0bc08d3f2a
Model: fpadovani/eng-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
6068 lines
164 KiB
JSON
6068 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.243374797187669,
|
|
"eval_steps": 1000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 6.804560613632202,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.676,
|
|
"mean_token_accuracy": 0.11936860606074333,
|
|
"num_tokens": 10948.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 6.834685564041138,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.6101,
|
|
"mean_token_accuracy": 0.12141277864575387,
|
|
"num_tokens": 21455.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 6.750869464874268,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.63,
|
|
"mean_token_accuracy": 0.12830424755811692,
|
|
"num_tokens": 32377.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 6.79157075881958,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.5238,
|
|
"mean_token_accuracy": 0.12687862142920495,
|
|
"num_tokens": 42276.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 6.750016450881958,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.5983,
|
|
"mean_token_accuracy": 0.12940823212265967,
|
|
"num_tokens": 55049.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.721223831176758,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.4942,
|
|
"mean_token_accuracy": 0.12893472835421563,
|
|
"num_tokens": 66792.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 6.726838397979736,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.6137,
|
|
"mean_token_accuracy": 0.1254408910870552,
|
|
"num_tokens": 78208.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 6.748009109497071,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.5285,
|
|
"mean_token_accuracy": 0.12195699289441109,
|
|
"num_tokens": 88871.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 6.831908369064331,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.5754,
|
|
"mean_token_accuracy": 0.13007976487278938,
|
|
"num_tokens": 100118.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 6.782376527786255,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.12892376035451888,
|
|
"num_tokens": 113120.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 6.78302035331726,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.5468,
|
|
"mean_token_accuracy": 0.1232769824564457,
|
|
"num_tokens": 123220.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 6.853044748306274,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.685,
|
|
"mean_token_accuracy": 0.12743987143039703,
|
|
"num_tokens": 134479.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 6.798919200897217,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.5239,
|
|
"mean_token_accuracy": 0.12954188883304596,
|
|
"num_tokens": 145189.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 6.781832551956176,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.495,
|
|
"mean_token_accuracy": 0.1263722836971283,
|
|
"num_tokens": 157354.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 6.827206039428711,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.6629,
|
|
"mean_token_accuracy": 0.1251331351697445,
|
|
"num_tokens": 168821.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 6.769263887405396,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.4675,
|
|
"mean_token_accuracy": 0.12909640297293662,
|
|
"num_tokens": 180554.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 6.827683067321777,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.5608,
|
|
"mean_token_accuracy": 0.12865682691335678,
|
|
"num_tokens": 191229.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.73335599899292,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.4681,
|
|
"mean_token_accuracy": 0.1287917375564575,
|
|
"num_tokens": 201903.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 6.850645112991333,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.6595,
|
|
"mean_token_accuracy": 0.12540490552783012,
|
|
"num_tokens": 214255.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 6.776821851730347,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.5262,
|
|
"mean_token_accuracy": 0.1281994625926018,
|
|
"num_tokens": 225423.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 6.793178844451904,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.6095,
|
|
"mean_token_accuracy": 0.12897214218974112,
|
|
"num_tokens": 237801.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.771815538406372,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.13965426087379457,
|
|
"num_tokens": 249195.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 6.6641730785369875,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.4097,
|
|
"mean_token_accuracy": 0.13700704053044319,
|
|
"num_tokens": 260129.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 6.825938653945923,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.6298,
|
|
"mean_token_accuracy": 0.12641703933477402,
|
|
"num_tokens": 271220.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 6.823924016952515,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.4819,
|
|
"mean_token_accuracy": 0.13169370368123054,
|
|
"num_tokens": 281560.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 6.640269660949707,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.4123,
|
|
"mean_token_accuracy": 0.13991473019123077,
|
|
"num_tokens": 292518.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 6.741516876220703,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.5952,
|
|
"mean_token_accuracy": 0.1250296004116535,
|
|
"num_tokens": 304865.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 6.848145580291748,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.4184,
|
|
"mean_token_accuracy": 0.13249706700444222,
|
|
"num_tokens": 315951.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 6.714532136917114,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.485,
|
|
"mean_token_accuracy": 0.13813133388757706,
|
|
"num_tokens": 327094.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.790360689163208,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.4913,
|
|
"mean_token_accuracy": 0.13367248624563216,
|
|
"num_tokens": 337801.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.683937454223633,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.4896,
|
|
"mean_token_accuracy": 0.13618724271655083,
|
|
"num_tokens": 349039.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 6.822566223144531,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.5093,
|
|
"mean_token_accuracy": 0.12809096947312354,
|
|
"num_tokens": 360639.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 6.866790294647217,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.5543,
|
|
"mean_token_accuracy": 0.13276145607233047,
|
|
"num_tokens": 371005.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 6.723278903961182,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.4563,
|
|
"mean_token_accuracy": 0.1335893027484417,
|
|
"num_tokens": 383340.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.638869333267212,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.1407618537545204,
|
|
"num_tokens": 394682.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 6.777361917495727,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.5174,
|
|
"mean_token_accuracy": 0.12884570136666298,
|
|
"num_tokens": 405713.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.754083967208862,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.448,
|
|
"mean_token_accuracy": 0.13516365364193916,
|
|
"num_tokens": 416208.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.655037212371826,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.3433,
|
|
"mean_token_accuracy": 0.1272922620177269,
|
|
"num_tokens": 427595.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 6.822840356826783,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.5328,
|
|
"mean_token_accuracy": 0.13492699712514877,
|
|
"num_tokens": 438273.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.719421339035034,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.453,
|
|
"mean_token_accuracy": 0.12706918492913247,
|
|
"num_tokens": 448948.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 6.7537905216217045,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.5158,
|
|
"mean_token_accuracy": 0.12623732313513755,
|
|
"num_tokens": 460838.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 6.820205020904541,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.5391,
|
|
"mean_token_accuracy": 0.12940281108021737,
|
|
"num_tokens": 473555.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.692381763458252,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.5015,
|
|
"mean_token_accuracy": 0.1293352037668228,
|
|
"num_tokens": 485221.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.703965425491333,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.3939,
|
|
"mean_token_accuracy": 0.13270398750901222,
|
|
"num_tokens": 496552.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.722942638397217,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.12913408428430556,
|
|
"num_tokens": 509760.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.653210401535034,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.3185,
|
|
"mean_token_accuracy": 0.14057503044605255,
|
|
"num_tokens": 521175.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.683060264587402,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.4599,
|
|
"mean_token_accuracy": 0.13481639921665192,
|
|
"num_tokens": 532597.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.669690656661987,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.5423,
|
|
"mean_token_accuracy": 0.13203370422124863,
|
|
"num_tokens": 545661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 6.889167976379395,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.4909,
|
|
"mean_token_accuracy": 0.1295941650867462,
|
|
"num_tokens": 557336.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.586929893493652,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.3887,
|
|
"mean_token_accuracy": 0.1374834731221199,
|
|
"num_tokens": 567398.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.73761625289917,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.4776,
|
|
"mean_token_accuracy": 0.13249947875738144,
|
|
"num_tokens": 579543.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.6058248519897464,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.3037,
|
|
"mean_token_accuracy": 0.14238273948431016,
|
|
"num_tokens": 589218.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.5944455623626705,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.3701,
|
|
"mean_token_accuracy": 0.13574394583702087,
|
|
"num_tokens": 600327.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.623188447952271,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.3389,
|
|
"mean_token_accuracy": 0.13717315047979356,
|
|
"num_tokens": 610923.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.801026058197022,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.4289,
|
|
"mean_token_accuracy": 0.1286413587629795,
|
|
"num_tokens": 621138.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.54342622756958,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.3733,
|
|
"mean_token_accuracy": 0.13763788491487502,
|
|
"num_tokens": 631440.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 6.745696115493774,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.4355,
|
|
"mean_token_accuracy": 0.13543301299214364,
|
|
"num_tokens": 642328.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.626775121688842,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.3643,
|
|
"mean_token_accuracy": 0.13606855794787406,
|
|
"num_tokens": 652875.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.605106496810913,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.3,
|
|
"mean_token_accuracy": 0.1362170659005642,
|
|
"num_tokens": 663726.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.8082743167877195,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.534,
|
|
"mean_token_accuracy": 0.12634207531809807,
|
|
"num_tokens": 675638.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.666491556167602,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.4047,
|
|
"mean_token_accuracy": 0.13120765015482902,
|
|
"num_tokens": 687239.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.707161617279053,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.4597,
|
|
"mean_token_accuracy": 0.12927801236510278,
|
|
"num_tokens": 698905.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 6.6740621566772464,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.4529,
|
|
"mean_token_accuracy": 0.13328550159931182,
|
|
"num_tokens": 710602.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.5762975215911865,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.4236,
|
|
"mean_token_accuracy": 0.13116890713572502,
|
|
"num_tokens": 721326.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.811364936828613,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.3957,
|
|
"mean_token_accuracy": 0.13776788860559464,
|
|
"num_tokens": 732972.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.571664810180664,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.3537,
|
|
"mean_token_accuracy": 0.13690979033708572,
|
|
"num_tokens": 743624.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.736306858062744,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.4756,
|
|
"mean_token_accuracy": 0.13334027379751207,
|
|
"num_tokens": 756384.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.744812917709351,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.4479,
|
|
"mean_token_accuracy": 0.1372112102806568,
|
|
"num_tokens": 767256.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.629975700378418,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.3612,
|
|
"mean_token_accuracy": 0.13806816861033439,
|
|
"num_tokens": 779059.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.734923458099365,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.4828,
|
|
"mean_token_accuracy": 0.13683631718158723,
|
|
"num_tokens": 790291.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.675401973724365,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.3718,
|
|
"mean_token_accuracy": 0.1384123720228672,
|
|
"num_tokens": 801450.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.519574356079102,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.3112,
|
|
"mean_token_accuracy": 0.14070284813642503,
|
|
"num_tokens": 812428.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.600682067871094,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.133209041506052,
|
|
"num_tokens": 823644.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.7434169292449955,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.3969,
|
|
"mean_token_accuracy": 0.1412114456295967,
|
|
"num_tokens": 834064.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.646493005752563,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.3464,
|
|
"mean_token_accuracy": 0.13672405928373338,
|
|
"num_tokens": 845207.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.678540420532227,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.4167,
|
|
"mean_token_accuracy": 0.13155203685164452,
|
|
"num_tokens": 855564.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.675069141387939,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.4551,
|
|
"mean_token_accuracy": 0.138312416523695,
|
|
"num_tokens": 866959.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.566254138946533,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.3763,
|
|
"mean_token_accuracy": 0.13844561353325843,
|
|
"num_tokens": 877319.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.7429297924041744,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.4585,
|
|
"mean_token_accuracy": 0.13310084044933318,
|
|
"num_tokens": 888037.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.519621229171753,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.216,
|
|
"mean_token_accuracy": 0.13677576705813407,
|
|
"num_tokens": 898775.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.66207971572876,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.3591,
|
|
"mean_token_accuracy": 0.13202459812164308,
|
|
"num_tokens": 909513.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.608506584167481,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.4058,
|
|
"mean_token_accuracy": 0.13273290768265725,
|
|
"num_tokens": 920730.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.621666431427002,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.3456,
|
|
"mean_token_accuracy": 0.14084025770425795,
|
|
"num_tokens": 931457.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.511780643463135,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.3503,
|
|
"mean_token_accuracy": 0.13648652881383896,
|
|
"num_tokens": 942572.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.6575798988342285,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.3487,
|
|
"mean_token_accuracy": 0.1328490473330021,
|
|
"num_tokens": 953132.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.612064790725708,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.3945,
|
|
"mean_token_accuracy": 0.13820873945951462,
|
|
"num_tokens": 963551.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.622423791885376,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.3555,
|
|
"mean_token_accuracy": 0.1371752694249153,
|
|
"num_tokens": 975297.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.592673206329346,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.3509,
|
|
"mean_token_accuracy": 0.13770653903484345,
|
|
"num_tokens": 986910.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.705222082138062,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.4895,
|
|
"mean_token_accuracy": 0.1382869116961956,
|
|
"num_tokens": 997058.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.570894861221314,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.3818,
|
|
"mean_token_accuracy": 0.13141772672533988,
|
|
"num_tokens": 1008261.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.588732576370239,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.2814,
|
|
"mean_token_accuracy": 0.14031188189983368,
|
|
"num_tokens": 1019193.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.581288385391235,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.4279,
