c33b01fdac
Model: fpadovani/ita-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed455 Source: Original Platform
2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0080685829551186,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.861198043823242,
|
|
"epoch": 0.005042864346949067,
|
|
"grad_norm": 17.75,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3685,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10258.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.86604413986206,
|
|
"epoch": 0.010085728693898134,
|
|
"grad_norm": 17.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.4019,
|
|
"mean_token_accuracy": 0.00016911421553231776,
|
|
"num_tokens": 21166.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.88087158203125,
|
|
"epoch": 0.015128593040847202,
|
|
"grad_norm": 19.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.2401,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 31390.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.931153154373169,
|
|
"epoch": 0.020171457387796268,
|
|
"grad_norm": 23.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.9419,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 40983.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.108882856369019,
|
|
"epoch": 0.025214321734745335,
|
|
"grad_norm": 32.5,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.5698,
|
|
"mean_token_accuracy": 0.0002748680766671896,
|
|
"num_tokens": 52673.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 5.765079593658447,
|
|
"epoch": 0.030257186081694403,
|
|
"grad_norm": 42.0,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 12.7033,
|
|
"mean_token_accuracy": 0.00023715496063232423,
|
|
"num_tokens": 64956.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 8.901754570007324,
|
|
"epoch": 0.03530005042864347,
|
|
"grad_norm": 5.0625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 11.2035,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 76492.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.649229621887207,
|
|
"epoch": 0.040342914775592535,
|
|
"grad_norm": 3.171875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.739,
|
|
"mean_token_accuracy": 5.611672531813383e-05,
|
|
"num_tokens": 89191.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.720369338989258,
|
|
"epoch": 0.0453857791225416,
|
|
"grad_norm": 3.5625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.6068,
|
|
"mean_token_accuracy": 0.002752099640201777,
|
|
"num_tokens": 100077.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.725068855285645,
|
|
"epoch": 0.05042864346949067,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.4492,
|
|
"mean_token_accuracy": 0.009254818642511963,
|
|
"num_tokens": 110530.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.668148040771484,
|
|
"epoch": 0.05547150781643974,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.3025,
|
|
"mean_token_accuracy": 0.008732812548987567,
|
|
"num_tokens": 121113.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.679216957092285,
|
|
"epoch": 0.060514372163388806,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1662,
|
|
"mean_token_accuracy": 0.013584407884627581,
|
|
"num_tokens": 133912.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.712758827209473,
|
|
"epoch": 0.06555723651033787,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0595,
|
|
"mean_token_accuracy": 0.015708898566663264,
|
|
"num_tokens": 145322.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.717932033538819,
|
|
"epoch": 0.07060010085728693,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.975,
|
|
"mean_token_accuracy": 0.02391951158642769,
|
|
"num_tokens": 156252.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.711066150665284,
|
|
"epoch": 0.07564296520423601,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.8386,
|
|
"mean_token_accuracy": 0.04081910476088524,
|
|
"num_tokens": 166622.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.719244766235352,
|
|
"epoch": 0.08068582955118507,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7625,
|
|
"mean_token_accuracy": 0.043442795425653456,
|
|
"num_tokens": 177591.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.715385818481446,
|
|
"epoch": 0.08572869389813415,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6713,
|
|
"mean_token_accuracy": 0.03803261611610651,
|
|
"num_tokens": 187764.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.708494853973388,
|
|
"epoch": 0.0907715582450832,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.5773,
|
|
"mean_token_accuracy": 0.04156861491501331,
|
|
"num_tokens": 197810.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.700858688354492,
|
|
"epoch": 0.09581442259203228,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.5398,
|
|
"mean_token_accuracy": 0.03897072076797485,
|
|
"num_tokens": 208552.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.691888046264648,
|
|
"epoch": 0.10085728693898134,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4336,
|
|
"mean_token_accuracy": 0.04147556833922863,
|
|
"num_tokens": 221628.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.673360824584961,
|
|
"epoch": 0.1059001512859304,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2838,
|
|
"mean_token_accuracy": 0.04640204608440399,
|
|
"num_tokens": 231033.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.63631467819214,
|
|
"epoch": 0.11094301563287948,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.2331,
|
|
"mean_token_accuracy": 0.039961645379662514,
|
|
"num_tokens": 242259.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.613164234161378,
|
|
"epoch": 0.11598587997982854,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1455,
|
|
"mean_token_accuracy": 0.04157233964651823,
|
|
"num_tokens": 254116.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.582231140136718,
|
|
"epoch": 0.12102874432677761,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9531,
|
|
"mean_token_accuracy": 0.04660932868719101,
