0b3600b7b3
Model: fpadovani/ita-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.5123550176500253,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.910352373123169,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 21.125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.2192,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11445.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.904936599731445,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 21.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.1143,
|
|
"mean_token_accuracy": 6.73400645609945e-05,
|
|
"num_tokens": 22095.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.916885423660278,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 22.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.1034,
|
|
"mean_token_accuracy": 0.00013709709164686502,
|
|
"num_tokens": 33847.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.0439094543457035,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 27.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.7381,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 44495.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.412609577178955,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 37.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.0504,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 55298.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.753208303451538,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 31.25,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 12.0538,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 66391.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 9.891419410705566,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 3.484375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.9607,
|
|
"mean_token_accuracy": 0.00013166556600481273,
|
|
"num_tokens": 77624.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.698719787597657,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 3.5625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.7344,
|
|
"mean_token_accuracy": 0.00032866199035197496,
|
|
"num_tokens": 89426.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.726075839996337,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.5926,
|
|
"mean_token_accuracy": 0.00417199133662507,
|
|
"num_tokens": 100252.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.712922477722168,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.3898,
|
|
"mean_token_accuracy": 0.014383823238313198,
|
|
"num_tokens": 110578.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.644971942901611,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.2471,
|
|
"mean_token_accuracy": 0.015645577292889356,
|
|
"num_tokens": 121631.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.69424057006836,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1162,
|
|
"mean_token_accuracy": 0.01469698864966631,
|
|
"num_tokens": 134254.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.732514667510987,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0187,
|
|
"mean_token_accuracy": 0.021757309418171646,
|
|
"num_tokens": 144928.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.716106986999511,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8923,
|
|
"mean_token_accuracy": 0.03248250037431717,
|
|
"num_tokens": 155518.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.709668159484863,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.8081,
|
|
"mean_token_accuracy": 0.032918563671410085,
|
|
"num_tokens": 165850.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.714706993103027,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7875,
|
|
"mean_token_accuracy": 0.03219325188547373,
|
|
"num_tokens": 176787.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.714124965667725,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.7108,
|
|
"mean_token_accuracy": 0.035928618907928464,
|
|
"num_tokens": 188345.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.705197811126709,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.6412,
|
|
"mean_token_accuracy": 0.0326865192502737,
|
|
"num_tokens": 199315.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.699298572540282,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4768,
|
|
"mean_token_accuracy": 0.03807435855269432,
|
|
"num_tokens": 209627.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.682019710540771,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3915,
|
|
"mean_token_accuracy": 0.036197875253856183,
|
|
"num_tokens": 220872.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.66194772720337,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2819,
|
|
"mean_token_accuracy": 0.04516553804278374,
|
|
"num_tokens": 233272.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.633659744262696,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1755,
|
|
"mean_token_accuracy": 0.05018138997256756,
|
|
"num_tokens": 243529.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.592283630371094,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0954,
|
|
"mean_token_accuracy": 0.04114586636424065,
|
|
"num_tokens": 254942.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.565635299682617,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9839,
|
|
"mean_token_accuracy": 0.048531080782413485,
|
|
"num_tokens": 266552.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.520887851715088,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8489,
|
|
"mean_token_accuracy": 0.048508816957473756,
|
|
"num_tokens": 277906.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.44383192062378,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7307,
|
|
"mean_token_accuracy": 0.05025425441563129,
|
|
"num_tokens": 289079.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.368312740325928,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6518,
|
|
"mean_token_accuracy": 0.04377192948013544,
|
|
"num_tokens": 300107.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.284347629547119,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4993,
|
|
"mean_token_accuracy": 0.050150758028030394,
|
|
"num_tokens": 310779.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.191308975219727,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.453,
|
|
"mean_token_accuracy": 0.04963008016347885,
|
|
"num_tokens": 323478.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.073376941680909,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.2947,
|
|
"mean_token_accuracy": 0.052390134707093236,
|
|
"num_tokens": 335255.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.928036785125732,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1781,
|
|
"mean_token_accuracy": 0.050037645548582074,
|
|
"num_tokens": 344824.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.778972053527832,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.1288,
|
|
"mean_token_accuracy": 0.04721053130924702,
|
|
"num_tokens": 356283.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.597450828552246,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.0566,
|
|
"mean_token_accuracy": 0.04783501923084259,
|
|
"num_tokens": 368098.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.383034706115723,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9436,
|
|
"mean_token_accuracy": 0.048526646196842195,
|
|
"num_tokens": 380948.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.10937566757202,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8313,
|
|
"mean_token_accuracy": 0.050237832218408586,
|
|
"num_tokens": 390392.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.893278980255127,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.8307,
|
|
"mean_token_accuracy": 0.050915177911520004,
|
|
