"mean_token_accuracy": 0.15829960033297538,
"num_tokens": 10396797.0,
"step": 4095
},
{
"entropy": 6.483374834060669,
"epoch": 0.47316791690709753,
"grad_norm": 0.80859375,
"learning_rate": 0.0004985479785991103,
"loss": 6.4509,
"mean_token_accuracy": 0.1525266408920288,
"num_tokens": 10409139.0,
"step": 4100
},
{
"entropy": 6.621217346191406,
"epoch": 0.47374495095210617,
"grad_norm": 0.84765625,
"learning_rate": 0.0004985432944342892,
"loss": 6.3974,
"mean_token_accuracy": 0.15837369561195375,
"num_tokens": 10421442.0,
"step": 4105
},
{
"entropy": 6.532503461837768,
"epoch": 0.4743219849971148,
"grad_norm": 0.89453125,
"learning_rate": 0.0004985386027507003,
"loss": 6.3919,
"mean_token_accuracy": 0.16080056130886078,
"num_tokens": 10433761.0,
"step": 4110
},
{
"entropy": 6.5195691108703615,
"epoch": 0.4748990190421235,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985339035485018,
"loss": 6.4894,
"mean_token_accuracy": 0.15351008325815202,
"num_tokens": 10446508.0,
"step": 4115
},
{
"entropy": 6.506284713745117,
"epoch": 0.47547605308713214,
"grad_norm": 0.80078125,
"learning_rate": 0.0004985291968278513,
"loss": 6.3903,
"mean_token_accuracy": 0.1637458860874176,
"num_tokens": 10459352.0,
"step": 4120
},
{
"entropy": 6.540371036529541,
"epoch": 0.4760530871321408,
"grad_norm": 0.8828125,
"learning_rate": 0.0004985244825889074,
"loss": 6.4646,
"mean_token_accuracy": 0.150443135201931,
"num_tokens": 10470324.0,
"step": 4125
},
{
"entropy": 6.587960052490234,
"epoch": 0.4766301211771495,
"grad_norm": 0.8125,
"learning_rate": 0.0004985197608318284,
"loss": 6.4331,
"mean_token_accuracy": 0.15636301636695862,
"num_tokens": 10482715.0,
"step": 4130
},
{
"entropy": 6.566105747222901,
"epoch": 0.4772071552221581,
"grad_norm": 0.90234375,
"learning_rate": 0.0004985150315567735,
"loss": 6.4303,
"mean_token_accuracy": 0.1571477860212326,
"num_tokens": 10493207.0,
"step": 4135
},
{
"entropy": 6.46118950843811,
"epoch": 0.47778418926716676,
"grad_norm": 0.859375,
"learning_rate": 0.0004985102947639013,
"loss": 6.3845,
"mean_token_accuracy": 0.1525936096906662,
"num_tokens": 10505187.0,
"step": 4140
},
{
"entropy": 6.585748958587646,
"epoch": 0.4783612233121754,
"grad_norm": 0.80078125,
"learning_rate": 0.0004985055504533715,
"loss": 6.422,
"mean_token_accuracy": 0.16249419003725052,
"num_tokens": 10518301.0,
"step": 4145
},
{
"entropy": 6.492847347259522,
"epoch": 0.4789382573571841,
"grad_norm": 0.85546875,
"learning_rate": 0.0004985007986253435,
"loss": 6.4236,
"mean_token_accuracy": 0.15842338651418686,
"num_tokens": 10531288.0,
"step": 4150
},
{
"entropy": 6.4829020500183105,
"epoch": 0.47951529140219273,
"grad_norm": 0.7734375,
"learning_rate": 0.0004984960392799772,
"loss": 6.3133,
"mean_token_accuracy": 0.16635616421699523,
"num_tokens": 10544120.0,
"step": 4155
},
{
"entropy": 6.4867268085479735,
"epoch": 0.4800923254472014,
"grad_norm": 0.8515625,
"learning_rate": 0.0004984912724174326,
"loss": 6.406,
"mean_token_accuracy": 0.16807787865400314,
"num_tokens": 10556453.0,
"step": 4160
},
{
"entropy": 6.576405239105225,
"epoch": 0.48066935949221,
"grad_norm": 0.828125,
"learning_rate": 0.0004984864980378701,
"loss": 6.5096,
"mean_token_accuracy": 0.14650319889187813,
"num_tokens": 10569404.0,
"step": 4165
},
{
"entropy": 6.551389598846436,
"epoch": 0.4812463935372187,
"grad_norm": 0.84375,
"learning_rate": 0.0004984817161414502,
"loss": 6.4266,
"mean_token_accuracy": 0.15406363904476167,
"num_tokens": 10582312.0,
"step": 4170
},
{
"entropy": 6.541248607635498,
"epoch": 0.48182342758222735,
"grad_norm": 0.8203125,
"learning_rate": 0.0004984769267283338,
"loss": 6.4758,
"mean_token_accuracy": 0.15606760680675508,
"num_tokens": 10594391.0,
"step": 4175
},
{
"entropy": 6.58020920753479,
"epoch": 0.482400461627236,
"grad_norm": 0.890625,
"learning_rate": 0.0004984721297986819,
"loss": 6.4124,
"mean_token_accuracy": 0.15352201610803604,
"num_tokens": 10606354.0,
"step": 4180
},
{
"entropy": 6.512746667861938,
"epoch": 0.4829774956722447,
"grad_norm": 0.9921875,
"learning_rate": 0.0004984673253526561,
"loss": 6.4036,
"mean_token_accuracy": 0.1524976074695587,
"num_tokens": 10618051.0,
"step": 4185
},
{
"entropy": 6.496927738189697,
"epoch": 0.4835545297172533,
"grad_norm": 0.8125,
"learning_rate": 0.0004984625133904176,
"loss": 6.3149,
"mean_token_accuracy": 0.16221490651369094,
"num_tokens": 10629688.0,
"step": 4190
},
{
"entropy": 6.55139570236206,
"epoch": 0.48413156376226196,
"grad_norm": 0.82421875,
"learning_rate": 0.0004984576939121286,
"loss": 6.4507,
"mean_token_accuracy": 0.1502139836549759,
"num_tokens": 10642249.0,
"step": 4195
},
{
"entropy": 6.47601842880249,
"epoch": 0.4847085978072706,
"grad_norm": 0.83984375,
"learning_rate": 0.0004984528669179511,
"loss": 6.3853,
"mean_token_accuracy": 0.1614321783185005,
"num_tokens": 10655231.0,
"step": 4200
},
{
"entropy": 6.433983659744262,
"epoch": 0.4852856318522793,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984480324080472,
"loss": 6.3893,
"mean_token_accuracy": 0.16549270153045653,
"num_tokens": 10668443.0,
"step": 4205
},
{
"entropy": 6.628914546966553,
"epoch": 0.48586266589728794,
"grad_norm": 0.84375,
"learning_rate": 0.00049844319038258,
"loss": 6.3931,
"mean_token_accuracy": 0.15670084208250046,
"num_tokens": 10680523.0,
"step": 4210
},
{
"entropy": 6.616771078109741,
"epoch": 0.4864396999422966,
"grad_norm": 0.75390625,
"learning_rate": 0.0004984383408417119,
"loss": 6.4454,
"mean_token_accuracy": 0.1573428303003311,
"num_tokens": 10693223.0,
"step": 4215
},
{
"entropy": 6.509914207458496,
"epoch": 0.4870167339873053,
"grad_norm": 0.82421875,
"learning_rate": 0.0004984334837856064,
"loss": 6.3977,
"mean_token_accuracy": 0.16693476736545562,
"num_tokens": 10705276.0,
"step": 4220
},
{
"entropy": 6.503002166748047,
"epoch": 0.4875937680323139,
"grad_norm": 0.796875,
"learning_rate": 0.0004984286192144263,
"loss": 6.434,
"mean_token_accuracy": 0.16314385533332826,
"num_tokens": 10718495.0,
"step": 4225
},
{
"entropy": 6.521592330932617,
"epoch": 0.48817080207732255,
"grad_norm": 0.81640625,
"learning_rate": 0.0004984237471283359,
"loss": 6.3447,
"mean_token_accuracy": 0.16712241023778915,
"num_tokens": 10730885.0,
"step": 4230
},
{
"entropy": 6.5000770568847654,
"epoch": 0.4887478361223312,
"grad_norm": 0.890625,
"learning_rate": 0.0004984188675274986,
"loss": 6.3908,
"mean_token_accuracy": 0.15794526487588884,
"num_tokens": 10742906.0,
"step": 4235
},
{
"entropy": 6.528513193130493,
"epoch": 0.4893248701673399,
"grad_norm": 0.8125,
"learning_rate": 0.0004984139804120787,
"loss": 6.3446,
"mean_token_accuracy": 0.1597832128405571,
"num_tokens": 10757050.0,
"step": 4240
},
{
"entropy": 6.557890224456787,
"epoch": 0.4899019042123485,
"grad_norm": 0.90234375,
"learning_rate": 0.0004984090857822406,
"loss": 6.3975,
"mean_token_accuracy": 0.1596458524465561,
"num_tokens": 10768667.0,
"step": 4245
},
{
"entropy": 6.591907787322998,
"epoch": 0.49047893825735717,
"grad_norm": 0.8046875,
"learning_rate": 0.0004984041836381488,
"loss": 6.4248,
"mean_token_accuracy": 0.15413296818733216,
"num_tokens": 10781687.0,
"step": 4250
},
{
"entropy": 6.402188920974732,
"epoch": 0.49105597230236586,
"grad_norm": 0.8359375,
"learning_rate": 0.0004983992739799682,
"loss": 6.3199,
"mean_token_accuracy": 0.16379206255078316,
"num_tokens": 10794966.0,
"step": 4255
},
{
"entropy": 6.588558530807495,
"epoch": 0.4916330063473745,
"grad_norm": 0.765625,
"learning_rate": 0.0004983943568078642,
"loss": 6.4368,
"mean_token_accuracy": 0.1489301383495331,
"num_tokens": 10807946.0,
"step": 4260
},
{
"entropy": 6.531991815567016,
"epoch": 0.49221004039238314,
"grad_norm": 0.81640625,
"learning_rate": 0.0004983894321220017,
"loss": 6.3998,
"mean_token_accuracy": 0.15459950715303422,
"num_tokens": 10819258.0,
"step": 4265
},
{
"entropy": 6.468906021118164,
"epoch": 0.4927870744373918,
"grad_norm": 0.88671875,
"learning_rate": 0.0004983844999225468,
"loss": 6.3978,
"mean_token_accuracy": 0.15649582892656327,
"num_tokens": 10831439.0,
"step": 4270
},
{
"entropy": 6.5455316543579105,
"epoch": 0.4933641084824005,
"grad_norm": 0.75390625,
"learning_rate": 0.0004983795602096651,
"loss": 6.3436,
"mean_token_accuracy": 0.16237957179546356,
"num_tokens": 10844214.0,
"step": 4275
},
{
"entropy": 6.526888799667359,
"epoch": 0.4939411425274091,
"grad_norm": 0.828125,
"learning_rate": 0.000498374612983523,
"loss": 6.4648,
"mean_token_accuracy": 0.15004564374685286,
"num_tokens": 10856675.0,
"step": 4280
},
{
"entropy": 6.541804265975952,
"epoch": 0.49451817657241776,
"grad_norm": 0.828125,
"learning_rate": 0.0004983696582442866,
"loss": 6.389,
"mean_token_accuracy": 0.15704446136951447,
"num_tokens": 10869296.0,
"step": 4285
},
{
"entropy": 6.478774929046631,
"epoch": 0.49509521061742645,
"grad_norm": 0.9140625,
"learning_rate": 0.0004983646959921227,
"loss": 6.4454,
"mean_token_accuracy": 0.16479874402284622,
"num_tokens": 10881484.0,
"step": 4290
},
{
"entropy": 6.578076696395874,
"epoch": 0.4956722446624351,
"grad_norm": 0.89453125,
"learning_rate": 0.0004983597262271984,
"loss": 6.4188,
"mean_token_accuracy": 0.16067702323198318,
"num_tokens": 10894322.0,
"step": 4295
},
{
"entropy": 6.481278276443481,
"epoch": 0.49624927870744373,
"grad_norm": 0.796875,
"learning_rate": 0.0004983547489496804,
"loss": 6.3652,
"mean_token_accuracy": 0.17005881816148757,
"num_tokens": 10907568.0,
"step": 4300
},
{
"entropy": 6.5737041473388675,
"epoch": 0.49682631275245237,
"grad_norm": 0.71875,
"learning_rate": 0.0004983497641597365,
"loss": 6.5261,
"mean_token_accuracy": 0.1484769120812416,
"num_tokens": 10921371.0,
"step": 4305
},
{
"entropy": 6.64163818359375,
"epoch": 0.49740334679746107,
"grad_norm": 0.86328125,
"learning_rate": 0.0004983447718575342,
"loss": 6.4558,
"mean_token_accuracy": 0.153278249502182,
"num_tokens": 10934725.0,
"step": 4310
},
{
"entropy": 6.529429626464844,
"epoch": 0.4979803808424697,
"grad_norm": 0.8125,
"learning_rate": 0.0004983397720432414,
"loss": 6.4106,
"mean_token_accuracy": 0.15111638754606246,
"num_tokens": 10947985.0,
"step": 4315
},
{
"entropy": 6.3701372146606445,
"epoch": 0.49855741488747835,
"grad_norm": 0.8828125,
"learning_rate": 0.0004983347647170264,
"loss": 6.2466,
"mean_token_accuracy": 0.17306942492723465,
"num_tokens": 10961112.0,
"step": 4320
},
{
"entropy": 6.4976856231689455,
"epoch": 0.49913444893248704,
"grad_norm": 0.78515625,
"learning_rate": 0.0004983297498790574,
"loss": 6.3554,
"mean_token_accuracy": 0.16256517618894578,
"num_tokens": 10973414.0,
"step": 4325
},
{
"entropy": 6.610036849975586,
"epoch": 0.4997114829774957,
"grad_norm": 0.8828125,
"learning_rate": 0.0004983247275295034,
"loss": 6.3368,
"mean_token_accuracy": 0.16400094330310822,
"num_tokens": 10985609.0,
"step": 4330
},
{
"entropy": 6.477565145492553,
"epoch": 0.5002885170225043,
"grad_norm": 0.8046875,
"learning_rate": 0.0004983196976685329,
"loss": 6.4223,
"mean_token_accuracy": 0.16130915582180022,
"num_tokens": 10998928.0,
"step": 4335
},
{
"entropy": 6.518366050720215,
"epoch": 0.500865551067513,
"grad_norm": 0.796875,
"learning_rate": 0.0004983146602963155,
"loss": 6.3347,
"mean_token_accuracy": 0.15564725548028946,
"num_tokens": 11012254.0,
"step": 4340
},
{
"entropy": 6.5654683113098145,
"epoch": 0.5014425851125216,
"grad_norm": 0.8046875,
"learning_rate": 0.0004983096154130203,
"loss": 6.3981,
"mean_token_accuracy": 0.15106508135795593,
"num_tokens": 11024040.0,
"step": 4345
},
{
"entropy": 6.603226995468139,
"epoch": 0.5020196191575302,
"grad_norm": 0.84375,
"learning_rate": 0.0004983045630188171,
"loss": 6.5357,
"mean_token_accuracy": 0.14584791511297227,
"num_tokens": 11037207.0,
"step": 4350
},
{
"entropy": 6.5520185947418215,
"epoch": 0.502596653202539,
"grad_norm": 0.828125,
"learning_rate": 0.0004982995031138759,
"loss": 6.4134,
"mean_token_accuracy": 0.1584774896502495,
"num_tokens": 11049990.0,
"step": 4355
},
{
"entropy": 6.566530513763428,
"epoch": 0.5031736872475476,
"grad_norm": 0.78515625,
"learning_rate": 0.0004982944356983666,
"loss": 6.4835,
"mean_token_accuracy": 0.15407072454690934,
"num_tokens": 11062365.0,
"step": 4360
},
{
"entropy": 6.509780597686768,
"epoch": 0.5037507212925563,
"grad_norm": 0.81640625,
"learning_rate": 0.0004982893607724602,
"loss": 6.3554,
"mean_token_accuracy": 0.1576795235276222,
"num_tokens": 11073721.0,
"step": 4365
},
{
"entropy": 6.46516637802124,
"epoch": 0.5043277553375649,
"grad_norm": 0.84375,
"learning_rate": 0.000498284278336327,
"loss": 6.3865,
"mean_token_accuracy": 0.15484800785779954,
"num_tokens": 11085834.0,
"step": 4370
},
{
"entropy": 6.561235237121582,
"epoch": 0.5049047893825735,
"grad_norm": 0.93359375,
"learning_rate": 0.0004982791883901379,
"loss": 6.4379,
"mean_token_accuracy": 0.1539237193763256,
"num_tokens": 11099432.0,
"step": 4375
},
{
"entropy": 6.417895078659058,
"epoch": 0.5054818234275822,
"grad_norm": 0.828125,
"learning_rate": 0.0004982740909340643,
"loss": 6.2179,
"mean_token_accuracy": 0.178082175552845,
"num_tokens": 11111169.0,
"step": 4380
},
{
"entropy": 6.4774627685546875,
"epoch": 0.5060588574725908,
"grad_norm": 0.76171875,
"learning_rate": 0.0004982689859682775,
"loss": 6.3629,
"mean_token_accuracy": 0.15544058978557587,
"num_tokens": 11124131.0,
"step": 4385
},
{
"entropy": 6.549461221694946,
"epoch": 0.5066358915175996,
"grad_norm": 0.78125,
"learning_rate": 0.0004982638734929492,
"loss": 6.3555,
"mean_token_accuracy": 0.15999710261821748,
"num_tokens": 11135810.0,
"step": 4390
},
{
"entropy": 6.5114202976226805,
"epoch": 0.5072129255626082,
"grad_norm": 0.84765625,
"learning_rate": 0.0004982587535082515,
"loss": 6.4399,
"mean_token_accuracy": 0.16011044085025788,
"num_tokens": 11148081.0,
"step": 4395
},
{
"entropy": 6.538697814941406,
"epoch": 0.5077899596076169,
"grad_norm": 0.7421875,
"learning_rate": 0.0004982536260143565,
"loss": 6.3381,
"mean_token_accuracy": 0.1549960732460022,
"num_tokens": 11161427.0,
"step": 4400
},
{
"entropy": 6.452679586410523,
"epoch": 0.5083669936526255,
"grad_norm": 0.82421875,
"learning_rate": 0.0004982484910114367,
"loss": 6.2565,
"mean_token_accuracy": 0.16689955294132233,
"num_tokens": 11174183.0,
"step": 4405
},
{
"entropy": 6.442296552658081,
"epoch": 0.5089440276976341,
"grad_norm": 0.81640625,
"learning_rate": 0.0004982433484996648,
"loss": 6.418,
"mean_token_accuracy": 0.15510635450482368,
"num_tokens": 11187151.0,
"step": 4410
},
{
"entropy": 6.545526695251465,
"epoch": 0.5095210617426428,
"grad_norm": 0.796875,
"learning_rate": 0.0004982381984792138,
"loss": 6.298,
"mean_token_accuracy": 0.16621606498956681,
"num_tokens": 11199603.0,
"step": 4415
},
{
"entropy": 6.4789220809936525,
"epoch": 0.5100980957876514,
"grad_norm": 0.87109375,
"learning_rate": 0.0004982330409502568,
"loss": 6.3847,
"mean_token_accuracy": 0.16437848657369614,
"num_tokens": 11211709.0,
"step": 4420
},
{
"entropy": 6.491673469543457,
"epoch": 0.5106751298326602,
"grad_norm": 0.79296875,
"learning_rate": 0.0004982278759129674,
"loss": 6.373,
"mean_token_accuracy": 0.16032237857580184,
"num_tokens": 11224192.0,
"step": 4425
},
{
"entropy": 6.517372989654541,
"epoch": 0.5112521638776688,
"grad_norm": 0.921875,
"learning_rate": 0.0004982227033675194,
"loss": 6.3591,
"mean_token_accuracy": 0.16160479038953782,
"num_tokens": 11237026.0,
"step": 4430
},
{
"entropy": 6.414926385879516,
"epoch": 0.5118291979226774,
"grad_norm": 0.80078125,
"learning_rate": 0.0004982175233140865,
"loss": 6.3635,
"mean_token_accuracy": 0.15989502817392348,
"num_tokens": 11250646.0,
"step": 4435
},
{
"entropy": 6.640326833724975,
"epoch": 0.5124062319676861,
"grad_norm": 0.88671875,
"learning_rate": 0.0004982123357528431,
"loss": 6.4629,
"mean_token_accuracy": 0.15322466492652892,
"num_tokens": 11262991.0,
"step": 4440
},
{
"entropy": 6.547874546051025,
"epoch": 0.5129832660126947,
"grad_norm": 0.82421875,
"learning_rate": 0.0004982071406839636,
"loss": 6.3598,
"mean_token_accuracy": 0.15413384437561034,
"num_tokens": 11275535.0,
"step": 4445
},
{
"entropy": 6.51468071937561,
"epoch": 0.5135603000577034,
"grad_norm": 0.8046875,
"learning_rate": 0.0004982019381076229,
"loss": 6.5046,
"mean_token_accuracy": 0.15111950039863586,
"num_tokens": 11288473.0,
"step": 4450
},
{
"entropy": 6.622634696960449,
"epoch": 0.514137334102712,
"grad_norm": 0.8125,
"learning_rate": 0.0004981967280239958,
"loss": 6.5094,
"mean_token_accuracy": 0.15655739828944207,
"num_tokens": 11300755.0,
"step": 4455
},
{
"entropy": 6.58624119758606,
"epoch": 0.5147143681477208,
"grad_norm": 0.79296875,
"learning_rate": 0.0004981915104332578,
"loss": 6.3661,
"mean_token_accuracy": 0.1594516083598137,
"num_tokens": 11312912.0,
"step": 4460
},
{
"entropy": 6.457141399383545,
"epoch": 0.5152914021927294,
"grad_norm": 0.81640625,
"learning_rate": 0.000498186285335584,
"loss": 6.4066,
"mean_token_accuracy": 0.1596623405814171,
"num_tokens": 11325714.0,
"step": 4465
},
{
"entropy": 6.578498840332031,
"epoch": 0.515868436237738,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981810527311505,
"loss": 6.3365,
"mean_token_accuracy": 0.16815614998340606,
"num_tokens": 11337820.0,
"step": 4470
},
{
"entropy": 6.45233736038208,
"epoch": 0.5164454702827467,
"grad_norm": 0.8671875,
"learning_rate": 0.000498175812620133,
"loss": 6.3944,
"mean_token_accuracy": 0.15770285874605178,
"num_tokens": 11350005.0,
"step": 4475
},
{
"entropy": 6.5402703285217285,
"epoch": 0.5170225043277553,
"grad_norm": 0.75390625,
"learning_rate": 0.0004981705650027081,
"loss": 6.4988,
"mean_token_accuracy": 0.15301365703344344,
"num_tokens": 11362918.0,
"step": 4480
},
{
"entropy": 6.550092267990112,
"epoch": 0.517599538372764,
"grad_norm": 0.8046875,
"learning_rate": 0.000498165309879052,
"loss": 6.3836,
"mean_token_accuracy": 0.15631941705942154,
"num_tokens": 11375186.0,
"step": 4485
},
{
"entropy": 6.527862882614135,
"epoch": 0.5181765724177726,
"grad_norm": 0.8984375,
"learning_rate": 0.0004981600472493415,
"loss": 6.4194,
"mean_token_accuracy": 0.15206652581691743,
"num_tokens": 11387607.0,
"step": 4490
},
{
"entropy": 6.546878957748413,
"epoch": 0.5187536064627813,
"grad_norm": 0.765625,
"learning_rate": 0.0004981547771137537,
"loss": 6.437,
"mean_token_accuracy": 0.1564124882221222,
"num_tokens": 11400090.0,
"step": 4495
},
{
"entropy": 6.549637079238892,
"epoch": 0.51933064050779,
"grad_norm": 1.0625,
"learning_rate": 0.0004981494994724658,
"loss": 6.4771,
"mean_token_accuracy": 0.15591733604669572,
"num_tokens": 11413220.0,
"step": 4500
},
{
"entropy": 6.531276607513428,
"epoch": 0.5199076745527986,
"grad_norm": 0.8515625,
"learning_rate": 0.0004981442143256554,
"loss": 6.3598,
"mean_token_accuracy": 0.1604811280965805,
"num_tokens": 11425043.0,
"step": 4505
},
{
"entropy": 6.486877870559693,
"epoch": 0.5204847085978073,
"grad_norm": 0.87890625,
"learning_rate": 0.0004981389216735001,
"loss": 6.3656,
"mean_token_accuracy": 0.16314008086919785,
"num_tokens": 11438097.0,
"step": 4510
},
{
"entropy": 6.48299241065979,
"epoch": 0.5210617426428159,
"grad_norm": 0.81640625,
"learning_rate": 0.000498133621516178,
"loss": 6.3548,
"mean_token_accuracy": 0.16053190380334853,
"num_tokens": 11450903.0,
"step": 4515
},
{
"entropy": 6.5034607410430905,
"epoch": 0.5216387766878245,
"grad_norm": 0.79296875,
"learning_rate": 0.0004981283138538675,
"loss": 6.4428,
"mean_token_accuracy": 0.15942756831645966,
"num_tokens": 11464566.0,
"step": 4520
},
{
"entropy": 6.546222782135009,
"epoch": 0.5222158107328332,
"grad_norm": 0.7734375,
"learning_rate": 0.000498122998686747,
"loss": 6.3884,
"mean_token_accuracy": 0.1575324922800064,
"num_tokens": 11476664.0,
"step": 4525
},
{
"entropy": 6.478293800354004,
"epoch": 0.5227928447778419,
"grad_norm": 0.875,
"learning_rate": 0.0004981176760149951,
"loss": 6.4246,
"mean_token_accuracy": 0.15483495444059373,
"num_tokens": 11489052.0,
"step": 4530
},
{
"entropy": 6.555764818191529,
"epoch": 0.5233698788228506,
"grad_norm": 0.9453125,
"learning_rate": 0.0004981123458387911,
"loss": 6.3773,
"mean_token_accuracy": 0.1606239289045334,
"num_tokens": 11501774.0,
"step": 4535
},
{
"entropy": 6.521160745620728,
"epoch": 0.5239469128678592,
"grad_norm": 0.87890625,
"learning_rate": 0.0004981070081583142,
"loss": 6.4976,
"mean_token_accuracy": 0.15571219325065613,
"num_tokens": 11514397.0,
"step": 4540
},
{
"entropy": 6.519374561309815,
"epoch": 0.5245239469128679,
"grad_norm": 0.80859375,
"learning_rate": 0.000498101662973744,
"loss": 6.3928,
"mean_token_accuracy": 0.15284379273653032,
"num_tokens": 11527413.0,
"step": 4545
},
{
"entropy": 6.563087224960327,
"epoch": 0.5251009809578765,
"grad_norm": 0.7890625,
"learning_rate": 0.00049809631028526,
"loss": 6.4284,
"mean_token_accuracy": 0.15203100591897964,
"num_tokens": 11540423.0,
"step": 4550
},
{
"entropy": 6.506260728836059,
"epoch": 0.5256780150028851,
"grad_norm": 0.765625,
"learning_rate": 0.0004980909500930424,
"loss": 6.3276,
"mean_token_accuracy": 0.16709806472063066,
"num_tokens": 11552825.0,
"step": 4555
},
{
"entropy": 6.4506385803222654,
"epoch": 0.5262550490478938,
"grad_norm": 0.82421875,
"learning_rate": 0.0004980855823972717,
"loss": 6.2924,
"mean_token_accuracy": 0.16380308270454408,
"num_tokens": 11565556.0,
"step": 4560
},
{
"entropy": 6.501029014587402,
"epoch": 0.5268320830929025,
"grad_norm": 0.76171875,
"learning_rate": 0.000498080207198128,
"loss": 6.382,
"mean_token_accuracy": 0.16203862726688384,
"num_tokens": 11580056.0,
"step": 4565
},
{
"entropy": 6.510374975204468,
"epoch": 0.5274091171379112,
"grad_norm": 0.7421875,
"learning_rate": 0.0004980748244957925,
"loss": 6.4179,
"mean_token_accuracy": 0.1628774508833885,
"num_tokens": 11592104.0,
"step": 4570
},
{
"entropy": 6.455078506469727,
"epoch": 0.5279861511829198,
"grad_norm": 0.9453125,
"learning_rate": 0.0004980694342904461,
"loss": 6.3118,
"mean_token_accuracy": 0.15631030946969987,
"num_tokens": 11605349.0,
"step": 4575
},
{
"entropy": 6.481748342514038,
"epoch": 0.5285631852279284,
"grad_norm": 0.9375,
"learning_rate": 0.0004980640365822701,
"loss": 6.2951,
"mean_token_accuracy": 0.1727045714855194,
"num_tokens": 11617787.0,
"step": 4580
},
{
"entropy": 6.375321769714356,
"epoch": 0.5291402192729371,
"grad_norm": 1.1953125,
"learning_rate": 0.000498058631371446,
"loss": 6.3024,
"mean_token_accuracy": 0.16541579365730286,
"num_tokens": 11630279.0,
"step": 4585
},
{
"entropy": 6.560987758636474,
"epoch": 0.5297172533179457,
"grad_norm": 0.8203125,
"learning_rate": 0.0004980532186581555,
"loss": 6.4075,
"mean_token_accuracy": 0.16303146183490752,
"num_tokens": 11643378.0,
"step": 4590
},
{
"entropy": 6.474534416198731,
"epoch": 0.5302942873629544,
"grad_norm": 0.83203125,
"learning_rate": 0.000498047798442581,
"loss": 6.3745,
"mean_token_accuracy": 0.16842037290334702,
"num_tokens": 11656472.0,
"step": 4595
},
{
"entropy": 6.472138118743897,
"epoch": 0.5308713214079631,
"grad_norm": 0.83203125,
"learning_rate": 0.0004980423707249044,
"loss": 6.3307,
"mean_token_accuracy": 0.16127256602048873,
"num_tokens": 11669265.0,
"step": 4600
},
{
"entropy": 6.472711849212646,
"epoch": 0.5314483554529718,
"grad_norm": 0.83984375,
"learning_rate": 0.0004980369355053086,
"loss": 6.3725,
"mean_token_accuracy": 0.15782807618379593,
"num_tokens": 11681620.0,
"step": 4605
},
{
"entropy": 6.379305410385132,
"epoch": 0.5320253894979804,
"grad_norm": 0.85546875,
"learning_rate": 0.0004980314927839763,
"loss": 6.2888,
"mean_token_accuracy": 0.16392303705215455,
"num_tokens": 11695720.0,
"step": 4610
},
{
"entropy": 6.473198652267456,
"epoch": 0.532602423542989,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980260425610903,
"loss": 6.3356,
"mean_token_accuracy": 0.15988775640726088,
"num_tokens": 11709563.0,
"step": 4615
},
{
"entropy": 6.553856039047242,
"epoch": 0.5331794575879977,
"grad_norm": 0.85546875,
"learning_rate": 0.0004980205848368343,
"loss": 6.3965,
"mean_token_accuracy": 0.16205898225307463,
"num_tokens": 11722401.0,
"step": 4620
},
{
"entropy": 6.4375251770019535,
"epoch": 0.5337564916330063,
"grad_norm": 0.77734375,
"learning_rate": 0.0004980151196113917,
"loss": 6.4087,
"mean_token_accuracy": 0.16084825843572617,
"num_tokens": 11735183.0,
"step": 4625
},
{
"entropy": 6.566055488586426,
"epoch": 0.534333525678015,
"grad_norm": 0.7890625,
"learning_rate": 0.0004980096468849464,
"loss": 6.3688,
"mean_token_accuracy": 0.15413255393505096,
"num_tokens": 11748282.0,
"step": 4630
},
{
"entropy": 6.47422194480896,
"epoch": 0.5349105597230237,
"grad_norm": 0.88671875,
"learning_rate": 0.0004980041666576823,
"loss": 6.3513,
"mean_token_accuracy": 0.15646012276411056,
"num_tokens": 11760208.0,
"step": 4635
},
{
"entropy": 6.466552066802978,
"epoch": 0.5354875937680323,
"grad_norm": 0.90625,
"learning_rate": 0.0004979986789297837,
"loss": 6.4018,
"mean_token_accuracy": 0.1543569967150688,
"num_tokens": 11772509.0,
"step": 4640
},
{
"entropy": 6.59739408493042,
"epoch": 0.536064627813041,
"grad_norm": 0.85546875,
"learning_rate": 0.0004979931837014355,
"loss": 6.3366,
"mean_token_accuracy": 0.1621736466884613,
"num_tokens": 11786101.0,
"step": 4645
},
{
"entropy": 6.488185453414917,
"epoch": 0.5366416618580496,
"grad_norm": 0.8359375,
"learning_rate": 0.0004979876809728223,
"loss": 6.3939,
"mean_token_accuracy": 0.15879629403352738,
"num_tokens": 11798325.0,
"step": 4650
},
{
"entropy": 6.516392993927002,
"epoch": 0.5372186959030583,
"grad_norm": 0.76953125,
"learning_rate": 0.0004979821707441292,
"loss": 6.3919,
"mean_token_accuracy": 0.15318617820739747,
"num_tokens": 11812905.0,
"step": 4655
},
{
"entropy": 6.4433094501495365,
"epoch": 0.5377957299480669,
"grad_norm": 0.859375,
"learning_rate": 0.0004979766530155416,
"loss": 6.2937,
"mean_token_accuracy": 0.15997868925333023,
"num_tokens": 11826495.0,
"step": 4660
},
{
"entropy": 6.516284656524658,
"epoch": 0.5383727639930755,
"grad_norm": 0.84375,
"learning_rate": 0.0004979711277872449,
"loss": 6.4204,
"mean_token_accuracy": 0.14770983308553695,
"num_tokens": 11840472.0,
"step": 4665
},
{
"entropy": 6.557969570159912,
"epoch": 0.5389497980380843,
"grad_norm": 0.9375,
"learning_rate": 0.0004979655950594252,
"loss": 6.3612,
"mean_token_accuracy": 0.1606568932533264,
"num_tokens": 11852800.0,
"step": 4670
},
{
"entropy": 6.572749853134155,
"epoch": 0.5395268320830929,
"grad_norm": 0.80078125,
"learning_rate": 0.0004979600548322684,
"loss": 6.3818,
"mean_token_accuracy": 0.15735995769500732,
"num_tokens": 11866111.0,
"step": 4675
},
{
"entropy": 6.396126937866211,
"epoch": 0.5401038661281016,
"grad_norm": 0.81640625,
"learning_rate": 0.000497954507105961,
"loss": 6.3845,
"mean_token_accuracy": 0.15580861121416092,
"num_tokens": 11878103.0,
"step": 4680
},
{
"entropy": 6.506599283218383,
"epoch": 0.5406809001731102,
"grad_norm": 0.8046875,
"learning_rate": 0.0004979489518806893,
"loss": 6.3564,
"mean_token_accuracy": 0.16247568577528,
"num_tokens": 11890851.0,
"step": 4685
},
{
"entropy": 6.555721282958984,
"epoch": 0.5412579342181189,
"grad_norm": 0.78125,
"learning_rate": 0.0004979433891566405,
"loss": 6.3961,
"mean_token_accuracy": 0.1531717985868454,
"num_tokens": 11903351.0,
"step": 4690
},
{
"entropy": 6.504356765747071,
"epoch": 0.5418349682631275,
"grad_norm": 0.77734375,
"learning_rate": 0.0004979378189340015,
"loss": 6.434,
"mean_token_accuracy": 0.1565772294998169,
"num_tokens": 11915239.0,
"step": 4695
},
{
"entropy": 6.47059440612793,
"epoch": 0.5424120023081361,
"grad_norm": 1.125,
"learning_rate": 0.0004979322412129597,
"loss": 6.307,
"mean_token_accuracy": 0.16283725053071976,
"num_tokens": 11928678.0,
"step": 4700
},
{
"entropy": 6.532925081253052,
"epoch": 0.5429890363531449,
"grad_norm": 0.9296875,
"learning_rate": 0.0004979266559937028,
"loss": 6.3443,
"mean_token_accuracy": 0.1592644363641739,
"num_tokens": 11940547.0,
"step": 4705
},
{
"entropy": 6.495752763748169,
"epoch": 0.5435660703981535,
"grad_norm": 0.859375,
"learning_rate": 0.0004979210632764185,
"loss": 6.3841,
