d2120b5eda
Model: fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
4057 lines
109 KiB
JSON
4057 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.9885629040278467,
|
|
"eval_steps": 1000,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 8.122408580780029,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.8555,
|
|
"mean_token_accuracy": 0.08926185071468354,
|
|
"num_tokens": 9311.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 8.126056098937989,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.9328,
|
|
"mean_token_accuracy": 0.08018440753221512,
|
|
"num_tokens": 20562.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 8.05552430152893,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.898,
|
|
"mean_token_accuracy": 0.0867466501891613,
|
|
"num_tokens": 31490.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 8.07232894897461,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.7975,
|
|
"mean_token_accuracy": 0.09018379971385002,
|
|
"num_tokens": 40741.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 8.081013774871826,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.7822,
|
|
"mean_token_accuracy": 0.09092294126749038,
|
|
"num_tokens": 50749.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.060551881790161,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.8658,
|
|
"mean_token_accuracy": 0.09334692656993866,
|
|
"num_tokens": 60296.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 8.036958074569702,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.8272,
|
|
"mean_token_accuracy": 0.0933995120227337,
|
|
"num_tokens": 70797.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 8.074491786956788,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.8404,
|
|
"mean_token_accuracy": 0.08989858850836754,
|
|
"num_tokens": 80815.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 8.017825746536255,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.813,
|
|
"mean_token_accuracy": 0.0953871764242649,
|
|
"num_tokens": 90326.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 8.046491861343384,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.8112,
|
|
"mean_token_accuracy": 0.08890505433082581,
|
|
"num_tokens": 99800.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 8.046163415908813,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.921,
|
|
"mean_token_accuracy": 0.09161711558699608,
|
|
"num_tokens": 110636.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 8.105820083618164,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.8157,
|
|
"mean_token_accuracy": 0.09141169264912605,
|
|
"num_tokens": 120269.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 8.087437438964844,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.8186,
|
|
"mean_token_accuracy": 0.08604214638471604,
|
|
"num_tokens": 130176.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 8.022943496704102,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.7724,
|
|
"mean_token_accuracy": 0.09266669675707817,
|
|
"num_tokens": 140020.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 8.029492521286011,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.8566,
|
|
"mean_token_accuracy": 0.09054757431149482,
|
|
"num_tokens": 150713.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 8.057504415512085,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.7773,
|
|
"mean_token_accuracy": 0.08817628473043441,
|
|
"num_tokens": 159828.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 8.043695306777954,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.808,
|
|
"mean_token_accuracy": 0.0962263211607933,
|
|
"num_tokens": 169909.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 8.03000283241272,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.8599,
|
|
"mean_token_accuracy": 0.0911699652671814,
|
|
"num_tokens": 179462.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 8.082845211029053,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.8531,
|
|
"mean_token_accuracy": 0.08622325584292412,
|
|
"num_tokens": 188965.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 8.165623950958253,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.8427,
|
|
"mean_token_accuracy": 0.08928864933550358,
|
|
"num_tokens": 200382.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 8.097312736511231,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.7246,
|
|
"mean_token_accuracy": 0.09705447182059287,
|
|
"num_tokens": 209718.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.990663337707519,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.724,
|
|
"mean_token_accuracy": 0.09378693476319314,
|
|
"num_tokens": 219216.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 8.002191734313964,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.7805,
|
|
"mean_token_accuracy": 0.0973575845360756,
|
|
"num_tokens": 228219.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 8.117930793762207,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.7441,
|
|
"mean_token_accuracy": 0.08997285291552544,
|
|
"num_tokens": 237776.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.958203458786011,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.7335,
|
|
"mean_token_accuracy": 0.09299578815698624,
|
|
"num_tokens": 248411.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 8.043806886672973,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8195,
|
|
"mean_token_accuracy": 0.08543559387326241,
|
|
"num_tokens": 258797.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 8.126793479919433,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7469,
|
|
"mean_token_accuracy": 0.09252887964248657,
|
|
"num_tokens": 269228.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.956741619110107,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8359,
|
|
"mean_token_accuracy": 0.09155542254447938,
|
|
"num_tokens": 279191.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.055004549026489,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7325,
|
|
"mean_token_accuracy": 0.09122926145792007,
|
|
"num_tokens": 289382.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.072270679473878,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6913,
|
|
"mean_token_accuracy": 0.09444988965988159,
|
|
"num_tokens": 299163.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.007786178588868,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7736,
|
|
"mean_token_accuracy": 0.08923879414796829,
|
|
"num_tokens": 308551.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.080718994140625,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.736,
|
|
"mean_token_accuracy": 0.08449316844344139,
|
|
"num_tokens": 318427.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.999086856842041,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8093,
|
|
"mean_token_accuracy": 0.08699049055576324,
|
|
"num_tokens": 327539.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.11481409072876,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7423,
|
|
"mean_token_accuracy": 0.09079778492450714,
|
|
"num_tokens": 336676.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.016329479217529,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.738,
|
|
"mean_token_accuracy": 0.10277078524231911,
|
|
"num_tokens": 345914.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.019076776504516,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7036,
|
|
"mean_token_accuracy": 0.09528676047921181,
|
|
"num_tokens": 355982.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.040568590164185,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7516,
|
|
"mean_token_accuracy": 0.08857373371720315,
|
|
"num_tokens": 365398.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.110048770904541,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.755,
|
|
"mean_token_accuracy": 0.0934124194085598,
|
|
"num_tokens": 374519.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.9319648265838625,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.6904,
|
|
"mean_token_accuracy": 0.09402457699179649,
|
|
"num_tokens": 383536.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.176937437057495,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.77,
|
|
"mean_token_accuracy": 0.0955814115703106,
|
|
"num_tokens": 393703.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.964634227752685,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.8693,
|
|
"mean_token_accuracy": 0.08404051922261715,
|
|
"num_tokens": 404065.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.064677190780639,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7777,
|
|
"mean_token_accuracy": 0.0903236448764801,
|
|
"num_tokens": 413869.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.066971015930175,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.7805,
|
|
"mean_token_accuracy": 0.0898888312280178,
|
|
"num_tokens": 424251.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.01867208480835,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.7275,
|
|
"mean_token_accuracy": 0.09023677036166192,
|
|
"num_tokens": 432530.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.00871753692627,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.6977,
|
|
"mean_token_accuracy": 0.09342340901494026,
|
|
"num_tokens": 442335.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.094160842895509,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.7204,
|
|
"mean_token_accuracy": 0.09416243061423302,
|
|
"num_tokens": 452048.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.901502513885498,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.646,
|
|
"mean_token_accuracy": 0.09958525598049164,
|
|
"num_tokens": 460840.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.078534698486328,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.7572,
|
|
"mean_token_accuracy": 0.09259009286761284,
|
|
"num_tokens": 470198.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.032262182235717,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.7012,
|
|
"mean_token_accuracy": 0.09810718074440956,
|
|
"num_tokens": 481030.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.884334325790405,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.6662,
|
|
"mean_token_accuracy": 0.0929951898753643,
