6046 lines
164 KiB
JSON
6046 lines
164 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.585308750365818,
|
|
"eval_steps": 3000,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692006397247315,
|
|
"epoch": 0.0009755145839430299,
|
|
"grad_norm": 12.9375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.892,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 17120.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691958236694337,
|
|
"epoch": 0.0019510291678860598,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.8306,
|
|
"mean_token_accuracy": 0.00028931311098858716,
|
|
"num_tokens": 34182.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.688035488128662,
|
|
"epoch": 0.0029265437518290896,
|
|
"grad_norm": 9.5625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.545,
|
|
"mean_token_accuracy": 0.0563898827880621,
|
|
"num_tokens": 51205.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.501210594177246,
|
|
"epoch": 0.0039020583357721197,
|
|
"grad_norm": 4.90625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.2388,
|
|
"mean_token_accuracy": 0.0683012217283249,
|
|
"num_tokens": 67722.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.27708158493042,
|
|
"epoch": 0.00487757291971515,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.0655,
|
|
"mean_token_accuracy": 0.06708129495382309,
|
|
"num_tokens": 84230.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.517713356018067,
|
|
"epoch": 0.005853087503658179,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9573,
|
|
"mean_token_accuracy": 0.06805420815944671,
|
|
"num_tokens": 100967.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.29547052383423,
|
|
"epoch": 0.00682860208760121,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.8479,
|
|
"mean_token_accuracy": 0.0726205624639988,
|
|
"num_tokens": 117450.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.387489223480225,
|
|
"epoch": 0.007804116671544239,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8467,
|
|
"mean_token_accuracy": 0.06873829215765,
|
|
"num_tokens": 135328.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.264148712158203,
|
|
"epoch": 0.00877963125548727,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8057,
|
|
"mean_token_accuracy": 0.07149567306041718,
|
|
"num_tokens": 152571.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.268710899353028,
|
|
"epoch": 0.0097551458394303,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7553,
|
|
"mean_token_accuracy": 0.07231884635984898,
|
|
"num_tokens": 170216.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.260807514190674,
|
|
"epoch": 0.010730660423373329,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.702,
|
|
"mean_token_accuracy": 0.07669694796204567,
|
|
"num_tokens": 186513.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.229700088500977,
|
|
"epoch": 0.011706175007316359,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.6889,
|
|
"mean_token_accuracy": 0.07528627961874008,
|
|
"num_tokens": 202600.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.081421279907227,
|
|
"epoch": 0.01268168959125939,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5969,
|
|
"mean_token_accuracy": 0.08673674017190933,
|
|
"num_tokens": 220270.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.137750053405762,
|
|
"epoch": 0.01365720417520242,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.5338,
|
|
"mean_token_accuracy": 0.08762008398771286,
|
|
"num_tokens": 236810.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.082253742218018,
|
|
"epoch": 0.014632718759145449,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5259,
|
|
"mean_token_accuracy": 0.08117572665214538,
|
|
"num_tokens": 252559.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.097311305999757,
|
|
"epoch": 0.015608233343088479,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.4073,
|
|
"mean_token_accuracy": 0.09147667810320854,
|
|
"num_tokens": 268507.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.028666019439697,
|
|
"epoch": 0.01658374792703151,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.406,
|
|
"mean_token_accuracy": 0.0866853080689907,
|
|
"num_tokens": 284776.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.004025745391846,
|
|
"epoch": 0.01755926251097454,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3519,
|
|
"mean_token_accuracy": 0.08328424319624901,
|
|
"num_tokens": 300937.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.073490333557128,
|
|
"epoch": 0.01853477709491757,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.2741,
|
|
"mean_token_accuracy": 0.08352594524621963,
|
|
"num_tokens": 317464.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.939503002166749,
|
|
"epoch": 0.0195102916788606,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.1496,
|
|
"mean_token_accuracy": 0.08892395347356796,
|
|
"num_tokens": 334022.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.905517959594727,
|
|
"epoch": 0.02048580626280363,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1039,
|
|
"mean_token_accuracy": 0.087137870490551,
|
|
"num_tokens": 350545.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.737315464019776,
|
|
"epoch": 0.021461320846746658,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0136,
|
|
"mean_token_accuracy": 0.09296058937907219,
|
|
"num_tokens": 368878.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.848572158813477,
|
|
"epoch": 0.022436835430689687,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0059,
|
|
"mean_token_accuracy": 0.08585245460271836,
|
|
"num_tokens": 388381.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.741306686401368,
|
|
"epoch": 0.023412350014632717,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9316,
|
|
"mean_token_accuracy": 0.08110709637403488,
|
|
"num_tokens": 403727.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.699060344696045,
|
|
"epoch": 0.02438786459857575,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.882,
|
|
"mean_token_accuracy": 0.08276206851005555,
|
|
"num_tokens": 419046.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.528530883789063,
|
|
"epoch": 0.02536337918251878,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.7349,
|
|
"mean_token_accuracy": 0.09832416251301765,
|
|
"num_tokens": 435186.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.434342861175537,
|
|
"epoch": 0.02633889376646181,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7342,
|
|
"mean_token_accuracy": 0.08667377829551696,
|
|
"num_tokens": 451552.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.417459678649902,
|
|
"epoch": 0.02731440835040484,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.7047,
|
|
"mean_token_accuracy": 0.08775919750332832,
|
|
"num_tokens": 468006.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.20961685180664,
|
|
"epoch": 0.02828992293434787,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6358,
|
|
"mean_token_accuracy": 0.08715632259845733,
|
|
"num_tokens": 484381.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.286615180969239,
|
|
"epoch": 0.029265437518290898,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.6472,
|
|
"mean_token_accuracy": 0.08763191923499107,
|
|
"num_tokens": 501309.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.081134414672851,
|
|
"epoch": 0.030240952102233928,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5847,
|
|
"mean_token_accuracy": 0.08041713163256645,
|
|
"num_tokens": 518290.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.070791053771973,
|
|
"epoch": 0.031216466686176957,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.5477,
|
|
"mean_token_accuracy": 0.0860756978392601,
|
|
"num_tokens": 533700.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.901391220092773,
|
|
"epoch": 0.03219198127011999,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4125,
|
|
"mean_token_accuracy": 0.09618512019515038,
|
|
"num_tokens": 548935.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.865772151947022,
|
|
"epoch": 0.03316749585406302,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4734,
|
|
"mean_token_accuracy": 0.08627908602356911,
|
|
"num_tokens": 565869.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.87720022201538,
|
|
"epoch": 0.03414301043800605,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.4862,
|
|
"mean_token_accuracy": 0.08944675847887992,
|
|
"num_tokens": 583065.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.795064163208007,
|
|
"epoch": 0.03511852502194908,
|
|
"grad_norm": 2.921875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.4763,
|
|
"mean_token_accuracy": 0.09086722284555435,
|
|
"num_tokens": 599875.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.737292766571045,
|
|
"epoch": 0.03609403960589211,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.4493,
|
|
"mean_token_accuracy": 0.0904251515865326,
|
|
"num_tokens": 616296.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.838920211791992,
|
|
"epoch": 0.03706955418983514,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.4171,
|
|
"mean_token_accuracy": 0.08522484824061394,
|
|
"num_tokens": 632548.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.708141040802001,
|
|
"epoch": 0.03804506877377817,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.4184,
|
|
"mean_token_accuracy": 0.09042482525110244,
|
|
"num_tokens": 650395.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.621811580657958,
|
|
"epoch": 0.0390205833577212,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.4313,
|
|
"mean_token_accuracy": 0.0962656319141388,
|
|
"num_tokens": 668453.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.641033744812011,
|
|
"epoch": 0.03999609794166423,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.3245,
|
|
"mean_token_accuracy": 0.09897583797574043,
|
|
"num_tokens": 683914.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.632667446136475,
|
|
"epoch": 0.04097161252560726,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.4474,
|
|
"mean_token_accuracy": 0.0904572695493698,
|
|
"num_tokens": 700480.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.652472400665284,
|
|
"epoch": 0.041947127109550286,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4236,
|
|
"mean_token_accuracy": 0.08843122720718384,
|
|
"num_tokens": 716604.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.608741188049317,
|
|
"epoch": 0.042922641693493316,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.4127,
|
|
"mean_token_accuracy": 0.0901436910033226,
|
|
"num_tokens": 732697.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.667686557769775,
|
|
"epoch": 0.043898156277436345,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.4916,
|
|
"mean_token_accuracy": 0.08182057477533818,
|
|
"num_tokens": 749072.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.605211448669433,
|
|
"epoch": 0.044873670861379375,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.4343,
|
|
"mean_token_accuracy": 0.08532521277666091,
|
|
"num_tokens": 767826.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.613678073883056,
|
|
"epoch": 0.045849185445322405,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.3376,
|
|
"mean_token_accuracy": 0.09638948515057563,
|
|
"num_tokens": 786863.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.50732488632202,
|
|
"epoch": 0.046824700029265434,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.3269,
|
|
"mean_token_accuracy": 0.10453230515122414,
|
|
"num_tokens": 803925.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.582274150848388,
|
|
"epoch": 0.04780021461320847,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.4043,
|
|
"mean_token_accuracy": 0.09170760661363601,
|
|
"num_tokens": 822088.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.597280120849609,
|
|
"epoch": 0.0487757291971515,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.4091,
|
|
"mean_token_accuracy": 0.08865795284509659,
|
|
"num_tokens": 837991.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.430963706970214,
|
|
"epoch": 0.04975124378109453,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3073,
|
|
"mean_token_accuracy": 0.10073584988713265,
|
|
"num_tokens": 855593.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.710581207275391,
|
|
"epoch": 0.05072675836503756,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3215,
|
|
"mean_token_accuracy": 0.09267524480819703,
|
|
"num_tokens": 873696.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.442234230041503,
|
|
"epoch": 0.05170227294898059,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2974,
|
|
"mean_token_accuracy": 0.098615263402462,
|
|
"num_tokens": 890023.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.602227687835693,
|
|
"epoch": 0.05267778753292362,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.4573,
|
|
"mean_token_accuracy": 0.08713396042585372,
|
|
"num_tokens": 906492.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.557075691223144,
|
|
"epoch": 0.05365330211686665,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.3801,
|
|
"mean_token_accuracy": 0.08883005455136299,
|
|
"num_tokens": 921837.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.565664386749267,
|
|
"epoch": 0.05462881670080968,
|
|
"grad_norm": 2.0,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.3169,
|
|
"mean_token_accuracy": 0.0994895301759243,
|
|
"num_tokens": 938771.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.539812088012695,
|
|
"epoch": 0.05560433128475271,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.3361,
|
|
"mean_token_accuracy": 0.09000369980931282,
|
|
"num_tokens": 954562.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.536813306808472,
|
|
"epoch": 0.05657984586869574,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.2737,
|
|
"mean_token_accuracy": 0.09747345894575118,
|
|
"num_tokens": 973700.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.417824840545654,
|
|
"epoch": 0.05755536045263877,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.3354,
|
|
"mean_token_accuracy": 0.09296533018350601,
|
|
"num_tokens": 990688.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.517854309082031,
|
|
"epoch": 0.058530875036581796,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.4108,
|
|
"mean_token_accuracy": 0.08671896606683731,
|
|
"num_tokens": 1008144.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.50785789489746,
|
|
"epoch": 0.059506389620524826,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.3529,
|
|
"mean_token_accuracy": 0.08922475874423981,
|
|
"num_tokens": 1027379.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.543958187103271,
|
|
"epoch": 0.060481904204467855,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.3276,
|
|
"mean_token_accuracy": 0.09632412865757942,
|
|
"num_tokens": 1043999.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.512003898620605,
|
|
"epoch": 0.061457418788410885,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.3399,
|
|
"mean_token_accuracy": 0.08813720867037773,
|
|
"num_tokens": 1062237.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.499386405944824,
|
|
"epoch": 0.062432933372353915,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.2911,
|
|
"mean_token_accuracy": 0.09350473955273628,
|
|
"num_tokens": 1078092.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.514391231536866,
|
|
"epoch": 0.06340844795629695,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.3255,
|
|
"mean_token_accuracy": 0.09594152867794037,
|
|
"num_tokens": 1095124.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.501588439941406,
|
|
"epoch": 0.06438396254023998,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.3112,
|
|
"mean_token_accuracy": 0.0899473249912262,
|
|
"num_tokens": 1112775.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.435393333435059,
|
|
"epoch": 0.06535947712418301,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.2395,
|
|
"mean_token_accuracy": 0.09844813719391823,
|
|
"num_tokens": 1129726.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.475375080108643,
|
|
"epoch": 0.06633499170812604,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.2963,
|
|
"mean_token_accuracy": 0.09962385520339012,
|
|
"num_tokens": 1144034.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.432106876373291,
|
|
"epoch": 0.06731050629206907,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.243,
|
|
"mean_token_accuracy": 0.0963106982409954,
|
|
"num_tokens": 1159633.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.410737228393554,
|
|
"epoch": 0.0682860208760121,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.2048,
|
|
"mean_token_accuracy": 0.09049743562936782,
|
|
"num_tokens": 1175350.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.411827564239502,
|
|
"epoch": 0.06926153545995513,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.1279,
|
|
"mean_token_accuracy": 0.09687257036566735,
|
|
"num_tokens": 1193121.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.431901168823241,
|
|
"epoch": 0.07023705004389816,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.2869,
|
|
"mean_token_accuracy": 0.090427827835083,
|
|
"num_tokens": 1209896.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.414903736114502,
|
|
"epoch": 0.07121256462784119,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.2795,
|
|
"mean_token_accuracy": 0.09120966792106629,
|
|
"num_tokens": 1227011.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.36677007675171,
|
|
"epoch": 0.07218807921178422,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.2051,
|
|
"mean_token_accuracy": 0.09367732107639312,
|
|
"num_tokens": 1241595.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.433219146728515,
|
|
"epoch": 0.07316359379572725,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.2106,
|
|
"mean_token_accuracy": 0.09003259316086769,
|
|
"num_tokens": 1256919.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.26855993270874,
|
|
"epoch": 0.07413910837967028,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.2039,
|
|
"mean_token_accuracy": 0.1087122842669487,
|
|
"num_tokens": 1274344.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.406241989135742,
|
|
"epoch": 0.0751146229636133,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.1847,
|
|
"mean_token_accuracy": 0.09858927503228188,
|
|
"num_tokens": 1290252.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.430280685424805,
|
|
"epoch": 0.07609013754755634,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.241,
|
|
"mean_token_accuracy": 0.09144357889890671,
|
|
"num_tokens": 1306987.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.38391466140747,
|
|
"epoch": 0.07706565213149937,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.2544,
|
|
"mean_token_accuracy": 0.0886089101433754,
|
|
"num_tokens": 1323031.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.422025775909423,
|
|
"epoch": 0.0780411667154424,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.1942,
|
|
"mean_token_accuracy": 0.09101833701133728,
|
|
"num_tokens": 1339592.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.359192848205566,
|
|
"epoch": 0.07901668129938542,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.1758,
|
|
"mean_token_accuracy": 0.10071646049618721,
|
|
"num_tokens": 1358131.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.330156660079956,
|
|
