65064b0c3b
Model: fpadovani/jpn-jpan-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
15200 lines
415 KiB
JSON
15200 lines
415 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 8.771929824561404,
|
|
"eval_steps": 500,
|
|
"global_step": 7500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.593726110458374,
|
|
"epoch": 0.005847953216374269,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.6031,
|
|
"mean_token_accuracy": 0.13570500910282135,
|
|
"num_tokens": 12656.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.586997556686401,
|
|
"epoch": 0.011695906432748537,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.6652,
|
|
"mean_token_accuracy": 0.12274946123361588,
|
|
"num_tokens": 25207.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.649361896514892,
|
|
"epoch": 0.017543859649122806,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.673,
|
|
"mean_token_accuracy": 0.12404287382960319,
|
|
"num_tokens": 38096.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.637357187271118,
|
|
"epoch": 0.023391812865497075,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.7506,
|
|
"mean_token_accuracy": 0.12453248649835587,
|
|
"num_tokens": 50626.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.673450565338134,
|
|
"epoch": 0.029239766081871343,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.6825,
|
|
"mean_token_accuracy": 0.12940182387828827,
|
|
"num_tokens": 63448.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.774927949905395,
|
|
"epoch": 0.03508771929824561,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.731,
|
|
"mean_token_accuracy": 0.12124454602599144,
|
|
"num_tokens": 75656.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.754840278625489,
|
|
"epoch": 0.04093567251461988,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.7155,
|
|
"mean_token_accuracy": 0.12517715990543365,
|
|
"num_tokens": 88756.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.778929853439331,
|
|
"epoch": 0.04678362573099415,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.6499,
|
|
"mean_token_accuracy": 0.12550911605358123,
|
|
"num_tokens": 100934.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.761408805847168,
|
|
"epoch": 0.05263157894736842,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.6072,
|
|
"mean_token_accuracy": 0.12898701280355454,
|
|
"num_tokens": 112212.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.740135717391968,
|
|
"epoch": 0.05847953216374269,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.6067,
|
|
"mean_token_accuracy": 0.12694446071982385,
|
|
"num_tokens": 126314.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.718768978118897,
|
|
"epoch": 0.06432748538011696,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.652,
|
|
"mean_token_accuracy": 0.13094937354326247,
|
|
"num_tokens": 138416.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.7235801219940186,
|
|
"epoch": 0.07017543859649122,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.6356,
|
|
"mean_token_accuracy": 0.12992877885699272,
|
|
"num_tokens": 149903.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.728446578979492,
|
|
"epoch": 0.07602339181286549,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.616,
|
|
"mean_token_accuracy": 0.13435307145118713,
|
|
"num_tokens": 163516.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.745584630966187,
|
|
"epoch": 0.08187134502923976,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.6154,
|
|
"mean_token_accuracy": 0.12847506403923034,
|
|
"num_tokens": 176024.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.79000825881958,
|
|
"epoch": 0.08771929824561403,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.6904,
|
|
"mean_token_accuracy": 0.12912894561886787,
|
|
"num_tokens": 189337.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.840251207351685,
|
|
"epoch": 0.0935672514619883,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.632,
|
|
"mean_token_accuracy": 0.1310859203338623,
|
|
"num_tokens": 201752.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.792728662490845,
|
|
"epoch": 0.09941520467836257,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.7007,
|
|
"mean_token_accuracy": 0.1200527586042881,
|
|
"num_tokens": 214676.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.769255495071411,
|
|
"epoch": 0.10526315789473684,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.6036,
|
|
"mean_token_accuracy": 0.1250310204923153,
|
|
"num_tokens": 225474.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.832526683807373,
|
|
"epoch": 0.1111111111111111,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.6196,
|
|
"mean_token_accuracy": 0.13330849632620811,
|
|
"num_tokens": 237345.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.786408853530884,
|
|
"epoch": 0.11695906432748537,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.5976,
|
|
"mean_token_accuracy": 0.13106755986809732,
|
|
"num_tokens": 249633.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.785315036773682,
|
|
"epoch": 0.12280701754385964,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.5707,
|
|
"mean_token_accuracy": 0.12854716926813126,
|
|
"num_tokens": 261840.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.7381500720977785,
|
|
"epoch": 0.1286549707602339,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.6581,
|
|
"mean_token_accuracy": 0.12695110440254212,
|
|
"num_tokens": 273838.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.7819578647613525,
|
|
"epoch": 0.13450292397660818,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.6314,
|
|
"mean_token_accuracy": 0.12518885657191275,
|
|
"num_tokens": 286001.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.892224454879761,
|
|
"epoch": 0.14035087719298245,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.5622,
|
|
"mean_token_accuracy": 0.1403429314494133,
|
|
"num_tokens": 299052.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.751383972167969,
|
|
"epoch": 0.14619883040935672,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.6007,
|
|
"mean_token_accuracy": 0.12698042243719102,
|
|
"num_tokens": 311803.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.862428855895996,
|
|
"epoch": 0.15204678362573099,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.6155,
|
|
"mean_token_accuracy": 0.1285979501903057,
|
|
"num_tokens": 323712.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.733092164993286,
|
|
"epoch": 0.15789473684210525,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.5735,
|
|
"mean_token_accuracy": 0.1340746872127056,
|
|
"num_tokens": 335171.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.840703058242798,
|
|
"epoch": 0.16374269005847952,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6469,
|
|
"mean_token_accuracy": 0.12687307894229888,
|
|
"num_tokens": 347185.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.806375932693482,
|
|
"epoch": 0.1695906432748538,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5806,
|
|
"mean_token_accuracy": 0.12792280316352844,
|
|
"num_tokens": 359695.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.806016826629639,
|
|
"epoch": 0.17543859649122806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6646,
|
|
"mean_token_accuracy": 0.13149451091885567,
|
|
"num_tokens": 373805.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.811039972305298,
|
|
"epoch": 0.18128654970760233,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.615,
|
|
"mean_token_accuracy": 0.13272017240524292,
|
|
"num_tokens": 385772.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.7556663990020756,
|
|
"epoch": 0.1871345029239766,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.5329,
|
|
"mean_token_accuracy": 0.13663637787103652,
|
|
"num_tokens": 397641.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.7673561573028564,
|
|
"epoch": 0.19298245614035087,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5761,
|
|
"mean_token_accuracy": 0.13438102826476098,
|
|
"num_tokens": 410241.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.851863145828247,
|
|
"epoch": 0.19883040935672514,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5371,
|
|
"mean_token_accuracy": 0.12968598157167435,
|
|
"num_tokens": 422079.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.722187614440918,
|
|
"epoch": 0.2046783625730994,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6151,
|
|
"mean_token_accuracy": 0.12510994151234628,
|
|
"num_tokens": 435187.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.802532577514649,
|
|
"epoch": 0.21052631578947367,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.59,
|
|
"mean_token_accuracy": 0.12483937367796898,
|
|
"num_tokens": 447329.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.7081256866455075,
|
|
"epoch": 0.21637426900584794,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5244,
|
|
"mean_token_accuracy": 0.13455075845122338,
|
|
"num_tokens": 459316.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.866452074050903,
|
|
"epoch": 0.2222222222222222,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.602,
|
|
"mean_token_accuracy": 0.13409386128187178,
|
|
"num_tokens": 471184.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.7785547256469725,
|
|
"epoch": 0.22807017543859648,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.5964,
|
|
"mean_token_accuracy": 0.13105484768748282,
|
|
"num_tokens": 483988.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.754951095581054,
|
|
"epoch": 0.23391812865497075,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.6021,
|
|
"mean_token_accuracy": 0.12733055427670478,
|
|
"num_tokens": 495995.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.760473966598511,
|
|
"epoch": 0.23976608187134502,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5481,
|
|
"mean_token_accuracy": 0.13547581657767296,
|
|
"num_tokens": 507382.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.720216274261475,
|
|
"epoch": 0.24561403508771928,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.5492,
|
|
"mean_token_accuracy": 0.1361517131328583,
|
|
"num_tokens": 518748.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.7840687274932865,
|
|
"epoch": 0.25146198830409355,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5697,
|
|
"mean_token_accuracy": 0.12845324501395225,
|
|
"num_tokens": 531571.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.810080337524414,
|
|
"epoch": 0.2573099415204678,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5872,
|
|
"mean_token_accuracy": 0.1274337127804756,
|
|
"num_tokens": 544699.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.742982816696167,
|
|
"epoch": 0.2631578947368421,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5439,
|
|
"mean_token_accuracy": 0.13237009420990944,
|
|
"num_tokens": 558110.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.847256898880005,
|
|
"epoch": 0.26900584795321636,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.6277,
|
|
"mean_token_accuracy": 0.12315782383084298,
|
|
"num_tokens": 571566.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.752783679962159,
|
|
"epoch": 0.27485380116959063,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.5427,
|
|
"mean_token_accuracy": 0.130996935069561,
|
|
"num_tokens": 583149.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.764012479782105,
|
|
"epoch": 0.2807017543859649,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.5734,
|
|
"mean_token_accuracy": 0.1311168886721134,
|
|
"num_tokens": 596360.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.776003646850586,
|
|
"epoch": 0.28654970760233917,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.57,
|
|
"mean_token_accuracy": 0.12710137963294982,
|
|
"num_tokens": 607719.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.803481483459473,
|
|
"epoch": 0.29239766081871343,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.5331,
|
|
"mean_token_accuracy": 0.13498716056346893,
|
|
"num_tokens": 619002.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.713655710220337,
|
|
"epoch": 0.2982456140350877,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.5076,
|
|
"mean_token_accuracy": 0.13216801062226297,
|
|
"num_tokens": 631437.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.640055465698242,
|
|
"epoch": 0.30409356725146197,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.5087,
|
|
"mean_token_accuracy": 0.13282301425933837,
|
|
"num_tokens": 642611.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.83185043334961,
|
|
"epoch": 0.30994152046783624,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.5447,
|
|
"mean_token_accuracy": 0.1268044739961624,
|
|
"num_tokens": 654874.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.68661732673645,
|
|
"epoch": 0.3157894736842105,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.629,
|
|
"mean_token_accuracy": 0.1293095037341118,
|
|
"num_tokens": 669141.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.781434631347656,
|
|
"epoch": 0.3216374269005848,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.5456,
|
|
"mean_token_accuracy": 0.13286237716674804,
|
|
"num_tokens": 681192.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.814055967330932,
|
|
"epoch": 0.32748538011695905,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.5423,
|
|
"mean_token_accuracy": 0.13368416726589202,
|
|
"num_tokens": 693191.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.664337110519409,
|
|
"epoch": 0.3333333333333333,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.5257,
|
|
"mean_token_accuracy": 0.1304435022175312,
|
|
"num_tokens": 705131.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.808525848388672,
|
|
"epoch": 0.3391812865497076,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.5226,
|
|
"mean_token_accuracy": 0.1330956645309925,
|
|
"num_tokens": 717873.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.779027700424194,
|
|
"epoch": 0.34502923976608185,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.5753,
|
|
"mean_token_accuracy": 0.12839027419686316,
|
|
"num_tokens": 730527.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.754869270324707,
|
|
"epoch": 0.3508771929824561,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.6158,
|
|
"mean_token_accuracy": 0.1226100891828537,
|
|
"num_tokens": 743263.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.853420972824097,
|
|
"epoch": 0.3567251461988304,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6122,
|
|
"mean_token_accuracy": 0.12751553878188132,
|
|
"num_tokens": 755422.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.644754314422608,
|
|
"epoch": 0.36257309941520466,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.4967,
|
|
"mean_token_accuracy": 0.13192886114120483,
|
|
"num_tokens": 767854.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.757552576065064,
|
|
"epoch": 0.3684210526315789,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.4794,
|
|
"mean_token_accuracy": 0.13279144391417502,
|
|
"num_tokens": 779969.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.734856033325196,
|
|
"epoch": 0.3742690058479532,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.5177,
|
|
"mean_token_accuracy": 0.12462197691202163,
|
|
"num_tokens": 792544.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.747901725769043,
|
|
"epoch": 0.38011695906432746,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.5284,
|
|
"mean_token_accuracy": 0.1278999462723732,
|
|
"num_tokens": 804603.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.6661285877227785,
|
|
"epoch": 0.38596491228070173,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.4698,
|
|
"mean_token_accuracy": 0.1412305675446987,
|
|
"num_tokens": 816006.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.718921327590943,
|
|
"epoch": 0.391812865497076,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.4472,
|
|
"mean_token_accuracy": 0.13868665173649788,
|
|
"num_tokens": 828003.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.650281476974487,
|
|
"epoch": 0.39766081871345027,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.5019,
|
|
"mean_token_accuracy": 0.13453598469495773,
|
|
"num_tokens": 840649.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.828660154342652,
|
|
"epoch": 0.40350877192982454,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.5351,
|
|
"mean_token_accuracy": 0.12318970710039139,
|
|
"num_tokens": 853736.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.61459264755249,
|
|
"epoch": 0.4093567251461988,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.5521,
|
|
"mean_token_accuracy": 0.1260755755007267,
|
|
"num_tokens": 866346.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.783421611785888,
|
|
"epoch": 0.4152046783625731,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.4784,
|
|
"mean_token_accuracy": 0.12796891778707503,
|
|
"num_tokens": 879163.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.739914035797119,
|
|
"epoch": 0.42105263157894735,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.5281,
|
|
"mean_token_accuracy": 0.1271045207977295,
|
|
"num_tokens": 891452.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.692205429077148,
|
|
"epoch": 0.4269005847953216,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.5409,
|
|
"mean_token_accuracy": 0.13060612305998803,
|
|
"num_tokens": 903351.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.750999116897583,
|
|
"epoch": 0.4327485380116959,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.4841,
|
|
"mean_token_accuracy": 0.1305551879107952,
|
|
"num_tokens": 914818.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.551722192764283,
|
|
"epoch": 0.43859649122807015,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.3844,
|
|
"mean_token_accuracy": 0.14969016388058662,
|
|
"num_tokens": 927337.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.822294139862061,
|
|
"epoch": 0.4444444444444444,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.5932,
|
|
"mean_token_accuracy": 0.1261659525334835,
|
|
"num_tokens": 940058.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.664493227005005,
|
|
"epoch": 0.4502923976608187,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.3961,
|
|
"mean_token_accuracy": 0.13526701480150222,
|
|
"num_tokens": 952820.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.564838886260986,
|
|
"epoch": 0.45614035087719296,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.4493,
|
|
"mean_token_accuracy": 0.13812072202563286,
|
|
"num_tokens": 965132.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.727004194259644,
|
|
"epoch": 0.4619883040935672,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.4848,
|
|
"mean_token_accuracy": 0.1320212572813034,
|
|
"num_tokens": 978695.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.706100749969482,
|
|
"epoch": 0.4678362573099415,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.4478,
|
|
"mean_token_accuracy": 0.13389745727181435,
|
|
"num_tokens": 992427.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.581472444534302,
|
|
"epoch": 0.47368421052631576,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.4625,
|
|
"mean_token_accuracy": 0.13692987337708473,
|
|
"num_tokens": 1005090.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.768648767471314,
|
|
"epoch": 0.47953216374269003,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.4654,
|
|
"mean_token_accuracy": 0.13497180566191674,
|
|
"num_tokens": 1017219.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.577150154113769,
|
|
"epoch": 0.4853801169590643,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.5118,
|
|
"mean_token_accuracy": 0.12923638001084328,
|
|
"num_tokens": 1029576.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.703565454483032,
|
|
"epoch": 0.49122807017543857,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5447,
|
|
"mean_token_accuracy": 0.12888929918408393,
|
|
"num_tokens": 1042259.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.760423231124878,
|
|
"epoch": 0.49707602339181284,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.4665,
|
|
"mean_token_accuracy": 0.12709181681275367,
|
|
"num_tokens": 1054965.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.540304946899414,
|
|
"epoch": 0.5029239766081871,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.4192,
|
|
"mean_token_accuracy": 0.13073362559080123,
|
|
"num_tokens": 1067346.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.69997730255127,
|
|
"epoch": 0.5087719298245614,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.4084,
|
|
"mean_token_accuracy": 0.12967713326215743,
|
|
"num_tokens": 1080366.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.659937810897827,
|
|
"epoch": 0.5146198830409356,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.3841,
|
|
"mean_token_accuracy": 0.1347927011549473,
|
|
"num_tokens": 1091904.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.573716306686402,
|
|
"epoch": 0.52046783625731,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.3977,
|
|
"mean_token_accuracy": 0.13465345576405524,
|
|
"num_tokens": 1103363.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.556374645233154,
|
|
"epoch": 0.5263157894736842,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.4271,
|
|
"mean_token_accuracy": 0.1363191269338131,
|
|
"num_tokens": 1115178.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.6787323474884035,
|
|
"epoch": 0.5321637426900585,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.364,
|
|
"mean_token_accuracy": 0.13259132504463195,
|
|
"num_tokens": 1128276.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.688528728485108,
|
|
"epoch": 0.5380116959064327,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.4907,
|
|
"mean_token_accuracy": 0.13230301365256308,
|
|
"num_tokens": 1142100.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.469075775146484,
|
|
"epoch": 0.543859649122807,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.4026,
|
|
"mean_token_accuracy": 0.13583226278424262,
|
|
"num_tokens": 1154928.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.702433824539185,
|
|
"epoch": 0.5497076023391813,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.4994,
|
|
"mean_token_accuracy": 0.1225062869489193,
|
|
"num_tokens": 1166937.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.537669372558594,
|
|
"epoch": 0.5555555555555556,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.4175,
|
|
"mean_token_accuracy": 0.1342865504324436,
|
|
"num_tokens": 1180008.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.674032211303711,
|
|
"epoch": 0.5614035087719298,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.4255,
|
|
"mean_token_accuracy": 0.12588841170072557,
|
|
"num_tokens": 1191926.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.737406444549561,
|
|
"epoch": 0.5672514619883041,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.4953,
|
|
"mean_token_accuracy": 0.12321070730686187,
|
|
"num_tokens": 1204904.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.525170803070068,
|
|
"epoch": 0.5730994152046783,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.3645,
|
|
"mean_token_accuracy": 0.13284065276384355,
|
|
"num_tokens": 1217006.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.552954483032226,
|
|
"epoch": 0.5789473684210527,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3799,
|
|
"mean_token_accuracy": 0.13598829582333566,
|
|
"num_tokens": 1228460.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.607265853881836,
|
|
"epoch": 0.5847953216374269,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.4021,
|
|
"mean_token_accuracy": 0.1343585431575775,
|
|
"num_tokens": 1240891.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5847953216374269,
|
|
"eval_entropy": 7.373745485041345,
|
|
"eval_loss": 7.422036647796631,
|
|
"eval_mean_token_accuracy": 0.12701122225660577,
|
|
"eval_num_tokens": 1240891.0,
|
|
"eval_runtime": 2.2961,
|
|
"eval_samples_per_second": 1178.964,
|
|
"eval_steps_per_second": 147.643,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.5875434398651125,
|
|
"epoch": 0.5906432748538012,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3656,
|
|
"mean_token_accuracy": 0.1350531853735447,
|
|
"num_tokens": 1254385.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.6494550704956055,
|
|
"epoch": 0.5964912280701754,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.4191,
|
|
"mean_token_accuracy": 0.13399955853819848,
|
|
"num_tokens": 1266809.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.640181350708008,
|
|
"epoch": 0.6023391812865497,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.4899,
|
|
"mean_token_accuracy": 0.12867182418704032,
|
|
"num_tokens": 1279478.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.59810357093811,
|
|
"epoch": 0.6081871345029239,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.4026,
|
|
"mean_token_accuracy": 0.13495590910315514,
|
|
"num_tokens": 1291241.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.57518630027771,
|
|
"epoch": 0.6140350877192983,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3959,
|
|
"mean_token_accuracy": 0.1318567432463169,
|
|
"num_tokens": 1303776.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.491529178619385,
|
|
"epoch": 0.6198830409356725,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3455,
|
|
"mean_token_accuracy": 0.13106245100498198,
|
|
"num_tokens": 1315970.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.7029557704925535,
|
|
"epoch": 0.6257309941520468,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.423,
|
|
"mean_token_accuracy": 0.12939090728759767,
|
|
"num_tokens": 1328142.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.564161109924316,
|
|
"epoch": 0.631578947368421,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.3336,
|
|
"mean_token_accuracy": 0.13661645874381065,
|
|
"num_tokens": 1341126.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.525344514846802,
|
|
"epoch": 0.6374269005847953,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.3498,
|
|
"mean_token_accuracy": 0.13808920085430146,
|
|
"num_tokens": 1353820.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.3739342212677,
|
|
"epoch": 0.6432748538011696,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.3519,
|
|
"mean_token_accuracy": 0.13071095943450928,
|
|
"num_tokens": 1365342.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.562578582763672,
|
|
"epoch": 0.6491228070175439,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.373,
|
|
"mean_token_accuracy": 0.13720307797193526,
|
|
"num_tokens": 1377965.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.589451599121094,
|
|
"epoch": 0.6549707602339181,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.4585,
|
|
"mean_token_accuracy": 0.12641436904668807,
|
|
"num_tokens": 1391128.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.503201961517334,
|
|
"epoch": 0.6608187134502924,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.3229,
|
|
"mean_token_accuracy": 0.13581799939274788,
|
|
"num_tokens": 1402866.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.522457122802734,
|
|
"epoch": 0.6666666666666666,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.4011,
|
|
"mean_token_accuracy": 0.13442004173994065,
|
|
"num_tokens": 1415314.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.550132846832275,
|
|
"epoch": 0.672514619883041,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.3021,
|
|
"mean_token_accuracy": 0.13721019327640532,
|
|
"num_tokens": 1427003.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.435861873626709,
|
|
"epoch": 0.6783625730994152,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.2802,
|
|
"mean_token_accuracy": 0.13575370907783507,
|
|
"num_tokens": 1438950.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.538718414306641,
|
|
"epoch": 0.6842105263157895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.3822,
|
|
"mean_token_accuracy": 0.13449773266911508,
|
|
"num_tokens": 1451640.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.539703464508056,
|
|
"epoch": 0.6900584795321637,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.3901,
|
|
"mean_token_accuracy": 0.13477873280644417,
|
|
"num_tokens": 1464282.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.420022916793823,
|
|
"epoch": 0.695906432748538,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.3199,
|
|
"mean_token_accuracy": 0.13606854826211928,
|
|
"num_tokens": 1476199.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.528499126434326,
|
|
"epoch": 0.7017543859649122,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.3494,
|
|
"mean_token_accuracy": 0.13913520947098731,
|
|
"num_tokens": 1488055.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.544654035568238,
|
|
"epoch": 0.7076023391812866,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.2989,
|
|
"mean_token_accuracy": 0.13840262964367867,
|
|
"num_tokens": 1500764.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.431989049911499,
|
|
"epoch": 0.7134502923976608,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.3203,
|
|
"mean_token_accuracy": 0.13689279779791833,
|
|
"num_tokens": 1513645.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.424747610092163,
|
|
"epoch": 0.7192982456140351,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.2418,
|
|
"mean_token_accuracy": 0.14617047160863877,
|
|
"num_tokens": 1528146.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.445183849334716,
|
|
"epoch": 0.7251461988304093,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.3038,
|
|
"mean_token_accuracy": 0.13819727301597595,
|
|
"num_tokens": 1541336.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.532771015167237,
|
|
"epoch": 0.7309941520467836,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.3238,
|
|
"mean_token_accuracy": 0.13210776224732398,
|
|
"num_tokens": 1554609.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.47041335105896,
|
|
"epoch": 0.7368421052631579,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.4036,
|
|
"mean_token_accuracy": 0.13117921128869056,
|
|
"num_tokens": 1567520.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.4976729393005375,
|
|
"epoch": 0.7426900584795322,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.2851,
|
|
"mean_token_accuracy": 0.14150782003998758,
|
|
"num_tokens": 1579783.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.376789665222168,
|
|
"epoch": 0.7485380116959064,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.2247,
|
|
"mean_token_accuracy": 0.13450378701090812,
|
|
"num_tokens": 1592830.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.538475275039673,
|
|
"epoch": 0.7543859649122807,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.4304,
|
|
"mean_token_accuracy": 0.12484153211116791,
|
|
"num_tokens": 1605470.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.442430639266968,
|
|
"epoch": 0.7602339181286549,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.2943,
|
|
"mean_token_accuracy": 0.13867733925580977,
|
|
"num_tokens": 1618106.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.355888032913208,
|
|
"epoch": 0.7660818713450293,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.3423,
|
|
"mean_token_accuracy": 0.12944767996668816,
|
|
"num_tokens": 1632043.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.49034104347229,
|
|
"epoch": 0.7719298245614035,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.2699,
|
|
"mean_token_accuracy": 0.1375485248863697,
|
|
"num_tokens": 1644671.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.438511610031128,
|
|
"epoch": 0.7777777777777778,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.2571,
|
|
"mean_token_accuracy": 0.13946201801300048,
|
|
"num_tokens": 1657000.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.363019323348999,
|
|
"epoch": 0.783625730994152,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.288,
|
|
"mean_token_accuracy": 0.1353791870176792,
|
|
"num_tokens": 1669874.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.394830369949341,
|
|
"epoch": 0.7894736842105263,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.3037,
|
|
"mean_token_accuracy": 0.13502334728837012,
|
|
"num_tokens": 1681238.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.389416217803955,
|
|
"epoch": 0.7953216374269005,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.1837,
|
|
"mean_token_accuracy": 0.1395817719399929,
|
|
"num_tokens": 1694887.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.460854482650757,
|
|
"epoch": 0.8011695906432749,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.3458,
|
|
"mean_token_accuracy": 0.13248592168092727,
|
|
"num_tokens": 1706888.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.458899688720703,
|
|
"epoch": 0.8070175438596491,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.3121,
|
|
"mean_token_accuracy": 0.13233812376856804,
|
|
"num_tokens": 1718681.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.411834287643432,
|
|
"epoch": 0.8128654970760234,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.3076,
|
|
"mean_token_accuracy": 0.13430518060922622,
|
|
"num_tokens": 1732073.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.387229108810425,
|
|
"epoch": 0.8187134502923976,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.2038,
|
|
"mean_token_accuracy": 0.13365850895643233,
|
|
"num_tokens": 1744176.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.384578371047974,
|
|
"epoch": 0.8245614035087719,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.2441,
|
|
"mean_token_accuracy": 0.1429324761033058,
|
|
"num_tokens": 1757138.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.342513751983643,
|
|
"epoch": 0.8304093567251462,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.193,
|
|
"mean_token_accuracy": 0.14023306518793105,
|
|
"num_tokens": 1768762.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.379601860046387,
|
|
"epoch": 0.8362573099415205,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.2776,
|
|
"mean_token_accuracy": 0.1369776740670204,
|
|
"num_tokens": 1781274.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.324306583404541,
|
|
"epoch": 0.8421052631578947,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.1528,
|
|
"mean_token_accuracy": 0.14969784170389175,
|
|
"num_tokens": 1794092.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.376630353927612,
|
|
"epoch": 0.847953216374269,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.3085,
|
|
"mean_token_accuracy": 0.13185275122523307,
|
|
"num_tokens": 1806605.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.482823085784912,
|
|
"epoch": 0.8538011695906432,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.3004,
|
|
"mean_token_accuracy": 0.12456657811999321,
|
|
"num_tokens": 1820064.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.312306928634643,
|
|
"epoch": 0.8596491228070176,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.1611,
|
|
"mean_token_accuracy": 0.1399262838065624,
|
|
"num_tokens": 1832899.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.350351285934448,
|
|
"epoch": 0.8654970760233918,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.2602,
|
|
"mean_token_accuracy": 0.13681291192770004,
|
|
"num_tokens": 1845701.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.113675498962403,
|
|
"epoch": 0.8713450292397661,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.0812,
|
|
"mean_token_accuracy": 0.14980586394667625,
|
|
"num_tokens": 1858143.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.359517431259155,
|
|
"epoch": 0.8771929824561403,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.2215,
|
|
"mean_token_accuracy": 0.13886995390057563,
|
|
"num_tokens": 1871111.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.3289408683776855,
|
|
"epoch": 0.8830409356725146,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.1813,
|
|
"mean_token_accuracy": 0.1427088886499405,
|
|
"num_tokens": 1882806.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.2833109378814695,
|
|
"epoch": 0.8888888888888888,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.1914,
|
|
"mean_token_accuracy": 0.14085574224591255,
|
|
"num_tokens": 1895451.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.306222057342529,
|
|
"epoch": 0.8947368421052632,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.1696,
|
|
"mean_token_accuracy": 0.1424693427979946,
|
|
"num_tokens": 1906959.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.268920803070069,
|
|
"epoch": 0.9005847953216374,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.2486,
|
|
"mean_token_accuracy": 0.13296061381697655,
|
|
"num_tokens": 1919889.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.426680231094361,
|
|
"epoch": 0.9064327485380117,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.2794,
|
|
"mean_token_accuracy": 0.13839673027396202,
|
|
"num_tokens": 1932152.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.396844387054443,
|
|
"epoch": 0.9122807017543859,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.2329,
|
|
"mean_token_accuracy": 0.1382187895476818,
|
|
"num_tokens": 1944874.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.191975879669189,
|
|
"epoch": 0.9181286549707602,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.2003,
|
|
"mean_token_accuracy": 0.1381520450115204,
|
|
"num_tokens": 1956698.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.353271389007569,
|
|
"epoch": 0.9239766081871345,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.1834,
|
|
"mean_token_accuracy": 0.1376000702381134,
|
|
"num_tokens": 1970553.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.22394437789917,
|
|
"epoch": 0.9298245614035088,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.1715,
|
|
"mean_token_accuracy": 0.13637209460139274,
|
|
"num_tokens": 1982836.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.331004190444946,
|
|
"epoch": 0.935672514619883,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.182,
|
|
"mean_token_accuracy": 0.14434827119112015,
|
|
"num_tokens": 1994012.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.191262865066529,
|
|
"epoch": 0.9415204678362573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.0765,
|
|
"mean_token_accuracy": 0.14728244692087172,
|
|
"num_tokens": 2007058.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.346480751037598,
|
|
"epoch": 0.9473684210526315,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.2807,
|
|
"mean_token_accuracy": 0.1287767119705677,
|
|
"num_tokens": 2019191.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.356786680221558,
|
|
"epoch": 0.9532163742690059,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.2192,
|
|
"mean_token_accuracy": 0.1383483201265335,
|
|
"num_tokens": 2031485.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.276351737976074,
|
|
"epoch": 0.9590643274853801,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.2438,
|
|
"mean_token_accuracy": 0.12981143370270729,
|
|
"num_tokens": 2044606.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.322518301010132,
|
|
"epoch": 0.9649122807017544,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.2186,
|
|
"mean_token_accuracy": 0.1239808440208435,
|
|
"num_tokens": 2056170.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.2173041820526125,
|
|
"epoch": 0.9707602339181286,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.1692,
|
|
"mean_token_accuracy": 0.13378976061940193,
|
|
"num_tokens": 2069257.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.278322315216064,
|
|
"epoch": 0.9766081871345029,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.1461,
|
|
"mean_token_accuracy": 0.13162968605756759,
|
|
"num_tokens": 2081347.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.183484840393066,
|
|
"epoch": 0.9824561403508771,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.0693,
|
|
"mean_token_accuracy": 0.14508000239729882,
|
|
"num_tokens": 2093121.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.276888608932495,
|
|
"epoch": 0.9883040935672515,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.154,
|
|
"mean_token_accuracy": 0.13984944075345992,
|
|
"num_tokens": 2105098.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.281503057479858,
|
|
"epoch": 0.9941520467836257,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.257,
|
|
"mean_token_accuracy": 0.12888796031475067,
|
|
"num_tokens": 2118305.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.19629979133606,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.1088,
|
|
"mean_token_accuracy": 0.14972682297229767,
|
|
"num_tokens": 2129484.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.273441553115845,
|
|
"epoch": 1.0058479532163742,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.9959,
|
|
"mean_token_accuracy": 0.13713507130742073,
|
|
"num_tokens": 2141996.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.136537981033325,
|
|
"epoch": 1.0116959064327484,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.9814,
|
|
"mean_token_accuracy": 0.14575768858194352,
|
|
"num_tokens": 2154396.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.241056442260742,
|
|
"epoch": 1.0175438596491229,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8824,
|
|
"mean_token_accuracy": 0.13679319769144058,
|
|
"num_tokens": 2166397.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.1693150997161865,
|
|
"epoch": 1.023391812865497,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.052,
|
|
"mean_token_accuracy": 0.13590832501649858,
|
|
"num_tokens": 2180441.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.1720458507537845,
|
|
"epoch": 1.0292397660818713,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.9798,
|
|
"mean_token_accuracy": 0.14448419958353043,
|
|
"num_tokens": 2192759.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.048953914642334,
|
|
"epoch": 1.0350877192982457,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9222,
|
|
"mean_token_accuracy": 0.14383947402238845,
|
|
"num_tokens": 2205989.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.172360372543335,
|
|
"epoch": 1.04093567251462,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.9968,
|
|
"mean_token_accuracy": 0.14286552369594574,
|
|
"num_tokens": 2218780.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.109023189544677,
|
|
"epoch": 1.0467836257309941,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.0301,
|
|
"mean_token_accuracy": 0.14032476022839546,
|
|
"num_tokens": 2231748.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.228937244415283,
|
|
"epoch": 1.0526315789473684,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.0109,
|
|
"mean_token_accuracy": 0.1399959348142147,
|
|
"num_tokens": 2244635.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.132290601730347,
|
|
"epoch": 1.0584795321637426,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.9031,
|
|
"mean_token_accuracy": 0.14366517663002015,
|
|
"num_tokens": 2255941.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.007370376586914,
|
|
"epoch": 1.064327485380117,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.9809,
|
|
"mean_token_accuracy": 0.14777857810258865,
|
|
"num_tokens": 2267162.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.138269233703613,
|
|
"epoch": 1.0701754385964912,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.9496,
|
|
"mean_token_accuracy": 0.14804905876517296,
|
|
"num_tokens": 2279290.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.11580114364624,
|
|
"epoch": 1.0760233918128654,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.0084,
|
|
"mean_token_accuracy": 0.13601162061095237,
|
|
"num_tokens": 2293053.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.015692281723022,
|
|
"epoch": 1.0818713450292399,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.9212,
|
|
"mean_token_accuracy": 0.14603447318077087,
|
|
"num_tokens": 2305489.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.051322793960571,
|
|
"epoch": 1.087719298245614,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.8687,
|
|
"mean_token_accuracy": 0.15775832384824753,
|
|
"num_tokens": 2319310.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.131311130523682,
|
|
"epoch": 1.0935672514619883,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.0298,
|
|
"mean_token_accuracy": 0.13980275765061378,
|
|
"num_tokens": 2332132.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.102800178527832,
|
|
"epoch": 1.0994152046783625,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.0008,
|
|
"mean_token_accuracy": 0.14390438795089722,
|
|
"num_tokens": 2343505.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.119720315933227,
|
|
"epoch": 1.1052631578947367,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.9893,
|
|
"mean_token_accuracy": 0.14247232899069787,
|
|
"num_tokens": 2355930.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.099494981765747,
|
|
"epoch": 1.1111111111111112,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9565,
|
|
"mean_token_accuracy": 0.14375333711504937,
|
|
"num_tokens": 2369820.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.016859483718872,
|
|
"epoch": 1.1169590643274854,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.9159,
|
|
"mean_token_accuracy": 0.14982443079352378,
|
|
"num_tokens": 2382375.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.053458166122437,
|
|
"epoch": 1.1228070175438596,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.0089,
|
|
"mean_token_accuracy": 0.14104411751031876,
|
|
"num_tokens": 2393391.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.047031879425049,
|
|
"epoch": 1.128654970760234,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.9345,
|
|
"mean_token_accuracy": 0.1424281544983387,
|
|
"num_tokens": 2405266.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.0777851104736325,
|
|
"epoch": 1.1345029239766082,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.9704,
|
|
"mean_token_accuracy": 0.14111316949129105,
|
|
"num_tokens": 2418620.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.027369260787964,
|
|
"epoch": 1.1403508771929824,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.065,
|
|
"mean_token_accuracy": 0.1403935208916664,
|
|
"num_tokens": 2431107.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.187720537185669,
|
|
"epoch": 1.1461988304093567,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.0144,
|
|
"mean_token_accuracy": 0.13802240937948226,
|
|
"num_tokens": 2443729.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.010544013977051,
|
|
"epoch": 1.1520467836257309,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.9195,
|
|
"mean_token_accuracy": 0.14878860712051392,
|
|
"num_tokens": 2456798.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.0476819515228275,
|
|
"epoch": 1.1578947368421053,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.009,
|
|
"mean_token_accuracy": 0.1457380011677742,
|
|
"num_tokens": 2468701.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.9938723087310795,
|
|
"epoch": 1.1637426900584795,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.9561,
|
|
"mean_token_accuracy": 0.14130357205867766,
|
|
"num_tokens": 2480391.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.076553392410278,
|
|
"epoch": 1.1695906432748537,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.9596,
|
|
"mean_token_accuracy": 0.14768824875354766,
|
|
"num_tokens": 2491887.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1695906432748537,
|
|
"eval_entropy": 6.926040534072921,
|
|
"eval_loss": 7.151692867279053,
|
|
"eval_mean_token_accuracy": 0.13112522253167752,
|
|
"eval_num_tokens": 2491887.0,
|
|
"eval_runtime": 2.2827,
|
|
"eval_samples_per_second": 1185.89,
|
|
"eval_steps_per_second": 148.51,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.03064193725586,
|
|
"epoch": 1.1754385964912282,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999996875147963,
|
|
"loss": 6.9415,
|
|
"mean_token_accuracy": 0.14060102105140687,
|
|
"num_tokens": 2503490.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.09430251121521,
|
|
"epoch": 1.1812865497076024,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999984180451436,
|
|
"loss": 7.0456,
|
|
"mean_token_accuracy": 0.13780836313962935,
|
|
"num_tokens": 2515330.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.0454943656921385,
|
|
"epoch": 1.1871345029239766,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999961720662221,
|
|
"loss": 6.9752,
|
|
"mean_token_accuracy": 0.1445000797510147,
|
|
"num_tokens": 2526743.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.988425970077515,
|
|
"epoch": 1.1929824561403508,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999929495877796,
|
|
"loss": 7.0217,
|
|
"mean_token_accuracy": 0.14440475553274154,
|
|
"num_tokens": 2538343.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.189151477813721,
|
|
"epoch": 1.198830409356725,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 0.0004999887506238018,
|
|
"loss": 7.1389,
|
|
"mean_token_accuracy": 0.13006046786904335,
|
|
"num_tokens": 2552687.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.918720817565918,
|
|
"epoch": 1.2046783625730995,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999835751925124,
|
|
"loss": 6.9506,
|
|
"mean_token_accuracy": 0.15075658932328223,
|
|
"num_tokens": 2565666.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.996754360198975,
|
|
"epoch": 1.2105263157894737,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999774233163734,
|
|
"loss": 6.9527,
|
|
"mean_token_accuracy": 0.1366286665201187,
|
|
"num_tokens": 2578233.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.082205200195313,
|
|
"epoch": 1.2163742690058479,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999702950220841,
|
|
"loss": 6.9552,
|
|
"mean_token_accuracy": 0.14596255272626876,
|
|
"num_tokens": 2590221.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.949472808837891,
|
|
"epoch": 1.2222222222222223,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999621903405819,
|
|