|
|
"num_tokens": 490411.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.049663257598876,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.681,
|
|
"mean_token_accuracy": 0.09713885933160782,
|
|
"num_tokens": 499551.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.01776294708252,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.7379,
|
|
"mean_token_accuracy": 0.08942644819617271,
|
|
"num_tokens": 508885.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.009857130050658,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7463,
|
|
"mean_token_accuracy": 0.08858433440327644,
|
|
"num_tokens": 518291.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.004399061203003,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.6323,
|
|
"mean_token_accuracy": 0.0969871662557125,
|
|
"num_tokens": 528035.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.935997676849365,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6993,
|
|
"mean_token_accuracy": 0.09522532522678376,
|
|
"num_tokens": 538410.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.997100067138672,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.6389,
|
|
"mean_token_accuracy": 0.09725429192185402,
|
|
"num_tokens": 549697.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.946566820144653,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.5314,
|
|
"mean_token_accuracy": 0.10031294599175453,
|
|
"num_tokens": 558932.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.803422832489014,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6543,
|
|
"mean_token_accuracy": 0.0901026301085949,
|
|
"num_tokens": 568441.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.110786485671998,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.755,
|
|
"mean_token_accuracy": 0.09423976689577103,
|
|
"num_tokens": 577861.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.907559442520141,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.731,
|
|
"mean_token_accuracy": 0.09102459624409676,
|
|
"num_tokens": 587858.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.018079805374146,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.7284,
|
|
"mean_token_accuracy": 0.09751355946063996,
|
|
"num_tokens": 597806.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.010554122924805,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.6704,
|
|
"mean_token_accuracy": 0.10095374062657356,
|
|
"num_tokens": 607408.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.945669937133789,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 2.875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6148,
|
|
"mean_token_accuracy": 0.09937258884310722,
|
|
"num_tokens": 617862.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.973281812667847,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.6229,
|
|
"mean_token_accuracy": 0.09138860106468201,
|
|
"num_tokens": 627128.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.833469247817993,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.627,
|
|
"mean_token_accuracy": 0.09354910627007484,
|
|
"num_tokens": 636775.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.112251043319702,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6205,
|
|
"mean_token_accuracy": 0.09625474289059639,
|
|
"num_tokens": 645906.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.940666913986206,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.5624,
|
|
"mean_token_accuracy": 0.09025543108582497,
|
|
"num_tokens": 655601.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.98874797821045,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.6814,
|
|
"mean_token_accuracy": 0.09709643945097923,
|
|
"num_tokens": 665682.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.868845844268799,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.6302,
|
|
"mean_token_accuracy": 0.09729701802134513,
|
|
"num_tokens": 675200.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.9250412940979,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.6475,
|
|
"mean_token_accuracy": 0.09828538969159126,
|
|
"num_tokens": 685590.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.931744909286499,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.6064,
|
|
"mean_token_accuracy": 0.10055591687560081,
|
|
"num_tokens": 695665.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.942970275878906,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.664,
|
|
"mean_token_accuracy": 0.09400481581687928,
|
|
"num_tokens": 705850.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.049799823760987,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.6738,
|
|
"mean_token_accuracy": 0.0901768684387207,
|
|
"num_tokens": 716010.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.923341178894043,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.6223,
|
|
"mean_token_accuracy": 0.09590414240956306,
|
|
"num_tokens": 724908.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.7986541271209715,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.5942,
|
|
"mean_token_accuracy": 0.09819011092185974,
|
|
"num_tokens": 733479.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.961906242370605,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.6145,
|
|
"mean_token_accuracy": 0.09582870304584504,
|
|
"num_tokens": 743365.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.91075963973999,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.6664,
|
|
"mean_token_accuracy": 0.09616638831794262,
|
|
"num_tokens": 753720.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.916628408432007,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5311,
|
|
"mean_token_accuracy": 0.0964023895561695,
|
|
"num_tokens": 762524.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.922145891189575,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.6111,
|
|
"mean_token_accuracy": 0.09766499325633049,
|
|
"num_tokens": 772549.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.79258098602295,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.6114,
|
|
"mean_token_accuracy": 0.09619558006525039,
|
|
"num_tokens": 782304.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.05925145149231,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.7212,
|
|
"mean_token_accuracy": 0.09696372747421264,
|
|
"num_tokens": 792010.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.807120895385742,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5492,
|
|
"mean_token_accuracy": 0.10038653984665871,
|
|
"num_tokens": 801667.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.899970769882202,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.6252,
|
|
"mean_token_accuracy": 0.09784654006361962,
|
|
"num_tokens": 812385.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.8692365169525145,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.535,
|
|
"mean_token_accuracy": 0.10144751518964767,
|
|
"num_tokens": 822177.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.8849406242370605,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.5996,
|
|
"mean_token_accuracy": 0.09166432395577431,
|
|
"num_tokens": 832576.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.780496835708618,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.566,
|
|
"mean_token_accuracy": 0.09900226145982742,
|
|
"num_tokens": 842290.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.872547626495361,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.585,
|
|
"mean_token_accuracy": 0.09592773094773292,
|
|
"num_tokens": 852500.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.809439420700073,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.4945,
|
|
"mean_token_accuracy": 0.09826857671141624,
|
|
"num_tokens": 861401.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.860213375091552,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.5047,
|
|
"mean_token_accuracy": 0.10223424881696701,
|
|
"num_tokens": 871523.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.898686456680298,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.605,
|
|
"mean_token_accuracy": 0.090828076004982,
|
|
"num_tokens": 881384.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.867025518417359,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.6307,
|
|
"mean_token_accuracy": 0.09517297595739364,
|
|
"num_tokens": 890511.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.833306884765625,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.563,
|
|
"mean_token_accuracy": 0.09683875888586044,
|
|
"num_tokens": 900064.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.857913303375244,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.6161,
|
|
"mean_token_accuracy": 0.09450365006923675,
|
|
"num_tokens": 910892.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.899250507354736,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.6154,
|
|
"mean_token_accuracy": 0.09202362224459648,
|
|
"num_tokens": 920566.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.7721343517303465,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.5319,
|
|
"mean_token_accuracy": 0.09768470898270606,
|
|
"num_tokens": 930609.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.801538562774658,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.565,
|
|
"mean_token_accuracy": 0.09313339218497277,
|
|
"num_tokens": 940383.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.941165828704834,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 2.6875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.5122,
|
|
"mean_token_accuracy": 0.09608981050550938,
|
|
"num_tokens": 951302.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.788536548614502,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.4818,
|
|
"mean_token_accuracy": 0.10193300023674964,
|
|
"num_tokens": 961126.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.839777183532715,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.5032,
|
|
"mean_token_accuracy": 0.09682125523686409,
|
|
"num_tokens": 970845.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.721405076980591,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.427,
|
|
"mean_token_accuracy": 0.10273272842168808,
|
|
