c3e92c4ecd
Model: fpadovani/eus-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
16211 lines
443 KiB
JSON
16211 lines
443 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.8732273313278895,
|
|
"eval_steps": 500,
|
|
"global_step": 8000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.318607425689697,
|
|
"epoch": 0.004297378599054577,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.4191,
|
|
"mean_token_accuracy": 0.09157993644475937,
|
|
"num_tokens": 8559.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.338263130187988,
|
|
"epoch": 0.008594757198109154,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.3752,
|
|
"mean_token_accuracy": 0.09438179731369019,
|
|
"num_tokens": 18235.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.412753295898438,
|
|
"epoch": 0.01289213579716373,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.378,
|
|
"mean_token_accuracy": 0.0941568873822689,
|
|
"num_tokens": 28257.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.402242660522461,
|
|
"epoch": 0.017189514396218308,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.3608,
|
|
"mean_token_accuracy": 0.09274673461914062,
|
|
"num_tokens": 38142.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.447778749465942,
|
|
"epoch": 0.021486892995272882,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.389,
|
|
"mean_token_accuracy": 0.09097675308585167,
|
|
"num_tokens": 47363.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.516436624526977,
|
|
"epoch": 0.02578427159432746,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.336,
|
|
"mean_token_accuracy": 0.10141076073050499,
|
|
"num_tokens": 57121.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.562677669525146,
|
|
"epoch": 0.030081650193382038,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.3441,
|
|
"mean_token_accuracy": 0.09505503103137017,
|
|
"num_tokens": 66785.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.522300148010254,
|
|
"epoch": 0.034379028792436615,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.2518,
|
|
"mean_token_accuracy": 0.1012355662882328,
|
|
"num_tokens": 76313.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.549300718307495,
|
|
"epoch": 0.03867640739149119,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.3611,
|
|
"mean_token_accuracy": 0.0938008576631546,
|
|
"num_tokens": 86151.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.5571035861969,
|
|
"epoch": 0.042973785990545764,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.3491,
|
|
"mean_token_accuracy": 0.09041639566421508,
|
|
"num_tokens": 94916.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.54373984336853,
|
|
"epoch": 0.047271164589600345,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.2944,
|
|
"mean_token_accuracy": 0.09890028536319732,
|
|
"num_tokens": 104592.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.575749158859253,
|
|
"epoch": 0.05156854318865492,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.2915,
|
|
"mean_token_accuracy": 0.10528360232710839,
|
|
"num_tokens": 113942.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.521316862106323,
|
|
"epoch": 0.055865921787709494,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.3286,
|
|
"mean_token_accuracy": 0.09918183535337448,
|
|
"num_tokens": 123190.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.60635232925415,
|
|
"epoch": 0.060163300386764075,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.5078,
|
|
"mean_token_accuracy": 0.08831268101930619,
|
|
"num_tokens": 132367.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.664913272857666,
|
|
"epoch": 0.06446067898581866,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.2721,
|
|
"mean_token_accuracy": 0.10505343601107597,
|
|
"num_tokens": 141375.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.561436700820923,
|
|
"epoch": 0.06875805758487323,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.2566,
|
|
"mean_token_accuracy": 0.10234465301036835,
|
|
"num_tokens": 150791.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.493262481689453,
|
|
"epoch": 0.0730554361839278,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.3507,
|
|
"mean_token_accuracy": 0.09232475236058235,
|
|
"num_tokens": 158686.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.559574556350708,
|
|
"epoch": 0.07735281478298238,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.2842,
|
|
"mean_token_accuracy": 0.09609017446637154,
|
|
"num_tokens": 167482.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.523833465576172,
|
|
"epoch": 0.08165019338203695,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.1847,
|
|
"mean_token_accuracy": 0.10569239854812622,
|
|
"num_tokens": 176598.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.474951887130738,
|
|
"epoch": 0.08594757198109153,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.1612,
|
|
"mean_token_accuracy": 0.10713009536266327,
|
|
"num_tokens": 185883.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.453275728225708,
|
|
"epoch": 0.09024495058014612,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.2229,
|
|
"mean_token_accuracy": 0.099655120074749,
|
|
"num_tokens": 194820.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.569756937026978,
|
|
"epoch": 0.09454232917920069,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.2667,
|
|
"mean_token_accuracy": 0.09399808198213577,
|
|
"num_tokens": 203710.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.516098976135254,
|
|
"epoch": 0.09883970777825526,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.2409,
|
|
"mean_token_accuracy": 0.10409615635871887,
|
|
"num_tokens": 212642.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.51554684638977,
|
|
"epoch": 0.10313708637730984,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.2735,
|
|
"mean_token_accuracy": 0.09576103389263153,
|
|
"num_tokens": 221411.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.539292192459106,
|
|
"epoch": 0.10743446497636441,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.1672,
|
|
"mean_token_accuracy": 0.10191577151417733,
|
|
"num_tokens": 230454.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.474944972991944,
|
|
"epoch": 0.11173184357541899,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.2508,
|
|
"mean_token_accuracy": 0.09938024953007699,
|
|
"num_tokens": 240272.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.49649453163147,
|
|
"epoch": 0.11602922217447358,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.2856,
|
|
"mean_token_accuracy": 0.10054537132382393,
|
|
"num_tokens": 249652.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.581856298446655,
|
|
"epoch": 0.12032660077352815,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.1517,
|
|
"mean_token_accuracy": 0.1040219210088253,
|
|
"num_tokens": 258359.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.510291481018067,
|
|
"epoch": 0.12462397937258272,
|
|
"grad_norm": 1.9296875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.1336,
|
|
"mean_token_accuracy": 0.10528543666005134,
|
|
"num_tokens": 266863.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.496348476409912,
|
|
"epoch": 0.1289213579716373,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.2272,
|
|
"mean_token_accuracy": 0.10392259955406188,
|
|
"num_tokens": 274694.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.5612743377685545,
|
|
"epoch": 0.1332187365706919,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3341,
|
|
"mean_token_accuracy": 0.09439479187130928,
|
|
"num_tokens": 285038.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.626920127868653,
|
|
"epoch": 0.13751611516974646,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.2748,
|
|
"mean_token_accuracy": 0.09491272494196892,
|
|
"num_tokens": 293451.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.397944831848145,
|
|
"epoch": 0.14181349376880104,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.1126,
|
|
"mean_token_accuracy": 0.10942281857132911,
|
|
"num_tokens": 302184.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.540780925750733,
|
|
"epoch": 0.1461108723678556,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2473,
|
|
"mean_token_accuracy": 0.10303145796060562,
|
|
"num_tokens": 311069.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.574706363677978,
|
|
"epoch": 0.15040825096691018,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.2454,
|
|
"mean_token_accuracy": 0.09808394089341163,
|
|
"num_tokens": 319841.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.498692369461059,
|
|
"epoch": 0.15470562956596476,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.225,
|
|
"mean_token_accuracy": 0.09445683807134628,
|
|
"num_tokens": 328621.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.5258214473724365,
|
|
"epoch": 0.15900300816501933,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.204,
|
|
"mean_token_accuracy": 0.10020598471164703,
|
|
"num_tokens": 336469.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.548258686065674,
|
|
"epoch": 0.1633003867640739,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2766,
|
|
"mean_token_accuracy": 0.09492596611380577,
|
|
"num_tokens": 344859.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.586396789550781,
|
|
"epoch": 0.16759776536312848,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.2416,
|
|
"mean_token_accuracy": 0.09392696470022202,
|
|
"num_tokens": 354536.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.494244813919067,
|
|
"epoch": 0.17189514396218306,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2726,
|
|
"mean_token_accuracy": 0.10099927261471749,
|
|
"num_tokens": 363592.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.5656694889068605,
|
|
"epoch": 0.17619252256123766,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.1745,
|
|
"mean_token_accuracy": 0.10011893808841706,
|
|
"num_tokens": 373278.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.549917793273925,
|
|
"epoch": 0.18048990116029223,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2723,
|
|
"mean_token_accuracy": 0.09688640609383584,
|
|
"num_tokens": 382498.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.570077800750733,
|
|
"epoch": 0.1847872797593468,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1941,
|
|
"mean_token_accuracy": 0.10630622431635857,
|
|
"num_tokens": 391273.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.451020860671997,
|
|
"epoch": 0.18908465835840138,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1551,
|
|
"mean_token_accuracy": 0.09831803515553475,
|
|
"num_tokens": 400340.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.513375282287598,
|
|
"epoch": 0.19338203695745596,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1945,
|
|
"mean_token_accuracy": 0.1025567151606083,
|
|
"num_tokens": 409162.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.531175994873047,
|
|
"epoch": 0.19767941555651053,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.1772,
|
|
"mean_token_accuracy": 0.103980852663517,
|
|
"num_tokens": 417345.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.446167516708374,
|
|
"epoch": 0.2019767941555651,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1774,
|
|
"mean_token_accuracy": 0.10338029116392136,
|
|
"num_tokens": 427488.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.533880424499512,
|
|
"epoch": 0.20627417275461968,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1371,
|
|
"mean_token_accuracy": 0.1055621288716793,
|
|
"num_tokens": 435496.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.404056406021118,
|
|
"epoch": 0.21057155135367425,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.089,
|
|
"mean_token_accuracy": 0.10534645691514015,
|
|
"num_tokens": 444290.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.516732883453369,
|
|
"epoch": 0.21486892995272883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2072,
|
|
"mean_token_accuracy": 0.1041240192949772,
|
|
"num_tokens": 452704.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.54860291481018,
|
|
"epoch": 0.2191663085517834,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1303,
|
|
"mean_token_accuracy": 0.10454478338360787,
|
|
"num_tokens": 461445.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.390961217880249,
|
|
"epoch": 0.22346368715083798,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1877,
|
|
"mean_token_accuracy": 0.10488151460886001,
|
|
"num_tokens": 470644.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.427588891983032,
|
|
"epoch": 0.22776106574989258,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1273,
|
|
"mean_token_accuracy": 0.10415012761950493,
|
|
"num_tokens": 479212.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.486822843551636,
|
|
"epoch": 0.23205844434894715,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2598,
|
|
"mean_token_accuracy": 0.09593637436628341,
|
|
"num_tokens": 489027.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.549335575103759,
|
|
"epoch": 0.23635582294800173,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2775,
|
|
"mean_token_accuracy": 0.10373037382960319,
|
|
"num_tokens": 497548.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.466809272766113,
|
|
"epoch": 0.2406532015470563,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0831,
|
|
"mean_token_accuracy": 0.10918937921524048,
|
|
"num_tokens": 505836.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.526752281188965,
|
|
"epoch": 0.24495058014611087,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.231,
|
|
"mean_token_accuracy": 0.09891332313418388,
|
|
"num_tokens": 515463.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.458172607421875,
|
|
"epoch": 0.24924795874516545,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1369,
|
|
"mean_token_accuracy": 0.10318988189101219,
|
|
"num_tokens": 524882.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.400683259963989,
|
|
"epoch": 0.25354533734422,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1488,
|
|
"mean_token_accuracy": 0.09986984804272651,
|
|
"num_tokens": 534523.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.545065259933471,
|
|
"epoch": 0.2578427159432746,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.1232,
|
|
"mean_token_accuracy": 0.10373991057276725,
|
|
"num_tokens": 543772.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.3497340202331545,
|
|
"epoch": 0.26214009454232917,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0968,
|
|
"mean_token_accuracy": 0.1055608831346035,
|
|
"num_tokens": 552024.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.485379552841186,
|
|
"epoch": 0.2664374731413838,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1063,
|
|
"mean_token_accuracy": 0.1070250503718853,
|
|
"num_tokens": 561114.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.523478364944458,
|
|
"epoch": 0.2707348517404383,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.1463,
|
|
"mean_token_accuracy": 0.10905889123678207,
|
|
"num_tokens": 569655.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.377780628204346,
|
|
"epoch": 0.2750322303394929,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1831,
|
|
"mean_token_accuracy": 0.10561417937278747,
|
|
"num_tokens": 578701.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.41802134513855,
|
|
"epoch": 0.27932960893854747,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.1493,
|
|
"mean_token_accuracy": 0.10375717356801033,
|
|
"num_tokens": 588166.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.434600496292115,
|
|
"epoch": 0.28362698753760207,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1779,
|
|
"mean_token_accuracy": 0.10254393517971039,
|
|
"num_tokens": 598862.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.429252767562867,
|
|
"epoch": 0.2879243661366566,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.1318,
|
|
"mean_token_accuracy": 0.10302076190710067,
|
|
"num_tokens": 607680.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.45730152130127,
|
|
"epoch": 0.2922217447357112,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1604,
|
|
"mean_token_accuracy": 0.09979369044303894,
|
|
"num_tokens": 617194.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.370795726776123,
|
|
"epoch": 0.29651912333476577,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1056,
|
|
"mean_token_accuracy": 0.10890763849020005,
|
|
"num_tokens": 625946.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.480315494537353,
|
|
"epoch": 0.30081650193382037,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0787,
|
|
"mean_token_accuracy": 0.10798180177807808,
|
|
"num_tokens": 634293.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.34269905090332,
|
|
"epoch": 0.30511388053287497,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0733,
|
|
"mean_token_accuracy": 0.106089236587286,
|
|
"num_tokens": 643479.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.386287689208984,
|
|
"epoch": 0.3094112591319295,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1175,
|
|
"mean_token_accuracy": 0.1021731823682785,
|
|
"num_tokens": 653599.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.3755628108978275,
|
|
"epoch": 0.3137086377309841,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0971,
|
|
"mean_token_accuracy": 0.10849021822214126,
|
|
"num_tokens": 663042.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.371647834777832,
|
|
"epoch": 0.31800601633003867,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.036,
|
|
"mean_token_accuracy": 0.10965228900313377,
|
|
"num_tokens": 672254.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.470899868011474,
|
|
"epoch": 0.32230339492909327,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0761,
|
|
"mean_token_accuracy": 0.10654274672269821,
|
|
"num_tokens": 681648.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.253026962280273,
|
|
"epoch": 0.3266007735281478,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0529,
|
|
"mean_token_accuracy": 0.11080088913440704,
|
|
"num_tokens": 691080.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.3884796619415285,
|
|
"epoch": 0.3308981521272024,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1391,
|
|
"mean_token_accuracy": 0.10383651852607727,
|
|
"num_tokens": 700549.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.5841371536254885,
|
|
"epoch": 0.33519553072625696,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1858,
|
|
"mean_token_accuracy": 0.10275669917464256,
|
|
"num_tokens": 710623.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.3249256134033205,
|
|
"epoch": 0.33949290932531156,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1082,
|
|
"mean_token_accuracy": 0.10990465432405472,
|
|
"num_tokens": 719695.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.283933448791504,
|
|
"epoch": 0.3437902879243661,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.088,
|
|
"mean_token_accuracy": 0.10904877632856369,
|
|
"num_tokens": 728998.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.4869218349456785,
|
|
"epoch": 0.3480876665234207,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.1604,
|
|
"mean_token_accuracy": 0.10160251557826996,
|
|
"num_tokens": 738537.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.35856614112854,
|
|
"epoch": 0.3523850451224753,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0856,
|
|
"mean_token_accuracy": 0.10455845221877098,
|
|
"num_tokens": 747555.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.3205808162689205,
|
|
"epoch": 0.35668242372152986,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0854,
|
|
"mean_token_accuracy": 0.10954482555389404,
|
|
"num_tokens": 756857.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.390940237045288,
|
|
"epoch": 0.36097980232058446,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0953,
|
|
"mean_token_accuracy": 0.11360994949936867,
|
|
"num_tokens": 766181.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.433085155487061,
|
|
"epoch": 0.365277180919639,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.1085,
|
|
"mean_token_accuracy": 0.10497047677636147,
|
|
"num_tokens": 774894.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.305972146987915,
|
|
"epoch": 0.3695745595186936,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1319,
|
|
"mean_token_accuracy": 0.1049921877682209,
|
|
"num_tokens": 784601.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.370691442489624,
|
|
"epoch": 0.37387193811774816,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1739,
|
|
"mean_token_accuracy": 0.10319070667028427,
|
|
"num_tokens": 794373.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.330904865264893,
|
|
"epoch": 0.37816931671680276,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0547,
|
|
"mean_token_accuracy": 0.10838651731610298,
|
|
"num_tokens": 803608.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.375745105743408,
|
|
"epoch": 0.3824666953158573,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0837,
|
|
"mean_token_accuracy": 0.10409105122089386,
|
|
"num_tokens": 812287.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.335326385498047,
|
|
"epoch": 0.3867640739149119,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0273,
|
|
"mean_token_accuracy": 0.10988899692893028,
|
|
"num_tokens": 821049.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.3366615772247314,
|
|
"epoch": 0.39106145251396646,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.037,
|
|
"mean_token_accuracy": 0.10783056244254112,
|
|
"num_tokens": 829148.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.3818999290466305,
|
|
"epoch": 0.39535883111302106,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0454,
|
|
"mean_token_accuracy": 0.10884898602962494,
|
|
"num_tokens": 838228.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.300606727600098,
|
|
"epoch": 0.39965620971207566,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0491,
|
|
"mean_token_accuracy": 0.10492841824889183,
|
|
"num_tokens": 846059.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.4146685123443605,
|
|
"epoch": 0.4039535883111302,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.1786,
|
|
"mean_token_accuracy": 0.10016006529331208,
|
|
"num_tokens": 856309.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.28831844329834,
|
|
"epoch": 0.4082509669101848,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.0303,
|
|
"mean_token_accuracy": 0.11658978387713433,
|
|
"num_tokens": 865921.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.264179229736328,
|
|
"epoch": 0.41254834550923936,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.031,
|
|
"mean_token_accuracy": 0.11011900380253792,
|
|
"num_tokens": 875076.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.346679544448852,
|
|
"epoch": 0.41684572410829396,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.0691,
|
|
"mean_token_accuracy": 0.10555051863193513,
|
|
"num_tokens": 884586.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.398708009719849,
|
|
"epoch": 0.4211431027073485,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.1919,
|
|
"mean_token_accuracy": 0.10402772873640061,
|
|
"num_tokens": 894192.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.372455215454101,
|
|
"epoch": 0.4254404813064031,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.083,
|
|
"mean_token_accuracy": 0.10693729296326637,
|
|
"num_tokens": 903542.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.1894408702850345,
|
|
"epoch": 0.42973785990545765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.9916,
|
|
"mean_token_accuracy": 0.11179914698004723,
|
|
"num_tokens": 912625.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.42973785990545765,
|
|
"eval_entropy": 7.3068524684991925,
|
|
"eval_loss": 7.0247063636779785,
|
|
"eval_mean_token_accuracy": 0.11340155569953961,
|
|
"eval_num_tokens": 912625.0,
|
|
"eval_runtime": 2.6673,
|
|
"eval_samples_per_second": 1330.579,
|
|
"eval_steps_per_second": 166.463,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.301834917068481,
|
|
"epoch": 0.43403523850451226,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.9753,
|
|
"mean_token_accuracy": 0.11066327542066574,
|
|
"num_tokens": 921438.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.299079513549804,
|
|
"epoch": 0.4383326171035668,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9705,
|
|
"mean_token_accuracy": 0.11115480959415436,
|
|
"num_tokens": 931097.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.25611834526062,
|
|
"epoch": 0.4426299957026214,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9377,
|
|
"mean_token_accuracy": 0.1125190131366253,
|
|
"num_tokens": 939375.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.2075482368469235,
|
|
"epoch": 0.44692737430167595,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9692,
|
|
"mean_token_accuracy": 0.11348672360181808,
|
|
"num_tokens": 948223.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.267970848083496,
|
|
"epoch": 0.45122475290073055,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9881,
|
|
"mean_token_accuracy": 0.10696823000907899,
|
|
"num_tokens": 956243.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.269130039215088,
|
|
"epoch": 0.45552213149978515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.9888,
|
|
"mean_token_accuracy": 0.11132953837513923,
|
|
"num_tokens": 964506.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.2140302658081055,
|
|
"epoch": 0.4598195100988397,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.0261,
|
|
"mean_token_accuracy": 0.11418202742934228,
|
|
"num_tokens": 973617.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.263705730438232,
|
|
"epoch": 0.4641168886978943,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9285,
|
|
"mean_token_accuracy": 0.11379739046096801,
|
|
"num_tokens": 981796.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.2532874584198,
|
|
"epoch": 0.46841426729694885,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9974,
|
|
"mean_token_accuracy": 0.11635445952415466,
|
|
"num_tokens": 990677.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.17174711227417,
|
|
"epoch": 0.47271164589600345,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.9292,
|
|
"mean_token_accuracy": 0.11551985442638397,
|
|
"num_tokens": 1001131.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.255539464950561,
|
|
"epoch": 0.477009024495058,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.0782,
|
|
"mean_token_accuracy": 0.1087973214685917,
|
|
"num_tokens": 1009925.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.204084014892578,
|
|
"epoch": 0.4813064030941126,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.9989,
|
|
"mean_token_accuracy": 0.11738647520542145,
|
|
"num_tokens": 1018998.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.313282203674317,
|
|
"epoch": 0.48560378169316715,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.9487,
|
|
"mean_token_accuracy": 0.11487800031900405,
|
|
"num_tokens": 1027496.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.188301038742066,
|
|
"epoch": 0.48990116029222175,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.9178,
|
|
"mean_token_accuracy": 0.1234424039721489,
|
|
"num_tokens": 1035669.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.197154760360718,
|
|
"epoch": 0.4941985388912763,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.9966,
|
|
"mean_token_accuracy": 0.11247355863451958,
|
|
"num_tokens": 1044416.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.310844898223877,
|
|
"epoch": 0.4984959174903309,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.0271,
|
|
"mean_token_accuracy": 0.11254049837589264,
|
|
"num_tokens": 1053500.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.146087980270385,
|
|
"epoch": 0.5027932960893855,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.9275,
|
|
"mean_token_accuracy": 0.11227174699306489,
|
|
"num_tokens": 1061758.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.159999990463257,
|
|
"epoch": 0.50709067468844,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9957,
|
|
"mean_token_accuracy": 0.11408972144126892,
|
|
"num_tokens": 1070570.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.258799076080322,
|
|
"epoch": 0.5113880532874946,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.9414,
|
|
"mean_token_accuracy": 0.11279132589697838,
|
|
"num_tokens": 1080640.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.13523416519165,
|
|
"epoch": 0.5156854318865493,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.9924,
|
|
"mean_token_accuracy": 0.11432184651494026,
|
|
"num_tokens": 1090680.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.3078405380249025,
|
|
"epoch": 0.5199828104856038,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.0484,
|
|
"mean_token_accuracy": 0.10550333335995674,
|
|
"num_tokens": 1100453.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.091824913024903,
|
|
"epoch": 0.5242801890846583,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8893,
|
|
"mean_token_accuracy": 0.11649776622653008,
|
|
"num_tokens": 1109085.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.119379043579102,
|
|
"epoch": 0.5285775676837129,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8739,
|
|
"mean_token_accuracy": 0.11886096745729446,
|
|
"num_tokens": 1118333.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.0497113227844235,
|
|
"epoch": 0.5328749462827675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.12016497254371643,
|
|
"num_tokens": 1127459.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.180634450912476,
|
|
"epoch": 0.5371723248818221,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.994,
|
|
"mean_token_accuracy": 0.11278382316231728,
|
|
"num_tokens": 1137071.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.2187317371368405,
|
|
"epoch": 0.5414697034808766,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.9744,
|
|
"mean_token_accuracy": 0.10984951481223107,
|
|
"num_tokens": 1146856.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.1333935260772705,
|
|
"epoch": 0.5457670820799312,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.0027,
|
|
"mean_token_accuracy": 0.11289475411176682,
|
|
"num_tokens": 1155847.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.205038070678711,
|
|
"epoch": 0.5500644606789858,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.9754,
|
|
"mean_token_accuracy": 0.11328396424651146,
|
|
"num_tokens": 1164922.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.175393056869507,
|
|
"epoch": 0.5543618392780404,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.9435,
|
|
"mean_token_accuracy": 0.11000620648264885,
|
|
"num_tokens": 1174389.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.118865060806274,
|
|
"epoch": 0.5586592178770949,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.8989,
|
|
"mean_token_accuracy": 0.11063683554530143,
|
|
"num_tokens": 1182959.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.158196449279785,
|
|
"epoch": 0.5629565964761496,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.9443,
|
|
"mean_token_accuracy": 0.11405061110854149,
|
|
"num_tokens": 1191777.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.121678113937378,
|
|
"epoch": 0.5672539750752041,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.8375,
|
|
"mean_token_accuracy": 0.12199422866106033,
|
|
"num_tokens": 1200658.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.01548957824707,
|
|
"epoch": 0.5715513536742587,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.8453,
|
|
"mean_token_accuracy": 0.12067256942391395,
|
|
"num_tokens": 1209607.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.2269895553588865,
|
|
"epoch": 0.5758487322733132,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.974,
|
|
"mean_token_accuracy": 0.10879514962434769,
|
|
"num_tokens": 1218146.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.988537359237671,
|
|
"epoch": 0.5801461108723679,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.8482,
|
|
"mean_token_accuracy": 0.11418939754366875,
|
|
"num_tokens": 1226886.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.105102682113648,
|
|
"epoch": 0.5844434894714224,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.8729,
|
|
"mean_token_accuracy": 0.1209171898663044,
|
|
"num_tokens": 1235408.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.0300126552581785,
|
|
"epoch": 0.588740868070477,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8501,
|
|
"mean_token_accuracy": 0.12092051953077317,
|
|
"num_tokens": 1244643.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.138811111450195,
|
|
"epoch": 0.5930382466695315,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.9133,
|
|
"mean_token_accuracy": 0.11425379812717437,
|
|
"num_tokens": 1254145.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.102302742004395,
|
|
"epoch": 0.5973356252685862,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.9522,
|
|
"mean_token_accuracy": 0.11475471332669258,
|
|
"num_tokens": 1263395.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.116247606277466,
|
|
"epoch": 0.6016330038676407,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.8717,
|
|
"mean_token_accuracy": 0.12142749726772309,
|
|
"num_tokens": 1272942.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.9945995807647705,
|
|
"epoch": 0.6059303824666953,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.8167,
|
|
"mean_token_accuracy": 0.12329712808132172,
|
|
"num_tokens": 1281481.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.154071187973022,
|
|
"epoch": 0.6102277610657499,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.1102140225470066,
|
|
"num_tokens": 1290455.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.065611410140991,
|
|
"epoch": 0.6145251396648045,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.8602,
|
|
"mean_token_accuracy": 0.11892874240875244,
|
|
"num_tokens": 1299241.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.104707860946656,
|
|
"epoch": 0.618822518263859,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.9291,
|
|
"mean_token_accuracy": 0.1116059735417366,
|
|
"num_tokens": 1308658.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.139448833465576,
|
|
"epoch": 0.6231198968629136,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.8789,
|
|
"mean_token_accuracy": 0.11212217062711716,
|
|
"num_tokens": 1316987.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.924467325210571,
|
|
"epoch": 0.6274172754619682,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.8228,
|
|
"mean_token_accuracy": 0.12177187651395797,
|
|
"num_tokens": 1325500.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.004321479797364,
|
|
"epoch": 0.6317146540610228,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.9595,
|
|
"mean_token_accuracy": 0.11682300940155983,
|
|
"num_tokens": 1335048.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.079775619506836,
|
|
"epoch": 0.6360120326600773,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.8921,
|
|
"mean_token_accuracy": 0.11796021088957787,
|
|
"num_tokens": 1343879.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.036481952667236,
|
|
"epoch": 0.6403094112591319,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.8026,
|
|
"mean_token_accuracy": 0.1175395205616951,
|
|
"num_tokens": 1353169.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.024873685836792,
|
|
"epoch": 0.6446067898581865,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.9047,
|
|
"mean_token_accuracy": 0.11657271012663842,
|
|
"num_tokens": 1362912.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.031742286682129,
|
|
"epoch": 0.6489041684572411,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.9109,
|
|
"mean_token_accuracy": 0.1104318842291832,
|
|
"num_tokens": 1372331.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.986195516586304,
|
|
"epoch": 0.6532015470562956,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.7911,
|
|
"mean_token_accuracy": 0.12072895392775536,
|
|
"num_tokens": 1381060.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.975994014739991,
|
|
"epoch": 0.6574989256553503,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.7951,
|
|
"mean_token_accuracy": 0.13102718442678452,
|
|
"num_tokens": 1388749.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.983586311340332,
|
|
"epoch": 0.6617963042544048,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.11960498094558716,
|
|
"num_tokens": 1398456.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.966071653366089,
|
|
"epoch": 0.6660936828534594,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.7798,
|
|
"mean_token_accuracy": 0.12140673324465752,
|
|
"num_tokens": 1407430.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.90265302658081,
|
|
"epoch": 0.6703910614525139,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.8787,
|
|
"mean_token_accuracy": 0.11302192658185958,
|
|
"num_tokens": 1416631.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.030915307998657,
|
|
"epoch": 0.6746884400515686,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.8116,
|
|
"mean_token_accuracy": 0.11736064329743386,
|
|
"num_tokens": 1426124.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.964035367965698,
|
|
"epoch": 0.6789858186506231,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.9457,
|
|
"mean_token_accuracy": 0.11582199558615684,
|
|
"num_tokens": 1436510.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.125880765914917,
|
|
"epoch": 0.6832831972496777,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.8824,
|
|
"mean_token_accuracy": 0.1163998268544674,
|
|
"num_tokens": 1445954.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.923286104202271,
|
|
"epoch": 0.6875805758487322,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.8985,
|
|
"mean_token_accuracy": 0.10873348712921142,
|
|
"num_tokens": 1454981.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.04893126487732,
|
|
"epoch": 0.6918779544477869,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.8641,
|
|
"mean_token_accuracy": 0.11823213398456574,
|
|
"num_tokens": 1464904.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.886806869506836,
|
|
"epoch": 0.6961753330468414,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.7107,
|
|
"mean_token_accuracy": 0.12331509366631507,
|
|
"num_tokens": 1473605.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.912139225006103,
|
|
"epoch": 0.700472711645896,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.759,
|
|
"mean_token_accuracy": 0.1245518982410431,
|
|
"num_tokens": 1482989.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.9620767593383786,
|
|
"epoch": 0.7047700902449506,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.8448,
|
|
"mean_token_accuracy": 0.11647504344582557,
|
|
"num_tokens": 1492514.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.943895387649536,
|
|
"epoch": 0.7090674688440052,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.8546,
|
|
"mean_token_accuracy": 0.12430108189582825,
|
|
"num_tokens": 1501127.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.962851333618164,
|
|
"epoch": 0.7133648474430597,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.8742,
|
|
"mean_token_accuracy": 0.11583281457424163,
|
|
"num_tokens": 1509698.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.987609577178955,
|
|
"epoch": 0.7176622260421143,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.8135,
|
|
"mean_token_accuracy": 0.12093358635902404,
|
|
"num_tokens": 1518224.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.959578323364258,
|
|
"epoch": 0.7219596046411689,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.8131,
|
|
"mean_token_accuracy": 0.12489956095814705,
|
|
"num_tokens": 1527163.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.942628192901611,
|
|
"epoch": 0.7262569832402235,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.8483,
|
|
"mean_token_accuracy": 0.11925563141703606,
|
|
"num_tokens": 1536980.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.907382965087891,
|
|
"epoch": 0.730554361839278,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.8019,
|
|
"mean_token_accuracy": 0.12201033234596252,
|
|
"num_tokens": 1546388.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.943301010131836,
|
|
"epoch": 0.7348517404383326,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.8442,
|
|
"mean_token_accuracy": 0.11446970775723457,
|
|
"num_tokens": 1555618.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.032028865814209,
|
|
"epoch": 0.7391491190373872,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.9204,
|
|
"mean_token_accuracy": 0.11842655092477798,
|
|
"num_tokens": 1565398.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.910132169723511,
|
|
"epoch": 0.7434464976364418,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.78,
|
|
"mean_token_accuracy": 0.1209449052810669,
|
|
"num_tokens": 1574052.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.863509893417358,
|
|
"epoch": 0.7477438762354963,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.7197,
|
|
"mean_token_accuracy": 0.12694448828697205,
|
|
"num_tokens": 1582930.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.87777304649353,
|
|
"epoch": 0.752041254834551,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.8381,
|
|
"mean_token_accuracy": 0.11370284855365753,
|
|
"num_tokens": 1592007.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.8960123538970945,
|
|
"epoch": 0.7563386334336055,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.6681,
|
|
"mean_token_accuracy": 0.13288158252835275,
|
|
"num_tokens": 1600532.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.847743368148803,
|
|
"epoch": 0.7606360120326601,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8122,
|
|
"mean_token_accuracy": 0.11822389364242554,
|
|
"num_tokens": 1610319.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.883726263046265,
|
|
"epoch": 0.7649333906317146,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.6769,
|
|
"mean_token_accuracy": 0.12348684296011925,
|
|
"num_tokens": 1618161.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.87597885131836,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.7947,
|
|
"mean_token_accuracy": 0.12304572686553002,
|
|
"num_tokens": 1627676.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.900651979446411,
|
|
"epoch": 0.7735281478298238,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.7425,
|
|
"mean_token_accuracy": 0.11917232349514961,
|
|
"num_tokens": 1636693.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.785532140731812,
|
|
"epoch": 0.7778255264288784,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.7689,
|
|
"mean_token_accuracy": 0.11988367959856987,
|
|
"num_tokens": 1645577.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.942232894897461,
|
|
"epoch": 0.7821229050279329,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.7411,
|
|
"mean_token_accuracy": 0.11902217343449592,
|
|
"num_tokens": 1654239.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.833974695205688,
|
|
"epoch": 0.7864202836269876,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.7464,
|
|
"mean_token_accuracy": 0.12057677283883095,
|
|
"num_tokens": 1663085.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.830857181549073,
|
|
"epoch": 0.7907176622260421,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.7733,
|
|
"mean_token_accuracy": 0.12257009446620941,
|
|
"num_tokens": 1671621.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.914818286895752,
|
|
"epoch": 0.7950150408250967,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.7718,
|
|
"mean_token_accuracy": 0.12270039841532707,
|
|
"num_tokens": 1680912.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.676487874984741,
|
|
"epoch": 0.7993124194241513,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.6962,
|
|
"mean_token_accuracy": 0.129410083591938,
|
|
"num_tokens": 1689772.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.942879056930542,
|
|
"epoch": 0.8036097980232059,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.8636,
|
|
"mean_token_accuracy": 0.11327557042241096,
|
|
"num_tokens": 1698985.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.8534533977508545,
|
|
"epoch": 0.8079071766222604,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.7291,
|
|
"mean_token_accuracy": 0.12363906055688859,
|
|
"num_tokens": 1707365.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.760773754119873,
|
|
"epoch": 0.812204555221315,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.7532,
|
|
"mean_token_accuracy": 0.1252680540084839,
|
|
"num_tokens": 1715858.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.933451700210571,
|
|
"epoch": 0.8165019338203696,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.7832,
|
|
"mean_token_accuracy": 0.11664097607135773,
|
|
"num_tokens": 1724721.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.852610778808594,
|
|
"epoch": 0.8207993124194242,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.7964,
|
|
"mean_token_accuracy": 0.12085579261183739,
|
|
"num_tokens": 1733700.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.8043561458587645,
|
|
"epoch": 0.8250966910184787,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.7446,
|
|
"mean_token_accuracy": 0.11831567585468292,
|
|
"num_tokens": 1742167.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.86628828048706,
|
|
"epoch": 0.8293940696175333,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.7552,
|
|
"mean_token_accuracy": 0.12031449228525162,
|
|
"num_tokens": 1752087.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.81239686012268,
|
|
"epoch": 0.8336914482165879,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.7122,
|
|
"mean_token_accuracy": 0.1284126915037632,
|
|
"num_tokens": 1760749.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.679265117645263,
|
|
"epoch": 0.8379888268156425,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.7459,
|
|
"mean_token_accuracy": 0.11835021451115608,
|
|
"num_tokens": 1770923.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.882328510284424,
|
|
"epoch": 0.842286205414697,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.7311,
|
|
"mean_token_accuracy": 0.12391256019473076,
|
|
"num_tokens": 1779233.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.815756559371948,
|
|
"epoch": 0.8465835840137517,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.7245,
|
|
"mean_token_accuracy": 0.1235703021287918,
|
|
"num_tokens": 1787593.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.834695529937744,
|
|
"epoch": 0.8508809626128062,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.773,
|
|
"mean_token_accuracy": 0.11845792606472969,
|
|
"num_tokens": 1796655.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.734475326538086,
|
|
"epoch": 0.8551783412118608,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.701,
|
|
"mean_token_accuracy": 0.12803028374910355,
|
|
"num_tokens": 1805790.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.814859628677368,
|
|
"epoch": 0.8594757198109153,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.683,
|
|
"mean_token_accuracy": 0.11991076469421387,
|
|
"num_tokens": 1813776.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8594757198109153,
|
|
"eval_entropy": 6.726756938942918,
|
|
"eval_loss": 6.731381416320801,
|
|
"eval_mean_token_accuracy": 0.12728899561271473,
|
|
"eval_num_tokens": 1813776.0,
|
|
"eval_runtime": 2.8327,
|
|
"eval_samples_per_second": 1252.854,
|
|
"eval_steps_per_second": 156.739,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.696360111236572,
|
|
"epoch": 0.86377309840997,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999998427807679,
|
|
"loss": 6.6901,
|
|
"mean_token_accuracy": 0.12232941612601281,
|
|
"num_tokens": 1823034.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.810961771011352,
|
|
"epoch": 0.8680704770090245,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999992040780138,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.11861197948455811,
|
|
"num_tokens": 1832320.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.716808557510376,
|
|
"epoch": 0.8723678556080791,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999980740669294,
|
|
"loss": 6.6632,
|
|
"mean_token_accuracy": 0.12679338902235032,
|
|
"num_tokens": 1841165.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.7952981948852536,
|
|
"epoch": 0.8766652342071336,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999964527499823,
|
|
"loss": 6.7338,
|
|
"mean_token_accuracy": 0.12241052910685539,
|
|
"num_tokens": 1850440.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.801466560363769,
|
|
"epoch": 0.8809626128061883,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004999943401307127,
|
|
"loss": 6.7461,
|
|
"mean_token_accuracy": 0.12134231030941009,
|
|
"num_tokens": 1859304.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.778349208831787,
|
|
"epoch": 0.8852599914052428,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999917362137337,
|
|
"loss": 6.6069,
|
|
"mean_token_accuracy": 0.13296075016260148,
|
|
"num_tokens": 1867560.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.69033408164978,
|
|
"epoch": 0.8895573700042974,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999886410047312,
|
|
"loss": 6.6847,
|
|
"mean_token_accuracy": 0.1256091572344303,
|
|
"num_tokens": 1876742.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.799950695037841,
|
|
"epoch": 0.8938547486033519,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999850545104638,
|
|
"loss": 6.6921,
|
|
"mean_token_accuracy": 0.12495070844888687,
|
|
"num_tokens": 1885738.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.801503467559814,
|
|
"epoch": 0.8981521272024066,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999809767387633,
|
|
"loss": 6.7228,
|
|
"mean_token_accuracy": 0.11731479540467263,
|
|
"num_tokens": 1894688.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.757845163345337,
|
|
"epoch": 0.9024495058014611,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004999764076985337,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.12538044303655624,
|
|
"num_tokens": 1903809.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.814986515045166,
|
|
"epoch": 0.9067468844005157,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999713473997519,
|
|
"loss": 6.747,
|
|
"mean_token_accuracy": 0.12134177088737488,
|
|
"num_tokens": 1913558.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.723377370834351,
|
|
"epoch": 0.9110442629995703,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999657958534677,
|
|
"loss": 6.7283,
|
|
"mean_token_accuracy": 0.11966701298952102,
|
|
"num_tokens": 1923324.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.708611631393433,
|
|
"epoch": 0.9153416415986249,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999597530718034,
|
|
"loss": 6.7075,
|
|
"mean_token_accuracy": 0.12495688870549201,
|
|
"num_tokens": 1932849.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.832595348358154,
|
|
"epoch": 0.9196390201976794,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499953219067954,
|
|
"loss": 6.8184,
|
|
"mean_token_accuracy": 0.12064682915806771,
|
|
"num_tokens": 1942678.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.740026760101318,
|
|
"epoch": 0.923936398796734,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999461938561873,
|
|
"loss": 6.6653,
|
|
"mean_token_accuracy": 0.11836395263671876,
|
|
"num_tokens": 1951670.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.813496685028076,
|
|
"epoch": 0.9282337773957886,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999386774518432,
|
|
"loss": 6.7849,
|
|
"mean_token_accuracy": 0.12696564868092536,
|
|
"num_tokens": 1961060.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.821066284179688,
|
|
"epoch": 0.9325311559948432,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999306698713349,
|
|
"loss": 6.7251,
|
|
"mean_token_accuracy": 0.12872326821088792,
|
|
"num_tokens": 1970790.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.802606868743896,
|
|
"epoch": 0.9368285345938977,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004999221711321477,
|
|
"loss": 6.7599,
|
|
"mean_token_accuracy": 0.12338658943772315,
|
|
"num_tokens": 1980016.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.666169691085815,
|
|
"epoch": 0.9411259131929522,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999131812528393,
|
|
"loss": 6.7137,
|
|
"mean_token_accuracy": 0.12647809982299804,
|
|
"num_tokens": 1989542.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.856275272369385,
|
|
"epoch": 0.9454232917920069,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00049990370025304,
|
|
"loss": 6.6789,
|
|
"mean_token_accuracy": 0.1320645920932293,
|
|
"num_tokens": 1999212.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.65680422782898,
|
|
"epoch": 0.9497206703910615,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998937281534526,
|
|
"loss": 6.6665,
|
|
"mean_token_accuracy": 0.12952047884464263,
|
|
"num_tokens": 2008378.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.683283519744873,
|
|
"epoch": 0.954018048990116,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004998832649758521,
|
|
"loss": 6.7497,
|
|
