5035 lines
136 KiB
JSON
5035 lines
136 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.1452770433216143,
|
|
"eval_steps": 500,
|
|
"global_step": 2500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692013454437255,
|
|
"epoch": 0.0002905540866432286,
|
|
"grad_norm": 12.125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8253,
|
|
"mean_token_accuracy": 0.00011534024961292743,
|
|
"num_tokens": 10788.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.692022037506103,
|
|
"epoch": 0.0005811081732864572,
|
|
"grad_norm": 12.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7714,
|
|
"mean_token_accuracy": 0.0003569915541447699,
|
|
"num_tokens": 21530.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691065692901612,
|
|
"epoch": 0.0008716622599296859,
|
|
"grad_norm": 9.0625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5249,
|
|
"mean_token_accuracy": 0.024039586074650288,
|
|
"num_tokens": 31586.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.681152439117431,
|
|
"epoch": 0.0011622163465729145,
|
|
"grad_norm": 6.59375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1914,
|
|
"mean_token_accuracy": 0.04179190285503864,
|
|
"num_tokens": 40558.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.632322597503663,
|
|
"epoch": 0.0014527704332161432,
|
|
"grad_norm": 3.84375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9485,
|
|
"mean_token_accuracy": 0.04313154201954603,
|
|
"num_tokens": 50140.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.600634860992432,
|
|
"epoch": 0.0017433245198593718,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7726,
|
|
"mean_token_accuracy": 0.04188743010163307,
|
|
"num_tokens": 59092.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.610234546661378,
|
|
"epoch": 0.0020338786065026006,
|
|
"grad_norm": 2.796875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7306,
|
|
"mean_token_accuracy": 0.03705091737210751,
|
|
"num_tokens": 68279.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.6115083694458,
|
|
"epoch": 0.002324432693145829,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7246,
|
|
"mean_token_accuracy": 0.03739018738269806,
|
|
"num_tokens": 79896.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.594012641906739,
|
|
"epoch": 0.0026149867797890577,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.6422,
|
|
"mean_token_accuracy": 0.04058306273072958,
|
|
"num_tokens": 89336.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.573887252807618,
|
|
"epoch": 0.0029055408664322865,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5546,
|
|
"mean_token_accuracy": 0.04180898889899254,
|
|
"num_tokens": 98466.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.593281364440918,
|
|
"epoch": 0.003196094953075515,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5105,
|
|
"mean_token_accuracy": 0.04817579984664917,
|
|
"num_tokens": 108540.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.526645183563232,
|
|
"epoch": 0.0034866490397187436,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4716,
|
|
"mean_token_accuracy": 0.050264496356248856,
|
|
"num_tokens": 118750.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.412051582336426,
|
|
"epoch": 0.0037772031263619723,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3319,
|
|
"mean_token_accuracy": 0.05254916958510876,
|
|
"num_tokens": 127809.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.48470401763916,
|
|
"epoch": 0.004067757213005201,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2973,
|
|
"mean_token_accuracy": 0.05050221066921949,
|
|
"num_tokens": 137280.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.446945762634277,
|
|
"epoch": 0.0043583112996484295,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2174,
|
|
"mean_token_accuracy": 0.050593961775302884,
|
|
"num_tokens": 147299.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.421859169006348,
|
|
"epoch": 0.004648865386291658,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.136,
|
|
"mean_token_accuracy": 0.05404375270009041,
|
|
"num_tokens": 157174.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.441582870483398,
|
|
"epoch": 0.004939419472934887,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0058,
|
|
"mean_token_accuracy": 0.05041108168661594,
|
|
"num_tokens": 167491.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.385116672515869,
|
|
"epoch": 0.005229973559578115,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9325,
|
|
"mean_token_accuracy": 0.05533732958137989,
|
|
"num_tokens": 177873.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.342092037200928,
|
|
"epoch": 0.005520527646221344,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.826,
|
|
"mean_token_accuracy": 0.0548530388623476,
|
|
"num_tokens": 187473.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.286836433410645,
|
|
"epoch": 0.005811081732864573,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.6569,
|
|
"mean_token_accuracy": 0.05997130684554577,
|
|
"num_tokens": 197515.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.200335216522216,
|
|
"epoch": 0.006101635819507801,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.5902,
|
|
"mean_token_accuracy": 0.058104698359966275,
|
|
"num_tokens": 208667.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.139002704620362,
|
|
"epoch": 0.00639218990615103,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4807,
|
|
"mean_token_accuracy": 0.06041117906570435,
|
|
"num_tokens": 218735.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.01154441833496,
|
|
"epoch": 0.006682743992794259,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3825,
|
|
"mean_token_accuracy": 0.06299497745931149,
|
|
"num_tokens": 228662.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.89796781539917,
|
|
"epoch": 0.006973298079437487,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2306,
|
|
"mean_token_accuracy": 0.0671528723090887,
|
|
"num_tokens": 238398.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.823051834106446,
|
|
"epoch": 0.007263852166080716,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.1653,
|
|
"mean_token_accuracy": 0.06395192183554173,
|
|
"num_tokens": 248686.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.683453178405761,
|
|
"epoch": 0.007554406252723945,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.047,
|
|
"mean_token_accuracy": 0.06545184403657914,
|
|
"num_tokens": 258243.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.465025901794434,
|
|
"epoch": 0.007844960339367173,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8093,
|
|
"mean_token_accuracy": 0.07491479925811291,
|
|
"num_tokens": 266700.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.230644702911377,
|
|
"epoch": 0.008135514426010402,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.7068,
|
|
"mean_token_accuracy": 0.0849619172513485,
|
|
"num_tokens": 275769.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.029471111297607,
|
|
"epoch": 0.00842606851265363,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7501,
|
|
"mean_token_accuracy": 0.06648054942488671,
|
|
"num_tokens": 286179.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.853238105773926,
|
|
"epoch": 0.008716622599296859,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6829,
|
|
"mean_token_accuracy": 0.06797396205365658,
|
|
"num_tokens": 295721.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.664832592010498,
|
|
"epoch": 0.009007176685940088,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5839,
|
|
"mean_token_accuracy": 0.06939561814069747,
|
|
"num_tokens": 304857.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.603672409057618,
|
|
"epoch": 0.009297730772583316,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.5881,
|
|
"mean_token_accuracy": 0.0636594358831644,
|
|
"num_tokens": 314711.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.415755271911621,
|
|
"epoch": 0.009588284859226545,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.4831,
|
|
"mean_token_accuracy": 0.07131265327334405,
|
|
"num_tokens": 324019.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.281650257110595,
|
|
"epoch": 0.009878838945869774,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.4462,
|
|
"mean_token_accuracy": 0.06885843947529793,
|
|
"num_tokens": 334619.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.176009941101075,
|
|
"epoch": 0.010169393032513002,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.516,
|
|
"mean_token_accuracy": 0.06891523078083991,
|
|
"num_tokens": 343788.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.143517398834229,
|
|
"epoch": 0.01045994711915623,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4422,
|
|
"mean_token_accuracy": 0.06905875466763974,
|
|
"num_tokens": 352974.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.098498773574828,
|
|
"epoch": 0.01075050120579946,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3605,
|
|
"mean_token_accuracy": 0.07677599526941777,
|
|
"num_tokens": 362342.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.9134715557098385,
|
|
"epoch": 0.011041055292442687,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3028,
|
|
"mean_token_accuracy": 0.07759733125567436,
|
|
"num_tokens": 372939.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.9448919773101805,
|
|
"epoch": 0.011331609379085917,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3574,
|
|
"mean_token_accuracy": 0.07698826119303703,
|
|
"num_tokens": 381957.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.866607666015625,
|
|
"epoch": 0.011622163465729146,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.359,
|
|
"mean_token_accuracy": 0.07136249989271164,
|
|
"num_tokens": 392592.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.92036657333374,
|
|
"epoch": 0.011912717552372373,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3296,
|
|
"mean_token_accuracy": 0.07443161755800247,
|
|
"num_tokens": 402921.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.811191749572754,
|
|
"epoch": 0.012203271639015602,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2769,
|
|
"mean_token_accuracy": 0.08031044751405716,
|
|
"num_tokens": 412036.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.7815025806427,
|
|
"epoch": 0.012493825725658832,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2312,
|
|
"mean_token_accuracy": 0.07401583753526211,
|
|
"num_tokens": 422098.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.856000328063965,
|
|
"epoch": 0.01278437981230206,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3411,
|
|
"mean_token_accuracy": 0.07584784552454948,
|
|
"num_tokens": 431146.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.768830442428589,
|
|
"epoch": 0.013074933898945288,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3719,
|
|
"mean_token_accuracy": 0.07549082823097705,
|
|
"num_tokens": 440737.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.7340232849121096,
|
|
"epoch": 0.013365487985588518,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2487,
|
|
"mean_token_accuracy": 0.07361838817596436,
|
|
"num_tokens": 451703.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.819036912918091,
|
|
"epoch": 0.013656042072231747,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2772,
|
|
"mean_token_accuracy": 0.07234742939472198,
|
|
"num_tokens": 461814.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.680500268936157,
|
|
"epoch": 0.013946596158874974,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1571,
|
|
"mean_token_accuracy": 0.08204417824745178,
|
|
"num_tokens": 472097.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.729534482955932,
|
|
"epoch": 0.014237150245518203,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1848,
|
|
"mean_token_accuracy": 0.08709119185805321,
|
|
"num_tokens": 481358.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.560984992980957,
|
|
"epoch": 0.014527704332161433,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2044,
|
|
"mean_token_accuracy": 0.07833344042301178,
|
|
"num_tokens": 491420.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.682111406326294,
|
|
"epoch": 0.01481825841880466,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1728,
|
|
"mean_token_accuracy": 0.08949121907353401,
|
|
"num_tokens": 499373.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.732259702682495,
|
|
"epoch": 0.01510881250544789,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.2722,
|
|
"mean_token_accuracy": 0.07445954978466034,
|
|
"num_tokens": 509300.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.59925479888916,
|
|
"epoch": 0.015399366592091119,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2628,
|
|
"mean_token_accuracy": 0.07740841507911682,
|
|
"num_tokens": 519383.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.5937131404876705,
|
|
"epoch": 0.015689920678734346,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2049,
|
|
"mean_token_accuracy": 0.07306748554110527,
|
|
"num_tokens": 528947.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.72738127708435,
|
|
"epoch": 0.015980474765377575,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1361,
|
|
"mean_token_accuracy": 0.08117511533200741,
|
|
"num_tokens": 538150.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.532864713668824,
|
|
"epoch": 0.016271028852020804,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1629,
|
|
"mean_token_accuracy": 0.07852732315659523,
|
|
"num_tokens": 547488.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.593377494812012,
|
|
"epoch": 0.016561582938664034,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.1742,
|
|
"mean_token_accuracy": 0.07643779553472996,
|
|
"num_tokens": 556242.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.6778340339660645,
|
|
"epoch": 0.01685213702530726,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1334,
|
|
"mean_token_accuracy": 0.07919927760958671,
|
|
"num_tokens": 566407.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.54665789604187,
|
|
"epoch": 0.01714269111195049,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.1739,
|
|
"mean_token_accuracy": 0.07580296993255616,
|
|
"num_tokens": 575820.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.5551825046539305,
|
|
"epoch": 0.017433245198593718,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2182,
|
|
"mean_token_accuracy": 0.0772180262953043,
|
|
"num_tokens": 585374.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.52297887802124,
|
|
"epoch": 0.017723799285236947,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0797,
|
|
"mean_token_accuracy": 0.07936501353979111,
|
|
"num_tokens": 595141.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.577914524078369,