|
|
"mean_token_accuracy": 0.13302782103419303,
|
|
"num_tokens": 1031227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.568547344207763,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.2321,
|
|
"mean_token_accuracy": 0.1408770777285099,
|
|
"num_tokens": 1042963.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.437390089035034,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.2996,
|
|
"mean_token_accuracy": 0.13700294718146325,
|
|
"num_tokens": 1053709.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.657422161102295,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.3475,
|
|
"mean_token_accuracy": 0.14993347823619843,
|
|
"num_tokens": 1064527.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.504825448989868,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4244,
|
|
"mean_token_accuracy": 0.13651368021965027,
|
|
"num_tokens": 1075127.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.6327752590179445,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.4171,
|
|
"mean_token_accuracy": 0.13323045745491982,
|
|
"num_tokens": 1086581.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.50992169380188,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.3437,
|
|
"mean_token_accuracy": 0.14702514410018921,
|
|
"num_tokens": 1097818.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.582889890670776,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.3189,
|
|
"mean_token_accuracy": 0.1353127583861351,
|
|
"num_tokens": 1108043.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.54462251663208,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.395,
|
|
"mean_token_accuracy": 0.13064431995153428,
|
|
"num_tokens": 1119875.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.528886127471924,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.2946,
|
|
"mean_token_accuracy": 0.13784321770071983,
|
|
"num_tokens": 1132406.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.733206176757813,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.4654,
|
|
"mean_token_accuracy": 0.1306557297706604,
|
|
"num_tokens": 1143828.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.470890140533447,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.2844,
|
|
"mean_token_accuracy": 0.134017276763916,
|
|
"num_tokens": 1155482.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.468265056610107,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.2269,
|
|
"mean_token_accuracy": 0.15050580948591233,
|
|
"num_tokens": 1166847.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.61746768951416,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.3131,
|
|
"mean_token_accuracy": 0.1418164536356926,
|
|
"num_tokens": 1177419.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.621824598312378,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4839,
|
|
"mean_token_accuracy": 0.1299498923122883,
|
|
"num_tokens": 1188975.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.526283550262451,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.3277,
|
|
"mean_token_accuracy": 0.1318104512989521,
|
|
"num_tokens": 1200252.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.502114009857178,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.3516,
|
|
"mean_token_accuracy": 0.14484424144029617,
|
|
"num_tokens": 1211864.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.527244329452515,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.2344,
|
|
"mean_token_accuracy": 0.13978444710373877,
|
|
"num_tokens": 1222429.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.539353752136231,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.3687,
|
|
"mean_token_accuracy": 0.13503918722271918,
|
|
"num_tokens": 1233782.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.604424047470093,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.3108,
|
|
"mean_token_accuracy": 0.13538552671670914,
|
|
"num_tokens": 1244309.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.405550956726074,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.2604,
|
|
"mean_token_accuracy": 0.13862796425819396,
|
|
"num_tokens": 1256403.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.593482112884521,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.368,
|
|
"mean_token_accuracy": 0.13131998851895332,
|
|
"num_tokens": 1268758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.455279159545898,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.2432,
|
|
"mean_token_accuracy": 0.14214460998773576,
|
|
"num_tokens": 1279355.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.446060514450073,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.2534,
|
|
"mean_token_accuracy": 0.13319372832775117,
|
|
"num_tokens": 1290619.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.472115468978882,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.2078,
|
|
"mean_token_accuracy": 0.1420151934027672,
|
|
"num_tokens": 1301851.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.567578840255737,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.3756,
|
|
"mean_token_accuracy": 0.13884048759937287,
|
|
"num_tokens": 1313260.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.437223625183106,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.3613,
|
|
"mean_token_accuracy": 0.13548871874809265,
|
|
"num_tokens": 1324189.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.612559270858765,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3834,
|
|
"mean_token_accuracy": 0.13582534343004227,
|
|
"num_tokens": 1335765.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.57055926322937,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.3301,
|
|
"mean_token_accuracy": 0.13537356853485108,
|
|
"num_tokens": 1347653.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.516417598724365,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3319,
|
|
"mean_token_accuracy": 0.13851706087589263,
|
|
"num_tokens": 1360201.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.3951996803283695,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.1915,
|
|
"mean_token_accuracy": 0.14084203839302062,
|
|
"num_tokens": 1370974.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.473271322250366,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.2639,
|
|
"mean_token_accuracy": 0.13946592584252357,
|
|
"num_tokens": 1383855.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.380508089065552,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.2487,
|
|
"mean_token_accuracy": 0.1441446803510189,
|
|
"num_tokens": 1395041.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.608717060089111,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.3495,
|
|
"mean_token_accuracy": 0.13589635416865348,
|
|
"num_tokens": 1406102.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.322340440750122,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.1684,
|
|
"mean_token_accuracy": 0.14366589039564132,
|
|
"num_tokens": 1417713.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.41874098777771,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3154,
|
|
"mean_token_accuracy": 0.14051032364368438,
|
|
"num_tokens": 1428526.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.511454772949219,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.2669,
|
|
"mean_token_accuracy": 0.1415833331644535,
|
|
"num_tokens": 1438849.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.425647258758545,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.338,
|
|
"mean_token_accuracy": 0.13376166373491288,
|
|
"num_tokens": 1450776.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.400399971008301,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.1926,
|
|
"mean_token_accuracy": 0.13401735424995423,
|
|
"num_tokens": 1462552.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.372816514968872,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.2145,
|
|
"mean_token_accuracy": 0.13779662400484086,
|
|
"num_tokens": 1473867.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.367090892791748,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.2229,
|
|
"mean_token_accuracy": 0.1417820245027542,
|
|
"num_tokens": 1485233.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.38919768333435,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.2657,
|
|
"mean_token_accuracy": 0.1480380691587925,
|
|
"num_tokens": 1496919.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.460553503036499,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.2319,
|
|
"mean_token_accuracy": 0.13744415640830993,
|
|
"num_tokens": 1508695.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.431367063522339,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.2857,
|
|
"mean_token_accuracy": 0.1416758455336094,
|
|
"num_tokens": 1519629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.417092514038086,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.1953,
|
|
"mean_token_accuracy": 0.14230941683053971,
|
|
"num_tokens": 1530144.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.333138132095337,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.2018,
|
|
"mean_token_accuracy": 0.14396312534809114,
|
|
"num_tokens": 1541013.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.468795728683472,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.2658,
|
|
"mean_token_accuracy": 0.1420371912419796,
|
|
"num_tokens": 1553344.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.4593830585479735,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.3229,
|
|
"mean_token_accuracy": 0.13928796947002411,
|
|
"num_tokens": 1565298.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.3176170825958256,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.2698,
|
|
"mean_token_accuracy": 0.14581286311149597,
|
|
"num_tokens": 1577645.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.400595617294312,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.2314,
|
|
"mean_token_accuracy": 0.1455103024840355,
|
|
"num_tokens": 1588387.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.489009809494019,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.14746614098548888,
|
|
"num_tokens": 1599837.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.347275733947754,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2675,
|
|
"mean_token_accuracy": 0.13686583638191224,
|
|
"num_tokens": 1610401.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.339430952072144,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.2135,
|
|
"mean_token_accuracy": 0.1422584317624569,
|
|
"num_tokens": 1622688.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.48252592086792,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.28,
|
|
"mean_token_accuracy": 0.13767832219600679,
|
|
"num_tokens": 1633594.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.333359479904175,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.1891,
|
|
"mean_token_accuracy": 0.13808258026838302,
|
|
"num_tokens": 1645861.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.202661991119385,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.0935,
|
|
"mean_token_accuracy": 0.14916127622127534,
|
|
"num_tokens": 1656903.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.368153619766235,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.14299265444278716,
|
|
"num_tokens": 1668130.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.201777791976928,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.1072,
|
|
"mean_token_accuracy": 0.14978721365332603,
|
|
"num_tokens": 1678228.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.457394170761108,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2464,
|
|
"mean_token_accuracy": 0.14466758221387863,
|
|
"num_tokens": 1690339.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.3576329231262205,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.2749,
|
|
"mean_token_accuracy": 0.13463479280471802,
|
|
"num_tokens": 1701067.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.3852208137512205,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.2136,
|
|
"mean_token_accuracy": 0.14947787374258042,
|
|
"num_tokens": 1712835.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.313982200622559,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.1699,
|
|
"mean_token_accuracy": 0.14509137719869614,
|
|
"num_tokens": 1723432.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.337073802947998,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.1998,
|
|
"mean_token_accuracy": 0.13908682987093926,
|
|
"num_tokens": 1734134.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.322796201705932,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.1652,
|
|
"mean_token_accuracy": 0.14878124743700027,
|
|
"num_tokens": 1744723.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.426623106002808,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.282,
|
|
"mean_token_accuracy": 0.1442626602947712,
|
|
"num_tokens": 1755269.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.393712663650513,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2924,
|
|
"mean_token_accuracy": 0.14282836467027665,
|
|
"num_tokens": 1766344.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.361828184127807,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2455,
|
|
"mean_token_accuracy": 0.13926497995853424,
|
|
"num_tokens": 1779052.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.1982386112213135,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.132,
|
|
"mean_token_accuracy": 0.1468164563179016,
|
|
"num_tokens": 1789754.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.213403940200806,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.0593,
|
|
"mean_token_accuracy": 0.15008397102355958,
|
|
"num_tokens": 1801161.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.34478988647461,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.1542,
|
|
"mean_token_accuracy": 0.14100898206233978,
|
|
"num_tokens": 1810892.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.226585960388183,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2234,
|
|
"mean_token_accuracy": 0.1414470836520195,
|
|
"num_tokens": 1821574.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.331795644760132,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.2499,
|
|
"mean_token_accuracy": 0.1461139790713787,
|
|
"num_tokens": 1833342.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.328761720657349,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.1284,
|
|
"mean_token_accuracy": 0.14916785582900047,
|
|
"num_tokens": 1843768.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.35804853439331,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2006,
|
|
"mean_token_accuracy": 0.14303460344672203,
|
|
"num_tokens": 1855569.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.284555959701538,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.1931,
|
|
"mean_token_accuracy": 0.14043574631214142,
|
|
"num_tokens": 1866684.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.273364686965943,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1105,
|
|
"mean_token_accuracy": 0.14270259737968444,
|
|
"num_tokens": 1878700.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.258121919631958,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1488,
|
|
"mean_token_accuracy": 0.1441320076584816,
|
|
"num_tokens": 1889970.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.198890638351441,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.0658,
|
|
"mean_token_accuracy": 0.14365847408771515,
|
|
"num_tokens": 1902008.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.253487491607666,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.2935,
|
|
"mean_token_accuracy": 0.1447810485959053,
|
|
"num_tokens": 1913253.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.276497077941895,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1373,
|
|
"mean_token_accuracy": 0.1510348603129387,
|
|
"num_tokens": 1923408.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.320691394805908,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.1986,
|
|
"mean_token_accuracy": 0.14775727093219757,
|
|
"num_tokens": 1935315.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.097941493988037,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.0456,
|
|
"mean_token_accuracy": 0.14678798317909242,
|
|
"num_tokens": 1946524.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.297728490829468,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.1813,
|
|
"mean_token_accuracy": 0.14240945130586624,
|
|
"num_tokens": 1957561.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.178823757171631,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.1188,
|
|
"mean_token_accuracy": 0.1454611212015152,
|
|
"num_tokens": 1969089.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.299928140640259,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.1458,
|
|
"mean_token_accuracy": 0.14649649932980538,
|
|
"num_tokens": 1980341.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.12959508895874,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.1144,
|
|
"mean_token_accuracy": 0.15559765547513962,
|
|
"num_tokens": 1991220.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.257497024536133,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2078,
|
|
"mean_token_accuracy": 0.14465426802635192,
|
|
"num_tokens": 2002650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.262146902084351,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1196,
|
|
"mean_token_accuracy": 0.15046928450465202,
|
|
"num_tokens": 2012884.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.315225553512573,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.1767,
|
|
"mean_token_accuracy": 0.14503859728574753,
|
|
"num_tokens": 2023543.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.3141711235046385,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.2486,
|
|
"mean_token_accuracy": 0.14016201198101044,
|
|
"num_tokens": 2034789.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.18101692199707,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.147,
|
|
"mean_token_accuracy": 0.1476250857114792,
|
|
"num_tokens": 2046719.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.13016300201416,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.1076,
|
|
"mean_token_accuracy": 0.14819848537445068,
|
|
"num_tokens": 2057016.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.18584303855896,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.0899,
|
|
