8035 lines
219 KiB
JSON
8035 lines
219 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.2679079736110646,
|
|
"eval_steps": 500,
|
|
"global_step": 4000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692002010345458,
|
|
"epoch": 0.00033488496701383075,
|
|
"grad_norm": 10.5625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.798,
|
|
"mean_token_accuracy": 0.00014925373252481223,
|
|
"num_tokens": 10113.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69187297821045,
|
|
"epoch": 0.0006697699340276615,
|
|
"grad_norm": 11.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.749,
|
|
"mean_token_accuracy": 0.000525744108017534,
|
|
"num_tokens": 19598.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690606594085693,
|
|
"epoch": 0.0010046549010414923,
|
|
"grad_norm": 8.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5809,
|
|
"mean_token_accuracy": 0.0301290619187057,
|
|
"num_tokens": 29705.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.670234203338623,
|
|
"epoch": 0.001339539868055323,
|
|
"grad_norm": 8.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.3539,
|
|
"mean_token_accuracy": 0.04047959893941879,
|
|
"num_tokens": 39871.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.557568454742432,
|
|
"epoch": 0.0016744248350691537,
|
|
"grad_norm": 4.3125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.1453,
|
|
"mean_token_accuracy": 0.04189771674573421,
|
|
"num_tokens": 49805.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.465806770324708,
|
|
"epoch": 0.0020093098020829846,
|
|
"grad_norm": 2.90625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.9507,
|
|
"mean_token_accuracy": 0.04296911805868149,
|
|
"num_tokens": 59845.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.461338233947753,
|
|
"epoch": 0.0023441947690968153,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.9614,
|
|
"mean_token_accuracy": 0.0368592556566,
|
|
"num_tokens": 69836.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.483565521240234,
|
|
"epoch": 0.002679079736110646,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.8929,
|
|
"mean_token_accuracy": 0.046063635125756265,
|
|
"num_tokens": 80088.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.410326480865479,
|
|
"epoch": 0.0030139647031244767,
|
|
"grad_norm": 2.359375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.8768,
|
|
"mean_token_accuracy": 0.04229466691613197,
|
|
"num_tokens": 89458.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.434312725067139,
|
|
"epoch": 0.0033488496701383074,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8494,
|
|
"mean_token_accuracy": 0.04813771955668926,
|
|
"num_tokens": 99523.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.504166984558106,
|
|
"epoch": 0.003683734637152138,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8242,
|
|
"mean_token_accuracy": 0.05405057594180107,
|
|
"num_tokens": 110080.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.344751262664795,
|
|
"epoch": 0.004018619604165969,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.7129,
|
|
"mean_token_accuracy": 0.05208342745900154,
|
|
"num_tokens": 119412.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.211891746520996,
|
|
"epoch": 0.0043535045711798,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.6776,
|
|
"mean_token_accuracy": 0.051375290378928185,
|
|
"num_tokens": 129432.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.302197170257568,
|
|
"epoch": 0.0046883895381936306,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6112,
|
|
"mean_token_accuracy": 0.059913215786218645,
|
|
"num_tokens": 139756.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.31943874359131,
|
|
"epoch": 0.005023274505207461,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5806,
|
|
"mean_token_accuracy": 0.05665694735944271,
|
|
"num_tokens": 148846.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.25365629196167,
|
|
"epoch": 0.005358159472221292,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.5087,
|
|
"mean_token_accuracy": 0.06129935123026371,
|
|
"num_tokens": 158647.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.23855152130127,
|
|
"epoch": 0.005693044439235123,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.3883,
|
|
"mean_token_accuracy": 0.060005851462483405,
|
|
"num_tokens": 167701.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.186964225769042,
|
|
"epoch": 0.006027929406248953,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3761,
|
|
"mean_token_accuracy": 0.06175476983189583,
|
|
"num_tokens": 178428.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.192708683013915,
|
|
"epoch": 0.006362814373262784,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.3126,
|
|
"mean_token_accuracy": 0.057852984964847566,
|
|
"num_tokens": 187737.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.148875999450684,
|
|
"epoch": 0.006697699340276615,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3047,
|
|
"mean_token_accuracy": 0.05756702236831188,
|
|
"num_tokens": 198374.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.143085479736328,
|
|
"epoch": 0.007032584307290445,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2348,
|
|
"mean_token_accuracy": 0.05544475130736828,
|
|
"num_tokens": 208483.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.065252590179444,
|
|
"epoch": 0.007367469274304276,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1179,
|
|
"mean_token_accuracy": 0.0532416757196188,
|
|
"num_tokens": 218715.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.994503688812255,
|
|
"epoch": 0.007702354241318108,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9596,
|
|
"mean_token_accuracy": 0.06599014401435851,
|
|
"num_tokens": 227777.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.866777515411377,
|
|
"epoch": 0.008037239208331938,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.982,
|
|
"mean_token_accuracy": 0.05981399100273847,
|
|
"num_tokens": 238192.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.84700288772583,
|
|
"epoch": 0.008372124175345769,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.827,
|
|
"mean_token_accuracy": 0.06517648994922638,
|
|
"num_tokens": 247259.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.726642894744874,
|
|
"epoch": 0.0087070091423596,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8551,
|
|
"mean_token_accuracy": 0.054852675274014474,
|
|
"num_tokens": 258100.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.696339225769043,
|
|
"epoch": 0.00904189410937343,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.7497,
|
|
"mean_token_accuracy": 0.061625052243471146,
|
|
"num_tokens": 267620.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.46251974105835,
|
|
"epoch": 0.009376779076387261,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.6451,
|
|
"mean_token_accuracy": 0.07104777507483959,
|
|
"num_tokens": 277654.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.472876358032227,
|
|
"epoch": 0.009711664043401092,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6622,
|
|
"mean_token_accuracy": 0.061386901512742044,
|
|
"num_tokens": 288195.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.343764114379884,
|
|
"epoch": 0.010046549010414923,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.6348,
|
|
"mean_token_accuracy": 0.0647747952491045,
|
|
"num_tokens": 298860.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.256326389312743,
|
|
"epoch": 0.010381433977428753,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5569,
|
|
"mean_token_accuracy": 0.06752786003053188,
|
|
"num_tokens": 307799.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.192852115631103,
|
|
"epoch": 0.010716318944442584,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.5493,
|
|
"mean_token_accuracy": 0.06289746835827828,
|
|
"num_tokens": 318257.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.057830619812012,
|
|
"epoch": 0.011051203911456415,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.5351,
|
|
"mean_token_accuracy": 0.06436791345477104,
|
|
"num_tokens": 329065.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.072613048553468,
|
|
"epoch": 0.011386088878470245,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4825,
|
|
"mean_token_accuracy": 0.06475077606737614,
|
|
"num_tokens": 338809.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.959860801696777,
|
|
"epoch": 0.011720973845484076,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.4588,
|
|
"mean_token_accuracy": 0.06541862487792968,
|
|
"num_tokens": 349318.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.849260616302491,
|
|
"epoch": 0.012055858812497907,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.4348,
|
|
"mean_token_accuracy": 0.06766009479761123,
|
|
"num_tokens": 358760.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.860241985321045,
|
|
"epoch": 0.012390743779511737,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3168,
|
|
"mean_token_accuracy": 0.06955580413341522,
|
|
"num_tokens": 367589.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.835146236419678,
|
|
"epoch": 0.012725628746525568,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.3624,
|
|
"mean_token_accuracy": 0.06997261717915534,
|
|
"num_tokens": 377012.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.665368556976318,
|
|
"epoch": 0.013060513713539399,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.3276,
|
|
"mean_token_accuracy": 0.07220411114394665,
|
|
"num_tokens": 386791.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.820245933532714,
|
|
"epoch": 0.01339539868055323,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.4404,
|
|
"mean_token_accuracy": 0.06381001807749272,
|
|
"num_tokens": 396620.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.68319730758667,
|
|
"epoch": 0.01373028364756706,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.3971,
|
|
"mean_token_accuracy": 0.06604428850114345,
|
|
"num_tokens": 406384.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.685719966888428,
|
|
"epoch": 0.01406516861458089,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.3622,
|
|
"mean_token_accuracy": 0.07016287520527839,
|
|
"num_tokens": 416481.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.711257934570312,
|
|
"epoch": 0.014400053581594721,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.3841,
|
|
"mean_token_accuracy": 0.07033630795776843,
|
|
"num_tokens": 425697.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.652897453308105,
|
|
"epoch": 0.014734938548608552,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2995,
|
|
"mean_token_accuracy": 0.0704294566065073,
|
|
"num_tokens": 435836.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.68178768157959,
|
|
"epoch": 0.015069823515622383,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3569,
|
|
"mean_token_accuracy": 0.06939754635095596,
|
|
"num_tokens": 445926.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.638516330718994,
|
|
"epoch": 0.015404708482636215,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.2983,
|
|
"mean_token_accuracy": 0.07235002666711807,
|
|
"num_tokens": 455594.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.570248413085938,
|
|
"epoch": 0.015739593449650046,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.3461,
|
|
"mean_token_accuracy": 0.0698947947472334,
|
|
"num_tokens": 465928.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.62375259399414,
|
|
"epoch": 0.016074478416663877,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.2851,
|
|
"mean_token_accuracy": 0.07294886186718941,
|
|
"num_tokens": 475946.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.62733097076416,
|
|
"epoch": 0.016409363383677707,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.2878,
|
|
"mean_token_accuracy": 0.07264770828187465,
|
|
"num_tokens": 485285.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.576191997528076,
|
|
"epoch": 0.016744248350691538,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.2942,
|
|
"mean_token_accuracy": 0.07046032212674617,
|
|
"num_tokens": 496516.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.584659576416016,
|
|
"epoch": 0.01707913331770537,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.2829,
|
|
"mean_token_accuracy": 0.07218111529946328,
|
|
"num_tokens": 506330.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.551284503936767,
|
|
"epoch": 0.0174140182847192,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3443,
|
|
"mean_token_accuracy": 0.07309619709849358,
|
|
"num_tokens": 515825.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.580291271209717,
|
|
"epoch": 0.01774890325173303,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2687,
|
|
"mean_token_accuracy": 0.06971881128847598,
|
|
"num_tokens": 526303.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.48112440109253,
|
|
"epoch": 0.01808378821874686,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.3759,
|
|
"mean_token_accuracy": 0.07506775110960007,
|
|
"num_tokens": 536805.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.600781631469726,
|
|
"epoch": 0.01841867318576069,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.2216,
|
|
"mean_token_accuracy": 0.07659760937094688,
|
|
"num_tokens": 546381.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.565075874328613,
|
|
"epoch": 0.018753558152774522,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.2444,
|
|
"mean_token_accuracy": 0.0764175396412611,
|
|
"num_tokens": 556278.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.52633352279663,
|
|
"epoch": 0.019088443119788353,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2569,
|
|
"mean_token_accuracy": 0.07796998918056489,
|
|
"num_tokens": 565312.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.471183109283448,
|
|
"epoch": 0.019423328086802184,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.2313,
|
|
"mean_token_accuracy": 0.07882754877209663,
|
|
"num_tokens": 575304.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.441308689117431,
|
|
"epoch": 0.019758213053816014,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.1941,
|
|
"mean_token_accuracy": 0.07772204428911209,
|
|
"num_tokens": 585262.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.575240802764892,
|
|
"epoch": 0.020093098020829845,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.2345,
|
|
"mean_token_accuracy": 0.07687565684318542,
|
|
"num_tokens": 595297.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.531679725646972,
|
|
"epoch": 0.020427982987843676,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.2113,
|
|
"mean_token_accuracy": 0.07619660049676895,
|
|
"num_tokens": 604354.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.408655452728272,
|
|
"epoch": 0.020762867954857506,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.3286,
|
|
"mean_token_accuracy": 0.07043087370693683,
|
|
"num_tokens": 616283.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.580455112457276,
|
|
"epoch": 0.021097752921871337,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.2259,
|
|
"mean_token_accuracy": 0.07452421076595783,
|
|
"num_tokens": 626059.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.43509645462036,
|
|
"epoch": 0.021432637888885168,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.1612,
|
|
"mean_token_accuracy": 0.07599973231554032,
|
|
"num_tokens": 634483.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.409455680847168,
|
|
"epoch": 0.021767522855899,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1607,
|
|
"mean_token_accuracy": 0.08002224043011666,
|
|
"num_tokens": 643111.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.471851444244384,
|
|
"epoch": 0.02210240782291283,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.1549,
|
|
"mean_token_accuracy": 0.08036141917109489,
|
|
"num_tokens": 653650.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.432329463958741,
|
|
"epoch": 0.02243729278992666,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1509,
|
|
"mean_token_accuracy": 0.08708992898464203,
|
|
"num_tokens": 663143.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.438332271575927,
|
|
"epoch": 0.02277217775694049,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.1221,
|
|
"mean_token_accuracy": 0.08310596123337746,
|
|
"num_tokens": 672434.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.37921953201294,
|
|
"epoch": 0.02310706272395432,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.1224,
|
|
"mean_token_accuracy": 0.07925173193216324,
|
|
"num_tokens": 682973.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.429475784301758,
|
|
"epoch": 0.023441947690968152,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.1876,
|
|
"mean_token_accuracy": 0.07883969098329544,
|
|
"num_tokens": 692136.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.41495771408081,
|
|
"epoch": 0.023776832657981983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.0919,
|
|
"mean_token_accuracy": 0.08360464051365853,
|
|
"num_tokens": 702257.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.438455581665039,
|
|
"epoch": 0.024111717624995813,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1034,
|
|
"mean_token_accuracy": 0.08223021924495696,
|
|
"num_tokens": 712444.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.402928638458253,
|
|
"epoch": 0.024446602592009644,
|
|
"grad_norm": 2.59375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.2461,
|
|
"mean_token_accuracy": 0.07820370681583881,
|
|
"num_tokens": 721614.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.494019985198975,
|
|
"epoch": 0.024781487559023475,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.2242,
|
|
"mean_token_accuracy": 0.07433447390794753,
|
|
"num_tokens": 731189.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.423804473876952,
|
|
"epoch": 0.025116372526037305,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.1387,
|
|
"mean_token_accuracy": 0.08087190091609955,
|
|
"num_tokens": 741555.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.242571067810058,
|
|
"epoch": 0.025451257493051136,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.0677,
|
|
"mean_token_accuracy": 0.084798514097929,
|
|
"num_tokens": 751560.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.45106019973755,
|
|
"epoch": 0.025786142460064967,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.1193,
|
|
"mean_token_accuracy": 0.07877458333969116,
|
|
"num_tokens": 761471.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.435086345672607,
|
|
"epoch": 0.026121027427078797,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.1184,
|
|
"mean_token_accuracy": 0.08121965527534485,
|
|
"num_tokens": 771565.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.258657455444336,
|
|
"epoch": 0.026455912394092628,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.0685,
|
|
"mean_token_accuracy": 0.08140704035758972,
|
|
"num_tokens": 781248.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.397636890411377,
|
|
"epoch": 0.02679079736110646,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.1233,
|
|
"mean_token_accuracy": 0.0783721823245287,
|
|
"num_tokens": 790907.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.232399845123291,
|
|
"epoch": 0.02712568232812029,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.1384,
|
|
"mean_token_accuracy": 0.07922295071184635,
|
|
"num_tokens": 800073.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.517311382293702,
|
|
"epoch": 0.02746056729513412,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.099,
|
|
"mean_token_accuracy": 0.0789324052631855,
|
|
"num_tokens": 809782.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.192907619476319,
|
|
"epoch": 0.02779545226214795,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0579,
|
|
"mean_token_accuracy": 0.08063366934657097,
|
|
"num_tokens": 820329.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.31554536819458,
|
|
"epoch": 0.02813033722916178,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9757,
|
|
"mean_token_accuracy": 0.09008945301175117,
|
|
"num_tokens": 829405.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.286150169372558,
|
|
"epoch": 0.028465222196175612,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.0418,
|
|
"mean_token_accuracy": 0.08079335764050484,
|
|
"num_tokens": 838700.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.263776969909667,
|
|
"epoch": 0.028800107163189443,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.1043,
|
|
"mean_token_accuracy": 0.083824248239398,
|
|
"num_tokens": 848892.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.295380115509033,
|
|
"epoch": 0.029134992130203274,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0354,
|
|
"mean_token_accuracy": 0.08374309539794922,
|
|
"num_tokens": 857737.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.229405975341797,
|
|
"epoch": 0.029469877097217104,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.968,
|
|
"mean_token_accuracy": 0.08430874347686768,
|
|
"num_tokens": 867758.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.303557300567627,
|
|
"epoch": 0.029804762064230935,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.9853,
|
|
"mean_token_accuracy": 0.08112291395664215,
|
|
"num_tokens": 877273.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.286294078826904,
|
|
"epoch": 0.030139647031244766,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.087,
|
|
"mean_token_accuracy": 0.08337052017450333,
|
|
"num_tokens": 887109.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.264797973632813,
|
|
"epoch": 0.030474531998258596,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.0057,
|
|
"mean_token_accuracy": 0.08394731357693672,
|
|
"num_tokens": 896157.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.19506025314331,
|
|
"epoch": 0.03080941696527243,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.011,
|
|
"mean_token_accuracy": 0.07882361188530922,
|
|
"num_tokens": 907231.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.185263442993165,
|
|
"epoch": 0.03114430193228626,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.9223,
|
|
"mean_token_accuracy": 0.0873487912118435,
|
|
"num_tokens": 916756.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.243685722351074,
|
|
"epoch": 0.03147918689930009,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.0922,
|
|
"mean_token_accuracy": 0.07646922953426838,
|
|
"num_tokens": 927714.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.29259443283081,
|
|
"epoch": 0.03181407186631392,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.9389,
|
|
"mean_token_accuracy": 0.08775902166962624,
|
|
"num_tokens": 936620.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.152761650085449,
|
|
"epoch": 0.03214895683332775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9753,
|
|
"mean_token_accuracy": 0.08129271119832993,
|
|
"num_tokens": 945942.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.177737617492676,
|
|
"epoch": 0.032483841800341584,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.0066,
|
|
"mean_token_accuracy": 0.08826668187975883,
|
|
"num_tokens": 956685.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.196662044525146,
|
|
"epoch": 0.032818726767355415,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.9442,
|
|
"mean_token_accuracy": 0.08767017722129822,
|
|
"num_tokens": 967453.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.19489345550537,
|
|
"epoch": 0.033153611734369245,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9764,
|
|
"mean_token_accuracy": 0.08438041880726814,
|
|
"num_tokens": 978220.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.135752773284912,
|
|
"epoch": 0.033488496701383076,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9185,
|
|
"mean_token_accuracy": 0.09252278879284859,
|
|
"num_tokens": 988120.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.218993663787842,
|
|
"epoch": 0.03382338166839691,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9388,
|
|
"mean_token_accuracy": 0.08044274970889091,
|
|
"num_tokens": 997733.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.14416561126709,
|
|
"epoch": 0.03415826663541074,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9004,
|
|
"mean_token_accuracy": 0.08713611736893653,
|
|
"num_tokens": 1007362.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.12597336769104,
|
|
"epoch": 0.03449315160242457,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8522,
|
|
"mean_token_accuracy": 0.08573579043149948,
|
|
"num_tokens": 1016352.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.978172826766968,
|
|
"epoch": 0.0348280365694384,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.8683,
|
|
"mean_token_accuracy": 0.08755372166633606,
|
|
"num_tokens": 1025944.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.23288698196411,
|
|
"epoch": 0.03516292153645223,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.9434,
|
|
