c33b01fdac
Model: fpadovani/ita-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed455 Source: Original Platform
4057 lines
109 KiB
JSON
4057 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.016137165910237,
|
|
"eval_steps": 1000,
|
|
"global_step": 2000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.5978460788726805,
|
|
"epoch": 1.0131114473020675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999997762223182,
|
|
"loss": 6.2404,
|
|
"mean_token_accuracy": 0.11512459516525268,
|
|
"num_tokens": 2220419.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.775608205795288,
|
|
"epoch": 1.0181543116490166,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999988671262489,
|
|
"loss": 6.3899,
|
|
"mean_token_accuracy": 0.10936411917209625,
|
|
"num_tokens": 2232671.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.581707525253296,
|
|
"epoch": 1.0231971759959657,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999972587285105,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.11535554900765418,
|
|
"num_tokens": 2243367.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.625203037261963,
|
|
"epoch": 1.0282400403429148,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999949510341015,
|
|
"loss": 6.2172,
|
|
"mean_token_accuracy": 0.11806095391511917,
|
|
"num_tokens": 2254120.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.559353828430176,
|
|
"epoch": 1.033282904689864,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999919440501947,
|
|
"loss": 6.2589,
|
|
"mean_token_accuracy": 0.11413354724645615,
|
|
"num_tokens": 2265430.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.629713344573974,
|
|
"epoch": 1.0383257690368128,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999882377861355,
|
|
"loss": 6.2949,
|
|
"mean_token_accuracy": 0.11334428489208222,
|
|
"num_tokens": 2277031.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.503676271438598,
|
|
"epoch": 1.043368633383762,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999838322534433,
|
|
"loss": 6.1731,
|
|
"mean_token_accuracy": 0.11683624684810638,
|
|
"num_tokens": 2287430.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.539786052703858,
|
|
"epoch": 1.048411497730711,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999787274658107,
|
|
"loss": 6.2532,
|
|
"mean_token_accuracy": 0.11774518638849259,
|
|
"num_tokens": 2298063.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.622480630874634,
|
|
"epoch": 1.0534543620776602,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999729234391032,
|
|
"loss": 6.2971,
|
|
"mean_token_accuracy": 0.1107066549360752,
|
|
"num_tokens": 2308684.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.555305910110474,
|
|
"epoch": 1.058497226424609,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00049996642019136,
|
|
"loss": 6.2187,
|
|
"mean_token_accuracy": 0.11692636534571647,
|
|
"num_tokens": 2319611.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.465685415267944,
|
|
"epoch": 1.0635400907715582,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999592177427933,
|
|
"loss": 6.2071,
|
|
"mean_token_accuracy": 0.11449091359972954,
|
|
"num_tokens": 2330655.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.55456166267395,
|
|
"epoch": 1.0685829551185073,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999513161157885,
|
|
"loss": 6.2084,
|
|
"mean_token_accuracy": 0.11772598922252656,
|
|
"num_tokens": 2341717.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.517464208602905,
|
|
"epoch": 1.0736258194654564,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999427153349041,
|
|
"loss": 6.2967,
|
|
"mean_token_accuracy": 0.11366510167717933,
|
|
"num_tokens": 2352614.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.499022531509399,
|
|
"epoch": 1.0786686838124055,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999334154268714,
|
|
"loss": 6.1994,
|
|
"mean_token_accuracy": 0.11385042443871499,
|
|
"num_tokens": 2362651.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.607080030441284,
|
|
"epoch": 1.0837115481593544,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999234164205949,
|
|
"loss": 6.1907,
|
|
"mean_token_accuracy": 0.11571867242455483,
|
|
"num_tokens": 2375518.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.455441665649414,
|
|
"epoch": 1.0887544125063036,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999127183471517,
|
|
"loss": 6.2486,
|
|
"mean_token_accuracy": 0.10975311547517777,
|
|
"num_tokens": 2386908.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.411809968948364,
|
|
"epoch": 1.0937972768532527,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999013212397915,
|
|
"loss": 6.0991,
|
|
"mean_token_accuracy": 0.12264346778392791,
|
|
"num_tokens": 2398992.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.534028816223144,
|
|
"epoch": 1.0988401412002018,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998892251339368,
|
|
"loss": 6.3038,
|
|
"mean_token_accuracy": 0.1096903495490551,
|
|
"num_tokens": 2410741.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.562671566009522,
|
|
"epoch": 1.1038830055471507,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998764300671827,
|
|
"loss": 6.2484,
|
|
"mean_token_accuracy": 0.10970004200935364,
|
|
"num_tokens": 2421702.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.49102144241333,
|
|
"epoch": 1.1089258698940998,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998629360792965,
|
|
"loss": 6.2651,
|
|
"mean_token_accuracy": 0.11268085315823555,
|
|
"num_tokens": 2433676.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.491061782836914,
|
|
"epoch": 1.113968734241049,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998487432122179,
|
|
"loss": 6.2049,
|
|
"mean_token_accuracy": 0.12022922188043594,
|
|
"num_tokens": 2445415.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.445030593872071,
|
|
"epoch": 1.119011598587998,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998338515100585,
|
|
"loss": 6.1118,
|
|
"mean_token_accuracy": 0.12043405771255493,
|
|
"num_tokens": 2456195.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.435121774673462,
|
|
"epoch": 1.1240544629349472,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998182610191022,
|
|
"loss": 6.1812,
|
|
"mean_token_accuracy": 0.11498187631368637,
|
|
"num_tokens": 2469448.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.392489862442017,
|
|
"epoch": 1.129097327281896,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998019717878047,
|
|
"loss": 6.1917,
|
|
"mean_token_accuracy": 0.1175561472773552,
|
|
"num_tokens": 2480532.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.50950813293457,
|
|
"epoch": 1.1341401916288452,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499784983866793,