"num_tokens": 980919.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.719191455841065,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.4516,
|
|
"mean_token_accuracy": 0.1017610639333725,
|
|
"num_tokens": 990659.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.737829732894897,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.514,
|
|
"mean_token_accuracy": 0.1017485462129116,
|
|
"num_tokens": 999540.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.7914680480957035,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.5122,
|
|
"mean_token_accuracy": 0.09308664873242378,
|
|
"num_tokens": 1008688.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.721870851516724,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.4086,
|
|
"mean_token_accuracy": 0.1110854797065258,
|
|
"num_tokens": 1018310.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.716399812698365,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.5061,
|
|
"mean_token_accuracy": 0.0994237333536148,
|
|
"num_tokens": 1027973.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.801226186752319,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.4382,
|
|
"mean_token_accuracy": 0.1021985501050949,
|
|
"num_tokens": 1038141.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.658663749694824,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.5474,
|
|
"mean_token_accuracy": 0.09069671258330345,
|
|
"num_tokens": 1047932.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.632207870483398,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.3242,
|
|
"mean_token_accuracy": 0.10905587524175644,
|
|
"num_tokens": 1056826.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.736492490768432,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.4832,
|
|
"mean_token_accuracy": 0.10051515996456147,
|
|
"num_tokens": 1066025.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.75524320602417,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.4987,
|
|
"mean_token_accuracy": 0.09955808669328689,
|
|
"num_tokens": 1075452.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.787500476837158,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5126,
|
|
"mean_token_accuracy": 0.09939858093857765,
|
|
"num_tokens": 1085300.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.606941556930542,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.3578,
|
|
"mean_token_accuracy": 0.10578031092882156,
|
|
"num_tokens": 1094548.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.804773807525635,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.4029,
|
|
"mean_token_accuracy": 0.09974669739603996,
|
|
"num_tokens": 1104321.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.712262105941773,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.544,
|
|
"mean_token_accuracy": 0.0965195819735527,
|
|
"num_tokens": 1115047.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.739543342590332,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.5043,
|
|
"mean_token_accuracy": 0.10095428079366683,
|
|
"num_tokens": 1124757.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.72175817489624,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.4489,
|
|
"mean_token_accuracy": 0.09865402802824974,
|
|
"num_tokens": 1134255.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.810949993133545,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.5124,
|
|
"mean_token_accuracy": 0.09687942638993263,
|
|
"num_tokens": 1143117.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.622174072265625,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.3751,
|
|
"mean_token_accuracy": 0.10171565413475037,
|
|
"num_tokens": 1153203.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.735700798034668,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.416,
|
|
"mean_token_accuracy": 0.10218172967433929,
|
|
"num_tokens": 1162839.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.567971992492676,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.3696,
|
|
"mean_token_accuracy": 0.1032779723405838,
|
|
"num_tokens": 1172580.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.7000068664550785,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.3688,
|
|
"mean_token_accuracy": 0.10539614930748939,
|
|
"num_tokens": 1183715.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.688623571395874,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.5444,
|
|
"mean_token_accuracy": 0.09419808611273765,
|
|
"num_tokens": 1194217.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.695876932144165,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.4023,
|
|
"mean_token_accuracy": 0.10362396463751793,
|
|
"num_tokens": 1203297.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.752481842041016,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.5044,
|
|
"mean_token_accuracy": 0.09252066910266876,
|
|
"num_tokens": 1213394.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.600792169570923,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.485,
|
|
"mean_token_accuracy": 0.10286026149988174,
|
|
"num_tokens": 1222771.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.684136581420899,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.4382,
|
|
"mean_token_accuracy": 0.09525436088442803,
|
|
"num_tokens": 1232171.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.607944822311401,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.3756,
|
|
"mean_token_accuracy": 0.1039860226213932,
|
|
"num_tokens": 1241640.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.603791999816894,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.3979,
|
|
"mean_token_accuracy": 0.10446851849555969,
|
|
"num_tokens": 1251785.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.632256364822387,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.3945,
|
|
"mean_token_accuracy": 0.09782340601086617,
|
|
"num_tokens": 1262327.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.627688646316528,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.3077,
|
|
"mean_token_accuracy": 0.10489342957735062,
|
|
"num_tokens": 1270871.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.533834218978882,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.3627,
|
|
"mean_token_accuracy": 0.10482602864503861,
|
|
"num_tokens": 1281563.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.566735506057739,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.3783,
|
|
"mean_token_accuracy": 0.09988497495651245,
|
|
"num_tokens": 1290867.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.6221600532531735,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.3859,
|
|
"mean_token_accuracy": 0.10277852416038513,
|
|
"num_tokens": 1300384.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.620599699020386,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.3065,
|
|
"mean_token_accuracy": 0.10421342998743058,
|
|
"num_tokens": 1310443.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.551718235015869,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.2883,
|
|
"mean_token_accuracy": 0.10518570244312286,
|
|
"num_tokens": 1320345.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.555121326446534,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.4677,
|
|
"mean_token_accuracy": 0.09977790042757988,
|
|
"num_tokens": 1330868.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.603720664978027,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.3716,
|
|
"mean_token_accuracy": 0.10149549767374992,
|
|
"num_tokens": 1340605.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.610010147094727,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.3887,
|
|
"mean_token_accuracy": 0.1037661112844944,
|
|
"num_tokens": 1350248.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.414504098892212,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.3681,
|
|
"mean_token_accuracy": 0.10223357081413269,
|
|
"num_tokens": 1360251.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.644025135040283,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.295,
|
|
"mean_token_accuracy": 0.10866039916872978,
|
|
"num_tokens": 1370156.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.5345031261444095,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4874,
|
|
"mean_token_accuracy": 0.09497825279831887,
|
|
"num_tokens": 1380346.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.57585654258728,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.3459,
|
|
"mean_token_accuracy": 0.10090088471770287,
|
|
"num_tokens": 1390958.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.568233871459961,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.4193,
|
|
"mean_token_accuracy": 0.10254513472318649,
|
|
"num_tokens": 1400324.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.632588481903076,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.3289,
|
|
"mean_token_accuracy": 0.10863603204488755,
|
|
"num_tokens": 1410272.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.395322036743164,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.3459,
|
|
"mean_token_accuracy": 0.1074515551328659,
|
|
"num_tokens": 1420061.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.52363076210022,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.3027,
|
|
"mean_token_accuracy": 0.10952795967459679,
|
|
"num_tokens": 1428771.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.477725315093994,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.3102,
|
|
"mean_token_accuracy": 0.1064012736082077,
|
|
"num_tokens": 1438154.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.440126705169678,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.2848,
|
|
"mean_token_accuracy": 0.10690619945526122,
|
|
"num_tokens": 1447403.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.531488990783691,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.3787,
|
|
"mean_token_accuracy": 0.09951569363474846,
|
|
"num_tokens": 1457443.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.509539699554443,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.3605,
|
|
"mean_token_accuracy": 0.10302009806036949,
|
|
"num_tokens": 1466783.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.600085258483887,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3581,
|
|
"mean_token_accuracy": 0.10676125213503837,
|