"mean_token_accuracy": 0.14715152531862258,
|
|
"num_tokens": 2067898.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.1806296242607965,
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.1076,
|
|
"mean_token_accuracy": 0.14723623626761967,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.192201852798462,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.011,
|
|
"mean_token_accuracy": 0.14255973398685456,
|
|
"num_tokens": 2089228.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.130853652954102,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.0111,
|
|
"mean_token_accuracy": 0.15055073350667952,
|
|
"num_tokens": 2101098.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.19484510421753,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 5.9498,
|
|
"mean_token_accuracy": 0.16138385385274887,
|
|
"num_tokens": 2113081.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.1329797267913815,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 5.929,
|
|
"mean_token_accuracy": 0.1544704854488373,
|
|
"num_tokens": 2123356.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.242762088775635,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.0284,
|
|
"mean_token_accuracy": 0.15365447849035263,
|
|
"num_tokens": 2134388.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.03142409324646,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 5.9001,
|
|
"mean_token_accuracy": 0.15889017432928085,
|
|
"num_tokens": 2145209.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.138263082504272,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 5.9587,
|
|
"mean_token_accuracy": 0.15847741961479186,
|
|
"num_tokens": 2157146.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.0894828796386715,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 5.973,
|
|
"mean_token_accuracy": 0.1494349643588066,
|
|
"num_tokens": 2168721.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.10938835144043,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 5.9746,
|
|
"mean_token_accuracy": 0.1536022260785103,
|
|
"num_tokens": 2179213.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.141197156906128,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.9961,
|
|
"mean_token_accuracy": 0.15018199980258942,
|
|
"num_tokens": 2191224.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.267806339263916,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.0593,
|
|
"mean_token_accuracy": 0.14988923594355583,
|
|
"num_tokens": 2203665.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.117225313186646,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.1537061482667923,
|
|
"num_tokens": 2215136.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.1260009765625,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.013,
|
|
"mean_token_accuracy": 0.15645022243261336,
|
|
"num_tokens": 2226375.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.05034761428833,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 5.862,
|
|
"mean_token_accuracy": 0.15826935172080994,
|
|
"num_tokens": 2236097.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.181932258605957,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.1105,
|
|
"mean_token_accuracy": 0.14561123251914979,
|
|
"num_tokens": 2247964.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.062780239365318,
|
|
"eval_loss": 6.287398338317871,
|
|
"eval_mean_token_accuracy": 0.14789545227858153,
|
|
"eval_num_tokens": 2247964.0,
|
|
"eval_runtime": 2.5923,
|
|
"eval_samples_per_second": 1355.956,
|
|
"eval_steps_per_second": 169.736,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.103185844421387,
|
|
"epoch": 1.0865332612222822,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999997383518051,
|
|
"loss": 6.0583,
|
|
"mean_token_accuracy": 0.15494063049554824,
|
|
"num_tokens": 2260467.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.160040998458863,
|
|
"epoch": 1.0919415900486749,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999986754070562,
|
|
"loss": 5.9518,
|
|
"mean_token_accuracy": 0.1582807868719101,
|
|
"num_tokens": 2272176.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.112463903427124,
|
|
"epoch": 1.0973499188750675,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999967948166009,
|
|
"loss": 5.9872,
|
|
"mean_token_accuracy": 0.15890037268400192,
|
|
"num_tokens": 2283491.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.08608865737915,
|
|
"epoch": 1.1027582477014604,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999940965872734,
|
|
"loss": 5.9013,
|
|
"mean_token_accuracy": 0.1505625694990158,
|
|
"num_tokens": 2295131.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.06984395980835,
|
|
"epoch": 1.1081665765278528,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999905807288788,
|
|
"loss": 5.9428,
|
|
"mean_token_accuracy": 0.15290083289146422,
|
|
"num_tokens": 2305832.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 5.9947999000549315,
|
|
"epoch": 1.1135749053542456,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499986247254194,
|
|
"loss": 5.9314,
|
|
"mean_token_accuracy": 0.1570906773209572,
|
|
"num_tokens": 2315874.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.243255710601806,
|
|
"epoch": 1.1189832341806383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999810961789667,
|
|
"loss": 6.1117,
|
|
"mean_token_accuracy": 0.14899385422468187,
|
|
"num_tokens": 2327580.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.1247154712677006,
|
|
"epoch": 1.124391563007031,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499975127521916,
|
|
"loss": 5.9593,
|
|
"mean_token_accuracy": 0.15359291732311248,
|
|
"num_tokens": 2340426.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.109557294845581,
|
|
"epoch": 1.1297998918334236,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999683413047319,
|
|
"loss": 6.087,
|
|
"mean_token_accuracy": 0.14736745953559877,
|
|
"num_tokens": 2353228.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.092878198623657,
|
|
"epoch": 1.1352082206598162,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999607375520759,
|
|
"loss": 5.8995,
|
|
"mean_token_accuracy": 0.15909954607486726,
|
|
"num_tokens": 2364238.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.086206674575806,
|
|
"epoch": 1.1406165494862088,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999523162915794,
|
|
"loss": 6.0444,
|
|
"mean_token_accuracy": 0.1497299149632454,
|
|
"num_tokens": 2375847.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.047412347793579,
|
|
"epoch": 1.1460248783126015,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999430775538459,
|
|
"loss": 5.9214,
|
|
"mean_token_accuracy": 0.15928174257278443,
|
|
"num_tokens": 2386489.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.080969953536988,
|
|
"epoch": 1.151433207138994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999330213724487,
|
|
"loss": 5.9616,
|
|
"mean_token_accuracy": 0.15427634492516518,
|
|
"num_tokens": 2398839.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.098521661758423,
|
|
"epoch": 1.1568415359653867,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999221477839319,
|
|
"loss": 5.9327,
|
|
"mean_token_accuracy": 0.16004678606987,
|
|
"num_tokens": 2410016.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.011932134628296,
|
|
"epoch": 1.1622498647917794,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999104568278104,
|
|
"loss": 5.9662,
|
|
"mean_token_accuracy": 0.1546537309885025,
|
|
"num_tokens": 2421202.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.13898868560791,
|
|
"epoch": 1.167658193618172,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499897948546569,
|
|
"loss": 6.0554,
|
|
"mean_token_accuracy": 0.14611297100782394,
|
|
"num_tokens": 2431943.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.0566650390625,
|
|
"epoch": 1.1730665224445647,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998846229856629,
|
|
"loss": 5.8776,
|
|
"mean_token_accuracy": 0.15458440482616426,
|
|
"num_tokens": 2443454.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.059741497039795,
|
|
"epoch": 1.1784748512709573,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998704801935171,
|
|
"loss": 6.0745,
|
|
"mean_token_accuracy": 0.14530793875455855,
|
|
"num_tokens": 2455909.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 5.996316003799438,
|
|
"epoch": 1.18388318009735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998555202215267,
|
|
"loss": 5.9358,
|
|
"mean_token_accuracy": 0.16019879430532455,
|
|
"num_tokens": 2467486.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.088795566558838,
|
|
"epoch": 1.1892915089237426,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998397431240561,
|
|
"loss": 5.8886,
|
|
"mean_token_accuracy": 0.16266787201166152,
|
|
"num_tokens": 2477560.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 5.929623699188232,
|
|
"epoch": 1.1946998377501352,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998231489584395,
|
|
"loss": 5.9373,
|
|
"mean_token_accuracy": 0.16155486851930617,
|
|
"num_tokens": 2487222.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.084240818023682,
|
|
"epoch": 1.2001081665765279,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00049980573778498,
|
|
"loss": 5.956,
|
|
"mean_token_accuracy": 0.15426427125930786,
|
|
"num_tokens": 2497474.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.036470985412597,
|
|
"epoch": 1.2055164954029205,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997875096669498,
|
|
"loss": 6.0157,
|
|
"mean_token_accuracy": 0.15495393723249434,
|
|
"num_tokens": 2508185.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.075341987609863,
|
|
"epoch": 1.2109248242293131,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004997684646705901,
|
|
"loss": 5.9611,
|
|
"mean_token_accuracy": 0.15263352990150453,
|
|
"num_tokens": 2518876.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.024726867675781,
|
|
"epoch": 1.2163331530557058,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997486028651103,
|
|
"loss": 5.9497,
|
|
"mean_token_accuracy": 0.1580318495631218,
|
|
"num_tokens": 2530554.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 5.956853342056275,
|
|
"epoch": 1.2217414818820984,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997279243226884,
|
|
"loss": 5.8434,
|
|
"mean_token_accuracy": 0.1560763031244278,
|
|
"num_tokens": 2541331.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.129449796676636,
|
|
"epoch": 1.227149810708491,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997064291184701,
|
|
"loss": 6.061,
|
|
"mean_token_accuracy": 0.14723667353391648,
|
|
"num_tokens": 2554225.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 5.949807167053223,
|
|
"epoch": 1.2325581395348837,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004996841173305691,
|
|
"loss": 5.9374,
|
|
"mean_token_accuracy": 0.157360303401947,
|
|
"num_tokens": 2565050.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.148911762237549,
|
|
"epoch": 1.2379664683612763,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996609890400665,
|
|
"loss": 6.029,
|
|
"mean_token_accuracy": 0.1493443951010704,
|
|
"num_tokens": 2576926.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.134435796737671,
|
|
"epoch": 1.243374797187669,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996370443310107,
|
|
"loss": 6.0976,
|
|
"mean_token_accuracy": 0.14909718930721283,
|
|
"num_tokens": 2588775.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 5.956080913543701,
|
|
"epoch": 1.2487831260140616,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996122832904165,
|
|
"loss": 5.9534,
|
|
"mean_token_accuracy": 0.1530027374625206,
|
|
"num_tokens": 2600534.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 5.9704314231872555,
|
|
"epoch": 1.2541914548404542,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995867060082658,
|
|
"loss": 5.7776,
|
|
"mean_token_accuracy": 0.1611389622092247,
|
|
"num_tokens": 2612350.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 5.943893909454346,
|
|
"epoch": 1.2595997836668469,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995603125775065,
|
|
"loss": 5.9646,
|
|
"mean_token_accuracy": 0.15136941969394685,
|
|
"num_tokens": 2623871.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.044587802886963,
|
|
"epoch": 1.2650081124932395,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995331030940523,
|
|
"loss": 5.9051,
|
|
"mean_token_accuracy": 0.15263695865869523,
|
|
"num_tokens": 2636371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.092774963378906,
|
|
"epoch": 1.2704164413196322,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995050776567826,
|
|
"loss": 5.9491,
|
|
"mean_token_accuracy": 0.15803743302822112,
|
|
"num_tokens": 2646794.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.90415997505188,
|
|
"epoch": 1.2758247701460248,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994762363675421,
|
|
"loss": 5.9559,
|
|
"mean_token_accuracy": 0.1607417270541191,
|
|
"num_tokens": 2657105.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.061801719665527,
|
|
"epoch": 1.2812330989724177,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004994465793311396,
|
|
"loss": 5.8831,
|
|
"mean_token_accuracy": 0.15674052089452745,
|
|
"num_tokens": 2668297.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.922903776168823,
|
|
"epoch": 1.28664142779881,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004994161066553493,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.15893352329730986,
|
|
"num_tokens": 2678834.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 5.943311405181885,
|
|
"epoch": 1.292049756625203,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993848184509086,
|
|
"loss": 5.9929,
|
|
"mean_token_accuracy": 0.15331263542175294,
|
|
"num_tokens": 2690392.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.8633359432220455,
|
|
"epoch": 1.2974580854515954,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004993527148315193,
|
|
"loss": 5.7922,
|
|
"mean_token_accuracy": 0.15931502282619475,
|
|
"num_tokens": 2701319.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.0737066745758055,
|
|
"epoch": 1.3028664142779882,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004993197959138458,
|
|
"loss": 5.9576,
|
|
"mean_token_accuracy": 0.15070452392101288,
|
|
"num_tokens": 2713018.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.998841667175293,
|
|
"epoch": 1.3082747431043806,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004992860618175156,
|
|
"loss": 6.077,
|
|
"mean_token_accuracy": 0.14975738078355788,
|
|
"num_tokens": 2724995.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.051755809783936,
|
|
"epoch": 1.3136830719307735,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992515126651184,
|
|
"loss": 5.8892,
|
|
"mean_token_accuracy": 0.15623158663511277,
|
|
"num_tokens": 2736733.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.874134159088134,
|
|
"epoch": 1.319091400757166,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992161485822059,
|
|
"loss": 5.8884,
|
|
"mean_token_accuracy": 0.1621337816119194,
|
|
"num_tokens": 2748436.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.914373540878296,
|
|
"epoch": 1.3244997295835588,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991799696972913,
|
|
"loss": 5.9013,
|
|
"mean_token_accuracy": 0.15741000026464463,
|
|
"num_tokens": 2758924.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.100838947296142,
|
|
"epoch": 1.3299080584099512,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004991429761418488,
|
|
"loss": 5.9506,
|
|
"mean_token_accuracy": 0.15151908099651337,
|
|
"num_tokens": 2769773.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.044694995880127,
|
|
"epoch": 1.335316387236344,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499105168050313,
|
|
"loss": 6.0952,
|
|
"mean_token_accuracy": 0.1586819291114807,
|
|
"num_tokens": 2782414.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.923618650436401,
|
|
"epoch": 1.3407247160627367,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990665455600787,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.15739361941814423,
|
|
"num_tokens": 2794325.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.849081230163574,
|
|
"epoch": 1.3461330448891293,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990271088114999,
|
|
"loss": 5.7123,
|
|
"mean_token_accuracy": 0.16516527980566026,
|
|
"num_tokens": 2805391.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.8755106925964355,
|
|
"epoch": 1.351541373715522,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989868579478901,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.15670257657766343,
|
|
"num_tokens": 2816633.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.043559789657593,
|
|
"epoch": 1.3569497025419146,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004989457931155209,
|
|
"loss": 5.9513,
|
|
"mean_token_accuracy": 0.15694985687732696,
|
|
"num_tokens": 2828395.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.018378067016601,
|
|
"epoch": 1.3623580313683072,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004989039144636219,
|
|
"loss": 5.9441,
|
|
"mean_token_accuracy": 0.15901106745004653,
|
|
"num_tokens": 2839592.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.976392936706543,
|
|
"epoch": 1.3677663601946999,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004988612221443804,
|
|
"loss": 5.9075,
|
|
"mean_token_accuracy": 0.16764327585697175,
|
|
"num_tokens": 2849557.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.009874248504639,
|
|
"epoch": 1.3731746890210925,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004988177163129403,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.15865644067525864,
|
|
"num_tokens": 2862030.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.910986471176147,
|
|
"epoch": 1.3785830178474852,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987733971274018,
|
|
"loss": 5.868,
|
|
"mean_token_accuracy": 0.16133906841278076,
|
|
"num_tokens": 2872379.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.0357307434082035,
|
|
"epoch": 1.3839913466738778,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987282647488211,
|
|
"loss": 5.9579,
|
|
"mean_token_accuracy": 0.16110740303993226,
|
|
"num_tokens": 2883170.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.914972877502441,
|
|