"loss": 6.9768,
|
|
"mean_token_accuracy": 0.13911690190434456,
|
|
"num_tokens": 2602807.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.050590658187867,
|
|
"epoch": 1.2280701754385965,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000499953109307042,
|
|
"loss": 7.0195,
|
|
"mean_token_accuracy": 0.13519042506814002,
|
|
"num_tokens": 2615273.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.007106781005859,
|
|
"epoch": 1.2339181286549707,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999430519608762,
|
|
"loss": 6.9877,
|
|
"mean_token_accuracy": 0.13718256056308747,
|
|
"num_tokens": 2627077.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.059272384643554,
|
|
"epoch": 1.239766081871345,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999320183457346,
|
|
"loss": 6.9618,
|
|
"mean_token_accuracy": 0.13905680626630784,
|
|
"num_tokens": 2640580.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.970555400848388,
|
|
"epoch": 1.2456140350877192,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999200085095035,
|
|
"loss": 7.0246,
|
|
"mean_token_accuracy": 0.13743191361427307,
|
|
"num_tokens": 2653906.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.9946081161499025,
|
|
"epoch": 1.2514619883040936,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999070225043068,
|
|
"loss": 6.97,
|
|
"mean_token_accuracy": 0.14154034852981567,
|
|
"num_tokens": 2666021.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.074818181991577,
|
|
"epoch": 1.2573099415204678,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998930603865042,
|
|
"loss": 6.9002,
|
|
"mean_token_accuracy": 0.14058925807476044,
|
|
"num_tokens": 2678542.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.897901296615601,
|
|
"epoch": 1.263157894736842,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998781222166929,
|
|
"loss": 6.943,
|
|
"mean_token_accuracy": 0.14766753390431403,
|
|
"num_tokens": 2691006.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.105810642242432,
|
|
"epoch": 1.2690058479532165,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998622080597052,
|
|
"loss": 7.0293,
|
|
"mean_token_accuracy": 0.14189143255352973,
|
|
"num_tokens": 2704064.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.975711107254028,
|
|
"epoch": 1.2748538011695907,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998453179846098,
|
|
"loss": 6.9269,
|
|
"mean_token_accuracy": 0.13873959332704544,
|
|
"num_tokens": 2717452.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.936954832077026,
|
|
"epoch": 1.280701754385965,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998274520647109,
|
|
"loss": 6.8894,
|
|
"mean_token_accuracy": 0.1417660728096962,
|
|
"num_tokens": 2730969.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.8774402141571045,
|
|
"epoch": 1.286549707602339,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998086103775478,
|
|
"loss": 6.8991,
|
|
"mean_token_accuracy": 0.1409339912235737,
|
|
"num_tokens": 2742648.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.919243955612183,
|
|
"epoch": 1.2923976608187133,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997887930048947,
|
|
"loss": 6.8285,
|
|
"mean_token_accuracy": 0.1467128537595272,
|
|
"num_tokens": 2754816.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.961157989501953,
|
|
"epoch": 1.2982456140350878,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004997680000327607,
|
|
"loss": 6.9841,
|
|
"mean_token_accuracy": 0.13582526817917823,
|
|
"num_tokens": 2767066.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.939070129394532,
|
|
"epoch": 1.304093567251462,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997462315513884,
|
|
"loss": 6.873,
|
|
"mean_token_accuracy": 0.14410377591848372,
|
|
"num_tokens": 2779251.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.994794654846191,
|
|
"epoch": 1.3099415204678362,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000499723487655255,
|
|
"loss": 6.8864,
|
|
"mean_token_accuracy": 0.1423727609217167,
|
|
"num_tokens": 2791743.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.952969026565552,
|
|
"epoch": 1.3157894736842106,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004996997684430703,
|
|
"loss": 6.9278,
|
|
"mean_token_accuracy": 0.14672429263591766,
|
|
"num_tokens": 2804694.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.930943393707276,
|
|
"epoch": 1.3216374269005848,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996750740177775,
|
|
"loss": 6.905,
|
|
"mean_token_accuracy": 0.14298794865608216,
|
|
"num_tokens": 2816484.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.945026159286499,
|
|
"epoch": 1.327485380116959,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996494044865521,
|
|
"loss": 6.9284,
|
|
"mean_token_accuracy": 0.1373042084276676,
|
|
"num_tokens": 2829353.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.875068235397339,
|
|
"epoch": 1.3333333333333333,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996227599608018,
|
|
"loss": 6.8588,
|
|
"mean_token_accuracy": 0.14598043411970138,
|
|
"num_tokens": 2842695.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.885642242431641,
|
|
"epoch": 1.3391812865497075,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995951405561658,
|
|
"loss": 6.836,
|
|
"mean_token_accuracy": 0.14788470193743705,
|
|
"num_tokens": 2855627.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.8490643978118895,
|
|
"epoch": 1.345029239766082,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995665463925142,
|
|
"loss": 6.8251,
|
|
"mean_token_accuracy": 0.15566134750843047,
|
|
"num_tokens": 2868132.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.898070859909057,
|
|
"epoch": 1.3508771929824561,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004995369775939478,
|
|
"loss": 6.8227,
|
|
"mean_token_accuracy": 0.13949502557516097,
|
|
"num_tokens": 2880641.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.863165760040284,
|
|
"epoch": 1.3567251461988303,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995064342887972,
|
|
"loss": 6.8693,
|
|
"mean_token_accuracy": 0.14175399467349054,
|
|
"num_tokens": 2892479.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.998213052749634,
|
|
"epoch": 1.3625730994152048,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004994749166096226,
|
|
"loss": 6.9284,
|
|
"mean_token_accuracy": 0.14056200981140138,
|
|
"num_tokens": 2905675.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.885084772109986,
|
|
"epoch": 1.368421052631579,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994424246932129,
|
|
"loss": 6.9458,
|
|
"mean_token_accuracy": 0.1405379667878151,
|
|
"num_tokens": 2918259.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.960570287704468,
|
|
"epoch": 1.3742690058479532,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004994089586805856,
|
|
"loss": 6.8332,
|
|
"mean_token_accuracy": 0.15375758558511735,
|
|
"num_tokens": 2932280.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.81732406616211,
|
|
"epoch": 1.3801169590643274,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993745187169852,
|
|
"loss": 6.877,
|
|
"mean_token_accuracy": 0.14667267948389054,
|
|
"num_tokens": 2944385.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.933884525299073,
|
|
"epoch": 1.3859649122807016,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499339104951884,
|
|
"loss": 6.8819,
|
|
"mean_token_accuracy": 0.13797343596816064,
|
|
"num_tokens": 2956239.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.908072471618652,
|
|
"epoch": 1.391812865497076,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00049930271753898,
|
|
"loss": 6.9006,
|
|
"mean_token_accuracy": 0.14377526491880416,
|
|
"num_tokens": 2967948.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.873384141921997,
|
|
"epoch": 1.3976608187134503,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004992653566361974,
|
|
"loss": 6.8464,
|
|
"mean_token_accuracy": 0.14522414803504943,
|
|
"num_tokens": 2979589.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.831958150863647,
|
|
"epoch": 1.4035087719298245,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004992270224056849,
|
|
"loss": 6.8398,
|
|
"mean_token_accuracy": 0.14143177419900893,
|
|
"num_tokens": 2991679.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.917820596694947,
|
|
"epoch": 1.409356725146199,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004991877150138162,
|
|
"loss": 6.8751,
|
|
"mean_token_accuracy": 0.13764828592538833,
|
|
"num_tokens": 3004037.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.871622037887573,
|
|
"epoch": 1.4152046783625731,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004991474346311879,
|
|
"loss": 6.8488,
|
|
"mean_token_accuracy": 0.14050350114703178,
|
|
"num_tokens": 3016632.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.868418455123901,
|
|
"epoch": 1.4210526315789473,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004991061814326198,
|
|
"loss": 6.8466,
|
|
"mean_token_accuracy": 0.14317608624696732,
|
|
"num_tokens": 3029095.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.901586484909058,
|
|
"epoch": 1.4269005847953216,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004990639555971539,
|
|
"loss": 6.8753,
|
|
"mean_token_accuracy": 0.13867598176002502,
|
|
"num_tokens": 3042467.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.995685815811157,
|
|
"epoch": 1.4327485380116958,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004990207573080533,
|
|
"loss": 6.9927,
|
|
"mean_token_accuracy": 0.13158960789442062,
|
|
"num_tokens": 3056233.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.840405225753784,
|
|
"epoch": 1.4385964912280702,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004989765867528014,
|
|
"loss": 6.9191,
|
|
"mean_token_accuracy": 0.14545614793896675,
|
|
"num_tokens": 3069216.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.882023048400879,
|
|
"epoch": 1.4444444444444444,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004989314441231019,
|
|
"loss": 6.8113,
|
|
"mean_token_accuracy": 0.15136598646640778,
|
|
"num_tokens": 3081116.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.936877727508545,
|
|
"epoch": 1.4502923976608186,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000498885329614877,
|
|
"loss": 6.9482,
|
|
"mean_token_accuracy": 0.13658374026417733,
|
|
"num_tokens": 3093284.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.829084968566894,
|
|
"epoch": 1.456140350877193,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988382434282668,
|
|
"loss": 6.856,
|
|
"mean_token_accuracy": 0.14588840380311013,
|
|
"num_tokens": 3105147.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.857968950271607,
|
|
"epoch": 1.4619883040935673,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000498790185767629,
|
|
"loss": 6.8391,
|
|
"mean_token_accuracy": 0.1462831415235996,
|
|
"num_tokens": 3116457.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.828664684295655,
|
|
"epoch": 1.4678362573099415,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000498741156841537,
|
|
"loss": 6.8415,
|
|
"mean_token_accuracy": 0.1462543435394764,
|
|
"num_tokens": 3128223.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.85268268585205,
|
|
"epoch": 1.4736842105263157,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049869115686278,
|
|
"loss": 6.8888,
|
|
"mean_token_accuracy": 0.14214529991149902,
|
|
"num_tokens": 3139886.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.931294393539429,
|
|
"epoch": 1.47953216374269,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004986401860483615,
|
|
"loss": 6.878,
|
|
"mean_token_accuracy": 0.13950230181217194,
|
|
"num_tokens": 3152624.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.762976217269897,
|
|
"epoch": 1.4853801169590644,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004985882446194984,
|
|
"loss": 6.7706,
|
|
"mean_token_accuracy": 0.14813553541898727,
|
|
"num_tokens": 3164881.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.894041299819946,
|
|
"epoch": 1.4912280701754386,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004985353328016204,
|
|
"loss": 6.8605,
|
|
"mean_token_accuracy": 0.13663011640310288,
|
|
"num_tokens": 3177967.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.873754501342773,
|
|
"epoch": 1.4970760233918128,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004984814508243685,
|
|
"loss": 6.8768,
|
|
"mean_token_accuracy": 0.14321380108594894,
|
|
"num_tokens": 3190646.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.8783287525177,
|
|
"epoch": 1.5029239766081872,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004984265989215942,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.13877278193831444,
|
|
"num_tokens": 3203299.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.81124324798584,
|
|
"epoch": 1.5087719298245614,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004983707773313589,
|
|
"loss": 6.9135,
|
|
"mean_token_accuracy": 0.13964597284793853,
|
|
"num_tokens": 3215992.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.783960151672363,
|
|
"epoch": 1.5146198830409356,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004983139862959322,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.14580969959497453,
|
|
"num_tokens": 3228989.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.869039011001587,
|
|
"epoch": 1.52046783625731,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004982562260617912,
|
|
"loss": 6.9118,
|
|
"mean_token_accuracy": 0.1454062916338444,
|
|
"num_tokens": 3241635.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.80722713470459,
|
|
"epoch": 1.526315789473684,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004981974968796197,
|
|
"loss": 6.843,
|
|
"mean_token_accuracy": 0.14277214854955672,
|
|
"num_tokens": 3253590.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.783811759948731,
|
|
"epoch": 1.5321637426900585,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004981377990043064,
|
|
"loss": 6.7871,
|
|
"mean_token_accuracy": 0.15090060085058213,
|
|
"num_tokens": 3264859.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.9159692287445065,
|
|
"epoch": 1.5380116959064327,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004980771326949442,
|
|
"loss": 6.8957,
|
|
"mean_token_accuracy": 0.14444180428981782,
|
|
"num_tokens": 3276353.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.745017957687378,
|
|
"epoch": 1.543859649122807,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004980154982148296,
|
|
"loss": 6.8208,
|
|
"mean_token_accuracy": 0.14608026593923568,
|
|
"num_tokens": 3288876.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.809013080596924,
|
|
"epoch": 1.5497076023391814,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004979528958314604,
|
|
"loss": 6.8369,
|
|
"mean_token_accuracy": 0.13851020634174346,
|
|
"num_tokens": 3300648.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.806869745254517,
|
|
"epoch": 1.5555555555555556,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004978893258165354,
|
|
"loss": 6.7788,
|
|
"mean_token_accuracy": 0.1443500228226185,
|
|
"num_tokens": 3313664.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.917032289505005,
|
|
"epoch": 1.5614035087719298,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004978247884459532,
|
|
"loss": 6.8891,
|
|
"mean_token_accuracy": 0.13166723549365997,
|
|
"num_tokens": 3326221.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.7701085090637205,
|
|
"epoch": 1.5672514619883042,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004977592839998105,
|
|
"loss": 6.8167,
|
|
"mean_token_accuracy": 0.14367288127541541,
|
|
"num_tokens": 3337470.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.8947234630584715,
|
|
"epoch": 1.5730994152046782,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004976928127624013,
|
|
"loss": 6.9031,
|
|
"mean_token_accuracy": 0.14372491762042044,
|
|
"num_tokens": 3350925.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.732043361663818,
|
|
"epoch": 1.5789473684210527,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004976253750222157,
|
|
"loss": 6.8264,
|
|
"mean_token_accuracy": 0.14307058081030846,
|
|
"num_tokens": 3363387.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.8033754348754885,
|
|
"epoch": 1.5847953216374269,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004975569710719382,
|
|
"loss": 6.8028,
|
|
"mean_token_accuracy": 0.14749399870634078,
|
|
"num_tokens": 3376640.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.904033422470093,
|
|
"epoch": 1.590643274853801,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004974876012084468,
|
|
"loss": 6.8209,
|
|
"mean_token_accuracy": 0.14032859951257706,
|
|
"num_tokens": 3389848.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.781014633178711,
|
|
"epoch": 1.5964912280701755,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004974172657328116,
|
|
"loss": 6.8459,
|
|
"mean_token_accuracy": 0.14404987692832946,
|
|
"num_tokens": 3401862.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.809361743927002,
|
|
"epoch": 1.6023391812865497,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004973459649502939,
|
|
"loss": 6.8454,
|
|
"mean_token_accuracy": 0.14433501437306404,
|
|
"num_tokens": 3414719.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.862741851806641,
|
|
"epoch": 1.608187134502924,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004972736991703439,
|
|
"loss": 6.7987,
|
|
"mean_token_accuracy": 0.14657391011714935,
|
|
"num_tokens": 3427391.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.804032707214356,
|
|
"epoch": 1.6140350877192984,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004972004687066005,
|
|
"loss": 6.8671,
|
|
"mean_token_accuracy": 0.14304091930389404,
|
|
"num_tokens": 3440203.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.806503057479858,
|
|
"epoch": 1.6198830409356724,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000497126273876889,
|
|
"loss": 6.7804,
|
|
"mean_token_accuracy": 0.14722208082675933,
|
|
"num_tokens": 3451801.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.850189828872681,
|
|
"epoch": 1.6257309941520468,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004970511150032205,
|
|
"loss": 6.7863,
|
|
"mean_token_accuracy": 0.1449687287211418,
|
|
"num_tokens": 3464039.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.808931350708008,
|
|
"epoch": 1.631578947368421,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004969749924117896,
|
|
"loss": 6.8233,
|
|
"mean_token_accuracy": 0.14308529123663902,
|
|
"num_tokens": 3476373.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.705229377746582,
|
|
"epoch": 1.6374269005847952,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004968979064329741,
|
|
"loss": 6.7733,
|
|
"mean_token_accuracy": 0.1510583981871605,
|
|
"num_tokens": 3487586.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.747954416275024,
|
|
"epoch": 1.6432748538011697,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004968198574013323,
|
|
"loss": 6.7678,
|
|
"mean_token_accuracy": 0.14269521757960318,
|
|
"num_tokens": 3501210.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.826955032348633,
|
|
"epoch": 1.6491228070175439,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000496740845655603,
|
|
"loss": 6.786,
|
|
"mean_token_accuracy": 0.143802759796381,
|
|
"num_tokens": 3513958.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.692679595947266,
|
|
"epoch": 1.654970760233918,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004966608715387029,
|
|
"loss": 6.7705,
|
|
"mean_token_accuracy": 0.15449799299240113,
|
|
"num_tokens": 3526297.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.825113105773926,
|
|
"epoch": 1.6608187134502925,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004965799353977248,
|
|
"loss": 6.7264,
|
|
"mean_token_accuracy": 0.1499396577477455,
|
|
"num_tokens": 3538308.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.663604831695556,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004964980375839379,
|
|
"loss": 6.7788,
|
|
"mean_token_accuracy": 0.14948643296957015,
|
|
"num_tokens": 3550240.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.810740041732788,
|
|
"epoch": 1.672514619883041,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004964151784527844,
|
|
"loss": 6.8177,
|
|
"mean_token_accuracy": 0.14038417786359786,
|
|
"num_tokens": 3562181.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.7995506763458256,
|
|
"epoch": 1.6783625730994152,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000496331358363879,
|
|
"loss": 6.8748,
|
|
"mean_token_accuracy": 0.14474193155765533,
|
|
"num_tokens": 3574643.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.722173357009888,
|
|
"epoch": 1.6842105263157894,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004962465776810066,
|
|
"loss": 6.6783,
|
|
"mean_token_accuracy": 0.14990489184856415,
|
|
"num_tokens": 3586415.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.80692310333252,
|
|
"epoch": 1.6900584795321638,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004961608367721217,
|
|
"loss": 6.796,
|
|
"mean_token_accuracy": 0.14830706417560577,
|
|
"num_tokens": 3598452.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.748047399520874,
|
|
"epoch": 1.695906432748538,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004960741360093459,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.1371912442147732,
|
|
"num_tokens": 3610899.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.785003566741944,
|
|
"epoch": 1.7017543859649122,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004959864757689669,
|
|
"loss": 6.7816,
|
|
"mean_token_accuracy": 0.14844242185354234,
|
|
"num_tokens": 3622475.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.718051433563232,
|
|
"epoch": 1.7076023391812867,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004958978564314361,
|
|
"loss": 6.8765,
|
|
"mean_token_accuracy": 0.1463383197784424,
|
|
"num_tokens": 3635860.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.84115424156189,
|
|
"epoch": 1.7134502923976607,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000495808278381368,
|
|
"loss": 6.8372,
|
|
"mean_token_accuracy": 0.13889722377061844,
|
|
"num_tokens": 3648718.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.708309984207153,
|
|
"epoch": 1.719298245614035,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004957177420075377,
|
|
"loss": 6.7553,
|
|
"mean_token_accuracy": 0.14848395586013793,
|
|
"num_tokens": 3660746.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.7734802722930905,
|
|
"epoch": 1.7251461988304093,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004956262477028797,
|
|
"loss": 6.7964,
|
|
"mean_token_accuracy": 0.14100225493311883,
|
|
"num_tokens": 3673573.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.736315965652466,
|
|
"epoch": 1.7309941520467835,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004955337958644857,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.1544882595539093,
|
|
"num_tokens": 3685771.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.674497795104981,
|
|
"epoch": 1.736842105263158,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004954403868936032,
|
|
"loss": 6.7876,
|
|
"mean_token_accuracy": 0.1484549582004547,
|
|
"num_tokens": 3698882.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.761597156524658,
|
|
"epoch": 1.7426900584795322,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495346021195634,
|
|
"loss": 6.8086,
|
|
"mean_token_accuracy": 0.14867511242628098,
|
|
"num_tokens": 3712820.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.776470708847046,
|
|
"epoch": 1.7485380116959064,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004952506991801319,
|
|
"loss": 6.7569,
|
|
"mean_token_accuracy": 0.14823356494307519,
|
|
"num_tokens": 3724845.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.657959222793579,
|
|
"epoch": 1.7543859649122808,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004951544212608015,
|
|
"loss": 6.6243,
|
|
"mean_token_accuracy": 0.16444699168205262,
|
|
"num_tokens": 3736309.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7543859649122808,
|
|
"eval_entropy": 6.543074199935328,
|
|
"eval_loss": 6.911302089691162,
|
|
"eval_mean_token_accuracy": 0.1365549994500733,
|
|
"eval_num_tokens": 3736309.0,
|
|
"eval_runtime": 2.2834,
|
|
"eval_samples_per_second": 1185.535,
|
|
"eval_steps_per_second": 148.466,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.784309244155883,
|
|
"epoch": 1.7602339181286548,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004950571878554956,
|
|
"loss": 6.792,
|
|
"mean_token_accuracy": 0.14341349452733992,
|
|
"num_tokens": 3748800.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.687272214889527,
|
|
"epoch": 1.7660818713450293,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004949589993862143,
|
|
"loss": 6.7631,
|
|
"mean_token_accuracy": 0.15106673985719682,
|
|
"num_tokens": 3760753.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.770102500915527,
|
|
"epoch": 1.7719298245614035,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004948598562791025,
|
|
"loss": 6.8495,
|
|
"mean_token_accuracy": 0.14250847771763803,
|
|
"num_tokens": 3773915.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.74422779083252,
|
|
"epoch": 1.7777777777777777,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004947597589644484,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.14317139536142348,
|
|
"num_tokens": 3785907.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.7505429744720455,
|
|
"epoch": 1.7836257309941521,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004946587078766818,
|
|
"loss": 6.6894,
|
|
"mean_token_accuracy": 0.15321093127131463,
|
|
"num_tokens": 3798407.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.542722606658936,
|
|
"epoch": 1.7894736842105263,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004945567034543712,
|
|
"loss": 6.704,
|
|
"mean_token_accuracy": 0.15031700879335402,
|
|
"num_tokens": 3810586.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.850691318511963,
|
|
"epoch": 1.7953216374269005,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004944537461402234,
|
|
"loss": 6.7884,
|
|
"mean_token_accuracy": 0.14434545189142228,
|
|
"num_tokens": 3823901.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.730458450317383,
|
|
"epoch": 1.801169590643275,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004943498363810807,
|
|
"loss": 6.8062,
|
|
"mean_token_accuracy": 0.13911711871623994,
|
|
"num_tokens": 3836926.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.723568391799927,
|
|
"epoch": 1.807017543859649,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004942449746279184,
|
|
"loss": 6.5866,
|
|
"mean_token_accuracy": 0.15482160598039627,
|
|
"num_tokens": 3849291.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.727809047698974,
|
|
"epoch": 1.8128654970760234,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004941391613358444,
|
|
"loss": 6.8201,
|
|
"mean_token_accuracy": 0.1427263304591179,
|
|
"num_tokens": 3862002.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.651541233062744,
|
|
"epoch": 1.8187134502923976,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000494032396964096,
|
|
"loss": 6.725,
|
|
"mean_token_accuracy": 0.1479753792285919,
|
|
"num_tokens": 3874538.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.704821348190308,
|
|
"epoch": 1.8245614035087718,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000493924681976038,
|
|
"loss": 6.6901,
|
|
"mean_token_accuracy": 0.1507028266787529,
|
|
"num_tokens": 3886682.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.611117172241211,
|
|
"epoch": 1.8304093567251463,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004938160168391614,
|
|
"loss": 6.6522,
|
|
"mean_token_accuracy": 0.14730015993118287,
|
|
"num_tokens": 3898144.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.8177355289459225,
|
|
"epoch": 1.8362573099415205,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004937064020250804,
|
|
"loss": 6.8447,
|
|
"mean_token_accuracy": 0.14146786481142043,
|
|
"num_tokens": 3910209.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.690582036972046,
|
|
"epoch": 1.8421052631578947,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004935958380095316,
|
|
"loss": 6.764,
|
|
"mean_token_accuracy": 0.15035033375024795,
|
|
"num_tokens": 3923401.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.772142171859741,
|
|
"epoch": 1.8479532163742691,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004934843252723702,
|
|
"loss": 6.6814,
|
|
"mean_token_accuracy": 0.1467684879899025,
|
|
"num_tokens": 3934970.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.598661994934082,
|
|
"epoch": 1.8538011695906431,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00049337186429757,
|
|
"loss": 6.5979,
|
|
"mean_token_accuracy": 0.15314390659332275,
|
|
"num_tokens": 3945543.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.618632888793945,
|
|
"epoch": 1.8596491228070176,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004932584555732192,
|
|
"loss": 6.7086,
|
|
"mean_token_accuracy": 0.15223760306835174,
|
|
"num_tokens": 3958152.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.754468250274658,
|
|
"epoch": 1.8654970760233918,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004931440995915203,
|
|
"loss": 6.7092,
|
|
"mean_token_accuracy": 0.14498373717069626,
|
|
"num_tokens": 3972173.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.704925966262818,
|
|
"epoch": 1.871345029239766,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000493028796848786,
|
|
"loss": 6.7804,
|
|
"mean_token_accuracy": 0.1387328766286373,
|
|
"num_tokens": 3985865.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.787711143493652,
|
|
"epoch": 1.8771929824561404,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004929125478454389,
|
|
"loss": 6.6433,
|
|
"mean_token_accuracy": 0.15223552137613297,
|
|
"num_tokens": 3999026.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.648384428024292,
|
|
"epoch": 1.8830409356725146,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004927953530860077,
|
|
"loss": 6.7554,
|
|
"mean_token_accuracy": 0.14936705231666564,
|
|
"num_tokens": 4011270.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.763438940048218,
|
|
"epoch": 1.8888888888888888,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004926772130791261,
|
|
"loss": 6.7552,
|
|
"mean_token_accuracy": 0.14583429992198943,
|
|
"num_tokens": 4024162.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.708201503753662,
|
|
"epoch": 1.8947368421052633,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004925581283375303,
|
|
"loss": 6.7003,
|
|
"mean_token_accuracy": 0.1484905406832695,
|
|
"num_tokens": 4037066.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.667509937286377,
|
|
"epoch": 1.9005847953216373,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004924380993780566,
|
|
"loss": 6.7202,
|
|
"mean_token_accuracy": 0.14748010113835336,
|
|
"num_tokens": 4050096.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.7155702114105225,
|
|
"epoch": 1.9064327485380117,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004923171267216392,
|
|
"loss": 6.7029,
|
|
"mean_token_accuracy": 0.14461838454008102,
|
|
"num_tokens": 4061691.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.605735349655151,
|
|
"epoch": 1.912280701754386,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004921952108933083,
|
|
"loss": 6.6427,
|
|
"mean_token_accuracy": 0.15174098312854767,
|
|
"num_tokens": 4073087.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.74940333366394,
|
|
"epoch": 1.9181286549707601,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004920723524221871,
|
|
"loss": 6.7546,
|
|
"mean_token_accuracy": 0.149676413834095,
|
|
"num_tokens": 4086055.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.657784175872803,
|
|
"epoch": 1.9239766081871346,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004919485518414902,
|
|
"loss": 6.6928,
|
|
"mean_token_accuracy": 0.1461845487356186,
|
|
"num_tokens": 4099258.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.716952657699585,
|
|
"epoch": 1.9298245614035088,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004918238096885213,
|
|
"loss": 6.7637,
|
|
"mean_token_accuracy": 0.14263566285371782,
|
|
"num_tokens": 4111545.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.661676263809204,
|
|
"epoch": 1.935672514619883,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00049169812650467,
|
|
"loss": 6.7076,
|
|
"mean_token_accuracy": 0.15014173984527587,
|
|
"num_tokens": 4123373.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.673471117019654,
|
|
"epoch": 1.9415204678362574,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004915715028354102,
|
|
"loss": 6.7194,
|
|
"mean_token_accuracy": 0.1462250053882599,
|
|
"num_tokens": 4136551.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.69158616065979,
|
|
"epoch": 1.9473684210526314,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004914439392302981,
|
|
"loss": 6.6811,
|
|
"mean_token_accuracy": 0.14650024399161338,
|
|
"num_tokens": 4149560.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.646466588973999,
|
|
"epoch": 1.9532163742690059,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004913154362429686,
|
|
"loss": 6.5991,
|
|
"mean_token_accuracy": 0.15584417134523393,
|
|
"num_tokens": 4161767.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.681569337844849,
|
|
"epoch": 1.95906432748538,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004911859944311341,
|
|
"loss": 6.6893,
|
|
"mean_token_accuracy": 0.15823563933372498,
|
|
"num_tokens": 4174203.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.643621110916138,
|
|
"epoch": 1.9649122807017543,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004910556143565812,
|
|
"loss": 6.7018,
|
|
"mean_token_accuracy": 0.15049159824848174,
|
|
"num_tokens": 4186725.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.665674304962158,
|
|
"epoch": 1.9707602339181287,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004909242965851688,
|
|
"loss": 6.7362,
|
|
"mean_token_accuracy": 0.15222102999687195,
|
|
"num_tokens": 4199278.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.739201164245605,
|
|
"epoch": 1.976608187134503,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004907920416868253,
|
|
"loss": 6.7216,
|
|
"mean_token_accuracy": 0.14601051211357116,
|
|
"num_tokens": 4213163.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.582593631744385,
|
|
"epoch": 1.9824561403508771,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004906588502355467,
|
|
"loss": 6.5807,
|
|
"mean_token_accuracy": 0.14897475093603135,
|
|
"num_tokens": 4225222.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.638074016571045,
|
|
"epoch": 1.9883040935672516,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004905247228093933,
|
|
"loss": 6.6304,
|
|
"mean_token_accuracy": 0.15521894097328187,
|
|
"num_tokens": 4236286.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.530340576171875,
|
|
"epoch": 1.9941520467836256,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004903896599904878,
|
|
"loss": 6.6647,
|
|
"mean_token_accuracy": 0.14791015088558196,
|
|
"num_tokens": 4247462.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.7336372375488285,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004902536623650125,
|
|
"loss": 6.7514,
|
|
"mean_token_accuracy": 0.1418044872581959,
|
|
"num_tokens": 4258968.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.613061475753784,
|
|
"epoch": 2.0058479532163744,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004901167305232068,
|
|
"loss": 6.3199,
|
|
"mean_token_accuracy": 0.15634880512952803,
|
|
"num_tokens": 4271022.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.678990364074707,
|
|
"epoch": 2.0116959064327484,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004899788650593648,
|
|
"loss": 6.3559,
|
|
"mean_token_accuracy": 0.1503312237560749,
|
|
"num_tokens": 4283348.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.541168832778931,
|
|
"epoch": 2.017543859649123,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004898400665718325,
|
|
"loss": 6.2881,
|
|
"mean_token_accuracy": 0.15968621373176575,
|
|
"num_tokens": 4295775.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.719831800460815,
|
|
"epoch": 2.023391812865497,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004897003356630052,
|
|
"loss": 6.3368,
|
|
"mean_token_accuracy": 0.1551282972097397,
|
|
"num_tokens": 4308408.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.478828239440918,
|
|
"epoch": 2.0292397660818713,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004895596729393252,
|
|
"loss": 6.2524,
|
|
"mean_token_accuracy": 0.1626296654343605,
|
|
"num_tokens": 4320614.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.523333215713501,
|
|
"epoch": 2.0350877192982457,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004894180790112787,
|
|
"loss": 6.2112,
|
|
"mean_token_accuracy": 0.16793770492076873,
|
|
"num_tokens": 4334118.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.555729389190674,
|
|
"epoch": 2.0409356725146197,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004892755544933934,
|
|
"loss": 6.3314,
|
|
"mean_token_accuracy": 0.1544162929058075,
|
|
"num_tokens": 4346486.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.517939233779908,
|
|
"epoch": 2.046783625730994,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004891321000042361,
|
|
"loss": 6.3986,
|
|
"mean_token_accuracy": 0.15014228150248526,
|
|
"num_tokens": 4358408.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.6038178443908695,
|
|
"epoch": 2.0526315789473686,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 6.3349,
|
|
"mean_token_accuracy": 0.15063187554478646,
|
|
"num_tokens": 4369816.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.564553737640381,
|
|
"epoch": 2.0584795321637426,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048884240360655,
|
|
"loss": 6.3834,
|
|
"mean_token_accuracy": 0.14581367149949073,
|
|
"num_tokens": 4382192.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.567172908782959,
|
|
"epoch": 2.064327485380117,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004886961629553242,
|
|
"loss": 6.3496,
|
|
"mean_token_accuracy": 0.15813596695661544,
|
|
"num_tokens": 4395546.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.526296615600586,
|
|
"epoch": 2.0701754385964914,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004885489948474272,
|
|
"loss": 6.3753,
|
|
"mean_token_accuracy": 0.15297993570566176,
|
|
"num_tokens": 4407988.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.5855114459991455,
|
|
"epoch": 2.0760233918128654,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004884008999215792,
|
|
"loss": 6.2988,
|
|
"mean_token_accuracy": 0.1582840144634247,
|
|
"num_tokens": 4419668.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.584148645401001,
|
|
"epoch": 2.08187134502924,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004882518788205225,
|
|
"loss": 6.3854,
|
|
"mean_token_accuracy": 0.14738436043262482,
|
|
"num_tokens": 4431493.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.5512090682983395,
|
|
"epoch": 2.087719298245614,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004881019321910198,
|
|
"loss": 6.353,
|
|
"mean_token_accuracy": 0.15487503558397292,
|
|
"num_tokens": 4443237.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.513033437728882,
|
|
"epoch": 2.0935672514619883,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048795106068384985,
|
|
"loss": 6.3583,
|
|
"mean_token_accuracy": 0.15871656984090804,
|
|
"num_tokens": 4455238.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.637407732009888,
|
|
"epoch": 2.0994152046783627,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000487799264953806,
|
|
"loss": 6.4203,
|
|
"mean_token_accuracy": 0.14721036553382874,
|
|
"num_tokens": 4469180.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.610451412200928,
|
|
"epoch": 2.1052631578947367,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004876465456596926,
|
|
"loss": 6.4533,
|
|
"mean_token_accuracy": 0.1517632856965065,
|
|
"num_tokens": 4481169.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.543406438827515,
|
|
"epoch": 2.111111111111111,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004874929034643223,
|
|
"loss": 6.3631,
|
|
"mean_token_accuracy": 0.14619019031524658,
|
|
"num_tokens": 4494051.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.454391813278198,
|
|
"epoch": 2.116959064327485,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004873383390345132,
|
|
"loss": 6.3206,
|
|
"mean_token_accuracy": 0.16392768919467926,
|
|
"num_tokens": 4508230.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.593087863922119,
|
|
"epoch": 2.1228070175438596,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00048718285304108613,
|
|
"loss": 6.3171,
|
|
"mean_token_accuracy": 0.15622008740901946,
|
|
"num_tokens": 4520555.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.444989585876465,
|
|
"epoch": 2.128654970760234,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048702644615886137,
|
|
"loss": 6.3646,
|
|
"mean_token_accuracy": 0.151460662484169,
|
|
"num_tokens": 4532126.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.621482419967651,
|
|
"epoch": 2.134502923976608,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048686911906665597,
|
|
"loss": 6.4521,
|
|
"mean_token_accuracy": 0.14833228290081024,
|
|
"num_tokens": 4543754.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.429253005981446,
|
|
"epoch": 2.1403508771929824,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048671087244728093,
|
|
"loss": 6.2887,
|
|
"mean_token_accuracy": 0.1584049344062805,
|
|
"num_tokens": 4555907.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.561292409896851,
|
|
"epoch": 2.146198830409357,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048655170698753774,
|
|
"loss": 6.4126,
|
|
"mean_token_accuracy": 0.15332231670618057,
|
|
"num_tokens": 4568461.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.525373888015747,
|
|
"epoch": 2.152046783625731,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000486391623378216,
|
|
"loss": 6.3628,
|
|
"mean_token_accuracy": 0.15623549669981002,
|
|
"num_tokens": 4581391.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.457386541366577,
|
|
"epoch": 2.1578947368421053,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004862306223140901,
|
|
"loss": 6.3492,
|
|
"mean_token_accuracy": 0.15107778757810592,
|
|
"num_tokens": 4594187.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.504654359817505,
|
|
"epoch": 2.1637426900584797,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048606870449391604,
|
|
"loss": 6.3854,
|
|
"mean_token_accuracy": 0.15558697730302812,
|
|
"num_tokens": 4607162.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.578766107559204,
|
|
"epoch": 2.1695906432748537,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048590587062042907,
|
|
"loss": 6.4904,
|
|
"mean_token_accuracy": 0.14629387259483337,
|
|
"num_tokens": 4619717.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.447008419036865,
|
|
"epoch": 2.175438596491228,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048574212140033965,
|
|
"loss": 6.2273,
|
|
"mean_token_accuracy": 0.16127332150936127,
|
|
"num_tokens": 4631938.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.438826036453247,
|
|
"epoch": 2.181286549707602,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004855774575443311,
|
|
"loss": 6.3345,
|
|
"mean_token_accuracy": 0.1587992399930954,
|
|
"num_tokens": 4644377.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.539951610565185,
|
|
"epoch": 2.1871345029239766,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048541187976705645,
|
|
"loss": 6.2964,
|
|
"mean_token_accuracy": 0.16265074014663697,
|
|
"num_tokens": 4656657.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.491901159286499,
|
|
"epoch": 2.192982456140351,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048524538878713517,
|
|
"loss": 6.2791,
|
|
"mean_token_accuracy": 0.16020860895514488,
|
|
"num_tokens": 4670156.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.5246663093566895,
|
|
"epoch": 2.198830409356725,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048507798532715005,
|
|
"loss": 6.3826,
|
|
"mean_token_accuracy": 0.14875538647174835,
|
|
"num_tokens": 4683564.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.444547891616821,
|
|
"epoch": 2.2046783625730995,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00048490967011364394,
|
|
"loss": 6.3352,
|
|
"mean_token_accuracy": 0.16067604571580887,
|
|
"num_tokens": 4696008.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.373475456237793,
|
|
"epoch": 2.2105263157894735,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004847404438771172,
|
|
"loss": 6.2113,
|
|
"mean_token_accuracy": 0.16405704319477082,
|
|
"num_tokens": 4707956.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.469513893127441,
|
|
"epoch": 2.216374269005848,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004845703073520236,
|
|
"loss": 6.4399,
|
|
"mean_token_accuracy": 0.14858177527785302,
|
|
"num_tokens": 4720683.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.590300607681274,
|
|
"epoch": 2.2222222222222223,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048439926127676796,
|
|
"loss": 6.3683,
|
|
"mean_token_accuracy": 0.15888102650642394,
|
|
"num_tokens": 4732524.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.465745878219605,
|
|
"epoch": 2.2280701754385963,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048422730639370256,
|
|
"loss": 6.4267,
|
|
"mean_token_accuracy": 0.15533056408166884,
|
|
"num_tokens": 4746483.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.493499422073365,
|
|
"epoch": 2.2339181286549707,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004840544434491238,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.15531336814165114,
|
|
"num_tokens": 4759543.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.377004480361938,
|
|
"epoch": 2.239766081871345,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048388067319326937,
|
|
"loss": 6.3158,
|
|
"mean_token_accuracy": 0.15956761240959166,
|
|
"num_tokens": 4770766.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.536472988128662,
|
|
"epoch": 2.245614035087719,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048370599638031464,
|
|
"loss": 6.3084,
|
|
"mean_token_accuracy": 0.16118213832378386,
|
|
"num_tokens": 4783157.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.363384056091308,
|
|
"epoch": 2.2514619883040936,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048353041376836937,
|
|
"loss": 6.3827,
|
|
"mean_token_accuracy": 0.16078438013792037,
|
|
"num_tokens": 4795874.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.52140212059021,
|
|
"epoch": 2.257309941520468,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004833539261194749,
|
|
"loss": 6.3334,
|
|
"mean_token_accuracy": 0.1594570055603981,
|
|
"num_tokens": 4806968.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.3824951171875,
|
|
"epoch": 2.263157894736842,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048317653419960007,
|
|
"loss": 6.3644,
|
|
"mean_token_accuracy": 0.155350361764431,
|
|
"num_tokens": 4819927.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.4782744407653805,
|
|
"epoch": 2.2690058479532165,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048299823877863857,
|
|
"loss": 6.3349,
|
|
"mean_token_accuracy": 0.160041144490242,
|
|
"num_tokens": 4831824.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.478532791137695,
|
|
"epoch": 2.2748538011695905,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004828190406304054,
|
|
"loss": 6.3812,
|
|
"mean_token_accuracy": 0.152671180665493,
|
|
"num_tokens": 4844405.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.451467847824096,
|
|
"epoch": 2.280701754385965,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004826389405326334,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.16256463527679443,
|
|
"num_tokens": 4856424.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.527578496932984,
|
|
"epoch": 2.2865497076023393,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004824579392669698,
|
|
"loss": 6.4768,
|
|
"mean_token_accuracy": 0.1487383469939232,
|
|
"num_tokens": 4869473.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.367396450042724,
|
|
"epoch": 2.2923976608187133,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048227603761897316,
|
|
"loss": 6.2911,
|
|
"mean_token_accuracy": 0.16065459549427033,
|
|
"num_tokens": 4881220.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.414208173751831,
|
|
"epoch": 2.2982456140350878,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004820932363781096,
|
|
"loss": 6.472,
|
|
"mean_token_accuracy": 0.14860487282276152,
|
|
"num_tokens": 4892766.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.45541353225708,
|
|
"epoch": 2.3040935672514617,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004819095363377496,
|
|
"loss": 6.3238,
|
|
"mean_token_accuracy": 0.15184853971004486,
|
|
"num_tokens": 4904549.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.419804286956787,
|
|
"epoch": 2.309941520467836,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00048172493829516465,
|
|
"loss": 6.3309,
|
|
"mean_token_accuracy": 0.15828023850917816,
|
|
"num_tokens": 4917204.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.4697606563568115,
|
|