|
|
"epoch": 0.018014353371880176,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0687,
|
|
"mean_token_accuracy": 0.0796559315174818,
|
|
"num_tokens": 604481.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.417883062362671,
|
|
"epoch": 0.018304907458523405,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.0375,
|
|
"mean_token_accuracy": 0.08207950145006179,
|
|
"num_tokens": 614063.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.440096426010132,
|
|
"epoch": 0.01859546154516663,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.0574,
|
|
"mean_token_accuracy": 0.07540913559496402,
|
|
"num_tokens": 624389.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.416240692138672,
|
|
"epoch": 0.01888601563180986,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.087,
|
|
"mean_token_accuracy": 0.08941936045885086,
|
|
"num_tokens": 634289.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.504856967926026,
|
|
"epoch": 0.01917656971845309,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.0857,
|
|
"mean_token_accuracy": 0.081592907756567,
|
|
"num_tokens": 643950.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.438478994369507,
|
|
"epoch": 0.01946712380509632,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.0506,
|
|
"mean_token_accuracy": 0.07806282639503478,
|
|
"num_tokens": 653210.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.493538475036621,
|
|
"epoch": 0.019757677891739548,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0527,
|
|
"mean_token_accuracy": 0.08196914717555046,
|
|
"num_tokens": 662862.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.4518084049224855,
|
|
"epoch": 0.020048231978382777,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.0057,
|
|
"mean_token_accuracy": 0.07910942509770394,
|
|
"num_tokens": 673323.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.414995431900024,
|
|
"epoch": 0.020338786065026003,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.055,
|
|
"mean_token_accuracy": 0.08215757757425309,
|
|
"num_tokens": 683532.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.348615980148315,
|
|
"epoch": 0.020629340151669232,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9814,
|
|
"mean_token_accuracy": 0.0809572272002697,
|
|
"num_tokens": 693162.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.25407395362854,
|
|
"epoch": 0.02091989423831246,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9904,
|
|
"mean_token_accuracy": 0.07966803461313247,
|
|
"num_tokens": 703783.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.457367944717407,
|
|
"epoch": 0.02121044832495569,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0101,
|
|
"mean_token_accuracy": 0.08167746141552926,
|
|
"num_tokens": 713083.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.2763917446136475,
|
|
"epoch": 0.02150100241159892,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9183,
|
|
"mean_token_accuracy": 0.08331615403294564,
|
|
"num_tokens": 722036.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.346493482589722,
|
|
"epoch": 0.02179155649824215,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0015,
|
|
"mean_token_accuracy": 0.08131363317370414,
|
|
"num_tokens": 732465.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.370974349975586,
|
|
"epoch": 0.022082110584885375,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.0694,
|
|
"mean_token_accuracy": 0.0784931555390358,
|
|
"num_tokens": 742592.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.243825960159302,
|
|
"epoch": 0.022372664671528604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9464,
|
|
"mean_token_accuracy": 0.08015396147966385,
|
|
"num_tokens": 752150.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.414292812347412,
|
|
"epoch": 0.022663218758171833,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0178,
|
|
"mean_token_accuracy": 0.0777292400598526,
|
|
"num_tokens": 761550.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.340526723861695,
|
|
"epoch": 0.022953772844815062,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9556,
|
|
"mean_token_accuracy": 0.07885948345065116,
|
|
"num_tokens": 771899.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.310690546035767,
|
|
"epoch": 0.02324432693145829,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1145,
|
|
"mean_token_accuracy": 0.07887880131602287,
|
|
"num_tokens": 782948.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.256437063217163,
|
|
"epoch": 0.02353488101810152,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.963,
|
|
"mean_token_accuracy": 0.08006646111607552,
|
|
"num_tokens": 792563.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.28327488899231,
|
|
"epoch": 0.023825435104744747,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8839,
|
|
"mean_token_accuracy": 0.0808573119342327,
|
|
"num_tokens": 801736.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.292191982269287,
|
|
"epoch": 0.024115989191387976,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9514,
|
|
"mean_token_accuracy": 0.07757750265300274,
|
|
"num_tokens": 811527.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.277588653564453,
|
|
"epoch": 0.024406543278031205,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9567,
|
|
"mean_token_accuracy": 0.08245478793978692,
|
|
"num_tokens": 822003.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.219156408309937,
|
|
"epoch": 0.024697097364674434,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8475,
|
|
"mean_token_accuracy": 0.08649195730686188,
|
|
"num_tokens": 831773.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.16748538017273,
|
|
"epoch": 0.024987651451317663,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.853,
|
|
"mean_token_accuracy": 0.08320235460996628,
|
|
"num_tokens": 841233.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.227472400665283,
|
|
"epoch": 0.025278205537960893,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.9353,
|
|
"mean_token_accuracy": 0.0826109692454338,
|
|
"num_tokens": 851851.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.172217082977295,
|
|
"epoch": 0.02556875962460412,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8521,
|
|
"mean_token_accuracy": 0.08434986025094986,
|
|
"num_tokens": 861898.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.161604738235473,
|
|
"epoch": 0.025859313711247348,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9483,
|
|
"mean_token_accuracy": 0.08010145872831345,
|
|
"num_tokens": 871037.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.2003782272338865,
|
|
"epoch": 0.026149867797890577,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8892,
|
|
"mean_token_accuracy": 0.08248139917850494,
|
|
"num_tokens": 882805.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.0999010562896725,
|
|
"epoch": 0.026440421884533806,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.8614,
|
|
"mean_token_accuracy": 0.0847211316227913,
|
|
"num_tokens": 893020.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.154237747192383,
|
|
"epoch": 0.026730975971177035,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8617,
|
|
"mean_token_accuracy": 0.08975208401679993,
|
|
"num_tokens": 902409.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.152691459655761,
|
|
"epoch": 0.027021530057820264,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8884,
|
|
"mean_token_accuracy": 0.08094722181558608,
|
|
"num_tokens": 912665.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.048449563980102,
|
|
"epoch": 0.027312084144463494,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7633,
|
|
"mean_token_accuracy": 0.0842330276966095,
|
|
"num_tokens": 921520.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.150616073608399,
|
|
"epoch": 0.02760263823110672,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7382,
|
|
"mean_token_accuracy": 0.08964505940675735,
|
|
"num_tokens": 931513.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.082989072799682,
|
|
"epoch": 0.02789319231774995,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8599,
|
|
"mean_token_accuracy": 0.08315788432955742,
|
|
"num_tokens": 940748.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.063059997558594,
|
|
"epoch": 0.028183746404393178,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.81,
|
|
"mean_token_accuracy": 0.08362271934747696,
|
|
"num_tokens": 949810.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.115930604934692,
|
|
"epoch": 0.028474300491036407,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7554,
|
|
"mean_token_accuracy": 0.09185877740383148,
|
|
"num_tokens": 959122.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.133077001571655,
|
|
"epoch": 0.028764854577679636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.887,
|
|
"mean_token_accuracy": 0.08635125607252121,
|
|
"num_tokens": 968974.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.069976043701172,
|
|
"epoch": 0.029055408664322865,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8548,
|
|
"mean_token_accuracy": 0.08358358740806579,
|
|
"num_tokens": 978444.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.1011241436004635,
|
|
"epoch": 0.02934596275096609,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7724,
|
|
"mean_token_accuracy": 0.09029116481542587,
|
|
"num_tokens": 988025.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.02938323020935,
|
|
"epoch": 0.02963651683760932,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8098,
|
|
"mean_token_accuracy": 0.0861921526491642,
|
|
"num_tokens": 997783.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.101745176315307,
|
|
"epoch": 0.02992707092425255,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7998,
|
|
"mean_token_accuracy": 0.08733523786067962,
|
|
"num_tokens": 1006208.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.995005512237549,
|
|
"epoch": 0.03021762501089578,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7581,
|
|
"mean_token_accuracy": 0.08793584778904914,
|
|
"num_tokens": 1016861.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.110492086410522,
|
|
"epoch": 0.030508179097539008,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7479,
|
|
"mean_token_accuracy": 0.08422723487019539,
|
|
"num_tokens": 1026276.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.966236734390259,
|
|
"epoch": 0.030798733184182237,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8109,
|
|
"mean_token_accuracy": 0.09106520414352418,
|
|
"num_tokens": 1035571.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.032464265823364,
|
|
"epoch": 0.031089287270825463,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8321,
|
|
"mean_token_accuracy": 0.08516258671879769,
|
|
"num_tokens": 1044794.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.120251321792603,
|
|
"epoch": 0.03137984135746869,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8057,
|
|
"mean_token_accuracy": 0.08295166790485382,
|
|
"num_tokens": 1054919.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.968770742416382,
|
|
"epoch": 0.031670395444111925,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.695,
|
|
"mean_token_accuracy": 0.09049282670021057,
|
|
"num_tokens": 1064220.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.963365888595581,
|
|
"epoch": 0.03196094953075515,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7935,
|
|
"mean_token_accuracy": 0.08347960636019706,
|
|
"num_tokens": 1075044.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.027530097961426,
|
|
"epoch": 0.032251503617398376,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7377,
|
|
"mean_token_accuracy": 0.09026792943477631,
|
|
"num_tokens": 1083608.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.977536869049072,
|
|
"epoch": 0.03254205770404161,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.7649,
|
|
"mean_token_accuracy": 0.09168017953634262,
|
|
"num_tokens": 1094270.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.97630295753479,
|
|
"epoch": 0.032832611790684835,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7807,
|
|
"mean_token_accuracy": 0.08632769882678985,
|
|
"num_tokens": 1103763.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.034525442123413,
|
|
"epoch": 0.03312316587732807,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7585,
|
|
"mean_token_accuracy": 0.09223658666014671,
|
|
"num_tokens": 1113117.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.9836554527282715,
|
|
"epoch": 0.03341371996397129,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8207,
|
|
"mean_token_accuracy": 0.0857993058860302,
|
|
"num_tokens": 1123254.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.890299654006958,
|
|
"epoch": 0.03370427405061452,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7576,
|
|
"mean_token_accuracy": 0.09352053552865983,
|
|
"num_tokens": 1132608.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.976414442062378,
|
|
"epoch": 0.03399482813725775,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6998,
|
|
"mean_token_accuracy": 0.09929749593138695,
|
|
"num_tokens": 1142525.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.91153621673584,
|
|
"epoch": 0.03428538222390098,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7312,
|
|
"mean_token_accuracy": 0.09118211716413498,
|
|
"num_tokens": 1151294.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.890883445739746,
|
|
"epoch": 0.03457593631054421,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6838,
|
|
"mean_token_accuracy": 0.09118377938866615,
|
|
"num_tokens": 1160388.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.041623067855835,
|
|
"epoch": 0.034866490397187436,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8162,
|
|
"mean_token_accuracy": 0.08897348642349243,
|
|
"num_tokens": 1169616.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.859910583496093,
|
|
"epoch": 0.03515704448383067,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6653,
|
|
"mean_token_accuracy": 0.09158354997634888,
|
|
"num_tokens": 1179715.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.859572076797486,
|
|
"epoch": 0.035447598570473894,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6291,
|
|
"mean_token_accuracy": 0.09261076152324677,
|
|
"num_tokens": 1188206.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.957557249069214,
|
|
"epoch": 0.03573815265711712,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7527,
|
|
"mean_token_accuracy": 0.08373265452682972,
|
|
"num_tokens": 1197372.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.925255870819091,
|
|
"epoch": 0.03602870674376035,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7477,
|
|