|
|
"loss": 6.1337,
|
|
"mean_token_accuracy": 0.11248333379626274,
|
|
"num_tokens": 2492203.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.380187177658081,
|
|
"epoch": 1.1391830559757943,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997672973088661,
|
|
"loss": 6.1358,
|
|
"mean_token_accuracy": 0.1200421243906021,
|
|
"num_tokens": 2502503.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.490655899047852,
|
|
"epoch": 1.1442259203227434,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997489121689944,
|
|
"loss": 6.1909,
|
|
"mean_token_accuracy": 0.11370846405625343,
|
|
"num_tokens": 2513981.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.496898746490478,
|
|
"epoch": 1.1492687846696923,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997298285043189,
|
|
"loss": 6.217,
|
|
"mean_token_accuracy": 0.11847558692097664,
|
|
"num_tokens": 2524572.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.521786737442016,
|
|
"epoch": 1.1543116490166414,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997100463741527,
|
|
"loss": 6.1756,
|
|
"mean_token_accuracy": 0.11777064353227615,
|
|
"num_tokens": 2534159.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.408275556564331,
|
|
"epoch": 1.1593545133635905,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996895658399785,
|
|
"loss": 6.1456,
|
|
"mean_token_accuracy": 0.11982325688004494,
|
|
"num_tokens": 2545486.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.426592206954956,
|
|
"epoch": 1.1643973777105396,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996683869654507,
|
|
"loss": 6.169,
|
|
"mean_token_accuracy": 0.12543639093637465,
|
|
"num_tokens": 2556288.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.44536771774292,
|
|
"epoch": 1.1694402420574885,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996465098163934,
|
|
"loss": 6.1312,
|
|
"mean_token_accuracy": 0.12022658362984658,
|
|
"num_tokens": 2567241.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.346808815002442,
|
|
"epoch": 1.1744831064044376,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996239344608015,
|
|
"loss": 6.164,
|
|
"mean_token_accuracy": 0.11732971742749214,
|
|
"num_tokens": 2577697.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.421145391464234,
|
|
"epoch": 1.1795259707513868,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996006609688398,
|
|
"loss": 6.1354,
|
|
"mean_token_accuracy": 0.12507255673408507,
|
|
"num_tokens": 2588231.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.482469606399536,
|
|
"epoch": 1.1845688350983359,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995766894128426,
|
|
"loss": 6.1123,
|
|
"mean_token_accuracy": 0.12405804842710495,
|
|
"num_tokens": 2598750.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.389590549468994,
|
|
"epoch": 1.189611699445285,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995520198673143,
|
|
"loss": 6.169,
|
|
"mean_token_accuracy": 0.11749346479773522,
|
|
"num_tokens": 2609481.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.441128349304199,
|
|
"epoch": 1.194654563792234,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995266524089283,
|
|
"loss": 6.1752,
|
|
"mean_token_accuracy": 0.11522530987858773,
|
|
"num_tokens": 2619746.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.325751733779907,
|
|
"epoch": 1.199697428139183,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995005871165271,
|
|
"loss": 6.031,
|
|
"mean_token_accuracy": 0.12297471836209298,
|
|
"num_tokens": 2629578.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.383418941497803,
|
|
"epoch": 1.2047402924861321,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004994738240711225,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.1259584166109562,
|
|
"num_tokens": 2639063.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.420254707336426,
|
|
"epoch": 1.2097831568330812,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004994463633558945,
|
|
"loss": 6.198,
|
|
"mean_token_accuracy": 0.11809922605752946,
|
|
"num_tokens": 2649304.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.4016345024108885,
|
|
"epoch": 1.2148260211800301,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004994182050561917,
|
|
"loss": 6.2067,
|
|
"mean_token_accuracy": 0.11969330683350562,
|
|
"num_tokens": 2659917.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.510196495056152,
|
|
"epoch": 1.2198688855269793,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004993893492595306,
|
|
"loss": 6.2814,
|
|
"mean_token_accuracy": 0.1167818121612072,
|
|
"num_tokens": 2672136.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.418848848342895,
|
|
"epoch": 1.2249117498739284,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004993597960555959,
|
|
"loss": 6.1967,
|
|
"mean_token_accuracy": 0.11458191946148873,
|
|
"num_tokens": 2683656.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.332972669601441,
|
|
"epoch": 1.2299546142208775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993295455362395,
|
|
"loss": 6.06,
|
|
"mean_token_accuracy": 0.12417502328753471,
|
|
"num_tokens": 2693289.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.447884559631348,
|
|
"epoch": 1.2349974785678266,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004992985977954808,
|
|
"loss": 6.1174,
|
|
"mean_token_accuracy": 0.11916331723332405,
|
|
"num_tokens": 2703442.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.469077157974243,
|
|
"epoch": 1.2400403429147757,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499266952929506,
|
|
"loss": 6.1648,
|
|
"mean_token_accuracy": 0.11862540766596794,
|
|
"num_tokens": 2715229.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.3187275409698485,
|
|
"epoch": 1.2450832072617246,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004992346110366682,
|
|
"loss": 6.107,
|
|
"mean_token_accuracy": 0.11733711212873459,
|
|
"num_tokens": 2725075.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.427688646316528,
|
|
"epoch": 1.2501260716086737,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004992015722174865,
|
|
"loss": 6.1036,
|
|
"mean_token_accuracy": 0.11654637530446052,
|
|
"num_tokens": 2735138.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.247444581985474,
|
|
"epoch": 1.2551689359556228,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004991678365746466,
|
|
"loss": 6.0342,
|
|
"mean_token_accuracy": 0.12928969264030457,
|
|
"num_tokens": 2745824.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.4426844120025635,
|
|
"epoch": 1.2602118003025717,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991334042129996,
|
|
"loss": 6.1156,
|
|
"mean_token_accuracy": 0.1282254844903946,
|
|
