6cc3d04e25
Model: fpadovani/jpn-jpan-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.7543859649122808,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 8.045226430892944,
|
|
"epoch": 0.005847953216374269,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.818,
|
|
"mean_token_accuracy": 0.11080007404088973,
|
|
"num_tokens": 13999.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 8.016419839859008,
|
|
"epoch": 0.011695906432748537,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.7674,
|
|
"mean_token_accuracy": 0.12552170008420943,
|
|
"num_tokens": 25888.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.983883237838745,
|
|
"epoch": 0.017543859649122806,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.7832,
|
|
"mean_token_accuracy": 0.12148510366678238,
|
|
"num_tokens": 37699.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 8.008830261230468,
|
|
"epoch": 0.023391812865497075,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.8146,
|
|
"mean_token_accuracy": 0.11834284961223603,
|
|
"num_tokens": 49648.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 8.017287445068359,
|
|
"epoch": 0.029239766081871343,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.7518,
|
|
"mean_token_accuracy": 0.12864405438303947,
|
|
"num_tokens": 62255.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.999194955825805,
|
|
"epoch": 0.03508771929824561,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.792,
|
|
"mean_token_accuracy": 0.11829871609807015,
|
|
"num_tokens": 76003.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.996608638763428,
|
|
"epoch": 0.04093567251461988,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.7807,
|
|
"mean_token_accuracy": 0.12491086795926094,
|
|
"num_tokens": 87585.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 8.053028106689453,
|
|
"epoch": 0.04678362573099415,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.7457,
|
|
"mean_token_accuracy": 0.1264454424381256,
|
|
"num_tokens": 100404.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 8.056242990493775,
|
|
"epoch": 0.05263157894736842,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.7331,
|
|
"mean_token_accuracy": 0.12377208396792412,
|
|
"num_tokens": 113130.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.999864816665649,
|
|
"epoch": 0.05847953216374269,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 7.6657,
|
|
"mean_token_accuracy": 0.13675192892551422,
|
|
"num_tokens": 125430.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 8.017994022369384,
|
|
"epoch": 0.06432748538011696,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.6988,
|
|
"mean_token_accuracy": 0.12444054186344147,
|
|
"num_tokens": 137550.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.978502178192139,
|
|
"epoch": 0.07017543859649122,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.7037,
|
|
"mean_token_accuracy": 0.13199957981705665,
|
|
"num_tokens": 149530.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 8.002739524841308,
|
|
"epoch": 0.07602339181286549,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.7433,
|
|
"mean_token_accuracy": 0.11834913790225983,
|
|
"num_tokens": 161671.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 8.044865322113036,
|
|
"epoch": 0.08187134502923976,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 7.7415,
|
|
"mean_token_accuracy": 0.12546497955918312,
|
|
"num_tokens": 174160.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 8.031145143508912,
|
|
"epoch": 0.08771929824561403,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.7769,
|
|
"mean_token_accuracy": 0.11971935406327247,
|
|
"num_tokens": 186969.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 8.05164942741394,
|
|
"epoch": 0.0935672514619883,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.8155,
|
|
"mean_token_accuracy": 0.12601222917437555,
|
|
"num_tokens": 199470.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 8.037533330917359,
|
|
"epoch": 0.09941520467836257,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.7126,
|
|
"mean_token_accuracy": 0.12613636404275894,
|
|
"num_tokens": 211306.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.987277841567993,
|
|
"epoch": 0.10526315789473684,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.6838,
|
|
"mean_token_accuracy": 0.11928426325321198,
|
|
"num_tokens": 223323.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.975043487548828,
|
|
"epoch": 0.1111111111111111,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.7213,
|
|
"mean_token_accuracy": 0.12729128673672677,
|
|
"num_tokens": 235896.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.983143472671509,
|
|
"epoch": 0.11695906432748537,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.7478,
|
|
"mean_token_accuracy": 0.11931364014744758,
|
|
"num_tokens": 247879.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 8.00935206413269,
|
|
"epoch": 0.12280701754385964,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.643,
|
|
"mean_token_accuracy": 0.12651290744543076,
|
|
"num_tokens": 259714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 8.024541139602661,
|
|
"epoch": 0.1286549707602339,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.7247,
|
|
"mean_token_accuracy": 0.13059264943003654,
|
|
"num_tokens": 272429.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 8.002756261825562,
|
|
"epoch": 0.13450292397660818,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 7.7266,
|
|
"mean_token_accuracy": 0.12073208466172218,
|
|
"num_tokens": 285924.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.962329769134522,
|
|
"epoch": 0.14035087719298245,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.7193,
|
|
"mean_token_accuracy": 0.12854890897870064,
|
|
"num_tokens": 299453.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 8.003749895095826,
|
|
"epoch": 0.14619883040935672,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.7368,
|
|
"mean_token_accuracy": 0.126718270778656,
|
|
"num_tokens": 312295.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 8.033454608917236,
|
|
"epoch": 0.15204678362573099,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.6836,
|
|
"mean_token_accuracy": 0.12897008657455444,
|
|
"num_tokens": 324730.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.970009756088257,
|
|
"epoch": 0.15789473684210525,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.6357,
|
|
"mean_token_accuracy": 0.1297009363770485,
|
|
"num_tokens": 336481.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.994677114486694,
|
|
"epoch": 0.16374269005847952,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6983,
|
|
"mean_token_accuracy": 0.1270280137658119,
|
|
"num_tokens": 348987.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.016926193237305,
|
|
"epoch": 0.1695906432748538,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6826,
|
|
"mean_token_accuracy": 0.12861572131514548,
|
|
"num_tokens": 360390.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.914753246307373,
|
|
"epoch": 0.17543859649122806,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.6898,
|
|
"mean_token_accuracy": 0.12696985229849816,
|
|
"num_tokens": 372456.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.051334762573243,
|
|
"epoch": 0.18128654970760233,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.6932,
|
|
"mean_token_accuracy": 0.1194346971809864,
|
|
"num_tokens": 384185.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.977176713943481,
|
|
"epoch": 0.1871345029239766,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.7047,
|
|
"mean_token_accuracy": 0.1230759084224701,
|
|
"num_tokens": 395399.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.035953950881957,
|
|
"epoch": 0.19298245614035087,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.7026,
|
|
"mean_token_accuracy": 0.12683228254318238,
|
|
"num_tokens": 408676.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.06921739578247,
|
|
"epoch": 0.19883040935672514,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.7717,
|
|
"mean_token_accuracy": 0.12356067821383476,
|
|
"num_tokens": 421861.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.930960464477539,
|
|
"epoch": 0.2046783625730994,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.7268,
|
|
"mean_token_accuracy": 0.12672705948352814,
|
|
"num_tokens": 434852.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.079622077941895,
|
|
"epoch": 0.21052631578947367,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7208,
|
|
"mean_token_accuracy": 0.12138797342777252,
|
|
"num_tokens": 447313.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.027794551849365,
|
|
"epoch": 0.21637426900584794,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.6783,
|
|
"mean_token_accuracy": 0.12779284864664078,
|
|
"num_tokens": 460271.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.856478452682495,
|
|
"epoch": 0.2222222222222222,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.681,
|
|
"mean_token_accuracy": 0.12920279279351235,
|
|
"num_tokens": 473006.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.003684854507446,
|
|
"epoch": 0.22807017543859648,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.654,
|
|
"mean_token_accuracy": 0.1259481579065323,
|
|
"num_tokens": 486100.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.972032690048218,
