0b3600b7b3
Model: fpadovani/ita-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
10145 lines
276 KiB
JSON
10145 lines
276 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.040342914775593,
|
|
"eval_steps": 500,
|
|
"global_step": 5000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.156036901473999,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 6.0108,
|
|
"mean_token_accuracy": 0.12939631938934326,
|
|
"num_tokens": 3326246.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.238477659225464,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 6.0146,
|
|
"mean_token_accuracy": 0.12993920966982841,
|
|
"num_tokens": 3336308.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.137201929092408,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 5.9478,
|
|
"mean_token_accuracy": 0.13382459282875062,
|
|
"num_tokens": 3347596.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.140363264083862,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 5.9261,
|
|
"mean_token_accuracy": 0.1316240534186363,
|
|
"num_tokens": 3358370.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.282736682891846,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 5.9539,
|
|
"mean_token_accuracy": 0.12983666807413102,
|
|
"num_tokens": 3368630.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.224072694778442,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 5.964,
|
|
"mean_token_accuracy": 0.13069895058870315,
|
|
"num_tokens": 3379952.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.137469005584717,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 5.9064,
|
|
"mean_token_accuracy": 0.1311415933072567,
|
|
"num_tokens": 3390097.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.11315426826477,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.9007,
|
|
"mean_token_accuracy": 0.13429841324687003,
|
|
"num_tokens": 3399778.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.219993686676025,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 6.0492,
|
|
"mean_token_accuracy": 0.1224518284201622,
|
|
"num_tokens": 3410213.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.288167142868042,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 6.0564,
|
|
"mean_token_accuracy": 0.12722626626491546,
|
|
"num_tokens": 3422071.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.228869485855102,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 6.024,
|
|
"mean_token_accuracy": 0.12989916950464248,
|
|
"num_tokens": 3432102.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.2041120529174805,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.12639209032058715,
|
|
"num_tokens": 3442821.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.140823221206665,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 5.9593,
|
|
"mean_token_accuracy": 0.1277303285896778,
|
|
"num_tokens": 3454378.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.172880506515503,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 5.9748,
|
|
"mean_token_accuracy": 0.13318561986088753,
|
|
"num_tokens": 3463933.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.177058744430542,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.9138,
|
|
"mean_token_accuracy": 0.12689697593450547,
|
|
"num_tokens": 3473993.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.158699464797974,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 6.0649,
|
|
"mean_token_accuracy": 0.12153254449367523,
|
|
"num_tokens": 3484882.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.223286962509155,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.9822,
|
|
"mean_token_accuracy": 0.12474498450756073,
|
|
"num_tokens": 3496503.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.219234228134155,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 6.0554,
|
|
"mean_token_accuracy": 0.12866610437631607,
|
|
"num_tokens": 3508696.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.2126219272613525,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.13273679241538047,
|
|
"num_tokens": 3519542.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.218908977508545,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.9417,
|
|
"mean_token_accuracy": 0.12883660346269607,
|
|
"num_tokens": 3531576.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.181266021728516,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.13329532369971275,
|
|
"num_tokens": 3542816.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.208526515960694,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.1266243115067482,
|
|
"num_tokens": 3554100.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.285739994049072,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.12605574503540992,
|
|
"num_tokens": 3565048.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.161268329620361,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.9799,
|
|
"mean_token_accuracy": 0.12972840890288354,
|
|
"num_tokens": 3575587.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.198655223846435,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 6.0417,
|
|
"mean_token_accuracy": 0.1309143140912056,
|
|
"num_tokens": 3587059.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.121437358856201,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.8684,
|
|
"mean_token_accuracy": 0.13101207837462425,
|
|
"num_tokens": 3598492.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.1245574951171875,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 5.9395,
|
|
"mean_token_accuracy": 0.12826012969017028,
|
|
"num_tokens": 3609513.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.186050367355347,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 5.9739,
|
|
"mean_token_accuracy": 0.13249149844050406,
|
|
"num_tokens": 3621327.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.106121635437011,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.887,
|
|
"mean_token_accuracy": 0.13323233276605606,
|
|
"num_tokens": 3632136.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.168408966064453,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.8437,
|
|
"mean_token_accuracy": 0.1332289457321167,
|
|
"num_tokens": 3642271.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.016178035736084,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.9801,
|
|
"mean_token_accuracy": 0.12791188880801202,
|
|
"num_tokens": 3652355.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.278066539764405,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.13114508166909217,
|
|
"num_tokens": 3663714.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.035008478164673,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 5.9067,
|
|
"mean_token_accuracy": 0.1274005278944969,
|
|
"num_tokens": 3674675.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.157523965835571,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.9846,
|
|
"mean_token_accuracy": 0.134155460447073,
|
|
"num_tokens": 3686403.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.123535585403443,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.8754,
|
|
"mean_token_accuracy": 0.14105360731482505,
|
|
"num_tokens": 3697259.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.131226682662964,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.9318,
|
|
"mean_token_accuracy": 0.13263342157006264,
|
|
"num_tokens": 3708536.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.21954231262207,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.13225506469607354,
|
|
"num_tokens": 3719181.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.048918867111206,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.7999,
|
|
"mean_token_accuracy": 0.13783433213829993,
|
|
"num_tokens": 3729325.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.120263910293579,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.9244,
|
|
"mean_token_accuracy": 0.12808132991194726,
|
|
"num_tokens": 3740462.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.212024211883545,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 6.0037,
|
|
"mean_token_accuracy": 0.13034727275371552,
|
|
"num_tokens": 3752523.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.034075117111206,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.8662,
|
|
"mean_token_accuracy": 0.13170775398612022,
|
|
"num_tokens": 3764120.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.1746021747589115,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.9938,
|
|
"mean_token_accuracy": 0.12606719359755517,
|
|
"num_tokens": 3776733.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.14895544052124,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.969,
|
|
"mean_token_accuracy": 0.13328379690647124,
|
|
"num_tokens": 3787799.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.066791725158692,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.8874,
|
|
"mean_token_accuracy": 0.13170522972941398,
|
|
"num_tokens": 3797995.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.203559875488281,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.1316309779882431,
|
|
"num_tokens": 3809580.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.136819458007812,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 6.002,
|
|
"mean_token_accuracy": 0.13009854927659034,
|
|
"num_tokens": 3820150.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.137077331542969,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 5.8522,
|
|
"mean_token_accuracy": 0.1314531646668911,
|
|
"num_tokens": 3830391.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.01658182144165,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.8555,
|
|
"mean_token_accuracy": 0.13897303864359856,
|
|
"num_tokens": 3842773.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.169336891174316,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.9072,
|
|
"mean_token_accuracy": 0.13076598420739174,
|
|
"num_tokens": 3853646.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.141702365875244,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.8672,
|
|
"mean_token_accuracy": 0.1321739748120308,
|
|
"num_tokens": 3865319.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.057036447525024,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.9197,
|
|
"mean_token_accuracy": 0.12733314409852028,
|
|
"num_tokens": 3877107.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.127471590042115,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.9594,
|
|
"mean_token_accuracy": 0.1326076626777649,
|
|
"num_tokens": 3889334.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.109628105163575,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.8521,
|
|
"mean_token_accuracy": 0.13769446834921836,
|
|
"num_tokens": 3899482.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.055948114395141,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.7791,
|
|
"mean_token_accuracy": 0.13959548696875573,
|
|
"num_tokens": 3910216.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.133174705505371,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.1310464061796665,
|
|
"num_tokens": 3922246.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.121117448806762,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.14053154289722442,
|
|
"num_tokens": 3933688.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.121936845779419,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.13004440814256668,
|
|
"num_tokens": 3944168.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.069381046295166,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.8582,
|
|
"mean_token_accuracy": 0.13762850761413575,
|
|
"num_tokens": 3954346.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.156457567214966,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.9712,
|
|
"mean_token_accuracy": 0.12970432341098787,
|
|
"num_tokens": 3966375.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.138356399536133,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.9372,
|
|
"mean_token_accuracy": 0.13539533764123918,
|
|
"num_tokens": 3977030.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.0566565990448,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.8489,
|
|
"mean_token_accuracy": 0.13358401656150817,
|
|
"num_tokens": 3988064.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.10338978767395,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.8977,
|
|
"mean_token_accuracy": 0.13457708284258843,
|
|
"num_tokens": 3997994.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.171204137802124,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.9426,
|
|
"mean_token_accuracy": 0.12793719545006751,
|
|
"num_tokens": 4008645.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.067741918563843,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.891,
|
|
"mean_token_accuracy": 0.13499666452407838,
|
|
"num_tokens": 4020387.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.933810758590698,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.7667,
|
|
"mean_token_accuracy": 0.1390545152127743,
|
|
"num_tokens": 4032720.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.113948631286621,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.13772407844662665,
|
|
"num_tokens": 4043438.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.984061193466187,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.8149,
|
|
"mean_token_accuracy": 0.14559042751789092,
|
|
"num_tokens": 4054467.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.068659162521362,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.8142,
|
|
"mean_token_accuracy": 0.13721002638339996,
|
|
"num_tokens": 4066418.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.145805835723877,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.9205,
|
|
"mean_token_accuracy": 0.13175863698124884,
|
|
"num_tokens": 4078770.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.075065279006958,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.9581,
|
|
"mean_token_accuracy": 0.12940308153629304,
|
|
"num_tokens": 4089998.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.088465833663941,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 5.8615,
|
|
"mean_token_accuracy": 0.13679955005645753,
|
|
"num_tokens": 4101409.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.117992305755616,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.9812,
|
|
"mean_token_accuracy": 0.12942875251173974,
|
|
"num_tokens": 4113598.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.047978687286377,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.9059,
|
|
"mean_token_accuracy": 0.13164842873811722,
|
|
"num_tokens": 4124639.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.22193021774292,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.9702,
|
|
"mean_token_accuracy": 0.12647129818797112,
|
|
"num_tokens": 4136280.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.056543493270874,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.8227,
|
|
"mean_token_accuracy": 0.1319723516702652,
|
|
"num_tokens": 4148251.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.069836711883545,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.9088,
|
|
"mean_token_accuracy": 0.13684113174676896,
|
|
"num_tokens": 4159298.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.108313369750976,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.8093,
|
|
"mean_token_accuracy": 0.13893356025218964,
|
|
"num_tokens": 4169974.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.052180051803589,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.8247,
|
|
"mean_token_accuracy": 0.14235706329345704,
|
|
"num_tokens": 4181442.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.107061386108398,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.8668,
|
|
"mean_token_accuracy": 0.13770687878131865,
|
|
"num_tokens": 4191535.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.075644683837891,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.8944,
|
|
"mean_token_accuracy": 0.13668857961893083,
|
|
"num_tokens": 4201741.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.0178862571716305,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.7844,
|
|
"mean_token_accuracy": 0.13908884152770043,
|
|
"num_tokens": 4211762.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.073857307434082,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.8684,
|
|
"mean_token_accuracy": 0.1342634789645672,
|
|
"num_tokens": 4222322.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.1019774913787845,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.9438,
|
|
"mean_token_accuracy": 0.13611835464835167,
|
|
"num_tokens": 4232166.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.0883242130279545,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.8871,
|
|
"mean_token_accuracy": 0.1351541370153427,
|
|
"num_tokens": 4243651.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.0817262649536135,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 5.814,
|
|
"mean_token_accuracy": 0.14198361784219743,
|
|
"num_tokens": 4255137.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.938052940368652,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.8279,
|
|
"mean_token_accuracy": 0.13684711009263992,
|
|
"num_tokens": 4264904.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.035186624526977,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.7888,
|
|
"mean_token_accuracy": 0.13858872056007385,
|
|
"num_tokens": 4275305.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.053544712066651,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.8805,
|
|
"mean_token_accuracy": 0.13514195308089255,
|
|
"num_tokens": 4287038.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.167809200286865,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.12615374028682708,
|
|
"num_tokens": 4298668.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.978178691864014,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.7499,
|
|
"mean_token_accuracy": 0.1383742056787014,
|
|
"num_tokens": 4309429.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.073020601272583,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.1353800967335701,
|
|
"num_tokens": 4321137.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.074612426757812,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.12799250036478044,
|
|
"num_tokens": 4331249.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.084528923034668,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.8556,
|
|
"mean_token_accuracy": 0.13444280624389648,
|
|
"num_tokens": 4342154.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.1047789573669435,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.13036157563328743,
|
|
"num_tokens": 4353906.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.083949089050293,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.1291046179831028,
|
|
"num_tokens": 4366416.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.076101493835449,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.8428,
|
|
"mean_token_accuracy": 0.13770969063043595,
|
|
"num_tokens": 4376743.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.963961230383979,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 5.636,
|
|
"mean_token_accuracy": 0.14866483873791164,
|
|
"num_tokens": 4385379.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.011814737319947,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.5817,
|
|
"mean_token_accuracy": 0.14293505102396012,
|
|
"num_tokens": 4396872.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.032746171951294,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.5241,
|
|
"mean_token_accuracy": 0.14778304547071458,
|
|
"num_tokens": 4407075.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.942056798934937,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.5591,
|
|
"mean_token_accuracy": 0.15164274126291274,
|
|
"num_tokens": 4418020.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.723532737363082,
|
|
"eval_loss": 5.9471025466918945,
|
|
"eval_mean_token_accuracy": 0.1362081553796472,
|
|
"eval_num_tokens": 4418020.0,
|
|
"eval_runtime": 2.045,
|
|
"eval_samples_per_second": 1534.452,
|
|
"eval_steps_per_second": 192.173,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.940024185180664,
|
|
"epoch": 2.0211800302571863,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004860483957827243,
|
|
"loss": 5.5724,
|
|
"mean_token_accuracy": 0.14962393194437026,
|
|
"num_tokens": 4429740.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.0355141162872314,
|
|
"epoch": 2.026222894604135,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004859105618998735,
|
|
"loss": 5.5662,
|
|
"mean_token_accuracy": 0.14324932247400285,
|
|
"num_tokens": 4440183.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.9932300567626955,
|
|
"epoch": 2.031265758951084,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004857720725021311,
|
|
"loss": 5.5371,
|
|
"mean_token_accuracy": 0.14567766189575196,
|
|
"num_tokens": 4451750.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.812967920303345,
|
|
"epoch": 2.036308623298033,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048563292801992534,
|
|
"loss": 5.4972,
|
|
"mean_token_accuracy": 0.14970174431800842,
|
|
"num_tokens": 4462071.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.048516750335693,
|
|
"epoch": 2.0413514876449823,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004854931288857203,
|
|
"loss": 5.6861,
|
|
"mean_token_accuracy": 0.1418111339211464,
|
|
"num_tokens": 4474241.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.946160411834716,
|
|
"epoch": 2.0463943519919314,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048535267553401505,
|
|
"loss": 5.5198,
|
|
"mean_token_accuracy": 0.14917661547660827,
|
|
"num_tokens": 4487083.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.8958487033844,
|
|
"epoch": 2.0514372163388805,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048521156840134193,
|
|
"loss": 5.5706,
|
|
"mean_token_accuracy": 0.1516394168138504,
|
|
"num_tokens": 4497181.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.988409042358398,
|
|
"epoch": 2.0564800806858297,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000485069807926265,
|
|
"loss": 5.5509,
|
|
"mean_token_accuracy": 0.14955907613039016,
|
|
"num_tokens": 4508250.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.9201305389404295,
|
|
"epoch": 2.0615229450327788,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004849273945493792,
|
|
"loss": 5.5546,
|
|
"mean_token_accuracy": 0.15315290838479995,
|
|
"num_tokens": 4519778.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.911941957473755,
|
|
"epoch": 2.066565809379728,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004847843287133084,
|
|
"loss": 5.5338,
|
|
