1035 lines
27 KiB
JSON
1035 lines
27 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.029055408664322865,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1462312691712000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|