"num_tokens": 2756194.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.358617973327637,
|
|
"epoch": 1.2652546646495209,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004990982752395618,
|
|
"loss": 6.2748,
|
|
"mean_token_accuracy": 0.12170547097921372,
|
|
"num_tokens": 2766811.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.469353485107422,
|
|
"epoch": 1.27029752899647,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004990624497635152,
|
|
"loss": 6.0757,
|
|
"mean_token_accuracy": 0.12089031934738159,
|
|
"num_tokens": 2777079.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.278725862503052,
|
|
"epoch": 1.275340393343419,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000499025927896206,
|
|
"loss": 6.122,
|
|
"mean_token_accuracy": 0.12426704019308091,
|
|
"num_tokens": 2788789.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.437152147293091,
|
|
"epoch": 1.2803832576903682,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004989887097511451,
|
|
"loss": 6.1902,
|
|
"mean_token_accuracy": 0.11868843510746956,
|
|
"num_tokens": 2801587.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.451790285110474,
|
|
"epoch": 1.2854261220373173,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004989507954440074,
|
|
"loss": 6.1,
|
|
"mean_token_accuracy": 0.11965932920575142,
|
|
"num_tokens": 2812027.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.1532281875610355,
|
|
"epoch": 1.2904689863842662,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989121850926313,
|
|
"loss": 6.1071,
|
|
"mean_token_accuracy": 0.11968399733304977,
|
|
"num_tokens": 2823791.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.457482099533081,
|
|
"epoch": 1.2955118507312153,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004988728788170188,
|
|
"loss": 6.1601,
|
|
"mean_token_accuracy": 0.12196319550275803,
|
|
"num_tokens": 2835015.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.410132265090942,
|
|
"epoch": 1.3005547150781644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004988328767393346,
|
|
"loss": 6.1115,
|
|
"mean_token_accuracy": 0.12226968184113503,
|
|
"num_tokens": 2846078.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.270524978637695,
|
|
"epoch": 1.3055975794251133,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004987921789839061,
|
|
"loss": 6.0785,
|
|
"mean_token_accuracy": 0.1250629886984825,
|
|
"num_tokens": 2856168.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.428250122070312,
|
|
"epoch": 1.3106404437720625,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004987507856772229,
|
|
"loss": 6.1623,
|
|
"mean_token_accuracy": 0.12155707478523255,
|
|
"num_tokens": 2868258.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.292591142654419,
|
|
"epoch": 1.3156833081190116,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004987086969479364,
|
|
"loss": 6.0768,
|
|
"mean_token_accuracy": 0.11958595067262649,
|
|
"num_tokens": 2879202.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.267068004608154,
|
|
"epoch": 1.3207261724659607,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004986659129268592,
|
|
"loss": 6.0257,
|
|
"mean_token_accuracy": 0.12108759433031083,
|
|
"num_tokens": 2890669.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.301166677474976,
|
|
"epoch": 1.3257690368129098,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004986224337469652,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.12250669449567794,
|
|
"num_tokens": 2900799.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.271768856048584,
|
|
"epoch": 1.330811901159859,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004985782595433886,
|
|
"loss": 6.1363,
|
|
"mean_token_accuracy": 0.1203561469912529,
|
|
"num_tokens": 2913166.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.377553939819336,
|
|
"epoch": 1.3358547655068078,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 6.125,
|
|
"mean_token_accuracy": 0.1276712730526924,
|
|
"num_tokens": 2924750.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.341912269592285,
|
|
"epoch": 1.340897629853757,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000498487826616525,
|
|
"loss": 6.0692,
|
|
"mean_token_accuracy": 0.12529361471533776,
|
|
"num_tokens": 2935343.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.308580684661865,
|
|
"epoch": 1.345940494200706,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984415681743056,
|
|
"loss": 6.0622,
|
|
"mean_token_accuracy": 0.1283659666776657,
|
|
"num_tokens": 2945420.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.413863706588745,
|
|
"epoch": 1.350983358547655,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004983946152705382,
|
|
"loss": 6.177,
|
|
"mean_token_accuracy": 0.12130876779556274,
|
|
"num_tokens": 2955580.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.38455638885498,
|
|
"epoch": 1.356026222894604,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004983469680511531,
|
|
"loss": 6.0229,
|
|
"mean_token_accuracy": 0.13240867108106613,
|
|
"num_tokens": 2966192.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.2409889698028564,
|
|
"epoch": 1.3610690872415532,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004982986266642393,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.12243769392371177,
|
|
"num_tokens": 2976715.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.392279195785522,
|
|
"epoch": 1.3661119515885023,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004982495912600426,
|
|
"loss": 6.1688,
|
|
"mean_token_accuracy": 0.11467552334070205,
|
|
"num_tokens": 2987763.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.220443630218506,
|
|
"epoch": 1.3711548159354514,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004981998619909664,
|
|
"loss": 5.9627,
|
|
"mean_token_accuracy": 0.1257205128669739,
|
|
"num_tokens": 2998518.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.389316844940185,
|
|
"epoch": 1.3761976802824005,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004981494390115704,
|
|
"loss": 6.1676,
|
|
"mean_token_accuracy": 0.12993667274713516,
|
|
"num_tokens": 3008953.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.2395531177520756,
|
|
"epoch": 1.3812405446293494,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004980983224785704,
|
|
"loss": 6.0096,
|
|
"mean_token_accuracy": 0.1318300373852253,
|
|
"num_tokens": 3018763.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.279836845397949,
|
|
"epoch": 1.3862834089762985,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004980465125508376,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.12234401479363441,
|
|
"num_tokens": 3030061.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.348916482925415,
|
|
"epoch": 1.3913262733232477,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004979940093893986,