|
|
"num_tokens": 264518.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.530023288726806,
|
|
"epoch": 0.1260716086737267,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8083,
|
|
"mean_token_accuracy": 0.045368751883506774,
|
|
"num_tokens": 273783.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.469050121307372,
|
|
"epoch": 0.13111447302067575,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.766,
|
|
"mean_token_accuracy": 0.049766818061470985,
|
|
"num_tokens": 284568.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.408735752105713,
|
|
"epoch": 0.1361573373676248,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.6705,
|
|
"mean_token_accuracy": 0.04497178308665752,
|
|
"num_tokens": 296499.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.32727508544922,
|
|
"epoch": 0.14120020171457387,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4921,
|
|
"mean_token_accuracy": 0.04762334674596787,
|
|
"num_tokens": 307013.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.229682350158692,
|
|
"epoch": 0.14624306606152296,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.4351,
|
|
"mean_token_accuracy": 0.04772641360759735,
|
|
"num_tokens": 318604.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.111124515533447,
|
|
"epoch": 0.15128593040847202,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.3261,
|
|
"mean_token_accuracy": 0.04569742903113365,
|
|
"num_tokens": 328968.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.94740915298462,
|
|
"epoch": 0.15632879475542108,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1758,
|
|
"mean_token_accuracy": 0.049343152344226836,
|
|
"num_tokens": 338950.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.782429695129395,
|
|
"epoch": 0.16137165910237014,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.1492,
|
|
"mean_token_accuracy": 0.046918957680463794,
|
|
"num_tokens": 350667.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.565590858459473,
|
|
"epoch": 0.1664145234493192,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.0798,
|
|
"mean_token_accuracy": 0.04624507054686546,
|
|
"num_tokens": 361781.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.344639873504638,
|
|
"epoch": 0.1714573877962683,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9366,
|
|
"mean_token_accuracy": 0.04954341873526573,
|
|
"num_tokens": 373410.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.09048776626587,
|
|
"epoch": 0.17650025214321735,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8345,
|
|
"mean_token_accuracy": 0.05336628369987011,
|
|
"num_tokens": 385302.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.843950366973877,
|
|
"epoch": 0.1815431164901664,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.8151,
|
|
"mean_token_accuracy": 0.045010988228023054,
|
|
"num_tokens": 395183.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.655001354217529,
|
|
"epoch": 0.18658598083711547,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7734,
|
|
"mean_token_accuracy": 0.04725377857685089,
|
|
"num_tokens": 405356.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.491672897338868,
|
|
"epoch": 0.19162884518406456,
|
|
"grad_norm": 0.734375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7416,
|
|
"mean_token_accuracy": 0.04555538780987263,
|
|
"num_tokens": 415864.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.389861965179444,
|
|
"epoch": 0.19667170953101362,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6937,
|
|
"mean_token_accuracy": 0.050760002061724666,
|
|
"num_tokens": 426225.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.245965766906739,
|
|
"epoch": 0.20171457387796268,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.7303,
|
|
"mean_token_accuracy": 0.048710960522294046,
|
|
"num_tokens": 436847.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.166359043121338,
|
|
"epoch": 0.20675743822491174,
|
|
"grad_norm": 0.58203125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7034,
|
|
"mean_token_accuracy": 0.04945480115711689,
|
|
"num_tokens": 448807.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.179372882843017,
|
|
"epoch": 0.2118003025718608,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6746,
|
|
"mean_token_accuracy": 0.045829778909683226,
|
|
"num_tokens": 459691.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.181133460998534,
|
|
"epoch": 0.2168431669188099,
|
|
"grad_norm": 0.63671875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.6769,
|
|
"mean_token_accuracy": 0.046935905888676646,
|
|
"num_tokens": 470097.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.214150047302246,
|
|
"epoch": 0.22188603126575895,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.6151,
|
|
"mean_token_accuracy": 0.05064198933541775,
|
|
"num_tokens": 482006.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.223631191253663,
|
|
"epoch": 0.22692889561270801,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.6701,
|
|
"mean_token_accuracy": 0.05050523765385151,
|
|
"num_tokens": 493134.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.13507890701294,
|
|
"epoch": 0.23197175995965708,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.641,
|
|
"mean_token_accuracy": 0.05316512696444988,
|
|
"num_tokens": 504801.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.20094051361084,
|
|
"epoch": 0.23701462430660616,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.687,
|
|
"mean_token_accuracy": 0.053307381272315976,
|
|
"num_tokens": 517523.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.198506450653076,
|
|
"epoch": 0.24205748865355523,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.5752,
|
|
"mean_token_accuracy": 0.056617460772395135,
|
|
"num_tokens": 528267.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.212173080444336,
|
|
"epoch": 0.24710035300050429,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.5815,
|
|
"mean_token_accuracy": 0.05843498557806015,