"epoch": 2.3157894736842106,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004815394430515234,
|
|
"loss": 6.4003,
|
|
"mean_token_accuracy": 0.16001321077346803,
|
|
"num_tokens": 4928838.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.530691719055175,
|
|
"epoch": 2.3216374269005846,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004813530514118884,
|
|
"loss": 6.3734,
|
|
"mean_token_accuracy": 0.1554233744740486,
|
|
"num_tokens": 4940281.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.41666464805603,
|
|
"epoch": 2.327485380116959,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004811657641852129,
|
|
"loss": 6.3576,
|
|
"mean_token_accuracy": 0.15298293232917787,
|
|
"num_tokens": 4953355.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.50026125907898,
|
|
"epoch": 2.3333333333333335,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00048097758218433687,
|
|
"loss": 6.3659,
|
|
"mean_token_accuracy": 0.15752369314432144,
|
|
"num_tokens": 4966128.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.474570655822754,
|
|
"epoch": 2.3391812865497075,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00048078850622598355,
|
|
"loss": 6.3719,
|
|
"mean_token_accuracy": 0.15596262663602828,
|
|
"num_tokens": 4979409.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.3391812865497075,
|
|
"eval_entropy": 6.244584498503919,
|
|
"eval_loss": 6.810573577880859,
|
|
"eval_mean_token_accuracy": 0.14119860577427268,
|
|
"eval_num_tokens": 4979409.0,
|
|
"eval_runtime": 2.2734,
|
|
"eval_samples_per_second": 1190.728,
|
|
"eval_steps_per_second": 149.116,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.445643615722656,
|
|
"epoch": 2.345029239766082,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048059853713075637,
|
|
"loss": 6.3789,
|
|
"mean_token_accuracy": 0.15238464921712874,
|
|
"num_tokens": 4991086.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.449284791946411,
|
|
"epoch": 2.3508771929824563,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004804076757231347,
|
|
"loss": 6.3238,
|
|
"mean_token_accuracy": 0.16143079102039337,
|
|
"num_tokens": 5002837.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.443182754516601,
|
|
"epoch": 2.3567251461988303,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048021592283147073,
|
|
"loss": 6.3949,
|
|
"mean_token_accuracy": 0.15568165928125383,
|
|
"num_tokens": 5014688.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.489663219451904,
|
|
"epoch": 2.3625730994152048,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004800232792879858,
|
|
"loss": 6.375,
|
|
"mean_token_accuracy": 0.1525643140077591,
|
|
"num_tokens": 5026677.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.34787015914917,
|
|
"epoch": 2.3684210526315788,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047982974592876665,
|
|
"loss": 6.3747,
|
|
"mean_token_accuracy": 0.16030002385377884,
|
|
"num_tokens": 5039504.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.453823661804199,
|
|
"epoch": 2.374269005847953,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000479635323593762,
|
|
"loss": 6.274,
|
|
"mean_token_accuracy": 0.16640568822622298,
|
|
"num_tokens": 5051071.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.334503221511841,
|
|
"epoch": 2.3801169590643276,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047944001312677867,
|
|
"loss": 6.3277,
|
|
"mean_token_accuracy": 0.16389417946338652,
|
|
"num_tokens": 5064489.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.433465576171875,
|
|
"epoch": 2.3859649122807016,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004792438153754782,
|
|
"loss": 6.3748,
|
|
"mean_token_accuracy": 0.1552420049905777,
|
|
"num_tokens": 5076742.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.431836032867432,
|
|
"epoch": 2.391812865497076,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004790467311913727,
|
|
"loss": 6.3354,
|
|
"mean_token_accuracy": 0.16166526228189468,
|
|
"num_tokens": 5089194.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.544153165817261,
|
|
"epoch": 2.39766081871345,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004788487614298219,
|
|
"loss": 6.4077,
|
|
"mean_token_accuracy": 0.15238103717565538,
|
|
"num_tokens": 5100712.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.286121654510498,
|
|
"epoch": 2.4035087719298245,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004786499069500286,
|
|
"loss": 6.1947,
|
|
"mean_token_accuracy": 0.1636320561170578,
|
|
"num_tokens": 5114844.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.388913488388061,
|
|
"epoch": 2.409356725146199,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047845016861503556,
|
|
"loss": 6.3178,
|
|
"mean_token_accuracy": 0.15902430713176727,
|
|
"num_tokens": 5126626.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.4127631187438965,
|
|
"epoch": 2.415204678362573,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047824954729172136,
|
|
"loss": 6.3931,
|
|
"mean_token_accuracy": 0.15777807235717772,
|
|
"num_tokens": 5138527.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.561508512496948,
|
|
"epoch": 2.4210526315789473,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004780480438507972,
|
|
"loss": 6.4737,
|
|
"mean_token_accuracy": 0.14927541017532348,
|
|
"num_tokens": 5152045.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.493700265884399,
|
|
"epoch": 2.426900584795322,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047784565916680216,
|
|
"loss": 6.4653,
|
|
"mean_token_accuracy": 0.14986254870891572,
|
|
"num_tokens": 5164806.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.4221197128295895,
|
|
"epoch": 2.4327485380116958,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004776423941181005,
|
|
"loss": 6.4082,
|
|
"mean_token_accuracy": 0.14851786941289902,
|
|
"num_tokens": 5178676.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.49404993057251,
|
|
"epoch": 2.43859649122807,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00047743824958687694,
|
|
"loss": 6.3918,
|
|
"mean_token_accuracy": 0.1562908411026001,
|
|
"num_tokens": 5191183.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.386559343338012,
|
|
"epoch": 2.4444444444444446,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047723322645913354,
|
|
"loss": 6.3147,
|
|
"mean_token_accuracy": 0.16154536008834838,
|
|
"num_tokens": 5203460.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.416320276260376,
|
|
"epoch": 2.4502923976608186,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004770273256246853,
|
|
"loss": 6.3038,
|
|
"mean_token_accuracy": 0.1643177568912506,
|
|
"num_tokens": 5215362.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.427623701095581,
|
|
"epoch": 2.456140350877193,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004768205479771566,
|
|
"loss": 6.4046,
|
|
"mean_token_accuracy": 0.15239704698324202,
|
|
"num_tokens": 5228592.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.423169565200806,
|
|
"epoch": 2.461988304093567,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047661289441397724,
|
|
"loss": 6.355,
|
|
"mean_token_accuracy": 0.15616060495376588,
|
|
"num_tokens": 5242337.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.388077688217163,
|
|
"epoch": 2.4678362573099415,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047640436583637865,
|
|
"loss": 6.2911,
|
|
"mean_token_accuracy": 0.15808799415826796,
|
|
"num_tokens": 5255566.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.470604658126831,
|
|
"epoch": 2.473684210526316,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047619496314938977,
|
|
"loss": 6.3989,
|
|
"mean_token_accuracy": 0.15252429395914077,
|
|
"num_tokens": 5268638.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.355663251876831,
|
|
"epoch": 2.47953216374269,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004759846872618332,
|
|
"loss": 6.3615,
|
|
"mean_token_accuracy": 0.15340882539749146,
|
|
"num_tokens": 5280031.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.407628536224365,
|
|
"epoch": 2.4853801169590644,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047577353908632156,
|
|
"loss": 6.3039,
|
|
"mean_token_accuracy": 0.1645380064845085,
|
|
"num_tokens": 5291792.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.404614114761353,
|
|
"epoch": 2.4912280701754383,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047556151953925287,
|
|
"loss": 6.4094,
|
|
"mean_token_accuracy": 0.15307499915361406,
|
|
"num_tokens": 5303752.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.465473699569702,
|
|
"epoch": 2.497076023391813,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004753486295408073,
|
|
"loss": 6.351,
|
|
"mean_token_accuracy": 0.15327188968658448,
|
|
"num_tokens": 5315711.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.426405763626098,
|
|
"epoch": 2.502923976608187,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047513487001494257,
|
|
"loss": 6.3683,
|
|
"mean_token_accuracy": 0.16236188709735871,
|
|
"num_tokens": 5329449.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.4586772441864015,
|
|
"epoch": 2.5087719298245617,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047492024188939056,
|
|
"loss": 6.3991,
|
|
"mean_token_accuracy": 0.1537608489394188,
|
|
"num_tokens": 5341015.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.481842708587647,
|
|
"epoch": 2.5146198830409356,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004747047460956525,
|
|
"loss": 6.3922,
|
|
"mean_token_accuracy": 0.15520621985197067,
|
|
"num_tokens": 5352537.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.437920093536377,
|
|
"epoch": 2.52046783625731,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004744883835689958,
|
|
"loss": 6.3519,
|
|
"mean_token_accuracy": 0.16303111612796783,
|
|
"num_tokens": 5364617.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.420978021621704,
|
|
"epoch": 2.526315789473684,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004742711552484492,
|
|
"loss": 6.3798,
|
|
"mean_token_accuracy": 0.1547342821955681,
|
|
"num_tokens": 5377939.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.408878135681152,
|
|
"epoch": 2.5321637426900585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047405306207679926,
|
|
"loss": 6.305,
|
|
"mean_token_accuracy": 0.163915978372097,
|
|
"num_tokens": 5390111.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.379895401000977,
|
|
"epoch": 2.538011695906433,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004738341050005859,
|
|
"loss": 6.3846,
|
|
"mean_token_accuracy": 0.152874419093132,
|
|
"num_tokens": 5403690.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.504248809814453,
|
|
"epoch": 2.543859649122807,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004736142849700986,
|
|
"loss": 6.3599,
|
|
"mean_token_accuracy": 0.1571785405278206,
|
|
"num_tokens": 5416456.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.42173318862915,
|
|
"epoch": 2.5497076023391814,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004733936029393721,
|
|
"loss": 6.474,
|
|
"mean_token_accuracy": 0.14887326657772065,
|
|
"num_tokens": 5429490.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.463390684127807,
|
|
"epoch": 2.5555555555555554,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004731720598661821,
|
|
"loss": 6.3615,
|
|
"mean_token_accuracy": 0.1581603243947029,
|
|
"num_tokens": 5441382.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.4675092697143555,
|
|
"epoch": 2.56140350877193,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004729496567120415,
|
|
"loss": 6.4369,
|
|
"mean_token_accuracy": 0.156419774889946,
|
|
"num_tokens": 5453135.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.351563835144043,
|
|
"epoch": 2.5672514619883042,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000472726394442196,
|
|
"loss": 6.2849,
|
|
"mean_token_accuracy": 0.17103164792060851,
|
|
"num_tokens": 5467026.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.407415962219238,
|
|
"epoch": 2.573099415204678,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047250227402561976,
|
|
"loss": 6.3949,
|
|
"mean_token_accuracy": 0.1573207288980484,
|
|
"num_tokens": 5479328.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.4740362644195555,
|
|
"epoch": 2.5789473684210527,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004722772964350115,
|
|
"loss": 6.3941,
|
|
"mean_token_accuracy": 0.15537292510271072,
|
|
"num_tokens": 5490843.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.365005779266357,
|
|
"epoch": 2.5847953216374266,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047205146264679015,
|
|
"loss": 6.308,
|
|
"mean_token_accuracy": 0.15925779044628144,
|
|
"num_tokens": 5502721.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.40563702583313,
|
|
"epoch": 2.590643274853801,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047182477364109045,
|
|
"loss": 6.3553,
|
|
"mean_token_accuracy": 0.15847325325012207,
|
|
"num_tokens": 5514190.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.4377704620361325,
|
|
"epoch": 2.5964912280701755,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047159723040175896,
|
|
"loss": 6.2704,
|
|
"mean_token_accuracy": 0.16171683073043824,
|
|
"num_tokens": 5528259.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.409121179580689,
|
|
"epoch": 2.60233918128655,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047136883391634966,
|
|
"loss": 6.4424,
|
|
"mean_token_accuracy": 0.15242915600538254,
|
|
"num_tokens": 5540375.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.443257904052734,
|
|
"epoch": 2.608187134502924,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004711395851761198,
|
|
"loss": 6.3391,
|
|
"mean_token_accuracy": 0.16303063035011292,
|
|
"num_tokens": 5552848.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.385362005233764,
|
|
"epoch": 2.6140350877192984,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004709094851760251,
|
|
"loss": 6.3218,
|
|
"mean_token_accuracy": 0.16016172021627426,
|
|
"num_tokens": 5565988.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.409554672241211,
|
|
"epoch": 2.6198830409356724,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004706785349147163,
|
|
"loss": 6.3978,
|
|
"mean_token_accuracy": 0.15372055619955063,
|
|
"num_tokens": 5578110.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.422667646408081,
|
|
"epoch": 2.625730994152047,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000470446735394534,
|
|
"loss": 6.3774,
|
|
"mean_token_accuracy": 0.1569847896695137,
|
|
"num_tokens": 5590997.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.401986503601075,
|
|
"epoch": 2.6315789473684212,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004702140876215047,
|
|
"loss": 6.2686,
|
|
"mean_token_accuracy": 0.16344404965639114,
|
|
"num_tokens": 5602407.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.382628488540649,
|
|
"epoch": 2.6374269005847952,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004699805926053365,
|
|
"loss": 6.3799,
|
|
"mean_token_accuracy": 0.15904879868030547,
|
|
"num_tokens": 5615298.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.403160190582275,
|
|
"epoch": 2.6432748538011697,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046974625135941445,
|
|
"loss": 6.3306,
|
|
"mean_token_accuracy": 0.1589551791548729,
|
|
"num_tokens": 5627197.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.531083774566651,
|
|
"epoch": 2.6491228070175437,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046951106490079636,
|
|
"loss": 6.3901,
|
|
"mean_token_accuracy": 0.14688626676797867,
|
|
"num_tokens": 5640345.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.354160737991333,
|
|
"epoch": 2.654970760233918,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046927503425020847,
|
|
"loss": 6.4245,
|
|
"mean_token_accuracy": 0.15789488106966018,
|
|
"num_tokens": 5653866.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.4712237358093265,
|
|
"epoch": 2.6608187134502925,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046903816043204045,
|
|
"loss": 6.3754,
|
|
"mean_token_accuracy": 0.1609050899744034,
|
|
"num_tokens": 5666618.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.276658153533935,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004688004444743419,
|
|
"loss": 6.3325,
|
|
"mean_token_accuracy": 0.16040945500135423,
|
|
"num_tokens": 5679165.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.471549844741821,
|
|
"epoch": 2.672514619883041,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000468561887408817,
|
|
"loss": 6.3708,
|
|
"mean_token_accuracy": 0.15915319621562957,
|
|
"num_tokens": 5691387.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.2920068264007565,
|
|
"epoch": 2.678362573099415,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004683224902708206,
|
|
"loss": 6.2909,
|
|
"mean_token_accuracy": 0.16518297046422958,
|
|
"num_tokens": 5703804.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.511602783203125,
|
|
"epoch": 2.6842105263157894,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004680822540993534,
|
|
"loss": 6.452,
|
|
"mean_token_accuracy": 0.15097396522760392,
|
|
"num_tokens": 5716338.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.409046220779419,
|
|
"epoch": 2.690058479532164,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046784117993705753,
|
|
"loss": 6.3906,
|
|
"mean_token_accuracy": 0.1572423741221428,
|
|
"num_tokens": 5729644.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.403546667098999,
|
|
"epoch": 2.6959064327485383,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046759926883021234,
|
|
"loss": 6.3193,
|
|
"mean_token_accuracy": 0.16220491826534272,
|
|
"num_tokens": 5741357.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.446309185028076,
|
|
"epoch": 2.7017543859649122,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046735652182872925,
|
|
"loss": 6.4356,
|
|
"mean_token_accuracy": 0.15013170689344407,
|
|
"num_tokens": 5754783.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.367806100845337,
|
|
"epoch": 2.7076023391812867,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004671129399861476,
|
|
"loss": 6.2865,
|
|
"mean_token_accuracy": 0.16693611145019532,
|
|
"num_tokens": 5766845.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.487013912200927,
|
|
"epoch": 2.7134502923976607,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00046686852435963016,
|
|
"loss": 6.442,
|
|
"mean_token_accuracy": 0.1500522792339325,
|
|
"num_tokens": 5779340.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.3810874938964846,
|
|
"epoch": 2.719298245614035,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004666232760099583,
|
|
"loss": 6.2877,
|
|
"mean_token_accuracy": 0.16884452253580093,
|
|
"num_tokens": 5791305.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.435328722000122,
|
|
"epoch": 2.7251461988304095,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046637719600152737,
|
|
"loss": 6.3852,
|
|
"mean_token_accuracy": 0.15519808381795883,
|
|
"num_tokens": 5805005.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.335537528991699,
|
|
"epoch": 2.7309941520467835,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046613028540234226,
|
|
"loss": 6.2757,
|
|
"mean_token_accuracy": 0.16379417330026627,
|
|
"num_tokens": 5817875.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.439946603775025,
|
|
"epoch": 2.736842105263158,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046588254528401264,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.15380895882844925,
|
|
"num_tokens": 5830529.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.383373785018921,
|
|
"epoch": 2.742690058479532,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004656339767217485,
|
|
"loss": 6.3354,
|
|
"mean_token_accuracy": 0.1607013463973999,
|
|
"num_tokens": 5842505.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.419260168075562,
|
|
"epoch": 2.7485380116959064,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046538458079435516,
|
|
"loss": 6.316,
|
|
"mean_token_accuracy": 0.15341950505971907,
|
|
"num_tokens": 5854215.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.422201204299927,
|
|
"epoch": 2.754385964912281,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004651343585842287,
|
|
"loss": 6.3595,
|
|
"mean_token_accuracy": 0.15550109148025512,
|
|
"num_tokens": 5866484.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.393207168579101,
|
|
"epoch": 2.760233918128655,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046488331117735176,
|
|
"loss": 6.3654,
|
|
"mean_token_accuracy": 0.15960022509098054,
|
|
"num_tokens": 5879131.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.420015954971314,
|
|
"epoch": 2.7660818713450293,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004646314396632878,
|
|
"loss": 6.3798,
|
|
"mean_token_accuracy": 0.15467930138111113,
|
|
"num_tokens": 5891776.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.420463514328003,
|
|
"epoch": 2.7719298245614032,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004643787451351774,
|
|
"loss": 6.2981,
|
|
"mean_token_accuracy": 0.15805638134479522,
|
|
"num_tokens": 5904246.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.319274616241455,
|
|
"epoch": 2.7777777777777777,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046412522868973297,
|
|
"loss": 6.3117,
|
|
"mean_token_accuracy": 0.16229696124792098,
|
|
"num_tokens": 5917278.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.393973112106323,
|
|
"epoch": 2.783625730994152,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046387089142723386,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.14831992760300636,
|
|
"num_tokens": 5930157.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.429065275192261,
|
|
"epoch": 2.7894736842105265,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00046361573445152224,
|
|
"loss": 6.2455,
|
|
"mean_token_accuracy": 0.16678779423236847,
|
|
"num_tokens": 5942821.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.359903049468994,
|
|
"epoch": 2.7953216374269005,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046335975886999754,
|
|
"loss": 6.3201,
|
|
"mean_token_accuracy": 0.16159365475177764,
|
|
"num_tokens": 5954123.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.392605876922607,
|
|
"epoch": 2.801169590643275,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046310296579361225,
|
|
"loss": 6.2983,
|
|
"mean_token_accuracy": 0.1590059816837311,
|
|
"num_tokens": 5965629.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.369566631317139,
|
|
"epoch": 2.807017543859649,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004628453563368666,
|
|
"loss": 6.3403,
|
|
"mean_token_accuracy": 0.16067956686019896,
|
|
"num_tokens": 5979157.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.424136638641357,
|
|
"epoch": 2.8128654970760234,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004625869316178043,
|
|
"loss": 6.2869,
|
|
"mean_token_accuracy": 0.17058369666337966,
|
|
"num_tokens": 5992729.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.247052431106567,
|
|
"epoch": 2.818713450292398,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000462327692758007,
|
|
"loss": 6.4088,
|
|
"mean_token_accuracy": 0.15289342552423477,
|
|
"num_tokens": 6004685.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.410907745361328,
|
|
"epoch": 2.824561403508772,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046206764088259003,
|
|
"loss": 6.3736,
|
|
"mean_token_accuracy": 0.16130629777908326,
|
|
"num_tokens": 6016139.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.358149909973145,
|
|
"epoch": 2.8304093567251463,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046180677712019707,
|
|
"loss": 6.3559,
|
|
"mean_token_accuracy": 0.1639443278312683,
|
|
"num_tokens": 6028236.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.369370174407959,
|
|
"epoch": 2.8362573099415203,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004615451026029957,
|
|
"loss": 6.3645,
|
|
"mean_token_accuracy": 0.1627436622977257,
|
|
"num_tokens": 6040017.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.4513365745544435,
|
|
"epoch": 2.8421052631578947,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000461282618466672,
|
|
"loss": 6.3813,
|
|
"mean_token_accuracy": 0.1559862807393074,
|
|
"num_tokens": 6052962.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.277508544921875,
|
|
"epoch": 2.847953216374269,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046101932585042597,
|
|
"loss": 6.35,
|
|
"mean_token_accuracy": 0.15423065721988677,
|
|
"num_tokens": 6065187.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.3818567276000975,
|
|
"epoch": 2.853801169590643,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046075522589696636,
|
|
"loss": 6.2982,
|
|
"mean_token_accuracy": 0.1641393393278122,
|
|
"num_tokens": 6078347.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.39556245803833,
|
|
"epoch": 2.8596491228070176,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000460490319752506,
|
|
"loss": 6.3867,
|
|
"mean_token_accuracy": 0.15675798803567886,
|
|
"num_tokens": 6091156.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.361089658737183,
|
|
"epoch": 2.8654970760233915,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00046022460856675663,
|
|
"loss": 6.3249,
|
|
"mean_token_accuracy": 0.16081140488386153,
|
|
"num_tokens": 6103800.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.35357403755188,
|
|
"epoch": 2.871345029239766,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004599580934929237,
|
|
"loss": 6.342,
|
|
"mean_token_accuracy": 0.16039023846387862,
|
|
"num_tokens": 6116469.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.349758434295654,
|
|
"epoch": 2.8771929824561404,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004596907756877018,
|
|
"loss": 6.3853,
|
|
"mean_token_accuracy": 0.15386279225349425,
|
|
"num_tokens": 6129297.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.435890197753906,
|
|
"epoch": 2.883040935672515,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045942265631126935,
|
|
"loss": 6.3576,
|
|
"mean_token_accuracy": 0.15894166380167007,
|
|
"num_tokens": 6141139.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.387912940979004,
|
|
"epoch": 2.888888888888889,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004591537365272837,
|
|
"loss": 6.3091,
|
|
"mean_token_accuracy": 0.16368301808834076,
|
|
"num_tokens": 6154271.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.409696483612061,
|
|
"epoch": 2.8947368421052633,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000458884017502876,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.15825672224164009,
|
|
"num_tokens": 6166296.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.250841951370239,
|
|
"epoch": 2.9005847953216373,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004586135004086461,
|
|
"loss": 6.1801,
|
|
"mean_token_accuracy": 0.16876088231801986,
|
|
"num_tokens": 6178146.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.41408371925354,
|
|
"epoch": 2.9064327485380117,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004583421864186577,
|
|
"loss": 6.3381,
|
|
"mean_token_accuracy": 0.15795808732509614,
|
|
"num_tokens": 6190591.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.353419542312622,
|
|
"epoch": 2.912280701754386,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004580700767104328,
|
|
"loss": 6.317,
|
|
"mean_token_accuracy": 0.16341704577207566,
|
|
"num_tokens": 6201996.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.394930791854859,
|
|
"epoch": 2.91812865497076,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045779717246494724,
|
|
"loss": 6.3896,
|
|
"mean_token_accuracy": 0.15675280690193177,
|
|
"num_tokens": 6214640.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.486262416839599,
|
|
"epoch": 2.9239766081871346,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004575234748666249,
|
|
"loss": 6.4475,
|
|
"mean_token_accuracy": 0.15161905139684678,
|
|
"num_tokens": 6227811.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.9239766081871346,
|
|
"eval_entropy": 6.3679620073256595,
|
|
"eval_loss": 6.7001118659973145,
|
|
"eval_mean_token_accuracy": 0.14387351291248668,
|
|
"eval_num_tokens": 6227811.0,
|
|
"eval_runtime": 2.2778,
|
|
"eval_samples_per_second": 1188.439,
|
|
"eval_steps_per_second": 148.829,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.447972249984741,
|
|
"epoch": 2.9298245614035086,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000457248985103333,
|
|
"loss": 6.3699,
|
|
"mean_token_accuracy": 0.1568946972489357,
|
|
"num_tokens": 6240897.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.364565420150757,
|
|
"epoch": 2.935672514619883,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004569737043663768,
|
|
"loss": 6.367,
|
|
"mean_token_accuracy": 0.15833095759153365,
|
|
"num_tokens": 6252416.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.411330461502075,
|
|
"epoch": 2.9415204678362574,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004566976338504945,
|
|
"loss": 6.4325,
|
|
"mean_token_accuracy": 0.14796094745397567,
|
|
"num_tokens": 6265203.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.372057485580444,
|
|
"epoch": 2.9473684210526314,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045642077475385186,
|
|
"loss": 6.3175,
|
|
"mean_token_accuracy": 0.16635856479406358,
|
|
"num_tokens": 6277384.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.38411660194397,
|
|
"epoch": 2.953216374269006,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004561431282780372,
|
|
"loss": 6.3569,
|
|
"mean_token_accuracy": 0.1604871705174446,
|
|
"num_tokens": 6291114.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.306672811508179,
|
|
"epoch": 2.95906432748538,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00045586469562805624,
|
|
"loss": 6.2933,
|
|
"mean_token_accuracy": 0.1560649335384369,
|
|
"num_tokens": 6304604.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.376500940322876,
|
|
"epoch": 2.9649122807017543,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004555854780123265,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.15167758017778396,
|
|
"num_tokens": 6316434.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.408717250823974,
|
|
"epoch": 2.9707602339181287,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004553054766426725,
|
|
"loss": 6.2689,
|
|
"mean_token_accuracy": 0.15962534844875337,
|
|
"num_tokens": 6328147.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.301998233795166,
|
|
"epoch": 2.976608187134503,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004550246927343203,
|
|
"loss": 6.2661,
|
|
"mean_token_accuracy": 0.16512599289417268,
|
|
"num_tokens": 6339836.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.4500223159790036,
|
|
"epoch": 2.982456140350877,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045474312750589226,
|
|
"loss": 6.3751,
|
|
"mean_token_accuracy": 0.1530895672738552,
|
|
"num_tokens": 6352093.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.309706306457519,
|
|
"epoch": 2.9883040935672516,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00045446078217940155,
|
|
"loss": 6.2957,
|
|
"mean_token_accuracy": 0.1603766605257988,
|
|
"num_tokens": 6364748.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.274790906906128,
|
|
"epoch": 2.9941520467836256,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045417765798024716,
|
|
"loss": 6.1146,
|
|
"mean_token_accuracy": 0.16290974020957946,
|
|
"num_tokens": 6376384.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.364281797409058,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004538937561372086,
|
|
"loss": 6.3485,
|
|
"mean_token_accuracy": 0.1564707338809967,
|
|
"num_tokens": 6388452.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.385696840286255,
|
|
"epoch": 3.0058479532163744,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004536090778824398,
|
|
"loss": 5.9455,
|
|
"mean_token_accuracy": 0.16394055783748626,
|
|
"num_tokens": 6401876.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.325968074798584,
|
|
"epoch": 3.0116959064327484,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004533236244514652,
|
|
"loss": 5.9787,
|
|
"mean_token_accuracy": 0.16571004986763,
|
|
"num_tokens": 6413607.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.363959264755249,
|
|
"epoch": 3.017543859649123,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045303739708317303,
|
|
"loss": 5.9905,
|
|
"mean_token_accuracy": 0.1685149371623993,
|
|
"num_tokens": 6426295.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.260405969619751,
|
|
"epoch": 3.023391812865497,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004527503970198105,
|
|
"loss": 5.8862,
|
|
"mean_token_accuracy": 0.17230862528085708,
|
|
"num_tokens": 6438257.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.326606035232544,
|
|
"epoch": 3.0292397660818713,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004524626255069786,
|
|
"loss": 6.0472,
|
|
"mean_token_accuracy": 0.1555703967809677,
|
|
"num_tokens": 6451020.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.297233438491821,
|
|
"epoch": 3.0350877192982457,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00045217408379362635,
|
|
"loss": 6.0091,
|
|
"mean_token_accuracy": 0.1690215915441513,
|
|
"num_tokens": 6463528.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.244835615158081,
|
|
"epoch": 3.0409356725146197,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045188477313204536,
|
|
"loss": 5.8807,
|
|
"mean_token_accuracy": 0.17526089102029802,
|
|
"num_tokens": 6475030.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.393086957931518,
|
|
"epoch": 3.046783625730994,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00045159469477786457,
|
|
"loss": 5.9884,
|
|
"mean_token_accuracy": 0.17051160335540771,
|
|
"num_tokens": 6487889.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.2137922763824465,
|
|
"epoch": 3.0526315789473686,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004513038499900451,
|
|
"loss": 5.9341,
|
|
"mean_token_accuracy": 0.1741265207529068,
|
|
"num_tokens": 6501806.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.225668668746948,
|
|
"epoch": 3.0584795321637426,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000451012240030874,
|
|
"loss": 5.8621,
|
|
"mean_token_accuracy": 0.17396994680166245,
|
|
"num_tokens": 6512980.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.362692213058471,
|
|
"epoch": 3.064327485380117,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045071986616595946,
|
|
"loss": 5.9823,
|
|
"mean_token_accuracy": 0.15765693187713622,
|
|
"num_tokens": 6525156.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.276885890960694,
|
|
"epoch": 3.0701754385964914,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00045042672966422506,
|
|
"loss": 5.9774,
|
|
"mean_token_accuracy": 0.17088396698236466,
|
|
"num_tokens": 6538048.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.162761640548706,
|
|
"epoch": 3.0760233918128654,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004501328317979041,
|
|
"loss": 5.8537,
|
|
"mean_token_accuracy": 0.17376164197921753,
|
|
"num_tokens": 6549088.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.2773816108703615,
|
|
"epoch": 3.08187134502924,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004498381738425346,
|
|
"loss": 6.0316,
|
|
"mean_token_accuracy": 0.15929657816886902,
|
|
"num_tokens": 6561597.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.277020883560181,
|
|
"epoch": 3.087719298245614,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044954275707695304,
|
|
"loss": 5.9434,
|
|
"mean_token_accuracy": 0.16779058426618576,
|
|
"num_tokens": 6573253.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.279333066940308,
|
|
"epoch": 3.0935672514619883,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004492465827832894,
|
|
"loss": 6.0267,
|
|
"mean_token_accuracy": 0.16542163789272307,
|
|
"num_tokens": 6584619.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.264507198333741,
|
|
"epoch": 3.0994152046783627,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044894965224696144,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.16943052113056184,
|
|
"num_tokens": 6597514.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.3290825366973875,
|
|
"epoch": 3.1052631578947367,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044865196675666886,
|
|
"loss": 6.0226,
|
|
"mean_token_accuracy": 0.16399950683116912,
|
|
"num_tokens": 6610486.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.303022146224976,
|
|
"epoch": 3.111111111111111,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004483535276043881,
|
|
"loss": 5.9443,
|
|
"mean_token_accuracy": 0.17359011471271515,
|
|
"num_tokens": 6622936.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.154239749908447,
|
|
"epoch": 3.116959064327485,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004480543360853665,
|
|
"loss": 5.8152,
|
|
"mean_token_accuracy": 0.1817859560251236,
|
|
"num_tokens": 6635841.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.322794437408447,
|
|
"epoch": 3.1228070175438596,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004477543934981167,
|
|
"loss": 6.0443,
|
|
"mean_token_accuracy": 0.16220160573720932,
|
|
"num_tokens": 6648611.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.237705373764038,
|
|
"epoch": 3.128654970760234,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044745370114441103,
|
|
"loss": 5.9954,
|
|
"mean_token_accuracy": 0.16805303543806077,
|
|
"num_tokens": 6660557.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.288750123977661,
|
|
"epoch": 3.134502923976608,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044715226032927585,
|
|
"loss": 5.995,
|
|
"mean_token_accuracy": 0.16606099158525467,
|
|
"num_tokens": 6672172.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.28822660446167,
|
|
"epoch": 3.1403508771929824,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004468500723609859,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.16786604523658752,
|
|
"num_tokens": 6685244.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.335166311264038,
|
|
"epoch": 3.146198830409357,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044654713855105864,
|
|
"loss": 6.0474,
|
|
"mean_token_accuracy": 0.168092343211174,
|
|
"num_tokens": 6698401.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.183216619491577,
|
|
"epoch": 3.152046783625731,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004462434602142486,
|
|
"loss": 5.9933,
|
|
"mean_token_accuracy": 0.16719063818454744,
|
|
"num_tokens": 6710562.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.300883626937866,
|
|
"epoch": 3.1578947368421053,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004459390386685414,
|
|
"loss": 5.9741,
|
|
"mean_token_accuracy": 0.16917198300361633,
|
|
"num_tokens": 6724046.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.205930137634278,
|
|
"epoch": 3.1637426900584797,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00044563387523514854,
|
|
"loss": 5.9398,
|
|
"mean_token_accuracy": 0.17038815468549728,
|
|
"num_tokens": 6736028.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.191318225860596,
|
|
"epoch": 3.1695906432748537,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004453279712385011,
|
|
"loss": 5.9298,
|
|
"mean_token_accuracy": 0.17975495308637618,
|
|
"num_tokens": 6748828.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.244331216812133,
|
|
"epoch": 3.175438596491228,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004450213280062444,
|
|
"loss": 6.0125,
|
|
"mean_token_accuracy": 0.16804856061935425,
|
|
"num_tokens": 6761029.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.214777278900146,
|
|
"epoch": 3.181286549707602,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044471394686923206,
|
|
"loss": 5.9348,
|
|
"mean_token_accuracy": 0.1671859934926033,
|
|
"num_tokens": 6772008.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.259116220474243,
|
|
"epoch": 3.1871345029239766,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004444058291615202,
|
|
"loss": 6.004,
|
|
"mean_token_accuracy": 0.1627207562327385,
|
|
"num_tokens": 6784913.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.246962594985962,
|
|
"epoch": 3.192982456140351,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004440969762203618,
|
|
"loss": 6.0286,
|
|
"mean_token_accuracy": 0.1614857569336891,
|
|
"num_tokens": 6797354.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.2715987205505375,
|
|
"epoch": 3.198830409356725,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004437873893862008,
|
|
"loss": 6.03,
|
|
"mean_token_accuracy": 0.1656293749809265,
|
|
"num_tokens": 6810010.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.213133382797241,
|
|
"epoch": 3.2046783625730995,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044347707000266627,
|
|
"loss": 5.8755,
|
|
"mean_token_accuracy": 0.17746605724096298,
|
|
"num_tokens": 6821264.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.190156173706055,
|
|
"epoch": 3.2105263157894735,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044316601941656655,
|
|
"loss": 6.0178,
|
|
"mean_token_accuracy": 0.16163998246192932,
|
|
"num_tokens": 6833578.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.2568563461303714,
|
|
"epoch": 3.216374269005848,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004428542389778834,
|
|
"loss": 5.8727,
|
|
"mean_token_accuracy": 0.17870430648326874,
|
|
"num_tokens": 6845744.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.1800566673278805,
|
|
"epoch": 3.2222222222222223,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004425417300397665,
|
|
"loss": 6.0205,
|
|
"mean_token_accuracy": 0.16778453886508943,
|
|
"num_tokens": 6857970.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.309784173965454,
|
|
"epoch": 3.2280701754385963,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044222849395852683,
|
|
"loss": 6.0367,
|
|
"mean_token_accuracy": 0.1630901277065277,
|
|
"num_tokens": 6870586.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.237935781478882,
|
|
"epoch": 3.2339181286549707,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044191453209363165,
|
|
"loss": 6.0678,
|
|
"mean_token_accuracy": 0.16293592303991317,
|
|
"num_tokens": 6883667.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.23095703125,
|
|
"epoch": 3.239766081871345,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004415998458076978,
|
|
"loss": 6.0468,
|
|
"mean_token_accuracy": 0.16571044474840163,
|
|
"num_tokens": 6896709.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.243925714492798,
|
|
"epoch": 3.245614035087719,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00044128443646648634,
|
|
"loss": 5.9543,
|
|
"mean_token_accuracy": 0.17564820051193236,
|
|
"num_tokens": 6909899.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.22199125289917,
|
|
"epoch": 3.2514619883040936,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044096830543889653,
|
|
"loss": 6.0247,
|
|
"mean_token_accuracy": 0.16570626199245453,
|
|
"num_tokens": 6922751.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.247598361968994,
|
|
"epoch": 3.257309941520468,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004406514540969596,
|
|
"loss": 5.9863,
|
|
"mean_token_accuracy": 0.17584278881549836,
|
|
"num_tokens": 6936011.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.248948287963867,
|
|
"epoch": 3.263157894736842,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044033388381583324,
|
|
"loss": 6.0991,
|
|
"mean_token_accuracy": 0.15915966033935547,
|
|
"num_tokens": 6948831.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.250845289230346,
|
|
"epoch": 3.2690058479532165,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044001559597379503,
|
|
"loss": 5.9181,
|
|
"mean_token_accuracy": 0.17828963398933412,
|
|
"num_tokens": 6961329.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.146090030670166,
|
|
"epoch": 3.2748538011695905,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004396965919522373,
|
|
"loss": 5.9652,
|
|
"mean_token_accuracy": 0.17339591979980468,
|
|
"num_tokens": 6973075.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.204458570480346,
|
|
"epoch": 3.280701754385965,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043937687313565997,
|
|
"loss": 5.9621,
|
|
"mean_token_accuracy": 0.17302401214838029,
|
|
"num_tokens": 6985592.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.256625604629517,
|
|
"epoch": 3.2865497076023393,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004390564409116661,
|
|
"loss": 6.0016,
|
|
"mean_token_accuracy": 0.17008765041828156,
|
|
"num_tokens": 6998075.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.181208372116089,
|
|
"epoch": 3.2923976608187133,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043873529667095423,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.17330780029296874,
|
|
"num_tokens": 7008790.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.264780902862549,
|
|
"epoch": 3.2982456140350878,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043841344180731346,
|
|
"loss": 6.0486,
|
|
"mean_token_accuracy": 0.1657038301229477,
|
|
"num_tokens": 7022031.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.227780055999756,
|
|
"epoch": 3.3040935672514617,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000438090877717617,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.16618477702140808,
|
|
"num_tokens": 7033491.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.263735961914063,
|
|
"epoch": 3.309941520467836,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004377676058018161,
|
|
"loss": 5.9906,
|
|
"mean_token_accuracy": 0.16891173124313355,
|
|
"num_tokens": 7045472.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.172624301910401,
|
|
"epoch": 3.3157894736842106,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004374436274629341,
|
|
"loss": 5.9318,
|
|
"mean_token_accuracy": 0.16880391538143158,
|
|
"num_tokens": 7059050.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.258979606628418,
|
|
"epoch": 3.3216374269005846,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004371189441070602,
|
|
"loss": 6.0633,
|
|
"mean_token_accuracy": 0.1651046544313431,
|
|
"num_tokens": 7070604.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.183944797515869,
|
|
"epoch": 3.327485380116959,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004367935571433433,
|
|
"loss": 5.982,
|
|
"mean_token_accuracy": 0.17618281543254852,
|
|
"num_tokens": 7083687.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.221978092193604,
|
|
"epoch": 3.3333333333333335,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004364674679839864,
|
|
"loss": 6.0221,
|
|
"mean_token_accuracy": 0.1694357842206955,
|
|
"num_tokens": 7095846.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.247653579711914,
|
|
"epoch": 3.3391812865497075,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004361406780442398,
|
|
"loss": 5.9794,
|
|
"mean_token_accuracy": 0.16668584048748017,
|
|
"num_tokens": 7108051.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.251958322525025,
|
|
"epoch": 3.345029239766082,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043581318874239504,
|
|
"loss": 6.0401,
|
|
"mean_token_accuracy": 0.161568284034729,
|
|
"num_tokens": 7120699.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.213390874862671,
|
|
"epoch": 3.3508771929824563,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00043548500149977946,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.17091115117073058,
|
|
"num_tokens": 7134039.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.268674278259278,
|
|
"epoch": 3.3567251461988303,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004351561177407491,
|
|
"loss": 6.0295,
|
|
"mean_token_accuracy": 0.16179833561182022,
|
|
"num_tokens": 7145196.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.158988666534424,
|
|
"epoch": 3.3625730994152048,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004348265388926831,
|
|
"loss": 6.0532,
|
|
"mean_token_accuracy": 0.15928393006324768,
|
|
"num_tokens": 7156818.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.207103300094604,
|
|
"epoch": 3.3684210526315788,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004344962663859771,
|
|
"loss": 5.9047,
|
|
"mean_token_accuracy": 0.16933748573064805,
|
|
"num_tokens": 7169282.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.218513298034668,
|
|
"epoch": 3.374269005847953,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043416530165403777,
|
|
"loss": 6.0129,
|
|
"mean_token_accuracy": 0.17008000761270523,
|
|
"num_tokens": 7180663.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.1580695629119875,
|
|
"epoch": 3.3801169590643276,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004338336461332756,
|
|
"loss": 6.0213,
|
|
"mean_token_accuracy": 0.16719587296247482,
|
|
"num_tokens": 7192529.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.2996687412261965,
|
|
"epoch": 3.3859649122807016,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043350130126309936,
|
|
"loss": 6.0187,
|
|
"mean_token_accuracy": 0.16427617371082306,
|
|
"num_tokens": 7204484.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.178333759307861,
|
|
"epoch": 3.391812865497076,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043316826848590955,
|
|
"loss": 5.9632,
|
|
"mean_token_accuracy": 0.16770129203796386,
|
|
"num_tokens": 7217312.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.126302576065063,
|
|
"epoch": 3.39766081871345,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004328345492470924,