"epoch": 0.07999219588332845,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0996,
|
|
"mean_token_accuracy": 0.10162749364972115,
|
|
"num_tokens": 1375705.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.238559198379516,
|
|
"epoch": 0.08096771046727148,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.1287,
|
|
"mean_token_accuracy": 0.10049679651856422,
|
|
"num_tokens": 1392736.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.364288711547852,
|
|
"epoch": 0.08194322505121451,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.1716,
|
|
"mean_token_accuracy": 0.09369641467928887,
|
|
"num_tokens": 1408536.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.287070560455323,
|
|
"epoch": 0.08291873963515754,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.1448,
|
|
"mean_token_accuracy": 0.09421766176819801,
|
|
"num_tokens": 1425423.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.344510269165038,
|
|
"epoch": 0.08389425421910057,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.1528,
|
|
"mean_token_accuracy": 0.09409154579043388,
|
|
"num_tokens": 1442420.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.356864070892334,
|
|
"epoch": 0.0848697688030436,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.1051,
|
|
"mean_token_accuracy": 0.09946276843547822,
|
|
"num_tokens": 1458835.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.345529747009277,
|
|
"epoch": 0.08584528338698663,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.2056,
|
|
"mean_token_accuracy": 0.09219177067279816,
|
|
"num_tokens": 1475563.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.245398807525635,
|
|
"epoch": 0.08682079797092966,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.1443,
|
|
"mean_token_accuracy": 0.10394385680556298,
|
|
"num_tokens": 1492687.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.398572254180909,
|
|
"epoch": 0.08779631255487269,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.1549,
|
|
"mean_token_accuracy": 0.09206878691911698,
|
|
"num_tokens": 1508394.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.238665962219239,
|
|
"epoch": 0.08877182713881572,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.1345,
|
|
"mean_token_accuracy": 0.1022073931992054,
|
|
"num_tokens": 1527112.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.340081214904785,
|
|
"epoch": 0.08974734172275875,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.172,
|
|
"mean_token_accuracy": 0.1001360334455967,
|
|
"num_tokens": 1545401.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.266607761383057,
|
|
"epoch": 0.09072285630670178,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.0632,
|
|
"mean_token_accuracy": 0.09910417199134827,
|
|
"num_tokens": 1561621.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.288278388977051,
|
|
"epoch": 0.09169837089064481,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.1701,
|
|
"mean_token_accuracy": 0.09583483785390853,
|
|
"num_tokens": 1579225.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.34649305343628,
|
|
"epoch": 0.09267388547458784,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.0698,
|
|
"mean_token_accuracy": 0.09566665440797806,
|
|
"num_tokens": 1596125.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.171308040618896,
|
|
"epoch": 0.09364940005853087,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.1873,
|
|
"mean_token_accuracy": 0.0900168351829052,
|
|
"num_tokens": 1613451.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.294004344940186,
|
|
"epoch": 0.09462491464247391,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.1328,
|
|
"mean_token_accuracy": 0.09555885493755341,
|
|
"num_tokens": 1630929.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.21247615814209,
|
|
"epoch": 0.09560042922641694,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.1292,
|
|
"mean_token_accuracy": 0.09596397504210472,
|
|
"num_tokens": 1647379.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.275317049026489,
|
|
"epoch": 0.09657594381035997,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9022,
|
|
"mean_token_accuracy": 0.10550546199083329,
|
|
"num_tokens": 1664415.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.096874141693116,
|
|
"epoch": 0.097551458394303,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.98,
|
|
"mean_token_accuracy": 0.10389248356223106,
|
|
"num_tokens": 1679556.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.084659719467163,
|
|
"epoch": 0.09852697297824603,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.929,
|
|
"mean_token_accuracy": 0.11411564722657204,
|
|
"num_tokens": 1696727.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.191386079788208,
|
|
"epoch": 0.09950248756218906,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9651,
|
|
"mean_token_accuracy": 0.10536571145057679,
|
|
"num_tokens": 1713918.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.281560707092286,
|
|
"epoch": 0.10047800214613209,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000257,
|
|
"loss": 8.1123,
|
|
"mean_token_accuracy": 0.08992677107453347,
|
|
"num_tokens": 1729581.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.201297378540039,
|
|
"epoch": 0.10145351673007512,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9692,
|
|
"mean_token_accuracy": 0.09487369433045387,
|
|
"num_tokens": 1746367.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.232381868362427,
|
|
"epoch": 0.10242903131401815,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 8.0664,
|
|
"mean_token_accuracy": 0.09260008409619332,
|
|
"num_tokens": 1765009.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.1582435131073,
|
|
"epoch": 0.10340454589796118,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 8.0738,
|
|
"mean_token_accuracy": 0.09549360871315002,
|
|
"num_tokens": 1782113.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.155733203887939,
|
|
"epoch": 0.10438006048190421,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.9716,
|
|
"mean_token_accuracy": 0.0997552827000618,
|
|
"num_tokens": 1798301.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.204453039169312,
|
|
"epoch": 0.10535557506584724,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 8.0163,
|
|
"mean_token_accuracy": 0.09422366693615913,
|
|
"num_tokens": 1815059.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.108425045013428,
|
|
"epoch": 0.10633108964979027,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.9208,
|
|
"mean_token_accuracy": 0.10392606258392334,
|
|
"num_tokens": 1831699.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.054630374908447,
|
|
"epoch": 0.1073066042337333,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.987,
|
|
"mean_token_accuracy": 0.09787993282079696,
|
|
"num_tokens": 1847993.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.20120029449463,
|
|
"epoch": 0.10828211881767633,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8415,
|
|
"mean_token_accuracy": 0.10329653769731521,
|
|
"num_tokens": 1864753.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.112896871566772,
|
|
"epoch": 0.10925763340161936,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.998,
|
|
"mean_token_accuracy": 0.10229669883847237,
|
|
"num_tokens": 1882154.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.059369564056396,
|
|
"epoch": 0.11023314798556239,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8607,
|
|
"mean_token_accuracy": 0.10597956404089928,
|
|
"num_tokens": 1899683.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.033463430404662,
|
|
"epoch": 0.11120866256950541,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8691,
|
|
"mean_token_accuracy": 0.10046507343649864,
|
|
"num_tokens": 1918489.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.10311417579651,
|
|
"epoch": 0.11218417715344844,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9332,
|
|
"mean_token_accuracy": 0.0978827528655529,
|
|
"num_tokens": 1935609.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.027519035339356,
|
|
"epoch": 0.11315969173739147,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.9663,
|
|
"mean_token_accuracy": 0.09261189103126526,
|
|
"num_tokens": 1951311.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.112423610687255,
|
|
"epoch": 0.1141352063213345,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 8.0193,
|
|
"mean_token_accuracy": 0.09270902574062348,
|
|
"num_tokens": 1968908.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.119378280639648,
|
|
"epoch": 0.11511072090527753,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.9161,
|
|
"mean_token_accuracy": 0.09896000772714615,
|
|
"num_tokens": 1983807.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.102399921417236,
|
|
"epoch": 0.11608623548922056,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.9479,
|
|
"mean_token_accuracy": 0.09910893440246582,
|
|
"num_tokens": 2000219.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.085544061660766,
|
|
"epoch": 0.11706175007316359,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.8969,
|
|
"mean_token_accuracy": 0.10566937029361725,
|
|
"num_tokens": 2017360.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.984587049484253,
|
|
"epoch": 0.11803726465710662,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8842,
|
|
"mean_token_accuracy": 0.10565442368388175,
|
|
"num_tokens": 2034953.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.064342927932739,
|
|
"epoch": 0.11901277924104965,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.8435,
|
|
"mean_token_accuracy": 0.09929553419351578,
|
|
"num_tokens": 2051551.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.023351669311523,
|
|
"epoch": 0.11998829382499268,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.8137,
|
|
"mean_token_accuracy": 0.10783486068248749,
|
|
"num_tokens": 2067188.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.9922370433807375,
|
|
"epoch": 0.12096380840893571,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.8723,
|
|
"mean_token_accuracy": 0.10359162315726281,
|
|
"num_tokens": 2082569.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.063414001464844,
|
|
"epoch": 0.12193932299287874,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.9427,
|
|
"mean_token_accuracy": 0.09755868315696717,
|
|
"num_tokens": 2099446.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.01604437828064,
|
|
"epoch": 0.12291483757682177,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8774,
|
|
"mean_token_accuracy": 0.1052703656256199,
|
|
"num_tokens": 2116526.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.07321276664734,
|
|
"epoch": 0.1238903521607648,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.9364,
|
|
"mean_token_accuracy": 0.09000732526183128,
|
|
"num_tokens": 2134624.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.8934488773345945,
|
|
"epoch": 0.12486586674470783,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7724,
|
|
"mean_token_accuracy": 0.10647233128547669,
|
|
"num_tokens": 2152872.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.017675352096557,
|
|
"epoch": 0.12584138132865086,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.9026,
|
|
"mean_token_accuracy": 0.09761283919215202,
|
|
"num_tokens": 2169480.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.964239072799683,
|
|
"epoch": 0.1268168959125939,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7652,
|
|
"mean_token_accuracy": 0.11001085266470909,
|
|
"num_tokens": 2186054.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.021305990219116,
|
|
"epoch": 0.12779241049653692,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.9007,
|
|
"mean_token_accuracy": 0.09257680848240853,
|
|
"num_tokens": 2202320.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.897371673583985,
|
|
"epoch": 0.12876792508047996,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7826,
|
|
"mean_token_accuracy": 0.10420317128300667,
|
|
"num_tokens": 2218817.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.001240921020507,
|
|
"epoch": 0.12974343966442298,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.9361,
|
|
"mean_token_accuracy": 0.09622591212391854,
|
|
"num_tokens": 2234845.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.018117094039917,
|
|
"epoch": 0.13071895424836602,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.87,
|
|
"mean_token_accuracy": 0.09835803210735321,
|
|
"num_tokens": 2251904.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.918103551864624,
|
|
"epoch": 0.13169446883230904,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.872,
|
|
"mean_token_accuracy": 0.10161498412489892,
|
|
"num_tokens": 2268343.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.039315509796143,
|
|
"epoch": 0.13266998341625208,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.8635,
|
|
"mean_token_accuracy": 0.10121119394898415,
|
|
"num_tokens": 2284524.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.767062854766846,
|
|
"epoch": 0.1336454980001951,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.7948,
|
|
"mean_token_accuracy": 0.10690312683582306,
|
|
"num_tokens": 2302350.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.020025968551636,
|
|
"epoch": 0.13462101258413814,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.8576,
|
|
"mean_token_accuracy": 0.09347154945135117,
|
|
"num_tokens": 2319953.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.949396181106567,
|
|
"epoch": 0.13559652716808115,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7541,
|
|
"mean_token_accuracy": 0.10058033019304276,
|
|
"num_tokens": 2336698.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.795568990707397,
|
|
"epoch": 0.1365720417520242,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7172,
|
|
"mean_token_accuracy": 0.10608202442526818,
|
|
"num_tokens": 2353986.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.892446756362915,
|
|
"epoch": 0.1375475563359672,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7931,
|
|
"mean_token_accuracy": 0.10464861243963242,
|
|
"num_tokens": 2370320.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.920102310180664,
|
|
"epoch": 0.13852307091991026,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.8371,
|
|
"mean_token_accuracy": 0.09619187116622925,
|
|
"num_tokens": 2385453.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.961038494110108,
|
|
"epoch": 0.13949858550385327,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.8199,
|
|
"mean_token_accuracy": 0.09596224501729012,
|
|
"num_tokens": 2401545.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.852643156051636,
|
|
"epoch": 0.14047410008779632,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.8045,
|
|
"mean_token_accuracy": 0.10015163868665695,
|
|
"num_tokens": 2418414.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.859485912322998,
|
|
"epoch": 0.14144961467173933,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.8569,
|
|
"mean_token_accuracy": 0.09870302975177765,
|
|
"num_tokens": 2434973.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.955648517608642,
|
|
"epoch": 0.14242512925568238,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.8625,
|
|
"mean_token_accuracy": 0.09559171572327614,
|
|
"num_tokens": 2451053.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.91767840385437,
|
|
"epoch": 0.1434006438396254,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.8198,
|
|
"mean_token_accuracy": 0.09590378031134605,
|
|
"num_tokens": 2467280.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.8793638229370115,
|
|
"epoch": 0.14437615842356843,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.7417,
|
|
"mean_token_accuracy": 0.10683688595890999,
|
|
"num_tokens": 2484588.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.828915596008301,
|
|
"epoch": 0.14535167300751145,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.7492,
|
|
"mean_token_accuracy": 0.10286132618784904,
|
|
"num_tokens": 2501200.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.78443865776062,
|
|
"epoch": 0.1463271875914545,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.7085,
|
|
"mean_token_accuracy": 0.11014421433210372,
|
|
"num_tokens": 2517199.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.8230297565460205,
|
|
"epoch": 0.1473027021753975,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.772,
|
|
"mean_token_accuracy": 0.0969228208065033,
|
|
"num_tokens": 2534824.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.725612926483154,
|
|
"epoch": 0.14827821675934055,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6539,
|
|
"mean_token_accuracy": 0.10735608860850335,
|
|
"num_tokens": 2550842.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.8303399085998535,
|
|
"epoch": 0.14925373134328357,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.5965,
|
|
"mean_token_accuracy": 0.10855040177702904,
|
|
"num_tokens": 2567878.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.791242980957032,
|
|
"epoch": 0.1502292459272266,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.7018,
|
|
"mean_token_accuracy": 0.1028899759054184,
|
|
"num_tokens": 2584148.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.784344911575317,
|
|
"epoch": 0.15120476051116966,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.7143,
|
|
"mean_token_accuracy": 0.10468425303697586,
|
|
"num_tokens": 2600067.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.857481384277344,
|
|
"epoch": 0.15218027509511267,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6988,
|
|
"mean_token_accuracy": 0.10016048923134804,
|
|
"num_tokens": 2615746.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.801036739349366,
|
|
"epoch": 0.15315578967905571,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.7356,
|
|
"mean_token_accuracy": 0.09950978606939316,
|
|
"num_tokens": 2631641.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.770744705200196,
|
|
"epoch": 0.15413130426299873,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.723,
|
|
"mean_token_accuracy": 0.10329516828060151,
|
|
"num_tokens": 2647015.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.763613557815551,
|
|
"epoch": 0.15510681884694177,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.6529,
|
|
"mean_token_accuracy": 0.09973687306046486,
|
|
"num_tokens": 2664045.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.779078197479248,
|
|
"epoch": 0.1560823334308848,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.6204,
|
|
"mean_token_accuracy": 0.10611663311719895,
|
|
"num_tokens": 2680774.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.753378868103027,
|
|
"epoch": 0.15705784801482783,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.7434,
|
|
"mean_token_accuracy": 0.10084796100854873,
|
|
"num_tokens": 2697719.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.735131311416626,
|
|
"epoch": 0.15803336259877085,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.6936,
|
|
"mean_token_accuracy": 0.09436290189623833,
|
|
"num_tokens": 2714518.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.7410211086273195,
|
|
"epoch": 0.1590088771827139,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.7422,
|
|
"mean_token_accuracy": 0.1037781998515129,
|
|
"num_tokens": 2729288.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.805609893798828,
|
|
"epoch": 0.1599843917666569,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.7328,
