22277 lines
610 KiB
JSON
22277 lines
610 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.113504556752279,
|
|
"eval_steps": 500,
|
|
"global_step": 11000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.7426025390625,
|
|
"epoch": 0.004142502071251036,
|
|
"grad_norm": 4.96875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7521,
|
|
"mean_token_accuracy": 9.009009227156639e-05,
|
|
"num_tokens": 9149.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742637062072754,
|
|
"epoch": 0.008285004142502071,
|
|
"grad_norm": 5.3125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7074,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 16800.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742610836029053,
|
|
"epoch": 0.012427506213753107,
|
|
"grad_norm": 4.90625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.743,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 26617.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742677021026612,
|
|
"epoch": 0.016570008285004142,
|
|
"grad_norm": 5.15625,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6772,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 35855.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742655372619629,
|
|
"epoch": 0.020712510356255178,
|
|
"grad_norm": 4.40625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5975,
|
|
"mean_token_accuracy": 0.002011376223526895,
|
|
"num_tokens": 45644.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742458724975586,
|
|
"epoch": 0.024855012427506214,
|
|
"grad_norm": 3.859375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4485,
|
|
"mean_token_accuracy": 0.019727057497948407,
|
|
"num_tokens": 54641.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.741547870635987,
|
|
"epoch": 0.02899751449875725,
|
|
"grad_norm": 3.40625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3138,
|
|
"mean_token_accuracy": 0.04090950228273869,
|
|
"num_tokens": 63924.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.739116668701172,
|
|
"epoch": 0.033140016570008285,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2108,
|
|
"mean_token_accuracy": 0.041662289202213286,
|
|
"num_tokens": 73452.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.735450172424317,
|
|
"epoch": 0.037282518641259324,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0929,
|
|
"mean_token_accuracy": 0.039270054548978806,
|
|
"num_tokens": 82731.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.733051109313966,
|
|
"epoch": 0.041425020712510356,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.0119,
|
|
"mean_token_accuracy": 0.04254167750477791,
|
|
"num_tokens": 92626.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.732081508636474,
|
|
"epoch": 0.045567522783761395,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9713,
|
|
"mean_token_accuracy": 0.04115805197507143,
|
|
"num_tokens": 101380.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.730613136291504,
|
|
"epoch": 0.04971002485501243,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8815,
|
|
"mean_token_accuracy": 0.041873040795326236,
|
|
"num_tokens": 109895.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.728298664093018,
|
|
"epoch": 0.053852526926263466,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8626,
|
|
"mean_token_accuracy": 0.04000609517097473,
|
|
"num_tokens": 119203.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.7264967918396,
|
|
"epoch": 0.0579950289975145,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7547,
|
|
"mean_token_accuracy": 0.03999764695763588,
|
|
"num_tokens": 127572.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.723791694641113,
|
|
"epoch": 0.06213753106876554,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7222,
|
|
"mean_token_accuracy": 0.042604190111160276,
|
|
"num_tokens": 137161.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.719281482696534,
|
|
"epoch": 0.06628003314001657,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6617,
|
|
"mean_token_accuracy": 0.04084589965641498,
|
|
"num_tokens": 145991.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.712574768066407,
|
|
"epoch": 0.07042253521126761,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.598,
|
|
"mean_token_accuracy": 0.041171186976134776,
|
|
"num_tokens": 154927.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.706455898284911,
|
|
"epoch": 0.07456503728251865,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4593,
|
|
"mean_token_accuracy": 0.042302171140909194,
|
|
"num_tokens": 163234.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.699676418304444,
|
|
"epoch": 0.07870753935376967,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4018,
|
|
"mean_token_accuracy": 0.04330670088529587,
|
|
"num_tokens": 171554.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.688521003723144,
|
|
"epoch": 0.08285004142502071,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3311,
|
|
"mean_token_accuracy": 0.047984727472066876,
|
|
"num_tokens": 180171.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.67230396270752,
|
|
"epoch": 0.08699254349627175,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2155,
|
|
"mean_token_accuracy": 0.057312173396348955,
|
|
"num_tokens": 189216.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.645287895202637,
|
|
"epoch": 0.09113504556752279,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.2219,
|
|
"mean_token_accuracy": 0.052681009843945505,
|
|
"num_tokens": 200022.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.617320728302001,
|
|
"epoch": 0.09527754763877382,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.0422,
|
|
"mean_token_accuracy": 0.05372999124228954,
|
|
"num_tokens": 208686.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.564020252227783,
|
|
"epoch": 0.09942004971002485,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.9322,
|
|
"mean_token_accuracy": 0.054267140850424764,
|
|
"num_tokens": 218266.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.5111478805542,
|
|
"epoch": 0.1035625517812759,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8288,
|
|
"mean_token_accuracy": 0.04845710247755051,
|
|
"num_tokens": 227658.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.454801654815673,
|
|
"epoch": 0.10770505385252693,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6344,
|
|
"mean_token_accuracy": 0.06026250720024109,
|
|
"num_tokens": 236301.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.368635272979736,
|
|
"epoch": 0.11184755592377796,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.644,
|
|
"mean_token_accuracy": 0.048985954746603964,
|
|
"num_tokens": 246355.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.31434211730957,
|
|
"epoch": 0.115990057995029,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4805,
|
|
"mean_token_accuracy": 0.05331834591925144,
|
|
"num_tokens": 256023.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.216940021514892,
|
|
"epoch": 0.12013256006628004,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.4309,
|
|
"mean_token_accuracy": 0.04547450765967369,
|
|
"num_tokens": 266893.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.09643030166626,
|
|
"epoch": 0.12427506213753108,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1871,
|
|
"mean_token_accuracy": 0.0535514272749424,
|
|
"num_tokens": 276075.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.940307807922363,
|
|
"epoch": 0.12841756420878211,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1018,
|
|
"mean_token_accuracy": 0.055915596336126326,
|
|
"num_tokens": 285529.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.774145030975342,
|
|
"epoch": 0.13256006628003314,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.0263,
|
|
"mean_token_accuracy": 0.05546179860830307,
|
|
"num_tokens": 294403.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.571882343292236,
|
|
"epoch": 0.13670256835128416,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9661,
|
|
"mean_token_accuracy": 0.05883818194270134,
|
|
"num_tokens": 303507.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.380561351776123,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7977,
|
|
"mean_token_accuracy": 0.06465772800147533,
|
|
"num_tokens": 312393.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.121123886108398,
|
|
"epoch": 0.14498757249378624,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7071,
|
|
"mean_token_accuracy": 0.06362416185438632,
|
|
"num_tokens": 321309.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.919694709777833,
|
|
"epoch": 0.1491300745650373,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7209,
|
|
"mean_token_accuracy": 0.06159159280359745,
|
|
"num_tokens": 330217.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.755216789245605,
|
|
"epoch": 0.15327257663628832,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.6795,
|
|
"mean_token_accuracy": 0.061438082903623584,
|
|
"num_tokens": 339587.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.5862530708313,
|
|
"epoch": 0.15741507870753935,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5864,
|
|
"mean_token_accuracy": 0.06275138966739177,
|
|
"num_tokens": 349367.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.414569091796874,
|
|
"epoch": 0.1615575807787904,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4852,
|
|
"mean_token_accuracy": 0.06507482491433621,
|
|
"num_tokens": 358272.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.34211015701294,
|
|
"epoch": 0.16570008285004142,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.6662,
|
|
"mean_token_accuracy": 0.06489905305206775,
|
|
"num_tokens": 367709.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.344364166259766,
|
|
"epoch": 0.16984258492129245,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5038,
|
|
"mean_token_accuracy": 0.06977917924523354,
|
|
"num_tokens": 376180.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.172031211853028,
|
|
"epoch": 0.1739850869925435,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6071,
|
|
"mean_token_accuracy": 0.06480413824319839,
|
|
"num_tokens": 385636.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.20558671951294,
|
|
"epoch": 0.17812758906379453,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.4616,
|
|
"mean_token_accuracy": 0.07232532948255539,
|
|
"num_tokens": 394486.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.116187858581544,
|
|
"epoch": 0.18227009113504558,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.4215,
|
|
"mean_token_accuracy": 0.06673903428018094,
|
|
"num_tokens": 403765.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.066840410232544,
|
|
"epoch": 0.1864125932062966,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.4037,
|
|
"mean_token_accuracy": 0.07279244847595692,
|
|
"num_tokens": 412988.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.047943782806396,
|
|
"epoch": 0.19055509527754763,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3598,
|
|
"mean_token_accuracy": 0.07073828913271427,
|
|
"num_tokens": 422096.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.999078845977783,
|
|
"epoch": 0.19469759734879868,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.441,
|
|
"mean_token_accuracy": 0.06653320901095867,
|
|
"num_tokens": 431333.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.124928903579711,
|
|
"epoch": 0.1988400994200497,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4798,
|
|
"mean_token_accuracy": 0.07552412785589695,
|
|
"num_tokens": 440382.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.967213916778564,
|
|
"epoch": 0.20298260149130073,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.3739,
|
|
"mean_token_accuracy": 0.07711177170276642,
|
|
"num_tokens": 448469.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.066874313354493,
|
|
"epoch": 0.2071251035625518,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4382,
|
|
"mean_token_accuracy": 0.07283047214150429,
|
|
"num_tokens": 457356.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.006765127182007,
|
|
"epoch": 0.2112676056338028,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.4372,
|
|
"mean_token_accuracy": 0.07371915206313133,
|
|
"num_tokens": 465928.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.962953233718872,
|
|
"epoch": 0.21541010770505387,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3086,
|
|
"mean_token_accuracy": 0.07748371884226798,
|
|
"num_tokens": 475269.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.846234321594238,
|
|
"epoch": 0.2195526097763049,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.2837,
|
|
"mean_token_accuracy": 0.07470350190997124,
|
|
"num_tokens": 485450.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.911148262023926,
|
|
"epoch": 0.22369511184755592,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3174,
|
|
"mean_token_accuracy": 0.07727364040911197,
|
|
"num_tokens": 494029.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.862251615524292,
|
|
"epoch": 0.22783761391880697,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.2856,
|
|
"mean_token_accuracy": 0.07773918211460114,
|
|
"num_tokens": 503253.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.789915609359741,
|
|
"epoch": 0.231980115990058,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2873,
|
|
"mean_token_accuracy": 0.07239806838333607,
|
|
"num_tokens": 512857.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.920849275588989,
|
|
"epoch": 0.23612261806130902,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3294,
|
|
"mean_token_accuracy": 0.07429277002811432,
|
|
"num_tokens": 522319.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.767004537582397,
|
|
"epoch": 0.24026512013256007,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2731,
|
|
"mean_token_accuracy": 0.07617050744593143,
|
|
"num_tokens": 531469.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.817885971069336,
|
|
"epoch": 0.2444076222038111,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2883,
|
|
"mean_token_accuracy": 0.07853077799081802,
|
|
"num_tokens": 540470.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.739614582061767,
|
|
"epoch": 0.24855012427506215,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2711,
|
|
"mean_token_accuracy": 0.07688240185379983,
|
|
"num_tokens": 550317.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.804083776473999,
|
|
"epoch": 0.2526926263463132,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2542,
|
|
"mean_token_accuracy": 0.07694350108504296,
|
|
"num_tokens": 559409.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.721450233459473,
|
|
"epoch": 0.25683512841756423,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2393,
|
|
"mean_token_accuracy": 0.0897148072719574,
|
|
"num_tokens": 568726.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.8209740161895756,
|
|
"epoch": 0.2609776304888152,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.26,
|
|
"mean_token_accuracy": 0.07505979798734189,
|
|
"num_tokens": 577926.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.7856145858764645,
|
|
"epoch": 0.2651201325600663,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2553,
|
|
"mean_token_accuracy": 0.07944877073168755,
|
|
"num_tokens": 586695.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.730410480499268,
|
|
"epoch": 0.26926263463131733,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.264,
|
|
"mean_token_accuracy": 0.07737773731350898,
|
|
"num_tokens": 596250.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.764742994308472,
|
|
"epoch": 0.27340513670256833,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2863,
|
|
"mean_token_accuracy": 0.07472797855734825,
|
|
"num_tokens": 605761.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.720160007476807,
|
|
"epoch": 0.2775476387738194,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.2106,
|
|
"mean_token_accuracy": 0.07840372547507286,
|
|
"num_tokens": 614313.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.699429607391357,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2329,
|
|
"mean_token_accuracy": 0.079537483304739,
|
|
"num_tokens": 624470.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.716359567642212,
|
|
"epoch": 0.28583264291632143,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1781,
|
|
"mean_token_accuracy": 0.07983956411480904,
|
|
"num_tokens": 633458.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.705728006362915,
|
|
"epoch": 0.2899751449875725,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1396,
|
|
"mean_token_accuracy": 0.08111847266554832,
|
|
"num_tokens": 643251.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.584441757202148,
|
|
"epoch": 0.29411764705882354,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.2011,
|
|
"mean_token_accuracy": 0.08007878810167313,
|
|
"num_tokens": 652005.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.649393129348755,
|
|
"epoch": 0.2982601491300746,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0934,
|
|
"mean_token_accuracy": 0.08189089968800545,
|
|
"num_tokens": 661003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.623857879638672,
|
|
"epoch": 0.3024026512013256,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0933,
|
|
"mean_token_accuracy": 0.08178795203566551,
|
|
"num_tokens": 670221.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.750538110733032,
|
|
"epoch": 0.30654515327257664,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.3477,
|
|
"mean_token_accuracy": 0.07552824206650258,
|
|
"num_tokens": 680756.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.595584869384766,
|
|
"epoch": 0.3106876553438277,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.0678,
|
|
"mean_token_accuracy": 0.08516154885292053,
|
|
"num_tokens": 689641.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.549710321426391,
|
|
"epoch": 0.3148301574150787,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.995,
|
|
"mean_token_accuracy": 0.0825497955083847,
|
|
"num_tokens": 698209.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.580410575866699,
|
|
"epoch": 0.31897265948632975,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9957,
|
|
"mean_token_accuracy": 0.08926377072930336,
|
|
"num_tokens": 706712.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.595743894577026,
|
|
"epoch": 0.3231151615575808,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1863,
|
|
"mean_token_accuracy": 0.08130085095763206,
|
|
"num_tokens": 716087.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.5478174686431885,
|
|
"epoch": 0.3272576636288318,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.0828,
|
|
"mean_token_accuracy": 0.07734477743506432,
|
|
"num_tokens": 725913.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.606314182281494,
|
|
"epoch": 0.33140016570008285,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1209,
|
|
"mean_token_accuracy": 0.0859277956187725,
|
|
"num_tokens": 735020.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.527130556106568,
|
|
"epoch": 0.3355426677713339,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.0165,
|
|
"mean_token_accuracy": 0.07862741872668266,
|
|
"num_tokens": 744299.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.538954830169677,
|
|
"epoch": 0.3396851698425849,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9993,
|
|
"mean_token_accuracy": 0.08304714113473892,
|
|
"num_tokens": 753231.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.5157403469085695,
|
|
"epoch": 0.34382767191383595,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.0619,
|
|
"mean_token_accuracy": 0.08629989027976989,
|
|
"num_tokens": 763043.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.627204179763794,
|
|
"epoch": 0.347970173985087,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0861,
|
|
"mean_token_accuracy": 0.08552133813500404,
|
|
"num_tokens": 771977.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.432950973510742,
|
|
"epoch": 0.352112676056338,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9878,
|
|
"mean_token_accuracy": 0.08852855414152146,
|
|
"num_tokens": 780285.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.566639614105225,
|
|
"epoch": 0.35625517812758906,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.047,
|
|
"mean_token_accuracy": 0.08173825293779373,
|
|
"num_tokens": 789980.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.483531761169433,
|
|
"epoch": 0.3603976801988401,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.0026,
|
|
"mean_token_accuracy": 0.09233896881341934,
|
|
"num_tokens": 798700.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.38806381225586,
|
|
"epoch": 0.36454018227009116,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.059,
|
|
"mean_token_accuracy": 0.08308355435729027,
|
|
"num_tokens": 808758.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.56745285987854,
|
|
"epoch": 0.36868268434134216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0227,
|
|
"mean_token_accuracy": 0.0889634720981121,
|
|
"num_tokens": 817692.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.4513147354125975,
|
|
"epoch": 0.3728251864125932,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.074,
|
|
"mean_token_accuracy": 0.09340473040938377,
|
|
"num_tokens": 827052.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.493870782852173,
|
|
"epoch": 0.37696768848384427,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.0443,
|
|
"mean_token_accuracy": 0.09431338384747505,
|
|
"num_tokens": 836442.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.542994403839112,
|
|
"epoch": 0.38111019055509526,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9475,
|
|
"mean_token_accuracy": 0.08114949613809586,
|
|
"num_tokens": 846441.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.422606658935547,
|
|
"epoch": 0.3852526926263463,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9848,
|
|
"mean_token_accuracy": 0.09037281647324562,
|
|
"num_tokens": 856108.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.409952163696289,
|
|
"epoch": 0.38939519469759737,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.9198,
|
|
"mean_token_accuracy": 0.08507395833730698,
|
|
"num_tokens": 865788.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.413805341720581,
|
|
"epoch": 0.39353769676884837,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9738,
|
|
"mean_token_accuracy": 0.0836036428809166,
|
|
"num_tokens": 874833.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.285798835754394,
|
|
"epoch": 0.3976801988400994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.9254,
|
|
"mean_token_accuracy": 0.08790519461035728,
|
|
"num_tokens": 883685.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.440755939483642,
|
|
"epoch": 0.40182270091135047,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9147,
|
|
"mean_token_accuracy": 0.08789389207959175,
|
|
"num_tokens": 893908.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.403847885131836,
|
|
"epoch": 0.40596520298260147,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0129,
|
|
"mean_token_accuracy": 0.08363201320171357,
|
|
"num_tokens": 903081.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.289476585388184,
|
|
"epoch": 0.4101077050538525,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.8789,
|
|
"mean_token_accuracy": 0.08956534340977669,
|
|
"num_tokens": 912515.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.4823840141296385,
|
|
"epoch": 0.4142502071251036,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.8952,
|
|
"mean_token_accuracy": 0.09587915241718292,
|
|
"num_tokens": 921056.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4142502071251036,
|
|
"eval_entropy": 7.1376285746153485,
|
|
"eval_loss": 7.034562110900879,
|
|
"eval_mean_token_accuracy": 0.08812946460839273,
|
|
"eval_num_tokens": 921056.0,
|
|
"eval_runtime": 2.122,
|
|
"eval_samples_per_second": 1765.767,
|
|
"eval_steps_per_second": 221.015,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.1874003410339355,
|
|
"epoch": 0.4183927091963546,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8368,
|
|
"mean_token_accuracy": 0.09126921370625496,
|
|
"num_tokens": 931112.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.335116291046143,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9556,
|
|
"mean_token_accuracy": 0.08759467005729675,
|
|
"num_tokens": 941120.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.3472545623779295,
|
|
"epoch": 0.4266777133388567,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9077,
|
|
"mean_token_accuracy": 0.08984342738986015,
|
|
"num_tokens": 950041.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.321388053894043,
|
|
"epoch": 0.43082021541010773,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8881,
|
|
"mean_token_accuracy": 0.09532358199357986,
|
|
"num_tokens": 958608.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.433567953109741,
|
|
"epoch": 0.43496271748135873,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.994,
|
|
"mean_token_accuracy": 0.08581585139036178,
|
|
"num_tokens": 966901.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.247237634658814,
|
|
"epoch": 0.4391052195526098,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8413,
|
|
"mean_token_accuracy": 0.09217345640063286,
|
|
"num_tokens": 976198.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.419485569000244,
|
|
"epoch": 0.44324772162386084,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8981,
|
|
"mean_token_accuracy": 0.0871151939034462,
|
|
"num_tokens": 985248.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.186427021026612,
|
|
"epoch": 0.44739022369511183,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8256,
|
|
"mean_token_accuracy": 0.09469342082738877,
|
|
"num_tokens": 993289.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.353890085220337,
|
|
"epoch": 0.4515327257663629,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.8839,
|
|
"mean_token_accuracy": 0.09135557413101196,
|
|
"num_tokens": 1001784.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.35116548538208,
|
|
"epoch": 0.45567522783761394,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8471,
|
|
"mean_token_accuracy": 0.09857566952705384,
|
|
"num_tokens": 1010445.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.249645519256592,
|
|
"epoch": 0.45981772990886494,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.8867,
|
|
"mean_token_accuracy": 0.09312620386481285,
|
|
"num_tokens": 1019084.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.222869443893432,
|
|
"epoch": 0.463960231980116,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8057,
|
|
"mean_token_accuracy": 0.09473200142383575,
|
|
"num_tokens": 1027263.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.348370885848999,
|
|
"epoch": 0.46810273405136704,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.91,
|
|
"mean_token_accuracy": 0.08753768503665924,
|
|
"num_tokens": 1036275.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.228622531890869,
|
|
"epoch": 0.47224523612261804,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.952,
|
|
"mean_token_accuracy": 0.08856018707156181,
|
|
"num_tokens": 1045208.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.336494016647339,
|
|
"epoch": 0.4763877381938691,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8686,
|
|
"mean_token_accuracy": 0.09350670352578164,
|
|
"num_tokens": 1054703.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.101537704467773,
|
|
"epoch": 0.48053024026512015,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7426,
|
|
"mean_token_accuracy": 0.09196772649884224,
|
|
"num_tokens": 1063652.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.175399208068848,
|
|
"epoch": 0.48467274233637114,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8364,
|
|
"mean_token_accuracy": 0.08494803085923194,
|
|
"num_tokens": 1072070.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.374863576889038,
|
|
"epoch": 0.4888152444076222,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.8737,
|
|
"mean_token_accuracy": 0.09276422187685966,
|
|
"num_tokens": 1079992.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.18324236869812,
|
|
"epoch": 0.49295774647887325,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.8582,
|
|
"mean_token_accuracy": 0.09373369887471199,
|
|
"num_tokens": 1088862.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.20685133934021,
|
|
"epoch": 0.4971002485501243,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.6698,
|
|
"mean_token_accuracy": 0.1027169682085514,
|
|
"num_tokens": 1097614.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.092188501358033,
|
|
"epoch": 0.5012427506213754,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.8528,
|
|
"mean_token_accuracy": 0.0964401125907898,
|
|
"num_tokens": 1107347.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.277201890945435,
|
|
"epoch": 0.5053852526926264,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.8723,
|
|
"mean_token_accuracy": 0.09177736937999725,
|
|
"num_tokens": 1116707.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.224560880661011,
|
|
"epoch": 0.5095277547638773,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8637,
|
|
"mean_token_accuracy": 0.09503144547343254,
|
|
"num_tokens": 1126356.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.175086212158203,
|
|
"epoch": 0.5136702568351285,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.8186,
|
|
"mean_token_accuracy": 0.0897763006389141,
|
|
"num_tokens": 1136281.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.1897046089172365,
|
|
"epoch": 0.5178127589063795,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7722,
|
|
"mean_token_accuracy": 0.09368982836604119,
|
|
"num_tokens": 1145466.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.105846500396728,
|
|
"epoch": 0.5219552609776305,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7725,
|
|
"mean_token_accuracy": 0.09914444759488106,
|
|
"num_tokens": 1154225.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.1858007431030275,
|
|
"epoch": 0.5260977630488816,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.0893120139837265,
|
|
"num_tokens": 1163872.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.202767562866211,
|
|
"epoch": 0.5302402651201326,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7379,
|
|
"mean_token_accuracy": 0.09808101058006287,
|
|
"num_tokens": 1172695.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.032454538345337,
|
|
"epoch": 0.5343827671913836,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.6203,
|
|
"mean_token_accuracy": 0.10113452076911926,
|
|
"num_tokens": 1181291.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.1764801979064945,
|
|
"epoch": 0.5385252692626347,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.788,
|
|
"mean_token_accuracy": 0.09804049134254456,
|
|
"num_tokens": 1189927.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.1191915512084964,
|
|
"epoch": 0.5426677713338857,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7234,
|
|
"mean_token_accuracy": 0.10310481786727906,
|
|
"num_tokens": 1197913.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.032878971099853,
|
|
"epoch": 0.5468102734051367,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.6523,
|
|
"mean_token_accuracy": 0.104758819937706,
|
|
"num_tokens": 1207032.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.1075170040130615,
|
|
"epoch": 0.5509527754763878,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7536,
|
|
"mean_token_accuracy": 0.09772960171103477,
|
|
"num_tokens": 1215710.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.11876540184021,
|
|
"epoch": 0.5550952775476388,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.7292,
|
|
"mean_token_accuracy": 0.0940351977944374,
|
|
"num_tokens": 1225619.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.023571300506592,
|
|
"epoch": 0.5592377796188898,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6637,
|
|
"mean_token_accuracy": 0.09711888134479522,
|
|
"num_tokens": 1235000.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.982565116882324,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.664,
|
|
"mean_token_accuracy": 0.098265540599823,
|
|
"num_tokens": 1243594.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.147447633743286,
|
|
"epoch": 0.5675227837613919,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8558,
|
|
"mean_token_accuracy": 0.09704434052109719,
|
|
"num_tokens": 1252279.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.213455104827881,
|
|
"epoch": 0.5716652858326429,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7541,
|
|
"mean_token_accuracy": 0.09795698821544647,
|
|
"num_tokens": 1261314.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.968819999694825,
|
|
"epoch": 0.575807787903894,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6257,
|
|
"mean_token_accuracy": 0.10265169814229011,
|
|
"num_tokens": 1270854.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.012129592895508,
|
|
"epoch": 0.579950289975145,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6127,
|
|
"mean_token_accuracy": 0.10243444740772248,
|
|
"num_tokens": 1279183.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.013506698608398,
|
|
"epoch": 0.584092792046396,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6971,
|
|
"mean_token_accuracy": 0.09987314119935035,
|
|
"num_tokens": 1288347.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.083859825134278,
|
|
"epoch": 0.5882352941176471,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.7262,
|
|
"mean_token_accuracy": 0.09877969771623611,
|
|
"num_tokens": 1297449.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.100916719436645,
|
|
"epoch": 0.5923777961888981,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.6759,
|
|
"mean_token_accuracy": 0.10137342810630798,
|
|
"num_tokens": 1306232.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.998146438598633,
|
|
"epoch": 0.5965202982601492,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6866,
|
|
"mean_token_accuracy": 0.10439095944166184,
|
|
"num_tokens": 1315073.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.042725706100464,
|
|
"epoch": 0.6006628003314002,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7278,
|
|
"mean_token_accuracy": 0.09690923467278481,
|
|
"num_tokens": 1324629.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.049885368347168,
|
|
"epoch": 0.6048053024026512,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.8315,
|
|
"mean_token_accuracy": 0.09335754811763763,
|
|
"num_tokens": 1334152.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.204215431213379,
|
|
"epoch": 0.6089478044739023,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.8026,
|
|
"mean_token_accuracy": 0.0956441655755043,
|
|
"num_tokens": 1343063.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.062389278411866,
|
|
"epoch": 0.6130903065451533,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.722,
|
|
"mean_token_accuracy": 0.09436356574296952,
|
|
"num_tokens": 1352632.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.928182935714721,
|
|
"epoch": 0.6172328086164043,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6537,
|
|
"mean_token_accuracy": 0.09909828305244446,
|
|
"num_tokens": 1361103.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.017486238479615,
|
|
"epoch": 0.6213753106876554,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6427,
|
|
"mean_token_accuracy": 0.10182388573884964,
|
|
"num_tokens": 1370658.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.007332801818848,
|
|
"epoch": 0.6255178127589064,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6679,
|
|
"mean_token_accuracy": 0.09760173633694649,
|
|
"num_tokens": 1379817.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.043180799484253,
|
|
"epoch": 0.6296603148301574,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6189,
|
|
"mean_token_accuracy": 0.10276637375354766,
|
|
"num_tokens": 1389329.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.83769268989563,
|
|
"epoch": 0.6338028169014085,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6539,
|
|
"mean_token_accuracy": 0.10125284790992736,
|
|
"num_tokens": 1399279.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.993398857116699,
|
|
"epoch": 0.6379453189726595,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.6017,
|
|
"mean_token_accuracy": 0.10439575612545013,
|
|
"num_tokens": 1408765.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.039279270172119,
|
|
"epoch": 0.6420878210439105,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.7302,
|
|
"mean_token_accuracy": 0.10342611595988274,
|
|
"num_tokens": 1418615.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.924956750869751,
|
|
"epoch": 0.6462303231151616,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6327,
|
|
"mean_token_accuracy": 0.09766181856393814,
|
|
"num_tokens": 1427359.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.031212186813354,
|
|
"epoch": 0.6503728251864126,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.7909,
|
|
"mean_token_accuracy": 0.10112168565392494,
|
|
"num_tokens": 1435516.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.065572786331177,
|
|
"epoch": 0.6545153272576636,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.7387,
|
|
"mean_token_accuracy": 0.09975770190358162,
|
|
"num_tokens": 1444438.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.0253180980682375,
|
|
"epoch": 0.6586578293289147,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.651,
|
|
"mean_token_accuracy": 0.10093981400132179,
|
|
"num_tokens": 1453740.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.016492176055908,
|
|
"epoch": 0.6628003314001657,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6457,
|
|
"mean_token_accuracy": 0.10038311630487443,
|
|
"num_tokens": 1462083.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.8830955028533936,
|
|
"epoch": 0.6669428334714167,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5764,
|
|
"mean_token_accuracy": 0.10810273066163063,
|
|
"num_tokens": 1471459.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.919638729095459,
|
|
"epoch": 0.6710853355426678,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.5688,
|
|
"mean_token_accuracy": 0.1042623020708561,
|
|
"num_tokens": 1480516.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.843572092056275,
|
|
"epoch": 0.6752278376139188,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4732,
|
|
"mean_token_accuracy": 0.1075613982975483,
|
|
"num_tokens": 1489268.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.020244693756103,
|
|
"epoch": 0.6793703396851698,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.6765,
|
|
"mean_token_accuracy": 0.10138711780309677,
|
|
"num_tokens": 1498819.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.910724115371704,
|
|
"epoch": 0.6835128417564209,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6586,
|
|
"mean_token_accuracy": 0.10495447292923928,
|
|
"num_tokens": 1507852.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.989867115020752,
|
|
"epoch": 0.6876553438276719,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.6807,
|
|
"mean_token_accuracy": 0.10375865250825882,
|
|
"num_tokens": 1517538.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.873890399932861,
|
|
"epoch": 0.6917978458989229,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.563,
|
|
"mean_token_accuracy": 0.1091550700366497,
|
|
"num_tokens": 1525389.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.926404094696045,
|
|
"epoch": 0.695940347970174,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5379,
|
|
"mean_token_accuracy": 0.10326163768768311,
|
|
"num_tokens": 1534117.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.861436605453491,
|
|
"epoch": 0.700082850041425,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.576,
|
|
"mean_token_accuracy": 0.10448229536414147,
|
|
"num_tokens": 1542610.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.804575872421265,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5549,
|
|
"mean_token_accuracy": 0.11098664253950119,
|
|
"num_tokens": 1551771.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.011364126205445,
|
|
"epoch": 0.7083678541839271,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.734,
|
|
"mean_token_accuracy": 0.09743175655603409,
|
|
"num_tokens": 1560231.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.8614123344421385,
|
|
"epoch": 0.7125103562551781,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6704,
|
|
"mean_token_accuracy": 0.10075936913490295,
|
|
"num_tokens": 1569578.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.939766216278076,
|
|
"epoch": 0.7166528583264291,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5619,
|
|
"mean_token_accuracy": 0.10681614726781845,
|
|
"num_tokens": 1579308.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.910393333435058,
|
|
"epoch": 0.7207953603976802,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.7379,
|
|
"mean_token_accuracy": 0.10018669590353965,
|
|
"num_tokens": 1589506.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.811976051330566,
|
|
"epoch": 0.7249378624689312,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5377,
|
|
"mean_token_accuracy": 0.10628587827086448,
|
|
"num_tokens": 1598127.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.834736251831055,
|
|
"epoch": 0.7290803645401823,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.549,
|
|
"mean_token_accuracy": 0.10746697857975959,
|
|
"num_tokens": 1608073.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.892040348052978,
|
|
"epoch": 0.7332228666114333,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.5821,
|
|
"mean_token_accuracy": 0.10727652609348297,
|
|
"num_tokens": 1616272.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.854152059555053,
|
|
"epoch": 0.7373653686826843,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.5853,
|
|
"mean_token_accuracy": 0.10497673824429513,
|
|
"num_tokens": 1625577.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.974736309051513,
|
|
"epoch": 0.7415078707539354,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6705,
|
|
"mean_token_accuracy": 0.09881450086832047,
|
|
"num_tokens": 1634902.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.89474573135376,
|
|
"epoch": 0.7456503728251864,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5895,
|
|
"mean_token_accuracy": 0.1074444942176342,
|
|
"num_tokens": 1644131.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.79931492805481,
|
|
"epoch": 0.7497928748964374,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5106,
|
|
"mean_token_accuracy": 0.11076099425554276,
|
|
"num_tokens": 1652272.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.829667091369629,
|
|
"epoch": 0.7539353769676885,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5935,
|
|
"mean_token_accuracy": 0.10850171595811844,
|
|
"num_tokens": 1660962.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.8230170726776125,
|
|
"epoch": 0.7580778790389395,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5328,
|
|
"mean_token_accuracy": 0.10780560970306396,
|
|
"num_tokens": 1670570.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.78682131767273,
|
|
"epoch": 0.7622203811101905,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5977,
|
|
"mean_token_accuracy": 0.10254386737942696,
|
|
"num_tokens": 1679674.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.933328247070312,
|
|
"epoch": 0.7663628831814416,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5662,
|
|
"mean_token_accuracy": 0.10486595705151558,
|
|
"num_tokens": 1688849.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.694953775405883,
|
|
"epoch": 0.7705053852526926,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.4459,
|
|
"mean_token_accuracy": 0.10817986652255059,
|
|
"num_tokens": 1698565.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.862428140640259,
|
|
"epoch": 0.7746478873239436,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.5348,
|
|
"mean_token_accuracy": 0.11153487414121628,
|
|
"num_tokens": 1708009.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.8652912139892575,
|
|
"epoch": 0.7787903893951947,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.544,
|
|
"mean_token_accuracy": 0.10657664686441422,
|
|
"num_tokens": 1716497.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.7234900951385494,
|
|
"epoch": 0.7829328914664457,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5287,
|
|
"mean_token_accuracy": 0.11238738149404526,
|
|
"num_tokens": 1725609.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.804434871673584,
|
|
"epoch": 0.7870753935376967,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4561,
|
|
"mean_token_accuracy": 0.11403901427984238,
|
|
"num_tokens": 1734165.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.788070774078369,
|
|
"epoch": 0.7912178956089478,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.58,
|
|
"mean_token_accuracy": 0.10270701348781586,
|
|
"num_tokens": 1744856.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.728517675399781,
|
|
"epoch": 0.7953603976801988,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5771,
|
|
"mean_token_accuracy": 0.10236030369997025,
|
|
"num_tokens": 1753386.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.854466104507447,
|
|
"epoch": 0.7995028997514498,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4332,
|
|
"mean_token_accuracy": 0.11357746422290801,
|
|
"num_tokens": 1762521.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.700516653060913,
|
|
"epoch": 0.8036454018227009,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4878,
|
|
"mean_token_accuracy": 0.1087129294872284,
|
|
"num_tokens": 1772403.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.739825487136841,
|
|
"epoch": 0.8077879038939519,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3968,
|
|
"mean_token_accuracy": 0.11042674705386162,
|
|
"num_tokens": 1780924.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.788566207885742,
|
|
"epoch": 0.8119304059652029,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5364,
|
|
"mean_token_accuracy": 0.10567695051431655,
|
|
"num_tokens": 1789992.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.741111326217651,
|
|
"epoch": 0.816072908036454,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.5032,
|
|
"mean_token_accuracy": 0.11023874580860138,
|
|
"num_tokens": 1799513.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.761337041854858,
|
|
"epoch": 0.820215410107705,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3597,
|
|
"mean_token_accuracy": 0.11022111549973487,
|
|
"num_tokens": 1807393.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.774386739730835,
|
|
"epoch": 0.824357912178956,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.6022,
|
|
"mean_token_accuracy": 0.10382014736533166,
|
|
"num_tokens": 1817284.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.767058658599853,
|
|
"epoch": 0.8285004142502072,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.375,
|
|
"mean_token_accuracy": 0.1134818322956562,
|
|
"num_tokens": 1825591.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8285004142502072,
|
|
"eval_entropy": 6.4873610498554415,
|
|
"eval_loss": 6.588028907775879,
|
|
"eval_mean_token_accuracy": 0.11022937250162747,
|
|
"eval_num_tokens": 1825591.0,
|
|
"eval_runtime": 2.1242,
|
|
"eval_samples_per_second": 1763.952,
|
|
"eval_steps_per_second": 220.788,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.784157371520996,
|
|
"epoch": 0.8326429163214581,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999998547681232,
|
|
"loss": 6.5023,
|
|
"mean_token_accuracy": 0.11439347863197327,
|
|
"num_tokens": 1834655.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.565783596038818,
|
|
"epoch": 0.8367854183927091,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999992647639447,
|
|
"loss": 6.3503,
|
|
"mean_token_accuracy": 0.1175505943596363,
|
|
"num_tokens": 1844187.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.671072387695313,
|
|
"epoch": 0.8409279204639603,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999982209116618,
|
|
"loss": 6.4409,
|
|
"mean_token_accuracy": 0.10719897747039794,
|
|
"num_tokens": 1852768.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.695590972900391,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999967232133796,
|
|
"loss": 6.4767,
|
|
"mean_token_accuracy": 0.1183364823460579,
|
|
"num_tokens": 1862095.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.7737802028656,
|
|
"epoch": 0.8492129246064622,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999947716721196,
|
|
"loss": 6.5222,
|
|
"mean_token_accuracy": 0.10554626062512398,
|
|
"num_tokens": 1871239.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.712663888931274,
|
|
"epoch": 0.8533554266777134,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999923662918179,
|
|
"loss": 6.515,
|
|
"mean_token_accuracy": 0.11144773066043853,
|
|
"num_tokens": 1880065.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.769164752960205,
|
|
"epoch": 0.8574979287489644,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999895070773266,
|
|
"loss": 6.5543,
|
|
"mean_token_accuracy": 0.10713088288903236,
|
|
"num_tokens": 1889065.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.731873178482056,
|
|
"epoch": 0.8616404308202155,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999861940344129,
|
|
"loss": 6.5513,
|
|
"mean_token_accuracy": 0.1086876131594181,
|
|
"num_tokens": 1898238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.776325654983521,
|
|
"epoch": 0.8657829328914665,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999824271697597,
|
|
"loss": 6.5038,
|
|
"mean_token_accuracy": 0.1181743711233139,
|
|
"num_tokens": 1906780.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.801963806152344,
|
|
"epoch": 0.8699254349627175,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000499978206490965,
|
|
"loss": 6.5841,
|
|
"mean_token_accuracy": 0.10914266481995583,
|
|
"num_tokens": 1916183.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.635178756713867,
|
|
"epoch": 0.8740679370339686,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999735320065427,
|
|
"loss": 6.4376,
|
|
"mean_token_accuracy": 0.11982565298676491,
|
|
"num_tokens": 1925382.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.6497776985168455,
|
|
"epoch": 0.8782104391052196,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999684037259216,
|
|
"loss": 6.2866,
|
|
"mean_token_accuracy": 0.1231508955359459,
|
|
"num_tokens": 1933639.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.584863090515137,
|
|
"epoch": 0.8823529411764706,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999628216594456,
|
|
"loss": 6.5596,
|
|
"mean_token_accuracy": 0.10988020077347756,
|
|
"num_tokens": 1943044.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.758437156677246,
|
|
"epoch": 0.8864954432477217,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000499956785818375,
|
|
"loss": 6.4002,
|
|
"mean_token_accuracy": 0.11250741928815841,
|
|
"num_tokens": 1951777.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.580097198486328,
|
|
"epoch": 0.8906379453189727,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999502962148843,
|
|
"loss": 6.4733,
|
|
"mean_token_accuracy": 0.1164925940334797,
|
|
"num_tokens": 1961356.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.8584808826446535,
|
|
"epoch": 0.8947804473902237,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499943352862064,
|
|
"loss": 6.5297,
|
|
"mean_token_accuracy": 0.11108277216553689,
|
|
"num_tokens": 1970341.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.662361335754395,
|
|
"epoch": 0.8989229494614748,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999359557739194,
|
|
"loss": 6.3848,
|
|
"mean_token_accuracy": 0.1149318166077137,
|
|
"num_tokens": 1978576.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.662960624694824,
|
|
"epoch": 0.9030654515327258,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999281049653713,
|
|
"loss": 6.4318,
|
|
"mean_token_accuracy": 0.11755185425281525,
|
|
"num_tokens": 1987229.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.685955619812011,
|
|
"epoch": 0.9072079536039768,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999198004522557,
|
|
"loss": 6.3431,
|
|
"mean_token_accuracy": 0.124117149412632,
|
|
"num_tokens": 1996473.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.601357793807983,
|
|
"epoch": 0.9113504556752279,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999110422513237,
|
|
"loss": 6.3803,
|
|
"mean_token_accuracy": 0.11419567912817001,
|
|
"num_tokens": 2005551.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.621028995513916,
|
|
"epoch": 0.9154929577464789,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999018303802414,
|
|
"loss": 6.4823,
|
|
"mean_token_accuracy": 0.11499169245362281,
|
|
"num_tokens": 2015012.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.696759366989136,
|
|
"epoch": 0.9196354598177299,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998921648575904,
|
|
"loss": 6.4869,
|
|
"mean_token_accuracy": 0.11509674936532974,
|
|
"num_tokens": 2024435.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.658921909332276,
|
|
"epoch": 0.923777961888981,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998820457028669,
|
|
"loss": 6.39,
|
|
"mean_token_accuracy": 0.11696809232234955,
|
|
"num_tokens": 2033991.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.638371849060059,
|
|
"epoch": 0.927920463960232,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998714729364824,
|
|
"loss": 6.3985,
|
|
"mean_token_accuracy": 0.12039039358496666,
|
|
"num_tokens": 2043117.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.695077419281006,
|
|
"epoch": 0.932062966031483,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998604465797634,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.11857732310891152,
|
|
"num_tokens": 2052304.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.52111029624939,
|
|
"epoch": 0.9362054681027341,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004998489666549512,
|
|
"loss": 6.3676,
|
|
"mean_token_accuracy": 0.12015296593308449,
|
|
"num_tokens": 2061217.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.554986667633057,
|
|
"epoch": 0.9403479701739851,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004998370331852021,
|
|
"loss": 6.3543,
|
|
"mean_token_accuracy": 0.11446144282817841,
|
|
"num_tokens": 2070796.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.708700561523438,
|
|
"epoch": 0.9444904722452361,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998246461945872,
|
|
"loss": 6.4765,
|
|
"mean_token_accuracy": 0.11509892418980598,
|
|
"num_tokens": 2080202.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.634687614440918,
|
|
"epoch": 0.9486329743164872,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998118057080923,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.11585810706019402,
|
|
"num_tokens": 2089788.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.633106660842896,
|
|
"epoch": 0.9527754763877382,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997985117516183,
|
|
"loss": 6.4111,
|
|
"mean_token_accuracy": 0.11044302135705948,
|
|
"num_tokens": 2098727.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.56185474395752,
|
|
"epoch": 0.9569179784589892,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997847643519803,
|
|
"loss": 6.3168,
|
|
"mean_token_accuracy": 0.11371051520109177,
|
|
"num_tokens": 2107645.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.496697521209716,
|
|
"epoch": 0.9610604805302403,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997705635369085,
|
|
"loss": 6.299,
|
|
"mean_token_accuracy": 0.11548860892653465,
|
|
"num_tokens": 2116613.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.596785545349121,
|
|
"epoch": 0.9652029826014913,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997559093350474,
|
|
"loss": 6.2959,
|
|
"mean_token_accuracy": 0.11894101276993752,
|
|
"num_tokens": 2125585.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.679542684555054,
|
|
"epoch": 0.9693454846727423,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997408017759562,
|
|
"loss": 6.4112,
|
|
"mean_token_accuracy": 0.11748032122850419,
|
|
"num_tokens": 2134795.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.610104560852051,
|
|
"epoch": 0.9734879867439934,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997252408901084,
|
|
"loss": 6.4127,
|
|
"mean_token_accuracy": 0.11549609526991844,
|
|
"num_tokens": 2143699.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.6923322677612305,
|
|
"epoch": 0.9776304888152444,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004997092267088922,
|
|
"loss": 6.359,
|
|
"mean_token_accuracy": 0.11639811396598816,
|
|
"num_tokens": 2152187.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.513570928573609,
|
|
"epoch": 0.9817729908864954,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996927592646097,
|
|
"loss": 6.4218,
|
|
"mean_token_accuracy": 0.11397558972239494,
|
|
"num_tokens": 2160941.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.680602073669434,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996758385904776,
|
|
"loss": 6.4021,
|
|
"mean_token_accuracy": 0.11822220534086228,
|
|
"num_tokens": 2168946.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.4418701171875,
|
|
"epoch": 0.9900579950289975,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996584647206269,
|
|
"loss": 6.307,
|
|
"mean_token_accuracy": 0.11671008840203285,
|
|
"num_tokens": 2177662.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.607698106765747,
|
|
"epoch": 0.9942004971002486,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996406376901026,
|
|
"loss": 6.4069,
|
|
"mean_token_accuracy": 0.12819217070937156,
|
|
"num_tokens": 2187461.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.504300785064697,
|
|
"epoch": 0.9983429991714996,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996223575348635,
|
|
"loss": 6.2895,
|
|
"mean_token_accuracy": 0.12742482796311377,
|
|
"num_tokens": 2196878.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.635723876953125,
|
|
"epoch": 1.0024855012427507,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996036242917828,
|
|
"loss": 6.3007,
|
|
"mean_token_accuracy": 0.11356158629059791,
|
|
"num_tokens": 2206178.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.521615219116211,
|
|
"epoch": 1.0066280033140016,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995844379986476,
|
|
"loss": 6.1736,
|
|
"mean_token_accuracy": 0.1223662681877613,
|
|
"num_tokens": 2215330.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.593561172485352,
|
|
"epoch": 1.0107705053852527,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995647986941587,
|
|
"loss": 6.1712,
|
|
"mean_token_accuracy": 0.12720630392432214,
|
|
"num_tokens": 2224146.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.5283712387084964,
|
|
"epoch": 1.0149130074565038,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995447064179308,
|
|
"loss": 6.1551,
|
|
"mean_token_accuracy": 0.13192698806524278,
|
|
"num_tokens": 2233646.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.364728593826294,
|
|
"epoch": 1.0190555095277547,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000499524161210492,
|
|
"loss": 6.0263,
|
|
"mean_token_accuracy": 0.1275729440152645,
|
|
"num_tokens": 2243663.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.589374303817749,
|
|
"epoch": 1.0231980115990058,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995031631132844,
|
|
"loss": 6.2014,
|
|
"mean_token_accuracy": 0.13359731659293175,
|
|
"num_tokens": 2252564.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.493811464309692,
|
|
"epoch": 1.027340513670257,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004994817121686633,
|
|
"loss": 6.2099,
|
|
"mean_token_accuracy": 0.12453857064247131,
|
|
"num_tokens": 2262504.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.53023624420166,
|
|
"epoch": 1.0314830157415078,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994598084198979,
|
|
"loss": 6.0827,
|
|
"mean_token_accuracy": 0.13783227279782295,
|
|
"num_tokens": 2270553.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.445224046707153,
|
|
"epoch": 1.035625517812759,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004994374519111701,
|
|
"loss": 6.1134,
|
|
"mean_token_accuracy": 0.12587740272283554,
|
|
"num_tokens": 2279319.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.521056652069092,
|
|
"epoch": 1.03976801988401,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994146426875757,
|
|
"loss": 6.2428,
|
|
"mean_token_accuracy": 0.1287379577755928,
|
|
"num_tokens": 2288749.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.496748733520508,
|
|
"epoch": 1.043910521955261,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004993913807951232,
|
|
"loss": 6.1241,
|
|
"mean_token_accuracy": 0.12613506391644477,
|
|
"num_tokens": 2298272.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.513681507110595,
|
|
"epoch": 1.048053024026512,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993676662807345,
|
|
"loss": 6.1752,
|
|
"mean_token_accuracy": 0.13274361863732337,
|
|
"num_tokens": 2307230.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.4921036720275875,
|
|
"epoch": 1.0521955260977631,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004993434991922443,
|
|
"loss": 6.251,
|
|
"mean_token_accuracy": 0.12720409855246545,
|
|
"num_tokens": 2316306.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.523912334442139,
|
|
"epoch": 1.056338028169014,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004993188795784002,
|
|
"loss": 6.1805,
|
|
"mean_token_accuracy": 0.1223751187324524,
|
|
"num_tokens": 2325672.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.428217792510987,
|
|
"epoch": 1.0604805302402651,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004992938074888628,
|
|
"loss": 6.1681,
|
|
"mean_token_accuracy": 0.1255595475435257,
|
|
"num_tokens": 2335251.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.473890256881714,
|
|
"epoch": 1.0646230323115162,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004992682829742052,
|
|
"loss": 6.125,
|
|
"mean_token_accuracy": 0.12794472500681878,
|
|
"num_tokens": 2344054.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.3825085163116455,
|
|
"epoch": 1.0687655343827671,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004992423060859132,
|
|
"loss": 6.1274,
|
|
"mean_token_accuracy": 0.12978773862123488,
|
|
"num_tokens": 2353493.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.579702091217041,
|
|
"epoch": 1.0729080364540182,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992158768763849,
|
|
"loss": 6.2352,
|
|
"mean_token_accuracy": 0.12934877946972848,
|
|
"num_tokens": 2362687.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.51726803779602,
|
|
"epoch": 1.0770505385252693,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991889953989308,
|
|
"loss": 6.2101,
|
|
"mean_token_accuracy": 0.13052549585700035,
|
|
"num_tokens": 2372325.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.475685262680054,
|
|
"epoch": 1.0811930405965202,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004991616617077741,
|
|
"loss": 6.2296,
|
|
"mean_token_accuracy": 0.12784886062145234,
|
|
"num_tokens": 2381987.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.639652824401855,
|
|
"epoch": 1.0853355426677713,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004991338758580496,
|
|
"loss": 6.1472,
|
|
"mean_token_accuracy": 0.1295457474887371,
|
|
"num_tokens": 2390901.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.358977031707764,
|
|
"epoch": 1.0894780447390224,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991056379058045,
|
|
"loss": 6.1883,
|
|
"mean_token_accuracy": 0.12295814007520675,
|
|
"num_tokens": 2400756.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.40067458152771,
|
|
"epoch": 1.0936205468102733,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004990769479079976,
|
|
"loss": 5.9995,
|
|
"mean_token_accuracy": 0.13081807941198348,
|
|
"num_tokens": 2409526.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.402806425094605,
|
|
"epoch": 1.0977630488815244,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004990478059225003,
|
|
"loss": 6.1774,
|
|
"mean_token_accuracy": 0.12853156328201293,
|
|
"num_tokens": 2419583.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.509050941467285,
|
|
"epoch": 1.1019055509527755,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004990182120080945,
|
|
"loss": 6.142,
|
|
"mean_token_accuracy": 0.13302397802472116,
|
|
"num_tokens": 2429439.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.362121295928955,
|
|
"epoch": 1.1060480530240264,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004989881662244748,
|
|
"loss": 6.128,
|
|
"mean_token_accuracy": 0.12412297874689102,
|
|
"num_tokens": 2438467.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.464797878265381,
|
|
"epoch": 1.1101905550952775,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989576686322468,
|
|
"loss": 6.0819,
|
|
"mean_token_accuracy": 0.13460188284516333,
|
|
"num_tokens": 2446619.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.365456485748291,
|
|
"epoch": 1.1143330571665286,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004989267192929272,
|
|
"loss": 6.1521,
|
|
"mean_token_accuracy": 0.12620313614606857,
|
|
"num_tokens": 2454998.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.402712297439575,
|
|
"epoch": 1.1184755592377795,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004988953182689445,
|
|
"loss": 6.1088,
|
|
"mean_token_accuracy": 0.13071864768862723,
|
|
"num_tokens": 2463497.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.442966413497925,
|
|
"epoch": 1.1226180613090306,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988634656236378,
|
|
"loss": 6.0379,
|
|
"mean_token_accuracy": 0.1323768101632595,
|
|
"num_tokens": 2471763.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.382192516326905,
|
|
"epoch": 1.1267605633802817,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004988311614212576,
|
|
"loss": 6.0982,
|
|
"mean_token_accuracy": 0.13209657073020936,
|
|
"num_tokens": 2479949.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.453052568435669,
|
|
"epoch": 1.1309030654515326,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987984057269648,
|
|
"loss": 6.1482,
|
|
"mean_token_accuracy": 0.1316144049167633,
|
|
"num_tokens": 2489097.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.231172561645508,
|
|
"epoch": 1.1350455675227837,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004987651986068311,
|
|
"loss": 5.9739,
|
|
"mean_token_accuracy": 0.13615623116493225,
|
|
"num_tokens": 2497625.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.410304975509644,
|
|
"epoch": 1.1391880695940348,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004987315401278394,
|
|
"loss": 6.2206,
|
|
"mean_token_accuracy": 0.12649596706032754,
|
|
"num_tokens": 2507107.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.553795623779297,
|
|
"epoch": 1.143330571665286,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004986974303578819,
|
|
"loss": 6.2198,
|
|
"mean_token_accuracy": 0.12320356070995331,
|
|
"num_tokens": 2516629.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.4057948112487795,
|
|
"epoch": 1.1474730737365368,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004986628693657621,
|
|
"loss": 6.1043,
|
|
"mean_token_accuracy": 0.13110435828566552,
|
|
"num_tokens": 2525924.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.529660415649414,
|
|
"epoch": 1.151615575807788,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004986278572211933,
|
|
"loss": 6.2581,
|
|
"mean_token_accuracy": 0.12878933995962144,
|
|
"num_tokens": 2535586.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.49504132270813,
|
|
"epoch": 1.1557580778790388,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004985923939947987,
|
|
"loss": 6.3228,
|
|
"mean_token_accuracy": 0.12478614002466201,
|
|
"num_tokens": 2546685.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.359566736221313,
|
|
"epoch": 1.15990057995029,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004985564797581115,
|
|
"loss": 6.0478,
|
|
"mean_token_accuracy": 0.13317504152655602,
|
|
"num_tokens": 2556873.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.434709787368774,
|
|
"epoch": 1.164043082021541,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004985201145835747,
|
|
"loss": 6.0913,
|
|
"mean_token_accuracy": 0.13183084428310393,
|
|
"num_tokens": 2565222.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.3213330745697025,
|
|
"epoch": 1.1681855840927922,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000498483298544541,
|
|
"loss": 6.0768,
|
|
"mean_token_accuracy": 0.12748995423316956,
|
|
"num_tokens": 2575467.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.398600482940674,
|
|
"epoch": 1.172328086164043,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000498446031715272,
|
|
"loss": 6.0972,
|
|
"mean_token_accuracy": 0.13319127038121223,
|
|
"num_tokens": 2583963.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.394404697418213,
|
|
"epoch": 1.1764705882352942,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004984083141709391,
|
|
"loss": 6.1374,
|
|
"mean_token_accuracy": 0.133284143358469,
|
|
"num_tokens": 2593350.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.300214338302612,
|
|
"epoch": 1.180613090306545,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000498370145987623,
|
|
"loss": 6.0387,
|
|
"mean_token_accuracy": 0.13577536046504973,
|
|
"num_tokens": 2602191.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.413609552383423,
|
|
"epoch": 1.1847555923777962,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004983315272423127,
|
|
"loss": 6.175,
|
|
"mean_token_accuracy": 0.1315205328166485,
|
|
"num_tokens": 2613590.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.330083179473877,
|
|
"epoch": 1.1888980944490473,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004982924580129066,
|
|
"loss": 5.9919,
|
|
"mean_token_accuracy": 0.13426781222224235,
|
|
"num_tokens": 2622072.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.408995532989502,
|
|
"epoch": 1.1930405965202984,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004982529383782116,
|
|
"loss": 6.1871,
|
|
"mean_token_accuracy": 0.1286151461303234,
|
|
"num_tokens": 2630904.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.3900384426116945,
|
|
"epoch": 1.1971830985915493,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004982129684179432,
|
|
"loss": 6.1579,
|
|
"mean_token_accuracy": 0.12926432937383653,
|
|
"num_tokens": 2639925.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.350859832763672,
|
|
"epoch": 1.2013256006628004,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004981725482127251,
|
|
"loss": 6.0588,
|
|
"mean_token_accuracy": 0.1335553601384163,
|
|
"num_tokens": 2649397.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.4356773853302,
|
|
"epoch": 1.2054681027340513,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004981316778440892,
|
|
"loss": 6.1985,
|
|
"mean_token_accuracy": 0.13050690963864325,
|
|
"num_tokens": 2658877.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.354866790771484,
|
|
"epoch": 1.2096106048053024,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004980903573944756,
|
|
"loss": 6.0988,
|
|
"mean_token_accuracy": 0.1269911654293537,
|
|
"num_tokens": 2668518.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.373867654800415,
|
|
"epoch": 1.2137531068765535,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004980485869472322,
|
|
"loss": 6.1852,
|
|
"mean_token_accuracy": 0.12791354209184647,
|
|
"num_tokens": 2677459.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.468676900863647,
|
|
"epoch": 1.2178956089478046,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004980063665866143,
|
|
"loss": 6.1557,
|
|
"mean_token_accuracy": 0.12465783432126046,
|
|
"num_tokens": 2686668.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.215289306640625,
|
|
"epoch": 1.2220381110190555,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004979636963977853,
|
|
"loss": 5.9811,
|
|
"mean_token_accuracy": 0.1315767101943493,
|
|
"num_tokens": 2695545.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.380659484863282,
|
|
"epoch": 1.2261806130903066,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004979205764668156,
|
|
"loss": 6.1323,
|
|
"mean_token_accuracy": 0.1333031803369522,
|
|
"num_tokens": 2705514.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.364181184768677,
|
|
"epoch": 1.2303231151615575,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978770068806825,
|
|
"loss": 6.0352,
|
|
"mean_token_accuracy": 0.13151699304580688,
|
|
"num_tokens": 2714373.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.372594594955444,
|
|
"epoch": 1.2344656172328086,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000497832987727271,
|
|
"loss": 6.1622,
|
|
"mean_token_accuracy": 0.1286887414753437,
|
|
"num_tokens": 2723887.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.338165140151977,
|
|
"epoch": 1.2386081193040597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004977885190953724,
|
|
"loss": 6.0249,
|
|
"mean_token_accuracy": 0.13291778564453124,
|
|
"num_tokens": 2733591.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.403629159927368,
|
|
