3035 lines
82 KiB
JSON
3035 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.08779631255487269,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.742569732666016,
|
|
"epoch": 0.000292654375182909,
|
|
"grad_norm": 4.8125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.776,
|
|
"mean_token_accuracy": 0.00037909409729763865,
|
|
"num_tokens": 10517.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742593669891358,
|
|
"epoch": 0.000585308750365818,
|
|
"grad_norm": 4.78125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7562,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 20499.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.74259843826294,
|
|
"epoch": 0.000877963125548727,
|
|
"grad_norm": 5.53125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.7254,
|
|
"mean_token_accuracy": 0.00031158956699073315,
|
|
"num_tokens": 30396.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742587089538574,
|
|
"epoch": 0.001170617500731636,
|
|
"grad_norm": 5.21875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6813,
|
|
"mean_token_accuracy": 0.0014116012142039835,
|
|
"num_tokens": 40862.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742495727539062,
|
|
"epoch": 0.0014632718759145448,
|
|
"grad_norm": 4.46875,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.573,
|
|
"mean_token_accuracy": 0.015958905406296253,
|
|
"num_tokens": 50482.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.74194974899292,
|
|
"epoch": 0.001755926251097454,
|
|
"grad_norm": 3.9375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4277,
|
|
"mean_token_accuracy": 0.046428434550762177,
|
|
"num_tokens": 59557.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.739844131469727,
|
|
"epoch": 0.002048580626280363,
|
|
"grad_norm": 3.4375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3155,
|
|
"mean_token_accuracy": 0.04745985008776188,
|
|
"num_tokens": 69864.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.732469940185547,
|
|
"epoch": 0.002341235001463272,
|
|
"grad_norm": 2.484375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1801,
|
|
"mean_token_accuracy": 0.056155404448509215,
|
|
"num_tokens": 80632.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.719615936279297,
|
|
"epoch": 0.0026338893766461808,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0892,
|
|
"mean_token_accuracy": 0.049328677356243134,
|
|
"num_tokens": 90889.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.709369468688966,
|
|
"epoch": 0.0029265437518290896,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9839,
|
|
"mean_token_accuracy": 0.05372027605772019,
|
|
"num_tokens": 101136.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.702022743225097,
|
|
"epoch": 0.003219198127011999,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9529,
|
|
"mean_token_accuracy": 0.05223828963935375,
|
|
"num_tokens": 111743.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.692447948455811,
|
|
"epoch": 0.003511852502194908,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.9212,
|
|
"mean_token_accuracy": 0.050166602805256845,
|
|
"num_tokens": 121968.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.686018753051759,
|
|
"epoch": 0.0038045068773778167,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8286,
|
|
"mean_token_accuracy": 0.055526862666010854,
|
|
"num_tokens": 131917.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.676072120666504,
|
|
"epoch": 0.004097161252560726,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7736,
|
|
"mean_token_accuracy": 0.05640910603106022,
|
|
"num_tokens": 141562.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.663375949859619,
|
|
"epoch": 0.004389815627743635,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.755,
|
|
"mean_token_accuracy": 0.05814168080687523,
|
|
"num_tokens": 152163.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.652446174621582,
|
|
"epoch": 0.004682470002926544,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.7531,
|
|
"mean_token_accuracy": 0.05046718679368496,
|
|
"num_tokens": 162684.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.65147132873535,
|
|
"epoch": 0.004975124378109453,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.6277,
|
|
"mean_token_accuracy": 0.055151918530464174,
|
|
"num_tokens": 173234.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.643102264404297,
|
|
"epoch": 0.0052677787532923615,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.6174,
|
|
"mean_token_accuracy": 0.05056312084197998,
|
|
"num_tokens": 183645.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.629851245880127,
|
|
"epoch": 0.00556043312847527,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.5472,
|
|
"mean_token_accuracy": 0.05371568165719509,
|
|
"num_tokens": 193874.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.60717601776123,
|
|
"epoch": 0.005853087503658179,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.4433,
|
|
"mean_token_accuracy": 0.05704141147434712,
|
|
"num_tokens": 203965.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.582340145111084,
|
|
"epoch": 0.006145741878841089,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.3847,
|
|
"mean_token_accuracy": 0.05947887413203716,
|
|
"num_tokens": 214334.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.55553674697876,
|
|
"epoch": 0.006438396254023998,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.3073,
|
|
"mean_token_accuracy": 0.06609507203102112,
|
|
"num_tokens": 225435.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.51070203781128,
|
|
"epoch": 0.006731050629206907,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1764,
|
|
"mean_token_accuracy": 0.06483110673725605,
|
|
"num_tokens": 235107.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.439863014221192,
|
|
"epoch": 0.007023705004389816,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.1014,
|
|
"mean_token_accuracy": 0.06455096490681171,
|
|
"num_tokens": 245660.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.376344776153564,
|
|
"epoch": 0.0073163593795727245,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.0183,
|
|
"mean_token_accuracy": 0.06346280165016652,
|
|
"num_tokens": 256077.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.332421588897706,
|
|
"epoch": 0.007609013754755633,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.9453,
|
|
"mean_token_accuracy": 0.0634394098073244,
|
|
"num_tokens": 265702.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.269437503814697,
|
|
"epoch": 0.007901668129938543,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.88,
|
|
"mean_token_accuracy": 0.06306366845965386,
|
|
"num_tokens": 276739.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.201168727874755,
|
|
"epoch": 0.008194322505121452,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.7255,
|
|
"mean_token_accuracy": 0.06702562645077706,
|
|
"num_tokens": 285370.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.072903537750244,
|
|
"epoch": 0.008486976880304361,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6633,
|
|
"mean_token_accuracy": 0.06559402719140053,
|
|
"num_tokens": 295080.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.988585090637207,
|
|
"epoch": 0.00877963125548727,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.5425,
|
|
"mean_token_accuracy": 0.06815728284418583,
|
|
"num_tokens": 304509.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.857632827758788,
|
|
"epoch": 0.009072285630670179,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.4977,
|
|
"mean_token_accuracy": 0.06578324660658837,
|
|
"num_tokens": 315214.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.758829021453858,
|
|
"epoch": 0.009364940005853088,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.4204,
|
|
"mean_token_accuracy": 0.06578456349670887,
|
|
"num_tokens": 325865.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.577184200286865,
|
|
"epoch": 0.009657594381035996,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.3182,
|
|
"mean_token_accuracy": 0.06833198070526122,
|
|
"num_tokens": 335490.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.448071479797363,
|
|
"epoch": 0.009950248756218905,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.3465,
|
|
"mean_token_accuracy": 0.0624022550880909,
|
|
"num_tokens": 345507.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.283161449432374,
|
|
"epoch": 0.010242903131401814,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.2345,
|
|
"mean_token_accuracy": 0.06619864143431187,
|
|
"num_tokens": 355129.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.061387634277343,
|
|
"epoch": 0.010535557506584723,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.1198,
|
|
"mean_token_accuracy": 0.0754898265004158,
|
|
"num_tokens": 364700.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.924063873291015,
|
|
"epoch": 0.010828211881767632,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.1966,