"mean_token_accuracy": 0.08828524276614189,
|
|
"num_tokens": 1035675.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.088372182846069,
|
|
"epoch": 0.03549780650346606,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.936,
|
|
"mean_token_accuracy": 0.08637615889310837,
|
|
"num_tokens": 1045944.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.136125946044922,
|
|
"epoch": 0.03583269147047989,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.8464,
|
|
"mean_token_accuracy": 0.09038148149847984,
|
|
"num_tokens": 1055721.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.160592937469483,
|
|
"epoch": 0.03616757643749372,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.978,
|
|
"mean_token_accuracy": 0.08611000180244446,
|
|
"num_tokens": 1065841.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.033481216430664,
|
|
"epoch": 0.03650246140450755,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.976,
|
|
"mean_token_accuracy": 0.08632155880331993,
|
|
"num_tokens": 1076574.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.213283681869507,
|
|
"epoch": 0.03683734637152138,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.9227,
|
|
"mean_token_accuracy": 0.08273187503218651,
|
|
"num_tokens": 1085761.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.176979160308838,
|
|
"epoch": 0.037172231338535214,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.9048,
|
|
"mean_token_accuracy": 0.08671903014183044,
|
|
"num_tokens": 1096036.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.981817817687988,
|
|
"epoch": 0.037507116305549044,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8203,
|
|
"mean_token_accuracy": 0.08757113367319107,
|
|
"num_tokens": 1105103.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.080931758880615,
|
|
"epoch": 0.037842001272562875,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.863,
|
|
"mean_token_accuracy": 0.09306468516588211,
|
|
"num_tokens": 1115355.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.037179470062256,
|
|
"epoch": 0.038176886239576706,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8377,
|
|
"mean_token_accuracy": 0.08505469113588333,
|
|
"num_tokens": 1125977.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.067077589035033,
|
|
"epoch": 0.03851177120659054,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.8574,
|
|
"mean_token_accuracy": 0.08890697509050369,
|
|
"num_tokens": 1135536.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.069751501083374,
|
|
"epoch": 0.03884665617360437,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.7972,
|
|
"mean_token_accuracy": 0.0888807274401188,
|
|
"num_tokens": 1144857.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.992953777313232,
|
|
"epoch": 0.0391815411406182,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.8085,
|
|
"mean_token_accuracy": 0.08944643214344979,
|
|
"num_tokens": 1154188.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.046015882492066,
|
|
"epoch": 0.03951642610763203,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.847,
|
|
"mean_token_accuracy": 0.08940242640674115,
|
|
"num_tokens": 1164583.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.920463609695434,
|
|
"epoch": 0.03985131107464586,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.8122,
|
|
"mean_token_accuracy": 0.09227430075407028,
|
|
"num_tokens": 1174163.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.987206220626831,
|
|
"epoch": 0.04018619604165969,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.8286,
|
|
"mean_token_accuracy": 0.08923885226249695,
|
|
"num_tokens": 1184065.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.923127269744873,
|
|
"epoch": 0.04052108100867352,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.6988,
|
|
"mean_token_accuracy": 0.1023092582821846,
|
|
"num_tokens": 1193263.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.989051961898804,
|
|
"epoch": 0.04085596597568735,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.7558,
|
|
"mean_token_accuracy": 0.09200508967041969,
|
|
"num_tokens": 1202940.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.819897556304932,
|
|
"epoch": 0.04119085094270118,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.6977,
|
|
"mean_token_accuracy": 0.09320536255836487,
|
|
"num_tokens": 1212978.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.914996862411499,
|
|
"epoch": 0.04152573590971501,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7375,
|
|
"mean_token_accuracy": 0.09381118640303612,
|
|
"num_tokens": 1222945.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.922825765609741,
|
|
"epoch": 0.04186062087672884,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.7243,
|
|
"mean_token_accuracy": 0.0925237312912941,
|
|
"num_tokens": 1232041.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.018640756607056,
|
|
"epoch": 0.042195505843742674,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.7608,
|
|
"mean_token_accuracy": 0.09631229490041733,
|
|
"num_tokens": 1241765.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.8229437351226805,
|
|
"epoch": 0.042530390810756505,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7353,
|
|
"mean_token_accuracy": 0.09919300377368927,
|
|
"num_tokens": 1250642.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.911471843719482,
|
|
"epoch": 0.042865275777770336,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7431,
|
|
"mean_token_accuracy": 0.09453340247273445,
|
|
"num_tokens": 1260250.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.082605934143066,
|
|
"epoch": 0.043200160744784166,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.8231,
|
|
"mean_token_accuracy": 0.08769241273403168,
|
|
"num_tokens": 1269786.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.858733367919922,
|
|
"epoch": 0.043535045711798,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7604,
|
|
"mean_token_accuracy": 0.08618547506630421,
|
|
"num_tokens": 1280029.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.007468605041504,
|
|
"epoch": 0.04386993067881183,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.718,
|
|
"mean_token_accuracy": 0.08944717943668365,
|
|
"num_tokens": 1288983.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.869023227691651,
|
|
"epoch": 0.04420481564582566,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7681,
|
|
"mean_token_accuracy": 0.09466379135847092,
|
|
"num_tokens": 1298629.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.920618724822998,
|
|
"epoch": 0.04453970061283949,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.7448,
|
|
"mean_token_accuracy": 0.08919373750686646,
|
|
"num_tokens": 1308611.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.893028736114502,
|
|
"epoch": 0.04487458557985332,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7868,
|
|
"mean_token_accuracy": 0.08869361504912376,
|
|
"num_tokens": 1318150.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.96576623916626,
|
|
"epoch": 0.04520947054686715,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7329,
|
|
"mean_token_accuracy": 0.0911085106432438,
|
|
"num_tokens": 1328933.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.807632970809936,
|
|
"epoch": 0.04554435551388098,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7145,
|
|
"mean_token_accuracy": 0.09275061488151551,
|
|
"num_tokens": 1338614.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.796882915496826,
|
|
"epoch": 0.04587924048089481,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6533,
|
|
"mean_token_accuracy": 0.08674826323986054,
|
|
"num_tokens": 1348292.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.877967834472656,
|
|
"epoch": 0.04621412544790864,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7007,
|
|
"mean_token_accuracy": 0.09283162206411362,
|
|
"num_tokens": 1358479.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.635259914398193,
|
|
"epoch": 0.04654901041492247,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.5801,
|
|
"mean_token_accuracy": 0.09727387279272079,
|
|
"num_tokens": 1367778.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.858510446548462,
|
|
"epoch": 0.046883895381936304,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.668,
|
|
"mean_token_accuracy": 0.09592612758278847,
|
|
"num_tokens": 1376826.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.8318394184112545,
|
|
"epoch": 0.047218780348950135,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.7229,
|
|
"mean_token_accuracy": 0.09366490095853805,
|
|
"num_tokens": 1386280.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.815138149261474,
|
|
"epoch": 0.047553665315963965,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7523,
|
|
"mean_token_accuracy": 0.09019050747156143,
|
|
"num_tokens": 1396062.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.9230225563049315,
|
|
"epoch": 0.047888550282977796,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.7367,
|
|
"mean_token_accuracy": 0.0929199367761612,
|
|
"num_tokens": 1406980.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.8387205600738525,
|
|
"epoch": 0.04822343524999163,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.7349,
|
|
"mean_token_accuracy": 0.09628795683383942,
|
|
"num_tokens": 1417341.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.781990957260132,
|
|
"epoch": 0.04855832021700546,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.7154,
|
|
"mean_token_accuracy": 0.09174965173006058,
|
|
"num_tokens": 1427066.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.9304931640625,
|
|
"epoch": 0.04889320518401929,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6933,
|
|
"mean_token_accuracy": 0.0906587302684784,
|
|
"num_tokens": 1437429.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.806986618041992,
|
|
"epoch": 0.04922809015103312,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.795,
|
|
"mean_token_accuracy": 0.08047471344470977,
|
|
"num_tokens": 1449128.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.767903661727905,
|
|
"epoch": 0.04956297511804695,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6489,
|
|
"mean_token_accuracy": 0.09558213129639626,
|
|
"num_tokens": 1459397.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.746322727203369,
|
|
"epoch": 0.04989786008506078,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6609,
|
|
"mean_token_accuracy": 0.09293203428387642,
|
|
"num_tokens": 1469424.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.720838212966919,
|
|
"epoch": 0.05023274505207461,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6218,
|
|
"mean_token_accuracy": 0.10069447308778763,
|
|
"num_tokens": 1478980.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.836196994781494,
|
|
"epoch": 0.05056763001908844,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.6726,
|
|
"mean_token_accuracy": 0.09147711992263793,
|
|
"num_tokens": 1488655.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.671448421478272,
|
|
"epoch": 0.05090251498610227,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6464,
|
|
"mean_token_accuracy": 0.0919630341231823,
|
|
"num_tokens": 1498248.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.777037143707275,
|
|
"epoch": 0.0512373999531161,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6067,
|
|
"mean_token_accuracy": 0.0970597043633461,
|
|
"num_tokens": 1509496.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.764213752746582,
|
|
"epoch": 0.051572284920129934,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6505,
|
|
"mean_token_accuracy": 0.0956808902323246,
|
|
"num_tokens": 1518741.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.794732761383057,
|
|
"epoch": 0.051907169887143764,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.6665,
|
|
"mean_token_accuracy": 0.09552088975906373,
|
|
"num_tokens": 1528928.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.712340307235718,
|
|
"epoch": 0.052242054854157595,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5846,
|
|
"mean_token_accuracy": 0.09882296845316887,
|
|
"num_tokens": 1538255.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.661662578582764,
|
|
"epoch": 0.052576939821171426,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.7101,
|
|
"mean_token_accuracy": 0.09138747528195382,
|
|
"num_tokens": 1549014.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.798562860488891,
|
|
"epoch": 0.052911824788185256,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.5892,
|
|
"mean_token_accuracy": 0.09761035665869713,
|
|
"num_tokens": 1558905.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.620625352859497,
|
|
"epoch": 0.05324670975519909,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5663,
|
|
"mean_token_accuracy": 0.09829202219843865,
|
|
"num_tokens": 1568605.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.721243143081665,
|
|
"epoch": 0.05358159472221292,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.5793,
|
|
"mean_token_accuracy": 0.09484488666057586,
|
|
"num_tokens": 1579330.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.629036903381348,
|
|
"epoch": 0.05391647968922675,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5517,
|
|
"mean_token_accuracy": 0.10525800734758377,
|
|
"num_tokens": 1588702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.6774791240692135,
|
|
"epoch": 0.05425136465624058,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5431,
|
|
"mean_token_accuracy": 0.09909340515732765,
|
|
"num_tokens": 1598789.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.629120206832885,
|
|
"epoch": 0.05458624962325441,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.5682,
|
|
"mean_token_accuracy": 0.09840805605053901,
|
|
"num_tokens": 1608580.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.713317251205444,
|
|
"epoch": 0.05492113459026824,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5903,
|
|
"mean_token_accuracy": 0.09436297565698623,
|
|
"num_tokens": 1618344.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.623961067199707,
|
|
"epoch": 0.05525601955728207,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5522,
|
|
"mean_token_accuracy": 0.09576157703995705,
|
|
"num_tokens": 1627654.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.673608446121216,
|
|
"epoch": 0.0555909045242959,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.7425,
|
|
"mean_token_accuracy": 0.09187466725707054,
|
|
"num_tokens": 1638591.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.863967800140381,
|
|
"epoch": 0.05592578949130973,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.7303,
|
|
"mean_token_accuracy": 0.08798938393592834,
|
|
"num_tokens": 1649384.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.7440564155578615,
|
|
"epoch": 0.05626067445832356,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.7142,
|
|
"mean_token_accuracy": 0.08833635300397873,
|
|
"num_tokens": 1659079.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.69318208694458,
|
|
"epoch": 0.056595559425337394,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.6531,
|
|
"mean_token_accuracy": 0.09180667996406555,
|
|
"num_tokens": 1669245.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.693623304367065,
|
|
"epoch": 0.056930444392351225,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.5619,
|
|
"mean_token_accuracy": 0.09287880584597588,
|
|
"num_tokens": 1678781.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.553192472457885,
|
|
"epoch": 0.057265329359365055,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5305,
|
|
"mean_token_accuracy": 0.09830001443624496,
|
|
"num_tokens": 1688192.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.70746922492981,
|
|
"epoch": 0.057600214326378886,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5205,
|
|
"mean_token_accuracy": 0.09981869980692863,
|
|
"num_tokens": 1698058.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.522954750061035,
|
|
"epoch": 0.05793509929339272,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5455,
|
|
"mean_token_accuracy": 0.09677255153656006,
|
|
"num_tokens": 1708568.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.559320640563965,
|
|
"epoch": 0.05826998426040655,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.4243,
|
|
"mean_token_accuracy": 0.10318205356597901,
|
|
"num_tokens": 1718854.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.73996000289917,
|
|
"epoch": 0.05860486922742038,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.5525,
|
|
"mean_token_accuracy": 0.09444483071565628,
|
|
"num_tokens": 1728549.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.554740953445434,
|
|
"epoch": 0.05893975419443421,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5246,
|
|
"mean_token_accuracy": 0.09602352306246757,
|
|
"num_tokens": 1737664.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.663396739959717,
|
|
"epoch": 0.05927463916144804,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.6325,
|
|
"mean_token_accuracy": 0.09179740771651268,
|
|
"num_tokens": 1748791.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.649189853668213,
|
|
"epoch": 0.05960952412846187,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5579,
|
|
"mean_token_accuracy": 0.09603997468948364,
|
|
"num_tokens": 1758323.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.5717028141021725,
|
|
"epoch": 0.0599444090954757,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.5289,
|
|
"mean_token_accuracy": 0.09905452281236649,
|
|
"num_tokens": 1768992.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.641081666946411,
|
|
"epoch": 0.06027929406248953,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.5959,
|
|
"mean_token_accuracy": 0.09036346450448036,
|
|
"num_tokens": 1779810.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.533492517471314,
|
|
"epoch": 0.06061417902950336,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4949,
|
|
"mean_token_accuracy": 0.10070249736309052,
|
|
"num_tokens": 1790071.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.578271436691284,
|
|
"epoch": 0.06094906399651719,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.5443,
|
|
"mean_token_accuracy": 0.09873676598072052,
|
|
"num_tokens": 1799818.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.532402181625367,
|
|
"epoch": 0.06128394896353103,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.5053,
|
|
"mean_token_accuracy": 0.09719008877873421,
|
|
"num_tokens": 1810559.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.512709522247315,
|
|
"epoch": 0.06161883393054486,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4387,
|
|
"mean_token_accuracy": 0.09774044081568718,
|
|
"num_tokens": 1819810.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.604777812957764,
|
|
"epoch": 0.06195371889755869,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.5699,
|
|
"mean_token_accuracy": 0.0973470464348793,
|
|
"num_tokens": 1830737.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.58216872215271,
|
|
"epoch": 0.06228860386457252,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.5539,
|
|
"mean_token_accuracy": 0.09564543291926383,
|
|
"num_tokens": 1841134.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.623823022842407,
|
|
"epoch": 0.06262348883158635,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4891,
|
|
"mean_token_accuracy": 0.1037807047367096,
|
|
"num_tokens": 1851962.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.517935132980346,
|
|
"epoch": 0.06295837379860018,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.5686,
|
|
"mean_token_accuracy": 0.09641919285058975,
|
|
"num_tokens": 1862748.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.5019701480865475,
|
|
"epoch": 0.06329325876561401,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.5053,
|
|
"mean_token_accuracy": 0.09677049964666366,
|
|
"num_tokens": 1872398.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.4540633201599125,
|
|
"epoch": 0.06362814373262785,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.3761,
|
|
"mean_token_accuracy": 0.10685181841254235,
|
|
"num_tokens": 1882263.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.627192592620849,
|
|
"epoch": 0.06396302869964167,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.5367,
|
|
"mean_token_accuracy": 0.09033481255173684,
|
|
"num_tokens": 1893007.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.4436921119689945,
|
|
"epoch": 0.0642979136666555,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.5383,
|
|
"mean_token_accuracy": 0.10231668278574943,
|
|
"num_tokens": 1903357.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.596249151229858,
|
|
"epoch": 0.06463279863366933,
|
|
"grad_norm": 2.46875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.4762,
|
|
"mean_token_accuracy": 0.09526494145393372,
|
|
"num_tokens": 1913793.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.469473552703858,
|
|
"epoch": 0.06496768360068317,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.452,
|
|
"mean_token_accuracy": 0.09744430407881736,
|
|
"num_tokens": 1922540.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.527045965194702,
|
|
"epoch": 0.06530256856769699,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.5215,
|
|
"mean_token_accuracy": 0.09597379714250565,
|
|
"num_tokens": 1933706.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.437564373016357,
|
|
"epoch": 0.06563745353471083,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.4096,
|
|
"mean_token_accuracy": 0.10182152763009071,
|
|
"num_tokens": 1944169.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.391198968887329,
|
|
"epoch": 0.06597233850172465,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3428,
|
|
"mean_token_accuracy": 0.1063389353454113,
|
|
"num_tokens": 1954017.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.518598890304565,
|
|
"epoch": 0.06630722346873849,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.5243,
|
|
"mean_token_accuracy": 0.0978463925421238,
|
|
"num_tokens": 1965075.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.42684416770935,
|
|
"epoch": 0.06664210843575231,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.4063,
|
|
"mean_token_accuracy": 0.10326866209506988,
|
|
"num_tokens": 1974808.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.512786722183227,
|
|
"epoch": 0.06697699340276615,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.4755,
|
|
"mean_token_accuracy": 0.09493202716112137,
|
|
"num_tokens": 1985172.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.5474714756011965,
|
|
"epoch": 0.06731187836977998,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.4858,
|
|
"mean_token_accuracy": 0.10103376507759095,
|
|
"num_tokens": 1996338.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.445226097106934,
|
|
"epoch": 0.06764676333679381,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.4732,
|
|
"mean_token_accuracy": 0.09459194168448448,
|
|
"num_tokens": 2005429.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.482739734649658,
|
|
"epoch": 0.06798164830380764,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.4646,
|
|
"mean_token_accuracy": 0.09422318562865258,
|
|
"num_tokens": 2015278.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.479223680496216,
|
|
"epoch": 0.06831653327082148,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.4532,
|
|
"mean_token_accuracy": 0.09788647592067719,
|
|
"num_tokens": 2025758.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.373387289047241,
|
|
"epoch": 0.0686514182378353,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.3723,
|
|
"mean_token_accuracy": 0.10488328263163567,
|
|
"num_tokens": 2035029.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.414871549606323,
|
|
"epoch": 0.06898630320484914,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.3718,
|
|
"mean_token_accuracy": 0.10269584804773331,
|
|
"num_tokens": 2044182.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.479082298278809,
|
|
"epoch": 0.06932118817186296,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.4598,
|
|
"mean_token_accuracy": 0.09844749867916107,
|
|
"num_tokens": 2054676.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.414534330368042,
|
|
"epoch": 0.0696560731388768,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.3981,
|
|
"mean_token_accuracy": 0.09748629033565522,
|
|
"num_tokens": 2062933.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.395028114318848,
|
|
"epoch": 0.06999095810589062,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.3612,
|
|
"mean_token_accuracy": 0.10393474400043487,
|
|
"num_tokens": 2073146.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.429117488861084,
|
|
"epoch": 0.07032584307290446,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.4926,
|
|
"mean_token_accuracy": 0.09564846009016037,
|
|
"num_tokens": 2083847.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.4130007266998295,
|
|
"epoch": 0.07066072803991828,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.3997,