|
|
"mean_token_accuracy": 0.10105982348322869,
|
|
"num_tokens": 2746782.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.723773622512818,
|
|
"epoch": 0.16095990635059995,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.7618,
|
|
"mean_token_accuracy": 0.09355797246098518,
|
|
"num_tokens": 2763159.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.758946180343628,
|
|
"epoch": 0.16193542093454297,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.7035,
|
|
"mean_token_accuracy": 0.10159995257854462,
|
|
"num_tokens": 2780338.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.690690183639527,
|
|
"epoch": 0.162910935518486,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.6255,
|
|
"mean_token_accuracy": 0.1065425306558609,
|
|
"num_tokens": 2796214.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.810243940353393,
|
|
"epoch": 0.16388645010242903,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.7369,
|
|
"mean_token_accuracy": 0.09597960785031319,
|
|
"num_tokens": 2812745.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.653784370422363,
|
|
"epoch": 0.16486196468637207,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.6108,
|
|
"mean_token_accuracy": 0.10254015624523163,
|
|
"num_tokens": 2828288.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.766561985015869,
|
|
"epoch": 0.16583747927031509,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.7323,
|
|
"mean_token_accuracy": 0.0973560057580471,
|
|
"num_tokens": 2843618.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.639662551879883,
|
|
"epoch": 0.16681299385425813,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5204,
|
|
"mean_token_accuracy": 0.11329260393977165,
|
|
"num_tokens": 2860659.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.678345632553101,
|
|
"epoch": 0.16778850843820114,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.6314,
|
|
"mean_token_accuracy": 0.09550370648503304,
|
|
"num_tokens": 2877482.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.740656137466431,
|
|
"epoch": 0.1687640230221442,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.7182,
|
|
"mean_token_accuracy": 0.0953593485057354,
|
|
"num_tokens": 2895868.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.680325174331665,
|
|
"epoch": 0.1697395376060872,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.7127,
|
|
"mean_token_accuracy": 0.09883771538734436,
|
|
"num_tokens": 2913103.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.702930736541748,
|
|
"epoch": 0.17071505219003025,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.6396,
|
|
"mean_token_accuracy": 0.1021553747355938,
|
|
"num_tokens": 2928291.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.616526365280151,
|
|
"epoch": 0.17169056677397326,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.6904,
|
|
"mean_token_accuracy": 0.10376504734158516,
|
|
"num_tokens": 2945115.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.760608196258545,
|
|
"epoch": 0.1726660813579163,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5907,
|
|
"mean_token_accuracy": 0.0988293744623661,
|
|
"num_tokens": 2962869.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.663983917236328,
|
|
"epoch": 0.17364159594185932,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.7038,
|
|
"mean_token_accuracy": 0.09704967588186264,
|
|
"num_tokens": 2979298.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.685730457305908,
|
|
"epoch": 0.17461711052580237,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5975,
|
|
"mean_token_accuracy": 0.10688030645251274,
|
|
"num_tokens": 2996146.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.635448312759399,
|
|
"epoch": 0.17559262510974538,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.679,
|
|
"mean_token_accuracy": 0.10033529624342918,
|
|
"num_tokens": 3013075.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.693193626403809,
|
|
"epoch": 0.17656813969368843,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.632,
|
|
"mean_token_accuracy": 0.09835075661540031,
|
|
"num_tokens": 3029494.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.6192834854125975,
|
|
"epoch": 0.17754365427763144,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.6371,
|
|
"mean_token_accuracy": 0.10126575529575348,
|
|
"num_tokens": 3046333.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.632585287094116,
|
|
"epoch": 0.17851916886157448,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.6164,
|
|
"mean_token_accuracy": 0.10273443460464478,
|
|
"num_tokens": 3062714.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.714128875732422,
|
|
"epoch": 0.1794946834455175,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.6159,
|
|
"mean_token_accuracy": 0.10113414600491524,
|
|
"num_tokens": 3080711.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.591054010391235,
|
|
"epoch": 0.18047019802946054,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.6972,
|
|
"mean_token_accuracy": 0.09632447808980941,
|
|
"num_tokens": 3096885.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.6689952373504635,
|
|
"epoch": 0.18144571261340356,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.5938,
|
|
"mean_token_accuracy": 0.10410150662064552,
|
|
"num_tokens": 3114536.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.593691968917847,
|
|
"epoch": 0.1824212271973466,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.5417,
|
|
"mean_token_accuracy": 0.10690793171525001,
|
|
"num_tokens": 3131921.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.663251686096191,
|
|
"epoch": 0.18339674178128962,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.6836,
|
|
"mean_token_accuracy": 0.09887410327792168,
|
|
"num_tokens": 3148064.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.519239473342895,
|
|
"epoch": 0.18437225636523266,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4983,
|
|
"mean_token_accuracy": 0.11145109161734582,
|
|
"num_tokens": 3165822.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.600449705123902,
|
|
"epoch": 0.18534777094917568,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.5992,
|
|
"mean_token_accuracy": 0.10686507672071457,
|
|
"num_tokens": 3183286.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.692311573028564,
|
|
"epoch": 0.18632328553311872,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.6311,
|
|
"mean_token_accuracy": 0.0974265143275261,
|
|
"num_tokens": 3198522.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.514405059814453,
|
|
"epoch": 0.18729880011706174,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.6467,
|
|
"mean_token_accuracy": 0.10179897472262382,
|
|
"num_tokens": 3218422.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.675178670883179,
|
|
"epoch": 0.18827431470100478,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.6831,
|
|
"mean_token_accuracy": 0.09527623429894447,
|
|
"num_tokens": 3236407.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.584980392456055,
|
|
"epoch": 0.18924982928494782,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.5995,
|
|
"mean_token_accuracy": 0.10131113454699517,
|
|
"num_tokens": 3252740.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.610750722885132,
|
|
"epoch": 0.19022534386889084,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.5638,
|
|
"mean_token_accuracy": 0.09960713386535644,
|
|
"num_tokens": 3269874.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.584094142913818,
|
|
"epoch": 0.19120085845283388,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.6135,
|
|
"mean_token_accuracy": 0.10098579525947571,
|
|
"num_tokens": 3286515.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.482845687866211,
|
|
"epoch": 0.1921763730367769,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3981,
|
|
"mean_token_accuracy": 0.12201494947075844,
|
|
"num_tokens": 3303700.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.54534764289856,
|
|
"epoch": 0.19315188762071994,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5804,
|
|
"mean_token_accuracy": 0.10665170550346374,
|
|
"num_tokens": 3319768.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.5603784084320065,
|
|
"epoch": 0.19412740220466296,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.4928,
|
|
"mean_token_accuracy": 0.1061321146786213,
|
|
"num_tokens": 3336555.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.4874776840209964,
|
|
"epoch": 0.195102916788606,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4912,
|
|
"mean_token_accuracy": 0.10724320039153099,
|
|
"num_tokens": 3353273.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.5069818019866945,
|
|
"epoch": 0.19607843137254902,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004999999929644166,
|
|
"loss": 7.3894,
|
|
"mean_token_accuracy": 0.11500273495912552,
|
|
"num_tokens": 3369408.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.454842758178711,
|
|
"epoch": 0.19705394595649206,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999999643823598,
|
|
"loss": 7.458,
|
|
"mean_token_accuracy": 0.10803161263465881,
|
|
"num_tokens": 3385834.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.504195785522461,
|
|
"epoch": 0.19802946054043508,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999999138141084,
|
|
"loss": 7.5588,
|
|
"mean_token_accuracy": 0.10538785606622696,
|
|
"num_tokens": 3401407.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.613107299804687,
|
|
"epoch": 0.19900497512437812,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999998412596673,
|
|
"loss": 7.5628,
|
|
"mean_token_accuracy": 0.09914690926671028,
|
|
"num_tokens": 3418256.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.500989294052124,
|
|
"epoch": 0.19998048970832114,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999997467190435,
|
|
"loss": 7.5344,
|
|
"mean_token_accuracy": 0.1071588896214962,
|
|
"num_tokens": 3433932.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.476874446868896,
|
|
"epoch": 0.20095600429226418,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999996301922465,
|
|
"loss": 7.5814,
|
|
"mean_token_accuracy": 0.09997389689087868,
|
|
"num_tokens": 3451723.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.533150291442871,
|
|
"epoch": 0.2019315188762072,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999994916792876,
|
|
"loss": 7.5493,
|
|
"mean_token_accuracy": 0.10057334378361701,
|
|
"num_tokens": 3469170.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.584330272674561,
|
|
"epoch": 0.20290703346015024,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999993311801802,
|
|
"loss": 7.5659,
|
|
"mean_token_accuracy": 0.09998832419514656,
|
|
"num_tokens": 3485282.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.521093273162842,
|
|
"epoch": 0.20388254804409325,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00049999914869494,
|
|
"loss": 7.571,
|
|
"mean_token_accuracy": 0.09400331526994705,
|
|
"num_tokens": 3502206.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.571083688735962,
|
|
"epoch": 0.2048580626280363,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000499998944223585,
|
|
"loss": 7.4951,
|
|
"mean_token_accuracy": 0.10485005304217339,
|
|
"num_tokens": 3520384.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.377060794830323,
|
|
"epoch": 0.2058335772119793,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004999987177661351,
|
|
"loss": 7.4949,
|
|
"mean_token_accuracy": 0.10567046105861663,
|
|
"num_tokens": 3536314.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.494265747070313,
|
|
"epoch": 0.20680909179592236,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999984693226124,
|
|
"loss": 7.4656,
|
|
"mean_token_accuracy": 0.11277214139699936,
|
|
"num_tokens": 3551510.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.456046295166016,
|
|
"epoch": 0.20778460637986537,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999981988930411,
|
|
"loss": 7.4626,
|
|
"mean_token_accuracy": 0.10642851740121842,
|
|
"num_tokens": 3568515.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.450523471832275,
|
|
"epoch": 0.20876012096380842,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999979064774478,
|
|
"loss": 7.4992,
|
|
"mean_token_accuracy": 0.10894627124071121,
|
|
"num_tokens": 3585183.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.419216012954712,
|
|
"epoch": 0.20973563554775143,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000499997592075861,
|
|
"loss": 7.4519,
|
|
"mean_token_accuracy": 0.10805186256766319,
|
|
"num_tokens": 3601898.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.5282000541687015,
|
|
"epoch": 0.21071115013169447,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999972556883113,
|
|
"loss": 7.5039,
|
|
"mean_token_accuracy": 0.103752601146698,
|
|
"num_tokens": 3616699.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.493114757537842,
|
|
"epoch": 0.2116866647156375,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999968973148318,
|
|
"loss": 7.5122,
|
|
"mean_token_accuracy": 0.10303746312856674,
|
|
"num_tokens": 3634030.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.544573545455933,
|
|
"epoch": 0.21266217929958053,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004999965169554573,
|
|
"loss": 7.5177,
|
|
"mean_token_accuracy": 0.1014569953083992,
|
|
"num_tokens": 3650495.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.461155843734741,
|
|
"epoch": 0.21363769388352355,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999961146102251,
|
|
"loss": 7.4642,
|
|
"mean_token_accuracy": 0.10736886784434319,
|
|
"num_tokens": 3666926.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.4098371982574465,
|
|
"epoch": 0.2146132084674666,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999956902791746,
|
|
"loss": 7.4399,
|
|
"mean_token_accuracy": 0.11416979283094406,
|
|
"num_tokens": 3683061.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.425881814956665,
|
|
"epoch": 0.2155887230514096,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004999952439623469,
|
|
"loss": 7.5083,
|
|
"mean_token_accuracy": 0.10643071681261063,
|
|
"num_tokens": 3703013.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.470410490036011,
|
|
"epoch": 0.21656423763535265,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000499994775659786,
|
|
"loss": 7.3552,
|
|
"mean_token_accuracy": 0.11576942354440689,
|
|
"num_tokens": 3719885.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.442147254943848,
|
|
"epoch": 0.21753975221929567,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999942853715375,
|
|
"loss": 7.5,
|
|
"mean_token_accuracy": 0.10365172252058982,
|
|
"num_tokens": 3735122.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.475345134735107,
|
|
"epoch": 0.2185152668032387,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004999937730976494,
|
|
"loss": 7.4711,
|
|
"mean_token_accuracy": 0.10123512893915176,
|
|
"num_tokens": 3751627.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.391668796539307,
|
|
"epoch": 0.21949078138718173,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004999932388381716,
|
|
"loss": 7.4941,
|
|
"mean_token_accuracy": 0.10175474062561989,
|
|
"num_tokens": 3768165.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.565304565429687,
|
|
"epoch": 0.22046629597112477,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999926825931564,
|
|
"loss": 7.5158,
|
|
"mean_token_accuracy": 0.1036800429224968,
|
|
"num_tokens": 3784245.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.275793266296387,
|
|
"epoch": 0.2214418105550678,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999921043626582,
|
|
"loss": 7.2665,
|
|
"mean_token_accuracy": 0.11247111111879349,
|
|
"num_tokens": 3802057.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.457341098785401,
|
|
"epoch": 0.22241732513901083,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999915041467333,
|
|
"loss": 7.4519,
|
|
"mean_token_accuracy": 0.10512350872159004,
|
|
"num_tokens": 3817104.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.422237396240234,
|
|
"epoch": 0.22339283972295385,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999908819454407,
|
|
"loss": 7.446,
|
|
"mean_token_accuracy": 0.10683505311608314,
|
|
"num_tokens": 3833037.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.36663784980774,
|
|
"epoch": 0.2243683543068969,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999902377588408,
|
|
"loss": 7.3331,
|
|
"mean_token_accuracy": 0.10857838094234466,
|
|
"num_tokens": 3849634.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.420496654510498,
|
|
"epoch": 0.2253438688908399,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.000499989571586997,
|
|
"loss": 7.4986,
|
|
"mean_token_accuracy": 0.10436328127980232,
|
|
"num_tokens": 3865949.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.3169677734375,
|
|
"epoch": 0.22631938347478295,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004999888834299739,
|
|
"loss": 7.3869,
|
|
"mean_token_accuracy": 0.11616735979914665,
|
|
"num_tokens": 3884646.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.491457986831665,
|
|
"epoch": 0.227294898058726,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999881732878391,
|
|
"loss": 7.466,
|
|
"mean_token_accuracy": 0.10703919902443886,
|
|
"num_tokens": 3900740.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.363152885437012,
|
|
"epoch": 0.228270412642669,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004999874411606618,
|
|
"loss": 7.4314,
|
|
"mean_token_accuracy": 0.11084038242697716,
|
|
"num_tokens": 3918619.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.45768780708313,
|
|
"epoch": 0.22924592722661205,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999866870485137,
|
|
"loss": 7.3911,
|
|
"mean_token_accuracy": 0.10921363979578018,
|
|
"num_tokens": 3938161.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.441078805923462,
|
|
"epoch": 0.23022144181055507,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999859109514684,
|
|
"loss": 7.4416,
|
|
"mean_token_accuracy": 0.1069450207054615,
|
|
"num_tokens": 3953582.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.39387698173523,
|
|
"epoch": 0.2311969563944981,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004999851128696017,
|
|
"loss": 7.409,
|
|
"mean_token_accuracy": 0.1000187411904335,
|
|
"num_tokens": 3969747.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.476375389099121,
|
|
"epoch": 0.23217247097844113,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999842928029917,
|
|
"loss": 7.4518,
|
|
"mean_token_accuracy": 0.10572833418846131,
|
|
"num_tokens": 3984646.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.347596120834351,
|
|
"epoch": 0.23314798556238417,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999834507517184,
|
|
"loss": 7.4403,
|
|
"mean_token_accuracy": 0.10399013683199883,
|
|
"num_tokens": 4003108.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.443073081970215,
|
|
"epoch": 0.23412350014632718,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999825867158642,
|
|
"loss": 7.5053,
|
|
"mean_token_accuracy": 0.10046004429459572,
|
|
"num_tokens": 4019944.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.434243679046631,
|
|
"epoch": 0.23509901473027023,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999817006955134,