"epoch": 1.2427506213753108,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004977436010746846,
|
|
"loss": 6.1543,
|
|
"mean_token_accuracy": 0.1351762518286705,
|
|
"num_tokens": 2741638.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2427506213753108,
|
|
"eval_entropy": 6.184635734761447,
|
|
"eval_loss": 6.306137561798096,
|
|
"eval_mean_token_accuracy": 0.12917301741871498,
|
|
"eval_num_tokens": 2741638.0,
|
|
"eval_runtime": 2.1134,
|
|
"eval_samples_per_second": 1772.967,
|
|
"eval_steps_per_second": 221.917,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.382915115356445,
|
|
"epoch": 1.2468931234465617,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004976982337558124,
|
|
"loss": 6.0264,
|
|
"mean_token_accuracy": 0.1349784404039383,
|
|
"num_tokens": 2749711.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.304248571395874,
|
|
"epoch": 1.2510356255178128,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004976524172302665,
|
|
"loss": 6.1943,
|
|
"mean_token_accuracy": 0.12665166333317757,
|
|
"num_tokens": 2758121.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.454039573669434,
|
|
"epoch": 1.2551781275890637,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004976061515904641,
|
|
"loss": 6.1996,
|
|
"mean_token_accuracy": 0.12957023978233337,
|
|
"num_tokens": 2766764.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.427061653137207,
|
|
"epoch": 1.2593206296603148,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004975594369297277,
|
|
"loss": 6.1446,
|
|
"mean_token_accuracy": 0.1343374729156494,
|
|
"num_tokens": 2775561.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.3574135303497314,
|
|
"epoch": 1.2634631317315659,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004975122733422861,
|
|
"loss": 6.0999,
|
|
"mean_token_accuracy": 0.12937545776367188,
|
|
"num_tokens": 2784691.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.3339677333831785,
|
|
"epoch": 1.267605633802817,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004974646609232732,
|
|
"loss": 6.07,
|
|
"mean_token_accuracy": 0.13353220298886298,
|
|
"num_tokens": 2793893.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.349122190475464,
|
|
"epoch": 1.2717481358740679,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004974165997687288,
|
|
"loss": 6.1495,
|
|
"mean_token_accuracy": 0.1352272853255272,
|
|
"num_tokens": 2804479.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.413073539733887,
|
|
"epoch": 1.275890637945319,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004973680899755972,
|
|
"loss": 6.1781,
|
|
"mean_token_accuracy": 0.1326884187757969,
|
|
"num_tokens": 2813966.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.329093027114868,
|
|
"epoch": 1.2800331400165699,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000497319131641728,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.13602066040039062,
|
|
"num_tokens": 2822328.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.286653518676758,
|
|
"epoch": 1.284175642087821,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004972697248658755,
|
|
"loss": 5.9838,
|
|
"mean_token_accuracy": 0.134714674949646,
|
|
"num_tokens": 2829845.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.249941444396972,
|
|
"epoch": 1.288318144159072,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004972198697476987,
|
|
"loss": 5.9553,
|
|
"mean_token_accuracy": 0.1375754125416279,
|
|
"num_tokens": 2839040.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.230720090866089,
|
|
"epoch": 1.2924606462303232,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004971695663877607,
|
|
"loss": 6.0501,
|
|
"mean_token_accuracy": 0.1403980202972889,
|
|
"num_tokens": 2848574.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.382760429382325,
|
|
"epoch": 1.296603148301574,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004971188148875289,
|
|
"loss": 6.0074,
|
|
"mean_token_accuracy": 0.1397629365324974,
|
|
"num_tokens": 2856812.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.279427480697632,
|
|
"epoch": 1.3007456503728252,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004970676153493748,
|
|
"loss": 6.1114,
|
|
"mean_token_accuracy": 0.13487065508961676,
|
|
"num_tokens": 2866001.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.338931465148926,
|
|
"epoch": 1.3048881524440763,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004970159678765731,
|
|
"loss": 6.1069,
|
|
"mean_token_accuracy": 0.12865526527166365,
|
|
"num_tokens": 2875181.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.307818031311035,
|
|
"epoch": 1.3090306545153272,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004969638725733028,
|
|
"loss": 6.0146,
|
|
"mean_token_accuracy": 0.1392890766263008,
|
|
"num_tokens": 2885103.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.189027595520019,
|
|
"epoch": 1.3131731565865783,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004969113295446455,
|
|
"loss": 5.9759,
|
|
"mean_token_accuracy": 0.13826468139886855,
|
|
"num_tokens": 2894055.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.370010614395142,
|
|
"epoch": 1.3173156586578294,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004968583388965863,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.13388948068022727,
|
|
"num_tokens": 2903391.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.315566825866699,
|
|
"epoch": 1.3214581607290803,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004968049007360132,
|
|
"loss": 6.05,
|
|
"mean_token_accuracy": 0.1396598182618618,
|
|
"num_tokens": 2911760.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.269933223724365,
|
|
"epoch": 1.3256006628003314,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004967510151707169,
|
|
"loss": 6.0144,
|
|
"mean_token_accuracy": 0.1340911418199539,
|
|
"num_tokens": 2920589.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.371693134307861,
|
|
"epoch": 1.3297431648715825,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966966823093903,
|
|
"loss": 6.1224,
|
|
"mean_token_accuracy": 0.13662606105208397,
|
|
"num_tokens": 2929635.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.30497727394104,
|
|
"epoch": 1.3338856669428334,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004966419022616287,
|
|
"loss": 6.0162,
|
|
"mean_token_accuracy": 0.13851367086172103,
|
|
"num_tokens": 2938778.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.233600711822509,
|
|
"epoch": 1.3380281690140845,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004965866751379296,
|
|
"loss": 6.1197,
|
|
"mean_token_accuracy": 0.13244155272841454,
|
|
"num_tokens": 2947836.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.397352600097657,
|
|
"epoch": 1.3421706710853356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004965310010496922,
|
|
"loss": 6.0784,
|
|
"mean_token_accuracy": 0.12939544543623924,
|
|
"num_tokens": 2957028.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.264967441558838,
|
|
"epoch": 1.3463131731565865,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004964748801092168,
|
|
"loss": 6.0393,
|
|
"mean_token_accuracy": 0.13198372200131417,
|
|
"num_tokens": 2966088.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.334882211685181,
|
|
"epoch": 1.3504556752278376,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004964183124297059,
|
|
"loss": 6.1223,
|
|
"mean_token_accuracy": 0.13003934547305107,
|
|
"num_tokens": 2975271.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.306211709976196,
|
|
"epoch": 1.3545981772990887,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004963612981252625,
|
|
"loss": 6.1261,
|
|
"mean_token_accuracy": 0.13269460275769235,
|
|
"num_tokens": 2983799.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.32020058631897,
|
|
"epoch": 1.3587406793703396,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004963038373108908,
|
|
"loss": 6.0904,
|
|
"mean_token_accuracy": 0.13578941598534583,
|
|
"num_tokens": 2992675.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.32526535987854,
|
|
"epoch": 1.3628831814415907,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004962459301024954,
|
|
"loss": 6.153,
|
|
"mean_token_accuracy": 0.1340462252497673,
|
|
"num_tokens": 3002483.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.328367424011231,
|
|
"epoch": 1.3670256835128418,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004961875766168816,
|
|
"loss": 6.0482,
|
|
"mean_token_accuracy": 0.13762085065245627,
|
|
"num_tokens": 3010611.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.296908664703369,
|
|
"epoch": 1.371168185584093,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004961287769717547,
|
|
"loss": 6.0196,
|
|
"mean_token_accuracy": 0.13391901478171347,
|
|
"num_tokens": 3019928.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.1809553623199465,
|
|
"epoch": 1.3753106876553438,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004960695312857199,
|
|
"loss": 6.0149,
|
|
"mean_token_accuracy": 0.1327693857252598,
|
|
"num_tokens": 3029003.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.265834045410156,
|
|
"epoch": 1.379453189726595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004960098396782824,
|
|
"loss": 6.0269,
|
|
"mean_token_accuracy": 0.14121449887752532,
|
|
"num_tokens": 3038049.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.215864753723144,
|
|
"epoch": 1.3835956917978458,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004959497022698466,
|
|
"loss": 6.0243,
|
|
"mean_token_accuracy": 0.1427525833249092,
|
|
"num_tokens": 3046674.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.269976186752319,
|
|
"epoch": 1.387738193869097,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004958891191817163,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.13031871765851974,
|
|
"num_tokens": 3056459.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.325995588302613,
|
|
"epoch": 1.391880695940348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000495828090536094,
|
|
"loss": 6.0442,
|
|
"mean_token_accuracy": 0.13008645325899124,
|
|
"num_tokens": 3066294.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.270032787322998,
|
|
"epoch": 1.3960231980115991,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004957666164560815,
|
|
"loss": 6.0061,
|
|
"mean_token_accuracy": 0.13253876194357872,
|
|
"num_tokens": 3075420.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.301951360702515,
|
|
"epoch": 1.40016570008285,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004957046970656784,
|
|
"loss": 6.1132,
|
|
"mean_token_accuracy": 0.13512104675173758,
|
|
"num_tokens": 3084373.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.254469299316407,
|
|
"epoch": 1.4043082021541011,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004956423324897831,
|
|
"loss": 6.086,
|
|
"mean_token_accuracy": 0.13642527535557747,
|
|
"num_tokens": 3094212.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.278657054901123,
|
|
"epoch": 1.408450704225352,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955795228541917,
|
|
"loss": 6.0177,
|
|
"mean_token_accuracy": 0.14363805800676346,
|
|
"num_tokens": 3102769.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.314499855041504,
|
|
"epoch": 1.4125932062966031,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000495516268285598,
|
|
"loss": 6.0502,
|
|
"mean_token_accuracy": 0.13782473877072335,
|
|
"num_tokens": 3111465.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.214475965499878,
|
|
"epoch": 1.4167357083678542,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004954525689115935,
|
|
"loss": 6.0005,
|
|
"mean_token_accuracy": 0.13500823974609374,
|
|
"num_tokens": 3121106.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.27736988067627,
|
|
"epoch": 1.4208782104391053,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004953884248606669,
|
|
"loss": 6.002,
|
|
"mean_token_accuracy": 0.14771596789360047,
|
|
"num_tokens": 3130662.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.197289609909058,
|
|
"epoch": 1.4250207125103562,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004953238362622034,
|
|
"loss": 6.046,
|
|
"mean_token_accuracy": 0.13686998263001443,
|
|
"num_tokens": 3139538.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.3258789539337155,
|
|
"epoch": 1.4291632145816073,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004952588032464855,
|
|
"loss": 6.0886,
|
|
"mean_token_accuracy": 0.13061245977878572,
|
|
"num_tokens": 3148552.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.150630235671997,
|
|
"epoch": 1.4333057166528582,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004951933259446919,
|
|
"loss": 5.9884,
|
|
"mean_token_accuracy": 0.14620922207832338,
|
|
"num_tokens": 3156549.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.3774982452392575,
|
|
"epoch": 1.4374482187241093,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004951274044888975,
|
|
"loss": 6.1652,
|
|
"mean_token_accuracy": 0.12070072740316391,
|
|
"num_tokens": 3166821.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.185049962997437,
|
|
"epoch": 1.4415907207953604,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004950610390120731,
|
|
"loss": 6.0523,
|
|
"mean_token_accuracy": 0.13229329884052277,
|
|
"num_tokens": 3174962.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.27318377494812,
|
|
"epoch": 1.4457332228666115,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000494994229648085,
|
|
"loss": 5.9639,
|
|
"mean_token_accuracy": 0.1355227366089821,
|
|
"num_tokens": 3184027.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.173826360702515,
|
|
"epoch": 1.4498757249378624,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004949269765316948,
|
|
"loss": 6.1056,
|
|
"mean_token_accuracy": 0.12917206510901452,
|
|
"num_tokens": 3193186.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.373425626754761,
|
|
"epoch": 1.4540182270091135,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 0.0004948592797985599,
|
|
"loss": 6.0952,
|
|
"mean_token_accuracy": 0.12932660430669785,
|
|
"num_tokens": 3202676.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.382624101638794,
|
|
"epoch": 1.4581607290803644,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004947911395852317,
|
|
"loss": 6.1966,
|
|
"mean_token_accuracy": 0.1361641675233841,
|
|
"num_tokens": 3211950.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.380419588088989,
|
|
"epoch": 1.4623032311516155,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004947225560291565,
|
|
"loss": 6.1404,
|
|
"mean_token_accuracy": 0.12988072633743286,
|
|
"num_tokens": 3221135.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.153287029266357,
|
|
"epoch": 1.4664457332228666,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004946535292686746,
|
|
"loss": 5.9467,
|
|
"mean_token_accuracy": 0.14215940982103348,
|
|
"num_tokens": 3230174.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.2029670715332035,
|
|
"epoch": 1.4705882352941178,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004945840594430209,
|
|
"loss": 5.9463,
|
|
"mean_token_accuracy": 0.13841561377048492,
|
|
"num_tokens": 3238898.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.249447679519653,
|
|
"epoch": 1.4747307373653686,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004945141466923235,
|
|
"loss": 5.9835,
|
|
"mean_token_accuracy": 0.14440677165985108,
|
|
"num_tokens": 3247550.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.36776909828186,
|
|
"epoch": 1.4788732394366197,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004944437911576041,
|
|
"loss": 6.1334,
|
|
"mean_token_accuracy": 0.12904864847660064,
|
|
"num_tokens": 3256810.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.134829092025757,
|
|
"epoch": 1.4830157415078706,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004943729929807772,
|
|
"loss": 5.9163,
|
|
"mean_token_accuracy": 0.13975635766983033,
|
|
"num_tokens": 3265886.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.258411979675293,
|
|
"epoch": 1.4871582435791217,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004943017523046507,
|
|
"loss": 5.9896,
|
|
"mean_token_accuracy": 0.14377044066786765,
|
|
"num_tokens": 3274522.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.268470525741577,
|
|
"epoch": 1.4913007456503728,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000494230069272925,
|
|
"loss": 6.0988,
|
|
"mean_token_accuracy": 0.13920872285962105,
|
|
"num_tokens": 3283388.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.251413249969483,
|
|
"epoch": 1.495443247721624,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004941579440301923,
|
|
"loss": 5.9634,
|
|
"mean_token_accuracy": 0.13360913395881652,
|
|
"num_tokens": 3291887.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.1874171733856205,
|
|
"epoch": 1.4995857497928748,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004940853767219372,
|
|
"loss": 6.0311,
|
|
"mean_token_accuracy": 0.13737098798155783,
|
|
"num_tokens": 3300674.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.158373260498047,
|
|
"epoch": 1.503728251864126,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004940123674945359,
|
|
"loss": 5.8806,
|
|
"mean_token_accuracy": 0.14576042145490647,
|
|
"num_tokens": 3309745.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.272955846786499,
|
|
"epoch": 1.5078707539353768,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004939389164952561,
|
|
"loss": 6.0288,
|
|
"mean_token_accuracy": 0.1426420822739601,
|
|
"num_tokens": 3318888.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.192076063156128,
|
|
"epoch": 1.512013256006628,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004938650238722564,
|
|
"loss": 6.0002,
|
|
"mean_token_accuracy": 0.13870967626571656,
|
|
"num_tokens": 3328274.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.2451879501342775,
|
|
"epoch": 1.516155758077879,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004937906897745861,
|
|
"loss": 6.0114,
|
|
"mean_token_accuracy": 0.1397687092423439,
|
|
"num_tokens": 3337579.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.244029188156128,
|
|
"epoch": 1.5202982601491302,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004937159143521856,
|
|
"loss": 5.9948,
|
|
"mean_token_accuracy": 0.13942682594060898,
|
|
"num_tokens": 3346262.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.220328426361084,
|
|
"epoch": 1.5244407622203813,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004936406977558847,
|
|
"loss": 6.063,
|
|
"mean_token_accuracy": 0.13193173110485076,
|
|
"num_tokens": 3354833.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.2797801971435545,
|
|
"epoch": 1.5285832642916322,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004935650401374039,
|
|
"loss": 6.1433,
|
|
"mean_token_accuracy": 0.14123231619596482,
|
|
"num_tokens": 3364314.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.2219626903533936,
|
|
"epoch": 1.532725766362883,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004934889416493524,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.13829006403684616,
|
|
"num_tokens": 3373648.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.146945238113403,
|
|
"epoch": 1.5368682684341342,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004934124024452295,
|
|
"loss": 5.8563,
|
|
"mean_token_accuracy": 0.1476227268576622,
|
|
"num_tokens": 3382919.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.159171915054321,
|
|
"epoch": 1.5410107705053853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004933354226794232,
|
|
"loss": 5.9906,
|
|
"mean_token_accuracy": 0.1392545849084854,
|
|
"num_tokens": 3391814.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.1792021751403805,
|
|
"epoch": 1.5451532725766364,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004932580025072098,
|
|
"loss": 6.0321,
|
|
"mean_token_accuracy": 0.13409882485866548,
|
|
"num_tokens": 3400392.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.255561637878418,
|
|
"epoch": 1.5492957746478875,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004931801420847545,
|
|
"loss": 6.0792,
|
|
"mean_token_accuracy": 0.12993714809417725,
|
|
"num_tokens": 3410430.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.133722019195557,
|
|
"epoch": 1.5534382767191384,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004931018415691102,
|
|
"loss": 5.9171,
|
|
"mean_token_accuracy": 0.138246750831604,
|
|
"num_tokens": 3420654.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.223542308807373,
|
|
"epoch": 1.5575807787903893,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004930231011182176,
|
|
"loss": 6.0585,
|
|
"mean_token_accuracy": 0.13670557215809823,
|
|
"num_tokens": 3430514.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.254469013214111,
|
|
"epoch": 1.5617232808616404,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004929439208909048,
|
|
"loss": 5.9879,
|
|
"mean_token_accuracy": 0.13579081147909164,
|
|
"num_tokens": 3439551.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.171637487411499,
|
|
"epoch": 1.5658657829328915,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004928643010468869,
|
|
"loss": 5.9557,
|
|
"mean_token_accuracy": 0.13981301635503768,
|
|
"num_tokens": 3448160.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.192162418365479,
|
|
"epoch": 1.5700082850041426,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004927842417467659,
|
|
"loss": 5.9886,
|
|
"mean_token_accuracy": 0.1458936534821987,
|
|
"num_tokens": 3457296.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.243723821640015,
|
|
"epoch": 1.5741507870753937,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004927037431520302,
|
|
"loss": 5.951,
|
|
"mean_token_accuracy": 0.1482045531272888,
|
|
"num_tokens": 3466108.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.144083166122437,
|
|
"epoch": 1.5782932891466446,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004926228054250541,
|
|
"loss": 6.0433,
|
|
"mean_token_accuracy": 0.14178147464990615,
|
|
"num_tokens": 3475336.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.324700212478637,
|
|
"epoch": 1.5824357912178955,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004925414287290979,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.1372267983853817,
|
|
"num_tokens": 3483963.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.0791584014892575,
|
|
"epoch": 1.5865782932891466,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004924596132283075,
|
|
"loss": 5.9327,
|
|
"mean_token_accuracy": 0.14885903000831605,
|
|
"num_tokens": 3492947.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.174278497695923,
|
|
"epoch": 1.5907207953603977,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004923773590877135,
|
|
"loss": 6.0027,
|
|
"mean_token_accuracy": 0.13670362904667854,
|
|
"num_tokens": 3501877.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.269541835784912,
|
|
"epoch": 1.5948632974316488,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004922946664732314,
|
|
"loss": 6.0663,
|
|
"mean_token_accuracy": 0.1364058144390583,
|
|
"num_tokens": 3510610.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.2451316833496096,
|
|
"epoch": 1.5990057995029,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004922115355516615,
|
|
"loss": 6.0414,
|
|
"mean_token_accuracy": 0.137686824798584,
|
|
"num_tokens": 3519829.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.232548189163208,
|
|
"epoch": 1.6031483015741508,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004921279664906877,
|
|
"loss": 6.0075,
|
|
"mean_token_accuracy": 0.13653246834874153,
|
|
"num_tokens": 3529408.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.147272729873658,
|
|
"epoch": 1.6072908036454017,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004920439594588781,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.1483641341328621,
|
|
"num_tokens": 3538068.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.2408874988555905,
|
|
"epoch": 1.6114333057166528,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000491959514625684,
|
|
"loss": 6.0614,
|
|
"mean_token_accuracy": 0.13784216567873955,
|
|
"num_tokens": 3547302.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.210910749435425,
|
|
"epoch": 1.6155758077879039,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00049187463216144,
|
|
"loss": 5.9494,
|
|
"mean_token_accuracy": 0.14787167757749559,
|
|
"num_tokens": 3556087.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.240442419052124,
|
|
"epoch": 1.619718309859155,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000491789312237363,
|
|
"loss": 6.0244,
|
|
"mean_token_accuracy": 0.14060693755745887,
|
|
"num_tokens": 3566235.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.1727869510650635,
|
|
"epoch": 1.623860811930406,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004917035550255528,
|
|
"loss": 5.9717,
|
|
"mean_token_accuracy": 0.14355085045099258,
|
|
"num_tokens": 3574762.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.142957496643066,
|
|
"epoch": 1.628003314001657,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491617360698991,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.14090993776917457,
|
|
"num_tokens": 3583359.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.10711522102356,
|
|
"epoch": 1.6321458160729079,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004915307294315412,
|
|
"loss": 5.8897,
|
|
"mean_token_accuracy": 0.14290473684668542,
|
|
"num_tokens": 3591436.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.16881251335144,
|
|
"epoch": 1.636288318144159,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000491443661397948,
|
|
"loss": 5.9887,
|
|
"mean_token_accuracy": 0.138546983897686,
|
|
"num_tokens": 3601054.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.1883124828338625,
|
|
"epoch": 1.64043082021541,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004913561567738368,
|
|
"loss": 5.9656,
|
|
"mean_token_accuracy": 0.14132906943559648,
|
|
"num_tokens": 3610641.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.135755014419556,
|
|
"epoch": 1.6445733222866612,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004912682157357146,
|
|
"loss": 5.8953,
|
|
"mean_token_accuracy": 0.13921359851956366,
|
|
"num_tokens": 3619351.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.056790399551391,
|
|
"epoch": 1.6487158243579123,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004911798384609675,
|
|
"loss": 5.8955,
|
|
"mean_token_accuracy": 0.14593140929937362,
|
|
"num_tokens": 3627341.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.211941194534302,
|
|
"epoch": 1.6528583264291632,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004910910251278626,
|
|
"loss": 5.944,
|
|
"mean_token_accuracy": 0.1503682926297188,
|
|
"num_tokens": 3637132.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.155584001541138,
|
|
"epoch": 1.6570008285004143,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004910017759155458,
|
|
"loss": 5.9901,
|
|
"mean_token_accuracy": 0.1395006664097309,
|
|
"num_tokens": 3646196.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6570008285004143,
|
|
"eval_entropy": 5.874153767567454,
|
|
"eval_loss": 6.1204142570495605,
|
|
"eval_mean_token_accuracy": 0.13903315654441492,
|
|
"eval_num_tokens": 3646196.0,
|
|
"eval_runtime": 2.1097,
|
|
"eval_samples_per_second": 1776.05,
|
|
"eval_steps_per_second": 222.302,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.080776929855347,
|
|
"epoch": 1.6611433305716652,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004909120910040427,
|
|
"loss": 5.9134,
|
|
"mean_token_accuracy": 0.14788957238197326,
|
|
"num_tokens": 3654658.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.328572368621826,
|
|
"epoch": 1.6652858326429163,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004908219705742574,
|
|
"loss": 6.1992,
|
|
"mean_token_accuracy": 0.1336714006960392,
|
|
"num_tokens": 3664509.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.124708271026611,
|
|
"epoch": 1.6694283347141674,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004907314148079727,
|
|
"loss": 5.9056,
|
|
"mean_token_accuracy": 0.144881671667099,
|
|
"num_tokens": 3672983.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.03100905418396,
|
|
"epoch": 1.6735708367854185,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004906404238878496,
|
|
"loss": 5.7658,
|
|
"mean_token_accuracy": 0.14805700182914733,
|
|
"num_tokens": 3681196.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.20078763961792,
|
|
"epoch": 1.6777133388566694,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004905489979974265,
|
|
"loss": 5.9474,
|
|
"mean_token_accuracy": 0.13829150870442392,
|
|
"num_tokens": 3690077.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.1578284740448,
|
|
"epoch": 1.6818558409279205,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004904571373211197,
|
|
"loss": 6.0161,
|
|
"mean_token_accuracy": 0.13646676018834114,
|
|
"num_tokens": 3700121.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.117290496826172,
|
|
"epoch": 1.6859983429991714,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000490364842044222,
|
|
"loss": 5.8542,
|
|
"mean_token_accuracy": 0.14759955778717995,
|
|
"num_tokens": 3709618.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.113605165481568,
|
|
"epoch": 1.6901408450704225,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004902721123529031,
|
|
"loss": 6.0396,
|
|
"mean_token_accuracy": 0.13407523557543755,
|
|
"num_tokens": 3719570.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.335561895370484,
|
|
"epoch": 1.6942833471416736,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000490178948434209,
|
|
"loss": 5.9767,
|
|
"mean_token_accuracy": 0.1504202052950859,
|
|
"num_tokens": 3729354.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.074905490875244,
|
|
"epoch": 1.6984258492129247,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004900853504760614,
|
|
"loss": 5.8788,
|
|
"mean_token_accuracy": 0.14436088725924492,
|
|
"num_tokens": 3738453.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.107450151443482,
|
|
"epoch": 1.7025683512841756,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004899913186672577,
|
|
"loss": 5.8637,
|
|
"mean_token_accuracy": 0.14600920379161836,
|
|
"num_tokens": 3747815.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.097668552398682,
|
|
"epoch": 1.7067108533554267,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004898968531974702,
|
|
"loss": 5.958,
|
|
"mean_token_accuracy": 0.14926589876413346,
|
|
"num_tokens": 3757298.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.155336618423462,
|
|
"epoch": 1.7108533554266776,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004898019542572462,
|
|
"loss": 5.9705,
|
|
"mean_token_accuracy": 0.14036926478147507,
|
|
"num_tokens": 3765939.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.2313727855682375,
|
|
"epoch": 1.7149958574979287,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004897066220380073,
|
|
"loss": 5.9794,
|
|
"mean_token_accuracy": 0.1348344750702381,
|
|
"num_tokens": 3775160.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.14722900390625,
|
|
"epoch": 1.7191383595691798,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004896108567320489,
|
|
"loss": 5.8782,
|
|
"mean_token_accuracy": 0.14222242087125778,
|
|
"num_tokens": 3785181.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.210811567306519,
|
|
"epoch": 1.723280861640431,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048951465853254,
|
|
"loss": 5.9942,
|
|
"mean_token_accuracy": 0.14354456216096878,
|
|
"num_tokens": 3794531.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.08496060371399,
|
|
"epoch": 1.7274233637116818,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004894180276335231,
|
|
"loss": 5.9737,
|
|
"mean_token_accuracy": 0.13818031474947928,
|
|
"num_tokens": 3804602.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.18796672821045,
|
|
"epoch": 1.731565865782933,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004893209642299132,
|
|
"loss": 6.0063,
|
|
"mean_token_accuracy": 0.14014846235513687,
|
|
"num_tokens": 3813394.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.179889488220215,
|
|
"epoch": 1.7357083678541838,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004892234685174976,
|
|
"loss": 5.9863,
|
|
"mean_token_accuracy": 0.1434843860566616,
|
|
"num_tokens": 3821604.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.189331436157227,
|
|
"epoch": 1.739850869925435,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004891255406929359,
|
|
"loss": 6.0006,
|
|
"mean_token_accuracy": 0.1404062420129776,
|
|
"num_tokens": 3831977.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.181633138656617,
|
|
"epoch": 1.743993371996686,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004890271809537595,
|
|
"loss": 5.9313,
|
|
"mean_token_accuracy": 0.1440660372376442,
|
|
"num_tokens": 3840684.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.090033292770386,
|
|
"epoch": 1.7481358740679371,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004889283894983704,
|
|
"loss": 5.8667,
|
|
"mean_token_accuracy": 0.13873573243618012,
|
|
"num_tokens": 3851147.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.180199146270752,
|
|
"epoch": 1.752278376139188,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000488829166526042,
|
|
"loss": 6.0012,
|
|
"mean_token_accuracy": 0.1477836012840271,
|
|
"num_tokens": 3859881.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.214416217803955,
|
|
"epoch": 1.7564208782104391,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004887295122369176,
|
|
"loss": 6.0564,
|
|
"mean_token_accuracy": 0.13372149169445038,
|
|
"num_tokens": 3870151.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.1593677520751955,
|
|
"epoch": 1.76056338028169,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000488629426832011,
|
|
"loss": 5.9708,
|
|
"mean_token_accuracy": 0.14135998338460923,
|
|
"num_tokens": 3878398.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.155690288543701,
|
|
"epoch": 1.7647058823529411,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004885289105132054,
|
|
"loss": 5.9198,
|
|
"mean_token_accuracy": 0.14018949940800668,
|
|
"num_tokens": 3888007.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.14916615486145,
|
|
"epoch": 1.7688483844241922,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004884279634832532,
|
|
"loss": 5.9146,
|
|
"mean_token_accuracy": 0.14503725469112397,
|
|
"num_tokens": 3896945.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.1898030757904055,
|
|
"epoch": 1.7729908864954433,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004883265859457754,
|
|
"loss": 5.9171,
|
|
"mean_token_accuracy": 0.15017052739858627,
|
|
"num_tokens": 3906049.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.06870813369751,
|
|
"epoch": 1.7771333885666944,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048822477810526186,
|
|
"loss": 5.894,
|
|
"mean_token_accuracy": 0.14274008721113204,
|
|
"num_tokens": 3915050.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.154209947586059,
|
|
"epoch": 1.7812758906379453,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048812254016706994,
|
|
"loss": 5.9662,
|
|
"mean_token_accuracy": 0.14590073004364967,
|
|
"num_tokens": 3924031.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.1595371723175045,
|
|
"epoch": 1.7854183927091962,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004880198723374247,
|
|
"loss": 6.0068,
|
|
"mean_token_accuracy": 0.1431715354323387,
|
|
"num_tokens": 3933756.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.157705354690552,
|
|
"epoch": 1.7895608947804473,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048791677482341845,
|
|
"loss": 5.9966,
|
|
"mean_token_accuracy": 0.13522283211350442,
|
|
"num_tokens": 3943129.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.053929376602173,
|
|
"epoch": 1.7937033968516984,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048781324783301016,
|
|
"loss": 5.8316,
|
|
"mean_token_accuracy": 0.14560012370347977,
|
|
"num_tokens": 3951494.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.1220502853393555,
|
|
"epoch": 1.7978458989229495,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048770929157502505,
|
|
"loss": 5.9907,
|
|
"mean_token_accuracy": 0.14323157966136932,
|
|
"num_tokens": 3960588.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.240426683425904,
|
|
"epoch": 1.8019884009942007,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004876049062591543,
|
|
"loss": 5.9738,
|
|
"mean_token_accuracy": 0.13992966637015342,
|
|
"num_tokens": 3970135.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.161054420471191,
|
|
"epoch": 1.8061309030654515,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048750009209595443,
|
|
"loss": 5.9449,
|
|
"mean_token_accuracy": 0.13832185566425323,
|
|
"num_tokens": 3979569.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.0378721237182615,
|
|
"epoch": 1.8102734051367024,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048739484929684714,
|
|
"loss": 5.8756,
|
|
"mean_token_accuracy": 0.1455282188951969,
|
|
"num_tokens": 3987885.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.15960202217102,
|
|
"epoch": 1.8144159072079535,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048728917807411856,
|
|
"loss": 6.0671,
|
|
"mean_token_accuracy": 0.13294378742575647,
|
|
"num_tokens": 3997726.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.212815666198731,
|
|
"epoch": 1.8185584092792046,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004871830786409191,
|
|
"loss": 5.8781,
|
|
"mean_token_accuracy": 0.14924565553665162,
|
|
"num_tokens": 4006437.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.0596874237060545,
|
|
"epoch": 1.8227009113504558,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.00048707655121126296,
|
|
"loss": 5.8415,
|
|
"mean_token_accuracy": 0.15142472386360167,
|
|
"num_tokens": 4015326.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.121099901199341,
|
|
"epoch": 1.8268434134217069,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048696959600002767,
|
|
"loss": 5.8944,
|
|
"mean_token_accuracy": 0.14363905936479568,
|
|
"num_tokens": 4024294.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.075892591476441,
|
|
"epoch": 1.8309859154929577,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048686221322295347,
|
|
"loss": 5.9181,
|
|
"mean_token_accuracy": 0.14851204603910445,
|
|
"num_tokens": 4033341.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.131572198867798,
|
|
"epoch": 1.8351284175642086,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048675440309664314,
|
|
"loss": 5.8773,
|
|
"mean_token_accuracy": 0.15065260529518126,
|
|
"num_tokens": 4042446.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.017001438140869,
|
|
"epoch": 1.8392709196354597,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048664616583856166,
|
|
"loss": 5.9139,
|
|
"mean_token_accuracy": 0.13879126980900763,
|
|
"num_tokens": 4051849.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.231928920745849,
|
|
"epoch": 1.8434134217067109,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048653750166703516,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.14212620705366136,
|
|
"num_tokens": 4061075.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.187917613983155,
|
|
"epoch": 1.847555923777962,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004864284108012513,
|
|
"loss": 5.9896,
|
|
"mean_token_accuracy": 0.134600942581892,
|
|
"num_tokens": 4071022.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.095667886734009,
|
|
"epoch": 1.851698425849213,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004863188934612583,
|
|
"loss": 5.8877,
|
|
"mean_token_accuracy": 0.14419684410095215,
|
|
"num_tokens": 4080728.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.050001907348633,
|
|
"epoch": 1.855840927920464,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004862089498679645,
|
|
"loss": 5.8452,
|
|
"mean_token_accuracy": 0.15132850408554077,
|
|
"num_tokens": 4088712.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.076476573944092,
|
|
"epoch": 1.8599834299917148,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004860985802431383,
|
|
"loss": 5.8787,
|
|
"mean_token_accuracy": 0.14517760127782822,
|
|
"num_tokens": 4097488.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.029026937484741,
|
|
"epoch": 1.864125932062966,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004859877848094071,
|
|
"loss": 5.8553,
|
|
"mean_token_accuracy": 0.1449560411274433,
|
|
"num_tokens": 4106485.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.144035577774048,
|
|
"epoch": 1.868268434134217,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004858765637902575,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.1511354774236679,
|
|
"num_tokens": 4115806.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.181568622589111,
|
|
"epoch": 1.8724109362054682,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004857649174100344,
|
|
"loss": 5.9816,
|
|
"mean_token_accuracy": 0.14631061926484107,
|
|
"num_tokens": 4124636.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.041752862930298,
|
|
"epoch": 1.8765534382767193,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004856528458939409,
|
|
"loss": 5.9225,
|
|
"mean_token_accuracy": 0.14047878235578537,
|
|
"num_tokens": 4133876.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.1034379482269285,
|
|
"epoch": 1.8806959403479702,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004855403494680373,
|
|
"loss": 5.838,
|
|
"mean_token_accuracy": 0.14113176017999648,
|
|
"num_tokens": 4143555.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.057332515716553,
|
|
"epoch": 1.884838442419221,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048542742835924133,
|
|
"loss": 5.8007,
|
|
"mean_token_accuracy": 0.1482735127210617,
|
|
"num_tokens": 4153007.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.935732364654541,
|
|
"epoch": 1.8889809444904722,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048531408279532705,
|
|
"loss": 5.8457,
|
|
"mean_token_accuracy": 0.14541342705488206,
|
|
"num_tokens": 4162107.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.094252443313598,
|
|
"epoch": 1.8931234465617233,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048520031300492507,
|
|
"loss": 5.8453,
|
|
"mean_token_accuracy": 0.14589096754789352,
|
|
"num_tokens": 4170920.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.151250314712525,
|
|
"epoch": 1.8972659486329744,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048508611921752124,
|
|
"loss": 5.8697,
|
|
"mean_token_accuracy": 0.14556538313627243,
|
|
"num_tokens": 4180226.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.978137493133545,
|
|
"epoch": 1.9014084507042255,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048497150166345703,
|
|
"loss": 5.7964,
|
|
"mean_token_accuracy": 0.14646997451782226,
|
|
"num_tokens": 4189283.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.02995023727417,
|
|
"epoch": 1.9055509527754764,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004848564605739287,
|
|
"loss": 5.8471,
|
|
"mean_token_accuracy": 0.14702720940113068,
|
|
"num_tokens": 4199017.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.117755317687989,
|
|
"epoch": 1.9096934548467275,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048474099618098644,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.1420730285346508,
|
|
"num_tokens": 4207756.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.051420879364014,
|
|
"epoch": 1.9138359569179784,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004846251087175348,
|
|
"loss": 5.8734,
|
|
"mean_token_accuracy": 0.14901573210954666,
|
|
"num_tokens": 4216721.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.079467821121216,
|
|
"epoch": 1.9179784589892295,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004845087984173312,
|
|
"loss": 5.8561,
|
|
"mean_token_accuracy": 0.14915963113307953,
|
|
"num_tokens": 4225596.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.1051336288452145,
|
|
"epoch": 1.9221209610604806,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048439206551498644,
|
|
"loss": 5.902,
|
|
"mean_token_accuracy": 0.1464046224951744,
|
|
"num_tokens": 4234500.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.042331075668335,
|
|
"epoch": 1.9262634631317317,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004842749102459635,
|
|
"loss": 5.8588,
|
|
"mean_token_accuracy": 0.14956332221627236,
|
|
"num_tokens": 4243240.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.009944772720337,
|
|
"epoch": 1.9304059652029826,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048415733284657727,
|
|
"loss": 5.8508,
|
|
"mean_token_accuracy": 0.15071502476930618,
|
|
"num_tokens": 4251892.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.074450302124023,
|
|
"epoch": 1.9345484672742337,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004840393335539943,
|
|
"loss": 5.8104,
|
|
"mean_token_accuracy": 0.14492977261543274,
|
|
"num_tokens": 4261447.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.101310682296753,
|
|
"epoch": 1.9386909693454846,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048392091260623195,
|
|
"loss": 5.953,
|
|
"mean_token_accuracy": 0.14990518093109131,
|
|
"num_tokens": 4270120.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.094955158233643,
|
|
"epoch": 1.9428334714167357,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048380207024215836,
|
|
"loss": 5.819,
|
|
"mean_token_accuracy": 0.14880019277334214,
|
|
"num_tokens": 4278300.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.136969947814942,
|
|
"epoch": 1.9469759734879868,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048368280670149145,
|
|
"loss": 5.8631,
|
|
"mean_token_accuracy": 0.15050907880067826,
|
|
"num_tokens": 4286454.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.95679988861084,
|
|
"epoch": 1.951118475559238,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004835631222247989,
|
|
"loss": 5.7432,
|
|
"mean_token_accuracy": 0.15112610012292862,
|
|
"num_tokens": 4294752.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.037843418121338,
|
|
"epoch": 1.9552609776304888,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048344301705349735,
|
|
"loss": 5.8094,
|
|
"mean_token_accuracy": 0.150174181163311,
|
|
"num_tokens": 4303195.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.089149713516235,
|
|
"epoch": 1.95940347970174,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048332249142985195,
|
|
"loss": 5.9467,
|
|
"mean_token_accuracy": 0.14172974601387978,
|
|
"num_tokens": 4311961.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.0431944847106935,
|
|
"epoch": 1.9635459817729908,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004832015455969763,
|
|
"loss": 5.799,
|
|
"mean_token_accuracy": 0.13888026773929596,
|
|
"num_tokens": 4322267.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.9912135124206545,
|
|
"epoch": 1.9676884838442419,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048308017979883114,
|
|
"loss": 5.9071,
|
|
"mean_token_accuracy": 0.1430051490664482,
|
|
"num_tokens": 4330584.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.976078844070434,
|
|
"epoch": 1.971830985915493,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048295839428022454,
|
|
"loss": 5.7541,
|
|
"mean_token_accuracy": 0.1447746202349663,
|
|
"num_tokens": 4339215.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.088702201843262,
|
|
"epoch": 1.975973487986744,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048283618928681127,
|
|
"loss": 5.7697,
|
|
"mean_token_accuracy": 0.15670698434114455,
|
|
"num_tokens": 4347711.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.980598735809326,
|
|
"epoch": 1.980115990057995,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004827135650650923,
|
|
"loss": 5.7687,
|
|
"mean_token_accuracy": 0.15119177773594855,
|
|
"num_tokens": 4356982.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.923292350769043,
|
|
"epoch": 1.984258492129246,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004825905218624139,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.1521460235118866,
|
|
"num_tokens": 4365378.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.200933170318604,
|
|
"epoch": 1.988400994200497,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048246705992696756,
|
|
"loss": 5.8582,
|
|
"mean_token_accuracy": 0.14180101156234742,
|
|
"num_tokens": 4374007.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.116976499557495,
|
|
"epoch": 1.992543496271748,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048234317950778975,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.14587089642882348,
|
|
"num_tokens": 4383909.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.156420421600342,
|
|
"epoch": 1.9966859983429992,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004822188808547607,
|
|
"loss": 6.012,
|
|
"mean_token_accuracy": 0.142410609126091,
|
|
"num_tokens": 4393145.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.003123235702515,
|
|
"epoch": 2.0008285004142503,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004820941642186043,
|
|
"loss": 5.7456,
|
|
"mean_token_accuracy": 0.1429907985031605,
|
|
"num_tokens": 4402024.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.080794095993042,
|
|
"epoch": 2.0049710024855014,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004819690298508879,
|
|
"loss": 5.6056,
|
|
"mean_token_accuracy": 0.1654096856713295,
|
|
"num_tokens": 4411377.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.021516275405884,
|
|
"epoch": 2.009113504556752,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048184347800402083,
|
|
"loss": 5.5028,
|
|
"mean_token_accuracy": 0.1704428642988205,
|
|
"num_tokens": 4419781.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.882586193084717,
|
|
"epoch": 2.013256006628003,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004817175089312551,
|
|
"loss": 5.4861,
|
|
"mean_token_accuracy": 0.168807016313076,
|
|
"num_tokens": 4428427.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.024956703186035,
|
|
"epoch": 2.0173985086992543,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048159112288668403,
|
|
"loss": 5.5291,
|
|
"mean_token_accuracy": 0.16385246813297272,
|
|
"num_tokens": 4438156.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.052538204193115,
|
|
"epoch": 2.0215410107705054,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048146432012524215,
|
|
"loss": 5.526,
|
|
"mean_token_accuracy": 0.16904665380716324,
|
|
"num_tokens": 4446342.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.926303672790527,
|
|
"epoch": 2.0256835128417565,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048133710090270446,
|
|
"loss": 5.5929,
|
|
"mean_token_accuracy": 0.157154780626297,
|
|
"num_tokens": 4455582.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.037592601776123,
|
|
"epoch": 2.0298260149130076,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048120946547568596,
|
|
"loss": 5.6155,
|
|
"mean_token_accuracy": 0.16731440871953965,
|
|
"num_tokens": 4464518.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.118183374404907,
|
|
"epoch": 2.0339685169842583,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048108141410164133,
|
|
"loss": 5.7143,
|
|
"mean_token_accuracy": 0.14911534935235976,
|
|
"num_tokens": 4475031.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.091432571411133,
|
|
"epoch": 2.0381110190555094,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004809529470388642,
|
|
"loss": 5.5973,
|
|
"mean_token_accuracy": 0.1624573528766632,
|
|
"num_tokens": 4484609.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.997477340698242,
|
|
"epoch": 2.0422535211267605,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004808240645464866,
|
|
"loss": 5.6256,
|
|
"mean_token_accuracy": 0.16317506730556489,
|
|
"num_tokens": 4494238.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.926339244842529,
|
|
"epoch": 2.0463960231980116,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004806947668844787,
|
|
"loss": 5.4941,
|
|
"mean_token_accuracy": 0.17045442909002304,
|
|
"num_tokens": 4503258.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.006639289855957,
|
|
"epoch": 2.0505385252692627,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048056505431364794,
|
|
"loss": 5.572,
|
|
"mean_token_accuracy": 0.16603607833385467,
|
|
"num_tokens": 4512071.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.930764484405517,
|
|
"epoch": 2.054681027340514,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048043492709563877,
|
|
"loss": 5.4034,
|
|
"mean_token_accuracy": 0.17490930408239364,
|
|
"num_tokens": 4520219.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.931649208068848,
|
|
"epoch": 2.0588235294117645,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048030438549293193,
|
|
"loss": 5.6195,
|
|
"mean_token_accuracy": 0.15992584079504013,
|
|
"num_tokens": 4529386.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.98941855430603,
|
|
"epoch": 2.0629660314830156,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048017342976884424,
|
|
"loss": 5.6158,
|
|
"mean_token_accuracy": 0.1569383054971695,
|
|
"num_tokens": 4538860.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.9581921100616455,
|
|
"epoch": 2.0671085335542667,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048004206018752773,
|
|
"loss": 5.5839,
|
|
"mean_token_accuracy": 0.15840956568717957,
|
|
"num_tokens": 4547842.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.900189781188965,
|
|
"epoch": 2.071251035625518,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047991027701396895,
|
|
"loss": 5.5734,
|
|
"mean_token_accuracy": 0.15809652507305144,
|
|
"num_tokens": 4556238.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.071251035625518,
|
|
"eval_entropy": 5.746327305653456,
|
|
"eval_loss": 6.01482629776001,
|
|
"eval_mean_token_accuracy": 0.1459539357755484,
|
|
"eval_num_tokens": 4556238.0,
|
|
"eval_runtime": 2.1145,
|
|
"eval_samples_per_second": 1772.047,
|
|
"eval_steps_per_second": 221.801,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.968504333496094,
|
|
"epoch": 2.075393537696769,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004797780805139892,
|
|
"loss": 5.5675,
|
|
"mean_token_accuracy": 0.16575732082128525,
|
|
"num_tokens": 4565684.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.919406795501709,
|
|
"epoch": 2.07953603976802,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004796454709542432,
|
|
"loss": 5.5206,
|
|
"mean_token_accuracy": 0.1661490336060524,
|
|
"num_tokens": 4575001.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.018925285339355,
|
|
"epoch": 2.083678541839271,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004795124486022189,
|
|
"loss": 5.6453,
|
|
"mean_token_accuracy": 0.15758625715970992,
|
|
"num_tokens": 4584128.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.9742213726043705,
|
|
"epoch": 2.087821043910522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047937901372623694,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.17302303165197372,
|
|
"num_tokens": 4593011.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.854166841506958,
|
|
"epoch": 2.091963545981773,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047924516659545005,
|
|
"loss": 5.5407,
|
|
"mean_token_accuracy": 0.16289012879133224,
|
|
"num_tokens": 4602676.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.035513019561767,
|
|
"epoch": 2.096106048053024,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004791109074798425,
|
|
"loss": 5.6102,
|
|
"mean_token_accuracy": 0.16002045273780824,
|
|
"num_tokens": 4611915.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.925036859512329,
|
|
"epoch": 2.100248550124275,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004789762366502296,
|
|
"loss": 5.5856,
|
|
"mean_token_accuracy": 0.16237015426158904,
|
|
"num_tokens": 4621396.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.973063850402832,
|
|
"epoch": 2.1043910521955262,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004788411543782571,
|
|
"loss": 5.5693,
|
|
"mean_token_accuracy": 0.1650550365447998,
|
|
"num_tokens": 4630533.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.918658590316772,
|
|
"epoch": 2.108533554266777,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004787056609364008,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.16128252744674682,
|
|
"num_tokens": 4639023.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.888208055496216,
|
|
"epoch": 2.112676056338028,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004785697565979657,
|
|
"loss": 5.5425,
|
|
"mean_token_accuracy": 0.165078566968441,
|
|
"num_tokens": 4647975.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.93974118232727,
|
|
"epoch": 2.116818558409279,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004784334416370857,
|
|
"loss": 5.553,
|
|
"mean_token_accuracy": 0.16065166890621185,
|
|
"num_tokens": 4657555.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.92600827217102,
|
|
"epoch": 2.1209610604805302,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004782967163287231,
|
|
"loss": 5.4749,
|
|
"mean_token_accuracy": 0.16520752906799316,
|
|
"num_tokens": 4667228.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.862721252441406,
|
|
"epoch": 2.1251035625517813,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047815958094866756,
|
|
"loss": 5.5451,
|
|
"mean_token_accuracy": 0.15971999764442443,
|
|
"num_tokens": 4676723.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.909252023696899,
|
|
"epoch": 2.1292460646230325,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004780220357735363,
|
|
"loss": 5.5297,
|
|
"mean_token_accuracy": 0.16149691641330718,
|
|
"num_tokens": 4686241.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.946609020233154,
|
|
"epoch": 2.1333885666942836,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047788408108077294,
|
|
"loss": 5.5983,
|
|
"mean_token_accuracy": 0.16169135719537736,
|
|
"num_tokens": 4695195.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.926496315002441,
|
|
"epoch": 2.1375310687655342,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004777457171486471,
|
|
"loss": 5.63,
|
|
"mean_token_accuracy": 0.1641976311802864,
|
|
"num_tokens": 4704716.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.916761875152588,
|
|
"epoch": 2.1416735708367853,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004776069442562541,
|
|
"loss": 5.611,
|
|
"mean_token_accuracy": 0.15986738204956055,
|
|
"num_tokens": 4713720.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.965861654281616,
|
|
"epoch": 2.1458160729080364,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004774677626835138,
|
|
"loss": 5.5809,
|
|
"mean_token_accuracy": 0.15728512555360794,
|
|
"num_tokens": 4722428.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.0213781833648685,
|
|
"epoch": 2.1499585749792876,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047732817271117084,
|
|
"loss": 5.5818,
|
|
"mean_token_accuracy": 0.16007290482521058,
|
|
"num_tokens": 4732395.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.908387517929077,
|
|
"epoch": 2.1541010770505387,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004771881746207934,
|
|
"loss": 5.5734,
|
|
"mean_token_accuracy": 0.1633661702275276,
|
|
"num_tokens": 4741071.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.949651002883911,
|
|
"epoch": 2.1582435791217898,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047704776869477285,
|
|
"loss": 5.5682,
|
|
"mean_token_accuracy": 0.15665056705474853,
|
|
"num_tokens": 4749876.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.9557836055755615,
|
|
"epoch": 2.1623860811930404,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047690695521632344,
|
|
"loss": 5.5276,
|
|
"mean_token_accuracy": 0.16186155378818512,
|
|
"num_tokens": 4758721.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.848019027709961,
|
|
"epoch": 2.1665285832642915,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047676573446948114,
|
|
"loss": 5.5518,
|
|
"mean_token_accuracy": 0.16213135570287704,
|
|
"num_tokens": 4767162.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.002100133895874,
|
|
"epoch": 2.1706710853355426,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047662410673910376,
|
|
"loss": 5.5617,
|
|
"mean_token_accuracy": 0.16884766817092894,
|
|
"num_tokens": 4776397.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.865268135070801,
|
|
"epoch": 2.1748135874067938,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047648207231087,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.16300645619630813,
|
|
"num_tokens": 4784822.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.916837501525879,
|
|
"epoch": 2.178956089478045,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004763396314712786,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.1626960352063179,
|
|
"num_tokens": 4794272.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.836860132217407,
|
|
"epoch": 2.183098591549296,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004761967845076484,
|
|
"loss": 5.5054,
|
|
"mean_token_accuracy": 0.15978458821773528,
|
|
"num_tokens": 4804314.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.909748697280884,
|
|
"epoch": 2.1872410936205466,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004760535317081174,
|
|
"loss": 5.5287,
|
|
"mean_token_accuracy": 0.1633632943034172,
|
|
"num_tokens": 4812685.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.921784448623657,
|
|
"epoch": 2.1913835956917977,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047590987336164207,
|
|
"loss": 5.6,
|
|
"mean_token_accuracy": 0.15942207127809524,
|
|
"num_tokens": 4821761.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.919152498245239,
|
|
"epoch": 2.195526097763049,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047576580975799705,
|
|
"loss": 5.6023,
|
|
"mean_token_accuracy": 0.1616935759782791,
|
|
"num_tokens": 4830515.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.972771167755127,
|
|
"epoch": 2.1996685998343,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004756213411877743,
|
|
"loss": 5.5531,
|
|
"mean_token_accuracy": 0.15991234183311462,
|
|
"num_tokens": 4840741.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.841622543334961,
|
|
"epoch": 2.203811101905551,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 5.511,
|
|
"mean_token_accuracy": 0.16772922724485398,
|
|
"num_tokens": 4849571.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.876320219039917,
|
|
"epoch": 2.207953603976802,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047533119031404757,
|
|
"loss": 5.4613,
|
|
"mean_token_accuracy": 0.17685824185609816,
|
|
"num_tokens": 4857194.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.845054769515992,
|
|
"epoch": 2.212096106048053,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047518550859580966,
|
|
"loss": 5.5546,
|
|
"mean_token_accuracy": 0.16370104104280472,
|
|
"num_tokens": 4865894.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.921027755737304,
|
|
"epoch": 2.216238608119304,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004750394230815249,
|
|
"loss": 5.5579,
|
|
"mean_token_accuracy": 0.16355429589748383,
|
|
"num_tokens": 4875131.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.816493511199951,
|
|
"epoch": 2.220381110190555,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047489293406586375,
|
|
"loss": 5.4248,
|
|
"mean_token_accuracy": 0.1692013368010521,
|
|
"num_tokens": 4884098.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.829316854476929,
|
|
"epoch": 2.224523612261806,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004747460418443106,
|
|
"loss": 5.4902,
|
|
"mean_token_accuracy": 0.16259288936853408,
|
|
"num_tokens": 4892963.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.944681453704834,
|
|
"epoch": 2.2286661143330573,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047459874671316305,
|
|
"loss": 5.5444,
|
|
"mean_token_accuracy": 0.1634556159377098,
|
|
"num_tokens": 4902097.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.78733811378479,
|
|
"epoch": 2.2328086164043084,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047445104896953153,
|
|
"loss": 5.4631,
|
|
"mean_token_accuracy": 0.17191764265298842,
|
|
"num_tokens": 4910469.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.055269050598144,
|
|
"epoch": 2.236951118475559,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047430294891133846,
|
|
"loss": 5.7319,
|
|
"mean_token_accuracy": 0.1553758502006531,
|
|
"num_tokens": 4921024.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.950701379776001,
|
|
"epoch": 2.24109362054681,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004741544468373179,
|
|
"loss": 5.5893,
|
|
"mean_token_accuracy": 0.1584577217698097,
|
|
"num_tokens": 4930297.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.880150985717774,
|
|
"epoch": 2.2452361226180613,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047400554304701474,
|
|
"loss": 5.5233,
|
|
"mean_token_accuracy": 0.16663775146007537,
|
|
"num_tokens": 4938971.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.836776924133301,
|
|
"epoch": 2.2493786246893124,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047385623784078427,
|
|
"loss": 5.5389,
|
|
"mean_token_accuracy": 0.16363985687494279,
|
|
"num_tokens": 4948531.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.934040164947509,
|
|
"epoch": 2.2535211267605635,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004737065315197913,
|
|
"loss": 5.6068,
|
|
"mean_token_accuracy": 0.16370140463113786,
|
|
"num_tokens": 4957430.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.97674069404602,
|
|
"epoch": 2.2576636288318146,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047355642438600987,
|
|
"loss": 5.6062,
|
|
"mean_token_accuracy": 0.1618436112999916,
|
|
"num_tokens": 4966681.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.824284076690674,
|
|
"epoch": 2.2618061309030653,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004734059167422225,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.16828390955924988,
|
|
"num_tokens": 4975327.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.818786573410034,
|
|
"epoch": 2.2659486329743164,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047325500889201956,
|
|
"loss": 5.5113,
|
|
"mean_token_accuracy": 0.1690054528415203,
|
|
"num_tokens": 4983741.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.873368835449218,
|
|
"epoch": 2.2700911350455675,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004731037011397986,
|
|
"loss": 5.6155,
|
|
"mean_token_accuracy": 0.15773370265960693,
|
|
"num_tokens": 4993171.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.998204851150513,
|
|
"epoch": 2.2742336371168186,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047295199379076403,
|
|
"loss": 5.6538,
|
|
"mean_token_accuracy": 0.15480987429618837,
|
|
"num_tokens": 5002810.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.870165014266968,
|
|
"epoch": 2.2783761391880697,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000472799887150926,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.1632503390312195,
|
|
"num_tokens": 5011875.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.883377504348755,
|
|
"epoch": 2.282518641259321,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004726473815271003,
|
|
"loss": 5.5717,
|
|
"mean_token_accuracy": 0.16406640112400056,
|
|
"num_tokens": 5020780.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.855701684951782,
|
|
"epoch": 2.286661143330572,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047249447722690747,
|
|
"loss": 5.4673,
|
|
"mean_token_accuracy": 0.16856606751680375,
|
|
"num_tokens": 5029896.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.917786359786987,
|
|
"epoch": 2.2908036454018226,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047234117455877215,
|
|
"loss": 5.5556,
|
|
"mean_token_accuracy": 0.15978156477212907,
|
|
"num_tokens": 5038956.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.8568426132202145,
|
|
"epoch": 2.2949461474730737,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004721874738319225,
|
|
"loss": 5.5006,
|
|
"mean_token_accuracy": 0.17033230215311052,
|
|
"num_tokens": 5047005.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.8400990009307865,
|
|
"epoch": 2.299088649544325,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004720333753563898,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.15479887425899505,
|
|
"num_tokens": 5056475.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.88454122543335,
|
|
"epoch": 2.303231151615576,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047187887944300747,
|
|
"loss": 5.5452,
|
|
"mean_token_accuracy": 0.1641763985157013,
|
|
"num_tokens": 5064978.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.871527290344238,
|
|
"epoch": 2.307373653686827,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004717239864034107,
|
|
"loss": 5.6136,
|
|
"mean_token_accuracy": 0.16027020514011384,
|
|
"num_tokens": 5074235.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.900209093093872,
|
|
"epoch": 2.3115161557580777,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047156869655003554,
|
|
"loss": 5.4717,
|
|
"mean_token_accuracy": 0.16512181013822555,
|
|
"num_tokens": 5082386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.812497186660766,
|
|
"epoch": 2.315658657829329,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047141301019611875,
|
|
"loss": 5.4607,
|
|
"mean_token_accuracy": 0.17094125151634215,
|
|
"num_tokens": 5090634.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.026579332351685,
|
|
"epoch": 2.31980115990058,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004712569276556967,
|
|
"loss": 5.624,
|
|
"mean_token_accuracy": 0.15455703735351561,
|
|
"num_tokens": 5099003.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.8493246078491214,
|
|
"epoch": 2.323943661971831,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00047110044924360484,
|
|
"loss": 5.7296,
|
|
"mean_token_accuracy": 0.1564518503844738,
|
|
"num_tokens": 5108631.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.867331886291504,
|
|
"epoch": 2.328086164043082,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047094357527547725,
|
|
"loss": 5.4878,
|
|
"mean_token_accuracy": 0.16400372236967087,
|
|
"num_tokens": 5117309.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.9126866340637205,
|
|
"epoch": 2.332228666114333,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004707863060677459,
|
|
"loss": 5.5899,
|
|
"mean_token_accuracy": 0.16657224744558335,
|
|
"num_tokens": 5126305.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.036413955688476,
|
|
"epoch": 2.3363711681855843,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047062864193764,
|
|
"loss": 5.667,
|
|
"mean_token_accuracy": 0.16238280534744262,
|
|
"num_tokens": 5135338.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.88200364112854,
|
|
"epoch": 2.340513670256835,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047047058320318525,
|
|
"loss": 5.5754,
|
|
"mean_token_accuracy": 0.16598308235406875,
|
|
"num_tokens": 5144518.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.895388412475586,
|
|
"epoch": 2.344656172328086,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004703121301832034,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.1575632408261299,
|
|
"num_tokens": 5154102.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.959476137161255,
|
|
"epoch": 2.348798674399337,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047015328319731153,
|
|
"loss": 5.6082,
|
|
"mean_token_accuracy": 0.15803269743919374,
|
|
"num_tokens": 5163719.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.876530647277832,
|
|
"epoch": 2.3529411764705883,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004699940425659215,
|
|
"loss": 5.5748,
|
|
"mean_token_accuracy": 0.16093172878026962,
|
|
"num_tokens": 5173192.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.8166028499603275,
|
|
"epoch": 2.3570836785418394,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046983440861023877,
|
|
"loss": 5.5116,
|
|
"mean_token_accuracy": 0.17299484312534333,
|
|
"num_tokens": 5182438.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.906814622879028,
|
|
"epoch": 2.36122618061309,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046967438165226275,
|
|
"loss": 5.6238,
|
|
"mean_token_accuracy": 0.1604929357767105,
|
|
"num_tokens": 5191595.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.841341638565064,
|
|
"epoch": 2.365368682684341,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004695139620147851,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.16756096631288528,
|
|
"num_tokens": 5200174.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.901956987380982,
|
|
"epoch": 2.3695111847555923,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046935315002138994,
|
|
"loss": 5.6061,
|
|
"mean_token_accuracy": 0.15783541053533554,
|
|
"num_tokens": 5210454.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.907624387741089,