|
|
"mean_token_accuracy": 0.09807902723550796,
|
|
"num_tokens": 2094969.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.458782052993774,
|
|
"epoch": 0.07099561300693212,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.4067,
|
|
"mean_token_accuracy": 0.0952724613249302,
|
|
"num_tokens": 2105275.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.375722408294678,
|
|
"epoch": 0.07133049797394594,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.4499,
|
|
"mean_token_accuracy": 0.09948632344603539,
|
|
"num_tokens": 2115833.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.466668558120728,
|
|
"epoch": 0.07166538294095978,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 7.3751,
|
|
"mean_token_accuracy": 0.1037701740860939,
|
|
"num_tokens": 2124747.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.423423814773559,
|
|
"epoch": 0.0720002679079736,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 7.4149,
|
|
"mean_token_accuracy": 0.09693367630243302,
|
|
"num_tokens": 2135114.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.397886610031128,
|
|
"epoch": 0.07233515287498744,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 7.433,
|
|
"mean_token_accuracy": 0.09895812720060349,
|
|
"num_tokens": 2144324.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.453474044799805,
|
|
"epoch": 0.07267003784200127,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 7.3563,
|
|
"mean_token_accuracy": 0.09745102897286415,
|
|
"num_tokens": 2153971.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.311913967132568,
|
|
"epoch": 0.0730049228090151,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 7.3283,
|
|
"mean_token_accuracy": 0.10133487731218338,
|
|
"num_tokens": 2164389.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.367936801910401,
|
|
"epoch": 0.07333980777602893,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 7.3145,
|
|
"mean_token_accuracy": 0.10260041207075118,
|
|
"num_tokens": 2173802.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.433255243301391,
|
|
"epoch": 0.07367469274304277,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.3249,
|
|
"mean_token_accuracy": 0.103522939234972,
|
|
"num_tokens": 2182590.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.250359439849854,
|
|
"epoch": 0.07400957771005659,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 7.3198,
|
|
"mean_token_accuracy": 0.1075370505452156,
|
|
"num_tokens": 2192025.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.5280996799469,
|
|
"epoch": 0.07434446267707043,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 7.4923,
|
|
"mean_token_accuracy": 0.10366464406251907,
|
|
"num_tokens": 2201999.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.378125333786011,
|
|
"epoch": 0.07467934764408425,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.09555793032050133,
|
|
"num_tokens": 2211804.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.355848169326782,
|
|
"epoch": 0.07501423261109809,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 7.2508,
|
|
"mean_token_accuracy": 0.1032866396009922,
|
|
"num_tokens": 2222079.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.278160572052002,
|
|
"epoch": 0.07534911757811191,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.3811,
|
|
"mean_token_accuracy": 0.10743188410997391,
|
|
"num_tokens": 2232213.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.39425253868103,
|
|
"epoch": 0.07568400254512575,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 7.3066,
|
|
"mean_token_accuracy": 0.10472791343927383,
|
|
"num_tokens": 2242017.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.2768415927886965,
|
|
"epoch": 0.07601888751213957,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.2968,
|
|
"mean_token_accuracy": 0.10491021052002907,
|
|
"num_tokens": 2251654.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.349087905883789,
|
|
"epoch": 0.07635377247915341,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 7.3466,
|
|
"mean_token_accuracy": 0.10458370298147202,
|
|
"num_tokens": 2262972.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.255107593536377,
|
|
"epoch": 0.07668865744616724,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 7.3488,
|
|
"mean_token_accuracy": 0.09580066353082657,
|
|
"num_tokens": 2273067.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.342799282073974,
|
|
"epoch": 0.07702354241318107,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 7.2318,
|
|
"mean_token_accuracy": 0.10805003568530083,
|
|
"num_tokens": 2282104.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.299344301223755,
|
|
"epoch": 0.0773584273801949,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 7.3451,
|
|
"mean_token_accuracy": 0.10275197774171829,
|
|
"num_tokens": 2291366.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.345579051971436,
|
|
"epoch": 0.07769331234720873,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 7.2663,
|
|
"mean_token_accuracy": 0.10576335266232491,
|
|
"num_tokens": 2300107.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.265869903564453,
|
|
"epoch": 0.07802819731422256,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 7.2317,
|
|
"mean_token_accuracy": 0.1063105285167694,
|
|
"num_tokens": 2309755.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.330099105834961,
|
|
"epoch": 0.0783630822812364,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 7.3643,
|
|
"mean_token_accuracy": 0.09808892235159875,
|
|
"num_tokens": 2320042.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.272184801101685,
|
|
"epoch": 0.07869796724825022,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 7.2137,
|
|
"mean_token_accuracy": 0.1104245975613594,
|
|
"num_tokens": 2329592.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.326994514465332,
|
|
"epoch": 0.07903285221526406,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 7.2911,
|
|
"mean_token_accuracy": 0.1015694797039032,
|
|
"num_tokens": 2339219.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.2799012660980225,
|
|
"epoch": 0.07936773718227788,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 7.2485,
|
|
"mean_token_accuracy": 0.10349665880203247,
|
|
"num_tokens": 2349037.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.291835355758667,
|
|
"epoch": 0.07970262214929172,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 7.2575,
|
|
"mean_token_accuracy": 0.10751553550362587,
|
|
"num_tokens": 2359321.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.34235315322876,
|
|
"epoch": 0.08003750711630554,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 7.3263,
|
|
"mean_token_accuracy": 0.10246394127607346,
|
|
"num_tokens": 2368978.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.249216890335083,
|
|
"epoch": 0.08037239208331938,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 7.2286,
|
|
"mean_token_accuracy": 0.10601548030972481,
|
|
"num_tokens": 2378429.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.3063520908355715,
|
|
"epoch": 0.0807072770503332,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 7.2869,
|
|
"mean_token_accuracy": 0.10153080821037293,
|
|
"num_tokens": 2389711.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.228058338165283,
|
|
"epoch": 0.08104216201734704,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 7.2833,
|
|
"mean_token_accuracy": 0.10485092476010323,
|
|
"num_tokens": 2399207.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.281473827362061,
|
|
"epoch": 0.08137704698436087,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 7.2453,
|
|
"mean_token_accuracy": 0.10985753983259201,
|
|
"num_tokens": 2409236.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.30497670173645,
|
|
"epoch": 0.0817119319513747,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 7.2938,
|
|
"mean_token_accuracy": 0.1052240014076233,
|
|
"num_tokens": 2419588.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.2253070831298825,
|
|
"epoch": 0.08204681691838854,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 7.1701,
|
|
"mean_token_accuracy": 0.10711280703544616,
|
|
"num_tokens": 2429651.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.329334735870361,
|
|
"epoch": 0.08238170188540236,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 7.3662,
|
|
"mean_token_accuracy": 0.0996719591319561,
|
|
"num_tokens": 2439942.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.277449989318848,
|
|
"epoch": 0.0827165868524162,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 7.263,
|
|
"mean_token_accuracy": 0.10270665660500526,
|
|
"num_tokens": 2448843.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.296933031082153,
|
|
"epoch": 0.08305147181943003,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 7.2922,
|
|
"mean_token_accuracy": 0.10227823108434678,
|
|
"num_tokens": 2458878.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.26516489982605,
|
|
"epoch": 0.08338635678644386,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 7.1605,
|
|
"mean_token_accuracy": 0.1101699061691761,
|
|
"num_tokens": 2468527.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.2820967674255375,
|
|
"epoch": 0.08372124175345769,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 7.3152,
|
|
"mean_token_accuracy": 0.10319520086050034,
|
|
"num_tokens": 2478996.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.2530265808105465,
|
|
"epoch": 0.08405612672047152,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 7.1887,
|
|
"mean_token_accuracy": 0.10442061051726341,
|
|
"num_tokens": 2488697.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.187231397628784,
|
|
"epoch": 0.08439101168748535,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 7.2401,
|
|
"mean_token_accuracy": 0.10387319102883338,
|
|
"num_tokens": 2498590.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.274334239959717,
|
|
"epoch": 0.08472589665449919,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 7.1824,
|
|
"mean_token_accuracy": 0.10959140658378601,
|
|
"num_tokens": 2507824.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.244437408447266,
|
|
"epoch": 0.08506078162151301,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 7.2469,
|
|
"mean_token_accuracy": 0.10558498799800872,
|
|
"num_tokens": 2516738.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.3016125679016115,
|
|
"epoch": 0.08539566658852685,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 7.2459,
|
|
"mean_token_accuracy": 0.10852355360984803,
|
|
"num_tokens": 2525843.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.2142653465271,
|
|
"epoch": 0.08573055155554067,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 7.2187,
|
|
"mean_token_accuracy": 0.10861733108758927,
|
|
"num_tokens": 2534877.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.3023340702056885,
|
|
"epoch": 0.08606543652255451,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 7.2457,
|
|
"mean_token_accuracy": 0.10270560383796692,
|
|
"num_tokens": 2545197.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.228833866119385,
|
|
"epoch": 0.08640032148956833,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 7.2251,
|
|
"mean_token_accuracy": 0.11042040884494782,
|
|
"num_tokens": 2555732.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.210680150985718,
|
|
"epoch": 0.08673520645658217,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 7.2745,
|
|
"mean_token_accuracy": 0.1047789104282856,
|
|
"num_tokens": 2565779.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.2319292545318605,
|
|
"epoch": 0.087070091423596,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 7.173,
|
|
"mean_token_accuracy": 0.11112685948610306,
|
|
"num_tokens": 2574822.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.255096387863159,
|
|
"epoch": 0.08740497639060983,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 7.2847,
|
|
"mean_token_accuracy": 0.10384105294942855,
|
|
"num_tokens": 2584780.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.21968297958374,
|
|
"epoch": 0.08773986135762366,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 7.2825,
|
|
"mean_token_accuracy": 0.10405526086688041,
|
|
"num_tokens": 2595080.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.281798458099365,
|
|
"epoch": 0.08807474632463749,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 7.3151,
|
|
"mean_token_accuracy": 0.10257799997925758,
|
|
"num_tokens": 2605221.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.314059209823609,
|
|
"epoch": 0.08840963129165132,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 7.2323,
|
|
"mean_token_accuracy": 0.10295548290014267,
|
|
"num_tokens": 2615915.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.2096466541290285,
|
|
"epoch": 0.08874451625866515,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 7.1915,
|
|
"mean_token_accuracy": 0.10639428049325943,
|
|
"num_tokens": 2626791.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.226166772842407,
|
|
"epoch": 0.08907940122567898,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 7.2656,
|
|
"mean_token_accuracy": 0.10837312117218971,
|
|
"num_tokens": 2636055.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.297124433517456,
|
|
"epoch": 0.08941428619269282,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 7.1916,
|
|
"mean_token_accuracy": 0.10522143691778182,
|
|
"num_tokens": 2645177.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.187276220321655,
|
|
"epoch": 0.08974917115970664,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 7.1588,
|
|
"mean_token_accuracy": 0.11183756738901138,
|
|
"num_tokens": 2655961.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.148924684524536,
|
|
"epoch": 0.09008405612672048,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 7.1541,
|
|
"mean_token_accuracy": 0.11793737411499024,
|
|
"num_tokens": 2666394.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.216915845870972,
|
|
"epoch": 0.0904189410937343,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 7.1927,
|
|
"mean_token_accuracy": 0.10960348248481751,
|
|
"num_tokens": 2675309.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.186840438842774,
|
|
"epoch": 0.09075382606074814,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 7.1826,
|
|
"mean_token_accuracy": 0.10617644041776657,
|
|
"num_tokens": 2685500.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.246086359024048,
|
|
"epoch": 0.09108871102776196,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 7.1035,
|
|
"mean_token_accuracy": 0.10482602342963218,
|
|
"num_tokens": 2694648.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.174089431762695,
|
|
"epoch": 0.0914235959947758,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 7.2129,
|
|
"mean_token_accuracy": 0.10653489902615547,
|
|
"num_tokens": 2703878.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.275624847412109,
|
|
"epoch": 0.09175848096178962,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.10273868069052697,
|
|
"num_tokens": 2714142.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.172191524505616,
|
|
"epoch": 0.09209336592880346,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 7.1153,
|
|
"mean_token_accuracy": 0.11495553776621818,
|
|
"num_tokens": 2723083.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.233070755004883,
|
|
"epoch": 0.09242825089581728,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 7.2819,
|
|
"mean_token_accuracy": 0.10039968714118004,
|
|
"num_tokens": 2733577.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.328128337860107,
|
|
"epoch": 0.09276313586283112,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 7.2459,
|
|
"mean_token_accuracy": 0.0923890821635723,
|
|
"num_tokens": 2744134.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.1128767967224125,
|
|
"epoch": 0.09309802082984495,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 7.1178,
|
|
"mean_token_accuracy": 0.10745147988200188,
|
|
"num_tokens": 2754460.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.22829155921936,
|
|
"epoch": 0.09343290579685878,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 7.2128,
|
|
"mean_token_accuracy": 0.10378189608454705,
|
|
"num_tokens": 2763935.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.211243963241577,
|
|
"epoch": 0.09376779076387261,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 7.1907,
|
|
"mean_token_accuracy": 0.10557325407862664,
|
|
"num_tokens": 2774945.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.171038055419922,
|
|
"epoch": 0.09410267573088645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 7.2125,
|
|
"mean_token_accuracy": 0.10536214634776116,
|
|
"num_tokens": 2784799.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.0606297016143795,
|
|
"epoch": 0.09443756069790027,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 7.0165,
|
|
"mean_token_accuracy": 0.11608987599611283,
|
|
"num_tokens": 2794506.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.236382102966308,
|
|
"epoch": 0.0947724456649141,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 7.276,
|
|
"mean_token_accuracy": 0.1041162945330143,
|
|
"num_tokens": 2804547.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.113600969314575,
|
|
"epoch": 0.09510733063192793,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 7.1121,
|
|
"mean_token_accuracy": 0.10910582020878792,
|
|
"num_tokens": 2815279.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.247256755828857,
|
|
"epoch": 0.09544221559894177,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 7.185,
|
|
"mean_token_accuracy": 0.09964644908905029,
|
|
"num_tokens": 2825754.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.25377254486084,
|
|
"epoch": 0.09577710056595559,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 7.3647,
|
|
"mean_token_accuracy": 0.10228754505515099,
|
|
"num_tokens": 2835775.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.1915538787841795,
|
|
"epoch": 0.09611198553296943,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 7.0842,
|
|
"mean_token_accuracy": 0.10943688377737999,
|
|
"num_tokens": 2845065.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.218343782424927,
|
|
"epoch": 0.09644687049998325,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 7.2454,
|
|
"mean_token_accuracy": 0.10062536150217057,
|
|
"num_tokens": 2855493.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.202933025360108,
|
|
"epoch": 0.09678175546699709,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 7.0416,
|
|
"mean_token_accuracy": 0.11148291304707528,
|
|
"num_tokens": 2864449.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.00218358039856,
|
|
"epoch": 0.09711664043401091,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 7.0897,
|
|
"mean_token_accuracy": 0.11226318851113319,
|
|
"num_tokens": 2874056.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.200784492492676,
|
|
"epoch": 0.09745152540102475,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 7.1136,
|
|
"mean_token_accuracy": 0.10861515179276467,
|
|
"num_tokens": 2884283.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.221148538589477,
|
|
"epoch": 0.09778641036803858,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 7.1259,
|
|
"mean_token_accuracy": 0.11199686899781228,
|
|
"num_tokens": 2892720.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.0525695323944095,
|
|
"epoch": 0.09812129533505241,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 7.1113,
|
|
"mean_token_accuracy": 0.10836914032697678,
|
|
"num_tokens": 2902671.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.207837247848511,
|
|
"epoch": 0.09845618030206624,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 7.1715,
|
|
"mean_token_accuracy": 0.10252555385231972,
|
|
"num_tokens": 2912683.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.156357526779175,
|
|
"epoch": 0.09879106526908008,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 7.1258,
|
|
"mean_token_accuracy": 0.10757871195673943,
|
|
"num_tokens": 2922888.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.196430444717407,
|
|
"epoch": 0.0991259502360939,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 7.1284,
|
|
"mean_token_accuracy": 0.10644729062914848,
|
|
"num_tokens": 2932284.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.186505126953125,
|
|
"epoch": 0.09946083520310774,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 7.1535,
|
|
"mean_token_accuracy": 0.11216149106621742,
|
|
"num_tokens": 2942061.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.131209659576416,
|
|
"epoch": 0.09979572017012156,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 7.0322,
|
|
"mean_token_accuracy": 0.11146841198205948,
|
|
"num_tokens": 2950216.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.070336866378784,
|
|
"epoch": 0.1001306051371354,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 7.0925,
|
|
"mean_token_accuracy": 0.11065334677696229,
|
|
"num_tokens": 2959578.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.151112222671509,
|
|
"epoch": 0.10046549010414922,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.11227871030569077,
|
|
"num_tokens": 2969043.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.07126636505127,
|
|
"epoch": 0.10080037507116306,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 7.0899,
|
|
"mean_token_accuracy": 0.11176811680197715,
|
|
"num_tokens": 2978765.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.129266357421875,
|
|
"epoch": 0.10113526003817688,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 7.0906,
|
|
"mean_token_accuracy": 0.11375992447137832,
|
|
"num_tokens": 2988878.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.214033937454223,
|
|
"epoch": 0.10147014500519072,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 7.1617,
|
|
"mean_token_accuracy": 0.10871290341019631,
|
|
"num_tokens": 2999048.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.157622814178467,
|
|
"epoch": 0.10180502997220454,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 7.0465,
|
|
"mean_token_accuracy": 0.11121024414896966,
|
|
"num_tokens": 3008847.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.073071575164795,
|
|
"epoch": 0.10213991493921838,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 7.0971,
|
|
"mean_token_accuracy": 0.10889534875750542,
|
|
"num_tokens": 3018514.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.136441946029663,
|
|
"epoch": 0.1024747999062322,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 7.0224,
|
|
"mean_token_accuracy": 0.10855127573013305,
|
|
"num_tokens": 3029096.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.046940040588379,
|
|
"epoch": 0.10280968487324604,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 7.1024,
|
|
"mean_token_accuracy": 0.11343139931559562,
|
|
"num_tokens": 3037163.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.228511238098145,
|
|
"epoch": 0.10314456984025987,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 7.1733,
|
|
"mean_token_accuracy": 0.10640680566430091,
|
|
"num_tokens": 3046763.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.183480834960937,
|
|
"epoch": 0.1034794548072737,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 7.0105,
|
|
"mean_token_accuracy": 0.11388382688164711,
|
|
"num_tokens": 3056250.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.958459711074829,
|
|
"epoch": 0.10381433977428753,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 7.0395,
|
|
"mean_token_accuracy": 0.10873231142759324,
|
|
"num_tokens": 3067238.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.212129783630371,
|
|
"epoch": 0.10414922474130137,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 7.1779,
|
|
"mean_token_accuracy": 0.10753423050045967,
|
|
"num_tokens": 3077479.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.078889799118042,
|
|
"epoch": 0.10448410970831519,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 7.1183,
|
|
"mean_token_accuracy": 0.11443167254328727,
|
|
"num_tokens": 3086639.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.310558414459228,
|
|
"epoch": 0.10481899467532903,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 7.1741,
|
|
"mean_token_accuracy": 0.10301887020468711,
|
|
"num_tokens": 3096501.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.1115892887115475,
|
|
"epoch": 0.10515387964234285,
|
|
"grad_norm": 2.390625,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 7.0055,
|
|
"mean_token_accuracy": 0.11831002086400985,
|
|
"num_tokens": 3106026.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.11304988861084,
|
|
"epoch": 0.10548876460935669,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 7.0505,
|
|
"mean_token_accuracy": 0.11222735717892647,
|
|
"num_tokens": 3114744.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.051252317428589,
|
|
"epoch": 0.10582364957637051,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 7.0637,
|
|
"mean_token_accuracy": 0.1122724510729313,
|
|