"num_tokens": 400328.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.709152698516846,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8375,
|
|
"mean_token_accuracy": 0.04621901027858257,
|
|
"num_tokens": 410909.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.523740673065186,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 0.58984375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7759,
|
|
"mean_token_accuracy": 0.047038092836737634,
|
|
"num_tokens": 422726.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.369872570037842,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6757,
|
|
"mean_token_accuracy": 0.05148980468511581,
|
|
"num_tokens": 433376.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.261030673980713,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.7332,
|
|
"mean_token_accuracy": 0.04960624761879444,
|
|
"num_tokens": 443940.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.258826065063477,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 0.58984375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.6401,
|
|
"mean_token_accuracy": 0.048060222342610356,
|
|
"num_tokens": 455016.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.232400035858154,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6756,
|
|
"mean_token_accuracy": 0.05087458901107311,
|
|
"num_tokens": 466380.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.21111011505127,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5595,
|
|
"mean_token_accuracy": 0.049945533461868766,
|
|
"num_tokens": 476915.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.243093967437744,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.6386,
|
|
"mean_token_accuracy": 0.057695179060101506,
|
|
"num_tokens": 488924.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.072712755203247,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5785,
|
|
"mean_token_accuracy": 0.05549042522907257,
|
|
"num_tokens": 498965.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.176321601867675,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.5965,
|
|
"mean_token_accuracy": 0.056973418965935704,
|
|
"num_tokens": 509755.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.122655010223388,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.5713,
|
|
"mean_token_accuracy": 0.05634952038526535,
|
|
"num_tokens": 521221.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.139823246002198,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.6557,
|
|
"mean_token_accuracy": 0.054410668835043904,
|
|
"num_tokens": 533595.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.137359714508056,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.5856,
|
|
"mean_token_accuracy": 0.05684904344379902,
|
|
"num_tokens": 543672.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.149091720581055,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4872,
|
|
"mean_token_accuracy": 0.06253227479755878,
|
|
"num_tokens": 554906.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.084903669357299,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.5417,
|
|
"mean_token_accuracy": 0.06494161449372768,
|
|
"num_tokens": 566419.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.07434377670288,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.5279,
|
|
"mean_token_accuracy": 0.062463199719786644,
|
|
"num_tokens": 578848.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.002639818191529,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4524,
|
|
"mean_token_accuracy": 0.06691358126699924,
|
|
"num_tokens": 590601.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.033667945861817,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.4989,
|
|
"mean_token_accuracy": 0.06721620820462704,
|
|
"num_tokens": 602021.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.998079681396485,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.5903,
|
|
"mean_token_accuracy": 0.06254490688443184,
|
|
"num_tokens": 613966.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.071601390838623,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.4139,
|
|
"mean_token_accuracy": 0.06817254088819028,
|
|
"num_tokens": 624851.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.997099876403809,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.4835,
|
|
"mean_token_accuracy": 0.0678780272603035,
|
|
"num_tokens": 635580.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.94349856376648,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.5534,
|
|
"mean_token_accuracy": 0.06504139304161072,
|
|
"num_tokens": 646927.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.021345949172973,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.4053,
|
|
"mean_token_accuracy": 0.0697047770023346,
|
|
"num_tokens": 658116.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.9564056396484375,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.3887,
|
|
"mean_token_accuracy": 0.06593849398195743,
|
|
"num_tokens": 669054.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.9183282375335695,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.4247,
|
|
"mean_token_accuracy": 0.07052628844976425,
|
|
"num_tokens": 680439.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.927852010726928,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.3485,
|
|
"mean_token_accuracy": 0.0716003067791462,
|
|
"num_tokens": 690909.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.843877267837525,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.3335,
|
|
"mean_token_accuracy": 0.0685461949557066,
|
|
"num_tokens": 701744.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.78822283744812,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.3599,
|
|
"mean_token_accuracy": 0.07047874368727207,
|
|
"num_tokens": 712268.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.879432201385498,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.3937,
|
|
"mean_token_accuracy": 0.06803249828517437,
|
|
"num_tokens": 722978.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.837024545669555,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.3977,
|
|
"mean_token_accuracy": 0.06485099494457244,
|
|
"num_tokens": 734278.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.870750522613525,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.3428,
|
|
"mean_token_accuracy": 0.06625398620963097,
|
|
"num_tokens": 746109.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.815386438369751,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2712,
|
|
"mean_token_accuracy": 0.0703127596527338,
|
|
"num_tokens": 757519.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.838667201995849,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.2327,
|
|
"mean_token_accuracy": 0.06976432576775551,
|
|
"num_tokens": 767665.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.770452690124512,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2605,
|
|
"mean_token_accuracy": 0.07452492453157902,
|
|
"num_tokens": 778880.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.671797084808349,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2654,
|
|
"mean_token_accuracy": 0.07736610174179077,
|
|
"num_tokens": 790360.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.728646850585937,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2348,
|
|
"mean_token_accuracy": 0.07338146939873695,
|
|
"num_tokens": 800889.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.675601196289063,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.2333,
|
|
"mean_token_accuracy": 0.07188612334430218,
|
|
"num_tokens": 811414.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.728438234329223,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.2265,
|
|
"mean_token_accuracy": 0.07272094301879406,
