de25a3f955
Model: fpadovani/tur-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
13178 lines
360 KiB
JSON
13178 lines
360 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.461461959224266,
|
|
"eval_steps": 500,
|
|
"global_step": 6500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.235357761383057,
|
|
"epoch": 2.987568373943312,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044771649108722695,
|
|
"loss": 6.1371,
|
|
"mean_token_accuracy": 0.15235839933156967,
|
|
"num_tokens": 5886434.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.248644781112671,
|
|
"epoch": 2.992541024365987,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004474659428000014,
|
|
"loss": 6.1344,
|
|
"mean_token_accuracy": 0.15656718909740447,
|
|
"num_tokens": 5895462.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.2803534984588625,
|
|
"epoch": 2.9975136747886624,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044721487494066254,
|
|
"loss": 6.0961,
|
|
"mean_token_accuracy": 0.1535220265388489,
|
|
"num_tokens": 5905274.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.144067605336507,
|
|
"epoch": 3.00198906016907,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000446963288265579,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.17379982107215458,
|
|
"num_tokens": 5914164.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.233295345306397,
|
|
"epoch": 3.0069617105917454,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044671118353268257,
|
|
"loss": 5.6706,
|
|
"mean_token_accuracy": 0.16814168095588683,
|
|
"num_tokens": 5923159.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.154281520843506,
|
|
"epoch": 3.0119343610144207,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044645856150146593,
|
|
"loss": 5.6456,
|
|
"mean_token_accuracy": 0.17532947957515715,
|
|
"num_tokens": 5932708.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.238128805160523,
|
|
"epoch": 3.016907011437096,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044620542293297974,
|
|
"loss": 5.6583,
|
|
"mean_token_accuracy": 0.16875627487897873,
|
|
"num_tokens": 5941370.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.142639303207398,
|
|
"epoch": 3.021879661859771,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004459517685898312,
|
|
"loss": 5.7012,
|
|
"mean_token_accuracy": 0.17632030993700026,
|
|
"num_tokens": 5950567.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.22289252281189,
|
|
"epoch": 3.0268523122824464,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00044569759923618113,
|
|
"loss": 5.777,
|
|
"mean_token_accuracy": 0.165495465695858,
|
|
"num_tokens": 5959242.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.239925289154053,
|
|
"epoch": 3.0318249627051217,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004454429156377419,
|
|
"loss": 5.7633,
|
|
"mean_token_accuracy": 0.17161031067371368,
|
|
"num_tokens": 5968365.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.166422128677368,
|
|
"epoch": 3.036797613127797,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004451877185617751,
|
|
"loss": 5.7402,
|
|
"mean_token_accuracy": 0.16547627598047257,
|
|
"num_tokens": 5978407.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.993821334838867,
|
|
"epoch": 3.041770263550472,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004449320087770894,
|
|
"loss": 5.4769,
|
|
"mean_token_accuracy": 0.19580196142196654,
|
|
"num_tokens": 5988865.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.163328218460083,
|
|
"epoch": 3.046742913973148,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004446757870540376,
|
|
"loss": 5.652,
|
|
"mean_token_accuracy": 0.17390889525413514,
|
|
"num_tokens": 5998873.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.154457044601441,
|
|
"epoch": 3.051715564395823,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004444190541645153,
|
|
"loss": 5.6993,
|
|
"mean_token_accuracy": 0.16924364119768143,
|
|
"num_tokens": 6007021.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.024441909790039,
|
|
"epoch": 3.0566882148184984,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.00044416181088195776,
|
|
"loss": 5.5318,
|
|
"mean_token_accuracy": 0.18400538265705108,
|
|
"num_tokens": 6015884.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.114338970184326,
|
|
"epoch": 3.0616608652411736,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004439040579813379,
|
|
"loss": 5.7725,
|
|
"mean_token_accuracy": 0.16868721544742585,
|
|
"num_tokens": 6026110.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.007745170593262,
|
|
"epoch": 3.066633515663849,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044364579623916385,
|
|
"loss": 5.544,
|
|
"mean_token_accuracy": 0.18894758224487304,
|
|
"num_tokens": 6037479.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.161260604858398,
|
|
"epoch": 3.071606166086524,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004433870264334767,
|
|
"loss": 5.6831,
|
|
"mean_token_accuracy": 0.16961449682712554,
|
|
"num_tokens": 6046743.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.092568731307983,
|
|
"epoch": 3.0765788165091994,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004431277493438484,
|
|
"loss": 5.7801,
|
|
"mean_token_accuracy": 0.16989244520664215,
|
|
"num_tokens": 6056982.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.166731929779052,
|
|
"epoch": 3.0815514669318746,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004428679657513789,
|
|
"loss": 5.7533,
|
|
"mean_token_accuracy": 0.16614997535943984,
|
|
"num_tokens": 6066016.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.066860151290894,
|
|
"epoch": 3.08652411735455,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004426076764386939,
|
|
"loss": 5.6727,
|
|
"mean_token_accuracy": 0.1643929287791252,
|
|
"num_tokens": 6075149.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.141622304916382,
|
|
"epoch": 3.091496767777225,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00044234688218994293,
|
|
"loss": 5.629,
|
|
"mean_token_accuracy": 0.17735926508903505,
|
|
"num_tokens": 6085736.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.0378796577453615,
|
|
"epoch": 3.0964694181999004,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004420855837907967,
|
|
"loss": 5.5717,
|
|
"mean_token_accuracy": 0.18653376698493956,
|
|
"num_tokens": 6096557.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.229071712493896,
|
|
"epoch": 3.1014420686225757,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044182378202844427,
|
|
"loss": 5.8708,
|
|
"mean_token_accuracy": 0.15957390666007995,
|
|
"num_tokens": 6105735.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.208705186843872,
|
|
"epoch": 3.106414719045251,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00044156147769159183,
|
|
"loss": 5.7223,
|
|
"mean_token_accuracy": 0.17071411162614822,
|
|
"num_tokens": 6114132.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.126375484466553,
|
|
"epoch": 3.1113873694679266,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004412986715704591,
|
|
"loss": 5.7394,
|
|
"mean_token_accuracy": 0.16539884358644485,
|
|
"num_tokens": 6124341.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.070566654205322,
|
|
"epoch": 3.116360019890602,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004410353644567777,
|
|
"loss": 5.6175,
|
|
"mean_token_accuracy": 0.1764797881245613,
|
|
"num_tokens": 6134418.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.107773113250732,
|
|
"epoch": 3.121332670313277,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044077155714378836,
|
|
"loss": 5.7152,
|
|
"mean_token_accuracy": 0.16651442646980286,
|
|
"num_tokens": 6143659.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.17090859413147,
|
|
"epoch": 3.1263053207359524,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004405072504262391,
|
|
"loss": 5.7469,
|
|
"mean_token_accuracy": 0.16660532504320144,
|
|
"num_tokens": 6152706.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.158785200119018,
|
|
"epoch": 3.1312779711586276,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000440242445100382,
|
|
"loss": 5.7806,
|
|
"mean_token_accuracy": 0.16242196559906005,
|
|
"num_tokens": 6161390.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.150398540496826,
|
|
"epoch": 3.136250621581303,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043997714196397143,
|
|
"loss": 5.7185,
|
|
"mean_token_accuracy": 0.17186514735221864,
|
|
"num_tokens": 6170982.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.216726207733155,
|
|
"epoch": 3.141223272003978,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00043971134181626173,
|
|
"loss": 5.7972,
|
|
"mean_token_accuracy": 0.1616930343210697,
|
|
"num_tokens": 6179356.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.163413715362549,
|
|
"epoch": 3.1461959224266534,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043944504545800396,
|
|
"loss": 5.7381,
|
|
"mean_token_accuracy": 0.16850098818540574,
|
|
"num_tokens": 6188270.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.092465925216675,
|
|
"epoch": 3.1511685728493286,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004391782536914447,
|
|
"loss": 5.6758,
|
|
"mean_token_accuracy": 0.1722916468977928,
|
|
"num_tokens": 6197624.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.181496238708496,
|
|
"epoch": 3.156141223272004,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00043891096732032257,
|
|
"loss": 5.7337,
|
|
"mean_token_accuracy": 0.1694842904806137,
|
|
"num_tokens": 6207174.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.00815978050232,
|
|
"epoch": 3.161113873694679,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043864318714986625,
|
|
"loss": 5.564,
|
|
"mean_token_accuracy": 0.18162263929843903,
|
|
"num_tokens": 6217939.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.153735589981079,
|
|
"epoch": 3.1660865241173544,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00043837491398679234,
|
|
"loss": 5.7577,
|
|
"mean_token_accuracy": 0.1657753676176071,
|
|
"num_tokens": 6227857.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.090896844863892,
|
|
"epoch": 3.17105917454003,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004381061486393025,
|
|
"loss": 5.8061,
|
|
"mean_token_accuracy": 0.15896081030368805,
|
|
"num_tokens": 6236832.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.110244846343994,
|
|
"epoch": 3.1760318249627053,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004378368919170809,
|
|
"loss": 5.7255,
|
|
"mean_token_accuracy": 0.16526849567890167,
|
|
"num_tokens": 6245879.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.082866477966308,
|
|
"epoch": 3.1810044753853806,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043756714463129244,
|
|
"loss": 5.7378,
|
|
"mean_token_accuracy": 0.16449848264455796,
|
|
"num_tokens": 6256350.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.039389896392822,
|
|
"epoch": 3.185977125808056,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004372969075945796,
|
|
"loss": 5.5248,
|
|
"mean_token_accuracy": 0.1876134157180786,
|
|
"num_tokens": 6266085.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.152119922637939,
|
|
"epoch": 3.190949776230731,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004370261816210604,
|
|
"loss": 5.7895,
|
|
"mean_token_accuracy": 0.16866564452648164,
|
|
"num_tokens": 6276499.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.1442320346832275,
|
|
"epoch": 3.1959224266534063,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004367549675263258,
|
|
"loss": 5.7917,
|
|
"mean_token_accuracy": 0.16600092500448227,
|
|
"num_tokens": 6286630.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.1032568454742435,
|
|
"epoch": 3.2008950770760816,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004364832661274373,
|
|
"loss": 5.7355,
|
|
"mean_token_accuracy": 0.16697604954242706,
|
|
"num_tokens": 6295907.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.103985786437988,
|
|
"epoch": 3.205867727498757,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043621107824292444,
|
|
"loss": 5.678,
|
|
"mean_token_accuracy": 0.16814650744199752,
|
|
"num_tokens": 6304304.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.0664064407348635,
|
|
"epoch": 3.210840377921432,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004359384046927824,
|
|
"loss": 5.7743,
|
|
"mean_token_accuracy": 0.1621297836303711,
|
|
"num_tokens": 6315041.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.094092607498169,
|
|
"epoch": 3.2158130283441073,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004356652462984693,
|
|
"loss": 5.6589,
|
|
"mean_token_accuracy": 0.17201083153486252,
|
|
"num_tokens": 6324551.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.112543964385987,
|
|
"epoch": 3.2207856787667826,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043539160388290416,
|
|
"loss": 5.716,
|
|
"mean_token_accuracy": 0.1722979485988617,
|
|
"num_tokens": 6335512.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.108734464645385,
|
|
"epoch": 3.225758329189458,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004351174782704641,
|
|
"loss": 5.784,
|
|
"mean_token_accuracy": 0.164656613022089,
|
|
"num_tokens": 6345987.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.979353904724121,
|
|
"epoch": 3.230730979612133,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043484287028698176,
|
|
"loss": 5.5847,
|
|
"mean_token_accuracy": 0.18731824606657027,
|
|
"num_tokens": 6357617.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.0212657928466795,
|
|
"epoch": 3.2357036300348083,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043456778075974306,
|
|
"loss": 5.671,
|
|
"mean_token_accuracy": 0.1743704840540886,
|
|
"num_tokens": 6366123.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.089354991912842,
|
|
"epoch": 3.240676280457484,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043429221051748487,
|
|
"loss": 5.7719,
|
|
"mean_token_accuracy": 0.16662058979272842,
|
|
"num_tokens": 6375776.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.183185720443726,
|
|
"epoch": 3.2456489308801593,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043401616039039174,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.15838052928447724,
|
|
"num_tokens": 6386235.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.1844994068145756,
|
|
"epoch": 3.2506215813028345,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004337396312100944,
|
|
"loss": 5.7773,
|
|
"mean_token_accuracy": 0.1660405546426773,
|
|
"num_tokens": 6396919.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.102126741409302,
|
|
"epoch": 3.2555942317255098,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004334626238096665,
|
|
"loss": 5.7532,
|
|
"mean_token_accuracy": 0.17060373574495316,
|
|
"num_tokens": 6406769.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.073059892654419,
|
|
"epoch": 3.260566882148185,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004331851390236227,
|
|
"loss": 5.7565,
|
|
"mean_token_accuracy": 0.1754371017217636,
|
|
"num_tokens": 6416208.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.117652130126953,
|
|
"epoch": 3.2655395325708603,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043290717768791533,
|
|
"loss": 5.7157,
|
|
"mean_token_accuracy": 0.16898192018270491,
|
|
"num_tokens": 6426125.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.063122653961182,
|
|
"epoch": 3.2705121829935355,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043262874063993285,
|
|
"loss": 5.6867,
|
|
"mean_token_accuracy": 0.16296105235815048,
|
|
"num_tokens": 6436671.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.089679527282715,
|
|
"epoch": 3.2754848334162108,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004323498287184967,
|
|
"loss": 5.803,
|
|
"mean_token_accuracy": 0.16432414650917054,
|
|
"num_tokens": 6447396.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.0989457130432125,
|
|
"epoch": 3.280457483838886,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043207044276385884,
|
|
"loss": 5.7261,
|
|
"mean_token_accuracy": 0.16696277707815171,
|
|
"num_tokens": 6457045.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.011654853820801,
|
|
"epoch": 3.2854301342615613,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043179058361769924,
|
|
"loss": 5.7724,
|
|
"mean_token_accuracy": 0.17085179090499877,
|
|
"num_tokens": 6466741.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.927375173568725,
|
|
"epoch": 3.2904027846842365,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004315102521231237,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.19177620410919188,
|
|
"num_tokens": 6477769.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.175060224533081,
|
|
"epoch": 3.295375435106912,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004312294491246609,
|
|
"loss": 5.8537,
|
|
"mean_token_accuracy": 0.16070080697536468,
|
|
"num_tokens": 6486991.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.111861801147461,
|
|
"epoch": 3.3003480855295875,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004309481754682597,
|
|
"loss": 5.7423,
|
|
"mean_token_accuracy": 0.16674381792545317,
|
|
"num_tokens": 6496772.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.116496229171753,
|
|
"epoch": 3.3053207359522627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043066643200128724,
|
|
"loss": 5.7858,
|
|
"mean_token_accuracy": 0.164015344530344,
|
|
"num_tokens": 6507577.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.110065650939942,
|
|
"epoch": 3.310293386374938,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004303842195725259,
|
|
"loss": 5.7782,
|
|
"mean_token_accuracy": 0.16365092992782593,
|
|
"num_tokens": 6518040.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.115998983383179,
|
|
"epoch": 3.3152660367976132,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043010153903217074,
|
|
"loss": 5.7313,
|
|
"mean_token_accuracy": 0.1665323406457901,
|
|
"num_tokens": 6529261.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.091701078414917,
|
|
"epoch": 3.3202386872202885,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004298183912318273,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.17246847599744797,
|
|
"num_tokens": 6537527.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.099134349822998,
|
|
"epoch": 3.3252113376429637,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00042953477702450845,
|
|
"loss": 5.7603,
|
|
"mean_token_accuracy": 0.1710022747516632,
|
|
"num_tokens": 6547205.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.08137698173523,
|
|
"epoch": 3.330183988065639,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004292506972646324,
|
|
"loss": 5.8454,
|
|
"mean_token_accuracy": 0.16369720846414565,
|
|
"num_tokens": 6556328.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.145713853836059,
|
|
"epoch": 3.3351566384883142,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042896615280802006,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.16524243652820586,
|
|
"num_tokens": 6566414.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.038229942321777,
|
|
"epoch": 3.3401292889109895,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042868114451189164,
|
|
"loss": 5.6283,
|
|
"mean_token_accuracy": 0.1868848443031311,
|
|
"num_tokens": 6577365.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.068049192428589,
|
|
"epoch": 3.3451019393336647,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004283956732348655,
|
|
"loss": 5.7172,
|
|
"mean_token_accuracy": 0.16948506683111192,
|
|
"num_tokens": 6586647.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.116658496856689,
|
|
"epoch": 3.35007458975634,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042810973983695426,
|
|
"loss": 5.7945,
|
|
"mean_token_accuracy": 0.16618717461824417,
|
|
"num_tokens": 6596955.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.156133222579956,
|
|
"epoch": 3.3550472401790152,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004278233451795629,
|
|
"loss": 5.7026,
|
|
"mean_token_accuracy": 0.17426156252622604,
|
|
"num_tokens": 6606161.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.948416805267334,
|
|
"epoch": 3.3600198906016905,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000427536490125486,
|
|
"loss": 5.5826,
|
|
"mean_token_accuracy": 0.18411061763763428,
|
|
"num_tokens": 6616540.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.042610549926758,
|
|
"epoch": 3.364992541024366,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042724917553890507,
|
|
"loss": 5.7216,
|
|
"mean_token_accuracy": 0.1715673863887787,
|
|
"num_tokens": 6626109.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.101540899276733,
|
|
"epoch": 3.3699651914470414,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042696140228538616,
|
|
"loss": 5.7235,
|
|
"mean_token_accuracy": 0.17070289552211762,
|
|
"num_tokens": 6635576.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.0797447681427,
|
|
"epoch": 3.3749378418697167,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042667317123187696,
|
|
"loss": 5.8241,
|
|
"mean_token_accuracy": 0.1661553755402565,
|
|
"num_tokens": 6645728.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.111036491394043,
|
|
"epoch": 3.379910492292392,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00042638448324670424,
|
|
"loss": 5.6965,
|
|
"mean_token_accuracy": 0.17327468246221542,
|
|
"num_tokens": 6654538.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.05802059173584,
|
|
"epoch": 3.384883142715067,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004260953391995716,
|
|
"loss": 5.6568,
|
|
"mean_token_accuracy": 0.1744958832859993,
|
|
"num_tokens": 6664811.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.063599061965943,
|
|
"epoch": 3.3898557931377424,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042580573996155646,
|
|
"loss": 5.765,
|
|
"mean_token_accuracy": 0.17499004453420638,
|
|
"num_tokens": 6674185.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.102707195281982,
|
|
"epoch": 3.3948284435604177,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004255156864051074,
|
|
"loss": 5.7595,
|
|
"mean_token_accuracy": 0.15964170247316362,
|
|
"num_tokens": 6684103.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.923308324813843,
|
|
"epoch": 3.399801093983093,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00042522517940404186,
|
|
"loss": 5.437,
|
|
"mean_token_accuracy": 0.19307186007499694,
|
|
"num_tokens": 6693631.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.087829494476319,
|
|
"epoch": 3.404773744405768,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004249342198335433,
|
|
"loss": 5.7911,
|
|
"mean_token_accuracy": 0.1642305538058281,
|
|
"num_tokens": 6704465.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.036604547500611,
|
|
"epoch": 3.4097463948284434,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042464280857015856,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.17335094511508942,
|
|
"num_tokens": 6713592.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.102187871932983,
|
|
"epoch": 3.4147190452511187,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00042435094649179525,
|
|
"loss": 5.7835,
|
|
"mean_token_accuracy": 0.16397027745842935,
|
|
"num_tokens": 6726252.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.121373844146729,
|
|
"epoch": 3.419691695673794,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004240586344777191,
|
|
"loss": 5.8536,
|
|
"mean_token_accuracy": 0.1597203016281128,
|
|
"num_tokens": 6736301.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.114486789703369,