|
|
"num_tokens": 538750.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.089133930206298,
|
|
"epoch": 0.2521432173474534,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.5768,
|
|
"mean_token_accuracy": 0.05827867574989796,
|
|
"num_tokens": 549380.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.13608980178833,
|
|
"epoch": 0.25718608169440244,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.5477,
|
|
"mean_token_accuracy": 0.061000148952007296,
|
|
"num_tokens": 559643.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.085019493103028,
|
|
"epoch": 0.2622289460413515,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.5338,
|
|
"mean_token_accuracy": 0.05923284590244293,
|
|
"num_tokens": 571250.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.101050186157227,
|
|
"epoch": 0.26727181038830056,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.5085,
|
|
"mean_token_accuracy": 0.0720983661711216,
|
|
"num_tokens": 582406.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.07942600250244,
|
|
"epoch": 0.2723146747352496,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.571,
|
|
"mean_token_accuracy": 0.06375469006597996,
|
|
"num_tokens": 593115.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.018554973602296,
|
|
"epoch": 0.2773575390821987,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.4442,
|
|
"mean_token_accuracy": 0.0684063233435154,
|
|
"num_tokens": 603938.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.983159112930298,
|
|
"epoch": 0.28240040342914774,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.4799,
|
|
"mean_token_accuracy": 0.06466774754226208,
|
|
"num_tokens": 614067.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.012670087814332,
|
|
"epoch": 0.2874432677760968,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.4875,
|
|
"mean_token_accuracy": 0.06548541337251663,
|
|
"num_tokens": 626271.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.980437278747559,
|
|
"epoch": 0.2924861321230459,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.4434,
|
|
"mean_token_accuracy": 0.0688519038259983,
|
|
"num_tokens": 636580.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.933881235122681,
|
|
"epoch": 0.297528996469995,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3571,
|
|
"mean_token_accuracy": 0.06868328899145126,
|
|
"num_tokens": 648202.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.9951011657714846,
|
|
"epoch": 0.30257186081694404,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.4313,
|
|
"mean_token_accuracy": 0.06523990221321582,
|
|
"num_tokens": 659274.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.9290464401245115,
|
|
"epoch": 0.3076147251638931,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.4182,
|
|
"mean_token_accuracy": 0.06868452057242394,
|
|
"num_tokens": 670782.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.905453872680664,
|
|
"epoch": 0.31265758951084216,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.3428,
|
|
"mean_token_accuracy": 0.06368493661284447,
|
|
"num_tokens": 681392.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.886255693435669,
|
|
"epoch": 0.3177004538577912,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.345,
|
|
"mean_token_accuracy": 0.07283873222768307,
|
|
"num_tokens": 692779.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.934207105636597,
|
|
"epoch": 0.3227433182047403,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.3834,
|
|
"mean_token_accuracy": 0.07133128270506858,
|
|
"num_tokens": 702200.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.806493806838989,
|
|
"epoch": 0.32778618255168934,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.3336,
|
|
"mean_token_accuracy": 0.06831678487360478,
|
|
"num_tokens": 714817.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.848582601547241,
|
|
"epoch": 0.3328290468986384,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2545,
|
|
"mean_token_accuracy": 0.06371853351593018,
|
|
"num_tokens": 725555.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.847062253952027,
|
|
"epoch": 0.3378719112455875,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.2501,
|
|
"mean_token_accuracy": 0.06582499109208584,
|
|
"num_tokens": 737372.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.837714433670044,
|
|
"epoch": 0.3429147755925366,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2999,
|
|
"mean_token_accuracy": 0.07149351052939892,
|
|
"num_tokens": 748515.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.744440412521362,
|
|
"epoch": 0.34795763993948564,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.3175,
|
|
"mean_token_accuracy": 0.06945871189236641,
|
|
"num_tokens": 760586.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.766058301925659,
|
|
"epoch": 0.3530005042864347,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2489,
|
|
"mean_token_accuracy": 0.07091887481510639,
|
|
"num_tokens": 772228.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.782631349563599,
|
|
"epoch": 0.35804336863338376,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.3217,
|
|
"mean_token_accuracy": 0.06774664521217347,
|
|
"num_tokens": 784035.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.645714569091797,
|
|
"epoch": 0.3630862329803328,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.1914,
|
|
"mean_token_accuracy": 0.07828422412276267,
|
|
"num_tokens": 794383.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.75626392364502,
|
|
"epoch": 0.3681290973272819,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1439,
|
|
"mean_token_accuracy": 0.07751081734895707,
|
|
"num_tokens": 805192.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.646742677688598,
|
|
"epoch": 0.37317196167423095,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.2756,
|
|
"mean_token_accuracy": 0.07152257226407528,
|
|
"num_tokens": 816794.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.783850574493409,
|