|
|
"loss": 7.3817,
|
|
"mean_token_accuracy": 0.11024417504668235,
|
|
"num_tokens": 4037617.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.375543785095215,
|
|
"epoch": 0.23607452931421324,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999807926907527,
|
|
"loss": 7.3821,
|
|
"mean_token_accuracy": 0.10390213131904602,
|
|
"num_tokens": 4053483.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.442840242385865,
|
|
"epoch": 0.2370500438981563,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999798627016707,
|
|
"loss": 7.4542,
|
|
"mean_token_accuracy": 0.10955530405044556,
|
|
"num_tokens": 4070165.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.404299592971801,
|
|
"epoch": 0.2380255584820993,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999789107283586,
|
|
"loss": 7.4285,
|
|
"mean_token_accuracy": 0.1026936151087284,
|
|
"num_tokens": 4086208.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.361278438568116,
|
|
"epoch": 0.23900107306604235,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000499977936770909,
|
|
"loss": 7.3886,
|
|
"mean_token_accuracy": 0.10849282145500183,
|
|
"num_tokens": 4101790.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.39232292175293,
|
|
"epoch": 0.23997658764998536,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999769408294173,
|
|
"loss": 7.3226,
|
|
"mean_token_accuracy": 0.11622612252831459,
|
|
"num_tokens": 4117633.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.36030535697937,
|
|
"epoch": 0.2409521022339284,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004999759229039807,
|
|
"loss": 7.3937,
|
|
"mean_token_accuracy": 0.10975055694580078,
|
|
"num_tokens": 4132634.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.391692495346069,
|
|
"epoch": 0.24192761681787142,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004999748829946988,
|
|
"loss": 7.4326,
|
|
"mean_token_accuracy": 0.10449747890233993,
|
|
"num_tokens": 4150702.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.378620338439942,
|
|
"epoch": 0.24290313140181446,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999738211016732,
|
|
"loss": 7.3054,
|
|
"mean_token_accuracy": 0.12116105854511261,
|
|
"num_tokens": 4168483.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.168521881103516,
|
|
"epoch": 0.24387864598575748,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999727372250076,
|
|
"loss": 7.269,
|
|
"mean_token_accuracy": 0.11330273225903512,
|
|
"num_tokens": 4186199.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.455726528167725,
|
|
"epoch": 0.24485416056970052,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999716313648079,
|
|
"loss": 7.4094,
|
|
"mean_token_accuracy": 0.10582706704735756,
|
|
"num_tokens": 4202303.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.416627550125122,
|
|
"epoch": 0.24582967515364354,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999705035211822,
|
|
"loss": 7.4334,
|
|
"mean_token_accuracy": 0.1056906208395958,
|
|
"num_tokens": 4218541.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.40962781906128,
|
|
"epoch": 0.24680518973758658,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999693536942406,
|
|
"loss": 7.4016,
|
|
"mean_token_accuracy": 0.10808240473270417,
|
|
"num_tokens": 4234025.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.359267616271973,
|
|
"epoch": 0.2477807043215296,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999681818840957,
|
|
"loss": 7.4212,
|
|
"mean_token_accuracy": 0.10803825408220291,
|
|
"num_tokens": 4250798.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.362655210494995,
|
|
"epoch": 0.24875621890547264,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000499966988090862,
|
|
"loss": 7.4198,
|
|
"mean_token_accuracy": 0.10368345007300377,
|
|
"num_tokens": 4266453.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.493310832977295,
|
|
"epoch": 0.24973173348941566,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999657723146558,
|
|
"loss": 7.3599,
|
|
"mean_token_accuracy": 0.11028055846691132,
|
|
"num_tokens": 4281825.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.1498860836029055,
|
|
"epoch": 0.2507072480733587,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004999645345555962,
|
|
"loss": 7.2152,
|
|
"mean_token_accuracy": 0.12320265024900437,
|
|
"num_tokens": 4298821.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.333085250854492,
|
|
"epoch": 0.2516827626573017,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999632748138041,
|
|
"loss": 7.2362,
|
|
"mean_token_accuracy": 0.11482708379626275,
|
|
"num_tokens": 4314076.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.318206071853638,
|
|
"epoch": 0.25265827724124473,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999619930894025,
|
|
"loss": 7.2854,
|
|
"mean_token_accuracy": 0.11372094079852105,
|
|
"num_tokens": 4330909.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.337229156494141,
|
|
"epoch": 0.2536337918251878,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999606893825168,
|
|
"loss": 7.3535,
|
|
"mean_token_accuracy": 0.11244488433003426,
|
|
"num_tokens": 4346927.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.334287166595459,
|
|
"epoch": 0.2546093064091308,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004999593636932742,
|
|
"loss": 7.4025,
|
|
"mean_token_accuracy": 0.106304831802845,
|
|
"num_tokens": 4364956.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.3552874565124515,
|
|
"epoch": 0.25558482099307384,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004999580160218044,
|
|
"loss": 7.3232,
|
|
"mean_token_accuracy": 0.11021103486418724,
|
|
"num_tokens": 4383014.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.275822448730469,
|
|
"epoch": 0.25656033557701685,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004999566463682391,
|
|
"loss": 7.4183,
|
|
"mean_token_accuracy": 0.10739389285445214,
|
|
"num_tokens": 4401050.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.3236278057098385,
|
|
"epoch": 0.2575358501609599,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004999552547327119,
|
|
"loss": 7.2918,
|
|
"mean_token_accuracy": 0.1109052024781704,
|
|
"num_tokens": 4415453.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.293913793563843,
|
|
"epoch": 0.25851136474490294,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999538411153591,
|
|
"loss": 7.2756,
|
|
"mean_token_accuracy": 0.1161400705575943,
|
|
"num_tokens": 4433107.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.332246017456055,
|
|
"epoch": 0.25948687932884595,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999524055163187,
|
|
"loss": 7.3661,
|
|
"mean_token_accuracy": 0.10503623932600022,
|
|
"num_tokens": 4449440.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.3815333366394045,
|
|
"epoch": 0.26046239391278897,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000499950947935731,
|
|
"loss": 7.3488,
|
|
"mean_token_accuracy": 0.10719693452119827,
|
|
"num_tokens": 4465327.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.370807838439942,
|
|
"epoch": 0.26143790849673204,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004999494683737384,
|
|
"loss": 7.3741,
|
|
"mean_token_accuracy": 0.1034908264875412,
|
|
"num_tokens": 4481518.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.375785827636719,
|
|
"epoch": 0.26241342308067506,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004999479668304854,
|
|
"loss": 7.3928,
|
|
"mean_token_accuracy": 0.10684486851096153,
|
|
"num_tokens": 4497632.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.132483434677124,
|
|
"epoch": 0.2633889376646181,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004999464433061188,
|
|
"loss": 7.1541,
|
|
"mean_token_accuracy": 0.12024063989520073,
|
|
"num_tokens": 4513790.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.308583927154541,
|
|
"epoch": 0.26436445224856114,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999448978007876,
|
|
"loss": 7.278,
|
|
"mean_token_accuracy": 0.112956852465868,
|
|
"num_tokens": 4530802.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.34735803604126,
|
|
"epoch": 0.26533996683250416,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999433303146426,
|
|
"loss": 7.3344,
|
|
"mean_token_accuracy": 0.10671353563666344,
|
|
"num_tokens": 4546208.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.294841766357422,
|
|
"epoch": 0.2663154814164472,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004999417408478371,
|
|
"loss": 7.3908,
|
|
"mean_token_accuracy": 0.10694388523697854,
|
|
"num_tokens": 4563753.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.300462913513184,
|
|
"epoch": 0.2672909960003902,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999401294005264,
|
|
"loss": 7.275,
|
|
"mean_token_accuracy": 0.11355202496051789,
|
|
"num_tokens": 4580274.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.285962104797363,
|
|
"epoch": 0.26826651058433326,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000499938495972868,
|
|
"loss": 7.3565,
|
|
"mean_token_accuracy": 0.10627115666866302,
|
|
"num_tokens": 4596085.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.340288925170898,
|
|
"epoch": 0.2692420251682763,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004999368405650214,
|
|
"loss": 7.3823,
|
|
"mean_token_accuracy": 0.10515163168311119,
|
|
"num_tokens": 4613258.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.229737424850464,
|
|
"epoch": 0.2702175397522193,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999351631771485,
|
|
"loss": 7.2144,
|
|
"mean_token_accuracy": 0.11676347106695176,
|
|
"num_tokens": 4630128.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.2856508731842045,
|
|
"epoch": 0.2711930543361623,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999334638094131,
|
|
"loss": 7.2457,
|
|
"mean_token_accuracy": 0.11199160739779472,
|
|
"num_tokens": 4646181.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.3243039608001705,
|
|
"epoch": 0.2721685689201054,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999317424619812,
|
|
"loss": 7.2958,
|
|
"mean_token_accuracy": 0.10610211119055749,
|
|
"num_tokens": 4661804.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.253617286682129,
|
|
"epoch": 0.2731440835040484,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004999299991350212,
|
|
"loss": 7.2073,
|
|
"mean_token_accuracy": 0.11567432507872581,
|
|
"num_tokens": 4678163.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.251808834075928,
|
|
"epoch": 0.2741195980879914,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999282338287034,
|
|
"loss": 7.2031,
|
|
"mean_token_accuracy": 0.11231900975108147,
|
|
"num_tokens": 4695469.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.328244829177857,
|
|
"epoch": 0.2750951126719344,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004999264465432001,
|
|
"loss": 7.3404,
|
|
"mean_token_accuracy": 0.11071868836879731,
|
|
"num_tokens": 4711599.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.275565099716187,
|
|
"epoch": 0.2760706272558775,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004999246372786862,
|
|
"loss": 7.3392,
|
|
"mean_token_accuracy": 0.10794105604290963,
|
|
"num_tokens": 4728381.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.267817783355713,
|
|
"epoch": 0.2770461418398205,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004999228060353385,
|
|
"loss": 7.3392,
|
|
"mean_token_accuracy": 0.10768465995788574,
|
|
"num_tokens": 4745348.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.346744680404663,
|
|
"epoch": 0.27802165642376353,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004999209528133357,
|
|
"loss": 7.341,
|
|
"mean_token_accuracy": 0.10165520682930947,
|
|
"num_tokens": 4763011.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.288636779785156,
|
|
"epoch": 0.27899717100770655,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000499919077612859,
|
|
"loss": 7.288,
|
|
"mean_token_accuracy": 0.10964113101363182,
|
|
"num_tokens": 4779462.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.215214252471924,
|
|
"epoch": 0.2799726855916496,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004999171804340917,
|
|
"loss": 7.2462,
|
|
"mean_token_accuracy": 0.11817943677306175,
|
|
"num_tokens": 4796844.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.266482639312744,
|
|
"epoch": 0.28094820017559263,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004999152612772191,
|
|
"loss": 7.2566,
|
|
"mean_token_accuracy": 0.1129236064851284,
|
|
"num_tokens": 4812851.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.202636861801148,
|
|
"epoch": 0.28192371475953565,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999133201424288,
|
|
"loss": 7.2044,
|
|
"mean_token_accuracy": 0.11233072057366371,
|
|
"num_tokens": 4830144.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.329087591171264,
|
|
"epoch": 0.28289922934347866,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004999113570299105,
|
|
"loss": 7.2945,
|
|
"mean_token_accuracy": 0.10283592864871025,
|
|
"num_tokens": 4846981.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.3973691940307615,
|
|
"epoch": 0.28387474392742174,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004999093719398561,
|
|
"loss": 7.4583,
|
|
"mean_token_accuracy": 0.09933406785130501,
|
|
"num_tokens": 4864404.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.256984806060791,
|
|
"epoch": 0.28485025851136475,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004999073648724592,
|
|
"loss": 7.3441,
|
|
"mean_token_accuracy": 0.10281749814748764,
|
|
"num_tokens": 4882004.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.35606198310852,
|
|
"epoch": 0.28582577309530777,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004999053358279164,
|
|
"loss": 7.2788,
|
|
"mean_token_accuracy": 0.10536540821194648,
|
|
"num_tokens": 4899444.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.294085502624512,
|
|
"epoch": 0.2868012876792508,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004999032848064256,
|
|
"loss": 7.2738,
|
|
"mean_token_accuracy": 0.10585445612668991,
|
|
"num_tokens": 4917214.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.19778618812561,
|
|
"epoch": 0.28777680226319385,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999012118081875,
|
|
"loss": 7.3654,
|
|
"mean_token_accuracy": 0.10196950063109397,
|
|
"num_tokens": 4934534.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.33796648979187,
|
|
"epoch": 0.28875231684713687,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998991168334045,
|
|
"loss": 7.2736,
|
|
"mean_token_accuracy": 0.10711966082453728,
|
|
"num_tokens": 4950740.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.282023906707764,
|
|
"epoch": 0.2897278314310799,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004998969998822814,
|
|
"loss": 7.357,
|
|
"mean_token_accuracy": 0.10374500527977944,
|
|
"num_tokens": 4966720.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.320036554336548,
|
|
"epoch": 0.2907033460150229,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998948609550249,
|
|
"loss": 7.2468,
|
|
"mean_token_accuracy": 0.10450642704963684,
|
|
"num_tokens": 4982195.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.250419092178345,
|
|
"epoch": 0.29167886059896597,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004998927000518442,
|
|
"loss": 7.3605,
|
|
"mean_token_accuracy": 0.10471020191907883,
|
|
"num_tokens": 4997487.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.259267807006836,
|
|
"epoch": 0.292654375182909,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004998905171729505,
|
|
"loss": 7.2455,
|
|
"mean_token_accuracy": 0.10793293043971061,
|
|
"num_tokens": 5012913.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.207444763183593,
|
|
"epoch": 0.293629889766852,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004998883123185568,
|
|
"loss": 7.2118,
|
|
"mean_token_accuracy": 0.1189083606004715,
|
|
"num_tokens": 5031462.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.273334741592407,
|
|
"epoch": 0.294605404350795,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998860854888788,
|
|
"loss": 7.2006,
|
|
"mean_token_accuracy": 0.11655357554554939,
|
|
"num_tokens": 5047354.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.1826136112213135,
|
|
"epoch": 0.2955809189347381,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004998838366841341,
|
|
"loss": 7.2666,
|
|
"mean_token_accuracy": 0.11233135238289833,
|
|
"num_tokens": 5065268.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.27263765335083,
|
|
"epoch": 0.2965564335186811,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998815659045423,
|
|
"loss": 7.2622,
|
|
"mean_token_accuracy": 0.10862576067447663,
|
|
"num_tokens": 5081066.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.27400689125061,
|
|
"epoch": 0.2975319481026241,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004998792731503254,
|
|
"loss": 7.2909,
|
|
"mean_token_accuracy": 0.11164338439702988,
|
|
"num_tokens": 5099440.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.246980094909668,
|
|
"epoch": 0.29850746268656714,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004998769584217074,
|
|
"loss": 7.2408,
|
|
"mean_token_accuracy": 0.11869253292679786,
|
|
"num_tokens": 5116573.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.087693595886231,
|
|
"epoch": 0.2994829772705102,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004998746217189145,
|
|
"loss": 7.0129,
|
|
"mean_token_accuracy": 0.12668982818722724,
|
|
"num_tokens": 5132920.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.322620582580567,
|
|
"epoch": 0.3004584918544532,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000499872263042175,
|
|
"loss": 7.2676,
|
|
"mean_token_accuracy": 0.10794816985726356,
|
|
"num_tokens": 5150681.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.098180484771729,
|
|
"epoch": 0.30143400643839624,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004998698823917196,
|
|
"loss": 7.0585,
|
|
"mean_token_accuracy": 0.1271660767495632,
|
|
"num_tokens": 5167473.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.247739934921265,
|
|
"epoch": 0.3024095210223393,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004998674797677805,
|
|
"loss": 7.3303,
|
|
"mean_token_accuracy": 0.10471464842557907,
|
|
"num_tokens": 5183847.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.218716812133789,
|
|
"epoch": 0.3033850356062823,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004998650551705929,
|
|
"loss": 7.2864,
|
|
"mean_token_accuracy": 0.1065748505294323,
|
|
"num_tokens": 5200019.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.149478721618652,
|
|
"epoch": 0.30436055019022534,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998626086003934,
|
|
"loss": 7.1586,
|
|
"mean_token_accuracy": 0.11314807906746864,
|
|
"num_tokens": 5217245.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.24084324836731,
|
|
"epoch": 0.30533606477416836,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004998601400574213,
|
|
"loss": 7.0673,
|
|
"mean_token_accuracy": 0.11648911908268929,
|
|
"num_tokens": 5234822.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.2059166431427,