|
|
"epoch": 3.4246643460964696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004237658734085514,
|
|
"loss": 5.8014,
|
|
"mean_token_accuracy": 0.16278771460056304,
|
|
"num_tokens": 6745706.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.070153045654297,
|
|
"epoch": 3.429636996519145,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042347266416626624,
|
|
"loss": 5.7956,
|
|
"mean_token_accuracy": 0.16099387854337693,
|
|
"num_tokens": 6755811.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.153490018844605,
|
|
"epoch": 3.43460964694182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004231790076341877,
|
|
"loss": 5.7343,
|
|
"mean_token_accuracy": 0.16927113607525826,
|
|
"num_tokens": 6766365.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.052618312835693,
|
|
"epoch": 3.4395822973644954,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004228849046969876,
|
|
"loss": 5.7081,
|
|
"mean_token_accuracy": 0.17573282122612,
|
|
"num_tokens": 6775806.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.065675735473633,
|
|
"epoch": 3.4445549477871706,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004225903562406826,
|
|
"loss": 5.7429,
|
|
"mean_token_accuracy": 0.17060302197933197,
|
|
"num_tokens": 6785996.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.07124400138855,
|
|
"epoch": 3.449527598209846,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004222953631526313,
|
|
"loss": 5.6984,
|
|
"mean_token_accuracy": 0.18066974878311157,
|
|
"num_tokens": 6795528.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.046488332748413,
|
|
"epoch": 3.454500248632521,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000421999926321532,
|
|
"loss": 5.7041,
|
|
"mean_token_accuracy": 0.17475352585315704,
|
|
"num_tokens": 6804592.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.988970375061035,
|
|
"epoch": 3.4594728990551964,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004217040466374199,
|
|
"loss": 5.7424,
|
|
"mean_token_accuracy": 0.16841422021389008,
|
|
"num_tokens": 6813242.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.118766021728516,
|
|
"epoch": 3.4644455494778716,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042140772499166403,
|
|
"loss": 5.7697,
|
|
"mean_token_accuracy": 0.17049301862716676,
|
|
"num_tokens": 6823223.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.945927953720092,
|
|
"epoch": 3.469418199900547,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004211109622769652,
|
|
"loss": 5.613,
|
|
"mean_token_accuracy": 0.18204609602689742,
|
|
"num_tokens": 6832425.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.005742502212525,
|
|
"epoch": 3.474390850323222,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00042081375938735283,
|
|
"loss": 5.7992,
|
|
"mean_token_accuracy": 0.1653480514883995,
|
|
"num_tokens": 6842549.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.080510568618775,
|
|
"epoch": 3.4793635007458974,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00042051611721818233,
|
|
"loss": 5.8713,
|
|
"mean_token_accuracy": 0.16462980061769486,
|
|
"num_tokens": 6851652.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 3.4793635007458974,
|
|
"eval_entropy": 6.044144784250567,
|
|
"eval_loss": 6.597715377807617,
|
|
"eval_mean_token_accuracy": 0.1396133896941963,
|
|
"eval_num_tokens": 6851652.0,
|
|
"eval_runtime": 9.7155,
|
|
"eval_samples_per_second": 356.646,
|
|
"eval_steps_per_second": 44.671,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.102687931060791,
|
|
"epoch": 3.4843361511685726,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042021803666613273,
|
|
"loss": 5.7222,
|
|
"mean_token_accuracy": 0.17163947224617004,
|
|
"num_tokens": 6860716.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.034015703201294,
|
|
"epoch": 3.489308801591248,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00041991951862920365,
|
|
"loss": 5.7076,
|
|
"mean_token_accuracy": 0.1728237673640251,
|
|
"num_tokens": 6870074.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.16734938621521,
|
|
"epoch": 3.4942814520139236,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004196205640067124,
|
|
"loss": 5.9609,
|
|
"mean_token_accuracy": 0.14933402240276336,
|
|
"num_tokens": 6880569.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.164630603790283,
|
|
"epoch": 3.499254102436599,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000419321173699292,
|
|
"loss": 5.8307,
|
|
"mean_token_accuracy": 0.15948736071586608,
|
|
"num_tokens": 6890522.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.1042969703674315,
|
|
"epoch": 3.504226752859274,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004190213486088877,
|
|
"loss": 5.7673,
|
|
"mean_token_accuracy": 0.16679638475179673,
|
|
"num_tokens": 6900388.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.047546291351319,
|
|
"epoch": 3.5091994032819493,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004187210896387546,
|
|
"loss": 5.7317,
|
|
"mean_token_accuracy": 0.16994239538908004,
|
|
"num_tokens": 6910467.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.016292953491211,
|
|
"epoch": 3.5141720537046246,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004184203976934552,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.1713675558567047,
|
|
"num_tokens": 6919370.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.977844285964966,
|
|
"epoch": 3.5191447041273,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041811927367885605,
|
|
"loss": 5.5475,
|
|
"mean_token_accuracy": 0.1895998015999794,
|
|
"num_tokens": 6929688.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.057902717590332,
|
|
"epoch": 3.524117354549975,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004178177185021254,
|
|
"loss": 5.7964,
|
|
"mean_token_accuracy": 0.17015004605054856,
|
|
"num_tokens": 6939861.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.100983381271362,
|
|
"epoch": 3.5290900049726504,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004175157330717305,
|
|
"loss": 5.8026,
|
|
"mean_token_accuracy": 0.15981769561767578,
|
|
"num_tokens": 6949746.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.151196479797363,
|
|
"epoch": 3.5340626553953256,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000417213318297435,
|
|
"loss": 5.7646,
|
|
"mean_token_accuracy": 0.1679669663310051,
|
|
"num_tokens": 6961408.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.079554224014283,
|
|
"epoch": 3.539035305818001,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041691047509029555,
|
|
"loss": 5.8155,
|
|
"mean_token_accuracy": 0.17046236693859101,
|
|
"num_tokens": 6972003.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.069966745376587,
|
|
"epoch": 3.5440079562406765,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00041660720436265987,
|
|
"loss": 5.7184,
|
|
"mean_token_accuracy": 0.17085322737693787,
|
|
"num_tokens": 6983528.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.0844159603118895,
|
|
"epoch": 3.548980606663352,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004163035070281634,
|
|
"loss": 5.7877,
|
|
"mean_token_accuracy": 0.17089734226465225,
|
|
"num_tokens": 6992766.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.002828788757324,
|
|
"epoch": 3.553953257086027,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00041599938400172686,
|
|
"loss": 5.6811,
|
|
"mean_token_accuracy": 0.1688712403178215,
|
|
"num_tokens": 7001797.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.075935888290405,
|
|
"epoch": 3.5589259075087023,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041569483619955354,
|
|
"loss": 5.728,
|
|
"mean_token_accuracy": 0.17270313054323197,
|
|
"num_tokens": 7011215.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.083470058441162,
|
|
"epoch": 3.5638985579313776,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041538986453912616,
|
|
"loss": 5.7596,
|
|
"mean_token_accuracy": 0.16251621693372725,
|
|
"num_tokens": 7021513.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.052660083770752,
|
|
"epoch": 3.568871208354053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004150844699392045,
|
|
"loss": 5.7848,
|
|
"mean_token_accuracy": 0.16535720378160476,
|
|
"num_tokens": 7031654.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.004251194000244,
|
|
"epoch": 3.573843858776728,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00041477865331982264,
|
|
"loss": 5.7499,
|
|
"mean_token_accuracy": 0.16939473748207093,
|
|
"num_tokens": 7040614.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.051688289642334,
|
|
"epoch": 3.5788165091994033,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004144724156022857,
|
|
"loss": 5.7432,
|
|
"mean_token_accuracy": 0.1719522327184677,
|
|
"num_tokens": 7050877.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.177400016784668,
|
|
"epoch": 3.5837891596220786,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00041416575770916763,
|
|
"loss": 5.9265,
|
|
"mean_token_accuracy": 0.1563877694308758,
|
|
"num_tokens": 7061330.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.123777055740357,
|
|
"epoch": 3.588761810044754,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041385868056430827,
|
|
"loss": 5.843,
|
|
"mean_token_accuracy": 0.16367755085229874,
|
|
"num_tokens": 7071290.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.082329893112183,
|
|
"epoch": 3.593734460467429,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004135511850928102,
|
|
"loss": 5.7905,
|
|
"mean_token_accuracy": 0.16273097246885299,
|
|
"num_tokens": 7080112.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.103447723388672,
|
|
"epoch": 3.5987071108901043,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00041324327222103655,
|
|
"loss": 5.952,
|
|
"mean_token_accuracy": 0.15705312192440032,
|
|
"num_tokens": 7091467.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.067325639724731,
|
|
"epoch": 3.6036797613127796,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00041293494287660785,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.18322497457265854,
|
|
"num_tokens": 7101744.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.078366708755493,
|
|
"epoch": 3.608652411735455,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041262619798839914,
|
|
"loss": 5.7757,
|
|
"mean_token_accuracy": 0.16733835637569427,
|
|
"num_tokens": 7112854.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.963122844696045,
|
|
"epoch": 3.61362506215813,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00041231703848653746,
|
|
"loss": 5.7601,
|
|
"mean_token_accuracy": 0.17440671622753143,
|
|
"num_tokens": 7121613.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.1137251377105715,
|
|
"epoch": 3.6185977125808053,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041200746530239903,
|
|
"loss": 5.8238,
|
|
"mean_token_accuracy": 0.16737723648548125,
|
|
"num_tokens": 7131357.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.038400888442993,
|
|
"epoch": 3.623570363003481,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004116974793686061,
|
|
"loss": 5.6868,
|
|
"mean_token_accuracy": 0.1773845911026001,
|
|
"num_tokens": 7140200.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.044875907897949,
|
|
"epoch": 3.6285430134261563,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004113870816190244,
|
|
"loss": 5.8051,
|
|
"mean_token_accuracy": 0.1665480002760887,
|
|
"num_tokens": 7149367.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.036097240447998,
|
|
"epoch": 3.6335156638488315,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004110762729887605,
|
|
"loss": 5.7584,
|
|
"mean_token_accuracy": 0.17131147533655167,
|
|
"num_tokens": 7158088.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.020812129974365,
|
|
"epoch": 3.6384883142715068,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00041076505441415873,
|
|
"loss": 5.76,
|
|
"mean_token_accuracy": 0.17685214430093765,
|
|
"num_tokens": 7168141.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.094634532928467,
|
|
"epoch": 3.643460964694182,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004104534268327982,
|
|
"loss": 5.6806,
|
|
"mean_token_accuracy": 0.17692216634750366,
|
|
"num_tokens": 7177646.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.127710580825806,
|
|
"epoch": 3.6484336151168573,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004101413911834904,
|
|
"loss": 5.8311,
|
|
"mean_token_accuracy": 0.16202280521392823,
|
|
"num_tokens": 7186445.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.074866104125976,
|
|
"epoch": 3.6534062655395325,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00040982894840627614,
|
|
"loss": 5.7458,
|
|
"mean_token_accuracy": 0.171629698574543,
|
|
"num_tokens": 7195812.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.987684345245361,
|
|
"epoch": 3.6583789159622078,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095160994424228,
|
|
"loss": 5.6373,
|
|
"mean_token_accuracy": 0.17803133130073548,
|
|
"num_tokens": 7204756.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.054418134689331,
|
|
"epoch": 3.663351566384883,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040920284523442125,
|
|
"loss": 5.8079,
|
|
"mean_token_accuracy": 0.17144977897405625,
|
|
"num_tokens": 7214708.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.069856023788452,
|
|
"epoch": 3.6683242168075583,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040888918672598337,
|
|
"loss": 5.8022,
|
|
"mean_token_accuracy": 0.16736897379159926,
|
|
"num_tokens": 7224623.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.083552503585816,
|
|
"epoch": 3.673296867230234,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000408575124862039,
|
|
"loss": 5.7558,
|
|
"mean_token_accuracy": 0.17243701219558716,
|
|
"num_tokens": 7235228.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.059838438034058,
|
|
"epoch": 3.678269517652909,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004082606605887331,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.17212211042642594,
|
|
"num_tokens": 7245318.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.78553352355957,
|
|
"epoch": 3.6832421680755845,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040794579485342304,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.20502729713916779,
|
|
"num_tokens": 7255468.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.086856269836426,
|
|
"epoch": 3.6882148184982597,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040763052860467547,
|
|
"loss": 5.7869,
|
|
"mean_token_accuracy": 0.16513276547193528,
|
|
"num_tokens": 7265006.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.055691146850586,
|
|
"epoch": 3.693187468920935,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004073148627922637,
|
|
"loss": 5.7666,
|
|
"mean_token_accuracy": 0.16470177322626114,
|
|
"num_tokens": 7273835.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.093978023529052,
|
|
"epoch": 3.69816011934361,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004069987983671648,
|
|
"loss": 5.8637,
|
|
"mean_token_accuracy": 0.1615389958024025,
|
|
"num_tokens": 7284617.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.003893423080444,
|
|
"epoch": 3.7031327697662855,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040668233628155664,
|
|
"loss": 5.5943,
|
|
"mean_token_accuracy": 0.18453547209501267,
|
|
"num_tokens": 7293707.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.126529598236084,
|
|
"epoch": 3.7081054201889607,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004063654774888153,
|
|
"loss": 5.8179,
|
|
"mean_token_accuracy": 0.1559830993413925,
|
|
"num_tokens": 7303227.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 5.988818979263305,
|
|
"epoch": 3.713078070611636,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004060482229435116,
|
|
"loss": 5.6855,
|
|
"mean_token_accuracy": 0.17775852233171463,
|
|
"num_tokens": 7312782.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.035627794265747,
|
|
"epoch": 3.7180507210343112,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004057305736014089,
|
|
"loss": 5.8255,
|
|
"mean_token_accuracy": 0.16572927832603454,
|
|
"num_tokens": 7322249.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.124061632156372,
|
|
"epoch": 3.7230233714569865,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004054125304194598,
|
|
"loss": 5.7706,
|
|
"mean_token_accuracy": 0.1727963089942932,
|
|
"num_tokens": 7331076.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.032123136520386,
|
|
"epoch": 3.7279960218796617,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00040509409435580355,
|
|
"loss": 5.7406,
|
|
"mean_token_accuracy": 0.16983142942190171,
|
|
"num_tokens": 7340039.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.021432495117187,
|
|
"epoch": 3.732968672302337,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004047752663697627,
|
|
"loss": 5.729,
|
|
"mean_token_accuracy": 0.17167635411024093,
|
|
"num_tokens": 7349300.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.189910221099853,
|
|
"epoch": 3.7379413227250122,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004044560474218409,
|
|
"loss": 5.8261,
|
|
"mean_token_accuracy": 0.1654362589120865,
|
|
"num_tokens": 7358363.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.022884130477905,
|
|
"epoch": 3.7429139731476875,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040413643847371923,
|
|
"loss": 5.746,
|
|
"mean_token_accuracy": 0.17480883300304412,
|
|
"num_tokens": 7369391.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.054754209518433,
|
|
"epoch": 3.7478866235703627,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004038164404882539,
|
|
"loss": 5.8252,
|
|
"mean_token_accuracy": 0.1592554122209549,
|
|
"num_tokens": 7378272.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.089888095855713,
|
|
"epoch": 3.7528592739930384,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004034960544294731,
|
|
"loss": 5.6801,
|
|
"mean_token_accuracy": 0.1753301739692688,
|
|
"num_tokens": 7387590.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.040151786804199,
|
|
"epoch": 3.7578319244157137,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040317528126257413,
|
|
"loss": 5.789,
|
|
"mean_token_accuracy": 0.16391418725252152,
|
|
"num_tokens": 7397733.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.023583459854126,
|
|
"epoch": 3.762804574838389,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004028541219539205,
|
|
"loss": 5.6459,
|
|
"mean_token_accuracy": 0.17921225726604462,
|
|
"num_tokens": 7406748.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.072662734985352,
|
|
"epoch": 3.767777225261064,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004025325774710391,
|
|
"loss": 5.8655,
|
|
"mean_token_accuracy": 0.16039032340049744,
|
|
"num_tokens": 7416419.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.022720575332642,
|
|
"epoch": 3.7727498756837394,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040221064878261684,
|
|
"loss": 5.7283,
|
|
"mean_token_accuracy": 0.17025696188211442,
|
|
"num_tokens": 7425257.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.114238214492798,
|
|
"epoch": 3.7777225261064147,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004018883368584985,
|
|
"loss": 5.8412,
|
|
"mean_token_accuracy": 0.16228442937135695,
|
|
"num_tokens": 7436914.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.067906379699707,
|
|
"epoch": 3.78269517652909,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00040156564266968336,
|
|
"loss": 5.8181,
|
|
"mean_token_accuracy": 0.1671231895685196,
|
|
"num_tokens": 7447568.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.012821531295776,
|
|
"epoch": 3.787667826951765,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00040124256718832214,
|
|
"loss": 5.7758,
|
|
"mean_token_accuracy": 0.1733666479587555,
|
|
"num_tokens": 7457717.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.0828978538513185,
|
|
"epoch": 3.7926404773744404,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004009191113877141,
|
|
"loss": 5.7503,
|
|
"mean_token_accuracy": 0.1689877226948738,
|
|
"num_tokens": 7467123.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.023863887786865,
|
|
"epoch": 3.797613127797116,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004005952762423048,
|
|
"loss": 5.7433,
|
|
"mean_token_accuracy": 0.17492797151207923,
|
|
"num_tokens": 7476987.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.060265111923218,
|
|
"epoch": 3.8025857782197914,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004002710627276821,
|
|
"loss": 5.8381,
|
|
"mean_token_accuracy": 0.17107998132705687,
|
|
"num_tokens": 7487024.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.082758569717408,
|
|
"epoch": 3.8075584286424666,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003999464718205739,
|
|
"loss": 5.8367,
|
|
"mean_token_accuracy": 0.16380092948675157,
|
|
"num_tokens": 7496826.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.077786064147949,
|
|
"epoch": 3.812531079065142,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003996215044988451,
|
|
"loss": 5.789,
|
|
"mean_token_accuracy": 0.16794931143522263,
|
|
"num_tokens": 7506981.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.003514766693115,
|
|
"epoch": 3.817503729487817,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.00039929616174149454,
|
|
"loss": 5.6004,
|
|
"mean_token_accuracy": 0.18044859319925308,
|
|
"num_tokens": 7516925.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.037631607055664,
|
|
"epoch": 3.8224763799104924,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039897044452865207,
|
|
"loss": 5.7413,
|
|
"mean_token_accuracy": 0.1756616488099098,
|
|
"num_tokens": 7525789.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.942282676696777,
|
|
"epoch": 3.8274490303331676,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003986443538415756,
|
|
"loss": 5.6884,
|
|
"mean_token_accuracy": 0.17516703009605408,
|
|
"num_tokens": 7535907.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.038372230529785,
|
|
"epoch": 3.832421680755843,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003983178906626484,
|
|
"loss": 5.7267,
|
|
"mean_token_accuracy": 0.1768231898546219,
|
|
"num_tokens": 7546985.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.937757301330566,
|
|
"epoch": 3.837394331178518,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039799105597537564,
|
|
"loss": 5.5636,
|
|
"mean_token_accuracy": 0.190547975897789,
|
|
"num_tokens": 7556912.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.0606911182403564,
|
|
"epoch": 3.8423669816011934,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003976638507643818,
|
|
"loss": 5.7837,
|
|
"mean_token_accuracy": 0.1704917550086975,
|
|
"num_tokens": 7567156.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.038426876068115,
|
|
"epoch": 3.8473396320238686,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003973362760154076,
|
|
"loss": 5.7305,
|
|
"mean_token_accuracy": 0.17417844831943513,
|
|
"num_tokens": 7576423.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.061949682235718,
|
|
"epoch": 3.852312282446544,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003970083327153071,
|
|
"loss": 5.7446,
|
|
"mean_token_accuracy": 0.17025127708911897,
|
|
"num_tokens": 7585612.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.958314657211304,
|
|
"epoch": 3.857284932869219,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003966800218520445,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.17344398945569992,
|
|