|
|
"epoch": 0.23391812865497075,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5933,
|
|
"mean_token_accuracy": 0.14003728628158568,
|
|
"num_tokens": 499460.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.951391363143921,
|
|
"epoch": 0.23976608187134502,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.6514,
|
|
"mean_token_accuracy": 0.12902599275112153,
|
|
"num_tokens": 511795.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.95180230140686,
|
|
"epoch": 0.24561403508771928,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7382,
|
|
"mean_token_accuracy": 0.12644921094179154,
|
|
"num_tokens": 524897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.979979419708252,
|
|
"epoch": 0.25146198830409355,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.6319,
|
|
"mean_token_accuracy": 0.1295425295829773,
|
|
"num_tokens": 538325.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.990107727050781,
|
|
"epoch": 0.2573099415204678,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.7107,
|
|
"mean_token_accuracy": 0.12212530672550201,
|
|
"num_tokens": 550979.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.008913707733154,
|
|
"epoch": 0.2631578947368421,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.6891,
|
|
"mean_token_accuracy": 0.12676209211349487,
|
|
"num_tokens": 563001.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.963163757324219,
|
|
"epoch": 0.26900584795321636,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.6043,
|
|
"mean_token_accuracy": 0.12866764068603515,
|
|
"num_tokens": 573600.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.968063545227051,
|
|
"epoch": 0.27485380116959063,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.6265,
|
|
"mean_token_accuracy": 0.1300235703587532,
|
|
"num_tokens": 585287.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.902215766906738,
|
|
"epoch": 0.2807017543859649,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.6554,
|
|
"mean_token_accuracy": 0.12321203649044037,
|
|
"num_tokens": 597386.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.978629684448242,
|
|
"epoch": 0.28654970760233917,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.6527,
|
|
"mean_token_accuracy": 0.12698477283120155,
|
|
"num_tokens": 609256.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.978091716766357,
|
|
"epoch": 0.29239766081871343,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.6533,
|
|
"mean_token_accuracy": 0.12933312207460404,
|
|
"num_tokens": 621506.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.905685138702393,
|
|
"epoch": 0.2982456140350877,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.5799,
|
|
"mean_token_accuracy": 0.13296303376555443,
|
|
"num_tokens": 633512.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.041613578796387,
|
|
"epoch": 0.30409356725146197,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.7242,
|
|
"mean_token_accuracy": 0.12240937054157257,
|
|
"num_tokens": 647098.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.9852275371551515,
|
|
"epoch": 0.30994152046783624,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.7257,
|
|
"mean_token_accuracy": 0.1233548603951931,
|
|
"num_tokens": 659559.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.069615888595582,
|
|
"epoch": 0.3157894736842105,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.6581,
|
|
"mean_token_accuracy": 0.12490400299429893,
|
|
"num_tokens": 672621.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.8597798347473145,
|
|
"epoch": 0.3216374269005848,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6462,
|
|
"mean_token_accuracy": 0.12783507704734803,
|
|
"num_tokens": 685652.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.071738576889038,
|
|
"epoch": 0.32748538011695905,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.7153,
|
|
"mean_token_accuracy": 0.12348538860678673,
|
|
"num_tokens": 696910.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.863040113449097,
|
|
"epoch": 0.3333333333333333,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.578,
|
|
"mean_token_accuracy": 0.13220044150948523,
|
|
"num_tokens": 709054.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.035344123840332,
|
|
"epoch": 0.3391812865497076,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.5782,
|
|
"mean_token_accuracy": 0.1288279578089714,
|
|
"num_tokens": 720902.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.9420812129974365,
|
|
"epoch": 0.34502923976608185,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.682,
|
|
"mean_token_accuracy": 0.12467397823929786,
|
|
"num_tokens": 733319.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.934081220626831,
|
|
"epoch": 0.3508771929824561,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.5585,
|
|
"mean_token_accuracy": 0.13054021298885346,
|
|
"num_tokens": 744817.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.9335449695587155,
|
|
"epoch": 0.3567251461988304,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.6303,
|
|
"mean_token_accuracy": 0.12146333456039429,
|
|
"num_tokens": 756582.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.948399925231934,
|
|
"epoch": 0.36257309941520466,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.5449,
|
|
"mean_token_accuracy": 0.132960344851017,
|
|
"num_tokens": 769273.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.933630418777466,
|
|
"epoch": 0.3684210526315789,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.6025,
|
|
"mean_token_accuracy": 0.1318463236093521,
|
|
"num_tokens": 782001.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.923580074310303,
|
|
"epoch": 0.3742690058479532,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.6304,
|
|
"mean_token_accuracy": 0.13039980456233025,
|
|
"num_tokens": 794575.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.862440013885498,
|
|
"epoch": 0.38011695906432746,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.6157,
|
|
"mean_token_accuracy": 0.12852895334362985,
|
|
"num_tokens": 805583.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.954799365997315,
|
|
"epoch": 0.38596491228070173,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.6656,
|
|
"mean_token_accuracy": 0.12845203652977943,
|
|
"num_tokens": 817617.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.96801438331604,
|
|
"epoch": 0.391812865497076,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.5744,
|
|
"mean_token_accuracy": 0.1288806565105915,
|
|
"num_tokens": 830043.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.787790775299072,
|
|
"epoch": 0.39766081871345027,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.5246,
|
|
"mean_token_accuracy": 0.13534748554229736,
|
|
"num_tokens": 843200.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.927206993103027,
|
|
"epoch": 0.40350877192982454,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.5919,
|
|
"mean_token_accuracy": 0.1352095700800419,
|
|
"num_tokens": 856031.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.932764101028442,
|
|
"epoch": 0.4093567251461988,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.5701,
|
|
"mean_token_accuracy": 0.13280727341771126,
|
|
"num_tokens": 869119.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.834122610092163,
|
|
"epoch": 0.4152046783625731,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.5864,
|
|
"mean_token_accuracy": 0.13350669145584107,
|
|
"num_tokens": 880802.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.999099969863892,
|
|
"epoch": 0.42105263157894735,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.6039,
|
|
"mean_token_accuracy": 0.13265340849757196,
|
|
"num_tokens": 894237.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.928555250167847,
|
|
"epoch": 0.4269005847953216,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.605,
|
|
"mean_token_accuracy": 0.12948731631040572,
|
|
"num_tokens": 906695.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.827572917938232,
|
|
"epoch": 0.4327485380116959,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.5653,
|
|
"mean_token_accuracy": 0.1352884441614151,
|
|
"num_tokens": 918955.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.882963037490844,
|
|
"epoch": 0.43859649122807015,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.6175,
|
|
"mean_token_accuracy": 0.13342018574476242,
|
|
"num_tokens": 931812.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.043187236785888,
|
|
"epoch": 0.4444444444444444,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.6117,
|
|
"mean_token_accuracy": 0.1259584277868271,
|
|
"num_tokens": 945261.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.798225545883179,
|
|
"epoch": 0.4502923976608187,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.5332,
|
|
"mean_token_accuracy": 0.13877285420894622,
|
|
"num_tokens": 958376.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.8840476989746096,
|
|
"epoch": 0.45614035087719296,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.5497,
|
|
"mean_token_accuracy": 0.1363872230052948,
|
|
"num_tokens": 971709.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.952695989608765,
|
|
"epoch": 0.4619883040935672,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.5272,
|
|
"mean_token_accuracy": 0.13542769774794577,
|
|
"num_tokens": 983847.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.852967643737793,
