d2120b5eda
Model: fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.9945300845350572,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3222032873472000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|