|
"num_tokens": 1476320.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.455271196365357,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.2729,
|
|
"mean_token_accuracy": 0.10698438361287117,
|
|
"num_tokens": 1487581.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.526057958602905,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.4161,
|
|
"mean_token_accuracy": 0.0985928162932396,
|
|
"num_tokens": 1496455.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.511812591552735,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.348,
|
|
"mean_token_accuracy": 0.10783538296818733,
|
|
"num_tokens": 1506422.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.514915466308594,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.3114,
|
|
"mean_token_accuracy": 0.10534781441092492,
|
|
"num_tokens": 1516593.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.418980598449707,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.3705,
|
|
"mean_token_accuracy": 0.10273290202021598,
|
|
"num_tokens": 1526011.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.522382354736328,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.2848,
|
|
"mean_token_accuracy": 0.1055482119321823,
|
|
"num_tokens": 1536018.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.341956090927124,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.224,
|
|
"mean_token_accuracy": 0.10807310789823532,
|
|
"num_tokens": 1545568.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.458911895751953,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.2246,
|
|
"mean_token_accuracy": 0.10711941570043564,
|
|
"num_tokens": 1556389.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.578403520584106,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.3641,
|
|
"mean_token_accuracy": 0.10444995686411858,
|
|
"num_tokens": 1565587.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.33792815208435,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.2441,
|
|
"mean_token_accuracy": 0.10755613222718238,
|
|
"num_tokens": 1574911.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.365066766738892,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.2242,
|
|
"mean_token_accuracy": 0.11039689257740974,
|
|
"num_tokens": 1584477.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.408384799957275,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.2922,
|
|
"mean_token_accuracy": 0.1128959059715271,
|
|
"num_tokens": 1594408.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.360699558258057,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.247,
|
|
"mean_token_accuracy": 0.10692444741725922,
|
|
"num_tokens": 1603494.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.462933492660523,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.35,
|
|
"mean_token_accuracy": 0.10176746174693108,
|
|
"num_tokens": 1612994.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.455133867263794,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.2032,
|
|
"mean_token_accuracy": 0.10662396997213364,
|
|
"num_tokens": 1622150.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.382295083999634,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.2319,
|
|
"mean_token_accuracy": 0.1111760102212429,
|
|
"num_tokens": 1631040.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.31751708984375,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.192,
|
|
"mean_token_accuracy": 0.11513797119259835,
|
|
"num_tokens": 1641519.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.354842281341552,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.2668,
|
|
"mean_token_accuracy": 0.10789724364876747,
|
|
"num_tokens": 1652218.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.601192665100098,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.3301,
|
|
"mean_token_accuracy": 0.09921003803610802,
|
|
"num_tokens": 1663763.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.362071323394775,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.2763,
|
|
"mean_token_accuracy": 0.10137885957956314,
|
|
"num_tokens": 1674160.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.43802547454834,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.4024,
|
|
"mean_token_accuracy": 0.10859992802143097,
|
|
"num_tokens": 1685138.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.443594169616699,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.2993,
|
|
"mean_token_accuracy": 0.10044958740472794,
|
|
"num_tokens": 1694847.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.433558607101441,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.2772,
|
|
"mean_token_accuracy": 0.10592831820249557,
|
|
"num_tokens": 1704503.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.368968296051025,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.2668,
|
|
"mean_token_accuracy": 0.10812032371759414,
|
|
"num_tokens": 1714011.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.3641283988952635,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.2193,
|
|
"mean_token_accuracy": 0.10835571512579918,
|
|
"num_tokens": 1724254.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.2866339683532715,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.2739,
|
|
"mean_token_accuracy": 0.10235930308699608,
|
|
"num_tokens": 1734696.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.365292119979858,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.2165,
|
|
"mean_token_accuracy": 0.11037806496024131,
|
|
"num_tokens": 1743650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.375092315673828,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.175,
|
|
"mean_token_accuracy": 0.11102150678634644,
|
|
"num_tokens": 1753449.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.370843553543091,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.322,
|
|
"mean_token_accuracy": 0.10787996426224708,
|
|
"num_tokens": 1764104.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.183559894561768,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.0971,
|
|
"mean_token_accuracy": 0.11445347592234612,
|
|
"num_tokens": 1773172.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.315752792358398,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.1655,
|
|
"mean_token_accuracy": 0.1090462140738964,
|
|
"num_tokens": 1783268.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.2884824752807615,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.2384,
|
|
"mean_token_accuracy": 0.1040517307817936,
|
|
"num_tokens": 1792169.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.355252885818482,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.2273,
|
|
"mean_token_accuracy": 0.10773444548249245,
|
|
"num_tokens": 1802280.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.304852056503296,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.2212,
|
|
"mean_token_accuracy": 0.10946610048413277,
|
|
"num_tokens": 1812086.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.223779201507568,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.2053,
|
|
"mean_token_accuracy": 0.11222834661602973,
|
|
"num_tokens": 1821554.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.391703653335571,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.2522,
|
|
"mean_token_accuracy": 0.1073651947081089,
|
|
"num_tokens": 1831088.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.282382965087891,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.2627,
|
|
"mean_token_accuracy": 0.11030717194080353,
|
|
"num_tokens": 1840606.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.240512847900391,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.1708,
|
|
"mean_token_accuracy": 0.11626142486929894,
|
|
"num_tokens": 1850343.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.2862786769866945,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.2798,
|
|
"mean_token_accuracy": 0.10937313660979271,
|
|
"num_tokens": 1860727.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.274987268447876,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.1902,
|
|
"mean_token_accuracy": 0.10906597152352333,
|
|
"num_tokens": 1870703.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.29812650680542,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.2196,
|
|
"mean_token_accuracy": 0.10876635760068894,
|
|
"num_tokens": 1880694.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.354408121109008,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.163,
|
|
"mean_token_accuracy": 0.1110567644238472,
|
|
"num_tokens": 1890195.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.202347040176392,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1253,
|
|
"mean_token_accuracy": 0.11104833781719207,
|
|
"num_tokens": 1900221.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.143579006195068,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.1785,
|
|
"mean_token_accuracy": 0.11059431210160256,
|
|
"num_tokens": 1910122.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.353754472732544,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.1831,
|
|
"mean_token_accuracy": 0.11084778457880021,
|
|
"num_tokens": 1918814.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.168116044998169,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1749,
|
|
"mean_token_accuracy": 0.11735359057784081,
|
|
"num_tokens": 1928446.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.272993516921997,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.1308,
|
|
"mean_token_accuracy": 0.1103288859128952,
|
|
"num_tokens": 1939060.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.2277168273925785,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.1353,
|
|
"mean_token_accuracy": 0.11369294598698616,
|
|
"num_tokens": 1948809.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.233875274658203,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.1353,
|
|
"mean_token_accuracy": 0.11033130958676338,
|
|
"num_tokens": 1959232.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.227099574655981,
|
|
"eval_loss": 7.268585205078125,
|
|
"eval_mean_token_accuracy": 0.10413067842820822,
|
|
"eval_num_tokens": 1959232.0,
|
|
"eval_runtime": 2.5901,
|
|
"eval_samples_per_second": 1337.795,
|
|