|
|
"epoch": 0.4678362573099415,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.6325,
|
|
"mean_token_accuracy": 0.12114153057336807,
|
|
"num_tokens": 997034.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.9122217178344725,
|
|
"epoch": 0.47368421052631576,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.6585,
|
|
"mean_token_accuracy": 0.12669240608811377,
|
|
"num_tokens": 1010213.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.966189193725586,
|
|
"epoch": 0.47953216374269003,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.5094,
|
|
"mean_token_accuracy": 0.1325475551187992,
|
|
"num_tokens": 1021618.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.752740716934204,
|
|
"epoch": 0.4853801169590643,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.5722,
|
|
"mean_token_accuracy": 0.12650121599435807,
|
|
"num_tokens": 1033809.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.968109512329102,
|
|
"epoch": 0.49122807017543857,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.5408,
|
|
"mean_token_accuracy": 0.13147031590342523,
|
|
"num_tokens": 1045291.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.82607216835022,
|
|
"epoch": 0.49707602339181284,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.5405,
|
|
"mean_token_accuracy": 0.13000031486153601,
|
|
"num_tokens": 1057516.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.792825746536255,
|
|
"epoch": 0.5029239766081871,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.4731,
|
|
"mean_token_accuracy": 0.14245186597108841,
|
|
"num_tokens": 1070675.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.928640842437744,
|
|
"epoch": 0.5087719298245614,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.5646,
|
|
"mean_token_accuracy": 0.13614390641450883,
|
|
"num_tokens": 1082965.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.782242202758789,
|
|
"epoch": 0.5146198830409356,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.5514,
|
|
"mean_token_accuracy": 0.12592832893133163,
|
|
"num_tokens": 1095922.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.94010591506958,
|
|
"epoch": 0.52046783625731,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.5318,
|
|
"mean_token_accuracy": 0.1322290450334549,
|
|
"num_tokens": 1109002.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.902499675750732,
|
|
"epoch": 0.5263157894736842,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.5483,
|
|
"mean_token_accuracy": 0.12942796126008033,
|
|
"num_tokens": 1121199.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.8024513721466064,
|
|
"epoch": 0.5321637426900585,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.498,
|
|
"mean_token_accuracy": 0.13429903537034987,
|
|
"num_tokens": 1134721.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.875277853012085,
|
|
"epoch": 0.5380116959064327,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.5853,
|
|
"mean_token_accuracy": 0.13090707287192344,
|
|
"num_tokens": 1147954.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.777953195571899,
|
|
"epoch": 0.543859649122807,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.4904,
|
|
"mean_token_accuracy": 0.13606545925140381,
|
|
"num_tokens": 1161208.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.901414012908935,
|
|
"epoch": 0.5497076023391813,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.6013,
|
|
"mean_token_accuracy": 0.12688619866967202,
|
|
"num_tokens": 1173789.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.890290832519531,
|
|
"epoch": 0.5555555555555556,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.5108,
|
|
"mean_token_accuracy": 0.13075993657112123,
|
|
"num_tokens": 1185765.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.818354701995849,
|
|
"epoch": 0.5614035087719298,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.5364,
|
|
"mean_token_accuracy": 0.13063954189419746,
|
|
"num_tokens": 1197954.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.836427164077759,
|
|
"epoch": 0.5672514619883041,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.4666,
|
|
"mean_token_accuracy": 0.13487003445625306,
|
|
"num_tokens": 1211285.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.822644567489624,
|
|
"epoch": 0.5730994152046783,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.5392,
|
|
"mean_token_accuracy": 0.12864260450005532,
|
|
"num_tokens": 1223197.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.805648565292358,
|
|
"epoch": 0.5789473684210527,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.4746,
|
|
"mean_token_accuracy": 0.13128072768449783,
|
|
"num_tokens": 1234817.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.884338331222534,
|
|
"epoch": 0.5847953216374269,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.5232,
|
|
"mean_token_accuracy": 0.14004627764225006,
|
|
"num_tokens": 1249305.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5847953216374269,
|
|
"eval_entropy": 7.763977130957409,
|
|
"eval_loss": 7.5110273361206055,
|
|
"eval_mean_token_accuracy": 0.12611977400252689,
|
|
"eval_num_tokens": 1249305.0,
|
|
"eval_runtime": 1.7446,
|
|
"eval_samples_per_second": 1551.671,
|
|
"eval_steps_per_second": 194.317,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.822003698348999,
|
|
"epoch": 0.5906432748538012,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.5147,
|
|
"mean_token_accuracy": 0.1345560982823372,
|
|
"num_tokens": 1262182.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.7908985137939455,
|
|
"epoch": 0.5964912280701754,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.5284,
|
|
"mean_token_accuracy": 0.13263775631785393,
|
|
"num_tokens": 1274616.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.79985671043396,
|
|
"epoch": 0.6023391812865497,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.5451,
|
|
"mean_token_accuracy": 0.12807372063398362,
|
|
"num_tokens": 1287306.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.839459228515625,
|
|
"epoch": 0.6081871345029239,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.4045,
|
|
"mean_token_accuracy": 0.1380430668592453,
|
|
"num_tokens": 1299472.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.752567291259766,
|
|
"epoch": 0.6140350877192983,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.379,
|
|
"mean_token_accuracy": 0.14084984213113785,
|
|
"num_tokens": 1312636.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.679509496688842,
|
|
"epoch": 0.6198830409356725,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.409,
|
|
"mean_token_accuracy": 0.141881512850523,
|
|
"num_tokens": 1324058.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.792089557647705,
|
|
"epoch": 0.6257309941520468,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.512,
|
|
"mean_token_accuracy": 0.12878955602645875,
|
|
"num_tokens": 1337605.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.908940649032592,
|
|
"epoch": 0.631578947368421,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.5055,
|
|
"mean_token_accuracy": 0.12933722659945487,
|
|
"num_tokens": 1350658.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.683991622924805,
|
|
"epoch": 0.6374269005847953,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.4509,
|
|
"mean_token_accuracy": 0.1343121938407421,
|
|
"num_tokens": 1362091.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.78796877861023,
|
|
"epoch": 0.6432748538011696,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.471,
|
|
"mean_token_accuracy": 0.13308046758174896,
|
|
"num_tokens": 1374055.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.648328351974487,
|
|
"epoch": 0.6491228070175439,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.3421,
|
|
"mean_token_accuracy": 0.13799523189663887,
|
|
"num_tokens": 1385269.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.785341358184814,
|
|
"epoch": 0.6549707602339181,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.4657,
|
|
"mean_token_accuracy": 0.13630712032318115,
|
|
"num_tokens": 1396559.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.649790620803833,
|
|
"epoch": 0.6608187134502924,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.4754,
|
|
"mean_token_accuracy": 0.13643540740013121,
|
|
"num_tokens": 1408838.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.809543752670288,
|
|
"epoch": 0.6666666666666666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.3656,
|
|
"mean_token_accuracy": 0.13893612772226333,
|
|
"num_tokens": 1421023.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.6780742645263675,
|
|
"epoch": 0.672514619883041,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.4352,
|
|
"mean_token_accuracy": 0.1361393950879574,
|
|
"num_tokens": 1433912.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.781989669799804,
|
|
"epoch": 0.6783625730994152,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.4323,
|
|
"mean_token_accuracy": 0.1340094953775406,
|
|
"num_tokens": 1445262.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.737918901443481,
|
|
"epoch": 0.6842105263157895,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.3966,
|
|
"mean_token_accuracy": 0.1352713793516159,
|
|
"num_tokens": 1457927.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.741323614120484,
|
|
"epoch": 0.6900584795321637,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.4588,