|
|
"mean_token_accuracy": 0.07078267820179462,
|
|
"num_tokens": 374920.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.817254543304443,
|
|
"epoch": 0.01112086625695054,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.9958,
|
|
"mean_token_accuracy": 0.0730692483484745,
|
|
"num_tokens": 384517.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.686346626281738,
|
|
"epoch": 0.01141352063213345,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.1191,
|
|
"mean_token_accuracy": 0.07759304121136665,
|
|
"num_tokens": 394118.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.639398574829102,
|
|
"epoch": 0.011706175007316359,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.1035,
|
|
"mean_token_accuracy": 0.07444557547569275,
|
|
"num_tokens": 404135.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.65887680053711,
|
|
"epoch": 0.01199882938249927,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.0896,
|
|
"mean_token_accuracy": 0.07085032351315021,
|
|
"num_tokens": 414198.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.562304019927979,
|
|
"epoch": 0.012291483757682178,
|
|
"grad_norm": 0.62109375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.0473,
|
|
"mean_token_accuracy": 0.07471234947443009,
|
|
"num_tokens": 424747.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.543915939331054,
|
|
"epoch": 0.012584138132865087,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.0208,
|
|
"mean_token_accuracy": 0.07326471023261547,
|
|
"num_tokens": 435597.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.513623428344726,
|
|
"epoch": 0.012876792508047996,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.064,
|
|
"mean_token_accuracy": 0.07332713790237903,
|
|
"num_tokens": 445262.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.550195503234864,
|
|
"epoch": 0.013169446883230905,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.094,
|
|
"mean_token_accuracy": 0.0686261247843504,
|
|
"num_tokens": 456455.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.483574199676514,
|
|
"epoch": 0.013462101258413814,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.9287,
|
|
"mean_token_accuracy": 0.07206557914614678,
|
|
"num_tokens": 465696.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.426221942901611,
|
|
"epoch": 0.013754755633596722,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.9791,
|
|
"mean_token_accuracy": 0.07656441517174244,
|
|
"num_tokens": 475171.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.437851619720458,
|
|
"epoch": 0.014047410008779631,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.9953,
|
|
"mean_token_accuracy": 0.07377254441380501,
|
|
"num_tokens": 484852.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.439754581451416,
|
|
"epoch": 0.01434006438396254,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.0016,
|
|
"mean_token_accuracy": 0.07172177135944366,
|
|
"num_tokens": 494539.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.422525405883789,
|
|
"epoch": 0.014632718759145449,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.9657,
|
|
"mean_token_accuracy": 0.06861439161002636,
|
|
"num_tokens": 504406.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.499318695068359,
|
|
"epoch": 0.014925373134328358,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.9752,
|
|
"mean_token_accuracy": 0.07367737665772438,
|
|
"num_tokens": 514905.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.41889181137085,
|
|
"epoch": 0.015218027509511267,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.936,
|
|
"mean_token_accuracy": 0.08054133653640747,
|
|
"num_tokens": 524904.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.374094104766845,
|
|
"epoch": 0.015510681884694176,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.9791,
|
|
"mean_token_accuracy": 0.07328154221177101,
|
|
"num_tokens": 534423.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.391270446777344,
|
|
"epoch": 0.015803336259877086,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.0123,
|
|
"mean_token_accuracy": 0.06913212612271309,
|
|
"num_tokens": 544468.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.460853576660156,
|
|
"epoch": 0.016095990635059995,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.9631,
|
|
"mean_token_accuracy": 0.07310855686664582,
|
|
"num_tokens": 555873.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.311201095581055,
|
|
"epoch": 0.016388645010242904,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.9352,
|
|
"mean_token_accuracy": 0.07407794632017613,
|
|
"num_tokens": 566228.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.375296783447265,
|
|
"epoch": 0.016681299385425813,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.8366,
|
|
"mean_token_accuracy": 0.07612851671874524,
|
|
"num_tokens": 576087.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.290613842010497,
|
|
"epoch": 0.016973953760608722,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.9593,
|
|
"mean_token_accuracy": 0.07361815236508847,
|
|
"num_tokens": 586768.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.414975547790528,
|
|
"epoch": 0.01726660813579163,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.9496,
|
|
"mean_token_accuracy": 0.07182935290038586,
|
|
"num_tokens": 597352.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.286548233032226,
|
|
"epoch": 0.01755926251097454,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.8397,
|
|
"mean_token_accuracy": 0.07391503937542439,
|
|
"num_tokens": 607460.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.280863857269287,
|
|
"epoch": 0.01785191688615745,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.8721,
|
|
"mean_token_accuracy": 0.07399217337369919,
|
|
"num_tokens": 618882.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.382979774475098,
|
|
"epoch": 0.018144571261340357,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.9746,
|
|
"mean_token_accuracy": 0.06754698418080807,
|
|
"num_tokens": 628378.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.238453578948974,
|
|
"epoch": 0.018437225636523266,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.913,
|
|
"mean_token_accuracy": 0.07458860278129578,
|
|
"num_tokens": 637339.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.266683578491211,
|
|
"epoch": 0.018729880011706175,
|
|
"grad_norm": 0.6953125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.9179,
|
|
"mean_token_accuracy": 0.07612942159175873,
|
|
"num_tokens": 648190.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.325155258178711,
|
|
"epoch": 0.019022534386889084,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.8386,
|
|
"mean_token_accuracy": 0.07890116423368454,
|
|
"num_tokens": 658704.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.223814582824707,
|
|
"epoch": 0.019315188762071993,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.835,
|
|
"mean_token_accuracy": 0.0768560890108347,
|
|
"num_tokens": 668888.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.254367637634278,
|
|
"epoch": 0.0196078431372549,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.8646,
|
|
"mean_token_accuracy": 0.07014993466436863,
|
|
"num_tokens": 679294.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.180438041687012,
|
|
"epoch": 0.01990049751243781,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.8385,
|
|
"mean_token_accuracy": 0.07648964524269104,
|
|
"num_tokens": 689662.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.204143142700195,
|
|
"epoch": 0.02019315188762072,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.8542,
|
|
"mean_token_accuracy": 0.07869461253285408,
|
|
"num_tokens": 700356.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.173739433288574,
|
|
"epoch": 0.02048580626280363,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.7786,
|
|
"mean_token_accuracy": 0.07795644551515579,
|
|
"num_tokens": 710831.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.16906099319458,
|
|
"epoch": 0.020778460637986537,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.7646,
|
|
"mean_token_accuracy": 0.07683240249752998,
|
|
"num_tokens": 720445.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.170342254638673,
|
|
"epoch": 0.021071115013169446,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.7871,
|
|
"mean_token_accuracy": 0.08130665272474288,
|
|
"num_tokens": 730355.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.122212743759155,
|
|
"epoch": 0.021363769388352355,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.7918,
|
|
"mean_token_accuracy": 0.08222091645002365,
|
|
"num_tokens": 740397.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.200673866271973,
|
|
"epoch": 0.021656423763535264,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.7934,
|
|
"mean_token_accuracy": 0.0756984956562519,
|
|
"num_tokens": 750277.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.109900617599488,