|
"epoch": 0.37821482602118,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.2993,
|
|
"mean_token_accuracy": 0.07075197845697404,
|
|
"num_tokens": 827505.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.719998741149903,
|
|
"epoch": 0.3832576903681291,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1357,
|
|
"mean_token_accuracy": 0.07439711764454841,
|
|
"num_tokens": 837462.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.620962858200073,
|
|
"epoch": 0.3883005547150782,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1072,
|
|
"mean_token_accuracy": 0.07790239676833152,
|
|
"num_tokens": 848309.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.697546195983887,
|
|
"epoch": 0.39334341906202724,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.2305,
|
|
"mean_token_accuracy": 0.07441804558038712,
|
|
"num_tokens": 859474.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.653050518035888,
|
|
"epoch": 0.3983862834089763,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1129,
|
|
"mean_token_accuracy": 0.07637989819049835,
|
|
"num_tokens": 871638.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.5871381759643555,
|
|
"epoch": 0.40342914775592537,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1506,
|
|
"mean_token_accuracy": 0.07561550661921501,
|
|
"num_tokens": 882480.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.699258184432983,
|
|
"epoch": 0.4084720121028744,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.1518,
|
|
"mean_token_accuracy": 0.07447279319167137,
|
|
"num_tokens": 893366.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.562766981124878,
|
|
"epoch": 0.4135148764498235,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.1168,
|
|
"mean_token_accuracy": 0.08146974667906762,
|
|
"num_tokens": 904678.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.616194534301758,
|
|
"epoch": 0.41855774079677255,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.088,
|
|
"mean_token_accuracy": 0.07914503738284111,
|
|
"num_tokens": 915756.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.499222898483277,
|
|
"epoch": 0.4236006051437216,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0765,
|
|
"mean_token_accuracy": 0.0707643773406744,
|
|
"num_tokens": 926204.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.52804388999939,
|
|
"epoch": 0.4286434694906707,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.11,
|
|
"mean_token_accuracy": 0.08115734942257405,
|
|
"num_tokens": 936894.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.593617248535156,
|
|
"epoch": 0.4336863338376198,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1412,
|
|
"mean_token_accuracy": 0.08076336197555065,
|
|
"num_tokens": 948987.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.59008822441101,
|
|
"epoch": 0.43872919818456885,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.124,
|
|
"mean_token_accuracy": 0.07760139107704163,
|
|
"num_tokens": 961036.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.5442492961883545,
|
|
"epoch": 0.4437720625315179,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.1393,
|
|
"mean_token_accuracy": 0.0726228266954422,
|
|
"num_tokens": 972445.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.532951831817627,
|
|
"epoch": 0.44881492687846697,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.1081,
|
|
"mean_token_accuracy": 0.07657850906252861,
|
|
"num_tokens": 982847.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.5546064376831055,
|
|
"epoch": 0.45385779122541603,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0673,
|
|
"mean_token_accuracy": 0.07684047818183899,
|
|
"num_tokens": 994055.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.523573160171509,
|
|
"epoch": 0.4589006555723651,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.0516,
|
|
"mean_token_accuracy": 0.0797266386449337,
|
|
"num_tokens": 1003565.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.470361423492432,
|
|
"epoch": 0.46394351991931415,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9668,
|
|
"mean_token_accuracy": 0.0837884522974491,
|
|
"num_tokens": 1013376.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.455811643600464,
|
|
"epoch": 0.4689863842662632,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0285,
|
|
"mean_token_accuracy": 0.08764110282063484,
|
|
"num_tokens": 1024580.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.433918380737305,
|
|
"epoch": 0.47402924861321233,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.0367,
|
|
"mean_token_accuracy": 0.07916341610252857,
|
|
"num_tokens": 1037172.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.462331533432007,
|
|
"epoch": 0.4790721129601614,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9474,
|
|
"mean_token_accuracy": 0.0814237765967846,
|
|
"num_tokens": 1046690.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.432046413421631,
|
|
"epoch": 0.48411497730711045,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9672,
|
|
"mean_token_accuracy": 0.07880407944321632,
|
|
"num_tokens": 1058341.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.456065654754639,
|
|
"epoch": 0.4891578416540595,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9077,
|
|
"mean_token_accuracy": 0.08365077599883079,
|
|
"num_tokens": 1067544.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.459004259109497,
|
|
"epoch": 0.49420070600100857,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0003,
|
|
"mean_token_accuracy": 0.07831413969397545,
|
|
"num_tokens": 1079028.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.444475555419922,
|
|
"epoch": 0.49924357034795763,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.0642,
|
|
"mean_token_accuracy": 0.08345532491803169,
|
|
"num_tokens": 1090709.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.374783658981324,
|
|
"epoch": 0.5042864346949067,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.9266,
|
|