|
|
"mean_token_accuracy": 0.13115558326244353,
|
|
"num_tokens": 1470116.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.70854115486145,
|
|
"epoch": 0.695906432748538,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.3566,
|
|
"mean_token_accuracy": 0.13966110795736314,
|
|
"num_tokens": 1482067.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.681605339050293,
|
|
"epoch": 0.7017543859649122,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.4402,
|
|
"mean_token_accuracy": 0.13019128441810607,
|
|
"num_tokens": 1493902.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.735611867904663,
|
|
"epoch": 0.7076023391812866,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.4067,
|
|
"mean_token_accuracy": 0.13062553629279136,
|
|
"num_tokens": 1505689.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.676931142807007,
|
|
"epoch": 0.7134502923976608,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.3517,
|
|
"mean_token_accuracy": 0.14924161732196808,
|
|
"num_tokens": 1518002.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.720028305053711,
|
|
"epoch": 0.7192982456140351,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.3775,
|
|
"mean_token_accuracy": 0.13257619887590408,
|
|
"num_tokens": 1529307.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.744207429885864,
|
|
"epoch": 0.7251461988304093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.5186,
|
|
"mean_token_accuracy": 0.12124198526144028,
|
|
"num_tokens": 1542351.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.635728359222412,
|
|
"epoch": 0.7309941520467836,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.3486,
|
|
"mean_token_accuracy": 0.13930814936757088,
|
|
"num_tokens": 1556263.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.675192213058471,
|
|
"epoch": 0.7368421052631579,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.3742,
|
|
"mean_token_accuracy": 0.1383934386074543,
|
|
"num_tokens": 1568906.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.6466429233551025,
|
|
"epoch": 0.7426900584795322,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.3503,
|
|
"mean_token_accuracy": 0.14224105402827264,
|
|
"num_tokens": 1580631.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.686527395248413,
|
|
"epoch": 0.7485380116959064,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.3641,
|
|
"mean_token_accuracy": 0.13352902382612228,
|
|
"num_tokens": 1592369.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.71283073425293,
|
|
"epoch": 0.7543859649122807,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.4124,
|
|
"mean_token_accuracy": 0.1391394928097725,
|
|
"num_tokens": 1604888.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.693008184432983,
|
|
"epoch": 0.7602339181286549,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.335,
|
|
"mean_token_accuracy": 0.13951285406947136,
|
|
"num_tokens": 1617128.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.620254135131836,
|
|
"epoch": 0.7660818713450293,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.3705,
|
|
"mean_token_accuracy": 0.13122161403298377,
|
|
"num_tokens": 1629350.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.713437891006469,
|
|
"epoch": 0.7719298245614035,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.4003,
|
|
"mean_token_accuracy": 0.1302126467227936,
|
|
"num_tokens": 1643082.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.659784364700317,
|
|
"epoch": 0.7777777777777778,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.3814,
|
|
"mean_token_accuracy": 0.13562896028161048,
|
|
"num_tokens": 1654785.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.717230129241943,
|
|
"epoch": 0.783625730994152,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.405,
|
|
"mean_token_accuracy": 0.13106834143400192,
|
|
"num_tokens": 1667704.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.781724691390991,
|
|
"epoch": 0.7894736842105263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.357,
|
|
"mean_token_accuracy": 0.13603928461670875,
|
|
"num_tokens": 1680459.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.590497255325317,
|
|
"epoch": 0.7953216374269005,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.2758,
|
|
"mean_token_accuracy": 0.13571080267429353,
|
|
"num_tokens": 1692912.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.570317316055298,
|
|
"epoch": 0.8011695906432749,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.3071,
|
|
"mean_token_accuracy": 0.13977469876408577,
|
|
"num_tokens": 1705329.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.754013347625732,
|
|
"epoch": 0.8070175438596491,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.3897,
|
|
"mean_token_accuracy": 0.1318515956401825,
|
|
"num_tokens": 1717283.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.542753267288208,
|
|
"epoch": 0.8128654970760234,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.3316,
|
|
"mean_token_accuracy": 0.1392970860004425,
|
|
"num_tokens": 1730308.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.708332633972168,
|
|
"epoch": 0.8187134502923976,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.3765,
|
|
"mean_token_accuracy": 0.13571367934346198,
|
|
"num_tokens": 1742716.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.568768167495728,
|
|
"epoch": 0.8245614035087719,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.3157,
|
|
"mean_token_accuracy": 0.14167511239647865,
|
|
"num_tokens": 1754183.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.634387493133545,
|
|
"epoch": 0.8304093567251462,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.3343,
|
|
"mean_token_accuracy": 0.1332976020872593,
|
|
"num_tokens": 1766601.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.588458633422851,
|
|
"epoch": 0.8362573099415205,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.2395,
|
|
"mean_token_accuracy": 0.14873942732810974,
|
|
"num_tokens": 1778756.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.434172487258911,
|
|
"epoch": 0.8421052631578947,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.3105,
|
|
"mean_token_accuracy": 0.14211145639419556,
|
|
"num_tokens": 1791485.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.56705412864685,
|
|
"epoch": 0.847953216374269,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.3197,
|
|
"mean_token_accuracy": 0.13355855271220207,
|
|
"num_tokens": 1803712.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.577960348129272,
|
|
"epoch": 0.8538011695906432,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.2931,
|
|
"mean_token_accuracy": 0.13647336065769194,
|
|
"num_tokens": 1818387.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.667880868911743,
|
|
"epoch": 0.8596491228070176,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.3142,
|
|
"mean_token_accuracy": 0.13656700849533082,
|
|
"num_tokens": 1830113.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.547010898590088,
|
|
"epoch": 0.8654970760233918,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.3893,
|
|
"mean_token_accuracy": 0.13586266636848449,
|
|
"num_tokens": 1841789.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.559738063812256,
|
|
"epoch": 0.8713450292397661,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.2702,
|
|
"mean_token_accuracy": 0.1386386424303055,
|
|
"num_tokens": 1853669.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.552060985565186,
|
|
"epoch": 0.8771929824561403,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.3102,
|
|
"mean_token_accuracy": 0.137037692964077,
|
|
"num_tokens": 1865322.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.495941781997681,
|
|
"epoch": 0.8830409356725146,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.2358,
|
|
"mean_token_accuracy": 0.14545494318008423,
|
|
"num_tokens": 1879039.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.621470403671265,
|
|
"epoch": 0.8888888888888888,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.362,
|
|
"mean_token_accuracy": 0.1306812882423401,
|
|
"num_tokens": 1893078.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.691838264465332,
|
|
"epoch": 0.8947368421052632,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.386,
|
|
"mean_token_accuracy": 0.1329762138426304,
|
|
"num_tokens": 1907399.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.540285968780518,
|
|
"epoch": 0.9005847953216374,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.3331,
|
|
"mean_token_accuracy": 0.1288886435329914,
|
|
"num_tokens": 1919290.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.695523977279663,
|
|
"epoch": 0.9064327485380117,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.301,
|
|
"mean_token_accuracy": 0.13419981822371482,
|
|
"num_tokens": 1931658.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.506599998474121,
|
|
"epoch": 0.9122807017543859,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.2914,
|
|
"mean_token_accuracy": 0.13456346243619918,
|
|
"num_tokens": 1943881.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.5762388706207275,
|
|
"epoch": 0.9181286549707602,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.2317,
|
|
"mean_token_accuracy": 0.13697463721036912,
|
|
"num_tokens": 1956288.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.47317852973938,
|
|
"epoch": 0.9239766081871345,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.1189,