|
|
"loss": 5.9567,
|
|
"mean_token_accuracy": 0.17035896927118302,
|
|
"num_tokens": 7229032.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.224781942367554,
|
|
"epoch": 3.4035087719298245,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004325001449950134,
|
|
"loss": 6.0456,
|
|
"mean_token_accuracy": 0.16447727084159852,
|
|
"num_tokens": 7242827.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.10618371963501,
|
|
"epoch": 3.409356725146199,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004321650571810109,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.1736912399530411,
|
|
"num_tokens": 7256152.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.269802236557007,
|
|
"epoch": 3.415204678362573,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004318292872593899,
|
|
"loss": 6.0582,
|
|
"mean_token_accuracy": 0.163247187435627,
|
|
"num_tokens": 7268122.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.287968540191651,
|
|
"epoch": 3.4210526315789473,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004314928366874162,
|
|
"loss": 6.0484,
|
|
"mean_token_accuracy": 0.17468223869800567,
|
|
"num_tokens": 7281914.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.052388477325439,
|
|
"epoch": 3.426900584795322,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004311557069253092,
|
|
"loss": 5.9439,
|
|
"mean_token_accuracy": 0.1708640545606613,
|
|
"num_tokens": 7294555.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.196170806884766,
|
|
"epoch": 3.4327485380116958,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004308178994362364,
|
|
"loss": 6.0224,
|
|
"mean_token_accuracy": 0.1732296958565712,
|
|
"num_tokens": 7307606.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.247723245620728,
|
|
"epoch": 3.43859649122807,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00043047941568630644,
|
|
"loss": 6.0571,
|
|
"mean_token_accuracy": 0.17093972414731978,
|
|
"num_tokens": 7322003.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.216898107528687,
|
|
"epoch": 3.4444444444444446,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00043014025714456305,
|
|
"loss": 6.0057,
|
|
"mean_token_accuracy": 0.16487190499901772,
|
|
"num_tokens": 7333932.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.269065189361572,
|
|
"epoch": 3.4502923976608186,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004298004252829787,
|
|
"loss": 6.0437,
|
|
"mean_token_accuracy": 0.16805607676506043,
|
|
"num_tokens": 7346849.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.2054437637329105,
|
|
"epoch": 3.456140350877193,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000429459921576448,
|
|
"loss": 6.0929,
|
|
"mean_token_accuracy": 0.16196269243955613,
|
|
"num_tokens": 7358940.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.141955709457397,
|
|
"epoch": 3.461988304093567,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042911874750278133,
|
|
"loss": 5.9353,
|
|
"mean_token_accuracy": 0.18545606434345246,
|
|
"num_tokens": 7371883.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.171265268325806,
|
|
"epoch": 3.4678362573099415,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00042877690454269884,
|
|
"loss": 5.9664,
|
|
"mean_token_accuracy": 0.17044519484043122,
|
|
"num_tokens": 7384203.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.18736047744751,
|
|
"epoch": 3.473684210526316,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00042843439417982344,
|
|
"loss": 5.994,
|
|
"mean_token_accuracy": 0.16554267406463624,
|
|
"num_tokens": 7397851.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.150423145294189,
|
|
"epoch": 3.47953216374269,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004280912179006746,
|
|
"loss": 5.9952,
|
|
"mean_token_accuracy": 0.17289814352989197,
|
|
"num_tokens": 7410684.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.178168869018554,
|
|
"epoch": 3.4853801169590644,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000427747377194662,
|
|
"loss": 6.0022,
|
|
"mean_token_accuracy": 0.17682780921459199,
|
|
"num_tokens": 7423393.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.163204765319824,
|
|
"epoch": 3.4912280701754383,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004274028735540789,
|
|
"loss": 5.951,
|
|
"mean_token_accuracy": 0.17342002987861632,
|
|
"num_tokens": 7436170.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.162869501113891,
|
|
"epoch": 3.497076023391813,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00042705770847409596,
|
|
"loss": 6.0194,
|
|
"mean_token_accuracy": 0.16856492310762405,
|
|
"num_tokens": 7448915.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.195002555847168,
|
|
"epoch": 3.502923976608187,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00042671188345275423,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.16502529233694077,
|
|
"num_tokens": 7461585.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.21580867767334,
|
|
"epoch": 3.5087719298245617,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004263653999909593,
|
|
"loss": 5.9884,
|
|
"mean_token_accuracy": 0.17214400917291642,
|
|
"num_tokens": 7473826.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.5087719298245617,
|
|
"eval_entropy": 6.110265312644936,
|
|
"eval_loss": 6.669482707977295,
|
|
"eval_mean_token_accuracy": 0.14564121225546595,
|
|
"eval_num_tokens": 7473826.0,
|
|
"eval_runtime": 2.2807,
|
|
"eval_samples_per_second": 1186.915,
|
|
"eval_steps_per_second": 148.638,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.168160915374756,
|
|
"epoch": 3.5146198830409356,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000426018259592474,
|
|
"loss": 5.9714,
|
|
"mean_token_accuracy": 0.17308609932661057,
|
|
"num_tokens": 7486271.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.158830356597901,
|
|
"epoch": 3.52046783625731,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004256704637639126,
|
|
"loss": 5.99,
|
|
"mean_token_accuracy": 0.16976851373910903,
|
|
"num_tokens": 7499160.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.174773645401001,
|
|
"epoch": 3.526315789473684,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004253220140147341,
|
|
"loss": 6.0677,
|
|
"mean_token_accuracy": 0.1640997663140297,
|
|
"num_tokens": 7511424.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.226050233840942,
|
|
"epoch": 3.5321637426900585,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004249729118572352,
|
|
"loss": 6.016,
|
|
"mean_token_accuracy": 0.1699744313955307,
|
|
"num_tokens": 7523113.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.267972946166992,
|
|
"epoch": 3.538011695906433,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004246231588065444,
|
|
"loss": 6.0965,
|
|
"mean_token_accuracy": 0.15748149454593657,
|
|
"num_tokens": 7536198.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.100395584106446,
|
|
"epoch": 3.543859649122807,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000424272756380615,
|
|
"loss": 5.917,
|
|
"mean_token_accuracy": 0.17440422251820564,
|
|
"num_tokens": 7548782.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.188682985305786,
|
|
"epoch": 3.5497076023391814,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004239217061002188,
|
|
"loss": 6.0274,
|
|
"mean_token_accuracy": 0.16385763883590698,
|
|
"num_tokens": 7560745.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.2717749118804935,
|
|
"epoch": 3.5555555555555554,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042357000948893894,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.16803848445415498,
|
|
"num_tokens": 7572095.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.16748628616333,
|
|
"epoch": 3.56140350877193,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00042321766807316405,
|
|
"loss": 6.0787,
|
|
"mean_token_accuracy": 0.1694350227713585,
|
|
"num_tokens": 7584462.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.280899858474731,
|
|
"epoch": 3.5672514619883042,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042286468338208113,
|
|
"loss": 6.007,
|
|
"mean_token_accuracy": 0.17000401318073272,
|
|
"num_tokens": 7596910.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.111650514602661,
|
|
"epoch": 3.573099415204678,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004225110569476691,
|
|
"loss": 6.0113,
|
|
"mean_token_accuracy": 0.16991945058107377,
|
|
"num_tokens": 7609298.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.139304828643799,
|
|
"epoch": 3.5789473684210527,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000422156790304692,
|
|
"loss": 6.0016,
|
|
"mean_token_accuracy": 0.17069095075130464,
|
|
"num_tokens": 7622945.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.227212619781494,
|
|
"epoch": 3.5847953216374266,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042180188499069246,
|
|
"loss": 5.935,
|
|
"mean_token_accuracy": 0.1692454621195793,
|
|
"num_tokens": 7635880.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.076358413696289,
|
|
"epoch": 3.590643274853801,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000421446342545985,
|
|
"loss": 5.9835,
|
|
"mean_token_accuracy": 0.169833804666996,
|
|
"num_tokens": 7648079.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.206840181350708,
|
|
"epoch": 3.5964912280701755,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042109016451364933,
|
|
"loss": 6.0421,
|
|
"mean_token_accuracy": 0.16122946441173552,
|
|
"num_tokens": 7660701.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.154277420043945,
|
|
"epoch": 3.60233918128655,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042073335243952354,
|
|
"loss": 5.8997,
|
|
"mean_token_accuracy": 0.1736776426434517,
|
|
"num_tokens": 7672496.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.14108510017395,
|
|
"epoch": 3.608187134502924,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004203759078721979,
|
|
"loss": 6.0705,
|
|
"mean_token_accuracy": 0.1707608222961426,
|
|
"num_tokens": 7684407.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.227617835998535,
|
|
"epoch": 3.6140350877192984,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004200178323630072,
|
|
"loss": 5.9926,
|
|
"mean_token_accuracy": 0.16486037969589235,
|
|
"num_tokens": 7697195.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.159926795959473,
|
|
"epoch": 3.6198830409356724,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041965912746602495,
|
|
"loss": 5.9801,
|
|
"mean_token_accuracy": 0.17005882859230043,
|
|
"num_tokens": 7709698.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.1045825481414795,
|
|
"epoch": 3.625730994152047,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00041929979473805606,
|
|
"loss": 5.9693,
|
|
"mean_token_accuracy": 0.17374220192432405,
|
|
"num_tokens": 7722638.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.214031934738159,
|
|
"epoch": 3.6315789473684212,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041893983573863044,
|
|
"loss": 5.9751,
|
|
"mean_token_accuracy": 0.17192549854516984,
|
|
"num_tokens": 7735513.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.179910898208618,
|
|
"epoch": 3.6374269005847952,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00041857925202999586,
|
|
"loss": 6.1075,
|
|
"mean_token_accuracy": 0.1613726243376732,
|
|
"num_tokens": 7748826.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.03430757522583,
|
|
"epoch": 3.6432748538011697,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004182180451771115,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.17266260087490082,
|
|
"num_tokens": 7761139.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.277319002151489,
|
|
"epoch": 3.6491228070175437,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004178562167476409,
|
|
"loss": 6.064,
|
|
"mean_token_accuracy": 0.1629991427063942,
|
|
"num_tokens": 7772723.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.239894437789917,
|
|
"epoch": 3.654970760233918,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041749376831194565,
|
|
"loss": 6.0519,
|
|
"mean_token_accuracy": 0.17082974016666413,
|
|
"num_tokens": 7784436.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.130610227584839,
|
|
"epoch": 3.6608187134502925,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004171307014430777,
|
|
"loss": 5.9757,
|
|
"mean_token_accuracy": 0.17419460713863372,
|
|
"num_tokens": 7797471.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.134858846664429,
|
|
"epoch": 3.6666666666666665,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00041676701771677343,
|
|
"loss": 5.9935,
|
|
"mean_token_accuracy": 0.16835572570562363,
|
|
"num_tokens": 7809923.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.257731676101685,
|
|
"epoch": 3.672514619883041,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004164027187114463,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.16347835212945938,
|
|
"num_tokens": 7821793.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.244524002075195,
|
|
"epoch": 3.678362573099415,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00041603780600818,
|
|
"loss": 6.0558,
|
|
"mean_token_accuracy": 0.16942404508590697,
|
|
"num_tokens": 7834263.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.115977811813354,
|
|
"epoch": 3.6842105263157894,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004156722811907219,
|
|
"loss": 5.9724,
|
|
"mean_token_accuracy": 0.1744808316230774,
|
|
"num_tokens": 7845533.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.131769132614136,
|
|
"epoch": 3.690058479532164,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041530614584547594,
|
|
"loss": 5.9743,
|
|
"mean_token_accuracy": 0.1665850654244423,
|
|
"num_tokens": 7857290.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.128918600082398,
|
|
"epoch": 3.6959064327485383,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041493940156149577,
|
|
"loss": 6.0232,
|
|
"mean_token_accuracy": 0.16824957877397537,
|
|
"num_tokens": 7870049.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.180130052566528,
|
|
"epoch": 3.7017543859649122,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000414572049930478,
|
|
"loss": 6.0084,
|
|
"mean_token_accuracy": 0.17029111087322235,
|
|
"num_tokens": 7882154.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.132463121414185,
|
|
"epoch": 3.7076023391812867,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00041420409254675465,
|
|
"loss": 6.0728,
|
|
"mean_token_accuracy": 0.1656474679708481,
|
|
"num_tokens": 7895173.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.204341125488281,
|
|
"epoch": 3.7134502923976607,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004138355310072876,
|
|
"loss": 6.0228,
|
|
"mean_token_accuracy": 0.16480461210012437,
|
|
"num_tokens": 7908032.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.26247353553772,
|
|
"epoch": 3.719298245614035,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041346636691166015,
|
|
"loss": 6.0733,
|
|
"mean_token_accuracy": 0.15896032750606537,
|
|
"num_tokens": 7921551.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.234937381744385,
|
|
"epoch": 3.7251461988304095,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00041309660186207093,
|
|
"loss": 6.0624,
|
|
"mean_token_accuracy": 0.16706914007663726,
|
|
"num_tokens": 7933986.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.208376455307007,
|
|
"epoch": 3.7309941520467835,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041272623746332687,
|
|
"loss": 6.021,
|
|
"mean_token_accuracy": 0.1685838967561722,
|
|
"num_tokens": 7946072.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.198312664031983,
|
|
"epoch": 3.736842105263158,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041235527532283603,
|
|
"loss": 6.0702,
|
|
"mean_token_accuracy": 0.16089875549077987,
|
|
"num_tokens": 7959779.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.221551370620728,
|
|
"epoch": 3.742690058479532,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004119837170506005,
|
|
"loss": 5.9995,
|
|
"mean_token_accuracy": 0.16619008779525757,
|
|
"num_tokens": 7972868.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.172733783721924,
|
|
"epoch": 3.7485380116959064,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00041161156425921004,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.17230560332536698,
|
|
"num_tokens": 7986160.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.086882019042969,
|
|
"epoch": 3.754385964912281,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041123881856383436,
|
|
"loss": 5.9981,
|
|
"mean_token_accuracy": 0.1715784415602684,
|
|
"num_tokens": 7999286.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.217818021774292,
|
|
"epoch": 3.760233918128655,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004108654815822165,
|
|
"loss": 6.0075,
|
|
"mean_token_accuracy": 0.16945263892412185,
|
|
"num_tokens": 8012745.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.156993055343628,
|
|
"epoch": 3.7660818713450293,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004104915549346658,
|
|
"loss": 6.0431,
|
|
"mean_token_accuracy": 0.1693377137184143,
|
|
"num_tokens": 8025865.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.175265645980835,
|
|
"epoch": 3.7719298245614032,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00041011704024405065,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.16357186883687974,
|
|
"num_tokens": 8038345.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.1215215682983395,
|
|
"epoch": 3.7777777777777777,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040974193913579174,
|
|
"loss": 5.8609,
|
|
"mean_token_accuracy": 0.17776030898094178,
|
|
"num_tokens": 8049861.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.072991895675659,
|
|
"epoch": 3.783625730994152,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004093662532378547,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.1761360615491867,
|
|
"num_tokens": 8061762.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.2837165832519535,
|
|
"epoch": 3.7894736842105265,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004089899841807433,
|
|
"loss": 6.0521,
|
|
"mean_token_accuracy": 0.1629977360367775,
|
|
"num_tokens": 8074074.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.106632947921753,
|
|
"epoch": 3.7953216374269005,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040861313359749217,
|
|
"loss": 5.9817,
|
|
"mean_token_accuracy": 0.16985149085521697,
|
|
"num_tokens": 8085664.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.165804529190064,
|
|
"epoch": 3.801169590643275,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040823570312366005,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.16323314905166625,
|
|
"num_tokens": 8097768.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.173748350143432,
|
|
"epoch": 3.807017543859649,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000407857694397322,
|
|
"loss": 5.9755,
|
|
"mean_token_accuracy": 0.17611339837312698,
|
|
"num_tokens": 8109240.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.164474678039551,
|
|
"epoch": 3.8128654970760234,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004074791090590632,
|
|
"loss": 6.0412,
|
|
"mean_token_accuracy": 0.1648976132273674,
|
|
"num_tokens": 8121658.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.179024171829224,
|
|
"epoch": 3.818713450292398,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004070999487519711,
|
|
"loss": 6.0579,
|
|
"mean_token_accuracy": 0.16155336946249008,
|
|
"num_tokens": 8133353.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.094624471664429,
|
|
"epoch": 3.824561403508772,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00040672021512162864,
|
|
"loss": 6.05,
|
|
"mean_token_accuracy": 0.1641142785549164,
|
|
"num_tokens": 8146589.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.264404773712158,
|
|
"epoch": 3.8304093567251463,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000406339909816107,
|
|
"loss": 6.0213,
|
|
"mean_token_accuracy": 0.16167487800121308,
|
|
"num_tokens": 8158676.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.129144954681396,
|
|
"epoch": 3.8362573099415203,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004059590344859584,
|
|
"loss": 5.893,
|
|
"mean_token_accuracy": 0.18017241209745408,
|
|
"num_tokens": 8170616.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.99580717086792,
|
|
"epoch": 3.8421052631578947,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004055775907842092,
|
|
"loss": 6.0134,
|
|
"mean_token_accuracy": 0.17614096999168397,
|
|
"num_tokens": 8183045.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.239017581939697,
|
|
"epoch": 3.847953216374269,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004051955803663523,
|
|
"loss": 6.0581,
|
|
"mean_token_accuracy": 0.17262273877859116,
|
|
"num_tokens": 8195215.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.170020008087159,
|
|
"epoch": 3.853801169590643,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004048130048903404,
|
|
"loss": 6.038,
|
|
"mean_token_accuracy": 0.16725364923477173,
|
|
"num_tokens": 8207963.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.154378414154053,
|
|
"epoch": 3.8596491228070176,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004044298660165786,
|
|
"loss": 5.9784,
|
|
"mean_token_accuracy": 0.17133220434188842,
|
|
"num_tokens": 8220908.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.240460634231567,
|
|
"epoch": 3.8654970760233915,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00040404616540791683,
|
|
"loss": 5.9752,
|
|
"mean_token_accuracy": 0.17119812965393066,
|
|
"num_tokens": 8233600.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.1052779197692875,
|
|
"epoch": 3.871345029239766,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004036619047296435,
|
|
"loss": 5.9502,
|
|
"mean_token_accuracy": 0.18522704094648362,
|
|
"num_tokens": 8246209.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.1566300868988035,
|
|
"epoch": 3.8771929824561404,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004032770856494772,
|
|
"loss": 6.0201,
|
|
"mean_token_accuracy": 0.1704757332801819,
|
|
"num_tokens": 8257655.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.18036150932312,
|
|
"epoch": 3.883040935672515,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00040289170983756075,
|
|
"loss": 6.0087,
|
|
"mean_token_accuracy": 0.17002097219228746,
|
|
"num_tokens": 8270701.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.148637771606445,
|
|
"epoch": 3.888888888888889,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004025057789664524,
|
|
"loss": 6.0335,
|
|
"mean_token_accuracy": 0.16643901467323302,
|
|
"num_tokens": 8283270.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.158613920211792,
|
|
"epoch": 3.8947368421052633,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004021192947111201,
|
|
"loss": 6.0065,
|
|
"mean_token_accuracy": 0.16925504952669143,
|
|
"num_tokens": 8294842.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.1078980445861815,
|
|
"epoch": 3.9005847953216373,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040173225874893305,
|
|
"loss": 5.8036,
|
|
"mean_token_accuracy": 0.18257610499858856,
|
|
"num_tokens": 8307527.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.118655967712402,
|
|
"epoch": 3.9064327485380117,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00040134467275965515,
|
|
"loss": 5.982,
|
|
"mean_token_accuracy": 0.17393046170473098,
|
|
"num_tokens": 8319220.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.233961963653565,
|
|
"epoch": 3.912280701754386,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004009565384254374,
|
|
"loss": 6.0261,
|
|
"mean_token_accuracy": 0.16972270756959915,
|
|
"num_tokens": 8331577.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.120598983764649,
|
|
"epoch": 3.91812865497076,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040056785743081065,
|
|
"loss": 6.0985,
|
|
"mean_token_accuracy": 0.16492811292409898,
|
|
"num_tokens": 8343327.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.216963291168213,
|
|
"epoch": 3.9239766081871346,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004001786314626783,
|
|
"loss": 6.0776,
|
|
"mean_token_accuracy": 0.16241157054901123,
|
|
"num_tokens": 8356063.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.181361246109009,
|
|
"epoch": 3.9298245614035086,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003997888622103088,
|
|
"loss": 6.0741,
|
|
"mean_token_accuracy": 0.15976226478815078,
|
|
"num_tokens": 8368547.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.10531849861145,
|
|
"epoch": 3.935672514619883,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003993985513653289,
|
|
"loss": 5.9001,
|
|
"mean_token_accuracy": 0.1757778897881508,
|
|
"num_tokens": 8381172.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.183521318435669,
|
|
"epoch": 3.9415204678362574,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003990077006217155,
|
|
"loss": 6.0812,
|
|
"mean_token_accuracy": 0.16544191539287567,
|
|
"num_tokens": 8392560.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.172083234786987,
|
|
"epoch": 3.9473684210526314,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000398616311675789,
|
|
"loss": 5.9894,
|
|
"mean_token_accuracy": 0.16490666568279266,
|
|
"num_tokens": 8405547.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.165727376937866,
|
|
"epoch": 3.953216374269006,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039822438622620523,
|
|
"loss": 6.0125,
|
|
"mean_token_accuracy": 0.1696140930056572,
|
|
"num_tokens": 8418749.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.113630151748657,
|
|
"epoch": 3.95906432748538,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039783192597394896,
|
|
"loss": 6.037,
|
|
"mean_token_accuracy": 0.17148204445838927,
|
|
"num_tokens": 8430979.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.1933770179748535,
|
|
"epoch": 3.9649122807017543,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039743893262232584,
|
|
"loss": 5.9959,
|
|
"mean_token_accuracy": 0.16409866660833358,
|
|
"num_tokens": 8443256.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.145831298828125,
|
|
"epoch": 3.9707602339181287,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003970454078769551,
|
|
"loss": 5.9922,
|
|
"mean_token_accuracy": 0.16444988250732423,
|
|
"num_tokens": 8454933.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.192234945297241,
|
|
"epoch": 3.976608187134503,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003966513534457623,
|
|
"loss": 6.0383,
|
|
"mean_token_accuracy": 0.1717495933175087,
|
|
"num_tokens": 8467517.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.0661486148834225,
|
|
"epoch": 3.982456140350877,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003962567710389721,
|
|
"loss": 6.0111,
|
|
"mean_token_accuracy": 0.1670329064130783,
|
|
"num_tokens": 8481084.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.217316913604736,
|
|
"epoch": 3.9883040935672516,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039586166236910033,
|
|
"loss": 6.0117,
|
|
"mean_token_accuracy": 0.1652706429362297,
|
|
"num_tokens": 8493236.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.004075002670288,
|
|
"epoch": 3.9941520467836256,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003954660291509469,
|
|
"loss": 5.8034,
|
|
"mean_token_accuracy": 0.1852240741252899,
|
|
"num_tokens": 8506152.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.148331594467163,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039506987310158854,
|
|
"loss": 5.9765,
|
|
"mean_token_accuracy": 0.16524131447076798,
|
|
"num_tokens": 8517936.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.208982992172241,
|
|
"epoch": 4.005847953216374,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039467319594037076,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.1820027008652687,
|
|
"num_tokens": 8530099.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.058341026306152,
|
|
"epoch": 4.011695906432749,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003942759993889009,
|
|
"loss": 5.6727,
|
|
"mean_token_accuracy": 0.1847613424062729,
|
|
"num_tokens": 8542145.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.1071876049041744,
|
|
"epoch": 4.017543859649122,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039387828517104044,
|
|
"loss": 5.713,
|
|
"mean_token_accuracy": 0.1795678973197937,
|
|
"num_tokens": 8553758.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.073211717605591,
|
|
"epoch": 4.023391812865497,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00039348005501289775,
|
|
"loss": 5.6637,
|
|
"mean_token_accuracy": 0.1778433293104172,
|
|
"num_tokens": 8566215.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.1634174346923825,
|
|
"epoch": 4.029239766081871,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039308131064282016,
|
|
"loss": 5.6641,
|
|
"mean_token_accuracy": 0.1766158401966095,
|
|
"num_tokens": 8578256.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.055958652496338,
|
|
"epoch": 4.035087719298246,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003926820537913868,
|
|
"loss": 5.6126,
|
|
"mean_token_accuracy": 0.18623143434524536,
|
|
"num_tokens": 8591235.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.961992931365967,
|
|
"epoch": 4.04093567251462,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003922822861914012,
|
|
"loss": 5.5232,
|
|
"mean_token_accuracy": 0.18734114915132521,
|
|
"num_tokens": 8603230.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.08734655380249,
|
|
"epoch": 4.046783625730994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003918820095778832,
|
|
"loss": 5.6244,
|
|
"mean_token_accuracy": 0.1852467656135559,
|
|
"num_tokens": 8615114.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.97623496055603,
|
|
"epoch": 4.052631578947368,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00039148122568806215,
|
|
"loss": 5.5808,
|
|
"mean_token_accuracy": 0.19530852288007736,
|
|
"num_tokens": 8627212.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.029066705703736,
|
|
"epoch": 4.058479532163743,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003910799362613688,
|
|
"loss": 5.6803,
|
|
"mean_token_accuracy": 0.17855599969625474,
|
|
"num_tokens": 8639887.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.140685224533081,
|
|
"epoch": 4.064327485380117,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00039067814303942806,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.18161029368638992,
|
|
"num_tokens": 8653097.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.050816392898559,
|
|
"epoch": 4.0701754385964914,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00039027584776605136,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.1780172884464264,
|
|
"num_tokens": 8666830.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.970085477828979,
|
|
"epoch": 4.076023391812866,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00038987305218722904,
|
|
"loss": 5.5956,
|
|
"mean_token_accuracy": 0.1876891404390335,
|
|
"num_tokens": 8679725.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.093829774856568,
|
|
"epoch": 4.081871345029239,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003894697580511226,
|
|
"loss": 5.7409,
|
|
"mean_token_accuracy": 0.1759042337536812,
|
|
"num_tokens": 8691280.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.0667116165161135,
|
|
"epoch": 4.087719298245614,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003890659671080579,
|
|
"loss": 5.7308,
|
|
"mean_token_accuracy": 0.17748655378818512,
|
|
"num_tokens": 8702555.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.158699989318848,
|
|
"epoch": 4.093567251461988,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003886616811105162,
|
|
"loss": 5.7297,
|
|
"mean_token_accuracy": 0.17546985894441605,
|
|
"num_tokens": 8714693.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 4.093567251461988,
|
|
"eval_entropy": 6.059477153429239,
|
|
"eval_loss": 6.620388031005859,
|
|
"eval_mean_token_accuracy": 0.14776903937040528,
|
|
"eval_num_tokens": 8714693.0,
|
|
"eval_runtime": 2.2886,
|
|
"eval_samples_per_second": 1182.803,
|
|
"eval_steps_per_second": 148.124,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.085285091400147,
|
|
"epoch": 4.099415204678363,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038825690181312793,
|
|
"loss": 5.6244,
|
|
"mean_token_accuracy": 0.17662072628736497,
|
|
"num_tokens": 8726967.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.0292808532714846,
|
|
"epoch": 4.105263157894737,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038785163097266424,
|
|
"loss": 5.5735,
|
|
"mean_token_accuracy": 0.19106259047985077,
|
|
"num_tokens": 8739617.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.087874221801758,
|
|
"epoch": 4.111111111111111,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003874458703480296,
|
|
"loss": 5.7334,
|
|
"mean_token_accuracy": 0.17352990806102753,
|
|
"num_tokens": 8751684.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.0835143566131595,
|
|
"epoch": 4.116959064327485,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003870396217002542,
|
|
"loss": 5.6771,
|
|
"mean_token_accuracy": 0.17921384871006013,
|
|
"num_tokens": 8763761.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.939363956451416,
|
|
"epoch": 4.12280701754386,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003866328867924864,
|
|
"loss": 5.5565,
|
|
"mean_token_accuracy": 0.19428135305643082,
|
|
"num_tokens": 8775274.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.044627571105957,
|
|
"epoch": 4.128654970760234,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003862256673899848,
|
|
"loss": 5.6757,
|
|
"mean_token_accuracy": 0.18519430607557297,
|
|
"num_tokens": 8787683.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.01921911239624,
|
|
"epoch": 4.1345029239766085,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003858179652601108,
|
|
"loss": 5.5418,
|
|
"mean_token_accuracy": 0.19873106628656387,
|
|
"num_tokens": 8800119.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.131620979309082,
|
|
"epoch": 4.140350877192983,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038540978217232086,
|
|
"loss": 5.6961,
|
|
"mean_token_accuracy": 0.18028391599655152,
|
|
"num_tokens": 8812017.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.018851900100708,
|
|
"epoch": 4.146198830409356,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038500111989815894,
|
|
"loss": 5.6414,
|
|
"mean_token_accuracy": 0.18363899886608123,
|
|
"num_tokens": 8823895.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.0797628402709964,
|
|
"epoch": 4.152046783625731,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038459198021124853,
|
|
"loss": 5.661,
|
|
"mean_token_accuracy": 0.18023838698863984,
|
|
"num_tokens": 8835750.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.020551967620849,
|
|
"epoch": 4.157894736842105,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003841823648872852,
|
|
"loss": 5.6764,
|
|
"mean_token_accuracy": 0.18082891702651976,
|
|
"num_tokens": 8847704.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.131635522842407,
|
|
"epoch": 4.16374269005848,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003837722757040289,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.17385272681713104,
|
|
"num_tokens": 8860092.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.00726284980774,
|
|
"epoch": 4.169590643274854,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003833617144412961,
|
|
"loss": 5.596,
|
|
"mean_token_accuracy": 0.18837205767631532,
|
|
"num_tokens": 8873090.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.045232725143433,
|
|
"epoch": 4.175438596491228,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003829506828809521,
|
|
"loss": 5.7513,
|
|
"mean_token_accuracy": 0.1748623937368393,
|
|
"num_tokens": 8885398.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.060342025756836,
|
|
"epoch": 4.181286549707602,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003825391828069032,
|
|
"loss": 5.6081,
|
|
"mean_token_accuracy": 0.19171490520238876,
|
|
"num_tokens": 8897438.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.055225515365601,
|
|
"epoch": 4.187134502923977,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003821272160050894,
|
|
"loss": 5.6972,
|
|
"mean_token_accuracy": 0.1824840009212494,
|
|
"num_tokens": 8909666.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.06932110786438,
|
|
"epoch": 4.192982456140351,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003817147842634762,
|
|
"loss": 5.5719,
|
|
"mean_token_accuracy": 0.18798228055238725,
|
|
"num_tokens": 8921403.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.074003791809082,
|
|
"epoch": 4.1988304093567255,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003813018893720468,
|
|
"loss": 5.7131,
|
|
"mean_token_accuracy": 0.17799114137887956,
|
|
"num_tokens": 8934623.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.081497097015381,
|
|
"epoch": 4.204678362573099,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00038088853312279473,
|
|
"loss": 5.6861,
|
|
"mean_token_accuracy": 0.1815410941839218,
|
|
"num_tokens": 8947626.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.9984534740447994,
|
|
"epoch": 4.2105263157894735,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003804747173097157,
|
|
"loss": 5.6305,
|
|
"mean_token_accuracy": 0.19211477041244507,
|
|
"num_tokens": 8959699.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.989898729324341,
|
|
"epoch": 4.216374269005848,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038006044372880006,
|
|
"loss": 5.6539,
|
|
"mean_token_accuracy": 0.18349657207727432,
|
|
"num_tokens": 8971650.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.986303424835205,
|
|
"epoch": 4.222222222222222,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037964571417802473,
|
|
"loss": 5.6467,
|
|
"mean_token_accuracy": 0.1821511596441269,
|
|
"num_tokens": 8984266.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.107998943328857,
|
|
"epoch": 4.228070175438597,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00037923053045734595,
|
|
"loss": 5.7822,
|
|
"mean_token_accuracy": 0.17441510707139968,
|
|
"num_tokens": 8996336.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.152076530456543,
|
|
"epoch": 4.23391812865497,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037881489436869054,
|
|
"loss": 5.8137,
|
|
"mean_token_accuracy": 0.17387254387140275,
|
|
"num_tokens": 9008486.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.005298805236817,
|
|
"epoch": 4.239766081871345,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003783988077159491,
|
|
"loss": 5.6748,
|
|
"mean_token_accuracy": 0.17437252700328826,
|
|
"num_tokens": 9020130.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.087873935699463,
|
|
"epoch": 4.245614035087719,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00037798227230496754,
|
|
"loss": 5.7121,
|
|
"mean_token_accuracy": 0.17935457825660706,
|
|
"num_tokens": 9032008.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.098003339767456,
|
|
"epoch": 4.251461988304094,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037756528994353945,
|
|
"loss": 5.755,
|
|
"mean_token_accuracy": 0.17370388507843018,
|
|
"num_tokens": 9045337.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.018971300125122,
|
|
"epoch": 4.257309941520468,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 5.6304,
|
|
"mean_token_accuracy": 0.18800177723169326,
|
|
"num_tokens": 9058033.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.044499349594116,
|
|
"epoch": 4.2631578947368425,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00037672999161020873,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.1770397886633873,
|
|
"num_tokens": 9070120.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.020751285552978,
|
|
"epoch": 4.269005847953216,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00037631167926356087,
|
|
"loss": 5.6901,
|
|
"mean_token_accuracy": 0.18077751249074936,
|
|
"num_tokens": 9082947.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.074623155593872,
|
|
"epoch": 4.2748538011695905,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003758929272169599,
|
|
"loss": 5.7544,
|
|
"mean_token_accuracy": 0.1766696408390999,
|
|
"num_tokens": 9096196.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.127846097946167,
|
|
"epoch": 4.280701754385965,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037547373728781996,
|
|
"loss": 5.7241,
|
|
"mean_token_accuracy": 0.17622800767421723,
|
|
"num_tokens": 9109403.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.094042015075684,
|
|
"epoch": 4.286549707602339,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003750541112954552,
|
|
"loss": 5.7758,
|
|
"mean_token_accuracy": 0.17313547879457475,
|
|
"num_tokens": 9122062.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.971199226379395,
|
|
"epoch": 4.292397660818714,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003746340510610727,
|
|
"loss": 5.5549,
|
|
"mean_token_accuracy": 0.1898258149623871,
|
|
"num_tokens": 9133790.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.08124361038208,
|
|
"epoch": 4.298245614035087,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00037421355840776396,
|
|
"loss": 5.6981,
|
|
"mean_token_accuracy": 0.1801225945353508,
|
|
"num_tokens": 9145651.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.0237737655639645,
|
|
"epoch": 4.304093567251462,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003737926351604971,
|
|
"loss": 5.7382,
|
|
"mean_token_accuracy": 0.17153676450252534,
|
|
"num_tokens": 9157899.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.968272304534912,
|
|
"epoch": 4.309941520467836,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00037337128314610926,
|
|
"loss": 5.5406,
|
|
"mean_token_accuracy": 0.18944045752286912,
|
|
"num_tokens": 9169413.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.010223245620727,
|
|
"epoch": 4.315789473684211,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003729495041932984,
|
|
"loss": 5.6903,
|
|
"mean_token_accuracy": 0.18027177453041077,
|
|
"num_tokens": 9181745.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.018995141983032,
|
|
"epoch": 4.321637426900585,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003725273001326155,
|
|
"loss": 5.6315,
|
|
"mean_token_accuracy": 0.18165598213672637,
|
|
"num_tokens": 9194419.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.056298828125,
|
|
"epoch": 4.3274853801169595,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003721046727964565,
|
|
"loss": 5.7336,
|
|
"mean_token_accuracy": 0.18085582703351974,
|
|
"num_tokens": 9206102.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.9240764617919925,
|
|
"epoch": 4.333333333333333,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037168162401905416,
|
|
"loss": 5.6536,
|
|
"mean_token_accuracy": 0.1859877422451973,
|
|
"num_tokens": 9218795.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.047742557525635,
|
|
"epoch": 4.3391812865497075,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00037125815563647063,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.18040397763252258,
|
|
"num_tokens": 9231640.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.10801510810852,
|
|
"epoch": 4.345029239766082,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00037083426948658903,
|
|
"loss": 5.6666,
|
|
"mean_token_accuracy": 0.1811925694346428,
|
|
"num_tokens": 9243639.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.094470977783203,
|
|
"epoch": 4.350877192982456,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037040996740910566,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.17913928180932998,
|
|
"num_tokens": 9256445.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.986751461029053,
|
|
"epoch": 4.356725146198831,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000369985251245522,
|
|
"loss": 5.6379,
|
|
"mean_token_accuracy": 0.18539317697286606,
|
|
"num_tokens": 9268407.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.129997825622558,
|
|
"epoch": 4.362573099415204,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00036956012283913655,
|
|
"loss": 5.8049,
|
|
"mean_token_accuracy": 0.17652171105146408,
|
|
"num_tokens": 9281956.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.1099484920501705,
|
|
"epoch": 4.368421052631579,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00036913458403503715,
|
|
"loss": 5.765,
|
|
"mean_token_accuracy": 0.17523328512907027,
|
|
"num_tokens": 9294616.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.025202322006225,
|
|
"epoch": 4.374269005847953,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003687086366800927,
|
|
"loss": 5.7227,
|
|
"mean_token_accuracy": 0.1798295259475708,
|
|
"num_tokens": 9306903.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.0017040252685545,
|
|
"epoch": 4.380116959064328,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003682822826229453,
|
|
"loss": 5.6572,
|
|
"mean_token_accuracy": 0.18211227506399155,
|
|
"num_tokens": 9319361.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.0340070724487305,
|
|
"epoch": 4.385964912280702,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00036785552371400206,
|
|
"loss": 5.6694,
|
|
"mean_token_accuracy": 0.17709777504205704,
|
|
"num_tokens": 9331396.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.112665700912475,
|
|
"epoch": 4.391812865497076,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003674283618054273,
|
|
"loss": 5.8333,
|
|
"mean_token_accuracy": 0.1700577914714813,
|
|
"num_tokens": 9343877.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.1644754886627195,
|
|
"epoch": 4.39766081871345,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003670007987511344,
|
|
"loss": 5.8422,
|
|
"mean_token_accuracy": 0.1769800066947937,
|
|
"num_tokens": 9356625.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.987030124664306,
|
|
"epoch": 4.4035087719298245,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003665728364067776,
|
|
"loss": 5.6526,
|
|
"mean_token_accuracy": 0.1835209310054779,
|
|
"num_tokens": 9369526.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.975714159011841,
|
|
"epoch": 4.409356725146199,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036614447662974413,
|
|
"loss": 5.7669,
|
|
"mean_token_accuracy": 0.1720859795808792,
|
|
"num_tokens": 9381013.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.1493980884552,
|
|
"epoch": 4.415204678362573,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003657157212791462,
|
|
"loss": 5.7356,
|
|
"mean_token_accuracy": 0.1803906038403511,
|
|
"num_tokens": 9394651.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.072501134872437,
|
|
"epoch": 4.421052631578947,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003652865722158128,
|
|
"loss": 5.7714,
|
|
"mean_token_accuracy": 0.17131675332784652,
|
|
"num_tokens": 9406181.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.02750334739685,
|
|
"epoch": 4.426900584795321,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00036485703130228147,
|
|
"loss": 5.7309,
|
|
"mean_token_accuracy": 0.184201155602932,
|
|
"num_tokens": 9419634.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.0360194683074955,
|
|
"epoch": 4.432748538011696,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00036442710040279086,
|
|
"loss": 5.7091,
|
|
"mean_token_accuracy": 0.17466941028833388,
|
|
"num_tokens": 9432513.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.147225475311279,
|
|
"epoch": 4.43859649122807,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003639967813832717,
|
|
"loss": 5.766,
|
|
"mean_token_accuracy": 0.17544693052768706,
|
|
"num_tokens": 9444600.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.934966802597046,
|
|
"epoch": 4.444444444444445,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003635660761113394,
|
|
"loss": 5.5927,
|
|
"mean_token_accuracy": 0.20057351440191268,
|
|
"num_tokens": 9457232.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.990604400634766,
|
|
"epoch": 4.450292397660819,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003631349864562857,
|
|
"loss": 5.7351,
|
|
"mean_token_accuracy": 0.17782551050186157,
|
|
"num_tokens": 9469792.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.1070178031921385,
|
|
"epoch": 4.456140350877193,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003627035142890708,
|
|
"loss": 5.6782,
|
|
"mean_token_accuracy": 0.18631869107484816,
|
|
"num_tokens": 9481844.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.06354341506958,
|
|
"epoch": 4.461988304093567,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003622716614823147,
|
|
"loss": 5.7733,
|
|
"mean_token_accuracy": 0.17527737617492675,
|
|
"num_tokens": 9495002.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.031888437271118,
|
|
"epoch": 4.4678362573099415,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003618394299102896,
|
|
"loss": 5.7132,
|
|
"mean_token_accuracy": 0.1771496430039406,
|
|
"num_tokens": 9506887.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.081401205062866,
|
|
"epoch": 4.473684210526316,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00036140682144891143,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.17240021228790284,
|
|
"num_tokens": 9518568.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.067675971984864,
|
|
