6cc3d04e25
Model: fpadovani/jpn-jpan-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
11156 lines
304 KiB
JSON
11156 lines
304 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.432748538011696,
|
|
"eval_steps": 500,
|
|
"global_step": 5500,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.998679256439209,
|
|
"epoch": 1.7602339181286548,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004950571878554956,
|
|
"loss": 6.8474,
|
|
"mean_token_accuracy": 0.14006248787045478,
|
|
"num_tokens": 3755608.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.015962743759156,
|
|
"epoch": 1.7660818713450293,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004949589993862143,
|
|
"loss": 6.7274,
|
|
"mean_token_accuracy": 0.1476696565747261,
|
|
"num_tokens": 3768208.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.909123659133911,
|
|
"epoch": 1.7719298245614035,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004948598562791025,
|
|
"loss": 6.7562,
|
|
"mean_token_accuracy": 0.14620479792356492,
|
|
"num_tokens": 3780550.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.959772539138794,
|
|
"epoch": 1.7777777777777777,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004947597589644484,
|
|
"loss": 6.7369,
|
|
"mean_token_accuracy": 0.14819346666336058,
|
|
"num_tokens": 3792417.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.95702714920044,
|
|
"epoch": 1.7836257309941521,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004946587078766818,
|
|
"loss": 6.7084,
|
|
"mean_token_accuracy": 0.14805916994810103,
|
|
"num_tokens": 3803887.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.961219358444214,
|
|
"epoch": 1.7894736842105263,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004945567034543712,
|
|
"loss": 6.8511,
|
|
"mean_token_accuracy": 0.1391899310052395,
|
|
"num_tokens": 3816534.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.987950134277344,
|
|
"epoch": 1.7953216374269005,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004944537461402234,
|
|
"loss": 6.7979,
|
|
"mean_token_accuracy": 0.14789642840623857,
|
|
"num_tokens": 3827790.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.950696229934692,
|
|
"epoch": 1.801169590643275,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004943498363810807,
|
|
"loss": 6.7444,
|
|
"mean_token_accuracy": 0.1527763783931732,
|
|
"num_tokens": 3839800.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.80357174873352,
|
|
"epoch": 1.807017543859649,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004942449746279184,
|
|
"loss": 6.6761,
|
|
"mean_token_accuracy": 0.15447496101260186,
|
|
"num_tokens": 3852621.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.954777812957763,
|
|
"epoch": 1.8128654970760234,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004941391613358444,
|
|
"loss": 6.7447,
|
|
"mean_token_accuracy": 0.14117342829704285,
|
|
"num_tokens": 3864184.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.999409198760986,
|
|
"epoch": 1.8187134502923976,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000494032396964096,
|
|
"loss": 6.7195,
|
|
"mean_token_accuracy": 0.14479989558458328,
|
|
"num_tokens": 3876750.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.909908151626587,
|
|
"epoch": 1.8245614035087718,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000493924681976038,
|
|
"loss": 6.765,
|
|
"mean_token_accuracy": 0.15119151771068573,
|
|
"num_tokens": 3888788.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.870877361297607,
|
|
"epoch": 1.8304093567251463,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004938160168391614,
|
|
"loss": 6.8113,
|
|
"mean_token_accuracy": 0.14277349263429642,
|
|
"num_tokens": 3901816.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.910139703750611,
|
|
"epoch": 1.8362573099415205,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004937064020250804,
|
|
"loss": 6.7192,
|
|
"mean_token_accuracy": 0.15223183929920198,
|
|
"num_tokens": 3913127.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.960526704788208,
|
|
"epoch": 1.8421052631578947,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004935958380095316,
|
|
"loss": 6.7258,
|
|
"mean_token_accuracy": 0.1503272145986557,
|
|
"num_tokens": 3924878.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.9379799365997314,
|
|
"epoch": 1.8479532163742691,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004934843252723702,
|
|
"loss": 6.8122,
|
|
"mean_token_accuracy": 0.14785331189632417,
|
|
"num_tokens": 3938334.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.9066784381866455,
|
|
"epoch": 1.8538011695906431,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00049337186429757,
|
|
"loss": 6.7632,
|
|
"mean_token_accuracy": 0.14622444063425064,
|
|
"num_tokens": 3949733.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.020902299880982,
|
|
"epoch": 1.8596491228070176,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004932584555732192,
|
|
"loss": 6.796,
|
|
"mean_token_accuracy": 0.14879967719316484,
|
|
"num_tokens": 3963109.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.93715009689331,
|
|
"epoch": 1.8654970760233918,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004931440995915203,
|
|
"loss": 6.7623,
|
|
"mean_token_accuracy": 0.14756920784711838,
|
|
"num_tokens": 3975477.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.860741233825683,
|
|
"epoch": 1.871345029239766,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000493028796848786,
|
|
"loss": 6.729,
|
|
"mean_token_accuracy": 0.15381080508232117,
|
|
"num_tokens": 3987757.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.963431406021118,
|
|
"epoch": 1.8771929824561404,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004929125478454389,
|
|
"loss": 6.6288,
|
|
"mean_token_accuracy": 0.15386789590120314,
|
|
"num_tokens": 3998807.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.80572738647461,
|
|
"epoch": 1.8830409356725146,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004927953530860077,
|
|
"loss": 6.6824,
|
|
"mean_token_accuracy": 0.15219125896692276,
|
|
"num_tokens": 4010320.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.929991149902344,
|
|
"epoch": 1.8888888888888888,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004926772130791261,
|
|
"loss": 6.8253,
|
|
"mean_token_accuracy": 0.14115398228168488,
|
|
"num_tokens": 4023432.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.975191450119018,
|
|
"epoch": 1.8947368421052633,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004925581283375303,
|
|
"loss": 6.8053,
|
|
"mean_token_accuracy": 0.14290309697389603,
|
|
"num_tokens": 4036282.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.837200069427491,
|
|
"epoch": 1.9005847953216373,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004924380993780566,
|
|
"loss": 6.6825,
|
|
"mean_token_accuracy": 0.15446478575468064,
|
|
"num_tokens": 4048164.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.882443904876709,
|
|
"epoch": 1.9064327485380117,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004923171267216392,
|
|
"loss": 6.7106,
|
|
"mean_token_accuracy": 0.15049419552087784,
|
|
"num_tokens": 4059287.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.894489479064942,
|
|
"epoch": 1.912280701754386,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004921952108933083,
|
|
"loss": 6.7098,
|
|
"mean_token_accuracy": 0.14405074566602707,
|
|
"num_tokens": 4071976.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.829108381271363,
|
|
"epoch": 1.9181286549707601,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004920723524221871,
|
|
"loss": 6.732,
|
|
"mean_token_accuracy": 0.1510712221264839,
|
|
"num_tokens": 4085116.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.996309614181518,
|
|
"epoch": 1.9239766081871346,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004919485518414902,
|
|
"loss": 6.7352,
|
|
"mean_token_accuracy": 0.14357687830924987,
|
|
"num_tokens": 4096778.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.72740797996521,
|
|
"epoch": 1.9298245614035088,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004918238096885213,
|
|
"loss": 6.6573,
|
|
"mean_token_accuracy": 0.14709970653057097,
|
|
"num_tokens": 4109490.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.898422527313232,
|
|
"epoch": 1.935672514619883,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00049169812650467,
|
|
"loss": 6.6938,
|
|
"mean_token_accuracy": 0.151672625541687,
|
|
"num_tokens": 4122832.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.936109113693237,
|
|
"epoch": 1.9415204678362574,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004915715028354102,
|
|
"loss": 6.7214,
|
|
"mean_token_accuracy": 0.14411992505192756,
|
|
"num_tokens": 4134585.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.8710979461669925,
|
|
"epoch": 1.9473684210526314,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004914439392302981,
|
|
"loss": 6.6729,
|
|
"mean_token_accuracy": 0.14913154765963554,
|
|
"num_tokens": 4145909.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.881339979171753,
|
|
"epoch": 1.9532163742690059,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004913154362429686,
|
|
"loss": 6.5815,
|
|
"mean_token_accuracy": 0.1624261111021042,
|
|
"num_tokens": 4157360.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.859464550018311,
|
|
"epoch": 1.95906432748538,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004911859944311341,
|
|
"loss": 6.7562,
|
|
"mean_token_accuracy": 0.14489269107580185,
|
|
"num_tokens": 4169789.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.925325965881347,
|
|
"epoch": 1.9649122807017543,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004910556143565812,
|
|
"loss": 6.7577,
|
|
"mean_token_accuracy": 0.1472465306520462,
|
|
"num_tokens": 4183131.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.8551278591156,
|
|
"epoch": 1.9707602339181287,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004909242965851688,
|
|
"loss": 6.7374,
|
|
"mean_token_accuracy": 0.14674712345004082,
|
|
"num_tokens": 4194865.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.911614084243775,
|
|
"epoch": 1.976608187134503,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004907920416868253,
|
|
"loss": 6.7426,
|
|
"mean_token_accuracy": 0.1460190460085869,
|
|
"num_tokens": 4207406.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.831352806091308,
|
|
"epoch": 1.9824561403508771,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004906588502355467,
|
|
"loss": 6.6926,
|
|
"mean_token_accuracy": 0.14983116388320922,
|
|
"num_tokens": 4221623.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.866110467910767,
|
|
"epoch": 1.9883040935672516,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004905247228093933,
|
|
"loss": 6.6396,
|
|
"mean_token_accuracy": 0.1614004448056221,
|
|
"num_tokens": 4233699.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.787024068832397,
|
|
"epoch": 1.9941520467836256,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004903896599904878,
|
|
"loss": 6.6382,
|
|
"mean_token_accuracy": 0.1509261205792427,
|
|
"num_tokens": 4247669.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.890104055404663,
|
|
"epoch": 2.0,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004902536623650125,
|
|
"loss": 6.6938,
|
|
"mean_token_accuracy": 0.1474872499704361,
|
|
"num_tokens": 4258968.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.827988338470459,
|
|
"epoch": 2.0058479532163744,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004901167305232068,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.1567946717143059,
|
|
"num_tokens": 4270684.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.813016700744629,
|
|
"epoch": 2.0116959064327484,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004899788650593648,
|
|
"loss": 6.3439,
|
|
"mean_token_accuracy": 0.15991095155477525,
|
|
"num_tokens": 4282567.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.813668727874756,
|
|
"epoch": 2.017543859649123,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004898400665718325,
|
|
"loss": 6.3958,
|
|
"mean_token_accuracy": 0.1543136343359947,
|
|
"num_tokens": 4294803.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.83916745185852,
|
|
"epoch": 2.023391812865497,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004897003356630052,
|
|
"loss": 6.4091,
|
|
"mean_token_accuracy": 0.1572381541132927,
|
|
"num_tokens": 4305898.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.75266432762146,
|
|
"epoch": 2.0292397660818713,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004895596729393252,
|
|
"loss": 6.3202,
|
|
"mean_token_accuracy": 0.16260852962732314,
|
|
"num_tokens": 4317985.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.707140350341797,
|
|
"epoch": 2.0350877192982457,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004894180790112787,
|
|
"loss": 6.3909,
|
|
"mean_token_accuracy": 0.1585234597325325,
|
|
"num_tokens": 4330482.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.88357458114624,
|
|
"epoch": 2.0409356725146197,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004892755544933934,
|
|
"loss": 6.4595,
|
|
"mean_token_accuracy": 0.14881203919649125,
|
|
"num_tokens": 4342782.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.811342191696167,
|
|
"epoch": 2.046783625730994,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004891321000042361,
|
|
"loss": 6.4566,
|
|
"mean_token_accuracy": 0.15072599798440933,
|
|
"num_tokens": 4356204.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.7372157096862795,
|
|
"epoch": 2.0526315789473686,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 6.3753,
|
|
"mean_token_accuracy": 0.15697392523288728,
|
|
"num_tokens": 4370003.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.821414375305176,
|
|
"epoch": 2.0584795321637426,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048884240360655,
|
|
"loss": 6.4135,
|
|
"mean_token_accuracy": 0.153708016872406,
|
|
"num_tokens": 4381972.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.72168664932251,
|
|
"epoch": 2.064327485380117,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004886961629553242,
|
|
"loss": 6.3326,
|
|
"mean_token_accuracy": 0.16419654041528703,
|
|
"num_tokens": 4394033.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.757396078109741,
|
|
"epoch": 2.0701754385964914,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004885489948474272,
|
|
"loss": 6.4438,
|
|
"mean_token_accuracy": 0.15901652127504348,
|
|
"num_tokens": 4406799.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.825146055221557,
|
|
"epoch": 2.0760233918128654,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004884008999215792,
|
|
"loss": 6.4505,
|
|
"mean_token_accuracy": 0.1513996809720993,
|
|
"num_tokens": 4420182.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.729333925247192,
|
|
"epoch": 2.08187134502924,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004882518788205225,
|
|
"loss": 6.4222,
|
|
"mean_token_accuracy": 0.15328143090009688,
|
|
"num_tokens": 4432014.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.8558149337768555,
|
|
"epoch": 2.087719298245614,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004881019321910198,
|
|
"loss": 6.4085,
|
|
"mean_token_accuracy": 0.1588073641061783,
|
|
"num_tokens": 4444755.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.713996887207031,
|
|
"epoch": 2.0935672514619883,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048795106068384985,
|
|
"loss": 6.4627,
|
|
"mean_token_accuracy": 0.15979111194610596,
|
|
