c33b01fdac
Model: fpadovani/ita-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed455 Source: Original Platform
10090 lines
275 KiB
JSON
10090 lines
275 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 5.040342914775593,
|
|
"eval_steps": 1000,
|
|
"global_step": 5000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.0552534580230715,
|
|
"epoch": 2.0211800302571863,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004860483957827243,
|
|
"loss": 5.5994,
|
|
"mean_token_accuracy": 0.15007765144109725,
|
|
"num_tokens": 4432237.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.877443599700928,
|
|
"epoch": 2.026222894604135,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004859105618998735,
|
|
"loss": 5.4805,
|
|
"mean_token_accuracy": 0.15430912971496583,
|
|
"num_tokens": 4441256.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.051828908920288,
|
|
"epoch": 2.031265758951084,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004857720725021311,
|
|
"loss": 5.631,
|
|
"mean_token_accuracy": 0.15027819573879242,
|
|
"num_tokens": 4452509.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.036274003982544,
|
|
"epoch": 2.036308623298033,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048563292801992534,
|
|
"loss": 5.5939,
|
|
"mean_token_accuracy": 0.14560532867908477,
|
|
"num_tokens": 4464819.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.864096164703369,
|
|
"epoch": 2.0413514876449823,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004854931288857203,
|
|
"loss": 5.5211,
|
|
"mean_token_accuracy": 0.15042279660701752,
|
|
"num_tokens": 4474867.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.974116468429566,
|
|
"epoch": 2.0463943519919314,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048535267553401505,
|
|
"loss": 5.5002,
|
|
"mean_token_accuracy": 0.1578679770231247,
|
|
"num_tokens": 4485229.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.9208403587341305,
|
|
"epoch": 2.0514372163388805,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048521156840134193,
|
|
"loss": 5.5047,
|
|
"mean_token_accuracy": 0.14796890020370485,
|
|
"num_tokens": 4496589.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.9155717372894285,
|
|
"epoch": 2.0564800806858297,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000485069807926265,
|
|
"loss": 5.5963,
|
|
"mean_token_accuracy": 0.15184325575828553,
|
|
"num_tokens": 4508179.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.968095779418945,
|
|
"epoch": 2.0615229450327788,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004849273945493792,
|
|
"loss": 5.5835,
|
|
"mean_token_accuracy": 0.14911481440067292,
|
|
"num_tokens": 4518796.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.962812185287476,
|
|
"epoch": 2.066565809379728,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004847843287133084,
|
|
"loss": 5.5885,
|
|
"mean_token_accuracy": 0.14909806177020074,
|
|
"num_tokens": 4530544.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.899384880065918,
|
|
"epoch": 2.0716086737266766,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004846406108627048,
|
|
"loss": 5.5116,
|
|
"mean_token_accuracy": 0.15038229301571845,
|
|
"num_tokens": 4541382.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.001844167709351,
|
|
"epoch": 2.0766515380736257,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048449624144424665,
|
|
"loss": 5.6939,
|
|
"mean_token_accuracy": 0.14810260832309724,
|
|
"num_tokens": 4552318.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.939915561676026,
|
|
"epoch": 2.081694402420575,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048435122090663744,
|
|
"loss": 5.5055,
|
|
"mean_token_accuracy": 0.15254699885845185,
|
|
"num_tokens": 4562746.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.882055282592773,
|
|
"epoch": 2.086737266767524,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048420554970060424,
|
|
"loss": 5.5871,
|
|
"mean_token_accuracy": 0.15026217699050903,
|
|
"num_tokens": 4574635.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.966374444961548,
|
|
"epoch": 2.091780131114473,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048405922827889677,
|
|
"loss": 5.6398,
|
|
"mean_token_accuracy": 0.15539249479770662,
|
|
"num_tokens": 4584735.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.965135002136231,
|
|
"epoch": 2.096822995461422,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048391225709628515,
|
|
"loss": 5.527,
|
|
"mean_token_accuracy": 0.1471520520746708,
|
|
"num_tokens": 4596090.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.8798826217651365,
|
|
"epoch": 2.1018658598083713,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004837646366095593,
|
|
"loss": 5.6354,
|
|
"mean_token_accuracy": 0.14944104999303817,
|
|
"num_tokens": 4607692.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.899916696548462,
|
|
"epoch": 2.1069087241553204,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048361636727752716,
|
|
"loss": 5.5062,
|
|
"mean_token_accuracy": 0.15163369178771974,
|
|
"num_tokens": 4618319.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.90755262374878,
|
|
"epoch": 2.1119515885022695,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004834674495610131,
|
|
"loss": 5.6344,
|
|
"mean_token_accuracy": 0.14891936182975768,
|
|
"num_tokens": 4628650.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.058501863479615,
|
|
"epoch": 2.116994452849218,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00048331788392285693,
|
|
"loss": 5.6365,
|
|
"mean_token_accuracy": 0.14694681018590927,
|
|
"num_tokens": 4639883.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.843246173858643,
|
|
"epoch": 2.1220373171961673,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048316767082791205,
|
|
"loss": 5.509,
|
|
"mean_token_accuracy": 0.1515124186873436,
|
|
"num_tokens": 4650041.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.929569482803345,
|
|
"epoch": 2.1270801815431164,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048301681074304427,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.14349849224090577,
|
|
"num_tokens": 4661226.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.9502081871032715,
|
|
"epoch": 2.1321230458900655,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048286530413713016,
|
|
"loss": 5.6538,
|
|
"mean_token_accuracy": 0.14704589173197746,
|
|
"num_tokens": 4672658.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.916981601715088,
|
|
"epoch": 2.1371659102370146,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048271315148105574,
|
|
"loss": 5.5242,
|
|
"mean_token_accuracy": 0.15154259353876115,
|
|
"num_tokens": 4683214.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.878285980224609,
|
|
"epoch": 2.1422087745839637,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000482560353247715,
|
|
"loss": 5.5644,
|
|
"mean_token_accuracy": 0.15382485687732697,
|
|
"num_tokens": 4694787.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.90423583984375,
|
|
"epoch": 2.147251638930913,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004824069099120084,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.1500967636704445,
|
|
"num_tokens": 4705827.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.854505443572998,
|
|
"epoch": 2.152294503277862,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004822528219508413,
|
|
"loss": 5.5784,
|
|
"mean_token_accuracy": 0.14981241375207902,
|
|
"num_tokens": 4717197.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.017052078247071,
|
|
"epoch": 2.157337367624811,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004820980898431227,
|
|
"loss": 5.6975,
|
|
"mean_token_accuracy": 0.14760568588972092,
|
|
"num_tokens": 4729000.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.89847846031189,
|
|
"epoch": 2.1623802319717598,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004819427140697635,
|
|
"loss": 5.5886,
|
|
"mean_token_accuracy": 0.14233839511871338,
|
|
"num_tokens": 4739999.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.943842554092408,
|
|
"epoch": 2.167423096318709,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004817866951136753,
|
|
"loss": 5.5741,
|
|
"mean_token_accuracy": 0.1452633783221245,
|
|
"num_tokens": 4752461.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.997279262542724,
|
|
"epoch": 2.172465960665658,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048163003345976847,
|
|
"loss": 5.6339,
|
|
"mean_token_accuracy": 0.14656942561268807,
|
|
"num_tokens": 4763114.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.925815391540527,
|
|
"epoch": 2.177508825012607,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004814727295949512,
|
|
"loss": 5.6482,
|
|
"mean_token_accuracy": 0.14550220668315889,
|
|
"num_tokens": 4773926.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.922710275650024,
|
|
"epoch": 2.1825516893595562,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048131478400812735,
|
|
"loss": 5.5627,
|
|
"mean_token_accuracy": 0.1553325057029724,
|
|
"num_tokens": 4784062.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.0119630813598635,
|
|
"epoch": 2.1875945537065054,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004811561971901956,
|
|
"loss": 5.5651,
|
|
"mean_token_accuracy": 0.14884001687169074,
|
|
"num_tokens": 4794763.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.896900796890259,
|
|
"epoch": 2.1926374180534545,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004809969696340474,
|
|
"loss": 5.5999,
|
|
"mean_token_accuracy": 0.1485906720161438,
|
|
"num_tokens": 4804317.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.835540294647217,
|
|
"epoch": 2.1976802824004036,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048083710183456554,
|
|
"loss": 5.5447,
|
|
"mean_token_accuracy": 0.14782537817955016,
|
|
"num_tokens": 4816229.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.907172441482544,
|
|
"epoch": 2.2027231467473527,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048067659428862306,
|
|
"loss": 5.5889,
|
|
"mean_token_accuracy": 0.15646623820066452,
|
|
"num_tokens": 4827317.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.898754835128784,
|
|
"epoch": 2.2077660110943014,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048051544749508096,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.14502012878656387,
|
|
"num_tokens": 4838519.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.895417547225952,
|
|
"epoch": 2.2128088754412505,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004803536619547872,
|
|
"loss": 5.5751,
|
|
"mean_token_accuracy": 0.1473906010389328,
|
|
"num_tokens": 4848916.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.999032783508301,
|
|
"epoch": 2.2178517397881996,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000480191238170575,
|
|
"loss": 5.6661,
|
|
"mean_token_accuracy": 0.14679714888334275,
|
|
"num_tokens": 4860977.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.803864336013794,
|
|
"epoch": 2.2228946041351487,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004800281766472612,
|
|
"loss": 5.4242,
|
|
"mean_token_accuracy": 0.15536852180957794,
|
|
"num_tokens": 4871332.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.974799108505249,
|
|
"epoch": 2.227937468482098,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047986447789164486,
|
|
"loss": 5.6471,
|
|
"mean_token_accuracy": 0.14665255099534988,
|
|
"num_tokens": 4883139.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.971863746643066,
|
|
"epoch": 2.232980332829047,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004797001424125054,
|
|
"loss": 5.6218,
|
|
"mean_token_accuracy": 0.14806001037359237,
|
|
"num_tokens": 4894892.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.882292938232422,
|
|
"epoch": 2.238023197175996,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047953517072060124,
|
|
"loss": 5.6297,
|
|
"mean_token_accuracy": 0.1497166186571121,
|
|
"num_tokens": 4906059.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.972710275650025,
|
|
"epoch": 2.243066061522945,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004793695633286684,
|
|
"loss": 5.5835,
|
|
"mean_token_accuracy": 0.1511883981525898,
|
|
"num_tokens": 4917527.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.890481567382812,
|
|
"epoch": 2.2481089258698943,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047920332075141825,
|
|
"loss": 5.5568,
|
|
"mean_token_accuracy": 0.15390919744968415,
|
|
"num_tokens": 4928502.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.889628076553345,
|
|
"epoch": 2.253151790216843,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004790364435055367,
|
|
"loss": 5.5158,
|
|
"mean_token_accuracy": 0.14949529021978378,
|
|
"num_tokens": 4938179.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.9327967166900635,
|
|
"epoch": 2.258194654563792,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000478868932109682,
|
|
"loss": 5.6155,
|
|
"mean_token_accuracy": 0.1446187674999237,
|
|
"num_tokens": 4948979.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.878043174743652,
|
|
"epoch": 2.263237518910741,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004787007870844835,
|
|
"loss": 5.5065,
|
|
"mean_token_accuracy": 0.148458269238472,
|
|
"num_tokens": 4958459.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.9287214279174805,
|
|
"epoch": 2.2682803832576903,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00047853200895253976,
|
|
"loss": 5.5901,
|
|
"mean_token_accuracy": 0.13930843025445938,
|
|
"num_tokens": 4970680.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.901183986663819,
|
|
"epoch": 2.2733232476046394,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047836259823841716,
|
|
"loss": 5.6461,
|
|
"mean_token_accuracy": 0.14999350756406785,
|
|
"num_tokens": 4981635.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.896735429763794,
|
|
"epoch": 2.2783661119515886,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047819255546864797,
|
|
"loss": 5.5778,
|
|
"mean_token_accuracy": 0.15206377655267717,
|
|
"num_tokens": 4992661.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.898317861557007,
|
|
"epoch": 2.2834089762985377,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004780218811717293,
|
|
"loss": 5.6334,
|
|
"mean_token_accuracy": 0.14954008609056474,
|
|
"num_tokens": 5004364.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.922018718719483,
|
|
"epoch": 2.288451840645487,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047785057587812057,
|
|
"loss": 5.6615,
|
|
"mean_token_accuracy": 0.15065123289823532,
|
|
"num_tokens": 5015575.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.913210201263428,
|
|
"epoch": 2.293494704992436,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00047767864012024266,
|
|
"loss": 5.5648,
|
|
"mean_token_accuracy": 0.15293241441249847,
|
|
"num_tokens": 5026984.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.891224336624146,
|
|
"epoch": 2.2985375693393846,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00047750607443247585,
|
|
"loss": 5.5696,
|
|
"mean_token_accuracy": 0.14152300506830215,
|
|
"num_tokens": 5039835.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.862606430053711,
|
|
"epoch": 2.3035804336863337,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047733287935115837,
|
|
"loss": 5.6331,
|
|
"mean_token_accuracy": 0.14703422784805298,
|
|
"num_tokens": 5052284.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.899987602233887,
|
|
"epoch": 2.308623298033283,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004771590554145844,
|
|
"loss": 5.5302,
|
|
"mean_token_accuracy": 0.15128262341022491,
|
|
"num_tokens": 5062535.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.828275728225708,
|
|
"epoch": 2.313666162380232,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047698460316300285,
|
|
"loss": 5.6422,
|
|
"mean_token_accuracy": 0.14976432174444199,
|
|
"num_tokens": 5074065.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.872760200500489,
|
|
"epoch": 2.318709026727181,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004768095231386153,
|
|
"loss": 5.5077,
|
|
"mean_token_accuracy": 0.14901819974184036,
|
|
"num_tokens": 5085269.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.880764389038086,
|
|
"epoch": 2.32375189107413,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004766338158855744,
|
|
"loss": 5.6058,
|
|
"mean_token_accuracy": 0.14826219528913498,
|
|
"num_tokens": 5096537.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.921002721786499,
|
|
"epoch": 2.3287947554210793,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004764574819499826,
|
|
"loss": 5.6333,
|
|
"mean_token_accuracy": 0.14573112279176711,
|
|
"num_tokens": 5109554.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.935253524780274,
|
|
"epoch": 2.3338376197680284,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047628052187988955,
|
|
"loss": 5.581,
|
|
"mean_token_accuracy": 0.14508368000388144,
|
|
"num_tokens": 5119740.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.782969999313354,
|
|
"epoch": 2.338880484114977,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004761029362252914,
|
|
"loss": 5.6456,
|
|
"mean_token_accuracy": 0.14789862781763077,
|
|
"num_tokens": 5130682.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.949564361572266,
|
|
"epoch": 2.343923348461926,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047592472553812856,
|
|
"loss": 5.5529,
|
|
"mean_token_accuracy": 0.15189136415719987,
|
|
"num_tokens": 5141290.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.781671762466431,
|
|
"epoch": 2.3489662128088753,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004757458903722837,
|
|
"loss": 5.5209,
|
|
"mean_token_accuracy": 0.1490607626736164,
|
|
"num_tokens": 5152930.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.8325927734375,
|
|
"epoch": 2.3540090771558244,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047556643128358075,
|
|
"loss": 5.5707,
|
|
"mean_token_accuracy": 0.14698617458343505,
|
|
"num_tokens": 5164031.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.897674417495727,
|
|
"epoch": 2.3590519415027735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047538634882978277,
|
|
"loss": 5.5536,
|
|
"mean_token_accuracy": 0.14963567852973939,
|
|
"num_tokens": 5174833.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.818548059463501,
|
|
"epoch": 2.3640948058497226,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047520564357059013,
|
|
"loss": 5.4996,
|
|
"mean_token_accuracy": 0.1549072280526161,
|
|
"num_tokens": 5184020.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.854113054275513,
|
|
"epoch": 2.3691376701966718,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004750243160676389,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.15382556170225142,
|
|
"num_tokens": 5193812.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.944058179855347,
|
|
"epoch": 2.374180534543621,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004748423668844992,
|
|
"loss": 5.7206,
|
|
"mean_token_accuracy": 0.15199883580207824,
|
|
"num_tokens": 5204663.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.909019804000854,
|
|
"epoch": 2.37922339889057,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004746597965866733,
|
|
"loss": 5.5072,
|
|
"mean_token_accuracy": 0.1537255123257637,
|
|
"num_tokens": 5215582.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.920667362213135,
|
|
"epoch": 2.384266263237519,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004744766057415938,
|
|
"loss": 5.583,
|
|
"mean_token_accuracy": 0.1504060596227646,
|
|
"num_tokens": 5227185.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.780981731414795,
|
|
"epoch": 2.389309127584468,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004742927949186222,
|
|
"loss": 5.5358,
|
|
"mean_token_accuracy": 0.1539474681019783,
|
|
"num_tokens": 5238800.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.8042927265167235,
|
|
"epoch": 2.394351991931417,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047410836468904664,
