c3e92c4ecd
Model: fpadovani/eus-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.8594757198109153,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 2739434655744000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|