"num_tokens": 4456390.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.802612447738648,
|
|
"epoch": 2.0994152046783627,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000487799264953806,
|
|
"loss": 6.3984,
|
|
"mean_token_accuracy": 0.1593334347009659,
|
|
"num_tokens": 4468567.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.708228302001953,
|
|
"epoch": 2.1052631578947367,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004876465456596926,
|
|
"loss": 6.355,
|
|
"mean_token_accuracy": 0.163933627307415,
|
|
"num_tokens": 4479553.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.73302903175354,
|
|
"epoch": 2.111111111111111,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004874929034643223,
|
|
"loss": 6.3748,
|
|
"mean_token_accuracy": 0.166480416059494,
|
|
"num_tokens": 4491505.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.732792568206787,
|
|
"epoch": 2.116959064327485,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004873383390345132,
|
|
"loss": 6.4515,
|
|
"mean_token_accuracy": 0.15464365035295485,
|
|
"num_tokens": 4504896.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.722332382202149,
|
|
"epoch": 2.1228070175438596,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048718285304108613,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.15354599058628082,
|
|
"num_tokens": 4516319.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.785979557037353,
|
|
"epoch": 2.128654970760234,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048702644615886137,
|
|
"loss": 6.444,
|
|
"mean_token_accuracy": 0.15366278141736983,
|
|
"num_tokens": 4529132.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.737329578399658,
|
|
"epoch": 2.134502923976608,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048686911906665597,
|
|
"loss": 6.4034,
|
|
"mean_token_accuracy": 0.15596025735139846,
|
|
"num_tokens": 4541080.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.7506248474121096,
|
|
"epoch": 2.1403508771929824,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00048671087244728093,
|
|
"loss": 6.4262,
|
|
"mean_token_accuracy": 0.15710604190826416,
|
|
"num_tokens": 4553442.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.656120491027832,
|
|
"epoch": 2.146198830409357,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00048655170698753774,
|
|
"loss": 6.3467,
|
|
"mean_token_accuracy": 0.1673656329512596,
|
|
"num_tokens": 4565617.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.828509378433227,
|
|
"epoch": 2.152046783625731,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000486391623378216,
|
|
"loss": 6.4452,
|
|
"mean_token_accuracy": 0.15144958794116975,
|
|
"num_tokens": 4577418.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.74020562171936,
|
|
"epoch": 2.1578947368421053,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004862306223140901,
|
|
"loss": 6.3782,
|
|
"mean_token_accuracy": 0.15981672555208207,
|
|
"num_tokens": 4589763.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.737591505050659,
|
|
"epoch": 2.1637426900584797,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048606870449391604,
|
|
"loss": 6.4601,
|
|
"mean_token_accuracy": 0.16001340448856355,
|
|
"num_tokens": 4602522.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.807626438140869,
|
|
"epoch": 2.1695906432748537,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00048590587062042907,
|
|
"loss": 6.4247,
|
|
"mean_token_accuracy": 0.16054905503988265,
|
|
"num_tokens": 4614680.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.736106109619141,
|
|
"epoch": 2.175438596491228,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048574212140033965,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.15460269302129745,
|
|
"num_tokens": 4628166.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.786086177825927,
|
|
"epoch": 2.181286549707602,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004855774575443311,
|
|
"loss": 6.467,
|
|
"mean_token_accuracy": 0.15853571593761445,
|
|
"num_tokens": 4640200.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.687403678894043,
|
|
"epoch": 2.1871345029239766,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00048541187976705645,
|
|
"loss": 6.4076,
|
|
"mean_token_accuracy": 0.15247596949338912,
|
|
"num_tokens": 4652968.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.758385753631591,
|
|
"epoch": 2.192982456140351,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048524538878713517,
|
|
"loss": 6.41,
|
|
"mean_token_accuracy": 0.15483911335468292,
|
|
"num_tokens": 4665741.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.741532802581787,
|
|
"epoch": 2.198830409356725,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00048507798532715005,
|
|
"loss": 6.4392,
|
|
"mean_token_accuracy": 0.1583689719438553,
|
|
"num_tokens": 4680074.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.715676784515381,
|
|
"epoch": 2.2046783625730995,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048490967011364394,
|
|
"loss": 6.464,
|
|
"mean_token_accuracy": 0.15492096841335296,
|
|
"num_tokens": 4692340.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.750712442398071,
|
|
"epoch": 2.2105263157894735,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004847404438771172,
|
|
"loss": 6.4514,
|
|
"mean_token_accuracy": 0.15536247342824935,
|
|
"num_tokens": 4704470.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.768217754364014,
|
|
"epoch": 2.216374269005848,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004845703073520236,
|
|
"loss": 6.4739,
|
|
"mean_token_accuracy": 0.15126457586884498,
|
|
"num_tokens": 4716253.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.783787202835083,
|
|
"epoch": 2.2222222222222223,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00048439926127676796,
|
|
"loss": 6.5792,
|
|
"mean_token_accuracy": 0.14901455491781235,
|
|
"num_tokens": 4730083.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.766217136383057,
|
|
"epoch": 2.2280701754385963,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00048422730639370256,
|
|
"loss": 6.3512,
|
|
"mean_token_accuracy": 0.16607776284217834,
|
|
"num_tokens": 4742770.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.684992122650146,
|
|
"epoch": 2.2339181286549707,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004840544434491238,
|
|
"loss": 6.4802,
|
|
"mean_token_accuracy": 0.15634565502405168,
|
|
"num_tokens": 4754299.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.754200983047485,
|
|
"epoch": 2.239766081871345,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048388067319326937,
|
|
"loss": 6.4477,
|
|
"mean_token_accuracy": 0.1498482957482338,
|
|
"num_tokens": 4766964.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.724980974197388,
|
|
"epoch": 2.245614035087719,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048370599638031464,
|
|
"loss": 6.4645,
|
|
"mean_token_accuracy": 0.15188252553343773,
|
|
"num_tokens": 4779921.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.7860071659088135,
|
|
"epoch": 2.2514619883040936,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048353041376836937,
|
|
"loss": 6.4507,
|
|
"mean_token_accuracy": 0.15500609576702118,
|
|
"num_tokens": 4792233.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.721638202667236,
|
|
"epoch": 2.257309941520468,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004833539261194749,
|
|
"loss": 6.4418,
|
|
"mean_token_accuracy": 0.16221449077129363,
|
|
"num_tokens": 4803348.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.746827745437622,
|
|
"epoch": 2.263157894736842,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00048317653419960007,
|
|
"loss": 6.5179,
|
|
"mean_token_accuracy": 0.14920955896377563,
|
|
"num_tokens": 4816574.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.650460290908813,
|
|
"epoch": 2.2690058479532165,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048299823877863857,
|
|
"loss": 6.4275,
|
|
"mean_token_accuracy": 0.16205373406410217,
|
|
"num_tokens": 4828008.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.807295799255371,
|
|
"epoch": 2.2748538011695905,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004828190406304054,
|
|
"loss": 6.5065,
|
|
"mean_token_accuracy": 0.1508103094995022,
|
|
"num_tokens": 4840621.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.71404767036438,
|
|
"epoch": 2.280701754385965,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004826389405326334,
|
|
"loss": 6.3941,
|
|
"mean_token_accuracy": 0.15502528846263885,
|
|
"num_tokens": 4853503.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.770388221740722,
|
|
"epoch": 2.2865497076023393,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004824579392669698,
|
|
"loss": 6.4145,
|
|
"mean_token_accuracy": 0.15558422803878785,
|
|
"num_tokens": 4865925.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.737335300445556,
|
|
"epoch": 2.2923976608187133,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048227603761897316,
|
|
"loss": 6.4974,
|
|
"mean_token_accuracy": 0.15278878808021545,
|
|
"num_tokens": 4878704.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.752950620651245,
|
|
"epoch": 2.2982456140350878,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004820932363781096,
|
|
"loss": 6.4382,
|
|
"mean_token_accuracy": 0.15426261425018312,
|
|
"num_tokens": 4891750.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.7050700187683105,
|
|
"epoch": 2.3040935672514617,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004819095363377496,
|
|
"loss": 6.4288,
|
|
"mean_token_accuracy": 0.16031316965818404,
|
|
"num_tokens": 4904539.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.66979079246521,
|
|
"epoch": 2.309941520467836,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048172493829516465,
|
|
"loss": 6.3265,
|
|
"mean_token_accuracy": 0.16127625554800035,
|
|
"num_tokens": 4916210.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.7233682632446286,
|
|
"epoch": 2.3157894736842106,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004815394430515234,
|
|
"loss": 6.3931,
|
|
"mean_token_accuracy": 0.15900776535272598,
|
|
"num_tokens": 4929011.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.632206392288208,
|
|
"epoch": 2.3216374269005846,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004813530514118884,
|
|
"loss": 6.3776,
|
|
"mean_token_accuracy": 0.15831104964017867,
|
|
"num_tokens": 4942484.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.69144344329834,
|
|
"epoch": 2.327485380116959,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004811657641852129,
|
|
"loss": 6.3611,
|
|
"mean_token_accuracy": 0.15635986775159835,
|
|
"num_tokens": 4955395.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.7046935081481935,
|
|
"epoch": 2.3333333333333335,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048097758218433687,
|
|
"loss": 6.395,
|
|
"mean_token_accuracy": 0.15986014455556868,
|
|
"num_tokens": 4966853.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.564965295791626,
|
|
"epoch": 2.3391812865497075,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048078850622598355,
|
|
"loss": 6.2784,
|
|
"mean_token_accuracy": 0.16841577887535095,
|
|
"num_tokens": 4978514.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.3391812865497075,
|
|
"eval_entropy": 6.477566384284546,
|
|
"eval_loss": 6.7816619873046875,
|
|
"eval_mean_token_accuracy": 0.14170035538826808,
|
|
"eval_num_tokens": 4978514.0,
|
|
"eval_runtime": 1.735,
|
|
"eval_samples_per_second": 1560.257,
|
|
"eval_steps_per_second": 195.392,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.68201093673706,
|
|
"epoch": 2.345029239766082,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048059853713075637,
|
|
"loss": 6.4089,
|
|
"mean_token_accuracy": 0.15335486680269242,
|
|
"num_tokens": 4991331.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.663340282440186,
|
|
"epoch": 2.3508771929824563,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004804076757231347,
|
|
"loss": 6.3966,
|
|
"mean_token_accuracy": 0.1558781698346138,
|
|
"num_tokens": 5004652.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.723086738586426,
|
|
"epoch": 2.3567251461988303,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048021592283147073,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.16744109243154526,
|
|
"num_tokens": 5017204.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.603569030761719,
|
|
"epoch": 2.3625730994152048,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004800232792879858,
|
|
"loss": 6.4342,
|
|
"mean_token_accuracy": 0.1633380264043808,
|
|
"num_tokens": 5029667.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.577605772018432,
|
|
"epoch": 2.3684210526315788,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047982974592876665,
|
|
"loss": 6.2965,
|
|
"mean_token_accuracy": 0.16584915220737456,
|
|
"num_tokens": 5041490.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.706831741333008,
|
|
"epoch": 2.374269005847953,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000479635323593762,
|
|
"loss": 6.4454,
|
|
"mean_token_accuracy": 0.15232941806316375,
|
|
"num_tokens": 5054353.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.669787502288818,
|
|
"epoch": 2.3801169590643276,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047944001312677867,
|
|
"loss": 6.3963,
|
|
"mean_token_accuracy": 0.15449538677930832,
|
|
"num_tokens": 5067322.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.760600280761719,
|
|
"epoch": 2.3859649122807016,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004792438153754782,
|
|
"loss": 6.4509,
|
|
"mean_token_accuracy": 0.16203467249870301,
|
|
"num_tokens": 5079957.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.617363405227661,
|
|
"epoch": 2.391812865497076,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004790467311913727,
|
|
"loss": 6.4151,
|
|
"mean_token_accuracy": 0.1563390463590622,
|
|
"num_tokens": 5091975.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.719149446487426,
|
|
"epoch": 2.39766081871345,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004788487614298219,
|
|
"loss": 6.4384,
|
|
"mean_token_accuracy": 0.1564145416021347,
|
|
"num_tokens": 5103847.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.63304591178894,
|
|
"epoch": 2.4035087719298245,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004786499069500286,
|
|
"loss": 6.3619,
|
|
"mean_token_accuracy": 0.16103654056787492,
|
|
"num_tokens": 5115153.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.5946839332580565,
|
|
"epoch": 2.409356725146199,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047845016861503556,
|
|
"loss": 6.3115,
|
|
"mean_token_accuracy": 0.17103153020143508,
|
|
"num_tokens": 5128008.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.595819759368896,
|
|
"epoch": 2.415204678362573,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047824954729172136,
|
|
"loss": 6.3053,
|
|
"mean_token_accuracy": 0.1675003096461296,
|
|
"num_tokens": 5140023.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.642523908615113,
|
|
"epoch": 2.4210526315789473,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004780480438507972,
|
|
"loss": 6.3409,
|
|
"mean_token_accuracy": 0.16591034084558487,
|
|
"num_tokens": 5152660.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.621842288970948,
|
|
"epoch": 2.426900584795322,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047784565916680216,
|
|
"loss": 6.3419,
|
|
"mean_token_accuracy": 0.15960612446069716,
|
|
"num_tokens": 5164481.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.694283151626587,
|
|
