0bc08d3f2a
Model: fpadovani/eng-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0811249323958896,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3510135917568000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|