"mean_token_accuracy": 0.15600561648607253,
"num_tokens": 11953069.0,
"step": 4710
},
{
"entropy": 6.4508391380310055,
"epoch": 0.5441431044431622,
"grad_norm": 0.8203125,
"learning_rate": 0.0004979154630612949,
"loss": 6.3361,
"mean_token_accuracy": 0.16159862279891968,
"num_tokens": 11964301.0,
"step": 4715
},
{
"entropy": 6.479263782501221,
"epoch": 0.5447201384881708,
"grad_norm": 0.8515625,
"learning_rate": 0.0004979098553485205,
"loss": 6.3493,
"mean_token_accuracy": 0.15931008011102676,
"num_tokens": 11976311.0,
"step": 4720
},
{
"entropy": 6.455762338638306,
"epoch": 0.5452971725331794,
"grad_norm": 0.76953125,
"learning_rate": 0.0004979042401382838,
"loss": 6.3316,
"mean_token_accuracy": 0.1660098060965538,
"num_tokens": 11989749.0,
"step": 4725
},
{
"entropy": 6.427380084991455,
"epoch": 0.5458742065781881,
"grad_norm": 0.79296875,
"learning_rate": 0.0004978986174307737,
"loss": 6.2826,
"mean_token_accuracy": 0.16364272236824035,
"num_tokens": 12002384.0,
"step": 4730
},
{
"entropy": 6.565292167663574,
"epoch": 0.5464512406231967,
"grad_norm": 0.85546875,
"learning_rate": 0.0004978929872261794,
"loss": 6.3748,
"mean_token_accuracy": 0.1662522479891777,
"num_tokens": 12014573.0,
"step": 4735
},
{
"entropy": 6.344450426101685,
"epoch": 0.5470282746682055,
"grad_norm": 0.8984375,
"learning_rate": 0.0004978873495246901,
"loss": 6.2243,
"mean_token_accuracy": 0.16644109189510345,
"num_tokens": 12027340.0,
"step": 4740
},
{
"entropy": 6.545307683944702,
"epoch": 0.5476053087132141,
"grad_norm": 0.84765625,
"learning_rate": 0.0004978817043264955,
"loss": 6.3823,
"mean_token_accuracy": 0.159617817401886,
"num_tokens": 12039498.0,
"step": 4745
},
{
"entropy": 6.530856990814209,
"epoch": 0.5481823427582228,
"grad_norm": 0.86328125,
"learning_rate": 0.0004978760516317856,
"loss": 6.3984,
"mean_token_accuracy": 0.15844097435474397,
"num_tokens": 12051426.0,
"step": 4750
},
{
"entropy": 6.496254634857178,
"epoch": 0.5487593768032314,
"grad_norm": 0.875,
"learning_rate": 0.0004978703914407503,
"loss": 6.3698,
"mean_token_accuracy": 0.15838246494531633,
"num_tokens": 12063975.0,
"step": 4755
},
{
"entropy": 6.514954042434693,
"epoch": 0.54933641084824,
"grad_norm": 0.86328125,
"learning_rate": 0.0004978647237535802,
"loss": 6.3347,
"mean_token_accuracy": 0.16198110431432725,
"num_tokens": 12075159.0,
"step": 4760
},
{
"entropy": 6.464260864257812,
"epoch": 0.5499134448932487,
"grad_norm": 0.7890625,
"learning_rate": 0.0004978590485704657,
"loss": 6.331,
"mean_token_accuracy": 0.16103683412075043,
"num_tokens": 12087052.0,
"step": 4765
},
{
"entropy": 6.4438018798828125,
"epoch": 0.5504904789382573,
"grad_norm": 0.87890625,
"learning_rate": 0.0004978533658915979,
"loss": 6.4177,
"mean_token_accuracy": 0.15159978345036507,
"num_tokens": 12100021.0,
"step": 4770
},
{
"entropy": 6.500536775588989,
"epoch": 0.5510675129832661,
"grad_norm": 0.80859375,
"learning_rate": 0.0004978476757171678,
"loss": 6.3866,
"mean_token_accuracy": 0.1626763090491295,
"num_tokens": 12113067.0,
"step": 4775
},
{
"entropy": 6.49054799079895,
"epoch": 0.5516445470282747,
"grad_norm": 0.84375,
"learning_rate": 0.0004978419780473668,
"loss": 6.3858,
"mean_token_accuracy": 0.16142747104167937,
"num_tokens": 12126108.0,
"step": 4780
},
{
"entropy": 6.436980056762695,
"epoch": 0.5522215810732833,
"grad_norm": 0.79296875,
"learning_rate": 0.0004978362728823865,
"loss": 6.2368,
"mean_token_accuracy": 0.16867295056581497,
"num_tokens": 12138821.0,
"step": 4785
},
{
"entropy": 6.561331415176392,
"epoch": 0.552798615118292,
"grad_norm": 0.8203125,
"learning_rate": 0.0004978305602224189,
"loss": 6.4602,
"mean_token_accuracy": 0.15451819002628325,
"num_tokens": 12151621.0,
"step": 4790
},
{
"entropy": 6.565451955795288,
"epoch": 0.5533756491633006,
"grad_norm": 0.91015625,
"learning_rate": 0.0004978248400676562,
"loss": 6.4049,
"mean_token_accuracy": 0.15947678834199905,
"num_tokens": 12163232.0,
"step": 4795
},
{
"entropy": 6.437511682510376,
"epoch": 0.5539526832083093,
"grad_norm": 0.82421875,
"learning_rate": 0.0004978191124182905,
"loss": 6.3774,
"mean_token_accuracy": 0.15531412959098817,
"num_tokens": 12176857.0,
"step": 4800
},
{
"entropy": 6.3844664096832275,
"epoch": 0.5545297172533179,
"grad_norm": 0.74609375,
"learning_rate": 0.0004978133772745149,
"loss": 6.2797,
"mean_token_accuracy": 0.16717115342617034,
"num_tokens": 12190311.0,
"step": 4805
},
{
"entropy": 6.522913789749145,
"epoch": 0.5551067512983267,
"grad_norm": 0.890625,
"learning_rate": 0.0004978076346365218,
"loss": 6.3607,
"mean_token_accuracy": 0.15984816551208497,
"num_tokens": 12202490.0,
"step": 4810
},
{
"entropy": 6.519549751281739,
"epoch": 0.5556837853433353,
"grad_norm": 0.765625,
"learning_rate": 0.0004978018845045047,
"loss": 6.3681,
"mean_token_accuracy": 0.1594349056482315,
"num_tokens": 12214759.0,
"step": 4815
},
{
"entropy": 6.502993392944336,
"epoch": 0.5562608193883439,
"grad_norm": 0.7890625,
"learning_rate": 0.0004977961268786568,
"loss": 6.3037,
"mean_token_accuracy": 0.1682477280497551,
"num_tokens": 12227639.0,
"step": 4820
},
{
"entropy": 6.485041999816895,
"epoch": 0.5568378534333526,
"grad_norm": 0.84375,
"learning_rate": 0.0004977903617591719,
"loss": 6.3981,
"mean_token_accuracy": 0.15737620294094085,
"num_tokens": 12241049.0,
"step": 4825
},
{
"entropy": 6.483181476593018,
"epoch": 0.5574148874783612,
"grad_norm": 0.859375,
"learning_rate": 0.0004977845891462439,
"loss": 6.2891,
"mean_token_accuracy": 0.16137616634368895,
"num_tokens": 12253980.0,
"step": 4830
},
{
"entropy": 6.440955638885498,
"epoch": 0.5579919215233698,
"grad_norm": 0.86328125,
"learning_rate": 0.0004977788090400668,
"loss": 6.3626,
"mean_token_accuracy": 0.16536644846200943,
"num_tokens": 12265103.0,
"step": 4835
},
{
"entropy": 6.425109720230102,
"epoch": 0.5585689555683785,
"grad_norm": 0.79296875,
"learning_rate": 0.0004977730214408352,
"loss": 6.3493,
"mean_token_accuracy": 0.1596547245979309,
"num_tokens": 12278574.0,
"step": 4840
},
{
"entropy": 6.427152872085571,
"epoch": 0.5591459896133872,
"grad_norm": 0.8359375,
"learning_rate": 0.0004977672263487435,
"loss": 6.3435,
"mean_token_accuracy": 0.16322456002235414,
"num_tokens": 12291038.0,
"step": 4845
},
{
"entropy": 6.527709913253784,
"epoch": 0.5597230236583959,
"grad_norm": 0.80859375,
"learning_rate": 0.000497761423763987,
"loss": 6.4031,
"mean_token_accuracy": 0.1564134478569031,
"num_tokens": 12304339.0,
"step": 4850
},
{
"entropy": 6.523615217208862,
"epoch": 0.5603000577034045,
"grad_norm": 0.81640625,
"learning_rate": 0.0004977556136867606,
"loss": 6.4104,
"mean_token_accuracy": 0.15531465411186218,
"num_tokens": 12317927.0,
"step": 4855
},
{
"entropy": 6.5276378154754635,
"epoch": 0.5608770917484132,
"grad_norm": 0.75390625,
"learning_rate": 0.0004977497961172598,
"loss": 6.3349,
"mean_token_accuracy": 0.15690107643604279,
"num_tokens": 12332904.0,
"step": 4860
},
{
"entropy": 6.502278852462768,
"epoch": 0.5614541257934218,
"grad_norm": 0.76953125,
"learning_rate": 0.0004977439710556802,
"loss": 6.2659,
"mean_token_accuracy": 0.16384697556495667,
"num_tokens": 12345449.0,
"step": 4865
},
{
"entropy": 6.3931052684783936,
"epoch": 0.5620311598384304,
"grad_norm": 0.85546875,
"learning_rate": 0.0004977381385022179,
"loss": 6.319,
"mean_token_accuracy": 0.1633118733763695,
"num_tokens": 12357334.0,
"step": 4870
},
{
"entropy": 6.561559820175171,
"epoch": 0.5626081938834391,
"grad_norm": 0.80859375,
"learning_rate": 0.0004977322984570688,
"loss": 6.3612,
"mean_token_accuracy": 0.15845298767089844,
"num_tokens": 12370094.0,
"step": 4875
},
{
"entropy": 6.466009855270386,
"epoch": 0.5631852279284478,
"grad_norm": 0.9453125,
"learning_rate": 0.0004977264509204296,
"loss": 6.3963,
"mean_token_accuracy": 0.15249388813972473,
"num_tokens": 12382875.0,
"step": 4880
},
{
"entropy": 6.431724596023559,
"epoch": 0.5637622619734565,
"grad_norm": 0.796875,
"learning_rate": 0.0004977205958924967,
"loss": 6.3243,
"mean_token_accuracy": 0.15520370602607728,
"num_tokens": 12395534.0,
"step": 4885
},
{
"entropy": 6.492003107070923,
"epoch": 0.5643392960184651,
"grad_norm": 0.8203125,
"learning_rate": 0.0004977147333734673,
"loss": 6.3458,
"mean_token_accuracy": 0.15953391343355178,
"num_tokens": 12407300.0,
"step": 4890
},
{
"entropy": 6.409242057800293,
"epoch": 0.5649163300634737,
"grad_norm": 0.79296875,
"learning_rate": 0.0004977088633635385,
"loss": 6.2632,
"mean_token_accuracy": 0.1659919276833534,
"num_tokens": 12419499.0,
"step": 4895
},
{
"entropy": 6.421889972686768,
"epoch": 0.5654933641084824,
"grad_norm": 0.91796875,
"learning_rate": 0.0004977029858629076,
"loss": 6.2121,
"mean_token_accuracy": 0.1685824289917946,
"num_tokens": 12431279.0,
"step": 4900
},
{
"entropy": 6.295782613754272,
"epoch": 0.566070398153491,
"grad_norm": 0.8046875,
"learning_rate": 0.0004976971008717726,
"loss": 6.1719,
"mean_token_accuracy": 0.17155274003744125,
"num_tokens": 12443976.0,
"step": 4905
},
{
"entropy": 6.393910932540893,
"epoch": 0.5666474321984997,
"grad_norm": 0.80859375,
"learning_rate": 0.0004976912083903309,
"loss": 6.3299,
"mean_token_accuracy": 0.16340630799531936,
"num_tokens": 12457752.0,
"step": 4910
},
{
"entropy": 6.492015361785889,
"epoch": 0.5672244662435084,
"grad_norm": 0.8359375,
"learning_rate": 0.0004976853084187814,
"loss": 6.34,
"mean_token_accuracy": 0.15709138363599778,
"num_tokens": 12469776.0,
"step": 4915
},
{
"entropy": 6.52558970451355,
"epoch": 0.5678015002885171,
"grad_norm": 0.8359375,
"learning_rate": 0.0004976794009573219,
"loss": 6.3669,
"mean_token_accuracy": 0.14939526617527008,
"num_tokens": 12483300.0,
"step": 4920
},
{
"entropy": 6.38107852935791,
"epoch": 0.5683785343335257,
"grad_norm": 0.86328125,
"learning_rate": 0.0004976734860061515,
"loss": 6.3129,
"mean_token_accuracy": 0.16156516522169112,
"num_tokens": 12494564.0,
"step": 4925
},
{
"entropy": 6.423398447036743,
"epoch": 0.5689555683785343,
"grad_norm": 0.80859375,
"learning_rate": 0.0004976675635654688,
"loss": 6.3394,
"mean_token_accuracy": 0.16027214378118515,
"num_tokens": 12507761.0,
"step": 4930
},
{
"entropy": 6.51806640625,
"epoch": 0.569532602423543,
"grad_norm": 0.84375,
"learning_rate": 0.0004976616336354733,
"loss": 6.331,
"mean_token_accuracy": 0.16270025670528412,
"num_tokens": 12520045.0,
"step": 4935
},
{
"entropy": 6.4369653224945065,
"epoch": 0.5701096364685516,
"grad_norm": 1.1484375,
"learning_rate": 0.0004976556962163645,
"loss": 6.2544,
"mean_token_accuracy": 0.16926467120647432,
"num_tokens": 12532471.0,
"step": 4940
},
{
"entropy": 6.468795585632324,
"epoch": 0.5706866705135603,
"grad_norm": 0.8515625,
"learning_rate": 0.0004976497513083419,
"loss": 6.3643,
"mean_token_accuracy": 0.15875103324651718,
"num_tokens": 12544610.0,
"step": 4945
},
{
"entropy": 6.472221326828003,
"epoch": 0.571263704558569,
"grad_norm": 0.875,
"learning_rate": 0.0004976437989116055,
"loss": 6.3786,
"mean_token_accuracy": 0.16317480951547622,
"num_tokens": 12558682.0,
"step": 4950
},
{
"entropy": 6.466315221786499,
"epoch": 0.5718407386035776,
"grad_norm": 0.87890625,
"learning_rate": 0.0004976378390263554,
"loss": 6.2759,
"mean_token_accuracy": 0.15960074067115784,
"num_tokens": 12570711.0,
"step": 4955
},
{
"entropy": 6.410310411453247,
"epoch": 0.5724177726485863,
"grad_norm": 0.87109375,
"learning_rate": 0.0004976318716527924,
"loss": 6.3023,
"mean_token_accuracy": 0.16249395459890364,
"num_tokens": 12582468.0,
"step": 4960
},
{
"entropy": 6.483807325363159,
"epoch": 0.5729948066935949,
"grad_norm": 0.84375,
"learning_rate": 0.0004976258967911168,
"loss": 6.3381,
"mean_token_accuracy": 0.16272847652435302,
"num_tokens": 12594991.0,
"step": 4965
},
{
"entropy": 6.498748826980591,
"epoch": 0.5735718407386036,
"grad_norm": 0.859375,
"learning_rate": 0.0004976199144415298,
"loss": 6.3175,
"mean_token_accuracy": 0.16240322440862656,
"num_tokens": 12606928.0,
"step": 4970
},
{
"entropy": 6.455127906799317,
"epoch": 0.5741488747836122,
"grad_norm": 0.77734375,
"learning_rate": 0.0004976139246042325,
"loss": 6.4012,
"mean_token_accuracy": 0.15524941086769103,
"num_tokens": 12620539.0,
"step": 4975
},
{
"entropy": 6.4773979663848875,
"epoch": 0.5747259088286208,
"grad_norm": 0.765625,
"learning_rate": 0.0004976079272794265,
"loss": 6.3446,
"mean_token_accuracy": 0.15420007407665254,
"num_tokens": 12632629.0,
"step": 4980
},
{
"entropy": 6.43410210609436,
"epoch": 0.5753029428736296,
"grad_norm": 0.8984375,
"learning_rate": 0.0004976019224673134,
"loss": 6.3069,
"mean_token_accuracy": 0.16904855519533157,
"num_tokens": 12646631.0,
"step": 4985
},
{
"entropy": 6.449543762207031,
"epoch": 0.5758799769186382,
"grad_norm": 0.84765625,
"learning_rate": 0.0004975959101680953,
"loss": 6.3019,
"mean_token_accuracy": 0.16361594051122666,
"num_tokens": 12659154.0,
"step": 4990
},
{
"entropy": 6.447082471847534,
"epoch": 0.5764570109636469,
"grad_norm": 0.84375,
"learning_rate": 0.0004975898903819742,
"loss": 6.3907,
"mean_token_accuracy": 0.1598036080598831,
"num_tokens": 12671796.0,
"step": 4995
},
{
"entropy": 6.5214070796966555,
"epoch": 0.5770340450086555,
"grad_norm": 0.875,
"learning_rate": 0.0004975838631091527,
"loss": 6.3528,
"mean_token_accuracy": 0.15806207805871964,
"num_tokens": 12684432.0,
"step": 5000
},
{
"entropy": 6.488364028930664,
"epoch": 0.5776110790536642,
"grad_norm": 0.859375,
"learning_rate": 0.0004975778283498336,
"loss": 6.3571,
"mean_token_accuracy": 0.16473528295755385,
"num_tokens": 12697585.0,
"step": 5005
},
{
"entropy": 6.400859832763672,
"epoch": 0.5781881130986728,
"grad_norm": 0.83984375,
"learning_rate": 0.0004975717861042198,
"loss": 6.2783,
"mean_token_accuracy": 0.1685378298163414,
"num_tokens": 12710654.0,
"step": 5010
},
{
"entropy": 6.401386976242065,
"epoch": 0.5787651471436814,
"grad_norm": 0.84765625,
"learning_rate": 0.0004975657363725146,
"loss": 6.3226,
"mean_token_accuracy": 0.15781749784946442,
"num_tokens": 12723141.0,
"step": 5015
},
{
"entropy": 6.534639596939087,
"epoch": 0.5793421811886902,
"grad_norm": 0.78515625,
"learning_rate": 0.0004975596791549213,
"loss": 6.4103,
"mean_token_accuracy": 0.15141381472349166,
"num_tokens": 12735921.0,
"step": 5020
},
{
"entropy": 6.395477533340454,
"epoch": 0.5799192152336988,
"grad_norm": 0.83203125,
"learning_rate": 0.0004975536144516437,
"loss": 6.3753,
"mean_token_accuracy": 0.15583303123712539,
"num_tokens": 12747704.0,
"step": 5025
},
{
"entropy": 6.563265943527222,
"epoch": 0.5804962492787075,
"grad_norm": 1.109375,
"learning_rate": 0.000497547542262886,
"loss": 6.3668,
"mean_token_accuracy": 0.15204933732748033,
"num_tokens": 12759368.0,
"step": 5030
},
{
"entropy": 6.520508766174316,
"epoch": 0.5810732833237161,
"grad_norm": 0.84765625,
"learning_rate": 0.0004975414625888521,
"loss": 6.3235,
"mean_token_accuracy": 0.16076601594686507,
"num_tokens": 12772226.0,
"step": 5035
},
{
"entropy": 6.515491724014282,
"epoch": 0.5816503173687247,
"grad_norm": 0.78515625,
"learning_rate": 0.0004975353754297468,
"loss": 6.3922,
"mean_token_accuracy": 0.15815138816833496,
"num_tokens": 12784817.0,
"step": 5040
},
{
"entropy": 6.4733335971832275,
"epoch": 0.5822273514137334,
"grad_norm": 0.859375,
"learning_rate": 0.0004975292807857745,
"loss": 6.3285,
"mean_token_accuracy": 0.16127829551696776,
"num_tokens": 12796996.0,
"step": 5045
},
{
"entropy": 6.462998390197754,
"epoch": 0.582804385458742,
"grad_norm": 0.88671875,
"learning_rate": 0.0004975231786571406,
"loss": 6.3803,
"mean_token_accuracy": 0.15985668301582337,
"num_tokens": 12809581.0,
"step": 5050
},
{
"entropy": 6.427072811126709,
"epoch": 0.5833814195037508,
"grad_norm": 0.80078125,
"learning_rate": 0.0004975170690440499,
"loss": 6.3413,
"mean_token_accuracy": 0.1596603661775589,
"num_tokens": 12822060.0,
"step": 5055
},
{
"entropy": 6.490553283691407,
"epoch": 0.5839584535487594,
"grad_norm": 0.8046875,
"learning_rate": 0.0004975109519467083,
"loss": 6.3249,
"mean_token_accuracy": 0.1605900213122368,
"num_tokens": 12835554.0,
"step": 5060
},
{
"entropy": 6.465603160858154,
"epoch": 0.584535487593768,
"grad_norm": 0.7890625,
"learning_rate": 0.0004975048273653212,
"loss": 6.2566,
"mean_token_accuracy": 0.17011079341173171,
"num_tokens": 12848187.0,
"step": 5065
},
{
"entropy": 6.394630336761475,
"epoch": 0.5851125216387767,
"grad_norm": 0.89453125,
"learning_rate": 0.0004974986953000948,
"loss": 6.2778,
"mean_token_accuracy": 0.16856665015220643,
"num_tokens": 12861498.0,
"step": 5070
},
{
"entropy": 6.557682037353516,
"epoch": 0.5856895556837853,
"grad_norm": 0.84765625,
"learning_rate": 0.0004974925557512354,
"loss": 6.459,
"mean_token_accuracy": 0.1463148832321167,
"num_tokens": 12874906.0,
"step": 5075
},
{
"entropy": 6.386854028701782,
"epoch": 0.586266589728794,
"grad_norm": 0.87109375,
"learning_rate": 0.0004974864087189493,
"loss": 6.2865,
"mean_token_accuracy": 0.16323204934597016,
"num_tokens": 12887169.0,
"step": 5080
},
{
"entropy": 6.466772794723511,
"epoch": 0.5868436237738026,
"grad_norm": 0.8359375,
"learning_rate": 0.0004974802542034434,
"loss": 6.3339,
"mean_token_accuracy": 0.16101205796003343,
"num_tokens": 12899639.0,
"step": 5085
},
{
"entropy": 6.465685510635376,
"epoch": 0.5874206578188114,
"grad_norm": 0.79296875,
"learning_rate": 0.0004974740922049246,
"loss": 6.2465,
"mean_token_accuracy": 0.16204580515623093,
"num_tokens": 12911913.0,
"step": 5090
},
{
"entropy": 6.532586860656738,
"epoch": 0.58799769186382,
"grad_norm": 0.74609375,
"learning_rate": 0.0004974679227236004,
"loss": 6.4113,
"mean_token_accuracy": 0.15404328852891921,
"num_tokens": 12924200.0,
"step": 5095
},
{
"entropy": 6.4691802024841305,
"epoch": 0.5885747259088286,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974617457596779,
"loss": 6.3971,
"mean_token_accuracy": 0.15657893866300582,
"num_tokens": 12936726.0,
"step": 5100
},
{
"entropy": 6.508554601669312,
"epoch": 0.5891517599538373,
"grad_norm": 0.8828125,
"learning_rate": 0.0004974555613133652,
"loss": 6.2811,
"mean_token_accuracy": 0.159566231071949,
"num_tokens": 12949751.0,
"step": 5105
},
{
"entropy": 6.475770330429077,
"epoch": 0.5897287939988459,
"grad_norm": 0.8203125,
"learning_rate": 0.0004974493693848702,
"loss": 6.3737,
"mean_token_accuracy": 0.1601461336016655,
"num_tokens": 12962573.0,
"step": 5110
},
{
"entropy": 6.506417083740234,
"epoch": 0.5903058280438546,
"grad_norm": 0.8828125,
"learning_rate": 0.0004974431699744011,
"loss": 6.4022,
"mean_token_accuracy": 0.15874024480581284,
"num_tokens": 12975635.0,
"step": 5115
},
{
"entropy": 6.464160919189453,
"epoch": 0.5908828620888632,
"grad_norm": 0.85546875,
"learning_rate": 0.0004974369630821665,
"loss": 6.3972,
"mean_token_accuracy": 0.15452799648046495,
"num_tokens": 12987822.0,
"step": 5120
},
{
"entropy": 6.50497260093689,
"epoch": 0.5914598961338718,
"grad_norm": 0.8828125,
"learning_rate": 0.0004974307487083752,
"loss": 6.347,
"mean_token_accuracy": 0.16242809891700744,
"num_tokens": 13000082.0,
"step": 5125
},
{
"entropy": 6.421209764480591,
"epoch": 0.5920369301788806,
"grad_norm": 0.8671875,
"learning_rate": 0.0004974245268532361,
"loss": 6.2443,
"mean_token_accuracy": 0.16429829150438308,
"num_tokens": 13011902.0,
"step": 5130
},
{
"entropy": 6.432503414154053,
"epoch": 0.5926139642238892,
"grad_norm": 0.75390625,
"learning_rate": 0.0004974182975169585,
"loss": 6.2979,
"mean_token_accuracy": 0.16235667765140532,
"num_tokens": 13024507.0,
"step": 5135
},
{
"entropy": 6.4553704261779785,
"epoch": 0.5931909982688979,
"grad_norm": 0.8125,
"learning_rate": 0.000497412060699752,
"loss": 6.4016,
"mean_token_accuracy": 0.16003545448184014,
"num_tokens": 13037955.0,
"step": 5140
},
{
"entropy": 6.525091123580933,
"epoch": 0.5937680323139065,
"grad_norm": 1.015625,
"learning_rate": 0.0004974058164018263,
"loss": 6.351,
"mean_token_accuracy": 0.1640540525317192,
"num_tokens": 13050631.0,
"step": 5145
},
{
"entropy": 6.477291822433472,
"epoch": 0.5943450663589152,
"grad_norm": 0.84375,
"learning_rate": 0.0004973995646233914,
"loss": 6.2996,
"mean_token_accuracy": 0.16100106835365297,
"num_tokens": 13062566.0,
"step": 5150
},
{
"entropy": 6.407656812667847,
"epoch": 0.5949221004039238,
"grad_norm": 0.8046875,
"learning_rate": 0.0004973933053646576,
"loss": 6.2747,
"mean_token_accuracy": 0.16398902982473373,
"num_tokens": 13075454.0,
"step": 5155
},
{
"entropy": 6.498917055130005,
"epoch": 0.5954991344489324,
"grad_norm": 0.8515625,
"learning_rate": 0.0004973870386258355,
"loss": 6.3468,
"mean_token_accuracy": 0.15738814622163771,
"num_tokens": 13087780.0,
"step": 5160
},
{
"entropy": 6.477192354202271,
"epoch": 0.5960761684939412,
"grad_norm": 0.765625,
"learning_rate": 0.0004973807644071357,
"loss": 6.3504,
"mean_token_accuracy": 0.15569237619638443,
"num_tokens": 13100957.0,
"step": 5165
},
{
"entropy": 6.460265016555786,
"epoch": 0.5966532025389498,
"grad_norm": 0.75390625,
"learning_rate": 0.0004973744827087695,
"loss": 6.2701,
"mean_token_accuracy": 0.16348017752170563,
"num_tokens": 13113232.0,
"step": 5170
},
{
"entropy": 6.407651424407959,
"epoch": 0.5972302365839585,
"grad_norm": 0.84375,
"learning_rate": 0.000497368193530948,
"loss": 6.2438,
"mean_token_accuracy": 0.16760450899600982,
"num_tokens": 13125950.0,
"step": 5175
},
{
"entropy": 6.380605173110962,
"epoch": 0.5978072706289671,
"grad_norm": 0.7734375,
"learning_rate": 0.0004973618968738828,
"loss": 6.2429,
"mean_token_accuracy": 0.16398492008447646,
"num_tokens": 13138936.0,
"step": 5180
},
{
"entropy": 6.498394584655761,
"epoch": 0.5983843046739757,
"grad_norm": 0.71875,
"learning_rate": 0.0004973555927377856,
"loss": 6.3801,
"mean_token_accuracy": 0.1577525481581688,
"num_tokens": 13152787.0,
"step": 5185
},
{
"entropy": 6.453382253646851,
"epoch": 0.5989613387189844,
"grad_norm": 0.78125,
"learning_rate": 0.0004973492811228685,
"loss": 6.3077,
"mean_token_accuracy": 0.162015600502491,
"num_tokens": 13165459.0,
"step": 5190
},
{
"entropy": 6.52237606048584,
"epoch": 0.599538372763993,
"grad_norm": 0.8203125,
"learning_rate": 0.0004973429620293438,
"loss": 6.3203,
"mean_token_accuracy": 0.1562004953622818,
"num_tokens": 13178766.0,
"step": 5195
},
{
"entropy": 6.502490377426147,
"epoch": 0.6001154068090018,
"grad_norm": 0.8046875,
"learning_rate": 0.0004973366354574239,
"loss": 6.4284,
"mean_token_accuracy": 0.1566441461443901,
"num_tokens": 13193712.0,
"step": 5200
},
{
"entropy": 6.505635213851929,
"epoch": 0.6006924408540104,
"grad_norm": 0.80078125,
"learning_rate": 0.0004973303014073219,
"loss": 6.4283,
"mean_token_accuracy": 0.1619483232498169,
"num_tokens": 13207363.0,
"step": 5205
},
{
"entropy": 6.441206932067871,
"epoch": 0.601269474899019,
"grad_norm": 0.890625,
"learning_rate": 0.0004973239598792504,
"loss": 6.2197,
"mean_token_accuracy": 0.16894505620002748,
"num_tokens": 13219131.0,
"step": 5210
},
{
"entropy": 6.462107229232788,
"epoch": 0.6018465089440277,
"grad_norm": 0.86328125,
"learning_rate": 0.0004973176108734232,
"loss": 6.3294,
"mean_token_accuracy": 0.1623179242014885,
"num_tokens": 13231595.0,
"step": 5215
},
{
"entropy": 6.4451995372772215,
"epoch": 0.6024235429890363,
"grad_norm": 0.80859375,
"learning_rate": 0.0004973112543900535,
"loss": 6.3424,
"mean_token_accuracy": 0.16196154356002807,
"num_tokens": 13243727.0,
"step": 5220
},
{
"entropy": 6.453670310974121,
"epoch": 0.603000577034045,
"grad_norm": 0.8046875,
"learning_rate": 0.0004973048904293551,
"loss": 6.3048,
"mean_token_accuracy": 0.16469427496194838,
"num_tokens": 13255002.0,
"step": 5225
},
{
"entropy": 6.488593482971192,
"epoch": 0.6035776110790536,
"grad_norm": 0.8046875,
"learning_rate": 0.0004972985189915422,
"loss": 6.3417,
"mean_token_accuracy": 0.16113921850919724,
"num_tokens": 13268523.0,
"step": 5230
},
{
"entropy": 6.368151140213013,
"epoch": 0.6041546451240624,
"grad_norm": 0.83203125,
"learning_rate": 0.000497292140076829,
"loss": 6.2508,
"mean_token_accuracy": 0.1658511444926262,
"num_tokens": 13280625.0,
"step": 5235
},
{
"entropy": 6.551406383514404,
"epoch": 0.604731679169071,
"grad_norm": 0.77734375,
"learning_rate": 0.0004972857536854301,
"loss": 6.4091,
"mean_token_accuracy": 0.15450685024261473,
"num_tokens": 13294880.0,
"step": 5240
},
{
"entropy": 6.51069598197937,
"epoch": 0.6053087132140796,
"grad_norm": 0.8515625,
"learning_rate": 0.0004972793598175603,
"loss": 6.3424,
"mean_token_accuracy": 0.15950118750333786,
"num_tokens": 13307875.0,
"step": 5245
},
{
"entropy": 6.417842197418213,
"epoch": 0.6058857472590883,
"grad_norm": 0.80859375,
"learning_rate": 0.0004972729584734347,
"loss": 6.2173,
"mean_token_accuracy": 0.16876524686813354,
"num_tokens": 13320067.0,
"step": 5250
},
{
"entropy": 6.375647735595703,
"epoch": 0.6064627813040969,
"grad_norm": 0.828125,
"learning_rate": 0.0004972665496532685,
"loss": 6.2833,
"mean_token_accuracy": 0.16125652939081192,
"num_tokens": 13332036.0,
"step": 5255
},
{
"entropy": 6.473684883117675,
"epoch": 0.6070398153491056,
"grad_norm": 0.828125,
"learning_rate": 0.0004972601333572774,
"loss": 6.3059,
"mean_token_accuracy": 0.1631132811307907,
"num_tokens": 13343831.0,
"step": 5260
},
{
"entropy": 6.508532238006592,
"epoch": 0.6076168493941142,
"grad_norm": 0.8515625,
"learning_rate": 0.0004972537095856769,
"loss": 6.3611,
"mean_token_accuracy": 0.15429210513830185,
"num_tokens": 13356452.0,
"step": 5265
},
{
"entropy": 6.461349248886108,
"epoch": 0.608193883439123,
"grad_norm": 0.9140625,
"learning_rate": 0.0004972472783386834,
"loss": 6.2744,
"mean_token_accuracy": 0.1614803835749626,
"num_tokens": 13369052.0,
"step": 5270
},
{
"entropy": 6.479144048690796,
"epoch": 0.6087709174841316,
"grad_norm": 0.83984375,
"learning_rate": 0.0004972408396165132,
"loss": 6.3656,
"mean_token_accuracy": 0.15780351161956788,
"num_tokens": 13381943.0,
"step": 5275
},
{
"entropy": 6.455678510665893,
"epoch": 0.6093479515291402,
"grad_norm": 0.828125,
"learning_rate": 0.0004972343934193826,
"loss": 6.3247,
"mean_token_accuracy": 0.15625445544719696,
"num_tokens": 13394069.0,
"step": 5280
},
{
"entropy": 6.525052213668824,
"epoch": 0.6099249855741489,
"grad_norm": 0.76953125,
"learning_rate": 0.0004972279397475086,
"loss": 6.3845,
"mean_token_accuracy": 0.1577905982732773,
"num_tokens": 13407397.0,
"step": 5285
},
{
"entropy": 6.425345516204834,
"epoch": 0.6105020196191575,
"grad_norm": 0.8046875,
"learning_rate": 0.0004972214786011083,
"loss": 6.3092,
"mean_token_accuracy": 0.1553454354405403,
"num_tokens": 13419716.0,
"step": 5290
},
{
"entropy": 6.471757507324218,
"epoch": 0.6110790536641661,
"grad_norm": 0.8359375,
"learning_rate": 0.000497215009980399,
"loss": 6.2705,
"mean_token_accuracy": 0.15723506659269332,
"num_tokens": 13432839.0,
"step": 5295
},
{
"entropy": 6.493448495864868,
"epoch": 0.6116560877091748,
"grad_norm": 0.7734375,
"learning_rate": 0.000497208533885598,
"loss": 6.314,
"mean_token_accuracy": 0.15734073221683503,
"num_tokens": 13445318.0,
"step": 5300
},
{
"entropy": 6.421839809417724,
"epoch": 0.6122331217541835,
"grad_norm": 0.8671875,
"learning_rate": 0.0004972020503169235,
"loss": 6.2926,
"mean_token_accuracy": 0.1610089898109436,
"num_tokens": 13456652.0,
"step": 5305
},
{
"entropy": 6.419455242156983,
"epoch": 0.6128101557991922,
"grad_norm": 0.796875,
"learning_rate": 0.0004971955592745934,
"loss": 6.2935,
"mean_token_accuracy": 0.15970533713698387,
"num_tokens": 13469133.0,
"step": 5310
},
{
"entropy": 6.470340442657471,
"epoch": 0.6133871898442008,
"grad_norm": 1.1015625,
"learning_rate": 0.000497189060758826,
"loss": 6.3026,
"mean_token_accuracy": 0.16227268874645234,
"num_tokens": 13481741.0,
"step": 5315
},
{
"entropy": 6.43679633140564,
"epoch": 0.6139642238892095,
"grad_norm": 0.984375,
"learning_rate": 0.00049718255476984,
"loss": 6.3954,
"mean_token_accuracy": 0.15243706703186036,
"num_tokens": 13494257.0,
"step": 5320
},
{
"entropy": 6.415068626403809,
"epoch": 0.6145412579342181,
"grad_norm": 0.83203125,
"learning_rate": 0.0004971760413078539,
"loss": 6.2579,