"num_tokens": 7595581.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.06782488822937,
|
|
"epoch": 3.8622575832918944,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003963513444146914,
|
|
"loss": 5.7427,
|
|
"mean_token_accuracy": 0.17112387865781784,
|
|
"num_tokens": 7605202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.086015844345093,
|
|
"epoch": 3.8672302337145696,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003960223013934239,
|
|
"loss": 5.6482,
|
|
"mean_token_accuracy": 0.18539201319217682,
|
|
"num_tokens": 7615783.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.093843460083008,
|
|
"epoch": 3.872202884137245,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003956928937795192,
|
|
"loss": 5.8533,
|
|
"mean_token_accuracy": 0.15679872334003447,
|
|
"num_tokens": 7626609.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.040159130096436,
|
|
"epoch": 3.8771755345599206,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000395363122565353,
|
|
"loss": 5.7344,
|
|
"mean_token_accuracy": 0.1774304687976837,
|
|
"num_tokens": 7636676.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.0875279903411865,
|
|
"epoch": 3.882148184982596,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039503298874439655,
|
|
"loss": 5.7626,
|
|
"mean_token_accuracy": 0.17330909818410872,
|
|
"num_tokens": 7646523.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.966638898849487,
|
|
"epoch": 3.887120835405271,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003947024933112131,
|
|
"loss": 5.6422,
|
|
"mean_token_accuracy": 0.1802971065044403,
|
|
"num_tokens": 7656267.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.018937349319458,
|
|
"epoch": 3.8920934858279463,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003943716372614556,
|
|
"loss": 5.7126,
|
|
"mean_token_accuracy": 0.175618451833725,
|
|
"num_tokens": 7665529.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.067586278915405,
|
|
"epoch": 3.8970661362506216,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003940404215918634,
|
|
"loss": 5.795,
|
|
"mean_token_accuracy": 0.16990861892700196,
|
|
"num_tokens": 7675477.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.860292720794678,
|
|
"epoch": 3.902038786673297,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00039370884730025897,
|
|
"loss": 5.4091,
|
|
"mean_token_accuracy": 0.20204151719808577,
|
|
"num_tokens": 7685675.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.95585298538208,
|
|
"epoch": 3.907011437095972,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003933769153855455,
|
|
"loss": 5.7813,
|
|
"mean_token_accuracy": 0.17225549817085267,
|
|
"num_tokens": 7695821.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.99542465209961,
|
|
"epoch": 3.9119840875186473,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003930446268477034,
|
|
"loss": 5.7817,
|
|
"mean_token_accuracy": 0.16887823939323426,
|
|
"num_tokens": 7706163.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.0233885765075685,
|
|
"epoch": 3.9169567379413226,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003927119826877875,
|
|
"loss": 5.6957,
|
|
"mean_token_accuracy": 0.17559928596019744,
|
|
"num_tokens": 7715125.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.015093278884888,
|
|
"epoch": 3.921929388363998,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003923789839079238,
|
|
"loss": 5.7142,
|
|
"mean_token_accuracy": 0.17194195538759233,
|
|
"num_tokens": 7724546.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.853858757019043,
|
|
"epoch": 3.9269020387866735,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00039204563151130683,
|
|
"loss": 5.544,
|
|
"mean_token_accuracy": 0.19260187298059464,
|
|
"num_tokens": 7734829.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.993786859512329,
|
|
"epoch": 3.931874689209349,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003917119265021965,
|
|
"loss": 5.7247,
|
|
"mean_token_accuracy": 0.16931314915418624,
|
|
"num_tokens": 7744612.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.06753191947937,
|
|
"epoch": 3.936847339632024,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003913778698859146,
|
|
"loss": 5.8116,
|
|
"mean_token_accuracy": 0.16965224146842955,
|
|
"num_tokens": 7754042.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.087614107131958,
|
|
"epoch": 3.9418199900546993,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003910434626688427,
|
|
"loss": 5.6937,
|
|
"mean_token_accuracy": 0.17561808079481125,
|
|
"num_tokens": 7762608.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.9966308116912845,
|
|
"epoch": 3.9467926404773745,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003907087058584183,
|
|
"loss": 5.6317,
|
|
"mean_token_accuracy": 0.18102066665887834,
|
|
"num_tokens": 7771477.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.9490820407867435,
|
|
"epoch": 3.95176529090005,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00039037360046313186,
|
|
"loss": 5.7338,
|
|
"mean_token_accuracy": 0.17638552114367484,
|
|
"num_tokens": 7782785.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.144771528244019,
|
|
"epoch": 3.956737941322725,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039003814749252454,
|
|
"loss": 5.9076,
|
|
"mean_token_accuracy": 0.15977851673960686,
|
|
"num_tokens": 7793572.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.069395065307617,
|
|
"epoch": 3.9617105917454003,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00038970234795718424,
|
|
"loss": 5.6742,
|
|
"mean_token_accuracy": 0.17439233213663102,
|
|
"num_tokens": 7803170.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.048822498321533,
|
|
"epoch": 3.9666832421680756,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000389366202868743,
|
|
"loss": 5.8353,
|
|
"mean_token_accuracy": 0.16539930254220964,
|
|
"num_tokens": 7813445.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.0415510654449465,
|
|
"epoch": 3.971655892590751,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003890297132398739,
|
|
"loss": 5.8421,
|
|
"mean_token_accuracy": 0.16893949508666992,
|
|
"num_tokens": 7823089.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.0344892024993895,
|
|
"epoch": 3.976628543013426,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003886928800842879,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.17397593557834626,
|
|
"num_tokens": 7832487.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.976628543013426,
|
|
"eval_entropy": 5.932080362249629,
|
|
"eval_loss": 6.540711402893066,
|
|
"eval_mean_token_accuracy": 0.14234826518093935,
|
|
"eval_num_tokens": 7832487.0,
|
|
"eval_runtime": 12.6455,
|
|
"eval_samples_per_second": 274.01,
|
|
"eval_steps_per_second": 34.32,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.9831901550292965,
|
|
"epoch": 3.9816011934361013,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038835570441673095,
|
|
"loss": 5.7308,
|
|
"mean_token_accuracy": 0.17587945461273194,
|
|
"num_tokens": 7842540.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.0749249935150145,
|
|
"epoch": 3.9865738438587766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003880181872529809,
|
|
"loss": 5.8054,
|
|
"mean_token_accuracy": 0.16680429726839066,
|
|
"num_tokens": 7852384.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.0260110855102536,
|
|
"epoch": 3.991546494281452,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003876803296098443,
|
|
"loss": 5.7897,
|
|
"mean_token_accuracy": 0.1721744254231453,
|
|
"num_tokens": 7864082.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.028295230865479,
|
|
"epoch": 3.996519144704127,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003873421325051535,
|
|
"loss": 5.7143,
|
|
"mean_token_accuracy": 0.16935357376933097,
|
|
"num_tokens": 7873677.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.080985334184435,
|
|
"epoch": 4.000994530084535,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038700359695776346,
|
|
"loss": 5.6925,
|
|
"mean_token_accuracy": 0.17726261251502567,
|
|
"num_tokens": 7882368.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.986006736755371,
|
|
"epoch": 4.00596718050721,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003866647239875487,
|
|
"loss": 5.3572,
|
|
"mean_token_accuracy": 0.18282776772975923,
|
|
"num_tokens": 7892327.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.950453233718872,
|
|
"epoch": 4.010939830929885,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038632551461540023,
|
|
"loss": 5.3429,
|
|
"mean_token_accuracy": 0.19138669222593307,
|
|
"num_tokens": 7902048.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.027964925765991,
|
|
"epoch": 4.015912481352561,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003859859698632228,
|
|
"loss": 5.3845,
|
|
"mean_token_accuracy": 0.19397752434015275,
|
|
"num_tokens": 7911512.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.018005228042602,
|
|
"epoch": 4.020885131775236,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003856460907539312,
|
|
"loss": 5.4799,
|
|
"mean_token_accuracy": 0.17938076853752136,
|
|
"num_tokens": 7921832.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.0080382347106935,
|
|
"epoch": 4.025857782197911,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003853058783114478,
|
|
"loss": 5.3953,
|
|
"mean_token_accuracy": 0.19186470955610274,
|
|
"num_tokens": 7931143.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 5.925951862335205,
|
|
"epoch": 4.030830432620586,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003849653335606987,
|
|
"loss": 5.2969,
|
|
"mean_token_accuracy": 0.19981253296136856,
|
|
"num_tokens": 7941108.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.928447484970093,
|
|
"epoch": 4.035803083043262,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003846244575276117,
|
|
"loss": 5.3743,
|
|
"mean_token_accuracy": 0.18678346425294876,
|
|
"num_tokens": 7951464.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.055474519729614,
|
|
"epoch": 4.040775733465938,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038428325123911205,
|
|
"loss": 5.4677,
|
|
"mean_token_accuracy": 0.17983518093824385,
|
|
"num_tokens": 7961626.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.02820053100586,
|
|
"epoch": 4.045748383888613,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003839417157231205,
|
|
"loss": 5.4726,
|
|
"mean_token_accuracy": 0.17989495992660523,
|
|
"num_tokens": 7970135.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.868865299224853,
|
|
"epoch": 4.050721034311288,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003835998520085492,
|
|
"loss": 5.3814,
|
|
"mean_token_accuracy": 0.18720358312129975,
|
|
"num_tokens": 7982177.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.974019861221313,
|
|
"epoch": 4.0556936847339635,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003832576611252993,
|
|
"loss": 5.3448,
|
|
"mean_token_accuracy": 0.19214671701192856,
|
|
"num_tokens": 7991845.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.947139692306519,
|
|
"epoch": 4.060666335156639,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003829151441042572,
|
|
"loss": 5.3217,
|
|
"mean_token_accuracy": 0.19296037703752517,
|
|
"num_tokens": 8000731.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.936928367614746,
|
|
"epoch": 4.065638985579314,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038257230197729237,
|
|
"loss": 5.3557,
|
|
"mean_token_accuracy": 0.18900153785943985,
|
|
"num_tokens": 8011137.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.958340787887574,
|
|
"epoch": 4.070611636001989,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038222913577725325,
|
|
"loss": 5.4103,
|
|
"mean_token_accuracy": 0.17967583388090133,
|
|
"num_tokens": 8020413.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.950023937225342,
|
|
"epoch": 4.0755842864246645,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038188564653796487,
|
|
"loss": 5.3399,
|
|
"mean_token_accuracy": 0.1946147635579109,
|
|
"num_tokens": 8029727.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.996586275100708,
|
|
"epoch": 4.08055693684734,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003815418352942252,
|
|
"loss": 5.4106,
|
|
"mean_token_accuracy": 0.19237400442361832,
|
|
"num_tokens": 8038707.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.842990159988403,
|
|
"epoch": 4.085529587270015,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003811977030818026,
|
|
"loss": 5.1438,
|
|
"mean_token_accuracy": 0.2157265216112137,
|
|
"num_tokens": 8048604.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.9085572242736815,
|
|
"epoch": 4.09050223769269,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000380853250937432,
|
|
"loss": 5.3868,
|
|
"mean_token_accuracy": 0.19575043320655822,
|
|
"num_tokens": 8058089.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 6.037813425064087,
|
|
"epoch": 4.0954748881153655,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00038050847989881236,
|
|
"loss": 5.3898,
|
|
"mean_token_accuracy": 0.18490490764379502,
|
|
"num_tokens": 8068078.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.89682674407959,
|
|
"epoch": 4.100447538538041,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003801633910046035,
|
|
"loss": 5.2827,
|
|
"mean_token_accuracy": 0.19786714464426042,
|
|
"num_tokens": 8076918.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.793061685562134,
|
|
"epoch": 4.105420188960716,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003798179852944225,
|
|
"loss": 5.1907,
|
|
"mean_token_accuracy": 0.21232437938451768,
|
|
"num_tokens": 8087104.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.936882352828979,
|
|
"epoch": 4.110392839383391,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037947226380884115,
|
|
"loss": 5.474,
|
|
"mean_token_accuracy": 0.17738474756479264,
|
|
"num_tokens": 8096976.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.9654487609863285,
|
|
"epoch": 4.1153654898060665,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037912622758938243,
|
|
"loss": 5.3635,
|
|
"mean_token_accuracy": 0.1959421753883362,
|
|
"num_tokens": 8106707.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.990051937103272,
|
|
"epoch": 4.120338140228742,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003787798776785174,
|
|
"loss": 5.4202,
|
|
"mean_token_accuracy": 0.18028839230537413,
|
|
"num_tokens": 8116822.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.844709873199463,
|
|
"epoch": 4.125310790651417,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003784332151196622,
|
|
"loss": 5.3191,
|
|
"mean_token_accuracy": 0.19868418276309968,
|
|
"num_tokens": 8125677.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.769356679916382,
|
|
"epoch": 4.130283441074092,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037808624095717503,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.2077832415699959,
|
|
"num_tokens": 8134878.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.943852806091309,
|
|
"epoch": 4.1352560914967675,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003777389562363526,
|
|
"loss": 5.3732,
|
|
"mean_token_accuracy": 0.20043659806251526,
|
|
"num_tokens": 8144491.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.003221559524536,
|
|
"epoch": 4.140228741919443,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037739136200342743,
|
|
"loss": 5.4257,
|
|
"mean_token_accuracy": 0.1835670828819275,
|
|
"num_tokens": 8155163.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.914414405822754,
|
|
"epoch": 4.145201392342118,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037704345930556426,
|
|
"loss": 5.3749,
|
|
"mean_token_accuracy": 0.18914274275302886,
|
|
"num_tokens": 8164626.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.006533193588257,
|
|
"epoch": 4.150174042764793,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037669524919085716,
|
|
"loss": 5.5437,
|
|
"mean_token_accuracy": 0.18296259492635727,
|
|
"num_tokens": 8174681.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.03918776512146,
|
|
"epoch": 4.1551466931874685,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003763467327083264,
|
|
"loss": 5.5833,
|
|
"mean_token_accuracy": 0.17702725976705552,
|
|
"num_tokens": 8184627.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.875203466415405,
|
|
"epoch": 4.160119343610144,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037599791090791527,
|
|
"loss": 5.2092,
|
|
"mean_token_accuracy": 0.20463866144418716,
|
|
"num_tokens": 8195029.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.904081678390503,
|
|
"epoch": 4.165091994032819,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 0.0003756487848404868,
|
|
"loss": 5.2877,
|
|
"mean_token_accuracy": 0.20360684245824814,
|
|
"num_tokens": 8204897.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 5.947072267532349,
|
|
"epoch": 4.170064644455495,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003752993555578205,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.17716439217329025,
|
|
"num_tokens": 8215859.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.971736860275269,
|
|
"epoch": 4.17503729487817,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003749496241126095,
|
|
"loss": 5.314,
|
|
"mean_token_accuracy": 0.195272795855999,
|
|
"num_tokens": 8225377.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.950549459457397,
|
|
"epoch": 4.180009945300846,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00037459959155845726,
|
|
"loss": 5.5088,
|
|
"mean_token_accuracy": 0.18158449232578278,
|
|
"num_tokens": 8235436.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.938119316101075,
|
|
"epoch": 4.184982595723521,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003742492589498744,
|
|
"loss": 5.4258,
|
|
"mean_token_accuracy": 0.18466287553310395,
|
|
"num_tokens": 8245129.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 5.912076234817505,
|
|
"epoch": 4.189955246146196,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00037389862734227525,
|
|
"loss": 5.3654,
|
|
"mean_token_accuracy": 0.19315945208072663,
|
|
"num_tokens": 8254870.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.9881573677062985,
|
|
"epoch": 4.194927896568871,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037354769779197514,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.1829470157623291,
|
|
"num_tokens": 8266078.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.004723453521729,
|
|
"epoch": 4.199900546991547,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037319647135618695,
|
|
"loss": 5.4747,
|
|
"mean_token_accuracy": 0.1898008942604065,
|
|
"num_tokens": 8275833.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.852568864822388,
|
|
"epoch": 4.204873197414222,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037284494909301793,
|
|
"loss": 5.3961,
|
|
"mean_token_accuracy": 0.18148481994867324,
|
|
"num_tokens": 8286866.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.992520713806153,
|
|
"epoch": 4.209845847836897,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003724931320614663,
|
|
"loss": 5.4066,
|
|
"mean_token_accuracy": 0.1872383952140808,
|
|
"num_tokens": 8296316.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.053286743164063,
|
|
"epoch": 4.214818498259572,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037214102132141874,
|
|
"loss": 5.5774,
|
|
"mean_token_accuracy": 0.1752246618270874,
|
|
"num_tokens": 8306023.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.864358949661255,
|
|
"epoch": 4.219791148682248,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003717886179336465,
|
|
"loss": 5.3101,
|
|
"mean_token_accuracy": 0.19370326399803162,
|
|
"num_tokens": 8315808.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.950646018981933,
|
|
"epoch": 4.224763799104923,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00037143592295980243,
|
|
"loss": 5.3995,
|
|
"mean_token_accuracy": 0.19023952931165694,
|
|
"num_tokens": 8324951.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.90330057144165,
|
|
"epoch": 4.229736449527598,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003710829374624181,
|
|
"loss": 5.3966,
|
|
"mean_token_accuracy": 0.18766625374555587,
|
|
"num_tokens": 8335019.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.95871114730835,
|
|
"epoch": 4.2347090999502734,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003707296625048999,
|
|
"loss": 5.3838,
|
|
"mean_token_accuracy": 0.19565755128860474,
|
|
"num_tokens": 8344496.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.971175050735473,
|
|
"epoch": 4.239681750372949,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003703760991515264,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.18413328677415847,
|
|
"num_tokens": 8353766.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.923512506484985,
|
|
"epoch": 4.244654400795624,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00037002224846744526,
|
|
"loss": 5.4076,
|
|
"mean_token_accuracy": 0.17967646718025207,
|
|
"num_tokens": 8363576.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.043702936172485,
|
|
"epoch": 4.249627051218299,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00036966811151866943,
|
|
"loss": 5.4945,
|
|
"mean_token_accuracy": 0.17869742065668107,
|
|
"num_tokens": 8372714.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.910320472717285,
|
|
"epoch": 4.2545997016409745,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036931368937207436,
|
|
"loss": 5.4033,
|
|
"mean_token_accuracy": 0.19694521129131318,
|
|
"num_tokens": 8383563.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.892403173446655,
|
|
"epoch": 4.25957235206365,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003689589830953948,
|
|
"loss": 5.3088,
|
|
"mean_token_accuracy": 0.2015668511390686,
|
|
"num_tokens": 8393352.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.935925483703613,
|
|
"epoch": 4.264545002486325,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003686039937572213,
|
|
"loss": 5.4514,
|
|
"mean_token_accuracy": 0.18705656230449677,
|
|
"num_tokens": 8403219.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.892400312423706,
|
|
"epoch": 4.269517652909,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003682487224269972,
|
|
"loss": 5.3208,
|
|
"mean_token_accuracy": 0.20029816776514053,
|
|
"num_tokens": 8411596.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.9044867038726805,
|
|
"epoch": 4.2744903033316755,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003678931701750155,
|
|
"loss": 5.4232,
|
|
"mean_token_accuracy": 0.18545915484428405,
|
|
"num_tokens": 8421472.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.8963850975036625,
|
|
"epoch": 4.279462953754351,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00036753733807241536,
|
|
"loss": 5.3322,
|
|
"mean_token_accuracy": 0.20247312039136886,
|
|
"num_tokens": 8431489.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.7858113765716555,
|
|
"epoch": 4.284435604177026,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00036718122719117907,
|
|
"loss": 5.1755,
|
|
"mean_token_accuracy": 0.2115116074681282,
|
|
"num_tokens": 8441108.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.808594846725464,
|
|
"epoch": 4.289408254599701,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003668248386041288,
|
|
"loss": 5.2432,
|
|
"mean_token_accuracy": 0.20886893570423126,
|
|
"num_tokens": 8451664.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 5.927479982376099,
|
|
"epoch": 4.294380905022377,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036646817338492334,
|
|
"loss": 5.4272,
|
|
"mean_token_accuracy": 0.18317549377679826,