"mean_token_accuracy": 0.08401187434792519,
|
|
"num_tokens": 1207698.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.975111675262451,
|
|
"epoch": 0.03631926083040358,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7088,
|
|
"mean_token_accuracy": 0.08878697119653226,
|
|
"num_tokens": 1217256.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.7793882369995115,
|
|
"epoch": 0.03660981491704681,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.5962,
|
|
"mean_token_accuracy": 0.08970977216959,
|
|
"num_tokens": 1226574.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.868472766876221,
|
|
"epoch": 0.03690036900369004,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.7118,
|
|
"mean_token_accuracy": 0.09969199374318123,
|
|
"num_tokens": 1234999.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.824588871002197,
|
|
"epoch": 0.03719092309033326,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6674,
|
|
"mean_token_accuracy": 0.09083129242062568,
|
|
"num_tokens": 1244390.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.80760760307312,
|
|
"epoch": 0.037481477176976495,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6152,
|
|
"mean_token_accuracy": 0.09922600984573364,
|
|
"num_tokens": 1253649.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.868955373764038,
|
|
"epoch": 0.03777203126361972,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6038,
|
|
"mean_token_accuracy": 0.09038031622767448,
|
|
"num_tokens": 1262715.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.799616193771362,
|
|
"epoch": 0.038062585350262954,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.6623,
|
|
"mean_token_accuracy": 0.09431513771414757,
|
|
"num_tokens": 1271384.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.898798608779908,
|
|
"epoch": 0.03835313943690618,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7402,
|
|
"mean_token_accuracy": 0.09072391986846924,
|
|
"num_tokens": 1281217.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.83271975517273,
|
|
"epoch": 0.03864369352354941,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7192,
|
|
"mean_token_accuracy": 0.08845826983451843,
|
|
"num_tokens": 1291419.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.803164100646972,
|
|
"epoch": 0.03893424761019264,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7509,
|
|
"mean_token_accuracy": 0.09317919313907623,
|
|
"num_tokens": 1300750.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.924269723892212,
|
|
"epoch": 0.03922480169683586,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.674,
|
|
"mean_token_accuracy": 0.09520297944545746,
|
|
"num_tokens": 1309975.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.822680854797364,
|
|
"epoch": 0.039515355783479096,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.6411,
|
|
"mean_token_accuracy": 0.09562266170978546,
|
|
"num_tokens": 1320984.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.780631351470947,
|
|
"epoch": 0.03980590987012232,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6535,
|
|
"mean_token_accuracy": 0.09362242966890336,
|
|
"num_tokens": 1329901.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.880793523788452,
|
|
"epoch": 0.040096463956765555,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7526,
|
|
"mean_token_accuracy": 0.09442069381475449,
|
|
"num_tokens": 1340185.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.807402038574219,
|
|
"epoch": 0.04038701804340878,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6918,
|
|
"mean_token_accuracy": 0.09226417914032936,
|
|
"num_tokens": 1349655.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.853033638000488,
|
|
"epoch": 0.040677572130052006,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6612,
|
|
"mean_token_accuracy": 0.08990521878004074,
|
|
"num_tokens": 1359193.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.800865554809571,
|
|
"epoch": 0.04096812621669524,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5969,
|
|
"mean_token_accuracy": 0.09179130122065544,
|
|
"num_tokens": 1368280.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.742464542388916,
|
|
"epoch": 0.041258680303338464,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6564,
|
|
"mean_token_accuracy": 0.08979732692241668,
|
|
"num_tokens": 1378006.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.877160882949829,
|
|
"epoch": 0.0415492343899817,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6672,
|
|
"mean_token_accuracy": 0.09487208649516106,
|
|
"num_tokens": 1387908.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.7572847366333,
|
|
"epoch": 0.04183978847662492,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6392,
|
|
"mean_token_accuracy": 0.09631308168172836,
|
|
"num_tokens": 1396865.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.801170921325683,
|
|
"epoch": 0.042130342563268156,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6724,
|
|
"mean_token_accuracy": 0.09610226228833199,
|
|
"num_tokens": 1405636.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.8957888126373295,
|
|
"epoch": 0.04242089664991138,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6814,
|
|
"mean_token_accuracy": 0.09137585759162903,
|
|
"num_tokens": 1414373.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.785763597488403,
|
|
"epoch": 0.04271145073655461,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6742,
|
|
"mean_token_accuracy": 0.09274838268756866,
|
|
"num_tokens": 1425925.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.702451086044311,
|
|
"epoch": 0.04300200482319784,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5992,
|
|
"mean_token_accuracy": 0.092991454154253,
|
|
"num_tokens": 1436539.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.715664482116699,
|
|
"epoch": 0.043292558909841065,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5928,
|
|
"mean_token_accuracy": 0.09597984626889229,
|
|
"num_tokens": 1445913.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.814825916290284,
|
|
"epoch": 0.0435831129964843,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6661,
|
|
"mean_token_accuracy": 0.08998511806130409,
|
|
"num_tokens": 1455387.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.769626665115356,
|
|
"epoch": 0.043873667083127524,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6905,
|
|
"mean_token_accuracy": 0.09334170669317246,
|
|
"num_tokens": 1465050.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.801269054412842,
|
|
"epoch": 0.04416422116977075,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6236,
|
|
"mean_token_accuracy": 0.0948921263217926,
|
|
"num_tokens": 1474448.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.676837682723999,
|
|
"epoch": 0.04445477525641398,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5515,
|
|
"mean_token_accuracy": 0.0982357420027256,
|
|
"num_tokens": 1485048.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.727758884429932,
|
|
"epoch": 0.04474532934305721,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5684,
|
|
"mean_token_accuracy": 0.09265084117650986,
|
|
"num_tokens": 1495441.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.6234723091125485,
|
|
"epoch": 0.04503588342970044,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4869,
|
|
"mean_token_accuracy": 0.10020759403705597,
|
|
"num_tokens": 1504273.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.7275268077850345,
|
|
"epoch": 0.045326437516343666,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6772,
|
|
"mean_token_accuracy": 0.0932683877646923,
|
|
"num_tokens": 1513631.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.699482059478759,
|
|
"epoch": 0.0456169916029869,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6101,
|
|
"mean_token_accuracy": 0.0970641016960144,
|
|
"num_tokens": 1524005.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.69863338470459,
|
|
"epoch": 0.045907545689630125,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5287,
|
|
"mean_token_accuracy": 0.09622322842478752,
|
|
"num_tokens": 1533970.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.781415510177612,
|
|
"epoch": 0.04619809977627335,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.614,
|
|
"mean_token_accuracy": 0.0962904892861843,
|
|
"num_tokens": 1543152.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.566462612152099,
|
|
"epoch": 0.04648865386291658,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.4738,
|
|
"mean_token_accuracy": 0.0966126188635826,
|
|
"num_tokens": 1552181.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.660043859481812,
|
|
"epoch": 0.04677920794955981,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.488,
|
|
"mean_token_accuracy": 0.10304239392280579,
|
|
"num_tokens": 1561384.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.574128723144531,
|
|
"epoch": 0.04706976203620304,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.10381592363119126,
|
|
"num_tokens": 1570517.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.631992340087891,
|
|
"epoch": 0.04736031612284627,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5389,
|
|
"mean_token_accuracy": 0.09975797310471535,
|
|
"num_tokens": 1579404.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.548789834976196,
|
|
"epoch": 0.04765087020948949,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.4271,
|
|
"mean_token_accuracy": 0.09895784109830856,
|
|
"num_tokens": 1589351.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.643519687652588,
|
|
"epoch": 0.047941424296132726,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.4288,
|
|
"mean_token_accuracy": 0.09936647340655327,
|
|
"num_tokens": 1598083.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.553793811798096,
|
|
"epoch": 0.04823197838277595,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4337,
|
|
"mean_token_accuracy": 0.10424233078956605,
|
|
"num_tokens": 1607063.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.551960134506226,
|
|
"epoch": 0.048522532469419184,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4999,
|
|
"mean_token_accuracy": 0.1029004231095314,
|
|
"num_tokens": 1616804.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.624791765213013,
|
|
"epoch": 0.04881308655606241,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5016,
|
|
"mean_token_accuracy": 0.10294707268476486,
|
|
"num_tokens": 1626005.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.597585964202881,
|
|
"epoch": 0.04910364064270564,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4951,
|
|
"mean_token_accuracy": 0.10337291657924652,
|
|
"num_tokens": 1635145.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.543288469314575,
|
|
"epoch": 0.04939419472934887,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.4558,
|
|
"mean_token_accuracy": 0.10390876084566117,
|
|
"num_tokens": 1644381.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.646559333801269,
|
|
"epoch": 0.049684748815992094,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.606,
|
|
"mean_token_accuracy": 0.09714684486389161,
|
|
"num_tokens": 1653965.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.615056085586548,
|
|
"epoch": 0.04997530290263533,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.4749,
|
|
"mean_token_accuracy": 0.10362254977226257,
|
|
"num_tokens": 1663120.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.491550159454346,
|
|
"epoch": 0.05026585698927855,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4611,
|
|
"mean_token_accuracy": 0.10190960243344308,
|
|
"num_tokens": 1672533.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.599954557418823,
|
|
"epoch": 0.050556411075921785,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.5367,
|
|
"mean_token_accuracy": 0.09783021062612533,
|
|
"num_tokens": 1682123.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.579014635086059,
|
|
"epoch": 0.05084696516256501,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4836,
|
|
"mean_token_accuracy": 0.09750580117106437,
|
|
"num_tokens": 1690996.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.623455667495728,
|
|
"epoch": 0.05113751924920824,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.6269,
|
|
"mean_token_accuracy": 0.09714528471231461,
|
|
"num_tokens": 1701264.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.633913516998291,
|
|
"epoch": 0.05142807333585147,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.53,
|
|
"mean_token_accuracy": 0.10271772965788842,
|
|
"num_tokens": 1711226.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.582356214523315,
|
|
"epoch": 0.051718627422494695,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4446,
|
|
"mean_token_accuracy": 0.10424281507730485,
|
|
"num_tokens": 1721241.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.601112365722656,
|
|
"epoch": 0.05200918150913793,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5286,
|
|
"mean_token_accuracy": 0.10329965353012086,
|
|
"num_tokens": 1730916.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.650094270706177,
|
|
"epoch": 0.052299735595781154,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5418,
|
|
"mean_token_accuracy": 0.10129279345273971,
|
|
"num_tokens": 1741590.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.553584146499634,
|
|
"epoch": 0.052590289682424386,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.532,
|
|
"mean_token_accuracy": 0.09813695922493934,
|
|
"num_tokens": 1751419.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.428548288345337,
|
|
"epoch": 0.05288084376906761,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3947,
|
|
"mean_token_accuracy": 0.10944176390767098,
|
|
"num_tokens": 1761947.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.63139066696167,
|
|
"epoch": 0.05317139785571084,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4533,
|
|
"mean_token_accuracy": 0.1052020475268364,
|
|
"num_tokens": 1773204.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.442141962051392,
|
|
"epoch": 0.05346195194235407,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4139,
|
|
"mean_token_accuracy": 0.10542120933532714,
|
|
"num_tokens": 1783110.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.468068027496338,
|
|
"epoch": 0.053752506028997296,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.3869,
|
|
"mean_token_accuracy": 0.10259869024157524,
|
|
"num_tokens": 1791332.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.4169505596160885,
|
|
"epoch": 0.05404306011564053,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4024,
|
|
"mean_token_accuracy": 0.10698127672076226,
|
|
"num_tokens": 1801302.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.594544839859009,
|
|