"mean_token_accuracy": 0.1738950192928314,
"num_tokens": 13507714.0,
"step": 5325
},
{
"entropy": 6.428360319137573,
"epoch": 0.6151182919792267,
"grad_norm": 0.80078125,
"learning_rate": 0.0004971695203730872,
"loss": 6.3064,
"mean_token_accuracy": 0.1591736286878586,
"num_tokens": 13519499.0,
"step": 5330
},
{
"entropy": 6.4912420272827145,
"epoch": 0.6156953260242354,
"grad_norm": 0.80078125,
"learning_rate": 0.000497162991965759,
"loss": 6.2825,
"mean_token_accuracy": 0.16679684966802596,
"num_tokens": 13531538.0,
"step": 5335
},
{
"entropy": 6.553953790664673,
"epoch": 0.6162723600692441,
"grad_norm": 0.78515625,
"learning_rate": 0.0004971564560860889,
"loss": 6.352,
"mean_token_accuracy": 0.15729986131191254,
"num_tokens": 13544576.0,
"step": 5340
},
{
"entropy": 6.415745735168457,
"epoch": 0.6168493941142528,
"grad_norm": 0.8359375,
"learning_rate": 0.0004971499127342968,
"loss": 6.297,
"mean_token_accuracy": 0.15716785192489624,
"num_tokens": 13557318.0,
"step": 5345
},
{
"entropy": 6.352963447570801,
"epoch": 0.6174264281592614,
"grad_norm": 1.2265625,
"learning_rate": 0.0004971433619106027,
"loss": 6.1886,
"mean_token_accuracy": 0.16966686099767686,
"num_tokens": 13569334.0,
"step": 5350
},
{
"entropy": 6.437016868591309,
"epoch": 0.61800346220427,
"grad_norm": 0.875,
"learning_rate": 0.000497136803615227,
"loss": 6.2662,
"mean_token_accuracy": 0.1659908726811409,
"num_tokens": 13581748.0,
"step": 5355
},
{
"entropy": 6.47385802268982,
"epoch": 0.6185804962492787,
"grad_norm": 0.8828125,
"learning_rate": 0.0004971302378483902,
"loss": 6.3469,
"mean_token_accuracy": 0.16016255766153337,
"num_tokens": 13593892.0,
"step": 5360
},
{
"entropy": 6.445749044418335,
"epoch": 0.6191575302942873,
"grad_norm": 0.84765625,
"learning_rate": 0.0004971236646103132,
"loss": 6.2198,
"mean_token_accuracy": 0.16264959871768953,
"num_tokens": 13605593.0,
"step": 5365
},
{
"entropy": 6.426254367828369,
"epoch": 0.619734564339296,
"grad_norm": 0.8828125,
"learning_rate": 0.0004971170839012171,
"loss": 6.402,
"mean_token_accuracy": 0.1573261246085167,
"num_tokens": 13618356.0,
"step": 5370
},
{
"entropy": 6.485814094543457,
"epoch": 0.6203115983843047,
"grad_norm": 0.87109375,
"learning_rate": 0.0004971104957213233,
"loss": 6.3457,
"mean_token_accuracy": 0.1604474663734436,
"num_tokens": 13629939.0,
"step": 5375
},
{
"entropy": 6.3256737232208256,
"epoch": 0.6208886324293134,
"grad_norm": 0.90234375,
"learning_rate": 0.0004971039000708531,
"loss": 6.1998,
"mean_token_accuracy": 0.17142249196767806,
"num_tokens": 13642123.0,
"step": 5380
},
{
"entropy": 6.41780834197998,
"epoch": 0.621465666474322,
"grad_norm": 0.91015625,
"learning_rate": 0.0004970972969500286,
"loss": 6.2908,
"mean_token_accuracy": 0.16614297032356262,
"num_tokens": 13654956.0,
"step": 5385
},
{
"entropy": 6.4260025978088375,
"epoch": 0.6220427005193306,
"grad_norm": 0.94921875,
"learning_rate": 0.000497090686359072,
"loss": 6.2563,
"mean_token_accuracy": 0.16691885590553285,
"num_tokens": 13667549.0,
"step": 5390
},
{
"entropy": 6.461333417892456,
"epoch": 0.6226197345643393,
"grad_norm": 0.828125,
"learning_rate": 0.0004970840682982054,
"loss": 6.3172,
"mean_token_accuracy": 0.16463381201028823,
"num_tokens": 13680747.0,
"step": 5395
},
{
"entropy": 6.487573194503784,
"epoch": 0.6231967686093479,
"grad_norm": 0.77734375,
"learning_rate": 0.0004970774427676514,
"loss": 6.3576,
"mean_token_accuracy": 0.15117861852049827,
"num_tokens": 13695500.0,
"step": 5400
},
{
"entropy": 6.485534191131592,
"epoch": 0.6237738026543566,
"grad_norm": 0.8359375,
"learning_rate": 0.000497070809767633,
"loss": 6.327,
"mean_token_accuracy": 0.15981363877654076,
"num_tokens": 13707807.0,
"step": 5405
},
{
"entropy": 6.407062005996704,
"epoch": 0.6243508366993653,
"grad_norm": 0.92578125,
"learning_rate": 0.0004970641692983731,
"loss": 6.2726,
"mean_token_accuracy": 0.16532813906669616,
"num_tokens": 13720401.0,
"step": 5410
},
{
"entropy": 6.414212942123413,
"epoch": 0.624927870744374,
"grad_norm": 0.796875,
"learning_rate": 0.0004970575213600954,
"loss": 6.2596,
"mean_token_accuracy": 0.16263928413391113,
"num_tokens": 13732611.0,
"step": 5415
},
{
"entropy": 6.413683795928955,
"epoch": 0.6255049047893826,
"grad_norm": 0.76953125,
"learning_rate": 0.0004970508659530231,
"loss": 6.3135,
"mean_token_accuracy": 0.1561568185687065,
"num_tokens": 13744876.0,
"step": 5420
},
{
"entropy": 6.420420742034912,
"epoch": 0.6260819388343912,
"grad_norm": 0.8828125,
"learning_rate": 0.0004970442030773802,
"loss": 6.3382,
"mean_token_accuracy": 0.16061879694461823,
"num_tokens": 13757265.0,
"step": 5425
},
{
"entropy": 6.492178726196289,
"epoch": 0.6266589728793999,
"grad_norm": 0.90234375,
"learning_rate": 0.0004970375327333909,
"loss": 6.2707,
"mean_token_accuracy": 0.16342684477567673,
"num_tokens": 13770101.0,
"step": 5430
},
{
"entropy": 6.485668134689331,
"epoch": 0.6272360069244085,
"grad_norm": 0.84375,
"learning_rate": 0.0004970308549212796,
"loss": 6.3115,
"mean_token_accuracy": 0.16133440434932708,
"num_tokens": 13784104.0,
"step": 5435
},
{
"entropy": 6.386432552337647,
"epoch": 0.6278130409694171,
"grad_norm": 0.83203125,
"learning_rate": 0.0004970241696412705,
"loss": 6.2359,
"mean_token_accuracy": 0.16974283754825592,
"num_tokens": 13796901.0,
"step": 5440
},
{
"entropy": 6.43957667350769,
"epoch": 0.6283900750144259,
"grad_norm": 0.8671875,
"learning_rate": 0.000497017476893589,
"loss": 6.3112,
"mean_token_accuracy": 0.1625409170985222,
"num_tokens": 13809640.0,
"step": 5445
},
{
"entropy": 6.53232970237732,
"epoch": 0.6289671090594345,
"grad_norm": 0.84375,
"learning_rate": 0.0004970107766784598,
"loss": 6.3116,
"mean_token_accuracy": 0.16599140763282777,
"num_tokens": 13821684.0,
"step": 5450
},
{
"entropy": 6.448949337005615,
"epoch": 0.6295441431044432,
"grad_norm": 0.8203125,
"learning_rate": 0.0004970040689961084,
"loss": 6.3363,
"mean_token_accuracy": 0.1553775981068611,
"num_tokens": 13833973.0,
"step": 5455
},
{
"entropy": 6.375447082519531,
"epoch": 0.6301211771494518,
"grad_norm": 0.80078125,
"learning_rate": 0.0004969973538467605,
"loss": 6.2622,
"mean_token_accuracy": 0.1685728073120117,
"num_tokens": 13845954.0,
"step": 5460
},
{
"entropy": 6.517121410369873,
"epoch": 0.6306982111944605,
"grad_norm": 0.78125,
"learning_rate": 0.0004969906312306419,
"loss": 6.3,
"mean_token_accuracy": 0.15561674758791924,
"num_tokens": 13858261.0,
"step": 5465
},
{
"entropy": 6.478764772415161,
"epoch": 0.6312752452394691,
"grad_norm": 0.79296875,
"learning_rate": 0.0004969839011479786,
"loss": 6.2791,
"mean_token_accuracy": 0.16333419382572173,
"num_tokens": 13870409.0,
"step": 5470
},
{
"entropy": 6.430654668807984,
"epoch": 0.6318522792844777,
"grad_norm": 0.78125,
"learning_rate": 0.000496977163598997,
"loss": 6.3614,
"mean_token_accuracy": 0.15769426077604293,
"num_tokens": 13883980.0,
"step": 5475
},
{
"entropy": 6.476246643066406,
"epoch": 0.6324293133294865,
"grad_norm": 0.80078125,
"learning_rate": 0.0004969704185839239,
"loss": 6.3736,
"mean_token_accuracy": 0.16204104721546173,
"num_tokens": 13897901.0,
"step": 5480
},
{
"entropy": 6.482294368743896,
"epoch": 0.6330063473744951,
"grad_norm": 0.82421875,
"learning_rate": 0.0004969636661029859,
"loss": 6.3509,
"mean_token_accuracy": 0.1583631828427315,
"num_tokens": 13909655.0,
"step": 5485
},
{
"entropy": 6.494528293609619,
"epoch": 0.6335833814195038,
"grad_norm": 0.94140625,
"learning_rate": 0.0004969569061564103,
"loss": 6.3003,
"mean_token_accuracy": 0.16088411957025528,
"num_tokens": 13923121.0,
"step": 5490
},
{
"entropy": 6.520718622207641,
"epoch": 0.6341604154645124,
"grad_norm": 0.81640625,
"learning_rate": 0.0004969501387444245,
"loss": 6.3925,
"mean_token_accuracy": 0.1511763960123062,
"num_tokens": 13936100.0,
"step": 5495
},
{
"entropy": 6.4032745361328125,
"epoch": 0.634737449509521,
"grad_norm": 0.82421875,
"learning_rate": 0.0004969433638672559,
"loss": 6.2821,
"mean_token_accuracy": 0.15816063433885574,
"num_tokens": 13949799.0,
"step": 5500
},
{
"entropy": 6.450932168960572,
"epoch": 0.6353144835545297,
"grad_norm": 0.859375,
"learning_rate": 0.0004969365815251325,
"loss": 6.267,
"mean_token_accuracy": 0.16805639117956161,
"num_tokens": 13961805.0,
"step": 5505
},
{
"entropy": 6.401394653320312,
"epoch": 0.6358915175995383,
"grad_norm": 0.82421875,
"learning_rate": 0.0004969297917182825,
"loss": 6.3204,
"mean_token_accuracy": 0.1655355915427208,
"num_tokens": 13974264.0,
"step": 5510
},
{
"entropy": 6.427269029617309,
"epoch": 0.6364685516445471,
"grad_norm": 0.77734375,
"learning_rate": 0.0004969229944469342,
"loss": 6.2694,
"mean_token_accuracy": 0.16473935395479203,
"num_tokens": 13986582.0,
"step": 5515
},
{
"entropy": 6.460540533065796,
"epoch": 0.6370455856895557,
"grad_norm": 0.82421875,
"learning_rate": 0.0004969161897113162,
"loss": 6.2857,
"mean_token_accuracy": 0.157159423828125,
"num_tokens": 13998675.0,
"step": 5520
},
{
"entropy": 6.466981744766235,
"epoch": 0.6376226197345644,
"grad_norm": 0.828125,
"learning_rate": 0.0004969093775116574,
"loss": 6.3225,
"mean_token_accuracy": 0.1617126539349556,
"num_tokens": 14010744.0,
"step": 5525
},
{
"entropy": 6.524676513671875,
"epoch": 0.638199653779573,
"grad_norm": 0.75,
"learning_rate": 0.0004969025578481869,
"loss": 6.3096,
"mean_token_accuracy": 0.1576560452580452,
"num_tokens": 14023721.0,
"step": 5530
},
{
"entropy": 6.431024599075317,
"epoch": 0.6387766878245816,
"grad_norm": 0.79296875,
"learning_rate": 0.000496895730721134,
"loss": 6.2184,
"mean_token_accuracy": 0.16521818935871124,
"num_tokens": 14035998.0,
"step": 5535
},
{
"entropy": 6.338489437103272,
"epoch": 0.6393537218695903,
"grad_norm": 0.8984375,
"learning_rate": 0.0004968888961307286,
"loss": 6.2027,
"mean_token_accuracy": 0.16932371407747268,
"num_tokens": 14048407.0,
"step": 5540
},
{
"entropy": 6.3543487071990965,
"epoch": 0.6399307559145989,
"grad_norm": 0.765625,
"learning_rate": 0.0004968820540772003,
"loss": 6.1883,
"mean_token_accuracy": 0.1674530401825905,
"num_tokens": 14060898.0,
"step": 5545
},
{
"entropy": 6.428665018081665,
"epoch": 0.6405077899596077,
"grad_norm": 0.8984375,
"learning_rate": 0.0004968752045607794,
"loss": 6.2746,
"mean_token_accuracy": 0.16548453867435456,
"num_tokens": 14073474.0,
"step": 5550
},
{
"entropy": 6.425108814239502,
"epoch": 0.6410848240046163,
"grad_norm": 0.8359375,
"learning_rate": 0.0004968683475816961,
"loss": 6.353,
"mean_token_accuracy": 0.15872218310832978,
"num_tokens": 14085229.0,
"step": 5555
},
{
"entropy": 6.432684373855591,
"epoch": 0.641661858049625,
"grad_norm": 0.86328125,
"learning_rate": 0.0004968614831401811,
"loss": 6.2776,
"mean_token_accuracy": 0.16353354454040528,
"num_tokens": 14097943.0,
"step": 5560
},
{
"entropy": 6.543388414382934,
"epoch": 0.6422388920946336,
"grad_norm": 0.86328125,
"learning_rate": 0.0004968546112364654,
"loss": 6.4176,
"mean_token_accuracy": 0.1529288560152054,
"num_tokens": 14112397.0,
"step": 5565
},
{
"entropy": 6.441380167007447,
"epoch": 0.6428159261396422,
"grad_norm": 0.83203125,
"learning_rate": 0.00049684773187078,
"loss": 6.2476,
"mean_token_accuracy": 0.16730485260486602,
"num_tokens": 14127096.0,
"step": 5570
},
{
"entropy": 6.497290849685669,
"epoch": 0.6433929601846509,
"grad_norm": 0.82421875,
"learning_rate": 0.0004968408450433565,
"loss": 6.3284,
"mean_token_accuracy": 0.15766475945711136,
"num_tokens": 14139325.0,
"step": 5575
},
{
"entropy": 6.50930609703064,
"epoch": 0.6439699942296595,
"grad_norm": 0.80859375,
"learning_rate": 0.0004968339507544263,
"loss": 6.3056,
"mean_token_accuracy": 0.15905881971120833,
"num_tokens": 14151334.0,
"step": 5580
},
{
"entropy": 6.486502265930175,
"epoch": 0.6445470282746683,
"grad_norm": 0.88671875,
"learning_rate": 0.0004968270490042212,
"loss": 6.2151,
"mean_token_accuracy": 0.17320423871278762,
"num_tokens": 14164195.0,
"step": 5585
},
{
"entropy": 6.368127059936524,
"epoch": 0.6451240623196769,
"grad_norm": 0.84375,
"learning_rate": 0.0004968201397929737,
"loss": 6.2717,
"mean_token_accuracy": 0.16237038969993592,
"num_tokens": 14175937.0,
"step": 5590
},
{
"entropy": 6.4224189758300785,
"epoch": 0.6457010963646855,
"grad_norm": 0.7890625,
"learning_rate": 0.0004968132231209158,
"loss": 6.2844,
"mean_token_accuracy": 0.16292650997638702,
"num_tokens": 14189288.0,
"step": 5595
},
{
"entropy": 6.4685217380523685,
"epoch": 0.6462781304096942,
"grad_norm": 0.828125,
"learning_rate": 0.0004968062989882803,
"loss": 6.2794,
"mean_token_accuracy": 0.16588214188814163,
"num_tokens": 14202428.0,
"step": 5600
},
{
"entropy": 6.327192735671997,
"epoch": 0.6468551644547028,
"grad_norm": 0.73828125,
"learning_rate": 0.0004967993673953003,
"loss": 6.273,
"mean_token_accuracy": 0.16544562429189683,
"num_tokens": 14216871.0,
"step": 5605
},
{
"entropy": 6.401887845993042,
"epoch": 0.6474321984997115,
"grad_norm": 0.81640625,
"learning_rate": 0.0004967924283422087,
"loss": 6.2498,
"mean_token_accuracy": 0.16043491661548615,
"num_tokens": 14228503.0,
"step": 5610
},
{
"entropy": 6.42364354133606,
"epoch": 0.6480092325447201,
"grad_norm": 0.83203125,
"learning_rate": 0.000496785481829239,
"loss": 6.1793,
"mean_token_accuracy": 0.16777475029230118,
"num_tokens": 14240239.0,
"step": 5615
},
{
"entropy": 6.395428085327149,
"epoch": 0.6485862665897288,
"grad_norm": 0.828125,
"learning_rate": 0.0004967785278566245,
"loss": 6.2541,
"mean_token_accuracy": 0.16892644315958022,
"num_tokens": 14253180.0,
"step": 5620
},
{
"entropy": 6.3999152183532715,
"epoch": 0.6491633006347375,
"grad_norm": 0.7109375,
"learning_rate": 0.0004967715664245997,
"loss": 6.3116,
"mean_token_accuracy": 0.15934029072523118,
"num_tokens": 14267275.0,
"step": 5625
},
{
"entropy": 6.455025959014892,
"epoch": 0.6497403346797461,
"grad_norm": 0.88671875,
"learning_rate": 0.0004967645975333983,
"loss": 6.3232,
"mean_token_accuracy": 0.1586598202586174,
"num_tokens": 14279650.0,
"step": 5630
},
{
"entropy": 6.418586301803589,
"epoch": 0.6503173687247548,
"grad_norm": 0.89453125,
"learning_rate": 0.0004967576211832548,
"loss": 6.2187,
"mean_token_accuracy": 0.16348557323217391,
"num_tokens": 14291647.0,
"step": 5635
},
{
"entropy": 6.325366878509522,
"epoch": 0.6508944027697634,
"grad_norm": 0.83984375,
"learning_rate": 0.0004967506373744039,
"loss": 6.2198,
"mean_token_accuracy": 0.16452054381370546,
"num_tokens": 14304335.0,
"step": 5640
},
{
"entropy": 6.430365800857544,
"epoch": 0.651471436814772,
"grad_norm": 0.8046875,
"learning_rate": 0.0004967436461070805,
"loss": 6.2481,
"mean_token_accuracy": 0.16254803836345671,
"num_tokens": 14317226.0,
"step": 5645
},
{
"entropy": 6.434140729904175,
"epoch": 0.6520484708597807,
"grad_norm": 0.8125,
"learning_rate": 0.0004967366473815196,
"loss": 6.2797,
"mean_token_accuracy": 0.1666984811425209,
"num_tokens": 14329000.0,
"step": 5650
},
{
"entropy": 6.386842870712281,
"epoch": 0.6526255049047894,
"grad_norm": 0.8125,
"learning_rate": 0.0004967296411979568,
"loss": 6.2577,
"mean_token_accuracy": 0.16278322041034698,
"num_tokens": 14341714.0,
"step": 5655
},
{
"entropy": 6.412582921981811,
"epoch": 0.6532025389497981,
"grad_norm": 0.8046875,
"learning_rate": 0.0004967226275566275,
"loss": 6.2462,
"mean_token_accuracy": 0.15906601399183273,
"num_tokens": 14353889.0,
"step": 5660
},
{
"entropy": 6.535210132598877,
"epoch": 0.6537795729948067,
"grad_norm": 0.98046875,
"learning_rate": 0.000496715606457768,
"loss": 6.3295,
"mean_token_accuracy": 0.1594246506690979,
"num_tokens": 14367093.0,
"step": 5665
},
{
"entropy": 6.274228048324585,
"epoch": 0.6543566070398154,
"grad_norm": 0.8671875,
"learning_rate": 0.0004967085779016142,
"loss": 6.0987,
"mean_token_accuracy": 0.17560223937034608,
"num_tokens": 14380064.0,
"step": 5670
},
{
"entropy": 6.444216394424439,
"epoch": 0.654933641084824,
"grad_norm": 0.8359375,
"learning_rate": 0.0004967015418884022,
"loss": 6.2893,
"mean_token_accuracy": 0.16281547397375107,
"num_tokens": 14392026.0,
"step": 5675
},
{
"entropy": 6.419415473937988,
"epoch": 0.6555106751298326,
"grad_norm": 0.8515625,
"learning_rate": 0.0004966944984183692,
"loss": 6.2463,
"mean_token_accuracy": 0.15920784771442414,
"num_tokens": 14404011.0,
"step": 5680
},
{
"entropy": 6.416391611099243,
"epoch": 0.6560877091748413,
"grad_norm": 0.86328125,
"learning_rate": 0.0004966874474917518,
"loss": 6.2416,
"mean_token_accuracy": 0.16538350880146027,
"num_tokens": 14416356.0,
"step": 5685
},
{
"entropy": 6.384862327575684,
"epoch": 0.65666474321985,
"grad_norm": 0.80859375,
"learning_rate": 0.0004966803891087873,
"loss": 6.2679,
"mean_token_accuracy": 0.15895518362522126,
"num_tokens": 14428873.0,
"step": 5690
},
{
"entropy": 6.373016309738159,
"epoch": 0.6572417772648587,
"grad_norm": 0.76953125,
"learning_rate": 0.000496673323269713,
"loss": 6.2474,
"mean_token_accuracy": 0.1652483820915222,
"num_tokens": 14441703.0,
"step": 5695
},
{
"entropy": 6.473770570755005,
"epoch": 0.6578188113098673,
"grad_norm": 0.9765625,
"learning_rate": 0.0004966662499747666,
"loss": 6.2524,
"mean_token_accuracy": 0.16153307110071183,
"num_tokens": 14453701.0,
"step": 5700
},
{
"entropy": 6.456718921661377,
"epoch": 0.6583958453548759,
"grad_norm": 1.09375,
"learning_rate": 0.0004966591692241859,
"loss": 6.2848,
"mean_token_accuracy": 0.1631843477487564,
"num_tokens": 14466108.0,
"step": 5705
},
{
"entropy": 6.405096006393433,
"epoch": 0.6589728793998846,
"grad_norm": 0.8125,
"learning_rate": 0.0004966520810182092,
"loss": 6.3424,
"mean_token_accuracy": 0.16256508827209473,
"num_tokens": 14479130.0,
"step": 5710
},
{
"entropy": 6.451769876480102,
"epoch": 0.6595499134448932,
"grad_norm": 0.828125,
"learning_rate": 0.0004966449853570749,
"loss": 6.2904,
"mean_token_accuracy": 0.16311692148447038,
"num_tokens": 14492797.0,
"step": 5715
},
{
"entropy": 6.406583213806153,
"epoch": 0.6601269474899019,
"grad_norm": 0.9609375,
"learning_rate": 0.0004966378822410215,
"loss": 6.2139,
"mean_token_accuracy": 0.1620306298136711,
"num_tokens": 14505279.0,
"step": 5720
},
{
"entropy": 6.40832781791687,
"epoch": 0.6607039815349106,
"grad_norm": 0.8671875,
"learning_rate": 0.0004966307716702881,
"loss": 6.3031,
"mean_token_accuracy": 0.16416211426258087,
"num_tokens": 14517867.0,
"step": 5725
},
{
"entropy": 6.400218391418457,
"epoch": 0.6612810155799193,
"grad_norm": 0.91015625,
"learning_rate": 0.0004966236536451138,
"loss": 6.2295,
"mean_token_accuracy": 0.16840277910232543,
"num_tokens": 14529804.0,
"step": 5730
},
{
"entropy": 6.332806348800659,
"epoch": 0.6618580496249279,
"grad_norm": 0.98828125,
"learning_rate": 0.000496616528165738,
"loss": 6.1157,
"mean_token_accuracy": 0.17764069586992265,
"num_tokens": 14541484.0,
"step": 5735
},
{
"entropy": 6.35449628829956,
"epoch": 0.6624350836699365,
"grad_norm": 0.93359375,
"learning_rate": 0.0004966093952324003,
"loss": 6.1861,
"mean_token_accuracy": 0.16744254529476166,
"num_tokens": 14554796.0,
"step": 5740
},
{
"entropy": 6.292174339294434,
"epoch": 0.6630121177149452,
"grad_norm": 0.8984375,
"learning_rate": 0.0004966022548453406,
"loss": 6.1281,
"mean_token_accuracy": 0.18883802443742753,
"num_tokens": 14567090.0,
"step": 5745
},
{
"entropy": 6.370087051391602,
"epoch": 0.6635891517599538,
"grad_norm": 0.8671875,
"learning_rate": 0.0004965951070047993,
"loss": 6.2575,
"mean_token_accuracy": 0.16039278656244277,
"num_tokens": 14580334.0,
"step": 5750
},
{
"entropy": 6.384637451171875,
"epoch": 0.6641661858049625,
"grad_norm": 0.86328125,
"learning_rate": 0.0004965879517110166,
"loss": 6.1818,
"mean_token_accuracy": 0.17177506238222123,
"num_tokens": 14593670.0,
"step": 5755
},
{
"entropy": 6.3392963886260985,
"epoch": 0.6647432198499712,
"grad_norm": 0.80078125,
"learning_rate": 0.0004965807889642333,
"loss": 6.2103,
"mean_token_accuracy": 0.16691275388002397,
"num_tokens": 14606612.0,
"step": 5760
},
{
"entropy": 6.490033864974976,
"epoch": 0.6653202538949798,
"grad_norm": 0.8671875,
"learning_rate": 0.00049657361876469,
"loss": 6.2239,
"mean_token_accuracy": 0.16626789271831513,
"num_tokens": 14618323.0,
"step": 5765
},
{
"entropy": 6.27760009765625,
"epoch": 0.6658972879399885,
"grad_norm": 0.95703125,
"learning_rate": 0.0004965664411126282,
"loss": 6.2311,
"mean_token_accuracy": 0.16586460769176484,
"num_tokens": 14631580.0,
"step": 5770
},
{
"entropy": 6.381122303009033,
"epoch": 0.6664743219849971,
"grad_norm": 0.921875,
"learning_rate": 0.0004965592560082894,
"loss": 6.1559,
"mean_token_accuracy": 0.16871390789747237,
"num_tokens": 14643429.0,
"step": 5775
},
{
"entropy": 6.4877759456634525,
"epoch": 0.6670513560300058,
"grad_norm": 0.89453125,
"learning_rate": 0.0004965520634519149,
"loss": 6.3704,
"mean_token_accuracy": 0.15916894674301146,
"num_tokens": 14656890.0,
"step": 5780
},
{
"entropy": 6.332120084762574,
"epoch": 0.6676283900750144,
"grad_norm": 1.15625,
"learning_rate": 0.0004965448634437468,
"loss": 6.2516,
"mean_token_accuracy": 0.1616871252655983,
"num_tokens": 14670721.0,
"step": 5785
},
{
"entropy": 6.45994815826416,
"epoch": 0.668205424120023,
"grad_norm": 0.86328125,
"learning_rate": 0.0004965376559840272,
"loss": 6.2788,
"mean_token_accuracy": 0.15587088614702224,
"num_tokens": 14682808.0,
"step": 5790
},
{
"entropy": 6.4104235649108885,
"epoch": 0.6687824581650318,
"grad_norm": 0.828125,
"learning_rate": 0.0004965304410729986,
"loss": 6.2592,
"mean_token_accuracy": 0.17194210290908812,
"num_tokens": 14694580.0,
"step": 5795
},
{
"entropy": 6.341176223754883,
"epoch": 0.6693594922100404,
"grad_norm": 0.8046875,
"learning_rate": 0.0004965232187109037,
"loss": 6.2736,
"mean_token_accuracy": 0.1583656281232834,
"num_tokens": 14707832.0,
"step": 5800
},
{
"entropy": 6.460653781890869,
"epoch": 0.6699365262550491,
"grad_norm": 0.8359375,
"learning_rate": 0.0004965159888979854,
"loss": 6.2805,
"mean_token_accuracy": 0.16277113258838655,
"num_tokens": 14719293.0,
"step": 5805
},
{
"entropy": 6.33139591217041,
"epoch": 0.6705135603000577,
"grad_norm": 0.80859375,
"learning_rate": 0.0004965087516344869,
"loss": 6.1755,
"mean_token_accuracy": 0.16717009246349335,
"num_tokens": 14731421.0,
"step": 5810
},
{
"entropy": 6.35786280632019,
"epoch": 0.6710905943450663,
"grad_norm": 0.79296875,
"learning_rate": 0.0004965015069206515,
"loss": 6.2799,
"mean_token_accuracy": 0.16111143082380294,
"num_tokens": 14744160.0,
"step": 5815
},
{
"entropy": 6.518506193161011,
"epoch": 0.671667628390075,
"grad_norm": 0.8984375,
"learning_rate": 0.000496494254756723,
"loss": 6.2824,
"mean_token_accuracy": 0.15849184840917588,
"num_tokens": 14756820.0,
"step": 5820
},
{
"entropy": 6.372296142578125,
"epoch": 0.6722446624350836,
"grad_norm": 0.85546875,
"learning_rate": 0.0004964869951429451,
"loss": 6.2227,
"mean_token_accuracy": 0.1700335294008255,
"num_tokens": 14770000.0,
"step": 5825
},
{
"entropy": 6.415471744537354,
"epoch": 0.6728216964800924,
"grad_norm": 0.91015625,
"learning_rate": 0.0004964797280795622,
"loss": 6.3075,
"mean_token_accuracy": 0.1579108163714409,
"num_tokens": 14782777.0,
"step": 5830
},
{
"entropy": 6.452236700057983,
"epoch": 0.673398730525101,
"grad_norm": 0.88671875,
"learning_rate": 0.0004964724535668188,
"loss": 6.3368,
"mean_token_accuracy": 0.16094402670860292,
"num_tokens": 14796561.0,
"step": 5835
},
{
"entropy": 6.412081909179688,
"epoch": 0.6739757645701097,
"grad_norm": 0.80078125,
"learning_rate": 0.0004964651716049593,
"loss": 6.2962,
"mean_token_accuracy": 0.16613052189350128,
"num_tokens": 14809975.0,
"step": 5840
},
{
"entropy": 6.4701982021331785,
"epoch": 0.6745527986151183,
"grad_norm": 0.8203125,
"learning_rate": 0.0004964578821942288,
"loss": 6.318,
"mean_token_accuracy": 0.15860529243946075,
"num_tokens": 14823578.0,
"step": 5845
},
{
"entropy": 6.476156282424927,
"epoch": 0.6751298326601269,
"grad_norm": 0.7890625,
"learning_rate": 0.0004964505853348724,
"loss": 6.2851,
"mean_token_accuracy": 0.16648429483175278,
"num_tokens": 14836520.0,
"step": 5850
},
{
"entropy": 6.431745862960815,
"epoch": 0.6757068667051356,
"grad_norm": 0.80859375,
"learning_rate": 0.0004964432810271356,
"loss": 6.321,
"mean_token_accuracy": 0.15974137783050538,
"num_tokens": 14847806.0,
"step": 5855
},
{
"entropy": 6.3844091415405275,
"epoch": 0.6762839007501442,
"grad_norm": 0.8515625,
"learning_rate": 0.0004964359692712641,
"loss": 6.2074,
"mean_token_accuracy": 0.16706834882497787,
"num_tokens": 14859751.0,
"step": 5860
},
{
"entropy": 6.481093883514404,
"epoch": 0.676860934795153,
"grad_norm": 0.78125,
"learning_rate": 0.0004964286500675037,
"loss": 6.2351,
"mean_token_accuracy": 0.165215802192688,
"num_tokens": 14872215.0,
"step": 5865
},
{
"entropy": 6.417674732208252,
"epoch": 0.6774379688401616,
"grad_norm": 0.8359375,
"learning_rate": 0.0004964213234161007,
"loss": 6.2849,
"mean_token_accuracy": 0.16408911049365998,
"num_tokens": 14884895.0,
"step": 5870
},
{
"entropy": 6.39830961227417,
"epoch": 0.6780150028851702,
"grad_norm": 0.82421875,
"learning_rate": 0.0004964139893173014,
"loss": 6.2486,
"mean_token_accuracy": 0.16585659235715866,
"num_tokens": 14897131.0,
"step": 5875
},
{
"entropy": 6.350745296478271,
"epoch": 0.6785920369301789,
"grad_norm": 0.84765625,
"learning_rate": 0.0004964066477713525,
"loss": 6.1867,
"mean_token_accuracy": 0.1712081626057625,
"num_tokens": 14908835.0,
"step": 5880
},
{
"entropy": 6.394719171524048,
"epoch": 0.6791690709751875,
"grad_norm": 0.90234375,
"learning_rate": 0.0004963992987785011,
"loss": 6.2861,
"mean_token_accuracy": 0.1636969566345215,
"num_tokens": 14920808.0,
"step": 5885
},
{
"entropy": 6.453951072692871,
"epoch": 0.6797461050201962,
"grad_norm": 0.8828125,
"learning_rate": 0.0004963919423389942,
"loss": 6.1362,
"mean_token_accuracy": 0.17062204480171203,
"num_tokens": 14932558.0,
"step": 5890
},
{
"entropy": 6.310215997695923,
"epoch": 0.6803231390652048,
"grad_norm": 0.8203125,
"learning_rate": 0.0004963845784530794,
"loss": 6.2496,
"mean_token_accuracy": 0.16623902767896653,
"num_tokens": 14946623.0,
"step": 5895
},
{
"entropy": 6.384698295593262,
"epoch": 0.6809001731102136,
"grad_norm": 0.8046875,
"learning_rate": 0.000496377207121004,
"loss": 6.2711,
"mean_token_accuracy": 0.1658199280500412,
"num_tokens": 14959168.0,
"step": 5900
},
{
"entropy": 6.477806615829468,
"epoch": 0.6814772071552222,
"grad_norm": 0.8828125,
"learning_rate": 0.0004963698283430164,
"loss": 6.3015,
"mean_token_accuracy": 0.15701866149902344,
"num_tokens": 14972710.0,
"step": 5905
},
{
"entropy": 6.399756193161011,
"epoch": 0.6820542412002308,
"grad_norm": 0.859375,
"learning_rate": 0.0004963624421193646,
"loss": 6.1501,
"mean_token_accuracy": 0.16826880425214769,
"num_tokens": 14984707.0,
"step": 5910
},
{
"entropy": 6.298456907272339,
"epoch": 0.6826312752452395,
"grad_norm": 0.86328125,
"learning_rate": 0.000496355048450297,
"loss": 6.1769,
"mean_token_accuracy": 0.17006256580352783,
"num_tokens": 14996623.0,
"step": 5915
},
{
"entropy": 6.430842161178589,
"epoch": 0.6832083092902481,
"grad_norm": 0.8671875,
"learning_rate": 0.0004963476473360623,
"loss": 6.2945,
"mean_token_accuracy": 0.1616099148988724,
"num_tokens": 15009579.0,
"step": 5920
},
{
"entropy": 6.403625202178955,
"epoch": 0.6837853433352568,
"grad_norm": 0.875,
"learning_rate": 0.0004963402387769094,
"loss": 6.2285,
"mean_token_accuracy": 0.16827364563941954,
"num_tokens": 15022239.0,
"step": 5925
},
{
"entropy": 6.363240528106689,
"epoch": 0.6843623773802654,
"grad_norm": 0.81640625,
"learning_rate": 0.0004963328227730876,
"loss": 6.2556,
"mean_token_accuracy": 0.16110678166151046,
"num_tokens": 15034724.0,
"step": 5930
},
{
"entropy": 6.31289849281311,
"epoch": 0.684939411425274,
"grad_norm": 0.80859375,
"learning_rate": 0.0004963253993248461,
"loss": 6.1712,
"mean_token_accuracy": 0.16620973199605943,
"num_tokens": 15047534.0,
"step": 5935
},
{
"entropy": 6.4068629264831545,
"epoch": 0.6855164454702828,
"grad_norm": 0.875,
"learning_rate": 0.0004963179684324349,
"loss": 6.2699,
