3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.029952674773857307,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692037105560303,
|
|
"epoch": 9.984224924619101e-05,
|
|
"grad_norm": 15.3125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.6672,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 5422.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691985416412354,
|
|
"epoch": 0.00019968449849238202,
|
|
"grad_norm": 15.125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.6253,
|
|
"mean_token_accuracy": 0.00012738853693008422,
|
|
"num_tokens": 11441.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690535736083984,
|
|
"epoch": 0.00029952674773857304,
|
|
"grad_norm": 12.375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.3697,
|
|
"mean_token_accuracy": 0.03621518523432314,
|
|
"num_tokens": 17413.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.663197803497315,
|
|
"epoch": 0.00039936899698476405,
|
|
"grad_norm": 7.4375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.0374,
|
|
"mean_token_accuracy": 0.05610103793442249,
|
|
"num_tokens": 23280.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.522790908813477,
|
|
"epoch": 0.0004992112462309551,
|
|
"grad_norm": 5.1875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.7254,
|
|
"mean_token_accuracy": 0.05437180008739233,
|
|
"num_tokens": 28643.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.49898328781128,
|
|
"epoch": 0.0005990534954771461,
|
|
"grad_norm": 4.21875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.5672,
|
|
"mean_token_accuracy": 0.059307342022657396,
|
|
"num_tokens": 34253.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.442592239379882,
|
|
"epoch": 0.0006988957447233371,
|
|
"grad_norm": 3.609375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.5534,
|
|
"mean_token_accuracy": 0.053936305642127993,
|
|
"num_tokens": 39990.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.463098812103272,
|
|
"epoch": 0.0007987379939695281,
|
|
"grad_norm": 3.0625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.5075,
|
|
"mean_token_accuracy": 0.06420534774661064,
|
|
"num_tokens": 45837.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.428625583648682,
|
|
"epoch": 0.0008985802432157192,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.4782,
|
|
"mean_token_accuracy": 0.07273747362196445,
|
|
"num_tokens": 51998.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.35524263381958,
|
|
"epoch": 0.0009984224924619101,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.3763,
|
|
"mean_token_accuracy": 0.07940605096518993,
|
|
"num_tokens": 58235.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.194870853424073,
|
|
"epoch": 0.0010982647417081012,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.2541,
|
|
"mean_token_accuracy": 0.07720154039561748,
|
|
"num_tokens": 63553.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.117318916320801,
|
|
"epoch": 0.0011981069909542921,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.3244,
|
|
"mean_token_accuracy": 0.07141049914062023,
|
|
"num_tokens": 69696.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.213671588897705,
|
|
"epoch": 0.0012979492402004833,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.1874,
|
|
"mean_token_accuracy": 0.07848358415067196,
|
|
"num_tokens": 75580.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.183634948730468,
|
|
"epoch": 0.0013977914894466742,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.0286,
|
|
"mean_token_accuracy": 0.07954556941986084,
|
|
"num_tokens": 80808.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.02001485824585,
|
|
"epoch": 0.0014976337386928653,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1145,
|
|
"mean_token_accuracy": 0.07110607773065566,
|
|
"num_tokens": 86260.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.129467391967774,
|
|
"epoch": 0.0015974759879390562,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0261,
|
|
"mean_token_accuracy": 0.08673112615942954,
|
|
"num_tokens": 92099.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.037476825714112,
|
|
"epoch": 0.0016973182371852473,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.9278,
|
|
"mean_token_accuracy": 0.08497167229652405,
|
|
"num_tokens": 97720.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.079257011413574,
|
|
"epoch": 0.0017971604864314384,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9373,
|
|
"mean_token_accuracy": 0.07472074516117573,
|
|
"num_tokens": 103907.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.07486629486084,
|
|
"epoch": 0.0018970027356776293,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8796,
|
|
"mean_token_accuracy": 0.07506735101342202,
|
|
"num_tokens": 109981.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.837776279449463,
|
|
"epoch": 0.0019968449849238202,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.6173,
|
|
"mean_token_accuracy": 0.09049530178308487,
|
|
"num_tokens": 114997.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.909902858734132,
|
|
"epoch": 0.0020966872341700114,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6543,
|
|
"mean_token_accuracy": 0.08456408381462097,
|
|
"num_tokens": 120762.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.8260329246521,
|
|
"epoch": 0.0021965294834162025,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5515,
|
|
"mean_token_accuracy": 0.08778458833694458,
|
|
"num_tokens": 126113.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.766840267181397,
|
|
"epoch": 0.0022963717326623936,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.5581,
|
|
"mean_token_accuracy": 0.080570487678051,
|
|
"num_tokens": 132303.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.73247833251953,
|
|
"epoch": 0.0023962139819085843,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4172,
|
|
"mean_token_accuracy": 0.08937932215631009,
|
|
"num_tokens": 138642.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.6115159034729,
|
|
"epoch": 0.0024960562311547754,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.3259,
|
|
"mean_token_accuracy": 0.08444585129618645,
|
|
"num_tokens": 144420.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.505456352233887,
|
|
"epoch": 0.0025958984804009665,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2552,
|
|
"mean_token_accuracy": 0.08594180978834629,
|
|
"num_tokens": 150126.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.33946189880371,
|
|
"epoch": 0.0026957407296471577,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.1841,
|
|
"mean_token_accuracy": 0.08540923669934272,
|
|
"num_tokens": 156394.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.154175662994385,
|
|
"epoch": 0.0027955829788933483,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.0888,
|
|
"mean_token_accuracy": 0.08764742240309716,
|
|
"num_tokens": 161888.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.103712558746338,
|
|
"epoch": 0.0028954252281395395,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.0941,
|
|
"mean_token_accuracy": 0.08791576996445656,
|
|
"num_tokens": 167479.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.953013134002685,
|
|
"epoch": 0.0029952674773857306,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9641,
|
|
"mean_token_accuracy": 0.08592916205525399,
|
|
"num_tokens": 173013.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.789963626861573,
|
|
"epoch": 0.0030951097266319217,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.9985,
|
|
"mean_token_accuracy": 0.08673654571175575,
|
|
"num_tokens": 178908.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.74723539352417,
|
|
"epoch": 0.0031949519758781124,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.8767,
|
|
"mean_token_accuracy": 0.09191061705350875,
|
|
"num_tokens": 184314.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.576598930358887,
|
|
"epoch": 0.0032947942251243035,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8791,
|
|
"mean_token_accuracy": 0.08847894445061684,
|
|
"num_tokens": 190217.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.493614959716798,
|
|
"epoch": 0.0033946364743704946,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8022,
|
|
"mean_token_accuracy": 0.09182382375001907,
|
|
"num_tokens": 196154.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.42578411102295,
|
|
"epoch": 0.0034944787236166857,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8988,
|
|
"mean_token_accuracy": 0.08841787204146385,
|
|
"num_tokens": 202043.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.447116088867187,
|
|
"epoch": 0.003594320972862877,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.8551,
|
|
"mean_token_accuracy": 0.08922517634928226,
|
|
"num_tokens": 208843.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.342639446258545,
|
|
"epoch": 0.0036941632221090676,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.8137,
|
|
"mean_token_accuracy": 0.09233534485101699,
|
|
"num_tokens": 213777.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.31864881515503,
|
|
"epoch": 0.0037940054713552587,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7358,
|
|
"mean_token_accuracy": 0.0952209122478962,
|
|
"num_tokens": 220209.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.204437160491944,
|
|
"epoch": 0.00389384772060145,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.747,
|
|
"mean_token_accuracy": 0.0910322405397892,
|
|
"num_tokens": 225900.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.255455112457275,
|
|
"epoch": 0.0039936899698476405,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.8,
|
|
"mean_token_accuracy": 0.08869038000702859,
|
|
"num_tokens": 231781.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.125996208190918,
|
|
"epoch": 0.004093532219093832,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7429,
|
|
"mean_token_accuracy": 0.08568662032485008,
|
|
"num_tokens": 236974.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.232761573791503,
|
|
"epoch": 0.004193374468340023,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7475,
|
|
"mean_token_accuracy": 0.08775760419666767,
|
|
"num_tokens": 242588.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.146185207366944,
|
|
"epoch": 0.004293216717586213,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.6947,
|
|
"mean_token_accuracy": 0.0925628162920475,
|
|
"num_tokens": 248736.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.136151313781738,
|
|
"epoch": 0.004393058966832405,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.7161,
|
|
"mean_token_accuracy": 0.08190489560365677,
|
|
"num_tokens": 254348.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.15009708404541,
|
|
"epoch": 0.004492901216078596,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.6301,
|
|
"mean_token_accuracy": 0.09184228777885436,
|
|
"num_tokens": 259415.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.953672695159912,
|
|
"epoch": 0.004592743465324787,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.8223,
|
|
"mean_token_accuracy": 0.0824849434196949,
|
|
"num_tokens": 265073.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.194051551818848,
|
|
"epoch": 0.004692585714570978,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7181,
|
|
"mean_token_accuracy": 0.08306673839688301,
|
|
"num_tokens": 270672.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.986516904830933,
|
|
"epoch": 0.004792427963817169,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.5703,
|
|
"mean_token_accuracy": 0.09606460258364677,
|
|
"num_tokens": 275971.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.076891374588012,
|
|
"epoch": 0.00489227021306336,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.6701,
|
|
"mean_token_accuracy": 0.09255184978246689,
|
|
"num_tokens": 281554.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.021392154693604,
|
|
"epoch": 0.004992112462309551,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.6684,
|
|
"mean_token_accuracy": 0.09083683714270592,
|
|
"num_tokens": 287733.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.066862726211548,
|
|
"epoch": 0.0050919547115557415,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.6995,
|
|
"mean_token_accuracy": 0.08480059802532196,
|
|
"num_tokens": 293958.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.096820306777953,
|
|
"epoch": 0.005191796960801933,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.6813,
|
|
"mean_token_accuracy": 0.08337552659213543,
|
|
"num_tokens": 299615.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.020620632171632,
|
|
"epoch": 0.005291639210048124,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.6583,
|
|
"mean_token_accuracy": 0.09328315630555153,
|
|
"num_tokens": 305401.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.00606279373169,
|
|
"epoch": 0.005391481459294315,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.5975,
|
|
"mean_token_accuracy": 0.09268495440483093,
|
|
"num_tokens": 310420.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.975163745880127,
|
|
"epoch": 0.005491323708540506,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6291,
|
|
"mean_token_accuracy": 0.08745413720607757,
|
|
"num_tokens": 315937.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.931376361846924,
|
|
"epoch": 0.005591165957786697,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.6787,
|
|
"mean_token_accuracy": 0.09163259714841843,
|
|
"num_tokens": 321055.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.027003908157349,
|
|
"epoch": 0.005691008207032888,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.5598,
|
|
"mean_token_accuracy": 0.09943108633160591,
|
|
"num_tokens": 327446.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.895185947418213,
|
|
"epoch": 0.005790850456279079,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6409,
|
|
"mean_token_accuracy": 0.09412241280078888,
|
|
"num_tokens": 333363.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.069699239730834,
|
|
"epoch": 0.0058906927055252705,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.6736,
|
|
"mean_token_accuracy": 0.09531420394778252,
|
|
"num_tokens": 339550.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.054055500030518,
|
|
"epoch": 0.005990534954771461,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.7013,
|
|
"mean_token_accuracy": 0.08208880387246609,
|
|
"num_tokens": 345472.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.895511817932129,
|
|
"epoch": 0.006090377204017652,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6018,
|
|
"mean_token_accuracy": 0.08789039179682731,
|
|
"num_tokens": 350984.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.9680193901062015,
|
|
"epoch": 0.006190219453263843,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.5038,
|
|
"mean_token_accuracy": 0.09207511395215988,
|
|
"num_tokens": 356054.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.946393775939941,
|
|
"epoch": 0.006290061702510034,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6509,
|
|
"mean_token_accuracy": 0.1025794766843319,
|
|
"num_tokens": 361985.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.917030382156372,
|
|
"epoch": 0.006389903951756225,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.6035,
|
|
"mean_token_accuracy": 0.0876886785030365,
|
|
"num_tokens": 367784.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.934617376327514,
|
|
"epoch": 0.006489746201002416,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.5577,
|
|
"mean_token_accuracy": 0.09602945894002915,
|
|
"num_tokens": 374012.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.939635705947876,
|
|
"epoch": 0.006589588450248607,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6332,
|
|
"mean_token_accuracy": 0.08928873538970947,
|
|
"num_tokens": 379623.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.943338441848755,
|
|
"epoch": 0.006689430699494799,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.5892,
|
|
"mean_token_accuracy": 0.09149550125002862,
|
|
"num_tokens": 384982.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.994742679595947,
|
|
"epoch": 0.006789272948740989,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.6425,
|
|
"mean_token_accuracy": 0.08843275494873523,
|
|
"num_tokens": 391107.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.823350954055786,
|
|
"epoch": 0.00688911519798718,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.509,
|
|
"mean_token_accuracy": 0.09060701727867126,
|
|
"num_tokens": 396516.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.931709098815918,
|
|
"epoch": 0.0069889574472333715,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.6268,
|
|
"mean_token_accuracy": 0.08907039240002632,
|
|
"num_tokens": 403216.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.828972816467285,
|
|
"epoch": 0.007088799696479562,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5647,
|
|
"mean_token_accuracy": 0.09698711633682251,
|
|
"num_tokens": 409757.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.901078701019287,
|
|
"epoch": 0.007188641945725754,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.5776,
|
|
"mean_token_accuracy": 0.09460580721497536,
|
|
"num_tokens": 415071.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.830478239059448,
|
|
"epoch": 0.007288484194971944,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.5183,
|
|
"mean_token_accuracy": 0.09679247066378593,
|
|
"num_tokens": 420660.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.885542583465576,
|
|
"epoch": 0.007388326444218135,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.5054,
|
|
"mean_token_accuracy": 0.09666672423481941,
|
|
"num_tokens": 427219.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.882419490814209,
|
|
"epoch": 0.007488168693464327,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.4902,
|
|
"mean_token_accuracy": 0.0948188878595829,
|
|
"num_tokens": 432938.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.615965557098389,
|
|
"epoch": 0.007588010942710517,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.4774,
|
|
"mean_token_accuracy": 0.10164310932159423,
|
|
"num_tokens": 438223.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.930799531936645,
|
|
"epoch": 0.007687853191956708,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.5398,
|
|
"mean_token_accuracy": 0.09940317496657372,
|
|
"num_tokens": 444432.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.903547239303589,
|
|
"epoch": 0.0077876954412029,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5144,
|
|
"mean_token_accuracy": 0.08792076036334037,
|
|
"num_tokens": 450497.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.71635160446167,
|
|
"epoch": 0.00788753769044909,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5483,
|
|
"mean_token_accuracy": 0.09417223706841468,
|
|
"num_tokens": 456358.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.832306432723999,
|
|
"epoch": 0.007987379939695281,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.4283,
|
|
"mean_token_accuracy": 0.09271656051278114,
|
|
"num_tokens": 461638.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.71069369316101,
|
|
"epoch": 0.008087222188941472,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.2931,
|
|
"mean_token_accuracy": 0.10623703300952911,
|
|
"num_tokens": 466383.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.75597825050354,
|
|
"epoch": 0.008187064438187664,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5048,
|
|
"mean_token_accuracy": 0.09718195348978043,
|
|
"num_tokens": 472325.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.676103734970093,
|
|
"epoch": 0.008286906687433855,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.3461,
|
|
"mean_token_accuracy": 0.09879515469074249,
|
|
"num_tokens": 477810.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.821485471725464,
|
|
"epoch": 0.008386748936680045,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.4209,
|
|
"mean_token_accuracy": 0.09962461516261101,
|
|
"num_tokens": 483375.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.718852472305298,
|
|
"epoch": 0.008486591185926236,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.4184,
|
|
"mean_token_accuracy": 0.09406220316886901,
|
|
"num_tokens": 489826.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.717019510269165,
|
|
"epoch": 0.008586433435172427,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.3222,
|
|
"mean_token_accuracy": 0.10805956795811653,
|
|
"num_tokens": 494797.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.657623624801635,
|
|
"epoch": 0.00868627568441862,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.3715,
|
|
"mean_token_accuracy": 0.11027861759066582,
|
|
"num_tokens": 499868.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.766957855224609,
|
|
"epoch": 0.00878611793366481,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.489,
|
|
"mean_token_accuracy": 0.09478027820587158,
|
|
"num_tokens": 505777.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.825216388702392,
|
|
"epoch": 0.008885960182911,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.09458905309438706,
|
|
"num_tokens": 511404.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.681342935562133,
|
|
"epoch": 0.008985802432157191,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.3926,
|
|
"mean_token_accuracy": 0.10864230245351791,
|
|
"num_tokens": 517255.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.636969661712646,
|
|
"epoch": 0.009085644681403382,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.418,
|
|
"mean_token_accuracy": 0.09724331274628639,
|
|
"num_tokens": 523299.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.716694116592407,
|
|
"epoch": 0.009185486930649574,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.4159,
|
|
"mean_token_accuracy": 0.09504513517022133,
|
|
"num_tokens": 528700.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.704607963562012,
|
|
"epoch": 0.009285329179895765,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.3906,
|
|
"mean_token_accuracy": 0.10305075645446778,
|
|
"num_tokens": 534456.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.687975072860718,
|
|
"epoch": 0.009385171429141956,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.392,
|
|
"mean_token_accuracy": 0.09684277400374412,
|
|
"num_tokens": 541492.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.614287185668945,
|
|
"epoch": 0.009485013678388146,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.4688,
|
|
"mean_token_accuracy": 0.09323694482445717,
|
|
"num_tokens": 547126.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.748909521102905,
|
|
"epoch": 0.009584855927634337,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.4187,
|
|
"mean_token_accuracy": 0.09110049903392792,
|
|
"num_tokens": 553522.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.599171686172485,
|
|
"epoch": 0.00968469817688053,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.3032,
|
|
"mean_token_accuracy": 0.09761992767453194,
|
|
"num_tokens": 559487.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.581656837463379,
|
|
"epoch": 0.00978454042612672,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.3578,
|
|
"mean_token_accuracy": 0.09209932833909988,
|
|
"num_tokens": 565483.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.675942230224609,
|
|
"epoch": 0.009884382675372911,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.3958,
|
|
"mean_token_accuracy": 0.1011126957833767,
|
|
"num_tokens": 571609.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.702158212661743,
|
|
"epoch": 0.009984224924619102,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.3743,
|
|
"mean_token_accuracy": 0.09817478843033314,
|
|
"num_tokens": 578310.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.673187351226806,
|
|
"epoch": 0.010084067173865292,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.3739,
|
|
"mean_token_accuracy": 0.10150901302695274,
|
|
"num_tokens": 584288.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.5667647361755375,
|
|
"epoch": 0.010183909423111483,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.3503,
|
|
"mean_token_accuracy": 0.10190780833363533,
|
|
"num_tokens": 590002.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.712408590316772,
|
|
"epoch": 0.010283751672357675,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.3543,
|
|
"mean_token_accuracy": 0.09027517363429069,
|
|
"num_tokens": 595376.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.502785348892212,
|
|
"epoch": 0.010383593921603866,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.3149,
|
|
"mean_token_accuracy": 0.09311652481555939,
|
|
"num_tokens": 601190.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.7128880500793455,
|
|
"epoch": 0.010483436170850057,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.4182,
|
|
"mean_token_accuracy": 0.08943627402186394,
|
|
"num_tokens": 606976.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.492193746566772,
|
|
"epoch": 0.010583278420096248,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.3876,
|
|
"mean_token_accuracy": 0.0967042751610279,
|
|
"num_tokens": 613859.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.769742202758789,
|
|
"epoch": 0.010683120669342438,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.3351,
|
|
"mean_token_accuracy": 0.10183999463915824,
|
|
"num_tokens": 620154.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.589622163772583,
|
|
"epoch": 0.01078296291858863,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.4284,
|
|
"mean_token_accuracy": 0.09763827249407768,
|
|
"num_tokens": 626088.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.531725645065308,
|
|
"epoch": 0.010882805167834821,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.2637,
|
|
"mean_token_accuracy": 0.09581549614667892,
|
|
"num_tokens": 631502.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.610470390319824,
|
|
"epoch": 0.010982647417081012,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.3997,
|
|
"mean_token_accuracy": 0.09115200191736221,
|
|
"num_tokens": 638113.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.597003698348999,
|
|
"epoch": 0.011082489666327203,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.4932,
|
|
"mean_token_accuracy": 0.09354109838604927,
|
|
"num_tokens": 644739.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.561251831054688,
|
|
"epoch": 0.011182331915573393,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.2814,
|
|
"mean_token_accuracy": 0.09907945692539215,
|
|
"num_tokens": 650488.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.5941314697265625,
|
|
"epoch": 0.011282174164819586,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.346,
|
|
"mean_token_accuracy": 0.10458883121609688,
|
|
"num_tokens": 656613.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.585415840148926,
|
|
"epoch": 0.011382016414065776,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.2214,
|
|
"mean_token_accuracy": 0.1068822205066681,
|
|
"num_tokens": 662377.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.463992738723755,
|
|
"epoch": 0.011481858663311967,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.3178,
|
|
"mean_token_accuracy": 0.09922432601451874,
|
|
"num_tokens": 668590.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.510423278808593,
|
|
"epoch": 0.011581700912558158,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.3292,
|
|
"mean_token_accuracy": 0.10266099199652672,
|
|
"num_tokens": 674125.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.558202886581421,
|
|
"epoch": 0.011681543161804349,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.2624,
|
|
"mean_token_accuracy": 0.10101615637540817,
|
|
"num_tokens": 680170.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.5202680110931395,
|
|
"epoch": 0.011781385411050541,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.3084,
|
|
"mean_token_accuracy": 0.09623364508152008,
|
|
"num_tokens": 685611.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.497081136703491,
|
|
"epoch": 0.011881227660296732,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.3122,
|
|
"mean_token_accuracy": 0.10387674495577812,
|
|
"num_tokens": 691755.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.541520500183106,
|
|
"epoch": 0.011981069909542922,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.2026,
|
|
"mean_token_accuracy": 0.09760682582855225,
|
|
"num_tokens": 697370.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.482969570159912,
|
|
"epoch": 0.012080912158789113,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.4197,
|
|
"mean_token_accuracy": 0.09226728752255439,
|
|
"num_tokens": 703872.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.466841697692871,
|
|
"epoch": 0.012180754408035304,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.2546,
|
|
"mean_token_accuracy": 0.10292266681790352,
|
|
"num_tokens": 709217.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.397413873672486,
|
|
"epoch": 0.012280596657281494,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.1903,
|
|
"mean_token_accuracy": 0.10792005062103271,
|
|
"num_tokens": 714644.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.464698457717896,
|
|
"epoch": 0.012380438906527687,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.1951,
|
|
"mean_token_accuracy": 0.10397047251462936,
|
|
"num_tokens": 720101.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.475820302963257,
|
|
"epoch": 0.012480281155773877,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.2862,
|
|
"mean_token_accuracy": 0.09931057021021843,
|
|
"num_tokens": 726456.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.414108514785767,
|
|
"epoch": 0.012580123405020068,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.1976,
|
|
"mean_token_accuracy": 0.10596151649951935,
|
|
"num_tokens": 732134.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.464668941497803,
|
|
"epoch": 0.012679965654266259,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.1947,
|
|
"mean_token_accuracy": 0.10172332227230071,
|
|
"num_tokens": 737709.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.350508165359497,
|
|
"epoch": 0.01277980790351245,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.3682,
|
|
"mean_token_accuracy": 0.09781662225723267,
|
|
"num_tokens": 743072.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.463224124908447,
|
|
"epoch": 0.012879650152758642,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.1883,
|
|
"mean_token_accuracy": 0.10707319229841232,
|
|
"num_tokens": 749535.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.501761102676392,
|
|
"epoch": 0.012979492402004833,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.2733,
|
|
"mean_token_accuracy": 0.10395639017224312,
|
|
"num_tokens": 755229.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.386087274551391,
|
|
"epoch": 0.013079334651251023,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.2134,
|
|
"mean_token_accuracy": 0.10286233350634574,
|
|
"num_tokens": 760848.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.381467628479004,
|
|
"epoch": 0.013179176900497214,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.269,
|
|
"mean_token_accuracy": 0.10547955185174943,
|
|
"num_tokens": 766124.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.397349739074707,
|
|
"epoch": 0.013279019149743405,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.1458,
|
|
"mean_token_accuracy": 0.10455328896641732,
|
|
"num_tokens": 771397.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.368815183639526,
|
|
"epoch": 0.013378861398989597,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.1752,
|
|
"mean_token_accuracy": 0.11119461134076118,
|
|
"num_tokens": 776925.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.324629020690918,
|
|
"epoch": 0.013478703648235788,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1141,
|
|
"mean_token_accuracy": 0.10647935196757316,
|
|
"num_tokens": 782273.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.373594522476196,
|
|
"epoch": 0.013578545897481979,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.1333,
|
|
"mean_token_accuracy": 0.10900317803025246,
|
|
"num_tokens": 787801.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.4068872928619385,
|
|
"epoch": 0.01367838814672817,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.1684,
|
|
"mean_token_accuracy": 0.10694136023521424,
|
|
"num_tokens": 793217.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.324366044998169,
|
|
"epoch": 0.01377823039597436,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.195,
|
|
"mean_token_accuracy": 0.10198777243494987,
|
|
"num_tokens": 799565.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.493931007385254,
|
|
"epoch": 0.013878072645220552,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.3115,
|
|
"mean_token_accuracy": 0.09703441336750984,
|
|
"num_tokens": 805521.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.291825580596924,
|
|
"epoch": 0.013977914894466743,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.2683,
|
|
"mean_token_accuracy": 0.10237760916352272,
|
|
"num_tokens": 811990.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.361256313323975,
|
|
"epoch": 0.014077757143712934,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.2956,
|
|
"mean_token_accuracy": 0.09967261478304863,
|
|
"num_tokens": 818726.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.394858312606812,
|
|
"epoch": 0.014177599392959124,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.2031,
|
|
"mean_token_accuracy": 0.10854720175266266,
|
|
"num_tokens": 824910.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.382597780227661,
|
|
"epoch": 0.014277441642205315,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.1747,
|
|
"mean_token_accuracy": 0.10250596627593041,
|
|
"num_tokens": 830344.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.238950634002686,
|
|
"epoch": 0.014377283891451507,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.1087,
|
|
"mean_token_accuracy": 0.11101329773664474,
|
|
"num_tokens": 836032.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.332369232177735,
|
|
"epoch": 0.014477126140697698,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.2624,
|
|
"mean_token_accuracy": 0.10453605726361274,
|
|
"num_tokens": 841585.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.365637540817261,
|
|
"epoch": 0.014576968389943889,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.229,
|
|
"mean_token_accuracy": 0.10223104059696198,
|
|
"num_tokens": 847945.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.209516954421997,
|
|
"epoch": 0.01467681063919008,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.2036,
|
|
"mean_token_accuracy": 0.10675105154514312,
|
|
"num_tokens": 853436.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.330843734741211,
|
|
"epoch": 0.01477665288843627,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.1663,
|
|
"mean_token_accuracy": 0.10609268397092819,
|
|
"num_tokens": 859119.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.330539321899414,
|
|
"epoch": 0.014876495137682461,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.2006,
|
|
"mean_token_accuracy": 0.10883735790848732,
|
|
"num_tokens": 865207.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.321112060546875,
|
|
"epoch": 0.014976337386928653,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.1784,
|
|
"mean_token_accuracy": 0.10204658731818199,
|
|
"num_tokens": 870446.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.339653825759887,
|
|
"epoch": 0.015076179636174844,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.0788,
|
|
"mean_token_accuracy": 0.11148179769515991,
|
|
"num_tokens": 875640.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.23172779083252,
|
|
"epoch": 0.015176021885421035,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.0821,
|
|
"mean_token_accuracy": 0.11134838536381722,
|
|
"num_tokens": 880803.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.295123052597046,
|
|
"epoch": 0.015275864134667225,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.248,
|
|
"mean_token_accuracy": 0.10265103802084923,
|
|
"num_tokens": 886584.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.201002311706543,
|
|
"epoch": 0.015375706383913416,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.1926,
|
|
"mean_token_accuracy": 0.10653134509921074,
|
|
"num_tokens": 892765.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.242213916778565,
|
|
"epoch": 0.015475548633159609,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.1753,
|
|
"mean_token_accuracy": 0.11137882471084595,
|
|
"num_tokens": 899505.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.3361810684204105,
|
|
"epoch": 0.0155753908824058,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.1253,
|
|
"mean_token_accuracy": 0.11177048087120056,
|
|
"num_tokens": 905459.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.215915870666504,
|
|
"epoch": 0.015675233131651988,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.1896,
|
|
"mean_token_accuracy": 0.1064237505197525,
|
|
"num_tokens": 910981.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.24966778755188,
|
|
"epoch": 0.01577507538089818,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.1012,
|
|
"mean_token_accuracy": 0.10888548716902732,
|
|
"num_tokens": 916455.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.214824104309082,
|
|
"epoch": 0.015874917630144373,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.9919,
|
|
"mean_token_accuracy": 0.11557956710457802,
|
|
"num_tokens": 921216.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.192123794555664,
|
|
"epoch": 0.015974759879390562,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.0767,
|
|
"mean_token_accuracy": 0.10509335845708848,
|
|
"num_tokens": 926908.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.1781196117401125,
|
|
"epoch": 0.016074602128636754,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.0638,
|
|
"mean_token_accuracy": 0.11034815311431885,
|
|
"num_tokens": 932243.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.189894533157348,
|
|
"epoch": 0.016174444377882943,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.1698,
|
|
"mean_token_accuracy": 0.1015389807522297,
|
|
"num_tokens": 937635.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.258892917633057,
|
|
"epoch": 0.016274286627129136,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.1481,
|
|
"mean_token_accuracy": 0.10326545163989068,
|
|
"num_tokens": 943366.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.294253206253051,
|
|
"epoch": 0.016374128876375328,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.0561,
|
|
"mean_token_accuracy": 0.10886959284543991,
|
|
"num_tokens": 948938.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.266506814956665,
|
|
"epoch": 0.016473971125621517,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.2073,
|
|
"mean_token_accuracy": 0.10875980257987976,
|
|
"num_tokens": 955141.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.174079132080078,
|
|
"epoch": 0.01657381337486771,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.1919,
|
|
"mean_token_accuracy": 0.1085533007979393,
|
|
"num_tokens": 960700.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.298240995407104,
|
|
"epoch": 0.0166736556241139,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.0432,
|
|
"mean_token_accuracy": 0.10635742172598839,
|
|
"num_tokens": 966048.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.215938520431519,
|
|
"epoch": 0.01677349787336009,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.0501,
|
|
"mean_token_accuracy": 0.11426909789443015,
|
|
"num_tokens": 972093.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.105905294418335,
|
|
"epoch": 0.016873340122606283,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.1019,
|
|
"mean_token_accuracy": 0.11449455171823501,
|
|
"num_tokens": 978008.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.254274368286133,
|
|
"epoch": 0.016973182371852472,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.0798,
|
|
"mean_token_accuracy": 0.10710402801632882,
|
|
"num_tokens": 983390.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.21899790763855,
|
|
"epoch": 0.017073024621098665,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.2222,
|
|
"mean_token_accuracy": 0.10271771252155304,
|
|
"num_tokens": 989131.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.1298003673553465,
|
|
"epoch": 0.017172866870344854,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.09,
|
|
"mean_token_accuracy": 0.11098077148199081,
|
|
"num_tokens": 994875.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.113921117782593,
|
|
"epoch": 0.017272709119591046,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.0757,
|
|
"mean_token_accuracy": 0.10846130549907684,
|
|
"num_tokens": 1001028.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.222342777252197,
|
|
"epoch": 0.01737255136883724,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.1236,
|
|
"mean_token_accuracy": 0.11109541580080987,
|
|
"num_tokens": 1006835.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.119646120071411,
|
|
"epoch": 0.017472393618083427,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.1362,
|
|
"mean_token_accuracy": 0.10868640765547752,
|
|
"num_tokens": 1013966.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.367539119720459,
|
|
"epoch": 0.01757223586732962,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.055,
|
|
"mean_token_accuracy": 0.11352950558066369,
|
|
"num_tokens": 1019199.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.148986005783081,
|
|
"epoch": 0.01767207811657581,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.013,
|
|
"mean_token_accuracy": 0.11290877237915993,
|
|
"num_tokens": 1024820.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.0365533351898195,
|
|
"epoch": 0.017771920365822,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.1368,
|
|
"mean_token_accuracy": 0.10875423103570939,
|
|
"num_tokens": 1030580.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.245150136947632,
|
|
"epoch": 0.017871762615068194,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.058,
|
|
"mean_token_accuracy": 0.11648992523550987,
|
|
"num_tokens": 1036113.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.180494165420532,
|
|
"epoch": 0.017971604864314383,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.0761,
|
|
"mean_token_accuracy": 0.11047818586230278,
|
|
"num_tokens": 1042393.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.1404953956604,
|
|
"epoch": 0.018071447113560575,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.0767,
|
|
"mean_token_accuracy": 0.10953986048698425,
|
|
"num_tokens": 1048946.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.169243335723877,
|
|
"epoch": 0.018171289362806764,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.0804,
|
|
"mean_token_accuracy": 0.10796508491039276,
|
|
"num_tokens": 1055382.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.1037177562713625,
|
|
"epoch": 0.018271131612052956,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.0204,
|
|
"mean_token_accuracy": 0.118329706043005,
|
|
"num_tokens": 1060938.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.105455350875855,
|
|
"epoch": 0.01837097386129915,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.1646,
|
|
"mean_token_accuracy": 0.10304270908236504,
|
|
"num_tokens": 1068009.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.208564186096192,
|
|
"epoch": 0.018470816110545338,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.0674,
|
|
"mean_token_accuracy": 0.11573286652565003,
|
|
"num_tokens": 1074609.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.129713106155395,
|
|
"epoch": 0.01857065835979153,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.1146,
|
|
"mean_token_accuracy": 0.11269782558083534,
|
|
"num_tokens": 1080711.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.081141948699951,
|
|
"epoch": 0.01867050060903772,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.9499,
|
|
"mean_token_accuracy": 0.1222643569111824,
|
|
"num_tokens": 1085785.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.1242677688598635,
|
|
"epoch": 0.01877034285828391,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.1394,
|
|
"mean_token_accuracy": 0.09848750457167625,
|
|
"num_tokens": 1091394.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.117886972427368,
|
|
"epoch": 0.018870185107530104,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.0096,
|
|
"mean_token_accuracy": 0.11437401026487351,
|
|
"num_tokens": 1096977.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.156701135635376,
|
|
"epoch": 0.018970027356776293,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.166,
|
|
"mean_token_accuracy": 0.10502217188477517,
|
|
"num_tokens": 1102891.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.151956176757812,
|
|
"epoch": 0.019069869606022485,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.1109,
|
|
"mean_token_accuracy": 0.10888380631804466,
|
|
"num_tokens": 1108865.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.112528324127197,
|
|
"epoch": 0.019169711855268674,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.052,
|
|
"mean_token_accuracy": 0.11442630961537362,
|
|
"num_tokens": 1115099.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.104516553878784,
|
|
"epoch": 0.019269554104514867,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.0321,
|
|
"mean_token_accuracy": 0.11025213748216629,
|
|
"num_tokens": 1120141.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.142838478088379,
|
|
"epoch": 0.01936939635376106,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.0369,
|
|
"mean_token_accuracy": 0.11430102586746216,
|
|
"num_tokens": 1125047.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.072331666946411,
|
|
"epoch": 0.019469238603007248,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.0064,
|
|
"mean_token_accuracy": 0.10542083233594894,
|
|
"num_tokens": 1131149.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.159235239028931,
|
|
"epoch": 0.01956908085225344,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.0413,
|
|
"mean_token_accuracy": 0.1097842164337635,
|
|
"num_tokens": 1137105.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.977080297470093,
|
|
"epoch": 0.01966892310149963,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.0466,
|
|
"mean_token_accuracy": 0.10660657659173012,
|
|
"num_tokens": 1143058.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.2410801410675045,
|
|
"epoch": 0.019768765350745822,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.0958,
|
|
"mean_token_accuracy": 0.10574808344244957,
|
|
"num_tokens": 1148708.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.155167055130005,
|
|
"epoch": 0.019868607599992014,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.1245,
|
|
"mean_token_accuracy": 0.10252380073070526,
|
|
"num_tokens": 1154728.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.1254583358764645,
|
|
"epoch": 0.019968449849238203,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.0231,
|
|
"mean_token_accuracy": 0.11203286051750183,
|
|
"num_tokens": 1160349.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.067977476119995,
|
|
"epoch": 0.020068292098484396,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.0819,
|
|
"mean_token_accuracy": 0.10922700464725495,
|
|
"num_tokens": 1166215.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.055725526809693,
|
|
"epoch": 0.020168134347730585,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.9405,
|
|
"mean_token_accuracy": 0.11441097259521485,
|
|
"num_tokens": 1171763.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.043396949768066,
|
|
"epoch": 0.020267976596976777,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.9167,
|
|
"mean_token_accuracy": 0.12178001627326011,
|
|
"num_tokens": 1177550.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.996433687210083,
|
|
"epoch": 0.020367818846222966,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.9145,
|
|
"mean_token_accuracy": 0.12212632149457932,
|
|
"num_tokens": 1182946.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.967729949951172,
|
|
"epoch": 0.02046766109546916,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.9436,
|
|
"mean_token_accuracy": 0.11037085354328155,
|
|
"num_tokens": 1188988.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.121986532211304,
|
|
"epoch": 0.02056750334471535,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.9991,
|
|
"mean_token_accuracy": 0.11994724944233895,
|
|
"num_tokens": 1194829.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.002535772323609,
|
|
"epoch": 0.02066734559396154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.9548,
|
|
"mean_token_accuracy": 0.11914273425936699,
|
|
"num_tokens": 1200179.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.036171579360962,
|
|
"epoch": 0.020767187843207732,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.0104,
|
|
"mean_token_accuracy": 0.11094862222671509,
|
|
"num_tokens": 1206086.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.199465036392212,
|
|
"epoch": 0.02086703009245392,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.0701,
|
|
"mean_token_accuracy": 0.10852830782532692,
|
|
"num_tokens": 1211921.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.952513790130615,
|
|
"epoch": 0.020966872341700114,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.9844,
|
|
"mean_token_accuracy": 0.12001401036977768,
|
|
"num_tokens": 1217389.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.009827756881714,
|
|
"epoch": 0.021066714590946306,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.8613,
|
|
"mean_token_accuracy": 0.12178552374243737,
|
|
"num_tokens": 1222740.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.95297474861145,
|
|
"epoch": 0.021166556840192495,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.9935,
|
|
"mean_token_accuracy": 0.1173486851155758,
|
|
"num_tokens": 1228281.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.151493835449219,
|
|
"epoch": 0.021266399089438687,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.0623,
|
|
"mean_token_accuracy": 0.10887223184108734,
|
|
"num_tokens": 1234284.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.047469758987427,
|
|
"epoch": 0.021366241338684876,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 7.0,
|
|
"mean_token_accuracy": 0.11805735900998116,
|
|
"num_tokens": 1239803.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.092536211013794,
|
|
"epoch": 0.02146608358793107,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.9886,
|
|
"mean_token_accuracy": 0.11481416448950768,
|
|
"num_tokens": 1245884.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.958631992340088,
|
|
"epoch": 0.02156592583717726,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.0566,
|
|
"mean_token_accuracy": 0.10592376589775085,
|
|
"num_tokens": 1252350.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.141324663162232,
|
|
"epoch": 0.02166576808642345,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.8914,
|
|
"mean_token_accuracy": 0.11976096481084823,
|
|
"num_tokens": 1257661.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.862487268447876,
|
|
"epoch": 0.021765610335669643,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.8798,
|
|
"mean_token_accuracy": 0.11734078601002693,
|
|
"num_tokens": 1262879.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.029018068313599,
|
|
"epoch": 0.02186545258491583,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.8818,
|
|
"mean_token_accuracy": 0.11813108623027802,
|
|
"num_tokens": 1268791.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.932131385803222,
|
|
"epoch": 0.021965294834162024,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.9512,
|
|
"mean_token_accuracy": 0.12142582163214684,
|
|
"num_tokens": 1274390.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.056042861938477,
|
|
"epoch": 0.022065137083408216,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.9611,
|
|
"mean_token_accuracy": 0.11118568778038025,
|
|
"num_tokens": 1281408.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.9724180698394775,
|
|
"epoch": 0.022164979332654405,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.9057,
|
|
"mean_token_accuracy": 0.11218234598636627,
|
|
"num_tokens": 1286603.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.921506690979004,
|
|
"epoch": 0.022264821581900598,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.9586,
|
|
"mean_token_accuracy": 0.11624157801270485,
|
|
"num_tokens": 1292709.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.959320068359375,
|
|
"epoch": 0.022364663831146787,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.7576,
|
|
"mean_token_accuracy": 0.12267198339104653,
|
|
"num_tokens": 1298542.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.927985811233521,
|
|
"epoch": 0.02246450608039298,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.9363,
|
|
"mean_token_accuracy": 0.11307285130023956,
|
|
"num_tokens": 1304545.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.0036454677581785,
|
|
"epoch": 0.02256434832963917,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.92,
|
|
"mean_token_accuracy": 0.11687800735235214,
|
|
"num_tokens": 1310751.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.875037097930909,
|
|
"epoch": 0.02266419057888536,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.8757,
|
|
"mean_token_accuracy": 0.12439063787460328,
|
|
"num_tokens": 1316473.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.030423259735107,
|
|
"epoch": 0.022764032828131553,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.9127,
|
|
"mean_token_accuracy": 0.12259516194462776,
|
|
"num_tokens": 1322453.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.999492931365967,
|
|
"epoch": 0.022863875077377742,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.9618,
|
|
"mean_token_accuracy": 0.11632513478398324,
|
|
"num_tokens": 1328206.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.076742887496948,
|
|
"epoch": 0.022963717326623934,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.9603,
|
|
"mean_token_accuracy": 0.11790076270699501,
|
|
"num_tokens": 1334287.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.867577743530274,
|
|
"epoch": 0.023063559575870127,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.8576,
|
|
"mean_token_accuracy": 0.11696006879210472,
|
|
"num_tokens": 1340209.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.0559654712677,
|
|
"epoch": 0.023163401825116316,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 7.0352,
|
|
"mean_token_accuracy": 0.11193358227610588,
|
|
"num_tokens": 1346222.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.895693016052246,
|
|
"epoch": 0.023263244074362508,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.8635,
|
|
"mean_token_accuracy": 0.11950421705842018,
|
|
"num_tokens": 1351674.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.051764965057373,
|
|
"epoch": 0.023363086323608697,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.9207,
|
|
"mean_token_accuracy": 0.1170286051928997,
|
|
"num_tokens": 1358020.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.953119373321533,
|
|
"epoch": 0.02346292857285489,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.9599,
|
|
"mean_token_accuracy": 0.11382025703787804,
|
|
"num_tokens": 1363595.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.040049028396607,
|
|
"epoch": 0.023562770822101082,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.9561,
|
|
"mean_token_accuracy": 0.126374813914299,
|
|
"num_tokens": 1368850.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.871081209182739,
|
|
"epoch": 0.02366261307134727,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.8119,
|
|
"mean_token_accuracy": 0.12303650677204132,
|
|
"num_tokens": 1374487.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.937151670455933,
|
|
"epoch": 0.023762455320593463,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.7714,
|
|
"mean_token_accuracy": 0.12604868933558463,
|
|
"num_tokens": 1379702.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.927495336532592,
|
|
"epoch": 0.023862297569839652,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.9194,
|
|
"mean_token_accuracy": 0.11645317524671554,
|
|
"num_tokens": 1385478.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.912464380264282,
|
|
"epoch": 0.023962139819085845,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.8441,
|
|
"mean_token_accuracy": 0.12719617113471032,
|
|
"num_tokens": 1391269.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.831350517272949,
|
|
"epoch": 0.024061982068332037,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.7614,
|
|
"mean_token_accuracy": 0.1196099117398262,
|
|
"num_tokens": 1396238.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.0208110332489015,
|
|
"epoch": 0.024161824317578226,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.8398,
|
|
"mean_token_accuracy": 0.12437132149934768,
|
|
"num_tokens": 1401755.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.01935043334961,
|
|
"epoch": 0.02426166656682442,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.9103,
|
|
"mean_token_accuracy": 0.11659247428178787,
|
|
"num_tokens": 1408360.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.974241542816162,
|
|
"epoch": 0.024361508816070607,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.9215,
|
|
"mean_token_accuracy": 0.11295409947633743,
|
|
"num_tokens": 1414050.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.985059976577759,
|
|
"epoch": 0.0244613510653168,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.9197,
|
|
"mean_token_accuracy": 0.11652319058775902,
|
|
"num_tokens": 1419618.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.8160017967224125,
|
|
"epoch": 0.02456119331456299,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.8269,
|
|
"mean_token_accuracy": 0.12137463390827179,
|
|
"num_tokens": 1425457.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.10415620803833,
|
|
"epoch": 0.02466103556380918,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.93,
|
|
"mean_token_accuracy": 0.11103903874754906,
|
|
"num_tokens": 1431346.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.850605487823486,
|
|
"epoch": 0.024760877813055374,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.7673,
|
|
"mean_token_accuracy": 0.13000494986772537,
|
|
"num_tokens": 1436487.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.948511695861816,
|
|
"epoch": 0.024860720062301563,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.8722,
|
|
"mean_token_accuracy": 0.11711608618497849,
|
|
"num_tokens": 1441770.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.966010427474975,
|
|
"epoch": 0.024960562311547755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.8854,
|
|
"mean_token_accuracy": 0.11426938027143478,
|
|
"num_tokens": 1447728.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.901544523239136,
|
|
"epoch": 0.025060404560793944,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.11754528582096099,
|
|
"num_tokens": 1453973.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.889706373214722,
|
|
"epoch": 0.025160246810040136,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.8752,
|
|
"mean_token_accuracy": 0.12020720615983009,
|
|
"num_tokens": 1459010.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.882485342025757,
|
|
"epoch": 0.02526008905928633,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.7915,
|
|
"mean_token_accuracy": 0.122784835845232,
|
|
"num_tokens": 1464451.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.974827814102173,
|
|
"epoch": 0.025359931308532518,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.8644,
|
|
"mean_token_accuracy": 0.11677931845188141,
|
|
"num_tokens": 1470606.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.937281560897827,
|
|
"epoch": 0.02545977355777871,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.9025,
|
|
"mean_token_accuracy": 0.12493098676204681,
|
|
"num_tokens": 1477131.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.765360355377197,
|
|
"epoch": 0.0255596158070249,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.728,
|
|
"mean_token_accuracy": 0.12518197447061538,
|
|
"num_tokens": 1482818.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.07388858795166,
|
|
"epoch": 0.02565945805627109,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.923,
|
|
"mean_token_accuracy": 0.11540115773677825,
|
|
"num_tokens": 1489302.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.910632991790772,
|
|
"epoch": 0.025759300305517284,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.8646,
|
|
"mean_token_accuracy": 0.1154491625726223,
|
|
"num_tokens": 1495380.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.0754852294921875,
|
|
"epoch": 0.025859142554763473,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.9706,
|
|
"mean_token_accuracy": 0.11140237599611283,
|
|
"num_tokens": 1501422.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.860459280014038,
|
|
"epoch": 0.025958984804009665,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.8683,
|
|
"mean_token_accuracy": 0.11450280100107194,
|
|
"num_tokens": 1506752.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.941881847381592,
|
|
"epoch": 0.026058827053255854,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.7661,
|
|
"mean_token_accuracy": 0.12358115315437317,
|
|
"num_tokens": 1512080.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.822068119049073,
|
|
"epoch": 0.026158669302502047,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.732,
|
|
"mean_token_accuracy": 0.12754540219902993,
|
|
"num_tokens": 1517082.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.787048101425171,
|
|
"epoch": 0.02625851155174824,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.7048,
|
|
"mean_token_accuracy": 0.1254524379968643,
|
|
"num_tokens": 1522691.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.843017673492431,
|
|
"epoch": 0.026358353800994428,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.8477,
|
|
"mean_token_accuracy": 0.12210826799273491,
|
|
"num_tokens": 1528196.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.8819899559021,
|
|
"epoch": 0.02645819605024062,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.7849,
|
|
"mean_token_accuracy": 0.12873289808630944,
|
|
"num_tokens": 1534020.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.828749370574951,
|
|
"epoch": 0.02655803829948681,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.7272,
|
|
"mean_token_accuracy": 0.12374946102499962,
|
|
"num_tokens": 1539240.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.769867038726806,
|
|
"epoch": 0.026657880548733002,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.8215,
|
|
"mean_token_accuracy": 0.1225133903324604,
|
|
"num_tokens": 1545030.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.99039044380188,
|
|
"epoch": 0.026757722797979194,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.884,
|
|
"mean_token_accuracy": 0.12295123115181923,
|
|
"num_tokens": 1551457.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.920403480529785,
|
|
"epoch": 0.026857565047225383,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.8168,
|
|
"mean_token_accuracy": 0.1248493030667305,
|
|
"num_tokens": 1557531.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.758120441436768,
|
|
"epoch": 0.026957407296471576,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.7921,
|
|
"mean_token_accuracy": 0.12435408905148507,
|
|
"num_tokens": 1563410.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.970003747940064,
|
|
"epoch": 0.027057249545717765,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.8861,
|
|
"mean_token_accuracy": 0.12241258472204208,
|
|
"num_tokens": 1569200.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.029645252227783,
|
|
"epoch": 0.027157091794963957,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.871,
|
|
"mean_token_accuracy": 0.12175140902400017,
|
|
"num_tokens": 1575292.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.84347939491272,
|
|
"epoch": 0.02725693404421015,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.8117,
|
|
"mean_token_accuracy": 0.12237219363451005,
|
|
"num_tokens": 1580772.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.96959810256958,
|
|
"epoch": 0.02735677629345634,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.9377,
|
|
"mean_token_accuracy": 0.12287636026740074,
|
|
"num_tokens": 1587388.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.934046268463135,
|
|
"epoch": 0.02745661854270253,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.7697,
|
|
"mean_token_accuracy": 0.12688597813248634,
|
|
"num_tokens": 1592489.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.854925727844238,
|
|
"epoch": 0.02755646079194872,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.7964,
|
|
"mean_token_accuracy": 0.1234237290918827,
|
|
"num_tokens": 1597995.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.998856925964356,
|
|
"epoch": 0.027656303041194912,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.8758,
|
|
"mean_token_accuracy": 0.12157108932733536,
|
|
"num_tokens": 1604057.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.926128005981445,
|
|
"epoch": 0.027756145290441105,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.7865,
|
|
"mean_token_accuracy": 0.12133876532316208,
|
|
"num_tokens": 1609775.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.80944447517395,
|
|
"epoch": 0.027855987539687294,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.7047,
|
|
"mean_token_accuracy": 0.12691160216927527,
|
|
"num_tokens": 1616355.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.781903266906738,
|
|
"epoch": 0.027955829788933486,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.7363,
|
|
"mean_token_accuracy": 0.1296573504805565,
|
|
"num_tokens": 1621826.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.780082511901855,
|
|
"epoch": 0.028055672038179675,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.604,
|
|
"mean_token_accuracy": 0.13054399192333221,
|
|
"num_tokens": 1627481.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.84660906791687,
|
|
"epoch": 0.028155514287425867,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.7943,
|
|
"mean_token_accuracy": 0.11441523656249046,
|
|
"num_tokens": 1633448.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.810406064987182,
|
|
"epoch": 0.02825535653667206,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.5316,
|
|
"mean_token_accuracy": 0.14356368407607079,
|
|
"num_tokens": 1638706.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.692267274856567,
|
|
"epoch": 0.02835519878591825,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.6827,
|
|
"mean_token_accuracy": 0.13166252374649048,
|
|
"num_tokens": 1644029.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.722214555740356,
|
|
"epoch": 0.02845504103516444,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.8181,
|
|
"mean_token_accuracy": 0.11837060526013374,
|
|
"num_tokens": 1650592.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.028738355636596,
|
|
"epoch": 0.02855488328441063,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.7622,
|
|
"mean_token_accuracy": 0.11889392361044884,
|
|
"num_tokens": 1656083.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.883346891403198,
|
|
"epoch": 0.028654725533656823,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.7748,
|
|
"mean_token_accuracy": 0.12805232256650925,
|
|
"num_tokens": 1661508.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.671160364151001,
|
|
"epoch": 0.028754567782903015,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.6402,
|
|
"mean_token_accuracy": 0.13186832368373871,
|
|
"num_tokens": 1667010.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.975370264053344,
|
|
"epoch": 0.028854410032149204,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.8019,
|
|
"mean_token_accuracy": 0.12405418083071709,
|
|
"num_tokens": 1672458.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.866298961639404,
|
|
"epoch": 0.028954252281395396,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.7412,
|
|
"mean_token_accuracy": 0.11540611907839775,
|
|
"num_tokens": 1678803.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.807646942138672,
|
|
"epoch": 0.029054094530641585,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.7019,
|
|
"mean_token_accuracy": 0.1331303209066391,
|
|
"num_tokens": 1683794.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.956542015075684,
|
|
"epoch": 0.029153936779887778,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.12255690768361091,
|
|
"num_tokens": 1689598.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.852446269989014,
|
|
"epoch": 0.029253779029133967,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.8298,
|
|
"mean_token_accuracy": 0.1206131137907505,
|
|
"num_tokens": 1695600.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.896799898147583,
|
|
"epoch": 0.02935362127838016,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.7175,
|
|
"mean_token_accuracy": 0.12855139449238778,
|
|
"num_tokens": 1700639.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.842949771881104,
|
|
"epoch": 0.02945346352762635,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.7827,
|
|
"mean_token_accuracy": 0.1200721763074398,
|
|
"num_tokens": 1706411.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.822641134262085,
|
|
"epoch": 0.02955330577687254,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.7148,
|
|
"mean_token_accuracy": 0.12686698734760285,
|
|
"num_tokens": 1712809.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.8841558456420895,
|
|
"epoch": 0.029653148026118733,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.8102,
|
|
"mean_token_accuracy": 0.12312434241175652,
|
|
"num_tokens": 1718543.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.962130069732666,
|
|
"epoch": 0.029752990275364922,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.7034,
|
|
"mean_token_accuracy": 0.12929870337247848,
|
|
"num_tokens": 1724081.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.757903623580932,
|
|
"epoch": 0.029852832524611114,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.7297,
|
|
"mean_token_accuracy": 0.1286991134285927,
|
|
"num_tokens": 1729724.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.853930139541626,
|
|
"epoch": 0.029952674773857307,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.6381,
|
|
"mean_token_accuracy": 0.13184052780270578,
|
|
"num_tokens": 1734667.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"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": 3049514809344000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|