|
|
"epoch": 0.021949078138718173,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.7735,
|
|
"mean_token_accuracy": 0.07954552099108696,
|
|
"num_tokens": 759764.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.258574295043946,
|
|
"epoch": 0.02224173251390108,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.8364,
|
|
"mean_token_accuracy": 0.0779568761587143,
|
|
"num_tokens": 769709.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.177908420562744,
|
|
"epoch": 0.02253438688908399,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.7721,
|
|
"mean_token_accuracy": 0.0785105787217617,
|
|
"num_tokens": 780267.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.026059818267822,
|
|
"epoch": 0.0228270412642669,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.7254,
|
|
"mean_token_accuracy": 0.0828751116991043,
|
|
"num_tokens": 790184.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.152198600769044,
|
|
"epoch": 0.023119695639449808,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.8261,
|
|
"mean_token_accuracy": 0.07748916000127792,
|
|
"num_tokens": 800645.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.190797138214112,
|
|
"epoch": 0.023412350014632717,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.7524,
|
|
"mean_token_accuracy": 0.08283478766679764,
|
|
"num_tokens": 810873.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.11506757736206,
|
|
"epoch": 0.02370500438981563,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.7456,
|
|
"mean_token_accuracy": 0.07929125241935253,
|
|
"num_tokens": 821661.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.140167808532714,
|
|
"epoch": 0.02399765876499854,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.7449,
|
|
"mean_token_accuracy": 0.07854388505220414,
|
|
"num_tokens": 831818.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.042124271392822,
|
|
"epoch": 0.024290313140181447,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.7535,
|
|
"mean_token_accuracy": 0.08163447454571723,
|
|
"num_tokens": 843030.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.099297571182252,
|
|
"epoch": 0.024582967515364356,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.7175,
|
|
"mean_token_accuracy": 0.08631875067949295,
|
|
"num_tokens": 852477.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.106180906295776,
|
|
"epoch": 0.024875621890547265,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.7318,
|
|
"mean_token_accuracy": 0.08200445771217346,
|
|
"num_tokens": 863018.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.047952270507812,
|
|
"epoch": 0.025168276265730174,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.6688,
|
|
"mean_token_accuracy": 0.082207240909338,
|
|
"num_tokens": 872732.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.055245399475098,
|
|
"epoch": 0.025460930640913083,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.7356,
|
|
"mean_token_accuracy": 0.0770014464855194,
|
|
"num_tokens": 883189.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.050556993484497,
|
|
"epoch": 0.02575358501609599,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.7489,
|
|
"mean_token_accuracy": 0.0801065780222416,
|
|
"num_tokens": 893874.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.03118143081665,
|
|
"epoch": 0.0260462393912789,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.6741,
|
|
"mean_token_accuracy": 0.08284056335687637,
|
|
"num_tokens": 904424.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.11718683242798,
|
|
"epoch": 0.02633889376646181,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.7431,
|
|
"mean_token_accuracy": 0.07791092544794083,
|
|
"num_tokens": 914601.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.059693479537964,
|
|
"epoch": 0.026631548141644718,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.7462,
|
|
"mean_token_accuracy": 0.07850845158100128,
|
|
"num_tokens": 925162.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.002055263519287,
|
|
"epoch": 0.026924202516827627,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.6594,
|
|
"mean_token_accuracy": 0.08002852946519852,
|
|
"num_tokens": 935182.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.099039840698243,
|
|
"epoch": 0.027216856892010536,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.6897,
|
|
"mean_token_accuracy": 0.08213006183505059,
|
|
"num_tokens": 944812.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.961377811431885,
|
|
"epoch": 0.027509511267193445,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.6964,
|
|
"mean_token_accuracy": 0.07877071015536785,
|
|
"num_tokens": 954532.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.022741889953613,
|
|
"epoch": 0.027802165642376354,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.639,
|
|
"mean_token_accuracy": 0.08602378293871879,
|
|
"num_tokens": 963611.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.00587968826294,
|
|
"epoch": 0.028094820017559263,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.688,
|
|
"mean_token_accuracy": 0.08180306293070316,
|
|
"num_tokens": 973177.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.030350875854491,
|
|
"epoch": 0.02838747439274217,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.6898,
|
|
"mean_token_accuracy": 0.0791365496814251,
|
|
"num_tokens": 983623.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.03669090270996,
|
|
"epoch": 0.02868012876792508,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.7472,
|
|
"mean_token_accuracy": 0.08025188818573951,
|
|
"num_tokens": 993810.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.088376188278199,
|
|
"epoch": 0.02897278314310799,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.6529,
|
|
"mean_token_accuracy": 0.0789113201200962,
|
|
"num_tokens": 1004155.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.990477657318115,
|
|
"epoch": 0.029265437518290898,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.612,
|
|
"mean_token_accuracy": 0.08940722271800042,
|
|
"num_tokens": 1014896.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.972944545745849,
|
|
"epoch": 0.029558091893473807,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.5516,
|
|
"mean_token_accuracy": 0.0878017283976078,
|
|
"num_tokens": 1024991.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.899525737762451,
|
|
"epoch": 0.029850746268656716,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.5796,
|
|
"mean_token_accuracy": 0.08802613019943237,
|
|
"num_tokens": 1036697.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.065279579162597,
|
|
"epoch": 0.030143400643839625,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.6774,
|
|
"mean_token_accuracy": 0.07723400816321373,
|
|
"num_tokens": 1046245.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.788718843460083,
|
|
"epoch": 0.030436055019022534,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.5633,
|
|
"mean_token_accuracy": 0.08955003842711448,
|
|
"num_tokens": 1056232.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.821068525314331,
|
|
"epoch": 0.030728709394205442,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.4477,
|
|
"mean_token_accuracy": 0.09515074491500855,
|
|
"num_tokens": 1065536.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.05577802658081,
|
|
"epoch": 0.03102136376938835,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.5794,
|
|
"mean_token_accuracy": 0.08964744806289673,
|
|
"num_tokens": 1075135.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.835528612136841,
|
|
"epoch": 0.031314018144571264,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.5717,
|
|
"mean_token_accuracy": 0.0826539345085621,
|
|
"num_tokens": 1085498.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.983124303817749,
|
|
"epoch": 0.03160667251975417,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.6005,
|
|
"mean_token_accuracy": 0.08875111043453217,
|
|
"num_tokens": 1095559.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.931251287460327,
|
|
"epoch": 0.03189932689493708,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.5456,
|
|
"mean_token_accuracy": 0.08640314042568206,
|
|
"num_tokens": 1105701.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.889260959625244,
|
|
"epoch": 0.03219198127011999,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.5431,
|
|
"mean_token_accuracy": 0.08784606605768204,
|
|
"num_tokens": 1116455.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.8568816661834715,
|
|
"epoch": 0.0324846356453029,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5697,
|
|
"mean_token_accuracy": 0.08820039629936219,
|
|
"num_tokens": 1126596.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.986862754821777,
|
|
"epoch": 0.03277729002048581,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.5865,
|
|
"mean_token_accuracy": 0.08685005903244018,
|
|