"mean_token_accuracy": 0.16617678850889206,
"num_tokens": 15060468.0,
"step": 5940
},
{
"entropy": 6.435975074768066,
"epoch": 0.6860934795152914,
"grad_norm": 0.890625,
"learning_rate": 0.0004963105300961036,
"loss": 6.2452,
"mean_token_accuracy": 0.1608870580792427,
"num_tokens": 15073424.0,
"step": 5945
},
{
"entropy": 6.380932950973511,
"epoch": 0.6866705135603001,
"grad_norm": 0.90625,
"learning_rate": 0.0004963030843161026,
"loss": 6.287,
"mean_token_accuracy": 0.15832511186599732,
"num_tokens": 15086582.0,
"step": 5950
},
{
"entropy": 6.479296588897705,
"epoch": 0.6872475476053087,
"grad_norm": 0.83203125,
"learning_rate": 0.0004962956310926823,
"loss": 6.2874,
"mean_token_accuracy": 0.1592211276292801,
"num_tokens": 15099288.0,
"step": 5955
},
{
"entropy": 6.414581060409546,
"epoch": 0.6878245816503173,
"grad_norm": 0.78125,
"learning_rate": 0.0004962881704260934,
"loss": 6.2051,
"mean_token_accuracy": 0.16755102276802064,
"num_tokens": 15111800.0,
"step": 5960
},
{
"entropy": 6.396193742752075,
"epoch": 0.688401615695326,
"grad_norm": 0.96875,
"learning_rate": 0.0004962807023165868,
"loss": 6.2563,
"mean_token_accuracy": 0.16351549625396727,
"num_tokens": 15123384.0,
"step": 5965
},
{
"entropy": 6.36266450881958,
"epoch": 0.6889786497403346,
"grad_norm": 0.85546875,
"learning_rate": 0.0004962732267644138,
"loss": 6.1621,
"mean_token_accuracy": 0.1664055049419403,
"num_tokens": 15134857.0,
"step": 5970
},
{
"entropy": 6.395646381378174,
"epoch": 0.6895556837853434,
"grad_norm": 0.76171875,
"learning_rate": 0.0004962657437698254,
"loss": 6.3036,
"mean_token_accuracy": 0.16814867109060289,
"num_tokens": 15149124.0,
"step": 5975
},
{
"entropy": 6.244737005233764,
"epoch": 0.690132717830352,
"grad_norm": 0.88671875,
"learning_rate": 0.0004962582533330739,
"loss": 6.0647,
"mean_token_accuracy": 0.18276433199644088,
"num_tokens": 15163315.0,
"step": 5980
},
{
"entropy": 6.433995866775513,
"epoch": 0.6907097518753607,
"grad_norm": 0.84765625,
"learning_rate": 0.0004962507554544109,
"loss": 6.2554,
"mean_token_accuracy": 0.16413442492485047,
"num_tokens": 15175371.0,
"step": 5985
},
{
"entropy": 6.423586463928222,
"epoch": 0.6912867859203693,
"grad_norm": 0.83203125,
"learning_rate": 0.0004962432501340886,
"loss": 6.2297,
"mean_token_accuracy": 0.16294970810413362,
"num_tokens": 15187633.0,
"step": 5990
},
{
"entropy": 6.39416732788086,
"epoch": 0.6918638199653779,
"grad_norm": 0.828125,
"learning_rate": 0.0004962357373723596,
"loss": 6.2701,
"mean_token_accuracy": 0.1605140045285225,
"num_tokens": 15199652.0,
"step": 5995
},
{
"entropy": 6.471811056137085,
"epoch": 0.6924408540103866,
"grad_norm": 0.75,
"learning_rate": 0.0004962282171694763,
"loss": 6.2885,
"mean_token_accuracy": 0.16474244892597198,
"num_tokens": 15213309.0,
"step": 6000
},
{
"epoch": 0.6924408540103866,
"eval_entropy": 6.305234524279722,
"eval_loss": 6.288375377655029,
"eval_mean_token_accuracy": 0.166228380397424,
"eval_num_tokens": 15213309.0,
"eval_runtime": 17.8169,
"eval_samples_per_second": 1579.178,
"eval_steps_per_second": 197.397,
"step": 6000
},
{
"entropy": 6.407313680648803,
"epoch": 0.6930178880553952,
"grad_norm": 0.8203125,
"learning_rate": 0.0004962206895256919,
"loss": 6.2349,
"mean_token_accuracy": 0.16205479800701142,
"num_tokens": 15226466.0,
"step": 6005
},
{
"entropy": 6.275579786300659,
"epoch": 0.693594922100404,
"grad_norm": 0.80859375,
"learning_rate": 0.0004962131544412595,
"loss": 6.1267,
"mean_token_accuracy": 0.17538843154907227,
"num_tokens": 15239306.0,
"step": 6010
},
{
"entropy": 6.371297407150268,
"epoch": 0.6941719561454126,
"grad_norm": 0.84765625,
"learning_rate": 0.0004962056119164325,
"loss": 6.2535,
"mean_token_accuracy": 0.16581086218357086,
"num_tokens": 15252365.0,
"step": 6015
},
{
"entropy": 6.46108570098877,
"epoch": 0.6947489901904212,
"grad_norm": 0.8359375,
"learning_rate": 0.0004961980619514646,
"loss": 6.2761,
"mean_token_accuracy": 0.16486027836799622,
"num_tokens": 15265770.0,
"step": 6020
},
{
"entropy": 6.398010969161987,
"epoch": 0.6953260242354299,
"grad_norm": 0.82421875,
"learning_rate": 0.0004961905045466098,
"loss": 6.2754,
"mean_token_accuracy": 0.16328403055667878,
"num_tokens": 15278766.0,
"step": 6025
},
{
"entropy": 6.503208446502685,
"epoch": 0.6959030582804385,
"grad_norm": 0.81640625,
"learning_rate": 0.0004961829397021222,
"loss": 6.2858,
"mean_token_accuracy": 0.15995303988456727,
"num_tokens": 15292180.0,
"step": 6030
},
{
"entropy": 6.451317501068115,
"epoch": 0.6964800923254472,
"grad_norm": 0.82421875,
"learning_rate": 0.0004961753674182564,
"loss": 6.3007,
"mean_token_accuracy": 0.15992565751075744,
"num_tokens": 15305313.0,
"step": 6035
},
{
"entropy": 6.455318593978882,
"epoch": 0.6970571263704558,
"grad_norm": 0.80078125,
"learning_rate": 0.0004961677876952669,
"loss": 6.3497,
"mean_token_accuracy": 0.15649957954883575,
"num_tokens": 15318734.0,
"step": 6040
},
{
"entropy": 6.367619514465332,
"epoch": 0.6976341604154646,
"grad_norm": 0.86328125,
"learning_rate": 0.0004961602005334087,
"loss": 6.2162,
"mean_token_accuracy": 0.1630083441734314,
"num_tokens": 15330915.0,
"step": 6045
},
{
"entropy": 6.420135116577148,
"epoch": 0.6982111944604732,
"grad_norm": 0.82421875,
"learning_rate": 0.000496152605932937,
"loss": 6.3083,
"mean_token_accuracy": 0.15993753224611282,
"num_tokens": 15343634.0,
"step": 6050
},
{
"entropy": 6.443638706207276,
"epoch": 0.6987882285054818,
"grad_norm": 0.80859375,
"learning_rate": 0.0004961450038941074,
"loss": 6.2466,
"mean_token_accuracy": 0.16260004937648773,
"num_tokens": 15356049.0,
"step": 6055
},
{
"entropy": 6.369221115112305,
"epoch": 0.6993652625504905,
"grad_norm": 0.82421875,
"learning_rate": 0.0004961373944171754,
"loss": 6.1004,
"mean_token_accuracy": 0.1690909057855606,
"num_tokens": 15367804.0,
"step": 6060
},
{
"entropy": 6.348468399047851,
"epoch": 0.6999422965954991,
"grad_norm": 0.83984375,
"learning_rate": 0.0004961297775023968,
"loss": 6.2341,
"mean_token_accuracy": 0.1663442924618721,
"num_tokens": 15379921.0,
"step": 6065
},
{
"entropy": 6.332316446304321,
"epoch": 0.7005193306405078,
"grad_norm": 0.7578125,
"learning_rate": 0.000496122153150028,
"loss": 6.2063,
"mean_token_accuracy": 0.1757096141576767,
"num_tokens": 15393043.0,
"step": 6070
},
{
"entropy": 6.452652359008789,
"epoch": 0.7010963646855164,
"grad_norm": 0.796875,
"learning_rate": 0.0004961145213603255,
"loss": 6.2926,
"mean_token_accuracy": 0.1563486099243164,
"num_tokens": 15407577.0,
"step": 6075
},
{
"entropy": 6.411776113510132,
"epoch": 0.7016733987305251,
"grad_norm": 0.83984375,
"learning_rate": 0.000496106882133546,
"loss": 6.2942,
"mean_token_accuracy": 0.16173667162656785,
"num_tokens": 15420157.0,
"step": 6080
},
{
"entropy": 6.3121175289154055,
"epoch": 0.7022504327755338,
"grad_norm": 0.875,
"learning_rate": 0.0004960992354699463,
"loss": 6.2521,
"mean_token_accuracy": 0.1679789364337921,
"num_tokens": 15432882.0,
"step": 6085
},
{
"entropy": 6.40244779586792,
"epoch": 0.7028274668205424,
"grad_norm": 0.81640625,
"learning_rate": 0.0004960915813697836,
"loss": 6.1713,
"mean_token_accuracy": 0.16617501974105836,
"num_tokens": 15444541.0,
"step": 6090
},
{
"entropy": 6.37771053314209,
"epoch": 0.7034045008655511,
"grad_norm": 0.86328125,
"learning_rate": 0.0004960839198333154,
"loss": 6.3018,
"mean_token_accuracy": 0.16228880733251572,
"num_tokens": 15457104.0,
"step": 6095
},
{
"entropy": 6.42156720161438,
"epoch": 0.7039815349105597,
"grad_norm": 0.8671875,
"learning_rate": 0.0004960762508607994,
"loss": 6.2456,
"mean_token_accuracy": 0.16384343653917313,
"num_tokens": 15469724.0,
"step": 6100
},
{
"entropy": 6.416327810287475,
"epoch": 0.7045585689555683,
"grad_norm": 0.8359375,
"learning_rate": 0.0004960685744524934,
"loss": 6.2881,
"mean_token_accuracy": 0.16480329632759094,
"num_tokens": 15481261.0,
"step": 6105
},
{
"entropy": 6.350079822540283,
"epoch": 0.705135603000577,
"grad_norm": 0.81640625,
"learning_rate": 0.0004960608906086558,
"loss": 6.2208,
"mean_token_accuracy": 0.16420309692621232,
"num_tokens": 15494996.0,
"step": 6110
},
{
"entropy": 6.404261541366577,
"epoch": 0.7057126370455857,
"grad_norm": 0.7890625,
"learning_rate": 0.000496053199329545,
"loss": 6.1642,
"mean_token_accuracy": 0.16822902262210845,
"num_tokens": 15506772.0,
"step": 6115
},
{
"entropy": 6.3732819080352785,
"epoch": 0.7062896710905944,
"grad_norm": 0.8203125,
"learning_rate": 0.0004960455006154196,
"loss": 6.2876,
"mean_token_accuracy": 0.16168780773878097,
"num_tokens": 15520336.0,
"step": 6120
},
{
"entropy": 6.474810409545898,
"epoch": 0.706866705135603,
"grad_norm": 0.84765625,
"learning_rate": 0.0004960377944665386,
"loss": 6.3679,
"mean_token_accuracy": 0.15211256816983224,
"num_tokens": 15535139.0,
"step": 6125
},
{
"entropy": 6.396150255203247,
"epoch": 0.7074437391806117,
"grad_norm": 0.8359375,
"learning_rate": 0.0004960300808831611,
"loss": 6.2355,
"mean_token_accuracy": 0.1614559680223465,
"num_tokens": 15547523.0,
"step": 6130
},
{
"entropy": 6.351924133300781,
"epoch": 0.7080207732256203,
"grad_norm": 0.80078125,
"learning_rate": 0.0004960223598655467,
"loss": 6.2385,
"mean_token_accuracy": 0.16700737476348876,
"num_tokens": 15560816.0,
"step": 6135
},
{
"entropy": 6.430610704421997,
"epoch": 0.7085978072706289,
"grad_norm": 0.7890625,
"learning_rate": 0.000496014631413955,
"loss": 6.2409,
"mean_token_accuracy": 0.1643073469400406,
"num_tokens": 15574364.0,
"step": 6140
},
{
"entropy": 6.307897710800171,
"epoch": 0.7091748413156376,
"grad_norm": 0.83203125,
"learning_rate": 0.0004960068955286459,
"loss": 6.0504,
"mean_token_accuracy": 0.18151674717664718,
"num_tokens": 15586969.0,
"step": 6145
},
{
"entropy": 6.4239410877227785,
"epoch": 0.7097518753606463,
"grad_norm": 0.8671875,
"learning_rate": 0.0004959991522098797,
"loss": 6.2119,
"mean_token_accuracy": 0.16022869125008582,
"num_tokens": 15600459.0,
"step": 6150
},
{
"entropy": 6.444924354553223,
"epoch": 0.710328909405655,
"grad_norm": 0.8359375,
"learning_rate": 0.0004959914014579168,
"loss": 6.2652,
"mean_token_accuracy": 0.15775836259126663,
"num_tokens": 15613484.0,
"step": 6155
},
{
"entropy": 6.353692579269409,
"epoch": 0.7109059434506636,
"grad_norm": 0.8828125,
"learning_rate": 0.0004959836432730178,
"loss": 6.282,
"mean_token_accuracy": 0.16222158819437027,
"num_tokens": 15626036.0,
"step": 6160
},
{
"entropy": 6.286346817016602,
"epoch": 0.7114829774956722,
"grad_norm": 1.171875,
"learning_rate": 0.0004959758776554438,
"loss": 6.2124,
"mean_token_accuracy": 0.17564715892076493,
"num_tokens": 15640229.0,
"step": 6165
},
{
"entropy": 6.401192998886108,
"epoch": 0.7120600115406809,
"grad_norm": 0.9140625,
"learning_rate": 0.000495968104605456,
"loss": 6.1071,
"mean_token_accuracy": 0.17238144278526307,
"num_tokens": 15652655.0,
"step": 6170
},
{
"entropy": 6.407490825653076,
"epoch": 0.7126370455856895,
"grad_norm": 0.859375,
"learning_rate": 0.0004959603241233157,
"loss": 6.2637,
"mean_token_accuracy": 0.16557672023773193,
"num_tokens": 15665535.0,
"step": 6175
},
{
"entropy": 6.40711145401001,
"epoch": 0.7132140796306982,
"grad_norm": 0.765625,
"learning_rate": 0.0004959525362092846,
"loss": 6.2342,
"mean_token_accuracy": 0.16254877895116807,
"num_tokens": 15678631.0,
"step": 6180
},
{
"entropy": 6.4322343349456785,
"epoch": 0.7137911136757069,
"grad_norm": 0.7890625,
"learning_rate": 0.0004959447408636247,
"loss": 6.3179,
"mean_token_accuracy": 0.16229903995990752,
"num_tokens": 15691662.0,
"step": 6185
},
{
"entropy": 6.395531558990479,
"epoch": 0.7143681477207156,
"grad_norm": 0.87890625,
"learning_rate": 0.0004959369380865983,
"loss": 6.2408,
"mean_token_accuracy": 0.1694466605782509,
"num_tokens": 15705017.0,
"step": 6190
},
{
"entropy": 6.42110686302185,
"epoch": 0.7149451817657242,
"grad_norm": 0.8203125,
"learning_rate": 0.0004959291278784676,
"loss": 6.2544,
"mean_token_accuracy": 0.16188978999853135,
"num_tokens": 15717537.0,
"step": 6195
},
{
"entropy": 6.332228946685791,
"epoch": 0.7155222158107328,
"grad_norm": 0.83203125,
"learning_rate": 0.0004959213102394954,
"loss": 6.0963,
"mean_token_accuracy": 0.16999780088663102,
"num_tokens": 15730075.0,
"step": 6200
},
{
"entropy": 6.376552534103394,
"epoch": 0.7160992498557415,
"grad_norm": 0.83203125,
"learning_rate": 0.0004959134851699449,
"loss": 6.1954,
"mean_token_accuracy": 0.16710759848356246,
"num_tokens": 15743187.0,
"step": 6205
},
{
"entropy": 6.411378002166748,
"epoch": 0.7166762839007501,
"grad_norm": 0.8828125,
"learning_rate": 0.0004959056526700788,
"loss": 6.2817,
"mean_token_accuracy": 0.1558884635567665,
"num_tokens": 15755459.0,
"step": 6210
},
{
"entropy": 6.355316114425659,
"epoch": 0.7172533179457588,
"grad_norm": 0.83984375,
"learning_rate": 0.0004958978127401609,
"loss": 6.1278,
"mean_token_accuracy": 0.16565006971359253,
"num_tokens": 15767929.0,
"step": 6215
},
{
"entropy": 6.33356614112854,
"epoch": 0.7178303519907675,
"grad_norm": 0.8125,
"learning_rate": 0.0004958899653804548,
"loss": 6.1179,
"mean_token_accuracy": 0.1736751303076744,
"num_tokens": 15780079.0,
"step": 6220
},
{
"entropy": 6.420310592651367,
"epoch": 0.7184073860357761,
"grad_norm": 0.86328125,
"learning_rate": 0.0004958821105912246,
"loss": 6.217,
"mean_token_accuracy": 0.16546618044376374,
"num_tokens": 15792417.0,
"step": 6225
},
{
"entropy": 6.43408350944519,
"epoch": 0.7189844200807848,
"grad_norm": 0.7734375,
"learning_rate": 0.000495874248372734,
"loss": 6.2172,
"mean_token_accuracy": 0.1711224213242531,
"num_tokens": 15804993.0,
"step": 6230
},
{
"entropy": 6.350424289703369,
"epoch": 0.7195614541257934,
"grad_norm": 0.77734375,
"learning_rate": 0.000495866378725248,
"loss": 6.2582,
"mean_token_accuracy": 0.16318586766719817,
"num_tokens": 15819765.0,
"step": 6235
},
{
"entropy": 6.361863899230957,
"epoch": 0.7201384881708021,
"grad_norm": 0.86328125,
"learning_rate": 0.000495858501649031,
"loss": 6.2619,
"mean_token_accuracy": 0.16482506394386293,
"num_tokens": 15831992.0,
"step": 6240
},
{
"entropy": 6.335883855819702,
"epoch": 0.7207155222158107,
"grad_norm": 0.81640625,
"learning_rate": 0.000495850617144348,
"loss": 6.2084,
"mean_token_accuracy": 0.17357497811317443,
"num_tokens": 15845348.0,
"step": 6245
},
{
"entropy": 6.4281090259552,
"epoch": 0.7212925562608193,
"grad_norm": 0.8203125,
"learning_rate": 0.0004958427252114643,
"loss": 6.2316,
"mean_token_accuracy": 0.16620133072137833,
"num_tokens": 15857677.0,
"step": 6250
},
{
"entropy": 6.310732650756836,
"epoch": 0.7218695903058281,
"grad_norm": 0.8359375,
"learning_rate": 0.000495834825850645,
"loss": 6.1912,
"mean_token_accuracy": 0.17365667521953582,
"num_tokens": 15871626.0,
"step": 6255
},
{
"entropy": 6.438030433654785,
"epoch": 0.7224466243508367,
"grad_norm": 0.7734375,
"learning_rate": 0.0004958269190621562,
"loss": 6.2999,
"mean_token_accuracy": 0.16112429201602935,
"num_tokens": 15884372.0,
"step": 6260
},
{
"entropy": 6.265491580963134,
"epoch": 0.7230236583958454,
"grad_norm": 0.89453125,
"learning_rate": 0.0004958190048462637,
"loss": 6.1757,
"mean_token_accuracy": 0.16929175704717636,
"num_tokens": 15896079.0,
"step": 6265
},
{
"entropy": 6.373610734939575,
"epoch": 0.723600692440854,
"grad_norm": 0.828125,
"learning_rate": 0.0004958110832032336,
"loss": 6.2789,
"mean_token_accuracy": 0.16542789340019226,
"num_tokens": 15909528.0,
"step": 6270
},
{
"entropy": 6.391331529617309,
"epoch": 0.7241777264858626,
"grad_norm": 0.88671875,
"learning_rate": 0.0004958031541333322,
"loss": 6.203,
"mean_token_accuracy": 0.16602921783924102,
"num_tokens": 15921247.0,
"step": 6275
},
{
"entropy": 6.411171197891235,
"epoch": 0.7247547605308713,
"grad_norm": 0.84375,
"learning_rate": 0.0004957952176368267,
"loss": 6.2785,
"mean_token_accuracy": 0.16443133652210234,
"num_tokens": 15933093.0,
"step": 6280
},
{
"entropy": 6.3585645198822025,
"epoch": 0.7253317945758799,
"grad_norm": 0.796875,
"learning_rate": 0.0004957872737139836,
"loss": 6.2239,
"mean_token_accuracy": 0.16355139017105103,
"num_tokens": 15944888.0,
"step": 6285
},
{
"entropy": 6.40592737197876,
"epoch": 0.7259088286208887,
"grad_norm": 0.78125,
"learning_rate": 0.0004957793223650702,
"loss": 6.2547,
"mean_token_accuracy": 0.16579299718141555,
"num_tokens": 15957793.0,
"step": 6290
},
{
"entropy": 6.399983835220337,
"epoch": 0.7264858626658973,
"grad_norm": 0.92578125,
"learning_rate": 0.000495771363590354,
"loss": 6.2577,
"mean_token_accuracy": 0.15640069246292115,
"num_tokens": 15970152.0,
"step": 6295
},
{
"entropy": 6.358456373214722,
"epoch": 0.727062896710906,
"grad_norm": 0.81640625,
"learning_rate": 0.0004957633973901027,
"loss": 6.2443,
"mean_token_accuracy": 0.16428245455026627,
"num_tokens": 15983347.0,
"step": 6300
},
{
"entropy": 6.383837699890137,
"epoch": 0.7276399307559146,
"grad_norm": 0.80078125,
"learning_rate": 0.0004957554237645841,
"loss": 6.212,
"mean_token_accuracy": 0.1730270951986313,
"num_tokens": 15996148.0,
"step": 6305
},
{
"entropy": 6.391459274291992,
"epoch": 0.7282169648009232,
"grad_norm": 0.8515625,
"learning_rate": 0.0004957474427140665,
"loss": 6.2195,
"mean_token_accuracy": 0.16915312260389329,
"num_tokens": 16008833.0,
"step": 6310
},
{
"entropy": 6.3193872451782225,
"epoch": 0.7287939988459319,
"grad_norm": 0.88671875,
"learning_rate": 0.0004957394542388182,
"loss": 6.1426,
"mean_token_accuracy": 0.172357177734375,
"num_tokens": 16020674.0,
"step": 6315
},
{
"entropy": 6.316094303131104,
"epoch": 0.7293710328909405,
"grad_norm": 1.1640625,
"learning_rate": 0.0004957314583391082,
"loss": 6.1191,
"mean_token_accuracy": 0.1790053740143776,
"num_tokens": 16033229.0,
"step": 6320
},
{
"entropy": 6.403400850296021,
"epoch": 0.7299480669359493,
"grad_norm": 0.78125,
"learning_rate": 0.0004957234550152052,
"loss": 6.1338,
"mean_token_accuracy": 0.16456628888845443,
"num_tokens": 16045646.0,
"step": 6325
},
{
"entropy": 6.404736471176148,
"epoch": 0.7305251009809579,
"grad_norm": 1.03125,
"learning_rate": 0.0004957154442673784,
"loss": 6.2346,
"mean_token_accuracy": 0.16127407997846605,
"num_tokens": 16058400.0,
"step": 6330
},
{
"entropy": 6.295368671417236,
"epoch": 0.7311021350259665,
"grad_norm": 0.86328125,
"learning_rate": 0.0004957074260958972,
"loss": 6.2097,
"mean_token_accuracy": 0.16328535377979278,
"num_tokens": 16071067.0,
"step": 6335
},
{
"entropy": 6.4290313720703125,
"epoch": 0.7316791690709752,
"grad_norm": 0.80859375,
"learning_rate": 0.0004956994005010315,
"loss": 6.2093,
"mean_token_accuracy": 0.16073839813470842,
"num_tokens": 16083760.0,
"step": 6340
},
{
"entropy": 6.383505821228027,
"epoch": 0.7322562031159838,
"grad_norm": 0.8671875,
"learning_rate": 0.000495691367483051,
"loss": 6.1714,
"mean_token_accuracy": 0.17427341043949127,
"num_tokens": 16095469.0,
"step": 6345
},
{
"entropy": 6.330617809295655,
"epoch": 0.7328332371609925,
"grad_norm": 0.91015625,
"learning_rate": 0.0004956833270422259,
"loss": 6.1945,
"mean_token_accuracy": 0.167451611161232,
"num_tokens": 16108215.0,
"step": 6350
},
{
"entropy": 6.456769227981567,
"epoch": 0.7334102712060011,
"grad_norm": 0.859375,
"learning_rate": 0.0004956752791788269,
"loss": 6.2532,
"mean_token_accuracy": 0.16016148626804352,
"num_tokens": 16121050.0,
"step": 6355
},
{
"entropy": 6.417612981796265,
"epoch": 0.7339873052510099,
"grad_norm": 0.85546875,
"learning_rate": 0.0004956672238931244,
"loss": 6.2578,
"mean_token_accuracy": 0.1622995361685753,
"num_tokens": 16133603.0,
"step": 6360
},
{
"entropy": 6.339549922943116,
"epoch": 0.7345643392960185,
"grad_norm": 0.83984375,
"learning_rate": 0.0004956591611853894,
"loss": 6.1409,
"mean_token_accuracy": 0.171478071808815,
"num_tokens": 16146797.0,
"step": 6365
},
{
"entropy": 6.4115142822265625,
"epoch": 0.7351413733410271,
"grad_norm": 0.73828125,
"learning_rate": 0.0004956510910558931,
"loss": 6.3133,
"mean_token_accuracy": 0.1641661122441292,
"num_tokens": 16159223.0,
"step": 6370
},
{
"entropy": 6.351845216751099,
"epoch": 0.7357184073860358,
"grad_norm": 0.8359375,
"learning_rate": 0.0004956430135049071,
"loss": 6.2143,
"mean_token_accuracy": 0.1620581865310669,
"num_tokens": 16172305.0,
"step": 6375
},
{
"entropy": 6.428864145278931,
"epoch": 0.7362954414310444,
"grad_norm": 0.98828125,
"learning_rate": 0.0004956349285327028,
"loss": 6.2265,
"mean_token_accuracy": 0.16274599581956864,
"num_tokens": 16185575.0,
"step": 6380
},
{
"entropy": 6.340291213989258,
"epoch": 0.7368724754760531,
"grad_norm": 0.921875,
"learning_rate": 0.0004956268361395523,
"loss": 6.1484,
"mean_token_accuracy": 0.17272731512784958,
"num_tokens": 16198014.0,
"step": 6385
},
{
"entropy": 6.40310697555542,
"epoch": 0.7374495095210617,
"grad_norm": 0.94921875,
"learning_rate": 0.0004956187363257278,
"loss": 6.2038,
"mean_token_accuracy": 0.16336548924446107,
"num_tokens": 16211717.0,
"step": 6390
},
{
"entropy": 6.410084342956543,
"epoch": 0.7380265435660704,
"grad_norm": 0.8984375,
"learning_rate": 0.0004956106290915017,
"loss": 6.2492,
"mean_token_accuracy": 0.16823570430278778,
"num_tokens": 16225039.0,
"step": 6395
},
{
"entropy": 6.312919092178345,
"epoch": 0.7386035776110791,
"grad_norm": 0.8671875,
"learning_rate": 0.0004956025144371467,
"loss": 6.1904,
"mean_token_accuracy": 0.16095079034566878,
"num_tokens": 16236832.0,
"step": 6400
},
{
"entropy": 6.389898061752319,
"epoch": 0.7391806116560877,
"grad_norm": 0.89453125,
"learning_rate": 0.0004955943923629356,
"loss": 6.1755,
"mean_token_accuracy": 0.1681758537888527,
"num_tokens": 16250093.0,
"step": 6405
},
{
"entropy": 6.388084173202515,
"epoch": 0.7397576457010964,
"grad_norm": 0.859375,
"learning_rate": 0.0004955862628691417,
"loss": 6.1642,
"mean_token_accuracy": 0.1677396222949028,
"num_tokens": 16263051.0,
"step": 6410
},
{
"entropy": 6.302446174621582,
"epoch": 0.740334679746105,
"grad_norm": 0.828125,
"learning_rate": 0.0004955781259560386,
"loss": 6.2202,
"mean_token_accuracy": 0.16396949291229249,
"num_tokens": 16276545.0,
"step": 6415
},
{
"entropy": 6.393322086334228,
"epoch": 0.7409117137911136,
"grad_norm": 0.78515625,
"learning_rate": 0.0004955699816238995,
"loss": 6.2185,
"mean_token_accuracy": 0.16904087066650392,
"num_tokens": 16288900.0,
"step": 6420
},
{
"entropy": 6.370112466812134,
"epoch": 0.7414887478361223,
"grad_norm": 0.83203125,
"learning_rate": 0.0004955618298729988,
"loss": 6.2267,
"mean_token_accuracy": 0.1618413209915161,
"num_tokens": 16302066.0,
"step": 6425
},
{
"entropy": 6.420699691772461,
"epoch": 0.742065781881131,
"grad_norm": 0.77734375,
"learning_rate": 0.0004955536707036105,
"loss": 6.3154,
"mean_token_accuracy": 0.1590120330452919,
"num_tokens": 16316255.0,
"step": 6430
},
{
"entropy": 6.3836956977844235,
"epoch": 0.7426428159261397,
"grad_norm": 0.796875,
"learning_rate": 0.000495545504116009,
"loss": 6.2086,
"mean_token_accuracy": 0.16407999843358995,
"num_tokens": 16328419.0,
"step": 6435
},
{
"entropy": 6.333625841140747,
"epoch": 0.7432198499711483,
"grad_norm": 0.9453125,
"learning_rate": 0.0004955373301104691,
"loss": 6.1446,
"mean_token_accuracy": 0.17361177057027816,
"num_tokens": 16339864.0,
"step": 6440
},
{
"entropy": 6.400269651412964,
"epoch": 0.743796884016157,
"grad_norm": 0.953125,
"learning_rate": 0.0004955291486872657,
"loss": 6.2265,
"mean_token_accuracy": 0.16552126109600068,
"num_tokens": 16354364.0,
"step": 6445
},
{
"entropy": 6.3924407958984375,
"epoch": 0.7443739180611656,
"grad_norm": 0.91015625,
"learning_rate": 0.0004955209598466738,
"loss": 6.2129,
"mean_token_accuracy": 0.15995777845382692,
"num_tokens": 16366118.0,
"step": 6450
},
{
"entropy": 6.413617372512817,
"epoch": 0.7449509521061742,
"grad_norm": 0.9140625,
"learning_rate": 0.000495512763588969,
"loss": 6.2586,
"mean_token_accuracy": 0.16330986469984055,
"num_tokens": 16378805.0,
"step": 6455
},
{
"entropy": 6.362340259552002,
"epoch": 0.7455279861511829,
"grad_norm": 0.82421875,
"learning_rate": 0.0004955045599144269,
"loss": 6.2099,
"mean_token_accuracy": 0.17093328684568404,
"num_tokens": 16391202.0,
"step": 6460
},
{
"entropy": 6.318493318557739,
"epoch": 0.7461050201961916,
"grad_norm": 0.796875,
"learning_rate": 0.0004954963488233235,
"loss": 6.1521,
"mean_token_accuracy": 0.16465499103069306,
"num_tokens": 16402409.0,
"step": 6465
},
{
"entropy": 6.378865432739258,
"epoch": 0.7466820542412003,
"grad_norm": 0.8203125,
"learning_rate": 0.000495488130315935,
"loss": 6.1678,
"mean_token_accuracy": 0.166618475317955,
"num_tokens": 16415710.0,
"step": 6470
},
{
"entropy": 6.359663486480713,
"epoch": 0.7472590882862089,
"grad_norm": 0.7734375,
"learning_rate": 0.0004954799043925377,
"loss": 6.1871,
"mean_token_accuracy": 0.1670057252049446,
"num_tokens": 16428135.0,
"step": 6475
},
{
"entropy": 6.366726970672607,
"epoch": 0.7478361223312175,
"grad_norm": 0.87109375,
"learning_rate": 0.0004954716710534082,
"loss": 6.2828,
"mean_token_accuracy": 0.16121697127819062,
"num_tokens": 16440052.0,
"step": 6480
},
{
"entropy": 6.375087308883667,
"epoch": 0.7484131563762262,
"grad_norm": 1.046875,
"learning_rate": 0.0004954634302988237,
"loss": 6.194,
"mean_token_accuracy": 0.1659095734357834,
"num_tokens": 16453188.0,
"step": 6485
},
{
"entropy": 6.43290228843689,
"epoch": 0.7489901904212348,
"grad_norm": 0.78125,
"learning_rate": 0.0004954551821290612,
"loss": 6.2665,
"mean_token_accuracy": 0.1626218318939209,
"num_tokens": 16464946.0,
"step": 6490
},
{
"entropy": 6.450923252105713,
"epoch": 0.7495672244662435,
"grad_norm": 0.859375,
"learning_rate": 0.0004954469265443981,
"loss": 6.2631,
"mean_token_accuracy": 0.1620316132903099,
"num_tokens": 16476957.0,
"step": 6495
},
{
"entropy": 6.407968378067016,
"epoch": 0.7501442585112522,
"grad_norm": 0.88671875,
"learning_rate": 0.0004954386635451123,
"loss": 6.1228,
"mean_token_accuracy": 0.17593680024147035,
"num_tokens": 16488847.0,
"step": 6500
},
{
"entropy": 6.364045333862305,
"epoch": 0.7507212925562609,
"grad_norm": 0.7890625,
"learning_rate": 0.0004954303931314814,
"loss": 6.1777,
"mean_token_accuracy": 0.17224719375371933,
"num_tokens": 16502497.0,
"step": 6505
},
{
"entropy": 6.375355052947998,
"epoch": 0.7512983266012695,
"grad_norm": 0.94140625,
"learning_rate": 0.0004954221153037837,
"loss": 6.2474,
"mean_token_accuracy": 0.16628716737031937,
"num_tokens": 16514799.0,
"step": 6510
},
{
"entropy": 6.4122003555297855,
"epoch": 0.7518753606462781,
"grad_norm": 0.8125,
"learning_rate": 0.0004954138300622978,
"loss": 6.2757,
"mean_token_accuracy": 0.16112708896398545,
"num_tokens": 16528051.0,
"step": 6515
},
{
"entropy": 6.426703834533692,
"epoch": 0.7524523946912868,
"grad_norm": 0.78515625,
"learning_rate": 0.000495405537407302,
"loss": 6.2066,
"mean_token_accuracy": 0.16361595392227174,
"num_tokens": 16540127.0,
"step": 6520
},
{
"entropy": 6.361682319641114,
"epoch": 0.7530294287362954,
"grad_norm": 0.84765625,
"learning_rate": 0.0004953972373390755,
"loss": 6.1774,
"mean_token_accuracy": 0.1759776309132576,
"num_tokens": 16553419.0,
"step": 6525
},
{
"entropy": 6.464717721939087,
"epoch": 0.753606462781304,
"grad_norm": 0.8359375,
"learning_rate": 0.0004953889298578975,
"loss": 6.2795,
"mean_token_accuracy": 0.16779824048280717,
"num_tokens": 16567068.0,
"step": 6530
},
{
"entropy": 6.353485202789306,
"epoch": 0.7541834968263128,
"grad_norm": 0.8671875,
"learning_rate": 0.0004953806149640473,
"loss": 6.1126,
"mean_token_accuracy": 0.1709718957543373,
"num_tokens": 16579516.0,
"step": 6535
},
{
"entropy": 6.367344331741333,
"epoch": 0.7547605308713214,
"grad_norm": 0.8671875,
"learning_rate": 0.0004953722926578045,
"loss": 6.2389,
"mean_token_accuracy": 0.1674762025475502,
"num_tokens": 16592489.0,
"step": 6540
},
{
"entropy": 6.301311111450195,
"epoch": 0.7553375649163301,
"grad_norm": 0.875,
"learning_rate": 0.0004953639629394492,
"loss": 6.1814,
"mean_token_accuracy": 0.17027597576379777,
"num_tokens": 16604848.0,
"step": 6545
},
{
"entropy": 6.436667346954346,
"epoch": 0.7559145989613387,
"grad_norm": 0.8828125,
"learning_rate": 0.0004953556258092613,