|
|
"epoch": 2.3736536868268434,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004691919459964525,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.16504106521606446,
|
|
"num_tokens": 5219243.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.749259567260742,
|
|
"epoch": 2.3777961888980945,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046903035026513894,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.16370561122894287,
|
|
"num_tokens": 5228656.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.798469972610474,
|
|
"epoch": 2.3819386909693456,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004688683631534056,
|
|
"loss": 5.5367,
|
|
"mean_token_accuracy": 0.16182190775871277,
|
|
"num_tokens": 5237807.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.007817935943604,
|
|
"epoch": 2.3860811930405967,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004687059849879981,
|
|
"loss": 5.6304,
|
|
"mean_token_accuracy": 0.1664149597287178,
|
|
"num_tokens": 5247432.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.7761283874511715,
|
|
"epoch": 2.3902236951118474,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046854321609645093,
|
|
"loss": 5.4659,
|
|
"mean_token_accuracy": 0.16753794699907304,
|
|
"num_tokens": 5256680.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.857513284683227,
|
|
"epoch": 2.3943661971830985,
|
|
"grad_norm": 2.125,
|
|
"learning_rate": 0.00046838005680708677,
|
|
"loss": 5.5792,
|
|
"mean_token_accuracy": 0.15362140089273452,
|
|
"num_tokens": 5266793.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.812308025360108,
|
|
"epoch": 2.3985086992543496,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046821650744901573,
|
|
"loss": 5.4872,
|
|
"mean_token_accuracy": 0.16628407388925553,
|
|
"num_tokens": 5275028.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.76289963722229,
|
|
"epoch": 2.4026512013256007,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004680525683521347,
|
|
"loss": 5.5511,
|
|
"mean_token_accuracy": 0.1570456638932228,
|
|
"num_tokens": 5284162.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.000487470626831,
|
|
"epoch": 2.406793703396852,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004678882398471269,
|
|
"loss": 5.6353,
|
|
"mean_token_accuracy": 0.152632637321949,
|
|
"num_tokens": 5293263.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.866671895980835,
|
|
"epoch": 2.4109362054681025,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046772352226546066,
|
|
"loss": 5.5094,
|
|
"mean_token_accuracy": 0.17525156289339067,
|
|
"num_tokens": 5301373.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.868894290924072,
|
|
"epoch": 2.4150787075393536,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004675584159393895,
|
|
"loss": 5.56,
|
|
"mean_token_accuracy": 0.1619205355644226,
|
|
"num_tokens": 5311009.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.8928765773773195,
|
|
"epoch": 2.4192212096106047,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046739292120195083,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.16602448076009751,
|
|
"num_tokens": 5320341.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.891959476470947,
|
|
"epoch": 2.423363711681856,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004672270383869657,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.16422820091247559,
|
|
"num_tokens": 5329250.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.778356838226318,
|
|
"epoch": 2.427506213753107,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004670607678290379,
|
|
"loss": 5.49,
|
|
"mean_token_accuracy": 0.16743973568081855,
|
|
"num_tokens": 5336515.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.824636936187744,
|
|
"epoch": 2.431648715824358,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004668941098635532,
|
|
"loss": 5.5694,
|
|
"mean_token_accuracy": 0.1667967051267624,
|
|
"num_tokens": 5346043.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.857447242736816,
|
|
"epoch": 2.435791217895609,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046672706482667897,
|
|
"loss": 5.5213,
|
|
"mean_token_accuracy": 0.165392604470253,
|
|
"num_tokens": 5355124.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.807597255706787,
|
|
"epoch": 2.43993371996686,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004665596330553633,
|
|
"loss": 5.5492,
|
|
"mean_token_accuracy": 0.1653759300708771,
|
|
"num_tokens": 5364556.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.913714218139648,
|
|
"epoch": 2.444076222038111,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046639181488733445,
|
|
"loss": 5.5868,
|
|
"mean_token_accuracy": 0.16553183048963546,
|
|
"num_tokens": 5373778.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.8565521240234375,
|
|
"epoch": 2.448218724109362,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004662236106610999,
|
|
"loss": 5.5554,
|
|
"mean_token_accuracy": 0.1615779146552086,
|
|
"num_tokens": 5382633.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.795263338088989,
|
|
"epoch": 2.452361226180613,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046605502071594604,
|
|
"loss": 5.4583,
|
|
"mean_token_accuracy": 0.1652618780732155,
|
|
"num_tokens": 5391442.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.796332740783692,
|
|
"epoch": 2.4565037282518642,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004658860453919371,
|
|
"loss": 5.4625,
|
|
"mean_token_accuracy": 0.17088422924280167,
|
|
"num_tokens": 5400504.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.796403884887695,
|
|
"epoch": 2.460646230323115,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004657166850299149,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.1733391761779785,
|
|
"num_tokens": 5408939.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.825985050201416,
|
|
"epoch": 2.464788732394366,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004655469399714978,
|
|
"loss": 5.5146,
|
|
"mean_token_accuracy": 0.16244352757930755,
|
|
"num_tokens": 5418934.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.7783496379852295,
|
|
"epoch": 2.468931234465617,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004653768105590799,
|
|
"loss": 5.5324,
|
|
"mean_token_accuracy": 0.16445797234773635,
|
|
"num_tokens": 5427894.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.853815174102783,
|
|
"epoch": 2.4730737365368682,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046520629713583115,
|
|
"loss": 5.514,
|
|
"mean_token_accuracy": 0.1653539478778839,
|
|
"num_tokens": 5436830.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.760482311248779,
|
|
"epoch": 2.4772162386081193,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046503540004569555,
|
|
"loss": 5.5041,
|
|
"mean_token_accuracy": 0.17337694615125657,
|
|
"num_tokens": 5445504.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.837162828445434,
|
|
"epoch": 2.4813587406793705,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046486411963339125,
|
|
"loss": 5.5684,
|
|
"mean_token_accuracy": 0.15989079922437668,
|
|
"num_tokens": 5455378.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.914019918441772,
|
|
"epoch": 2.4855012427506216,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046469245624440955,
|
|
"loss": 5.6317,
|
|
"mean_token_accuracy": 0.1610162228345871,
|
|
"num_tokens": 5465334.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.4855012427506216,
|
|
"eval_entropy": 5.710150802059214,
|
|
"eval_loss": 5.937281131744385,
|
|
"eval_mean_token_accuracy": 0.1504057919118069,
|
|
"eval_num_tokens": 5465334.0,
|
|
"eval_runtime": 2.1173,
|
|
"eval_samples_per_second": 1769.724,
|
|
"eval_steps_per_second": 221.511,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.926897859573364,
|
|
"epoch": 2.4896437448218722,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046452041022501423,
|
|
"loss": 5.6075,
|
|
"mean_token_accuracy": 0.16159055083990098,
|
|
"num_tokens": 5475701.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.838952112197876,
|
|
"epoch": 2.4937862468931233,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004643479819222411,
|
|
"loss": 5.51,
|
|
"mean_token_accuracy": 0.16615799814462662,
|
|
"num_tokens": 5485244.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.78017258644104,
|
|
"epoch": 2.4979287489643744,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046417517168389664,
|
|
"loss": 5.4661,
|
|
"mean_token_accuracy": 0.17169951945543288,
|
|
"num_tokens": 5493587.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.816814327239991,
|
|
"epoch": 2.5020712510356256,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004640019798585581,
|
|
"loss": 5.604,
|
|
"mean_token_accuracy": 0.16011267006397248,
|
|
"num_tokens": 5503117.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.915554571151733,
|
|
"epoch": 2.5062137531068767,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046382840679557233,
|
|
"loss": 5.5685,
|
|
"mean_token_accuracy": 0.1609106868505478,
|
|
"num_tokens": 5512742.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.85755558013916,
|
|
"epoch": 2.5103562551781273,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046365445284505516,
|
|
"loss": 5.5291,
|
|
"mean_token_accuracy": 0.16356179416179656,
|
|
"num_tokens": 5521862.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.7801220417022705,
|
|
"epoch": 2.5144987572493784,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004634801183578907,
|
|
"loss": 5.5125,
|
|
"mean_token_accuracy": 0.16270552426576615,
|
|
"num_tokens": 5531244.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.8893523693084715,
|
|
"epoch": 2.5186412593206295,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046330540368573063,
|
|
"loss": 5.5694,
|
|
"mean_token_accuracy": 0.15749936699867248,
|
|
"num_tokens": 5540786.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.881492185592651,
|
|
"epoch": 2.5227837613918807,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046313030918099365,
|
|
"loss": 5.548,
|
|
"mean_token_accuracy": 0.1654050186276436,
|
|
"num_tokens": 5549961.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.876528787612915,
|
|
"epoch": 2.5269262634631318,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046295483519686437,
|
|
"loss": 5.5095,
|
|
"mean_token_accuracy": 0.1663644015789032,
|
|
"num_tokens": 5558941.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.836793231964111,
|
|
"epoch": 2.531068765534383,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046277898208729306,
|
|
"loss": 5.5097,
|
|
"mean_token_accuracy": 0.16931835263967515,
|
|
"num_tokens": 5567630.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.802315425872803,
|
|
"epoch": 2.535211267605634,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004626027502069947,
|
|
"loss": 5.4744,
|
|
"mean_token_accuracy": 0.16817359030246734,
|
|
"num_tokens": 5576430.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.852225017547608,
|
|
"epoch": 2.539353769676885,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046242613991144814,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.16349476873874663,
|
|
"num_tokens": 5584785.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.715366697311401,
|
|
"epoch": 2.5434962717481358,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046224915155689586,
|
|
"loss": 5.497,
|
|
"mean_token_accuracy": 0.16514008194208146,
|
|
"num_tokens": 5594583.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.911450958251953,
|
|
"epoch": 2.547638773819387,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046207178550034244,
|
|
"loss": 5.4651,
|
|
"mean_token_accuracy": 0.16805889904499055,
|
|
"num_tokens": 5603498.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.896241235733032,
|
|
"epoch": 2.551781275890638,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004618940420995547,
|
|
"loss": 5.6092,
|
|
"mean_token_accuracy": 0.15735539495944978,
|
|
"num_tokens": 5613101.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.856862783432007,
|
|
"epoch": 2.555923777961889,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004617159217130606,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.15954624712467194,
|
|
"num_tokens": 5621821.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.858033895492554,
|
|
"epoch": 2.5600662800331397,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004615374247001483,
|
|
"loss": 5.5965,
|
|
"mean_token_accuracy": 0.163676854968071,
|
|
"num_tokens": 5631006.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.821109580993652,
|
|
"epoch": 2.564208782104391,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004613585514208659,
|
|
"loss": 5.584,
|
|
"mean_token_accuracy": 0.15758810192346573,
|
|
"num_tokens": 5640159.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.809291362762451,
|
|
"epoch": 2.568351284175642,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046117930223602016,
|
|
"loss": 5.4131,
|
|
"mean_token_accuracy": 0.17075895965099336,
|
|
"num_tokens": 5648369.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.7870830535888675,
|
|
"epoch": 2.572493786246893,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004609996775071765,
|
|
"loss": 5.497,
|
|
"mean_token_accuracy": 0.1685049504041672,
|
|
"num_tokens": 5657669.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.8332267761230465,
|
|
"epoch": 2.576636288318144,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004608196775966575,
|
|
"loss": 5.5545,
|
|
"mean_token_accuracy": 0.1708658054471016,
|
|
"num_tokens": 5666601.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.808360290527344,
|
|
"epoch": 2.5807787903893953,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046063930286754267,
|
|
"loss": 5.4682,
|
|
"mean_token_accuracy": 0.16872803270816802,
|
|
"num_tokens": 5675243.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.720640659332275,
|
|
"epoch": 2.5849212924606464,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046045855368366756,
|
|
"loss": 5.4876,
|
|
"mean_token_accuracy": 0.16355924159288407,
|
|
"num_tokens": 5683646.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.864615392684937,
|
|
"epoch": 2.5890637945318975,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000460277430409623,
|
|
"loss": 5.5573,
|
|
"mean_token_accuracy": 0.17474440187215806,
|
|
"num_tokens": 5693069.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.744599151611328,
|
|
"epoch": 2.593206296603148,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046009593341075444,
|
|
"loss": 5.4103,
|
|
"mean_token_accuracy": 0.17491702139377593,
|
|
"num_tokens": 5701527.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.771568346023559,
|
|
"epoch": 2.5973487986743993,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004599140630531612,
|
|
"loss": 5.5715,
|
|
"mean_token_accuracy": 0.163454470038414,
|
|
"num_tokens": 5710379.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.902339601516724,
|
|
"epoch": 2.6014913007456504,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004597318197036956,
|
|
"loss": 5.4832,
|
|
"mean_token_accuracy": 0.16780744194984437,
|
|
"num_tokens": 5719217.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.821617698669433,
|
|
"epoch": 2.6056338028169015,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004595492037299624,
|
|
"loss": 5.5687,
|
|
"mean_token_accuracy": 0.16430330276489258,
|
|
"num_tokens": 5728331.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.791578912734986,
|
|
"epoch": 2.6097763048881526,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045936621550031806,
|
|
"loss": 5.5365,
|
|
"mean_token_accuracy": 0.16430482566356658,
|
|
"num_tokens": 5737696.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.895357322692871,
|
|
"epoch": 2.6139188069594033,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045918285538386973,
|
|
"loss": 5.6027,
|
|
"mean_token_accuracy": 0.16820576936006545,
|
|
"num_tokens": 5746191.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.865847158432007,
|
|
"epoch": 2.6180613090306544,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045899912375047485,
|
|
"loss": 5.5974,
|
|
"mean_token_accuracy": 0.16639549508690835,
|
|
"num_tokens": 5754824.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.825492715835571,
|
|
"epoch": 2.6222038111019055,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045881502097074023,
|
|
"loss": 5.5633,
|
|
"mean_token_accuracy": 0.1547349810600281,
|
|
"num_tokens": 5764863.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.7731404304504395,
|
|
"epoch": 2.6263463131731566,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004586305474160212,
|
|
"loss": 5.5672,
|
|
"mean_token_accuracy": 0.1573731079697609,
|
|
"num_tokens": 5774273.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.809962558746338,
|
|
"epoch": 2.6304888152444077,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045844570345842116,
|
|
"loss": 5.537,
|
|
"mean_token_accuracy": 0.1640781581401825,
|
|
"num_tokens": 5784085.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.81432318687439,
|
|
"epoch": 2.634631317315659,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004582604894707906,
|
|
"loss": 5.5467,
|
|
"mean_token_accuracy": 0.17041893154382706,
|
|
"num_tokens": 5793921.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.827432489395141,
|
|
"epoch": 2.63877381938691,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045807490582672634,
|
|
"loss": 5.5111,
|
|
"mean_token_accuracy": 0.15524204820394516,
|
|
"num_tokens": 5803902.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.774159717559814,
|
|
"epoch": 2.6429163214581606,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045788895290057083,
|
|
"loss": 5.4285,
|
|
"mean_token_accuracy": 0.17325582951307297,
|
|
"num_tokens": 5812089.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.891901969909668,
|
|
"epoch": 2.6470588235294117,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045770263106741137,
|
|
"loss": 5.6104,
|
|
"mean_token_accuracy": 0.1625809907913208,
|
|
"num_tokens": 5821072.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.821473503112793,
|
|
"epoch": 2.651201325600663,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004575159407030797,
|
|
"loss": 5.4379,
|
|
"mean_token_accuracy": 0.16588133871555327,
|
|
"num_tokens": 5829725.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.675117111206054,
|
|
"epoch": 2.655343827671914,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004573288821841505,
|
|
"loss": 5.5024,
|
|
"mean_token_accuracy": 0.16882624328136445,
|
|
"num_tokens": 5838153.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.7418474674224855,
|
|
"epoch": 2.659486329743165,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004571414558879413,
|
|
"loss": 5.4539,
|
|
"mean_token_accuracy": 0.16360593885183333,
|
|
"num_tokens": 5847270.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.831032466888428,
|
|
"epoch": 2.6636288318144157,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004569536621925114,
|
|
"loss": 5.5782,
|
|
"mean_token_accuracy": 0.15531986355781555,
|
|
"num_tokens": 5855954.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.787395000457764,
|
|
"epoch": 2.667771333885667,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004567655014766612,
|
|
"loss": 5.4812,
|
|
"mean_token_accuracy": 0.16389115303754806,
|
|
"num_tokens": 5864435.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.83512053489685,
|
|
"epoch": 2.671913835956918,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004565769741199314,
|
|
"loss": 5.5554,
|
|
"mean_token_accuracy": 0.16735721826553346,
|
|
"num_tokens": 5873964.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.89815936088562,
|
|
"epoch": 2.676056338028169,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045638808050260237,
|
|
"loss": 5.5158,
|
|
"mean_token_accuracy": 0.1644240438938141,
|
|
"num_tokens": 5882738.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.745797443389892,
|
|
"epoch": 2.68019884009942,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045619882100569317,
|
|
"loss": 5.6269,
|
|
"mean_token_accuracy": 0.16016174107789993,
|
|
"num_tokens": 5891749.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.860331964492798,
|
|
"epoch": 2.684341342170671,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004560091960109608,
|
|
"loss": 5.5678,
|
|
"mean_token_accuracy": 0.15677377581596375,
|
|
"num_tokens": 5900551.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.871357536315918,
|
|
"epoch": 2.6884838442419223,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004558192059008996,
|
|
"loss": 5.5854,
|
|
"mean_token_accuracy": 0.15826322734355927,
|
|
"num_tokens": 5909254.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.788536500930786,
|
|
"epoch": 2.692626346313173,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004556288510587406,
|
|
"loss": 5.5187,
|
|
"mean_token_accuracy": 0.16280469447374343,
|
|
"num_tokens": 5918357.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.865714168548584,
|
|
"epoch": 2.696768848384424,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045543813186845004,
|
|
"loss": 5.527,
|
|
"mean_token_accuracy": 0.15844795256853103,
|
|
"num_tokens": 5927742.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.7749183177948,
|
|
"epoch": 2.700911350455675,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004552470487147294,
|
|
"loss": 5.6026,
|
|
"mean_token_accuracy": 0.1588852234184742,
|
|
"num_tokens": 5937025.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.915446710586548,
|
|
"epoch": 2.7050538525269263,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045505560198301443,
|
|
"loss": 5.5988,
|
|
"mean_token_accuracy": 0.15682683289051055,
|
|
"num_tokens": 5946176.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.871020174026489,
|
|
"epoch": 2.7091963545981774,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045486379205947396,
|
|
"loss": 5.5523,
|
|
"mean_token_accuracy": 0.16398580521345138,
|
|
"num_tokens": 5955550.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.758722591400146,
|
|
"epoch": 2.713338856669428,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004546716193310098,
|
|
"loss": 5.4882,
|
|
"mean_token_accuracy": 0.1692626565694809,
|
|
"num_tokens": 5964467.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.771913814544678,
|
|
"epoch": 2.717481358740679,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000454479084185255,
|
|
"loss": 5.5132,
|
|
"mean_token_accuracy": 0.16013832688331603,
|
|
"num_tokens": 5973906.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.870921897888183,
|
|
"epoch": 2.7216238608119303,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004542861870105743,
|
|
"loss": 5.5602,
|
|
"mean_token_accuracy": 0.15937376916408538,
|
|
"num_tokens": 5983701.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.809008455276489,
|
|
"epoch": 2.7257663628831814,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00045409292819606224,
|
|
"loss": 5.515,
|
|
"mean_token_accuracy": 0.16878179907798768,
|
|
"num_tokens": 5993367.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.755811977386474,
|
|
"epoch": 2.7299088649544325,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045389930813154297,
|
|
"loss": 5.4848,
|
|
"mean_token_accuracy": 0.1650410145521164,
|
|
"num_tokens": 6003752.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.856390428543091,
|
|
"epoch": 2.7340513670256836,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004537053272075693,
|
|
"loss": 5.5551,
|
|
"mean_token_accuracy": 0.16377977281808853,
|
|
"num_tokens": 6012486.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.740771055221558,
|
|
"epoch": 2.7381938690969347,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045351098581542204,
|
|
"loss": 5.489,
|
|
"mean_token_accuracy": 0.17124730348587036,
|
|
"num_tokens": 6020742.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.818489265441895,
|
|
"epoch": 2.742336371168186,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004533162843471089,
|
|
"loss": 5.5387,
|
|
"mean_token_accuracy": 0.1596716359257698,
|
|
"num_tokens": 6030248.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.813158750534058,
|
|
"epoch": 2.7464788732394365,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004531212231953641,
|
|
"loss": 5.5722,
|
|
"mean_token_accuracy": 0.15962957441806794,
|
|
"num_tokens": 6039293.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.779981088638306,
|
|
"epoch": 2.7506213753106876,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004529258027536473,
|
|
"loss": 5.4464,
|
|
"mean_token_accuracy": 0.17154055833816528,
|
|
"num_tokens": 6048081.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.797445249557495,
|
|
"epoch": 2.7547638773819387,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004527300234161428,
|
|
"loss": 5.5916,
|
|
"mean_token_accuracy": 0.16878824979066848,
|
|
"num_tokens": 6057549.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.834072637557983,
|
|
"epoch": 2.75890637945319,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004525338855777591,
|
|
"loss": 5.5875,
|
|
"mean_token_accuracy": 0.15878960192203523,
|
|
"num_tokens": 6067372.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.791154527664185,
|
|
"epoch": 2.7630488815244405,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045233738963412746,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.17494594007730485,
|
|
"num_tokens": 6075532.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.758261442184448,
|
|
"epoch": 2.7671913835956916,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045214053598160173,
|
|
"loss": 5.4585,
|
|
"mean_token_accuracy": 0.16725144833326339,
|
|
"num_tokens": 6084389.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.75204610824585,
|
|
"epoch": 2.7713338856669427,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045194332501725725,
|
|
"loss": 5.4895,
|
|
"mean_token_accuracy": 0.1700201228260994,
|
|
"num_tokens": 6093536.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.848310661315918,
|
|
"epoch": 2.775476387738194,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045174575713889014,
|
|
"loss": 5.5422,
|
|
"mean_token_accuracy": 0.1688771940767765,
|
|
"num_tokens": 6102677.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.802322435379028,
|
|
"epoch": 2.779618889809445,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004515478327450164,
|
|
"loss": 5.6236,
|
|
"mean_token_accuracy": 0.160081122815609,
|
|
"num_tokens": 6112430.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.809485006332397,
|
|
"epoch": 2.783761391880696,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045134955223487093,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.16987434178590774,
|
|
"num_tokens": 6121408.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.800628662109375,
|
|
"epoch": 2.787903893951947,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045115091600840757,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.1611669510602951,
|
|
"num_tokens": 6131339.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.820253705978393,
|
|
"epoch": 2.7920463960231983,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045095192446629697,
|
|
"loss": 5.5177,
|
|
"mean_token_accuracy": 0.16406785994768142,
|
|
"num_tokens": 6140590.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.92112512588501,
|
|
"epoch": 2.796188898094449,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000450752578009927,
|
|
"loss": 5.5965,
|
|
"mean_token_accuracy": 0.1622408464550972,
|
|
"num_tokens": 6149958.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.754869365692139,
|
|
"epoch": 2.8003314001657,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045055287704140115,
|
|
"loss": 5.4434,
|
|
"mean_token_accuracy": 0.16395304203033448,
|
|
"num_tokens": 6159411.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.766679859161377,
|
|
"epoch": 2.804473902236951,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004503528219635381,
|
|
"loss": 5.5309,
|
|
"mean_token_accuracy": 0.1593677282333374,
|
|
"num_tokens": 6169264.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.821379137039185,
|
|
"epoch": 2.8086164043082023,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004501524131798709,
|
|
"loss": 5.5095,
|
|
"mean_token_accuracy": 0.1638065531849861,
|
|
"num_tokens": 6178374.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.865478467941284,
|
|
"epoch": 2.812758906379453,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000449951651094646,
|
|
"loss": 5.5153,
|
|
"mean_token_accuracy": 0.15953427255153657,
|
|
"num_tokens": 6186918.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.715597677230835,
|
|
"epoch": 2.816901408450704,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044975053611282217,
|
|
"loss": 5.5713,
|
|
"mean_token_accuracy": 0.16546980440616607,
|
|
"num_tokens": 6196400.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.685925531387329,
|
|
"epoch": 2.821043910521955,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004495490686400706,
|
|
"loss": 5.4138,
|
|
"mean_token_accuracy": 0.1680915981531143,
|
|
"num_tokens": 6204849.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.814952373504639,
|
|
"epoch": 2.8251864125932062,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044934724908277313,
|
|
"loss": 5.5133,
|
|
"mean_token_accuracy": 0.16838027983903886,
|
|
"num_tokens": 6214121.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.767223501205445,
|
|
"epoch": 2.8293289146644574,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044914507784802184,
|
|
"loss": 5.506,
|
|
"mean_token_accuracy": 0.1660693109035492,
|
|
"num_tokens": 6223163.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.7958411693573,
|
|
"epoch": 2.8334714167357085,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004489425553436182,
|
|
"loss": 5.5504,
|
|
"mean_token_accuracy": 0.16537869721651077,
|
|
"num_tokens": 6231581.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.861129331588745,
|
|
"epoch": 2.8376139188069596,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044873968197807213,
|
|
"loss": 5.605,
|
|
"mean_token_accuracy": 0.16778871268033982,
|
|
"num_tokens": 6240015.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.894214105606079,
|
|
"epoch": 2.8417564208782107,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044853645816060156,
|
|
"loss": 5.6105,
|
|
"mean_token_accuracy": 0.16154228299856185,
|
|
"num_tokens": 6249083.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.889301395416259,
|
|
"epoch": 2.8458989229494613,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044833288430113104,
|
|
"loss": 5.5802,
|
|
"mean_token_accuracy": 0.16691730171442032,
|
|
"num_tokens": 6258798.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.787882900238037,
|
|
"epoch": 2.8500414250207124,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044812896081029136,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.16846126168966294,
|
|
"num_tokens": 6267819.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.799396944046021,
|
|
"epoch": 2.8541839270919636,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004479246880994183,
|
|
"loss": 5.5435,
|
|
"mean_token_accuracy": 0.16353802010416985,
|
|
"num_tokens": 6276915.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.8152307033538815,
|
|
"epoch": 2.8583264291632147,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004477200665805525,
|
|
"loss": 5.5442,
|
|
"mean_token_accuracy": 0.16179773956537247,
|
|
"num_tokens": 6285880.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.81086540222168,
|
|
"epoch": 2.8624689312344658,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044751509666643777,
|
|
"loss": 5.6586,
|
|
"mean_token_accuracy": 0.15439242944121362,
|
|
"num_tokens": 6295408.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.7647120475769045,
|
|
"epoch": 2.8666114333057164,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044730977877052084,
|
|
"loss": 5.4169,
|
|
"mean_token_accuracy": 0.16521361470222473,
|
|
"num_tokens": 6304110.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.833881855010986,
|
|
"epoch": 2.8707539353769675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044710411330695045,
|
|
"loss": 5.568,
|
|
"mean_token_accuracy": 0.1649985432624817,
|
|
"num_tokens": 6312577.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.951030969619751,
|
|
"epoch": 2.8748964374482187,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044689810069057615,
|
|
"loss": 5.6516,
|
|
"mean_token_accuracy": 0.15365287512540818,
|
|
"num_tokens": 6321766.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.880116176605225,
|
|
"epoch": 2.8790389395194698,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004466917413369479,
|
|
"loss": 5.5967,
|
|
"mean_token_accuracy": 0.16034159660339356,
|
|
"num_tokens": 6330327.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.778976106643677,
|
|
"epoch": 2.883181441590721,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004464850356623152,
|
|
"loss": 5.5348,
|
|
"mean_token_accuracy": 0.16730698198080063,
|
|
"num_tokens": 6339596.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.77376914024353,
|
|
"epoch": 2.887323943661972,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044627798408362597,
|
|
"loss": 5.4292,
|
|
"mean_token_accuracy": 0.17025616765022278,
|
|
"num_tokens": 6348080.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.812956380844116,
|
|
"epoch": 2.891466445733223,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004460705870185258,
|
|
"loss": 5.555,
|
|
"mean_token_accuracy": 0.1581030383706093,
|
|
"num_tokens": 6357725.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.778322029113769,
|
|
"epoch": 2.8956089478044738,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004458628448853572,
|
|
"loss": 5.4931,
|
|
"mean_token_accuracy": 0.16816313415765763,
|
|
"num_tokens": 6366821.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.932670783996582,
|
|
"epoch": 2.899751449875725,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004456547581031589,
|
|
"loss": 5.5567,
|
|
"mean_token_accuracy": 0.16425554677844048,
|
|
"num_tokens": 6375641.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.899751449875725,
|
|
"eval_entropy": 5.551940224572285,
|
|
"eval_loss": 5.856940746307373,
|
|
"eval_mean_token_accuracy": 0.15571236410247746,
|
|
"eval_num_tokens": 6375641.0,
|
|
"eval_runtime": 2.1164,
|
|
"eval_samples_per_second": 1770.42,
|
|
"eval_steps_per_second": 221.598,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.725690841674805,
|
|
"epoch": 2.903893951946976,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004454463270916646,
|
|
"loss": 5.4236,
|
|
"mean_token_accuracy": 0.1692560002207756,
|
|
"num_tokens": 6384596.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.945777463912964,
|
|
"epoch": 2.908036454018227,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004452375522713023,
|
|
"loss": 5.7809,
|
|
"mean_token_accuracy": 0.14747752845287324,
|
|
"num_tokens": 6395048.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.800138902664185,
|
|
"epoch": 2.912178956089478,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004450284340631938,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.15904973149299623,
|
|
"num_tokens": 6404828.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.837668132781983,
|
|
"epoch": 2.916321458160729,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044481897288915334,
|
|
"loss": 5.485,
|
|
"mean_token_accuracy": 0.1632108837366104,
|
|
"num_tokens": 6414670.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.610891962051392,
|
|
"epoch": 2.92046396023198,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044460916917168697,
|
|
"loss": 5.4041,
|
|
"mean_token_accuracy": 0.17070720791816713,
|
|
"num_tokens": 6423446.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.696707391738892,
|
|
"epoch": 2.924606462303231,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004443990233339918,
|
|
"loss": 5.4998,
|
|
"mean_token_accuracy": 0.15940917879343033,
|
|
"num_tokens": 6433412.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.852585029602051,
|
|
"epoch": 2.928748964374482,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004441885357999548,
|
|
"loss": 5.4658,
|
|
"mean_token_accuracy": 0.17039099633693694,
|
|
"num_tokens": 6443520.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.802426815032959,
|
|
"epoch": 2.9328914664457333,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004439777069941524,
|
|
"loss": 5.5817,
|
|
"mean_token_accuracy": 0.15985384210944176,
|
|
"num_tokens": 6452933.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.731008005142212,
|
|
"epoch": 2.9370339685169844,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044376653734184935,
|
|
"loss": 5.4004,
|
|
"mean_token_accuracy": 0.16928976774215698,
|
|
"num_tokens": 6462113.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.657544660568237,
|
|
"epoch": 2.9411764705882355,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004435550272689979,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.17204506397247316,
|
|
"num_tokens": 6471411.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.846671056747437,
|
|
"epoch": 2.945318972659486,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044334317720223706,
|
|
"loss": 5.5639,
|
|
"mean_token_accuracy": 0.16564810276031494,
|
|
"num_tokens": 6480682.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.8505572319030765,
|
|
"epoch": 2.9494614747307373,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044313098756889154,
|
|
"loss": 5.5983,
|
|
"mean_token_accuracy": 0.16059545576572418,
|
|
"num_tokens": 6490614.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.79144139289856,
|
|
"epoch": 2.9536039768019884,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004429184587969708,
|
|
"loss": 5.4556,
|
|
"mean_token_accuracy": 0.16850335150957108,
|
|
"num_tokens": 6499079.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.83440933227539,
|
|
"epoch": 2.9577464788732395,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004427055913151689,
|
|
"loss": 5.5556,
|
|
"mean_token_accuracy": 0.15629634037613868,
|
|
"num_tokens": 6508556.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.852693748474121,
|
|
"epoch": 2.9618889809444906,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004424923855528626,
|
|
"loss": 5.5747,
|
|
"mean_token_accuracy": 0.16567572355270385,
|
|
"num_tokens": 6517934.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.789475774765014,
|
|
"epoch": 2.9660314830157413,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004422788419401114,
|
|
"loss": 5.4591,
|
|
"mean_token_accuracy": 0.17523163408041,
|
|
"num_tokens": 6526995.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.817619132995605,
|
|
"epoch": 2.9701739850869924,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044206496090765595,
|
|
"loss": 5.5147,
|
|
"mean_token_accuracy": 0.15977577418088912,
|
|
"num_tokens": 6535264.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.831095218658447,
|
|
"epoch": 2.9743164871582435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044185074288691776,
|
|
"loss": 5.4146,
|
|
"mean_token_accuracy": 0.17837767899036408,
|
|
"num_tokens": 6544498.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.684333181381225,
|
|
"epoch": 2.9784589892294946,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044163618830999794,
|
|
"loss": 5.5215,
|
|
"mean_token_accuracy": 0.16589969992637635,
|
|
"num_tokens": 6554761.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.786956310272217,
|
|
"epoch": 2.9826014913007457,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00044142129760967643,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.15644074678421022,
|
|
"num_tokens": 6564039.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.83938570022583,
|
|
"epoch": 2.986743993371997,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004412060712194115,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.15914641916751862,
|
|
"num_tokens": 6572559.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.759703350067139,
|
|
"epoch": 2.990886495443248,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004409905095733381,
|
|
"loss": 5.5356,
|
|
"mean_token_accuracy": 0.16635193228721618,
|
|
"num_tokens": 6581621.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.760700654983521,
|
|
"epoch": 2.995028997514499,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044077461310626764,
|
|
"loss": 5.4775,
|
|
"mean_token_accuracy": 0.16566378027200698,
|
|
"num_tokens": 6590353.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.788886833190918,
|
|
"epoch": 2.9991714995857497,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000440558382253687,
|
|
"loss": 5.5022,
|
|
"mean_token_accuracy": 0.16478997617959976,
|
|
"num_tokens": 6598822.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.766598749160766,
|
|
"epoch": 3.003314001657001,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004403418174517573,
|
|
"loss": 5.2733,
|
|
"mean_token_accuracy": 0.17376732379198073,
|
|
"num_tokens": 6606728.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.812176895141602,
|
|
"epoch": 3.007456503728252,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044012491913731353,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.18124154359102249,
|
|
"num_tokens": 6616226.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.710838270187378,
|
|
"epoch": 3.011599005799503,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004399076877478632,
|
|
"loss": 5.075,
|
|
"mean_token_accuracy": 0.19328108429908752,
|
|
"num_tokens": 6624801.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.790155744552612,
|
|
"epoch": 3.015741507870754,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004396901237215858,
|
|
"loss": 5.269,
|
|
"mean_token_accuracy": 0.17623631209135054,
|
|
"num_tokens": 6633043.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.765642833709717,
|
|
"epoch": 3.019884009942005,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043947222749733173,
|
|
"loss": 5.2102,
|
|
"mean_token_accuracy": 0.18271743953227998,
|
|
"num_tokens": 6641197.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.805044937133789,
|
|
"epoch": 3.024026512013256,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043925399951462153,
|
|
"loss": 5.2399,
|
|
"mean_token_accuracy": 0.18738753497600555,
|
|
"num_tokens": 6650411.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.6769920825958256,
|
|
"epoch": 3.028169014084507,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004390354402136449,
|
|
"loss": 5.1248,
|
|
"mean_token_accuracy": 0.18881589770317078,
|
|
"num_tokens": 6659405.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.736174297332764,
|
|
"epoch": 3.032311516155758,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004388165500352597,
|
|
"loss": 5.2136,
|
|
"mean_token_accuracy": 0.18298795074224472,
|
|
"num_tokens": 6667701.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.7536540031433105,
|
|
"epoch": 3.036454018227009,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004385973294209915,
|
|
"loss": 5.2331,
|
|
"mean_token_accuracy": 0.18401437103748322,
|
|
"num_tokens": 6676801.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.681411933898926,
|
|
"epoch": 3.0405965202982603,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004383777788130321,
|
|
"loss": 5.1045,
|
|
"mean_token_accuracy": 0.18881960958242416,
|
|
"num_tokens": 6686005.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.664102697372437,
|
|
"epoch": 3.044739022369511,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004381578986542393,
|
|
"loss": 5.1442,
|
|
"mean_token_accuracy": 0.18728526830673217,
|
|
"num_tokens": 6694884.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.671980381011963,
|
|
"epoch": 3.048881524440762,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004379376893881352,
|
|
"loss": 5.2569,
|
|
"mean_token_accuracy": 0.18485986441373825,
|
|
"num_tokens": 6703476.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.757300901412964,
|
|
"epoch": 3.053024026512013,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004377171514589059,
|
|
"loss": 5.276,
|
|
"mean_token_accuracy": 0.18426604121923446,
|
|
"num_tokens": 6712576.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.629332256317139,
|
|
"epoch": 3.0571665285832643,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004374962853114007,
|
|
"loss": 5.0989,
|
|
"mean_token_accuracy": 0.18679540753364562,
|
|
"num_tokens": 6721532.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.704257583618164,
|
|
"epoch": 3.0613090306545154,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004372750913911306,
|
|
"loss": 5.1983,
|
|
"mean_token_accuracy": 0.18960632532835006,
|
|
"num_tokens": 6730055.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.631225967407227,
|
|
"epoch": 3.0654515327257665,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004370535701442679,
|
|
"loss": 5.1999,
|
|
"mean_token_accuracy": 0.1866339772939682,
|
|
"num_tokens": 6739310.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.770854473114014,
|
|
"epoch": 3.069594034797017,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043683172201764537,
|
|
"loss": 5.2841,
|
|
"mean_token_accuracy": 0.17963071763515473,
|
|
"num_tokens": 6748976.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.659300661087036,
|
|
"epoch": 3.0737365368682683,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004366095474587547,
|
|
"loss": 5.1374,
|
|
"mean_token_accuracy": 0.18832063376903535,
|
|
"num_tokens": 6758043.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.707594966888427,
|
|
"epoch": 3.0778790389395194,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043638704691574646,
|
|
"loss": 5.2043,
|
|
"mean_token_accuracy": 0.18486624360084533,
|
|
"num_tokens": 6767554.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.746665573120117,
|
|
"epoch": 3.0820215410107705,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004361642208374283,
|
|
"loss": 5.2268,
|
|
"mean_token_accuracy": 0.17804842293262482,
|
|
"num_tokens": 6776161.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.666239452362061,
|
|
"epoch": 3.0861640430820216,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043594106967326495,
|
|
"loss": 5.1496,
|
|
"mean_token_accuracy": 0.18773341476917266,
|
|
"num_tokens": 6785371.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.761974048614502,
|
|
"epoch": 3.0903065451532727,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043571759387337665,
|
|
"loss": 5.1761,
|
|
"mean_token_accuracy": 0.18921129703521727,
|
|
"num_tokens": 6795242.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.707759094238281,
|
|
"epoch": 3.0944490472245234,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004354937938885385,
|
|
"loss": 5.2582,
|
|
"mean_token_accuracy": 0.18611662089824677,
|
|
"num_tokens": 6805423.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.70939302444458,
|
|
"epoch": 3.0985915492957745,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004352696701701796,
|
|
"loss": 5.1808,
|
|
"mean_token_accuracy": 0.18869266510009766,
|
|
"num_tokens": 6814377.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.685007333755493,
|
|
"epoch": 3.1027340513670256,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004350452231703819,
|
|
"loss": 5.2471,
|
|
"mean_token_accuracy": 0.18337070047855378,
|
|
"num_tokens": 6824025.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.6758904457092285,
|
|
"epoch": 3.1068765534382767,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043482045334187954,
|
|
"loss": 5.1716,
|
|
"mean_token_accuracy": 0.18803014159202575,
|
|
"num_tokens": 6833071.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.586541986465454,
|
|
"epoch": 3.111019055509528,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043459536113805785,
|
|
"loss": 5.1525,
|
|
"mean_token_accuracy": 0.1817285716533661,
|
|
"num_tokens": 6841983.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.664303827285766,
|
|
"epoch": 3.115161557580779,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043436994701295237,
|
|
"loss": 5.1358,
|
|
"mean_token_accuracy": 0.1896913841366768,
|
|
"num_tokens": 6851436.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.661314010620117,
|
|
"epoch": 3.1193040596520296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043414421142124804,
|
|
"loss": 5.221,
|
|
"mean_token_accuracy": 0.17457636296749116,
|
|
"num_tokens": 6861810.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.687656593322754,
|
|
"epoch": 3.1234465617232807,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043391815481827814,
|
|
"loss": 5.1126,
|
|
"mean_token_accuracy": 0.19503262490034104,
|
|
"num_tokens": 6870418.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.569714593887329,
|
|
"epoch": 3.127589063794532,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004336917776600237,
|
|
"loss": 5.1071,
|
|
"mean_token_accuracy": 0.1894732654094696,
|
|
"num_tokens": 6879419.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.6053571701049805,
|
|
"epoch": 3.131731565865783,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004334650804031122,
|
|
"loss": 5.2332,
|
|
"mean_token_accuracy": 0.18138375431299208,
|
|
"num_tokens": 6888768.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.720597743988037,
|
|
"epoch": 3.135874067937034,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004332380635048165,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.18394508212804794,
|
|
"num_tokens": 6897133.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.733305883407593,
|
|
"epoch": 3.140016570008285,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043301072742305474,
|
|
"loss": 5.2058,
|
|
"mean_token_accuracy": 0.18487854897975922,
|
|
"num_tokens": 6904976.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.676326942443848,
|
|
"epoch": 3.1441590720795363,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004327830726163886,
|
|
"loss": 5.2272,
|
|
"mean_token_accuracy": 0.18749673962593078,
|
|
"num_tokens": 6915138.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.7293699264526365,
|
|
"epoch": 3.148301574150787,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004325550995440226,
|
|
"loss": 5.1886,
|
|
"mean_token_accuracy": 0.18212902545928955,
|
|
"num_tokens": 6923962.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.540950870513916,
|
|
"epoch": 3.152444076222038,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043232680866580356,
|
|
"loss": 5.1693,
|
|
"mean_token_accuracy": 0.18463869094848634,
|
|
"num_tokens": 6932102.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.671437835693359,
|
|
"epoch": 3.156586578293289,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00043209820044221893,
|
|
"loss": 5.2108,
|
|
"mean_token_accuracy": 0.1867389351129532,
|
|
"num_tokens": 6940354.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.669532251358032,
|
|
"epoch": 3.1607290803645403,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043186927533439654,
|
|
"loss": 5.2209,
|
|
"mean_token_accuracy": 0.17929926812648772,
|
|
"num_tokens": 6949779.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.762870788574219,
|
|
"epoch": 3.1648715824357914,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043164003380410335,
|
|
"loss": 5.2644,
|
|
"mean_token_accuracy": 0.1840151846408844,
|
|
"num_tokens": 6958760.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.666829538345337,
|
|
"epoch": 3.169014084507042,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004314104763137445,
|
|
"loss": 5.1239,
|
|
"mean_token_accuracy": 0.1888977989554405,
|
|
"num_tokens": 6967535.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.6747228622436525,
|
|
"epoch": 3.173156586578293,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043118060332636264,
|
|
"loss": 5.2419,
|
|
"mean_token_accuracy": 0.18832369148731232,
|
|
"num_tokens": 6976211.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.735997343063355,
|
|
"epoch": 3.1772990886495442,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043095041530563665,
|
|
"loss": 5.257,
|
|
"mean_token_accuracy": 0.18535635769367217,
|
|
"num_tokens": 6986001.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.660528087615967,
|
|
"epoch": 3.1814415907207954,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043071991271588104,
|
|
"loss": 5.2432,
|
|
"mean_token_accuracy": 0.19192676693201066,
|
|
"num_tokens": 6994568.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 5.719581174850464,
|
|
"epoch": 3.1855840927920465,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004304890960220446,
|
|
"loss": 5.3453,
|
|
"mean_token_accuracy": 0.17908592373132706,
|
|
"num_tokens": 7004744.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.704136514663697,
|
|
"epoch": 3.1897265948632976,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043025796568970975,
|
|
"loss": 5.2962,
|
|
"mean_token_accuracy": 0.1782546654343605,
|
|
"num_tokens": 7014315.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.721152257919312,
|
|
"epoch": 3.1938690969345487,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043002652218509184,
|
|
"loss": 5.1558,
|
|
"mean_token_accuracy": 0.18557834774255752,
|
|
"num_tokens": 7022258.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.681660079956055,
|
|
"epoch": 3.1980115990057993,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004297947659750376,
|
|
"loss": 5.298,
|
|
"mean_token_accuracy": 0.17832167595624923,
|
|
"num_tokens": 7031732.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.690945100784302,
|
|
"epoch": 3.2021541010770505,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004295626975270246,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.18352857530117034,
|
|
"num_tokens": 7040478.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.71664719581604,
|
|
"epoch": 3.2062966031483016,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042933031730916024,
|
|
"loss": 5.1764,
|
|
"mean_token_accuracy": 0.1858328476548195,
|
|
"num_tokens": 7049360.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.701043748855591,
|
|
"epoch": 3.2104391052195527,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000429097625790181,
|
|
"loss": 5.266,
|
|
"mean_token_accuracy": 0.1839570701122284,
|
|
"num_tokens": 7059271.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.62371506690979,
|
|
"epoch": 3.2145816072908038,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004288646234394508,
|
|
"loss": 5.2333,
|
|
"mean_token_accuracy": 0.18282438665628434,
|
|
"num_tokens": 7069038.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.714450120925903,
|
|
"epoch": 3.218724109362055,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042863131072696115,
|
|
"loss": 5.1774,
|
|
"mean_token_accuracy": 0.1850357949733734,
|
|
"num_tokens": 7077971.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.668226528167724,
|
|
"epoch": 3.2228666114333056,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042839768812332905,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.18207817077636718,
|
|
"num_tokens": 7086821.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.643896484375,
|
|
"epoch": 3.2270091135045567,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000428163756099797,
|
|
"loss": 5.2174,
|
|
"mean_token_accuracy": 0.18410703390836716,
|
|
"num_tokens": 7095407.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.59422926902771,
|
|
"epoch": 3.2311516155758078,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004279295151282314,
|
|
"loss": 5.1473,
|
|
"mean_token_accuracy": 0.19247130453586578,
|
|
"num_tokens": 7103964.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.716597986221314,
|
|
"epoch": 3.235294117647059,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004276949656811217,
|
|
"loss": 5.2694,
|
|
"mean_token_accuracy": 0.18162517994642258,
|
|
"num_tokens": 7113571.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.735844564437866,
|
|
"epoch": 3.23943661971831,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042746010823158006,
|
|
"loss": 5.2946,
|
|
"mean_token_accuracy": 0.17452100813388824,
|
|
"num_tokens": 7122844.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.633665657043457,
|
|
"epoch": 3.243579121789561,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042722494325333945,
|
|
"loss": 5.1803,
|
|
"mean_token_accuracy": 0.1806490018963814,
|
|
"num_tokens": 7132123.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.678092288970947,
|
|
"epoch": 3.2477216238608118,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004269894712207533,
|
|
"loss": 5.1969,
|
|
"mean_token_accuracy": 0.18957881033420562,
|
|
"num_tokens": 7141068.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.6144341945648195,
|
|
"epoch": 3.251864125932063,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004267536926087947,
|
|
"loss": 5.2075,
|
|
"mean_token_accuracy": 0.17903397977352142,
|
|
"num_tokens": 7150829.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.648570394515991,
|
|
"epoch": 3.256006628003314,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004265176078930545,
|
|
"loss": 5.2198,
|
|
"mean_token_accuracy": 0.18345784097909928,
|
|
"num_tokens": 7160814.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.726291084289551,
|
|
"epoch": 3.260149130074565,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00042628121754974166,
|
|
"loss": 5.2979,
|
|
"mean_token_accuracy": 0.17981989234685897,
|
|
"num_tokens": 7170085.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.748935651779175,
|
|
"epoch": 3.264291632145816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004260445220556811,
|
|
"loss": 5.2492,
|
|
"mean_token_accuracy": 0.1845031350851059,
|
|
"num_tokens": 7179366.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.692015218734741,
|
|
"epoch": 3.2684341342170673,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042580752188831376,
|
|
"loss": 5.3121,
|
|
"mean_token_accuracy": 0.17713498771190644,
|
|
"num_tokens": 7188342.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.772071313858032,
|
|
"epoch": 3.272576636288318,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004255702175256946,
|
|
"loss": 5.3298,
|
|
"mean_token_accuracy": 0.1851222664117813,
|
|
"num_tokens": 7197466.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.664972019195557,
|
|
"epoch": 3.276719138359569,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042533260944649273,
|
|
"loss": 5.191,
|
|
"mean_token_accuracy": 0.18613426834344865,
|
|
"num_tokens": 7205462.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.668264102935791,
|
|
"epoch": 3.28086164043082,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004250946981299893,
|
|
"loss": 5.2157,
|
|
"mean_token_accuracy": 0.19075654298067093,
|
|
"num_tokens": 7215115.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.68201322555542,
|
|
"epoch": 3.2850041425020713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042485648405607776,
|
|
"loss": 5.167,
|
|
"mean_token_accuracy": 0.18031159937381744,
|
|
"num_tokens": 7224909.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.544072580337525,
|
|
"epoch": 3.2891466445733224,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004246179677052617,
|
|
"loss": 5.0989,
|
|
"mean_token_accuracy": 0.19356471449136733,
|
|
"num_tokens": 7232946.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.621029090881348,
|
|
"epoch": 3.2932891466445735,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004243791495586547,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.19010450690984726,
|
|
"num_tokens": 7241104.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.761633396148682,
|
|
"epoch": 3.297431648715824,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004241400300979791,
|
|
"loss": 5.2672,
|
|
"mean_token_accuracy": 0.18169665187597275,
|
|
"num_tokens": 7250749.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 5.651895189285279,
|
|
"epoch": 3.3015741507870753,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004239006098055651,
|
|
"loss": 5.1474,
|
|
"mean_token_accuracy": 0.18569830507040025,
|
|
"num_tokens": 7258687.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.56197509765625,
|
|
"epoch": 3.3057166528583264,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004236608891643494,
|
|
"loss": 5.1985,
|
|
"mean_token_accuracy": 0.1832697108387947,
|
|
"num_tokens": 7267316.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.618909740447998,
|
|
"epoch": 3.3098591549295775,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004234208686578748,
|
|
"loss": 5.2157,
|
|
"mean_token_accuracy": 0.1854364439845085,
|
|
"num_tokens": 7276734.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.66032657623291,
|
|
"epoch": 3.3140016570008286,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042318054877028876,
|
|
"loss": 5.1969,
|
|
"mean_token_accuracy": 0.18412411212921143,
|
|
"num_tokens": 7285916.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3140016570008286,
|
|
"eval_entropy": 5.480584607195499,
|
|
"eval_loss": 5.844878196716309,
|
|
"eval_mean_token_accuracy": 0.15718900104130762,
|
|
"eval_num_tokens": 7285916.0,
|
|
"eval_runtime": 2.122,
|
|
"eval_samples_per_second": 1765.792,
|
|
"eval_steps_per_second": 221.019,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.7113525390625,
|
|
"epoch": 3.3181441590720797,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042293992998634314,
|
|
"loss": 5.1827,
|
|
"mean_token_accuracy": 0.19378558099269866,
|
|
"num_tokens": 7294184.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.622339963912964,
|
|
"epoch": 3.3222866611433304,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004226990127913918,
|
|
"loss": 5.2039,
|
|
"mean_token_accuracy": 0.18599776178598404,
|
|
"num_tokens": 7302647.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.648343420028686,
|
|
"epoch": 3.3264291632145815,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004224577976713915,
|
|
"loss": 5.2959,
|
|
"mean_token_accuracy": 0.17115701138973236,
|
|
"num_tokens": 7311455.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.760056209564209,
|
|
"epoch": 3.3305716652858326,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00042221628511289927,
|
|
"loss": 5.3463,
|
|
"mean_token_accuracy": 0.16830217987298965,
|
|
"num_tokens": 7320978.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.823963212966919,
|
|
"epoch": 3.3347141673570837,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004219744756030723,
|
|
"loss": 5.3529,
|
|
"mean_token_accuracy": 0.1746882513165474,
|
|
"num_tokens": 7330202.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.610805368423462,
|
|
"epoch": 3.338856669428335,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042173236962966706,
|
|
"loss": 5.1985,
|
|
"mean_token_accuracy": 0.18864554166793823,
|
|
"num_tokens": 7339121.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.660786533355713,
|
|
"epoch": 3.342999171499586,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004214899676810375,
|
|
"loss": 5.2694,
|
|
"mean_token_accuracy": 0.17870924919843673,
|
|
"num_tokens": 7348631.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.669182443618775,
|
|
"epoch": 3.347141673570837,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004212472702461348,
|
|
"loss": 5.2381,
|
|
"mean_token_accuracy": 0.1864793688058853,
|
|
"num_tokens": 7358354.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.617064189910889,
|
|
"epoch": 3.3512841756420877,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004210042778145064,
|
|
"loss": 5.2278,
|
|
"mean_token_accuracy": 0.1866077035665512,
|
|
"num_tokens": 7367031.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.653417682647705,
|
|
"epoch": 3.355426677713339,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00042076099087629443,
|
|
"loss": 5.2353,
|
|
"mean_token_accuracy": 0.18688128292560577,
|
|
"num_tokens": 7375996.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.769751787185669,
|
|
"epoch": 3.35956917978459,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042051740992223534,
|
|
"loss": 5.3014,
|
|
"mean_token_accuracy": 0.17178498655557634,
|
|
"num_tokens": 7385355.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.578042650222779,
|
|
"epoch": 3.363711681855841,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004202735354436586,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.19076684415340422,
|
|
"num_tokens": 7394384.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.68038010597229,
|
|
"epoch": 3.367854183927092,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042002936793248546,
|
|
"loss": 5.272,
|
|
"mean_token_accuracy": 0.18589885234832765,
|
|
"num_tokens": 7403141.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.672816896438599,
|
|
"epoch": 3.371996685998343,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004197849078812286,
|
|
"loss": 5.2358,
|
|
"mean_token_accuracy": 0.18239964097738265,
|
|
"num_tokens": 7413331.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.675556421279907,
|
|
"epoch": 3.376139188069594,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004195401557829907,
|
|
"loss": 5.278,
|
|
"mean_token_accuracy": 0.17857578694820403,
|
|
"num_tokens": 7422362.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.621066522598267,
|
|
"epoch": 3.380281690140845,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041929511213146343,
|
|
"loss": 5.1897,
|
|
"mean_token_accuracy": 0.18869503438472748,
|
|
"num_tokens": 7431350.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.682591724395752,
|
|
"epoch": 3.384424192212096,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004190497774209266,
|
|
"loss": 5.1908,
|
|
"mean_token_accuracy": 0.18529084771871568,
|
|
"num_tokens": 7440382.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.69628119468689,
|
|
"epoch": 3.3885666942833472,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041880415214624715,
|
|
"loss": 5.29,
|
|
"mean_token_accuracy": 0.18324529826641084,
|
|
"num_tokens": 7449357.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.678526210784912,
|
|
"epoch": 3.3927091963545983,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004185582368028781,
|
|
"loss": 5.1925,
|
|
"mean_token_accuracy": 0.18658399432897568,
|
|
"num_tokens": 7457544.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.6309898853302,
|
|
"epoch": 3.3968516984258494,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004183120318868577,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.19819343388080596,
|
|
"num_tokens": 7465915.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.638827323913574,
|
|
"epoch": 3.4009942004971,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00041806553789480787,
|
|
"loss": 5.1593,
|
|
"mean_token_accuracy": 0.18596962988376617,
|
|
"num_tokens": 7474551.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.659074878692627,
|
|
"epoch": 3.405136702568351,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041781875532393423,
|
|
"loss": 5.2461,
|
|
"mean_token_accuracy": 0.18592666387557982,
|
|
"num_tokens": 7483271.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.70736665725708,
|
|
"epoch": 3.4092792046396023,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00041757168467202404,
|
|
"loss": 5.2647,
|
|
"mean_token_accuracy": 0.1779997631907463,
|
|
"num_tokens": 7493179.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.64538722038269,
|
|
"epoch": 3.4134217067108534,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041732432643744576,
|
|
"loss": 5.2456,
|
|
"mean_token_accuracy": 0.18239352852106094,
|
|
"num_tokens": 7501858.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.722668647766113,
|
|
"epoch": 3.4175642087821045,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004170766811191481,
|
|
"loss": 5.3119,
|
|
"mean_token_accuracy": 0.17607627511024476,
|
|
"num_tokens": 7511817.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.67455530166626,
|
|
"epoch": 3.421706710853355,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004168287492166587,
|
|
"loss": 5.277,
|
|
"mean_token_accuracy": 0.18296863585710527,
|
|
"num_tokens": 7521655.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.6682655811309814,
|
|
"epoch": 3.4258492129246063,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041658053123008325,
|
|
"loss": 5.2358,
|
|
"mean_token_accuracy": 0.17903233766555787,
|
|
"num_tokens": 7531321.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.68986177444458,
|
|
"epoch": 3.4299917149958574,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004163320276601046,
|
|
"loss": 5.2495,
|
|
"mean_token_accuracy": 0.18417581021785737,
|
|
"num_tokens": 7540613.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.656633615493774,
|
|
"epoch": 3.4341342170671085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004160832390079816,
|
|
"loss": 5.305,
|
|
"mean_token_accuracy": 0.18491521924734117,
|
|
"num_tokens": 7549598.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.634925127029419,
|
|
"epoch": 3.4382767191383596,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041583416577554815,
|
|
"loss": 5.1918,
|
|
"mean_token_accuracy": 0.1878269463777542,
|
|
"num_tokens": 7558738.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.6589844703674315,
|
|
"epoch": 3.4424192212096107,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041558480846521213,
|
|
"loss": 5.2305,
|
|
"mean_token_accuracy": 0.1870835915207863,
|
|
"num_tokens": 7566873.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.583999967575073,
|
|
"epoch": 3.446561723280862,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041533516757995455,
|
|
"loss": 5.1972,
|
|
"mean_token_accuracy": 0.18694443106651307,
|
|
"num_tokens": 7576645.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.673832082748413,
|
|
"epoch": 3.4507042253521125,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004150852436233283,
|
|
"loss": 5.2077,
|
|
"mean_token_accuracy": 0.18114688992500305,
|
|
"num_tokens": 7584854.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.651709461212159,
|
|
"epoch": 3.4548467274233636,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004148350370994574,
|
|
"loss": 5.3285,
|
|
"mean_token_accuracy": 0.1787771075963974,
|
|
"num_tokens": 7594022.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.652410697937012,
|
|
"epoch": 3.4589892294946147,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004145845485130357,
|
|
"loss": 5.2018,
|
|
"mean_token_accuracy": 0.18653349131345748,
|
|
"num_tokens": 7603259.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.638375234603882,
|
|
"epoch": 3.463131731565866,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00041433377836932594,
|
|
"loss": 5.3126,
|
|
"mean_token_accuracy": 0.17587198615074157,
|
|
"num_tokens": 7613451.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.779592370986938,
|
|
"epoch": 3.467274233637117,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041408272717415913,
|
|
"loss": 5.2771,
|
|
"mean_token_accuracy": 0.17856950759887696,
|
|
"num_tokens": 7622218.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.590104722976685,
|
|
"epoch": 3.4714167357083676,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004138313954339327,
|
|
"loss": 5.1375,
|
|
"mean_token_accuracy": 0.18517837077379226,
|
|
"num_tokens": 7630508.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.607667016983032,
|
|
"epoch": 3.4755592377796187,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004135797836556106,
|
|
"loss": 5.2209,
|
|
"mean_token_accuracy": 0.18611387461423873,
|
|
"num_tokens": 7640141.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.665380144119263,
|
|
"epoch": 3.47970173985087,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041332789234672094,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.1789573237299919,
|
|
"num_tokens": 7648764.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.6518947124481205,
|
|
"epoch": 3.483844241922121,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004130757220153561,
|
|
"loss": 5.2294,
|
|
"mean_token_accuracy": 0.18545118123292922,
|
|
"num_tokens": 7658642.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.696086883544922,
|
|
"epoch": 3.487986743993372,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004128232731701712,
|
|
"loss": 5.3863,
|
|
"mean_token_accuracy": 0.17482188493013381,
|
|
"num_tokens": 7667524.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.686117458343506,
|
|
"epoch": 3.492129246064623,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004125705463203833,
|
|
"loss": 5.227,
|
|
"mean_token_accuracy": 0.18984778821468354,
|
|
"num_tokens": 7675800.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.660667371749878,
|
|
"epoch": 3.4962717481358743,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041231754197576986,
|
|
"loss": 5.2643,
|
|
"mean_token_accuracy": 0.18582551032304764,
|
|
"num_tokens": 7683742.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.574977111816406,
|
|
"epoch": 3.5004142502071254,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004120642606466685,
|
|
"loss": 5.2475,
|
|
"mean_token_accuracy": 0.18301280289888383,
|
|
