c3e92c4ecd
Model: fpadovani/eus-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.42973785990545765,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1381344823296000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|