de25a3f955
Model: fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.9825957235206366,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 8.055328845977783,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.7699,
|
|
"mean_token_accuracy": 0.09289914444088936,
|
|
"num_tokens": 9089.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.9934410572052,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.9292,
|
|
"mean_token_accuracy": 0.0899883709847927,
|
|
"num_tokens": 21650.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 8.043873596191407,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.8569,
|
|
"mean_token_accuracy": 0.0829312488436699,
|
|
"num_tokens": 31067.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.9948217391967775,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.8229,
|
|
"mean_token_accuracy": 0.08587435036897659,
|
|
"num_tokens": 40816.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 8.020876693725587,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.8415,
|
|
"mean_token_accuracy": 0.08777489773929119,
|
|
"num_tokens": 51065.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.016945314407348,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.8082,
|
|
"mean_token_accuracy": 0.09032100439071655,
|
|
"num_tokens": 60539.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 8.016125869750976,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.8391,
|
|
"mean_token_accuracy": 0.08789670392870903,
|
|
"num_tokens": 71010.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 8.02649121284485,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.9238,
|
|
"mean_token_accuracy": 0.08401809930801392,
|
|
"num_tokens": 81394.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 8.087946605682372,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.9606,
|
|
"mean_token_accuracy": 0.08528241962194442,
|
|
"num_tokens": 92491.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 8.078335189819336,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.788,
|
|
"mean_token_accuracy": 0.09470873698592186,
|
|
"num_tokens": 102113.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 8.062228298187256,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.9462,
|
|
"mean_token_accuracy": 0.08278393298387528,
|
|
"num_tokens": 112211.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 8.023606491088866,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.7747,
|
|
"mean_token_accuracy": 0.0901342697441578,
|
|
"num_tokens": 122166.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 8.003111219406128,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.7581,
|
|
"mean_token_accuracy": 0.09486681148409844,
|
|
"num_tokens": 131016.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.973823165893554,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.8401,
|
|
"mean_token_accuracy": 0.09549511298537254,
|
|
"num_tokens": 141369.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 8.026473474502563,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.7438,
|
|
"mean_token_accuracy": 0.09063734859228134,
|
|
"num_tokens": 150939.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 8.038272190093995,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.7447,
|
|
"mean_token_accuracy": 0.09616999849677085,
|
|
"num_tokens": 159981.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.98382682800293,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.7952,
|
|
"mean_token_accuracy": 0.08998179510235786,
|
|
"num_tokens": 170139.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 8.070371198654176,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.7446,
|
|
"mean_token_accuracy": 0.09454798772931099,
|
|
"num_tokens": 180746.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 8.05376615524292,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.6953,
|
|
"mean_token_accuracy": 0.09375635012984276,
|
|
"num_tokens": 190250.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.960572814941406,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.851,
|
|
"mean_token_accuracy": 0.08996001482009888,
|
|
"num_tokens": 201185.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 8.003534412384033,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.8286,
|
|
"mean_token_accuracy": 0.08292106837034226,
|
|
"num_tokens": 211476.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 8.028951740264892,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.7808,
|
|
"mean_token_accuracy": 0.08876777328550815,
|
|
"num_tokens": 220965.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 8.002874565124511,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.8037,
|
|
"mean_token_accuracy": 0.08823544830083847,
|
|
"num_tokens": 231539.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 8.01924533843994,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.7812,
|
|
"mean_token_accuracy": 0.08827254623174667,
|
|
"num_tokens": 241206.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.982138299942017,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.7187,
|
|
"mean_token_accuracy": 0.09137163013219833,
|
|
"num_tokens": 250629.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 8.054266786575317,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8412,
|
|
"mean_token_accuracy": 0.09213275685906411,
|
|
"num_tokens": 260705.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 8.1447190284729,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7988,
|
|
"mean_token_accuracy": 0.09144297614693642,
|
|
"num_tokens": 270499.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.07759428024292,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.8214,
|
|
"mean_token_accuracy": 0.08136414065957069,
|
|
"num_tokens": 279998.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.998452997207641,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.7959,
|
|
"mean_token_accuracy": 0.09346884936094284,
|
|
"num_tokens": 290322.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.026957654953003,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.741,
|
|
"mean_token_accuracy": 0.09033143669366836,
|
|
"num_tokens": 299893.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.940546894073487,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7927,
|
|
"mean_token_accuracy": 0.09222338274121285,
|
|
"num_tokens": 309191.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.115492153167725,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7385,
|
|
"mean_token_accuracy": 0.09139059036970139,
|
|
"num_tokens": 318494.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.960078859329224,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7788,
|
|
"mean_token_accuracy": 0.0923788644373417,
|
|
"num_tokens": 328299.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.144480419158935,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8496,
|
|
"mean_token_accuracy": 0.08552220612764358,
|
|
"num_tokens": 338878.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.092591524124146,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7971,
|
|
"mean_token_accuracy": 0.08752194195985794,
|
|
"num_tokens": 348508.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.965788459777832,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7449,
|
|
"mean_token_accuracy": 0.08802809938788414,
|
|
"num_tokens": 358233.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.010887241363525,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7062,
|
|
"mean_token_accuracy": 0.09302832186222076,
|
|
"num_tokens": 367297.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.047781038284302,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7319,
|
|
"mean_token_accuracy": 0.08684638142585754,
|
|
"num_tokens": 376880.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.015840673446656,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.7585,
|
|
"mean_token_accuracy": 0.09129639342427254,
|
|
"num_tokens": 386806.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.969292211532593,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.7064,
|
|
"mean_token_accuracy": 0.0939807079732418,
|
|
"num_tokens": 396161.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.018733930587768,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7352,
|
|
"mean_token_accuracy": 0.09067206978797912,
|
|
"num_tokens": 405643.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.042714643478394,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7571,
|
|
"mean_token_accuracy": 0.09590560421347619,
|
|
"num_tokens": 414367.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.020010566711425,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.7188,
|
|
"mean_token_accuracy": 0.09117217287421227,
|
|
"num_tokens": 424430.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.063075876235962,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.7448,
|
|
"mean_token_accuracy": 0.09226949140429497,
|
|
"num_tokens": 435176.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.007707548141479,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.8191,
|
|
"mean_token_accuracy": 0.09036103039979934,
|
|
"num_tokens": 445611.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.024854516983032,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.732,
|
|
"mean_token_accuracy": 0.09320453926920891,
|
|
"num_tokens": 455143.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.067090892791748,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7643,
|
|
"mean_token_accuracy": 0.09478386417031288,
|
|
"num_tokens": 465507.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.055903482437134,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.8112,
|
|
"mean_token_accuracy": 0.08999223560094834,
|
|
"num_tokens": 476152.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.923675012588501,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.6274,
|
|
"mean_token_accuracy": 0.09889271780848503,
|
|
"num_tokens": 486341.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.941055393218994,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.5837,
|
|
"mean_token_accuracy": 0.09929879531264305,
|
|
"num_tokens": 495915.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.947218942642212,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.7126,
|
|
"mean_token_accuracy": 0.09258004277944565,
|
|
"num_tokens": 505666.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.92978081703186,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.5981,
|
|
"mean_token_accuracy": 0.09562918171286583,
|
|
"num_tokens": 515411.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.961107540130615,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7116,
|
|
"mean_token_accuracy": 0.09054910317063332,
|
|
"num_tokens": 525018.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.943254613876343,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.6326,
|
|
"mean_token_accuracy": 0.09534763172268867,
|
|
"num_tokens": 535056.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.991715574264527,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6291,
|
|
"mean_token_accuracy": 0.09336584955453872,
|
|
"num_tokens": 545115.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.080969667434692,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.6495,
|
|
"mean_token_accuracy": 0.09909325316548348,
|
|
"num_tokens": 554429.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.849652528762817,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.5983,
|
|
"mean_token_accuracy": 0.10442101955413818,
|
|
"num_tokens": 563357.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.9693444728851315,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6591,
|
|
"mean_token_accuracy": 0.09824569448828697,
|
|
"num_tokens": 574480.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.956569957733154,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.6345,
|
|
"mean_token_accuracy": 0.0936326451599598,
|
|
"num_tokens": 583722.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.912858390808106,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.5976,
|
|
"mean_token_accuracy": 0.10199215114116669,
|
|
"num_tokens": 593644.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.891701698303223,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6595,
|
|
"mean_token_accuracy": 0.09747006073594093,
|
|
"num_tokens": 603547.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.942541217803955,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.6749,
|
|
"mean_token_accuracy": 0.09077658727765084,
|
|
"num_tokens": 613880.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.01622257232666,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6791,
|
|
"mean_token_accuracy": 0.09485002011060714,
|
|
"num_tokens": 624213.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.940730810165405,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.7188,
|
|
"mean_token_accuracy": 0.09535183757543564,
|
|
"num_tokens": 634350.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.01810598373413,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.6764,
|
|
"mean_token_accuracy": 0.09407038241624832,
|
|
"num_tokens": 644040.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.90651569366455,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6464,
|
|
"mean_token_accuracy": 0.09886646866798401,
|
|
"num_tokens": 652915.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.997863626480102,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.663,
|
|
"mean_token_accuracy": 0.09337838441133499,
|
|
"num_tokens": 662665.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.880599927902222,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.6934,
|
|
"mean_token_accuracy": 0.09364098682999611,
|
|
"num_tokens": 671667.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.07152271270752,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.6974,
|
|
"mean_token_accuracy": 0.08758332729339599,
|
|
"num_tokens": 680432.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.930040788650513,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.7059,
|
|
"mean_token_accuracy": 0.09117975533008575,
|
|
"num_tokens": 690184.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.958481740951538,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5809,
|
|
"mean_token_accuracy": 0.09757120013237,
|
|
"num_tokens": 699147.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.778140354156494,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.6241,
|
|
"mean_token_accuracy": 0.09795154929161072,
|
|
"num_tokens": 708696.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.94059271812439,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.5868,
|
|
"mean_token_accuracy": 0.09651872366666794,
|
|
"num_tokens": 718438.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.958788824081421,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.7347,
|
|
"mean_token_accuracy": 0.08726413100957871,
|
|
"num_tokens": 728157.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.920962810516357,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.6345,
|
|
"mean_token_accuracy": 0.09856286793947219,
|
|
"num_tokens": 737007.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.974191665649414,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.5487,
|
|
"mean_token_accuracy": 0.1027902714908123,
|
|
"num_tokens": 746873.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.777252101898194,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.4764,
|
|
"mean_token_accuracy": 0.10671861693263054,
|
|
"num_tokens": 755486.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.895697546005249,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.6079,
|
|
"mean_token_accuracy": 0.09227629154920577,
|
|
"num_tokens": 765481.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.837667560577392,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5488,
|
|
"mean_token_accuracy": 0.09985539466142654,
|
|
"num_tokens": 774335.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.047065830230713,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.6338,
|
|
"mean_token_accuracy": 0.09441028982400894,
|
|
"num_tokens": 784307.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.9048755168914795,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.6761,
|
|
"mean_token_accuracy": 0.09370537623763084,
|
|
"num_tokens": 795058.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.047796821594238,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.6291,
|
|
"mean_token_accuracy": 0.09702575877308846,
|
|
"num_tokens": 805687.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.804214286804199,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.6174,
|
|
"mean_token_accuracy": 0.09554455727338791,
|
|
"num_tokens": 815138.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.951880931854248,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5742,
|
|
"mean_token_accuracy": 0.09677162915468215,
|
|
"num_tokens": 825009.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.795875835418701,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.5637,