|
|
"mean_token_accuracy": 0.14074339866638183,
|
|
"num_tokens": 1967265.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.464909362792969,
|
|
"epoch": 0.9298245614035088,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.2537,
|
|
"mean_token_accuracy": 0.136859755218029,
|
|
"num_tokens": 1980039.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.564469909667968,
|
|
"epoch": 0.935672514619883,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.2341,
|
|
"mean_token_accuracy": 0.13667627722024916,
|
|
"num_tokens": 1992735.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.528766298294068,
|
|
"epoch": 0.9415204678362573,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.2443,
|
|
"mean_token_accuracy": 0.14110952094197274,
|
|
"num_tokens": 2004545.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.447480821609497,
|
|
"epoch": 0.9473684210526315,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.2416,
|
|
"mean_token_accuracy": 0.13759677186608316,
|
|
"num_tokens": 2017100.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.442005157470703,
|
|
"epoch": 0.9532163742690059,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.1837,
|
|
"mean_token_accuracy": 0.1485427126288414,
|
|
"num_tokens": 2029868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.489367485046387,
|
|
"epoch": 0.9590643274853801,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.2581,
|
|
"mean_token_accuracy": 0.13477363735437392,
|
|
"num_tokens": 2041334.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.601362180709839,
|
|
"epoch": 0.9649122807017544,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.3372,
|
|
"mean_token_accuracy": 0.13630635887384415,
|
|
"num_tokens": 2054797.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.494726514816284,
|
|
"epoch": 0.9707602339181286,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.2108,
|
|
"mean_token_accuracy": 0.13611383736133575,
|
|
"num_tokens": 2068337.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.5208252429962155,
|
|
"epoch": 0.9766081871345029,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.246,
|
|
"mean_token_accuracy": 0.1431644394993782,
|
|
"num_tokens": 2079984.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.471163415908814,
|
|
"epoch": 0.9824561403508771,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.2326,
|
|
"mean_token_accuracy": 0.14368441179394723,
|
|
"num_tokens": 2093069.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.470370626449585,
|
|
"epoch": 0.9883040935672515,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.1945,
|
|
"mean_token_accuracy": 0.14249885827302933,
|
|
"num_tokens": 2104753.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.432940340042114,
|
|
"epoch": 0.9941520467836257,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.2132,
|
|
"mean_token_accuracy": 0.13686279132962226,
|
|
"num_tokens": 2116576.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.479133415222168,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.2019,
|
|
"mean_token_accuracy": 0.13272840678691863,
|
|
"num_tokens": 2129484.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.452660703659058,
|
|
"epoch": 1.0058479532163742,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.0498,
|
|
"mean_token_accuracy": 0.15022014528512956,
|
|
"num_tokens": 2142000.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.31855959892273,
|
|
"epoch": 1.0116959064327484,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.0071,
|
|
"mean_token_accuracy": 0.15096633061766623,
|
|
"num_tokens": 2152965.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.434969520568847,
|
|
"epoch": 1.0175438596491229,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.0943,
|
|
"mean_token_accuracy": 0.13671399429440498,
|
|
"num_tokens": 2165078.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.309150886535645,
|
|
"epoch": 1.023391812865497,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.0169,
|
|
"mean_token_accuracy": 0.14662902280688286,
|
|
"num_tokens": 2176706.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.506210374832153,
|
|
"epoch": 1.0292397660818713,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.0426,
|
|
"mean_token_accuracy": 0.14441205859184264,
|
|
"num_tokens": 2190157.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.395573568344116,
|
|
"epoch": 1.0350877192982457,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.0273,
|
|
"mean_token_accuracy": 0.14645258858799934,
|
|
"num_tokens": 2201821.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.278140211105347,
|
|
"epoch": 1.04093567251462,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.9704,
|
|
"mean_token_accuracy": 0.15083624422550201,
|
|
"num_tokens": 2214504.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.436160612106323,
|
|
"epoch": 1.0467836257309941,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.0734,
|
|
"mean_token_accuracy": 0.14177168905735016,
|
|
"num_tokens": 2226318.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.396118307113648,
|
|
"epoch": 1.0526315789473684,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.1067,
|
|
"mean_token_accuracy": 0.13549974411725998,
|
|
"num_tokens": 2237892.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.349207878112793,
|
|
"epoch": 1.0584795321637426,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.0502,
|
|
"mean_token_accuracy": 0.14364722073078157,
|
|
"num_tokens": 2250656.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.357096576690674,
|
|
"epoch": 1.064327485380117,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.07,
|
|
"mean_token_accuracy": 0.1390746206045151,
|
|
"num_tokens": 2263685.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.401983308792114,
|
|
"epoch": 1.0701754385964912,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.0579,
|
|
"mean_token_accuracy": 0.13671347573399545,
|
|
"num_tokens": 2275413.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.350112628936768,
|
|
"epoch": 1.0760233918128654,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.0372,
|
|
"mean_token_accuracy": 0.14231925532221795,
|
|
"num_tokens": 2288165.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.44090895652771,
|
|
"epoch": 1.0818713450292399,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.0406,
|
|
"mean_token_accuracy": 0.13954761996865273,
|
|
"num_tokens": 2300574.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.308967161178589,
|
|
"epoch": 1.087719298245614,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.0408,
|
|
"mean_token_accuracy": 0.14026969522237778,
|
|
"num_tokens": 2311631.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.384764003753662,
|
|
"epoch": 1.0935672514619883,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.0431,
|
|
"mean_token_accuracy": 0.14462336376309395,
|
|
"num_tokens": 2324312.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.358921909332276,
|
|
"epoch": 1.0994152046783625,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.1237,
|
|
"mean_token_accuracy": 0.1363001599907875,
|
|
"num_tokens": 2337747.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.332943248748779,
|
|
"epoch": 1.1052631578947367,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.0609,
|
|
"mean_token_accuracy": 0.1438799172639847,
|
|
"num_tokens": 2349830.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.372280883789062,
|
|
"epoch": 1.1111111111111112,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.9989,
|
|
"mean_token_accuracy": 0.15035150349140167,
|
|
"num_tokens": 2361825.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.287798833847046,
|
|
"epoch": 1.1169590643274854,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.9498,
|
|
"mean_token_accuracy": 0.1477070778608322,
|
|
"num_tokens": 2373963.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.3268575191497805,
|
|
"epoch": 1.1228070175438596,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.1058,
|
|
"mean_token_accuracy": 0.13842709735035896,
|
|
"num_tokens": 2386348.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.385518598556518,
|
|
"epoch": 1.128654970760234,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.0137,
|
|
"mean_token_accuracy": 0.14070830792188643,
|
|
"num_tokens": 2398338.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.311811065673828,
|
|
"epoch": 1.1345029239766082,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.9778,
|
|
"mean_token_accuracy": 0.146958364546299,
|
|
"num_tokens": 2411855.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.2889463901519775,
|
|
"epoch": 1.1403508771929824,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.035,
|
|
"mean_token_accuracy": 0.14933026432991028,
|
|
"num_tokens": 2424271.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.293201684951782,
|
|
"epoch": 1.1461988304093567,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.05,
|
|
"mean_token_accuracy": 0.14216556251049042,
|
|
"num_tokens": 2436668.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.34758563041687,
|
|
"epoch": 1.1520467836257309,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.9859,
|
|
"mean_token_accuracy": 0.14464660882949829,
|
|
"num_tokens": 2449451.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.311631584167481,
|
|
"epoch": 1.1578947368421053,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.107,