"num_tokens": 1136969.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.84932541847229,
|
|
"epoch": 0.03306994439566872,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.5699,
|
|
"mean_token_accuracy": 0.08707550391554833,
|
|
"num_tokens": 1147175.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.847287702560425,
|
|
"epoch": 0.033362598770851626,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.527,
|
|
"mean_token_accuracy": 0.09069008976221085,
|
|
"num_tokens": 1156539.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.910270690917969,
|
|
"epoch": 0.033655253146034535,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.5107,
|
|
"mean_token_accuracy": 0.0893574945628643,
|
|
"num_tokens": 1166630.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.857235383987427,
|
|
"epoch": 0.033947907521217444,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.5453,
|
|
"mean_token_accuracy": 0.08516263216733932,
|
|
"num_tokens": 1176279.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.920040893554687,
|
|
"epoch": 0.03424056189640035,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.543,
|
|
"mean_token_accuracy": 0.08200450427830219,
|
|
"num_tokens": 1186321.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.912441968917847,
|
|
"epoch": 0.03453321627158326,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.5193,
|
|
"mean_token_accuracy": 0.08341860473155975,
|
|
"num_tokens": 1198287.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.7401360988616945,
|
|
"epoch": 0.03482587064676617,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.5164,
|
|
"mean_token_accuracy": 0.08626452758908272,
|
|
"num_tokens": 1208618.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.83452959060669,
|
|
"epoch": 0.03511852502194908,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.5024,
|
|
"mean_token_accuracy": 0.08562399223446845,
|
|
"num_tokens": 1217932.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.012685871124267,
|
|
"epoch": 0.03541117939713199,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.5414,
|
|
"mean_token_accuracy": 0.08212736770510673,
|
|
"num_tokens": 1227965.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.799877643585205,
|
|
"epoch": 0.0357038337723149,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.485,
|
|
"mean_token_accuracy": 0.08530130311846733,
|
|
"num_tokens": 1237879.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.793827056884766,
|
|
"epoch": 0.035996488147497806,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.4804,
|
|
"mean_token_accuracy": 0.0855767272412777,
|
|
"num_tokens": 1248547.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.876594161987304,
|
|
"epoch": 0.036289142522680715,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.5034,
|
|
"mean_token_accuracy": 0.08643778935074806,
|
|
"num_tokens": 1259462.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.922547101974487,
|
|
"epoch": 0.036581796897863623,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.5425,
|
|
"mean_token_accuracy": 0.0930032640695572,
|
|
"num_tokens": 1270668.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.821293497085572,
|
|
"epoch": 0.03687445127304653,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.506,
|
|
"mean_token_accuracy": 0.08481546565890312,
|
|
"num_tokens": 1280459.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.859448432922363,
|
|
"epoch": 0.03716710564822944,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.4332,
|
|
"mean_token_accuracy": 0.0836933895945549,
|
|
"num_tokens": 1291114.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.879277753829956,
|
|
"epoch": 0.03745976002341235,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.5801,
|
|
"mean_token_accuracy": 0.07807724177837372,
|
|
"num_tokens": 1301442.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.79024305343628,
|
|
"epoch": 0.03775241439859526,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.3892,
|
|
"mean_token_accuracy": 0.09089659005403519,
|
|
"num_tokens": 1311315.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.697639131546021,
|
|
"epoch": 0.03804506877377817,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4088,
|
|
"mean_token_accuracy": 0.09008980989456176,
|
|
"num_tokens": 1320962.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.803319597244263,
|
|
"epoch": 0.03833772314896108,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.4448,
|
|
"mean_token_accuracy": 0.08984620422124863,
|
|
"num_tokens": 1331006.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.771277570724488,
|
|
"epoch": 0.038630377524143986,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.425,
|
|
"mean_token_accuracy": 0.08424968719482422,
|
|
"num_tokens": 1341342.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.751668787002563,
|
|
"epoch": 0.038923031899326895,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.3853,
|
|
"mean_token_accuracy": 0.08578916639089584,
|
|
"num_tokens": 1350484.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.757629871368408,
|
|
"epoch": 0.0392156862745098,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.4261,
|
|
"mean_token_accuracy": 0.08848872929811477,
|
|
"num_tokens": 1360422.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.717638921737671,
|
|
"epoch": 0.03950834064969271,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.3313,
|
|
"mean_token_accuracy": 0.0953003279864788,
|
|
"num_tokens": 1370653.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.675066566467285,
|
|
"epoch": 0.03980099502487562,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.4036,
|
|
"mean_token_accuracy": 0.08738565891981125,
|
|
"num_tokens": 1379904.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.810481309890747,
|
|
"epoch": 0.04009364940005853,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.3752,
|
|
"mean_token_accuracy": 0.088729477673769,
|
|
"num_tokens": 1390292.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.815591716766358,
|
|
"epoch": 0.04038630377524144,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.3868,
|
|
"mean_token_accuracy": 0.09195382669568061,
|
|
"num_tokens": 1400089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.60293927192688,
|
|
"epoch": 0.04067895815042435,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.4062,
|
|
"mean_token_accuracy": 0.08808108903467655,
|
|
"num_tokens": 1409486.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.7336822032928465,
|
|
"epoch": 0.04097161252560726,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.4036,
|
|
"mean_token_accuracy": 0.09051820784807205,
|
|
"num_tokens": 1419834.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.825760173797607,
|
|
"epoch": 0.041264266900790166,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4213,
|
|
"mean_token_accuracy": 0.08384566307067871,
|
|
"num_tokens": 1430045.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.65966100692749,
|
|
"epoch": 0.041556921275973074,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.3022,
|
|
"mean_token_accuracy": 0.09362379759550095,
|
|
"num_tokens": 1439394.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.688346433639526,
|
|
"epoch": 0.04184957565115598,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.4215,
|
|
"mean_token_accuracy": 0.09055223762989044,
|
|
"num_tokens": 1449768.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.685902833938599,
|
|
"epoch": 0.04214223002633889,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.3477,
|
|
"mean_token_accuracy": 0.09224402233958244,
|
|
"num_tokens": 1460716.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.657220983505249,
|
|
"epoch": 0.0424348844015218,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.3045,
|
|
"mean_token_accuracy": 0.09458784088492393,
|
|
"num_tokens": 1470447.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.609696722030639,
|
|
"epoch": 0.04272753877670471,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.3177,
|
|
"mean_token_accuracy": 0.09160968512296677,
|
|
"num_tokens": 1481195.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.631748199462891,
|
|
"epoch": 0.04302019315188762,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.3056,
|
|
"mean_token_accuracy": 0.0939846970140934,
|
|
"num_tokens": 1491679.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.620209503173828,
|
|
"epoch": 0.04331284752707053,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.2829,
|
|
"mean_token_accuracy": 0.09510877132415771,
|
|
"num_tokens": 1501444.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.611186599731445,
|
|
"epoch": 0.043605501902253437,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.2859,
|
|
"mean_token_accuracy": 0.09353157803416252,
|
|
"num_tokens": 1511299.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.654992341995239,
|
|
"epoch": 0.043898156277436345,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.3184,