"eval_steps_per_second": 167.562,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.177479839324951,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.0681,
|
|
"mean_token_accuracy": 0.11539435237646103,
|
|
"num_tokens": 1969117.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.297568957010905,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 7.1097,
|
|
"mean_token_accuracy": 0.11202286763323678,
|
|
"num_tokens": 1978202.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.039242362976074,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 6.8775,
|
|
"mean_token_accuracy": 0.12471100017428398,
|
|
"num_tokens": 1988085.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.2135011672973635,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 6.8566,
|
|
"mean_token_accuracy": 0.12106488719582557,
|
|
"num_tokens": 1996977.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.1463054656982425,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 6.9291,
|
|
"mean_token_accuracy": 0.11656595915555953,
|
|
"num_tokens": 2006622.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.221590375900268,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 7.0479,
|
|
"mean_token_accuracy": 0.11268379613757133,
|
|
"num_tokens": 2016390.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.181693983078003,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 6.8967,
|
|
"mean_token_accuracy": 0.11597555577754974,
|
|
"num_tokens": 2026439.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.143760633468628,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 6.9743,
|
|
"mean_token_accuracy": 0.12400304973125457,
|
|
"num_tokens": 2035611.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.12859525680542,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 6.9878,
|
|
"mean_token_accuracy": 0.11458344608545304,
|
|
"num_tokens": 2047166.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.1092852592468265,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 6.9031,
|
|
"mean_token_accuracy": 0.11905743703246116,
|
|
"num_tokens": 2057216.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.121430969238281,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 6.9688,
|
|
"mean_token_accuracy": 0.11538073271512986,
|
|
"num_tokens": 2067339.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.104109287261963,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 6.851,
|
|
"mean_token_accuracy": 0.11959391385316849,
|
|
"num_tokens": 2077815.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.016418695449829,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 6.9525,
|
|
"mean_token_accuracy": 0.11269129067659378,
|
|
"num_tokens": 2087738.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.061359930038452,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 6.885,
|
|
"mean_token_accuracy": 0.11644371598958969,
|
|
"num_tokens": 2097246.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.8907383441925045,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 6.8094,
|
|
"mean_token_accuracy": 0.12704239040613174,
|
|
"num_tokens": 2106415.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.081757068634033,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 6.7604,
|
|
"mean_token_accuracy": 0.12543603628873826,
|
|
"num_tokens": 2115950.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.131565284729004,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 7.0008,
|
|
"mean_token_accuracy": 0.12020133659243584,
|
|
"num_tokens": 2127896.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.041161298751831,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 6.9288,
|
|
"mean_token_accuracy": 0.11448188871145248,
|
|
"num_tokens": 2137238.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.940253067016601,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 3.796875,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 6.9006,
|
|
"mean_token_accuracy": 0.1144180566072464,
|
|
"num_tokens": 2147107.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.133631563186645,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 6.9704,
|
|
"mean_token_accuracy": 0.11777857169508935,
|
|
"num_tokens": 2157128.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.987091588973999,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 6.9113,
|
|
"mean_token_accuracy": 0.11355555281043053,
|
|
"num_tokens": 2167981.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.0585156917572025,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 6.8561,
|
|
"mean_token_accuracy": 0.11993846967816353,
|
|
"num_tokens": 2177659.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.066860008239746,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 6.8788,
|
|
"mean_token_accuracy": 0.1190800815820694,
|
|
"num_tokens": 2186936.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.104629707336426,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 6.9139,
|
|
"mean_token_accuracy": 0.12040532752871513,
|
|
"num_tokens": 2197126.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.996875047683716,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 6.9362,
|
|
"mean_token_accuracy": 0.11848222240805625,
|
|
"num_tokens": 2206337.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.024795484542847,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 6.8698,
|
|
"mean_token_accuracy": 0.115912526845932,
|
|
"num_tokens": 2217112.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.095630693435669,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 6.9678,
|
|
"mean_token_accuracy": 0.11383809894323349,
|
|
"num_tokens": 2226239.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.096014213562012,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 6.9894,
|
|
"mean_token_accuracy": 0.11798062771558762,
|
|
"num_tokens": 2235702.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.03745436668396,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 6.9407,
|
|
"mean_token_accuracy": 0.12207928895950318,
|
|
"num_tokens": 2246005.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.008592653274536,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 6.903,
|
|
"mean_token_accuracy": 0.12116350829601288,
|
|
"num_tokens": 2255534.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.113993358612061,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 6.977,
|
|
"mean_token_accuracy": 0.1165345698595047,
|
|
"num_tokens": 2266374.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.927402591705322,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 6.7571,
|
|
"mean_token_accuracy": 0.12876896783709527,
|
|
"num_tokens": 2275614.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.9787169933319095,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 6.8386,
|
|
"mean_token_accuracy": 0.13057343736290933,
|
|
"num_tokens": 2284403.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.033973646163941,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 6.9209,
|
|
"mean_token_accuracy": 0.11778316348791122,
|
|
"num_tokens": 2294153.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.958228731155396,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 6.8837,
|
|
"mean_token_accuracy": 0.12165310680866241,
|
|
"num_tokens": 2302897.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.121665811538696,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 6.9827,
|
|
"mean_token_accuracy": 0.11410253793001175,
|
|
"num_tokens": 2312483.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.747711133956909,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 6.779,
|
|
"mean_token_accuracy": 0.12839893773198127,
|
|
"num_tokens": 2321911.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.108477783203125,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 7.0208,
|
|
"mean_token_accuracy": 0.12018021941184998,
|
|
"num_tokens": 2331932.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.837444925308228,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 6.8088,
|
|
"mean_token_accuracy": 0.126504997164011,
|
|
"num_tokens": 2342264.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.097235155105591,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 6.9754,
|
|
"mean_token_accuracy": 0.11833128929138184,
|
|
"num_tokens": 2353068.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.940999603271484,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 6.7782,
|
|
"mean_token_accuracy": 0.12161566987633705,
|
|
"num_tokens": 2361765.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.015862894058228,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 6.8349,
|
|
"mean_token_accuracy": 0.11880438923835754,
|
|
"num_tokens": 2372721.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.931281423568725,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 6.9136,
|
|
"mean_token_accuracy": 0.1202242910861969,
|
|
"num_tokens": 2382288.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.987937498092651,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 6.9225,
|
|
"mean_token_accuracy": 0.1201055109500885,
|
|
"num_tokens": 2391953.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.024634742736817,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 6.9557,
|
|
"mean_token_accuracy": 0.11614167094230651,
|
|
"num_tokens": 2402705.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.973006725311279,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 6.9007,
|
|
"mean_token_accuracy": 0.11774003282189369,
|
|
"num_tokens": 2412654.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.076304149627686,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 7.0036,
|
|
"mean_token_accuracy": 0.12298540249466897,
|
|
"num_tokens": 2422958.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.919585371017456,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 6.8278,
|
|
"mean_token_accuracy": 0.1256534494459629,
|
|
"num_tokens": 2432274.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.986346006393433,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 6.9947,
|
|
"mean_token_accuracy": 0.11662818863987923,
|
|
"num_tokens": 2443818.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.016325521469116,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 6.8953,
|
|
"mean_token_accuracy": 0.11740255281329155,
|
|