|
|
"mean_token_accuracy": 0.09957250207662582,
|
|
"num_tokens": 834708.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.925895738601684,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.6117,
|
|
"mean_token_accuracy": 0.0967609629034996,
|
|
"num_tokens": 843888.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.853721952438354,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.6049,
|
|
"mean_token_accuracy": 0.09610669240355492,
|
|
"num_tokens": 853057.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.923298740386963,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.6546,
|
|
"mean_token_accuracy": 0.09225371703505517,
|
|
"num_tokens": 862701.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.815232610702514,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.549,
|
|
"mean_token_accuracy": 0.10032494440674782,
|
|
"num_tokens": 872129.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.797481489181519,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.5867,
|
|
"mean_token_accuracy": 0.0956971988081932,
|
|
"num_tokens": 883166.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.8453751564025875,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.4193,
|
|
"mean_token_accuracy": 0.10789886862039566,
|
|
"num_tokens": 894159.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.877769851684571,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.5812,
|
|
"mean_token_accuracy": 0.09706785902380943,
|
|
"num_tokens": 904588.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.8562257289886475,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.5406,
|
|
"mean_token_accuracy": 0.09594281017780304,
|
|
"num_tokens": 915554.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.823633670806885,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.593,
|
|
"mean_token_accuracy": 0.09437503591179848,
|
|
"num_tokens": 925546.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.860237646102905,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.5458,
|
|
"mean_token_accuracy": 0.09933575987815857,
|
|
"num_tokens": 936615.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.841644430160523,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.5854,
|
|
"mean_token_accuracy": 0.09283004775643348,
|
|
"num_tokens": 947374.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.803491544723511,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.5083,
|
|
"mean_token_accuracy": 0.09690938144922256,
|
|
"num_tokens": 957772.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.840833377838135,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.6166,
|
|
"mean_token_accuracy": 0.09776200726628304,
|
|
"num_tokens": 967724.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.732542610168457,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.5253,
|
|
"mean_token_accuracy": 0.0953903116285801,
|
|
"num_tokens": 976497.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.777756214141846,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.5113,
|
|
"mean_token_accuracy": 0.09828274846076965,
|
|
"num_tokens": 986207.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4972650422675286,
|
|
"eval_entropy": 7.791762825530795,
|
|
"eval_loss": 7.572336673736572,
|
|
"eval_mean_token_accuracy": 0.09290712282559427,
|
|
"eval_num_tokens": 986207.0,
|
|
"eval_runtime": 10.6134,
|
|
"eval_samples_per_second": 326.475,
|
|
"eval_steps_per_second": 40.892,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.700582551956177,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.4704,
|
|
"mean_token_accuracy": 0.10335623621940612,
|
|
"num_tokens": 995568.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.818715000152588,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.5242,
|
|
"mean_token_accuracy": 0.09710540696978569,
|
|
"num_tokens": 1005829.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.748081254959106,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.4813,
|
|
"mean_token_accuracy": 0.0990183673799038,
|
|
"num_tokens": 1015290.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.806796503067017,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.4393,
|
|
"mean_token_accuracy": 0.10151970684528351,
|
|
"num_tokens": 1024479.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.689154100418091,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.3823,
|
|
"mean_token_accuracy": 0.10462581366300583,
|
|
"num_tokens": 1034092.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.866795063018799,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.5921,
|
|
"mean_token_accuracy": 0.09865057915449142,
|
|
"num_tokens": 1043844.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.7049336433410645,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.438,
|
|
"mean_token_accuracy": 0.09978735372424126,
|
|
"num_tokens": 1052646.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.812493181228637,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.4984,
|
|
"mean_token_accuracy": 0.10014364868402481,
|
|
"num_tokens": 1062345.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.697603464126587,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.4877,
|
|
"mean_token_accuracy": 0.10163192600011825,
|
|
"num_tokens": 1071116.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.705530023574829,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.4518,
|
|
"mean_token_accuracy": 0.10024819299578666,
|
|
"num_tokens": 1080562.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.769873380661011,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5085,
|
|
"mean_token_accuracy": 0.09623854458332062,
|
|
"num_tokens": 1091944.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.738706064224243,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.428,
|
|
"mean_token_accuracy": 0.09941570535302162,
|
|
"num_tokens": 1101412.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.751716041564942,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.5273,
|
|
"mean_token_accuracy": 0.09751861318945884,
|
|
"num_tokens": 1110652.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.731394052505493,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.4286,
|
|
"mean_token_accuracy": 0.10633731186389923,
|
|
"num_tokens": 1120420.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.672761011123657,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.4971,
|
|
"mean_token_accuracy": 0.09915391579270363,
|
|
"num_tokens": 1130046.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.727224063873291,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.4616,
|
|
"mean_token_accuracy": 0.09992800429463386,
|
|
"num_tokens": 1138946.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.676466178894043,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.2396,
|
|
"mean_token_accuracy": 0.10890419781208038,
|
|
"num_tokens": 1149440.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.697867679595947,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.5005,
|
|
"mean_token_accuracy": 0.09577652290463448,
|
|
"num_tokens": 1158636.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.611849689483643,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.3645,
|
|
"mean_token_accuracy": 0.10494528487324714,
|
|
"num_tokens": 1167776.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.796306943893432,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.3803,
|
|
"mean_token_accuracy": 0.10176268145442009,
|
|
"num_tokens": 1177919.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.503759813308716,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.4353,
|
|
"mean_token_accuracy": 0.10978803336620331,
|
|
"num_tokens": 1187363.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.82352294921875,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.6222,
|
|
"mean_token_accuracy": 0.0936784565448761,
|
|
"num_tokens": 1197329.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.739608573913574,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.5092,
|
|
"mean_token_accuracy": 0.10064340829849243,
|
|
"num_tokens": 1207408.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.672548484802246,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.4761,
|
|
"mean_token_accuracy": 0.09743527770042419,
|
|
"num_tokens": 1219258.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.729985666275025,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.4926,
|
|
"mean_token_accuracy": 0.10201280266046524,
|
|
"num_tokens": 1228561.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.63199667930603,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.4172,
|
|
"mean_token_accuracy": 0.10181611031293869,
|
|
"num_tokens": 1239665.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.664997196197509,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.4299,
|
|
"mean_token_accuracy": 0.10080647617578506,
|
|
"num_tokens": 1249490.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.661840343475342,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.444,
|
|
"mean_token_accuracy": 0.09851267859339714,
|
|
"num_tokens": 1259328.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.728383016586304,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.3382,
|
|
"mean_token_accuracy": 0.10014413371682167,
|
|
"num_tokens": 1269523.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.566671514511109,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4814,
|
|
"mean_token_accuracy": 0.09888403639197349,
|
|
"num_tokens": 1277961.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.71548638343811,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.4241,
|
|
"mean_token_accuracy": 0.10168351605534554,
|
|
"num_tokens": 1288228.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.617433977127075,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.3612,
|
|
"mean_token_accuracy": 0.10187801942229271,
|
|
"num_tokens": 1299636.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.645993185043335,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.4156,
|
|
"mean_token_accuracy": 0.1010623887181282,
|
|
"num_tokens": 1310035.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.518844509124756,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.4243,
|
|
"mean_token_accuracy": 0.10684464648365974,
|
|
"num_tokens": 1320751.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.560531044006348,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.3123,
|
|
"mean_token_accuracy": 0.1015079528093338,
|
|
"num_tokens": 1331023.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.549887800216675,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.4162,
|
|
"mean_token_accuracy": 0.0934910848736763,
|
|
"num_tokens": 1341814.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.5861584663391115,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.3119,
|
|
"mean_token_accuracy": 0.10053951293230057,
|
|
"num_tokens": 1351026.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.675773239135742,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.4987,
|
|
"mean_token_accuracy": 0.09008960351347924,
|
|
"num_tokens": 1361004.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.494913244247437,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.3688,
|
|
"mean_token_accuracy": 0.09771248623728752,
|
|
"num_tokens": 1370729.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.63529224395752,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.3343,
|
|
"mean_token_accuracy": 0.10306410938501358,
|
|
"num_tokens": 1379525.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.504175186157227,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4201,
|
|
"mean_token_accuracy": 0.09782344102859497,
|
|
"num_tokens": 1389863.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.53595929145813,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.5219,
|
|
"mean_token_accuracy": 0.09682246819138526,
|
|
"num_tokens": 1399468.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.5885009765625,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.3137,
|
|
"mean_token_accuracy": 0.1067239873111248,
|
|
"num_tokens": 1408785.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.479005336761475,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.3641,
|
|
"mean_token_accuracy": 0.106873170286417,
|
|
"num_tokens": 1418666.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.666646814346313,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.456,
|
|
"mean_token_accuracy": 0.10155233442783355,
|
|
"num_tokens": 1428699.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.585138511657715,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.3622,
|
|
"mean_token_accuracy": 0.1017698161303997,
|
|
"num_tokens": 1437354.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.572378015518188,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.3688,
|
|
"mean_token_accuracy": 0.1023016132414341,
|
|
"num_tokens": 1447327.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.603588342666626,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.4419,
|
|
"mean_token_accuracy": 0.103351029753685,
|
|
"num_tokens": 1457793.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.511776924133301,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.3378,
|
|
"mean_token_accuracy": 0.10458240360021591,
|
|
"num_tokens": 1467143.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.545070505142212,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.4118,
|
|
"mean_token_accuracy": 0.1036214292049408,
|
|
"num_tokens": 1476422.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.486561870574951,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3393,
|
|
"mean_token_accuracy": 0.10872172713279724,
|
|
"num_tokens": 1486253.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.460987997055054,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.3504,
|
|
"mean_token_accuracy": 0.10164174363017082,
|
|
"num_tokens": 1496080.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.541303873062134,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.2914,
|
|
"mean_token_accuracy": 0.10655068382620811,
|
|
"num_tokens": 1505532.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.487213468551635,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.2302,
|
|
"mean_token_accuracy": 0.10376732945442199,
|
|
"num_tokens": 1515705.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.382348251342774,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.3104,
|
|
"mean_token_accuracy": 0.10365234911441804,
|
|
"num_tokens": 1525651.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.496106147766113,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.2997,
|
|
"mean_token_accuracy": 0.10095278844237328,
|
|
"num_tokens": 1534576.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.319676780700684,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 3.75,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.1693,
|
|
"mean_token_accuracy": 0.1153423972427845,
|
|
"num_tokens": 1544765.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.5089842796325685,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.2884,
|
|
"mean_token_accuracy": 0.11112908571958542,
|
|
"num_tokens": 1553636.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.429884243011474,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.2552,
|
|
"mean_token_accuracy": 0.10479394719004631,
|
|
"num_tokens": 1563356.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.357166290283203,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.2638,
|
|
"mean_token_accuracy": 0.10951004400849343,
|
|
"num_tokens": 1572508.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.495957851409912,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.3002,
|
|
"mean_token_accuracy": 0.10887534394860268,
|
|
"num_tokens": 1581583.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.4536933422088625,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.3065,
|
|
"mean_token_accuracy": 0.10426458492875099,
|
|
"num_tokens": 1592114.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.473225355148315,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.2075,
|
|
"mean_token_accuracy": 0.10634186267852783,
|
|
"num_tokens": 1602417.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.363744163513184,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.3076,
|
|
"mean_token_accuracy": 0.10592522397637368,
|
|
"num_tokens": 1611753.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.4581629753112795,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.3749,
|
|
"mean_token_accuracy": 0.10747737661004067,
|
|
"num_tokens": 1621959.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.386861944198609,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.2712,
|
|
"mean_token_accuracy": 0.10451980754733085,
|
|
"num_tokens": 1630909.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.430747222900391,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.231,
|
|
"mean_token_accuracy": 0.10984086617827415,
|
|