|
|
"epoch": 0.30631157935811143,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004998576495419177,
|
|
"loss": 7.1987,
|
|
"mean_token_accuracy": 0.110284673422575,
|
|
"num_tokens": 5251282.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.137545871734619,
|
|
"epoch": 0.30728709394205445,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000499855137054126,
|
|
"loss": 7.1793,
|
|
"mean_token_accuracy": 0.10772244781255721,
|
|
"num_tokens": 5269630.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.247171068191529,
|
|
"epoch": 0.30826260852599746,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998526025942917,
|
|
"loss": 7.2419,
|
|
"mean_token_accuracy": 0.10699235051870346,
|
|
"num_tokens": 5287106.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.254577684402466,
|
|
"epoch": 0.3092381231099405,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004998500461626625,
|
|
"loss": 7.2723,
|
|
"mean_token_accuracy": 0.10746567994356156,
|
|
"num_tokens": 5303387.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.319872236251831,
|
|
"epoch": 0.31021363769388355,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.000499847467759488,
|
|
"loss": 7.2542,
|
|
"mean_token_accuracy": 0.10725675225257873,
|
|
"num_tokens": 5320482.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.208133268356323,
|
|
"epoch": 0.31118915227782656,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004998448673850205,
|
|
"loss": 7.223,
|
|
"mean_token_accuracy": 0.11193624213337898,
|
|
"num_tokens": 5336420.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.094880104064941,
|
|
"epoch": 0.3121646668617696,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004998422450395139,
|
|
"loss": 7.1597,
|
|
"mean_token_accuracy": 0.11134122982621193,
|
|
"num_tokens": 5353464.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.231945180892945,
|
|
"epoch": 0.3131401814457126,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998396007232244,
|
|
"loss": 7.2267,
|
|
"mean_token_accuracy": 0.11414750963449478,
|
|
"num_tokens": 5372607.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.20907883644104,
|
|
"epoch": 0.31411569602965567,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998369344364105,
|
|
"loss": 7.2419,
|
|
"mean_token_accuracy": 0.11458229944109917,
|
|
"num_tokens": 5390134.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.192635631561279,
|
|
"epoch": 0.3150912106135987,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004998342461793328,
|
|
"loss": 7.1426,
|
|
"mean_token_accuracy": 0.11570450440049171,
|
|
"num_tokens": 5407083.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.023532438278198,
|
|
"epoch": 0.3160667251975417,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998315359522538,
|
|
"loss": 7.0462,
|
|
"mean_token_accuracy": 0.12723288238048552,
|
|
"num_tokens": 5424617.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.210883092880249,
|
|
"epoch": 0.3170422397814847,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004998288037554383,
|
|
"loss": 7.1847,
|
|
"mean_token_accuracy": 0.12012180760502815,
|
|
"num_tokens": 5440913.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.20393934249878,
|
|
"epoch": 0.3180177543654278,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004998260495891535,
|
|
"loss": 7.2107,
|
|
"mean_token_accuracy": 0.11585899069905281,
|
|
"num_tokens": 5457157.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.133607053756714,
|
|
"epoch": 0.3189932689493708,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004998232734536685,
|
|
"loss": 7.1437,
|
|
"mean_token_accuracy": 0.11597841158509255,
|
|
"num_tokens": 5474520.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.240679121017456,
|
|
"epoch": 0.3199687835333138,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998204753492544,
|
|
"loss": 7.2093,
|
|
"mean_token_accuracy": 0.11443925127387047,
|
|
"num_tokens": 5490599.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.190227603912353,
|
|
"epoch": 0.32094429811725683,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004998176552761848,
|
|
"loss": 7.2699,
|
|
"mean_token_accuracy": 0.10237837061285973,
|
|
"num_tokens": 5506714.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.265681362152099,
|
|
"epoch": 0.3219198127011999,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004998148132347352,
|
|
"loss": 7.2306,
|
|
"mean_token_accuracy": 0.10756946429610252,
|
|
"num_tokens": 5522894.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.134102153778076,
|
|
"epoch": 0.3228953272851429,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004998119492251832,
|
|
"loss": 7.0984,
|
|
"mean_token_accuracy": 0.11984493806958199,
|
|
"num_tokens": 5540054.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.262777042388916,
|
|
"epoch": 0.32387084186908593,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004998090632478088,
|
|
"loss": 7.2625,
|
|
"mean_token_accuracy": 0.10252039134502411,
|
|
"num_tokens": 5556189.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.238120269775391,
|
|
"epoch": 0.32484635645302895,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000499806155302894,
|
|
"loss": 7.2793,
|
|
"mean_token_accuracy": 0.1049114279448986,
|
|
"num_tokens": 5572562.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.277428388595581,
|
|
"epoch": 0.325821871036972,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000499803225390723,
|
|
"loss": 7.2699,
|
|
"mean_token_accuracy": 0.1055513508617878,
|
|
"num_tokens": 5588992.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.255199003219604,
|
|
"epoch": 0.32679738562091504,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004998002735115819,
|
|
"loss": 7.1826,
|
|
"mean_token_accuracy": 0.10926883295178413,
|
|
"num_tokens": 5604143.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.182808113098145,
|
|
"epoch": 0.32777290020485805,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004997972996657593,
|
|
"loss": 7.295,
|
|
"mean_token_accuracy": 0.10058706179261208,
|
|
"num_tokens": 5620887.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.215796041488647,
|
|
"epoch": 0.32874841478880107,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004997943038535459,
|
|
"loss": 7.2408,
|
|
"mean_token_accuracy": 0.10989224687218666,
|
|
"num_tokens": 5637156.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.2290528297424315,
|
|
"epoch": 0.32972392937274414,
|
|
"grad_norm": 0.7578125,
|
|
"learning_rate": 0.0004997912860752341,
|
|
"loss": 7.1963,
|
|
"mean_token_accuracy": 0.10859450697898865,
|
|
"num_tokens": 5654024.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.167445707321167,
|
|
"epoch": 0.33069944395668716,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997882463311192,
|
|
"loss": 7.1421,
|
|
"mean_token_accuracy": 0.10873658806085587,
|
|
"num_tokens": 5671538.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.191862487792969,
|
|
"epoch": 0.33167495854063017,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004997851846214979,
|
|
"loss": 7.1813,
|
|
"mean_token_accuracy": 0.11439736038446427,
|
|
"num_tokens": 5688788.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.1439920425415036,
|
|
"epoch": 0.3326504731245732,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004997821009466695,
|
|
"loss": 7.1274,
|
|
"mean_token_accuracy": 0.11521929055452347,
|
|
"num_tokens": 5706066.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.128738975524902,
|
|
"epoch": 0.33362598770851626,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997789953069354,
|
|
"loss": 7.1059,
|
|
"mean_token_accuracy": 0.11363393142819404,
|
|
"num_tokens": 5722100.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.202400350570679,
|
|
"epoch": 0.3346015022924593,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004997758677025989,
|
|
"loss": 7.2373,
|
|
"mean_token_accuracy": 0.10656889006495476,
|
|
"num_tokens": 5739068.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.182660055160523,
|
|
"epoch": 0.3355770168764023,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.0004997727181339658,
|
|
"loss": 7.2038,
|
|
"mean_token_accuracy": 0.10887709036469459,
|
|
"num_tokens": 5756147.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.202863168716431,
|
|
"epoch": 0.3365525314603453,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004997695466013438,
|
|
"loss": 7.1747,
|
|
"mean_token_accuracy": 0.10922196507453918,
|
|
"num_tokens": 5772157.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.202521228790284,
|
|
"epoch": 0.3375280460442884,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997663531050427,
|
|
"loss": 7.1983,
|
|
"mean_token_accuracy": 0.11102297827601433,
|
|
"num_tokens": 5788128.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.124049186706543,
|
|
"epoch": 0.3385035606282314,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004997631376453749,
|
|
"loss": 7.1898,
|
|
"mean_token_accuracy": 0.11080607250332833,
|
|
"num_tokens": 5804793.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.14783296585083,
|
|
"epoch": 0.3394790752121744,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004997599002226541,
|
|
"loss": 7.1129,
|
|
"mean_token_accuracy": 0.11176546439528465,
|
|
"num_tokens": 5822409.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.101404285430908,
|
|
"epoch": 0.3404545897961175,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499756640837197,
|
|
"loss": 7.1547,
|
|
"mean_token_accuracy": 0.10845724120736122,
|
|
"num_tokens": 5839023.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.241454839706421,
|
|
"epoch": 0.3414301043800605,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000499753359489322,
|
|
"loss": 7.169,
|
|
"mean_token_accuracy": 0.10654053464531898,
|
|
"num_tokens": 5855804.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.1083615779876705,
|
|
"epoch": 0.3424056189640035,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004997500561793498,
|
|
"loss": 7.1759,
|
|
"mean_token_accuracy": 0.11318699792027473,
|
|
"num_tokens": 5872372.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.232053995132446,
|
|
"epoch": 0.3433811335479465,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004997467309076031,
|
|
"loss": 7.2216,
|
|
"mean_token_accuracy": 0.10516873672604561,
|
|
"num_tokens": 5888666.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.054588890075683,
|
|
"epoch": 0.3443566481318896,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004997433836744069,
|
|
"loss": 7.1628,
|
|
"mean_token_accuracy": 0.11004716753959656,
|
|
"num_tokens": 5905577.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.211126518249512,
|
|
"epoch": 0.3453321627158326,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004997400144800882,
|
|
"loss": 7.1273,
|
|
"mean_token_accuracy": 0.11347045078873634,
|
|
"num_tokens": 5922570.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.046509408950806,
|
|
"epoch": 0.34630767729977563,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997366233249762,
|
|
"loss": 7.0896,
|
|
"mean_token_accuracy": 0.12237504422664643,
|
|
"num_tokens": 5938995.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.139289665222168,
|
|
"epoch": 0.34728319188371864,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004997332102094024,
|
|
"loss": 7.1051,
|
|
"mean_token_accuracy": 0.11859508603811264,
|
|
"num_tokens": 5954869.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.272542238235474,
|
|
"epoch": 0.3482587064676617,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004997297751337004,
|
|
"loss": 7.2014,
|
|
"mean_token_accuracy": 0.11317256689071656,
|
|
"num_tokens": 5972750.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.0967755794525145,
|
|
"epoch": 0.34923422105160473,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004997263180982055,
|
|
"loss": 7.1296,
|
|
"mean_token_accuracy": 0.1137120395898819,
|
|
"num_tokens": 5989004.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.098430013656616,
|
|
"epoch": 0.35020973563554775,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004997228391032559,
|
|
"loss": 7.1462,
|
|
"mean_token_accuracy": 0.11115683764219284,
|
|
"num_tokens": 6004749.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.090719604492188,
|
|
"epoch": 0.35118525021949076,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004997193381491913,
|
|
"loss": 7.019,
|
|
"mean_token_accuracy": 0.11676162034273148,
|
|
"num_tokens": 6022235.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.0829535007476805,
|
|
"epoch": 0.35216076480343383,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000499715815236354,
|
|
"loss": 7.1947,
|
|
"mean_token_accuracy": 0.10693453401327133,
|
|
"num_tokens": 6039194.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.189563322067261,
|
|
"epoch": 0.35313627938737685,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000499712270365088,
|
|
"loss": 7.0593,
|
|
"mean_token_accuracy": 0.10776279419660569,
|
|
"num_tokens": 6055376.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.125262308120727,
|
|
"epoch": 0.35411179397131987,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004997087035357398,
|
|
"loss": 7.1277,
|
|
"mean_token_accuracy": 0.11357001885771752,
|
|
"num_tokens": 6071105.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.074201965332032,
|
|
"epoch": 0.3550873085552629,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997051147486582,
|
|
"loss": 7.0519,
|
|
"mean_token_accuracy": 0.11605791896581649,
|
|
"num_tokens": 6087964.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.086669158935547,
|
|
"epoch": 0.35606282313920595,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004997015040041935,
|
|
"loss": 7.1199,
|
|
"mean_token_accuracy": 0.10704236179590225,
|
|
"num_tokens": 6105245.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.13190975189209,
|
|
"epoch": 0.35703833772314897,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004996978713026986,
|
|
"loss": 7.0724,
|
|
"mean_token_accuracy": 0.12007546350359917,
|
|
"num_tokens": 6121736.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.104649305343628,
|
|
"epoch": 0.358013852307092,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004996942166445287,
|
|
"loss": 7.1355,
|
|
"mean_token_accuracy": 0.11229356825351715,
|
|
"num_tokens": 6137751.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.145856428146362,
|
|
"epoch": 0.358989366891035,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004996905400300407,
|
|
"loss": 7.1839,
|
|
"mean_token_accuracy": 0.1106605127453804,
|
|
"num_tokens": 6154655.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.193540096282959,
|
|
"epoch": 0.35996488147497807,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499686841459594,
|
|
"loss": 7.1746,
|
|
"mean_token_accuracy": 0.10867514237761497,
|
|
"num_tokens": 6172230.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.10019702911377,
|
|
"epoch": 0.3609403960589211,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004996831209335497,
|
|
"loss": 7.1168,
|
|
"mean_token_accuracy": 0.10731762573122978,
|
|
"num_tokens": 6188520.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.129278135299683,
|
|
"epoch": 0.3619159106428641,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004996793784522719,
|
|
"loss": 7.1545,
|
|
"mean_token_accuracy": 0.10917501226067543,
|
|
"num_tokens": 6205555.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.180557060241699,
|
|
"epoch": 0.3628914252268071,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004996756140161259,
|
|
"loss": 7.1412,
|
|
"mean_token_accuracy": 0.11165231317281724,
|
|
"num_tokens": 6223548.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.114638328552246,
|
|
"epoch": 0.3638669398107502,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004996718276254797,
|
|
"loss": 7.151,
|
|
"mean_token_accuracy": 0.10733274817466736,
|
|
"num_tokens": 6239266.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.132656908035278,
|
|
"epoch": 0.3648424543946932,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004996680192807032,
|
|
"loss": 7.115,
|
|
"mean_token_accuracy": 0.11217067912220954,
|
|
"num_tokens": 6255020.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.115752077102661,
|
|
"epoch": 0.3658179689786362,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004996641889821686,
|
|
"loss": 7.0536,
|
|
"mean_token_accuracy": 0.10760100483894348,
|
|
"num_tokens": 6272774.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.088830280303955,
|
|
"epoch": 0.36679348356257924,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004996603367302502,
|
|
"loss": 7.0727,
|
|
"mean_token_accuracy": 0.11251697391271591,
|
|
"num_tokens": 6290478.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.100843334197998,
|
|
"epoch": 0.3677689981465223,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004996564625253244,
|
|
"loss": 7.1506,
|
|
"mean_token_accuracy": 0.11167759746313095,
|
|
"num_tokens": 6308003.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.098121881484985,
|
|
"epoch": 0.3687445127304653,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004996525663677697,
|
|
"loss": 7.1054,
|
|
"mean_token_accuracy": 0.10613081008195877,
|
|
"num_tokens": 6325748.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 7.079687166213989,
|
|
"epoch": 0.36972002731440834,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000499648648257967,
|
|
"loss": 7.0579,
|
|
"mean_token_accuracy": 0.12034444063901902,
|
|
"num_tokens": 6344031.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.065695476531983,
|
|
"epoch": 0.37069554189835136,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000499644708196299,
|
|
"loss": 7.1106,
|
|
"mean_token_accuracy": 0.11427498012781143,
|
|
"num_tokens": 6359821.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.036489105224609,
|
|
"epoch": 0.3716710564822944,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996407461831507,
|
|
"loss": 6.9544,
|
|
"mean_token_accuracy": 0.12395858988165856,
|
|
"num_tokens": 6376464.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.124951982498169,
|
|
"epoch": 0.37264657106623744,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004996367622189094,
|
|
"loss": 7.1286,
|
|
"mean_token_accuracy": 0.109093027561903,
|
|
"num_tokens": 6392415.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.078305053710937,
|
|
"epoch": 0.37362208565018046,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004996327563039643,
|
|
"loss": 7.1238,
|
|
"mean_token_accuracy": 0.1082805871963501,
|
|
"num_tokens": 6409151.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 7.166975975036621,
|
|
"epoch": 0.3745976002341235,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004996287284387068,
|
|
"loss": 7.0814,
|
|
"mean_token_accuracy": 0.11464927569031716,
|
|
"num_tokens": 6426997.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.088303327560425,
|
|
"epoch": 0.37557311481806654,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004996246786235306,
|
|
"loss": 7.0782,
|
|
"mean_token_accuracy": 0.11261907815933228,
|
|
"num_tokens": 6442871.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.04116382598877,
|
|
"epoch": 0.37654862940200956,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004996206068588314,
|
|