"epoch": 1.3893996755002704,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004986823193412093,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.15728989392518997,
|
|
"num_tokens": 2894013.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.044302940368652,
|
|
"epoch": 1.394808004326663,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000498635561071532,
|
|
"loss": 5.9602,
|
|
"mean_token_accuracy": 0.15943014323711396,
|
|
"num_tokens": 2905073.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.024225664138794,
|
|
"epoch": 1.4002163331530557,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000498587990109709,
|
|
"loss": 5.9323,
|
|
"mean_token_accuracy": 0.15316786766052246,
|
|
"num_tokens": 2916579.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.833254861831665,
|
|
"epoch": 1.4056246619794484,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004985396066286133,
|
|
"loss": 5.8355,
|
|
"mean_token_accuracy": 0.15808981508016587,
|
|
"num_tokens": 2928522.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.022909688949585,
|
|
"epoch": 1.411032990805841,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004984904108040706,
|
|
"loss": 5.9247,
|
|
"mean_token_accuracy": 0.16038866490125656,
|
|
"num_tokens": 2938754.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.792943954467773,
|
|
"epoch": 1.4164413196322336,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004984404028148585,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.16507242918014525,
|
|
"num_tokens": 2949696.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.921805334091187,
|
|
"epoch": 1.4218496484586263,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004983895828427062,
|
|
"loss": 5.9148,
|
|
"mean_token_accuracy": 0.15484894514083863,
|
|
"num_tokens": 2961144.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.958304309844971,
|
|
"epoch": 1.427257977285019,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983379510722935,
|
|
"loss": 5.9021,
|
|
"mean_token_accuracy": 0.1613841623067856,
|
|
"num_tokens": 2972368.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.935693836212158,
|
|
"epoch": 1.4326663061114115,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004982855076912506,
|
|
"loss": 5.8272,
|
|
"mean_token_accuracy": 0.157490174472332,
|
|
"num_tokens": 2984853.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.975592565536499,
|
|
"epoch": 1.4380746349378042,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004982322528901568,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.161625674366951,
|
|
"num_tokens": 2995095.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.916803073883057,
|
|
"epoch": 1.4434829637641968,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00049817818686254,
|
|
"loss": 5.9433,
|
|
"mean_token_accuracy": 0.15642223358154297,
|
|
"num_tokens": 3005926.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.045813846588135,
|
|
"epoch": 1.4488912925905895,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004981233098048765,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.156609845161438,
|
|
"num_tokens": 3017647.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.861057567596435,
|
|
"epoch": 1.454299621416982,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004980676219165894,
|
|
"loss": 5.8174,
|
|
"mean_token_accuracy": 0.16126608699560166,
|
|
"num_tokens": 3028904.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.0215582847595215,
|
|
"epoch": 1.4597079502433747,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498011123400049,
|
|
"loss": 5.8676,
|
|
"mean_token_accuracy": 0.1609919086098671,
|
|
"num_tokens": 3039347.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.861247110366821,
|
|
"epoch": 1.4651162790697674,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004979538144605707,
|
|
"loss": 5.8372,
|
|
"mean_token_accuracy": 0.16235667169094087,
|
|
"num_tokens": 3049951.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.96915078163147,
|
|
"epoch": 1.47052460789616,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004978956953064154,
|
|
"loss": 5.894,
|
|
"mean_token_accuracy": 0.15848492085933685,
|
|
"num_tokens": 3061894.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.92190408706665,
|
|
"epoch": 1.4759329367225527,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004978367661487883,
|
|
"loss": 5.8533,
|
|
"mean_token_accuracy": 0.16612305045127868,
|
|
"num_tokens": 3072393.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.887731313705444,
|
|
"epoch": 1.4813412655489453,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004977770272018379,
|
|
"loss": 5.8958,
|
|
"mean_token_accuracy": 0.16117597371339798,
|
|
"num_tokens": 3083541.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.8998823165893555,
|
|
"epoch": 1.486749594375338,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497716478682656,
|
|
"loss": 5.7312,
|
|
"mean_token_accuracy": 0.16381334215402604,
|
|
"num_tokens": 3093851.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.803137588500976,
|
|
"epoch": 1.4921579232017308,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976551208112759,
|
|
"loss": 5.8132,
|
|
"mean_token_accuracy": 0.1614048346877098,
|
|
"num_tokens": 3105703.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.030934000015259,
|
|
"epoch": 1.4975662520281232,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004975929538106721,
|
|
"loss": 5.9342,
|
|
"mean_token_accuracy": 0.15392773896455764,
|
|
"num_tokens": 3117063.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.88799262046814,
|
|
"epoch": 1.502974580854516,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004975299779067598,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.15933458805084227,
|
|
"num_tokens": 3129020.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.959277105331421,
|
|
"epoch": 1.5083829096809085,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004974661933283936,
|
|
"loss": 5.8976,
|
|
"mean_token_accuracy": 0.15693499147891998,
|
|
"num_tokens": 3140627.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.932709836959839,
|
|
"epoch": 1.5137912385073014,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004974016003073667,
|
|
"loss": 5.9111,
|
|
"mean_token_accuracy": 0.15645995289087294,
|
|
"num_tokens": 3152294.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.866684675216675,
|
|
"epoch": 1.5191995673336938,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004973361990784104,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.15972761809825897,
|
|
"num_tokens": 3163770.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.999703359603882,
|
|
"epoch": 1.5246078961600866,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004972699898791927,
|
|
"loss": 5.8653,
|
|
"mean_token_accuracy": 0.15475109070539475,
|
|
"num_tokens": 3175804.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.959162330627441,
|
|
"epoch": 1.530016224986479,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004972029729503182,
|
|
"loss": 5.8715,
|
|
"mean_token_accuracy": 0.16324462890625,
|
|
"num_tokens": 3187233.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.809819269180298,
|
|
"epoch": 1.535424553812872,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004971351485353265,
|
|
"loss": 5.8495,
|
|
"mean_token_accuracy": 0.16579529494047165,
|
|
"num_tokens": 3198884.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.859294652938843,
|
|
"epoch": 1.5408328826392643,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004970665168806917,
|
|
"loss": 5.8231,
|
|
"mean_token_accuracy": 0.15876574665308,
|
|
"num_tokens": 3209836.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.875070858001709,
|
|
"epoch": 1.5462412114656572,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004969970782358214,
|
|
"loss": 5.8374,
|
|
"mean_token_accuracy": 0.16300275176763535,
|
|
"num_tokens": 3221295.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.844339418411255,
|
|
"epoch": 1.5516495402920496,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004969268328530557,
|
|
"loss": 5.7734,
|
|
"mean_token_accuracy": 0.1712643414735794,
|
|
"num_tokens": 3231832.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.921893072128296,
|
|
"epoch": 1.5570578691184425,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004968557809876666,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.1570061594247818,
|
|
"num_tokens": 3243483.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.895966863632202,
|
|
"epoch": 1.562466197944835,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004967839228978568,
|
|
"loss": 5.9079,
|
|
"mean_token_accuracy": 0.15856830030679703,
|
|
"num_tokens": 3255785.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.881782865524292,
|
|
"epoch": 1.5678745267712277,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004967112588447585,
|
|
"loss": 5.8009,
|
|
"mean_token_accuracy": 0.15945337414741517,
|
|
"num_tokens": 3267035.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.979495429992676,
|
|
"epoch": 1.5732828555976204,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004966377890924332,
|
|
"loss": 5.8888,
|
|
"mean_token_accuracy": 0.1595017358660698,
|
|
"num_tokens": 3277848.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.839128541946411,
|
|
"epoch": 1.578691184424013,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004965635139078704,
|
|
"loss": 5.8046,
|
|
"mean_token_accuracy": 0.16240252256393434,
|
|
"num_tokens": 3289864.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.943506908416748,
|
|
"epoch": 1.5840995132504057,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004964884335609857,
|
|
"loss": 5.9401,
|
|
"mean_token_accuracy": 0.15871844738721846,
|
|
"num_tokens": 3301441.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.836030387878418,
|
|
"epoch": 1.5895078420767983,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004964125483246218,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.15844986885786055,
|
|
"num_tokens": 3313216.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.828057146072387,
|
|
"epoch": 1.594916170903191,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004963358584745457,
|
|
"loss": 5.7186,
|
|
"mean_token_accuracy": 0.16143687814474106,
|
|
"num_tokens": 3323760.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.829604816436768,
|
|
"epoch": 1.6003244997295836,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004962583642894482,
|
|
"loss": 5.779,
|
|
"mean_token_accuracy": 0.16349702924489976,
|
|
"num_tokens": 3334785.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.889637184143067,
|
|
"epoch": 1.6057328285559762,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004961800660509437,
|
|
"loss": 5.8413,
|
|
"mean_token_accuracy": 0.16552892476320266,
|
|
"num_tokens": 3345741.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.864379262924194,
|
|
"epoch": 1.6111411573823688,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000496100964043568,
|
|
"loss": 5.7981,
|
|
"mean_token_accuracy": 0.1618591532111168,
|
|
"num_tokens": 3356118.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.8000274181365965,
|
|
"epoch": 1.6165494862087615,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000496021058554778,
|
|
"loss": 5.7479,
|
|
"mean_token_accuracy": 0.16704143285751344,
|
|
"num_tokens": 3368107.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.87118935585022,
|
|
"epoch": 1.6219578150351541,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004959403498749505,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.16020423620939256,
|
|
"num_tokens": 3379367.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.761199283599853,
|
|
"epoch": 1.6273661438615468,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004958588382973809,
|
|
"loss": 5.89,
|
|
"mean_token_accuracy": 0.1578308969736099,
|
|
"num_tokens": 3390597.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.950701808929443,
|
|
"epoch": 1.6327744726879394,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004957765241182826,
|
|
"loss": 5.8029,
|
|
"mean_token_accuracy": 0.16140344589948655,
|
|
"num_tokens": 3402618.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.664628601074218,
|
|
"epoch": 1.638182801514332,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004956934076367854,
|
|
"loss": 5.6986,
|
|
"mean_token_accuracy": 0.1686057612299919,
|
|
"num_tokens": 3413162.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.872065114974975,
|
|
"epoch": 1.6435911303407247,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495609489154935,
|
|
"loss": 5.7733,
|
|
"mean_token_accuracy": 0.1699829012155533,
|
|
"num_tokens": 3424331.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.8806146621704105,
|
|
"epoch": 1.6489994591671173,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004955247689776911,
|
|
"loss": 5.9017,
|
|
"mean_token_accuracy": 0.15300843268632888,
|
|
"num_tokens": 3434739.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.007542371749878,
|
|
"epoch": 1.65440778799351,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000495439247412927,
|
|
"loss": 5.9874,
|
|
"mean_token_accuracy": 0.15355408191680908,
|
|
"num_tokens": 3445984.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.71329665184021,
|
|
"epoch": 1.6598161168199026,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004953529247714285,
|
|
"loss": 5.705,
|
|
"mean_token_accuracy": 0.16740665286779405,
|
|
"num_tokens": 3456955.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.989834642410278,
|
|
"epoch": 1.6652244456462952,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004952658013668922,
|
|
"loss": 5.8546,
|
|
"mean_token_accuracy": 0.16043079793453216,
|
|
"num_tokens": 3468895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.721366119384766,
|
|
"epoch": 1.670632774472688,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004951778775159246,
|
|
"loss": 5.7625,
|
|
"mean_token_accuracy": 0.16552325934171677,
|
|
"num_tokens": 3480083.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.908278322219848,
|
|
"epoch": 1.6760411032990805,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004950891535380414,
|
|
"loss": 5.913,
|
|
"mean_token_accuracy": 0.15520185530185698,
|
|
"num_tokens": 3492242.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.804476690292359,
|
|
"epoch": 1.6814494321254734,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004949996297556656,
|
|
"loss": 5.7314,
|
|
"mean_token_accuracy": 0.16728926002979277,
|
|
"num_tokens": 3501926.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.754759883880615,
|
|
"epoch": 1.6868577609518658,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004949093064941268,
|
|
"loss": 5.7959,
|
|
"mean_token_accuracy": 0.16971654891967775,
|
|
"num_tokens": 3511421.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.902746534347534,
|
|
"epoch": 1.6922660897782587,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004948181840816601,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.16682299673557283,
|
|
"num_tokens": 3522033.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.776981258392334,
|
|
"epoch": 1.697674418604651,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004947262628494045,
|
|
"loss": 5.7762,
|
|
"mean_token_accuracy": 0.1661134257912636,
|
|
"num_tokens": 3533232.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.829985189437866,
|
|
"epoch": 1.703082747431044,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004946335431314018,
|
|
"loss": 5.7862,
|
|
"mean_token_accuracy": 0.16318220347166063,
|
|
"num_tokens": 3544904.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.848809146881104,
|
|
"epoch": 1.7084910762574363,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004945400252645959,
|
|
"loss": 5.8647,
|
|
"mean_token_accuracy": 0.16345469206571578,
|
|
"num_tokens": 3556545.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.818371438980103,
|
|
"epoch": 1.7138994050838292,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004944457095888308,
|
|
"loss": 5.7634,
|
|
"mean_token_accuracy": 0.16968527734279631,
|
|
"num_tokens": 3567361.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.850625848770141,
|
|
"epoch": 1.7193077339102216,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00049435059644685,
|
|
"loss": 5.7807,
|
|
"mean_token_accuracy": 0.1620098665356636,
|
|
"num_tokens": 3578943.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.839304876327515,
|
|
"epoch": 1.7247160627366145,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004942546861842947,
|
|
"loss": 5.7631,
|
|
"mean_token_accuracy": 0.16221804022789002,
|
|
"num_tokens": 3589868.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.768456363677979,
|
|
"epoch": 1.730124391563007,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004941579791497032,
|
|
"loss": 5.766,
|
|
"mean_token_accuracy": 0.16636517345905305,
|
|
"num_tokens": 3601245.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.874995183944702,
|
|
"epoch": 1.7355327203893998,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004940604756945091,
|
|
"loss": 5.8115,
|
|
"mean_token_accuracy": 0.16220273673534394,
|
|
"num_tokens": 3612168.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.753401184082032,
|
|
"epoch": 1.7409410492157922,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004939621761730399,
|
|
"loss": 5.7446,
|
|
"mean_token_accuracy": 0.1687006637454033,
|
|
"num_tokens": 3624699.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.846948862075806,
|
|
"epoch": 1.746349378042185,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004938630809425166,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.16605381965637206,
|
|
"num_tokens": 3635554.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.79630708694458,
|
|
"epoch": 1.7517577068685775,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004937631903630513,
|
|
"loss": 5.7771,
|
|
"mean_token_accuracy": 0.16420649141073226,
|
|
"num_tokens": 3646776.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.782458209991455,
|
|
"epoch": 1.7571660356949703,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004936625047976468,
|
|
"loss": 5.7101,
|
|
"mean_token_accuracy": 0.1681812509894371,
|
|
"num_tokens": 3657976.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.878973960876465,
|
|
"epoch": 1.7625743645213627,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004935610246121944,