"epoch": 0.054333614202283755,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.10172575935721398,
|
|
"num_tokens": 1810949.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.589707708358764,
|
|
"epoch": 0.05462416828892699,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.482,
|
|
"mean_token_accuracy": 0.10701743364334107,
|
|
"num_tokens": 1820042.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.3840772151947025,
|
|
"epoch": 0.05491472237557021,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4086,
|
|
"mean_token_accuracy": 0.11226747334003448,
|
|
"num_tokens": 1829079.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.685991048812866,
|
|
"epoch": 0.05520527646221344,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.6022,
|
|
"mean_token_accuracy": 0.09806751757860184,
|
|
"num_tokens": 1839635.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.483322238922119,
|
|
"epoch": 0.05549583054885667,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.458,
|
|
"mean_token_accuracy": 0.10805773884057998,
|
|
"num_tokens": 1849651.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.516399383544922,
|
|
"epoch": 0.0557863846354999,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4495,
|
|
"mean_token_accuracy": 0.098360076546669,
|
|
"num_tokens": 1859132.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.511109638214111,
|
|
"epoch": 0.05607693872214313,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4723,
|
|
"mean_token_accuracy": 0.10554241985082627,
|
|
"num_tokens": 1868063.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.4864051818847654,
|
|
"epoch": 0.056367492808786356,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4308,
|
|
"mean_token_accuracy": 0.10283332020044327,
|
|
"num_tokens": 1877204.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.515411281585694,
|
|
"epoch": 0.05665804689542958,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.4761,
|
|
"mean_token_accuracy": 0.10713874846696854,
|
|
"num_tokens": 1886987.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.520045566558838,
|
|
"epoch": 0.056948600982072814,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5135,
|
|
"mean_token_accuracy": 0.1081652283668518,
|
|
"num_tokens": 1896791.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.366655588150024,
|
|
"epoch": 0.05723915506871604,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3233,
|
|
"mean_token_accuracy": 0.11065977737307549,
|
|
"num_tokens": 1905878.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.603346538543701,
|
|
"epoch": 0.05752970915535927,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.5566,
|
|
"mean_token_accuracy": 0.10223873779177665,
|
|
"num_tokens": 1914742.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.4896119117736815,
|
|
"epoch": 0.0578202632420025,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3861,
|
|
"mean_token_accuracy": 0.10645260885357857,
|
|
"num_tokens": 1924175.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.419984769821167,
|
|
"epoch": 0.05811081732864573,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3878,
|
|
"mean_token_accuracy": 0.11293421387672424,
|
|
"num_tokens": 1933724.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.515293407440185,
|
|
"epoch": 0.05840137141528896,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.5608,
|
|
"mean_token_accuracy": 0.10285485908389091,
|
|
"num_tokens": 1943987.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.596132755279541,
|
|
"epoch": 0.05869192550193218,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.5914,
|
|
"mean_token_accuracy": 0.10778910219669342,
|
|
"num_tokens": 1953754.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.429438877105713,
|
|
"epoch": 0.058982479588575415,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.4654,
|
|
"mean_token_accuracy": 0.106864894926548,
|
|
"num_tokens": 1963630.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.527030992507934,
|
|
"epoch": 0.05927303367521864,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.4443,
|
|
"mean_token_accuracy": 0.11184166073799133,
|
|
"num_tokens": 1973325.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.455427885055542,
|
|
"epoch": 0.05956358776186187,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.364,
|
|
"mean_token_accuracy": 0.11303301379084588,
|
|
"num_tokens": 1983269.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.333305406570434,
|
|
"epoch": 0.0598541418485051,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.3086,
|
|
"mean_token_accuracy": 0.10803757533431053,
|
|
"num_tokens": 1992269.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.459513521194458,
|
|
"epoch": 0.060144695935148325,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.3731,
|
|
"mean_token_accuracy": 0.10919068828225136,
|
|
"num_tokens": 2001520.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.40694751739502,
|
|
"epoch": 0.06043525002179156,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.4015,
|
|
"mean_token_accuracy": 0.10682632625102997,
|
|
"num_tokens": 2009787.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.541155672073364,
|
|
"epoch": 0.06072580410843478,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.4651,
|
|
"mean_token_accuracy": 0.1081154353916645,
|
|
"num_tokens": 2018830.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.345413541793823,
|
|
"epoch": 0.061016358195078016,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.3822,
|
|
"mean_token_accuracy": 0.10369199216365814,
|
|
"num_tokens": 2029222.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.523482990264893,
|
|
"epoch": 0.06130691228172124,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.4571,
|
|
"mean_token_accuracy": 0.0996652863919735,
|
|
"num_tokens": 2037777.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.409976768493652,
|
|
"epoch": 0.061597466368364474,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.4149,
|
|
"mean_token_accuracy": 0.10962732359766961,
|
|
"num_tokens": 2048137.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.4193802833557125,
|
|
"epoch": 0.0618880204550077,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.10600300803780556,
|
|
"num_tokens": 2058379.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.587002468109131,
|
|
"epoch": 0.062178574541650926,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.5367,
|
|
"mean_token_accuracy": 0.10047689005732537,
|
|
"num_tokens": 2068844.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.351005172729492,
|
|
"epoch": 0.06246912862829416,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.3839,
|
|
"mean_token_accuracy": 0.11975573673844338,
|
|
"num_tokens": 2078859.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.442875385284424,
|
|
"epoch": 0.06275968271493738,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.10901174843311309,
|
|
"num_tokens": 2087621.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.451695346832276,
|
|
"epoch": 0.06305023680158062,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.10689355432987213,
|
|
"num_tokens": 2096878.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.527059412002563,
|
|
"epoch": 0.06334079088822385,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.10578126087784767,
|
|
"num_tokens": 2106018.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.425774145126343,
|
|
"epoch": 0.06363134497486707,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.3355,
|
|
"mean_token_accuracy": 0.11034304574131966,
|
|
"num_tokens": 2114980.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.324463939666748,
|
|
"epoch": 0.0639218990615103,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.3556,
|
|
"mean_token_accuracy": 0.10826032757759094,
|
|
"num_tokens": 2123655.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.5232044696807865,
|
|
"epoch": 0.06421245314815353,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.10399825125932693,
|
|
"num_tokens": 2133543.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.313161087036133,
|
|
"epoch": 0.06450300723479675,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.3388,
|
|
"mean_token_accuracy": 0.11555257141590118,
|
|
"num_tokens": 2142187.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.3854879379272464,
|
|
"epoch": 0.06479356132143999,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.3585,
|
|
"mean_token_accuracy": 0.10878968238830566,
|
|
"num_tokens": 2151894.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.38906192779541,
|
|
"epoch": 0.06508411540808322,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.2847,
|
|
"mean_token_accuracy": 0.11355354338884353,
|
|
"num_tokens": 2161930.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.289382266998291,
|
|
"epoch": 0.06537466949472644,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.3569,
|
|
"mean_token_accuracy": 0.1129719614982605,
|
|
"num_tokens": 2171023.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.4400489807128904,
|
|
"epoch": 0.06566522358136967,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.4426,
|
|
"mean_token_accuracy": 0.11219586357474327,
|
|
"num_tokens": 2180657.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.376230049133301,
|
|
"epoch": 0.0659557776680129,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.3582,
|
|
"mean_token_accuracy": 0.10606605932116508,
|
|
"num_tokens": 2191341.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.304773330688477,
|
|
"epoch": 0.06624633175465613,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.3338,
|
|
"mean_token_accuracy": 0.10943459793925285,
|
|
"num_tokens": 2200206.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.41196608543396,
|
|
"epoch": 0.06653688584129935,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.3872,
|
|
"mean_token_accuracy": 0.11151469200849533,
|
|
"num_tokens": 2209840.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.328424167633057,
|
|
"epoch": 0.06682743992794259,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.1169155366718769,
|
|
"num_tokens": 2219304.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.443253898620606,
|
|
"epoch": 0.06711799401458582,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.3727,
|
|
"mean_token_accuracy": 0.11276882886886597,
|
|
"num_tokens": 2229184.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.262433385848999,
|
|
"epoch": 0.06740854810122904,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.2706,
|
|
"mean_token_accuracy": 0.11013494282960892,
|
|
"num_tokens": 2238444.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.4703967571258545,
|
|
"epoch": 0.06769910218787227,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.4657,
|
|
"mean_token_accuracy": 0.11107994392514228,
|
|
"num_tokens": 2247346.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.371312046051026,
|
|
"epoch": 0.0679896562745155,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.329,
|
|
"mean_token_accuracy": 0.11129426136612892,
|
|
"num_tokens": 2256532.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.257063865661621,
|
|
"epoch": 0.06828021036115874,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.11564552634954453,
|
|
"num_tokens": 2266254.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.30529637336731,
|
|
"epoch": 0.06857076444780195,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.2288,
|
|
"mean_token_accuracy": 0.11662948206067085,
|
|
"num_tokens": 2275285.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.3822582244873045,
|
|
"epoch": 0.06886131853444519,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.3453,
|
|
"mean_token_accuracy": 0.11292837411165238,
|
|
"num_tokens": 2285448.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.24997091293335,
|
|
"epoch": 0.06915187262108842,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.2455,
|
|
"mean_token_accuracy": 0.1130777969956398,
|
|
"num_tokens": 2295896.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.280991077423096,
|
|
"epoch": 0.06944242670773164,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.3089,
|
|
"mean_token_accuracy": 0.10868189558386802,
|
|
"num_tokens": 2306873.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.47158374786377,
|
|
"epoch": 0.06973298079437487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.4186,
|
|
"mean_token_accuracy": 0.10683797150850297,
|
|
"num_tokens": 2317843.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.257269239425659,
|
|
"epoch": 0.0700235348810181,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.2455,
|
|
"mean_token_accuracy": 0.11401106640696526,
|
|
"num_tokens": 2327047.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.321083211898804,
|
|
"epoch": 0.07031408896766134,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.3372,
|
|
"mean_token_accuracy": 0.11759073585271836,
|
|
"num_tokens": 2336893.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.334356689453125,
|
|
"epoch": 0.07060464305430456,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.2793,
|
|
"mean_token_accuracy": 0.11712626814842224,
|
|
"num_tokens": 2346788.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.306163549423218,
|
|
"epoch": 0.07089519714094779,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.285,
|
|
"mean_token_accuracy": 0.11232926696538925,
|
|
"num_tokens": 2356261.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.299500322341919,
|
|
"epoch": 0.07118575122759102,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.2777,
|
|
"mean_token_accuracy": 0.11824780255556107,
|
|
"num_tokens": 2366038.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.250729322433472,
|
|
"epoch": 0.07147630531423424,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.2505,
|
|
"mean_token_accuracy": 0.11852356940507888,
|
|
"num_tokens": 2375839.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.387542676925659,
|
|
"epoch": 0.07176685940087747,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.2843,
|
|
"mean_token_accuracy": 0.11505208611488342,
|
|
"num_tokens": 2384862.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.261922121047974,
|
|
"epoch": 0.0720574134875207,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.3258,
|
|
"mean_token_accuracy": 0.10773891285061836,
|
|
"num_tokens": 2395086.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.389956378936768,
|
|
"epoch": 0.07234796757416392,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.3688,
|
|
"mean_token_accuracy": 0.11114799231290817,
|
|
"num_tokens": 2405601.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.205085325241089,
|
|
"epoch": 0.07263852166080716,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.2512,
|
|
"mean_token_accuracy": 0.1158894307911396,