|
|
"num_tokens": 8460531.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.948016119003296,
|
|
"epoch": 4.299353555445053,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003661112326080548,
|
|
"loss": 5.3926,
|
|
"mean_token_accuracy": 0.19342503547668458,
|
|
"num_tokens": 8470018.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.865046262741089,
|
|
"epoch": 4.304326205867728,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003657540173488455,
|
|
"loss": 5.3409,
|
|
"mean_token_accuracy": 0.19755968153476716,
|
|
"num_tokens": 8480095.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.966604995727539,
|
|
"epoch": 4.309298856290403,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003653965286834446,
|
|
"loss": 5.4864,
|
|
"mean_token_accuracy": 0.18020871579647063,
|
|
"num_tokens": 8489817.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.006196451187134,
|
|
"epoch": 4.314271506713078,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003650387676888251,
|
|
"loss": 5.4903,
|
|
"mean_token_accuracy": 0.17801908552646636,
|
|
"num_tokens": 8499333.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.975729370117188,
|
|
"epoch": 4.319244157135754,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003646807354427802,
|
|
"loss": 5.4657,
|
|
"mean_token_accuracy": 0.1757289931178093,
|
|
"num_tokens": 8508732.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.913530874252319,
|
|
"epoch": 4.324216807558429,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036432243302392025,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.18204899430274962,
|
|
"num_tokens": 8519525.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.854799747467041,
|
|
"epoch": 4.329189457981104,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036396386151166975,
|
|
"loss": 5.3349,
|
|
"mean_token_accuracy": 0.20804867893457413,
|
|
"num_tokens": 8529559.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.958149147033692,
|
|
"epoch": 4.334162108403779,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003636050219862638,
|
|
"loss": 5.385,
|
|
"mean_token_accuracy": 0.18482863157987595,
|
|
"num_tokens": 8538615.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.895068359375,
|
|
"epoch": 4.339134758826455,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036324591552874493,
|
|
"loss": 5.3375,
|
|
"mean_token_accuracy": 0.18990998864173889,
|
|
"num_tokens": 8548432.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.825157928466797,
|
|
"epoch": 4.34410740924913,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036288654322095973,
|
|
"loss": 5.3249,
|
|
"mean_token_accuracy": 0.19903762638568878,
|
|
"num_tokens": 8558216.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.933069086074829,
|
|
"epoch": 4.349080059671805,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036252690614555557,
|
|
"loss": 5.2857,
|
|
"mean_token_accuracy": 0.19398757368326186,
|
|
"num_tokens": 8568234.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.9437792778015135,
|
|
"epoch": 4.35405271009448,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003621670053859778,
|
|
"loss": 5.4305,
|
|
"mean_token_accuracy": 0.18686269074678422,
|
|
"num_tokens": 8578140.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 5.90855188369751,
|
|
"epoch": 4.359025360517156,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000361806842026466,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.18879732340574265,
|
|
"num_tokens": 8587976.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.887794733047485,
|
|
"epoch": 4.363998010939831,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00036144641715205077,
|
|
"loss": 5.4467,
|
|
"mean_token_accuracy": 0.18811312913894654,
|
|
"num_tokens": 8597739.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.96191258430481,
|
|
"epoch": 4.368970661362506,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003610857318485507,
|
|
"loss": 5.5042,
|
|
"mean_token_accuracy": 0.18207356035709382,
|
|
"num_tokens": 8608609.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.931263732910156,
|
|
"epoch": 4.373943311785181,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036072478720256885,
|
|
"loss": 5.4156,
|
|
"mean_token_accuracy": 0.1854161724448204,
|
|
"num_tokens": 8617870.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.909566593170166,
|
|
"epoch": 4.378915962207857,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003603635843014895,
|
|
"loss": 5.3904,
|
|
"mean_token_accuracy": 0.19139891266822814,
|
|
"num_tokens": 8627811.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 5.979073429107666,
|
|
"epoch": 4.383888612630532,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00036000212423347523,
|
|
"loss": 5.4727,
|
|
"mean_token_accuracy": 0.18408248871564864,
|
|
"num_tokens": 8636989.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.914165782928467,
|
|
"epoch": 4.388861263053207,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000359640408087463,
|
|
"loss": 5.3181,
|
|
"mean_token_accuracy": 0.19404911547899245,
|
|
"num_tokens": 8646090.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.874205112457275,
|
|
"epoch": 4.393833913475882,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003592784369531615,
|
|
"loss": 5.3333,
|
|
"mean_token_accuracy": 0.19463734328746796,
|
|
"num_tokens": 8656244.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.78307089805603,
|
|
"epoch": 4.398806563898558,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00035891621192104764,
|
|
"loss": 5.2923,
|
|
"mean_token_accuracy": 0.2012870654463768,
|
|
"num_tokens": 8665783.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.9753392219543455,
|
|
"epoch": 4.403779214321233,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00035855373408236294,
|
|
"loss": 5.5099,
|
|
"mean_token_accuracy": 0.1828705608844757,
|
|
"num_tokens": 8676736.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.956588411331177,
|
|
"epoch": 4.408751864743908,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003581910045291107,
|
|
"loss": 5.4345,
|
|
"mean_token_accuracy": 0.18966108411550522,
|
|
"num_tokens": 8687035.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.981082201004028,
|
|
"epoch": 4.413724515166583,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003578280243540526,
|
|
"loss": 5.4704,
|
|
"mean_token_accuracy": 0.18727438598871232,
|
|
"num_tokens": 8696939.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.885527849197388,
|
|
"epoch": 4.4186971655892595,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00035746479465070515,
|
|
"loss": 5.4492,
|
|
"mean_token_accuracy": 0.18996560871601104,
|
|
"num_tokens": 8707213.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.878031635284424,
|
|
"epoch": 4.423669816011935,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00035710131651333683,
|
|
"loss": 5.395,
|
|
"mean_token_accuracy": 0.1882211059331894,
|
|
"num_tokens": 8717066.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.963189029693604,
|
|
"epoch": 4.42864246643461,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035673759103696447,
|
|
"loss": 5.4221,
|
|
"mean_token_accuracy": 0.1869526296854019,
|
|
"num_tokens": 8727421.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.950403213500977,
|
|
"epoch": 4.433615116857285,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003563736193173499,
|
|
"loss": 5.4718,
|
|
"mean_token_accuracy": 0.18542165011167527,
|
|
"num_tokens": 8736086.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.97789421081543,
|
|
"epoch": 4.4385877672799605,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003560094024509969,
|
|
"loss": 5.4847,
|
|
"mean_token_accuracy": 0.19082806408405303,
|
|
"num_tokens": 8746496.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.9135480403900145,
|
|
"epoch": 4.443560417702636,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035564494153514777,
|
|
"loss": 5.3399,
|
|
"mean_token_accuracy": 0.19800452291965484,
|
|
"num_tokens": 8755528.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.947375106811523,
|
|
"epoch": 4.448533068125311,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00035528023766778,
|
|
"loss": 5.4509,
|
|
"mean_token_accuracy": 0.18112998604774475,
|
|
"num_tokens": 8765486.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.960193204879761,
|
|
"epoch": 4.453505718547986,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003549152919476031,
|
|
"loss": 5.476,
|
|
"mean_token_accuracy": 0.18522600829601288,
|
|
"num_tokens": 8775315.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.9549473285675045,
|
|
"epoch": 4.4584783689706615,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035455010547405513,
|
|
"loss": 5.4858,
|
|
"mean_token_accuracy": 0.1813558518886566,
|
|
"num_tokens": 8784976.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.91564245223999,
|
|
"epoch": 4.463451019393337,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035418467934729934,
|
|
"loss": 5.5115,
|
|
"mean_token_accuracy": 0.17875563055276872,
|
|
"num_tokens": 8795412.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.8935364246368405,
|
|
"epoch": 4.468423669816012,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00035381901466822113,
|
|
"loss": 5.3662,
|
|
"mean_token_accuracy": 0.19603478014469147,
|
|
"num_tokens": 8805277.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.89279465675354,
|
|
"epoch": 4.473396320238687,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003534531125384244,
|
|
"loss": 5.4277,
|
|
"mean_token_accuracy": 0.18796817362308502,
|
|
"num_tokens": 8814308.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 4.473396320238687,
|
|
"eval_entropy": 5.822956018184187,
|
|
"eval_loss": 6.54652214050293,
|
|
"eval_mean_token_accuracy": 0.1425970572600579,
|
|
"eval_num_tokens": 8814308.0,
|
|
"eval_runtime": 13.6428,
|
|
"eval_samples_per_second": 253.98,
|
|
"eval_steps_per_second": 31.812,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.9118977069854735,
|
|
"epoch": 4.4783689706613625,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035308697406022865,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.18483713418245315,
|
|
"num_tokens": 8824836.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.850842905044556,
|
|
"epoch": 4.483341621084038,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035272060033666503,
|
|
"loss": 5.3172,
|
|
"mean_token_accuracy": 0.1941648930311203,
|
|
"num_tokens": 8834390.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.994395732879639,
|
|
"epoch": 4.488314271506713,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003523539924714738,
|
|
"loss": 5.5171,
|
|
"mean_token_accuracy": 0.18296067267656327,
|
|
"num_tokens": 8844266.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.958738946914673,
|
|
"epoch": 4.493286921929388,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00035198715156910016,
|
|
"loss": 5.3995,
|
|
"mean_token_accuracy": 0.18978705108165742,
|
|
"num_tokens": 8852876.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.736412906646729,
|
|
"epoch": 4.4982595723520635,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035162007873469175,
|
|
"loss": 5.1331,
|
|
"mean_token_accuracy": 0.21699246168136596,
|
|
"num_tokens": 8862861.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.892131757736206,
|
|
"epoch": 4.503232222774739,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035125277507409483,
|
|
"loss": 5.3792,
|
|
"mean_token_accuracy": 0.19149543195962906,
|
|
"num_tokens": 8872177.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.793962049484253,
|
|
"epoch": 4.508204873197414,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00035088524169385075,
|
|
"loss": 5.4128,
|
|
"mean_token_accuracy": 0.1900569647550583,
|
|
"num_tokens": 8882202.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.896869277954101,
|
|
"epoch": 4.513177523620089,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00035051747970119354,
|
|
"loss": 5.4284,
|
|
"mean_token_accuracy": 0.19023434817790985,
|
|
"num_tokens": 8892361.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.012280988693237,
|
|
"epoch": 4.5181501740427645,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00035014949020404523,
|
|
"loss": 5.5037,
|
|
"mean_token_accuracy": 0.18297217786312103,
|
|
"num_tokens": 8903014.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.922554397583008,
|
|
"epoch": 4.52312282446544,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034978127431101365,
|
|
"loss": 5.4871,
|
|
"mean_token_accuracy": 0.17989742010831833,
|
|
"num_tokens": 8912145.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.86684741973877,
|
|
"epoch": 4.528095474888115,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034941283313138865,
|
|
"loss": 5.3972,
|
|
"mean_token_accuracy": 0.19028664082288743,
|
|
"num_tokens": 8921748.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.925622463226318,
|
|
"epoch": 4.53306812531079,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003490441677751387,
|
|
"loss": 5.4922,
|
|
"mean_token_accuracy": 0.18012350648641587,
|
|
"num_tokens": 8932327.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.973186683654785,
|
|
"epoch": 4.5380407757334655,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034867527935290773,
|
|
"loss": 5.508,
|
|
"mean_token_accuracy": 0.1775151878595352,
|
|
"num_tokens": 8942781.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 5.968158483505249,
|
|
"epoch": 4.543013426156142,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034830616897601146,
|
|
"loss": 5.4861,
|
|
"mean_token_accuracy": 0.1806615859270096,
|
|
"num_tokens": 8951887.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.943377065658569,
|
|
"epoch": 4.547986076578816,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003479368377564344,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.17926227897405625,
|
|
"num_tokens": 8961454.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.922529458999634,
|
|
"epoch": 4.552958727001492,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003475672868068265,
|
|
"loss": 5.3685,
|
|
"mean_token_accuracy": 0.18969711661338806,
|
|
"num_tokens": 8970218.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.933936643600464,
|
|
"epoch": 4.557931377424167,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00034719751724049955,
|
|
"loss": 5.4057,
|
|
"mean_token_accuracy": 0.18795102685689927,
|
|
"num_tokens": 8980141.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.853798723220825,
|
|
"epoch": 4.562904027846843,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034682753017142387,
|
|
"loss": 5.4101,
|
|
"mean_token_accuracy": 0.19265685379505157,
|
|
"num_tokens": 8989514.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.896053171157837,
|
|
"epoch": 4.567876678269518,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034645732671422517,
|
|
"loss": 5.5179,
|
|
"mean_token_accuracy": 0.18127399832010269,
|
|
"num_tokens": 9000054.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.932244491577149,
|
|
"epoch": 4.572849328692193,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003460869079841809,
|
|
"loss": 5.5184,
|
|
"mean_token_accuracy": 0.18970169126987457,
|
|
"num_tokens": 9010070.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.995617341995239,
|
|
"epoch": 4.577821979114868,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003457162750972172,
|
|
"loss": 5.4887,
|
|
"mean_token_accuracy": 0.18080704808235168,
|
|
"num_tokens": 9019785.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.925299119949341,
|
|
"epoch": 4.582794629537544,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00034534542916990534,
|
|
"loss": 5.4679,
|
|
"mean_token_accuracy": 0.18388414680957793,
|
|
"num_tokens": 9028281.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.776969528198242,
|
|
"epoch": 4.587767279960219,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034497437131945837,
|
|
"loss": 5.2566,
|
|
"mean_token_accuracy": 0.2058468982577324,
|
|
"num_tokens": 9037628.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 6.006793212890625,
|
|
"epoch": 4.592739930382894,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003446031026637277,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.18150182515382768,
|
|
"num_tokens": 9047939.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.969013977050781,
|
|
"epoch": 4.597712580805569,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00034423162432119995,
|
|
"loss": 5.4554,
|
|
"mean_token_accuracy": 0.18919723480939865,
|
|
"num_tokens": 9056583.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.830774879455566,
|
|
"epoch": 4.602685231228245,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003438599374109934,
|
|
"loss": 5.4228,
|
|
"mean_token_accuracy": 0.19052909314632416,
|
|
"num_tokens": 9067610.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.943630695343018,
|
|
"epoch": 4.60765788165092,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034348804305285466,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.18154219835996627,
|
|
"num_tokens": 9077958.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.962500333786011,
|
|
"epoch": 4.612630532073595,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003431159423671552,
|
|
"loss": 5.5074,
|
|
"mean_token_accuracy": 0.18074077665805816,
|
|
"num_tokens": 9088821.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 5.886448907852173,
|
|
"epoch": 4.61760318249627,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034274363647488826,
|
|
"loss": 5.3673,
|
|
"mean_token_accuracy": 0.19161462485790254,
|
|
"num_tokens": 9098495.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.90208625793457,
|
|
"epoch": 4.622575832918946,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003423711264976652,
|
|
"loss": 5.4093,
|
|
"mean_token_accuracy": 0.1892756849527359,
|
|
"num_tokens": 9107279.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.944086599349975,
|
|
"epoch": 4.627548483341621,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034199841355771226,
|
|
"loss": 5.4621,
|
|
"mean_token_accuracy": 0.1810019999742508,
|
|
"num_tokens": 9118182.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 5.904891538619995,
|
|
"epoch": 4.632521133764296,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034162549877786685,
|
|
"loss": 5.4005,
|
|
"mean_token_accuracy": 0.19171113073825835,
|
|
"num_tokens": 9128299.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.987291431427002,
|
|
"epoch": 4.6374937841869714,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003412523832815748,
|
|
"loss": 5.5583,
|
|
"mean_token_accuracy": 0.1722610667347908,
|
|
"num_tokens": 9138097.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.824917840957641,
|
|
"epoch": 4.642466434609647,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003408790681928865,
|
|
"loss": 5.3217,
|
|
"mean_token_accuracy": 0.19245031476020813,
|
|
"num_tokens": 9148951.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.949419593811035,
|
|
"epoch": 4.647439085032322,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003405055546364535,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.19261041283607483,
|
|
"num_tokens": 9158372.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.811802101135254,
|
|
"epoch": 4.652411735454997,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034013184373752557,
|
|
"loss": 5.2917,
|
|
"mean_token_accuracy": 0.1995917961001396,
|
|
"num_tokens": 9168308.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.784817886352539,
|
|
"epoch": 4.6573843858776725,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003397579366219466,
|
|
"loss": 5.3162,
|
|
"mean_token_accuracy": 0.20091514736413957,
|
|
"num_tokens": 9178200.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.904370737075806,
|
|
"epoch": 4.662357036300348,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003393838344161519,
|
|
"loss": 5.4868,
|
|
"mean_token_accuracy": 0.18642538636922837,
|
|
"num_tokens": 9187759.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.674773740768432,
|
|
"epoch": 4.667329686723024,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00033900953824716443,
|
|
"loss": 5.1095,
|
|
"mean_token_accuracy": 0.21913074105978012,
|
|
"num_tokens": 9198530.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.891289520263672,
|
|
"epoch": 4.672302337145698,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003386350492425913,
|
|
"loss": 5.4286,
|
|
"mean_token_accuracy": 0.18640197664499283,
|
|
"num_tokens": 9207668.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.966985082626342,
|
|
"epoch": 4.677274987568374,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033826036853062095,
|
|
"loss": 5.5658,
|
|
"mean_token_accuracy": 0.1805041417479515,
|
|
"num_tokens": 9216856.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 5.924076080322266,
|
|
"epoch": 4.68224763799105,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033788549724001897,
|
|
"loss": 5.4624,
|
|
"mean_token_accuracy": 0.1825665295124054,
|
|
"num_tokens": 9226307.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.8213876247406,
|
|
"epoch": 4.687220288413725,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033751043650012534,
|
|
"loss": 5.428,
|
|
"mean_token_accuracy": 0.18701232969760895,
|
|
"num_tokens": 9235919.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.757311201095581,
|
|
"epoch": 4.6921929388364,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003371351874408506,
|
|
"loss": 5.1717,
|
|
"mean_token_accuracy": 0.20693135112524033,
|
|
"num_tokens": 9245542.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.006253719329834,
|
|
"epoch": 4.697165589259075,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003367597511926728,
|
|
"loss": 5.5284,
|
|
"mean_token_accuracy": 0.18441032469272614,
|
|
"num_tokens": 9256146.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.9626096248626705,
|
|
"epoch": 4.702138239681751,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033638412888663374,
|
|
"loss": 5.5684,
|
|
"mean_token_accuracy": 0.17925261706113815,
|
|
"num_tokens": 9266694.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.8828061580657955,
|
|
"epoch": 4.707110890104426,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003360083216543358,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.18103974610567092,
|
|
"num_tokens": 9276155.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.930776691436767,
|
|
"epoch": 4.712083540527101,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003356323306279387,
|
|
"loss": 5.4335,
|
|
"mean_token_accuracy": 0.18625507056713103,
|
|
"num_tokens": 9286048.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 6.016036081314087,
|
|
"epoch": 4.717056190949776,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00033525615694015545,
|
|
"loss": 5.5538,
|
|
"mean_token_accuracy": 0.1771557793021202,
|
|
"num_tokens": 9296515.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.764087295532226,
|
|
"epoch": 4.722028841372452,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033487980172424955,
|
|
"loss": 5.3379,
|
|
"mean_token_accuracy": 0.20236023813486098,
|
|
"num_tokens": 9306737.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.9372289180755615,
|
|
"epoch": 4.727001491795127,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033450326611403136,
|
|
"loss": 5.4871,
|
|
"mean_token_accuracy": 0.18094107061624526,
|
|
"num_tokens": 9316453.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.953107309341431,
|
|
"epoch": 4.731974142217802,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033412655124385494,
|
|
"loss": 5.4438,
|
|