|
|
"num_tokens": 822822.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.590001153945923,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1402,
|
|
"mean_token_accuracy": 0.07824505940079689,
|
|
"num_tokens": 833145.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.747368717193604,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.2475,
|
|
"mean_token_accuracy": 0.0740690290927887,
|
|
"num_tokens": 844015.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.641835784912109,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1787,
|
|
"mean_token_accuracy": 0.07636387497186661,
|
|
"num_tokens": 855806.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.64897780418396,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.098,
|
|
"mean_token_accuracy": 0.07710095196962356,
|
|
"num_tokens": 865750.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.62802677154541,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.2246,
|
|
"mean_token_accuracy": 0.07233364507555962,
|
|
"num_tokens": 877630.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.64155797958374,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1193,
|
|
"mean_token_accuracy": 0.07785095870494843,
|
|
"num_tokens": 888667.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.625726747512817,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.197,
|
|
"mean_token_accuracy": 0.07570679485797882,
|
|
"num_tokens": 900989.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.635288333892822,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0775,
|
|
"mean_token_accuracy": 0.07197484150528907,
|
|
"num_tokens": 911479.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.58501181602478,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1204,
|
|
"mean_token_accuracy": 0.07360159009695053,
|
|
"num_tokens": 922434.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.5460936546325685,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.1144,
|
|
"mean_token_accuracy": 0.07529124468564988,
|
|
"num_tokens": 933405.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.6064252853393555,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.0975,
|
|
"mean_token_accuracy": 0.07624291330575943,
|
|
"num_tokens": 943567.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.480703639984131,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.136,
|
|
"mean_token_accuracy": 0.07384473346173763,
|
|
"num_tokens": 955123.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.631748962402344,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1533,
|
|
"mean_token_accuracy": 0.07767361216247082,
|
|
"num_tokens": 965565.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.526004362106323,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.1039,
|
|
"mean_token_accuracy": 0.07384410724043847,
|
|
"num_tokens": 977101.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.576969480514526,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.1376,
|
|
"mean_token_accuracy": 0.07549539841711521,
|
|
"num_tokens": 987445.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.526590204238891,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0196,
|
|
"mean_token_accuracy": 0.08274288028478623,
|
|
"num_tokens": 999379.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.471691417694092,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.968,
|
|
"mean_token_accuracy": 0.08329149559140206,
|
|
"num_tokens": 1010744.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.54520697593689,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0416,
|
|
"mean_token_accuracy": 0.07827929966151714,
|
|
"num_tokens": 1021808.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.477932929992676,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0811,
|
|
"mean_token_accuracy": 0.07999052368104458,
|
|
"num_tokens": 1033694.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.520880126953125,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.0456,
|
|
"mean_token_accuracy": 0.07764143832027912,
|
|
"num_tokens": 1044690.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.542422342300415,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.0482,
|
|
"mean_token_accuracy": 0.07791347205638885,
|
|
"num_tokens": 1055557.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.409520292282105,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9957,
|
|
"mean_token_accuracy": 0.08100500479340553,
|
|
"num_tokens": 1067304.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.459061002731323,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.021,
|
|
"mean_token_accuracy": 0.0783581517636776,
|
|
"num_tokens": 1078955.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.449764680862427,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0517,
|
|
"mean_token_accuracy": 0.08230436109006405,
|
|
"num_tokens": 1091405.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.463499355316162,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.0248,
|
|
"mean_token_accuracy": 0.08033888339996338,
|
|
"num_tokens": 1103227.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.433458662033081,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.9657,
|
|
"mean_token_accuracy": 0.0861723892390728,
|
|
"num_tokens": 1114485.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5042864346949067,
|
|
"eval_entropy": 7.395277802270788,
|
|
"eval_loss": 7.067502975463867,
|
|
"eval_mean_token_accuracy": 0.08697184946852482,
|
|
"eval_num_tokens": 1114485.0,
|
|
"eval_runtime": 2.1239,
|
|
"eval_samples_per_second": 1477.45,
|
|
"eval_steps_per_second": 185.034,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.405847024917603,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.984,
|
|
"mean_token_accuracy": 0.08936863020062447,
|
|
"num_tokens": 1126789.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.460533142089844,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.966,
|
|
"mean_token_accuracy": 0.08199240490794182,
|
|
"num_tokens": 1138192.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.3393408298492435,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9098,
|
|
"mean_token_accuracy": 0.08162495531141759,
|
|
"num_tokens": 1148526.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.458217191696167,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.0153,
|
|
"mean_token_accuracy": 0.07633640989661217,
|
|
"num_tokens": 1159194.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.33443512916565,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.906,
|
|
"mean_token_accuracy": 0.0897760234773159,
|
|
"num_tokens": 1170601.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.390394878387451,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.9091,
|
|
"mean_token_accuracy": 0.08475598022341728,
|
|
"num_tokens": 1182441.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.325603246688843,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8008,
|
|
"mean_token_accuracy": 0.09372416436672211,
|
|
"num_tokens": 1192530.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.284105587005615,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.9239,
|
|
"mean_token_accuracy": 0.08224731758236885,
|
|
"num_tokens": 1203831.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.360192966461182,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8601,
|
|
"mean_token_accuracy": 0.08579591512680054,
|
|
"num_tokens": 1214164.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.267377424240112,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8831,
|
|
"mean_token_accuracy": 0.08585480228066444,
|
|
"num_tokens": 1226932.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.379234743118286,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.942,
|
|