"loss": 7.0759,
|
|
"mean_token_accuracy": 0.11092812642455101,
|
|
"num_tokens": 6459677.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.167269468307495,
|
|
"epoch": 0.3775241439859526,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004996165131450071,
|
|
"loss": 7.2127,
|
|
"mean_token_accuracy": 0.10652901008725166,
|
|
"num_tokens": 6477239.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 7.098695421218872,
|
|
"epoch": 0.37849965856989565,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004996123974824576,
|
|
"loss": 7.0695,
|
|
"mean_token_accuracy": 0.11821612566709519,
|
|
"num_tokens": 6493277.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 7.047782516479492,
|
|
"epoch": 0.37947517315383866,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004996082598715852,
|
|
"loss": 7.0966,
|
|
"mean_token_accuracy": 0.11363401412963867,
|
|
"num_tokens": 6509204.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 7.014270877838134,
|
|
"epoch": 0.3804506877377817,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000499604100312794,
|
|
"loss": 6.9519,
|
|
"mean_token_accuracy": 0.12616565227508544,
|
|
"num_tokens": 6525486.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.046218776702881,
|
|
"epoch": 0.3814262023217247,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004995999188064908,
|
|
"loss": 7.158,
|
|
"mean_token_accuracy": 0.10741189196705818,
|
|
"num_tokens": 6542227.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.1583123207092285,
|
|
"epoch": 0.38240171690566777,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000499595715353084,
|
|
"loss": 7.2279,
|
|
"mean_token_accuracy": 0.10246274694800377,
|
|
"num_tokens": 6558951.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.150441884994507,
|
|
"epoch": 0.3833772314896108,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995914899529844,
|
|
"loss": 7.1286,
|
|
"mean_token_accuracy": 0.11037195920944214,
|
|
"num_tokens": 6574696.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.180084657669068,
|
|
"epoch": 0.3843527460735538,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004995872426066048,
|
|
"loss": 7.1448,
|
|
"mean_token_accuracy": 0.11020474806427956,
|
|
"num_tokens": 6591628.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.103500747680664,
|
|
"epoch": 0.3853282606574968,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995829733143602,
|
|
"loss": 7.0431,
|
|
"mean_token_accuracy": 0.11289738193154335,
|
|
"num_tokens": 6608107.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 7.082343816757202,
|
|
"epoch": 0.3863037752414399,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499578682076668,
|
|
"loss": 7.1855,
|
|
"mean_token_accuracy": 0.11477380469441414,
|
|
"num_tokens": 6626699.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 7.221960496902466,
|
|
"epoch": 0.3872792898253829,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004995743688939474,
|
|
"loss": 7.1891,
|
|
"mean_token_accuracy": 0.1018472857773304,
|
|
"num_tokens": 6644540.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.113204669952393,
|
|
"epoch": 0.3882548044093259,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004995700337666199,
|
|
"loss": 7.1306,
|
|
"mean_token_accuracy": 0.10720818117260933,
|
|
"num_tokens": 6660653.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 7.155580043792725,
|
|
"epoch": 0.38923031899326893,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000499565676695109,
|
|
"loss": 7.1317,
|
|
"mean_token_accuracy": 0.10808075144886971,
|
|
"num_tokens": 6677553.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.119460582733154,
|
|
"epoch": 0.390205833577212,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995612976798406,
|
|
"loss": 7.1175,
|
|
"mean_token_accuracy": 0.1091853991150856,
|
|
"num_tokens": 6695274.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 7.064525413513183,
|
|
"epoch": 0.391181348161155,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004995568967212425,
|
|
"loss": 7.0522,
|
|
"mean_token_accuracy": 0.11336143016815185,
|
|
"num_tokens": 6711166.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 7.021340322494507,
|
|
"epoch": 0.39215686274509803,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004995524738197448,
|
|
"loss": 7.028,
|
|
"mean_token_accuracy": 0.11226390227675438,
|
|
"num_tokens": 6728827.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.996756649017334,
|
|
"epoch": 0.39313237732904105,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004995480289757796,
|
|
"loss": 7.0064,
|
|
"mean_token_accuracy": 0.12143127471208573,
|
|
"num_tokens": 6745901.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 7.007443380355835,
|
|
"epoch": 0.3941078919129841,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004995435621897815,
|
|
"loss": 6.985,
|
|
"mean_token_accuracy": 0.11859407424926757,
|
|
"num_tokens": 6762176.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 7.087689161300659,
|
|
"epoch": 0.39508340649692714,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004995390734621867,
|
|
"loss": 7.139,
|
|
"mean_token_accuracy": 0.10878413170576096,
|
|
"num_tokens": 6779322.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 7.182950258255005,
|
|
"epoch": 0.39605892108087015,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004995345627934338,
|
|
"loss": 7.0747,
|
|
"mean_token_accuracy": 0.1140649899840355,
|
|
"num_tokens": 6794293.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.839075326919556,
|
|
"epoch": 0.39703443566481317,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004995300301839637,
|
|
"loss": 6.9084,
|
|
"mean_token_accuracy": 0.1312996357679367,
|
|
"num_tokens": 6812880.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 7.124480581283569,
|
|
"epoch": 0.39800995024875624,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004995254756342194,
|
|
"loss": 7.1075,
|
|
"mean_token_accuracy": 0.10809694975614548,
|
|
"num_tokens": 6829230.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 7.089977121353149,
|
|
"epoch": 0.39898546483269925,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004995208991446458,
|
|
"loss": 7.102,
|
|
"mean_token_accuracy": 0.11135954707860947,
|
|
"num_tokens": 6846787.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 7.006796836853027,
|
|
"epoch": 0.39996097941664227,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995163007156902,
|
|
"loss": 6.9749,
|
|
"mean_token_accuracy": 0.11075088903307914,
|
|
"num_tokens": 6862990.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 7.126283025741577,
|
|
"epoch": 0.4009364940005853,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004995116803478018,
|
|
"loss": 7.1272,
|
|
"mean_token_accuracy": 0.1081840455532074,
|
|
"num_tokens": 6879170.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 7.005995607376098,
|
|
"epoch": 0.40191200858452836,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004995070380414321,
|
|
"loss": 7.1092,
|
|
"mean_token_accuracy": 0.10969639047980309,
|
|
"num_tokens": 6895635.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 7.09612774848938,
|
|
"epoch": 0.4028875231684714,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000499502373797035,
|
|
"loss": 7.0205,
|
|
"mean_token_accuracy": 0.11540707424283028,
|
|
"num_tokens": 6912836.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 7.0344482421875,
|
|
"epoch": 0.4038630377524144,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000499497687615066,
|
|
"loss": 7.0476,
|
|
"mean_token_accuracy": 0.11897711828351021,
|
|
"num_tokens": 6929507.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 7.0960619926452635,
|
|
"epoch": 0.4048385523363574,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000499492979495983,
|
|
"loss": 7.0645,
|
|
"mean_token_accuracy": 0.10878148227930069,
|
|
"num_tokens": 6946527.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 7.08494553565979,
|
|
"epoch": 0.4058140669203005,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994882494402462,
|
|
"loss": 7.0716,
|
|
"mean_token_accuracy": 0.11425348073244095,
|
|
"num_tokens": 6962582.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 7.053049182891845,
|
|
"epoch": 0.4067895815042435,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004994834974483178,
|
|
"loss": 7.0636,
|
|
"mean_token_accuracy": 0.11429096385836601,
|
|
"num_tokens": 6979389.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.987600946426392,
|
|
"epoch": 0.4077650960881865,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004994787235206621,
|
|
"loss": 7.0396,
|
|
"mean_token_accuracy": 0.11151788830757141,
|
|
"num_tokens": 6996637.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 7.187871599197388,
|
|
"epoch": 0.4087406106721295,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004994739276577456,
|
|
"loss": 7.1287,
|
|
"mean_token_accuracy": 0.10643508210778237,
|
|
"num_tokens": 7013013.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.9453386783599855,
|
|
"epoch": 0.4097161252560726,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004994691098600369,
|
|
"loss": 7.0909,
|
|
"mean_token_accuracy": 0.11552847772836686,
|
|
"num_tokens": 7031476.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 7.183104228973389,
|
|
"epoch": 0.4106916398400156,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004994642701280068,
|
|
"loss": 7.0001,
|
|
"mean_token_accuracy": 0.11294645816087723,
|
|
"num_tokens": 7047424.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.992200660705566,
|
|
"epoch": 0.4116671544239586,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004994594084621282,
|
|
"loss": 7.0871,
|
|
"mean_token_accuracy": 0.10682062655687333,
|
|
"num_tokens": 7063793.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 7.219247531890869,
|
|
"epoch": 0.41264266900790164,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004994545248628762,
|
|
"loss": 7.1364,
|
|
"mean_token_accuracy": 0.10866177827119827,
|
|
"num_tokens": 7079026.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 7.020672225952149,
|
|
"epoch": 0.4136181835918447,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000499449619330728,
|
|
"loss": 7.0572,
|
|
"mean_token_accuracy": 0.1168307289481163,
|
|
"num_tokens": 7097524.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 7.141960477828979,
|
|
"epoch": 0.41459369817578773,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000499444691866163,
|
|
"loss": 7.1313,
|
|
"mean_token_accuracy": 0.10980155616998673,
|
|
"num_tokens": 7115514.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 7.095918035507202,
|
|
"epoch": 0.41556921275973074,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004994397424696625,
|
|
"loss": 7.0673,
|
|
"mean_token_accuracy": 0.11292248368263244,
|
|
"num_tokens": 7132589.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 7.082249641418457,
|
|
"epoch": 0.4165447273436738,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004994347711417105,
|
|
"loss": 7.082,
|
|
"mean_token_accuracy": 0.10995210707187653,
|
|
"num_tokens": 7149416.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.970026350021362,
|
|
"epoch": 0.41752024192761683,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004994297778827924,
|
|
"loss": 7.018,
|
|
"mean_token_accuracy": 0.11910998299717904,
|
|
"num_tokens": 7165763.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 7.037493944168091,
|
|
"epoch": 0.41849575651155985,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004994247626933964,
|
|
"loss": 7.0108,
|
|
"mean_token_accuracy": 0.12116987556219101,
|
|
"num_tokens": 7183131.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.9932863235473635,
|
|
"epoch": 0.41947127109550286,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004994197255740124,
|
|
"loss": 7.0341,
|
|
"mean_token_accuracy": 0.11638646721839904,
|
|
"num_tokens": 7200467.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 7.079003763198853,
|
|
"epoch": 0.42044678567944593,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004994146665251327,
|
|
"loss": 7.0269,
|
|
"mean_token_accuracy": 0.11671786457300186,
|
|
"num_tokens": 7217311.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.985736274719239,
|
|
"epoch": 0.42142230026338895,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004994095855472516,
|
|
"loss": 7.0325,
|
|
"mean_token_accuracy": 0.11662963330745697,
|
|
"num_tokens": 7235244.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 7.125861310958863,
|
|
"epoch": 0.42239781484733196,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004994044826408656,
|
|
"loss": 7.0606,
|
|
"mean_token_accuracy": 0.10941600650548935,
|
|
"num_tokens": 7252235.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 7.026695823669433,
|
|
"epoch": 0.423373329431275,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004993993578064733,
|
|
"loss": 7.0796,
|
|
"mean_token_accuracy": 0.10855171978473663,
|
|
"num_tokens": 7269367.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 7.164697599411011,
|
|
"epoch": 0.42434884401521805,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993942110445756,
|
|
"loss": 7.0575,
|
|
"mean_token_accuracy": 0.10973318442702293,
|
|
"num_tokens": 7287198.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 7.023011207580566,
|
|
"epoch": 0.42532435859916107,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004993890423556753,
|
|
"loss": 6.9969,
|
|
"mean_token_accuracy": 0.11209728047251702,
|
|
"num_tokens": 7302356.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 7.013686943054199,
|
|
"epoch": 0.4262998731831041,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004993838517402775,
|
|
"loss": 6.992,
|
|
"mean_token_accuracy": 0.11840955764055253,
|
|
"num_tokens": 7318530.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 7.065437269210816,
|
|
"epoch": 0.4272753877670471,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0004993786391988894,
|
|
"loss": 6.9724,
|
|
"mean_token_accuracy": 0.11313622668385506,
|
|
"num_tokens": 7335195.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.926114702224732,
|
|
"epoch": 0.42825090235099017,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004993734047320206,
|
|
"loss": 6.9729,
|
|
"mean_token_accuracy": 0.11200029700994492,
|
|
"num_tokens": 7352233.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 7.071037006378174,
|
|
"epoch": 0.4292264169349332,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004993681483401821,
|
|
"loss": 7.027,
|
|
"mean_token_accuracy": 0.11576457694172859,
|
|
"num_tokens": 7368569.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 7.000790357589722,
|
|
"epoch": 0.4302019315188762,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004993628700238879,
|
|
"loss": 7.0716,
|
|
"mean_token_accuracy": 0.11266482919454575,
|
|
"num_tokens": 7384669.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 7.18484787940979,
|
|
"epoch": 0.4311774461028192,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004993575697836537,
|
|
"loss": 7.0834,
|
|
"mean_token_accuracy": 0.10938332602381706,
|
|
"num_tokens": 7401833.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.963419961929321,
|
|
"epoch": 0.4321529606867623,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004993522476199975,
|
|
"loss": 6.978,
|
|
"mean_token_accuracy": 0.1122715525329113,
|
|
"num_tokens": 7418663.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 7.067373323440552,
|
|
"epoch": 0.4331284752707053,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004993469035334392,
|
|
"loss": 6.9573,
|
|
"mean_token_accuracy": 0.12277843132615089,
|
|
"num_tokens": 7434701.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.984041833877564,
|
|
"epoch": 0.4341039898546483,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000499341537524501,
|
|
"loss": 7.0441,
|
|
"mean_token_accuracy": 0.10746957212686539,
|
|
"num_tokens": 7450154.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.906111478805542,
|
|
"epoch": 0.43507950443859134,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004993361495937074,
|
|
"loss": 6.8559,
|
|
"mean_token_accuracy": 0.13145155981183052,
|
|
"num_tokens": 7468110.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 7.01196985244751,
|
|
"epoch": 0.4360550190225344,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004993307397415848,
|
|
"loss": 6.9797,
|
|
"mean_token_accuracy": 0.11334640309214591,
|
|
"num_tokens": 7482962.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 7.0998986721038815,
|
|
"epoch": 0.4370305336064774,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004993253079686619,
|
|
"loss": 7.1178,
|
|
"mean_token_accuracy": 0.10759365484118462,
|
|
"num_tokens": 7499116.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 7.059480142593384,
|
|
"epoch": 0.43800604819042044,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004993198542754692,
|
|
"loss": 7.0707,
|
|
"mean_token_accuracy": 0.11301630288362503,
|
|
"num_tokens": 7515765.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 7.048842477798462,
|
|
"epoch": 0.43898156277436345,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004993143786625401,
|
|
"loss": 7.0522,
|
|
"mean_token_accuracy": 0.11276320442557335,
|
|
"num_tokens": 7532074.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.91264853477478,
|
|
"epoch": 0.4399570773583065,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004993088811304092,
|
|
"loss": 6.8962,
|
|
"mean_token_accuracy": 0.12252458482980728,
|
|
"num_tokens": 7547704.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 7.015114450454712,
|
|
"epoch": 0.44093259194224954,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004993033616796139,
|
|
"loss": 7.0158,
|
|
"mean_token_accuracy": 0.11404076218605042,
|
|
"num_tokens": 7563317.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.961794090270996,
|
|
"epoch": 0.44190810652619256,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004992978203106936,
|
|
"loss": 7.0042,
|
|
"mean_token_accuracy": 0.11175615265965462,
|
|
"num_tokens": 7580770.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 7.016242122650146,
|
|
"epoch": 0.4428836211101356,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004992922570241896,
|
|
"loss": 6.9809,
|
|
"mean_token_accuracy": 0.11705166772007942,
|
|
"num_tokens": 7598181.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 7.028905820846558,
|
|
"epoch": 0.44385913569407864,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004992866718206457,
|
|
"loss": 6.9722,
|
|
"mean_token_accuracy": 0.11778600364923478,
|
|
"num_tokens": 7613739.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 7.025203371047974,
|
|
"epoch": 0.44483465027802166,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004992810647006077,
|
|
"loss": 7.0563,
|
|
"mean_token_accuracy": 0.11031661182641983,
|
|
"num_tokens": 7629297.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 7.087608432769775,
|
|
"epoch": 0.4458101648619647,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004992754356646234,
|
|
"loss": 7.049,
|
|
"mean_token_accuracy": 0.110834738612175,
|
|
"num_tokens": 7645273.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 7.067881631851196,
|
|
"epoch": 0.4467856794459077,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004992697847132427,
|
|