"mean_token_accuracy": 0.11866686791181565,
|
|
"num_tokens": 2018985.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.716252756118775,
|
|
"epoch": 0.9583154275891707,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998723107430862,
|
|
"loss": 6.6294,
|
|
"mean_token_accuracy": 0.12505461126565934,
|
|
"num_tokens": 2027274.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.733844661712647,
|
|
"epoch": 0.9626128061882252,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998608654790741,
|
|
"loss": 6.7276,
|
|
"mean_token_accuracy": 0.12319627404212952,
|
|
"num_tokens": 2036892.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.80896143913269,
|
|
"epoch": 0.9669101847872797,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499848929208808,
|
|
"loss": 6.7032,
|
|
"mean_token_accuracy": 0.1196390450000763,
|
|
"num_tokens": 2046042.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.681768321990967,
|
|
"epoch": 0.9712075633863343,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998365019583519,
|
|
"loss": 6.656,
|
|
"mean_token_accuracy": 0.1205221801996231,
|
|
"num_tokens": 2054576.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.673296976089477,
|
|
"epoch": 0.975504941985389,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998235837548417,
|
|
"loss": 6.6372,
|
|
"mean_token_accuracy": 0.13002716973423958,
|
|
"num_tokens": 2063552.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.650518178939819,
|
|
"epoch": 0.9798023205844435,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499810174626486,
|
|
"loss": 6.6236,
|
|
"mean_token_accuracy": 0.1278505653142929,
|
|
"num_tokens": 2072965.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.689944839477539,
|
|
"epoch": 0.984099699183498,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004997962746025646,
|
|
"loss": 6.6977,
|
|
"mean_token_accuracy": 0.12715715765953065,
|
|
"num_tokens": 2082461.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.717270183563232,
|
|
"epoch": 0.9883970777825526,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004997818837134298,
|
|
"loss": 6.6697,
|
|
"mean_token_accuracy": 0.12612896114587785,
|
|
"num_tokens": 2091022.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.634740400314331,
|
|
"epoch": 0.9926944563816072,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004997670019905057,
|
|
"loss": 6.5604,
|
|
"mean_token_accuracy": 0.13522578105330468,
|
|
"num_tokens": 2099896.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.714453172683716,
|
|
"epoch": 0.9969918349806618,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997516294662876,
|
|
"loss": 6.671,
|
|
"mean_token_accuracy": 0.12892441749572753,
|
|
"num_tokens": 2109122.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.7551817364162865,
|
|
"epoch": 1.0008594757198108,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997357661743433,
|
|
"loss": 6.6203,
|
|
"mean_token_accuracy": 0.12951675057411194,
|
|
"num_tokens": 2117879.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.554825782775879,
|
|
"epoch": 1.0051568543188656,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997194121493118,
|
|
"loss": 6.3746,
|
|
"mean_token_accuracy": 0.13388435691595077,
|
|
"num_tokens": 2127552.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.640289449691773,
|
|
"epoch": 1.0094542329179201,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997025674269037,
|
|
"loss": 6.3859,
|
|
"mean_token_accuracy": 0.13642154037952423,
|
|
"num_tokens": 2136648.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.596437311172485,
|
|
"epoch": 1.0137516115169747,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996852320439013,
|
|
"loss": 6.4447,
|
|
"mean_token_accuracy": 0.13524584099650383,
|
|
"num_tokens": 2144852.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.616053581237793,
|
|
"epoch": 1.0180489901160292,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996674060381578,
|
|
"loss": 6.4566,
|
|
"mean_token_accuracy": 0.1324625827372074,
|
|
"num_tokens": 2154567.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.667360925674439,
|
|
"epoch": 1.0223463687150838,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996490894485985,
|
|
"loss": 6.4445,
|
|
"mean_token_accuracy": 0.1303381770849228,
|
|
"num_tokens": 2164260.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.516767692565918,
|
|
"epoch": 1.0266437473141383,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996302823152193,
|
|
"loss": 6.4128,
|
|
"mean_token_accuracy": 0.14060178995132447,
|
|
"num_tokens": 2173247.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.566610860824585,
|
|
"epoch": 1.0309411259131929,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996109846790873,
|
|
"loss": 6.4723,
|
|
"mean_token_accuracy": 0.12932706177234649,
|
|
"num_tokens": 2181595.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.6948295593261715,
|
|
"epoch": 1.0352385045122476,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995911965823412,
|
|
"loss": 6.3895,
|
|
"mean_token_accuracy": 0.1299505650997162,
|
|
"num_tokens": 2190608.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.462952327728272,
|
|
"epoch": 1.0395358831113022,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995709180681899,
|
|
"loss": 6.3982,
|
|
"mean_token_accuracy": 0.13950304985046386,
|
|
"num_tokens": 2199131.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.655004644393921,
|
|
"epoch": 1.0438332617103567,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499550149180914,
|
|
"loss": 6.5499,
|
|
"mean_token_accuracy": 0.1290518805384636,
|
|
"num_tokens": 2208365.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.545926237106324,
|
|
"epoch": 1.0481306403094113,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995288899658641,
|
|
"loss": 6.3942,
|
|
"mean_token_accuracy": 0.13622732385993003,
|
|
"num_tokens": 2216851.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.64662799835205,
|
|
"epoch": 1.0524280189084658,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995071404694619,
|
|
"loss": 6.4725,
|
|
"mean_token_accuracy": 0.12700082510709762,
|
|
"num_tokens": 2225309.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.5808178901672365,
|
|
"epoch": 1.0567253975075204,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004994849007391996,
|
|
"loss": 6.4282,
|
|
"mean_token_accuracy": 0.13448652923107146,
|
|
"num_tokens": 2235036.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.624714040756226,
|
|
"epoch": 1.061022776106575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004994621708236401,
|
|
"loss": 6.4285,
|
|
"mean_token_accuracy": 0.13216220885515212,
|
|
"num_tokens": 2244038.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.5215829372406,
|
|
"epoch": 1.0653201547056295,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499438950772416,
|
|
"loss": 6.3825,
|
|
"mean_token_accuracy": 0.13749199584126473,
|
|
"num_tokens": 2252828.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.589654731750488,
|
|
"epoch": 1.0696175333046842,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004994152406362311,
|
|
"loss": 6.3526,
|
|
"mean_token_accuracy": 0.13534352406859398,
|
|
"num_tokens": 2261709.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.544873857498169,
|
|
"epoch": 1.0739149119037388,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993910404668586,
|
|
"loss": 6.402,
|
|
"mean_token_accuracy": 0.13456418067216874,
|
|
"num_tokens": 2269801.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.497649669647217,
|
|
"epoch": 1.0782122905027933,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499366350317142,
|
|
"loss": 6.3616,
|
|
"mean_token_accuracy": 0.13916796073317528,
|
|
"num_tokens": 2280587.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.4322058200836185,
|
|
"epoch": 1.0825096691018479,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004993411702409948,
|
|
"loss": 6.3588,
|
|
"mean_token_accuracy": 0.1342848613858223,
|
|
"num_tokens": 2289783.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.5220903873443605,
|
|
"epoch": 1.0868070477009024,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004993155002934002,
|
|
"loss": 6.4414,
|
|
"mean_token_accuracy": 0.1303260490298271,
|
|
"num_tokens": 2298785.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.592111110687256,
|
|
"epoch": 1.091104426299957,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004992893405304111,
|
|
"loss": 6.5385,
|
|
"mean_token_accuracy": 0.1299525059759617,
|
|
"num_tokens": 2308400.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.564047622680664,
|
|
"epoch": 1.0954018048990115,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00049926269100915,
|
|
"loss": 6.3635,
|
|
"mean_token_accuracy": 0.14365571066737176,
|
|
"num_tokens": 2317440.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.465882968902588,
|
|
"epoch": 1.0996991834980663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004992355517878087,
|
|
"loss": 6.3992,
|
|
"mean_token_accuracy": 0.13191932812333107,
|
|
"num_tokens": 2326730.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.493737697601318,
|
|
"epoch": 1.1039965620971208,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992079229256484,
|
|
"loss": 6.3817,
|
|
"mean_token_accuracy": 0.14110777899622917,
|
|
"num_tokens": 2336126.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.5359251499176025,
|
|
"epoch": 1.1082939406961754,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004991798044829996,
|
|
"loss": 6.4005,
|
|
"mean_token_accuracy": 0.13598520383238794,
|
|
"num_tokens": 2345144.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.489447927474975,
|
|
"epoch": 1.11259131929523,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991511965212618,
|
|
"loss": 6.3612,
|
|
"mean_token_accuracy": 0.14140836670994758,
|
|
"num_tokens": 2353851.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.433896398544311,
|
|
"epoch": 1.1168886978942845,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991220991029032,
|
|
"loss": 6.3495,
|
|
"mean_token_accuracy": 0.14358441829681395,
|
|
"num_tokens": 2362316.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.580701875686645,
|
|
"epoch": 1.121186076493339,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499092512291461,
|
|
"loss": 6.4558,
|
|
"mean_token_accuracy": 0.13336558341979982,
|
|
"num_tokens": 2373350.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.581955623626709,
|
|
"epoch": 1.1254834550923936,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499062436151541,
|
|
"loss": 6.3937,
|
|
"mean_token_accuracy": 0.13979026824235916,
|
|
"num_tokens": 2381753.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.408409786224365,
|
|
"epoch": 1.129780833691448,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004990318707488173,
|
|
"loss": 6.4101,
|
|
"mean_token_accuracy": 0.13570450320839883,
|
|
"num_tokens": 2390093.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.653066492080688,
|
|
"epoch": 1.1340782122905029,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004990008161500327,
|
|
"loss": 6.4404,
|
|
"mean_token_accuracy": 0.13621403872966767,
|
|
"num_tokens": 2399103.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.472879123687744,
|
|
"epoch": 1.1383755908895574,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000498969272422998,
|
|
"loss": 6.4251,
|
|
"mean_token_accuracy": 0.1384277917444706,
|
|
"num_tokens": 2407530.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.486513137817383,
|
|
"epoch": 1.142672969488612,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989372396365921,
|
|
"loss": 6.3886,
|
|
"mean_token_accuracy": 0.14020788073539733,
|
|
"num_tokens": 2416264.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.486220169067383,
|
|
"epoch": 1.1469703480876665,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989047178607618,
|
|
"loss": 6.4839,
|
|
"mean_token_accuracy": 0.13277149870991706,
|
|
"num_tokens": 2425256.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.470200538635254,
|
|
"epoch": 1.151267726686721,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004988717071665215,
|
|
"loss": 6.4321,
|
|
"mean_token_accuracy": 0.13158162906765938,
|
|
"num_tokens": 2434343.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.559522008895874,
|
|
"epoch": 1.1555651052857756,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004988382076259537,
|
|
"loss": 6.3565,
|
|
"mean_token_accuracy": 0.14069283604621888,
|
|
"num_tokens": 2442864.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.418056011199951,
|
|
"epoch": 1.1598624838848304,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004988042193122077,
|
|
"loss": 6.3991,
|
|
"mean_token_accuracy": 0.13773582503199577,
|
|
"num_tokens": 2451701.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.470896196365357,
|
|
"epoch": 1.164159862483885,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987697422995005,
|
|
"loss": 6.3886,
|
|
"mean_token_accuracy": 0.13572183027863502,
|
|
"num_tokens": 2460943.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.459086990356445,
|
|
"epoch": 1.1684572410829395,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004987347766631161,
|
|
"loss": 6.4124,
|
|
"mean_token_accuracy": 0.1375699020922184,
|
|
"num_tokens": 2470294.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.532469606399536,
|
|
"epoch": 1.172754619681994,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0004986993224794055,
|
|
"loss": 6.4782,
|
|
"mean_token_accuracy": 0.12745001912117004,
|
|
"num_tokens": 2479530.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.472117757797241,
|
|
"epoch": 1.1770519982810486,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004986633798257865,
|
|
"loss": 6.2547,
|
|
"mean_token_accuracy": 0.14262767881155014,
|
|
"num_tokens": 2488511.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.497004985809326,
|
|
"epoch": 1.181349376880103,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004986269487807434,
|
|
"loss": 6.4471,
|
|
"mean_token_accuracy": 0.13357599899172784,
|
|
"num_tokens": 2497868.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.477199649810791,
|
|
"epoch": 1.1856467554791577,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000498590029423827,
|
|
"loss": 6.3789,
|
|
"mean_token_accuracy": 0.13786919862031938,
|
|
"num_tokens": 2506873.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.5761964321136475,
|
|
"epoch": 1.1899441340782122,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004985526218356546,
|
|
"loss": 6.5068,
|
|
"mean_token_accuracy": 0.13015856072306634,
|
|
"num_tokens": 2517317.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.4771264553070065,
|
|
"epoch": 1.1942415126772667,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004985147260979093,
|
|
"loss": 6.4256,
|
|
"mean_token_accuracy": 0.14017087668180467,
|
|
"num_tokens": 2527348.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.5237367153167725,
|
|
"epoch": 1.1985388912763215,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984763422933402,
|
|
"loss": 6.3999,
|
|
"mean_token_accuracy": 0.1334431551396847,
|
|
"num_tokens": 2536277.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.4698676586151125,
|
|
"epoch": 1.202836269875376,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004984374705057623,
|
|
"loss": 6.4717,
|
|
"mean_token_accuracy": 0.12729324474930764,
|
|
"num_tokens": 2545638.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.3455939292907715,
|
|
"epoch": 1.2071336484744306,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983981108200561,
|
|
"loss": 6.3065,
|
|
"mean_token_accuracy": 0.14563890397548676,
|
|
"num_tokens": 2555694.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.531454277038574,
|
|
"epoch": 1.2114310270734852,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983582633221672,
|
|
"loss": 6.3456,
|
|
"mean_token_accuracy": 0.13223223984241486,
|
|
"num_tokens": 2564344.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.432409477233887,
|
|
"epoch": 1.2157284056725397,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004983179280991068,
|
|
"loss": 6.3542,
|
|
"mean_token_accuracy": 0.13943792656064033,
|
|
"num_tokens": 2573726.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.452390909194946,
|
|
"epoch": 1.2200257842715942,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004982771052389508,
|
|
"loss": 6.4114,
|
|
"mean_token_accuracy": 0.13722645938396455,
|
|
"num_tokens": 2582845.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.4626904964447025,
|
|
"epoch": 1.224323162870649,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004982357948308401,
|
|
"loss": 6.2976,
|
|
"mean_token_accuracy": 0.1429854430258274,
|
|
"num_tokens": 2591258.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.509098243713379,
|
|
"epoch": 1.2286205414697036,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004981939969649799,
|
|
"loss": 6.3916,
|
|
"mean_token_accuracy": 0.1318533204495907,
|
|
"num_tokens": 2600099.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.4922661781311035,
|
|
"epoch": 1.232917920068758,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004981517117326404,
|
|
"loss": 6.3898,
|
|
"mean_token_accuracy": 0.14044687375426293,
|
|
"num_tokens": 2608845.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.376135206222534,
|
|
"epoch": 1.2372152986678127,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004981089392261553,
|
|
"loss": 6.3137,
|
|
"mean_token_accuracy": 0.13984026685357093,
|
|
"num_tokens": 2618723.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.430540895462036,
|
|
"epoch": 1.2415126772668672,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000498065679538923,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.1358903966844082,
|
|
"num_tokens": 2628542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.445869493484497,
|
|
"epoch": 1.2458100558659218,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980219327654049,
|
|
"loss": 6.3441,
|
|
"mean_token_accuracy": 0.14282078370451928,
|
|
"num_tokens": 2637395.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.394225835800171,
|
|
"epoch": 1.2501074344649763,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497977699001127,
|
|
"loss": 6.1983,
|
|
"mean_token_accuracy": 0.14897849634289742,
|
|
"num_tokens": 2645545.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.3761859893798825,
|
|
"epoch": 1.2544048130640308,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004979329783426778,
|
|
"loss": 6.3791,
|
|
"mean_token_accuracy": 0.14151204004883766,
|
|
"num_tokens": 2654000.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.437392139434815,
|
|
"epoch": 1.2587021916630854,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978877708877094,
|
|
"loss": 6.3721,
|
|
"mean_token_accuracy": 0.13939365819096566,
|
|
"num_tokens": 2664017.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.415973091125489,
|
|
"epoch": 1.2629995702621402,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004978420767349368,
|
|
"loss": 6.3758,
|
|
"mean_token_accuracy": 0.1392499901354313,
|
|
"num_tokens": 2673250.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.470103645324707,
|
|
"epoch": 1.2672969488611947,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004977958959841379,
|
|
"loss": 6.4096,
|
|
"mean_token_accuracy": 0.13894930183887483,
|
|
"num_tokens": 2682190.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.462930536270141,
|
|
"epoch": 1.2715943274602493,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000497749228736153,
|
|
"loss": 6.3257,
|
|
"mean_token_accuracy": 0.13693501725792884,
|
|
"num_tokens": 2691425.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.487558031082154,
|
|
"epoch": 1.2758917060593038,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004977020750928845,
|
|
"loss": 6.4314,
|
|
"mean_token_accuracy": 0.13328145816922188,
|
|
"num_tokens": 2701123.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.339822912216187,
|
|
"epoch": 1.2801890846583583,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004976544351572973,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.13217268735170365,
|
|
"num_tokens": 2710829.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.5346620082855225,
|
|
"epoch": 1.2844864632574131,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004976063090334179,
|
|
"loss": 6.382,
|
|
"mean_token_accuracy": 0.13936254531145095,
|
|
"num_tokens": 2720313.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.357281064987182,
|
|
"epoch": 1.2887838418564677,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004975576968263346,
|
|
"loss": 6.2811,
|
|
"mean_token_accuracy": 0.14250001460313796,
|
|
"num_tokens": 2729734.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2887838418564677,
|
|
"eval_entropy": 6.225875508677852,
|
|
"eval_loss": 6.502307415008545,
|
|
"eval_mean_token_accuracy": 0.1380094436106381,
|
|
"eval_num_tokens": 2729734.0,
|
|
"eval_runtime": 2.6325,
|
|
"eval_samples_per_second": 1348.163,
|
|
"eval_steps_per_second": 168.663,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.396690702438354,
|
|
"epoch": 1.2930812204555222,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497508598642197,
|
|
"loss": 6.3825,
|
|
"mean_token_accuracy": 0.13525475636124612,
|
|
"num_tokens": 2739001.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.442233562469482,
|
|
"epoch": 1.2973785990545768,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000497459014588216,
|
|
"loss": 6.2613,
|
|
"mean_token_accuracy": 0.1440258502960205,
|
|
"num_tokens": 2748460.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.321086597442627,
|
|
"epoch": 1.3016759776536313,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497408944772663,
|
|
"loss": 6.3893,
|
|
"mean_token_accuracy": 0.1376827046275139,
|
|
"num_tokens": 2757993.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.437920141220093,
|
|
"epoch": 1.3059733562526858,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004973583893048707,
|
|
"loss": 6.4098,
|
|
"mean_token_accuracy": 0.13957171514630318,
|
|
"num_tokens": 2767380.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.435287952423096,
|
|
"epoch": 1.3102707348517404,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004973073482952321,
|
|
"loss": 6.3164,
|
|
"mean_token_accuracy": 0.14111845567822456,
|
|
"num_tokens": 2776207.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.3955525875091555,
|
|
"epoch": 1.314568113450795,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972558218552004,
|
|
"loss": 6.3726,
|
|
"mean_token_accuracy": 0.14018303379416466,
|
|
"num_tokens": 2785857.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.43671350479126,
|
|
"epoch": 1.3188654920498495,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004972038100972885,
|
|
"loss": 6.4667,
|
|
"mean_token_accuracy": 0.13432941138744353,
|
|
"num_tokens": 2795500.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.449034547805786,
|
|
"epoch": 1.323162870648904,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004971513131350697,
|
|
"loss": 6.3399,
|
|
"mean_token_accuracy": 0.1369171641767025,
|
|
"num_tokens": 2804188.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.387170505523682,
|
|
"epoch": 1.3274602492479588,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004970983310831759,
|
|
"loss": 6.3547,
|
|
"mean_token_accuracy": 0.13035919144749641,
|
|
"num_tokens": 2814255.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.4166727542877195,
|
|
"epoch": 1.3317576278470133,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004970448640572989,
|
|
"loss": 6.3329,
|
|
"mean_token_accuracy": 0.14074430763721466,
|
|
"num_tokens": 2823059.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.386816072463989,
|
|
"epoch": 1.336055006446068,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004969909121741895,
|
|
"loss": 6.2582,
|
|
"mean_token_accuracy": 0.14372179433703422,
|
|
"num_tokens": 2831729.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.37970085144043,
|
|
"epoch": 1.3403523850451224,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004969364755516569,
|
|
"loss": 6.3209,
|
|
"mean_token_accuracy": 0.14519608169794082,
|
|
"num_tokens": 2839597.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.372022533416748,
|
|
"epoch": 1.344649763644177,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968815543085689,
|
|
"loss": 6.1988,
|
|
"mean_token_accuracy": 0.14956804513931274,
|
|
"num_tokens": 2847964.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.318734455108642,
|
|
"epoch": 1.3489471422432318,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004968261485648516,
|
|
"loss": 6.3975,
|
|
"mean_token_accuracy": 0.13511656373739242,
|
|
"num_tokens": 2857712.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.415015649795532,
|
|
"epoch": 1.3532445208422863,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000496770258441489,
|
|
"loss": 6.4039,
|
|
"mean_token_accuracy": 0.13150390312075616,
|
|
"num_tokens": 2866724.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.480535697937012,
|
|
"epoch": 1.3575418994413408,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004967138840605228,
|
|
"loss": 6.2821,
|
|
"mean_token_accuracy": 0.1422867201268673,
|
|
"num_tokens": 2875318.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.371291399002075,
|
|
"epoch": 1.3618392780403954,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000496657025545052,
|
|
"loss": 6.326,
|
|
"mean_token_accuracy": 0.1330809101462364,
|
|
"num_tokens": 2883794.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.419414758682251,
|
|
"epoch": 1.36613665663945,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000496599683019233,
|
|
"loss": 6.3467,
|
|
"mean_token_accuracy": 0.1338125415146351,
|
|
"num_tokens": 2891996.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.391923189163208,
|
|
"epoch": 1.3704340352385045,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000496541856608279,
|
|
"loss": 6.4484,
|
|
"mean_token_accuracy": 0.1343950994312763,
|
|
"num_tokens": 2901718.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.449080562591552,
|
|
"epoch": 1.374731413837559,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004964835464384595,
|
|
"loss": 6.2728,
|
|
"mean_token_accuracy": 0.1431309886276722,
|
|
"num_tokens": 2910924.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.311095952987671,
|
|
"epoch": 1.3790287924366136,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000496424752637101,
|
|
"loss": 6.3674,
|
|
"mean_token_accuracy": 0.1359468601644039,
|
|
"num_tokens": 2920075.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.367361640930175,
|
|
"epoch": 1.3833261710356681,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004963654753325853,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.1429993435740471,
|
|
"num_tokens": 2928946.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.329522371292114,
|
|
"epoch": 1.387623549634723,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004963057146543505,
|
|
"loss": 6.3067,
|
|
"mean_token_accuracy": 0.15131649672985076,
|
|
"num_tokens": 2936817.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.382766580581665,
|
|
"epoch": 1.3919209282337774,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00049624547073289,
|
|
"loss": 6.2892,
|
|
"mean_token_accuracy": 0.14377383068203925,
|
|
"num_tokens": 2946096.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.362337255477906,
|
|
"epoch": 1.396218306832832,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004961847436997526,
|
|
"loss": 6.2334,
|
|
"mean_token_accuracy": 0.14999890178442002,
|
|
"num_tokens": 2955632.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.366307544708252,
|
|
"epoch": 1.4005156854318865,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004961235336875416,
|
|
"loss": 6.2579,
|
|
"mean_token_accuracy": 0.14550368934869767,
|
|
"num_tokens": 2963779.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.358678722381592,
|
|
"epoch": 1.404813064030941,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004960618408299154,
|
|
"loss": 6.2494,
|
|
"mean_token_accuracy": 0.1504204198718071,
|
|
"num_tokens": 2972778.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.318516445159912,
|
|
"epoch": 1.4091104426299956,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004959996652615865,
|
|
"loss": 6.3797,
|
|
"mean_token_accuracy": 0.13637638315558434,
|
|
"num_tokens": 2982318.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.44262466430664,
|
|
"epoch": 1.4134078212290504,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004959370071183216,
|
|
"loss": 6.3439,
|
|
"mean_token_accuracy": 0.14733938425779342,
|
|
"num_tokens": 2991822.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.3717328071594235,
|
|
"epoch": 1.417705199828105,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004958738665369407,
|
|
"loss": 6.3792,
|
|
"mean_token_accuracy": 0.13971310779452323,
|
|
"num_tokens": 3001203.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.349838733673096,
|
|
"epoch": 1.4220025784271595,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004958102436553179,
|
|
"loss": 6.2995,
|
|
"mean_token_accuracy": 0.13951244726777076,
|
|
"num_tokens": 3010445.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.399106740951538,
|
|
"epoch": 1.426299957026214,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049574613861238,
|
|
"loss": 6.3232,
|
|
"mean_token_accuracy": 0.13808537349104882,
|
|
"num_tokens": 3019919.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.476325130462646,
|
|
"epoch": 1.4305973356252686,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004956815515481069,
|
|
"loss": 6.4575,
|
|
"mean_token_accuracy": 0.13107064962387086,
|
|
"num_tokens": 3029469.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.404677104949951,
|
|
"epoch": 1.4348947142243231,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004956164826035309,
|
|
"loss": 6.3246,
|
|
"mean_token_accuracy": 0.14275707602500914,
|
|
"num_tokens": 3038627.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.359204912185669,
|
|
"epoch": 1.4391920928233777,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004955509319207363,
|
|
"loss": 6.3652,
|
|
"mean_token_accuracy": 0.138480856269598,
|
|
"num_tokens": 3048124.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.322403764724731,
|
|
"epoch": 1.4434894714224322,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004954848996428601,
|
|
"loss": 6.2081,
|
|
"mean_token_accuracy": 0.14859130680561067,
|
|
"num_tokens": 3056394.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.297450351715088,
|
|
"epoch": 1.4477868500214868,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00049541838591409,
|
|
"loss": 6.3257,
|
|
"mean_token_accuracy": 0.1398971401154995,
|
|
"num_tokens": 3066521.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.366423177719116,
|
|
"epoch": 1.4520842286205415,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004953513908796657,
|
|
"loss": 6.3707,
|
|
"mean_token_accuracy": 0.14321988448500633,
|
|
"num_tokens": 3076571.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.381562280654907,
|
|
"epoch": 1.456381607219596,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004952839146858773,
|
|
"loss": 6.428,
|
|
"mean_token_accuracy": 0.1346765175461769,
|
|
"num_tokens": 3086086.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.335469150543213,
|
|
"epoch": 1.4606789858186506,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004952159574800658,
|
|
"loss": 6.2967,
|
|
"mean_token_accuracy": 0.1462803915143013,
|
|
"num_tokens": 3095046.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.314669752120972,
|
|
"epoch": 1.4649763644177052,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004951475194106229,
|
|
"loss": 6.2718,
|
|
"mean_token_accuracy": 0.14232781156897545,
|
|
"num_tokens": 3104193.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.4825750350952145,
|
|
"epoch": 1.4692737430167597,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004950786006269898,
|
|
"loss": 6.3566,
|
|
"mean_token_accuracy": 0.1422339826822281,
|
|
"num_tokens": 3112965.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.331253480911255,
|
|
"epoch": 1.4735711216158143,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004950092012796576,
|
|
"loss": 6.4137,
|
|
"mean_token_accuracy": 0.1361984558403492,
|
|
"num_tokens": 3122883.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.377304458618164,
|
|
"epoch": 1.477868500214869,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004949393215201666,
|
|
"loss": 6.2783,
|
|
"mean_token_accuracy": 0.15220774635672568,
|
|
"num_tokens": 3132182.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.340102863311768,
|
|
"epoch": 1.4821658788139236,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004948689615011065,
|
|
"loss": 6.3395,
|
|
"mean_token_accuracy": 0.14068598598241805,
|
|
"num_tokens": 3141957.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.31355299949646,
|
|
"epoch": 1.4864632574129781,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004947981213761154,
|
|
"loss": 6.2533,
|
|
"mean_token_accuracy": 0.14872836396098138,
|
|
"num_tokens": 3151374.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.266488409042358,
|
|
"epoch": 1.4907606360120327,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004947268012998797,
|
|
"loss": 6.2864,
|
|
"mean_token_accuracy": 0.1442766495049,
|
|
"num_tokens": 3160815.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.43033537864685,
|
|
"epoch": 1.4950580146110872,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000494655001428134,
|
|
"loss": 6.2953,
|
|
"mean_token_accuracy": 0.14629183933138848,
|
|
"num_tokens": 3169684.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.341884183883667,
|
|
"epoch": 1.4993553932101418,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945827219176604,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.14152552783489228,
|
|
"num_tokens": 3178754.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.314790964126587,
|
|
"epoch": 1.5036527718091963,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004945099629262888,
|
|
"loss": 6.352,
|
|
"mean_token_accuracy": 0.14713623374700546,
|
|
"num_tokens": 3187315.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.367385196685791,
|
|
"epoch": 1.5079501504082509,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004944367246128954,
|
|
"loss": 6.1877,
|
|
"mean_token_accuracy": 0.15467803478240966,
|
|
"num_tokens": 3195975.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.279762935638428,
|
|
"epoch": 1.5122475290073054,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004943630071374036,
|
|
"loss": 6.2884,
|
|
"mean_token_accuracy": 0.14560731798410415,
|
|
"num_tokens": 3204071.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.352357530593872,
|
|
"epoch": 1.51654490760636,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004942888106607828,
|
|
"loss": 6.2805,
|
|
"mean_token_accuracy": 0.14571768492460252,
|
|
"num_tokens": 3213001.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.385111236572266,
|
|
"epoch": 1.5208422862054147,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004942141353450486,
|
|
"loss": 6.3184,
|
|
"mean_token_accuracy": 0.13302915021777154,
|
|
"num_tokens": 3222576.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.397508764266968,
|
|
"epoch": 1.5251396648044693,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004941389813532619,
|
|
"loss": 6.3244,
|
|
"mean_token_accuracy": 0.14305002242326736,
|
|
"num_tokens": 3230474.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.347569131851197,
|
|
"epoch": 1.5294370434035238,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000494063348849529,
|
|
"loss": 6.3197,
|
|
"mean_token_accuracy": 0.14401185810565947,
|
|
"num_tokens": 3238723.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.265641689300537,
|
|
"epoch": 1.5337344220025786,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004939872379990011,
|
|
"loss": 6.2394,
|
|
"mean_token_accuracy": 0.1483248919248581,
|
|
"num_tokens": 3247844.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.41879940032959,
|
|
"epoch": 1.5380318006016331,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004939106489678739,
|
|
"loss": 6.3428,
|
|
"mean_token_accuracy": 0.14301740303635596,
|
|
"num_tokens": 3257228.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.408779001235962,
|
|
"epoch": 1.5423291792006877,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000493833581923387,
|
|
"loss": 6.3231,
|
|
"mean_token_accuracy": 0.13824094384908675,
|
|
"num_tokens": 3266655.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.3060040950775145,
|
|
"epoch": 1.5466265577997422,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004937560370338244,
|
|
"loss": 6.2434,
|
|
"mean_token_accuracy": 0.14510375261306763,
|
|
"num_tokens": 3275213.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.408991718292237,
|
|
"epoch": 1.5509239363987968,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000493678014468513,
|
|
"loss": 6.3779,
|
|
"mean_token_accuracy": 0.13703251108527184,
|
|
"num_tokens": 3284447.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.350679492950439,
|
|
"epoch": 1.5552213149978513,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004935995143978227,
|
|
"loss": 6.3041,
|
|
"mean_token_accuracy": 0.13779493197798728,
|
|
"num_tokens": 3293568.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.285771989822388,
|
|
"epoch": 1.5595186935969059,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004935205369931664,
|
|
"loss": 6.2543,
|
|
"mean_token_accuracy": 0.14422329068183898,
|
|
"num_tokens": 3302055.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.356205034255981,
|
|
"epoch": 1.5638160721959604,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004934410824269992,
|
|
"loss": 6.293,
|
|
"mean_token_accuracy": 0.14307598844170571,
|
|
"num_tokens": 3310672.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.324226570129395,
|
|
"epoch": 1.568113450795015,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004933611508728182,
|
|
"loss": 6.2441,
|
|
"mean_token_accuracy": 0.14398171454668046,
|
|
"num_tokens": 3319666.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.338666772842407,
|
|
"epoch": 1.5724108293940695,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000493280742505162,
|
|
"loss": 6.2517,
|
|
"mean_token_accuracy": 0.14428381100296975,
|
|
"num_tokens": 3329121.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.26886887550354,
|
|
"epoch": 1.576708207993124,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004931998574996102,
|
|
"loss": 6.253,
|
|
"mean_token_accuracy": 0.14581965357065202,
|
|
"num_tokens": 3337868.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.31952691078186,
|
|
"epoch": 1.5810055865921788,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004931184960327832,
|
|
"loss": 6.2743,
|
|
"mean_token_accuracy": 0.14775650948286057,
|
|
"num_tokens": 3346447.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.299622392654419,
|
|
"epoch": 1.5853029651912334,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004930366582823421,
|
|
"loss": 6.2391,
|
|
"mean_token_accuracy": 0.14039799869060515,
|
|
"num_tokens": 3355187.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.348891973495483,
|
|
"epoch": 1.589600343790288,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004929543444269879,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.12413048818707466,
|
|
"num_tokens": 3365978.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.397365522384644,
|
|
"epoch": 1.5938977223893425,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000492871554646461,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.14158129766583444,
|
|
"num_tokens": 3374607.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.336241769790649,
|
|
"epoch": 1.5981951009883972,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004927882891215413,
|
|
"loss": 6.3163,
|
|
"mean_token_accuracy": 0.13903330489993096,
|
|
"num_tokens": 3384541.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.361058235168457,
|
|
"epoch": 1.6024924795874518,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004927045480340475,
|
|
"loss": 6.3238,
|
|
"mean_token_accuracy": 0.13128943145275115,
|
|
"num_tokens": 3394009.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.38292350769043,
|
|
"epoch": 1.6067898581865063,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004926203315668363,
|
|
"loss": 6.3203,
|
|
"mean_token_accuracy": 0.14705114364624022,
|
|
"num_tokens": 3402550.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.333295106887817,
|
|
"epoch": 1.6110872367855609,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004925356399038032,
|
|
"loss": 6.2854,
|
|
"mean_token_accuracy": 0.14575798213481903,
|
|
"num_tokens": 3413230.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.227672910690307,
|
|
"epoch": 1.6153846153846154,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004924504732298808,
|
|
"loss": 6.2675,
|
|
"mean_token_accuracy": 0.1498914524912834,
|
|
"num_tokens": 3422325.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.411320638656616,
|
|
"epoch": 1.61968199398367,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004923648317310391,
|
|
"loss": 6.3458,
|
|
"mean_token_accuracy": 0.13707044050097467,
|
|
"num_tokens": 3431758.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.262428092956543,
|
|
"epoch": 1.6239793725827245,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004922787155942849,
|
|
"loss": 6.2385,
|
|
"mean_token_accuracy": 0.14557855874300002,
|
|
"num_tokens": 3441281.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.317093467712402,
|
|
"epoch": 1.628276751181779,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004921921250076611,
|
|
"loss": 6.2754,
|
|
"mean_token_accuracy": 0.14766671732068062,
|
|
"num_tokens": 3449822.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.347660970687866,
|
|
"epoch": 1.6325741297808336,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004921050601602475,
|
|
"loss": 6.2806,
|
|
"mean_token_accuracy": 0.14015369564294816,
|
|
"num_tokens": 3459437.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.3261857509613035,
|
|
"epoch": 1.6368715083798882,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004920175212421587,
|
|
"loss": 6.2958,
|
|
"mean_token_accuracy": 0.14800578653812407,
|
|
"num_tokens": 3469546.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.411861324310303,
|
|
"epoch": 1.6411688869789427,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004919295084445445,
|
|
"loss": 6.2441,
|
|
"mean_token_accuracy": 0.13949010968208314,
|
|
"num_tokens": 3478521.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.252915906906128,
|
|
"epoch": 1.6454662655779975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004918410219595899,
|
|
"loss": 6.149,
|
|
"mean_token_accuracy": 0.1499532699584961,
|
|
"num_tokens": 3487958.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.255596494674682,
|
|
"epoch": 1.649763644177052,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000491752061980514,
|
|
"loss": 6.2476,
|
|
"mean_token_accuracy": 0.14446067586541175,
|
|
"num_tokens": 3495941.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.297366428375244,
|
|
"epoch": 1.6540610227761066,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004916626287015697,
|
|
"loss": 6.2807,
|
|
"mean_token_accuracy": 0.1459622099995613,
|
|
"num_tokens": 3505172.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.296765279769898,
|
|
"epoch": 1.658358401375161,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004915727223180436,
|
|
"loss": 6.2546,
|
|
"mean_token_accuracy": 0.14922861829400064,
|
|
"num_tokens": 3514084.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.42270679473877,
|
|
"epoch": 1.6626557799742159,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004914823430262554,
|
|
"loss": 6.3903,
|
|
"mean_token_accuracy": 0.1387872278690338,
|
|
"num_tokens": 3523318.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.304086303710937,
|
|
"epoch": 1.6669531585732704,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004913914910235573,
|
|
"loss": 6.2144,
|
|
"mean_token_accuracy": 0.14882446974515914,
|
|
"num_tokens": 3531681.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.264715385437012,
|
|
"epoch": 1.671250537172325,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004913001665083337,
|
|
"loss": 6.2909,
|
|
"mean_token_accuracy": 0.14123921766877173,
|
|
"num_tokens": 3540326.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.336367464065551,
|
|
"epoch": 1.6755479157713795,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004912083696800008,
|
|
"loss": 6.2498,
|
|
"mean_token_accuracy": 0.1417755626142025,
|
|
"num_tokens": 3548319.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.2093394756317135,
|
|
"epoch": 1.679845294370434,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004911161007390063,
|
|
"loss": 6.0905,
|
|
"mean_token_accuracy": 0.15048674941062928,
|
|
"num_tokens": 3557743.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.2037858963012695,
|
|
"epoch": 1.6841426729694886,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004910233598868287,
|
|
"loss": 6.223,
|
|
"mean_token_accuracy": 0.15063542425632476,
|
|
"num_tokens": 3566810.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.319229221343994,
|
|
"epoch": 1.6884400515685432,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004909301473259769,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.14927951395511627,
|
|
"num_tokens": 3576273.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.224372720718383,
|
|
"epoch": 1.6927374301675977,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004908364632599899,
|
|
"loss": 6.276,
|
|
"mean_token_accuracy": 0.14561723917722702,
|
|
"num_tokens": 3585362.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.2979576110839846,
|
|
"epoch": 1.6970348087666522,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004907423078934362,
|
|
"loss": 6.2566,
|
|
"mean_token_accuracy": 0.1375780612230301,
|
|
"num_tokens": 3594445.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.296867942810058,
|
|
"epoch": 1.7013321873657068,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004906476814319134,
|
|
"loss": 6.2747,
|
|
"mean_token_accuracy": 0.13942770212888717,
|
|
"num_tokens": 3604046.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.365720796585083,
|
|
"epoch": 1.7056295659647613,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004905525840820481,
|
|
"loss": 6.289,
|
|
"mean_token_accuracy": 0.14392739534378052,
|
|
"num_tokens": 3613264.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.303918886184692,
|
|
"epoch": 1.709926944563816,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004904570160514948,
|
|
"loss": 6.1724,
|
|
"mean_token_accuracy": 0.14787200540304185,
|
|
"num_tokens": 3620938.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.262858819961548,
|
|
"epoch": 1.7142243231628707,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004903609775489358,
|
|
"loss": 6.276,
|
|
"mean_token_accuracy": 0.14155241400003432,
|
|
"num_tokens": 3629782.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.363343238830566,
|
|
"epoch": 1.7185217017619252,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004902644687840809,
|
|
"loss": 6.3347,
|
|
"mean_token_accuracy": 0.13947545289993285,
|
|
"num_tokens": 3638389.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.7185217017619252,
|
|
"eval_entropy": 6.028549555185679,
|
|
"eval_loss": 6.352311611175537,
|
|
"eval_mean_token_accuracy": 0.14357087935681814,
|
|
"eval_num_tokens": 3638389.0,
|
|
"eval_runtime": 2.6487,
|
|
"eval_samples_per_second": 1339.891,
|
|
"eval_steps_per_second": 167.628,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.196940898895264,
|
|
"epoch": 1.7228190803609797,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004901674899676667,
|
|
"loss": 6.1708,
|
|
"mean_token_accuracy": 0.15208130180835724,
|
|
"num_tokens": 3647196.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.18595404624939,
|
|
"epoch": 1.7271164589600345,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004900700413114561,
|
|
"loss": 6.1128,
|
|
"mean_token_accuracy": 0.15374353379011155,
|
|
"num_tokens": 3655931.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.310988998413086,
|
|
"epoch": 1.731413837559089,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000489972123028238,
|
|
"loss": 6.2676,
|
|
"mean_token_accuracy": 0.1467567965388298,
|
|
"num_tokens": 3665340.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.268195533752442,
|
|
"epoch": 1.7357112161581436,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004898737353318268,
|
|
"loss": 6.2224,
|
|
"mean_token_accuracy": 0.15013988465070724,
|
|
"num_tokens": 3674674.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.301918649673462,
|
|
"epoch": 1.7400085947571982,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489774878437062,
|
|
"loss": 6.2572,
|
|
"mean_token_accuracy": 0.15027557015419007,
|
|
"num_tokens": 3683910.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.267896747589111,
|
|
"epoch": 1.7443059733562527,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004896755525598074,
|
|
"loss": 6.1509,
|
|
"mean_token_accuracy": 0.14787854254245758,
|
|
"num_tokens": 3692772.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.2744200229644775,
|
|
"epoch": 1.7486033519553073,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895757579169511,
|
|
"loss": 6.1827,
|
|
"mean_token_accuracy": 0.15500140488147734,
|
|
"num_tokens": 3701875.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.258078145980835,
|
|
"epoch": 1.7529007305543618,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004894754947264047,
|
|
"loss": 6.1938,
|
|
"mean_token_accuracy": 0.14737177044153213,
|
|
"num_tokens": 3711059.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.273642301559448,
|
|
"epoch": 1.7571981091534163,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000489374763207103,
|
|
"loss": 6.1763,
|
|
"mean_token_accuracy": 0.15221278220415116,
|
|
"num_tokens": 3719667.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.176022243499756,
|
|
"epoch": 1.761495487752471,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004892735635790033,
|
|
"loss": 6.1691,
|
|
"mean_token_accuracy": 0.14955660551786423,
|
|
"num_tokens": 3728905.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.270573949813842,
|
|
"epoch": 1.7657928663515254,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000489171896063085,
|
|
"loss": 6.2023,
|
|
"mean_token_accuracy": 0.14582386016845703,
|
|
"num_tokens": 3737973.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.205034065246582,
|
|
"epoch": 1.77009024495058,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004890697608813495,
|
|
"loss": 6.1614,
|
|
"mean_token_accuracy": 0.149971167743206,
|
|
"num_tokens": 3747160.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.202639961242676,
|
|
"epoch": 1.7743876235496348,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004889671582568193,
|
|
"loss": 6.1251,
|
|
"mean_token_accuracy": 0.14935097396373748,
|
|
"num_tokens": 3754908.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.299329328536987,
|
|
"epoch": 1.7786850021486893,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004888640884135374,
|
|
"loss": 6.2506,
|
|
"mean_token_accuracy": 0.13549285531044006,
|
|
"num_tokens": 3763588.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.21603832244873,
|
|
"epoch": 1.7829823807477438,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004887605515765671,
|
|
"loss": 6.128,
|
|
"mean_token_accuracy": 0.15028423219919204,
|
|
"num_tokens": 3771932.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.286951112747192,
|
|
"epoch": 1.7872797593467986,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004886565479719914,
|
|
"loss": 6.2865,
|
|
"mean_token_accuracy": 0.1433756723999977,
|
|
"num_tokens": 3780761.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.220171737670898,
|
|
"epoch": 1.7915771379458532,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004885520778269128,
|
|
"loss": 6.0901,
|
|
"mean_token_accuracy": 0.15441024005413057,
|
|
"num_tokens": 3789511.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.20865387916565,
|
|
"epoch": 1.7958745165449077,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004884471413694523,
|
|
"loss": 6.2819,
|
|
"mean_token_accuracy": 0.1363023564219475,
|
|
"num_tokens": 3799068.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.3188676834106445,
|
|
"epoch": 1.8001718951439623,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004883417388287491,
|
|
"loss": 6.1667,
|
|
"mean_token_accuracy": 0.1450318641960621,
|
|
"num_tokens": 3807850.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.26737003326416,
|
|
"epoch": 1.8044692737430168,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004882358704349603,