"mean_token_accuracy": 0.0817857712507248,
|
|
"num_tokens": 1237212.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.295353126525879,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8459,
|
|
"mean_token_accuracy": 0.08828911706805229,
|
|
"num_tokens": 1249340.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.293833637237549,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8602,
|
|
"mean_token_accuracy": 0.0935377262532711,
|
|
"num_tokens": 1260778.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.2916364669799805,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.8675,
|
|
"mean_token_accuracy": 0.0834194079041481,
|
|
"num_tokens": 1272216.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.310083103179932,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.9369,
|
|
"mean_token_accuracy": 0.08788081333041191,
|
|
"num_tokens": 1283558.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.404259157180786,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.9825,
|
|
"mean_token_accuracy": 0.08486812300980091,
|
|
"num_tokens": 1294638.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.435716342926026,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8912,
|
|
"mean_token_accuracy": 0.087406737357378,
|
|
"num_tokens": 1304697.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.253348064422608,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8609,
|
|
"mean_token_accuracy": 0.08688995763659477,
|
|
"num_tokens": 1316618.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.336230707168579,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.9537,
|
|
"mean_token_accuracy": 0.08084713369607925,
|
|
"num_tokens": 1328134.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.371943950653076,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8797,
|
|
"mean_token_accuracy": 0.08257691636681556,
|
|
"num_tokens": 1339101.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.2372519969940186,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.7574,
|
|
"mean_token_accuracy": 0.09110561460256576,
|
|
"num_tokens": 1349181.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.217679405212403,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8774,
|
|
"mean_token_accuracy": 0.08801197558641434,
|
|
"num_tokens": 1361443.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.238507413864136,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8177,
|
|
"mean_token_accuracy": 0.09036832302808762,
|
|
"num_tokens": 1372807.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.285562324523926,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8942,
|
|
"mean_token_accuracy": 0.08551665917038917,
|
|
"num_tokens": 1385509.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.294302272796631,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7994,
|
|
"mean_token_accuracy": 0.09008246585726738,
|
|
"num_tokens": 1396035.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.160209274291992,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7335,
|
|
"mean_token_accuracy": 0.0913459986448288,
|
|
"num_tokens": 1406839.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.1683602809906,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7121,
|
|
"mean_token_accuracy": 0.09201777651906014,
|
|
"num_tokens": 1417708.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.210118246078491,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.08575858026742936,
|
|
"num_tokens": 1428985.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.197644662857056,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.09253060892224312,
|
|
"num_tokens": 1440232.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.127893161773682,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6128,
|
|
"mean_token_accuracy": 0.09752296134829522,
|
|
"num_tokens": 1451735.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.12758059501648,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7665,
|
|
"mean_token_accuracy": 0.0905262552201748,
|
|
"num_tokens": 1463477.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.16529951095581,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7643,
|
|
"mean_token_accuracy": 0.08742906525731087,
|
|
"num_tokens": 1473802.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.178797626495362,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7625,
|
|
"mean_token_accuracy": 0.09097197279334068,
|
|
"num_tokens": 1485519.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.134563350677491,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7571,
|
|
"mean_token_accuracy": 0.09082425311207772,
|
|
"num_tokens": 1496957.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.125902271270752,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7513,
|
|
"mean_token_accuracy": 0.09530752003192902,
|
|
"num_tokens": 1508336.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.134354066848755,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7368,
|
|
"mean_token_accuracy": 0.09136363193392753,
|
|
"num_tokens": 1520016.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.047116708755493,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.5969,
|
|
"mean_token_accuracy": 0.10183367505669594,
|
|
"num_tokens": 1531877.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.992479372024536,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.5929,
|
|
"mean_token_accuracy": 0.09865794181823731,
|
|
"num_tokens": 1541553.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.179808568954468,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.7486,
|
|
"mean_token_accuracy": 0.09588267058134078,
|
|
"num_tokens": 1552491.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.9392517566680905,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6438,
|
|
"mean_token_accuracy": 0.09699270352721215,
|
|
"num_tokens": 1563695.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.107646083831787,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6743,
|
|
"mean_token_accuracy": 0.0914404772222042,
|
|
"num_tokens": 1573467.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.988015794754029,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6471,
|
|
"mean_token_accuracy": 0.10227895081043244,
|
|
"num_tokens": 1584367.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.015262031555176,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5912,
|
|
"mean_token_accuracy": 0.10431398674845696,
|
|
"num_tokens": 1595301.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.049318599700928,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6489,
|
|
"mean_token_accuracy": 0.09553751796483993,
|
|
"num_tokens": 1605916.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.039408111572266,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6366,
|
|
"mean_token_accuracy": 0.09584024250507354,
|
|
"num_tokens": 1616455.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.976061582565308,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6249,
|
|
"mean_token_accuracy": 0.10145259872078896,
|
|
"num_tokens": 1627396.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.117921972274781,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.7499,
|
|
"mean_token_accuracy": 0.09127496853470803,
|
|
"num_tokens": 1639073.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.956200408935547,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.09801545068621635,
|
|
"num_tokens": 1649787.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.9809126377105715,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5655,
|
|
"mean_token_accuracy": 0.09974676370620728,
|
|
"num_tokens": 1658883.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.860607004165649,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5439,