|
|
"num_tokens": 2416195.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.327854824066162,
|
|
"epoch": 0.07292907574745039,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.1885,
|
|
"mean_token_accuracy": 0.11666455045342446,
|
|
"num_tokens": 2424810.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.270368146896362,
|
|
"epoch": 0.07321962983409362,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.3428,
|
|
"mean_token_accuracy": 0.11471045464277267,
|
|
"num_tokens": 2434367.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.3180938243865965,
|
|
"epoch": 0.07351018392073684,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.2969,
|
|
"mean_token_accuracy": 0.11603154689073562,
|
|
"num_tokens": 2443610.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.276625680923462,
|
|
"epoch": 0.07380073800738007,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.2087,
|
|
"mean_token_accuracy": 0.11644172966480255,
|
|
"num_tokens": 2453481.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.126973104476929,
|
|
"epoch": 0.0740912920940233,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.2526,
|
|
"mean_token_accuracy": 0.12029187902808189,
|
|
"num_tokens": 2463104.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.366124963760376,
|
|
"epoch": 0.07438184618066652,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.2256,
|
|
"mean_token_accuracy": 0.11856506243348122,
|
|
"num_tokens": 2472860.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.325764846801758,
|
|
"epoch": 0.07467240026730976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.11878090873360633,
|
|
"num_tokens": 2482643.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.244407844543457,
|
|
"epoch": 0.07496295435395299,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.2963,
|
|
"mean_token_accuracy": 0.11809295788407326,
|
|
"num_tokens": 2491895.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.29679856300354,
|
|
"epoch": 0.07525350844059622,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.2099,
|
|
"mean_token_accuracy": 0.11272355318069457,
|
|
"num_tokens": 2501080.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.253429985046386,
|
|
"epoch": 0.07554406252723944,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.2024,
|
|
"mean_token_accuracy": 0.11589482352137566,
|
|
"num_tokens": 2510341.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.217455101013184,
|
|
"epoch": 0.07583461661388267,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.3344,
|
|
"mean_token_accuracy": 0.117685367166996,
|
|
"num_tokens": 2520655.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.29218373298645,
|
|
"epoch": 0.07612517070052591,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.259,
|
|
"mean_token_accuracy": 0.11394123807549476,
|
|
"num_tokens": 2530022.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.272565269470215,
|
|
"epoch": 0.07641572478716913,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.2531,
|
|
"mean_token_accuracy": 0.11075997799634933,
|
|
"num_tokens": 2541451.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.214617490768433,
|
|
"epoch": 0.07670627887381236,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.1515,
|
|
"mean_token_accuracy": 0.11713779196143151,
|
|
"num_tokens": 2550299.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.279712581634522,
|
|
"epoch": 0.07699683296045559,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.2833,
|
|
"mean_token_accuracy": 0.11670693308115006,
|
|
"num_tokens": 2560422.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.251215887069702,
|
|
"epoch": 0.07728738704709882,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.1963,
|
|
"mean_token_accuracy": 0.12052965685725212,
|
|
"num_tokens": 2570327.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.290622997283935,
|
|
"epoch": 0.07757794113374204,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.303,
|
|
"mean_token_accuracy": 0.12087919563055038,
|
|
"num_tokens": 2580026.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.2676429271698,
|
|
"epoch": 0.07786849522038528,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.3344,
|
|
"mean_token_accuracy": 0.11165543794631957,
|
|
"num_tokens": 2591172.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.31669340133667,
|
|
"epoch": 0.07815904930702851,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.2188,
|
|
"mean_token_accuracy": 0.11619681417942047,
|
|
"num_tokens": 2601725.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.1492432117462155,
|
|
"epoch": 0.07844960339367173,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.2285,
|
|
"mean_token_accuracy": 0.10701777413487434,
|
|
"num_tokens": 2611909.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.327322006225586,
|
|
"epoch": 0.07874015748031496,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.2134,
|
|
"mean_token_accuracy": 0.11941884234547614,
|
|
"num_tokens": 2621273.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.27476658821106,
|
|
"epoch": 0.07903071156695819,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.2408,
|
|
"mean_token_accuracy": 0.11688115298748017,
|
|
"num_tokens": 2631072.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.332375431060791,
|
|
"epoch": 0.07932126565360141,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.3279,
|
|
"mean_token_accuracy": 0.10464770346879959,
|
|
"num_tokens": 2641868.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.315234565734864,
|
|
"epoch": 0.07961181974024464,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.2211,
|
|
"mean_token_accuracy": 0.11905064806342125,
|
|
"num_tokens": 2651495.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.127819538116455,
|
|
"epoch": 0.07990237382688788,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.1337,
|
|
"mean_token_accuracy": 0.12542947009205818,
|
|
"num_tokens": 2660418.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.218345642089844,
|
|
"epoch": 0.08019292791353111,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.3179,
|
|
"mean_token_accuracy": 0.11425245255231857,
|
|
"num_tokens": 2670656.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.304035377502442,
|
|
"epoch": 0.08048348200017433,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.1612,
|
|
"mean_token_accuracy": 0.12277446761727333,
|
|
"num_tokens": 2679806.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.233858823776245,
|
|
"epoch": 0.08077403608681756,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.2365,
|
|
"mean_token_accuracy": 0.11297848522663116,
|
|
"num_tokens": 2689690.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.260920190811158,
|
|
"epoch": 0.0810645901734608,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.2509,
|
|
"mean_token_accuracy": 0.11679463014006615,
|
|
"num_tokens": 2699923.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.293380784988403,
|
|
"epoch": 0.08135514426010401,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.2307,
|
|
"mean_token_accuracy": 0.12221046313643455,
|
|
"num_tokens": 2709476.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.20638747215271,
|
|
"epoch": 0.08164569834674724,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.243,
|
|
"mean_token_accuracy": 0.12326459586620331,
|
|
"num_tokens": 2719235.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.2035150051116945,
|
|
"epoch": 0.08193625243339048,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.1337,
|
|
"mean_token_accuracy": 0.12361931875348091,
|
|
"num_tokens": 2728275.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.172535991668701,
|
|
"epoch": 0.08222680652003371,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.2901,
|
|
"mean_token_accuracy": 0.11154365614056587,
|
|
"num_tokens": 2739138.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.309834671020508,
|
|
"epoch": 0.08251736060667693,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.1975,
|
|
"mean_token_accuracy": 0.12241730839014053,
|
|
"num_tokens": 2749077.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.187080812454224,
|
|
"epoch": 0.08280791469332016,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.1652,
|
|
"mean_token_accuracy": 0.12800487205386163,
|
|
"num_tokens": 2758005.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.194875431060791,
|
|
"epoch": 0.0830984687799634,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.1579,
|
|
"mean_token_accuracy": 0.12136282622814179,
|
|
"num_tokens": 2768012.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.2406840324401855,
|
|
"epoch": 0.08338902286660661,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.2887,
|
|
"mean_token_accuracy": 0.11709356978535652,
|
|
"num_tokens": 2778694.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.249140071868896,
|
|
"epoch": 0.08367957695324985,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.1696,
|
|
"mean_token_accuracy": 0.12004066854715348,
|
|
"num_tokens": 2788970.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.281752204895019,
|
|
"epoch": 0.08397013103989308,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.1724,
|
|
"mean_token_accuracy": 0.126155137270689,
|
|
"num_tokens": 2797712.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.065038061141967,
|
|
"epoch": 0.08426068512653631,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.1986,
|
|
"mean_token_accuracy": 0.1191088818013668,
|
|
"num_tokens": 2807812.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.368619108200074,
|
|
"epoch": 0.08455123921317953,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.2814,
|
|
"mean_token_accuracy": 0.11855203360319137,
|
|
"num_tokens": 2817983.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.161626720428467,
|
|
"epoch": 0.08484179329982276,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.1715,
|
|
"mean_token_accuracy": 0.11652336046099662,
|
|
"num_tokens": 2827936.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.251077365875244,
|
|
"epoch": 0.085132347386466,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.1981,
|
|
"mean_token_accuracy": 0.13417380750179292,
|
|
"num_tokens": 2838088.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.252039957046509,
|
|
"epoch": 0.08542290147310921,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.1931,
|
|
"mean_token_accuracy": 0.12354014292359353,
|
|
"num_tokens": 2847463.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.080291652679444,
|
|
"epoch": 0.08571345555975245,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.1409,
|
|
"mean_token_accuracy": 0.12890383899211882,
|
|
"num_tokens": 2857021.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.2580098628997805,
|
|
"epoch": 0.08600400964639568,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.1901,
|
|
"mean_token_accuracy": 0.12243498712778092,
|
|
"num_tokens": 2867577.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.258398723602295,
|
|
"epoch": 0.08629456373303891,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.2612,
|
|
"mean_token_accuracy": 0.1226688154041767,
|
|
"num_tokens": 2877508.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.195114278793335,
|
|
"epoch": 0.08658511781968213,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.1484,
|
|
"mean_token_accuracy": 0.12138795107603073,
|
|
"num_tokens": 2887193.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.263925457000733,
|
|
"epoch": 0.08687567190632536,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.1099,
|
|
"mean_token_accuracy": 0.12464227676391601,
|
|
"num_tokens": 2896345.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.047787523269653,
|
|
"epoch": 0.0871662259929686,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.0758,
|
|
"mean_token_accuracy": 0.12386289611458778,
|
|
"num_tokens": 2905855.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.170946264266968,
|
|
"epoch": 0.08745678007961181,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.1479,
|
|
"mean_token_accuracy": 0.12517718896269797,
|
|
"num_tokens": 2914747.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.053203916549682,
|
|
"epoch": 0.08774733416625505,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.0277,
|
|
"mean_token_accuracy": 0.1249882385134697,
|
|
"num_tokens": 2924085.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.226366186141968,
|
|
"epoch": 0.08803788825289828,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.184,
|
|
"mean_token_accuracy": 0.12349544689059258,
|
|
"num_tokens": 2934137.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.120088529586792,
|
|
"epoch": 0.0883284423395415,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.0708,
|
|
"mean_token_accuracy": 0.12396376803517342,
|
|
"num_tokens": 2943868.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.102652740478516,
|
|
"epoch": 0.08861899642618473,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.1155236929655075,
|
|
"num_tokens": 2953399.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.192172336578369,
|
|
"epoch": 0.08890955051282796,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.1235,
|
|
"mean_token_accuracy": 0.12404483556747437,
|
|
"num_tokens": 2963356.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.156837511062622,
|
|
"epoch": 0.0892001045994712,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.1485,
|
|
"mean_token_accuracy": 0.11465330868959427,
|
|
"num_tokens": 2973111.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.199955844879151,
|
|
"epoch": 0.08949065868611442,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.2266,
|
|
"mean_token_accuracy": 0.12046344429254532,
|
|
"num_tokens": 2983410.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.216725921630859,
|
|
"epoch": 0.08978121277275765,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.1542,
|
|
"mean_token_accuracy": 0.12176189124584198,
|
|
"num_tokens": 2992168.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.1444254398345945,
|
|
"epoch": 0.09007176685940088,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.124,
|
|
"mean_token_accuracy": 0.11986778303980827,
|
|
"num_tokens": 3001813.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.1660035133361815,
|
|
"epoch": 0.0903623209460441,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.2097,
|
|
"mean_token_accuracy": 0.12152451127767563,
|
|
"num_tokens": 3010646.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.243027400970459,
|
|
"epoch": 0.09065287503268733,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.1377,
|
|
"mean_token_accuracy": 0.12305564433336258,
|
|
"num_tokens": 3019647.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.047219085693359,
|
|