"epoch": 2.4327485380116958,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004776423941181005,
|
|
"loss": 6.4148,
|
|
"mean_token_accuracy": 0.1627780683338642,
|
|
"num_tokens": 5177178.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.662910604476929,
|
|
"epoch": 2.43859649122807,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047743824958687694,
|
|
"loss": 6.3242,
|
|
"mean_token_accuracy": 0.1637122377753258,
|
|
"num_tokens": 5189385.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.671608304977417,
|
|
"epoch": 2.4444444444444446,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047723322645913354,
|
|
"loss": 6.3967,
|
|
"mean_token_accuracy": 0.15904711037874222,
|
|
"num_tokens": 5201839.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.640072441101074,
|
|
"epoch": 2.4502923976608186,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004770273256246853,
|
|
"loss": 6.4296,
|
|
"mean_token_accuracy": 0.15411125719547272,
|
|
"num_tokens": 5214565.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.666627836227417,
|
|
"epoch": 2.456140350877193,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004768205479771566,
|
|
"loss": 6.3979,
|
|
"mean_token_accuracy": 0.1613724946975708,
|
|
"num_tokens": 5227029.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.71558108329773,
|
|
"epoch": 2.461988304093567,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00047661289441397724,
|
|
"loss": 6.4165,
|
|
"mean_token_accuracy": 0.16418871134519578,
|
|
"num_tokens": 5239208.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.596176862716675,
|
|
"epoch": 2.4678362573099415,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047640436583637865,
|
|
"loss": 6.3858,
|
|
"mean_token_accuracy": 0.16437364965677262,
|
|
"num_tokens": 5252272.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.737638711929321,
|
|
"epoch": 2.473684210526316,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00047619496314938977,
|
|
"loss": 6.5001,
|
|
"mean_token_accuracy": 0.1528855338692665,
|
|
"num_tokens": 5264387.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.590871715545655,
|
|
"epoch": 2.47953216374269,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.0004759846872618332,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.1567104309797287,
|
|
"num_tokens": 5276662.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.714838886260987,
|
|
"epoch": 2.4853801169590644,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047577353908632156,
|
|
"loss": 6.4057,
|
|
"mean_token_accuracy": 0.16444096565246583,
|
|
"num_tokens": 5290734.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.559372758865356,
|
|
"epoch": 2.4912280701754383,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047556151953925287,
|
|
"loss": 6.2934,
|
|
"mean_token_accuracy": 0.16795622557401657,
|
|
"num_tokens": 5302510.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.728880214691162,
|
|
"epoch": 2.497076023391813,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004753486295408073,
|
|
"loss": 6.367,
|
|
"mean_token_accuracy": 0.1610953539609909,
|
|
"num_tokens": 5313243.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.527223587036133,
|
|
"epoch": 2.502923976608187,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047513487001494257,
|
|
"loss": 6.4541,
|
|
"mean_token_accuracy": 0.15720518827438354,
|
|
"num_tokens": 5325485.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.80053596496582,
|
|
"epoch": 2.5087719298245617,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00047492024188939056,
|
|
"loss": 6.4904,
|
|
"mean_token_accuracy": 0.15193130373954772,
|
|
"num_tokens": 5339335.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.724518489837647,
|
|
"epoch": 2.5146198830409356,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004747047460956525,
|
|
"loss": 6.4582,
|
|
"mean_token_accuracy": 0.15313778072595596,
|
|
"num_tokens": 5351195.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.576109123229981,
|
|
"epoch": 2.52046783625731,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004744883835689958,
|
|
"loss": 6.3918,
|
|
"mean_token_accuracy": 0.16303829550743104,
|
|
"num_tokens": 5362884.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.630385446548462,
|
|
"epoch": 2.526315789473684,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004742711552484492,
|
|
"loss": 6.4316,
|
|
"mean_token_accuracy": 0.15381407290697097,
|
|
"num_tokens": 5375770.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.7010797500610355,
|
|
"epoch": 2.5321637426900585,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00047405306207679926,
|
|
"loss": 6.4131,
|
|
"mean_token_accuracy": 0.15431758612394333,
|
|
"num_tokens": 5388673.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.5139001369476315,
|
|
"epoch": 2.538011695906433,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004738341050005859,
|
|
"loss": 6.2541,
|
|
"mean_token_accuracy": 0.169919852912426,
|
|
"num_tokens": 5400454.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.658669376373291,
|
|
"epoch": 2.543859649122807,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004736142849700986,
|
|
"loss": 6.3808,
|
|
"mean_token_accuracy": 0.1611814633011818,
|
|
"num_tokens": 5413740.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.70971646308899,
|
|
"epoch": 2.5497076023391814,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004733936029393721,
|
|
"loss": 6.4029,
|
|
"mean_token_accuracy": 0.16251128613948823,
|
|
"num_tokens": 5426089.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.631553030014038,
|
|
"epoch": 2.5555555555555554,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004731720598661821,
|
|
"loss": 6.4487,
|
|
"mean_token_accuracy": 0.15272153913974762,
|
|
"num_tokens": 5438835.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.610339641571045,
|
|
"epoch": 2.56140350877193,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004729496567120415,
|
|
"loss": 6.3975,
|
|
"mean_token_accuracy": 0.1609238341450691,
|
|
"num_tokens": 5452054.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.617287492752075,
|
|
"epoch": 2.5672514619883042,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000472726394442196,
|
|
"loss": 6.3596,
|
|
"mean_token_accuracy": 0.15381914675235747,
|
|
"num_tokens": 5466455.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.618226528167725,
|
|
"epoch": 2.573099415204678,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047250227402561976,
|
|
"loss": 6.3663,
|
|
"mean_token_accuracy": 0.16046204268932343,
|
|
"num_tokens": 5479990.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.672522830963135,
|
|
"epoch": 2.5789473684210527,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004722772964350115,
|
|
"loss": 6.4165,
|
|
"mean_token_accuracy": 0.16307691484689713,
|
|
"num_tokens": 5493035.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.564087581634522,
|
|
"epoch": 2.5847953216374266,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047205146264679015,
|
|
"loss": 6.4402,
|
|
"mean_token_accuracy": 0.1571527138352394,
|
|
"num_tokens": 5506093.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.61999020576477,
|
|
"epoch": 2.590643274853801,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00047182477364109045,
|
|
"loss": 6.4048,
|
|
"mean_token_accuracy": 0.16972629576921464,
|
|
"num_tokens": 5518552.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.619844722747803,
|
|
"epoch": 2.5964912280701755,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047159723040175896,
|
|
"loss": 6.4267,
|
|
"mean_token_accuracy": 0.16126966029405593,
|
|
"num_tokens": 5530123.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.721620512008667,
|
|
"epoch": 2.60233918128655,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00047136883391634966,
|
|
"loss": 6.4362,
|
|
"mean_token_accuracy": 0.15450142174959183,
|
|
"num_tokens": 5542769.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.470245790481568,
|
|
"epoch": 2.608187134502924,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004711395851761198,
|
|
"loss": 6.2405,
|
|
"mean_token_accuracy": 0.1802494168281555,
|
|
"num_tokens": 5555975.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.587647151947022,
|
|
"epoch": 2.6140350877192984,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004709094851760251,
|
|
"loss": 6.3537,
|
|
"mean_token_accuracy": 0.16206099689006806,
|
|
"num_tokens": 5569059.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.68525652885437,
|
|
"epoch": 2.6198830409356724,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004706785349147163,
|
|
"loss": 6.4279,
|
|
"mean_token_accuracy": 0.15201290100812911,
|
|
"num_tokens": 5581358.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.574829769134522,
|
|
"epoch": 2.625730994152047,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000470446735394534,
|
|
"loss": 6.3709,
|
|
"mean_token_accuracy": 0.15791332423686982,
|
|
"num_tokens": 5594256.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.701315546035767,
|
|
"epoch": 2.6315789473684212,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004702140876215047,
|
|
"loss": 6.3875,
|
|
"mean_token_accuracy": 0.1589572623372078,
|
|
"num_tokens": 5607236.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.651389122009277,
|
|
"epoch": 2.6374269005847952,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004699805926053365,
|
|
"loss": 6.4792,
|
|
"mean_token_accuracy": 0.15163235664367675,
|
|
"num_tokens": 5619550.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.604261875152588,
|
|
"epoch": 2.6432748538011697,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00046974625135941445,
|
|
"loss": 6.3421,
|
|
"mean_token_accuracy": 0.16901533007621766,
|
|
"num_tokens": 5632721.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.528126525878906,
|
|
"epoch": 2.6491228070175437,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00046951106490079636,
|
|
"loss": 6.2096,
|
|
"mean_token_accuracy": 0.16352829337120056,
|
|
"num_tokens": 5643672.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.604205131530762,
|
|
"epoch": 2.654970760233918,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00046927503425020847,
|
|
"loss": 6.34,
|
|
"mean_token_accuracy": 0.15846267193555832,
|
|
"num_tokens": 5656072.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.614410305023194,
|
|
"epoch": 2.6608187134502925,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046903816043204045,
|
|
"loss": 6.3677,
|
|
"mean_token_accuracy": 0.16216813772916794,
|
|
"num_tokens": 5668365.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.641722393035889,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004688004444743419,
|
|
"loss": 6.3445,
|
|
"mean_token_accuracy": 0.1614544212818146,
|
|
"num_tokens": 5680013.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.5340320587158205,
|
|
"epoch": 2.672514619883041,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000468561887408817,
|
|
"loss": 6.356,
|
|
"mean_token_accuracy": 0.1662822589278221,
|
|
"num_tokens": 5691884.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.6201173782348635,
|
|
"epoch": 2.678362573099415,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004683224902708206,
|
|
"loss": 6.4189,
|
|
"mean_token_accuracy": 0.155883726477623,
|
|
"num_tokens": 5704340.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.716411542892456,
|
|
"epoch": 2.6842105263157894,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004680822540993534,
|
|
"loss": 6.4739,
|
|
"mean_token_accuracy": 0.1568793460726738,
|
|
"num_tokens": 5717792.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.584109830856323,
|
|
"epoch": 2.690058479532164,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046784117993705753,
|
|
"loss": 6.3408,
|
|
"mean_token_accuracy": 0.1581699401140213,
|
|
"num_tokens": 5730237.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.569747686386108,
|
|
"epoch": 2.6959064327485383,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046759926883021234,
|
|
"loss": 6.3398,
|
|
"mean_token_accuracy": 0.16386526226997375,
|
|
"num_tokens": 5743953.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.524936771392822,
|
|
"epoch": 2.7017543859649122,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046735652182872925,
|
|
"loss": 6.3135,
|
|
"mean_token_accuracy": 0.15978483110666275,
|
|
"num_tokens": 5756148.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.598800230026245,
|
|
"epoch": 2.7076023391812867,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004671129399861476,
|
|
"loss": 6.3401,
|
|
"mean_token_accuracy": 0.16510675847530365,
|
|
"num_tokens": 5769163.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.624055099487305,
|
|
"epoch": 2.7134502923976607,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046686852435963016,
|
|
"loss": 6.3743,
|
|
"mean_token_accuracy": 0.158755099773407,
|
|
"num_tokens": 5781832.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.613272619247437,
|
|
"epoch": 2.719298245614035,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004666232760099583,
|
|
"loss": 6.3601,
|
|
"mean_token_accuracy": 0.1566229984164238,
|
|
"num_tokens": 5794803.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.5775844097137455,
|
|
"epoch": 2.7251461988304095,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046637719600152737,
|
|
"loss": 6.413,
|
|
"mean_token_accuracy": 0.15503155887126924,
|
|
"num_tokens": 5806768.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.711832046508789,
|
|
"epoch": 2.7309941520467835,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046613028540234226,
|
|
"loss": 6.4006,
|
|
"mean_token_accuracy": 0.1553715169429779,
|
|
"num_tokens": 5819113.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.615337562561035,
|
|
"epoch": 2.736842105263158,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046588254528401264,
|
|
"loss": 6.38,
|
|
"mean_token_accuracy": 0.1609787106513977,
|
|
"num_tokens": 5830067.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.585383749008178,
|
|
"epoch": 2.742690058479532,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004656339767217485,
|
|
"loss": 6.4119,
|
|
"mean_token_accuracy": 0.16197725534439086,
|
|
"num_tokens": 5842882.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.586494827270508,
|
|
"epoch": 2.7485380116959064,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046538458079435516,
|
|
"loss": 6.376,
|
|
"mean_token_accuracy": 0.1608525738120079,
|
|
"num_tokens": 5854041.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.59232006072998,
|
|
"epoch": 2.754385964912281,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004651343585842287,
|
|
"loss": 6.3945,
|
|
"mean_token_accuracy": 0.1616936817765236,
|
|
"num_tokens": 5866027.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.615852785110474,
|
|
"epoch": 2.760233918128655,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046488331117735176,
|
|
"loss": 6.4221,
|
|
"mean_token_accuracy": 0.159077088534832,
|
|
"num_tokens": 5877064.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.664672040939331,
|
|
"epoch": 2.7660818713450293,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004646314396632878,
|
|
"loss": 6.3829,
|
|
"mean_token_accuracy": 0.1577585145831108,
|
|
"num_tokens": 5889190.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.573435640335083,
|
|
"epoch": 2.7719298245614032,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004643787451351774,
|
|
"loss": 6.3555,
|
|
"mean_token_accuracy": 0.16325586289167404,
|
|
"num_tokens": 5901615.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.535305595397949,
|
|