|
|
"mean_token_accuracy": 0.1046278566122055,
|
|
"num_tokens": 1669099.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.945769739151001,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5476,
|
|
"mean_token_accuracy": 0.09769671782851219,
|
|
"num_tokens": 1680227.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.982058095932007,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.10293467938899994,
|
|
"num_tokens": 1689901.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.859155321121216,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5986,
|
|
"mean_token_accuracy": 0.09999214485287666,
|
|
"num_tokens": 1700498.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.901654386520386,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5141,
|
|
"mean_token_accuracy": 0.10296274051070213,
|
|
"num_tokens": 1710715.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.941842555999756,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.6341,
|
|
"mean_token_accuracy": 0.09950209185481071,
|
|
"num_tokens": 1723858.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.89209246635437,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5935,
|
|
"mean_token_accuracy": 0.09372718334197998,
|
|
"num_tokens": 1734678.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.904127931594848,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5056,
|
|
"mean_token_accuracy": 0.1047722913324833,
|
|
"num_tokens": 1743869.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.984526538848877,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.578,
|
|
"mean_token_accuracy": 0.09974783658981323,
|
|
"num_tokens": 1755266.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.783455896377563,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5364,
|
|
"mean_token_accuracy": 0.09974951297044754,
|
|
"num_tokens": 1766089.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.944797992706299,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5431,
|
|
"mean_token_accuracy": 0.10037100091576576,
|
|
"num_tokens": 1776057.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.836930894851685,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5737,
|
|
"mean_token_accuracy": 0.09728048965334893,
|
|
"num_tokens": 1786729.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.9329804420471195,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.508,
|
|
"mean_token_accuracy": 0.10167987123131753,
|
|
"num_tokens": 1797708.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.859876441955566,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5453,
|
|
"mean_token_accuracy": 0.10706769153475762,
|
|
"num_tokens": 1807326.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.783277606964111,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4844,
|
|
"mean_token_accuracy": 0.1034775085747242,
|
|
"num_tokens": 1817266.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.894398736953735,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5,
|
|
"mean_token_accuracy": 0.10923754200339317,
|
|
"num_tokens": 1827840.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.709803342819214,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4748,
|
|
"mean_token_accuracy": 0.1036040298640728,
|
|
"num_tokens": 1838158.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.877896070480347,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5591,
|
|
"mean_token_accuracy": 0.09829444661736489,
|
|
"num_tokens": 1848317.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.768724632263184,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4806,
|
|
"mean_token_accuracy": 0.10388926193118095,
|
|
"num_tokens": 1858751.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.846293878555298,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.493,
|
|
"mean_token_accuracy": 0.10360832959413528,
|
|
"num_tokens": 1870739.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.887697744369507,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.6064,
|
|
"mean_token_accuracy": 0.10239030867815017,
|
|
"num_tokens": 1880995.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.8503830909729,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5576,
|
|
"mean_token_accuracy": 0.1004607506096363,
|
|
"num_tokens": 1892633.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.927896118164062,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6157,
|
|
"mean_token_accuracy": 0.09754905551671982,
|
|
"num_tokens": 1904802.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.896012830734253,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.1034881442785263,
|
|
"num_tokens": 1915755.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.753184366226196,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3595,
|
|
"mean_token_accuracy": 0.1010950192809105,
|
|
"num_tokens": 1926098.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.809195327758789,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5846,
|
|
"mean_token_accuracy": 0.1017203226685524,
|
|
"num_tokens": 1936607.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.739422082901001,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4709,
|
|
"mean_token_accuracy": 0.10663464665412903,
|
|
"num_tokens": 1948596.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.76585898399353,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.5381,
|
|
"mean_token_accuracy": 0.1049344077706337,
|
|
"num_tokens": 1960322.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.844669008255005,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4564,
|
|
"mean_token_accuracy": 0.10585802644491196,
|
|
"num_tokens": 1971052.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.804664039611817,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4861,
|
|
"mean_token_accuracy": 0.09895869046449661,
|
|
"num_tokens": 1980917.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.729268407821655,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4384,
|
|
"mean_token_accuracy": 0.10832197889685631,
|
|
"num_tokens": 1991035.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.7104339599609375,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.458,
|
|
"mean_token_accuracy": 0.10516942515969277,
|
|
"num_tokens": 2000587.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.7780543804168705,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4273,
|
|
"mean_token_accuracy": 0.10677500739693642,
|
|
"num_tokens": 2011087.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.705195903778076,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4673,
|
|
"mean_token_accuracy": 0.10288073197007179,
|
|
"num_tokens": 2023472.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.75528621673584,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.454,
|
|
"mean_token_accuracy": 0.10575027167797088,
|
|
"num_tokens": 2033783.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.657633304595947,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4729,
|
|
"mean_token_accuracy": 0.10640397965908051,
|
|
"num_tokens": 2045882.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.779510593414306,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4781,
|
|
"mean_token_accuracy": 0.10185832753777505,
|
|
"num_tokens": 2056549.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.75550947189331,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4788,
|
|
"mean_token_accuracy": 0.10527065321803093,
|
|
"num_tokens": 2067540.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.6518762588500975,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.3729,
|
|
"mean_token_accuracy": 0.10782607644796371,
|
|
"num_tokens": 2077386.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.694978809356689,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3588,