"epoch": 0.09094342911933057,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.0258,
|
|
"mean_token_accuracy": 0.12844225764274597,
|
|
"num_tokens": 3029071.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.0907470226287845,
|
|
"epoch": 0.0912339832059738,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.0384,
|
|
"mean_token_accuracy": 0.13014808297157288,
|
|
"num_tokens": 3037978.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.1422034740448,
|
|
"epoch": 0.09152453729261702,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.1474,
|
|
"mean_token_accuracy": 0.12799782454967498,
|
|
"num_tokens": 3047312.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.136872339248657,
|
|
"epoch": 0.09181509137926025,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.0871,
|
|
"mean_token_accuracy": 0.12448323741555214,
|
|
"num_tokens": 3055657.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.184458923339844,
|
|
"epoch": 0.09210564546590348,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.2451,
|
|
"mean_token_accuracy": 0.1185241773724556,
|
|
"num_tokens": 3066344.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.301413631439209,
|
|
"epoch": 0.0923961995525467,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.2121,
|
|
"mean_token_accuracy": 0.11998013481497764,
|
|
"num_tokens": 3076481.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.087073802947998,
|
|
"epoch": 0.09268675363918993,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.0722,
|
|
"mean_token_accuracy": 0.12851928919553757,
|
|
"num_tokens": 3085647.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.2406998634338375,
|
|
"epoch": 0.09297730772583317,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.1909,
|
|
"mean_token_accuracy": 0.12459061220288277,
|
|
"num_tokens": 3095085.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.156445837020874,
|
|
"epoch": 0.0932678618124764,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.1138,
|
|
"mean_token_accuracy": 0.12587291449308396,
|
|
"num_tokens": 3104313.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.215060758590698,
|
|
"epoch": 0.09355841589911962,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.1969,
|
|
"mean_token_accuracy": 0.11667787209153176,
|
|
"num_tokens": 3115071.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.139194774627685,
|
|
"epoch": 0.09384896998576285,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.12204712033271789,
|
|
"num_tokens": 3124152.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.224874258041382,
|
|
"epoch": 0.09413952407240608,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.1518,
|
|
"mean_token_accuracy": 0.12817504107952118,
|
|
"num_tokens": 3133733.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.059720706939697,
|
|
"epoch": 0.0944300781590493,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.0599,
|
|
"mean_token_accuracy": 0.12859237790107728,
|
|
"num_tokens": 3142843.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.157731437683106,
|
|
"epoch": 0.09472063224569253,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.0586,
|
|
"mean_token_accuracy": 0.12594763189554214,
|
|
"num_tokens": 3152091.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.018420934677124,
|
|
"epoch": 0.09501118633233577,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.1296,
|
|
"mean_token_accuracy": 0.12261961698532105,
|
|
"num_tokens": 3162594.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.239905881881714,
|
|
"epoch": 0.09530174041897899,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.1522,
|
|
"mean_token_accuracy": 0.1294144131243229,
|
|
"num_tokens": 3172575.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.194077634811402,
|
|
"epoch": 0.09559229450562222,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.1598,
|
|
"mean_token_accuracy": 0.12255481034517288,
|
|
"num_tokens": 3182066.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.107237339019775,
|
|
"epoch": 0.09588284859226545,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.0264,
|
|
"mean_token_accuracy": 0.12209519520401954,
|
|
"num_tokens": 3191093.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.18548550605774,
|
|
"epoch": 0.09617340267890868,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.1164,
|
|
"mean_token_accuracy": 0.1266675502061844,
|
|
"num_tokens": 3200466.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.148427295684814,
|
|
"epoch": 0.0964639567655519,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.0961,
|
|
"mean_token_accuracy": 0.1309407591819763,
|
|
"num_tokens": 3209962.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.137912273406982,
|
|
"epoch": 0.09675451085219514,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.0753,
|
|
"mean_token_accuracy": 0.13115072175860404,
|
|
"num_tokens": 3219735.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.088384389877319,
|
|
"epoch": 0.09704506493883837,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.125,
|
|
"mean_token_accuracy": 0.1225659005343914,
|
|
"num_tokens": 3228793.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.067927408218384,
|
|
"epoch": 0.09733561902548159,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 5.9736,
|
|
"mean_token_accuracy": 0.1300351545214653,
|
|
"num_tokens": 3238045.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.178382873535156,
|
|
"epoch": 0.09762617311212482,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.1776,
|
|
"mean_token_accuracy": 0.12008099108934403,
|
|
"num_tokens": 3248238.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.156318283081054,
|
|
"epoch": 0.09791672719876805,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.0686,
|
|
"mean_token_accuracy": 0.1366035208106041,
|
|
"num_tokens": 3256445.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.028420305252075,
|
|
"epoch": 0.09820728128541129,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.0412,
|
|
"mean_token_accuracy": 0.1275148406624794,
|
|
"num_tokens": 3266342.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.081392288208008,
|
|
"epoch": 0.0984978353720545,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.0836,
|
|
"mean_token_accuracy": 0.12850913628935814,
|
|
"num_tokens": 3276081.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.141415071487427,
|
|
"epoch": 0.09878838945869774,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.094,
|
|
"mean_token_accuracy": 0.12524966821074485,
|
|
"num_tokens": 3285228.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.129530048370361,
|
|
"epoch": 0.09907894354534097,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.1391,
|
|
"mean_token_accuracy": 0.1269663617014885,
|
|
"num_tokens": 3294392.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.269520378112793,
|
|
"epoch": 0.09936949763198419,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.2016,
|
|
"mean_token_accuracy": 0.11826551407575607,
|
|
"num_tokens": 3304255.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.058578109741211,
|
|
"epoch": 0.09966005171862742,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 5.9811,
|
|
"mean_token_accuracy": 0.1295688658952713,
|
|
"num_tokens": 3312731.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.168110990524292,
|
|
"epoch": 0.09995060580527065,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.0579,
|
|
"mean_token_accuracy": 0.12608980387449265,
|
|
"num_tokens": 3322178.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.064984083175659,
|
|
"epoch": 0.10024115989191389,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.0326,
|
|
"mean_token_accuracy": 0.12514449581503867,
|
|
"num_tokens": 3331688.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.188555145263672,
|
|
"epoch": 0.1005317139785571,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.2908,
|
|
"mean_token_accuracy": 0.11966812238097191,
|
|
"num_tokens": 3342151.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.153595590591431,
|
|
"epoch": 0.10082226806520034,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.06,
|
|
"mean_token_accuracy": 0.13309167325496674,
|
|
"num_tokens": 3351405.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.072870206832886,
|
|
"epoch": 0.10111282215184357,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.0799,
|
|
"mean_token_accuracy": 0.12330102473497391,
|
|
"num_tokens": 3362122.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.11573429107666,
|
|
"epoch": 0.10140337623848679,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.1035,
|
|
"mean_token_accuracy": 0.125773186981678,
|
|
"num_tokens": 3372797.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.074742650985717,
|
|
"epoch": 0.10169393032513002,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 5.999,
|
|
"mean_token_accuracy": 0.13261257037520408,
|
|
"num_tokens": 3382574.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.100408744812012,
|
|
"epoch": 0.10198448441177325,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.1297650747001171,
|
|
"num_tokens": 3392128.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.075628852844238,
|
|
"epoch": 0.10227503849841647,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 5.9802,
|
|
"mean_token_accuracy": 0.1275343604385853,
|
|
"num_tokens": 3401154.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.036476182937622,
|
|
"epoch": 0.1025655925850597,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.0298,
|
|
"mean_token_accuracy": 0.13419338539242745,
|
|
"num_tokens": 3410133.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.148456811904907,
|
|
"epoch": 0.10285614667170294,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.0791,
|
|
"mean_token_accuracy": 0.12794992104172706,
|
|
"num_tokens": 3420270.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.103959608078003,
|
|
"epoch": 0.10314670075834617,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.0662,
|
|
"mean_token_accuracy": 0.12644353434443473,
|
|
"num_tokens": 3430854.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.200176763534546,
|
|
"epoch": 0.10343725484498939,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.035,
|
|
"mean_token_accuracy": 0.12726576030254363,
|
|
"num_tokens": 3441280.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.975127696990967,
|
|
"epoch": 0.10372780893163262,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 5.9856,
|
|
"mean_token_accuracy": 0.13601229563355446,
|
|
"num_tokens": 3450103.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.0330243587493895,
|
|
"epoch": 0.10401836301827586,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.0696,
|
|
"mean_token_accuracy": 0.12510816380381584,
|
|
"num_tokens": 3461013.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.096307325363159,
|
|
"epoch": 0.10430891710491907,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 5.9965,
|
|
"mean_token_accuracy": 0.13395112082362176,
|
|
"num_tokens": 3470120.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.193869876861572,
|
|
"epoch": 0.10459947119156231,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.1934,
|
|
"mean_token_accuracy": 0.12477752342820167,
|
|
"num_tokens": 3480010.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.083474636077881,
|
|
"epoch": 0.10489002527820554,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.0506,
|
|
"mean_token_accuracy": 0.12901688143610954,
|
|
"num_tokens": 3490140.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.154728937149048,
|
|
"epoch": 0.10518057936484877,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.0767,
|
|
"mean_token_accuracy": 0.12849546670913697,
|
|
"num_tokens": 3501120.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.052251720428467,
|
|
"epoch": 0.10547113345149199,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 5.8971,
|
|
"mean_token_accuracy": 0.13428612872958184,
|
|
"num_tokens": 3510199.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.032751893997192,
|
|
"epoch": 0.10576168753813522,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.0217,
|
|
"mean_token_accuracy": 0.12719288766384124,
|
|
"num_tokens": 3519146.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.124502849578858,
|
|
"epoch": 0.10605224162477846,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.0924,
|
|
"mean_token_accuracy": 0.13152782544493674,
|
|
"num_tokens": 3529281.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.984056615829468,
|
|
"epoch": 0.10634279571142168,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.0373,
|
|
"mean_token_accuracy": 0.1306003786623478,
|
|
"num_tokens": 3539371.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.161852884292602,
|
|
"epoch": 0.10663334979806491,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.1009,
|
|
"mean_token_accuracy": 0.12435954585671424,
|
|
"num_tokens": 3549375.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.991615581512451,
|
|
"epoch": 0.10692390388470814,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.0098,
|
|
"mean_token_accuracy": 0.12970533594489098,
|
|
"num_tokens": 3558935.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.1774444580078125,
|
|
"epoch": 0.10721445797135137,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.0616,
|
|
"mean_token_accuracy": 0.12848633974790574,
|
|
"num_tokens": 3568947.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.936046028137207,
|
|
"epoch": 0.10750501205799459,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 5.9816,
|
|
"mean_token_accuracy": 0.1268470034003258,
|
|
"num_tokens": 3579479.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.122731304168701,
|
|
"epoch": 0.10779556614463782,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 5.9675,
|
|
"mean_token_accuracy": 0.12663150280714036,
|
|
"num_tokens": 3588901.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.087626218795776,
|
|
"epoch": 0.10808612023128106,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 5.968,
|
|
"mean_token_accuracy": 0.13817571178078653,
|
|
"num_tokens": 3598290.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.992796802520752,
|
|
"epoch": 0.10837667431792428,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.0643,
|
|
"mean_token_accuracy": 0.12368949353694916,
|
|
"num_tokens": 3608000.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.115261077880859,
|
|
"epoch": 0.10866722840456751,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.0252,
|
|
"mean_token_accuracy": 0.12702150121331215,
|
|
"num_tokens": 3617557.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.0366315841674805,
|
|
"epoch": 0.10895778249121074,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.0296,
|
|
"mean_token_accuracy": 0.12519691735506058,
|
|