"loss": 6.2132,
"mean_token_accuracy": 0.16695142388343812,
"num_tokens": 16617426.0,
"step": 6550
},
{
"entropy": 6.425419187545776,
"epoch": 0.7564916330063474,
"grad_norm": 0.8828125,
"learning_rate": 0.0004953472812675216,
"loss": 6.217,
"mean_token_accuracy": 0.1640801399946213,
"num_tokens": 16629676.0,
"step": 6555
},
{
"entropy": 6.4231878280639645,
"epoch": 0.757068667051356,
"grad_norm": 0.8671875,
"learning_rate": 0.0004953389293145104,
"loss": 6.2477,
"mean_token_accuracy": 0.15641462951898574,
"num_tokens": 16642113.0,
"step": 6560
},
{
"entropy": 6.338461065292359,
"epoch": 0.7576457010963646,
"grad_norm": 0.859375,
"learning_rate": 0.0004953305699505088,
"loss": 6.179,
"mean_token_accuracy": 0.16644170880317688,
"num_tokens": 16653866.0,
"step": 6565
},
{
"entropy": 6.3428082942962645,
"epoch": 0.7582227351413734,
"grad_norm": 0.83984375,
"learning_rate": 0.0004953222031757979,
"loss": 6.2116,
"mean_token_accuracy": 0.16792171001434325,
"num_tokens": 16666917.0,
"step": 6570
},
{
"entropy": 6.365523910522461,
"epoch": 0.758799769186382,
"grad_norm": 0.86328125,
"learning_rate": 0.0004953138289906592,
"loss": 6.155,
"mean_token_accuracy": 0.16653495728969575,
"num_tokens": 16678752.0,
"step": 6575
},
{
"entropy": 6.335975551605225,
"epoch": 0.7593768032313907,
"grad_norm": 0.9296875,
"learning_rate": 0.0004953054473953742,
"loss": 6.1744,
"mean_token_accuracy": 0.17191429883241655,
"num_tokens": 16690942.0,
"step": 6580
},
{
"entropy": 6.373892307281494,
"epoch": 0.7599538372763993,
"grad_norm": 0.890625,
"learning_rate": 0.0004952970583902251,
"loss": 6.1776,
"mean_token_accuracy": 0.1657567486166954,
"num_tokens": 16703009.0,
"step": 6585
},
{
"entropy": 6.329607057571411,
"epoch": 0.760530871321408,
"grad_norm": 0.7265625,
"learning_rate": 0.0004952886619754937,
"loss": 6.1528,
"mean_token_accuracy": 0.16590441018342972,
"num_tokens": 16714881.0,
"step": 6590
},
{
"entropy": 6.353391504287719,
"epoch": 0.7611079053664166,
"grad_norm": 0.765625,
"learning_rate": 0.0004952802581514625,
"loss": 6.2282,
"mean_token_accuracy": 0.16431566178798676,
"num_tokens": 16729892.0,
"step": 6595
},
{
"entropy": 6.42394437789917,
"epoch": 0.7616849394114252,
"grad_norm": 0.7890625,
"learning_rate": 0.0004952718469184144,
"loss": 6.1992,
"mean_token_accuracy": 0.16453457325696946,
"num_tokens": 16742974.0,
"step": 6600
},
{
"entropy": 6.389783954620361,
"epoch": 0.762261973456434,
"grad_norm": 0.8359375,
"learning_rate": 0.0004952634282766322,
"loss": 6.2503,
"mean_token_accuracy": 0.158389113843441,
"num_tokens": 16756758.0,
"step": 6605
},
{
"entropy": 6.463902425765991,
"epoch": 0.7628390075014426,
"grad_norm": 0.78125,
"learning_rate": 0.0004952550022263988,
"loss": 6.2623,
"mean_token_accuracy": 0.16274497509002686,
"num_tokens": 16768875.0,
"step": 6610
},
{
"entropy": 6.353977489471435,
"epoch": 0.7634160415464513,
"grad_norm": 0.8203125,
"learning_rate": 0.0004952465687679979,
"loss": 6.1515,
"mean_token_accuracy": 0.16281863003969194,
"num_tokens": 16780526.0,
"step": 6615
},
{
"entropy": 6.3285585880279545,
"epoch": 0.7639930755914599,
"grad_norm": 0.8671875,
"learning_rate": 0.000495238127901713,
"loss": 6.1239,
"mean_token_accuracy": 0.1764816462993622,
"num_tokens": 16793778.0,
"step": 6620
},
{
"entropy": 6.302238130569458,
"epoch": 0.7645701096364685,
"grad_norm": 0.79296875,
"learning_rate": 0.0004952296796278282,
"loss": 6.1349,
"mean_token_accuracy": 0.16890315860509872,
"num_tokens": 16805367.0,
"step": 6625
},
{
"entropy": 6.379828214645386,
"epoch": 0.7651471436814772,
"grad_norm": 0.890625,
"learning_rate": 0.0004952212239466274,
"loss": 6.1833,
"mean_token_accuracy": 0.16319627091288566,
"num_tokens": 16818298.0,
"step": 6630
},
{
"entropy": 6.403966951370239,
"epoch": 0.7657241777264858,
"grad_norm": 0.83984375,
"learning_rate": 0.0004952127608583953,
"loss": 6.2453,
"mean_token_accuracy": 0.16557506173849107,
"num_tokens": 16831031.0,
"step": 6635
},
{
"entropy": 6.322995996475219,
"epoch": 0.7663012117714946,
"grad_norm": 0.76171875,
"learning_rate": 0.0004952042903634162,
"loss": 6.2186,
"mean_token_accuracy": 0.17293581366539001,
"num_tokens": 16844143.0,
"step": 6640
},
{
"entropy": 6.311652565002442,
"epoch": 0.7668782458165032,
"grad_norm": 0.8359375,
"learning_rate": 0.0004951958124619752,
"loss": 6.1241,
"mean_token_accuracy": 0.18038636595010757,
"num_tokens": 16857639.0,
"step": 6645
},
{
"entropy": 6.274792861938477,
"epoch": 0.7674552798615119,
"grad_norm": 0.79296875,
"learning_rate": 0.0004951873271543573,
"loss": 6.1833,
"mean_token_accuracy": 0.1697569027543068,
"num_tokens": 16870479.0,
"step": 6650
},
{
"entropy": 6.442476463317871,
"epoch": 0.7680323139065205,
"grad_norm": 0.859375,
"learning_rate": 0.0004951788344408483,
"loss": 6.1996,
"mean_token_accuracy": 0.17086183428764343,
"num_tokens": 16883876.0,
"step": 6655
},
{
"entropy": 6.388926219940186,
"epoch": 0.7686093479515291,
"grad_norm": 0.87890625,
"learning_rate": 0.0004951703343217335,
"loss": 6.1687,
"mean_token_accuracy": 0.1668828547000885,
"num_tokens": 16895516.0,
"step": 6660
},
{
"entropy": 6.394172859191895,
"epoch": 0.7691863819965378,
"grad_norm": 0.8828125,
"learning_rate": 0.0004951618267972987,
"loss": 6.231,
"mean_token_accuracy": 0.1687481701374054,
"num_tokens": 16906558.0,
"step": 6665
},
{
"entropy": 6.362077713012695,
"epoch": 0.7697634160415464,
"grad_norm": 1.0234375,
"learning_rate": 0.0004951533118678304,
"loss": 6.1972,
"mean_token_accuracy": 0.17087253034114838,
"num_tokens": 16919951.0,
"step": 6670
},
{
"entropy": 6.459950017929077,
"epoch": 0.7703404500865552,
"grad_norm": 0.84765625,
"learning_rate": 0.0004951447895336147,
"loss": 6.2211,
"mean_token_accuracy": 0.16551271378993987,
"num_tokens": 16932885.0,
"step": 6675
},
{
"entropy": 6.363070869445801,
"epoch": 0.7709174841315638,
"grad_norm": 0.80859375,
"learning_rate": 0.0004951362597949384,
"loss": 6.2169,
"mean_token_accuracy": 0.1604614496231079,
"num_tokens": 16944769.0,
"step": 6680
},
{
"entropy": 6.330994749069214,
"epoch": 0.7714945181765724,
"grad_norm": 0.8359375,
"learning_rate": 0.0004951277226520883,
"loss": 6.2045,
"mean_token_accuracy": 0.162946917116642,
"num_tokens": 16958652.0,
"step": 6685
},
{
"entropy": 6.24558916091919,
"epoch": 0.7720715522215811,
"grad_norm": 0.87109375,
"learning_rate": 0.0004951191781053517,
"loss": 6.0816,
"mean_token_accuracy": 0.18285587728023528,
"num_tokens": 16971850.0,
"step": 6690
},
{
"entropy": 6.400682544708252,
"epoch": 0.7726485862665897,
"grad_norm": 0.99609375,
"learning_rate": 0.0004951106261550157,
"loss": 6.1938,
"mean_token_accuracy": 0.16207986921072007,
"num_tokens": 16984164.0,
"step": 6695
},
{
"entropy": 6.372224807739258,
"epoch": 0.7732256203115984,
"grad_norm": 0.78515625,
"learning_rate": 0.0004951020668013683,
"loss": 6.2855,
"mean_token_accuracy": 0.15490710884332656,
"num_tokens": 16999140.0,
"step": 6700
},
{
"entropy": 6.408423328399659,
"epoch": 0.773802654356607,
"grad_norm": 0.796875,
"learning_rate": 0.0004950935000446972,
"loss": 6.1631,
"mean_token_accuracy": 0.17136792242527008,
"num_tokens": 17012137.0,
"step": 6705
},
{
"entropy": 6.317019891738892,
"epoch": 0.7743796884016156,
"grad_norm": 0.82421875,
"learning_rate": 0.0004950849258852905,
"loss": 6.1201,
"mean_token_accuracy": 0.1736213445663452,
"num_tokens": 17024594.0,
"step": 6710
},
{
"entropy": 6.341596508026123,
"epoch": 0.7749567224466244,
"grad_norm": 0.8515625,
"learning_rate": 0.0004950763443234366,
"loss": 6.2444,
"mean_token_accuracy": 0.1684948444366455,
"num_tokens": 17036927.0,
"step": 6715
},
{
"entropy": 6.291156196594239,
"epoch": 0.775533756491633,
"grad_norm": 0.8984375,
"learning_rate": 0.0004950677553594243,
"loss": 6.1711,
"mean_token_accuracy": 0.17052215486764907,
"num_tokens": 17049393.0,
"step": 6720
},
{
"entropy": 6.433130264282227,
"epoch": 0.7761107905366417,
"grad_norm": 0.8359375,
"learning_rate": 0.0004950591589935422,
"loss": 6.2294,
"mean_token_accuracy": 0.16084382981061934,
"num_tokens": 17063053.0,
"step": 6725
},
{
"entropy": 6.465151453018189,
"epoch": 0.7766878245816503,
"grad_norm": 0.921875,
"learning_rate": 0.0004950505552260798,
"loss": 6.2362,
"mean_token_accuracy": 0.16392787098884581,
"num_tokens": 17075700.0,
"step": 6730
},
{
"entropy": 6.315752601623535,
"epoch": 0.777264858626659,
"grad_norm": 0.84765625,
"learning_rate": 0.0004950419440573261,
"loss": 6.1095,
"mean_token_accuracy": 0.16548786759376527,
"num_tokens": 17088089.0,
"step": 6735
},
{
"entropy": 6.378903341293335,
"epoch": 0.7778418926716676,
"grad_norm": 0.8828125,
"learning_rate": 0.000495033325487571,
"loss": 6.1886,
"mean_token_accuracy": 0.17116651982069014,
"num_tokens": 17100769.0,
"step": 6740
},
{
"entropy": 6.349705028533935,
"epoch": 0.7784189267166762,
"grad_norm": 0.828125,
"learning_rate": 0.0004950246995171042,
"loss": 6.2394,
"mean_token_accuracy": 0.1664668083190918,
"num_tokens": 17113107.0,
"step": 6745
},
{
"entropy": 6.290965127944946,
"epoch": 0.778995960761685,
"grad_norm": 0.81640625,
"learning_rate": 0.000495016066146216,
"loss": 6.1825,
"mean_token_accuracy": 0.17528556734323503,
"num_tokens": 17125856.0,
"step": 6750
},
{
"entropy": 6.3704639911651615,
"epoch": 0.7795729948066936,
"grad_norm": 0.80859375,
"learning_rate": 0.0004950074253751968,
"loss": 6.1766,
"mean_token_accuracy": 0.16756103485822677,
"num_tokens": 17139931.0,
"step": 6755
},
{
"entropy": 6.3566728115081785,
"epoch": 0.7801500288517023,
"grad_norm": 0.7890625,
"learning_rate": 0.000494998777204337,
"loss": 6.1251,
"mean_token_accuracy": 0.16913940459489823,
"num_tokens": 17151669.0,
"step": 6760
},
{
"entropy": 6.3416680812835695,
"epoch": 0.7807270628967109,
"grad_norm": 0.796875,
"learning_rate": 0.0004949901216339276,
"loss": 6.142,
"mean_token_accuracy": 0.17505425065755845,
"num_tokens": 17164434.0,
"step": 6765
},
{
"entropy": 6.324961948394775,
"epoch": 0.7813040969417195,
"grad_norm": 0.83984375,
"learning_rate": 0.0004949814586642598,
"loss": 6.2053,
"mean_token_accuracy": 0.16748605370521547,
"num_tokens": 17176732.0,
"step": 6770
},
{
"entropy": 6.379989337921143,
"epoch": 0.7818811309867282,
"grad_norm": 0.8515625,
"learning_rate": 0.000494972788295625,
"loss": 6.1319,
"mean_token_accuracy": 0.16905532032251358,
"num_tokens": 17188698.0,
"step": 6775
},
{
"entropy": 6.338295555114746,
"epoch": 0.7824581650317368,
"grad_norm": 0.7890625,
"learning_rate": 0.0004949641105283144,
"loss": 6.1267,
"mean_token_accuracy": 0.17444328367710113,
"num_tokens": 17201219.0,
"step": 6780
},
{
"entropy": 6.366789197921753,
"epoch": 0.7830351990767456,
"grad_norm": 0.82421875,
"learning_rate": 0.0004949554253626205,
"loss": 6.235,
"mean_token_accuracy": 0.16587198823690413,
"num_tokens": 17213318.0,
"step": 6785
},
{
"entropy": 6.349771308898926,
"epoch": 0.7836122331217542,
"grad_norm": 0.86328125,
"learning_rate": 0.0004949467327988351,
"loss": 6.1603,
"mean_token_accuracy": 0.16665525138378143,
"num_tokens": 17226080.0,
"step": 6790
},
{
"entropy": 6.377096557617188,
"epoch": 0.7841892671667628,
"grad_norm": 0.80859375,
"learning_rate": 0.0004949380328372505,
"loss": 6.1844,
"mean_token_accuracy": 0.16439106017351152,
"num_tokens": 17237840.0,
"step": 6795
},
{
"entropy": 6.384302854537964,
"epoch": 0.7847663012117715,
"grad_norm": 0.9296875,
"learning_rate": 0.0004949293254781595,
"loss": 6.1254,
"mean_token_accuracy": 0.17288285493850708,
"num_tokens": 17249972.0,
"step": 6800
},
{
"entropy": 6.378346014022827,
"epoch": 0.7853433352567801,
"grad_norm": 0.87890625,
"learning_rate": 0.0004949206107218547,
"loss": 6.1792,
"mean_token_accuracy": 0.16759558916091918,
"num_tokens": 17261605.0,
"step": 6805
},
{
"entropy": 6.337935590744019,
"epoch": 0.7859203693017888,
"grad_norm": 0.81640625,
"learning_rate": 0.0004949118885686296,
"loss": 6.1345,
"mean_token_accuracy": 0.16815117448568345,
"num_tokens": 17273992.0,
"step": 6810
},
{
"entropy": 6.391584205627441,
"epoch": 0.7864974033467974,
"grad_norm": 0.859375,
"learning_rate": 0.0004949031590187774,
"loss": 6.174,
"mean_token_accuracy": 0.16896240413188934,
"num_tokens": 17286978.0,
"step": 6815
},
{
"entropy": 6.266331195831299,
"epoch": 0.7870744373918062,
"grad_norm": 0.88671875,
"learning_rate": 0.0004948944220725915,
"loss": 6.1817,
"mean_token_accuracy": 0.17290505915880203,
"num_tokens": 17298998.0,
"step": 6820
},
{
"entropy": 6.404459142684937,
"epoch": 0.7876514714368148,
"grad_norm": 0.8359375,
"learning_rate": 0.000494885677730366,
"loss": 6.2042,
"mean_token_accuracy": 0.16686858087778092,
"num_tokens": 17311614.0,
"step": 6825
},
{
"entropy": 6.361331844329834,
"epoch": 0.7882285054818234,
"grad_norm": 0.81640625,
"learning_rate": 0.000494876925992395,
"loss": 6.1912,
"mean_token_accuracy": 0.16285034716129304,
"num_tokens": 17323603.0,
"step": 6830
},
{
"entropy": 6.365416574478149,
"epoch": 0.7888055395268321,
"grad_norm": 0.859375,
"learning_rate": 0.0004948681668589728,
"loss": 6.1573,
"mean_token_accuracy": 0.17288437932729722,
"num_tokens": 17334916.0,
"step": 6835
},
{
"entropy": 6.240475463867187,
"epoch": 0.7893825735718407,
"grad_norm": 0.81640625,
"learning_rate": 0.000494859400330394,
"loss": 6.1895,
"mean_token_accuracy": 0.17153511196374893,
"num_tokens": 17346777.0,
"step": 6840
},
{
"entropy": 6.372337913513183,
"epoch": 0.7899596076168494,
"grad_norm": 0.89453125,
"learning_rate": 0.0004948506264069535,
"loss": 6.1922,
"mean_token_accuracy": 0.17216622829437256,
"num_tokens": 17359144.0,
"step": 6845
},
{
"entropy": 6.4244777202606205,
"epoch": 0.790536641661858,
"grad_norm": 0.890625,
"learning_rate": 0.0004948418450889464,
"loss": 6.2456,
"mean_token_accuracy": 0.16362111568450927,
"num_tokens": 17372992.0,
"step": 6850
},
{
"entropy": 6.280022144317627,
"epoch": 0.7911136757068667,
"grad_norm": 0.78125,
"learning_rate": 0.000494833056376668,
"loss": 6.0957,
"mean_token_accuracy": 0.17566343545913696,
"num_tokens": 17385413.0,
"step": 6855
},
{
"entropy": 6.367961072921753,
"epoch": 0.7916907097518754,
"grad_norm": 0.83984375,
"learning_rate": 0.0004948242602704139,
"loss": 6.2238,
"mean_token_accuracy": 0.16973874419927598,
"num_tokens": 17397882.0,
"step": 6860
},
{
"entropy": 6.288860702514649,
"epoch": 0.792267743796884,
"grad_norm": 0.87890625,
"learning_rate": 0.0004948154567704801,
"loss": 6.1115,
"mean_token_accuracy": 0.16626325845718384,
"num_tokens": 17410662.0,
"step": 6865
},
{
"entropy": 6.3847719669342045,
"epoch": 0.7928447778418927,
"grad_norm": 1.03125,
"learning_rate": 0.0004948066458771625,
"loss": 6.2049,
"mean_token_accuracy": 0.16260354667901994,
"num_tokens": 17423301.0,
"step": 6870
},
{
"entropy": 6.400874137878418,
"epoch": 0.7934218118869013,
"grad_norm": 0.83203125,
"learning_rate": 0.0004947978275907577,
"loss": 6.1772,
"mean_token_accuracy": 0.1704396441578865,
"num_tokens": 17435733.0,
"step": 6875
},
{
"entropy": 6.368968200683594,
"epoch": 0.79399884593191,
"grad_norm": 0.828125,
"learning_rate": 0.000494789001911562,
"loss": 6.2607,
"mean_token_accuracy": 0.15757424235343934,
"num_tokens": 17448516.0,
"step": 6880
},
{
"entropy": 6.3357648849487305,
"epoch": 0.7945758799769186,
"grad_norm": 0.85546875,
"learning_rate": 0.0004947801688398725,
"loss": 6.1557,
"mean_token_accuracy": 0.16735616326332092,
"num_tokens": 17462199.0,
"step": 6885
},
{
"entropy": 6.445990657806396,
"epoch": 0.7951529140219273,
"grad_norm": 0.828125,
"learning_rate": 0.0004947713283759862,
"loss": 6.2338,
"mean_token_accuracy": 0.16647788584232331,
"num_tokens": 17475651.0,
"step": 6890
},
{
"entropy": 6.448093700408935,
"epoch": 0.795729948066936,
"grad_norm": 0.8125,
"learning_rate": 0.0004947624805202003,
"loss": 6.2342,
"mean_token_accuracy": 0.16142688542604447,
"num_tokens": 17488264.0,
"step": 6895
},
{
"entropy": 6.368506669998169,
"epoch": 0.7963069821119446,
"grad_norm": 0.83203125,
"learning_rate": 0.0004947536252728126,
"loss": 6.1983,
"mean_token_accuracy": 0.16493333429098128,
"num_tokens": 17500691.0,
"step": 6900
},
{
"entropy": 6.353544139862061,
"epoch": 0.7968840161569533,
"grad_norm": 0.7578125,
"learning_rate": 0.0004947447626341209,
"loss": 6.1957,
"mean_token_accuracy": 0.1679762914776802,
"num_tokens": 17513957.0,
"step": 6905
},
{
"entropy": 6.4467274188995365,
"epoch": 0.7974610502019619,
"grad_norm": 0.83203125,
"learning_rate": 0.0004947358926044232,
"loss": 6.2185,
"mean_token_accuracy": 0.16547130346298217,
"num_tokens": 17527057.0,
"step": 6910
},
{
"entropy": 6.373523235321045,
"epoch": 0.7980380842469705,
"grad_norm": 0.828125,
"learning_rate": 0.0004947270151840179,
"loss": 6.1973,
"mean_token_accuracy": 0.15839087814092637,
"num_tokens": 17541215.0,
"step": 6915
},
{
"entropy": 6.35720386505127,
"epoch": 0.7986151182919792,
"grad_norm": 0.87109375,
"learning_rate": 0.0004947181303732038,
"loss": 6.21,
"mean_token_accuracy": 0.16345522552728653,
"num_tokens": 17553116.0,
"step": 6920
},
{
"entropy": 6.306585454940796,
"epoch": 0.7991921523369879,
"grad_norm": 0.765625,
"learning_rate": 0.0004947092381722793,
"loss": 6.1259,
"mean_token_accuracy": 0.16889618337154388,
"num_tokens": 17565480.0,
"step": 6925
},
{
"entropy": 6.295399761199951,
"epoch": 0.7997691863819966,
"grad_norm": 0.796875,
"learning_rate": 0.0004947003385815438,
"loss": 6.2127,
"mean_token_accuracy": 0.16810959428548813,
"num_tokens": 17577706.0,
"step": 6930
},
{
"entropy": 6.434894371032715,
"epoch": 0.8003462204270052,
"grad_norm": 0.8515625,
"learning_rate": 0.0004946914316012964,
"loss": 6.2355,
"mean_token_accuracy": 0.16640186607837676,
"num_tokens": 17591469.0,
"step": 6935
},
{
"entropy": 6.428550434112549,
"epoch": 0.8009232544720138,
"grad_norm": 0.7890625,
"learning_rate": 0.000494682517231837,
"loss": 6.164,
"mean_token_accuracy": 0.17068625688552858,
"num_tokens": 17605225.0,
"step": 6940
},
{
"entropy": 6.245584726333618,
"epoch": 0.8015002885170225,
"grad_norm": 0.91796875,
"learning_rate": 0.0004946735954734652,
"loss": 6.0642,
"mean_token_accuracy": 0.17588913589715957,
"num_tokens": 17618509.0,
"step": 6945
},
{
"entropy": 6.3537209033966064,
"epoch": 0.8020773225620311,
"grad_norm": 0.83203125,
"learning_rate": 0.0004946646663264811,
"loss": 6.2486,
"mean_token_accuracy": 0.1653627946972847,
"num_tokens": 17630652.0,
"step": 6950
},
{
"entropy": 6.327255773544311,
"epoch": 0.8026543566070398,
"grad_norm": 0.9140625,
"learning_rate": 0.0004946557297911852,
"loss": 6.079,
"mean_token_accuracy": 0.17674667686223983,
"num_tokens": 17641935.0,
"step": 6955
},
{
"entropy": 6.3670917510986325,
"epoch": 0.8032313906520485,
"grad_norm": 0.8046875,
"learning_rate": 0.0004946467858678777,
"loss": 6.2835,
"mean_token_accuracy": 0.15880746990442277,
"num_tokens": 17654043.0,
"step": 6960
},
{
"entropy": 6.446195363998413,
"epoch": 0.8038084246970572,
"grad_norm": 0.8125,
"learning_rate": 0.0004946378345568597,
"loss": 6.2089,
"mean_token_accuracy": 0.16398643404245378,
"num_tokens": 17666566.0,
"step": 6965
},
{
"entropy": 6.234373331069946,
"epoch": 0.8043854587420658,
"grad_norm": 0.9921875,
"learning_rate": 0.0004946288758584324,
"loss": 6.0275,
"mean_token_accuracy": 0.17151330709457396,
"num_tokens": 17678463.0,
"step": 6970
},
{
"entropy": 6.405322551727295,
"epoch": 0.8049624927870744,
"grad_norm": 0.79296875,
"learning_rate": 0.0004946199097728968,
"loss": 6.2667,
"mean_token_accuracy": 0.16178204566240312,
"num_tokens": 17691663.0,
"step": 6975
},
{
"entropy": 6.395699882507325,
"epoch": 0.8055395268320831,
"grad_norm": 0.8515625,
"learning_rate": 0.0004946109363005548,
"loss": 6.2324,
"mean_token_accuracy": 0.16503103375434874,
"num_tokens": 17705048.0,
"step": 6980
},
{
"entropy": 6.363728809356689,
"epoch": 0.8061165608770917,
"grad_norm": 0.87109375,
"learning_rate": 0.000494601955441708,
"loss": 6.211,
"mean_token_accuracy": 0.16915686428546906,
"num_tokens": 17717865.0,
"step": 6985
},
{
"entropy": 6.375814628601074,
"epoch": 0.8066935949221004,
"grad_norm": 0.79296875,
"learning_rate": 0.0004945929671966585,
"loss": 6.2272,
"mean_token_accuracy": 0.16043869405984879,
"num_tokens": 17730568.0,
"step": 6990
},
{
"entropy": 6.314331912994385,
"epoch": 0.8072706289671091,
"grad_norm": 0.80859375,
"learning_rate": 0.0004945839715657085,
"loss": 6.1815,
"mean_token_accuracy": 0.16977579444646834,
"num_tokens": 17743303.0,
"step": 6995
},
{
"entropy": 6.427794599533081,
"epoch": 0.8078476630121177,
"grad_norm": 0.78125,
"learning_rate": 0.0004945749685491607,
"loss": 6.2577,
"mean_token_accuracy": 0.1618587702512741,
"num_tokens": 17756552.0,
"step": 7000
},
{
"entropy": 6.3551558494567875,
"epoch": 0.8084246970571264,
"grad_norm": 0.7890625,
"learning_rate": 0.0004945659581473181,
"loss": 6.2785,
"mean_token_accuracy": 0.16330808997154236,
"num_tokens": 17770783.0,
"step": 7005
},
{
"entropy": 6.383954095840454,
"epoch": 0.809001731102135,
"grad_norm": 0.84375,
"learning_rate": 0.0004945569403604833,
"loss": 6.1562,
"mean_token_accuracy": 0.16762078553438187,
"num_tokens": 17784573.0,
"step": 7010
},
{
"entropy": 6.424306869506836,
"epoch": 0.8095787651471437,
"grad_norm": 0.84765625,
"learning_rate": 0.0004945479151889601,
"loss": 6.1923,
"mean_token_accuracy": 0.16073544770479203,
"num_tokens": 17797304.0,
"step": 7015
},
{
"entropy": 6.291573762893677,
"epoch": 0.8101557991921523,
"grad_norm": 0.76171875,
"learning_rate": 0.0004945388826330517,
"loss": 6.0622,
"mean_token_accuracy": 0.17693958580493926,
"num_tokens": 17810401.0,
"step": 7020
},
{
"entropy": 6.305244064331054,
"epoch": 0.8107328332371609,
"grad_norm": 0.7421875,
"learning_rate": 0.0004945298426930621,
"loss": 6.1434,
"mean_token_accuracy": 0.17025542855262757,
"num_tokens": 17824120.0,
"step": 7025
},
{
"entropy": 6.380581474304199,
"epoch": 0.8113098672821697,
"grad_norm": 0.82421875,
"learning_rate": 0.0004945207953692952,
"loss": 6.1799,
"mean_token_accuracy": 0.16469264179468154,
"num_tokens": 17836865.0,
"step": 7030
},
{
"entropy": 6.311848449707031,
"epoch": 0.8118869013271783,
"grad_norm": 0.84375,
"learning_rate": 0.0004945117406620556,
"loss": 6.2074,
"mean_token_accuracy": 0.1644309490919113,
"num_tokens": 17849128.0,
"step": 7035
},
{
"entropy": 6.367767953872681,
"epoch": 0.812463935372187,
"grad_norm": 0.8359375,
"learning_rate": 0.0004945026785716474,
"loss": 6.1677,
"mean_token_accuracy": 0.1677388444542885,
"num_tokens": 17861151.0,
"step": 7040
},
{
"entropy": 6.327512121200561,
"epoch": 0.8130409694171956,
"grad_norm": 0.875,
"learning_rate": 0.0004944936090983757,
"loss": 6.1501,
"mean_token_accuracy": 0.1745579496026039,
"num_tokens": 17874043.0,
"step": 7045
},
{
"entropy": 6.328273725509644,
"epoch": 0.8136180034622043,
"grad_norm": 0.8984375,
"learning_rate": 0.0004944845322425455,
"loss": 6.1685,
"mean_token_accuracy": 0.1651006817817688,
"num_tokens": 17887176.0,
"step": 7050
},
{
"entropy": 6.4299534320831295,
"epoch": 0.8141950375072129,
"grad_norm": 0.8359375,
"learning_rate": 0.0004944754480044621,
"loss": 6.2311,
"mean_token_accuracy": 0.1592079609632492,
"num_tokens": 17899708.0,
"step": 7055
},
{
"entropy": 6.320512151718139,
"epoch": 0.8147720715522215,
"grad_norm": 0.85546875,
"learning_rate": 0.0004944663563844311,
"loss": 6.1268,
"mean_token_accuracy": 0.17118301540613173,
"num_tokens": 17912303.0,
"step": 7060
},
{
"entropy": 6.379102373123169,
"epoch": 0.8153491055972303,
"grad_norm": 0.7734375,
"learning_rate": 0.0004944572573827581,
"loss": 6.2577,
"mean_token_accuracy": 0.15897056609392166,
"num_tokens": 17925963.0,
"step": 7065
},
{
"entropy": 6.393832969665527,
"epoch": 0.8159261396422389,
"grad_norm": 0.890625,
"learning_rate": 0.0004944481509997494,
"loss": 6.2386,
"mean_token_accuracy": 0.16631796956062317,
"num_tokens": 17938786.0,
"step": 7070
},
{
"entropy": 6.40287504196167,
"epoch": 0.8165031736872476,
"grad_norm": 0.8203125,
"learning_rate": 0.000494439037235711,
"loss": 6.1682,
"mean_token_accuracy": 0.1683258295059204,
"num_tokens": 17950646.0,
"step": 7075
},
{
"entropy": 6.379344654083252,
"epoch": 0.8170802077322562,
"grad_norm": 0.76953125,
"learning_rate": 0.0004944299160909495,
"loss": 6.1901,
"mean_token_accuracy": 0.16389957815408707,
"num_tokens": 17963115.0,
"step": 7080
},
{
"entropy": 6.328940725326538,
"epoch": 0.8176572417772648,
"grad_norm": 0.8125,
"learning_rate": 0.000494420787565772,
"loss": 6.1938,
"mean_token_accuracy": 0.1683003857731819,
"num_tokens": 17976068.0,
"step": 7085
},
{
"entropy": 6.358946800231934,
"epoch": 0.8182342758222735,
"grad_norm": 0.85546875,
"learning_rate": 0.0004944116516604852,
"loss": 6.1481,
"mean_token_accuracy": 0.17259256839752196,
"num_tokens": 17988487.0,
"step": 7090
},
{
"entropy": 6.357552099227905,
"epoch": 0.8188113098672821,
"grad_norm": 0.87109375,
"learning_rate": 0.0004944025083753965,
"loss": 6.1936,
"mean_token_accuracy": 0.1638806402683258,
"num_tokens": 18001664.0,
"step": 7095
},
{
"entropy": 6.372899293899536,
"epoch": 0.8193883439122909,
"grad_norm": 0.82421875,
"learning_rate": 0.0004943933577108133,
"loss": 6.2075,
"mean_token_accuracy": 0.16567729860544206,
"num_tokens": 18014384.0,
"step": 7100
},
{
"entropy": 6.366517114639282,
"epoch": 0.8199653779572995,
"grad_norm": 0.984375,
"learning_rate": 0.0004943841996670436,
"loss": 6.1411,
"mean_token_accuracy": 0.16836489140987396,
"num_tokens": 18027516.0,
"step": 7105
},
{
"entropy": 6.323846769332886,
"epoch": 0.8205424120023082,
"grad_norm": 0.890625,
"learning_rate": 0.0004943750342443953,
"loss": 6.0784,
"mean_token_accuracy": 0.1761428400874138,
"num_tokens": 18039771.0,
"step": 7110
},
{
"entropy": 6.335892629623413,
"epoch": 0.8211194460473168,
"grad_norm": 0.86328125,
"learning_rate": 0.0004943658614431767,
"loss": 6.2552,
"mean_token_accuracy": 0.1610368736088276,
"num_tokens": 18052620.0,
"step": 7115
},
{
"entropy": 6.236982154846191,
"epoch": 0.8216964800923254,
"grad_norm": 0.828125,
"learning_rate": 0.0004943566812636963,
"loss": 6.0156,
"mean_token_accuracy": 0.17757391780614853,
"num_tokens": 18065723.0,
"step": 7120
},
{
"entropy": 6.309774351119995,
"epoch": 0.8222735141373341,
"grad_norm": 0.8671875,
"learning_rate": 0.000494347493706263,
"loss": 6.1139,
"mean_token_accuracy": 0.16816986054182054,
"num_tokens": 18078335.0,
"step": 7125
},
{
"entropy": 6.3064737796783445,
"epoch": 0.8228505481823427,
"grad_norm": 0.85546875,
"learning_rate": 0.0004943382987711856,
"loss": 6.1359,
"mean_token_accuracy": 0.17049853056669234,
"num_tokens": 18089943.0,
"step": 7130
},
{
"entropy": 6.325173425674438,
"epoch": 0.8234275822273515,
"grad_norm": 0.828125,
"learning_rate": 0.0004943290964587734,
"loss": 6.0903,
"mean_token_accuracy": 0.16795280575752258,
"num_tokens": 18101843.0,
"step": 7135
},
{
"entropy": 6.241851854324341,
"epoch": 0.8240046162723601,
"grad_norm": 0.78515625,
"learning_rate": 0.0004943198867693361,
"loss": 6.0009,
"mean_token_accuracy": 0.18136388063430786,
"num_tokens": 18113170.0,
"step": 7140
},
{
"entropy": 6.325794649124146,
"epoch": 0.8245816503173687,
"grad_norm": 0.85546875,
"learning_rate": 0.0004943106697031833,
"loss": 6.1785,
"mean_token_accuracy": 0.16728792637586593,
"num_tokens": 18124936.0,
"step": 7145
},
{
"entropy": 6.3345221996307375,
"epoch": 0.8251586843623774,
"grad_norm": 0.85546875,
"learning_rate": 0.0004943014452606251,
"loss": 6.1313,
"mean_token_accuracy": 0.16424822360277175,
"num_tokens": 18137894.0,
"step": 7150
},
{
"entropy": 6.357614278793335,
"epoch": 0.825735718407386,
"grad_norm": 0.88671875,
"learning_rate": 0.0004942922134419717,
"loss": 6.1923,
"mean_token_accuracy": 0.16184918731451034,
"num_tokens": 18149170.0,
"step": 7155
},
{
"entropy": 6.346024084091186,
"epoch": 0.8263127524523947,
"grad_norm": 0.8203125,
"learning_rate": 0.0004942829742475336,
"loss": 6.2033,
"mean_token_accuracy": 0.1654985547065735,
"num_tokens": 18160694.0,
"step": 7160
},
{
"entropy": 6.423761940002441,
"epoch": 0.8268897864974033,
"grad_norm": 0.8125,
"learning_rate": 0.0004942737276776217,