|
|
"loss": 6.0695,
|
|
"mean_token_accuracy": 0.12201000452041626,
|
|
"num_tokens": 3040646.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.368468952178955,
|
|
"epoch": 1.3963691376701965,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004979408131574345,
|
|
"loss": 6.1544,
|
|
"mean_token_accuracy": 0.11946613788604736,
|
|
"num_tokens": 3051470.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.276396369934082,
|
|
"epoch": 1.4014120020171457,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004978869240202802,
|
|
"loss": 6.1367,
|
|
"mean_token_accuracy": 0.1224789097905159,
|
|
"num_tokens": 3063539.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.297502517700195,
|
|
"epoch": 1.4064548663640948,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004978323421454246,
|
|
"loss": 5.9703,
|
|
"mean_token_accuracy": 0.13029929473996163,
|
|
"num_tokens": 3073945.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.269286251068115,
|
|
"epoch": 1.411497730711044,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004977770677025092,
|
|
"loss": 6.0557,
|
|
"mean_token_accuracy": 0.125611362606287,
|
|
"num_tokens": 3085276.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.317072153091431,
|
|
"epoch": 1.416540595057993,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977211008633284,
|
|
"loss": 6.0779,
|
|
"mean_token_accuracy": 0.12258708700537682,
|
|
"num_tokens": 3097599.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.210897493362427,
|
|
"epoch": 1.4215834594049421,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004976644418018285,
|
|
"loss": 6.0269,
|
|
"mean_token_accuracy": 0.13576313257217407,
|
|
"num_tokens": 3109536.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.3146326541900635,
|
|
"epoch": 1.426626323751891,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000497607090694107,
|
|
"loss": 5.994,
|
|
"mean_token_accuracy": 0.1293320521712303,
|
|
"num_tokens": 3119178.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.206587886810302,
|
|
"epoch": 1.4316691880988401,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004975490477184126,
|
|
"loss": 6.0707,
|
|
"mean_token_accuracy": 0.1234750546514988,
|
|
"num_tokens": 3131183.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.338142347335816,
|
|
"epoch": 1.4367120524457893,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497490313055144,
|
|
"loss": 6.0437,
|
|
"mean_token_accuracy": 0.12467821165919304,
|
|
"num_tokens": 3142038.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.280117177963257,
|
|
"epoch": 1.4417549167927382,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974308868868501,
|
|
"loss": 6.0538,
|
|
"mean_token_accuracy": 0.128610360622406,
|
|
"num_tokens": 3152878.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.2970295429229735,
|
|
"epoch": 1.4467977811396873,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004973707693982287,
|
|
"loss": 6.025,
|
|
"mean_token_accuracy": 0.12771496102213858,
|
|
"num_tokens": 3165485.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.257808780670166,
|
|
"epoch": 1.4518406454866364,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004973099607761264,
|
|
"loss": 6.0463,
|
|
"mean_token_accuracy": 0.12606460675597192,
|
|
"num_tokens": 3175134.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.246237754821777,
|
|
"epoch": 1.4568835098335855,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004972484612095378,
|
|
"loss": 6.0275,
|
|
"mean_token_accuracy": 0.12163080051541328,
|
|
"num_tokens": 3185319.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.289009523391724,
|
|
"epoch": 1.4619263741805346,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497186270889605,
|
|
"loss": 6.087,
|
|
"mean_token_accuracy": 0.12769344821572304,
|
|
"num_tokens": 3196427.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.200272750854492,
|
|
"epoch": 1.4669692385274837,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004971233900096169,
|
|
"loss": 5.9995,
|
|
"mean_token_accuracy": 0.12756021395325662,
|
|
"num_tokens": 3207233.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.335526371002198,
|
|
"epoch": 1.4720121028744326,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004970598187650085,
|
|
"loss": 6.0253,
|
|
"mean_token_accuracy": 0.12155926004052162,
|
|
"num_tokens": 3217999.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.204497385025024,
|
|
"epoch": 1.4770549672213817,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000496995557353361,
|
|
"loss": 6.0792,
|
|
"mean_token_accuracy": 0.11796228364109992,
|
|
"num_tokens": 3228965.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.2911436557769775,
|
|
"epoch": 1.4820978315683309,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004969306059744004,
|
|
"loss": 6.0286,
|
|
"mean_token_accuracy": 0.12896939292550086,
|
|
"num_tokens": 3239960.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.252200365066528,
|
|
"epoch": 1.4871406959152798,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004968649648299968,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.12258177921175957,
|
|
"num_tokens": 3249250.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.185096216201782,
|
|
"epoch": 1.4921835602622289,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004967986341241646,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.12548948973417282,
|
|
"num_tokens": 3260936.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.2398033618927,
|
|
"epoch": 1.497226424609178,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004967316140630611,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.1348661631345749,
|
|
"num_tokens": 3271571.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.1721444606781,
|
|
"epoch": 1.502269288956127,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004966639048549862,
|
|
"loss": 6.0033,
|
|
"mean_token_accuracy": 0.12591522112488746,
|
|
"num_tokens": 3282216.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.171544027328491,
|
|
"epoch": 1.5073121533030762,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004965955067103816,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.1273512251675129,
|
|
"num_tokens": 3293791.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.191919231414795,
|
|
"epoch": 1.5123550176500253,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004965264198418305,
|
|
"loss": 5.9528,
|
|
"mean_token_accuracy": 0.13013582155108452,
|
|
"num_tokens": 3304260.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.202256298065185,
|
|
"epoch": 1.5173978819969742,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004964566444640564,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.1315968282520771,