|
|
"loss": 5.5143,
|
|
"mean_token_accuracy": 0.15056025832891465,
|
|
"num_tokens": 5249149.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.800653028488159,
|
|
"epoch": 2.399394856278366,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047392331562608073,
|
|
"loss": 5.4598,
|
|
"mean_token_accuracy": 0.15479990243911743,
|
|
"num_tokens": 5259753.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.819547033309936,
|
|
"epoch": 2.404437720625315,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004737376483048609,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.1512117490172386,
|
|
"num_tokens": 5269954.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.868506288528442,
|
|
"epoch": 2.4094805849722642,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004735513633024458,
|
|
"loss": 5.5635,
|
|
"mean_token_accuracy": 0.15017117112874984,
|
|
"num_tokens": 5280519.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.782323026657105,
|
|
"epoch": 2.4145234493192134,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004733644611978134,
|
|
"loss": 5.5817,
|
|
"mean_token_accuracy": 0.15218057483434677,
|
|
"num_tokens": 5291729.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.835575723648072,
|
|
"epoch": 2.4195663136661625,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004731769425718597,
|
|
"loss": 5.5403,
|
|
"mean_token_accuracy": 0.1511191189289093,
|
|
"num_tokens": 5302277.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.735279512405396,
|
|
"epoch": 2.4246091780131116,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000472988808007397,
|
|
"loss": 5.5069,
|
|
"mean_token_accuracy": 0.15890975594520568,
|
|
"num_tokens": 5314432.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.742378854751587,
|
|
"epoch": 2.4296520423600603,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047280005808915176,
|
|
"loss": 5.5104,
|
|
"mean_token_accuracy": 0.1485556811094284,
|
|
"num_tokens": 5324707.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.898344039916992,
|
|
"epoch": 2.4346949067070094,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004726106934037631,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.15054061114788056,
|
|
"num_tokens": 5334414.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.778563833236694,
|
|
"epoch": 2.4397377710539585,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004724207145397809,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.15101237297058107,
|
|
"num_tokens": 5346372.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.777136564254761,
|
|
"epoch": 2.4447806354009076,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047223012208766384,
|
|
"loss": 5.5569,
|
|
"mean_token_accuracy": 0.14512302875518798,
|
|
"num_tokens": 5356720.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.980841493606567,
|
|
"epoch": 2.4498234997478567,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004720389166397777,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.14773910045623778,
|
|
"num_tokens": 5367959.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.804137086868286,
|
|
"epoch": 2.454866364094806,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047184709879039325,
|
|
"loss": 5.5195,
|
|
"mean_token_accuracy": 0.1527034819126129,
|
|
"num_tokens": 5378490.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.728920125961304,
|
|
"epoch": 2.459909228441755,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047165466913568497,
|
|
"loss": 5.4778,
|
|
"mean_token_accuracy": 0.15172019302845002,
|
|
"num_tokens": 5389326.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.830344295501709,
|
|
"epoch": 2.464952092788704,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004714616282737285,
|
|
"loss": 5.529,
|
|
"mean_token_accuracy": 0.15059774518013,
|
|
"num_tokens": 5402289.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.850159168243408,
|
|
"epoch": 2.469994957135653,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004712679768044995,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.14636312052607536,
|
|
"num_tokens": 5412487.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.841196727752686,
|
|
"epoch": 2.4750378214826023,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000471073715329871,
|
|
"loss": 5.5363,
|
|
"mean_token_accuracy": 0.14737867712974548,
|
|
"num_tokens": 5423939.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.794162082672119,
|
|
"epoch": 2.4800806858295514,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004708788444536123,
|
|
"loss": 5.597,
|
|
"mean_token_accuracy": 0.1550315424799919,
|
|
"num_tokens": 5434623.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.844778347015381,
|
|
"epoch": 2.4851235501765,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004706833647813865,
|
|
"loss": 5.5545,
|
|
"mean_token_accuracy": 0.14855214953422546,
|
|
"num_tokens": 5445368.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.857855415344238,
|
|
"epoch": 2.4901664145234492,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047048727692074893,
|
|
"loss": 5.5509,
|
|
"mean_token_accuracy": 0.14679374620318414,
|
|
"num_tokens": 5456678.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.816537570953369,
|
|
"epoch": 2.4952092788703983,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004702905814811452,
|
|
"loss": 5.5658,
|
|
"mean_token_accuracy": 0.14878202676773072,
|
|
"num_tokens": 5468181.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.903571844100952,
|
|
"epoch": 2.5002521432173475,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047009327907390926,
|
|
"loss": 5.6824,
|
|
"mean_token_accuracy": 0.14749020785093309,
|
|
"num_tokens": 5479858.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.829712390899658,
|
|
"epoch": 2.5052950075642966,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004698953703122616,
|
|
"loss": 5.5209,
|
|
"mean_token_accuracy": 0.14475518167018891,
|
|
"num_tokens": 5489885.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.84113187789917,
|
|
"epoch": 2.5103378719112457,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046969685581130715,
|
|
"loss": 5.4854,
|
|
"mean_token_accuracy": 0.1478098750114441,
|
|
"num_tokens": 5500597.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.709924793243408,
|
|
"epoch": 2.515380736258195,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00046949773618803353,
|
|
"loss": 5.4721,
|
|
"mean_token_accuracy": 0.15299285352230071,
|
|
"num_tokens": 5513830.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.699863624572754,
|
|
"epoch": 2.5204236006051435,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046929801206130925,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.14502166733145713,
|
|
"num_tokens": 5525318.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.867627096176148,
|
|
"epoch": 2.5254664649520926,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046909768405188135,
|
|
"loss": 5.5174,
|
|
"mean_token_accuracy": 0.15620429664850236,
|
|
"num_tokens": 5536338.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.785635614395142,
|
|
"epoch": 2.5305093292990417,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00046889675278237404,
|
|
"loss": 5.5396,
|
|
"mean_token_accuracy": 0.14952272772789002,
|
|
"num_tokens": 5550114.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.821239185333252,
|
|
"epoch": 2.535552193645991,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004686952188772863,
|
|
"loss": 5.5774,
|
|
"mean_token_accuracy": 0.1446179799735546,
|
|
"num_tokens": 5562066.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.830097436904907,
|
|
"epoch": 2.54059505799294,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046849308296299005,
|
|
"loss": 5.4994,
|
|
"mean_token_accuracy": 0.15423471182584764,
|
|
"num_tokens": 5573177.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.745027828216553,
|
|
"epoch": 2.545637922339889,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004682903456677285,
|
|
"loss": 5.485,
|
|
"mean_token_accuracy": 0.15495067834854126,
|
|
"num_tokens": 5583612.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.79877462387085,
|
|
"epoch": 2.550680786686838,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046808700762161385,
|
|
"loss": 5.5197,
|
|
"mean_token_accuracy": 0.1541713669896126,
|
|
"num_tokens": 5594188.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.7926177978515625,
|
|
"epoch": 2.5557236510337873,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004678830694566253,
|
|
"loss": 5.5584,
|
|
"mean_token_accuracy": 0.15409241616725922,
|
|
"num_tokens": 5604779.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.809142065048218,
|
|
"epoch": 2.5607665153807364,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004676785318066075,
|
|
"loss": 5.6185,
|
|
"mean_token_accuracy": 0.14566346108913422,
|
|
"num_tokens": 5615769.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.822908926010132,
|
|
"epoch": 2.5658093797276855,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004674733953072682,
|
|
"loss": 5.4845,
|
|
"mean_token_accuracy": 0.1612080842256546,
|
|
"num_tokens": 5626402.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.72437801361084,
|
|
"epoch": 2.5708522440746346,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004672676605961763,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.15387741476297379,
|
|
"num_tokens": 5637001.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.896881866455078,
|
|
"epoch": 2.5758951084215833,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046706132831276007,
|
|
"loss": 5.598,
|
|
"mean_token_accuracy": 0.14745208919048308,
|
|
"num_tokens": 5649265.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.775057935714722,
|
|
"epoch": 2.5809379727685324,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046685439909830486,
|
|
"loss": 5.5167,
|
|
"mean_token_accuracy": 0.1562715619802475,
|
|
"num_tokens": 5660755.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.778666305541992,
|
|
"epoch": 2.5859808371154815,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004666468735959516,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.1540301576256752,
|
|
"num_tokens": 5673084.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.800733518600464,
|
|
"epoch": 2.5910237014624307,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004664387524506943,
|
|
"loss": 5.5191,
|
|
"mean_token_accuracy": 0.15298924446105958,
|
|
"num_tokens": 5683524.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.761188268661499,
|
|
"epoch": 2.59606656580938,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004662300363093784,
|
|
"loss": 5.4562,
|
|
"mean_token_accuracy": 0.15839844942092896,
|
|
"num_tokens": 5693847.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.821705389022827,
|
|
"epoch": 2.601109430156329,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004660207258206983,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.15692463517189026,
|
|
"num_tokens": 5704915.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.829744863510132,
|
|
"epoch": 2.606152294503278,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00046581082163519585,
|
|
"loss": 5.5935,
|
|
"mean_token_accuracy": 0.1491970643401146,
|
|
"num_tokens": 5716783.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.812629508972168,
|
|
"epoch": 2.6111951588502267,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004656003244052582,
|
|
"loss": 5.5693,
|
|
"mean_token_accuracy": 0.14537739381194115,
|
|
"num_tokens": 5728993.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.864555215835571,
|
|
"epoch": 2.616238023197176,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004653892347851155,
|
|
"loss": 5.5596,
|
|
"mean_token_accuracy": 0.14665171056985854,
|
|
"num_tokens": 5740340.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.771637439727783,
|
|
"epoch": 2.621280887544125,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004651775534308392,
|
|
"loss": 5.5591,
|
|
"mean_token_accuracy": 0.15334956645965575,
|
|
"num_tokens": 5751545.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.742058229446411,
|
|
"epoch": 2.626323751891074,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004649652810003398,
|
|
"loss": 5.4808,
|
|
"mean_token_accuracy": 0.1505816325545311,
|
|
"num_tokens": 5761604.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.77313871383667,
|
|
"epoch": 2.631366616238023,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046475241815336504,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.15121986716985703,
|
|
"num_tokens": 5772350.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.787600040435791,
|
|
"epoch": 2.6364094805849723,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004645389655514974,
|
|
"loss": 5.582,
|
|
"mean_token_accuracy": 0.14978788793087006,
|
|
"num_tokens": 5783675.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.882177448272705,
|
|
"epoch": 2.6414523449319214,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004643249238581525,
|
|
"loss": 5.5893,
|
|
"mean_token_accuracy": 0.14864657074213028,
|
|
"num_tokens": 5794430.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.811367797851562,
|
|
"epoch": 2.6464952092788705,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004641102937385769,
|
|
"loss": 5.6118,
|
|
"mean_token_accuracy": 0.14719293788075447,
|
|
"num_tokens": 5806175.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.82002625465393,
|
|
"epoch": 2.6515380736258196,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004638950758598461,
|
|
"loss": 5.6089,
|
|
"mean_token_accuracy": 0.1414988726377487,
|
|
"num_tokens": 5817410.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.828128719329834,
|
|
"epoch": 2.6565809379727687,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046367927089086203,
|
|
"loss": 5.4959,
|
|
"mean_token_accuracy": 0.15349043309688568,
|
|
"num_tokens": 5828912.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.791985702514649,
|
|
"epoch": 2.661623802319718,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046346287950235163,
|
|
"loss": 5.5844,
|
|
"mean_token_accuracy": 0.15193765759468078,
|
|
"num_tokens": 5839258.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.800520324707032,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004632459023668642,
|
|
"loss": 5.5682,
|
|
"mean_token_accuracy": 0.14980051368474961,
|
|
"num_tokens": 5849959.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.832754135131836,
|
|
"epoch": 2.6717095310136156,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004630283401587699,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.15486376732587814,
|
|
"num_tokens": 5860426.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.908721876144409,
|
|
"epoch": 2.6767523953605648,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004628101935542568,
|
|
"loss": 5.6051,
|
|
"mean_token_accuracy": 0.148734749853611,
|
|
"num_tokens": 5871388.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.7463287830352785,
|
|
"epoch": 2.681795259707514,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00046259146323132986,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.14766682982444762,
|
|
"num_tokens": 5881949.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.859231948852539,
|
|
"epoch": 2.686838124054463,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046237214986980776,
|
|
"loss": 5.6022,
|
|
"mean_token_accuracy": 0.14296458661556244,
|
|
"num_tokens": 5893144.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.81245641708374,
|
|
"epoch": 2.691880988401412,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004621522541513216,
|
|
"loss": 5.4865,
|
|
"mean_token_accuracy": 0.14680671393871308,
|
|
"num_tokens": 5903305.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.688008594512939,
|
|
"epoch": 2.696923852748361,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004619317767593123,
|
|
"loss": 5.5526,
|
|
"mean_token_accuracy": 0.14722446650266646,
|
|
"num_tokens": 5914713.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.848729848861694,
|
|
"epoch": 2.70196671709531,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046171071837902873,
|
|
"loss": 5.5321,
|
|
"mean_token_accuracy": 0.15542553961277009,
|
|
"num_tokens": 5925576.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.76939058303833,
|
|
"epoch": 2.707009581442259,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004614890796975254,
|
|
"loss": 5.5409,
|
|
"mean_token_accuracy": 0.14616062939167024,
|
|
"num_tokens": 5935686.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.795203161239624,
|
|
"epoch": 2.712052445789208,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004612668614036605,
|
|
"loss": 5.5171,
|
|
"mean_token_accuracy": 0.14774811863899232,
|
|
"num_tokens": 5947741.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.742692279815674,
|
|
"epoch": 2.7170953101361572,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004610440641880934,
|
|
"loss": 5.5483,
|
|
"mean_token_accuracy": 0.1518729194998741,
|
|
"num_tokens": 5958288.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.796945953369141,
|
|
"epoch": 2.7221381744831064,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00046082068874328324,
|
|
"loss": 5.5054,
|
|
"mean_token_accuracy": 0.15013954192399978,
|
|
"num_tokens": 5969815.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.803002071380615,
|
|
"epoch": 2.7271810388300555,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046059673576348596,
|
|
"loss": 5.5495,
|
|
"mean_token_accuracy": 0.15700096487998963,
|
|
"num_tokens": 5980795.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.7719989776611325,
|
|
"epoch": 2.7322239031770046,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046037220594475264,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.14789892435073854,
|
|
"num_tokens": 5992284.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.7986085414886475,
|
|
"epoch": 2.7372667675239537,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000460147099984927,
|
|
"loss": 5.4559,
|
|
"mean_token_accuracy": 0.1507952943444252,
|
|
"num_tokens": 6002161.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.780350208282471,
|
|
"epoch": 2.742309631870903,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004599214185836438,
|
|
"loss": 5.4476,
|
|
"mean_token_accuracy": 0.15204527378082275,
|
|
"num_tokens": 6013154.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.766159725189209,
|
|
"epoch": 2.747352496217852,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045969516244232574,
|
|
"loss": 5.5534,
|
|
"mean_token_accuracy": 0.146834909170866,
|
|
"num_tokens": 6023567.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.874376249313355,
|
|
"epoch": 2.752395360564801,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045946833226418235,
|
|
"loss": 5.5572,
|
|
"mean_token_accuracy": 0.15636955946683884,
|
|
"num_tokens": 6034381.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.828603649139405,
|
|
"epoch": 2.7574382249117497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004592409287542069,
|
|
"loss": 5.5901,
|
|
"mean_token_accuracy": 0.15142763406038284,
|
|
"num_tokens": 6045899.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.657381296157837,
|
|
"epoch": 2.762481089258699,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004590129526191748,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.15433629155158995,
|
|
"num_tokens": 6056852.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.7464581489562985,
|
|
"epoch": 2.767523953605648,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004587844045676413,
|
|
"loss": 5.5434,
|
|
"mean_token_accuracy": 0.15072519481182098,
|
|
"num_tokens": 6067345.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.878564357757568,
|
|
"epoch": 2.772566817952597,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00045855528530993874,
|
|
"loss": 5.6189,
|
|
"mean_token_accuracy": 0.1496897079050541,
|
|
"num_tokens": 6078749.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.754133176803589,
|
|