|
|
"loss": 5.8478,
|
|
"mean_token_accuracy": 0.15901172906160355,
|
|
"num_tokens": 3668242.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.888310766220092,
|
|
"epoch": 1.7679826933477556,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004934587501754736,
|
|
"loss": 5.8561,
|
|
"mean_token_accuracy": 0.16048324555158616,
|
|
"num_tokens": 3679200.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.772162866592407,
|
|
"epoch": 1.7733910221741482,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004933556818591498,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.17035170644521713,
|
|
"num_tokens": 3690227.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.913128471374511,
|
|
"epoch": 1.7787993510005409,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932518200377736,
|
|
"loss": 5.7978,
|
|
"mean_token_accuracy": 0.16582833528518676,
|
|
"num_tokens": 3701821.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.7263421535491945,
|
|
"epoch": 1.7842076798269335,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004931471650887789,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.16091262698173522,
|
|
"num_tokens": 3712420.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.855186462402344,
|
|
"epoch": 1.7896160086533262,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004930417173924822,
|
|
"loss": 5.7914,
|
|
"mean_token_accuracy": 0.16447875946760177,
|
|
"num_tokens": 3722211.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.774428272247315,
|
|
"epoch": 1.7950243374797188,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004929354773320806,
|
|
"loss": 5.763,
|
|
"mean_token_accuracy": 0.15943235605955125,
|
|
"num_tokens": 3732822.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.866081047058105,
|
|
"epoch": 1.8004326663061114,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004928284452936506,
|
|
"loss": 5.8421,
|
|
"mean_token_accuracy": 0.16382588148117067,
|
|
"num_tokens": 3744262.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.791400051116943,
|
|
"epoch": 1.805840995132504,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004927206216661473,
|
|
"loss": 5.7213,
|
|
"mean_token_accuracy": 0.1668461889028549,
|
|
"num_tokens": 3755713.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.7758666515350345,
|
|
"epoch": 1.8112493239588967,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004926120068414015,
|
|
"loss": 5.7534,
|
|
"mean_token_accuracy": 0.1650092586874962,
|
|
"num_tokens": 3767179.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.7698955059051515,
|
|
"epoch": 1.8166576527852893,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004925026012141201,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.16847193390130996,
|
|
"num_tokens": 3777714.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.894411087036133,
|
|
"epoch": 1.822065981611682,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004923924051818832,
|
|
"loss": 5.7355,
|
|
"mean_token_accuracy": 0.1707264557480812,
|
|
"num_tokens": 3788092.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.727274370193482,
|
|
"epoch": 1.8274743104380746,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004922814191451435,
|
|
"loss": 5.7112,
|
|
"mean_token_accuracy": 0.1640962466597557,
|
|
"num_tokens": 3798080.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.843620157241821,
|
|
"epoch": 1.8328826392644673,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004921696435072245,
|
|
"loss": 5.7534,
|
|
"mean_token_accuracy": 0.17095378488302232,
|
|
"num_tokens": 3809547.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.798443365097046,
|
|
"epoch": 1.83829096809086,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004920570786743191,
|
|
"loss": 5.742,
|
|
"mean_token_accuracy": 0.16847866624593735,
|
|
"num_tokens": 3820681.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.768449497222901,
|
|
"epoch": 1.8436992969172525,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004919437250554882,
|
|
"loss": 5.7321,
|
|
"mean_token_accuracy": 0.1685010939836502,
|
|
"num_tokens": 3831598.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.7867199897766115,
|
|
"epoch": 1.8491076257436452,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004918295830626592,
|
|
"loss": 5.7813,
|
|
"mean_token_accuracy": 0.16900016963481904,
|
|
"num_tokens": 3843437.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.7356771469116214,
|
|
"epoch": 1.8545159545700378,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004917146531106242,
|
|
"loss": 5.6663,
|
|
"mean_token_accuracy": 0.16742974072694777,
|
|
"num_tokens": 3855387.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.836302280426025,
|
|
"epoch": 1.8599242833964305,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004915989356170389,
|
|
"loss": 5.8346,
|
|
"mean_token_accuracy": 0.15981754139065743,
|
|
"num_tokens": 3866999.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.778556537628174,
|
|
"epoch": 1.865332612222823,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004914824310024213,
|
|
"loss": 5.7481,
|
|
"mean_token_accuracy": 0.16680270582437515,
|
|
"num_tokens": 3877721.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.802711105346679,
|
|
"epoch": 1.870740941049216,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004913651396901489,
|
|
"loss": 5.8292,
|
|
"mean_token_accuracy": 0.15776942074298858,
|
|
"num_tokens": 3888831.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.798004865646362,
|
|
"epoch": 1.8761492698756084,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000491247062106459,
|
|
"loss": 5.7687,
|
|
"mean_token_accuracy": 0.16842874735593796,
|
|
"num_tokens": 3900389.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.762385892868042,
|
|
"epoch": 1.8815575987020012,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004911281986804455,
|
|
"loss": 5.7169,
|
|
"mean_token_accuracy": 0.16832340657711028,
|
|
"num_tokens": 3911641.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.643378114700317,
|
|
"epoch": 1.8869659275283936,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004910085498440587,
|
|
"loss": 5.6163,
|
|
"mean_token_accuracy": 0.17025550752878188,
|
|
"num_tokens": 3922877.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.7903642654418945,
|
|
"epoch": 1.8923742563547865,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004908881160321025,
|
|
"loss": 5.6854,
|
|
"mean_token_accuracy": 0.1704636514186859,
|
|
"num_tokens": 3933573.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.707590818405151,
|
|
"epoch": 1.897782585181179,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004907668976822338,
|
|
"loss": 5.6487,
|
|
"mean_token_accuracy": 0.17355832904577256,
|
|
"num_tokens": 3944339.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.822390556335449,
|
|
"epoch": 1.9031909140075718,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004906448952349603,
|
|
"loss": 5.7435,
|
|
"mean_token_accuracy": 0.1827266827225685,
|
|
"num_tokens": 3955459.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.681776237487793,
|
|
"epoch": 1.9085992428339642,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004905221091336392,
|
|
"loss": 5.6507,
|
|
"mean_token_accuracy": 0.170497328042984,
|
|
"num_tokens": 3966782.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.770271635055542,
|
|
"epoch": 1.914007571660357,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004903985398244753,
|
|
"loss": 5.7813,
|
|
"mean_token_accuracy": 0.16678330600261687,
|
|
"num_tokens": 3978794.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.878634405136109,
|
|
"epoch": 1.9194159004867495,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004902741877565201,
|
|
"loss": 5.7904,
|
|
"mean_token_accuracy": 0.1669964611530304,
|
|
"num_tokens": 3990838.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.765031194686889,
|
|
"epoch": 1.9248242293131423,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000490149053381669,
|
|
"loss": 5.8163,
|
|
"mean_token_accuracy": 0.16891681253910065,
|
|
"num_tokens": 4000630.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.817885828018189,
|
|
"epoch": 1.9302325581395348,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004900231371546608,
|
|
"loss": 5.6884,
|
|
"mean_token_accuracy": 0.1711932376027107,
|
|
"num_tokens": 4011433.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.759051847457886,
|
|
"epoch": 1.9356408869659276,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004898964395330753,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.16207005232572555,
|
|
"num_tokens": 4023671.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.770365142822266,
|
|
"epoch": 1.94104921579232,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004897689609773318,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.1738724648952484,
|
|
"num_tokens": 4034686.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.781462383270264,
|
|
"epoch": 1.946457544618713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004896407019506879,
|
|
"loss": 5.7426,
|
|
"mean_token_accuracy": 0.1736023336648941,
|
|
"num_tokens": 4046387.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.717059755325318,
|
|
"epoch": 1.9518658734451053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895116629192371,
|
|
"loss": 5.6605,
|
|
"mean_token_accuracy": 0.17057515382766725,
|
|
"num_tokens": 4056103.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.701630592346191,
|
|
"epoch": 1.9572742022714982,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004893818443519074,
|
|
"loss": 5.6277,
|
|
"mean_token_accuracy": 0.1794825553894043,
|
|
"num_tokens": 4066940.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.701979112625122,
|
|
"epoch": 1.9626825310978906,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048925124672046,
|
|
"loss": 5.7611,
|
|
"mean_token_accuracy": 0.1745072513818741,
|
|
"num_tokens": 4077232.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.801124811172485,
|
|
"epoch": 1.9680908599242835,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004891198704994868,
|
|
"loss": 5.8269,
|
|
"mean_token_accuracy": 0.1630716234445572,
|
|
"num_tokens": 4089322.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.815474891662598,
|
|
"epoch": 1.973499188750676,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 5.6626,
|
|
"mean_token_accuracy": 0.17274468690156936,
|
|
"num_tokens": 4099737.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.672778224945068,
|
|
"epoch": 1.9789075175770687,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004888547842014771,
|
|
"loss": 5.6799,
|
|
"mean_token_accuracy": 0.16707670241594313,
|
|
"num_tokens": 4110793.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.743105173110962,
|
|
"epoch": 1.9843158464034614,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004887210750877648,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.1723834455013275,
|
|
"num_tokens": 4121481.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.727146005630493,
|
|
"epoch": 1.989724175229854,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004885865893111716,
|
|
"loss": 5.6526,
|
|
"mean_token_accuracy": 0.17971729189157487,
|
|
"num_tokens": 4132701.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.734009075164795,
|
|
"epoch": 1.9951325040562466,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004884513273604195,
|
|
"loss": 5.7394,
|
|
"mean_token_accuracy": 0.17172908633947373,
|
|
"num_tokens": 4143764.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.72644387351142,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 0.0004883152897270502,
|
|
"loss": 5.7053,
|
|
"mean_token_accuracy": 0.16676922970347935,
|
|
"num_tokens": 4153778.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.748702573776245,
|
|
"epoch": 2.005408328826393,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004881784769054253,
|
|
"loss": 5.3936,
|
|
"mean_token_accuracy": 0.18091860860586167,
|
|
"num_tokens": 4165099.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.768449211120606,
|
|
"epoch": 2.0108166576527853,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004880408893927226,
|
|
"loss": 5.4829,
|
|
"mean_token_accuracy": 0.1814225897192955,
|
|
"num_tokens": 4178051.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.723996829986572,
|
|
"epoch": 2.016224986479178,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004879025276889355,
|
|
"loss": 5.3897,
|
|
"mean_token_accuracy": 0.1843633994460106,
|
|
"num_tokens": 4189987.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.766743803024292,
|
|
"epoch": 2.0216333153055706,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048776339229687077,
|
|
"loss": 5.4189,
|
|
"mean_token_accuracy": 0.18033189177513123,
|
|
"num_tokens": 4200999.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.573055553436279,
|
|
"epoch": 2.0270416441319634,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048762348372214683,
|
|
"loss": 5.3626,
|
|
"mean_token_accuracy": 0.17809585779905318,
|
|
"num_tokens": 4212122.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.7209272384643555,
|
|
"epoch": 2.032449972958356,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048748280247319165,
|
|
"loss": 5.4228,
|
|
"mean_token_accuracy": 0.18461056053638458,
|
|
"num_tokens": 4222580.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.6292177677154545,
|
|
"epoch": 2.0378583017847487,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004873413490612412,
|
|
"loss": 5.2663,
|
|
"mean_token_accuracy": 0.18493057936429977,
|
|
"num_tokens": 4233756.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.656900072097779,
|
|
"epoch": 2.043266630611141,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048719912400033757,
|
|
"loss": 5.4161,
|
|
"mean_token_accuracy": 0.17914021164178848,
|
|
"num_tokens": 4245625.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.780725002288818,
|
|
"epoch": 2.048674959437534,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000487056127807327,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.17479688674211502,
|
|
"num_tokens": 4257788.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.6406174659729,
|
|
"epoch": 2.0540832882639264,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048691236100185795,
|
|
"loss": 5.3735,
|
|
"mean_token_accuracy": 0.18234946876764296,
|
|
"num_tokens": 4269641.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.679793357849121,
|
|
"epoch": 2.0594916170903192,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004867678241063792,
|
|
"loss": 5.427,
|
|
"mean_token_accuracy": 0.1848498284816742,
|
|
"num_tokens": 4280952.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.694111633300781,
|
|
"epoch": 2.0648999459167117,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048662251764613833,
|
|
"loss": 5.3791,
|
|
"mean_token_accuracy": 0.1879727154970169,
|
|
"num_tokens": 4294371.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.510202121734619,
|
|
"epoch": 2.0703082747431045,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004864764421491793,
|
|
"loss": 5.3847,
|
|
"mean_token_accuracy": 0.1767863377928734,
|
|
"num_tokens": 4307112.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.711728143692016,
|
|
"epoch": 2.075716603569497,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004863295981463408,
|
|
"loss": 5.3529,
|
|
"mean_token_accuracy": 0.18180871307849883,
|
|
"num_tokens": 4317199.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.588171482086182,
|
|
"epoch": 2.08112493239589,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004861819861712543,
|
|
"loss": 5.3677,
|
|
"mean_token_accuracy": 0.18255759179592132,
|
|
"num_tokens": 4328546.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.7120726108551025,
|
|
"epoch": 2.086533261222282,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048603360676034203,
|
|
"loss": 5.4257,
|
|
"mean_token_accuracy": 0.18246274292469025,
|
|
"num_tokens": 4338795.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.55946455001831,
|
|
"epoch": 2.091941590048675,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048588446045281505,
|
|
"loss": 5.4013,
|
|
"mean_token_accuracy": 0.1848064348101616,
|
|
"num_tokens": 4350373.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.613817882537842,
|
|
"epoch": 2.0973499188750675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004857345477906714,
|
|
"loss": 5.3848,
|
|
"mean_token_accuracy": 0.18297084271907807,
|
|
"num_tokens": 4362801.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.709928321838379,
|
|
"epoch": 2.1027582477014604,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048558386931869417,
|
|
"loss": 5.4869,
|
|
"mean_token_accuracy": 0.18310190439224244,
|
|
"num_tokens": 4375782.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.575471258163452,
|
|
"epoch": 2.1081665765278528,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004854324255844492,
|
|
"loss": 5.3756,
|
|
"mean_token_accuracy": 0.18223719894886017,
|
|
"num_tokens": 4385580.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.65678300857544,
|
|
"epoch": 2.1135749053542456,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048528021713828336,
|
|
"loss": 5.3938,
|
|
"mean_token_accuracy": 0.1827685907483101,
|
|
"num_tokens": 4397434.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.660488605499268,
|
|
"epoch": 2.118983234180638,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004851272445333225,
|
|
"loss": 5.4412,
|
|
"mean_token_accuracy": 0.1808636575937271,
|
|
"num_tokens": 4407888.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.57245717048645,
|
|
"epoch": 2.124391563007031,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048497350832546937,
|
|
"loss": 5.3489,
|
|
"mean_token_accuracy": 0.19036678671836854,
|
|
"num_tokens": 4420010.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.63389401435852,
|
|
"epoch": 2.1297998918334233,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004848190090734018,
|
|
"loss": 5.3931,
|
|
"mean_token_accuracy": 0.18125024288892747,
|
|
"num_tokens": 4430730.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.567873191833496,
|
|
"epoch": 2.135208220659816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048466374733857035,
|
|
"loss": 5.3847,
|
|
"mean_token_accuracy": 0.18848742842674254,
|
|