"num_tokens": 7692379.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.737248182296753,
|
|
"epoch": 3.504556752278376,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004118107028439751,
|
|
"loss": 5.2413,
|
|
"mean_token_accuracy": 0.1811201259493828,
|
|
"num_tokens": 7702156.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.683750247955322,
|
|
"epoch": 3.508699254349627,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041155686907914365,
|
|
"loss": 5.2398,
|
|
"mean_token_accuracy": 0.181915944814682,
|
|
"num_tokens": 7712135.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.589407920837402,
|
|
"epoch": 3.5128417564208783,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004113027598641845,
|
|
"loss": 5.2116,
|
|
"mean_token_accuracy": 0.18613894730806352,
|
|
"num_tokens": 7721347.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.676157474517822,
|
|
"epoch": 3.5169842584921294,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004110483757116637,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.18491633534431456,
|
|
"num_tokens": 7731351.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.6546203136444095,
|
|
"epoch": 3.52112676056338,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00041079371713470195,
|
|
"loss": 5.1363,
|
|
"mean_token_accuracy": 0.19584661722183228,
|
|
"num_tokens": 7740413.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.587917232513428,
|
|
"epoch": 3.525269262634631,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041053878464697343,
|
|
"loss": 5.2268,
|
|
"mean_token_accuracy": 0.18845798522233964,
|
|
"num_tokens": 7749319.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.67224760055542,
|
|
"epoch": 3.5294117647058822,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004102835787627047,
|
|
"loss": 5.2755,
|
|
"mean_token_accuracy": 0.17681072950363158,
|
|
"num_tokens": 7757936.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.743300294876098,
|
|
"epoch": 3.5335542667771334,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004100280999966741,
|
|
"loss": 5.3364,
|
|
"mean_token_accuracy": 0.18500278890132904,
|
|
"num_tokens": 7767349.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.596862173080444,
|
|
"epoch": 3.5376967688483845,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040977234886421007,
|
|
"loss": 5.1507,
|
|
"mean_token_accuracy": 0.18324362635612487,
|
|
"num_tokens": 7776448.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.578992700576782,
|
|
"epoch": 3.5418392709196356,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040951632588119064,
|
|
"loss": 5.2281,
|
|
"mean_token_accuracy": 0.18678826540708543,
|
|
"num_tokens": 7785768.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.78763222694397,
|
|
"epoch": 3.5459817729908867,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004092600315640422,
|
|
"loss": 5.3267,
|
|
"mean_token_accuracy": 0.18010974675416946,
|
|
"num_tokens": 7794803.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.623131370544433,
|
|
"epoch": 3.550124275062138,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004090034664297382,
|
|
"loss": 5.163,
|
|
"mean_token_accuracy": 0.18634875416755675,
|
|
"num_tokens": 7803533.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.609513854980468,
|
|
"epoch": 3.5542667771333885,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040874663099579883,
|
|
"loss": 5.2574,
|
|
"mean_token_accuracy": 0.18191962838172912,
|
|
"num_tokens": 7812854.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.675520229339599,
|
|
"epoch": 3.5584092792046396,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040848952578028915,
|
|
"loss": 5.2222,
|
|
"mean_token_accuracy": 0.18348516374826432,
|
|
"num_tokens": 7821683.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.619959354400635,
|
|
"epoch": 3.5625517812758907,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004082321513018183,
|
|
"loss": 5.2052,
|
|
"mean_token_accuracy": 0.18874328285455705,
|
|
"num_tokens": 7830967.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.664625263214111,
|
|
"epoch": 3.566694283347142,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040797450807953897,
|
|
"loss": 5.2697,
|
|
"mean_token_accuracy": 0.17909055948257446,
|
|
"num_tokens": 7841179.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.687614870071411,
|
|
"epoch": 3.5708367854183924,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004077165966331454,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.19027931541204451,
|
|
"num_tokens": 7849841.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.514700508117675,
|
|
"epoch": 3.5749792874896436,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004074584174828735,
|
|
"loss": 5.2316,
|
|
"mean_token_accuracy": 0.1888791725039482,
|
|
"num_tokens": 7859390.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.630543899536133,
|
|
"epoch": 3.5791217895608947,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040719997114949856,
|
|
"loss": 5.2142,
|
|
"mean_token_accuracy": 0.18585167080163956,
|
|
"num_tokens": 7868848.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.661330223083496,
|
|
"epoch": 3.5832642916321458,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040694125815433533,
|
|
"loss": 5.2638,
|
|
"mean_token_accuracy": 0.1847008377313614,
|
|
"num_tokens": 7877521.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.698994159698486,
|
|
"epoch": 3.587406793703397,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004066822790192359,
|
|
"loss": 5.2392,
|
|
"mean_token_accuracy": 0.17902445048093796,
|
|
"num_tokens": 7886355.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.634966850280762,
|
|
"epoch": 3.591549295774648,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040642303426658967,
|
|
"loss": 5.28,
|
|
"mean_token_accuracy": 0.1777811899781227,
|
|
"num_tokens": 7895028.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.580403518676758,
|
|
"epoch": 3.595691797845899,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040616352441932163,
|
|
"loss": 5.1559,
|
|
"mean_token_accuracy": 0.18351313918828965,
|
|
"num_tokens": 7904192.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.600250864028931,
|
|
"epoch": 3.59983429991715,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00040590375000089146,
|
|
"loss": 5.2365,
|
|
"mean_token_accuracy": 0.1771611213684082,
|
|
"num_tokens": 7912328.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.6892577648162845,
|
|
"epoch": 3.603976801988401,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00040564371153529256,
|
|
"loss": 5.2749,
|
|
"mean_token_accuracy": 0.1780044376850128,
|
|
"num_tokens": 7922594.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.605875635147095,
|
|
"epoch": 3.608119304059652,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004053834095470509,
|
|
"loss": 5.2497,
|
|
"mean_token_accuracy": 0.16639523804187775,
|
|
"num_tokens": 7932345.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.67349944114685,
|
|
"epoch": 3.612261806130903,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040512284456122407,
|
|
"loss": 5.2285,
|
|
"mean_token_accuracy": 0.18600521683692933,
|
|
"num_tokens": 7942800.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.601762056350708,
|
|
"epoch": 3.616404308202154,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040486201710340016,
|
|
"loss": 5.1506,
|
|
"mean_token_accuracy": 0.1878705933690071,
|
|
"num_tokens": 7951094.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.613385343551636,
|
|
"epoch": 3.6205468102734053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040460092769969654,
|
|
"loss": 5.227,
|
|
"mean_token_accuracy": 0.17774293273687364,
|
|
"num_tokens": 7960652.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.734005498886108,
|
|
"epoch": 3.624689312344656,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004043395768767592,
|
|
"loss": 5.3036,
|
|
"mean_token_accuracy": 0.176109878718853,
|
|
"num_tokens": 7969350.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.672399997711182,
|
|
"epoch": 3.628831814415907,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004040779651617612,
|
|
"loss": 5.21,
|
|
"mean_token_accuracy": 0.1864970713853836,
|
|
"num_tokens": 7978969.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.565951108932495,
|
|
"epoch": 3.632974316487158,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040381609308240207,
|
|
"loss": 5.2501,
|
|
"mean_token_accuracy": 0.18693784326314927,
|
|
"num_tokens": 7988531.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.645549488067627,
|
|
"epoch": 3.6371168185584093,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040355396116690634,
|
|
"loss": 5.2108,
|
|
"mean_token_accuracy": 0.17983887642621993,
|
|
"num_tokens": 7996576.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.650477170944214,
|
|
"epoch": 3.6412593206296604,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004032915699440227,
|
|
"loss": 5.2751,
|
|
"mean_token_accuracy": 0.1771684393286705,
|
|
"num_tokens": 8006279.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.658308124542236,
|
|
"epoch": 3.6454018227009115,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000403028919943023,
|
|
"loss": 5.226,
|
|
"mean_token_accuracy": 0.18415955156087876,
|
|
"num_tokens": 8015476.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.686865234375,
|
|
"epoch": 3.6495443247721626,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004027660116937009,
|
|
"loss": 5.252,
|
|
"mean_token_accuracy": 0.1796737015247345,
|
|
"num_tokens": 8025450.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.609664487838745,
|
|
"epoch": 3.6536868268434133,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004025028457263712,
|
|
"loss": 5.2591,
|
|
"mean_token_accuracy": 0.18740882724523544,
|
|
"num_tokens": 8034850.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.588315343856811,
|
|
"epoch": 3.6578293289146644,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004022394225718684,
|
|
"loss": 5.1971,
|
|
"mean_token_accuracy": 0.18252264261245726,
|
|
"num_tokens": 8044526.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.598998069763184,
|
|
"epoch": 3.6619718309859155,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004019757427615456,
|
|
"loss": 5.1502,
|
|
"mean_token_accuracy": 0.1838933378458023,
|
|
"num_tokens": 8054033.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.6652288913726805,
|
|
"epoch": 3.6661143330571666,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004017118068272739,
|
|
"loss": 5.2649,
|
|
"mean_token_accuracy": 0.18519273847341539,
|
|
"num_tokens": 8063535.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.6589026927948,
|
|
"epoch": 3.6702568351284177,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040144761530144096,
|
|
"loss": 5.2475,
|
|
"mean_token_accuracy": 0.190778848528862,
|
|
"num_tokens": 8073079.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.590589570999145,
|
|
"epoch": 3.6743993371996684,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040118316871695,
|
|
"loss": 5.1685,
|
|
"mean_token_accuracy": 0.19072185605764388,
|
|
"num_tokens": 8082925.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.6333975315094,
|
|
"epoch": 3.6785418392709195,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040091846760721857,
|
|
"loss": 5.2783,
|
|
"mean_token_accuracy": 0.1795778527855873,
|
|
"num_tokens": 8092038.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.633273315429688,
|
|
"epoch": 3.6826843413421706,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040065351250617774,
|
|
"loss": 5.2441,
|
|
"mean_token_accuracy": 0.17714082300662995,
|
|
"num_tokens": 8101581.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.616935348510742,
|
|
"epoch": 3.6868268434134217,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040038830394827107,
|
|
"loss": 5.3074,
|
|
"mean_token_accuracy": 0.17179063707590103,
|
|
"num_tokens": 8111392.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.696556234359742,
|
|
"epoch": 3.690969345484673,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000400122842468453,
|
|
"loss": 5.3128,
|
|
"mean_token_accuracy": 0.17595641911029816,
|
|
"num_tokens": 8120941.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.62767481803894,
|
|
"epoch": 3.695111847555924,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003998571286021884,
|
|
"loss": 5.2114,
|
|
"mean_token_accuracy": 0.1770459607243538,
|
|
"num_tokens": 8129983.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.6886271953582765,
|
|
"epoch": 3.699254349627175,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003995911628854512,
|
|
"loss": 5.3284,
|
|
"mean_token_accuracy": 0.18379758894443513,
|
|
"num_tokens": 8139428.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.66134614944458,
|
|
"epoch": 3.7033968516984257,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003993249458547234,
|
|
"loss": 5.2626,
|
|
"mean_token_accuracy": 0.17668695300817489,
|
|
"num_tokens": 8148520.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.559238624572754,
|
|
"epoch": 3.707539353769677,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039905847804699363,
|
|
"loss": 5.1047,
|
|
"mean_token_accuracy": 0.19523490369319915,
|
|
"num_tokens": 8156908.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.597048902511597,
|
|
"epoch": 3.711681855840928,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039879175999975677,
|
|
"loss": 5.2233,
|
|
"mean_token_accuracy": 0.18122218549251556,
|
|
"num_tokens": 8165677.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.650084352493286,
|
|
"epoch": 3.715824357912179,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039852479225101217,
|
|
"loss": 5.2066,
|
|
"mean_token_accuracy": 0.18009962886571884,
|
|
"num_tokens": 8174786.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.695609188079834,
|
|
"epoch": 3.71996685998343,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039825757533926305,
|
|
"loss": 5.2716,
|
|
"mean_token_accuracy": 0.17197834104299545,
|
|
"num_tokens": 8184150.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.5949602127075195,
|
|
"epoch": 3.724109362054681,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039799010980351494,
|
|
"loss": 5.1866,
|
|
"mean_token_accuracy": 0.19903383255004883,
|
|
"num_tokens": 8192257.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.638699245452881,
|
|
"epoch": 3.728251864125932,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039772239618327535,
|
|
"loss": 5.2503,
|
|
"mean_token_accuracy": 0.18292991518974305,
|
|
"num_tokens": 8201041.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.728251864125932,
|
|
"eval_entropy": 5.3940906728000275,
|
|
"eval_loss": 5.802577018737793,
|
|
"eval_mean_token_accuracy": 0.160384202769189,
|
|
"eval_num_tokens": 8201041.0,
|
|
"eval_runtime": 2.1087,
|
|
"eval_samples_per_second": 1776.895,
|
|
"eval_steps_per_second": 222.408,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.626062822341919,
|
|
"epoch": 3.732394366197183,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039745443501855156,
|
|
"loss": 5.2187,
|
|
"mean_token_accuracy": 0.19095283299684523,
|
|
"num_tokens": 8209881.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.646627902984619,
|
|
"epoch": 3.736536868268434,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039718622684985075,
|
|
"loss": 5.2556,
|
|
"mean_token_accuracy": 0.18988119661808014,
|
|
"num_tokens": 8218504.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.606738662719726,
|
|
"epoch": 3.7406793703396852,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000396917772218178,
|
|
"loss": 5.0775,
|
|
"mean_token_accuracy": 0.19684855937957763,
|
|
"num_tokens": 8226785.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.677734756469727,
|
|
"epoch": 3.7448218724109363,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003966490716650357,
|
|
"loss": 5.2242,
|
|
"mean_token_accuracy": 0.18311616480350495,
|
|
"num_tokens": 8236092.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.610603761672974,
|
|
"epoch": 3.7489643744821874,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00039638012573242225,
|
|
"loss": 5.2148,
|
|
"mean_token_accuracy": 0.1902775928378105,
|
|
"num_tokens": 8245297.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.670913505554199,
|
|
"epoch": 3.7531068765534386,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003961109349628309,
|
|
"loss": 5.2875,
|
|
"mean_token_accuracy": 0.17370747476816178,
|
|
"num_tokens": 8254772.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.6657225608825685,
|
|
"epoch": 3.757249378624689,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000395841499899249,
|
|
"loss": 5.3296,
|
|
"mean_token_accuracy": 0.17300237864255905,
|
|
"num_tokens": 8264310.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.656255912780762,
|
|
"epoch": 3.7613918806959403,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039557182108515624,
|
|
"loss": 5.2646,
|
|
"mean_token_accuracy": 0.18701519668102265,
|
|
"num_tokens": 8274184.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 5.633713483810425,
|
|
"epoch": 3.7655343827671914,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039530189906452457,
|
|
"loss": 5.19,
|
|
"mean_token_accuracy": 0.1922944039106369,
|
|
"num_tokens": 8283151.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.662411165237427,
|
|
"epoch": 3.7696768848384425,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039503173438181605,
|
|
"loss": 5.2768,
|
|
"mean_token_accuracy": 0.1810198575258255,
|
|
"num_tokens": 8292574.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.561538934707642,
|
|
"epoch": 3.773819386909693,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039476132758198236,
|
|
"loss": 5.1895,
|
|
"mean_token_accuracy": 0.1845146358013153,
|
|
"num_tokens": 8301281.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.63593373298645,
|
|
"epoch": 3.7779618889809443,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039449067921046353,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.18710329681634902,
|
|
"num_tokens": 8310201.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.685689258575439,
|
|
"epoch": 3.7821043910521954,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039421978981318703,
|
|
"loss": 5.2757,
|
|
"mean_token_accuracy": 0.18318944573402404,
|
|
"num_tokens": 8320369.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.662524271011352,
|
|
"epoch": 3.7862468931234465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039394865993656625,
|
|
"loss": 5.2463,
|
|
"mean_token_accuracy": 0.1847994029521942,
|
|
"num_tokens": 8329519.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.594557428359986,
|
|
"epoch": 3.7903893951946976,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003936772901274999,
|
|
"loss": 5.2452,
|
|
"mean_token_accuracy": 0.17869513630867004,
|
|
"num_tokens": 8338282.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.63024377822876,
|
|
"epoch": 3.7945318972659487,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003934056809333705,
|
|
"loss": 5.2639,
|
|
"mean_token_accuracy": 0.18257157355546952,
|
|
"num_tokens": 8346861.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.7111217975616455,
|
|
"epoch": 3.7986743993372,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039313383290204333,
|
|
"loss": 5.2974,
|
|
"mean_token_accuracy": 0.18029544800519942,
|
|
"num_tokens": 8356078.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.6479004383087155,
|
|
"epoch": 3.802816901408451,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003928617465818658,
|
|
"loss": 5.2577,
|
|
"mean_token_accuracy": 0.17393672913312913,
|
|
"num_tokens": 8365487.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.648684549331665,
|
|
"epoch": 3.8069594034797016,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039258942252166554,
|
|
"loss": 5.2384,
|
|
"mean_token_accuracy": 0.18395584672689438,
|
|
"num_tokens": 8374622.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.645304155349732,
|
|
"epoch": 3.8111019055509527,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003923168612707502,
|
|
"loss": 5.173,
|
|
"mean_token_accuracy": 0.19261859506368637,
|
|
"num_tokens": 8383312.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.576239204406738,
|
|
"epoch": 3.815244407622204,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003920440633789053,
|
|
"loss": 5.2017,
|
|
"mean_token_accuracy": 0.17927682548761367,
|
|
"num_tokens": 8391987.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.587787914276123,
|
|
"epoch": 3.819386909693455,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00039177102939639424,
|
|
"loss": 5.2031,
|
|
"mean_token_accuracy": 0.17870101630687713,
|
|
"num_tokens": 8401263.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.718034076690674,
|
|
"epoch": 3.8235294117647056,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003914977598739563,
|
|
"loss": 5.2357,
|
|
"mean_token_accuracy": 0.19321168810129166,
|
|
"num_tokens": 8410110.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.698187637329101,
|
|
"epoch": 3.8276719138359567,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003912242553628059,
|
|
"loss": 5.3202,
|
|
"mean_token_accuracy": 0.17658483386039733,
|
|
"num_tokens": 8419321.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.68555326461792,
|
|
"epoch": 3.831814415907208,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003909505164146318,
|
|
"loss": 5.2886,
|
|
"mean_token_accuracy": 0.181088188290596,
|
|
"num_tokens": 8428376.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.642216205596924,
|
|
"epoch": 3.835956917978459,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000390676543581595,
|
|
"loss": 5.2141,
|
|
"mean_token_accuracy": 0.18256331235170364,
|
|
"num_tokens": 8437263.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.596312761306763,
|
|
"epoch": 3.84009942004971,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003904023374163289,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.19174929857254028,
|
|
"num_tokens": 8445889.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.696162319183349,
|
|
"epoch": 3.844241922120961,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003901278984719372,
|
|
"loss": 5.4047,
|
|
"mean_token_accuracy": 0.17020085826516151,
|
|
"num_tokens": 8456086.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.706778192520142,
|
|
"epoch": 3.8483844241922123,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038985322730199326,
|
|
"loss": 5.2302,
|
|
"mean_token_accuracy": 0.18645505011081695,
|
|
"num_tokens": 8465150.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.608067655563355,
|
|
"epoch": 3.8525269262634634,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003895783244605389,
|
|
"loss": 5.2835,
|
|
"mean_token_accuracy": 0.1789731577038765,
|
|
"num_tokens": 8474172.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.5726934432983395,
|
|
"epoch": 3.856669428334714,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038930319050208315,
|
|
"loss": 5.1738,
|
|
"mean_token_accuracy": 0.19118774682283401,
|
|
"num_tokens": 8483636.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.606748342514038,
|
|
"epoch": 3.860811930405965,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038902782598160127,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.19268114864826202,
|
|
"num_tokens": 8491560.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.693191862106323,
|
|
"epoch": 3.8649544324772163,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038875223145453346,
|
|
"loss": 5.4268,
|
|
"mean_token_accuracy": 0.18574169427156448,
|
|
"num_tokens": 8501685.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.741757440567016,
|
|
"epoch": 3.8690969345484674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003884764074767842,
|
|
"loss": 5.2715,
|
|
"mean_token_accuracy": 0.1783949464559555,
|
|
"num_tokens": 8510580.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.6376340866088865,
|
|
"epoch": 3.873239436619718,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003882003546047205,
|
|
"loss": 5.2305,
|
|
"mean_token_accuracy": 0.17803009897470473,
|
|
"num_tokens": 8519595.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.715295362472534,
|
|
"epoch": 3.877381938690969,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00038792407339517123,
|
|
"loss": 5.3209,
|
|
"mean_token_accuracy": 0.17905150651931762,
|
|
"num_tokens": 8529118.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.672022533416748,
|
|
"epoch": 3.8815244407622203,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038764756440542574,
|
|
"loss": 5.2627,
|
|
"mean_token_accuracy": 0.1794586092233658,
|
|
"num_tokens": 8539322.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.592021083831787,
|
|
"epoch": 3.8856669428334714,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038737082819323285,
|
|
"loss": 5.1975,
|
|
"mean_token_accuracy": 0.18603272438049318,
|
|
"num_tokens": 8548707.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.604137516021728,
|
|
"epoch": 3.8898094449047225,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038709386531679974,
|
|
"loss": 5.2883,
|
|
"mean_token_accuracy": 0.18267909586429595,
|
|
"num_tokens": 8557500.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.62609601020813,
|
|
"epoch": 3.8939519469759736,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038681667633479086,
|
|
"loss": 5.298,
|
|
"mean_token_accuracy": 0.17353631258010865,
|
|
"num_tokens": 8567465.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.576682090759277,
|
|
"epoch": 3.8980944490472247,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038653926180632665,
|
|
"loss": 5.1237,
|
|
"mean_token_accuracy": 0.19197191298007965,
|
|
"num_tokens": 8576680.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.689194202423096,
|
|
"epoch": 3.902236951118476,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038626162229098257,
|
|
"loss": 5.2908,
|
|
"mean_token_accuracy": 0.17073331773281097,
|
|
"num_tokens": 8585539.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.648666191101074,
|
|
"epoch": 3.9063794531897265,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003859837583487877,
|
|
"loss": 5.1412,
|
|
"mean_token_accuracy": 0.19408987760543822,
|
|
"num_tokens": 8593598.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.624051809310913,
|
|
"epoch": 3.9105219552609776,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003857056705402241,
|
|
"loss": 5.2655,
|
|
"mean_token_accuracy": 0.1789528176188469,
|
|
"num_tokens": 8603440.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.5498459815979,
|
|
"epoch": 3.9146644573322287,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003854273594262253,
|
|
"loss": 5.095,
|
|
"mean_token_accuracy": 0.19449471831321716,
|
|
"num_tokens": 8612490.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.56399450302124,
|
|
"epoch": 3.91880695940348,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003851488255681752,
|
|
"loss": 5.176,
|
|
"mean_token_accuracy": 0.18269762992858887,
|
|
"num_tokens": 8621994.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.603197956085205,
|
|
"epoch": 3.922949461474731,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000384870069527907,
|
|
"loss": 5.2652,
|
|
"mean_token_accuracy": 0.1824132487177849,
|
|
"num_tokens": 8631185.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.626424407958984,
|
|
"epoch": 3.9270919635459816,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038459109186770213,
|
|
"loss": 5.1436,
|
|
"mean_token_accuracy": 0.1938893973827362,
|
|
"num_tokens": 8639018.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.590229749679565,
|
|
"epoch": 3.9312344656172327,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038431189315028905,
|
|
"loss": 5.1562,
|
|
"mean_token_accuracy": 0.18676035702228547,
|
|
"num_tokens": 8647672.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.6201681137084964,
|
|
"epoch": 3.9353769676884838,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003840324739388419,
|
|
"loss": 5.2768,
|
|
"mean_token_accuracy": 0.17821766436100006,
|
|
"num_tokens": 8656638.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.629859209060669,
|
|
"epoch": 3.939519469759735,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00038375283479698,
|
|
"loss": 5.25,
|
|
"mean_token_accuracy": 0.18878189325332642,
|
|
"num_tokens": 8666167.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.720983791351318,
|
|
"epoch": 3.943661971830986,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003834729762887659,
|
|
"loss": 5.3304,
|
|
"mean_token_accuracy": 0.17489232867956161,
|
|
"num_tokens": 8676435.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.623570442199707,
|
|
"epoch": 3.947804473902237,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003831928989787049,
|
|
"loss": 5.2081,
|
|
"mean_token_accuracy": 0.1888226866722107,
|
|
"num_tokens": 8686011.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.6070746898651125,
|
|
"epoch": 3.951946975973488,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003829126034317435,
|
|
"loss": 5.2308,
|
|
"mean_token_accuracy": 0.18148026317358018,
|
|
"num_tokens": 8695063.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.665477609634399,
|
|
"epoch": 3.956089478044739,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003826320902132684,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.17441263496875764,
|
|
"num_tokens": 8704585.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.6900327682495115,
|
|
"epoch": 3.96023198011599,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038235135988910545,
|
|
"loss": 5.237,
|
|
"mean_token_accuracy": 0.1779341459274292,
|
|
"num_tokens": 8714097.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.633197546005249,
|
|
"epoch": 3.964374482187241,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038207041302551844,
|
|
"loss": 5.1669,
|
|
"mean_token_accuracy": 0.18947037011384965,
|
|
"num_tokens": 8722890.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.554238796234131,
|
|
"epoch": 3.968516984258492,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003817892501892078,
|
|
"loss": 5.1761,
|
|
"mean_token_accuracy": 0.1865229845046997,
|
|
"num_tokens": 8732046.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.572015953063965,
|
|
"epoch": 3.9726594863297433,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038150787194730977,
|
|
"loss": 5.2016,
|
|
"mean_token_accuracy": 0.18636801689863206,
|
|
"num_tokens": 8741750.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.630955505371094,
|
|
"epoch": 3.976801988400994,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000381226278867395,
|
|
"loss": 5.2052,
|
|
"mean_token_accuracy": 0.1785096377134323,
|
|
"num_tokens": 8750141.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.601733636856079,
|
|
"epoch": 3.980944490472245,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038094447151746746,
|
|
"loss": 5.1218,
|
|
"mean_token_accuracy": 0.19307401925325393,
|
|
"num_tokens": 8758567.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.635490083694458,
|
|
"epoch": 3.985086992543496,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003806624504659635,
|
|
"loss": 5.2125,
|
|
"mean_token_accuracy": 0.18852995336055756,
|
|
"num_tokens": 8767829.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.577333688735962,
|
|
"epoch": 3.9892294946147473,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003803802162817503,
|
|
"loss": 5.1161,
|
|
"mean_token_accuracy": 0.19472022205591202,
|
|
"num_tokens": 8776911.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.570671701431275,
|
|
"epoch": 3.9933719966859984,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000380097769534125,
|
|
"loss": 5.222,
|
|
"mean_token_accuracy": 0.18993626534938812,
|
|
"num_tokens": 8786117.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.598080682754516,
|
|
"epoch": 3.9975144987572495,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003798151107928136,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.19167835563421248,
|
|
"num_tokens": 8795647.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.545804214477539,
|
|
"epoch": 4.001657000828501,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037953224062796967,
|
|
"loss": 5.1417,
|
|
"mean_token_accuracy": 0.18513134568929673,
|
|
"num_tokens": 8803480.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.682935667037964,
|
|
"epoch": 4.005799502899752,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003792491596101733,
|
|
"loss": 4.9957,
|
|
"mean_token_accuracy": 0.2028227984905243,
|
|
"num_tokens": 8812825.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.684786796569824,
|
|
"epoch": 4.009942004971003,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003789658683104298,
|
|
"loss": 5.0615,
|
|
"mean_token_accuracy": 0.2027088701725006,
|
|
"num_tokens": 8821911.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.644535160064697,
|
|
"epoch": 4.014084507042254,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037868236730016866,
|
|
"loss": 4.9841,
|
|
"mean_token_accuracy": 0.20158157348632813,
|
|
"num_tokens": 8830691.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.604446601867676,
|
|
"epoch": 4.018227009113504,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037839865715124234,
|
|
"loss": 4.9726,
|
|
"mean_token_accuracy": 0.19711296558380126,
|
|
"num_tokens": 8840410.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.563369703292847,
|
|
"epoch": 4.022369511184755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037811473843592534,
|
|
"loss": 4.9613,
|
|
"mean_token_accuracy": 0.19836542755365372,
|
|
"num_tokens": 8849662.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.6024778366088865,
|
|
"epoch": 4.026512013256006,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037783061172691273,
|
|
"loss": 4.9569,
|
|
"mean_token_accuracy": 0.20249882638454436,
|
|
"num_tokens": 8858666.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.658640146255493,
|
|
"epoch": 4.0306545153272575,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003775462775973191,
|
|
"loss": 5.0009,
|
|
"mean_token_accuracy": 0.19936698526144028,
|
|
"num_tokens": 8867885.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.570419073104858,
|
|
"epoch": 4.034797017398509,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003772617366206773,
|
|
"loss": 4.9574,
|
|
"mean_token_accuracy": 0.2057440310716629,
|
|
"num_tokens": 8877458.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.516879367828369,
|
|
"epoch": 4.03893951946976,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037697698937093783,
|
|
"loss": 4.9682,
|
|
"mean_token_accuracy": 0.19634950160980225,
|
|
"num_tokens": 8886995.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.6331878185272215,
|
|
"epoch": 4.043082021541011,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00037669203642246685,
|
|
"loss": 5.0637,
|
|
"mean_token_accuracy": 0.19491669833660125,
|
|
"num_tokens": 8895224.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.63622932434082,
|
|
"epoch": 4.047224523612262,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037640687835004567,
|
|
"loss": 4.9991,
|
|
"mean_token_accuracy": 0.19510312974452973,
|
|
"num_tokens": 8905216.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.604790544509887,
|
|
"epoch": 4.051367025683513,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003761215157288692,
|
|
"loss": 5.0103,
|
|
"mean_token_accuracy": 0.20593261122703552,
|
|
"num_tokens": 8915493.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.527111196517945,
|
|
"epoch": 4.055509527754764,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037583594913454514,
|
|
"loss": 4.9674,
|
|
"mean_token_accuracy": 0.20826881378889084,
|
|
"num_tokens": 8924740.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.603532028198242,
|
|
"epoch": 4.059652029826015,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003755501791430925,
|
|
"loss": 5.0094,
|
|
"mean_token_accuracy": 0.19447989612817765,
|
|
"num_tokens": 8934726.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.6424449443817135,
|
|
"epoch": 4.063794531897266,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037526420633094044,
|
|
"loss": 5.0198,
|
|
"mean_token_accuracy": 0.19403040409088135,
|
|
"num_tokens": 8944301.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.647685289382935,
|
|
"epoch": 4.067937033968517,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037497803127492757,
|
|
"loss": 5.0333,
|
|
"mean_token_accuracy": 0.203432360291481,
|
|
"num_tokens": 8953967.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.515998983383179,
|
|
"epoch": 4.072079536039768,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003746916545523001,
|
|
"loss": 4.941,
|
|
"mean_token_accuracy": 0.20445602238178254,
|
|
"num_tokens": 8962983.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.6274574279785154,
|
|
"epoch": 4.076222038111019,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003744050767407113,
|
|
"loss": 5.0654,
|
|
"mean_token_accuracy": 0.19855775237083434,
|
|
"num_tokens": 8972019.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.607837820053101,
|
|
"epoch": 4.08036454018227,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003741182984182198,
|
|
"loss": 4.9404,
|
|
"mean_token_accuracy": 0.20956467241048812,
|
|
"num_tokens": 8981751.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.579764032363892,
|
|
"epoch": 4.084507042253521,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003738313201632889,
|
|
"loss": 4.9429,
|
|
"mean_token_accuracy": 0.2060052052140236,
|
|
"num_tokens": 8990330.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.477473354339599,
|
|
"epoch": 4.088649544324772,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003735441425547851,
|
|
"loss": 4.9414,
|
|
"mean_token_accuracy": 0.19764771014451982,
|
|
"num_tokens": 8998366.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.491153383255005,
|
|
"epoch": 4.092792046396023,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00037325676617197697,
|
|
"loss": 4.8846,
|
|
"mean_token_accuracy": 0.2112745687365532,
|
|
"num_tokens": 9006881.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.507876968383789,
|
|
"epoch": 4.096934548467274,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003729691915945342,
|
|
"loss": 4.9264,
|
|
"mean_token_accuracy": 0.20645105987787246,
|
|
"num_tokens": 9015415.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.621544408798218,
|
|
"epoch": 4.1010770505385254,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000372681419402526,
|
|
"loss": 4.9893,
|
|
"mean_token_accuracy": 0.19992004036903382,
|
|
"num_tokens": 9024578.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 5.589077043533325,
|
|
"epoch": 4.105219552609777,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003723934501764203,
|
|
"loss": 5.0008,
|
|
"mean_token_accuracy": 0.19620827287435533,
|
|
"num_tokens": 9032637.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.576468515396118,
|
|
"epoch": 4.109362054681028,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003721052844970826,
|
|
"loss": 4.931,
|
|
"mean_token_accuracy": 0.19871418178081512,
|
|
"num_tokens": 9041513.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.585641813278198,
|
|
"epoch": 4.113504556752279,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037181692294577457,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.19696544408798217,
|
|
"num_tokens": 9051024.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.473770570755005,
|
|
"epoch": 4.117647058823529,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037152836610415284,
|
|
"loss": 4.9364,
|
|
"mean_token_accuracy": 0.20744020342826844,
|
|
"num_tokens": 9059494.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.519936466217041,
|
|
"epoch": 4.12178956089478,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037123961455426836,
|
|
"loss": 4.8569,
|
|
"mean_token_accuracy": 0.20487651526927947,
|
|
"num_tokens": 9068166.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 5.555421686172485,
|
|
"epoch": 4.125932062966031,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037095066887856434,
|
|
"loss": 4.9462,
|
|
"mean_token_accuracy": 0.2053873598575592,
|
|
"num_tokens": 9077283.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.5558021068573,
|
|
"epoch": 4.130074565037282,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003706615296598758,
|
|
"loss": 4.9534,
|
|
"mean_token_accuracy": 0.20764309167861938,
|
|
"num_tokens": 9086605.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.510453844070435,
|
|
"epoch": 4.134217067108533,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003703721974814282,
|
|
"loss": 4.9484,
|
|
"mean_token_accuracy": 0.20199305713176727,
|
|
"num_tokens": 9095567.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.524186134338379,
|
|
"epoch": 4.1383595691797845,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037008267292683615,
|
|
"loss": 4.9411,
|
|
"mean_token_accuracy": 0.19794709980487823,
|
|
"num_tokens": 9104457.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.542298316955566,
|
|
"epoch": 4.142502071251036,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00036979295658010227,
|
|
"loss": 4.9917,
|
|
"mean_token_accuracy": 0.20759812742471695,
|
|
"num_tokens": 9113945.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.142502071251036,
|
|
"eval_entropy": 5.325104786643087,
|
|
"eval_loss": 5.791159152984619,
|
|
"eval_mean_token_accuracy": 0.16121668379698226,
|
|
"eval_num_tokens": 9113945.0,
|
|
"eval_runtime": 2.1084,
|
|
"eval_samples_per_second": 1777.154,
|
|
"eval_steps_per_second": 222.441,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.575279664993286,
|
|
"epoch": 4.146644573322287,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003695030490256161,
|
|
"loss": 5.0636,
|
|
"mean_token_accuracy": 0.19771672040224075,
|
|
"num_tokens": 9122996.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.588223266601562,
|
|
"epoch": 4.150787075393538,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000369212950848153,
|
|
"loss": 4.9147,
|
|
"mean_token_accuracy": 0.20448946952819824,
|
|
"num_tokens": 9132079.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.555638122558594,
|
|
"epoch": 4.154929577464789,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003689226626328725,
|
|
"loss": 5.038,
|
|
"mean_token_accuracy": 0.194102942943573,
|
|
"num_tokens": 9141335.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.584480428695679,
|
|
"epoch": 4.15907207953604,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036863218496531767,
|
|
"loss": 5.0626,
|
|
"mean_token_accuracy": 0.20296033322811127,
|
|
"num_tokens": 9150049.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.644832754135132,
|
|
"epoch": 4.163214581607291,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003683415184314139,
|
|
"loss": 4.9523,
|
|
"mean_token_accuracy": 0.20458456128835678,
|
|
"num_tokens": 9159245.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.494403553009033,
|
|
"epoch": 4.167357083678542,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036805066361746716,
|
|
"loss": 4.9568,
|
|
"mean_token_accuracy": 0.20082859545946122,
|
|
"num_tokens": 9167892.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.494073057174683,
|
|
"epoch": 4.1714995857497925,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036775962111016346,
|
|
"loss": 4.9657,
|
|
"mean_token_accuracy": 0.20667509436607362,
|
|
"num_tokens": 9176067.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.568603563308716,
|
|
"epoch": 4.175642087821044,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036746839149656734,
|
|
"loss": 4.9455,
|
|
"mean_token_accuracy": 0.20248468816280366,
|
|
"num_tokens": 9185318.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.510792827606201,
|
|
"epoch": 4.179784589892295,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003671769753641207,
|
|
"loss": 4.9885,
|
|
"mean_token_accuracy": 0.19469943195581435,
|
|
"num_tokens": 9194885.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.572036552429199,
|
|
"epoch": 4.183927091963546,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036688537330064187,
|
|
"loss": 5.0333,
|
|
"mean_token_accuracy": 0.19983012974262238,
|
|
"num_tokens": 9203906.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.487776470184326,
|
|
"epoch": 4.188069594034797,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000366593585894324,
|
|
"loss": 4.8766,
|
|
"mean_token_accuracy": 0.20726116001605988,
|
|
"num_tokens": 9213351.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.530978584289551,
|
|
"epoch": 4.192212096106048,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003663016137337341,
|
|
"loss": 4.9792,
|
|
"mean_token_accuracy": 0.19911393970251084,
|
|
"num_tokens": 9222191.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.586823797225952,
|
|
"epoch": 4.196354598177299,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036600945740781197,
|
|
"loss": 5.01,
|
|
"mean_token_accuracy": 0.20222293436527253,
|
|
"num_tokens": 9231517.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.599796485900879,
|
|
"epoch": 4.20049710024855,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00036571711750586896,
|
|
"loss": 5.024,
|
|
"mean_token_accuracy": 0.20583473294973373,
|
|
"num_tokens": 9241208.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.588336086273193,
|
|
"epoch": 4.204639602319801,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003654245946175864,
|
|
"loss": 4.9336,
|
|
"mean_token_accuracy": 0.20594659298658372,
|
|
"num_tokens": 9250237.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.587030982971191,
|
|
"epoch": 4.2087821043910525,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036513188933301504,
|
|
"loss": 5.0924,
|
|
"mean_token_accuracy": 0.19034477919340134,
|
|
"num_tokens": 9260273.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.572637367248535,
|
|
"epoch": 4.212924606462304,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003648390022425734,
|
|
"loss": 4.943,
|
|
"mean_token_accuracy": 0.20188928842544557,
|
|
"num_tokens": 9268878.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.533031892776489,
|
|
"epoch": 4.217067108533554,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003645459339370468,
|
|
"loss": 4.959,
|
|
"mean_token_accuracy": 0.20898036956787108,
|
|
"num_tokens": 9276324.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.523786640167236,
|
|
"epoch": 4.221209610604805,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036425268500758594,
|
|
"loss": 5.0062,
|
|
"mean_token_accuracy": 0.20836644917726516,
|
|
"num_tokens": 9285052.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.5912247657775875,
|
|
"epoch": 4.225352112676056,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00036395925604570614,
|
|
"loss": 4.9173,
|
|
"mean_token_accuracy": 0.20793739706277847,
|
|
"num_tokens": 9293324.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.586169958114624,
|
|
"epoch": 4.229494614747307,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003636656476432854,
|
|
"loss": 4.9853,
|
|
"mean_token_accuracy": 0.1950943261384964,
|
|
"num_tokens": 9302131.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.501743984222412,
|
|
"epoch": 4.233637116818558,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003633718603925642,
|
|
"loss": 4.8791,
|
|
"mean_token_accuracy": 0.20414563715457917,
|
|
"num_tokens": 9311128.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.625845241546631,
|
|
"epoch": 4.237779618889809,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003630778948861434,
|
|
"loss": 4.9794,
|
|
"mean_token_accuracy": 0.19475948065519333,
|
|
"num_tokens": 9319849.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.529236555099487,
|
|
"epoch": 4.2419221209610605,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003627837517169836,
|
|
"loss": 4.9886,
|
|
"mean_token_accuracy": 0.1945464313030243,
|
|
"num_tokens": 9328355.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.518364286422729,
|
|
"epoch": 4.246064623032312,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003624894314784037,
|
|
"loss": 4.9541,
|
|
"mean_token_accuracy": 0.2013275533914566,
|
|
"num_tokens": 9337562.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.537717294692993,
|
|
"epoch": 4.250207125103563,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003621949347640797,
|
|
"loss": 4.9913,
|
|
"mean_token_accuracy": 0.2016964539885521,
|
|
"num_tokens": 9347851.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.588918209075928,
|
|
"epoch": 4.254349627174814,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003619002621680439,
|
|
"loss": 4.9649,
|
|
"mean_token_accuracy": 0.20827952921390533,
|
|
"num_tokens": 9356524.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.641050243377686,
|
|
"epoch": 4.258492129246065,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003616054142846829,
|
|
"loss": 5.0672,
|
|
"mean_token_accuracy": 0.1944969415664673,
|
|
"num_tokens": 9365521.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.469163179397583,
|
|
"epoch": 4.262634631317316,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003613103917087371,
|
|
"loss": 4.9721,
|
|
"mean_token_accuracy": 0.1978716567158699,
|
|
"num_tokens": 9374088.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.503856420516968,
|
|
"epoch": 4.266777133388567,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003610151950352994,
|
|
"loss": 5.0212,
|
|
"mean_token_accuracy": 0.19691963344812394,
|
|
"num_tokens": 9382683.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.511172199249268,
|
|
"epoch": 4.270919635459817,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003607198248598137,
|
|
"loss": 4.9445,
|
|
"mean_token_accuracy": 0.20874057114124298,
|
|
"num_tokens": 9391790.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.6230686664581295,
|
|
"epoch": 4.2750621375310685,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003604242817780739,
|
|
"loss": 5.0295,
|
|
"mean_token_accuracy": 0.1953863501548767,
|
|
"num_tokens": 9400234.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.589998579025268,
|
|
"epoch": 4.27920463960232,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00036012856638622273,
|
|
"loss": 4.9456,
|
|
"mean_token_accuracy": 0.19400485008955,
|
|
"num_tokens": 9409513.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.492519187927246,
|
|
"epoch": 4.283347141673571,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003598326792807504,
|
|
"loss": 5.0253,
|
|
"mean_token_accuracy": 0.18992377817630768,
|
|
"num_tokens": 9418608.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.573359203338623,
|
|
"epoch": 4.287489643744822,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035953662105849357,
|
|
"loss": 5.0417,
|
|
"mean_token_accuracy": 0.20090628862380983,
|
|
"num_tokens": 9428705.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.5551652908325195,
|
|
"epoch": 4.291632145816073,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000359240392316634,
|
|
"loss": 4.9569,
|
|
"mean_token_accuracy": 0.20240781903266908,
|
|
"num_tokens": 9438149.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.573009824752807,
|
|
"epoch": 4.295774647887324,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003589439936526974,
|
|
"loss": 4.9338,
|
|
"mean_token_accuracy": 0.20852490663528442,
|
|
"num_tokens": 9447566.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.497342491149903,
|
|
"epoch": 4.299917149958575,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003586474256645525,
|
|
"loss": 5.0049,
|
|
"mean_token_accuracy": 0.1969831570982933,
|
|
"num_tokens": 9458156.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.604018926620483,
|
|
"epoch": 4.304059652029826,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035835068895040905,
|
|
"loss": 5.0498,
|
|
"mean_token_accuracy": 0.1937399759888649,
|
|
"num_tokens": 9467388.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.5714545249938965,
|
|
"epoch": 4.308202154101077,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00035805378410881753,
|
|
"loss": 5.0098,
|
|
"mean_token_accuracy": 0.2038671389222145,
|
|
"num_tokens": 9477515.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.557762479782104,
|
|
"epoch": 4.312344656172328,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003577567117386675,
|
|
"loss": 5.0075,
|
|
"mean_token_accuracy": 0.2024889886379242,
|
|
"num_tokens": 9485851.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.61469874382019,
|
|
"epoch": 4.3164871582435795,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003574594724391863,
|
|
"loss": 5.0252,
|
|
"mean_token_accuracy": 0.19521711468696595,
|
|
"num_tokens": 9495523.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.548380613327026,
|
|
"epoch": 4.32062966031483,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003571620668099382,
|
|
"loss": 4.9604,
|
|
"mean_token_accuracy": 0.20205987989902496,
|
|
"num_tokens": 9504195.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.605651140213013,
|
|
"epoch": 4.324772162386081,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003568644954508227,
|
|
"loss": 5.0015,
|
|
"mean_token_accuracy": 0.2055728778243065,
|
|
"num_tokens": 9513313.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.580483150482178,
|
|
"epoch": 4.328914664457332,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003565667589620739,
|
|
"loss": 5.0711,
|
|
"mean_token_accuracy": 0.19164893180131912,
|
|
"num_tokens": 9522387.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.540576028823852,
|
|
"epoch": 4.333057166528583,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003562688579442588,
|
|
"loss": 5.0031,
|
|
"mean_token_accuracy": 0.19807538241147996,
|
|
"num_tokens": 9531522.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.5370134830474855,
|
|
"epoch": 4.337199668599834,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035597079299827625,
|
|
"loss": 5.0288,
|
|
"mean_token_accuracy": 0.1964710086584091,
|
|
"num_tokens": 9540248.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.541212797164917,
|
|
"epoch": 4.341342170671085,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035567256472535585,
|
|
"loss": 4.9891,
|
|
"mean_token_accuracy": 0.19899856001138688,
|
|
"num_tokens": 9549850.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.571442937850952,
|
|
"epoch": 4.345484672742336,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003553741737270566,
|
|
"loss": 4.9579,
|
|
"mean_token_accuracy": 0.20556673258543015,
|
|
"num_tokens": 9558929.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.5093913078308105,
|
|
"epoch": 4.3496271748135875,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035507562060526574,
|
|
"loss": 4.9745,
|
|
"mean_token_accuracy": 0.20543653815984725,
|
|
"num_tokens": 9567627.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.574807167053223,
|
|
"epoch": 4.353769676884839,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035477690596219763,
|
|
"loss": 5.1068,
|
|
"mean_token_accuracy": 0.1915406808257103,
|
|
"num_tokens": 9576223.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.599922704696655,
|
|
"epoch": 4.35791217895609,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00035447803040039225,
|
|
"loss": 5.0454,
|
|
"mean_token_accuracy": 0.19947913736104966,
|
|
"num_tokens": 9585510.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.568687009811401,
|
|
"epoch": 4.362054681027341,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003541789945227143,
|
|
"loss": 4.9672,
|
|
"mean_token_accuracy": 0.208879691362381,
|
|
"num_tokens": 9594843.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.575809907913208,
|
|
"epoch": 4.366197183098592,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00035387979893235184,
|
|
"loss": 4.9962,
|
|
"mean_token_accuracy": 0.21260460317134858,
|
|
"num_tokens": 9604142.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.522830629348755,
|
|
"epoch": 4.370339685169842,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003535804442328151,
|
|
"loss": 5.0354,
|
|
"mean_token_accuracy": 0.2037716880440712,
|
|
"num_tokens": 9613669.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.591111326217652,
|
|
"epoch": 4.374482187241093,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035328093102793506,
|
|
"loss": 5.0159,
|
|
"mean_token_accuracy": 0.2032582700252533,
|
|
"num_tokens": 9623502.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.614077425003051,
|
|
"epoch": 4.378624689312344,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035298125992186283,
|
|
"loss": 5.0345,
|
|
"mean_token_accuracy": 0.19720663130283356,
|
|
"num_tokens": 9632041.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.576364946365357,
|
|
"epoch": 4.3827671913835955,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035268143151906764,
|
|
"loss": 5.0338,
|
|
"mean_token_accuracy": 0.19800903052091598,
|
|
"num_tokens": 9641442.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.500249147415161,
|
|
"epoch": 4.386909693454847,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00035238144642433616,
|
|
"loss": 4.9945,
|
|
"mean_token_accuracy": 0.20094509571790695,
|
|
"num_tokens": 9650857.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.591351556777954,
|
|
"epoch": 4.391052195526098,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000352081305242771,
|
|
"loss": 5.0329,
|
|
"mean_token_accuracy": 0.20460001677274703,
|
|
"num_tokens": 9659173.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.582446336746216,
|
|
"epoch": 4.395194697597349,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035178100857978994,
|
|
"loss": 4.9506,
|
|
"mean_token_accuracy": 0.20382338017225266,
|
|
"num_tokens": 9668114.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.480424118041992,
|
|
"epoch": 4.3993371996686,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035148055704112405,
|
|
"loss": 5.0155,
|
|
"mean_token_accuracy": 0.20217379927635193,
|
|
"num_tokens": 9676628.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.447292280197144,
|
|
"epoch": 4.403479701739851,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003511799512328168,
|
|
"loss": 4.9324,
|
|
"mean_token_accuracy": 0.20503744930028917,
|
|
"num_tokens": 9685801.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.543638420104981,
|
|
"epoch": 4.407622203811102,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003508791917612232,
|
|
"loss": 4.9923,
|
|
"mean_token_accuracy": 0.20178699046373366,
|
|
"num_tokens": 9694237.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.610736083984375,
|
|
"epoch": 4.411764705882353,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035057827923300785,
|
|
"loss": 4.9445,
|
|
"mean_token_accuracy": 0.20458553433418275,
|
|
"num_tokens": 9701904.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.492824554443359,
|
|
"epoch": 4.415907207953604,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035027721425514414,
|
|
"loss": 4.9069,
|
|
"mean_token_accuracy": 0.20770732909440995,
|
|
"num_tokens": 9710784.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.505400037765503,
|
|
"epoch": 4.4200497100248555,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003499759974349131,
|
|
"loss": 5.0245,
|
|
"mean_token_accuracy": 0.1987316682934761,
|
|
"num_tokens": 9720452.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.518406343460083,
|
|
"epoch": 4.424192212096106,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034967462937990194,
|
|
"loss": 5.0113,
|
|
"mean_token_accuracy": 0.19568999707698823,
|
|
"num_tokens": 9729122.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.619096994400024,
|
|
"epoch": 4.428334714167357,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000349373110698003,
|
|
"loss": 5.0754,
|
|
"mean_token_accuracy": 0.19746872037649155,
|
|
"num_tokens": 9738756.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.572315883636475,
|
|
"epoch": 4.432477216238608,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003490714419974124,
|
|
"loss": 5.0321,
|
|
"mean_token_accuracy": 0.19607203900814058,
|
|
"num_tokens": 9748547.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.6057971000671385,
|
|
"epoch": 4.436619718309859,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034876962388662885,
|
|
"loss": 5.0936,
|
|
"mean_token_accuracy": 0.18962884545326233,
|
|
"num_tokens": 9758421.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.501942920684814,
|
|
"epoch": 4.44076222038111,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003484676569744525,
|
|
"loss": 4.9115,
|
|
"mean_token_accuracy": 0.1995246559381485,
|
|
"num_tokens": 9766855.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.524267101287842,
|
|
"epoch": 4.444904722452361,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003481655418699836,
|
|
"loss": 4.9928,
|
|
"mean_token_accuracy": 0.2063850000500679,
|
|
"num_tokens": 9776157.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.552674388885498,
|
|
"epoch": 4.449047224523612,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034786327918262126,
|
|
"loss": 4.9,
|
|
"mean_token_accuracy": 0.20996199995279313,
|
|
"num_tokens": 9784386.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.5498682975769045,
|
|
"epoch": 4.4531897265948635,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034756086952206255,
|
|
"loss": 4.9424,
|
|
"mean_token_accuracy": 0.20693851709365846,
|
|
"num_tokens": 9793838.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.590666389465332,
|
|
"epoch": 4.457332228666115,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003472583134983005,
|
|
"loss": 5.0379,
|
|
"mean_token_accuracy": 0.1982978790998459,
|
|
"num_tokens": 9802595.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.589692783355713,
|
|
"epoch": 4.461474730737366,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034695561172162396,
|
|
"loss": 5.0169,
|
|
"mean_token_accuracy": 0.19614625424146653,
|
|
"num_tokens": 9811309.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.4726921081542965,
|
|
"epoch": 4.465617232808617,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034665276480261526,
|
|
"loss": 4.9542,
|
|
"mean_token_accuracy": 0.1961188718676567,
|
|
"num_tokens": 9820327.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.557860612869263,
|
|
"epoch": 4.469759734879867,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003463497733521499,
|
|
"loss": 5.0536,
|
|
"mean_token_accuracy": 0.19261214137077332,
|
|
"num_tokens": 9829588.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.639744138717651,
|
|
"epoch": 4.473902236951118,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003460466379813947,
|
|
"loss": 5.0243,
|
|
"mean_token_accuracy": 0.19458882361650467,
|
|
"num_tokens": 9838548.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.520778465270996,
|
|
"epoch": 4.478044739022369,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003457433593018068,
|
|
"loss": 5.042,
|
|
"mean_token_accuracy": 0.2011490061879158,
|
|
"num_tokens": 9847925.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.577949142456054,
|
|
"epoch": 4.48218724109362,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034543993792513243,
|
|
"loss": 5.0647,
|
|
"mean_token_accuracy": 0.1944203719496727,
|
|
"num_tokens": 9856839.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.651066493988037,
|
|
"epoch": 4.486329743164871,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003451363744634057,
|
|
"loss": 5.0217,
|
|
"mean_token_accuracy": 0.2023582398891449,
|
|
"num_tokens": 9866519.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.600418615341186,
|
|
"epoch": 4.4904722452361225,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034483266952894725,
|
|
"loss": 5.1015,
|
|
"mean_token_accuracy": 0.19910694062709808,
|
|
"num_tokens": 9876866.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.57288932800293,
|
|
"epoch": 4.494614747307374,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003445288237343632,
|
|
"loss": 5.051,
|
|
"mean_token_accuracy": 0.20302271693944932,
|
|
"num_tokens": 9884639.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.531723403930664,
|
|
"epoch": 4.498757249378625,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003442248376925436,
|
|
"loss": 5.0277,
|
|
"mean_token_accuracy": 0.19613326638936995,
|
|
"num_tokens": 9893819.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.462950372695923,
|
|
"epoch": 4.502899751449876,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034392071201666167,
|
|
"loss": 4.9569,
|
|
"mean_token_accuracy": 0.20001253634691238,
|
|
"num_tokens": 9902419.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.584561157226562,
|
|
"epoch": 4.507042253521127,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000343616447320172,
|
|
"loss": 5.0132,
|
|
"mean_token_accuracy": 0.20294599682092668,
|
|
"num_tokens": 9912434.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.596906995773315,
|
|
"epoch": 4.511184755592378,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003433120442168097,
|
|
"loss": 4.958,
|
|
"mean_token_accuracy": 0.20808692574501036,
|
|
"num_tokens": 9921800.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.527103757858276,
|
|
"epoch": 4.515327257663629,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00034300750332058924,
|
|
"loss": 5.0373,
|
|
"mean_token_accuracy": 0.19972340315580367,
|
|
"num_tokens": 9930863.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.536216163635254,
|
|
"epoch": 4.51946975973488,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003427028252458028,
|
|
"loss": 5.029,
|
|
"mean_token_accuracy": 0.19840795844793319,
|
|
"num_tokens": 9940524.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.572382497787475,
|
|
"epoch": 4.5236122618061305,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003423980106070194,
|
|
"loss": 5.0176,
|
|
"mean_token_accuracy": 0.19764320999383928,
|
|
"num_tokens": 9948949.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.5633138656616214,
|
|
"epoch": 4.527754763877382,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003420930600190834,
|
|
"loss": 5.0359,
|
|
"mean_token_accuracy": 0.19476404041051865,
|
|
"num_tokens": 9958154.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.5475754737854,
|
|
"epoch": 4.531897265948633,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003417879740971135,
|
|
"loss": 4.9549,
|
|
"mean_token_accuracy": 0.20078898519277572,
|
|
"num_tokens": 9967023.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.561546325683594,
|
|
"epoch": 4.536039768019884,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003414827534565014,
|
|
"loss": 5.0182,
|
|
"mean_token_accuracy": 0.1946766644716263,
|
|
"num_tokens": 9976596.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.540077161788941,
|
|
"epoch": 4.540182270091135,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034117739871291036,
|
|
"loss": 4.9524,
|
|
"mean_token_accuracy": 0.20736970007419586,
|
|
"num_tokens": 9985330.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.588950824737549,
|
|
"epoch": 4.544324772162386,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00034087191048227445,
|
|
"loss": 5.0582,
|
|
"mean_token_accuracy": 0.19825471490621566,
|
|
"num_tokens": 9996053.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.51121883392334,
|
|
"epoch": 4.548467274233637,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003405662893807967,
|
|
"loss": 4.9823,
|
|
"mean_token_accuracy": 0.20189203768968583,
|
|
"num_tokens": 10005517.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.5827068328857425,
|
|
"epoch": 4.552609776304888,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003402605360249482,
|
|
"loss": 5.0816,
|
|
"mean_token_accuracy": 0.1970222756266594,
|
|
"num_tokens": 10014354.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.522285223007202,
|
|
"epoch": 4.556752278376139,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000339954651031467,
|
|
"loss": 4.9655,
|
|
"mean_token_accuracy": 0.20246757715940475,
|
|
"num_tokens": 10023194.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.556752278376139,
|
|
"eval_entropy": 5.332227236172284,
|
|
"eval_loss": 5.7590413093566895,
|
|
"eval_mean_token_accuracy": 0.16357732894641758,
|
|
"eval_num_tokens": 10023194.0,
|
|
"eval_runtime": 2.1158,
|
|
"eval_samples_per_second": 1770.98,
|
|
"eval_steps_per_second": 221.668,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.4891246318817135,
|
|
"epoch": 4.5608947804473905,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033964863501735647,
|
|
"loss": 4.9613,
|
|
"mean_token_accuracy": 0.20575889945030212,
|
|
"num_tokens": 10032490.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.522625207901001,
|
|
"epoch": 4.565037282518642,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033934248859988453,
|
|
"loss": 4.973,
|
|
"mean_token_accuracy": 0.20597781538963317,
|
|
"num_tokens": 10041613.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.4799727439880375,
|
|
"epoch": 4.569179784589892,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003390362123965817,
|
|
"loss": 4.9177,
|
|
"mean_token_accuracy": 0.19979257583618165,
|
|
"num_tokens": 10050500.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.608070373535156,
|
|
"epoch": 4.573322286661144,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003387298070252407,
|
|
"loss": 4.9914,
|
|
"mean_token_accuracy": 0.2010465681552887,
|
|
"num_tokens": 10059650.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.531474351882935,
|
|
"epoch": 4.577464788732394,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003384232731039146,
|
|
"loss": 5.0378,
|
|
"mean_token_accuracy": 0.2063387379050255,
|
|
"num_tokens": 10067882.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.472613763809204,
|
|
"epoch": 4.581607290803645,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033811661125091584,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.20119038969278336,
|
|
"num_tokens": 10077147.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.529345560073852,
|
|
"epoch": 4.585749792874896,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033780982208481494,
|
|
"loss": 5.0354,
|
|
"mean_token_accuracy": 0.1956279009580612,
|
|
"num_tokens": 10086149.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.516835260391235,
|
|
"epoch": 4.589892294946147,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003375029062244392,
|
|
"loss": 4.9391,
|
|
"mean_token_accuracy": 0.20938792824745178,
|
|
"num_tokens": 10094997.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.64341287612915,
|
|
"epoch": 4.5940347970173985,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033719586428887125,
|
|
"loss": 5.0314,
|
|
"mean_token_accuracy": 0.1985064297914505,
|
|
"num_tokens": 10103585.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.514911842346192,
|
|
"epoch": 4.59817729908865,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033688869689744836,
|
|
"loss": 4.9148,
|
|
"mean_token_accuracy": 0.20262313038110732,
|
|
"num_tokens": 10113877.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.488924884796143,
|
|
"epoch": 4.602319801159901,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003365814046697608,
|
|
"loss": 5.0134,
|
|
"mean_token_accuracy": 0.19243194311857223,
|
|
"num_tokens": 10123127.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.610398006439209,
|
|
"epoch": 4.606462303231152,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003362739882256504,
|
|
"loss": 5.0896,
|
|
"mean_token_accuracy": 0.18990264236927032,
|
|
"num_tokens": 10132301.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.597383308410644,
|
|
"epoch": 4.610604805302403,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003359664481852099,
|
|
"loss": 5.0718,
|
|
"mean_token_accuracy": 0.19178640842437744,
|
|
"num_tokens": 10140985.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.497363471984864,
|
|
"epoch": 4.614747307373654,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033565878516878103,
|
|
"loss": 4.9725,
|
|
"mean_token_accuracy": 0.1992477670311928,
|
|
"num_tokens": 10149668.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.491865444183349,
|
|
"epoch": 4.618889809444905,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003353509997969537,
|
|
"loss": 5.0445,
|
|
"mean_token_accuracy": 0.19564536660909654,
|
|
"num_tokens": 10158584.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.557589101791382,
|
|
"epoch": 4.623032311516155,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033504309269056484,
|
|
"loss": 4.9896,
|
|
"mean_token_accuracy": 0.2018134504556656,
|
|
"num_tokens": 10167657.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.466531991958618,
|
|
"epoch": 4.6271748135874065,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003347350644706965,
|
|
"loss": 4.886,
|
|
"mean_token_accuracy": 0.2023467630147934,
|
|
"num_tokens": 10176657.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.580476856231689,
|
|
"epoch": 4.631317315658658,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003344269157586754,
|
|
"loss": 5.0069,
|
|
"mean_token_accuracy": 0.20492542237043382,
|
|
"num_tokens": 10185132.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.5081945896148685,