"num_tokens": 1641316.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.31324348449707,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.162,
|
|
"mean_token_accuracy": 0.11068534851074219,
|
|
"num_tokens": 1651506.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.327385950088501,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.1863,
|
|
"mean_token_accuracy": 0.1077701598405838,
|
|
"num_tokens": 1659871.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.48043303489685,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.3945,
|
|
"mean_token_accuracy": 0.1023897610604763,
|
|
"num_tokens": 1670178.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.385392951965332,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.2603,
|
|
"mean_token_accuracy": 0.10792220383882523,
|
|
"num_tokens": 1680131.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.27159686088562,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.1722,
|
|
"mean_token_accuracy": 0.10633963197469712,
|
|
"num_tokens": 1689568.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.463675117492675,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.2091,
|
|
"mean_token_accuracy": 0.11588807627558709,
|
|
"num_tokens": 1698039.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.324467611312866,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.2559,
|
|
"mean_token_accuracy": 0.11219154074788093,
|
|
"num_tokens": 1707659.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.355623626708985,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.1977,
|
|
"mean_token_accuracy": 0.1129053220152855,
|
|
"num_tokens": 1716952.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.395571517944336,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.2837,
|
|
"mean_token_accuracy": 0.10170883014798164,
|
|
"num_tokens": 1726323.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.420442390441894,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.2683,
|
|
"mean_token_accuracy": 0.10080605000257492,
|
|
"num_tokens": 1736265.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.318033504486084,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.2224,
|
|
"mean_token_accuracy": 0.1113182045519352,
|
|
"num_tokens": 1745150.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.413629579544067,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.3464,
|
|
"mean_token_accuracy": 0.103422349691391,
|
|
"num_tokens": 1754641.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.228156900405883,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.1335,
|
|
"mean_token_accuracy": 0.11196400448679925,
|
|
"num_tokens": 1763468.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.363440370559692,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.2689,
|
|
"mean_token_accuracy": 0.1072551429271698,
|
|
"num_tokens": 1774845.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.326597213745117,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.2023,
|
|
"mean_token_accuracy": 0.10989816784858704,
|
|
"num_tokens": 1784606.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.279292297363281,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.2172,
|
|
"mean_token_accuracy": 0.11103402152657509,
|
|
"num_tokens": 1793893.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.359324836730957,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.2327,
|
|
"mean_token_accuracy": 0.11132587045431137,
|
|
"num_tokens": 1804095.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.383117151260376,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.2325,
|
|
"mean_token_accuracy": 0.11229527816176414,
|
|
"num_tokens": 1813475.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.1859687805175785,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.1476,
|
|
"mean_token_accuracy": 0.11770151481032372,
|
|
"num_tokens": 1822779.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.370684814453125,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.1676,
|
|
"mean_token_accuracy": 0.1098080925643444,
|
|
"num_tokens": 1832239.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.159108829498291,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.1597,
|
|
"mean_token_accuracy": 0.10881416276097297,
|
|
"num_tokens": 1843395.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.331651067733764,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.2928,
|
|
"mean_token_accuracy": 0.10963825955986976,
|
|
"num_tokens": 1853861.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.316252088546753,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.1811,
|
|
"mean_token_accuracy": 0.11016927137970925,
|
|
"num_tokens": 1862992.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.308121156692505,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.2191,
|
|
"mean_token_accuracy": 0.10547153875231743,
|
|
"num_tokens": 1871684.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.327257347106934,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.2388,
|
|
"mean_token_accuracy": 0.10471320077776909,
|
|
"num_tokens": 1881956.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.2531952381134035,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.3038,
|
|
"mean_token_accuracy": 0.1041056789457798,
|
|
"num_tokens": 1891860.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.325952672958374,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.198,
|
|
"mean_token_accuracy": 0.11258542090654373,
|
|
"num_tokens": 1901521.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.231018781661987,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.1664,
|
|
"mean_token_accuracy": 0.11102576702833175,
|
|
"num_tokens": 1910438.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.181985664367676,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.204,
|
|
"mean_token_accuracy": 0.11018524467945098,
|
|
"num_tokens": 1922045.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.2546672344207765,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.0342,
|
|
"mean_token_accuracy": 0.12166386395692826,
|
|
"num_tokens": 1930330.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.023459434509277,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.0554,
|
|
"mean_token_accuracy": 0.11036473363637925,
|
|
"num_tokens": 1939485.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.2998260974884035,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.1335,
|
|
"mean_token_accuracy": 0.10684073865413665,
|
|
"num_tokens": 1949448.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.224824094772339,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.2343,
|
|
"mean_token_accuracy": 0.10612170770764351,
|
|
"num_tokens": 1959923.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.155893044537663,
|
|
"eval_loss": 7.266453742980957,
|
|
"eval_mean_token_accuracy": 0.10303841252309111,
|
|
"eval_num_tokens": 1959923.0,
|
|
"eval_runtime": 17.9035,
|
|
"eval_samples_per_second": 193.537,
|
|
"eval_steps_per_second": 24.241,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.241476678848267,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.1783,
|
|
"mean_token_accuracy": 0.11285145059227944,
|
|
"num_tokens": 1969105.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.1709286901685925,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 6.9728,
|
|
"mean_token_accuracy": 0.11111065910922156,
|
|
"num_tokens": 1979124.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.242545461654663,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 6.9845,
|
|
"mean_token_accuracy": 0.11652704775333404,
|
|
"num_tokens": 1987813.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.173238372802734,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 6.9709,
|
|
"mean_token_accuracy": 0.10800814852118493,
|
|
"num_tokens": 1996743.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.169902324676514,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 6.9781,
|
|
"mean_token_accuracy": 0.11889770552515984,
|
|
"num_tokens": 2006499.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.142313623428345,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 6.938,
|
|
"mean_token_accuracy": 0.11886951848864555,
|
|
"num_tokens": 2015103.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.117598581314087,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 6.8916,
|
|
"mean_token_accuracy": 0.11825998350977898,
|
|
"num_tokens": 2024681.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.157252025604248,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 6.9673,
|
|
"mean_token_accuracy": 0.10880548655986785,
|
|
"num_tokens": 2034388.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.067089462280274,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 6.8928,
|
|
"mean_token_accuracy": 0.11720709502696991,
|
|
"num_tokens": 2042729.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.257500410079956,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 7.0073,
|
|
"mean_token_accuracy": 0.10822945088148117,
|
|
"num_tokens": 2051669.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.103691864013672,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 6.9321,
|
|
"mean_token_accuracy": 0.11978971287608146,
|
|
"num_tokens": 2061002.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.070641994476318,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 6.8387,
|
|
"mean_token_accuracy": 0.12227742224931717,
|
|
"num_tokens": 2070222.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.130647850036621,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 7.0321,
|
|
"mean_token_accuracy": 0.11664234772324562,
|
|
"num_tokens": 2079683.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.198758506774903,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 7.01,
|
|
"mean_token_accuracy": 0.11501247957348823,
|
|
"num_tokens": 2089659.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.092316198348999,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 6.972,
|
|
"mean_token_accuracy": 0.11740045994520187,
|
|
"num_tokens": 2099766.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.127299070358276,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 7.0844,
|
|
"mean_token_accuracy": 0.10562424808740616,
|
|
"num_tokens": 2110244.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.178738021850586,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 6.995,
|
|
"mean_token_accuracy": 0.11412457302212715,
|
|
"num_tokens": 2120068.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.041600370407105,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 6.9809,
|
|
"mean_token_accuracy": 0.11297853365540504,
|
|
"num_tokens": 2128862.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.311644601821899,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 6.9714,
|
|
"mean_token_accuracy": 0.10598357245326043,
|
|
"num_tokens": 2140618.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.031351327896118,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 6.9358,
|
|
"mean_token_accuracy": 0.11448398232460022,
|
|
"num_tokens": 2151890.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.107405376434326,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 6.9429,
|
|
"mean_token_accuracy": 0.12040551453828811,
|
|
"num_tokens": 2160862.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.082157278060913,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 6.972,
|
|
"mean_token_accuracy": 0.1206208162009716,
|
|
"num_tokens": 2171631.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.101116418838501,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 6.9039,
|
|
"mean_token_accuracy": 0.11683834940195084,
|
|
"num_tokens": 2181564.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.0888330936431885,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 6.9483,
|
|
"mean_token_accuracy": 0.1219956248998642,
|
|
"num_tokens": 2191288.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.077724838256836,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 6.9091,
|
|
"mean_token_accuracy": 0.11639309674501419,
|
|
"num_tokens": 2201254.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.03474612236023,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 6.8604,
|
|
"mean_token_accuracy": 0.12253397256135941,
|
|
"num_tokens": 2210325.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.013818979263306,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 6.9783,
|
|
"mean_token_accuracy": 0.11415349319577217,
|
|
"num_tokens": 2220357.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.125763177871704,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 6.9136,
|
|
"mean_token_accuracy": 0.12185454815626144,
|
|
"num_tokens": 2229208.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.976495838165283,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 6.9066,
|
|
"mean_token_accuracy": 0.1208449885249138,
|
|
"num_tokens": 2238780.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.023766660690308,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 6.9321,
|
|
"mean_token_accuracy": 0.11635106280446053,
|
|
"num_tokens": 2249457.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.03515591621399,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 6.9243,
|
|
"mean_token_accuracy": 0.12004116848111153,
|
|
"num_tokens": 2259565.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.0158350467681885,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 6.8362,
|
|
"mean_token_accuracy": 0.1266941860318184,
|
|
"num_tokens": 2268301.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.0392069816589355,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 6.9599,
|
|
"mean_token_accuracy": 0.12106132805347443,
|
|
"num_tokens": 2278298.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.984105157852173,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 6.9147,
|
|
"mean_token_accuracy": 0.11764297261834145,
|
|
"num_tokens": 2288442.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.064280462265015,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 6.8592,
|
|
"mean_token_accuracy": 0.11508701145648956,
|
|
"num_tokens": 2298785.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.0413929462432865,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 6.9754,
|
|
"mean_token_accuracy": 0.11580611690878868,
|
|
"num_tokens": 2308881.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.988332509994507,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 6.8893,
|
|
"mean_token_accuracy": 0.12281140014529228,
|
|
"num_tokens": 2318447.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.054516458511353,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 6.8734,
|
|
"mean_token_accuracy": 0.1290799207985401,
|
|
"num_tokens": 2327162.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.068665790557861,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 6.9212,
|
|
"mean_token_accuracy": 0.11114579290151597,
|
|
"num_tokens": 2337207.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.90760908126831,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 6.915,
|
|
"mean_token_accuracy": 0.1156627170741558,
|
|
"num_tokens": 2347381.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.137280416488648,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 6.9566,
|
|
"mean_token_accuracy": 0.11516986638307572,
|
|
"num_tokens": 2357942.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.938902187347412,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 6.8771,
|
|
"mean_token_accuracy": 0.12062501758337021,
|
|
"num_tokens": 2367242.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.070527029037476,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 6.9193,
|
|
"mean_token_accuracy": 0.1242191381752491,
|
|
"num_tokens": 2376344.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.961205291748047,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 6.9481,
|
|
"mean_token_accuracy": 0.12189279198646545,
|
|
"num_tokens": 2387206.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.075365877151489,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 6.9677,
|
|
"mean_token_accuracy": 0.11550175845623016,
|
|
"num_tokens": 2397532.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.982903003692627,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 6.8367,
|
|
"mean_token_accuracy": 0.11436758115887642,
|
|
"num_tokens": 2409165.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.945490407943725,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 6.9284,
|
|
"mean_token_accuracy": 0.11924622058868409,
|
|
"num_tokens": 2420657.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.063077449798584,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 6.8696,
|
|
"mean_token_accuracy": 0.11942782625555992,
|
|
"num_tokens": 2429936.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.959188175201416,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 6.8697,
|
|
"mean_token_accuracy": 0.1241569198668003,
|
|
"num_tokens": 2440386.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.9969260692596436,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 7.0142,