"num_tokens": 4442296.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.64949598312378,
|
|
"epoch": 2.1406165494862086,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004845077236851966,
|
|
"loss": 5.4189,
|
|
"mean_token_accuracy": 0.18026936799287796,
|
|
"num_tokens": 4453996.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.557392597198486,
|
|
"epoch": 2.1460248783126015,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048435093868027076,
|
|
"loss": 5.4207,
|
|
"mean_token_accuracy": 0.18153801262378694,
|
|
"num_tokens": 4465779.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.549536085128784,
|
|
"epoch": 2.151433207138994,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048419339289354995,
|
|
"loss": 5.3565,
|
|
"mean_token_accuracy": 0.19019488394260406,
|
|
"num_tokens": 4477113.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.617552423477173,
|
|
"epoch": 2.1568415359653867,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000484035086897556,
|
|
"loss": 5.2979,
|
|
"mean_token_accuracy": 0.1857837811112404,
|
|
"num_tokens": 4487920.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.642735195159912,
|
|
"epoch": 2.162249864791779,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004838760212675732,
|
|
"loss": 5.5028,
|
|
"mean_token_accuracy": 0.17568326443433763,
|
|
"num_tokens": 4499049.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.162249864791779,
|
|
"eval_entropy": 5.524916348674081,
|
|
"eval_loss": 6.017584323883057,
|
|
"eval_mean_token_accuracy": 0.16565372737293893,
|
|
"eval_num_tokens": 4499049.0,
|
|
"eval_runtime": 2.7188,
|
|
"eval_samples_per_second": 1292.867,
|
|
"eval_steps_per_second": 161.838,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.604034423828125,
|
|
"epoch": 2.167658193618172,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004837161965816464,
|
|
"loss": 5.3194,
|
|
"mean_token_accuracy": 0.18912744671106338,
|
|
"num_tokens": 4508944.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.575043106079102,
|
|
"epoch": 2.1730665224445644,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048355561342057886,
|
|
"loss": 5.3506,
|
|
"mean_token_accuracy": 0.18395831286907197,
|
|
"num_tokens": 4518845.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.485327100753784,
|
|
"epoch": 2.1784748512709573,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048339427236793024,
|
|
"loss": 5.4136,
|
|
"mean_token_accuracy": 0.17852067798376084,
|
|
"num_tokens": 4530416.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.687785625457764,
|
|
"epoch": 2.1838831800973497,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004832321740100143,
|
|
"loss": 5.3827,
|
|
"mean_token_accuracy": 0.18364240527153014,
|
|
"num_tokens": 4542630.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.506225728988648,
|
|
"epoch": 2.1892915089237426,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048306931893589666,
|
|
"loss": 5.3769,
|
|
"mean_token_accuracy": 0.1830895036458969,
|
|
"num_tokens": 4553270.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.4672914981842045,
|
|
"epoch": 2.194699837750135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048290570773739327,
|
|
"loss": 5.2511,
|
|
"mean_token_accuracy": 0.19056000113487243,
|
|
"num_tokens": 4564647.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.688305234909057,
|
|
"epoch": 2.200108166576528,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048274134100906747,
|
|
"loss": 5.4387,
|
|
"mean_token_accuracy": 0.17347942888736725,
|
|
"num_tokens": 4575507.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.58884973526001,
|
|
"epoch": 2.2055164954029207,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048257621934822835,
|
|
"loss": 5.4172,
|
|
"mean_token_accuracy": 0.18910931646823884,
|
|
"num_tokens": 4586761.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.551853990554809,
|
|
"epoch": 2.210924824229313,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004824103433549285,
|
|
"loss": 5.3407,
|
|
"mean_token_accuracy": 0.18681126683950425,
|
|
"num_tokens": 4597689.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.5847193717956545,
|
|
"epoch": 2.2163331530557056,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004822437136319616,
|
|
"loss": 5.3805,
|
|
"mean_token_accuracy": 0.1807395339012146,
|
|
"num_tokens": 4608886.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.582241678237915,
|
|
"epoch": 2.2217414818820984,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004820763307848604,
|
|
"loss": 5.4395,
|
|
"mean_token_accuracy": 0.17519450187683105,
|
|
"num_tokens": 4619989.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.5418031215667725,
|
|
"epoch": 2.2271498107084913,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048190819542189456,
|
|
"loss": 5.3161,
|
|
"mean_token_accuracy": 0.19040639996528624,
|
|
"num_tokens": 4630218.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.512099266052246,
|
|
"epoch": 2.2325581395348837,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048173930815406847,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.1735458716750145,
|
|
"num_tokens": 4642024.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.616588354110718,
|
|
"epoch": 2.2379664683612766,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004815696695951187,
|
|
"loss": 5.406,
|
|
"mean_token_accuracy": 0.18752684891223909,
|
|
"num_tokens": 4653370.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.567294025421143,
|
|
"epoch": 2.243374797187669,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004813992803615123,
|
|
"loss": 5.4055,
|
|
"mean_token_accuracy": 0.1839067354798317,
|
|
"num_tokens": 4664522.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.642352819442749,
|
|
"epoch": 2.248783126014062,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048122814107244414,
|
|
"loss": 5.4551,
|
|
"mean_token_accuracy": 0.18220221549272536,
|
|
"num_tokens": 4676542.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.495743179321289,
|
|
"epoch": 2.2541914548404542,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048105625234983466,
|
|
"loss": 5.3812,
|
|
"mean_token_accuracy": 0.18212060779333114,
|
|
"num_tokens": 4687810.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.6320196151733395,
|
|
"epoch": 2.259599783666847,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048088361481832807,
|
|
"loss": 5.3594,
|
|
"mean_token_accuracy": 0.18495688140392302,
|
|
"num_tokens": 4697661.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.454736804962158,
|
|
"epoch": 2.2650081124932395,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048071022910528934,
|
|
"loss": 5.3165,
|
|
"mean_token_accuracy": 0.18503704965114592,
|
|
"num_tokens": 4708585.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.6714404106140135,
|
|
"epoch": 2.2704164413196324,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048053609584080274,
|
|
"loss": 5.4704,
|
|
"mean_token_accuracy": 0.18292915672063828,
|
|
"num_tokens": 4719848.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.63643217086792,
|
|
"epoch": 2.275824770146025,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048036121565766876,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.174479441344738,
|
|
"num_tokens": 4731474.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.570285511016846,
|
|
"epoch": 2.2812330989724177,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004801855891914026,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.18509435057640075,
|
|
"num_tokens": 4743081.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.564574766159057,
|
|
"epoch": 2.28664142779881,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048000921708023117,
|
|
"loss": 5.3924,
|
|
"mean_token_accuracy": 0.17817198634147643,
|
|
"num_tokens": 4754829.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.56991925239563,
|
|
"epoch": 2.292049756625203,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004798320999650911,
|
|
"loss": 5.361,
|
|
"mean_token_accuracy": 0.1823549672961235,
|
|
"num_tokens": 4765522.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.5502685546875,
|
|
"epoch": 2.2974580854515954,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047965423848962636,
|
|
"loss": 5.3533,
|
|
"mean_token_accuracy": 0.1851808175444603,
|
|
"num_tokens": 4776246.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.533322668075561,
|
|
"epoch": 2.302866414277988,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000479475633300186,
|
|
"loss": 5.4117,
|
|
"mean_token_accuracy": 0.18015872836112976,
|
|
"num_tokens": 4787595.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.652274560928345,
|
|
"epoch": 2.3082747431043806,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047929628504582173,
|
|
"loss": 5.465,
|
|
"mean_token_accuracy": 0.17422494292259216,
|
|
"num_tokens": 4798877.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.565356683731079,
|
|
"epoch": 2.3136830719307735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004791161943782856,
|
|
"loss": 5.4572,
|
|
"mean_token_accuracy": 0.18433694094419478,
|
|
"num_tokens": 4809542.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.696715927124023,
|
|
"epoch": 2.319091400757166,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047893536195202726,
|
|
"loss": 5.5696,
|
|
"mean_token_accuracy": 0.17354034632444382,
|
|
"num_tokens": 4820982.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.489514541625977,
|
|
"epoch": 2.3244997295835588,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004787537884241923,
|
|
"loss": 5.4121,
|
|
"mean_token_accuracy": 0.18239568322896957,
|
|
"num_tokens": 4831543.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.772100210189819,
|
|
"epoch": 2.329908058409951,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047857147445461943,
|
|
"loss": 5.4564,
|
|
"mean_token_accuracy": 0.1802772104740143,
|
|
"num_tokens": 4843221.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.442333984375,
|
|
"epoch": 2.335316387236344,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004783884207058379,
|
|
"loss": 5.338,
|
|
"mean_token_accuracy": 0.18246717751026154,
|
|
"num_tokens": 4855201.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.627194118499756,
|
|
"epoch": 2.3407247160627365,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047820462784306555,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.18327535390853883,
|
|
"num_tokens": 4866387.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.51273627281189,
|
|
"epoch": 2.3461330448891293,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000478020096534206,
|
|
"loss": 5.3328,
|
|
"mean_token_accuracy": 0.18196618258953096,
|
|
"num_tokens": 4877702.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.599471759796143,
|
|
"epoch": 2.3515413737155217,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047783482744984644,
|
|
"loss": 5.4109,
|
|
"mean_token_accuracy": 0.18699432611465455,
|
|
"num_tokens": 4888652.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.61759991645813,
|
|
"epoch": 2.3569497025419146,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047764882126325527,
|
|
"loss": 5.524,
|
|
"mean_token_accuracy": 0.17653899490833283,
|
|
"num_tokens": 4900995.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.505554151535034,
|
|
"epoch": 2.362358031368307,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004774620786503793,
|
|
"loss": 5.2946,
|
|
"mean_token_accuracy": 0.19195163995027542,
|
|
"num_tokens": 4912173.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.440061807632446,
|
|
"epoch": 2.3677663601947,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004772746002898417,
|
|
"loss": 5.3204,
|
|
"mean_token_accuracy": 0.1913541927933693,
|
|
"num_tokens": 4923149.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.566332292556763,
|
|
"epoch": 2.3731746890210923,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047708638686293903,
|
|
"loss": 5.3236,
|
|
"mean_token_accuracy": 0.18964710980653762,
|
|
"num_tokens": 4934156.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.595124626159668,
|
|
"epoch": 2.378583017847485,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047689743905363963,
|
|
"loss": 5.5318,
|
|
"mean_token_accuracy": 0.1755328118801117,
|
|
"num_tokens": 4946690.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.581760025024414,
|
|
"epoch": 2.383991346673878,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004767077575485801,
|
|
"loss": 5.4237,
|
|
"mean_token_accuracy": 0.19032906591892243,
|
|
"num_tokens": 4958208.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.522891426086426,
|
|
"epoch": 2.3893996755002704,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004765173430370635,
|
|
"loss": 5.3611,
|
|
"mean_token_accuracy": 0.18898999094963073,
|
|
"num_tokens": 4968815.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.581397438049317,
|
|
"epoch": 2.394808004326663,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004763261962110567,
|
|
"loss": 5.3611,
|
|
"mean_token_accuracy": 0.18661250919103622,
|
|
"num_tokens": 4979468.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.552277040481568,
|
|
"epoch": 2.4002163331530557,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004761343177651875,
|
|
"loss": 5.4611,
|
|
"mean_token_accuracy": 0.17783761769533157,
|
|
"num_tokens": 4990841.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.698723840713501,
|
|
"epoch": 2.4056246619794486,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004759417083967428,
|
|
"loss": 5.4771,
|
|
"mean_token_accuracy": 0.1798073634505272,
|
|
"num_tokens": 5001567.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.483736896514893,
|
|
"epoch": 2.411032990805841,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047574836880566544,
|
|
"loss": 5.3586,
|
|
"mean_token_accuracy": 0.1799778699874878,
|
|
"num_tokens": 5013219.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.54263825416565,
|
|
"epoch": 2.4164413196322334,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004755542996945519,
|
|
"loss": 5.4762,
|
|
"mean_token_accuracy": 0.17981167882680893,
|
|
"num_tokens": 5025771.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.5824333190917965,
|
|
"epoch": 2.4218496484586263,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047535950176864976,
|
|
"loss": 5.4361,
|
|
"mean_token_accuracy": 0.1832021951675415,
|
|
"num_tokens": 5038205.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.606994247436523,
|
|
"epoch": 2.427257977285019,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047516397573585525,
|
|
"loss": 5.4078,
|
|
"mean_token_accuracy": 0.1840317055583,
|
|
"num_tokens": 5048428.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.55131311416626,
|
|
"epoch": 2.4326663061114115,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047496772230671034,
|
|
"loss": 5.3719,
|
|
"mean_token_accuracy": 0.18645645827054977,
|
|
"num_tokens": 5058391.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.52794394493103,
|
|
"epoch": 2.4380746349378044,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004747707421944004,
|
|
"loss": 5.4058,
|
|
"mean_token_accuracy": 0.18165958523750306,
|
|
"num_tokens": 5069832.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.566888332366943,
|
|
"epoch": 2.443482963764197,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004745730361147517,
|
|
"loss": 5.3825,
|
|
"mean_token_accuracy": 0.19324786514043807,
|
|
"num_tokens": 5080198.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.512354946136474,
|
|
"epoch": 2.4488912925905897,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004743746047862286,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.1786731258034706,
|
|
"num_tokens": 5090817.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.495129728317261,
|
|
"epoch": 2.454299621416982,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004741754489299309,
|
|
"loss": 5.3458,
|
|
"mean_token_accuracy": 0.18243062496185303,
|
|
"num_tokens": 5102699.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.50686993598938,
|
|
"epoch": 2.459707950243375,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047397556926959155,
|
|
"loss": 5.3458,
|
|
"mean_token_accuracy": 0.18208036720752716,
|
|
"num_tokens": 5113922.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.574401044845581,
|
|
"epoch": 2.4651162790697674,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047377496653157374,
|
|
"loss": 5.4411,
|
|
"mean_token_accuracy": 0.18198768198490142,
|
|
"num_tokens": 5124920.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.544427347183228,
|
|
"epoch": 2.4705246078961602,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047357364144486806,
|
|
"loss": 5.432,
|
|
"mean_token_accuracy": 0.1753288611769676,
|
|
"num_tokens": 5136304.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.587402820587158,
|
|
"epoch": 2.4759329367225527,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047337159474109057,
|
|
"loss": 5.5094,
|
|
"mean_token_accuracy": 0.1784786969423294,
|
|
"num_tokens": 5148101.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.622994709014892,
|
|
"epoch": 2.4813412655489455,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047316882715447934,
|
|
"loss": 5.4272,
|
|
"mean_token_accuracy": 0.18538385778665542,
|
|
"num_tokens": 5159186.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.511073350906372,
|
|
"epoch": 2.486749594375338,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004729653394218922,
|
|
"loss": 5.4135,
|
|
"mean_token_accuracy": 0.1818036988377571,
|
|
"num_tokens": 5170542.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.5297465324401855,
|
|
"epoch": 2.492157923201731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047276113228280427,
|
|
"loss": 5.4959,
|
|
"mean_token_accuracy": 0.1765660211443901,
|
|
"num_tokens": 5182685.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.614099979400635,
|
|
"epoch": 2.497566252028123,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004725562064793045,
|
|
"loss": 5.3181,
|
|
"mean_token_accuracy": 0.18960487246513366,
|
|
"num_tokens": 5192622.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.534894275665283,
|
|
"epoch": 2.502974580854516,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047235056275609397,
|
|
"loss": 5.4906,
|
|
"mean_token_accuracy": 0.17734816819429397,
|
|