"num_tokens": 3123996.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.191004085540771,
|
|
"epoch": 0.10615853454338435,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 7.0931,
|
|
"mean_token_accuracy": 0.11069929674267769,
|
|
"num_tokens": 3133301.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.092730140686035,
|
|
"epoch": 0.10649341951039817,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 7.0279,
|
|
"mean_token_accuracy": 0.11370506212115288,
|
|
"num_tokens": 3142505.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.134963846206665,
|
|
"epoch": 0.10682830447741201,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 7.0849,
|
|
"mean_token_accuracy": 0.10752070397138595,
|
|
"num_tokens": 3152858.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.055032539367676,
|
|
"epoch": 0.10716318944442584,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 7.0931,
|
|
"mean_token_accuracy": 0.10625340268015862,
|
|
"num_tokens": 3163241.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.144267892837524,
|
|
"epoch": 0.10749807441143967,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 7.0218,
|
|
"mean_token_accuracy": 0.11668949276208877,
|
|
"num_tokens": 3172461.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.057622337341309,
|
|
"epoch": 0.1078329593784535,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 7.0481,
|
|
"mean_token_accuracy": 0.11361603289842606,
|
|
"num_tokens": 3181686.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.912009286880493,
|
|
"epoch": 0.10816784434546733,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.9843,
|
|
"mean_token_accuracy": 0.1171471931040287,
|
|
"num_tokens": 3191123.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.154270648956299,
|
|
"epoch": 0.10850272931248116,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.9257,
|
|
"mean_token_accuracy": 0.11631628349423409,
|
|
"num_tokens": 3200346.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.113286304473877,
|
|
"epoch": 0.108837614279495,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 7.1185,
|
|
"mean_token_accuracy": 0.11047174260020257,
|
|
"num_tokens": 3211345.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.09348497390747,
|
|
"epoch": 0.10917249924650882,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 7.1419,
|
|
"mean_token_accuracy": 0.10584981217980385,
|
|
"num_tokens": 3221621.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.246621990203858,
|
|
"epoch": 0.10950738421352266,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 7.1149,
|
|
"mean_token_accuracy": 0.10783747285604477,
|
|
"num_tokens": 3231535.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.982515430450439,
|
|
"epoch": 0.10984226918053648,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.9956,
|
|
"mean_token_accuracy": 0.11259618178009986,
|
|
"num_tokens": 3241649.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.1504510879516605,
|
|
"epoch": 0.11017715414755032,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 7.0252,
|
|
"mean_token_accuracy": 0.11361673399806023,
|
|
"num_tokens": 3251690.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.053209447860718,
|
|
"epoch": 0.11051203911456414,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.9969,
|
|
"mean_token_accuracy": 0.11527483984827995,
|
|
"num_tokens": 3260852.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.068172883987427,
|
|
"epoch": 0.11084692408157798,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 7.0595,
|
|
"mean_token_accuracy": 0.1096741497516632,
|
|
"num_tokens": 3270695.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.072029256820679,
|
|
"epoch": 0.1111818090485918,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.9683,
|
|
"mean_token_accuracy": 0.10862559601664543,
|
|
"num_tokens": 3279417.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.998540782928467,
|
|
"epoch": 0.11151669401560564,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 7.0489,
|
|
"mean_token_accuracy": 0.10989508256316186,
|
|
"num_tokens": 3289787.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.159190273284912,
|
|
"epoch": 0.11185157898261947,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.9637,
|
|
"mean_token_accuracy": 0.11042561084032058,
|
|
"num_tokens": 3298454.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.955928802490234,
|
|
"epoch": 0.1121864639496333,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.9832,
|
|
"mean_token_accuracy": 0.11283597946166993,
|
|
"num_tokens": 3308794.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.128731632232666,
|
|
"epoch": 0.11252134891664713,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 7.065,
|
|
"mean_token_accuracy": 0.1055634468793869,
|
|
"num_tokens": 3318371.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.0393671035766605,
|
|
"epoch": 0.11285623388366096,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 7.0061,
|
|
"mean_token_accuracy": 0.11146225556731224,
|
|
"num_tokens": 3327717.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.034864616394043,
|
|
"epoch": 0.11319111885067479,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.9333,
|
|
"mean_token_accuracy": 0.11891898959875107,
|
|
"num_tokens": 3337188.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.019638633728027,
|
|
"epoch": 0.11352600381768863,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.9712,
|
|
"mean_token_accuracy": 0.11413623094558716,
|
|
"num_tokens": 3345665.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.039876556396484,
|
|
"epoch": 0.11386088878470245,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 7.0175,
|
|
"mean_token_accuracy": 0.10979360565543175,
|
|
"num_tokens": 3355178.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.1303102493286135,
|
|
"epoch": 0.11419577375171629,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 7.0396,
|
|
"mean_token_accuracy": 0.11044175028800965,
|
|
"num_tokens": 3364379.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.106704568862915,
|
|
"epoch": 0.11453065871873011,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 7.016,
|
|
"mean_token_accuracy": 0.11462057530879974,
|
|
"num_tokens": 3374187.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.016075277328492,
|
|
"epoch": 0.11486554368574395,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.9737,
|
|
"mean_token_accuracy": 0.11734921336174012,
|
|
"num_tokens": 3383602.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.07158637046814,
|
|
"epoch": 0.11520042865275777,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 7.0599,
|
|
"mean_token_accuracy": 0.11306147575378418,
|
|
"num_tokens": 3395542.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.086615419387817,
|
|
"epoch": 0.11553531361977161,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 7.0251,
|
|
"mean_token_accuracy": 0.11459362357854844,
|
|
"num_tokens": 3404528.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.106108665466309,
|
|
"epoch": 0.11587019858678543,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 7.0324,
|
|
"mean_token_accuracy": 0.10542695000767707,
|
|
"num_tokens": 3414525.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.0047276496887205,
|
|
"epoch": 0.11620508355379927,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.9753,
|
|
"mean_token_accuracy": 0.11097218319773675,
|
|
"num_tokens": 3424743.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.031596994400024,
|
|
"epoch": 0.1165399685208131,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.9906,
|
|
"mean_token_accuracy": 0.1132536143064499,
|
|
"num_tokens": 3435516.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.027150964736938,
|
|
"epoch": 0.11687485348782693,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 7.0282,
|
|
"mean_token_accuracy": 0.10911838114261627,
|
|
"num_tokens": 3445017.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.069865942001343,
|
|
"epoch": 0.11720973845484076,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.9641,
|
|
"mean_token_accuracy": 0.11882567778229713,
|
|
"num_tokens": 3454200.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.010887384414673,
|
|
"epoch": 0.1175446234218546,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.9487,
|
|
"mean_token_accuracy": 0.11610497310757636,
|
|
"num_tokens": 3464631.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.955662679672241,
|
|
"epoch": 0.11787950838886842,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.8773,
|
|
"mean_token_accuracy": 0.1225242219865322,
|
|
"num_tokens": 3474564.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.984214448928833,
|
|
"epoch": 0.11821439335588226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.9992,
|
|
"mean_token_accuracy": 0.11146586537361144,
|
|
"num_tokens": 3485464.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.085714197158813,
|
|
"epoch": 0.11854927832289608,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.9085,
|
|
"mean_token_accuracy": 0.11654480174183846,
|
|
"num_tokens": 3495194.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.960865259170532,
|
|
"epoch": 0.11888416328990992,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.9723,
|
|
"mean_token_accuracy": 0.11530765071511269,
|
|
"num_tokens": 3505737.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.033911085128784,
|
|
"epoch": 0.11921904825692374,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 7.0055,
|
|
"mean_token_accuracy": 0.11757232770323753,
|
|
"num_tokens": 3515520.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.031638526916504,
|
|
"epoch": 0.11955393322393758,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.8168,
|
|
"mean_token_accuracy": 0.11526884287595748,
|
|
"num_tokens": 3524795.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.886267566680909,
|
|
"epoch": 0.1198888181909514,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.8974,
|
|
"mean_token_accuracy": 0.12006916925311088,
|
|
"num_tokens": 3535022.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.068918228149414,
|
|
"epoch": 0.12022370315796524,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 7.0184,
|
|
"mean_token_accuracy": 0.11066200956702232,
|
|
"num_tokens": 3544560.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 7.047748041152954,
|
|
"epoch": 0.12055858812497906,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.9715,
|
|
"mean_token_accuracy": 0.11284061819314957,
|
|
"num_tokens": 3554298.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.993912839889527,
|
|
"epoch": 0.1208934730919929,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.9752,
|
|
"mean_token_accuracy": 0.11206858605146408,
|
|
"num_tokens": 3563574.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.031573247909546,
|
|
"epoch": 0.12122835805900672,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 7.0121,
|
|
"mean_token_accuracy": 0.10989024490118027,
|
|
"num_tokens": 3573529.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.085992765426636,
|
|
"epoch": 0.12156324302602056,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 7.0094,
|
|
"mean_token_accuracy": 0.11554570496082306,
|
|
"num_tokens": 3583604.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.971385431289673,
|
|
"epoch": 0.12189812799303439,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.11939797103404999,
|
|
"num_tokens": 3593341.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.971127367019653,
|
|
"epoch": 0.12223301296004822,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 7.0266,
|
|
"mean_token_accuracy": 0.1068960227072239,
|
|
"num_tokens": 3603654.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.259721422195435,
|
|
"epoch": 0.12256789792706206,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 7.151,
|
|
"mean_token_accuracy": 0.09942766651511192,
|
|
"num_tokens": 3613651.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.01947603225708,
|
|
"epoch": 0.12290278289407588,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.9189,
|
|
"mean_token_accuracy": 0.11156605258584022,
|
|
"num_tokens": 3623860.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.9894308090209964,
|
|
"epoch": 0.12323766786108972,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.9501,
|
|
"mean_token_accuracy": 0.11726592108607292,
|
|
"num_tokens": 3633614.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.994549131393432,
|
|
"epoch": 0.12357255282810355,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.11003849059343337,
|
|
"num_tokens": 3644282.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.083448171615601,
|
|
"epoch": 0.12390743779511738,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 7.0166,
|
|
"mean_token_accuracy": 0.11039167046546935,
|
|
"num_tokens": 3653572.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.020518732070923,
|
|
"epoch": 0.12424232276213121,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.8674,
|
|
"mean_token_accuracy": 0.1183643437922001,
|
|
"num_tokens": 3664732.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.974581718444824,
|
|
"epoch": 0.12457720772914505,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.9522,
|
|
"mean_token_accuracy": 0.11369702592492104,
|
|
"num_tokens": 3674457.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.985838460922241,
|
|
"epoch": 0.12491209269615887,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.865,
|
|
"mean_token_accuracy": 0.11783641502261162,
|
|
"num_tokens": 3683754.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.998713445663452,
|
|
"epoch": 0.1252469776631727,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.9441,
|
|
"mean_token_accuracy": 0.11969761848449707,
|
|
"num_tokens": 3692529.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.0662554740905765,
|
|
"epoch": 0.12558186263018653,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.9909,
|
|
"mean_token_accuracy": 0.11368770599365234,
|
|
"num_tokens": 3702013.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.048868989944458,
|
|
"epoch": 0.12591674759720037,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.9005,
|
|
"mean_token_accuracy": 0.11901575177907944,
|
|
"num_tokens": 3712548.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.905255222320557,
|
|
"epoch": 0.1262516325642142,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.8767,
|
|
"mean_token_accuracy": 0.1174386665225029,
|
|
"num_tokens": 3722307.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.953826427459717,
|
|
"epoch": 0.12658651753122802,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.8451,
|
|
"mean_token_accuracy": 0.12337855622172356,
|
|
"num_tokens": 3732665.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.961658620834351,
|
|
"epoch": 0.12692140249824185,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.9002,
|
|
"mean_token_accuracy": 0.11729438528418541,
|
|
"num_tokens": 3742667.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.926886653900146,
|
|
"epoch": 0.1272562874652557,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.8903,
|
|
"mean_token_accuracy": 0.1160149559378624,
|
|
"num_tokens": 3753961.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 7.047531175613403,
|
|
"epoch": 0.12759117243226953,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 7.0601,
|
|
"mean_token_accuracy": 0.11376734524965286,
|
|
"num_tokens": 3763493.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.049573469161987,
|
|
"epoch": 0.12792605739928334,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.8542,
|
|
"mean_token_accuracy": 0.11631640791893005,
|
|
"num_tokens": 3772673.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.898791646957397,
|
|
"epoch": 0.12826094236629718,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.9467,
|
|
"mean_token_accuracy": 0.11332176104187966,
|
|
"num_tokens": 3782601.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.967033576965332,
|
|
"epoch": 0.128595827333311,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.8646,
|
|
"mean_token_accuracy": 0.12189628630876541,
|
|
"num_tokens": 3793277.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.094461297988891,
|
|
"epoch": 0.12893071230032485,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.9216,
|
|
"mean_token_accuracy": 0.11160155087709427,
|
|
"num_tokens": 3802882.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 7.000693368911743,
|
|
"epoch": 0.12926559726733866,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.8803,
|
|
"mean_token_accuracy": 0.11363799199461937,
|
|
"num_tokens": 3812520.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.914379072189331,
|
|
"epoch": 0.1296004822343525,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.8686,
|
|
"mean_token_accuracy": 0.11704775542020798,
|
|
"num_tokens": 3822943.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.935992240905762,
|
|
"epoch": 0.12993536720136634,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.8581,
|
|
"mean_token_accuracy": 0.11949142143130302,
|
|
"num_tokens": 3832762.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.948669004440307,
|
|
"epoch": 0.13027025216838017,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.848,
|
|
"mean_token_accuracy": 0.12257980108261109,
|
|
"num_tokens": 3842631.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.975763320922852,
|
|
"epoch": 0.13060513713539398,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.925,
|
|
"mean_token_accuracy": 0.1129115492105484,
|
|
"num_tokens": 3852508.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.912820339202881,
|
|
"epoch": 0.13094002210240782,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.7904,
|
|
"mean_token_accuracy": 0.12319999411702157,
|
|
"num_tokens": 3861305.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.938683223724365,
|
|
"epoch": 0.13127490706942166,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.9031,
|
|
"mean_token_accuracy": 0.10886179655790329,
|
|
"num_tokens": 3870786.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.972001314163208,
|
|
"epoch": 0.1316097920364355,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.8129,
|
|
"mean_token_accuracy": 0.11861697286367416,
|
|
"num_tokens": 3879723.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.981768417358398,
|
|
"epoch": 0.1319446770034493,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.8908,
|
|
"mean_token_accuracy": 0.12219975739717484,
|
|
"num_tokens": 3888755.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.862844944000244,
|
|
"epoch": 0.13227956197046314,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.8609,
|
|
"mean_token_accuracy": 0.114610805362463,
|
|
"num_tokens": 3898543.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.927087640762329,
|
|
"epoch": 0.13261444693747698,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.8759,
|
|
"mean_token_accuracy": 0.1202179692685604,
|
|
"num_tokens": 3909130.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.91467547416687,
|
|
"epoch": 0.13294933190449082,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.7716,
|
|
"mean_token_accuracy": 0.12905031815171242,
|
|
"num_tokens": 3918214.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.910163879394531,
|
|
"epoch": 0.13328421687150463,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.8678,
|
|
"mean_token_accuracy": 0.1201781801879406,
|
|
"num_tokens": 3927629.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.862524509429932,
|
|
"epoch": 0.13361910183851847,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.8341,
|
|
"mean_token_accuracy": 0.11810301542282105,
|
|
"num_tokens": 3937048.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 7.062771224975586,
|
|
"epoch": 0.1339539868055323,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.9325,
|
|
"mean_token_accuracy": 0.11639236509799958,
|
|
"num_tokens": 3947615.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 7.0152489185333256,
|
|
"epoch": 0.13428887177254614,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 6.8097,
|
|
"mean_token_accuracy": 0.12037209495902061,
|
|
"num_tokens": 3957430.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.870607852935791,
|
|
"epoch": 0.13462375673955995,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 6.8064,
|
|
"mean_token_accuracy": 0.12359208092093468,
|
|
"num_tokens": 3968690.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.935949039459229,
|
|
"epoch": 0.1349586417065738,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 7.023,
|
|
"mean_token_accuracy": 0.11046545654535293,
|
|
"num_tokens": 3977933.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.983598899841309,
|
|
"epoch": 0.13529352667358763,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 6.7921,
|
|
"mean_token_accuracy": 0.11740928962826729,
|
|
"num_tokens": 3987164.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 7.006376457214356,
|
|
"epoch": 0.13562841164060147,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 6.8961,
|
|
"mean_token_accuracy": 0.11724071651697159,
|
|
"num_tokens": 3996918.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.911764287948609,
|
|
"epoch": 0.13596329660761527,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 6.8712,
|
|
"mean_token_accuracy": 0.11954173818230629,
|
|
"num_tokens": 4006995.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 7.045343685150146,
|
|
"epoch": 0.1362981815746291,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 6.9961,
|
|
"mean_token_accuracy": 0.10837767869234086,
|
|
"num_tokens": 4017263.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.982536745071411,
|
|
"epoch": 0.13663306654164295,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 6.8525,
|
|
"mean_token_accuracy": 0.1231170006096363,
|
|
"num_tokens": 4026851.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.881824588775634,
|
|
"epoch": 0.1369679515086568,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 6.8325,
|
|
"mean_token_accuracy": 0.12139289081096649,
|
|
"num_tokens": 4037216.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 7.082671356201172,
|
|
"epoch": 0.1373028364756706,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 6.882,
|
|
"mean_token_accuracy": 0.11201102063059806,
|
|
"num_tokens": 4047997.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.906302165985108,
|
|
"epoch": 0.13763772144268444,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 6.7949,
|
|
"mean_token_accuracy": 0.12172465845942497,
|
|
"num_tokens": 4057611.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.788911533355713,
|
|
"epoch": 0.13797260640969827,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 6.7323,
|
|
"mean_token_accuracy": 0.1269633650779724,
|
|
"num_tokens": 4067162.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 7.006991052627564,
|
|
"epoch": 0.1383074913767121,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 6.9344,
|
|
"mean_token_accuracy": 0.1172435287386179,
|
|
"num_tokens": 4075821.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.994809246063232,
|
|
"epoch": 0.13864237634372592,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 6.8232,
|
|
"mean_token_accuracy": 0.11947978660464287,
|
|
"num_tokens": 4085229.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.859807014465332,
|
|
"epoch": 0.13897726131073976,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 6.8603,
|
|
"mean_token_accuracy": 0.11699308604001998,
|
|
"num_tokens": 4095015.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.953826522827148,
|
|
"epoch": 0.1393121462777536,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 6.9417,
|
|
"mean_token_accuracy": 0.11828644946217537,
|
|
"num_tokens": 4105043.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.98751130104065,
|
|
"epoch": 0.13964703124476743,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 6.8126,
|
|
"mean_token_accuracy": 0.11838952526450157,
|
|
"num_tokens": 4114798.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.94891128540039,
|
|
"epoch": 0.13998191621178124,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 6.9153,
|
|
"mean_token_accuracy": 0.11310055032372475,
|
|
"num_tokens": 4124528.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.954400062561035,
|
|
"epoch": 0.14031680117879508,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 6.8028,
|
|
"mean_token_accuracy": 0.12207239344716073,
|
|
"num_tokens": 4134615.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.973540353775024,
|
|
"epoch": 0.14065168614580892,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.8371,
|
|
"mean_token_accuracy": 0.11877345144748688,
|
|
"num_tokens": 4144923.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.875203943252563,
|
|
"epoch": 0.14098657111282276,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 6.7978,
|
|
"mean_token_accuracy": 0.12377667054533958,
|
|
"num_tokens": 4153805.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.967620134353638,
|
|
"epoch": 0.14132145607983657,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 6.9717,
|
|