|
|
"mean_token_accuracy": 0.11339911371469498,
|
|
"num_tokens": 2449086.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.175140237808227,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 7.0377,
|
|
"mean_token_accuracy": 0.11450302749872207,
|
|
"num_tokens": 2459380.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.951718759536743,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 6.8848,
|
|
"mean_token_accuracy": 0.12214633822441101,
|
|
"num_tokens": 2468623.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.023490524291992,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 6.8114,
|
|
"mean_token_accuracy": 0.11972368806600571,
|
|
"num_tokens": 2479817.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.9894520282745365,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 6.8848,
|
|
"mean_token_accuracy": 0.12344507575035095,
|
|
"num_tokens": 2490544.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.928281021118164,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 6.8455,
|
|
"mean_token_accuracy": 0.1291354365646839,
|
|
"num_tokens": 2500516.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.020780372619629,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 6.9773,
|
|
"mean_token_accuracy": 0.12090862020850182,
|
|
"num_tokens": 2510508.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.979570388793945,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 6.9218,
|
|
"mean_token_accuracy": 0.12041369453072548,
|
|
"num_tokens": 2520532.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.023465776443482,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 6.887,
|
|
"mean_token_accuracy": 0.11943913698196411,
|
|
"num_tokens": 2529912.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.949926042556763,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 6.9624,
|
|
"mean_token_accuracy": 0.11250624060630798,
|
|
"num_tokens": 2540700.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.981410264968872,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 6.8543,
|
|
"mean_token_accuracy": 0.12385431155562401,
|
|
"num_tokens": 2549343.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.865144109725952,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 6.8602,
|
|
"mean_token_accuracy": 0.12558662444353103,
|
|
"num_tokens": 2558630.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.992158222198486,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 6.8568,
|
|
"mean_token_accuracy": 0.12431244775652886,
|
|
"num_tokens": 2568559.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.963652801513672,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 6.9209,
|
|
"mean_token_accuracy": 0.1234417900443077,
|
|
"num_tokens": 2578088.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.930180501937866,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 6.8663,
|
|
"mean_token_accuracy": 0.12683536410331725,
|
|
"num_tokens": 2588518.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.067978286743164,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 6.9898,
|
|
"mean_token_accuracy": 0.11430260315537452,
|
|
"num_tokens": 2598850.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.021387243270874,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 6.9686,
|
|
"mean_token_accuracy": 0.1073531225323677,
|
|
"num_tokens": 2610667.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.954224586486816,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 6.8191,
|
|
"mean_token_accuracy": 0.1314563162624836,
|
|
"num_tokens": 2620430.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.77024655342102,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 6.7138,
|
|
"mean_token_accuracy": 0.12848498672246933,
|
|
"num_tokens": 2629433.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.951243495941162,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 6.8848,
|
|
"mean_token_accuracy": 0.11360282599925994,
|
|
"num_tokens": 2639611.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.018729877471924,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 6.8893,
|
|
"mean_token_accuracy": 0.11612915620207787,
|
|
"num_tokens": 2649494.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.918116617202759,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 6.7948,
|
|
"mean_token_accuracy": 0.12346037551760673,
|
|
"num_tokens": 2658723.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.913340854644775,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 6.8747,
|
|
"mean_token_accuracy": 0.12230955138802528,
|
|
"num_tokens": 2668530.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.951289796829224,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 6.9435,
|
|
"mean_token_accuracy": 0.12191602662205696,
|
|
"num_tokens": 2678788.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.9707471370697025,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 6.838,
|
|
"mean_token_accuracy": 0.12125239670276641,
|
|
"num_tokens": 2688308.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.735447740554809,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 6.7852,
|
|
"mean_token_accuracy": 0.1300181344151497,
|
|
"num_tokens": 2698146.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.007236576080322,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 6.8533,
|
|
"mean_token_accuracy": 0.12391836419701577,
|
|
"num_tokens": 2707336.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.90031623840332,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 6.8073,
|
|
"mean_token_accuracy": 0.1273755319416523,
|
|
"num_tokens": 2717573.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.875842666625976,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 6.8247,
|
|
"mean_token_accuracy": 0.12682975456118584,
|
|
"num_tokens": 2727488.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.834414958953857,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 6.8143,
|
|
"mean_token_accuracy": 0.13222975805401802,
|
|
"num_tokens": 2738010.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.945574712753296,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 6.8677,
|
|
"mean_token_accuracy": 0.12154300883412361,
|
|
"num_tokens": 2746984.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.899899244308472,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 6.8654,
|
|
"mean_token_accuracy": 0.11535304635763169,
|
|
"num_tokens": 2756578.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.906005001068115,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 6.8023,
|
|
"mean_token_accuracy": 0.12816528901457785,
|
|
"num_tokens": 2765773.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.878059434890747,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 6.8035,
|
|
"mean_token_accuracy": 0.12617505267262458,
|
|
"num_tokens": 2775002.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.936922693252564,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 6.873,
|
|
"mean_token_accuracy": 0.12435229048132897,
|
|
"num_tokens": 2784682.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.884806394577026,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 6.7813,
|
|
"mean_token_accuracy": 0.13019969910383225,
|
|
"num_tokens": 2794371.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.919467496871948,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 6.8578,
|
|
"mean_token_accuracy": 0.11788132414221764,
|
|
"num_tokens": 2803840.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.835882663726807,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.7454,
|
|
"mean_token_accuracy": 0.13257994279265403,
|
|
"num_tokens": 2813536.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.843277454376221,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 6.8069,
|
|
"mean_token_accuracy": 0.12237441018223763,
|
|
"num_tokens": 2822741.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.898312330245972,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 6.7842,
|
|
"mean_token_accuracy": 0.12550862208008767,
|
|
"num_tokens": 2832208.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.686010265350342,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.663,
|
|
"mean_token_accuracy": 0.1326310396194458,
|
|
"num_tokens": 2841672.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.867441606521607,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.7861,
|
|
"mean_token_accuracy": 0.12769792675971986,
|
|
"num_tokens": 2852323.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.86247444152832,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 6.8429,
|
|
"mean_token_accuracy": 0.1261087991297245,
|
|
"num_tokens": 2861863.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.926764297485351,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 6.7867,
|
|
"mean_token_accuracy": 0.12548809126019478,
|
|
"num_tokens": 2871608.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.9585823059082035,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.7525,
|
|
"mean_token_accuracy": 0.1302333578467369,
|
|
"num_tokens": 2882091.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.817167139053344,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.7359,
|
|
"mean_token_accuracy": 0.13480845838785172,
|
|
"num_tokens": 2892499.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.845238065719604,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.7645,
|
|
"mean_token_accuracy": 0.12246781513094902,
|
|
"num_tokens": 2902208.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.857409286499023,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 6.7555,
|
|
"mean_token_accuracy": 0.1244891531765461,
|
|
"num_tokens": 2911196.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.825583457946777,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 6.8108,
|
|
"mean_token_accuracy": 0.12157977223396302,
|
|
"num_tokens": 2921339.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.829862213134765,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 6.7633,
|
|
"mean_token_accuracy": 0.12412352785468102,
|
|
"num_tokens": 2930795.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.815878105163574,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.1268233872950077,
|
|
"num_tokens": 2939647.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.4912978617603183,
|
|
"eval_entropy": 6.714339637536607,
|
|
"eval_loss": 7.027554512023926,
|
|
"eval_mean_token_accuracy": 0.1158115748350384,
|
|
"eval_num_tokens": 2939647.0,
|
|
"eval_runtime": 13.3039,
|
|
"eval_samples_per_second": 260.45,
|
|
"eval_steps_per_second": 32.622,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.835459756851196,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.7481,
|
|
"mean_token_accuracy": 0.12527029886841773,
|
|
"num_tokens": 2948365.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.746442461013794,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.7688,
|
|
"mean_token_accuracy": 0.12213929519057273,
|
|
"num_tokens": 2957447.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.8704711437225345,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.7511,
|
|
"mean_token_accuracy": 0.1285943053662777,
|
|
"num_tokens": 2965938.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.894935607910156,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.7707,
|
|
"mean_token_accuracy": 0.12603728622198104,
|
|
"num_tokens": 2975830.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.8132643699646,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 6.7491,
|
|
"mean_token_accuracy": 0.12986776381731033,
|
|
"num_tokens": 2985391.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.82594780921936,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.744,
|
|
"mean_token_accuracy": 0.1368634097278118,
|
|
"num_tokens": 2994397.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.81362738609314,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 6.7869,
|
|
"mean_token_accuracy": 0.12196057885885239,
|
|
"num_tokens": 3005085.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.807890462875366,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.771,
|
|
"mean_token_accuracy": 0.12297940105199814,
|
|
"num_tokens": 3015500.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.883302497863769,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 6.8824,
|
|
"mean_token_accuracy": 0.12056382745504379,
|
|
"num_tokens": 3024862.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.900200939178466,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.7142,
|
|
"mean_token_accuracy": 0.13087142780423164,
|
|
"num_tokens": 3034775.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.749657487869262,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.7871,
|
|
"mean_token_accuracy": 0.12653839439153672,
|
|
"num_tokens": 3044048.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.909590578079223,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.7998,
|
|
"mean_token_accuracy": 0.1275485150516033,
|
|
"num_tokens": 3054237.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.612554597854614,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.6053,
|
|
"mean_token_accuracy": 0.13363586664199828,
|
|
"num_tokens": 3064787.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.7135475158691404,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.7758,
|
|
"mean_token_accuracy": 0.1243907667696476,
|
|
"num_tokens": 3074002.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.749949312210083,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.6525,
|
|
"mean_token_accuracy": 0.129579795897007,
|
|
"num_tokens": 3084270.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.872102737426758,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.7541,
|
|
"mean_token_accuracy": 0.13147178888320923,
|
|
"num_tokens": 3095236.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.74047703742981,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 6.7342,
|
|
"mean_token_accuracy": 0.12891987785696984,
|
|
"num_tokens": 3104912.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.77790150642395,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.7544,
|
|
"mean_token_accuracy": 0.1278040237724781,
|
|
"num_tokens": 3114782.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.925691080093384,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.7819,
|
|
"mean_token_accuracy": 0.13000695928931236,
|
|
"num_tokens": 3123606.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.682368040084839,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.6029,
|
|
"mean_token_accuracy": 0.13756740540266038,
|
|
"num_tokens": 3133343.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.910130405426026,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 6.8401,
|
|
"mean_token_accuracy": 0.11867891177535057,
|
|
"num_tokens": 3143137.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.797189426422119,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.7056,
|
|
"mean_token_accuracy": 0.12898564338684082,
|
|
"num_tokens": 3152595.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.758291721343994,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.8001,
|
|
"mean_token_accuracy": 0.12771535739302636,
|
|
"num_tokens": 3162333.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.749180459976197,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.6662,
|
|
"mean_token_accuracy": 0.12852055877447127,
|
|
"num_tokens": 3172765.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.855949831008911,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.8392,
|
|
"mean_token_accuracy": 0.12078842222690582,
|
|
"num_tokens": 3183613.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.871716928482056,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.7259,
|
|
"mean_token_accuracy": 0.1271841213107109,
|
|
"num_tokens": 3193020.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.740650987625122,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.7367,
|
|
"mean_token_accuracy": 0.12784550338983536,
|
|
"num_tokens": 3202845.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.698820781707764,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.6635,
|
|
"mean_token_accuracy": 0.12537247836589813,
|
|
"num_tokens": 3211195.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.819862747192383,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.7534,
|
|
"mean_token_accuracy": 0.13159084394574166,
|
|
"num_tokens": 3221549.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.688119506835937,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.6053,
|
|
"mean_token_accuracy": 0.13880132585763932,
|
|
"num_tokens": 3232206.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.809904432296753,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.7225,
|
|
"mean_token_accuracy": 0.1224221020936966,
|
|
"num_tokens": 3241613.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.65362229347229,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.6798,
|
|
"mean_token_accuracy": 0.13001496493816375,
|
|