"loss": 6.3124,
"mean_token_accuracy": 0.16496190130710603,
"num_tokens": 18173619.0,
"step": 7165
},
{
"entropy": 6.449676036834717,
"epoch": 0.827466820542412,
"grad_norm": 1.0390625,
"learning_rate": 0.0004942644737325468,
"loss": 6.2614,
"mean_token_accuracy": 0.16386640071868896,
"num_tokens": 18186751.0,
"step": 7170
},
{
"entropy": 6.325207805633545,
"epoch": 0.8280438545874207,
"grad_norm": 0.9453125,
"learning_rate": 0.0004942552124126202,
"loss": 6.1525,
"mean_token_accuracy": 0.16702402085065843,
"num_tokens": 18198701.0,
"step": 7175
},
{
"entropy": 6.350707149505615,
"epoch": 0.8286208886324293,
"grad_norm": 0.84765625,
"learning_rate": 0.0004942459437181535,
"loss": 6.1251,
"mean_token_accuracy": 0.17409539818763733,
"num_tokens": 18211422.0,
"step": 7180
},
{
"entropy": 6.320394611358642,
"epoch": 0.829197922677438,
"grad_norm": 0.90234375,
"learning_rate": 0.0004942366676494584,
"loss": 6.2097,
"mean_token_accuracy": 0.1659229278564453,
"num_tokens": 18223427.0,
"step": 7185
},
{
"entropy": 6.277689838409424,
"epoch": 0.8297749567224466,
"grad_norm": 0.859375,
"learning_rate": 0.0004942273842068469,
"loss": 6.1679,
"mean_token_accuracy": 0.17429596334695815,
"num_tokens": 18235234.0,
"step": 7190
},
{
"entropy": 6.374852085113526,
"epoch": 0.8303519907674553,
"grad_norm": 0.890625,
"learning_rate": 0.0004942180933906311,
"loss": 6.0967,
"mean_token_accuracy": 0.17455530613660813,
"num_tokens": 18247669.0,
"step": 7195
},
{
"entropy": 6.261248683929443,
"epoch": 0.8309290248124639,
"grad_norm": 0.90234375,
"learning_rate": 0.0004942087952011237,
"loss": 6.0792,
"mean_token_accuracy": 0.17736732363700866,
"num_tokens": 18260987.0,
"step": 7200
},
{
"entropy": 6.378260564804077,
"epoch": 0.8315060588574726,
"grad_norm": 0.9140625,
"learning_rate": 0.0004941994896386374,
"loss": 6.1867,
"mean_token_accuracy": 0.1577957659959793,
"num_tokens": 18274805.0,
"step": 7205
},
{
"entropy": 6.360244607925415,
"epoch": 0.8320830929024813,
"grad_norm": 0.78125,
"learning_rate": 0.0004941901767034851,
"loss": 6.1673,
"mean_token_accuracy": 0.16796040832996367,
"num_tokens": 18286438.0,
"step": 7210
},
{
"entropy": 6.357167530059814,
"epoch": 0.8326601269474899,
"grad_norm": 0.8671875,
"learning_rate": 0.0004941808563959802,
"loss": 6.1856,
"mean_token_accuracy": 0.16499463766813277,
"num_tokens": 18299310.0,
"step": 7215
},
{
"entropy": 6.29741940498352,
"epoch": 0.8332371609924986,
"grad_norm": 0.8984375,
"learning_rate": 0.0004941715287164359,
"loss": 6.1245,
"mean_token_accuracy": 0.16438130140304566,
"num_tokens": 18311286.0,
"step": 7220
},
{
"entropy": 6.32746958732605,
"epoch": 0.8338141950375072,
"grad_norm": 0.75,
"learning_rate": 0.0004941621936651661,
"loss": 6.1803,
"mean_token_accuracy": 0.16897405236959456,
"num_tokens": 18325759.0,
"step": 7225
},
{
"entropy": 6.34986686706543,
"epoch": 0.8343912290825158,
"grad_norm": 0.90234375,
"learning_rate": 0.0004941528512424849,
"loss": 6.1252,
"mean_token_accuracy": 0.1749942496418953,
"num_tokens": 18338907.0,
"step": 7230
},
{
"entropy": 6.402791833877563,
"epoch": 0.8349682631275245,
"grad_norm": 0.859375,
"learning_rate": 0.0004941435014487064,
"loss": 6.1905,
"mean_token_accuracy": 0.16975402384996413,
"num_tokens": 18351760.0,
"step": 7235
},
{
"entropy": 6.343279838562012,
"epoch": 0.8355452971725332,
"grad_norm": 0.91796875,
"learning_rate": 0.000494134144284145,
"loss": 6.1873,
"mean_token_accuracy": 0.1620405212044716,
"num_tokens": 18363348.0,
"step": 7240
},
{
"entropy": 6.356585359573364,
"epoch": 0.8361223312175419,
"grad_norm": 0.78125,
"learning_rate": 0.0004941247797491156,
"loss": 6.2152,
"mean_token_accuracy": 0.16385446041822432,
"num_tokens": 18375233.0,
"step": 7245
},
{
"entropy": 6.361816120147705,
"epoch": 0.8366993652625505,
"grad_norm": 0.80859375,
"learning_rate": 0.0004941154078439329,
"loss": 6.129,
"mean_token_accuracy": 0.16602645069360733,
"num_tokens": 18388001.0,
"step": 7250
},
{
"entropy": 6.210663080215454,
"epoch": 0.8372763993075591,
"grad_norm": 0.890625,
"learning_rate": 0.0004941060285689126,
"loss": 6.0004,
"mean_token_accuracy": 0.1854765772819519,
"num_tokens": 18399647.0,
"step": 7255
},
{
"entropy": 6.286527299880982,
"epoch": 0.8378534333525678,
"grad_norm": 0.87109375,
"learning_rate": 0.0004940966419243695,
"loss": 6.1585,
"mean_token_accuracy": 0.17007975280284882,
"num_tokens": 18412441.0,
"step": 7260
},
{
"entropy": 6.3775622844696045,
"epoch": 0.8384304673975764,
"grad_norm": 0.85546875,
"learning_rate": 0.00049408724791062,
"loss": 6.1714,
"mean_token_accuracy": 0.17022394239902497,
"num_tokens": 18425232.0,
"step": 7265
},
{
"entropy": 6.389389944076538,
"epoch": 0.8390075014425851,
"grad_norm": 0.87890625,
"learning_rate": 0.0004940778465279797,
"loss": 6.1677,
"mean_token_accuracy": 0.1612836867570877,
"num_tokens": 18438148.0,
"step": 7270
},
{
"entropy": 6.276881313323974,
"epoch": 0.8395845354875938,
"grad_norm": 0.875,
"learning_rate": 0.0004940684377767647,
"loss": 6.006,
"mean_token_accuracy": 0.17631301879882813,
"num_tokens": 18449955.0,
"step": 7275
},
{
"entropy": 6.251585054397583,
"epoch": 0.8401615695326025,
"grad_norm": 0.8671875,
"learning_rate": 0.0004940590216572916,
"loss": 6.105,
"mean_token_accuracy": 0.17243984788656236,
"num_tokens": 18463125.0,
"step": 7280
},
{
"entropy": 6.367015790939331,
"epoch": 0.8407386035776111,
"grad_norm": 0.78515625,
"learning_rate": 0.0004940495981698772,
"loss": 6.2058,
"mean_token_accuracy": 0.16317611634731294,
"num_tokens": 18475280.0,
"step": 7285
},
{
"entropy": 6.35017147064209,
"epoch": 0.8413156376226197,
"grad_norm": 0.75390625,
"learning_rate": 0.0004940401673148384,
"loss": 6.162,
"mean_token_accuracy": 0.16571533679962158,
"num_tokens": 18487406.0,
"step": 7290
},
{
"entropy": 6.275267219543457,
"epoch": 0.8418926716676284,
"grad_norm": 0.9140625,
"learning_rate": 0.0004940307290924923,
"loss": 6.1234,
"mean_token_accuracy": 0.17094551920890808,
"num_tokens": 18500288.0,
"step": 7295
},
{
"entropy": 6.404674673080445,
"epoch": 0.842469705712637,
"grad_norm": 0.85546875,
"learning_rate": 0.0004940212835031565,
"loss": 6.1988,
"mean_token_accuracy": 0.1616864025592804,
"num_tokens": 18514254.0,
"step": 7300
},
{
"entropy": 6.36431851387024,
"epoch": 0.8430467397576457,
"grad_norm": 0.90625,
"learning_rate": 0.0004940118305471486,
"loss": 6.2378,
"mean_token_accuracy": 0.15897418856620787,
"num_tokens": 18526536.0,
"step": 7305
},
{
"entropy": 6.427651834487915,
"epoch": 0.8436237738026544,
"grad_norm": 0.8515625,
"learning_rate": 0.0004940023702247866,
"loss": 6.1441,
"mean_token_accuracy": 0.16780431121587752,
"num_tokens": 18540137.0,
"step": 7310
},
{
"entropy": 6.333643674850464,
"epoch": 0.844200807847663,
"grad_norm": 0.8125,
"learning_rate": 0.0004939929025363886,
"loss": 6.1381,
"mean_token_accuracy": 0.17276106625795365,
"num_tokens": 18552975.0,
"step": 7315
},
{
"entropy": 6.247806549072266,
"epoch": 0.8447778418926717,
"grad_norm": 0.85546875,
"learning_rate": 0.0004939834274822731,
"loss": 6.0625,
"mean_token_accuracy": 0.17385534346103668,
"num_tokens": 18564653.0,
"step": 7320
},
{
"entropy": 6.364656734466553,
"epoch": 0.8453548759376803,
"grad_norm": 0.8203125,
"learning_rate": 0.0004939739450627588,
"loss": 6.2033,
"mean_token_accuracy": 0.16594440788030623,
"num_tokens": 18577226.0,
"step": 7325
},
{
"entropy": 6.347729778289795,
"epoch": 0.845931909982689,
"grad_norm": 0.80078125,
"learning_rate": 0.0004939644552781647,
"loss": 6.1359,
"mean_token_accuracy": 0.16554813832044601,
"num_tokens": 18591206.0,
"step": 7330
},
{
"entropy": 6.305720424652099,
"epoch": 0.8465089440276976,
"grad_norm": 0.875,
"learning_rate": 0.0004939549581288099,
"loss": 6.138,
"mean_token_accuracy": 0.17282395362854003,
"num_tokens": 18604182.0,
"step": 7335
},
{
"entropy": 6.379542970657349,
"epoch": 0.8470859780727062,
"grad_norm": 0.83203125,
"learning_rate": 0.0004939454536150138,
"loss": 6.2233,
"mean_token_accuracy": 0.16193835139274598,
"num_tokens": 18615830.0,
"step": 7340
},
{
"entropy": 6.351224136352539,
"epoch": 0.847663012117715,
"grad_norm": 0.828125,
"learning_rate": 0.000493935941737096,
"loss": 6.1833,
"mean_token_accuracy": 0.16340697109699248,
"num_tokens": 18628372.0,
"step": 7345
},
{
"entropy": 6.373519563674927,
"epoch": 0.8482400461627236,
"grad_norm": 0.8046875,
"learning_rate": 0.0004939264224953768,
"loss": 6.2525,
"mean_token_accuracy": 0.1579662710428238,
"num_tokens": 18641470.0,
"step": 7350
},
{
"entropy": 6.371709108352661,
"epoch": 0.8488170802077323,
"grad_norm": 0.8046875,
"learning_rate": 0.000493916895890176,
"loss": 6.0974,
"mean_token_accuracy": 0.1789488598704338,
"num_tokens": 18654660.0,
"step": 7355
},
{
"entropy": 6.333353853225708,
"epoch": 0.8493941142527409,
"grad_norm": 0.83203125,
"learning_rate": 0.000493907361921814,
"loss": 6.1924,
"mean_token_accuracy": 0.16414279490709305,
"num_tokens": 18665626.0,
"step": 7360
},
{
"entropy": 6.318211460113526,
"epoch": 0.8499711482977496,
"grad_norm": 0.85546875,
"learning_rate": 0.0004938978205906118,
"loss": 6.1137,
"mean_token_accuracy": 0.17439971268177032,
"num_tokens": 18677175.0,
"step": 7365
},
{
"entropy": 6.343401908874512,
"epoch": 0.8505481823427582,
"grad_norm": 0.828125,
"learning_rate": 0.0004938882718968901,
"loss": 6.1332,
"mean_token_accuracy": 0.17032305002212525,
"num_tokens": 18689005.0,
"step": 7370
},
{
"entropy": 6.2724262237548825,
"epoch": 0.8511252163877668,
"grad_norm": 0.80078125,
"learning_rate": 0.0004938787158409702,
"loss": 6.181,
"mean_token_accuracy": 0.1706418961286545,
"num_tokens": 18702085.0,
"step": 7375
},
{
"entropy": 6.351395463943481,
"epoch": 0.8517022504327756,
"grad_norm": 0.80859375,
"learning_rate": 0.0004938691524231733,
"loss": 6.1819,
"mean_token_accuracy": 0.16577364206314088,
"num_tokens": 18715064.0,
"step": 7380
},
{
"entropy": 6.367648410797119,
"epoch": 0.8522792844777842,
"grad_norm": 0.88671875,
"learning_rate": 0.0004938595816438212,
"loss": 6.1981,
"mean_token_accuracy": 0.16841965466737746,
"num_tokens": 18728105.0,
"step": 7385
},
{
"entropy": 6.286236524581909,
"epoch": 0.8528563185227929,
"grad_norm": 0.8515625,
"learning_rate": 0.0004938500035032358,
"loss": 6.1076,
"mean_token_accuracy": 0.1745881989598274,
"num_tokens": 18741021.0,
"step": 7390
},
{
"entropy": 6.283361291885376,
"epoch": 0.8534333525678015,
"grad_norm": 0.96875,
"learning_rate": 0.0004938404180017392,
"loss": 6.1466,
"mean_token_accuracy": 0.17935479879379274,
"num_tokens": 18753271.0,
"step": 7395
},
{
"entropy": 6.326036405563355,
"epoch": 0.8540103866128101,
"grad_norm": 0.890625,
"learning_rate": 0.0004938308251396539,
"loss": 6.1516,
"mean_token_accuracy": 0.167026287317276,
"num_tokens": 18765631.0,
"step": 7400
},
{
"entropy": 6.335333061218262,
"epoch": 0.8545874206578188,
"grad_norm": 0.82421875,
"learning_rate": 0.0004938212249173024,
"loss": 6.1485,
"mean_token_accuracy": 0.1670267701148987,
"num_tokens": 18778472.0,
"step": 7405
},
{
"entropy": 6.418694257736206,
"epoch": 0.8551644547028274,
"grad_norm": 0.81640625,
"learning_rate": 0.0004938116173350078,
"loss": 6.1868,
"mean_token_accuracy": 0.16128545999526978,
"num_tokens": 18791564.0,
"step": 7410
},
{
"entropy": 6.310502290725708,
"epoch": 0.8557414887478362,
"grad_norm": 0.7578125,
"learning_rate": 0.0004938020023930932,
"loss": 6.1428,
"mean_token_accuracy": 0.16949567347764968,
"num_tokens": 18804749.0,
"step": 7415
},
{
"entropy": 6.373205137252808,
"epoch": 0.8563185227928448,
"grad_norm": 0.87109375,
"learning_rate": 0.0004937923800918817,
"loss": 6.1656,
"mean_token_accuracy": 0.16341724693775178,
"num_tokens": 18816863.0,
"step": 7420
},
{
"entropy": 6.354863500595092,
"epoch": 0.8568955568378535,
"grad_norm": 0.8515625,
"learning_rate": 0.0004937827504316974,
"loss": 6.1286,
"mean_token_accuracy": 0.17136932164430618,
"num_tokens": 18831543.0,
"step": 7425
},
{
"entropy": 6.3266009330749515,
"epoch": 0.8574725908828621,
"grad_norm": 0.83203125,
"learning_rate": 0.0004937731134128639,
"loss": 6.1649,
"mean_token_accuracy": 0.16626855731010437,
"num_tokens": 18844242.0,
"step": 7430
},
{
"entropy": 6.309177446365356,
"epoch": 0.8580496249278707,
"grad_norm": 0.94140625,
"learning_rate": 0.0004937634690357054,
"loss": 6.1785,
"mean_token_accuracy": 0.16604990065097808,
"num_tokens": 18856213.0,
"step": 7435
},
{
"entropy": 6.389643764495849,
"epoch": 0.8586266589728794,
"grad_norm": 0.80078125,
"learning_rate": 0.0004937538173005462,
"loss": 6.1669,
"mean_token_accuracy": 0.16924736946821212,
"num_tokens": 18869680.0,
"step": 7440
},
{
"entropy": 6.371203804016114,
"epoch": 0.859203693017888,
"grad_norm": 0.76953125,
"learning_rate": 0.0004937441582077111,
"loss": 6.1415,
"mean_token_accuracy": 0.17059220224618912,
"num_tokens": 18882360.0,
"step": 7445
},
{
"entropy": 6.28338508605957,
"epoch": 0.8597807270628968,
"grad_norm": 0.89453125,
"learning_rate": 0.0004937344917575249,
"loss": 6.1129,
"mean_token_accuracy": 0.16984111964702606,
"num_tokens": 18895138.0,
"step": 7450
},
{
"entropy": 6.285325384140014,
"epoch": 0.8603577611079054,
"grad_norm": 0.76953125,
"learning_rate": 0.0004937248179503127,
"loss": 6.1584,
"mean_token_accuracy": 0.16774938404560089,
"num_tokens": 18908849.0,
"step": 7455
},
{
"entropy": 6.353943300247193,
"epoch": 0.860934795152914,
"grad_norm": 0.859375,
"learning_rate": 0.0004937151367863998,
"loss": 6.1267,
"mean_token_accuracy": 0.17231214940547943,
"num_tokens": 18920757.0,
"step": 7460
},
{
"entropy": 6.2311420917510985,
"epoch": 0.8615118291979227,
"grad_norm": 0.859375,
"learning_rate": 0.000493705448266112,
"loss": 6.0605,
"mean_token_accuracy": 0.17295785695314408,
"num_tokens": 18933239.0,
"step": 7465
},
{
"entropy": 6.357706356048584,
"epoch": 0.8620888632429313,
"grad_norm": 0.87109375,
"learning_rate": 0.0004936957523897752,
"loss": 6.177,
"mean_token_accuracy": 0.16270439624786376,
"num_tokens": 18945174.0,
"step": 7470
},
{
"entropy": 6.391672039031983,
"epoch": 0.86266589728794,
"grad_norm": 0.80859375,
"learning_rate": 0.0004936860491577153,
"loss": 6.2446,
"mean_token_accuracy": 0.15656707137823106,
"num_tokens": 18957478.0,
"step": 7475
},
{
"entropy": 6.338565874099731,
"epoch": 0.8632429313329486,
"grad_norm": 0.8046875,
"learning_rate": 0.0004936763385702588,
"loss": 6.1722,
"mean_token_accuracy": 0.16844886690378189,
"num_tokens": 18969461.0,
"step": 7480
},
{
"entropy": 6.447391748428345,
"epoch": 0.8638199653779572,
"grad_norm": 0.88671875,
"learning_rate": 0.0004936666206277322,
"loss": 6.1648,
"mean_token_accuracy": 0.16605869233608245,
"num_tokens": 18981891.0,
"step": 7485
},
{
"entropy": 6.316598176956177,
"epoch": 0.864396999422966,
"grad_norm": 0.86328125,
"learning_rate": 0.0004936568953304624,
"loss": 6.039,
"mean_token_accuracy": 0.17837050408124924,
"num_tokens": 18993376.0,
"step": 7490
},
{
"entropy": 6.228720474243164,
"epoch": 0.8649740334679746,
"grad_norm": 0.8359375,
"learning_rate": 0.0004936471626787766,
"loss": 6.1226,
"mean_token_accuracy": 0.17042396664619447,
"num_tokens": 19005485.0,
"step": 7495
},
{
"entropy": 6.285957956314087,
"epoch": 0.8655510675129833,
"grad_norm": 0.9296875,
"learning_rate": 0.0004936374226730022,
"loss": 6.1756,
"mean_token_accuracy": 0.16250619143247605,
"num_tokens": 19018509.0,
"step": 7500
},
{
"entropy": 6.430623006820679,
"epoch": 0.8661281015579919,
"grad_norm": 0.8828125,
"learning_rate": 0.0004936276753134666,
"loss": 6.1405,
"mean_token_accuracy": 0.16987382471561432,
"num_tokens": 19030595.0,
"step": 7505
},
{
"entropy": 6.373871994018555,
"epoch": 0.8667051356030006,
"grad_norm": 0.8515625,
"learning_rate": 0.0004936179206004976,
"loss": 6.2206,
"mean_token_accuracy": 0.16224805116653443,
"num_tokens": 19043441.0,
"step": 7510
},
{
"entropy": 6.303015947341919,
"epoch": 0.8672821696480092,
"grad_norm": 0.859375,
"learning_rate": 0.0004936081585344236,
"loss": 6.1184,
"mean_token_accuracy": 0.16949329972267152,
"num_tokens": 19055761.0,
"step": 7515
},
{
"entropy": 6.329971075057983,
"epoch": 0.8678592036930178,
"grad_norm": 0.8515625,
"learning_rate": 0.0004935983891155727,
"loss": 6.1628,
"mean_token_accuracy": 0.16706838011741637,
"num_tokens": 19068211.0,
"step": 7520
},
{
"entropy": 6.333680200576782,
"epoch": 0.8684362377380266,
"grad_norm": 0.8828125,
"learning_rate": 0.0004935886123442737,
"loss": 6.1094,
"mean_token_accuracy": 0.16620445251464844,
"num_tokens": 19081487.0,
"step": 7525
},
{
"entropy": 6.3789692401885985,
"epoch": 0.8690132717830352,
"grad_norm": 0.77734375,
"learning_rate": 0.0004935788282208551,
"loss": 6.1258,
"mean_token_accuracy": 0.16772944927215577,
"num_tokens": 19094779.0,
"step": 7530
},
{
"entropy": 6.302552080154419,
"epoch": 0.8695903058280439,
"grad_norm": 0.7890625,
"learning_rate": 0.0004935690367456464,
"loss": 6.1898,
"mean_token_accuracy": 0.1666984111070633,
"num_tokens": 19106999.0,
"step": 7535
},
{
"entropy": 6.372557210922241,
"epoch": 0.8701673398730525,
"grad_norm": 0.921875,
"learning_rate": 0.0004935592379189766,
"loss": 6.1381,
"mean_token_accuracy": 0.1672473892569542,
"num_tokens": 19119570.0,
"step": 7540
},
{
"entropy": 6.3990599632263185,
"epoch": 0.8707443739180611,
"grad_norm": 0.875,
"learning_rate": 0.0004935494317411754,
"loss": 6.1411,
"mean_token_accuracy": 0.16526359766721727,
"num_tokens": 19131101.0,
"step": 7545
},
{
"entropy": 6.244806814193725,
"epoch": 0.8713214079630698,
"grad_norm": 0.83203125,
"learning_rate": 0.0004935396182125726,
"loss": 6.1125,
"mean_token_accuracy": 0.16791403144598008,
"num_tokens": 19144159.0,
"step": 7550
},
{
"entropy": 6.387464237213135,
"epoch": 0.8718984420080784,
"grad_norm": 0.7734375,
"learning_rate": 0.0004935297973334983,
"loss": 6.1054,
"mean_token_accuracy": 0.1751489594578743,
"num_tokens": 19157034.0,
"step": 7555
},
{
"entropy": 6.3096170902252195,
"epoch": 0.8724754760530872,
"grad_norm": 0.8046875,
"learning_rate": 0.0004935199691042828,
"loss": 6.1511,
"mean_token_accuracy": 0.17347051203250885,
"num_tokens": 19170813.0,
"step": 7560
},
{
"entropy": 6.277336311340332,
"epoch": 0.8730525100980958,
"grad_norm": 0.8515625,
"learning_rate": 0.0004935101335252568,
"loss": 6.093,
"mean_token_accuracy": 0.17822133302688598,
"num_tokens": 19183060.0,
"step": 7565
},
{
"entropy": 6.308446788787842,
"epoch": 0.8736295441431045,
"grad_norm": 0.8515625,
"learning_rate": 0.0004935002905967509,
"loss": 6.0312,
"mean_token_accuracy": 0.17758481651544572,
"num_tokens": 19195615.0,
"step": 7570
},
{
"entropy": 6.345848464965821,
"epoch": 0.8742065781881131,
"grad_norm": 0.9453125,
"learning_rate": 0.0004934904403190963,
"loss": 6.1979,
"mean_token_accuracy": 0.16513141840696335,
"num_tokens": 19209531.0,
"step": 7575
},
{
"entropy": 6.288311004638672,
"epoch": 0.8747836122331217,
"grad_norm": 0.875,
"learning_rate": 0.0004934805826926242,
"loss": 6.1377,
"mean_token_accuracy": 0.17015064060688018,
"num_tokens": 19221658.0,
"step": 7580
},
{
"entropy": 6.373033761978149,
"epoch": 0.8753606462781304,
"grad_norm": 0.86328125,
"learning_rate": 0.0004934707177176663,
"loss": 6.1451,
"mean_token_accuracy": 0.16882509142160415,
"num_tokens": 19233491.0,
"step": 7585
},
{
"entropy": 6.36579999923706,
"epoch": 0.875937680323139,
"grad_norm": 0.875,
"learning_rate": 0.0004934608453945543,
"loss": 6.2309,
"mean_token_accuracy": 0.1647283762693405,
"num_tokens": 19246318.0,
"step": 7590
},
{
"entropy": 6.334402990341187,
"epoch": 0.8765147143681478,
"grad_norm": 0.80078125,
"learning_rate": 0.0004934509657236203,
"loss": 6.165,
"mean_token_accuracy": 0.16658859997987746,
"num_tokens": 19258735.0,
"step": 7595
},
{
"entropy": 6.417309999465942,
"epoch": 0.8770917484131564,
"grad_norm": 0.8671875,
"learning_rate": 0.0004934410787051965,
"loss": 6.2063,
"mean_token_accuracy": 0.16228333860635757,
"num_tokens": 19271640.0,
"step": 7600
},
{
"entropy": 6.300780439376831,
"epoch": 0.877668782458165,
"grad_norm": 0.9140625,
"learning_rate": 0.0004934311843396157,
"loss": 6.1118,
"mean_token_accuracy": 0.17644276469945908,
"num_tokens": 19283824.0,
"step": 7605
},
{
"entropy": 6.311488914489746,
"epoch": 0.8782458165031737,
"grad_norm": 0.83984375,
"learning_rate": 0.0004934212826272104,
"loss": 6.0685,
"mean_token_accuracy": 0.17539282739162446,
"num_tokens": 19298072.0,
"step": 7610
},
{
"entropy": 6.370089292526245,
"epoch": 0.8788228505481823,
"grad_norm": 0.76171875,
"learning_rate": 0.0004934113735683137,
"loss": 6.1521,
"mean_token_accuracy": 0.16698871701955795,
"num_tokens": 19311132.0,
"step": 7615
},
{
"entropy": 6.336485719680786,
"epoch": 0.879399884593191,
"grad_norm": 0.80859375,
"learning_rate": 0.000493401457163259,
"loss": 6.1213,
"mean_token_accuracy": 0.16742272078990936,
"num_tokens": 19324342.0,
"step": 7620
},
{
"entropy": 6.300466108322143,
"epoch": 0.8799769186381996,
"grad_norm": 0.8671875,
"learning_rate": 0.0004933915334123798,
"loss": 6.1194,
"mean_token_accuracy": 0.16596459597349167,
"num_tokens": 19336629.0,
"step": 7625
},
{
"entropy": 6.315993547439575,
"epoch": 0.8805539526832084,
"grad_norm": 0.8828125,
"learning_rate": 0.0004933816023160099,
"loss": 6.1693,
"mean_token_accuracy": 0.15985054075717925,
"num_tokens": 19348247.0,
"step": 7630
},
{
"entropy": 6.3706707000732425,
"epoch": 0.881130986728217,
"grad_norm": 0.84375,
"learning_rate": 0.0004933716638744832,
"loss": 6.0678,
"mean_token_accuracy": 0.17529698759317397,
"num_tokens": 19361230.0,
"step": 7635
},
{
"entropy": 6.294079065322876,
"epoch": 0.8817080207732256,
"grad_norm": 0.7890625,
"learning_rate": 0.000493361718088134,
"loss": 6.0966,
"mean_token_accuracy": 0.177022348344326,
"num_tokens": 19373803.0,
"step": 7640
},
{
"entropy": 6.273023843765259,
"epoch": 0.8822850548182343,
"grad_norm": 0.91015625,
"learning_rate": 0.000493351764957297,
"loss": 6.051,
"mean_token_accuracy": 0.17366888225078583,
"num_tokens": 19385960.0,
"step": 7645
},
{
"entropy": 6.286736965179443,
"epoch": 0.8828620888632429,
"grad_norm": 0.7890625,
"learning_rate": 0.0004933418044823068,
"loss": 6.1875,
"mean_token_accuracy": 0.1685008630156517,
"num_tokens": 19399649.0,
"step": 7650
},
{
"entropy": 6.313032007217407,
"epoch": 0.8834391229082516,
"grad_norm": 0.90625,
"learning_rate": 0.0004933318366634984,
"loss": 6.1604,
"mean_token_accuracy": 0.1714610606431961,
"num_tokens": 19413964.0,
"step": 7655
},
{
"entropy": 6.344218349456787,
"epoch": 0.8840161569532602,
"grad_norm": 0.90625,
"learning_rate": 0.0004933218615012072,
"loss": 6.1798,
"mean_token_accuracy": 0.1704179748892784,
"num_tokens": 19425715.0,
"step": 7660
},
{
"entropy": 6.287836360931396,
"epoch": 0.8845931909982689,
"grad_norm": 0.83984375,
"learning_rate": 0.0004933118789957687,
"loss": 6.169,
"mean_token_accuracy": 0.1677010625600815,
"num_tokens": 19438733.0,
"step": 7665
},
{
"entropy": 6.394447135925293,
"epoch": 0.8851702250432776,
"grad_norm": 0.78125,
"learning_rate": 0.0004933018891475186,
"loss": 6.2412,
"mean_token_accuracy": 0.1652499720454216,
"num_tokens": 19451924.0,
"step": 7670
},
{
"entropy": 6.310375356674195,
"epoch": 0.8857472590882862,
"grad_norm": 0.92578125,
"learning_rate": 0.0004932918919567927,
"loss": 6.0992,
"mean_token_accuracy": 0.1740289106965065,
"num_tokens": 19464331.0,
"step": 7675
},
{
"entropy": 6.290035152435303,
"epoch": 0.8863242931332949,
"grad_norm": 0.88671875,
"learning_rate": 0.0004932818874239275,
"loss": 6.1197,
"mean_token_accuracy": 0.16606962829828262,
"num_tokens": 19477127.0,
"step": 7680
},
{
"entropy": 6.310290384292602,
"epoch": 0.8869013271783035,
"grad_norm": 0.7890625,
"learning_rate": 0.0004932718755492595,
"loss": 6.089,
"mean_token_accuracy": 0.17557645589113235,
"num_tokens": 19490281.0,
"step": 7685
},
{
"entropy": 6.233003997802735,
"epoch": 0.8874783612233121,
"grad_norm": 0.81640625,
"learning_rate": 0.0004932618563331254,
"loss": 6.1019,
"mean_token_accuracy": 0.17658591121435166,
"num_tokens": 19503154.0,
"step": 7690
},
{
"entropy": 6.3082092761993405,
"epoch": 0.8880553952683208,
"grad_norm": 0.83203125,
"learning_rate": 0.0004932518297758622,
"loss": 6.2139,
"mean_token_accuracy": 0.17067087143659593,
"num_tokens": 19517363.0,
"step": 7695
},
{
"entropy": 6.33934588432312,
"epoch": 0.8886324293133295,
"grad_norm": 0.8125,
"learning_rate": 0.0004932417958778071,
"loss": 6.1096,
"mean_token_accuracy": 0.17136658877134323,
"num_tokens": 19529785.0,
"step": 7700
},
{
"entropy": 6.30635895729065,
"epoch": 0.8892094633583382,
"grad_norm": 1.109375,
"learning_rate": 0.0004932317546392976,
"loss": 6.173,
"mean_token_accuracy": 0.17024747133255005,
"num_tokens": 19542025.0,
"step": 7705
},
{
"entropy": 6.36300106048584,
"epoch": 0.8897864974033468,
"grad_norm": 0.890625,
"learning_rate": 0.0004932217060606714,
"loss": 6.1706,
"mean_token_accuracy": 0.16287696957588196,
"num_tokens": 19553969.0,
"step": 7710
},
{
"entropy": 6.350517797470093,
"epoch": 0.8903635314483554,
"grad_norm": 0.8671875,
"learning_rate": 0.0004932116501422665,
"loss": 6.0277,
"mean_token_accuracy": 0.17318399399518966,
"num_tokens": 19566568.0,
"step": 7715
},
{
"entropy": 6.2909540176391605,
"epoch": 0.8909405654933641,
"grad_norm": 0.8046875,
"learning_rate": 0.0004932015868844211,
"loss": 6.1513,
"mean_token_accuracy": 0.16795647740364075,
"num_tokens": 19579947.0,
"step": 7720
},
{
"entropy": 6.366373825073242,
"epoch": 0.8915175995383727,
"grad_norm": 0.84375,
"learning_rate": 0.0004931915162874738,
"loss": 6.1511,
"mean_token_accuracy": 0.16321159303188323,
"num_tokens": 19592457.0,
"step": 7725
},
{
"entropy": 6.3742717742919925,
"epoch": 0.8920946335833814,
"grad_norm": 1.2265625,
"learning_rate": 0.0004931814383517632,
"loss": 6.1153,
"mean_token_accuracy": 0.1651516377925873,
"num_tokens": 19604165.0,
"step": 7730
},
{
"entropy": 6.28874454498291,
"epoch": 0.8926716676283901,
"grad_norm": 0.8515625,
"learning_rate": 0.0004931713530776284,
"loss": 6.1508,
"mean_token_accuracy": 0.16553493738174438,
"num_tokens": 19617060.0,
"step": 7735
},
{
"entropy": 6.343902587890625,
"epoch": 0.8932487016733988,
"grad_norm": 0.890625,
"learning_rate": 0.0004931612604654083,
"loss": 6.171,
"mean_token_accuracy": 0.16089639961719512,
"num_tokens": 19628724.0,
"step": 7740
},
{
"entropy": 6.39453706741333,
"epoch": 0.8938257357184074,
"grad_norm": 0.8203125,
"learning_rate": 0.0004931511605154428,
"loss": 6.1865,
"mean_token_accuracy": 0.1709512323141098,
"num_tokens": 19642281.0,
"step": 7745
},
{
"entropy": 6.229511404037476,
"epoch": 0.894402769763416,
"grad_norm": 0.91015625,
"learning_rate": 0.0004931410532280711,
"loss": 6.0995,
"mean_token_accuracy": 0.17549168616533278,
"num_tokens": 19655503.0,
"step": 7750
},
{
"entropy": 6.351932621002197,
"epoch": 0.8949798038084247,
"grad_norm": 0.93359375,
"learning_rate": 0.0004931309386036337,
"loss": 6.1062,
"mean_token_accuracy": 0.16583985090255737,
"num_tokens": 19667758.0,
"step": 7755
},
{
"entropy": 6.333207893371582,
"epoch": 0.8955568378534333,
"grad_norm": 0.84375,
"learning_rate": 0.0004931208166424704,
"loss": 6.1706,
"mean_token_accuracy": 0.17188671976327896,
"num_tokens": 19679878.0,
"step": 7760
},
{
"entropy": 6.334847640991211,
"epoch": 0.896133871898442,
"grad_norm": 1.1171875,
"learning_rate": 0.0004931106873449219,
"loss": 6.2049,
"mean_token_accuracy": 0.16228338330984116,
"num_tokens": 19692278.0,
"step": 7765
},
{
"entropy": 6.2812264919281,
"epoch": 0.8967109059434507,
"grad_norm": 0.80859375,
"learning_rate": 0.0004931005507113285,
"loss": 6.0222,
"mean_token_accuracy": 0.18351362347602845,
"num_tokens": 19705319.0,
"step": 7770
},
{
"entropy": 6.338000631332397,
"epoch": 0.8972879399884593,
"grad_norm": 0.8203125,
"learning_rate": 0.0004930904067420317,
"loss": 6.0739,
"mean_token_accuracy": 0.16478749513626098,
"num_tokens": 19717908.0,
"step": 7775
},
{
"entropy": 6.409122657775879,
"epoch": 0.897864974033468,
"grad_norm": 0.8828125,
"learning_rate": 0.0004930802554373723,
"loss": 6.2259,