"mean_token_accuracy": 0.11000535413622856,
|
|
"num_tokens": 4164019.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 7.187637710571289,
|
|
"epoch": 0.1416563410468504,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 6.9861,
|
|
"mean_token_accuracy": 0.11090470850467682,
|
|
"num_tokens": 4174937.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.919124603271484,
|
|
"epoch": 0.14199122601386424,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 6.8021,
|
|
"mean_token_accuracy": 0.11856020465493203,
|
|
"num_tokens": 4184890.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.9292703628540036,
|
|
"epoch": 0.14232611098087808,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 6.8861,
|
|
"mean_token_accuracy": 0.11611267998814583,
|
|
"num_tokens": 4194879.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.931923198699951,
|
|
"epoch": 0.1426609959478919,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 6.8132,
|
|
"mean_token_accuracy": 0.12112772390246392,
|
|
"num_tokens": 4205438.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.931875848770142,
|
|
"epoch": 0.14299588091490573,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.9146,
|
|
"mean_token_accuracy": 0.11795324608683586,
|
|
"num_tokens": 4215344.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.962862920761109,
|
|
"epoch": 0.14333076588191956,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 6.7975,
|
|
"mean_token_accuracy": 0.11945817843079568,
|
|
"num_tokens": 4224788.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.956465482711792,
|
|
"epoch": 0.1436656508489334,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 6.829,
|
|
"mean_token_accuracy": 0.11241990551352501,
|
|
"num_tokens": 4234553.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.921896600723267,
|
|
"epoch": 0.1440005358159472,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 6.8248,
|
|
"mean_token_accuracy": 0.11565629690885544,
|
|
"num_tokens": 4246056.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 7.010245037078858,
|
|
"epoch": 0.14433542078296105,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 6.8635,
|
|
"mean_token_accuracy": 0.11582673490047454,
|
|
"num_tokens": 4256444.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.949187517166138,
|
|
"epoch": 0.1446703057499749,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 6.8588,
|
|
"mean_token_accuracy": 0.11878125369548798,
|
|
"num_tokens": 4266317.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.899581384658814,
|
|
"epoch": 0.14500519071698872,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 6.853,
|
|
"mean_token_accuracy": 0.11252575814723968,
|
|
"num_tokens": 4276489.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.956885910034179,
|
|
"epoch": 0.14534007568400253,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 6.752,
|
|
"mean_token_accuracy": 0.12423492670059204,
|
|
"num_tokens": 4286227.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.831689929962158,
|
|
"epoch": 0.14567496065101637,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 6.8009,
|
|
"mean_token_accuracy": 0.12221949622035026,
|
|
"num_tokens": 4296830.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.9344555854797365,
|
|
"epoch": 0.1460098456180302,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 6.8809,
|
|
"mean_token_accuracy": 0.11556645929813385,
|
|
"num_tokens": 4306127.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 7.055763053894043,
|
|
"epoch": 0.14634473058504405,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 6.7785,
|
|
"mean_token_accuracy": 0.12510929331183435,
|
|
"num_tokens": 4315383.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.827849006652832,
|
|
"epoch": 0.14667961555205786,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 6.7213,
|
|
"mean_token_accuracy": 0.12449167892336846,
|
|
"num_tokens": 4325781.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 7.007912826538086,
|
|
"epoch": 0.1470145005190717,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 6.8945,
|
|
"mean_token_accuracy": 0.11278351321816445,
|
|
"num_tokens": 4334817.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.941176557540894,
|
|
"epoch": 0.14734938548608553,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 6.7884,
|
|
"mean_token_accuracy": 0.12627619057893752,
|
|
"num_tokens": 4344255.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.870765495300293,
|
|
"epoch": 0.14768427045309937,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 6.7868,
|
|
"mean_token_accuracy": 0.1239921011030674,
|
|
"num_tokens": 4353350.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.907273149490356,
|
|
"epoch": 0.14801915542011318,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 6.7844,
|
|
"mean_token_accuracy": 0.11687748283147811,
|
|
"num_tokens": 4363616.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.857398796081543,
|
|
"epoch": 0.14835404038712702,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 6.7562,
|
|
"mean_token_accuracy": 0.11924734488129615,
|
|
"num_tokens": 4372675.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.899632787704467,
|
|
"epoch": 0.14868892535414086,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 6.7947,
|
|
"mean_token_accuracy": 0.12011732533574104,
|
|
"num_tokens": 4383353.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.965082883834839,
|
|
"epoch": 0.1490238103211547,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 6.8141,
|
|
"mean_token_accuracy": 0.11718386486172676,
|
|
"num_tokens": 4393921.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.814152336120605,
|
|
"epoch": 0.1493586952881685,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 6.8036,
|
|
"mean_token_accuracy": 0.12445863336324692,
|
|
"num_tokens": 4402499.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.908452796936035,
|
|
"epoch": 0.14969358025518234,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 6.7534,
|
|
"mean_token_accuracy": 0.13187158554792405,
|
|
"num_tokens": 4412076.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 7.001694250106811,
|
|
"epoch": 0.15002846522219618,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 6.7998,
|
|
"mean_token_accuracy": 0.12347555607557296,
|
|
"num_tokens": 4421668.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.822030305862427,
|
|
"epoch": 0.15036335018921002,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 6.6848,
|
|
"mean_token_accuracy": 0.12389474511146545,
|
|
"num_tokens": 4430581.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.82412576675415,
|
|
"epoch": 0.15069823515622383,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 6.8022,
|
|
"mean_token_accuracy": 0.11891614273190498,
|
|
"num_tokens": 4441694.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.908269643783569,
|
|
"epoch": 0.15103312012323766,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 6.8176,
|
|
"mean_token_accuracy": 0.12239727601408959,
|
|
"num_tokens": 4451379.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.839193725585938,
|
|
"epoch": 0.1513680050902515,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 6.6092,
|
|
"mean_token_accuracy": 0.1298626258969307,
|
|
"num_tokens": 4460133.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.887795448303223,
|
|
"epoch": 0.15170289005726534,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 6.8094,
|
|
"mean_token_accuracy": 0.11962502151727676,
|
|
"num_tokens": 4469737.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.947560119628906,
|
|
"epoch": 0.15203777502427915,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 6.7375,
|
|
"mean_token_accuracy": 0.12183001786470413,
|
|
"num_tokens": 4478769.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.870909881591797,
|
|
"epoch": 0.15237265999129299,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 6.7532,
|
|
"mean_token_accuracy": 0.1212999165058136,
|
|
"num_tokens": 4487259.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.8715455532073975,
|
|
"epoch": 0.15270754495830682,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 6.8108,
|
|
"mean_token_accuracy": 0.11858168765902519,
|
|
"num_tokens": 4496719.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.888361215591431,
|
|
"epoch": 0.15304242992532066,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.12574247270822525,
|
|
"num_tokens": 4506911.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.922957706451416,
|
|
"epoch": 0.15337731489233447,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 6.7656,
|
|
"mean_token_accuracy": 0.12273657768964767,
|
|
"num_tokens": 4517200.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.814663934707641,
|
|
"epoch": 0.1537121998593483,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 6.6935,
|
|
"mean_token_accuracy": 0.1277276173233986,
|
|
"num_tokens": 4525716.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.809919309616089,
|
|
"epoch": 0.15404708482636215,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 6.7201,
|
|
"mean_token_accuracy": 0.12629192546010018,
|
|
"num_tokens": 4534940.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.868072128295898,
|
|
"epoch": 0.15438196979337598,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 6.8202,
|
|
"mean_token_accuracy": 0.1203918308019638,
|
|
"num_tokens": 4544500.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.883258199691772,
|
|
"epoch": 0.1547168547603898,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 6.7057,
|
|
"mean_token_accuracy": 0.12403135448694229,
|
|
"num_tokens": 4553085.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.877315044403076,
|
|
"epoch": 0.15505173972740363,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 6.6414,
|
|
"mean_token_accuracy": 0.12905535995960235,
|
|
"num_tokens": 4562025.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.80784764289856,
|
|
"epoch": 0.15538662469441747,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 6.7272,
|
|
"mean_token_accuracy": 0.1252925232052803,
|
|
"num_tokens": 4571487.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.866438388824463,
|
|
"epoch": 0.1557215096614313,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 6.752,
|
|
"mean_token_accuracy": 0.12572527676820755,
|
|
"num_tokens": 4581735.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.933021211624146,
|
|
"epoch": 0.15605639462844512,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 6.743,
|
|
"mean_token_accuracy": 0.12504383623600007,
|
|
"num_tokens": 4591583.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.792250442504883,
|
|
"epoch": 0.15639127959545895,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 6.65,
|
|
"mean_token_accuracy": 0.13422503992915152,
|
|
"num_tokens": 4600785.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.798924398422241,
|
|
"epoch": 0.1567261645624728,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 6.7049,
|
|
"mean_token_accuracy": 0.12252919971942902,
|
|
"num_tokens": 4610198.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.916207885742187,
|
|
"epoch": 0.15706104952948663,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 6.708,
|
|
"mean_token_accuracy": 0.1256384737789631,
|
|
"num_tokens": 4618623.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.871882009506225,
|
|
"epoch": 0.15739593449650044,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 6.7224,
|
|
"mean_token_accuracy": 0.12050877287983894,
|
|
"num_tokens": 4627489.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.786143064498901,
|
|
"epoch": 0.15773081946351428,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 6.6627,
|
|
"mean_token_accuracy": 0.1244005523622036,
|
|
"num_tokens": 4636085.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.8336621761322025,
|
|
"epoch": 0.15806570443052811,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 6.8076,
|
|
"mean_token_accuracy": 0.12189308553934097,
|
|
"num_tokens": 4646743.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.9684895992279055,
|
|
"epoch": 0.15840058939754195,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 6.7745,
|
|
"mean_token_accuracy": 0.12153615579009056,
|
|
"num_tokens": 4656540.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.858010530471802,
|
|
"epoch": 0.15873547436455576,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 6.7227,
|
|
"mean_token_accuracy": 0.12424319982528687,
|
|
"num_tokens": 4666871.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.8657575130462645,
|
|
"epoch": 0.1590703593315696,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 6.8593,
|
|
"mean_token_accuracy": 0.12255802899599075,
|
|
"num_tokens": 4676956.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.944288158416748,
|
|
"epoch": 0.15940524429858344,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 6.7636,
|
|
"mean_token_accuracy": 0.11702195033431054,
|
|
"num_tokens": 4686678.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.960586786270142,
|
|
"epoch": 0.15974012926559727,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 6.8094,
|
|
"mean_token_accuracy": 0.12440174296498299,
|
|
"num_tokens": 4697301.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.818715524673462,
|
|
"epoch": 0.16007501423261108,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 6.7751,
|
|
"mean_token_accuracy": 0.11601589918136597,
|
|
"num_tokens": 4707350.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.9010858058929445,
|
|
"epoch": 0.16040989919962492,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 6.6889,
|
|
"mean_token_accuracy": 0.11968213543295861,
|
|
"num_tokens": 4717543.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.897911262512207,
|
|
"epoch": 0.16074478416663876,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 6.7427,
|
|
"mean_token_accuracy": 0.12274712696671486,
|
|
"num_tokens": 4727989.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.8522521495819095,
|
|
"epoch": 0.1610796691336526,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 6.7568,
|
|
"mean_token_accuracy": 0.12302400320768356,
|
|
"num_tokens": 4738307.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.829119443893433,
|
|
"epoch": 0.1614145541006664,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 6.6793,
|
|
"mean_token_accuracy": 0.12769056260585784,
|
|
"num_tokens": 4747634.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.853943252563477,
|
|
"epoch": 0.16174943906768025,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 6.6551,
|
|
"mean_token_accuracy": 0.1255294993519783,
|
|
"num_tokens": 4757132.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.829272603988647,
|
|
"epoch": 0.16208432403469408,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 6.6416,
|
|
"mean_token_accuracy": 0.12597506865859032,
|
|
"num_tokens": 4766994.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.802279901504517,
|
|
"epoch": 0.16241920900170792,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 6.679,
|
|
"mean_token_accuracy": 0.11738070249557495,
|
|
"num_tokens": 4777809.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.85419602394104,
|
|
"epoch": 0.16275409396872173,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 6.6607,
|
|
"mean_token_accuracy": 0.12683946564793586,
|
|
"num_tokens": 4787324.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.866635847091675,
|
|
"epoch": 0.16308897893573557,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 6.8143,
|
|
"mean_token_accuracy": 0.12405446171760559,
|
|
"num_tokens": 4797912.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.871827507019043,
|
|
"epoch": 0.1634238639027494,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 6.7162,
|
|
"mean_token_accuracy": 0.1233360968530178,
|
|
"num_tokens": 4807074.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.9017274379730225,
|
|
"epoch": 0.16375874886976324,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 6.6601,
|
|
"mean_token_accuracy": 0.12303908988833427,
|
|
"num_tokens": 4816464.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.804702043533325,
|
|
"epoch": 0.16409363383677708,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 6.6576,
|
|
"mean_token_accuracy": 0.12317432612180709,
|
|
"num_tokens": 4827114.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.784649467468261,
|
|
"epoch": 0.1644285188037909,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 6.6043,
|
|
"mean_token_accuracy": 0.13119509741663932,
|
|
"num_tokens": 4836989.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.833284568786621,
|
|
"epoch": 0.16476340377080473,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 6.6919,
|
|
"mean_token_accuracy": 0.12436783984303475,
|
|
"num_tokens": 4846538.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.839111614227295,
|
|
"epoch": 0.16509828873781857,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 6.6854,
|
|
"mean_token_accuracy": 0.12325471937656403,
|
|
"num_tokens": 4855401.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.898120355606079,
|
|
"epoch": 0.1654331737048324,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 6.854,
|
|
"mean_token_accuracy": 0.12041018605232238,
|
|
"num_tokens": 4864604.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.969759750366211,
|
|
"epoch": 0.1657680586718462,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 6.7646,
|
|
"mean_token_accuracy": 0.12048763707280159,
|
|
"num_tokens": 4874789.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.79827036857605,
|
|
"epoch": 0.16610294363886005,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 6.6983,
|
|
"mean_token_accuracy": 0.1236626923084259,
|
|
"num_tokens": 4886121.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.723689603805542,
|
|
"epoch": 0.1664378286058739,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 6.6166,
|
|
"mean_token_accuracy": 0.13452839255332946,
|
|
"num_tokens": 4895307.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.896350431442261,
|
|
"epoch": 0.16677271357288773,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 6.6957,
|
|
"mean_token_accuracy": 0.12570520788431166,
|
|
"num_tokens": 4905591.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.877748250961304,
|
|
"epoch": 0.16710759853990154,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 6.7398,
|
|
"mean_token_accuracy": 0.12403841838240623,
|
|
"num_tokens": 4914834.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.79852409362793,
|
|
"epoch": 0.16744248350691537,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 6.6806,
|
|
"mean_token_accuracy": 0.12610427811741828,
|
|
"num_tokens": 4925082.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.793560743331909,
|
|
"epoch": 0.1677773684739292,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002740575249600342,
|
|
"loss": 6.6622,
|
|
"mean_token_accuracy": 0.1314407117664814,
|
|
"num_tokens": 4933958.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.86600661277771,
|
|
"epoch": 0.16811225344094305,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00027287945635254263,
|
|
"loss": 6.6227,
|
|
"mean_token_accuracy": 0.12924253568053246,
|
|
"num_tokens": 4943055.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.850647974014282,
|
|
"epoch": 0.16844713840795686,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00027170144588082635,
|
|
"loss": 6.6538,
|
|
"mean_token_accuracy": 0.128940337151289,
|
|
"num_tokens": 4952303.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.880576086044312,
|
|
"epoch": 0.1687820233749707,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00027052352584063763,
|
|
"loss": 6.7145,
|
|
"mean_token_accuracy": 0.12662334814667703,
|
|
"num_tokens": 4962858.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.8835780143737795,
|
|
"epoch": 0.16911690834198453,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00026934572852524907,
|
|
"loss": 6.7221,
|
|
"mean_token_accuracy": 0.11757861375808716,
|
|
"num_tokens": 4973064.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.823378467559815,
|
|
"epoch": 0.16945179330899837,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00026816808622456937,
|
|
"loss": 6.6898,
|
|
"mean_token_accuracy": 0.1307799570262432,
|
|
"num_tokens": 4983745.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.784839391708374,
|
|
"epoch": 0.16978667827601218,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002669906312242569,
|
|
"loss": 6.584,
|
|
"mean_token_accuracy": 0.12814493253827094,
|
|
"num_tokens": 4992957.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.876389694213867,
|
|
"epoch": 0.17012156324302602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026581339580483525,
|
|
"loss": 6.754,
|
|
"mean_token_accuracy": 0.11539148837327957,
|
|
"num_tokens": 5002529.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.8306896686553955,
|
|
"epoch": 0.17045644821003986,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002646364122408082,
|
|
"loss": 6.6743,
|
|
"mean_token_accuracy": 0.12661109492182732,
|
|
"num_tokens": 5012207.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.80672812461853,
|
|
"epoch": 0.1707913331770537,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002634597127997749,
|
|
"loss": 6.6072,
|
|
"mean_token_accuracy": 0.12907114252448082,
|
|
"num_tokens": 5022852.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.81147518157959,
|
|
"epoch": 0.1711262181440675,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002622833297415445,
|
|
"loss": 6.7125,
|
|
"mean_token_accuracy": 0.12396204024553299,
|
|
"num_tokens": 5032670.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.821378755569458,
|
|
"epoch": 0.17146110311108134,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002611072953172531,
|
|
"loss": 6.6237,
|
|
"mean_token_accuracy": 0.13157539442181587,
|
|
"num_tokens": 5042409.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.829177522659302,
|
|
"epoch": 0.17179598807809518,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00025993164176847845,
|
|
"loss": 6.6101,
|
|
"mean_token_accuracy": 0.1319831594824791,
|
|
"num_tokens": 5051036.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.838034152984619,
|
|
"epoch": 0.17213087304510902,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002587564013263564,
|
|
"loss": 6.6585,
|
|
"mean_token_accuracy": 0.12493582740426064,
|
|
"num_tokens": 5059958.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.816223049163819,
|
|
"epoch": 0.17246575801212283,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002575816062106974,
|
|
"loss": 6.6236,
|
|
"mean_token_accuracy": 0.12678119912743568,
|
|
"num_tokens": 5070436.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.853928470611573,
|
|
"epoch": 0.17280064297913666,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00025640728862910293,
|
|
"loss": 6.6837,
|
|
"mean_token_accuracy": 0.1225144974887371,
|
|
"num_tokens": 5080999.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.8469517707824705,
|
|
"epoch": 0.1731355279461505,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00025523348077608285,
|
|
"loss": 6.691,
|
|
"mean_token_accuracy": 0.123951056599617,
|
|
"num_tokens": 5090968.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.867240524291992,
|
|
"epoch": 0.17347041291316434,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00025406021483217225,
|
|
"loss": 6.7541,
|
|
"mean_token_accuracy": 0.12274921983480454,
|
|
"num_tokens": 5101222.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.858665084838867,
|
|
"epoch": 0.17380529788017815,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00025288752296304963,
|
|
"loss": 6.6296,
|
|
"mean_token_accuracy": 0.12856790646910668,
|
|
"num_tokens": 5109933.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.871017265319824,
|
|
"epoch": 0.174140182847192,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000251715437318655,
|
|
"loss": 6.7004,
|
|
"mean_token_accuracy": 0.12015999183058738,
|
|
"num_tokens": 5119371.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.8718102931976315,
|
|
"epoch": 0.17447506781420583,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002505439900323084,
|
|
"loss": 6.6955,
|
|
"mean_token_accuracy": 0.12642580419778823,
|
|
"num_tokens": 5129299.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.807115173339843,
|
|
"epoch": 0.17480995278121966,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00024937321321982894,
|
|
"loss": 6.6398,
|
|
"mean_token_accuracy": 0.12406356632709503,
|
|
"num_tokens": 5139317.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.802812623977661,
|
|
"epoch": 0.17514483774823347,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00024820313897865433,
|
|
"loss": 6.6253,
|
|
"mean_token_accuracy": 0.13387617096304893,
|
|
"num_tokens": 5149179.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.815228986740112,
|
|
"epoch": 0.1754797227152473,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00024703379938696105,
|
|
"loss": 6.6942,
|
|