|
|
"mean_token_accuracy": 0.08825154900550843,
|
|
"num_tokens": 1522389.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.5400854587554935,
|
|
"epoch": 0.044190810652619254,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.2799,
|
|
"mean_token_accuracy": 0.08888843208551407,
|
|
"num_tokens": 1533493.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.710229063034058,
|
|
"epoch": 0.04448346502780216,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.3453,
|
|
"mean_token_accuracy": 0.08668449223041534,
|
|
"num_tokens": 1542668.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.633497714996338,
|
|
"epoch": 0.04477611940298507,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.2267,
|
|
"mean_token_accuracy": 0.09399316534399986,
|
|
"num_tokens": 1552221.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.639019966125488,
|
|
"epoch": 0.04506877377816798,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.3691,
|
|
"mean_token_accuracy": 0.08336025923490524,
|
|
"num_tokens": 1562807.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.568796062469483,
|
|
"epoch": 0.04536142815335089,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.3082,
|
|
"mean_token_accuracy": 0.08972151428461075,
|
|
"num_tokens": 1572751.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.611496067047119,
|
|
"epoch": 0.0456540825285338,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.3434,
|
|
"mean_token_accuracy": 0.08387525528669357,
|
|
"num_tokens": 1582802.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.618696594238282,
|
|
"epoch": 0.04594673690371671,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.2633,
|
|
"mean_token_accuracy": 0.0899326853454113,
|
|
"num_tokens": 1593780.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.53293023109436,
|
|
"epoch": 0.046239391278899616,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.3148,
|
|
"mean_token_accuracy": 0.09066571295261383,
|
|
"num_tokens": 1604386.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.6753880977630615,
|
|
"epoch": 0.046532045654082525,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.295,
|
|
"mean_token_accuracy": 0.0937283106148243,
|
|
"num_tokens": 1614359.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.544372034072876,
|
|
"epoch": 0.046824700029265434,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.176,
|
|
"mean_token_accuracy": 0.09263980463147163,
|
|
"num_tokens": 1624389.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.56023473739624,
|
|
"epoch": 0.04711735440444834,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.2833,
|
|
"mean_token_accuracy": 0.09009978026151658,
|
|
"num_tokens": 1634608.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.455692481994629,
|
|
"epoch": 0.04741000877963126,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.2045,
|
|
"mean_token_accuracy": 0.09845735877752304,
|
|
"num_tokens": 1645723.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.554573822021484,
|
|
"epoch": 0.04770266315481417,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.1817,
|
|
"mean_token_accuracy": 0.10012574419379235,
|
|
"num_tokens": 1655182.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.52402024269104,
|
|
"epoch": 0.04799531752999708,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.1881,
|
|
"mean_token_accuracy": 0.09166754521429539,
|
|
"num_tokens": 1664597.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.551714515686035,
|
|
"epoch": 0.048287971905179985,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.2815,
|
|
"mean_token_accuracy": 0.09629782885313035,
|
|
"num_tokens": 1674809.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.4392908096313475,
|
|
"epoch": 0.048580626280362894,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.1535,
|
|
"mean_token_accuracy": 0.09503219947218895,
|
|
"num_tokens": 1684082.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.506573343276978,
|
|
"epoch": 0.0488732806555458,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.2511,
|
|
"mean_token_accuracy": 0.09144405052065849,
|
|
"num_tokens": 1695547.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.713369989395142,
|
|
"epoch": 0.04916593503072871,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.2906,
|
|
"mean_token_accuracy": 0.097728531062603,
|
|
"num_tokens": 1706434.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.475108194351196,
|
|
"epoch": 0.04945858940591162,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.226,
|
|
"mean_token_accuracy": 0.0920279160141945,
|
|
"num_tokens": 1718063.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.463070058822632,
|
|
"epoch": 0.04975124378109453,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.2122,
|
|
"mean_token_accuracy": 0.08780092373490334,
|
|
"num_tokens": 1728594.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.555836248397827,
|
|
"epoch": 0.05004389815627744,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.2233,
|
|
"mean_token_accuracy": 0.09364095106720924,
|
|
"num_tokens": 1738879.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.437271451950073,
|
|
"epoch": 0.05033655253146035,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.1784,
|
|
"mean_token_accuracy": 0.09283483177423477,
|
|
"num_tokens": 1748850.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.491488265991211,
|
|
"epoch": 0.050629206906643257,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.23,
|
|
"mean_token_accuracy": 0.09758371859788895,
|
|
"num_tokens": 1759992.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.6118079662323,
|
|
"epoch": 0.050921861281826165,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.2323,
|
|
"mean_token_accuracy": 0.09015733227133751,
|
|
"num_tokens": 1769879.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.352298641204834,
|
|
"epoch": 0.051214515657009074,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.1161,
|
|
"mean_token_accuracy": 0.10056154727935791,
|
|
"num_tokens": 1779429.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.476632738113404,
|
|
"epoch": 0.05150717003219198,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.2185,
|
|
"mean_token_accuracy": 0.08883849456906319,
|
|
"num_tokens": 1788655.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.499739456176758,
|
|
"epoch": 0.05179982440737489,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.2176,
|
|
"mean_token_accuracy": 0.0874781295657158,
|
|
"num_tokens": 1798454.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.425588893890381,
|
|
"epoch": 0.0520924787825578,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.1901,
|
|
"mean_token_accuracy": 0.09141783937811851,
|
|
"num_tokens": 1809016.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.4348835945129395,
|
|
"epoch": 0.05238513315774071,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.0729,
|
|
"mean_token_accuracy": 0.10189558044075966,
|
|
"num_tokens": 1819992.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.461593580245972,
|
|
"epoch": 0.05267778753292362,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.1446,
|
|
"mean_token_accuracy": 0.0971047930419445,
|
|
"num_tokens": 1829932.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.391701364517212,
|
|
"epoch": 0.05297044190810653,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.106,
|
|
"mean_token_accuracy": 0.09629447236657143,
|
|
"num_tokens": 1839877.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.366179513931274,
|
|
"epoch": 0.053263096283289436,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.1228,
|
|
"mean_token_accuracy": 0.09160000830888748,
|
|
"num_tokens": 1849619.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.25721960067749,
|
|
"epoch": 0.053555750658472345,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.0871,
|
|
"mean_token_accuracy": 0.10284551829099656,
|
|
"num_tokens": 1859513.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.4424947738647464,
|
|
"epoch": 0.053848405033655254,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.0591,
|
|
"mean_token_accuracy": 0.09884816706180573,
|
|
"num_tokens": 1869849.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.435005235671997,
|
|
"epoch": 0.05414105940883816,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.174,
|
|
"mean_token_accuracy": 0.09173808693885803,
|
|
"num_tokens": 1879385.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.384939336776734,
|
|
"epoch": 0.05443371378402107,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.1714,
|
|
"mean_token_accuracy": 0.0962887905538082,
|
|
"num_tokens": 1889415.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.449399948120117,
|
|
"epoch": 0.05472636815920398,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.2205,
|
|
"mean_token_accuracy": 0.09307129457592964,
|
|
"num_tokens": 1901145.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.3746397495269775,
|
|