"mean_token_accuracy": 0.19174361526966094,
|
|
"num_tokens": 9327422.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.915654802322388,
|
|
"epoch": 4.736946792640477,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003337496582486139,
|
|
"loss": 5.3968,
|
|
"mean_token_accuracy": 0.19034271091222763,
|
|
"num_tokens": 9336890.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.934427928924561,
|
|
"epoch": 4.741919443063153,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033337258826373896,
|
|
"loss": 5.4673,
|
|
"mean_token_accuracy": 0.18873524516820908,
|
|
"num_tokens": 9346807.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.967929840087891,
|
|
"epoch": 4.746892093485828,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033299534242519355,
|
|
"loss": 5.4236,
|
|
"mean_token_accuracy": 0.1856658712029457,
|
|
"num_tokens": 9356343.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.932782793045044,
|
|
"epoch": 4.751864743908503,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033261792186947133,
|
|
"loss": 5.473,
|
|
"mean_token_accuracy": 0.18221897929906844,
|
|
"num_tokens": 9366936.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.92764892578125,
|
|
"epoch": 4.756837394331178,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00033224032773359203,
|
|
"loss": 5.4199,
|
|
"mean_token_accuracy": 0.1950580134987831,
|
|
"num_tokens": 9377520.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.9402083396911625,
|
|
"epoch": 4.761810044753854,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033186256115509845,
|
|
"loss": 5.5463,
|
|
"mean_token_accuracy": 0.18014792799949647,
|
|
"num_tokens": 9386498.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 5.921793508529663,
|
|
"epoch": 4.766782695176529,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003314846232720528,
|
|
"loss": 5.4755,
|
|
"mean_token_accuracy": 0.18086444437503815,
|
|
"num_tokens": 9396153.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 5.964250135421753,
|
|
"epoch": 4.771755345599204,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033110651522303346,
|
|
"loss": 5.4783,
|
|
"mean_token_accuracy": 0.18877972215414046,
|
|
"num_tokens": 9406019.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 5.871837186813354,
|
|
"epoch": 4.776727996021879,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003307282381471314,
|
|
"loss": 5.4505,
|
|
"mean_token_accuracy": 0.18326770067214965,
|
|
"num_tokens": 9416019.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.8644036769866945,
|
|
"epoch": 4.781700646444555,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00033034979318394676,
|
|
"loss": 5.4281,
|
|
"mean_token_accuracy": 0.19091861695051193,
|
|
"num_tokens": 9425874.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.960529232025147,
|
|
"epoch": 4.78667329686723,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00032997118147358554,
|
|
"loss": 5.4798,
|
|
"mean_token_accuracy": 0.18944692760705947,
|
|
"num_tokens": 9435252.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 5.949538660049439,
|
|
"epoch": 4.791645947289906,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000329592404156656,
|
|
"loss": 5.4218,
|
|
"mean_token_accuracy": 0.1829541727900505,
|
|
"num_tokens": 9444696.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 5.899854564666748,
|
|
"epoch": 4.79661859771258,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032921346237426536,
|
|
"loss": 5.4608,
|
|
"mean_token_accuracy": 0.1830321729183197,
|
|
"num_tokens": 9454163.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 5.824109220504761,
|
|
"epoch": 4.8015912481352565,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032883435726801635,
|
|
"loss": 5.2583,
|
|
"mean_token_accuracy": 0.2064215898513794,
|
|
"num_tokens": 9463058.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.845318031311035,
|
|
"epoch": 4.806563898557932,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003284550899800036,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.19094939529895782,
|
|
"num_tokens": 9472302.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 5.960778713226318,
|
|
"epoch": 4.811536548980607,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00032807566165281055,
|
|
"loss": 5.4564,
|
|
"mean_token_accuracy": 0.18604158461093903,
|
|
"num_tokens": 9482607.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.963613796234131,
|
|
"epoch": 4.816509199403282,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003276960734295054,
|
|
"loss": 5.4465,
|
|
"mean_token_accuracy": 0.1893915370106697,
|
|
"num_tokens": 9492451.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.9191468238830565,
|
|
"epoch": 4.8214818498259575,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032731632645363856,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.17867228090763093,
|
|
"num_tokens": 9502633.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 5.833761167526245,
|
|
"epoch": 4.826454500248633,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003269364218692384,
|
|
"loss": 5.3791,
|
|
"mean_token_accuracy": 0.1943543002009392,
|
|
"num_tokens": 9512887.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.870547199249268,
|
|
"epoch": 4.831427150671308,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003265563608208081,
|
|
"loss": 5.3981,
|
|
"mean_token_accuracy": 0.194460828602314,
|
|
"num_tokens": 9522210.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.814197301864624,
|
|
"epoch": 4.836399801093983,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003261761444533221,
|
|
"loss": 5.2675,
|
|
"mean_token_accuracy": 0.20554081946611405,
|
|
"num_tokens": 9532790.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.923008441925049,
|
|
"epoch": 4.8413724515166585,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003257957739122231,
|
|
"loss": 5.4272,
|
|
"mean_token_accuracy": 0.19084182530641555,
|
|
"num_tokens": 9543429.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.950304079055786,
|
|
"epoch": 4.846345101939334,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032541525034341805,
|
|
"loss": 5.5166,
|
|
"mean_token_accuracy": 0.17799354940652848,
|
|
"num_tokens": 9552722.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.890521574020386,
|
|
"epoch": 4.851317752362009,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003250345748932749,
|
|
"loss": 5.4011,
|
|
"mean_token_accuracy": 0.18751320689916612,
|
|
"num_tokens": 9560880.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 5.957213354110718,
|
|
"epoch": 4.856290402784684,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032465374870861913,
|
|
"loss": 5.4735,
|
|
"mean_token_accuracy": 0.18342356532812118,
|
|
"num_tokens": 9570553.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.8826569557189945,
|
|
"epoch": 4.8612630532073595,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032427277293673053,
|
|
"loss": 5.5101,
|
|
"mean_token_accuracy": 0.1779240369796753,
|
|
"num_tokens": 9581313.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.902159309387207,
|
|
"epoch": 4.866235703630035,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003238916487253393,
|
|
"loss": 5.4534,
|
|
"mean_token_accuracy": 0.1881001278758049,
|
|
"num_tokens": 9589545.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 5.919522285461426,
|
|
"epoch": 4.87120835405271,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003235103772226229,
|
|
"loss": 5.4802,
|
|
"mean_token_accuracy": 0.1821992054581642,
|
|
"num_tokens": 9599489.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.904637622833252,
|
|
"epoch": 4.876181004475385,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003231289595772026,
|
|
"loss": 5.5462,
|
|
"mean_token_accuracy": 0.18283843249082565,
|
|
"num_tokens": 9609346.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 5.721666479110718,
|
|
"epoch": 4.8811536548980605,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003227473969381398,
|
|
"loss": 5.1932,
|
|
"mean_token_accuracy": 0.21505673080682755,
|
|
"num_tokens": 9618808.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 5.910362243652344,
|
|
"epoch": 4.886126305320736,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032236569045493283,
|
|
"loss": 5.3849,
|
|
"mean_token_accuracy": 0.1890556052327156,
|
|
"num_tokens": 9629312.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 5.98698205947876,
|
|
"epoch": 4.891098955743411,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032198384127751344,
|
|
"loss": 5.5828,
|
|
"mean_token_accuracy": 0.17863494008779526,
|
|
"num_tokens": 9639439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 5.9798455238342285,
|
|
"epoch": 4.896071606166086,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000321601850556243,
|
|
"loss": 5.4098,
|
|
"mean_token_accuracy": 0.18349761217832566,
|
|
"num_tokens": 9647824.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 5.8910962581634525,
|
|
"epoch": 4.9010442565887615,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003212197194419095,
|
|
"loss": 5.4487,
|
|
"mean_token_accuracy": 0.18705722242593764,
|
|
"num_tokens": 9656962.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.889167737960816,
|
|
"epoch": 4.906016907011437,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003208374490857238,
|
|
"loss": 5.5136,
|
|
"mean_token_accuracy": 0.18630672693252565,
|
|
"num_tokens": 9666360.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 5.948098516464233,
|
|
"epoch": 4.910989557434112,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003204550406393164,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.18107982277870177,
|
|
"num_tokens": 9676355.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 5.964854669570923,
|
|
"epoch": 4.915962207856787,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003200724952547335,
|
|
"loss": 5.4852,
|
|
"mean_token_accuracy": 0.1879056379199028,
|
|
"num_tokens": 9686361.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 5.893513584136963,
|
|
"epoch": 4.9209348582794625,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003196898140844341,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.19281947314739228,
|
|
"num_tokens": 9696571.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 6.013906478881836,
|
|
"epoch": 4.925907508702139,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00031930699828128614,
|
|
"loss": 5.6164,
|
|
"mean_token_accuracy": 0.17802755609154702,
|
|
"num_tokens": 9707595.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 5.946589183807373,
|
|
"epoch": 4.930880159124813,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003189240489985633,
|
|
"loss": 5.4923,
|
|
"mean_token_accuracy": 0.18939886391162872,
|
|
"num_tokens": 9717094.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.875442790985107,
|
|
"epoch": 4.935852809547489,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003185409673899412,
|
|
"loss": 5.4416,
|
|
"mean_token_accuracy": 0.1899618536233902,
|
|
"num_tokens": 9727329.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 5.8156603336334225,
|
|
"epoch": 4.940825459970164,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00031815775460949426,
|
|
"loss": 5.3153,
|
|
"mean_token_accuracy": 0.20046168118715285,
|
|
"num_tokens": 9736189.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 5.931135416030884,
|
|
"epoch": 4.94579811039284,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003177744118116919,
|
|
"loss": 5.397,
|
|
"mean_token_accuracy": 0.1889800548553467,
|
|
"num_tokens": 9745760.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 6.02621898651123,
|
|
"epoch": 4.950770760815515,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003173909401513954,
|
|
"loss": 5.5614,
|
|
"mean_token_accuracy": 0.17368809431791304,
|
|
"num_tokens": 9755941.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 5.888241195678711,
|
|
"epoch": 4.95574341123819,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00031700734078385416,
|
|
"loss": 5.4018,
|
|
"mean_token_accuracy": 0.18955527991056442,
|
|
"num_tokens": 9764546.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 5.9381632804870605,
|
|
"epoch": 4.960716061660865,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00031662361486470243,
|
|
"loss": 5.4497,
|
|
"mean_token_accuracy": 0.18426193296909332,
|
|
"num_tokens": 9773728.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.861423301696777,
|
|
"epoch": 4.965688712083541,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003162397635499555,
|
|
"loss": 5.4397,
|
|
"mean_token_accuracy": 0.18276682496070862,
|
|
"num_tokens": 9783878.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 5.926243305206299,
|
|
"epoch": 4.970661362506216,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00031585578799600653,
|
|
"loss": 5.4812,
|
|
"mean_token_accuracy": 0.18322205245494844,
|
|
"num_tokens": 9793926.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.970661362506216,
|
|
"eval_entropy": 5.845926248532836,
|
|
"eval_loss": 6.493011474609375,
|
|
"eval_mean_token_accuracy": 0.1452222225869993,
|
|
"eval_num_tokens": 9793926.0,
|
|
"eval_runtime": 12.6365,
|
|
"eval_samples_per_second": 274.205,
|
|
"eval_steps_per_second": 34.345,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.893375396728516,
|
|
"epoch": 4.975634012928891,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031547168935962304,
|
|
"loss": 5.3851,
|
|
"mean_token_accuracy": 0.1958031475543976,
|
|
"num_tokens": 9804213.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 5.889259576797485,
|
|
"epoch": 4.980606663351566,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003150874687979433,
|
|
"loss": 5.4537,
|
|
"mean_token_accuracy": 0.18845126032829285,
|
|
"num_tokens": 9814151.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.870977878570557,
|
|
"epoch": 4.985579313774242,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031470312746847284,
|
|
"loss": 5.4221,
|
|
"mean_token_accuracy": 0.18838190138339997,
|
|
"num_tokens": 9822646.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 5.7266449451446535,
|
|
"epoch": 4.990551964196917,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031431866652908107,
|
|
"loss": 5.193,
|
|
"mean_token_accuracy": 0.21573361456394197,
|
|
"num_tokens": 9833204.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.801643180847168,
|
|
"epoch": 4.995524614619592,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003139340871379977,
|
|
"loss": 5.3408,
|
|
"mean_token_accuracy": 0.2173926427960396,
|
|
"num_tokens": 9843150.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.915289296044244,
|
|
"epoch": 5.0,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 0.0003135493904538093,
|
|
"loss": 5.397,
|
|
"mean_token_accuracy": 0.18821010490258536,
|
|
"num_tokens": 9850520.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.893287944793701,
|
|
"epoch": 5.004972650422675,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003131645776354559,
|
|
"loss": 5.1388,
|
|
"mean_token_accuracy": 0.20298085808753968,
|
|
"num_tokens": 9860876.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 5.806403350830078,
|
|
"epoch": 5.0099453008453505,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0003127796498422272,
|
|
"loss": 5.0393,
|
|
"mean_token_accuracy": 0.22022406458854676,
|
|
"num_tokens": 9870944.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 5.861393117904663,
|
|
"epoch": 5.014917951268026,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031239460823375933,
|
|
"loss": 5.0954,
|
|
"mean_token_accuracy": 0.2075657308101654,
|
|
"num_tokens": 9880615.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 5.720591688156128,
|
|
"epoch": 5.019890601690701,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031200945397003157,
|
|
"loss": 4.9349,
|
|
"mean_token_accuracy": 0.22961988896131516,
|
|
"num_tokens": 9889933.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.913609313964844,
|
|
"epoch": 5.024863252113376,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003116241882113623,
|
|
"loss": 5.1931,
|
|
"mean_token_accuracy": 0.19516998380422593,
|
|
"num_tokens": 9900166.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 5.787868499755859,
|
|
"epoch": 5.0298359025360515,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00031123881211840565,
|
|
"loss": 5.1057,
|
|
"mean_token_accuracy": 0.2080114886164665,
|
|
"num_tokens": 9909747.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 5.839036273956299,
|
|
"epoch": 5.034808552958727,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003108533268521485,
|
|
"loss": 5.1459,
|
|
"mean_token_accuracy": 0.21180519461631775,
|
|
"num_tokens": 9919039.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.9464012622833256,
|
|
"epoch": 5.039781203381402,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003104677335739065,
|
|
"loss": 5.1578,
|
|
"mean_token_accuracy": 0.21298085302114486,
|
|
"num_tokens": 9927933.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.886098957061767,
|
|
"epoch": 5.044753853804077,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00031008203344532067,
|
|
"loss": 5.1624,
|
|
"mean_token_accuracy": 0.19933608621358873,
|
|
"num_tokens": 9939264.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.846206474304199,
|
|
"epoch": 5.0497265042267525,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000309696227628354,
|
|
"loss": 5.1852,
|
|
"mean_token_accuracy": 0.2022075355052948,
|
|
"num_tokens": 9948375.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 5.86159873008728,
|
|
"epoch": 5.054699154649428,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030931031728528754,
|
|
"loss": 5.0578,
|
|
"mean_token_accuracy": 0.21181212663650512,
|
|
"num_tokens": 9957792.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 5.84744963645935,
|
|
"epoch": 5.059671805072103,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003089243035787178,
|
|
"loss": 5.1361,
|
|
"mean_token_accuracy": 0.20890870243310927,
|
|
"num_tokens": 9967213.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.844092750549317,
|
|
"epoch": 5.064644455494779,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00030853818767155225,
|
|
"loss": 5.1961,
|
|
"mean_token_accuracy": 0.1964746206998825,
|
|
"num_tokens": 9976886.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 5.801842117309571,
|
|
"epoch": 5.069617105917454,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003081519707270065,
|
|
"loss": 5.0867,
|
|
"mean_token_accuracy": 0.21339723318815232,
|
|
"num_tokens": 9986702.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 5.853747844696045,
|
|
"epoch": 5.07458975634013,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030776565390860017,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.2179371550679207,
|
|
"num_tokens": 9996597.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 5.8521381378173825,
|
|
"epoch": 5.079562406762805,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003073792383801543,
|
|
"loss": 5.1651,
|
|
"mean_token_accuracy": 0.20414186269044876,
|
|
"num_tokens": 10006440.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 5.84061107635498,
|
|
"epoch": 5.08453505718548,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00030699272530578697,
|
|
"loss": 5.1841,
|
|
"mean_token_accuracy": 0.20523302257061005,
|
|
"num_tokens": 10016198.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 5.913068056106567,
|
|
"epoch": 5.089507707608155,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00030660611584991013,
|
|
"loss": 5.1902,
|
|
"mean_token_accuracy": 0.20201814025640488,
|
|
"num_tokens": 10025795.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 5.8320948600769045,
|
|
"epoch": 5.094480358030831,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00030621941117722615,
|
|
"loss": 5.0726,
|
|
"mean_token_accuracy": 0.21997185945510864,
|
|
"num_tokens": 10035448.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 5.794463634490967,
|
|
"epoch": 5.099453008453506,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003058326124527243,
|
|
"loss": 5.1414,
|
|
"mean_token_accuracy": 0.2136662110686302,
|
|
"num_tokens": 10046335.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.9148650646209715,
|
|
"epoch": 5.104425658876181,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003054457208416771,
|
|
"loss": 5.1882,
|
|
"mean_token_accuracy": 0.2012014627456665,
|
|
"num_tokens": 10056346.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 5.717076110839844,
|
|
"epoch": 5.109398309298856,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00030505873750963705,
|
|
"loss": 5.0108,
|
|
"mean_token_accuracy": 0.2263475239276886,
|
|
"num_tokens": 10065948.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 5.749404668807983,
|
|
"epoch": 5.114370959721532,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.0003046716636224327,
|
|
"loss": 5.0834,
|
|
"mean_token_accuracy": 0.21022767722606658,
|
|
"num_tokens": 10076680.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 5.86261887550354,
|
|
"epoch": 5.119343610144207,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003042845003461657,
|
|
"loss": 5.1535,
|
|
"mean_token_accuracy": 0.20200339704751968,
|
|
"num_tokens": 10086373.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 5.836791515350342,
|
|
"epoch": 5.124316260566882,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030389724884720664,
|
|
"loss": 5.1323,
|
|
"mean_token_accuracy": 0.20291296690702437,
|
|
"num_tokens": 10095415.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.659631633758545,
|
|
"epoch": 5.129288910989557,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030350991029219236,
|
|
"loss": 4.9987,
|
|
"mean_token_accuracy": 0.22584066092967986,
|
|
"num_tokens": 10104709.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 5.878235864639282,
|
|
"epoch": 5.134261561412233,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00030312248584802166,
|
|
"loss": 5.1378,
|
|
"mean_token_accuracy": 0.20545922815799714,
|
|
"num_tokens": 10113539.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 5.729328203201294,
|
|
"epoch": 5.139234211834908,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003027349766818521,
|
|
"loss": 5.0442,
|
|
"mean_token_accuracy": 0.2209361657500267,
|
|
"num_tokens": 10123914.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 5.88569827079773,
|
|
"epoch": 5.144206862257583,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003023473839610965,
|
|
"loss": 5.1992,
|
|
"mean_token_accuracy": 0.1998208537697792,
|
|
"num_tokens": 10133572.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 5.891399192810058,
|
|
"epoch": 5.149179512680258,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003019597088534195,
|
|
"loss": 5.2592,
|
|
"mean_token_accuracy": 0.20073730647563934,
|
|
"num_tokens": 10142973.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 5.89968581199646,
|
|
"epoch": 5.154152163102934,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003015719525267338,
|
|
"loss": 5.2193,
|
|
"mean_token_accuracy": 0.1999717578291893,
|
|
"num_tokens": 10153322.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.815464210510254,
|
|
"epoch": 5.159124813525609,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003011841161491969,
|
|
"loss": 5.1288,
|
|
"mean_token_accuracy": 0.20926322638988495,
|
|
"num_tokens": 10163176.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 5.861962461471558,
|
|
"epoch": 5.164097463948284,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00030079620088920727,
|
|
"loss": 5.142,
|
|
"mean_token_accuracy": 0.20740548074245452,
|
|
"num_tokens": 10172616.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 5.893119192123413,
