10145 lines
276 KiB
JSON
10145 lines
276 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.142502071251036,
|
|
"eval_steps": 500,
|
|
"global_step": 5000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 2018763363962880.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|