"mean_token_accuracy": 0.14484044089913367,
|
|
"num_tokens": 4530879.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.95937180519104,
|
|
"epoch": 2.0716086737266766,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004846406108627048,
|
|
"loss": 5.6061,
|
|
"mean_token_accuracy": 0.15488057881593703,
|
|
"num_tokens": 4542560.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.874370336532593,
|
|
"epoch": 2.0766515380736257,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048449624144424665,
|
|
"loss": 5.504,
|
|
"mean_token_accuracy": 0.15539904236793517,
|
|
"num_tokens": 4552213.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.914573717117309,
|
|
"epoch": 2.081694402420575,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048435122090663744,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.15514215230941772,
|
|
"num_tokens": 4562798.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.906200838088989,
|
|
"epoch": 2.086737266767524,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048420554970060424,
|
|
"loss": 5.5632,
|
|
"mean_token_accuracy": 0.142371766269207,
|
|
"num_tokens": 4574534.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.931839179992676,
|
|
"epoch": 2.091780131114473,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048405922827889677,
|
|
"loss": 5.5342,
|
|
"mean_token_accuracy": 0.14170490056276322,
|
|
"num_tokens": 4586239.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.883896541595459,
|
|
"epoch": 2.096822995461422,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048391225709628515,
|
|
"loss": 5.5777,
|
|
"mean_token_accuracy": 0.14845428690314294,
|
|
"num_tokens": 4598018.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.90255184173584,
|
|
"epoch": 2.1018658598083713,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004837646366095593,
|
|
"loss": 5.5498,
|
|
"mean_token_accuracy": 0.14273814857006073,
|
|
"num_tokens": 4608985.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.972240877151489,
|
|
"epoch": 2.1069087241553204,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048361636727752716,
|
|
"loss": 5.6074,
|
|
"mean_token_accuracy": 0.1534451276063919,
|
|
"num_tokens": 4621103.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.869103240966797,
|
|
"epoch": 2.1119515885022695,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004834674495610131,
|
|
"loss": 5.5826,
|
|
"mean_token_accuracy": 0.1498872831463814,
|
|
"num_tokens": 4632244.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.895373868942261,
|
|
"epoch": 2.116994452849218,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048331788392285693,
|
|
"loss": 5.5021,
|
|
"mean_token_accuracy": 0.15359969437122345,
|
|
"num_tokens": 4642387.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.9281104564666744,
|
|
"epoch": 2.1220373171961673,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048316767082791205,
|
|
"loss": 5.6388,
|
|
"mean_token_accuracy": 0.14638803526759148,
|
|
"num_tokens": 4655181.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.84404764175415,
|
|
"epoch": 2.1270801815431164,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048301681074304427,
|
|
"loss": 5.5393,
|
|
"mean_token_accuracy": 0.14985166043043135,
|
|
"num_tokens": 4665664.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.846418714523315,
|
|
"epoch": 2.1321230458900655,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048286530413713016,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.15592919886112214,
|
|
"num_tokens": 4677683.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.902777290344238,
|
|
"epoch": 2.1371659102370146,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048271315148105574,
|
|
"loss": 5.5582,
|
|
"mean_token_accuracy": 0.1461687430739403,
|
|
"num_tokens": 4688891.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.846302080154419,
|
|
"epoch": 2.1422087745839637,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000482560353247715,
|
|
"loss": 5.5815,
|
|
"mean_token_accuracy": 0.14932963848114014,
|
|
"num_tokens": 4700765.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.894233274459839,
|
|
"epoch": 2.147251638930913,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004824069099120084,
|
|
"loss": 5.5988,
|
|
"mean_token_accuracy": 0.14528623223304749,
|
|
"num_tokens": 4712004.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.949212646484375,
|
|
"epoch": 2.152294503277862,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.15119172781705856,
|
|
"num_tokens": 4722427.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.893136692047119,
|
|
"epoch": 2.157337367624811,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004820980898431227,
|
|
"loss": 5.5911,
|
|
"mean_token_accuracy": 0.15657991245388986,
|
|
"num_tokens": 4733815.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.882743215560913,
|
|
"epoch": 2.1623802319717598,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004819427140697635,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.14681099206209183,
|
|
"num_tokens": 4745070.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.947874927520752,
|
|
"epoch": 2.167423096318709,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004817866951136753,
|
|
"loss": 5.6263,
|
|
"mean_token_accuracy": 0.14205105900764464,
|
|
"num_tokens": 4756066.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.946091842651367,
|
|
"epoch": 2.172465960665658,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00048163003345976847,
|
|
"loss": 5.6249,
|
|
"mean_token_accuracy": 0.14056657999753952,
|
|
"num_tokens": 4767153.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.883247137069702,
|
|
"epoch": 2.177508825012607,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004814727295949512,
|
|
"loss": 5.5406,
|
|
"mean_token_accuracy": 0.14822641015052795,
|
|
"num_tokens": 4777554.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.855099630355835,
|
|
"epoch": 2.1825516893595562,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048131478400812735,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.14793250113725662,
|
|
"num_tokens": 4788881.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.950084686279297,
|
|
"epoch": 2.1875945537065054,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004811561971901956,
|
|
"loss": 5.5617,
|
|
"mean_token_accuracy": 0.14551590234041215,
|
|
"num_tokens": 4798680.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.90962929725647,
|
|
"epoch": 2.1926374180534545,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004809969696340474,
|
|
"loss": 5.5424,
|
|
"mean_token_accuracy": 0.14731819033622742,
|
|
"num_tokens": 4808096.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.853575611114502,
|
|
"epoch": 2.1976802824004036,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048083710183456554,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.15830671787261963,
|
|
"num_tokens": 4818778.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.8463743209838865,
|
|
"epoch": 2.2027231467473527,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048067659428862306,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.15291596204042435,
|
|
"num_tokens": 4828586.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.879240798950195,
|
|
"epoch": 2.2077660110943014,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048051544749508096,
|
|
"loss": 5.5964,
|
|
"mean_token_accuracy": 0.1415431432425976,
|
|
"num_tokens": 4839419.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.8528117656707765,
|
|
"epoch": 2.2128088754412505,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004803536619547872,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.1551996648311615,
|
|
"num_tokens": 4849700.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.927838945388794,
|
|
"epoch": 2.2178517397881996,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000480191238170575,
|
|
"loss": 5.6865,
|
|
"mean_token_accuracy": 0.14072581827640535,
|
|
"num_tokens": 4860261.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.942123126983643,
|
|
"epoch": 2.2228946041351487,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004800281766472612,
|
|
"loss": 5.606,
|
|
"mean_token_accuracy": 0.14752493649721146,
|
|
"num_tokens": 4872277.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.9327787399292,
|
|
"epoch": 2.227937468482098,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00047986447789164486,
|
|
"loss": 5.6514,
|
|
"mean_token_accuracy": 0.1517902985215187,
|
|
"num_tokens": 4883512.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.868672037124634,
|
|
"epoch": 2.232980332829047,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004797001424125054,
|
|
"loss": 5.5791,
|
|
"mean_token_accuracy": 0.15303744301199912,
|
|
"num_tokens": 4896197.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.887090826034546,
|
|
"epoch": 2.238023197175996,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047953517072060124,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.15226840674877168,
|
|
"num_tokens": 4907397.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.871148490905762,
|
|
"epoch": 2.243066061522945,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004793695633286684,
|
|
"loss": 5.6034,
|
|
"mean_token_accuracy": 0.151021346449852,
|
|
"num_tokens": 4919008.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.838630294799804,
|
|
"epoch": 2.2481089258698943,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047920332075141825,
|
|
"loss": 5.5393,
|
|
"mean_token_accuracy": 0.1499991774559021,
|
|
"num_tokens": 4930877.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.875366067886352,
|
|
"epoch": 2.253151790216843,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004790364435055367,
|
|
"loss": 5.4496,
|
|
"mean_token_accuracy": 0.1593221455812454,
|
|
"num_tokens": 4942274.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.812536191940308,
|
|
"epoch": 2.258194654563792,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000478868932109682,
|
|
"loss": 5.5358,
|
|
"mean_token_accuracy": 0.1527387946844101,
|
|
"num_tokens": 4953740.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.902262592315674,
|
|
"epoch": 2.263237518910741,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004787007870844835,
|
|
"loss": 5.5712,
|
|
"mean_token_accuracy": 0.1538744106888771,
|
|
"num_tokens": 4964760.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.868859958648682,
|
|
"epoch": 2.2682803832576903,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047853200895253976,
|
|
"loss": 5.6361,
|
|
"mean_token_accuracy": 0.14738022089004515,
|
|
"num_tokens": 4976404.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.898247623443604,
|
|
"epoch": 2.2733232476046394,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00047836259823841716,
|
|
"loss": 5.517,
|
|
"mean_token_accuracy": 0.14728200286626816,
|
|
"num_tokens": 4987552.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.844000768661499,
|
|
"epoch": 2.2783661119515886,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047819255546864797,
|
|
"loss": 5.5286,
|
|
"mean_token_accuracy": 0.14984559565782546,
|
|
"num_tokens": 4998912.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.887818384170532,
|
|
"epoch": 2.2834089762985377,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004780218811717293,
|
|
"loss": 5.5116,
|
|
"mean_token_accuracy": 0.1581410750746727,
|
|
"num_tokens": 5009556.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.834558296203613,
|
|
"epoch": 2.288451840645487,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047785057587812057,
|
|
"loss": 5.5505,
|
|
"mean_token_accuracy": 0.15399980396032334,
|
|
"num_tokens": 5021569.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.890166759490967,
|
|
"epoch": 2.293494704992436,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047767864012024266,
|
|
"loss": 5.6262,
|
|
"mean_token_accuracy": 0.14727912992238998,
|
|
"num_tokens": 5032994.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.945313930511475,
|
|
"epoch": 2.2985375693393846,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047750607443247585,
|
|
"loss": 5.5581,
|
|
"mean_token_accuracy": 0.14870548546314238,
|
|
"num_tokens": 5044681.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.793630743026734,
|
|
"epoch": 2.3035804336863337,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047733287935115837,
|
|
"loss": 5.4719,
|
|
"mean_token_accuracy": 0.15815871804952622,
|
|
"num_tokens": 5055509.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.76172251701355,
|
|
"epoch": 2.308623298033283,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004771590554145844,
|
|
"loss": 5.5002,
|
|
"mean_token_accuracy": 0.16068245470523834,
|
|
"num_tokens": 5066903.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.854470109939575,
|
|
"epoch": 2.313666162380232,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047698460316300285,
|
|
"loss": 5.5504,
|
|
"mean_token_accuracy": 0.14702785909175872,
|
|
"num_tokens": 5078411.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.862625408172607,
|
|
"epoch": 2.318709026727181,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004768095231386153,
|
|
"loss": 5.5565,
|
|
"mean_token_accuracy": 0.14897067844867706,
|
|
"num_tokens": 5089811.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.883305549621582,
|
|
"epoch": 2.32375189107413,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004766338158855744,
|
|
"loss": 5.5733,
|
|
"mean_token_accuracy": 0.15708377063274384,
|
|
"num_tokens": 5101196.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.832779026031494,
|
|
"epoch": 2.3287947554210793,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004764574819499826,
|
|
"loss": 5.5436,
|
|
"mean_token_accuracy": 0.15225967913866043,
|
|
"num_tokens": 5112379.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.895620441436767,
|
|
"epoch": 2.3338376197680284,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047628052187988955,
|
|
"loss": 5.6705,
|
|
"mean_token_accuracy": 0.141206493973732,
|
|
"num_tokens": 5125065.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.997529363632202,
|
|
"epoch": 2.338880484114977,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004761029362252914,
|
|
"loss": 5.6291,
|
|
"mean_token_accuracy": 0.14648888558149337,
|
|
"num_tokens": 5137471.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.9020641326904295,
|
|
"epoch": 2.343923348461926,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047592472553812856,
|
|
"loss": 5.5518,
|
|
"mean_token_accuracy": 0.15038047060370446,
|
|
"num_tokens": 5147682.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.817275094985962,
|
|
"epoch": 2.3489662128088753,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004757458903722837,
|
|
"loss": 5.5417,
|
|
"mean_token_accuracy": 0.14693662226200105,
|
|
"num_tokens": 5156143.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.934580278396607,
|
|
"epoch": 2.3540090771558244,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00047556643128358075,
|
|
"loss": 5.5985,
|
|
"mean_token_accuracy": 0.14841473698616028,
|
|
"num_tokens": 5167462.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.744073915481567,
|
|
"epoch": 2.3590519415027735,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047538634882978277,
|
|
"loss": 5.5169,
|
|
"mean_token_accuracy": 0.15127300173044206,
|
|
"num_tokens": 5178030.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.785821199417114,
|
|
"epoch": 2.3640948058497226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047520564357059013,
|
|
"loss": 5.5364,
|
|
"mean_token_accuracy": 0.15155871361494064,
|
|
"num_tokens": 5188296.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.808376026153565,
|
|
"epoch": 2.3691376701966718,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004750243160676389,
|
|
"loss": 5.5265,
|
|
"mean_token_accuracy": 0.1540423169732094,
|
|
"num_tokens": 5199058.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.812775659561157,
|
|
"epoch": 2.374180534543621,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004748423668844992,
|
|
"loss": 5.498,
|
|
"mean_token_accuracy": 0.15063866078853608,
|
|
"num_tokens": 5209314.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.90487060546875,
|
|
"epoch": 2.37922339889057,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004746597965866733,
|
|
"loss": 5.6177,
|
|
"mean_token_accuracy": 0.14394214525818824,
|
|
"num_tokens": 5220248.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.914368104934693,
|
|
"epoch": 2.384266263237519,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004744766057415938,
|
|
"loss": 5.5947,
|
|
"mean_token_accuracy": 0.1400570899248123,
|
|
"num_tokens": 5230874.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.826127243041992,
|
|
"epoch": 2.389309127584468,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004742927949186222,
|
|
"loss": 5.5239,
|
|
"mean_token_accuracy": 0.15305102318525315,
|
|
"num_tokens": 5243328.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.763830518722534,
|
|
"epoch": 2.394351991931417,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047410836468904664,
|
|
"loss": 5.4773,
|
|
"mean_token_accuracy": 0.14843705296516418,
|
|
"num_tokens": 5254357.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.795266103744507,
|
|
"epoch": 2.399394856278366,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047392331562608073,
|
|
"loss": 5.4802,
|
|
"mean_token_accuracy": 0.15369822680950165,
|
|
"num_tokens": 5264492.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.744703674316407,
|
|
"epoch": 2.404437720625315,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004737376483048609,
|
|
"loss": 5.443,
|
|
"mean_token_accuracy": 0.1589686468243599,
|
|
"num_tokens": 5274816.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.8192870140075685,
|
|
"epoch": 2.4094805849722642,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004735513633024458,
|
|
"loss": 5.5429,
|
|
"mean_token_accuracy": 0.14981934279203415,
|
|
"num_tokens": 5286855.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.903432035446167,
|
|
"epoch": 2.4145234493192134,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004733644611978134,
|
|
"loss": 5.6537,
|
|
"mean_token_accuracy": 0.13768935427069665,
|
|
"num_tokens": 5299594.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.8120519638061525,
|
|
"epoch": 2.4195663136661625,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004731769425718597,
|
|
"loss": 5.4793,
|
|
"mean_token_accuracy": 0.15393216013908387,
|
|
"num_tokens": 5310078.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.746055555343628,
|
|
"epoch": 2.4246091780131116,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000472988808007397,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.15920527279376984,
|
|
"num_tokens": 5319544.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.794832992553711,
|
|
"epoch": 2.4296520423600603,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00047280005808915176,
|
|
"loss": 5.568,
|
|
"mean_token_accuracy": 0.14523061513900756,
|
|
"num_tokens": 5331025.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.790472984313965,
|
|
"epoch": 2.4346949067070094,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004726106934037631,
|
|
"loss": 5.5414,
|
|
"mean_token_accuracy": 0.15091371536254883,
|
|
"num_tokens": 5343086.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.923413705825806,
|
|
"epoch": 2.4397377710539585,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004724207145397809,
|
|
"loss": 5.5993,
|
|
"mean_token_accuracy": 0.15322286635637283,
|
|
"num_tokens": 5353598.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.803432273864746,
|
|
"epoch": 2.4447806354009076,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047223012208766384,
|
|
"loss": 5.4755,
|
|
"mean_token_accuracy": 0.14711593165993692,
|
|
"num_tokens": 5363607.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.779847431182861,
|
|
"epoch": 2.4498234997478567,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004720389166397777,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.15192387402057647,
|
|
"num_tokens": 5375198.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.7408387660980225,
|
|
"epoch": 2.454866364094806,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00047184709879039325,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.149568809568882,
|
|
"num_tokens": 5386058.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.9129396915435795,
|
|
"epoch": 2.459909228441755,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047165466913568497,
|
|
"loss": 5.5563,
|
|
"mean_token_accuracy": 0.1510582074522972,
|
|
"num_tokens": 5395720.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.781081771850586,
|
|
"epoch": 2.464952092788704,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004714616282737285,
|
|
"loss": 5.5995,
|
|
"mean_token_accuracy": 0.14857714772224426,
|
|
"num_tokens": 5406252.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.839043188095093,
|
|
"epoch": 2.469994957135653,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004712679768044995,
|
|
"loss": 5.603,
|
|
"mean_token_accuracy": 0.14704830050468445,
|
|
"num_tokens": 5418125.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.849780893325805,
|
|
"epoch": 2.4750378214826023,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000471073715329871,
|
|
"loss": 5.5515,
|
|
"mean_token_accuracy": 0.15433166921138763,
|
|
"num_tokens": 5428095.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.860912895202636,
|
|
"epoch": 2.4800806858295514,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004708788444536123,
|
|
"loss": 5.5654,
|
|
"mean_token_accuracy": 0.1464594319462776,
|
|
"num_tokens": 5438359.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.805293464660645,
|
|
"epoch": 2.4851235501765,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004706833647813865,
|
|
"loss": 5.4992,
|
|
"mean_token_accuracy": 0.15880793184041977,
|
|
"num_tokens": 5448559.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.751124143600464,
|
|
"epoch": 2.4901664145234492,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00047048727692074893,
|
|
"loss": 5.5168,
|
|
"mean_token_accuracy": 0.14838636070489883,
|
|
"num_tokens": 5459739.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.7891887664794925,
|
|
"epoch": 2.4952092788703983,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004702905814811452,
|
|
"loss": 5.4923,
|
|
"mean_token_accuracy": 0.14752392023801802,
|
|
"num_tokens": 5471160.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.707677030563355,
