1035 lines
27 KiB
JSON
1035 lines
27 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.011532163203173651,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742021656036377,
|
|
"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
|
|
}
|
|
],
|
|
"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": 81741851688960.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|