"num_tokens": 2453474.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.8944995403289795,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 6.835,
|
|
"mean_token_accuracy": 0.12458662986755371,
|
|
"num_tokens": 2462715.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.846959495544434,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 6.7497,
|
|
"mean_token_accuracy": 0.1315823934972286,
|
|
"num_tokens": 2473256.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.0362790584564205,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 6.7982,
|
|
"mean_token_accuracy": 0.11804969012737274,
|
|
"num_tokens": 2482475.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.89988112449646,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 6.8586,
|
|
"mean_token_accuracy": 0.12154440209269524,
|
|
"num_tokens": 2493163.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.071360063552857,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 6.8924,
|
|
"mean_token_accuracy": 0.12033604830503464,
|
|
"num_tokens": 2502062.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.961255645751953,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 6.8963,
|
|
"mean_token_accuracy": 0.12024620845913887,
|
|
"num_tokens": 2511196.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.960355758666992,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 6.8045,
|
|
"mean_token_accuracy": 0.12246129810810089,
|
|
"num_tokens": 2520492.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.869314336776734,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 6.8433,
|
|
"mean_token_accuracy": 0.12053112015128135,
|
|
"num_tokens": 2530368.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.951426267623901,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 6.8433,
|
|
"mean_token_accuracy": 0.11951586231589317,
|
|
"num_tokens": 2540734.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.865332794189453,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 6.8849,
|
|
"mean_token_accuracy": 0.11997138857841491,
|
|
"num_tokens": 2550907.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.091476535797119,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 6.9347,
|
|
"mean_token_accuracy": 0.12072236984968185,
|
|
"num_tokens": 2560062.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.950492095947266,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 6.8496,
|
|
"mean_token_accuracy": 0.11858923509716987,
|
|
"num_tokens": 2568840.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.028611564636231,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 6.9239,
|
|
"mean_token_accuracy": 0.11663276925683022,
|
|
"num_tokens": 2578566.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.926668787002564,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 6.9353,
|
|
"mean_token_accuracy": 0.1164866603910923,
|
|
"num_tokens": 2589853.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.004153776168823,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 6.8875,
|
|
"mean_token_accuracy": 0.11925893649458885,
|
|
"num_tokens": 2600848.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.871169805526733,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 6.8388,
|
|
"mean_token_accuracy": 0.11952893510460853,
|
|
"num_tokens": 2610730.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.973673725128174,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 6.9268,
|
|
"mean_token_accuracy": 0.11996840909123421,
|
|
"num_tokens": 2621129.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.997471475601197,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 6.8324,
|
|
"mean_token_accuracy": 0.12776239216327667,
|
|
"num_tokens": 2630306.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.920943546295166,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 6.9681,
|
|
"mean_token_accuracy": 0.11704278215765954,
|
|
"num_tokens": 2639874.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.954730463027954,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 6.9154,
|
|
"mean_token_accuracy": 0.11681979894638062,
|
|
"num_tokens": 2650340.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.899462127685547,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 6.6792,
|
|
"mean_token_accuracy": 0.13105636909604074,
|
|
"num_tokens": 2659541.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.897839117050171,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 6.891,
|
|
"mean_token_accuracy": 0.119564588367939,
|
|
"num_tokens": 2669151.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.863831377029419,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 6.7588,
|
|
"mean_token_accuracy": 0.13353405073285102,
|
|
"num_tokens": 2678867.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.90164623260498,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 6.8977,
|
|
"mean_token_accuracy": 0.11816785335540772,
|
|
"num_tokens": 2687572.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.916720294952393,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 6.8263,
|
|
"mean_token_accuracy": 0.11733598187565804,
|
|
"num_tokens": 2696922.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.887785625457764,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 6.8546,
|
|
"mean_token_accuracy": 0.12436464950442314,
|
|
"num_tokens": 2707465.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.900615215301514,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 6.7507,
|
|
"mean_token_accuracy": 0.12554468885064124,
|
|
"num_tokens": 2718708.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.850574159622193,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 6.8303,
|
|
"mean_token_accuracy": 0.11606418639421463,
|
|
"num_tokens": 2728862.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.966866397857666,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 6.7908,
|
|
"mean_token_accuracy": 0.12374083399772644,
|
|
"num_tokens": 2738546.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.810509490966797,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 6.7312,
|
|
"mean_token_accuracy": 0.1258455790579319,
|
|
"num_tokens": 2747885.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.827880191802978,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 6.7847,
|
|
"mean_token_accuracy": 0.12803196012973786,
|
|
"num_tokens": 2756851.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.945564031600952,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 6.8056,
|
|
"mean_token_accuracy": 0.12532249093055725,
|
|
"num_tokens": 2766309.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.826288318634033,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 6.7852,
|
|
"mean_token_accuracy": 0.12112483680248261,
|
|
"num_tokens": 2775679.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.870833444595337,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 6.749,
|
|
"mean_token_accuracy": 0.12467086315155029,
|
|
"num_tokens": 2784610.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.896092796325684,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 6.8682,
|
|
"mean_token_accuracy": 0.1267649620771408,
|
|
"num_tokens": 2793969.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.831112432479858,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 6.6917,
|
|
"mean_token_accuracy": 0.12754441648721696,
|
|
"num_tokens": 2803096.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.769429063796997,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.7791,
|
|
"mean_token_accuracy": 0.12707841470837594,
|
|
"num_tokens": 2812320.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.009070777893067,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 6.8511,
|
|
"mean_token_accuracy": 0.12180934101343155,
|
|
"num_tokens": 2821393.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.779871034622192,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.13057589083909987,
|
|
"num_tokens": 2831182.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.900580501556396,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.7846,
|
|
"mean_token_accuracy": 0.12680001556873322,
|
|
"num_tokens": 2840429.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.815707778930664,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.13107500970363617,
|
|
"num_tokens": 2849443.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.8182614803314205,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 6.7516,
|
|
"mean_token_accuracy": 0.12866144329309465,
|
|
"num_tokens": 2858908.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.914551210403443,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 6.8311,
|
|
"mean_token_accuracy": 0.12273252010345459,
|
|
"num_tokens": 2869615.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.870243167877197,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.7585,
|
|
"mean_token_accuracy": 0.12355621457099915,
|
|
"num_tokens": 2879007.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.766189193725586,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.6833,
|
|
"mean_token_accuracy": 0.13059235736727715,
|
|
"num_tokens": 2887527.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.728798818588257,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.723,
|
|
"mean_token_accuracy": 0.1263209819793701,
|
|
"num_tokens": 2897137.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.824571895599365,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 6.8092,
|
|
"mean_token_accuracy": 0.1324581488966942,
|
|
"num_tokens": 2908346.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.87175030708313,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 6.8296,
|
|
"mean_token_accuracy": 0.1234806202352047,
|
|
"num_tokens": 2918438.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.954668474197388,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 6.945,
|
|
"mean_token_accuracy": 0.11962197124958038,
|
|
"num_tokens": 2928096.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.853156518936157,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.7823,
|
|
"mean_token_accuracy": 0.12796442881226539,
|
|
"num_tokens": 2938150.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.786738729476928,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.7274,