|
|
"mean_token_accuracy": 0.10642436146736145,
|
|
"num_tokens": 2087777.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.664656162261963,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.351,
|
|
"mean_token_accuracy": 0.10316140428185464,
|
|
"num_tokens": 2099071.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.4923890113830565,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4144,
|
|
"mean_token_accuracy": 0.10309024602174759,
|
|
"num_tokens": 2109168.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.833002996444702,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5138,
|
|
"mean_token_accuracy": 0.1049713984131813,
|
|
"num_tokens": 2120696.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.668496990203858,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4429,
|
|
"mean_token_accuracy": 0.10518230870366096,
|
|
"num_tokens": 2132427.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.701524496078491,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4353,
|
|
"mean_token_accuracy": 0.10702200308442116,
|
|
"num_tokens": 2143120.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.689940118789673,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.10534614622592926,
|
|
"num_tokens": 2154154.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.696073293685913,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5324,
|
|
"mean_token_accuracy": 0.0996658444404602,
|
|
"num_tokens": 2167000.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.711336994171143,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4649,
|
|
"mean_token_accuracy": 0.10657059699296952,
|
|
"num_tokens": 2178100.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.600247192382812,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.10585904195904731,
|
|
"num_tokens": 2189063.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.7853367063734265,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3816,
|
|
"mean_token_accuracy": 0.10893791500065061,
|
|
"num_tokens": 2198302.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.630726861953735,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3312,
|
|
"mean_token_accuracy": 0.1090955302119255,
|
|
"num_tokens": 2209648.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.540408185420145,
|
|
"eval_loss": 6.461416721343994,
|
|
"eval_mean_token_accuracy": 0.11113336287040747,
|
|
"eval_num_tokens": 2209648.0,
|
|
"eval_runtime": 2.0341,
|
|
"eval_samples_per_second": 1542.682,
|
|
"eval_steps_per_second": 193.204,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.596870613098145,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.2431,
|
|
"mean_token_accuracy": 0.11784344986081123,
|
|
"num_tokens": 2220806.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.553163862228393,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.2489,
|
|
"mean_token_accuracy": 0.11016370132565498,
|
|
"num_tokens": 2231806.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.60176043510437,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.2911,
|
|
"mean_token_accuracy": 0.11688948720693589,
|
|
"num_tokens": 2243835.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.443943977355957,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.11431019976735116,
|
|
"num_tokens": 2254722.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.5499800682067875,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 6.1699,
|
|
"mean_token_accuracy": 0.12210036814212799,
|
|
"num_tokens": 2264806.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.655332374572754,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.3597,
|
|
"mean_token_accuracy": 0.10959624648094177,
|
|
"num_tokens": 2277037.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.563287734985352,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.10317928344011307,
|
|
"num_tokens": 2289829.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.605117607116699,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.1877,
|
|
"mean_token_accuracy": 0.11521424874663352,
|
|
"num_tokens": 2300340.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.508002901077271,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.2135,
|
|
"mean_token_accuracy": 0.11831212639808655,
|
|
"num_tokens": 2310009.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.558585786819458,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.2327,
|
|
"mean_token_accuracy": 0.11367388069629669,
|
|
"num_tokens": 2319290.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.563412809371949,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.2095,
|
|
"mean_token_accuracy": 0.11704154834151267,
|
|
"num_tokens": 2329170.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.4528532981872555,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.2172,
|
|
"mean_token_accuracy": 0.11061301082372665,
|
|
"num_tokens": 2339617.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.584239149093628,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.2238,
|
|
"mean_token_accuracy": 0.11742238327860832,
|
|
"num_tokens": 2349548.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.460854816436767,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.18,
|
|
"mean_token_accuracy": 0.11255253925919532,
|
|
"num_tokens": 2359116.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.515554475784302,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.2383,
|
|
"mean_token_accuracy": 0.12132272943854332,
|
|
"num_tokens": 2369412.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.517651605606079,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.3072,
|
|
"mean_token_accuracy": 0.11325916647911072,
|
|
"num_tokens": 2381276.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.56121187210083,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 6.2166,
|
|
"mean_token_accuracy": 0.11995315775275231,
|
|
"num_tokens": 2392135.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.375766229629517,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 6.1213,
|
|
"mean_token_accuracy": 0.12119341865181923,
|
|
"num_tokens": 2402957.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.571408224105835,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.251,
|
|
"mean_token_accuracy": 0.12106918767094613,
|
|
"num_tokens": 2413386.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.413351535797119,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.2446,
|
|
"mean_token_accuracy": 0.11359023898839951,
|
|
"num_tokens": 2424441.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.525660133361816,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.2587,
|
|
"mean_token_accuracy": 0.11170726269483566,
|
|
"num_tokens": 2436144.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.5052775859832765,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 6.1793,
|
|
"mean_token_accuracy": 0.1222833052277565,
|
|
"num_tokens": 2447393.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.421037435531616,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 6.1303,
|
|
"mean_token_accuracy": 0.1208079345524311,
|
|
"num_tokens": 2458632.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.469361782073975,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.221,
|
|
"mean_token_accuracy": 0.11303923651576042,
|
|
"num_tokens": 2469331.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.545333385467529,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 6.2689,
|
|
"mean_token_accuracy": 0.11166509613394737,
|
|
"num_tokens": 2479501.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.431850719451904,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 6.1293,
|
|
"mean_token_accuracy": 0.11632360890507698,
|
|