"num_tokens": 3251968.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.732522678375244,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.6513,
|
|
"mean_token_accuracy": 0.13061657845973967,
|
|
"num_tokens": 3261969.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.7879266262054445,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 6.713,
|
|
"mean_token_accuracy": 0.1265723451972008,
|
|
"num_tokens": 3270894.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.723002815246582,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.7505,
|
|
"mean_token_accuracy": 0.12020857334136963,
|
|
"num_tokens": 3280949.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.85877799987793,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.7185,
|
|
"mean_token_accuracy": 0.12710705101490022,
|
|
"num_tokens": 3291449.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.644477128982544,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.7215,
|
|
"mean_token_accuracy": 0.12685354053974152,
|
|
"num_tokens": 3301632.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.804497814178466,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.5871,
|
|
"mean_token_accuracy": 0.1358786255121231,
|
|
"num_tokens": 3310802.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.711171293258667,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.6606,
|
|
"mean_token_accuracy": 0.13076258674263955,
|
|
"num_tokens": 3320833.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.716874408721924,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.6396,
|
|
"mean_token_accuracy": 0.13465334549546243,
|
|
"num_tokens": 3330030.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.516720962524414,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.44,
|
|
"mean_token_accuracy": 0.14811225682497026,
|
|
"num_tokens": 3338745.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.724125480651855,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.6863,
|
|
"mean_token_accuracy": 0.1328687660396099,
|
|
"num_tokens": 3348067.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.690582752227783,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.5785,
|
|
"mean_token_accuracy": 0.13962048292160034,
|
|
"num_tokens": 3357061.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.695200443267822,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.596,
|
|
"mean_token_accuracy": 0.1310545489192009,
|
|
"num_tokens": 3366573.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.709780216217041,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.6797,
|
|
"mean_token_accuracy": 0.12514400482177734,
|
|
"num_tokens": 3376996.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.850689458847046,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.7079,
|
|
"mean_token_accuracy": 0.12260405942797661,
|
|
"num_tokens": 3387615.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.5351707458496096,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.5019,
|
|
"mean_token_accuracy": 0.15043683126568794,
|
|
"num_tokens": 3397578.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.659417963027954,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.6746,
|
|
"mean_token_accuracy": 0.12657460197806358,
|
|
"num_tokens": 3406697.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.783771085739136,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.6459,
|
|
"mean_token_accuracy": 0.13219962567090987,
|
|
"num_tokens": 3416631.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.719735622406006,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.7346,
|
|
"mean_token_accuracy": 0.12242367416620255,
|
|
"num_tokens": 3426564.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.769449996948242,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.7388,
|
|
"mean_token_accuracy": 0.1309608444571495,
|
|
"num_tokens": 3436200.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.741757011413574,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.7252,
|
|
"mean_token_accuracy": 0.1283721849322319,
|
|
"num_tokens": 3446460.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.724111890792846,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 1.921875,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.5913,
|
|
"mean_token_accuracy": 0.1381916008889675,
|
|
"num_tokens": 3457629.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.721476221084595,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.7723,
|
|
"mean_token_accuracy": 0.12472076639533043,
|
|
"num_tokens": 3467754.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.6909526824951175,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.6901,
|
|
"mean_token_accuracy": 0.12899342849850653,
|
|
"num_tokens": 3477094.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.709241199493408,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.649,
|
|
"mean_token_accuracy": 0.13431572914123535,
|
|
"num_tokens": 3487276.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.6987536430358885,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.6413,
|
|
"mean_token_accuracy": 0.133719053119421,
|
|
"num_tokens": 3498118.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.633240509033203,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.4763,
|
|
"mean_token_accuracy": 0.13459513634443282,
|
|
"num_tokens": 3508286.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.6068065643310545,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.5962,
|
|
"mean_token_accuracy": 0.13357006162405013,
|
|
"num_tokens": 3517979.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.689920902252197,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.6253,
|
|
"mean_token_accuracy": 0.13220322877168655,
|
|
"num_tokens": 3528003.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.789654922485352,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.6844,
|
|
"mean_token_accuracy": 0.12885999754071237,
|
|
"num_tokens": 3537932.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.6286862850189205,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.8181,
|
|
"mean_token_accuracy": 0.12549626901745797,
|
|
"num_tokens": 3548584.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.8323548316955565,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.6232,
|
|
"mean_token_accuracy": 0.13544862642884253,
|
|
"num_tokens": 3558117.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.687149953842163,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.7517,
|
|
"mean_token_accuracy": 0.12215576767921447,
|
|
"num_tokens": 3569226.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.743102121353149,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.5999,
|
|
"mean_token_accuracy": 0.13134600892663,
|
|
"num_tokens": 3578375.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.685059261322022,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.6304,
|
|
"mean_token_accuracy": 0.13430401384830476,
|
|
"num_tokens": 3587356.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.726788187026978,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.6957,
|
|
"mean_token_accuracy": 0.13085027858614923,
|
|
"num_tokens": 3596785.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.662399053573608,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.5802,
|
|
"mean_token_accuracy": 0.13663700297474862,
|
|
"num_tokens": 3605357.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.723802614212036,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.5997,
|
|
"mean_token_accuracy": 0.1368691235780716,
|
|
"num_tokens": 3614979.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.673075103759766,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.583,
|
|
"mean_token_accuracy": 0.13440734073519706,
|
|
"num_tokens": 3624594.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.635775375366211,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.6129,
|
|
"mean_token_accuracy": 0.1337726093828678,
|
|
"num_tokens": 3633998.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.698388481140137,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.5638,
|
|
"mean_token_accuracy": 0.13645450323820113,
|
|
"num_tokens": 3642406.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.578488683700561,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.633,
|
|
"mean_token_accuracy": 0.12992775663733483,
|
|
"num_tokens": 3652905.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.679895401000977,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.5709,
|
|
"mean_token_accuracy": 0.13101621717214584,
|
|
"num_tokens": 3662518.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.699575424194336,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.6244,
|
|
"mean_token_accuracy": 0.13423072993755342,
|
|
"num_tokens": 3672102.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.703206348419189,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.6389,
|
|
"mean_token_accuracy": 0.13154254257678985,
|
|
"num_tokens": 3681821.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.7686851978302,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.7336,
|
|
"mean_token_accuracy": 0.12900978326797485,
|
|
"num_tokens": 3690753.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.642085838317871,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.12560946121811867,
|
|
"num_tokens": 3701694.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.69543228149414,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.6102,
|
|
"mean_token_accuracy": 0.13482431396842004,
|
|
"num_tokens": 3710894.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.643682813644409,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.5572,
|
|
"mean_token_accuracy": 0.1428975597023964,
|
|
"num_tokens": 3720081.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.644357204437256,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.6654,
|
|
"mean_token_accuracy": 0.12970876395702363,
|
|
"num_tokens": 3731168.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.65168194770813,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.453,
|
|
"mean_token_accuracy": 0.14205871373414994,
|
|
"num_tokens": 3740533.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.578559398651123,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.6241,
|
|
"mean_token_accuracy": 0.12968793734908104,
|
|
"num_tokens": 3750523.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.690841817855835,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.6616,
|
|
"mean_token_accuracy": 0.13006103932857513,
|
|
"num_tokens": 3760323.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.4785408020019535,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.4094,
|
|
"mean_token_accuracy": 0.15436080545186998,
|
|
"num_tokens": 3770970.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.657366704940796,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.6583,
|
|
"mean_token_accuracy": 0.13346222266554833,
|
|
"num_tokens": 3780810.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.691922521591186,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.5706,
|
|
"mean_token_accuracy": 0.13881206288933753,
|
|
"num_tokens": 3791461.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.514286136627197,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.3493,
|
|
"mean_token_accuracy": 0.15353642255067826,
|
|
"num_tokens": 3800353.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.594685983657837,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.5385,
|
|
"mean_token_accuracy": 0.1411295749247074,
|
|
"num_tokens": 3810633.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.682897472381592,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.6878,
|
|
"mean_token_accuracy": 0.13393995463848113,
|
|
"num_tokens": 3820254.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.710420083999634,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.6716,
|
|
"mean_token_accuracy": 0.13213822916150092,
|
|
"num_tokens": 3829976.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.63531084060669,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.616,
|
|
"mean_token_accuracy": 0.12971105426549911,
|
|
"num_tokens": 3840593.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.6207164287567135,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.5933,
|
|
"mean_token_accuracy": 0.14084266349673272,
|
|
"num_tokens": 3850063.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.678275203704834,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.6364,
|
|
"mean_token_accuracy": 0.13240463137626649,
|
|
"num_tokens": 3860132.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.625484371185303,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.5804,
|
|
"mean_token_accuracy": 0.13869195356965064,
|
|
"num_tokens": 3869914.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.618179845809936,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.5323,
|
|
"mean_token_accuracy": 0.146308434009552,
|
|
"num_tokens": 3879825.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.647174978256226,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.6094,
|
|
"mean_token_accuracy": 0.14014348685741423,
|
|
"num_tokens": 3889318.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.63746223449707,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.587,
|
|
"mean_token_accuracy": 0.13484067544341088,
|
|
"num_tokens": 3899017.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.648053073883057,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.5633,
|
|
"mean_token_accuracy": 0.13868496119976043,
|
|
"num_tokens": 3907766.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.609647560119629,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.5422,
|
|
"mean_token_accuracy": 0.1334989458322525,
|
|
"num_tokens": 3916323.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.389401653395271,
|
|
"eval_loss": 6.812779426574707,
|
|
"eval_mean_token_accuracy": 0.12556906916578794,
|
|
"eval_num_tokens": 3916323.0,
|
|
"eval_runtime": 10.4359,
|
|
"eval_samples_per_second": 332.027,
|
|
"eval_steps_per_second": 41.587,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.5790094375610355,
|
|
"epoch": 1.993535554450522,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004864723636325361,
|
|
"loss": 6.6133,
|
|
"mean_token_accuracy": 0.13446975126862526,
|
|
"num_tokens": 3927639.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.747440910339355,
|
|
"epoch": 1.9985082048731975,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 0.00048633867439566855,
|
|
"loss": 6.5045,
|
|
"mean_token_accuracy": 0.135795433819294,
|
|
"num_tokens": 3937452.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.634132014380561,
|
|
"epoch": 2.002983590253605,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004862043484784227,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.13990397502978644,
|
|
"num_tokens": 3946737.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.65202693939209,
|
|
"epoch": 2.0079562406762803,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048606938628546975,
|
|
"loss": 6.2441,
|
|
"mean_token_accuracy": 0.13903627321124076,
|
|
"num_tokens": 3957541.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.680914115905762,
|
|
"epoch": 2.0129288910989556,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004859337882233978,
|
|
"loss": 6.3243,
|
|
"mean_token_accuracy": 0.13173834905028342,
|
|
"num_tokens": 3967621.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.637338161468506,
|
|
"epoch": 2.017901541521631,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004857975547007106,
|
|
"loss": 6.2998,
|
|
"mean_token_accuracy": 0.14607882052659987,
|
|
"num_tokens": 3977920.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.406887578964233,
|
|
"epoch": 2.0228741919443065,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048566068612782595,
|
|
"loss": 5.9681,
|
|
"mean_token_accuracy": 0.16418219357728958,
|
|
"num_tokens": 3988128.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.57090950012207,
|
|
"epoch": 2.0278468423669818,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048552318291707533,
|
|
"loss": 6.163,
|
|
"mean_token_accuracy": 0.1462465599179268,
|
|
"num_tokens": 3998467.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.573382568359375,
|
|
"epoch": 2.032819492789657,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048538504548270177,
|
|
"loss": 6.159,
|
|
"mean_token_accuracy": 0.1430424600839615,
|
|
"num_tokens": 4007513.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.460464191436768,
|
|
"epoch": 2.0377921432123323,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048524627424085913,
|
|
"loss": 6.1935,
|
|
"mean_token_accuracy": 0.15068715214729309,
|
|
"num_tokens": 4016538.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.477964115142822,
|
|
"epoch": 2.0427647936350075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048510686960961057,
|
|
"loss": 6.1198,
|
|
"mean_token_accuracy": 0.15172877311706542,
|
|
"num_tokens": 4026850.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.528733730316162,