"epoch": 4.47953216374269,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00036097383797573205,
|
|
"loss": 5.7489,
|
|
"mean_token_accuracy": 0.17744352668523788,
|
|
"num_tokens": 9531645.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.062890481948853,
|
|
"epoch": 4.485380116959064,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00036054048136993075,
|
|
"loss": 5.6717,
|
|
"mean_token_accuracy": 0.1851318210363388,
|
|
"num_tokens": 9543794.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.053694152832032,
|
|
"epoch": 4.491228070175438,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00036010675351230616,
|
|
"loss": 5.7694,
|
|
"mean_token_accuracy": 0.18543318063020706,
|
|
"num_tokens": 9554860.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.064020776748658,
|
|
"epoch": 4.497076023391813,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003596726562852684,
|
|
"loss": 5.7093,
|
|
"mean_token_accuracy": 0.18126840591430665,
|
|
"num_tokens": 9567832.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.9396082878112795,
|
|
"epoch": 4.502923976608187,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035923819157283074,
|
|
"loss": 5.5907,
|
|
"mean_token_accuracy": 0.18436191082000733,
|
|
"num_tokens": 9581270.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.049287176132202,
|
|
"epoch": 4.508771929824562,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00035880336126060074,
|
|
"loss": 5.7072,
|
|
"mean_token_accuracy": 0.17770397961139678,
|
|
"num_tokens": 9593894.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.071543788909912,
|
|
"epoch": 4.514619883040936,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00035836816723577354,
|
|
"loss": 5.7273,
|
|
"mean_token_accuracy": 0.177946075797081,
|
|
"num_tokens": 9605723.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.977633762359619,
|
|
"epoch": 4.52046783625731,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003579326113871222,
|
|
"loss": 5.6752,
|
|
"mean_token_accuracy": 0.18044619113206864,
|
|
"num_tokens": 9617515.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.06384687423706,
|
|
"epoch": 4.526315789473684,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003574966956049904,
|
|
"loss": 5.7071,
|
|
"mean_token_accuracy": 0.1821513816714287,
|
|
"num_tokens": 9630112.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.00923638343811,
|
|
"epoch": 4.5321637426900585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035706042178128397,
|
|
"loss": 5.7477,
|
|
"mean_token_accuracy": 0.17877199053764342,
|
|
"num_tokens": 9642162.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.066380023956299,
|
|
"epoch": 4.538011695906433,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003566237918094625,
|
|
"loss": 5.685,
|
|
"mean_token_accuracy": 0.18580681830644608,
|
|
"num_tokens": 9654769.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.04279088973999,
|
|
"epoch": 4.543859649122807,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003561868075845313,
|
|
"loss": 5.7415,
|
|
"mean_token_accuracy": 0.17682476639747619,
|
|
"num_tokens": 9667385.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.987998628616333,
|
|
"epoch": 4.549707602339181,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00035574947100303327,
|
|
"loss": 5.7193,
|
|
"mean_token_accuracy": 0.17582241743803023,
|
|
"num_tokens": 9679867.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.0535664558410645,
|
|
"epoch": 4.555555555555555,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00035531178396304083,
|
|
"loss": 5.7301,
|
|
"mean_token_accuracy": 0.1778804838657379,
|
|
"num_tokens": 9693813.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.0113687992095945,
|
|
"epoch": 4.56140350877193,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00035487374836414676,
|
|
"loss": 5.7822,
|
|
"mean_token_accuracy": 0.17001428008079528,
|
|
"num_tokens": 9706800.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.104873466491699,
|
|
"epoch": 4.567251461988304,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035443536610745707,
|
|
"loss": 5.7163,
|
|
"mean_token_accuracy": 0.1862686812877655,
|
|
"num_tokens": 9719466.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.0243443012237545,
|
|
"epoch": 4.573099415204679,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00035399663909558226,
|
|
"loss": 5.76,
|
|
"mean_token_accuracy": 0.17816496938467025,
|
|
"num_tokens": 9731575.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.0728837013244625,
|
|
"epoch": 4.578947368421053,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035355756923262897,
|
|
"loss": 5.7602,
|
|
"mean_token_accuracy": 0.1771169677376747,
|
|
"num_tokens": 9744399.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.074622535705567,
|
|
"epoch": 4.584795321637427,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003531181584241919,
|
|
"loss": 5.6872,
|
|
"mean_token_accuracy": 0.1787990599870682,
|
|
"num_tokens": 9756329.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.054456949234009,
|
|
"epoch": 4.590643274853801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003526784085773455,
|
|
"loss": 5.7732,
|
|
"mean_token_accuracy": 0.17451857030391693,
|
|
"num_tokens": 9768723.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.083310270309449,
|
|
"epoch": 4.5964912280701755,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003522383216006356,
|
|
"loss": 5.7952,
|
|
"mean_token_accuracy": 0.17585009783506395,
|
|
"num_tokens": 9781232.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.000090742111206,
|
|
"epoch": 4.60233918128655,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003517978994040713,
|
|
"loss": 5.6429,
|
|
"mean_token_accuracy": 0.1819620206952095,
|
|
"num_tokens": 9794209.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.004682970046997,
|
|
"epoch": 4.6081871345029235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00035135714389911637,
|
|
"loss": 5.6978,
|
|
"mean_token_accuracy": 0.18210337311029434,
|
|
"num_tokens": 9806128.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.060727071762085,
|
|
"epoch": 4.614035087719298,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003509160569986814,
|
|
"loss": 5.7665,
|
|
"mean_token_accuracy": 0.17622111588716508,
|
|
"num_tokens": 9818929.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.998124027252198,
|
|
"epoch": 4.619883040935672,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00035047464061711527,
|
|
"loss": 5.7115,
|
|
"mean_token_accuracy": 0.1826970934867859,
|
|
"num_tokens": 9831887.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.020193386077881,
|
|
"epoch": 4.625730994152047,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00035003289667019655,
|
|
"loss": 5.6719,
|
|
"mean_token_accuracy": 0.18340953886508943,
|
|
"num_tokens": 9844180.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.004499626159668,
|
|
"epoch": 4.631578947368421,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003495908270751257,
|
|
"loss": 5.7767,
|
|
"mean_token_accuracy": 0.17524271756410598,
|
|
"num_tokens": 9856897.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.111381196975708,
|
|
"epoch": 4.637426900584796,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003491484337505166,
|
|
"loss": 5.764,
|
|
"mean_token_accuracy": 0.17700181603431703,
|
|
"num_tokens": 9869079.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.119686889648437,
|
|
"epoch": 4.643274853801169,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00034870571861638785,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.17430781126022338,
|
|
"num_tokens": 9882819.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.101339817047119,
|
|
"epoch": 4.649122807017544,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00034826268359415487,
|
|
"loss": 5.7991,
|
|
"mean_token_accuracy": 0.172004035115242,
|
|
"num_tokens": 9895491.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.0609513282775875,
|
|
"epoch": 4.654970760233918,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003478193306066215,
|
|
"loss": 5.7348,
|
|
"mean_token_accuracy": 0.1806287556886673,
|
|
"num_tokens": 9907260.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.996920442581176,
|
|
"epoch": 4.6608187134502925,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00034737566157797147,
|
|
"loss": 5.7677,
|
|
"mean_token_accuracy": 0.1805923879146576,
|
|
"num_tokens": 9919588.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.075291728973388,
|
|
"epoch": 4.666666666666667,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034693167843376016,
|
|
"loss": 5.7144,
|
|
"mean_token_accuracy": 0.18659944087266922,
|
|
"num_tokens": 9930711.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.0045171737670895,
|
|
"epoch": 4.6725146198830405,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034648738310090634,
|
|
"loss": 5.6739,
|
|
"mean_token_accuracy": 0.1824045106768608,
|
|
"num_tokens": 9941870.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.0736244201660154,
|
|
"epoch": 4.678362573099415,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003460427775076835,
|
|
"loss": 5.758,
|
|
"mean_token_accuracy": 0.17529213428497314,
|
|
"num_tokens": 9954004.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.678362573099415,
|
|
"eval_entropy": 6.0285666094417065,
|
|
"eval_loss": 6.584229469299316,
|
|
"eval_mean_token_accuracy": 0.14919248640888913,
|
|
"eval_num_tokens": 9954004.0,
|
|
"eval_runtime": 2.2833,
|
|
"eval_samples_per_second": 1185.557,
|
|
"eval_steps_per_second": 148.468,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.069556713104248,
|
|
"epoch": 4.684210526315789,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034559786358371187,
|
|
"loss": 5.7536,
|
|
"mean_token_accuracy": 0.17481254190206527,
|
|
"num_tokens": 9966729.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.997095346450806,
|
|
"epoch": 4.690058479532164,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034515264325994975,
|
|
"loss": 5.6777,
|
|
"mean_token_accuracy": 0.18319383263587952,
|
|
"num_tokens": 9978378.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.912587976455688,
|
|
"epoch": 4.695906432748538,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00034470711846868525,
|
|
"loss": 5.6596,
|
|
"mean_token_accuracy": 0.18406610041856766,
|
|
"num_tokens": 9990590.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.076769161224365,
|
|
"epoch": 4.701754385964913,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00034426129114352807,
|
|
"loss": 5.6629,
|
|
"mean_token_accuracy": 0.17711972296237946,
|
|
"num_tokens": 10003436.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.03622875213623,
|
|
"epoch": 4.707602339181286,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003438151632194006,
|
|
"loss": 5.7385,
|
|
"mean_token_accuracy": 0.1793992593884468,
|
|
"num_tokens": 10015446.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.04246883392334,
|
|
"epoch": 4.713450292397661,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034336873663252997,
|
|
"loss": 5.7266,
|
|
"mean_token_accuracy": 0.1793872207403183,
|
|
"num_tokens": 10027671.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.043845701217651,
|
|
"epoch": 4.719298245614035,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034292201332043976,
|
|
"loss": 5.7667,
|
|
"mean_token_accuracy": 0.1812325268983841,
|
|
"num_tokens": 10040823.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.955125856399536,
|
|
"epoch": 4.7251461988304095,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003424749952219411,
|
|
"loss": 5.5921,
|
|
"mean_token_accuracy": 0.18806560784578324,
|
|
"num_tokens": 10052642.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.001202869415283,
|
|
"epoch": 4.730994152046784,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00034202768427712456,
|
|
"loss": 5.7412,
|
|
"mean_token_accuracy": 0.1794249251484871,
|
|
"num_tokens": 10065679.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.0669468402862545,
|
|
"epoch": 4.7368421052631575,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003415800824273518,
|
|
"loss": 5.7766,
|
|
"mean_token_accuracy": 0.1780404955148697,
|
|
"num_tokens": 10077757.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.095100975036621,
|
|
"epoch": 4.742690058479532,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003411321916152469,
|
|
"loss": 5.7458,
|
|
"mean_token_accuracy": 0.17781118750572206,
|
|
"num_tokens": 10091027.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 6.018962001800537,
|
|
"epoch": 4.748538011695906,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003406840137846881,
|
|
"loss": 5.7547,
|
|
"mean_token_accuracy": 0.17874414771795272,
|
|
"num_tokens": 10103020.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.030233097076416,
|
|
"epoch": 4.754385964912281,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034023555088079924,
|
|
"loss": 5.6978,
|
|
"mean_token_accuracy": 0.1779080718755722,
|
|
"num_tokens": 10115581.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.039458179473877,
|
|
"epoch": 4.760233918128655,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003397868048499416,
|
|
"loss": 5.7382,
|
|
"mean_token_accuracy": 0.1720757707953453,
|
|
"num_tokens": 10126560.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 6.051576232910156,
|
|
"epoch": 4.76608187134503,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00033933777763970506,
|
|
"loss": 5.7023,
|
|
"mean_token_accuracy": 0.18598945587873458,
|
|
"num_tokens": 10139593.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.038228559494018,
|
|
"epoch": 4.771929824561403,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003388884711989,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.17908376902341844,
|
|
"num_tokens": 10151254.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.948408079147339,
|
|
"epoch": 4.777777777777778,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003384388874775486,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.19270180761814118,
|
|
"num_tokens": 10162860.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.994518852233886,
|
|
"epoch": 4.783625730994152,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033798902842687627,
|
|
"loss": 5.6909,
|
|
"mean_token_accuracy": 0.18395201712846757,
|
|
"num_tokens": 10174781.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.998277091979981,
|
|
"epoch": 4.7894736842105265,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003375388959993037,
|
|
"loss": 5.7097,
|
|
"mean_token_accuracy": 0.1802419066429138,
|
|
"num_tokens": 10187105.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 6.034826850891113,
|
|
"epoch": 4.7953216374269,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003370884921484379,
|
|
"loss": 5.6477,
|
|
"mean_token_accuracy": 0.1862107500433922,
|
|
"num_tokens": 10200774.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.991315603256226,
|
|
"epoch": 4.8011695906432745,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003366378188290638,
|
|
"loss": 5.7251,
|
|
"mean_token_accuracy": 0.17497428208589555,
|
|
"num_tokens": 10213577.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.994521045684815,
|
|
"epoch": 4.807017543859649,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003361868779971361,
|
|
"loss": 5.602,
|
|
"mean_token_accuracy": 0.20564654171466829,
|
|
"num_tokens": 10228205.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 6.018125820159912,
|
|
"epoch": 4.812865497076023,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033573567160977017,
|
|
"loss": 5.8112,
|
|
"mean_token_accuracy": 0.17021705359220504,
|
|
"num_tokens": 10241305.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.981740951538086,
|
|
"epoch": 4.818713450292398,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000335284201625234,
|
|
"loss": 5.6559,
|
|
"mean_token_accuracy": 0.18922288119792938,
|
|
"num_tokens": 10255092.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.92077112197876,
|
|
"epoch": 4.824561403508772,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00033483247000294,
|
|
"loss": 5.6334,
|
|
"mean_token_accuracy": 0.1897211492061615,
|
|
"num_tokens": 10267744.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 6.024300241470337,
|
|
"epoch": 4.830409356725146,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003343804787034357,
|
|
"loss": 5.7136,
|
|
"mean_token_accuracy": 0.18470683097839355,
|
|
"num_tokens": 10280243.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 6.055981540679932,
|
|
"epoch": 4.83625730994152,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003339282296883956,
|
|
"loss": 5.7066,
|
|
"mean_token_accuracy": 0.18061566799879075,
|
|
"num_tokens": 10293881.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.934822988510132,
|
|
"epoch": 4.842105263157895,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000333475724920613,
|
|
"loss": 5.6804,
|
|
"mean_token_accuracy": 0.18286727070808412,
|
|
"num_tokens": 10305585.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.97342963218689,
|
|
"epoch": 4.847953216374269,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003330229663639909,
|
|
"loss": 5.7057,
|
|
"mean_token_accuracy": 0.1788641855120659,
|
|
"num_tokens": 10317406.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 6.025168466567993,
|
|
"epoch": 4.853801169590644,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000332569955983534,
|
|
"loss": 5.7227,
|
|
"mean_token_accuracy": 0.18043506443500518,
|
|
"num_tokens": 10330298.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 6.011347532272339,
|
|
"epoch": 4.859649122807017,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003321166957453398,
|
|
"loss": 5.6403,
|
|
"mean_token_accuracy": 0.19065832793712617,
|
|
"num_tokens": 10342309.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.037061595916748,
|
|
"epoch": 4.8654970760233915,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003316631876165904,
|
|
"loss": 5.7203,
|
|
"mean_token_accuracy": 0.18328420519828797,
|
|
"num_tokens": 10353887.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.9679934024810795,
|
|
"epoch": 4.871345029239766,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033120943356554334,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.18036941736936568,
|
|
"num_tokens": 10367993.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.985843992233276,
|
|
"epoch": 4.87719298245614,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000330755435561524,
|
|
"loss": 5.7456,
|
|
"mean_token_accuracy": 0.18387797474861145,
|
|
"num_tokens": 10380500.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.039897632598877,
|
|
"epoch": 4.883040935672515,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00033030119557491604,
|
|
"loss": 5.7155,
|
|
"mean_token_accuracy": 0.1831536427140236,
|
|
"num_tokens": 10393494.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.001364088058471,
|
|
"epoch": 4.888888888888889,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003298467155771537,
|
|
"loss": 5.6996,
|
|
"mean_token_accuracy": 0.18229294419288636,
|
|
"num_tokens": 10406485.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.991800212860108,
|
|
"epoch": 4.894736842105263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032939199754071274,
|
|
"loss": 5.6996,
|
|
"mean_token_accuracy": 0.18020350039005278,
|
|
"num_tokens": 10418317.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 6.000661325454712,
|
|
"epoch": 4.900584795321637,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000328937043439102,
|
|
"loss": 5.7182,
|
|
"mean_token_accuracy": 0.18871968388557434,
|
|
"num_tokens": 10431672.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.099198484420777,
|
|
"epoch": 4.906432748538012,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032848185524685504,
|
|
"loss": 5.8081,
|
|
"mean_token_accuracy": 0.17294216454029082,
|
|
"num_tokens": 10443747.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 6.000463008880615,
|
|
"epoch": 4.912280701754386,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032802643493952106,
|
|
"loss": 5.6424,
|
|
"mean_token_accuracy": 0.18660420030355454,
|
|
"num_tokens": 10457822.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.006785249710083,
|
|
"epoch": 4.918128654970761,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000327570784493657,
|
|
"loss": 5.7092,
|
|
"mean_token_accuracy": 0.18135093599557878,
|
|
"num_tokens": 10471940.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.944002294540406,
|
|
"epoch": 4.923976608187134,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032711490588681845,
|
|
"loss": 5.6367,
|
|
"mean_token_accuracy": 0.19120272248983383,
|
|
"num_tokens": 10483869.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.925238847732544,
|
|
"epoch": 4.9298245614035086,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003266588010975512,
|
|
"loss": 5.6747,
|
|
"mean_token_accuracy": 0.18364477753639222,
|
|
"num_tokens": 10497223.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.031793642044067,
|
|
"epoch": 4.935672514619883,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003262024721053827,
|
|
"loss": 5.7378,
|
|
"mean_token_accuracy": 0.1792234867811203,
|
|
"num_tokens": 10508642.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.033666801452637,
|
|
"epoch": 4.941520467836257,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00032574592089081375,
|
|
"loss": 5.7215,
|
|
"mean_token_accuracy": 0.18227653652429582,
|
|
"num_tokens": 10521056.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 6.106046390533447,
|
|
"epoch": 4.947368421052632,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032528914943530905,
|
|
"loss": 5.7624,
|
|
"mean_token_accuracy": 0.17589445859193803,
|
|
"num_tokens": 10532719.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.036490249633789,
|
|
"epoch": 4.953216374269006,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003248321597212896,
|
|
"loss": 5.7847,
|
|
"mean_token_accuracy": 0.18094717264175414,
|
|
"num_tokens": 10545751.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.987871932983398,
|
|
"epoch": 4.95906432748538,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003243749537321236,
|
|
"loss": 5.7583,
|
|
"mean_token_accuracy": 0.1749679997563362,
|
|
"num_tokens": 10559215.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 6.045964145660401,
|
|
"epoch": 4.964912280701754,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032391753345211767,
|
|
"loss": 5.7417,
|
|
"mean_token_accuracy": 0.17879257202148438,
|
|
"num_tokens": 10571643.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.122284364700318,
|
|
"epoch": 4.970760233918129,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003234599008665088,
|
|
"loss": 5.7863,
|
|
"mean_token_accuracy": 0.17536696195602416,
|
|
"num_tokens": 10584048.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 6.0027031898498535,
|
|
"epoch": 4.976608187134503,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003230020579614551,
|
|
"loss": 5.7648,
|
|
"mean_token_accuracy": 0.17808147370815278,
|
|
"num_tokens": 10596306.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.876022577285767,
|
|
"epoch": 4.982456140350877,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003225440067240276,
|
|
"loss": 5.612,
|
|
"mean_token_accuracy": 0.19063190817832948,
|
|
"num_tokens": 10609417.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.057668018341064,
|
|
"epoch": 4.988304093567251,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032208574914220167,
|
|
"loss": 5.7627,
|
|
"mean_token_accuracy": 0.17722997069358826,
|
|
"num_tokens": 10623157.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.922485971450806,
|
|
"epoch": 4.994152046783626,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003216272872048479,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.1865217000246048,
|
|
"num_tokens": 10635584.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.025067329406738,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003211686229017241,
|
|
"loss": 5.7845,
|
|
"mean_token_accuracy": 0.1800237014889717,
|
|
"num_tokens": 10647420.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 6.089043378829956,
|
|
"epoch": 5.005847953216374,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003207097582234662,
|
|
"loss": 5.5156,
|
|
"mean_token_accuracy": 0.18573589771986007,
|
|
"num_tokens": 10661511.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.901047658920288,
|
|
"epoch": 5.011695906432749,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003202506951615798,
|
|
"loss": 5.2518,
|
|
"mean_token_accuracy": 0.2090180605649948,
|
|
"num_tokens": 10673858.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.989277362823486,
|
|
"epoch": 5.017543859649122,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00031979143570843145,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.18558239340782165,
|
|
"num_tokens": 10686620.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 6.021729326248169,
|
|
"epoch": 5.023391812865497,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00031933198185724016,
|
|
"loss": 5.4414,
|
|
"mean_token_accuracy": 0.1897239491343498,
|
|
"num_tokens": 10699452.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.9853307723999025,
|
|
"epoch": 5.029239766081871,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00031887233560206864,
|
|
"loss": 5.4354,
|
|
"mean_token_accuracy": 0.1913822427392006,
|
|
"num_tokens": 10711327.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 6.045548057556152,
|
|
"epoch": 5.035087719298246,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031841249893781447,
|
|
"loss": 5.4897,
|
|
"mean_token_accuracy": 0.18177365362644196,
|
|
"num_tokens": 10723487.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.99133677482605,
|
|
"epoch": 5.04093567251462,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003179524738602019,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.18719770312309264,
|
|
"num_tokens": 10735960.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.883477973937988,
|
|
"epoch": 5.046783625730994,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00031749226236577277,
|
|
"loss": 5.3788,
|
|
"mean_token_accuracy": 0.19386433213949203,
|
|
"num_tokens": 10749256.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.050616216659546,
|
|
"epoch": 5.052631578947368,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031703186645187776,
|
|
"loss": 5.4302,
|
|
"mean_token_accuracy": 0.1898864194750786,
|
|
"num_tokens": 10761295.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.9561910152435305,
|
|
"epoch": 5.058479532163743,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00031657128811666836,
|
|
"loss": 5.4003,
|
|
"mean_token_accuracy": 0.19574057161808014,
|
|
"num_tokens": 10773874.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 6.030394506454468,
|
|
"epoch": 5.064327485380117,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00031611052935908736,
|
|
"loss": 5.4785,
|
|
"mean_token_accuracy": 0.1891598716378212,
|
|
"num_tokens": 10786749.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.970602703094483,
|
|
"epoch": 5.0701754385964914,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000315649592178861,
|
|
"loss": 5.3958,
|
|
"mean_token_accuracy": 0.1949351578950882,
|
|
"num_tokens": 10798722.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.8993137836456295,
|
|
"epoch": 5.076023391812866,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003151884785764897,
|
|
"loss": 5.3744,
|
|
"mean_token_accuracy": 0.19463745206594468,
|
|
"num_tokens": 10810461.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.0198029518127445,
|
|
"epoch": 5.081871345029239,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003147271905532394,
|
|
"loss": 5.462,
|
|
"mean_token_accuracy": 0.19428947418928147,
|
|
"num_tokens": 10823797.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.988357877731323,
|
|
"epoch": 5.087719298245614,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003142657301111333,
|
|
"loss": 5.4217,
|
|
"mean_token_accuracy": 0.1948299378156662,
|
|
"num_tokens": 10835909.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.958567714691162,
|
|
"epoch": 5.093567251461988,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003138040992529428,
|
|
"loss": 5.4741,
|
|
"mean_token_accuracy": 0.1908164232969284,
|
|
"num_tokens": 10848541.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.039626836776733,
|
|
"epoch": 5.099415204678363,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000313342299982179,
|
|
"loss": 5.4962,
|
|
"mean_token_accuracy": 0.18445653915405275,
|
|
"num_tokens": 10862412.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.997631454467774,
|
|
"epoch": 5.105263157894737,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003128803343030836,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.1841598242521286,
|
|
"num_tokens": 10875300.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 6.014658069610595,
|
|
"epoch": 5.111111111111111,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031241820422062117,
|
|
"loss": 5.4734,
|
|
"mean_token_accuracy": 0.19178768545389174,
|
|
"num_tokens": 10887716.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 6.012116813659668,
|
|
"epoch": 5.116959064327485,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00031195591174046914,
|
|
"loss": 5.4476,
|
|
"mean_token_accuracy": 0.19168466776609422,
|
|
"num_tokens": 10899945.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.989486598968506,
|
|
"epoch": 5.12280701754386,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00031149345886901,
|
|
"loss": 5.4629,
|
|
"mean_token_accuracy": 0.19434204399585725,
|
|
"num_tokens": 10912609.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 6.036200475692749,
|
|
"epoch": 5.128654970760234,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003110308476133225,
|
|
"loss": 5.4458,
|
|
"mean_token_accuracy": 0.18503202348947526,
|
|
"num_tokens": 10925852.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.887316036224365,
|
|
"epoch": 5.1345029239766085,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00031056807998117256,
|
|
"loss": 5.3064,
|
|
"mean_token_accuracy": 0.2038549080491066,
|
|
"num_tokens": 10939438.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.965604019165039,
|
|
"epoch": 5.140350877192983,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000310105157981005,
|
|
"loss": 5.4635,
|
|
"mean_token_accuracy": 0.18920304626226425,
|
|
"num_tokens": 10952082.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.890659379959106,
|
|
"epoch": 5.146198830409356,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00030964208362193436,
|
|
"loss": 5.3196,
|
|
"mean_token_accuracy": 0.20706926584243773,
|
|
"num_tokens": 10964373.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.9448442459106445,
|
|
"epoch": 5.152046783625731,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00030917885891373664,
|
|
"loss": 5.4147,
|
|
"mean_token_accuracy": 0.19285420030355455,
|
|
"num_tokens": 10975511.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.925690412521362,
|
|
"epoch": 5.157894736842105,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003087154858668404,
|
|
"loss": 5.4574,
|
|
"mean_token_accuracy": 0.19780198037624358,
|
|
"num_tokens": 10987241.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.9963782787322994,
|
|
"epoch": 5.16374269005848,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003082519664923177,
|
|
"loss": 5.4191,
|
|
"mean_token_accuracy": 0.19637430161237718,
|
|
"num_tokens": 10998610.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.886073303222656,
|
|
"epoch": 5.169590643274854,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000307788302801876,
|
|
"loss": 5.3352,
|
|
"mean_token_accuracy": 0.20464493930339814,
|
|
"num_tokens": 11010847.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.972019672393799,
|
|
"epoch": 5.175438596491228,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00030732449680784886,
|
|
"loss": 5.451,
|
|
"mean_token_accuracy": 0.18870837837457657,
|
|
"num_tokens": 11022364.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.991866397857666,
|
|
"epoch": 5.181286549707602,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003068605505231877,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.19285177439451218,
|
|
"num_tokens": 11036842.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.925907993316651,
|
|
"epoch": 5.187134502923977,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00030639646596145243,
|
|
"loss": 5.4119,
|
|
"mean_token_accuracy": 0.19109603613615037,
|
|
"num_tokens": 11049024.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.842715263366699,
|
|
"epoch": 5.192982456140351,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00030593224513680357,
|
|
"loss": 5.3913,
|
|
"mean_token_accuracy": 0.20553675144910813,
|
|
"num_tokens": 11061357.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.9635398387908936,
|
|
"epoch": 5.1988304093567255,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003054678900639927,
|
|
"loss": 5.4861,
|
|
"mean_token_accuracy": 0.1839690238237381,
|
|
"num_tokens": 11074995.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 6.011612367630005,
|
|
"epoch": 5.204678362573099,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00030500340275835405,
|
|
"loss": 5.4425,
|
|
"mean_token_accuracy": 0.18970632702112197,
|
|
"num_tokens": 11087435.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 6.0321893215179445,
|
|
"epoch": 5.2105263157894735,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00030453878523579596,
|
|
"loss": 5.489,
|
|
"mean_token_accuracy": 0.1906844675540924,
|
|
"num_tokens": 11100182.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 6.001064109802246,
|
|
"epoch": 5.216374269005848,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00030407403951279174,
|
|
"loss": 5.4542,
|
|
"mean_token_accuracy": 0.18916164040565492,
|
|
"num_tokens": 11112855.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.960168409347534,
|
|
"epoch": 5.222222222222222,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003036091676063711,
|
|
"loss": 5.4846,
|
|
"mean_token_accuracy": 0.18928785771131515,
|
|
"num_tokens": 11125620.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.974697494506836,
|
|
"epoch": 5.228070175438597,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003031441715341115,
|
|
"loss": 5.4814,
|
|
"mean_token_accuracy": 0.19442566037178038,
|
|
"num_tokens": 11138999.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 6.003526067733764,
|
|
"epoch": 5.23391812865497,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003026790533141293,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.19007026255130768,
|
|
"num_tokens": 11150755.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.912185192108154,
|
|
"epoch": 5.239766081871345,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003022138149650708,
|
|
"loss": 5.384,
|
|
"mean_token_accuracy": 0.20422173887491227,
|
|
"num_tokens": 11162296.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.926997089385987,
|
|
"epoch": 5.245614035087719,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003017484585061039,
|
|
"loss": 5.3766,
|
|
"mean_token_accuracy": 0.19927339851856232,
|
|
"num_tokens": 11173875.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.89944896697998,
|
|
"epoch": 5.251461988304094,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00030128298595690905,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.19918446242809296,
|
|
"num_tokens": 11185717.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.95864520072937,
|
|
"epoch": 5.257309941520468,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00030081739933767034,
|
|
"loss": 5.4173,
|
|
"mean_token_accuracy": 0.19788180142641068,
|
|
"num_tokens": 11196964.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.9288076877594,
|
|
"epoch": 5.2631578947368425,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003003517006690672,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.18629933446645736,
|
|
"num_tokens": 11209193.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 5.2631578947368425,
|
|
"eval_entropy": 5.876284589458004,
|
|
"eval_loss": 6.573401927947998,
|
|
"eval_mean_token_accuracy": 0.1507907477166273,
|
|
"eval_num_tokens": 11209193.0,
|
|
"eval_runtime": 2.2823,
|
|
"eval_samples_per_second": 1186.082,
|
|
"eval_steps_per_second": 148.534,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.948824071884156,
|
|
"epoch": 5.269005847953216,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00029988589197226515,
|
|
"loss": 5.4393,
|
|
"mean_token_accuracy": 0.19501043409109114,
|
|
"num_tokens": 11221927.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 6.035808038711548,
|
|
"epoch": 5.2748538011695905,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00029941997526890745,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.18238018155097963,
|
|
"num_tokens": 11235035.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 6.004548025131226,
|
|
"epoch": 5.280701754385965,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000298953952581106,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.19265481531620027,
|
|
"num_tokens": 11248059.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.984726333618164,
|
|
"epoch": 5.286549707602339,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00029848782593143246,
|
|
"loss": 5.4744,
|
|
"mean_token_accuracy": 0.19058849811553955,
|
|
"num_tokens": 11260262.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.858472013473511,
|
|
"epoch": 5.292397660818714,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00029802159734291015,
|
|
"loss": 5.4209,
|
|
"mean_token_accuracy": 0.19406241178512573,
|
|
"num_tokens": 11272940.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 6.025643968582154,
|
|
"epoch": 5.298245614035087,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002975552688390045,
|
|
"loss": 5.5177,
|
|
"mean_token_accuracy": 0.19065407514572144,
|
|
"num_tokens": 11286934.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.938787794113159,
|
|
"epoch": 5.304093567251462,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00029708884244361474,
|
|
"loss": 5.41,
|
|
"mean_token_accuracy": 0.19796527326107025,
|
|
"num_tokens": 11300213.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.819243288040161,
|
|
"epoch": 5.309941520467836,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00029662232018106473,
|
|
"loss": 5.355,
|
|
"mean_token_accuracy": 0.20654971599578859,
|
|
"num_tokens": 11312084.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.964603900909424,
|
|
"epoch": 5.315789473684211,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00029615570407609467,
|
|
"loss": 5.4633,
|
|
"mean_token_accuracy": 0.19208070933818816,
|
|
"num_tokens": 11324928.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.947579383850098,
|
|
"epoch": 5.321637426900585,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00029568899615385174,
|
|
"loss": 5.3998,
|
|
"mean_token_accuracy": 0.19761320799589158,
|
|
"num_tokens": 11336270.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.817259836196899,
|
|
"epoch": 5.3274853801169595,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000295222198439882,
|
|
"loss": 5.3044,
|
|
"mean_token_accuracy": 0.2075532019138336,
|
|
"num_tokens": 11348849.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.0071252346038815,
|
|
"epoch": 5.333333333333333,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002947553129601209,
|
|
"loss": 5.4701,
|
|
"mean_token_accuracy": 0.1949871227145195,
|
|
"num_tokens": 11360879.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.914417171478272,
|
|
"epoch": 5.3391812865497075,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002942883417408849,
|
|
"loss": 5.4545,
|
|
"mean_token_accuracy": 0.19224474728107452,
|
|
"num_tokens": 11372604.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.992934799194336,
|
|
"epoch": 5.345029239766082,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002938212868088626,
|
|
"loss": 5.5186,
|
|
"mean_token_accuracy": 0.19104269444942473,
|
|
"num_tokens": 11384607.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.927836990356445,
|
|
"epoch": 5.350877192982456,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00029335415019110593,
|
|
"loss": 5.3674,
|
|
"mean_token_accuracy": 0.20213899612426758,
|
|
"num_tokens": 11397620.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.941346263885498,
|
|
"epoch": 5.356725146198831,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002928869339150212,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.18442373126745223,
|
|
"num_tokens": 11410614.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 6.0068481922149655,
|
|
"epoch": 5.362573099415204,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002924196400083607,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.18756728023290634,
|
|
"num_tokens": 11422828.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.975446939468384,
|
|
"epoch": 5.368421052631579,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00029195227049921336,
|
|
"loss": 5.4582,
|
|
"mean_token_accuracy": 0.19261221587657928,
|
|
"num_tokens": 11435335.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.975659036636353,
|
|
"epoch": 5.374269005847953,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00029148482741599634,
|
|
"loss": 5.5342,
|
|
"mean_token_accuracy": 0.1838696762919426,
|
|
"num_tokens": 11448804.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.979130220413208,
|
|
"epoch": 5.380116959064328,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00029101731278744636,
|
|
"loss": 5.4568,
|
|
"mean_token_accuracy": 0.19432918578386307,
|
|
"num_tokens": 11460459.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.883121490478516,
|
|
"epoch": 5.385964912280702,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002905497286426102,
|
|
"loss": 5.4357,
|
|
"mean_token_accuracy": 0.19023000746965407,
|
|
"num_tokens": 11471684.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.93282527923584,
|
|
"epoch": 5.391812865497076,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00029008207701083677,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.20480825901031494,
|
|
"num_tokens": 11484303.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.8664806365966795,
|
|
"epoch": 5.39766081871345,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002896143599217674,
|
|
"loss": 5.3587,
|
|
"mean_token_accuracy": 0.205398990213871,
|
|
"num_tokens": 11497283.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.909426355361939,
|
|
"epoch": 5.4035087719298245,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002891465794053282,
|
|
"loss": 5.4046,
|
|
"mean_token_accuracy": 0.19378584921360015,
|
|
"num_tokens": 11509157.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.983506965637207,
|
|
"epoch": 5.409356725146199,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002886787374917198,
|
|
"loss": 5.5498,
|
|
"mean_token_accuracy": 0.18307142853736877,
|
|
"num_tokens": 11521510.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.804906415939331,
|
|
"epoch": 5.415204678362573,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028821083621140986,
|
|
"loss": 5.3244,
|
|
"mean_token_accuracy": 0.2069457694888115,
|
|
"num_tokens": 11533948.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.932200193405151,
|
|
"epoch": 5.421052631578947,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002877428775951234,
|
|
"loss": 5.4561,
|
|
"mean_token_accuracy": 0.2003613978624344,
|
|
"num_tokens": 11545718.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.967337894439697,
|
|
"epoch": 5.426900584795321,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002872748636738344,
|
|
"loss": 5.4805,
|
|
"mean_token_accuracy": 0.19335651993751526,
|
|
"num_tokens": 11558167.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.94414324760437,
|
|
"epoch": 5.432748538011696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002868067964787568,
|
|
"loss": 5.5579,
|
|
"mean_token_accuracy": 0.19486698359251023,
|
|
"num_tokens": 11573188.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.935049676895142,
|
|
"epoch": 5.43859649122807,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00028633867804133583,
|
|
"loss": 5.4848,
|
|
"mean_token_accuracy": 0.18951261937618255,
|
|
"num_tokens": 11584751.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.033418941497803,
|
|
"epoch": 5.444444444444445,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002858705103932391,
|
|
"loss": 5.4996,
|
|
"mean_token_accuracy": 0.18893657624721527,
|
|
"num_tokens": 11597378.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.940593290328979,
|
|
"epoch": 5.450292397660819,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00028540229556634785,
|
|
"loss": 5.4788,
|
|
"mean_token_accuracy": 0.18982810229063035,
|
|
"num_tokens": 11609479.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.964474868774414,
|
|
"epoch": 5.456140350877193,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002849340355927477,
|
|
"loss": 5.4918,
|
|
"mean_token_accuracy": 0.19218603968620301,
|
|
"num_tokens": 11621321.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.940859794616699,
|
|
"epoch": 5.461988304093567,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002844657325047208,
|
|
"loss": 5.4221,
|
|
"mean_token_accuracy": 0.19410355091094972,
|
|
"num_tokens": 11633397.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.8772368907928465,
|
|
"epoch": 5.4678362573099415,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002839973883347361,
|
|
"loss": 5.5071,
|
|
"mean_token_accuracy": 0.18924573361873626,
|
|
"num_tokens": 11645229.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 6.039159679412842,
|
|
"epoch": 5.473684210526316,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028352900511544076,
|
|
"loss": 5.5839,
|
|
"mean_token_accuracy": 0.18156208246946334,
|
|
"num_tokens": 11657873.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.897280645370484,
|
|
"epoch": 5.47953216374269,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00028306058487965156,
|
|
"loss": 5.3598,
|
|
"mean_token_accuracy": 0.2037922516465187,
|
|
"num_tokens": 11670620.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.877945280075073,
|
|
"epoch": 5.485380116959064,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002825921296603461,
|
|
"loss": 5.4683,
|
|
"mean_token_accuracy": 0.19935393184423447,
|
|
"num_tokens": 11683090.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 6.013880586624145,
|
|
"epoch": 5.491228070175438,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00028212364149065347,
|
|
"loss": 5.5973,
|
|
"mean_token_accuracy": 0.18494570553302764,
|
|
"num_tokens": 11694949.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 6.058586692810058,
|
|
"epoch": 5.497076023391813,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002816551224038458,
|
|
"loss": 5.5876,
|
|
"mean_token_accuracy": 0.17943034768104554,
|
|
"num_tokens": 11708258.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 6.065577173233033,
|
|
"epoch": 5.502923976608187,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00028118657443332984,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.18721984922885895,
|
|
"num_tokens": 11720024.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.956423091888428,
|
|
"epoch": 5.508771929824562,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002807179996126371,
|
|
"loss": 5.4816,
|
|
"mean_token_accuracy": 0.19129722267389299,
|
|
"num_tokens": 11733323.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.91726508140564,
|
|
"epoch": 5.514619883040936,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00028024939997541606,
|
|
"loss": 5.4485,
|
|
"mean_token_accuracy": 0.19177333861589432,
|
|
"num_tokens": 11745733.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.981409931182862,
|
|
"epoch": 5.52046783625731,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00027978077755542264,
|
|
"loss": 5.5176,
|
|
"mean_token_accuracy": 0.18772226721048355,
|
|
"num_tokens": 11758307.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.97986855506897,
|
|
"epoch": 5.526315789473684,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002793121343865119,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.1938734084367752,
|
|
"num_tokens": 11769411.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.852325868606568,
|
|
"epoch": 5.5321637426900585,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002788434725026286,
|
|
"loss": 5.3743,
|
|
"mean_token_accuracy": 0.2037036031484604,
|
|
"num_tokens": 11781980.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.944281005859375,
|
|
"epoch": 5.538011695906433,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000278374793937799,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.19133008569478988,
|
|
"num_tokens": 11794529.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 6.024075508117676,
|
|
"epoch": 5.543859649122807,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00027790610072612174,
|
|
"loss": 5.5521,
|
|
"mean_token_accuracy": 0.1857476696372032,
|
|
"num_tokens": 11807535.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.920930862426758,
|
|
"epoch": 5.549707602339181,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00027743739490175903,
|
|
"loss": 5.3655,
|
|
"mean_token_accuracy": 0.20021205246448517,
|
|
"num_tokens": 11820393.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.927069568634034,
|
|