"num_tokens": 2490087.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.484920310974121,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.2024,
|
|
"mean_token_accuracy": 0.11425360813736915,
|
|
"num_tokens": 2501130.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.501006603240967,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.2999,
|
|
"mean_token_accuracy": 0.11134214028716087,
|
|
"num_tokens": 2513999.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.474369859695434,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 6.2194,
|
|
"mean_token_accuracy": 0.11561294570565224,
|
|
"num_tokens": 2525579.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.4915367603302006,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 6.1905,
|
|
"mean_token_accuracy": 0.11502130404114723,
|
|
"num_tokens": 2536091.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.468376922607422,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.2268,
|
|
"mean_token_accuracy": 0.1178181879222393,
|
|
"num_tokens": 2546073.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.495087146759033,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.1908,
|
|
"mean_token_accuracy": 0.11532010957598686,
|
|
"num_tokens": 2556626.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.426233196258545,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.1999,
|
|
"mean_token_accuracy": 0.1225629799067974,
|
|
"num_tokens": 2567304.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.414515161514283,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 6.0725,
|
|
"mean_token_accuracy": 0.11471373662352562,
|
|
"num_tokens": 2577242.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.41071925163269,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 6.1714,
|
|
"mean_token_accuracy": 0.11379559189081193,
|
|
"num_tokens": 2588330.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.4306882381439205,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 6.1998,
|
|
"mean_token_accuracy": 0.11733074858784676,
|
|
"num_tokens": 2598471.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.489285898208618,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 6.257,
|
|
"mean_token_accuracy": 0.1135383665561676,
|
|
"num_tokens": 2610238.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.389137125015258,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 6.1335,
|
|
"mean_token_accuracy": 0.12369364947080612,
|
|
"num_tokens": 2619870.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.471128797531128,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 6.2621,
|
|
"mean_token_accuracy": 0.11531401872634887,
|
|
"num_tokens": 2631541.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.359742164611816,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 6.0995,
|
|
"mean_token_accuracy": 0.12339793145656586,
|
|
"num_tokens": 2641896.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.430509996414185,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.1842,
|
|
"mean_token_accuracy": 0.1165244683623314,
|
|
"num_tokens": 2654208.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.394809579849243,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.11764412522315978,
|
|
"num_tokens": 2665115.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.455261135101319,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.2527,
|
|
"mean_token_accuracy": 0.11351834759116172,
|
|
"num_tokens": 2676093.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.452110052108765,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 6.197,
|
|
"mean_token_accuracy": 0.12444573938846588,
|
|
"num_tokens": 2687866.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.458801126480102,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 6.2066,
|
|
"mean_token_accuracy": 0.11660922318696976,
|
|
"num_tokens": 2699814.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.44323582649231,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.1839,
|
|
"mean_token_accuracy": 0.12365676537156105,
|
|
"num_tokens": 2710384.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.4240641593933105,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 6.1554,
|
|
"mean_token_accuracy": 0.11770453900098801,
|
|
"num_tokens": 2721442.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.377686071395874,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 6.1657,
|
|
"mean_token_accuracy": 0.11785626113414764,
|
|
"num_tokens": 2733605.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.43232831954956,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 6.2089,
|
|
"mean_token_accuracy": 0.12633543610572814,
|
|
"num_tokens": 2744853.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.320886278152466,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 6.1113,
|
|
"mean_token_accuracy": 0.1273192882537842,
|
|
"num_tokens": 2754477.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.442186880111694,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 6.1173,
|
|
"mean_token_accuracy": 0.1199355199933052,
|
|
"num_tokens": 2765895.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.348427820205688,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 6.1468,
|
|
"mean_token_accuracy": 0.119571004062891,
|
|
"num_tokens": 2776897.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.411334657669068,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.1993,
|
|
"mean_token_accuracy": 0.1178061805665493,
|
|
"num_tokens": 2788666.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.403590869903565,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 6.0816,
|
|
"mean_token_accuracy": 0.11880902722477912,
|
|
"num_tokens": 2799167.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.2923743724823,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 6.0803,
|
|
"mean_token_accuracy": 0.11575406789779663,
|
|
"num_tokens": 2809774.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.41163649559021,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 6.1993,
|
|
"mean_token_accuracy": 0.11564520001411438,
|
|
"num_tokens": 2820515.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.407106590270996,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 6.1398,
|
|
"mean_token_accuracy": 0.11928385570645332,
|
|
"num_tokens": 2832163.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.310460901260376,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 6.1142,
|
|
"mean_token_accuracy": 0.12122111395001411,
|
|
"num_tokens": 2843717.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.301966953277588,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 6.0834,
|
|
"mean_token_accuracy": 0.1179940603673458,
|
|
"num_tokens": 2854489.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.371503496170044,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 6.1238,
|
|
"mean_token_accuracy": 0.12143627852201462,
|
|
"num_tokens": 2865317.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.281271934509277,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 6.0504,
|
|
"mean_token_accuracy": 0.1238139659166336,
|
|
"num_tokens": 2877466.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.346838331222534,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 6.0828,
|
|
"mean_token_accuracy": 0.1219179905951023,
|
|
"num_tokens": 2887698.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.200007104873658,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 6.0367,
|
|
"mean_token_accuracy": 0.12488842830061912,
|
|
"num_tokens": 2899339.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.343645191192627,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 6.0781,
|
|
"mean_token_accuracy": 0.12387225553393363,
|
|