"epoch": 2.777609682299546,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045832559555817525,
|
|
"loss": 5.4752,
|
|
"mean_token_accuracy": 0.15643274337053298,
|
|
"num_tokens": 6089577.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.725897407531738,
|
|
"epoch": 2.7826525466464953,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00045809533602623185,
|
|
"loss": 5.488,
|
|
"mean_token_accuracy": 0.15202998816967012,
|
|
"num_tokens": 6101253.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.834066534042359,
|
|
"epoch": 2.7876954109934444,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004578645074297603,
|
|
"loss": 5.543,
|
|
"mean_token_accuracy": 0.14669007956981658,
|
|
"num_tokens": 6113342.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.754017639160156,
|
|
"epoch": 2.792738275340393,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00045763311048618144,
|
|
"loss": 5.4574,
|
|
"mean_token_accuracy": 0.1498860999941826,
|
|
"num_tokens": 6124785.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.706370162963867,
|
|
"epoch": 2.797781139687342,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004574011459146822,
|
|
"loss": 5.5046,
|
|
"mean_token_accuracy": 0.15086573660373687,
|
|
"num_tokens": 6137592.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.786870956420898,
|
|
"epoch": 2.8028240040342913,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00045716861443621386,
|
|
"loss": 5.5262,
|
|
"mean_token_accuracy": 0.15265323370695114,
|
|
"num_tokens": 6148667.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.832633399963379,
|
|
"epoch": 2.8078668683812404,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00045693551677348967,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.14994143545627595,
|
|
"num_tokens": 6160713.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.790002489089966,
|
|
"epoch": 2.8129097327281896,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004567018536509823,
|
|
"loss": 5.6289,
|
|
"mean_token_accuracy": 0.14836482107639312,
|
|
"num_tokens": 6172272.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.765521478652954,
|
|
"epoch": 2.8179525970751387,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045646762579492245,
|
|
"loss": 5.4725,
|
|
"mean_token_accuracy": 0.14509377405047416,
|
|
"num_tokens": 6183785.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.713761329650879,
|
|
"epoch": 2.822995461422088,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045623283393329546,
|
|
"loss": 5.4211,
|
|
"mean_token_accuracy": 0.15780846476554872,
|
|
"num_tokens": 6194464.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.793527507781983,
|
|
"epoch": 2.828038325769037,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045599747879584,
|
|
"loss": 5.4798,
|
|
"mean_token_accuracy": 0.14737002104520797,
|
|
"num_tokens": 6204671.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.737512636184692,
|
|
"epoch": 2.833081190115986,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004557615611140451,
|
|
"loss": 5.4596,
|
|
"mean_token_accuracy": 0.15752613693475723,
|
|
"num_tokens": 6215385.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.683342123031617,
|
|
"epoch": 2.838124054462935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004555250816211485,
|
|
"loss": 5.3996,
|
|
"mean_token_accuracy": 0.15441375076770783,
|
|
"num_tokens": 6225912.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.859903478622437,
|
|
"epoch": 2.8431669188098843,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045528804105213395,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.1575697124004364,
|
|
"num_tokens": 6236559.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.750939083099365,
|
|
"epoch": 2.848209783156833,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004550504401437289,
|
|
"loss": 5.4778,
|
|
"mean_token_accuracy": 0.1542205587029457,
|
|
"num_tokens": 6247711.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.738182830810547,
|
|
"epoch": 2.853252647503782,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045481227963440235,
|
|
"loss": 5.5831,
|
|
"mean_token_accuracy": 0.14905681312084199,
|
|
"num_tokens": 6258221.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.674665546417236,
|
|
"epoch": 2.858295511850731,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00045457356026436295,
|
|
"loss": 5.4458,
|
|
"mean_token_accuracy": 0.15977903604507446,
|
|
"num_tokens": 6270839.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.79375810623169,
|
|
"epoch": 2.8633383761976803,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045433428277555577,
|
|
"loss": 5.5393,
|
|
"mean_token_accuracy": 0.15413761287927627,
|
|
"num_tokens": 6281014.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.784238481521607,
|
|
"epoch": 2.8683812405446294,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004540944479116609,
|
|
"loss": 5.4007,
|
|
"mean_token_accuracy": 0.16100236773490906,
|
|
"num_tokens": 6291585.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.736752510070801,
|
|
"epoch": 2.8734241048915785,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004538540564180906,
|
|
"loss": 5.4882,
|
|
"mean_token_accuracy": 0.15270894691348075,
|
|
"num_tokens": 6302474.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.705942535400391,
|
|
"epoch": 2.8784669692385276,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004536131090419871,
|
|
"loss": 5.4876,
|
|
"mean_token_accuracy": 0.1556615948677063,
|
|
"num_tokens": 6314294.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.750041437149048,
|
|
"epoch": 2.8835098335854763,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004533716065322205,
|
|
"loss": 5.5287,
|
|
"mean_token_accuracy": 0.15640472173690795,
|
|
"num_tokens": 6325465.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.736567258834839,
|
|
"epoch": 2.8885526979324254,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004531295496393861,
|
|
"loss": 5.4374,
|
|
"mean_token_accuracy": 0.15409981161355973,
|
|
"num_tokens": 6335010.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.7275248050689695,
|
|
"epoch": 2.8935955622793745,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004528869391158023,
|
|
"loss": 5.5051,
|
|
"mean_token_accuracy": 0.16013409495353698,
|
|
"num_tokens": 6345640.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.701203298568726,
|
|
"epoch": 2.8986384266263236,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045264377571550834,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.16127857714891433,
|
|
"num_tokens": 6355539.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.691991996765137,
|
|
"epoch": 2.9036812909732728,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004524000601942615,
|
|
"loss": 5.3838,
|
|
"mean_token_accuracy": 0.15436727553606033,
|
|
"num_tokens": 6366542.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.788708353042603,
|
|
"epoch": 2.908724155320222,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004521557933095354,
|
|
"loss": 5.569,
|
|
"mean_token_accuracy": 0.1509890839457512,
|
|
"num_tokens": 6376735.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.70527663230896,
|
|
"epoch": 2.913767019667171,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000451910975820517,
|
|
"loss": 5.4097,
|
|
"mean_token_accuracy": 0.1568297877907753,
|
|
"num_tokens": 6386663.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.6970068454742435,
|
|
"epoch": 2.91880988401412,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00045166560848810476,
|
|
"loss": 5.4967,
|
|
"mean_token_accuracy": 0.15476052463054657,
|
|
"num_tokens": 6397886.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.7294145107269285,
|
|
"epoch": 2.9238527483610692,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045141969207490606,
|
|
"loss": 5.4553,
|
|
"mean_token_accuracy": 0.1620788887143135,
|
|
"num_tokens": 6408768.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.771146011352539,
|
|
"epoch": 2.9288956127080183,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004511732273452347,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.1514406159520149,
|
|
"num_tokens": 6419427.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.772919988632202,
|
|
"epoch": 2.9339384770549675,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004509262150651089,
|
|
"loss": 5.5391,
|
|
"mean_token_accuracy": 0.14892581328749657,
|
|
"num_tokens": 6430135.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.789680433273316,
|
|
"epoch": 2.938981341401916,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045067865600224833,
|
|
"loss": 5.4722,
|
|
"mean_token_accuracy": 0.14810145050287246,
|
|
"num_tokens": 6440977.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.71174840927124,
|
|
"epoch": 2.9440242057488653,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004504305509260724,
|
|
"loss": 5.5292,
|
|
"mean_token_accuracy": 0.1523166388273239,
|
|
"num_tokens": 6452319.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.866940307617187,
|
|
"epoch": 2.9490670700958144,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004501819006076974,
|
|
"loss": 5.5375,
|
|
"mean_token_accuracy": 0.15753424614667894,
|
|
"num_tokens": 6463970.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.828211402893066,
|
|
"epoch": 2.9541099344427635,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044993270581993413,
|
|
"loss": 5.5438,
|
|
"mean_token_accuracy": 0.1473902255296707,
|
|
"num_tokens": 6474113.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.746021556854248,
|
|
"epoch": 2.9591527987897126,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00044968296733728587,
|
|
"loss": 5.5207,
|
|
"mean_token_accuracy": 0.14902456998825073,
|
|
"num_tokens": 6485596.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.766293334960937,
|
|
"epoch": 2.9641956631366617,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004494326859359456,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.15855055004358293,
|
|
"num_tokens": 6497521.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.655600547790527,
|
|
"epoch": 2.969238527483611,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044918186239379363,
|
|
"loss": 5.3383,
|
|
"mean_token_accuracy": 0.16225115805864335,
|
|
"num_tokens": 6508034.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.6600078582763675,
|
|
"epoch": 2.9742813918305595,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004489304974903953,
|
|
"loss": 5.5183,
|
|
"mean_token_accuracy": 0.15390594750642778,
|
|
"num_tokens": 6518640.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.823737573623657,
|
|
"epoch": 2.9793242561775086,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004486785920069985,
|
|
"loss": 5.5205,
|
|
"mean_token_accuracy": 0.1515997052192688,
|
|
"num_tokens": 6528856.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.7825037956237795,
|
|
"epoch": 2.9843671205244577,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004484261467265313,
|
|
"loss": 5.5475,
|
|
"mean_token_accuracy": 0.15371402800083162,
|
|
"num_tokens": 6539385.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.6707898616790775,
|
|
"epoch": 2.989409984871407,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00044817316243359956,
|
|
"loss": 5.4999,
|
|
"mean_token_accuracy": 0.1549810454249382,
|
|
"num_tokens": 6550327.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.877950143814087,
|
|
"epoch": 2.994452849218356,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00044791963991448427,
|
|
"loss": 5.592,
|
|
"mean_token_accuracy": 0.1392840065062046,
|
|
"num_tokens": 6563230.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.734732627868652,
|
|
"epoch": 2.999495713565305,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00044766557995713915,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.15572975128889083,
|
|
"num_tokens": 6573947.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.821922938028972,
|
|
"epoch": 3.0040342914775593,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004474109833511885,
|
|
"loss": 5.3518,
|
|
"mean_token_accuracy": 0.1613782677385542,
|
|
"num_tokens": 6583671.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.781922054290772,
|
|
"epoch": 3.0090771558245084,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004471558508879244,
|
|
"loss": 5.1668,
|
|
"mean_token_accuracy": 0.16712762117385865,
|
|
"num_tokens": 6594270.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.673750400543213,
|
|
"epoch": 3.0141200201714575,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004469001833603045,
|
|
"loss": 5.2479,
|
|
"mean_token_accuracy": 0.17177896797657013,
|
|
"num_tokens": 6605849.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.7892529487609865,
|
|
"epoch": 3.0191628845184066,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004466439815629493,
|
|
"loss": 5.1923,
|
|
"mean_token_accuracy": 0.16783121824264527,
|
|
"num_tokens": 6615630.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.667241144180298,
|
|
"epoch": 3.0242057488653553,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044638724629214,
|
|
"loss": 5.2048,
|
|
"mean_token_accuracy": 0.17333222627639772,
|
|
"num_tokens": 6626467.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.0242057488653553,
|
|
"eval_entropy": 5.504511266567628,
|
|
"eval_loss": 5.790870666503906,
|
|
"eval_mean_token_accuracy": 0.14608151404988068,
|
|
"eval_num_tokens": 6626467.0,
|
|
"eval_runtime": 2.0251,
|
|
"eval_samples_per_second": 1549.556,
|
|
"eval_steps_per_second": 194.065,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.6987756252288815,
|
|
"epoch": 3.0292486132123044,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00044612997834581557,
|
|
"loss": 5.2257,
|
|
"mean_token_accuracy": 0.16876233369112015,
|
|
"num_tokens": 6638599.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.69688229560852,
|
|
"epoch": 3.0342914775592535,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044587217852357086,
|
|
"loss": 5.1674,
|
|
"mean_token_accuracy": 0.1671883672475815,
|
|
"num_tokens": 6648375.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.657116079330445,
|
|
"epoch": 3.0393343419062027,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00044561384762665363,
|
|
"loss": 5.2711,
|
|
"mean_token_accuracy": 0.168925179541111,
|
|
"num_tokens": 6659910.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.746803426742554,
|
|
"epoch": 3.044377206253152,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004453549864579623,
|
|
"loss": 5.2674,
|
|
"mean_token_accuracy": 0.17047284841537474,
|
|
"num_tokens": 6670719.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.622771978378296,
|
|
"epoch": 3.049420070600101,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004450955958220433,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.1788124457001686,
|
|
"num_tokens": 6681168.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.675848865509034,
|
|
"epoch": 3.05446293494705,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044483567652508864,
|
|
"loss": 5.2701,
|
|
"mean_token_accuracy": 0.1658952757716179,
|
|
"num_tokens": 6694026.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.699475622177124,
|
|
"epoch": 3.059505799293999,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445752293749337,
|
|
"loss": 5.1471,
|
|
"mean_token_accuracy": 0.17167919725179673,
|
|
"num_tokens": 6703666.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.648286247253418,
|
|
"epoch": 3.0645486636409482,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004443142551810541,
|
|
"loss": 5.1549,
|
|
"mean_token_accuracy": 0.1759031593799591,
|
|
"num_tokens": 6713506.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.671035957336426,
|
|
"epoch": 3.069591527987897,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044405275475456365,
|
|
"loss": 5.1574,
|
|
"mean_token_accuracy": 0.17762884497642517,
|
|
"num_tokens": 6724184.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.565056419372558,
|
|
"epoch": 3.074634392334846,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004437907289082116,
|
|
"loss": 5.2606,
|
|
"mean_token_accuracy": 0.16472148448228835,
|
|
"num_tokens": 6734843.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.785818195343017,
|
|
"epoch": 3.079677256681795,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00044352817845638053,
|
|
"loss": 5.2553,
|
|
"mean_token_accuracy": 0.16973095536231994,
|
|
"num_tokens": 6746194.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.709447288513184,
|
|
"epoch": 3.0847201210287443,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004432651042150833,
|
|
"loss": 5.2239,
|
|
"mean_token_accuracy": 0.1714300900697708,
|
|
"num_tokens": 6757617.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.69810266494751,
|
|
"epoch": 3.0897629853756934,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044300150700196054,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.1714382514357567,
|
|
"num_tokens": 6770026.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.6673102378845215,
|
|
"epoch": 3.0948058497226425,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004427373876362786,
|
|
"loss": 5.2005,
|
|
"mean_token_accuracy": 0.17543677538633345,
|
|
"num_tokens": 6781072.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.631171655654907,
|
|
"epoch": 3.0998487140695916,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004424727469389265,
|
|
"loss": 5.2515,
|
|
"mean_token_accuracy": 0.16655698120594026,
|
|
"num_tokens": 6790758.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.652545595169068,
|
|
"epoch": 3.1048915784165407,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004422075857324137,
|
|
"loss": 5.1749,
|
|
"mean_token_accuracy": 0.17081340700387954,
|
|
"num_tokens": 6800929.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.683443975448609,
|
|
"epoch": 3.10993444276349,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004419419048408673,
|
|
"loss": 5.2641,
|
|
"mean_token_accuracy": 0.16619861274957656,
|
|
"num_tokens": 6811565.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.633646154403687,
|
|
"epoch": 3.1149773071104385,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004416757050900297,
|
|
"loss": 5.1929,
|
|
"mean_token_accuracy": 0.17773550599813462,
|
|
"num_tokens": 6822088.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.638602066040039,
|
|
"epoch": 3.1200201714573876,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004414089873072557,
|
|
"loss": 5.274,
|
|
"mean_token_accuracy": 0.1683272659778595,
|
|
"num_tokens": 6833032.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.666909122467041,
|
|
"epoch": 3.1250630358043368,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004411417523215105,
|
|
"loss": 5.1799,
|
|
"mean_token_accuracy": 0.17280325144529343,
|
|
"num_tokens": 6844224.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.659029579162597,
|
|
"epoch": 3.130105900151286,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004408740009633665,
|
|
"loss": 5.2337,
|
|
"mean_token_accuracy": 0.1697711542248726,
|
|
"num_tokens": 6855620.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.72894172668457,
|
|
"epoch": 3.135148764498235,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044060573406500117,
|
|
"loss": 5.2624,
|
|
"mean_token_accuracy": 0.1718868523836136,
|
|
"num_tokens": 6869080.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.6096104145050045,
|
|
"epoch": 3.140191628845184,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004403369524601941,
|
|
"loss": 5.1303,
|
|
"mean_token_accuracy": 0.18251424580812453,
|
|
"num_tokens": 6878893.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.603878736495972,
|
|
"epoch": 3.145234493192133,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004400676569843249,
|
|
"loss": 5.2431,
|
|
"mean_token_accuracy": 0.16451599895954133,
|
|
"num_tokens": 6889641.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.636689710617065,
|
|
"epoch": 3.1502773575390823,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043979784847436994,