|
|
"epoch": 2.5002521432173475,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00047009327907390926,
|
|
"loss": 5.5503,
|
|
"mean_token_accuracy": 0.14980040788650512,
|
|
"num_tokens": 5482344.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.856988191604614,
|
|
"epoch": 2.5052950075642966,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004698953703122616,
|
|
"loss": 5.5528,
|
|
"mean_token_accuracy": 0.1503003254532814,
|
|
"num_tokens": 5493111.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.655979204177856,
|
|
"epoch": 2.5103378719112457,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046969685581130715,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.1560885176062584,
|
|
"num_tokens": 5503877.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.831366395950317,
|
|
"epoch": 2.515380736258195,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046949773618803353,
|
|
"loss": 5.5051,
|
|
"mean_token_accuracy": 0.15249394997954369,
|
|
"num_tokens": 5514067.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.860077047348023,
|
|
"epoch": 2.5204236006051435,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00046929801206130925,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.1510472685098648,
|
|
"num_tokens": 5525534.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.5204236006051435,
|
|
"eval_entropy": 5.669652560285029,
|
|
"eval_loss": 5.856481552124023,
|
|
"eval_mean_token_accuracy": 0.14118836258962258,
|
|
"eval_num_tokens": 5525534.0,
|
|
"eval_runtime": 2.0404,
|
|
"eval_samples_per_second": 1537.926,
|
|
"eval_steps_per_second": 192.608,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.819248008728027,
|
|
"epoch": 2.5254664649520926,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046909768405188135,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.15700920820236205,
|
|
"num_tokens": 5535525.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.769597578048706,
|
|
"epoch": 2.5305093292990417,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046889675278237404,
|
|
"loss": 5.52,
|
|
"mean_token_accuracy": 0.15728203654289247,
|
|
"num_tokens": 5546129.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.912272644042969,
|
|
"epoch": 2.535552193645991,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004686952188772863,
|
|
"loss": 5.6652,
|
|
"mean_token_accuracy": 0.14891352951526643,
|
|
"num_tokens": 5557548.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.8772557258605955,
|
|
"epoch": 2.54059505799294,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046849308296299005,
|
|
"loss": 5.5734,
|
|
"mean_token_accuracy": 0.15616171360015868,
|
|
"num_tokens": 5567367.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.810514688491821,
|
|
"epoch": 2.545637922339889,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004682903456677285,
|
|
"loss": 5.4811,
|
|
"mean_token_accuracy": 0.15235700085759163,
|
|
"num_tokens": 5577489.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.77357177734375,
|
|
"epoch": 2.550680786686838,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00046808700762161385,
|
|
"loss": 5.5288,
|
|
"mean_token_accuracy": 0.1467547282576561,
|
|
"num_tokens": 5587877.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.770196580886841,
|
|
"epoch": 2.5557236510337873,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004678830694566253,
|
|
"loss": 5.5394,
|
|
"mean_token_accuracy": 0.15670244991779328,
|
|
"num_tokens": 5597863.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.807074022293091,
|
|
"epoch": 2.5607665153807364,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.153423710167408,
|
|
"num_tokens": 5608141.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.787803220748901,
|
|
"epoch": 2.5658093797276855,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004674733953072682,
|
|
"loss": 5.6115,
|
|
"mean_token_accuracy": 0.14963685870170593,
|
|
"num_tokens": 5619660.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.947618913650513,
|
|
"epoch": 2.5708522440746346,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004672676605961763,
|
|
"loss": 5.6018,
|
|
"mean_token_accuracy": 0.14351876080036163,
|
|
"num_tokens": 5630271.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.773510217666626,
|
|
"epoch": 2.5758951084215833,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046706132831276007,
|
|
"loss": 5.587,
|
|
"mean_token_accuracy": 0.15054969489574432,
|
|
"num_tokens": 5642594.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.778890609741211,
|
|
"epoch": 2.5809379727685324,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046685439909830486,
|
|
"loss": 5.5241,
|
|
"mean_token_accuracy": 0.15210181772708892,
|
|
"num_tokens": 5653361.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.767368650436401,
|
|
"epoch": 2.5859808371154815,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004666468735959516,
|
|
"loss": 5.438,
|
|
"mean_token_accuracy": 0.1562154620885849,
|
|
"num_tokens": 5664167.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.761620044708252,
|
|
"epoch": 2.5910237014624307,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004664387524506943,
|
|
"loss": 5.5465,
|
|
"mean_token_accuracy": 0.14990140795707702,
|
|
"num_tokens": 5675903.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.845316219329834,
|
|
"epoch": 2.59606656580938,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004662300363093784,
|
|
"loss": 5.522,
|
|
"mean_token_accuracy": 0.15081020295619965,
|
|
"num_tokens": 5686319.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.9148040294647215,
|
|
"epoch": 2.601109430156329,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004660207258206983,
|
|
"loss": 5.6162,
|
|
"mean_token_accuracy": 0.14688173905014992,
|
|
"num_tokens": 5696000.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.7588094711303714,
|
|
"epoch": 2.606152294503278,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00046581082163519585,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.15163008272647857,
|
|
"num_tokens": 5708145.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.725979423522949,
|
|
"epoch": 2.6111951588502267,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004656003244052582,
|
|
"loss": 5.4874,
|
|
"mean_token_accuracy": 0.1598169982433319,
|
|
"num_tokens": 5718053.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.767901945114136,
|
|
"epoch": 2.616238023197176,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004653892347851155,
|
|
"loss": 5.5493,
|
|
"mean_token_accuracy": 0.15450764447450638,
|
|
"num_tokens": 5728723.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.755375957489013,
|
|
"epoch": 2.621280887544125,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004651775534308392,
|
|
"loss": 5.475,
|
|
"mean_token_accuracy": 0.16065322309732438,
|
|
"num_tokens": 5739414.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.794290542602539,
|
|
"epoch": 2.626323751891074,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004649652810003398,
|
|
"loss": 5.625,
|
|
"mean_token_accuracy": 0.14363892748951912,
|
|
"num_tokens": 5751247.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.791951465606689,
|
|
"epoch": 2.631366616238023,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00046475241815336504,
|
|
"loss": 5.4823,
|
|
"mean_token_accuracy": 0.15644117295742035,
|
|
"num_tokens": 5761791.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.740055847167969,
|
|
"epoch": 2.6364094805849723,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004645389655514974,
|
|
"loss": 5.4744,
|
|
"mean_token_accuracy": 0.14935198575258254,
|
|
"num_tokens": 5773597.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.794382095336914,
|
|
"epoch": 2.6414523449319214,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004643249238581525,
|
|
"loss": 5.4981,
|
|
"mean_token_accuracy": 0.15026383846998215,
|
|
"num_tokens": 5782963.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.6868040561676025,
|
|
"epoch": 2.6464952092788705,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004641102937385769,
|
|
"loss": 5.5234,
|
|
"mean_token_accuracy": 0.14880652129650115,
|
|
"num_tokens": 5793474.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.7942564487457275,
|
|
"epoch": 2.6515380736258196,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004638950758598461,
|
|
"loss": 5.5086,
|
|
"mean_token_accuracy": 0.1518484354019165,
|
|
"num_tokens": 5804018.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.857076978683471,
|
|
"epoch": 2.6565809379727687,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046367927089086203,
|
|
"loss": 5.5453,
|
|
"mean_token_accuracy": 0.14937240481376649,
|
|
"num_tokens": 5815897.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.783289623260498,
|
|
"epoch": 2.661623802319718,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046346287950235163,
|
|
"loss": 5.5718,
|
|
"mean_token_accuracy": 0.14845054149627684,
|
|
"num_tokens": 5825883.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.835809564590454,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004632459023668642,
|
|
"loss": 5.534,
|
|
"mean_token_accuracy": 0.15753208100795746,
|
|
"num_tokens": 5836668.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.812521171569824,
|
|
"epoch": 2.6717095310136156,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004630283401587699,
|
|
"loss": 5.6635,
|
|
"mean_token_accuracy": 0.14314896017313003,
|
|
"num_tokens": 5848497.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.7993275165557865,
|
|
"epoch": 2.6767523953605648,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004628101935542568,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.15989356487989426,
|
|
"num_tokens": 5859231.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.795991325378418,
|
|
"epoch": 2.681795259707514,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046259146323132986,
|
|
"loss": 5.6306,
|
|
"mean_token_accuracy": 0.14957452565431595,
|
|
"num_tokens": 5871190.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.786652040481568,
|
|
"epoch": 2.686838124054463,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046237214986980776,
|
|
"loss": 5.4843,
|
|
"mean_token_accuracy": 0.15117899924516678,
|
|
"num_tokens": 5881788.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.762012195587158,
|
|
"epoch": 2.691880988401412,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004621522541513216,
|
|
"loss": 5.4693,
|
|
"mean_token_accuracy": 0.1548840656876564,
|
|
"num_tokens": 5891289.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.71791353225708,
|
|
"epoch": 2.696923852748361,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004619317767593123,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.15003726333379747,
|
|
"num_tokens": 5903594.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.874940586090088,
|
|
"epoch": 2.70196671709531,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046171071837902873,
|
|
"loss": 5.6303,
|
|
"mean_token_accuracy": 0.15058379024267196,
|
|
"num_tokens": 5914592.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.747613334655762,
|
|
"epoch": 2.707009581442259,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004614890796975254,
|
|
"loss": 5.4503,
|
|
"mean_token_accuracy": 0.15491787791252137,
|
|
"num_tokens": 5927137.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.760874032974243,
|
|
"epoch": 2.712052445789208,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004612668614036605,
|
|
"loss": 5.5606,
|
|
"mean_token_accuracy": 0.15194035768508912,
|
|
"num_tokens": 5938730.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.808653068542481,
|
|
"epoch": 2.7170953101361572,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004610440641880934,
|
|
"loss": 5.5755,
|
|
"mean_token_accuracy": 0.14960916787385942,
|
|
"num_tokens": 5948912.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.733116531372071,
|
|
"epoch": 2.7221381744831064,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00046082068874328324,
|
|
"loss": 5.5035,
|
|
"mean_token_accuracy": 0.1540943205356598,
|
|
"num_tokens": 5960253.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.814432811737061,
|
|
"epoch": 2.7271810388300555,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046059673576348596,
|
|
"loss": 5.5983,
|
|
"mean_token_accuracy": 0.14576624780893327,
|
|
"num_tokens": 5972615.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.867447233200073,
|
|
"epoch": 2.7322239031770046,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00046037220594475264,
|
|
"loss": 5.5646,
|
|
"mean_token_accuracy": 0.1537248432636261,
|
|
"num_tokens": 5985023.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.686407423019409,
|
|
"epoch": 2.7372667675239537,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000460147099984927,
|
|
"loss": 5.4651,
|
|
"mean_token_accuracy": 0.1570134088397026,
|
|
"num_tokens": 5996445.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.789271497726441,
|
|
"epoch": 2.742309631870903,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004599214185836438,
|
|
"loss": 5.5116,
|
|
"mean_token_accuracy": 0.14551102817058564,
|
|
"num_tokens": 6007385.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.780018663406372,
|
|
"epoch": 2.747352496217852,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00045969516244232574,
|
|
"loss": 5.5484,
|
|
"mean_token_accuracy": 0.15154093354940415,
|
|
"num_tokens": 6018174.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.726502370834351,
|
|
"epoch": 2.752395360564801,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045946833226418235,
|
|
"loss": 5.469,
|
|
"mean_token_accuracy": 0.14863964468240737,
|
|
"num_tokens": 6030049.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.716791200637817,
|
|
"epoch": 2.7574382249117497,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004592409287542069,
|
|
"loss": 5.5229,
|
|
"mean_token_accuracy": 0.15207778811454772,
|
|
"num_tokens": 6040155.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.703477764129639,
|
|
"epoch": 2.762481089258699,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004590129526191748,
|
|
"loss": 5.5284,
|
|
"mean_token_accuracy": 0.15693228095769882,
|
|
"num_tokens": 6050569.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.795703887939453,
|
|
"epoch": 2.767523953605648,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004587844045676413,
|
|
"loss": 5.4791,
|
|
"mean_token_accuracy": 0.15692086666822433,
|
|
"num_tokens": 6060871.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.748446369171143,
|
|
"epoch": 2.772566817952597,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00045855528530993874,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.16212048679590224,
|
|
"num_tokens": 6073045.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.743357181549072,
|
|
"epoch": 2.777609682299546,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045832559555817525,
|
|
"loss": 5.5294,
|
|
"mean_token_accuracy": 0.15036371648311614,
|
|
"num_tokens": 6084798.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.762336921691895,
|
|
"epoch": 2.7826525466464953,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045809533602623185,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.15212155729532242,
|
|
"num_tokens": 6095831.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.75358247756958,
|
|
"epoch": 2.7876954109934444,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004578645074297603,
|
|
"loss": 5.4389,
|
|
"mean_token_accuracy": 0.1555684432387352,
|
|
"num_tokens": 6106958.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.71989336013794,
|
|
"epoch": 2.792738275340393,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045763311048618144,
|
|
"loss": 5.473,
|
|
"mean_token_accuracy": 0.1608845993876457,
|
|
"num_tokens": 6118165.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.665394067764282,
|
|
"epoch": 2.797781139687342,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004574011459146822,
|
|
"loss": 5.4641,
|
|
"mean_token_accuracy": 0.1566280245780945,
|
|
"num_tokens": 6129297.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.766199827194214,
|
|
"epoch": 2.8028240040342913,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00045716861443621386,
|
|
"loss": 5.4923,
|
|
"mean_token_accuracy": 0.15070500820875168,
|
|
"num_tokens": 6140071.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.747651100158691,
|
|
"epoch": 2.8078668683812404,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00045693551677348967,
|
|
"loss": 5.4975,
|
|
"mean_token_accuracy": 0.15279183983802797,
|
|
"num_tokens": 6150940.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.80690107345581,
|
|
"epoch": 2.8129097327281896,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004567018536509823,
|
|
"loss": 5.4803,
|
|
"mean_token_accuracy": 0.15254825353622437,
|
|
"num_tokens": 6161966.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.824843215942383,
|
|
"epoch": 2.8179525970751387,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00045646762579492245,
|
|
"loss": 5.5348,
|
|
"mean_token_accuracy": 0.15181106626987456,
|
|
"num_tokens": 6174455.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.693896532058716,
|
|
"epoch": 2.822995461422088,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045623283393329546,
|
|
"loss": 5.4716,
|
|
"mean_token_accuracy": 0.15222429782152175,
|
|
"num_tokens": 6184789.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.757958793640137,
|
|
"epoch": 2.828038325769037,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045599747879584,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.1462457463145256,
|
|
"num_tokens": 6195704.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.788491535186767,
|
|
"epoch": 2.833081190115986,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004557615611140451,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.15523429661989213,
|
|
"num_tokens": 6206290.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.7461964130401615,
|
|
"epoch": 2.838124054462935,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004555250816211485,
|
|
"loss": 5.5513,
|
|
"mean_token_accuracy": 0.1454817198216915,
|
|
"num_tokens": 6216660.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.743470144271851,
|
|
"epoch": 2.8431669188098843,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00045528804105213395,
|
|
"loss": 5.4974,
|
|
"mean_token_accuracy": 0.15376585721969604,
|
|
"num_tokens": 6228005.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.701382303237915,
|
|
"epoch": 2.848209783156833,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004550504401437289,
|
|
"loss": 5.452,
|
|
"mean_token_accuracy": 0.15123870223760605,
|
|
"num_tokens": 6239447.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.765235710144043,
|
|
"epoch": 2.853252647503782,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00045481227963440235,
|
|
"loss": 5.5613,
|
|
"mean_token_accuracy": 0.14397587329149247,
|
|
"num_tokens": 6252312.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.774335527420044,
|
|
"epoch": 2.858295511850731,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00045457356026436295,
|
|
"loss": 5.4998,
|
|
"mean_token_accuracy": 0.1525854289531708,
|
|
"num_tokens": 6264131.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.689094018936157,
|
|
"epoch": 2.8633383761976803,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045433428277555577,
|
|
"loss": 5.4477,
|
|
"mean_token_accuracy": 0.15649830549955368,
|
|
"num_tokens": 6275161.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.7043664932250975,
|
|
"epoch": 2.8683812405446294,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004540944479116609,
|
|
"loss": 5.4985,
|
|
"mean_token_accuracy": 0.1511241763830185,
|
|
"num_tokens": 6285655.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.798206949234009,
|
|
"epoch": 2.8734241048915785,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004538540564180906,
|
|
"loss": 5.5862,
|
|
"mean_token_accuracy": 0.15425564870238304,
|
|
"num_tokens": 6296532.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.878371000289917,
|
|
"epoch": 2.8784669692385276,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004536131090419871,
|
|
"loss": 5.6173,
|
|
"mean_token_accuracy": 0.14916557520627977,
|
|
"num_tokens": 6308229.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.743052625656128,
|
|
"epoch": 2.8835098335854763,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004533716065322205,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.15372492074966432,
|
|
"num_tokens": 6321257.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.792555570602417,
|
|
"epoch": 2.8885526979324254,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004531295496393861,
|
|
"loss": 5.492,
|
|
"mean_token_accuracy": 0.15237339586019516,
|
|
"num_tokens": 6332040.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.679398250579834,
|
|
"epoch": 2.8935955622793745,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004528869391158023,
|
|
"loss": 5.3587,
|
|
"mean_token_accuracy": 0.16076318472623824,
|
|
"num_tokens": 6342925.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.777550983428955,
|
|
"epoch": 2.8986384266263236,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00045264377571550834,
|
|
"loss": 5.5478,
|
|
"mean_token_accuracy": 0.15306758880615234,
|
|
"num_tokens": 6354263.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.823110675811767,
|
|
"epoch": 2.9036812909732728,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004524000601942615,
|
|
"loss": 5.5441,
|
|
"mean_token_accuracy": 0.14977220743894576,
|
|
"num_tokens": 6365497.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.803255605697632,
|
|
"epoch": 2.908724155320222,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004521557933095354,
|
|
"loss": 5.594,
|
|
"mean_token_accuracy": 0.1505901463329792,
|
|
"num_tokens": 6376501.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.776741027832031,
|
|
"epoch": 2.913767019667171,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000451910975820517,
|
|
"loss": 5.5313,
|
|
"mean_token_accuracy": 0.1469830572605133,
|
|
"num_tokens": 6387632.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.669276332855224,
|
|
"epoch": 2.91880988401412,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045166560848810476,
|
|
"loss": 5.5222,
|
|
"mean_token_accuracy": 0.16028138548135756,
|
|
"num_tokens": 6398813.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.841599655151367,
|
|
"epoch": 2.9238527483610692,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045141969207490606,