"epoch": 2.7777777777777777,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046412522868973297,
|
|
"loss": 6.366,
|
|
"mean_token_accuracy": 0.16434068381786346,
|
|
"num_tokens": 5913797.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.705916213989258,
|
|
"epoch": 2.783625730994152,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046387089142723386,
|
|
"loss": 6.4472,
|
|
"mean_token_accuracy": 0.1538482964038849,
|
|
"num_tokens": 5927607.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.476776790618897,
|
|
"epoch": 2.7894736842105265,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00046361573445152224,
|
|
"loss": 6.2802,
|
|
"mean_token_accuracy": 0.1694086194038391,
|
|
"num_tokens": 5940254.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.487308406829834,
|
|
"epoch": 2.7953216374269005,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046335975886999754,
|
|
"loss": 6.1334,
|
|
"mean_token_accuracy": 0.18231945633888244,
|
|
"num_tokens": 5953341.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.451533412933349,
|
|
"epoch": 2.801169590643275,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046310296579361225,
|
|
"loss": 6.1586,
|
|
"mean_token_accuracy": 0.17581101804971694,
|
|
"num_tokens": 5965148.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.554260492324829,
|
|
"epoch": 2.807017543859649,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004628453563368666,
|
|
"loss": 6.4399,
|
|
"mean_token_accuracy": 0.15495024621486664,
|
|
"num_tokens": 5979143.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.657009410858154,
|
|
"epoch": 2.8128654970760234,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004625869316178043,
|
|
"loss": 6.3971,
|
|
"mean_token_accuracy": 0.16090115159749985,
|
|
"num_tokens": 5992735.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.577971267700195,
|
|
"epoch": 2.818713450292398,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000462327692758007,
|
|
"loss": 6.4575,
|
|
"mean_token_accuracy": 0.15103900134563447,
|
|
"num_tokens": 6005033.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.611282587051392,
|
|
"epoch": 2.824561403508772,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046206764088259003,
|
|
"loss": 6.3029,
|
|
"mean_token_accuracy": 0.1640724003314972,
|
|
"num_tokens": 6016746.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.5664506435394285,
|
|
"epoch": 2.8304093567251463,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00046180677712019707,
|
|
"loss": 6.4126,
|
|
"mean_token_accuracy": 0.16085123866796494,
|
|
"num_tokens": 6029774.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.5928404331207275,
|
|
"epoch": 2.8362573099415203,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004615451026029957,
|
|
"loss": 6.4392,
|
|
"mean_token_accuracy": 0.1511187493801117,
|
|
"num_tokens": 6043773.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.600831651687622,
|
|
"epoch": 2.8421052631578947,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000461282618466672,
|
|
"loss": 6.3484,
|
|
"mean_token_accuracy": 0.1549486830830574,
|
|
"num_tokens": 6055840.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.543839311599731,
|
|
"epoch": 2.847953216374269,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00046101932585042597,
|
|
"loss": 6.2775,
|
|
"mean_token_accuracy": 0.16883161664009094,
|
|
"num_tokens": 6069170.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.5523152351379395,
|
|
"epoch": 2.853801169590643,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046075522589696636,
|
|
"loss": 6.3542,
|
|
"mean_token_accuracy": 0.16195902973413467,
|
|
"num_tokens": 6080888.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.5835165023803714,
|
|
"epoch": 2.8596491228070176,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000460490319752506,
|
|
"loss": 6.2847,
|
|
"mean_token_accuracy": 0.16598695069551467,
|
|
"num_tokens": 6093035.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.537801218032837,
|
|
"epoch": 2.8654970760233915,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00046022460856675663,
|
|
"loss": 6.3605,
|
|
"mean_token_accuracy": 0.1592122033238411,
|
|
"num_tokens": 6105963.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.675387907028198,
|
|
"epoch": 2.871345029239766,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004599580934929237,
|
|
"loss": 6.3764,
|
|
"mean_token_accuracy": 0.1580013170838356,
|
|
"num_tokens": 6118541.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.545244836807251,
|
|
"epoch": 2.8771929824561404,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004596907756877018,
|
|
"loss": 6.3093,
|
|
"mean_token_accuracy": 0.15826527327299117,
|
|
"num_tokens": 6131317.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.548814535140991,
|
|
"epoch": 2.883040935672515,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045942265631126935,
|
|
"loss": 6.3647,
|
|
"mean_token_accuracy": 0.15769590437412262,
|
|
"num_tokens": 6143721.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.619016218185425,
|
|
"epoch": 2.888888888888889,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004591537365272837,
|
|
"loss": 6.3556,
|
|
"mean_token_accuracy": 0.15823782980442047,
|
|
"num_tokens": 6156497.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.520663642883301,
|
|
"epoch": 2.8947368421052633,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000458884017502876,
|
|
"loss": 6.3415,
|
|
"mean_token_accuracy": 0.16449064016342163,
|
|
"num_tokens": 6168140.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.606345558166504,
|
|
"epoch": 2.9005847953216373,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004586135004086461,
|
|
"loss": 6.328,
|
|
"mean_token_accuracy": 0.16131319850683212,
|
|
"num_tokens": 6181067.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.569672298431397,
|
|
"epoch": 2.9064327485380117,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004583421864186577,
|
|
"loss": 6.3114,
|
|
"mean_token_accuracy": 0.1567941665649414,
|
|
"num_tokens": 6194072.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.60381669998169,
|
|
"epoch": 2.912280701754386,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004580700767104328,
|
|
"loss": 6.3763,
|
|
"mean_token_accuracy": 0.15945695638656615,
|
|
"num_tokens": 6206711.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.483018159866333,
|
|
"epoch": 2.91812865497076,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045779717246494724,
|
|
"loss": 6.308,
|
|
"mean_token_accuracy": 0.15928689241409302,
|
|
"num_tokens": 6219903.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.5889068126678465,
|
|
"epoch": 2.9239766081871346,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004575234748666249,
|
|
"loss": 6.3549,
|
|
"mean_token_accuracy": 0.15934692174196244,
|
|
"num_tokens": 6232947.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.9239766081871346,
|
|
"eval_entropy": 6.326272354013448,
|
|
"eval_loss": 6.665432453155518,
|
|
"eval_mean_token_accuracy": 0.1448239707099521,
|
|
"eval_num_tokens": 6232947.0,
|
|
"eval_runtime": 1.7388,
|
|
"eval_samples_per_second": 1556.785,
|
|
"eval_steps_per_second": 194.958,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.545487737655639,
|
|
"epoch": 2.9298245614035086,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000457248985103333,
|
|
"loss": 6.3147,
|
|
"mean_token_accuracy": 0.16264024376869202,
|
|
"num_tokens": 6245597.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.515110015869141,
|
|
"epoch": 2.935672514619883,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004569737043663768,
|
|
"loss": 6.2636,
|
|
"mean_token_accuracy": 0.15790152698755264,
|
|
"num_tokens": 6257745.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.564759111404419,
|
|
"epoch": 2.9415204678362574,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004566976338504945,
|
|
"loss": 6.3523,
|
|
"mean_token_accuracy": 0.16369810551404954,
|
|
"num_tokens": 6268871.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.513104486465454,
|
|
"epoch": 2.9473684210526314,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045642077475385186,
|
|
"loss": 6.3137,
|
|
"mean_token_accuracy": 0.16343397349119188,
|
|
"num_tokens": 6281487.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.464796352386474,
|
|
"epoch": 2.953216374269006,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004561431282780372,
|
|
"loss": 6.3129,
|
|
"mean_token_accuracy": 0.16285736113786697,
|
|
"num_tokens": 6292928.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.517546510696411,
|
|
"epoch": 2.95906432748538,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045586469562805624,
|
|
"loss": 6.2631,
|
|
"mean_token_accuracy": 0.17902406454086303,
|
|
"num_tokens": 6304340.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.526446771621704,
|
|
"epoch": 2.9649122807017543,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004555854780123265,
|
|
"loss": 6.3417,
|
|
"mean_token_accuracy": 0.1617569476366043,
|
|
"num_tokens": 6315972.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.5971986770629885,
|
|
"epoch": 2.9707602339181287,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004553054766426725,
|
|
"loss": 6.3391,
|
|
"mean_token_accuracy": 0.16493984907865525,
|
|
"num_tokens": 6327482.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.499934244155884,
|
|
"epoch": 2.976608187134503,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004550246927343203,
|
|
"loss": 6.3281,
|
|
"mean_token_accuracy": 0.1609764203429222,
|
|
"num_tokens": 6338932.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.5534508228302,
|
|
"epoch": 2.982456140350877,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045474312750589226,
|
|
"loss": 6.3516,
|
|
"mean_token_accuracy": 0.16287137418985367,
|
|
"num_tokens": 6350770.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.566776990890503,
|
|
"epoch": 2.9883040935672516,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00045446078217940155,
|
|
"loss": 6.3704,
|
|
"mean_token_accuracy": 0.15735775828361512,
|
|
"num_tokens": 6363472.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.5520330429077145,
|
|
"epoch": 2.9941520467836256,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045417765798024716,
|
|
"loss": 6.3431,
|
|
"mean_token_accuracy": 0.15789056420326233,
|
|
"num_tokens": 6377174.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.522563076019287,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004538937561372086,
|
|
"loss": 6.3152,
|
|
"mean_token_accuracy": 0.15764351636171342,
|
|
"num_tokens": 6388452.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.6392518997192385,
|
|
"epoch": 3.0058479532163744,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004536090778824398,
|
|
"loss": 6.0416,
|
|
"mean_token_accuracy": 0.17430478781461717,
|
|
"num_tokens": 6401740.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.456202220916748,
|
|
"epoch": 3.0116959064327484,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004533236244514652,
|
|
"loss": 6.0128,
|
|
"mean_token_accuracy": 0.18396907448768615,
|
|
"num_tokens": 6414850.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.533134031295776,
|
|
"epoch": 3.017543859649123,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00045303739708317303,
|
|
"loss": 6.1569,
|
|
"mean_token_accuracy": 0.1593727260828018,
|
|
"num_tokens": 6429250.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.558267545700073,
|
|
"epoch": 3.023391812865497,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004527503970198105,
|
|
"loss": 5.9858,
|
|
"mean_token_accuracy": 0.1749659836292267,
|
|
"num_tokens": 6440583.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.430190944671631,
|
|
"epoch": 3.0292397660818713,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004524626255069786,
|
|
"loss": 5.9912,
|
|
"mean_token_accuracy": 0.17568861842155456,
|
|
"num_tokens": 6452678.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.538709592819214,
|
|
"epoch": 3.0350877192982457,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045217408379362635,
|
|
"loss": 6.0489,
|
|
"mean_token_accuracy": 0.1696217104792595,
|
|
"num_tokens": 6464751.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.4175807476043705,
|
|
"epoch": 3.0409356725146197,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00045188477313204536,
|
|
"loss": 6.0121,
|
|
"mean_token_accuracy": 0.1693053662776947,
|
|
"num_tokens": 6476017.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.589514923095703,
|
|
"epoch": 3.046783625730994,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045159469477786457,
|
|
"loss": 6.0376,
|
|
"mean_token_accuracy": 0.1720902919769287,
|
|
"num_tokens": 6487404.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.536309623718262,
|
|
"epoch": 3.0526315789473686,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004513038499900451,
|
|
"loss": 6.0948,
|
|
"mean_token_accuracy": 0.16542150229215621,
|
|
"num_tokens": 6500027.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.467177772521973,
|
|
"epoch": 3.0584795321637426,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000451012240030874,
|
|
"loss": 6.087,
|
|
"mean_token_accuracy": 0.16320331692695617,
|
|
"num_tokens": 6513067.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.534950160980225,
|
|
"epoch": 3.064327485380117,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045071986616595946,
|
|
"loss": 6.0231,
|
|
"mean_token_accuracy": 0.16768090277910233,
|
|
"num_tokens": 6524932.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.526103448867798,
|
|
"epoch": 3.0701754385964914,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045042672966422506,
|
|
"loss": 6.0608,
|
|
"mean_token_accuracy": 0.16547864377498628,
|
|
"num_tokens": 6536843.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.5405725002288815,
|
|
"epoch": 3.0760233918128654,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004501328317979041,
|
|
"loss": 5.9799,
|
|
"mean_token_accuracy": 0.17787163257598876,
|
|
"num_tokens": 6549414.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.437455129623413,
|
|
"epoch": 3.08187134502924,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004498381738425346,
|
|
"loss": 6.0621,
|
|
"mean_token_accuracy": 0.1710044726729393,
|
|
"num_tokens": 6561492.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.456179237365722,
|
|
"epoch": 3.087719298245614,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00044954275707695304,
|
|
"loss": 5.969,
|
|
"mean_token_accuracy": 0.1766466438770294,
|
|
"num_tokens": 6572873.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.4630388736724855,
|
|
"epoch": 3.0935672514619883,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004492465827832894,
|
|
"loss": 6.1049,
|
|
"mean_token_accuracy": 0.17341775745153426,
|
|
"num_tokens": 6584902.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.575921821594238,
|
|
"epoch": 3.0994152046783627,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044894965224696144,
|
|
"loss": 6.1188,
|
|
"mean_token_accuracy": 0.16425863206386565,
|
|
"num_tokens": 6596792.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.493634605407715,
|
|
"epoch": 3.1052631578947367,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044865196675666886,
|
|
"loss": 6.052,
|
|
"mean_token_accuracy": 0.17168822437524794,
|
|
"num_tokens": 6608559.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.443542098999023,
|
|
"epoch": 3.111111111111111,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004483535276043881,
|
|
"loss": 6.0973,
|
|
"mean_token_accuracy": 0.16921755522489548,
|
|
