65064b0c3b
Model: fpadovani/jpn-jpan-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.5087719298245617,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1.0817856546048e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|