|
|
"loss": 5.4377,
|
|
"mean_token_accuracy": 0.1571269378066063,
|
|
"num_tokens": 6409532.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.751657485961914,
|
|
"epoch": 2.9288956127080183,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004511732273452347,
|
|
"loss": 5.5746,
|
|
"mean_token_accuracy": 0.14663779512047767,
|
|
"num_tokens": 6420755.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.788761901855469,
|
|
"epoch": 2.9339384770549675,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004509262150651089,
|
|
"loss": 5.6045,
|
|
"mean_token_accuracy": 0.15201486200094222,
|
|
"num_tokens": 6430662.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.798057413101196,
|
|
"epoch": 2.938981341401916,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045067865600224833,
|
|
"loss": 5.4817,
|
|
"mean_token_accuracy": 0.1538977727293968,
|
|
"num_tokens": 6441230.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.799156188964844,
|
|
"epoch": 2.9440242057488653,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004504305509260724,
|
|
"loss": 5.6031,
|
|
"mean_token_accuracy": 0.14440229386091233,
|
|
"num_tokens": 6452574.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.729610824584961,
|
|
"epoch": 2.9490670700958144,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004501819006076974,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.15835335105657578,
|
|
"num_tokens": 6462562.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.711238670349121,
|
|
"epoch": 2.9541099344427635,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00044993270581993413,
|
|
"loss": 5.4884,
|
|
"mean_token_accuracy": 0.15400700718164445,
|
|
"num_tokens": 6473594.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.761794567108154,
|
|
"epoch": 2.9591527987897126,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00044968296733728587,
|
|
"loss": 5.5127,
|
|
"mean_token_accuracy": 0.15556546151638032,
|
|
"num_tokens": 6484129.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.786276531219483,
|
|
"epoch": 2.9641956631366617,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004494326859359456,
|
|
"loss": 5.4905,
|
|
"mean_token_accuracy": 0.14901109039783478,
|
|
"num_tokens": 6495584.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.75063648223877,
|
|
"epoch": 2.969238527483611,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.5692,
|
|
"mean_token_accuracy": 0.1525542750954628,
|
|
"num_tokens": 6507872.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.712465095520019,
|
|
"epoch": 2.9742813918305595,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004489304974903953,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.15936507284641266,
|
|
"num_tokens": 6519172.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.729972171783447,
|
|
"epoch": 2.9793242561775086,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004486785920069985,
|
|
"loss": 5.5385,
|
|
"mean_token_accuracy": 0.15972347259521485,
|
|
"num_tokens": 6530151.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.8090736865997314,
|
|
"epoch": 2.9843671205244577,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004484261467265313,
|
|
"loss": 5.5077,
|
|
"mean_token_accuracy": 0.1548163615167141,
|
|
"num_tokens": 6541881.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.719935655593872,
|
|
"epoch": 2.989409984871407,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00044817316243359956,
|
|
"loss": 5.5406,
|
|
"mean_token_accuracy": 0.15042774602770806,
|
|
"num_tokens": 6553390.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.774695491790771,
|
|
"epoch": 2.994452849218356,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044791963991448427,
|
|
"loss": 5.3679,
|
|
"mean_token_accuracy": 0.1631093814969063,
|
|
"num_tokens": 6563754.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.682390308380127,
|
|
"epoch": 2.999495713565305,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044766557995713915,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.1516885504126549,
|
|
"num_tokens": 6574124.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.759289847479926,
|
|
"epoch": 3.0040342914775593,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004474109833511885,
|
|
"loss": 5.2792,
|
|
"mean_token_accuracy": 0.1634783943494161,
|
|
"num_tokens": 6582795.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.714936923980713,
|
|
"epoch": 3.0090771558245084,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004471558508879244,
|
|
"loss": 5.2117,
|
|
"mean_token_accuracy": 0.1708674907684326,
|
|
"num_tokens": 6593649.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.6874351978302,
|
|
"epoch": 3.0141200201714575,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004469001833603045,
|
|
"loss": 5.2149,
|
|
"mean_token_accuracy": 0.17300137430429458,
|
|
"num_tokens": 6606370.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.653540849685669,
|
|
"epoch": 3.0191628845184066,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004466439815629493,
|
|
"loss": 5.1578,
|
|
"mean_token_accuracy": 0.17079648822546006,
|
|
"num_tokens": 6616599.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.7674613952636715,
|
|
"epoch": 3.0242057488653553,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044638724629214,
|
|
"loss": 5.2355,
|
|
"mean_token_accuracy": 0.16412699967622757,
|
|
"num_tokens": 6627707.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0242057488653553,
|
|
"eval_entropy": 5.550955645000662,
|
|
"eval_loss": 5.774935245513916,
|
|
"eval_mean_token_accuracy": 0.14621713238059716,
|
|
"eval_num_tokens": 6627707.0,
|
|
"eval_runtime": 2.0361,
|
|
"eval_samples_per_second": 1541.159,
|
|
"eval_steps_per_second": 193.013,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.6996255874633786,
|
|
"epoch": 3.0292486132123044,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044612997834581557,
|
|
"loss": 5.1529,
|
|
"mean_token_accuracy": 0.17394840568304062,
|
|
"num_tokens": 6639001.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.701579523086548,
|
|
"epoch": 3.0342914775592535,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044587217852357086,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.1673397347331047,
|
|
"num_tokens": 6649318.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.657785272598266,
|
|
"epoch": 3.0393343419062027,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044561384762665363,
|
|
"loss": 5.2632,
|
|
"mean_token_accuracy": 0.1646798849105835,
|
|
"num_tokens": 6660584.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.613374662399292,
|
|
"epoch": 3.044377206253152,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004453549864579623,
|
|
"loss": 5.0913,
|
|
"mean_token_accuracy": 0.1776850089430809,
|
|
"num_tokens": 6671361.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.696940183639526,
|
|
"epoch": 3.049420070600101,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004450955958220433,
|
|
"loss": 5.3329,
|
|
"mean_token_accuracy": 0.16356808543205262,
|
|
"num_tokens": 6681896.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.701268482208252,
|
|
"epoch": 3.05446293494705,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00044483567652508864,
|
|
"loss": 5.1842,
|
|
"mean_token_accuracy": 0.16942804604768752,
|
|
"num_tokens": 6693360.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.665386390686035,
|
|
"epoch": 3.059505799293999,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004445752293749337,
|
|
"loss": 5.2571,
|
|
"mean_token_accuracy": 0.16399620920419694,
|
|
"num_tokens": 6704921.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.693010950088501,
|
|
"epoch": 3.0645486636409482,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004443142551810541,
|
|
"loss": 5.2148,
|
|
"mean_token_accuracy": 0.1694386675953865,
|
|
"num_tokens": 6715659.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.593599557876587,
|
|
"epoch": 3.069591527987897,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044405275475456365,
|
|
"loss": 5.126,
|
|
"mean_token_accuracy": 0.1801849588751793,
|
|
"num_tokens": 6726431.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.649939250946045,
|
|
"epoch": 3.074634392334846,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004437907289082116,
|
|
"loss": 5.2022,
|
|
"mean_token_accuracy": 0.17012674659490584,
|
|
"num_tokens": 6738002.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.657159042358399,
|
|
"epoch": 3.079677256681795,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00044352817845638053,
|
|
"loss": 5.1971,
|
|
"mean_token_accuracy": 0.16744016408920287,
|
|
"num_tokens": 6748509.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.687095594406128,
|
|
"epoch": 3.0847201210287443,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004432651042150833,
|
|
"loss": 5.2427,
|
|
"mean_token_accuracy": 0.1699002429842949,
|
|
"num_tokens": 6759258.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.641091060638428,
|
|
"epoch": 3.0897629853756934,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044300150700196054,
|
|
"loss": 5.2279,
|
|
"mean_token_accuracy": 0.17298754155635834,
|
|
"num_tokens": 6771263.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.6690654277801515,
|
|
"epoch": 3.0948058497226425,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004427373876362786,
|
|
"loss": 5.2101,
|
|
"mean_token_accuracy": 0.167030069231987,
|
|
"num_tokens": 6779965.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.671913099288941,
|
|
"epoch": 3.0998487140695916,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004424727469389265,
|
|
"loss": 5.1788,
|
|
"mean_token_accuracy": 0.16704553961753846,
|
|
"num_tokens": 6791521.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.699006128311157,
|
|
"epoch": 3.1048915784165407,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 5.2602,
|
|
"mean_token_accuracy": 0.16808926612138747,
|
|
"num_tokens": 6803537.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.655160617828369,
|
|
"epoch": 3.10993444276349,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004419419048408673,
|
|
"loss": 5.2629,
|
|
"mean_token_accuracy": 0.16783561706542968,
|
|
"num_tokens": 6813811.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.640321731567383,
|
|
"epoch": 3.1149773071104385,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004416757050900297,
|
|
"loss": 5.185,
|
|
"mean_token_accuracy": 0.17083243727684022,
|
|
"num_tokens": 6825422.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.772631692886352,
|
|
"epoch": 3.1200201714573876,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004414089873072557,
|
|
"loss": 5.3198,
|
|
"mean_token_accuracy": 0.16099680662155152,
|
|
"num_tokens": 6837086.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.663555145263672,
|
|
"epoch": 3.1250630358043368,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004411417523215105,
|
|
"loss": 5.1729,
|
|
"mean_token_accuracy": 0.17167633026838303,
|
|
"num_tokens": 6847195.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.600523853302002,
|
|
"epoch": 3.130105900151286,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004408740009633665,
|
|
"loss": 5.1639,
|
|
"mean_token_accuracy": 0.1768056869506836,
|
|
"num_tokens": 6857492.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.670230293273926,
|
|
"epoch": 3.135148764498235,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00044060573406500117,
|
|
"loss": 5.3509,
|
|
"mean_token_accuracy": 0.16179293692111968,
|
|
"num_tokens": 6869190.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.66859188079834,
|
|
"epoch": 3.140191628845184,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004403369524601941,
|
|
"loss": 5.1614,
|
|
"mean_token_accuracy": 0.16615760028362275,
|
|
"num_tokens": 6880728.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.568915605545044,
|
|
"epoch": 3.145234493192133,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004400676569843249,
|
|
"loss": 5.1865,
|
|
"mean_token_accuracy": 0.16992096602916718,
|
|
"num_tokens": 6892043.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.664674949645996,
|
|
"epoch": 3.1502773575390823,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043979784847436994,
|
|
"loss": 5.2156,
|
|
"mean_token_accuracy": 0.1754838913679123,
|
|
"num_tokens": 6903913.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.530419921875,
|
|
"epoch": 3.1553202218860315,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004395275277689004,
|
|
"loss": 5.1804,
|
|
"mean_token_accuracy": 0.1666817456483841,
|
|
"num_tokens": 6913951.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.55042519569397,
|
|
"epoch": 3.16036308623298,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004392566957080793,
|
|
"loss": 5.1931,
|
|
"mean_token_accuracy": 0.16878220587968826,
|
|
"num_tokens": 6924762.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.6149509906768795,
|
|
"epoch": 3.1654059505799292,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00043898535313365884,
|
|
"loss": 5.1803,
|
|
"mean_token_accuracy": 0.17030122727155686,
|
|
"num_tokens": 6935845.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.6124639987945555,
|
|
"epoch": 3.1704488149268784,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000438713500888978,
|
|
"loss": 5.1967,
|
|
"mean_token_accuracy": 0.17351544946432113,
|
|
"num_tokens": 6947338.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.64640326499939,
|
|
"epoch": 3.1754916792738275,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004384411398189599,
|
|
"loss": 5.2366,
|
|
"mean_token_accuracy": 0.17089353054761885,
|
|
"num_tokens": 6958084.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.615694141387939,
|
|
"epoch": 3.1805345436207766,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000438168270770109,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.17027769088745118,
|
|
"num_tokens": 6968063.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.64213342666626,
|
|
"epoch": 3.1855774079677257,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004378948945905085,
|
|
"loss": 5.1711,
|
|
"mean_token_accuracy": 0.17558542191982268,
|
|
"num_tokens": 6978424.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.526204586029053,
|
|
"epoch": 3.190620272314675,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00043762101212981805,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.16841456294059753,
|
|
"num_tokens": 6991067.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.586109018325805,
|
|
"epoch": 3.195663136661624,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043734662423927054,
|
|
"loss": 5.2506,
|
|
"mean_token_accuracy": 0.17426552772521972,
|
|
"num_tokens": 7000642.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.699530220031738,
|
|
"epoch": 3.200706001008573,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004370717317716699,
|
|
"loss": 5.2937,
|
|
"mean_token_accuracy": 0.16766004413366317,
|
|
"num_tokens": 7011305.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.63154559135437,
|
|
"epoch": 3.205748865355522,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004367963355813882,
|
|
"loss": 5.2288,
|
|
"mean_token_accuracy": 0.16498964726924897,
|
|
"num_tokens": 7021452.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.57880973815918,
|
|
"epoch": 3.210791729702471,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043652043652436323,
|
|
"loss": 5.2948,
|
|
"mean_token_accuracy": 0.16183717250823976,
|
|
"num_tokens": 7033189.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.734592485427856,
|
|
"epoch": 3.21583459404942,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004362440354580956,
|
|
"loss": 5.2501,
|
|
"mean_token_accuracy": 0.16903538703918458,
|
|
"num_tokens": 7044153.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.65629997253418,
|
|
"epoch": 3.220877458396369,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004359671332416462,
|
|
"loss": 5.2856,
|
|
"mean_token_accuracy": 0.16131064891815186,
|
|
"num_tokens": 7056679.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.612703657150268,
|
|
"epoch": 3.225920322743318,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004356897307356335,
|
|
"loss": 5.1575,
|
|
"mean_token_accuracy": 0.17671172618865966,
|
|
"num_tokens": 7067428.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.645615196228027,
|
|
"epoch": 3.2309631870902673,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004354118288022309,
|
|
"loss": 5.2659,
|
|
"mean_token_accuracy": 0.16396681368350982,
|
|
"num_tokens": 7079196.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.637091732025146,
|
|
"epoch": 3.2360060514372164,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043513342830516405,
|
|
"loss": 5.1711,
|
|
"mean_token_accuracy": 0.17157938182353974,
|
|
"num_tokens": 7088936.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.623890018463134,
|
|
"epoch": 3.2410489157841655,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000434854530109708,
|
|
"loss": 5.2732,
|
|
"mean_token_accuracy": 0.16736230850219727,
|
|
"num_tokens": 7101504.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.7050158977508545,
|
|
"epoch": 3.2460917801311147,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00043457513508268487,
|
|
"loss": 5.2356,
|
|
"mean_token_accuracy": 0.16740143150091172,
|
|
"num_tokens": 7114121.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.51312255859375,
|
|
"epoch": 3.2511346444780633,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043429524409246083,
|
|
"loss": 5.1568,
|
|
"mean_token_accuracy": 0.16881831139326095,
|
|
"num_tokens": 7123321.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.617919874191284,
|
|
"epoch": 3.2561775088250124,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004340148580089436,
|
|
"loss": 5.2248,
|
|
"mean_token_accuracy": 0.1702039957046509,
|
|
"num_tokens": 7134145.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.745028924942017,
|
|
"epoch": 3.2612203731719616,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004337339777035794,
|
|
"loss": 5.2635,
|
|
"mean_token_accuracy": 0.16553854197263718,
|
|
"num_tokens": 7144727.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.501512956619263,
|
|
"epoch": 3.2662632375189107,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000433452604049351,
|
|
"loss": 5.1741,
|
|
"mean_token_accuracy": 0.17596300542354584,
|
|
"num_tokens": 7155531.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.562725067138672,
|
|
"epoch": 3.27130610186586,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00043317073792077394,
|
|
"loss": 5.2121,
|
|
"mean_token_accuracy": 0.1719446986913681,
|
|
"num_tokens": 7167234.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.676653480529785,
|
|
"epoch": 3.276348966212809,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004328883801938947,
|
|
"loss": 5.288,
|
|
"mean_token_accuracy": 0.165134260058403,
|
|
"num_tokens": 7178556.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.5326354026794435,
|
|
"epoch": 3.281391830559758,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043260553174628776,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.17600880414247513,
|
|
"num_tokens": 7189074.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.646118402481079,
|
|
"epoch": 3.286434694906707,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004323221934570524,
|
|
"loss": 5.2805,
|
|
"mean_token_accuracy": 0.17150572836399078,
|
|
"num_tokens": 7200567.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.665420341491699,
|
|
"epoch": 3.2914775592536563,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043203836620681074,
|
|
"loss": 5.3452,
|
|
"mean_token_accuracy": 0.16221626698970795,
|
|
"num_tokens": 7211012.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.580119180679321,
|
|
"epoch": 3.2965204236006054,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043175405087770435,
|
|
"loss": 5.2305,
|
|
"mean_token_accuracy": 0.16943196207284927,
|
|
"num_tokens": 7222271.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.627090692520142,
|
|
"epoch": 3.301563287947554,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00043146924835339186,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.17381926774978637,
|
|
"num_tokens": 7235067.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.587334442138672,
|
|
"epoch": 3.306606152294503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043118395951904605,
|
|
"loss": 5.2415,
|
|
"mean_token_accuracy": 0.1683749184012413,
|
|
"num_tokens": 7246159.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.741242361068726,
|
|
"epoch": 3.3116490166414523,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004308981852613512,
|
|
"loss": 5.2901,
|
|
"mean_token_accuracy": 0.1656254991889,
|
|
"num_tokens": 7255944.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.622401428222656,
|
|
"epoch": 3.3166918809884014,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004306119264685003,
|
|
"loss": 5.2015,
|
|
"mean_token_accuracy": 0.1710934042930603,
|
|
"num_tokens": 7267273.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.579175853729248,
|
|
"epoch": 3.3217347453353505,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004303251840301923,
|
|
"loss": 5.2176,
|
|
"mean_token_accuracy": 0.16749075651168824,
|
|
"num_tokens": 7277622.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.585457992553711,
|
|
"epoch": 3.3267776096822996,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00043003795883762933,
|
|
"loss": 5.2478,
|
|
"mean_token_accuracy": 0.16704524010419847,
|
|
"num_tokens": 7288851.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.696379375457764,
|
|
"epoch": 3.3318204740292487,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00042975025178351386,
|
|
"loss": 5.3028,
|
|
"mean_token_accuracy": 0.1655774787068367,
|
|
"num_tokens": 7302175.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.607939577102661,
|
|
"epoch": 3.336863338376198,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000429462063762046,
|
|
"loss": 5.1924,
|
|
"mean_token_accuracy": 0.173699052631855,
|
|
"num_tokens": 7312693.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.57006344795227,
|
|
"epoch": 3.3419062027231465,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042917339566892095,
|
|
"loss": 5.2148,
|
|
"mean_token_accuracy": 0.16248608380556107,
|
|
"num_tokens": 7323790.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.58256950378418,
|
|
"epoch": 3.3469490670700957,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004288842484013257,
|
|
"loss": 5.1972,
|
|
"mean_token_accuracy": 0.17196149975061417,