|
|
"epoch": 2.0477374440576828,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004849668320089276,
|
|
"loss": 6.1176,
|
|
"mean_token_accuracy": 0.14902114272117614,
|
|
"num_tokens": 4037105.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.468921232223511,
|
|
"epoch": 2.052710094480358,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048482616186068826,
|
|
"loss": 6.1672,
|
|
"mean_token_accuracy": 0.15710178166627883,
|
|
"num_tokens": 4045694.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.41478328704834,
|
|
"epoch": 2.0576827449030333,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048468485958867664,
|
|
"loss": 6.0695,
|
|
"mean_token_accuracy": 0.1532224215567112,
|
|
"num_tokens": 4056425.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.460195922851563,
|
|
"epoch": 2.0626553953257085,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048454292561858083,
|
|
"loss": 6.0962,
|
|
"mean_token_accuracy": 0.15494642853736879,
|
|
"num_tokens": 4066257.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.469780683517456,
|
|
"epoch": 2.0676280457483838,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004844003603779921,
|
|
"loss": 6.1758,
|
|
"mean_token_accuracy": 0.14712971001863479,
|
|
"num_tokens": 4076835.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.535259866714478,
|
|
"epoch": 2.072600696171059,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004842571642964036,
|
|
"loss": 6.2805,
|
|
"mean_token_accuracy": 0.13958214595913887,
|
|
"num_tokens": 4087644.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.509330368041992,
|
|
"epoch": 2.0775733465937343,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048411333780520876,
|
|
"loss": 6.2137,
|
|
"mean_token_accuracy": 0.14953471198678017,
|
|
"num_tokens": 4097040.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.446858310699463,
|
|
"epoch": 2.0825459970164095,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048396888133770034,
|
|
"loss": 6.1975,
|
|
"mean_token_accuracy": 0.1492289237678051,
|
|
"num_tokens": 4106698.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.545877838134766,
|
|
"epoch": 2.087518647439085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048382379532906897,
|
|
"loss": 6.1988,
|
|
"mean_token_accuracy": 0.14872195571660995,
|
|
"num_tokens": 4116804.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.470265865325928,
|
|
"epoch": 2.0924912978617605,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004836780802164016,
|
|
"loss": 6.2987,
|
|
"mean_token_accuracy": 0.1431171678006649,
|
|
"num_tokens": 4126658.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.455013799667358,
|
|
"epoch": 2.0974639482844357,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048353173643868064,
|
|
"loss": 6.0883,
|
|
"mean_token_accuracy": 0.1502581126987934,
|
|
"num_tokens": 4136323.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.409735441207886,
|
|
"epoch": 2.102436598707111,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004833847644367826,
|
|
"loss": 6.2264,
|
|
"mean_token_accuracy": 0.15136316567659377,
|
|
"num_tokens": 4145714.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.606991863250732,
|
|
"epoch": 2.107409249129786,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048323716465347615,
|
|
"loss": 6.2202,
|
|
"mean_token_accuracy": 0.14936506152153015,
|
|
"num_tokens": 4156059.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.423093795776367,
|
|
"epoch": 2.1123818995524615,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004830889375334215,
|
|
"loss": 6.2496,
|
|
"mean_token_accuracy": 0.14233115315437317,
|
|
"num_tokens": 4166990.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.468916702270508,
|
|
"epoch": 2.1173545499751367,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048294008352316875,
|
|
"loss": 6.238,
|
|
"mean_token_accuracy": 0.15066475570201873,
|
|
"num_tokens": 4175716.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.47969913482666,
|
|
"epoch": 2.122327200397812,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004827906030711564,
|
|
"loss": 6.2402,
|
|
"mean_token_accuracy": 0.14436411261558532,
|
|
"num_tokens": 4184832.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.308349561691284,
|
|
"epoch": 2.1272998508204872,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048264049662771053,
|
|
"loss": 6.0225,
|
|
"mean_token_accuracy": 0.16351697593927383,
|
|
"num_tokens": 4195678.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.463945198059082,
|
|
"epoch": 2.1322725012431625,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048248976464504266,
|
|
"loss": 6.18,
|
|
"mean_token_accuracy": 0.15189888030290605,
|
|
"num_tokens": 4205357.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.534983730316162,
|
|
"epoch": 2.1372451516658377,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048233840757724926,
|
|
"loss": 6.2837,
|
|
"mean_token_accuracy": 0.1422731302678585,
|
|
"num_tokens": 4215416.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.478677129745483,
|
|
"epoch": 2.142217802088513,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048218642588030943,
|
|
"loss": 6.3119,
|
|
"mean_token_accuracy": 0.13605532199144363,
|
|
"num_tokens": 4225894.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.54672269821167,
|
|
"epoch": 2.1471904525111887,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048203382001208453,
|
|
"loss": 6.2345,
|
|
"mean_token_accuracy": 0.15075795128941535,
|
|
"num_tokens": 4235973.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.457994222640991,
|
|
"epoch": 2.152163102933864,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048188059043231585,
|
|
"loss": 6.1567,
|
|
"mean_token_accuracy": 0.15078617855906487,
|
|
"num_tokens": 4246411.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.496095752716064,
|
|
"epoch": 2.157135753356539,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048172673760262414,
|
|
"loss": 6.2164,
|
|
"mean_token_accuracy": 0.14968225806951524,
|
|
"num_tokens": 4255754.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.439593505859375,
|
|
"epoch": 2.1621084037792144,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048157226198650735,
|
|
"loss": 6.1541,
|
|
"mean_token_accuracy": 0.14534053802490235,
|
|
"num_tokens": 4265263.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.571390581130982,
|
|
"epoch": 2.1670810542018897,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004814171640493398,
|
|
"loss": 6.3208,
|
|
"mean_token_accuracy": 0.14245227798819543,
|
|
"num_tokens": 4275262.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.426617956161499,
|
|
"epoch": 2.172053704624565,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004812614442583708,
|
|
"loss": 6.2288,
|
|
"mean_token_accuracy": 0.1431797794997692,
|
|
"num_tokens": 4285467.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.475995397567749,
|
|
"epoch": 2.17702635504724,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004811051030827227,
|
|
"loss": 6.1586,
|
|
"mean_token_accuracy": 0.14471708834171296,
|
|
"num_tokens": 4294680.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.461446905136109,
|
|
"epoch": 2.1819990054699154,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048094814099339006,
|
|
"loss": 6.2413,
|
|
"mean_token_accuracy": 0.14151463583111762,
|
|
"num_tokens": 4304605.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.370043134689331,
|
|
"epoch": 2.1869716558925907,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004807905584632381,
|
|
"loss": 6.0797,
|
|
"mean_token_accuracy": 0.15645029544830322,
|
|
"num_tokens": 4314112.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.33198618888855,
|
|
"epoch": 2.191944306315266,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048063235596700087,
|
|
"loss": 6.1265,
|
|
"mean_token_accuracy": 0.14964040219783784,
|
|
"num_tokens": 4323872.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.431909799575806,
|
|
"epoch": 2.196916956737941,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048047353398128056,
|
|
"loss": 6.0861,
|
|
"mean_token_accuracy": 0.15557749271392823,
|
|
"num_tokens": 4333002.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.442919683456421,
|
|
"epoch": 2.2018896071606164,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004803140929845452,
|
|
"loss": 6.2004,
|
|
"mean_token_accuracy": 0.14600576907396318,
|
|
"num_tokens": 4342012.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.454645442962646,
|
|
"epoch": 2.2068622575832917,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048015403345712796,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.146176515519619,
|
|
"num_tokens": 4351959.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.441257619857788,
|
|
"epoch": 2.2118349080059674,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004799933558812252,
|
|
"loss": 6.1308,
|
|
"mean_token_accuracy": 0.14963359832763673,
|
|
"num_tokens": 4361267.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.423764610290528,
|
|
"epoch": 2.2168075584286426,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047983206074089546,
|
|
"loss": 6.1904,
|
|
"mean_token_accuracy": 0.1483028084039688,
|
|
"num_tokens": 4370104.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.152798080444336,
|
|
"epoch": 2.221780208851318,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004796701485220576,
|
|
"loss": 5.9275,
|
|
"mean_token_accuracy": 0.17749467715620995,
|
|
"num_tokens": 4380724.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.269147348403931,
|
|
"epoch": 2.226752859273993,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004795076197124893,
|
|
"loss": 6.0093,
|
|
"mean_token_accuracy": 0.16409922391176224,
|
|
"num_tokens": 4390521.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.44737401008606,
|
|
"epoch": 2.2317255096966684,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047934447480182626,
|
|
"loss": 6.1985,
|
|
"mean_token_accuracy": 0.15125566273927687,
|
|
"num_tokens": 4400075.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.338036823272705,
|
|
"epoch": 2.2366981601193436,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004791807142815598,
|
|
"loss": 6.1475,
|
|
"mean_token_accuracy": 0.146401809155941,
|
|
"num_tokens": 4409913.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.463025665283203,
|
|
"epoch": 2.241670810542019,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047901633864503614,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.15036816596984864,
|
|
"num_tokens": 4418551.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.4096095085144045,
|
|
"epoch": 2.246643460964694,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047885134838745433,
|
|
"loss": 6.36,
|
|
"mean_token_accuracy": 0.142999829351902,
|
|
"num_tokens": 4428129.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.544295930862427,
|
|
"epoch": 2.2516161113873694,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004786857440058654,
|
|
"loss": 6.2492,
|
|
"mean_token_accuracy": 0.14261592403054238,
|
|
"num_tokens": 4437311.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.3263102054595945,
|
|
"epoch": 2.2565887618100446,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004785195259991701,
|
|
"loss": 6.0584,
|
|
"mean_token_accuracy": 0.1550035446882248,
|
|
"num_tokens": 4447387.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.3363120555877686,
|
|
"epoch": 2.26156141223272,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000478352694868118,
|
|
"loss": 6.2548,
|
|
"mean_token_accuracy": 0.142509326338768,
|
|
"num_tokens": 4456435.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.281604385375976,
|
|
"epoch": 2.266534062655395,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004781852511153058,
|
|
"loss": 6.1017,
|
|
"mean_token_accuracy": 0.15582385510206223,
|
|
"num_tokens": 4466464.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.406975030899048,
|
|
"epoch": 2.271506713078071,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047801719524517573,
|
|
"loss": 6.1907,
|
|
"mean_token_accuracy": 0.15059518292546273,
|
|
"num_tokens": 4476793.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.384379816055298,
|
|
"epoch": 2.276479363500746,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047784852776401403,
|
|
"loss": 6.149,
|
|
"mean_token_accuracy": 0.15004813000559808,
|
|
"num_tokens": 4485811.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.413690042495728,
|
|
"epoch": 2.2814520139234213,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004776792491799495,
|
|
"loss": 6.2864,
|
|
"mean_token_accuracy": 0.14758335873484613,
|
|
"num_tokens": 4495018.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.483861589431763,
|
|
"epoch": 2.2864246643460966,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047750936000295213,
|
|
"loss": 6.2308,
|
|
"mean_token_accuracy": 0.14389350041747093,
|
|
"num_tokens": 4505847.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.383129501342774,
|
|
"epoch": 2.291397314768772,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047733886074483115,
|
|
"loss": 6.2554,
|
|
"mean_token_accuracy": 0.1428247258067131,
|
|
"num_tokens": 4516104.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.283045530319214,
|
|
"epoch": 2.296369965191447,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004771677519192339,
|
|
"loss": 6.0002,
|
|
"mean_token_accuracy": 0.1624878838658333,
|
|
"num_tokens": 4524875.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.400923585891723,
|
|
"epoch": 2.3013426156141223,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004769960340416439,
|
|
"loss": 6.2688,
|
|
"mean_token_accuracy": 0.14108621925115586,
|
|
"num_tokens": 4533836.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.191251802444458,
|
|
"epoch": 2.3063152660367976,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047682370762937976,
|
|
"loss": 5.7457,
|
|
"mean_token_accuracy": 0.18288942873477937,
|
|
"num_tokens": 4544323.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.342076396942138,
|
|
"epoch": 2.311287916459473,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047665077320159315,
|
|
"loss": 6.1642,
|
|
"mean_token_accuracy": 0.1523999348282814,
|
|
"num_tokens": 4553811.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.2140544891357425,
|
|
"epoch": 2.316260566882148,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 0.00047647723127926765,
|
|
"loss": 6.0228,
|
|
"mean_token_accuracy": 0.17004263699054717,
|
|
"num_tokens": 4563098.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.4378420352935795,
|
|
"epoch": 2.3212332173048233,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004763030823852168,
|
|
"loss": 6.2858,
|
|
"mean_token_accuracy": 0.14585834592580796,
|
|
"num_tokens": 4573589.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.41130223274231,
|
|
"epoch": 2.3262058677274986,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004761283270440829,
|
|
"loss": 6.2011,
|
|
"mean_token_accuracy": 0.14975759536027908,
|
|
"num_tokens": 4583059.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.498157787322998,
|
|
"epoch": 2.331178518150174,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004759529657823349,
|
|
"loss": 6.3694,
|
|
"mean_token_accuracy": 0.14244077056646348,
|
|
"num_tokens": 4593371.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.411780071258545,
|
|
"epoch": 2.336151168572849,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047577699912826755,
|
|
"loss": 6.283,
|
|
"mean_token_accuracy": 0.14445912837982178,
|
|
"num_tokens": 4602862.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.397918319702148,
|
|
"epoch": 2.341123818995525,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047560042761199915,
|
|
"loss": 6.1501,
|
|
"mean_token_accuracy": 0.15046437829732895,
|
|
"num_tokens": 4612827.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.373718881607056,
|
|
"epoch": 2.3460964694182,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004754232517654704,
|
|
"loss": 6.2178,
|
|
"mean_token_accuracy": 0.14718493670225144,
|
|
"num_tokens": 4621706.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.439148187637329,
|
|
"epoch": 2.3510691198408753,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004752454721224424,
|
|
"loss": 6.1576,
|
|
"mean_token_accuracy": 0.14772356003522874,
|
|
"num_tokens": 4630878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.45961627960205,
|
|
"epoch": 2.3560417702635505,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047506708921849535,
|
|
"loss": 6.2997,
|
|
"mean_token_accuracy": 0.14015229567885398,
|
|
"num_tokens": 4640582.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.363547325134277,
|
|
"epoch": 2.361014420686226,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004748881035910269,
|
|
"loss": 6.2217,
|
|
"mean_token_accuracy": 0.15062569975852966,
|
|
"num_tokens": 4650799.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.413719129562378,