|
|
"num_tokens": 3315882.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.294244718551636,
|
|
"epoch": 1.5224407463439233,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004963861807939227,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.12783424481749533,
|
|
"num_tokens": 3326649.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.161068487167358,
|
|
"epoch": 1.5274836106908725,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004963150290504321,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.12348355501890182,
|
|
"num_tokens": 3338670.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.307290506362915,
|
|
"epoch": 1.5325264750378214,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004962431894547259,
|
|
"loss": 6.0966,
|
|
"mean_token_accuracy": 0.12335565835237502,
|
|
"num_tokens": 3350360.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.216002511978149,
|
|
"epoch": 1.5375693393847705,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004961706622300834,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.14111101776361465,
|
|
"num_tokens": 3360647.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.0693176746368405,
|
|
"epoch": 1.5426122037317196,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004960974476019206,
|
|
"loss": 6.0163,
|
|
"mean_token_accuracy": 0.1311452269554138,
|
|
"num_tokens": 3371153.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.3018128871917725,
|
|
"epoch": 1.5476550680786687,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004960235457977905,
|
|
"loss": 6.0078,
|
|
"mean_token_accuracy": 0.1318977639079094,
|
|
"num_tokens": 3382027.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.211627340316772,
|
|
"epoch": 1.5526979324256178,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004959489570473814,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.13007684871554376,
|
|
"num_tokens": 3393122.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.161399507522583,
|
|
"epoch": 1.557740796772567,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000495873681582517,
|
|
"loss": 5.9511,
|
|
"mean_token_accuracy": 0.12621625512838364,
|
|
"num_tokens": 3403992.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.325139665603638,
|
|
"epoch": 1.5627836611195158,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004957977196371552,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.1289366900920868,
|
|
"num_tokens": 3415375.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.303205871582032,
|
|
"epoch": 1.567826525466465,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004957210714473873,
|
|
"loss": 6.0724,
|
|
"mean_token_accuracy": 0.12386729642748832,
|
|
"num_tokens": 3426670.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.233889436721801,
|
|
"epoch": 1.572869389813414,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004956437372514378,
|
|
"loss": 6.0126,
|
|
"mean_token_accuracy": 0.1264229014515877,
|
|
"num_tokens": 3438262.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.208000659942627,
|
|
"epoch": 1.577912254160363,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004955657172896631,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.12889024913311004,
|
|
"num_tokens": 3451442.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.176012802124023,
|
|
"epoch": 1.582955118507312,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004954870118045511,
|
|
"loss": 5.9453,
|
|
"mean_token_accuracy": 0.13044202998280524,
|
|
"num_tokens": 3462022.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.182885599136353,
|
|
"epoch": 1.5879979828542612,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004954076210407203,
|
|
"loss": 5.926,
|
|
"mean_token_accuracy": 0.12430310323834419,
|
|
"num_tokens": 3472777.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.167458534240723,
|
|
"epoch": 1.5930408472012103,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004953275452449188,
|
|
"loss": 5.9126,
|
|
"mean_token_accuracy": 0.13071943670511246,
|
|
"num_tokens": 3482946.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.124200487136841,
|
|
"epoch": 1.5980837115481594,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004952467846660243,
|
|
"loss": 5.9521,
|
|
"mean_token_accuracy": 0.12621726244688034,
|
|
"num_tokens": 3493930.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.248808479309082,
|
|
"epoch": 1.6031265758951085,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004951653395550423,
|
|
"loss": 6.0274,
|
|
"mean_token_accuracy": 0.13040645346045493,
|
|
"num_tokens": 3506139.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.14754204750061,
|
|
"epoch": 1.6081694402420574,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004950832101651062,
|
|
"loss": 5.8785,
|
|
"mean_token_accuracy": 0.13420721665024757,
|
|
"num_tokens": 3516844.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.1516115188598635,
|
|
"epoch": 1.6132123045890066,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004950003967514762,
|
|
"loss": 5.9405,
|
|
"mean_token_accuracy": 0.12669654116034507,
|
|
"num_tokens": 3526815.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.122528409957885,
|
|
"epoch": 1.6182551689359557,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000494916899571538,
|
|
"loss": 5.9078,
|
|
"mean_token_accuracy": 0.13329293578863144,
|
|
"num_tokens": 3538041.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.184952688217163,
|
|
"epoch": 1.6232980332829046,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004948327188848029,
|
|
"loss": 5.9254,
|
|
"mean_token_accuracy": 0.13670434802770615,
|
|
"num_tokens": 3549359.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.08763689994812,
|
|
"epoch": 1.6283408976298537,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004947478549529065,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.13631670400500298,
|
|
"num_tokens": 3561902.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.208184385299683,
|
|
"epoch": 1.6333837619768028,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004946623080396077,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.1355512760579586,
|
|
"num_tokens": 3572584.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.185911369323731,
|
|
"epoch": 1.638426626323752,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004945760784107884,
|
|
"loss": 5.9123,
|
|
"mean_token_accuracy": 0.1353035695850849,
|
|
"num_tokens": 3582635.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.138697481155395,
|
|
"epoch": 1.643469490670701,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004944891663344522,
|
|
"loss": 5.9255,
|
|
"mean_token_accuracy": 0.13605186343193054,
|
|