|
|
"loss": 5.2371,
|
|
"mean_token_accuracy": 0.16656252294778823,
|
|
"num_tokens": 6900930.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.794859313964844,
|
|
"epoch": 3.1553202218860315,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004395275277689004,
|
|
"loss": 5.3309,
|
|
"mean_token_accuracy": 0.16986626088619233,
|
|
"num_tokens": 6912083.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.619329929351807,
|
|
"epoch": 3.16036308623298,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004392566957080793,
|
|
"loss": 5.2973,
|
|
"mean_token_accuracy": 0.1596845045685768,
|
|
"num_tokens": 6924677.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.6448863506317135,
|
|
"epoch": 3.1654059505799292,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00043898535313365884,
|
|
"loss": 5.2109,
|
|
"mean_token_accuracy": 0.17366909682750703,
|
|
"num_tokens": 6936148.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.675040006637573,
|
|
"epoch": 3.1704488149268784,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000438713500888978,
|
|
"loss": 5.2041,
|
|
"mean_token_accuracy": 0.17503865510225297,
|
|
"num_tokens": 6945956.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.575944328308106,
|
|
"epoch": 3.1754916792738275,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004384411398189599,
|
|
"loss": 5.1579,
|
|
"mean_token_accuracy": 0.17190011441707612,
|
|
"num_tokens": 6956785.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.653457689285278,
|
|
"epoch": 3.1805345436207766,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000438168270770109,
|
|
"loss": 5.2785,
|
|
"mean_token_accuracy": 0.16781815737485886,
|
|
"num_tokens": 6967652.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.65847954750061,
|
|
"epoch": 3.1855774079677257,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004378948945905085,
|
|
"loss": 5.2563,
|
|
"mean_token_accuracy": 0.16726302206516266,
|
|
"num_tokens": 6979682.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.622790575027466,
|
|
"epoch": 3.190620272314675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043762101212981805,
|
|
"loss": 5.2364,
|
|
"mean_token_accuracy": 0.17226118892431258,
|
|
"num_tokens": 6989733.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.5481565475463865,
|
|
"epoch": 3.195663136661624,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043734662423927054,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.16597001552581786,
|
|
"num_tokens": 6999713.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.65051040649414,
|
|
"epoch": 3.200706001008573,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004370717317716699,
|
|
"loss": 5.218,
|
|
"mean_token_accuracy": 0.16898648142814637,
|
|
"num_tokens": 7010066.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.63144588470459,
|
|
"epoch": 3.205748865355522,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004367963355813882,
|
|
"loss": 5.2454,
|
|
"mean_token_accuracy": 0.16329686641693114,
|
|
"num_tokens": 7020149.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.607960557937622,
|
|
"epoch": 3.210791729702471,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043652043652436323,
|
|
"loss": 5.2657,
|
|
"mean_token_accuracy": 0.16739798486232757,
|
|
"num_tokens": 7031417.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.703138732910157,
|
|
"epoch": 3.21583459404942,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004362440354580956,
|
|
"loss": 5.2727,
|
|
"mean_token_accuracy": 0.16097553819417953,
|
|
"num_tokens": 7042619.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.62476110458374,
|
|
"epoch": 3.220877458396369,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004359671332416462,
|
|
"loss": 5.2381,
|
|
"mean_token_accuracy": 0.1659028485417366,
|
|
"num_tokens": 7054003.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.640838193893432,
|
|
"epoch": 3.225920322743318,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004356897307356335,
|
|
"loss": 5.2966,
|
|
"mean_token_accuracy": 0.16723308116197586,
|
|
"num_tokens": 7065837.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.66088833808899,
|
|
"epoch": 3.2309631870902673,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004354118288022309,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.16727673262357712,
|
|
"num_tokens": 7076059.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.648311758041382,
|
|
"epoch": 3.2360060514372164,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043513342830516405,
|
|
"loss": 5.2473,
|
|
"mean_token_accuracy": 0.1683645948767662,
|
|
"num_tokens": 7086163.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.606088876724243,
|
|
"epoch": 3.2410489157841655,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000434854530109708,
|
|
"loss": 5.1644,
|
|
"mean_token_accuracy": 0.17205434888601304,
|
|
"num_tokens": 7096044.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.580563640594482,
|
|
"epoch": 3.2460917801311147,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00043457513508268487,
|
|
"loss": 5.2674,
|
|
"mean_token_accuracy": 0.16422151178121566,
|
|
"num_tokens": 7109189.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.6042952060699465,
|
|
"epoch": 3.2511346444780633,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00043429524409246083,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.16773535460233688,
|
|
"num_tokens": 7120418.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.649179935455322,
|
|
"epoch": 3.2561775088250124,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004340148580089436,
|
|
"loss": 5.2777,
|
|
"mean_token_accuracy": 0.1653271198272705,
|
|
"num_tokens": 7131895.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.589054298400879,
|
|
"epoch": 3.2612203731719616,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004337339777035794,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.17108710408210753,
|
|
"num_tokens": 7141608.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.708654880523682,
|
|
"epoch": 3.2662632375189107,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000433452604049351,
|
|
"loss": 5.3014,
|
|
"mean_token_accuracy": 0.16750621497631074,
|
|
"num_tokens": 7153550.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.65173077583313,
|
|
"epoch": 3.27130610186586,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043317073792077394,
|
|
"loss": 5.1852,
|
|
"mean_token_accuracy": 0.17359477430582046,
|
|
"num_tokens": 7162945.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.583182287216187,
|
|
"epoch": 3.276348966212809,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004328883801938947,
|
|
"loss": 5.1921,
|
|
"mean_token_accuracy": 0.1746538758277893,
|
|
"num_tokens": 7173426.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.583803176879883,
|
|
"epoch": 3.281391830559758,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043260553174628776,
|
|
"loss": 5.1638,
|
|
"mean_token_accuracy": 0.16372706592082978,
|
|
"num_tokens": 7184335.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.567218589782715,
|
|
"epoch": 3.286434694906707,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004323221934570524,
|
|
"loss": 5.2151,
|
|
"mean_token_accuracy": 0.16582974791526794,
|
|
"num_tokens": 7195587.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.705914926528931,
|
|
"epoch": 3.2914775592536563,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043203836620681074,
|
|
"loss": 5.2697,
|
|
"mean_token_accuracy": 0.16241308450698852,
|
|
"num_tokens": 7205616.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.57292160987854,
|
|
"epoch": 3.2965204236006054,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043175405087770435,
|
|
"loss": 5.1365,
|
|
"mean_token_accuracy": 0.17185734063386918,
|
|
"num_tokens": 7215642.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.554850244522095,
|
|
"epoch": 3.301563287947554,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00043146924835339186,
|
|
"loss": 5.1639,
|
|
"mean_token_accuracy": 0.16475398540496827,
|
|
"num_tokens": 7226607.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.60114860534668,
|
|
"epoch": 3.306606152294503,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00043118395951904605,
|
|
"loss": 5.243,
|
|
"mean_token_accuracy": 0.1650063782930374,
|
|
"num_tokens": 7238035.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.676028156280518,
|
|
"epoch": 3.3116490166414523,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004308981852613512,
|
|
"loss": 5.231,
|
|
"mean_token_accuracy": 0.16254142820835113,
|
|
"num_tokens": 7248183.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.620025014877319,
|
|
"epoch": 3.3166918809884014,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004306119264685003,
|
|
"loss": 5.2539,
|
|
"mean_token_accuracy": 0.16837486177682875,
|
|
"num_tokens": 7260092.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.655144500732422,
|
|
"epoch": 3.3217347453353505,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004303251840301923,
|
|
"loss": 5.2445,
|
|
"mean_token_accuracy": 0.16475940644741058,
|
|
"num_tokens": 7272262.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.6170656204223635,
|
|
"epoch": 3.3267776096822996,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00043003795883762933,
|
|
"loss": 5.2972,
|
|
"mean_token_accuracy": 0.17337860614061357,
|
|
"num_tokens": 7283061.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.6957062721252445,
|
|
"epoch": 3.3318204740292487,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042975025178351386,
|
|
"loss": 5.2965,
|
|
"mean_token_accuracy": 0.1646546885371208,
|
|
"num_tokens": 7292904.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.648697423934936,
|
|
"epoch": 3.336863338376198,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000429462063762046,
|
|
"loss": 5.22,
|
|
"mean_token_accuracy": 0.16747646033763885,
|
|
"num_tokens": 7304368.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.633818340301514,
|
|
"epoch": 3.3419062027231465,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00042917339566892095,
|
|
"loss": 5.3334,
|
|
"mean_token_accuracy": 0.16680456697940826,
|
|
"num_tokens": 7315500.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.657038640975952,
|
|
"epoch": 3.3469490670700957,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004288842484013257,
|
|
"loss": 5.2111,
|
|
"mean_token_accuracy": 0.17002334743738173,
|
|
"num_tokens": 7326479.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.65393099784851,
|
|
"epoch": 3.3519919314170448,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004285946228579365,
|
|
"loss": 5.3089,
|
|
"mean_token_accuracy": 0.16525124460458757,
|
|
"num_tokens": 7338284.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.668390703201294,
|
|
"epoch": 3.357034795763994,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004283045199389164,
|
|
"loss": 5.2307,
|
|
"mean_token_accuracy": 0.16933902204036713,
|
|
"num_tokens": 7348601.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.630464124679565,
|
|
"epoch": 3.362077660110943,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00042801394054591197,
|
|
"loss": 5.2619,
|
|
"mean_token_accuracy": 0.16639727354049683,
|
|
"num_tokens": 7360282.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.67201075553894,
|
|
"epoch": 3.367120524457892,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042772288558205057,
|
|
"loss": 5.2487,
|
|
"mean_token_accuracy": 0.16803878545761108,
|
|
"num_tokens": 7371377.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.532483053207398,
|
|
"epoch": 3.3721633888048412,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004274313559519377,
|
|
"loss": 5.1877,
|
|
"mean_token_accuracy": 0.17245490849018097,
|
|
"num_tokens": 7381895.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.643547010421753,
|
|
"epoch": 3.3772062531517903,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004271393525616544,
|
|
"loss": 5.2907,
|
|
"mean_token_accuracy": 0.16530250310897826,
|
|
"num_tokens": 7393211.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.634685897827149,
|
|
"epoch": 3.3822491174987395,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042684687631875377,
|
|
"loss": 5.2297,
|
|
"mean_token_accuracy": 0.16384206265211104,
|
|
"num_tokens": 7404618.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.635930728912354,
|
|
"epoch": 3.3872919818456886,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042655392813225864,
|
|
"loss": 5.192,
|
|
"mean_token_accuracy": 0.17214861512184143,
|
|
"num_tokens": 7415467.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.5750932693481445,
|
|
"epoch": 3.3923348461926373,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042626050891265884,
|
|
"loss": 5.2423,
|
|
"mean_token_accuracy": 0.1681326672434807,
|
|
"num_tokens": 7426594.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.689355421066284,
|
|
"epoch": 3.3973777105395864,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004259666195719081,
|
|
"loss": 5.2783,
|
|
"mean_token_accuracy": 0.16297342032194137,
|
|
"num_tokens": 7436920.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.671323347091675,
|
|
"epoch": 3.4024205748865355,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042567226102342134,
|
|
"loss": 5.3422,
|
|
"mean_token_accuracy": 0.168358151614666,
|
|
"num_tokens": 7448499.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.595593738555908,
|
|
"epoch": 3.4074634392334846,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004253774341820715,
|
|
"loss": 5.2544,
|
|
"mean_token_accuracy": 0.17542664408683778,
|
|
"num_tokens": 7459788.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.693252325057983,
|
|
"epoch": 3.4125063035804337,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004250821399641874,
|
|
"loss": 5.2294,
|
|
"mean_token_accuracy": 0.1667968213558197,
|
|
"num_tokens": 7472387.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.591354703903198,
|
|
"epoch": 3.417549167927383,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004247863792875502,
|
|
"loss": 5.2268,
|
|
"mean_token_accuracy": 0.16651408672332763,
|
|
"num_tokens": 7483242.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.555906581878662,
|
|
"epoch": 3.422592032274332,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042449015307139076,
|
|
"loss": 5.2518,
|
|
"mean_token_accuracy": 0.16951095014810563,
|
|
"num_tokens": 7494908.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.650915050506592,
|
|
"epoch": 3.427634896621281,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042419346223638725,
|
|
"loss": 5.3099,
|
|
"mean_token_accuracy": 0.16329773068428038,
|
|
"num_tokens": 7505468.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.6585064888000485,
|
|
"epoch": 3.4326777609682297,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004238963077046616,
|
|
"loss": 5.2732,
|
|
"mean_token_accuracy": 0.1615394189953804,
|
|
"num_tokens": 7517281.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.575494766235352,
|
|
"epoch": 3.437720625315179,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004235986903997767,
|
|
"loss": 5.2611,
|
|
"mean_token_accuracy": 0.16711698174476625,
|
|
"num_tokens": 7528680.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.592639923095703,
|
|
"epoch": 3.442763489662128,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042330061124673437,
|
|
"loss": 5.2476,
|
|
"mean_token_accuracy": 0.16082952469587325,
|
|
"num_tokens": 7540927.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.688634586334229,
|
|
"epoch": 3.447806354009077,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042300207117197116,
|
|
"loss": 5.2796,
|
|
"mean_token_accuracy": 0.17175026684999467,
|
|
"num_tokens": 7551827.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.578490495681763,
|
|
"epoch": 3.452849218356026,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042270307110335657,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.17266012132167816,
|
|
"num_tokens": 7561721.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.588348388671875,
|
|
"epoch": 3.4578920827029753,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004224036119701899,
|
|
"loss": 5.2426,
|
|
"mean_token_accuracy": 0.16711491495370864,
|
|
"num_tokens": 7571932.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.647637367248535,
|
|
"epoch": 3.4629349470499244,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042210369470319694,
|
|
"loss": 5.2233,
|
|
"mean_token_accuracy": 0.1681799590587616,
|
|
"num_tokens": 7581859.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.527620792388916,
|
|
"epoch": 3.4679778113968736,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004218033202345274,
|
|
"loss": 5.2101,
|
|
"mean_token_accuracy": 0.16634268462657928,
|
|
"num_tokens": 7592811.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.665244197845459,
|
|
"epoch": 3.4730206757438227,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004215024894977522,
|
|
"loss": 5.2993,
|
|
"mean_token_accuracy": 0.16073853224515916,
|
|
"num_tokens": 7604332.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.588452911376953,
|
|
"epoch": 3.478063540090772,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004212012034278601,
|
|
"loss": 5.2283,
|
|
"mean_token_accuracy": 0.1654362678527832,
|
|
"num_tokens": 7615387.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.604885053634644,
|
|
"epoch": 3.4831064044377205,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042089946296125513,
|
|
"loss": 5.2938,
|
|
"mean_token_accuracy": 0.16449018716812133,
|
|
"num_tokens": 7627381.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.693688488006591,
|
|
"epoch": 3.4881492687846696,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042059726903575373,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.16380953043699265,
|
|
"num_tokens": 7637327.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.454297828674316,
|
|
"epoch": 3.4931921331316187,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004202946225905815,
|
|
"loss": 5.1721,
|
|
"mean_token_accuracy": 0.17511086910963058,
|
|
"num_tokens": 7647709.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.590501928329468,
|
|
"epoch": 3.498234997478568,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004199915245663706,
|
|
"loss": 5.2635,
|
|
"mean_token_accuracy": 0.16579607129096985,
|
|
"num_tokens": 7660003.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.624577188491822,
|
|
"epoch": 3.503277861825517,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041968797590515673,
|
|
"loss": 5.193,
|
|
"mean_token_accuracy": 0.17277942150831221,
|
|
"num_tokens": 7671667.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.564048910140992,
|
|
"epoch": 3.508320726172466,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004193839775503761,
|
|
"loss": 5.2307,
|
|
"mean_token_accuracy": 0.17198647558689117,
|
|
"num_tokens": 7682653.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.58487286567688,
|
|
"epoch": 3.513363590519415,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041907953044686253,
|
|
"loss": 5.244,
|
|
"mean_token_accuracy": 0.17172675281763078,
|
|
"num_tokens": 7693834.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.6659149646759035,
|
|
"epoch": 3.518406454866364,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004187746355408448,
|
|
"loss": 5.3355,
|
|
"mean_token_accuracy": 0.16576267629861832,
|
|
"num_tokens": 7706323.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.651409387588501,
|
|
"epoch": 3.523449319213313,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004184692937799431,
|
|
"loss": 5.2576,
|
|
"mean_token_accuracy": 0.17061871737241746,
|
|
"num_tokens": 7719019.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.576673984527588,
|
|
"epoch": 3.528492183560262,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041816350611316693,
|
|
"loss": 5.2084,
|
|
"mean_token_accuracy": 0.16736092716455458,
|
|
"num_tokens": 7729312.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.5250208377838135,