"num_tokens": 2908340.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.327138948440552,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.1782,
|
|
"mean_token_accuracy": 0.12270187363028526,
|
|
"num_tokens": 2920432.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.36779146194458,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 6.1233,
|
|
"mean_token_accuracy": 0.1162177599966526,
|
|
"num_tokens": 2930385.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.394891881942749,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 6.0822,
|
|
"mean_token_accuracy": 0.11857689172029495,
|
|
"num_tokens": 2940760.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.301519727706909,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 6.0849,
|
|
"mean_token_accuracy": 0.12730747163295747,
|
|
"num_tokens": 2952441.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.3621299266815186,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 6.1216,
|
|
"mean_token_accuracy": 0.12057074010372162,
|
|
"num_tokens": 2962766.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.23132815361023,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 6.0094,
|
|
"mean_token_accuracy": 0.1286078929901123,
|
|
"num_tokens": 2974239.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.28206877708435,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 6.0404,
|
|
"mean_token_accuracy": 0.12513453513383865,
|
|
"num_tokens": 2984497.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.3009498596191404,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 6.0282,
|
|
"mean_token_accuracy": 0.12919085174798967,
|
|
"num_tokens": 2995367.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.347341632843017,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 6.073,
|
|
"mean_token_accuracy": 0.12306225821375846,
|
|
"num_tokens": 3007687.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.3034430027008055,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 6.0466,
|
|
"mean_token_accuracy": 0.12317413464188576,
|
|
"num_tokens": 3017505.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.194773483276367,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 5.9897,
|
|
"mean_token_accuracy": 0.1256312608718872,
|
|
"num_tokens": 3028271.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.2702476501464846,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 6.0498,
|
|
"mean_token_accuracy": 0.1274739347398281,
|
|
"num_tokens": 3040440.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.246666383743286,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 6.0101,
|
|
"mean_token_accuracy": 0.12436403408646583,
|
|
"num_tokens": 3050999.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.263060140609741,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 6.0682,
|
|
"mean_token_accuracy": 0.12510961294174194,
|
|
"num_tokens": 3063374.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.368999242782593,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 6.094,
|
|
"mean_token_accuracy": 0.12669392451643943,
|
|
"num_tokens": 3075464.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.1546022415161135,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 5.9634,
|
|
"mean_token_accuracy": 0.12526423931121827,
|
|
"num_tokens": 3086398.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.278340196609497,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 6.1006,
|
|
"mean_token_accuracy": 0.11808581873774529,
|
|
"num_tokens": 3097777.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.323656940460205,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 6.0605,
|
|
"mean_token_accuracy": 0.1321638911962509,
|
|
"num_tokens": 3108711.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.269476699829101,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 6.0955,
|
|
"mean_token_accuracy": 0.1233494944870472,
|
|
"num_tokens": 3119589.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.261565446853638,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.13225764781236649,
|
|
"num_tokens": 3131744.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.313161945343017,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 6.0621,
|
|
"mean_token_accuracy": 0.12190337777137757,
|
|
"num_tokens": 3143526.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.223160982131958,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 5.9978,
|
|
"mean_token_accuracy": 0.12351568043231964,
|
|
"num_tokens": 3154561.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.3151466846466064,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 6.1643,
|
|
"mean_token_accuracy": 0.11945203095674514,
|
|
"num_tokens": 3166052.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.319597864151001,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 6.1199,
|
|
"mean_token_accuracy": 0.12031879499554635,
|
|
"num_tokens": 3179335.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.243300199508667,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.12574711814522743,
|
|
"num_tokens": 3192129.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.348361444473267,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 6.1172,
|
|
"mean_token_accuracy": 0.1257016494870186,
|
|
"num_tokens": 3202792.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.289998245239258,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 6.0474,
|
|
"mean_token_accuracy": 0.12485183402895927,
|
|
"num_tokens": 3213707.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.2773839950561525,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 6.0406,
|
|
"mean_token_accuracy": 0.12844547405838966,
|
|
"num_tokens": 3224531.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.3069030284881595,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 6.0255,
|
|
"mean_token_accuracy": 0.13030240088701248,
|
|
"num_tokens": 3234656.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.21710901260376,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 6.0545,
|
|
"mean_token_accuracy": 0.12597464323043822,
|
|
"num_tokens": 3246745.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.216967058181763,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 6.0314,
|
|
"mean_token_accuracy": 0.12574660703539847,
|
|
"num_tokens": 3257277.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.296077060699463,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 6.001,
|
|
"mean_token_accuracy": 0.1272491253912449,
|
|
"num_tokens": 3267704.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.2856720924377445,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 6.0725,
|
|
"mean_token_accuracy": 0.12544610872864723,
|
|
"num_tokens": 3279772.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.18753604888916,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 5.9753,
|
|
"mean_token_accuracy": 0.1282753072679043,
|
|
"num_tokens": 3290536.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.208110904693603,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 6.0397,
|
|
"mean_token_accuracy": 0.13054657354950905,
|
|
"num_tokens": 3302051.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.247125911712646,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 6.0154,
|
|
"mean_token_accuracy": 0.11876789480447769,
|
|
"num_tokens": 3314324.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5123550176500253,
|
|
"eval_entropy": 6.089648048386319,
|
|
"eval_loss": 6.147760391235352,
|
|
"eval_mean_token_accuracy": 0.12741849596112131,
|
|
"eval_num_tokens": 3314324.0,
|
|
"eval_runtime": 2.0304,
|
|
"eval_samples_per_second": 1545.537,
|
|
"eval_steps_per_second": 193.561,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9910,
|
|
"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": 825837351966720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|