"epoch": 5.555555555555555,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002769686784989276,
|
|
"loss": 5.5665,
|
|
"mean_token_accuracy": 0.19014804065227509,
|
|
"num_tokens": 11833146.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.8506035804748535,
|
|
"epoch": 5.56140350877193,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00027649995355189026,
|
|
"loss": 5.3235,
|
|
"mean_token_accuracy": 0.20847362130880356,
|
|
"num_tokens": 11844753.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.82243275642395,
|
|
"epoch": 5.567251461988304,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000276031222094947,
|
|
"loss": 5.2858,
|
|
"mean_token_accuracy": 0.20902137607336044,
|
|
"num_tokens": 11856549.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.873478174209595,
|
|
"epoch": 5.573099415204679,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002755624861624259,
|
|
"loss": 5.3827,
|
|
"mean_token_accuracy": 0.20422631800174712,
|
|
"num_tokens": 11868475.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.865388011932373,
|
|
"epoch": 5.578947368421053,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002750937477886745,
|
|
"loss": 5.3934,
|
|
"mean_token_accuracy": 0.19930585473775864,
|
|
"num_tokens": 11880781.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 6.013696956634521,
|
|
"epoch": 5.584795321637427,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000274625009008051,
|
|
"loss": 5.5631,
|
|
"mean_token_accuracy": 0.18276871293783187,
|
|
"num_tokens": 11892959.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.99136266708374,
|
|
"epoch": 5.590643274853801,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002741562718549152,
|
|
"loss": 5.4937,
|
|
"mean_token_accuracy": 0.19513768404722215,
|
|
"num_tokens": 11906062.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.8985576152801515,
|
|
"epoch": 5.5964912280701755,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00027368753836362023,
|
|
"loss": 5.4074,
|
|
"mean_token_accuracy": 0.19351859241724015,
|
|
"num_tokens": 11919114.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.938813304901123,
|
|
"epoch": 5.60233918128655,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002732188105685027,
|
|
"loss": 5.479,
|
|
"mean_token_accuracy": 0.19051125943660735,
|
|
"num_tokens": 11931654.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.898592567443847,
|
|
"epoch": 5.6081871345029235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002727500905038752,
|
|
"loss": 5.4668,
|
|
"mean_token_accuracy": 0.19124536365270614,
|
|
"num_tokens": 11944450.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.002505350112915,
|
|
"epoch": 5.614035087719298,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002722813802040161,
|
|
"loss": 5.5338,
|
|
"mean_token_accuracy": 0.18638397753238678,
|
|
"num_tokens": 11957420.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.898351001739502,
|
|
"epoch": 5.619883040935672,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00027181268170316193,
|
|
"loss": 5.3827,
|
|
"mean_token_accuracy": 0.20197507739067078,
|
|
"num_tokens": 11970218.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.81345911026001,
|
|
"epoch": 5.625730994152047,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002713439970354978,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.19874269515275955,
|
|
"num_tokens": 11982285.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 6.004039001464844,
|
|
"epoch": 5.631578947368421,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002708753282351486,
|
|
"loss": 5.5029,
|
|
"mean_token_accuracy": 0.19823363572359085,
|
|
"num_tokens": 11995456.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.893980550765991,
|
|
"epoch": 5.637426900584796,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002704066773361706,
|
|
"loss": 5.4922,
|
|
"mean_token_accuracy": 0.19233156889677047,
|
|
"num_tokens": 12008036.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 6.015719270706176,
|
|
"epoch": 5.643274853801169,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002699380463725425,
|
|
"loss": 5.5817,
|
|
"mean_token_accuracy": 0.17800767123699188,
|
|
"num_tokens": 12021267.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.969558572769165,
|
|
"epoch": 5.649122807017544,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000269469437378156,
|
|
"loss": 5.4277,
|
|
"mean_token_accuracy": 0.19691437631845474,
|
|
"num_tokens": 12033369.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.902233457565307,
|
|
"epoch": 5.654970760233918,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026900085238680787,
|
|
"loss": 5.4861,
|
|
"mean_token_accuracy": 0.19313597530126572,
|
|
"num_tokens": 12046330.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.962917613983154,
|
|
"epoch": 5.6608187134502925,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002685322934321907,
|
|
"loss": 5.459,
|
|
"mean_token_accuracy": 0.1935941219329834,
|
|
"num_tokens": 12059220.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.872378015518189,
|
|
"epoch": 5.666666666666667,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002680637625478837,
|
|
"loss": 5.4114,
|
|
"mean_token_accuracy": 0.19792089164257048,
|
|
"num_tokens": 12071695.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.860135078430176,
|
|
"epoch": 5.6725146198830405,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002675952617673446,
|
|
"loss": 5.3795,
|
|
"mean_token_accuracy": 0.1924514502286911,
|
|
"num_tokens": 12084792.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.958759403228759,
|
|
"epoch": 5.678362573099415,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002671267931239006,
|
|
"loss": 5.407,
|
|
"mean_token_accuracy": 0.20442104637622832,
|
|
"num_tokens": 12096660.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.816103172302246,
|
|
"epoch": 5.684210526315789,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026665835865073896,
|
|
"loss": 5.3671,
|
|
"mean_token_accuracy": 0.1999454081058502,
|
|
"num_tokens": 12107880.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.906464099884033,
|
|
"epoch": 5.690058479532164,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002661899603808989,
|
|
"loss": 5.575,
|
|
"mean_token_accuracy": 0.187053319811821,
|
|
"num_tokens": 12121987.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.9588244438171385,
|
|
"epoch": 5.695906432748538,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002657216003472626,
|
|
"loss": 5.4292,
|
|
"mean_token_accuracy": 0.19662821143865586,
|
|
"num_tokens": 12133375.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.953715372085571,
|
|
"epoch": 5.701754385964913,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026525328058254604,
|
|
"loss": 5.5211,
|
|
"mean_token_accuracy": 0.1949773296713829,
|
|
"num_tokens": 12146016.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.9281025409698485,
|
|
"epoch": 5.707602339181286,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026478500311929076,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.19733416885137559,
|
|
"num_tokens": 12158306.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 6.00237889289856,
|
|
"epoch": 5.713450292397661,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026431676998985444,
|
|
"loss": 5.483,
|
|
"mean_token_accuracy": 0.19225390255451202,
|
|
"num_tokens": 12171934.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.918964385986328,
|
|
"epoch": 5.719298245614035,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002638485832264024,
|
|
"loss": 5.4303,
|
|
"mean_token_accuracy": 0.200017748773098,
|
|
"num_tokens": 12185924.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.942403364181518,
|
|
"epoch": 5.7251461988304095,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026338044486089895,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.19176100492477416,
|
|
"num_tokens": 12198451.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.94615912437439,
|
|
"epoch": 5.730994152046784,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026291235692509795,
|
|
"loss": 5.4919,
|
|
"mean_token_accuracy": 0.19788724035024643,
|
|
"num_tokens": 12210909.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.876265811920166,
|
|
"epoch": 5.7368421052631575,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026244432145053467,
|
|
"loss": 5.4054,
|
|
"mean_token_accuracy": 0.19663514494895934,
|
|
"num_tokens": 12222635.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.928455257415772,
|
|
"epoch": 5.742690058479532,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002619763404685167,
|
|
"loss": 5.4403,
|
|
"mean_token_accuracy": 0.1949448272585869,
|
|
"num_tokens": 12234868.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.964970302581787,
|
|
"epoch": 5.748538011695906,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026150841601011495,
|
|
"loss": 5.5037,
|
|
"mean_token_accuracy": 0.19281102865934371,
|
|
"num_tokens": 12246667.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 6.019935178756714,
|
|
"epoch": 5.754385964912281,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0002610405501061552,
|
|
"loss": 5.5129,
|
|
"mean_token_accuracy": 0.18634127527475358,
|
|
"num_tokens": 12259164.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.886829042434693,
|
|
"epoch": 5.760233918128655,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002605727447872091,
|
|
"loss": 5.4072,
|
|
"mean_token_accuracy": 0.2007581502199173,
|
|
"num_tokens": 12271681.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.921840858459473,
|
|
"epoch": 5.76608187134503,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002601050020835851,
|
|
"loss": 5.5149,
|
|
"mean_token_accuracy": 0.19163930416107178,
|
|
"num_tokens": 12284773.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.995645380020141,
|
|
"epoch": 5.771929824561403,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00025963732402532026,
|
|
"loss": 5.5362,
|
|
"mean_token_accuracy": 0.18075649440288544,
|
|
"num_tokens": 12297768.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.943518877029419,
|
|
"epoch": 5.777777777777778,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002591697126421707,
|
|
"loss": 5.4524,
|
|
"mean_token_accuracy": 0.18932004123926163,
|
|
"num_tokens": 12309816.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.9088483333587645,
|
|
"epoch": 5.783625730994152,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002587021699636035,
|
|
"loss": 5.5251,
|
|
"mean_token_accuracy": 0.18891255408525467,
|
|
"num_tokens": 12321944.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.978242349624634,
|
|
"epoch": 5.7894736842105265,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00025823469801878734,
|
|
"loss": 5.5113,
|
|
"mean_token_accuracy": 0.1878558248281479,
|
|
"num_tokens": 12333483.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.975175952911377,
|
|
"epoch": 5.7953216374269,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00025776729883658406,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.2002690389752388,
|
|
"num_tokens": 12345300.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.95694637298584,
|
|
"epoch": 5.8011695906432745,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002572999744455395,
|
|
"loss": 5.5021,
|
|
"mean_token_accuracy": 0.19280034750699998,
|
|
"num_tokens": 12357143.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.980992698669434,
|
|
"epoch": 5.807017543859649,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00025683272687387513,
|
|
"loss": 5.5508,
|
|
"mean_token_accuracy": 0.1847701609134674,
|
|
"num_tokens": 12369524.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 6.030041980743408,
|
|
"epoch": 5.812865497076023,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002563655581494789,
|
|
"loss": 5.507,
|
|
"mean_token_accuracy": 0.19017528593540192,
|
|
"num_tokens": 12380403.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.905593013763427,
|
|
"epoch": 5.818713450292398,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00025589847029989653,
|
|
"loss": 5.4561,
|
|
"mean_token_accuracy": 0.19175576120615007,
|
|
"num_tokens": 12391839.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.969892358779907,
|
|
"epoch": 5.824561403508772,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00025543146535232295,
|
|
"loss": 5.5395,
|
|
"mean_token_accuracy": 0.19064286947250367,
|
|
"num_tokens": 12403431.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.958486461639405,
|
|
"epoch": 5.830409356725146,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00025496454533359284,
|
|
"loss": 5.5212,
|
|
"mean_token_accuracy": 0.1901782289147377,
|
|
"num_tokens": 12416368.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.855278253555298,
|
|
"epoch": 5.83625730994152,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00025449771227017265,
|
|
"loss": 5.4345,
|
|
"mean_token_accuracy": 0.19709573686122894,
|
|
"num_tokens": 12429022.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.9747887134552,
|
|
"epoch": 5.842105263157895,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002540309681881516,
|
|
"loss": 5.4825,
|
|
"mean_token_accuracy": 0.19121056646108628,
|
|
"num_tokens": 12443035.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.930963706970215,
|
|
"epoch": 5.847953216374269,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002535643151132321,
|
|
"loss": 5.4442,
|
|
"mean_token_accuracy": 0.1950688824057579,
|
|
"num_tokens": 12454647.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.847953216374269,
|
|
"eval_entropy": 5.8640149265615635,
|
|
"eval_loss": 6.544374465942383,
|
|
"eval_mean_token_accuracy": 0.15245128400997426,
|
|
"eval_num_tokens": 12454647.0,
|
|
"eval_runtime": 2.4726,
|
|
"eval_samples_per_second": 1094.779,
|
|
"eval_steps_per_second": 137.1,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.990120840072632,
|
|
"epoch": 5.853801169590644,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00025309775507072234,
|
|
"loss": 5.5373,
|
|
"mean_token_accuracy": 0.18221253901720047,
|
|
"num_tokens": 12467165.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.929445219039917,
|
|
"epoch": 5.859649122807017,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002526312900855261,
|
|
"loss": 5.4761,
|
|
"mean_token_accuracy": 0.1993470385670662,
|
|
"num_tokens": 12481419.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.96486873626709,
|
|
"epoch": 5.8654970760233915,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002521649221821349,
|
|
"loss": 5.4894,
|
|
"mean_token_accuracy": 0.18978445678949357,
|
|
"num_tokens": 12493211.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.944071817398071,
|
|
"epoch": 5.871345029239766,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00025169865338461894,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.19630177915096284,
|
|
"num_tokens": 12504967.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.991852426528931,
|
|
"epoch": 5.87719298245614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002512324857166182,
|
|
"loss": 5.5185,
|
|
"mean_token_accuracy": 0.19692197293043137,
|
|
"num_tokens": 12517555.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.868841505050659,
|
|
"epoch": 5.883040935672515,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002507664212013337,
|
|
"loss": 5.4311,
|
|
"mean_token_accuracy": 0.19558733105659484,
|
|
"num_tokens": 12529749.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.89975323677063,
|
|
"epoch": 5.888888888888889,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00025030046186151915,
|
|
"loss": 5.4981,
|
|
"mean_token_accuracy": 0.1928455874323845,
|
|
"num_tokens": 12542403.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.88417067527771,
|
|
"epoch": 5.894736842105263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002498346097194712,
|
|
"loss": 5.4286,
|
|
"mean_token_accuracy": 0.19918245524168016,
|
|
"num_tokens": 12555061.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.932634401321411,
|
|
"epoch": 5.900584795321637,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00024936886679702167,
|
|
"loss": 5.5092,
|
|
"mean_token_accuracy": 0.19107322990894318,
|
|
"num_tokens": 12569315.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.947402667999268,
|
|
"epoch": 5.906432748538012,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00024890323511552825,
|
|
"loss": 5.4932,
|
|
"mean_token_accuracy": 0.19295616894960405,
|
|
"num_tokens": 12581494.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.977280998229981,
|
|
"epoch": 5.912280701754386,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00024843771669586575,
|
|
"loss": 5.4055,
|
|
"mean_token_accuracy": 0.20475989282131196,
|
|
"num_tokens": 12593997.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.940289688110352,
|
|
"epoch": 5.918128654970761,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002479723135584176,
|
|
"loss": 5.5169,
|
|
"mean_token_accuracy": 0.1893742099404335,
|
|
"num_tokens": 12606160.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.929003143310547,
|
|
"epoch": 5.923976608187134,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002475070277230668,
|
|
"loss": 5.5073,
|
|
"mean_token_accuracy": 0.18923161327838897,
|
|
"num_tokens": 12617735.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.949045038223266,
|
|
"epoch": 5.9298245614035086,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002470418612091871,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.18994081467390062,
|
|
"num_tokens": 12629867.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.94757194519043,
|
|
"epoch": 5.935672514619883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002465768160356347,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.19230643957853316,
|
|
"num_tokens": 12641677.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.942834424972534,
|
|
"epoch": 5.941520467836257,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002461118942207388,
|
|
"loss": 5.4947,
|
|
"mean_token_accuracy": 0.1864371806383133,
|
|
"num_tokens": 12652772.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.850364685058594,
|
|
"epoch": 5.947368421052632,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024564709778229353,
|
|
"loss": 5.3228,
|
|
"mean_token_accuracy": 0.2117023468017578,
|
|
"num_tokens": 12665139.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.926506662368775,
|
|
"epoch": 5.953216374269006,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024518242873754887,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.19036668688058853,
|
|
"num_tokens": 12677788.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.86831259727478,
|
|
"epoch": 5.95906432748538,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00024471788910320163,
|
|
"loss": 5.3564,
|
|
"mean_token_accuracy": 0.20837045460939407,
|
|
"num_tokens": 12689898.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.907672452926636,
|
|
"epoch": 5.964912280701754,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00024425348089538714,
|
|
"loss": 5.4324,
|
|
"mean_token_accuracy": 0.19490057080984116,
|
|
"num_tokens": 12702078.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.949279117584228,
|
|
"epoch": 5.970760233918129,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002437892061296705,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.1934816613793373,
|
|
"num_tokens": 12713802.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.844375133514404,
|
|
"epoch": 5.976608187134503,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00024332506682103722,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.19540058821439743,
|
|
"num_tokens": 12725757.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.964664602279663,
|
|
"epoch": 5.982456140350877,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00024286106498388533,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.18981240391731263,
|
|
"num_tokens": 12738665.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 6.025056648254394,
|
|
"epoch": 5.988304093567251,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002423972026320162,
|
|
"loss": 5.5432,
|
|
"mean_token_accuracy": 0.18444295972585678,
|
|
"num_tokens": 12752938.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.8327264308929445,
|
|
"epoch": 5.994152046783626,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00024193348177862551,
|
|
"loss": 5.3983,
|
|
"mean_token_accuracy": 0.20766942650079728,
|
|
"num_tokens": 12764968.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.93146243095398,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00024146990443629508,
|
|
"loss": 5.4708,
|
|
"mean_token_accuracy": 0.19202574640512465,
|
|
"num_tokens": 12776904.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.971071767807007,
|
|
"epoch": 6.005847953216374,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002410064726169838,
|
|
"loss": 5.2921,
|
|
"mean_token_accuracy": 0.2047412022948265,
|
|
"num_tokens": 12789603.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.88391342163086,
|
|
"epoch": 6.011695906432749,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00024054318833201894,
|
|
"loss": 5.1771,
|
|
"mean_token_accuracy": 0.21095502227544785,
|
|
"num_tokens": 12801569.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.884033155441284,
|
|
"epoch": 6.017543859649122,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00024008005359208774,
|
|
"loss": 5.2288,
|
|
"mean_token_accuracy": 0.204602912068367,
|
|
"num_tokens": 12813617.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.986951589584351,
|
|
"epoch": 6.023391812865497,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00023961707040722792,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.2075929433107376,
|
|
"num_tokens": 12826202.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.9600403785705565,
|
|
"epoch": 6.029239766081871,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00023915424078681987,
|
|
"loss": 5.2931,
|
|
"mean_token_accuracy": 0.2009615033864975,
|
|
"num_tokens": 12838285.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.915965604782104,
|
|
"epoch": 6.035087719298246,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00023869156673957733,
|
|
"loss": 5.227,
|
|
"mean_token_accuracy": 0.2054852619767189,
|
|
"num_tokens": 12850517.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.946900463104248,
|
|
"epoch": 6.04093567251462,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00023822905027353863,
|
|
"loss": 5.2965,
|
|
"mean_token_accuracy": 0.19976856261491777,
|
|
"num_tokens": 12862244.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.921419334411621,
|
|
"epoch": 6.046783625730994,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002377666933960586,
|
|
"loss": 5.3224,
|
|
"mean_token_accuracy": 0.2036558657884598,
|
|
"num_tokens": 12874602.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.873494100570679,
|
|
"epoch": 6.052631578947368,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002373044981137994,
|
|
"loss": 5.1898,
|
|
"mean_token_accuracy": 0.20810480117797853,
|
|
"num_tokens": 12885878.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.932150936126709,
|
|
"epoch": 6.058479532163743,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023684246643272132,
|
|
"loss": 5.269,
|
|
"mean_token_accuracy": 0.2106764331459999,
|
|
"num_tokens": 12898120.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.911846494674682,
|
|
"epoch": 6.064327485380117,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002363806003580752,
|
|
"loss": 5.2094,
|
|
"mean_token_accuracy": 0.21006779372692108,
|
|
"num_tokens": 12909776.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.862727499008178,
|
|
"epoch": 6.0701754385964914,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023591890189439296,
|
|
"loss": 5.2972,
|
|
"mean_token_accuracy": 0.20344337224960327,
|
|
"num_tokens": 12923518.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.883879709243774,
|
|
"epoch": 6.076023391812866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023545737304547893,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.20876469761133193,
|
|
"num_tokens": 12935416.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.901695108413696,
|
|
"epoch": 6.081871345029239,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002349960158144015,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.21417600363492967,
|
|
"num_tokens": 12947426.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.871595859527588,
|
|
"epoch": 6.087719298245614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000234534832203484,
|
|
"loss": 5.1904,
|
|
"mean_token_accuracy": 0.2131078138947487,
|
|
"num_tokens": 12959510.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.772227811813354,
|
|
"epoch": 6.093567251461988,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002340738242142965,
|
|
"loss": 5.1812,
|
|
"mean_token_accuracy": 0.21209583729505538,
|
|
"num_tokens": 12971479.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.9340564727783205,
|
|
"epoch": 6.099415204678363,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023361299384764677,
|
|
"loss": 5.2993,
|
|
"mean_token_accuracy": 0.2027627557516098,
|
|
"num_tokens": 12984473.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 6.013600254058838,
|
|
"epoch": 6.105263157894737,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023315234310357147,
|
|
"loss": 5.3663,
|
|
"mean_token_accuracy": 0.19507211744785308,
|
|
"num_tokens": 12997239.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.920335006713867,
|
|
"epoch": 6.111111111111111,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023269187398132802,
|
|
"loss": 5.259,
|
|
"mean_token_accuracy": 0.20516862124204635,
|
|
"num_tokens": 13010115.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.840462112426758,
|
|
"epoch": 6.116959064327485,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023223158847938543,
|
|
"loss": 5.147,
|
|
"mean_token_accuracy": 0.20946022123098373,
|
|
"num_tokens": 13022456.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.889286136627197,
|
|
"epoch": 6.12280701754386,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023177148859541565,
|
|
"loss": 5.2507,
|
|
"mean_token_accuracy": 0.2058929443359375,
|
|
"num_tokens": 13034860.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.9037909507751465,
|
|
"epoch": 6.128654970760234,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023131157632628525,
|
|
"loss": 5.28,
|
|
"mean_token_accuracy": 0.20258750915527343,
|
|
"num_tokens": 13047677.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.916197824478149,
|
|
"epoch": 6.1345029239766085,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023085185366804656,
|
|
"loss": 5.2087,
|
|
"mean_token_accuracy": 0.22138769179582596,
|
|
"num_tokens": 13059773.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.892022228240966,
|
|
"epoch": 6.140350877192983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023039232261592868,
|
|
"loss": 5.2513,
|
|
"mean_token_accuracy": 0.20069521963596343,
|
|
"num_tokens": 13070949.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.801952314376831,
|
|
"epoch": 6.146198830409356,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022993298516432937,
|
|
"loss": 5.1946,
|
|
"mean_token_accuracy": 0.20589322000741958,
|
|
"num_tokens": 13082494.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.934401130676269,
|
|
"epoch": 6.152046783625731,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022947384330680636,
|
|
"loss": 5.2663,
|
|
"mean_token_accuracy": 0.20695140957832336,
|
|
"num_tokens": 13093617.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.936961698532104,
|
|
"epoch": 6.157894736842105,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00022901489903606794,
|
|
"loss": 5.2667,
|
|
"mean_token_accuracy": 0.20977372229099273,
|
|
"num_tokens": 13105854.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.8561930656433105,
|
|
"epoch": 6.16374269005848,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022855615434396527,
|
|
"loss": 5.1672,
|
|
"mean_token_accuracy": 0.20685282051563264,
|
|
"num_tokens": 13118555.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.802619218826294,
|
|
"epoch": 6.169590643274854,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002280976112214832,
|
|
"loss": 5.1928,
|
|
"mean_token_accuracy": 0.20864183604717254,
|
|
"num_tokens": 13130369.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.828814840316772,
|
|
"epoch": 6.175438596491228,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022763927165873177,
|
|
"loss": 5.2102,
|
|
"mean_token_accuracy": 0.21179423332214356,
|
|
"num_tokens": 13141823.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.845408535003662,
|
|
"epoch": 6.181286549707602,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002271811376449376,
|
|
"loss": 5.1808,
|
|
"mean_token_accuracy": 0.2110128790140152,
|
|
"num_tokens": 13154874.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.898601293563843,
|
|
"epoch": 6.187134502923977,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00022672321116843497,
|
|
"loss": 5.2548,
|
|
"mean_token_accuracy": 0.20522923916578292,
|
|
"num_tokens": 13167514.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.931172180175781,
|
|
"epoch": 6.192982456140351,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00022626549421665788,
|
|
"loss": 5.2681,
|
|
"mean_token_accuracy": 0.20112829357385636,
|
|
"num_tokens": 13180114.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.916411113739014,
|
|
"epoch": 6.1988304093567255,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00022580798877613062,
|
|
"loss": 5.2848,
|
|
"mean_token_accuracy": 0.1989371657371521,
|
|
"num_tokens": 13193362.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.9057870388031,
|
|
"epoch": 6.204678362573099,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022535069683245952,
|
|
"loss": 5.2826,
|
|
"mean_token_accuracy": 0.20734949111938478,
|
|
"num_tokens": 13205540.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.871886491775513,
|
|
"epoch": 6.2105263157894735,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022489362037032446,
|
|
"loss": 5.1793,
|
|
"mean_token_accuracy": 0.21506163775920867,
|
|
"num_tokens": 13217559.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.840169668197632,
|
|
"epoch": 6.216374269005848,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002244367613734702,
|
|
"loss": 5.2023,
|
|
"mean_token_accuracy": 0.21939359307289125,
|
|
"num_tokens": 13229639.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.860301351547241,
|
|
"epoch": 6.222222222222222,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022398012182469734,
|
|
"loss": 5.2367,
|
|
"mean_token_accuracy": 0.20788523256778718,
|
|
"num_tokens": 13242464.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.824774932861328,
|
|
"epoch": 6.228070175438597,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002235237037058544,
|
|
"loss": 5.1646,
|
|
"mean_token_accuracy": 0.20927339494228364,
|
|
"num_tokens": 13254640.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.992959213256836,
|
|
"epoch": 6.23391812865497,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022306750899782875,
|
|
"loss": 5.3445,
|
|
"mean_token_accuracy": 0.19701227843761443,
|
|
"num_tokens": 13268438.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.921886825561524,
|
|
"epoch": 6.239766081871345,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000222611539680538,
|
|
"loss": 5.3281,
|
|
"mean_token_accuracy": 0.1959202319383621,
|
|
"num_tokens": 13280481.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 6.006784582138062,
|
|
"epoch": 6.245614035087719,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022215579773292188,
|
|
"loss": 5.3676,
|
|
"mean_token_accuracy": 0.1935841977596283,
|
|
"num_tokens": 13293775.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.888484334945678,
|
|
"epoch": 6.251461988304094,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00022170028513293288,
|
|
"loss": 5.2391,
|
|
"mean_token_accuracy": 0.21241168230772017,
|
|
"num_tokens": 13306570.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.893971681594849,
|
|
"epoch": 6.257309941520468,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002212450038575285,
|
|
"loss": 5.2545,
|
|
"mean_token_accuracy": 0.20222436636686325,
|
|
"num_tokens": 13318531.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.9497781753540036,
|
|
"epoch": 6.2631578947368425,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022078995588266227,
|
|
"loss": 5.3176,
|
|
"mean_token_accuracy": 0.20109234303236007,
|
|
"num_tokens": 13330827.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.902797555923462,
|
|
"epoch": 6.269005847953216,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002203351431832748,
|
|
"loss": 5.2305,
|
|
"mean_token_accuracy": 0.20438866317272186,
|
|
"num_tokens": 13344220.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.8953464984893795,
|
|
"epoch": 6.2748538011695905,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002198805677332858,
|
|
"loss": 5.2446,
|
|
"mean_token_accuracy": 0.20323979407548903,
|
|
"num_tokens": 13357045.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.914409589767456,
|
|
"epoch": 6.280701754385965,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021942623150558556,
|
|
"loss": 5.2514,
|
|
"mean_token_accuracy": 0.2149972915649414,
|
|
"num_tokens": 13369627.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.88154182434082,
|
|
"epoch": 6.286549707602339,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00021897213647202558,
|
|
"loss": 5.2658,
|
|
"mean_token_accuracy": 0.2015674591064453,
|
|
"num_tokens": 13381804.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.875191783905029,
|
|
"epoch": 6.292397660818714,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002185182846034111,
|
|
"loss": 5.2236,
|
|
"mean_token_accuracy": 0.20801629424095153,
|
|
"num_tokens": 13394946.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.981801843643188,
|
|
"epoch": 6.298245614035087,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002180646778694918,
|
|
"loss": 5.3345,
|
|
"mean_token_accuracy": 0.19815392941236495,
|
|
"num_tokens": 13407155.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.826538610458374,
|
|
"epoch": 6.304093567251462,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00021761131823895324,
|
|
"loss": 5.1961,
|
|
"mean_token_accuracy": 0.21220326125621797,
|
|
"num_tokens": 13419456.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.767988681793213,
|
|
"epoch": 6.309941520467836,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002171582076794088,
|
|
"loss": 5.1604,
|
|
"mean_token_accuracy": 0.21559286266565322,
|
|
"num_tokens": 13431653.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.901519632339477,
|
|
"epoch": 6.315789473684211,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002167053481573909,
|
|
"loss": 5.3145,
|
|
"mean_token_accuracy": 0.19939769953489303,
|
|
"num_tokens": 13444188.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.901363277435303,
|
|
"epoch": 6.321637426900585,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00021625274163834224,
|
|
"loss": 5.2439,
|
|
"mean_token_accuracy": 0.20560989379882813,
|
|
"num_tokens": 13455674.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.865947437286377,
|
|
"epoch": 6.3274853801169595,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00021580039008660762,
|
|
"loss": 5.2761,
|
|
"mean_token_accuracy": 0.20248084664344787,
|
|
"num_tokens": 13467415.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.868316316604615,
|
|
"epoch": 6.333333333333333,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00021534829546542518,
|
|
"loss": 5.243,
|
|
"mean_token_accuracy": 0.2154111847281456,
|
|
"num_tokens": 13480583.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.950709009170533,
|
|
"epoch": 6.3391812865497075,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002148964597369179,
|
|
"loss": 5.2793,
|
|
"mean_token_accuracy": 0.2050291046500206,
|
|
"num_tokens": 13492952.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.889143323898315,
|
|
"epoch": 6.345029239766082,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021444488486208545,
|
|
"loss": 5.38,
|
|
"mean_token_accuracy": 0.19696755707263947,
|
|
"num_tokens": 13505515.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.866916131973267,
|
|
"epoch": 6.350877192982456,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00021399357280079502,
|
|
"loss": 5.1936,
|
|
"mean_token_accuracy": 0.21434470713138581,
|
|
"num_tokens": 13518215.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.882256746292114,
|
|
"epoch": 6.356725146198831,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00021354252551177333,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.21647098809480667,
|
|
"num_tokens": 13530426.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.880256986618042,
|
|
"epoch": 6.362573099415204,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00021309174495259814,
|
|
"loss": 5.2876,
|
|
"mean_token_accuracy": 0.20116450339555741,
|
|
"num_tokens": 13543310.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.90527868270874,
|
|
"epoch": 6.368421052631579,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00021264123307968913,
|
|
"loss": 5.2898,
|
|
"mean_token_accuracy": 0.20875199884176254,
|
|
"num_tokens": 13555377.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.95016884803772,
|
|
"epoch": 6.374269005847953,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002121909918483002,
|
|
"loss": 5.3087,
|
|
"mean_token_accuracy": 0.2037610501050949,
|
|
"num_tokens": 13567246.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.934789943695068,
|
|
"epoch": 6.380116959064328,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00021174102321251082,
|
|
"loss": 5.3343,
|
|
"mean_token_accuracy": 0.20247435867786406,
|
|
"num_tokens": 13580073.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.939570713043213,
|
|
"epoch": 6.385964912280702,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002112913291252168,
|
|
"loss": 5.2609,
|
|
"mean_token_accuracy": 0.19968079924583435,
|
|
"num_tokens": 13593392.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.898805141448975,
|
|
"epoch": 6.391812865497076,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000210841911538123,
|
|
"loss": 5.2149,
|
|
"mean_token_accuracy": 0.20809216797351837,
|
|
"num_tokens": 13606530.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.872603464126587,
|
|
"epoch": 6.39766081871345,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00021039277240173372,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.20336396992206573,
|
|
"num_tokens": 13619879.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.890434741973877,
|
|
"epoch": 6.4035087719298245,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002099439136653452,
|
|
"loss": 5.2463,
|
|
"mean_token_accuracy": 0.20971034467220306,
|
|
"num_tokens": 13633516.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 6.0390478610992435,
|
|
"epoch": 6.409356725146199,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00020949533727703657,
|
|
"loss": 5.4093,
|
|
"mean_token_accuracy": 0.19657751619815828,
|
|
"num_tokens": 13646767.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.914198780059815,
|
|
"epoch": 6.415204678362573,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00020904704518366136,
|
|
"loss": 5.2746,
|
|
"mean_token_accuracy": 0.20094932317733766,
|
|
"num_tokens": 13657988.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.9170006275177,
|
|
"epoch": 6.421052631578947,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00020859903933083952,
|
|
"loss": 5.2857,
|
|
"mean_token_accuracy": 0.20173392146825792,
|
|
"num_tokens": 13670244.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.837920761108398,
|
|
"epoch": 6.426900584795321,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002081513216629486,
|
|
"loss": 5.2385,
|
|
"mean_token_accuracy": 0.20759102553129197,
|
|
"num_tokens": 13683035.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.8975193977355955,
|
|
"epoch": 6.432748538011696,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002077038941231153,
|
|
"loss": 5.2899,
|
|
"mean_token_accuracy": 0.20170291960239412,
|
|
"num_tokens": 13696528.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 6.432748538011696,
|
|
"eval_entropy": 5.893075248133116,
|
|
"eval_loss": 6.540669918060303,
|
|
"eval_mean_token_accuracy": 0.15225270315784972,
|
|
"eval_num_tokens": 13696528.0,
|
|
"eval_runtime": 2.2834,
|
|
"eval_samples_per_second": 1185.499,
|
|
"eval_steps_per_second": 148.461,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.895291566848755,
|
|
"epoch": 6.43859649122807,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00020725675865320721,
|
|
"loss": 5.3033,
|
|
"mean_token_accuracy": 0.20175388008356093,
|
|
"num_tokens": 13709596.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.9250648021698,
|
|
"epoch": 6.444444444444445,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00020680991719382446,
|
|
"loss": 5.2777,
|
|
"mean_token_accuracy": 0.208809557557106,
|
|
"num_tokens": 13722341.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.950402498245239,
|
|
"epoch": 6.450292397660819,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00020636337168429092,
|
|
"loss": 5.3513,
|
|
"mean_token_accuracy": 0.19857801645994186,
|
|
"num_tokens": 13734389.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.880336427688599,
|
|
"epoch": 6.456140350877193,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00020591712406264613,
|
|
"loss": 5.2604,
|
|
"mean_token_accuracy": 0.20674069821834565,
|
|
"num_tokens": 13746020.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.875683259963989,
|
|
"epoch": 6.461988304093567,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002054711762656369,
|
|
"loss": 5.3192,
|
|
"mean_token_accuracy": 0.20653996765613555,
|
|
"num_tokens": 13758057.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.885418701171875,
|
|
"epoch": 6.4678362573099415,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002050255302287084,
|
|
"loss": 5.2689,
|
|
"mean_token_accuracy": 0.205387482047081,
|
|
"num_tokens": 13771485.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.935099744796753,
|
|
"epoch": 6.473684210526316,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002045801878859967,
|
|
"loss": 5.3007,
|
|
"mean_token_accuracy": 0.2012038365006447,
|
|
"num_tokens": 13784546.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.937276029586792,
|
|
"epoch": 6.47953216374269,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00020413515117031917,
|
|
"loss": 5.3122,
|
|
"mean_token_accuracy": 0.1989278644323349,
|
|
"num_tokens": 13797014.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.911808824539184,
|
|
"epoch": 6.485380116959064,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00020369042201316722,
|
|
"loss": 5.2805,
|
|
"mean_token_accuracy": 0.20476778447628022,
|
|
"num_tokens": 13810068.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.860673713684082,
|
|
"epoch": 6.491228070175438,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00020324600234469743,
|
|
"loss": 5.2905,
|
|
"mean_token_accuracy": 0.20078159868717194,
|
|
"num_tokens": 13822699.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.856434106826782,
|
|
"epoch": 6.497076023391813,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00020280189409372274,
|
|
"loss": 5.2464,
|
|
"mean_token_accuracy": 0.20916291326284409,
|
|
"num_tokens": 13835345.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.961838293075561,
|
|
"epoch": 6.502923976608187,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020235809918770504,
|
|
"loss": 5.3668,
|
|
"mean_token_accuracy": 0.19513495415449142,
|
|
"num_tokens": 13848282.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.880169534683228,
|
|
"epoch": 6.508771929824562,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00020191461955274597,
|
|
"loss": 5.2464,
|
|
"mean_token_accuracy": 0.21040370762348176,
|
|
"num_tokens": 13861665.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.93856110572815,
|
|
"epoch": 6.514619883040936,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002014714571135789,
|
|
"loss": 5.3066,
|
|
"mean_token_accuracy": 0.19565294235944747,
|
|
"num_tokens": 13874073.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.9494531631469725,
|
|
"epoch": 6.52046783625731,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00020102861379356064,
|
|
"loss": 5.2837,
|
|
"mean_token_accuracy": 0.1988810420036316,
|
|
"num_tokens": 13887305.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.916756200790405,
|
|
"epoch": 6.526315789473684,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000200586091514663,
|
|
"loss": 5.3784,
|
|
"mean_token_accuracy": 0.19814997166395187,
|
|
"num_tokens": 13900989.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.9187170505523685,
|
|
"epoch": 6.5321637426900585,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00020014389219746438,
|
|
"loss": 5.3579,
|
|
"mean_token_accuracy": 0.20408453047275543,
|
|
"num_tokens": 13913189.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.91787371635437,
|
|
"epoch": 6.538011695906433,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001997020177611414,
|
|
"loss": 5.2556,
|
|
"mean_token_accuracy": 0.20817144960165024,
|
|
"num_tokens": 13924826.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.816835975646972,
|
|
"epoch": 6.543859649122807,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019926047012346098,
|
|
"loss": 5.2107,
|
|
"mean_token_accuracy": 0.21737461984157563,
|
|
"num_tokens": 13937816.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.880726766586304,
|
|
"epoch": 6.549707602339181,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00019881925120077143,
|
|
"loss": 5.302,
|
|
"mean_token_accuracy": 0.199569408595562,
|
|
"num_tokens": 13949755.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.851153898239136,
|
|
"epoch": 6.555555555555555,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019837836290799454,
|
|
"loss": 5.1376,
|
|
"mean_token_accuracy": 0.21568706035614013,
|
|
"num_tokens": 13962281.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.985859394073486,
|
|
"epoch": 6.56140350877193,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001979378071586171,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.19286799132823945,
|
|
"num_tokens": 13976162.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.782685232162476,
|
|
"epoch": 6.567251461988304,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001974975858646827,
|
|
"loss": 5.1112,
|
|
"mean_token_accuracy": 0.22031393349170686,
|
|
"num_tokens": 13988185.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.9385453224182125,
|
|
"epoch": 6.573099415204679,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019705770093678333,
|
|
"loss": 5.2879,
|
|
"mean_token_accuracy": 0.19739507734775544,
|
|
"num_tokens": 14000489.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.952527570724487,
|
|
"epoch": 6.578947368421053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019661815428405094,
|
|
"loss": 5.3157,
|
|
"mean_token_accuracy": 0.2126591607928276,
|
|
"num_tokens": 14012806.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.867352867126465,
|
|
"epoch": 6.584795321637427,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019617894781414968,
|
|
"loss": 5.2335,
|
|
"mean_token_accuracy": 0.20621028542518616,
|
|
"num_tokens": 14024364.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.795357370376587,
|
|
"epoch": 6.590643274853801,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019574008343326726,
|
|
"loss": 5.1902,
|
|
"mean_token_accuracy": 0.2127966746687889,
|
|
"num_tokens": 14036933.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.856552219390869,
|
|
"epoch": 6.5964912280701755,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00019530156304610623,
|
|
"loss": 5.1639,
|
|
"mean_token_accuracy": 0.2132712721824646,
|
|
"num_tokens": 14048971.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.923597764968872,
|
|
"epoch": 6.60233918128655,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00019486338855587683,
|
|
"loss": 5.3446,
|
|
"mean_token_accuracy": 0.19792177379131318,
|
|
"num_tokens": 14060769.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.764224052429199,
|
|
"epoch": 6.6081871345029235,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019442556186428762,
|
|
"loss": 5.1014,
|
|
"mean_token_accuracy": 0.2157956212759018,
|
|
"num_tokens": 14072432.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.790011787414551,
|
|
"epoch": 6.614035087719298,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001939880848715378,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.22169077396392822,
|
|
"num_tokens": 14084768.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.918018960952759,
|
|
"epoch": 6.619883040935672,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00019355095947630897,
|
|
"loss": 5.2867,
|
|
"mean_token_accuracy": 0.20190161913633348,
|
|
"num_tokens": 14096814.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.9388690948486325,
|
|
"epoch": 6.625730994152047,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019311418757575672,
|
|
"loss": 5.337,
|
|
"mean_token_accuracy": 0.19692451059818267,
|
|
"num_tokens": 14110782.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.834782600402832,
|
|
"epoch": 6.631578947368421,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00019267777106550229,