|
|
"epoch": 4.635459817729909,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033411864717607114,
|
|
"loss": 4.897,
|
|
"mean_token_accuracy": 0.2007085993885994,
|
|
"num_tokens": 10193428.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.575933790206909,
|
|
"epoch": 4.63960231980116,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003338102593446951,
|
|
"loss": 5.0736,
|
|
"mean_token_accuracy": 0.1944304257631302,
|
|
"num_tokens": 10202549.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.564479160308838,
|
|
"epoch": 4.643744821872411,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00033350175288659925,
|
|
"loss": 4.9483,
|
|
"mean_token_accuracy": 0.20355320423841478,
|
|
"num_tokens": 10211161.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.515966939926147,
|
|
"epoch": 4.647887323943662,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033319312842407483,
|
|
"loss": 5.0342,
|
|
"mean_token_accuracy": 0.19618548899888993,
|
|
"num_tokens": 10220988.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.466572570800781,
|
|
"epoch": 4.652029826014913,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003328843865796511,
|
|
"loss": 4.9418,
|
|
"mean_token_accuracy": 0.20542856454849243,
|
|
"num_tokens": 10229383.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.562775993347168,
|
|
"epoch": 4.656172328086164,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033257552797609426,
|
|
"loss": 5.0006,
|
|
"mean_token_accuracy": 0.19591869562864303,
|
|
"num_tokens": 10238622.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.737648439407349,
|
|
"epoch": 4.660314830157415,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033226655323640553,
|
|
"loss": 5.1217,
|
|
"mean_token_accuracy": 0.19706439673900605,
|
|
"num_tokens": 10248201.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.661566257476807,
|
|
"epoch": 4.664457332228666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000331957462983821,
|
|
"loss": 5.1309,
|
|
"mean_token_accuracy": 0.18809773921966552,
|
|
"num_tokens": 10258239.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.530985260009766,
|
|
"epoch": 4.668599834299917,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003316482578418094,
|
|
"loss": 5.0175,
|
|
"mean_token_accuracy": 0.19841015636920928,
|
|
"num_tokens": 10266915.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.550752067565918,
|
|
"epoch": 4.672742336371169,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033133893843407126,
|
|
"loss": 5.1315,
|
|
"mean_token_accuracy": 0.19444346874952317,
|
|
"num_tokens": 10276826.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.550286769866943,
|
|
"epoch": 4.676884838442419,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033102950538453776,
|
|
"loss": 5.052,
|
|
"mean_token_accuracy": 0.19384968280792236,
|
|
"num_tokens": 10286041.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.559438800811767,
|
|
"epoch": 4.68102734051367,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033071995931736897,
|
|
"loss": 5.0491,
|
|
"mean_token_accuracy": 0.1995423913002014,
|
|
"num_tokens": 10295258.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.554120969772339,
|
|
"epoch": 4.685169842584921,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00033041030085695323,
|
|
"loss": 4.9691,
|
|
"mean_token_accuracy": 0.20475178360939025,
|
|
"num_tokens": 10304726.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.53778109550476,
|
|
"epoch": 4.689312344656172,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003301005306279055,
|
|
"loss": 4.95,
|
|
"mean_token_accuracy": 0.20653076171875,
|
|
"num_tokens": 10312912.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.535728025436401,
|
|
"epoch": 4.693454846727423,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032979064925506627,
|
|
"loss": 5.01,
|
|
"mean_token_accuracy": 0.2014824703335762,
|
|
"num_tokens": 10322046.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.4885228157043455,
|
|
"epoch": 4.697597348798674,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003294806573634999,
|
|
"loss": 4.9727,
|
|
"mean_token_accuracy": 0.2114190548658371,
|
|
"num_tokens": 10331375.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.560163784027099,
|
|
"epoch": 4.7017398508699255,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003291705555784942,
|
|
"loss": 5.0704,
|
|
"mean_token_accuracy": 0.1980523183941841,
|
|
"num_tokens": 10341465.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.662457609176636,
|
|
"epoch": 4.705882352941177,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032886034452555814,
|
|
"loss": 5.1056,
|
|
"mean_token_accuracy": 0.18925056457519532,
|
|
"num_tokens": 10350137.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.498206424713135,
|
|
"epoch": 4.710024855012428,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003285500248304214,
|
|
"loss": 5.0074,
|
|
"mean_token_accuracy": 0.19517462700605392,
|
|
"num_tokens": 10358666.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.508872318267822,
|
|
"epoch": 4.714167357083679,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003282395971190328,
|
|
"loss": 4.9121,
|
|
"mean_token_accuracy": 0.2035570815205574,
|
|
"num_tokens": 10366910.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.586439085006714,
|
|
"epoch": 4.71830985915493,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032792906201755876,
|
|
"loss": 5.0596,
|
|
"mean_token_accuracy": 0.19604087620973587,
|
|
"num_tokens": 10376435.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.511652183532715,
|
|
"epoch": 4.72245236122618,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00032761842015238256,
|
|
"loss": 4.9586,
|
|
"mean_token_accuracy": 0.20873773097991943,
|
|
"num_tokens": 10384670.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.485471248626709,
|
|
"epoch": 4.726594863297431,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003273076721501029,
|
|
"loss": 4.9492,
|
|
"mean_token_accuracy": 0.20404998511075972,
|
|
"num_tokens": 10393731.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.585121297836304,
|
|
"epoch": 4.730737365368682,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003269968186375325,
|
|
"loss": 5.0627,
|
|
"mean_token_accuracy": 0.19787506014108658,
|
|
"num_tokens": 10402369.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.5847282886505125,
|
|
"epoch": 4.7348798674399335,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003266858602416967,
|
|
"loss": 5.0238,
|
|
"mean_token_accuracy": 0.1989196866750717,
|
|
"num_tokens": 10411566.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.597936582565308,
|
|
"epoch": 4.739022369511185,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00032637479758983255,
|
|
"loss": 5.0562,
|
|
"mean_token_accuracy": 0.19529632478952408,
|
|
"num_tokens": 10420086.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.571719741821289,
|
|
"epoch": 4.743164871582436,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032606363130938766,
|
|
"loss": 5.1032,
|
|
"mean_token_accuracy": 0.19663951247930528,
|
|
"num_tokens": 10429746.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.5720525741577145,
|
|
"epoch": 4.747307373653687,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003257523620280181,
|
|
"loss": 4.9837,
|
|
"mean_token_accuracy": 0.19508006423711777,
|
|
"num_tokens": 10438536.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.548521852493286,
|
|
"epoch": 4.751449875724938,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032544099037358834,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.20414538383483888,
|
|
"num_tokens": 10447471.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.529682397842407,
|
|
"epoch": 4.755592377796189,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003251295169741688,
|
|
"loss": 5.0076,
|
|
"mean_token_accuracy": 0.20929636806249619,
|
|
"num_tokens": 10457179.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.5125141620635985,
|
|
"epoch": 4.75973487986744,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032481794245803536,
|
|
"loss": 5.0622,
|
|
"mean_token_accuracy": 0.19754002541303634,
|
|
"num_tokens": 10466291.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.48158483505249,
|
|
"epoch": 4.763877381938691,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032450626745366806,
|
|
"loss": 4.9481,
|
|
"mean_token_accuracy": 0.19789758175611497,
|
|
"num_tokens": 10475452.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.558969926834107,
|
|
"epoch": 4.768019884009942,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032419449258974924,
|
|
"loss": 4.9495,
|
|
"mean_token_accuracy": 0.20292745977640153,
|
|
"num_tokens": 10484381.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.6061560153961185,
|
|
"epoch": 4.7721623860811935,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000323882618495163,
|
|
"loss": 5.1193,
|
|
"mean_token_accuracy": 0.19379464834928511,
|
|
"num_tokens": 10495132.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.556785631179809,
|
|
"epoch": 4.776304888152444,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003235706457989934,
|
|
"loss": 5.0625,
|
|
"mean_token_accuracy": 0.19120130091905593,
|
|
"num_tokens": 10504519.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.592218971252441,
|
|
"epoch": 4.780447390223695,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032325857513052354,
|
|
"loss": 5.0232,
|
|
"mean_token_accuracy": 0.198995740711689,
|
|
"num_tokens": 10513575.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.594635343551635,
|
|
"epoch": 4.784589892294946,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000322946407119234,
|
|
"loss": 5.1003,
|
|
"mean_token_accuracy": 0.1960811883211136,
|
|
"num_tokens": 10522148.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.524764347076416,
|
|
"epoch": 4.788732394366197,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032263414239480187,
|
|
"loss": 5.0244,
|
|
"mean_token_accuracy": 0.19612499326467514,
|
|
"num_tokens": 10531183.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.5439550399780275,
|
|
"epoch": 4.792874896437448,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000322321781587099,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.2006504863500595,
|
|
"num_tokens": 10540677.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.511647796630859,
|
|
"epoch": 4.797017398508699,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003220093253261915,
|
|
"loss": 4.9785,
|
|
"mean_token_accuracy": 0.19443445056676864,
|
|
"num_tokens": 10550009.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.526203346252442,
|
|
"epoch": 4.80115990057995,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003216967742423377,
|
|
"loss": 5.0176,
|
|
"mean_token_accuracy": 0.19927203506231309,
|
|
"num_tokens": 10559694.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.560089778900147,
|
|
"epoch": 4.8053024026512015,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003213841289659872,
|
|
"loss": 5.0009,
|
|
"mean_token_accuracy": 0.21151264309883117,
|
|
"num_tokens": 10567965.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.547700500488281,
|
|
"epoch": 4.809444904722453,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003210713901277799,
|
|
"loss": 5.0147,
|
|
"mean_token_accuracy": 0.20137341767549516,
|
|
"num_tokens": 10577653.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.573288679122925,
|
|
"epoch": 4.813587406793704,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000320758558358544,
|
|
"loss": 5.014,
|
|
"mean_token_accuracy": 0.19640549272298813,
|
|
"num_tokens": 10586903.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.511530590057373,
|
|
"epoch": 4.817729908864955,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003204456342892955,
|
|
"loss": 4.9184,
|
|
"mean_token_accuracy": 0.20262907445430756,
|
|
"num_tokens": 10595411.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.449669265747071,
|
|
"epoch": 4.821872410936205,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032013261855123637,
|
|
"loss": 5.0033,
|
|
"mean_token_accuracy": 0.1966636672616005,
|
|
"num_tokens": 10604147.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.566453266143799,
|
|
"epoch": 4.826014913007457,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00031981951177575366,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.1945537209510803,
|
|
"num_tokens": 10614048.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.622162485122681,
|
|
"epoch": 4.830157415078707,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003195063145944178,
|
|
"loss": 5.1257,
|
|
"mean_token_accuracy": 0.19485061317682267,
|
|
"num_tokens": 10623399.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.586702489852906,
|
|
"epoch": 4.834299917149958,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003191930276389819,
|
|
"loss": 4.962,
|
|
"mean_token_accuracy": 0.2041936695575714,
|
|
"num_tokens": 10632627.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.540104532241822,
|
|
"epoch": 4.838442419221209,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031887965154137987,
|
|
"loss": 5.0164,
|
|
"mean_token_accuracy": 0.19275878369808197,
|
|
"num_tokens": 10641155.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.596768856048584,
|
|
"epoch": 4.8425849212924605,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003185661869337255,
|
|
"loss": 5.0463,
|
|
"mean_token_accuracy": 0.20041119754314424,
|
|
"num_tokens": 10649661.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.476096487045288,
|
|
"epoch": 4.846727423363712,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00031825263444831137,
|
|
"loss": 4.9589,
|
|
"mean_token_accuracy": 0.20016977488994597,
|
|
"num_tokens": 10658652.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.552750158309936,
|
|
"epoch": 4.850869925434963,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003179389947176069,
|
|
"loss": 5.0708,
|
|
"mean_token_accuracy": 0.19576626867055893,
|
|
"num_tokens": 10668703.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.5536716938018795,
|
|
"epoch": 4.855012427506214,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003176252683742578,
|
|
"loss": 4.9951,
|
|
"mean_token_accuracy": 0.19600070267915726,
|
|
"num_tokens": 10677740.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.5348320484161375,
|
|
"epoch": 4.859154929577465,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003173114560510843,
|
|
"loss": 4.9898,
|
|
"mean_token_accuracy": 0.20137470364570617,
|
|
"num_tokens": 10686429.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.597005939483642,
|
|
"epoch": 4.863297431648716,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031699755838108027,
|
|
"loss": 5.0803,
|
|
"mean_token_accuracy": 0.1996299922466278,
|
|
"num_tokens": 10696367.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.577630662918091,
|
|
"epoch": 4.867439933719967,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003166835759974115,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.19527925848960875,
|
|
"num_tokens": 10705712.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.557117652893067,
|
|
"epoch": 4.871582435791218,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031636950953341486,
|
|
"loss": 5.0525,
|
|
"mean_token_accuracy": 0.1936440125107765,
|
|
"num_tokens": 10715897.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.518204307556152,
|
|
"epoch": 4.8757249378624685,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031605535962259666,
|
|
"loss": 4.9791,
|
|
"mean_token_accuracy": 0.18888992071151733,
|
|
"num_tokens": 10724810.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.569813013076782,
|
|
"epoch": 4.87986743993372,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003157411268986316,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.20020650029182435,
|
|
"num_tokens": 10734661.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.519001817703247,
|
|
"epoch": 4.884009942004971,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003154268119953615,
|
|
"loss": 4.9876,
|
|
"mean_token_accuracy": 0.20336902588605882,
|
|
"num_tokens": 10743642.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.560329580307007,
|
|
"epoch": 4.888152444076222,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003151124155467938,
|
|
"loss": 5.012,
|
|
"mean_token_accuracy": 0.20232246220111846,
|
|
"num_tokens": 10752812.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.528487777709961,
|
|
"epoch": 4.892294946147473,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031479793818710056,
|
|
"loss": 5.0302,
|
|
"mean_token_accuracy": 0.19297149181365966,
|
|
"num_tokens": 10762117.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.483056592941284,
|
|
"epoch": 4.896437448218724,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031448338055061685,
|
|
"loss": 4.9433,
|
|
"mean_token_accuracy": 0.20823530703783036,
|
|
"num_tokens": 10771515.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.542733860015869,
|
|
"epoch": 4.900579950289975,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031416874327183994,
|
|
"loss": 4.9776,
|
|
"mean_token_accuracy": 0.20258747637271882,
|
|
"num_tokens": 10780610.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.474884271621704,
|
|
"epoch": 4.904722452361226,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031385402698542753,
|
|
"loss": 4.9729,
|
|
"mean_token_accuracy": 0.20315501540899278,
|
|
"num_tokens": 10789169.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.521940803527832,
|
|
"epoch": 4.908864954432477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031353923232619674,
|
|
"loss": 4.958,
|
|
"mean_token_accuracy": 0.2027697592973709,
|
|
"num_tokens": 10798636.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.535996770858764,
|
|
"epoch": 4.9130074565037285,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003132243599291229,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.20239778608083725,
|
|
"num_tokens": 10808245.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.566502094268799,
|
|
"epoch": 4.91714995857498,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031290941042933797,
|
|
"loss": 5.0048,
|
|
"mean_token_accuracy": 0.19770491272211074,
|
|
"num_tokens": 10817164.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.52999324798584,
|
|
"epoch": 4.92129246064623,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031259438446212953,
|
|
"loss": 5.0402,
|
|
"mean_token_accuracy": 0.19190698713064194,
|
|
"num_tokens": 10826098.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.504102373123169,
|
|
"epoch": 4.925434962717482,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031227928266293933,
|
|
"loss": 5.0635,
|
|
"mean_token_accuracy": 0.19432169646024705,
|
|
"num_tokens": 10836427.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.53760461807251,
|
|
"epoch": 4.929577464788732,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031196410566736215,
|
|
"loss": 5.0184,
|
|
"mean_token_accuracy": 0.20221779346466065,
|
|
"num_tokens": 10845804.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.579085159301758,
|
|
"epoch": 4.933719966859983,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00031164885411114453,
|
|
"loss": 4.9999,
|
|
"mean_token_accuracy": 0.1958640471100807,
|
|
"num_tokens": 10855951.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.535676527023315,
|
|
"epoch": 4.937862468931234,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00031133352863018315,
|
|
"loss": 5.116,
|
|
"mean_token_accuracy": 0.19759414047002793,
|
|
"num_tokens": 10865528.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.571910285949707,
|
|
"epoch": 4.942004971002485,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031101812986052406,
|
|
"loss": 5.0522,
|
|
"mean_token_accuracy": 0.20647826939821243,
|
|
"num_tokens": 10874365.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.5368860244750975,
|
|
"epoch": 4.9461474730737365,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031070265843836107,
|
|
"loss": 4.9708,
|
|
"mean_token_accuracy": 0.19827270805835723,
|
|
"num_tokens": 10883034.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.569532203674316,
|
|
"epoch": 4.950289975144988,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003103871150000345,
|
|
"loss": 5.0335,
|
|
"mean_token_accuracy": 0.20150164067745208,
|
|
"num_tokens": 10892307.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.5545354843139645,
|
|
"epoch": 4.954432477216239,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031007150018202983,
|
|
"loss": 5.0835,
|
|
"mean_token_accuracy": 0.19231959879398347,
|
|
"num_tokens": 10902074.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.531843662261963,
|
|
"epoch": 4.95857497928749,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030975581462097673,
|
|
"loss": 4.9453,
|
|
"mean_token_accuracy": 0.19770769029855728,
|
|
"num_tokens": 10911205.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.518322992324829,
|
|
"epoch": 4.962717481358741,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030944005895364754,
|
|
"loss": 4.9717,
|
|
"mean_token_accuracy": 0.19631796479225158,
|
|
"num_tokens": 10920371.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.504184055328369,
|
|
"epoch": 4.966859983429992,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030912423381695584,
|
|
"loss": 4.9599,
|
|
"mean_token_accuracy": 0.20894330590963364,
|
|
"num_tokens": 10928741.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.511369419097901,
|
|
"epoch": 4.971002485501243,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00030880833984795554,
|
|
"loss": 4.9728,
|
|
"mean_token_accuracy": 0.20293036103248596,
|
|
"num_tokens": 10938074.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 4.971002485501243,
|
|
"eval_entropy": 5.3413487080572,
|
|
"eval_loss": 5.731026649475098,
|
|
"eval_mean_token_accuracy": 0.16661989340967714,
|
|
"eval_num_tokens": 10938074.0,
|
|
"eval_runtime": 2.1129,
|
|
"eval_samples_per_second": 1773.371,
|
|
"eval_steps_per_second": 221.967,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.469365835189819,
|
|
"epoch": 4.975144987572493,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003084923776838392,
|
|
"loss": 5.0005,
|
|
"mean_token_accuracy": 0.19742163717746736,
|
|
"num_tokens": 10946756.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.503800773620606,
|
|
"epoch": 4.9792874896437445,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030817634796193705,
|
|
"loss": 4.9623,
|
|
"mean_token_accuracy": 0.20523536801338196,
|
|
"num_tokens": 10956313.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.6429239273071286,
|
|
"epoch": 4.983429991714996,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00030786025131971555,
|
|
"loss": 5.035,
|
|
"mean_token_accuracy": 0.2020207926630974,
|
|
"num_tokens": 10964333.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.525502061843872,
|
|
"epoch": 4.987572493786247,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003075440883947764,
|
|
"loss": 4.9536,
|
|
"mean_token_accuracy": 0.20119588375091552,
|
|
"num_tokens": 10974110.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.479152011871338,
|
|
"epoch": 4.991714995857498,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003072278598248544,
|
|
"loss": 4.9927,
|
|
"mean_token_accuracy": 0.199561271071434,
|
|
"num_tokens": 10983186.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.495191049575806,
|
|
"epoch": 4.995857497928749,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00030691156624781734,
|
|
"loss": 4.9967,
|
|
"mean_token_accuracy": 0.20461216270923616,
|
|
"num_tokens": 10992244.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.516499280929565,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.000306595208301664,
|
|
"loss": 4.9655,
|
|
"mean_token_accuracy": 0.20728713124990464,
|
|
"num_tokens": 11000005.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.57286434173584,
|
|
"epoch": 5.004142502071251,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030627878662452273,
|
|
"loss": 4.8415,
|
|
"mean_token_accuracy": 0.20957619100809097,
|
|
"num_tokens": 11009346.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.592178583145142,
|
|
"epoch": 5.008285004142502,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003059623018546508,
|
|
"loss": 4.8078,
|
|
"mean_token_accuracy": 0.22071910798549652,
|
|
"num_tokens": 11017844.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.607227039337158,
|
|
"epoch": 5.012427506213753,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003056457546304326,
|
|
"loss": 4.8318,
|
|
"mean_token_accuracy": 0.21769062280654908,
|
|
"num_tokens": 11026840.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.495561027526856,
|
|
"epoch": 5.016570008285004,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003053291455903782,
|
|
"loss": 4.7958,
|
|
"mean_token_accuracy": 0.2191757082939148,
|
|
"num_tokens": 11034906.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.520737552642823,
|
|
"epoch": 5.0207125103562555,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00030501247537312293,
|
|
"loss": 4.8435,
|
|
"mean_token_accuracy": 0.212516850233078,
|
|
"num_tokens": 11044403.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.518807792663575,
|
|
"epoch": 5.024855012427507,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003046957446174251,
|
|
"loss": 4.7887,
|
|
"mean_token_accuracy": 0.2178909435868263,
|
|
"num_tokens": 11052685.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.470620441436767,
|
|
"epoch": 5.028997514498757,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030437895396216524,
|
|
"loss": 4.6533,
|
|
"mean_token_accuracy": 0.23105335831642151,
|
|
"num_tokens": 11060830.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.576980066299439,
|
|
"epoch": 5.033140016570008,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030406210404634475,
|
|
"loss": 4.8448,
|
|
"mean_token_accuracy": 0.20931386649608613,
|
|
"num_tokens": 11070220.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.531246280670166,
|
|
"epoch": 5.037282518641259,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003037451955090846,
|
|
"loss": 4.7675,
|
|
"mean_token_accuracy": 0.2134119153022766,
|
|
"num_tokens": 11079171.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.540034818649292,
|
|
"epoch": 5.04142502071251,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00030342822898962377,
|
|
"loss": 4.8399,
|
|
"mean_token_accuracy": 0.21195529997348786,
|
|
"num_tokens": 11089205.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.523284673690796,
|
|
"epoch": 5.045567522783761,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003031112051273185,
|
|
"loss": 4.8156,
|
|
"mean_token_accuracy": 0.21725714355707168,
|
|
"num_tokens": 11098339.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.525024032592773,
|
|
"epoch": 5.049710024855012,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003027941245616405,
|
|
"loss": 4.7344,
|
|
"mean_token_accuracy": 0.22797350436449051,
|
|
"num_tokens": 11107218.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.517011022567749,
|
|
"epoch": 5.0538525269262635,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003024769879321761,
|
|
"loss": 4.8405,
|
|
"mean_token_accuracy": 0.21409492939710617,
|
|
"num_tokens": 11116457.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.497248220443725,
|
|
"epoch": 5.057995028997515,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003021597958786243,
|
|
"loss": 4.8015,
|
|
"mean_token_accuracy": 0.21317162215709687,
|
|
"num_tokens": 11124834.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.510232019424438,
|
|
"epoch": 5.062137531068766,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003018425490407962,
|
|
"loss": 4.8159,
|
|
"mean_token_accuracy": 0.22010593861341476,
|
|
"num_tokens": 11133920.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.596433496475219,
|
|
"epoch": 5.066280033140017,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00030152524805861344,
|
|
"loss": 4.85,
|
|
"mean_token_accuracy": 0.21346355825662613,
|
|
"num_tokens": 11143573.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.527303218841553,
|
|
"epoch": 5.070422535211268,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030120789357210667,
|
|
"loss": 4.8888,
|
|
"mean_token_accuracy": 0.2141047865152359,
|
|
"num_tokens": 11154255.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.519129991531372,
|
|
"epoch": 5.074565037282519,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030089048622141453,
|
|
"loss": 4.8088,
|
|
"mean_token_accuracy": 0.2143345817923546,
|
|
"num_tokens": 11163306.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.570961284637451,
|
|
"epoch": 5.078707539353769,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030057302664678243,
|
|
"loss": 4.839,
|
|
"mean_token_accuracy": 0.2142380803823471,
|
|
"num_tokens": 11172763.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.495805454254151,
|
|
"epoch": 5.08285004142502,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00030025551548856105,
|
|
"loss": 4.8023,
|
|
"mean_token_accuracy": 0.22165123820304872,
|
|
"num_tokens": 11182196.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.490293169021607,
|
|
"epoch": 5.0869925434962715,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002999379533872051,
|
|
"loss": 4.8391,
|
|
"mean_token_accuracy": 0.21249848753213882,
|
|
"num_tokens": 11192462.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.489285707473755,
|
|
"epoch": 5.091135045567523,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002996203409832719,
|
|
"loss": 4.8193,
|
|
"mean_token_accuracy": 0.22059475630521774,
|
|
"num_tokens": 11201805.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.510881185531616,
|
|
"epoch": 5.095277547638774,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002993026789174205,
|
|
"loss": 4.8005,
|
|
"mean_token_accuracy": 0.2264577254652977,
|
|
"num_tokens": 11210114.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.568402051925659,
|
|
"epoch": 5.099420049710025,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002989849678304099,
|
|
"loss": 4.8905,
|
|
"mean_token_accuracy": 0.2115759313106537,
|
|
"num_tokens": 11219725.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.540450477600098,
|
|
"epoch": 5.103562551781276,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00029866720836309817,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.21033845245838165,
|
|
"num_tokens": 11228583.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.505786609649658,
|
|
"epoch": 5.107705053852527,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00029834940115644086,
|
|
"loss": 4.8117,
|
|
"mean_token_accuracy": 0.2107660323381424,
|
|
"num_tokens": 11237787.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.464326572418213,
|
|
"epoch": 5.111847555923778,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002980315468514898,
|
|
"loss": 4.8541,
|
|
"mean_token_accuracy": 0.205310919880867,
|
|
"num_tokens": 11247136.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.443206024169922,
|
|
"epoch": 5.115990057995029,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00029771364608939196,
|
|
"loss": 4.7667,
|
|
"mean_token_accuracy": 0.21348730325698853,
|
|
"num_tokens": 11256931.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.543191576004029,
|
|
"epoch": 5.12013256006628,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002973956995113878,
|
|
"loss": 4.7785,
|
|
"mean_token_accuracy": 0.21458605527877808,
|
|
"num_tokens": 11266457.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.467357873916626,
|
|
"epoch": 5.1242750621375315,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002970777077588103,
|
|
"loss": 4.7213,
|
|
"mean_token_accuracy": 0.22049364000558852,
|
|
"num_tokens": 11274239.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.509474802017212,
|
|
"epoch": 5.128417564208782,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002967596714730837,
|
|
"loss": 4.8522,
|
|
"mean_token_accuracy": 0.21160443723201752,
|
|
"num_tokens": 11283239.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.491240406036377,
|
|
"epoch": 5.132560066280033,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00029644159129572185,
|
|
"loss": 4.8095,
|
|
"mean_token_accuracy": 0.21354417502880096,
|
|
"num_tokens": 11291662.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.47540431022644,
|
|
"epoch": 5.136702568351284,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002961234678683272,
|
|
"loss": 4.8104,
|
|
"mean_token_accuracy": 0.2211032137274742,
|
|
"num_tokens": 11300165.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.558735466003418,
|
|
"epoch": 5.140845070422535,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002958053018325896,
|
|
"loss": 4.799,
|
|
"mean_token_accuracy": 0.22224198281764984,
|
|
"num_tokens": 11309729.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.537526321411133,
|
|
"epoch": 5.144987572493786,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002954870938302846,
|
|
"loss": 4.8343,
|
|
"mean_token_accuracy": 0.21238924711942672,
|
|
"num_tokens": 11318674.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.539915418624878,
|
|
"epoch": 5.149130074565037,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002951688445032726,
|
|
"loss": 4.8113,
|
|
"mean_token_accuracy": 0.21471831053495408,
|
|
"num_tokens": 11327583.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.557190561294556,
|
|
"epoch": 5.153272576636288,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002948505544934971,
|
|
"loss": 4.8412,
|
|
"mean_token_accuracy": 0.21318480521440505,
|
|
"num_tokens": 11338132.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.531205177307129,
|
|
"epoch": 5.1574150787075395,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000294532224442984,
|
|
"loss": 4.8737,
|
|
"mean_token_accuracy": 0.2153089702129364,
|
|
"num_tokens": 11347210.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.485893440246582,
|
|
"epoch": 5.161557580778791,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029421385499383976,
|
|
"loss": 4.7605,
|
|
"mean_token_accuracy": 0.22600371092557908,
|
|
"num_tokens": 11355479.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.5649974822998045,
|
|
"epoch": 5.165700082850042,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029389544678825035,
|
|
"loss": 4.9015,
|
|
"mean_token_accuracy": 0.20461722165346147,
|
|
"num_tokens": 11364518.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.440051937103272,
|
|
"epoch": 5.169842584921293,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00029357700046847985,
|
|
"loss": 4.8173,
|
|
"mean_token_accuracy": 0.21370338499546052,
|
|
"num_tokens": 11373589.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.577805280685425,
|
|
"epoch": 5.173985086992544,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002932585166768694,
|
|
"loss": 4.8767,
|
|
"mean_token_accuracy": 0.21139735281467437,
|
|
"num_tokens": 11383209.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.506186008453369,
|
|
"epoch": 5.178127589063794,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029293999605583545,
|
|
"loss": 4.8014,
|
|
"mean_token_accuracy": 0.21928748488426208,
|
|
"num_tokens": 11392012.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.537167072296143,
|
|
"epoch": 5.182270091135045,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029262143924786905,
|
|
"loss": 4.8575,
|
|
"mean_token_accuracy": 0.21298470348119736,
|
|
"num_tokens": 11400489.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.483784770965576,
|
|
"epoch": 5.186412593206296,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029230284689553404,
|
|
"loss": 4.7469,
|
|
"mean_token_accuracy": 0.2189180552959442,
|
|
"num_tokens": 11409337.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.5318530082702635,
|
|
"epoch": 5.190555095277547,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029198421964146605,
|
|
"loss": 4.8229,
|
|
"mean_token_accuracy": 0.22308073788881302,
|
|
"num_tokens": 11418650.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.47744665145874,
|
|
"epoch": 5.1946975973487985,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002916655581283709,
|
|
"loss": 4.8144,
|
|
"mean_token_accuracy": 0.20934932678937912,
|
|
"num_tokens": 11427516.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.429421663284302,
|
|
"epoch": 5.19884009942005,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00029134686299902397,
|
|
"loss": 4.7906,
|
|
"mean_token_accuracy": 0.2170763835310936,
|
|
"num_tokens": 11435676.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.517792367935181,
|
|
"epoch": 5.202982601491301,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002910281348962679,
|
|
"loss": 4.8545,
|
|
"mean_token_accuracy": 0.19837765246629716,
|
|
"num_tokens": 11444520.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.54579725265503,
|
|
"epoch": 5.207125103562552,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00029070937446301225,
|
|
"loss": 4.8269,
|
|
"mean_token_accuracy": 0.20962976664304733,
|
|
"num_tokens": 11453759.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.646638250350952,
|
|
"epoch": 5.211267605633803,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002903905823422317,
|
|
"loss": 4.969,
|
|
"mean_token_accuracy": 0.19739352613687516,
|
|
"num_tokens": 11463409.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.510572814941407,
|
|
"epoch": 5.215410107705054,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002900717591769646,
|
|
"loss": 4.85,
|
|
"mean_token_accuracy": 0.21569056957960128,
|
|
"num_tokens": 11473440.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.482707977294922,
|
|
"epoch": 5.219552609776305,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002897529056103122,
|
|
"loss": 4.7113,
|
|
"mean_token_accuracy": 0.22551114857196808,
|
|
"num_tokens": 11482353.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.464768409729004,
|
|
"epoch": 5.223695111847556,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002894340222854371,
|
|
"loss": 4.8479,
|
|
"mean_token_accuracy": 0.21127415895462037,
|
|
"num_tokens": 11491331.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.452624320983887,
|
|
"epoch": 5.2278376139188065,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002891151098455617,
|
|
"loss": 4.7739,
|
|
"mean_token_accuracy": 0.2104616105556488,
|
|
"num_tokens": 11500553.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.398191452026367,
|
|
"epoch": 5.231980115990058,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002887961689339672,
|
|
"loss": 4.7028,
|
|
"mean_token_accuracy": 0.2209431067109108,
|
|
"num_tokens": 11509146.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.551649713516236,
|
|
"epoch": 5.236122618061309,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002884772001939924,
|
|
"loss": 4.8358,
|
|
"mean_token_accuracy": 0.21035597324371338,
|
|
"num_tokens": 11518172.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.487992477416992,
|
|
"epoch": 5.24026512013256,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000288158204269032,
|
|
"loss": 4.8179,
|
|
"mean_token_accuracy": 0.2106998324394226,
|
|
"num_tokens": 11527291.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.491569900512696,
|
|
"epoch": 5.244407622203811,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028783918180253567,
|
|
"loss": 4.8467,
|
|
"mean_token_accuracy": 0.2201030433177948,
|
|
"num_tokens": 11536306.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.513833808898926,
|
|
"epoch": 5.248550124275062,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002875201334380066,
|
|
"loss": 4.8875,
|
|
"mean_token_accuracy": 0.21367935836315155,
|
|
"num_tokens": 11545879.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.457678031921387,
|
|
"epoch": 5.252692626346313,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002872010598190002,
|
|
"loss": 4.7843,
|
|
"mean_token_accuracy": 0.22213353514671325,
|
|
"num_tokens": 11555067.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.484569454193116,
|
|
"epoch": 5.256835128417564,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002868819615891229,
|
|
"loss": 4.7671,
|
|
"mean_token_accuracy": 0.21414062529802322,
|
|
"num_tokens": 11563419.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.447198820114136,
|
|
"epoch": 5.260977630488815,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002865628393920305,
|
|
"loss": 4.8449,
|
|
"mean_token_accuracy": 0.2085914582014084,
|
|
"num_tokens": 11573018.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.453502941131592,
|
|
"epoch": 5.2651201325600665,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002862436938714276,
|
|
"loss": 4.7746,
|
|
"mean_token_accuracy": 0.21958879232406617,
|
|
"num_tokens": 11581996.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.5813130855560305,
|
|
"epoch": 5.269262634631318,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002859245256710654,
|
|
"loss": 4.837,
|
|
"mean_token_accuracy": 0.21546498090028762,
|
|
"num_tokens": 11591706.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.501917123794556,
|
|
"epoch": 5.273405136702569,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002856053354347411,
|
|
"loss": 4.8609,
|
|
"mean_token_accuracy": 0.2159338191151619,
|
|
"num_tokens": 11600616.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.44673752784729,
|
|
"epoch": 5.27754763877382,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002852861238062963,
|
|
"loss": 4.7486,
|
|
"mean_token_accuracy": 0.22136152535676956,
|
|
"num_tokens": 11609650.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.504470109939575,
|
|
"epoch": 5.28169014084507,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002849668914296157,
|
|
"loss": 4.7631,
|
|
"mean_token_accuracy": 0.21626771241426468,
|
|
"num_tokens": 11618785.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.504045486450195,
|
|
"epoch": 5.285832642916321,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000284647638948626,
|
|
"loss": 4.8493,
|
|
"mean_token_accuracy": 0.21150384247303008,
|
|
"num_tokens": 11628089.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.49078483581543,
|
|
"epoch": 5.289975144987572,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002843283670072942,
|
|
"loss": 4.924,
|
|
"mean_token_accuracy": 0.20615437030792236,
|
|
"num_tokens": 11636954.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.501015090942383,
|
|
"epoch": 5.294117647058823,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002840090762496268,
|
|
"loss": 4.8583,
|
|
"mean_token_accuracy": 0.21323447078466415,
|
|
"num_tokens": 11646719.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.488340902328491,
|
|
"epoch": 5.2982601491300745,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002836897673196681,
|
|
"loss": 4.764,
|
|
"mean_token_accuracy": 0.21574227064847945,
|
|
"num_tokens": 11655589.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.467871379852295,
|
|
"epoch": 5.302402651201326,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00028337044086149913,
|
|
"loss": 4.8033,
|
|
"mean_token_accuracy": 0.21337209790945053,
|
|
"num_tokens": 11663703.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.534906387329102,
|
|
"epoch": 5.306545153272577,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028305109751923616,
|
|
"loss": 4.8922,
|
|
"mean_token_accuracy": 0.21264195591211318,
|
|
"num_tokens": 11672643.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.483319234848023,
|
|
"epoch": 5.310687655343828,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00028273173793702967,
|
|
"loss": 4.7531,
|
|
"mean_token_accuracy": 0.21593598425388336,
|
|
"num_tokens": 11681449.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.501971292495727,
|
|
"epoch": 5.314830157415079,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002824123627590629,
|
|
"loss": 4.8222,
|
|
"mean_token_accuracy": 0.20788242667913437,
|
|
"num_tokens": 11690820.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.488041019439697,
|
|
"epoch": 5.31897265948633,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002820929726295504,
|
|
"loss": 4.8707,
|
|
"mean_token_accuracy": 0.21694955825805665,
|
|
"num_tokens": 11699705.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.489735794067383,
|
|
"epoch": 5.323115161557581,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002817735681927371,
|
|
"loss": 4.8146,
|
|
"mean_token_accuracy": 0.22264870256185532,
|
|
"num_tokens": 11709417.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.531863689422607,
|
|
"epoch": 5.327257663628831,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00028145415009289644,
|
|
"loss": 4.8264,
|
|
"mean_token_accuracy": 0.2141342505812645,
|
|
"num_tokens": 11718661.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.465619039535523,
|
|
"epoch": 5.3314001657000825,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002811347189743297,
|
|
"loss": 4.828,
|
|
"mean_token_accuracy": 0.21483712196350097,
|
|
"num_tokens": 11727568.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.476656818389893,
|
|
"epoch": 5.335542667771334,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028081527548136446,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.22993845045566558,
|
|
"num_tokens": 11736615.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.564050245285034,
|
|
"epoch": 5.339685169842585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002804958202583531,
|
|
"loss": 4.8794,
|
|
"mean_token_accuracy": 0.20408553779125213,
|
|
"num_tokens": 11747458.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.506606101989746,
|
|
"epoch": 5.343827671913836,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002801763539496718,
|
|
"loss": 4.8651,
|
|
"mean_token_accuracy": 0.2152437835931778,
|
|
"num_tokens": 11756279.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.4696571826934814,
|
|
"epoch": 5.347970173985087,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002798568771997188,
|
|
"loss": 4.8431,
|
|
"mean_token_accuracy": 0.2145977035164833,
|
|
"num_tokens": 11765022.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.409263658523559,
|
|
"epoch": 5.352112676056338,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002795373906529137,
|
|
"loss": 4.7929,
|
|
"mean_token_accuracy": 0.21652219742536544,
|
|
"num_tokens": 11774704.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.485758352279663,
|
|
"epoch": 5.356255178127589,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002792178949536959,
|
|
"loss": 4.7555,
|
|
"mean_token_accuracy": 0.22120640724897384,
|
|
"num_tokens": 11782309.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.485047292709351,
|
|
"epoch": 5.36039768019884,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000278898390746523,
|
|
"loss": 4.807,
|
|
"mean_token_accuracy": 0.21500315368175507,
|
|
"num_tokens": 11792595.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.436954355239868,
|
|
"epoch": 5.364540182270091,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027857887867587005,
|
|
"loss": 4.8055,
|
|
"mean_token_accuracy": 0.21833099871873857,
|
|
"num_tokens": 11801536.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.5313948631286625,
|
|
"epoch": 5.368682684341342,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002782593593862276,
|
|
"loss": 4.8678,
|
|
"mean_token_accuracy": 0.21031728833913804,
|
|
"num_tokens": 11810222.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.49743161201477,
|
|
"epoch": 5.3728251864125935,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002779398335221011,
|
|
"loss": 4.8813,
|
|
"mean_token_accuracy": 0.2098690688610077,
|
|
"num_tokens": 11819310.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.488875150680542,
|
|
"epoch": 5.376967688483845,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002776203017280092,
|
|
"loss": 4.8311,
|
|
"mean_token_accuracy": 0.2115108385682106,
|
|
"num_tokens": 11827916.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.4597855567932125,
|
|
"epoch": 5.381110190555095,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00027730076464848225,
|
|
"loss": 4.8237,
|
|
"mean_token_accuracy": 0.21887100785970687,
|
|
"num_tokens": 11836770.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.499082231521607,
|
|
"epoch": 5.385252692626346,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027698122292806174,
|
|
"loss": 4.7867,
|
|
"mean_token_accuracy": 0.21705404222011565,
|
|
"num_tokens": 11845917.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.385252692626346,
|
|
"eval_entropy": 5.333525238769141,
|
|
"eval_loss": 5.743383407592773,
|
|
"eval_mean_token_accuracy": 0.1653154025485775,
|
|
"eval_num_tokens": 11845917.0,
|
|
"eval_runtime": 2.1107,
|
|
"eval_samples_per_second": 1775.269,
|
|
"eval_steps_per_second": 222.205,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 5.462417125701904,
|
|
"epoch": 5.389395194697597,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000276661677211298,
|
|
"loss": 4.8087,
|
|
"mean_token_accuracy": 0.21588131189346313,
|
|
"num_tokens": 11855118.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 5.506772184371949,
|
|
"epoch": 5.393537696768848,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002763421281427497,
|
|
"loss": 4.7607,
|
|
"mean_token_accuracy": 0.21787813305854797,
|
|
"num_tokens": 11863793.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 5.602477502822876,
|
|
"epoch": 5.397680198840099,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002760225763669823,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.20351433306932448,
|
|
"num_tokens": 11873194.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.5684339046478275,
|
|
"epoch": 5.40182270091135,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027570302252856666,
|
|
"loss": 4.8375,
|
|
"mean_token_accuracy": 0.2130189910531044,
|
|
"num_tokens": 11882566.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 5.463960313796997,
|
|
"epoch": 5.4059652029826015,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00027538346727207767,
|
|
"loss": 4.8029,
|
|
"mean_token_accuracy": 0.2232893541455269,
|
|
"num_tokens": 11892267.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 5.500541019439697,
|
|
"epoch": 5.410107705053853,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002750639112420934,
|
|
"loss": 4.8242,
|
|
"mean_token_accuracy": 0.21305467635393144,
|
|
"num_tokens": 11901440.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 5.511700582504273,
|
|
"epoch": 5.414250207125104,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00027474435508319304,
|
|
"loss": 4.916,
|
|
"mean_token_accuracy": 0.21176356077194214,
|
|
"num_tokens": 11911259.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 5.48839225769043,
|
|
"epoch": 5.418392709196355,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027442479943995645,
|
|
"loss": 4.8548,
|
|
"mean_token_accuracy": 0.20965074449777604,
|
|
"num_tokens": 11920467.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 5.503031206130982,
|
|
"epoch": 5.422535211267606,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002741052449569624,
|
|
"loss": 4.8714,
|
|
"mean_token_accuracy": 0.21195811182260513,
|
|
"num_tokens": 11929971.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 5.515838956832885,
|
|
"epoch": 5.426677713338857,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002737856922787869,
|
|
"loss": 4.8743,
|
|
"mean_token_accuracy": 0.20683438181877137,
|
|
"num_tokens": 11938585.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 5.48805422782898,
|
|
"epoch": 5.430820215410108,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027346614205000277,
|
|
"loss": 4.8651,
|
|
"mean_token_accuracy": 0.20904596894979477,
|
|
"num_tokens": 11948264.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 5.503715705871582,
|
|
"epoch": 5.434962717481358,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027314659491517777,
|
|
"loss": 4.8676,
|
|
"mean_token_accuracy": 0.2091337963938713,
|
|
"num_tokens": 11958459.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 5.481890726089477,
|
|
"epoch": 5.4391052195526095,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002728270515188733,
|
|
"loss": 4.8349,
|
|
"mean_token_accuracy": 0.2113039717078209,
|
|
"num_tokens": 11967994.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 5.582008981704712,
|
|
"epoch": 5.443247721623861,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002725075125056435,
|
|
"loss": 4.8744,
|
|
"mean_token_accuracy": 0.21958480924367904,
|
|
"num_tokens": 11976564.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 5.517498111724853,
|
|
"epoch": 5.447390223695112,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002721879785200333,
|
|
"loss": 4.8173,
|
|
"mean_token_accuracy": 0.22103905975818633,
|
|
"num_tokens": 11985682.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 5.535716772079468,
|
|
"epoch": 5.451532725766363,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002718684502065777,
|
|
"loss": 4.888,
|
|
"mean_token_accuracy": 0.21553197652101516,
|
|
"num_tokens": 11995098.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 5.53657169342041,
|
|
"epoch": 5.455675227837614,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027154892820980024,
|
|
"loss": 4.875,
|
|
"mean_token_accuracy": 0.2154003009200096,
|
|
"num_tokens": 12004538.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 5.461729049682617,
|
|
"epoch": 5.459817729908865,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027122941317421175,
|
|
"loss": 4.8144,
|
|
"mean_token_accuracy": 0.2193264588713646,
|
|
"num_tokens": 12012568.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 5.49633378982544,
|
|
"epoch": 5.463960231980116,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000270909905744309,
|
|
"loss": 4.8485,
|
|
"mean_token_accuracy": 0.22042151540517807,
|
|
"num_tokens": 12021934.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 5.538760137557984,
|
|
"epoch": 5.468102734051367,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00027059040656457316,
|
|
"loss": 4.8603,
|
|
"mean_token_accuracy": 0.2072578638792038,
|
|
"num_tokens": 12031158.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 5.584251928329468,
|
|
"epoch": 5.472245236122618,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002702709162794692,
|
|
"loss": 4.875,
|
|
"mean_token_accuracy": 0.2107881337404251,
|
|
"num_tokens": 12040392.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 5.482973337173462,
|
|
"epoch": 5.4763877381938695,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026995143553344394,
|
|
"loss": 4.7353,
|
|
"mean_token_accuracy": 0.22332855463027954,
|
|
"num_tokens": 12048412.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 5.466313409805298,
|
|
"epoch": 5.48053024026512,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00026963196497092483,
|
|
"loss": 4.8213,
|
|
"mean_token_accuracy": 0.21503878980875016,
|
|
"num_tokens": 12057681.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 5.498620176315308,
|
|
"epoch": 5.484672742336371,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00026931250523631905,
|
|
"loss": 4.8156,
|
|
"mean_token_accuracy": 0.21374430507421494,
|
|
"num_tokens": 12066243.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 5.43012318611145,
|
|
"epoch": 5.488815244407622,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026899305697401175,
|
|
"loss": 4.8238,
|
|
"mean_token_accuracy": 0.21214921921491622,
|
|
"num_tokens": 12075312.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 5.459860706329346,
|
|
"epoch": 5.492957746478873,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002686736208283649,
|
|
"loss": 4.8284,
|
|
"mean_token_accuracy": 0.2131899908185005,
|
|
"num_tokens": 12083990.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 5.6119121551513675,
|
|
"epoch": 5.497100248550124,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00026835419744371627,
|
|
"loss": 4.9403,
|
|
"mean_token_accuracy": 0.20740652680397034,
|
|
"num_tokens": 12093301.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 5.48863639831543,
|
|
"epoch": 5.501242750621375,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002680347874643775,
|
|
"loss": 4.777,
|
|
"mean_token_accuracy": 0.22249195277690886,
|
|
"num_tokens": 12102054.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 5.459077930450439,
|
|
"epoch": 5.505385252692626,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002677153915346336,
|
|
"loss": 4.8368,
|
|
"mean_token_accuracy": 0.21560850292444228,
|
|
"num_tokens": 12111432.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 5.432029294967651,
|
|
"epoch": 5.5095277547638775,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002673960102987412,
|
|
"loss": 4.7855,
|
|
"mean_token_accuracy": 0.22020294070243834,
|
|
"num_tokens": 12120424.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 5.487680244445801,
|
|
"epoch": 5.513670256835129,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026707664440092676,
|
|
"loss": 4.9132,
|
|
"mean_token_accuracy": 0.21376733481884003,
|
|
"num_tokens": 12130210.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 5.625579452514648,
|
|
"epoch": 5.51781275890638,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026675729448538655,
|
|
"loss": 4.9222,
|
|
"mean_token_accuracy": 0.2186216413974762,
|
|
"num_tokens": 12139366.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 5.474394273757935,
|
|
"epoch": 5.521955260977631,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026643796119628404,
|
|
"loss": 4.7145,
|
|
"mean_token_accuracy": 0.22230644077062606,
|
|
"num_tokens": 12148385.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 5.556215000152588,
|
|
"epoch": 5.526097763048882,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002661186451777495,
|
|
"loss": 4.85,
|
|
"mean_token_accuracy": 0.21862132996320724,
|
|
"num_tokens": 12156640.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 5.536400318145752,
|
|
"epoch": 5.530240265120133,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002657993470738781,
|
|
"loss": 4.9425,
|
|
"mean_token_accuracy": 0.20092385858297349,
|
|
"num_tokens": 12166532.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 5.442637586593628,
|
|
"epoch": 5.534382767191383,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026548006752872915,
|
|
"loss": 4.7666,
|
|
"mean_token_accuracy": 0.22753115743398666,
|
|
"num_tokens": 12175239.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 5.425423574447632,
|
|
"epoch": 5.538525269262634,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002651608071863244,
|
|
"loss": 4.8197,
|
|
"mean_token_accuracy": 0.21672474443912507,
|
|
"num_tokens": 12184588.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 5.548989629745483,
|
|
"epoch": 5.542667771333885,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026484156669064665,
|
|
"loss": 4.8847,
|
|
"mean_token_accuracy": 0.20640380829572677,
|
|
"num_tokens": 12194140.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.535123872756958,
|
|
"epoch": 5.5468102734051365,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000264522346685639,
|
|
"loss": 4.8828,
|
|
"mean_token_accuracy": 0.21611355096101761,
|
|
"num_tokens": 12203504.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 5.5147271156311035,
|
|
"epoch": 5.550952775476388,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002642031478152031,
|
|
"loss": 4.8391,
|
|
"mean_token_accuracy": 0.2181399032473564,
|
|
"num_tokens": 12212594.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 5.505739402770996,
|
|
"epoch": 5.555095277547639,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026388397072319786,
|
|
"loss": 4.8878,
|
|
"mean_token_accuracy": 0.2140798255801201,
|
|
"num_tokens": 12222101.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 5.558458852767944,
|
|
"epoch": 5.55923777961889,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002635648160534385,
|
|
"loss": 4.8929,
|
|
"mean_token_accuracy": 0.2096651092171669,
|
|
"num_tokens": 12232118.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 5.573447179794312,
|
|
"epoch": 5.563380281690141,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026324568444969473,
|
|
"loss": 4.8123,
|
|
"mean_token_accuracy": 0.21913018226623535,
|
|
"num_tokens": 12241275.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 5.469147872924805,
|
|
"epoch": 5.567522783761392,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026292657655569,
|
|
"loss": 4.8162,
|
|
"mean_token_accuracy": 0.21792408227920532,
|
|
"num_tokens": 12251125.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 5.442272472381592,
|
|
"epoch": 5.571665285832643,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026260749301509954,
|
|
"loss": 4.8097,
|
|
"mean_token_accuracy": 0.21625008136034013,
|
|
"num_tokens": 12260613.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 5.556268548965454,
|
|
"epoch": 5.575807787903894,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026228843447155005,
|
|
"loss": 4.8118,
|
|
"mean_token_accuracy": 0.22149630039930343,
|
|
"num_tokens": 12269878.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 5.480302190780639,
|
|
"epoch": 5.5799502899751445,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00026196940156861736,
|
|
"loss": 4.8626,
|
|
"mean_token_accuracy": 0.2077437236905098,
|
|
"num_tokens": 12278616.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 5.401016473770142,
|
|
"epoch": 5.584092792046396,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026165039494982555,
|
|
"loss": 4.7936,
|
|
"mean_token_accuracy": 0.21300697475671768,
|
|
"num_tokens": 12287104.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 5.399640226364136,
|
|
"epoch": 5.588235294117647,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026133141525864605,
|
|
"loss": 4.7914,
|
|
"mean_token_accuracy": 0.21519204676151277,
|
|
"num_tokens": 12296503.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 5.477622556686401,
|
|
"epoch": 5.592377796188898,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026101246313849557,
|
|
"loss": 4.7665,
|
|
"mean_token_accuracy": 0.21981689482927322,
|
|
"num_tokens": 12306298.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 5.498661088943481,
|
|
"epoch": 5.596520298260149,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002606935392327356,
|
|
"loss": 4.834,
|
|
"mean_token_accuracy": 0.2095463752746582,
|
|
"num_tokens": 12314826.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 5.492101812362671,
|
|
"epoch": 5.6006628003314,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026037464418467035,
|
|
"loss": 4.7773,
|
|
"mean_token_accuracy": 0.21948902159929276,
|
|
"num_tokens": 12323776.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 5.53430495262146,
|
|
"epoch": 5.604805302402651,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002600557786375459,
|
|
"loss": 4.8693,
|
|
"mean_token_accuracy": 0.21544408053159714,
|
|
"num_tokens": 12332374.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 5.4284251689910885,
|
|
"epoch": 5.608947804473902,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00025973694323454916,
|
|
"loss": 4.7535,
|
|
"mean_token_accuracy": 0.22376630455255508,
|
|
"num_tokens": 12341007.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 5.4040874481201175,
|
|
"epoch": 5.613090306545153,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00025941813861880575,
|
|
"loss": 4.7702,
|
|
"mean_token_accuracy": 0.22251395881175995,
|
|
"num_tokens": 12350330.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 5.495104074478149,
|
|
"epoch": 5.6172328086164045,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025909936543337957,
|
|
"loss": 4.8266,
|
|
"mean_token_accuracy": 0.20475833863019943,
|
|
"num_tokens": 12358592.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 5.53274941444397,
|
|
"epoch": 5.621375310687656,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002587806243212709,
|
|
"loss": 4.8493,
|
|
"mean_token_accuracy": 0.2120782643556595,
|
|
"num_tokens": 12368195.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 5.472455644607544,
|
|
"epoch": 5.625517812758907,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002584619159254154,
|
|
"loss": 4.7734,
|
|
"mean_token_accuracy": 0.21647787392139434,
|
|
"num_tokens": 12377206.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 5.43663330078125,
|
|
"epoch": 5.629660314830158,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025814324088868276,
|
|
"loss": 4.8403,
|
|
"mean_token_accuracy": 0.21724140793085098,
|
|
"num_tokens": 12386595.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 5.430670499801636,
|
|
"epoch": 5.633802816901408,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025782459985387526,
|
|
"loss": 4.7265,
|
|
"mean_token_accuracy": 0.22419483363628387,
|
|
"num_tokens": 12395923.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 5.5203405857086185,
|
|
"epoch": 5.637945318972659,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002575059934637268,
|
|
"loss": 4.8865,
|
|
"mean_token_accuracy": 0.21158853322267532,
|
|
"num_tokens": 12405094.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 5.4943098545074465,
|
|
"epoch": 5.64208782104391,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002571874223609012,
|
|
"loss": 4.8755,
|
|
"mean_token_accuracy": 0.21089241653680801,
|
|
"num_tokens": 12414834.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 5.507603025436401,
|
|
"epoch": 5.646230323115161,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025686888718799136,
|
|
"loss": 4.823,
|
|
"mean_token_accuracy": 0.2115618422627449,
|
|
"num_tokens": 12423485.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 5.565799236297607,
|
|
"epoch": 5.6503728251864125,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002565503885875173,
|
|
"loss": 4.869,
|
|
"mean_token_accuracy": 0.21416962444782256,
|
|
"num_tokens": 12432149.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 5.512131118774414,
|
|
"epoch": 5.654515327257664,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00025623192720192564,
|
|
"loss": 4.8274,
|
|
"mean_token_accuracy": 0.21803111284971238,
|
|
"num_tokens": 12441272.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 5.460856962203979,
|
|
"epoch": 5.658657829328915,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002559135036735877,
|
|
"loss": 4.8054,
|
|
"mean_token_accuracy": 0.21020345240831376,
|
|
"num_tokens": 12449318.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 5.413202619552612,
|
|
"epoch": 5.662800331400166,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00025559511864479864,
|
|
"loss": 4.7522,
|
|
"mean_token_accuracy": 0.22357650697231293,
|
|
"num_tokens": 12458473.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 5.512956333160401,
|
|
"epoch": 5.666942833471417,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002552767727577758,
|
|
"loss": 4.8934,
|
|
"mean_token_accuracy": 0.2064364641904831,
|
|
"num_tokens": 12467703.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 5.513271951675415,
|
|
"epoch": 5.671085335542668,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00025495846665465774,
|
|
"loss": 4.7627,
|
|
"mean_token_accuracy": 0.22760390341281891,
|
|
"num_tokens": 12476166.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 5.411068725585937,
|
|
"epoch": 5.675227837613919,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002546402009775024,
|
|
"loss": 4.8029,
|
|
"mean_token_accuracy": 0.21550437957048416,
|
|
"num_tokens": 12484880.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 5.520448160171509,
|
|
"epoch": 5.679370339685169,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025432197636828663,
|
|
"loss": 4.8915,
|
|
"mean_token_accuracy": 0.2122902661561966,
|
|
"num_tokens": 12494208.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 5.541320848464966,
|
|
"epoch": 5.683512841756421,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002540037934689042,
|
|
"loss": 4.8777,
|
|
"mean_token_accuracy": 0.2063458487391472,
|
|
"num_tokens": 12503647.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 5.493671321868897,
|
|
"epoch": 5.687655343827672,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00025368565292116475,
|
|
"loss": 4.8901,
|
|
"mean_token_accuracy": 0.2099316820502281,
|
|
"num_tokens": 12512869.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 5.486989212036133,
|
|
"epoch": 5.691797845898923,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002533675553667927,
|
|
"loss": 4.9453,
|
|
"mean_token_accuracy": 0.199296210706234,
|
|
"num_tokens": 12523222.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 5.461559200286866,
|
|
"epoch": 5.695940347970174,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00025304950144742523,
|
|
"loss": 4.8183,
|
|
"mean_token_accuracy": 0.21540112048387527,
|
|
"num_tokens": 12532343.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 5.481991910934449,
|
|
"epoch": 5.700082850041425,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002527314918046122,
|
|
"loss": 4.8235,
|
|
"mean_token_accuracy": 0.22275232672691345,
|
|
"num_tokens": 12541508.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.460350370407104,
|
|
"epoch": 5.704225352112676,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002524135270798136,
|
|
"loss": 4.8645,
|
|
"mean_token_accuracy": 0.20687949657440186,
|
|
"num_tokens": 12550837.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 5.495569801330566,
|
|
"epoch": 5.708367854183927,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025209560791439914,
|
|
"loss": 4.7849,
|
|
"mean_token_accuracy": 0.22157662063837053,
|
|
"num_tokens": 12559859.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 5.48965630531311,
|
|
"epoch": 5.712510356255178,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002517777349496465,
|
|
"loss": 4.8797,
|
|
"mean_token_accuracy": 0.2130148306488991,
|
|
"num_tokens": 12569137.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 5.535360336303711,
|
|
"epoch": 5.716652858326429,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002514599088267401,
|
|
"loss": 4.8829,
|
|
"mean_token_accuracy": 0.2087639316916466,
|
|
"num_tokens": 12578111.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 5.531719064712524,
|
|
"epoch": 5.72079536039768,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025114213018677004,
|
|
"loss": 4.8396,
|
|
"mean_token_accuracy": 0.21515557914972305,
|
|
"num_tokens": 12587604.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 5.535152864456177,
|
|
"epoch": 5.7249378624689315,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002508243996707305,
|
|
"loss": 4.8796,
|
|
"mean_token_accuracy": 0.20691268742084504,
|
|
"num_tokens": 12596620.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 5.427631092071533,
|
|
"epoch": 5.729080364540183,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00025050671791951887,
|
|
"loss": 4.8316,
|
|
"mean_token_accuracy": 0.21652291864156722,
|
|
"num_tokens": 12605452.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 5.396881818771362,
|
|
"epoch": 5.733222866611433,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002501890855739337,
|
|
"loss": 4.7379,
|
|
"mean_token_accuracy": 0.22244983464479445,
|
|
"num_tokens": 12614009.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 5.510207796096802,
|
|
"epoch": 5.737365368682684,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024987150327467414,
|
|
"loss": 4.874,
|
|
"mean_token_accuracy": 0.2146199882030487,
|
|
"num_tokens": 12624135.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 5.46781849861145,
|
|
"epoch": 5.741507870753935,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00024955397166233855,
|
|
"loss": 4.7342,
|
|
"mean_token_accuracy": 0.22886782437562941,
|
|
"num_tokens": 12632870.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 5.511617422103882,
|
|
"epoch": 5.745650372825186,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00024923649137742276,
|
|
"loss": 4.9179,
|
|
"mean_token_accuracy": 0.20559141635894776,
|
|
"num_tokens": 12643606.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 5.562722110748291,
|
|
"epoch": 5.749792874896437,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002489190630603194,
|
|
"loss": 4.9668,
|
|
"mean_token_accuracy": 0.2096701055765152,
|
|
"num_tokens": 12652926.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 5.54264121055603,
|
|
"epoch": 5.753935376967688,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002486016873513159,
|
|
"loss": 4.8925,
|
|
"mean_token_accuracy": 0.21575085818767548,
|
|
"num_tokens": 12661841.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.46075291633606,
|
|
"epoch": 5.7580778790389395,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002482843648905937,
|
|
"loss": 4.8244,
|
|
"mean_token_accuracy": 0.2096736326813698,
|
|
"num_tokens": 12670527.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 5.45245246887207,
|
|
"epoch": 5.762220381110191,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002479670963182271,
|
|
"loss": 4.8107,
|
|
"mean_token_accuracy": 0.20710556954145432,
|
|