|
|
"epoch": 2.365987071108901,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004747085157792504,
|
|
"loss": 6.1592,
|
|
"mean_token_accuracy": 0.15538667291402816,
|
|
"num_tokens": 4660566.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.415573406219482,
|
|
"epoch": 2.3709597215315763,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004745283263241933,
|
|
"loss": 6.2586,
|
|
"mean_token_accuracy": 0.14485761076211928,
|
|
"num_tokens": 4670234.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.433115100860595,
|
|
"epoch": 2.3759323719542516,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004743475357686957,
|
|
"loss": 6.1999,
|
|
"mean_token_accuracy": 0.14847781062126159,
|
|
"num_tokens": 4681114.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.2774864673614506,
|
|
"epoch": 2.380905022376927,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047416614465740843,
|
|
"loss": 6.0094,
|
|
"mean_token_accuracy": 0.17164644598960876,
|
|
"num_tokens": 4691970.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.327695655822754,
|
|
"epoch": 2.385877672799602,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047398415353679167,
|
|
"loss": 6.1969,
|
|
"mean_token_accuracy": 0.14662425667047502,
|
|
"num_tokens": 4701723.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.360439395904541,
|
|
"epoch": 2.3908503232222773,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004738015629551132,
|
|
"loss": 6.173,
|
|
"mean_token_accuracy": 0.15016620755195617,
|
|
"num_tokens": 4711356.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.403611898422241,
|
|
"epoch": 2.395822973644953,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047361837346244655,
|
|
"loss": 6.2228,
|
|
"mean_token_accuracy": 0.15572515726089478,
|
|
"num_tokens": 4721765.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.342036199569702,
|
|
"epoch": 2.4007956240676283,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004734345856106698,
|
|
"loss": 6.104,
|
|
"mean_token_accuracy": 0.1529115065932274,
|
|
"num_tokens": 4731678.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.359368753433228,
|
|
"epoch": 2.4057682744903035,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047325019995346355,
|
|
"loss": 6.1151,
|
|
"mean_token_accuracy": 0.152193820476532,
|
|
"num_tokens": 4740960.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.300037288665772,
|
|
"epoch": 2.4107409249129788,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004730652170463091,
|
|
"loss": 6.2634,
|
|
"mean_token_accuracy": 0.14104650765657425,
|
|
"num_tokens": 4751753.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.398259496688842,
|
|
"epoch": 2.415713575335654,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004728796374464875,
|
|
"loss": 6.1169,
|
|
"mean_token_accuracy": 0.15501195341348648,
|
|
"num_tokens": 4761274.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.37506217956543,
|
|
"epoch": 2.4206862257583293,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047269346171307716,
|
|
"loss": 6.2809,
|
|
"mean_token_accuracy": 0.14763263687491418,
|
|
"num_tokens": 4770852.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.450977325439453,
|
|
"epoch": 2.4256588761810045,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004725066904069523,
|
|
"loss": 6.1225,
|
|
"mean_token_accuracy": 0.15091280341148378,
|
|
"num_tokens": 4780857.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.27843542098999,
|
|
"epoch": 2.4306315266036798,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004723193240907814,
|
|
"loss": 6.0938,
|
|
"mean_token_accuracy": 0.14641971737146378,
|
|
"num_tokens": 4790018.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.343325138092041,
|
|
"epoch": 2.435604177026355,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004721313633290257,
|
|
"loss": 6.1805,
|
|
"mean_token_accuracy": 0.1517829641699791,
|
|
"num_tokens": 4799747.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.434195137023925,
|
|
"epoch": 2.4405768274490303,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000471942808687937,
|
|
"loss": 6.2111,
|
|
"mean_token_accuracy": 0.1466923885047436,
|
|
"num_tokens": 4809710.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.313620233535767,
|
|
"epoch": 2.4455494778717055,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047175366073555654,
|
|
"loss": 6.1618,
|
|
"mean_token_accuracy": 0.15373150259256363,
|
|
"num_tokens": 4819387.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.303595972061157,
|
|
"epoch": 2.4505221282943808,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004715639200417125,
|
|
"loss": 6.1211,
|
|
"mean_token_accuracy": 0.14637587666511537,
|
|
"num_tokens": 4829541.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.3491682529449465,
|
|
"epoch": 2.455494778717056,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004713735871780192,
|
|
"loss": 6.1111,
|
|
"mean_token_accuracy": 0.156888185441494,
|
|
"num_tokens": 4838699.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.350799989700318,
|
|
"epoch": 2.4604674291397313,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004711826627178747,
|
|
"loss": 6.2429,
|
|
"mean_token_accuracy": 0.14678731113672255,
|
|
"num_tokens": 4848295.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.335339164733886,
|
|
"epoch": 2.4654400795624065,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047099114723645944,
|
|
"loss": 6.1023,
|
|
"mean_token_accuracy": 0.1573378786444664,
|
|
"num_tokens": 4857649.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.369188547134399,
|
|
"epoch": 2.470412729985082,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047079904131073434,
|
|
"loss": 6.1453,
|
|
"mean_token_accuracy": 0.1550306923687458,
|
|
"num_tokens": 4867332.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.371166753768921,
|
|
"epoch": 2.4753853804077575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004706063455194389,
|
|
"loss": 6.2051,
|
|
"mean_token_accuracy": 0.14223534017801284,
|
|
"num_tokens": 4878726.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.364006805419922,
|
|
"epoch": 2.4803580308304327,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047041306044308986,
|
|
"loss": 6.1844,
|
|
"mean_token_accuracy": 0.14895544126629828,
|
|
"num_tokens": 4888030.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.35819182395935,
|
|
"epoch": 2.485330681253108,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004702191866639792,
|
|
"loss": 6.2028,
|
|
"mean_token_accuracy": 0.14386930018663407,
|
|
"num_tokens": 4897191.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.485330681253108,
|
|
"eval_entropy": 6.212806101768248,
|
|
"eval_loss": 6.726681232452393,
|
|
"eval_mean_token_accuracy": 0.13171183960812707,
|
|
"eval_num_tokens": 4897191.0,
|
|
"eval_runtime": 12.996,
|
|
"eval_samples_per_second": 266.62,
|
|
"eval_steps_per_second": 33.395,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.342361116409302,
|
|
"epoch": 2.490303331675783,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004700247247661726,
|
|
"loss": 6.1875,
|
|
"mean_token_accuracy": 0.1494624748826027,
|
|
"num_tokens": 4907154.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.185926723480224,
|
|
"epoch": 2.4952759820984585,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004698296753355072,
|
|
"loss": 5.822,
|
|
"mean_token_accuracy": 0.17507386356592178,
|
|
"num_tokens": 4917582.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.222843074798584,
|
|
"epoch": 2.5002486325211337,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004696340389595903,
|
|
"loss": 6.1136,
|
|
"mean_token_accuracy": 0.15697015970945358,
|
|
"num_tokens": 4927473.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.338209581375122,
|
|
"epoch": 2.505221282943809,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046943781622779733,
|
|
"loss": 6.2545,
|
|
"mean_token_accuracy": 0.14296423271298409,
|
|
"num_tokens": 4938037.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.359233474731445,
|
|
"epoch": 2.510193933366484,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046924100773127055,
|
|
"loss": 6.2179,
|
|
"mean_token_accuracy": 0.1430736795067787,
|
|
"num_tokens": 4948009.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.323881912231445,
|
|
"epoch": 2.5151665837891595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046904361406291633,
|
|
"loss": 6.1876,
|
|
"mean_token_accuracy": 0.14997006952762604,
|
|
"num_tokens": 4957971.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.4062830924987795,
|
|
"epoch": 2.520139234211835,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004688456358174044,
|
|
"loss": 6.1737,
|
|
"mean_token_accuracy": 0.13893459215760232,
|
|
"num_tokens": 4968040.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.359023714065552,
|
|
"epoch": 2.5251118846345104,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004686470735911653,
|
|
"loss": 6.099,
|
|
"mean_token_accuracy": 0.15624479651451112,
|
|
"num_tokens": 4977142.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.336380481719971,
|
|
"epoch": 2.5300845350571857,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004684479279823891,
|
|
"loss": 6.2671,
|
|
"mean_token_accuracy": 0.14676804095506668,
|
|
"num_tokens": 4986726.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.389647340774536,
|
|
"epoch": 2.535057185479861,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004682481995910232,
|
|
"loss": 6.1624,
|
|
"mean_token_accuracy": 0.14428184404969216,
|
|
"num_tokens": 4996690.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.371994686126709,
|
|
"epoch": 2.540029835902536,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046804788901877084,
|
|
"loss": 6.2152,
|
|
"mean_token_accuracy": 0.15031069070100783,
|
|
"num_tokens": 5005804.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.311955547332763,
|
|
"epoch": 2.5450024863252114,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046784699686908886,
|
|
"loss": 6.1814,
|
|
"mean_token_accuracy": 0.15370430797338486,
|
|
"num_tokens": 5015666.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.297400856018067,
|
|
"epoch": 2.5499751367478867,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046764552374718656,
|
|
"loss": 6.1105,
|
|
"mean_token_accuracy": 0.15727180242538452,
|
|
"num_tokens": 5025897.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.322667932510376,
|
|
"epoch": 2.554947787170562,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046744347026002315,
|
|
"loss": 6.1584,
|
|
"mean_token_accuracy": 0.15125731974840165,
|
|
"num_tokens": 5035195.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.316974830627442,
|
|
"epoch": 2.559920437593237,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004672408370163066,
|
|
"loss": 6.2099,
|
|
"mean_token_accuracy": 0.15157809108495712,
|
|
"num_tokens": 5044472.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.299518203735351,
|
|
"epoch": 2.5648930880159124,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000467037624626491,
|
|
"loss": 6.2525,
|
|
"mean_token_accuracy": 0.14488816410303115,
|
|
"num_tokens": 5053593.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.353642845153809,
|
|
"epoch": 2.5698657384385877,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046683383370277554,
|
|
"loss": 6.1999,
|
|
"mean_token_accuracy": 0.14531988576054572,
|
|
"num_tokens": 5063604.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.303402614593506,
|
|
"epoch": 2.574838388861263,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046662946485910226,
|
|
"loss": 6.09,
|
|
"mean_token_accuracy": 0.14906830638647078,
|
|
"num_tokens": 5073013.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.286452150344848,
|
|
"epoch": 2.579811039283938,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000466424518711154,
|
|
"loss": 6.1773,
|
|
"mean_token_accuracy": 0.15088475197553636,
|
|
"num_tokens": 5082590.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.383158779144287,
|
|
"epoch": 2.5847836897066134,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004662189958763531,
|
|
"loss": 6.1582,
|
|
"mean_token_accuracy": 0.14844950437545776,
|
|
"num_tokens": 5091860.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.335248327255249,
|
|
"epoch": 2.5897563401292887,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004660128969738591,
|
|
"loss": 6.072,
|
|
"mean_token_accuracy": 0.16079305559396745,
|
|
"num_tokens": 5101164.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.212764739990234,
|
|
"epoch": 2.594728990551964,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004658062226245668,
|
|
"loss": 6.1704,
|
|
"mean_token_accuracy": 0.14848436266183854,
|
|
"num_tokens": 5112885.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.348287916183471,
|
|
"epoch": 2.5997016409746396,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004655989734511052,
|
|
"loss": 6.1581,
|
|
"mean_token_accuracy": 0.15465263426303863,
|
|
"num_tokens": 5124063.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.164831733703613,
|
|
"epoch": 2.604674291397315,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004653911500778343,
|
|
"loss": 5.996,
|
|
"mean_token_accuracy": 0.16380439549684525,
|
|
"num_tokens": 5134286.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.2939306735992435,
|
|
"epoch": 2.60964694181999,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004651827531308443,
|
|
"loss": 6.2519,
|
|
"mean_token_accuracy": 0.14676303565502166,
|
|
"num_tokens": 5144421.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.46519365310669,
|
|
"epoch": 2.6146195922426654,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004649737832379534,
|
|
"loss": 6.172,
|
|
"mean_token_accuracy": 0.151572185754776,
|
|
"num_tokens": 5153466.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.245690155029297,
|
|
"epoch": 2.6195922426653406,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004647642410287057,
|
|
"loss": 6.2571,
|
|
"mean_token_accuracy": 0.14614539518952369,
|
|
"num_tokens": 5164034.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.315093851089477,
|
|
"epoch": 2.624564893088016,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004645541271343696,
|
|
"loss": 6.1746,
|
|
"mean_token_accuracy": 0.15348961800336838,
|
|
"num_tokens": 5173867.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.3680578708648685,
|
|
"epoch": 2.629537543510691,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046434344218793566,
|
|
"loss": 6.2472,
|
|
"mean_token_accuracy": 0.14953457862138747,
|
|
"num_tokens": 5183182.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.317287302017212,
|
|
"epoch": 2.6345101939333664,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046413218682411485,
|
|
"loss": 6.1787,
|
|
"mean_token_accuracy": 0.1503915973007679,
|
|
"num_tokens": 5192537.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.41108922958374,
|
|
"epoch": 2.6394828443560416,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004639203616793365,
|
|
"loss": 6.149,
|
|
"mean_token_accuracy": 0.15816644579172134,
|
|
"num_tokens": 5201879.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.329562854766846,
|
|
"epoch": 2.6444554947787173,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046370796739174657,
|
|
"loss": 6.1664,
|
|
"mean_token_accuracy": 0.1496497370302677,
|
|
"num_tokens": 5211555.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.23956151008606,
|
|
"epoch": 2.6494281452013926,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046349500460120556,
|
|
"loss": 6.0716,
|
|
"mean_token_accuracy": 0.15945928990840913,
|
|
"num_tokens": 5220895.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.398822927474976,
|
|
"epoch": 2.654400795624068,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046328147394928684,
|
|
"loss": 6.2225,
|
|
"mean_token_accuracy": 0.14685262143611907,
|
|
"num_tokens": 5230101.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.388316822052002,
|
|
"epoch": 2.659373446046743,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004630673760792741,
|
|
"loss": 6.2074,
|
|
"mean_token_accuracy": 0.1387428343296051,
|
|
"num_tokens": 5239472.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.231325578689575,
|
|
"epoch": 2.6643460964694183,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004628527116361603,
|
|
"loss": 6.1486,
|
|
"mean_token_accuracy": 0.15218093991279602,
|
|
"num_tokens": 5248823.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.295870733261109,
|
|
"epoch": 2.6693187468920936,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004626374812666449,
|
|
"loss": 6.1647,
|
|
"mean_token_accuracy": 0.1490417644381523,
|
|
"num_tokens": 5258551.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.286609315872193,
|
|
"epoch": 2.674291397314769,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046242168561913256,
|
|
"loss": 5.9462,
|
|