|
|
"epoch": 5.169070114370959,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003004082079154013,
|
|
"loss": 5.1802,
|
|
"mean_token_accuracy": 0.20273840874433519,
|
|
"num_tokens": 10182243.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.860502099990844,
|
|
"epoch": 5.174042764793635,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030002013839664925,
|
|
"loss": 5.1379,
|
|
"mean_token_accuracy": 0.20685796737670897,
|
|
"num_tokens": 10191394.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 5.632623815536499,
|
|
"epoch": 5.17901541521631,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002996319935020522,
|
|
"loss": 5.0052,
|
|
"mean_token_accuracy": 0.23010420352220534,
|
|
"num_tokens": 10202167.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 5.803150463104248,
|
|
"epoch": 5.183988065638985,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002992437744009381,
|
|
"loss": 5.0741,
|
|
"mean_token_accuracy": 0.21144769340753555,
|
|
"num_tokens": 10211250.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 5.929183578491211,
|
|
"epoch": 5.188960716061661,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029885548226285855,
|
|
"loss": 5.2343,
|
|
"mean_token_accuracy": 0.19720856547355653,
|
|
"num_tokens": 10220722.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 5.855019807815552,
|
|
"epoch": 5.193933366484337,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002984671182575851,
|
|
"loss": 5.2267,
|
|
"mean_token_accuracy": 0.20069740563631058,
|
|
"num_tokens": 10229880.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 5.829359436035157,
|
|
"epoch": 5.198906016907012,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002980786835551061,
|
|
"loss": 5.2274,
|
|
"mean_token_accuracy": 0.20405944287776948,
|
|
"num_tokens": 10240939.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 5.903607654571533,
|
|
"epoch": 5.203878667329687,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002976901793256225,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.20207842886447908,
|
|
"num_tokens": 10249383.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 5.958780431747437,
|
|
"epoch": 5.208851317752362,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000297301606739545,
|
|
"loss": 5.219,
|
|
"mean_token_accuracy": 0.20224751234054567,
|
|
"num_tokens": 10259166.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 5.920119714736939,
|
|
"epoch": 5.213823968175038,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029691296696748994,
|
|
"loss": 5.2784,
|
|
"mean_token_accuracy": 0.1929038941860199,
|
|
"num_tokens": 10269053.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.8165388107299805,
|
|
"epoch": 5.218796618597713,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00029652426118027635,
|
|
"loss": 5.1596,
|
|
"mean_token_accuracy": 0.2129316359758377,
|
|
"num_tokens": 10279148.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 5.759281635284424,
|
|
"epoch": 5.223769269020388,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029613549054892206,
|
|
"loss": 5.1266,
|
|
"mean_token_accuracy": 0.20290495306253434,
|
|
"num_tokens": 10288719.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 5.79491868019104,
|
|
"epoch": 5.228741919443063,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002957466562446401,
|
|
"loss": 5.1336,
|
|
"mean_token_accuracy": 0.20628678053617477,
|
|
"num_tokens": 10300089.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.849854040145874,
|
|
"epoch": 5.233714569865739,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002953577594388355,
|
|
"loss": 5.1537,
|
|
"mean_token_accuracy": 0.2066730350255966,
|
|
"num_tokens": 10309400.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 5.875857019424439,
|
|
"epoch": 5.238687220288414,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00029496880130310145,
|
|
"loss": 5.1989,
|
|
"mean_token_accuracy": 0.2089892402291298,
|
|
"num_tokens": 10319115.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 5.882859706878662,
|
|
"epoch": 5.243659870711089,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002945797830092161,
|
|
"loss": 5.1027,
|
|
"mean_token_accuracy": 0.2098475068807602,
|
|
"num_tokens": 10327174.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 5.8527733325958256,
|
|
"epoch": 5.248632521133764,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002941907057291386,
|
|
"loss": 5.1803,
|
|
"mean_token_accuracy": 0.20339882671833037,
|
|
"num_tokens": 10335608.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.744290590286255,
|
|
"epoch": 5.25360517155644,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029380157063500594,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.21159339547157288,
|
|
"num_tokens": 10345592.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 5.784367227554322,
|
|
"epoch": 5.258577821979115,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002934123788991291,
|
|
"loss": 5.106,
|
|
"mean_token_accuracy": 0.21375617533922195,
|
|
"num_tokens": 10355913.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 5.877681303024292,
|
|
"epoch": 5.26355047240179,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00029302313169399,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.2061230018734932,
|
|
"num_tokens": 10365419.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 5.8195634365081785,
|
|
"epoch": 5.268523122824465,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002926338301922376,
|
|
"loss": 5.1869,
|
|
"mean_token_accuracy": 0.20282365381717682,
|
|
"num_tokens": 10373939.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 5.880087518692017,
|
|
"epoch": 5.273495773247141,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00029224447556668417,
|
|
"loss": 5.2294,
|
|
"mean_token_accuracy": 0.19475735276937484,
|
|
"num_tokens": 10383932.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 5.893941497802734,
|
|
"epoch": 5.278468423669816,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002918550689903022,
|
|
"loss": 5.1631,
|
|
"mean_token_accuracy": 0.20375477522611618,
|
|
"num_tokens": 10393319.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.800909376144409,
|
|
"epoch": 5.283441074092491,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00029146561163622076,
|
|
"loss": 5.2984,
|
|
"mean_token_accuracy": 0.19060192704200746,
|
|
"num_tokens": 10404102.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 5.796519470214844,
|
|
"epoch": 5.288413724515166,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002910761046777218,
|
|
"loss": 5.1522,
|
|
"mean_token_accuracy": 0.21030451357364655,
|
|
"num_tokens": 10412718.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 5.942877388000488,
|
|
"epoch": 5.293386374937842,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00029068654928823666,
|
|
"loss": 5.2484,
|
|
"mean_token_accuracy": 0.19571566283702851,
|
|
"num_tokens": 10422546.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 5.885830068588257,
|
|
"epoch": 5.298359025360517,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00029029694664134284,
|
|
"loss": 5.2181,
|
|
"mean_token_accuracy": 0.19823373407125472,
|
|
"num_tokens": 10431498.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 5.871678352355957,
|
|
"epoch": 5.303331675783192,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002899072979107597,
|
|
"loss": 5.2182,
|
|
"mean_token_accuracy": 0.19903253018856049,
|
|
"num_tokens": 10441713.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.931542778015137,
|
|
"epoch": 5.308304326205867,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002895176042703461,
|
|
"loss": 5.241,
|
|
"mean_token_accuracy": 0.19700314849615097,
|
|
"num_tokens": 10451570.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.878738355636597,
|
|
"epoch": 5.3132769766285435,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00028912786689409554,
|
|
"loss": 5.1935,
|
|
"mean_token_accuracy": 0.20382671654224396,
|
|
"num_tokens": 10461442.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 5.922053098678589,
|
|
"epoch": 5.318249627051218,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002887380869561338,
|
|
"loss": 5.2793,
|
|
"mean_token_accuracy": 0.19343662708997728,
|
|
"num_tokens": 10471223.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 5.794125127792358,
|
|
"epoch": 5.323222277473894,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002883482656307145,
|
|
"loss": 5.1185,
|
|
"mean_token_accuracy": 0.2036026895046234,
|
|
"num_tokens": 10480469.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 5.820155048370362,
|
|
"epoch": 5.328194927896569,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002879584040922162,
|
|
"loss": 5.2063,
|
|
"mean_token_accuracy": 0.2026737317442894,
|
|
"num_tokens": 10490320.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.843921041488647,
|
|
"epoch": 5.3331675783192445,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002875685035151386,
|
|
"loss": 5.2208,
|
|
"mean_token_accuracy": 0.19339460432529448,
|
|
"num_tokens": 10500471.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 5.9464186668396,
|
|
"epoch": 5.33814022874192,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00028717856507409886,
|
|
"loss": 5.3,
|
|
"mean_token_accuracy": 0.1931556209921837,
|
|
"num_tokens": 10510956.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 5.9070676326751705,
|
|
"epoch": 5.343112879164595,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002867885899438281,
|
|
"loss": 5.2562,
|
|
"mean_token_accuracy": 0.20198429077863694,
|
|
"num_tokens": 10520838.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 5.910067510604859,
|
|
"epoch": 5.34808552958727,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002863985792991683,
|
|
"loss": 5.218,
|
|
"mean_token_accuracy": 0.20493451803922652,
|
|
"num_tokens": 10530879.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 5.722791385650635,
|
|
"epoch": 5.3530581800099455,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002860085343150683,
|
|
"loss": 5.0096,
|
|
"mean_token_accuracy": 0.22839111387729644,
|
|
"num_tokens": 10540161.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.858139705657959,
|
|
"epoch": 5.358030830432621,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028561845616658033,
|
|
"loss": 5.2234,
|
|
"mean_token_accuracy": 0.20006729066371917,
|
|
"num_tokens": 10551173.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 5.839142084121704,
|
|
"epoch": 5.363003480855296,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00028522834602885656,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.20221744179725648,
|
|
"num_tokens": 10560882.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.924833154678344,
|
|
"epoch": 5.367976131277971,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002848382050771455,
|
|
"loss": 5.3153,
|
|
"mean_token_accuracy": 0.19415909051895142,
|
|
"num_tokens": 10570837.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 5.885377407073975,
|
|
"epoch": 5.3729487817006465,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002844480344867886,
|
|
"loss": 5.1995,
|
|
"mean_token_accuracy": 0.20630669593811035,
|
|
"num_tokens": 10580335.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.858039808273316,
|
|
"epoch": 5.377921432123322,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002840578354332165,
|
|
"loss": 5.1902,
|
|
"mean_token_accuracy": 0.20606424063444137,
|
|
"num_tokens": 10589030.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.802721834182739,
|
|
"epoch": 5.382894082545997,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002836676090919456,
|
|
"loss": 5.201,
|
|
"mean_token_accuracy": 0.19860343337059022,
|
|
"num_tokens": 10599309.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 5.811599636077881,
|
|
"epoch": 5.387866732968672,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002832773566385746,
|
|
"loss": 5.1427,
|
|
"mean_token_accuracy": 0.20634075105190278,
|
|
"num_tokens": 10608781.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.851524829864502,
|
|
"epoch": 5.3928393833913475,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00028288707924878056,
|
|
"loss": 5.0935,
|
|
"mean_token_accuracy": 0.21324850767850875,
|
|
"num_tokens": 10618512.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 5.876164388656616,
|
|
"epoch": 5.397812033814023,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002824967780983162,
|
|
"loss": 5.2118,
|
|
"mean_token_accuracy": 0.2038364365696907,
|
|
"num_tokens": 10629420.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 5.854246664047241,
|
|
"epoch": 5.402784684236698,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028210645436300524,
|
|
"loss": 5.2113,
|
|
"mean_token_accuracy": 0.20821097046136855,
|
|
"num_tokens": 10639625.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 5.80420446395874,
|
|
"epoch": 5.407757334659373,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00028171610921873995,
|
|
"loss": 5.1066,
|
|
"mean_token_accuracy": 0.20922330617904664,
|
|
"num_tokens": 10648633.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.732096385955811,
|
|
"epoch": 5.4127299850820485,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002813257438414766,
|
|
"loss": 5.0301,
|
|
"mean_token_accuracy": 0.22040063440799712,
|
|
"num_tokens": 10657981.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 5.756271839141846,
|
|
"epoch": 5.417702635504724,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000280935359407233,
|
|
"loss": 5.0671,
|
|
"mean_token_accuracy": 0.2121916174888611,
|
|
"num_tokens": 10669734.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 5.827439022064209,
|
|
"epoch": 5.422675285927399,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002805449570920838,
|
|
"loss": 5.1664,
|
|
"mean_token_accuracy": 0.19690950065851212,
|
|
"num_tokens": 10679486.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 5.857757711410523,
|
|
"epoch": 5.427647936350074,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00028015453807215786,
|
|
"loss": 5.2141,
|
|
"mean_token_accuracy": 0.1967525899410248,
|
|
"num_tokens": 10689551.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 5.820701313018799,
|
|
"epoch": 5.4326205867727495,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027976410352363416,
|
|
"loss": 5.1478,
|
|
"mean_token_accuracy": 0.20198413878679275,
|
|
"num_tokens": 10699173.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 5.791245222091675,
|
|
"epoch": 5.437593237195425,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027937365462273864,
|
|
"loss": 5.1288,
|
|
"mean_token_accuracy": 0.20913468152284623,
|
|
"num_tokens": 10709050.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.84505033493042,
|
|
"epoch": 5.4425658876181,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027898319254574043,
|
|
"loss": 5.2237,
|
|
"mean_token_accuracy": 0.19869889318943024,
|
|
"num_tokens": 10718397.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 5.849493408203125,
|
|
"epoch": 5.447538538040776,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002785927184689483,
|
|
"loss": 5.2548,
|
|
"mean_token_accuracy": 0.19637724608182908,
|
|
"num_tokens": 10729305.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.888776063919067,
|
|
"epoch": 5.452511188463451,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002782022335687073,
|
|
"loss": 5.1663,
|
|
"mean_token_accuracy": 0.20197226107120514,
|
|
"num_tokens": 10738602.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 5.834828042984009,
|
|
"epoch": 5.457483838886127,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00027781173902139485,
|
|
"loss": 5.1796,
|
|
"mean_token_accuracy": 0.20987022966146468,
|
|
"num_tokens": 10748373.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 5.8571155071258545,
|
|
"epoch": 5.462456489308802,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002774212360034176,
|
|
"loss": 5.21,
|
|
"mean_token_accuracy": 0.20093724578619004,
|
|
"num_tokens": 10756879.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 5.780397891998291,
|
|
"epoch": 5.467429139731477,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002770307256912078,
|
|
"loss": 5.0508,
|
|
"mean_token_accuracy": 0.2135750398039818,
|
|
"num_tokens": 10766969.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 5.467429139731477,
|
|
"eval_entropy": 5.77161213989082,
|
|
"eval_loss": 6.5008320808410645,
|
|
"eval_mean_token_accuracy": 0.14712006491415786,
|
|
"eval_num_tokens": 10766969.0,
|
|
"eval_runtime": 11.7101,
|
|
"eval_samples_per_second": 295.899,
|
|
"eval_steps_per_second": 37.062,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 5.902654123306275,
|
|
"epoch": 5.472401790154152,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027664020926121944,
|
|
"loss": 5.2178,
|
|
"mean_token_accuracy": 0.19934790581464767,
|
|
"num_tokens": 10776780.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 5.836739015579224,
|
|
"epoch": 5.477374440576828,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002762496878899251,
|
|
"loss": 5.2038,
|
|
"mean_token_accuracy": 0.20035247653722763,
|
|
"num_tokens": 10786341.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 5.861342906951904,
|
|
"epoch": 5.482347090999503,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00027585916275381226,
|
|
"loss": 5.1258,
|
|
"mean_token_accuracy": 0.20893691033124923,
|
|
"num_tokens": 10795754.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 5.836158609390258,
|
|
"epoch": 5.487319741422178,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002754686350293797,
|
|
"loss": 5.2931,
|
|
"mean_token_accuracy": 0.1926471173763275,
|
|
"num_tokens": 10805796.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 5.7477398872375485,
|
|
"epoch": 5.492292391844853,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002750781058931339,
|
|
"loss": 5.0278,
|
|
"mean_token_accuracy": 0.22331394553184508,
|
|
"num_tokens": 10815427.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 5.843760585784912,
|
|
"epoch": 5.497265042267529,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002746875765215857,
|
|
"loss": 5.1658,
|
|
"mean_token_accuracy": 0.20294297337532044,
|
|
"num_tokens": 10825501.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 5.923003244400024,
|
|
"epoch": 5.502237692690204,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027429704809124665,
|
|
"loss": 5.2801,
|
|
"mean_token_accuracy": 0.19684297144412993,
|
|
"num_tokens": 10834961.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.850966835021973,
|
|
"epoch": 5.507210343112879,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002739065217786255,
|
|
"loss": 5.1782,
|
|
"mean_token_accuracy": 0.2017541855573654,
|
|
"num_tokens": 10845491.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 5.8192174434661865,
|
|
"epoch": 5.512182993535554,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002735159987602246,
|
|
"loss": 5.244,
|
|
"mean_token_accuracy": 0.2001827046275139,
|
|
"num_tokens": 10855045.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 5.907034158706665,
|
|
"epoch": 5.51715564395823,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002731254802125362,
|
|
"loss": 5.2939,
|
|
"mean_token_accuracy": 0.19034968167543412,
|
|
"num_tokens": 10864254.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.8851776123046875,
|
|
"epoch": 5.522128294380905,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027273496731203945,
|
|
"loss": 5.1836,
|
|
"mean_token_accuracy": 0.20778014063835143,
|
|
"num_tokens": 10874768.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 5.845489072799682,
|
|
"epoch": 5.52710094480358,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000272344461235196,
|
|
"loss": 5.138,
|
|
"mean_token_accuracy": 0.21296205520629882,
|
|
"num_tokens": 10884820.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 5.683800888061524,
|
|
"epoch": 5.532073595226255,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027195396315844733,
|
|
"loss": 5.0758,
|
|
"mean_token_accuracy": 0.22607096880674363,
|
|
"num_tokens": 10895393.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.810624504089356,
|
|
"epoch": 5.537046245648931,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002715634742582107,
|
|
"loss": 5.0842,
|
|
"mean_token_accuracy": 0.21691976487636566,
|
|
"num_tokens": 10903875.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 5.918364524841309,
|
|
"epoch": 5.542018896071606,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00027117299571087576,
|
|
"loss": 5.2589,
|
|
"mean_token_accuracy": 0.19621842801570893,
|
|
"num_tokens": 10913861.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 5.841112184524536,
|
|
"epoch": 5.546991546494281,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002707825286928008,
|
|
"loss": 5.2446,
|
|
"mean_token_accuracy": 0.19139108508825303,
|
|
"num_tokens": 10923597.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 5.814586162567139,
|
|
"epoch": 5.551964196916956,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002703920743803097,
|
|
"loss": 5.2486,
|
|
"mean_token_accuracy": 0.1902356520295143,
|
|
"num_tokens": 10934094.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.936142253875732,
|
|
"epoch": 5.556936847339632,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00027000163394968763,
|
|
"loss": 5.3128,
|
|
"mean_token_accuracy": 0.19154981672763824,
|
|
"num_tokens": 10944121.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.817663288116455,
|
|
"epoch": 5.561909497762308,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026961120857717843,
|
|
"loss": 5.1138,
|
|
"mean_token_accuracy": 0.2095634549856186,
|
|
"num_tokens": 10955117.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.764522457122803,
|
|
"epoch": 5.566882148184982,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026922079943898016,
|
|
"loss": 5.0858,
|
|
"mean_token_accuracy": 0.21865837275981903,
|
|
"num_tokens": 10965248.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 5.890922594070434,
|
|
"epoch": 5.571854798607658,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002688304077112421,
|
|
"loss": 5.2973,
|
|
"mean_token_accuracy": 0.19830915927886963,
|
|
"num_tokens": 10975610.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 5.7830939292907715,
|
|
"epoch": 5.576827449030333,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002684400345700613,
|
|
"loss": 5.152,
|
|
"mean_token_accuracy": 0.2124311000108719,
|
|
"num_tokens": 10984421.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 5.698187685012817,
|
|
"epoch": 5.581800099453009,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002680496811914786,
|
|
"loss": 5.0057,
|
|
"mean_token_accuracy": 0.2276494473218918,
|
|
"num_tokens": 10993740.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 5.817315292358399,
|
|
"epoch": 5.586772749875684,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026765934875147513,
|
|
"loss": 5.0933,
|
|
"mean_token_accuracy": 0.2124271109700203,
|
|
"num_tokens": 11003382.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 5.814067554473877,
|
|
"epoch": 5.591745400298359,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002672690384259693,
|
|
"loss": 5.1275,
|
|
"mean_token_accuracy": 0.21623673737049104,
|
|
"num_tokens": 11013336.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.821282005310058,
|
|
"epoch": 5.596718050721035,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026687875139081287,
|
|
"loss": 5.2159,
|
|
"mean_token_accuracy": 0.19692184180021285,
|
|
"num_tokens": 11023800.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.960972642898559,
|
|