"num_tokens": 12680346.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 5.5696563720703125,
|
|
"epoch": 5.766362883181442,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024764988227418114,
|
|
"loss": 4.9159,
|
|
"mean_token_accuracy": 0.21850883811712266,
|
|
"num_tokens": 12691088.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 5.563732290267945,
|
|
"epoch": 5.770505385252693,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002473327233983116,
|
|
"loss": 4.8554,
|
|
"mean_token_accuracy": 0.21547263711690903,
|
|
"num_tokens": 12700296.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 5.421257543563843,
|
|
"epoch": 5.774647887323944,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002470156203303622,
|
|
"loss": 4.857,
|
|
"mean_token_accuracy": 0.2154713451862335,
|
|
"num_tokens": 12709595.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 5.484141778945923,
|
|
"epoch": 5.778790389395195,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002466985737099649,
|
|
"loss": 4.902,
|
|
"mean_token_accuracy": 0.21215634793043137,
|
|
"num_tokens": 12719052.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 5.4773119449615475,
|
|
"epoch": 5.782932891466446,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002463815841766371,
|
|
"loss": 4.7678,
|
|
"mean_token_accuracy": 0.22186898738145827,
|
|
"num_tokens": 12727287.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 5.483669662475586,
|
|
"epoch": 5.787075393537696,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002460646523697813,
|
|
"loss": 4.8269,
|
|
"mean_token_accuracy": 0.21340495645999907,
|
|
"num_tokens": 12736511.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 5.557035779953003,
|
|
"epoch": 5.7912178956089475,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024574777892868386,
|
|
"loss": 4.8588,
|
|
"mean_token_accuracy": 0.2163700729608536,
|
|
"num_tokens": 12745597.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 5.4778656482696535,
|
|
"epoch": 5.795360397680199,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024543096449251286,
|
|
"loss": 4.7497,
|
|
"mean_token_accuracy": 0.22163933962583543,
|
|
"num_tokens": 12753492.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 5.5171465396881105,
|
|
"epoch": 5.79950289975145,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002451142097003177,
|
|
"loss": 4.8472,
|
|
"mean_token_accuracy": 0.21241547018289567,
|
|
"num_tokens": 12761999.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.79950289975145,
|
|
"eval_entropy": 5.323617148246846,
|
|
"eval_loss": 5.724344253540039,
|
|
"eval_mean_token_accuracy": 0.16743344090767762,
|
|
"eval_num_tokens": 12761999.0,
|
|
"eval_runtime": 2.1098,
|
|
"eval_samples_per_second": 1776.02,
|
|
"eval_steps_per_second": 222.299,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 5.509900617599487,
|
|
"epoch": 5.803645401822701,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002447975151910276,
|
|
"loss": 4.7851,
|
|
"mean_token_accuracy": 0.21556495726108552,
|
|
"num_tokens": 12770922.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.428083658218384,
|
|
"epoch": 5.807787903893952,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024448088160344974,
|
|
"loss": 4.7956,
|
|
"mean_token_accuracy": 0.21692176163196564,
|
|
"num_tokens": 12779539.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 5.46669373512268,
|
|
"epoch": 5.811930405965203,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002441643095762688,
|
|
"loss": 4.7937,
|
|
"mean_token_accuracy": 0.22022989243268967,
|
|
"num_tokens": 12788121.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 5.410281991958618,
|
|
"epoch": 5.816072908036454,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000243847799748045,
|
|
"loss": 4.7041,
|
|
"mean_token_accuracy": 0.22062321454286576,
|
|
"num_tokens": 12797390.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 5.458677530288696,
|
|
"epoch": 5.820215410107705,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00024353135275721356,
|
|
"loss": 4.9262,
|
|
"mean_token_accuracy": 0.2066818505525589,
|
|
"num_tokens": 12807876.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 5.426154088973999,
|
|
"epoch": 5.824357912178956,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024321496924208235,
|
|
"loss": 4.7512,
|
|
"mean_token_accuracy": 0.2136765480041504,
|
|
"num_tokens": 12816496.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 5.4880571365356445,
|
|
"epoch": 5.8285004142502075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00024289864984083166,
|
|
"loss": 4.8379,
|
|
"mean_token_accuracy": 0.218058480322361,
|
|
"num_tokens": 12825925.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 5.572530698776245,
|
|
"epoch": 5.832642916321458,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002425823951915122,
|
|
"loss": 4.86,
|
|
"mean_token_accuracy": 0.21090638041496276,
|
|
"num_tokens": 12835479.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 5.450262069702148,
|
|
"epoch": 5.836785418392709,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002422662059320442,
|
|
"loss": 4.7482,
|
|
"mean_token_accuracy": 0.22353963553905487,
|
|
"num_tokens": 12844024.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 5.452236270904541,
|
|
"epoch": 5.84092792046396,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024195008270021595,
|
|
"loss": 4.7697,
|
|
"mean_token_accuracy": 0.2159978985786438,
|
|
"num_tokens": 12852673.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 5.422987794876098,
|
|
"epoch": 5.845070422535211,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024163402613368252,
|
|
"loss": 4.7578,
|
|
"mean_token_accuracy": 0.2247852623462677,
|
|
"num_tokens": 12861191.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 5.520094347000122,
|
|
"epoch": 5.849212924606462,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024131803686996457,
|
|
"loss": 4.9277,
|
|
"mean_token_accuracy": 0.20994001179933547,
|
|
"num_tokens": 12870874.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 5.516187620162964,
|
|
"epoch": 5.853355426677713,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00024100211554644696,
|
|
"loss": 4.8484,
|
|
"mean_token_accuracy": 0.21837085336446763,
|
|
"num_tokens": 12879896.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 5.504858016967773,
|
|
"epoch": 5.857497928748964,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024068626280037753,
|
|
"loss": 4.86,
|
|
"mean_token_accuracy": 0.21108729243278504,
|
|
"num_tokens": 12888665.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 5.455237245559692,
|
|
"epoch": 5.8616404308202155,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024037047926886578,
|
|
"loss": 4.8124,
|
|
"mean_token_accuracy": 0.21359451115131378,
|
|
"num_tokens": 12897477.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 5.45536789894104,
|
|
"epoch": 5.865782932891467,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00024005476558888172,
|
|
"loss": 4.8778,
|
|
"mean_token_accuracy": 0.20298874974250794,
|
|
"num_tokens": 12905934.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 5.491654014587402,
|
|
"epoch": 5.869925434962718,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002397391223972541,
|
|
"loss": 4.8626,
|
|
"mean_token_accuracy": 0.20993417352437974,
|
|
"num_tokens": 12915092.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 5.50332293510437,
|
|
"epoch": 5.874067937033969,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023942355033066993,
|
|
"loss": 4.9306,
|
|
"mean_token_accuracy": 0.21204065084457396,
|
|
"num_tokens": 12924775.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 5.489796209335327,
|
|
"epoch": 5.87821043910522,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023910805002567244,
|
|
"loss": 4.853,
|
|
"mean_token_accuracy": 0.20865764021873473,
|
|
"num_tokens": 12933677.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 5.5174377918243405,
|
|
"epoch": 5.882352941176471,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002387926221186603,
|
|
"loss": 4.8742,
|
|
"mean_token_accuracy": 0.20817747563123704,
|
|
"num_tokens": 12942416.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 5.48937726020813,
|
|
"epoch": 5.886495443247721,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023847726724588608,
|
|
"loss": 4.8657,
|
|
"mean_token_accuracy": 0.21319568604230882,
|
|
"num_tokens": 12952284.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 5.496838855743408,
|
|
"epoch": 5.890637945318972,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002381619860434549,
|
|
"loss": 4.8908,
|
|
"mean_token_accuracy": 0.20573581010103226,
|
|
"num_tokens": 12960694.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 5.39988260269165,
|
|
"epoch": 5.894780447390223,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023784677914732356,
|
|
"loss": 4.8139,
|
|
"mean_token_accuracy": 0.2149551033973694,
|
|
"num_tokens": 12969138.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 5.468914890289307,
|
|
"epoch": 5.8989229494614746,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023753164719329868,
|
|
"loss": 4.838,
|
|
"mean_token_accuracy": 0.21042930483818054,
|
|
"num_tokens": 12978518.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 5.512823867797851,
|
|
"epoch": 5.903065451532726,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023721659081703595,
|
|
"loss": 4.8704,
|
|
"mean_token_accuracy": 0.21749993711709975,
|
|
"num_tokens": 12987506.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 5.499007701873779,
|
|
"epoch": 5.907207953603977,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023690161065403838,
|
|
"loss": 4.8558,
|
|
"mean_token_accuracy": 0.20485332012176513,
|
|
"num_tokens": 12996259.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 5.520441961288452,
|
|
"epoch": 5.911350455675228,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002365867073396555,
|
|
"loss": 4.8758,
|
|
"mean_token_accuracy": 0.207315531373024,
|
|
"num_tokens": 13005430.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 5.476233386993409,
|
|
"epoch": 5.915492957746479,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023627188150908174,
|
|
"loss": 4.822,
|
|
"mean_token_accuracy": 0.2248324304819107,
|
|
"num_tokens": 13014915.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 5.453920078277588,
|
|
"epoch": 5.91963545981773,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023595713379735494,
|
|
"loss": 4.8191,
|
|
"mean_token_accuracy": 0.21181789487600328,
|
|
"num_tokens": 13023471.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 5.516880321502685,
|
|
"epoch": 5.923777961888981,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023564246483935602,
|
|
"loss": 4.8205,
|
|
"mean_token_accuracy": 0.21136183738708497,
|
|
"num_tokens": 13033270.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 5.503847694396972,
|
|
"epoch": 5.927920463960232,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002353278752698063,
|
|
"loss": 4.7797,
|
|
"mean_token_accuracy": 0.21452480107545852,
|
|
"num_tokens": 13043300.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 5.494083976745605,
|
|
"epoch": 5.9320629660314825,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002350133657232676,
|
|
"loss": 4.8569,
|
|
"mean_token_accuracy": 0.21026458144187926,
|
|
"num_tokens": 13052152.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 5.48362021446228,
|
|
"epoch": 5.9362054681027345,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002346989368341399,
|
|
"loss": 4.8835,
|
|
"mean_token_accuracy": 0.21113257110118866,
|
|
"num_tokens": 13061704.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 5.506317615509033,
|
|
"epoch": 5.940347970173985,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00023438458923666057,
|
|
"loss": 4.8109,
|
|
"mean_token_accuracy": 0.22241746187210082,
|
|
"num_tokens": 13071055.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 5.468820238113404,
|
|
"epoch": 5.944490472245236,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002340703235649032,
|
|
"loss": 4.8981,
|
|
"mean_token_accuracy": 0.20368586033582686,
|
|
"num_tokens": 13080023.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 5.526552343368531,
|
|
"epoch": 5.948632974316487,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023375614045277593,
|
|
"loss": 4.8164,
|
|
"mean_token_accuracy": 0.21295997947454454,
|
|
"num_tokens": 13088690.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 5.4384277820587155,
|
|
"epoch": 5.952775476387738,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023344204053402047,
|
|
"loss": 4.8154,
|
|
"mean_token_accuracy": 0.2159508615732193,
|
|
"num_tokens": 13097116.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 5.534406757354736,
|
|
"epoch": 5.956917978458989,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023312802444221066,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.2157169297337532,
|
|
"num_tokens": 13106463.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 5.489149951934815,
|
|
"epoch": 5.96106048053024,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002328140928107511,
|
|
"loss": 4.7938,
|
|
"mean_token_accuracy": 0.21611494868993758,
|
|
"num_tokens": 13115769.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 5.557583236694336,
|
|
"epoch": 5.965202982601491,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002325002462728766,
|
|
"loss": 4.8971,
|
|
"mean_token_accuracy": 0.2161348953843117,
|
|
"num_tokens": 13124928.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 5.492957210540771,
|
|
"epoch": 5.9693454846727425,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023218648546164945,
|
|
"loss": 4.8121,
|
|
"mean_token_accuracy": 0.2155376672744751,
|
|
"num_tokens": 13134106.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 5.632611703872681,
|
|
"epoch": 5.973487986743994,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002318728110099599,
|
|
"loss": 5.0229,
|
|
"mean_token_accuracy": 0.20594469904899598,
|
|
"num_tokens": 13143019.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 5.547437715530395,
|
|
"epoch": 5.977630488815245,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.00023155922355052339,
|
|
"loss": 4.781,
|
|
"mean_token_accuracy": 0.23284861147403718,
|
|
"num_tokens": 13152675.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 5.49273452758789,
|
|
"epoch": 5.981772990886496,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023124572371588,
|
|
"loss": 4.801,
|
|
"mean_token_accuracy": 0.2149716079235077,
|
|
"num_tokens": 13161094.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 5.430967712402344,
|
|
"epoch": 5.985915492957746,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023093231213839344,
|
|
"loss": 4.8617,
|
|
"mean_token_accuracy": 0.2128844678401947,
|
|
"num_tokens": 13170858.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 5.463623666763306,
|
|
"epoch": 5.990057995028997,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002306189894502488,
|
|
"loss": 4.9458,
|
|
"mean_token_accuracy": 0.19705432057380676,
|
|
"num_tokens": 13179668.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 5.515454387664795,
|
|
"epoch": 5.994200497100248,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023030575628345246,
|
|
"loss": 4.7877,
|
|
"mean_token_accuracy": 0.21718480288982392,
|
|
"num_tokens": 13188312.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 5.468227815628052,
|
|
"epoch": 5.998342999171499,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022999261326982985,
|
|
"loss": 4.8308,
|
|
"mean_token_accuracy": 0.21415364146232604,
|
|
"num_tokens": 13196989.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 5.459630537033081,
|
|
"epoch": 6.0024855012427505,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022967956104102445,
|
|
"loss": 4.7274,
|
|
"mean_token_accuracy": 0.22485609948635102,
|
|
"num_tokens": 13205830.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 5.562932920455933,
|
|
"epoch": 6.006628003314002,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022936660022849714,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.23390730619430541,
|
|
"num_tokens": 13214711.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 5.486841011047363,
|
|
"epoch": 6.010770505385253,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022905373146352382,
|
|
"loss": 4.7412,
|
|
"mean_token_accuracy": 0.22181791365146636,
|
|
"num_tokens": 13223736.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 5.474657583236694,
|
|
"epoch": 6.014913007456504,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002287409553771952,
|
|
"loss": 4.676,
|
|
"mean_token_accuracy": 0.2246609553694725,
|
|
"num_tokens": 13232666.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 5.472076463699341,
|
|
"epoch": 6.019055509527755,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022842827260041455,
|
|
"loss": 4.7131,
|
|
"mean_token_accuracy": 0.21984125524759293,
|
|
"num_tokens": 13241770.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 5.49608588218689,
|
|
"epoch": 6.023198011599006,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022811568376389745,
|
|
"loss": 4.6444,
|
|
"mean_token_accuracy": 0.2331205502152443,
|
|
"num_tokens": 13250475.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 5.446823310852051,
|
|
"epoch": 6.027340513670257,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022780318949816963,
|
|
"loss": 4.6178,
|
|
"mean_token_accuracy": 0.22951310873031616,
|
|
"num_tokens": 13259587.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 5.4599121570587155,
|
|
"epoch": 6.031483015741508,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022749079043356623,
|
|
"loss": 4.6356,
|
|
"mean_token_accuracy": 0.24199324250221252,
|
|
"num_tokens": 13269681.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 5.42302885055542,
|
|
"epoch": 6.0356255178127585,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00022717848720023022,
|
|
"loss": 4.6471,
|
|
"mean_token_accuracy": 0.23642077147960663,
|
|
"num_tokens": 13278186.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 5.442330026626587,
|
|
"epoch": 6.03976801988401,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022686628042811142,
|
|
"loss": 4.7348,
|
|
"mean_token_accuracy": 0.22516871839761735,
|
|
"num_tokens": 13287194.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 5.536360216140747,
|
|
"epoch": 6.043910521955261,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000226554170746965,
|
|
"loss": 4.7833,
|
|
"mean_token_accuracy": 0.21911916136741638,
|
|
"num_tokens": 13296774.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 5.563199520111084,
|
|
"epoch": 6.048053024026512,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022624215878635023,
|
|
"loss": 4.6997,
|
|
"mean_token_accuracy": 0.22683680057525635,
|
|
"num_tokens": 13306368.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 5.560784149169922,
|
|
"epoch": 6.052195526097763,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022593024517562945,
|
|
"loss": 4.6891,
|
|
"mean_token_accuracy": 0.2304449200630188,
|
|
"num_tokens": 13315393.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 5.460399389266968,
|
|
"epoch": 6.056338028169014,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022561843054396636,
|
|
"loss": 4.6547,
|
|
"mean_token_accuracy": 0.23610471785068513,
|
|
"num_tokens": 13324384.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 5.469268894195556,
|
|
"epoch": 6.060480530240265,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022530671552032521,
|
|
"loss": 4.7192,
|
|
"mean_token_accuracy": 0.22683825045824052,
|
|
"num_tokens": 13333191.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 5.44218602180481,
|
|
"epoch": 6.064623032311516,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002249951007334693,
|
|
"loss": 4.6978,
|
|
"mean_token_accuracy": 0.22503953129053117,
|
|
"num_tokens": 13342911.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 5.53337984085083,
|
|
"epoch": 6.068765534382767,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022468358681195966,
|
|
"loss": 4.6902,
|
|
"mean_token_accuracy": 0.23153300881385802,
|
|
"num_tokens": 13352464.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 5.503551387786866,
|
|
"epoch": 6.072908036454018,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022437217438415393,
|
|
"loss": 4.6876,
|
|
"mean_token_accuracy": 0.22718945592641832,
|
|
"num_tokens": 13361812.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 5.530497407913208,
|
|
"epoch": 6.0770505385252696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00022406086407820503,
|
|
"loss": 4.7636,
|
|
"mean_token_accuracy": 0.22589852660894394,
|
|
"num_tokens": 13372080.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 5.523541975021362,
|
|
"epoch": 6.081193040596521,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022374965652205975,
|
|
"loss": 4.8005,
|
|
"mean_token_accuracy": 0.2187222346663475,
|
|
"num_tokens": 13380897.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 5.583355236053467,
|
|
"epoch": 6.085335542667772,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022343855234345796,
|
|
"loss": 4.8082,
|
|
"mean_token_accuracy": 0.21540041267871857,
|
|
"num_tokens": 13389885.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 5.461841011047364,
|
|
"epoch": 6.089478044739022,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00022312755216993065,
|
|
"loss": 4.734,
|
|
"mean_token_accuracy": 0.22350315600633622,
|
|
"num_tokens": 13399734.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 5.4164313793182375,
|
|
"epoch": 6.093620546810273,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022281665662879903,
|
|
"loss": 4.6613,
|
|
"mean_token_accuracy": 0.23128788769245148,
|
|
"num_tokens": 13408641.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 5.443401432037353,
|
|
"epoch": 6.097763048881524,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022250586634717362,
|
|
"loss": 4.7118,
|
|
"mean_token_accuracy": 0.22124956250190736,
|
|
"num_tokens": 13418585.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 5.473458194732666,
|
|
"epoch": 6.101905550952775,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022219518195195216,
|
|
"loss": 4.6289,
|
|
"mean_token_accuracy": 0.23267924338579177,
|
|
"num_tokens": 13427897.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 5.376785850524902,
|
|
"epoch": 6.106048053024026,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022188460406981925,
|
|
"loss": 4.609,
|
|
"mean_token_accuracy": 0.23319688886404039,
|
|
"num_tokens": 13437311.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 5.4937516212463375,
|
|
"epoch": 6.1101905550952775,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00022157413332724436,
|
|
"loss": 4.6795,
|
|
"mean_token_accuracy": 0.22955696582794188,
|
|
"num_tokens": 13446359.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 5.508681011199951,
|
|
"epoch": 6.114333057166529,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00022126377035048072,
|
|
"loss": 4.7167,
|
|
"mean_token_accuracy": 0.2286304861307144,
|
|
"num_tokens": 13455302.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 5.6163163661956785,
|
|
"epoch": 6.11847555923778,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002209535157655646,
|
|
"loss": 4.8246,
|
|
"mean_token_accuracy": 0.22315991520881653,
|
|
"num_tokens": 13466391.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 5.500481986999512,
|
|
"epoch": 6.122618061309031,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022064337019831322,
|
|
"loss": 4.6376,
|
|
"mean_token_accuracy": 0.2347148135304451,
|
|
"num_tokens": 13476130.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 5.406926774978638,
|
|
"epoch": 6.126760563380282,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00022033333427432422,
|
|
"loss": 4.6415,
|
|
"mean_token_accuracy": 0.2326324239373207,
|
|
"num_tokens": 13485105.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 5.471008729934693,
|
|
"epoch": 6.130903065451533,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002200234086189738,
|
|
"loss": 4.704,
|
|
"mean_token_accuracy": 0.2203171044588089,
|
|
"num_tokens": 13494914.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 5.4667762279510494,
|
|
"epoch": 6.135045567522784,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00021971359385741608,
|
|
"loss": 4.7316,
|
|
"mean_token_accuracy": 0.22485463321208954,
|
|
"num_tokens": 13504293.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 5.582313823699951,
|
|
"epoch": 6.139188069594034,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021940389061458115,
|
|
"loss": 4.8251,
|
|
"mean_token_accuracy": 0.22064273357391356,
|
|
"num_tokens": 13513972.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 5.503071403503418,
|
|
"epoch": 6.1433305716652855,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002190942995151743,
|
|
"loss": 4.7262,
|
|
"mean_token_accuracy": 0.22276137471199037,
|
|
"num_tokens": 13522508.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 5.532184648513794,
|
|
"epoch": 6.147473073736537,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002187848211836746,
|
|
"loss": 4.6358,
|
|
"mean_token_accuracy": 0.23082863241434098,
|
|
"num_tokens": 13531235.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 5.476139545440674,
|
|
"epoch": 6.151615575807788,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002184754562443337,
|
|
"loss": 4.679,
|
|
"mean_token_accuracy": 0.2319875955581665,
|
|
"num_tokens": 13541023.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 5.435385417938233,
|
|
"epoch": 6.155758077879039,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021816620532117448,
|
|
"loss": 4.6841,
|
|
"mean_token_accuracy": 0.22998475283384323,
|
|
"num_tokens": 13550622.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 5.451135158538818,
|
|
"epoch": 6.15990057995029,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002178570690379899,
|
|
"loss": 4.6985,
|
|
"mean_token_accuracy": 0.22150947898626328,
|
|
"num_tokens": 13559423.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 5.480345249176025,
|
|
"epoch": 6.164043082021541,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021754804801834144,
|
|
"loss": 4.685,
|
|
"mean_token_accuracy": 0.22463978081941605,
|
|
"num_tokens": 13569306.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 5.444636869430542,
|
|
"epoch": 6.168185584092792,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002172391428855585,
|
|
"loss": 4.6563,
|
|
"mean_token_accuracy": 0.23197539746761323,
|
|
"num_tokens": 13578367.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 5.452653074264527,
|
|
"epoch": 6.172328086164043,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021693035426273633,
|
|
"loss": 4.7161,
|
|
"mean_token_accuracy": 0.22137341946363448,
|
|
"num_tokens": 13586930.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 5.4562568187713625,
|
|
"epoch": 6.176470588235294,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00021662168277273547,
|
|
"loss": 4.6904,
|
|
"mean_token_accuracy": 0.23341362923383713,
|
|
"num_tokens": 13595891.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 5.53175311088562,
|
|
"epoch": 6.1806130903065455,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002163131290381799,
|
|
"loss": 4.7666,
|
|
"mean_token_accuracy": 0.2249235674738884,
|
|
"num_tokens": 13605786.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 5.442884874343872,
|
|
"epoch": 6.184755592377797,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00021600469368145624,
|
|
"loss": 4.6511,
|
|
"mean_token_accuracy": 0.23911345899105071,
|
|
"num_tokens": 13614561.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 5.477076768875122,
|
|
"epoch": 6.188898094449047,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002156963773247124,
|
|
"loss": 4.6858,
|
|
"mean_token_accuracy": 0.23836783319711685,
|
|
"num_tokens": 13622853.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 5.452009534835815,
|
|
"epoch": 6.193040596520298,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002153881805898561,
|
|
"loss": 4.6538,
|
|
"mean_token_accuracy": 0.22906040847301484,
|
|
"num_tokens": 13631709.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 5.517543697357178,
|
|
"epoch": 6.197183098591549,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002150801040985539,
|
|
"loss": 4.7203,
|
|
"mean_token_accuracy": 0.2307763457298279,
|
|
"num_tokens": 13640854.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 5.421875858306885,
|
|
"epoch": 6.2013256006628,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021477214847222958,
|
|
"loss": 4.6647,
|
|
"mean_token_accuracy": 0.23170070201158524,
|
|
"num_tokens": 13649318.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 5.475580501556396,
|
|
"epoch": 6.205468102734051,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021446431433206348,
|
|
"loss": 4.7563,
|
|
"mean_token_accuracy": 0.22821633517742157,
|
|
"num_tokens": 13658738.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 5.3724822998046875,
|
|
"epoch": 6.209610604805302,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002141566022989907,
|
|
"loss": 4.6617,
|
|
"mean_token_accuracy": 0.23219722360372544,
|
|
"num_tokens": 13667315.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 5.482559776306152,
|
|
"epoch": 6.2137531068765535,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002138490129936998,
|
|
"loss": 4.7565,
|
|
"mean_token_accuracy": 0.22553036063909532,
|
|
"num_tokens": 13676665.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.2137531068765535,
|
|
"eval_entropy": 5.289943446991032,
|
|
"eval_loss": 5.740652561187744,
|
|
"eval_mean_token_accuracy": 0.16698340100965012,
|
|
"eval_num_tokens": 13676665.0,
|
|
"eval_runtime": 2.1038,
|
|
"eval_samples_per_second": 1781.049,
|
|
"eval_steps_per_second": 222.928,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 5.446432971954346,
|
|
"epoch": 6.217895608947805,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002135415470366323,
|
|
"loss": 4.704,
|
|
"mean_token_accuracy": 0.2294152483344078,
|
|
"num_tokens": 13685774.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 5.498067951202392,
|
|
"epoch": 6.222038111019056,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00021323420504798047,
|
|
"loss": 4.7332,
|
|
"mean_token_accuracy": 0.22531790882349015,
|
|
"num_tokens": 13695700.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 5.501949977874756,
|
|
"epoch": 6.226180613090307,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021292698764768675,
|
|
"loss": 4.7442,
|
|
"mean_token_accuracy": 0.2276514306664467,
|
|
"num_tokens": 13704675.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 5.471911525726318,
|
|
"epoch": 6.230323115161558,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00021261989545544218,
|
|
"loss": 4.7947,
|
|
"mean_token_accuracy": 0.2248911440372467,
|
|
"num_tokens": 13714562.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 5.552870893478394,
|
|
"epoch": 6.234465617232809,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00021231292909068534,
|
|
"loss": 4.759,
|
|
"mean_token_accuracy": 0.22857129573822021,
|
|
"num_tokens": 13723917.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 5.470516586303711,
|
|
"epoch": 6.238608119304059,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021200608917260082,
|
|
"loss": 4.6892,
|
|
"mean_token_accuracy": 0.23517653942108155,
|
|
"num_tokens": 13732482.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 5.48828444480896,
|
|
"epoch": 6.24275062137531,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00021169937632011838,
|
|
"loss": 4.7495,
|
|
"mean_token_accuracy": 0.22435679286718369,
|
|
"num_tokens": 13742110.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 5.435388565063477,
|
|
"epoch": 6.2468931234465614,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021139279115191123,
|
|
"loss": 4.6838,
|
|
"mean_token_accuracy": 0.21725301295518876,
|
|
"num_tokens": 13750501.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 5.443965101242066,
|
|
"epoch": 6.2510356255178126,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021108633428639534,
|
|
"loss": 4.6633,
|
|
"mean_token_accuracy": 0.23451380878686906,
|
|
"num_tokens": 13759786.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 5.469325971603394,
|
|
"epoch": 6.255178127589064,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002107800063417275,
|
|
"loss": 4.73,
|
|
"mean_token_accuracy": 0.22246698886156083,
|
|
"num_tokens": 13768476.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 5.471257543563842,
|
|
"epoch": 6.259320629660315,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002104738079358048,
|
|
"loss": 4.7561,
|
|
"mean_token_accuracy": 0.2139316529035568,
|
|
"num_tokens": 13778721.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 5.535071659088135,
|
|
"epoch": 6.263463131731566,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002101677396862628,
|
|
"loss": 4.7408,
|
|
"mean_token_accuracy": 0.2270593374967575,
|
|
"num_tokens": 13787108.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 5.450421524047852,
|
|
"epoch": 6.267605633802817,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020986180221047463,
|
|
"loss": 4.6253,
|
|
"mean_token_accuracy": 0.23937228173017502,
|
|
"num_tokens": 13795021.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 5.398626136779785,
|
|
"epoch": 6.271748135874068,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020955599612554966,
|
|
"loss": 4.7103,
|
|
"mean_token_accuracy": 0.2249516651034355,
|
|
"num_tokens": 13804750.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 5.419511699676514,
|
|
"epoch": 6.275890637945319,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020925032204833206,
|
|
"loss": 4.6879,
|
|
"mean_token_accuracy": 0.23139946460723876,
|
|
"num_tokens": 13813910.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 5.53766713142395,
|
|
"epoch": 6.28003314001657,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020894478059539995,
|
|
"loss": 4.6545,
|
|
"mean_token_accuracy": 0.2312415987253189,
|
|
"num_tokens": 13822832.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 5.430802297592163,
|
|
"epoch": 6.284175642087821,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020863937238306375,
|
|
"loss": 4.6528,
|
|
"mean_token_accuracy": 0.23278989791870117,
|
|
"num_tokens": 13831951.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 5.387748765945434,
|
|
"epoch": 6.2883181441590725,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020833409802736535,
|
|
"loss": 4.6437,
|
|
"mean_token_accuracy": 0.23334923684597014,
|
|
"num_tokens": 13840422.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 5.491671228408814,
|
|
"epoch": 6.292460646230323,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020802895814407636,
|
|
"loss": 4.7267,
|
|
"mean_token_accuracy": 0.2283321738243103,
|
|
"num_tokens": 13850903.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 5.506031656265259,
|
|
"epoch": 6.296603148301574,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002077239533486972,
|
|
"loss": 4.7968,
|
|
"mean_token_accuracy": 0.21267668455839156,
|
|
"num_tokens": 13859972.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 5.491634273529053,
|
|
"epoch": 6.300745650372825,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00020741908425645606,
|
|
"loss": 4.7632,
|
|
"mean_token_accuracy": 0.22508847266435622,
|
|
"num_tokens": 13868980.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 5.458075666427613,
|
|
"epoch": 6.304888152444076,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020711435148230707,
|
|
"loss": 4.6901,
|
|
"mean_token_accuracy": 0.22455791532993316,
|
|
"num_tokens": 13877901.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 5.411118507385254,
|
|
"epoch": 6.309030654515327,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020680975564092968,
|
|
"loss": 4.6941,
|
|
"mean_token_accuracy": 0.22158022671937944,
|
|
"num_tokens": 13887081.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 5.521286582946777,
|
|
"epoch": 6.313173156586578,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00020650529734672686,
|
|
"loss": 4.7765,
|
|
"mean_token_accuracy": 0.22457414120435715,
|
|
"num_tokens": 13896645.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 5.579707193374634,
|
|
"epoch": 6.317315658657829,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020620097721382424,
|
|
"loss": 4.7548,
|
|
"mean_token_accuracy": 0.2235257148742676,
|
|
"num_tokens": 13905975.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 5.5240696430206295,
|
|
"epoch": 6.3214581607290805,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002058967958560689,
|
|
"loss": 4.7073,
|
|
"mean_token_accuracy": 0.2291066586971283,
|
|
"num_tokens": 13914605.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 5.446935749053955,
|
|
"epoch": 6.325600662800332,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020559275388702775,
|
|
"loss": 4.6499,
|
|
"mean_token_accuracy": 0.23807752430438994,
|
|
"num_tokens": 13924592.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 5.386942768096924,
|
|
"epoch": 6.329743164871583,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00020528885191998682,
|
|
"loss": 4.6997,
|
|
"mean_token_accuracy": 0.22407541126012803,
|
|
"num_tokens": 13933513.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 5.366648149490357,
|
|
"epoch": 6.333885666942834,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00020498509056794952,
|
|
"loss": 4.578,
|
|
"mean_token_accuracy": 0.2412107393145561,
|
|
"num_tokens": 13942173.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 5.3916966915130615,
|
|
"epoch": 6.338028169014084,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020468147044363556,
|
|
"loss": 4.5923,
|
|
"mean_token_accuracy": 0.23077704310417174,
|
|
"num_tokens": 13950205.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 5.41549015045166,
|
|
"epoch": 6.342170671085335,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020437799215948017,
|
|
"loss": 4.6846,
|
|
"mean_token_accuracy": 0.23519167304039001,
|
|
"num_tokens": 13958956.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 5.411104726791382,
|
|
"epoch": 6.346313173156586,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002040746563276319,
|
|
"loss": 4.6913,
|
|
"mean_token_accuracy": 0.22375471591949464,
|
|
"num_tokens": 13967718.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 5.4366086483001705,
|
|
"epoch": 6.350455675227837,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020377146355995257,
|
|
"loss": 4.7823,
|
|
"mean_token_accuracy": 0.21871876269578933,
|
|
"num_tokens": 13976841.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 5.51446418762207,
|
|
"epoch": 6.3545981772990885,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00020346841446801495,
|
|
"loss": 4.6563,
|
|
"mean_token_accuracy": 0.2334139347076416,
|
|
"num_tokens": 13984946.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 5.4362020015716555,
|
|
"epoch": 6.35874067937034,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020316550966310222,
|
|
"loss": 4.7184,
|
|
"mean_token_accuracy": 0.22419632375240325,
|
|
"num_tokens": 13993521.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 5.411268281936645,
|
|
"epoch": 6.362883181441591,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002028627497562065,
|
|
"loss": 4.7034,
|
|
"mean_token_accuracy": 0.2253747507929802,
|
|
"num_tokens": 14002777.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 5.467538118362427,
|
|
"epoch": 6.367025683512842,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020256013535802748,
|
|
"loss": 4.7245,
|
|
"mean_token_accuracy": 0.2179402604699135,
|
|
"num_tokens": 14012708.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 5.512913608551026,
|
|
"epoch": 6.371168185584093,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002022576670789717,
|
|
"loss": 4.7452,
|
|
"mean_token_accuracy": 0.21670423150062562,
|
|
"num_tokens": 14021088.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 5.478987550735473,
|
|
"epoch": 6.375310687655344,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002019553455291505,
|
|
"loss": 4.7574,
|
|
"mean_token_accuracy": 0.2214504823088646,
|
|
"num_tokens": 14031650.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 5.455540800094605,
|
|
"epoch": 6.379453189726595,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002016531713183797,
|
|
"loss": 4.7627,
|
|
"mean_token_accuracy": 0.22557770162820817,
|
|
"num_tokens": 14040741.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 5.46314492225647,
|
|
"epoch": 6.383595691797846,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00020135114505617752,
|
|
"loss": 4.7312,
|
|
"mean_token_accuracy": 0.2182163864374161,
|
|
"num_tokens": 14049913.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 5.479154109954834,
|
|
"epoch": 6.387738193869097,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020104926735176393,
|
|
"loss": 4.7015,
|
|
"mean_token_accuracy": 0.23123564720153808,
|
|
"num_tokens": 14059583.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 5.535675859451294,
|
|
"epoch": 6.391880695940348,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020074753881405949,
|
|
"loss": 4.6647,
|
|
"mean_token_accuracy": 0.23026125133037567,
|
|
"num_tokens": 14067376.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 5.462156772613525,
|
|
"epoch": 6.396023198011599,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002004459600516833,
|
|
"loss": 4.7344,
|
|
"mean_token_accuracy": 0.22997039407491685,
|
|
"num_tokens": 14076415.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 5.5187239170074465,
|
|
"epoch": 6.40016570008285,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00020014453167295298,
|
|
"loss": 4.7294,
|
|
"mean_token_accuracy": 0.21980999261140824,
|
|
"num_tokens": 14085418.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 5.479145431518555,
|
|
"epoch": 6.404308202154101,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019984325428588229,
|
|
"loss": 4.7072,
|
|
"mean_token_accuracy": 0.22885724455118178,
|
|
"num_tokens": 14094454.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 5.464708566665649,
|
|
"epoch": 6.408450704225352,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019954212849818077,
|
|
"loss": 4.6438,
|
|
"mean_token_accuracy": 0.23752079904079437,
|
|
"num_tokens": 14103611.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 5.390991163253784,
|
|
"epoch": 6.412593206296603,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019924115491725197,
|
|
"loss": 4.5797,
|
|
"mean_token_accuracy": 0.23385887891054152,
|
|
"num_tokens": 14112145.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 5.4284874439239506,
|
|
"epoch": 6.416735708367854,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019894033415019253,
|
|
"loss": 4.7136,
|
|
"mean_token_accuracy": 0.2247097373008728,
|
|
"num_tokens": 14121215.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 5.39704270362854,
|
|
"epoch": 6.420878210439105,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019863966680379086,
|
|
"loss": 4.5765,
|
|
"mean_token_accuracy": 0.23517947942018508,
|
|
"num_tokens": 14130106.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 5.388026905059815,
|
|
"epoch": 6.4250207125103564,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001983391534845258,
|
|
"loss": 4.642,
|
|
"mean_token_accuracy": 0.23924634456634522,
|
|
"num_tokens": 14139071.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 5.421600294113159,
|
|
"epoch": 6.4291632145816076,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00019803879479856547,
|
|
"loss": 4.6998,
|
|
"mean_token_accuracy": 0.22814930081367493,
|
|
"num_tokens": 14148018.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 5.429807710647583,
|
|
"epoch": 6.433305716652859,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019773859135176626,
|
|
"loss": 4.6458,
|
|
"mean_token_accuracy": 0.22997224926948548,
|
|
"num_tokens": 14156369.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 5.48077974319458,
|
|
"epoch": 6.43744821872411,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001974385437496713,
|
|
"loss": 4.7386,
|
|
"mean_token_accuracy": 0.22237730026245117,
|
|
"num_tokens": 14165930.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 5.4337592124938965,
|
|
"epoch": 6.44159072079536,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019713865259750945,
|
|
"loss": 4.6847,
|
|
"mean_token_accuracy": 0.22869787365198135,
|
|
"num_tokens": 14175557.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 5.48696117401123,
|
|
"epoch": 6.445733222866611,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019683891850019391,
|
|
"loss": 4.7541,
|
|
"mean_token_accuracy": 0.22052708864212037,
|
|
"num_tokens": 14185532.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 5.572324991226196,
|
|
"epoch": 6.449875724937862,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019653934206232098,
|
|
"loss": 4.8428,
|
|
"mean_token_accuracy": 0.22178803384304047,
|
|
"num_tokens": 14194307.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 5.515306568145752,
|
|
"epoch": 6.454018227009113,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019623992388816935,
|
|
"loss": 4.7958,
|
|
"mean_token_accuracy": 0.22667295932769777,
|
|
"num_tokens": 14203900.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 5.4851343631744385,
|
|
"epoch": 6.458160729080364,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001959406645816979,
|
|
"loss": 4.6923,
|
|
"mean_token_accuracy": 0.22562823444604874,
|
|
"num_tokens": 14213232.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 5.447917652130127,
|
|
"epoch": 6.4623032311516155,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019564156474654554,
|
|
"loss": 4.6473,
|
|
"mean_token_accuracy": 0.22318599969148636,
|
|
"num_tokens": 14221365.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 5.383006620407104,
|
|
"epoch": 6.466445733222867,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019534262498602928,
|
|
"loss": 4.6858,
|
|
"mean_token_accuracy": 0.22617679536342622,
|
|
"num_tokens": 14229878.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 5.487738132476807,
|
|
"epoch": 6.470588235294118,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00019504384590314318,
|
|
"loss": 4.7268,
|
|
"mean_token_accuracy": 0.22530025243759155,
|
|
"num_tokens": 14239132.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 5.488894414901734,
|
|
"epoch": 6.474730737365369,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019474522810055745,
|
|
"loss": 4.7386,
|
|
"mean_token_accuracy": 0.22939057499170304,
|
|
"num_tokens": 14247116.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 5.406441879272461,
|
|
"epoch": 6.47887323943662,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00019444677218061673,
|
|
"loss": 4.7039,
|
|
"mean_token_accuracy": 0.2315358877182007,
|
|
"num_tokens": 14255423.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 5.461256265640259,
|
|
"epoch": 6.483015741507871,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001941484787453393,
|
|
"loss": 4.6792,
|
|
"mean_token_accuracy": 0.22345255464315414,
|
|
"num_tokens": 14263836.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 5.44575719833374,
|
|
"epoch": 6.487158243579122,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00019385034839641558,
|
|
"loss": 4.7608,
|
|
"mean_token_accuracy": 0.22295506298542023,
|
|
"num_tokens": 14273460.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 5.476379251480102,
|
|
"epoch": 6.491300745650372,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001935523817352071,
|
|
"loss": 4.8025,
|
|
"mean_token_accuracy": 0.22025205790996552,
|
|
"num_tokens": 14282925.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 5.490783023834228,
|
|
"epoch": 6.4954432477216235,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019325457936274516,
|
|
"loss": 4.7073,
|
|
"mean_token_accuracy": 0.2324343666434288,
|
|
"num_tokens": 14292059.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 5.491448640823364,
|
|
"epoch": 6.499585749792875,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019295694187972955,
|
|
"loss": 4.7306,
|
|
"mean_token_accuracy": 0.22499118149280548,
|
|
"num_tokens": 14301276.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 5.5004771709442135,
|
|
"epoch": 6.503728251864126,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019265946988652783,
|
|
"loss": 4.7255,
|
|
"mean_token_accuracy": 0.22655408680438996,
|
|
"num_tokens": 14310259.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 5.420364284515381,
|
|
"epoch": 6.507870753935377,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001923621639831733,
|
|
"loss": 4.6252,
|
|
"mean_token_accuracy": 0.2297055870294571,
|
|
"num_tokens": 14318782.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 5.41087498664856,
|
|
"epoch": 6.512013256006628,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019206502476936467,
|
|
"loss": 4.5778,
|
|
"mean_token_accuracy": 0.24084218442440034,
|
|
"num_tokens": 14327571.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 5.376150035858155,
|
|
"epoch": 6.516155758077879,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00019176805284446414,
|
|
"loss": 4.6398,
|
|
"mean_token_accuracy": 0.23546205312013627,
|
|
"num_tokens": 14336974.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 5.480745124816894,
|
|
"epoch": 6.52029826014913,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019147124880749633,
|
|
"loss": 4.802,
|
|
"mean_token_accuracy": 0.2192357748746872,
|
|
"num_tokens": 14346367.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 5.477602577209472,
|
|
"epoch": 6.524440762220381,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00019117461325714768,
|
|
"loss": 4.724,
|
|
"mean_token_accuracy": 0.22668899297714235,
|
|
"num_tokens": 14355641.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 5.48706431388855,
|
|
"epoch": 6.528583264291632,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019087814679176436,
|
|
"loss": 4.6713,
|
|
"mean_token_accuracy": 0.23021564781665801,
|
|
"num_tokens": 14364927.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 5.461276483535767,
|
|
"epoch": 6.5327257663628835,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019058185000935175,
|
|
"loss": 4.6585,
|
|
"mean_token_accuracy": 0.23257556855678557,
|
|
"num_tokens": 14373370.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 5.474993133544922,
|
|
"epoch": 6.536868268434135,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00019028572350757278,
|
|
"loss": 4.7305,
|
|
"mean_token_accuracy": 0.2261441856622696,
|
|
"num_tokens": 14381820.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 5.4712308883667,
|
|
"epoch": 6.541010770505386,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018998976788374682,
|
|
"loss": 4.709,
|
|
"mean_token_accuracy": 0.22573544085025787,
|
|
"num_tokens": 14391158.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 5.439238119125366,
|
|
"epoch": 6.545153272576636,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018969398373484898,
|
|
"loss": 4.7276,
|
|
"mean_token_accuracy": 0.2247076541185379,
|
|
"num_tokens": 14399945.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 5.457880878448487,
|
|
"epoch": 6.549295774647887,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018939837165750796,
|
|
"loss": 4.6638,
|
|
"mean_token_accuracy": 0.23126714527606965,
|
|
"num_tokens": 14410275.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 5.52023868560791,
|
|
"epoch": 6.553438276719138,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018910293224800584,
|
|
"loss": 4.7484,
|
|
"mean_token_accuracy": 0.2301405593752861,
|
|
"num_tokens": 14420369.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 5.481889486312866,
|
|
"epoch": 6.557580778790389,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018880766610227608,
|
|
"loss": 4.7186,
|
|
"mean_token_accuracy": 0.23138007521629333,
|
|
"num_tokens": 14429078.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 5.4427628993988035,
|
|
"epoch": 6.56172328086164,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00018851257381590264,
|
|
"loss": 4.6466,
|
|
"mean_token_accuracy": 0.23252028077840806,
|
|
"num_tokens": 14438920.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 5.547739887237549,
|
|
"epoch": 6.5658657829328915,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00018821765598411917,
|
|
"loss": 4.8686,
|
|
"mean_token_accuracy": 0.22076121717691422,
|
|
"num_tokens": 14449164.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 5.445230865478516,
|
|
"epoch": 6.570008285004143,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00018792291320180693,
|
|
"loss": 4.7872,
|
|
"mean_token_accuracy": 0.21768543422222136,
|
|
"num_tokens": 14459418.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 5.5161356925964355,
|
|
"epoch": 6.574150787075394,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018762834606349455,
|
|
"loss": 4.7279,
|
|
"mean_token_accuracy": 0.22668305039405823,
|
|
"num_tokens": 14468691.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 5.54826078414917,
|
|
"epoch": 6.578293289146645,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00018733395516335588,
|
|
"loss": 4.7903,
|
|
"mean_token_accuracy": 0.2204552561044693,
|
|
"num_tokens": 14477936.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 5.532195329666138,
|
|
"epoch": 6.582435791217896,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018703974109520976,
|
|
"loss": 4.7658,
|
|
"mean_token_accuracy": 0.2294721856713295,
|
|
"num_tokens": 14487191.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 5.432251882553101,
|
|
"epoch": 6.586578293289147,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018674570445251798,
|
|
"loss": 4.6938,
|
|
"mean_token_accuracy": 0.22849086225032805,
|
|
"num_tokens": 14495731.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 5.38211030960083,
|
|
"epoch": 6.590720795360397,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018645184582838465,
|
|
"loss": 4.6649,
|
|
"mean_token_accuracy": 0.22463564276695253,
|
|
"num_tokens": 14505009.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 5.3984674453735355,
|
|
"epoch": 6.594863297431648,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018615816581555466,
|
|
"loss": 4.7297,
|
|
"mean_token_accuracy": 0.21958644390106202,
|
|
"num_tokens": 14513843.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 5.440894412994385,
|
|
"epoch": 6.5990057995028994,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001858646650064128,
|
|
"loss": 4.7321,
|
|
"mean_token_accuracy": 0.23181678652763366,
|
|
"num_tokens": 14523405.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 5.524806642532349,
|
|
"epoch": 6.603148301574151,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018557134399298215,
|
|
"loss": 4.7903,
|
|
"mean_token_accuracy": 0.22481401711702348,
|
|
"num_tokens": 14532615.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 5.484766960144043,
|
|
"epoch": 6.607290803645402,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001852782033669233,
|
|
"loss": 4.7131,
|
|
"mean_token_accuracy": 0.2321273133158684,
|
|
"num_tokens": 14541297.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 5.439073467254639,
|
|
"epoch": 6.611433305716653,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018498524371953295,
|
|
"loss": 4.6763,
|
|
"mean_token_accuracy": 0.22429613918066024,
|
|
"num_tokens": 14549867.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 5.460627841949463,
|
|
"epoch": 6.615575807787904,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018469246564174275,
|
|
"loss": 4.771,
|
|
"mean_token_accuracy": 0.21486834138631822,
|
|
"num_tokens": 14558708.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 5.449061727523803,
|
|
"epoch": 6.619718309859155,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000184399869724118,
|
|
"loss": 4.7424,
|
|
"mean_token_accuracy": 0.21969811171293258,
|
|
"num_tokens": 14567917.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 5.407015132904053,
|
|
"epoch": 6.623860811930406,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018410745655685662,
|
|
"loss": 4.6238,
|
|
"mean_token_accuracy": 0.23875953108072281,
|
|
"num_tokens": 14576417.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 5.397463035583496,
|
|
"epoch": 6.628003314001657,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00018381522672978806,
|
|
"loss": 4.7139,
|
|
"mean_token_accuracy": 0.22609803229570388,
|
|
"num_tokens": 14585172.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.628003314001657,
|
|
"eval_entropy": 5.287970195192772,
|
|
"eval_loss": 5.729158878326416,
|
|
"eval_mean_token_accuracy": 0.16776795118157542,
|
|
"eval_num_tokens": 14585172.0,
|
|
"eval_runtime": 2.112,
|
|
"eval_samples_per_second": 1774.189,
|
|
"eval_steps_per_second": 222.07,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 5.518819141387939,
|
|
"epoch": 6.632145816072908,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018352318083237167,
|
|
"loss": 4.7385,
|
|
"mean_token_accuracy": 0.21797265112400055,
|
|
"num_tokens": 14593758.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 5.482235670089722,
|
|
"epoch": 6.636288318144159,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00018323131945369603,
|
|
"loss": 4.6943,
|
|
"mean_token_accuracy": 0.22733736634254456,
|
|
"num_tokens": 14602558.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 5.5073973655700685,
|
|
"epoch": 6.6404308202154105,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018293964318247735,
|
|
"loss": 4.7209,
|
|
"mean_token_accuracy": 0.2269052043557167,
|
|
"num_tokens": 14611463.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 5.467496013641357,
|
|
"epoch": 6.644573322286661,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00018264815260705863,
|
|
"loss": 4.7226,
|
|
"mean_token_accuracy": 0.22509234994649888,
|
|
"num_tokens": 14619951.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 5.509167289733886,
|
|
"epoch": 6.648715824357912,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018235684831540826,
|
|
"loss": 4.7773,
|
|
"mean_token_accuracy": 0.2251000627875328,
|
|
"num_tokens": 14629332.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 5.451407718658447,
|
|
"epoch": 6.652858326429163,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018206573089511862,
|
|
"loss": 4.7148,
|
|
"mean_token_accuracy": 0.22758610546588898,
|
|
"num_tokens": 14639010.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 5.457706260681152,
|
|
"epoch": 6.657000828500414,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00018177480093340564,
|
|
"loss": 4.7379,
|
|
"mean_token_accuracy": 0.21934054791927338,
|
|
"num_tokens": 14648507.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 5.450697517395019,
|
|
"epoch": 6.661143330571665,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001814840590171065,
|
|
"loss": 4.73,
|
|
"mean_token_accuracy": 0.22598262429237365,
|
|
"num_tokens": 14657892.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 5.459395933151245,
|
|
"epoch": 6.665285832642916,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00018119350573267979,
|
|
"loss": 4.7094,
|
|
"mean_token_accuracy": 0.22731710821390153,
|
|
"num_tokens": 14666460.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 5.460178470611572,
|
|
"epoch": 6.669428334714167,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00018090314166620302,
|
|
"loss": 4.7002,
|
|
"mean_token_accuracy": 0.2245880290865898,
|
|
"num_tokens": 14675046.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 5.457176256179809,
|
|
"epoch": 6.6735708367854185,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018061296740337225,
|
|
"loss": 4.7316,
|
|
"mean_token_accuracy": 0.22238862365484238,
|
|
"num_tokens": 14685055.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 5.472867631912232,
|
|
"epoch": 6.67771333885667,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018032298352950077,
|
|
"loss": 4.7166,
|
|
"mean_token_accuracy": 0.2237478882074356,
|
|
"num_tokens": 14695371.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 5.411253213882446,
|
|
"epoch": 6.681855840927921,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00018003319062951756,
|
|
"loss": 4.686,
|
|
"mean_token_accuracy": 0.22918927669525146,
|
|
"num_tokens": 14703671.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 5.454790353775024,
|
|
"epoch": 6.685998342999172,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017974358928796682,
|
|
"loss": 4.7059,
|
|
"mean_token_accuracy": 0.23365746289491654,
|
|
"num_tokens": 14712527.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 5.456111001968384,
|
|
"epoch": 6.690140845070422,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00017945418008900594,
|
|
"loss": 4.7472,
|
|
"mean_token_accuracy": 0.22565506994724274,
|
|
"num_tokens": 14720711.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 5.494932508468628,
|
|
"epoch": 6.694283347141674,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001791649636164049,
|
|
"loss": 4.7811,
|
|
"mean_token_accuracy": 0.2167421191930771,
|
|
"num_tokens": 14729667.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 5.54667649269104,
|
|
"epoch": 6.698425849212924,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000178875940453545,
|
|
"loss": 4.7039,
|
|
"mean_token_accuracy": 0.2284021258354187,
|
|
"num_tokens": 14738614.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 5.500935173034668,
|
|
"epoch": 6.702568351284175,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017858711118341742,
|
|
"loss": 4.7282,
|
|
"mean_token_accuracy": 0.2218704491853714,
|
|
"num_tokens": 14747724.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 5.424642896652221,
|
|
"epoch": 6.7067108533554265,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00017829847638862255,
|
|
"loss": 4.6752,
|
|
"mean_token_accuracy": 0.22767631560564042,
|
|
"num_tokens": 14757436.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 5.404869079589844,
|
|
"epoch": 6.710853355426678,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017801003665136823,
|
|
"loss": 4.6704,
|
|
"mean_token_accuracy": 0.2330010563135147,
|
|
"num_tokens": 14765948.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 5.44160795211792,
|
|
"epoch": 6.714995857497929,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017772179255346883,
|
|
"loss": 4.7046,
|
|
"mean_token_accuracy": 0.22521086633205414,
|
|
"num_tokens": 14775397.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 5.443006563186645,
|
|
"epoch": 6.71913835956918,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017743374467634433,
|
|
"loss": 4.6716,
|
|
"mean_token_accuracy": 0.2303983062505722,
|
|
"num_tokens": 14783605.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 5.4643017768859865,
|
|
"epoch": 6.723280861640431,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017714589360101862,
|
|
"loss": 4.7749,
|
|
"mean_token_accuracy": 0.22290083169937133,
|
|
"num_tokens": 14792210.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 5.438012170791626,
|
|
"epoch": 6.727423363711682,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000176858239908119,
|
|
"loss": 4.6713,
|
|
"mean_token_accuracy": 0.2300460159778595,
|
|
"num_tokens": 14801186.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 5.402060174942017,
|
|
"epoch": 6.731565865782933,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017657078417787414,
|
|
"loss": 4.6175,
|
|
"mean_token_accuracy": 0.2323254868388176,
|
|
"num_tokens": 14810100.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 5.37335991859436,
|
|
"epoch": 6.735708367854184,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017628352699011396,
|
|
"loss": 4.6081,
|
|
"mean_token_accuracy": 0.22928288727998733,
|
|
"num_tokens": 14819335.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 5.4116315841674805,
|
|
"epoch": 6.739850869925435,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001759964689242674,
|
|
"loss": 4.7139,
|
|
"mean_token_accuracy": 0.23641964942216873,
|
|
"num_tokens": 14828686.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 5.468182754516602,
|
|
"epoch": 6.743993371996686,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001757096105593619,
|
|
"loss": 4.7227,
|
|
"mean_token_accuracy": 0.22019149512052535,
|
|
"num_tokens": 14838312.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 5.3723437786102295,
|
|
"epoch": 6.748135874067937,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017542295247402234,
|
|
"loss": 4.6665,
|
|
"mean_token_accuracy": 0.22680625319480896,
|
|
"num_tokens": 14846838.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 5.444209432601928,
|
|
"epoch": 6.752278376139188,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017513649524646919,
|
|
"loss": 4.7246,
|
|
"mean_token_accuracy": 0.21581420600414275,
|
|
"num_tokens": 14857427.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 5.44220461845398,
|
|
"epoch": 6.756420878210439,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001748502394545181,
|
|
"loss": 4.7304,
|
|
"mean_token_accuracy": 0.21652780324220658,
|
|
"num_tokens": 14866244.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 5.44826111793518,
|
|
"epoch": 6.76056338028169,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001745641856755783,
|
|
"loss": 4.7259,
|
|
"mean_token_accuracy": 0.2213211938738823,
|
|
"num_tokens": 14875697.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 5.4401023387908936,
|
|
"epoch": 6.764705882352941,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017427833448665136,
|
|
"loss": 4.7445,
|
|
"mean_token_accuracy": 0.22263761311769487,
|
|
"num_tokens": 14884650.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 5.428892183303833,
|
|
"epoch": 6.768848384424192,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017399268646433065,
|
|
"loss": 4.6972,
|
|
"mean_token_accuracy": 0.22623355239629744,
|
|
"num_tokens": 14894058.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 5.48489785194397,
|
|
"epoch": 6.772990886495443,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017370724218479912,
|
|
"loss": 4.792,
|
|
"mean_token_accuracy": 0.2254809856414795,
|
|
"num_tokens": 14903219.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 5.465812635421753,
|
|
"epoch": 6.7771333885666944,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00017342200222382934,
|
|
"loss": 4.72,
|
|
"mean_token_accuracy": 0.22732471525669098,
|
|
"num_tokens": 14913051.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 5.511168622970581,
|
|
"epoch": 6.781275890637946,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017313696715678134,
|
|
"loss": 4.7498,
|
|
"mean_token_accuracy": 0.22207529544830323,
|
|
"num_tokens": 14921777.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 5.539793682098389,
|
|
"epoch": 6.785418392709197,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.000172852137558602,
|
|
"loss": 4.883,
|
|
"mean_token_accuracy": 0.21283642202615738,
|
|
"num_tokens": 14931558.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 5.501955938339234,
|
|
"epoch": 6.789560894780448,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017256751400382384,
|
|
"loss": 4.7597,
|
|
"mean_token_accuracy": 0.2309992417693138,
|
|
"num_tokens": 14940584.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 5.51120228767395,
|
|
"epoch": 6.793703396851699,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017228309706656354,
|
|
"loss": 4.7246,
|
|
"mean_token_accuracy": 0.22900386899709702,
|
|
"num_tokens": 14950830.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 5.5506373882293705,
|
|
"epoch": 6.797845898922949,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001719988873205213,
|
|
"loss": 4.7241,
|
|
"mean_token_accuracy": 0.21901852935552596,
|
|
"num_tokens": 14958990.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 5.426819467544556,
|
|
"epoch": 6.8019884009942,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017171488533897916,
|
|
"loss": 4.7347,
|
|
"mean_token_accuracy": 0.22732140868902206,
|
|
"num_tokens": 14967451.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 5.415303134918213,
|
|
"epoch": 6.806130903065451,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017143109169480005,
|
|
"loss": 4.6768,
|
|
"mean_token_accuracy": 0.22993717789649964,
|
|
"num_tokens": 14976157.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 5.471483182907105,
|
|
"epoch": 6.810273405136702,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017114750696042695,
|
|
"loss": 4.8007,
|
|
"mean_token_accuracy": 0.21850461810827254,
|
|
"num_tokens": 14985591.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 5.450443172454834,
|
|
"epoch": 6.8144159072079535,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00017086413170788107,
|
|
"loss": 4.6882,
|
|
"mean_token_accuracy": 0.2305337205529213,
|
|
"num_tokens": 14994513.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 5.470681095123291,
|
|
"epoch": 6.818558409279205,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00017058096650876143,
|
|
"loss": 4.7557,
|
|
"mean_token_accuracy": 0.22366218268871307,
|
|
"num_tokens": 15003252.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 5.497387838363648,
|
|
"epoch": 6.822700911350456,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00017029801193424313,
|
|
"loss": 4.6906,
|
|
"mean_token_accuracy": 0.2275169923901558,
|
|
"num_tokens": 15012342.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 5.380797386169434,
|
|
"epoch": 6.826843413421707,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00017001526855507632,
|
|
"loss": 4.6469,
|
|
"mean_token_accuracy": 0.22846923172473907,
|
|
"num_tokens": 15021336.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 5.421555089950561,
|
|
"epoch": 6.830985915492958,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016973273694158555,
|
|
"loss": 4.6755,
|
|
"mean_token_accuracy": 0.21999928653240203,
|
|
"num_tokens": 15031016.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 5.505602788925171,
|
|
"epoch": 6.835128417564209,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016945041766366773,
|
|
"loss": 4.7392,
|
|
"mean_token_accuracy": 0.2208612933754921,
|
|
"num_tokens": 15039181.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 5.493553733825683,
|
|
"epoch": 6.83927091963546,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016916831129079187,
|
|
"loss": 4.755,
|
|
"mean_token_accuracy": 0.2161984220147133,
|
|
"num_tokens": 15048563.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 5.496672058105469,
|
|
"epoch": 6.84341342170671,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001688864183919973,
|
|
"loss": 4.7042,
|
|
"mean_token_accuracy": 0.22479698359966277,
|
|
"num_tokens": 15056960.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 5.517833232879639,
|
|
"epoch": 6.8475559237779615,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016860473953589266,
|
|
"loss": 4.7307,
|
|
"mean_token_accuracy": 0.22456844598054887,
|
|
"num_tokens": 15065645.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 5.496301317214966,
|
|
"epoch": 6.851698425849213,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016832327529065522,
|
|
"loss": 4.7143,
|
|
"mean_token_accuracy": 0.22681891024112702,
|
|
"num_tokens": 15074544.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 5.480351972579956,
|
|
"epoch": 6.855840927920464,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001680420262240289,
|
|
"loss": 4.7436,
|
|
"mean_token_accuracy": 0.2264260694384575,
|
|
"num_tokens": 15083543.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 5.416957712173462,
|
|
"epoch": 6.859983429991715,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016776099290332397,
|
|
"loss": 4.7236,
|
|
"mean_token_accuracy": 0.22866163849830629,
|
|
"num_tokens": 15092291.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 5.539653015136719,
|
|
"epoch": 6.864125932062966,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00016748017589541525,
|
|
"loss": 4.779,
|
|
"mean_token_accuracy": 0.2211819499731064,
|
|
"num_tokens": 15100875.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 5.4702881336212155,
|
|
"epoch": 6.868268434134217,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001671995757667414,
|
|
"loss": 4.7428,
|
|
"mean_token_accuracy": 0.22622203826904297,
|
|
"num_tokens": 15111385.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 5.541867256164551,
|
|
"epoch": 6.872410936205468,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016691919308330343,
|
|
"loss": 4.7783,
|
|
"mean_token_accuracy": 0.22476228922605515,
|
|
"num_tokens": 15121153.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 5.476632165908813,
|
|
"epoch": 6.876553438276719,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016663902841066404,
|
|
"loss": 4.7414,
|
|
"mean_token_accuracy": 0.219778436422348,
|
|
"num_tokens": 15130132.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 5.427923536300659,
|
|
"epoch": 6.88069594034797,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016635908231394587,
|
|
"loss": 4.724,
|
|
"mean_token_accuracy": 0.2216860234737396,
|
|
"num_tokens": 15138878.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 5.447377061843872,
|
|