|
|
"loss": 6.1858,
|
|
"mean_token_accuracy": 0.15362504720687867,
|
|
"num_tokens": 3816232.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.204585838317871,
|
|
"epoch": 1.8087666523420713,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004881295364192601,
|
|
"loss": 6.2312,
|
|
"mean_token_accuracy": 0.13994672447443007,
|
|
"num_tokens": 3825334.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.213073301315307,
|
|
"epoch": 1.813064030941126,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004880227370138394,
|
|
"loss": 6.2034,
|
|
"mean_token_accuracy": 0.1517607569694519,
|
|
"num_tokens": 3834756.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.338291358947754,
|
|
"epoch": 1.8173614095401804,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004879154724519057,
|
|
"loss": 6.245,
|
|
"mean_token_accuracy": 0.14460117518901824,
|
|
"num_tokens": 3843852.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.261815309524536,
|
|
"epoch": 1.821658788139235,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004878077429676816,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.1576495312154293,
|
|
"num_tokens": 3852213.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.227415990829468,
|
|
"epoch": 1.8259561667382895,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004876995487964054,
|
|
"loss": 6.2609,
|
|
"mean_token_accuracy": 0.13798727393150328,
|
|
"num_tokens": 3861522.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.2565147399902346,
|
|
"epoch": 1.830253545337344,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048759089017432996,
|
|
"loss": 6.2082,
|
|
"mean_token_accuracy": 0.15773860663175582,
|
|
"num_tokens": 3870328.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.246215200424194,
|
|
"epoch": 1.8345509239363988,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004874817673387222,
|
|
"loss": 6.2124,
|
|
"mean_token_accuracy": 0.14310137778520585,
|
|
"num_tokens": 3880399.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.259421873092651,
|
|
"epoch": 1.8388483025354534,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048737218052786275,
|
|
"loss": 6.1678,
|
|
"mean_token_accuracy": 0.14924097061157227,
|
|
"num_tokens": 3889859.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.1444484233856205,
|
|
"epoch": 1.843145681134508,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048726212998104554,
|
|
"loss": 6.1267,
|
|
"mean_token_accuracy": 0.1519687682390213,
|
|
"num_tokens": 3899609.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.270504760742187,
|
|
"epoch": 1.8474430597335625,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004871516159385768,
|
|
"loss": 6.2271,
|
|
"mean_token_accuracy": 0.1401534102857113,
|
|
"num_tokens": 3909452.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.335358810424805,
|
|
"epoch": 1.8517404383326173,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004870406386417752,
|
|
"loss": 6.269,
|
|
"mean_token_accuracy": 0.14715837240219115,
|
|
"num_tokens": 3918579.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.262742137908935,
|
|
"epoch": 1.8560378169316718,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004869291983329707,
|
|
"loss": 6.1624,
|
|
"mean_token_accuracy": 0.15219318568706514,
|
|
"num_tokens": 3926951.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.181666231155395,
|
|
"epoch": 1.8603351955307263,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004868172952555044,
|
|
"loss": 6.2002,
|
|
"mean_token_accuracy": 0.1487558200955391,
|
|
"num_tokens": 3935771.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.242819786071777,
|
|
"epoch": 1.864632574129781,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004867049296537278,
|
|
"loss": 6.0966,
|
|
"mean_token_accuracy": 0.15747302025556564,
|
|
"num_tokens": 3944577.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.281427431106567,
|
|
"epoch": 1.8689299527288354,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004865921017730027,
|
|
"loss": 6.267,
|
|
"mean_token_accuracy": 0.14637882634997368,
|
|
"num_tokens": 3954499.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.216175651550293,
|
|
"epoch": 1.87322733132789,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048647881185969995,
|
|
"loss": 6.1862,
|
|
"mean_token_accuracy": 0.14351969063282013,
|
|
"num_tokens": 3963953.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.210945272445679,
|
|
"epoch": 1.8775247099269445,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004863650601611994,
|
|
"loss": 6.1971,
|
|
"mean_token_accuracy": 0.14761989265680314,
|
|
"num_tokens": 3973925.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.239703273773193,
|
|
"epoch": 1.881822088525999,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048625084692588937,
|
|
"loss": 6.1881,
|
|
"mean_token_accuracy": 0.14801398664712906,
|
|
"num_tokens": 3983280.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.194201278686523,
|
|
"epoch": 1.8861194671250536,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048613617240316593,
|
|
"loss": 6.1691,
|
|
"mean_token_accuracy": 0.14983253479003905,
|
|
"num_tokens": 3992767.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.32085862159729,
|
|
"epoch": 1.8904168457241082,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004860210368434323,
|
|
"loss": 6.2578,
|
|
"mean_token_accuracy": 0.13981493264436723,
|
|
"num_tokens": 4001600.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.232434463500977,
|
|
"epoch": 1.8947142243231627,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00048590544049809857,
|
|
"loss": 6.2388,
|
|
"mean_token_accuracy": 0.14904941469430924,
|
|
"num_tokens": 4010852.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.2208820343017575,
|
|
"epoch": 1.8990116029222175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000485789383619581,
|
|
"loss": 6.1344,
|
|
"mean_token_accuracy": 0.14979886412620544,
|
|
"num_tokens": 4020280.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.244888353347778,
|
|
"epoch": 1.903308981521272,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004856728664613015,
|
|
"loss": 6.1359,
|
|
"mean_token_accuracy": 0.15172455683350564,
|
|
"num_tokens": 4029283.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.2939231395721436,
|
|
"epoch": 1.9076063601203266,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048555588927768674,
|
|
"loss": 6.2399,
|
|
"mean_token_accuracy": 0.15116785317659379,
|
|
"num_tokens": 4038704.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.17348461151123,
|
|
"epoch": 1.9119037387193811,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004854384523241683,
|
|
"loss": 6.0965,
|
|
"mean_token_accuracy": 0.14770452305674553,
|
|
"num_tokens": 4047492.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.251682472229004,
|
|
"epoch": 1.916201117318436,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048532055585718143,
|
|
"loss": 6.2895,
|
|
"mean_token_accuracy": 0.14247784093022348,
|
|
"num_tokens": 4056792.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.257444572448731,
|
|
"epoch": 1.9204984959174904,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048520220013416505,
|
|
"loss": 6.1858,
|
|
"mean_token_accuracy": 0.15522112101316451,
|
|
"num_tokens": 4065284.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.119591903686524,
|
|
"epoch": 1.924795874516545,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004850833854135607,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.149675352871418,
|
|
"num_tokens": 4074727.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.340999507904053,
|
|
"epoch": 1.9290932531155995,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004849641119548122,
|
|
"loss": 6.2279,
|
|
"mean_token_accuracy": 0.1487278312444687,
|
|
"num_tokens": 4083360.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.237739086151123,
|
|
"epoch": 1.933390631714654,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000484844380018365,
|
|
"loss": 6.3064,
|
|
"mean_token_accuracy": 0.13851014897227287,
|
|
"num_tokens": 4092431.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.2545881271362305,
|
|
"epoch": 1.9376880103137086,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000484724189865666,
|
|
"loss": 6.1942,
|
|
"mean_token_accuracy": 0.14804239720106124,
|
|
"num_tokens": 4102198.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.2334897994995115,
|
|
"epoch": 1.9419853889127632,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004846035417591624,
|
|
"loss": 6.1817,
|
|
"mean_token_accuracy": 0.1523241475224495,
|
|
"num_tokens": 4111568.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.23221755027771,
|
|
"epoch": 1.9462827675118177,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004844824359623014,
|
|
"loss": 6.1998,
|
|
"mean_token_accuracy": 0.14418465942144393,
|
|
"num_tokens": 4121151.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.278671407699585,
|
|
"epoch": 1.9505801461108723,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048436087273952966,
|
|
"loss": 6.1418,
|
|
"mean_token_accuracy": 0.14655561298131942,
|
|
"num_tokens": 4130197.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.086948871612549,
|
|
"epoch": 1.9548775247099268,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048423885235629265,
|
|
"loss": 6.1494,
|
|
"mean_token_accuracy": 0.1391369014978409,
|
|
"num_tokens": 4138310.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.2306829452514645,
|
|
"epoch": 1.9591749033089814,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004841163750790342,
|
|
"loss": 6.16,
|
|
"mean_token_accuracy": 0.1557426020503044,
|
|
"num_tokens": 4148270.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.210293531417847,
|
|
"epoch": 1.9634722819080361,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048399344117519555,
|
|
"loss": 6.1914,
|
|
"mean_token_accuracy": 0.1499612420797348,
|
|
"num_tokens": 4157276.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.134465265274048,
|
|
"epoch": 1.9677696605070907,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048387005091321544,
|
|
"loss": 6.2405,
|
|
"mean_token_accuracy": 0.15338025540113448,
|
|
"num_tokens": 4166144.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.35327787399292,
|
|
"epoch": 1.9720670391061452,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048374620456252877,
|
|
"loss": 6.2072,
|
|
"mean_token_accuracy": 0.14432276636362076,
|
|
"num_tokens": 4175146.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.164651250839233,
|
|
"epoch": 1.9763644177052,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048362190239356644,
|
|
"loss": 6.261,
|
|
"mean_token_accuracy": 0.14050934612751007,
|
|
"num_tokens": 4184089.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.261477375030518,
|
|
"epoch": 1.9806617963042545,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00048349714467775474,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.14885230213403702,
|
|
"num_tokens": 4192453.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.221670866012573,
|
|
"epoch": 1.984959174903309,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048337193168751464,
|
|
"loss": 6.1047,
|
|
"mean_token_accuracy": 0.15655480176210404,
|
|
"num_tokens": 4201323.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.134425449371338,
|
|
"epoch": 1.9892565535023636,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004832462636962613,
|
|
"loss": 6.1612,
|
|
"mean_token_accuracy": 0.15031953081488608,
|
|
"num_tokens": 4209805.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.2537988185882565,
|
|
"epoch": 1.9935539321014182,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004831201409784034,
|
|
"loss": 6.1351,
|
|
"mean_token_accuracy": 0.14936504513025284,
|
|
"num_tokens": 4218562.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.205300283432007,
|
|
"epoch": 1.9978513107004727,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004829935638093424,
|
|
"loss": 6.1914,
|
|
"mean_token_accuracy": 0.1510600835084915,
|
|
"num_tokens": 4227605.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.255237208472358,
|
|
"epoch": 2.0017189514396216,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004828665324654724,
|
|
"loss": 6.0293,
|
|
"mean_token_accuracy": 0.15552079843150246,
|
|
"num_tokens": 4235441.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.2011861324310305,
|
|
"epoch": 2.006016330038676,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004827390472241791,
|
|
"loss": 5.8218,
|
|
"mean_token_accuracy": 0.1603270322084427,
|
|
"num_tokens": 4243982.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.090994930267334,
|
|
"epoch": 2.010313708637731,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004826111083638392,
|
|
"loss": 5.9169,
|
|
"mean_token_accuracy": 0.15214518457651138,
|
|
"num_tokens": 4253032.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.203659296035767,
|
|
"epoch": 2.0146110872367857,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048248271616382,
|
|
"loss": 5.8146,
|
|
"mean_token_accuracy": 0.16601352095603944,
|
|
"num_tokens": 4262000.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.0723145484924315,
|
|
"epoch": 2.0189084658358403,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00048235387090447894,
|
|
"loss": 5.8301,
|
|
"mean_token_accuracy": 0.15925069227814675,
|
|
"num_tokens": 4271089.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.14722900390625,
|
|
"epoch": 2.023205844434895,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048222457286716235,
|
|
"loss": 5.8384,
|
|
"mean_token_accuracy": 0.16075642555952072,
|
|
"num_tokens": 4280064.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.140841007232666,
|
|
"epoch": 2.0275032230339494,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048209482233420564,
|
|
"loss": 5.8164,
|
|
"mean_token_accuracy": 0.1619458571076393,
|
|
"num_tokens": 4289096.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.0976330757141115,
|
|
"epoch": 2.031800601633004,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000481964619588932,
|
|
"loss": 5.8376,
|
|
"mean_token_accuracy": 0.15225235521793365,
|
|
"num_tokens": 4298341.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.118788385391236,
|
|
"epoch": 2.0360979802320585,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004818339649156523,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.14997191429138185,
|
|
"num_tokens": 4307466.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.107895755767823,
|
|
"epoch": 2.040395358831113,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048170285859966395,
|
|
"loss": 5.8004,
|
|
"mean_token_accuracy": 0.1605072647333145,
|
|
"num_tokens": 4316559.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.136744546890259,
|
|
"epoch": 2.0446927374301676,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048157130092725087,
|
|
"loss": 5.8229,
|
|
"mean_token_accuracy": 0.15814021229743958,
|
|
"num_tokens": 4326002.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.073250246047974,
|
|
"epoch": 2.048990116029222,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004814392921856824,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.16814109385013581,
|
|
"num_tokens": 4333767.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.123864555358887,
|
|
"epoch": 2.0532874946282766,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004813068326632128,
|
|
"loss": 5.8407,
|
|
"mean_token_accuracy": 0.1677692249417305,
|
|
"num_tokens": 4342726.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.03325662612915,
|
|
"epoch": 2.057584873227331,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004811739226490809,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.164510577917099,
|
|
"num_tokens": 4351278.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.029361057281494,
|
|
"epoch": 2.0618822518263857,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048104056243350896,
|
|
"loss": 5.7613,
|
|
"mean_token_accuracy": 0.1730365440249443,
|
|
"num_tokens": 4360146.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.069676494598388,
|
|
"epoch": 2.0661796304254403,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004809067523077023,
|
|
"loss": 5.81,
|
|
"mean_token_accuracy": 0.1618461474776268,
|
|
"num_tokens": 4368474.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.122001266479492,
|
|
"epoch": 2.0704770090244953,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048077249256384884,
|
|
"loss": 5.889,
|
|
"mean_token_accuracy": 0.15631877779960632,
|
|
"num_tokens": 4378430.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.074360942840576,
|
|
"epoch": 2.07477438762355,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004806377834951182,
|
|
"loss": 5.8821,
|
|
"mean_token_accuracy": 0.15965779721736909,
|
|
"num_tokens": 4387661.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.1412880420684814,
|
|
"epoch": 2.0790717662226044,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048050262539566104,
|
|
"loss": 5.8314,
|
|
"mean_token_accuracy": 0.1613117054104805,
|
|
"num_tokens": 4396568.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.996071815490723,
|
|
"epoch": 2.083369144821659,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004803670185606087,
|
|
"loss": 5.7781,
|
|
"mean_token_accuracy": 0.16476011425256729,
|
|
"num_tokens": 4405113.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.1124523162841795,
|
|
"epoch": 2.0876665234207135,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004802309632860724,
|
|
"loss": 5.8673,
|
|
"mean_token_accuracy": 0.1617572069168091,
|
|
"num_tokens": 4414930.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.101985788345337,
|
|
"epoch": 2.091963902019768,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048009445986914236,
|
|
"loss": 5.8943,
|
|
"mean_token_accuracy": 0.15561344623565673,
|
|
"num_tokens": 4424200.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.046984767913818,
|
|
"epoch": 2.0962612806188226,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047995750860788756,
|
|
"loss": 5.8467,
|
|
"mean_token_accuracy": 0.1636588454246521,
|
|
"num_tokens": 4433533.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.107754755020141,
|
|
"epoch": 2.100558659217877,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004798201098013547,
|
|
"loss": 5.8871,
|
|
"mean_token_accuracy": 0.16322858333587648,
|
|
"num_tokens": 4442752.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.040629911422729,
|
|
"epoch": 2.1048560378169316,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047968226374956797,
|
|
"loss": 5.9185,
|
|
"mean_token_accuracy": 0.1524975925683975,
|
|
"num_tokens": 4453624.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.105652713775635,
|
|
"epoch": 2.109153416415986,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047954397075352794,
|
|
"loss": 5.9152,
|
|
"mean_token_accuracy": 0.1547384575009346,
|
|
"num_tokens": 4462101.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.125913143157959,
|
|
"epoch": 2.1134507950150407,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047940523111521136,
|
|
"loss": 5.9316,
|
|
"mean_token_accuracy": 0.1496809035539627,
|
|
"num_tokens": 4471274.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.113536024093628,
|
|
"epoch": 2.1177481736140953,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004792660451375701,
|
|
"loss": 5.8882,
|
|
"mean_token_accuracy": 0.15914952903985977,
|
|
"num_tokens": 4480281.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.109311866760254,
|
|
"epoch": 2.12204555221315,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047912641312453064,
|
|
"loss": 5.8202,
|
|
"mean_token_accuracy": 0.1615729033946991,
|
|
"num_tokens": 4489325.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.092775630950928,
|
|
"epoch": 2.1263429308122044,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047898633538099363,
|
|
"loss": 5.8065,
|
|
"mean_token_accuracy": 0.1640813648700714,
|
|
"num_tokens": 4497977.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.024752140045166,
|
|
"epoch": 2.130640309411259,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004788458122128327,
|
|
"loss": 5.9332,
|
|
"mean_token_accuracy": 0.15297108441591262,
|
|
"num_tokens": 4509133.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.080051136016846,
|
|
"epoch": 2.134937688010314,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047870484392689434,
|
|
"loss": 5.8706,
|
|
"mean_token_accuracy": 0.1625355988740921,
|
|
"num_tokens": 4518431.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.0470091819763185,
|
|
"epoch": 2.1392350666093685,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000478563430830997,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.16243069469928742,
|
|
"num_tokens": 4528489.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.141346740722656,
|
|
"epoch": 2.143532445208423,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047842157323393035,
|
|
"loss": 5.772,
|
|
"mean_token_accuracy": 0.1622501254081726,
|
|
"num_tokens": 4537068.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.048506736755371,
|
|
"epoch": 2.1478298238074776,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004782792714454547,
|
|
"loss": 5.8698,
|
|
"mean_token_accuracy": 0.16409993469715117,
|
|
"num_tokens": 4545773.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.1478298238074776,
|
|
"eval_entropy": 5.904563934953363,
|
|
"eval_loss": 6.252427101135254,
|
|
"eval_mean_token_accuracy": 0.1526490680135048,
|
|
"eval_num_tokens": 4545773.0,
|
|
"eval_runtime": 2.6419,
|
|
"eval_samples_per_second": 1343.356,
|
|
"eval_steps_per_second": 168.061,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.071846055984497,
|
|
"epoch": 2.152127202406532,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004781365257763002,
|
|
"loss": 5.9078,
|
|
"mean_token_accuracy": 0.15740047991275788,
|
|
"num_tokens": 4555358.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.051876068115234,
|
|
"epoch": 2.1564245810055866,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047799333653816633,
|
|
"loss": 5.7342,
|
|
"mean_token_accuracy": 0.16167396157979966,
|
|
"num_tokens": 4563550.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.0851177215576175,
|
|
"epoch": 2.160721959604641,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047784970404372124,
|
|
"loss": 5.8633,
|
|
"mean_token_accuracy": 0.15467997789382934,
|
|
"num_tokens": 4572517.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.993022775650024,
|
|
"epoch": 2.1650193382036957,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047770562860660083,
|
|
"loss": 5.7783,
|
|
"mean_token_accuracy": 0.1633142426609993,
|
|
"num_tokens": 4581611.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.051138591766358,
|
|
"epoch": 2.1693167168027503,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004775611105414083,
|
|
"loss": 5.8011,
|
|
"mean_token_accuracy": 0.16601613759994507,
|
|
"num_tokens": 4590433.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.099878931045533,
|
|
"epoch": 2.173614095401805,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004774161501637133,
|
|
"loss": 5.9225,
|
|
"mean_token_accuracy": 0.15691807717084885,
|
|
"num_tokens": 4599363.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.089349889755249,
|
|
"epoch": 2.1779114740008594,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004772707477900514,
|
|
"loss": 5.8646,
|
|
"mean_token_accuracy": 0.15824637860059737,
|
|
"num_tokens": 4608647.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.087095451354981,
|
|
"epoch": 2.182208852599914,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004771249037379232,
|
|
"loss": 5.8386,
|
|
"mean_token_accuracy": 0.15788670778274536,
|
|
"num_tokens": 4617395.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.0928810119628904,
|
|
"epoch": 2.1865062311989685,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004769786183257939,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.1567574843764305,
|
|
"num_tokens": 4627905.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.062714338302612,
|
|
"epoch": 2.190803609798023,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004768318918730924,
|
|
"loss": 5.8809,
|
|
"mean_token_accuracy": 0.15509230345487596,
|
|
"num_tokens": 4636679.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.076866483688354,
|
|
"epoch": 2.195100988397078,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047668472470021044,
|
|
"loss": 5.8577,
|
|
"mean_token_accuracy": 0.16147154271602632,
|
|
"num_tokens": 4645963.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.9767965316772464,
|
|
"epoch": 2.1993983669961326,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004765371171285025,
|
|
"loss": 5.8373,
|
|
"mean_token_accuracy": 0.1610219269990921,
|
|
"num_tokens": 4656335.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.011840438842773,
|
|
"epoch": 2.203695745595187,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047638906948028445,
|
|
"loss": 5.8077,
|
|
"mean_token_accuracy": 0.1693956509232521,
|
|
"num_tokens": 4664924.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.089924144744873,
|
|
"epoch": 2.2079931241942417,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047624058207883317,
|
|
"loss": 5.8463,
|
|
"mean_token_accuracy": 0.16108024567365647,
|
|
"num_tokens": 4674411.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.920033025741577,
|
|
"epoch": 2.212290502793296,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047609165524838576,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.16555635184049605,
|
|
"num_tokens": 4682856.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.064959621429443,
|
|
"epoch": 2.2165878813923507,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004759422893141389,
|
|
"loss": 5.864,
|
|
"mean_token_accuracy": 0.1585097059607506,
|
|
"num_tokens": 4691652.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.0054035663604735,
|
|
"epoch": 2.2208852599914053,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004757924846022482,
|
|
"loss": 5.7852,
|
|
"mean_token_accuracy": 0.16821536868810655,
|
|
"num_tokens": 4700934.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.044727849960327,
|
|
"epoch": 2.22518263859046,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047564224143982714,
|
|
"loss": 5.8701,
|
|
"mean_token_accuracy": 0.1659425362944603,
|
|
"num_tokens": 4709195.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.051371097564697,
|
|
"epoch": 2.2294800171895144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047549156015494676,
|
|
"loss": 5.9571,
|
|
"mean_token_accuracy": 0.15879578590393068,
|
|
"num_tokens": 4719153.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.180147266387939,
|
|
"epoch": 2.233777395788569,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047534044107663484,
|
|
"loss": 5.8897,
|
|
"mean_token_accuracy": 0.15573096573352813,
|
|
"num_tokens": 4727672.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.027987289428711,
|
|
"epoch": 2.2380747743876235,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047518888453487496,
|
|
"loss": 5.815,
|
|
"mean_token_accuracy": 0.1587490826845169,
|
|
"num_tokens": 4735893.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.009792184829712,
|
|
"epoch": 2.242372152986678,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004750368908606061,
|
|
"loss": 5.8351,
|
|
"mean_token_accuracy": 0.1585497111082077,
|
|
"num_tokens": 4744022.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.052504634857177,
|
|
"epoch": 2.2466695315857326,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00047488446038572164,
|
|
"loss": 5.9233,
|
|
"mean_token_accuracy": 0.15686811581254007,
|
|
"num_tokens": 4752996.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.069022941589355,
|
|
"epoch": 2.250966910184787,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004747315934430688,
|
|
"loss": 5.8228,
|
|
"mean_token_accuracy": 0.16428692042827606,
|
|
"num_tokens": 4762116.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.985714483261108,
|
|
"epoch": 2.2552642887838417,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000474578290366448,
|
|
"loss": 5.8516,
|
|
"mean_token_accuracy": 0.16084216237068177,
|
|
"num_tokens": 4771154.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.015341901779175,
|
|
"epoch": 2.259561667382896,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004744245514906117,
|
|
"loss": 5.8852,
|
|
"mean_token_accuracy": 0.16002282053232192,
|
|
"num_tokens": 4779691.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.119104862213135,
|
|
"epoch": 2.263859045981951,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047427037715126426,
|
|
"loss": 5.8451,
|
|
"mean_token_accuracy": 0.16192725151777268,
|
|
"num_tokens": 4788033.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.9754243850708,
|
|
"epoch": 2.2681564245810057,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004741157676850608,
|
|
"loss": 5.8391,
|
|
"mean_token_accuracy": 0.16340215802192687,
|
|
"num_tokens": 4797495.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.059681034088134,
|
|
"epoch": 2.2724538031800603,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047396072342960663,
|
|
"loss": 5.8809,
|
|
"mean_token_accuracy": 0.1521667018532753,
|
|
"num_tokens": 4806598.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.053537321090698,
|
|
"epoch": 2.276751181779115,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047380524472345645,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.15857099145650863,
|
|
"num_tokens": 4816135.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.061359167098999,
|
|
"epoch": 2.2810485603781694,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004736493319061134,
|
|
"loss": 5.8807,
|
|
"mean_token_accuracy": 0.16125693172216415,
|
|
"num_tokens": 4825927.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.045619821548462,
|
|
"epoch": 2.285345938977224,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004734929853180291,
|
|
"loss": 5.812,
|
|
"mean_token_accuracy": 0.16086436361074447,
|
|
"num_tokens": 4834068.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.985046768188477,
|
|
"epoch": 2.2896433175762785,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047333620530060175,
|
|
"loss": 5.7847,
|
|
"mean_token_accuracy": 0.16806644648313523,
|
|
"num_tokens": 4843096.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.978422594070435,
|
|
"epoch": 2.293940696175333,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0004731789921961764,
|
|
"loss": 5.8517,
|
|
"mean_token_accuracy": 0.15997182875871657,
|
|
"num_tokens": 4850948.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.107482051849365,
|
|
"epoch": 2.2982380747743876,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004730213463480434,
|
|
"loss": 5.9563,
|
|
"mean_token_accuracy": 0.16070407629013062,
|
|
"num_tokens": 4860416.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.077378416061402,
|
|
"epoch": 2.302535453373442,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047286326810043857,
|
|
"loss": 5.8143,
|
|
"mean_token_accuracy": 0.16285796463489532,
|
|
"num_tokens": 4868815.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.011205101013184,
|
|
"epoch": 2.3068328319724967,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047270475779854137,
|
|
"loss": 5.7873,
|
|
"mean_token_accuracy": 0.16168000251054765,
|
|
"num_tokens": 4877973.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.025539493560791,
|
|
"epoch": 2.311130210571551,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047254581578847507,
|
|
"loss": 5.8485,
|
|
"mean_token_accuracy": 0.16380018293857573,
|
|
"num_tokens": 4887294.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.984267330169677,
|
|
"epoch": 2.3154275891706058,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004723864424173055,
|
|
"loss": 5.8287,
|
|
"mean_token_accuracy": 0.16092878431081772,
|
|
"num_tokens": 4895618.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.986261415481567,
|
|
"epoch": 2.3197249677696608,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004722266380330403,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.16197982728481292,
|
|
"num_tokens": 4904702.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.0529139041900635,
|
|
"epoch": 2.3240223463687153,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047206640298462857,
|
|
"loss": 5.8894,
|
|
"mean_token_accuracy": 0.17073851823806763,
|
|
"num_tokens": 4913677.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.063980150222778,
|
|
"epoch": 2.32831972496777,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00047190573762195945,
|
|
"loss": 6.0403,
|
|
"mean_token_accuracy": 0.14613149017095567,
|
|
"num_tokens": 4924192.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.0608070373535154,
|
|
"epoch": 2.3326171035668244,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047174464229586186,
|
|
"loss": 5.7443,
|
|
"mean_token_accuracy": 0.17208250164985656,
|
|
"num_tokens": 4933424.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.004354619979859,
|
|
"epoch": 2.336914482165879,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004715831173581036,
|
|
"loss": 5.8841,
|
|
"mean_token_accuracy": 0.15948930084705354,
|
|
"num_tokens": 4942695.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.082979011535644,
|
|
"epoch": 2.3412118607649335,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047142116316139073,
|
|
"loss": 5.9818,
|
|
"mean_token_accuracy": 0.160202556848526,
|
|
"num_tokens": 4952951.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.969216823577881,
|
|
"epoch": 2.345509239363988,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004712587800593663,
|
|
"loss": 5.8281,
|
|
"mean_token_accuracy": 0.16207036674022673,
|
|
"num_tokens": 4961750.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.071132469177246,
|
|
"epoch": 2.3498066179630426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004710959684066102,
|
|
"loss": 5.9409,
|
|
"mean_token_accuracy": 0.15581327080726623,
|
|
"num_tokens": 4970999.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.987265491485596,
|
|
"epoch": 2.354103996562097,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047093272855863803,
|
|
"loss": 5.7371,
|
|
"mean_token_accuracy": 0.16952835246920586,
|
|
"num_tokens": 4979703.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.004208040237427,
|
|
"epoch": 2.3584013751611517,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004707690608719003,
|
|
"loss": 5.8161,
|
|
"mean_token_accuracy": 0.16796157360076905,
|
|
"num_tokens": 4988666.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.986488962173462,
|
|
"epoch": 2.362698753760206,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004706049657037818,
|
|
"loss": 5.8503,
|
|
"mean_token_accuracy": 0.16016759127378463,
|
|
"num_tokens": 4997741.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.038318729400634,
|
|
"epoch": 2.3669961323592608,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004704404434126009,
|
|
"loss": 5.7894,
|
|
"mean_token_accuracy": 0.1624614730477333,
|
|
"num_tokens": 5006067.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.032493543624878,
|
|
"epoch": 2.3712935109583153,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047027549435760843,
|
|
"loss": 5.7965,
|
|
"mean_token_accuracy": 0.15930217355489731,
|
|
"num_tokens": 5015212.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.014610290527344,
|
|
"epoch": 2.37559088955737,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004701101188989872,
|
|
"loss": 5.9034,
|
|
"mean_token_accuracy": 0.16026949286460876,
|
|
"num_tokens": 5025461.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.03846001625061,
|
|
"epoch": 2.3798882681564244,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00046994431739785114,
|
|
"loss": 5.8775,
|
|
"mean_token_accuracy": 0.1588368222117424,
|
|
"num_tokens": 5033479.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.028546094894409,
|
|
"epoch": 2.384185646755479,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046977809021624454,
|
|
"loss": 5.9274,
|
|
"mean_token_accuracy": 0.16070414036512376,
|
|
"num_tokens": 5043098.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.025289678573609,
|
|
"epoch": 2.3884830253545335,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004696114377171409,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.16839157789945602,
|
|
"num_tokens": 5052785.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.994251012802124,
|
|
"epoch": 2.3927804039535885,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004694443602644429,
|
|
"loss": 5.8354,
|
|
"mean_token_accuracy": 0.16162479519844056,
|
|
"num_tokens": 5061947.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.045731544494629,
|
|
"epoch": 2.397077782552643,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004692768582229808,
|
|
"loss": 5.8513,
|
|
"mean_token_accuracy": 0.16438515037298201,
|
|
"num_tokens": 5070624.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.072655868530274,
|
|
"epoch": 2.4013751611516976,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00046910893195851213,
|
|
"loss": 5.8882,
|
|
"mean_token_accuracy": 0.16443493068218232,
|
|
"num_tokens": 5079290.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.928467702865601,
|
|
"epoch": 2.405672539750752,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046894058183772074,
|
|
"loss": 5.7806,
|
|
"mean_token_accuracy": 0.16285862177610397,
|
|
"num_tokens": 5088214.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.986145830154419,
|
|
"epoch": 2.4099699183498067,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000468771808228216,
|
|
"loss": 5.8065,
|
|
"mean_token_accuracy": 0.16282968968153,
|
|
"num_tokens": 5096752.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.038999795913696,
|
|
"epoch": 2.414267296948861,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00046860261149853197,
|
|
"loss": 5.9055,
|
|
"mean_token_accuracy": 0.1631876140832901,
|
|
"num_tokens": 5105204.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.035766458511352,
|
|
"epoch": 2.4185646755479158,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004684329920181268,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.16995367854833604,
|
|
"num_tokens": 5114306.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.994801473617554,
|
|
"epoch": 2.4228620541469703,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046826295015738154,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.17038536816835403,
|
|
"num_tokens": 5122894.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.962534093856812,
|
|
"epoch": 2.427159432746025,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004680924862875996,
|
|
"loss": 5.9293,
|
|
"mean_token_accuracy": 0.16031765043735505,
|
|
"num_tokens": 5131660.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.0869300365448,
|
|
"epoch": 2.4314568113450794,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046792160078100605,
|
|
"loss": 5.8735,
|
|
"mean_token_accuracy": 0.16528491526842118,
|
|
"num_tokens": 5140165.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.027950811386108,
|
|
"epoch": 2.435754189944134,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046775029401074653,
|
|
"loss": 5.8481,
|
|
"mean_token_accuracy": 0.16267438381910324,
|
|
"num_tokens": 5149998.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.000285339355469,
|
|
"epoch": 2.4400515685431885,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046757856635088645,
|
|
"loss": 5.8275,
|
|
"mean_token_accuracy": 0.15965816080570222,
|
|
"num_tokens": 5159182.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.107571029663086,
|
|
"epoch": 2.444348947142243,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004674064181764105,
|
|
"loss": 5.9902,
|
|
"mean_token_accuracy": 0.1484472408890724,
|
|
"num_tokens": 5169087.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.016386556625366,
|
|
"epoch": 2.448646325741298,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046723384986322147,
|
|
"loss": 5.9129,
|
|
"mean_token_accuracy": 0.16184251755475998,
|
|
"num_tokens": 5177670.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.000235605239868,
|
|
"epoch": 2.4529437043403526,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004670608617881395,
|
|
"loss": 5.9895,
|
|
"mean_token_accuracy": 0.15571076422929764,
|
|
"num_tokens": 5187273.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.049048662185669,
|
|
"epoch": 2.457241082939407,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004668874543289014,
|
|
"loss": 5.8521,
|
|
"mean_token_accuracy": 0.1639543741941452,
|
|
"num_tokens": 5196546.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.047182607650757,
|
|
"epoch": 2.4615384615384617,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046671362786415986,
|
|
"loss": 5.868,
|
|
"mean_token_accuracy": 0.16193134039640428,
|
|
"num_tokens": 5205966.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.986069393157959,
|
|
"epoch": 2.465835840137516,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00046653938277348237,
|
|
"loss": 5.9371,
|
|
"mean_token_accuracy": 0.1564442351460457,
|
|
"num_tokens": 5216042.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.9859216690063475,
|
|
"epoch": 2.4701332187365708,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004663647194373505,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.1584170863032341,
|
|
"num_tokens": 5225514.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.070285224914551,
|
|
"epoch": 2.4744305973356253,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046618963823715913,
|
|
"loss": 5.8435,
|
|
"mean_token_accuracy": 0.1591175839304924,
|
|
"num_tokens": 5233705.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.038730239868164,
|
|
"epoch": 2.47872797593468,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046601413955521575,
|
|
"loss": 5.8938,
|
|
"mean_token_accuracy": 0.15731203854084014,
|
|
"num_tokens": 5242820.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.0063272476196286,
|
|
"epoch": 2.4830253545337344,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004658382237747393,
|
|
"loss": 5.9412,
|
|
"mean_token_accuracy": 0.16072043478488923,
|
|
"num_tokens": 5252036.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.932094287872315,
|
|
"epoch": 2.487322733132789,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046566189127985946,
|
|
"loss": 5.7774,
|
|
"mean_token_accuracy": 0.16504888385534286,
|
|
"num_tokens": 5260734.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.045875883102417,
|
|
"epoch": 2.4916201117318435,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000465485142455616,
|
|
"loss": 5.9137,
|
|
"mean_token_accuracy": 0.1599499188363552,
|
|
"num_tokens": 5269764.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.092646837234497,
|
|
"epoch": 2.495917490330898,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046530797768795765,
|
|
"loss": 5.8703,
|
|
"mean_token_accuracy": 0.16274474710226058,
|
|
"num_tokens": 5278585.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.939150762557984,
|
|
"epoch": 2.5002148689299526,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046513039736374153,
|
|
"loss": 5.9042,
|
|
"mean_token_accuracy": 0.1589077964425087,
|
|
"num_tokens": 5287335.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.120346546173096,
|
|
"epoch": 2.504512247529007,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004649524018707319,
|
|
"loss": 6.0321,
|
|
"mean_token_accuracy": 0.1470044195652008,
|
|
"num_tokens": 5297977.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.078829956054688,
|
|
"epoch": 2.5088096261280617,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00046477399159759996,
|
|
"loss": 5.8862,
|
|
"mean_token_accuracy": 0.15342434048652648,
|
|
"num_tokens": 5307148.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.070769643783569,
|
|
"epoch": 2.5131070047271162,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046459516693392246,
|
|
"loss": 5.9408,
|
|
"mean_token_accuracy": 0.15786074995994567,
|
|
"num_tokens": 5317046.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.1084699630737305,
|
|
"epoch": 2.517404383326171,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004644159282701808,
|
|
"loss": 5.9463,
|
|
"mean_token_accuracy": 0.16232722848653794,
|
|
"num_tokens": 5325978.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.984037685394287,
|
|
"epoch": 2.5217017619252258,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046423627599776076,
|
|
"loss": 5.8166,
|
|
"mean_token_accuracy": 0.16753552109003067,
|
|
"num_tokens": 5334804.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.011123323440552,
|
|
"epoch": 2.5259991405242803,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000464056210508951,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.16593734920024872,
|
|
"num_tokens": 5343575.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.933651447296143,
|
|
"epoch": 2.530296519123335,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004638757321969426,
|
|
"loss": 5.8152,
|
|
"mean_token_accuracy": 0.16034290194511414,
|
|
"num_tokens": 5352151.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.013668155670166,
|
|
"epoch": 2.5345938977223894,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046369484145582815,
|
|
"loss": 5.979,
|
|
"mean_token_accuracy": 0.14990964382886887,
|
|
"num_tokens": 5363048.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.094151353836059,
|
|
"epoch": 2.538891276321444,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046351353868060054,
|
|
"loss": 5.915,
|
|
"mean_token_accuracy": 0.15911020636558532,
|
|
"num_tokens": 5372365.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.9932348251342775,
|
|
"epoch": 2.5431886549204985,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046333182426715273,
|
|
"loss": 5.8556,
|
|
"mean_token_accuracy": 0.1618341937661171,
|
|
"num_tokens": 5381583.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.938994073867798,
|
|
"epoch": 2.547486033519553,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046314969861227626,
|
|
"loss": 5.8308,
|
|
"mean_token_accuracy": 0.1632826253771782,
|
|
"num_tokens": 5390746.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.052537298202514,
|
|
"epoch": 2.5517834121186076,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004629671621136608,
|
|
"loss": 5.871,
|
|
"mean_token_accuracy": 0.1576824888586998,
|
|
"num_tokens": 5399715.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.043849658966065,
|
|
"epoch": 2.556080790717662,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004627842151698931,
|
|
"loss": 5.8969,
|
|
"mean_token_accuracy": 0.16249217838048935,
|
|
"num_tokens": 5407906.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.964152765274048,
|
|
"epoch": 2.5603781693167167,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00046260085818045625,
|
|
"loss": 5.837,
|
|
"mean_token_accuracy": 0.16123417466878892,
|
|
"num_tokens": 5416935.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.996441411972046,
|
|
"epoch": 2.5646755479157712,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004624170915457284,
|
|
"loss": 5.8913,
|
|
"mean_token_accuracy": 0.15868400782346725,
|
|
"num_tokens": 5426061.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.013878536224365,
|
|
"epoch": 2.5689729265148262,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046223291566698264,
|
|
"loss": 5.8837,
|
|
"mean_token_accuracy": 0.15597704648971558,
|
|
"num_tokens": 5435546.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.004412698745727,
|
|
"epoch": 2.5732703051138808,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004620483309463855,
|
|
"loss": 5.8658,
|
|
"mean_token_accuracy": 0.16499627828598024,
|
|
"num_tokens": 5445026.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.009965515136718,
|
|
"epoch": 2.5775676837129353,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004618633377869961,
|
|
"loss": 5.8086,
|
|
"mean_token_accuracy": 0.1655992992222309,
|
|
"num_tokens": 5453936.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.5775676837129353,
|
|
"eval_entropy": 5.927217808929649,
|
|
"eval_loss": 6.199429988861084,
|
|
"eval_mean_token_accuracy": 0.15587029703431302,
|
|
"eval_num_tokens": 5453936.0,
|
|
"eval_runtime": 2.6386,
|
|
"eval_samples_per_second": 1345.011,
|
|
"eval_steps_per_second": 168.268,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.040154361724854,
|
|
"epoch": 2.58186506231199,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004616779365927656,
|
|
"loss": 5.8739,
|
|
"mean_token_accuracy": 0.1727909952402115,
|
|
"num_tokens": 5462313.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.9107612609863285,
|
|
"epoch": 2.5861624409110444,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004614921277685361,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.16819749325513839,
|
|
"num_tokens": 5471500.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.942871713638306,
|
|
"epoch": 2.590459819510099,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046130591172003976,
|
|
"loss": 5.7968,
|
|
"mean_token_accuracy": 0.17107525765895842,
|
|
"num_tokens": 5480317.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.950167942047119,
|
|
"epoch": 2.5947571981091535,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004611192888538981,
|
|
"loss": 5.8505,
|
|
"mean_token_accuracy": 0.15793465673923493,
|
|
"num_tokens": 5490159.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.067723894119263,
|
|
"epoch": 2.599054576708208,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00046093225957762084,
|
|
"loss": 5.8182,
|
|
"mean_token_accuracy": 0.16943113654851913,
|
|
"num_tokens": 5499055.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.040552759170533,
|
|
"epoch": 2.6033519553072626,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004607448242996051,
|
|
"loss": 5.9241,
|
|
"mean_token_accuracy": 0.16228144615888596,
|
|
"num_tokens": 5508710.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.96411509513855,
|
|
"epoch": 2.607649333906317,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004605569834291347,
|
|
"loss": 5.8694,
|
|
"mean_token_accuracy": 0.16797070503234862,
|
|
"num_tokens": 5517060.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.886189937591553,
|
|
"epoch": 2.6119467125053717,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046036873737637904,
|
|
"loss": 5.7903,
|
|
"mean_token_accuracy": 0.1609458565711975,
|
|
"num_tokens": 5525652.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.049513244628907,
|
|
"epoch": 2.6162440911044262,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004601800865523921,
|
|
"loss": 5.9254,
|
|
"mean_token_accuracy": 0.15415238887071608,
|
|
"num_tokens": 5535150.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.997943162918091,
|
|
"epoch": 2.620541469703481,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00045999103136911204,
|
|
"loss": 5.912,
|
|
"mean_token_accuracy": 0.16251415014266968,
|
|
"num_tokens": 5545121.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.913229990005493,
|
|
"epoch": 2.6248388483025353,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045980157223935965,
|
|
"loss": 5.7407,
|
|
"mean_token_accuracy": 0.16793070435523988,
|
|
"num_tokens": 5553109.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.917746305465698,
|
|
"epoch": 2.62913622690159,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045961170957683806,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.1712057888507843,
|
|
"num_tokens": 5562020.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.967804384231568,
|
|
"epoch": 2.6334336055006444,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045942144379613147,
|
|
"loss": 5.8134,
|
|
"mean_token_accuracy": 0.1609429031610489,
|
|
"num_tokens": 5571188.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.9468371868133545,
|
|
"epoch": 2.637730984099699,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045923077531270426,
|
|
"loss": 5.8814,
|
|
"mean_token_accuracy": 0.16267036646604538,
|
|
"num_tokens": 5580481.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.058697366714478,
|
|
"epoch": 2.6420283626987535,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004590397045429001,
|
|
"loss": 5.8284,
|
|
"mean_token_accuracy": 0.16753861606121062,
|
|
"num_tokens": 5589710.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.9068666934967045,
|
|
"epoch": 2.646325741297808,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045884823190394134,
|
|
"loss": 5.8227,
|
|
"mean_token_accuracy": 0.16702128946781158,
|
|
"num_tokens": 5599083.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.957175970077515,
|
|
"epoch": 2.650623119896863,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004586563578139275,
|
|
"loss": 5.7608,
|
|
"mean_token_accuracy": 0.17045541107654572,
|
|