"num_tokens": 3594140.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.07603087425232,
|
|
"epoch": 1.6485123550176501,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004944015720807239,
|
|
"loss": 5.8809,
|
|
"mean_token_accuracy": 0.13401841074228288,
|
|
"num_tokens": 3605748.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.134157705307007,
|
|
"epoch": 1.653555219364599,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004943132959218484,
|
|
"loss": 6.0377,
|
|
"mean_token_accuracy": 0.1261787310242653,
|
|
"num_tokens": 3618304.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.240805006027221,
|
|
"epoch": 1.6585980837115482,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00049422433813219,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.13287644907832147,
|
|
"num_tokens": 3630708.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.146529674530029,
|
|
"epoch": 1.6636409480584973,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004941346989882318,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.12477361038327217,
|
|
"num_tokens": 3643188.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.0920877933502195,
|
|
"epoch": 1.6686838124054462,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004940443787685739,
|
|
"loss": 5.8116,
|
|
"mean_token_accuracy": 0.1345119781792164,
|
|
"num_tokens": 3653491.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.0734718322753904,
|
|
"epoch": 1.6737266767523953,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004939533777539341,
|
|
"loss": 5.8212,
|
|
"mean_token_accuracy": 0.13908676952123641,
|
|
"num_tokens": 3663695.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.225847434997559,
|
|
"epoch": 1.6787695410993444,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004938616962271453,
|
|
"loss": 6.0323,
|
|
"mean_token_accuracy": 0.13221884593367578,
|
|
"num_tokens": 3674093.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.098683214187622,
|
|
"epoch": 1.6838124054462935,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004937693344731561,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.13245968520641327,
|
|
"num_tokens": 3685362.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.2160602569580075,
|
|
"epoch": 1.6888552697932426,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004936762927790286,
|
|
"loss": 5.9689,
|
|
"mean_token_accuracy": 0.13073147758841513,
|
|
"num_tokens": 3695963.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.16051287651062,
|
|
"epoch": 1.6938981341401917,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004935825714339389,
|
|
"loss": 5.929,
|
|
"mean_token_accuracy": 0.12861507833004,
|
|
"num_tokens": 3706805.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.203017616271973,
|
|
"epoch": 1.6989409984871406,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004934881707291748,
|
|
"loss": 6.0305,
|
|
"mean_token_accuracy": 0.12621263936161994,
|
|
"num_tokens": 3718822.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.157417583465576,
|
|
"epoch": 1.7039838628340898,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004933930909581362,
|
|
"loss": 5.972,
|
|
"mean_token_accuracy": 0.13590837195515632,
|
|
"num_tokens": 3730170.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.205537796020508,
|
|
"epoch": 1.7090267271810389,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004932973324163328,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.12582333013415337,
|
|
"num_tokens": 3741293.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.160839653015136,
|
|
"epoch": 1.7140695915279878,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004932008954013846,
|
|
"loss": 5.9679,
|
|
"mean_token_accuracy": 0.12444445788860321,
|
|
"num_tokens": 3753521.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.1997740268707275,
|
|
"epoch": 1.7191124558749369,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004931037802130201,
|
|
"loss": 5.963,
|
|
"mean_token_accuracy": 0.13262831941246986,
|
|
"num_tokens": 3765767.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.179419755935669,
|
|
"epoch": 1.724155320221886,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004930059871530755,
|
|
"loss": 5.9409,
|
|
"mean_token_accuracy": 0.129554682970047,
|
|
"num_tokens": 3776330.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.163461828231812,
|
|
"epoch": 1.7291981845688351,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004929075165254938,
|
|
"loss": 5.9281,
|
|
"mean_token_accuracy": 0.13296380937099456,
|
|
"num_tokens": 3786937.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.106232500076294,
|
|
"epoch": 1.7342410489157842,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004928083686363241,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.12930018305778504,
|
|
"num_tokens": 3798546.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.1347733497619625,
|
|
"epoch": 1.7392839132627333,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004927085437937204,
|
|
"loss": 5.8981,
|
|
"mean_token_accuracy": 0.13048461303114892,
|
|
"num_tokens": 3809893.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.0981621742248535,
|
|
"epoch": 1.7443267776096822,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 5.9091,
|
|
"mean_token_accuracy": 0.13413122594356536,
|
|
"num_tokens": 3820570.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.125863027572632,
|
|
"epoch": 1.7493696419566314,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000492506864491345,
|
|
"loss": 5.9699,
|
|
"mean_token_accuracy": 0.12728568613529206,
|
|
"num_tokens": 3833048.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.212840175628662,
|
|
"epoch": 1.7544125063035805,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004924050106583976,
|
|
"loss": 5.9172,
|
|
"mean_token_accuracy": 0.12656415328383447,
|
|
"num_tokens": 3844375.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.069183397293091,
|
|
"epoch": 1.7594553706505294,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004923024811256618,
|
|
"loss": 5.8212,
|
|
"mean_token_accuracy": 0.14076320379972457,
|
|
"num_tokens": 3855103.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.154578733444214,
|
|
"epoch": 1.7644982349974785,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000492199276211802,
|
|
"loss": 5.8599,
|
|
"mean_token_accuracy": 0.1319786749780178,
|
|
"num_tokens": 3865303.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.077178335189819,
|
|
"epoch": 1.7695410993444276,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004920953962375813,
|
|
"loss": 5.9282,
|
|
"mean_token_accuracy": 0.13082144781947136,
|
|
"num_tokens": 3877344.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.11976375579834,
|
|