"epoch": 0.05501902253438689,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.0894,
|
|
"mean_token_accuracy": 0.10347672551870346,
|
|
"num_tokens": 1910355.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.481390714645386,
|
|
"epoch": 0.0553116769095698,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.1573,
|
|
"mean_token_accuracy": 0.09576819315552712,
|
|
"num_tokens": 1920526.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.3511247634887695,
|
|
"epoch": 0.05560433128475271,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.1049,
|
|
"mean_token_accuracy": 0.09833246991038322,
|
|
"num_tokens": 1929884.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.348301696777344,
|
|
"epoch": 0.055896985659935616,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.1044,
|
|
"mean_token_accuracy": 0.09435090646147729,
|
|
"num_tokens": 1940707.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.472948408126831,
|
|
"epoch": 0.056189640035118525,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.1739,
|
|
"mean_token_accuracy": 0.09039252474904061,
|
|
"num_tokens": 1951024.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.3457207679748535,
|
|
"epoch": 0.056482294410301434,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.0498,
|
|
"mean_token_accuracy": 0.09302077516913414,
|
|
"num_tokens": 1961553.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.3738847255706785,
|
|
"epoch": 0.05677494878548434,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1534,
|
|
"mean_token_accuracy": 0.09500665068626404,
|
|
"num_tokens": 1971297.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.282409811019898,
|
|
"epoch": 0.05706760316066725,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.0597,
|
|
"mean_token_accuracy": 0.09678901359438896,
|
|
"num_tokens": 1981753.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.471026945114136,
|
|
"epoch": 0.05736025753585016,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.1558,
|
|
"mean_token_accuracy": 0.10005593970417977,
|
|
"num_tokens": 1992590.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.350202465057373,
|
|
"epoch": 0.05765291191103307,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1311,
|
|
"mean_token_accuracy": 0.09674226641654968,
|
|
"num_tokens": 2001745.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.262063264846802,
|
|
"epoch": 0.05794556628621598,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.0455,
|
|
"mean_token_accuracy": 0.09895217716693878,
|
|
"num_tokens": 2011171.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.362540626525879,
|
|
"epoch": 0.05823822066139889,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.0997674971818924,
|
|
"num_tokens": 2020732.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.338001441955567,
|
|
"epoch": 0.058530875036581796,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.0761,
|
|
"mean_token_accuracy": 0.09528696537017822,
|
|
"num_tokens": 2030702.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.215022277832031,
|
|
"epoch": 0.058823529411764705,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 7.0855,
|
|
"mean_token_accuracy": 0.09225433394312858,
|
|
"num_tokens": 2040933.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.375273084640503,
|
|
"epoch": 0.059116183786947614,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 7.1447,
|
|
"mean_token_accuracy": 0.09505715742707252,
|
|
"num_tokens": 2051083.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.354562759399414,
|
|
"epoch": 0.05940883816213052,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 7.0474,
|
|
"mean_token_accuracy": 0.09751659147441387,
|
|
"num_tokens": 2060290.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.2843845844268795,
|
|
"epoch": 0.05970149253731343,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 7.0758,
|
|
"mean_token_accuracy": 0.10457687452435493,
|
|
"num_tokens": 2070537.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.208773374557495,
|
|
"epoch": 0.05999414691249634,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 7.0188,
|
|
"mean_token_accuracy": 0.09619513675570487,
|
|
"num_tokens": 2081020.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.309514284133911,
|
|
"epoch": 0.06028680128767925,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 7.0259,
|
|
"mean_token_accuracy": 0.10205215588212013,
|
|
"num_tokens": 2090731.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.4033431053161625,
|
|
"epoch": 0.06057945566286216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 7.1142,
|
|
"mean_token_accuracy": 0.09757147133350372,
|
|
"num_tokens": 2100248.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.219016265869141,
|
|
"epoch": 0.06087211003804507,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 7.0196,
|
|
"mean_token_accuracy": 0.09904577806591988,
|
|
"num_tokens": 2110518.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.363824844360352,
|
|
"epoch": 0.061164764413227976,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 7.0895,
|
|
"mean_token_accuracy": 0.0941918708384037,
|
|
"num_tokens": 2121424.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.209250974655151,
|
|
"epoch": 0.061457418788410885,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 7.0093,
|
|
"mean_token_accuracy": 0.1055798962712288,
|
|
"num_tokens": 2130904.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.318933916091919,
|
|
"epoch": 0.061750073163593794,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 7.0483,
|
|
"mean_token_accuracy": 0.10186817795038224,
|
|
"num_tokens": 2140710.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.373817396163941,
|
|
"epoch": 0.0620427275387767,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 7.1132,
|
|
"mean_token_accuracy": 0.0974596157670021,
|
|
"num_tokens": 2150767.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.216788959503174,
|
|
"epoch": 0.06233538191395961,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 7.0526,
|
|
"mean_token_accuracy": 0.09709591269493104,
|
|
"num_tokens": 2161414.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.242707395553589,
|
|
"epoch": 0.06262803628914253,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.9802,
|
|
"mean_token_accuracy": 0.10571711510419846,
|
|
"num_tokens": 2171960.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.240592432022095,
|
|
"epoch": 0.06292069066432543,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.9542,
|
|
"mean_token_accuracy": 0.09980396181344986,
|
|
"num_tokens": 2182344.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.144010734558106,
|
|
"epoch": 0.06321334503950835,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.9743,
|
|
"mean_token_accuracy": 0.09961154088377952,
|
|
"num_tokens": 2194829.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.1723168849945065,
|
|
"epoch": 0.06350599941469125,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.9419,
|
|
"mean_token_accuracy": 0.10114665776491165,
|
|
"num_tokens": 2204662.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.227386426925659,
|
|
"epoch": 0.06379865378987416,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.9674,
|
|
"mean_token_accuracy": 0.10895995870232582,
|
|
"num_tokens": 2213815.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.223431777954102,
|
|
"epoch": 0.06409130816505706,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.9819,
|
|
"mean_token_accuracy": 0.10199102833867073,
|
|
"num_tokens": 2223026.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.193548965454101,
|
|
"epoch": 0.06438396254023998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 7.0315,
|
|
"mean_token_accuracy": 0.09989764392375947,
|
|
"num_tokens": 2233575.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.221850919723511,
|
|
"epoch": 0.06467661691542288,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.9146,
|
|
"mean_token_accuracy": 0.10782215669751168,
|
|
"num_tokens": 2244171.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.2020580768585205,
|
|
"epoch": 0.0649692712906058,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.9815,
|
|
"mean_token_accuracy": 0.0972609356045723,
|
|
"num_tokens": 2253972.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.1520631313323975,
|
|
"epoch": 0.0652619256657887,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 7.0108,
|
|
"mean_token_accuracy": 0.10005398914217949,
|
|
"num_tokens": 2264699.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.216694593429565,
|
|
"epoch": 0.06555458004097162,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.9772,
|
|
"mean_token_accuracy": 0.10254745483398438,
|
|
"num_tokens": 2274894.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.193087959289551,
|
|
"epoch": 0.06584723441615452,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 7.0038,
|
|
"mean_token_accuracy": 0.09882058501243592,
|
|