"mean_token_accuracy": 0.12468233704566956,
|
|
"num_tokens": 5159995.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.824119138717651,
|
|
"epoch": 0.17581460768226115,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00024586522650278447,
|
|
"loss": 6.7064,
|
|
"mean_token_accuracy": 0.1234874352812767,
|
|
"num_tokens": 5170188.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.727022552490235,
|
|
"epoch": 0.17614949264927499,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00024469745236314064,
|
|
"loss": 6.4856,
|
|
"mean_token_accuracy": 0.14216285943984985,
|
|
"num_tokens": 5179313.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.894363975524902,
|
|
"epoch": 0.1764843776162888,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00024353050898314767,
|
|
"loss": 6.633,
|
|
"mean_token_accuracy": 0.12777856811881066,
|
|
"num_tokens": 5190775.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.823488235473633,
|
|
"epoch": 0.17681926258330263,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00024236442835514743,
|
|
"loss": 6.6161,
|
|
"mean_token_accuracy": 0.13103508874773978,
|
|
"num_tokens": 5200315.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.7732930183410645,
|
|
"epoch": 0.17715414755031647,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00024119924244782965,
|
|
"loss": 6.6877,
|
|
"mean_token_accuracy": 0.1246482066810131,
|
|
"num_tokens": 5209637.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.806252145767212,
|
|
"epoch": 0.1774890325173303,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00024003498320535462,
|
|
"loss": 6.6007,
|
|
"mean_token_accuracy": 0.12928060591220855,
|
|
"num_tokens": 5218767.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.8419249057769775,
|
|
"epoch": 0.17782391748434412,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023887168254647727,
|
|
"loss": 6.6966,
|
|
"mean_token_accuracy": 0.12198827490210533,
|
|
"num_tokens": 5228411.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.818643522262573,
|
|
"epoch": 0.17815880245135796,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00023770937236367308,
|
|
"loss": 6.5399,
|
|
"mean_token_accuracy": 0.13580606505274773,
|
|
"num_tokens": 5237885.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.771623039245606,
|
|
"epoch": 0.1784936874183718,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00023654808452226278,
|
|
"loss": 6.6216,
|
|
"mean_token_accuracy": 0.12874399200081826,
|
|
"num_tokens": 5248210.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.841006183624268,
|
|
"epoch": 0.17882857238538563,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023538785085953912,
|
|
"loss": 6.7179,
|
|
"mean_token_accuracy": 0.12152459099888802,
|
|
"num_tokens": 5258716.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.947375917434693,
|
|
"epoch": 0.17916345735239944,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023422870318389404,
|
|
"loss": 6.7584,
|
|
"mean_token_accuracy": 0.11797030940651894,
|
|
"num_tokens": 5270181.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.838764190673828,
|
|
"epoch": 0.17949834231941328,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002330706732739468,
|
|
"loss": 6.608,
|
|
"mean_token_accuracy": 0.1317827507853508,
|
|
"num_tokens": 5279735.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.782953977584839,
|
|
"epoch": 0.17983322728642712,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00023191379287767211,
|
|
"loss": 6.6768,
|
|
"mean_token_accuracy": 0.12713976725935935,
|
|
"num_tokens": 5290082.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.799340438842774,
|
|
"epoch": 0.18016811225344095,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002307580937115305,
|
|
"loss": 6.5897,
|
|
"mean_token_accuracy": 0.12767886444926263,
|
|
"num_tokens": 5299387.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.8211243629455565,
|
|
"epoch": 0.18050299722045476,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00022960360745959846,
|
|
"loss": 6.6863,
|
|
"mean_token_accuracy": 0.12651853859424592,
|
|
"num_tokens": 5309187.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.901071548461914,
|
|
"epoch": 0.1808378821874686,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00022845036577269972,
|
|
"loss": 6.6017,
|
|
"mean_token_accuracy": 0.12703219577670097,
|
|
"num_tokens": 5318334.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.833128595352173,
|
|
"epoch": 0.18117276715448244,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00022729840026753777,
|
|
"loss": 6.5916,
|
|
"mean_token_accuracy": 0.12880082949995994,
|
|
"num_tokens": 5327934.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.680203676223755,
|
|
"epoch": 0.18150765212149628,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002261477425258287,
|
|
"loss": 6.6388,
|
|
"mean_token_accuracy": 0.12967443689703942,
|
|
"num_tokens": 5338545.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.836952781677246,
|
|
"epoch": 0.1818425370885101,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0002249984240934358,
|
|
"loss": 6.6742,
|
|
"mean_token_accuracy": 0.12668901532888413,
|
|
"num_tokens": 5348590.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.86365442276001,
|
|
"epoch": 0.18217742205552392,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00022385047647950464,
|
|
"loss": 6.6662,
|
|
"mean_token_accuracy": 0.13221725150942804,
|
|
"num_tokens": 5359116.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.816820096969605,
|
|
"epoch": 0.18251230702253776,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002227039311555986,
|
|
"loss": 6.5816,
|
|
"mean_token_accuracy": 0.12717426493763923,
|
|
"num_tokens": 5368645.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.736215972900391,
|
|
"epoch": 0.1828471919895516,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002215588195548372,
|
|
"loss": 6.6042,
|
|
"mean_token_accuracy": 0.1240386612713337,
|
|
"num_tokens": 5379276.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.852612161636353,
|
|
"epoch": 0.1831820769565654,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022041517307103337,
|
|
"loss": 6.6055,
|
|
"mean_token_accuracy": 0.12126688584685326,
|
|
"num_tokens": 5389420.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.810447835922242,
|
|
"epoch": 0.18351696192357925,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002192730230578331,
|
|
"loss": 6.5729,
|
|
"mean_token_accuracy": 0.13156581223011016,
|
|
"num_tokens": 5398522.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.714840316772461,
|
|
"epoch": 0.18385184689059308,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002181324008278559,
|
|
"loss": 6.5134,
|
|
"mean_token_accuracy": 0.12573120072484018,
|
|
"num_tokens": 5408285.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.7645124912261965,
|
|
"epoch": 0.18418673185760692,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00021699333765183655,
|
|
"loss": 6.6557,
|
|
"mean_token_accuracy": 0.1297672800719738,
|
|
"num_tokens": 5417613.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.833173131942749,
|
|
"epoch": 0.18452161682462073,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002158558647577673,
|
|
"loss": 6.5478,
|
|
"mean_token_accuracy": 0.12156279236078263,
|
|
"num_tokens": 5427562.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.790528774261475,
|
|
"epoch": 0.18485650179163457,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021472001333004215,
|
|
"loss": 6.607,
|
|
"mean_token_accuracy": 0.1342087969183922,
|
|
"num_tokens": 5436824.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.776734685897827,
|
|
"epoch": 0.1851913867586484,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00021358581450860186,
|
|
"loss": 6.588,
|
|
"mean_token_accuracy": 0.13091426938772202,
|
|
"num_tokens": 5446694.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.744842576980591,
|
|
"epoch": 0.18552627172566225,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002124532993880799,
|
|
"loss": 6.5367,
|
|
"mean_token_accuracy": 0.13016503900289536,
|
|
"num_tokens": 5457201.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.819921684265137,
|
|
"epoch": 0.18586115669267605,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00021132249901695044,
|
|
"loss": 6.6072,
|
|
"mean_token_accuracy": 0.1310259610414505,
|
|
"num_tokens": 5466727.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.791133451461792,
|
|
"epoch": 0.1861960416596899,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00021019344439667705,
|
|
"loss": 6.6041,
|
|
"mean_token_accuracy": 0.1303398057818413,
|
|
"num_tokens": 5475924.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.814376735687256,
|
|
"epoch": 0.18653092662670373,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00020906616648086213,
|
|
"loss": 6.6259,
|
|
"mean_token_accuracy": 0.1274624541401863,
|
|
"num_tokens": 5485186.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.867758893966675,
|
|
"epoch": 0.18686581159371757,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00020794069617439942,
|
|
"loss": 6.6386,
|
|
"mean_token_accuracy": 0.12785386741161348,
|
|
"num_tokens": 5495425.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.8394591331481935,
|
|
"epoch": 0.18720069656073138,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00020681706433262593,
|
|
"loss": 6.5887,
|
|
"mean_token_accuracy": 0.13350537568330764,
|
|
"num_tokens": 5504822.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.748140525817871,
|
|
"epoch": 0.18753558152774522,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00020569530176047602,
|
|
"loss": 6.6664,
|
|
"mean_token_accuracy": 0.12352422326803207,
|
|
"num_tokens": 5513896.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.769604635238648,
|
|
"epoch": 0.18787046649475905,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002045754392116374,
|
|
"loss": 6.632,
|
|
"mean_token_accuracy": 0.12628040462732315,
|
|
"num_tokens": 5524746.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.810101413726807,
|
|
"epoch": 0.1882053514617729,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020345750738770757,
|
|
"loss": 6.5922,
|
|
"mean_token_accuracy": 0.12662632763385773,
|
|
"num_tokens": 5534479.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.846392440795898,
|
|
"epoch": 0.1885402364287867,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00020234153693735214,
|
|
"loss": 6.6596,
|
|
"mean_token_accuracy": 0.12164091020822525,
|
|
"num_tokens": 5545252.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.885404300689697,
|
|
"epoch": 0.18887512139580054,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002012275584554647,
|
|
"loss": 6.6396,
|
|
"mean_token_accuracy": 0.12538378760218621,
|
|
"num_tokens": 5555156.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.812751007080078,
|
|
"epoch": 0.18921000636281438,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00020011560248232803,
|
|
"loss": 6.6319,
|
|
"mean_token_accuracy": 0.12105094343423843,
|
|
"num_tokens": 5566828.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.815339756011963,
|
|
"epoch": 0.1895448913298282,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019900569950277692,
|
|
"loss": 6.6075,
|
|
"mean_token_accuracy": 0.12889422550797464,
|
|
"num_tokens": 5576168.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.727983856201172,
|
|
"epoch": 0.18987977629684202,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00019789787994536228,
|
|
"loss": 6.4988,
|
|
"mean_token_accuracy": 0.14106391817331315,
|
|
"num_tokens": 5584944.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.843129587173462,
|
|
"epoch": 0.19021466126385586,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00019679217418151667,
|
|
"loss": 6.6073,
|
|
"mean_token_accuracy": 0.12682909443974494,
|
|
"num_tokens": 5593953.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.818669223785401,
|
|
"epoch": 0.1905495462308697,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00019568861252472236,
|
|
"loss": 6.6886,
|
|
"mean_token_accuracy": 0.1252566806972027,
|
|
"num_tokens": 5603241.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.829478216171265,
|
|
"epoch": 0.19088443119788354,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00019458722522967952,
|
|
"loss": 6.6345,
|
|
"mean_token_accuracy": 0.12553785666823386,
|
|
"num_tokens": 5613724.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.809851169586182,
|
|
"epoch": 0.19121931616489735,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00019348804249147723,
|
|
"loss": 6.6064,
|
|
"mean_token_accuracy": 0.12687140852212905,
|
|
"num_tokens": 5622422.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.794346857070923,
|
|
"epoch": 0.19155420113191118,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001923910944447655,
|
|
"loss": 6.6097,
|
|
"mean_token_accuracy": 0.12917831987142564,
|
|
"num_tokens": 5633054.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.781935501098633,
|
|
"epoch": 0.19188908609892502,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00019129641116292928,
|
|
"loss": 6.558,
|
|
"mean_token_accuracy": 0.13445394337177277,
|
|
"num_tokens": 5643655.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.826231145858765,
|
|
"epoch": 0.19222397106593886,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00019020402265726343,
|
|
"loss": 6.6112,
|
|
"mean_token_accuracy": 0.12430167943239212,
|
|
"num_tokens": 5654186.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.738478469848633,
|
|
"epoch": 0.19255885603295267,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001891139588761509,
|
|
"loss": 6.5893,
|
|
"mean_token_accuracy": 0.1296643413603306,
|
|
"num_tokens": 5664155.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.77187762260437,
|
|
"epoch": 0.1928937409999665,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00018802624970424076,
|
|
"loss": 6.6262,
|
|
"mean_token_accuracy": 0.1317816123366356,
|
|
"num_tokens": 5674320.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.811074161529541,
|
|
"epoch": 0.19322862596698034,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00018694092496162945,
|
|
"loss": 6.5645,
|
|
"mean_token_accuracy": 0.12772737443447113,
|
|
"num_tokens": 5684795.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.801860189437866,
|
|
"epoch": 0.19356351093399418,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00018585801440304306,
|
|
"loss": 6.5186,
|
|
"mean_token_accuracy": 0.13562109619379042,
|
|
"num_tokens": 5694683.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.806711196899414,
|
|
"epoch": 0.193898395901008,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00018477754771702165,
|
|
"loss": 6.6508,
|
|
"mean_token_accuracy": 0.12221604064106942,
|
|
"num_tokens": 5704739.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.830066728591919,
|
|
"epoch": 0.19423328086802183,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00018369955452510506,
|
|
"loss": 6.6421,
|
|
"mean_token_accuracy": 0.12659963890910148,
|
|
"num_tokens": 5714256.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.8282102108001705,
|
|
"epoch": 0.19456816583503567,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001826240643810212,
|
|
"loss": 6.6148,
|
|
"mean_token_accuracy": 0.13070322200655937,
|
|
"num_tokens": 5724553.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.801303148269653,
|
|
"epoch": 0.1949030508020495,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001815511067698758,
|
|
"loss": 6.5688,
|
|
"mean_token_accuracy": 0.13162247389554976,
|
|
"num_tokens": 5734529.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.800146865844726,
|
|
"epoch": 0.19523793576906331,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001804807111073436,
|
|
"loss": 6.5548,
|
|
"mean_token_accuracy": 0.12798983603715897,
|
|
"num_tokens": 5744650.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.833826017379761,
|
|
"epoch": 0.19557282073607715,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001794129067388625,
|
|
"loss": 6.5777,
|
|
"mean_token_accuracy": 0.12912197709083556,
|
|
"num_tokens": 5754214.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.809359788894653,
|
|
"epoch": 0.195907705703091,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00017834772293882868,
|
|
"loss": 6.72,
|
|
"mean_token_accuracy": 0.11678024381399155,
|
|
"num_tokens": 5764660.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.851030015945435,
|
|
"epoch": 0.19624259067010483,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000177285188909794,
|
|
"loss": 6.6705,
|
|
"mean_token_accuracy": 0.1260698400437832,
|
|
"num_tokens": 5774710.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.815495586395263,
|
|
"epoch": 0.19657747563711864,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001762253337816656,
|
|
"loss": 6.6,
|
|
"mean_token_accuracy": 0.13109977841377257,
|
|
"num_tokens": 5784147.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.842741632461548,
|
|
"epoch": 0.19691236060413247,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00017516818661090738,
|
|
"loss": 6.6937,
|
|
"mean_token_accuracy": 0.13180825263261794,
|
|
"num_tokens": 5793695.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.803332710266114,
|
|
"epoch": 0.1972472455711463,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001741137763797428,
|
|
"loss": 6.6552,
|
|
"mean_token_accuracy": 0.1228181965649128,
|
|
"num_tokens": 5803436.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.858682489395141,
|
|
"epoch": 0.19758213053816015,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00017306213199536115,
|
|
"loss": 6.6661,
|
|
"mean_token_accuracy": 0.12328538373112678,
|
|
"num_tokens": 5814783.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.829118061065674,
|
|
"epoch": 0.19791701550517396,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001720132822891243,
|
|
"loss": 6.6133,
|
|
"mean_token_accuracy": 0.12435527741909028,
|
|
"num_tokens": 5824715.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.829468059539795,
|
|
"epoch": 0.1982519004721878,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001709672560157769,
|
|
"loss": 6.5189,
|
|
"mean_token_accuracy": 0.13034008517861367,
|
|
"num_tokens": 5832056.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.856924724578858,
|
|
"epoch": 0.19858678543920164,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00016992408185265758,
|
|
"loss": 6.6522,
|
|
"mean_token_accuracy": 0.12381615936756134,
|
|
"num_tokens": 5842430.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.810846519470215,
|
|
"epoch": 0.19892167040621547,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00016888378839891298,
|
|
"loss": 6.6752,
|
|
"mean_token_accuracy": 0.12158726155757904,
|
|
"num_tokens": 5852115.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.722003555297851,
|
|
"epoch": 0.19925655537322928,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001678464041747137,
|
|
"loss": 6.5469,
|
|
"mean_token_accuracy": 0.13217816799879073,
|
|
"num_tokens": 5863019.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.799395942687989,
|
|
"epoch": 0.19959144034024312,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016681195762047223,
|
|
"loss": 6.5504,
|
|
"mean_token_accuracy": 0.13168118968605996,
|
|
"num_tokens": 5873025.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.856830787658692,
|
|
"epoch": 0.19992632530725696,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00016578047709606337,
|
|
"loss": 6.519,
|
|
"mean_token_accuracy": 0.13400563970208168,
|
|
"num_tokens": 5881946.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.761207103729248,
|
|
"epoch": 0.2002612102742708,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00016475199088004678,
|
|
"loss": 6.5859,
|
|
"mean_token_accuracy": 0.12711762860417367,
|
|
"num_tokens": 5891159.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.792570781707764,
|
|
"epoch": 0.2005960952412846,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00016372652716889163,
|
|
"loss": 6.6242,
|
|
"mean_token_accuracy": 0.12485459521412849,
|
|
"num_tokens": 5901279.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.787021446228027,
|
|
"epoch": 0.20093098020829844,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001627041140762035,
|
|
"loss": 6.4947,
|
|
"mean_token_accuracy": 0.13184405416250228,
|
|
"num_tokens": 5911446.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.7908485412597654,
|
|
"epoch": 0.20126586517531228,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00016168477963195382,
|
|
"loss": 6.5958,
|
|
"mean_token_accuracy": 0.12908268123865127,
|
|
"num_tokens": 5921433.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.823174047470093,
|
|
"epoch": 0.20160075014232612,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001606685517817114,
|
|
"loss": 6.5735,
|
|
"mean_token_accuracy": 0.13237855955958366,
|
|
"num_tokens": 5930820.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.824475479125977,
|
|
"epoch": 0.20193563510933993,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00015965545838587592,
|
|
"loss": 6.5891,
|
|
"mean_token_accuracy": 0.12329481318593025,
|
|
"num_tokens": 5941460.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.807975006103516,
|
|
"epoch": 0.20227052007635377,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00015864552721891467,
|
|
"loss": 6.5664,
|
|
"mean_token_accuracy": 0.12724735885858535,
|
|
"num_tokens": 5950262.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.792840147018433,
|
|
"epoch": 0.2026054050433676,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00015763878596860076,
|
|
"loss": 6.6069,
|
|
"mean_token_accuracy": 0.12816847562789918,
|
|
"num_tokens": 5959531.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.756864070892334,
|
|
"epoch": 0.20294029001038144,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00015663526223525412,
|
|
"loss": 6.5376,
|
|
"mean_token_accuracy": 0.13237527310848235,
|
|
"num_tokens": 5969692.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.738115882873535,
|
|
"epoch": 0.20327517497739525,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001556349835309848,
|
|
"loss": 6.5253,
|
|
"mean_token_accuracy": 0.12704963907599448,
|
|
"num_tokens": 5978722.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.843791580200195,
|
|
"epoch": 0.2036100599444091,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001546379772789389,
|
|
"loss": 6.7126,
|
|
"mean_token_accuracy": 0.13372190073132514,
|
|
"num_tokens": 5988931.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.837690639495849,
|
|
"epoch": 0.20394494491142293,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00015364427081254622,
|
|
"loss": 6.6053,
|
|
"mean_token_accuracy": 0.1307089202105999,
|
|
"num_tokens": 5999469.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.784780073165893,
|
|
"epoch": 0.20427982987843676,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015265389137477165,
|
|
"loss": 6.4825,
|
|
"mean_token_accuracy": 0.13612487763166428,
|
|
"num_tokens": 6009163.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.752602815628052,
|
|
"epoch": 0.2046147148454506,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00015166686611736786,
|
|
"loss": 6.4907,
|
|
"mean_token_accuracy": 0.1331620492041111,
|
|
"num_tokens": 6019084.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.811483240127563,
|
|
"epoch": 0.2049495998124644,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00015068322210013064,
|
|
"loss": 6.6831,
|
|
"mean_token_accuracy": 0.1276084303855896,
|
|
"num_tokens": 6028991.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.8156044483184814,
|
|
"epoch": 0.20528448477947825,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001497029862901578,
|
|
"loss": 6.5827,
|
|
"mean_token_accuracy": 0.13622187227010726,
|
|
"num_tokens": 6038353.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.799685859680176,
|
|
"epoch": 0.2056193697464921,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00014872618556110905,
|
|
"loss": 6.5275,
|
|
"mean_token_accuracy": 0.13983718305826187,
|
|
"num_tokens": 6048196.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.804253673553466,
|
|
"epoch": 0.20595425471350592,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00014775284669246992,
|
|
"loss": 6.646,
|
|
"mean_token_accuracy": 0.12736983150243758,
|
|
"num_tokens": 6059490.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.786814451217651,
|
|
"epoch": 0.20628913968051973,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00014678299636881716,
|
|
"loss": 6.5864,
|
|
"mean_token_accuracy": 0.12696156799793243,
|
|
"num_tokens": 6070260.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.736616897583008,
|
|