"epoch": 6.8848384424192215,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001660793553578307,
|
|
"loss": 4.7861,
|
|
"mean_token_accuracy": 0.21774494498968125,
|
|
"num_tokens": 15148727.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 5.457162666320801,
|
|
"epoch": 6.888980944490473,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00016579984810655856,
|
|
"loss": 4.7446,
|
|
"mean_token_accuracy": 0.22254077792167665,
|
|
"num_tokens": 15158107.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 5.478066539764404,
|
|
"epoch": 6.893123446561724,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016552056112392594,
|
|
"loss": 4.7389,
|
|
"mean_token_accuracy": 0.21979377716779708,
|
|
"num_tokens": 15168101.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 5.490390205383301,
|
|
"epoch": 6.897265948632974,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001652414949732854,
|
|
"loss": 4.6719,
|
|
"mean_token_accuracy": 0.22484802603721618,
|
|
"num_tokens": 15176376.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 5.449921178817749,
|
|
"epoch": 6.901408450704225,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016496265021754365,
|
|
"loss": 4.6966,
|
|
"mean_token_accuracy": 0.22801542282104492,
|
|
"num_tokens": 15184936.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 5.512594938278198,
|
|
"epoch": 6.905550952775476,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016468402741916124,
|
|
"loss": 4.7814,
|
|
"mean_token_accuracy": 0.2254841923713684,
|
|
"num_tokens": 15194846.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 5.426405668258667,
|
|
"epoch": 6.909693454846727,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001644056271401506,
|
|
"loss": 4.7144,
|
|
"mean_token_accuracy": 0.2276289790868759,
|
|
"num_tokens": 15204543.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 5.466154527664185,
|
|
"epoch": 6.913835956917978,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016412744994207573,
|
|
"loss": 4.7078,
|
|
"mean_token_accuracy": 0.22184761613607407,
|
|
"num_tokens": 15213459.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 5.47257924079895,
|
|
"epoch": 6.9179784589892295,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00016384949638605025,
|
|
"loss": 4.7848,
|
|
"mean_token_accuracy": 0.20956427454948426,
|
|
"num_tokens": 15223841.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 5.441487646102905,
|
|
"epoch": 6.922120961060481,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016357176703273696,
|
|
"loss": 4.7293,
|
|
"mean_token_accuracy": 0.2279518201947212,
|
|
"num_tokens": 15233594.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 5.447011280059814,
|
|
"epoch": 6.926263463131732,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001632942624423463,
|
|
"loss": 4.6618,
|
|
"mean_token_accuracy": 0.23597458451986314,
|
|
"num_tokens": 15242876.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 5.527422523498535,
|
|
"epoch": 6.930405965202983,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00016301698317463525,
|
|
"loss": 4.7637,
|
|
"mean_token_accuracy": 0.22331327497959136,
|
|
"num_tokens": 15251581.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 5.495326089859009,
|
|
"epoch": 6.934548467274234,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00016273992978890658,
|
|
"loss": 4.6403,
|
|
"mean_token_accuracy": 0.23078704327344896,
|
|
"num_tokens": 15260601.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 5.447265338897705,
|
|
"epoch": 6.938690969345485,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001624631028440071,
|
|
"loss": 4.6497,
|
|
"mean_token_accuracy": 0.22823804020881652,
|
|
"num_tokens": 15270123.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 5.492997741699218,
|
|
"epoch": 6.942833471416735,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001621865028983271,
|
|
"loss": 4.7637,
|
|
"mean_token_accuracy": 0.22195232212543486,
|
|
"num_tokens": 15278623.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 5.401391935348511,
|
|
"epoch": 6.946975973487987,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001619101305097988,
|
|
"loss": 4.6801,
|
|
"mean_token_accuracy": 0.2280568525195122,
|
|
"num_tokens": 15287789.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 5.348383092880249,
|
|
"epoch": 6.9511184755592375,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00016163398623589563,
|
|
"loss": 4.634,
|
|
"mean_token_accuracy": 0.23254870027303695,
|
|
"num_tokens": 15296266.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 5.5018696784973145,
|
|
"epoch": 6.955260977630489,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00016135807063363074,
|
|
"loss": 4.7336,
|
|
"mean_token_accuracy": 0.2199333652853966,
|
|
"num_tokens": 15304785.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 5.389359617233277,
|
|
"epoch": 6.95940347970174,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016108238425955584,
|
|
"loss": 4.6437,
|
|
"mean_token_accuracy": 0.22871606498956681,
|
|
"num_tokens": 15313727.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 5.4866879940032955,
|
|
"epoch": 6.963545981772991,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001608069276697608,
|
|
"loss": 4.8143,
|
|
"mean_token_accuracy": 0.22128498703241348,
|
|
"num_tokens": 15322911.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 5.506561660766602,
|
|
"epoch": 6.967688483844242,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001605317014198714,
|
|
"loss": 4.7271,
|
|
"mean_token_accuracy": 0.22974718362092972,
|
|
"num_tokens": 15330882.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 5.437571811676025,
|
|
"epoch": 6.971830985915493,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016025670606504926,
|
|
"loss": 4.6322,
|
|
"mean_token_accuracy": 0.2328172042965889,
|
|
"num_tokens": 15339915.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 5.4235039234161375,
|
|
"epoch": 6.975973487986744,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001599819421599899,
|
|
"loss": 4.6914,
|
|
"mean_token_accuracy": 0.22509037852287292,
|
|
"num_tokens": 15348181.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 5.412591743469238,
|
|
"epoch": 6.980115990057995,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015970741025892233,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.23317933082580566,
|
|
"num_tokens": 15357128.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 5.438268423080444,
|
|
"epoch": 6.984258492129246,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00015943311091560721,
|
|
"loss": 4.7622,
|
|
"mean_token_accuracy": 0.22604166120290756,
|
|
"num_tokens": 15366878.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 5.550321102142334,
|
|
"epoch": 6.988400994200497,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00015915904468333635,
|
|
"loss": 4.7154,
|
|
"mean_token_accuracy": 0.23511925637722014,
|
|
"num_tokens": 15376101.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 5.4086997509002686,
|
|
"epoch": 6.9925434962717485,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015888521211493135,
|
|
"loss": 4.651,
|
|
"mean_token_accuracy": 0.22697234153747559,
|
|
"num_tokens": 15384942.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 5.400546455383301,
|
|
"epoch": 6.996685998342999,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001586116137627423,
|
|
"loss": 4.6714,
|
|
"mean_token_accuracy": 0.2327042728662491,
|
|
"num_tokens": 15394380.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 5.524434995651245,
|
|
"epoch": 7.00082850041425,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00015833825017864711,
|
|
"loss": 4.6303,
|
|
"mean_token_accuracy": 0.23847290873527527,
|
|
"num_tokens": 15402180.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 5.469347906112671,
|
|
"epoch": 7.004971002485501,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015806512191405,
|
|
"loss": 4.6972,
|
|
"mean_token_accuracy": 0.2366890490055084,
|
|
"num_tokens": 15411434.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 5.469150495529175,
|
|
"epoch": 7.009113504556752,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001577922295198803,
|
|
"loss": 4.6282,
|
|
"mean_token_accuracy": 0.23492867201566697,
|
|
"num_tokens": 15419905.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 5.427287864685058,
|
|
"epoch": 7.013256006628003,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015751957354659206,
|
|
"loss": 4.6597,
|
|
"mean_token_accuracy": 0.23350475877523422,
|
|
"num_tokens": 15429200.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 5.449121427536011,
|
|
"epoch": 7.017398508699254,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015724715454416206,
|
|
"loss": 4.6276,
|
|
"mean_token_accuracy": 0.24014855474233626,
|
|
"num_tokens": 15439328.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 5.434017896652222,
|
|
"epoch": 7.021541010770505,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00015697497306208926,
|
|
"loss": 4.5939,
|
|
"mean_token_accuracy": 0.24022023528814315,
|
|
"num_tokens": 15447739.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 5.491503381729126,
|
|
"epoch": 7.0256835128417565,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001567030296493934,
|
|
"loss": 4.5539,
|
|
"mean_token_accuracy": 0.2426705002784729,
|
|
"num_tokens": 15456598.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 5.4921211242675785,
|
|
"epoch": 7.029826014913008,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015643132485461395,
|
|
"loss": 4.6763,
|
|
"mean_token_accuracy": 0.23373716175556183,
|
|
"num_tokens": 15466633.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 5.4619190216064455,
|
|
"epoch": 7.033968516984259,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001561598592258094,
|
|
"loss": 4.624,
|
|
"mean_token_accuracy": 0.24350053519010545,
|
|
"num_tokens": 15476809.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 5.390859746932984,
|
|
"epoch": 7.03811101905551,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00015588863331055541,
|
|
"loss": 4.5765,
|
|
"mean_token_accuracy": 0.23694039583206178,
|
|
"num_tokens": 15484964.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 5.398795461654663,
|
|
"epoch": 7.042253521126761,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001556176476559444,
|
|
"loss": 4.5825,
|
|
"mean_token_accuracy": 0.24087686985731124,
|
|
"num_tokens": 15493600.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.042253521126761,
|
|
"eval_entropy": 5.2711568976770335,
|
|
"eval_loss": 5.730020999908447,
|
|
"eval_mean_token_accuracy": 0.16879743175593012,
|
|
"eval_num_tokens": 15493600.0,
|
|
"eval_runtime": 2.1073,
|
|
"eval_samples_per_second": 1778.143,
|
|
"eval_steps_per_second": 222.565,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 5.383858346939087,
|
|
"epoch": 7.046396023198011,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015534690280858405,
|
|
"loss": 4.5991,
|
|
"mean_token_accuracy": 0.24276501834392547,
|
|
"num_tokens": 15502614.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 5.536665153503418,
|
|
"epoch": 7.050538525269262,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00015507639931459615,
|
|
"loss": 4.6855,
|
|
"mean_token_accuracy": 0.23332052379846574,
|
|
"num_tokens": 15511521.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 5.426029920578003,
|
|
"epoch": 7.054681027340513,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000154806137719616,
|
|
"loss": 4.6216,
|
|
"mean_token_accuracy": 0.23898648768663405,
|
|
"num_tokens": 15520220.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 5.41043529510498,
|
|
"epoch": 7.0588235294117645,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001545361185687906,
|
|
"loss": 4.5817,
|
|
"mean_token_accuracy": 0.24333779215812684,
|
|
"num_tokens": 15528924.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 5.361557197570801,
|
|
"epoch": 7.062966031483016,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00015426634240677824,
|
|
"loss": 4.5539,
|
|
"mean_token_accuracy": 0.23756618052721024,
|
|
"num_tokens": 15539722.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 5.474828577041626,
|
|
"epoch": 7.067108533554267,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015399680977774689,
|
|
"loss": 4.6342,
|
|
"mean_token_accuracy": 0.23686929643154145,
|
|
"num_tokens": 15548682.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 5.507525062561035,
|
|
"epoch": 7.071251035625518,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015372752122537318,
|
|
"loss": 4.6339,
|
|
"mean_token_accuracy": 0.2357267141342163,
|
|
"num_tokens": 15557508.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 5.478341341018677,
|
|
"epoch": 7.075393537696769,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015345847729284168,
|
|
"loss": 4.6486,
|
|
"mean_token_accuracy": 0.23708010613918304,
|
|
"num_tokens": 15566120.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 5.436391830444336,
|
|
"epoch": 7.07953603976802,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015318967852284335,
|
|
"loss": 4.5506,
|
|
"mean_token_accuracy": 0.24444466978311538,
|
|
"num_tokens": 15575166.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 5.397804641723633,
|
|
"epoch": 7.083678541839271,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00015292112545757476,
|
|
"loss": 4.572,
|
|
"mean_token_accuracy": 0.23554302453994752,
|
|
"num_tokens": 15585063.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 5.410696601867675,
|
|
"epoch": 7.087821043910522,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00015265281863873663,
|
|
"loss": 4.6244,
|
|
"mean_token_accuracy": 0.2350102946162224,
|
|
"num_tokens": 15594386.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 5.405501079559326,
|
|
"epoch": 7.091963545981773,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00015238475860753338,
|
|
"loss": 4.6256,
|
|
"mean_token_accuracy": 0.23459437638521194,
|
|
"num_tokens": 15603796.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 5.450531816482544,
|
|
"epoch": 7.096106048053024,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015211694590467118,
|
|
"loss": 4.568,
|
|
"mean_token_accuracy": 0.24386134445667268,
|
|
"num_tokens": 15612752.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 5.436820125579834,
|
|
"epoch": 7.100248550124275,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001518493810703575,
|
|
"loss": 4.6505,
|
|
"mean_token_accuracy": 0.22592575550079347,
|
|
"num_tokens": 15621995.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 5.395336627960205,
|
|
"epoch": 7.104391052195526,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015158206464430002,
|
|
"loss": 4.6068,
|
|
"mean_token_accuracy": 0.24140643328428268,
|
|
"num_tokens": 15630390.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 5.336005163192749,
|
|
"epoch": 7.108533554266777,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000151314997165705,
|
|
"loss": 4.5381,
|
|
"mean_token_accuracy": 0.2429471269249916,
|
|
"num_tokens": 15639723.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 5.365355730056763,
|
|
"epoch": 7.112676056338028,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00015104817917327694,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.2499431386590004,
|
|
"num_tokens": 15647755.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 5.4321998119354244,
|
|
"epoch": 7.116818558409279,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00015078161120521676,
|
|
"loss": 4.6283,
|
|
"mean_token_accuracy": 0.24222247898578644,
|
|
"num_tokens": 15656949.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 5.40409026145935,
|
|
"epoch": 7.12096106048053,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001505152937992211,
|
|
"loss": 4.57,
|
|
"mean_token_accuracy": 0.23804070204496383,
|
|
"num_tokens": 15666290.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 5.487924003601075,
|
|
"epoch": 7.125103562551781,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00015024922749248143,
|
|
"loss": 4.6373,
|
|
"mean_token_accuracy": 0.237720388174057,
|
|
"num_tokens": 15675728.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 5.448507928848267,
|
|
"epoch": 7.1292460646230325,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014998341282168243,
|
|
"loss": 4.5502,
|
|
"mean_token_accuracy": 0.24467460364103316,
|
|
"num_tokens": 15684519.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 5.464306402206421,
|
|
"epoch": 7.133388566694284,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001497178503230015,
|
|
"loss": 4.6546,
|
|
"mean_token_accuracy": 0.22924972623586654,
|
|
"num_tokens": 15693813.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 5.450425004959106,
|
|
"epoch": 7.137531068765535,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014945254053210712,
|
|
"loss": 4.6374,
|
|
"mean_token_accuracy": 0.23202786594629288,
|
|
"num_tokens": 15702245.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 5.456525897979736,
|
|
"epoch": 7.141673570836786,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014918748398415801,
|
|
"loss": 4.6436,
|
|
"mean_token_accuracy": 0.2358775779604912,
|
|
"num_tokens": 15712246.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 5.417597770690918,
|
|
"epoch": 7.145816072908037,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00014892268121380244,
|
|
"loss": 4.5713,
|
|
"mean_token_accuracy": 0.23698565661907195,
|
|
"num_tokens": 15720904.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 5.454194068908691,
|
|
"epoch": 7.149958574979287,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00014865813275517637,
|
|
"loss": 4.5816,
|
|
"mean_token_accuracy": 0.24436526894569396,
|
|
"num_tokens": 15730595.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 5.374666976928711,
|
|
"epoch": 7.154101077050538,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014839383914190302,
|
|
"loss": 4.5455,
|
|
"mean_token_accuracy": 0.24027044475078582,
|
|
"num_tokens": 15739475.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 5.411549997329712,
|
|
"epoch": 7.158243579121789,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001481298009070915,
|
|
"loss": 4.5797,
|
|
"mean_token_accuracy": 0.24344454258680343,
|
|
"num_tokens": 15748672.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 5.479576492309571,
|
|
"epoch": 7.16238608119304,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014786601858333566,
|
|
"loss": 4.5908,
|
|
"mean_token_accuracy": 0.24053577482700347,
|
|
"num_tokens": 15757540.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 5.4145832538604735,
|
|
"epoch": 7.1665285832642915,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00014760249270271341,
|
|
"loss": 4.61,
|
|
"mean_token_accuracy": 0.23149538934230804,
|
|
"num_tokens": 15767198.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 5.390424346923828,
|
|
"epoch": 7.170671085335543,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00014733922379678513,
|
|
"loss": 4.5911,
|
|
"mean_token_accuracy": 0.23455866575241088,
|
|
"num_tokens": 15776818.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 5.513827466964722,
|
|
"epoch": 7.174813587406794,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00014707621239659312,
|
|
"loss": 4.7107,
|
|
"mean_token_accuracy": 0.22772708237171174,
|
|
"num_tokens": 15786054.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 5.458751773834228,
|
|
"epoch": 7.178956089478045,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001468134590326601,
|
|
"loss": 4.6819,
|
|
"mean_token_accuracy": 0.22936149090528488,
|
|
"num_tokens": 15795367.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 5.493890047073364,
|
|
"epoch": 7.183098591549296,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001465509642349881,
|
|
"loss": 4.6556,
|
|
"mean_token_accuracy": 0.23393623232841493,
|
|
"num_tokens": 15805006.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 5.519836664199829,
|
|
"epoch": 7.187241093620547,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014628872853305808,
|
|
"loss": 4.6976,
|
|
"mean_token_accuracy": 0.23083092123270035,
|
|
"num_tokens": 15813978.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 5.494243812561035,
|
|
"epoch": 7.191383595691798,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014602675245582795,
|
|
"loss": 4.6557,
|
|
"mean_token_accuracy": 0.23307705223560332,
|
|
"num_tokens": 15824156.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 5.530265522003174,
|
|
"epoch": 7.195526097763049,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014576503653173219,
|
|
"loss": 4.7257,
|
|
"mean_token_accuracy": 0.22632366567850112,
|
|
"num_tokens": 15833514.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 5.448892831802368,
|
|
"epoch": 7.1996685998342995,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00014550358128868036,
|
|
"loss": 4.6625,
|
|
"mean_token_accuracy": 0.23346309512853622,
|
|
"num_tokens": 15842047.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 5.486913204193115,
|
|
"epoch": 7.203811101905551,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014524238725405615,
|
|
"loss": 4.6937,
|
|
"mean_token_accuracy": 0.2285568356513977,
|
|
"num_tokens": 15851916.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 5.463122749328614,
|
|
"epoch": 7.207953603976802,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00014498145495471664,
|
|
"loss": 4.6267,
|
|
"mean_token_accuracy": 0.23774217963218688,
|
|
"num_tokens": 15860776.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 5.452417802810669,
|
|
"epoch": 7.212096106048053,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014472078491699076,
|
|
"loss": 4.6864,
|
|
"mean_token_accuracy": 0.23721128106117248,
|
|
"num_tokens": 15870272.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 5.458945846557617,
|
|
"epoch": 7.216238608119304,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001444603776666784,
|
|
"loss": 4.6089,
|
|
"mean_token_accuracy": 0.24200923293828963,
|
|
"num_tokens": 15879448.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 5.527491235733033,
|
|
"epoch": 7.220381110190555,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014420023372904944,
|
|
"loss": 4.6923,
|
|
"mean_token_accuracy": 0.2308870568871498,
|
|
"num_tokens": 15888806.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 5.4437816619873045,
|
|
"epoch": 7.224523612261806,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014394035362884274,
|
|
"loss": 4.6556,
|
|
"mean_token_accuracy": 0.23654720783233643,
|
|
"num_tokens": 15898443.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 5.41554479598999,
|
|
"epoch": 7.228666114333057,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014368073789026482,
|
|
"loss": 4.5573,
|
|
"mean_token_accuracy": 0.2407093659043312,
|
|
"num_tokens": 15907759.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 5.450989866256714,
|
|
"epoch": 7.232808616404308,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014342138703698882,
|
|
"loss": 4.5645,
|
|
"mean_token_accuracy": 0.24144845455884933,
|
|
"num_tokens": 15916687.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 5.448304224014282,
|
|
"epoch": 7.2369511184755595,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000143162301592154,
|
|
"loss": 4.6154,
|
|
"mean_token_accuracy": 0.23854661881923675,
|
|
"num_tokens": 15925422.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 5.446715545654297,
|
|
"epoch": 7.241093620546811,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00014290348207836385,
|
|
"loss": 4.6208,
|
|
"mean_token_accuracy": 0.23098595738410949,
|
|
"num_tokens": 15934279.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 5.439447355270386,
|
|
"epoch": 7.245236122618062,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001426449290176857,
|
|
"loss": 4.5834,
|
|
"mean_token_accuracy": 0.23927582204341888,
|
|
"num_tokens": 15943565.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 5.384623622894287,
|
|
"epoch": 7.249378624689312,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014238664293164934,
|
|
"loss": 4.5918,
|
|
"mean_token_accuracy": 0.24010520279407502,
|
|
"num_tokens": 15952360.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 5.447373104095459,
|
|
"epoch": 7.253521126760563,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001421286243412459,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.23103807717561722,
|
|
"num_tokens": 15960715.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 5.491124677658081,
|
|
"epoch": 7.257663628831814,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00014187087376692724,
|
|
"loss": 4.6506,
|
|
"mean_token_accuracy": 0.23849486708641052,
|
|
"num_tokens": 15969543.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 5.37595796585083,
|
|
"epoch": 7.261806130903065,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001416133917286043,
|
|
"loss": 4.5747,
|
|
"mean_token_accuracy": 0.23537552803754808,
|
|
"num_tokens": 15978256.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 5.3844517230987545,
|
|
"epoch": 7.265948632974316,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001413561787456466,
|
|
"loss": 4.5669,
|
|
"mean_token_accuracy": 0.2360729083418846,
|
|
"num_tokens": 15987646.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 5.41820707321167,
|
|
"epoch": 7.2700911350455675,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014109923533688068,
|
|
"loss": 4.5932,
|
|
"mean_token_accuracy": 0.2336271658539772,
|
|
"num_tokens": 15996767.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 5.452511930465699,
|
|
"epoch": 7.274233637116819,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001408425620205897,
|
|
"loss": 4.6172,
|
|
"mean_token_accuracy": 0.2332585021853447,
|
|
"num_tokens": 16007170.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 5.492886304855347,
|
|
"epoch": 7.27837613918807,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001405861593145116,
|
|
"loss": 4.6505,
|
|
"mean_token_accuracy": 0.23124625235795976,
|
|
"num_tokens": 16015354.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 5.505690622329712,
|
|
"epoch": 7.282518641259321,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014033002773583872,
|
|
"loss": 4.6403,
|
|
"mean_token_accuracy": 0.22802675366401673,
|
|
"num_tokens": 16024600.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 5.492707061767578,
|
|
"epoch": 7.286661143330572,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001400741678012164,
|
|
"loss": 4.6233,
|
|
"mean_token_accuracy": 0.24178002029657364,
|
|
"num_tokens": 16033301.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 5.445309114456177,
|
|
"epoch": 7.290803645401823,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00013981858002674223,
|
|
"loss": 4.6619,
|
|
"mean_token_accuracy": 0.23175255507230758,
|
|
"num_tokens": 16043539.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 5.495531320571899,
|
|
"epoch": 7.294946147473074,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001395632649279645,
|
|
"loss": 4.6995,
|
|
"mean_token_accuracy": 0.2296041488647461,
|
|
"num_tokens": 16052861.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 5.490782690048218,
|
|
"epoch": 7.299088649544325,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013930822301988177,
|
|
"loss": 4.587,
|
|
"mean_token_accuracy": 0.24263111054897307,
|
|
"num_tokens": 16062454.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 5.422387981414795,
|
|
"epoch": 7.3032311516155755,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013905345481694147,
|
|
"loss": 4.6084,
|
|
"mean_token_accuracy": 0.23536452502012253,
|
|
"num_tokens": 16071432.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 5.420614576339721,
|
|
"epoch": 7.307373653686827,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001387989608330388,
|
|
"loss": 4.5814,
|
|
"mean_token_accuracy": 0.23629239201545715,
|
|
"num_tokens": 16080411.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 5.454785251617432,
|
|
"epoch": 7.311516155758078,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013854474158151602,
|
|
"loss": 4.6206,
|
|
"mean_token_accuracy": 0.23535432666540146,
|
|
"num_tokens": 16089396.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 5.494325160980225,
|
|
"epoch": 7.315658657829329,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013829079757516106,
|
|
"loss": 4.7136,
|
|
"mean_token_accuracy": 0.2294663444161415,
|
|
"num_tokens": 16099308.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 5.443635034561157,
|
|
"epoch": 7.31980115990058,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013803712932620681,
|
|
"loss": 4.6432,
|
|
"mean_token_accuracy": 0.22834567874670028,
|
|
"num_tokens": 16109336.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 5.489700174331665,
|
|
"epoch": 7.323943661971831,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001377837373463298,
|
|
"loss": 4.6137,
|
|
"mean_token_accuracy": 0.23566730171442032,
|
|
"num_tokens": 16118483.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 5.47642822265625,
|
|
"epoch": 7.328086164043082,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001375306221466494,
|
|
"loss": 4.6214,
|
|
"mean_token_accuracy": 0.2339439496397972,
|
|
"num_tokens": 16127302.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 5.4224936962127686,
|
|
"epoch": 7.332228666114333,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013727778423772654,
|
|
"loss": 4.673,
|
|
"mean_token_accuracy": 0.2349598616361618,
|
|
"num_tokens": 16136361.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 5.5139096736907955,
|
|
"epoch": 7.336371168185584,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013702522412956304,
|
|
"loss": 4.7221,
|
|
"mean_token_accuracy": 0.22534615844488143,
|
|
"num_tokens": 16146071.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 5.491855621337891,
|
|
"epoch": 7.340513670256835,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013677294233160016,
|
|
"loss": 4.6585,
|
|
"mean_token_accuracy": 0.227964885532856,
|
|
"num_tokens": 16154606.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 5.41273078918457,
|
|
"epoch": 7.3446561723280865,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00013652093935271776,
|
|
"loss": 4.6161,
|
|
"mean_token_accuracy": 0.24378422498703003,
|
|
"num_tokens": 16163648.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 5.490383958816528,
|
|
"epoch": 7.348798674399337,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013626921570123357,
|
|
"loss": 4.6659,
|
|
"mean_token_accuracy": 0.2314506232738495,
|
|
"num_tokens": 16172693.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 5.4631383419036865,
|
|
"epoch": 7.352941176470588,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00013601777188490156,
|
|
"loss": 4.5853,
|
|
"mean_token_accuracy": 0.24082445949316025,
|
|
"num_tokens": 16181689.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 5.441583490371704,
|
|
"epoch": 7.357083678541839,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001357666084109115,
|
|
"loss": 4.6806,
|
|
"mean_token_accuracy": 0.23397230952978135,
|
|
"num_tokens": 16191254.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 5.455285549163818,
|
|
"epoch": 7.36122618061309,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001355157257858875,
|
|
"loss": 4.5999,
|
|
"mean_token_accuracy": 0.2364482119679451,
|
|
"num_tokens": 16199613.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 5.398629522323608,
|
|
"epoch": 7.365368682684341,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00013526512451588718,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.23717987686395645,
|
|
"num_tokens": 16208345.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 5.4645223140716555,
|
|
"epoch": 7.369511184755592,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001350148051064008,
|
|
"loss": 4.6747,
|
|
"mean_token_accuracy": 0.22994847893714904,
|
|
"num_tokens": 16217964.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 5.482628726959229,
|
|
"epoch": 7.373653686826843,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013476476806234983,
|
|
"loss": 4.7264,
|
|
"mean_token_accuracy": 0.230340938270092,
|
|
"num_tokens": 16227140.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 5.55241026878357,
|
|
"epoch": 7.3777961888980945,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001345150138880865,
|
|
"loss": 4.6689,
|
|
"mean_token_accuracy": 0.22891770154237748,
|
|
"num_tokens": 16237416.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 5.5532080173492435,
|
|
"epoch": 7.381938690969346,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013426554308739214,
|
|
"loss": 4.6946,
|
|
"mean_token_accuracy": 0.23257962018251419,
|
|
"num_tokens": 16246682.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 5.459280014038086,
|
|
"epoch": 7.386081193040597,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013401635616347655,
|
|
"loss": 4.6812,
|
|
"mean_token_accuracy": 0.23148078620433807,
|
|
"num_tokens": 16255450.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 5.351534700393676,
|
|
"epoch": 7.390223695111848,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00013376745361897712,
|
|
"loss": 4.562,
|
|
"mean_token_accuracy": 0.23984588384628297,
|
|
"num_tokens": 16265014.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 5.494270610809326,
|
|
"epoch": 7.394366197183099,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013351883595595743,
|
|
"loss": 4.6493,
|
|
"mean_token_accuracy": 0.23885316550731658,
|
|
"num_tokens": 16273309.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 5.431078720092773,
|
|
"epoch": 7.39850869925435,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001332705036759065,
|
|
"loss": 4.6143,
|
|
"mean_token_accuracy": 0.23021501749753953,
|
|
"num_tokens": 16282129.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 5.390668678283691,
|
|
"epoch": 7.4026512013256,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001330224572797377,
|
|
"loss": 4.5848,
|
|
"mean_token_accuracy": 0.23649993240833284,
|
|
"num_tokens": 16291208.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 5.489054441452026,
|
|
"epoch": 7.406793703396851,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013277469726778756,
|
|
"loss": 4.5869,
|
|
"mean_token_accuracy": 0.2421462282538414,
|
|
"num_tokens": 16300111.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 5.451315450668335,
|
|
"epoch": 7.4109362054681025,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001325272241398153,
|
|
"loss": 4.5642,
|
|
"mean_token_accuracy": 0.24056823402643204,
|
|
"num_tokens": 16308653.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 5.506957864761352,
|
|
"epoch": 7.415078707539354,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013228003839500113,
|
|
"loss": 4.6589,
|
|
"mean_token_accuracy": 0.22678677141666412,
|
|
"num_tokens": 16317986.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 5.446587467193604,
|
|
"epoch": 7.419221209610605,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001320331405319458,
|
|
"loss": 4.6413,
|
|
"mean_token_accuracy": 0.2311987429857254,
|
|
"num_tokens": 16326889.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 5.38463397026062,
|
|
"epoch": 7.423363711681856,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00013178653104866926,
|
|
"loss": 4.6528,
|
|
"mean_token_accuracy": 0.23329097628593445,
|
|
"num_tokens": 16336121.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 5.441002988815308,
|
|
"epoch": 7.427506213753107,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013154021044260966,
|
|
"loss": 4.7162,
|
|
"mean_token_accuracy": 0.22670302391052247,
|
|
"num_tokens": 16345216.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 5.409039497375488,
|
|
"epoch": 7.431648715824358,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00013129417921062275,
|
|
"loss": 4.5635,
|
|
"mean_token_accuracy": 0.24468129873275757,
|
|
"num_tokens": 16353735.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 5.437806415557861,
|
|
"epoch": 7.435791217895609,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001310484378489803,
|
|
"loss": 4.5827,
|
|
"mean_token_accuracy": 0.24198826551437377,
|
|
"num_tokens": 16362474.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 5.458545398712158,
|
|
"epoch": 7.43993371996686,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013080298685336957,
|
|
"loss": 4.7054,
|
|
"mean_token_accuracy": 0.22612921744585038,
|
|
"num_tokens": 16371948.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 5.395973587036133,
|
|
"epoch": 7.444076222038111,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013055782671889205,
|
|
"loss": 4.5851,
|
|
"mean_token_accuracy": 0.23771227151155472,
|
|
"num_tokens": 16380849.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 5.470195198059082,
|
|
"epoch": 7.448218724109362,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001303129579400624,
|
|
"loss": 4.6977,
|
|
"mean_token_accuracy": 0.22957099825143815,
|
|
"num_tokens": 16389495.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 5.46120433807373,
|
|
"epoch": 7.452361226180613,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001300683810108078,
|
|
"loss": 4.5979,
|
|
"mean_token_accuracy": 0.23696892708539963,
|
|
"num_tokens": 16397744.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 5.386603116989136,
|
|
"epoch": 7.456503728251864,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001298240964244666,
|
|
"loss": 4.614,
|
|
"mean_token_accuracy": 0.2303352400660515,
|
|
"num_tokens": 16406785.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.456503728251864,
|
|
"eval_entropy": 5.255923632111377,
|
|
"eval_loss": 5.731689929962158,
|
|
"eval_mean_token_accuracy": 0.1686617727758788,
|
|
"eval_num_tokens": 16406785.0,
|
|
"eval_runtime": 2.11,
|
|
"eval_samples_per_second": 1775.802,
|
|
"eval_steps_per_second": 222.271,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 5.433587312698364,
|
|
"epoch": 7.460646230323115,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001295801046737876,
|
|
"loss": 4.6418,
|
|
"mean_token_accuracy": 0.24157550632953645,
|
|
"num_tokens": 16415768.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 5.371090459823608,
|
|
"epoch": 7.464788732394366,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.00012933640625092867,
|
|
"loss": 4.5976,
|
|
"mean_token_accuracy": 0.2386163964867592,
|
|
"num_tokens": 16424722.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 5.452657985687256,
|
|
"epoch": 7.468931234465617,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012909300164745636,
|
|
"loss": 4.6141,
|
|
"mean_token_accuracy": 0.2396414503455162,
|
|
"num_tokens": 16433797.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 5.422679328918457,
|
|
"epoch": 7.473073736536868,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012884989135434423,
|
|
"loss": 4.5352,
|
|
"mean_token_accuracy": 0.24652525186538696,
|
|
"num_tokens": 16442537.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 5.46554388999939,
|
|
"epoch": 7.477216238608119,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00012860707586197218,
|
|
"loss": 4.652,
|
|
"mean_token_accuracy": 0.23123886734247207,
|
|
"num_tokens": 16451440.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 5.42999529838562,
|
|
"epoch": 7.4813587406793705,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012836455566012586,
|
|
"loss": 4.6375,
|
|
"mean_token_accuracy": 0.23561885952949524,
|
|
"num_tokens": 16459358.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 5.416983366012573,
|
|
"epoch": 7.485501242750622,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00012812233123799481,
|
|
"loss": 4.5356,
|
|
"mean_token_accuracy": 0.24424079060554504,
|
|
"num_tokens": 16467584.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 5.450838232040406,
|
|
"epoch": 7.489643744821873,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012788040308417232,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.23527542799711226,
|
|
"num_tokens": 16476674.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 5.419613695144653,
|
|
"epoch": 7.493786246893124,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012763877168665383,
|
|
"loss": 4.5275,
|
|
"mean_token_accuracy": 0.24421215057373047,
|
|
"num_tokens": 16485422.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 5.432162284851074,
|
|
"epoch": 7.497928748964375,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012739743753283617,
|
|
"loss": 4.7129,
|
|
"mean_token_accuracy": 0.23073383122682573,
|
|
"num_tokens": 16494024.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 5.407644653320313,
|
|
"epoch": 7.502071251035625,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001271564011095169,
|
|
"loss": 4.6975,
|
|
"mean_token_accuracy": 0.23130497038364412,
|
|
"num_tokens": 16503805.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 5.461945056915283,
|
|
"epoch": 7.506213753106876,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001269156629028927,
|
|
"loss": 4.6448,
|
|
"mean_token_accuracy": 0.2288455680012703,
|
|
"num_tokens": 16512559.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 5.508998966217041,
|
|
"epoch": 7.510356255178127,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001266752233985589,
|
|
"loss": 4.6945,
|
|
"mean_token_accuracy": 0.22597112357616425,
|
|
"num_tokens": 16521910.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 5.45227427482605,
|
|
"epoch": 7.514498757249378,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012643508308150822,
|
|
"loss": 4.6279,
|
|
"mean_token_accuracy": 0.23708249181509017,
|
|
"num_tokens": 16530802.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 5.409143352508545,
|
|
"epoch": 7.5186412593206295,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001261952424361299,
|
|
"loss": 4.6146,
|
|
"mean_token_accuracy": 0.2355431005358696,
|
|
"num_tokens": 16540519.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 5.420194435119629,
|
|
"epoch": 7.522783761391881,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001259557019462088,
|
|
"loss": 4.5893,
|
|
"mean_token_accuracy": 0.2345808893442154,
|
|
"num_tokens": 16549433.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 5.43223204612732,
|
|
"epoch": 7.526926263463132,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012571646209492417,
|
|
"loss": 4.6348,
|
|
"mean_token_accuracy": 0.23276626765727998,
|
|
"num_tokens": 16558128.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 5.436632251739502,
|
|
"epoch": 7.531068765534383,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012547752336484908,
|
|
"loss": 4.5723,
|
|
"mean_token_accuracy": 0.24057172983884811,
|
|
"num_tokens": 16566636.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 5.491369819641113,
|
|
"epoch": 7.535211267605634,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012523888623794892,
|
|
"loss": 4.5974,
|
|
"mean_token_accuracy": 0.23670983761548997,
|
|
"num_tokens": 16575036.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 5.4118022441864015,
|
|
"epoch": 7.539353769676885,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001250005511955809,
|
|
"loss": 4.6119,
|
|
"mean_token_accuracy": 0.23590679317712784,
|
|
"num_tokens": 16584146.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 5.489098739624024,
|
|
"epoch": 7.543496271748136,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001247625187184929,
|
|
"loss": 4.7092,
|
|
"mean_token_accuracy": 0.23371747136116028,
|
|
"num_tokens": 16593915.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 5.460393476486206,
|
|
"epoch": 7.547638773819386,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001245247892868223,
|
|
"loss": 4.5509,
|
|
"mean_token_accuracy": 0.25090702474117277,
|
|
"num_tokens": 16601490.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 5.426677465438843,
|
|
"epoch": 7.551781275890638,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001242873633800955,
|
|
"loss": 4.645,
|
|
"mean_token_accuracy": 0.2287042945623398,
|
|
"num_tokens": 16610972.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 5.465074300765991,
|
|
"epoch": 7.555923777961889,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00012405024147722643,
|
|
"loss": 4.6217,
|
|
"mean_token_accuracy": 0.23180095702409745,
|
|
"num_tokens": 16619388.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 5.346891832351685,
|
|
"epoch": 7.56006628003314,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001238134240565158,
|
|
"loss": 4.564,
|
|
"mean_token_accuracy": 0.23870298713445665,
|
|
"num_tokens": 16628265.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 5.432445383071899,
|
|
"epoch": 7.564208782104391,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0001235769115956504,
|
|
"loss": 4.5537,
|
|
"mean_token_accuracy": 0.24641286283731462,
|
|
"num_tokens": 16636809.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 5.49163784980774,
|
|
"epoch": 7.568351284175642,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00012334070457170151,
|
|
"loss": 4.7344,
|
|
"mean_token_accuracy": 0.22605728954076768,
|
|
"num_tokens": 16646084.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 5.487151098251343,
|
|
"epoch": 7.572493786246893,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012310480346112472,
|
|
"loss": 4.7237,
|
|
"mean_token_accuracy": 0.2246138334274292,
|
|
"num_tokens": 16656196.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 5.418950366973877,
|
|
"epoch": 7.576636288318144,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00012286920873975815,
|
|
"loss": 4.6597,
|
|
"mean_token_accuracy": 0.23062705844640732,
|
|
"num_tokens": 16665770.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 5.4313013553619385,
|
|
"epoch": 7.580778790389395,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012263392088282226,
|
|
"loss": 4.5651,
|
|
"mean_token_accuracy": 0.23793004602193832,
|
|
"num_tokens": 16674206.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 5.394683933258056,
|
|
"epoch": 7.584921292460646,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012239894036491833,
|
|
"loss": 4.6086,
|
|
"mean_token_accuracy": 0.2338030979037285,
|
|
"num_tokens": 16683977.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 5.404988098144531,
|
|
"epoch": 7.5890637945318975,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012216426766002762,
|
|
"loss": 4.59,
|
|
"mean_token_accuracy": 0.23571064621210097,
|
|
"num_tokens": 16693117.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 5.431485509872436,
|
|
"epoch": 7.593206296603149,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012192990324151079,
|
|
"loss": 4.6031,
|
|
"mean_token_accuracy": 0.23800773471593856,
|
|
"num_tokens": 16701631.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 5.463937091827392,
|
|
"epoch": 7.5973487986744,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012169584758210635,
|
|
"loss": 4.6925,
|
|
"mean_token_accuracy": 0.23966999500989913,
|
|
"num_tokens": 16710337.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 5.409084272384644,
|
|
"epoch": 7.60149130074565,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012146210115393025,
|
|
"loss": 4.5858,
|
|
"mean_token_accuracy": 0.23898185640573502,
|
|
"num_tokens": 16719046.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 5.474574232101441,
|
|
"epoch": 7.605633802816901,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00012122866442847455,
|
|
"loss": 4.6746,
|
|
"mean_token_accuracy": 0.23025610148906708,
|
|
"num_tokens": 16727860.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 5.486311626434326,
|
|
"epoch": 7.609776304888152,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012099553787660654,
|
|
"loss": 4.7082,
|
|
"mean_token_accuracy": 0.23690710812807084,
|
|
"num_tokens": 16737839.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 5.364979791641235,
|
|
"epoch": 7.613918806959403,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00012076272196856808,
|
|
"loss": 4.5959,
|
|
"mean_token_accuracy": 0.23230910003185273,
|
|
"num_tokens": 16746171.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 5.401865386962891,
|
|
"epoch": 7.618061309030654,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001205302171739742,
|
|
"loss": 4.572,
|
|
"mean_token_accuracy": 0.24215463250875474,
|
|
"num_tokens": 16755236.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 5.452185010910034,
|
|
"epoch": 7.6222038111019055,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012029802396181256,
|
|
"loss": 4.6221,
|
|
"mean_token_accuracy": 0.2302182748913765,
|
|
"num_tokens": 16764126.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 5.443011713027954,
|
|
"epoch": 7.626346313173157,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00012006614280044225,
|
|
"loss": 4.6567,
|
|
"mean_token_accuracy": 0.2415397509932518,
|
|
"num_tokens": 16772861.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 5.436422204971313,
|
|
"epoch": 7.630488815244408,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011983457415759274,
|
|
"loss": 4.6423,
|
|
"mean_token_accuracy": 0.24018560498952865,
|
|
"num_tokens": 16782119.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 5.486966228485107,
|
|
"epoch": 7.634631317315659,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00011960331850036352,
|
|
"loss": 4.6646,
|
|
"mean_token_accuracy": 0.22786333858966829,
|
|
"num_tokens": 16790531.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 5.414906978607178,
|
|
"epoch": 7.63877381938691,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00011937237629522233,
|
|
"loss": 4.603,
|
|
"mean_token_accuracy": 0.23567911386489868,
|
|
"num_tokens": 16799516.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 5.437748289108276,
|
|
"epoch": 7.642916321458161,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011914174800800501,
|
|
"loss": 4.6188,
|
|
"mean_token_accuracy": 0.2296470195055008,
|
|
"num_tokens": 16807830.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 5.467829465866089,
|
|
"epoch": 7.647058823529412,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011891143410391387,
|
|
"loss": 4.6986,
|
|
"mean_token_accuracy": 0.2282799482345581,
|
|
"num_tokens": 16816987.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 5.376076936721802,
|
|
"epoch": 7.651201325600663,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011868143504751732,
|
|
"loss": 4.533,
|
|
"mean_token_accuracy": 0.24274307042360305,
|
|
"num_tokens": 16825723.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 5.434685707092285,
|
|
"epoch": 7.6553438276719135,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00011845175130274848,
|
|
"loss": 4.6625,
|
|
"mean_token_accuracy": 0.23135142922401428,
|
|
"num_tokens": 16835429.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 5.430702781677246,
|
|
"epoch": 7.659486329743165,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011822238333290475,
|
|
"loss": 4.5898,
|
|
"mean_token_accuracy": 0.2488136112689972,
|
|
"num_tokens": 16844693.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 5.44289608001709,
|
|
"epoch": 7.663628831814416,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011799333160064617,
|
|
"loss": 4.6239,
|
|
"mean_token_accuracy": 0.22939321398735046,
|
|
"num_tokens": 16853058.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 5.516484928131104,
|
|
"epoch": 7.667771333885667,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001177645965679953,
|
|
"loss": 4.6759,
|
|
"mean_token_accuracy": 0.22799726575613022,
|
|
"num_tokens": 16863017.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 5.395643615722657,
|
|
"epoch": 7.671913835956918,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00011753617869633561,
|
|
"loss": 4.5332,
|
|
"mean_token_accuracy": 0.2447441890835762,
|
|
"num_tokens": 16872054.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 5.4911168098449705,
|
|
"epoch": 7.676056338028169,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011730807844641085,
|
|
"loss": 4.6855,
|
|
"mean_token_accuracy": 0.23557499796152115,
|
|
"num_tokens": 16881554.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 5.4015625,
|
|
"epoch": 7.68019884009942,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011708029627832431,
|
|
"loss": 4.6502,
|
|
"mean_token_accuracy": 0.2340620458126068,
|
|
"num_tokens": 16890050.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 5.474047422409058,
|
|
"epoch": 7.684341342170671,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011685283265153732,
|
|
"loss": 4.5811,
|
|
"mean_token_accuracy": 0.23611383140087128,
|
|
"num_tokens": 16899298.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 5.513465309143067,
|
|
"epoch": 7.688483844241922,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011662568802486905,
|
|
"loss": 4.7006,
|
|
"mean_token_accuracy": 0.2284679189324379,
|
|
"num_tokens": 16908227.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 5.488403844833374,
|
|
"epoch": 7.692626346313173,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00011639886285649489,
|
|
"loss": 4.6848,
|
|
"mean_token_accuracy": 0.22802948951721191,
|
|
"num_tokens": 16917708.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 5.397838878631592,
|
|
"epoch": 7.6967688483844245,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00011617235760394614,
|
|
"loss": 4.5183,
|
|
"mean_token_accuracy": 0.2461710587143898,
|
|
"num_tokens": 16926108.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 5.464581394195557,
|
|
"epoch": 7.700911350455675,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001159461727241085,
|
|
"loss": 4.6237,
|
|
"mean_token_accuracy": 0.2309143528342247,
|
|
"num_tokens": 16935821.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 5.41339693069458,
|
|
"epoch": 7.705053852526927,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001157203086732217,
|
|
"loss": 4.669,
|
|
"mean_token_accuracy": 0.2354589283466339,
|
|
"num_tokens": 16944899.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 5.42128357887268,
|
|
"epoch": 7.709196354598177,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00011549476590687811,
|
|
"loss": 4.6236,
|
|
"mean_token_accuracy": 0.2403522863984108,
|
|
"num_tokens": 16953745.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 5.540648794174194,
|
|
"epoch": 7.713338856669428,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011526954488002227,
|
|
"loss": 4.7159,
|
|
"mean_token_accuracy": 0.23205580562353134,
|
|
"num_tokens": 16962940.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 5.447509717941284,
|
|
"epoch": 7.717481358740679,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00011504464604694953,
|
|
"loss": 4.6411,
|
|
"mean_token_accuracy": 0.23500789403915406,
|
|
"num_tokens": 16971322.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 5.4797309875488285,
|
|
"epoch": 7.72162386081193,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011482006986130533,
|
|
"loss": 4.5727,
|
|
"mean_token_accuracy": 0.23681646287441255,
|
|
"num_tokens": 16980066.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 5.4371950149536135,
|
|
"epoch": 7.725766362883181,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011459581677608453,
|
|
"loss": 4.6289,
|
|
"mean_token_accuracy": 0.2348506584763527,
|
|
"num_tokens": 16989202.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 5.4079631805419925,
|
|
"epoch": 7.7299088649544325,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011437188724362997,
|
|
"loss": 4.6596,
|
|
"mean_token_accuracy": 0.22531077563762664,
|
|
"num_tokens": 16999267.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 5.369661712646485,
|
|
"epoch": 7.734051367025684,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011414828171563214,
|
|
"loss": 4.6129,
|
|
"mean_token_accuracy": 0.23921015113592148,
|
|
"num_tokens": 17008252.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 5.418490695953369,
|
|
"epoch": 7.738193869096935,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011392500064312773,
|
|
"loss": 4.6147,
|
|
"mean_token_accuracy": 0.23070580065250396,
|
|
"num_tokens": 17017568.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 5.46760778427124,
|
|
"epoch": 7.742336371168186,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.000113702044476499,
|
|
"loss": 4.6311,
|
|
"mean_token_accuracy": 0.234436696767807,
|
|
"num_tokens": 17026206.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 5.438302898406983,
|
|
"epoch": 7.746478873239437,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011347941366547303,
|
|
"loss": 4.6118,
|
|
"mean_token_accuracy": 0.23505968004465103,
|
|
"num_tokens": 17035114.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 5.464243650436401,
|
|
"epoch": 7.750621375310688,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001132571086591204,
|
|
"loss": 4.6888,
|
|
"mean_token_accuracy": 0.22960633039474487,
|
|
"num_tokens": 17044459.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 5.4405186653137205,
|
|
"epoch": 7.754763877381938,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001130351299058547,
|
|
"loss": 4.6673,
|
|
"mean_token_accuracy": 0.22523288875818254,
|
|
"num_tokens": 17053985.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 5.406127977371216,
|
|
"epoch": 7.758906379453189,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011281347785343127,
|
|
"loss": 4.5134,
|
|
"mean_token_accuracy": 0.2389586240053177,
|
|
"num_tokens": 17063136.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 5.444184350967407,
|
|
"epoch": 7.7630488815244405,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00011259215294894647,
|
|
"loss": 4.6701,
|
|
"mean_token_accuracy": 0.22709797322750092,
|
|
"num_tokens": 17072423.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 5.425132465362549,
|
|
"epoch": 7.767191383595692,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011237115563883693,
|
|
"loss": 4.6355,
|
|
"mean_token_accuracy": 0.23451708257198334,
|
|
"num_tokens": 17080578.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 5.445489454269409,
|
|
"epoch": 7.771333885666943,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011215048636887825,
|
|
"loss": 4.5682,
|
|
"mean_token_accuracy": 0.24168250113725662,
|
|
"num_tokens": 17089995.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 5.387733268737793,
|
|
"epoch": 7.775476387738194,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001119301455841846,
|
|
"loss": 4.556,
|
|
"mean_token_accuracy": 0.23807462900877,
|
|
"num_tokens": 17099659.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 5.486556625366211,
|
|
"epoch": 7.779618889809445,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00011171013372920735,
|
|
"loss": 4.6661,
|
|
"mean_token_accuracy": 0.23664889186620713,
|
|
"num_tokens": 17109972.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 5.527633714675903,
|
|
"epoch": 7.783761391880696,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00011149045124773441,
|
|
"loss": 4.7499,
|
|
"mean_token_accuracy": 0.22407512813806535,
|
|
"num_tokens": 17120288.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 5.490897941589355,
|
|
"epoch": 7.787903893951947,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00011127109858288945,
|
|
"loss": 4.6728,
|
|
"mean_token_accuracy": 0.23673882186412812,
|
|
"num_tokens": 17130577.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 5.481154918670654,
|
|
"epoch": 7.792046396023198,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011105207617713065,
|
|
"loss": 4.7391,
|
|
"mean_token_accuracy": 0.22611385434865952,
|
|
"num_tokens": 17140782.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 5.386972045898437,
|
|
"epoch": 7.796188898094449,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011083338447225027,
|
|
"loss": 4.6282,
|
|
"mean_token_accuracy": 0.2377781793475151,
|
|
"num_tokens": 17150013.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 5.495413208007813,
|
|
"epoch": 7.8003314001657,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011061502390937331,
|
|
"loss": 4.6628,
|
|
"mean_token_accuracy": 0.22644497007131575,
|
|
"num_tokens": 17159349.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 5.429597425460815,
|
|
"epoch": 7.804473902236952,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001103969949289568,
|
|
"loss": 4.7043,
|
|
"mean_token_accuracy": 0.23896088302135468,
|
|
"num_tokens": 17168063.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 5.447285461425781,
|
|
"epoch": 7.808616404308202,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00011017929797078919,
|
|
"loss": 4.6341,
|
|
"mean_token_accuracy": 0.23011545687913895,
|
|
"num_tokens": 17177087.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 5.454278421401978,
|
|
"epoch": 7.812758906379453,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00010996193347398886,
|
|
"loss": 4.6362,
|
|
"mean_token_accuracy": 0.23783269375562668,
|
|
"num_tokens": 17186875.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 5.4562290668487545,
|
|
"epoch": 7.816901408450704,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010974490187700394,
|
|
"loss": 4.6588,
|
|
"mean_token_accuracy": 0.2312362477183342,
|
|
"num_tokens": 17195666.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 5.452163457870483,
|
|
"epoch": 7.821043910521955,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001095282036176107,
|
|
"loss": 4.6217,
|
|
"mean_token_accuracy": 0.2354704335331917,
|
|
"num_tokens": 17205145.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 5.504744243621826,
|
|
"epoch": 7.825186412593206,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010931183913291335,
|
|
"loss": 4.6672,
|
|
"mean_token_accuracy": 0.2313088670372963,
|
|
"num_tokens": 17214106.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 5.3981689453125,
|
|
"epoch": 7.829328914664457,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010909580885934265,
|
|
"loss": 4.5494,
|
|
"mean_token_accuracy": 0.24952173680067063,
|
|
"num_tokens": 17223343.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 5.43416395187378,
|
|
"epoch": 7.8334714167357085,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00010888011323265517,
|
|
"loss": 4.678,
|
|
"mean_token_accuracy": 0.22725510597229004,
|
|
"num_tokens": 17233528.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 5.405489921569824,
|
|
"epoch": 7.83761391880696,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010866475268793275,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.23560418784618378,
|
|
"num_tokens": 17242026.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 5.3866559028625485,
|
|
"epoch": 7.841756420878211,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010844972765958096,
|
|
"loss": 4.5971,
|
|
"mean_token_accuracy": 0.24127698838710784,
|
|
"num_tokens": 17251203.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 5.339055490493775,
|
|
"epoch": 7.845898922949462,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010823503858132894,
|
|
"loss": 4.5132,
|
|
"mean_token_accuracy": 0.24420754760503768,
|
|
"num_tokens": 17260072.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 5.426773405075073,
|
|
"epoch": 7.850041425020713,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010802068588622793,
|
|
"loss": 4.6467,
|
|
"mean_token_accuracy": 0.23799521028995513,
|
|
"num_tokens": 17269015.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 5.406404209136963,
|
|
"epoch": 7.854183927091963,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010780667000665067,
|
|
"loss": 4.5925,
|
|
"mean_token_accuracy": 0.24103019684553145,
|
|
"num_tokens": 17277721.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 5.4329689025878904,
|
|
"epoch": 7.858326429163214,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010759299137429073,
|
|
"loss": 4.6635,
|
|
"mean_token_accuracy": 0.23397628366947174,
|
|
"num_tokens": 17286805.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 5.494572114944458,
|
|
"epoch": 7.862468931234465,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00010737965042016116,
|
|
"loss": 4.7347,
|
|
"mean_token_accuracy": 0.2213120937347412,
|
|
"num_tokens": 17297577.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 5.447954797744751,
|
|
"epoch": 7.866611433305716,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010716664757459404,
|
|
"loss": 4.6412,
|
|
"mean_token_accuracy": 0.23594612777233123,
|
|
"num_tokens": 17306620.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 5.453318309783936,
|
|
"epoch": 7.8707539353769675,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010695398326723938,
|
|
"loss": 4.6569,
|
|
"mean_token_accuracy": 0.23203144967556,
|
|
"num_tokens": 17314796.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 7.8707539353769675,
|
|
"eval_entropy": 5.2538599510436885,
|
|
"eval_loss": 5.728179454803467,
|
|
"eval_mean_token_accuracy": 0.16892587557149086,
|
|
"eval_num_tokens": 17314796.0,
|
|
"eval_runtime": 2.1167,
|
|
"eval_samples_per_second": 1770.236,
|
|
"eval_steps_per_second": 221.575,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 5.446596288681031,
|
|
"epoch": 7.874896437448219,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00010674165792706423,
|
|
"loss": 4.6656,
|
|
"mean_token_accuracy": 0.23972989469766617,
|
|
"num_tokens": 17324048.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 5.445090913772583,
|
|
"epoch": 7.87903893951947,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010652967198235214,
|
|
"loss": 4.6259,
|
|
"mean_token_accuracy": 0.23765333294868468,
|
|
"num_tokens": 17333354.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 5.40987753868103,
|
|
"epoch": 7.883181441590721,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001063180258607018,
|
|
"loss": 4.5564,
|
|
"mean_token_accuracy": 0.25185338854789735,
|
|
"num_tokens": 17343026.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 5.370121622085572,
|
|
"epoch": 7.887323943661972,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010610671998902671,
|
|
"loss": 4.5407,
|
|
"mean_token_accuracy": 0.24592981338500977,
|
|
"num_tokens": 17351161.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 5.44391360282898,
|
|
"epoch": 7.891466445733223,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010589575479355379,
|
|
"loss": 4.596,
|
|
"mean_token_accuracy": 0.23580829054117203,
|
|
"num_tokens": 17360697.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 5.475640201568604,
|
|
"epoch": 7.895608947804474,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010568513069982289,
|
|
"loss": 4.6811,
|
|
"mean_token_accuracy": 0.23255289793014527,
|
|
"num_tokens": 17370303.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 5.426811027526855,
|
|
"epoch": 7.899751449875725,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00010547484813268594,
|
|
"loss": 4.6329,
|
|
"mean_token_accuracy": 0.230610391497612,
|
|
"num_tokens": 17379557.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 5.454054355621338,
|
|
"epoch": 7.903893951946976,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00010526490751630568,
|
|
"loss": 4.6866,
|
|
"mean_token_accuracy": 0.23392173051834106,
|
|
"num_tokens": 17388206.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 5.50150842666626,
|
|
"epoch": 7.908036454018227,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00010505530927415544,
|
|
"loss": 4.7331,
|
|
"mean_token_accuracy": 0.22603118121623994,
|
|
"num_tokens": 17398055.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 5.3604248523712155,
|
|
"epoch": 7.912178956089478,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010484605382901772,
|
|
"loss": 4.5262,
|
|
"mean_token_accuracy": 0.23282116204500197,
|
|
"num_tokens": 17407214.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 5.467148733139038,
|
|
"epoch": 7.916321458160729,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00010463714160298354,
|
|
"loss": 4.6681,
|
|
"mean_token_accuracy": 0.23354340046644212,
|
|
"num_tokens": 17416447.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 5.460616159439087,
|
|
"epoch": 7.92046396023198,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00010442857301745184,
|
|
"loss": 4.6498,
|
|
"mean_token_accuracy": 0.22852215468883513,
|
|
"num_tokens": 17424921.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 5.393813943862915,
|
|
"epoch": 7.924606462303231,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001042203484931281,
|
|
"loss": 4.5031,
|
|
"mean_token_accuracy": 0.24518180191516875,
|
|
"num_tokens": 17434425.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 5.392105293273926,
|
|
"epoch": 7.928748964374482,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010401246845002407,
|
|
"loss": 4.6109,
|
|
"mean_token_accuracy": 0.237376469373703,
|
|
"num_tokens": 17443879.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 5.390809774398804,
|
|
"epoch": 7.932891466445733,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010380493330745641,
|
|
"loss": 4.6049,
|
|
"mean_token_accuracy": 0.2331545904278755,
|
|
"num_tokens": 17452276.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 5.425653791427612,
|
|
"epoch": 7.937033968516984,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00010359774348404626,
|
|
"loss": 4.637,
|
|
"mean_token_accuracy": 0.23898312747478484,
|
|
"num_tokens": 17460706.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 5.468666744232178,
|
|
"epoch": 7.9411764705882355,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00010339089939771814,
|
|
"loss": 4.6806,
|
|
"mean_token_accuracy": 0.24359708428382873,
|
|
"num_tokens": 17469754.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 5.474518489837647,
|
|
"epoch": 7.945318972659487,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010318440146569904,
|
|
"loss": 4.7722,
|
|
"mean_token_accuracy": 0.22114842087030412,
|
|
"num_tokens": 17480057.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 5.411310148239136,
|
|
"epoch": 7.949461474730738,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00010297825010451806,
|
|
"loss": 4.6109,
|
|
"mean_token_accuracy": 0.23354050815105437,
|
|
"num_tokens": 17490427.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 5.451214599609375,
|
|
"epoch": 7.953603976801988,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001027724457300048,
|
|
"loss": 4.6233,
|
|
"mean_token_accuracy": 0.22287411987781525,
|
|
"num_tokens": 17499105.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 5.438440322875977,
|
|
"epoch": 7.957746478873239,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00010256698875728938,
|
|
"loss": 4.6968,
|
|
"mean_token_accuracy": 0.22868555337190627,
|
|
"num_tokens": 17508499.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 5.3884806632995605,
|
|
"epoch": 7.96188898094449,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010236187960080084,
|
|
"loss": 4.5548,
|
|
"mean_token_accuracy": 0.23828063756227494,
|
|
"num_tokens": 17516885.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 5.445450782775879,
|
|
"epoch": 7.966031483015741,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00010215711867426678,
|
|
"loss": 4.5528,
|
|
"mean_token_accuracy": 0.23936398774385453,
|
|
"num_tokens": 17524958.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 5.438009023666382,
|
|
"epoch": 7.970173985086992,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010195270639071243,
|
|
"loss": 4.6187,
|
|
"mean_token_accuracy": 0.2327953100204468,
|
|
"num_tokens": 17534185.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 5.4450901508331295,
|
|
"epoch": 7.9743164871582435,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001017486431624596,
|
|
"loss": 4.6378,
|
|
"mean_token_accuracy": 0.23655385822057723,
|
|
"num_tokens": 17542696.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 5.438672780990601,
|
|
"epoch": 7.978458989229495,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010154492940112624,
|
|
"loss": 4.5601,
|
|
"mean_token_accuracy": 0.24715670198202133,
|
|
"num_tokens": 17552751.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 5.5221234321594235,
|
|
"epoch": 7.982601491300746,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00010134156551762519,
|
|
"loss": 4.6662,
|
|
"mean_token_accuracy": 0.22336154133081437,
|
|
"num_tokens": 17561973.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 5.502470207214356,
|
|
"epoch": 7.986743993371997,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010113855192216359,
|
|
"loss": 4.7031,
|
|
"mean_token_accuracy": 0.226225808262825,
|
|
"num_tokens": 17571238.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 5.454457712173462,
|
|
"epoch": 7.990886495443248,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010093588902424217,
|
|
"loss": 4.5942,
|
|
"mean_token_accuracy": 0.24229961186647414,
|
|
"num_tokens": 17580189.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 5.4261950016021725,
|
|
"epoch": 7.995028997514499,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00010073357723265398,
|
|
"loss": 4.6749,
|
|
"mean_token_accuracy": 0.23186081796884536,
|
|
"num_tokens": 17590051.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 5.484763860702515,
|
|
"epoch": 7.99917149958575,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00010053161695548418,
|
|
"loss": 4.665,
|
|
"mean_token_accuracy": 0.2366202712059021,
|
|
"num_tokens": 17599036.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 5.442663097381592,
|
|