|
|
"loss": 5.2357,
|
|
"mean_token_accuracy": 0.20637393295764922,
|
|
"num_tokens": 14123154.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.895966005325318,
|
|
"epoch": 6.637426900584796,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00019224171183962474,
|
|
"loss": 5.3215,
|
|
"mean_token_accuracy": 0.20241911858320236,
|
|
"num_tokens": 14137182.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.942614984512329,
|
|
"epoch": 6.643274853801169,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00019180601179065232,
|
|
"loss": 5.3006,
|
|
"mean_token_accuracy": 0.20410328209400178,
|
|
"num_tokens": 14149560.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.90175142288208,
|
|
"epoch": 6.649122807017544,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001913706728095545,
|
|
"loss": 5.2894,
|
|
"mean_token_accuracy": 0.20331071615219115,
|
|
"num_tokens": 14162402.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.8765514373779295,
|
|
"epoch": 6.654970760233918,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019093569678573375,
|
|
"loss": 5.3282,
|
|
"mean_token_accuracy": 0.20350528061389922,
|
|
"num_tokens": 14174907.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.947376775741577,
|
|
"epoch": 6.6608187134502925,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019050108560701706,
|
|
"loss": 5.3348,
|
|
"mean_token_accuracy": 0.20112716406583786,
|
|
"num_tokens": 14187349.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.970543718338012,
|
|
"epoch": 6.666666666666667,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00019006684115964828,
|
|
"loss": 5.3657,
|
|
"mean_token_accuracy": 0.19752688854932784,
|
|
"num_tokens": 14198839.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.795181703567505,
|
|
"epoch": 6.6725146198830405,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00018963296532827935,
|
|
"loss": 5.3088,
|
|
"mean_token_accuracy": 0.19839034229516983,
|
|
"num_tokens": 14210879.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.857154655456543,
|
|
"epoch": 6.678362573099415,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018919945999596244,
|
|
"loss": 5.2745,
|
|
"mean_token_accuracy": 0.2048317402601242,
|
|
"num_tokens": 14224167.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.973622369766235,
|
|
"epoch": 6.684210526315789,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001887663270441418,
|
|
"loss": 5.2833,
|
|
"mean_token_accuracy": 0.20041715502738952,
|
|
"num_tokens": 14236735.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.967438220977783,
|
|
"epoch": 6.690058479532164,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00018833356835264576,
|
|
"loss": 5.3058,
|
|
"mean_token_accuracy": 0.20106961727142333,
|
|
"num_tokens": 14249507.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.811670112609863,
|
|
"epoch": 6.695906432748538,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00018790118579967762,
|
|
"loss": 5.2194,
|
|
"mean_token_accuracy": 0.20451067239046097,
|
|
"num_tokens": 14261626.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.912743139266968,
|
|
"epoch": 6.701754385964913,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000187469181261809,
|
|
"loss": 5.2817,
|
|
"mean_token_accuracy": 0.20487589836120607,
|
|
"num_tokens": 14274031.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.92514214515686,
|
|
"epoch": 6.707602339181286,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001870375566139706,
|
|
"loss": 5.2789,
|
|
"mean_token_accuracy": 0.19910192340612412,
|
|
"num_tokens": 14286014.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.8768819808959964,
|
|
"epoch": 6.713450292397661,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00018660631372944403,
|
|
"loss": 5.2993,
|
|
"mean_token_accuracy": 0.20433387011289597,
|
|
"num_tokens": 14297526.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.859215164184571,
|
|
"epoch": 6.719298245614035,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00018617545447985461,
|
|
"loss": 5.2306,
|
|
"mean_token_accuracy": 0.20605453848838806,
|
|
"num_tokens": 14310531.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.9304108142852785,
|
|
"epoch": 6.7251461988304095,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00018574498073516227,
|
|
"loss": 5.2661,
|
|
"mean_token_accuracy": 0.2013670861721039,
|
|
"num_tokens": 14322820.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.90087399482727,
|
|
"epoch": 6.730994152046784,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018531489436365395,
|
|
"loss": 5.3154,
|
|
"mean_token_accuracy": 0.20114836245775222,
|
|
"num_tokens": 14335778.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.836156034469605,
|
|
"epoch": 6.7368421052631575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018488519723193543,
|
|
"loss": 5.3161,
|
|
"mean_token_accuracy": 0.20043632090091706,
|
|
"num_tokens": 14347535.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.930212354660034,
|
|
"epoch": 6.742690058479532,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001844558912049229,
|
|
"loss": 5.3058,
|
|
"mean_token_accuracy": 0.20272140204906464,
|
|
"num_tokens": 14360094.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.985506820678711,
|
|
"epoch": 6.748538011695906,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018402697814583535,
|
|
"loss": 5.2579,
|
|
"mean_token_accuracy": 0.202339506149292,
|
|
"num_tokens": 14371040.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.851832628250122,
|
|
"epoch": 6.754385964912281,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001835984599161864,
|
|
"loss": 5.2732,
|
|
"mean_token_accuracy": 0.20662306398153304,
|
|
"num_tokens": 14382682.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.790254020690918,
|
|
"epoch": 6.760233918128655,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00018317033837577564,
|
|
"loss": 5.1802,
|
|
"mean_token_accuracy": 0.22153317779302598,
|
|
"num_tokens": 14395992.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.9642270565032955,
|
|
"epoch": 6.76608187134503,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001827426153826814,
|
|
"loss": 5.3632,
|
|
"mean_token_accuracy": 0.18804966956377028,
|
|
"num_tokens": 14407862.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.912167263031006,
|
|
"epoch": 6.771929824561403,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001823152927932521,
|
|
"loss": 5.2885,
|
|
"mean_token_accuracy": 0.2038208544254303,
|
|
"num_tokens": 14419855.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.925345325469971,
|
|
"epoch": 6.777777777777778,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001818883724620985,
|
|
"loss": 5.3263,
|
|
"mean_token_accuracy": 0.20725452452898024,
|
|
"num_tokens": 14432837.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.831614685058594,
|
|
"epoch": 6.783625730994152,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00018146185624208527,
|
|
"loss": 5.2183,
|
|
"mean_token_accuracy": 0.21333656460046768,
|
|
"num_tokens": 14444295.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.9824628829956055,
|
|
"epoch": 6.7894736842105265,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00018103574598432365,
|
|
"loss": 5.4436,
|
|
"mean_token_accuracy": 0.1923542007803917,
|
|
"num_tokens": 14458845.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.88775668144226,
|
|
"epoch": 6.7953216374269,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00018061004353816246,
|
|
"loss": 5.2469,
|
|
"mean_token_accuracy": 0.20238619893789292,
|
|
"num_tokens": 14470583.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.922159624099732,
|
|
"epoch": 6.8011695906432745,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000180184750751181,
|
|
"loss": 5.3163,
|
|
"mean_token_accuracy": 0.20422874987125397,
|
|
"num_tokens": 14483322.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.769148063659668,
|
|
"epoch": 6.807017543859649,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001797598694691804,
|
|
"loss": 5.1409,
|
|
"mean_token_accuracy": 0.21671497523784639,
|
|
"num_tokens": 14496199.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.949480628967285,
|
|
"epoch": 6.812865497076023,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00017933540153617579,
|
|
"loss": 5.422,
|
|
"mean_token_accuracy": 0.1976809710264206,
|
|
"num_tokens": 14508784.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.902020454406738,
|
|
"epoch": 6.818713450292398,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001789113487943888,
|
|
"loss": 5.3008,
|
|
"mean_token_accuracy": 0.19925431907176971,
|
|
"num_tokens": 14521386.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.968377113342285,
|
|
"epoch": 6.824561403508772,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00017848771308423835,
|
|
"loss": 5.3881,
|
|
"mean_token_accuracy": 0.19098925739526748,
|
|
"num_tokens": 14533839.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.913421201705932,
|
|
"epoch": 6.830409356725146,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00017806449624433384,
|
|
"loss": 5.2744,
|
|
"mean_token_accuracy": 0.20087118148803712,
|
|
"num_tokens": 14545713.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.923596668243408,
|
|
"epoch": 6.83625730994152,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000177641700111467,
|
|
"loss": 5.3347,
|
|
"mean_token_accuracy": 0.19933082163333893,
|
|
"num_tokens": 14558500.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.9302808284759525,
|
|
"epoch": 6.842105263157895,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000177219326520603,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.20167507976293564,
|
|
"num_tokens": 14571137.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.903340148925781,
|
|
"epoch": 6.847953216374269,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017679737730487378,
|
|
"loss": 5.2798,
|
|
"mean_token_accuracy": 0.2084101065993309,
|
|
"num_tokens": 14583251.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.900283527374268,
|
|
"epoch": 6.853801169590644,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017637585429556912,
|
|
"loss": 5.3197,
|
|
"mean_token_accuracy": 0.2054581895470619,
|
|
"num_tokens": 14594874.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.891225051879883,
|
|
"epoch": 6.859649122807017,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017595475932212908,
|
|
"loss": 5.25,
|
|
"mean_token_accuracy": 0.20416252315044403,
|
|
"num_tokens": 14606564.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.802581405639648,
|
|
"epoch": 6.8654970760233915,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017553409421213606,
|
|
"loss": 5.1711,
|
|
"mean_token_accuracy": 0.21535236090421678,
|
|
"num_tokens": 14619228.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.877661037445068,
|
|
"epoch": 6.871345029239766,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00017511386079130666,
|
|
"loss": 5.2809,
|
|
"mean_token_accuracy": 0.204811729490757,
|
|
"num_tokens": 14631993.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.893503856658936,
|
|
"epoch": 6.87719298245614,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00017469406088348427,
|
|
"loss": 5.2995,
|
|
"mean_token_accuracy": 0.20578158646821976,
|
|
"num_tokens": 14645494.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.831784200668335,
|
|
"epoch": 6.883040935672515,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00017427469631063055,
|
|
"loss": 5.2186,
|
|
"mean_token_accuracy": 0.20952749848365784,
|
|
"num_tokens": 14658019.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.872379493713379,
|
|
"epoch": 6.888888888888889,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00017385576889281774,
|
|
"loss": 5.2527,
|
|
"mean_token_accuracy": 0.21388380378484725,
|
|
"num_tokens": 14671031.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.912283515930175,
|
|
"epoch": 6.894736842105263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00017343728044822093,
|
|
"loss": 5.3527,
|
|
"mean_token_accuracy": 0.2019163876771927,
|
|
"num_tokens": 14683819.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.827155113220215,
|
|
"epoch": 6.900584795321637,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017301923279310996,
|
|
"loss": 5.2548,
|
|
"mean_token_accuracy": 0.21541957855224608,
|
|
"num_tokens": 14697477.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.814873123168946,
|
|
"epoch": 6.906432748538012,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00017260162774184167,
|
|
"loss": 5.2352,
|
|
"mean_token_accuracy": 0.209805665910244,
|
|
"num_tokens": 14709964.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.85977725982666,
|
|
"epoch": 6.912280701754386,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00017218446710685176,
|
|
"loss": 5.2243,
|
|
"mean_token_accuracy": 0.21051897704601288,
|
|
"num_tokens": 14722021.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.866635942459107,
|
|
"epoch": 6.918128654970761,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017176775269864763,
|
|
"loss": 5.2138,
|
|
"mean_token_accuracy": 0.21107935905456543,
|
|
"num_tokens": 14734578.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.868919944763183,
|
|
"epoch": 6.923976608187134,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001713514863257994,
|
|
"loss": 5.2138,
|
|
"mean_token_accuracy": 0.20756827145814896,
|
|
"num_tokens": 14746602.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.834951257705688,
|
|
"epoch": 6.9298245614035086,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00017093566979493302,
|
|
"loss": 5.2579,
|
|
"mean_token_accuracy": 0.20670852661132813,
|
|
"num_tokens": 14759802.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 6.011142921447754,
|
|
"epoch": 6.935672514619883,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0001705203049107223,
|
|
"loss": 5.3484,
|
|
"mean_token_accuracy": 0.19380707889795304,
|
|
"num_tokens": 14772559.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.910616397857666,
|
|
"epoch": 6.941520467836257,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017010539347588023,
|
|
"loss": 5.3095,
|
|
"mean_token_accuracy": 0.19598183333873748,
|
|
"num_tokens": 14783993.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.89832124710083,
|
|
"epoch": 6.947368421052632,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016969093729115253,
|
|
"loss": 5.3196,
|
|
"mean_token_accuracy": 0.20121149271726607,
|
|
"num_tokens": 14796087.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.964419603347778,
|
|
"epoch": 6.953216374269006,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016927693815530854,
|
|
"loss": 5.3629,
|
|
"mean_token_accuracy": 0.2021169811487198,
|
|
"num_tokens": 14807782.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.877931690216064,
|
|
"epoch": 6.95906432748538,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016886339786513428,
|
|
"loss": 5.3009,
|
|
"mean_token_accuracy": 0.20411295294761658,
|
|
"num_tokens": 14820909.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.878330135345459,
|
|
"epoch": 6.964912280701754,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001684503182154244,
|
|
"loss": 5.2561,
|
|
"mean_token_accuracy": 0.2068962648510933,
|
|
"num_tokens": 14833151.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.875535774230957,
|
|
"epoch": 6.970760233918129,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016803770099897385,
|
|
"loss": 5.2647,
|
|
"mean_token_accuracy": 0.21388636529445648,
|
|
"num_tokens": 14844783.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.851003360748291,
|
|
"epoch": 6.976608187134503,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016762554800657132,
|
|
"loss": 5.2155,
|
|
"mean_token_accuracy": 0.21056125015020372,
|
|
"num_tokens": 14856635.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.906779766082764,
|
|
"epoch": 6.982456140350877,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001672138610269901,
|
|
"loss": 5.2734,
|
|
"mean_token_accuracy": 0.20429999381303787,
|
|
"num_tokens": 14868513.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.893103218078613,
|
|
"epoch": 6.988304093567251,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016680264184698135,
|
|
"loss": 5.3152,
|
|
"mean_token_accuracy": 0.20282178074121476,
|
|
"num_tokens": 14880899.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.865803956985474,
|
|
"epoch": 6.994152046783626,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016639189225126574,
|
|
"loss": 5.1948,
|
|
"mean_token_accuracy": 0.20718448907136916,
|
|
"num_tokens": 14894131.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.8531835079193115,
|
|
"epoch": 7.0,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00016598161402252596,
|
|
"loss": 5.2542,
|
|
"mean_token_accuracy": 0.21230687350034713,
|
|
"num_tokens": 14906388.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.880353546142578,
|
|
"epoch": 7.005847953216374,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00016557180894139888,
|
|
"loss": 5.0652,
|
|
"mean_token_accuracy": 0.2162859708070755,
|
|
"num_tokens": 14918621.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.837734127044678,
|
|
"epoch": 7.011695906432749,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016516247878646805,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.21722232848405837,
|
|
"num_tokens": 14930757.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.792285442352295,
|
|
"epoch": 7.017543859649122,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016475362533425574,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.21750809103250504,
|
|
"num_tokens": 14943264.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 7.017543859649122,
|
|
"eval_entropy": 5.896686929517088,
|
|
"eval_loss": 6.52704381942749,
|
|
"eval_mean_token_accuracy": 0.15279760371047488,
|
|
"eval_num_tokens": 14943264.0,
|
|
"eval_runtime": 2.2852,
|
|
"eval_samples_per_second": 1184.579,
|
|
"eval_steps_per_second": 148.346,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.888935947418213,
|
|
"epoch": 7.023391812865497,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016434525035921517,
|
|
"loss": 5.1067,
|
|
"mean_token_accuracy": 0.21514792740345,
|
|
"num_tokens": 14954097.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.884676694869995,
|
|
"epoch": 7.029239766081871,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00016393735563372292,
|
|
"loss": 5.1073,
|
|
"mean_token_accuracy": 0.21550192236900328,
|
|
"num_tokens": 14967188.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.917854261398316,
|
|
"epoch": 7.035087719298246,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001635299429280715,
|
|
"loss": 5.1197,
|
|
"mean_token_accuracy": 0.21152072846889497,
|
|
"num_tokens": 14979698.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.908101177215576,
|
|
"epoch": 7.04093567251462,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001631230140104611,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.2121755599975586,
|
|
"num_tokens": 14992649.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.923228359222412,
|
|
"epoch": 7.046783625730994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001627165706469927,
|
|
"loss": 5.1563,
|
|
"mean_token_accuracy": 0.21694395393133165,
|
|
"num_tokens": 15006284.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.872251653671265,
|
|
"epoch": 7.052631578947368,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016231061460165926,
|
|
"loss": 5.1538,
|
|
"mean_token_accuracy": 0.21399132758378983,
|
|
"num_tokens": 15018017.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.867130088806152,
|
|
"epoch": 7.058479532163743,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016190514763633946,
|
|
"loss": 5.1539,
|
|
"mean_token_accuracy": 0.2111828252673149,
|
|
"num_tokens": 15029177.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.870800542831421,
|
|
"epoch": 7.064327485380117,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016150017151078894,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.21520036906003953,
|
|
"num_tokens": 15041720.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.933644247055054,
|
|
"epoch": 7.0701754385964914,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016109568798263285,
|
|
"loss": 5.2182,
|
|
"mean_token_accuracy": 0.20751896649599075,
|
|
"num_tokens": 15053847.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.889568710327149,
|
|
"epoch": 7.076023391812866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000160691698807359,
|
|
"loss": 5.0596,
|
|
"mean_token_accuracy": 0.22248515337705613,
|
|
"num_tokens": 15066211.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.867597436904907,
|
|
"epoch": 7.081871345029239,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016028820573830936,
|
|
"loss": 5.1463,
|
|
"mean_token_accuracy": 0.2143215224146843,
|
|
"num_tokens": 15077713.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.861376333236694,
|
|
"epoch": 7.087719298245614,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00015988521052667266,
|
|
"loss": 5.1481,
|
|
"mean_token_accuracy": 0.21356315165758133,
|
|
"num_tokens": 15090196.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.85325984954834,
|
|
"epoch": 7.093567251461988,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00015948271492147707,
|
|
"loss": 5.1627,
|
|
"mean_token_accuracy": 0.21280159503221513,
|
|
"num_tokens": 15101975.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.966574430465698,
|
|
"epoch": 7.099415204678363,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001590807206695824,
|
|
"loss": 5.1551,
|
|
"mean_token_accuracy": 0.20622991025447845,
|
|
"num_tokens": 15113594.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.858206653594971,
|
|
"epoch": 7.105263157894737,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015867922951567253,
|
|
"loss": 5.0872,
|
|
"mean_token_accuracy": 0.22163581103086472,
|
|
"num_tokens": 15126260.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.891089296340942,
|
|
"epoch": 7.111111111111111,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015827824320224781,
|
|
"loss": 5.1613,
|
|
"mean_token_accuracy": 0.20995641648769378,
|
|
"num_tokens": 15138287.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.928001356124878,
|
|
"epoch": 7.116959064327485,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00015787776346961764,
|
|
"loss": 5.1665,
|
|
"mean_token_accuracy": 0.20894373804330826,
|
|
"num_tokens": 15150175.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.879667615890503,
|
|
"epoch": 7.12280701754386,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00015747779205589269,
|
|
"loss": 5.1833,
|
|
"mean_token_accuracy": 0.20957615226507187,
|
|
"num_tokens": 15161953.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.900380659103393,
|
|
"epoch": 7.128654970760234,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001570783306969778,
|
|
"loss": 5.1107,
|
|
"mean_token_accuracy": 0.21315825879573821,
|
|
"num_tokens": 15174950.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.940548944473266,
|
|
"epoch": 7.1345029239766085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00015667938112656372,
|
|
"loss": 5.1792,
|
|
"mean_token_accuracy": 0.21149321496486664,
|
|
"num_tokens": 15188450.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.888680505752563,
|
|
"epoch": 7.140350877192983,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001562809450761201,
|
|
"loss": 5.1268,
|
|
"mean_token_accuracy": 0.2108440101146698,
|
|
"num_tokens": 15200493.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.824927854537964,
|
|
"epoch": 7.146198830409356,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015588302427488833,
|
|
"loss": 5.0966,
|
|
"mean_token_accuracy": 0.2165909931063652,
|
|
"num_tokens": 15212465.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.852302074432373,
|
|
"epoch": 7.152046783625731,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001554856204498728,
|
|
"loss": 5.0856,
|
|
"mean_token_accuracy": 0.21985691487789155,
|
|
"num_tokens": 15224628.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.898373746871949,
|
|
"epoch": 7.157894736842105,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001550887353258349,
|
|
"loss": 5.1819,
|
|
"mean_token_accuracy": 0.210515196621418,
|
|
"num_tokens": 15236325.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.856766080856323,
|
|
"epoch": 7.16374269005848,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00015469237062528445,
|
|
"loss": 5.1516,
|
|
"mean_token_accuracy": 0.2195606991648674,
|
|
"num_tokens": 15248969.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.890830230712891,
|
|
"epoch": 7.169590643274854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00015429652806847266,
|
|
"loss": 5.1493,
|
|
"mean_token_accuracy": 0.20845075845718383,
|
|
"num_tokens": 15262360.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.930358600616455,
|
|
"epoch": 7.175438596491228,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015390120937338459,
|
|
"loss": 5.1675,
|
|
"mean_token_accuracy": 0.20810847282409667,
|
|
"num_tokens": 15275429.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.869526433944702,
|
|
"epoch": 7.181286549707602,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001535064162557318,
|
|
"loss": 5.1117,
|
|
"mean_token_accuracy": 0.21576884835958482,
|
|
"num_tokens": 15287768.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.86055064201355,
|
|
"epoch": 7.187134502923977,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00015311215042894462,
|
|
"loss": 5.1003,
|
|
"mean_token_accuracy": 0.21992170661687852,
|
|
"num_tokens": 15300215.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.820063972473145,
|
|
"epoch": 7.192982456140351,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00015271841360416506,
|
|
"loss": 5.0378,
|
|
"mean_token_accuracy": 0.22423748821020126,
|
|
"num_tokens": 15311297.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.863794326782227,
|
|
"epoch": 7.1988304093567255,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00015232520749023924,
|
|
"loss": 5.0972,
|
|
"mean_token_accuracy": 0.22574346214532853,
|
|
"num_tokens": 15323226.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.869160890579224,
|
|
"epoch": 7.204678362573099,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001519325337937098,
|
|
"loss": 5.0734,
|
|
"mean_token_accuracy": 0.2184156820178032,
|
|
"num_tokens": 15336546.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.827308320999146,
|
|
"epoch": 7.2105263157894735,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001515403942188088,
|
|
"loss": 5.0808,
|
|
"mean_token_accuracy": 0.22181283831596374,
|
|
"num_tokens": 15350502.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.835464763641357,
|
|
"epoch": 7.216374269005848,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015114879046745005,
|
|
"loss": 5.0798,
|
|
"mean_token_accuracy": 0.2204731211066246,
|
|
"num_tokens": 15361579.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.888103246688843,
|
|
"epoch": 7.222222222222222,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001507577242392219,
|
|
"loss": 5.1447,
|
|
"mean_token_accuracy": 0.21713362485170365,
|
|
"num_tokens": 15376198.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.900730562210083,
|
|
"epoch": 7.228070175438597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00015036719723138006,
|
|
"loss": 5.1736,
|
|
"mean_token_accuracy": 0.2101849779486656,
|
|
"num_tokens": 15389259.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.923346662521363,
|
|
"epoch": 7.23391812865497,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00014997721113883946,
|
|
"loss": 5.2352,
|
|
"mean_token_accuracy": 0.20305581986904145,
|
|
"num_tokens": 15401568.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.883099794387817,
|
|
"epoch": 7.239766081871345,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00014958776765416776,
|
|
"loss": 5.1652,
|
|
"mean_token_accuracy": 0.21461765021085738,
|
|
"num_tokens": 15413606.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.88554892539978,
|
|
"epoch": 7.245614035087719,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001491988684675778,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.21019780039787292,
|
|
"num_tokens": 15425016.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.8895481586456295,
|
|
"epoch": 7.251461988304094,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00014881051526691963,
|
|
"loss": 5.157,
|
|
"mean_token_accuracy": 0.20731311291456223,
|
|
"num_tokens": 15438321.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.903854417800903,
|
|
"epoch": 7.257309941520468,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00014842270973767436,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.21231588423252107,
|
|
"num_tokens": 15450500.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.850664377212524,
|
|
"epoch": 7.2631578947368425,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001480354535629454,
|
|
"loss": 5.1163,
|
|
"mean_token_accuracy": 0.2162235051393509,
|
|
"num_tokens": 15462171.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.823956680297852,
|
|
"epoch": 7.269005847953216,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001476487484234526,
|
|
"loss": 5.0423,
|
|
"mean_token_accuracy": 0.21934671252965926,
|
|
"num_tokens": 15475183.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.870025873184204,
|
|
"epoch": 7.2748538011695905,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000147262595997524,
|
|
"loss": 5.1264,
|
|
"mean_token_accuracy": 0.2172160640358925,
|
|
"num_tokens": 15486441.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.883719110488892,
|
|
"epoch": 7.280701754385965,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00014687699796108854,
|
|
"loss": 5.1663,
|
|
"mean_token_accuracy": 0.20960905253887177,
|
|
"num_tokens": 15499421.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.876193332672119,
|
|
"epoch": 7.286549707602339,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00014649195598766967,
|
|
"loss": 5.0659,
|
|
"mean_token_accuracy": 0.2214240089058876,
|
|
"num_tokens": 15512316.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.892332887649536,
|
|
"epoch": 7.292397660818714,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00014610747174837702,
|
|
"loss": 5.1639,
|
|
"mean_token_accuracy": 0.21489033699035645,
|
|
"num_tokens": 15524961.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.858953189849854,
|
|
"epoch": 7.298245614035087,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00014572354691189977,
|
|
"loss": 5.1023,
|
|
"mean_token_accuracy": 0.21444770246744155,
|
|
"num_tokens": 15536633.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.870029449462891,
|
|
"epoch": 7.304093567251462,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001453401831444992,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.214753457903862,
|
|
"num_tokens": 15549283.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.934257984161377,
|
|
"epoch": 7.309941520467836,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014495738211000165,
|
|
"loss": 5.1867,
|
|
"mean_token_accuracy": 0.20582164376974105,
|
|
"num_tokens": 15560751.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.8405389308929445,
|
|
"epoch": 7.315789473684211,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000144575145469791,
|
|
"loss": 5.1304,
|
|
"mean_token_accuracy": 0.2162856787443161,
|
|
"num_tokens": 15572795.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.8723958969116214,
|
|
"epoch": 7.321637426900585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001441934748828016,
|
|
"loss": 5.1624,
|
|
"mean_token_accuracy": 0.21612410396337509,
|
|
"num_tokens": 15584666.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.860116672515869,
|
|
"epoch": 7.3274853801169595,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00014381237200551138,
|
|
"loss": 5.0689,
|
|
"mean_token_accuracy": 0.22249128967523574,
|
|
"num_tokens": 15597349.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.880211877822876,
|
|
"epoch": 7.333333333333333,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00014343183849193402,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.20980648845434188,
|
|
"num_tokens": 15609945.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.857202100753784,
|
|
"epoch": 7.3391812865497075,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00014305187599361223,
|
|
"loss": 5.1411,
|
|
"mean_token_accuracy": 0.2162948414683342,
|
|
"num_tokens": 15623167.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.907794713973999,
|
|
"epoch": 7.345029239766082,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00014267248615961047,
|
|
"loss": 5.1576,
|
|
"mean_token_accuracy": 0.21542315632104875,
|
|
"num_tokens": 15634751.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.831524896621704,
|
|
"epoch": 7.350877192982456,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00014229367063650788,
|
|
"loss": 5.0576,
|
|
"mean_token_accuracy": 0.22013792693614959,
|
|
"num_tokens": 15646857.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.862893295288086,
|
|
"epoch": 7.356725146198831,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00014191543106839116,
|
|
"loss": 5.0566,
|
|
"mean_token_accuracy": 0.21519360989332198,
|
|
"num_tokens": 15658612.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.828307867050171,
|
|
"epoch": 7.362573099415204,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00014153776909684685,
|
|
"loss": 5.1392,
|
|
"mean_token_accuracy": 0.21256667822599412,
|
|
"num_tokens": 15670330.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.838017845153809,
|
|
"epoch": 7.368421052631579,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00014116068636095522,
|
|
"loss": 5.1378,
|
|
"mean_token_accuracy": 0.21530213654041291,
|
|
"num_tokens": 15682536.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.824683570861817,
|
|
"epoch": 7.374269005847953,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00014078418449728236,
|
|
"loss": 4.9984,
|
|
"mean_token_accuracy": 0.2226453721523285,
|
|
"num_tokens": 15694643.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.8506995677948,
|
|
"epoch": 7.380116959064328,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00014040826513987318,
|
|
"loss": 5.1637,
|
|
"mean_token_accuracy": 0.21298383921384811,
|
|
"num_tokens": 15707005.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.856111001968384,
|
|
"epoch": 7.385964912280702,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001400329299202448,
|
|
"loss": 5.0857,
|
|
"mean_token_accuracy": 0.2217256173491478,
|
|
"num_tokens": 15719540.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.84460563659668,
|
|
"epoch": 7.391812865497076,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000139658180467379,
|
|
"loss": 5.0673,
|
|
"mean_token_accuracy": 0.21891699582338334,
|
|
"num_tokens": 15733722.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.902045631408692,
|
|
"epoch": 7.39766081871345,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001392840184077152,
|
|
"loss": 5.1511,
|
|
"mean_token_accuracy": 0.20928291231393814,
|
|
"num_tokens": 15745798.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.883737850189209,
|
|
"epoch": 7.4035087719298245,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00013891044536514367,
|
|
"loss": 5.1108,
|
|
"mean_token_accuracy": 0.21429356038570405,
|
|
"num_tokens": 15757705.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.8871925354003904,
|
|
"epoch": 7.409356725146199,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00013853746296099812,
|
|
"loss": 5.1939,
|
|
"mean_token_accuracy": 0.2060455784201622,
|
|
"num_tokens": 15771215.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.897227764129639,
|
|
"epoch": 7.415204678362573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00013816507281404894,
|
|
"loss": 5.1955,
|
|
"mean_token_accuracy": 0.21535377204418182,
|
|
"num_tokens": 15784540.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.913020849227905,
|
|
"epoch": 7.421052631578947,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00013779327654049606,
|
|
"loss": 5.1862,
|
|
"mean_token_accuracy": 0.2083567500114441,
|
|
"num_tokens": 15796892.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.8245758533477785,
|
|
"epoch": 7.426900584795321,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000137422075753962,
|
|
"loss": 5.1456,
|
|
"mean_token_accuracy": 0.22150931060314177,
|
|
"num_tokens": 15809587.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.843892765045166,
|
|
"epoch": 7.432748538011696,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001370514720654848,
|
|
"loss": 5.1225,
|
|
"mean_token_accuracy": 0.2093445912003517,
|
|
"num_tokens": 15821048.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.915178918838501,
|
|
"epoch": 7.43859649122807,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013668146708351104,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.21172926872968673,
|
|
"num_tokens": 15834358.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.768803977966309,
|
|
"epoch": 7.444444444444445,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013631206241388881,
|
|
"loss": 5.0651,
|
|
"mean_token_accuracy": 0.22636858820915223,
|
|
"num_tokens": 15846852.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.8744724750518795,
|
|
"epoch": 7.450292397660819,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013594325965986088,
|
|
"loss": 5.1478,
|
|
"mean_token_accuracy": 0.2131851077079773,
|
|
"num_tokens": 15859242.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.9372169971466064,
|
|
"epoch": 7.456140350877193,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00013557506042205782,
|
|
"loss": 5.1803,
|
|
"mean_token_accuracy": 0.21097951680421828,
|
|
"num_tokens": 15873524.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.885216951370239,
|
|
"epoch": 7.461988304093567,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00013520746629849044,
|
|
"loss": 5.1492,
|
|
"mean_token_accuracy": 0.2190225675702095,
|
|
"num_tokens": 15886641.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.837817144393921,
|
|
"epoch": 7.4678362573099415,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00013484047888454387,
|
|
"loss": 5.0797,
|
|
"mean_token_accuracy": 0.222073557972908,
|
|
"num_tokens": 15899066.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.881857109069824,
|
|
"epoch": 7.473684210526316,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001344740997729697,
|
|
"loss": 5.1897,
|
|
"mean_token_accuracy": 0.2119835287332535,
|
|
"num_tokens": 15910362.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.9012187957763675,
|
|
"epoch": 7.47953216374269,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00013410833055387943,
|
|
"loss": 5.1973,
|
|
"mean_token_accuracy": 0.20555586069822313,
|
|
"num_tokens": 15921930.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.887296867370606,
|
|
"epoch": 7.485380116959064,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00013374317281473783,
|
|
"loss": 5.1789,
|
|
"mean_token_accuracy": 0.21076648235321044,
|
|
"num_tokens": 15936492.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.898040008544922,
|
|
"epoch": 7.491228070175438,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00013337862814035564,
|
|
"loss": 5.2087,
|
|
"mean_token_accuracy": 0.21235450506210327,
|
|
"num_tokens": 15948994.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.859059572219849,
|
|
"epoch": 7.497076023391813,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013301469811288294,
|
|
"loss": 5.1005,
|
|
"mean_token_accuracy": 0.2122577026486397,
|
|
"num_tokens": 15962118.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.8275007724761965,
|
|
"epoch": 7.502923976608187,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00013265138431180207,
|
|
"loss": 5.0394,
|
|
"mean_token_accuracy": 0.22327614277601243,
|
|
"num_tokens": 15975065.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.88095908164978,
|
|
"epoch": 7.508771929824562,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000132288688313921,
|
|
"loss": 5.0855,
|
|
"mean_token_accuracy": 0.22290834784507751,
|
|
"num_tokens": 15987683.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.88097186088562,
|
|
"epoch": 7.514619883040936,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00013192661169336638,
|
|
"loss": 5.151,
|
|
"mean_token_accuracy": 0.21593787968158723,
|
|
"num_tokens": 15999875.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.850462913513184,
|
|
"epoch": 7.52046783625731,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00013156515602157664,
|
|
"loss": 5.1657,
|
|
"mean_token_accuracy": 0.21407234966754912,
|
|
"num_tokens": 16011285.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.834169340133667,
|
|
"epoch": 7.526315789473684,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00013120432286729548,
|
|
"loss": 5.136,
|
|
"mean_token_accuracy": 0.2122172698378563,
|
|
"num_tokens": 16024033.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.899155855178833,
|
|
"epoch": 7.5321637426900585,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00013084411379656453,
|
|
"loss": 5.1684,
|
|
"mean_token_accuracy": 0.20536281317472457,
|
|
"num_tokens": 16036415.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.885315895080566,
|
|
"epoch": 7.538011695906433,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00013048453037271696,
|
|
"loss": 5.1252,
|
|
"mean_token_accuracy": 0.21676210016012193,
|
|
"num_tokens": 16048601.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.917536067962646,
|
|
"epoch": 7.543859649122807,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00013012557415637073,
|
|
"loss": 5.1717,
|
|
"mean_token_accuracy": 0.2149989664554596,
|
|
"num_tokens": 16060226.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.83276777267456,
|
|
"epoch": 7.549707602339181,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001297672467054214,
|
|
"loss": 5.1122,
|
|
"mean_token_accuracy": 0.22261519432067872,
|
|
"num_tokens": 16072184.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.838424491882324,
|
|
"epoch": 7.555555555555555,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001294095495750361,
|
|
"loss": 5.0769,
|
|
"mean_token_accuracy": 0.22516507506370545,
|
|
"num_tokens": 16084138.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.841777420043945,
|
|
"epoch": 7.56140350877193,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00012905248431764566,
|
|
"loss": 5.1365,
|
|
"mean_token_accuracy": 0.21234174966812133,
|
|
"num_tokens": 16096894.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.926416540145874,
|
|
"epoch": 7.567251461988304,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00012869605248293914,
|
|
"loss": 5.244,
|
|
"mean_token_accuracy": 0.20191602706909179,
|
|
"num_tokens": 16109096.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.894947242736817,
|
|
"epoch": 7.573099415204679,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001283402556178563,
|
|
"loss": 5.2296,
|
|
"mean_token_accuracy": 0.20527111887931823,
|
|
"num_tokens": 16122717.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.89430422782898,
|
|
"epoch": 7.578947368421053,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00012798509526658076,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.21209409534931184,
|
|
"num_tokens": 16135643.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.918552732467651,
|
|
"epoch": 7.584795321637427,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00012763057297053427,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.214497672021389,
|
|
"num_tokens": 16148771.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.879001903533935,
|
|
"epoch": 7.590643274853801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00012727669026836862,
|
|
"loss": 5.1696,
|
|
"mean_token_accuracy": 0.21058829128742218,
|
|
"num_tokens": 16161078.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.871562337875366,
|
|
"epoch": 7.5964912280701755,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00012692344869596054,
|
|
"loss": 5.2028,
|
|
"mean_token_accuracy": 0.21265196204185485,
|
|
"num_tokens": 16173621.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.888844108581543,
|
|
"epoch": 7.60233918128655,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001265708497864035,
|
|
"loss": 5.1213,
|
|
"mean_token_accuracy": 0.22004827857017517,
|
|
"num_tokens": 16185874.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 7.60233918128655,
|
|
"eval_entropy": 5.864921841297881,
|
|
"eval_loss": 6.5312700271606445,
|
|
"eval_mean_token_accuracy": 0.15328266779317967,
|
|
"eval_num_tokens": 16185874.0,
|
|
"eval_runtime": 2.287,
|
|
"eval_samples_per_second": 1183.658,
|
|
"eval_steps_per_second": 148.23,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.87133264541626,
|
|
"epoch": 7.6081871345029235,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00012621889507000218,
|
|
"loss": 5.1552,
|
|
"mean_token_accuracy": 0.21903764307498932,
|
|
"num_tokens": 16198398.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.9023881435394285,
|
|
"epoch": 7.614035087719298,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00012586758607426523,
|
|
"loss": 5.1813,
|
|
"mean_token_accuracy": 0.2113264873623848,
|
|
"num_tokens": 16211134.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.861917161941529,
|
|
"epoch": 7.619883040935672,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00012551692432389888,
|
|
"loss": 5.1769,
|
|
"mean_token_accuracy": 0.21022562682628632,
|
|
"num_tokens": 16223707.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.909878349304199,
|
|
"epoch": 7.625730994152047,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001251669113408003,
|
|
"loss": 5.1722,
|
|
"mean_token_accuracy": 0.210222464799881,
|
|
"num_tokens": 16236444.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.963279294967651,
|
|
"epoch": 7.631578947368421,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00012481754864405092,
|
|
"loss": 5.2825,
|
|
"mean_token_accuracy": 0.19405649602413177,
|
|
"num_tokens": 16249499.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.922503232955933,
|
|
"epoch": 7.637426900584796,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00012446883774990992,
|
|
"loss": 5.2019,
|
|
"mean_token_accuracy": 0.20359573215246202,
|
|
"num_tokens": 16261582.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.821486711502075,
|
|
"epoch": 7.643274853801169,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012412078017180758,
|
|
"loss": 5.0782,
|
|
"mean_token_accuracy": 0.2195780321955681,
|
|
"num_tokens": 16274392.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.866240167617798,
|
|
"epoch": 7.649122807017544,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00012377337742033864,
|
|
"loss": 5.1786,
|
|
"mean_token_accuracy": 0.20985517352819444,
|
|
"num_tokens": 16286898.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.855369758605957,
|
|
"epoch": 7.654970760233918,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001234266310032562,
|
|
"loss": 5.0939,
|
|
"mean_token_accuracy": 0.21389817744493483,
|
|
"num_tokens": 16299716.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.949271869659424,
|
|
"epoch": 7.6608187134502925,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00012308054242546427,
|
|
"loss": 5.2771,
|
|
"mean_token_accuracy": 0.20447617620229722,
|
|
"num_tokens": 16313620.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.896162319183349,
|
|
"epoch": 7.666666666666667,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001227351131890124,
|
|
"loss": 5.1184,
|
|
"mean_token_accuracy": 0.21929934918880462,
|
|
"num_tokens": 16325471.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.828251838684082,
|
|
"epoch": 7.6725146198830405,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00012239034479308804,
|
|
"loss": 5.1216,
|
|
"mean_token_accuracy": 0.22113340348005295,
|
|
"num_tokens": 16337877.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.715599536895752,
|
|
"epoch": 7.678362573099415,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00012204623873401073,
|
|
"loss": 4.9768,
|
|
"mean_token_accuracy": 0.2329079270362854,
|
|
"num_tokens": 16351488.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.875384902954101,
|
|
"epoch": 7.684210526315789,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00012170279650522568,
|
|
"loss": 5.1787,
|
|
"mean_token_accuracy": 0.20896910279989242,
|
|
"num_tokens": 16364078.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.75007176399231,
|
|
"epoch": 7.690058479532164,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00012136001959729643,
|
|
"loss": 4.999,
|
|
"mean_token_accuracy": 0.22751859128475188,
|
|
"num_tokens": 16376866.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.865355777740478,
|
|
"epoch": 7.695906432748538,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00012101790949789968,
|
|
"loss": 5.165,
|
|
"mean_token_accuracy": 0.20557050704956054,
|
|
"num_tokens": 16390107.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.8646430492401125,
|
|
"epoch": 7.701754385964913,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00012067646769181761,
|
|
"loss": 5.1401,
|
|
"mean_token_accuracy": 0.21675122082233428,
|
|
"num_tokens": 16402647.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.87313666343689,
|
|