"mean_token_accuracy": 0.08068954423069954,
|
|
"num_tokens": 1102987.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.51790189743042,
|
|
"epoch": 0.5093292990418558,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0313,
|
|
"mean_token_accuracy": 0.0794920887798071,
|
|
"num_tokens": 1113370.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.340002822875976,
|
|
"epoch": 0.5143721633888049,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8681,
|
|
"mean_token_accuracy": 0.09227448850870132,
|
|
"num_tokens": 1124054.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.339383506774903,
|
|
"epoch": 0.5194150277357539,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.8567,
|
|
"mean_token_accuracy": 0.0867280587553978,
|
|
"num_tokens": 1134530.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.385304355621338,
|
|
"epoch": 0.524457892082703,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8604,
|
|
"mean_token_accuracy": 0.08653979301452637,
|
|
"num_tokens": 1146011.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.396727609634399,
|
|
"epoch": 0.529500756429652,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9462,
|
|
"mean_token_accuracy": 0.08311982899904251,
|
|
"num_tokens": 1155822.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.31582670211792,
|
|
"epoch": 0.5345436207766011,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8457,
|
|
"mean_token_accuracy": 0.0868146151304245,
|
|
"num_tokens": 1167398.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.356871652603149,
|
|
"epoch": 0.5395864851235502,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.9224,
|
|
"mean_token_accuracy": 0.0868445247411728,
|
|
"num_tokens": 1179706.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.2655237197875975,
|
|
"epoch": 0.5446293494704992,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.0891579382121563,
|
|
"num_tokens": 1190159.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.255985355377197,
|
|
"epoch": 0.5496722138174484,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.7742,
|
|
"mean_token_accuracy": 0.0909526750445366,
|
|
"num_tokens": 1200223.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.292838096618652,
|
|
"epoch": 0.5547150781643974,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8718,
|
|
"mean_token_accuracy": 0.08809854686260224,
|
|
"num_tokens": 1212108.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.266604948043823,
|
|
"epoch": 0.5597579425113465,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8207,
|
|
"mean_token_accuracy": 0.08790944889187813,
|
|
"num_tokens": 1221940.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.333012628555298,
|
|
"epoch": 0.5648008068582955,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8863,
|
|
"mean_token_accuracy": 0.08708575740456581,
|
|
"num_tokens": 1232884.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.273598337173462,
|
|
"epoch": 0.5698436712052446,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7908,
|
|
"mean_token_accuracy": 0.08785182014107704,
|
|
"num_tokens": 1242758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.219170236587525,
|
|
"epoch": 0.5748865355521936,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7871,
|
|
"mean_token_accuracy": 0.08846997991204261,
|
|
"num_tokens": 1254787.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.317018318176269,
|
|
"epoch": 0.5799293998991427,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.09232769683003425,
|
|
"num_tokens": 1266364.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.174505281448364,
|
|
"epoch": 0.5849722642460918,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7925,
|
|
"mean_token_accuracy": 0.08852695003151893,
|
|
"num_tokens": 1277105.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.396231412887573,
|
|
"epoch": 0.5900151285930408,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.9658,
|
|
"mean_token_accuracy": 0.08576021790504455,
|
|
"num_tokens": 1288470.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.2577635765075685,
|
|
"epoch": 0.59505799293999,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8688,
|
|
"mean_token_accuracy": 0.08686786629259587,
|
|
"num_tokens": 1301596.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.261093330383301,
|
|
"epoch": 0.600100857286939,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.9126,
|
|
"mean_token_accuracy": 0.08601501062512398,
|
|
"num_tokens": 1314102.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.230313920974732,
|
|
"epoch": 0.6051437216338881,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7932,
|
|
"mean_token_accuracy": 0.09197813868522645,
|
|
"num_tokens": 1326939.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.304633665084839,
|
|
"epoch": 0.6101865859808371,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.8429,
|
|
"mean_token_accuracy": 0.08887591883540154,
|
|
"num_tokens": 1337877.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.212860345840454,
|
|
"epoch": 0.6152294503277862,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.746,
|
|
"mean_token_accuracy": 0.09503204077482223,
|
|
"num_tokens": 1348425.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.211596918106079,
|
|
"epoch": 0.6202723146747352,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7806,
|
|
"mean_token_accuracy": 0.09582387432456016,
|
|
"num_tokens": 1359216.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.14785475730896,
|
|
"epoch": 0.6253151790216843,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7742,
|
|
"mean_token_accuracy": 0.0927226796746254,
|
|
"num_tokens": 1371181.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.265208148956299,
|
|
"epoch": 0.6303580433686334,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7655,
|
|
"mean_token_accuracy": 0.09117320030927659,
|
|
"num_tokens": 1382371.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.17245774269104,
|
|
