65064b0c3b
Model: fpadovani/jpn-jpan-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.1695906432748537,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3617710663680000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|