"num_tokens": 5607984.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.983618211746216,
|
|
"epoch": 2.6549204984959176,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00045846408269183505,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.16661562472581865,
|
|
"num_tokens": 5617470.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.072385835647583,
|
|
"epoch": 2.659217877094972,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045827140695751603,
|
|
"loss": 5.9119,
|
|
"mean_token_accuracy": 0.1572274535894394,
|
|
"num_tokens": 5626982.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.896085548400879,
|
|
"epoch": 2.6635152556940267,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004580783310316971,
|
|
"loss": 5.8128,
|
|
"mean_token_accuracy": 0.16616976857185364,
|
|
"num_tokens": 5635758.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.960517024993896,
|
|
"epoch": 2.6678126342930812,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00045788485533597895,
|
|
"loss": 5.8195,
|
|
"mean_token_accuracy": 0.16438868939876555,
|
|
"num_tokens": 5644890.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.991481637954712,
|
|
"epoch": 2.672110012892136,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045769098029283526,
|
|
"loss": 5.8077,
|
|
"mean_token_accuracy": 0.16789740025997163,
|
|
"num_tokens": 5653627.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.89129638671875,
|
|
"epoch": 2.6764073914911903,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004574967063256115,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.16654334515333175,
|
|
"num_tokens": 5662513.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.916149091720581,
|
|
"epoch": 2.680704770090245,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045730203385852447,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.1616238087415695,
|
|
"num_tokens": 5671458.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.026506137847901,
|
|
"epoch": 2.6850021486892994,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000457106963316661,
|
|
"loss": 5.8611,
|
|
"mean_token_accuracy": 0.16229415982961654,
|
|
"num_tokens": 5679728.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.925787544250488,
|
|
"epoch": 2.689299527288354,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045691149512597717,
|
|
"loss": 5.8691,
|
|
"mean_token_accuracy": 0.1628123015165329,
|
|
"num_tokens": 5688662.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.969518184661865,
|
|
"epoch": 2.6935969058874085,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045671562971329736,
|
|
"loss": 5.8832,
|
|
"mean_token_accuracy": 0.15795952081680298,
|
|
"num_tokens": 5697442.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.109791374206543,
|
|
"epoch": 2.6978942844864635,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045651936750631337,
|
|
"loss": 5.8931,
|
|
"mean_token_accuracy": 0.15932094007730485,
|
|
"num_tokens": 5706496.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.985749340057373,
|
|
"epoch": 2.702191663085518,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045632270893358333,
|
|
"loss": 5.9209,
|
|
"mean_token_accuracy": 0.16183084100484849,
|
|
"num_tokens": 5715954.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.065780544281006,
|
|
"epoch": 2.7064890416845726,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004561256544245312,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.15481958016753197,
|
|
"num_tokens": 5725276.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.002734851837158,
|
|
"epoch": 2.710786420283627,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000455928204409445,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.15934307724237443,
|
|
"num_tokens": 5734497.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.921933841705322,
|
|
"epoch": 2.7150837988826817,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045573035931947684,
|
|
"loss": 5.7649,
|
|
"mean_token_accuracy": 0.16860159635543823,
|
|
"num_tokens": 5743539.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.99278039932251,
|
|
"epoch": 2.7193811774817362,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004555321195866411,
|
|
"loss": 5.8654,
|
|
"mean_token_accuracy": 0.16891386210918427,
|
|
"num_tokens": 5752527.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.973156404495239,
|
|
"epoch": 2.723678556080791,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004553334856438143,
|
|
"loss": 5.8003,
|
|
"mean_token_accuracy": 0.16235947906970977,
|
|
"num_tokens": 5761890.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.030607748031616,
|
|
"epoch": 2.7279759346798453,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045513445792473356,
|
|
"loss": 5.8964,
|
|
"mean_token_accuracy": 0.15456002056598664,
|
|
"num_tokens": 5771041.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.929169940948486,
|
|
"epoch": 2.7322733132789,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004549350368639958,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.16194687634706498,
|
|
"num_tokens": 5779869.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.971911287307739,
|
|
"epoch": 2.7365706918779544,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00045473522289705693,
|
|
"loss": 5.7381,
|
|
"mean_token_accuracy": 0.1728900045156479,
|
|
"num_tokens": 5788850.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.005989217758179,
|
|
"epoch": 2.740868070477009,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045453501646023085,
|
|
"loss": 6.0005,
|
|
"mean_token_accuracy": 0.15093248039484025,
|
|
"num_tokens": 5798712.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.0704199314117435,
|
|
"epoch": 2.7451654490760635,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00045433441799068837,
|
|
"loss": 5.9725,
|
|
"mean_token_accuracy": 0.15257609635591507,
|
|
"num_tokens": 5808567.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.029580593109131,
|
|
"epoch": 2.749462827675118,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004541334279264562,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.1613762080669403,
|
|
"num_tokens": 5817810.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.981862306594849,
|
|
"epoch": 2.7537602062741726,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00045393204670641656,
|
|
"loss": 5.8176,
|
|
"mean_token_accuracy": 0.16276164203882218,
|
|
"num_tokens": 5827381.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.979506635665894,
|
|
"epoch": 2.758057584873227,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004537302747703055,
|
|
"loss": 5.8922,
|
|
"mean_token_accuracy": 0.16632017791271209,
|
|
"num_tokens": 5836325.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.023560333251953,
|
|
"epoch": 2.7623549634722817,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045352811255871216,
|
|
"loss": 5.8553,
|
|
"mean_token_accuracy": 0.1606232836842537,
|
|
"num_tokens": 5845125.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.966276597976685,
|
|
"epoch": 2.7666523420713363,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045332556051307804,
|
|
"loss": 5.7583,
|
|
"mean_token_accuracy": 0.16876666247844696,
|
|
"num_tokens": 5853837.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.997716283798217,
|
|
"epoch": 2.770949720670391,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00045312261907569585,
|
|
"loss": 5.8452,
|
|
"mean_token_accuracy": 0.16041913032531738,
|
|
"num_tokens": 5863832.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.9576208114624025,
|
|
"epoch": 2.775247099269446,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00045291928868970867,
|
|
"loss": 5.8876,
|
|
"mean_token_accuracy": 0.15914502441883088,
|
|
"num_tokens": 5872616.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.129185628890991,
|
|
"epoch": 2.7795444778685003,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004527155697991087,
|
|
"loss": 5.9726,
|
|
"mean_token_accuracy": 0.1521813154220581,
|
|
"num_tokens": 5882153.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.972490978240967,
|
|
"epoch": 2.783841856467555,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004525114628487365,
|
|
"loss": 5.8153,
|
|
"mean_token_accuracy": 0.1679455280303955,
|
|
"num_tokens": 5890650.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.901459693908691,
|
|
"epoch": 2.7881392350666094,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045230696828428026,
|
|
"loss": 5.8027,
|
|
"mean_token_accuracy": 0.167825648188591,
|
|
"num_tokens": 5899501.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.966943550109863,
|
|
"epoch": 2.792436613665664,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004521020865522742,
|
|
"loss": 5.8297,
|
|
"mean_token_accuracy": 0.16191235184669495,
|
|
"num_tokens": 5908040.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.932198715209961,
|
|
"epoch": 2.7967339922647185,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.00045189681810009827,
|
|
"loss": 5.709,
|
|
"mean_token_accuracy": 0.17586348950862885,
|
|
"num_tokens": 5916690.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.90502290725708,
|
|
"epoch": 2.801031370863773,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045169116337597653,
|
|
"loss": 5.8373,
|
|
"mean_token_accuracy": 0.1634879156947136,
|
|
"num_tokens": 5925228.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.973164701461792,
|
|
"epoch": 2.8053287494628276,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000451485122828977,
|
|
"loss": 5.7939,
|
|
"mean_token_accuracy": 0.16546232253313065,
|
|
"num_tokens": 5934289.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.986188840866089,
|
|
"epoch": 2.809626128061882,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045127869690900956,
|
|
"loss": 5.8559,
|
|
"mean_token_accuracy": 0.15859055817127227,
|
|
"num_tokens": 5943516.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.882618999481201,
|
|
"epoch": 2.8139235066609367,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045107188606682613,
|
|
"loss": 5.7903,
|
|
"mean_token_accuracy": 0.16830744594335556,
|
|
"num_tokens": 5951549.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.006636285781861,
|
|
"epoch": 2.8182208852599913,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004508646907540188,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.1634303957223892,
|
|
"num_tokens": 5960297.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.052435398101807,
|
|
"epoch": 2.8225182638590463,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004506571114230195,
|
|
"loss": 5.8999,
|
|
"mean_token_accuracy": 0.15769678801298143,
|
|
"num_tokens": 5969714.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.975861597061157,
|
|
"epoch": 2.826815642458101,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045044914852709824,
|
|
"loss": 5.851,
|
|
"mean_token_accuracy": 0.16105720549821853,
|
|
"num_tokens": 5978328.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.9491846561431885,
|
|
"epoch": 2.8311130210571553,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004502408025203631,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.15761610865592957,
|
|
"num_tokens": 5987010.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.962584829330444,
|
|
"epoch": 2.83541039965621,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004500320738577584,
|
|
"loss": 5.8199,
|
|
"mean_token_accuracy": 0.16470251828432084,
|
|
"num_tokens": 5996146.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.989235591888428,
|
|
"epoch": 2.8397077782552644,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044982296299506407,
|
|
"loss": 5.9413,
|
|
"mean_token_accuracy": 0.1508522428572178,
|
|
"num_tokens": 6006231.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.0159444332122805,
|
|
"epoch": 2.844005156854319,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004496134703888948,
|
|
"loss": 5.9457,
|
|
"mean_token_accuracy": 0.1512417882680893,
|
|
"num_tokens": 6015795.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.980672025680542,
|
|
"epoch": 2.8483025354533735,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00044940359649669846,
|
|
"loss": 5.7366,
|
|
"mean_token_accuracy": 0.1697762504220009,
|
|
"num_tokens": 6024370.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.073631238937378,
|
|
"epoch": 2.852599914052428,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00044919334177675595,
|
|
"loss": 5.8761,
|
|
"mean_token_accuracy": 0.15818028897047043,
|
|
"num_tokens": 6032511.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.898993730545044,
|
|
"epoch": 2.8568972926514826,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044898270668817955,
|
|
"loss": 5.8179,
|
|
"mean_token_accuracy": 0.16413381844758987,
|
|
"num_tokens": 6042039.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.970605754852295,
|
|
"epoch": 2.861194671250537,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000448771691690912,
|
|
"loss": 5.8935,
|
|
"mean_token_accuracy": 0.16170217096805573,
|
|
"num_tokens": 6051998.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.00086236000061,
|
|
"epoch": 2.8654920498495917,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004485602972457257,
|
|
"loss": 5.8414,
|
|
"mean_token_accuracy": 0.16919506341218948,
|
|
"num_tokens": 6060774.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.008049631118775,
|
|
"epoch": 2.8697894284486463,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00044834852381422165,
|
|
"loss": 5.8737,
|
|
"mean_token_accuracy": 0.16143593192100525,
|
|
"num_tokens": 6070915.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.8532084941864015,
|
|
"epoch": 2.874086807047701,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044813637185882836,
|
|
"loss": 5.8672,
|
|
"mean_token_accuracy": 0.1582166537642479,
|
|
"num_tokens": 6080609.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.980355739593506,
|
|
"epoch": 2.8783841856467554,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044792384184280106,
|
|
"loss": 5.8818,
|
|
"mean_token_accuracy": 0.16303323209285736,
|
|
"num_tokens": 6089231.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.953743362426758,
|
|
"epoch": 2.88268156424581,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044771093423022013,
|
|
"loss": 5.757,
|
|
"mean_token_accuracy": 0.17002557963132858,
|
|
"num_tokens": 6097835.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.008882141113281,
|
|
"epoch": 2.8869789428448644,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004474976494859909,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.1650140479207039,
|
|
"num_tokens": 6107141.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.017732191085815,
|
|
"epoch": 2.891276321443919,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004472839880758419,
|
|
"loss": 5.9082,
|
|
"mean_token_accuracy": 0.1615462228655815,
|
|
"num_tokens": 6116144.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.044085359573364,
|
|
"epoch": 2.8955737000429735,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 0.0004470699504663242,
|
|
"loss": 6.0668,
|
|
"mean_token_accuracy": 0.15092794522643088,
|
|
"num_tokens": 6126388.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.90790581703186,
|
|
"epoch": 2.899871078642028,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004468555371248104,
|
|
"loss": 5.8182,
|
|
"mean_token_accuracy": 0.17277036011219024,
|
|
"num_tokens": 6135347.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.053342056274414,
|
|
"epoch": 2.904168457241083,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004466407485194937,
|
|
"loss": 5.8804,
|
|
"mean_token_accuracy": 0.16284884214401246,
|
|
"num_tokens": 6144706.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.9684162616729735,
|
|
"epoch": 2.9084658358401376,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004464255851193864,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.16734683364629746,
|
|
"num_tokens": 6155417.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.938163900375367,
|
|
"epoch": 2.912763214439192,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004462100473943194,
|
|
"loss": 5.7289,
|
|
"mean_token_accuracy": 0.1742040902376175,
|
|
"num_tokens": 6165014.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.914160919189453,
|
|
"epoch": 2.9170605930382467,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000445994135814941,
|
|
"loss": 5.8186,
|
|
"mean_token_accuracy": 0.1670326218008995,
|
|
"num_tokens": 6174605.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.033651113510132,
|
|
"epoch": 2.9213579716373013,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044577785085271566,
|
|
"loss": 5.8807,
|
|
"mean_token_accuracy": 0.16132031083106996,
|
|
"num_tokens": 6183419.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.013253879547119,
|
|
"epoch": 2.925655350236356,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004455611929799235,
|
|
"loss": 5.8291,
|
|
"mean_token_accuracy": 0.16594379097223283,
|
|
"num_tokens": 6193050.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.981416988372803,
|
|
"epoch": 2.9299527288354104,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004453441626696585,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.1655766874551773,
|
|
"num_tokens": 6202330.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.975859355926514,
|
|
"epoch": 2.934250107434465,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00044512676039582823,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.1621957764029503,
|
|
"num_tokens": 6211195.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.999545860290527,
|
|
"epoch": 2.9385474860335195,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004449089866331524,
|
|
"loss": 5.8736,
|
|
"mean_token_accuracy": 0.15388306379318237,
|
|
"num_tokens": 6221116.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.940148782730103,
|
|
"epoch": 2.942844864632574,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004446908418571617,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.17225155532360076,
|
|
"num_tokens": 6229270.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.961176967620849,
|
|
"epoch": 2.9471422432316285,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004444723265441973,
|
|
"loss": 5.8327,
|
|
"mean_token_accuracy": 0.1613776430487633,
|
|
"num_tokens": 6238687.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.020242738723755,
|
|
"epoch": 2.9514396218306835,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004442534411714092,
|
|
"loss": 5.8065,
|
|
"mean_token_accuracy": 0.1669040083885193,
|
|
"num_tokens": 6247659.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.966691398620606,
|
|
"epoch": 2.955737000429738,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00044403418621675555,
|
|
"loss": 5.8771,
|
|
"mean_token_accuracy": 0.1676014930009842,
|
|
"num_tokens": 6256382.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.91684627532959,
|
|
"epoch": 2.9600343790287926,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004438145621590017,
|
|
"loss": 5.7602,
|
|
"mean_token_accuracy": 0.17016249001026154,
|
|
"num_tokens": 6264899.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.980312871932983,
|
|
"epoch": 2.964331757627847,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00044359456947771857,
|
|
"loss": 5.7983,
|
|
"mean_token_accuracy": 0.16945598572492598,
|
|
"num_tokens": 6273557.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.847117757797241,
|
|
"epoch": 2.9686291362269017,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004433742086532824,
|
|
"loss": 5.7581,
|
|
"mean_token_accuracy": 0.1744436129927635,
|
|
"num_tokens": 6283065.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.965201044082642,
|
|
"epoch": 2.9729265148259563,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00044315348016687317,
|
|
"loss": 5.8379,
|
|
"mean_token_accuracy": 0.1646779239177704,
|
|
"num_tokens": 6291800.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.966613721847534,
|
|
"epoch": 2.977223893425011,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004429323845004736,
|
|
"loss": 5.8599,
|
|
"mean_token_accuracy": 0.16521086245775224,
|
|
"num_tokens": 6300812.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.915075969696045,
|
|
"epoch": 2.9815212720240654,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044271092213686824,
|
|
"loss": 5.7683,
|
|
"mean_token_accuracy": 0.1773418366909027,
|
|
"num_tokens": 6309407.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.984325218200683,
|
|
"epoch": 2.98581865062312,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00044248909355964247,
|
|
"loss": 5.7903,
|
|
"mean_token_accuracy": 0.17371242493391037,
|
|
"num_tokens": 6317896.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.9681439876556395,
|
|
"epoch": 2.9901160292221745,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00044226689925318117,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.1653551772236824,
|
|
"num_tokens": 6326492.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.92264404296875,
|
|
"epoch": 2.994413407821229,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00044204433970266785,
|
|
"loss": 5.842,
|
|
"mean_token_accuracy": 0.1604343369603157,
|
|
"num_tokens": 6335552.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.000030946731568,
|
|
"epoch": 2.9987107864202835,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004418214153940837,
|
|
"loss": 5.8316,
|
|
"mean_token_accuracy": 0.16376646757125854,
|
|
"num_tokens": 6344620.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.953235891130236,
|
|
"epoch": 3.002578427159433,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044159812681420624,
|
|
"loss": 5.6261,
|
|
"mean_token_accuracy": 0.17379294335842133,
|
|
"num_tokens": 6352626.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.907259798049926,
|
|
"epoch": 3.0068758057584875,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004413744744506086,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.1750764086842537,
|
|
"num_tokens": 6361618.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.0068758057584875,
|
|
"eval_entropy": 5.784829720720515,
|
|
"eval_loss": 6.12911319732666,
|
|
"eval_mean_token_accuracy": 0.16059024073183537,
|
|
"eval_num_tokens": 6361618.0,
|
|
"eval_runtime": 2.6393,
|
|
"eval_samples_per_second": 1344.685,
|
|
"eval_steps_per_second": 168.228,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.976041603088379,
|
|
"epoch": 3.011173184357542,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00044115045879165806,
|
|
"loss": 5.5617,
|
|
"mean_token_accuracy": 0.17643589228391648,
|
|
"num_tokens": 6370729.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.912074232101441,
|
|
"epoch": 3.0154705629565965,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00044092608032651515,
|
|
"loss": 5.4723,
|
|
"mean_token_accuracy": 0.18105302900075912,
|
|
"num_tokens": 6380043.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.887026357650757,
|
|
"epoch": 3.019767941555651,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00044070133954513305,
|
|
"loss": 5.4741,
|
|
"mean_token_accuracy": 0.17589628100395202,
|
|
"num_tokens": 6389369.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.882784414291382,
|
|
"epoch": 3.0240653201547056,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004404762369382555,
|
|
"loss": 5.5424,
|
|
"mean_token_accuracy": 0.17499220222234727,
|
|
"num_tokens": 6399619.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.961531066894532,
|
|
"epoch": 3.02836269875376,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00044025077299741683,
|
|
"loss": 5.5977,
|
|
"mean_token_accuracy": 0.16877810806035995,
|
|
"num_tokens": 6408435.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.916531753540039,
|
|
"epoch": 3.0326600773528147,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00044002494821494007,
|
|
"loss": 5.5585,
|
|
"mean_token_accuracy": 0.1758269652724266,
|
|
"num_tokens": 6417572.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.956817102432251,
|
|
"epoch": 3.0369574559518693,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00043979876308393635,
|
|
"loss": 5.5645,
|
|
"mean_token_accuracy": 0.17251426726579666,
|
|
"num_tokens": 6426153.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.912615776062012,
|
|
"epoch": 3.041254834550924,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004395722180983036,
|
|
"loss": 5.6201,
|
|
"mean_token_accuracy": 0.1683608129620552,
|
|
"num_tokens": 6435258.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.929599142074585,
|
|
"epoch": 3.0455522131499784,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043934531375272535,
|
|
"loss": 5.6498,
|
|
"mean_token_accuracy": 0.1656801536679268,
|
|
"num_tokens": 6445170.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.905686235427856,
|
|
"epoch": 3.049849591749033,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00043911805054267015,
|
|
"loss": 5.592,
|
|
"mean_token_accuracy": 0.1732991725206375,
|
|
"num_tokens": 6455118.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.942510938644409,
|
|
"epoch": 3.0541469703480875,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00043889042896439004,
|
|
"loss": 5.5831,
|
|
"mean_token_accuracy": 0.1653883323073387,
|
|
"num_tokens": 6464168.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.909526205062866,
|
|
"epoch": 3.0584443489471425,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043866244951491946,
|
|
"loss": 5.5874,
|
|
"mean_token_accuracy": 0.17523137331008912,
|
|
"num_tokens": 6473079.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.939188241958618,
|
|
"epoch": 3.062741727546197,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00043843411269207445,
|
|
"loss": 5.607,
|
|
"mean_token_accuracy": 0.16826117187738418,
|
|
"num_tokens": 6482601.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.8631750583648685,
|
|
"epoch": 3.0670391061452515,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004382054189944514,
|
|
"loss": 5.4282,
|
|
"mean_token_accuracy": 0.18423998802900315,
|
|
"num_tokens": 6491089.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.787638378143311,
|
|
"epoch": 3.071336484744306,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004379763689214259,
|
|
"loss": 5.4461,
|
|
"mean_token_accuracy": 0.18675337582826615,
|
|
"num_tokens": 6499484.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.871266174316406,
|
|
"epoch": 3.0756338633433606,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004377469629731518,
|
|
"loss": 5.5753,
|
|
"mean_token_accuracy": 0.17729860842227935,
|
|
"num_tokens": 6508529.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.909856748580933,
|
|
"epoch": 3.079931241942415,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004375172016505599,
|
|
"loss": 5.6064,
|
|
"mean_token_accuracy": 0.1750326007604599,
|
|
"num_tokens": 6517966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.917290830612183,
|
|
"epoch": 3.0842286205414697,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004372870854553572,
|
|
"loss": 5.5829,
|
|
"mean_token_accuracy": 0.17315875887870788,
|
|
"num_tokens": 6527397.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.824261617660523,
|
|
"epoch": 3.0885259991405243,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004370566148900255,
|
|
"loss": 5.4634,
|
|
"mean_token_accuracy": 0.18077112585306168,
|
|
"num_tokens": 6535452.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.864097642898559,
|
|
"epoch": 3.092823377739579,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043682579045782024,
|
|
"loss": 5.6187,
|
|
"mean_token_accuracy": 0.17499222755432128,
|
|
"num_tokens": 6544591.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.793846464157104,
|
|
"epoch": 3.0971207563386334,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004365946126627699,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.1851562961935997,
|
|
"num_tokens": 6553200.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.854694652557373,
|
|
"epoch": 3.101418134937688,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043636308200967433,
|
|
"loss": 5.5159,
|
|
"mean_token_accuracy": 0.18134936541318894,
|
|
"num_tokens": 6562632.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.880624532699585,
|
|
"epoch": 3.1057155135367425,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004361311990041039,
|
|
"loss": 5.5495,
|
|
"mean_token_accuracy": 0.1787429004907608,
|
|
"num_tokens": 6571898.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.840307092666626,
|
|
"epoch": 3.110012892135797,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00043589896415239843,
|
|
"loss": 5.5639,
|
|
"mean_token_accuracy": 0.18298432976007462,
|
|
"num_tokens": 6581347.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.86344051361084,
|
|
"epoch": 3.1143102707348516,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00043566637796166595,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.18044697493314743,
|
|
"num_tokens": 6590148.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.85330810546875,
|
|
"epoch": 3.118607649333906,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 0.00043543344093978186,
|
|
"loss": 5.5653,
|
|
"mean_token_accuracy": 0.1742002323269844,
|
|
"num_tokens": 6598954.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.884604120254517,
|
|
"epoch": 3.122905027932961,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043520015359538745,
|
|
"loss": 5.5493,
|
|
"mean_token_accuracy": 0.17791756391525268,
|
|
"num_tokens": 6607787.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.849917030334472,
|
|
"epoch": 3.1272024065320156,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004349665164378891,
|
|
"loss": 5.539,
|
|
"mean_token_accuracy": 0.17442221343517303,
|
|
"num_tokens": 6617143.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.816744327545166,
|
|
"epoch": 3.13149978513107,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00043473252997745684,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.17419631034135818,
|
|
"num_tokens": 6625849.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.896743869781494,
|
|
"epoch": 3.1357971637301247,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043449819472502366,
|
|
"loss": 5.6037,
|
|
"mean_token_accuracy": 0.1719396159052849,
|
|
"num_tokens": 6634413.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.848897647857666,
|
|
"epoch": 3.1400945423291793,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004342635111922841,
|
|
"loss": 5.4441,
|
|
"mean_token_accuracy": 0.18833121359348298,
|
|
"num_tokens": 6643785.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.860336828231811,
|
|
"epoch": 3.144391920928234,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004340284798916931,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.18128058463335037,
|
|
"num_tokens": 6652802.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.971281385421753,
|
|
"epoch": 3.1486892995272884,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004337931013364653,
|
|
"loss": 5.6049,
|
|
"mean_token_accuracy": 0.17096141576766968,
|
|
"num_tokens": 6662844.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.851257371902466,
|
|
"epoch": 3.152986678126343,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000433557376040573,
|
|
"loss": 5.508,
|
|
"mean_token_accuracy": 0.17657318860292434,
|
|
"num_tokens": 6671563.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.776889896392822,
|
|
"epoch": 3.1572840567253975,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00043332130451874645,
|
|
"loss": 5.5256,
|
|
"mean_token_accuracy": 0.1795828402042389,
|
|
"num_tokens": 6680637.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.957527208328247,
|
|
"epoch": 3.161581435324452,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00043308488728647127,
|
|
"loss": 5.5655,
|
|
"mean_token_accuracy": 0.1760145381093025,
|
|
"num_tokens": 6688892.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.803671026229859,
|
|
"epoch": 3.1658788139235066,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004328481248599882,
|
|
"loss": 5.5947,
|
|
"mean_token_accuracy": 0.17668567150831221,
|
|
"num_tokens": 6698376.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.888890504837036,
|
|
"epoch": 3.170176192522561,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004326110177562918,
|
|
"loss": 5.6182,
|
|
"mean_token_accuracy": 0.17022019177675246,
|
|
"num_tokens": 6708279.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.837576055526734,
|
|
"epoch": 3.1744735711216157,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043237356649312926,
|
|
"loss": 5.4624,
|
|
"mean_token_accuracy": 0.18940459191799164,
|
|
"num_tokens": 6716648.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.802555084228516,
|
|
"epoch": 3.17877094972067,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004321357715889991,
|
|
"loss": 5.5063,
|
|
"mean_token_accuracy": 0.180134916305542,
|
|
"num_tokens": 6725134.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.87052845954895,
|
|
"epoch": 3.1830683283197247,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004318976335631505,
|
|
"loss": 5.5555,
|
|
"mean_token_accuracy": 0.17391526848077773,
|
|
"num_tokens": 6734404.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.896699523925781,
|
|
"epoch": 3.1873657069187797,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00043165915293558155,
|
|
"loss": 5.5749,
|
|
"mean_token_accuracy": 0.17638049721717836,
|
|
"num_tokens": 6742805.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.881629133224488,
|
|
"epoch": 3.1916630855178343,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004314203302270388,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.1817157968878746,
|
|
"num_tokens": 6752178.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.823851680755615,
|
|
"epoch": 3.195960464116889,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004311811659590154,
|
|
"loss": 5.5629,
|
|
"mean_token_accuracy": 0.17258817553520203,
|
|
"num_tokens": 6762018.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.779849195480347,
|
|
"epoch": 3.2002578427159434,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004309416606537507,
|
|
"loss": 5.4537,
|
|
"mean_token_accuracy": 0.1850757271051407,
|
|
"num_tokens": 6770110.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.7685325145721436,
|
|
"epoch": 3.204555221314998,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00043070181483422843,
|
|
"loss": 5.5234,
|
|
"mean_token_accuracy": 0.18397075831890106,
|
|
"num_tokens": 6778364.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.797230243682861,
|
|
"epoch": 3.2088525999140525,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000430461629024176,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.18117326945066453,
|
|
"num_tokens": 6786860.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.934008836746216,
|
|
"epoch": 3.213149978513107,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004302211037480634,
|
|
"loss": 5.6041,
|
|
"mean_token_accuracy": 0.172060327231884,
|
|
"num_tokens": 6795983.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.835339641571045,
|
|
"epoch": 3.2174473571121616,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004299802395311015,
|
|
"loss": 5.5304,
|
|
"mean_token_accuracy": 0.18076492846012115,
|
|
"num_tokens": 6804702.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.906740188598633,
|
|
"epoch": 3.221744735711216,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.0004297390368992414,
|
|
"loss": 5.5768,
|
|
"mean_token_accuracy": 0.1725253313779831,
|
|
"num_tokens": 6813361.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.802566814422607,
|
|
"epoch": 3.2260421143102707,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00042949749637917353,
|
|
"loss": 5.5771,
|
|
"mean_token_accuracy": 0.17496639490127563,
|
|
"num_tokens": 6821937.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.801820421218872,
|
|
"epoch": 3.230339492909325,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 0.0004292556184983256,
|
|
"loss": 5.6218,
|
|
"mean_token_accuracy": 0.16864633709192275,
|
|
"num_tokens": 6830930.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.908778047561645,
|
|
"epoch": 3.2346368715083798,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004290134037848623,
|
|
"loss": 5.5633,
|
|
"mean_token_accuracy": 0.17439839690923692,
|
|
"num_tokens": 6839815.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.8427082061767575,
|
|
"epoch": 3.2389342501074343,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00042877085276768386,
|
|
"loss": 5.6024,
|
|
"mean_token_accuracy": 0.17073359787464143,
|
|
"num_tokens": 6849453.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.799373722076416,
|
|
"epoch": 3.243231628706489,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00042852796597642455,
|
|
"loss": 5.4733,
|
|
"mean_token_accuracy": 0.1789889931678772,
|
|
"num_tokens": 6858693.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.860775279998779,
|
|
"epoch": 3.247529007305544,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004282847439414522,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.1849004626274109,
|
|
"num_tokens": 6868476.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.844316816329956,
|
|
"epoch": 3.2518263859045984,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004280411871938664,
|
|
"loss": 5.5109,
|
|
"mean_token_accuracy": 0.18599154204130172,
|
|
"num_tokens": 6877471.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.800557708740234,
|
|
"epoch": 3.256123764503653,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004277972962654979,
|
|
"loss": 5.5296,
|
|
"mean_token_accuracy": 0.1853979155421257,
|
|
"num_tokens": 6886453.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.8934258937835695,
|
|
"epoch": 3.2604211431027075,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004275530716889069,
|
|
"loss": 5.5961,
|
|
"mean_token_accuracy": 0.17265416085720062,
|
|
"num_tokens": 6895638.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.857405805587769,
|
|
"epoch": 3.264718521701762,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004273085139973822,
|
|
"loss": 5.5978,
|
|
"mean_token_accuracy": 0.1795324355363846,
|
|
"num_tokens": 6905127.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.888291549682617,
|
|
"epoch": 3.2690159003008166,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004270636237249401,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.1745821550488472,
|
|
"num_tokens": 6914316.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.855546331405639,
|
|
"epoch": 3.273313278899871,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042681840140632314,
|
|
"loss": 5.647,
|
|
"mean_token_accuracy": 0.17424414306879044,
|
|
"num_tokens": 6924447.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.807538747787476,
|
|
"epoch": 3.2776106574989257,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004265728475769989,
|
|
"loss": 5.5985,
|
|
"mean_token_accuracy": 0.17095187455415725,
|
|
"num_tokens": 6933279.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.846661043167114,
|
|
"epoch": 3.28190803609798,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004263269627731586,
|
|
"loss": 5.607,
|
|
"mean_token_accuracy": 0.17249199748039246,
|
|
"num_tokens": 6941810.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.8836431980133055,
|
|
"epoch": 3.2862054146970348,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004260807475317164,
|
|
"loss": 5.6031,
|
|
"mean_token_accuracy": 0.17557015866041184,
|
|
"num_tokens": 6951507.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.855884456634522,
|
|
"epoch": 3.2905027932960893,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004258342023903081,
|
|
"loss": 5.5966,
|
|
"mean_token_accuracy": 0.1747473195195198,
|
|
"num_tokens": 6961047.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.8552453994750975,
|
|
"epoch": 3.294800171895144,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00042558732788728975,
|
|
"loss": 5.5528,
|
|
"mean_token_accuracy": 0.17688599079847336,
|
|
"num_tokens": 6969837.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.85599684715271,
|
|
"epoch": 3.2990975504941984,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00042534012456173643,
|
|
"loss": 5.6531,
|
|
"mean_token_accuracy": 0.1686916783452034,
|
|
"num_tokens": 6978581.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.748218631744384,
|
|
"epoch": 3.303394929093253,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00042509259295344157,
|
|
"loss": 5.4981,
|
|
"mean_token_accuracy": 0.18584417998790742,
|
|
"num_tokens": 6988929.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.842048501968383,
|
|
"epoch": 3.3076923076923075,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00042484473360291514,
|
|
"loss": 5.5342,
|
|
"mean_token_accuracy": 0.1767965465784073,
|
|
"num_tokens": 6997729.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.8437494277954105,
|
|
"epoch": 3.311989686291362,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00042459654705138294,
|
|
"loss": 5.5965,
|
|
"mean_token_accuracy": 0.1742642492055893,
|
|
"num_tokens": 7007233.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.8709276676177975,
|
|
"epoch": 3.316287064890417,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004243480338407853,
|
|
"loss": 5.4606,
|
|
"mean_token_accuracy": 0.18200965374708175,
|
|
"num_tokens": 7015279.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.831651830673218,
|
|
"epoch": 3.3205844434894716,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004240991945137755,
|
|
"loss": 5.6536,
|
|
"mean_token_accuracy": 0.16545216739177704,
|
|
"num_tokens": 7024579.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.826376104354859,
|
|
"epoch": 3.324881822088526,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00042385002961371944,
|
|
"loss": 5.5665,
|
|
"mean_token_accuracy": 0.18016621917486192,
|
|
"num_tokens": 7033561.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.8217737674713135,
|
|
"epoch": 3.3291792006875807,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004236005396846935,
|
|
"loss": 5.5435,
|
|
"mean_token_accuracy": 0.1759042486548424,
|
|
"num_tokens": 7042695.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.8712411403656,
|
|
"epoch": 3.333476579286635,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00042335072527148406,
|
|
"loss": 5.5875,
|
|
"mean_token_accuracy": 0.16748757809400558,
|
|
"num_tokens": 7051166.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.8449126243591305,
|
|
"epoch": 3.3377739578856898,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004231005869195859,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.18884171396493912,
|
|
"num_tokens": 7059916.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.83662166595459,
|
|
"epoch": 3.3420713364847443,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004228501251752011,
|
|
"loss": 5.6415,
|
|
"mean_token_accuracy": 0.1701444074511528,
|
|
"num_tokens": 7068813.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.830411911010742,
|
|
"epoch": 3.346368715083799,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00042259934058523814,
|
|
"loss": 5.585,
|
|
"mean_token_accuracy": 0.17672917544841765,
|
|
"num_tokens": 7078378.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.799500131607056,
|
|
"epoch": 3.3506660936828534,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00042234823369731027,
|
|
"loss": 5.6336,
|
|
"mean_token_accuracy": 0.17052182853221892,
|
|
"num_tokens": 7088099.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.9160895347595215,
|
|
"epoch": 3.354963472281908,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00042209680505973465,
|
|
"loss": 5.6218,
|
|
"mean_token_accuracy": 0.1696390464901924,
|
|
"num_tokens": 7096911.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.792940998077393,
|
|
"epoch": 3.3592608508809625,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004218450552215308,
|
|
"loss": 5.4391,
|
|
"mean_token_accuracy": 0.18800032138824463,
|
|
"num_tokens": 7106370.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.834222412109375,
|
|
"epoch": 3.363558229480017,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004215929847324199,
|
|
"loss": 5.562,
|
|
"mean_token_accuracy": 0.17508285641670226,
|
|
"num_tokens": 7115388.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.805608797073364,
|
|
"epoch": 3.3678556080790716,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000421340594142823,
|
|
"loss": 5.5205,
|
|
"mean_token_accuracy": 0.1819288268685341,
|
|
"num_tokens": 7124015.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.830291509628296,
|
|
"epoch": 3.3721529866781266,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00042108788400386035,
|
|
"loss": 5.6372,
|
|
"mean_token_accuracy": 0.17691589295864105,
|
|
"num_tokens": 7132774.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.889302587509155,
|
|
"epoch": 3.376450365277181,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004208348548673498,
|
|
"loss": 5.6014,
|
|
"mean_token_accuracy": 0.17239128798246384,
|
|
"num_tokens": 7141806.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.933525419235229,
|
|
"epoch": 3.3807477438762357,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000420581507285806,
|
|
"loss": 5.5606,
|
|
"mean_token_accuracy": 0.17666767090559005,
|
|
"num_tokens": 7150570.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.788972234725952,
|
|
"epoch": 3.38504512247529,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004203278418124386,
|
|
"loss": 5.6721,
|
|
"mean_token_accuracy": 0.1664616048336029,
|
|
"num_tokens": 7159811.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.817618274688721,
|
|
"epoch": 3.3893425010743448,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004200738590011518,
|
|
"loss": 5.5039,
|
|
"mean_token_accuracy": 0.17234873920679092,
|
|
"num_tokens": 7169034.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.897416114807129,
|
|
"epoch": 3.3936398796733993,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00041981955940654245,
|
|
"loss": 5.616,
|
|
"mean_token_accuracy": 0.16870101690292358,
|
|
"num_tokens": 7177384.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.768745803833008,
|
|
"epoch": 3.397937258272454,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004195649435838992,
|
|
"loss": 5.6187,
|
|
"mean_token_accuracy": 0.17308720350265502,
|
|
"num_tokens": 7185774.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.842696857452393,
|
|
"epoch": 3.4022346368715084,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004193100120892013,
|
|
"loss": 5.5944,
|
|
"mean_token_accuracy": 0.17756400406360626,
|
|
"num_tokens": 7195677.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.871747159957886,
|
|
"epoch": 3.406532015470563,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004190547654791172,
|
|
"loss": 5.6139,
|
|
"mean_token_accuracy": 0.17074085175991058,
|
|
"num_tokens": 7205285.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.842202949523926,
|
|
"epoch": 3.4108293940696175,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00041879920431100347,
|
|
"loss": 5.4791,
|
|
"mean_token_accuracy": 0.1830856204032898,
|
|
"num_tokens": 7213550.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.802624797821045,
|
|
"epoch": 3.415126772668672,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004185433291429036,
|
|
"loss": 5.6402,
|
|
"mean_token_accuracy": 0.18000059574842453,
|
|
"num_tokens": 7222799.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.8716301918029785,
|
|
"epoch": 3.4194241512677266,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041828714053354665,
|
|
"loss": 5.5612,
|
|
"mean_token_accuracy": 0.1815226063132286,
|
|
"num_tokens": 7232041.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.892048406600952,
|
|
"epoch": 3.423721529866781,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004180306390423462,
|
|
"loss": 5.6296,
|
|
"mean_token_accuracy": 0.16942311376333236,
|
|
"num_tokens": 7241236.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.915034246444702,
|
|
"epoch": 3.4280189084658357,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00041777382522939884,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.17629202008247374,
|
|
"num_tokens": 7250108.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.794379329681396,
|
|
"epoch": 3.4323162870648902,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00041751669965548344,
|
|
"loss": 5.5433,
|
|
"mean_token_accuracy": 0.17530697584152222,
|
|
"num_tokens": 7258760.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.797320556640625,
|
|
"epoch": 3.4366136656639448,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00041725926288205945,
|
|
"loss": 5.6492,
|
|
"mean_token_accuracy": 0.17336636781692505,
|
|
"num_tokens": 7267337.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.4366136656639448,
|
|
"eval_entropy": 5.680117178607631,
|
|
"eval_loss": 6.116849422454834,
|
|
"eval_mean_token_accuracy": 0.1617838227869691,
|
|
"eval_num_tokens": 7267337.0,
|
|
"eval_runtime": 2.8249,
|
|
"eval_samples_per_second": 1256.325,
|
|
"eval_steps_per_second": 157.173,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.789460706710815,
|
|
"epoch": 3.4409110442629998,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004170015154712658,
|
|
"loss": 5.5477,
|
|
"mean_token_accuracy": 0.18614174127578736,
|
|
"num_tokens": 7276640.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.862052392959595,
|
|
"epoch": 3.4452084228620543,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00041674345798591993,
|
|
"loss": 5.6024,
|
|
"mean_token_accuracy": 0.17119915783405304,
|
|
"num_tokens": 7285538.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.805997037887574,
|
|
"epoch": 3.449505801461109,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004164850909895161,
|
|
"loss": 5.5496,
|
|
"mean_token_accuracy": 0.1798061028122902,
|
|
"num_tokens": 7294596.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.793576383590699,
|
|
"epoch": 3.4538031800601634,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004162264150462247,
|
|
"loss": 5.6236,
|
|
"mean_token_accuracy": 0.16988409012556077,
|
|
"num_tokens": 7304463.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.825239562988282,
|
|
"epoch": 3.458100558659218,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00041596743072089065,
|
|
"loss": 5.5327,
|
|
"mean_token_accuracy": 0.18108410388231277,
|
|
"num_tokens": 7314372.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.844225788116455,
|
|
"epoch": 3.4623979372582725,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000415708138579032,
|
|
"loss": 5.5355,
|
|
"mean_token_accuracy": 0.1855032667517662,
|
|
"num_tokens": 7322606.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.842318964004517,
|
|
"epoch": 3.466695315857327,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00041544853918683923,
|
|
"loss": 5.6399,
|
|
"mean_token_accuracy": 0.17741980254650117,
|
|
"num_tokens": 7331798.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.79937162399292,
|
|
"epoch": 3.4709926944563816,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004151886331111737,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.1791357234120369,
|
|
"num_tokens": 7341182.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.889120292663574,
|
|
"epoch": 3.475290073055436,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00041492842091956646,
|
|
"loss": 5.5002,
|
|
"mean_token_accuracy": 0.1823188081383705,
|
|
"num_tokens": 7350419.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.768772029876709,
|
|
"epoch": 3.4795874516544907,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004146679031802167,
|
|
"loss": 5.5894,
|
|
"mean_token_accuracy": 0.1724338635802269,
|
|
"num_tokens": 7359101.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.823584461212159,
|
|
"epoch": 3.4838848302535452,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00041440708046199123,
|
|
"loss": 5.5929,
|
|
"mean_token_accuracy": 0.17613361775875092,
|
|
"num_tokens": 7368376.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.836838912963867,
|
|
"epoch": 3.4881822088525998,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004141459533344226,
|
|
"loss": 5.5702,
|
|
"mean_token_accuracy": 0.18107785135507584,
|
|
"num_tokens": 7377991.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.793580818176269,
|
|
"epoch": 3.4924795874516543,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041388452236770795,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.17755429595708846,
|
|
"num_tokens": 7387712.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.868026351928711,
|
|
"epoch": 3.4967769660507093,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00041362278813270823,
|
|
"loss": 5.4719,