"epoch": 5.60169070114371,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00026648848882178703,
|
|
"loss": 5.3747,
|
|
"mean_token_accuracy": 0.19402886778116227,
|
|
"num_tokens": 11033250.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 5.816966104507446,
|
|
"epoch": 5.606663351566385,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002660982518945998,
|
|
"loss": 5.1843,
|
|
"mean_token_accuracy": 0.20584758520126342,
|
|
"num_tokens": 11042045.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 5.783524227142334,
|
|
"epoch": 5.61163600198906,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026570804178488146,
|
|
"loss": 5.1135,
|
|
"mean_token_accuracy": 0.21406568884849547,
|
|
"num_tokens": 11051349.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 5.827799320220947,
|
|
"epoch": 5.616608652411736,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002653178596681819,
|
|
"loss": 5.1547,
|
|
"mean_token_accuracy": 0.2102708712220192,
|
|
"num_tokens": 11061550.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 5.818497180938721,
|
|
"epoch": 5.621581302834411,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002649277067199665,
|
|
"loss": 5.1489,
|
|
"mean_token_accuracy": 0.20555733293294906,
|
|
"num_tokens": 11071887.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 5.753476524353028,
|
|
"epoch": 5.626553953257086,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002645375841156127,
|
|
"loss": 5.1101,
|
|
"mean_token_accuracy": 0.2094201847910881,
|
|
"num_tokens": 11081276.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 5.82939338684082,
|
|
"epoch": 5.631526603679761,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002641474930304067,
|
|
"loss": 5.2757,
|
|
"mean_token_accuracy": 0.18970660716295243,
|
|
"num_tokens": 11091761.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 5.733616495132447,
|
|
"epoch": 5.636499254102437,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002637574346395396,
|
|
"loss": 4.9951,
|
|
"mean_token_accuracy": 0.22388276010751723,
|
|
"num_tokens": 11101986.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 5.897250699996948,
|
|
"epoch": 5.641471904525112,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000263367410118104,
|
|
"loss": 5.1823,
|
|
"mean_token_accuracy": 0.2069081798195839,
|
|
"num_tokens": 11111835.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 5.7403820037841795,
|
|
"epoch": 5.646444554947787,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002629774206410906,
|
|
"loss": 5.1396,
|
|
"mean_token_accuracy": 0.20626671314239503,
|
|
"num_tokens": 11122646.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 5.911836910247803,
|
|
"epoch": 5.651417205370462,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002625874673833844,
|
|
"loss": 5.2108,
|
|
"mean_token_accuracy": 0.20220037251710893,
|
|
"num_tokens": 11132693.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 5.915971660614014,
|
|
"epoch": 5.656389855793138,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002621975515197613,
|
|
"loss": 5.2412,
|
|
"mean_token_accuracy": 0.19521611034870148,
|
|
"num_tokens": 11142285.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 5.826210784912109,
|
|
"epoch": 5.661362506215813,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026180767422488466,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.20359499603509904,
|
|
"num_tokens": 11150657.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 5.832951164245605,
|
|
"epoch": 5.666335156638488,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026141783667330156,
|
|
"loss": 5.1876,
|
|
"mean_token_accuracy": 0.20809621065855027,
|
|
"num_tokens": 11160391.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 5.900640916824341,
|
|
"epoch": 5.671307807061163,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002610280400394393,
|
|
"loss": 5.3368,
|
|
"mean_token_accuracy": 0.19245774298906326,
|
|
"num_tokens": 11169919.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 5.894226789474487,
|
|
"epoch": 5.676280457483839,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026063828549760195,
|
|
"loss": 5.2183,
|
|
"mean_token_accuracy": 0.1994325652718544,
|
|
"num_tokens": 11179948.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 5.701592063903808,
|
|
"epoch": 5.681253107906514,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002602485742219667,
|
|
"loss": 4.984,
|
|
"mean_token_accuracy": 0.22620704472064973,
|
|
"num_tokens": 11190875.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 5.696918582916259,
|
|
"epoch": 5.686225758329189,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00025985890738658077,
|
|
"loss": 5.0173,
|
|
"mean_token_accuracy": 0.220773184299469,
|
|
"num_tokens": 11201134.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.810629081726074,
|
|
"epoch": 5.691198408751864,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000259469286165357,
|
|
"loss": 5.1286,
|
|
"mean_token_accuracy": 0.21587330847978592,
|
|
"num_tokens": 11209656.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.771128177642822,
|
|
"epoch": 5.6961710591745405,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025907971173207107,
|
|
"loss": 5.1621,
|
|
"mean_token_accuracy": 0.2055776983499527,
|
|
"num_tokens": 11218817.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 5.8924461841583256,
|
|
"epoch": 5.701143709597215,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002586901852603576,
|
|
"loss": 5.2241,
|
|
"mean_token_accuracy": 0.20427747964859008,
|
|
"num_tokens": 11229291.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.866375255584717,
|
|
"epoch": 5.706116360019891,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002583007079237069,
|
|
"loss": 5.2367,
|
|
"mean_token_accuracy": 0.20185985863208772,
|
|
"num_tokens": 11238877.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 5.832436752319336,
|
|
"epoch": 5.711089010442566,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025791128089546107,
|
|
"loss": 5.2349,
|
|
"mean_token_accuracy": 0.20021228343248368,
|
|
"num_tokens": 11248885.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 5.874657440185547,
|
|
"epoch": 5.7160616608652415,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00025752190534881065,
|
|
"loss": 5.2037,
|
|
"mean_token_accuracy": 0.2058560222387314,
|
|
"num_tokens": 11259423.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 5.873690032958985,
|
|
"epoch": 5.721034311287917,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025713258245679127,
|
|
"loss": 5.2545,
|
|
"mean_token_accuracy": 0.1992800697684288,
|
|
"num_tokens": 11268900.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 5.7710059642791744,
|
|
"epoch": 5.726006961710592,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00025674331339227966,
|
|
"loss": 5.0526,
|
|
"mean_token_accuracy": 0.2191447600722313,
|
|
"num_tokens": 11277997.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 5.9117179870605465,
|
|
"epoch": 5.730979612133267,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00025635409932799074,
|
|
"loss": 5.3483,
|
|
"mean_token_accuracy": 0.19733596742153167,
|
|
"num_tokens": 11288788.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 5.7406778812408445,
|
|
"epoch": 5.7359522625559425,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002559649414364735,
|
|
"loss": 5.1178,
|
|
"mean_token_accuracy": 0.2173623636364937,
|
|
"num_tokens": 11297785.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 5.717997789382935,
|
|
"epoch": 5.740924912978618,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002555758408901076,
|
|
"loss": 5.1189,
|
|
"mean_token_accuracy": 0.2166953757405281,
|
|
"num_tokens": 11307851.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 5.823109912872314,
|
|
"epoch": 5.745897563401293,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00025518679886110023,
|
|
"loss": 5.2159,
|
|
"mean_token_accuracy": 0.20569314062595367,
|
|
"num_tokens": 11318616.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 5.8675061702728275,
|
|
"epoch": 5.750870213823968,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002547978165214822,
|
|
"loss": 5.189,
|
|
"mean_token_accuracy": 0.20990708321332932,
|
|
"num_tokens": 11328751.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 5.639439821243286,
|
|
"epoch": 5.7558428642466435,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025440889504310416,
|
|
"loss": 4.9346,
|
|
"mean_token_accuracy": 0.23592369854450226,
|
|
"num_tokens": 11338054.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 5.847891998291016,
|
|
"epoch": 5.760815514669319,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025402003559763393,
|
|
"loss": 5.2288,
|
|
"mean_token_accuracy": 0.19818786680698394,
|
|
"num_tokens": 11348341.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 5.858513832092285,
|
|
"epoch": 5.765788165091994,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00025363123935655214,
|
|
"loss": 5.1906,
|
|
"mean_token_accuracy": 0.20603644847869873,
|
|
"num_tokens": 11357496.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 5.813236045837402,
|
|
"epoch": 5.770760815514669,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002532425074911491,
|
|
"loss": 5.1617,
|
|
"mean_token_accuracy": 0.20833520740270614,
|
|
"num_tokens": 11366915.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 5.839464235305786,
|
|
"epoch": 5.7757334659373445,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002528538411725212,
|
|
"loss": 5.2784,
|
|
"mean_token_accuracy": 0.19643164575099945,
|
|
"num_tokens": 11376920.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 5.688136720657349,
|
|
"epoch": 5.78070611636002,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002524652415715672,
|
|
"loss": 5.0041,
|
|
"mean_token_accuracy": 0.2286743313074112,
|
|
"num_tokens": 11387405.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 5.9042212009429935,
|
|
"epoch": 5.785678766782695,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002520767098589852,
|
|
"loss": 5.338,
|
|
"mean_token_accuracy": 0.19609687924385072,
|
|
"num_tokens": 11396553.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 5.907504510879517,
|
|
"epoch": 5.79065141720537,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002516882472052683,
|
|
"loss": 5.2958,
|
|
"mean_token_accuracy": 0.19307634830474854,
|
|
"num_tokens": 11406964.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 5.887142372131348,
|
|
"epoch": 5.7956240676280455,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002512998547807022,
|
|
"loss": 5.2563,
|
|
"mean_token_accuracy": 0.2034821629524231,
|
|
"num_tokens": 11416550.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 5.819599628448486,
|
|
"epoch": 5.800596718050721,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00025091153375536026,
|
|
"loss": 5.0842,
|
|
"mean_token_accuracy": 0.21409160345792771,
|
|
"num_tokens": 11425477.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.849649286270141,
|
|
"epoch": 5.805569368473396,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00025052328529910135,
|
|
"loss": 5.1585,
|
|
"mean_token_accuracy": 0.2032954588532448,
|
|
"num_tokens": 11435153.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 5.7422192096710205,
|
|
"epoch": 5.810542018896071,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002501351105815655,
|
|
"loss": 5.1249,
|
|
"mean_token_accuracy": 0.21837035566568375,
|
|
"num_tokens": 11445635.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 5.879411506652832,
|
|
"epoch": 5.8155146693187465,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00024974701077217057,
|
|
"loss": 5.3078,
|
|
"mean_token_accuracy": 0.19504621028900146,
|
|
"num_tokens": 11455526.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 5.814843940734863,
|
|
"epoch": 5.820487319741423,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024935898704010875,
|
|
"loss": 5.1703,
|
|
"mean_token_accuracy": 0.2027420461177826,
|
|
"num_tokens": 11464115.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 5.687883710861206,
|
|
"epoch": 5.825459970164097,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024897104055434304,
|
|
"loss": 5.0852,
|
|
"mean_token_accuracy": 0.2194989338517189,
|
|
"num_tokens": 11475456.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 5.667568922042847,
|
|
"epoch": 5.830432620586773,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024858317248360387,
|
|
"loss": 4.9882,
|
|
"mean_token_accuracy": 0.23221344202756883,
|
|
"num_tokens": 11485796.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 5.890707397460938,
|
|
"epoch": 5.835405271009448,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002481953839963851,
|
|
"loss": 5.2468,
|
|
"mean_token_accuracy": 0.1959501177072525,
|
|
"num_tokens": 11494904.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 5.679607295989991,
|
|
"epoch": 5.840377921432124,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002478076762609413,
|
|
"loss": 5.0045,
|
|
"mean_token_accuracy": 0.22588608264923096,
|
|
"num_tokens": 11504429.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 5.8378581523895265,
|
|
"epoch": 5.845350571854799,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002474200504452834,
|
|
"loss": 5.2415,
|
|
"mean_token_accuracy": 0.19515056908130646,
|
|
"num_tokens": 11515557.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 5.876676940917969,
|
|
"epoch": 5.850323222277474,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024703250771717576,
|
|
"loss": 5.2718,
|
|
"mean_token_accuracy": 0.1987658753991127,
|
|
"num_tokens": 11525613.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 5.848644638061524,
|
|
"epoch": 5.855295872700149,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024664504924413224,
|
|
"loss": 5.1914,
|
|
"mean_token_accuracy": 0.2007001295685768,
|
|
"num_tokens": 11535463.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 5.838754606246948,
|
|
"epoch": 5.860268523122825,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002462576761934129,
|
|
"loss": 5.1856,
|
|
"mean_token_accuracy": 0.20157656371593474,
|
|
"num_tokens": 11546277.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 5.82364444732666,
|
|
"epoch": 5.8652411735455,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002458703897320208,
|
|
"loss": 5.1769,
|
|
"mean_token_accuracy": 0.20194328725337982,
|
|
"num_tokens": 11555361.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 5.807602882385254,
|
|
"epoch": 5.870213823968175,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00024548319102669745,
|
|
"loss": 5.092,
|
|
"mean_token_accuracy": 0.20934535413980485,
|
|
"num_tokens": 11564976.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.735958814620972,
|
|
"epoch": 5.87518647439085,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00024509608124392065,
|
|
"loss": 5.0581,
|
|
"mean_token_accuracy": 0.21383206099271773,
|
|
"num_tokens": 11574574.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 5.858099126815796,
|
|
"epoch": 5.880159124813526,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002447090615499,
|
|
"loss": 5.2045,
|
|
"mean_token_accuracy": 0.19665164053440093,
|
|
"num_tokens": 11584571.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 5.917815256118774,
|
|
"epoch": 5.885131775236201,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00024432213311057383,
|
|
"loss": 5.2985,
|
|
"mean_token_accuracy": 0.1945430874824524,
|
|
"num_tokens": 11594829.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 5.825990867614746,
|
|
"epoch": 5.890104425658876,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00024393529709160544,
|
|
"loss": 5.2552,
|
|
"mean_token_accuracy": 0.20461108535528183,
|
|
"num_tokens": 11604520.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.872145414352417,
|
|
"epoch": 5.895077076081551,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00024354855465837967,
|
|
"loss": 5.1615,
|
|
"mean_token_accuracy": 0.20431189835071564,
|
|
"num_tokens": 11612909.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.89687671661377,
|
|
"epoch": 5.900049726504227,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024316190697599955,
|
|
"loss": 5.3112,
|
|
"mean_token_accuracy": 0.19648130983114243,
|
|
"num_tokens": 11622928.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 5.892285776138306,
|
|
"epoch": 5.905022376926902,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00024277535520928257,
|
|
"loss": 5.3286,
|
|
"mean_token_accuracy": 0.1936645656824112,
|
|
"num_tokens": 11632192.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 5.834741544723511,
|
|
"epoch": 5.909995027349577,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00024238890052275742,
|
|
"loss": 5.1758,
|
|
"mean_token_accuracy": 0.21049728691577912,
|
|
"num_tokens": 11642346.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.825209903717041,
|
|
"epoch": 5.914967677772252,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024200254408066014,
|
|
"loss": 5.1262,
|
|
"mean_token_accuracy": 0.20988789647817613,
|
|
"num_tokens": 11651527.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 5.886900091171265,
|
|
"epoch": 5.919940328194928,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002416162870469308,
|
|
"loss": 5.2309,
|
|
"mean_token_accuracy": 0.19503704458475113,
|
|
"num_tokens": 11660920.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 5.801678657531738,
|
|
"epoch": 5.924912978617603,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002412301305852102,
|
|
"loss": 5.1938,
|
|
"mean_token_accuracy": 0.20512580126523972,
|
|
"num_tokens": 11670220.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 5.686042785644531,
|
|
"epoch": 5.929885629040278,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024084407585883588,
|
|
"loss": 5.0015,
|
|
"mean_token_accuracy": 0.23620261400938034,
|
|
"num_tokens": 11680922.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.820929288864136,
|
|
"epoch": 5.934858279462953,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002404581240308391,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.2024478405714035,
|
|
"num_tokens": 11691138.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 5.848070764541626,
|
|
"epoch": 5.939830929885629,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000240072276263941,
|
|
"loss": 5.2106,
|
|
"mean_token_accuracy": 0.199624465405941,
|
|
"num_tokens": 11700379.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 5.924061250686646,
|
|
"epoch": 5.944803580308305,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00023968653372054933,
|
|
"loss": 5.223,
|
|
"mean_token_accuracy": 0.19713016748428344,
|
|
"num_tokens": 11709678.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 5.767105960845948,
|
|
"epoch": 5.949776230730979,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023930089756275474,
|
|
"loss": 5.0211,
|
|
"mean_token_accuracy": 0.2173046052455902,
|
|
"num_tokens": 11719700.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 5.810717487335205,
|
|
"epoch": 5.954748881153655,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002389153689523275,
|
|
"loss": 5.1955,
|
|
"mean_token_accuracy": 0.20299695581197738,
|
|
"num_tokens": 11730087.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 5.824495267868042,
|
|
"epoch": 5.9597215315763306,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023852994905071385,
|
|
"loss": 5.2169,
|
|
"mean_token_accuracy": 0.2046458199620247,
|
|
"num_tokens": 11740063.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 5.969442892074585,
|
|
"epoch": 5.964694181999006,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023814463901903238,
|
|
"loss": 5.3063,
|
|
"mean_token_accuracy": 0.19607568383216858,
|
|
"num_tokens": 11750213.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.964694181999006,
|
|
"eval_entropy": 5.835118677209599,
|
|
"eval_loss": 6.473224639892578,
|
|
"eval_mean_token_accuracy": 0.14957943952681008,
|
|
"eval_num_tokens": 11750213.0,
|
|
"eval_runtime": 11.9891,
|
|
"eval_samples_per_second": 289.013,
|
|
"eval_steps_per_second": 36.2,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 5.8844921588897705,
|
|
"epoch": 5.969666832421681,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023775944001807077,
|
|
"loss": 5.2182,
|
|
"mean_token_accuracy": 0.20003747045993805,
|
|
"num_tokens": 11759686.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 5.834766530990601,
|
|
"epoch": 5.974639482844356,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023737435320828244,
|
|
"loss": 5.1857,
|
|
"mean_token_accuracy": 0.21080127656459807,
|
|
"num_tokens": 11770199.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 5.534712314605713,
|
|
"epoch": 5.979612133267032,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023698937974978257,
|
|
"loss": 4.8863,
|
|
"mean_token_accuracy": 0.24927745759487152,
|
|
"num_tokens": 11780585.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 5.8577704429626465,
|
|
"epoch": 5.984584783689707,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023660452080234473,
|
|
"loss": 5.2899,
|
|
"mean_token_accuracy": 0.19142373651266098,
|
|
"num_tokens": 11789938.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 5.826484489440918,
|
|
"epoch": 5.989557434112382,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00023621977752539785,
|
|
"loss": 5.1463,
|
|
"mean_token_accuracy": 0.21256843358278274,
|
|
"num_tokens": 11799783.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 5.977162265777588,
|
|
"epoch": 5.994530084535057,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002358351510780223,
|
|
"loss": 5.3332,
|
|
"mean_token_accuracy": 0.18900174051523208,
|
|
"num_tokens": 11809334.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 5.8778234958648685,
|
|
"epoch": 5.999502734957733,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00023545064261894623,
|
|
"loss": 5.2114,
|
|
"mean_token_accuracy": 0.19600799381732942,
|
|
"num_tokens": 11820136.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 5.81966945860121,
|
|
"epoch": 6.00397812033814,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023506625330654263,
|
|
"loss": 5.0313,
|
|
"mean_token_accuracy": 0.2187221199274063,
|
|
"num_tokens": 11828628.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 5.7761188507080075,
|
|
"epoch": 6.008950770760816,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023468198429882548,
|
|
"loss": 4.9825,
|
|
"mean_token_accuracy": 0.21956740766763688,
|
|
"num_tokens": 11837964.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 5.839637231826782,
|
|
"epoch": 6.013923421183491,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002342978367534462,
|
|
"loss": 4.9961,
|
|
"mean_token_accuracy": 0.21759188622236253,
|
|
"num_tokens": 11847801.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 5.830140447616577,
|
|
"epoch": 6.018896071606166,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002339138118276905,
|
|
"loss": 4.9439,
|
|
"mean_token_accuracy": 0.22285042256116866,
|
|
"num_tokens": 11857836.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 5.880048608779907,
|
|
"epoch": 6.023868722028841,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023352991067847456,
|
|
"loss": 5.0453,
|
|
"mean_token_accuracy": 0.2141704097390175,
|
|
"num_tokens": 11867809.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 5.825541639328003,
|
|
"epoch": 6.028841372451517,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023314613446234162,
|
|
"loss": 4.9889,
|
|
"mean_token_accuracy": 0.21569908261299134,
|
|
"num_tokens": 11876869.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 5.8893026351928714,
|
|
"epoch": 6.033814022874192,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023276248433545867,