"epoch": 0.6354009077155824,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7371,
|
|
"mean_token_accuracy": 0.09117325544357299,
|
|
"num_tokens": 1392981.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.17719988822937,
|
|
"epoch": 0.6404437720625316,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7773,
|
|
"mean_token_accuracy": 0.09603490680456161,
|
|
"num_tokens": 1405782.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.185710000991821,
|
|
"epoch": 0.6454866364094806,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7145,
|
|
"mean_token_accuracy": 0.09524220377206802,
|
|
"num_tokens": 1415809.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.095476579666138,
|
|
"epoch": 0.6505295007564297,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7341,
|
|
"mean_token_accuracy": 0.09393899962306022,
|
|
"num_tokens": 1427845.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.129822778701782,
|
|
"epoch": 0.6555723651033787,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7511,
|
|
"mean_token_accuracy": 0.08939804583787918,
|
|
"num_tokens": 1439761.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.1000738620758055,
|
|
"epoch": 0.6606152294503278,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.715,
|
|
"mean_token_accuracy": 0.09929769858717918,
|
|
"num_tokens": 1450373.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.087963581085205,
|
|
"epoch": 0.6656580937972768,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.5962,
|
|
"mean_token_accuracy": 0.09877277910709381,
|
|
"num_tokens": 1462134.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.068597888946533,
|
|
"epoch": 0.6707009581442259,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7009,
|
|
"mean_token_accuracy": 0.09789054095745087,
|
|
"num_tokens": 1473600.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.079747152328491,
|
|
"epoch": 0.675743822491175,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6277,
|
|
"mean_token_accuracy": 0.09865574762225152,
|
|
"num_tokens": 1484313.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.130831575393676,
|
|
"epoch": 0.680786686838124,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7239,
|
|
"mean_token_accuracy": 0.09442749768495559,
|
|
"num_tokens": 1495407.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.127714824676514,
|
|
"epoch": 0.6858295511850732,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7836,
|
|
"mean_token_accuracy": 0.0916206806898117,
|
|
"num_tokens": 1506707.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.020711421966553,
|
|
"epoch": 0.6908724155320222,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.626,
|
|
"mean_token_accuracy": 0.09958503022789955,
|
|
"num_tokens": 1517991.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.049675178527832,
|
|
"epoch": 0.6959152798789713,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6744,
|
|
"mean_token_accuracy": 0.09645361006259918,
|
|
"num_tokens": 1529538.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.978589725494385,
|
|
"epoch": 0.7009581442259203,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6592,
|
|
"mean_token_accuracy": 0.0996648609638214,
|
|
"num_tokens": 1540183.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.107329940795898,
|
|
"epoch": 0.7060010085728694,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.705,
|
|
"mean_token_accuracy": 0.09683080986142159,
|
|
"num_tokens": 1552899.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.037663221359253,
|
|
"epoch": 0.7110438729198184,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6628,
|
|
"mean_token_accuracy": 0.09990261644124984,
|
|
"num_tokens": 1563219.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.026231288909912,
|
|
"epoch": 0.7160867372667675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6104,
|
|
"mean_token_accuracy": 0.09809769243001938,
|
|
"num_tokens": 1573508.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.034958696365356,
|
|
"epoch": 0.7211296016137166,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6829,
|
|
"mean_token_accuracy": 0.09331593811511993,
|
|
"num_tokens": 1584319.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.9749456405639645,
|
|
"epoch": 0.7261724659606656,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5512,
|
|
"mean_token_accuracy": 0.10765914916992188,
|
|
"num_tokens": 1595986.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.940080738067627,
|
|
"epoch": 0.7312153303076148,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6202,
|
|
"mean_token_accuracy": 0.09700567573308945,
|
|
"num_tokens": 1606742.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.984588623046875,
|
|
"epoch": 0.7362581946545638,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.596,
|
|
"mean_token_accuracy": 0.09722576439380645,
|
|
"num_tokens": 1618325.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.967196083068847,
|
|
"epoch": 0.7413010590015129,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5904,
|
|
"mean_token_accuracy": 0.09871675670146943,
|
|
"num_tokens": 1627693.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.934539794921875,
|
|
"epoch": 0.7463439233484619,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5436,
|
|
"mean_token_accuracy": 0.09849280044436455,
|
|
"num_tokens": 1637707.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.968688011169434,
|
|
"epoch": 0.751386787695411,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.555,
|
|
"mean_token_accuracy": 0.10745493024587631,
|
|
"num_tokens": 1647277.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.964085435867309,
|
|
"epoch": 0.75642965204236,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6135,
|
|
"mean_token_accuracy": 0.0993950642645359,
|
|
"num_tokens": 1657919.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.996398067474365,
|
|
"epoch": 0.7614725163893091,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6403,
|
|
"mean_token_accuracy": 0.09593973234295845,
|
|