|
|
"num_tokens": 7333797.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.58911018371582,
|
|
"epoch": 3.3519919314170448,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004285946228579365,
|
|
"loss": 5.237,
|
|
"mean_token_accuracy": 0.17319759577512742,
|
|
"num_tokens": 7344674.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.62064037322998,
|
|
"epoch": 3.357034795763994,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004283045199389164,
|
|
"loss": 5.2529,
|
|
"mean_token_accuracy": 0.1656620755791664,
|
|
"num_tokens": 7355978.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.699236822128296,
|
|
"epoch": 3.362077660110943,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042801394054591197,
|
|
"loss": 5.3326,
|
|
"mean_token_accuracy": 0.1691778153181076,
|
|
"num_tokens": 7367054.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.6522870540618895,
|
|
"epoch": 3.367120524457892,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042772288558205057,
|
|
"loss": 5.3007,
|
|
"mean_token_accuracy": 0.16377128660678864,
|
|
"num_tokens": 7379588.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.566358804702759,
|
|
"epoch": 3.3721633888048412,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004274313559519377,
|
|
"loss": 5.2621,
|
|
"mean_token_accuracy": 0.1644855633378029,
|
|
"num_tokens": 7392411.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.590692329406738,
|
|
"epoch": 3.3772062531517903,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 5.1591,
|
|
"mean_token_accuracy": 0.17070212066173554,
|
|
"num_tokens": 7402801.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.593620634078979,
|
|
"epoch": 3.3822491174987395,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00042684687631875377,
|
|
"loss": 5.2185,
|
|
"mean_token_accuracy": 0.16406720578670503,
|
|
"num_tokens": 7413084.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.603054094314575,
|
|
"epoch": 3.3872919818456886,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042655392813225864,
|
|
"loss": 5.1696,
|
|
"mean_token_accuracy": 0.1752352610230446,
|
|
"num_tokens": 7424004.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.554675626754761,
|
|
"epoch": 3.3923348461926373,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042626050891265884,
|
|
"loss": 5.1964,
|
|
"mean_token_accuracy": 0.17738047987222672,
|
|
"num_tokens": 7434461.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.63235297203064,
|
|
"epoch": 3.3973777105395864,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004259666195719081,
|
|
"loss": 5.2572,
|
|
"mean_token_accuracy": 0.16517373472452163,
|
|
"num_tokens": 7445552.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.601845550537109,
|
|
"epoch": 3.4024205748865355,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042567226102342134,
|
|
"loss": 5.242,
|
|
"mean_token_accuracy": 0.16415649354457856,
|
|
"num_tokens": 7457073.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.614507341384888,
|
|
"epoch": 3.4074634392334846,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004253774341820715,
|
|
"loss": 5.2451,
|
|
"mean_token_accuracy": 0.16758612245321275,
|
|
"num_tokens": 7468197.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.632359313964844,
|
|
"epoch": 3.4125063035804337,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004250821399641874,
|
|
"loss": 5.2078,
|
|
"mean_token_accuracy": 0.17132595479488372,
|
|
"num_tokens": 7480751.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.5805529117584225,
|
|
"epoch": 3.417549167927383,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004247863792875502,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.16826128512620925,
|
|
"num_tokens": 7492259.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.620969533920288,
|
|
"epoch": 3.422592032274332,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042449015307139076,
|
|
"loss": 5.2937,
|
|
"mean_token_accuracy": 0.16408336013555527,
|
|
"num_tokens": 7504317.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.585500192642212,
|
|
"epoch": 3.427634896621281,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042419346223638725,
|
|
"loss": 5.1932,
|
|
"mean_token_accuracy": 0.16455380171537398,
|
|
"num_tokens": 7516039.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.64280481338501,
|
|
"epoch": 3.4326777609682297,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004238963077046616,
|
|
"loss": 5.2558,
|
|
"mean_token_accuracy": 0.16976338177919387,
|
|
"num_tokens": 7526095.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.553738594055176,
|
|
"epoch": 3.437720625315179,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004235986903997767,
|
|
"loss": 5.2085,
|
|
"mean_token_accuracy": 0.1659422919154167,
|
|
"num_tokens": 7537320.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.584099388122558,
|
|
"epoch": 3.442763489662128,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042330061124673437,
|
|
"loss": 5.2741,
|
|
"mean_token_accuracy": 0.1687419593334198,
|
|
"num_tokens": 7548100.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.688243293762207,
|
|
"epoch": 3.447806354009077,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042300207117197116,
|
|
"loss": 5.2879,
|
|
"mean_token_accuracy": 0.1642666921019554,
|
|
"num_tokens": 7560483.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.626182937622071,
|
|
"epoch": 3.452849218356026,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042270307110335657,
|
|
"loss": 5.1978,
|
|
"mean_token_accuracy": 0.17239670753479003,
|
|
"num_tokens": 7571256.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.607506322860718,
|
|
"epoch": 3.4578920827029753,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004224036119701899,
|
|
"loss": 5.1782,
|
|
"mean_token_accuracy": 0.17157293111085892,
|
|
"num_tokens": 7581716.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.608083629608155,
|
|
"epoch": 3.4629349470499244,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042210369470319694,
|
|
"loss": 5.2193,
|
|
"mean_token_accuracy": 0.17466395646333693,
|
|
"num_tokens": 7591932.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.588224744796753,
|
|
"epoch": 3.4679778113968736,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004218033202345274,
|
|
"loss": 5.2207,
|
|
"mean_token_accuracy": 0.16988266557455062,
|
|
"num_tokens": 7602842.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.543607568740844,
|
|
"epoch": 3.4730206757438227,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004215024894977522,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.1639377847313881,
|
|
"num_tokens": 7614075.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.6733174324035645,
|
|
"epoch": 3.478063540090772,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004212012034278601,
|
|
"loss": 5.3562,
|
|
"mean_token_accuracy": 0.16089720278978348,
|
|
"num_tokens": 7625642.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.639229249954224,
|
|
"epoch": 3.4831064044377205,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042089946296125513,
|
|
"loss": 5.2061,
|
|
"mean_token_accuracy": 0.16926859617233275,
|
|
"num_tokens": 7636447.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.522497224807739,
|
|
"epoch": 3.4881492687846696,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00042059726903575373,
|
|
"loss": 5.1439,
|
|
"mean_token_accuracy": 0.1721067100763321,
|
|
"num_tokens": 7648684.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.5845441818237305,
|
|
"epoch": 3.4931921331316187,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004202946225905815,
|
|
"loss": 5.2047,
|
|
"mean_token_accuracy": 0.16719142645597457,
|
|
"num_tokens": 7660864.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.572006368637085,
|
|
"epoch": 3.498234997478568,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004199915245663706,
|
|
"loss": 5.2261,
|
|
"mean_token_accuracy": 0.170198518037796,
|
|
"num_tokens": 7671939.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.735914707183838,
|
|
"epoch": 3.503277861825517,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041968797590515673,
|
|
"loss": 5.2621,
|
|
"mean_token_accuracy": 0.16947703808546066,
|
|
"num_tokens": 7681536.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.608390808105469,
|
|
"epoch": 3.508320726172466,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004193839775503761,
|
|
"loss": 5.1963,
|
|
"mean_token_accuracy": 0.1742884948849678,
|
|
"num_tokens": 7693674.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.554369401931763,
|
|
"epoch": 3.513363590519415,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00041907953044686253,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.1787819653749466,
|
|
"num_tokens": 7704533.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.629007482528687,
|
|
"epoch": 3.518406454866364,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004187746355408448,
|
|
"loss": 5.277,
|
|
"mean_token_accuracy": 0.1636540949344635,
|
|
"num_tokens": 7715448.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.600923776626587,
|
|
"epoch": 3.523449319213313,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004184692937799431,
|
|
"loss": 5.2469,
|
|
"mean_token_accuracy": 0.1666495382785797,
|
|
"num_tokens": 7726808.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.763300275802612,
|
|
"epoch": 3.528492183560262,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041816350611316693,
|
|
"loss": 5.2437,
|
|
"mean_token_accuracy": 0.16646722853183746,
|
|
"num_tokens": 7737899.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.528492183560262,
|
|
"eval_entropy": 5.484399833145336,
|
|
"eval_loss": 5.744884490966797,
|
|
"eval_mean_token_accuracy": 0.1476797835555392,
|
|
"eval_num_tokens": 7737899.0,
|
|
"eval_runtime": 2.0183,
|
|
"eval_samples_per_second": 1554.797,
|
|
"eval_steps_per_second": 194.721,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.574797105789185,
|
|
"epoch": 3.533535047907211,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00041785727349091117,
|
|
"loss": 5.1887,
|
|
"mean_token_accuracy": 0.16970745772123336,
|
|
"num_tokens": 7749268.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.608426189422607,
|
|
"epoch": 3.5385779122541603,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000417550596864954,
|
|
"loss": 5.2756,
|
|
"mean_token_accuracy": 0.1616047978401184,
|
|
"num_tokens": 7760312.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.63349666595459,
|
|
"epoch": 3.5436207766011094,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041724347718845344,
|
|
"loss": 5.27,
|
|
"mean_token_accuracy": 0.17193859815597534,
|
|
"num_tokens": 7771130.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.630754470825195,
|
|
"epoch": 3.5486636409480585,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041693591541594455,
|
|
"loss": 5.2243,
|
|
"mean_token_accuracy": 0.16470727920532227,
|
|
"num_tokens": 7783068.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.59250111579895,
|
|
"epoch": 3.5537065052950076,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004166279125033363,
|
|
"loss": 5.3309,
|
|
"mean_token_accuracy": 0.1666709691286087,
|
|
"num_tokens": 7794314.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.699529790878296,
|
|
"epoch": 3.5587493696419568,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00041631946940790893,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.16615971326828002,
|
|
"num_tokens": 7806888.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.566280126571655,
|
|
"epoch": 3.563792233988906,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041601058708831077,
|
|
"loss": 5.2634,
|
|
"mean_token_accuracy": 0.16525338739156722,
|
|
"num_tokens": 7818690.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.59206748008728,
|
|
"epoch": 3.568835098335855,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000415701266504555,
|
|
"loss": 5.1829,
|
|
"mean_token_accuracy": 0.1650310680270195,
|
|
"num_tokens": 7829321.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.58051176071167,
|
|
"epoch": 3.5738779626828037,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004153915086180173,
|
|
"loss": 5.2532,
|
|
"mean_token_accuracy": 0.17087870687246323,
|
|
"num_tokens": 7840223.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.60385046005249,
|
|
"epoch": 3.578920827029753,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004150813143914323,
|
|
"loss": 5.2415,
|
|
"mean_token_accuracy": 0.16197880804538728,
|
|
"num_tokens": 7850999.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.68939471244812,
|
|
"epoch": 3.583963691376702,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004147706847888908,
|
|
"loss": 5.2917,
|
|
"mean_token_accuracy": 0.16539207845926285,
|
|
"num_tokens": 7861935.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.550705480575561,
|
|
"epoch": 3.589006555723651,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004144596207758367,
|
|
"loss": 5.291,
|
|
"mean_token_accuracy": 0.16611184030771256,
|
|
"num_tokens": 7872960.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.618452167510986,
|
|
"epoch": 3.5940494200706,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004141481233190642,
|
|
"loss": 5.3225,
|
|
"mean_token_accuracy": 0.16165329217910768,
|
|
"num_tokens": 7883881.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.679779863357544,
|
|
"epoch": 3.5990922844175492,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004138361933867145,
|
|
"loss": 5.3049,
|
|
"mean_token_accuracy": 0.16277416199445724,
|
|
"num_tokens": 7895106.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.564934492111206,
|
|
"epoch": 3.6041351487644984,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004135238319482731,
|
|
"loss": 5.2007,
|
|
"mean_token_accuracy": 0.1703033149242401,
|
|
"num_tokens": 7906020.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.558734607696533,
|
|
"epoch": 3.609178013111447,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004132110399745666,
|
|
"loss": 5.2902,
|
|
"mean_token_accuracy": 0.16478729844093323,
|
|
"num_tokens": 7918959.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.683089971542358,
|
|
"epoch": 3.614220877458396,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004128978184377596,
|
|
"loss": 5.2487,
|
|
"mean_token_accuracy": 0.16195191591978073,
|
|
"num_tokens": 7929804.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.573867559432983,
|
|
"epoch": 3.6192637418053453,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004125841683113518,
|
|
"loss": 5.2478,
|
|
"mean_token_accuracy": 0.16878835409879683,
|
|
"num_tokens": 7941558.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.591611671447754,
|
|
"epoch": 3.6243066061522944,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041227009057017523,
|
|
"loss": 5.2641,
|
|
"mean_token_accuracy": 0.1688944950699806,
|
|
"num_tokens": 7951737.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.567723512649536,
|
|
"epoch": 3.6293494704992435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004119555861903906,
|
|
"loss": 5.1879,
|
|
"mean_token_accuracy": 0.17909695953130722,
|
|
"num_tokens": 7963164.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.61468653678894,
|
|
"epoch": 3.6343923348461926,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004116406561494847,
|
|
"loss": 5.262,
|
|
"mean_token_accuracy": 0.16693048924207687,
|
|
"num_tokens": 7973275.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.587840414047241,
|
|
"epoch": 3.6394351991931417,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041132530142626756,
|
|
"loss": 5.2172,
|
|
"mean_token_accuracy": 0.17658254951238633,
|
|
"num_tokens": 7984568.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.618630743026733,
|
|
"epoch": 3.644478063540091,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041100952300086884,
|
|
"loss": 5.2442,
|
|
"mean_token_accuracy": 0.1672349214553833,
|
|
"num_tokens": 7994263.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.62628402709961,
|
|
"epoch": 3.64952092788704,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041069332185473514,
|
|
"loss": 5.2333,
|
|
"mean_token_accuracy": 0.17175150960683822,
|
|
"num_tokens": 8004881.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.5773115158081055,
|
|
"epoch": 3.654563792233989,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000410376698970627,
|
|
"loss": 5.2232,
|
|
"mean_token_accuracy": 0.17307843416929244,
|
|
"num_tokens": 8015955.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.548713493347168,
|
|
"epoch": 3.659606656580938,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041005965533261554,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.1704103320837021,
|
|
"num_tokens": 8027010.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.557852745056152,
|
|
"epoch": 3.6646495209278873,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040974219192607973,
|
|
"loss": 5.1775,
|
|
"mean_token_accuracy": 0.1694122388958931,
|
|
"num_tokens": 8038564.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.59435658454895,
|
|
"epoch": 3.669692385274836,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00040942430973770323,
|
|
"loss": 5.2326,
|
|
"mean_token_accuracy": 0.17044240683317186,
|
|
"num_tokens": 8049975.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.636951017379761,
|
|
"epoch": 3.674735249621785,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004091060097554712,
|
|
"loss": 5.251,
|
|
"mean_token_accuracy": 0.1727294698357582,
|
|
"num_tokens": 8061042.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.571521377563476,
|
|
"epoch": 3.679778113968734,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004087872929686673,
|
|
"loss": 5.1912,
|
|
"mean_token_accuracy": 0.17105354517698287,
|
|
"num_tokens": 8071777.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.561955690383911,
|
|
"epoch": 3.6848209783156833,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00040846816036787056,
|
|
"loss": 5.2815,
|
|
"mean_token_accuracy": 0.16815441548824311,
|
|
"num_tokens": 8082387.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.65773344039917,
|
|
"epoch": 3.6898638426626325,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004081486129449527,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.16909808814525604,
|
|
"num_tokens": 8091818.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.635347509384156,
|
|
"epoch": 3.6949067070095816,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00040782865169307423,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.16841011196374894,
|
|
"num_tokens": 8103377.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.607952070236206,
|
|
"epoch": 3.6999495713565302,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004075082776066822,
|
|
"loss": 5.2984,
|
|
"mean_token_accuracy": 0.16228461563587188,
|
|
"num_tokens": 8114723.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.517363023757935,
|
|
"epoch": 3.7049924357034794,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004071874916815068,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.17213794142007827,
|
|
"num_tokens": 8124530.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.50477786064148,
|
|
"epoch": 3.7100353000504285,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040686629491455787,
|
|
"loss": 5.2305,
|
|
"mean_token_accuracy": 0.1722032830119133,
|
|
"num_tokens": 8134894.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.565054416656494,
|
|
"epoch": 3.7150781643973776,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040654468830412247,
|
|
"loss": 5.1796,
|
|
"mean_token_accuracy": 0.17571796625852584,
|
|
"num_tokens": 8143499.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.5682692527771,
|
|
"epoch": 3.7201210287443267,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00040622267284976134,
|
|
"loss": 5.2196,
|
|
"mean_token_accuracy": 0.1687147483229637,
|
|
"num_tokens": 8154556.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.564177989959717,
|
|
"epoch": 3.725163893091276,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040590024955230596,
|
|
"loss": 5.3217,
|
|
"mean_token_accuracy": 0.16327638328075408,
|
|
"num_tokens": 8166047.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.644197559356689,
|
|
"epoch": 3.730206757438225,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004055774194138554,
|
|
"loss": 5.2384,
|
|
"mean_token_accuracy": 0.16980254799127578,
|
|
"num_tokens": 8177068.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.584859323501587,
|
|
"epoch": 3.735249621785174,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004052541834377731,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.17090732455253602,
|
|
"num_tokens": 8187316.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.6213079452514645,
|
|
"epoch": 3.740292486132123,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00040493054262868397,
|
|
"loss": 5.3333,
|
|
"mean_token_accuracy": 0.1658324584364891,
|
|
"num_tokens": 8198642.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.5504120826721195,
|
|
"epoch": 3.7453353504790723,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040460649799247116,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.17417289614677428,
|
|
"num_tokens": 8208476.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.558039999008178,
|
|
"epoch": 3.7503782148260214,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004042820505362728,
|
|
"loss": 5.2684,
|
|
"mean_token_accuracy": 0.16562940031290055,
|
|
"num_tokens": 8219590.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.577512741088867,
|
|
"epoch": 3.7554210791729705,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004039572012684792,
|
|
"loss": 5.1885,
|
|
"mean_token_accuracy": 0.1750597521662712,
|
|
"num_tokens": 8230707.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.653000259399414,
|
|
"epoch": 3.760463943519919,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00040363195119872916,
|
|
"loss": 5.2328,
|
|
"mean_token_accuracy": 0.16398619413375853,
|
|
"num_tokens": 8241774.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.633572244644165,
|
|
"epoch": 3.7655068078668683,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00040330630133790757,
|
|
"loss": 5.3728,
|
|
"mean_token_accuracy": 0.16030244082212447,
|
|
"num_tokens": 8254314.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.589096784591675,
|
|
"epoch": 3.7705496722138174,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040298025269814165,
|
|
"loss": 5.2104,
|
|
"mean_token_accuracy": 0.17091599702835084,
|
|