"num_tokens": 2285541.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.2195906162261965,
|
|
"epoch": 0.06613988879133743,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.9395,
|
|
"mean_token_accuracy": 0.10300094708800316,
|
|
"num_tokens": 2295443.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.242094087600708,
|
|
"epoch": 0.06643254316652034,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 7.0233,
|
|
"mean_token_accuracy": 0.09846225753426552,
|
|
"num_tokens": 2306149.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.102969694137573,
|
|
"epoch": 0.06672519754170325,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.9776,
|
|
"mean_token_accuracy": 0.09972455576062203,
|
|
"num_tokens": 2316589.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.240374565124512,
|
|
"epoch": 0.06701785191688615,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.9679,
|
|
"mean_token_accuracy": 0.10801631659269333,
|
|
"num_tokens": 2326340.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.192791986465454,
|
|
"epoch": 0.06731050629206907,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.9389,
|
|
"mean_token_accuracy": 0.10559422895312309,
|
|
"num_tokens": 2336078.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.1362364292144775,
|
|
"epoch": 0.06760316066725197,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.8247,
|
|
"mean_token_accuracy": 0.10476716533303261,
|
|
"num_tokens": 2346017.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.143903064727783,
|
|
"epoch": 0.06789581504243489,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.9748,
|
|
"mean_token_accuracy": 0.09448844268918037,
|
|
"num_tokens": 2355970.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.164862585067749,
|
|
"epoch": 0.06818846941761779,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.9061,
|
|
"mean_token_accuracy": 0.09887387901544571,
|
|
"num_tokens": 2366010.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.173722934722901,
|
|
"epoch": 0.0684811237928007,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.9059,
|
|
"mean_token_accuracy": 0.10019839331507682,
|
|
"num_tokens": 2376431.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.145708847045898,
|
|
"epoch": 0.0687737781679836,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.9018,
|
|
"mean_token_accuracy": 0.10499030724167824,
|
|
"num_tokens": 2387027.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.116997337341308,
|
|
"epoch": 0.06906643254316652,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.9054,
|
|
"mean_token_accuracy": 0.10259901955723763,
|
|
"num_tokens": 2396746.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.146070146560669,
|
|
"epoch": 0.06935908691834942,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.917,
|
|
"mean_token_accuracy": 0.10224924832582474,
|
|
"num_tokens": 2406340.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.133272552490235,
|
|
"epoch": 0.06965174129353234,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.9518,
|
|
"mean_token_accuracy": 0.10214994624257087,
|
|
"num_tokens": 2417441.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.15489559173584,
|
|
"epoch": 0.06994439566871524,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.843,
|
|
"mean_token_accuracy": 0.10828633531928063,
|
|
"num_tokens": 2427365.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.020367860794067,
|
|
"epoch": 0.07023705004389816,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.8627,
|
|
"mean_token_accuracy": 0.10394505336880684,
|
|
"num_tokens": 2437236.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.1537285327911375,
|
|
"epoch": 0.07052970441908106,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.9642,
|
|
"mean_token_accuracy": 0.09915822148323059,
|
|
"num_tokens": 2448000.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.071624374389648,
|
|
"epoch": 0.07082235879426398,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.8886,
|
|
"mean_token_accuracy": 0.10841592401266098,
|
|
"num_tokens": 2458627.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.154271030426026,
|
|
"epoch": 0.07111501316944688,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.8187,
|
|
"mean_token_accuracy": 0.11104752346873284,
|
|
"num_tokens": 2468110.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.091701221466065,
|
|
"epoch": 0.0714076675446298,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.9753,
|
|
"mean_token_accuracy": 0.09984883964061737,
|
|
"num_tokens": 2479527.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.113129329681397,
|
|
"epoch": 0.0717003219198127,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.857,
|
|
"mean_token_accuracy": 0.11230505779385566,
|
|
"num_tokens": 2489856.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.067525911331177,
|
|
"epoch": 0.07199297629499561,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.9131,
|
|
"mean_token_accuracy": 0.10126614049077035,
|
|
"num_tokens": 2500164.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.203033304214477,
|
|
"epoch": 0.07228563067017851,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.9133,
|
|
"mean_token_accuracy": 0.10488039031624793,
|
|
"num_tokens": 2510535.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.032154607772827,
|
|
"epoch": 0.07257828504536143,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.8737,
|
|
"mean_token_accuracy": 0.10379046872258187,
|
|
"num_tokens": 2521334.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.104176473617554,
|
|
"epoch": 0.07287093942054433,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.8486,
|
|
"mean_token_accuracy": 0.10444203391671181,
|
|
"num_tokens": 2531990.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.063020420074463,
|
|
"epoch": 0.07316359379572725,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.9315,
|
|
"mean_token_accuracy": 0.10278217419981957,
|
|
"num_tokens": 2542418.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.165763187408447,
|
|
"epoch": 0.07345624817091015,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.9004,
|
|
"mean_token_accuracy": 0.10728387162089348,
|
|
"num_tokens": 2553944.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.088944149017334,
|
|
"epoch": 0.07374890254609306,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.8639,
|
|
"mean_token_accuracy": 0.10208542868494988,
|
|
"num_tokens": 2564577.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.134335517883301,
|
|
"epoch": 0.07404155692127597,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.9696,
|
|
"mean_token_accuracy": 0.09385665506124496,
|
|
"num_tokens": 2574003.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.100490999221802,
|
|
"epoch": 0.07433421129645888,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.8053,
|
|
"mean_token_accuracy": 0.10951183810830116,
|
|
"num_tokens": 2585367.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.030397129058838,
|
|
"epoch": 0.07462686567164178,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.7917,
|
|
"mean_token_accuracy": 0.10875893980264664,
|
|
"num_tokens": 2595282.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.015108489990235,
|
|
"epoch": 0.0749195200468247,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.7653,
|
|
"mean_token_accuracy": 0.11425828039646149,
|
|
"num_tokens": 2605220.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.989345264434815,
|
|
"epoch": 0.0752121744220076,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.8498,
|
|
"mean_token_accuracy": 0.10785233229398727,
|
|
"num_tokens": 2616492.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.082758522033691,
|
|
"epoch": 0.07550482879719052,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.8475,
|
|
"mean_token_accuracy": 0.10241246297955513,
|
|
"num_tokens": 2626133.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.978100776672363,
|
|
"epoch": 0.07579748317237343,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.8281,
|
|
"mean_token_accuracy": 0.10710375010967255,
|
|
"num_tokens": 2636661.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.118194580078125,
|
|
"epoch": 0.07609013754755634,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.8386,
|
|
"mean_token_accuracy": 0.10455987602472305,
|
|
"num_tokens": 2647598.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.000920104980469,
|
|
"epoch": 0.07638279192273925,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.7492,
|
|
"mean_token_accuracy": 0.10895679742097855,
|
|
"num_tokens": 2657267.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.016765689849853,
|
|
"epoch": 0.07667544629792215,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.8488,
|
|
"mean_token_accuracy": 0.11024482920765877,
|
|
"num_tokens": 2667048.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.150480651855469,
|
|
"epoch": 0.07696810067310507,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.8433,
|
|