|
|
"mean_token_accuracy": 0.1297776661813259,
|
|
"num_tokens": 2949410.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.7901389598846436,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.6949,
|
|
"mean_token_accuracy": 0.13579546809196472,
|
|
"num_tokens": 2958564.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.7910431861877445,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.7786,
|
|
"mean_token_accuracy": 0.1290195368230343,
|
|
"num_tokens": 2968750.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.848161697387695,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.7246,
|
|
"mean_token_accuracy": 0.13478757292032242,
|
|
"num_tokens": 2978397.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.8014851093292235,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 6.8567,
|
|
"mean_token_accuracy": 0.12458442598581314,
|
|
"num_tokens": 2987625.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.826836442947387,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.12829354032874107,
|
|
"num_tokens": 2997266.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.847477674484253,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 6.7884,
|
|
"mean_token_accuracy": 0.12614995688199998,
|
|
"num_tokens": 3007474.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.923624086380005,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.7673,
|
|
"mean_token_accuracy": 0.123389732837677,
|
|
"num_tokens": 3016270.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.7074995040893555,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 6.7666,
|
|
"mean_token_accuracy": 0.12175339832901955,
|
|
"num_tokens": 3027397.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.9206921577453615,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.12787529453635216,
|
|
"num_tokens": 3037324.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.7620690822601315,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.7484,
|
|
"mean_token_accuracy": 0.12835074141621589,
|
|
"num_tokens": 3048315.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.7305238246917725,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.6496,
|
|
"mean_token_accuracy": 0.13294167146086694,
|
|
"num_tokens": 3057875.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.8799608707427975,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.7548,
|
|
"mean_token_accuracy": 0.125601577013731,
|
|
"num_tokens": 3068006.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.78745174407959,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.7997,
|
|
"mean_token_accuracy": 0.1332696110010147,
|
|
"num_tokens": 3079450.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.740923690795898,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.5381,
|
|
"mean_token_accuracy": 0.13897247165441512,
|
|
"num_tokens": 3088385.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.808420610427857,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.7121,
|
|
"mean_token_accuracy": 0.12634384483098984,
|
|
"num_tokens": 3097854.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.894461917877197,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 6.8638,
|
|
"mean_token_accuracy": 0.11956453770399093,
|
|
"num_tokens": 3107877.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.788162040710449,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.8179,
|
|
"mean_token_accuracy": 0.12026642337441444,
|
|
"num_tokens": 3117967.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.7056066513061525,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.7008,
|
|
"mean_token_accuracy": 0.1319211445748806,
|
|
"num_tokens": 3129827.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.832995510101318,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.6689,
|
|
"mean_token_accuracy": 0.12922804206609725,
|
|
"num_tokens": 3138779.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.703251361846924,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 6.7135,
|
|
"mean_token_accuracy": 0.1345683753490448,
|
|
"num_tokens": 3148741.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.753573036193847,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.7635,
|
|
"mean_token_accuracy": 0.133981604129076,
|
|
"num_tokens": 3158964.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.690367984771728,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.5609,
|
|
"mean_token_accuracy": 0.13681313022971153,
|
|
"num_tokens": 3168347.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.72955288887024,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.774,
|
|
"mean_token_accuracy": 0.12525363489985467,
|
|
"num_tokens": 3178276.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.75451922416687,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.7262,
|
|
"mean_token_accuracy": 0.12230646312236786,
|
|
"num_tokens": 3186975.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.737055873870849,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.5717,
|
|
"mean_token_accuracy": 0.14679480865597724,
|
|
"num_tokens": 3197311.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.6865979671478275,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.7164,
|
|
"mean_token_accuracy": 0.1300855614244938,
|
|
"num_tokens": 3208534.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.861715793609619,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.6607,
|
|
"mean_token_accuracy": 0.13207977190613746,
|
|
"num_tokens": 3217440.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.711705112457276,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.7775,
|
|
"mean_token_accuracy": 0.126724024862051,
|
|
"num_tokens": 3225858.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.6331892013549805,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.5618,
|
|
"mean_token_accuracy": 0.14867160394787787,
|
|
"num_tokens": 3235921.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.721046876907349,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.6597,
|
|
"mean_token_accuracy": 0.1392642341554165,
|
|
"num_tokens": 3245123.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.782358407974243,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.6534,
|
|
"mean_token_accuracy": 0.13713510632514953,
|
|
"num_tokens": 3254246.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.758942079544068,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.6484,
|
|
"mean_token_accuracy": 0.1331721879541874,
|
|
"num_tokens": 3263602.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.770178031921387,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 6.8251,
|
|
"mean_token_accuracy": 0.12867565378546714,
|
|
"num_tokens": 3274430.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.788376569747925,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.5735,
|
|
"mean_token_accuracy": 0.13186567202210425,
|
|
"num_tokens": 3283587.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.684832429885864,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.7188,
|
|
"mean_token_accuracy": 0.12207710668444634,
|
|
"num_tokens": 3294165.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.785898447036743,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.6989,
|
|
"mean_token_accuracy": 0.13502319902181625,
|
|
"num_tokens": 3302756.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.692429256439209,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.6267,
|
|
"mean_token_accuracy": 0.13172637969255446,
|
|
"num_tokens": 3313108.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.766379261016846,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.7195,
|
|
"mean_token_accuracy": 0.1308559313416481,
|
|
"num_tokens": 3323008.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.7030027389526365,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.638,
|
|
"mean_token_accuracy": 0.13223350718617438,
|
|
"num_tokens": 3332490.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.676813220977783,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.6968,
|
|
"mean_token_accuracy": 0.1267889440059662,
|
|
"num_tokens": 3343069.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.711957550048828,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.7402,
|
|
"mean_token_accuracy": 0.1320636212825775,
|
|
"num_tokens": 3351834.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.790498113632202,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.6777,
|
|
"mean_token_accuracy": 0.1267648383975029,
|
|
"num_tokens": 3360915.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.693454313278198,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.6693,
|
|
"mean_token_accuracy": 0.13222509175539016,
|
|
"num_tokens": 3370184.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.74412693977356,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.7416,
|
|
"mean_token_accuracy": 0.1286899082362652,
|
|
"num_tokens": 3379545.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.738655376434326,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.692,
|
|
"mean_token_accuracy": 0.1320427715778351,
|
|
"num_tokens": 3390238.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.755091857910156,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.6941,
|
|
"mean_token_accuracy": 0.13341115564107894,
|
|
"num_tokens": 3400437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.583831262588501,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.6544,
|
|
"mean_token_accuracy": 0.13529954701662064,
|
|
"num_tokens": 3410982.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.798423385620117,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.7576,
|
|
"mean_token_accuracy": 0.12626220062375068,
|
|
"num_tokens": 3420898.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.699107265472412,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.6068,
|
|
"mean_token_accuracy": 0.13776216059923171,
|
|
"num_tokens": 3429144.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.696451807022095,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.6618,
|
|
"mean_token_accuracy": 0.1360717937350273,
|
|
"num_tokens": 3439427.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.782762908935547,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.6461,