"mean_token_accuracy": 0.1664321541786194,
"num_tokens": 19730810.0,
"step": 7780
},
{
"entropy": 6.3239624977111815,
"epoch": 0.8984420080784766,
"grad_norm": 0.83203125,
"learning_rate": 0.0004930700967976918,
"loss": 6.1376,
"mean_token_accuracy": 0.17125671803951265,
"num_tokens": 19744176.0,
"step": 7785
},
{
"entropy": 6.298233366012573,
"epoch": 0.8990190421234853,
"grad_norm": 0.8515625,
"learning_rate": 0.000493059930823332,
"loss": 6.0621,
"mean_token_accuracy": 0.1712489977478981,
"num_tokens": 19756072.0,
"step": 7790
},
{
"entropy": 6.259371900558472,
"epoch": 0.8995960761684939,
"grad_norm": 0.91015625,
"learning_rate": 0.0004930497575146346,
"loss": 6.1716,
"mean_token_accuracy": 0.16485102325677872,
"num_tokens": 19768170.0,
"step": 7795
},
{
"entropy": 6.293583345413208,
"epoch": 0.9001731102135025,
"grad_norm": 0.77734375,
"learning_rate": 0.0004930395768719421,
"loss": 6.1325,
"mean_token_accuracy": 0.16966002881526948,
"num_tokens": 19782029.0,
"step": 7800
},
{
"entropy": 6.358855533599853,
"epoch": 0.9007501442585113,
"grad_norm": 0.828125,
"learning_rate": 0.0004930293888955966,
"loss": 6.0704,
"mean_token_accuracy": 0.16987146884202958,
"num_tokens": 19793973.0,
"step": 7805
},
{
"entropy": 6.232981252670288,
"epoch": 0.9013271783035199,
"grad_norm": 0.83984375,
"learning_rate": 0.000493019193585941,
"loss": 6.062,
"mean_token_accuracy": 0.17212969660758973,
"num_tokens": 19806806.0,
"step": 7810
},
{
"entropy": 6.338475561141967,
"epoch": 0.9019042123485286,
"grad_norm": 0.80859375,
"learning_rate": 0.000493008990943318,
"loss": 6.1776,
"mean_token_accuracy": 0.16324291676282882,
"num_tokens": 19821025.0,
"step": 7815
},
{
"entropy": 6.312364482879639,
"epoch": 0.9024812463935372,
"grad_norm": 0.87890625,
"learning_rate": 0.0004929987809680709,
"loss": 6.0055,
"mean_token_accuracy": 0.17505942583084105,
"num_tokens": 19832909.0,
"step": 7820
},
{
"entropy": 6.22635293006897,
"epoch": 0.9030582804385459,
"grad_norm": 1.0390625,
"learning_rate": 0.0004929885636605432,
"loss": 6.0019,
"mean_token_accuracy": 0.1759738802909851,
"num_tokens": 19846813.0,
"step": 7825
},
{
"entropy": 6.345417594909668,
"epoch": 0.9036353144835545,
"grad_norm": 0.96875,
"learning_rate": 0.0004929783390210784,
"loss": 6.1491,
"mean_token_accuracy": 0.16306580901145934,
"num_tokens": 19860005.0,
"step": 7830
},
{
"entropy": 6.362852287292481,
"epoch": 0.9042123485285631,
"grad_norm": 0.84765625,
"learning_rate": 0.0004929681070500204,
"loss": 6.0974,
"mean_token_accuracy": 0.16966380774974824,
"num_tokens": 19871931.0,
"step": 7835
},
{
"entropy": 6.256046772003174,
"epoch": 0.9047893825735719,
"grad_norm": 0.82421875,
"learning_rate": 0.0004929578677477135,
"loss": 6.0799,
"mean_token_accuracy": 0.17365067452192307,
"num_tokens": 19884066.0,
"step": 7840
},
{
"entropy": 6.341046953201294,
"epoch": 0.9053664166185805,
"grad_norm": 0.875,
"learning_rate": 0.0004929476211145019,
"loss": 6.153,
"mean_token_accuracy": 0.16422106325626373,
"num_tokens": 19896416.0,
"step": 7845
},
{
"entropy": 6.42669677734375,
"epoch": 0.9059434506635892,
"grad_norm": 0.80859375,
"learning_rate": 0.0004929373671507303,
"loss": 6.1642,
"mean_token_accuracy": 0.16458971053361893,
"num_tokens": 19911036.0,
"step": 7850
},
{
"entropy": 6.351240587234497,
"epoch": 0.9065204847085978,
"grad_norm": 0.87890625,
"learning_rate": 0.0004929271058567436,
"loss": 6.1577,
"mean_token_accuracy": 0.1675029769539833,
"num_tokens": 19924516.0,
"step": 7855
},
{
"entropy": 6.252199649810791,
"epoch": 0.9070975187536064,
"grad_norm": 0.87109375,
"learning_rate": 0.000492916837232887,
"loss": 6.0347,
"mean_token_accuracy": 0.17800809144973756,
"num_tokens": 19936994.0,
"step": 7860
},
{
"entropy": 6.266292953491211,
"epoch": 0.9076745527986151,
"grad_norm": 0.86328125,
"learning_rate": 0.0004929065612795058,
"loss": 6.0865,
"mean_token_accuracy": 0.17676833420991897,
"num_tokens": 19950304.0,
"step": 7865
},
{
"entropy": 6.254460954666138,
"epoch": 0.9082515868436237,
"grad_norm": 0.84375,
"learning_rate": 0.0004928962779969456,
"loss": 6.1386,
"mean_token_accuracy": 0.16655431985855101,
"num_tokens": 19963842.0,
"step": 7870
},
{
"entropy": 6.3657478332519535,
"epoch": 0.9088286208886325,
"grad_norm": 0.8984375,
"learning_rate": 0.0004928859873855524,
"loss": 6.1584,
"mean_token_accuracy": 0.16760774701833725,
"num_tokens": 19975806.0,
"step": 7875
},
{
"entropy": 6.231832122802734,
"epoch": 0.9094056549336411,
"grad_norm": 0.8515625,
"learning_rate": 0.0004928756894456723,
"loss": 6.092,
"mean_token_accuracy": 0.17502667307853698,
"num_tokens": 19987647.0,
"step": 7880
},
{
"entropy": 6.357949447631836,
"epoch": 0.9099826889786498,
"grad_norm": 0.9296875,
"learning_rate": 0.0004928653841776515,
"loss": 6.1628,
"mean_token_accuracy": 0.16325012296438218,
"num_tokens": 20000591.0,
"step": 7885
},
{
"entropy": 6.406267166137695,
"epoch": 0.9105597230236584,
"grad_norm": 0.8984375,
"learning_rate": 0.0004928550715818368,
"loss": 6.0812,
"mean_token_accuracy": 0.1728819116950035,
"num_tokens": 20012657.0,
"step": 7890
},
{
"entropy": 6.382717943191528,
"epoch": 0.911136757068667,
"grad_norm": 0.8046875,
"learning_rate": 0.0004928447516585749,
"loss": 6.1694,
"mean_token_accuracy": 0.16989215463399887,
"num_tokens": 20024752.0,
"step": 7895
},
{
"entropy": 6.273079204559326,
"epoch": 0.9117137911136757,
"grad_norm": 0.8515625,
"learning_rate": 0.000492834424408213,
"loss": 6.0692,
"mean_token_accuracy": 0.1694467097520828,
"num_tokens": 20036896.0,
"step": 7900
},
{
"entropy": 6.329956150054931,
"epoch": 0.9122908251586843,
"grad_norm": 0.88671875,
"learning_rate": 0.0004928240898310984,
"loss": 6.1359,
"mean_token_accuracy": 0.16939880549907685,
"num_tokens": 20049421.0,
"step": 7905
},
{
"entropy": 6.349690055847168,
"epoch": 0.9128678592036931,
"grad_norm": 0.90234375,
"learning_rate": 0.0004928137479275789,
"loss": 6.1758,
"mean_token_accuracy": 0.1654338553547859,
"num_tokens": 20062298.0,
"step": 7910
},
{
"entropy": 6.3017576217651365,
"epoch": 0.9134448932487017,
"grad_norm": 0.7734375,
"learning_rate": 0.000492803398698002,
"loss": 6.0854,
"mean_token_accuracy": 0.1678279921412468,
"num_tokens": 20075954.0,
"step": 7915
},
{
"entropy": 6.2847202777862545,
"epoch": 0.9140219272937103,
"grad_norm": 0.875,
"learning_rate": 0.0004927930421427161,
"loss": 6.0208,
"mean_token_accuracy": 0.17434524595737458,
"num_tokens": 20088035.0,
"step": 7920
},
{
"entropy": 6.294939088821411,
"epoch": 0.914598961338719,
"grad_norm": 0.92578125,
"learning_rate": 0.0004927826782620694,
"loss": 6.0776,
"mean_token_accuracy": 0.17393822073936463,
"num_tokens": 20100081.0,
"step": 7925
},
{
"entropy": 6.360169124603272,
"epoch": 0.9151759953837276,
"grad_norm": 0.8125,
"learning_rate": 0.0004927723070564104,
"loss": 6.1337,
"mean_token_accuracy": 0.17027704119682313,
"num_tokens": 20113159.0,
"step": 7930
},
{
"entropy": 6.286882972717285,
"epoch": 0.9157530294287363,
"grad_norm": 0.84765625,
"learning_rate": 0.0004927619285260881,
"loss": 6.0953,
"mean_token_accuracy": 0.1708056688308716,
"num_tokens": 20125741.0,
"step": 7935
},
{
"entropy": 6.271196699142456,
"epoch": 0.9163300634737449,
"grad_norm": 0.80859375,
"learning_rate": 0.0004927515426714515,
"loss": 6.0599,
"mean_token_accuracy": 0.17156231701374053,
"num_tokens": 20137481.0,
"step": 7940
},
{
"entropy": 6.336545038223266,
"epoch": 0.9169070975187537,
"grad_norm": 0.8046875,
"learning_rate": 0.00049274114949285,
"loss": 6.1312,
"mean_token_accuracy": 0.1683913141489029,
"num_tokens": 20150188.0,
"step": 7945
},
{
"entropy": 6.371749639511108,
"epoch": 0.9174841315637623,
"grad_norm": 0.83203125,
"learning_rate": 0.000492730748990633,
"loss": 6.1699,
"mean_token_accuracy": 0.16632280945777894,
"num_tokens": 20163012.0,
"step": 7950
},
{
"entropy": 6.326217842102051,
"epoch": 0.9180611656087709,
"grad_norm": 0.8203125,
"learning_rate": 0.0004927203411651504,
"loss": 6.1006,
"mean_token_accuracy": 0.1692713424563408,
"num_tokens": 20175468.0,
"step": 7955
},
{
"entropy": 6.294983530044556,
"epoch": 0.9186381996537796,
"grad_norm": 0.85546875,
"learning_rate": 0.0004927099260167523,
"loss": 6.1273,
"mean_token_accuracy": 0.16911787688732147,
"num_tokens": 20188382.0,
"step": 7960
},
{
"entropy": 6.343044328689575,
"epoch": 0.9192152336987882,
"grad_norm": 0.82421875,
"learning_rate": 0.0004926995035457889,
"loss": 6.0923,
"mean_token_accuracy": 0.17835812717676164,
"num_tokens": 20200091.0,
"step": 7965
},
{
"entropy": 6.284211587905884,
"epoch": 0.9197922677437969,
"grad_norm": 0.8203125,
"learning_rate": 0.000492689073752611,
"loss": 6.1333,
"mean_token_accuracy": 0.16809267848730086,
"num_tokens": 20212004.0,
"step": 7970
},
{
"entropy": 6.268933963775635,
"epoch": 0.9203693017888055,
"grad_norm": 0.84765625,
"learning_rate": 0.0004926786366375691,
"loss": 6.1317,
"mean_token_accuracy": 0.1755678579211235,
"num_tokens": 20225214.0,
"step": 7975
},
{
"entropy": 6.2912781715393065,
"epoch": 0.9209463358338142,
"grad_norm": 0.828125,
"learning_rate": 0.0004926681922010145,
"loss": 6.0733,
"mean_token_accuracy": 0.17149922698736192,
"num_tokens": 20236595.0,
"step": 7980
},
{
"entropy": 6.288493919372558,
"epoch": 0.9215233698788229,
"grad_norm": 0.93359375,
"learning_rate": 0.0004926577404432982,
"loss": 6.0899,
"mean_token_accuracy": 0.16908173710107804,
"num_tokens": 20248601.0,
"step": 7985
},
{
"entropy": 6.310372066497803,
"epoch": 0.9221004039238315,
"grad_norm": 0.77734375,
"learning_rate": 0.0004926472813647721,
"loss": 6.106,
"mean_token_accuracy": 0.16341881453990936,
"num_tokens": 20261314.0,
"step": 7990
},
{
"entropy": 6.356702184677124,
"epoch": 0.9226774379688402,
"grad_norm": 0.8828125,
"learning_rate": 0.0004926368149657876,
"loss": 6.1312,
"mean_token_accuracy": 0.17094669938087464,
"num_tokens": 20274797.0,
"step": 7995
},
{
"entropy": 6.321406745910645,
"epoch": 0.9232544720138488,
"grad_norm": 0.8671875,
"learning_rate": 0.0004926263412466972,
"loss": 6.1863,
"mean_token_accuracy": 0.16469254940748215,
"num_tokens": 20287031.0,
"step": 8000
},
{
"entropy": 6.2833436012268065,
"epoch": 0.9238315060588574,
"grad_norm": 0.82421875,
"learning_rate": 0.0004926158602078527,
"loss": 6.0814,
"mean_token_accuracy": 0.16865895092487335,
"num_tokens": 20298716.0,
"step": 8005
},
{
"entropy": 6.311143016815185,
"epoch": 0.9244085401038661,
"grad_norm": 0.8984375,
"learning_rate": 0.0004926053718496069,
"loss": 6.0557,
"mean_token_accuracy": 0.18285336643457412,
"num_tokens": 20311129.0,
"step": 8010
},
{
"entropy": 6.161644124984742,
"epoch": 0.9249855741488748,
"grad_norm": 0.87890625,
"learning_rate": 0.0004925948761723126,
"loss": 5.9655,
"mean_token_accuracy": 0.17335004657506942,
"num_tokens": 20323067.0,
"step": 8015
},
{
"entropy": 6.329570674896241,
"epoch": 0.9255626081938835,
"grad_norm": 0.88671875,
"learning_rate": 0.0004925843731763226,
"loss": 6.1137,
"mean_token_accuracy": 0.1678135350346565,
"num_tokens": 20335710.0,
"step": 8020
},
{
"entropy": 6.3780488014221195,
"epoch": 0.9261396422388921,
"grad_norm": 1.0,
"learning_rate": 0.0004925738628619904,
"loss": 6.1529,
"mean_token_accuracy": 0.1679707333445549,
"num_tokens": 20348436.0,
"step": 8025
},
{
"entropy": 6.2938000679016115,
"epoch": 0.9267166762839008,
"grad_norm": 0.8359375,
"learning_rate": 0.0004925633452296693,
"loss": 6.0995,
"mean_token_accuracy": 0.16210879683494567,
"num_tokens": 20362182.0,
"step": 8030
},
{
"entropy": 6.302653074264526,
"epoch": 0.9272937103289094,
"grad_norm": 0.87890625,
"learning_rate": 0.0004925528202797131,
"loss": 6.1194,
"mean_token_accuracy": 0.16243739277124405,
"num_tokens": 20374674.0,
"step": 8035
},
{
"entropy": 6.288832092285157,
"epoch": 0.927870744373918,
"grad_norm": 0.8671875,
"learning_rate": 0.0004925422880124761,
"loss": 6.0563,
"mean_token_accuracy": 0.1735551044344902,
"num_tokens": 20388918.0,
"step": 8040
},
{
"entropy": 6.276844120025634,
"epoch": 0.9284477784189267,
"grad_norm": 0.81640625,
"learning_rate": 0.0004925317484283121,
"loss": 6.0697,
"mean_token_accuracy": 0.17131037414073944,
"num_tokens": 20401428.0,
"step": 8045
},
{
"entropy": 6.325076341629028,
"epoch": 0.9290248124639354,
"grad_norm": 0.859375,
"learning_rate": 0.0004925212015275758,
"loss": 6.1002,
"mean_token_accuracy": 0.17513058632612227,
"num_tokens": 20414276.0,
"step": 8050
},
{
"entropy": 6.340042304992676,
"epoch": 0.9296018465089441,
"grad_norm": 0.8515625,
"learning_rate": 0.000492510647310622,
"loss": 6.0794,
"mean_token_accuracy": 0.16827186644077302,
"num_tokens": 20425515.0,
"step": 8055
},
{
"entropy": 6.31017689704895,
"epoch": 0.9301788805539527,
"grad_norm": 0.93359375,
"learning_rate": 0.0004925000857778055,
"loss": 6.1302,
"mean_token_accuracy": 0.16977558434009551,
"num_tokens": 20439137.0,
"step": 8060
},
{
"entropy": 6.269415760040284,
"epoch": 0.9307559145989613,
"grad_norm": 0.8828125,
"learning_rate": 0.0004924895169294818,
"loss": 6.0194,
"mean_token_accuracy": 0.1778138369321823,
"num_tokens": 20452059.0,
"step": 8065
},
{
"entropy": 6.357288503646851,
"epoch": 0.93133294864397,
"grad_norm": 0.75390625,
"learning_rate": 0.0004924789407660062,
"loss": 6.192,
"mean_token_accuracy": 0.15842437297105788,
"num_tokens": 20465110.0,
"step": 8070
},
{
"entropy": 6.180405855178833,
"epoch": 0.9319099826889786,
"grad_norm": 0.8359375,
"learning_rate": 0.0004924683572877345,
"loss": 5.9699,
"mean_token_accuracy": 0.17301704585552216,
"num_tokens": 20477615.0,
"step": 8075
},
{
"entropy": 6.211012697219848,
"epoch": 0.9324870167339873,
"grad_norm": 0.859375,
"learning_rate": 0.0004924577664950225,
"loss": 5.9691,
"mean_token_accuracy": 0.18344345539808274,
"num_tokens": 20490054.0,
"step": 8080
},
{
"entropy": 6.34092378616333,
"epoch": 0.933064050778996,
"grad_norm": 0.859375,
"learning_rate": 0.0004924471683882266,
"loss": 6.1438,
"mean_token_accuracy": 0.16662690043449402,
"num_tokens": 20502439.0,
"step": 8085
},
{
"entropy": 6.311378908157349,
"epoch": 0.9336410848240047,
"grad_norm": 0.8515625,
"learning_rate": 0.0004924365629677031,
"loss": 6.1245,
"mean_token_accuracy": 0.1725110724568367,
"num_tokens": 20515094.0,
"step": 8090
},
{
"entropy": 6.306700801849365,
"epoch": 0.9342181188690133,
"grad_norm": 0.89453125,
"learning_rate": 0.000492425950233809,
"loss": 6.04,
"mean_token_accuracy": 0.1781902179121971,
"num_tokens": 20527025.0,
"step": 8095
},
{
"entropy": 6.32002215385437,
"epoch": 0.9347951529140219,
"grad_norm": 0.82421875,
"learning_rate": 0.000492415330186901,
"loss": 6.1692,
"mean_token_accuracy": 0.16498290449380876,
"num_tokens": 20540793.0,
"step": 8100
},
{
"entropy": 6.3359020233154295,
"epoch": 0.9353721869590306,
"grad_norm": 0.9453125,
"learning_rate": 0.0004924047028273364,
"loss": 6.1655,
"mean_token_accuracy": 0.16523907482624053,
"num_tokens": 20553425.0,
"step": 8105
},
{
"entropy": 6.360561370849609,
"epoch": 0.9359492210040392,
"grad_norm": 0.921875,
"learning_rate": 0.0004923940681554725,
"loss": 6.1151,
"mean_token_accuracy": 0.17208221107721328,
"num_tokens": 20566794.0,
"step": 8110
},
{
"entropy": 6.330394124984741,
"epoch": 0.9365262550490479,
"grad_norm": 0.80078125,
"learning_rate": 0.0004923834261716672,
"loss": 6.1838,
"mean_token_accuracy": 0.163845494389534,
"num_tokens": 20579756.0,
"step": 8115
},
{
"entropy": 6.349153900146485,
"epoch": 0.9371032890940566,
"grad_norm": 0.8046875,
"learning_rate": 0.0004923727768762782,
"loss": 6.1509,
"mean_token_accuracy": 0.16518385112285613,
"num_tokens": 20592626.0,
"step": 8120
},
{
"entropy": 6.3262717723846436,
"epoch": 0.9376803231390652,
"grad_norm": 0.82421875,
"learning_rate": 0.000492362120269664,
"loss": 6.0708,
"mean_token_accuracy": 0.17791670858860015,
"num_tokens": 20605023.0,
"step": 8125
},
{
"entropy": 6.345363521575928,
"epoch": 0.9382573571840739,
"grad_norm": 0.77734375,
"learning_rate": 0.0004923514563521827,
"loss": 6.1924,
"mean_token_accuracy": 0.16382016837596894,
"num_tokens": 20617668.0,
"step": 8130
},
{
"entropy": 6.29886908531189,
"epoch": 0.9388343912290825,
"grad_norm": 0.85546875,
"learning_rate": 0.0004923407851241933,
"loss": 6.0769,
"mean_token_accuracy": 0.1696800783276558,
"num_tokens": 20629848.0,
"step": 8135
},
{
"entropy": 6.284991836547851,
"epoch": 0.9394114252740912,
"grad_norm": 0.91796875,
"learning_rate": 0.0004923301065860545,
"loss": 6.0843,
"mean_token_accuracy": 0.17413091361522676,
"num_tokens": 20642695.0,
"step": 8140
},
{
"entropy": 6.334108829498291,
"epoch": 0.9399884593190998,
"grad_norm": 0.83203125,
"learning_rate": 0.0004923194207381254,
"loss": 6.1192,
"mean_token_accuracy": 0.17119077295064927,
"num_tokens": 20655844.0,
"step": 8145
},
{
"entropy": 6.342758464813232,
"epoch": 0.9405654933641084,
"grad_norm": 0.81640625,
"learning_rate": 0.0004923087275807656,
"loss": 6.2023,
"mean_token_accuracy": 0.16785278022289277,
"num_tokens": 20668380.0,
"step": 8150
},
{
"entropy": 6.322146558761597,
"epoch": 0.9411425274091172,
"grad_norm": 0.78125,
"learning_rate": 0.0004922980271143347,
"loss": 6.0803,
"mean_token_accuracy": 0.17897505462169647,
"num_tokens": 20680766.0,
"step": 8155
},
{
"entropy": 6.282084703445435,
"epoch": 0.9417195614541258,
"grad_norm": 0.85546875,
"learning_rate": 0.0004922873193391927,
"loss": 6.0809,
"mean_token_accuracy": 0.16945219039916992,
"num_tokens": 20692538.0,
"step": 8160
},
{
"entropy": 6.333216905593872,
"epoch": 0.9422965954991345,
"grad_norm": 0.8203125,
"learning_rate": 0.0004922766042556994,
"loss": 6.0826,
"mean_token_accuracy": 0.17635788023471832,
"num_tokens": 20705832.0,
"step": 8165
},
{
"entropy": 6.336708974838257,
"epoch": 0.9428736295441431,
"grad_norm": 0.87109375,
"learning_rate": 0.0004922658818642156,
"loss": 6.1351,
"mean_token_accuracy": 0.16822911351919173,
"num_tokens": 20718741.0,
"step": 8170
},
{
"entropy": 6.332752132415772,
"epoch": 0.9434506635891518,
"grad_norm": 0.8828125,
"learning_rate": 0.0004922551521651018,
"loss": 6.1075,
"mean_token_accuracy": 0.1693040058016777,
"num_tokens": 20731516.0,
"step": 8175
},
{
"entropy": 6.334755182266235,
"epoch": 0.9440276976341604,
"grad_norm": 0.80078125,
"learning_rate": 0.0004922444151587189,
"loss": 6.1353,
"mean_token_accuracy": 0.17082017064094543,
"num_tokens": 20744835.0,
"step": 8180
},
{
"entropy": 6.292046594619751,
"epoch": 0.944604731679169,
"grad_norm": 0.89453125,
"learning_rate": 0.0004922336708454279,
"loss": 6.1189,
"mean_token_accuracy": 0.16721362620592117,
"num_tokens": 20758967.0,
"step": 8185
},
{
"entropy": 6.280726623535156,
"epoch": 0.9451817657241778,
"grad_norm": 0.87109375,
"learning_rate": 0.0004922229192255903,
"loss": 6.075,
"mean_token_accuracy": 0.17387010008096696,
"num_tokens": 20771660.0,
"step": 8190
},
{
"entropy": 6.358853816986084,
"epoch": 0.9457587997691864,
"grad_norm": 0.828125,
"learning_rate": 0.0004922121602995678,
"loss": 6.1631,
"mean_token_accuracy": 0.17015791386365892,
"num_tokens": 20785366.0,
"step": 8195
},
{
"entropy": 6.381282901763916,
"epoch": 0.9463358338141951,
"grad_norm": 0.8203125,
"learning_rate": 0.0004922013940677221,
"loss": 6.1817,
"mean_token_accuracy": 0.16708896309137344,
"num_tokens": 20797996.0,
"step": 8200
},
{
"entropy": 6.308471012115478,
"epoch": 0.9469128678592037,
"grad_norm": 0.80859375,
"learning_rate": 0.0004921906205304155,
"loss": 6.1403,
"mean_token_accuracy": 0.16666304767131807,
"num_tokens": 20810921.0,
"step": 8205
},
{
"entropy": 6.313723754882813,
"epoch": 0.9474899019042123,
"grad_norm": 0.828125,
"learning_rate": 0.0004921798396880101,
"loss": 5.995,
"mean_token_accuracy": 0.17673814296722412,
"num_tokens": 20823143.0,
"step": 8210
},
{
"entropy": 6.302704572677612,
"epoch": 0.948066935949221,
"grad_norm": 0.8671875,
"learning_rate": 0.0004921690515408688,
"loss": 6.1101,
"mean_token_accuracy": 0.16627644300460814,
"num_tokens": 20834604.0,
"step": 8215
},
{
"entropy": 6.333921718597412,
"epoch": 0.9486439699942296,
"grad_norm": 0.82421875,
"learning_rate": 0.0004921582560893543,
"loss": 6.1562,
"mean_token_accuracy": 0.1714571610093117,
"num_tokens": 20847224.0,
"step": 8220
},
{
"entropy": 6.32955641746521,
"epoch": 0.9492210040392384,
"grad_norm": 0.8828125,
"learning_rate": 0.0004921474533338297,
"loss": 6.14,
"mean_token_accuracy": 0.1752069413661957,
"num_tokens": 20859477.0,
"step": 8225
},
{
"entropy": 6.285461854934693,
"epoch": 0.949798038084247,
"grad_norm": 0.875,
"learning_rate": 0.0004921366432746585,
"loss": 6.0942,
"mean_token_accuracy": 0.17483866810798646,
"num_tokens": 20872144.0,
"step": 8230
},
{
"entropy": 6.339507579803467,
"epoch": 0.9503750721292556,
"grad_norm": 0.78515625,
"learning_rate": 0.000492125825912204,
"loss": 6.0751,
"mean_token_accuracy": 0.17300888895988464,
"num_tokens": 20885168.0,
"step": 8235
},
{
"entropy": 6.290178442001343,
"epoch": 0.9509521061742643,
"grad_norm": 0.83203125,
"learning_rate": 0.0004921150012468302,
"loss": 6.0957,
"mean_token_accuracy": 0.1704855278134346,
"num_tokens": 20896650.0,
"step": 8240
},
{
"entropy": 6.356605482101441,
"epoch": 0.9515291402192729,
"grad_norm": 0.828125,
"learning_rate": 0.0004921041692789013,
"loss": 6.1853,
"mean_token_accuracy": 0.16309679448604583,
"num_tokens": 20908764.0,
"step": 8245
},
{
"entropy": 6.388557767868042,
"epoch": 0.9521061742642816,
"grad_norm": 0.90234375,
"learning_rate": 0.0004920933300087813,
"loss": 6.1032,
"mean_token_accuracy": 0.16476467102766038,
"num_tokens": 20920852.0,
"step": 8250
},
{
"entropy": 6.310107326507568,
"epoch": 0.9526832083092902,
"grad_norm": 0.98046875,
"learning_rate": 0.0004920824834368353,
"loss": 6.0673,
"mean_token_accuracy": 0.17406057864427565,
"num_tokens": 20933747.0,
"step": 8255
},
{
"entropy": 6.298854780197144,
"epoch": 0.953260242354299,
"grad_norm": 0.91796875,
"learning_rate": 0.0004920716295634275,
"loss": 6.1238,
"mean_token_accuracy": 0.1746158018708229,
"num_tokens": 20945563.0,
"step": 8260
},
{
"entropy": 6.278277587890625,
"epoch": 0.9538372763993076,
"grad_norm": 0.8203125,
"learning_rate": 0.0004920607683889235,
"loss": 6.09,
"mean_token_accuracy": 0.16672398447990416,
"num_tokens": 20957449.0,
"step": 8265
},
{
"entropy": 6.282410764694214,
"epoch": 0.9544143104443162,
"grad_norm": 0.875,
"learning_rate": 0.0004920498999136882,
"loss": 6.044,
"mean_token_accuracy": 0.17534780949354173,
"num_tokens": 20969548.0,
"step": 8270
},
{
"entropy": 6.251287317276001,
"epoch": 0.9549913444893249,
"grad_norm": 0.8828125,
"learning_rate": 0.0004920390241380874,
"loss": 6.0257,
"mean_token_accuracy": 0.17555872350931168,
"num_tokens": 20981712.0,
"step": 8275
},
{
"entropy": 6.305266666412353,
"epoch": 0.9555683785343335,
"grad_norm": 0.9140625,
"learning_rate": 0.0004920281410624868,
"loss": 6.1334,
"mean_token_accuracy": 0.1694267451763153,
"num_tokens": 20993997.0,
"step": 8280
},
{
"entropy": 6.3954089164733885,
"epoch": 0.9561454125793422,
"grad_norm": 0.86328125,
"learning_rate": 0.0004920172506872525,
"loss": 6.172,
"mean_token_accuracy": 0.1699483409523964,
"num_tokens": 21007158.0,
"step": 8285
},
{
"entropy": 6.303517913818359,
"epoch": 0.9567224466243508,
"grad_norm": 0.93359375,
"learning_rate": 0.0004920063530127508,
"loss": 6.0658,
"mean_token_accuracy": 0.16866770535707473,
"num_tokens": 21019592.0,
"step": 8290
},
{
"entropy": 6.278334474563598,
"epoch": 0.9572994806693594,
"grad_norm": 0.83984375,
"learning_rate": 0.0004919954480393482,
"loss": 6.0971,
"mean_token_accuracy": 0.17727649658918382,
"num_tokens": 21031850.0,
"step": 8295
},
{
"entropy": 6.324184608459473,
"epoch": 0.9578765147143682,
"grad_norm": 0.7890625,
"learning_rate": 0.0004919845357674114,
"loss": 6.0575,
"mean_token_accuracy": 0.17361602932214737,
"num_tokens": 21044348.0,
"step": 8300
},
{
"entropy": 6.316019916534424,
"epoch": 0.9584535487593768,
"grad_norm": 0.8046875,
"learning_rate": 0.0004919736161973076,
"loss": 6.1495,
"mean_token_accuracy": 0.17223343402147293,
"num_tokens": 21058017.0,
"step": 8305
},
{
"entropy": 6.355439186096191,
"epoch": 0.9590305828043855,
"grad_norm": 0.8828125,
"learning_rate": 0.000491962689329404,
"loss": 6.0759,
"mean_token_accuracy": 0.1674254924058914,
"num_tokens": 21070443.0,
"step": 8310
},
{
"entropy": 6.29806694984436,
"epoch": 0.9596076168493941,
"grad_norm": 0.8125,
"learning_rate": 0.0004919517551640681,
"loss": 6.1501,
"mean_token_accuracy": 0.1736349418759346,
"num_tokens": 21084033.0,
"step": 8315
},
{
"entropy": 6.336664342880249,
"epoch": 0.9601846508944027,
"grad_norm": 0.859375,
"learning_rate": 0.0004919408137016677,
"loss": 6.0935,
"mean_token_accuracy": 0.1669437140226364,
"num_tokens": 21095737.0,
"step": 8320
},
{
"entropy": 6.397348928451538,
"epoch": 0.9607616849394114,
"grad_norm": 0.83984375,
"learning_rate": 0.0004919298649425708,
"loss": 6.0639,
"mean_token_accuracy": 0.1697271764278412,
"num_tokens": 21107769.0,
"step": 8325
},
{
"entropy": 6.216275596618653,
"epoch": 0.96133871898442,
"grad_norm": 0.88671875,
"learning_rate": 0.0004919189088871456,
"loss": 6.0161,
"mean_token_accuracy": 0.17900092750787736,
"num_tokens": 21119008.0,
"step": 8330
},
{
"entropy": 6.246855115890503,
"epoch": 0.9619157530294288,
"grad_norm": 0.84375,
"learning_rate": 0.0004919079455357608,
"loss": 6.1334,
"mean_token_accuracy": 0.16993842422962188,
"num_tokens": 21131433.0,
"step": 8335
},
{
"entropy": 6.372685050964355,
"epoch": 0.9624927870744374,
"grad_norm": 0.90625,
"learning_rate": 0.0004918969748887847,
"loss": 6.1928,
"mean_token_accuracy": 0.16099000126123428,
"num_tokens": 21144237.0,
"step": 8340
},
{
"entropy": 6.383633422851562,
"epoch": 0.9630698211194461,
"grad_norm": 0.84375,
"learning_rate": 0.0004918859969465868,
"loss": 6.1689,
"mean_token_accuracy": 0.16766799092292786,
"num_tokens": 21157795.0,
"step": 8345
},
{
"entropy": 6.290271472930908,
"epoch": 0.9636468551644547,
"grad_norm": 0.8359375,
"learning_rate": 0.0004918750117095361,
"loss": 6.0985,
"mean_token_accuracy": 0.1680177181959152,
"num_tokens": 21169943.0,
"step": 8350
},
{
"entropy": 6.281228685379029,
"epoch": 0.9642238892094633,
"grad_norm": 0.8125,
"learning_rate": 0.0004918640191780021,
"loss": 6.0017,
"mean_token_accuracy": 0.17994622588157655,
"num_tokens": 21184432.0,
"step": 8355
},
{
"entropy": 6.307895565032959,
"epoch": 0.964800923254472,
"grad_norm": 0.84765625,
"learning_rate": 0.0004918530193523546,
"loss": 6.0218,
"mean_token_accuracy": 0.17425339967012404,
"num_tokens": 21196539.0,
"step": 8360
},
{
"entropy": 6.293627548217773,
"epoch": 0.9653779572994806,
"grad_norm": 0.85546875,
"learning_rate": 0.0004918420122329634,
"loss": 6.1277,
"mean_token_accuracy": 0.1632673978805542,
"num_tokens": 21209550.0,
"step": 8365
},
{
"entropy": 6.234926080703735,
"epoch": 0.9659549913444894,
"grad_norm": 0.8671875,
"learning_rate": 0.0004918309978201989,
"loss": 5.986,
"mean_token_accuracy": 0.18019188344478607,
"num_tokens": 21220747.0,
"step": 8370
},
{
"entropy": 6.2483526229858395,
"epoch": 0.966532025389498,
"grad_norm": 0.85546875,
"learning_rate": 0.0004918199761144315,
"loss": 5.9756,
"mean_token_accuracy": 0.18189812898635865,
"num_tokens": 21233229.0,
"step": 8375
},
{
"entropy": 6.232093477249146,
"epoch": 0.9671090594345066,
"grad_norm": 0.8515625,
"learning_rate": 0.0004918089471160318,
"loss": 6.0097,
"mean_token_accuracy": 0.17824846804141997,
"num_tokens": 21244656.0,
"step": 8380
},
{
"entropy": 6.1882080078125,
"epoch": 0.9676860934795153,
"grad_norm": 0.90234375,
"learning_rate": 0.0004917979108253709,
"loss": 6.0632,
"mean_token_accuracy": 0.16929151117801666,
"num_tokens": 21257602.0,
"step": 8385
},
{
"entropy": 6.268950080871582,
"epoch": 0.9682631275245239,
"grad_norm": 0.83203125,
"learning_rate": 0.00049178686724282,
"loss": 6.0456,
"mean_token_accuracy": 0.1763706237077713,
"num_tokens": 21270098.0,
"step": 8390
},
{
"entropy": 6.3219743251800535,
"epoch": 0.9688401615695326,
"grad_norm": 0.859375,
"learning_rate": 0.0004917758163687506,
"loss": 6.1007,