"epoch": 7.707602339181286,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00012033569566093223,
|
|
"loss": 5.1245,
|
|
"mean_token_accuracy": 0.21492659747600557,
|
|
"num_tokens": 16414435.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.940963935852051,
|
|
"epoch": 7.713450292397661,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00011999559488421852,
|
|
"loss": 5.244,
|
|
"mean_token_accuracy": 0.20350893586874008,
|
|
"num_tokens": 16426132.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.92445478439331,
|
|
"epoch": 7.719298245614035,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00011965616683773828,
|
|
"loss": 5.1771,
|
|
"mean_token_accuracy": 0.2077413335442543,
|
|
"num_tokens": 16438435.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.879776048660278,
|
|
"epoch": 7.7251461988304095,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00011931741299463354,
|
|
"loss": 5.1885,
|
|
"mean_token_accuracy": 0.21370921581983565,
|
|
"num_tokens": 16451175.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.775099515914917,
|
|
"epoch": 7.730994152046784,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00011897933482512031,
|
|
"loss": 5.1272,
|
|
"mean_token_accuracy": 0.22172015458345412,
|
|
"num_tokens": 16464833.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.859751892089844,
|
|
"epoch": 7.7368421052631575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00011864193379648208,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.21544105857610701,
|
|
"num_tokens": 16476672.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.81311182975769,
|
|
"epoch": 7.742690058479532,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011830521137306347,
|
|
"loss": 5.0568,
|
|
"mean_token_accuracy": 0.22072460800409316,
|
|
"num_tokens": 16488548.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.887657737731933,
|
|
"epoch": 7.748538011695906,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000117969169016264,
|
|
"loss": 5.1493,
|
|
"mean_token_accuracy": 0.21189361661672593,
|
|
"num_tokens": 16501614.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.863572502136231,
|
|
"epoch": 7.754385964912281,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011763380818453156,
|
|
"loss": 5.148,
|
|
"mean_token_accuracy": 0.21005476713180543,
|
|
"num_tokens": 16513061.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.868269968032837,
|
|
"epoch": 7.760233918128655,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00011729913033335616,
|
|
"loss": 5.148,
|
|
"mean_token_accuracy": 0.2131228744983673,
|
|
"num_tokens": 16525711.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.88983473777771,
|
|
"epoch": 7.76608187134503,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00011696513691526386,
|
|
"loss": 5.1166,
|
|
"mean_token_accuracy": 0.20983960330486298,
|
|
"num_tokens": 16537981.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.853209209442139,
|
|
"epoch": 7.771929824561403,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00011663182937980984,
|
|
"loss": 5.1196,
|
|
"mean_token_accuracy": 0.2168682411313057,
|
|
"num_tokens": 16550373.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.868277359008789,
|
|
"epoch": 7.777777777777778,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00011629920917357273,
|
|
"loss": 5.1161,
|
|
"mean_token_accuracy": 0.21858368813991547,
|
|
"num_tokens": 16564275.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.899106550216675,
|
|
"epoch": 7.783625730994152,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00011596727774014825,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.21568656265735625,
|
|
"num_tokens": 16575570.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.8177062511444095,
|
|
"epoch": 7.7894736842105265,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00011563603652014235,
|
|
"loss": 5.1794,
|
|
"mean_token_accuracy": 0.2080831781029701,
|
|
"num_tokens": 16588276.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.7727405548095705,
|
|
"epoch": 7.7953216374269,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00011530548695116594,
|
|
"loss": 4.989,
|
|
"mean_token_accuracy": 0.2253058597445488,
|
|
"num_tokens": 16600476.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.859190607070923,
|
|
"epoch": 7.8011695906432745,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001149756304678277,
|
|
"loss": 5.1387,
|
|
"mean_token_accuracy": 0.22045039087533952,
|
|
"num_tokens": 16613161.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.904561185836792,
|
|
"epoch": 7.807017543859649,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001146464685017285,
|
|
"loss": 5.2039,
|
|
"mean_token_accuracy": 0.2126063659787178,
|
|
"num_tokens": 16626615.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.900233316421509,
|
|
"epoch": 7.812865497076023,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001143180024814549,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.21391227692365647,
|
|
"num_tokens": 16637575.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.784756469726562,
|
|
"epoch": 7.818713450292398,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00011399023383257278,
|
|
"loss": 5.0937,
|
|
"mean_token_accuracy": 0.21939405500888826,
|
|
"num_tokens": 16650464.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.828171634674073,
|
|
"epoch": 7.824561403508772,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00011366316397762179,
|
|
"loss": 5.1379,
|
|
"mean_token_accuracy": 0.21435327976942062,
|
|
"num_tokens": 16663562.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.891318416595459,
|
|
"epoch": 7.830409356725146,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00011333679433610849,
|
|
"loss": 5.2207,
|
|
"mean_token_accuracy": 0.20312694758176802,
|
|
"num_tokens": 16676148.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.848797130584717,
|
|
"epoch": 7.83625730994152,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00011301112632450052,
|
|
"loss": 5.0924,
|
|
"mean_token_accuracy": 0.21762339174747466,
|
|
"num_tokens": 16688726.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.849908065795899,
|
|
"epoch": 7.842105263157895,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00011268616135622042,
|
|
"loss": 5.1566,
|
|
"mean_token_accuracy": 0.21155306249856948,
|
|
"num_tokens": 16701826.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.811604452133179,
|
|
"epoch": 7.847953216374269,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00011236190084163945,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.22032492905855178,
|
|
"num_tokens": 16713013.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.81961464881897,
|
|
"epoch": 7.853801169590644,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00011203834618807146,
|
|
"loss": 5.0926,
|
|
"mean_token_accuracy": 0.2198270320892334,
|
|
"num_tokens": 16725298.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.780805873870849,
|
|
"epoch": 7.859649122807017,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00011171549879976684,
|
|
"loss": 5.1178,
|
|
"mean_token_accuracy": 0.2203054755926132,
|
|
"num_tokens": 16738889.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.823562669754028,
|
|
"epoch": 7.8654970760233915,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00011139336007790657,
|
|
"loss": 5.1017,
|
|
"mean_token_accuracy": 0.22032118141651152,
|
|
"num_tokens": 16750848.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.874448442459107,
|
|
"epoch": 7.871345029239766,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00011107193142059554,
|
|
"loss": 5.1838,
|
|
"mean_token_accuracy": 0.21215973645448685,
|
|
"num_tokens": 16762294.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.874540138244629,
|
|
"epoch": 7.87719298245614,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00011075121422285736,
|
|
"loss": 5.1498,
|
|
"mean_token_accuracy": 0.21484171003103256,
|
|
"num_tokens": 16774433.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.81175684928894,
|
|
"epoch": 7.883040935672515,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00011043120987662771,
|
|
"loss": 5.0688,
|
|
"mean_token_accuracy": 0.21893557757139206,
|
|
"num_tokens": 16786828.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.937405347824097,
|
|
"epoch": 7.888888888888889,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00011011191977074819,
|
|
"loss": 5.2401,
|
|
"mean_token_accuracy": 0.20962440818548203,
|
|
"num_tokens": 16801221.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.920902347564697,
|
|
"epoch": 7.894736842105263,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000109793345290961,
|
|
"loss": 5.1732,
|
|
"mean_token_accuracy": 0.21226705759763717,
|
|
"num_tokens": 16814682.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.802679395675659,
|
|
"epoch": 7.900584795321637,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00010947548781990192,
|
|
"loss": 5.0837,
|
|
"mean_token_accuracy": 0.2145965039730072,
|
|
"num_tokens": 16827472.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.895700407028198,
|
|
"epoch": 7.906432748538012,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00010915834873709546,
|
|
"loss": 5.2266,
|
|
"mean_token_accuracy": 0.20440954566001893,
|
|
"num_tokens": 16839537.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.943037319183349,
|
|
"epoch": 7.912280701754386,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00010884192941894786,
|
|
"loss": 5.2319,
|
|
"mean_token_accuracy": 0.2041351467370987,
|
|
"num_tokens": 16850806.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.849862766265869,
|
|
"epoch": 7.918128654970761,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00010852623123874167,
|
|
"loss": 5.0716,
|
|
"mean_token_accuracy": 0.22146487832069398,
|
|
"num_tokens": 16862795.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.862993288040161,
|
|
"epoch": 7.923976608187134,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00010821125556662957,
|
|
"loss": 5.1705,
|
|
"mean_token_accuracy": 0.20833792239427568,
|
|
"num_tokens": 16874560.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.836628818511963,
|
|
"epoch": 7.9298245614035086,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00010789700376962867,
|
|
"loss": 5.0952,
|
|
"mean_token_accuracy": 0.22067254185676574,
|
|
"num_tokens": 16887577.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.822258186340332,
|
|
"epoch": 7.935672514619883,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010758347721161421,
|
|
"loss": 5.0908,
|
|
"mean_token_accuracy": 0.21994723826646806,
|
|
"num_tokens": 16899795.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.92198338508606,
|
|
"epoch": 7.941520467836257,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000107270677253314,
|
|
"loss": 5.1428,
|
|
"mean_token_accuracy": 0.21429868638515473,
|
|
"num_tokens": 16911582.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.811557865142822,
|
|
"epoch": 7.947368421052632,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00010695860525230224,
|
|
"loss": 5.0283,
|
|
"mean_token_accuracy": 0.21970626711845398,
|
|
"num_tokens": 16924658.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.8855987071990965,
|
|
"epoch": 7.953216374269006,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00010664726256299385,
|
|
"loss": 5.1759,
|
|
"mean_token_accuracy": 0.21461226791143417,
|
|
"num_tokens": 16936363.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.904529762268067,
|
|
"epoch": 7.95906432748538,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00010633665053663826,
|
|
"loss": 5.1916,
|
|
"mean_token_accuracy": 0.20434122979640962,
|
|
"num_tokens": 16949060.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.80406608581543,
|
|
"epoch": 7.964912280701754,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00010602677052131426,
|
|
"loss": 5.0984,
|
|
"mean_token_accuracy": 0.2208440274000168,
|
|
"num_tokens": 16961504.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.8398665428161625,
|
|
"epoch": 7.970760233918129,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00010571762386192302,
|
|
"loss": 5.134,
|
|
"mean_token_accuracy": 0.21571238934993744,
|
|
"num_tokens": 16975318.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.9180577278137205,
|
|
"epoch": 7.976608187134503,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00010540921190018343,
|
|
"loss": 5.2085,
|
|
"mean_token_accuracy": 0.20918190032243728,
|
|
"num_tokens": 16987764.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.854423093795776,
|
|
"epoch": 7.982456140350877,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00010510153597462538,
|
|
"loss": 5.1785,
|
|
"mean_token_accuracy": 0.20582091212272643,
|
|
"num_tokens": 17000015.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.894194650650024,
|
|
"epoch": 7.988304093567251,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010479459742058445,
|
|
"loss": 5.161,
|
|
"mean_token_accuracy": 0.21388570219278336,
|
|
"num_tokens": 17011517.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.91975736618042,
|
|
"epoch": 7.994152046783626,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00010448839757019616,
|
|
"loss": 5.2469,
|
|
"mean_token_accuracy": 0.20013497322797774,
|
|
"num_tokens": 17023525.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.768091011047363,
|
|
"epoch": 8.0,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00010418293775238952,
|
|
"loss": 5.0477,
|
|
"mean_token_accuracy": 0.2255009561777115,
|
|
"num_tokens": 17035872.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.786633110046386,
|
|
"epoch": 8.005847953216374,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00010387821929288228,
|
|
"loss": 4.9331,
|
|
"mean_token_accuracy": 0.23065799176692964,
|
|
"num_tokens": 17048069.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.8900469779968265,
|
|
"epoch": 8.011695906432749,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010357424351417433,
|
|
"loss": 5.1201,
|
|
"mean_token_accuracy": 0.21105236560106277,
|
|
"num_tokens": 17060956.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.832896518707275,
|
|
"epoch": 8.017543859649123,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00010327101173554236,
|
|
"loss": 4.9689,
|
|
"mean_token_accuracy": 0.22977243065834047,
|
|
"num_tokens": 17074546.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.823164939880371,
|
|
"epoch": 8.023391812865498,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000102968525273034,
|
|
"loss": 5.0011,
|
|
"mean_token_accuracy": 0.22379224747419357,
|
|
"num_tokens": 17088136.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.9150745391845705,
|
|
"epoch": 8.029239766081872,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010266678543946223,
|
|
"loss": 5.1114,
|
|
"mean_token_accuracy": 0.20976382791996,
|
|
"num_tokens": 17101814.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.877884292602539,
|
|
"epoch": 8.035087719298245,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00010236579354439956,
|
|
"loss": 5.0735,
|
|
"mean_token_accuracy": 0.21655522733926774,
|
|
"num_tokens": 17113592.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.9318647384643555,
|
|
"epoch": 8.04093567251462,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010206555089417235,
|
|
"loss": 5.1432,
|
|
"mean_token_accuracy": 0.21336976885795594,
|
|
"num_tokens": 17126288.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.879508352279663,
|
|
"epoch": 8.046783625730994,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001017660587918553,
|
|
"loss": 5.0657,
|
|
"mean_token_accuracy": 0.22253027111291884,
|
|
"num_tokens": 17138224.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.9390171527862545,
|
|
"epoch": 8.052631578947368,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00010146731853726554,
|
|
"loss": 5.1479,
|
|
"mean_token_accuracy": 0.2098224401473999,
|
|
"num_tokens": 17150705.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.876451110839843,
|
|
"epoch": 8.058479532163743,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00010116933142695728,
|
|
"loss": 5.0176,
|
|
"mean_token_accuracy": 0.22558841556310655,
|
|
"num_tokens": 17162957.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.8783165454864506,
|
|
"epoch": 8.064327485380117,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00010087209875421586,
|
|
"loss": 5.0663,
|
|
"mean_token_accuracy": 0.2212154299020767,
|
|
"num_tokens": 17175798.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.867189168930054,
|
|
"epoch": 8.070175438596491,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010057562180905237,
|
|
"loss": 5.0047,
|
|
"mean_token_accuracy": 0.2238185927271843,
|
|
"num_tokens": 17187449.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.8715380191802975,
|
|
"epoch": 8.076023391812866,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00010027990187819823,
|
|
"loss": 5.0307,
|
|
"mean_token_accuracy": 0.21817032098770142,
|
|
"num_tokens": 17199186.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.747433376312256,
|
|
"epoch": 8.08187134502924,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 9.998494024509894e-05,
|
|
"loss": 4.959,
|
|
"mean_token_accuracy": 0.23788644671440123,
|
|
"num_tokens": 17212630.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.781167125701904,
|
|
"epoch": 8.087719298245615,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.969073818990909e-05,
|
|
"loss": 5.0453,
|
|
"mean_token_accuracy": 0.22481624335050582,
|
|
"num_tokens": 17225457.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.827449464797974,
|
|
"epoch": 8.093567251461987,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.939729698948697e-05,
|
|
"loss": 5.0921,
|
|
"mean_token_accuracy": 0.22043644040822982,
|
|
"num_tokens": 17238530.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.778930807113648,
|
|
"epoch": 8.099415204678362,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.910461791738809e-05,
|
|
"loss": 4.9507,
|
|
"mean_token_accuracy": 0.2305397644639015,
|
|
"num_tokens": 17250953.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.850144624710083,
|
|
"epoch": 8.105263157894736,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.881270224386088e-05,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.220398710668087,
|
|
"num_tokens": 17263554.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.881121826171875,
|
|
"epoch": 8.11111111111111,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.852155123584018e-05,
|
|
"loss": 5.0902,
|
|
"mean_token_accuracy": 0.2177823856472969,
|
|
"num_tokens": 17275125.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.8802258491516115,
|
|
"epoch": 8.116959064327485,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.823116615694227e-05,
|
|
"loss": 5.0719,
|
|
"mean_token_accuracy": 0.21906458884477614,
|
|
"num_tokens": 17288132.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.780382537841797,
|
|
"epoch": 8.12280701754386,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 9.794154826745919e-05,
|
|
"loss": 5.0048,
|
|
"mean_token_accuracy": 0.2206939086318016,
|
|
"num_tokens": 17300645.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.8360830783844,
|
|
"epoch": 8.128654970760234,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.76526988243534e-05,
|
|
"loss": 5.0316,
|
|
"mean_token_accuracy": 0.22519539594650267,
|
|
"num_tokens": 17313207.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.7475690841674805,
|
|
"epoch": 8.134502923976608,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 9.736461908125214e-05,
|
|
"loss": 4.9076,
|
|
"mean_token_accuracy": 0.229633566737175,
|
|
"num_tokens": 17325173.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.856137371063232,
|
|
"epoch": 8.140350877192983,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 9.707731028844216e-05,
|
|
"loss": 5.0224,
|
|
"mean_token_accuracy": 0.22434089034795762,
|
|
"num_tokens": 17336308.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.860585737228393,
|
|
"epoch": 8.146198830409357,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.679077369286426e-05,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.22777694165706636,
|
|
"num_tokens": 17348845.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.894954109191895,
|
|
"epoch": 8.152046783625732,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.650501053810778e-05,
|
|
"loss": 5.1133,
|
|
"mean_token_accuracy": 0.21308251470327377,
|
|
"num_tokens": 17362525.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.785934209823608,
|
|
"epoch": 8.157894736842104,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 9.622002206440533e-05,
|
|
"loss": 4.9782,
|
|
"mean_token_accuracy": 0.22514349818229676,
|
|
"num_tokens": 17374973.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.859144496917724,
|
|
"epoch": 8.163742690058479,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.59358095086273e-05,
|
|
"loss": 4.9972,
|
|
"mean_token_accuracy": 0.23405203372240066,
|
|
"num_tokens": 17387431.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.916590690612793,
|
|
"epoch": 8.169590643274853,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.565237410427647e-05,
|
|
"loss": 5.1814,
|
|
"mean_token_accuracy": 0.2117987632751465,
|
|
"num_tokens": 17400471.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.909224891662598,
|
|
"epoch": 8.175438596491228,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 9.536971708148309e-05,
|
|
"loss": 5.0635,
|
|
"mean_token_accuracy": 0.21622217893600465,
|
|
"num_tokens": 17412165.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.939524507522583,
|
|
"epoch": 8.181286549707602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.508783966699867e-05,
|
|
"loss": 5.1274,
|
|
"mean_token_accuracy": 0.21066079437732696,
|
|
"num_tokens": 17424981.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.752725076675415,
|
|
"epoch": 8.187134502923977,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.48067430841914e-05,
|
|
"loss": 4.9436,
|
|
"mean_token_accuracy": 0.22999241203069687,
|
|
"num_tokens": 17437131.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 8.187134502923977,
|
|
"eval_entropy": 5.839937274786575,
|
|
"eval_loss": 6.529723167419434,
|
|
"eval_mean_token_accuracy": 0.15336025438125858,
|
|
"eval_num_tokens": 17437131.0,
|
|
"eval_runtime": 4.5225,
|
|
"eval_samples_per_second": 598.565,
|
|
"eval_steps_per_second": 74.959,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.88728518486023,
|
|
"epoch": 8.192982456140351,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.452642855304079e-05,
|
|
"loss": 5.1302,
|
|
"mean_token_accuracy": 0.2100319504737854,
|
|
"num_tokens": 17449015.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.870621871948242,
|
|
"epoch": 8.198830409356725,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 9.424689729013178e-05,
|
|
"loss": 5.0224,
|
|
"mean_token_accuracy": 0.22068232148885727,
|
|
"num_tokens": 17462547.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.901155567169189,
|
|
"epoch": 8.2046783625731,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.396815050865032e-05,
|
|
"loss": 5.0481,
|
|
"mean_token_accuracy": 0.2207955613732338,
|
|
"num_tokens": 17474630.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.864612054824829,
|
|
"epoch": 8.210526315789474,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.369018941837722e-05,
|
|
"loss": 5.0733,
|
|
"mean_token_accuracy": 0.21479740738868713,
|
|
"num_tokens": 17487015.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.870889663696289,
|
|
"epoch": 8.216374269005849,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.341301522568368e-05,
|
|
"loss": 5.0986,
|
|
"mean_token_accuracy": 0.21623831391334533,
|
|
"num_tokens": 17500772.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.80329008102417,
|
|
"epoch": 8.222222222222221,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.313662913352561e-05,
|
|
"loss": 4.9851,
|
|
"mean_token_accuracy": 0.22955896854400634,
|
|
"num_tokens": 17512818.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.842295122146607,
|
|
"epoch": 8.228070175438596,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 9.286103234143825e-05,
|
|
"loss": 5.0672,
|
|
"mean_token_accuracy": 0.21872319132089615,
|
|
"num_tokens": 17525022.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.884825992584228,
|
|
"epoch": 8.23391812865497,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.258622604553166e-05,
|
|
"loss": 5.0875,
|
|
"mean_token_accuracy": 0.21564254015684128,
|
|
"num_tokens": 17537248.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.876265001296997,
|
|
"epoch": 8.239766081871345,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.231221143848478e-05,
|
|
"loss": 5.0093,
|
|
"mean_token_accuracy": 0.22541218847036362,
|
|
"num_tokens": 17551003.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.8797142028808596,
|
|
"epoch": 8.24561403508772,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.20389897095407e-05,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.21421503126621247,
|
|
"num_tokens": 17562962.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.923748350143432,
|
|
"epoch": 8.251461988304094,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.176656204450129e-05,
|
|
"loss": 5.1317,
|
|
"mean_token_accuracy": 0.2144555926322937,
|
|
"num_tokens": 17576013.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.830680322647095,
|
|
"epoch": 8.257309941520468,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 9.149492962572212e-05,
|
|
"loss": 4.954,
|
|
"mean_token_accuracy": 0.2304137885570526,
|
|
"num_tokens": 17588733.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.8911439895629885,
|
|
"epoch": 8.263157894736842,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.122409363210738e-05,
|
|
"loss": 5.0972,
|
|
"mean_token_accuracy": 0.21913481652736663,
|
|
"num_tokens": 17600300.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.823686361312866,
|
|
"epoch": 8.269005847953217,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.095405523910468e-05,
|
|
"loss": 5.0738,
|
|
"mean_token_accuracy": 0.21685327142477034,
|
|
"num_tokens": 17611227.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.844291305541992,
|
|
"epoch": 8.274853801169591,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 9.068481561870018e-05,
|
|
"loss": 5.0497,
|
|
"mean_token_accuracy": 0.2260137602686882,
|
|
"num_tokens": 17624004.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.846800947189331,
|
|
"epoch": 8.280701754385966,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 9.041637593941292e-05,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.218025279045105,
|
|
"num_tokens": 17637557.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.935098028182983,
|
|
"epoch": 8.286549707602338,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.014873736629045e-05,
|
|
"loss": 5.1952,
|
|
"mean_token_accuracy": 0.2054765686392784,
|
|
"num_tokens": 17649129.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.884419775009155,
|
|
"epoch": 8.292397660818713,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.988190106090338e-05,
|
|
"loss": 5.1249,
|
|
"mean_token_accuracy": 0.21575379222631455,
|
|
"num_tokens": 17660881.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.892684602737427,
|
|
"epoch": 8.298245614035087,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.961586818134034e-05,
|
|
"loss": 5.0576,
|
|
"mean_token_accuracy": 0.22092251628637313,
|
|
"num_tokens": 17673560.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.855010652542115,
|
|
"epoch": 8.304093567251462,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 8.935063988220326e-05,
|
|
"loss": 5.0704,
|
|
"mean_token_accuracy": 0.21574662178754805,
|
|
"num_tokens": 17686071.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.740685749053955,
|
|
"epoch": 8.309941520467836,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.908621731460173e-05,
|
|
"loss": 4.9583,
|
|
"mean_token_accuracy": 0.23101436644792556,
|
|
"num_tokens": 17699357.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.738882398605346,
|
|
"epoch": 8.31578947368421,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 8.882260162614891e-05,
|
|
"loss": 4.8603,
|
|
"mean_token_accuracy": 0.23724617809057236,
|
|
"num_tokens": 17711626.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.855595922470092,
|
|
"epoch": 8.321637426900585,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.855979396095573e-05,
|
|
"loss": 5.0706,
|
|
"mean_token_accuracy": 0.22469092905521393,
|
|
"num_tokens": 17724661.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.799742221832275,
|
|
"epoch": 8.32748538011696,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.829779545962612e-05,
|
|
"loss": 4.9327,
|
|
"mean_token_accuracy": 0.22774261236190796,
|
|
"num_tokens": 17736881.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.840867328643799,
|
|
"epoch": 8.333333333333334,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.803660725925255e-05,
|
|
"loss": 4.9516,
|
|
"mean_token_accuracy": 0.22723323255777358,
|
|
"num_tokens": 17749255.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.947588348388672,
|
|
"epoch": 8.339181286549708,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.777623049341053e-05,
|
|
"loss": 5.0765,
|
|
"mean_token_accuracy": 0.2234731689095497,
|
|
"num_tokens": 17762663.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.9109718799591064,
|
|
"epoch": 8.345029239766081,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 8.75166662921538e-05,
|
|
"loss": 5.1229,
|
|
"mean_token_accuracy": 0.21286655515432357,
|
|
"num_tokens": 17775177.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.817645454406739,
|
|
"epoch": 8.350877192982455,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.725791578200967e-05,
|
|
"loss": 5.0607,
|
|
"mean_token_accuracy": 0.2219332531094551,
|
|
"num_tokens": 17788014.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.854659223556519,
|
|
"epoch": 8.35672514619883,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 8.699998008597391e-05,
|
|
"loss": 5.0823,
|
|
"mean_token_accuracy": 0.21687652170658112,
|
|
"num_tokens": 17803133.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.911325311660766,
|
|
"epoch": 8.362573099415204,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.674286032350593e-05,
|
|
"loss": 5.1144,
|
|
"mean_token_accuracy": 0.21549211889505387,
|
|
"num_tokens": 17815904.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.861735391616821,
|
|
"epoch": 8.368421052631579,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.648655761052393e-05,
|
|
"loss": 5.0712,
|
|
"mean_token_accuracy": 0.2190709099173546,
|
|
"num_tokens": 17827270.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.867135524749756,
|
|
"epoch": 8.374269005847953,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.623107305940009e-05,
|
|
"loss": 5.0716,
|
|
"mean_token_accuracy": 0.21449553221464157,
|
|
"num_tokens": 17839664.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.809323835372925,
|
|
"epoch": 8.380116959064328,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.59764077789556e-05,
|
|
"loss": 4.9773,
|
|
"mean_token_accuracy": 0.23139702081680297,
|
|
"num_tokens": 17851809.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.846773338317871,
|
|
"epoch": 8.385964912280702,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 8.572256287445614e-05,
|
|
"loss": 5.0196,
|
|
"mean_token_accuracy": 0.22606425136327743,
|
|
"num_tokens": 17865232.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.853825855255127,
|
|
"epoch": 8.391812865497077,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.546953944760677e-05,
|
|
"loss": 5.0114,
|
|
"mean_token_accuracy": 0.22210316658020018,
|
|
"num_tokens": 17878002.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.856227684020996,
|
|
"epoch": 8.397660818713451,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.521733859654721e-05,
|
|
"loss": 5.1016,
|
|
"mean_token_accuracy": 0.2158212885260582,
|
|
"num_tokens": 17889802.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.835464859008789,
|
|
"epoch": 8.403508771929825,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.496596141584743e-05,
|
|
"loss": 5.0831,
|
|
"mean_token_accuracy": 0.2117304563522339,
|
|
"num_tokens": 17902674.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.89712290763855,
|
|
"epoch": 8.409356725146198,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 8.471540899650221e-05,
|
|
"loss": 5.036,
|
|
"mean_token_accuracy": 0.21899518817663194,
|
|
"num_tokens": 17914399.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.821024990081787,
|
|
"epoch": 8.415204678362572,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.446568242592717e-05,
|
|
"loss": 5.0066,
|
|
"mean_token_accuracy": 0.22638234198093415,
|
|
"num_tokens": 17927882.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.868737888336182,
|
|
"epoch": 8.421052631578947,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.421678278795355e-05,
|
|
"loss": 5.0402,
|
|
"mean_token_accuracy": 0.21828827261924744,
|
|
"num_tokens": 17940443.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.798571252822876,
|
|
"epoch": 8.426900584795321,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 8.396871116282342e-05,
|
|
"loss": 5.0401,
|
|
"mean_token_accuracy": 0.22734010368585586,
|
|
"num_tokens": 17954363.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.906239414215088,
|
|
"epoch": 8.432748538011696,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.372146862718566e-05,
|
|
"loss": 5.0868,
|
|
"mean_token_accuracy": 0.21265804916620254,
|
|
"num_tokens": 17967231.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.883874225616455,
|
|
"epoch": 8.43859649122807,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.347505625409047e-05,
|
|
"loss": 5.1061,
|
|
"mean_token_accuracy": 0.2180918425321579,
|
|
"num_tokens": 17979735.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.915627527236938,
|
|
"epoch": 8.444444444444445,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.322947511298528e-05,
|
|
"loss": 5.1434,
|
|
"mean_token_accuracy": 0.21399013549089432,
|
|
"num_tokens": 17991400.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.875761222839356,
|
|
"epoch": 8.450292397660819,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.298472626970982e-05,
|
|
"loss": 5.0336,
|
|
"mean_token_accuracy": 0.22028183043003083,
|
|
"num_tokens": 18004023.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.824430322647094,
|
|
"epoch": 8.456140350877194,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 8.274081078649161e-05,
|
|
"loss": 5.0127,
|
|
"mean_token_accuracy": 0.22810006588697435,
|
|
"num_tokens": 18017052.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.864003658294678,
|
|
"epoch": 8.461988304093568,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 8.249772972194137e-05,
|
|
"loss": 5.1103,
|
|
"mean_token_accuracy": 0.21580332964658738,
|
|
"num_tokens": 18029279.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.787270212173462,
|
|
"epoch": 8.46783625730994,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 8.225548413104835e-05,
|
|
"loss": 4.9383,
|
|
"mean_token_accuracy": 0.23066811710596086,
|
|
"num_tokens": 18042063.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.884958744049072,
|
|
"epoch": 8.473684210526315,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 8.201407506517577e-05,
|
|
"loss": 5.0539,
|
|
"mean_token_accuracy": 0.2187667116522789,
|
|
"num_tokens": 18053330.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.836078262329101,
|
|
"epoch": 8.47953216374269,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.177350357205628e-05,
|
|
"loss": 5.0848,
|
|
"mean_token_accuracy": 0.21785496473312377,
|
|
"num_tokens": 18065761.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.852014446258545,
|
|
"epoch": 8.485380116959064,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 8.153377069578743e-05,
|
|
"loss": 5.0414,
|
|
"mean_token_accuracy": 0.22282142341136932,
|
|
"num_tokens": 18078037.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.799770259857178,
|
|
"epoch": 8.491228070175438,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.129487747682708e-05,
|
|
"loss": 4.9585,
|
|
"mean_token_accuracy": 0.2308633252978325,
|
|
"num_tokens": 18089811.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.930783271789551,
|
|
"epoch": 8.497076023391813,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.105682495198891e-05,
|
|
"loss": 5.1509,
|
|
"mean_token_accuracy": 0.21274857819080353,
|
|
"num_tokens": 18102107.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.835594558715821,
|
|
"epoch": 8.502923976608187,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 8.081961415443807e-05,
|
|
"loss": 5.0366,
|
|
"mean_token_accuracy": 0.22726909965276718,
|
|
"num_tokens": 18113749.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.812910652160644,
|
|
"epoch": 8.508771929824562,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 8.058324611368623e-05,
|
|
"loss": 5.0086,
|
|
"mean_token_accuracy": 0.22358557134866713,
|
|
"num_tokens": 18125511.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.842990684509277,
|
|
"epoch": 8.514619883040936,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 8.034772185558778e-05,
|
|
"loss": 5.0826,
|
|
"mean_token_accuracy": 0.21770238131284714,
|
|
"num_tokens": 18137284.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.905038595199585,
|
|
"epoch": 8.52046783625731,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.011304240233494e-05,
|
|
"loss": 5.1352,
|
|
"mean_token_accuracy": 0.20967480838298796,
|
|
"num_tokens": 18149564.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.860393142700195,
|
|
"epoch": 8.526315789473685,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.987920877245315e-05,
|
|
"loss": 5.0686,
|
|
"mean_token_accuracy": 0.22001402229070663,
|
|
"num_tokens": 18161825.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.9023401737213135,
|
|
"epoch": 8.53216374269006,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.964622198079738e-05,
|
|
"loss": 5.1204,
|
|
"mean_token_accuracy": 0.21550590097904204,
|
|
"num_tokens": 18175671.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.691943025588989,
|
|
"epoch": 8.538011695906432,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.941408303854678e-05,
|
|
"loss": 4.845,
|
|
"mean_token_accuracy": 0.24925006031990052,
|
|
"num_tokens": 18189083.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.90712890625,
|
|
"epoch": 8.543859649122806,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.918279295320116e-05,
|
|
"loss": 5.1209,
|
|
"mean_token_accuracy": 0.22101209610700606,
|
|
"num_tokens": 18203677.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.851021718978882,
|
|
"epoch": 8.549707602339181,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.895235272857606e-05,
|
|
"loss": 5.1,
|
|
"mean_token_accuracy": 0.21695491969585418,
|
|
"num_tokens": 18215785.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.867477893829346,
|
|
"epoch": 8.555555555555555,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 7.872276336479855e-05,
|
|
"loss": 5.0173,
|
|
"mean_token_accuracy": 0.2295249491930008,
|
|
"num_tokens": 18226619.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.884191989898682,
|
|
"epoch": 8.56140350877193,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.849402585830292e-05,
|
|
"loss": 5.1167,
|
|
"mean_token_accuracy": 0.21625495105981826,
|
|
"num_tokens": 18238428.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.876255989074707,
|
|
"epoch": 8.567251461988304,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.826614120182642e-05,
|
|
"loss": 5.0491,
|
|
"mean_token_accuracy": 0.2250838413834572,
|
|
"num_tokens": 18250769.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.879823350906372,
|
|
"epoch": 8.573099415204679,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.803911038440472e-05,
|
|
"loss": 5.0666,
|
|
"mean_token_accuracy": 0.21774054318666458,
|
|
"num_tokens": 18263340.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.923393964767456,
|
|
"epoch": 8.578947368421053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.781293439136781e-05,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.2112282246351242,
|
|
"num_tokens": 18276988.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.920970392227173,
|
|
"epoch": 8.584795321637428,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.75876142043359e-05,
|
|
"loss": 5.1194,
|
|
"mean_token_accuracy": 0.21236435770988465,
|
|
"num_tokens": 18289047.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.899970149993896,
|
|
"epoch": 8.590643274853802,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.73631508012146e-05,
|
|
"loss": 5.1452,
|
|
"mean_token_accuracy": 0.2154174491763115,
|
|
"num_tokens": 18302071.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.8553132057189945,
|
|
"epoch": 8.596491228070175,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.713954515619121e-05,
|
|
"loss": 5.0601,
|
|
"mean_token_accuracy": 0.21494852006435394,
|
|
"num_tokens": 18314416.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.819533491134644,
|
|
"epoch": 8.602339181286549,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.69167982397303e-05,
|
|
"loss": 5.0177,
|
|
"mean_token_accuracy": 0.22382608503103257,
|
|
"num_tokens": 18327214.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.802679824829101,
|
|
"epoch": 8.608187134502923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.669491101856936e-05,
|
|
"loss": 5.0138,
|
|
"mean_token_accuracy": 0.2232530012726784,
|
|
"num_tokens": 18339729.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.74157452583313,
|
|
"epoch": 8.614035087719298,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.647388445571502e-05,
|
|
"loss": 4.9694,
|
|
"mean_token_accuracy": 0.23078500479459763,
|
|
"num_tokens": 18351285.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.892966032028198,
|
|
"epoch": 8.619883040935672,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.625371951043824e-05,
|
|
"loss": 5.1556,
|
|
"mean_token_accuracy": 0.21145638823509216,
|
|
"num_tokens": 18364129.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.851832151412964,
|
|
"epoch": 8.625730994152047,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.603441713827075e-05,
|
|
"loss": 5.0812,
|
|
"mean_token_accuracy": 0.22541395127773284,
|
|
"num_tokens": 18376375.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.888142728805542,
|
|
"epoch": 8.631578947368421,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 7.581597829100062e-05,
|
|
"loss": 5.1215,
|
|
"mean_token_accuracy": 0.2174041748046875,
|
|
"num_tokens": 18388782.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.866147708892822,
|
|
"epoch": 8.637426900584796,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.559840391666798e-05,
|
|
"loss": 5.056,
|
|
"mean_token_accuracy": 0.2246631219983101,
|
|
"num_tokens": 18402392.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.845260572433472,
|
|
"epoch": 8.64327485380117,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.538169495956132e-05,
|
|
"loss": 5.0434,
|
|
"mean_token_accuracy": 0.21785633862018586,
|
|
"num_tokens": 18414238.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.871580171585083,
|
|
"epoch": 8.649122807017545,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.5165852360213e-05,
|
|
"loss": 5.1766,
|
|
"mean_token_accuracy": 0.2170020490884781,
|
|
"num_tokens": 18426096.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.8628324508667,
|
|
"epoch": 8.654970760233919,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.495087705539545e-05,
|
|
"loss": 5.0948,
|
|
"mean_token_accuracy": 0.21370044797658921,
|
|
"num_tokens": 18436929.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.8713050365448,
|
|
"epoch": 8.660818713450292,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.473676997811683e-05,
|
|
"loss": 5.0418,
|
|
"mean_token_accuracy": 0.2180508241057396,
|
|
"num_tokens": 18448206.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.840951061248779,
|
|
"epoch": 8.666666666666666,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.452353205761725e-05,
|
|
"loss": 5.0645,
|
|
"mean_token_accuracy": 0.21862475126981734,
|
|
"num_tokens": 18460313.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.898037576675415,
|
|
"epoch": 8.67251461988304,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.431116421936456e-05,
|
|
"loss": 5.0828,
|
|
"mean_token_accuracy": 0.21444789171218873,
|
|
"num_tokens": 18471984.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.994082260131836,
|
|
"epoch": 8.678362573099415,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.40996673850504e-05,
|
|
"loss": 5.1837,
|
|
"mean_token_accuracy": 0.20337404757738115,
|
|
"num_tokens": 18484310.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.858205413818359,
|
|
"epoch": 8.68421052631579,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.388904247258624e-05,
|
|
"loss": 5.1399,
|
|
"mean_token_accuracy": 0.21255170702934265,
|
|
"num_tokens": 18495939.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.837630987167358,
|
|
"epoch": 8.690058479532164,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.367929039609924e-05,
|
|
"loss": 5.0909,
|
|
"mean_token_accuracy": 0.22057269066572188,
|
|
"num_tokens": 18508409.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.864380645751953,
|
|
"epoch": 8.695906432748538,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.347041206592851e-05,
|
|
"loss": 5.0872,
|
|
"mean_token_accuracy": 0.21938099712133408,
|
|
"num_tokens": 18520465.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.890483760833741,
|
|
"epoch": 8.701754385964913,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7.326240838862097e-05,
|
|
"loss": 5.0958,
|
|
"mean_token_accuracy": 0.21648602336645126,
|
|
"num_tokens": 18533997.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.859398746490479,
|
|
"epoch": 8.707602339181287,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.30552802669274e-05,
|
|
"loss": 5.0895,
|
|
"mean_token_accuracy": 0.2143501430749893,
|
|
"num_tokens": 18546273.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.905642127990722,
|
|
"epoch": 8.713450292397662,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.284902859979891e-05,
|
|
"loss": 5.1934,
|
|
"mean_token_accuracy": 0.21309753656387329,
|
|
"num_tokens": 18560240.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.850403690338135,
|
|
"epoch": 8.719298245614034,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.264365428238232e-05,
|
|
"loss": 5.0735,
|
|
"mean_token_accuracy": 0.21909735053777696,
|
|
"num_tokens": 18572422.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.782764101028443,
|
|
"epoch": 8.725146198830409,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 7.243915820601699e-05,
|
|
"loss": 4.9837,
|
|
"mean_token_accuracy": 0.22374816685914994,
|
|
"num_tokens": 18584825.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.839371109008789,
|
|
"epoch": 8.730994152046783,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 7.223554125823058e-05,
|
|
"loss": 5.0547,
|
|
"mean_token_accuracy": 0.21895267814397812,
|
|
"num_tokens": 18596800.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.886599397659301,
|
|
"epoch": 8.736842105263158,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 7.203280432273513e-05,
|
|
"loss": 5.1605,
|
|
"mean_token_accuracy": 0.2147248551249504,
|
|
"num_tokens": 18610805.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.8841489315032955,
|
|
"epoch": 8.742690058479532,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7.18309482794236e-05,
|
|
"loss": 5.0697,
|
|
"mean_token_accuracy": 0.21852149665355683,
|
|
"num_tokens": 18622715.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.886079788208008,
|
|
"epoch": 8.748538011695906,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 7.162997400436563e-05,
|
|
"loss": 5.077,
|
|
"mean_token_accuracy": 0.21902636289596558,
|
|
"num_tokens": 18635293.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.906182336807251,
|
|
"epoch": 8.75438596491228,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 7.142988236980402e-05,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.20919781029224396,
|
|
"num_tokens": 18647202.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.870751190185547,
|
|
"epoch": 8.760233918128655,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.123067424415078e-05,
|
|
"loss": 5.0624,
|
|
"mean_token_accuracy": 0.21903820782899858,
|
|
"num_tokens": 18659576.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.881886768341064,
|
|
"epoch": 8.76608187134503,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.103235049198349e-05,
|
|
"loss": 5.109,
|
|
"mean_token_accuracy": 0.21400770545005798,
|
|
"num_tokens": 18672559.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.91015739440918,
|
|
"epoch": 8.771929824561404,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.083491197404143e-05,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.21434546113014222,
|
|
"num_tokens": 18685241.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 8.771929824561404,
|
|
"eval_entropy": 5.858551187149543,
|
|
"eval_loss": 6.524693489074707,
|
|
"eval_mean_token_accuracy": 0.15369601500465102,
|
|
"eval_num_tokens": 18685241.0,
|
|
"eval_runtime": 2.2742,
|
|
"eval_samples_per_second": 1190.326,
|
|
"eval_steps_per_second": 149.066,
|
|
"step": 7500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 8540,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 2.7041431862016e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|