"num_tokens": 1669107.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.903709173202515,
|
|
"epoch": 0.7665153807362582,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.65,
|
|
"mean_token_accuracy": 0.1014159619808197,
|
|
"num_tokens": 1681338.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.944526243209839,
|
|
"epoch": 0.7715582450832073,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.4847,
|
|
"mean_token_accuracy": 0.10121021494269371,
|
|
"num_tokens": 1691288.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.795492315292359,
|
|
"epoch": 0.7766011094301564,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5786,
|
|
"mean_token_accuracy": 0.10151153206825256,
|
|
"num_tokens": 1702864.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.9314769268035885,
|
|
"epoch": 0.7816439737771054,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4859,
|
|
"mean_token_accuracy": 0.10732312425971031,
|
|
"num_tokens": 1712454.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.922724056243896,
|
|
"epoch": 0.7866868381240545,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6473,
|
|
"mean_token_accuracy": 0.10383656024932861,
|
|
"num_tokens": 1723925.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.883858871459961,
|
|
"epoch": 0.7917297024710035,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5651,
|
|
"mean_token_accuracy": 0.10197380036115647,
|
|
"num_tokens": 1735624.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.861594009399414,
|
|
"epoch": 0.7967725668179526,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5574,
|
|
"mean_token_accuracy": 0.10136305913329124,
|
|
"num_tokens": 1746288.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.003063011169433,
|
|
"epoch": 0.8018154311649016,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6895,
|
|
"mean_token_accuracy": 0.09840247258543969,
|
|
"num_tokens": 1758145.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.8502415180206295,
|
|
"epoch": 0.8068582955118507,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5186,
|
|
"mean_token_accuracy": 0.10396859720349312,
|
|
"num_tokens": 1768586.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.890841293334961,
|
|
"epoch": 0.8119011598587998,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5309,
|
|
"mean_token_accuracy": 0.1057729996740818,
|
|
"num_tokens": 1780502.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.796362972259521,
|
|
"epoch": 0.8169440242057489,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5482,
|
|
"mean_token_accuracy": 0.10388528555631638,
|
|
"num_tokens": 1791450.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.87352409362793,
|
|
"epoch": 0.821986888552698,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5693,
|
|
"mean_token_accuracy": 0.10493593215942383,
|
|
"num_tokens": 1803690.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.796584367752075,
|
|
"epoch": 0.827029752899647,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4314,
|
|
"mean_token_accuracy": 0.10723827853798866,
|
|
"num_tokens": 1813630.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.848276948928833,
|
|
"epoch": 0.8320726172465961,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4771,
|
|
"mean_token_accuracy": 0.10020639672875405,
|
|
"num_tokens": 1824537.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.8475054740905765,
|
|
"epoch": 0.8371154815935451,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5398,
|
|
"mean_token_accuracy": 0.10186406672000885,
|
|
"num_tokens": 1834404.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.8914487838745115,
|
|
"epoch": 0.8421583459404942,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5033,
|
|
"mean_token_accuracy": 0.11444203406572342,
|
|
"num_tokens": 1845371.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.808951568603516,
|
|
"epoch": 0.8472012102874432,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5374,
|
|
"mean_token_accuracy": 0.10144101455807686,
|
|
"num_tokens": 1856749.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.753617286682129,
|
|
"epoch": 0.8522440746343923,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5239,
|
|
"mean_token_accuracy": 0.09771231561899185,
|
|
"num_tokens": 1866577.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.879329538345337,
|
|
"epoch": 0.8572869389813415,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4736,
|
|
"mean_token_accuracy": 0.10464069321751594,
|
|
"num_tokens": 1877123.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.824967718124389,
|
|
"epoch": 0.8623298033282905,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5271,
|
|
"mean_token_accuracy": 0.10140521749854088,
|
|
"num_tokens": 1888171.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.704531288146972,
|
|
"epoch": 0.8673726676752396,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4226,
|
|
"mean_token_accuracy": 0.1075032263994217,
|
|
"num_tokens": 1898150.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.856900072097778,
|
|
"epoch": 0.8724155320221886,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5061,
|
|
"mean_token_accuracy": 0.10565795376896858,
|
|
"num_tokens": 1907641.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.737218999862671,
|
|
"epoch": 0.8774583963691377,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4628,
|
|
"mean_token_accuracy": 0.10985867083072662,
|
|
"num_tokens": 1918297.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.803271055221558,
|
|
"epoch": 0.8825012607160867,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4939,
|
|
"mean_token_accuracy": 0.1006560243666172,
|
|
"num_tokens": 1930268.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.7526568412780765,
|
|
"epoch": 0.8875441250630358,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4368,
|
|
"mean_token_accuracy": 0.1081591635942459,
|
|
"num_tokens": 1941762.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.780880975723266,
|
|