|
|
"loss": 4.9514,
|
|
"mean_token_accuracy": 0.22375447899103165,
|
|
"num_tokens": 11886816.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 5.885336589813233,
|
|
"epoch": 6.038786673296867,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023237896145361276,
|
|
"loss": 5.1126,
|
|
"mean_token_accuracy": 0.20843871533870698,
|
|
"num_tokens": 11896209.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 5.796182155609131,
|
|
"epoch": 6.043759323719542,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023199556697220758,
|
|
"loss": 4.8916,
|
|
"mean_token_accuracy": 0.23365502953529357,
|
|
"num_tokens": 11906522.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 5.840322828292846,
|
|
"epoch": 6.048731974142218,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023161230204626015,
|
|
"loss": 4.9777,
|
|
"mean_token_accuracy": 0.22239890843629836,
|
|
"num_tokens": 11916040.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 5.583293437957764,
|
|
"epoch": 6.053704624564893,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00023122916783039694,
|
|
"loss": 4.7169,
|
|
"mean_token_accuracy": 0.2422255903482437,
|
|
"num_tokens": 11927138.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 5.809394407272339,
|
|
"epoch": 6.058677274987568,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002308461654788509,
|
|
"loss": 4.9313,
|
|
"mean_token_accuracy": 0.22453774511814117,
|
|
"num_tokens": 11937385.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 5.81418137550354,
|
|
"epoch": 6.063649925410243,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023046329614545758,
|
|
"loss": 4.926,
|
|
"mean_token_accuracy": 0.22273582518100737,
|
|
"num_tokens": 11946621.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 5.816492033004761,
|
|
"epoch": 6.068622575832919,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00023008056098365183,
|
|
"loss": 4.9909,
|
|
"mean_token_accuracy": 0.21247692704200744,
|
|
"num_tokens": 11956744.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 5.917932224273682,
|
|
"epoch": 6.073595226255594,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022969796114646424,
|
|
"loss": 5.0794,
|
|
"mean_token_accuracy": 0.21473651081323625,
|
|
"num_tokens": 11967702.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 5.83793683052063,
|
|
"epoch": 6.078567876678269,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022931549778651783,
|
|
"loss": 4.9302,
|
|
"mean_token_accuracy": 0.218222413957119,
|
|
"num_tokens": 11976895.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 5.8298581600189205,
|
|
"epoch": 6.083540527100944,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022893317205602444,
|
|
"loss": 5.0032,
|
|
"mean_token_accuracy": 0.2199178695678711,
|
|
"num_tokens": 11985893.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 5.789108848571777,
|
|
"epoch": 6.08851317752362,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022855098510678125,
|
|
"loss": 4.988,
|
|
"mean_token_accuracy": 0.21786531656980515,
|
|
"num_tokens": 11995254.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 5.833800745010376,
|
|
"epoch": 6.093485827946296,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022816893809016737,
|
|
"loss": 4.983,
|
|
"mean_token_accuracy": 0.20698680281639098,
|
|
"num_tokens": 12006983.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 5.773342800140381,
|
|
"epoch": 6.098458478368971,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022778703215714014,
|
|
"loss": 4.9033,
|
|
"mean_token_accuracy": 0.22475162744522095,
|
|
"num_tokens": 12019428.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 5.804765701293945,
|
|
"epoch": 6.103431128791646,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022740526845823223,
|
|
"loss": 4.9647,
|
|
"mean_token_accuracy": 0.21662858724594117,
|
|
"num_tokens": 12029613.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 5.897802305221558,
|
|
"epoch": 6.1084037792143215,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002270236481435477,
|
|
"loss": 5.0798,
|
|
"mean_token_accuracy": 0.20984634310007094,
|
|
"num_tokens": 12039328.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 5.818811655044556,
|
|
"epoch": 6.113376429636997,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00022664217236275847,
|
|
"loss": 4.8813,
|
|
"mean_token_accuracy": 0.22902658432722092,
|
|
"num_tokens": 12048894.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 5.611724758148194,
|
|
"epoch": 6.118349080059672,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022626084226510124,
|
|
"loss": 4.8418,
|
|
"mean_token_accuracy": 0.23866482675075532,
|
|
"num_tokens": 12059316.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 5.812175846099853,
|
|
"epoch": 6.123321730482347,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022587965899937364,
|
|
"loss": 4.9935,
|
|
"mean_token_accuracy": 0.22490330934524536,
|
|
"num_tokens": 12069142.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 5.766677331924439,
|
|
"epoch": 6.1282943809050225,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002254986237139312,
|
|
"loss": 4.9122,
|
|
"mean_token_accuracy": 0.23111393302679062,
|
|
"num_tokens": 12077893.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 5.839347887039184,
|
|
"epoch": 6.133267031327698,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00022511773755668342,
|
|
"loss": 5.1256,
|
|
"mean_token_accuracy": 0.2138102188706398,
|
|
"num_tokens": 12087811.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 5.844075727462768,
|
|
"epoch": 6.138239681750373,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002247370016750907,
|
|
"loss": 4.9971,
|
|
"mean_token_accuracy": 0.21946714222431182,
|
|
"num_tokens": 12097172.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 5.868205499649048,
|
|
"epoch": 6.143212332173048,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002243564172161605,
|
|
"loss": 5.0409,
|
|
"mean_token_accuracy": 0.21779171228408814,
|
|
"num_tokens": 12106787.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 5.811212253570557,
|
|
"epoch": 6.1481849825957235,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00022397598532644441,
|
|
"loss": 5.0082,
|
|
"mean_token_accuracy": 0.2161289259791374,
|
|
"num_tokens": 12116815.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 5.796535682678223,
|
|
"epoch": 6.153157633018399,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022359570715203408,
|
|
"loss": 5.0287,
|
|
"mean_token_accuracy": 0.2108641803264618,
|
|
"num_tokens": 12127439.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 5.827324104309082,
|
|
"epoch": 6.158130283441074,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022321558383855828,
|
|
"loss": 4.9719,
|
|
"mean_token_accuracy": 0.22453310042619706,
|
|
"num_tokens": 12136828.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.672471380233764,
|
|
"epoch": 6.163102933863749,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002228356165311791,
|
|
"loss": 4.8745,
|
|
"mean_token_accuracy": 0.2392685055732727,
|
|
"num_tokens": 12146593.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 5.792682456970215,
|
|
"epoch": 6.1680755842864246,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022245580637458883,
|
|
"loss": 5.0064,
|
|
"mean_token_accuracy": 0.21721436828374863,
|
|
"num_tokens": 12157623.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 5.802333116531372,
|
|
"epoch": 6.1730482347091,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022207615451300616,
|
|
"loss": 4.9546,
|
|
"mean_token_accuracy": 0.22223811149597167,
|
|
"num_tokens": 12167353.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 5.818595218658447,
|
|
"epoch": 6.178020885131775,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022169666209017304,
|
|
"loss": 4.9918,
|
|
"mean_token_accuracy": 0.21979388743638992,
|
|
"num_tokens": 12176669.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 5.761445236206055,
|
|
"epoch": 6.18299353555445,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0002213173302493509,
|
|
"loss": 4.9087,
|
|
"mean_token_accuracy": 0.23236635625362395,
|
|
"num_tokens": 12185789.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 5.807548332214355,
|
|
"epoch": 6.1879661859771256,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022093816013331763,
|
|
"loss": 5.0266,
|
|
"mean_token_accuracy": 0.21907133013010024,
|
|
"num_tokens": 12195289.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 5.802649974822998,
|
|
"epoch": 6.192938836399801,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00022055915288436362,
|
|
"loss": 4.9888,
|
|
"mean_token_accuracy": 0.21946765184402467,
|
|
"num_tokens": 12204499.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 5.629654502868652,
|
|
"epoch": 6.197911486822476,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022018030964428897,
|
|
"loss": 4.8696,
|
|
"mean_token_accuracy": 0.24005589485168458,
|
|
"num_tokens": 12214739.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 5.827194261550903,
|
|
"epoch": 6.202884137245151,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021980163155439943,
|
|
"loss": 4.9285,
|
|
"mean_token_accuracy": 0.22152439802885054,
|
|
"num_tokens": 12224027.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 5.823512744903565,
|
|
"epoch": 6.207856787667827,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021942311975550323,
|
|
"loss": 5.0302,
|
|
"mean_token_accuracy": 0.2157731682062149,
|
|
"num_tokens": 12234317.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 5.799914884567261,
|
|
"epoch": 6.212829438090502,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021904477538790774,
|
|
"loss": 4.8622,
|
|
"mean_token_accuracy": 0.22958147078752517,
|
|
"num_tokens": 12244561.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 5.555161619186402,
|
|
"epoch": 6.217802088513178,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021866659959141578,
|
|
"loss": 4.7133,
|
|
"mean_token_accuracy": 0.25613122433423996,
|
|
"num_tokens": 12254480.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 5.852056503295898,
|
|
"epoch": 6.222774738935853,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002182885935053225,
|
|
"loss": 5.0552,
|
|
"mean_token_accuracy": 0.21163981556892394,
|
|
"num_tokens": 12263531.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 5.867765712738037,
|
|
"epoch": 6.2277473893585285,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021791075826841166,
|
|
"loss": 5.0735,
|
|
"mean_token_accuracy": 0.21568623632192613,
|
|
"num_tokens": 12273189.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 5.771522903442383,
|
|
"epoch": 6.232720039781204,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002175330950189523,
|
|
"loss": 4.998,
|
|
"mean_token_accuracy": 0.2195636585354805,
|
|
"num_tokens": 12282241.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 5.822233963012695,
|
|
"epoch": 6.237692690203879,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002171556048946955,
|
|
"loss": 4.9438,
|
|
"mean_token_accuracy": 0.22124121338129044,
|
|
"num_tokens": 12291803.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 5.832151317596436,
|
|
"epoch": 6.242665340626554,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021677828903287063,
|
|
"loss": 5.0315,
|
|
"mean_token_accuracy": 0.2133589819073677,
|
|
"num_tokens": 12302086.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 5.845774602890015,
|
|
"epoch": 6.2476379910492295,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021640114857018205,
|
|
"loss": 4.9964,
|
|
"mean_token_accuracy": 0.21931133717298507,
|
|
"num_tokens": 12311651.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 5.779177808761597,
|
|
"epoch": 6.252610641471905,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021602418464280586,
|
|
"loss": 4.9962,
|
|
"mean_token_accuracy": 0.22462982535362244,
|
|
"num_tokens": 12321583.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 5.80717282295227,
|
|
"epoch": 6.25758329189458,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021564739838638615,
|
|
"loss": 5.0161,
|
|
"mean_token_accuracy": 0.22011105716228485,
|
|
"num_tokens": 12331139.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 5.721974945068359,
|
|
"epoch": 6.262555942317255,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00021527079093603202,
|
|
"loss": 4.8623,
|
|
"mean_token_accuracy": 0.22778724879026413,
|
|
"num_tokens": 12341509.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 5.8388206481933596,
|
|
"epoch": 6.2675285927399305,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021489436342631352,
|
|
"loss": 5.0608,
|
|
"mean_token_accuracy": 0.20397014766931534,
|
|
"num_tokens": 12350450.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 5.747814321517945,
|
|
"epoch": 6.272501243162606,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021451811699125883,
|
|
"loss": 4.9071,
|
|
"mean_token_accuracy": 0.2334297090768814,
|
|
"num_tokens": 12359168.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 5.79835696220398,
|
|
"epoch": 6.277473893585281,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00021414205276435073,
|
|
"loss": 4.9937,
|
|
"mean_token_accuracy": 0.2089177891612053,
|
|
"num_tokens": 12367831.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 5.823434734344483,
|
|
"epoch": 6.282446544007956,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002137661718785229,
|
|
"loss": 5.0694,
|
|
"mean_token_accuracy": 0.21166798025369643,
|
|
"num_tokens": 12377804.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 5.855309581756591,
|
|
"epoch": 6.2874191944306315,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021339047546615675,
|
|
"loss": 5.0266,
|
|
"mean_token_accuracy": 0.21380686312913894,
|
|
"num_tokens": 12387022.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 5.738504409790039,
|
|
"epoch": 6.292391844853307,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021301496465907794,
|
|
"loss": 4.9194,
|
|
"mean_token_accuracy": 0.2189393773674965,
|
|
"num_tokens": 12395935.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 5.861141014099121,
|
|
"epoch": 6.297364495275982,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021263964058855294,
|
|
"loss": 5.0469,
|
|
"mean_token_accuracy": 0.21514024585485458,
|
|
"num_tokens": 12405760.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 5.6332872867584225,
|
|
"epoch": 6.302337145698657,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021226450438528572,
|
|
"loss": 4.7753,
|
|
"mean_token_accuracy": 0.23923778533935547,
|
|
"num_tokens": 12415573.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 5.816941976547241,
|
|
"epoch": 6.3073097961213325,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021188955717941428,
|
|
"loss": 5.0158,
|
|
"mean_token_accuracy": 0.22000337839126588,
|
|
"num_tokens": 12426532.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 5.789715147018432,
|
|
"epoch": 6.312282446544008,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002115148001005071,
|
|
"loss": 5.0385,
|
|
"mean_token_accuracy": 0.2157644659280777,
|
|
"num_tokens": 12438319.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 5.763112020492554,
|
|
"epoch": 6.317255096966683,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021114023427756012,
|
|
"loss": 4.93,
|
|
"mean_token_accuracy": 0.23545843213796616,
|
|
"num_tokens": 12448470.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 5.829544353485107,
|
|
"epoch": 6.322227747389358,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021076586083899296,
|
|
"loss": 5.025,
|
|
"mean_token_accuracy": 0.21232481449842452,
|
|
"num_tokens": 12457647.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 5.879931259155273,
|
|
"epoch": 6.3272003978120335,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0002103916809126456,
|
|
"loss": 5.0736,
|
|
"mean_token_accuracy": 0.21089781820774078,
|
|
"num_tokens": 12467841.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 5.8575845718383786,
|
|
"epoch": 6.332173048234709,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00021001769562577517,
|
|
"loss": 5.0739,
|
|
"mean_token_accuracy": 0.2036380872130394,
|
|
"num_tokens": 12478404.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.825278663635254,
|
|
"epoch": 6.337145698657384,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002096439061050523,
|
|
"loss": 4.9822,
|
|
"mean_token_accuracy": 0.2150880992412567,
|
|
"num_tokens": 12486770.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 5.792035484313965,
|
|
"epoch": 6.34211834908006,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000209270313476558,
|
|
"loss": 5.0096,
|
|
"mean_token_accuracy": 0.2180551528930664,
|
|
"num_tokens": 12496769.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 5.782818460464478,
|
|
"epoch": 6.347090999502735,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020889691886578004,
|
|
"loss": 4.9776,
|
|
"mean_token_accuracy": 0.2149469941854477,
|
|
"num_tokens": 12505552.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 5.604223203659058,
|
|
"epoch": 6.352063649925411,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020852372339760973,
|
|
"loss": 4.7253,
|
|
"mean_token_accuracy": 0.2522709876298904,
|
|
"num_tokens": 12515391.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 5.872190475463867,
|
|
"epoch": 6.357036300348086,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0002081507281963383,
|
|
"loss": 5.1023,
|
|
"mean_token_accuracy": 0.2058992013335228,
|
|
"num_tokens": 12525001.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 5.789206790924072,
|
|
"epoch": 6.362008950770761,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020777793438565368,
|
|
"loss": 4.952,
|
|
"mean_token_accuracy": 0.2231757387518883,
|
|
"num_tokens": 12534393.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 5.715649557113648,
|
|
"epoch": 6.366981601193436,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00020740534308863717,
|
|
"loss": 4.8399,
|
|
"mean_token_accuracy": 0.23858658075332642,
|
|
"num_tokens": 12544358.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 5.841217184066773,
|
|
"epoch": 6.371954251616112,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020703295542775992,
|
|
"loss": 5.0594,
|
|
"mean_token_accuracy": 0.21320519000291824,
|
|
"num_tokens": 12555113.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 5.818130397796631,
|
|
"epoch": 6.376926902038787,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020666077252487968,
|
|
"loss": 5.013,
|
|
"mean_token_accuracy": 0.21977995336055756,
|
|
"num_tokens": 12565204.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.813720989227295,
|
|
"epoch": 6.381899552461462,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002062887955012372,
|
|
"loss": 4.923,
|
|
"mean_token_accuracy": 0.22735073715448378,
|
|
"num_tokens": 12574789.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 5.760911750793457,
|
|
"epoch": 6.386872202884137,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020591702547745306,
|
|
"loss": 4.9233,
|
|
"mean_token_accuracy": 0.22629654556512832,
|
|
"num_tokens": 12583865.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 5.758836269378662,
|
|
"epoch": 6.391844853306813,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020554546357352433,
|
|
"loss": 5.0219,
|
|
"mean_token_accuracy": 0.2102368652820587,
|
|
"num_tokens": 12593778.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 5.815315198898316,
|
|
"epoch": 6.396817503729488,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000205174110908821,
|
|
"loss": 5.005,
|
|
"mean_token_accuracy": 0.22070749700069428,
|
|
"num_tokens": 12603571.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 5.889243698120117,
|
|
"epoch": 6.401790154152163,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.0002048029686020827,
|
|
"loss": 5.0961,
|
|
"mean_token_accuracy": 0.2070106953382492,
|
|
"num_tokens": 12613850.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 5.835537719726562,
|
|
"epoch": 6.406762804574838,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002044320377714154,
|
|
"loss": 5.0386,
|
|
"mean_token_accuracy": 0.21194053143262864,
|
|
"num_tokens": 12623587.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 5.725809764862061,
|
|
"epoch": 6.411735454997514,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00020406131953428785,
|
|
"loss": 4.9078,
|
|
"mean_token_accuracy": 0.22267274260520936,
|
|
"num_tokens": 12634702.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 5.774286270141602,
|
|
"epoch": 6.416708105420189,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002036908150075285,
|
|
"loss": 4.9311,
|
|
"mean_token_accuracy": 0.22669970542192458,
|
|
"num_tokens": 12643905.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 5.876151609420776,
|
|
"epoch": 6.421680755842864,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002033205253073218,
|
|
"loss": 5.0874,
|
|
"mean_token_accuracy": 0.20803386121988296,
|
|
"num_tokens": 12653794.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 5.85568585395813,
|
|
"epoch": 6.426653406265539,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020295045154920528,
|
|
"loss": 5.032,
|
|
"mean_token_accuracy": 0.21960904002189635,
|
|
"num_tokens": 12663744.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 5.833763074874878,
|
|
"epoch": 6.431626056688215,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020258059484806564,
|
|
"loss": 4.9774,
|
|
"mean_token_accuracy": 0.22288461476564408,
|
|
"num_tokens": 12673503.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 5.674204301834107,
|
|
"epoch": 6.43659870711089,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00020221095631813574,
|
|
"loss": 4.9142,
|
|
"mean_token_accuracy": 0.22419310510158538,
|
|
"num_tokens": 12683534.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 5.7701514720916744,
|
|
"epoch": 6.441571357533565,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002018415370729913,
|
|
"loss": 5.0521,
|
|
"mean_token_accuracy": 0.21040195524692534,
|
|
"num_tokens": 12693356.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 5.748992681503296,
|
|
"epoch": 6.44654400795624,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020147233822554722,
|
|
"loss": 4.8869,
|
|
"mean_token_accuracy": 0.22637738138437272,
|
|
"num_tokens": 12702332.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 5.791528701782227,
|
|
"epoch": 6.451516658378916,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020110336088805463,
|
|
"loss": 4.9556,
|
|
"mean_token_accuracy": 0.22013868093490602,
|
|
"num_tokens": 12710759.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 5.811703634262085,
|
|
"epoch": 6.456489308801591,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020073460617209732,
|
|
"loss": 5.129,
|
|
"mean_token_accuracy": 0.20397570431232454,
|
|
"num_tokens": 12720435.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 5.848632764816284,
|
|
"epoch": 6.461461959224266,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020036607518858824,
|
|
"loss": 5.0567,
|
|
"mean_token_accuracy": 0.2102351650595665,
|
|
"num_tokens": 12730004.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 6.461461959224266,
|
|
"eval_entropy": 5.761959979061706,
|
|
"eval_loss": 6.476923942565918,
|
|
"eval_mean_token_accuracy": 0.14885088609110925,
|
|
"eval_num_tokens": 12730004.0,
|
|
"eval_runtime": 11.9696,
|
|
"eval_samples_per_second": 289.484,
|
|
"eval_steps_per_second": 36.259,
|
|
"step": 6500
|
|
}
|
|
],
|
|
"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": 2.0854642823424e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|