|
|
"epoch": 3.533535047907211,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041785727349091117,
|
|
"loss": 5.1583,
|
|
"mean_token_accuracy": 0.17586343586444855,
|
|
"num_tokens": 7739049.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.599133110046386,
|
|
"epoch": 3.5385779122541603,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000417550596864954,
|
|
"loss": 5.3516,
|
|
"mean_token_accuracy": 0.1626273900270462,
|
|
"num_tokens": 7750076.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.646395254135132,
|
|
"epoch": 3.5436207766011094,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041724347718845344,
|
|
"loss": 5.226,
|
|
"mean_token_accuracy": 0.1665637582540512,
|
|
"num_tokens": 7761442.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.620016193389892,
|
|
"epoch": 3.5486636409480585,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00041693591541594455,
|
|
"loss": 5.213,
|
|
"mean_token_accuracy": 0.17034838497638702,
|
|
"num_tokens": 7772061.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.5513238430023195,
|
|
"epoch": 3.5537065052950076,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004166279125033363,
|
|
"loss": 5.2113,
|
|
"mean_token_accuracy": 0.16602938920259475,
|
|
"num_tokens": 7783200.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.645471858978271,
|
|
"epoch": 3.5587493696419568,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041631946940790893,
|
|
"loss": 5.3003,
|
|
"mean_token_accuracy": 0.16540544331073762,
|
|
"num_tokens": 7793737.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.619654655456543,
|
|
"epoch": 3.563792233988906,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00041601058708831077,
|
|
"loss": 5.2423,
|
|
"mean_token_accuracy": 0.16466260999441146,
|
|
"num_tokens": 7804894.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.567594003677368,
|
|
"epoch": 3.568835098335855,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000415701266504555,
|
|
"loss": 5.2195,
|
|
"mean_token_accuracy": 0.16465836018323898,
|
|
"num_tokens": 7815773.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.591540956497193,
|
|
"epoch": 3.5738779626828037,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004153915086180173,
|
|
"loss": 5.2434,
|
|
"mean_token_accuracy": 0.1661566138267517,
|
|
"num_tokens": 7825857.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.582561445236206,
|
|
"epoch": 3.578920827029753,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004150813143914323,
|
|
"loss": 5.2845,
|
|
"mean_token_accuracy": 0.16402859538793563,
|
|
"num_tokens": 7836878.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.619180488586426,
|
|
"epoch": 3.583963691376702,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004147706847888908,
|
|
"loss": 5.2335,
|
|
"mean_token_accuracy": 0.1697487637400627,
|
|
"num_tokens": 7848382.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.586086225509644,
|
|
"epoch": 3.589006555723651,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004144596207758367,
|
|
"loss": 5.1462,
|
|
"mean_token_accuracy": 0.17295999079942703,
|
|
"num_tokens": 7858736.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.507045745849609,
|
|
"epoch": 3.5940494200706,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004141481233190642,
|
|
"loss": 5.2076,
|
|
"mean_token_accuracy": 0.16352104097604753,
|
|
"num_tokens": 7869396.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.593862533569336,
|
|
"epoch": 3.5990922844175492,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004138361933867145,
|
|
"loss": 5.2196,
|
|
"mean_token_accuracy": 0.17079371362924575,
|
|
"num_tokens": 7880072.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.650041580200195,
|
|
"epoch": 3.6041351487644984,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004135238319482731,
|
|
"loss": 5.2709,
|
|
"mean_token_accuracy": 0.16798788011074067,
|
|
"num_tokens": 7892011.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.600645875930786,
|
|
"epoch": 3.609178013111447,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004132110399745666,
|
|
"loss": 5.2663,
|
|
"mean_token_accuracy": 0.166333569586277,
|
|
"num_tokens": 7903586.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.5365297317504885,
|
|
"epoch": 3.614220877458396,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004128978184377596,
|
|
"loss": 5.2123,
|
|
"mean_token_accuracy": 0.1653869405388832,
|
|
"num_tokens": 7912999.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.589789819717407,
|
|
"epoch": 3.6192637418053453,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004125841683113518,
|
|
"loss": 5.2601,
|
|
"mean_token_accuracy": 0.16760523468255997,
|
|
"num_tokens": 7924969.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.671685361862183,
|
|
"epoch": 3.6243066061522944,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041227009057017523,
|
|
"loss": 5.28,
|
|
"mean_token_accuracy": 0.16978633850812913,
|
|
"num_tokens": 7936811.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.573072910308838,
|
|
"epoch": 3.6293494704992435,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004119555861903906,
|
|
"loss": 5.1493,
|
|
"mean_token_accuracy": 0.17094312459230424,
|
|
"num_tokens": 7946886.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.640597486495972,
|
|
"epoch": 3.6343923348461926,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004116406561494847,
|
|
"loss": 5.2874,
|
|
"mean_token_accuracy": 0.1683449164032936,
|
|
"num_tokens": 7960485.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.5896141052246096,
|
|
"epoch": 3.6394351991931417,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00041132530142626756,
|
|
"loss": 5.2273,
|
|
"mean_token_accuracy": 0.17194492667913436,
|
|
"num_tokens": 7971460.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.613400602340699,
|
|
"epoch": 3.644478063540091,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041100952300086884,
|
|
"loss": 5.2806,
|
|
"mean_token_accuracy": 0.16401728391647338,
|
|
"num_tokens": 7984060.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.629595518112183,
|
|
"epoch": 3.64952092788704,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00041069332185473514,
|
|
"loss": 5.3091,
|
|
"mean_token_accuracy": 0.16997109800577165,
|
|
"num_tokens": 7996547.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.632047557830811,
|
|
"epoch": 3.654563792233989,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000410376698970627,
|
|
"loss": 5.2113,
|
|
"mean_token_accuracy": 0.16881775707006455,
|
|
"num_tokens": 8006952.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.585224390029907,
|
|
"epoch": 3.659606656580938,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00041005965533261554,
|
|
"loss": 5.356,
|
|
"mean_token_accuracy": 0.1635224476456642,
|
|
"num_tokens": 8020293.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.552936553955078,
|
|
"epoch": 3.6646495209278873,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040974219192607973,
|
|
"loss": 5.2114,
|
|
"mean_token_accuracy": 0.1683351442217827,
|
|
"num_tokens": 8030736.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.562558317184449,
|
|
"epoch": 3.669692385274836,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040942430973770323,
|
|
"loss": 5.2073,
|
|
"mean_token_accuracy": 0.1711008444428444,
|
|
"num_tokens": 8041206.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.636075115203857,
|
|
"epoch": 3.674735249621785,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004091060097554712,
|
|
"loss": 5.2505,
|
|
"mean_token_accuracy": 0.164190311729908,
|
|
"num_tokens": 8052549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.633226680755615,
|
|
"epoch": 3.679778113968734,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004087872929686673,
|
|
"loss": 5.285,
|
|
"mean_token_accuracy": 0.16565053164958954,
|
|
"num_tokens": 8064011.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.550704383850098,
|
|
"epoch": 3.6848209783156833,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00040846816036787056,
|
|
"loss": 5.1438,
|
|
"mean_token_accuracy": 0.1691153019666672,
|
|
"num_tokens": 8074824.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.590891790390015,
|
|
"epoch": 3.6898638426626325,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004081486129449527,
|
|
"loss": 5.2612,
|
|
"mean_token_accuracy": 0.16238484233617784,
|
|
"num_tokens": 8086207.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.634239959716797,
|
|
"epoch": 3.6949067070095816,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00040782865169307423,
|
|
"loss": 5.1349,
|
|
"mean_token_accuracy": 0.1794676348567009,
|
|
"num_tokens": 8097384.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.499998903274536,
|
|
"epoch": 3.6999495713565302,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004075082776066822,
|
|
"loss": 5.1429,
|
|
"mean_token_accuracy": 0.17033801525831221,
|
|
"num_tokens": 8110306.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.51184492111206,
|
|
"epoch": 3.7049924357034794,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004071874916815068,
|
|
"loss": 5.2301,
|
|
"mean_token_accuracy": 0.16277444511651992,
|
|
"num_tokens": 8121120.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.63077187538147,
|
|
"epoch": 3.7100353000504285,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00040686629491455787,
|
|
"loss": 5.241,
|
|
"mean_token_accuracy": 0.1600613683462143,
|
|
"num_tokens": 8131657.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.585669374465942,
|
|
"epoch": 3.7150781643973776,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040654468830412247,
|
|
"loss": 5.1347,
|
|
"mean_token_accuracy": 0.17947703152894973,
|
|
"num_tokens": 8141846.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.606108140945435,
|
|
"epoch": 3.7201210287443267,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040622267284976134,
|
|
"loss": 5.2862,
|
|
"mean_token_accuracy": 0.16393069326877593,
|
|
"num_tokens": 8152713.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.4805840969085695,
|
|
"epoch": 3.725163893091276,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00040590024955230596,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.1743374988436699,
|
|
"num_tokens": 8164805.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.67301869392395,
|
|
"epoch": 3.730206757438225,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004055774194138554,
|
|
"loss": 5.365,
|
|
"mean_token_accuracy": 0.16485568583011628,
|
|
"num_tokens": 8176932.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.620328712463379,
|
|
"epoch": 3.735249621785174,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004052541834377731,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.17944686412811278,
|
|
"num_tokens": 8189199.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.47112135887146,
|
|
"epoch": 3.740292486132123,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00040493054262868397,
|
|
"loss": 5.1525,
|
|
"mean_token_accuracy": 0.1761087268590927,
|
|
"num_tokens": 8200132.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.596665573120117,
|
|
"epoch": 3.7453353504790723,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00040460649799247116,
|
|
"loss": 5.2448,
|
|
"mean_token_accuracy": 0.16757119446992874,
|
|
"num_tokens": 8211392.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.569138813018799,
|
|
"epoch": 3.7503782148260214,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004042820505362728,
|
|
"loss": 5.2093,
|
|
"mean_token_accuracy": 0.1687075525522232,
|
|
"num_tokens": 8222831.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.554853677749634,
|
|
"epoch": 3.7554210791729705,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004039572012684792,
|
|
"loss": 5.2463,
|
|
"mean_token_accuracy": 0.1688399389386177,
|
|
"num_tokens": 8234077.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.636492443084717,
|
|
"epoch": 3.760463943519919,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040363195119872916,
|
|
"loss": 5.2969,
|
|
"mean_token_accuracy": 0.1703490912914276,
|
|
"num_tokens": 8245419.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.676793813705444,
|
|
"epoch": 3.7655068078668683,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040330630133790757,
|
|
"loss": 5.3543,
|
|
"mean_token_accuracy": 0.16485787481069564,
|
|
"num_tokens": 8256185.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.51821346282959,
|
|
"epoch": 3.7705496722138174,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040298025269814165,
|
|
"loss": 5.1689,
|
|
"mean_token_accuracy": 0.17099626511335372,
|
|
"num_tokens": 8265812.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.60272855758667,
|
|
"epoch": 3.7755925365607665,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004026538062927981,
|
|
"loss": 5.1583,
|
|
"mean_token_accuracy": 0.17004178911447526,
|
|
"num_tokens": 8275544.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.525404214859009,
|
|
"epoch": 3.7806354009077157,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00040232696313647985,
|
|
"loss": 5.2166,
|
|
"mean_token_accuracy": 0.16815973669290543,
|
|
"num_tokens": 8286774.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.540351009368896,
|
|
"epoch": 3.7856782652546648,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004019997242450231,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.16658818423748017,
|
|
"num_tokens": 8298027.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.658772611618042,
|
|
"epoch": 3.7907211296016134,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004016720906354937,
|
|
"loss": 5.3168,
|
|
"mean_token_accuracy": 0.16622681319713592,
|
|
"num_tokens": 8308574.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.58692569732666,
|
|
"epoch": 3.7957639939485626,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004013440633261845,
|
|
"loss": 5.2219,
|
|
"mean_token_accuracy": 0.1759239062666893,
|
|
"num_tokens": 8318085.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.559729671478271,
|
|
"epoch": 3.8008068582955117,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040101564333661213,
|
|
"loss": 5.1447,
|
|
"mean_token_accuracy": 0.17531963735818862,
|
|
"num_tokens": 8328470.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.6063823223114015,
|
|
"epoch": 3.805849722642461,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040068683168751316,
|
|
"loss": 5.2914,
|
|
"mean_token_accuracy": 0.17011469155550002,
|
|
"num_tokens": 8338680.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.579203414916992,
|
|
"epoch": 3.81089258698941,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040035762940084205,
|
|
"loss": 5.2113,
|
|
"mean_token_accuracy": 0.17002039551734924,
|
|
"num_tokens": 8349926.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.618601751327515,
|
|
"epoch": 3.815935451336359,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000400028037499767,
|
|
"loss": 5.3005,
|
|
"mean_token_accuracy": 0.16290369182825087,
|
|
"num_tokens": 8361035.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.6035080909729,
|
|
"epoch": 3.820978315683308,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003996980570086673,
|
|
"loss": 5.2035,
|
|
"mean_token_accuracy": 0.17239801436662675,
|
|
"num_tokens": 8371201.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.485435199737549,
|
|
"epoch": 3.8260211800302573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039936768895312987,
|
|
"loss": 5.1131,
|
|
"mean_token_accuracy": 0.1814608931541443,
|
|
"num_tokens": 8382078.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.522430562973023,
|
|
"epoch": 3.8310640443772064,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00039903693435994626,
|
|
"loss": 5.1615,
|
|
"mean_token_accuracy": 0.1676882177591324,
|
|
"num_tokens": 8394072.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.571205091476441,
|
|
"epoch": 3.8361069087241555,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00039870579425710937,
|
|
"loss": 5.2938,
|
|
"mean_token_accuracy": 0.16642478704452515,
|
|
"num_tokens": 8405077.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.616708183288575,
|
|
"epoch": 3.8411497730711046,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003983742696738103,
|
|
"loss": 5.2832,
|
|
"mean_token_accuracy": 0.1668484941124916,
|
|
"num_tokens": 8416009.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.63858380317688,
|
|
"epoch": 3.8461926374180537,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003980423616404351,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.16477254629135132,
|
|
"num_tokens": 8426756.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.631947422027588,
|
|
"epoch": 3.8512355017650024,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003977100711885616,
|
|
"loss": 5.238,
|
|
"mean_token_accuracy": 0.1708429202437401,
|
|
"num_tokens": 8438052.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.567348003387451,
|
|
"epoch": 3.8562783661119515,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039737739935095614,
|
|
"loss": 5.2242,
|
|
"mean_token_accuracy": 0.16843420416116714,
|
|
"num_tokens": 8448481.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.575160503387451,
|
|
"epoch": 3.8613212304589006,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003970443471615705,
|
|
"loss": 5.2907,
|
|
"mean_token_accuracy": 0.16154980212450026,
|
|
"num_tokens": 8460199.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.603337097167969,
|
|
"epoch": 3.8663640948058497,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003967109156555386,
|
|
"loss": 5.2315,
|
|
"mean_token_accuracy": 0.1721283406019211,
|
|
"num_tokens": 8470842.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.625171375274658,
|
|
"epoch": 3.871406959152799,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039637710586917324,
|
|
"loss": 5.3198,
|
|
"mean_token_accuracy": 0.16095141023397447,
|
|
"num_tokens": 8482631.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.591945838928223,
|
|
"epoch": 3.876449823499748,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00039604291883996287,
|
|
"loss": 5.1773,
|
|
"mean_token_accuracy": 0.1684859037399292,
|
|
"num_tokens": 8495178.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.586579370498657,
|
|
"epoch": 3.8814926878466967,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039570835560656854,
|
|
"loss": 5.1957,
|
|
"mean_token_accuracy": 0.17589065581560134,
|
|
"num_tokens": 8505257.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.584151554107666,
|
|
"epoch": 3.8865355521936458,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003953734172088205,
|
|
"loss": 5.3147,
|
|
"mean_token_accuracy": 0.16295218765735625,
|
|
"num_tokens": 8517124.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.477700996398926,
|
|
"epoch": 3.891578416540595,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00039503810468771494,
|
|
"loss": 5.1784,
|
|
"mean_token_accuracy": 0.1702326253056526,
|
|
"num_tokens": 8528201.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.622996044158936,
|
|
"epoch": 3.896621280887544,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003947024190854111,
|
|
"loss": 5.3182,
|
|
"mean_token_accuracy": 0.16580865383148194,
|
|
"num_tokens": 8539900.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.6353349685668945,
|
|
"epoch": 3.901664145234493,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039436636144522737,
|
|
"loss": 5.2588,
|
|
"mean_token_accuracy": 0.16809331327676774,
|
|
"num_tokens": 8549199.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.577757883071899,
|
|
"epoch": 3.9067070095814422,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039402993281163875,
|
|
"loss": 5.1439,
|
|
"mean_token_accuracy": 0.16727978736162186,
|
|
"num_tokens": 8560626.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.622360181808472,
|
|
"epoch": 3.9117498739283914,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003936931342302732,
|
|
"loss": 5.342,
|
|
"mean_token_accuracy": 0.16621735394001008,
|
|