"num_tokens": 8265650.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.567480754852295,
|
|
"epoch": 3.7755925365607665,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004026538062927981,
|
|
"loss": 5.1846,
|
|
"mean_token_accuracy": 0.17226023823022843,
|
|
"num_tokens": 8276808.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.521521854400635,
|
|
"epoch": 3.7806354009077157,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040232696313647985,
|
|
"loss": 5.1229,
|
|
"mean_token_accuracy": 0.171603961288929,
|
|
"num_tokens": 8287867.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.561845302581787,
|
|
"epoch": 3.7856782652546648,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 5.1857,
|
|
"mean_token_accuracy": 0.17129729986190795,
|
|
"num_tokens": 8298707.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.518676090240478,
|
|
"epoch": 3.7907211296016134,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004016720906354937,
|
|
"loss": 5.1446,
|
|
"mean_token_accuracy": 0.174617962539196,
|
|
"num_tokens": 8309363.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.6099536418914795,
|
|
"epoch": 3.7957639939485626,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004013440633261845,
|
|
"loss": 5.3125,
|
|
"mean_token_accuracy": 0.1652904987335205,
|
|
"num_tokens": 8319376.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.56193437576294,
|
|
"epoch": 3.8008068582955117,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00040101564333661213,
|
|
"loss": 5.2106,
|
|
"mean_token_accuracy": 0.1721050351858139,
|
|
"num_tokens": 8330842.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.550498008728027,
|
|
"epoch": 3.805849722642461,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040068683168751316,
|
|
"loss": 5.1951,
|
|
"mean_token_accuracy": 0.17202721238136293,
|
|
"num_tokens": 8341588.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.5382551670074465,
|
|
"epoch": 3.81089258698941,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00040035762940084205,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.16737455725669861,
|
|
"num_tokens": 8351664.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.61973466873169,
|
|
"epoch": 3.815935451336359,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000400028037499767,
|
|
"loss": 5.3063,
|
|
"mean_token_accuracy": 0.16244014501571655,
|
|
"num_tokens": 8362803.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.526738023757934,
|
|
"epoch": 3.820978315683308,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003996980570086673,
|
|
"loss": 5.2347,
|
|
"mean_token_accuracy": 0.1708271712064743,
|
|
"num_tokens": 8373791.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.604276943206787,
|
|
"epoch": 3.8260211800302573,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039936768895312987,
|
|
"loss": 5.2225,
|
|
"mean_token_accuracy": 0.17140780091285707,
|
|
"num_tokens": 8385738.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.556724786758423,
|
|
"epoch": 3.8310640443772064,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039903693435994626,
|
|
"loss": 5.2225,
|
|
"mean_token_accuracy": 0.16848905086517335,
|
|
"num_tokens": 8396924.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.590852117538452,
|
|
"epoch": 3.8361069087241555,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039870579425710937,
|
|
"loss": 5.2958,
|
|
"mean_token_accuracy": 0.1621718280017376,
|
|
"num_tokens": 8409040.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.473066186904907,
|
|
"epoch": 3.8411497730711046,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003983742696738103,
|
|
"loss": 5.1598,
|
|
"mean_token_accuracy": 0.1753212630748749,
|
|
"num_tokens": 8419630.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.615228700637817,
|
|
"epoch": 3.8461926374180537,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003980423616404351,
|
|
"loss": 5.289,
|
|
"mean_token_accuracy": 0.16702399402856827,
|
|
"num_tokens": 8430036.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.617081546783448,
|
|
"epoch": 3.8512355017650024,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003977100711885616,
|
|
"loss": 5.1678,
|
|
"mean_token_accuracy": 0.17435339838266373,
|
|
"num_tokens": 8439829.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.512828731536866,
|
|
"epoch": 3.8562783661119515,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039737739935095614,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.17132361531257628,
|
|
"num_tokens": 8450492.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.525384902954102,
|
|
"epoch": 3.8613212304589006,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0003970443471615705,
|
|
"loss": 5.2602,
|
|
"mean_token_accuracy": 0.16383477598428725,
|
|
"num_tokens": 8462200.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.570471000671387,
|
|
"epoch": 3.8663640948058497,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003967109156555386,
|
|
"loss": 5.2107,
|
|
"mean_token_accuracy": 0.17096201330423355,
|
|
"num_tokens": 8473393.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.515159368515015,
|
|
"epoch": 3.871406959152799,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039637710586917324,
|
|
"loss": 5.1883,
|
|
"mean_token_accuracy": 0.17095932066440583,
|
|
"num_tokens": 8484646.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.57587366104126,
|
|
"epoch": 3.876449823499748,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039604291883996287,
|
|
"loss": 5.2479,
|
|
"mean_token_accuracy": 0.17198000252246856,
|
|
"num_tokens": 8495371.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.520472478866577,
|
|
"epoch": 3.8814926878466967,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039570835560656854,
|
|
"loss": 5.242,
|
|
"mean_token_accuracy": 0.17214390486478806,
|
|
"num_tokens": 8507976.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.595182800292969,
|
|
"epoch": 3.8865355521936458,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003953734172088205,
|
|
"loss": 5.2429,
|
|
"mean_token_accuracy": 0.1732884928584099,
|
|
"num_tokens": 8519120.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.5920994758605955,
|
|
"epoch": 3.891578416540595,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039503810468771494,
|
|
"loss": 5.2319,
|
|
"mean_token_accuracy": 0.16978574693202972,
|
|
"num_tokens": 8529727.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.505200147628784,
|
|
"epoch": 3.896621280887544,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003947024190854111,
|
|
"loss": 5.1814,
|
|
"mean_token_accuracy": 0.1698161318898201,
|
|
"num_tokens": 8540202.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.60473222732544,
|
|
"epoch": 3.901664145234493,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039436636144522737,
|
|
"loss": 5.1893,
|
|
"mean_token_accuracy": 0.17414570748806,
|
|
"num_tokens": 8550810.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.519709539413452,
|
|
"epoch": 3.9067070095814422,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039402993281163875,
|
|
"loss": 5.2091,
|
|
"mean_token_accuracy": 0.17127603739500047,
|
|
"num_tokens": 8561120.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.557254457473755,
|
|
"epoch": 3.9117498739283914,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003936931342302732,
|
|
"loss": 5.2454,
|
|
"mean_token_accuracy": 0.16454725265502929,
|
|
"num_tokens": 8572153.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.566335964202881,
|
|
"epoch": 3.9167927382753405,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003933559667479084,
|
|
"loss": 5.2335,
|
|
"mean_token_accuracy": 0.1681883603334427,
|
|
"num_tokens": 8583740.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.578996562957764,
|
|
"epoch": 3.9218356026222896,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039301843141246876,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.17075487971305847,
|
|
"num_tokens": 8596017.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.582157611846924,
|
|
"epoch": 3.9268784669692387,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039268052927302185,
|
|
"loss": 5.2261,
|
|
"mean_token_accuracy": 0.16905294954776764,
|
|
"num_tokens": 8608121.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.559178972244263,
|
|
"epoch": 3.931921331316188,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003923422613797753,
|
|
"loss": 5.2648,
|
|
"mean_token_accuracy": 0.17171190828084945,
|
|
"num_tokens": 8621446.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.56966757774353,
|
|
"epoch": 3.936964195663137,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003920036287840734,
|
|
"loss": 5.2469,
|
|
"mean_token_accuracy": 0.17294939756393432,
|
|
"num_tokens": 8632088.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.553457164764405,
|
|
"epoch": 3.9420070600100856,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00039166463253839434,
|
|
"loss": 5.1226,
|
|
"mean_token_accuracy": 0.18189847618341445,
|
|
"num_tokens": 8642546.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.513269901275635,
|
|
"epoch": 3.9470499243570347,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003913252736963462,
|
|
"loss": 5.231,
|
|
"mean_token_accuracy": 0.1656072124838829,
|
|
"num_tokens": 8652869.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.6004723072052,
|
|
"epoch": 3.952092788703984,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000390985553312664,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.17172389924526216,
|
|
"num_tokens": 8662834.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.577038097381592,
|
|
"epoch": 3.957135653050933,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003906454724432065,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.16459181308746337,
|
|
"num_tokens": 8673372.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.54888801574707,
|
|
"epoch": 3.962178517397882,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039030503214495296,
|
|
"loss": 5.2375,
|
|
"mean_token_accuracy": 0.16985922455787658,
|
|
"num_tokens": 8683660.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.592769813537598,
|
|
"epoch": 3.967221381744831,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003899642334759996,
|
|
"loss": 5.2466,
|
|
"mean_token_accuracy": 0.16771800220012664,
|
|
"num_tokens": 8694044.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.573861980438233,
|
|
"epoch": 3.97226424609178,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003896230774955567,
|
|
"loss": 5.2698,
|
|
"mean_token_accuracy": 0.1620016112923622,
|
|
"num_tokens": 8705014.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.550318574905395,
|
|
"epoch": 3.977307110438729,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038928156526394476,
|
|
"loss": 5.1968,
|
|
"mean_token_accuracy": 0.1656872048974037,
|
|
"num_tokens": 8716007.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.584542179107666,
|
|
"epoch": 3.982349974785678,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003889396978425918,
|
|
"loss": 5.2298,
|
|
"mean_token_accuracy": 0.1759941816329956,
|
|
"num_tokens": 8728390.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.570604228973389,
|
|
"epoch": 3.987392839132627,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003885974762940295,
|
|
"loss": 5.187,
|
|
"mean_token_accuracy": 0.17041437178850175,
|
|
"num_tokens": 8738806.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.493031644821167,
|
|
"epoch": 3.9924357034795763,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0003882549016818905,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.16945813596248627,
|
|
"num_tokens": 8749667.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.545030832290649,
|
|
"epoch": 3.9974785678265254,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00038791197507090456,
|
|
"loss": 5.232,
|
|
"mean_token_accuracy": 0.1699605718255043,
|
|
"num_tokens": 8761380.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.535213788350423,
|
|
"epoch": 4.00201714573878,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038756869752689527,
|
|
"loss": 5.0957,
|
|
"mean_token_accuracy": 0.17898427777820164,
|
|
"num_tokens": 8770921.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.477807950973511,
|
|
"epoch": 4.007060010085729,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003872250701167775,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.18703764379024507,
|
|
"num_tokens": 8782706.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.5072112560272215,
|
|
"epoch": 4.012102874432678,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00038688109390855297,
|
|
"loss": 4.9909,
|
|
"mean_token_accuracy": 0.18445588946342467,
|
|
"num_tokens": 8794600.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.518926477432251,
|
|
"epoch": 4.017145738779627,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003865367699713077,
|
|
"loss": 4.8957,
|
|
"mean_token_accuracy": 0.18901741057634353,
|
|
"num_tokens": 8804149.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.536908149719238,
|
|
"epoch": 4.022188603126576,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00038619209937520836,
|
|
"loss": 4.9078,
|
|
"mean_token_accuracy": 0.1938214957714081,
|
|
"num_tokens": 8814723.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.520428085327149,
|
|
"epoch": 4.027231467473525,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00038584708319149924,
|
|
"loss": 4.9978,
|
|
"mean_token_accuracy": 0.18627551048994065,
|
|
"num_tokens": 8827258.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.513300991058349,
|
|
"epoch": 4.032274331820474,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038550172249249856,
|
|
"loss": 4.9602,
|
|
"mean_token_accuracy": 0.18729876279830932,
|
|
"num_tokens": 8837528.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.032274331820474,
|
|
"eval_entropy": 5.355193902518003,
|
|
"eval_loss": 5.706085205078125,
|
|
"eval_mean_token_accuracy": 0.1511400248713166,
|
|
"eval_num_tokens": 8837528.0,
|
|
"eval_runtime": 2.0593,
|
|
"eval_samples_per_second": 1523.83,
|
|
"eval_steps_per_second": 190.843,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.518103790283203,
|
|
"epoch": 4.0373171961674235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003851560183515953,
|
|
"loss": 4.9449,
|
|
"mean_token_accuracy": 0.18659648448228836,
|
|
"num_tokens": 8848154.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.5106823444366455,
|
|
"epoch": 4.042360060514373,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000384809971843246,
|
|
"loss": 4.9316,
|
|
"mean_token_accuracy": 0.1924380913376808,
|
|
"num_tokens": 8858396.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.565646982192993,
|
|
"epoch": 4.047402924861321,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000384463584042971,
|
|
"loss": 4.9404,
|
|
"mean_token_accuracy": 0.19597283750772476,
|
|
"num_tokens": 8868590.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.4950035572052,
|
|
"epoch": 4.05244578920827,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038411685602735175,
|
|
"loss": 4.9456,
|
|
"mean_token_accuracy": 0.18894167393445968,
|
|
"num_tokens": 8878955.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.455377054214478,
|
|
"epoch": 4.057488653555219,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038376978887402687,
|
|
"loss": 4.9203,
|
|
"mean_token_accuracy": 0.19065040946006775,
|
|
"num_tokens": 8889402.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.509929084777832,
|
|
"epoch": 4.062531517902168,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003834223836616891,
|
|
"loss": 4.9379,
|
|
"mean_token_accuracy": 0.185760560631752,
|
|
"num_tokens": 8901433.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.416840982437134,
|
|
"epoch": 4.067574382249117,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038307464147008176,
|
|
"loss": 4.9687,
|
|
"mean_token_accuracy": 0.1869459792971611,
|
|
"num_tokens": 8912895.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.514374399185181,
|
|
"epoch": 4.072617246596066,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003827265633799957,
|
|
"loss": 4.978,
|
|
"mean_token_accuracy": 0.1872596487402916,
|
|
"num_tokens": 8923529.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.545920038223267,
|
|
"epoch": 4.0776601109430155,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003823781504732657,
|
|
"loss": 4.8868,
|
|
"mean_token_accuracy": 0.1910176992416382,
|
|
"num_tokens": 8933279.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.529631757736206,
|
|
"epoch": 4.082702975289965,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003820294038327669,
|
|
"loss": 4.996,
|
|
"mean_token_accuracy": 0.18962214440107344,
|
|
"num_tokens": 8944396.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.415576601028443,
|
|
"epoch": 4.087745839636914,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003816803245424121,
|
|
"loss": 4.7918,
|
|
"mean_token_accuracy": 0.1988392174243927,
|
|
"num_tokens": 8955535.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.499550914764404,
|
|
"epoch": 4.092788703983863,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003813309136871478,
|
|
"loss": 4.9721,
|
|
"mean_token_accuracy": 0.1848311498761177,
|
|
"num_tokens": 8965859.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.510080146789551,
|
|
"epoch": 4.097831568330812,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038098117235295083,
|
|
"loss": 5.0124,
|
|
"mean_token_accuracy": 0.18439542949199678,
|
|
"num_tokens": 8976580.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.468481492996216,
|
|
"epoch": 4.102874432677761,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003806311016268254,
|
|
"loss": 4.958,
|
|
"mean_token_accuracy": 0.18880272954702376,
|
|
"num_tokens": 8988154.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.505822849273682,
|
|
"epoch": 4.10791729702471,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003802807025967995,
|
|
"loss": 4.9789,
|
|
"mean_token_accuracy": 0.1805400624871254,
|
|
"num_tokens": 8999758.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.458872604370117,
|
|
"epoch": 4.112960161371659,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003799299763519213,
|
|
"loss": 4.9481,
|
|
"mean_token_accuracy": 0.18584952205419542,
|
|
"num_tokens": 9010371.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.466066455841064,
|
|
"epoch": 4.118003025718608,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003795789239822561,
|
|
"loss": 4.9325,
|
|
"mean_token_accuracy": 0.19052418023347856,
|
|
"num_tokens": 9021224.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.477115106582642,
|
|
"epoch": 4.1230458900655576,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003792275465788826,
|
|
"loss": 4.9919,
|
|
"mean_token_accuracy": 0.1877814292907715,
|
|
"num_tokens": 9033569.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.5263927459716795,
|
|
"epoch": 4.128088754412507,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037887584523389013,
|
|
"loss": 5.0065,
|
|
"mean_token_accuracy": 0.18378187865018844,
|
|
"num_tokens": 9046166.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.512891054153442,
|
|
"epoch": 4.133131618759456,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00037852382104037454,
|
|
"loss": 5.0062,
|
|
"mean_token_accuracy": 0.190128456056118,
|
|
"num_tokens": 9056210.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.471432065963745,
|
|
"epoch": 4.138174483106404,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00037817147509243497,
|
|
"loss": 4.9562,
|
|
"mean_token_accuracy": 0.18378653079271318,
|
|
"num_tokens": 9067473.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.553931856155396,
|
|
"epoch": 4.143217347453353,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003778188084851709,
|
|
"loss": 5.0062,
|
|
"mean_token_accuracy": 0.1818464905023575,
|
|
"num_tokens": 9077744.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.5054083347320555,
|
|
"epoch": 4.148260211800302,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037746582231467816,
|
|
"loss": 4.9952,
|
|
"mean_token_accuracy": 0.18302546441555023,
|
|
"num_tokens": 9088988.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.54720549583435,
|
|
"epoch": 4.153303076147251,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003771125176780461,
|
|
"loss": 5.0187,
|
|
"mean_token_accuracy": 0.1811775103211403,
|
|
"num_tokens": 9100584.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.4618175506591795,
|
|
"epoch": 4.1583459404942005,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00037675889567335345,
|
|
"loss": 4.9863,
|
|
"mean_token_accuracy": 0.18026272505521773,
|
|
"num_tokens": 9112442.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.458131599426269,
|
|
"epoch": 4.16338880484115,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037640495739966573,
|
|
"loss": 5.0371,
|
|
"mean_token_accuracy": 0.17680447697639465,
|
|
"num_tokens": 9122903.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.587776231765747,
|
|
"epoch": 4.168431669188099,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037605070395703124,
|
|
"loss": 5.0522,
|
|
"mean_token_accuracy": 0.18395502120256424,
|
|
"num_tokens": 9134301.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.631263208389282,
|
|
"epoch": 4.173474533535048,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037569613644647783,
|
|
"loss": 5.0457,
|
|
"mean_token_accuracy": 0.18186109960079194,
|
|
"num_tokens": 9144503.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.418415975570679,
|
|
"epoch": 4.178517397881997,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003753412559700094,
|
|
"loss": 4.8677,
|
|
"mean_token_accuracy": 0.19117171466350555,
|
|
"num_tokens": 9154173.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.524201202392578,
|
|
"epoch": 4.183560262228946,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037498606363060285,
|
|
"loss": 5.0248,
|
|
"mean_token_accuracy": 0.18463052213191986,
|
|
"num_tokens": 9163772.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.527320814132691,
|
|
"epoch": 4.188603126575895,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003746305605322041,
|
|
"loss": 5.0272,
|
|
"mean_token_accuracy": 0.1839022859930992,
|
|
"num_tokens": 9174618.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.614655780792236,
|
|
"epoch": 4.193645990922844,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 5.0878,
|
|
"mean_token_accuracy": 0.1827521562576294,
|
|
"num_tokens": 9185199.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.519039011001587,
|
|