"num_tokens": 5205068.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.525508260726928,
|
|
"epoch": 2.5083829096809085,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004721442018604822,
|
|
"loss": 5.3922,
|
|
"mean_token_accuracy": 0.18073588758707046,
|
|
"num_tokens": 5216049.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.484037590026856,
|
|
"epoch": 2.5137912385073014,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004719371245423853,
|
|
"loss": 5.3479,
|
|
"mean_token_accuracy": 0.18653271347284317,
|
|
"num_tokens": 5227024.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.583204650878907,
|
|
"epoch": 2.5191995673336938,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047172933155432273,
|
|
"loss": 5.4864,
|
|
"mean_token_accuracy": 0.18093232810497284,
|
|
"num_tokens": 5238862.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.62364821434021,
|
|
"epoch": 2.5246078961600866,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047152082365141466,
|
|
"loss": 5.4943,
|
|
"mean_token_accuracy": 0.17634447962045668,
|
|
"num_tokens": 5250020.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.616364765167236,
|
|
"epoch": 2.530016224986479,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004713116015913794,
|
|
"loss": 5.4216,
|
|
"mean_token_accuracy": 0.17771199941635132,
|
|
"num_tokens": 5260281.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.5874124526977536,
|
|
"epoch": 2.535424553812872,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047110166613453016,
|
|
"loss": 5.5424,
|
|
"mean_token_accuracy": 0.17905232012271882,
|
|
"num_tokens": 5272909.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.4743036270141605,
|
|
"epoch": 2.5408328826392643,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047089101804377305,
|
|
"loss": 5.3546,
|
|
"mean_token_accuracy": 0.19149015545845033,
|
|
"num_tokens": 5285629.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.654702043533325,
|
|
"epoch": 2.546241211465657,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004706796580846037,
|
|
"loss": 5.4197,
|
|
"mean_token_accuracy": 0.18404433727264405,
|
|
"num_tokens": 5296104.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.680897045135498,
|
|
"epoch": 2.5516495402920496,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047046758702510465,
|
|
"loss": 5.5471,
|
|
"mean_token_accuracy": 0.172174771130085,
|
|
"num_tokens": 5307148.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.514196300506592,
|
|
"epoch": 2.5570578691184425,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047025480563594264,
|
|
"loss": 5.3755,
|
|
"mean_token_accuracy": 0.17709315121173858,
|
|
"num_tokens": 5317878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.597324991226197,
|
|
"epoch": 2.5624661979448353,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047004131469036576,
|
|
"loss": 5.4643,
|
|
"mean_token_accuracy": 0.18067093342542648,
|
|
"num_tokens": 5329603.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.513444423675537,
|
|
"epoch": 2.5678745267712277,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046982711496420057,
|
|
"loss": 5.4229,
|
|
"mean_token_accuracy": 0.18200160264968873,
|
|
"num_tokens": 5340611.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.552023315429688,
|
|
"epoch": 2.57328285559762,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046961220723584925,
|
|
"loss": 5.3261,
|
|
"mean_token_accuracy": 0.18667233735322952,
|
|
"num_tokens": 5352017.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.510668420791626,
|
|
"epoch": 2.578691184424013,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046939659228628725,
|
|
"loss": 5.4074,
|
|
"mean_token_accuracy": 0.18748226165771484,
|
|
"num_tokens": 5362482.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.493450450897217,
|
|
"epoch": 2.584099513250406,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046918027089905956,
|
|
"loss": 5.3891,
|
|
"mean_token_accuracy": 0.18690418750047683,
|
|
"num_tokens": 5374074.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.460361528396606,
|
|
"epoch": 2.5895078420767983,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046896324386027873,
|
|
"loss": 5.2081,
|
|
"mean_token_accuracy": 0.19546242356300353,
|
|
"num_tokens": 5384650.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.387950086593628,
|
|
"epoch": 2.5949161709031907,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046874551195862145,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.18182933032512666,
|
|
"num_tokens": 5395184.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.488055944442749,
|
|
"epoch": 2.6003244997295836,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000468527075985326,
|
|
"loss": 5.2814,
|
|
"mean_token_accuracy": 0.1902179628610611,
|
|
"num_tokens": 5405947.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.505107736587524,
|
|
"epoch": 2.6057328285559764,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046830793673418924,
|
|
"loss": 5.3761,
|
|
"mean_token_accuracy": 0.1837602362036705,
|
|
"num_tokens": 5418237.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.459736585617065,
|
|
"epoch": 2.611141157382369,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046808809500156355,
|
|
"loss": 5.4371,
|
|
"mean_token_accuracy": 0.17828247994184493,
|
|
"num_tokens": 5429336.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.564971303939819,
|
|
"epoch": 2.6165494862087613,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004678675515863545,
|
|
"loss": 5.401,
|
|
"mean_token_accuracy": 0.17862969338893891,
|
|
"num_tokens": 5440257.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.631397151947022,
|
|
"epoch": 2.621957815035154,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004676463072900173,
|
|
"loss": 5.5125,
|
|
"mean_token_accuracy": 0.1811617508530617,
|
|
"num_tokens": 5453079.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.480101776123047,
|
|
"epoch": 2.627366143861547,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046742436291655415,
|
|
"loss": 5.386,
|
|
"mean_token_accuracy": 0.18305414617061616,
|
|
"num_tokens": 5464551.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.573856782913208,
|
|
"epoch": 2.6327744726879394,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046720171927251155,
|
|
"loss": 5.4218,
|
|
"mean_token_accuracy": 0.1902390941977501,
|
|
"num_tokens": 5474797.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.505781888961792,
|
|
"epoch": 2.638182801514332,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004669783771669769,
|
|
"loss": 5.3653,
|
|
"mean_token_accuracy": 0.18573900908231736,
|
|
"num_tokens": 5485760.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.551184606552124,
|
|
"epoch": 2.6435911303407247,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000466754337411576,
|
|
"loss": 5.3589,
|
|
"mean_token_accuracy": 0.19332992434501647,
|
|
"num_tokens": 5496182.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.498781967163086,
|
|
"epoch": 2.6489994591671175,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046652960082046985,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.18994317054748536,
|
|
"num_tokens": 5506848.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.619220304489136,
|
|
"epoch": 2.65440778799351,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046630416821035164,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.18141047060489654,
|
|
"num_tokens": 5518430.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.582853746414185,
|
|
"epoch": 2.6598161168199024,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004660780404004441,
|
|
"loss": 5.4211,
|
|
"mean_token_accuracy": 0.1833420991897583,
|
|
"num_tokens": 5529741.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.4642658710479735,
|
|
"epoch": 2.6652244456462952,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046585121821249623,
|
|
"loss": 5.4159,
|
|
"mean_token_accuracy": 0.18699712604284285,
|
|
"num_tokens": 5541835.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.553615570068359,
|
|
"epoch": 2.670632774472688,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046562370247078025,
|
|
"loss": 5.3128,
|
|
"mean_token_accuracy": 0.18640302270650863,
|
|
"num_tokens": 5552177.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.5714109420776365,
|
|
"epoch": 2.6760411032990805,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000465395494002089,
|
|
"loss": 5.5325,
|
|
"mean_token_accuracy": 0.17443085312843323,
|
|
"num_tokens": 5564440.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.559375143051147,
|
|
"epoch": 2.6814494321254734,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004651665936357325,
|
|
"loss": 5.3458,
|
|
"mean_token_accuracy": 0.1874310165643692,
|
|
"num_tokens": 5574905.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.50521411895752,
|
|
"epoch": 2.686857760951866,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004649370022035354,
|
|
"loss": 5.4066,
|
|
"mean_token_accuracy": 0.179879067838192,
|
|
"num_tokens": 5585583.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.494187307357788,
|
|
"epoch": 2.6922660897782587,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004647067205398332,
|
|
"loss": 5.4063,
|
|
"mean_token_accuracy": 0.1847532644867897,
|
|
"num_tokens": 5597077.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.517140531539917,
|
|
"epoch": 2.697674418604651,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046447574948147027,
|
|
"loss": 5.2459,
|
|
"mean_token_accuracy": 0.19212979227304458,
|
|
"num_tokens": 5607904.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.440673208236694,
|
|
"epoch": 2.703082747431044,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046424408986779587,
|
|
"loss": 5.433,
|
|
"mean_token_accuracy": 0.18221829384565352,
|
|
"num_tokens": 5619611.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.57274022102356,
|
|
"epoch": 2.7084910762574363,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004640117425406615,
|
|
"loss": 5.4387,
|
|
"mean_token_accuracy": 0.18775810301303864,
|
|
"num_tokens": 5630365.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.57887897491455,
|
|
"epoch": 2.713899405083829,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000463778708344418,
|
|
"loss": 5.3936,
|
|
"mean_token_accuracy": 0.18564871847629547,
|
|
"num_tokens": 5641008.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.479671478271484,
|
|
"epoch": 2.7193077339102216,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046354498812591207,
|
|
"loss": 5.43,
|
|
"mean_token_accuracy": 0.18488810509443282,
|
|
"num_tokens": 5652633.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.496960258483886,
|
|
"epoch": 2.7247160627366145,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046331058273448364,
|
|
"loss": 5.4384,
|
|
"mean_token_accuracy": 0.17253482937812806,
|
|
"num_tokens": 5664108.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.56645450592041,
|
|
"epoch": 2.730124391563007,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004630754930219624,
|
|
"loss": 5.3922,
|
|
"mean_token_accuracy": 0.1884406566619873,
|
|
"num_tokens": 5675701.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.412143135070801,
|
|
"epoch": 2.7355327203893998,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004628397198426648,
|
|
"loss": 5.3262,
|
|
"mean_token_accuracy": 0.18637054562568664,
|
|
"num_tokens": 5687292.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.520063066482544,
|
|
"epoch": 2.740941049215792,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046260326405339126,
|
|
"loss": 5.3895,
|
|
"mean_token_accuracy": 0.1865263342857361,
|
|
"num_tokens": 5698905.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.421196746826172,
|
|
"epoch": 2.746349378042185,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004623661265134226,
|
|
"loss": 5.2612,
|
|
"mean_token_accuracy": 0.18892876654863358,
|
|
"num_tokens": 5709678.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.513219022750855,
|
|
"epoch": 2.7517577068685775,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004621283080845172,
|
|
"loss": 5.3838,
|
|
"mean_token_accuracy": 0.19197018146514894,
|
|
"num_tokens": 5720460.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.468096828460693,
|
|
"epoch": 2.7571660356949703,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046188980963090783,
|
|
"loss": 5.4211,
|
|
"mean_token_accuracy": 0.17634700685739518,
|
|
"num_tokens": 5731701.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.484721851348877,
|
|
"epoch": 2.7625743645213627,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046165063201929825,
|
|
"loss": 5.4175,
|
|
"mean_token_accuracy": 0.1817382499575615,
|
|
"num_tokens": 5741512.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.508628082275391,
|
|
"epoch": 2.7679826933477556,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046141077611886074,
|
|
"loss": 5.3293,
|
|
"mean_token_accuracy": 0.1865936040878296,
|
|
"num_tokens": 5751882.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.462914276123047,
|
|
"epoch": 2.773391022174148,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004611702428012319,
|
|
"loss": 5.4498,
|
|
"mean_token_accuracy": 0.17966431975364686,
|
|
"num_tokens": 5761989.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.497539472579956,
|
|
"epoch": 2.778799351000541,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046092903294051057,
|
|
"loss": 5.3486,
|
|
"mean_token_accuracy": 0.18025930225849152,
|
|
"num_tokens": 5772532.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.552466154098511,
|
|
"epoch": 2.7842076798269337,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046068714741325385,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.18622040301561354,
|
|
"num_tokens": 5784152.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.559518909454345,
|
|
"epoch": 2.789616008653326,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004604445870984743,
|
|
"loss": 5.4234,
|
|
"mean_token_accuracy": 0.18217067122459413,
|
|
"num_tokens": 5793748.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.43218150138855,
|
|
"epoch": 2.7950243374797186,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004602013528776367,
|
|
"loss": 5.3697,
|
|
"mean_token_accuracy": 0.18369193822145463,
|
|
"num_tokens": 5804371.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.542735147476196,
|
|
"epoch": 2.8004326663061114,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004599574456346548,
|
|
"loss": 5.376,
|
|
"mean_token_accuracy": 0.1859410285949707,
|
|
"num_tokens": 5815470.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.579844665527344,
|
|
"epoch": 2.8058409951325043,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045971286625588796,
|
|
"loss": 5.4718,
|
|
"mean_token_accuracy": 0.18081397265195848,
|
|
"num_tokens": 5826037.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.496999359130859,
|
|
"epoch": 2.8112493239588967,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004594676156301383,
|
|
"loss": 5.3356,
|
|
"mean_token_accuracy": 0.18775255531072615,
|
|
"num_tokens": 5836468.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.434178256988526,
|
|
"epoch": 2.816657652785289,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045922169464864725,
|
|
"loss": 5.241,
|
|
"mean_token_accuracy": 0.18995754718780516,
|
|
"num_tokens": 5846371.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.343031692504883,
|
|
"epoch": 2.822065981611682,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000458975104205092,
|
|
"loss": 5.3279,
|
|
"mean_token_accuracy": 0.19117088466882706,
|
|
"num_tokens": 5856890.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.499229478836059,
|
|
"epoch": 2.827474310438075,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045872784519558304,
|
|
"loss": 5.3797,
|
|
"mean_token_accuracy": 0.1885200023651123,
|
|
"num_tokens": 5868075.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.582245922088623,
|
|
"epoch": 2.8328826392644673,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004584799185186599,
|
|
"loss": 5.4675,
|
|
"mean_token_accuracy": 0.18507401570677756,
|
|
"num_tokens": 5879579.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.481081104278564,
|
|
"epoch": 2.8382909680908597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045823132507528903,
|
|
"loss": 5.3213,
|
|
"mean_token_accuracy": 0.18382314443588257,
|
|
"num_tokens": 5890648.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.424989843368531,
|
|
"epoch": 2.8436992969172525,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004579820657688593,
|
|
"loss": 5.3664,
|
|
"mean_token_accuracy": 0.18504354506731033,
|
|
"num_tokens": 5902914.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.520438480377197,
|
|
"epoch": 2.8491076257436454,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004577321415051798,
|
|
"loss": 5.3963,
|
|
"mean_token_accuracy": 0.17876323461532592,
|
|
"num_tokens": 5915801.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.574653148651123,
|
|
"epoch": 2.854515954570038,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045748155319247564,
|
|
"loss": 5.4709,
|
|
"mean_token_accuracy": 0.17966972887516022,
|
|
"num_tokens": 5928973.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.530942440032959,
|
|
"epoch": 2.8599242833964302,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004572303017413856,
|
|
"loss": 5.4778,
|
|
"mean_token_accuracy": 0.17940681278705597,
|
|
"num_tokens": 5939720.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.567597436904907,
|
|
"epoch": 2.865332612222823,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004569783880649578,
|
|
"loss": 5.4319,
|
|
"mean_token_accuracy": 0.18098780512809753,
|
|
"num_tokens": 5951956.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.600171613693237,
|
|
"epoch": 2.870740941049216,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004567258130786474,
|
|
"loss": 5.5704,
|
|
"mean_token_accuracy": 0.17420514225959777,
|
|
"num_tokens": 5963735.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.549901247024536,
|
|
"epoch": 2.8761492698756084,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045647257770031215,
|
|
"loss": 5.3269,
|
|
"mean_token_accuracy": 0.18786317855119705,
|
|
"num_tokens": 5974567.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.49153995513916,
|
|