|
|
"mean_token_accuracy": 0.18063639253377914,
|
|
"num_tokens": 7396151.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.877557373046875,
|
|
"epoch": 3.501074344649764,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00041336075120094616,
|
|
"loss": 5.6213,
|
|
"mean_token_accuracy": 0.16375590413808822,
|
|
"num_tokens": 7405090.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.826443767547607,
|
|
"epoch": 3.5053717232488184,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00041309841214460586,
|
|
"loss": 5.5514,
|
|
"mean_token_accuracy": 0.17841600775718688,
|
|
"num_tokens": 7413243.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.788078308105469,
|
|
"epoch": 3.509669101847873,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004128357715365309,
|
|
"loss": 5.5499,
|
|
"mean_token_accuracy": 0.17876799106597902,
|
|
"num_tokens": 7421382.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.85180253982544,
|
|
"epoch": 3.5139664804469275,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00041257282995022345,
|
|
"loss": 5.6733,
|
|
"mean_token_accuracy": 0.1687352865934372,
|
|
"num_tokens": 7431401.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.762858581542969,
|
|
"epoch": 3.518263859045982,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004123095879598426,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.17950685918331147,
|
|
"num_tokens": 7441092.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.724184989929199,
|
|
"epoch": 3.5225612376450366,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00041204604614020397,
|
|
"loss": 5.5935,
|
|
"mean_token_accuracy": 0.17831090837717056,
|
|
"num_tokens": 7450558.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.885694789886474,
|
|
"epoch": 3.526858616244091,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004117822050667773,
|
|
"loss": 5.6019,
|
|
"mean_token_accuracy": 0.17594044655561447,
|
|
"num_tokens": 7459622.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.8979674816131595,
|
|
"epoch": 3.5311559948431457,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00041151806531568617,
|
|
"loss": 5.5557,
|
|
"mean_token_accuracy": 0.17839221507310868,
|
|
"num_tokens": 7468311.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.774103832244873,
|
|
"epoch": 3.5354533734422002,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00041125362746370625,
|
|
"loss": 5.5858,
|
|
"mean_token_accuracy": 0.1679152801632881,
|
|
"num_tokens": 7477049.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.788746166229248,
|
|
"epoch": 3.5397507520412548,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004109888920882639,
|
|
"loss": 5.5841,
|
|
"mean_token_accuracy": 0.17650095671415328,
|
|
"num_tokens": 7486332.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.872855854034424,
|
|
"epoch": 3.5440481306403093,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004107238597674356,
|
|
"loss": 5.5463,
|
|
"mean_token_accuracy": 0.17802504152059556,
|
|
"num_tokens": 7494713.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.825634288787842,
|
|
"epoch": 3.548345509239364,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000410458531079946,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.17894839495420456,
|
|
"num_tokens": 7504359.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.77186336517334,
|
|
"epoch": 3.5526428878384184,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004101929066051668,
|
|
"loss": 5.5394,
|
|
"mean_token_accuracy": 0.18013097047805787,
|
|
"num_tokens": 7512792.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.746202230453491,
|
|
"epoch": 3.556940266437473,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004099269869231157,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.17920746207237243,
|
|
"num_tokens": 7522617.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.836572551727295,
|
|
"epoch": 3.5612376450365275,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00040966077261445495,
|
|
"loss": 5.5722,
|
|
"mean_token_accuracy": 0.17907663732767104,
|
|
"num_tokens": 7531123.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.801541805267334,
|
|
"epoch": 3.565535023635582,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004093942642604904,
|
|
"loss": 5.6537,
|
|
"mean_token_accuracy": 0.17167993485927582,
|
|
"num_tokens": 7541521.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.796760225296021,
|
|
"epoch": 3.5698324022346366,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00040912746244316944,
|
|
"loss": 5.5076,
|
|
"mean_token_accuracy": 0.18444719016551972,
|
|
"num_tokens": 7550343.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.842436408996582,
|
|
"epoch": 3.5741297808336916,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00040886036774508095,
|
|
"loss": 5.7609,
|
|
"mean_token_accuracy": 0.15603764057159425,
|
|
"num_tokens": 7560399.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.8594526767730715,
|
|
"epoch": 3.578427159432746,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004085929807494527,
|
|
"loss": 5.5231,
|
|
"mean_token_accuracy": 0.17736161053180693,
|
|
"num_tokens": 7569982.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.688966608047485,
|
|
"epoch": 3.5827245380318007,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004083253020401512,
|
|
"loss": 5.4461,
|
|
"mean_token_accuracy": 0.18296868354082108,
|
|
"num_tokens": 7578954.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.804179763793945,
|
|
"epoch": 3.5870219166308552,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004080573322016797,
|
|
"loss": 5.5521,
|
|
"mean_token_accuracy": 0.1864866405725479,
|
|
"num_tokens": 7587424.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.716727209091187,
|
|
"epoch": 3.59131929522991,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004077890718191773,
|
|
"loss": 5.4522,
|
|
"mean_token_accuracy": 0.18167345225811005,
|
|
"num_tokens": 7596900.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.8346515655517575,
|
|
"epoch": 3.5956166738289643,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00040752052147841733,
|
|
"loss": 5.6654,
|
|
"mean_token_accuracy": 0.17199787348508835,
|
|
"num_tokens": 7606774.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.819213581085205,
|
|
"epoch": 3.599914052428019,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004072516817658065,
|
|
"loss": 5.5836,
|
|
"mean_token_accuracy": 0.1770189255475998,
|
|
"num_tokens": 7615691.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.860805511474609,
|
|
"epoch": 3.6042114310270734,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004069825532683831,
|
|
"loss": 5.5657,
|
|
"mean_token_accuracy": 0.18359059542417527,
|
|
"num_tokens": 7625294.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.803269147872925,
|
|
"epoch": 3.608508809626128,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00040671313657381645,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.17565432339906692,
|
|
"num_tokens": 7634563.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.808817434310913,
|
|
"epoch": 3.6128061882251825,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00040644343227040473,
|
|
"loss": 5.6077,
|
|
"mean_token_accuracy": 0.17300346344709397,
|
|
"num_tokens": 7643860.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.8044181823730465,
|
|
"epoch": 3.617103566824237,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004061734409470745,
|
|
"loss": 5.5844,
|
|
"mean_token_accuracy": 0.17170806974172592,
|
|
"num_tokens": 7654142.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.825724792480469,
|
|
"epoch": 3.621400945423292,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004059031631933788,
|
|
"loss": 5.5765,
|
|
"mean_token_accuracy": 0.18030764013528824,
|
|
"num_tokens": 7662910.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.800470018386841,
|
|
"epoch": 3.6256983240223466,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00040563259959949615,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.18108874857425689,
|
|
"num_tokens": 7670764.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.901778936386108,
|
|
"epoch": 3.629995702621401,
|
|
"grad_norm": 3.1875,
|
|
"learning_rate": 0.0004053617507562295,
|
|
"loss": 5.6179,
|
|
"mean_token_accuracy": 0.16800165474414824,
|
|
"num_tokens": 7679422.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.805789852142334,
|
|
"epoch": 3.6342930812204557,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00040509061725500426,
|
|
"loss": 5.5436,
|
|
"mean_token_accuracy": 0.17577001452445984,
|
|
"num_tokens": 7687992.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.822667741775513,
|
|
"epoch": 3.6385904598195102,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004048191996878677,
|
|
"loss": 5.6183,
|
|
"mean_token_accuracy": 0.17157122641801834,
|
|
"num_tokens": 7696607.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.914057874679566,
|
|
"epoch": 3.642887838418565,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040454749864748734,
|
|
"loss": 5.6509,
|
|
"mean_token_accuracy": 0.17557104229927062,
|
|
"num_tokens": 7705280.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.846065902709961,
|
|
"epoch": 3.6471852170176193,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004042755147271496,
|
|
"loss": 5.5181,
|
|
"mean_token_accuracy": 0.1821632355451584,
|
|
"num_tokens": 7714120.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.7978815078735355,
|
|
"epoch": 3.651482595616674,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004040032485207587,
|
|
"loss": 5.5749,
|
|
"mean_token_accuracy": 0.17602317035198212,
|
|
"num_tokens": 7724627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.791724586486817,
|
|
"epoch": 3.6557799742157284,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004037307006228352,
|
|
"loss": 5.6219,
|
|
"mean_token_accuracy": 0.1713547483086586,
|
|
"num_tokens": 7734650.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.819452476501465,
|
|
"epoch": 3.660077352814783,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004034578716285147,
|
|
"loss": 5.4856,
|
|
"mean_token_accuracy": 0.1901700899004936,
|
|
"num_tokens": 7744357.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.849957990646362,
|
|
"epoch": 3.6643747314138375,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004031847621335467,
|
|
"loss": 5.5276,
|
|
"mean_token_accuracy": 0.17219772338867187,
|
|
"num_tokens": 7753207.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.857378816604614,
|
|
"epoch": 3.668672110012892,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004029113727342933,
|
|
"loss": 5.6388,
|
|
"mean_token_accuracy": 0.17394320219755172,
|
|
"num_tokens": 7762750.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.8012917041778564,
|
|
"epoch": 3.6729694886119466,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00040263770402772746,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.18198629468679428,
|
|
"num_tokens": 7771539.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.843603467941284,
|
|
"epoch": 3.677266867211001,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004023637566114325,
|
|
"loss": 5.6619,
|
|
"mean_token_accuracy": 0.172455270588398,
|
|
"num_tokens": 7781447.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.857791805267334,
|
|
"epoch": 3.6815642458100557,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004020895310835999,
|
|
"loss": 5.5406,
|
|
"mean_token_accuracy": 0.18245439678430558,
|
|
"num_tokens": 7789977.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.864632606506348,
|
|
"epoch": 3.6858616244091102,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00040181502804302865,
|
|
"loss": 5.6803,
|
|
"mean_token_accuracy": 0.1658618316054344,
|
|
"num_tokens": 7800000.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.849373197555542,
|
|
"epoch": 3.690159003008165,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00040154024808912377,
|
|
"loss": 5.6873,
|
|
"mean_token_accuracy": 0.16825657933950425,
|
|
"num_tokens": 7809062.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.805650234222412,
|
|
"epoch": 3.6944563816072193,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0004012651918218947,
|
|
"loss": 5.4895,
|
|
"mean_token_accuracy": 0.185204353928566,
|
|
"num_tokens": 7817136.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.825238275527954,
|
|
"epoch": 3.6987537602062743,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.0004009898598419544,
|
|
"loss": 5.5941,
|
|
"mean_token_accuracy": 0.17511084377765657,
|
|
"num_tokens": 7825278.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.778262996673584,
|
|
"epoch": 3.703051138805329,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000400714252750518,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.17554006427526475,
|
|
"num_tokens": 7833448.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.7640807151794435,
|
|
"epoch": 3.7073485174043834,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0004004383711494011,
|
|
"loss": 5.5261,
|
|
"mean_token_accuracy": 0.1812538981437683,
|
|
"num_tokens": 7841919.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.869648504257202,
|
|
"epoch": 3.711645896003438,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0004001622156410189,
|
|
"loss": 5.7183,
|
|
"mean_token_accuracy": 0.17098257690668106,
|
|
"num_tokens": 7851584.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.8714433193206785,
|
|
"epoch": 3.7159432746024925,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00039988578682838467,
|
|
"loss": 5.5752,
|
|
"mean_token_accuracy": 0.1817497968673706,
|
|
"num_tokens": 7860329.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.79450216293335,
|
|
"epoch": 3.720240653201547,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00039960908531510843,
|
|
"loss": 5.6851,
|
|
"mean_token_accuracy": 0.174725678563118,
|
|
"num_tokens": 7870393.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.82914457321167,
|
|
"epoch": 3.7245380318006016,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003993321117053956,
|
|
"loss": 5.5946,
|
|
"mean_token_accuracy": 0.17997961342334748,
|
|
"num_tokens": 7878776.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.953162097930909,
|
|
"epoch": 3.728835410399656,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00039905486660404604,
|
|
"loss": 5.6787,
|
|
"mean_token_accuracy": 0.16963492333889008,
|
|
"num_tokens": 7888232.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.871613168716431,
|
|
"epoch": 3.7331327889987107,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00039877735061645206,
|
|
"loss": 5.6501,
|
|
"mean_token_accuracy": 0.1689319372177124,
|
|
"num_tokens": 7898951.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.852160978317261,
|
|
"epoch": 3.7374301675977653,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003984995643485977,
|
|
"loss": 5.6948,
|
|
"mean_token_accuracy": 0.17225231677293779,
|
|
"num_tokens": 7908162.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.773387813568116,
|
|
"epoch": 3.74172754619682,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039822150840705716,
|
|
"loss": 5.5639,
|
|
"mean_token_accuracy": 0.18139102011919023,
|
|
"num_tokens": 7916953.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.836595630645752,
|
|
"epoch": 3.746024924795875,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00039794318339899347,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.17648792564868926,
|
|
"num_tokens": 7925622.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.8513264656066895,
|
|
"epoch": 3.7503223033949293,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00039766458993215726,
|
|
"loss": 5.6085,
|
|
"mean_token_accuracy": 0.1730761244893074,
|
|
"num_tokens": 7934807.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.84073920249939,
|
|
"epoch": 3.754619681993984,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00039738572861488527,
|
|
"loss": 5.6474,
|
|
"mean_token_accuracy": 0.16410866752266884,
|
|
"num_tokens": 7945864.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.764331912994384,
|
|
"epoch": 3.7589170605930384,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000397106600056099,
|
|
"loss": 5.4787,
|
|
"mean_token_accuracy": 0.1886591359972954,
|
|
"num_tokens": 7954571.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.781460809707641,
|
|
"epoch": 3.763214439192093,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003968272048653039,
|
|
"loss": 5.4881,
|
|
"mean_token_accuracy": 0.18367257565259934,
|
|
"num_tokens": 7963092.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.797197008132935,
|
|
"epoch": 3.7675118177911475,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003965475436525873,
|
|
"loss": 5.5939,
|
|
"mean_token_accuracy": 0.17113098949193956,
|
|
"num_tokens": 7972509.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.7959370613098145,
|
|
"epoch": 3.771809196390202,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003962676170286174,
|
|
"loss": 5.5298,
|
|
"mean_token_accuracy": 0.18198893517255782,
|
|
"num_tokens": 7981494.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.757343387603759,
|
|
"epoch": 3.7761065749892566,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039598742560464223,
|
|
"loss": 5.5462,
|
|
"mean_token_accuracy": 0.17990134805440902,
|
|
"num_tokens": 7991788.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.73751540184021,
|
|
"epoch": 3.780403953588311,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0003957069699924877,
|
|
"loss": 5.4827,
|
|
"mean_token_accuracy": 0.1839318424463272,
|
|
"num_tokens": 8000528.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.824782514572144,
|
|
"epoch": 3.7847013321873657,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000395426250804557,
|
|
"loss": 5.5656,
|
|
"mean_token_accuracy": 0.18069888204336165,
|
|
"num_tokens": 8008729.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.824069023132324,
|
|
"epoch": 3.7889987107864203,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039514526865382847,
|
|
"loss": 5.6502,
|
|
"mean_token_accuracy": 0.16986860185861588,
|
|
"num_tokens": 8017391.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.839191484451294,
|
|
"epoch": 3.793296089385475,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0003948640241538548,
|
|
"loss": 5.5914,
|
|
"mean_token_accuracy": 0.17269910275936126,
|
|
"num_tokens": 8026772.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.898277664184571,
|
|
"epoch": 3.7975934679845293,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0003945825179187617,
|
|
"loss": 5.6773,
|
|
"mean_token_accuracy": 0.16648378521203994,
|
|
"num_tokens": 8035231.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.792491245269775,
|
|
"epoch": 3.801890846583584,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00039430075056324604,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.1733570098876953,
|
|
"num_tokens": 8044097.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.777822637557984,
|
|
"epoch": 3.8061882251826384,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00039401872270257546,
|
|
"loss": 5.6008,
|
|
"mean_token_accuracy": 0.17433446943759917,
|
|
"num_tokens": 8053511.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.9238283157348635,
|
|
"epoch": 3.810485603781693,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00039373643495258567,
|
|
"loss": 5.6608,
|
|
"mean_token_accuracy": 0.17477619647979736,
|
|
"num_tokens": 8063237.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.7283531665802006,
|
|
"epoch": 3.8147829823807475,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039345388792968056,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.17828034609556198,
|
|
"num_tokens": 8072098.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.814003896713257,
|
|
"epoch": 3.819080360979802,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039317108225082984,
|
|
"loss": 5.6322,
|
|
"mean_token_accuracy": 0.1731913909316063,
|
|
"num_tokens": 8080789.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.832900285720825,
|
|
"epoch": 3.8233777395788566,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039288801853356806,
|
|
"loss": 5.6241,
|
|
"mean_token_accuracy": 0.17220719009637833,
|
|
"num_tokens": 8091013.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.842153787612915,
|
|
"epoch": 3.8276751181779116,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003926046973959932,
|
|
"loss": 5.5414,
|
|
"mean_token_accuracy": 0.17905485332012178,
|
|
"num_tokens": 8099857.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.781813907623291,
|
|
"epoch": 3.831972496776966,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003923211194567654,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.17909391820430756,
|
|
"num_tokens": 8109268.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.81556658744812,
|
|
"epoch": 3.8362698753760207,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00039203728533510556,
|
|
"loss": 5.6081,
|
|
"mean_token_accuracy": 0.16844937950372696,
|
|
"num_tokens": 8119242.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.885900020599365,
|
|
"epoch": 3.8405672539750753,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000391753195650794,
|
|
"loss": 5.6246,
|
|
"mean_token_accuracy": 0.17332198321819306,
|
|
"num_tokens": 8129031.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.816825246810913,
|
|
"epoch": 3.84486463257413,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00039146885102416895,
|
|
"loss": 5.5852,
|
|
"mean_token_accuracy": 0.18095089495182037,
|
|
"num_tokens": 8137550.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.8004976272583,
|
|
"epoch": 3.8491620111731844,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00039118425207612553,
|
|
"loss": 5.5873,
|
|
"mean_token_accuracy": 0.17801908105611802,
|
|
"num_tokens": 8146410.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.78707914352417,
|
|
"epoch": 3.853459389772239,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00039089939942811396,
|
|
"loss": 5.4724,
|
|
"mean_token_accuracy": 0.17898129671812057,
|
|
"num_tokens": 8155126.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.680876779556274,
|
|
"epoch": 3.8577567683712934,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00039061429370213863,
|
|
"loss": 5.489,
|
|
"mean_token_accuracy": 0.1902177229523659,
|
|
"num_tokens": 8163623.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.719233655929566,
|
|
"epoch": 3.862054146970348,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00039032893552075646,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.18113213330507277,
|
|
"num_tokens": 8172446.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.764508628845215,
|
|
"epoch": 3.8663515255694025,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003900433255070758,
|
|
"loss": 5.6329,
|
|
"mean_token_accuracy": 0.1745102360844612,
|
|
"num_tokens": 8181278.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.8663515255694025,
|
|
"eval_entropy": 5.753025039896235,
|
|
"eval_loss": 6.068426132202148,
|
|
"eval_mean_token_accuracy": 0.1652604920437207,
|
|
"eval_num_tokens": 8181278.0,
|
|
"eval_runtime": 2.6353,
|
|
"eval_samples_per_second": 1346.697,
|
|
"eval_steps_per_second": 168.479,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.872699546813965,
|
|
"epoch": 3.870648904168457,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.00038975746428475454,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.18006572872400284,
|
|
"num_tokens": 8189981.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.824715709686279,
|
|
"epoch": 3.874946282767512,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00038947135247799955,
|
|
"loss": 5.6112,
|
|
"mean_token_accuracy": 0.17716822326183318,
|
|
"num_tokens": 8198674.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.777155494689941,
|
|
"epoch": 3.8792436613665666,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00038918499071156443,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.1758745104074478,
|
|
"num_tokens": 8208012.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.821870946884156,
|
|
"epoch": 3.883541039965621,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000388898379610749,
|
|
"loss": 5.5049,
|
|
"mean_token_accuracy": 0.18129893392324448,
|
|
"num_tokens": 8216811.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.853569602966308,
|
|
"epoch": 3.8878384185646757,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003886115198013973,
|
|
"loss": 5.7064,
|
|
"mean_token_accuracy": 0.16721303313970565,
|
|
"num_tokens": 8226637.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.838116788864136,
|
|
"epoch": 3.8921357971637303,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.0003883244119098965,
|
|
"loss": 5.6082,
|
|
"mean_token_accuracy": 0.17507980316877364,
|
|
"num_tokens": 8235443.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.858975648880005,
|
|
"epoch": 3.896433175762785,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003880370565631754,
|
|
"loss": 5.6013,
|
|
"mean_token_accuracy": 0.1741223856806755,
|
|
"num_tokens": 8243791.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.8384264469146725,
|
|
"epoch": 3.9007305543618394,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00038774945438870337,
|
|
"loss": 5.5771,
|
|
"mean_token_accuracy": 0.16520297676324844,
|
|
"num_tokens": 8252831.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.769410848617554,
|
|
"epoch": 3.905027932960894,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038746160601448845,
|
|
"loss": 5.6043,
|
|
"mean_token_accuracy": 0.1715276539325714,
|
|
"num_tokens": 8262281.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.800005292892456,
|
|
"epoch": 3.9093253115599484,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003871735120690766,
|
|
"loss": 5.4992,
|
|
"mean_token_accuracy": 0.18514232337474823,
|
|
"num_tokens": 8270834.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.800684928894043,
|
|
"epoch": 3.913622690159003,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003868851731815497,
|
|
"loss": 5.5763,
|
|
"mean_token_accuracy": 0.17813455760478974,
|
|
"num_tokens": 8280674.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.734304428100586,
|
|
"epoch": 3.9179200687580575,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003865965899815247,
|
|
"loss": 5.5801,
|
|
"mean_token_accuracy": 0.17343587577342987,
|
|
"num_tokens": 8289662.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.802282047271729,
|
|
"epoch": 3.922217447357112,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003863077630991518,
|
|
"loss": 5.5317,
|
|
"mean_token_accuracy": 0.17601389735937117,
|
|
"num_tokens": 8298795.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.8081879138946535,
|
|
"epoch": 3.9265148259561666,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0003860186931651139,
|
|
"loss": 5.5671,
|
|
"mean_token_accuracy": 0.17194166481494905,
|
|
"num_tokens": 8306855.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.836734867095947,
|
|
"epoch": 3.930812204555221,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003857293808106238,
|
|
"loss": 5.6094,
|
|
"mean_token_accuracy": 0.1730263873934746,
|
|
"num_tokens": 8316041.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.8036181926727295,
|
|
"epoch": 3.9351095831542757,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003854398266674241,
|
|
"loss": 5.5949,
|
|
"mean_token_accuracy": 0.17814355939626694,
|
|
"num_tokens": 8325873.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.760510492324829,
|
|
"epoch": 3.9394069617533303,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00038515003136778544,
|
|
"loss": 5.5104,
|
|
"mean_token_accuracy": 0.17347990572452546,
|
|
"num_tokens": 8334786.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.771294689178466,
|
|
"epoch": 3.943704340352385,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038485999554450483,
|
|
"loss": 5.5239,
|
|
"mean_token_accuracy": 0.18543218821287155,
|
|
"num_tokens": 8345189.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.763915920257569,
|
|
"epoch": 3.9480017189514394,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00038456971983090454,
|
|
"loss": 5.566,
|
|
"mean_token_accuracy": 0.17913423478603363,
|
|
"num_tokens": 8353537.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.834560966491699,
|
|
"epoch": 3.9522990975504944,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003842792048608309,
|
|
"loss": 5.5753,
|
|
"mean_token_accuracy": 0.17576515376567842,
|
|
"num_tokens": 8363477.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.767901134490967,
|
|
"epoch": 3.956596476149549,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0003839884512686523,
|
|
"loss": 5.5918,
|
|
"mean_token_accuracy": 0.17605188935995103,
|
|
"num_tokens": 8372501.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.745550155639648,
|
|
"epoch": 3.9608938547486034,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00038369745968925846,
|
|
"loss": 5.603,
|
|
"mean_token_accuracy": 0.18295123279094697,
|
|
"num_tokens": 8381457.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.828930997848511,
|
|
"epoch": 3.965191233347658,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00038340623075805875,
|
|
"loss": 5.6259,
|
|
"mean_token_accuracy": 0.17518098950386046,
|
|
"num_tokens": 8390506.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.843565988540649,
|
|
"epoch": 3.9694886119467125,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00038311476511098053,
|
|
"loss": 5.6834,
|
|
"mean_token_accuracy": 0.16934775561094284,
|
|
"num_tokens": 8399803.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.765578079223633,
|
|
"epoch": 3.973785990545767,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.0003828230633844685,
|
|
"loss": 5.5455,
|
|
"mean_token_accuracy": 0.18085446506738662,
|
|
"num_tokens": 8408531.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.686979532241821,
|
|
"epoch": 3.9780833691448216,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00038253112621548243,
|
|
"loss": 5.4734,
|
|
"mean_token_accuracy": 0.19051893651485444,
|
|
"num_tokens": 8416721.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.751924467086792,
|
|
"epoch": 3.982380747743876,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003822389542414966,
|
|
"loss": 5.4921,
|
|
"mean_token_accuracy": 0.18940197676420212,
|
|
"num_tokens": 8425896.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.8088000297546385,
|
|
"epoch": 3.9866781263429307,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00038194654810049775,
|
|
"loss": 5.5747,
|
|
"mean_token_accuracy": 0.174156029522419,
|
|
"num_tokens": 8434695.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.808415126800537,
|
|
"epoch": 3.9909755049419853,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000381653908430984,
|
|
"loss": 5.668,
|
|
"mean_token_accuracy": 0.1626859039068222,
|
|
"num_tokens": 8443696.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.833459711074829,
|
|
"epoch": 3.99527288354104,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003813610358719634,
|
|
"loss": 5.5772,
|
|
"mean_token_accuracy": 0.18208468109369277,
|
|
"num_tokens": 8453429.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.896011066436768,
|
|
"epoch": 3.999570262140095,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038106793106295266,
|
|
"loss": 5.6138,
|
|
"mean_token_accuracy": 0.17296209633350373,
|
|
"num_tokens": 8462823.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.785070101420085,
|
|
"epoch": 4.003437902879243,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003807745946439754,
|
|
"loss": 5.5021,
|
|
"mean_token_accuracy": 0.1812961714135276,
|
|
"num_tokens": 8471681.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.785916996002197,
|
|
"epoch": 4.007735281478298,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003804810272555612,
|
|
"loss": 5.3272,
|
|
"mean_token_accuracy": 0.18606556206941605,
|
|
"num_tokens": 8480819.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.704463100433349,
|
|
"epoch": 4.012032660077352,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0003801872295387439,
|
|
"loss": 5.2663,
|
|
"mean_token_accuracy": 0.20178429931402206,
|
|
"num_tokens": 8489692.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.77840895652771,
|
|
"epoch": 4.016330038676408,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003798932021350603,
|
|
"loss": 5.3609,
|
|
"mean_token_accuracy": 0.18003070652484893,
|
|
"num_tokens": 8498523.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.78088607788086,
|
|
"epoch": 4.020627417275462,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00037959894568654864,
|
|
"loss": 5.3854,
|
|
"mean_token_accuracy": 0.18569311797618865,
|
|
"num_tokens": 8507641.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.721806430816651,
|
|
"epoch": 4.024924795874517,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003793044608357474,
|
|
"loss": 5.2523,
|
|
"mean_token_accuracy": 0.1940135583281517,
|
|
"num_tokens": 8516082.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.861346769332886,
|
|
"epoch": 4.0292221744735714,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003790097482256939,
|
|
"loss": 5.4127,
|
|
"mean_token_accuracy": 0.18142988830804824,
|
|
"num_tokens": 8526424.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.786660146713257,
|
|
"epoch": 4.033519553072626,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003787148084999225,
|
|
"loss": 5.2836,
|
|
"mean_token_accuracy": 0.19199537336826325,
|
|
"num_tokens": 8536729.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.7474586963653564,
|
|
"epoch": 4.0378169316716805,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00037841964230246394,
|
|
"loss": 5.3701,
|
|
"mean_token_accuracy": 0.18352359384298325,
|
|
"num_tokens": 8546250.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.745085668563843,
|
|
"epoch": 4.042114310270735,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003781242502778429,
|
|
"loss": 5.3313,
|
|
"mean_token_accuracy": 0.1962459549307823,
|
|
"num_tokens": 8555067.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.74356746673584,
|
|
"epoch": 4.04641168886979,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00037782863307107785,
|
|
"loss": 5.2623,
|
|
"mean_token_accuracy": 0.19740765988826753,
|
|
"num_tokens": 8563638.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.752368068695068,
|
|
"epoch": 4.050709067468844,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00037753279132767833,
|
|
"loss": 5.3617,
|
|
"mean_token_accuracy": 0.18477154076099395,
|
|
"num_tokens": 8572233.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.75670108795166,
|
|
"epoch": 4.055006446067899,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00037723672569364453,
|
|
"loss": 5.3116,
|
|
"mean_token_accuracy": 0.19277881383895873,
|
|
"num_tokens": 8580995.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.716158151626587,
|
|
"epoch": 4.059303824666953,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00037694043681546545,
|
|
"loss": 5.3091,
|
|
"mean_token_accuracy": 0.1887386381626129,
|
|
"num_tokens": 8590250.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.7661590576171875,
|
|
"epoch": 4.063601203266008,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 0.0003766439253401177,
|
|
"loss": 5.3211,
|
|
"mean_token_accuracy": 0.19219609647989272,
|
|
"num_tokens": 8599183.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.735136604309082,
|
|
"epoch": 4.067898581865062,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00037634719191506367,
|
|
"loss": 5.2917,
|
|
"mean_token_accuracy": 0.1975168526172638,
|
|
"num_tokens": 8607085.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.718432426452637,
|
|
"epoch": 4.072195960464117,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00037605023718825065,
|
|
"loss": 5.3303,
|
|
"mean_token_accuracy": 0.1862225890159607,
|
|
"num_tokens": 8616308.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.823801708221436,
|
|
"epoch": 4.0764933390631715,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 0.000375753061808109,
|
|
"loss": 5.4226,
|
|
"mean_token_accuracy": 0.18610190898180007,
|
|
"num_tokens": 8625337.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.80818772315979,
|
|
"epoch": 4.080790717662226,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00037545566642355107,
|
|
"loss": 5.3634,
|
|
"mean_token_accuracy": 0.18869078159332275,
|
|
"num_tokens": 8634389.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.748035573959351,
|
|
"epoch": 4.0850880962612806,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003751580516839695,
|
|
"loss": 5.4805,
|
|
"mean_token_accuracy": 0.1765426829457283,
|
|
"num_tokens": 8644691.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.717017745971679,
|
|
"epoch": 4.089385474860335,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00037486021823923574,
|
|
"loss": 5.3737,
|
|
"mean_token_accuracy": 0.18233591318130493,
|
|
"num_tokens": 8653775.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.819413709640503,
|
|
"epoch": 4.09368285345939,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00037456216673969925,
|
|
"loss": 5.4217,
|
|
"mean_token_accuracy": 0.1816790908575058,
|
|
"num_tokens": 8661958.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.846973037719726,
|
|
"epoch": 4.097980232058444,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003742638978361851,
|
|
"loss": 5.322,
|
|
"mean_token_accuracy": 0.18977204710245132,
|
|
"num_tokens": 8671432.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.731178522109985,
|
|
"epoch": 4.102277610657499,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00037396541217999367,
|
|
"loss": 5.3221,
|
|
"mean_token_accuracy": 0.19207046329975128,
|
|
"num_tokens": 8681778.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.6961040019989015,
|
|
"epoch": 4.106574989256553,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003736667104228981,
|
|
"loss": 5.3705,
|
|
"mean_token_accuracy": 0.18717817813158036,
|
|
"num_tokens": 8690769.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.7984966278076175,
|
|
"epoch": 4.110872367855608,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037336779321714376,
|
|
"loss": 5.3476,
|
|
"mean_token_accuracy": 0.1864815980195999,
|
|
"num_tokens": 8699986.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.672695159912109,
|
|
"epoch": 4.115169746454662,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00037306866121544633,
|
|
"loss": 5.2734,
|
|
"mean_token_accuracy": 0.20004051625728608,
|
|
"num_tokens": 8708517.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.748255920410156,
|
|
"epoch": 4.119467125053717,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003727693150709904,
|
|
"loss": 5.4146,
|
|
"mean_token_accuracy": 0.19286408126354218,
|
|
"num_tokens": 8717556.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.72618350982666,
|
|
"epoch": 4.1237645036527715,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00037246975543742843,
|
|
"loss": 5.3114,
|
|
"mean_token_accuracy": 0.187771038711071,
|
|
"num_tokens": 8726499.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.726382923126221,
|
|
"epoch": 4.128061882251826,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000372169982968879,
|
|
"loss": 5.3155,
|
|
"mean_token_accuracy": 0.19927477687597275,
|
|
"num_tokens": 8734449.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.747401857376099,
|
|
"epoch": 4.132359260850881,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003718699983199252,
|
|
"loss": 5.3723,
|
|
"mean_token_accuracy": 0.1890871897339821,
|
|
"num_tokens": 8744098.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.678901624679566,
|
|
"epoch": 4.136656639449935,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.0003715698021456137,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.2021241933107376,
|
|
"num_tokens": 8753719.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.790144777297973,
|
|
"epoch": 4.1409540180489905,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.00037126939510145294,
|
|
"loss": 5.3301,
|
|
"mean_token_accuracy": 0.19362499862909316,
|
|
"num_tokens": 8761883.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.6856646060943605,
|
|
"epoch": 4.145251396648045,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003709687778434118,
|
|
"loss": 5.2654,
|
|
"mean_token_accuracy": 0.19640474766492844,
|
|
"num_tokens": 8770594.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.674073505401611,
|
|
"epoch": 4.1495487752471,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003706679510279183,
|
|
"loss": 5.2902,
|
|
"mean_token_accuracy": 0.19421906918287277,
|
|
"num_tokens": 8780314.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.755329704284668,
|
|
"epoch": 4.153846153846154,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003703669153118578,
|
|
"loss": 5.332,
|
|
"mean_token_accuracy": 0.1903119459748268,
|
|
"num_tokens": 8788832.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.7171543598175045,
|
|
"epoch": 4.158143532445209,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00037006567135257216,
|
|
"loss": 5.3378,
|
|
"mean_token_accuracy": 0.18289393335580825,
|
|
"num_tokens": 8797457.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.723929405212402,
|
|
"epoch": 4.162440911044263,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00036976421980785764,
|
|
"loss": 5.4053,
|
|
"mean_token_accuracy": 0.19152163863182067,
|
|
"num_tokens": 8806760.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.751377296447754,
|
|
"epoch": 4.166738289643318,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003694625613359641,
|
|
"loss": 5.2334,
|
|
"mean_token_accuracy": 0.19349892437458038,
|
|
"num_tokens": 8815641.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.715363359451294,
|
|
"epoch": 4.171035668242372,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003691606965955929,
|
|
"loss": 5.2654,
|
|
"mean_token_accuracy": 0.19013406187295914,
|
|
"num_tokens": 8824532.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.790452289581299,
|
|
"epoch": 4.175333046841427,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000368858626245896,
|
|
"loss": 5.4205,
|
|
"mean_token_accuracy": 0.17900191843509675,
|
|
"num_tokens": 8834896.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.770239639282226,
|
|
"epoch": 4.1796304254404815,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003685563509464744,
|
|
"loss": 5.339,
|
|
"mean_token_accuracy": 0.18509139120578766,
|
|
"num_tokens": 8844320.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.701741456985474,
|
|
"epoch": 4.183927804039536,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00036825387135737647,
|
|
"loss": 5.2413,
|
|
"mean_token_accuracy": 0.18954674601554872,
|
|
"num_tokens": 8853400.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.729306411743164,
|
|
"epoch": 4.188225182638591,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00036795118813909674,
|
|
"loss": 5.3283,
|
|
"mean_token_accuracy": 0.19178215712308883,
|
|
"num_tokens": 8861930.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.664833068847656,
|
|
"epoch": 4.192522561237645,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 0.00036764830195257437,
|
|
"loss": 5.4136,
|
|
"mean_token_accuracy": 0.18483411222696305,
|
|
"num_tokens": 8871103.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.766524028778076,
|
|
"epoch": 4.1968199398367,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.0003673452134591918,
|
|
"loss": 5.4422,
|
|
"mean_token_accuracy": 0.17678992152214051,
|
|
"num_tokens": 8879875.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.755912590026855,
|
|
"epoch": 4.201117318435754,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000367041923320773,
|
|
"loss": 5.3775,
|
|
"mean_token_accuracy": 0.1837664231657982,
|
|
"num_tokens": 8888992.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.7815014839172365,
|
|
"epoch": 4.205414697034809,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036673843219958257,
|
|
"loss": 5.3935,
|
|
"mean_token_accuracy": 0.1891205683350563,
|
|
"num_tokens": 8898609.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.762997579574585,
|
|
"epoch": 4.209712075633863,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003664347407583238,
|
|
"loss": 5.3758,
|
|
"mean_token_accuracy": 0.1859879970550537,
|
|
"num_tokens": 8907998.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.7384720802307125,
|
|
"epoch": 4.214009454232918,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0003661308496601373,
|
|
"loss": 5.364,
|
|
"mean_token_accuracy": 0.19431136548519135,
|
|
"num_tokens": 8916369.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.798846960067749,
|
|
"epoch": 4.218306832831972,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00036582675956859983,
|
|
"loss": 5.3859,
|
|
"mean_token_accuracy": 0.1881580486893654,
|
|
"num_tokens": 8926482.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.666932821273804,
|
|
"epoch": 4.222604211431027,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.00036552247114772263,
|
|
"loss": 5.2143,
|
|
"mean_token_accuracy": 0.20004839450120926,
|
|
"num_tokens": 8934329.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.736501169204712,
|
|
"epoch": 4.2269015900300815,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00036521798506194996,
|
|
"loss": 5.5424,
|
|
"mean_token_accuracy": 0.1650074765086174,
|
|
"num_tokens": 8943606.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.800119638442993,
|
|
"epoch": 4.231198968629136,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036491330197615775,
|
|
"loss": 5.395,
|
|
"mean_token_accuracy": 0.18160003274679185,
|
|
"num_tokens": 8953384.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.736812114715576,
|
|
"epoch": 4.235496347228191,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00036460842255565197,
|
|
"loss": 5.3129,
|
|
"mean_token_accuracy": 0.19640697836875914,
|
|
"num_tokens": 8962269.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.766232872009278,
|
|
"epoch": 4.239793725827245,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003643033474661676,
|
|
"loss": 5.3286,
|
|
"mean_token_accuracy": 0.18668037950992583,
|
|
"num_tokens": 8970476.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.722537469863892,
|
|
"epoch": 4.2440911044263,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036399807737386657,
|
|
"loss": 5.3935,
|
|
"mean_token_accuracy": 0.18381021022796631,
|
|
"num_tokens": 8980322.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.703783702850342,
|
|
"epoch": 4.248388483025354,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003636926129453368,
|
|
"loss": 5.356,
|
|
"mean_token_accuracy": 0.19123856574296952,
|
|
"num_tokens": 8989575.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.711059474945069,
|
|
"epoch": 4.252685861624409,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003633869548475904,
|
|
"loss": 5.2539,
|
|
"mean_token_accuracy": 0.18876997083425523,
|
|
"num_tokens": 8998027.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.760339450836182,
|
|
"epoch": 4.256983240223463,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003630811037480627,
|
|
"loss": 5.341,
|
|
"mean_token_accuracy": 0.18147286921739578,
|
|
"num_tokens": 9006942.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.695712375640869,
|
|
"epoch": 4.261280618822518,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003627750603146101,
|
|
"loss": 5.3212,
|
|
"mean_token_accuracy": 0.1893489122390747,
|
|
"num_tokens": 9016863.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.758656358718872,
|
|
"epoch": 4.265577997421573,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003624688252155091,
|
|
"loss": 5.3777,
|
|
"mean_token_accuracy": 0.18578420132398604,
|
|
"num_tokens": 9025837.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.8174535751342775,
|
|
"epoch": 4.269875376020628,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003621623991194549,
|
|
"loss": 5.3123,
|
|
"mean_token_accuracy": 0.19068270474672316,
|
|
"num_tokens": 9034625.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.769172239303589,
|
|
"epoch": 4.274172754619682,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003618557826955594,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.18804719001054765,
|
|
"num_tokens": 9043547.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.775179767608643,
|
|
"epoch": 4.278470133218737,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00036154897661335063,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.18669925183057784,
|
|
"num_tokens": 9052857.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.861923122406006,
|
|
"epoch": 4.2827675118177915,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003612419815427702,
|
|
"loss": 5.3577,
|
|
"mean_token_accuracy": 0.18439370393753052,
|
|
"num_tokens": 9062823.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.7063226222991945,
|
|
"epoch": 4.287064890416846,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003609347981541726,
|
|
"loss": 5.3464,
|
|
"mean_token_accuracy": 0.18418395072221755,
|
|
"num_tokens": 9071209.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.651985692977905,
|
|
"epoch": 4.291362269015901,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036062742711832376,
|
|
"loss": 5.3726,
|
|
"mean_token_accuracy": 0.18854329884052276,
|
|
"num_tokens": 9082699.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.753876113891602,
|
|
"epoch": 4.295659647614955,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003603198691063991,
|
|
"loss": 5.3351,
|
|
"mean_token_accuracy": 0.18982358127832413,
|
|
"num_tokens": 9091116.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.295659647614955,
|
|
"eval_entropy": 5.591331935680665,
|
|
"eval_loss": 6.053544521331787,
|
|
"eval_mean_token_accuracy": 0.16821764060505875,
|
|
"eval_num_tokens": 9091116.0,
|
|
"eval_runtime": 2.6562,
|
|
"eval_samples_per_second": 1336.128,
|
|
"eval_steps_per_second": 167.157,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.678605365753174,
|
|
"epoch": 4.29995702621401,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003600121247899824,
|
|
"loss": 5.2646,
|
|
"mean_token_accuracy": 0.19501282572746276,
|
|
"num_tokens": 9099470.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.668772459030151,
|
|
"epoch": 4.304254404813064,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035970419484106404,
|
|
"loss": 5.3036,
|
|
"mean_token_accuracy": 0.19127827882766724,
|
|
"num_tokens": 9108367.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.737813091278076,
|
|
"epoch": 4.308551783412119,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 0.0003593960799320402,
|
|
"loss": 5.3298,
|
|
"mean_token_accuracy": 0.18940934389829636,
|
|
"num_tokens": 9116560.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.694883489608765,
|
|
"epoch": 4.312849162011173,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003590877807357107,
|
|
"loss": 5.3762,
|
|
"mean_token_accuracy": 0.18794360309839248,
|
|
"num_tokens": 9124634.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.738213729858399,
|
|
"epoch": 4.317146540610228,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003587792979252776,
|
|
"loss": 5.3718,
|
|
"mean_token_accuracy": 0.18815282881259918,
|
|
"num_tokens": 9133246.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.703302431106567,
|
|
"epoch": 4.321443919209282,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003584706321743442,
|
|
"loss": 5.3256,
|
|
"mean_token_accuracy": 0.18445513844490052,
|
|
"num_tokens": 9142374.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.790320825576782,
|
|
"epoch": 4.325741297808337,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000358161784156913,
|
|
"loss": 5.3657,
|
|
"mean_token_accuracy": 0.18346769809722902,
|
|
"num_tokens": 9151657.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.714098358154297,
|
|
"epoch": 4.3300386764073915,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00035785275454738456,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.18277858197689056,
|
|
"num_tokens": 9159897.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.639693975448608,
|
|
"epoch": 4.334336055006446,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035754354402055635,
|
|
"loss": 5.4149,
|
|
"mean_token_accuracy": 0.18041964769363403,
|
|
"num_tokens": 9169798.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.88186502456665,
|
|
"epoch": 4.338633433605501,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.0003572341532516202,
|
|
"loss": 5.5088,
|
|
"mean_token_accuracy": 0.17149559259414673,
|
|
"num_tokens": 9179107.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.795796489715576,
|
|
"epoch": 4.342930812204555,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003569245829161622,