"num_tokens": 8571706.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.540307474136353,
|
|
"epoch": 3.9167927382753405,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003933559667479084,
|
|
"loss": 5.2593,
|
|
"mean_token_accuracy": 0.16392770111560823,
|
|
"num_tokens": 8583101.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.64545168876648,
|
|
"epoch": 3.9218356026222896,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039301843141246876,
|
|
"loss": 5.3366,
|
|
"mean_token_accuracy": 0.16268704384565352,
|
|
"num_tokens": 8594742.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.639380693435669,
|
|
"epoch": 3.9268784669692387,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00039268052927302185,
|
|
"loss": 5.2268,
|
|
"mean_token_accuracy": 0.16962742656469346,
|
|
"num_tokens": 8606609.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.5454483985900875,
|
|
"epoch": 3.931921331316188,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003923422613797753,
|
|
"loss": 5.2504,
|
|
"mean_token_accuracy": 0.16871908456087112,
|
|
"num_tokens": 8618291.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.645219993591309,
|
|
"epoch": 3.936964195663137,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003920036287840734,
|
|
"loss": 5.2942,
|
|
"mean_token_accuracy": 0.17185911387205124,
|
|
"num_tokens": 8629850.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.557175874710083,
|
|
"epoch": 3.9420070600100856,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00039166463253839434,
|
|
"loss": 5.2889,
|
|
"mean_token_accuracy": 0.1692992016673088,
|
|
"num_tokens": 8640363.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.552072334289551,
|
|
"epoch": 3.9470499243570347,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003913252736963462,
|
|
"loss": 5.2813,
|
|
"mean_token_accuracy": 0.16939892917871474,
|
|
"num_tokens": 8651869.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.615259170532227,
|
|
"epoch": 3.952092788703984,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000390985553312664,
|
|
"loss": 5.1347,
|
|
"mean_token_accuracy": 0.1722140207886696,
|
|
"num_tokens": 8662032.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.585288047790527,
|
|
"epoch": 3.957135653050933,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003906454724432065,
|
|
"loss": 5.1984,
|
|
"mean_token_accuracy": 0.17580096125602723,
|
|
"num_tokens": 8673186.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.560485219955444,
|
|
"epoch": 3.962178517397882,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00039030503214495296,
|
|
"loss": 5.1472,
|
|
"mean_token_accuracy": 0.17943393886089326,
|
|
"num_tokens": 8683176.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.516346788406372,
|
|
"epoch": 3.967221381744831,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0003899642334759996,
|
|
"loss": 5.1966,
|
|
"mean_token_accuracy": 0.17157244235277175,
|
|
"num_tokens": 8694799.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.624952554702759,
|
|
"epoch": 3.97226424609178,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003896230774955567,
|
|
"loss": 5.2613,
|
|
"mean_token_accuracy": 0.17060886174440384,
|
|
"num_tokens": 8706356.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.526438665390015,
|
|
"epoch": 3.977307110438729,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00038928156526394476,
|
|
"loss": 5.1756,
|
|
"mean_token_accuracy": 0.17259202301502227,
|
|
"num_tokens": 8717340.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.499563789367675,
|
|
"epoch": 3.982349974785678,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003889396978425918,
|
|
"loss": 5.1292,
|
|
"mean_token_accuracy": 0.17198775112628936,
|
|
"num_tokens": 8728902.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.540572023391723,
|
|
"epoch": 3.987392839132627,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003885974762940295,
|
|
"loss": 5.201,
|
|
"mean_token_accuracy": 0.1716202676296234,
|
|
"num_tokens": 8739033.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.5737879276275635,
|
|
"epoch": 3.9924357034795763,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003882549016818905,
|
|
"loss": 5.1929,
|
|
"mean_token_accuracy": 0.1709715649485588,
|
|
"num_tokens": 8750107.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.504067993164062,
|
|
"epoch": 3.9974785678265254,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00038791197507090456,
|
|
"loss": 5.2097,
|
|
"mean_token_accuracy": 0.16755173206329346,
|
|
"num_tokens": 8762486.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.504860666063097,
|
|
"epoch": 4.00201714573878,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00038756869752689527,
|
|
"loss": 5.0875,
|
|
"mean_token_accuracy": 0.17902690172195435,
|
|
"num_tokens": 8770890.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.547015142440796,
|
|
"epoch": 4.007060010085729,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003872250701167775,
|
|
"loss": 4.9953,
|
|
"mean_token_accuracy": 0.178935244679451,
|
|
"num_tokens": 8780880.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.54614486694336,
|
|
"epoch": 4.012102874432678,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00038688109390855297,
|
|
"loss": 4.9247,
|
|
"mean_token_accuracy": 0.18966569155454635,
|
|
"num_tokens": 8792616.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.529557561874389,
|
|
"epoch": 4.017145738779627,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003865367699713077,
|
|
"loss": 5.0109,
|
|
"mean_token_accuracy": 0.18950262367725373,
|
|
"num_tokens": 8803630.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.540474700927734,
|
|
"epoch": 4.022188603126576,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038619209937520836,
|
|
"loss": 4.9769,
|
|
"mean_token_accuracy": 0.18979049175977708,
|
|
"num_tokens": 8814824.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.606575536727905,
|
|
"epoch": 4.027231467473525,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038584708319149924,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.18345706760883332,
|
|
"num_tokens": 8824240.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.43360743522644,
|
|
"epoch": 4.032274331820474,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038550172249249856,
|
|
"loss": 4.9255,
|
|
"mean_token_accuracy": 0.1855061873793602,
|
|
"num_tokens": 8835129.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.032274331820474,
|
|
"eval_entropy": 5.3326384165814815,
|
|
"eval_loss": 5.726597785949707,
|
|
"eval_mean_token_accuracy": 0.15028117305146527,
|
|
"eval_num_tokens": 8835129.0,
|
|
"eval_runtime": 2.032,
|
|
"eval_samples_per_second": 1544.281,
|
|
"eval_steps_per_second": 193.404,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.539226388931274,
|
|
"epoch": 4.0373171961674235,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003851560183515953,
|
|
"loss": 4.9432,
|
|
"mean_token_accuracy": 0.19216325134038925,
|
|
"num_tokens": 8846883.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.688214540481567,
|
|
"epoch": 4.042360060514373,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000384809971843246,
|
|
"loss": 5.0739,
|
|
"mean_token_accuracy": 0.18137820214033126,
|
|
"num_tokens": 8857592.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.483003902435303,
|
|
"epoch": 4.047402924861321,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000384463584042971,
|
|
"loss": 4.9369,
|
|
"mean_token_accuracy": 0.19277067929506303,
|
|
"num_tokens": 8868527.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.520571613311768,
|
|
"epoch": 4.05244578920827,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038411685602735175,
|
|
"loss": 4.97,
|
|
"mean_token_accuracy": 0.18508448153734208,
|
|
"num_tokens": 8879416.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.473328638076782,
|
|
"epoch": 4.057488653555219,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038376978887402687,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.18569054156541825,
|
|
"num_tokens": 8889452.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.41745958328247,
|
|
"epoch": 4.062531517902168,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003834223836616891,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.19481220841407776,
|
|
"num_tokens": 8900447.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.603893375396728,
|
|
"epoch": 4.067574382249117,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038307464147008176,
|
|
"loss": 5.0318,
|
|
"mean_token_accuracy": 0.17619509845972062,
|
|
"num_tokens": 8911298.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.449928855895996,
|
|
"epoch": 4.072617246596066,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003827265633799957,
|
|
"loss": 4.9762,
|
|
"mean_token_accuracy": 0.184056793153286,
|
|
"num_tokens": 8922279.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.528772926330566,
|
|
"epoch": 4.0776601109430155,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003823781504732657,
|
|
"loss": 4.9933,
|
|
"mean_token_accuracy": 0.18298246711492538,
|
|
"num_tokens": 8934780.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.584649181365966,
|
|
"epoch": 4.082702975289965,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003820294038327669,
|
|
"loss": 4.9536,
|
|
"mean_token_accuracy": 0.1811712458729744,
|
|
"num_tokens": 8946114.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.456589698791504,
|
|
"epoch": 4.087745839636914,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003816803245424121,
|
|
"loss": 4.9449,
|
|
"mean_token_accuracy": 0.18967065960168839,
|
|
"num_tokens": 8957830.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.506550693511963,
|
|
"epoch": 4.092788703983863,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003813309136871478,
|
|
"loss": 5.0107,
|
|
"mean_token_accuracy": 0.18825370073318481,
|
|
"num_tokens": 8968312.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.488216972351074,
|
|
"epoch": 4.097831568330812,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038098117235295083,
|
|
"loss": 4.9547,
|
|
"mean_token_accuracy": 0.18724346160888672,
|
|
"num_tokens": 8979502.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.509997081756592,
|
|
"epoch": 4.102874432677761,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003806311016268254,
|
|
"loss": 4.9773,
|
|
"mean_token_accuracy": 0.18443292528390884,
|
|
"num_tokens": 8991934.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.5160528182983395,
|
|
"epoch": 4.10791729702471,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003802807025967995,
|
|
"loss": 4.9081,
|
|
"mean_token_accuracy": 0.19007430225610733,
|
|
"num_tokens": 9001669.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.453746366500854,
|
|
"epoch": 4.112960161371659,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003799299763519213,
|
|
"loss": 4.9433,
|
|
"mean_token_accuracy": 0.18578447848558427,
|
|
"num_tokens": 9011904.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.488339900970459,
|
|
"epoch": 4.118003025718608,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003795789239822561,
|
|
"loss": 4.9625,
|
|
"mean_token_accuracy": 0.18474705815315245,
|
|
"num_tokens": 9023024.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.519349765777588,
|
|
"epoch": 4.1230458900655576,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003792275465788826,
|
|
"loss": 5.0835,
|
|
"mean_token_accuracy": 0.17843087464571,
|
|
"num_tokens": 9035721.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.553795719146729,
|
|
"epoch": 4.128088754412507,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037887584523389013,
|
|
"loss": 4.9883,
|
|
"mean_token_accuracy": 0.18789806216955185,
|
|
"num_tokens": 9046541.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.518153667449951,
|
|
"epoch": 4.133131618759456,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037852382104037454,
|
|
"loss": 4.9646,
|
|
"mean_token_accuracy": 0.18353476226329804,
|
|
"num_tokens": 9057696.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.533792543411255,
|
|
"epoch": 4.138174483106404,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037817147509243497,
|
|
"loss": 5.1306,
|
|
"mean_token_accuracy": 0.17284587323665618,
|
|
"num_tokens": 9070737.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.561934280395508,
|
|
"epoch": 4.143217347453353,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003778188084851709,
|
|
"loss": 4.9912,
|
|
"mean_token_accuracy": 0.18854273110628128,
|
|
"num_tokens": 9081582.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.581554651260376,
|
|
"epoch": 4.148260211800302,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00037746582231467816,
|
|
"loss": 5.044,
|
|
"mean_token_accuracy": 0.17540753036737441,
|
|
"num_tokens": 9092571.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.474611139297485,
|
|
"epoch": 4.153303076147251,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003771125176780461,
|
|
"loss": 5.0132,
|
|
"mean_token_accuracy": 0.17701371312141417,
|
|
"num_tokens": 9103172.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.526391553878784,
|
|
"epoch": 4.1583459404942005,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00037675889567335345,
|
|
"loss": 4.9791,
|
|
"mean_token_accuracy": 0.18532492220401764,
|
|
"num_tokens": 9114407.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.478158855438233,
|
|
"epoch": 4.16338880484115,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00037640495739966573,
|
|
"loss": 4.9269,
|
|
"mean_token_accuracy": 0.18747835159301757,
|
|
"num_tokens": 9126175.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.496936368942261,
|
|
"epoch": 4.168431669188099,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00037605070395703124,
|
|
"loss": 5.0761,
|
|
"mean_token_accuracy": 0.17672075629234313,
|
|
"num_tokens": 9139046.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.5157920837402346,
|
|
"epoch": 4.173474533535048,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037569613644647783,
|
|
"loss": 4.9576,
|
|
"mean_token_accuracy": 0.18726904690265656,
|
|
"num_tokens": 9150413.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.571929407119751,
|
|
"epoch": 4.178517397881997,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003753412559700094,
|
|
"loss": 4.9563,
|
|
"mean_token_accuracy": 0.19210157096385955,
|
|
"num_tokens": 9161831.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.467312288284302,
|
|
"epoch": 4.183560262228946,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037498606363060285,
|
|
"loss": 4.9674,
|
|
"mean_token_accuracy": 0.19206145107746125,
|
|
"num_tokens": 9172381.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.369944667816162,
|
|
"epoch": 4.188603126575895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003746305605322041,
|
|
"loss": 4.915,
|
|
"mean_token_accuracy": 0.19153508991003038,
|
|
"num_tokens": 9182212.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.508538675308228,
|
|
"epoch": 4.193645990922844,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037427474777972477,
|
|
"loss": 4.9961,
|
|
"mean_token_accuracy": 0.18655004501342773,
|
|
"num_tokens": 9192207.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.567970180511475,
|
|
"epoch": 4.198688855269793,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00037391862647903943,
|
|
"loss": 5.0636,
|
|
"mean_token_accuracy": 0.18121236711740493,
|
|
"num_tokens": 9203693.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.517024421691895,
|
|
"epoch": 4.2037317196167425,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00037356219773698106,
|
|
"loss": 4.9756,
|
|
"mean_token_accuracy": 0.18997201770544053,
|
|
"num_tokens": 9215201.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.399962377548218,
|
|
"epoch": 4.208774583963692,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037320546266133854,
|
|
"loss": 4.9267,
|
|
"mean_token_accuracy": 0.18798813223838806,
|
|
"num_tokens": 9227705.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.533685874938965,
|
|
"epoch": 4.213817448310641,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003728484223608526,
|
|
"loss": 5.0687,
|
|
"mean_token_accuracy": 0.18384936153888704,
|
|
"num_tokens": 9238372.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.51551661491394,
|
|
"epoch": 4.21886031265759,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00037249107794521286,
|
|
"loss": 5.0059,
|
|
"mean_token_accuracy": 0.17918152809143068,
|
|
"num_tokens": 9250303.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.445101547241211,
|
|
"epoch": 4.223903177004539,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037213343052505394,
|
|
"loss": 5.0713,
|
|
"mean_token_accuracy": 0.1797373652458191,
|
|
"num_tokens": 9261750.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.578331565856933,
|
|
"epoch": 4.228946041351487,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003717754812119523,
|
|
"loss": 5.0012,
|
|
"mean_token_accuracy": 0.18506240993738174,
|
|
"num_tokens": 9273808.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.54116153717041,
|
|
"epoch": 4.233988905698436,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003714172311184226,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.18534486144781112,
|
|
"num_tokens": 9285099.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.527420282363892,
|
|
"epoch": 4.2390317700453854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003710586813579144,
|
|
"loss": 5.081,
|
|
"mean_token_accuracy": 0.176472045481205,
|
|
"num_tokens": 9295513.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.529643297195435,
|
|
"epoch": 4.244074634392335,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037069983304480875,
|
|
"loss": 5.0581,
|
|
"mean_token_accuracy": 0.17323246747255325,
|
|
"num_tokens": 9307243.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.4460673332214355,
|
|
"epoch": 4.249117498739284,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003703406872944144,
|
|
"loss": 5.0115,
|
|
"mean_token_accuracy": 0.18663803488016129,
|
|
"num_tokens": 9319289.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.559715938568115,
|
|
"epoch": 4.254160363086233,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003699812452229645,
|
|
"loss": 5.1124,
|
|
"mean_token_accuracy": 0.18264835476875305,
|
|
"num_tokens": 9332273.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.5281421661376955,
|
|
"epoch": 4.259203227433182,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036962150794761354,
|
|
"loss": 5.0083,
|
|
"mean_token_accuracy": 0.18391186594963074,
|
|
"num_tokens": 9343703.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.424625444412231,
|
|
"epoch": 4.264246091780131,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003692614765864331,
|
|
"loss": 4.988,
|
|
"mean_token_accuracy": 0.1832963854074478,
|
|
"num_tokens": 9355371.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.47719521522522,
|
|
"epoch": 4.26928895612708,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00036890115225840904,
|
|
"loss": 5.0198,
|
|
"mean_token_accuracy": 0.1821829319000244,
|
|
"num_tokens": 9367337.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.565412330627441,
|
|
"epoch": 4.274331820474029,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003685405360834376,
|
|
"loss": 5.0419,
|
|
"mean_token_accuracy": 0.18643008470535277,
|
|
"num_tokens": 9378613.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.547741794586182,
|
|
"epoch": 4.279374684820978,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00036817962918232234,
|
|
"loss": 5.0092,
|
|
"mean_token_accuracy": 0.18363531976938247,
|
|
"num_tokens": 9390576.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.524372625350952,
|
|
"epoch": 4.2844175491679275,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036781843267677,
|
|
"loss": 5.0471,
|
|
"mean_token_accuracy": 0.18128075897693635,
|
|
"num_tokens": 9401828.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.490870809555053,
|
|