|
|
"mean_token_accuracy": 0.13736561238765715,
|
|
"num_tokens": 3448218.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.727622175216675,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.7015,
|
|
"mean_token_accuracy": 0.1316203609108925,
|
|
"num_tokens": 3457656.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.7920112133026125,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.68,
|
|
"mean_token_accuracy": 0.13132236450910567,
|
|
"num_tokens": 3467414.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.60561580657959,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.6319,
|
|
"mean_token_accuracy": 0.13211813122034072,
|
|
"num_tokens": 3477195.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.766419887542725,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.7036,
|
|
"mean_token_accuracy": 0.12701989114284515,
|
|
"num_tokens": 3487497.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.729431629180908,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.6678,
|
|
"mean_token_accuracy": 0.1326162837445736,
|
|
"num_tokens": 3497363.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.664000749588013,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.6669,
|
|
"mean_token_accuracy": 0.13428003191947938,
|
|
"num_tokens": 3506969.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.758932733535767,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.6374,
|
|
"mean_token_accuracy": 0.133426945656538,
|
|
"num_tokens": 3516418.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.646256637573242,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.579,
|
|
"mean_token_accuracy": 0.13459737226366997,
|
|
"num_tokens": 3528023.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.680403089523315,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.6791,
|
|
"mean_token_accuracy": 0.126181248575449,
|
|
"num_tokens": 3537946.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.629174423217774,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.6441,
|
|
"mean_token_accuracy": 0.1346987560391426,
|
|
"num_tokens": 3547299.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.778659057617188,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.719,
|
|
"mean_token_accuracy": 0.13088061064481735,
|
|
"num_tokens": 3556075.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.693529462814331,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.6839,
|
|
"mean_token_accuracy": 0.13213445693254472,
|
|
"num_tokens": 3566314.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.683442831039429,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.5771,
|
|
"mean_token_accuracy": 0.1374658614397049,
|
|
"num_tokens": 3575095.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.673960494995117,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.6844,
|
|
"mean_token_accuracy": 0.13227922096848488,
|
|
"num_tokens": 3584375.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.6529711246490475,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.5489,
|
|
"mean_token_accuracy": 0.1394191190600395,
|
|
"num_tokens": 3594504.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.631583499908447,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.69,
|
|
"mean_token_accuracy": 0.12915903329849243,
|
|
"num_tokens": 3604050.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.778188943862915,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.5926,
|
|
"mean_token_accuracy": 0.1395821340382099,
|
|
"num_tokens": 3613388.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.655749797821045,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.7733,
|
|
"mean_token_accuracy": 0.12011321932077408,
|
|
"num_tokens": 3623758.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.818503284454346,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.6445,
|
|
"mean_token_accuracy": 0.13539304807782174,
|
|
"num_tokens": 3633040.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.634677410125732,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.5565,
|
|
"mean_token_accuracy": 0.13505534678697587,
|
|
"num_tokens": 3642980.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.705596256256103,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.7149,
|
|
"mean_token_accuracy": 0.13585179373621942,
|
|
"num_tokens": 3653817.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.733653402328491,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.6359,
|
|
"mean_token_accuracy": 0.130605985224247,
|
|
"num_tokens": 3664460.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.723941230773926,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.6779,
|
|
"mean_token_accuracy": 0.12275734916329384,
|
|
"num_tokens": 3674221.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.593170547485352,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.5576,
|
|
"mean_token_accuracy": 0.14246022552251816,
|
|
"num_tokens": 3683130.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.71184229850769,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.5075,
|
|
"mean_token_accuracy": 0.13699922263622283,
|
|
"num_tokens": 3694544.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.519984292984009,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.468,
|
|
"mean_token_accuracy": 0.13796097338199614,
|
|
"num_tokens": 3704857.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.704261016845703,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.6079,
|
|
"mean_token_accuracy": 0.13686281070113182,
|
|
"num_tokens": 3714623.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.6676782131195065,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.5741,
|
|
"mean_token_accuracy": 0.14162902683019638,
|
|
"num_tokens": 3723349.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.558374261856079,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.5519,
|
|
"mean_token_accuracy": 0.1372409552335739,
|
|
"num_tokens": 3732990.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.652611255645752,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.5476,
|
|
"mean_token_accuracy": 0.13684199526906013,
|
|
"num_tokens": 3742343.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.596817064285278,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.5754,
|
|
"mean_token_accuracy": 0.13304830119013786,
|
|
"num_tokens": 3752128.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.615978622436524,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.4941,
|
|
"mean_token_accuracy": 0.1367519110441208,
|
|
"num_tokens": 3760431.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.6409385204315186,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.5984,
|
|
"mean_token_accuracy": 0.1343713827431202,
|
|
"num_tokens": 3769623.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.536288595199585,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.5451,
|
|
"mean_token_accuracy": 0.13436267524957657,
|
|
"num_tokens": 3779383.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.681822633743286,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.7086,
|
|
"mean_token_accuracy": 0.12788542956113816,
|
|
"num_tokens": 3788847.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.528082084655762,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.5596,
|
|
"mean_token_accuracy": 0.14175273850560188,
|
|
"num_tokens": 3798711.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.6422826766967775,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.5227,
|
|
"mean_token_accuracy": 0.13906078785657883,
|
|
"num_tokens": 3808837.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.590198564529419,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.4968,
|
|
"mean_token_accuracy": 0.13925287276506423,
|
|
"num_tokens": 3818261.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.578291654586792,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.5356,
|
|
"mean_token_accuracy": 0.13477599993348122,
|
|
"num_tokens": 3828146.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.733463144302368,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.7972,
|
|
"mean_token_accuracy": 0.12044794335961342,
|
|
"num_tokens": 3838658.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.618522310256958,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.5364,
|
|
"mean_token_accuracy": 0.13059837222099305,
|
|
"num_tokens": 3849097.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.653698873519898,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.6806,
|
|
"mean_token_accuracy": 0.12555816173553466,
|
|
"num_tokens": 3858645.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.6428728103637695,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.5302,
|
|
"mean_token_accuracy": 0.14176158830523491,
|
|
"num_tokens": 3869232.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.662562847137451,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.6339,
|
|
"mean_token_accuracy": 0.13388122096657754,
|
|
"num_tokens": 3879453.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.6446600437164305,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.5457,
|
|
"mean_token_accuracy": 0.13349285572767258,
|
|
"num_tokens": 3888257.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.642933034896851,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.5935,
|
|
"mean_token_accuracy": 0.1399511255323887,
|
|
"num_tokens": 3897980.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.54600772857666,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.4067,
|
|
"mean_token_accuracy": 0.14303143247961997,
|
|
"num_tokens": 3907479.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.577399110794067,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.6357,
|
|
"mean_token_accuracy": 0.13481335565447808,
|
|
"num_tokens": 3917249.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.484592456422094,
|
|
"eval_loss": 6.79269552230835,
|
|
"eval_mean_token_accuracy": 0.12670518923980026,
|
|
"eval_num_tokens": 3917249.0,
|
|
"eval_runtime": 2.6856,
|
|
"eval_samples_per_second": 1290.212,
|
|
"eval_steps_per_second": 161.602,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 6453048170880000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|