"num_tokens": 6619938.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.5081884384155275,
|
|
"epoch": 3.116959064327485,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004480543360853665,
|
|
"loss": 6.0601,
|
|
"mean_token_accuracy": 0.17169931828975676,
|
|
"num_tokens": 6632850.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.390790128707886,
|
|
"epoch": 3.1228070175438596,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004477543934981167,
|
|
"loss": 5.9842,
|
|
"mean_token_accuracy": 0.18593875020742417,
|
|
"num_tokens": 6645761.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.481044292449951,
|
|
"epoch": 3.128654970760234,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044745370114441103,
|
|
"loss": 6.0533,
|
|
"mean_token_accuracy": 0.17378513216972352,
|
|
"num_tokens": 6657861.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.475371408462524,
|
|
"epoch": 3.134502923976608,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044715226032927585,
|
|
"loss": 6.0302,
|
|
"mean_token_accuracy": 0.1815857082605362,
|
|
"num_tokens": 6670495.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.2983275890350345,
|
|
"epoch": 3.1403508771929824,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004468500723609859,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.17493653446435928,
|
|
"num_tokens": 6683127.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.53174319267273,
|
|
"epoch": 3.146198830409357,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044654713855105864,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.16416590958833693,
|
|
"num_tokens": 6696159.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.477815103530884,
|
|
"epoch": 3.152046783625731,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004462434602142486,
|
|
"loss": 6.1359,
|
|
"mean_token_accuracy": 0.16184422969818116,
|
|
"num_tokens": 6708993.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.485203790664673,
|
|
"epoch": 3.1578947368421053,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004459390386685414,
|
|
"loss": 6.1362,
|
|
"mean_token_accuracy": 0.1668482318520546,
|
|
"num_tokens": 6722121.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.446034336090088,
|
|
"epoch": 3.1637426900584797,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00044563387523514854,
|
|
"loss": 5.9641,
|
|
"mean_token_accuracy": 0.17729679346084595,
|
|
"num_tokens": 6734167.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.373656702041626,
|
|
"epoch": 3.1695906432748537,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004453279712385011,
|
|
"loss": 6.054,
|
|
"mean_token_accuracy": 0.16502795666456221,
|
|
"num_tokens": 6745720.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.46063084602356,
|
|
"epoch": 3.175438596491228,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004450213280062444,
|
|
"loss": 6.0408,
|
|
"mean_token_accuracy": 0.1700156345963478,
|
|
"num_tokens": 6757523.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.50075249671936,
|
|
"epoch": 3.181286549707602,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044471394686923206,
|
|
"loss": 6.0614,
|
|
"mean_token_accuracy": 0.16797684282064437,
|
|
"num_tokens": 6768775.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.417105579376221,
|
|
"epoch": 3.1871345029239766,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004444058291615202,
|
|
"loss": 6.0529,
|
|
"mean_token_accuracy": 0.16829179525375365,
|
|
"num_tokens": 6781547.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.470510053634643,
|
|
"epoch": 3.192982456140351,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004440969762203618,
|
|
"loss": 6.1053,
|
|
"mean_token_accuracy": 0.1649450898170471,
|
|
"num_tokens": 6794708.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.408028221130371,
|
|
"epoch": 3.198830409356725,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004437873893862008,
|
|
"loss": 6.0263,
|
|
"mean_token_accuracy": 0.1764855980873108,
|
|
"num_tokens": 6807738.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.468166160583496,
|
|
"epoch": 3.2046783625730995,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044347707000266627,
|
|
"loss": 6.0724,
|
|
"mean_token_accuracy": 0.16998974829912186,
|
|
"num_tokens": 6820851.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.464267206192017,
|
|
"epoch": 3.2105263157894735,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00044316601941656655,
|
|
"loss": 6.0391,
|
|
"mean_token_accuracy": 0.18092983961105347,
|
|
"num_tokens": 6833888.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.4288331985473635,
|
|
"epoch": 3.216374269005848,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004428542389778834,
|
|
"loss": 6.0685,
|
|
"mean_token_accuracy": 0.17276390939950942,
|
|
"num_tokens": 6846104.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.459614324569702,
|
|
"epoch": 3.2222222222222223,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004425417300397665,
|
|
"loss": 6.0536,
|
|
"mean_token_accuracy": 0.17244808971881867,
|
|
"num_tokens": 6857458.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.400099849700927,
|
|
"epoch": 3.2280701754385963,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044222849395852683,
|
|
"loss": 6.0137,
|
|
"mean_token_accuracy": 0.17610430866479873,
|
|
"num_tokens": 6870455.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.447430515289307,
|
|
"epoch": 3.2339181286549707,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00044191453209363165,
|
|
"loss": 6.0807,
|
|
"mean_token_accuracy": 0.1680549591779709,
|
|
"num_tokens": 6883359.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.424415588378906,
|
|
"epoch": 3.239766081871345,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004415998458076978,
|
|
"loss": 6.0234,
|
|
"mean_token_accuracy": 0.1768151640892029,
|
|
"num_tokens": 6894940.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.46516523361206,
|
|
"epoch": 3.245614035087719,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044128443646648634,
|
|
"loss": 6.0762,
|
|
"mean_token_accuracy": 0.17116446942090988,
|
|
"num_tokens": 6905422.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.469412517547608,
|
|
"epoch": 3.2514619883040936,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044096830543889653,
|
|
"loss": 6.0779,
|
|
"mean_token_accuracy": 0.17375174462795256,
|
|
"num_tokens": 6918106.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.464608812332154,
|
|
"epoch": 3.257309941520468,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004406514540969596,
|
|
"loss": 6.0504,
|
|
"mean_token_accuracy": 0.17326418459415435,
|
|
"num_tokens": 6930975.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.496636581420899,
|
|
"epoch": 3.263157894736842,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044033388381583324,
|
|
"loss": 6.13,
|
|
"mean_token_accuracy": 0.15955014824867247,
|
|
"num_tokens": 6944436.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.482188129425049,
|
|
"epoch": 3.2690058479532165,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00044001559597379503,
|
|
"loss": 6.1077,
|
|
"mean_token_accuracy": 0.16621807664632798,
|
|
"num_tokens": 6957095.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.470199632644653,
|
|
"epoch": 3.2748538011695905,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004396965919522373,
|
|
"loss": 6.0339,
|
|
"mean_token_accuracy": 0.1778212830424309,
|
|
"num_tokens": 6969792.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.383312082290649,
|
|
"epoch": 3.280701754385965,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00043937687313565997,
|
|
"loss": 6.0942,
|
|
"mean_token_accuracy": 0.16948736160993577,
|
|
"num_tokens": 6982054.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.463279914855957,
|
|
"epoch": 3.2865497076023393,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004390564409116661,
|
|
"loss": 6.0835,
|
|
"mean_token_accuracy": 0.16666572242975236,
|
|
"num_tokens": 6994057.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.420546960830689,
|
|
"epoch": 3.2923976608187133,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.00043873529667095423,
|
|
"loss": 6.077,
|
|
"mean_token_accuracy": 0.17153801918029785,
|
|
"num_tokens": 7006417.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.456496238708496,
|
|
"epoch": 3.2982456140350878,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043841344180731346,
|
|
"loss": 6.1032,
|
|
"mean_token_accuracy": 0.16475788950920106,
|
|
"num_tokens": 7018331.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.443552398681641,
|
|
"epoch": 3.3040935672514617,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000438090877717617,
|
|
"loss": 5.9781,
|
|
"mean_token_accuracy": 0.17624930739402772,
|
|
"num_tokens": 7031654.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.506548070907593,
|
|
"epoch": 3.309941520467836,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004377676058018161,
|
|
"loss": 6.0848,
|
|
"mean_token_accuracy": 0.17138757109642028,
|
|
"num_tokens": 7044921.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.386026239395141,
|
|
"epoch": 3.3157894736842106,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004374436274629341,
|
|
"loss": 6.0196,
|
|
"mean_token_accuracy": 0.16961200386285782,
|
|
"num_tokens": 7056700.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.414784955978393,
|
|
"epoch": 3.3216374269005846,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004371189441070602,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.17018587440252303,
|
|
"num_tokens": 7069196.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.381654357910156,
|
|
"epoch": 3.327485380116959,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004367935571433433,
|
|
"loss": 6.0108,
|
|
"mean_token_accuracy": 0.17265803664922713,
|
|
"num_tokens": 7081528.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.482018709182739,
|
|
"epoch": 3.3333333333333335,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004364674679839864,
|
|
"loss": 6.1139,
|
|
"mean_token_accuracy": 0.1615537166595459,
|
|
"num_tokens": 7093474.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.332666635513306,
|
|
"epoch": 3.3391812865497075,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004361406780442398,
|
|
"loss": 6.0373,
|
|
"mean_token_accuracy": 0.1735363021492958,
|
|
"num_tokens": 7105508.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.403873348236084,
|
|
"epoch": 3.345029239766082,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00043581318874239504,
|
|
"loss": 6.0455,
|
|
"mean_token_accuracy": 0.17201557606458664,
|
|
"num_tokens": 7117952.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.40864930152893,
|
|
"epoch": 3.3508771929824563,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043548500149977946,
|
|
"loss": 6.0206,
|
|
"mean_token_accuracy": 0.17704936861991882,
|
|
"num_tokens": 7130666.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.416865301132202,
|
|
"epoch": 3.3567251461988303,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004351561177407491,
|
|
"loss": 6.0587,
|
|
"mean_token_accuracy": 0.17262544333934784,
|
|
"num_tokens": 7142717.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.4950133800506595,
|
|
"epoch": 3.3625730994152048,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004348265388926831,
|
|
"loss": 6.1161,
|
|
"mean_token_accuracy": 0.1695236951112747,
|
|
"num_tokens": 7154794.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.442908000946045,
|
|
"epoch": 3.3684210526315788,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004344962663859771,
|
|
"loss": 6.1571,
|
|
"mean_token_accuracy": 0.16535353511571885,
|
|
"num_tokens": 7168907.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.3863331317901615,
|
|
"epoch": 3.374269005847953,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043416530165403777,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.18584391623735427,
|
|
"num_tokens": 7180828.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.311170339584351,
|
|
"epoch": 3.3801169590643276,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004338336461332756,
|
|
"loss": 5.9883,
|
|
"mean_token_accuracy": 0.17260439395904542,
|
|
"num_tokens": 7193493.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.4398633480072025,
|
|
"epoch": 3.3859649122807016,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00043350130126309936,
|
|
"loss": 6.0268,
|
|
"mean_token_accuracy": 0.16570234894752503,
|
|
"num_tokens": 7206033.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.439795255661011,
|
|
"epoch": 3.391812865497076,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00043316826848590955,
|
|
"loss": 6.0643,
|
|
"mean_token_accuracy": 0.16937600821256638,
|
|
"num_tokens": 7218813.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.402001762390137,
|
|
"epoch": 3.39766081871345,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004328345492470924,
|
|
"loss": 6.0442,
|
|
"mean_token_accuracy": 0.177463199198246,
|
|
"num_tokens": 7230698.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.417586994171143,
|
|
"epoch": 3.4035087719298245,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004325001449950134,
|
|
"loss": 6.0687,
|
|
"mean_token_accuracy": 0.1662218913435936,
|
|
"num_tokens": 7243354.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.365819215774536,
|
|
"epoch": 3.409356725146199,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004321650571810109,
|
|
"loss": 6.0137,
|
|
"mean_token_accuracy": 0.18044356554746627,
|
|
"num_tokens": 7255855.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.392031860351563,
|
|
"epoch": 3.415204678362573,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004318292872593899,
|
|
"loss": 6.111,
|
|
"mean_token_accuracy": 0.17002128064632416,
|
|
"num_tokens": 7268921.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.4622111320495605,
|
|
"epoch": 3.4210526315789473,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004314928366874162,
|
|
"loss": 6.0869,
|
|
"mean_token_accuracy": 0.16710357069969178,
|
|
"num_tokens": 7281467.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.354787969589234,
|
|
"epoch": 3.426900584795322,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004311557069253092,
|
|
"loss": 6.0173,
|
|
"mean_token_accuracy": 0.17542969286441804,
|
|
"num_tokens": 7294887.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.502463483810425,
|
|
"epoch": 3.4327485380116958,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004308178994362364,
|
|
"loss": 6.1314,
|
|
"mean_token_accuracy": 0.17358484417200087,
|
|
"num_tokens": 7307009.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.373237991333008,
|
|
"epoch": 3.43859649122807,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00043047941568630644,
|
|
"loss": 6.0509,
|
|
"mean_token_accuracy": 0.16770925521850585,
|
|
"num_tokens": 7319751.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.4928570747375485,
|
|
"epoch": 3.4444444444444446,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00043014025714456305,
|
|
"loss": 6.171,
|
|
"mean_token_accuracy": 0.16604739129543306,
|
|
"num_tokens": 7333952.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.37247142791748,
|
|
"epoch": 3.4502923976608186,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004298004252829787,
|
|
"loss": 5.9403,
|
|
"mean_token_accuracy": 0.18072114288806915,
|
|
"num_tokens": 7346834.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.4232381820678714,
|
|
"epoch": 3.456140350877193,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000429459921576448,