"epoch": 0.20662402464753357,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001458166611790873,
|
|
"loss": 6.4936,
|
|
"mean_token_accuracy": 0.1394593209028244,
|
|
"num_tokens": 6079656.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.76190710067749,
|
|
"epoch": 0.2069589096145474,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00014485386761584773,
|
|
"loss": 6.5857,
|
|
"mean_token_accuracy": 0.12915114015340806,
|
|
"num_tokens": 6089313.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.805394077301026,
|
|
"epoch": 0.20729379458156125,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00014389464207457042,
|
|
"loss": 6.5439,
|
|
"mean_token_accuracy": 0.1319348469376564,
|
|
"num_tokens": 6099013.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.82142972946167,
|
|
"epoch": 0.20762867954857506,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00014293901085290795,
|
|
"loss": 6.6034,
|
|
"mean_token_accuracy": 0.12586347237229348,
|
|
"num_tokens": 6109869.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.873842811584472,
|
|
"epoch": 0.2079635645155889,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00014198700014997307,
|
|
"loss": 6.6921,
|
|
"mean_token_accuracy": 0.124761401116848,
|
|
"num_tokens": 6120113.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.843233585357666,
|
|
"epoch": 0.20829844948260273,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00014103863606562016,
|
|
"loss": 6.6372,
|
|
"mean_token_accuracy": 0.12196057364344597,
|
|
"num_tokens": 6131282.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.8211523532867435,
|
|
"epoch": 0.20863333444961657,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00014009394459972964,
|
|
"loss": 6.5522,
|
|
"mean_token_accuracy": 0.12481173872947693,
|
|
"num_tokens": 6141405.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.843969678878784,
|
|
"epoch": 0.20896821941663038,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00013915295165149513,
|
|
"loss": 6.5458,
|
|
"mean_token_accuracy": 0.12349180802702904,
|
|
"num_tokens": 6150586.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.753193616867065,
|
|
"epoch": 0.20930310438364422,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00013821568301871384,
|
|
"loss": 6.5855,
|
|
"mean_token_accuracy": 0.13214803636074066,
|
|
"num_tokens": 6160023.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.754095840454101,
|
|
"epoch": 0.20963798935065805,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00013728216439707862,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.13311260268092157,
|
|
"num_tokens": 6171038.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.800445556640625,
|
|
"epoch": 0.2099728743176719,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013635242137947419,
|
|
"loss": 6.5688,
|
|
"mean_token_accuracy": 0.12753629684448242,
|
|
"num_tokens": 6179821.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.744899702072144,
|
|
"epoch": 0.2103077592846857,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00013542647945527498,
|
|
"loss": 6.5299,
|
|
"mean_token_accuracy": 0.132371436804533,
|
|
"num_tokens": 6190443.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.828707456588745,
|
|
"epoch": 0.21064264425169954,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001345043640096465,
|
|
"loss": 6.6344,
|
|
"mean_token_accuracy": 0.1293755479156971,
|
|
"num_tokens": 6199783.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.804296827316284,
|
|
"epoch": 0.21097752921871338,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013358610032284957,
|
|
"loss": 6.4981,
|
|
"mean_token_accuracy": 0.13093448728322982,
|
|
"num_tokens": 6209253.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.727840518951416,
|
|
"epoch": 0.21131241418572722,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000132671713569547,
|
|
"loss": 6.5692,
|
|
"mean_token_accuracy": 0.13439862951636314,
|
|
"num_tokens": 6220252.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.806533575057983,
|
|
"epoch": 0.21164729915274103,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001317612288181136,
|
|
"loss": 6.5471,
|
|
"mean_token_accuracy": 0.1257143087685108,
|
|
"num_tokens": 6229385.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.7884642601013185,
|
|
"epoch": 0.21198218411975486,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00013085467102994864,
|
|
"loss": 6.5607,
|
|
"mean_token_accuracy": 0.13153301626443864,
|
|
"num_tokens": 6239758.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.771036577224732,
|
|
"epoch": 0.2123170690867687,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00012995206505879198,
|
|
"loss": 6.4624,
|
|
"mean_token_accuracy": 0.1326538510620594,
|
|
"num_tokens": 6249544.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.813483333587646,
|
|
"epoch": 0.21265195405378254,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0001290534356500421,
|
|
"loss": 6.6152,
|
|
"mean_token_accuracy": 0.12935134023427963,
|
|
"num_tokens": 6260052.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.728772115707398,
|
|
"epoch": 0.21298683902079635,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00012815880744007827,
|
|
"loss": 6.5161,
|
|
"mean_token_accuracy": 0.1393673837184906,
|
|
"num_tokens": 6269627.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.8218001365661625,
|
|
"epoch": 0.21332172398781019,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00012726820495558483,
|
|
"loss": 6.539,
|
|
"mean_token_accuracy": 0.12464180812239647,
|
|
"num_tokens": 6279170.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.7959307670593265,
|
|
"epoch": 0.21365660895482402,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00012638165261287868,
|
|
"loss": 6.5407,
|
|
"mean_token_accuracy": 0.1338554173707962,
|
|
"num_tokens": 6288574.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.77916259765625,
|
|
"epoch": 0.21399149392183786,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001254991747172402,
|
|
"loss": 6.5957,
|
|
"mean_token_accuracy": 0.13496605828404426,
|
|
"num_tokens": 6299492.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.743713665008545,
|
|
"epoch": 0.21432637888885167,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00012462079546224662,
|
|
"loss": 6.5311,
|
|
"mean_token_accuracy": 0.13045995235443114,
|
|
"num_tokens": 6310440.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.844704055786133,
|
|
"epoch": 0.2146612638558655,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00012374653892910896,
|
|
"loss": 6.5876,
|
|
"mean_token_accuracy": 0.13698795959353446,
|
|
"num_tokens": 6319476.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.788089466094971,
|
|
"epoch": 0.21499614882287935,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00012287642908601166,
|
|
"loss": 6.6685,
|
|
"mean_token_accuracy": 0.1300518959760666,
|
|
"num_tokens": 6329988.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.850390291213989,
|
|
"epoch": 0.21533103378989318,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00012201048978745569,
|
|
"loss": 6.6404,
|
|
"mean_token_accuracy": 0.12627878934144973,
|
|
"num_tokens": 6338663.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.849464702606201,
|
|
"epoch": 0.215665918756907,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00012114874477360427,
|
|
"loss": 6.4997,
|
|
"mean_token_accuracy": 0.13264379724860192,
|
|
"num_tokens": 6347535.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.745803594589233,
|
|
"epoch": 0.21600080372392083,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00012029121766963236,
|
|
"loss": 6.5317,
|
|
"mean_token_accuracy": 0.13268077075481416,
|
|
"num_tokens": 6356795.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.7572681427001955,
|
|
"epoch": 0.21633568869093467,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011943793198507858,
|
|
"loss": 6.4757,
|
|
"mean_token_accuracy": 0.13390135541558265,
|
|
"num_tokens": 6368092.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.787887191772461,
|
|
"epoch": 0.2166705736579485,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011858891111320104,
|
|
"loss": 6.5645,
|
|
"mean_token_accuracy": 0.13271910920739174,
|
|
"num_tokens": 6377646.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.873563766479492,
|
|
"epoch": 0.21700545862496232,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001177441783303359,
|
|
"loss": 6.6369,
|
|
"mean_token_accuracy": 0.12911410853266717,
|
|
"num_tokens": 6386888.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.8552556991577145,
|
|
"epoch": 0.21734034359197615,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00011690375679525896,
|
|
"loss": 6.4937,
|
|
"mean_token_accuracy": 0.136123950779438,
|
|
"num_tokens": 6396616.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.7596766471862795,
|
|
"epoch": 0.21767522855899,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00011606766954855124,
|
|
"loss": 6.4677,
|
|
"mean_token_accuracy": 0.13728993386030197,
|
|
"num_tokens": 6405474.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.722172832489013,
|
|
"epoch": 0.21801011352600383,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00011523593951196702,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.13620053455233574,
|
|
"num_tokens": 6414574.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.743352651596069,
|
|
"epoch": 0.21834499849301764,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00011440858948780523,
|
|
"loss": 6.5184,
|
|
"mean_token_accuracy": 0.1357611134648323,
|
|
"num_tokens": 6424545.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.768523120880127,
|
|
"epoch": 0.21867988346003148,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00011358564215828484,
|
|
"loss": 6.4864,
|
|
"mean_token_accuracy": 0.13317656964063646,
|
|
"num_tokens": 6433704.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.731975507736206,
|
|
"epoch": 0.21901476842704531,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00011276712008492254,
|
|
"loss": 6.4403,
|
|
"mean_token_accuracy": 0.13454940617084504,
|
|
"num_tokens": 6442343.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.8199234962463375,
|
|
"epoch": 0.21934965339405915,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00011195304570791451,
|
|
"loss": 6.6351,
|
|
"mean_token_accuracy": 0.12604313343763351,
|
|
"num_tokens": 6452755.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.710698509216309,
|
|
"epoch": 0.21968453836107296,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00011114344134552094,
|
|
"loss": 6.4581,
|
|
"mean_token_accuracy": 0.1343326598405838,
|
|
"num_tokens": 6462060.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.774784898757934,
|
|
"epoch": 0.2200194233280868,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001103383291934545,
|
|
"loss": 6.444,
|
|
"mean_token_accuracy": 0.14563580229878426,
|
|
"num_tokens": 6472051.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.7286536693573,
|
|
"epoch": 0.22035430829510064,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00010953773132427141,
|
|
"loss": 6.4827,
|
|
"mean_token_accuracy": 0.1401687055826187,
|
|
"num_tokens": 6482510.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.759169816970825,
|
|
"epoch": 0.22068919326211447,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00010874166968676677,
|
|
"loss": 6.4378,
|
|
"mean_token_accuracy": 0.134411983191967,
|
|
"num_tokens": 6491687.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.803877449035644,
|
|
"epoch": 0.22102407822912828,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010795016610537251,
|
|
"loss": 6.5339,
|
|
"mean_token_accuracy": 0.1297886326909065,
|
|
"num_tokens": 6501619.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.860368299484253,
|
|
"epoch": 0.22135896319614212,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010716324227955904,
|
|
"loss": 6.5072,
|
|
"mean_token_accuracy": 0.13070680797100068,
|
|
"num_tokens": 6510769.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.702731370925903,
|
|
"epoch": 0.22169384816315596,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001063809197832406,
|
|
"loss": 6.455,
|
|
"mean_token_accuracy": 0.13381422609090804,
|
|
"num_tokens": 6519699.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.764220523834228,
|
|
"epoch": 0.2220287331301698,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010560322006418368,
|
|
"loss": 6.6001,
|
|
"mean_token_accuracy": 0.12459379881620407,
|
|
"num_tokens": 6530973.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.76298770904541,
|
|
"epoch": 0.2223636180971836,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00010483016444341887,
|
|
"loss": 6.5266,
|
|
"mean_token_accuracy": 0.1310693420469761,
|
|
"num_tokens": 6540434.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.807111167907715,
|
|
"epoch": 0.22269850306419745,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00010406177411465654,
|
|
"loss": 6.5785,
|
|
"mean_token_accuracy": 0.13130905777215957,
|
|
"num_tokens": 6550515.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.801345539093018,
|
|
"epoch": 0.22303338803121128,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00010329807014370562,
|
|
"loss": 6.5429,
|
|
"mean_token_accuracy": 0.13474279195070266,
|
|
"num_tokens": 6561423.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.824318695068359,
|
|
"epoch": 0.22336827299822512,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00010253907346789632,
|
|
"loss": 6.4758,
|
|
"mean_token_accuracy": 0.1319676823914051,
|
|
"num_tokens": 6570378.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.742883157730103,
|
|
"epoch": 0.22370315796523893,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00010178480489550596,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.14064373224973678,
|
|
"num_tokens": 6580477.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.789406108856201,
|
|
"epoch": 0.22403804293225277,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00010103528510518836,
|
|
"loss": 6.6035,
|
|
"mean_token_accuracy": 0.11799206510186196,
|
|
"num_tokens": 6591368.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.776973485946655,
|
|
"epoch": 0.2243729278992666,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0001002905346454073,
|
|
"loss": 6.5327,
|
|
"mean_token_accuracy": 0.13617580980062485,
|
|
"num_tokens": 6601819.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.776618242263794,
|
|
"epoch": 0.22470781286628044,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.955057393387285e-05,
|
|
"loss": 6.4818,
|
|
"mean_token_accuracy": 0.12940553873777388,
|
|
"num_tokens": 6611208.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.770179367065429,
|
|
"epoch": 0.22504269783329425,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.88154232569816e-05,
|
|
"loss": 6.6091,
|
|
"mean_token_accuracy": 0.13247944489121438,
|
|
"num_tokens": 6621273.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.7875950813293455,
|
|
"epoch": 0.2253775828003081,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.808510276926075e-05,
|
|
"loss": 6.5607,
|
|
"mean_token_accuracy": 0.12952352091670036,
|
|
"num_tokens": 6630565.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.785948610305786,
|
|
"epoch": 0.22571246776732193,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 9.735963249281549e-05,
|
|
"loss": 6.5312,
|
|
"mean_token_accuracy": 0.1321560762822628,
|
|
"num_tokens": 6640116.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.693677854537964,
|
|
"epoch": 0.22604735273433577,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.663903231677974e-05,
|
|
"loss": 6.476,
|
|
"mean_token_accuracy": 0.13775476962327957,
|
|
"num_tokens": 6650537.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.803247785568237,
|
|
"epoch": 0.22638223770134958,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 9.592332199677145e-05,
|
|
"loss": 6.571,
|
|
"mean_token_accuracy": 0.12863265946507454,
|
|
"num_tokens": 6659827.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.804413509368897,
|
|
"epoch": 0.2267171226683634,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.521252115435061e-05,
|
|
"loss": 6.5474,
|
|
"mean_token_accuracy": 0.13004275858402253,
|
|
"num_tokens": 6669836.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.794536161422729,
|
|
"epoch": 0.22705200763537725,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.450664927648126e-05,
|
|
"loss": 6.582,
|
|
"mean_token_accuracy": 0.13132933303713798,
|
|
"num_tokens": 6680249.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.792027139663697,
|
|
"epoch": 0.2273868926023911,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.380572571499758e-05,
|
|
"loss": 6.5397,
|
|
"mean_token_accuracy": 0.13384176567196845,
|
|
"num_tokens": 6689613.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.876494359970093,
|
|
"epoch": 0.2277217775694049,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 9.310976968607307e-05,
|
|
"loss": 6.638,
|
|
"mean_token_accuracy": 0.12734776139259338,
|
|
"num_tokens": 6700253.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.809216499328613,
|
|
"epoch": 0.22805666253641874,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.241880026969381e-05,
|
|
"loss": 6.5323,
|
|
"mean_token_accuracy": 0.13138106614351272,
|
|
"num_tokens": 6709704.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.815900659561157,
|
|
"epoch": 0.22839154750343257,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 9.173283640913537e-05,
|
|
"loss": 6.4628,
|
|
"mean_token_accuracy": 0.13635657355189323,
|
|
"num_tokens": 6719814.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.794958543777466,
|
|
"epoch": 0.2287264324704464,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.10518969104436e-05,
|
|
"loss": 6.5045,
|
|
"mean_token_accuracy": 0.12916772514581681,
|
|
"num_tokens": 6729432.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.771754360198974,
|
|
"epoch": 0.22906131743746022,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 9.037600044191868e-05,
|
|
"loss": 6.5177,
|
|
"mean_token_accuracy": 0.13270103111863135,
|
|
"num_tokens": 6740204.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.725033521652222,
|
|
"epoch": 0.22939620240447406,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.970516553360383e-05,
|
|
"loss": 6.4926,
|
|
"mean_token_accuracy": 0.13345336839556693,
|
|
"num_tokens": 6750756.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.731901931762695,
|
|
"epoch": 0.2297310873714879,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.903941057677692e-05,
|
|
"loss": 6.5313,
|
|
"mean_token_accuracy": 0.13500786423683167,
|
|
"num_tokens": 6760455.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.779512166976929,
|
|
"epoch": 0.23006597233850173,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.837875382344635e-05,
|
|
"loss": 6.4663,
|
|
"mean_token_accuracy": 0.13911383524537085,
|
|
"num_tokens": 6769622.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.862916803359985,
|
|
"epoch": 0.23040085730551554,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.772321338585076e-05,
|
|
"loss": 6.6939,
|
|
"mean_token_accuracy": 0.12039006575942039,
|
|
"num_tokens": 6780051.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.8672120571136475,
|
|
"epoch": 0.23073574227252938,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.707280723596242e-05,
|
|
"loss": 6.6187,
|
|
"mean_token_accuracy": 0.12534988820552825,
|
|
"num_tokens": 6790209.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.831813764572144,
|
|
"epoch": 0.23107062723954322,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 8.64275532049944e-05,
|
|
"loss": 6.5413,
|
|
"mean_token_accuracy": 0.12511147931218147,
|
|
"num_tokens": 6801428.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.814356803894043,
|
|
"epoch": 0.23140551220655706,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.578746898291198e-05,
|
|
"loss": 6.5258,
|
|
"mean_token_accuracy": 0.1360584504902363,
|
|
"num_tokens": 6810184.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.852376556396484,
|
|
"epoch": 0.23174039717357087,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.515257211794742e-05,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.13506544306874274,
|
|
"num_tokens": 6819685.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.751736211776733,
|
|
"epoch": 0.2320752821405847,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.452288001611896e-05,
|
|
"loss": 6.49,
|
|
"mean_token_accuracy": 0.14031795561313629,
|
|
"num_tokens": 6829096.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.770990943908691,
|
|
"epoch": 0.23241016710759854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.389840994075379e-05,
|
|
"loss": 6.4955,
|
|
"mean_token_accuracy": 0.1357216477394104,
|
|
"num_tokens": 6838425.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.774834728240966,
|
|
"epoch": 0.23274505207461238,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 8.327917901201435e-05,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.13582499623298644,
|
|
"num_tokens": 6848184.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.826266241073609,
|
|
"epoch": 0.2330799370416262,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.266520420642931e-05,
|
|
"loss": 6.5265,
|
|
"mean_token_accuracy": 0.13699022382497789,
|
|
"num_tokens": 6857805.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.80344557762146,
|
|
"epoch": 0.23341482200864003,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.205650235642828e-05,
|
|
"loss": 6.6381,
|
|
"mean_token_accuracy": 0.12946054264903067,
|
|
"num_tokens": 6868002.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.8383880138397215,
|
|
"epoch": 0.23374970697565386,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.145309014987978e-05,
|
|
"loss": 6.5747,
|
|
"mean_token_accuracy": 0.12304935231804848,
|
|
"num_tokens": 6877400.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.764590501785278,
|
|
"epoch": 0.2340845919426677,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.085498412963437e-05,
|
|
"loss": 6.4044,
|
|
"mean_token_accuracy": 0.14621803015470505,
|
|
"num_tokens": 6886771.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.766050338745117,
|
|
"epoch": 0.2344194769096815,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 8.026220069307078e-05,
|
|
"loss": 6.542,
|
|
"mean_token_accuracy": 0.1415340781211853,
|
|
"num_tokens": 6896579.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.619394969940186,
|
|
"epoch": 0.23475436187669535,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 7.967475609164621e-05,
|
|
"loss": 6.3001,
|
|
"mean_token_accuracy": 0.1499709464609623,
|
|
"num_tokens": 6906400.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.7810791492462155,
|
|
"epoch": 0.2350892468437092,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.909266643045124e-05,
|
|
"loss": 6.5501,
|
|
"mean_token_accuracy": 0.13241417482495307,
|
|
"num_tokens": 6915597.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.782319259643555,
|
|
"epoch": 0.23542413181072303,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 7.851594766776802e-05,
|
|
"loss": 6.4769,
|
|
"mean_token_accuracy": 0.13908046260476112,
|
|
"num_tokens": 6924925.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.823849487304687,
|
|
"epoch": 0.23575901677773684,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.794461561463265e-05,
|
|
"loss": 6.5116,
|
|
"mean_token_accuracy": 0.1363851711153984,
|
|
"num_tokens": 6933912.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.8144848346710205,
|
|
"epoch": 0.23609390174475067,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.7378685934402e-05,
|
|
"loss": 6.4956,
|
|
"mean_token_accuracy": 0.132178645581007,
|
|
"num_tokens": 6943904.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.761655712127686,
|
|
"epoch": 0.2364287867117645,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.68181741423242e-05,
|
|
"loss": 6.4195,
|
|
"mean_token_accuracy": 0.13411911204457283,
|
|
"num_tokens": 6953549.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.771081829071045,
|
|
"epoch": 0.23676367167877835,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.626309560511313e-05,
|
|
"loss": 6.44,
|
|
"mean_token_accuracy": 0.1305633530020714,
|
|
"num_tokens": 6962778.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.815757989883423,
|
|
"epoch": 0.23709855664579216,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 7.571346554052724e-05,
|
|
"loss": 6.5598,
|
|
"mean_token_accuracy": 0.1300210751593113,
|
|
"num_tokens": 6974052.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.794507741928101,
|
|
"epoch": 0.237433441612806,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 7.516929901695249e-05,
|
|
"loss": 6.6488,
|
|
"mean_token_accuracy": 0.13233589828014375,
