de25a3f955
Model: fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
2057 lines
54 KiB
JSON
2057 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.9945300845350572,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 3220538609664000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|