|
|
"loss": 6.0777,
|
|
"mean_token_accuracy": 0.16881246417760848,
|
|
"num_tokens": 7359165.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.427524900436401,
|
|
"epoch": 3.461988304093567,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00042911874750278133,
|
|
"loss": 6.1442,
|
|
"mean_token_accuracy": 0.16581834703683854,
|
|
"num_tokens": 7371585.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.406796836853028,
|
|
"epoch": 3.4678362573099415,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042877690454269884,
|
|
"loss": 6.0647,
|
|
"mean_token_accuracy": 0.1748057186603546,
|
|
"num_tokens": 7383455.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.393041896820068,
|
|
"epoch": 3.473684210526316,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00042843439417982344,
|
|
"loss": 6.1054,
|
|
"mean_token_accuracy": 0.16994200497865677,
|
|
"num_tokens": 7395918.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.446899175643921,
|
|
"epoch": 3.47953216374269,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004280912179006746,
|
|
"loss": 6.0891,
|
|
"mean_token_accuracy": 0.1722549319267273,
|
|
"num_tokens": 7408053.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.460547208786011,
|
|
"epoch": 3.4853801169590644,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000427747377194662,
|
|
"loss": 6.1294,
|
|
"mean_token_accuracy": 0.1644581973552704,
|
|
"num_tokens": 7419947.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.36607403755188,
|
|
"epoch": 3.4912280701754383,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004274028735540789,
|
|
"loss": 6.0274,
|
|
"mean_token_accuracy": 0.16511478424072265,
|
|
"num_tokens": 7433022.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.352493619918823,
|
|
"epoch": 3.497076023391813,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042705770847409596,
|
|
"loss": 6.0171,
|
|
"mean_token_accuracy": 0.17755896002054214,
|
|
"num_tokens": 7445280.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.488861083984375,
|
|
"epoch": 3.502923976608187,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042671188345275423,
|
|
"loss": 6.0134,
|
|
"mean_token_accuracy": 0.1712557315826416,
|
|
"num_tokens": 7457324.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.396337080001831,
|
|
"epoch": 3.5087719298245617,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004263653999909593,
|
|
"loss": 6.0548,
|
|
"mean_token_accuracy": 0.16896977573633193,
|
|
"num_tokens": 7469070.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.5087719298245617,
|
|
"eval_entropy": 6.2064107835820295,
|
|
"eval_loss": 6.635471343994141,
|
|
"eval_mean_token_accuracy": 0.14681223468926452,
|
|
"eval_num_tokens": 7469070.0,
|
|
"eval_runtime": 1.7349,
|
|
"eval_samples_per_second": 1560.33,
|
|
"eval_steps_per_second": 195.402,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.398986387252807,
|
|
"epoch": 3.5146198830409356,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000426018259592474,
|
|
"loss": 6.1174,
|
|
"mean_token_accuracy": 0.16569647639989854,
|
|
"num_tokens": 7480988.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.42080659866333,
|
|
"epoch": 3.52046783625731,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004256704637639126,
|
|
"loss": 6.0295,
|
|
"mean_token_accuracy": 0.17857254147529603,
|
|
"num_tokens": 7493557.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.3781335830688475,
|
|
"epoch": 3.526315789473684,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004253220140147341,
|
|
"loss": 6.0832,
|
|
"mean_token_accuracy": 0.16816118657588958,
|
|
"num_tokens": 7507446.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.47371096611023,
|
|
"epoch": 3.5321637426900585,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004249729118572352,
|
|
"loss": 6.0245,
|
|
"mean_token_accuracy": 0.16584294736385347,
|
|
"num_tokens": 7520054.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.332010269165039,
|
|
"epoch": 3.538011695906433,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004246231588065444,
|
|
"loss": 5.9172,
|
|
"mean_token_accuracy": 0.1832406997680664,
|
|
"num_tokens": 7533141.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.397120428085327,
|
|
"epoch": 3.543859649122807,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000424272756380615,
|
|
"loss": 6.0855,
|
|
"mean_token_accuracy": 0.16364741027355195,
|
|
"num_tokens": 7545544.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.381061029434204,
|
|
"epoch": 3.5497076023391814,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004239217061002188,
|
|
"loss": 6.0672,
|
|
"mean_token_accuracy": 0.17535382956266404,
|
|
"num_tokens": 7558055.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.39494833946228,
|
|
"epoch": 3.5555555555555554,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00042357000948893894,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.1831562414765358,
|
|
"num_tokens": 7569951.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.3724018096923825,
|
|
"epoch": 3.56140350877193,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00042321766807316405,
|
|
"loss": 6.1251,
|
|
"mean_token_accuracy": 0.1649981915950775,
|
|
"num_tokens": 7581810.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.420140409469605,
|
|
"epoch": 3.5672514619883042,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00042286468338208113,
|
|
"loss": 6.0753,
|
|
"mean_token_accuracy": 0.17272226363420487,
|
|
"num_tokens": 7594896.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.406652307510376,
|
|
"epoch": 3.573099415204678,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004225110569476691,
|
|
"loss": 6.0969,
|
|
"mean_token_accuracy": 0.17346913516521453,
|
|
"num_tokens": 7606620.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.322535753250122,
|
|
"epoch": 3.5789473684210527,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000422156790304692,
|
|
"loss": 5.9476,
|
|
"mean_token_accuracy": 0.18171590715646743,
|
|
"num_tokens": 7619493.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.3500754833221436,
|
|
"epoch": 3.5847953216374266,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00042180188499069246,
|
|
"loss": 5.9981,
|
|
"mean_token_accuracy": 0.16691530644893646,
|
|
"num_tokens": 7632711.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.413538789749145,
|
|
"epoch": 3.590643274853801,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000421446342545985,
|
|
"loss": 6.0439,
|
|
"mean_token_accuracy": 0.17270101308822633,
|
|
"num_tokens": 7645209.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.45819730758667,
|
|
"epoch": 3.5964912280701755,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042109016451364933,
|
|
"loss": 6.1397,
|
|
"mean_token_accuracy": 0.16349208652973174,
|
|
"num_tokens": 7657347.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.54139814376831,
|
|
"epoch": 3.60233918128655,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00042073335243952354,
|
|
"loss": 6.0768,
|
|
"mean_token_accuracy": 0.16341819763183593,
|
|
"num_tokens": 7669593.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.355091190338134,
|
|
"epoch": 3.608187134502924,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004203759078721979,
|
|
"loss": 6.1021,
|
|
"mean_token_accuracy": 0.1635132610797882,
|
|
"num_tokens": 7682548.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.436393737792969,
|
|
"epoch": 3.6140350877192984,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004200178323630072,
|
|
"loss": 6.1074,
|
|
"mean_token_accuracy": 0.1676407665014267,
|
|
"num_tokens": 7694554.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.421990251541137,
|
|
"epoch": 3.6198830409356724,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00041965912746602495,
|
|
"loss": 6.1379,
|
|
"mean_token_accuracy": 0.16801537424325944,
|
|
"num_tokens": 7707747.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.453145360946655,
|
|
"epoch": 3.625730994152047,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00041929979473805606,
|
|
"loss": 6.1199,
|
|
"mean_token_accuracy": 0.17281144857406616,
|
|
"num_tokens": 7719487.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.456283950805664,
|
|
"epoch": 3.6315789473684212,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041893983573863044,
|
|
"loss": 6.1554,
|
|
"mean_token_accuracy": 0.16115709990262986,
|
|
"num_tokens": 7732531.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.39797158241272,
|
|
"epoch": 3.6374269005847952,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041857925202999586,
|
|
"loss": 6.0621,
|
|
"mean_token_accuracy": 0.17226134985685349,
|
|
"num_tokens": 7745034.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.433007383346558,
|
|
"epoch": 3.6432748538011697,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004182180451771115,
|
|
"loss": 6.07,
|
|
"mean_token_accuracy": 0.1699773356318474,
|
|
"num_tokens": 7757442.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.349019289016724,
|
|
"epoch": 3.6491228070175437,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004178562167476409,
|
|
"loss": 6.0827,
|
|
"mean_token_accuracy": 0.1698372945189476,
|
|
"num_tokens": 7770358.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.38358063697815,
|
|
"epoch": 3.654970760233918,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00041749376831194565,
|
|
"loss": 6.0655,
|
|
"mean_token_accuracy": 0.169904862344265,
|
|
"num_tokens": 7783611.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.38002724647522,
|
|
"epoch": 3.6608187134502925,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004171307014430777,
|
|
"loss": 6.0765,
|
|
"mean_token_accuracy": 0.16511097252368928,
|
|
"num_tokens": 7795522.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.484520769119262,
|
|
"epoch": 3.6666666666666665,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041676701771677343,
|
|
"loss": 6.0449,
|
|
"mean_token_accuracy": 0.17537835836410523,
|
|
"num_tokens": 7806801.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.384499883651733,
|
|
"epoch": 3.672514619883041,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004164027187114463,
|
|
"loss": 6.1404,
|
|
"mean_token_accuracy": 0.16712912172079086,
|
|
"num_tokens": 7821084.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.256851482391357,
|
|
"epoch": 3.678362573099415,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00041603780600818,
|
|
"loss": 5.8681,
|
|
"mean_token_accuracy": 0.19157647639513015,
|
|
"num_tokens": 7834065.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.389277458190918,
|
|
"epoch": 3.6842105263157894,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004156722811907219,
|
|
"loss": 6.0556,
|
|
"mean_token_accuracy": 0.17556868642568588,
|
|
"num_tokens": 7846378.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.368703365325928,
|
|
"epoch": 3.690058479532164,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041530614584547594,
|
|
"loss": 6.0881,
|
|
"mean_token_accuracy": 0.16810568422079086,
|
|
"num_tokens": 7858933.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.32564697265625,
|
|
"epoch": 3.6959064327485383,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00041493940156149577,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.17576699554920197,
|
|
"num_tokens": 7870408.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.478543758392334,
|
|
"epoch": 3.7017543859649122,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000414572049930478,
|
|
"loss": 6.0911,
|
|
"mean_token_accuracy": 0.16700346469879152,
|
|
"num_tokens": 7882778.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.3552347183227536,
|
|
"epoch": 3.7076023391812867,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041420409254675465,
|
|
"loss": 6.0638,
|
|
"mean_token_accuracy": 0.17098980247974396,
|
|
"num_tokens": 7895169.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.384609889984131,
|
|
"epoch": 3.7134502923976607,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004138355310072876,
|
|
"loss": 6.1144,
|
|
"mean_token_accuracy": 0.16919229030609131,
|
|
"num_tokens": 7908607.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.426222276687622,
|
|
"epoch": 3.719298245614035,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00041346636691166015,
|
|
"loss": 6.0718,
|
|
"mean_token_accuracy": 0.1694900795817375,
|
|
"num_tokens": 7922269.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.350055170059204,
|
|
"epoch": 3.7251461988304095,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00041309660186207093,
|
|
"loss": 6.0173,
|
|
"mean_token_accuracy": 0.174677112698555,
|
|
"num_tokens": 7935205.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.380976772308349,
|
|
"epoch": 3.7309941520467835,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041272623746332687,
|
|
"loss": 6.0703,
|
|
"mean_token_accuracy": 0.16969146579504013,
|
|
"num_tokens": 7947970.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.266904258728028,
|
|
"epoch": 3.736842105263158,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00041235527532283603,
|
|
"loss": 5.9171,
|
|
"mean_token_accuracy": 0.18555708527565,
|
|
"num_tokens": 7961978.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.421753978729248,
|
|
"epoch": 3.742690058479532,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004119837170506005,
|
|
"loss": 6.0917,
|
|
"mean_token_accuracy": 0.17158013582229614,
|
|
"num_tokens": 7974708.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.390347623825074,
|
|
"epoch": 3.7485380116959064,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041161156425921004,
|
|
"loss": 6.0666,
|
|
"mean_token_accuracy": 0.1749900311231613,
|
|
"num_tokens": 7986179.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.284877920150757,
|
|
"epoch": 3.754385964912281,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00041123881856383436,
|
|
"loss": 6.0946,
|
|
"mean_token_accuracy": 0.17346256375312805,
|
|
"num_tokens": 7997816.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.431673383712768,
|
|
"epoch": 3.760233918128655,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004108654815822165,
|
|
"loss": 6.022,
|
|
"mean_token_accuracy": 0.17136942446231843,
|
|
"num_tokens": 8009407.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.396068716049195,
|
|
"epoch": 3.7660818713450293,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004104915549346658,
|
|
"loss": 6.0956,
|
|
"mean_token_accuracy": 0.16537972092628478,
|
|
"num_tokens": 8021446.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.387154531478882,
|
|
"epoch": 3.7719298245614032,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041011704024405065,
|
|
"loss": 6.1229,
|
|
"mean_token_accuracy": 0.16637109518051146,
|
|
"num_tokens": 8035219.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.423195838928223,
|
|
"epoch": 3.7777777777777777,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00040974193913579174,
|
|
"loss": 5.9223,
|
|
"mean_token_accuracy": 0.18564422577619552,
|
|
"num_tokens": 8047619.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.421533870697021,
|
|
"epoch": 3.783625730994152,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004093662532378547,
|
|
"loss": 6.1346,
|
|
"mean_token_accuracy": 0.1697626069188118,
|
|
"num_tokens": 8059241.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.384479284286499,
|
|
"epoch": 3.7894736842105265,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004089899841807433,
|
|
"loss": 6.0698,
|
|
"mean_token_accuracy": 0.1731516972184181,
|
|