"epoch": 4.289460413514877,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003674569476893878,
|
|
"loss": 5.0327,
|
|
"mean_token_accuracy": 0.18399339765310288,
|
|
"num_tokens": 9414065.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.525549697875976,
|
|
"epoch": 4.294503277861826,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003670951753436794,
|
|
"loss": 5.0399,
|
|
"mean_token_accuracy": 0.18734060376882553,
|
|
"num_tokens": 9425025.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.483308267593384,
|
|
"epoch": 4.299546142208775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036673311676404163,
|
|
"loss": 4.9936,
|
|
"mean_token_accuracy": 0.18617091774940492,
|
|
"num_tokens": 9435562.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.528381824493408,
|
|
"epoch": 4.304589006555724,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003663707730757608,
|
|
"loss": 4.9923,
|
|
"mean_token_accuracy": 0.17679995745420457,
|
|
"num_tokens": 9445963.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.443905448913574,
|
|
"epoch": 4.309631870902673,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036600814540500956,
|
|
"loss": 4.9593,
|
|
"mean_token_accuracy": 0.1831197440624237,
|
|
"num_tokens": 9456810.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.42639651298523,
|
|
"epoch": 4.314674735249622,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000365645234878843,
|
|
"loss": 4.9882,
|
|
"mean_token_accuracy": 0.1854613497853279,
|
|
"num_tokens": 9466527.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.490806579589844,
|
|
"epoch": 4.31971759959657,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00036528204262519545,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.1850216194987297,
|
|
"num_tokens": 9477770.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.532266759872437,
|
|
"epoch": 4.3247604639435195,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003649185697728768,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.1845906123518944,
|
|
"num_tokens": 9487723.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.52371392250061,
|
|
"epoch": 4.329803328290469,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036455481745156906,
|
|
"loss": 4.9601,
|
|
"mean_token_accuracy": 0.1868498608469963,
|
|
"num_tokens": 9497286.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.387016916275025,
|
|
"epoch": 4.334846192637418,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003641907867918228,
|
|
"loss": 4.9161,
|
|
"mean_token_accuracy": 0.1925663098692894,
|
|
"num_tokens": 9508535.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.4750732421875,
|
|
"epoch": 4.339889056984367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003638264789250538,
|
|
"loss": 4.9911,
|
|
"mean_token_accuracy": 0.1869424894452095,
|
|
"num_tokens": 9518165.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.508026742935181,
|
|
"epoch": 4.344931921331316,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036346189498353914,
|
|
"loss": 4.995,
|
|
"mean_token_accuracy": 0.17981656938791274,
|
|
"num_tokens": 9528197.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.51187686920166,
|
|
"epoch": 4.349974785678265,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003630970361004143,
|
|
"loss": 5.003,
|
|
"mean_token_accuracy": 0.19398315101861954,
|
|
"num_tokens": 9539024.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.51695146560669,
|
|
"epoch": 4.355017650025214,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003627319034096691,
|
|
"loss": 4.9768,
|
|
"mean_token_accuracy": 0.19208788871765137,
|
|
"num_tokens": 9548701.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.36809196472168,
|
|
"epoch": 4.360060514372163,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003623664980461443,
|
|
"loss": 4.9388,
|
|
"mean_token_accuracy": 0.19122428596019744,
|
|
"num_tokens": 9558576.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.482896232604981,
|
|
"epoch": 4.3651033787191125,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00036200082114552806,
|
|
"loss": 5.004,
|
|
"mean_token_accuracy": 0.18354242146015168,
|
|
"num_tokens": 9567464.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.512559986114502,
|
|
"epoch": 4.370146243066062,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036163487384435294,
|
|
"loss": 4.9394,
|
|
"mean_token_accuracy": 0.191547292470932,
|
|
"num_tokens": 9578011.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.546701431274414,
|
|
"epoch": 4.375189107413011,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003612686572799914,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.17945659905672073,
|
|
"num_tokens": 9589279.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.538714647293091,
|
|
"epoch": 4.38023197175996,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003609021725906529,
|
|
"loss": 4.9824,
|
|
"mean_token_accuracy": 0.18380559235811234,
|
|
"num_tokens": 9600170.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.452891826629639,
|
|
"epoch": 4.385274836106909,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00036053542091538045,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.19126289635896682,
|
|
"num_tokens": 9611006.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.48802342414856,
|
|
"epoch": 4.390317700453858,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036016840339404664,
|
|
"loss": 5.0516,
|
|
"mean_token_accuracy": 0.18311115503311157,
|
|
"num_tokens": 9621645.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.502034997940063,
|
|
"epoch": 4.395360564800807,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003598011211673505,
|
|
"loss": 5.0572,
|
|
"mean_token_accuracy": 0.1818081632256508,
|
|
"num_tokens": 9633221.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.548558139801026,
|
|
"epoch": 4.400403429147756,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003594335753768135,
|
|
"loss": 5.0519,
|
|
"mean_token_accuracy": 0.18390922546386718,
|
|
"num_tokens": 9644616.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.500851154327393,
|
|
"epoch": 4.405446293494705,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003590657671647766,
|
|
"loss": 4.991,
|
|
"mean_token_accuracy": 0.1860932633280754,
|
|
"num_tokens": 9656072.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.451617431640625,
|
|
"epoch": 4.4104891578416545,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00035869769767439606,
|
|
"loss": 5.0359,
|
|
"mean_token_accuracy": 0.17706296294927598,
|
|
"num_tokens": 9667252.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.573953008651733,
|
|
"epoch": 4.415532022188603,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003583293680496404,
|
|
"loss": 5.0765,
|
|
"mean_token_accuracy": 0.1832370176911354,
|
|
"num_tokens": 9677810.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.518440532684326,
|
|
"epoch": 4.420574886535552,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003579607794352867,
|
|
"loss": 5.0322,
|
|
"mean_token_accuracy": 0.1833367243409157,
|
|
"num_tokens": 9688950.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.544857931137085,
|
|
"epoch": 4.425617750882501,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00035759193297691674,
|
|
"loss": 5.0529,
|
|
"mean_token_accuracy": 0.17716529220342636,
|
|
"num_tokens": 9699961.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.466248273849487,
|
|
"epoch": 4.43066061522945,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00035722282982091373,
|
|
"loss": 5.015,
|
|
"mean_token_accuracy": 0.18353263139724732,
|
|
"num_tokens": 9711548.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.519790983200073,
|
|
"epoch": 4.435703479576399,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003568534711144591,
|
|
"loss": 5.0491,
|
|
"mean_token_accuracy": 0.18694446831941605,
|
|
"num_tokens": 9722237.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.512378454208374,
|
|
"epoch": 4.440746343923348,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000356483858005528,
|
|
"loss": 5.0576,
|
|
"mean_token_accuracy": 0.1866755500435829,
|
|
"num_tokens": 9733654.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.443044757843017,
|
|
"epoch": 4.445789208270297,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00035611399164288654,
|
|
"loss": 4.9132,
|
|
"mean_token_accuracy": 0.1905991852283478,
|
|
"num_tokens": 9745012.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.465333318710327,
|
|
"epoch": 4.450832072617247,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003557438731760879,
|
|
"loss": 4.9902,
|
|
"mean_token_accuracy": 0.1781729832291603,
|
|
"num_tokens": 9755851.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.472749805450439,
|
|
"epoch": 4.455874936964196,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003553735037554689,
|
|
"loss": 4.9268,
|
|
"mean_token_accuracy": 0.18698177188634874,
|
|
"num_tokens": 9765652.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.526891469955444,
|
|
"epoch": 4.460917801311145,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003550028845321461,
|
|
"loss": 4.9929,
|
|
"mean_token_accuracy": 0.18309144228696822,
|
|
"num_tokens": 9776618.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.4950353622436525,
|
|
"epoch": 4.465960665658094,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003546320166580127,
|
|
"loss": 4.9565,
|
|
"mean_token_accuracy": 0.18619473725557328,
|
|
"num_tokens": 9786523.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.5177699565887455,
|
|
"epoch": 4.471003530005043,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035426090128573464,
|
|
"loss": 5.0376,
|
|
"mean_token_accuracy": 0.1834830731153488,
|
|
"num_tokens": 9797013.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.523991298675537,
|
|
"epoch": 4.476046394351992,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00035388953956874686,
|
|
"loss": 5.0358,
|
|
"mean_token_accuracy": 0.1820090815424919,
|
|
"num_tokens": 9807431.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.492027044296265,
|
|
"epoch": 4.481089258698941,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003535179326612504,
|
|
"loss": 5.0585,
|
|
"mean_token_accuracy": 0.1843281388282776,
|
|
"num_tokens": 9817560.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.489924192428589,
|
|
"epoch": 4.48613212304589,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00035314608171820787,
|
|
"loss": 4.9647,
|
|
"mean_token_accuracy": 0.1837001472711563,
|
|
"num_tokens": 9828233.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.455750513076782,
|
|
"epoch": 4.4911749873928395,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00035277398789534075,
|
|
"loss": 4.9899,
|
|
"mean_token_accuracy": 0.18740071803331376,
|
|
"num_tokens": 9838699.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.471644067764283,
|
|
"epoch": 4.496217851739789,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003524016523491252,
|
|
"loss": 5.0176,
|
|
"mean_token_accuracy": 0.17613862007856368,
|
|
"num_tokens": 9849635.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.49004898071289,
|
|
"epoch": 4.501260716086737,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00035202907623678865,
|
|
"loss": 4.9488,
|
|
"mean_token_accuracy": 0.18466950505971907,
|
|
"num_tokens": 9861822.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.481436109542846,
|
|
"epoch": 4.506303580433686,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003516562607163062,
|
|
"loss": 5.0238,
|
|
"mean_token_accuracy": 0.18602396100759505,
|
|
"num_tokens": 9872341.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.562046670913697,
|
|
"epoch": 4.511346444780635,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003512832069463971,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.17845325618982316,
|
|
"num_tokens": 9883078.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.460926914215088,
|
|
"epoch": 4.516389309127584,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00035090991608652125,
|
|
"loss": 5.1115,
|
|
"mean_token_accuracy": 0.17897260189056396,
|
|
"num_tokens": 9894688.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.434860038757324,
|
|
"epoch": 4.521432173474533,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035053638929687504,
|
|
"loss": 4.9177,
|
|
"mean_token_accuracy": 0.18693222254514694,
|
|
"num_tokens": 9904409.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.5169844150543215,
|
|
"epoch": 4.526475037821482,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003501626277383884,
|
|
"loss": 5.1095,
|
|
"mean_token_accuracy": 0.17296307235956193,
|
|
"num_tokens": 9915204.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.572116088867188,
|
|
"epoch": 4.5315179021684315,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000349788632572721,
|
|
"loss": 5.0262,
|
|
"mean_token_accuracy": 0.1805413156747818,
|
|
"num_tokens": 9927354.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.435087203979492,
|
|
"epoch": 4.536560766515381,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003494144049622582,
|
|
"loss": 4.9501,
|
|
"mean_token_accuracy": 0.1838911220431328,
|
|
"num_tokens": 9939009.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.44829568862915,
|
|
"epoch": 4.54160363086233,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003490399460701083,
|
|
"loss": 5.0559,
|
|
"mean_token_accuracy": 0.18399010002613067,
|
|
"num_tokens": 9950332.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.488559341430664,
|
|
"epoch": 4.546646495209279,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000348665257060098,
|
|
"loss": 4.9616,
|
|
"mean_token_accuracy": 0.18975978642702102,
|
|
"num_tokens": 9961206.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.50782036781311,
|
|
"epoch": 4.551689359556228,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003482903390967695,
|
|
"loss": 4.963,
|
|
"mean_token_accuracy": 0.1917782485485077,
|
|
"num_tokens": 9972272.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.396845531463623,
|
|
"epoch": 4.556732223903177,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034791519334537635,
|
|
"loss": 4.9784,
|
|
"mean_token_accuracy": 0.18735261410474777,
|
|
"num_tokens": 9982189.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.449600124359131,
|
|
"epoch": 4.561775088250126,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003475398209718804,
|
|
"loss": 4.9252,
|
|
"mean_token_accuracy": 0.18916368931531907,
|
|
"num_tokens": 9993122.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.49206657409668,
|
|
"epoch": 4.566817952597075,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003471642231429475,
|
|
"loss": 4.9398,
|
|
"mean_token_accuracy": 0.19178151339292526,
|
|
"num_tokens": 10004043.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.459736442565918,
|
|
"epoch": 4.5718608169440245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003467884010259443,
|
|
"loss": 4.9554,
|
|
"mean_token_accuracy": 0.19185522049665452,
|
|
"num_tokens": 10014042.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.3617125988006595,
|
|
"epoch": 4.576903681290974,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00034641235578893476,
|
|
"loss": 4.9069,
|
|
"mean_token_accuracy": 0.19162076115608215,
|
|
"num_tokens": 10025742.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.444120597839356,
|
|
"epoch": 4.581946545637923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003460360886006761,
|
|
"loss": 5.0016,
|
|
"mean_token_accuracy": 0.1868985116481781,
|
|
"num_tokens": 10036727.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.513584995269776,
|
|
"epoch": 4.586989409984872,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034565960063061534,
|
|
"loss": 4.9649,
|
|
"mean_token_accuracy": 0.17937788516283035,
|
|
"num_tokens": 10047246.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.412352085113525,
|
|
"epoch": 4.592032274331821,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003452828930488859,
|
|
"loss": 4.977,
|
|
"mean_token_accuracy": 0.18880498707294463,
|
|
"num_tokens": 10059750.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.476818704605103,
|
|
"epoch": 4.597075138678769,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00034490596702630355,
|
|
"loss": 5.0245,
|
|
"mean_token_accuracy": 0.18169016838073732,
|
|
"num_tokens": 10071431.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.505073595046997,
|
|
"epoch": 4.602118003025718,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 5.011,
|
|
"mean_token_accuracy": 0.18945524990558624,
|
|
"num_tokens": 10081567.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.541516065597534,
|
|
"epoch": 4.607160867372667,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00034415146434523463,
|
|
"loss": 5.0594,
|
|
"mean_token_accuracy": 0.18242636919021607,
|
|
"num_tokens": 10092506.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.508158111572266,
|
|
"epoch": 4.6122037317196165,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034377389003175984,
|
|
"loss": 5.028,
|
|
"mean_token_accuracy": 0.18220201581716539,
|
|
"num_tokens": 10103896.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.420201778411865,
|
|
"epoch": 4.617246596066566,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003433961019674483,
|
|
"loss": 5.0091,
|
|
"mean_token_accuracy": 0.18821532428264617,
|
|
"num_tokens": 10113646.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.435951232910156,
|
|
"epoch": 4.622289460413515,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003430181013264741,
|
|
"loss": 5.0545,
|
|
"mean_token_accuracy": 0.1781825065612793,
|
|
"num_tokens": 10125097.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.51766357421875,
|
|
"epoch": 4.627332324760464,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003426398892836719,
|
|
"loss": 5.0156,
|
|
"mean_token_accuracy": 0.17955711781978606,
|
|
"num_tokens": 10137396.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.530993509292602,
|
|
"epoch": 4.632375189107413,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003422614670145336,
|
|
"loss": 5.121,
|
|
"mean_token_accuracy": 0.17590828686952592,
|
|
"num_tokens": 10147483.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.475261163711548,
|
|
"epoch": 4.637418053454362,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003418828356952041,
|
|
"loss": 5.0105,
|
|
"mean_token_accuracy": 0.17809143662452698,
|
|
"num_tokens": 10159693.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.4684546947479244,
|
|
"epoch": 4.642460917801311,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034150399650247846,
|
|
"loss": 4.8984,
|
|
"mean_token_accuracy": 0.18903608173131942,
|
|
"num_tokens": 10170371.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.441571235656738,
|
|
"epoch": 4.64750378214826,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00034112495061379764,
|
|
"loss": 4.9923,
|
|
"mean_token_accuracy": 0.19029690474271774,
|
|
"num_tokens": 10181487.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.470997095108032,
|
|
"epoch": 4.652546646495209,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034074569920724483,
|
|
"loss": 5.0176,
|
|
"mean_token_accuracy": 0.18029228150844573,
|
|
"num_tokens": 10193386.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.552959537506103,
|
|
"epoch": 4.6575895108421586,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003403662434615423,
|
|
"loss": 5.0873,
|
|
"mean_token_accuracy": 0.18091829121112823,
|
|
"num_tokens": 10205531.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.503431558609009,
|
|
"epoch": 4.662632375189108,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00033998658455604736,
|
|
"loss": 5.0558,
|
|
"mean_token_accuracy": 0.18431001305580139,
|
|
"num_tokens": 10216586.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.477294254302978,
|
|
"epoch": 4.667675239536057,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033960672367074827,
|
|
"loss": 4.9512,
|
|
"mean_token_accuracy": 0.18683320730924607,
|
|
"num_tokens": 10227330.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.467446899414062,