"num_tokens": 3628784.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.1003316879272464,
|
|
"epoch": 0.10924833657785397,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 5.9738,
|
|
"mean_token_accuracy": 0.1348504453897476,
|
|
"num_tokens": 3637362.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.078695774078369,
|
|
"epoch": 0.1095388906644972,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.0113,
|
|
"mean_token_accuracy": 0.13162249848246574,
|
|
"num_tokens": 3647236.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.04180326461792,
|
|
"epoch": 0.10982944475114043,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 5.9672,
|
|
"mean_token_accuracy": 0.13503263145685196,
|
|
"num_tokens": 3656778.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.088074684143066,
|
|
"epoch": 0.11011999883778366,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.1134,
|
|
"mean_token_accuracy": 0.12615369856357575,
|
|
"num_tokens": 3665659.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.993572330474853,
|
|
"epoch": 0.11041055292442688,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.13086738660931588,
|
|
"num_tokens": 3675998.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.057390117645264,
|
|
"epoch": 0.11070110701107011,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 5.9815,
|
|
"mean_token_accuracy": 0.12661390975117684,
|
|
"num_tokens": 3686429.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.126709651947022,
|
|
"epoch": 0.11099166109771334,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.0305,
|
|
"mean_token_accuracy": 0.13001628071069718,
|
|
"num_tokens": 3696611.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.0224635124206545,
|
|
"epoch": 0.11128221518435656,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 5.8559,
|
|
"mean_token_accuracy": 0.14033628329634668,
|
|
"num_tokens": 3706164.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.912648868560791,
|
|
"epoch": 0.1115727692709998,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 5.9874,
|
|
"mean_token_accuracy": 0.1366017334163189,
|
|
"num_tokens": 3715656.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.123447513580322,
|
|
"epoch": 0.11186332335764303,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.037,
|
|
"mean_token_accuracy": 0.13282812014222145,
|
|
"num_tokens": 3724989.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.025955486297607,
|
|
"epoch": 0.11215387744428626,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 5.9432,
|
|
"mean_token_accuracy": 0.1403130918741226,
|
|
"num_tokens": 3733997.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.990260124206543,
|
|
"epoch": 0.11244443153092948,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.0374,
|
|
"mean_token_accuracy": 0.1268382787704468,
|
|
"num_tokens": 3743987.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.108919668197632,
|
|
"epoch": 0.11273498561757271,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.0427,
|
|
"mean_token_accuracy": 0.12940073609352112,
|
|
"num_tokens": 3753455.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.063518142700195,
|
|
"epoch": 0.11302553970421594,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 5.9631,
|
|
"mean_token_accuracy": 0.13474399223923683,
|
|
"num_tokens": 3762574.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.006039571762085,
|
|
"epoch": 0.11331609379085916,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.9105,
|
|
"mean_token_accuracy": 0.13433051109313965,
|
|
"num_tokens": 3771030.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.05625548362732,
|
|
"epoch": 0.1136066478775024,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.0116,
|
|
"mean_token_accuracy": 0.12894580587744714,
|
|
"num_tokens": 3780165.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.104951047897339,
|
|
"epoch": 0.11389720196414563,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 5.9758,
|
|
"mean_token_accuracy": 0.1321735166013241,
|
|
"num_tokens": 3789668.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.019385576248169,
|
|
"epoch": 0.11418775605078886,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.0284,
|
|
"mean_token_accuracy": 0.1343548469245434,
|
|
"num_tokens": 3799521.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.024891090393067,
|
|
"epoch": 0.11447831013743208,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.8961,
|
|
"mean_token_accuracy": 0.14202071875333785,
|
|
"num_tokens": 3808626.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.0244913578033445,
|
|
"epoch": 0.11476886422407531,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 5.9947,
|
|
"mean_token_accuracy": 0.125540604442358,
|
|
"num_tokens": 3817822.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.006051397323608,
|
|
"epoch": 0.11505941831071854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.9396,
|
|
"mean_token_accuracy": 0.14070837944746017,
|
|
"num_tokens": 3827221.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.057012033462525,
|
|
"epoch": 0.11534997239736176,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 5.9645,
|
|
"mean_token_accuracy": 0.12805723547935485,
|
|
"num_tokens": 3836430.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.025897645950318,
|
|
"epoch": 0.115640526484005,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 5.9427,
|
|
"mean_token_accuracy": 0.13494896292686462,
|
|
"num_tokens": 3846131.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.960288715362549,
|
|
"epoch": 0.11593108057064823,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 5.8965,
|
|
"mean_token_accuracy": 0.13898257911205292,
|
|
"num_tokens": 3855807.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.005527400970459,
|
|
"epoch": 0.11622163465729146,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 5.9919,
|
|
"mean_token_accuracy": 0.1341300755739212,
|
|
"num_tokens": 3865577.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.020885372161866,
|
|
"epoch": 0.11651218874393468,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 6.0277,
|
|
"mean_token_accuracy": 0.12657435163855552,
|
|
"num_tokens": 3875005.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.156516599655151,
|
|
"epoch": 0.11680274283057791,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 6.0878,
|
|
"mean_token_accuracy": 0.12202497869729996,
|
|
"num_tokens": 3884037.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.17493462562561,
|
|
"epoch": 0.11709329691722115,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 6.1017,
|
|
"mean_token_accuracy": 0.12757360935211182,
|
|
"num_tokens": 3893965.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.048020505905152,
|
|
"epoch": 0.11738385100386436,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 5.9497,
|
|
"mean_token_accuracy": 0.14000242203474045,
|
|
"num_tokens": 3903703.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.98990831375122,
|
|
"epoch": 0.1176744050905076,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 5.9054,
|
|
"mean_token_accuracy": 0.1310841754078865,
|
|
"num_tokens": 3913045.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.00819730758667,
|
|
"epoch": 0.11796495917715083,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.13758676499128342,
|
|
"num_tokens": 3922410.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.037257862091065,
|
|
"epoch": 0.11825551326379405,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 5.9736,
|
|
"mean_token_accuracy": 0.130282711237669,
|
|
"num_tokens": 3931458.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.077354669570923,
|
|
"epoch": 0.11854606735043728,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 6.0508,
|
|
"mean_token_accuracy": 0.13168376535177231,
|
|
"num_tokens": 3942121.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.044887590408325,
|
|
"epoch": 0.11883662143708051,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 5.9959,
|
|
"mean_token_accuracy": 0.13084948435425758,
|
|
"num_tokens": 3952253.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.054182767868042,
|
|
"epoch": 0.11912717552372375,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 6.0084,
|
|
"mean_token_accuracy": 0.12859943136572838,
|
|
"num_tokens": 3961789.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.001371192932129,
|
|
"epoch": 0.11941772961036697,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 5.9063,
|
|
"mean_token_accuracy": 0.1340605929493904,
|
|
"num_tokens": 3971730.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.027997779846191,
|
|
"epoch": 0.1197082836970102,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 5.9682,
|
|
"mean_token_accuracy": 0.12761855274438857,
|
|
"num_tokens": 3982356.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.0816364765167235,
|
|
"epoch": 0.11999883778365343,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 5.9698,
|
|
"mean_token_accuracy": 0.12905391156673432,
|
|
"num_tokens": 3992619.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.949817657470703,
|
|
"epoch": 0.12028939187029665,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 5.8642,
|
|
"mean_token_accuracy": 0.13732533380389214,
|
|
"num_tokens": 4002233.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.997682762145996,
|
|
"epoch": 0.12057994595693988,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 6.0073,
|
|
"mean_token_accuracy": 0.12828967198729516,
|
|
"num_tokens": 4012265.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.046585321426392,
|
|
"epoch": 0.12087050004358312,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 5.9949,
|
|
"mean_token_accuracy": 0.13329361379146576,
|
|
"num_tokens": 4021899.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.136781024932861,
|
|
"epoch": 0.12116105413022635,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 6.1281,
|
|
"mean_token_accuracy": 0.12372240424156189,
|
|
"num_tokens": 4031951.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.959255409240723,
|
|
"epoch": 0.12145160821686957,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 5.7985,
|
|
"mean_token_accuracy": 0.14132873937487603,
|
|
"num_tokens": 4040535.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.875403928756714,
|
|
"epoch": 0.1217421623035128,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 5.8698,
|
|
"mean_token_accuracy": 0.14000626653432846,
|
|
"num_tokens": 4049951.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.094307279586792,
|
|
"epoch": 0.12203271639015603,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.0269,
|
|
"mean_token_accuracy": 0.12536544725298882,
|
|
"num_tokens": 4059623.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.034672784805298,
|
|
"epoch": 0.12232327047679925,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 5.8494,
|
|
"mean_token_accuracy": 0.14006635397672654,
|
|
"num_tokens": 4068820.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.914458560943603,
|
|
"epoch": 0.12261382456344248,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 5.8758,
|
|
"mean_token_accuracy": 0.13851067423820496,
|
|
"num_tokens": 4078699.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.967116165161133,
|
|
"epoch": 0.12290437865008572,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 5.8401,
|
|
"mean_token_accuracy": 0.14074087589979173,
|
|
"num_tokens": 4087998.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.043182373046875,
|
|
"epoch": 0.12319493273672895,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 5.9877,
|
|
"mean_token_accuracy": 0.13383241593837739,
|
|
"num_tokens": 4097091.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.932327079772949,
|
|
"epoch": 0.12348548682337217,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 5.9291,
|
|
"mean_token_accuracy": 0.13196048736572266,
|
|
"num_tokens": 4106964.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.049034261703492,
|
|
"epoch": 0.1237760409100154,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 5.9481,
|
|
"mean_token_accuracy": 0.1403219759464264,
|
|
"num_tokens": 4116898.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.968407535552979,
|
|
"epoch": 0.12406659499665863,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.1425372302532196,
|
|
"num_tokens": 4125706.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.0740570545196535,
|
|
"epoch": 0.12435714908330185,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 6.0003,
|
|
"mean_token_accuracy": 0.1340542823076248,
|
|
"num_tokens": 4135601.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.985355806350708,
|
|
"epoch": 0.12464770316994508,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 5.9452,
|
|
"mean_token_accuracy": 0.13526148721575737,
|
|
"num_tokens": 4145049.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.050859975814819,
|
|
"epoch": 0.12493825725658832,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 5.9104,
|
|
"mean_token_accuracy": 0.1342906855046749,
|
|
"num_tokens": 4154089.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.029706764221191,
|
|
"epoch": 0.12522881134323155,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 5.9294,
|
|
"mean_token_accuracy": 0.1347600869834423,
|
|
"num_tokens": 4163729.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.928568124771118,
|
|
"epoch": 0.12551936542987477,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 5.8799,
|
|
"mean_token_accuracy": 0.13473999574780465,
|
|
"num_tokens": 4173027.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.970612621307373,
|
|
"epoch": 0.125809919516518,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 5.9384,
|
|
"mean_token_accuracy": 0.13225939273834228,
|
|
"num_tokens": 4182703.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.1540978908538815,
|
|
"epoch": 0.12610047360316123,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 5.985,
|
|
"mean_token_accuracy": 0.13003097623586654,
|
|
"num_tokens": 4192927.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.967820882797241,
|
|
"epoch": 0.12639102768980445,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 5.9551,
|
|
"mean_token_accuracy": 0.1343395359814167,
|
|
"num_tokens": 4203562.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.968146181106567,
|
|
"epoch": 0.1266815817764477,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 5.909,
|
|
"mean_token_accuracy": 0.14129758030176162,
|
|
"num_tokens": 4213223.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.010010147094727,
|
|
"epoch": 0.12697213586309092,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.8735,
|
|
"mean_token_accuracy": 0.13559180721640587,
|
|
"num_tokens": 4222676.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.019833612442016,
|
|