"epoch": 4.198688855269793,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00037391862647903943,
|
|
"loss": 4.9852,
|
|
"mean_token_accuracy": 0.19007405042648315,
|
|
"num_tokens": 9197329.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.422666215896607,
|
|
"epoch": 4.2037317196167425,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00037356219773698106,
|
|
"loss": 4.9716,
|
|
"mean_token_accuracy": 0.187738536298275,
|
|
"num_tokens": 9207516.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.488312101364135,
|
|
"epoch": 4.208774583963692,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00037320546266133854,
|
|
"loss": 5.0216,
|
|
"mean_token_accuracy": 0.17582017183303833,
|
|
"num_tokens": 9219598.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.461294841766358,
|
|
"epoch": 4.213817448310641,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003728484223608526,
|
|
"loss": 4.9196,
|
|
"mean_token_accuracy": 0.18632453233003615,
|
|
"num_tokens": 9229675.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.379378461837769,
|
|
"epoch": 4.21886031265759,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037249107794521286,
|
|
"loss": 4.9681,
|
|
"mean_token_accuracy": 0.1856491819024086,
|
|
"num_tokens": 9240308.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.440922212600708,
|
|
"epoch": 4.223903177004539,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037213343052505394,
|
|
"loss": 4.9516,
|
|
"mean_token_accuracy": 0.18670842200517654,
|
|
"num_tokens": 9250299.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.514345741271972,
|
|
"epoch": 4.228946041351487,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003717754812119523,
|
|
"loss": 5.0002,
|
|
"mean_token_accuracy": 0.1842057466506958,
|
|
"num_tokens": 9262659.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.516497659683227,
|
|
"epoch": 4.233988905698436,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003714172311184226,
|
|
"loss": 4.9535,
|
|
"mean_token_accuracy": 0.18644220381975174,
|
|
"num_tokens": 9274430.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.468480825424194,
|
|
"epoch": 4.2390317700453854,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003710586813579144,
|
|
"loss": 5.0481,
|
|
"mean_token_accuracy": 0.18256663233041764,
|
|
"num_tokens": 9285484.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.495090579986572,
|
|
"epoch": 4.244074634392335,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037069983304480875,
|
|
"loss": 4.9589,
|
|
"mean_token_accuracy": 0.18800730407238006,
|
|
"num_tokens": 9296253.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.495601987838745,
|
|
"epoch": 4.249117498739284,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003703406872944144,
|
|
"loss": 4.9854,
|
|
"mean_token_accuracy": 0.18881929963827132,
|
|
"num_tokens": 9308491.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.547918033599854,
|
|
"epoch": 4.254160363086233,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003699812452229645,
|
|
"loss": 4.9531,
|
|
"mean_token_accuracy": 0.18905131965875627,
|
|
"num_tokens": 9319722.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.439584493637085,
|
|
"epoch": 4.259203227433182,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00036962150794761354,
|
|
"loss": 4.8958,
|
|
"mean_token_accuracy": 0.1938430666923523,
|
|
"num_tokens": 9329324.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.429581165313721,
|
|
"epoch": 4.264246091780131,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003692614765864331,
|
|
"loss": 5.0048,
|
|
"mean_token_accuracy": 0.18257844895124437,
|
|
"num_tokens": 9340060.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.48242678642273,
|
|
"epoch": 4.26928895612708,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00036890115225840904,
|
|
"loss": 5.0127,
|
|
"mean_token_accuracy": 0.181241013109684,
|
|
"num_tokens": 9351071.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.568946695327758,
|
|
"epoch": 4.274331820474029,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003685405360834376,
|
|
"loss": 5.05,
|
|
"mean_token_accuracy": 0.17752183526754378,
|
|
"num_tokens": 9361910.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.508288717269897,
|
|
"epoch": 4.279374684820978,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036817962918232234,
|
|
"loss": 4.9868,
|
|
"mean_token_accuracy": 0.18908119350671768,
|
|
"num_tokens": 9371964.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.489854192733764,
|
|
"epoch": 4.2844175491679275,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00036781843267677,
|
|
"loss": 4.9553,
|
|
"mean_token_accuracy": 0.19056928306818008,
|
|
"num_tokens": 9381922.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.452848720550537,
|
|
"epoch": 4.289460413514877,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003674569476893878,
|
|
"loss": 4.9158,
|
|
"mean_token_accuracy": 0.1938249558210373,
|
|
"num_tokens": 9393256.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.425813388824463,
|
|
"epoch": 4.294503277861826,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003670951753436794,
|
|
"loss": 4.9552,
|
|
"mean_token_accuracy": 0.1870001882314682,
|
|
"num_tokens": 9403703.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.53969521522522,
|
|
"epoch": 4.299546142208775,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00036673311676404163,
|
|
"loss": 5.0322,
|
|
"mean_token_accuracy": 0.18738039582967758,
|
|
"num_tokens": 9414454.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.410726642608642,
|
|
"epoch": 4.304589006555724,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003663707730757608,
|
|
"loss": 4.9377,
|
|
"mean_token_accuracy": 0.19340676814317703,
|
|
"num_tokens": 9425566.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.4349579334259035,
|
|
"epoch": 4.309631870902673,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036600814540500956,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.17907942831516266,
|
|
"num_tokens": 9435555.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.585208606719971,
|
|
"epoch": 4.314674735249622,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000365645234878843,
|
|
"loss": 5.0811,
|
|
"mean_token_accuracy": 0.17448912411928177,
|
|
"num_tokens": 9447784.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.525291872024536,
|
|
"epoch": 4.31971759959657,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036528204262519545,
|
|
"loss": 4.9784,
|
|
"mean_token_accuracy": 0.18703167587518693,
|
|
"num_tokens": 9458243.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.48011474609375,
|
|
"epoch": 4.3247604639435195,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003649185697728768,
|
|
"loss": 4.979,
|
|
"mean_token_accuracy": 0.18327850699424744,
|
|
"num_tokens": 9470258.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.509270238876343,
|
|
"epoch": 4.329803328290469,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036455481745156906,
|
|
"loss": 4.9961,
|
|
"mean_token_accuracy": 0.1858487159013748,
|
|
"num_tokens": 9480787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.45818305015564,
|
|
"epoch": 4.334846192637418,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003641907867918228,
|
|
"loss": 4.9637,
|
|
"mean_token_accuracy": 0.18921741098165512,
|
|
"num_tokens": 9492459.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.5414228439331055,
|
|
"epoch": 4.339889056984367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003638264789250538,
|
|
"loss": 5.0678,
|
|
"mean_token_accuracy": 0.17808520197868347,
|
|
"num_tokens": 9504600.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.559147310256958,
|
|
"epoch": 4.344931921331316,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036346189498353914,
|
|
"loss": 5.0682,
|
|
"mean_token_accuracy": 0.18137387484312056,
|
|
"num_tokens": 9515853.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.463085317611695,
|
|
"epoch": 4.349974785678265,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003630970361004143,
|
|
"loss": 5.0339,
|
|
"mean_token_accuracy": 0.18236847966909409,
|
|
"num_tokens": 9528109.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.499902009963989,
|
|
"epoch": 4.355017650025214,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003627319034096691,
|
|
"loss": 4.9943,
|
|
"mean_token_accuracy": 0.18708615750074387,
|
|
"num_tokens": 9539795.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.4575550079345705,
|
|
"epoch": 4.360060514372163,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003623664980461443,
|
|
"loss": 5.0165,
|
|
"mean_token_accuracy": 0.1849098801612854,
|
|
"num_tokens": 9551020.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.514564275741577,
|
|
"epoch": 4.3651033787191125,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036200082114552806,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.18440216928720474,
|
|
"num_tokens": 9562540.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.549724483489991,
|
|
"epoch": 4.370146243066062,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00036163487384435294,
|
|
"loss": 5.0333,
|
|
"mean_token_accuracy": 0.18049701154232026,
|
|
"num_tokens": 9574122.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.5271683692932125,
|
|
"epoch": 4.375189107413011,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003612686572799914,
|
|
"loss": 5.0173,
|
|
"mean_token_accuracy": 0.1872974917292595,
|
|
"num_tokens": 9584781.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.560951232910156,
|
|
"epoch": 4.38023197175996,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003609021725906529,
|
|
"loss": 5.0268,
|
|
"mean_token_accuracy": 0.18644319027662276,
|
|
"num_tokens": 9595230.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.479853296279908,
|
|
"epoch": 4.385274836106909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036053542091538045,
|
|
"loss": 4.9737,
|
|
"mean_token_accuracy": 0.1884053334593773,
|
|
"num_tokens": 9605814.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.442518711090088,
|
|
"epoch": 4.390317700453858,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00036016840339404664,
|
|
"loss": 4.9625,
|
|
"mean_token_accuracy": 0.18475103229284287,
|
|
"num_tokens": 9617900.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.482942056655884,
|
|
"epoch": 4.395360564800807,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003598011211673505,
|
|
"loss": 5.0144,
|
|
"mean_token_accuracy": 0.18438477367162703,
|
|
"num_tokens": 9628987.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.467824029922485,
|
|
"epoch": 4.400403429147756,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003594335753768135,
|
|
"loss": 4.9739,
|
|
"mean_token_accuracy": 0.18842060416936873,
|
|
"num_tokens": 9638419.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.421893644332886,
|
|
"epoch": 4.405446293494705,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003590657671647766,
|
|
"loss": 4.9945,
|
|
"mean_token_accuracy": 0.18281660676002504,
|
|
"num_tokens": 9648975.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.482951354980469,
|
|
"epoch": 4.4104891578416545,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035869769767439606,
|
|
"loss": 4.9572,
|
|
"mean_token_accuracy": 0.191358682513237,
|
|
"num_tokens": 9659935.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.550790596008301,
|
|
"epoch": 4.415532022188603,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003583293680496404,
|
|
"loss": 5.0787,
|
|
"mean_token_accuracy": 0.17775934487581252,
|
|
"num_tokens": 9671269.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.435133171081543,
|
|
"epoch": 4.420574886535552,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003579607794352867,
|
|
"loss": 4.9875,
|
|
"mean_token_accuracy": 0.1856317326426506,
|
|
"num_tokens": 9682627.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.402055406570435,
|
|
"epoch": 4.425617750882501,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035759193297691674,
|
|
"loss": 4.9155,
|
|
"mean_token_accuracy": 0.18509576171636583,
|
|
"num_tokens": 9693482.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.481759977340698,
|
|
"epoch": 4.43066061522945,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00035722282982091373,
|
|
"loss": 5.0274,
|
|
"mean_token_accuracy": 0.1875482365489006,
|
|
"num_tokens": 9705279.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.487551307678222,
|
|
"epoch": 4.435703479576399,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003568534711144591,
|
|
"loss": 5.0211,
|
|
"mean_token_accuracy": 0.18040550649166107,
|
|
"num_tokens": 9716695.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.430243253707886,
|
|
"epoch": 4.440746343923348,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000356483858005528,
|
|
"loss": 4.9491,
|
|
"mean_token_accuracy": 0.19337646961212157,
|
|
"num_tokens": 9727960.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.459237718582154,
|
|
"epoch": 4.445789208270297,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00035611399164288654,
|
|
"loss": 5.0253,
|
|
"mean_token_accuracy": 0.17951168417930602,
|
|
"num_tokens": 9738341.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.503083372116089,
|
|
"epoch": 4.450832072617247,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003557438731760879,
|
|
"loss": 5.0255,
|
|
"mean_token_accuracy": 0.17970775067806244,
|
|
"num_tokens": 9748261.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.5371991157531735,
|
|
"epoch": 4.455874936964196,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003553735037554689,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.18527282029390335,
|
|
"num_tokens": 9759109.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.48226227760315,
|
|
"epoch": 4.460917801311145,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003550028845321461,
|
|
"loss": 4.9765,
|
|
"mean_token_accuracy": 0.18806328624486923,
|
|
"num_tokens": 9770431.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.431528949737549,
|
|
"epoch": 4.465960665658094,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003546320166580127,
|
|
"loss": 5.0287,
|
|
"mean_token_accuracy": 0.175430928170681,
|
|
"num_tokens": 9783459.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.499318790435791,
|
|
"epoch": 4.471003530005043,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035426090128573464,
|
|
"loss": 4.9561,
|
|
"mean_token_accuracy": 0.18869741559028624,
|
|
"num_tokens": 9793961.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.528396844863892,
|
|
"epoch": 4.476046394351992,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035388953956874686,
|
|
"loss": 5.0411,
|
|
"mean_token_accuracy": 0.17767144590616227,
|
|
"num_tokens": 9803964.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.458376216888428,
|
|
"epoch": 4.481089258698941,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003535179326612504,
|
|
"loss": 4.9536,
|
|
"mean_token_accuracy": 0.18931579291820527,
|
|
"num_tokens": 9816176.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.402630996704102,
|
|
"epoch": 4.48613212304589,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035314608171820787,
|
|
"loss": 4.9223,
|
|
"mean_token_accuracy": 0.19561639428138733,
|
|
"num_tokens": 9826649.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.509682273864746,
|
|
"epoch": 4.4911749873928395,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00035277398789534075,
|
|
"loss": 5.0828,
|
|
"mean_token_accuracy": 0.17876080870628358,
|
|
"num_tokens": 9838396.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.528962326049805,
|
|
"epoch": 4.496217851739789,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003524016523491252,
|
|
"loss": 5.0222,
|
|
"mean_token_accuracy": 0.18479302227497102,
|
|
"num_tokens": 9848863.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.382427024841308,
|
|
"epoch": 4.501260716086737,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00035202907623678865,
|
|
"loss": 4.9361,
|
|
"mean_token_accuracy": 0.1845976322889328,
|
|
"num_tokens": 9860673.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.473905801773071,
|
|
"epoch": 4.506303580433686,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003516562607163062,
|
|
"loss": 4.9988,
|
|
"mean_token_accuracy": 0.18442064970731736,
|
|
"num_tokens": 9871246.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.484776496887207,
|
|
"epoch": 4.511346444780635,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003512832069463971,
|
|
"loss": 5.0619,
|
|
"mean_token_accuracy": 0.1783318355679512,
|
|
"num_tokens": 9883352.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.48721399307251,
|
|
"epoch": 4.516389309127584,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035090991608652125,
|
|
"loss": 5.0588,
|
|
"mean_token_accuracy": 0.18417815566062928,
|
|
"num_tokens": 9894797.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.483404302597046,
|
|
"epoch": 4.521432173474533,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00035053638929687504,
|
|
"loss": 4.9957,
|
|
"mean_token_accuracy": 0.1791810855269432,
|
|
"num_tokens": 9905618.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.4427581310272215,
|
|
"epoch": 4.526475037821482,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003501626277383884,
|
|
"loss": 4.9951,
|
|
"mean_token_accuracy": 0.1821364253759384,
|
|
"num_tokens": 9916817.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.479065608978272,
|
|
"epoch": 4.5315179021684315,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000349788632572721,
|
|
"loss": 4.9795,
|
|
"mean_token_accuracy": 0.18556778728961945,
|
|
"num_tokens": 9929052.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.427140951156616,
|
|
"epoch": 4.536560766515381,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003494144049622582,
|
|
"loss": 4.8985,
|
|
"mean_token_accuracy": 0.1868479385972023,
|
|
"num_tokens": 9939513.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.536560766515381,
|
|
"eval_entropy": 5.31253418303628,
|
|
"eval_loss": 5.68939733505249,
|
|
"eval_mean_token_accuracy": 0.15258948582653478,
|
|
"eval_num_tokens": 9939513.0,
|
|
"eval_runtime": 2.2623,
|
|
"eval_samples_per_second": 1387.111,
|
|
"eval_steps_per_second": 173.72,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.377797412872314,
|
|
"epoch": 4.54160363086233,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003490399460701083,
|
|
"loss": 4.9599,
|
|
"mean_token_accuracy": 0.18414584249258042,
|
|
"num_tokens": 9950149.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.548014974594116,
|
|
"epoch": 4.546646495209279,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000348665257060098,
|
|
"loss": 5.0159,
|
|
"mean_token_accuracy": 0.18414955139160155,
|
|
"num_tokens": 9961696.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.50719895362854,
|
|
"epoch": 4.551689359556228,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003482903390967695,
|
|
"loss": 5.0301,
|
|
"mean_token_accuracy": 0.1839794009923935,
|
|
"num_tokens": 9971446.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.493980073928833,
|
|
"epoch": 4.556732223903177,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034791519334537635,
|
|
"loss": 4.9593,
|
|
"mean_token_accuracy": 0.19213175475597383,
|
|
"num_tokens": 9982992.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.410910034179688,
|
|
"epoch": 4.561775088250126,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003475398209718804,
|
|
"loss": 4.9151,
|
|
"mean_token_accuracy": 0.18857127726078032,
|
|
"num_tokens": 9993443.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.436291933059692,
|
|
"epoch": 4.566817952597075,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003471642231429475,
|
|
"loss": 5.0268,
|
|
"mean_token_accuracy": 0.18810340911149978,
|
|
"num_tokens": 10004531.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.496992778778076,
|
|
"epoch": 4.5718608169440245,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003467884010259443,
|
|
"loss": 4.9803,
|
|
"mean_token_accuracy": 0.185042504966259,
|
|
"num_tokens": 10016455.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.49594407081604,
|
|
"epoch": 4.576903681290974,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00034641235578893476,
|
|
"loss": 4.9148,
|
|
"mean_token_accuracy": 0.1935173898935318,
|
|
"num_tokens": 10028067.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.495612621307373,
|
|
"epoch": 4.581946545637923,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003460360886006761,
|
|
"loss": 5.0887,
|
|
"mean_token_accuracy": 0.18369213938713075,
|
|
"num_tokens": 10038615.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.502183532714843,
|
|
"epoch": 4.586989409984872,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034565960063061534,
|
|
"loss": 5.0289,
|
|
"mean_token_accuracy": 0.18505603224039077,
|
|
"num_tokens": 10050022.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.430886268615723,
|
|
"epoch": 4.592032274331821,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003452828930488859,
|
|
"loss": 4.971,
|
|
"mean_token_accuracy": 0.18751503229141236,
|
|
"num_tokens": 10062565.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.570018196105957,
|
|
"epoch": 4.597075138678769,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034490596702630355,
|
|
"loss": 5.1199,
|
|
"mean_token_accuracy": 0.17767640799283982,
|
|
"num_tokens": 10075553.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.49906439781189,
|
|
"epoch": 4.602118003025718,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 5.0251,
|
|
"mean_token_accuracy": 0.18141546547412873,
|
|
"num_tokens": 10086144.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.399090337753296,
|
|
"epoch": 4.607160867372667,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00034415146434523463,
|
|
"loss": 5.0148,
|
|
"mean_token_accuracy": 0.1811163231730461,
|
|
"num_tokens": 10097218.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.431238460540771,
|
|
"epoch": 4.6122037317196165,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034377389003175984,
|
|
"loss": 5.0081,
|
|
"mean_token_accuracy": 0.1780278652906418,
|
|
"num_tokens": 10107614.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.582207059860229,
|
|
"epoch": 4.617246596066566,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003433961019674483,
|
|
"loss": 5.0547,
|
|
"mean_token_accuracy": 0.17949339598417283,
|
|
"num_tokens": 10118976.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.514779329299927,
|
|
"epoch": 4.622289460413515,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003430181013264741,
|