|
|
"mean_token_accuracy": 0.1359217494726181,
|
|
"num_tokens": 2463107.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.347780799865722,
|
|
"epoch": 1.1637426900584795,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.0446,
|
|
"mean_token_accuracy": 0.14254012927412987,
|
|
"num_tokens": 2475443.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.332783555984497,
|
|
"epoch": 1.1695906432748537,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.9846,
|
|
"mean_token_accuracy": 0.14220065772533416,
|
|
"num_tokens": 2486884.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1695906432748537,
|
|
"eval_entropy": 7.181294761927782,
|
|
"eval_loss": 7.163454055786133,
|
|
"eval_mean_token_accuracy": 0.1324625013418122,
|
|
"eval_num_tokens": 2486884.0,
|
|
"eval_runtime": 1.7467,
|
|
"eval_samples_per_second": 1549.805,
|
|
"eval_steps_per_second": 194.083,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.23369197845459,
|
|
"epoch": 1.1754385964912282,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999996875147963,
|
|
"loss": 6.9283,
|
|
"mean_token_accuracy": 0.15480874925851823,
|
|
"num_tokens": 2499310.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.339850234985351,
|
|
"epoch": 1.1812865497076024,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999984180451436,
|
|
"loss": 7.0605,
|
|
"mean_token_accuracy": 0.13472636938095092,
|
|
"num_tokens": 2512230.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.260563325881958,
|
|
"epoch": 1.1871345029239766,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999961720662221,
|
|
"loss": 7.1069,
|
|
"mean_token_accuracy": 0.13796906247735025,
|
|
"num_tokens": 2525255.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.321180963516236,
|
|
"epoch": 1.1929824561403508,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999929495877796,
|
|
"loss": 6.9933,
|
|
"mean_token_accuracy": 0.14501239061355592,
|
|
"num_tokens": 2537672.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.1882189273834225,
|
|
"epoch": 1.198830409356725,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999887506238018,
|
|
"loss": 7.0226,
|
|
"mean_token_accuracy": 0.141133663803339,
|
|
"num_tokens": 2550371.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.301802682876587,
|
|
"epoch": 1.2046783625730995,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999835751925124,
|
|
"loss": 7.0414,
|
|
"mean_token_accuracy": 0.13682982176542283,
|
|
"num_tokens": 2563708.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.325049591064453,
|
|
"epoch": 1.2105263157894737,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999774233163734,
|
|
"loss": 7.0121,
|
|
"mean_token_accuracy": 0.13734989762306213,
|
|
"num_tokens": 2576529.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.25710163116455,
|
|
"epoch": 1.2163742690058479,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999702950220841,
|
|
"loss": 7.0276,
|
|
"mean_token_accuracy": 0.1384323477745056,
|
|
"num_tokens": 2588970.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.273972702026367,
|
|
"epoch": 1.2222222222222223,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999621903405819,
|
|
"loss": 6.9847,
|
|
"mean_token_accuracy": 0.14251093938946724,
|
|
"num_tokens": 2601134.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.237666416168213,
|
|
"epoch": 1.2280701754385965,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499953109307042,
|
|
"loss": 7.0436,
|
|
"mean_token_accuracy": 0.13975274115800856,
|
|
"num_tokens": 2612670.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.238549900054932,
|
|
"epoch": 1.2339181286549707,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999430519608762,
|
|
"loss": 7.001,
|
|
"mean_token_accuracy": 0.1480314277112484,
|
|
"num_tokens": 2625157.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.279294061660766,
|
|
"epoch": 1.239766081871345,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004999320183457346,
|
|
"loss": 6.9897,
|
|
"mean_token_accuracy": 0.13806044161319733,
|
|
"num_tokens": 2638318.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.243882465362549,
|
|
"epoch": 1.2456140350877192,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999200085095035,
|
|
"loss": 6.9805,
|
|
"mean_token_accuracy": 0.14396776705980302,
|
|
"num_tokens": 2651024.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.0745433330535885,
|
|
"epoch": 1.2514619883040936,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999070225043068,
|
|
"loss": 6.908,
|
|
"mean_token_accuracy": 0.14976727068424225,
|
|
"num_tokens": 2665568.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.382400941848755,
|
|
"epoch": 1.2573099415204678,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998930603865042,
|
|
"loss": 7.0444,
|
|
"mean_token_accuracy": 0.1408499911427498,
|
|
"num_tokens": 2678748.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.243338489532471,
|
|
"epoch": 1.263157894736842,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004998781222166929,
|
|
"loss": 6.9727,
|
|
"mean_token_accuracy": 0.14128838628530502,
|
|
"num_tokens": 2690682.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.114080858230591,
|
|
"epoch": 1.2690058479532165,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004998622080597052,
|
|
"loss": 7.0118,
|
|
"mean_token_accuracy": 0.13777799755334855,
|
|
"num_tokens": 2703077.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.226661300659179,
|
|
"epoch": 1.2748538011695907,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998453179846098,
|
|
"loss": 6.9777,
|
|
"mean_token_accuracy": 0.1415637917816639,
|
|
"num_tokens": 2714609.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.2645416259765625,
|
|
"epoch": 1.280701754385965,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998274520647109,
|
|
"loss": 6.9431,
|
|
"mean_token_accuracy": 0.14714309424161912,
|
|
"num_tokens": 2726907.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.118586158752441,
|
|
"epoch": 1.286549707602339,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998086103775478,
|
|
"loss": 6.9558,
|
|
"mean_token_accuracy": 0.13884984627366065,
|
|
"num_tokens": 2739248.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.270521402359009,
|
|
"epoch": 1.2923976608187133,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004997887930048947,
|
|
"loss": 6.9136,
|
|
"mean_token_accuracy": 0.14900762140750884,
|
|
"num_tokens": 2753842.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.079877853393555,
|
|
"epoch": 1.2982456140350878,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004997680000327607,
|
|
"loss": 6.9682,
|
|
"mean_token_accuracy": 0.13750396743416787,
|
|
"num_tokens": 2767249.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.311344146728516,
|
|
"epoch": 1.304093567251462,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997462315513884,
|
|
"loss": 6.9739,
|
|
"mean_token_accuracy": 0.14469007700681685,
|
|
"num_tokens": 2779657.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.179875421524048,
|
|
"epoch": 1.3099415204678362,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.000499723487655255,
|
|
"loss": 6.8902,
|
|
"mean_token_accuracy": 0.14483164623379707,
|
|
"num_tokens": 2792760.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.133006525039673,
|
|
"epoch": 1.3157894736842106,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004996997684430703,
|
|
"loss": 6.919,
|
|
"mean_token_accuracy": 0.1405123069882393,
|
|
"num_tokens": 2805277.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.2041998386383055,
|
|
"epoch": 1.3216374269005848,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004996750740177775,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.14064337760210038,
|
|
"num_tokens": 2818341.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.228417444229126,
|
|
"epoch": 1.327485380116959,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996494044865521,
|
|
"loss": 6.9637,
|
|
"mean_token_accuracy": 0.142166855186224,
|
|
"num_tokens": 2830199.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.088128757476807,
|
|
"epoch": 1.3333333333333333,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004996227599608018,
|
|
"loss": 6.8593,
|
|
"mean_token_accuracy": 0.1505097910761833,
|
|
"num_tokens": 2842141.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.227330923080444,
|
|
"epoch": 1.3391812865497075,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995951405561658,
|
|
"loss": 6.962,
|
|
"mean_token_accuracy": 0.13390366807579995,
|
|
"num_tokens": 2854929.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.122727584838867,
|
|
"epoch": 1.345029239766082,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004995665463925142,
|
|
"loss": 6.956,
|
|
"mean_token_accuracy": 0.14488306567072867,
|
|
"num_tokens": 2867034.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.140340137481689,
|
|
"epoch": 1.3508771929824561,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004995369775939478,
|
|
"loss": 6.8525,
|
|
"mean_token_accuracy": 0.155217444896698,
|
|
"num_tokens": 2881254.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.168582820892334,
|
|
"epoch": 1.3567251461988303,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004995064342887972,
|
|
"loss": 6.9836,
|
|
"mean_token_accuracy": 0.14219432771205903,