"epoch": 1.7745839636913767,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004919908415258612,
|
|
"loss": 5.8774,
|
|
"mean_token_accuracy": 0.13460170179605485,
|
|
"num_tokens": 3887471.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.022991371154785,
|
|
"epoch": 1.7796268280383258,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004918856124016002,
|
|
"loss": 5.8169,
|
|
"mean_token_accuracy": 0.14026920720934868,
|
|
"num_tokens": 3898486.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.090184545516967,
|
|
"epoch": 1.784669692385275,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004917797091918526,
|
|
"loss": 5.929,
|
|
"mean_token_accuracy": 0.1333512157201767,
|
|
"num_tokens": 3909864.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.166698837280274,
|
|
"epoch": 1.789712556732224,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004916731322257683,
|
|
"loss": 6.0132,
|
|
"mean_token_accuracy": 0.12205543667078018,
|
|
"num_tokens": 3921308.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.197019815444946,
|
|
"epoch": 1.794755421079173,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004915658818345908,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.13583694100379945,
|
|
"num_tokens": 3931469.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.988247108459473,
|
|
"epoch": 1.799798285426122,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004914579583516568,
|
|
"loss": 5.9568,
|
|
"mean_token_accuracy": 0.12513967156410216,
|
|
"num_tokens": 3943348.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.2157975196838375,
|
|
"epoch": 1.804841149773071,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000491349362112395,
|
|
"loss": 5.8597,
|
|
"mean_token_accuracy": 0.13096435815095903,
|
|
"num_tokens": 3954491.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.002476024627685,
|
|
"epoch": 1.80988401412002,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000491240093454325,
|
|
"loss": 5.7933,
|
|
"mean_token_accuracy": 0.13584209233522415,
|
|
"num_tokens": 3965151.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.162328100204467,
|
|
"epoch": 1.8149268784669692,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004911301527170563,
|
|
"loss": 5.9067,
|
|
"mean_token_accuracy": 0.1303732991218567,
|
|
"num_tokens": 3975482.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.0661835193634035,
|
|
"epoch": 1.8199697428139183,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004910195402422873,
|
|
"loss": 5.8759,
|
|
"mean_token_accuracy": 0.13615478053689004,
|
|
"num_tokens": 3985295.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.116996097564697,
|
|
"epoch": 1.8250126071608674,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004909082563738042,
|
|
"loss": 5.9292,
|
|
"mean_token_accuracy": 0.12562229484319687,
|
|
"num_tokens": 3997058.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.0792460441589355,
|
|
"epoch": 1.8300554715078166,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004907963014574797,
|
|
"loss": 5.8711,
|
|
"mean_token_accuracy": 0.1312556892633438,
|
|
"num_tokens": 4008224.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.027435636520385,
|
|
"epoch": 1.8350983358547657,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004906836758412722,
|
|
"loss": 5.8081,
|
|
"mean_token_accuracy": 0.13197171092033386,
|
|
"num_tokens": 4019264.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.102999448776245,
|
|
"epoch": 1.8401412002017146,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004905703798752248,
|
|
"loss": 5.8579,
|
|
"mean_token_accuracy": 0.13721298277378083,
|
|
"num_tokens": 4031974.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.052859258651734,
|
|
"epoch": 1.8451840645486637,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000490456413911464,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.13526368364691735,
|
|
"num_tokens": 4042023.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.177751207351685,
|
|
"epoch": 1.8502269288956126,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004903417783041987,
|
|
"loss": 5.9103,
|
|
"mean_token_accuracy": 0.1314178705215454,
|
|
"num_tokens": 4053488.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.049738836288452,
|
|
"epoch": 1.8552697932425617,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004902264734097188,
|
|
"loss": 5.754,
|
|
"mean_token_accuracy": 0.1436818316578865,
|
|
"num_tokens": 4063576.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.0148999214172365,
|
|
"epoch": 1.8603126575895108,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004901104995863948,
|
|
"loss": 5.819,
|
|
"mean_token_accuracy": 0.1303223468363285,
|
|
"num_tokens": 4073767.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.132151794433594,
|
|
"epoch": 1.86535552193646,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004899938571946759,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.13405972570180893,
|
|
"num_tokens": 4086112.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.055962467193604,
|
|
"epoch": 1.870398386283409,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004898765465970894,
|
|
"loss": 5.81,
|
|
"mean_token_accuracy": 0.1344530016183853,
|
|
"num_tokens": 4097172.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.022813367843628,
|
|
"epoch": 1.8754412506303582,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004897585681582393,
|
|
"loss": 5.8513,
|
|
"mean_token_accuracy": 0.13490026369690894,
|
|
"num_tokens": 4108034.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.124302816390991,
|
|
"epoch": 1.8804841149773073,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004896399222448053,
|
|
"loss": 5.9894,
|
|
"mean_token_accuracy": 0.12775079905986786,
|
|
"num_tokens": 4118787.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.162458753585815,
|
|
"epoch": 1.8855269793242562,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004895206092255417,
|
|
"loss": 5.9185,
|
|
"mean_token_accuracy": 0.13882039487361908,
|
|
"num_tokens": 4129427.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.992573070526123,
|
|
"epoch": 1.8905698436712053,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004894006294712761,
|
|
"loss": 5.799,
|
|
"mean_token_accuracy": 0.1407463550567627,
|
|
"num_tokens": 4139750.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.129534530639648,
|
|
"epoch": 1.8956127080181542,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004892799833549083,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.1315428465604782,
|
|
"num_tokens": 4150966.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.042229413986206,
|
|
"epoch": 1.9006555723651033,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004891586712514091,
|
|