|
"loss": 5.0,
|
|
"mean_token_accuracy": 0.18489600718021393,
|
|
"num_tokens": 10129914.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.500811767578125,
|
|
"epoch": 4.627332324760464,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003426398892836719,
|
|
"loss": 5.0131,
|
|
"mean_token_accuracy": 0.18103212714195252,
|
|
"num_tokens": 10139490.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.519602060317993,
|
|
"epoch": 4.632375189107413,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003422614670145336,
|
|
"loss": 5.0575,
|
|
"mean_token_accuracy": 0.1879510059952736,
|
|
"num_tokens": 10149874.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.475231695175171,
|
|
"epoch": 4.637418053454362,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003418828356952041,
|
|
"loss": 5.0746,
|
|
"mean_token_accuracy": 0.18149272501468658,
|
|
"num_tokens": 10161373.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.451475286483765,
|
|
"epoch": 4.642460917801311,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034150399650247846,
|
|
"loss": 4.9877,
|
|
"mean_token_accuracy": 0.18731293976306915,
|
|
"num_tokens": 10172044.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.47752013206482,
|
|
"epoch": 4.64750378214826,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00034112495061379764,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.19030793458223344,
|
|
"num_tokens": 10183127.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.519724178314209,
|
|
"epoch": 4.652546646495209,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00034074569920724483,
|
|
"loss": 5.047,
|
|
"mean_token_accuracy": 0.18437152802944184,
|
|
"num_tokens": 10194366.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.488125038146973,
|
|
"epoch": 4.6575895108421586,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003403662434615423,
|
|
"loss": 5.0245,
|
|
"mean_token_accuracy": 0.18091187626123428,
|
|
"num_tokens": 10205655.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.4745954990386965,
|
|
"epoch": 4.662632375189108,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033998658455604736,
|
|
"loss": 5.0059,
|
|
"mean_token_accuracy": 0.19196947664022446,
|
|
"num_tokens": 10215596.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.422599220275879,
|
|
"epoch": 4.667675239536057,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033960672367074827,
|
|
"loss": 5.002,
|
|
"mean_token_accuracy": 0.18458994925022126,
|
|
"num_tokens": 10225298.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.601092147827148,
|
|
"epoch": 4.672718103883006,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003392266619862619,
|
|
"loss": 5.11,
|
|
"mean_token_accuracy": 0.18092309683561325,
|
|
"num_tokens": 10236979.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.472450923919678,
|
|
"epoch": 4.677760968229954,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003388464006838284,
|
|
"loss": 4.9482,
|
|
"mean_token_accuracy": 0.18817544877529144,
|
|
"num_tokens": 10248109.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.453419160842896,
|
|
"epoch": 4.682803832576903,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033846594094530874,
|
|
"loss": 5.0483,
|
|
"mean_token_accuracy": 0.1877111718058586,
|
|
"num_tokens": 10258931.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.479018115997315,
|
|
"epoch": 4.687846696923852,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003380852839531806,
|
|
"loss": 5.0723,
|
|
"mean_token_accuracy": 0.18389774858951569,
|
|
"num_tokens": 10272521.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.472506809234619,
|
|
"epoch": 4.6928895612708015,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033770443089053476,
|
|
"loss": 4.97,
|
|
"mean_token_accuracy": 0.18354106396436692,
|
|
"num_tokens": 10283481.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.4647801399230955,
|
|
"epoch": 4.697932425617751,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033732338294107116,
|
|
"loss": 5.0381,
|
|
"mean_token_accuracy": 0.18187450617551804,
|
|
"num_tokens": 10294258.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.520396423339844,
|
|
"epoch": 4.7029752899647,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033694214128909567,
|
|
"loss": 5.0184,
|
|
"mean_token_accuracy": 0.19066052585840226,
|
|
"num_tokens": 10304464.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.495455598831176,
|
|
"epoch": 4.708018154311649,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033656070711951633,
|
|
"loss": 5.0206,
|
|
"mean_token_accuracy": 0.17870270162820817,
|
|
"num_tokens": 10316343.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.504643392562866,
|
|
"epoch": 4.713061018658598,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003361790816178391,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.1840691387653351,
|
|
"num_tokens": 10327057.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.535696125030517,
|
|
"epoch": 4.718103883005547,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003357972659701649,
|
|
"loss": 5.0607,
|
|
"mean_token_accuracy": 0.18565811663866044,
|
|
"num_tokens": 10338483.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.489884614944458,
|
|
"epoch": 4.723146747352496,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00033541526136318564,
|
|
"loss": 4.9798,
|
|
"mean_token_accuracy": 0.1904754728078842,
|
|
"num_tokens": 10349522.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.41412386894226,
|
|
"epoch": 4.728189611699445,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0003350330689841804,
|
|
"loss": 4.9777,
|
|
"mean_token_accuracy": 0.18288532346487046,
|
|
"num_tokens": 10361036.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.537215852737427,
|
|
"epoch": 4.733232476046394,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003346506900210118,
|
|
"loss": 5.1009,
|
|
"mean_token_accuracy": 0.18369718343019487,
|
|
"num_tokens": 10372673.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.506333637237549,
|
|
"epoch": 4.7382753403933435,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003342681256621227,
|
|
"loss": 5.1216,
|
|
"mean_token_accuracy": 0.18331604450941086,
|
|
"num_tokens": 10384100.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.501450443267823,
|
|
"epoch": 4.743318204740293,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003338853770965319,
|
|
"loss": 5.0224,
|
|
"mean_token_accuracy": 0.1799652710556984,
|
|
"num_tokens": 10395503.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.489291143417359,
|
|
"epoch": 4.748361069087242,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033350244551383064,
|
|
"loss": 4.9826,
|
|
"mean_token_accuracy": 0.18880001753568648,
|
|
"num_tokens": 10406147.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.448668718338013,
|
|
"epoch": 4.753403933434191,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003331193321041793,
|
|
"loss": 4.9133,
|
|
"mean_token_accuracy": 0.19162613302469253,
|
|
"num_tokens": 10416659.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.447510385513306,
|
|
"epoch": 4.75844679778114,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00033273603805830296,
|
|
"loss": 5.017,
|
|
"mean_token_accuracy": 0.18302143216133118,
|
|
"num_tokens": 10427765.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.495143079757691,
|
|
"epoch": 4.763489662128089,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033235256456748847,
|
|
"loss": 4.968,
|
|
"mean_token_accuracy": 0.19394523948431014,
|
|
"num_tokens": 10439376.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.441805267333985,
|
|
"epoch": 4.768532526475038,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003319689128235804,
|
|
"loss": 5.0332,
|
|
"mean_token_accuracy": 0.18095729649066924,
|
|
"num_tokens": 10449232.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.386273813247681,
|
|
"epoch": 4.773575390821987,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033158508401897707,
|
|
"loss": 4.9594,
|
|
"mean_token_accuracy": 0.18342937231063844,
|
|
"num_tokens": 10459258.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.4791380882263185,
|
|
"epoch": 4.778618255168936,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003312010793466273,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.18463625460863115,
|
|
"num_tokens": 10469909.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.5161216259002686,
|
|
"epoch": 4.783661119515885,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003308169000000265,
|
|
"loss": 4.9804,
|
|
"mean_token_accuracy": 0.187069633603096,
|
|
"num_tokens": 10479848.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.489686107635498,
|
|
"epoch": 4.788703983862834,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003304325471732129,
|
|
"loss": 5.023,
|
|
"mean_token_accuracy": 0.1826684832572937,
|
|
"num_tokens": 10490352.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.384819364547729,
|
|
"epoch": 4.793746848209783,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033004802206076396,
|
|
"loss": 4.9376,
|
|
"mean_token_accuracy": 0.19013286381959915,
|
|
"num_tokens": 10501905.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.473294591903686,
|
|
"epoch": 4.798789712556732,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032966332585779263,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.1794707238674164,
|
|
"num_tokens": 10512491.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.493734693527221,
|
|
"epoch": 4.803832576903681,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00032927845975994356,
|
|
"loss": 4.8843,
|
|
"mean_token_accuracy": 0.20153878331184388,
|
|
"num_tokens": 10522686.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.427979230880737,
|
|
"epoch": 4.80887544125063,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032889342496338953,
|
|
"loss": 4.9755,
|
|
"mean_token_accuracy": 0.18965854346752167,
|
|
"num_tokens": 10534240.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.464452028274536,
|
|
"epoch": 4.813918305597579,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003285082226648273,
|
|
"loss": 5.0975,
|
|
"mean_token_accuracy": 0.179216468334198,
|
|
"num_tokens": 10545096.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.554968786239624,
|
|
"epoch": 4.8189611699445285,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003281228540614749,
|
|
"loss": 5.0812,
|
|
"mean_token_accuracy": 0.1851471796631813,
|
|
"num_tokens": 10557076.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.406831836700439,
|
|
"epoch": 4.824004034291478,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032773732035106657,
|
|
"loss": 4.8959,
|
|
"mean_token_accuracy": 0.18923303633928298,
|
|
"num_tokens": 10567635.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.529042100906372,
|
|
"epoch": 4.829046898638427,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032735162273185006,
|
|
"loss": 5.1076,
|
|
"mean_token_accuracy": 0.17858670949935912,
|
|
"num_tokens": 10579488.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.456575202941894,
|
|
"epoch": 4.834089762985376,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003269657624025824,
|
|
"loss": 4.9714,
|
|
"mean_token_accuracy": 0.18321664482355118,
|
|
"num_tokens": 10591092.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.438525056838989,
|
|
"epoch": 4.839132627332325,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003265797405625265,
|
|
"loss": 5.0281,
|
|
"mean_token_accuracy": 0.18071822375059127,
|
|
"num_tokens": 10603165.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.401812267303467,
|
|
"epoch": 4.844175491679274,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003261935584114471,
|
|
"loss": 4.9175,
|
|
"mean_token_accuracy": 0.19631239920854568,
|
|
"num_tokens": 10615149.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.571744012832641,
|
|
"epoch": 4.849218356026223,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003258072171496072,
|
|
"loss": 5.093,
|
|
"mean_token_accuracy": 0.18332237750291824,
|
|
"num_tokens": 10626786.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.500848627090454,
|
|
"epoch": 4.854261220373172,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003254207179777644,
|
|
"loss": 4.9911,
|
|
"mean_token_accuracy": 0.18388643711805344,
|
|
"num_tokens": 10637107.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.395619106292725,
|
|
"epoch": 4.8593040847201205,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000325034062097167,
|
|
"loss": 4.9915,
|
|
"mean_token_accuracy": 0.18710375726222991,
|
|
"num_tokens": 10648198.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.557878065109253,
|
|
"epoch": 4.86434694906707,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032464725070955046,
|
|
"loss": 5.1159,
|
|
"mean_token_accuracy": 0.178499798476696,
|
|
"num_tokens": 10659867.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.529823064804077,
|
|
"epoch": 4.869389813414019,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032426028501713347,
|
|
"loss": 5.0439,
|
|
"mean_token_accuracy": 0.1776138499379158,
|
|
"num_tokens": 10672117.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.440957927703858,
|
|
"epoch": 4.874432677760968,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032387316622261425,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.190153568983078,
|
|
"num_tokens": 10682014.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.51104588508606,
|
|
"epoch": 4.879475542107917,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032348589552916703,
|
|
"loss": 5.1082,
|
|
"mean_token_accuracy": 0.1756524071097374,
|
|
"num_tokens": 10693330.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.479088926315308,
|
|
"epoch": 4.884518406454866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003230984741404381,
|
|
"loss": 4.9707,
|
|
"mean_token_accuracy": 0.191126349568367,
|
|
"num_tokens": 10703772.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.515173673629761,
|
|
"epoch": 4.889561270801815,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003227109032605419,
|
|
"loss": 5.0429,
|
|
"mean_token_accuracy": 0.18115611523389816,
|
|
"num_tokens": 10714088.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.369802856445313,
|
|
"epoch": 4.894604135148764,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003223231840940578,
|
|
"loss": 4.8587,
|
|
"mean_token_accuracy": 0.19165623635053636,
|
|
"num_tokens": 10724701.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.456402683258057,
|
|
"epoch": 4.8996469994957135,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032193531784602595,
|
|
"loss": 4.9891,
|
|
"mean_token_accuracy": 0.1859570935368538,
|
|
"num_tokens": 10735361.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.523557186126709,
|
|
"epoch": 4.904689863842663,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032154730572194346,
|
|
"loss": 4.9989,
|
|
"mean_token_accuracy": 0.1783282518386841,
|
|
"num_tokens": 10745909.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.480124759674072,
|
|
"epoch": 4.909732728189612,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000321159148927761,
|
|
"loss": 5.1088,
|
|
"mean_token_accuracy": 0.18037252575159074,
|
|
"num_tokens": 10759305.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.44937653541565,
|
|
"epoch": 4.914775592536561,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000320770848669879,
|
|
"loss": 5.0645,
|
|
"mean_token_accuracy": 0.18601060956716536,
|
|
"num_tokens": 10770451.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.483103322982788,
|
|
"epoch": 4.91981845688351,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032038240615514345,
|
|
"loss": 4.9667,
|
|
"mean_token_accuracy": 0.18633958846330642,
|
|
"num_tokens": 10782376.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.467893648147583,
|
|
"epoch": 4.924861321230459,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003199938225908426,
|
|
"loss": 5.0918,
|
|
"mean_token_accuracy": 0.18133595138788222,
|
|
"num_tokens": 10794252.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.487560796737671,
|
|
"epoch": 4.929904185577408,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003196050991847033,
|
|
"loss": 5.0599,
|
|
"mean_token_accuracy": 0.1813935562968254,
|
|
"num_tokens": 10804117.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.480642938613892,
|
|
"epoch": 4.934947049924357,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003192162371448868,
|
|
"loss": 4.9675,
|
|
"mean_token_accuracy": 0.18790970742702484,
|
|
"num_tokens": 10815798.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.429426193237305,
|
|
"epoch": 4.939989914271306,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00031882723767998517,
|
|
"loss": 5.0025,
|
|
"mean_token_accuracy": 0.1859897643327713,
|
|
"num_tokens": 10827135.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.4674952030181885,
|
|
"epoch": 4.9450327786182555,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00031843810199901795,
|
|
"loss": 5.0954,
|
|
"mean_token_accuracy": 0.1781628742814064,
|
|
"num_tokens": 10838988.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.483274459838867,
|
|
"epoch": 4.950075642965205,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003180488313114277,
|
|
"loss": 4.9206,
|
|
"mean_token_accuracy": 0.1916445016860962,
|
|
"num_tokens": 10848223.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.493588304519653,
|
|
"epoch": 4.955118507312154,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003176594268270766,
|
|
"loss": 4.971,
|
|
"mean_token_accuracy": 0.1868692457675934,
|
|
"num_tokens": 10859052.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.462699604034424,
|
|
"epoch": 4.960161371659103,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00031726988975624293,
|
|
"loss": 5.0236,
|
|
"mean_token_accuracy": 0.18071905225515367,
|
|
"num_tokens": 10869108.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.514888858795166,
|
|
"epoch": 4.965204236006051,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00031688022130961675,
|
|
"loss": 5.0306,
|
|
"mean_token_accuracy": 0.18339766561985016,
|
|
"num_tokens": 10880691.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.435796070098877,
|
|
"epoch": 4.970247100353,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003164904226982967,
|
|
"loss": 4.9201,
|
|
"mean_token_accuracy": 0.1864425867795944,
|
|
"num_tokens": 10892206.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.45585527420044,
|
|
"epoch": 4.975289964699949,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031610049513378574,
|
|
"loss": 4.9012,
|
|
"mean_token_accuracy": 0.19809649437665938,
|
|
"num_tokens": 10902130.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.465006542205811,
|
|
"epoch": 4.9803328290468984,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00031571043982798786,
|
|
"loss": 5.0311,
|
|
"mean_token_accuracy": 0.18464192152023315,
|
|
"num_tokens": 10913424.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.463282108306885,
|
|
"epoch": 4.985375693393848,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003153202579932039,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.1808464601635933,
|
|
"num_tokens": 10925718.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.463268184661866,
|
|
"epoch": 4.990418557740797,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031492995084212794,
|
|
"loss": 4.9748,
|
|
"mean_token_accuracy": 0.18657407611608506,
|
|
"num_tokens": 10937102.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.464972114562988,
|
|
"epoch": 4.995461422087746,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003145395195878437,
|
|
"loss": 4.9743,
|
|
"mean_token_accuracy": 0.18652818351984024,
|
|
"num_tokens": 10948679.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.423767142825657,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 0.00031414896544382047,
|
|
"loss": 4.9168,
|
|
"mean_token_accuracy": 0.1881958411799537,
|
|
"num_tokens": 10958200.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.453699445724487,
|
|
"epoch": 5.005042864346949,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003137582896239095,
|
|
"loss": 4.7505,
|
|
"mean_token_accuracy": 0.20444145202636718,
|
|
"num_tokens": 10969775.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.427678203582763,
|
|
"epoch": 5.010085728693898,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003133674933423404,
|
|
"loss": 4.7377,
|
|
"mean_token_accuracy": 0.20255281031131744,
|
|
"num_tokens": 10980423.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.397994327545166,
|
|
"epoch": 5.015128593040847,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00031297657781371687,
|
|
"loss": 4.7564,
|
|
"mean_token_accuracy": 0.20450910776853562,
|
|
"num_tokens": 10992716.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.51829719543457,
|
|
"epoch": 5.0201714573877965,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003125855442530135,
|
|
"loss": 4.8431,
|
|
"mean_token_accuracy": 0.20113045126199722,
|
|
"num_tokens": 11004597.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.4910666942596436,
|
|
"epoch": 5.025214321734746,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031219439387557155,
|
|
"loss": 4.8115,
|
|
"mean_token_accuracy": 0.19921248853206636,
|
|
"num_tokens": 11015423.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.455689191818237,
|
|
"epoch": 5.030257186081695,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00031180312789709535,
|
|
"loss": 4.8465,
|
|
"mean_token_accuracy": 0.2048381209373474,
|
|
"num_tokens": 11028191.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.418379163742065,
|
|
"epoch": 5.035300050428644,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003114117475336487,
|
|
"loss": 4.8474,
|
|
"mean_token_accuracy": 0.19876948446035386,
|
|
"num_tokens": 11040690.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.410862779617309,
|
|
"epoch": 5.040342914775593,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003110202540016509,
|
|
"loss": 4.7206,
|
|
"mean_token_accuracy": 0.20554412752389908,
|
|
"num_tokens": 11052580.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.040342914775593,
|
|
"eval_entropy": 5.313159577415797,
|
|
"eval_loss": 5.675646781921387,
|
|
"eval_mean_token_accuracy": 0.15337524381075196,
|
|
"eval_num_tokens": 11052580.0,
|
|
"eval_runtime": 2.0792,
|
|
"eval_samples_per_second": 1509.241,
|
|
"eval_steps_per_second": 189.016,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"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": 2740966200668160.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|