|
|
"num_tokens": 2894592.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.14756326675415,
|
|
"epoch": 1.3625730994152048,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004994749166096226,
|
|
"loss": 6.8909,
|
|
"mean_token_accuracy": 0.14301346614956856,
|
|
"num_tokens": 2907196.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.098900032043457,
|
|
"epoch": 1.368421052631579,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004994424246932129,
|
|
"loss": 6.9694,
|
|
"mean_token_accuracy": 0.14125872850418092,
|
|
"num_tokens": 2920398.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.166225385665894,
|
|
"epoch": 1.3742690058479532,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004994089586805856,
|
|
"loss": 6.8781,
|
|
"mean_token_accuracy": 0.14636376053094863,
|
|
"num_tokens": 2932052.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.2833469867706295,
|
|
"epoch": 1.3801169590643274,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004993745187169852,
|
|
"loss": 6.9581,
|
|
"mean_token_accuracy": 0.14133822321891784,
|
|
"num_tokens": 2943620.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.041484117507935,
|
|
"epoch": 1.3859649122807016,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499339104951884,
|
|
"loss": 6.8894,
|
|
"mean_token_accuracy": 0.15769957304000853,
|
|
"num_tokens": 2956159.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.195301103591919,
|
|
"epoch": 1.391812865497076,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00049930271753898,
|
|
"loss": 6.9178,
|
|
"mean_token_accuracy": 0.14382270127534866,
|
|
"num_tokens": 2967332.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.145396423339844,
|
|
"epoch": 1.3976608187134503,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004992653566361974,
|
|
"loss": 6.9224,
|
|
"mean_token_accuracy": 0.1394818566739559,
|
|
"num_tokens": 2980168.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.163400173187256,
|
|
"epoch": 1.4035087719298245,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004992270224056849,
|
|
"loss": 6.9953,
|
|
"mean_token_accuracy": 0.1428208038210869,
|
|
"num_tokens": 2993509.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.177978467941284,
|
|
"epoch": 1.409356725146199,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004991877150138162,
|
|
"loss": 6.9178,
|
|
"mean_token_accuracy": 0.14307066798210144,
|
|
"num_tokens": 3005827.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.057064533233643,
|
|
"epoch": 1.4152046783625731,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004991474346311879,
|
|
"loss": 6.8952,
|
|
"mean_token_accuracy": 0.14541105851531028,
|
|
"num_tokens": 3018968.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.2221822261810305,
|
|
"epoch": 1.4210526315789473,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991061814326198,
|
|
"loss": 6.9432,
|
|
"mean_token_accuracy": 0.13681356310844422,
|
|
"num_tokens": 3031892.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.1108252048492435,
|
|
"epoch": 1.4269005847953216,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.0004990639555971539,
|
|
"loss": 6.8643,
|
|
"mean_token_accuracy": 0.1525570660829544,
|
|
"num_tokens": 3045994.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.054197692871094,
|
|
"epoch": 1.4327485380116958,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004990207573080533,
|
|
"loss": 6.8826,
|
|
"mean_token_accuracy": 0.14980751127004624,
|
|
"num_tokens": 3058304.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.114181137084961,
|
|
"epoch": 1.4385964912280702,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004989765867528014,
|
|
"loss": 6.8822,
|
|
"mean_token_accuracy": 0.14745792597532273,
|
|
"num_tokens": 3071102.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.0125102519989015,
|
|
"epoch": 1.4444444444444444,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004989314441231019,
|
|
"loss": 6.9189,
|
|
"mean_token_accuracy": 0.140367042273283,
|
|
"num_tokens": 3084653.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.203199291229248,
|
|
"epoch": 1.4502923976608186,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000498885329614877,
|
|
"loss": 6.9363,
|
|
"mean_token_accuracy": 0.1474703371524811,
|
|
"num_tokens": 3097425.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.045669031143189,
|
|
"epoch": 1.456140350877193,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004988382434282668,
|
|
"loss": 6.9427,
|
|
"mean_token_accuracy": 0.14563710391521453,
|
|
"num_tokens": 3111898.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.218185138702393,
|
|
"epoch": 1.4619883040935673,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498790185767629,
|
|
"loss": 6.8721,
|
|
"mean_token_accuracy": 0.14673303216695785,
|
|
"num_tokens": 3124271.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.061633968353272,
|
|
"epoch": 1.4678362573099415,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000498741156841537,
|
|
"loss": 6.9632,
|
|
"mean_token_accuracy": 0.1429533690214157,
|
|
"num_tokens": 3136344.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.175415992736816,
|
|
"epoch": 1.4736842105263157,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00049869115686278,
|
|
"loss": 6.862,
|
|
"mean_token_accuracy": 0.14745716527104377,
|
|
"num_tokens": 3148299.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.136249303817749,
|
|
"epoch": 1.47953216374269,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004986401860483615,
|
|
"loss": 6.9367,
|
|
"mean_token_accuracy": 0.1429634302854538,
|
|
"num_tokens": 3161537.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.095357179641724,
|
|
"epoch": 1.4853801169590644,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004985882446194984,
|
|
"loss": 6.8773,
|
|
"mean_token_accuracy": 0.1452200636267662,
|
|
"num_tokens": 3173676.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.1800398349761965,
|
|
"epoch": 1.4912280701754386,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 0.0004985353328016204,
|
|
"loss": 6.832,
|
|
"mean_token_accuracy": 0.15665666610002518,
|
|
"num_tokens": 3186542.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.005247592926025,
|
|
"epoch": 1.4970760233918128,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004984814508243685,
|
|
"loss": 6.8534,
|
|
"mean_token_accuracy": 0.1438918486237526,
|
|
"num_tokens": 3197867.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.137544727325439,
|
|
"epoch": 1.5029239766081872,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004984265989215942,
|
|
"loss": 6.8381,
|
|
"mean_token_accuracy": 0.14461125507950784,
|
|
"num_tokens": 3209671.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.019346570968628,
|
|
"epoch": 1.5087719298245614,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004983707773313589,
|
|
"loss": 6.8255,
|
|
"mean_token_accuracy": 0.14992894530296325,
|
|
"num_tokens": 3220707.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.017501497268677,
|
|
"epoch": 1.5146198830409356,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004983139862959322,
|
|
"loss": 6.8087,
|
|
"mean_token_accuracy": 0.15319009572267533,
|
|
"num_tokens": 3235122.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.046554613113403,
|
|
"epoch": 1.52046783625731,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004982562260617912,
|
|
"loss": 6.8076,
|
|
"mean_token_accuracy": 0.14265020713210105,
|
|
"num_tokens": 3246833.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.087526750564575,
|
|
"epoch": 1.526315789473684,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004981974968796197,
|
|
"loss": 6.9071,
|
|
"mean_token_accuracy": 0.14019423797726632,
|
|
"num_tokens": 3260493.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.032801389694214,
|
|
"epoch": 1.5321637426900585,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004981377990043064,
|
|
"loss": 6.8352,
|
|
"mean_token_accuracy": 0.14121123254299164,
|
|
"num_tokens": 3272445.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.047143602371216,
|
|
"epoch": 1.5380116959064327,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004980771326949442,
|
|
"loss": 6.8525,
|
|
"mean_token_accuracy": 0.1418970376253128,
|
|
"num_tokens": 3284360.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.10487470626831,
|
|
"epoch": 1.543859649122807,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004980154982148296,
|
|
"loss": 6.8378,
|
|
"mean_token_accuracy": 0.14478948563337327,
|
|
"num_tokens": 3296841.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.979324865341186,
|
|
"epoch": 1.5497076023391814,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004979528958314604,
|
|
"loss": 6.838,
|
|
"mean_token_accuracy": 0.14271837919950486,
|
|
"num_tokens": 3309393.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.140212059020996,
|
|
"epoch": 1.5555555555555556,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004978893258165354,
|
|
"loss": 6.9083,
|
|
"mean_token_accuracy": 0.14211394041776657,
|
|
"num_tokens": 3322094.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.07458724975586,
|
|
"epoch": 1.5614035087719298,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004978247884459532,
|
|
"loss": 6.8339,
|
|
"mean_token_accuracy": 0.1480436772108078,