"loss": 7.0014,
|
|
"mean_token_accuracy": 0.10971406400203705,
|
|
"num_tokens": 7661296.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.86855297088623,
|
|
"epoch": 0.44776119402985076,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000499264111847018,
|
|
"loss": 7.0023,
|
|
"mean_token_accuracy": 0.11386808529496192,
|
|
"num_tokens": 7678658.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 7.125934982299805,
|
|
"epoch": 0.4487367086137938,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004992584170665037,
|
|
"loss": 7.0306,
|
|
"mean_token_accuracy": 0.1094054788351059,
|
|
"num_tokens": 7693678.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.985922002792359,
|
|
"epoch": 0.4497122231977368,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000499252700372256,
|
|
"loss": 7.0232,
|
|
"mean_token_accuracy": 0.11486716344952583,
|
|
"num_tokens": 7709526.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 7.020515871047974,
|
|
"epoch": 0.4506877377816798,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004992469617648337,
|
|
"loss": 6.9265,
|
|
"mean_token_accuracy": 0.11158730164170265,
|
|
"num_tokens": 7725055.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.868546867370606,
|
|
"epoch": 0.4516632523656229,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004992412012447976,
|
|
"loss": 6.8277,
|
|
"mean_token_accuracy": 0.13100250884890557,
|
|
"num_tokens": 7742047.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 7.032267808914185,
|
|
"epoch": 0.4526387669495659,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004992354188127105,
|
|
"loss": 7.0045,
|
|
"mean_token_accuracy": 0.11193139106035233,
|
|
"num_tokens": 7758283.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.975220394134522,
|
|
"epoch": 0.4536142815335089,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004992296144691375,
|
|
"loss": 6.9442,
|
|
"mean_token_accuracy": 0.1183014951646328,
|
|
"num_tokens": 7774660.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.93505334854126,
|
|
"epoch": 0.454589796117452,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992237882146456,
|
|
"loss": 6.9561,
|
|
"mean_token_accuracy": 0.11839842200279235,
|
|
"num_tokens": 7792147.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.9103350162506105,
|
|
"epoch": 0.455565310701395,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004992179400498044,
|
|
"loss": 6.8987,
|
|
"mean_token_accuracy": 0.1161189466714859,
|
|
"num_tokens": 7809021.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 7.007769536972046,
|
|
"epoch": 0.456540825285338,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0004992120699751852,
|
|
"loss": 6.9624,
|
|
"mean_token_accuracy": 0.11379743441939354,
|
|
"num_tokens": 7825074.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.837625646591187,
|
|
"epoch": 0.45751633986928103,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004992061779913616,
|
|
"loss": 6.8918,
|
|
"mean_token_accuracy": 0.12118960246443748,
|
|
"num_tokens": 7843413.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 7.060509872436524,
|
|
"epoch": 0.4584918544532241,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004992002640989094,
|
|
"loss": 6.9842,
|
|
"mean_token_accuracy": 0.10777736902236938,
|
|
"num_tokens": 7859874.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.888943815231324,
|
|
"epoch": 0.4594673690371671,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004991943282984065,
|
|
"loss": 6.9176,
|
|
"mean_token_accuracy": 0.11951580569148064,
|
|
"num_tokens": 7877302.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 7.015088510513306,
|
|
"epoch": 0.46044288362111013,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004991883705904328,
|
|
"loss": 7.0317,
|
|
"mean_token_accuracy": 0.10920384526252747,
|
|
"num_tokens": 7893207.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.905005931854248,
|
|
"epoch": 0.46141839820505315,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004991823909755705,
|
|
"loss": 6.8904,
|
|
"mean_token_accuracy": 0.12214830294251441,
|
|
"num_tokens": 7910239.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 7.027743482589722,
|
|
"epoch": 0.4623939127889962,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004991763894544041,
|
|
"loss": 7.0056,
|
|
"mean_token_accuracy": 0.11172575429081917,
|
|
"num_tokens": 7928030.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.973042011260986,
|
|
"epoch": 0.46336942737293924,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004991703660275199,
|
|
"loss": 6.9695,
|
|
"mean_token_accuracy": 0.11384573951363564,
|
|
"num_tokens": 7943653.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 7.05312213897705,
|
|
"epoch": 0.46434494195688225,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004991643206955064,
|
|
"loss": 7.0329,
|
|
"mean_token_accuracy": 0.11237467974424362,
|
|
"num_tokens": 7960564.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 7.014554071426391,
|
|
"epoch": 0.46532045654082527,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004991582534589544,
|
|
"loss": 6.9782,
|
|
"mean_token_accuracy": 0.11644384115934373,
|
|
"num_tokens": 7978045.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 7.0076799392700195,
|
|
"epoch": 0.46629597112476834,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004991521643184567,
|
|
"loss": 7.0233,
|
|
"mean_token_accuracy": 0.11004308089613915,
|
|
"num_tokens": 7995074.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 7.057988786697388,
|
|
"epoch": 0.46727148570871135,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004991460532746087,
|
|
"loss": 7.0417,
|
|
"mean_token_accuracy": 0.10691535025835038,
|
|
"num_tokens": 8011800.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 7.042518377304077,
|
|
"epoch": 0.46824700029265437,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000499139920328007,
|
|
"loss": 7.0164,
|
|
"mean_token_accuracy": 0.11115067452192307,
|
|
"num_tokens": 8028090.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.991872024536133,
|
|
"epoch": 0.4692225148765974,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004991337654792512,
|
|
"loss": 6.9636,
|
|
"mean_token_accuracy": 0.1162964753806591,
|
|
"num_tokens": 8044140.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.921262788772583,
|
|
"epoch": 0.47019802946054046,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004991275887289427,
|
|
"loss": 6.9011,
|
|
"mean_token_accuracy": 0.11688902750611305,
|
|
"num_tokens": 8060030.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.954582357406617,
|
|
"epoch": 0.47117354404448347,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004991213900776849,
|
|
"loss": 6.9477,
|
|
"mean_token_accuracy": 0.11353798434138299,
|
|
"num_tokens": 8075445.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 7.039027690887451,
|
|
"epoch": 0.4721490586284265,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004991151695260838,
|
|
"loss": 6.9547,
|
|
"mean_token_accuracy": 0.11307838335633277,
|
|
"num_tokens": 8093250.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.871022033691406,
|
|
"epoch": 0.4731245732123695,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000499108927074747,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.1290725477039814,
|
|
"num_tokens": 8109615.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.977431726455689,
|
|
"epoch": 0.4741000877963126,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004991026627242846,
|
|
"loss": 7.007,
|
|
"mean_token_accuracy": 0.1094487626105547,
|
|
"num_tokens": 8126591.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.984453248977661,
|
|
"epoch": 0.4750756023802556,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004990963764753088,
|
|
"loss": 6.9201,
|
|
"mean_token_accuracy": 0.11406318247318267,
|
|
"num_tokens": 8141360.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.96773853302002,
|
|
"epoch": 0.4760511169641986,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004990900683284337,
|
|
"loss": 7.0228,
|
|
"mean_token_accuracy": 0.11445973664522172,
|
|
"num_tokens": 8159688.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.959125900268555,
|
|
"epoch": 0.4770266315481416,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004990837382842758,
|
|
"loss": 6.9831,
|
|
"mean_token_accuracy": 0.11357531175017357,
|
|
"num_tokens": 8176887.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.870701313018799,
|
|
"epoch": 0.4780021461320847,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004990773863434537,
|
|
"loss": 6.8034,
|
|
"mean_token_accuracy": 0.12336878180503845,
|
|
"num_tokens": 8193372.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.963567733764648,
|
|
"epoch": 0.4789776607160277,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004990710125065879,
|
|
"loss": 7.0044,
|
|
"mean_token_accuracy": 0.1109432339668274,
|
|
"num_tokens": 8209713.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 7.021219301223755,
|
|
"epoch": 0.4799531752999707,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004990646167743015,
|
|
"loss": 7.0058,
|
|
"mean_token_accuracy": 0.11337054893374443,
|
|
"num_tokens": 8226359.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.926394844055176,
|
|
"epoch": 0.48092868988391374,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004990581991472193,
|
|
"loss": 6.9675,
|
|
"mean_token_accuracy": 0.11650057286024093,
|
|
"num_tokens": 8243182.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.998376560211182,
|
|
"epoch": 0.4819042044678568,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004990517596259684,
|
|
"loss": 6.9843,
|
|
"mean_token_accuracy": 0.11314982771873475,
|
|
"num_tokens": 8259185.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 7.04877028465271,
|
|
"epoch": 0.4828797190517998,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004990452982111781,
|
|
"loss": 7.0446,
|
|
"mean_token_accuracy": 0.10870871022343635,
|
|
"num_tokens": 8275941.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.96405930519104,
|
|
"epoch": 0.48385523363574284,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004990388149034797,
|
|
"loss": 6.9846,
|
|
"mean_token_accuracy": 0.11402082219719886,
|
|
"num_tokens": 8292535.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.987648439407349,
|
|
"epoch": 0.48483074821968586,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004990323097035069,
|
|
"loss": 6.9301,
|
|
"mean_token_accuracy": 0.11991023421287536,
|
|
"num_tokens": 8310562.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.940241813659668,
|
|
"epoch": 0.48580626280362893,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.0004990257826118952,
|
|
"loss": 6.9636,
|
|
"mean_token_accuracy": 0.10830035284161568,
|
|
"num_tokens": 8327241.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.913412094116211,
|
|
"epoch": 0.48678177738757195,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.0004990192336292824,
|
|
"loss": 6.9088,
|
|
"mean_token_accuracy": 0.11822234392166138,
|
|
"num_tokens": 8343516.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.8137157440185545,
|
|
"epoch": 0.48775729197151496,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004990126627563087,
|
|
"loss": 6.7738,
|
|
"mean_token_accuracy": 0.12430693060159684,
|
|
"num_tokens": 8361809.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.875748491287231,
|
|
"epoch": 0.488732806555458,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004990060699936159,
|
|
"loss": 6.8307,
|
|
"mean_token_accuracy": 0.12605192884802818,
|
|
"num_tokens": 8378632.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.968320083618164,
|
|
"epoch": 0.48970832113940105,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004989994553418483,
|
|
"loss": 6.9914,
|
|
"mean_token_accuracy": 0.10726674050092697,
|
|
"num_tokens": 8396201.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.940235614776611,
|
|
"epoch": 0.49068383572334406,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004989928188016522,
|
|
"loss": 6.8782,
|
|
"mean_token_accuracy": 0.11905882507562637,
|
|
"num_tokens": 8412851.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.7344177722930905,
|
|
"epoch": 0.4916593503072871,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004989861603736762,
|
|
"loss": 6.7232,
|
|
"mean_token_accuracy": 0.12667507529258729,
|
|
"num_tokens": 8430026.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.928332138061523,
|
|
"epoch": 0.49263486489123015,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004989794800585711,
|
|
"loss": 6.9486,
|
|
"mean_token_accuracy": 0.11339501067996025,
|
|
"num_tokens": 8446696.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 7.059392595291138,
|
|
"epoch": 0.49361037947517317,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004989727778569893,
|
|
"loss": 7.0275,
|
|
"mean_token_accuracy": 0.11218074709177017,
|
|
"num_tokens": 8464827.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.951289224624634,
|
|
"epoch": 0.4945858940591162,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004989660537695859,
|
|
"loss": 6.9867,
|
|
"mean_token_accuracy": 0.1131703607738018,
|
|
"num_tokens": 8480185.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.900561904907226,
|
|
"epoch": 0.4955614086430592,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.000498959307797018,
|
|
"loss": 6.8754,
|
|
"mean_token_accuracy": 0.12406175062060357,
|
|
"num_tokens": 8497511.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.917745304107666,
|
|
"epoch": 0.49653692322700227,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004989525399399447,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.11667040959000588,
|
|
"num_tokens": 8514394.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.989028596878052,
|
|
"epoch": 0.4975124378109453,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004989457501990275,
|
|
"loss": 6.9368,
|
|
"mean_token_accuracy": 0.11208094283938408,
|
|
"num_tokens": 8531659.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.9539367198944095,
|
|
"epoch": 0.4984879523948883,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004989389385749297,
|
|
"loss": 6.9859,
|
|
"mean_token_accuracy": 0.11457662954926491,
|
|
"num_tokens": 8547986.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 7.001014041900635,
|
|
"epoch": 0.4994634669788313,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.0004989321050683169,
|
|
"loss": 6.9339,
|
|
"mean_token_accuracy": 0.11468060165643693,
|
|
"num_tokens": 8565021.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.915047597885132,
|
|
"epoch": 0.5004389815627743,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000498925249679857,
|
|
"loss": 6.9396,
|
|
"mean_token_accuracy": 0.11381619796156883,
|
|
"num_tokens": 8582135.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.90760645866394,
|
|
"epoch": 0.5014144961467174,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004989183724102196,
|
|
"loss": 6.9589,
|
|
"mean_token_accuracy": 0.11705165952444077,
|
|
"num_tokens": 8599946.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 7.0192211151123045,
|
|
"epoch": 0.5023900107306605,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0004989114732600771,
|
|
"loss": 6.983,
|
|
"mean_token_accuracy": 0.110711020976305,
|
|
"num_tokens": 8615831.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.89142541885376,
|
|
"epoch": 0.5033655253146034,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004989045522301033,
|
|
"loss": 6.8671,
|
|
"mean_token_accuracy": 0.11825199350714684,
|
|
"num_tokens": 8631886.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.981450748443604,
|
|
"epoch": 0.5043410398985465,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004988976093209747,
|
|
"loss": 6.9144,
|
|
"mean_token_accuracy": 0.11438037380576134,
|
|
"num_tokens": 8647520.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.870293664932251,
|
|
"epoch": 0.5053165544824895,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004988906445333698,
|
|
"loss": 6.8368,
|
|
"mean_token_accuracy": 0.12467891722917557,
|
|
"num_tokens": 8663995.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.890131139755249,
|
|
"epoch": 0.5062920690664325,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004988836578679691,
|
|
"loss": 6.9143,
|
|
"mean_token_accuracy": 0.11203564777970314,
|
|
"num_tokens": 8680201.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.963378143310547,
|
|
"epoch": 0.5072675836503756,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0004988766493254552,
|
|
"loss": 6.9014,
|
|
"mean_token_accuracy": 0.12187715098261834,
|
|
"num_tokens": 8696978.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.8637247562408445,
|
|
"epoch": 0.5082430982343186,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000498869618906513,
|
|
"loss": 6.9428,
|
|
"mean_token_accuracy": 0.11771589368581772,
|
|
"num_tokens": 8713335.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.99650011062622,
|
|
"epoch": 0.5092186128182616,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004988625666118296,
|
|
"loss": 6.9406,
|
|
"mean_token_accuracy": 0.11646446585655212,
|
|
"num_tokens": 8730722.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.877693843841553,
|
|
"epoch": 0.5101941274022047,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000498855492442094,
|
|
"loss": 6.8641,
|
|
"mean_token_accuracy": 0.11773835569620132,
|
|
"num_tokens": 8747430.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.937593746185303,
|
|
"epoch": 0.5111696419861477,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004988483963979975,
|
|
"loss": 6.9526,
|
|
"mean_token_accuracy": 0.11680929362773895,
|
|
"num_tokens": 8764817.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 7.029442834854126,
|
|
"epoch": 0.5121451565700907,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004988412784802336,
|
|
"loss": 6.9674,
|
|
"mean_token_accuracy": 0.11305484101176262,
|
|
"num_tokens": 8780412.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.956655883789063,
|
|
"epoch": 0.5131206711540337,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004988341386894977,
|
|
"loss": 7.0367,
|
|
"mean_token_accuracy": 0.11347447261214257,
|
|
"num_tokens": 8795526.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.934908723831176,
|
|
"epoch": 0.5140961857379768,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004988269770264876,
|
|
"loss": 6.8335,
|
|
"mean_token_accuracy": 0.12252563685178756,
|
|
"num_tokens": 8813568.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.978274250030518,
|
|
"epoch": 0.5150717003219198,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004988197934919029,
|
|
"loss": 6.984,
|
|
"mean_token_accuracy": 0.11146572455763817,
|
|
"num_tokens": 8831807.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 7.024217653274536,
|
|
"epoch": 0.5160472149058628,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004988125880864458,
|
|
"loss": 6.9811,
|
|
"mean_token_accuracy": 0.11375662460923194,
|
|
"num_tokens": 8848875.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.793162870407104,
|
|
"epoch": 0.5170227294898059,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004988053608108203,
|
|
"loss": 6.6888,
|
|