"num_tokens": 8071543.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.374402332305908,
|
|
"epoch": 3.7953216374269005,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00040861313359749217,
|
|
"loss": 6.1187,
|
|
"mean_token_accuracy": 0.16990459859371185,
|
|
"num_tokens": 8083515.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.343055152893067,
|
|
"epoch": 3.801169590643275,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00040823570312366005,
|
|
"loss": 6.0495,
|
|
"mean_token_accuracy": 0.17494190484285355,
|
|
"num_tokens": 8095703.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.433814764022827,
|
|
"epoch": 3.807017543859649,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000407857694397322,
|
|
"loss": 6.1564,
|
|
"mean_token_accuracy": 0.16517484486103057,
|
|
"num_tokens": 8108285.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.34254789352417,
|
|
"epoch": 3.8128654970760234,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004074791090590632,
|
|
"loss": 6.0688,
|
|
"mean_token_accuracy": 0.16645632386207582,
|
|
"num_tokens": 8120943.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.397362470626831,
|
|
"epoch": 3.818713450292398,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004070999487519711,
|
|
"loss": 6.0986,
|
|
"mean_token_accuracy": 0.16318689584732055,
|
|
"num_tokens": 8133086.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.42890362739563,
|
|
"epoch": 3.824561403508772,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00040672021512162864,
|
|
"loss": 6.1308,
|
|
"mean_token_accuracy": 0.1646733433008194,
|
|
"num_tokens": 8147320.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.449703645706177,
|
|
"epoch": 3.8304093567251463,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000406339909816107,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.17072792649269103,
|
|
"num_tokens": 8160796.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.462924432754517,
|
|
"epoch": 3.8362573099415203,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004059590344859584,
|
|
"loss": 6.0435,
|
|
"mean_token_accuracy": 0.17428453266620636,
|
|
"num_tokens": 8173087.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.279750156402588,
|
|
"epoch": 3.8421052631578947,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004055775907842092,
|
|
"loss": 6.0096,
|
|
"mean_token_accuracy": 0.17417635917663574,
|
|
"num_tokens": 8185274.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.343102073669433,
|
|
"epoch": 3.847953216374269,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004051955803663523,
|
|
"loss": 6.0559,
|
|
"mean_token_accuracy": 0.17414364367723464,
|
|
"num_tokens": 8197753.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.383964681625367,
|
|
"epoch": 3.853801169590643,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004048130048903404,
|
|
"loss": 6.0632,
|
|
"mean_token_accuracy": 0.1722998633980751,
|
|
"num_tokens": 8211882.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.261384916305542,
|
|
"epoch": 3.8596491228070176,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004044298660165786,
|
|
"loss": 5.9382,
|
|
"mean_token_accuracy": 0.18322304189205169,
|
|
"num_tokens": 8225089.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.359839296340942,
|
|
"epoch": 3.8654970760233915,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040404616540791683,
|
|
"loss": 6.1141,
|
|
"mean_token_accuracy": 0.16743968725204467,
|
|
"num_tokens": 8238808.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.338085174560547,
|
|
"epoch": 3.871345029239766,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004036619047296435,
|
|
"loss": 6.0314,
|
|
"mean_token_accuracy": 0.17654839158058167,
|
|
"num_tokens": 8252482.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.352648687362671,
|
|
"epoch": 3.8771929824561404,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004032770856494772,
|
|
"loss": 6.0906,
|
|
"mean_token_accuracy": 0.16877461969852448,
|
|
"num_tokens": 8264232.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.369353914260865,
|
|
"epoch": 3.883040935672515,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00040289170983756075,
|
|
"loss": 6.0465,
|
|
"mean_token_accuracy": 0.1727171167731285,
|
|
"num_tokens": 8276784.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.398533916473388,
|
|
"epoch": 3.888888888888889,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004025057789664524,
|
|
"loss": 6.006,
|
|
"mean_token_accuracy": 0.17280499637126923,
|
|
"num_tokens": 8287887.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.223586702346802,
|
|
"epoch": 3.8947368421052633,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004021192947111201,
|
|
"loss": 5.865,
|
|
"mean_token_accuracy": 0.1852436348795891,
|
|
"num_tokens": 8301258.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.346120357513428,
|
|
"epoch": 3.9005847953216373,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00040173225874893305,
|
|
"loss": 6.0773,
|
|
"mean_token_accuracy": 0.17311538755893707,
|
|
"num_tokens": 8313520.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.514972925186157,
|
|
"epoch": 3.9064327485380117,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040134467275965515,
|
|
"loss": 6.1338,
|
|
"mean_token_accuracy": 0.16871004402637482,
|
|
"num_tokens": 8325597.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.4392743587493895,
|
|
"epoch": 3.912280701754386,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004009565384254374,
|
|
"loss": 6.1526,
|
|
"mean_token_accuracy": 0.16689565628767014,
|
|
"num_tokens": 8337449.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.369372987747193,
|
|
"epoch": 3.91812865497076,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040056785743081065,
|
|
"loss": 6.059,
|
|
"mean_token_accuracy": 0.1730581060051918,
|
|
"num_tokens": 8348802.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.340158748626709,
|
|
"epoch": 3.9239766081871346,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004001786314626783,
|
|
"loss": 6.0524,
|
|
"mean_token_accuracy": 0.16562539637088775,
|
|
"num_tokens": 8360284.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.391693639755249,
|
|
"epoch": 3.9298245614035086,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003997888622103088,
|
|
"loss": 6.0944,
|
|
"mean_token_accuracy": 0.16789348274469376,
|
|
"num_tokens": 8372080.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.392306470870972,
|
|
"epoch": 3.935672514619883,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003993985513653289,
|
|
"loss": 6.0402,
|
|
"mean_token_accuracy": 0.17357247173786164,
|
|
"num_tokens": 8385178.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.336736726760864,
|
|
"epoch": 3.9415204678362574,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003990077006217155,
|
|
"loss": 6.0276,
|
|
"mean_token_accuracy": 0.17882578521966935,
|
|
"num_tokens": 8397735.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.3677750587463375,
|
|
"epoch": 3.9473684210526314,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000398616311675789,
|
|
"loss": 6.1194,
|
|
"mean_token_accuracy": 0.16743769943714143,
|
|
"num_tokens": 8409916.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.432484912872314,
|
|
"epoch": 3.953216374269006,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039822438622620523,
|
|
"loss": 6.1176,
|
|
"mean_token_accuracy": 0.16798022091388704,
|
|
"num_tokens": 8421883.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.296114730834961,
|
|
"epoch": 3.95906432748538,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039783192597394896,
|
|
"loss": 5.9797,
|
|
"mean_token_accuracy": 0.18106727451086044,
|
|
"num_tokens": 8433610.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.356964111328125,
|
|
"epoch": 3.9649122807017543,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039743893262232584,
|
|
"loss": 6.055,
|
|
"mean_token_accuracy": 0.1702302724123001,
|
|
"num_tokens": 8446194.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.383318519592285,
|
|
"epoch": 3.9707602339181287,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003970454078769551,
|
|
"loss": 6.0254,
|
|
"mean_token_accuracy": 0.18046456724405288,
|
|
"num_tokens": 8458846.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.36809811592102,
|
|
"epoch": 3.976608187134503,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003966513534457623,
|
|
"loss": 6.0289,
|
|
"mean_token_accuracy": 0.16990351378917695,
|
|
"num_tokens": 8470087.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.360463333129883,
|
|
"epoch": 3.982456140350877,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003962567710389721,
|
|
"loss": 6.0587,
|
|
"mean_token_accuracy": 0.16591945588588713,
|
|
"num_tokens": 8481293.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.424599599838257,
|
|
"epoch": 3.9883040935672516,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039586166236910033,
|
|
"loss": 6.0746,
|
|
"mean_token_accuracy": 0.17950529456138611,
|
|
"num_tokens": 8493376.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.2540582656860355,
|
|
"epoch": 3.9941520467836256,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003954660291509469,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.17669039964675903,
|
|
"num_tokens": 8506132.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.374604606628418,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039506987310158854,
|
|
"loss": 6.0642,
|
|
"mean_token_accuracy": 0.16954705715179444,
|
|
"num_tokens": 8517936.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.440650177001953,
|
|
"epoch": 4.005847953216374,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039467319594037076,
|
|
"loss": 5.7523,
|
|
"mean_token_accuracy": 0.18199204057455062,
|
|
"num_tokens": 8529863.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.287904930114746,
|
|
"epoch": 4.011695906432749,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003942759993889009,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.18134903758764268,
|
|
"num_tokens": 8543621.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.32343544960022,
|
|
"epoch": 4.017543859649122,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00039387828517104044,
|
|
"loss": 5.747,
|
|
"mean_token_accuracy": 0.18953610509634017,
|
|
"num_tokens": 8557865.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.310224342346191,
|
|
"epoch": 4.023391812865497,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039348005501289775,
|
|
"loss": 5.6635,
|
|
"mean_token_accuracy": 0.18840035796165466,
|
|
"num_tokens": 8570203.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.20900592803955,
|
|
"epoch": 4.029239766081871,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00039308131064282016,
|
|
"loss": 5.7573,
|
|
"mean_token_accuracy": 0.18991120159626007,
|
|
"num_tokens": 8582638.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.371462488174439,
|
|
"epoch": 4.035087719298246,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003926820537913868,
|
|
"loss": 5.8399,
|
|
"mean_token_accuracy": 0.18359654098749162,
|
|
"num_tokens": 8594973.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.307078886032104,
|
|
"epoch": 4.04093567251462,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003922822861914012,
|
|
"loss": 5.7047,
|
|
"mean_token_accuracy": 0.19176373183727263,
|
|
"num_tokens": 8607216.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.380091571807862,
|
|
"epoch": 4.046783625730994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003918820095778832,
|
|
"loss": 5.9065,
|
|
"mean_token_accuracy": 0.17466982305049897,
|
|
"num_tokens": 8620833.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.3202704906463625,
|
|
"epoch": 4.052631578947368,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00039148122568806215,
|
|
"loss": 5.7286,
|
|
"mean_token_accuracy": 0.19510964304208755,
|
|
"num_tokens": 8632931.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.263472175598144,
|
|
"epoch": 4.058479532163743,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003910799362613688,
|
|
"loss": 5.8119,
|
|
"mean_token_accuracy": 0.1810266464948654,
|
|
"num_tokens": 8644226.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.242630481719971,
|
|
"epoch": 4.064327485380117,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039067814303942806,
|
|
"loss": 5.6201,
|
|
"mean_token_accuracy": 0.20384632647037507,
|
|
"num_tokens": 8656569.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.200847578048706,
|
|
"epoch": 4.0701754385964914,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039027584776605136,
|
|
"loss": 5.649,
|
|
"mean_token_accuracy": 0.1954135701060295,
|
|
"num_tokens": 8668077.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.340264749526978,
|
|
"epoch": 4.076023391812866,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00038987305218722904,
|
|
"loss": 5.7373,
|
|
"mean_token_accuracy": 0.1845865711569786,
|
|
"num_tokens": 8680227.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.334174776077271,
|
|
"epoch": 4.081871345029239,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003894697580511226,
|
|
"loss": 5.8268,
|
|
"mean_token_accuracy": 0.18544215708971024,
|
|
"num_tokens": 8692354.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.300135374069214,
|
|
"epoch": 4.087719298245614,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003890659671080579,
|
|
"loss": 5.7836,
|
|
"mean_token_accuracy": 0.17149038314819337,
|
|
"num_tokens": 8704634.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.360431241989136,
|
|
"epoch": 4.093567251461988,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003886616811105162,
|
|
"loss": 5.8245,
|
|
"mean_token_accuracy": 0.18058949559926987,
|
|
"num_tokens": 8717253.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 4.093567251461988,
|
|
"eval_entropy": 6.157160685829005,
|
|
"eval_loss": 6.595019817352295,
|
|
"eval_mean_token_accuracy": 0.14759186410227004,
|
|
"eval_num_tokens": 8717253.0,
|
|
"eval_runtime": 1.946,
|
|
"eval_samples_per_second": 1391.09,
|
|
"eval_steps_per_second": 174.207,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.384776067733765,
|
|
"epoch": 4.099415204678363,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038825690181312793,
|
|
"loss": 5.8123,
|
|
"mean_token_accuracy": 0.1791504666209221,
|
|
"num_tokens": 8729840.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.275667190551758,
|
|
"epoch": 4.105263157894737,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038785163097266424,
|
|
"loss": 5.6653,
|
|
"mean_token_accuracy": 0.19373680651187897,
|
|
"num_tokens": 8742065.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.273963785171508,
|
|
"epoch": 4.111111111111111,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003874458703480296,
|
|
"loss": 5.8457,
|
|
"mean_token_accuracy": 0.18192827701568604,
|
|
"num_tokens": 8754113.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.408229064941406,
|
|
"epoch": 4.116959064327485,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003870396217002542,
|
|
"loss": 5.8813,
|
|
"mean_token_accuracy": 0.1725478783249855,
|
|
"num_tokens": 8767255.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.310436391830445,
|
|
"epoch": 4.12280701754386,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003866328867924864,
|
|
"loss": 5.8007,
|
|
"mean_token_accuracy": 0.17622703462839126,
|
|
"num_tokens": 8781010.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.35017580986023,
|
|
"epoch": 4.128654970760234,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003862256673899848,