|
|
"num_tokens": 3334468.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.014019727706909,
|
|
"epoch": 1.5672514619883042,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004977592839998105,
|
|
"loss": 6.8153,
|
|
"mean_token_accuracy": 0.14521542564034462,
|
|
"num_tokens": 3346367.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.983550500869751,
|
|
"epoch": 1.5730994152046782,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004976928127624013,
|
|
"loss": 6.8099,
|
|
"mean_token_accuracy": 0.15263736099004746,
|
|
"num_tokens": 3357463.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.006190252304077,
|
|
"epoch": 1.5789473684210527,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004976253750222157,
|
|
"loss": 6.8012,
|
|
"mean_token_accuracy": 0.14718832671642304,
|
|
"num_tokens": 3369465.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.040311193466186,
|
|
"epoch": 1.5847953216374269,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004975569710719382,
|
|
"loss": 6.9177,
|
|
"mean_token_accuracy": 0.13584902808070182,
|
|
"num_tokens": 3382859.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.103788089752197,
|
|
"epoch": 1.590643274853801,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004974876012084468,
|
|
"loss": 6.8631,
|
|
"mean_token_accuracy": 0.1401561588048935,
|
|
"num_tokens": 3395431.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.0188991069793705,
|
|
"epoch": 1.5964912280701755,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004974172657328116,
|
|
"loss": 6.7815,
|
|
"mean_token_accuracy": 0.1481684312224388,
|
|
"num_tokens": 3408388.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.982085657119751,
|
|
"epoch": 1.6023391812865497,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004973459649502939,
|
|
"loss": 6.7716,
|
|
"mean_token_accuracy": 0.15199798196554185,
|
|
"num_tokens": 3421275.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.018893909454346,
|
|
"epoch": 1.608187134502924,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004972736991703439,
|
|
"loss": 6.8261,
|
|
"mean_token_accuracy": 0.14990319609642028,
|
|
"num_tokens": 3433537.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.0294441223144535,
|
|
"epoch": 1.6140350877192984,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004972004687066005,
|
|
"loss": 6.815,
|
|
"mean_token_accuracy": 0.14912112429738045,
|
|
"num_tokens": 3445167.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.945680761337281,
|
|
"epoch": 1.6198830409356724,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000497126273876889,
|
|
"loss": 6.7912,
|
|
"mean_token_accuracy": 0.15700830221176149,
|
|
"num_tokens": 3458529.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.987841701507568,
|
|
"epoch": 1.6257309941520468,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004970511150032205,
|
|
"loss": 6.7995,
|
|
"mean_token_accuracy": 0.15150927156209945,
|
|
"num_tokens": 3470315.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.013829851150513,
|
|
"epoch": 1.631578947368421,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004969749924117896,
|
|
"loss": 6.7354,
|
|
"mean_token_accuracy": 0.1469464585185051,
|
|
"num_tokens": 3481899.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.945081186294556,
|
|
"epoch": 1.6374269005847952,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004968979064329741,
|
|
"loss": 6.8463,
|
|
"mean_token_accuracy": 0.15086084008216857,
|
|
"num_tokens": 3494076.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.1430888175964355,
|
|
"epoch": 1.6432748538011697,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004968198574013323,
|
|
"loss": 6.8567,
|
|
"mean_token_accuracy": 0.14028689563274382,
|
|
"num_tokens": 3507670.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.98843731880188,
|
|
"epoch": 1.6491228070175439,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000496740845655603,
|
|
"loss": 6.765,
|
|
"mean_token_accuracy": 0.14957668334245683,
|
|
"num_tokens": 3519704.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.061826419830322,
|
|
"epoch": 1.654970760233918,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004966608715387029,
|
|
"loss": 6.8341,
|
|
"mean_token_accuracy": 0.14651029407978058,
|
|
"num_tokens": 3531665.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.909399461746216,
|
|
"epoch": 1.6608187134502925,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004965799353977248,
|
|
"loss": 6.8201,
|
|
"mean_token_accuracy": 0.15219928324222565,
|
|
"num_tokens": 3543876.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.972405481338501,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004964980375839379,
|
|
"loss": 6.7517,
|
|
"mean_token_accuracy": 0.15124111026525497,
|
|
"num_tokens": 3556517.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.9292216300964355,
|
|
"epoch": 1.672514619883041,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004964151784527844,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.15513818413019181,
|
|
"num_tokens": 3569378.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.04847559928894,
|
|
"epoch": 1.6783625730994152,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000496331358363879,
|
|
"loss": 6.7612,
|
|
"mean_token_accuracy": 0.14962893724441528,
|
|
"num_tokens": 3581684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.899519062042236,
|
|
"epoch": 1.6842105263157894,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004962465776810066,
|
|
"loss": 6.6936,
|
|
"mean_token_accuracy": 0.1610645532608032,
|
|
"num_tokens": 3594961.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.936596918106079,
|
|
"epoch": 1.6900584795321638,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004961608367721217,
|
|
"loss": 6.7638,
|
|
"mean_token_accuracy": 0.14676244854927062,
|
|
"num_tokens": 3607197.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.986419010162353,
|
|
"epoch": 1.695906432748538,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004960741360093459,
|
|
"loss": 6.7792,
|
|
"mean_token_accuracy": 0.1482525236904621,
|
|
"num_tokens": 3618728.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.044631910324097,
|
|
"epoch": 1.7017543859649122,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004959864757689669,
|
|
"loss": 6.8896,
|
|
"mean_token_accuracy": 0.1363150730729103,
|
|
"num_tokens": 3630872.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.024134588241577,
|
|
"epoch": 1.7076023391812867,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004958978564314361,
|
|
"loss": 6.7727,
|
|
"mean_token_accuracy": 0.15003089904785155,
|
|
"num_tokens": 3642657.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.922481203079224,
|
|
"epoch": 1.7134502923976607,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000495808278381368,
|
|
"loss": 6.7874,
|
|
"mean_token_accuracy": 0.14720278084278107,
|
|
"num_tokens": 3654505.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.898501586914063,
|
|
"epoch": 1.719298245614035,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004957177420075377,
|
|
"loss": 6.7039,
|
|
"mean_token_accuracy": 0.1616862505674362,
|
|
"num_tokens": 3668554.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.935784864425659,
|
|
"epoch": 1.7251461988304093,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004956262477028797,
|
|
"loss": 6.8482,
|
|
"mean_token_accuracy": 0.14284570664167404,
|
|
"num_tokens": 3681448.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.9588398933410645,
|
|
"epoch": 1.7309941520467835,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004955337958644857,
|
|
"loss": 6.7823,
|
|
"mean_token_accuracy": 0.14315327405929565,
|
|
"num_tokens": 3693677.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.937526559829712,
|
|
"epoch": 1.736842105263158,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004954403868936032,
|
|
"loss": 6.772,
|
|
"mean_token_accuracy": 0.15349175333976744,
|
|
"num_tokens": 3706204.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.011352682113648,
|
|
"epoch": 1.7426900584795322,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495346021195634,
|
|
"loss": 6.7382,
|
|
"mean_token_accuracy": 0.153244549036026,
|
|
"num_tokens": 3718255.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.956983232498169,
|
|
"epoch": 1.7485380116959064,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004952506991801319,
|
|
"loss": 6.7591,
|
|
"mean_token_accuracy": 0.1534613236784935,
|
|
"num_tokens": 3730623.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.98687424659729,
|
|
"epoch": 1.7543859649122808,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004951544212608015,
|
|
"loss": 6.7806,
|
|
"mean_token_accuracy": 0.14909627735614778,
|
|
"num_tokens": 3743636.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7543859649122808,
|
|
"eval_entropy": 6.749826677429289,
|
|
"eval_loss": 6.9042558670043945,
|
|
"eval_mean_token_accuracy": 0.1385386679733617,
|
|
"eval_num_tokens": 3743636.0,
|
|
"eval_runtime": 1.7314,
|
|
"eval_samples_per_second": 1563.503,
|
|
"eval_steps_per_second": 195.799,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 8540,
|
|
"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": 803006462730240.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|