"loss": 5.8827,
|
|
"mean_token_accuracy": 0.13546333983540534,
|
|
"num_tokens": 4162002.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.12616753578186,
|
|
"epoch": 1.9056984367120524,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004890366935378196,
|
|
"loss": 5.9401,
|
|
"mean_token_accuracy": 0.1307598277926445,
|
|
"num_tokens": 4173529.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.042671060562133,
|
|
"epoch": 1.9107413010590015,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004889140505932492,
|
|
"loss": 5.9541,
|
|
"mean_token_accuracy": 0.1275738313794136,
|
|
"num_tokens": 4184743.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.098602342605591,
|
|
"epoch": 1.9157841654059506,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004887907427988749,
|
|
"loss": 5.8032,
|
|
"mean_token_accuracy": 0.13772272765636445,
|
|
"num_tokens": 4195484.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.0307262420654295,
|
|
"epoch": 1.9208270297528998,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004886667705379403,
|
|
"loss": 5.8928,
|
|
"mean_token_accuracy": 0.13492754027247428,
|
|
"num_tokens": 4207564.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.070850229263305,
|
|
"epoch": 1.9258698940998489,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004885421341957541,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.13223434910178183,
|
|
"num_tokens": 4219165.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.033602094650268,
|
|
"epoch": 1.9309127584467978,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004884168341596888,
|
|
"loss": 5.8581,
|
|
"mean_token_accuracy": 0.13476209864020347,
|
|
"num_tokens": 4229444.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.0170416831970215,
|
|
"epoch": 1.9359556227937469,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004882908708191797,
|
|
"loss": 5.8266,
|
|
"mean_token_accuracy": 0.13455113247036934,
|
|
"num_tokens": 4240682.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.165860414505005,
|
|
"epoch": 1.9409984871406958,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048816424456572403,
|
|
"loss": 6.0864,
|
|
"mean_token_accuracy": 0.12815777510404586,
|
|
"num_tokens": 4253469.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.168189477920532,
|
|
"epoch": 1.946041351487645,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048803695579287905,
|
|
"loss": 5.9077,
|
|
"mean_token_accuracy": 0.13655197918415068,
|
|
"num_tokens": 4264173.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.087283897399902,
|
|
"epoch": 1.951084215834594,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048790900489626104,
|
|
"loss": 5.8836,
|
|
"mean_token_accuracy": 0.13402508944272995,
|
|
"num_tokens": 4275748.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.078026676177979,
|
|
"epoch": 1.9561270801815431,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004877803922735445,
|
|
"loss": 5.824,
|
|
"mean_token_accuracy": 0.1348547838628292,
|
|
"num_tokens": 4286624.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.069396734237671,
|
|
"epoch": 1.9611699445284922,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004876511183244603,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.13319926783442498,
|
|
"num_tokens": 4297722.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.090659189224243,
|
|
"epoch": 1.9662128088754414,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048752118345079496,
|
|
"loss": 5.8733,
|
|
"mean_token_accuracy": 0.13454780206084252,
|
|
"num_tokens": 4307718.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.0161665916442875,
|
|
"epoch": 1.9712556732223905,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048739058805638905,
|
|
"loss": 5.8686,
|
|
"mean_token_accuracy": 0.12878879010677338,
|
|
"num_tokens": 4320384.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.095842742919922,
|
|
"epoch": 1.9762985375693394,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000487259332547136,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.13814392685890198,
|
|
"num_tokens": 4331300.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.982296133041382,
|
|
"epoch": 1.9813414019162885,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000487127417330981,
|
|
"loss": 5.8092,
|
|
"mean_token_accuracy": 0.13893615901470185,
|
|
"num_tokens": 4341678.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.098065662384033,
|
|
"epoch": 1.9863842662632374,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004869948428179196,
|
|
"loss": 5.873,
|
|
"mean_token_accuracy": 0.13273516818881034,
|
|
"num_tokens": 4352924.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.043451976776123,
|
|
"epoch": 1.9914271306101865,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004868616094199964,
|
|
"loss": 5.8164,
|
|
"mean_token_accuracy": 0.1380403943359852,
|
|
"num_tokens": 4363133.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.0680694580078125,
|
|
"epoch": 1.9964699949571356,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000486727717551304,
|
|
"loss": 5.8742,
|
|
"mean_token_accuracy": 0.13288742080330848,
|
|
"num_tokens": 4376024.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.002887619866265,
|
|
"epoch": 2.00100857286939,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004865931676279813,
|
|
"loss": 5.7085,
|
|
"mean_token_accuracy": 0.14310597214433882,
|
|
"num_tokens": 4385440.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.066258478164673,
|
|
"epoch": 2.006051437216339,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004864579600682125,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.14719632714986802,
|
|
"num_tokens": 4397208.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.854479265213013,
|
|
"epoch": 2.011094301563288,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048632209529222607,
|
|
"loss": 5.4859,
|
|
"mean_token_accuracy": 0.15549872666597367,
|
|
"num_tokens": 4406863.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.0857807159423825,
|
|
"epoch": 2.016137165910237,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048618557372229253,
|
|
"loss": 5.6465,
|
|
"mean_token_accuracy": 0.14307259395718575,
|
|
"num_tokens": 4419069.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.016137165910237,
|
|
"eval_entropy": 5.91173945795792,
|
|
"eval_loss": 5.959563732147217,
|
|
"eval_mean_token_accuracy": 0.13467797030322728,
|
|
"eval_num_tokens": 4419069.0,
|
|
"eval_runtime": 2.0294,
|
|
"eval_samples_per_second": 1546.266,
|
|
"eval_steps_per_second": 193.653,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"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": 1093211539752960.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|