|
|
"loss": 5.3676,
|
|
"mean_token_accuracy": 0.1848204553127289,
|
|
"num_tokens": 9188295.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.703272533416748,
|
|
"epoch": 4.34722819080361,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.00035661483369016004,
|
|
"loss": 5.335,
|
|
"mean_token_accuracy": 0.1941680446267128,
|
|
"num_tokens": 9197603.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.7337511539459225,
|
|
"epoch": 4.351525569402664,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003563049062499822,
|
|
"loss": 5.349,
|
|
"mean_token_accuracy": 0.18076217025518418,
|
|
"num_tokens": 9206413.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.793175077438354,
|
|
"epoch": 4.355822948001719,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003559948012723865,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.1787755846977234,
|
|
"num_tokens": 9215376.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.805943632125855,
|
|
"epoch": 4.360120326600773,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003556845194345181,
|
|
"loss": 5.4361,
|
|
"mean_token_accuracy": 0.17978136539459227,
|
|
"num_tokens": 9224418.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.717202043533325,
|
|
"epoch": 4.364417705199828,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003553740614139086,
|
|
"loss": 5.3098,
|
|
"mean_token_accuracy": 0.1895645707845688,
|
|
"num_tokens": 9235083.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.7130590915679935,
|
|
"epoch": 4.368715083798882,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003550634278884742,
|
|
"loss": 5.2915,
|
|
"mean_token_accuracy": 0.19898206144571304,
|
|
"num_tokens": 9244576.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.686877250671387,
|
|
"epoch": 4.373012462397937,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00035475261953651433,
|
|
"loss": 5.3524,
|
|
"mean_token_accuracy": 0.190613454580307,
|
|
"num_tokens": 9253969.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.741063404083252,
|
|
"epoch": 4.3773098409969915,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035444163703671026,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.18438142836093901,
|
|
"num_tokens": 9264723.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.780627870559693,
|
|
"epoch": 4.381607219596046,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035413048106812357,
|
|
"loss": 5.3162,
|
|
"mean_token_accuracy": 0.19334478825330734,
|
|
"num_tokens": 9273863.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.761202239990235,
|
|
"epoch": 4.385904598195101,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00035381915231019425,
|
|
"loss": 5.3875,
|
|
"mean_token_accuracy": 0.18619555830955506,
|
|
"num_tokens": 9282633.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.750490999221801,
|
|
"epoch": 4.390201976794156,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003535076514427401,
|
|
"loss": 5.4514,
|
|
"mean_token_accuracy": 0.1834085300564766,
|
|
"num_tokens": 9291575.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.754965114593506,
|
|
"epoch": 4.39449935539321,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00035319597914595436,
|
|
"loss": 5.3417,
|
|
"mean_token_accuracy": 0.18682506084442138,
|
|
"num_tokens": 9299883.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.744359064102173,
|
|
"epoch": 4.398796733992265,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003528841361004049,
|
|
"loss": 5.3665,
|
|
"mean_token_accuracy": 0.18644660860300064,
|
|
"num_tokens": 9308828.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.748482131958008,
|
|
"epoch": 4.40309411259132,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003525721229870323,
|
|
"loss": 5.3968,
|
|
"mean_token_accuracy": 0.18078760355710982,
|
|
"num_tokens": 9318458.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.7246194839477536,
|
|
"epoch": 4.407391491190374,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00035225994048714823,
|
|
"loss": 5.328,
|
|
"mean_token_accuracy": 0.19000641852617264,
|
|
"num_tokens": 9327811.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.707337236404419,
|
|
"epoch": 4.411688869789429,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0003519475892824348,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.1916312411427498,
|
|
"num_tokens": 9336169.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.690235900878906,
|
|
"epoch": 4.415986248388483,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003516350700549419,
|
|
"loss": 5.3425,
|
|
"mean_token_accuracy": 0.19429921954870225,
|
|
"num_tokens": 9345141.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.669893503189087,
|
|
"epoch": 4.420283626987538,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00035132238348708697,
|
|
"loss": 5.243,
|
|
"mean_token_accuracy": 0.19799396842718125,
|
|
"num_tokens": 9353982.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.751178216934204,
|
|
"epoch": 4.424581005586592,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00035100953026165224,
|
|
"loss": 5.3157,
|
|
"mean_token_accuracy": 0.20062485635280608,
|
|
"num_tokens": 9362991.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.621764326095581,
|
|
"epoch": 4.428878384185647,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003506965110617841,
|
|
"loss": 5.3264,
|
|
"mean_token_accuracy": 0.19708930552005768,
|
|
"num_tokens": 9372310.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.723211717605591,
|
|
"epoch": 4.4331757627847015,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003503833265709915,
|
|
"loss": 5.3703,
|
|
"mean_token_accuracy": 0.18472055941820145,
|
|
"num_tokens": 9381268.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.786382436752319,
|
|
"epoch": 4.437473141383756,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00035006997747314404,
|
|
"loss": 5.3296,
|
|
"mean_token_accuracy": 0.19576110690832138,
|
|
"num_tokens": 9389670.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.674931287765503,
|
|
"epoch": 4.441770519982811,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00034975646445247106,
|
|
"loss": 5.2994,
|
|
"mean_token_accuracy": 0.19079017341136933,
|
|
"num_tokens": 9397520.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.691173028945923,
|
|
"epoch": 4.446067898581865,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0003494427881935596,
|
|
"loss": 5.3158,
|
|
"mean_token_accuracy": 0.19079864174127578,
|
|
"num_tokens": 9406848.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.690839052200317,
|
|
"epoch": 4.45036527718092,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00034912894938135325,
|
|
"loss": 5.314,
|
|
"mean_token_accuracy": 0.18439335823059083,
|
|
"num_tokens": 9415820.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.663485097885132,
|
|
"epoch": 4.454662655779974,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0003488149487011506,
|
|
"loss": 5.2766,
|
|
"mean_token_accuracy": 0.19169690907001496,
|
|
"num_tokens": 9424907.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.699732112884521,
|
|
"epoch": 4.458960034379029,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00034850078683860346,
|
|
"loss": 5.363,
|
|
"mean_token_accuracy": 0.19870943576097488,
|
|
"num_tokens": 9433882.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.786200141906738,
|
|
"epoch": 4.463257412978083,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 0.0003481864644797159,
|
|
"loss": 5.3657,
|
|
"mean_token_accuracy": 0.18985778987407684,
|
|
"num_tokens": 9443544.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.745025539398194,
|
|
"epoch": 4.467554791577138,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003478719823108424,
|
|
"loss": 5.394,
|
|
"mean_token_accuracy": 0.17771507203578948,
|
|
"num_tokens": 9452473.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.721456098556518,
|
|
"epoch": 4.471852170176192,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00034755734101868613,
|
|
"loss": 5.3534,
|
|
"mean_token_accuracy": 0.18390342593193054,
|
|
"num_tokens": 9461392.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.700122308731079,
|
|
"epoch": 4.476149548775247,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00034724254129029795,
|
|
"loss": 5.3392,
|
|
"mean_token_accuracy": 0.19301667660474778,
|
|
"num_tokens": 9470087.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.655315637588501,
|
|
"epoch": 4.4804469273743015,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 0.0003469275838130748,
|
|
"loss": 5.3097,
|
|
"mean_token_accuracy": 0.19494091272354125,
|
|
"num_tokens": 9479654.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.697892379760742,
|
|
"epoch": 4.484744305973356,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0003466124692747577,
|
|
"loss": 5.3843,
|
|
"mean_token_accuracy": 0.18161331713199616,
|
|
"num_tokens": 9488116.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.6924889087677,
|
|
"epoch": 4.489041684572411,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00034629719836343106,
|
|
"loss": 5.3157,
|
|
"mean_token_accuracy": 0.19479441940784453,
|
|
"num_tokens": 9496853.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.723341560363769,
|
|
"epoch": 4.493339063171465,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003459817717675203,
|
|
"loss": 5.3245,
|
|
"mean_token_accuracy": 0.19436981976032258,
|
|
"num_tokens": 9507085.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.695764446258545,
|
|
"epoch": 4.49763644177052,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0003456661901757913,
|
|
"loss": 5.3846,
|
|
"mean_token_accuracy": 0.18006172329187392,
|
|
"num_tokens": 9516393.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.684157800674439,
|
|
"epoch": 4.501933820369574,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00034535045427734796,
|
|
"loss": 5.3743,
|
|
"mean_token_accuracy": 0.18558215498924255,
|
|
"num_tokens": 9525306.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.7475286483764645,
|
|
"epoch": 4.506231198968629,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003450345647616313,
|
|
"loss": 5.3902,
|
|
"mean_token_accuracy": 0.1826770693063736,
|
|
"num_tokens": 9534401.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.782970285415649,
|
|
"epoch": 4.510528577567683,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003447185223184177,
|
|
"loss": 5.4414,
|
|
"mean_token_accuracy": 0.18412038087844848,
|
|
"num_tokens": 9544651.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.710432672500611,
|
|
"epoch": 4.514825956166739,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00034440232763781765,
|
|
"loss": 5.3324,
|
|
"mean_token_accuracy": 0.19580706506967543,
|
|
"num_tokens": 9552903.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.6136516571044925,
|
|
"epoch": 4.519123334765792,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000344085981410274,
|
|
"loss": 5.3398,
|
|
"mean_token_accuracy": 0.19176888316869736,
|
|
"num_tokens": 9561129.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.737872314453125,
|
|
"epoch": 4.523420713364848,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00034376948432656036,
|
|
"loss": 5.3576,
|
|
"mean_token_accuracy": 0.18605385720729828,
|
|
"num_tokens": 9570075.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.7929657936096195,
|
|
"epoch": 4.527718091963902,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003434528370777798,
|
|
"loss": 5.3024,
|
|
"mean_token_accuracy": 0.19521451741456985,
|
|
"num_tokens": 9578352.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.657254362106324,
|
|
"epoch": 4.532015470562957,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00034313604035536344,
|
|
"loss": 5.3978,
|
|
"mean_token_accuracy": 0.1889173671603203,
|
|
"num_tokens": 9587049.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.7483906745910645,
|
|
"epoch": 4.5363128491620115,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003428190948510687,
|
|
"loss": 5.3705,
|
|
"mean_token_accuracy": 0.18301943838596343,
|
|
"num_tokens": 9595856.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.652540922164917,
|
|
"epoch": 4.540610227761066,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003425020012569778,
|
|
"loss": 5.2978,
|
|
"mean_token_accuracy": 0.19216183423995972,
|
|
"num_tokens": 9604258.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.745301532745361,
|
|
"epoch": 4.544907606360121,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.00034218476026549665,
|
|
"loss": 5.3693,
|
|
"mean_token_accuracy": 0.18393879681825637,
|
|
"num_tokens": 9613961.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.778255891799927,
|
|
"epoch": 4.549204984959175,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003418673725693524,
|
|
"loss": 5.4234,
|
|
"mean_token_accuracy": 0.17763426899909973,
|
|
"num_tokens": 9623440.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.739652061462403,
|
|
"epoch": 4.55350236355823,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003415498388615932,
|
|
"loss": 5.3994,
|
|
"mean_token_accuracy": 0.18231528550386428,
|
|
"num_tokens": 9632988.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.71859221458435,
|
|
"epoch": 4.557799742157284,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003412321598355857,
|
|
"loss": 5.4083,
|
|
"mean_token_accuracy": 0.19213815927505493,
|
|
"num_tokens": 9641479.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.755429601669311,
|
|
"epoch": 4.562097120756339,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003409143361850139,
|
|
"loss": 5.3844,
|
|
"mean_token_accuracy": 0.18924206793308257,
|
|
"num_tokens": 9651088.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.65693473815918,
|
|
"epoch": 4.566394499355393,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0003405963686038775,
|
|
"loss": 5.2964,
|
|
"mean_token_accuracy": 0.1913794070482254,
|
|
"num_tokens": 9659969.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.753949785232544,
|
|
"epoch": 4.570691877954448,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003402782577864908,
|
|
"loss": 5.4404,
|
|
"mean_token_accuracy": 0.18301080763339997,
|
|
"num_tokens": 9669509.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.689779901504517,
|
|
"epoch": 4.574989256553502,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033996000442748056,
|
|
"loss": 5.3115,
|
|
"mean_token_accuracy": 0.1891514018177986,
|
|
"num_tokens": 9677758.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.734157419204712,
|
|
"epoch": 4.579286635152557,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00033964160922178495,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.18529036343097688,
|
|
"num_tokens": 9686622.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.695058965682984,
|
|
"epoch": 4.5835840137516115,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0003393230728646518,
|
|
"loss": 5.3165,
|
|
"mean_token_accuracy": 0.18462513834238053,
|
|
"num_tokens": 9694726.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.666723155975342,
|
|
"epoch": 4.587881392350666,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00033900439605163724,
|
|
"loss": 5.3229,
|
|
"mean_token_accuracy": 0.19404973238706588,
|
|
"num_tokens": 9703137.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.7576799392700195,
|
|
"epoch": 4.592178770949721,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033868557947860407,
|
|
"loss": 5.345,
|
|
"mean_token_accuracy": 0.1878753423690796,
|
|
"num_tokens": 9711818.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.70137300491333,
|
|
"epoch": 4.596476149548775,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00033836662384172014,
|
|
"loss": 5.3665,
|
|
"mean_token_accuracy": 0.19357299357652663,
|
|
"num_tokens": 9720678.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.747474527359008,
|
|
"epoch": 4.60077352814783,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003380475298374573,
|
|
"loss": 5.3913,
|
|
"mean_token_accuracy": 0.18863837271928788,
|
|
"num_tokens": 9729704.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.719842529296875,
|
|
"epoch": 4.605070906746884,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000337728298162589,
|
|
"loss": 5.3899,
|
|
"mean_token_accuracy": 0.18907782882452012,
|
|
"num_tokens": 9739223.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.759377527236938,
|
|
"epoch": 4.609368285345939,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00033740892951418993,
|
|
"loss": 5.3755,
|
|
"mean_token_accuracy": 0.19428345561027527,
|
|
"num_tokens": 9748905.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.721805810928345,
|
|
"epoch": 4.613665663944993,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0003370894245896333,
|
|
"loss": 5.4203,
|
|
"mean_token_accuracy": 0.18931963443756103,
|
|
"num_tokens": 9758156.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.671443319320678,
|
|
"epoch": 4.617963042544048,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00033676978408659047,
|
|
"loss": 5.3817,
|
|
"mean_token_accuracy": 0.18796280920505523,
|
|
"num_tokens": 9767547.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.702896690368652,
|
|
"epoch": 4.622260421143102,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003364500087030283,
|
|
"loss": 5.4522,
|
|
"mean_token_accuracy": 0.1870797097682953,
|
|
"num_tokens": 9777448.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.785913515090942,
|
|
"epoch": 4.626557799742157,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00033613009913720845,
|
|
"loss": 5.413,
|
|
"mean_token_accuracy": 0.18920918554067612,
|
|
"num_tokens": 9785856.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.80258002281189,
|
|
"epoch": 4.6308551783412115,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033581005608768563,
|
|
"loss": 5.415,
|
|
"mean_token_accuracy": 0.1848719224333763,
|
|
"num_tokens": 9794806.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.763333034515381,
|
|
"epoch": 4.635152556940266,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0003354898802533058,
|
|
"loss": 5.3631,
|
|
"mean_token_accuracy": 0.1853545919060707,
|
|
"num_tokens": 9804105.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.774251222610474,
|
|
"epoch": 4.6394499355393215,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003351695723332051,
|
|
"loss": 5.4696,
|
|
"mean_token_accuracy": 0.17461379766464233,
|
|
"num_tokens": 9812875.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.663911581039429,
|
|
"epoch": 4.643747314138375,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00033484913302680807,
|
|
"loss": 5.2352,
|
|
"mean_token_accuracy": 0.19657050222158431,
|
|
"num_tokens": 9821542.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.643954420089722,
|
|
"epoch": 4.648044692737431,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00033452856303382595,
|
|
"loss": 5.2794,
|
|
"mean_token_accuracy": 0.2007174998521805,
|
|
"num_tokens": 9830924.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.708545541763305,
|
|
"epoch": 4.652342071336484,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0003342078630542555,
|
|
"loss": 5.3675,
|
|
"mean_token_accuracy": 0.19253413379192352,
|
|
"num_tokens": 9839967.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.768747568130493,
|
|
"epoch": 4.65663944993554,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00033388703378837737,
|
|
"loss": 5.3874,
|
|
"mean_token_accuracy": 0.18171902298927306,
|
|
"num_tokens": 9849129.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.724219465255738,
|
|
"epoch": 4.660936828534594,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003335660759367544,
|
|
"loss": 5.3794,
|
|
"mean_token_accuracy": 0.1904281958937645,
|
|
"num_tokens": 9858303.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.739745092391968,
|
|
"epoch": 4.665234207133649,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033324499020023025,
|
|
"loss": 5.3741,
|
|
"mean_token_accuracy": 0.17980744540691376,
|
|
"num_tokens": 9866886.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.685865592956543,
|
|
"epoch": 4.669531585732703,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003329237772799277,
|
|
"loss": 5.294,
|
|
"mean_token_accuracy": 0.19730629324913024,
|
|
"num_tokens": 9875474.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.649552726745606,
|
|
"epoch": 4.673828964331758,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003326024378772477,
|
|
"loss": 5.2806,
|
|
"mean_token_accuracy": 0.1952342763543129,
|
|
"num_tokens": 9885104.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.74921760559082,
|
|
"epoch": 4.678126342930812,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003322809726938667,
|
|
"loss": 5.4086,
|
|
"mean_token_accuracy": 0.18478808104991912,
|
|
"num_tokens": 9894698.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.747147750854492,
|
|
"epoch": 4.682423721529867,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00033195938243173645,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.1903951957821846,
|
|
"num_tokens": 9903850.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.767203664779663,
|
|
"epoch": 4.6867211001289215,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003316376677930814,
|
|
"loss": 5.435,
|
|
"mean_token_accuracy": 0.1796480268239975,
|
|
"num_tokens": 9912823.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.697248315811157,
|
|
"epoch": 4.691018478727976,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003313158294803977,
|
|
"loss": 5.3899,
|
|
"mean_token_accuracy": 0.18389313369989396,
|
|
"num_tokens": 9922279.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.69881682395935,
|
|
"epoch": 4.695315857327031,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00033099386819645176,
|
|
"loss": 5.4661,
|
|
"mean_token_accuracy": 0.1810116797685623,
|
|
"num_tokens": 9931646.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.791141653060913,
|
|
"epoch": 4.699613235926085,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.0003306717846442782,
|
|
"loss": 5.4216,
|
|
"mean_token_accuracy": 0.1830429956316948,
|
|
"num_tokens": 9940672.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.684693336486816,
|
|
"epoch": 4.70391061452514,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003303495795271788,
|
|
"loss": 5.2702,
|
|
"mean_token_accuracy": 0.19574187994003295,
|
|
"num_tokens": 9948954.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.7547111988067625,
|
|
"epoch": 4.708207993124194,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033002725354872075,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.1855279877781868,
|
|
"num_tokens": 9959132.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.71658411026001,
|
|
"epoch": 4.712505371723249,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032970480741273514,
|
|
"loss": 5.4641,
|
|
"mean_token_accuracy": 0.18244242072105407,
|
|
"num_tokens": 9969143.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.741818523406982,
|
|
"epoch": 4.716802750322303,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003293822418233155,
|
|
"loss": 5.4122,
|
|
"mean_token_accuracy": 0.1828443467617035,
|
|
"num_tokens": 9978462.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.82344536781311,
|
|
"epoch": 4.721100128921358,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003290595574848161,
|
|
"loss": 5.4281,
|
|
"mean_token_accuracy": 0.18423805981874466,
|
|
"num_tokens": 9987543.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.813386249542236,
|
|
"epoch": 4.725397507520412,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003287367551018505,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.17934356182813643,
|
|
"num_tokens": 9996573.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.725397507520412,
|
|
"eval_entropy": 5.611860520667858,
|
|
"eval_loss": 6.019075393676758,
|
|
"eval_mean_token_accuracy": 0.1705198659486062,
|
|
"eval_num_tokens": 9996573.0,
|
|
"eval_runtime": 2.8198,
|
|
"eval_samples_per_second": 1258.597,
|
|
"eval_steps_per_second": 157.458,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.754193830490112,
|
|
"epoch": 4.729694886119467,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003284138353792903,
|
|
"loss": 5.3461,
|
|
"mean_token_accuracy": 0.18097725957632066,
|
|
"num_tokens": 10005954.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.64770359992981,
|
|
"epoch": 4.7339922647185215,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003280907990222628,
|
|
"loss": 5.305,
|
|
"mean_token_accuracy": 0.18924011439085006,
|
|
"num_tokens": 10014596.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.676145601272583,
|
|
"epoch": 4.738289643317576,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00032776764673615055,
|
|
"loss": 5.3786,
|
|
"mean_token_accuracy": 0.1847744882106781,
|
|
"num_tokens": 10023787.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.689545011520385,
|
|
"epoch": 4.742587021916631,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.0003274443792265888,
|
|
"loss": 5.358,
|
|
"mean_token_accuracy": 0.1837344542145729,
|
|
"num_tokens": 10031470.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.686351633071899,
|
|
"epoch": 4.746884400515685,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00032712099719946474,
|
|
"loss": 5.2844,
|
|
"mean_token_accuracy": 0.20021736919879912,
|
|
"num_tokens": 10039558.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.7209248542785645,
|
|
"epoch": 4.75118177911474,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00032679750136091533,
|
|
"loss": 5.3668,
|
|
"mean_token_accuracy": 0.18134486228227614,
|
|
"num_tokens": 10049244.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.704904174804687,
|
|
"epoch": 4.755479157713794,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003264738924173262,
|
|
"loss": 5.3619,
|
|
"mean_token_accuracy": 0.1795000821352005,
|
|
"num_tokens": 10058109.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.685872745513916,
|
|
"epoch": 4.759776536312849,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00032615017107533,
|
|
"loss": 5.3242,
|
|
"mean_token_accuracy": 0.1950099751353264,
|
|
"num_tokens": 10066824.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.652995443344116,
|
|
"epoch": 4.764073914911903,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003258263380418047,
|
|
"loss": 5.301,
|
|
"mean_token_accuracy": 0.20084263235330582,
|
|
"num_tokens": 10075399.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.736227750778198,
|
|
"epoch": 4.768371293510958,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032550239402387226,
|
|
"loss": 5.3276,
|
|
"mean_token_accuracy": 0.19647764712572097,
|
|
"num_tokens": 10083846.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.711696481704712,
|
|
"epoch": 4.772668672110013,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00032517833972889695,
|
|
"loss": 5.3413,
|
|
"mean_token_accuracy": 0.19022325575351715,
|
|
"num_tokens": 10092238.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.681711435317993,
|
|
"epoch": 4.776966050709067,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032485417586448375,
|
|
"loss": 5.3388,
|
|
"mean_token_accuracy": 0.19440205544233322,
|
|
"num_tokens": 10101195.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.732127952575683,
|
|
"epoch": 4.781263429308122,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000324529903138477,
|
|
"loss": 5.418,
|
|
"mean_token_accuracy": 0.18390129804611205,
|
|
"num_tokens": 10110226.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.650450992584228,
|
|
"epoch": 4.785560807907177,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003242055222589587,
|
|
"loss": 5.2979,
|
|
"mean_token_accuracy": 0.19801206439733504,
|
|
"num_tokens": 10119263.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.749388742446899,
|
|
"epoch": 4.7898581865062315,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000323881033934247,
|
|
"loss": 5.3358,
|
|
"mean_token_accuracy": 0.18700821250677108,
|
|
"num_tokens": 10128548.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.722526788711548,
|
|
"epoch": 4.794155565105286,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00032355643887289486,
|
|
"loss": 5.3103,
|
|
"mean_token_accuracy": 0.18732613176107407,
|
|
"num_tokens": 10137455.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.663701391220092,
|
|
"epoch": 4.798452943704341,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003232317377836881,
|
|
"loss": 5.3145,
|
|
"mean_token_accuracy": 0.19629506021738052,
|
|
"num_tokens": 10146469.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.685229349136352,
|
|
"epoch": 4.802750322303395,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.000322906931375644,
|
|
"loss": 5.3579,
|
|
"mean_token_accuracy": 0.1901021659374237,
|
|
"num_tokens": 10155898.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.708633041381836,
|
|
"epoch": 4.80704770090245,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00032258202035801,
|
|
"loss": 5.4213,
|
|
"mean_token_accuracy": 0.18825796395540237,
|
|
"num_tokens": 10165099.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.712029600143433,
|
|
"epoch": 4.811345079501504,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000322257005440262,
|
|
"loss": 5.3207,
|
|
"mean_token_accuracy": 0.19260724931955336,
|
|
"num_tokens": 10174240.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.684773683547974,
|
|
"epoch": 4.815642458100559,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003219318873321025,
|
|
"loss": 5.3934,
|
|
"mean_token_accuracy": 0.1948417142033577,
|
|
"num_tokens": 10183389.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.7410002708435055,
|
|
"epoch": 4.819939836699613,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00032160666674345954,
|
|
"loss": 5.3496,
|
|
"mean_token_accuracy": 0.19851730316877364,
|
|
"num_tokens": 10191749.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.6198793888092045,
|
|
"epoch": 4.824237215298668,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00032128134438448504,
|
|
"loss": 5.3568,
|
|
"mean_token_accuracy": 0.18850409984588623,
|
|
"num_tokens": 10200759.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.677628517150879,
|
|
"epoch": 4.828534593897722,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032095592096555284,
|
|
"loss": 5.3722,
|
|
"mean_token_accuracy": 0.18909865468740464,
|
|
"num_tokens": 10211141.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.800096559524536,
|
|
"epoch": 4.832831972496777,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003206303971972577,
|
|
"loss": 5.361,
|
|
"mean_token_accuracy": 0.18772295713424683,
|
|
"num_tokens": 10221412.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.70741834640503,
|
|
"epoch": 4.8371293510958315,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003203047737904134,
|
|
"loss": 5.3497,
|
|
"mean_token_accuracy": 0.19154001623392106,
|
|
"num_tokens": 10229894.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.7038774490356445,
|
|
"epoch": 4.841426729694886,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031997905145605135,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.17673951089382173,
|
|
"num_tokens": 10239636.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.750853061676025,
|
|
"epoch": 4.845724108293941,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00031965323090541874,
|
|
"loss": 5.3874,
|
|
"mean_token_accuracy": 0.19039532095193862,
|
|
"num_tokens": 10248694.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.762895965576172,
|
|
"epoch": 4.850021486892995,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003193273128499777,
|
|
"loss": 5.3585,
|
|
"mean_token_accuracy": 0.1840496689081192,
|
|
"num_tokens": 10258102.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.618077659606934,
|
|
"epoch": 4.85431886549205,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00031900129800140287,
|
|
"loss": 5.292,
|
|
"mean_token_accuracy": 0.1895531252026558,
|
|
"num_tokens": 10266752.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.627787446975708,
|
|
"epoch": 4.858616244091104,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00031867518707158027,
|
|
"loss": 5.3225,
|
|
"mean_token_accuracy": 0.18836573809385299,
|
|
"num_tokens": 10276199.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.73261833190918,
|
|
"epoch": 4.862913622690159,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000318348980772606,
|
|
"loss": 5.3333,
|
|
"mean_token_accuracy": 0.19033699631690978,
|
|
"num_tokens": 10285160.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.6856592178344725,
|
|
"epoch": 4.867211001289213,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031802267981678414,
|
|
"loss": 5.4164,
|
|
"mean_token_accuracy": 0.1895129695534706,
|
|
"num_tokens": 10295266.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.6361165046691895,
|
|
"epoch": 4.871508379888268,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031769628491662563,
|
|
"loss": 5.4527,
|
|
"mean_token_accuracy": 0.1891954943537712,
|
|
"num_tokens": 10305155.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.800771188735962,
|
|
"epoch": 4.8758057584873224,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00031736979678484634,
|
|
"loss": 5.5017,
|
|
"mean_token_accuracy": 0.17717777490615844,
|
|
"num_tokens": 10315052.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.8810230731964115,
|
|
"epoch": 4.880103137086377,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00031704321613436597,
|
|
"loss": 5.3839,
|
|
"mean_token_accuracy": 0.18088440150022506,
|
|
"num_tokens": 10324032.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.681092691421509,
|
|
"epoch": 4.8844005156854315,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003167165436783061,
|
|
"loss": 5.3745,
|
|
"mean_token_accuracy": 0.19031289517879485,
|
|
"num_tokens": 10333274.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.592399740219117,
|
|
"epoch": 4.888697894284486,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00031638978012998875,
|
|
"loss": 5.3472,
|
|
"mean_token_accuracy": 0.18405700474977493,
|
|
"num_tokens": 10341873.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.751446342468261,
|
|
"epoch": 4.892995272883541,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000316062926202935,
|
|
"loss": 5.3573,
|
|
"mean_token_accuracy": 0.1915464535355568,
|
|
"num_tokens": 10350579.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.711439514160157,
|
|
"epoch": 4.897292651482596,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003157359826108632,
|
|
"loss": 5.3888,
|
|
"mean_token_accuracy": 0.19404646307229995,
|
|
"num_tokens": 10359615.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.661752033233642,
|
|
"epoch": 4.90159003008165,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00031540895006768727,
|
|
"loss": 5.2951,
|
|
"mean_token_accuracy": 0.1917607694864273,
|
|
"num_tokens": 10367856.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.703397369384765,
|
|
"epoch": 4.905887408680705,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003150818292875158,
|
|
"loss": 5.3981,
|
|
"mean_token_accuracy": 0.1901896044611931,
|
|
"num_tokens": 10377471.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.758474588394165,
|
|
"epoch": 4.91018478727976,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003147546209846497,
|
|
"loss": 5.4325,
|
|
"mean_token_accuracy": 0.18345605134963988,
|
|
"num_tokens": 10387162.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.7711403369903564,
|
|
"epoch": 4.914482165878814,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0003144273258735812,
|
|
"loss": 5.3477,
|
|
"mean_token_accuracy": 0.19151308834552766,
|
|
"num_tokens": 10395995.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.75006070137024,
|
|
"epoch": 4.918779544477869,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003140999446689919,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.18897058814764023,
|
|
"num_tokens": 10405151.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.774607133865357,
|
|
"epoch": 4.923076923076923,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003137724780857516,
|
|
"loss": 5.3237,
|
|
"mean_token_accuracy": 0.19335650503635407,
|
|
"num_tokens": 10414811.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.7470215320587155,
|
|
"epoch": 4.927374301675978,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00031344492683891634,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.180099418759346,
|
|
"num_tokens": 10423822.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.5987835884094235,
|
|
"epoch": 4.931671680275032,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003131172916437272,
|
|
"loss": 5.2227,
|
|
"mean_token_accuracy": 0.2038162738084793,
|
|
"num_tokens": 10433014.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.662222051620484,
|
|
"epoch": 4.935969058874087,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031278957321560845,
|
|
"loss": 5.4094,
|
|
"mean_token_accuracy": 0.18789215236902237,
|
|
"num_tokens": 10442307.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.654838800430298,
|
|
"epoch": 4.9402664374731415,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00031246177227016615,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.19563332051038743,
|
|
"num_tokens": 10451780.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.770848369598388,
|
|
"epoch": 4.944563816072196,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00031213388952318653,
|
|
"loss": 5.3769,
|
|
"mean_token_accuracy": 0.18879292160272598,
|
|
"num_tokens": 10461441.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.726441955566406,
|
|
"epoch": 4.948861194671251,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 0.0003118059256906345,
|
|
"loss": 5.3602,
|
|
"mean_token_accuracy": 0.19396070688962935,
|
|
"num_tokens": 10470872.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.644023513793945,
|
|
"epoch": 4.953158573270305,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00031147788148865204,
|
|
"loss": 5.3263,
|
|
"mean_token_accuracy": 0.19643885940313338,
|
|
"num_tokens": 10480414.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.719807863235474,
|
|
"epoch": 4.95745595186936,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003111497576335564,
|
|
"loss": 5.4583,
|
|
"mean_token_accuracy": 0.18395932614803315,
|
|
"num_tokens": 10489837.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.729775428771973,
|
|
"epoch": 4.961753330468414,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003108215548418391,
|
|
"loss": 5.3859,
|
|
"mean_token_accuracy": 0.19093636125326158,
|
|
"num_tokens": 10498757.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.721962594985962,
|
|
"epoch": 4.966050709067469,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003104932738301637,
|
|
"loss": 5.3215,
|
|
"mean_token_accuracy": 0.19310247898101807,
|
|
"num_tokens": 10509057.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.771958589553833,
|
|
"epoch": 4.970348087666523,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00031016491531536477,
|
|
"loss": 5.4552,
|
|
"mean_token_accuracy": 0.18591174632310867,
|
|
"num_tokens": 10518207.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.665564250946045,
|
|
"epoch": 4.974645466265578,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003098364800144462,
|
|
"loss": 5.3996,
|
|
"mean_token_accuracy": 0.18913112133741378,
|
|
"num_tokens": 10527564.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.712208604812622,
|
|
"epoch": 4.9789428448646325,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003095079686445792,
|
|
"loss": 5.3376,
|
|
"mean_token_accuracy": 0.18524964153766632,
|
|
"num_tokens": 10536317.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.79870491027832,
|
|
"epoch": 4.983240223463687,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030917938192310146,
|
|
"loss": 5.4429,
|
|
"mean_token_accuracy": 0.17885127514600754,
|
|
"num_tokens": 10545451.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.732014751434326,
|
|
"epoch": 4.9875376020627415,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00030885072056751494,
|
|
"loss": 5.4222,
|
|
"mean_token_accuracy": 0.18293115198612214,
|
|
"num_tokens": 10554244.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.686966180801392,
|
|
"epoch": 4.991834980661796,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00030852198529548476,
|
|
"loss": 5.3861,
|
|
"mean_token_accuracy": 0.1893837794661522,
|
|
"num_tokens": 10562815.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.687817716598511,
|
|
"epoch": 4.996132359260851,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003081931768248373,
|
|
"loss": 5.3104,
|
|
"mean_token_accuracy": 0.20240730047225952,
|
|
"num_tokens": 10571136.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.694052113427056,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 0.0003078642958735588,
|
|
"loss": 5.281,
|
|
"mean_token_accuracy": 0.19216842121548122,
|
|
"num_tokens": 10579660.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.636717987060547,
|
|
"epoch": 5.0042973785990545,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00030753534315979393,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.20959500968456268,
|
|
"num_tokens": 10587594.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.651668214797974,
|
|
"epoch": 5.008594757198109,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0003072063194018438,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.19930772185325624,
|
|
"num_tokens": 10596694.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.711828994750976,
|
|
"epoch": 5.012892135797164,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003068772253181648,
|
|
"loss": 5.1189,
|
|
"mean_token_accuracy": 0.2065938577055931,
|
|
"num_tokens": 10605367.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.701521015167236,
|
|
"epoch": 5.017189514396218,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003065480616273671,
|
|
"loss": 5.1219,
|
|
"mean_token_accuracy": 0.19666921496391296,
|
|
"num_tokens": 10614094.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.72240686416626,
|
|
"epoch": 5.021486892995273,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003062188290482123,
|
|
"loss": 5.1721,
|
|
"mean_token_accuracy": 0.20151389688253402,
|
|
"num_tokens": 10624336.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.722569990158081,
|
|
"epoch": 5.025784271594327,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00030588952829961304,
|
|
"loss": 5.191,
|
|
"mean_token_accuracy": 0.1966630145907402,
|
|
"num_tokens": 10633461.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.672954797744751,
|
|
"epoch": 5.030081650193382,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003055601601006303,
|
|
"loss": 5.1555,
|
|
"mean_token_accuracy": 0.2011461764574051,
|
|
"num_tokens": 10641604.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.644674301147461,
|
|
"epoch": 5.034379028792436,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003052307251704728,
|
|
"loss": 5.169,
|
|
"mean_token_accuracy": 0.20120813697576523,
|
|
"num_tokens": 10650628.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.636267948150635,
|
|
"epoch": 5.038676407391491,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003049012242284946,
|
|
"loss": 5.0641,
|
|
"mean_token_accuracy": 0.20909030437469484,
|
|
"num_tokens": 10659887.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.766625213623047,
|
|
"epoch": 5.0429737859905455,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003045716579941941,
|
|
"loss": 5.2645,
|
|
"mean_token_accuracy": 0.19006333351135254,
|
|
"num_tokens": 10669401.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.697568321228028,
|
|
"epoch": 5.0472711645896,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00030424202718721215,
|
|
"loss": 5.145,
|
|
"mean_token_accuracy": 0.19573341459035873,
|
|
"num_tokens": 10676575.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.681671094894409,
|
|
"epoch": 5.051568543188655,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00030391233252733085,
|
|
"loss": 5.149,
|
|
"mean_token_accuracy": 0.20113655179738998,
|
|
"num_tokens": 10685190.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.7113268852233885,
|
|
"epoch": 5.055865921787709,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030358257473447144,
|
|
"loss": 5.2342,
|
|
"mean_token_accuracy": 0.1949695974588394,
|
|
"num_tokens": 10694143.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.675754833221435,
|
|
"epoch": 5.060163300386764,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030325275452869316,
|
|
"loss": 5.2206,
|
|
"mean_token_accuracy": 0.19502812772989273,
|
|
"num_tokens": 10703271.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.7130814552307125,
|
|
"epoch": 5.064460678985819,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00030292287263019153,
|
|
"loss": 5.2601,
|
|
"mean_token_accuracy": 0.18616272360086442,
|
|
"num_tokens": 10713526.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.7932346820831295,
|
|
"epoch": 5.068758057584874,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030259292975929675,
|
|
"loss": 5.2107,
|
|
"mean_token_accuracy": 0.1971692055463791,
|
|
"num_tokens": 10722685.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.720710277557373,
|
|
"epoch": 5.073055436183928,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003022629266364723,
|
|
"loss": 5.1419,
|
|
"mean_token_accuracy": 0.20456222593784332,
|
|
"num_tokens": 10731861.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.654185819625854,
|
|
"epoch": 5.077352814782983,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00030193286398231276,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.20370947271585466,
|
|
"num_tokens": 10740291.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.6524590969085695,
|
|
"epoch": 5.081650193382037,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030160274251754337,
|
|
"loss": 5.1319,
|
|
"mean_token_accuracy": 0.20393863916397095,
|
|
"num_tokens": 10749340.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.694770383834839,
|
|
"epoch": 5.085947571981092,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00030127256296301724,
|
|
"loss": 5.135,
|
|
"mean_token_accuracy": 0.20143632590770721,
|
|
"num_tokens": 10758948.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.708119010925293,
|
|
"epoch": 5.090244950580146,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003009423260397148,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.18842663317918779,
|
|
"num_tokens": 10768534.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.70088415145874,
|
|
"epoch": 5.094542329179201,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00030061203246874125,
|
|
"loss": 5.1411,
|
|
"mean_token_accuracy": 0.2087824136018753,
|
|
"num_tokens": 10777570.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.721751642227173,
|
|
"epoch": 5.0988397077782555,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00030028168297132593,
|
|
"loss": 5.1912,
|
|
"mean_token_accuracy": 0.20279961675405503,
|
|
"num_tokens": 10786952.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.717559576034546,
|
|
"epoch": 5.10313708637731,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002999512782688199,
|
|
"loss": 5.2532,
|
|
"mean_token_accuracy": 0.19311114996671677,
|
|
"num_tokens": 10795401.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.612150096893311,
|
|
"epoch": 5.1074344649763646,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 0.0002996208190826951,
|
|
"loss": 5.1404,
|
|
"mean_token_accuracy": 0.20555511862039566,
|
|
"num_tokens": 10805459.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.677275943756103,
|
|
"epoch": 5.111731843575419,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00029929030613454227,
|
|
"loss": 5.1793,
|
|
"mean_token_accuracy": 0.19179816097021102,
|
|
"num_tokens": 10814635.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.740015602111816,
|
|
"epoch": 5.116029222174474,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002989597401460697,
|
|
"loss": 5.2383,
|
|
"mean_token_accuracy": 0.19447495490312577,
|
|
"num_tokens": 10824256.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.682909250259399,
|
|
"epoch": 5.120326600773528,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00029862912183910105,
|
|
"loss": 5.1741,
|
|
"mean_token_accuracy": 0.19891832917928695,
|
|
"num_tokens": 10833919.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.645577001571655,
|
|
"epoch": 5.124623979372583,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00029829845193557496,
|
|
"loss": 5.1,
|
|
"mean_token_accuracy": 0.20446018427610396,
|
|
"num_tokens": 10842109.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.729709148406982,
|
|
"epoch": 5.128921357971637,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002979677311575421,
|
|
"loss": 5.2971,
|
|
"mean_token_accuracy": 0.19588488936424256,
|
|
"num_tokens": 10851185.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.628516674041748,
|
|
"epoch": 5.133218736570692,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002976369602271646,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.21069692969322204,
|
|
"num_tokens": 10859822.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.6155538082122805,
|
|
"epoch": 5.137516115169746,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002973061398667138,
|
|
"loss": 5.1491,
|
|
"mean_token_accuracy": 0.2032279059290886,
|
|
"num_tokens": 10868663.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.686222505569458,
|
|
"epoch": 5.141813493768801,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00029697527079856916,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.20866820216178894,
|
|
"num_tokens": 10876888.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.7373316287994385,
|
|
"epoch": 5.1461108723678555,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00029664435374521665,
|
|
"loss": 5.2344,
|
|
"mean_token_accuracy": 0.19479000568389893,
|
|
"num_tokens": 10885559.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.690717077255249,
|
|
"epoch": 5.15040825096691,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00029631338942924664,
|
|
"loss": 5.2184,
|
|
"mean_token_accuracy": 0.2004997491836548,
|
|
"num_tokens": 10893843.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.631839752197266,
|
|
"epoch": 5.154705629565965,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0002959823785733531,
|
|
"loss": 5.1568,
|
|
"mean_token_accuracy": 0.20057498812675476,
|
|
"num_tokens": 10903272.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.154705629565965,
|
|
"eval_entropy": 5.535785226134567,
|
|
"eval_loss": 6.002467632293701,
|
|
"eval_mean_token_accuracy": 0.17269592760718083,
|
|
"eval_num_tokens": 10903272.0,
|
|
"eval_runtime": 2.6512,
|
|
"eval_samples_per_second": 1338.65,
|
|
"eval_steps_per_second": 167.473,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.649084901809692,
|
|
"epoch": 5.159003008165019,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002956513219003312,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.19946276843547822,
|
|
"num_tokens": 10912630.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.700127983093262,
|
|
"epoch": 5.163300386764074,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029532022013307666,
|
|
"loss": 5.0771,
|
|
"mean_token_accuracy": 0.20625242888927459,
|
|
"num_tokens": 10921491.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.696587753295899,
|
|
"epoch": 5.167597765363128,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00029498907399458325,
|
|
"loss": 5.2006,
|
|
"mean_token_accuracy": 0.19744562804698945,
|
|
"num_tokens": 10931686.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.70154128074646,
|
|
"epoch": 5.171895143962183,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002946578842079418,
|
|
"loss": 5.2994,
|
|
"mean_token_accuracy": 0.1840354785323143,
|
|
"num_tokens": 10942273.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.704367303848267,
|
|
"epoch": 5.176192522561237,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002943266514963384,
|
|
"loss": 5.2671,
|
|
"mean_token_accuracy": 0.20017221421003342,
|
|
"num_tokens": 10950736.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.727904987335205,
|
|
"epoch": 5.180489901160292,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002939953765830529,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.1977004736661911,
|
|
"num_tokens": 10959789.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.678816652297973,
|
|
"epoch": 5.184787279759346,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029366406019145735,
|
|
"loss": 5.1742,
|
|
"mean_token_accuracy": 0.19149168580770493,
|
|
"num_tokens": 10968842.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.692221307754517,
|
|