"epoch": 0.8925869894099848,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4951,
|
|
"mean_token_accuracy": 0.10922890678048133,
|
|
"num_tokens": 1952416.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.708825016021729,
|
|
"epoch": 0.8976298537569339,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4435,
|
|
"mean_token_accuracy": 0.10499039888381959,
|
|
"num_tokens": 1963451.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.827254772186279,
|
|
"epoch": 0.902672718103883,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.4522,
|
|
"mean_token_accuracy": 0.11010371595621109,
|
|
"num_tokens": 1974360.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.77542781829834,
|
|
"epoch": 0.9077155824508321,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5687,
|
|
"mean_token_accuracy": 0.09718061909079552,
|
|
"num_tokens": 1985823.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.88060450553894,
|
|
"epoch": 0.9127584467977812,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5969,
|
|
"mean_token_accuracy": 0.10037398338317871,
|
|
"num_tokens": 1997475.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.803939294815064,
|
|
"epoch": 0.9178013111447302,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.447,
|
|
"mean_token_accuracy": 0.1048206314444542,
|
|
"num_tokens": 2009029.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.610732984542847,
|
|
"epoch": 0.9228441754916793,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.316,
|
|
"mean_token_accuracy": 0.10846489518880845,
|
|
"num_tokens": 2019564.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.762038087844848,
|
|
"epoch": 0.9278870398386283,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5572,
|
|
"mean_token_accuracy": 0.10404286086559296,
|
|
"num_tokens": 2031394.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.7797784328460695,
|
|
"epoch": 0.9329299041855774,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4583,
|
|
"mean_token_accuracy": 0.09987704530358314,
|
|
"num_tokens": 2042132.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.709231615066528,
|
|
"epoch": 0.9379727685325264,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4284,
|
|
"mean_token_accuracy": 0.10404183194041253,
|
|
"num_tokens": 2054469.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.638057374954224,
|
|
"epoch": 0.9430156328794755,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.10199068039655686,
|
|
"num_tokens": 2065001.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.732149457931518,
|
|
"epoch": 0.9480584972264247,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4278,
|
|
"mean_token_accuracy": 0.10125001892447472,
|
|
"num_tokens": 2075786.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.6592841148376465,
|
|
"epoch": 0.9531013615733737,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.3855,
|
|
"mean_token_accuracy": 0.10808129534125328,
|
|
"num_tokens": 2087818.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.751112985610962,
|
|
"epoch": 0.9581442259203228,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4603,
|
|
"mean_token_accuracy": 0.10283722653985024,
|
|
"num_tokens": 2098771.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.616975784301758,
|
|
"epoch": 0.9631870902672718,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.11221800073981285,
|
|
"num_tokens": 2110036.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.715431022644043,
|
|
"epoch": 0.9682299546142209,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4653,
|
|
"mean_token_accuracy": 0.11255124062299729,
|
|
"num_tokens": 2120411.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.637408781051636,
|
|
"epoch": 0.9732728189611699,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3735,
|
|
"mean_token_accuracy": 0.11180601567029953,
|
|
"num_tokens": 2132831.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.633380985260009,
|
|
"epoch": 0.978315683308119,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.1062696933746338,
|
|
"num_tokens": 2143166.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.632574987411499,
|
|
"epoch": 0.983358547655068,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4127,
|
|
"mean_token_accuracy": 0.10715299546718597,
|
|
"num_tokens": 2154366.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.666649913787841,
|
|
"epoch": 0.9884014120020171,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4129,
|
|
"mean_token_accuracy": 0.10922959744930268,
|
|
"num_tokens": 2164396.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.718012857437134,
|
|
"epoch": 0.9934442763489663,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4886,
|
|
"mean_token_accuracy": 0.10265103131532669,
|
|
"num_tokens": 2177295.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.6520209312438965,
|
|
"epoch": 0.9984871406959153,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3321,
|
|
"mean_token_accuracy": 0.11074534058570862,
|
|
"num_tokens": 2188832.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.64670647515191,
|
|
"epoch": 1.0030257186081695,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2959,
|
|
"mean_token_accuracy": 0.10541310575273302,
|
|
"num_tokens": 2197926.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.533407592773438,
|
|
"epoch": 1.0080685829551186,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3318,
|
|
"mean_token_accuracy": 0.10802690610289574,
|
|
"num_tokens": 2209156.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0080685829551186,
|
|
"eval_entropy": 6.521555552955802,
|
|
"eval_loss": 6.444716930389404,
|
|
"eval_mean_token_accuracy": 0.10976675038081392,
|
|
"eval_num_tokens": 2209156.0,
|
|
"eval_runtime": 2.1191,
|
|
"eval_samples_per_second": 1480.82,
|
|
"eval_steps_per_second": 185.456,
|
|
"step": 1000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9910,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 546050741299200.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|