"epoch": 8.003314001657001,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010033000860010866,
|
|
"loss": 4.5562,
|
|
"mean_token_accuracy": 0.24225867986679078,
|
|
"num_tokens": 17608433.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 5.451050043106079,
|
|
"epoch": 8.007456503728251,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001001287525731934,
|
|
"loss": 4.5841,
|
|
"mean_token_accuracy": 0.24033490121364592,
|
|
"num_tokens": 17617696.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 5.457457542419434,
|
|
"epoch": 8.011599005799503,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.992784928069397e-05,
|
|
"loss": 4.632,
|
|
"mean_token_accuracy": 0.23403769582509995,
|
|
"num_tokens": 17626619.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 5.447802877426147,
|
|
"epoch": 8.015741507870754,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.972729912785411e-05,
|
|
"loss": 4.5163,
|
|
"mean_token_accuracy": 0.2534124761819839,
|
|
"num_tokens": 17635306.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 5.414555072784424,
|
|
"epoch": 8.019884009942006,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 9.95271025192055e-05,
|
|
"loss": 4.4862,
|
|
"mean_token_accuracy": 0.25084113627672194,
|
|
"num_tokens": 17643526.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 5.467756175994873,
|
|
"epoch": 8.024026512013256,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.93272598585665e-05,
|
|
"loss": 4.6567,
|
|
"mean_token_accuracy": 0.22692112028598785,
|
|
"num_tokens": 17652474.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 5.414951610565185,
|
|
"epoch": 8.028169014084508,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.912777154904171e-05,
|
|
"loss": 4.5832,
|
|
"mean_token_accuracy": 0.24183519929647446,
|
|
"num_tokens": 17662025.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 5.33813705444336,
|
|
"epoch": 8.032311516155758,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.89286379930207e-05,
|
|
"loss": 4.4989,
|
|
"mean_token_accuracy": 0.24851167649030687,
|
|
"num_tokens": 17670908.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 5.383984708786011,
|
|
"epoch": 8.036454018227008,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.872985959217779e-05,
|
|
"loss": 4.5351,
|
|
"mean_token_accuracy": 0.2490708351135254,
|
|
"num_tokens": 17681242.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 5.409505558013916,
|
|
"epoch": 8.04059652029826,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.853143674747053e-05,
|
|
"loss": 4.5017,
|
|
"mean_token_accuracy": 0.24562000185251237,
|
|
"num_tokens": 17689816.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 5.449247217178344,
|
|
"epoch": 8.04473902236951,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.833336985913965e-05,
|
|
"loss": 4.5814,
|
|
"mean_token_accuracy": 0.249982613325119,
|
|
"num_tokens": 17699251.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 5.4540496349334715,
|
|
"epoch": 8.048881524440763,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.813565932670767e-05,
|
|
"loss": 4.5937,
|
|
"mean_token_accuracy": 0.24556749165058137,
|
|
"num_tokens": 17708906.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 5.4913458824157715,
|
|
"epoch": 8.053024026512013,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.793830554897826e-05,
|
|
"loss": 4.6176,
|
|
"mean_token_accuracy": 0.23052370101213454,
|
|
"num_tokens": 17718071.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 5.400876188278199,
|
|
"epoch": 8.057166528583265,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.774130892403566e-05,
|
|
"loss": 4.5021,
|
|
"mean_token_accuracy": 0.24932561218738555,
|
|
"num_tokens": 17727690.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 5.384219408035278,
|
|
"epoch": 8.061309030654515,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.754466984924348e-05,
|
|
"loss": 4.4994,
|
|
"mean_token_accuracy": 0.24482243955135347,
|
|
"num_tokens": 17735934.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 5.391112041473389,
|
|
"epoch": 8.065451532725767,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.734838872124432e-05,
|
|
"loss": 4.5133,
|
|
"mean_token_accuracy": 0.24244650304317475,
|
|
"num_tokens": 17744323.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 5.348169994354248,
|
|
"epoch": 8.069594034797017,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.715246593595859e-05,
|
|
"loss": 4.492,
|
|
"mean_token_accuracy": 0.25319374799728395,
|
|
"num_tokens": 17753335.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 5.390994739532471,
|
|
"epoch": 8.07373653686827,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.695690188858408e-05,
|
|
"loss": 4.5569,
|
|
"mean_token_accuracy": 0.24261512905359267,
|
|
"num_tokens": 17761952.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 5.47481050491333,
|
|
"epoch": 8.07787903893952,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 9.676169697359469e-05,
|
|
"loss": 4.6283,
|
|
"mean_token_accuracy": 0.24092841893434525,
|
|
"num_tokens": 17771341.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 5.360014295578003,
|
|
"epoch": 8.08202154101077,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.656685158474018e-05,
|
|
"loss": 4.4963,
|
|
"mean_token_accuracy": 0.24764938205480574,
|
|
"num_tokens": 17780289.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 5.457157468795776,
|
|
"epoch": 8.086164043082022,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.63723661150449e-05,
|
|
"loss": 4.6013,
|
|
"mean_token_accuracy": 0.2351839154958725,
|
|
"num_tokens": 17789317.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 5.48147406578064,
|
|
"epoch": 8.090306545153272,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.617824095680742e-05,
|
|
"loss": 4.6478,
|
|
"mean_token_accuracy": 0.23927124291658403,
|
|
"num_tokens": 17797875.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 5.423786544799805,
|
|
"epoch": 8.094449047224524,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.598447650159931e-05,
|
|
"loss": 4.5897,
|
|
"mean_token_accuracy": 0.24284026324748992,
|
|
"num_tokens": 17806603.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 5.525707244873047,
|
|
"epoch": 8.098591549295774,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.579107314026459e-05,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.22701942175626755,
|
|
"num_tokens": 17815747.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 5.443429660797119,
|
|
"epoch": 8.102734051367026,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.559803126291911e-05,
|
|
"loss": 4.536,
|
|
"mean_token_accuracy": 0.2422915741801262,
|
|
"num_tokens": 17824568.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 5.475485944747925,
|
|
"epoch": 8.106876553438276,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.540535125894931e-05,
|
|
"loss": 4.6175,
|
|
"mean_token_accuracy": 0.242696350812912,
|
|
"num_tokens": 17834244.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 5.420765733718872,
|
|
"epoch": 8.111019055509528,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.521303351701193e-05,
|
|
"loss": 4.5205,
|
|
"mean_token_accuracy": 0.24311172515153884,
|
|
"num_tokens": 17843285.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 5.477534246444702,
|
|
"epoch": 8.115161557580779,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.502107842503282e-05,
|
|
"loss": 4.6189,
|
|
"mean_token_accuracy": 0.24500531256198882,
|
|
"num_tokens": 17852379.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 5.4197441101074215,
|
|
"epoch": 8.11930405965203,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.48294863702063e-05,
|
|
"loss": 4.5387,
|
|
"mean_token_accuracy": 0.2542175889015198,
|
|
"num_tokens": 17861948.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 5.4329001903533936,
|
|
"epoch": 8.12344656172328,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.463825773899465e-05,
|
|
"loss": 4.5079,
|
|
"mean_token_accuracy": 0.24993507117033004,
|
|
"num_tokens": 17870445.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 5.480579137802124,
|
|
"epoch": 8.127589063794533,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.444739291712673e-05,
|
|
"loss": 4.5912,
|
|
"mean_token_accuracy": 0.2537703275680542,
|
|
"num_tokens": 17879380.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 5.440218257904053,
|
|
"epoch": 8.131731565865783,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.425689228959791e-05,
|
|
"loss": 4.5896,
|
|
"mean_token_accuracy": 0.24253950417041778,
|
|
"num_tokens": 17888631.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 5.416433048248291,
|
|
"epoch": 8.135874067937033,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.406675624066868e-05,
|
|
"loss": 4.5446,
|
|
"mean_token_accuracy": 0.24484400749206542,
|
|
"num_tokens": 17898541.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 5.464384603500366,
|
|
"epoch": 8.140016570008285,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 9.38769851538642e-05,
|
|
"loss": 4.5591,
|
|
"mean_token_accuracy": 0.24426549822092056,
|
|
"num_tokens": 17906664.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 5.4251776218414305,
|
|
"epoch": 8.144159072079535,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.368757941197356e-05,
|
|
"loss": 4.5653,
|
|
"mean_token_accuracy": 0.24238588362932206,
|
|
"num_tokens": 17916201.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 5.464312505722046,
|
|
"epoch": 8.148301574150787,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 9.34985393970488e-05,
|
|
"loss": 4.5918,
|
|
"mean_token_accuracy": 0.23939154148101807,
|
|
"num_tokens": 17925700.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 5.394193410873413,
|
|
"epoch": 8.152444076222038,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.330986549040432e-05,
|
|
"loss": 4.5416,
|
|
"mean_token_accuracy": 0.2434779167175293,
|
|
"num_tokens": 17934293.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 5.393955945968628,
|
|
"epoch": 8.15658657829329,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.312155807261602e-05,
|
|
"loss": 4.5662,
|
|
"mean_token_accuracy": 0.23979576379060746,
|
|
"num_tokens": 17943541.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 5.478118991851806,
|
|
"epoch": 8.16072908036454,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.293361752352045e-05,
|
|
"loss": 4.5739,
|
|
"mean_token_accuracy": 0.24406632035970688,
|
|
"num_tokens": 17952616.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 5.456351280212402,
|
|
"epoch": 8.164871582435792,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.274604422221433e-05,
|
|
"loss": 4.5145,
|
|
"mean_token_accuracy": 0.24423680901527406,
|
|
"num_tokens": 17961414.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 5.46937313079834,
|
|
"epoch": 8.169014084507042,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.255883854705344e-05,
|
|
"loss": 4.5957,
|
|
"mean_token_accuracy": 0.24432494938373567,
|
|
"num_tokens": 17972077.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 5.453628778457642,
|
|
"epoch": 8.173156586578294,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.237200087565217e-05,
|
|
"loss": 4.5573,
|
|
"mean_token_accuracy": 0.24323742389678954,
|
|
"num_tokens": 17980303.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 5.506000709533692,
|
|
"epoch": 8.177299088649544,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.218553158488238e-05,
|
|
"loss": 4.6179,
|
|
"mean_token_accuracy": 0.23840177655220032,
|
|
"num_tokens": 17990410.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 5.400051164627075,
|
|
"epoch": 8.181441590720794,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.199943105087319e-05,
|
|
"loss": 4.51,
|
|
"mean_token_accuracy": 0.24932715743780137,
|
|
"num_tokens": 17998931.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 5.4265378475189205,
|
|
"epoch": 8.185584092792046,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.181369964900957e-05,
|
|
"loss": 4.5716,
|
|
"mean_token_accuracy": 0.24133008420467378,
|
|
"num_tokens": 18007853.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 5.414221096038818,
|
|
"epoch": 8.189726594863297,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.162833775393209e-05,
|
|
"loss": 4.5193,
|
|
"mean_token_accuracy": 0.24494489282369614,
|
|
"num_tokens": 18016517.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 5.48521203994751,
|
|
"epoch": 8.193869096934549,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.1443345739536e-05,
|
|
"loss": 4.6226,
|
|
"mean_token_accuracy": 0.24047563076019288,
|
|
"num_tokens": 18025057.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 5.434843254089356,
|
|
"epoch": 8.198011599005799,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.12587239789703e-05,
|
|
"loss": 4.6103,
|
|
"mean_token_accuracy": 0.2359030395746231,
|
|
"num_tokens": 18034371.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 5.394551515579224,
|
|
"epoch": 8.202154101077051,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.10744728446374e-05,
|
|
"loss": 4.5704,
|
|
"mean_token_accuracy": 0.24470848143100737,
|
|
"num_tokens": 18043354.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 5.3827441215515135,
|
|
"epoch": 8.206296603148301,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.089059270819187e-05,
|
|
"loss": 4.5327,
|
|
"mean_token_accuracy": 0.2453048512339592,
|
|
"num_tokens": 18051879.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 5.496877813339234,
|
|
"epoch": 8.210439105219553,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.070708394054005e-05,
|
|
"loss": 4.6115,
|
|
"mean_token_accuracy": 0.23936769664287566,
|
|
"num_tokens": 18062011.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 5.4792884349822994,
|
|
"epoch": 8.214581607290803,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 9.052394691183921e-05,
|
|
"loss": 4.651,
|
|
"mean_token_accuracy": 0.23042578846216202,
|
|
"num_tokens": 18071692.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 5.390723180770874,
|
|
"epoch": 8.218724109362055,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.03411819914967e-05,
|
|
"loss": 4.5183,
|
|
"mean_token_accuracy": 0.2438945561647415,
|
|
"num_tokens": 18080646.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 5.4198588848114015,
|
|
"epoch": 8.222866611433306,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.015878954816941e-05,
|
|
"loss": 4.5187,
|
|
"mean_token_accuracy": 0.2470941349864006,
|
|
"num_tokens": 18088788.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 5.4500532150268555,
|
|
"epoch": 8.227009113504558,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.997676994976281e-05,
|
|
"loss": 4.5745,
|
|
"mean_token_accuracy": 0.23440682291984558,
|
|
"num_tokens": 18098319.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 5.434911489486694,
|
|
"epoch": 8.231151615575808,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.979512356343023e-05,
|
|
"loss": 4.6134,
|
|
"mean_token_accuracy": 0.24138568192720414,
|
|
"num_tokens": 18107740.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 5.422667980194092,
|
|
"epoch": 8.235294117647058,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.961385075557243e-05,
|
|
"loss": 4.6133,
|
|
"mean_token_accuracy": 0.23571241348981858,
|
|
"num_tokens": 18117762.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 5.392668724060059,
|
|
"epoch": 8.23943661971831,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.943295189183632e-05,
|
|
"loss": 4.4763,
|
|
"mean_token_accuracy": 0.24688512831926346,
|
|
"num_tokens": 18126212.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 5.471997499465942,
|
|
"epoch": 8.24357912178956,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 8.925242733711484e-05,
|
|
"loss": 4.6372,
|
|
"mean_token_accuracy": 0.23967914283275604,
|
|
"num_tokens": 18135622.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 5.469031953811646,
|
|
"epoch": 8.247721623860812,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.90722774555456e-05,
|
|
"loss": 4.595,
|
|
"mean_token_accuracy": 0.23535001128911973,
|
|
"num_tokens": 18144768.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 5.433808612823486,
|
|
"epoch": 8.251864125932062,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.889250261051063e-05,
|
|
"loss": 4.6139,
|
|
"mean_token_accuracy": 0.24655994027853012,
|
|
"num_tokens": 18155107.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 5.419907283782959,
|
|
"epoch": 8.256006628003314,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.871310316463554e-05,
|
|
"loss": 4.5753,
|
|
"mean_token_accuracy": 0.24290321618318558,
|
|
"num_tokens": 18164064.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 5.413675022125244,
|
|
"epoch": 8.260149130074565,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.853407947978845e-05,
|
|
"loss": 4.5994,
|
|
"mean_token_accuracy": 0.2364619955420494,
|
|
"num_tokens": 18172714.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 5.404689741134644,
|
|
"epoch": 8.264291632145817,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 8.835543191707982e-05,
|
|
"loss": 4.5808,
|
|
"mean_token_accuracy": 0.23887405544519424,
|
|
"num_tokens": 18182158.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 5.438195705413818,
|
|
"epoch": 8.268434134217067,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 8.81771608368613e-05,
|
|
"loss": 4.5784,
|
|
"mean_token_accuracy": 0.23557898104190828,
|
|
"num_tokens": 18190605.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 5.457483291625977,
|
|
"epoch": 8.272576636288319,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.799926659872507e-05,
|
|
"loss": 4.7285,
|
|
"mean_token_accuracy": 0.2230786442756653,
|
|
"num_tokens": 18199992.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 5.506515884399414,
|
|
"epoch": 8.276719138359569,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.782174956150335e-05,
|
|
"loss": 4.6091,
|
|
"mean_token_accuracy": 0.24097050875425338,
|
|
"num_tokens": 18208641.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 5.471822834014892,
|
|
"epoch": 8.280861640430821,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.764461008326723e-05,
|
|
"loss": 4.5953,
|
|
"mean_token_accuracy": 0.23975688964128494,
|
|
"num_tokens": 18218355.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 5.439051103591919,
|
|
"epoch": 8.285004142502071,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.746784852132661e-05,
|
|
"loss": 4.5984,
|
|
"mean_token_accuracy": 0.23919067680835723,
|
|
"num_tokens": 18227217.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 8.285004142502071,
|
|
"eval_entropy": 5.284503511782648,
|
|
"eval_loss": 5.734222888946533,
|
|
"eval_mean_token_accuracy": 0.1684083980696796,
|
|
"eval_num_tokens": 18227217.0,
|
|
"eval_runtime": 2.1112,
|
|
"eval_samples_per_second": 1774.787,
|
|
"eval_steps_per_second": 222.144,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 5.420424270629883,
|
|
"epoch": 8.289146644573322,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.729146523222877e-05,
|
|
"loss": 4.5418,
|
|
"mean_token_accuracy": 0.24030141085386275,
|
|
"num_tokens": 18236410.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 5.452866172790527,
|
|
"epoch": 8.293289146644574,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 8.711546057175804e-05,
|
|
"loss": 4.6128,
|
|
"mean_token_accuracy": 0.24361779242753984,
|
|
"num_tokens": 18244626.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 5.482981443405151,
|
|
"epoch": 8.297431648715824,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.69398348949351e-05,
|
|
"loss": 4.6251,
|
|
"mean_token_accuracy": 0.23858213573694229,
|
|
"num_tokens": 18253938.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 5.427640581130982,
|
|
"epoch": 8.301574150787076,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.676458855601612e-05,
|
|
"loss": 4.6068,
|
|
"mean_token_accuracy": 0.23559614568948745,
|
|
"num_tokens": 18262674.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 5.474660348892212,
|
|
"epoch": 8.305716652858326,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.658972190849215e-05,
|
|
"loss": 4.6564,
|
|
"mean_token_accuracy": 0.23526842892169952,
|
|
"num_tokens": 18272558.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 5.517200708389282,
|
|
"epoch": 8.309859154929578,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.641523530508827e-05,
|
|
"loss": 4.6738,
|
|
"mean_token_accuracy": 0.2396675616502762,
|
|
"num_tokens": 18282522.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 5.48144211769104,
|
|
"epoch": 8.314001657000828,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 8.624112909776314e-05,
|
|
"loss": 4.5797,
|
|
"mean_token_accuracy": 0.23627237379550933,
|
|
"num_tokens": 18290933.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 5.440665006637573,
|
|
"epoch": 8.31814415907208,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.606740363770788e-05,
|
|
"loss": 4.5613,
|
|
"mean_token_accuracy": 0.24098197966814042,
|
|
"num_tokens": 18300705.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 5.455484867095947,
|
|
"epoch": 8.32228666114333,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.589405927534571e-05,
|
|
"loss": 4.5549,
|
|
"mean_token_accuracy": 0.2460627317428589,
|
|
"num_tokens": 18309889.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 5.433671665191651,
|
|
"epoch": 8.326429163214582,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.57210963603313e-05,
|
|
"loss": 4.5892,
|
|
"mean_token_accuracy": 0.23434714376926422,
|
|
"num_tokens": 18319282.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 5.4790229320526125,
|
|
"epoch": 8.330571665285833,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.554851524154961e-05,
|
|
"loss": 4.64,
|
|
"mean_token_accuracy": 0.2349641889333725,
|
|
"num_tokens": 18328713.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 5.417065954208374,
|
|
"epoch": 8.334714167357085,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 8.537631626711573e-05,
|
|
"loss": 4.5068,
|
|
"mean_token_accuracy": 0.2498706042766571,
|
|
"num_tokens": 18337262.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 5.455858135223389,
|
|
"epoch": 8.338856669428335,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 8.520449978437372e-05,
|
|
"loss": 4.6325,
|
|
"mean_token_accuracy": 0.2378760650753975,
|
|
"num_tokens": 18346699.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 5.416524219512939,
|
|
"epoch": 8.342999171499585,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.50330661398962e-05,
|
|
"loss": 4.5344,
|
|
"mean_token_accuracy": 0.23782323449850082,
|
|
"num_tokens": 18355466.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 5.4285016536712645,
|
|
"epoch": 8.347141673570837,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.486201567948363e-05,
|
|
"loss": 4.5422,
|
|
"mean_token_accuracy": 0.2476327031850815,
|
|
"num_tokens": 18363988.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 5.4681884288787845,
|
|
"epoch": 8.351284175642087,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 8.469134874816342e-05,
|
|
"loss": 4.6605,
|
|
"mean_token_accuracy": 0.23449433594942093,
|
|
"num_tokens": 18373784.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 5.452970361709594,
|
|
"epoch": 8.35542667771334,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.45210656901895e-05,
|
|
"loss": 4.5277,
|
|
"mean_token_accuracy": 0.24586998224258422,
|
|
"num_tokens": 18382929.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 5.492053127288818,
|
|
"epoch": 8.35956917978459,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.435116684904136e-05,
|
|
"loss": 4.6127,
|
|
"mean_token_accuracy": 0.23736173659563065,
|
|
"num_tokens": 18392399.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 5.489425134658814,
|
|
"epoch": 8.363711681855841,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.418165256742346e-05,
|
|
"loss": 4.7164,
|
|
"mean_token_accuracy": 0.2306162565946579,
|
|
"num_tokens": 18402255.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 5.442831373214721,
|
|
"epoch": 8.367854183927092,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 8.401252318726473e-05,
|
|
"loss": 4.5869,
|
|
"mean_token_accuracy": 0.24043678343296052,
|
|
"num_tokens": 18411677.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 5.386390972137451,
|
|
"epoch": 8.371996685998344,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.384377904971753e-05,
|
|
"loss": 4.5516,
|
|
"mean_token_accuracy": 0.23836876899003984,
|
|
"num_tokens": 18420332.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 5.353835105895996,
|
|
"epoch": 8.376139188069594,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 8.367542049515733e-05,
|
|
"loss": 4.477,
|
|
"mean_token_accuracy": 0.24212346374988555,
|
|
"num_tokens": 18429542.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 5.4741583347320555,
|
|
"epoch": 8.380281690140846,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.350744786318168e-05,
|
|
"loss": 4.5896,
|
|
"mean_token_accuracy": 0.24692442864179612,
|
|
"num_tokens": 18439064.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 5.4267271041870115,
|
|
"epoch": 8.384424192212096,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.33398614926097e-05,
|
|
"loss": 4.6009,
|
|
"mean_token_accuracy": 0.23683178722858428,
|
|
"num_tokens": 18449001.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 5.397810602188111,
|
|
"epoch": 8.388566694283346,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 8.317266172148145e-05,
|
|
"loss": 4.5594,
|
|
"mean_token_accuracy": 0.2442532956600189,
|
|
"num_tokens": 18458105.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 5.438517045974732,
|
|
"epoch": 8.392709196354598,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.300584888705709e-05,
|
|
"loss": 4.5938,
|
|
"mean_token_accuracy": 0.23512856662273407,
|
|
"num_tokens": 18466975.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 5.365755653381347,
|
|
"epoch": 8.396851698425849,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.283942332581644e-05,
|
|
"loss": 4.4835,
|
|
"mean_token_accuracy": 0.257667638361454,
|
|
"num_tokens": 18475543.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 5.337273359298706,
|
|
"epoch": 8.4009942004971,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.267338537345792e-05,
|
|
"loss": 4.4566,
|
|
"mean_token_accuracy": 0.2502560868859291,
|
|
"num_tokens": 18484399.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 5.36476731300354,
|
|
"epoch": 8.40513670256835,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 8.250773536489833e-05,
|
|
"loss": 4.591,
|
|
"mean_token_accuracy": 0.23123996555805207,
|
|
"num_tokens": 18493168.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 5.337420701980591,
|
|
"epoch": 8.409279204639603,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.234247363427174e-05,
|
|
"loss": 4.5463,
|
|
"mean_token_accuracy": 0.24021608233451844,
|
|
"num_tokens": 18502624.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 5.4171491146087645,
|
|
"epoch": 8.413421706710853,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.217760051492926e-05,
|
|
"loss": 4.5405,
|
|
"mean_token_accuracy": 0.24212121814489365,
|
|
"num_tokens": 18510990.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 5.383139181137085,
|
|
"epoch": 8.417564208782105,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 8.201311633943788e-05,
|
|
"loss": 4.5309,
|
|
"mean_token_accuracy": 0.24388076812028886,
|
|
"num_tokens": 18519490.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 5.435222578048706,
|
|
"epoch": 8.421706710853355,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.184902143958013e-05,
|
|
"loss": 4.5874,
|
|
"mean_token_accuracy": 0.23824901580810548,
|
|
"num_tokens": 18528827.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 5.377360200881958,
|
|
"epoch": 8.425849212924607,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 8.168531614635347e-05,
|
|
"loss": 4.5611,
|
|
"mean_token_accuracy": 0.24246266782283782,
|
|
"num_tokens": 18537592.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 5.4337913513183596,
|
|
"epoch": 8.429991714995857,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.152200078996924e-05,
|
|
"loss": 4.6468,
|
|
"mean_token_accuracy": 0.24592253267765046,
|
|
"num_tokens": 18547253.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 5.394612836837768,
|
|
"epoch": 8.434134217067108,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.135907569985248e-05,
|
|
"loss": 4.6071,
|
|
"mean_token_accuracy": 0.2302077069878578,
|
|
"num_tokens": 18556658.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 5.415404272079468,
|
|
"epoch": 8.43827671913836,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.119654120464077e-05,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.23810053169727324,
|
|
"num_tokens": 18565594.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 5.4217134475708,
|
|
"epoch": 8.44241922120961,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.103439763218411e-05,
|
|
"loss": 4.5982,
|
|
"mean_token_accuracy": 0.24319261461496353,
|
|
"num_tokens": 18574033.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 5.424828720092774,
|
|
"epoch": 8.446561723280862,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.087264530954362e-05,
|
|
"loss": 4.5759,
|
|
"mean_token_accuracy": 0.23843355029821395,
|
|
"num_tokens": 18583064.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 5.554442453384399,
|
|
"epoch": 8.450704225352112,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.071128456299159e-05,
|
|
"loss": 4.6625,
|
|
"mean_token_accuracy": 0.23185975551605226,
|
|
"num_tokens": 18591910.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 5.487932252883911,
|
|
"epoch": 8.454846727423364,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 8.055031571801013e-05,
|
|
"loss": 4.5933,
|
|
"mean_token_accuracy": 0.24012340903282164,
|
|
"num_tokens": 18600880.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 5.5398636817932125,
|
|
"epoch": 8.458989229494614,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 8.038973909929112e-05,
|
|
"loss": 4.7733,
|
|
"mean_token_accuracy": 0.22599243372678757,
|
|
"num_tokens": 18610407.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 5.42407636642456,
|
|
"epoch": 8.463131731565866,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.022955503073502e-05,
|
|
"loss": 4.6233,
|
|
"mean_token_accuracy": 0.2369710922241211,
|
|
"num_tokens": 18620166.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 5.381896448135376,
|
|
"epoch": 8.467274233637117,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.006976383545071e-05,
|
|
"loss": 4.5419,
|
|
"mean_token_accuracy": 0.2381254330277443,
|
|
"num_tokens": 18628851.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 5.432081985473633,
|
|
"epoch": 8.471416735708369,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.991036583575448e-05,
|
|
"loss": 4.6,
|
|
"mean_token_accuracy": 0.23973309397697448,
|
|
"num_tokens": 18637921.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 5.4369255065917965,
|
|
"epoch": 8.475559237779619,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.975136135316941e-05,
|
|
"loss": 4.5298,
|
|
"mean_token_accuracy": 0.24518140256404877,
|
|
"num_tokens": 18647291.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 5.378586959838867,
|
|
"epoch": 8.47970173985087,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.959275070842502e-05,
|
|
"loss": 4.5376,
|
|
"mean_token_accuracy": 0.23944890946149827,
|
|
"num_tokens": 18656252.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 5.360685968399048,
|
|
"epoch": 8.483844241922121,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.943453422145626e-05,
|
|
"loss": 4.5561,
|
|
"mean_token_accuracy": 0.24402323216199875,
|
|
"num_tokens": 18666199.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 5.478400325775146,
|
|
"epoch": 8.487986743993371,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.927671221140317e-05,
|
|
"loss": 4.6409,
|
|
"mean_token_accuracy": 0.23700555860996247,
|
|
"num_tokens": 18675573.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 5.455059385299682,
|
|
"epoch": 8.492129246064623,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.911928499660986e-05,
|
|
"loss": 4.588,
|
|
"mean_token_accuracy": 0.23940761983394623,
|
|
"num_tokens": 18684627.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 5.447060251235962,
|
|
"epoch": 8.496271748135873,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 7.896225289462426e-05,
|
|
"loss": 4.596,
|
|
"mean_token_accuracy": 0.23441989719867706,
|
|
"num_tokens": 18694287.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 5.416751432418823,
|
|
"epoch": 8.500414250207125,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.880561622219735e-05,
|
|
"loss": 4.5893,
|
|
"mean_token_accuracy": 0.23685378581285477,
|
|
"num_tokens": 18704014.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 5.43733286857605,
|
|
"epoch": 8.504556752278376,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.864937529528232e-05,
|
|
"loss": 4.582,
|
|
"mean_token_accuracy": 0.24251590967178344,
|
|
"num_tokens": 18713030.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 5.455956649780274,
|
|
"epoch": 8.508699254349628,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.849353042903428e-05,
|
|
"loss": 4.5994,
|
|
"mean_token_accuracy": 0.24521316289901735,
|
|
"num_tokens": 18721911.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 5.415904426574707,
|
|
"epoch": 8.512841756420878,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.833808193780934e-05,
|
|
"loss": 4.5403,
|
|
"mean_token_accuracy": 0.24652504026889802,
|
|
"num_tokens": 18730688.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 5.4866614818573,
|
|
"epoch": 8.51698425849213,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.818303013516401e-05,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.24073482155799866,
|
|
"num_tokens": 18740214.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 5.402558040618897,
|
|
"epoch": 8.52112676056338,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 7.802837533385482e-05,
|
|
"loss": 4.5972,
|
|
"mean_token_accuracy": 0.23633088618516923,
|
|
"num_tokens": 18748837.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 5.386799144744873,
|
|
"epoch": 8.525269262634632,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.787411784583733e-05,
|
|
"loss": 4.4821,
|
|
"mean_token_accuracy": 0.2533039405941963,
|
|
"num_tokens": 18757482.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 5.328194713592529,
|
|
"epoch": 8.529411764705882,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.772025798226582e-05,
|
|
"loss": 4.5458,
|
|
"mean_token_accuracy": 0.2409552589058876,
|
|
"num_tokens": 18766326.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 5.507105588912964,
|
|
"epoch": 8.533554266777134,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.756679605349241e-05,
|
|
"loss": 4.6618,
|
|
"mean_token_accuracy": 0.23643684685230254,
|
|
"num_tokens": 18774287.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 5.460562658309937,
|
|
"epoch": 8.537696768848384,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 7.741373236906658e-05,
|
|
"loss": 4.6349,
|
|
"mean_token_accuracy": 0.24208370894193648,
|
|
"num_tokens": 18785172.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 5.391373205184936,
|
|
"epoch": 8.541839270919635,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.726106723773452e-05,
|
|
"loss": 4.6379,
|
|
"mean_token_accuracy": 0.2384229362010956,
|
|
"num_tokens": 18794294.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 5.474083662033081,
|
|
"epoch": 8.545981772990887,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.710880096743845e-05,
|
|
"loss": 4.6388,
|
|
"mean_token_accuracy": 0.23752205520868303,
|
|
"num_tokens": 18803108.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 5.446882247924805,
|
|
"epoch": 8.550124275062137,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.695693386531612e-05,
|
|
"loss": 4.5704,
|
|
"mean_token_accuracy": 0.24286768585443497,
|
|
"num_tokens": 18811215.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 5.4512176513671875,
|
|
"epoch": 8.554266777133389,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.680546623770001e-05,
|
|
"loss": 4.673,
|
|
"mean_token_accuracy": 0.23019793778657913,
|
|
"num_tokens": 18820796.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 5.568665075302124,
|
|
"epoch": 8.55840927920464,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.665439839011701e-05,
|
|
"loss": 4.7207,
|
|
"mean_token_accuracy": 0.22016632109880446,
|
|
"num_tokens": 18830673.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 5.444123315811157,
|
|
"epoch": 8.562551781275891,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.650373062728735e-05,
|
|
"loss": 4.599,
|
|
"mean_token_accuracy": 0.23663514256477355,
|
|
"num_tokens": 18839898.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 5.427448558807373,
|
|
"epoch": 8.566694283347141,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.635346325312437e-05,
|
|
"loss": 4.5523,
|
|
"mean_token_accuracy": 0.2399025022983551,
|
|
"num_tokens": 18848380.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 5.394487619400024,
|
|
"epoch": 8.570836785418393,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.62035965707339e-05,
|
|
"loss": 4.5218,
|
|
"mean_token_accuracy": 0.240800641477108,
|
|
"num_tokens": 18857157.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 5.467054033279419,
|
|
"epoch": 8.574979287489644,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.605413088241325e-05,
|
|
"loss": 4.6081,
|
|
"mean_token_accuracy": 0.24326770305633544,
|
|
"num_tokens": 18866544.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 5.449553060531616,
|
|
"epoch": 8.579121789560896,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.59050664896512e-05,
|
|
"loss": 4.6272,
|
|
"mean_token_accuracy": 0.23857941776514052,
|
|
"num_tokens": 18876052.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 5.433203792572021,
|
|
"epoch": 8.583264291632146,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 7.575640369312684e-05,
|
|
"loss": 4.5175,
|
|
"mean_token_accuracy": 0.24317475855350495,
|
|
"num_tokens": 18885141.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 5.415943670272827,
|
|
"epoch": 8.587406793703398,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.560814279270928e-05,
|
|
"loss": 4.5079,
|
|
"mean_token_accuracy": 0.2512508869171143,
|
|
"num_tokens": 18893108.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 5.455328893661499,
|
|
"epoch": 8.591549295774648,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.5460284087457e-05,
|
|
"loss": 4.5953,
|
|
"mean_token_accuracy": 0.24100806564092636,
|
|
"num_tokens": 18901436.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 5.434124088287353,
|
|
"epoch": 8.595691797845898,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.53128278756171e-05,
|
|
"loss": 4.5575,
|
|
"mean_token_accuracy": 0.252792164683342,
|
|
"num_tokens": 18910057.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 5.319125032424926,
|
|
"epoch": 8.59983429991715,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.516577445462493e-05,
|
|
"loss": 4.4418,
|
|
"mean_token_accuracy": 0.2546115592122078,
|
|
"num_tokens": 18918270.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 5.451590347290039,
|
|
"epoch": 8.6039768019884,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.501912412110331e-05,
|
|
"loss": 4.6501,
|
|
"mean_token_accuracy": 0.2343801200389862,
|
|
"num_tokens": 18928202.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 5.391668224334717,
|
|
"epoch": 8.608119304059652,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.487287717086193e-05,
|
|
"loss": 4.5947,
|
|
"mean_token_accuracy": 0.24063229858875274,
|
|
"num_tokens": 18937837.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 5.421950483322144,
|
|
"epoch": 8.612261806130903,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.472703389889695e-05,
|
|
"loss": 4.5483,
|
|
"mean_token_accuracy": 0.24551957845687866,
|
|
"num_tokens": 18946391.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 5.4456462383270265,
|
|
"epoch": 8.616404308202155,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.458159459939007e-05,
|
|
"loss": 4.5898,
|
|
"mean_token_accuracy": 0.23844190537929535,
|
|
"num_tokens": 18955269.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 5.425407886505127,
|
|
"epoch": 8.620546810273405,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.443655956570842e-05,
|
|
"loss": 4.528,
|
|
"mean_token_accuracy": 0.250596122443676,
|
|
"num_tokens": 18964030.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 5.406321907043457,
|
|
"epoch": 8.624689312344657,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.429192909040339e-05,
|
|
"loss": 4.5413,
|
|
"mean_token_accuracy": 0.24719631671905518,
|
|
"num_tokens": 18973690.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 5.441375780105591,
|
|
"epoch": 8.628831814415907,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.414770346521043e-05,
|
|
"loss": 4.5442,
|
|
"mean_token_accuracy": 0.24629451036453248,
|
|
"num_tokens": 18981339.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 5.43851318359375,
|
|
"epoch": 8.632974316487159,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.400388298104851e-05,
|
|
"loss": 4.5752,
|
|
"mean_token_accuracy": 0.23718754202127457,
|
|
"num_tokens": 18990141.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 5.369984292984009,
|
|
"epoch": 8.63711681855841,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.386046792801911e-05,
|
|
"loss": 4.531,
|
|
"mean_token_accuracy": 0.24998438358306885,
|
|
"num_tokens": 18999217.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 5.437046813964844,
|
|
"epoch": 8.64125932062966,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.371745859540622e-05,
|
|
"loss": 4.6471,
|
|
"mean_token_accuracy": 0.24133840948343277,
|
|
"num_tokens": 19008554.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 5.392385959625244,
|
|
"epoch": 8.645401822700912,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.357485527167522e-05,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.24621531665325164,
|
|
"num_tokens": 19018159.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 5.413475561141968,
|
|
"epoch": 8.649544324772162,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 7.343265824447254e-05,
|
|
"loss": 4.5616,
|
|
"mean_token_accuracy": 0.23825883865356445,
|
|
"num_tokens": 19026985.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 5.448697090148926,
|
|
"epoch": 8.653686826843414,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 7.329086780062522e-05,
|
|
"loss": 4.6131,
|
|
"mean_token_accuracy": 0.23503292948007584,
|
|
"num_tokens": 19036745.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 5.427451753616333,
|
|
"epoch": 8.657829328914664,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.314948422613998e-05,
|
|
"loss": 4.5475,
|
|
"mean_token_accuracy": 0.24619056135416031,
|
|
"num_tokens": 19045200.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 5.4715550422668455,
|
|
"epoch": 8.661971830985916,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 7.30085078062031e-05,
|
|
"loss": 4.5535,
|
|
"mean_token_accuracy": 0.24589183032512665,
|
|
"num_tokens": 19054102.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 5.436909627914429,
|
|
"epoch": 8.666114333057166,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.286793882517922e-05,
|
|
"loss": 4.6502,
|
|
"mean_token_accuracy": 0.23714553117752074,
|
|
"num_tokens": 19062895.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 5.408593320846558,
|
|
"epoch": 8.670256835128418,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.272777756661152e-05,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.25257229804992676,
|
|
"num_tokens": 19071523.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 5.376763391494751,
|
|
"epoch": 8.674399337199668,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.258802431322047e-05,
|
|
"loss": 4.5609,
|
|
"mean_token_accuracy": 0.24235949516296387,
|
|
"num_tokens": 19080765.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 5.460604953765869,
|
|
"epoch": 8.67854183927092,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.244867934690364e-05,
|
|
"loss": 4.6237,
|
|
"mean_token_accuracy": 0.23099300563335418,
|
|
"num_tokens": 19089871.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 5.394731569290161,
|
|
"epoch": 8.68268434134217,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.230974294873513e-05,
|
|
"loss": 4.544,
|
|
"mean_token_accuracy": 0.2325541764497757,
|
|
"num_tokens": 19098928.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 5.448122358322143,
|
|
"epoch": 8.68682684341342,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.217121539896476e-05,
|
|
"loss": 4.6361,
|
|
"mean_token_accuracy": 0.23209084570407867,
|
|
"num_tokens": 19107795.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 5.54832763671875,
|
|
"epoch": 8.690969345484673,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.203309697701784e-05,
|
|
"loss": 4.6681,
|
|
"mean_token_accuracy": 0.22867024838924407,
|
|
"num_tokens": 19117060.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 5.437559509277344,
|
|
"epoch": 8.695111847555923,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.189538796149424e-05,
|
|
"loss": 4.5943,
|
|
"mean_token_accuracy": 0.23705730438232422,
|
|
"num_tokens": 19126876.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 5.535960388183594,
|
|
"epoch": 8.699254349627175,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 7.175808863016806e-05,
|
|
"loss": 4.6713,
|
|
"mean_token_accuracy": 0.22964454889297486,
|
|
"num_tokens": 19136888.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 8.699254349627175,
|
|
"eval_entropy": 5.279432446209352,
|
|
"eval_loss": 5.734774589538574,
|
|
"eval_mean_token_accuracy": 0.1685351318896198,
|
|
"eval_num_tokens": 19136888.0,
|
|
"eval_runtime": 2.1056,
|
|
"eval_samples_per_second": 1779.569,
|
|
"eval_steps_per_second": 222.743,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 5.472884654998779,
|
|
"epoch": 8.703396851698425,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.162119925998714e-05,
|
|
"loss": 4.6429,
|
|
"mean_token_accuracy": 0.22809533029794693,
|
|
"num_tokens": 19146727.0,
|
|
"step": 10505
|
|
},
|
|
{
|
|
"entropy": 5.410800790786743,
|
|
"epoch": 8.707539353769677,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.148472012707225e-05,
|
|
"loss": 4.5718,
|
|
"mean_token_accuracy": 0.24027592241764067,
|
|
"num_tokens": 19157322.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 5.377881479263306,
|
|
"epoch": 8.711681855840927,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.134865150671676e-05,
|
|
"loss": 4.5524,
|
|
"mean_token_accuracy": 0.23891821056604384,
|
|
"num_tokens": 19166302.0,
|
|
"step": 10515
|
|
},
|
|
{
|
|
"entropy": 5.459302186965942,
|
|
"epoch": 8.71582435791218,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.121299367338593e-05,
|
|
"loss": 4.5972,
|
|
"mean_token_accuracy": 0.2426257163286209,
|
|
"num_tokens": 19174883.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 5.405632448196411,
|
|
"epoch": 8.71996685998343,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.107774690071635e-05,
|
|
"loss": 4.5243,
|
|
"mean_token_accuracy": 0.2377026841044426,
|
|
"num_tokens": 19184410.0,
|
|
"step": 10525
|
|
},
|
|
{
|
|
"entropy": 5.4462676525115965,
|
|
"epoch": 8.724109362054682,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.094291146151568e-05,
|
|
"loss": 4.6157,
|
|
"mean_token_accuracy": 0.2407502979040146,
|
|
"num_tokens": 19193559.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 5.508722400665283,
|
|
"epoch": 8.728251864125932,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 7.080848762776159e-05,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.23857323974370956,
|
|
"num_tokens": 19202520.0,
|
|
"step": 10535
|
|
},
|
|
{
|
|
"entropy": 5.369424629211426,
|
|
"epoch": 8.732394366197184,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.067447567060178e-05,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.2404316708445549,
|
|
"num_tokens": 19212357.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 5.466802358627319,
|
|
"epoch": 8.736536868268434,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.054087586035295e-05,
|
|
"loss": 4.6131,
|
|
"mean_token_accuracy": 0.23553002625703812,
|
|
"num_tokens": 19222069.0,
|
|
"step": 10545
|
|
},
|
|
{
|
|
"entropy": 5.497820854187012,
|
|
"epoch": 8.740679370339684,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.040768846650046e-05,
|
|
"loss": 4.6646,
|
|
"mean_token_accuracy": 0.23103345334529876,
|
|
"num_tokens": 19232274.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 5.468303918838501,
|
|
"epoch": 8.744821872410936,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.027491375769795e-05,
|
|
"loss": 4.6493,
|
|
"mean_token_accuracy": 0.23974855989217758,
|
|
"num_tokens": 19241877.0,
|
|
"step": 10555
|
|
},
|
|
{
|
|
"entropy": 5.425574588775635,
|
|
"epoch": 8.748964374482187,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.014255200176643e-05,
|
|
"loss": 4.5451,
|
|
"mean_token_accuracy": 0.24277597218751906,
|
|
"num_tokens": 19250229.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 5.410130500793457,
|
|
"epoch": 8.753106876553439,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.001060346569408e-05,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.23725207895040512,
|
|
"num_tokens": 19259061.0,
|
|
"step": 10565
|
|
},
|
|
{
|
|
"entropy": 5.488286685943604,
|
|
"epoch": 8.757249378624689,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.987906841563552e-05,
|
|
"loss": 4.633,
|
|
"mean_token_accuracy": 0.2310020536184311,
|
|
"num_tokens": 19268262.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 5.410633516311646,
|
|
"epoch": 8.76139188069594,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.974794711691125e-05,
|
|
"loss": 4.6054,
|
|
"mean_token_accuracy": 0.24446650594472885,
|
|
"num_tokens": 19276444.0,
|
|
"step": 10575
|
|
},
|
|
{
|
|
"entropy": 5.451984596252442,
|
|
"epoch": 8.765534382767191,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.961723983400733e-05,
|
|
"loss": 4.5911,
|
|
"mean_token_accuracy": 0.23680124431848526,
|
|
"num_tokens": 19286384.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 5.3822297096252445,
|
|
"epoch": 8.769676884838443,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.948694683057456e-05,
|
|
"loss": 4.5988,
|
|
"mean_token_accuracy": 0.23448602706193925,
|
|
"num_tokens": 19295251.0,
|
|
"step": 10585
|
|
},
|
|
{
|
|
"entropy": 5.45160026550293,
|
|
"epoch": 8.773819386909693,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.935706836942826e-05,
|
|
"loss": 4.5854,
|
|
"mean_token_accuracy": 0.24204438477754592,
|
|
"num_tokens": 19304288.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 5.477918720245361,
|
|
"epoch": 8.777961888980945,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.922760471254738e-05,
|
|
"loss": 4.6,
|
|
"mean_token_accuracy": 0.2403819426894188,
|
|
"num_tokens": 19312269.0,
|
|
"step": 10595
|
|
},
|
|
{
|
|
"entropy": 5.3520337581634525,
|
|
"epoch": 8.782104391052195,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.909855612107437e-05,
|
|
"loss": 4.5122,
|
|
"mean_token_accuracy": 0.2527250975370407,
|
|
"num_tokens": 19320856.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 5.393644237518311,
|
|
"epoch": 8.786246893123447,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.896992285531424e-05,
|
|
"loss": 4.5782,
|
|
"mean_token_accuracy": 0.24646031856536865,
|
|
"num_tokens": 19329427.0,
|
|
"step": 10605
|
|
},
|
|
{
|
|
"entropy": 5.375364017486572,
|
|
"epoch": 8.790389395194698,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.884170517473444e-05,
|
|
"loss": 4.5858,
|
|
"mean_token_accuracy": 0.24824906140565872,
|
|
"num_tokens": 19337772.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 5.495497751235962,
|
|
"epoch": 8.794531897265948,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.871390333796398e-05,
|
|
"loss": 4.645,
|
|
"mean_token_accuracy": 0.24597989320755004,
|
|
"num_tokens": 19346693.0,
|
|
"step": 10615
|
|
},
|
|
{
|
|
"entropy": 5.439974641799926,
|
|
"epoch": 8.7986743993372,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.85865176027931e-05,
|
|
"loss": 4.5672,
|
|
"mean_token_accuracy": 0.2392740324139595,
|
|
"num_tokens": 19355606.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 5.416589832305908,
|
|
"epoch": 8.80281690140845,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.845954822617284e-05,
|
|
"loss": 4.6036,
|
|
"mean_token_accuracy": 0.23645109236240386,
|
|
"num_tokens": 19365552.0,
|
|
"step": 10625
|
|
},
|
|
{
|
|
"entropy": 5.445793914794922,
|
|
"epoch": 8.806959403479702,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.833299546421423e-05,
|
|
"loss": 4.652,
|
|
"mean_token_accuracy": 0.2280827596783638,
|
|
"num_tokens": 19374869.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 5.443768882751465,
|
|
"epoch": 8.811101905550952,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.820685957218808e-05,
|
|
"loss": 4.6247,
|
|
"mean_token_accuracy": 0.2365737348794937,
|
|
"num_tokens": 19384609.0,
|
|
"step": 10635
|
|
},
|
|
{
|
|
"entropy": 5.409268236160278,
|
|
"epoch": 8.815244407622204,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.80811408045242e-05,
|
|
"loss": 4.5358,
|
|
"mean_token_accuracy": 0.24484186619520187,
|
|
"num_tokens": 19392618.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 5.560499048233032,
|
|
"epoch": 8.819386909693455,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.795583941481113e-05,
|
|
"loss": 4.6439,
|
|
"mean_token_accuracy": 0.23958196640014648,
|
|
"num_tokens": 19402341.0,
|
|
"step": 10645
|
|
},
|
|
{
|
|
"entropy": 5.471864080429077,
|
|
"epoch": 8.823529411764707,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.783095565579546e-05,
|
|
"loss": 4.5964,
|
|
"mean_token_accuracy": 0.23994804918766022,
|
|
"num_tokens": 19411834.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 5.40471076965332,
|
|
"epoch": 8.827671913835957,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.77064897793814e-05,
|
|
"loss": 4.5297,
|
|
"mean_token_accuracy": 0.24119991511106492,
|
|
"num_tokens": 19420820.0,
|
|
"step": 10655
|
|
},
|
|
{
|
|
"entropy": 5.460609817504883,
|
|
"epoch": 8.831814415907209,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.758244203663015e-05,
|
|
"loss": 4.6736,
|
|
"mean_token_accuracy": 0.22851051688194274,
|
|
"num_tokens": 19430321.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 5.418300914764404,
|
|
"epoch": 8.835956917978459,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.745881267775967e-05,
|
|
"loss": 4.5727,
|
|
"mean_token_accuracy": 0.24511416256427765,
|
|
"num_tokens": 19439134.0,
|
|
"step": 10665
|
|
},
|
|
{
|
|
"entropy": 5.419695615768433,
|
|
"epoch": 8.840099420049711,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.733560195214384e-05,
|
|
"loss": 4.5664,
|
|
"mean_token_accuracy": 0.24506309181451796,
|
|
"num_tokens": 19448503.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 5.481495332717896,
|
|
"epoch": 8.844241922120961,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.721281010831209e-05,
|
|
"loss": 4.6738,
|
|
"mean_token_accuracy": 0.23042037189006806,
|
|
"num_tokens": 19458423.0,
|
|
"step": 10675
|
|
},
|
|
{
|
|
"entropy": 5.384313488006592,
|
|
"epoch": 8.848384424192211,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.709043739394909e-05,
|
|
"loss": 4.5674,
|
|
"mean_token_accuracy": 0.23893739581108092,
|
|
"num_tokens": 19467346.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 5.420519590377808,
|
|
"epoch": 8.852526926263463,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.696848405589389e-05,
|
|
"loss": 4.5434,
|
|
"mean_token_accuracy": 0.24067894369363785,
|
|
"num_tokens": 19476659.0,
|
|
"step": 10685
|
|
},
|
|
{
|
|
"entropy": 5.395295524597168,
|
|
"epoch": 8.856669428334714,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.68469503401398e-05,
|
|
"loss": 4.5258,
|
|
"mean_token_accuracy": 0.24832364469766616,
|
|
"num_tokens": 19486036.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 5.410879325866699,
|
|
"epoch": 8.860811930405966,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.672583649183346e-05,
|
|
"loss": 4.5857,
|
|
"mean_token_accuracy": 0.24163961559534072,
|
|
"num_tokens": 19494964.0,
|
|
"step": 10695
|
|
},
|
|
{
|
|
"entropy": 5.393184185028076,
|
|
"epoch": 8.864954432477216,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.66051427552749e-05,
|
|
"loss": 4.583,
|
|
"mean_token_accuracy": 0.2414379522204399,
|
|
"num_tokens": 19503293.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 5.384077262878418,
|
|
"epoch": 8.869096934548468,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.648486937391651e-05,
|
|
"loss": 4.5666,
|
|
"mean_token_accuracy": 0.24330753982067108,
|
|
"num_tokens": 19512286.0,
|
|
"step": 10705
|
|
},
|
|
{
|
|
"entropy": 5.565299844741821,
|
|
"epoch": 8.873239436619718,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.636501659036278e-05,
|
|
"loss": 4.7238,
|
|
"mean_token_accuracy": 0.23481334894895553,
|
|
"num_tokens": 19522244.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 5.4990989685058596,
|
|
"epoch": 8.87738193869097,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.624558464636999e-05,
|
|
"loss": 4.6581,
|
|
"mean_token_accuracy": 0.2363555446267128,
|
|
"num_tokens": 19531811.0,
|
|
"step": 10715
|
|
},
|
|
{
|
|
"entropy": 5.477078771591186,
|
|
"epoch": 8.88152444076222,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.612657378284528e-05,
|
|
"loss": 4.6759,
|
|
"mean_token_accuracy": 0.22979843020439147,
|
|
"num_tokens": 19541569.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 5.418045949935913,
|
|
"epoch": 8.885666942833472,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.600798423984671e-05,
|
|
"loss": 4.5657,
|
|
"mean_token_accuracy": 0.24805757999420167,
|
|
"num_tokens": 19549904.0,
|
|
"step": 10725
|
|
},
|
|
{
|
|
"entropy": 5.394879388809204,
|
|
"epoch": 8.889809444904722,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.588981625658227e-05,
|
|
"loss": 4.5475,
|
|
"mean_token_accuracy": 0.23965625762939452,
|
|
"num_tokens": 19558804.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 5.441050434112549,
|
|
"epoch": 8.893951946975973,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.577207007140969e-05,
|
|
"loss": 4.5394,
|
|
"mean_token_accuracy": 0.2405978634953499,
|
|
"num_tokens": 19567060.0,
|
|
"step": 10735
|
|
},
|
|
{
|
|
"entropy": 5.432073259353638,
|
|
"epoch": 8.898094449047225,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 6.565474592183594e-05,
|
|
"loss": 4.6214,
|
|
"mean_token_accuracy": 0.22916834205389022,
|
|
"num_tokens": 19576813.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 5.388404798507691,
|
|
"epoch": 8.902236951118475,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 6.553784404451656e-05,
|
|
"loss": 4.5475,
|
|
"mean_token_accuracy": 0.24201963543891908,
|
|
"num_tokens": 19585551.0,
|
|
"step": 10745
|
|
},
|
|
{
|
|
"entropy": 5.444418859481812,
|
|
"epoch": 8.906379453189727,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.542136467525555e-05,
|
|
"loss": 4.5835,
|
|
"mean_token_accuracy": 0.24189568012952806,
|
|
"num_tokens": 19595085.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 5.43736834526062,
|
|
"epoch": 8.910521955260977,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.530530804900441e-05,
|
|
"loss": 4.5522,
|
|
"mean_token_accuracy": 0.2458910286426544,
|
|
"num_tokens": 19603260.0,
|
|
"step": 10755
|
|
},
|
|
{
|
|
"entropy": 5.466627597808838,
|
|
"epoch": 8.91466445733223,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.518967439986209e-05,
|
|
"loss": 4.6267,
|
|
"mean_token_accuracy": 0.2414685904979706,
|
|
"num_tokens": 19611608.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 5.401951551437378,
|
|
"epoch": 8.91880695940348,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.507446396107429e-05,
|
|
"loss": 4.5259,
|
|
"mean_token_accuracy": 0.24022713750600816,
|
|
"num_tokens": 19620214.0,
|
|
"step": 10765
|
|
},
|
|
{
|
|
"entropy": 5.439483690261841,
|
|
"epoch": 8.922949461474731,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.495967696503303e-05,
|
|
"loss": 4.5907,
|
|
"mean_token_accuracy": 0.24184695333242417,
|
|
"num_tokens": 19629227.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 5.506908655166626,
|
|
"epoch": 8.927091963545982,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.484531364327629e-05,
|
|
"loss": 4.6179,
|
|
"mean_token_accuracy": 0.24500985741615294,
|
|
"num_tokens": 19638348.0,
|
|
"step": 10775
|
|
},
|
|
{
|
|
"entropy": 5.403257179260254,
|
|
"epoch": 8.931234465617234,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 6.473137422648728e-05,
|
|
"loss": 4.4776,
|
|
"mean_token_accuracy": 0.25743522942066194,
|
|
"num_tokens": 19646480.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 5.4452352046966555,
|
|
"epoch": 8.935376967688484,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.461785894449434e-05,
|
|
"loss": 4.5809,
|
|
"mean_token_accuracy": 0.24043982475996017,
|
|
"num_tokens": 19656188.0,
|
|
"step": 10785
|
|
},
|
|
{
|
|
"entropy": 5.3688891410827635,
|
|
"epoch": 8.939519469759734,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.45047680262702e-05,
|
|
"loss": 4.6162,
|
|
"mean_token_accuracy": 0.2312155023217201,
|
|
"num_tokens": 19664849.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 5.436012125015258,
|
|
"epoch": 8.943661971830986,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 6.439210169993153e-05,
|
|
"loss": 4.5836,
|
|
"mean_token_accuracy": 0.24008898735046386,
|
|
"num_tokens": 19675104.0,
|
|
"step": 10795
|
|
},
|
|
{
|
|
"entropy": 5.429963064193726,
|
|
"epoch": 8.947804473902236,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.427986019273873e-05,
|
|
"loss": 4.6023,
|
|
"mean_token_accuracy": 0.23461055159568786,
|
|
"num_tokens": 19685009.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 5.453448104858398,
|
|
"epoch": 8.951946975973488,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.416804373109507e-05,
|
|
"loss": 4.617,
|
|
"mean_token_accuracy": 0.2422842413187027,
|
|
"num_tokens": 19694190.0,
|
|
"step": 10805
|
|
},
|
|
{
|
|
"entropy": 5.39255199432373,
|
|
"epoch": 8.956089478044738,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.405665254054666e-05,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.23304208517074584,
|
|
"num_tokens": 19703571.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 5.466146755218506,
|
|
"epoch": 8.96023198011599,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.394568684578165e-05,
|
|
"loss": 4.5684,
|
|
"mean_token_accuracy": 0.24163964092731477,
|
|
"num_tokens": 19713365.0,
|
|
"step": 10815
|
|
},
|
|
{
|
|
"entropy": 5.541957521438599,
|
|
"epoch": 8.96437448218724,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.383514687063e-05,
|
|
"loss": 4.6999,
|
|
"mean_token_accuracy": 0.23662541657686234,
|
|
"num_tokens": 19722535.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 5.353496503829956,
|
|
"epoch": 8.968516984258493,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.372503283806291e-05,
|
|
"loss": 4.5133,
|
|
"mean_token_accuracy": 0.25290452837944033,
|
|
"num_tokens": 19731563.0,
|
|
"step": 10825
|
|
},
|
|
{
|
|
"entropy": 5.446975660324097,
|
|
"epoch": 8.972659486329743,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.36153449701924e-05,
|
|
"loss": 4.6548,
|
|
"mean_token_accuracy": 0.236767016351223,
|
|
"num_tokens": 19741093.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 5.448146390914917,
|
|
"epoch": 8.976801988400995,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.350608348827096e-05,
|
|
"loss": 4.6065,
|
|
"mean_token_accuracy": 0.2352151095867157,
|
|
"num_tokens": 19749489.0,
|
|
"step": 10835
|
|
},
|
|
{
|
|
"entropy": 5.4187112808227536,
|
|
"epoch": 8.980944490472245,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.339724861269085e-05,
|
|
"loss": 4.6218,
|
|
"mean_token_accuracy": 0.23796432912349702,
|
|
"num_tokens": 19758745.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 5.473947906494141,
|
|
"epoch": 8.985086992543497,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.328884056298393e-05,
|
|
"loss": 4.6655,
|
|
"mean_token_accuracy": 0.23023081868886947,
|
|
"num_tokens": 19767713.0,
|
|
"step": 10845
|
|
},
|
|
{
|
|
"entropy": 5.415466356277466,
|
|
"epoch": 8.989229494614747,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.31808595578211e-05,
|
|
"loss": 4.6217,
|
|
"mean_token_accuracy": 0.2348709747195244,
|
|
"num_tokens": 19777236.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 5.3990740299224855,
|
|
"epoch": 8.993371996685998,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.307330581501183e-05,
|
|
"loss": 4.5409,
|
|
"mean_token_accuracy": 0.24719997197389604,
|
|
"num_tokens": 19785981.0,
|
|
"step": 10855
|
|
},
|
|
{
|
|
"entropy": 5.409685277938843,
|
|
"epoch": 8.99751449875725,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 6.29661795515038e-05,
|
|
"loss": 4.5981,
|
|
"mean_token_accuracy": 0.24582814127206803,
|
|
"num_tokens": 19795046.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 5.510007858276367,
|
|
"epoch": 9.0016570008285,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 6.285948098338238e-05,
|
|
"loss": 4.6309,
|
|
"mean_token_accuracy": 0.2314115658402443,
|
|
"num_tokens": 19803108.0,
|
|
"step": 10865
|
|
},
|
|
{
|
|
"entropy": 5.496815490722656,
|
|
"epoch": 9.005799502899752,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.27532103258702e-05,
|
|
"loss": 4.613,
|
|
"mean_token_accuracy": 0.2355228453874588,
|
|
"num_tokens": 19812751.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 5.493884325027466,
|
|
"epoch": 9.009942004971002,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.26473677933269e-05,
|
|
"loss": 4.5955,
|
|
"mean_token_accuracy": 0.2399630591273308,
|
|
"num_tokens": 19821634.0,
|
|
"step": 10875
|
|
},
|
|
{
|
|
"entropy": 5.43362021446228,
|
|
"epoch": 9.014084507042254,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.254195359924827e-05,
|
|
"loss": 4.5416,
|
|
"mean_token_accuracy": 0.24990837275981903,
|
|
"num_tokens": 19830760.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 5.3868170261383055,
|
|
"epoch": 9.018227009113504,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.24369679562664e-05,
|
|
"loss": 4.4616,
|
|
"mean_token_accuracy": 0.2515067979693413,
|
|
"num_tokens": 19839463.0,
|
|
"step": 10885
|
|
},
|
|
{
|
|
"entropy": 5.418480014801025,
|
|
"epoch": 9.022369511184756,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.233241107614875e-05,
|
|
"loss": 4.5266,
|
|
"mean_token_accuracy": 0.2483330950140953,
|
|
"num_tokens": 19848256.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 5.464984035491943,
|
|
"epoch": 9.026512013256006,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.222828316979798e-05,
|
|
"loss": 4.6009,
|
|
"mean_token_accuracy": 0.2431696280837059,
|
|
"num_tokens": 19857572.0,
|
|
"step": 10895
|
|
},
|
|
{
|
|
"entropy": 5.439783573150635,
|
|
"epoch": 9.030654515327258,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.212458444725146e-05,
|
|
"loss": 4.5731,
|
|
"mean_token_accuracy": 0.2408907949924469,
|
|
"num_tokens": 19866825.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 5.533265447616577,
|
|
"epoch": 9.034797017398509,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.202131511768079e-05,
|
|
"loss": 4.6896,
|
|
"mean_token_accuracy": 0.23642997443675995,
|
|
"num_tokens": 19875970.0,
|
|
"step": 10905
|
|
},
|
|
{
|
|
"entropy": 5.433733129501343,
|
|
"epoch": 9.03893951946976,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.19184753893916e-05,
|
|
"loss": 4.5614,
|
|
"mean_token_accuracy": 0.24125870317220688,
|
|
"num_tokens": 19884910.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 5.480454063415527,
|
|
"epoch": 9.04308202154101,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.181606546982274e-05,
|
|
"loss": 4.6122,
|
|
"mean_token_accuracy": 0.23832860887050628,
|
|
"num_tokens": 19894838.0,
|
|
"step": 10915
|
|
},
|
|
{
|
|
"entropy": 5.527755165100098,
|
|
"epoch": 9.047224523612261,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.171408556554633e-05,
|
|
"loss": 4.6843,
|
|
"mean_token_accuracy": 0.23294905722141265,
|
|
"num_tokens": 19904809.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 5.352613353729248,
|
|
"epoch": 9.051367025683513,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.16125358822669e-05,
|
|
"loss": 4.4414,
|
|
"mean_token_accuracy": 0.2558871150016785,
|
|
"num_tokens": 19913549.0,
|
|
"step": 10925
|
|
},
|
|
{
|
|
"entropy": 5.4163788795471195,
|
|
"epoch": 9.055509527754763,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 6.151141662482126e-05,
|
|
"loss": 4.5559,
|
|
"mean_token_accuracy": 0.24614081233739854,
|
|
"num_tokens": 19922572.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 5.4976154327392575,
|
|
"epoch": 9.059652029826015,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 6.141072799717805e-05,
|
|
"loss": 4.6439,
|
|
"mean_token_accuracy": 0.23508770316839217,
|
|
"num_tokens": 19932058.0,
|
|
"step": 10935
|
|
},
|
|
{
|
|
"entropy": 5.371648168563842,
|
|
"epoch": 9.063794531897265,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.131047020243717e-05,
|
|
"loss": 4.4995,
|
|
"mean_token_accuracy": 0.2456317812204361,
|
|
"num_tokens": 19940406.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 5.422821378707885,
|
|
"epoch": 9.067937033968517,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.121064344282962e-05,
|
|
"loss": 4.5485,
|
|
"mean_token_accuracy": 0.24659007787704468,
|
|
"num_tokens": 19949181.0,
|
|
"step": 10945
|
|
},
|
|
{
|
|
"entropy": 5.480816507339478,
|
|
"epoch": 9.072079536039768,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.111124791971687e-05,
|
|
"loss": 4.6167,
|
|
"mean_token_accuracy": 0.23715528547763826,
|
|
"num_tokens": 19958159.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 5.423428773880005,
|
|
"epoch": 9.07622203811102,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.1012283833590465e-05,
|
|
"loss": 4.5577,
|
|
"mean_token_accuracy": 0.23919959962368012,
|
|
"num_tokens": 19967614.0,
|
|
"step": 10955
|
|
},
|
|
{
|
|
"entropy": 5.410855293273926,
|
|
"epoch": 9.08036454018227,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.091375138407189e-05,
|
|
"loss": 4.5405,
|
|
"mean_token_accuracy": 0.24402375072240828,
|
|
"num_tokens": 19977697.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 5.545088863372802,
|
|
"epoch": 9.084507042253522,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.0815650769911774e-05,
|
|
"loss": 4.6417,
|
|
"mean_token_accuracy": 0.2320782095193863,
|
|
"num_tokens": 19986152.0,
|
|
"step": 10965
|
|
},
|
|
{
|
|
"entropy": 5.454336929321289,
|
|
"epoch": 9.088649544324772,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.0717982188989884e-05,
|
|
"loss": 4.5799,
|
|
"mean_token_accuracy": 0.23144979029893875,
|
|
"num_tokens": 19995904.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 5.402749347686767,
|
|
"epoch": 9.092792046396022,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.062074583831432e-05,
|
|
"loss": 4.6115,
|
|
"mean_token_accuracy": 0.24551478326320647,
|
|
"num_tokens": 20005516.0,
|
|
"step": 10975
|
|
},
|
|
{
|
|
"entropy": 5.400489330291748,
|
|
"epoch": 9.096934548467274,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.052394191402145e-05,
|
|
"loss": 4.522,
|
|
"mean_token_accuracy": 0.2455115333199501,
|
|
"num_tokens": 20014183.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 5.397538757324218,
|
|
"epoch": 9.101077050538525,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.042757061137545e-05,
|
|
"loss": 4.5768,
|
|
"mean_token_accuracy": 0.23675381988286973,
|
|
"num_tokens": 20023537.0,
|
|
"step": 10985
|
|
},
|
|
{
|
|
"entropy": 5.5187798023223875,
|
|
"epoch": 9.105219552609777,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.033163212476762e-05,
|
|
"loss": 4.6365,
|
|
"mean_token_accuracy": 0.23603421002626418,
|
|
"num_tokens": 20032639.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 5.4630982875823975,
|
|
"epoch": 9.109362054681027,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.023612664771651e-05,
|
|
"loss": 4.599,
|
|
"mean_token_accuracy": 0.24760505706071853,
|
|
"num_tokens": 20041273.0,
|
|
"step": 10995
|
|
},
|
|
{
|
|
"entropy": 5.508844423294067,
|
|
"epoch": 9.113504556752279,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.014105437286705e-05,
|
|
"loss": 4.5189,
|
|
"mean_token_accuracy": 0.2503361001610756,
|
|
"num_tokens": 20049730.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"epoch": 9.113504556752279,
|
|
"eval_entropy": 5.266478903512202,
|
|
"eval_loss": 5.735865592956543,
|
|
"eval_mean_token_accuracy": 0.1687001271415621,
|
|
"eval_num_tokens": 20049730.0,
|
|
"eval_runtime": 2.111,
|
|
"eval_samples_per_second": 1774.994,
|
|
"eval_steps_per_second": 222.17,
|
|
"step": 11000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12060,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 4435584930846720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|