|
|
"epoch": 4.672718103883006,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003392266619862619,
|
|
"loss": 5.077,
|
|
"mean_token_accuracy": 0.1728449732065201,
|
|
"num_tokens": 10238295.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.485948705673218,
|
|
"epoch": 4.677760968229954,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003388464006838284,
|
|
"loss": 5.048,
|
|
"mean_token_accuracy": 0.18787312060594558,
|
|
"num_tokens": 10249011.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.533275318145752,
|
|
"epoch": 4.682803832576903,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033846594094530874,
|
|
"loss": 4.9934,
|
|
"mean_token_accuracy": 0.18459066301584243,
|
|
"num_tokens": 10260986.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.465487003326416,
|
|
"epoch": 4.687846696923852,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003380852839531806,
|
|
"loss": 5.0169,
|
|
"mean_token_accuracy": 0.1808803305029869,
|
|
"num_tokens": 10271557.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.546116828918457,
|
|
"epoch": 4.6928895612708015,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033770443089053476,
|
|
"loss": 5.029,
|
|
"mean_token_accuracy": 0.18636609762907028,
|
|
"num_tokens": 10281458.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.4822780132293705,
|
|
"epoch": 4.697932425617751,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00033732338294107116,
|
|
"loss": 5.0136,
|
|
"mean_token_accuracy": 0.18492239117622375,
|
|
"num_tokens": 10293336.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.428778791427613,
|
|
"epoch": 4.7029752899647,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033694214128909567,
|
|
"loss": 5.0164,
|
|
"mean_token_accuracy": 0.17988018691539764,
|
|
"num_tokens": 10303633.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.475651597976684,
|
|
"epoch": 4.708018154311649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033656070711951633,
|
|
"loss": 5.0174,
|
|
"mean_token_accuracy": 0.18051771968603134,
|
|
"num_tokens": 10314776.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.460410165786743,
|
|
"epoch": 4.713061018658598,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003361790816178391,
|
|
"loss": 4.9704,
|
|
"mean_token_accuracy": 0.18928206264972686,
|
|
"num_tokens": 10325792.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.4019224643707275,
|
|
"epoch": 4.718103883005547,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003357972659701649,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.17767725735902787,
|
|
"num_tokens": 10339024.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.494750070571899,
|
|
"epoch": 4.723146747352496,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033541526136318564,
|
|
"loss": 5.012,
|
|
"mean_token_accuracy": 0.18619493246078492,
|
|
"num_tokens": 10349198.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.5273853778839115,
|
|
"epoch": 4.728189611699445,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003350330689841804,
|
|
"loss": 5.0843,
|
|
"mean_token_accuracy": 0.1764678657054901,
|
|
"num_tokens": 10360925.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.5187419891357425,
|
|
"epoch": 4.733232476046394,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003346506900210118,
|
|
"loss": 5.0267,
|
|
"mean_token_accuracy": 0.1847263902425766,
|
|
"num_tokens": 10372074.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.431736326217651,
|
|
"epoch": 4.7382753403933435,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003342681256621227,
|
|
"loss": 4.9168,
|
|
"mean_token_accuracy": 0.19285121709108352,
|
|
"num_tokens": 10384050.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.504525804519654,
|
|
"epoch": 4.743318204740293,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003338853770965319,
|
|
"loss": 4.9877,
|
|
"mean_token_accuracy": 0.17984011471271516,
|
|
"num_tokens": 10394466.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.47388277053833,
|
|
"epoch": 4.748361069087242,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033350244551383064,
|
|
"loss": 5.0447,
|
|
"mean_token_accuracy": 0.18421263992786407,
|
|
"num_tokens": 10405279.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.432354688644409,
|
|
"epoch": 4.753403933434191,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003331193321041793,
|
|
"loss": 4.9932,
|
|
"mean_token_accuracy": 0.18496614098548889,
|
|
"num_tokens": 10417504.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.432405281066894,
|
|
"epoch": 4.75844679778114,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00033273603805830296,
|
|
"loss": 5.0242,
|
|
"mean_token_accuracy": 0.1841118261218071,
|
|
"num_tokens": 10428824.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.502999925613404,
|
|
"epoch": 4.763489662128089,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033235256456748847,
|
|
"loss": 5.0685,
|
|
"mean_token_accuracy": 0.180811707675457,
|
|
"num_tokens": 10440164.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.548851442337036,
|
|
"epoch": 4.768532526475038,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003319689128235804,
|
|
"loss": 4.9868,
|
|
"mean_token_accuracy": 0.18743729889392852,
|
|
"num_tokens": 10450216.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.518535089492798,
|
|
"epoch": 4.773575390821987,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00033158508401897707,
|
|
"loss": 4.9972,
|
|
"mean_token_accuracy": 0.1878269076347351,
|
|
"num_tokens": 10461521.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.469465827941894,
|
|
"epoch": 4.778618255168936,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003312010793466273,
|
|
"loss": 5.1089,
|
|
"mean_token_accuracy": 0.17105515599250792,
|
|
"num_tokens": 10473052.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.439676952362061,
|
|
"epoch": 4.783661119515885,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003308169000000265,
|
|
"loss": 4.9654,
|
|
"mean_token_accuracy": 0.19206648021936418,
|
|
"num_tokens": 10483197.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.500624084472657,
|
|
"epoch": 4.788703983862834,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003304325471732129,
|
|
"loss": 4.9691,
|
|
"mean_token_accuracy": 0.18224557340145112,
|
|
"num_tokens": 10493877.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.470517253875732,
|
|
"epoch": 4.793746848209783,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00033004802206076396,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.18219218105077745,
|
|
"num_tokens": 10505457.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.46956148147583,
|
|
"epoch": 4.798789712556732,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032966332585779263,
|
|
"loss": 5.0602,
|
|
"mean_token_accuracy": 0.18441519290208816,
|
|
"num_tokens": 10516746.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.526749610900879,
|
|
"epoch": 4.803832576903681,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032927845975994356,
|
|
"loss": 5.0191,
|
|
"mean_token_accuracy": 0.18645018488168716,
|
|
"num_tokens": 10527879.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.507454919815063,
|
|
"epoch": 4.80887544125063,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032889342496338953,
|
|
"loss": 5.0895,
|
|
"mean_token_accuracy": 0.1788397803902626,
|
|
"num_tokens": 10539729.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.469819164276123,
|
|
"epoch": 4.813918305597579,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003285082226648273,
|
|
"loss": 5.1034,
|
|
"mean_token_accuracy": 0.18164454251527787,
|
|
"num_tokens": 10551343.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.49449725151062,
|
|
"epoch": 4.8189611699445285,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003281228540614749,
|
|
"loss": 5.0047,
|
|
"mean_token_accuracy": 0.1870117962360382,
|
|
"num_tokens": 10562362.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.481515312194825,
|
|
"epoch": 4.824004034291478,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032773732035106657,
|
|
"loss": 5.0015,
|
|
"mean_token_accuracy": 0.18595098555088044,
|
|
"num_tokens": 10574203.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.504070711135864,
|
|
"epoch": 4.829046898638427,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032735162273185006,
|
|
"loss": 5.0395,
|
|
"mean_token_accuracy": 0.17628868967294692,
|
|
"num_tokens": 10585732.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.542615747451782,
|
|
"epoch": 4.834089762985376,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003269657624025824,
|
|
"loss": 5.0725,
|
|
"mean_token_accuracy": 0.18423852026462556,
|
|
"num_tokens": 10596414.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.511446905136109,
|
|
"epoch": 4.839132627332325,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003265797405625265,
|
|
"loss": 4.9827,
|
|
"mean_token_accuracy": 0.1850873351097107,
|
|
"num_tokens": 10608893.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.391744184494018,
|
|
"epoch": 4.844175491679274,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003261935584114471,
|
|
"loss": 4.9088,
|
|
"mean_token_accuracy": 0.18824505656957627,
|
|
"num_tokens": 10619229.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.4464919090271,
|
|
"epoch": 4.849218356026223,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003258072171496072,
|
|
"loss": 5.0337,
|
|
"mean_token_accuracy": 0.17979249358177185,
|
|
"num_tokens": 10630811.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.496828556060791,
|
|
"epoch": 4.854261220373172,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003254207179777644,
|
|
"loss": 5.0284,
|
|
"mean_token_accuracy": 0.18516197353601455,
|
|
"num_tokens": 10641191.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.491752099990845,
|
|
"epoch": 4.8593040847201205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000325034062097167,
|
|
"loss": 5.0847,
|
|
"mean_token_accuracy": 0.18095747679471968,
|
|
"num_tokens": 10652019.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.484639120101929,
|
|
"epoch": 4.86434694906707,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032464725070955046,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.175615131855011,
|
|
"num_tokens": 10664820.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.530506801605225,
|
|
"epoch": 4.869389813414019,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032426028501713347,
|
|
"loss": 5.0235,
|
|
"mean_token_accuracy": 0.18118857741355895,
|
|
"num_tokens": 10676343.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.486395931243896,
|
|
"epoch": 4.874432677760968,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032387316622261425,
|
|
"loss": 5.0531,
|
|
"mean_token_accuracy": 0.1816219598054886,
|
|
"num_tokens": 10687464.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.45659384727478,
|
|
"epoch": 4.879475542107917,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032348589552916703,
|
|
"loss": 4.996,
|
|
"mean_token_accuracy": 0.18037255555391313,
|
|
"num_tokens": 10698796.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.452222681045532,
|
|
"epoch": 4.884518406454866,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003230984741404381,
|
|
"loss": 5.0022,
|
|
"mean_token_accuracy": 0.18359864503145218,
|
|
"num_tokens": 10708759.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.564523506164551,
|
|
"epoch": 4.889561270801815,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003227109032605419,
|
|
"loss": 5.0498,
|
|
"mean_token_accuracy": 0.1842788890004158,
|
|
"num_tokens": 10718518.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.537766695022583,
|
|
"epoch": 4.894604135148764,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003223231840940578,
|
|
"loss": 5.0933,
|
|
"mean_token_accuracy": 0.1779206320643425,
|
|
"num_tokens": 10730091.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.421212673187256,
|
|
"epoch": 4.8996469994957135,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032193531784602595,
|
|
"loss": 5.0338,
|
|
"mean_token_accuracy": 0.18051906377077104,
|
|
"num_tokens": 10739664.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.462425899505615,
|
|
"epoch": 4.904689863842663,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032154730572194346,
|
|
"loss": 4.9839,
|
|
"mean_token_accuracy": 0.18871878683567048,
|
|
"num_tokens": 10749573.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.5808899879455565,
|
|
"epoch": 4.909732728189612,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000321159148927761,
|
|
"loss": 5.1064,
|
|
"mean_token_accuracy": 0.18219306617975234,
|
|
"num_tokens": 10760455.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.545758867263794,
|
|
"epoch": 4.914775592536561,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000320770848669879,
|
|
"loss": 5.0735,
|
|
"mean_token_accuracy": 0.17791444659233094,
|
|
"num_tokens": 10771845.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.440776395797729,
|
|
"epoch": 4.91981845688351,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032038240615514345,
|
|
"loss": 5.022,
|
|
"mean_token_accuracy": 0.18617533445358275,
|
|
"num_tokens": 10783812.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.488300704956055,
|
|
"epoch": 4.924861321230459,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003199938225908426,
|
|
"loss": 4.9668,
|
|
"mean_token_accuracy": 0.18300664573907852,
|
|
"num_tokens": 10794390.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.480578517913818,
|
|
"epoch": 4.929904185577408,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003196050991847033,
|
|
"loss": 5.0115,
|
|
"mean_token_accuracy": 0.1819770187139511,
|
|
"num_tokens": 10805521.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.448656415939331,
|
|
"epoch": 4.934947049924357,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003192162371448868,
|
|
"loss": 5.0077,
|
|
"mean_token_accuracy": 0.18004132062196732,
|
|
"num_tokens": 10816265.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.450580167770386,
|
|
"epoch": 4.939989914271306,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031882723767998517,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.18321274667978288,
|
|
"num_tokens": 10826285.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.456393718719482,
|
|
"epoch": 4.9450327786182555,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00031843810199901795,
|
|
"loss": 4.9817,
|
|
"mean_token_accuracy": 0.1874566450715065,
|
|
"num_tokens": 10836817.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.47975263595581,
|
|
"epoch": 4.950075642965205,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003180488313114277,
|
|
"loss": 5.0022,
|
|
"mean_token_accuracy": 0.18786964267492295,
|
|
"num_tokens": 10848186.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.396901988983155,
|
|
"epoch": 4.955118507312154,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003176594268270766,
|
|
"loss": 5.0125,
|
|
"mean_token_accuracy": 0.18755416125059127,
|
|
"num_tokens": 10858851.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.506841707229614,
|
|
"epoch": 4.960161371659103,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00031726988975624293,
|
|
"loss": 5.0532,
|
|
"mean_token_accuracy": 0.18426744341850282,
|
|
"num_tokens": 10870242.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.482984590530395,
|
|
"epoch": 4.965204236006051,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031688022130961675,
|
|
"loss": 4.9781,
|
|
"mean_token_accuracy": 0.18386798352003098,
|
|
"num_tokens": 10881580.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.467860555648803,
|
|
"epoch": 4.970247100353,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003164904226982967,
|
|
"loss": 5.0395,
|
|
"mean_token_accuracy": 0.1865370824933052,
|
|
"num_tokens": 10893140.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.506410217285156,
|
|
"epoch": 4.975289964699949,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00031610049513378574,
|
|
"loss": 5.0336,
|
|
"mean_token_accuracy": 0.1879288896918297,
|
|
"num_tokens": 10904373.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.575470209121704,
|
|
"epoch": 4.9803328290468984,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031571043982798786,
|
|
"loss": 5.073,
|
|
"mean_token_accuracy": 0.18261750787496567,
|
|
"num_tokens": 10914155.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.516875410079956,
|
|
"epoch": 4.985375693393848,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003153202579932039,
|
|
"loss": 5.0538,
|
|
"mean_token_accuracy": 0.1847434476017952,
|
|
"num_tokens": 10926325.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.540553092956543,
|
|
"epoch": 4.990418557740797,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00031492995084212794,
|
|
"loss": 5.0917,
|
|
"mean_token_accuracy": 0.18534403741359712,
|
|
"num_tokens": 10938799.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.500184202194214,
|
|
"epoch": 4.995461422087746,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003145395195878437,
|
|
"loss": 4.8968,
|
|
"mean_token_accuracy": 0.18723651468753816,
|
|
"num_tokens": 10949918.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.423155413733588,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 0.00031414896544382047,
|
|
"loss": 4.905,
|
|
"mean_token_accuracy": 0.18644427590899998,
|
|
"num_tokens": 10958200.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.420553588867188,
|
|
"epoch": 5.005042864346949,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003137582896239095,
|
|
"loss": 4.8167,
|
|
"mean_token_accuracy": 0.1957211285829544,
|
|
"num_tokens": 10969020.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.411706924438477,
|
|
"epoch": 5.010085728693898,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003133674933423404,
|
|
"loss": 4.7441,
|
|
"mean_token_accuracy": 0.20455391108989715,
|
|
"num_tokens": 10978191.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.43629264831543,
|
|
"epoch": 5.015128593040847,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00031297657781371687,
|
|
"loss": 4.8202,
|
|
"mean_token_accuracy": 0.20507970452308655,
|
|
"num_tokens": 10989022.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.518202257156372,
|
|
"epoch": 5.0201714573877965,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003125855442530135,
|
|
"loss": 4.8736,
|
|
"mean_token_accuracy": 0.19305703788995743,
|
|
"num_tokens": 11002727.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.480440282821656,
|
|
"epoch": 5.025214321734746,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00031219439387557155,
|
|
"loss": 4.8215,
|
|
"mean_token_accuracy": 0.1949317440390587,
|
|
"num_tokens": 11013723.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.472077703475952,
|
|
"epoch": 5.030257186081695,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031180312789709535,
|
|
"loss": 4.7827,
|
|
"mean_token_accuracy": 0.2025404617190361,
|
|
"num_tokens": 11024676.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.449752902984619,
|
|
"epoch": 5.035300050428644,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003114117475336487,
|
|
"loss": 4.8321,
|
|
"mean_token_accuracy": 0.19745156913995743,
|
|
"num_tokens": 11036314.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.463897800445556,
|
|
"epoch": 5.040342914775593,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003110202540016509,
|
|
"loss": 4.8121,
|
|
"mean_token_accuracy": 0.19937098771333694,
|
|
"num_tokens": 11046321.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.040342914775593,
|
|
"eval_entropy": 5.316186170844935,
|
|
"eval_loss": 5.684955596923828,
|
|
"eval_mean_token_accuracy": 0.15281682478562567,
|
|
"eval_num_tokens": 11046321.0,
|
|
"eval_runtime": 2.0275,
|
|
"eval_samples_per_second": 1547.745,
|
|
"eval_steps_per_second": 193.838,
|
|
"step": 5000
|
|
}
|
|
],
|
|
"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": 2731819559731200.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|