"epoch": 2.8815575987020012,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045621868285021023,
|
|
"loss": 5.4557,
|
|
"mean_token_accuracy": 0.18465250581502915,
|
|
"num_tokens": 5985800.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.470799446105957,
|
|
"epoch": 2.8869659275283936,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045596412945099615,
|
|
"loss": 5.3125,
|
|
"mean_token_accuracy": 0.18825740218162537,
|
|
"num_tokens": 5997388.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.562892007827759,
|
|
"epoch": 2.8923742563547865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045570891842771727,
|
|
"loss": 5.4069,
|
|
"mean_token_accuracy": 0.18840867429971694,
|
|
"num_tokens": 6008739.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.439795112609863,
|
|
"epoch": 2.897782585181179,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045545305070781135,
|
|
"loss": 5.3061,
|
|
"mean_token_accuracy": 0.1903103232383728,
|
|
"num_tokens": 6019490.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.425024318695068,
|
|
"epoch": 2.903190914007572,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045519652722110206,
|
|
"loss": 5.4163,
|
|
"mean_token_accuracy": 0.18717291206121445,
|
|
"num_tokens": 6030712.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.5159917831420895,
|
|
"epoch": 2.908599242833964,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045493934889979655,
|
|
"loss": 5.3062,
|
|
"mean_token_accuracy": 0.19533758014440536,
|
|
"num_tokens": 6042161.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.42613468170166,
|
|
"epoch": 2.914007571660357,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045468151667848116,
|
|
"loss": 5.3403,
|
|
"mean_token_accuracy": 0.181694895029068,
|
|
"num_tokens": 6054107.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.506204509735108,
|
|
"epoch": 2.9194159004867495,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000454423031494119,
|
|
"loss": 5.4279,
|
|
"mean_token_accuracy": 0.1853549972176552,
|
|
"num_tokens": 6065232.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.4872571468353275,
|
|
"epoch": 2.9248242293131423,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004541638942860458,
|
|
"loss": 5.3717,
|
|
"mean_token_accuracy": 0.18070090413093567,
|
|
"num_tokens": 6076294.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.429201602935791,
|
|
"epoch": 2.9302325581395348,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004539041059959667,
|
|
"loss": 5.2899,
|
|
"mean_token_accuracy": 0.20034124553203583,
|
|
"num_tokens": 6087700.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.462101459503174,
|
|
"epoch": 2.9356408869659276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045364366756795295,
|
|
"loss": 5.3568,
|
|
"mean_token_accuracy": 0.18671762496232985,
|
|
"num_tokens": 6098561.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.5582990646362305,
|
|
"epoch": 2.94104921579232,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045338257994843845,
|
|
"loss": 5.3993,
|
|
"mean_token_accuracy": 0.18982526957988738,
|
|
"num_tokens": 6109729.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.426860189437866,
|
|
"epoch": 2.946457544618713,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004531208440862162,
|
|
"loss": 5.3531,
|
|
"mean_token_accuracy": 0.19022171944379807,
|
|
"num_tokens": 6120468.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.442908239364624,
|
|
"epoch": 2.9518658734451053,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045285846093243485,
|
|
"loss": 5.3039,
|
|
"mean_token_accuracy": 0.19160391241312028,
|
|
"num_tokens": 6132457.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.413277912139892,
|
|
"epoch": 2.957274202271498,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045259543144059557,
|
|
"loss": 5.3636,
|
|
"mean_token_accuracy": 0.18272600769996644,
|
|
"num_tokens": 6142289.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.508983135223389,
|
|
"epoch": 2.9626825310978906,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004523317565665479,
|
|
"loss": 5.3546,
|
|
"mean_token_accuracy": 0.18970296233892442,
|
|
"num_tokens": 6153786.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.51864914894104,
|
|
"epoch": 2.9680908599242835,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045206743726848716,
|
|
"loss": 5.3599,
|
|
"mean_token_accuracy": 0.18873424232006072,
|
|
"num_tokens": 6163988.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.409710311889649,
|
|
"epoch": 2.973499188750676,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004518024745069502,
|
|
"loss": 5.4538,
|
|
"mean_token_accuracy": 0.18343908041715623,
|
|
"num_tokens": 6175859.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.699503564834595,
|
|
"epoch": 2.9789075175770687,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004515368692448123,
|
|
"loss": 5.6266,
|
|
"mean_token_accuracy": 0.17070456892251967,
|
|
"num_tokens": 6188116.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.486202669143677,
|
|
"epoch": 2.9843158464034616,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045127062244728353,
|
|
"loss": 5.4231,
|
|
"mean_token_accuracy": 0.18883605748414994,
|
|
"num_tokens": 6198606.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.586038541793823,
|
|
"epoch": 2.989724175229854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004510037350819055,
|
|
"loss": 5.4669,
|
|
"mean_token_accuracy": 0.18225214928388594,
|
|
"num_tokens": 6210366.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.4436369895935055,
|
|
"epoch": 2.9951325040562464,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045073620811854744,
|
|
"loss": 5.339,
|
|
"mean_token_accuracy": 0.18868414610624312,
|
|
"num_tokens": 6221376.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.520844247606066,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.00045046804252940294,
|
|
"loss": 5.3398,
|
|
"mean_token_accuracy": 0.18265912764602238,
|
|
"num_tokens": 6230667.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.414664220809937,
|
|
"epoch": 3.005408328826393,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045019923928898644,
|
|
"loss": 4.9735,
|
|
"mean_token_accuracy": 0.20398972183465958,
|
|
"num_tokens": 6241158.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.505496788024902,
|
|
"epoch": 3.0108166576527853,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004499297993741294,
|
|
"loss": 5.0853,
|
|
"mean_token_accuracy": 0.19250927865505219,
|
|
"num_tokens": 6253811.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.418458414077759,
|
|
"epoch": 3.016224986479178,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004496597237639773,
|
|
"loss": 5.0442,
|
|
"mean_token_accuracy": 0.19874765872955322,
|
|
"num_tokens": 6265066.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.414768218994141,
|
|
"epoch": 3.0216333153055706,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044938901343998536,
|
|
"loss": 5.0245,
|
|
"mean_token_accuracy": 0.19345808327198027,
|
|
"num_tokens": 6275889.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.421528053283692,
|
|
"epoch": 3.0270416441319634,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004491176693859156,
|
|
"loss": 4.9777,
|
|
"mean_token_accuracy": 0.2064252883195877,
|
|
"num_tokens": 6287576.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.392618036270141,
|
|
"epoch": 3.032449972958356,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004488456925878329,
|
|
"loss": 5.0524,
|
|
"mean_token_accuracy": 0.19955354928970337,
|
|
"num_tokens": 6299419.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.407972717285157,
|
|
"epoch": 3.0378583017847487,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044857308403410163,
|
|
"loss": 5.038,
|
|
"mean_token_accuracy": 0.20420097708702087,
|
|
"num_tokens": 6310464.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.443487882614136,
|
|
"epoch": 3.043266630611141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044829984471538183,
|
|
"loss": 5.0161,
|
|
"mean_token_accuracy": 0.20679845660924911,
|
|
"num_tokens": 6321442.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.44049711227417,
|
|
"epoch": 3.048674959437534,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004480259756246258,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.2004274383187294,
|
|
"num_tokens": 6332868.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.499447202682495,
|
|
"epoch": 3.0540832882639264,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044775147775707465,
|
|
"loss": 5.0699,
|
|
"mean_token_accuracy": 0.19943254590034484,
|
|
"num_tokens": 6345232.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.449938154220581,
|
|
"epoch": 3.0594916170903192,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000447476352110254,
|
|
"loss": 5.1128,
|
|
"mean_token_accuracy": 0.19331368207931518,
|
|
"num_tokens": 6356524.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.436954355239868,
|
|
"epoch": 3.0648999459167117,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004472005996839712,
|
|
"loss": 5.043,
|
|
"mean_token_accuracy": 0.20366497337818146,
|
|
"num_tokens": 6367590.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.4483569145202635,
|
|
"epoch": 3.0703082747431045,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004469242214803112,
|
|
"loss": 5.0747,
|
|
"mean_token_accuracy": 0.19170792251825333,
|
|
"num_tokens": 6378211.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.349018192291259,
|
|
"epoch": 3.075716603569497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004466472185036331,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.20370565503835678,
|
|
"num_tokens": 6389595.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.42221508026123,
|
|
"epoch": 3.08112493239589,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004463695917605663,
|
|
"loss": 5.0637,
|
|
"mean_token_accuracy": 0.19848927855491638,
|
|
"num_tokens": 6401286.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.421692705154419,
|
|
"epoch": 3.086533261222282,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000446091342260007,
|
|
"loss": 5.0249,
|
|
"mean_token_accuracy": 0.20935140252113343,
|
|
"num_tokens": 6411946.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.390236663818359,
|
|
"epoch": 3.091941590048675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004458124710131145,
|
|
"loss": 4.9916,
|
|
"mean_token_accuracy": 0.2087801143527031,
|
|
"num_tokens": 6422463.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.318904638290405,
|
|
"epoch": 3.0973499188750675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044553297903330766,
|
|
"loss": 5.0005,
|
|
"mean_token_accuracy": 0.2030511677265167,
|
|
"num_tokens": 6433576.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.342587852478028,
|
|
"epoch": 3.1027582477014604,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044525286733626085,
|
|
"loss": 4.9642,
|
|
"mean_token_accuracy": 0.2040652632713318,
|
|
"num_tokens": 6444058.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.46890811920166,
|
|
"epoch": 3.1081665765278528,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004449721369399008,
|
|
"loss": 5.153,
|
|
"mean_token_accuracy": 0.1962315946817398,
|
|
"num_tokens": 6453872.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.3616416454315186,
|
|
"epoch": 3.1135749053542456,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044469078886440227,
|
|
"loss": 5.0491,
|
|
"mean_token_accuracy": 0.1998698353767395,
|
|
"num_tokens": 6465367.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.3856360912323,
|
|
"epoch": 3.118983234180638,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044440882413218484,
|
|
"loss": 4.9952,
|
|
"mean_token_accuracy": 0.214361210167408,
|
|
"num_tokens": 6476655.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.3557243824005125,
|
|
"epoch": 3.124391563007031,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044412624376790905,
|
|
"loss": 5.0583,
|
|
"mean_token_accuracy": 0.1975081071257591,
|
|
"num_tokens": 6488348.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.366420650482178,
|
|
"epoch": 3.1297998918334233,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004438430487984726,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.19896784275770188,
|
|
"num_tokens": 6500127.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.366138601303101,
|
|
"epoch": 3.135208220659816,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004435592402530066,
|
|
"loss": 5.0381,
|
|
"mean_token_accuracy": 0.20011679977178573,
|
|
"num_tokens": 6512772.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.410118818283081,
|
|
"epoch": 3.1406165494862086,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000443274819162872,
|
|
"loss": 5.0619,
|
|
"mean_token_accuracy": 0.20104195177555084,
|
|
"num_tokens": 6523359.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.317583656311035,
|
|
"epoch": 3.1460248783126015,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004429897865616557,
|
|
"loss": 5.0372,
|
|
"mean_token_accuracy": 0.19881036579608918,
|
|
"num_tokens": 6533679.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.3567657470703125,
|
|
"epoch": 3.151433207138994,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004427041434851668,
|
|
"loss": 4.9986,
|
|
"mean_token_accuracy": 0.19929321706295014,
|
|
"num_tokens": 6544694.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.355186080932617,
|
|
"epoch": 3.1568415359653867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044241789097143293,
|
|
"loss": 5.036,
|
|
"mean_token_accuracy": 0.2012198194861412,
|
|
"num_tokens": 6555858.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.362930536270142,
|
|
"epoch": 3.162249864791779,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00044213103006069635,
|
|
"loss": 5.0759,
|
|
"mean_token_accuracy": 0.19714714139699935,
|
|
"num_tokens": 6568214.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.331308889389038,
|
|
"epoch": 3.167658193618172,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044184356179541035,
|
|
"loss": 5.0226,
|
|
"mean_token_accuracy": 0.20453886687755585,
|
|
"num_tokens": 6579646.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.405777883529663,
|
|
"epoch": 3.1730665224445644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004415554872202352,
|
|
"loss": 4.9792,
|
|
"mean_token_accuracy": 0.21176978200674057,
|
|
"num_tokens": 6590683.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.405206203460693,
|
|
"epoch": 3.1784748512709573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00044126680738203466,
|
|
"loss": 5.1325,
|
|
"mean_token_accuracy": 0.19892753213644027,
|
|
"num_tokens": 6601197.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.365461921691894,
|
|
"epoch": 3.1838831800973497,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004409775233298718,
|
|
"loss": 5.1719,
|
|
"mean_token_accuracy": 0.18689680397510527,
|
|
"num_tokens": 6613564.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.417956876754761,
|
|
"epoch": 3.1892915089237426,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004406876361150056,
|
|
"loss": 5.0349,
|
|
"mean_token_accuracy": 0.19886281788349153,
|
|
"num_tokens": 6624696.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.366645622253418,
|
|
"epoch": 3.194699837750135,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004403971467908869,
|
|
"loss": 5.0357,
|
|
"mean_token_accuracy": 0.21102752834558486,
|
|
"num_tokens": 6636302.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.415676689147949,
|
|
"epoch": 3.200108166576528,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004401060564131545,
|
|
"loss": 5.0805,
|
|
"mean_token_accuracy": 0.1987275093793869,
|
|
"num_tokens": 6646517.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.336449050903321,
|
|
"epoch": 3.2055164954029207,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004398143660396315,
|
|
"loss": 5.058,
|
|
"mean_token_accuracy": 0.19970374256372453,
|
|
"num_tokens": 6657768.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.282119178771973,
|
|
"epoch": 3.210924824229313,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004395220767303215,
|
|
"loss": 4.9904,
|
|
"mean_token_accuracy": 0.20733337104320526,
|
|
"num_tokens": 6669110.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.3756097793579105,
|
|
"epoch": 3.2163331530557056,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004392291895474045,
|
|
"loss": 5.0565,
|
|
"mean_token_accuracy": 0.19415468722581863,
|
|
"num_tokens": 6680925.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.4291376113891605,
|
|
"epoch": 3.2217414818820984,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004389357055552331,
|
|
"loss": 5.1549,
|
|
"mean_token_accuracy": 0.19940261989831926,
|
|
"num_tokens": 6692923.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.397581148147583,
|
|
"epoch": 3.2271498107084913,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043864162582032897,
|
|
"loss": 5.0149,
|
|
"mean_token_accuracy": 0.2041090801358223,
|
|
"num_tokens": 6702692.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.3515589237213135,
|
|
"epoch": 3.2325581395348837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004383469514113784,
|
|
"loss": 5.1192,
|
|
"mean_token_accuracy": 0.19674407541751862,
|
|
"num_tokens": 6713830.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.407912492752075,
|
|
"epoch": 3.2379664683612766,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004380516833992289,
|
|
"loss": 5.0986,
|
|
"mean_token_accuracy": 0.19508025646209717,
|
|
"num_tokens": 6725564.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.349242067337036,
|
|
"epoch": 3.243374797187669,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000437755822856885,
|
|
"loss": 5.0611,
|
|
"mean_token_accuracy": 0.19773777574300766,
|
|
"num_tokens": 6736563.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.243374797187669,
|
|
"eval_entropy": 5.326288795471191,
|
|
"eval_loss": 5.915134906768799,
|
|
"eval_mean_token_accuracy": 0.1746227607469667,
|
|
"eval_num_tokens": 6736563.0,
|
|
"eval_runtime": 2.5292,
|
|
"eval_samples_per_second": 1389.784,
|
|
"eval_steps_per_second": 173.97,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.05475223808e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|