"mean_token_accuracy": 0.16948021054267884,
|
|
"num_tokens": 5267933.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.385824251174927,
|
|
"epoch": 2.679264047737444,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046220532534373067,
|
|
"loss": 6.209,
|
|
"mean_token_accuracy": 0.15343319475650788,
|
|
"num_tokens": 5277856.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.335033226013183,
|
|
"epoch": 2.6842366981601193,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004619884010922477,
|
|
"loss": 6.1923,
|
|
"mean_token_accuracy": 0.1467955894768238,
|
|
"num_tokens": 5288127.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.342468738555908,
|
|
"epoch": 2.6892093485827946,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004617709135181913,
|
|
"loss": 6.1771,
|
|
"mean_token_accuracy": 0.14629826247692107,
|
|
"num_tokens": 5297143.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.2246088027954105,
|
|
"epoch": 2.69418199900547,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046155286327676594,
|
|
"loss": 6.1227,
|
|
"mean_token_accuracy": 0.14895061925053596,
|
|
"num_tokens": 5307067.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.265856122970581,
|
|
"epoch": 2.699154649428145,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046133425102487134,
|
|
"loss": 6.0995,
|
|
"mean_token_accuracy": 0.1525820627808571,
|
|
"num_tokens": 5317471.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.3970927715301515,
|
|
"epoch": 2.7041272998508203,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046111507742110035,
|
|
"loss": 6.161,
|
|
"mean_token_accuracy": 0.15613785684108733,
|
|
"num_tokens": 5326803.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.249770259857177,
|
|
"epoch": 2.7090999502734956,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000460895343125737,
|
|
"loss": 6.1525,
|
|
"mean_token_accuracy": 0.1544552817940712,
|
|
"num_tokens": 5336342.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.13524808883667,
|
|
"epoch": 2.714072600696171,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046067504880075425,
|
|
"loss": 5.981,
|
|
"mean_token_accuracy": 0.17633682042360305,
|
|
"num_tokens": 5345694.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.31252155303955,
|
|
"epoch": 2.719045251118846,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004604541951098122,
|
|
"loss": 6.0507,
|
|
"mean_token_accuracy": 0.16010746657848357,
|
|
"num_tokens": 5355277.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.235470962524414,
|
|
"epoch": 2.724017901541522,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004602327827182565,
|
|
"loss": 6.1092,
|
|
"mean_token_accuracy": 0.1569782666862011,
|
|
"num_tokens": 5364650.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.291445255279541,
|
|
"epoch": 2.728990551964197,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004600108122931156,
|
|
"loss": 6.0605,
|
|
"mean_token_accuracy": 0.1518268406391144,
|
|
"num_tokens": 5373151.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.258221912384033,
|
|
"epoch": 2.7339632023868723,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00045978828450309907,
|
|
"loss": 6.2039,
|
|
"mean_token_accuracy": 0.15100904777646065,
|
|
"num_tokens": 5383093.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.3790418148040775,
|
|
"epoch": 2.7389358528095475,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 0.0004595652000185958,
|
|
"loss": 6.0765,
|
|
"mean_token_accuracy": 0.15718968361616134,
|
|
"num_tokens": 5392060.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.334905433654785,
|
|
"epoch": 2.743908503232223,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045934155951167163,
|
|
"loss": 6.2231,
|
|
"mean_token_accuracy": 0.1474114865064621,
|
|
"num_tokens": 5402119.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.299063777923584,
|
|
"epoch": 2.748881153654898,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004591173636560675,
|
|
"loss": 6.1342,
|
|
"mean_token_accuracy": 0.15938253104686737,
|
|
"num_tokens": 5411804.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.292401647567749,
|
|
"epoch": 2.7538538040775733,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004588926131271974,
|
|
"loss": 6.1958,
|
|
"mean_token_accuracy": 0.14893215149641037,
|
|
"num_tokens": 5423808.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.301559066772461,
|
|
"epoch": 2.7588264545002485,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045866730860214655,
|
|
"loss": 6.1608,
|
|
"mean_token_accuracy": 0.1546696111559868,
|
|
"num_tokens": 5432154.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.162888717651367,
|
|
"epoch": 2.763799104922924,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004584414507596687,
|
|
"loss": 5.9041,
|
|
"mean_token_accuracy": 0.17434341013431548,
|
|
"num_tokens": 5443006.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.258318090438843,
|
|
"epoch": 2.768771755345599,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045821504028018494,
|
|
"loss": 6.0718,
|
|
"mean_token_accuracy": 0.1543208807706833,
|
|
"num_tokens": 5453596.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.3162116527557375,
|
|
"epoch": 2.7737444057682747,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004579880778457811,
|
|
"loss": 6.1971,
|
|
"mean_token_accuracy": 0.14643352553248407,
|
|
"num_tokens": 5463048.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.267004823684692,
|
|
"epoch": 2.77871705619095,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004577605641402056,
|
|
"loss": 6.134,
|
|
"mean_token_accuracy": 0.1571312204003334,
|
|
"num_tokens": 5472409.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.290575981140137,
|
|
"epoch": 2.7836897066136252,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004575324998488681,
|
|
"loss": 6.0568,
|
|
"mean_token_accuracy": 0.15990764200687407,
|
|
"num_tokens": 5481380.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.2788111686706545,
|
|
"epoch": 2.7886623570363005,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004573038856588364,
|
|
"loss": 6.2577,
|
|
"mean_token_accuracy": 0.14428368359804153,
|
|
"num_tokens": 5491157.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.303381490707397,
|
|
"epoch": 2.7936350074589757,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004570747222588354,
|
|
"loss": 6.15,
|
|
"mean_token_accuracy": 0.1519258663058281,
|
|
"num_tokens": 5500856.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.306287527084351,
|
|
"epoch": 2.798607657881651,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045684501033924443,
|
|
"loss": 6.0499,
|
|
"mean_token_accuracy": 0.15974352210760118,
|
|
"num_tokens": 5509654.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.215396404266357,
|
|
"epoch": 2.8035803083043263,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045661475059209503,
|
|
"loss": 6.0408,
|
|
"mean_token_accuracy": 0.1627000704407692,
|
|
"num_tokens": 5519009.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.224809455871582,
|
|
"epoch": 2.8085529587270015,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004563839437110695,
|
|
"loss": 6.0609,
|
|
"mean_token_accuracy": 0.16025128960609436,
|
|
"num_tokens": 5528316.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.346328353881836,
|
|
"epoch": 2.8135256091496768,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045615259039149826,
|
|
"loss": 6.1754,
|
|
"mean_token_accuracy": 0.15645836144685746,
|
|
"num_tokens": 5538416.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.2237225532531735,
|
|
"epoch": 2.818498259572352,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045592069133035783,
|
|
"loss": 6.1377,
|
|
"mean_token_accuracy": 0.15551433488726615,
|
|
"num_tokens": 5548600.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.298158884048462,
|
|
"epoch": 2.8234709099950273,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004556882472262691,
|
|
"loss": 6.232,
|
|
"mean_token_accuracy": 0.14831071943044663,
|
|
"num_tokens": 5559454.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.31885929107666,
|
|
"epoch": 2.8284435604177025,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004554552587794948,
|
|
"loss": 6.1581,
|
|
"mean_token_accuracy": 0.14877900034189223,
|
|
"num_tokens": 5567748.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.279418277740478,
|
|
"epoch": 2.8334162108403778,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004552217266919375,
|
|
"loss": 6.25,
|
|
"mean_token_accuracy": 0.14875746816396712,
|
|
"num_tokens": 5577617.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.329247093200683,
|
|
"epoch": 2.838388861263053,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004549876516671376,
|
|
"loss": 6.2173,
|
|
"mean_token_accuracy": 0.14989273250102997,
|
|
"num_tokens": 5586600.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.331892013549805,
|
|
"epoch": 2.8433615116857283,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045475303441027114,
|
|
"loss": 6.0628,
|
|
"mean_token_accuracy": 0.16511600315570832,
|
|
"num_tokens": 5596526.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.272082901000976,
|
|
"epoch": 2.8483341621084035,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004545178756281479,
|
|
"loss": 6.1327,
|
|
"mean_token_accuracy": 0.1555761381983757,
|
|
"num_tokens": 5605780.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.242947769165039,
|
|
"epoch": 2.853306812531079,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045428217602920876,
|
|
"loss": 6.1603,
|
|
"mean_token_accuracy": 0.14795016199350358,
|
|
"num_tokens": 5616109.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.269103336334228,
|
|
"epoch": 2.8582794629537545,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045404593632352385,
|
|
"loss": 6.1324,
|
|
"mean_token_accuracy": 0.15838288515806198,
|
|
"num_tokens": 5625423.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.3219382762908936,
|
|
"epoch": 2.8632521133764297,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0004538091572227909,
|
|
"loss": 6.2055,
|
|
"mean_token_accuracy": 0.15248353332281112,
|
|
"num_tokens": 5634538.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.178640460968017,
|
|
"epoch": 2.868224763799105,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004535718394403319,
|
|
"loss": 5.9766,
|
|
"mean_token_accuracy": 0.17341183871030807,
|
|
"num_tokens": 5644087.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.129975605010986,
|
|
"epoch": 2.87319741422178,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004533339836910924,
|
|
"loss": 5.9945,
|
|
"mean_token_accuracy": 0.16715046614408494,
|
|
"num_tokens": 5654241.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.223379421234131,
|
|
"epoch": 2.8781700646444555,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004530955906916382,
|
|
"loss": 6.057,
|
|
"mean_token_accuracy": 0.15955688506364823,
|
|
"num_tokens": 5663827.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.271706581115723,
|
|
"epoch": 2.8831427150671307,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004528566611601536,
|
|
"loss": 6.1512,
|
|
"mean_token_accuracy": 0.15101373866200446,
|
|
"num_tokens": 5673427.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.222475481033325,
|
|
"epoch": 2.888115365489806,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004526171958164396,
|
|
"loss": 6.0228,
|
|
"mean_token_accuracy": 0.1544105350971222,
|
|
"num_tokens": 5684762.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.151531457901001,
|
|
"epoch": 2.893088015912481,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045237719538191107,
|
|
"loss": 6.1463,
|
|
"mean_token_accuracy": 0.15518633425235748,
|
|
"num_tokens": 5694938.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.249429655075073,
|
|
"epoch": 2.898060666335157,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000452136660579595,
|
|
"loss": 6.0412,
|
|
"mean_token_accuracy": 0.16044676899909974,
|
|
"num_tokens": 5704910.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.263382387161255,
|
|
"epoch": 2.903033316757832,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045189559213412825,
|
|
"loss": 6.1282,
|
|
"mean_token_accuracy": 0.1527381181716919,
|
|
"num_tokens": 5714325.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.240692567825318,
|
|
"epoch": 2.9080059671805074,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004516539907717553,
|
|
"loss": 6.0564,
|
|
"mean_token_accuracy": 0.1665084958076477,
|
|
"num_tokens": 5724382.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.218776559829712,
|
|
"epoch": 2.9129786176031827,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045141185722032617,
|
|
"loss": 6.102,
|
|
"mean_token_accuracy": 0.15742503553628923,
|
|
"num_tokens": 5734277.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.28412356376648,
|
|
"epoch": 2.917951268025858,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045116919220929407,
|
|
"loss": 6.1075,
|
|
"mean_token_accuracy": 0.14839262813329696,
|
|
"num_tokens": 5742772.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.162795639038086,
|
|
"epoch": 2.922923918448533,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004509259964697133,
|
|
"loss": 6.1237,
|
|
"mean_token_accuracy": 0.15320040434598922,
|
|
"num_tokens": 5754011.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.327255058288574,
|
|
"epoch": 2.9278965688712084,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000450682270734237,
|
|
"loss": 6.2212,
|
|
"mean_token_accuracy": 0.1451749987900257,
|
|
"num_tokens": 5765445.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.266171455383301,
|
|
"epoch": 2.9328692192938837,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004504380157371152,
|
|
"loss": 6.1385,
|
|
"mean_token_accuracy": 0.1579059898853302,
|
|
"num_tokens": 5775529.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.278435897827149,
|
|
"epoch": 2.937841869716559,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045019323221419204,
|
|
"loss": 6.2697,
|
|
"mean_token_accuracy": 0.14517680332064628,
|
|
"num_tokens": 5785594.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.222329425811767,
|
|
"epoch": 2.942814520139234,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044994792090290425,
|
|
"loss": 5.9411,
|
|
"mean_token_accuracy": 0.17010942548513414,
|
|
"num_tokens": 5796071.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.17365460395813,
|
|
"epoch": 2.9477871705619094,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044970208254227836,
|
|
"loss": 6.1061,
|
|
"mean_token_accuracy": 0.16210683137178422,
|
|
"num_tokens": 5805127.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.282905101776123,
|
|
"epoch": 2.9527598209845847,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044945571787292866,
|
|
"loss": 6.1842,
|
|
"mean_token_accuracy": 0.15130163878202438,
|
|
"num_tokens": 5815001.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.251187705993653,
|
|
"epoch": 2.95773247140726,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004492088276370552,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.14748269841074943,
|
|
"num_tokens": 5826915.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.281232452392578,
|
|
"epoch": 2.962705121829935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004489614125784411,
|
|
"loss": 6.1242,
|
|
"mean_token_accuracy": 0.1522660583257675,
|
|
"num_tokens": 5837444.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.148655080795288,
|
|
"epoch": 2.9676777722526104,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004487134734424509,
|
|
"loss": 6.0328,
|
|
"mean_token_accuracy": 0.1518077090382576,
|
|
"num_tokens": 5846514.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.2909081935882565,
|
|
"epoch": 2.9726504226752857,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004484650109760276,
|
|
"loss": 6.2748,
|
|
"mean_token_accuracy": 0.14825819209218025,
|
|
"num_tokens": 5857227.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.3236823081970215,
|
|
"epoch": 2.9776230730979614,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004482160259276911,
|
|
"loss": 6.2522,
|
|
"mean_token_accuracy": 0.14799594432115554,
|
|
"num_tokens": 5866877.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.2676714897155765,
|
|
"epoch": 2.9825957235206366,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004479665190475355,
|
|
"loss": 6.127,
|
|
"mean_token_accuracy": 0.15272125601768494,
|
|
"num_tokens": 5877127.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9825957235206366,
|
|
"eval_entropy": 6.129418342344223,
|
|
"eval_loss": 6.622411251068115,
|
|
"eval_mean_token_accuracy": 0.1357593302387521,
|
|
"eval_num_tokens": 5877127.0,
|
|
"eval_runtime": 10.3913,
|
|
"eval_samples_per_second": 333.453,
|
|
"eval_steps_per_second": 41.766,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 9643545851904000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|