"epoch": 5.189084658358402,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002933327030450143,
|
|
"loss": 5.1888,
|
|
"mean_token_accuracy": 0.19528967291116714,
|
|
"num_tokens": 10977799.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.712386417388916,
|
|
"epoch": 5.193382036957456,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029300130586727545,
|
|
"loss": 5.2405,
|
|
"mean_token_accuracy": 0.1896895572543144,
|
|
"num_tokens": 10987279.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.67851037979126,
|
|
"epoch": 5.197679415556511,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00029266986938187943,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.20856768935918807,
|
|
"num_tokens": 10996577.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.680726337432861,
|
|
"epoch": 5.2019767941555655,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002923383943125514,
|
|
"loss": 5.2664,
|
|
"mean_token_accuracy": 0.19332842230796815,
|
|
"num_tokens": 11005818.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.6977214336395265,
|
|
"epoch": 5.20627417275462,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002920068813831002,
|
|
"loss": 5.1186,
|
|
"mean_token_accuracy": 0.20581428557634354,
|
|
"num_tokens": 11014613.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.628753423690796,
|
|
"epoch": 5.210571551353675,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002916753313174178,
|
|
"loss": 5.2046,
|
|
"mean_token_accuracy": 0.19042902439832687,
|
|
"num_tokens": 11024327.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.643909311294555,
|
|
"epoch": 5.214868929952729,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002913437448394768,
|
|
"loss": 5.2954,
|
|
"mean_token_accuracy": 0.19764266014099122,
|
|
"num_tokens": 11034558.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.621911334991455,
|
|
"epoch": 5.219166308551784,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00029101212267332955,
|
|
"loss": 5.0866,
|
|
"mean_token_accuracy": 0.20859397649765016,
|
|
"num_tokens": 11042948.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.71375994682312,
|
|
"epoch": 5.223463687150838,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00029068046554310637,
|
|
"loss": 5.2041,
|
|
"mean_token_accuracy": 0.19912176877260207,
|
|
"num_tokens": 11052529.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.725802087783814,
|
|
"epoch": 5.227761065749893,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002903487741730139,
|
|
"loss": 5.2802,
|
|
"mean_token_accuracy": 0.18881526887416838,
|
|
"num_tokens": 11062464.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.697055673599243,
|
|
"epoch": 5.232058444348947,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00029001704928733354,
|
|
"loss": 5.1329,
|
|
"mean_token_accuracy": 0.20530512779951096,
|
|
"num_tokens": 11070522.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.618717765808105,
|
|
"epoch": 5.236355822948002,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002896852916104198,
|
|
"loss": 5.1022,
|
|
"mean_token_accuracy": 0.20654451847076416,
|
|
"num_tokens": 11080227.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.710249280929565,
|
|
"epoch": 5.240653201547056,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002893535018666988,
|
|
"loss": 5.2385,
|
|
"mean_token_accuracy": 0.19757603704929352,
|
|
"num_tokens": 11089040.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.709861898422242,
|
|
"epoch": 5.244950580146111,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00028902168078066674,
|
|
"loss": 5.2474,
|
|
"mean_token_accuracy": 0.1964607760310173,
|
|
"num_tokens": 11098788.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.720152378082275,
|
|
"epoch": 5.2492479587451655,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002886898290768883,
|
|
"loss": 5.1238,
|
|
"mean_token_accuracy": 0.20276738256216048,
|
|
"num_tokens": 11107399.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.639698886871338,
|
|
"epoch": 5.25354533734422,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000288357947479995,
|
|
"loss": 5.2127,
|
|
"mean_token_accuracy": 0.1937815859913826,
|
|
"num_tokens": 11116833.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.662129735946655,
|
|
"epoch": 5.257842715943275,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00028802603671468347,
|
|
"loss": 5.202,
|
|
"mean_token_accuracy": 0.19818273931741714,
|
|
"num_tokens": 11126027.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.717524576187134,
|
|
"epoch": 5.262140094542329,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00028769409750571413,
|
|
"loss": 5.2526,
|
|
"mean_token_accuracy": 0.18824333399534227,
|
|
"num_tokens": 11135873.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.619460201263427,
|
|
"epoch": 5.266437473141384,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00028736213057790975,
|
|
"loss": 5.0093,
|
|
"mean_token_accuracy": 0.2108587771654129,
|
|
"num_tokens": 11144470.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.666223812103271,
|
|
"epoch": 5.270734851740438,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002870301366561533,
|
|
"loss": 5.1886,
|
|
"mean_token_accuracy": 0.2039417341351509,
|
|
"num_tokens": 11153531.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.717132568359375,
|
|
"epoch": 5.275032230339493,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002866981164653867,
|
|
"loss": 5.2321,
|
|
"mean_token_accuracy": 0.19243954867124557,
|
|
"num_tokens": 11163365.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.646639490127564,
|
|
"epoch": 5.279329608938547,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002863660707306095,
|
|
"loss": 5.1142,
|
|
"mean_token_accuracy": 0.19977502822875975,
|
|
"num_tokens": 11171907.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.626325416564941,
|
|
"epoch": 5.283626987537602,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00028603400017687675,
|
|
"loss": 5.1911,
|
|
"mean_token_accuracy": 0.19853340089321136,
|
|
"num_tokens": 11180804.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.804350757598877,
|
|
"epoch": 5.287924366136656,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00028570190552929794,
|
|
"loss": 5.2537,
|
|
"mean_token_accuracy": 0.19246581792831421,
|
|
"num_tokens": 11189762.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.68079137802124,
|
|
"epoch": 5.292221744735711,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000285369787513035,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.1947798416018486,
|
|
"num_tokens": 11197983.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.646259784698486,
|
|
"epoch": 5.2965191233347655,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00028503764685330077,
|
|
"loss": 5.0951,
|
|
"mean_token_accuracy": 0.21035149097442626,
|
|
"num_tokens": 11206286.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.67607889175415,
|
|
"epoch": 5.30081650193382,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028470548427535794,
|
|
"loss": 5.2301,
|
|
"mean_token_accuracy": 0.20003851503133774,
|
|
"num_tokens": 11216444.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.651372385025025,
|
|
"epoch": 5.305113880532875,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00028437330050451654,
|
|
"loss": 5.1444,
|
|
"mean_token_accuracy": 0.20452726632356644,
|
|
"num_tokens": 11225378.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.68706636428833,
|
|
"epoch": 5.309411259131929,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002840410962661334,
|
|
"loss": 5.2729,
|
|
"mean_token_accuracy": 0.18928681463003158,
|
|
"num_tokens": 11234352.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.659502983093262,
|
|
"epoch": 5.313708637730985,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002837088722856098,
|
|
"loss": 5.1822,
|
|
"mean_token_accuracy": 0.20295837819576262,
|
|
"num_tokens": 11243350.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.695145034790039,
|
|
"epoch": 5.318006016330038,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00028337662928838996,
|
|
"loss": 5.2145,
|
|
"mean_token_accuracy": 0.19444998800754548,
|
|
"num_tokens": 11252746.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.700344038009644,
|
|
"epoch": 5.322303394929094,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028304436799995986,
|
|
"loss": 5.1535,
|
|
"mean_token_accuracy": 0.1991518408060074,
|
|
"num_tokens": 11262905.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.683503198623657,
|
|
"epoch": 5.326600773528148,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00028271208914584534,
|
|
"loss": 5.1735,
|
|
"mean_token_accuracy": 0.20304387360811232,
|
|
"num_tokens": 11270925.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.6509486675262455,
|
|
"epoch": 5.330898152127203,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028237979345161065,
|
|
"loss": 5.1855,
|
|
"mean_token_accuracy": 0.19876890927553176,
|
|
"num_tokens": 11279689.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.593795824050903,
|
|
"epoch": 5.335195530726257,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002820474816428568,
|
|
"loss": 5.1251,
|
|
"mean_token_accuracy": 0.21047322303056717,
|
|
"num_tokens": 11288783.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.556586408615113,
|
|
"epoch": 5.339492909325312,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002817151544452198,
|
|
"loss": 5.0849,
|
|
"mean_token_accuracy": 0.2196614682674408,
|
|
"num_tokens": 11297903.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.6540771484375,
|
|
"epoch": 5.343790287924366,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00028138281258436947,
|
|
"loss": 5.1389,
|
|
"mean_token_accuracy": 0.21052396446466445,
|
|
"num_tokens": 11306591.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.66905460357666,
|
|
"epoch": 5.348087666523421,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002810504567860078,
|
|
"loss": 5.2087,
|
|
"mean_token_accuracy": 0.2011790990829468,
|
|
"num_tokens": 11316772.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.6996422290802,
|
|
"epoch": 5.3523850451224755,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002807180877758667,
|
|
"loss": 5.1918,
|
|
"mean_token_accuracy": 0.2027275264263153,
|
|
"num_tokens": 11326842.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.631223583221436,
|
|
"epoch": 5.35668242372153,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00028038570627970754,
|
|
"loss": 5.2208,
|
|
"mean_token_accuracy": 0.20187463313341142,
|
|
"num_tokens": 11335765.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.703919219970703,
|
|
"epoch": 5.360979802320585,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 0.0002800533130233184,
|
|
"loss": 5.2645,
|
|
"mean_token_accuracy": 0.19333812594413757,
|
|
"num_tokens": 11345140.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.675189447402954,
|
|
"epoch": 5.365277180919639,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002797209087325135,
|
|
"loss": 5.1372,
|
|
"mean_token_accuracy": 0.20060636550188066,
|
|
"num_tokens": 11354585.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.7378476619720455,
|
|
"epoch": 5.369574559518694,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027938849413313083,
|
|
"loss": 5.3251,
|
|
"mean_token_accuracy": 0.19005185067653657,
|
|
"num_tokens": 11364606.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.650345754623413,
|
|
"epoch": 5.373871938117748,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000279056069951031,
|
|
"loss": 5.2421,
|
|
"mean_token_accuracy": 0.1923617973923683,
|
|
"num_tokens": 11374399.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.664040231704712,
|
|
"epoch": 5.378169316716803,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00027872363691209564,
|
|
"loss": 5.0985,
|
|
"mean_token_accuracy": 0.21121095418930053,
|
|
"num_tokens": 11383040.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.6200624942779545,
|
|
"epoch": 5.382466695315857,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002783911957422256,
|
|
"loss": 5.2387,
|
|
"mean_token_accuracy": 0.19980725646018982,
|
|
"num_tokens": 11392088.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.644261455535888,
|
|
"epoch": 5.386764073914912,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002780587471673394,
|
|
"loss": 5.1813,
|
|
"mean_token_accuracy": 0.19385674297809602,
|
|
"num_tokens": 11401400.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.619362211227417,
|
|
"epoch": 5.391061452513966,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00027772629191337206,
|
|
"loss": 5.1767,
|
|
"mean_token_accuracy": 0.20414769202470778,
|
|
"num_tokens": 11410205.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.7334969997406,
|
|
"epoch": 5.395358831113021,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027739383070627283,
|
|
"loss": 5.1726,
|
|
"mean_token_accuracy": 0.20797506123781204,
|
|
"num_tokens": 11419208.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.568106842041016,
|
|
"epoch": 5.3996562097120755,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002770613642720041,
|
|
"loss": 5.0142,
|
|
"mean_token_accuracy": 0.22051802426576614,
|
|
"num_tokens": 11427386.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.529188442230224,
|
|
"epoch": 5.40395358831113,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00027672889333653984,
|
|
"loss": 5.1557,
|
|
"mean_token_accuracy": 0.20316692143678666,
|
|
"num_tokens": 11436203.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.643898248672485,
|
|
"epoch": 5.408250966910185,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002763964186258635,
|
|
"loss": 5.2422,
|
|
"mean_token_accuracy": 0.19205624908208846,
|
|
"num_tokens": 11445682.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.6721007347106935,
|
|
"epoch": 5.412548345509239,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0002760639408659671,
|
|
"loss": 5.1532,
|
|
"mean_token_accuracy": 0.201337830722332,
|
|
"num_tokens": 11454622.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.601822185516357,
|
|
"epoch": 5.416845724108294,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002757314607828489,
|
|
"loss": 5.1233,
|
|
"mean_token_accuracy": 0.20339442640542985,
|
|
"num_tokens": 11463539.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.637900400161743,
|
|
"epoch": 5.421143102707348,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00027539897910251293,
|
|
"loss": 5.1203,
|
|
"mean_token_accuracy": 0.20832913517951965,
|
|
"num_tokens": 11472491.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.559300327301026,
|
|
"epoch": 5.425440481306403,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00027506649655096595,
|
|
"loss": 5.1855,
|
|
"mean_token_accuracy": 0.20213525593280793,
|
|
"num_tokens": 11480899.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.640084934234619,
|
|
"epoch": 5.429737859905457,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002747340138542171,
|
|
"loss": 5.165,
|
|
"mean_token_accuracy": 0.19372519552707673,
|
|
"num_tokens": 11490525.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.592509746551514,
|
|
"epoch": 5.434035238504512,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002744015317382757,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.19918297082185746,
|
|
"num_tokens": 11499589.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.702159214019775,
|
|
"epoch": 5.438332617103566,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002740690509291498,
|
|
"loss": 5.2453,
|
|
"mean_token_accuracy": 0.19494837671518325,
|
|
"num_tokens": 11508425.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.685543727874756,
|
|
"epoch": 5.442629995702621,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002737365721528445,
|
|
"loss": 5.2412,
|
|
"mean_token_accuracy": 0.19615920782089233,
|
|
"num_tokens": 11517723.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.60957145690918,
|
|
"epoch": 5.446927374301676,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002734040961353607,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.20812844187021257,
|
|
"num_tokens": 11525882.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.6621442317962645,
|
|
"epoch": 5.451224752900731,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000273071623602693,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.1919061541557312,
|
|
"num_tokens": 11535646.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.657871866226197,
|
|
"epoch": 5.4555221314997855,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027273915528082865,
|
|
"loss": 5.1163,
|
|
"mean_token_accuracy": 0.19977849572896958,
|
|
"num_tokens": 11545413.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.58711576461792,
|
|
"epoch": 5.45981951009884,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002724066918957455,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.21105781346559524,
|
|
"num_tokens": 11554680.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.699238252639771,
|
|
"epoch": 5.464116888697895,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002720742341734107,
|
|
"loss": 5.2649,
|
|
"mean_token_accuracy": 0.19819252490997313,
|
|
"num_tokens": 11564607.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.592772150039673,
|
|
"epoch": 5.468414267296949,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027174178283977904,
|
|
"loss": 5.219,
|
|
"mean_token_accuracy": 0.19916775524616243,
|
|
"num_tokens": 11574097.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.622210311889648,
|
|
"epoch": 5.472711645896004,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00027140933862079136,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.20623194724321364,
|
|
"num_tokens": 11582561.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.5986252784729,
|
|
"epoch": 5.477009024495058,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000271076902242373,
|
|
"loss": 5.2148,
|
|
"mean_token_accuracy": 0.19745754301548005,
|
|
"num_tokens": 11591747.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.672637510299682,
|
|
"epoch": 5.481306403094113,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000270744474430432,
|
|
"loss": 5.2185,
|
|
"mean_token_accuracy": 0.19644926339387894,
|
|
"num_tokens": 11599928.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.618949937820434,
|
|
"epoch": 5.485603781693167,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000270412055910858,
|
|
"loss": 5.1405,
|
|
"mean_token_accuracy": 0.2050392061471939,
|
|
"num_tokens": 11607912.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.647641658782959,
|
|
"epoch": 5.489901160292222,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002700796474095201,
|
|
"loss": 5.2561,
|
|
"mean_token_accuracy": 0.19126703441143036,
|
|
"num_tokens": 11617407.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.600618028640747,
|
|
"epoch": 5.494198538891276,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002697472496522656,
|
|
"loss": 5.1032,
|
|
"mean_token_accuracy": 0.20891058593988418,
|
|
"num_tokens": 11626280.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.669522333145141,
|
|
"epoch": 5.498495917490331,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002694148633649184,
|
|
"loss": 5.1497,
|
|
"mean_token_accuracy": 0.2001833662390709,
|
|
"num_tokens": 11634959.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.662327671051026,
|
|
"epoch": 5.5027932960893855,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002690824892732772,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.1971360847353935,
|
|
"num_tokens": 11644044.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.648580121994018,
|
|
"epoch": 5.50709067468844,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002687501281031142,
|
|
"loss": 5.1394,
|
|
"mean_token_accuracy": 0.2053503856062889,
|
|
"num_tokens": 11653622.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.725681495666504,
|
|
"epoch": 5.511388053287495,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002684177805801734,
|
|
"loss": 5.2783,
|
|
"mean_token_accuracy": 0.1839623287320137,
|
|
"num_tokens": 11663716.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.743385267257691,
|
|
"epoch": 5.515685431886549,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00026808544743016886,
|
|
"loss": 5.3195,
|
|
"mean_token_accuracy": 0.19182008057832717,
|
|
"num_tokens": 11672769.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.69994592666626,
|
|
"epoch": 5.519982810485604,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002677531293787835,
|
|
"loss": 5.2283,
|
|
"mean_token_accuracy": 0.20069448202848433,
|
|
"num_tokens": 11682492.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.608594942092895,
|
|
"epoch": 5.524280189084658,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000267420827151667,
|
|
"loss": 5.1682,
|
|
"mean_token_accuracy": 0.20189118534326553,
|
|
"num_tokens": 11690738.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.538219976425171,
|
|
"epoch": 5.528577567683713,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 0.0002670885414744347,
|
|
"loss": 5.0926,
|
|
"mean_token_accuracy": 0.20586565285921096,
|
|
"num_tokens": 11699925.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.657249593734742,
|
|
"epoch": 5.532874946282767,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002667562730726655,
|
|
"loss": 5.1693,
|
|
"mean_token_accuracy": 0.19973579943180084,
|
|
"num_tokens": 11709291.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.665799808502197,
|
|
"epoch": 5.537172324881822,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026642402267190095,
|
|
"loss": 5.207,
|
|
"mean_token_accuracy": 0.20679962038993835,
|
|
"num_tokens": 11717958.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.645871591567993,
|
|
"epoch": 5.541469703480876,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026609179099764313,
|
|
"loss": 5.2004,
|
|
"mean_token_accuracy": 0.19209784865379334,
|
|
"num_tokens": 11726776.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.62501859664917,
|
|
"epoch": 5.545767082079931,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026575957877535323,
|
|
"loss": 5.1724,
|
|
"mean_token_accuracy": 0.20808785855770112,
|
|
"num_tokens": 11735913.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.720390892028808,
|
|
"epoch": 5.5500644606789855,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00026542738673044985,
|
|
"loss": 5.1989,
|
|
"mean_token_accuracy": 0.1987552285194397,
|
|
"num_tokens": 11744728.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.716229724884033,
|
|
"epoch": 5.55436183927804,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002650952155883077,
|
|
"loss": 5.2247,
|
|
"mean_token_accuracy": 0.19197028279304504,
|
|
"num_tokens": 11753440.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.624403810501098,
|
|
"epoch": 5.558659217877095,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.0002647630660742559,
|
|
"loss": 5.1601,
|
|
"mean_token_accuracy": 0.2036215618252754,
|
|
"num_tokens": 11762061.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.669667339324951,
|
|
"epoch": 5.56295659647615,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000264430938913576,
|
|
"loss": 5.2475,
|
|
"mean_token_accuracy": 0.19204566925764083,
|
|
"num_tokens": 11771421.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.580520868301392,
|
|
"epoch": 5.567253975075204,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00026409883483150123,
|
|
"loss": 5.1256,
|
|
"mean_token_accuracy": 0.2094772517681122,
|
|
"num_tokens": 11781062.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.684006690979004,
|
|
"epoch": 5.571551353674259,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000263766754553214,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.1969390332698822,
|
|
"num_tokens": 11790188.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.656520175933838,
|
|
"epoch": 5.575848732273313,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002634346988038448,
|
|
"loss": 5.21,
|
|
"mean_token_accuracy": 0.19641240388154985,
|
|
"num_tokens": 11799093.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.687589693069458,
|
|
"epoch": 5.580146110872368,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026310266830847093,
|
|
"loss": 5.1521,
|
|
"mean_token_accuracy": 0.2081759601831436,
|
|
"num_tokens": 11808357.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.6259664535522464,
|
|
"epoch": 5.584443489471423,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026277066379211406,
|
|
"loss": 5.1754,
|
|
"mean_token_accuracy": 0.20285745561122895,
|
|
"num_tokens": 11817229.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.584443489471423,
|
|
"eval_entropy": 5.532415933437176,
|
|
"eval_loss": 5.985991954803467,
|
|
"eval_mean_token_accuracy": 0.17413052456738712,
|
|
"eval_num_tokens": 11817229.0,
|
|
"eval_runtime": 2.6399,
|
|
"eval_samples_per_second": 1344.357,
|
|
"eval_steps_per_second": 168.187,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.653585147857666,
|
|
"epoch": 5.588740868070477,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002624386859797396,
|
|
"loss": 5.2292,
|
|
"mean_token_accuracy": 0.19886101633310319,
|
|
"num_tokens": 11826952.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.6561119079589846,
|
|
"epoch": 5.593038246669532,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026210673559625406,
|
|
"loss": 5.2051,
|
|
"mean_token_accuracy": 0.20010110437870027,
|
|
"num_tokens": 11836411.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.704966688156128,
|
|
"epoch": 5.597335625268586,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002617748133665047,
|
|
"loss": 5.2628,
|
|
"mean_token_accuracy": 0.1965386301279068,
|
|
"num_tokens": 11846140.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.7157862186431885,
|
|
"epoch": 5.601633003867641,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002614429200152768,
|
|
"loss": 5.1945,
|
|
"mean_token_accuracy": 0.19961169362068176,
|
|
"num_tokens": 11855057.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.711543893814087,
|
|
"epoch": 5.6059303824666955,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000261111056267293,
|
|
"loss": 5.1253,
|
|
"mean_token_accuracy": 0.20506765991449355,
|
|
"num_tokens": 11863257.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.686924171447754,
|
|
"epoch": 5.61022776106575,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00026077922284721084,
|
|
"loss": 5.2786,
|
|
"mean_token_accuracy": 0.19427892863750457,
|
|
"num_tokens": 11872706.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.586248731613159,
|
|
"epoch": 5.614525139664805,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026044742047962206,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.2048526093363762,
|
|
"num_tokens": 11882692.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.655636501312256,
|
|
"epoch": 5.618822518263859,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026011564988905023,
|
|
"loss": 5.1575,
|
|
"mean_token_accuracy": 0.20574906170368196,
|
|
"num_tokens": 11893152.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.76156268119812,
|
|
"epoch": 5.623119896862914,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002597839117999499,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.19877577275037767,
|
|
"num_tokens": 11902193.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.668945932388306,
|
|
"epoch": 5.627417275461968,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002594522069367044,
|
|
"loss": 5.1903,
|
|
"mean_token_accuracy": 0.19848927557468415,
|
|
"num_tokens": 11910909.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.593928956985474,
|
|
"epoch": 5.631714654061023,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002591205360236245,
|
|
"loss": 5.2036,
|
|
"mean_token_accuracy": 0.20328882932662964,
|
|
"num_tokens": 11919518.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.640859842300415,
|
|
"epoch": 5.636012032660077,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000258788899784947,
|
|
"loss": 5.1994,
|
|
"mean_token_accuracy": 0.1996496945619583,
|
|
"num_tokens": 11928735.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.644925546646118,
|
|
"epoch": 5.640309411259132,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00025845729894483283,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.2066833034157753,
|
|
"num_tokens": 11937266.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.576916265487671,
|
|
"epoch": 5.644606789858186,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002581257342273657,
|
|
"loss": 5.1278,
|
|
"mean_token_accuracy": 0.2007118970155716,
|
|
"num_tokens": 11946583.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.694371747970581,
|
|
"epoch": 5.648904168457241,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002577942063565504,
|
|
"loss": 5.2141,
|
|
"mean_token_accuracy": 0.19705425351858138,
|
|
"num_tokens": 11955883.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.731935834884643,
|
|
"epoch": 5.6532015470562955,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002574627160563114,
|
|
"loss": 5.1733,
|
|
"mean_token_accuracy": 0.2075083076953888,
|
|
"num_tokens": 11965340.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.595839357376098,
|
|
"epoch": 5.65749892565535,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 0.0002571312640504909,
|
|
"loss": 5.102,
|
|
"mean_token_accuracy": 0.2094941884279251,
|
|
"num_tokens": 11974389.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.658647584915161,
|
|
"epoch": 5.661796304254405,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002567998510628476,
|
|
"loss": 5.27,
|
|
"mean_token_accuracy": 0.19054170697927475,
|
|
"num_tokens": 11983872.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.662417650222778,
|
|
"epoch": 5.666093682853459,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025646847781705506,
|
|
"loss": 5.1273,
|
|
"mean_token_accuracy": 0.2063935026526451,
|
|
"num_tokens": 11992318.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.6555342197418215,
|
|
"epoch": 5.670391061452514,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002561371450367,
|
|
"loss": 5.1742,
|
|
"mean_token_accuracy": 0.20117734372615814,
|
|
"num_tokens": 12001926.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.663101577758789,
|
|
"epoch": 5.674688440051568,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00025580585344528076,
|
|
"loss": 5.2748,
|
|
"mean_token_accuracy": 0.19326671808958054,
|
|
"num_tokens": 12011505.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.647570562362671,
|
|
"epoch": 5.678985818650623,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002554746037662058,
|
|
"loss": 5.1243,
|
|
"mean_token_accuracy": 0.20442164987325667,
|
|
"num_tokens": 12020969.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.681697082519531,
|
|
"epoch": 5.683283197249677,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002551433967227919,
|
|
"loss": 5.1944,
|
|
"mean_token_accuracy": 0.20292714536190032,
|
|
"num_tokens": 12029786.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.7089861869812015,
|
|
"epoch": 5.687580575848732,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000254812233038263,
|
|
"loss": 5.2898,
|
|
"mean_token_accuracy": 0.19138347506523132,
|
|
"num_tokens": 12039081.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.647166347503662,
|
|
"epoch": 5.691877954447786,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00025448111343574813,
|
|
"loss": 5.2112,
|
|
"mean_token_accuracy": 0.20239898562431335,
|
|
"num_tokens": 12049129.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.645027160644531,
|
|
"epoch": 5.696175333046842,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0002541500386382802,
|
|
"loss": 5.0399,
|
|
"mean_token_accuracy": 0.22197454124689103,
|
|
"num_tokens": 12058243.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.606950092315674,
|
|
"epoch": 5.7004727116458955,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00025381900936879433,
|
|
"loss": 5.054,
|
|
"mean_token_accuracy": 0.21612686663866043,
|
|
"num_tokens": 12067468.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.665207433700561,
|
|
"epoch": 5.704770090244951,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0002534880263501259,
|
|
"loss": 5.1797,
|
|
"mean_token_accuracy": 0.19831924736499787,
|
|
"num_tokens": 12076090.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.683203172683716,
|
|
"epoch": 5.7090674688440055,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002531570903050097,
|
|
"loss": 5.2533,
|
|
"mean_token_accuracy": 0.19505767226219178,
|
|
"num_tokens": 12085614.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.623214435577393,
|
|
"epoch": 5.71336484744306,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002528262019560776,
|
|
"loss": 5.1587,
|
|
"mean_token_accuracy": 0.20216239839792252,
|
|
"num_tokens": 12094291.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.677948665618897,
|
|
"epoch": 5.717662226042115,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002524953620258579,
|
|
"loss": 5.1935,
|
|
"mean_token_accuracy": 0.199533911049366,
|
|
"num_tokens": 12102478.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.710334205627442,
|
|
"epoch": 5.721959604641169,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002521645712367724,
|
|
"loss": 5.3202,
|
|
"mean_token_accuracy": 0.1888586938381195,
|
|
"num_tokens": 12112554.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.731233024597168,
|
|
"epoch": 5.726256983240224,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025183383031113606,
|
|
"loss": 5.2571,
|
|
"mean_token_accuracy": 0.1966889277100563,
|
|
"num_tokens": 12121685.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.730410528182984,
|
|
"epoch": 5.730554361839278,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025150313997115476,
|
|
"loss": 5.2525,
|
|
"mean_token_accuracy": 0.19456737190485002,
|
|
"num_tokens": 12131980.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.682445621490478,
|
|
"epoch": 5.734851740438333,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002511725009389244,
|
|
"loss": 5.2207,
|
|
"mean_token_accuracy": 0.2021138682961464,
|
|
"num_tokens": 12141457.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.752025127410889,
|
|
"epoch": 5.739149119037387,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000250841913936428,
|
|
"loss": 5.2585,
|
|
"mean_token_accuracy": 0.19461566507816314,
|
|
"num_tokens": 12149803.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.71280083656311,
|
|
"epoch": 5.743446497636442,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002505113796855357,
|
|
"loss": 5.1339,
|
|
"mean_token_accuracy": 0.20337906032800673,
|
|
"num_tokens": 12157904.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.617925786972046,
|
|
"epoch": 5.747743876235496,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00025018089890800225,
|
|
"loss": 5.1682,
|
|
"mean_token_accuracy": 0.20402243435382844,
|
|
"num_tokens": 12166540.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.585759210586548,
|
|
"epoch": 5.752041254834551,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024985047232546544,
|
|
"loss": 5.1803,
|
|
"mean_token_accuracy": 0.20675259977579116,
|
|
"num_tokens": 12175345.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.712131977081299,
|
|
"epoch": 5.7563386334336055,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024952010065944485,
|
|
"loss": 5.2415,
|
|
"mean_token_accuracy": 0.19719504714012145,
|
|
"num_tokens": 12184971.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.668954992294312,
|
|
"epoch": 5.76063601203266,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002491897846313402,
|
|
"loss": 5.1166,
|
|
"mean_token_accuracy": 0.21340720504522323,
|
|
"num_tokens": 12193643.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.6524131298065186,
|
|
"epoch": 5.764933390631715,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002488595249624297,
|
|
"loss": 5.2692,
|
|
"mean_token_accuracy": 0.1923430010676384,
|
|
"num_tokens": 12203073.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.634185171127319,
|
|
"epoch": 5.769230769230769,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024852932237386837,
|
|
"loss": 5.0889,
|
|
"mean_token_accuracy": 0.2056573748588562,
|
|
"num_tokens": 12211708.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.561472606658936,
|
|
"epoch": 5.773528147829824,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00024819917758668673,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.2066326007246971,
|
|
"num_tokens": 12219980.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.690144777297974,
|
|
"epoch": 5.777825526428878,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024786909132178906,
|
|
"loss": 5.2873,
|
|
"mean_token_accuracy": 0.1852530613541603,
|
|
"num_tokens": 12229523.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.694196701049805,
|
|
"epoch": 5.782122905027933,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024753906429995194,
|
|
"loss": 5.1182,
|
|
"mean_token_accuracy": 0.21192766726016998,
|
|
"num_tokens": 12238275.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.59233775138855,
|
|
"epoch": 5.786420283626987,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002472090972418222,
|
|
"loss": 5.1523,
|
|
"mean_token_accuracy": 0.2049255669116974,
|
|
"num_tokens": 12247569.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.6153425693511965,
|
|
"epoch": 5.790717662226042,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002468791908679163,
|
|
"loss": 5.2537,
|
|
"mean_token_accuracy": 0.1959172785282135,
|
|
"num_tokens": 12256923.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.659212636947632,
|
|
"epoch": 5.795015040825096,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002465493458986175,
|
|
"loss": 5.1509,
|
|
"mean_token_accuracy": 0.2078832983970642,
|
|
"num_tokens": 12265607.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.688326692581176,
|
|
"epoch": 5.799312419424151,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024621956305417587,
|
|
"loss": 5.1641,
|
|
"mean_token_accuracy": 0.20824072808027266,
|
|
"num_tokens": 12274348.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.642323017120361,
|
|
"epoch": 5.8036097980232055,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000245889843054705,
|
|
"loss": 5.1709,
|
|
"mean_token_accuracy": 0.2029065877199173,
|
|
"num_tokens": 12283072.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.604979610443115,
|
|
"epoch": 5.80790717662226,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024556018662018163,
|
|
"loss": 5.1839,
|
|
"mean_token_accuracy": 0.2049268439412117,
|
|
"num_tokens": 12292390.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.639643669128418,
|
|
"epoch": 5.812204555221315,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024523059447044377,
|
|
"loss": 5.2093,
|
|
"mean_token_accuracy": 0.19724443703889846,
|
|
"num_tokens": 12301394.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.706829071044922,
|
|
"epoch": 5.816501933820369,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002449010673251887,
|
|
"loss": 5.2095,
|
|
"mean_token_accuracy": 0.1985938772559166,
|
|
"num_tokens": 12310926.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.6958917617797855,
|
|
"epoch": 5.820799312419425,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002445716059039723,
|
|
"loss": 5.2136,
|
|
"mean_token_accuracy": 0.1975451961159706,
|
|
"num_tokens": 12319797.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.691281366348266,
|
|
"epoch": 5.825096691018478,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00024424221092620644,
|
|
"loss": 5.2724,
|
|
"mean_token_accuracy": 0.19195980876684188,
|
|
"num_tokens": 12329776.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.65817346572876,
|
|
"epoch": 5.829394069617534,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024391288311115822,
|
|
"loss": 5.2376,
|
|
"mean_token_accuracy": 0.19146959036588668,
|
|
"num_tokens": 12339679.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.686721420288086,
|
|
"epoch": 5.833691448216588,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002435836231779478,
|
|
"loss": 5.154,
|
|
"mean_token_accuracy": 0.20800569355487825,
|
|
"num_tokens": 12347647.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.681309080123901,
|
|
"epoch": 5.837988826815643,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024325443184554724,
|
|
"loss": 5.186,
|
|
"mean_token_accuracy": 0.19682217985391617,
|
|
"num_tokens": 12356567.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.695335245132446,
|
|
"epoch": 5.842286205414697,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00024292530983277904,
|
|
"loss": 5.2178,
|
|
"mean_token_accuracy": 0.19641996771097184,
|
|
"num_tokens": 12365434.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.655298042297363,
|
|
"epoch": 5.846583584013752,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00024259625785831408,
|
|
"loss": 5.2408,
|
|
"mean_token_accuracy": 0.19243062883615494,
|
|
"num_tokens": 12373600.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.707991600036621,
|
|
"epoch": 5.850880962612806,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00024226727664067023,
|
|
"loss": 5.221,
|
|
"mean_token_accuracy": 0.2026127129793167,
|
|
"num_tokens": 12382979.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.742899465560913,
|
|
"epoch": 5.855178341211861,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00024193836689821109,
|
|
"loss": 5.23,
|
|
"mean_token_accuracy": 0.19287777841091155,
|
|
"num_tokens": 12392142.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.6446703433990475,
|
|
"epoch": 5.8594757198109155,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002416095293491439,
|
|
"loss": 5.1635,
|
|
"mean_token_accuracy": 0.20380785763263704,
|
|
"num_tokens": 12401366.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.548672866821289,
|
|
"epoch": 5.86377309840997,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002412807647115186,
|
|
"loss": 5.0968,
|
|
"mean_token_accuracy": 0.20895102471113206,
|
|
"num_tokens": 12410081.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.576362037658692,
|
|
"epoch": 5.868070477009025,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024095207370322574,
|
|
"loss": 5.0801,
|
|
"mean_token_accuracy": 0.21353731006383897,
|
|
"num_tokens": 12419662.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.6391455173492435,
|
|
"epoch": 5.872367855608079,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024062345704199507,
|
|
"loss": 5.2492,
|
|
"mean_token_accuracy": 0.19254004061222077,
|
|
"num_tokens": 12429202.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.664939594268799,
|
|
"epoch": 5.876665234207134,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024029491544539405,
|
|
"loss": 5.163,
|
|
"mean_token_accuracy": 0.20638501644134521,
|
|
"num_tokens": 12438060.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.600986814498901,
|
|
"epoch": 5.880962612806188,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023996644963082616,
|
|
"loss": 5.115,
|
|
"mean_token_accuracy": 0.20695357173681259,
|
|
"num_tokens": 12446897.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.636267185211182,
|
|
"epoch": 5.885259991405243,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023963806031552948,
|
|
"loss": 5.2429,
|
|
"mean_token_accuracy": 0.19562375247478486,
|
|
"num_tokens": 12456273.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.678662633895874,
|
|
"epoch": 5.889557370004297,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023930974821657504,
|
|
"loss": 5.1946,
|
|
"mean_token_accuracy": 0.19977156817913055,
|
|
"num_tokens": 12465183.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.6863175392150875,
|
|
"epoch": 5.893854748603352,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023898151405086533,
|
|
"loss": 5.267,
|
|
"mean_token_accuracy": 0.1930649384856224,
|
|
"num_tokens": 12474248.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.6957320213317875,
|
|
"epoch": 5.8981521272024064,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023865335853513232,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.20737779140472412,
|
|
"num_tokens": 12483383.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.653290176391602,
|
|
"epoch": 5.902449505801461,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023832528238593677,
|
|
"loss": 5.1467,
|
|
"mean_token_accuracy": 0.20596271753311157,
|
|
"num_tokens": 12491707.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.656571531295777,
|
|
"epoch": 5.9067468844005155,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023799728631966556,
|
|
"loss": 5.1873,
|
|
"mean_token_accuracy": 0.19782646745443344,
|
|
"num_tokens": 12501277.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.665939998626709,
|
|
"epoch": 5.91104426299957,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.0002376693710525313,
|
|
"loss": 5.1767,
|
|
"mean_token_accuracy": 0.19610284864902497,
|
|
"num_tokens": 12509655.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.659619760513306,
|
|
"epoch": 5.915341641598625,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023734153730056967,
|
|
"loss": 5.1644,
|
|
"mean_token_accuracy": 0.2057501807808876,
|
|
"num_tokens": 12518408.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.693774700164795,
|
|
"epoch": 5.919639020197679,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023701378577963873,
|
|
"loss": 5.2122,
|
|
"mean_token_accuracy": 0.19942811280488967,
|
|
"num_tokens": 12528502.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.603091096878051,
|
|
"epoch": 5.923936398796734,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002366861172054166,
|
|
"loss": 5.1772,
|
|
"mean_token_accuracy": 0.2009565070271492,
|
|
"num_tokens": 12537803.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.7119873523712155,
|
|
"epoch": 5.928233777395788,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00023635853229340054,
|
|
"loss": 5.1754,
|
|
"mean_token_accuracy": 0.20153832137584687,
|
|
"num_tokens": 12547432.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.730851364135742,
|
|
"epoch": 5.932531155994843,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023603103175890512,
|
|
"loss": 5.2611,
|
|
"mean_token_accuracy": 0.19561851471662522,
|
|
"num_tokens": 12557630.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.627208471298218,
|
|
"epoch": 5.936828534593897,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00023570361631706062,
|
|
"loss": 5.1805,
|
|
"mean_token_accuracy": 0.20009257197380065,
|
|
"num_tokens": 12566168.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.632932138442993,
|
|
"epoch": 5.941125913192952,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023537628668281142,
|
|
"loss": 5.2077,
|
|
"mean_token_accuracy": 0.1941782847046852,
|
|
"num_tokens": 12575523.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.645141744613648,
|
|
"epoch": 5.945423291792007,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023504904357091468,
|
|
"loss": 5.0974,
|
|
"mean_token_accuracy": 0.2138304740190506,
|
|
"num_tokens": 12583598.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.58675742149353,
|
|
"epoch": 5.949720670391061,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002347218876959384,
|
|
"loss": 5.145,
|
|
"mean_token_accuracy": 0.20165867656469344,
|
|
"num_tokens": 12592627.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.569883060455322,
|
|
"epoch": 5.954018048990116,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.0002343948197722604,
|
|
"loss": 5.1719,
|
|
"mean_token_accuracy": 0.20010288059711456,
|
|
"num_tokens": 12601566.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.617793989181519,
|
|
"epoch": 5.958315427589171,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023406784051406638,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.19585224092006684,
|
|
"num_tokens": 12610526.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.642573118209839,
|
|
"epoch": 5.9626128061882255,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023374095063534816,
|
|
"loss": 5.1898,
|
|
"mean_token_accuracy": 0.19371045678853988,
|
|
"num_tokens": 12618892.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.67659821510315,
|
|
"epoch": 5.96691018478728,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023341415084990276,
|
|
"loss": 5.2364,
|
|
"mean_token_accuracy": 0.19624975472688674,
|
|
"num_tokens": 12627905.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.689128303527832,
|
|
"epoch": 5.971207563386335,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00023308744187132996,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.20014787316322327,
|
|
"num_tokens": 12636593.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.677709770202637,
|
|
"epoch": 5.975504941985389,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023276082441303197,
|
|
"loss": 5.1605,
|
|
"mean_token_accuracy": 0.20327988713979722,
|
|
"num_tokens": 12646055.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.700808382034301,
|
|
"epoch": 5.979802320584444,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00023243429918821056,
|
|
"loss": 5.2122,
|
|
"mean_token_accuracy": 0.20267895013093948,
|
|
"num_tokens": 12655125.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.639043807983398,
|
|
"epoch": 5.984099699183498,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023210786690986646,
|
|
"loss": 5.2462,
|
|
"mean_token_accuracy": 0.19144875854253768,
|
|
"num_tokens": 12663465.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.637978267669678,
|
|
"epoch": 5.988397077782553,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023178152829079712,
|
|
"loss": 5.1367,
|
|
"mean_token_accuracy": 0.20766632854938508,
|
|
"num_tokens": 12671573.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.610234212875366,
|
|
"epoch": 5.992694456381607,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00023145528404359562,
|
|
"loss": 5.0977,
|
|
"mean_token_accuracy": 0.21312466114759446,
|
|
"num_tokens": 12680865.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.687933015823364,
|
|
"epoch": 5.996991834980662,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0002311291348806492,
|
|
"loss": 5.1694,
|
|
"mean_token_accuracy": 0.2000417336821556,
|
|
"num_tokens": 12689790.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.581232441796197,
|
|
"epoch": 6.000859475719811,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002308030815141372,
|
|
"loss": 5.0931,
|
|
"mean_token_accuracy": 0.2127225250005722,
|
|
"num_tokens": 12697458.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.591152238845825,
|
|
"epoch": 6.005156854318866,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023047712465602976,
|
|
"loss": 4.9808,
|
|
"mean_token_accuracy": 0.2181651085615158,
|
|
"num_tokens": 12707122.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.591065883636475,
|
|
"epoch": 6.00945423291792,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023015126501808641,
|
|
"loss": 4.9688,
|
|
"mean_token_accuracy": 0.22145629525184632,
|
|
"num_tokens": 12715887.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.613132953643799,
|
|
"epoch": 6.013751611516975,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022982550331185437,
|
|
"loss": 4.9394,
|
|
"mean_token_accuracy": 0.22312914878129958,
|
|
"num_tokens": 12724630.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.013751611516975,
|
|
"eval_entropy": 5.533377574370788,
|
|
"eval_loss": 5.957110404968262,
|
|
"eval_mean_token_accuracy": 0.1760772148164006,
|
|
"eval_num_tokens": 12724630.0,
|
|
"eval_runtime": 2.8258,
|
|
"eval_samples_per_second": 1255.914,
|
|
"eval_steps_per_second": 157.122,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.643786382675171,
|
|
"epoch": 6.0180489901160295,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022949984024866704,
|
|
"loss": 5.083,
|
|
"mean_token_accuracy": 0.20120505541563033,
|
|
"num_tokens": 12734480.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.6155963897705075,
|
|
"epoch": 6.022346368715084,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002291742765396424,
|
|
"loss": 4.8932,
|
|
"mean_token_accuracy": 0.2182772383093834,
|
|
"num_tokens": 12742755.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.636760139465332,
|
|
"epoch": 6.0266437473141385,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022884881289568133,
|
|
"loss": 5.1041,
|
|
"mean_token_accuracy": 0.21231634318828582,
|
|
"num_tokens": 12751721.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.623049354553222,
|
|
"epoch": 6.030941125913193,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002285234500274665,
|
|
"loss": 4.8917,
|
|
"mean_token_accuracy": 0.22830037623643876,
|
|
"num_tokens": 12759982.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.683165121078491,