"mean_token_accuracy": 0.1630837947130203,
"num_tokens": 21283516.0,
"step": 8395
},
{
"entropy": 6.242742538452148,
"epoch": 0.9694171956145412,
"grad_norm": 0.9296875,
"learning_rate": 0.0004917647582035341,
"loss": 6.026,
"mean_token_accuracy": 0.17450403422117233,
"num_tokens": 21296146.0,
"step": 8400
},
{
"entropy": 6.303759431838989,
"epoch": 0.96999422965955,
"grad_norm": 0.8046875,
"learning_rate": 0.0004917536927475428,
"loss": 6.1105,
"mean_token_accuracy": 0.17302547097206117,
"num_tokens": 21308537.0,
"step": 8405
},
{
"entropy": 6.3395380020141605,
"epoch": 0.9705712637045586,
"grad_norm": 0.7734375,
"learning_rate": 0.0004917426200011486,
"loss": 6.141,
"mean_token_accuracy": 0.16933179795742034,
"num_tokens": 21322375.0,
"step": 8410
},
{
"entropy": 6.35033950805664,
"epoch": 0.9711482977495672,
"grad_norm": 0.81640625,
"learning_rate": 0.000491731539964724,
"loss": 6.1101,
"mean_token_accuracy": 0.17082908302545546,
"num_tokens": 21335884.0,
"step": 8415
},
{
"entropy": 6.2293980598449705,
"epoch": 0.9717253317945759,
"grad_norm": 0.80078125,
"learning_rate": 0.0004917204526386418,
"loss": 5.9787,
"mean_token_accuracy": 0.1721679225564003,
"num_tokens": 21349106.0,
"step": 8420
},
{
"entropy": 6.288054132461548,
"epoch": 0.9723023658395845,
"grad_norm": 0.78515625,
"learning_rate": 0.0004917093580232748,
"loss": 6.095,
"mean_token_accuracy": 0.1732641115784645,
"num_tokens": 21362306.0,
"step": 8425
},
{
"entropy": 6.362192916870117,
"epoch": 0.9728793998845932,
"grad_norm": 0.890625,
"learning_rate": 0.0004916982561189962,
"loss": 6.1051,
"mean_token_accuracy": 0.16863009333610535,
"num_tokens": 21375672.0,
"step": 8430
},
{
"entropy": 6.2904290676116945,
"epoch": 0.9734564339296018,
"grad_norm": 0.8359375,
"learning_rate": 0.0004916871469261794,
"loss": 6.0344,
"mean_token_accuracy": 0.1826200544834137,
"num_tokens": 21389174.0,
"step": 8435
},
{
"entropy": 6.269189786911011,
"epoch": 0.9740334679746105,
"grad_norm": 0.890625,
"learning_rate": 0.000491676030445198,
"loss": 6.1619,
"mean_token_accuracy": 0.1706463173031807,
"num_tokens": 21400027.0,
"step": 8440
},
{
"entropy": 6.293677568435669,
"epoch": 0.9746105020196192,
"grad_norm": 0.83203125,
"learning_rate": 0.0004916649066764259,
"loss": 6.1077,
"mean_token_accuracy": 0.16729268431663513,
"num_tokens": 21412843.0,
"step": 8445
},
{
"entropy": 6.347714185714722,
"epoch": 0.9751875360646278,
"grad_norm": 0.83984375,
"learning_rate": 0.0004916537756202375,
"loss": 6.1912,
"mean_token_accuracy": 0.16289388090372087,
"num_tokens": 21425253.0,
"step": 8450
},
{
"entropy": 6.295068073272705,
"epoch": 0.9757645701096365,
"grad_norm": 0.953125,
"learning_rate": 0.0004916426372770069,
"loss": 5.997,
"mean_token_accuracy": 0.1836947813630104,
"num_tokens": 21436428.0,
"step": 8455
},
{
"entropy": 6.229881477355957,
"epoch": 0.9763416041546451,
"grad_norm": 0.83203125,
"learning_rate": 0.0004916314916471087,
"loss": 6.0812,
"mean_token_accuracy": 0.17769705057144164,
"num_tokens": 21448680.0,
"step": 8460
},
{
"entropy": 6.283309888839722,
"epoch": 0.9769186381996537,
"grad_norm": 0.85546875,
"learning_rate": 0.0004916203387309179,
"loss": 6.0943,
"mean_token_accuracy": 0.17553262412548065,
"num_tokens": 21460935.0,
"step": 8465
},
{
"entropy": 6.324648523330689,
"epoch": 0.9774956722446624,
"grad_norm": 0.90625,
"learning_rate": 0.0004916091785288096,
"loss": 6.0911,
"mean_token_accuracy": 0.17453473955392837,
"num_tokens": 21473624.0,
"step": 8470
},
{
"entropy": 6.35996379852295,
"epoch": 0.9780727062896711,
"grad_norm": 0.81640625,
"learning_rate": 0.0004915980110411593,
"loss": 6.1947,
"mean_token_accuracy": 0.16287709027528763,
"num_tokens": 21486684.0,
"step": 8475
},
{
"entropy": 6.29085898399353,
"epoch": 0.9786497403346798,
"grad_norm": 0.87109375,
"learning_rate": 0.0004915868362683425,
"loss": 6.0445,
"mean_token_accuracy": 0.18065717071294785,
"num_tokens": 21500955.0,
"step": 8480
},
{
"entropy": 6.333687591552734,
"epoch": 0.9792267743796884,
"grad_norm": 0.8515625,
"learning_rate": 0.0004915756542107349,
"loss": 6.1112,
"mean_token_accuracy": 0.16784068048000336,
"num_tokens": 21512632.0,
"step": 8485
},
{
"entropy": 6.274033880233764,
"epoch": 0.979803808424697,
"grad_norm": 0.83984375,
"learning_rate": 0.0004915644648687127,
"loss": 6.0322,
"mean_token_accuracy": 0.17534481436014177,
"num_tokens": 21526455.0,
"step": 8490
},
{
"entropy": 6.342768144607544,
"epoch": 0.9803808424697057,
"grad_norm": 0.95703125,
"learning_rate": 0.0004915532682426524,
"loss": 6.114,
"mean_token_accuracy": 0.16444663405418397,
"num_tokens": 21538594.0,
"step": 8495
},
{
"entropy": 6.200225162506103,
"epoch": 0.9809578765147143,
"grad_norm": 0.82421875,
"learning_rate": 0.0004915420643329306,
"loss": 5.9968,
"mean_token_accuracy": 0.1741709068417549,
"num_tokens": 21551071.0,
"step": 8500
},
{
"entropy": 6.265279054641724,
"epoch": 0.981534910559723,
"grad_norm": 0.90234375,
"learning_rate": 0.000491530853139924,
"loss": 6.0004,
"mean_token_accuracy": 0.17739108949899673,
"num_tokens": 21563241.0,
"step": 8505
},
{
"entropy": 6.25150089263916,
"epoch": 0.9821119446047317,
"grad_norm": 0.8359375,
"learning_rate": 0.0004915196346640095,
"loss": 6.0184,
"mean_token_accuracy": 0.17257041484117508,
"num_tokens": 21575634.0,
"step": 8510
},
{
"entropy": 6.281124877929687,
"epoch": 0.9826889786497404,
"grad_norm": 0.89453125,
"learning_rate": 0.0004915084089055648,
"loss": 6.093,
"mean_token_accuracy": 0.17048918604850768,
"num_tokens": 21587703.0,
"step": 8515
},
{
"entropy": 6.330684232711792,
"epoch": 0.983266012694749,
"grad_norm": 0.859375,
"learning_rate": 0.0004914971758649673,
"loss": 6.1036,
"mean_token_accuracy": 0.16494657844305038,
"num_tokens": 21599048.0,
"step": 8520
},
{
"entropy": 6.28722071647644,
"epoch": 0.9838430467397576,
"grad_norm": 0.86328125,
"learning_rate": 0.0004914859355425948,
"loss": 5.978,
"mean_token_accuracy": 0.17772875279188155,
"num_tokens": 21611547.0,
"step": 8525
},
{
"entropy": 6.248496627807617,
"epoch": 0.9844200807847663,
"grad_norm": 0.875,
"learning_rate": 0.0004914746879388255,
"loss": 6.0668,
"mean_token_accuracy": 0.18192058950662612,
"num_tokens": 21623343.0,
"step": 8530
},
{
"entropy": 6.28045015335083,
"epoch": 0.9849971148297749,
"grad_norm": 0.90625,
"learning_rate": 0.0004914634330540373,
"loss": 6.0795,
"mean_token_accuracy": 0.16923675686120987,
"num_tokens": 21635318.0,
"step": 8535
},
{
"entropy": 6.336523151397705,
"epoch": 0.9855741488747836,
"grad_norm": 0.8671875,
"learning_rate": 0.0004914521708886093,
"loss": 6.0891,
"mean_token_accuracy": 0.16801240742206575,
"num_tokens": 21647539.0,
"step": 8540
},
{
"entropy": 6.288561105728149,
"epoch": 0.9861511829197923,
"grad_norm": 1.1484375,
"learning_rate": 0.0004914409014429201,
"loss": 6.0868,
"mean_token_accuracy": 0.1714344561100006,
"num_tokens": 21660587.0,
"step": 8545
},
{
"entropy": 6.315939426422119,
"epoch": 0.986728216964801,
"grad_norm": 0.8984375,
"learning_rate": 0.0004914296247173486,
"loss": 6.0893,
"mean_token_accuracy": 0.17230225205421448,
"num_tokens": 21672123.0,
"step": 8550
},
{
"entropy": 6.329123783111572,
"epoch": 0.9873052510098096,
"grad_norm": 0.8515625,
"learning_rate": 0.0004914183407122741,
"loss": 6.0382,
"mean_token_accuracy": 0.170340496301651,
"num_tokens": 21685430.0,
"step": 8555
},
{
"entropy": 6.360370874404907,
"epoch": 0.9878822850548182,
"grad_norm": 0.90625,
"learning_rate": 0.0004914070494280763,
"loss": 6.1429,
"mean_token_accuracy": 0.16646113991737366,
"num_tokens": 21698462.0,
"step": 8560
},
{
"entropy": 6.341304540634155,
"epoch": 0.9884593190998269,
"grad_norm": 0.8671875,
"learning_rate": 0.0004913957508651349,
"loss": 6.1809,
"mean_token_accuracy": 0.1604655459523201,
"num_tokens": 21711741.0,
"step": 8565
},
{
"entropy": 6.34149489402771,
"epoch": 0.9890363531448355,
"grad_norm": 0.85546875,
"learning_rate": 0.0004913844450238299,
"loss": 6.0853,
"mean_token_accuracy": 0.16677425801753998,
"num_tokens": 21723983.0,
"step": 8570
},
{
"entropy": 6.281299877166748,
"epoch": 0.9896133871898442,
"grad_norm": 0.86328125,
"learning_rate": 0.0004913731319045416,
"loss": 6.1262,
"mean_token_accuracy": 0.17223995923995972,
"num_tokens": 21736399.0,
"step": 8575
},
{
"entropy": 6.331571197509765,
"epoch": 0.9901904212348529,
"grad_norm": 0.80078125,
"learning_rate": 0.0004913618115076505,
"loss": 6.146,
"mean_token_accuracy": 0.165971776843071,
"num_tokens": 21748705.0,
"step": 8580
},
{
"entropy": 6.25674090385437,
"epoch": 0.9907674552798615,
"grad_norm": 0.9140625,
"learning_rate": 0.0004913504838335372,
"loss": 6.017,
"mean_token_accuracy": 0.18154898434877395,
"num_tokens": 21761659.0,
"step": 8585
},
{
"entropy": 6.277780055999756,
"epoch": 0.9913444893248702,
"grad_norm": 0.859375,
"learning_rate": 0.0004913391488825829,
"loss": 6.0384,
"mean_token_accuracy": 0.1767553433775902,
"num_tokens": 21774217.0,
"step": 8590
},
{
"entropy": 6.284101390838623,
"epoch": 0.9919215233698788,
"grad_norm": 0.91015625,
"learning_rate": 0.0004913278066551689,
"loss": 6.0883,
"mean_token_accuracy": 0.17325786352157593,
"num_tokens": 21787417.0,
"step": 8595
},
{
"entropy": 6.308177947998047,
"epoch": 0.9924985574148875,
"grad_norm": 0.84375,
"learning_rate": 0.0004913164571516765,
"loss": 6.1252,
"mean_token_accuracy": 0.17321580201387404,
"num_tokens": 21798869.0,
"step": 8600
},
{
"entropy": 6.301731824874878,
"epoch": 0.9930755914598961,
"grad_norm": 0.82421875,
"learning_rate": 0.0004913051003724876,
"loss": 6.0519,
"mean_token_accuracy": 0.1723179042339325,
"num_tokens": 21811622.0,
"step": 8605
},
{
"entropy": 6.314857387542725,
"epoch": 0.9936526255049047,
"grad_norm": 0.8984375,
"learning_rate": 0.0004912937363179839,
"loss": 6.0687,
"mean_token_accuracy": 0.17025351524353027,
"num_tokens": 21823401.0,
"step": 8610
},
{
"entropy": 6.277297592163086,
"epoch": 0.9942296595499135,
"grad_norm": 0.8125,
"learning_rate": 0.000491282364988548,
"loss": 6.064,
"mean_token_accuracy": 0.17257231920957566,
"num_tokens": 21835460.0,
"step": 8615
},
{
"entropy": 6.230294036865234,
"epoch": 0.9948066935949221,
"grad_norm": 0.80859375,
"learning_rate": 0.0004912709863845621,
"loss": 6.0106,
"mean_token_accuracy": 0.17783847004175185,
"num_tokens": 21847521.0,
"step": 8620
},
{
"entropy": 6.343156576156616,
"epoch": 0.9953837276399308,
"grad_norm": 0.86328125,
"learning_rate": 0.000491259600506409,
"loss": 6.151,
"mean_token_accuracy": 0.16660192161798476,
"num_tokens": 21860861.0,
"step": 8625
},
{
"entropy": 6.298342561721801,
"epoch": 0.9959607616849394,
"grad_norm": 0.9609375,
"learning_rate": 0.0004912482073544716,
"loss": 5.988,
"mean_token_accuracy": 0.1818713441491127,
"num_tokens": 21872380.0,
"step": 8630
},
{
"entropy": 6.234196710586548,
"epoch": 0.996537795729948,
"grad_norm": 0.81640625,
"learning_rate": 0.0004912368069291333,
"loss": 6.085,
"mean_token_accuracy": 0.17358871251344682,
"num_tokens": 21885322.0,
"step": 8635
},
{
"entropy": 6.356041002273559,
"epoch": 0.9971148297749567,
"grad_norm": 0.875,
"learning_rate": 0.0004912253992307773,
"loss": 6.1464,
"mean_token_accuracy": 0.1712245300412178,
"num_tokens": 21898243.0,
"step": 8640
},
{
"entropy": 6.352480220794678,
"epoch": 0.9976918638199653,
"grad_norm": 0.86328125,
"learning_rate": 0.0004912139842597875,
"loss": 6.0877,
"mean_token_accuracy": 0.16762156933546066,
"num_tokens": 21911467.0,
"step": 8645
},
{
"entropy": 6.307034111022949,
"epoch": 0.9982688978649741,
"grad_norm": 0.84765625,
"learning_rate": 0.0004912025620165477,
"loss": 6.0945,
"mean_token_accuracy": 0.16763445287942885,
"num_tokens": 21925583.0,
"step": 8650
},
{
"entropy": 6.285261440277099,
"epoch": 0.9988459319099827,
"grad_norm": 0.8125,
"learning_rate": 0.0004911911325014421,
"loss": 6.1352,
"mean_token_accuracy": 0.16875722408294677,
"num_tokens": 21937627.0,
"step": 8655
},
{
"entropy": 6.432320833206177,
"epoch": 0.9994229659549914,
"grad_norm": 0.90625,
"learning_rate": 0.0004911796957148552,
"loss": 6.1302,
"mean_token_accuracy": 0.16755087822675704,
"num_tokens": 21950376.0,
"step": 8660
},
{
"entropy": 6.299717378616333,
"epoch": 1.0,
"grad_norm": 0.81640625,
"learning_rate": 0.0004911682516571715,
"loss": 6.0704,
"mean_token_accuracy": 0.17948030233383178,
"num_tokens": 21963117.0,
"step": 8665
},
{
"entropy": 6.186800336837768,
"epoch": 1.0005770340450086,
"grad_norm": 0.8984375,
"learning_rate": 0.0004911568003287761,
"loss": 5.9359,
"mean_token_accuracy": 0.1857394814491272,
"num_tokens": 21976256.0,
"step": 8670
},
{
"entropy": 6.2768176078796385,
"epoch": 1.0011540680900173,
"grad_norm": 0.91796875,
"learning_rate": 0.0004911453417300541,
"loss": 5.9421,
"mean_token_accuracy": 0.1747553050518036,
"num_tokens": 21988768.0,
"step": 8675
},
{
"entropy": 6.32557692527771,
"epoch": 1.001731102135026,
"grad_norm": 0.85546875,
"learning_rate": 0.0004911338758613909,
"loss": 5.9736,
"mean_token_accuracy": 0.17850890904664993,
"num_tokens": 22000570.0,
"step": 8680
},
{
"entropy": 6.319621515274048,
"epoch": 1.0023081361800346,
"grad_norm": 0.8828125,
"learning_rate": 0.0004911224027231722,
"loss": 6.0171,
"mean_token_accuracy": 0.17065517157316207,
"num_tokens": 22012914.0,
"step": 8685
},
{
"entropy": 6.350254249572754,
"epoch": 1.0028851702250432,
"grad_norm": 0.92578125,
"learning_rate": 0.0004911109223157838,
"loss": 6.0062,
"mean_token_accuracy": 0.16737566739320756,
"num_tokens": 22025474.0,
"step": 8690
},
{
"entropy": 6.246565246582032,
"epoch": 1.0034622042700518,
"grad_norm": 0.82421875,
"learning_rate": 0.0004910994346396118,
"loss": 5.873,
"mean_token_accuracy": 0.1791422635316849,
"num_tokens": 22039228.0,
"step": 8695
},
{
"entropy": 6.249841737747192,
"epoch": 1.0040392383150605,
"grad_norm": 0.9296875,
"learning_rate": 0.0004910879396950427,
"loss": 6.0373,
"mean_token_accuracy": 0.17558213770389558,
"num_tokens": 22050153.0,
"step": 8700
},
{
"entropy": 6.333817386627198,
"epoch": 1.0046162723600693,
"grad_norm": 0.8203125,
"learning_rate": 0.000491076437482463,
"loss": 6.0313,
"mean_token_accuracy": 0.1722121447324753,
"num_tokens": 22063214.0,
"step": 8705
},
{
"entropy": 6.260972261428833,
"epoch": 1.005193306405078,
"grad_norm": 0.91015625,
"learning_rate": 0.0004910649280022599,
"loss": 5.9334,
"mean_token_accuracy": 0.17726417928934096,
"num_tokens": 22076032.0,
"step": 8710
},
{
"entropy": 6.274173450469971,
"epoch": 1.0057703404500866,
"grad_norm": 0.8203125,
"learning_rate": 0.0004910534112548201,
"loss": 5.9598,
"mean_token_accuracy": 0.1748859703540802,
"num_tokens": 22089134.0,
"step": 8715
},
{
"entropy": 6.288155460357666,
"epoch": 1.0063473744950953,
"grad_norm": 0.83984375,
"learning_rate": 0.0004910418872405312,
"loss": 5.9799,
"mean_token_accuracy": 0.17057327181100845,
"num_tokens": 22102772.0,
"step": 8720
},
{
"entropy": 6.227526712417602,
"epoch": 1.006924408540104,
"grad_norm": 0.796875,
"learning_rate": 0.0004910303559597806,
"loss": 5.9699,
"mean_token_accuracy": 0.1796274870634079,
"num_tokens": 22117586.0,
"step": 8725
},
{
"entropy": 6.263715505599976,
"epoch": 1.0075014425851125,
"grad_norm": 0.8671875,
"learning_rate": 0.0004910188174129564,
"loss": 6.032,
"mean_token_accuracy": 0.17084551006555557,
"num_tokens": 22129229.0,
"step": 8730
},
{
"entropy": 6.3296393871307375,
"epoch": 1.0080784766301212,
"grad_norm": 0.953125,
"learning_rate": 0.0004910072716004466,
"loss": 5.9848,
"mean_token_accuracy": 0.17309292703866958,
"num_tokens": 22142269.0,
"step": 8735
},
{
"entropy": 6.32510347366333,
"epoch": 1.0086555106751298,
"grad_norm": 0.8359375,
"learning_rate": 0.0004909957185226394,
"loss": 5.9946,
"mean_token_accuracy": 0.17437432706356049,
"num_tokens": 22153744.0,
"step": 8740
},
{
"entropy": 6.208776569366455,
"epoch": 1.0092325447201385,
"grad_norm": 1.4140625,
"learning_rate": 0.0004909841581799236,
"loss": 5.9027,
"mean_token_accuracy": 0.17276026010513307,
"num_tokens": 22166927.0,
"step": 8745
},
{
"entropy": 6.29905138015747,
"epoch": 1.009809578765147,
"grad_norm": 0.8359375,
"learning_rate": 0.0004909725905726879,
"loss": 5.9909,
"mean_token_accuracy": 0.1712697610259056,
"num_tokens": 22179003.0,
"step": 8750
},
{
"entropy": 6.315261030197144,
"epoch": 1.0103866128101557,
"grad_norm": 0.90625,
"learning_rate": 0.0004909610157013214,
"loss": 6.0611,
"mean_token_accuracy": 0.17488325536251068,
"num_tokens": 22191221.0,
"step": 8755
},
{
"entropy": 6.272414875030518,
"epoch": 1.0109636468551644,
"grad_norm": 0.84765625,
"learning_rate": 0.0004909494335662134,
"loss": 6.0023,
"mean_token_accuracy": 0.1766509383916855,
"num_tokens": 22203387.0,
"step": 8760
},
{
"entropy": 6.285401821136475,
"epoch": 1.011540680900173,
"grad_norm": 0.82421875,
"learning_rate": 0.0004909378441677535,
"loss": 5.9543,
"mean_token_accuracy": 0.1828874871134758,
"num_tokens": 22215225.0,
"step": 8765
},
{
"entropy": 6.282279682159424,
"epoch": 1.0121177149451817,
"grad_norm": 0.91796875,
"learning_rate": 0.0004909262475063314,
"loss": 5.9417,
"mean_token_accuracy": 0.18124219477176667,
"num_tokens": 22226390.0,
"step": 8770
},
{
"entropy": 6.24841890335083,
"epoch": 1.0126947489901905,
"grad_norm": 0.9140625,
"learning_rate": 0.0004909146435823373,
"loss": 6.0019,
"mean_token_accuracy": 0.176693457365036,
"num_tokens": 22238884.0,
"step": 8775
},
{
"entropy": 6.337589979171753,
"epoch": 1.0132717830351992,
"grad_norm": 0.84375,
"learning_rate": 0.0004909030323961613,
"loss": 6.1228,
"mean_token_accuracy": 0.16325929015874863,
"num_tokens": 22252214.0,
"step": 8780
},
{
"entropy": 6.35567684173584,
"epoch": 1.0138488170802078,
"grad_norm": 0.859375,
"learning_rate": 0.0004908914139481942,
"loss": 5.9934,
"mean_token_accuracy": 0.1741134375333786,
"num_tokens": 22264814.0,
"step": 8785
},
{
"entropy": 6.22755389213562,
"epoch": 1.0144258511252164,
"grad_norm": 0.7734375,
"learning_rate": 0.0004908797882388265,
"loss": 5.9401,
"mean_token_accuracy": 0.18325002491474152,
"num_tokens": 22278737.0,
"step": 8790
},
{
"entropy": 6.229189825057984,
"epoch": 1.015002885170225,
"grad_norm": 0.921875,
"learning_rate": 0.0004908681552684495,
"loss": 5.989,
"mean_token_accuracy": 0.17604121416807175,
"num_tokens": 22290974.0,
"step": 8795
},
{
"entropy": 6.2954566955566404,
"epoch": 1.0155799192152337,
"grad_norm": 0.9375,
"learning_rate": 0.0004908565150374542,
"loss": 5.9761,
"mean_token_accuracy": 0.1750330537557602,
"num_tokens": 22303080.0,
"step": 8800
},
{
"entropy": 6.286450815200806,
"epoch": 1.0161569532602424,
"grad_norm": 0.8359375,
"learning_rate": 0.0004908448675462323,
"loss": 6.0355,
"mean_token_accuracy": 0.1755423828959465,
"num_tokens": 22315536.0,
"step": 8805
},
{
"entropy": 6.214225816726684,
"epoch": 1.016733987305251,
"grad_norm": 0.921875,
"learning_rate": 0.0004908332127951756,
"loss": 5.9927,
"mean_token_accuracy": 0.1759694129228592,
"num_tokens": 22328114.0,
"step": 8810
},
{
"entropy": 6.273038387298584,
"epoch": 1.0173110213502596,
"grad_norm": 0.953125,
"learning_rate": 0.0004908215507846757,
"loss": 5.9105,
"mean_token_accuracy": 0.18296556174755096,
"num_tokens": 22339608.0,
"step": 8815
},
{
"entropy": 6.29830846786499,
"epoch": 1.0178880553952683,
"grad_norm": 0.83984375,
"learning_rate": 0.0004908098815151254,
"loss": 5.9654,
"mean_token_accuracy": 0.1758978173136711,
"num_tokens": 22351594.0,
"step": 8820
},
{
"entropy": 6.278672885894776,
"epoch": 1.018465089440277,
"grad_norm": 0.78125,
"learning_rate": 0.0004907982049869168,
"loss": 6.0408,
"mean_token_accuracy": 0.17886385768651963,
"num_tokens": 22363978.0,
"step": 8825
},
{
"entropy": 6.371420907974243,
"epoch": 1.0190421234852856,
"grad_norm": 0.81640625,
"learning_rate": 0.0004907865212004428,
"loss": 6.027,
"mean_token_accuracy": 0.17178633213043212,
"num_tokens": 22377845.0,
"step": 8830
},
{
"entropy": 6.272815847396851,
"epoch": 1.0196191575302942,
"grad_norm": 0.953125,
"learning_rate": 0.0004907748301560963,
"loss": 5.9555,
"mean_token_accuracy": 0.1750282183289528,
"num_tokens": 22392654.0,
"step": 8835
},
{
"entropy": 6.312655305862426,
"epoch": 1.0201961915753028,
"grad_norm": 0.87109375,
"learning_rate": 0.0004907631318542707,
"loss": 6.0007,
"mean_token_accuracy": 0.17550213485956193,
"num_tokens": 22404713.0,
"step": 8840
},
{
"entropy": 6.37878384590149,
"epoch": 1.0207732256203117,
"grad_norm": 0.8515625,
"learning_rate": 0.0004907514262953594,
"loss": 6.0247,
"mean_token_accuracy": 0.1659175708889961,
"num_tokens": 22417879.0,
"step": 8845
},
{
"entropy": 6.241927146911621,
"epoch": 1.0213502596653203,
"grad_norm": 0.828125,
"learning_rate": 0.0004907397134797559,
"loss": 5.9787,
"mean_token_accuracy": 0.17025523334741594,
"num_tokens": 22430674.0,
"step": 8850
},
{
"entropy": 6.322333812713623,
"epoch": 1.021927293710329,
"grad_norm": 0.93359375,
"learning_rate": 0.0004907279934078543,
"loss": 5.952,
"mean_token_accuracy": 0.18116206377744676,
"num_tokens": 22443135.0,
"step": 8855
},
{
"entropy": 6.23037805557251,
"epoch": 1.0225043277553376,
"grad_norm": 0.85546875,
"learning_rate": 0.0004907162660800488,
"loss": 5.9345,
"mean_token_accuracy": 0.17716160118579866,
"num_tokens": 22455832.0,
"step": 8860
},
{
"entropy": 6.340549373626709,
"epoch": 1.0230813618003463,
"grad_norm": 0.83984375,
"learning_rate": 0.0004907045314967338,
"loss": 5.992,
"mean_token_accuracy": 0.17303016632795334,
"num_tokens": 22467466.0,
"step": 8865
},
{
"entropy": 6.338985824584961,
"epoch": 1.023658395845355,
"grad_norm": 0.828125,
"learning_rate": 0.0004906927896583041,
"loss": 6.0278,
"mean_token_accuracy": 0.1670863524079323,
"num_tokens": 22481127.0,
"step": 8870
},
{
"entropy": 6.263080072402954,
"epoch": 1.0242354298903635,
"grad_norm": 0.9140625,
"learning_rate": 0.0004906810405651546,
"loss": 6.0232,
"mean_token_accuracy": 0.17175393402576447,
"num_tokens": 22492894.0,
"step": 8875
},
{
"entropy": 6.383795213699341,
"epoch": 1.0248124639353722,
"grad_norm": 0.83984375,
"learning_rate": 0.0004906692842176803,
"loss": 6.0205,
"mean_token_accuracy": 0.1698385149240494,
"num_tokens": 22505369.0,
"step": 8880
},
{
"entropy": 6.239025115966797,
"epoch": 1.0253894979803808,
"grad_norm": 0.8671875,
"learning_rate": 0.0004906575206162769,
"loss": 5.9382,
"mean_token_accuracy": 0.1769947186112404,
"num_tokens": 22517163.0,
"step": 8885
},
{
"entropy": 6.221582269668579,
"epoch": 1.0259665320253895,
"grad_norm": 0.83984375,
"learning_rate": 0.0004906457497613399,
"loss": 6.017,
"mean_token_accuracy": 0.17070773839950562,
"num_tokens": 22531132.0,
"step": 8890
},
{
"entropy": 6.361068964004517,
"epoch": 1.026543566070398,
"grad_norm": 0.8671875,
"learning_rate": 0.0004906339716532651,
"loss": 6.0213,
"mean_token_accuracy": 0.1671837493777275,
"num_tokens": 22542663.0,
"step": 8895
},
{
"entropy": 6.330655860900879,
"epoch": 1.0271206001154067,
"grad_norm": 0.8046875,
"learning_rate": 0.0004906221862924488,
"loss": 6.0019,
"mean_token_accuracy": 0.17416499108076094,
"num_tokens": 22555397.0,
"step": 8900
},
{
"entropy": 6.191199731826782,
"epoch": 1.0276976341604154,
"grad_norm": 0.84375,
"learning_rate": 0.0004906103936792875,
"loss": 5.9479,
"mean_token_accuracy": 0.17581079453229903,
"num_tokens": 22567373.0,
"step": 8905
},
{
"entropy": 6.29790563583374,
"epoch": 1.028274668205424,
"grad_norm": 0.859375,
"learning_rate": 0.0004905985938141777,
"loss": 5.9528,
"mean_token_accuracy": 0.1770506665110588,
"num_tokens": 22579200.0,
"step": 8910
},
{
"entropy": 6.292901706695557,
"epoch": 1.0288517022504329,
"grad_norm": 0.8984375,
"learning_rate": 0.0004905867866975163,
"loss": 5.9836,
"mean_token_accuracy": 0.17233404964208604,
"num_tokens": 22592940.0,
"step": 8915
},
{
"entropy": 6.292080307006836,
"epoch": 1.0294287362954415,
"grad_norm": 0.87109375,
"learning_rate": 0.0004905749723297003,
"loss": 5.9885,
"mean_token_accuracy": 0.17801400125026703,
"num_tokens": 22605440.0,
"step": 8920
},
{
"entropy": 6.242578601837158,
"epoch": 1.0300057703404502,
"grad_norm": 0.90625,
"learning_rate": 0.0004905631507111271,
"loss": 5.9847,
"mean_token_accuracy": 0.17813263237476348,
"num_tokens": 22618976.0,
"step": 8925
},
{
"entropy": 6.223312425613403,
"epoch": 1.0305828043854588,
"grad_norm": 0.8984375,
"learning_rate": 0.0004905513218421946,
"loss": 5.8729,
"mean_token_accuracy": 0.18013641089200974,
"num_tokens": 22632356.0,
"step": 8930
},
{
"entropy": 6.253937244415283,
"epoch": 1.0311598384304674,
"grad_norm": 0.83984375,
"learning_rate": 0.0004905394857233004,
"loss": 5.9736,
"mean_token_accuracy": 0.1722057729959488,
"num_tokens": 22645272.0,
"step": 8935
},
{
"entropy": 6.306738805770874,
"epoch": 1.031736872475476,
"grad_norm": 0.875,
"learning_rate": 0.0004905276423548426,
"loss": 5.9495,
"mean_token_accuracy": 0.1715457633137703,
"num_tokens": 22657101.0,
"step": 8940
},
{
"entropy": 6.301843118667603,
"epoch": 1.0323139065204847,
"grad_norm": 0.875,
"learning_rate": 0.0004905157917372197,
"loss": 5.9317,
"mean_token_accuracy": 0.17873359471559525,
"num_tokens": 22669107.0,
"step": 8945
},
{
"entropy": 6.304869365692139,
"epoch": 1.0328909405654934,
"grad_norm": 0.91796875,
"learning_rate": 0.0004905039338708302,
"loss": 5.9678,
"mean_token_accuracy": 0.17514553815126419,
"num_tokens": 22681633.0,
"step": 8950
},
{
"entropy": 6.355607652664185,
"epoch": 1.033467974610502,
"grad_norm": 0.78515625,
"learning_rate": 0.0004904920687560728,
"loss": 6.0256,
"mean_token_accuracy": 0.17806240171194077,
"num_tokens": 22694985.0,
"step": 8955
},
{
"entropy": 6.259906005859375,
"epoch": 1.0340450086555106,
"grad_norm": 0.79296875,
"learning_rate": 0.0004904801963933469,
"loss": 5.9828,
"mean_token_accuracy": 0.174012191593647,
"num_tokens": 22707333.0,
"step": 8960
},
{
"entropy": 6.276315689086914,
"epoch": 1.0346220427005193,
"grad_norm": 0.9140625,
"learning_rate": 0.0004904683167830515,
"loss": 5.9586,
"mean_token_accuracy": 0.17578708678483962,
"num_tokens": 22721030.0,
"step": 8965
},
{
"entropy": 6.2770734310150145,
"epoch": 1.035199076745528,
"grad_norm": 0.8984375,
"learning_rate": 0.0004904564299255862,
"loss": 5.9283,
"mean_token_accuracy": 0.17963338494300843,
"num_tokens": 22732711.0,
"step": 8970
},
{
"entropy": 6.2471513748168945,
"epoch": 1.0357761107905366,
"grad_norm": 0.87109375,
"learning_rate": 0.0004904445358213511,
"loss": 5.9418,
"mean_token_accuracy": 0.18144085705280305,
"num_tokens": 22745815.0,
"step": 8975
},
{
"entropy": 6.251951503753662,
"epoch": 1.0363531448355452,
"grad_norm": 0.87109375,
"learning_rate": 0.0004904326344707461,
"loss": 5.9865,
"mean_token_accuracy": 0.1752885952591896,
"num_tokens": 22757590.0,
"step": 8980
},
{
"entropy": 6.382661581039429,
"epoch": 1.036930178880554,
"grad_norm": 0.85546875,
"learning_rate": 0.0004904207258741712,
"loss": 6.044,
"mean_token_accuracy": 0.16637400537729263,
"num_tokens": 22769841.0,
"step": 8985
},
{
"entropy": 6.220981550216675,
"epoch": 1.0375072129255627,
"grad_norm": 0.89453125,
"learning_rate": 0.0004904088100320274,
"loss": 5.9489,
"mean_token_accuracy": 0.18653181344270706,
"num_tokens": 22781642.0,
"step": 8990
},
{
"entropy": 6.30796365737915,
"epoch": 1.0380842469705713,
"grad_norm": 0.89453125,
"learning_rate": 0.0004903968869447151,
"loss": 5.9683,
"mean_token_accuracy": 0.18010073900222778,
"num_tokens": 22794903.0,
"step": 8995
},
{
"entropy": 6.284050559997558,
"epoch": 1.03866128101558,
"grad_norm": 1.0234375,
"learning_rate": 0.0004903849566126356,
"loss": 6.0023,
"mean_token_accuracy": 0.18212944269180298,
"num_tokens": 22807839.0,
"step": 9000
},
{
"epoch": 1.03866128101558,
"eval_entropy": 6.075341097021225,
"eval_loss": 6.107968330383301,
"eval_mean_token_accuracy": 0.17483100661346443,
"eval_num_tokens": 22807839.0,
"eval_runtime": 17.5957,
"eval_samples_per_second": 1599.026,
"eval_steps_per_second": 199.878,
"step": 9000
}
],
"logging_steps": 5,
"max_steps": 86650,
"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": 4989038575288320.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink