3035 lines
81 KiB
JSON
3035 lines
81 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.034596489609520956,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742021656036377,
|
|
"epoch": 0.0001153216320317365,
|
|
"grad_norm": 6.875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.5028,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 3965.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742010688781738,
|
|
"epoch": 0.000230643264063473,
|
|
"grad_norm": 7.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.4561,
|
|
"mean_token_accuracy": 0.00022522523067891598,
|
|
"num_tokens": 7510.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742045021057129,
|
|
"epoch": 0.0003459648960952095,
|
|
"grad_norm": 6.71875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4301,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11084.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742029666900635,
|
|
"epoch": 0.000461286528126946,
|
|
"grad_norm": 6.59375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3211,
|
|
"mean_token_accuracy": 0.00036764706019312145,
|
|
"num_tokens": 14319.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742019271850586,
|
|
"epoch": 0.0005766081601586826,
|
|
"grad_norm": 5.9375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.2251,
|
|
"mean_token_accuracy": 0.003067592834122479,
|
|
"num_tokens": 17670.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.741859626770019,
|
|
"epoch": 0.000691929792190419,
|
|
"grad_norm": 5.78125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.1371,
|
|
"mean_token_accuracy": 0.027211554162204266,
|
|
"num_tokens": 21177.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.7411319732666,
|
|
"epoch": 0.0008072514242221556,
|
|
"grad_norm": 4.6875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9456,
|
|
"mean_token_accuracy": 0.05532368682324886,
|
|
"num_tokens": 24417.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.739075660705566,
|
|
"epoch": 0.000922573056253892,
|
|
"grad_norm": 4.03125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8362,
|
|
"mean_token_accuracy": 0.05229359455406666,
|
|
"num_tokens": 27824.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.734847354888917,
|
|
"epoch": 0.0010378946882856287,
|
|
"grad_norm": 3.375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.7625,
|
|
"mean_token_accuracy": 0.045908345840871334,
|
|
"num_tokens": 31712.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.730545330047608,
|
|
"epoch": 0.0011532163203173652,
|
|
"grad_norm": 3.0625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6394,
|
|
"mean_token_accuracy": 0.050624264404177666,
|
|
"num_tokens": 35207.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.728482437133788,
|
|
"epoch": 0.0012685379523491016,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6232,
|
|
"mean_token_accuracy": 0.04945621863007545,
|
|
"num_tokens": 39066.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.727034091949463,
|
|
"epoch": 0.001383859584380838,
|
|
"grad_norm": 2.53125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4208,
|
|
"mean_token_accuracy": 0.05683497190475464,
|
|
"num_tokens": 42389.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.72294578552246,
|
|
"epoch": 0.0014991812164125747,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4122,
|
|
"mean_token_accuracy": 0.060580461472272876,
|
|
"num_tokens": 45778.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.71625804901123,
|
|
"epoch": 0.0016145028484443112,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2958,
|
|
"mean_token_accuracy": 0.06154703982174396,
|
|
"num_tokens": 48918.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.70692949295044,
|
|
"epoch": 0.0017298244804760477,
|
|
"grad_norm": 2.671875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2954,
|
|
"mean_token_accuracy": 0.056437340751290324,
|
|
"num_tokens": 52456.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.700276470184326,
|
|
"epoch": 0.001845146112507784,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.237,
|
|
"mean_token_accuracy": 0.057351690530776975,
|
|
"num_tokens": 56193.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.697480201721191,
|
|
"epoch": 0.0019604677445395208,
|
|
"grad_norm": 2.921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1742,
|
|
"mean_token_accuracy": 0.06542446799576282,
|
|
"num_tokens": 59977.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.679913997650146,
|
|
"epoch": 0.0020757893765712574,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1228,
|
|
"mean_token_accuracy": 0.060923734307289125,
|
|
"num_tokens": 63844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.644670963287354,
|
|
"epoch": 0.0021911110086029937,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.0052,
|
|
"mean_token_accuracy": 0.06680413372814656,
|
|
"num_tokens": 67356.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.607680988311767,
|
|
"epoch": 0.0023064326406347303,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8542,
|
|
"mean_token_accuracy": 0.07436448931694031,
|
|
"num_tokens": 70767.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.569982147216797,
|
|
"epoch": 0.0024217542726664666,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.8446,
|
|
"mean_token_accuracy": 0.07276930250227451,
|
|
"num_tokens": 74437.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.519275093078614,
|
|
"epoch": 0.0025370759046982033,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.7407,
|
|
"mean_token_accuracy": 0.07625615149736405,
|
|
"num_tokens": 77940.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.484683418273926,
|
|
"epoch": 0.00265239753672994,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.6783,
|
|
"mean_token_accuracy": 0.06630247980356216,
|
|
"num_tokens": 81669.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.430662536621094,
|
|
"epoch": 0.002767719168761676,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.6506,
|
|
"mean_token_accuracy": 0.0662218876183033,
|
|
"num_tokens": 86135.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.37105360031128,
|
|
"epoch": 0.002883040800793413,
|
|
"grad_norm": 1.96875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.438,
|
|
"mean_token_accuracy": 0.07672500349581242,
|
|
"num_tokens": 89620.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.254761791229248,
|
|
"epoch": 0.0029983624328251495,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2973,
|
|
"mean_token_accuracy": 0.07640205062925816,
|
|
"num_tokens": 93245.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.16719446182251,
|
|
"epoch": 0.0031136840648568857,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.2293,
|
|
"mean_token_accuracy": 0.07873419560492038,
|
|
"num_tokens": 96754.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.051060390472411,
|
|
"epoch": 0.0032290056968886224,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.1469,
|
|
"mean_token_accuracy": 0.08079516477882862,
|
|
"num_tokens": 100266.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.90192928314209,
|
|
"epoch": 0.003344327328920359,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.0086,
|
|
"mean_token_accuracy": 0.07754942215979099,
|
|
"num_tokens": 103824.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.758885765075684,
|
|
"epoch": 0.0034596489609520953,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9559,
|
|
"mean_token_accuracy": 0.07231159210205078,
|
|
"num_tokens": 107659.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.57344217300415,
|
|
"epoch": 0.003574970592983832,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7848,
|
|
"mean_token_accuracy": 0.07904883921146393,
|
|
"num_tokens": 111348.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.393678665161133,
|
|
"epoch": 0.003690292225015568,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.781,
|
|
"mean_token_accuracy": 0.07496269196271896,
|
|
"num_tokens": 115386.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.191406345367431,
|
|
"epoch": 0.003805613857047305,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.6314,
|
|
"mean_token_accuracy": 0.08151335380971432,
|
|
"num_tokens": 118772.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.00878620147705,
|
|
"epoch": 0.0039209354890790415,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.6988,
|
|
"mean_token_accuracy": 0.08032485917210579,
|
|
"num_tokens": 122680.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.929794025421142,
|
|
"epoch": 0.004036257121110778,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6361,
|
|
"mean_token_accuracy": 0.06929410509765148,
|
|
"num_tokens": 127162.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.736303424835205,
|
|
"epoch": 0.004151578753142515,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4417,
|
|
"mean_token_accuracy": 0.08295803517103195,
|
|
"num_tokens": 130684.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.531741046905518,
|
|
"epoch": 0.004266900385174251,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.08261465467512608,
|
|
"num_tokens": 134154.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.454952239990234,
|
|
"epoch": 0.004382222017205987,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.4515,
|
|
"mean_token_accuracy": 0.08016656301915645,
|
|
"num_tokens": 137811.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.346348667144776,
|
|
"epoch": 0.004497543649237724,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3446,
|
|
"mean_token_accuracy": 0.08212785348296166,
|
|
"num_tokens": 141220.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.332176971435548,
|
|
"epoch": 0.004612865281269461,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.3456,
|
|
"mean_token_accuracy": 0.09070984274148941,
|
|
"num_tokens": 144768.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.172410202026366,
|
|
"epoch": 0.004728186913301197,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2619,
|
|
"mean_token_accuracy": 0.09590649902820587,
|
|
"num_tokens": 148173.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.156721019744873,
|
|
"epoch": 0.004843508545332933,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2489,
|
|
"mean_token_accuracy": 0.09932834953069687,
|
|
"num_tokens": 151861.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.06724648475647,
|
|
"epoch": 0.00495883017736467,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.1793,
|
|
"mean_token_accuracy": 0.0932161957025528,
|
|
"num_tokens": 155191.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.029554653167725,
|
|
"epoch": 0.0050741518093964065,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.251,
|
|
"mean_token_accuracy": 0.08765031173825263,
|
|
"num_tokens": 158655.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.052795362472533,
|
|
"epoch": 0.005189473441428143,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.2193,
|
|
"mean_token_accuracy": 0.08930162936449051,
|
|
"num_tokens": 162244.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.028767347335815,
|
|
"epoch": 0.00530479507345988,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3646,
|
|
"mean_token_accuracy": 0.09494938254356385,
|
|
"num_tokens": 165887.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.922196388244629,
|
|
"epoch": 0.0054201167054916165,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1272,
|
|
"mean_token_accuracy": 0.09443809911608696,
|
|
"num_tokens": 169248.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.923435258865356,
|
|
"epoch": 0.005535438337523352,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2258,
|
|
"mean_token_accuracy": 0.08999792523682118,
|
|
"num_tokens": 172559.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.926740837097168,
|
|
"epoch": 0.005650759969555089,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1897,
|
|
"mean_token_accuracy": 0.09816687665879727,
|
|
"num_tokens": 175961.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.846644592285156,
|
|
"epoch": 0.005766081601586826,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.2414,
|
|
"mean_token_accuracy": 0.09427700191736221,
|
|
"num_tokens": 179611.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.870696067810059,
|
|
"epoch": 0.005881403233618562,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.192,
|
|
"mean_token_accuracy": 0.09950039684772491,
|
|
"num_tokens": 183105.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.846232604980469,
|
|
"epoch": 0.005996724865650299,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.1981,
|
|
"mean_token_accuracy": 0.09012417867779732,
|
|
"num_tokens": 186682.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.812780809402466,
|
|
"epoch": 0.006112046497682035,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2472,
|
|
"mean_token_accuracy": 0.09005217626690865,
|
|
"num_tokens": 190487.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.831189346313477,
|
|
"epoch": 0.0062273681297137715,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.2184,
|
|
"mean_token_accuracy": 0.09543947987258435,
|
|
"num_tokens": 194445.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.788592338562012,
|
|
"epoch": 0.006342689761745508,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1396,
|
|
"mean_token_accuracy": 0.09726608917117119,
|
|
"num_tokens": 197864.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.855714988708496,
|
|
"epoch": 0.006458011393777245,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.0759,
|
|
"mean_token_accuracy": 0.09497340768575668,
|
|
"num_tokens": 201465.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.6531273365020756,
|
|
"epoch": 0.0065733330258089815,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0626,
|
|
"mean_token_accuracy": 0.09218045473098754,
|
|
"num_tokens": 204918.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.693502759933471,
|
|
"epoch": 0.006688654657840718,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1291,
|
|
"mean_token_accuracy": 0.10356841832399369,
|
|
"num_tokens": 208725.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.6476373195648195,
|
|
"epoch": 0.006803976289872454,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0995,
|
|
"mean_token_accuracy": 0.10565761625766754,
|
|
"num_tokens": 212286.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.739715099334717,
|
|
"epoch": 0.006919297921904191,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0161,
|
|
"mean_token_accuracy": 0.10625262334942817,
|
|
"num_tokens": 215900.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.547334480285644,
|
|
"epoch": 0.007034619553935927,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.0704,
|
|
"mean_token_accuracy": 0.09989853128790856,
|
|
"num_tokens": 219495.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.692840814590454,
|
|
"epoch": 0.007149941185967664,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.1342,
|
|
"mean_token_accuracy": 0.08733554258942604,
|
|
"num_tokens": 223358.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.652688360214233,
|
|
"epoch": 0.007265262817999401,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9969,
|
|
"mean_token_accuracy": 0.11347669884562492,
|
|
"num_tokens": 226549.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.5864927768707275,
|
|
"epoch": 0.007380584450031136,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.1201,
|
|
"mean_token_accuracy": 0.10434366092085838,
|
|
"num_tokens": 230102.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.69258713722229,
|
|
"epoch": 0.007495906082062873,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.0423,
|
|
"mean_token_accuracy": 0.10628816038370133,
|
|
"num_tokens": 233652.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.698617792129516,
|
|
"epoch": 0.00761122771409461,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.019,
|
|
"mean_token_accuracy": 0.10741165056824684,
|
|
"num_tokens": 237100.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.569851207733154,
|
|
"epoch": 0.007726549346126346,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9191,
|
|
"mean_token_accuracy": 0.10246064662933349,
|
|
"num_tokens": 240390.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.401850080490112,
|
|
"epoch": 0.007841870978158083,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.0172,
|
|
"mean_token_accuracy": 0.10681112185120582,
|
|
"num_tokens": 243955.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.614628744125366,
|
|
"epoch": 0.007957192610189819,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9765,
|
|
"mean_token_accuracy": 0.11208773031830788,
|
|
"num_tokens": 247371.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.602944850921631,
|
|
"epoch": 0.008072514242221556,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.0097,
|
|
"mean_token_accuracy": 0.10460884720087052,
|
|
"num_tokens": 250794.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.622193384170532,
|
|
"epoch": 0.008187835874253292,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0425,
|
|
"mean_token_accuracy": 0.10371957495808601,
|
|
"num_tokens": 254434.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.536931991577148,
|
|
"epoch": 0.00830315750628503,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0482,
|
|
"mean_token_accuracy": 0.10129153206944466,
|
|
"num_tokens": 258039.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.673097133636475,
|
|
"epoch": 0.008418479138316766,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0407,
|
|
"mean_token_accuracy": 0.11341009587049485,
|
|
"num_tokens": 261851.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.472537088394165,
|
|
"epoch": 0.008533800770348501,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0025,
|
|
"mean_token_accuracy": 0.10991203933954238,
|
|
"num_tokens": 265903.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.68609390258789,
|
|
"epoch": 0.008649122402380239,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0629,
|
|
"mean_token_accuracy": 0.10821591913700104,
|
|
"num_tokens": 269471.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.463858795166016,
|
|
"epoch": 0.008764444034411975,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.9775,
|
|
"mean_token_accuracy": 0.1095884695649147,
|
|
"num_tokens": 272945.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.583744382858276,
|
|
"epoch": 0.008879765666443712,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9999,
|
|
"mean_token_accuracy": 0.1146959975361824,
|
|
"num_tokens": 276300.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.481926393508911,
|
|
"epoch": 0.008995087298475448,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.0227,
|
|
"mean_token_accuracy": 0.0975736752152443,
|
|
"num_tokens": 279930.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.492558908462525,
|
|
"epoch": 0.009110408930507184,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9798,
|
|
"mean_token_accuracy": 0.1104181706905365,
|
|
"num_tokens": 283709.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.592624568939209,
|
|
"epoch": 0.009225730562538921,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1179,
|
|
"mean_token_accuracy": 0.10322732701897622,
|
|
"num_tokens": 287880.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.545290756225586,
|
|
"epoch": 0.009341052194570657,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8691,
|
|
"mean_token_accuracy": 0.11618536859750747,
|
|
"num_tokens": 291367.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.633703374862671,
|
|
"epoch": 0.009456373826602395,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0449,
|
|
"mean_token_accuracy": 0.10262716971337796,
|
|
"num_tokens": 294920.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.408407926559448,
|
|
"epoch": 0.00957169545863413,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.949,
|
|
"mean_token_accuracy": 0.11361531317234039,
|
|
"num_tokens": 298347.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.308802890777588,
|
|
"epoch": 0.009687017090665866,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.844,
|
|
"mean_token_accuracy": 0.11743432730436325,
|
|
"num_tokens": 301787.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.408286190032959,
|
|
"epoch": 0.009802338722697604,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9024,
|
|
"mean_token_accuracy": 0.11428986191749572,
|
|
"num_tokens": 305284.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.571081972122192,
|
|
"epoch": 0.00991766035472934,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0042,
|
|
"mean_token_accuracy": 0.10862119793891907,
|
|
"num_tokens": 308963.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.332292938232422,
|
|
"epoch": 0.010032981986761077,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8493,
|
|
"mean_token_accuracy": 0.11822831407189369,
|
|
"num_tokens": 312422.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.354525423049926,
|
|
"epoch": 0.010148303618792813,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.8159,
|
|
"mean_token_accuracy": 0.11766216531395912,
|
|
"num_tokens": 316194.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.4436431407928465,
|
|
"epoch": 0.010263625250824549,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.9387,
|
|
"mean_token_accuracy": 0.11376798823475838,
|
|
"num_tokens": 319824.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.281066131591797,
|
|
"epoch": 0.010378946882856286,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8218,
|
|
"mean_token_accuracy": 0.12678420096635817,
|
|
"num_tokens": 323503.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.375249338150025,
|
|
"epoch": 0.010494268514888022,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.9538,
|
|
"mean_token_accuracy": 0.11232749670743943,
|
|
"num_tokens": 326858.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.4025633335113525,
|
|
"epoch": 0.01060959014691976,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9062,
|
|
"mean_token_accuracy": 0.11565143764019012,
|
|
"num_tokens": 330370.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.458907747268677,
|
|
"epoch": 0.010724911778951495,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.8451,
|
|
"mean_token_accuracy": 0.11405345499515533,
|
|
"num_tokens": 334041.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.328810930252075,
|
|
"epoch": 0.010840233410983233,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.8215,
|
|
"mean_token_accuracy": 0.11701643615961074,
|
|
"num_tokens": 337724.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.327921342849732,
|
|
"epoch": 0.010955555043014969,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7295,
|
|
"mean_token_accuracy": 0.1177863135933876,
|
|
"num_tokens": 340946.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.270324945449829,
|
|
"epoch": 0.011070876675046705,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8849,
|
|
"mean_token_accuracy": 0.12444331720471383,
|
|
"num_tokens": 344603.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.37480788230896,
|
|
"epoch": 0.011186198307078442,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8489,
|
|
"mean_token_accuracy": 0.12077230662107467,
|
|
"num_tokens": 348153.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.291937065124512,
|
|
"epoch": 0.011301519939110178,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9269,
|
|
"mean_token_accuracy": 0.11545819267630578,
|
|
"num_tokens": 351682.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.421288204193115,
|
|
"epoch": 0.011416841571141916,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7485,
|
|
"mean_token_accuracy": 0.12845203131437302,
|
|
"num_tokens": 355245.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.289664030075073,
|
|
"epoch": 0.011532163203173651,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8448,
|
|
"mean_token_accuracy": 0.12541893050074576,
|
|
"num_tokens": 358888.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.256032466888428,
|
|
"epoch": 0.011647484835205387,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.895,
|
|
"mean_token_accuracy": 0.1204397052526474,
|
|
"num_tokens": 362626.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.243162775039673,
|
|
"epoch": 0.011762806467237125,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7621,
|
|
"mean_token_accuracy": 0.1269581660628319,
|
|
"num_tokens": 366038.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.213675785064697,
|
|
"epoch": 0.01187812809926886,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7473,
|
|
"mean_token_accuracy": 0.1257144331932068,
|
|
"num_tokens": 369594.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.182213354110718,
|
|
"epoch": 0.011993449731300598,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.7638,
|
|
"mean_token_accuracy": 0.1182183250784874,
|
|
"num_tokens": 373104.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.295378351211548,
|
|
"epoch": 0.012108771363332334,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7695,
|
|
"mean_token_accuracy": 0.13031666204333306,
|
|
"num_tokens": 376595.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.238579559326172,
|
|
"epoch": 0.01222409299536407,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.732,
|
|
"mean_token_accuracy": 0.1308875136077404,
|
|
"num_tokens": 380156.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.260405206680298,
|
|
"epoch": 0.012339414627395807,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.7805,
|
|
"mean_token_accuracy": 0.12662938982248306,
|
|
"num_tokens": 384027.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.102445602416992,
|
|
"epoch": 0.012454736259427543,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7174,
|
|
"mean_token_accuracy": 0.1260022670030594,
|
|
"num_tokens": 387550.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.094203948974609,
|
|
"epoch": 0.01257005789145928,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6872,
|
|
"mean_token_accuracy": 0.1364239566028118,
|
|
"num_tokens": 390963.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.21281795501709,
|
|
"epoch": 0.012685379523491016,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7373,
|
|
"mean_token_accuracy": 0.12801600620150566,
|
|
"num_tokens": 394925.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.168295288085938,
|
|
"epoch": 0.012800701155522754,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6973,
|
|
"mean_token_accuracy": 0.13043423891067504,
|
|
"num_tokens": 398277.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.060368776321411,
|
|
"epoch": 0.01291602278755449,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6855,
|
|
"mean_token_accuracy": 0.13090824484825134,
|
|
"num_tokens": 402031.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.3014678955078125,
|
|
"epoch": 0.013031344419586225,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7816,
|
|
"mean_token_accuracy": 0.1199102945625782,
|
|
"num_tokens": 405965.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.974482870101928,
|
|
"epoch": 0.013146666051617963,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6566,
|
|
"mean_token_accuracy": 0.13885213509202005,
|
|
"num_tokens": 409030.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.199598693847657,
|
|
"epoch": 0.013261987683649699,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.7265,
|
|
"mean_token_accuracy": 0.13360763862729072,
|
|
"num_tokens": 412644.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.124316883087158,
|
|
"epoch": 0.013377309315681436,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8504,
|
|
"mean_token_accuracy": 0.12711855992674828,
|
|
"num_tokens": 416572.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.1720640659332275,
|
|
"epoch": 0.013492630947713172,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7699,
|
|
"mean_token_accuracy": 0.12232115417718888,
|
|
"num_tokens": 420282.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.183737564086914,
|
|
"epoch": 0.013607952579744908,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.7675,
|
|
"mean_token_accuracy": 0.13052476122975348,
|
|
"num_tokens": 423856.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.144857501983642,
|
|
"epoch": 0.013723274211776645,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7764,
|
|
"mean_token_accuracy": 0.12236267328262329,
|
|
"num_tokens": 427505.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.138488531112671,
|
|
"epoch": 0.013838595843808381,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.795,
|
|
"mean_token_accuracy": 0.1210820995271206,
|
|
"num_tokens": 431019.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.971019458770752,
|
|
"epoch": 0.013953917475840119,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.7612,
|
|
"mean_token_accuracy": 0.11316960081458091,
|
|
"num_tokens": 434958.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.194779729843139,
|
|
"epoch": 0.014069239107871855,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6872,
|
|
"mean_token_accuracy": 0.13698609694838523,
|
|
"num_tokens": 438819.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.021146535873413,
|
|
"epoch": 0.01418456073990359,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.6825,
|
|
"mean_token_accuracy": 0.13391250744462013,
|
|
"num_tokens": 442673.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.011246109008789,
|
|
"epoch": 0.014299882371935328,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6265,
|
|
"mean_token_accuracy": 0.13872859627008438,
|
|
"num_tokens": 446022.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.936792039871216,
|
|
"epoch": 0.014415204003967064,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5687,
|
|
"mean_token_accuracy": 0.13359753265976906,
|
|
"num_tokens": 449417.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.843828582763672,
|
|
"epoch": 0.014530525635998801,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.623,
|
|
"mean_token_accuracy": 0.14156124815344812,
|
|
"num_tokens": 452864.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.110574722290039,
|
|
"epoch": 0.014645847268030537,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6629,
|
|
"mean_token_accuracy": 0.13592152297496796,
|
|
"num_tokens": 456271.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.8960436344146725,
|
|
"epoch": 0.014761168900062273,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.6724,
|
|
"mean_token_accuracy": 0.1281282015144825,
|
|
"num_tokens": 459829.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.219209623336792,
|
|
"epoch": 0.01487649053209401,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7325,
|
|
"mean_token_accuracy": 0.12644472867250442,
|
|
"num_tokens": 463533.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.049443197250366,
|
|
"epoch": 0.014991812164125746,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.8064,
|
|
"mean_token_accuracy": 0.13613916411995888,
|
|
"num_tokens": 467332.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.084133577346802,
|
|
"epoch": 0.015107133796157484,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5823,
|
|
"mean_token_accuracy": 0.1375665083527565,
|
|
"num_tokens": 470873.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.115225028991699,
|
|
"epoch": 0.01522245542818922,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7965,
|
|
"mean_token_accuracy": 0.12169739827513695,
|
|
"num_tokens": 474820.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.148219966888428,
|
|
"epoch": 0.015337777060220957,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.6386,
|
|
"mean_token_accuracy": 0.1283111959695816,
|
|
"num_tokens": 478395.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.927125406265259,
|
|
"epoch": 0.015453098692252693,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6057,
|
|
"mean_token_accuracy": 0.13224285766482352,
|
|
"num_tokens": 482046.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.033944511413575,
|
|
"epoch": 0.015568420324284429,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7945,
|
|
"mean_token_accuracy": 0.1299009971320629,
|
|
"num_tokens": 485673.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.172901773452759,
|
|
"epoch": 0.015683741956316166,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.7287,
|
|
"mean_token_accuracy": 0.12920700535178184,
|
|
"num_tokens": 489481.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.01988525390625,
|
|
"epoch": 0.015799063588347904,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6851,
|
|
"mean_token_accuracy": 0.12269095107913017,
|
|
"num_tokens": 493190.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.000023746490479,
|
|
"epoch": 0.015914385220379638,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6705,
|
|
"mean_token_accuracy": 0.13163023814558983,
|
|
"num_tokens": 496998.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.026926565170288,
|
|
"epoch": 0.016029706852411375,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6056,
|
|
"mean_token_accuracy": 0.13071495220065116,
|
|
"num_tokens": 500711.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.882460069656372,
|
|
"epoch": 0.016145028484443113,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6217,
|
|
"mean_token_accuracy": 0.13791571259498597,
|
|
"num_tokens": 504320.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.196972036361695,
|
|
"epoch": 0.016260350116474847,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.7158,
|
|
"mean_token_accuracy": 0.12625801786780358,
|
|
"num_tokens": 508591.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.642935752868652,
|
|
"epoch": 0.016375671748506584,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5201,
|
|
"mean_token_accuracy": 0.1374307781457901,
|
|
"num_tokens": 511954.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.1502011775970455,
|
|
"epoch": 0.016490993380538322,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.7295,
|
|
"mean_token_accuracy": 0.12733924239873887,
|
|
"num_tokens": 515674.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.986428070068359,
|
|
"epoch": 0.01660631501257006,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5921,
|
|
"mean_token_accuracy": 0.13627244234085084,
|
|
"num_tokens": 519263.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.887809419631958,
|
|
"epoch": 0.016721636644601794,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5426,
|
|
"mean_token_accuracy": 0.13855416104197502,
|
|
"num_tokens": 522852.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.8790075302124025,
|
|
"epoch": 0.01683695827663353,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3634,
|
|
"mean_token_accuracy": 0.15066966861486436,
|
|
"num_tokens": 525895.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.919354200363159,
|
|
"epoch": 0.01695227990866527,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.6803,
|
|
"mean_token_accuracy": 0.13804770186543464,
|
|
"num_tokens": 529767.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.9072585105896,
|
|
"epoch": 0.017067601540697003,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6582,
|
|
"mean_token_accuracy": 0.1369399406015873,
|
|
"num_tokens": 533460.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.891188287734986,
|
|
"epoch": 0.01718292317272874,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5457,
|
|
"mean_token_accuracy": 0.12860406264662744,
|
|
"num_tokens": 536858.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.973286294937134,
|
|
"epoch": 0.017298244804760478,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5516,
|
|
"mean_token_accuracy": 0.1398191474378109,
|
|
"num_tokens": 540385.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.866931486129761,
|
|
"epoch": 0.017413566436792212,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6073,
|
|
"mean_token_accuracy": 0.1349424861371517,
|
|
"num_tokens": 543900.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.864158821105957,
|
|
"epoch": 0.01752888806882395,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4816,
|
|
"mean_token_accuracy": 0.14627346023917198,
|
|
"num_tokens": 547698.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.8848350048065186,
|
|
"epoch": 0.017644209700855687,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.436,
|
|
"mean_token_accuracy": 0.14554693475365638,
|
|
"num_tokens": 551343.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.843238782882691,
|
|
"epoch": 0.017759531332887424,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5787,
|
|
"mean_token_accuracy": 0.13887985348701476,
|
|
"num_tokens": 554719.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.003432083129883,
|
|
"epoch": 0.01787485296491916,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.4253,
|
|
"mean_token_accuracy": 0.1476987384259701,
|
|
"num_tokens": 558065.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.605137872695923,
|
|
"epoch": 0.017990174596950896,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4838,
|
|
"mean_token_accuracy": 0.13948625475168228,
|
|
"num_tokens": 561838.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.009051418304443,
|
|
"epoch": 0.018105496228982634,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.5758,
|
|
"mean_token_accuracy": 0.14208441823720933,
|
|
"num_tokens": 565377.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.8511049270629885,
|
|
"epoch": 0.018220817861014368,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.5599,
|
|
"mean_token_accuracy": 0.1428572192788124,
|
|
"num_tokens": 568820.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.7770233154296875,
|
|
"epoch": 0.018336139493046105,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5236,
|
|
"mean_token_accuracy": 0.14241749197244644,
|
|
"num_tokens": 572524.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.862948751449585,
|
|
"epoch": 0.018451461125077843,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5636,
|
|
"mean_token_accuracy": 0.141788749396801,
|
|
"num_tokens": 575928.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.711058568954468,
|
|
"epoch": 0.01856678275710958,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.13463225737214088,
|
|
"num_tokens": 579592.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.781000709533691,
|
|
"epoch": 0.018682104389141314,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5346,
|
|
"mean_token_accuracy": 0.14024169445037843,
|
|
"num_tokens": 582901.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.882710933685303,
|
|
"epoch": 0.018797426021173052,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.5651,
|
|
"mean_token_accuracy": 0.14071007296442986,
|
|
"num_tokens": 586314.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.879060459136963,
|
|
"epoch": 0.01891274765320479,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5933,
|
|
"mean_token_accuracy": 0.1301053397357464,
|
|
"num_tokens": 590184.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.8879400253295895,
|
|
"epoch": 0.019028069285236524,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5387,
|
|
"mean_token_accuracy": 0.14789728671312333,
|
|
"num_tokens": 594206.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.814955615997315,
|
|
"epoch": 0.01914339091726826,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5642,
|
|
"mean_token_accuracy": 0.1346995562314987,
|
|
"num_tokens": 597907.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.744997978210449,
|
|
"epoch": 0.0192587125493,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4204,
|
|
"mean_token_accuracy": 0.14953070133924484,
|
|
"num_tokens": 601184.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.857750368118286,
|
|
"epoch": 0.019374034181331733,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4381,
|
|
"mean_token_accuracy": 0.14911116808652877,
|
|
"num_tokens": 604461.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.7806343078613285,
|
|
"epoch": 0.01948935581336347,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.1534685216844082,
|
|
"num_tokens": 608212.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.673570203781128,
|
|
"epoch": 0.019604677445395208,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3905,
|
|
"mean_token_accuracy": 0.15328768342733384,
|
|
"num_tokens": 611699.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.649495840072632,
|
|
"epoch": 0.019719999077426945,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.4736,
|
|
"mean_token_accuracy": 0.14484781697392463,
|
|
"num_tokens": 615250.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.805553245544433,
|
|
"epoch": 0.01983532070945868,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5798,
|
|
"mean_token_accuracy": 0.135956335067749,
|
|
"num_tokens": 619212.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.718591499328613,
|
|
"epoch": 0.019950642341490417,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4765,
|
|
"mean_token_accuracy": 0.1490676708519459,
|
|
"num_tokens": 622968.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.70408000946045,
|
|
"epoch": 0.020065963973522154,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3869,
|
|
"mean_token_accuracy": 0.14821098297834395,
|
|
"num_tokens": 626246.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.600178813934326,
|
|
"epoch": 0.02018128560555389,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4465,
|
|
"mean_token_accuracy": 0.1421129934489727,
|
|
"num_tokens": 629587.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.742484140396118,
|
|
"epoch": 0.020296607237585626,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3582,
|
|
"mean_token_accuracy": 0.15670016407966614,
|
|
"num_tokens": 633145.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.748065948486328,
|
|
"epoch": 0.020411928869617364,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.5601,
|
|
"mean_token_accuracy": 0.13865177184343339,
|
|
"num_tokens": 636833.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.831122016906738,
|
|
"epoch": 0.020527250501649098,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4765,
|
|
"mean_token_accuracy": 0.14462925493717194,
|
|
"num_tokens": 640416.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.613678026199341,
|
|
"epoch": 0.020642572133680835,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.3885,
|
|
"mean_token_accuracy": 0.1487666428089142,
|
|
"num_tokens": 643749.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.652020597457886,
|
|
"epoch": 0.020757893765712573,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4584,
|
|
"mean_token_accuracy": 0.14836200848221778,
|
|
"num_tokens": 647324.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.816301536560059,
|
|
"epoch": 0.02087321539774431,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5459,
|
|
"mean_token_accuracy": 0.14273923486471177,
|
|
"num_tokens": 650845.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.7125591278076175,
|
|
"epoch": 0.020988537029776044,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.3995,
|
|
"mean_token_accuracy": 0.14679229110479355,
|
|
"num_tokens": 654241.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.702474308013916,
|
|
"epoch": 0.021103858661807782,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5317,
|
|
"mean_token_accuracy": 0.14224857985973358,
|
|
"num_tokens": 658325.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.863994789123535,
|
|
"epoch": 0.02121918029383952,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5804,
|
|
"mean_token_accuracy": 0.13422438502311707,
|
|
"num_tokens": 661858.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.687291622161865,
|
|
"epoch": 0.021334501925871253,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4122,
|
|
"mean_token_accuracy": 0.14931200668215752,
|
|
"num_tokens": 665241.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.521294641494751,
|
|
"epoch": 0.02144982355790299,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2195,
|
|
"mean_token_accuracy": 0.16844909489154816,
|
|
"num_tokens": 668551.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.580444097518921,
|
|
"epoch": 0.02156514518993473,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4084,
|
|
"mean_token_accuracy": 0.1554088607430458,
|
|
"num_tokens": 672042.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.566297435760498,
|
|
"epoch": 0.021680466821966466,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2744,
|
|
"mean_token_accuracy": 0.1657500222325325,
|
|
"num_tokens": 675439.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.597789096832275,
|
|
"epoch": 0.0217957884539982,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.14348055571317672,
|
|
"num_tokens": 678922.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.507852172851562,
|
|
"epoch": 0.021911110086029938,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.3393,
|
|
"mean_token_accuracy": 0.1491332769393921,
|
|
"num_tokens": 682158.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.679322528839111,
|
|
"epoch": 0.022026431718061675,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.3042,
|
|
"mean_token_accuracy": 0.1576960414648056,
|
|
"num_tokens": 685852.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.703399705886841,
|
|
"epoch": 0.02214175335009341,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5342,
|
|
"mean_token_accuracy": 0.13639016672968865,
|
|
"num_tokens": 689757.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.55503625869751,
|
|
"epoch": 0.022257074982125147,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3188,
|
|
"mean_token_accuracy": 0.1594477578997612,
|
|
"num_tokens": 692721.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.620656299591064,
|
|
"epoch": 0.022372396614156884,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.51,
|
|
"mean_token_accuracy": 0.14588528275489807,
|
|
"num_tokens": 696739.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.659407997131348,
|
|
"epoch": 0.02248771824618862,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3588,
|
|
"mean_token_accuracy": 0.1492749720811844,
|
|
"num_tokens": 700203.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.545972156524658,
|
|
"epoch": 0.022603039878220356,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4537,
|
|
"mean_token_accuracy": 0.1438838504254818,
|
|
"num_tokens": 703675.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.6947386264801025,
|
|
"epoch": 0.022718361510252093,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.429,
|
|
"mean_token_accuracy": 0.14915493726730347,
|
|
"num_tokens": 707367.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.619061231613159,
|
|
"epoch": 0.02283368314228383,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4636,
|
|
"mean_token_accuracy": 0.14335431456565856,
|
|
"num_tokens": 711686.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.607592630386352,
|
|
"epoch": 0.022949004774315565,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.15773587226867675,
|
|
"num_tokens": 715183.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.534726762771607,
|
|
"epoch": 0.023064326406347303,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.3405,
|
|
"mean_token_accuracy": 0.15368402749300003,
|
|
"num_tokens": 718730.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.733155822753906,
|
|
"epoch": 0.02317964803837904,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.4205,
|
|
"mean_token_accuracy": 0.1442691281437874,
|
|
"num_tokens": 722187.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.555506229400635,
|
|
"epoch": 0.023294969670410774,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.3153,
|
|
"mean_token_accuracy": 0.15435170233249665,
|
|
"num_tokens": 725554.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.581414604187012,
|
|
"epoch": 0.023410291302442512,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.3737,
|
|
"mean_token_accuracy": 0.14934570640325545,
|
|
"num_tokens": 728875.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.493687105178833,
|
|
"epoch": 0.02352561293447425,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.1896,
|
|
"mean_token_accuracy": 0.1592538207769394,
|
|
"num_tokens": 731990.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.340290260314942,
|
|
"epoch": 0.023640934566505987,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.328,
|
|
"mean_token_accuracy": 0.1510552443563938,
|
|
"num_tokens": 735199.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.565967416763305,
|
|
"epoch": 0.02375625619853772,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.2783,
|
|
"mean_token_accuracy": 0.15099184811115265,
|
|
"num_tokens": 738783.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.403811693191528,
|
|
"epoch": 0.02387157783056946,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.16056827455759048,
|
|
"num_tokens": 742223.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.605123805999756,
|
|
"epoch": 0.023986899462601196,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.4548,
|
|
"mean_token_accuracy": 0.14856265038251876,
|
|
"num_tokens": 746360.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.615904331207275,
|
|
"epoch": 0.02410222109463293,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.3408,
|
|
"mean_token_accuracy": 0.14928023889660835,
|
|
"num_tokens": 749832.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.564470386505127,
|
|
"epoch": 0.024217542726664668,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.4036,
|
|
"mean_token_accuracy": 0.1470089204609394,
|
|
"num_tokens": 753658.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.350269460678101,
|
|
"epoch": 0.024332864358696405,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.3151,
|
|
"mean_token_accuracy": 0.14769368395209312,
|
|
"num_tokens": 756801.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.491814279556275,
|
|
"epoch": 0.02444818599072814,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.2635,
|
|
"mean_token_accuracy": 0.14811939746141434,
|
|
"num_tokens": 760156.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.578709697723388,
|
|
"epoch": 0.024563507622759877,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.2967,
|
|
"mean_token_accuracy": 0.15609683692455292,
|
|
"num_tokens": 763698.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.444866037368774,
|
|
"epoch": 0.024678829254791614,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.4254,
|
|
"mean_token_accuracy": 0.1493972808122635,
|
|
"num_tokens": 767064.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.564963006973267,
|
|
"epoch": 0.024794150886823352,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.318,
|
|
"mean_token_accuracy": 0.14573369473218917,
|
|
"num_tokens": 770789.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.5673754692077635,
|
|
"epoch": 0.024909472518855086,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.3552,
|
|
"mean_token_accuracy": 0.1587481126189232,
|
|
"num_tokens": 774560.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.565120553970337,
|
|
"epoch": 0.025024794150886823,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.1671,
|
|
"mean_token_accuracy": 0.1629626825451851,
|
|
"num_tokens": 777789.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.329736518859863,
|
|
"epoch": 0.02514011578291856,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.2969,
|
|
"mean_token_accuracy": 0.16039335876703262,
|
|
"num_tokens": 781328.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.625707387924194,
|
|
"epoch": 0.025255437414950295,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.2725,
|
|
"mean_token_accuracy": 0.15458450838923454,
|
|
"num_tokens": 784852.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.370968818664551,
|
|
"epoch": 0.025370759046982033,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.2534,
|
|
"mean_token_accuracy": 0.15902130156755448,
|
|
"num_tokens": 788309.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.412897634506225,
|
|
"epoch": 0.02548608067901377,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.2843,
|
|
"mean_token_accuracy": 0.15403042435646058,
|
|
"num_tokens": 791916.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.457622051239014,
|
|
"epoch": 0.025601402311045508,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.2666,
|
|
"mean_token_accuracy": 0.15358882620930672,
|
|
"num_tokens": 795592.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.4621740818023685,
|
|
"epoch": 0.02571672394307724,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.3631,
|
|
"mean_token_accuracy": 0.144256030023098,
|
|
"num_tokens": 799422.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.535422849655151,
|
|
"epoch": 0.02583204557510898,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.3138,
|
|
"mean_token_accuracy": 0.16079722866415977,
|
|
"num_tokens": 803083.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.581855201721192,
|
|
"epoch": 0.025947367207140717,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.1882,
|
|
"mean_token_accuracy": 0.16269952207803726,
|
|
"num_tokens": 806512.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.266746091842651,
|
|
"epoch": 0.02606268883917245,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.3644,
|
|
"mean_token_accuracy": 0.15194013267755507,
|
|
"num_tokens": 810081.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.515526151657104,
|
|
"epoch": 0.02617801047120419,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.325,
|
|
"mean_token_accuracy": 0.15144732221961021,
|
|
"num_tokens": 813844.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.55663833618164,
|
|
"epoch": 0.026293332103235926,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.3479,
|
|
"mean_token_accuracy": 0.15247188955545427,
|
|
"num_tokens": 817334.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.441150236129761,
|
|
"epoch": 0.02640865373526766,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.2644,
|
|
"mean_token_accuracy": 0.15573957711458206,
|
|
"num_tokens": 821106.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.452196168899536,
|
|
"epoch": 0.026523975367299397,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.2624,
|
|
"mean_token_accuracy": 0.15255917683243753,
|
|
"num_tokens": 824625.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.423743200302124,
|
|
"epoch": 0.026639296999331135,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.16335436403751374,
|
|
"num_tokens": 827879.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.420325756072998,
|
|
"epoch": 0.026754618631362873,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.3283,
|
|
"mean_token_accuracy": 0.15196822658181192,
|
|
"num_tokens": 831320.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.467994832992554,
|
|
"epoch": 0.026869940263394607,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.3333,
|
|
"mean_token_accuracy": 0.14823555946350098,
|
|
"num_tokens": 834454.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.357031345367432,
|
|
"epoch": 0.026985261895426344,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.281,
|
|
"mean_token_accuracy": 0.14480863139033318,
|
|
"num_tokens": 838182.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.462740278244018,
|
|
"epoch": 0.02710058352745808,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.1131,
|
|
"mean_token_accuracy": 0.16392626315355302,
|
|
"num_tokens": 841684.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.513607311248779,
|
|
"epoch": 0.027215905159489816,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.3624,
|
|
"mean_token_accuracy": 0.15705716609954834,
|
|
"num_tokens": 845335.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.240785837173462,
|
|
"epoch": 0.027331226791521553,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.2473,
|
|
"mean_token_accuracy": 0.15439498722553252,
|
|
"num_tokens": 848901.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.475341510772705,
|
|
"epoch": 0.02744654842355329,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.3809,
|
|
"mean_token_accuracy": 0.14640313088893891,
|
|
"num_tokens": 852567.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.593037557601929,
|
|
"epoch": 0.02756187005558503,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.4614,
|
|
"mean_token_accuracy": 0.14304601326584815,
|
|
"num_tokens": 856798.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.563302755355835,
|
|
"epoch": 0.027677191687616762,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.3882,
|
|
"mean_token_accuracy": 0.1500474914908409,
|
|
"num_tokens": 860463.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.443354368209839,
|
|
"epoch": 0.0277925133196485,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.357,
|
|
"mean_token_accuracy": 0.15508870482444764,
|
|
"num_tokens": 864384.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.41762490272522,
|
|
"epoch": 0.027907834951680238,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.2378,
|
|
"mean_token_accuracy": 0.16346798688173295,
|
|
"num_tokens": 867928.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.332534742355347,
|
|
"epoch": 0.02802315658371197,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.1574,
|
|
"mean_token_accuracy": 0.16043390184640885,
|
|
"num_tokens": 871373.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.275541162490844,
|
|
"epoch": 0.02813847821574371,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.2439,
|
|
"mean_token_accuracy": 0.15379589051008224,
|
|
"num_tokens": 875043.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.4980240821838375,
|
|
"epoch": 0.028253799847775447,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.3772,
|
|
"mean_token_accuracy": 0.15180049538612367,
|
|
"num_tokens": 878329.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.336836481094361,
|
|
"epoch": 0.02836912147980718,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.1803,
|
|
"mean_token_accuracy": 0.15510529577732085,
|
|
"num_tokens": 881719.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.452073669433593,
|
|
"epoch": 0.028484443111838918,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.2149,
|
|
"mean_token_accuracy": 0.17287414222955705,
|
|
"num_tokens": 885558.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.363374042510986,
|
|
"epoch": 0.028599764743870656,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.2344,
|
|
"mean_token_accuracy": 0.16227788776159285,
|
|
"num_tokens": 888882.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.407885885238647,
|
|
"epoch": 0.028715086375902393,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.2726,
|
|
"mean_token_accuracy": 0.14910536557435988,
|
|
"num_tokens": 892599.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.36954026222229,
|
|
"epoch": 0.028830408007934127,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.1148,
|
|
"mean_token_accuracy": 0.1696225017309189,
|
|
"num_tokens": 896314.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.212534475326538,
|
|
"epoch": 0.028945729639965865,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.0984,
|
|
"mean_token_accuracy": 0.16258733570575715,
|
|
"num_tokens": 899822.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.2822267532348635,
|
|
"epoch": 0.029061051271997602,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.2221,
|
|
"mean_token_accuracy": 0.15711940824985504,
|
|
"num_tokens": 903076.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.249537801742553,
|
|
"epoch": 0.029176372904029337,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.0233,
|
|
"mean_token_accuracy": 0.159354268014431,
|
|
"num_tokens": 906270.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.102555179595948,
|
|
"epoch": 0.029291694536061074,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.1359,
|
|
"mean_token_accuracy": 0.16118508726358413,
|
|
"num_tokens": 909749.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.432159471511841,
|
|
"epoch": 0.02940701616809281,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.3627,
|
|
"mean_token_accuracy": 0.14390599578619004,
|
|
"num_tokens": 913513.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.547852230072022,
|
|
"epoch": 0.029522337800124546,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.2054,
|
|
"mean_token_accuracy": 0.15653806552290916,
|
|
"num_tokens": 917233.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.256132745742798,
|
|
"epoch": 0.029637659432156283,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.1938,
|
|
"mean_token_accuracy": 0.1588376834988594,
|
|
"num_tokens": 920667.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.267224502563477,
|
|
"epoch": 0.02975298106418802,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.1545,
|
|
"mean_token_accuracy": 0.1703912615776062,
|
|
"num_tokens": 924089.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.17415566444397,
|
|
"epoch": 0.02986830269621976,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.1166,
|
|
"mean_token_accuracy": 0.16841672509908676,
|
|
"num_tokens": 927677.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.437988567352295,
|
|
"epoch": 0.029983624328251492,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.1931,
|
|
"mean_token_accuracy": 0.15562398582696915,
|
|
"num_tokens": 931103.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.3561567783355715,
|
|
"epoch": 0.03009894596028323,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.21,
|
|
"mean_token_accuracy": 0.15286269336938857,
|
|
"num_tokens": 934898.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.381216096878052,
|
|
"epoch": 0.030214267592314967,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.16,
|
|
"mean_token_accuracy": 0.16273058503866195,
|
|
"num_tokens": 938065.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.238500690460205,
|
|
"epoch": 0.0303295892243467,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.2126,
|
|
"mean_token_accuracy": 0.16215117052197456,
|
|
"num_tokens": 941841.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.372848749160767,
|
|
"epoch": 0.03044491085637844,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.2133,
|
|
"mean_token_accuracy": 0.15352949500083923,
|
|
"num_tokens": 945498.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.382011842727661,
|
|
"epoch": 0.030560232488410177,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.1665,
|
|
"mean_token_accuracy": 0.15650930255651474,
|
|
"num_tokens": 949043.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.320769596099853,
|
|
"epoch": 0.030675554120441914,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.1856,
|
|
"mean_token_accuracy": 0.15382006615400315,
|
|
"num_tokens": 952868.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.251468706130981,
|
|
"epoch": 0.030790875752473648,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 5.9936,
|
|
"mean_token_accuracy": 0.17058618664741515,
|
|
"num_tokens": 956112.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.1639808177947994,
|
|
"epoch": 0.030906197384505386,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.1816,
|
|
"mean_token_accuracy": 0.15354717746376992,
|
|
"num_tokens": 959742.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.466977548599243,
|
|
"epoch": 0.031021519016537123,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.1836,
|
|
"mean_token_accuracy": 0.14992421269416809,
|
|
"num_tokens": 963153.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.324162340164184,
|
|
"epoch": 0.031136840648568857,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.1243,
|
|
"mean_token_accuracy": 0.15615553557872772,
|
|
"num_tokens": 966531.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.207290983200073,
|
|
"epoch": 0.0312521622806006,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.1917,
|
|
"mean_token_accuracy": 0.15856592804193498,
|
|
"num_tokens": 970441.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.47190260887146,
|
|
"epoch": 0.03136748391263233,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.2361,
|
|
"mean_token_accuracy": 0.15151861011981965,
|
|
"num_tokens": 974081.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.372641849517822,
|
|
"epoch": 0.031482805544664066,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.1708,
|
|
"mean_token_accuracy": 0.16533790826797484,
|
|
"num_tokens": 977477.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.199554920196533,
|
|
"epoch": 0.03159812717669581,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.2106,
|
|
"mean_token_accuracy": 0.1591222569346428,
|
|
"num_tokens": 980672.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.334047889709472,
|
|
"epoch": 0.03171344880872754,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.1362,
|
|
"mean_token_accuracy": 0.16225132793188096,
|
|
"num_tokens": 984148.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.244274473190307,
|
|
"epoch": 0.031828770440759276,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.0759,
|
|
"mean_token_accuracy": 0.1592021256685257,
|
|
"num_tokens": 987532.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.233033657073975,
|
|
"epoch": 0.03194409207279102,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.1987,
|
|
"mean_token_accuracy": 0.15314276814460753,
|
|
"num_tokens": 991300.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.3514176368713375,
|
|
"epoch": 0.03205941370482275,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.0972,
|
|
"mean_token_accuracy": 0.16658444255590438,
|
|
"num_tokens": 994860.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.299854564666748,
|
|
"epoch": 0.032174735336854485,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.0349,
|
|
"mean_token_accuracy": 0.16576472967863082,
|
|
"num_tokens": 998320.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.200660228729248,
|
|
"epoch": 0.032290056968886226,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.0562,
|
|
"mean_token_accuracy": 0.16334922313690187,
|
|
"num_tokens": 1001510.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.216449165344239,
|
|
"epoch": 0.03240537860091796,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.0977,
|
|
"mean_token_accuracy": 0.159677754342556,
|
|
"num_tokens": 1005051.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.143203020095825,
|
|
"epoch": 0.032520700232949694,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.0517,
|
|
"mean_token_accuracy": 0.1645487815141678,
|
|
"num_tokens": 1008339.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.3322797298431395,
|
|
"epoch": 0.032636021864981435,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.284,
|
|
"mean_token_accuracy": 0.1555835984647274,
|
|
"num_tokens": 1012072.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.48833703994751,
|
|
"epoch": 0.03275134349701317,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.1539,
|
|
"mean_token_accuracy": 0.16940778195858003,
|
|
"num_tokens": 1015733.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.182392120361328,
|
|
"epoch": 0.0328666651290449,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.2657,
|
|
"mean_token_accuracy": 0.14596002101898192,
|
|
"num_tokens": 1019723.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.272395658493042,
|
|
"epoch": 0.032981986761076644,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.0907,
|
|
"mean_token_accuracy": 0.16454826593399047,
|
|
"num_tokens": 1023399.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.4120201587677,
|
|
"epoch": 0.03309730839310838,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.286,
|
|
"mean_token_accuracy": 0.14998917058110237,
|
|
"num_tokens": 1027654.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.292629241943359,
|
|
"epoch": 0.03321263002514012,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.2046,
|
|
"mean_token_accuracy": 0.15546465963125228,
|
|
"num_tokens": 1031093.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.23744764328003,
|
|
"epoch": 0.03332795165717185,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.121,
|
|
"mean_token_accuracy": 0.16061940789222717,
|
|
"num_tokens": 1034283.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.334246635437012,
|
|
"epoch": 0.03344327328920359,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.0285,
|
|
"mean_token_accuracy": 0.1709355279803276,
|
|
"num_tokens": 1038026.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.312613296508789,
|
|
"epoch": 0.03355859492123533,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.144,
|
|
"mean_token_accuracy": 0.16978459805250168,
|
|
"num_tokens": 1041691.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.297551918029785,
|
|
"epoch": 0.03367391655326706,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.248,
|
|
"mean_token_accuracy": 0.1555471897125244,
|
|
"num_tokens": 1045355.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.275693273544311,
|
|
"epoch": 0.033789238185298796,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.21,
|
|
"mean_token_accuracy": 0.15987666249275206,
|
|
"num_tokens": 1048708.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.252631616592407,
|
|
"epoch": 0.03390455981733054,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.0775,
|
|
"mean_token_accuracy": 0.15647277384996414,
|
|
"num_tokens": 1052193.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.326667451858521,
|
|
"epoch": 0.03401988144936227,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.067,
|
|
"mean_token_accuracy": 0.16924083083868027,
|
|
"num_tokens": 1055680.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.380450296401977,
|
|
"epoch": 0.034135203081394005,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.2271,
|
|
"mean_token_accuracy": 0.14974521473050117,
|
|
"num_tokens": 1059594.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.145453453063965,
|
|
"epoch": 0.034250524713425747,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.0015,
|
|
"mean_token_accuracy": 0.1716235727071762,
|
|
"num_tokens": 1062911.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.197411680221558,
|
|
"epoch": 0.03436584634545748,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.117,
|
|
"mean_token_accuracy": 0.16592582762241365,
|
|
"num_tokens": 1066682.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.303461742401123,
|
|
"epoch": 0.034481167977489215,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.3243,
|
|
"mean_token_accuracy": 0.14860715121030807,
|
|
"num_tokens": 1070559.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.478640699386597,
|
|
"epoch": 0.034596489609520956,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.1401,
|
|
"mean_token_accuracy": 0.15192170217633247,
|
|
"num_tokens": 1074079.0,
|
|
"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": 243138456944640.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|