|
|
"epoch": 6.035238504512248,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022819818864546016,
|
|
"loss": 5.0226,
|
|
"mean_token_accuracy": 0.21362431347370148,
|
|
"num_tokens": 12768874.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.5635591506958,
|
|
"epoch": 6.039535883111302,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022787302945990345,
|
|
"loss": 4.9707,
|
|
"mean_token_accuracy": 0.22218442112207412,
|
|
"num_tokens": 12777613.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.559458589553833,
|
|
"epoch": 6.043833261710357,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00022754797318081383,
|
|
"loss": 5.0023,
|
|
"mean_token_accuracy": 0.2189720094203949,
|
|
"num_tokens": 12785436.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.680919361114502,
|
|
"epoch": 6.048130640309411,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022722302051798442,
|
|
"loss": 5.0986,
|
|
"mean_token_accuracy": 0.2048298805952072,
|
|
"num_tokens": 12795334.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.618054628372192,
|
|
"epoch": 6.052428018908466,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002268981721809819,
|
|
"loss": 4.945,
|
|
"mean_token_accuracy": 0.22455888539552687,
|
|
"num_tokens": 12804579.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.640971660614014,
|
|
"epoch": 6.05672539750752,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002265734288791451,
|
|
"loss": 5.1089,
|
|
"mean_token_accuracy": 0.19927433133125305,
|
|
"num_tokens": 12813557.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.607983112335205,
|
|
"epoch": 6.061022776106575,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00022624879132158377,
|
|
"loss": 5.0562,
|
|
"mean_token_accuracy": 0.205833400785923,
|
|
"num_tokens": 12822588.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.6992696762084964,
|
|
"epoch": 6.0653201547056295,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00022592426021717654,
|
|
"loss": 5.1129,
|
|
"mean_token_accuracy": 0.21240501403808593,
|
|
"num_tokens": 12832543.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.703296422958374,
|
|
"epoch": 6.069617533304684,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0002255998362745696,
|
|
"loss": 5.0297,
|
|
"mean_token_accuracy": 0.2095639556646347,
|
|
"num_tokens": 12841237.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.705530500411987,
|
|
"epoch": 6.073914911903739,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022527552020217513,
|
|
"loss": 5.1382,
|
|
"mean_token_accuracy": 0.20859335511922836,
|
|
"num_tokens": 12850626.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.592755365371704,
|
|
"epoch": 6.078212290502793,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002249513127081697,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.2061067745089531,
|
|
"num_tokens": 12859290.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.650101232528686,
|
|
"epoch": 6.082509669101848,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022462721450049316,
|
|
"loss": 4.977,
|
|
"mean_token_accuracy": 0.2154896855354309,
|
|
"num_tokens": 12867828.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.589192199707031,
|
|
"epoch": 6.086807047700902,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002243032262868464,
|
|
"loss": 4.9809,
|
|
"mean_token_accuracy": 0.21581656187772752,
|
|
"num_tokens": 12876186.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.580323934555054,
|
|
"epoch": 6.091104426299957,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022397934877469,
|
|
"loss": 4.9976,
|
|
"mean_token_accuracy": 0.2120466485619545,
|
|
"num_tokens": 12884657.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.587834024429322,
|
|
"epoch": 6.095401804899011,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002236555826712432,
|
|
"loss": 4.9596,
|
|
"mean_token_accuracy": 0.21566486060619355,
|
|
"num_tokens": 12893760.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.662276983261108,
|
|
"epoch": 6.099699183498066,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022333192868348152,
|
|
"loss": 5.0771,
|
|
"mean_token_accuracy": 0.203414948284626,
|
|
"num_tokens": 12903263.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.622202301025391,
|
|
"epoch": 6.10399656209712,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022300838751813606,
|
|
"loss": 4.9864,
|
|
"mean_token_accuracy": 0.21558584421873092,
|
|
"num_tokens": 12912497.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.631495571136474,
|
|
"epoch": 6.108293940696175,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00022268495988169145,
|
|
"loss": 5.0449,
|
|
"mean_token_accuracy": 0.2076230064034462,
|
|
"num_tokens": 12921977.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.558540964126587,
|
|
"epoch": 6.1125913192952295,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022236164648038433,
|
|
"loss": 5.012,
|
|
"mean_token_accuracy": 0.20613891929388045,
|
|
"num_tokens": 12931845.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.7064347743988035,
|
|
"epoch": 6.116888697894285,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002220384480202019,
|
|
"loss": 5.0321,
|
|
"mean_token_accuracy": 0.21196336150169373,
|
|
"num_tokens": 12941613.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.690691089630127,
|
|
"epoch": 6.1211860764933395,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022171536520688046,
|
|
"loss": 5.013,
|
|
"mean_token_accuracy": 0.21545159965753555,
|
|
"num_tokens": 12950570.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.626392841339111,
|
|
"epoch": 6.125483455092394,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022139239874590362,
|
|
"loss": 5.0225,
|
|
"mean_token_accuracy": 0.20732492059469224,
|
|
"num_tokens": 12959576.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.608592796325683,
|
|
"epoch": 6.1297808336914485,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002210695493425013,
|
|
"loss": 4.9589,
|
|
"mean_token_accuracy": 0.2213190570473671,
|
|
"num_tokens": 12968333.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.616888856887817,
|
|
"epoch": 6.134078212290503,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00022074681770164735,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.20927069187164307,
|
|
"num_tokens": 12977213.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.472827053070068,
|
|
"epoch": 6.138375590889558,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022042420452805868,
|
|
"loss": 4.8732,
|
|
"mean_token_accuracy": 0.2373045191168785,
|
|
"num_tokens": 12986289.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.619239044189453,
|
|
"epoch": 6.142672969488612,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022010171052619365,
|
|
"loss": 4.952,
|
|
"mean_token_accuracy": 0.2176075041294098,
|
|
"num_tokens": 12994532.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.635759115219116,
|
|
"epoch": 6.146970348087667,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021977933640025,
|
|
"loss": 5.0266,
|
|
"mean_token_accuracy": 0.21576329469680786,
|
|
"num_tokens": 13003951.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.597749137878418,
|
|
"epoch": 6.151267726686721,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021945708285416434,
|
|
"loss": 5.0177,
|
|
"mean_token_accuracy": 0.2157116338610649,
|
|
"num_tokens": 13013123.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.637640428543091,
|
|
"epoch": 6.155565105285776,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002191349505916093,
|
|
"loss": 5.0426,
|
|
"mean_token_accuracy": 0.2115068703889847,
|
|
"num_tokens": 13021451.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.630251264572143,
|
|
"epoch": 6.15986248388483,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021881294031599318,
|
|
"loss": 4.9621,
|
|
"mean_token_accuracy": 0.22164122015237808,
|
|
"num_tokens": 13030894.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.632130861282349,
|
|
"epoch": 6.164159862483885,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002184910527304576,
|
|
"loss": 5.0124,
|
|
"mean_token_accuracy": 0.21388618648052216,
|
|
"num_tokens": 13039930.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.644307708740234,
|
|
"epoch": 6.1684572410829395,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00021816928853787636,
|
|
"loss": 5.0692,
|
|
"mean_token_accuracy": 0.20602648705244064,
|
|
"num_tokens": 13048575.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.714348697662354,
|
|
"epoch": 6.172754619681994,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021784764844085398,
|
|
"loss": 5.1424,
|
|
"mean_token_accuracy": 0.19582845866680146,
|
|
"num_tokens": 13057624.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.581179761886597,
|
|
"epoch": 6.177051998281049,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002175261331417238,
|
|
"loss": 4.9605,
|
|
"mean_token_accuracy": 0.21767899841070176,
|
|
"num_tokens": 13066556.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.566150093078614,
|
|
"epoch": 6.181349376880103,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021720474334254675,
|
|
"loss": 4.9856,
|
|
"mean_token_accuracy": 0.21944366693496703,
|
|
"num_tokens": 13075617.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.592297124862671,
|
|
"epoch": 6.185646755479158,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 0.00021688347974510962,
|
|
"loss": 5.0305,
|
|
"mean_token_accuracy": 0.21369386315345765,
|
|
"num_tokens": 13084528.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.509733724594116,
|
|
"epoch": 6.189944134078212,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021656234305092377,
|
|
"loss": 4.9808,
|
|
"mean_token_accuracy": 0.2168891489505768,
|
|
"num_tokens": 13093841.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.563720655441284,
|
|
"epoch": 6.194241512677267,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002162413339612234,
|
|
"loss": 5.0544,
|
|
"mean_token_accuracy": 0.20949698835611344,
|
|
"num_tokens": 13103104.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.663839673995971,
|
|
"epoch": 6.198538891276321,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 0.00021592045317696406,
|
|
"loss": 5.0397,
|
|
"mean_token_accuracy": 0.2100674256682396,
|
|
"num_tokens": 13112721.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.703644847869873,
|
|
"epoch": 6.202836269875376,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00021559970139882102,
|
|
"loss": 5.0939,
|
|
"mean_token_accuracy": 0.2089846983551979,
|
|
"num_tokens": 13121949.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.687656259536743,
|
|
"epoch": 6.20713364847443,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002152790793271881,
|
|
"loss": 5.0865,
|
|
"mean_token_accuracy": 0.20369456559419633,
|
|
"num_tokens": 13130693.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.6849222660064695,
|
|
"epoch": 6.211431027073485,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021495858766217558,
|
|
"loss": 5.0989,
|
|
"mean_token_accuracy": 0.20224303007125854,
|
|
"num_tokens": 13140438.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.6058190822601315,
|
|
"epoch": 6.2157284056725395,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021463822710360932,
|
|
"loss": 5.0062,
|
|
"mean_token_accuracy": 0.2131658360362053,
|
|
"num_tokens": 13149621.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.595905208587647,
|
|
"epoch": 6.220025784271594,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021431799835102867,
|
|
"loss": 5.0294,
|
|
"mean_token_accuracy": 0.21656057834625245,
|
|
"num_tokens": 13158366.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.647145795822143,
|
|
"epoch": 6.224323162870649,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021399790210368524,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.20274337828159333,
|
|
"num_tokens": 13167303.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.648982286453247,
|
|
"epoch": 6.228620541469703,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00021367793906054133,
|
|
"loss": 5.0516,
|
|
"mean_token_accuracy": 0.21569620668888093,
|
|
"num_tokens": 13175276.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.698874807357788,
|
|
"epoch": 6.232917920068758,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00021335810992026823,
|
|
"loss": 5.0679,
|
|
"mean_token_accuracy": 0.20422574281692504,
|
|
"num_tokens": 13184352.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.581685304641724,
|
|
"epoch": 6.237215298667812,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00021303841538124497,
|
|
"loss": 4.9913,
|
|
"mean_token_accuracy": 0.21835901439189911,
|
|
"num_tokens": 13192772.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.619719505310059,
|
|
"epoch": 6.241512677266867,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021271885614155685,
|
|
"loss": 5.0566,
|
|
"mean_token_accuracy": 0.21381517946720124,
|
|
"num_tokens": 13201574.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.601267528533936,
|
|
"epoch": 6.245810055865922,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002123994328989932,
|
|
"loss": 5.0269,
|
|
"mean_token_accuracy": 0.20712526887655258,
|
|
"num_tokens": 13211572.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.64578332901001,
|
|
"epoch": 6.250107434464977,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00021208014635104688,
|
|
"loss": 5.0702,
|
|
"mean_token_accuracy": 0.2048632487654686,
|
|
"num_tokens": 13220614.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.667975997924804,
|
|
"epoch": 6.254404813064031,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021176099719491209,
|
|
"loss": 5.0186,
|
|
"mean_token_accuracy": 0.22564091682434081,
|
|
"num_tokens": 13229484.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.595759534835816,
|
|
"epoch": 6.258702191663086,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00021144198612748312,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.21381184458732605,
|
|
"num_tokens": 13239649.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.589407968521118,
|
|
"epoch": 6.26299957026214,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00021112311384535243,
|
|
"loss": 4.9724,
|
|
"mean_token_accuracy": 0.2134211167693138,
|
|
"num_tokens": 13248199.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.650906133651733,
|
|
"epoch": 6.267296948861195,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00021080438104480976,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.20605516731739043,
|
|
"num_tokens": 13257254.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.634106636047363,
|
|
"epoch": 6.2715943274602495,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00021048578842184019,
|
|
"loss": 5.0281,
|
|
"mean_token_accuracy": 0.20817134529352188,
|
|
"num_tokens": 13266272.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.628504705429077,
|
|
"epoch": 6.275891706059304,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00021016733667212245,
|
|
"loss": 5.0184,
|
|
"mean_token_accuracy": 0.22011199593544006,
|
|
"num_tokens": 13274871.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.705023670196534,
|
|
"epoch": 6.280189084658359,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020984902649102806,
|
|
"loss": 5.1136,
|
|
"mean_token_accuracy": 0.20444865226745607,
|
|
"num_tokens": 13284844.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.649649238586425,
|
|
"epoch": 6.284486463257413,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00020953085857361924,
|
|
"loss": 5.0665,
|
|
"mean_token_accuracy": 0.2061414062976837,
|
|
"num_tokens": 13293977.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.651197290420532,
|
|
"epoch": 6.288783841856468,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00020921283361464754,
|
|
"loss": 5.1164,
|
|
"mean_token_accuracy": 0.20141676515340806,
|
|
"num_tokens": 13303359.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.688200426101685,
|
|
"epoch": 6.293081220455522,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020889495230855232,
|
|
"loss": 5.0812,
|
|
"mean_token_accuracy": 0.2116152212023735,
|
|
"num_tokens": 13311965.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.701021575927735,
|
|
"epoch": 6.297378599054577,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00020857721534945923,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.2124857261776924,
|
|
"num_tokens": 13321784.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.627671527862549,
|
|
"epoch": 6.301675977653631,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002082596234311789,
|
|
"loss": 5.0049,
|
|
"mean_token_accuracy": 0.21768666207790374,
|
|
"num_tokens": 13330141.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.564412593841553,
|
|
"epoch": 6.305973356252686,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0002079421772472051,
|
|
"loss": 5.0512,
|
|
"mean_token_accuracy": 0.20882692635059358,
|
|
"num_tokens": 13339303.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.634327507019043,
|
|
"epoch": 6.31027073485174,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002076248774907134,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.20862669944763185,
|
|
"num_tokens": 13348574.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.7153911113739015,
|
|
"epoch": 6.314568113450795,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020730772485455962,
|
|
"loss": 5.0429,
|
|
"mean_token_accuracy": 0.21003767251968383,
|
|
"num_tokens": 13356770.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.6841834545135494,
|
|
"epoch": 6.3188654920498495,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002069907200312785,
|
|
"loss": 5.1072,
|
|
"mean_token_accuracy": 0.20742908269166946,
|
|
"num_tokens": 13366973.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.675436639785767,
|
|
"epoch": 6.323162870648904,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020667386371308162,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.20942332446575165,
|
|
"num_tokens": 13375849.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.596558570861816,
|
|
"epoch": 6.327460249247959,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020635715659185673,
|
|
"loss": 4.9789,
|
|
"mean_token_accuracy": 0.2148328974843025,
|
|
"num_tokens": 13384051.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.615690517425537,
|
|
"epoch": 6.331757627847013,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00020604059935916551,
|
|
"loss": 5.1375,
|
|
"mean_token_accuracy": 0.20746109783649444,
|
|
"num_tokens": 13393194.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.6018767833709715,
|
|
"epoch": 6.336055006446068,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00020572419270624255,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.21225236505270004,
|
|
"num_tokens": 13402701.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.556215286254883,
|
|
"epoch": 6.340352385045122,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00020540793732399339,
|
|
"loss": 4.9253,
|
|
"mean_token_accuracy": 0.2242536649107933,
|
|
"num_tokens": 13411583.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.63846435546875,
|
|
"epoch": 6.344649763644177,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020509183390299325,
|
|
"loss": 5.0518,
|
|
"mean_token_accuracy": 0.2039532944560051,
|
|
"num_tokens": 13420787.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.635910415649414,
|
|
"epoch": 6.348947142243231,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020477588313348594,
|
|
"loss": 5.0742,
|
|
"mean_token_accuracy": 0.20989339649677277,
|
|
"num_tokens": 13428903.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.61276216506958,
|
|
"epoch": 6.353244520842286,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00020446008570538154,
|
|
"loss": 5.0637,
|
|
"mean_token_accuracy": 0.2056269735097885,
|
|
"num_tokens": 13438444.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.593193244934082,
|
|
"epoch": 6.35754189944134,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002041444423082554,
|
|
"loss": 5.0675,
|
|
"mean_token_accuracy": 0.2046339526772499,
|
|
"num_tokens": 13447188.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.641726875305176,
|
|
"epoch": 6.361839278040395,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00020382895363134652,
|
|
"loss": 5.0638,
|
|
"mean_token_accuracy": 0.20288414657115936,
|
|
"num_tokens": 13455980.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.647207164764405,
|
|
"epoch": 6.3661366566394495,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020351362036355602,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.20901528298854827,
|
|
"num_tokens": 13465029.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.655359888076783,
|
|
"epoch": 6.370434035238505,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002031984431934459,
|
|
"loss": 5.0989,
|
|
"mean_token_accuracy": 0.20772464126348494,
|
|
"num_tokens": 13474366.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.677274131774903,
|
|
"epoch": 6.3747314138375595,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00020288342280923695,
|
|
"loss": 5.0687,
|
|
"mean_token_accuracy": 0.2046821668744087,
|
|
"num_tokens": 13483186.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.664965534210205,
|
|
"epoch": 6.379028792436614,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020256855989880785,
|
|
"loss": 5.1207,
|
|
"mean_token_accuracy": 0.2006380647420883,
|
|
"num_tokens": 13492066.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.633226108551026,
|
|
"epoch": 6.383326171035669,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020225385514969336,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.2073546901345253,
|
|
"num_tokens": 13501671.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.646773624420166,
|
|
"epoch": 6.387623549634723,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00020193930924908277,
|
|
"loss": 4.9984,
|
|
"mean_token_accuracy": 0.2163984075188637,
|
|
"num_tokens": 13510323.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.635648775100708,
|
|
"epoch": 6.391920928233778,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00020162492288381867,
|
|
"loss": 5.0298,
|
|
"mean_token_accuracy": 0.2154936209321022,
|
|
"num_tokens": 13518658.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.574488735198974,
|
|
"epoch": 6.396218306832832,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002013106967403953,
|
|
"loss": 5.0944,
|
|
"mean_token_accuracy": 0.20916002243757248,
|
|
"num_tokens": 13528053.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.623544931411743,
|
|
"epoch": 6.400515685431887,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002009966315049569,
|
|
"loss": 5.0082,
|
|
"mean_token_accuracy": 0.21387167572975158,
|
|
"num_tokens": 13537125.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.676896858215332,
|
|
"epoch": 6.404813064030941,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002006827278632964,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.20302549451589585,
|
|
"num_tokens": 13547859.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.657016849517822,
|
|
"epoch": 6.409110442629996,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020036898650085377,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.21022023111581803,
|
|
"num_tokens": 13557545.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.690197515487671,
|
|
"epoch": 6.41340782122905,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020005540810271493,
|
|
"loss": 5.1268,
|
|
"mean_token_accuracy": 0.2055033728480339,
|
|
"num_tokens": 13566281.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.623619413375854,
|
|
"epoch": 6.417705199828105,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00019974199335360976,
|
|
"loss": 4.9998,
|
|
"mean_token_accuracy": 0.21737098097801208,
|
|
"num_tokens": 13575461.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.600651741027832,
|
|
"epoch": 6.4220025784271595,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00019942874293791068,
|
|
"loss": 5.007,
|
|
"mean_token_accuracy": 0.2195130094885826,
|
|
"num_tokens": 13585107.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.630463743209839,
|
|
"epoch": 6.426299957026214,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019911565753963145,
|
|
"loss": 4.9885,
|
|
"mean_token_accuracy": 0.21782609969377517,
|
|
"num_tokens": 13593787.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.609920978546143,
|
|
"epoch": 6.430597335625269,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001988027378424254,
|
|
"loss": 5.0691,
|
|
"mean_token_accuracy": 0.20439999401569367,
|
|
"num_tokens": 13602956.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.593091821670532,
|
|
"epoch": 6.434894714224323,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019848998452958429,
|
|
"loss": 5.045,
|
|
"mean_token_accuracy": 0.2129646435379982,
|
|
"num_tokens": 13612380.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.604695415496826,
|
|
"epoch": 6.439192092823378,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019817739828403602,
|
|
"loss": 5.0732,
|
|
"mean_token_accuracy": 0.21138588786125184,
|
|
"num_tokens": 13622049.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.680410385131836,
|
|
"epoch": 6.443489471422432,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019786497978834422,
|
|
"loss": 5.0841,
|
|
"mean_token_accuracy": 0.20210780054330826,
|
|
"num_tokens": 13631497.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.443489471422432,
|
|
"eval_entropy": 5.5225924457515685,
|
|
"eval_loss": 5.952114105224609,
|
|
"eval_mean_token_accuracy": 0.17705098398633906,
|
|
"eval_num_tokens": 13631497.0,
|
|
"eval_runtime": 2.6364,
|
|
"eval_samples_per_second": 1346.172,
|
|
"eval_steps_per_second": 168.414,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.639434385299682,
|
|
"epoch": 6.447786850021487,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00019755272972470602,
|
|
"loss": 5.0537,
|
|
"mean_token_accuracy": 0.20617803633213044,
|
|
"num_tokens": 13639767.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.636047649383545,
|
|
"epoch": 6.452084228620541,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019724064877495057,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.2049650251865387,
|
|
"num_tokens": 13649255.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.707356643676758,
|
|
"epoch": 6.456381607219596,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00019692873762053808,
|
|
"loss": 5.1381,
|
|
"mean_token_accuracy": 0.19920433461666107,
|
|
"num_tokens": 13657764.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.648329210281372,
|
|
"epoch": 6.46067898581865,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019661699694255785,
|
|
"loss": 5.0851,
|
|
"mean_token_accuracy": 0.20550374686717987,
|
|
"num_tokens": 13667922.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.641249370574951,
|
|
"epoch": 6.464976364417705,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019630542742172692,
|
|
"loss": 5.1066,
|
|
"mean_token_accuracy": 0.20587169677019118,
|
|
"num_tokens": 13678196.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.645707273483277,
|
|
"epoch": 6.4692737430167595,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019599402973838854,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.20322970002889634,
|
|
"num_tokens": 13687605.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.620758533477783,
|
|
"epoch": 6.473571121615814,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001956828045725107,
|
|
"loss": 4.9789,
|
|
"mean_token_accuracy": 0.22267434895038604,
|
|
"num_tokens": 13695790.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.644010448455811,
|
|
"epoch": 6.477868500214869,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001953717526036849,
|
|
"loss": 5.0751,
|
|
"mean_token_accuracy": 0.21198058724403382,
|
|
"num_tokens": 13705608.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.588969039916992,
|
|
"epoch": 6.482165878813923,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00019506087451112437,
|
|
"loss": 4.9635,
|
|
"mean_token_accuracy": 0.21701718866825104,
|
|
"num_tokens": 13714286.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.646505641937256,
|
|
"epoch": 6.486463257412978,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00019475017097366244,
|
|
"loss": 5.0024,
|
|
"mean_token_accuracy": 0.21229881793260574,
|
|
"num_tokens": 13722575.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.640190362930298,
|
|
"epoch": 6.490760636012032,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00019443964266975156,
|
|
"loss": 5.0692,
|
|
"mean_token_accuracy": 0.2061358943581581,
|
|
"num_tokens": 13732342.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.611169004440308,
|
|
"epoch": 6.495058014611088,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001941292902774614,
|
|
"loss": 4.9584,
|
|
"mean_token_accuracy": 0.2255277454853058,
|
|
"num_tokens": 13741386.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.6245041847229,
|
|
"epoch": 6.499355393210142,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019381911447447742,
|
|
"loss": 5.0584,
|
|
"mean_token_accuracy": 0.2153615340590477,
|
|
"num_tokens": 13750755.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.638590955734253,
|
|
"epoch": 6.503652771809197,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019350911593809977,
|
|
"loss": 4.9892,
|
|
"mean_token_accuracy": 0.21671021729707718,
|
|
"num_tokens": 13759903.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.609411334991455,
|
|
"epoch": 6.507950150408251,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019319929534524128,
|
|
"loss": 4.963,
|
|
"mean_token_accuracy": 0.21525610983371735,
|
|
"num_tokens": 13768066.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.577180099487305,
|
|
"epoch": 6.512247529007306,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00019288965337242636,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.2171801969408989,
|
|
"num_tokens": 13778010.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.660821771621704,
|
|
"epoch": 6.51654490760636,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019258019069578924,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.20383259803056716,
|
|
"num_tokens": 13787415.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.587430477142334,
|
|
"epoch": 6.520842286205415,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019227090799107266,
|
|
"loss": 4.9987,
|
|
"mean_token_accuracy": 0.21421348303556442,
|
|
"num_tokens": 13796937.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.685607051849365,
|
|
"epoch": 6.5251396648044695,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001919618059336265,
|
|
"loss": 5.0888,
|
|
"mean_token_accuracy": 0.20820480585098267,
|
|
"num_tokens": 13805759.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.647010040283203,
|
|
"epoch": 6.529437043403524,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019165288519840617,
|
|
"loss": 5.0114,
|
|
"mean_token_accuracy": 0.21360699981451034,
|
|
"num_tokens": 13814915.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.590897083282471,
|
|
"epoch": 6.533734422002579,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000191344146459971,
|
|
"loss": 5.0431,
|
|
"mean_token_accuracy": 0.21282268315553665,
|
|
"num_tokens": 13824334.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.621375226974488,
|
|
"epoch": 6.538031800601633,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 0.00019103559039248302,
|
|
"loss": 5.046,
|
|
"mean_token_accuracy": 0.20823441296815873,
|
|
"num_tokens": 13834124.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.584835720062256,
|
|
"epoch": 6.542329179200688,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001907272176697052,
|
|
"loss": 4.9471,
|
|
"mean_token_accuracy": 0.2210870087146759,
|
|
"num_tokens": 13843240.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.700993156433105,
|
|
"epoch": 6.546626557799742,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019041902896500059,
|
|
"loss": 5.1154,
|
|
"mean_token_accuracy": 0.20768067985773087,
|
|
"num_tokens": 13852837.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.657913732528686,
|
|
"epoch": 6.550923936398797,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00019011102495132993,
|
|
"loss": 5.0609,
|
|
"mean_token_accuracy": 0.20558776706457138,
|
|
"num_tokens": 13861650.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.606899881362915,
|
|
"epoch": 6.555221314997851,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018980320630125104,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.2085315242409706,
|
|
"num_tokens": 13871939.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.734761667251587,
|
|
"epoch": 6.559518693596906,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018949557368691666,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.20380180478096008,
|
|
"num_tokens": 13881132.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.58612699508667,
|
|
"epoch": 6.56381607219596,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018918812778007343,
|
|
"loss": 4.9718,
|
|
"mean_token_accuracy": 0.21432769149541855,
|
|
"num_tokens": 13890476.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.563734769821167,
|
|
"epoch": 6.568113450795015,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018888086925206054,
|
|
"loss": 5.0001,
|
|
"mean_token_accuracy": 0.2179549440741539,
|
|
"num_tokens": 13900720.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.605028581619263,
|
|
"epoch": 6.5724108293940695,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018857379877380763,
|
|
"loss": 5.0651,
|
|
"mean_token_accuracy": 0.2087094008922577,
|
|
"num_tokens": 13909461.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.634809494018555,
|
|
"epoch": 6.576708207993124,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018826691701583404,
|
|
"loss": 5.0291,
|
|
"mean_token_accuracy": 0.2082165375351906,
|
|
"num_tokens": 13918136.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.630088949203492,
|
|
"epoch": 6.581005586592179,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018796022464824663,
|
|
"loss": 5.1157,
|
|
"mean_token_accuracy": 0.20387773662805558,
|
|
"num_tokens": 13926965.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.715147066116333,
|
|
"epoch": 6.585302965191233,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018765372234073912,
|
|
"loss": 5.1567,
|
|
"mean_token_accuracy": 0.1988100677728653,
|
|
"num_tokens": 13936468.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.637403535842895,
|
|
"epoch": 6.589600343790288,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018734741076259005,
|
|
"loss": 4.9988,
|
|
"mean_token_accuracy": 0.21581486463546753,
|
|
"num_tokens": 13945424.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.618121910095215,
|
|
"epoch": 6.593897722389342,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018704129058266152,
|
|
"loss": 5.0357,
|
|
"mean_token_accuracy": 0.21465452760457993,
|
|
"num_tokens": 13955129.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.673315095901489,
|
|
"epoch": 6.598195100988397,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00018673536246939743,
|
|
"loss": 5.1315,
|
|
"mean_token_accuracy": 0.20281609445810317,
|
|
"num_tokens": 13964316.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.614810562133789,
|
|
"epoch": 6.602492479587451,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018642962709082274,
|
|
"loss": 5.0309,
|
|
"mean_token_accuracy": 0.2119555190205574,
|
|
"num_tokens": 13973173.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.645280218124389,
|
|
"epoch": 6.606789858186506,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00018612408511454103,
|
|
"loss": 5.0466,
|
|
"mean_token_accuracy": 0.2133909285068512,
|
|
"num_tokens": 13981531.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.633004856109619,
|
|
"epoch": 6.61108723678556,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018581873720773423,
|
|
"loss": 5.0742,
|
|
"mean_token_accuracy": 0.20953545868396758,
|
|
"num_tokens": 13990202.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.663352012634277,
|
|
"epoch": 6.615384615384615,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018551358403715989,
|
|
"loss": 5.0232,
|
|
"mean_token_accuracy": 0.2134046211838722,
|
|
"num_tokens": 13998851.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.629218864440918,
|
|
"epoch": 6.61968199398367,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018520862626915052,
|
|
"loss": 5.037,
|
|
"mean_token_accuracy": 0.21289423257112502,
|
|
"num_tokens": 14008120.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.6156574249267575,
|
|
"epoch": 6.623979372582724,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018490386456961223,
|
|
"loss": 5.1078,
|
|
"mean_token_accuracy": 0.20287092477083207,
|
|
"num_tokens": 14017435.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.649461317062378,
|
|
"epoch": 6.6282767511817795,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001845992996040224,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.19904009103775025,
|
|
"num_tokens": 14026260.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.7406963348388675,
|
|
"epoch": 6.632574129780834,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018429493203742946,
|
|
"loss": 5.1145,
|
|
"mean_token_accuracy": 0.20093687921762465,
|
|
"num_tokens": 14034854.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.600214052200317,
|
|
"epoch": 6.636871508379889,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018399076253445052,
|
|
"loss": 5.0542,
|
|
"mean_token_accuracy": 0.20597160011529922,
|
|
"num_tokens": 14043486.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.59062876701355,
|
|
"epoch": 6.641168886978943,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018368679175927012,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.20478217750787736,
|
|
"num_tokens": 14052903.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.594202327728271,
|
|
"epoch": 6.645466265577998,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018338302037563885,
|
|
"loss": 5.0383,
|
|
"mean_token_accuracy": 0.2146869868040085,
|
|
"num_tokens": 14061679.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.691454267501831,
|
|
"epoch": 6.649763644177052,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018307944904687211,
|
|
"loss": 5.0886,
|
|
"mean_token_accuracy": 0.2146966576576233,
|
|
"num_tokens": 14070549.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.584806203842163,
|
|
"epoch": 6.654061022776107,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001827760784358483,
|
|
"loss": 4.9574,
|
|
"mean_token_accuracy": 0.22807697057724,
|
|
"num_tokens": 14079321.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.660425472259521,
|
|
"epoch": 6.658358401375161,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018247290920500776,
|
|
"loss": 5.0916,
|
|
"mean_token_accuracy": 0.2055915892124176,
|
|
"num_tokens": 14087951.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.499595880508423,
|
|
"epoch": 6.662655779974216,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018216994201635062,
|
|
"loss": 4.995,
|
|
"mean_token_accuracy": 0.2179614320397377,
|
|
"num_tokens": 14097563.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.634319829940796,
|
|
"epoch": 6.66695315857327,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00018186717753143633,
|
|
"loss": 5.0786,
|
|
"mean_token_accuracy": 0.20810858011245728,
|
|
"num_tokens": 14106961.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.649790239334107,
|
|
"epoch": 6.671250537172325,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018156461641138133,
|
|
"loss": 5.0952,
|
|
"mean_token_accuracy": 0.2077918082475662,
|
|
"num_tokens": 14115854.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.699000835418701,
|
|
"epoch": 6.6755479157713795,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00018126225931685836,
|
|
"loss": 5.1044,
|
|
"mean_token_accuracy": 0.20795947462320327,
|
|
"num_tokens": 14124778.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.602504062652588,
|
|
"epoch": 6.679845294370434,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018096010690809444,
|
|
"loss": 5.0142,
|
|
"mean_token_accuracy": 0.2161995381116867,
|
|
"num_tokens": 14133708.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.524826383590698,
|
|
"epoch": 6.684142672969489,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00018065815984486962,
|
|
"loss": 4.9467,
|
|
"mean_token_accuracy": 0.22502322793006896,
|
|
"num_tokens": 14142885.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.587267112731934,
|
|
"epoch": 6.688440051568543,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018035641878651548,
|
|
"loss": 5.0549,
|
|
"mean_token_accuracy": 0.2125038519501686,
|
|
"num_tokens": 14151156.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.6342926025390625,
|
|
"epoch": 6.692737430167598,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018005488439191408,
|
|
"loss": 5.0639,
|
|
"mean_token_accuracy": 0.20955566316843033,
|
|
"num_tokens": 14160482.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.6495038032531735,
|
|
"epoch": 6.697034808766652,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001797535573194959,
|
|
"loss": 5.004,
|
|
"mean_token_accuracy": 0.21074233204126358,
|
|
"num_tokens": 14169637.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.597209548950195,
|
|
"epoch": 6.701332187365707,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001794524382272389,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.20124897062778474,
|
|
"num_tokens": 14179215.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.695490884780884,
|
|
"epoch": 6.705629565964761,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001791515277726667,
|
|
"loss": 5.1175,
|
|
"mean_token_accuracy": 0.20180365294218064,
|
|
"num_tokens": 14188484.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.623464488983155,
|
|
"epoch": 6.709926944563816,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017885082661284763,
|
|
"loss": 5.0344,
|
|
"mean_token_accuracy": 0.21944525092840195,
|
|
"num_tokens": 14197275.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.616363382339477,
|
|
"epoch": 6.71422432316287,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017855033540439274,
|
|
"loss": 5.0142,
|
|
"mean_token_accuracy": 0.2155451610684395,
|
|
"num_tokens": 14206260.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.652454376220703,
|
|
"epoch": 6.718521701761925,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017825005480345463,
|
|
"loss": 5.1188,
|
|
"mean_token_accuracy": 0.1999969884753227,
|
|
"num_tokens": 14215758.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.578374576568604,
|
|
"epoch": 6.7228190803609795,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017794998546572627,
|
|
"loss": 5.0269,
|
|
"mean_token_accuracy": 0.21094758957624435,
|
|
"num_tokens": 14225147.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.60685544013977,
|
|
"epoch": 6.727116458960034,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001776501280464391,
|
|
"loss": 5.0734,
|
|
"mean_token_accuracy": 0.20803066343069077,
|
|
"num_tokens": 14233567.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.595700979232788,
|
|
"epoch": 6.731413837559089,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017735048320036197,
|
|
"loss": 4.9722,
|
|
"mean_token_accuracy": 0.22230015695095062,
|
|
"num_tokens": 14242101.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.647860383987426,
|
|
"epoch": 6.735711216158143,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017705105158179917,
|
|
"loss": 5.0221,
|
|
"mean_token_accuracy": 0.21438967138528825,
|
|
"num_tokens": 14251502.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.583736515045166,
|
|
"epoch": 6.740008594757198,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017675183384458987,
|
|
"loss": 4.9584,
|
|
"mean_token_accuracy": 0.22263683527708053,
|
|
"num_tokens": 14260059.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.685175132751465,
|
|
"epoch": 6.744305973356253,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017645283064210616,
|
|
"loss": 5.0451,
|
|
"mean_token_accuracy": 0.20948705077171326,
|
|
"num_tokens": 14268672.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.560036373138428,
|
|
"epoch": 6.748603351955307,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017615404262725132,
|
|
"loss": 4.9044,
|
|
"mean_token_accuracy": 0.22161513268947602,
|
|
"num_tokens": 14277636.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.525001001358032,
|
|
"epoch": 6.752900730554362,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017585547045245885,
|
|
"loss": 4.9534,
|
|
"mean_token_accuracy": 0.22249802201986313,
|
|
"num_tokens": 14286417.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.603499984741211,
|
|
"epoch": 6.757198109153417,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00017555711476969138,
|
|
"loss": 5.0141,
|
|
"mean_token_accuracy": 0.2166544422507286,
|
|
"num_tokens": 14294725.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.599425172805786,
|
|
"epoch": 6.761495487752471,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017525897623043806,
|
|
"loss": 5.0233,
|
|
"mean_token_accuracy": 0.21701672524213791,
|
|
"num_tokens": 14303228.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.623521375656128,
|
|
"epoch": 6.765792866351526,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017496105548571472,
|
|
"loss": 5.0556,
|
|
"mean_token_accuracy": 0.20667367279529572,
|
|
"num_tokens": 14312334.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.654414367675781,
|
|
"epoch": 6.77009024495058,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017466335318606086,
|
|
"loss": 5.0923,
|
|
"mean_token_accuracy": 0.21041001677513121,
|
|
"num_tokens": 14321635.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.690842485427856,
|
|
"epoch": 6.774387623549635,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017436586998153947,
|
|
"loss": 5.1584,
|
|
"mean_token_accuracy": 0.2019079014658928,
|
|
"num_tokens": 14332456.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.63312668800354,
|
|
"epoch": 6.7786850021486895,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00017406860652173495,
|
|
"loss": 4.9893,
|
|
"mean_token_accuracy": 0.21592702567577363,
|
|
"num_tokens": 14341594.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.668723630905151,
|
|
"epoch": 6.782982380747744,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017377156345575176,
|
|
"loss": 5.0984,
|
|
"mean_token_accuracy": 0.2052740067243576,
|
|
"num_tokens": 14351123.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.591213321685791,
|
|
"epoch": 6.787279759346799,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017347474143221338,
|
|
"loss": 5.0159,
|
|
"mean_token_accuracy": 0.21256013661623002,
|
|
"num_tokens": 14360586.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.601018667221069,
|
|
"epoch": 6.791577137945853,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.00017317814109926044,
|
|
"loss": 5.0428,
|
|
"mean_token_accuracy": 0.21152911633253096,
|
|
"num_tokens": 14370101.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.6621605396270756,
|
|
"epoch": 6.795874516544908,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001728817631045495,
|
|
"loss": 5.1173,
|
|
"mean_token_accuracy": 0.21263852417469026,
|
|
"num_tokens": 14380255.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.6238607406616214,
|
|
"epoch": 6.800171895143962,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001725856080952516,
|
|
"loss": 5.0382,
|
|
"mean_token_accuracy": 0.21011119186878205,
|
|
"num_tokens": 14389563.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.593264245986939,
|
|
"epoch": 6.804469273743017,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001722896767180509,
|
|
"loss": 4.9787,
|
|
"mean_token_accuracy": 0.21862697154283522,
|
|
"num_tokens": 14398582.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.552109670639038,
|
|
"epoch": 6.808766652342071,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017199396961914334,
|
|
"loss": 5.0602,
|
|
"mean_token_accuracy": 0.20496242195367814,
|
|
"num_tokens": 14407739.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.656761312484742,
|
|
"epoch": 6.813064030941126,
|
|
"grad_norm": 2.015625,
|
|
"learning_rate": 0.00017169848744423506,
|
|
"loss": 5.0463,
|
|
"mean_token_accuracy": 0.20379938036203385,
|
|
"num_tokens": 14417624.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.690058326721191,
|
|
"epoch": 6.81736140954018,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017140323083854076,
|
|
"loss": 5.1293,
|
|
"mean_token_accuracy": 0.20183698683977128,
|
|
"num_tokens": 14427446.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.630348825454712,
|
|
"epoch": 6.821658788139235,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017110820044678317,
|
|
"loss": 4.9987,
|
|
"mean_token_accuracy": 0.22038342505693437,
|
|
"num_tokens": 14437274.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.584008455276489,
|
|
"epoch": 6.8259561667382895,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017081339691319054,
|
|
"loss": 4.9781,
|
|
"mean_token_accuracy": 0.21599197387695312,
|
|
"num_tokens": 14445786.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.599910116195678,
|
|
"epoch": 6.830253545337344,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00017051882088149612,
|
|
"loss": 5.0143,
|
|
"mean_token_accuracy": 0.21191873997449875,
|
|
"num_tokens": 14454571.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.598539686203003,
|
|
"epoch": 6.834550923936399,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00017022447299493599,
|
|
"loss": 5.0255,
|
|
"mean_token_accuracy": 0.21042714715003968,
|
|
"num_tokens": 14463645.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.576424837112427,
|
|
"epoch": 6.838848302535453,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016993035389624854,
|
|
"loss": 4.972,
|
|
"mean_token_accuracy": 0.22430160790681838,
|
|
"num_tokens": 14472737.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.612935304641724,
|
|
"epoch": 6.843145681134508,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001696364642276722,
|
|
"loss": 5.0689,
|
|
"mean_token_accuracy": 0.20361082404851913,
|
|
"num_tokens": 14482053.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.652880716323852,
|
|
"epoch": 6.847443059733562,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016934280463094448,
|
|
"loss": 5.0964,
|
|
"mean_token_accuracy": 0.2069859728217125,
|
|
"num_tokens": 14491955.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.6282188415527346,
|
|
"epoch": 6.851740438332617,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016904937574730062,
|
|
"loss": 5.003,
|
|
"mean_token_accuracy": 0.2141893520951271,
|
|
"num_tokens": 14501149.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.654011392593384,
|
|
"epoch": 6.856037816931671,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016875617821747208,
|
|
"loss": 5.0484,
|
|
"mean_token_accuracy": 0.214228817820549,
|
|
"num_tokens": 14510134.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.572317886352539,
|
|
"epoch": 6.860335195530726,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016846321268168508,
|
|
"loss": 5.0171,
|
|
"mean_token_accuracy": 0.21570573151111602,
|
|
"num_tokens": 14519842.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.617733335494995,
|
|
"epoch": 6.8646325741297805,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016817047977965905,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.2101440206170082,
|
|
"num_tokens": 14528790.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.656684160232544,
|
|
"epoch": 6.868929952728836,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001678779801506058,
|
|
"loss": 5.1187,
|
|
"mean_token_accuracy": 0.19708767235279084,
|
|
"num_tokens": 14537389.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.633894538879394,
|
|
"epoch": 6.8732273313278895,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00016758571443322774,
|
|
"loss": 4.9895,
|
|
"mean_token_accuracy": 0.21556791812181472,
|
|
"num_tokens": 14545095.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.8732273313278895,
|
|
"eval_entropy": 5.523246242119385,
|
|
"eval_loss": 5.93520975112915,
|
|
"eval_mean_token_accuracy": 0.1779590876319924,
|
|
"eval_num_tokens": 14545095.0,
|
|
"eval_runtime": 2.8329,
|
|
"eval_samples_per_second": 1252.782,
|
|
"eval_steps_per_second": 156.73,
|
|
"step": 8000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11630,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2.2051888527744e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|