|
|
"num_tokens": 6983471.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.768416547775269,
|
|
"epoch": 0.23776832657981983,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 7.463061095298893e-05,
|
|
"loss": 6.5128,
|
|
"mean_token_accuracy": 0.13784375190734863,
|
|
"num_tokens": 6993629.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.790668487548828,
|
|
"epoch": 0.23810321154683367,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.409741611704198e-05,
|
|
"loss": 6.4862,
|
|
"mean_token_accuracy": 0.1340516798198223,
|
|
"num_tokens": 7001519.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.843166542053223,
|
|
"epoch": 0.23843809651384748,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 7.35697291269174e-05,
|
|
"loss": 6.5882,
|
|
"mean_token_accuracy": 0.12695166170597078,
|
|
"num_tokens": 7011501.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.7557963848114015,
|
|
"epoch": 0.23877298148086132,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7.304756444942056e-05,
|
|
"loss": 6.4999,
|
|
"mean_token_accuracy": 0.13288233801722527,
|
|
"num_tokens": 7021396.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.751760864257813,
|
|
"epoch": 0.23910786644787516,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 7.253093639995994e-05,
|
|
"loss": 6.4636,
|
|
"mean_token_accuracy": 0.13693431839346887,
|
|
"num_tokens": 7031365.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.81280550956726,
|
|
"epoch": 0.239442751414889,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.20198591421544e-05,
|
|
"loss": 6.5538,
|
|
"mean_token_accuracy": 0.12958500310778617,
|
|
"num_tokens": 7040184.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.813288307189941,
|
|
"epoch": 0.2397776363819028,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 7.151434668744517e-05,
|
|
"loss": 6.515,
|
|
"mean_token_accuracy": 0.12613528594374657,
|
|
"num_tokens": 7049738.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.825316667556763,
|
|
"epoch": 0.24011252134891664,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.101441289471153e-05,
|
|
"loss": 6.517,
|
|
"mean_token_accuracy": 0.1323557436466217,
|
|
"num_tokens": 7059408.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.795191240310669,
|
|
"epoch": 0.24044740631593048,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 7.052007146989098e-05,
|
|
"loss": 6.5562,
|
|
"mean_token_accuracy": 0.12746668756008148,
|
|
"num_tokens": 7068852.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.763497257232666,
|
|
"epoch": 0.24078229128294432,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 7.003133596560341e-05,
|
|
"loss": 6.498,
|
|
"mean_token_accuracy": 0.13406475707888604,
|
|
"num_tokens": 7077854.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.750020265579224,
|
|
"epoch": 0.24111717624995813,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 6.954821978077952e-05,
|
|
"loss": 6.424,
|
|
"mean_token_accuracy": 0.13775352835655214,
|
|
"num_tokens": 7087014.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.74197359085083,
|
|
"epoch": 0.24145206121697196,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 6.907073616029356e-05,
|
|
"loss": 6.4368,
|
|
"mean_token_accuracy": 0.14112297743558883,
|
|
"num_tokens": 7097055.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.717160654067993,
|
|
"epoch": 0.2417869461839858,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.85988981946002e-05,
|
|
"loss": 6.4551,
|
|
"mean_token_accuracy": 0.14057836383581163,
|
|
"num_tokens": 7106076.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.761683702468872,
|
|
"epoch": 0.24212183115099964,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 6.813271881937564e-05,
|
|
"loss": 6.5358,
|
|
"mean_token_accuracy": 0.13456962928175925,
|
|
"num_tokens": 7115131.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.806710481643677,
|
|
"epoch": 0.24245671611801345,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 6.767221081516286e-05,
|
|
"loss": 6.5199,
|
|
"mean_token_accuracy": 0.1338411808013916,
|
|
"num_tokens": 7124943.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.74193229675293,
|
|
"epoch": 0.2427916010850273,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 6.72173868070215e-05,
|
|
"loss": 6.5297,
|
|
"mean_token_accuracy": 0.14725314006209372,
|
|
"num_tokens": 7134701.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.779576206207276,
|
|
"epoch": 0.24312648605204112,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 6.676825926418149e-05,
|
|
"loss": 6.4876,
|
|
"mean_token_accuracy": 0.13964010179042816,
|
|
"num_tokens": 7144939.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.815815877914429,
|
|
"epoch": 0.24346137101905496,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.632484049970122e-05,
|
|
"loss": 6.5553,
|
|
"mean_token_accuracy": 0.12960358187556267,
|
|
"num_tokens": 7154420.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.823481178283691,
|
|
"epoch": 0.24379625598606877,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.588714267013019e-05,
|
|
"loss": 6.5534,
|
|
"mean_token_accuracy": 0.1280705638229847,
|
|
"num_tokens": 7164160.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.767119121551514,
|
|
"epoch": 0.2441311409530826,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 6.545517777517544e-05,
|
|
"loss": 6.4802,
|
|
"mean_token_accuracy": 0.1350911557674408,
|
|
"num_tokens": 7173249.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.782384061813355,
|
|
"epoch": 0.24446602592009645,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 6.502895765737281e-05,
|
|
"loss": 6.4483,
|
|
"mean_token_accuracy": 0.13474421724677085,
|
|
"num_tokens": 7181549.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.752167701721191,
|
|
"epoch": 0.24480091088711028,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 6.460849400176212e-05,
|
|
"loss": 6.4703,
|
|
"mean_token_accuracy": 0.13417914360761643,
|
|
"num_tokens": 7192608.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.768589448928833,
|
|
"epoch": 0.24513579585412412,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 6.419379833556694e-05,
|
|
"loss": 6.4752,
|
|
"mean_token_accuracy": 0.13951172679662704,
|
|
"num_tokens": 7202352.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.715800189971924,
|
|
"epoch": 0.24547068082113793,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 6.378488202787835e-05,
|
|
"loss": 6.467,
|
|
"mean_token_accuracy": 0.1401784211397171,
|
|
"num_tokens": 7211803.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.769896936416626,
|
|
"epoch": 0.24580556578815177,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.33817562893435e-05,
|
|
"loss": 6.4598,
|
|
"mean_token_accuracy": 0.13421895503997802,
|
|
"num_tokens": 7220042.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.7630424976348875,
|
|
"epoch": 0.2461404507551656,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.29844321718582e-05,
|
|
"loss": 6.4167,
|
|
"mean_token_accuracy": 0.13856958076357842,
|
|
"num_tokens": 7229388.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.651368284225464,
|
|
"epoch": 0.24647533572217944,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 6.259292056826383e-05,
|
|
"loss": 6.3175,
|
|
"mean_token_accuracy": 0.1456449620425701,
|
|
"num_tokens": 7239711.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.7098479747772215,
|
|
"epoch": 0.24681022068919325,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 6.220723221204873e-05,
|
|
"loss": 6.4377,
|
|
"mean_token_accuracy": 0.14019205644726754,
|
|
"num_tokens": 7248677.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.75185375213623,
|
|
"epoch": 0.2471451056562071,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 6.182737767705406e-05,
|
|
"loss": 6.4213,
|
|
"mean_token_accuracy": 0.1440846912562847,
|
|
"num_tokens": 7258100.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.802224636077881,
|
|
"epoch": 0.24747999062322093,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 6.145336737718375e-05,
|
|
"loss": 6.5147,
|
|
"mean_token_accuracy": 0.13601332157850266,
|
|
"num_tokens": 7268512.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.761419486999512,
|
|
"epoch": 0.24781487559023477,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 6.10852115661191e-05,
|
|
"loss": 6.4822,
|
|
"mean_token_accuracy": 0.13151130005717276,
|
|
"num_tokens": 7278403.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.767429494857788,
|
|
"epoch": 0.24814976055724858,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 6.072292033703766e-05,
|
|
"loss": 6.5395,
|
|
"mean_token_accuracy": 0.1296151578426361,
|
|
"num_tokens": 7288197.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.783853816986084,
|
|
"epoch": 0.24848464552426242,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 6.036650362233648e-05,
|
|
"loss": 6.5209,
|
|
"mean_token_accuracy": 0.13040124699473382,
|
|
"num_tokens": 7298411.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.759630346298218,
|
|
"epoch": 0.24881953049127625,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 6.0015971193359824e-05,
|
|
"loss": 6.4612,
|
|
"mean_token_accuracy": 0.1391223095357418,
|
|
"num_tokens": 7307513.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.67000904083252,
|
|
"epoch": 0.2491544154582901,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.9671332660131306e-05,
|
|
"loss": 6.3389,
|
|
"mean_token_accuracy": 0.13932134881615638,
|
|
"num_tokens": 7317177.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.710117816925049,
|
|
"epoch": 0.2494893004253039,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 5.933259747109042e-05,
|
|
"loss": 6.3515,
|
|
"mean_token_accuracy": 0.1432790733873844,
|
|
"num_tokens": 7327884.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.769780969619751,
|
|
"epoch": 0.24982418539231774,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 5.899977491283351e-05,
|
|
"loss": 6.4868,
|
|
"mean_token_accuracy": 0.137585336714983,
|
|
"num_tokens": 7337335.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.7806394577026365,
|
|
"epoch": 0.25015907035933155,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 5.867287410985908e-05,
|
|
"loss": 6.5071,
|
|
"mean_token_accuracy": 0.13311323150992393,
|
|
"num_tokens": 7347297.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.7589140892028805,
|
|
"epoch": 0.2504939553263454,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 5.835190402431779e-05,
|
|
"loss": 6.4553,
|
|
"mean_token_accuracy": 0.13403227776288987,
|
|
"num_tokens": 7355806.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.7398083209991455,
|
|
"epoch": 0.2508288402933592,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 5.803687345576673e-05,
|
|
"loss": 6.5699,
|
|
"mean_token_accuracy": 0.13343106284737588,
|
|
"num_tokens": 7365990.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.778984212875367,
|
|
"epoch": 0.25116372526037306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 5.7727791040927977e-05,
|
|
"loss": 6.4572,
|
|
"mean_token_accuracy": 0.13448210880160333,
|
|
"num_tokens": 7375100.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.770331335067749,
|
|
"epoch": 0.2514986102273869,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 5.742466525345213e-05,
|
|
"loss": 6.4228,
|
|
"mean_token_accuracy": 0.1388053260743618,
|
|
"num_tokens": 7385594.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.788948726654053,
|
|
"epoch": 0.25183349519440074,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 5.7127504403685775e-05,
|
|
"loss": 6.5798,
|
|
"mean_token_accuracy": 0.13059211820363997,
|
|
"num_tokens": 7396379.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.8178774356842045,
|
|
"epoch": 0.2521683801614146,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 5.6836316638443664e-05,
|
|
"loss": 6.5223,
|
|
"mean_token_accuracy": 0.1305044874548912,
|
|
"num_tokens": 7406403.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.785275077819824,
|
|
"epoch": 0.2525032651284284,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 5.655110994078553e-05,
|
|
"loss": 6.4852,
|
|
"mean_token_accuracy": 0.1422339215874672,
|
|
"num_tokens": 7415574.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.789396047592163,
|
|
"epoch": 0.2528381500954422,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 5.6271892129797056e-05,
|
|
"loss": 6.4768,
|
|
"mean_token_accuracy": 0.13738657534122467,
|
|
"num_tokens": 7424565.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.739390754699707,
|
|
"epoch": 0.25317303506245603,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.599867086037556e-05,
|
|
"loss": 6.4599,
|
|
"mean_token_accuracy": 0.13474917858839036,
|
|
"num_tokens": 7433692.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.758706665039062,
|
|
"epoch": 0.25350792002946987,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.573145362302012e-05,
|
|
"loss": 6.3798,
|
|
"mean_token_accuracy": 0.14275875464081764,
|
|
"num_tokens": 7442101.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.780257892608643,
|
|
"epoch": 0.2538428049964837,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 5.5470247743626404e-05,
|
|
"loss": 6.5241,
|
|
"mean_token_accuracy": 0.12314473614096641,
|
|
"num_tokens": 7451452.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.765549421310425,
|
|
"epoch": 0.25417768996349754,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 5.5215060383285414e-05,
|
|
"loss": 6.5275,
|
|
"mean_token_accuracy": 0.1351473353803158,
|
|
"num_tokens": 7461506.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.760830354690552,
|
|
"epoch": 0.2545125749305114,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.496589853808759e-05,
|
|
"loss": 6.5136,
|
|
"mean_token_accuracy": 0.13388604745268823,
|
|
"num_tokens": 7471465.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.791913747787476,
|
|
"epoch": 0.2548474598975252,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.47227690389308e-05,
|
|
"loss": 6.4081,
|
|
"mean_token_accuracy": 0.14004301354289056,
|
|
"num_tokens": 7480441.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.7986242294311525,
|
|
"epoch": 0.25518234486453906,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 5.448567855133306e-05,
|
|
"loss": 6.5241,
|
|
"mean_token_accuracy": 0.13097715452313424,
|
|
"num_tokens": 7489967.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.814978504180909,
|
|
"epoch": 0.25551722983155284,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 5.425463357524986e-05,
|
|
"loss": 6.5427,
|
|
"mean_token_accuracy": 0.13041020110249518,
|
|
"num_tokens": 7500027.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.808373355865479,
|
|
"epoch": 0.2558521147985667,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 5.402964044489591e-05,
|
|
"loss": 6.464,
|
|
"mean_token_accuracy": 0.13352129757404327,
|
|
"num_tokens": 7509734.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.771735191345215,
|
|
"epoch": 0.2561869997655805,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 5.381070532857153e-05,
|
|
"loss": 6.4631,
|
|
"mean_token_accuracy": 0.13360599279403687,
|
|
"num_tokens": 7519020.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.774158954620361,
|
|
"epoch": 0.25652188473259435,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.359783422849357e-05,
|
|
"loss": 6.4214,
|
|
"mean_token_accuracy": 0.1385476030409336,
|
|
"num_tokens": 7527971.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.781787967681884,
|
|
"epoch": 0.2568567696996082,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.3391032980630736e-05,
|
|
"loss": 6.57,
|
|
"mean_token_accuracy": 0.13175242990255356,
|
|
"num_tokens": 7538199.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.7477575778961185,
|
|
"epoch": 0.257191654666622,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 5.31903072545437e-05,
|
|
"loss": 6.4332,
|
|
"mean_token_accuracy": 0.13715762495994568,
|
|
"num_tokens": 7547550.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.7370538234710695,
|
|
"epoch": 0.25752653963363586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.29956625532297e-05,
|
|
"loss": 6.3493,
|
|
"mean_token_accuracy": 0.1425662726163864,
|
|
"num_tokens": 7556479.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.725951147079468,
|
|
"epoch": 0.2578614246006497,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.280710421297146e-05,
|
|
"loss": 6.4147,
|
|
"mean_token_accuracy": 0.13666378185153008,
|
|
"num_tokens": 7566512.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.750552988052368,
|
|
"epoch": 0.2581963095676635,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 5.2624637403191165e-05,
|
|
"loss": 6.4856,
|
|
"mean_token_accuracy": 0.14022355601191522,
|
|
"num_tokens": 7575814.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.774207592010498,
|
|
"epoch": 0.2585311945346773,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.2448267126308605e-05,
|
|
"loss": 6.5371,
|
|
"mean_token_accuracy": 0.13062327057123185,
|
|
"num_tokens": 7585838.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.7971278667449955,
|
|
"epoch": 0.25886607950169116,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 5.2277998217603954e-05,
|
|
"loss": 6.4468,
|
|
"mean_token_accuracy": 0.13751493766903877,
|
|
"num_tokens": 7595775.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.772003221511841,
|
|
"epoch": 0.259200964468705,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.211383534508541e-05,
|
|
"loss": 6.5027,
|
|
"mean_token_accuracy": 0.1348140373826027,
|
|
"num_tokens": 7606899.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.80793514251709,
|
|
"epoch": 0.25953584943571884,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 5.1955783009361044e-05,
|
|
"loss": 6.5387,
|
|
"mean_token_accuracy": 0.13264777660369872,
|
|
"num_tokens": 7616960.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.764792156219483,
|
|
"epoch": 0.2598707344027327,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 5.180384554351543e-05,
|
|
"loss": 6.473,
|
|
"mean_token_accuracy": 0.13825612589716912,
|
|
"num_tokens": 7626461.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.765843343734741,
|
|
"epoch": 0.2602056193697465,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 5.1658027112990976e-05,
|
|
"loss": 6.5356,
|
|
"mean_token_accuracy": 0.13480739966034888,
|
|
"num_tokens": 7635714.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.726837491989135,
|
|
"epoch": 0.26054050433676035,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 5.151833171547365e-05,
|
|
"loss": 6.4661,
|
|
"mean_token_accuracy": 0.13827862814068795,
|
|
"num_tokens": 7646281.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.772043752670288,
|
|
"epoch": 0.26087538930377413,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 5.1384763180783274e-05,
|
|
"loss": 6.4824,
|
|
"mean_token_accuracy": 0.13868937864899636,
|
|
"num_tokens": 7655099.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.783730697631836,
|
|
"epoch": 0.26121027427078797,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.125732517076876e-05,
|
|
"loss": 6.5056,
|
|
"mean_token_accuracy": 0.1275267072021961,
|
|
"num_tokens": 7663967.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.823947763442993,
|
|
"epoch": 0.2615451592378018,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 5.113602117920747e-05,
|
|
"loss": 6.4972,
|
|
"mean_token_accuracy": 0.13771382421255113,
|
|
"num_tokens": 7674426.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.836914443969727,
|
|
"epoch": 0.26188004420481564,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 5.102085453170966e-05,
|
|
"loss": 6.5102,
|
|
"mean_token_accuracy": 0.12969584837555886,
|
|
"num_tokens": 7684506.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.775651741027832,
|
|
"epoch": 0.2622149291718295,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 5.091182838562709e-05,
|
|
"loss": 6.4801,
|
|
"mean_token_accuracy": 0.1315931849181652,
|
|
"num_tokens": 7694230.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.822858476638794,
|
|
"epoch": 0.2625498141388433,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.0808945729966716e-05,
|
|
"loss": 6.5988,
|
|
"mean_token_accuracy": 0.12708214297890663,
|
|
"num_tokens": 7704520.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.7741209983825685,
|
|
"epoch": 0.26288469910585716,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 5.071220938530844e-05,
|
|
"loss": 6.4409,
|
|
"mean_token_accuracy": 0.14077352732419968,
|
|
"num_tokens": 7713428.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.766770315170288,
|
|
"epoch": 0.263219584072871,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 5.0621622003728026e-05,
|
|
"loss": 6.548,
|
|
"mean_token_accuracy": 0.13225557506084443,
|
|
"num_tokens": 7722858.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.810342407226562,
|
|
"epoch": 0.2635544690398848,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.053718606872433e-05,
|
|
"loss": 6.494,
|
|
"mean_token_accuracy": 0.1297669731080532,
|
|
"num_tokens": 7733742.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.778527355194091,
|
|
"epoch": 0.2638893540068986,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 5.0458903895151134e-05,
|
|
"loss": 6.4738,
|
|
"mean_token_accuracy": 0.13769279047846794,
|
|
"num_tokens": 7743439.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.805730724334717,
|
|
"epoch": 0.26422423897391245,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 5.038677762915381e-05,
|
|
"loss": 6.5302,
|
|
"mean_token_accuracy": 0.12858908101916314,
|
|
"num_tokens": 7753998.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.801490592956543,
|
|
"epoch": 0.2645591239409263,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 5.032080924811033e-05,
|
|
"loss": 6.4831,
|
|
"mean_token_accuracy": 0.13166142255067825,
|
|
"num_tokens": 7763404.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.814216184616089,
|
|
"epoch": 0.2648940089079401,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 5.026100056057718e-05,
|
|
"loss": 6.5391,
|
|
"mean_token_accuracy": 0.12888559177517891,
|
|
"num_tokens": 7773491.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.816267633438111,
|
|
"epoch": 0.26522889387495396,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.0207353206239764e-05,
|
|
"loss": 6.5792,
|
|
"mean_token_accuracy": 0.1289291948080063,
|
|
"num_tokens": 7783161.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.8376367568969725,
|
|
"epoch": 0.2655637788419678,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.0159868655867436e-05,
|
|
"loss": 6.5698,
|
|
"mean_token_accuracy": 0.12774034067988396,
|
|
"num_tokens": 7793570.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.79964017868042,
|
|
"epoch": 0.26589866380898164,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 5.011854821127305e-05,
|
|
"loss": 6.5081,
|
|
"mean_token_accuracy": 0.1318251185119152,
|
|
"num_tokens": 7803835.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.794245958328247,
|
|
"epoch": 0.2662335487759954,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 5.008339300527755e-05,
|
|
"loss": 6.5205,
|
|
"mean_token_accuracy": 0.1366777576506138,
|
|
"num_tokens": 7813283.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.731097888946533,
|
|
"epoch": 0.26656843374300926,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 5.005440400167859e-05,
|
|
"loss": 6.4415,
|
|
"mean_token_accuracy": 0.13983828127384185,
|
|
"num_tokens": 7823322.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.769813394546508,
|
|
"epoch": 0.2669033187100231,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 5.003158199522442e-05,
|
|
"loss": 6.5394,
|
|
"mean_token_accuracy": 0.13228418529033661,
|
|
"num_tokens": 7833140.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.773566007614136,
|
|
"epoch": 0.26723820367703693,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 5.0014927611591806e-05,
|
|
"loss": 6.5057,
|
|
"mean_token_accuracy": 0.1311323657631874,
|
|
"num_tokens": 7843399.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.758834028244019,
|
|
"epoch": 0.26757308864405077,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 5.000444130736916e-05,
|
|
"loss": 6.4784,
|
|
"mean_token_accuracy": 0.13214338421821595,
|
|
"num_tokens": 7853680.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.7637241840362545,
|
|
"epoch": 0.2679079736110646,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.0000123370043736e-05,
|
|
"loss": 6.5295,
|
|
"mean_token_accuracy": 0.1373077802360058,
|
|
"num_tokens": 7864558.0,
|
|
"step": 4000
|
|
}
|
|
],
|
|
"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": true
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 1.2536032315392e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|