"epoch": 0.12726268994973414,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 5.9847,
|
|
"mean_token_accuracy": 0.1309241108596325,
|
|
"num_tokens": 4232854.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.01093168258667,
|
|
"epoch": 0.12755324403637738,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 5.8711,
|
|
"mean_token_accuracy": 0.14472563490271567,
|
|
"num_tokens": 4242702.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.029361248016357,
|
|
"epoch": 0.1278437981230206,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 5.9419,
|
|
"mean_token_accuracy": 0.13157446980476378,
|
|
"num_tokens": 4252837.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.959269094467163,
|
|
"epoch": 0.12813435220966382,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 5.8399,
|
|
"mean_token_accuracy": 0.1406804084777832,
|
|
"num_tokens": 4261958.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.050781154632569,
|
|
"epoch": 0.12842490629630707,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.14063480123877525,
|
|
"num_tokens": 4270916.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.006767225265503,
|
|
"epoch": 0.1287154603829503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 5.9463,
|
|
"mean_token_accuracy": 0.1407653845846653,
|
|
"num_tokens": 4280614.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.892512798309326,
|
|
"epoch": 0.1290060144695935,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 5.8542,
|
|
"mean_token_accuracy": 0.13779841586947442,
|
|
"num_tokens": 4290511.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.083350658416748,
|
|
"epoch": 0.12929656855623675,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 5.9252,
|
|
"mean_token_accuracy": 0.13738311529159547,
|
|
"num_tokens": 4300101.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.020438623428345,
|
|
"epoch": 0.12958712264287997,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 5.898,
|
|
"mean_token_accuracy": 0.13345685228705406,
|
|
"num_tokens": 4310263.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.936104154586792,
|
|
"epoch": 0.1298776767295232,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 5.9102,
|
|
"mean_token_accuracy": 0.13602888882160186,
|
|
"num_tokens": 4319472.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.055332803726197,
|
|
"epoch": 0.13016823081616644,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 5.8534,
|
|
"mean_token_accuracy": 0.1360207498073578,
|
|
"num_tokens": 4329228.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.921377229690552,
|
|
"epoch": 0.13045878490280965,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.13516351208090782,
|
|
"num_tokens": 4339816.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.94791316986084,
|
|
"epoch": 0.13074933898945287,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 5.8351,
|
|
"mean_token_accuracy": 0.13289576545357704,
|
|
"num_tokens": 4349152.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.988745975494385,
|
|
"epoch": 0.13103989307609612,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 5.9542,
|
|
"mean_token_accuracy": 0.13524625673890114,
|
|
"num_tokens": 4359073.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.959168386459351,
|
|
"epoch": 0.13133044716273934,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 5.9377,
|
|
"mean_token_accuracy": 0.136678983271122,
|
|
"num_tokens": 4368437.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.065321207046509,
|
|
"epoch": 0.13162100124938259,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.13410670906305314,
|
|
"num_tokens": 4378430.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.087664270401001,
|
|
"epoch": 0.1319115553360258,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 5.9672,
|
|
"mean_token_accuracy": 0.13350549340248108,
|
|
"num_tokens": 4388500.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.964868593215942,
|
|
"epoch": 0.13220210942266902,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 5.8422,
|
|
"mean_token_accuracy": 0.14961724877357482,
|
|
"num_tokens": 4398232.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.969161415100098,
|
|
"epoch": 0.13249266350931227,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 5.9124,
|
|
"mean_token_accuracy": 0.1390094742178917,
|
|
"num_tokens": 4406569.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.083000707626343,
|
|
"epoch": 0.1327832175959555,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.14335579425096512,
|
|
"num_tokens": 4415634.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.942263126373291,
|
|
"epoch": 0.1330737716825987,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 5.9028,
|
|
"mean_token_accuracy": 0.1399638831615448,
|
|
"num_tokens": 4426079.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.997076416015625,
|
|
"epoch": 0.13336432576924195,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 5.9007,
|
|
"mean_token_accuracy": 0.1342967502772808,
|
|
"num_tokens": 4435587.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.9411468505859375,
|
|
"epoch": 0.13365487985588517,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 5.852,
|
|
"mean_token_accuracy": 0.13950249254703523,
|
|
"num_tokens": 4445466.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.994159173965454,
|
|
"epoch": 0.1339454339425284,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.1332622177898884,
|
|
"num_tokens": 4455342.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.0287621974945065,
|
|
"epoch": 0.13423598802917164,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 5.8712,
|
|
"mean_token_accuracy": 0.13551222681999206,
|
|
"num_tokens": 4464358.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.003249263763427,
|
|
"epoch": 0.13452654211581486,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 5.85,
|
|
"mean_token_accuracy": 0.1360574722290039,
|
|
"num_tokens": 4473419.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.895709609985351,
|
|
"epoch": 0.13481709620245808,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 5.8087,
|
|
"mean_token_accuracy": 0.13682912141084672,
|
|
"num_tokens": 4483426.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.04387469291687,
|
|
"epoch": 0.13510765028910132,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 5.9602,
|
|
"mean_token_accuracy": 0.13195119872689248,
|
|
"num_tokens": 4493213.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.024215412139893,
|
|
"epoch": 0.13539820437574454,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.14107481837272645,
|
|
"num_tokens": 4503189.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.925279331207276,
|
|
"epoch": 0.13568875846238776,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 5.9355,
|
|
"mean_token_accuracy": 0.12918271347880364,
|
|
"num_tokens": 4513253.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.961361694335937,
|
|
"epoch": 0.135979312549031,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 5.7951,
|
|
"mean_token_accuracy": 0.1382595956325531,
|
|
"num_tokens": 4523419.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.926408576965332,
|
|
"epoch": 0.13626986663567422,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 5.9564,
|
|
"mean_token_accuracy": 0.13528416752815248,
|
|
"num_tokens": 4533585.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.992029476165771,
|
|
"epoch": 0.13656042072231747,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.14141826182603837,
|
|
"num_tokens": 4541863.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.003516006469726,
|
|
"epoch": 0.1368509748089607,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 5.8371,
|
|
"mean_token_accuracy": 0.13811903372406958,
|
|
"num_tokens": 4551044.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.98145227432251,
|
|
"epoch": 0.1371415288956039,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 5.8757,
|
|
"mean_token_accuracy": 0.14308483600616456,
|
|
"num_tokens": 4561525.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.967531442642212,
|
|
"epoch": 0.13743208298224716,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 5.8273,
|
|
"mean_token_accuracy": 0.14423855245113373,
|
|
"num_tokens": 4571012.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.948016929626465,
|
|
"epoch": 0.13772263706889037,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 5.8638,
|
|
"mean_token_accuracy": 0.13877268359065056,
|
|
"num_tokens": 4580840.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.953177118301392,
|
|
"epoch": 0.1380131911555336,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 5.9276,
|
|
"mean_token_accuracy": 0.13606794849038123,
|
|
"num_tokens": 4591093.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.916067981719971,
|
|
"epoch": 0.13830374524217684,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 5.8848,
|
|
"mean_token_accuracy": 0.1396774709224701,
|
|
"num_tokens": 4601449.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.057155704498291,
|
|
"epoch": 0.13859429932882006,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 5.9087,
|
|
"mean_token_accuracy": 0.1398300603032112,
|
|
"num_tokens": 4611314.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.022958469390869,
|
|
"epoch": 0.13888485341546328,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 5.9069,
|
|
"mean_token_accuracy": 0.13639484643936156,
|
|
"num_tokens": 4620461.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.024860906600952,
|
|
"epoch": 0.13917540750210652,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 5.8656,
|
|
"mean_token_accuracy": 0.14114097729325295,
|
|
"num_tokens": 4630198.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.89768385887146,
|
|
"epoch": 0.13946596158874974,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 5.7678,
|
|
"mean_token_accuracy": 0.1444575846195221,
|
|
"num_tokens": 4639275.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.9961847305297855,
|
|
"epoch": 0.13975651567539296,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 5.8535,
|
|
"mean_token_accuracy": 0.13934674188494683,
|
|
"num_tokens": 4648914.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.936286783218383,
|
|
"epoch": 0.1400470697620362,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 5.765,
|
|
"mean_token_accuracy": 0.14789522886276246,
|
|
"num_tokens": 4658351.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.933369731903076,
|
|
"epoch": 0.14033762384867943,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 5.779,
|
|
"mean_token_accuracy": 0.13668361231684684,
|
|
"num_tokens": 4667832.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.889568758010864,
|
|
"epoch": 0.14062817793532267,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.14445363059639932,
|
|
"num_tokens": 4678007.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.9436970233917235,
|
|
"epoch": 0.1409187320219659,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 5.9731,
|
|
"mean_token_accuracy": 0.1374736785888672,
|
|
"num_tokens": 4687980.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.0395537376403805,
|
|
"epoch": 0.1412092861086091,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 5.9202,
|
|
"mean_token_accuracy": 0.1379444718360901,
|
|
"num_tokens": 4699183.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.017334127426148,
|
|
"epoch": 0.14149984019525236,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 5.8333,
|
|
"mean_token_accuracy": 0.13580716103315355,
|
|
"num_tokens": 4708877.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.877793979644776,
|
|
"epoch": 0.14179039428189558,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 5.7952,
|
|
"mean_token_accuracy": 0.14586628898978232,
|
|
"num_tokens": 4718456.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.883253383636474,
|
|
"epoch": 0.1420809483685388,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 5.7874,
|
|
"mean_token_accuracy": 0.14279707446694373,
|
|
"num_tokens": 4728536.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.97650408744812,
|
|
"epoch": 0.14237150245518204,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 5.9474,
|
|
"mean_token_accuracy": 0.13733349069952966,
|
|
"num_tokens": 4738474.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.021295547485352,
|
|
"epoch": 0.14266205654182526,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 5.8469,
|
|
"mean_token_accuracy": 0.13698814958333969,
|
|
"num_tokens": 4748242.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.994341087341309,
|
|
"epoch": 0.14295261062846848,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.846,
|
|
"mean_token_accuracy": 0.14408844113349914,
|
|
"num_tokens": 4757787.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.905521297454834,
|
|
"epoch": 0.14324316471511173,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 5.7272,
|
|
"mean_token_accuracy": 0.14780522286891937,
|
|
"num_tokens": 4767178.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.858093166351319,
|
|
"epoch": 0.14353371880175494,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 5.7769,
|
|
"mean_token_accuracy": 0.1437581166625023,
|
|
"num_tokens": 4776463.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.954789066314698,
|
|
"epoch": 0.14382427288839816,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 5.815,
|
|
"mean_token_accuracy": 0.14446621090173722,
|
|
"num_tokens": 4785972.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.911713743209839,
|
|
"epoch": 0.1441148269750414,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 5.7652,
|
|
"mean_token_accuracy": 0.14482518881559373,
|
|
"num_tokens": 4794986.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.940880584716797,
|
|
"epoch": 0.14440538106168463,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 5.8926,
|
|
"mean_token_accuracy": 0.13828022554516792,
|
|
"num_tokens": 4804625.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.915801620483398,
|
|
"epoch": 0.14469593514832785,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 5.7716,
|
|
"mean_token_accuracy": 0.1409230798482895,
|
|
"num_tokens": 4813901.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.948210859298706,
|
|
"epoch": 0.1449864892349711,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 5.8636,
|
|
"mean_token_accuracy": 0.14530568122863768,
|
|
"num_tokens": 4823986.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.922775316238403,
|
|
"epoch": 0.1452770433216143,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 5.7392,
|
|
"mean_token_accuracy": 0.14593456238508223,
|
|
"num_tokens": 4832915.0,
|
|
"step": 2500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 7121498001408000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|