"mean_token_accuracy": 0.12523928061127662,
|
|
"num_tokens": 8866317.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.868563032150268,
|
|
"epoch": 0.517998244073749,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004987981116657325,
|
|
"loss": 6.8371,
|
|
"mean_token_accuracy": 0.11552531272172928,
|
|
"num_tokens": 8882543.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.922529649734497,
|
|
"epoch": 0.5189737586576919,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004987908406518909,
|
|
"loss": 6.8399,
|
|
"mean_token_accuracy": 0.11445399224758149,
|
|
"num_tokens": 8898640.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.936803960800171,
|
|
"epoch": 0.519949273241635,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000498783547770006,
|
|
"loss": 6.9548,
|
|
"mean_token_accuracy": 0.11621619910001754,
|
|
"num_tokens": 8914649.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.978597927093506,
|
|
"epoch": 0.5209247878255779,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987762330207903,
|
|
"loss": 6.9706,
|
|
"mean_token_accuracy": 0.11318715661764145,
|
|
"num_tokens": 8931585.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.846628665924072,
|
|
"epoch": 0.521900302409521,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004987688964049587,
|
|
"loss": 6.8613,
|
|
"mean_token_accuracy": 0.12094739824533463,
|
|
"num_tokens": 8948827.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.8830482959747314,
|
|
"epoch": 0.5228758169934641,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000498761537923228,
|
|
"loss": 6.8239,
|
|
"mean_token_accuracy": 0.12365922555327416,
|
|
"num_tokens": 8965680.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.981524515151977,
|
|
"epoch": 0.523851331577407,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 0.0004987541575763175,
|
|
"loss": 6.9672,
|
|
"mean_token_accuracy": 0.1103598140180111,
|
|
"num_tokens": 8982813.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.949263191223144,
|
|
"epoch": 0.5248268461613501,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000498746755364948,
|
|
"loss": 6.9504,
|
|
"mean_token_accuracy": 0.11520707830786706,
|
|
"num_tokens": 9000497.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.882326602935791,
|
|
"epoch": 0.5258023607452932,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004987393312898431,
|
|
"loss": 6.8758,
|
|
"mean_token_accuracy": 0.11944987401366233,
|
|
"num_tokens": 9016195.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.863258600234985,
|
|
"epoch": 0.5267778753292361,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004987318853517282,
|
|
"loss": 6.8335,
|
|
"mean_token_accuracy": 0.125432301312685,
|
|
"num_tokens": 9033295.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.933648490905762,
|
|
"epoch": 0.5277533899131792,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004987244175513308,
|
|
"loss": 6.9282,
|
|
"mean_token_accuracy": 0.11364365071058273,
|
|
"num_tokens": 9050725.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.941001844406128,
|
|
"epoch": 0.5287289044971223,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004987169278893807,
|
|
"loss": 6.8772,
|
|
"mean_token_accuracy": 0.11788134574890137,
|
|
"num_tokens": 9067842.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.962023401260376,
|
|
"epoch": 0.5297044190810652,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004987094163666098,
|
|
"loss": 6.9033,
|
|
"mean_token_accuracy": 0.10963810235261917,
|
|
"num_tokens": 9084602.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.882829856872559,
|
|
"epoch": 0.5306799336650083,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000498701882983752,
|
|
"loss": 6.8256,
|
|
"mean_token_accuracy": 0.12473871782422066,
|
|
"num_tokens": 9100530.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.901415729522705,
|
|
"epoch": 0.5316554482489513,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004986943277415435,
|
|
"loss": 6.9164,
|
|
"mean_token_accuracy": 0.1212962508201599,
|
|
"num_tokens": 9117304.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.846544599533081,
|
|
"epoch": 0.5326309628328944,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0004986867506407225,
|
|
"loss": 6.8622,
|
|
"mean_token_accuracy": 0.13095585703849794,
|
|
"num_tokens": 9133974.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.914484453201294,
|
|
"epoch": 0.5336064774168374,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004986791516820295,
|
|
"loss": 6.8267,
|
|
"mean_token_accuracy": 0.12075289562344552,
|
|
"num_tokens": 9150600.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.834219551086425,
|
|
"epoch": 0.5345819920007804,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.000498671530866207,
|
|
"loss": 6.8344,
|
|
"mean_token_accuracy": 0.11510834842920303,
|
|
"num_tokens": 9167449.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.90480670928955,
|
|
"epoch": 0.5355575065847235,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004986638881939997,
|
|
"loss": 6.7687,
|
|
"mean_token_accuracy": 0.12122940719127655,
|
|
"num_tokens": 9184570.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.853910255432129,
|
|
"epoch": 0.5365330211686665,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004986562236661544,
|
|
"loss": 6.961,
|
|
"mean_token_accuracy": 0.10868783816695213,
|
|
"num_tokens": 9200537.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 7.009225177764892,
|
|
"epoch": 0.5375085357526095,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00049864853728342,
|
|
"loss": 6.866,
|
|
"mean_token_accuracy": 0.12115592211484909,
|
|
"num_tokens": 9217252.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.878779077529908,
|
|
"epoch": 0.5384840503365526,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004986408290465476,
|
|
"loss": 6.9131,
|
|
"mean_token_accuracy": 0.11528742164373398,
|
|
"num_tokens": 9234108.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.938559627532959,
|
|
"epoch": 0.5394595649204955,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004986330989562906,
|
|
"loss": 6.8108,
|
|
"mean_token_accuracy": 0.12083505392074585,
|
|
"num_tokens": 9251017.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.805785608291626,
|
|
"epoch": 0.5404350795044386,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004986253470134042,
|
|
"loss": 6.7762,
|
|
"mean_token_accuracy": 0.1270703725516796,
|
|
"num_tokens": 9267922.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.761190700531006,
|
|
"epoch": 0.5414105940883817,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0004986175732186459,
|
|
"loss": 6.7694,
|
|
"mean_token_accuracy": 0.1298803262412548,
|
|
"num_tokens": 9284954.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.956725168228149,
|
|
"epoch": 0.5423861086723246,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004986097775727752,
|
|
"loss": 6.8884,
|
|
"mean_token_accuracy": 0.11942665055394172,
|
|
"num_tokens": 9303306.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.79859299659729,
|
|
"epoch": 0.5433616232562677,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004986019600765541,
|
|
"loss": 6.7824,
|
|
"mean_token_accuracy": 0.1234261803328991,
|
|
"num_tokens": 9319754.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.857258653640747,
|
|
"epoch": 0.5443371378402108,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004985941207307462,
|
|
"loss": 6.8724,
|
|
"mean_token_accuracy": 0.120131453871727,
|
|
"num_tokens": 9335736.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.854156827926635,
|
|
"epoch": 0.5453126524241537,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004985862595361179,
|
|
"loss": 6.8434,
|
|
"mean_token_accuracy": 0.12304849475622177,
|
|
"num_tokens": 9350435.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.9249341011047365,
|
|
"epoch": 0.5462881670080968,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004985783764934372,
|
|
"loss": 6.9191,
|
|
"mean_token_accuracy": 0.11444221660494805,
|
|
"num_tokens": 9367613.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.848372459411621,
|
|
"epoch": 0.5472636815920398,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004985704716034744,
|
|
"loss": 6.7859,
|
|
"mean_token_accuracy": 0.12150981798768043,
|
|
"num_tokens": 9385590.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.905705404281616,
|
|
"epoch": 0.5482391961759828,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004985625448670019,
|
|
"loss": 6.9263,
|
|
"mean_token_accuracy": 0.10945390835404396,
|
|
"num_tokens": 9403286.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.994100189208984,
|
|
"epoch": 0.5492147107599259,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004985545962847943,
|
|
"loss": 6.9189,
|
|
"mean_token_accuracy": 0.11298392787575721,
|
|
"num_tokens": 9419880.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.9102226257324215,
|
|
"epoch": 0.5501902253438689,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004985466258576283,
|
|
"loss": 6.8775,
|
|
"mean_token_accuracy": 0.11068576276302337,
|
|
"num_tokens": 9436579.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.943197536468506,
|
|
"epoch": 0.5511657399278119,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004985386335862829,
|
|
"loss": 6.9511,
|
|
"mean_token_accuracy": 0.11212421581149101,
|
|
"num_tokens": 9452802.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.9377796173095705,
|
|
"epoch": 0.552141254511755,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004985306194715387,
|
|
"loss": 6.9462,
|
|
"mean_token_accuracy": 0.11190766394138336,
|
|
"num_tokens": 9468297.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.920965957641601,
|
|
"epoch": 0.553116769095698,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004985225835141792,
|
|
"loss": 6.9673,
|
|
"mean_token_accuracy": 0.10958375483751297,
|
|
"num_tokens": 9484954.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 7.036347913742065,
|
|
"epoch": 0.554092283679641,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004985145257149895,
|
|
"loss": 7.0055,
|
|
"mean_token_accuracy": 0.10926211625337601,
|
|
"num_tokens": 9502211.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.92486252784729,
|
|
"epoch": 0.555067798263584,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004985064460747569,
|
|
"loss": 6.9012,
|
|
"mean_token_accuracy": 0.11126281246542931,
|
|
"num_tokens": 9518103.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.856705045700073,
|
|
"epoch": 0.5560433128475271,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000498498344594271,
|
|
"loss": 6.8237,
|
|
"mean_token_accuracy": 0.11587432101368904,
|
|
"num_tokens": 9534155.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.8256289005279545,
|
|
"epoch": 0.5570188274314701,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004984902212743234,
|
|
"loss": 6.8099,
|
|
"mean_token_accuracy": 0.12739480584859847,
|
|
"num_tokens": 9549972.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.8014082431793215,
|
|
"epoch": 0.5579943420154131,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004984820761157079,
|
|
"loss": 6.736,
|
|
"mean_token_accuracy": 0.13063260465860366,
|
|
"num_tokens": 9567306.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.940472412109375,
|
|
"epoch": 0.5589698565993562,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004984739091192205,
|
|
"loss": 6.9327,
|
|
"mean_token_accuracy": 0.1139756128191948,
|
|
"num_tokens": 9583488.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.727579021453858,
|
|
"epoch": 0.5599453711832992,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004984657202856592,
|
|
"loss": 6.7233,
|
|
"mean_token_accuracy": 0.12913710996508598,
|
|
"num_tokens": 9600030.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.848877954483032,
|
|
"epoch": 0.5609208857672422,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000498457509615824,
|
|
"loss": 6.842,
|
|
"mean_token_accuracy": 0.12119656652212143,
|
|
"num_tokens": 9615791.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.978024339675903,
|
|
"epoch": 0.5618964003511853,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004984492771105175,
|
|
"loss": 6.9141,
|
|
"mean_token_accuracy": 0.11272858902812004,
|
|
"num_tokens": 9630955.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.791843271255493,
|
|
"epoch": 0.5628719149351282,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004984410227705439,
|
|
"loss": 6.8307,
|
|
"mean_token_accuracy": 0.12661578208208085,
|
|
"num_tokens": 9647743.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.833824110031128,
|
|
"epoch": 0.5638474295190713,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.0004984327465967101,
|
|
"loss": 6.7632,
|
|
"mean_token_accuracy": 0.12368033602833747,
|
|
"num_tokens": 9665495.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.867427635192871,
|
|
"epoch": 0.5648229441030144,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004984244485898246,
|
|
"loss": 6.7957,
|
|
"mean_token_accuracy": 0.12146139666438102,
|
|
"num_tokens": 9683489.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.681371450424194,
|
|
"epoch": 0.5657984586869573,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0004984161287506982,
|
|
"loss": 6.7306,
|
|
"mean_token_accuracy": 0.13925604596734048,
|
|
"num_tokens": 9700491.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.976580095291138,
|
|
"epoch": 0.5667739732709004,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000498407787080144,
|
|
"loss": 6.8928,
|
|
"mean_token_accuracy": 0.11349690333008766,
|
|
"num_tokens": 9715694.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.8997095108032225,
|
|
"epoch": 0.5677494878548435,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000498399423578977,
|
|
"loss": 6.9487,
|
|
"mean_token_accuracy": 0.11223901212215423,
|
|
"num_tokens": 9731317.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 7.015579509735107,
|
|
"epoch": 0.5687250024387864,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004983910382480146,
|
|
"loss": 6.919,
|
|
"mean_token_accuracy": 0.11128888130187989,
|
|
"num_tokens": 9747928.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.893016815185547,
|
|
"epoch": 0.5697005170227295,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004983826310880762,
|
|
"loss": 6.8544,
|
|
"mean_token_accuracy": 0.11194843426346779,
|
|
"num_tokens": 9765454.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.965768766403198,
|
|
"epoch": 0.5706760316066726,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004983742020999832,
|
|
"loss": 6.9344,
|
|
"mean_token_accuracy": 0.10903514623641967,
|
|
"num_tokens": 9782255.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.727192163467407,
|
|
"epoch": 0.5716515461906155,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004983657512845593,
|
|
"loss": 6.7417,
|
|
"mean_token_accuracy": 0.1272973582148552,
|
|
"num_tokens": 9798680.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.887448167800903,
|
|
"epoch": 0.5726270607745586,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004983572786426302,
|
|
"loss": 6.882,
|
|
"mean_token_accuracy": 0.115676748752594,
|
|
"num_tokens": 9813955.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.7411586284637455,
|
|
"epoch": 0.5736025753585016,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004983487841750239,
|
|
"loss": 6.7617,
|
|
"mean_token_accuracy": 0.1255027547478676,
|
|
"num_tokens": 9830692.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.937388372421265,
|
|
"epoch": 0.5745780899424446,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004983402678825705,
|
|
"loss": 6.885,
|
|
"mean_token_accuracy": 0.11392842903733254,
|
|
"num_tokens": 9848825.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.873228216171265,
|
|
"epoch": 0.5755536045263877,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004983317297661021,
|
|
"loss": 6.8859,
|
|
"mean_token_accuracy": 0.11650989279150963,
|
|
"num_tokens": 9865802.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.91488995552063,
|
|
"epoch": 0.5765291191103307,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004983231698264531,
|
|
"loss": 6.9293,
|
|
"mean_token_accuracy": 0.11482540965080261,
|
|
"num_tokens": 9883540.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.936852312088012,
|
|
"epoch": 0.5775046336942737,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004983145880644597,
|
|
"loss": 6.8625,
|
|
"mean_token_accuracy": 0.11717101335525512,
|
|
"num_tokens": 9897724.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.896431303024292,
|
|
"epoch": 0.5784801482782168,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004983059844809608,
|
|
"loss": 6.8137,
|
|
"mean_token_accuracy": 0.12274272590875626,
|
|
"num_tokens": 9914710.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.854437398910522,
|
|
"epoch": 0.5794556628621598,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004982973590767969,
|
|
"loss": 6.9261,
|
|
"mean_token_accuracy": 0.11889142394065857,
|
|
"num_tokens": 9932362.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.84909086227417,
|
|
"epoch": 0.5804311774461028,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000498288711852811,
|
|
"loss": 6.7269,
|
|
"mean_token_accuracy": 0.12317892163991928,
|
|
"num_tokens": 9949409.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.850289630889892,
|
|
"epoch": 0.5814066920300458,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000498280042809848,
|
|
"loss": 6.8803,
|
|
"mean_token_accuracy": 0.11638240292668342,
|
|
"num_tokens": 9967467.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.904487228393554,
|
|
"epoch": 0.5823822066139889,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.0004982713519487549,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.1210539199411869,
|
|
"num_tokens": 9983082.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.827433156967163,
|
|
"epoch": 0.5833577211979319,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004982626392703812,
|
|
"loss": 6.8852,
|
|
"mean_token_accuracy": 0.11658342108130455,
|
|
"num_tokens": 10000046.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.890923309326172,
|
|
"epoch": 0.5843332357818749,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000498253904775578,
|
|
"loss": 6.8249,
|
|
"mean_token_accuracy": 0.11711288690567016,
|
|
"num_tokens": 10014674.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.777483081817627,
|
|
"epoch": 0.585308750365818,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004982451484651988,
|
|
"loss": 6.7075,
|
|
"mean_token_accuracy": 0.12117704004049301,
|
|
"num_tokens": 10030647.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 0.585308750365818,
|
|
"eval_entropy": 6.687477521946967,
|
|
"eval_loss": 6.875298976898193,
|
|
"eval_mean_token_accuracy": 0.1196487419652266,
|
|
"eval_num_tokens": 10030647.0,
|
|
"eval_runtime": 15.4221,
|
|
"eval_samples_per_second": 1074.886,
|
|
"eval_steps_per_second": 134.417,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 51250,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 3000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.3581706696704e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|