"mean_token_accuracy": 0.09998077973723411,
|
|
"num_tokens": 2677386.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.942998266220092,
|
|
"epoch": 0.07726075504828797,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.8162,
|
|
"mean_token_accuracy": 0.10883545204997062,
|
|
"num_tokens": 2687655.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.089137840270996,
|
|
"epoch": 0.07755340942347089,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.8018,
|
|
"mean_token_accuracy": 0.11154846027493477,
|
|
"num_tokens": 2698098.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.039312267303467,
|
|
"epoch": 0.07784606379865379,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.7993,
|
|
"mean_token_accuracy": 0.10543053150177002,
|
|
"num_tokens": 2708119.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.011804723739624,
|
|
"epoch": 0.0781387181738367,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.7769,
|
|
"mean_token_accuracy": 0.11111046820878982,
|
|
"num_tokens": 2717036.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.052466917037964,
|
|
"epoch": 0.0784313725490196,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.9079,
|
|
"mean_token_accuracy": 0.10972103923559189,
|
|
"num_tokens": 2728290.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.033558797836304,
|
|
"epoch": 0.07872402692420252,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.8055,
|
|
"mean_token_accuracy": 0.1085546374320984,
|
|
"num_tokens": 2738373.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.0267448902130125,
|
|
"epoch": 0.07901668129938542,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.8276,
|
|
"mean_token_accuracy": 0.11423720940947532,
|
|
"num_tokens": 2748232.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.0601318359375,
|
|
"epoch": 0.07930933567456834,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.8273,
|
|
"mean_token_accuracy": 0.10952704623341561,
|
|
"num_tokens": 2758138.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.004944372177124,
|
|
"epoch": 0.07960199004975124,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.7868,
|
|
"mean_token_accuracy": 0.11433181166648865,
|
|
"num_tokens": 2768497.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.992530298233032,
|
|
"epoch": 0.07989464442493416,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.8227,
|
|
"mean_token_accuracy": 0.10943501815199852,
|
|
"num_tokens": 2778956.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.0625303268432615,
|
|
"epoch": 0.08018729880011706,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.7757,
|
|
"mean_token_accuracy": 0.11099240705370902,
|
|
"num_tokens": 2788295.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.017833948135376,
|
|
"epoch": 0.08047995317529998,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.7851,
|
|
"mean_token_accuracy": 0.10521746203303337,
|
|
"num_tokens": 2797736.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.936165142059326,
|
|
"epoch": 0.08077260755048288,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.6646,
|
|
"mean_token_accuracy": 0.11661427244544029,
|
|
"num_tokens": 2806933.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.9978080749511715,
|
|
"epoch": 0.0810652619256658,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.829,
|
|
"mean_token_accuracy": 0.10937268137931824,
|
|
"num_tokens": 2817471.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.003310823440552,
|
|
"epoch": 0.0813579163008487,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.9589,
|
|
"mean_token_accuracy": 0.10846946313977242,
|
|
"num_tokens": 2828016.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.050713396072387,
|
|
"epoch": 0.08165057067603161,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.7987,
|
|
"mean_token_accuracy": 0.10719454810023307,
|
|
"num_tokens": 2838011.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.951278257369995,
|
|
"epoch": 0.08194322505121451,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.68,
|
|
"mean_token_accuracy": 0.11497046947479247,
|
|
"num_tokens": 2848281.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.931306600570679,
|
|
"epoch": 0.08223587942639743,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.7977,
|
|
"mean_token_accuracy": 0.11288361549377442,
|
|
"num_tokens": 2859338.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.084950971603393,
|
|
"epoch": 0.08252853380158033,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.7972,
|
|
"mean_token_accuracy": 0.1125359758734703,
|
|
"num_tokens": 2869238.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.982119035720825,
|
|
"epoch": 0.08282118817676325,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.7834,
|
|
"mean_token_accuracy": 0.1123705931007862,
|
|
"num_tokens": 2879312.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.892581462860107,
|
|
"epoch": 0.08311384255194615,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.7231,
|
|
"mean_token_accuracy": 0.10808718279004097,
|
|
"num_tokens": 2888974.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.027051877975464,
|
|
"epoch": 0.08340649692712906,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.7598,
|
|
"mean_token_accuracy": 0.1084697999060154,
|
|
"num_tokens": 2899501.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.986096715927124,
|
|
"epoch": 0.08369915130231197,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.7968,
|
|
"mean_token_accuracy": 0.10647018775343894,
|
|
"num_tokens": 2909210.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.951934385299682,
|
|
"epoch": 0.08399180567749488,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.7246,
|
|
"mean_token_accuracy": 0.10852879583835602,
|
|
"num_tokens": 2919470.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.909590435028076,
|
|
"epoch": 0.08428446005267778,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.6972,
|
|
"mean_token_accuracy": 0.1139543168246746,
|
|
"num_tokens": 2929526.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.91973295211792,
|
|
"epoch": 0.0845771144278607,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.7107,
|
|
"mean_token_accuracy": 0.1111906498670578,
|
|
"num_tokens": 2938944.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.030971336364746,
|
|
"epoch": 0.0848697688030436,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.7415,
|
|
"mean_token_accuracy": 0.11075187176465988,
|
|
"num_tokens": 2948334.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.899282360076905,
|
|
"epoch": 0.08516242317822652,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.6948,
|
|
"mean_token_accuracy": 0.11553604677319526,
|
|
"num_tokens": 2957687.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.967754316329956,
|
|
"epoch": 0.08545507755340942,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.6961,
|
|
"mean_token_accuracy": 0.11643371209502221,
|
|
"num_tokens": 2968291.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.844262170791626,
|
|
"epoch": 0.08574773192859234,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.6833,
|
|
"mean_token_accuracy": 0.11676188111305237,
|
|
"num_tokens": 2978809.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.051819658279419,
|
|
"epoch": 0.08604038630377524,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.828,
|
|
"mean_token_accuracy": 0.10943237319588661,
|
|
"num_tokens": 2988839.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.964070463180542,
|
|
"epoch": 0.08633304067895815,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.7741,
|
|
"mean_token_accuracy": 0.11031899750232696,
|
|
"num_tokens": 2999380.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.931664991378784,
|
|
"epoch": 0.08662569505414106,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.6807,
|
|
"mean_token_accuracy": 0.11252299547195435,
|
|
"num_tokens": 3009298.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.891529750823975,
|
|
"epoch": 0.08691834942932397,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.6839,
|
|
"mean_token_accuracy": 0.11446837261319161,
|
|
"num_tokens": 3019747.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.922841024398804,
|
|
"epoch": 0.08721100380450687,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.6891,
|
|
"mean_token_accuracy": 0.11496640294790268,
|
|
"num_tokens": 3029777.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.863205766677856,
|
|
"epoch": 0.08750365817968979,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.8019,
|
|
"mean_token_accuracy": 0.10379430875182152,
|
|
"num_tokens": 3040184.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.04589695930481,
|
|
"epoch": 0.08779631255487269,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.6707,
|
|
"mean_token_accuracy": 0.1120702162384987,
|
|
"num_tokens": 3049183.0,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 673794932244480.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|