|
|
"loss": 5.8205,
|
|
"mean_token_accuracy": 0.18056162744760512,
|
|
"num_tokens": 8794393.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.31821985244751,
|
|
"epoch": 4.1345029239766085,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003858179652601108,
|
|
"loss": 5.8988,
|
|
"mean_token_accuracy": 0.17191343605518342,
|
|
"num_tokens": 8806496.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.297108268737793,
|
|
"epoch": 4.140350877192983,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038540978217232086,
|
|
"loss": 5.6853,
|
|
"mean_token_accuracy": 0.18976295441389085,
|
|
"num_tokens": 8818411.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.317528295516968,
|
|
"epoch": 4.146198830409356,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00038500111989815894,
|
|
"loss": 5.7947,
|
|
"mean_token_accuracy": 0.18120289146900176,
|
|
"num_tokens": 8829714.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.263475370407105,
|
|
"epoch": 4.152046783625731,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038459198021124853,
|
|
"loss": 5.763,
|
|
"mean_token_accuracy": 0.18518542051315307,
|
|
"num_tokens": 8842393.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.339914703369141,
|
|
"epoch": 4.157894736842105,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003841823648872852,
|
|
"loss": 5.7902,
|
|
"mean_token_accuracy": 0.18060382604598998,
|
|
"num_tokens": 8855263.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.320138025283813,
|
|
"epoch": 4.16374269005848,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003837722757040289,
|
|
"loss": 5.8351,
|
|
"mean_token_accuracy": 0.18005656450986862,
|
|
"num_tokens": 8868220.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.318368339538575,
|
|
"epoch": 4.169590643274854,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003833617144412961,
|
|
"loss": 5.7718,
|
|
"mean_token_accuracy": 0.18751097321510315,
|
|
"num_tokens": 8879106.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.280430793762207,
|
|
"epoch": 4.175438596491228,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0003829506828809521,
|
|
"loss": 5.8549,
|
|
"mean_token_accuracy": 0.18114213347435,
|
|
"num_tokens": 8891358.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.360328531265258,
|
|
"epoch": 4.181286549707602,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003825391828069032,
|
|
"loss": 5.782,
|
|
"mean_token_accuracy": 0.18212279975414275,
|
|
"num_tokens": 8904327.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.26585054397583,
|
|
"epoch": 4.187134502923977,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003821272160050894,
|
|
"loss": 5.7474,
|
|
"mean_token_accuracy": 0.18088108003139497,
|
|
"num_tokens": 8916515.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.294408988952637,
|
|
"epoch": 4.192982456140351,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003817147842634762,
|
|
"loss": 5.8318,
|
|
"mean_token_accuracy": 0.1794787049293518,
|
|
"num_tokens": 8929482.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.390404176712036,
|
|
"epoch": 4.1988304093567255,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003813018893720468,
|
|
"loss": 5.8745,
|
|
"mean_token_accuracy": 0.17279811948537827,
|
|
"num_tokens": 8942328.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.278195858001709,
|
|
"epoch": 4.204678362573099,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038088853312279473,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.18261341750621796,
|
|
"num_tokens": 8955116.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.369569635391235,
|
|
"epoch": 4.2105263157894735,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003804747173097157,
|
|
"loss": 5.8735,
|
|
"mean_token_accuracy": 0.17344097197055816,
|
|
"num_tokens": 8967459.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.296767377853394,
|
|
"epoch": 4.216374269005848,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038006044372880006,
|
|
"loss": 5.7822,
|
|
"mean_token_accuracy": 0.18315625339746475,
|
|
"num_tokens": 8980079.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.301019239425659,
|
|
"epoch": 4.222222222222222,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037964571417802473,
|
|
"loss": 5.825,
|
|
"mean_token_accuracy": 0.1857357770204544,
|
|
"num_tokens": 8992764.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.2863012790679935,
|
|
"epoch": 4.228070175438597,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037923053045734595,
|
|
"loss": 5.7133,
|
|
"mean_token_accuracy": 0.17899733036756516,
|
|
"num_tokens": 9004904.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.400509357452393,
|
|
"epoch": 4.23391812865497,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00037881489436869054,
|
|
"loss": 5.9219,
|
|
"mean_token_accuracy": 0.16955226808786392,
|
|
"num_tokens": 9018411.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.288416004180908,
|
|
"epoch": 4.239766081871345,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003783988077159491,
|
|
"loss": 5.7271,
|
|
"mean_token_accuracy": 0.19147274047136306,
|
|
"num_tokens": 9030593.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.20024471282959,
|
|
"epoch": 4.245614035087719,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037798227230496754,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.19359563589096068,
|
|
"num_tokens": 9043715.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.190207529067993,
|
|
"epoch": 4.251461988304094,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037756528994353945,
|
|
"loss": 5.7068,
|
|
"mean_token_accuracy": 0.19130271375179292,
|
|
"num_tokens": 9056384.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.294029712677002,
|
|
"epoch": 4.257309941520468,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000377147862441398,
|
|
"loss": 5.8131,
|
|
"mean_token_accuracy": 0.1839430958032608,
|
|
"num_tokens": 9069444.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.303832483291626,
|
|
"epoch": 4.2631578947368425,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037672999161020873,
|
|
"loss": 5.7358,
|
|
"mean_token_accuracy": 0.1842231497168541,
|
|
"num_tokens": 9080933.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.280602169036865,
|
|
"epoch": 4.269005847953216,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037631167926356087,
|
|
"loss": 5.8008,
|
|
"mean_token_accuracy": 0.18425682783126832,
|
|
"num_tokens": 9093405.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.265407419204712,
|
|
"epoch": 4.2748538011695905,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003758929272169599,
|
|
"loss": 5.7756,
|
|
"mean_token_accuracy": 0.18727248758077622,
|
|
"num_tokens": 9105707.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.33483247756958,
|
|
"epoch": 4.280701754385965,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00037547373728781996,
|
|
"loss": 5.7426,
|
|
"mean_token_accuracy": 0.18454704731702803,
|
|
"num_tokens": 9119476.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.239254713058472,
|
|
"epoch": 4.286549707602339,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003750541112954552,
|
|
"loss": 5.7142,
|
|
"mean_token_accuracy": 0.18759913742542267,
|
|
"num_tokens": 9131788.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.28587532043457,
|
|
"epoch": 4.292397660818714,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003746340510610727,
|
|
"loss": 5.8259,
|
|
"mean_token_accuracy": 0.17954733371734619,
|
|
"num_tokens": 9144340.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.3251869678497314,
|
|
"epoch": 4.298245614035087,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037421355840776396,
|
|
"loss": 5.7912,
|
|
"mean_token_accuracy": 0.1825256571173668,
|
|
"num_tokens": 9156757.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.197872352600098,
|
|
"epoch": 4.304093567251462,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003737926351604971,
|
|
"loss": 5.7708,
|
|
"mean_token_accuracy": 0.18350359946489334,
|
|
"num_tokens": 9169975.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.324824714660645,
|
|
"epoch": 4.309941520467836,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00037337128314610926,
|
|
"loss": 5.7841,
|
|
"mean_token_accuracy": 0.18487975746393204,
|
|
"num_tokens": 9181891.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.316333389282226,
|
|
"epoch": 4.315789473684211,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003729495041932984,
|
|
"loss": 5.8104,
|
|
"mean_token_accuracy": 0.17722948044538497,
|
|
"num_tokens": 9194068.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.262782621383667,
|
|
"epoch": 4.321637426900585,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003725273001326155,
|
|
"loss": 5.8336,
|
|
"mean_token_accuracy": 0.1793358787894249,
|
|
"num_tokens": 9205530.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.318678617477417,
|
|
"epoch": 4.3274853801169595,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003721046727964565,
|
|
"loss": 5.7562,
|
|
"mean_token_accuracy": 0.188470958173275,
|
|
"num_tokens": 9218444.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.278766298294068,
|
|
"epoch": 4.333333333333333,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00037168162401905416,
|
|
"loss": 5.8551,
|
|
"mean_token_accuracy": 0.18399844020605088,
|
|
"num_tokens": 9229507.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.324070310592651,
|
|
"epoch": 4.3391812865497075,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00037125815563647063,
|
|
"loss": 5.8967,
|
|
"mean_token_accuracy": 0.17816781848669053,
|
|
"num_tokens": 9242128.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.341136693954468,
|
|
"epoch": 4.345029239766082,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00037083426948658903,
|
|
"loss": 5.847,
|
|
"mean_token_accuracy": 0.17691442668437957,
|
|
"num_tokens": 9255598.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.288648796081543,
|
|
"epoch": 4.350877192982456,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00037040996740910566,
|
|
"loss": 5.7908,
|
|
"mean_token_accuracy": 0.184350822865963,
|
|
"num_tokens": 9267568.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.314821386337281,
|
|
"epoch": 4.356725146198831,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000369985251245522,
|
|
"loss": 5.8195,
|
|
"mean_token_accuracy": 0.18325863629579545,
|
|
"num_tokens": 9280223.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.270903396606445,
|
|
"epoch": 4.362573099415204,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00036956012283913655,
|
|
"loss": 5.8757,
|
|
"mean_token_accuracy": 0.17867998033761978,
|
|
"num_tokens": 9292933.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.291424798965454,
|
|
"epoch": 4.368421052631579,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00036913458403503715,
|
|
"loss": 5.8225,
|
|
"mean_token_accuracy": 0.18914655447006226,
|
|
"num_tokens": 9306739.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.352597713470459,
|
|
"epoch": 4.374269005847953,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003687086366800927,
|
|
"loss": 5.7873,
|
|
"mean_token_accuracy": 0.18101000934839248,
|
|
"num_tokens": 9319151.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.308052587509155,
|
|
"epoch": 4.380116959064328,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003682822826229453,
|
|
"loss": 5.8528,
|
|
"mean_token_accuracy": 0.1763809308409691,
|
|
"num_tokens": 9332035.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.342720365524292,
|
|
"epoch": 4.385964912280702,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036785552371400206,
|
|
"loss": 5.89,
|
|
"mean_token_accuracy": 0.17496565133333206,
|
|
"num_tokens": 9345248.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.316687822341919,
|
|
"epoch": 4.391812865497076,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003674283618054273,
|
|
"loss": 5.8401,
|
|
"mean_token_accuracy": 0.18606477826833726,
|
|
"num_tokens": 9357514.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.298449611663818,
|
|
"epoch": 4.39766081871345,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003670007987511344,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.17373306006193162,
|
|
"num_tokens": 9368627.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.215012836456299,
|
|
"epoch": 4.4035087719298245,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003665728364067776,
|
|
"loss": 5.7748,
|
|
"mean_token_accuracy": 0.18543099761009216,
|
|
"num_tokens": 9380591.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.281973457336425,
|
|
"epoch": 4.409356725146199,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00036614447662974413,
|
|
"loss": 5.7592,
|
|
"mean_token_accuracy": 0.1860760360956192,
|
|
"num_tokens": 9392873.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.385033702850341,
|
|
"epoch": 4.415204678362573,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003657157212791462,
|
|
"loss": 5.9039,
|
|
"mean_token_accuracy": 0.173318113386631,
|
|
"num_tokens": 9404252.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.23109130859375,
|
|
"epoch": 4.421052631578947,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003652865722158128,
|
|
"loss": 5.8346,
|
|
"mean_token_accuracy": 0.18261070996522905,
|
|
"num_tokens": 9416091.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.317514944076538,
|
|
"epoch": 4.426900584795321,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00036485703130228147,
|
|
"loss": 5.7585,
|
|
"mean_token_accuracy": 0.1898007020354271,
|
|
"num_tokens": 9429913.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.285356760025024,
|
|
"epoch": 4.432748538011696,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036442710040279086,
|
|
"loss": 5.7386,
|
|
"mean_token_accuracy": 0.19126825034618378,
|
|
"num_tokens": 9442413.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.106606674194336,
|
|
"epoch": 4.43859649122807,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003639967813832717,
|
|
"loss": 5.6783,
|
|
"mean_token_accuracy": 0.19114170968532562,
|
|
"num_tokens": 9453636.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.251889944076538,
|
|
"epoch": 4.444444444444445,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003635660761113394,
|
|
"loss": 5.7704,
|
|
"mean_token_accuracy": 0.18708461672067642,
|
|
"num_tokens": 9466577.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.2538471698760985,
|
|
"epoch": 4.450292397660819,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003631349864562857,
|
|
"loss": 5.7819,
|
|
"mean_token_accuracy": 0.1778549924492836,
|
|
"num_tokens": 9479852.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.2790015697479244,
|
|
"epoch": 4.456140350877193,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003627035142890708,
|
|
"loss": 5.7376,
|
|
"mean_token_accuracy": 0.18533434122800826,
|
|
"num_tokens": 9493373.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.287004137039185,
|
|
"epoch": 4.461988304093567,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003622716614823147,
|
|
"loss": 5.8262,
|
|
"mean_token_accuracy": 0.18904732465744017,
|
|
"num_tokens": 9505793.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.257832002639771,
|
|
"epoch": 4.4678362573099415,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003618394299102896,
|
|
"loss": 5.858,
|
|
"mean_token_accuracy": 0.17263268828392028,
|
|
"num_tokens": 9517810.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.311147308349609,
|
|
"epoch": 4.473684210526316,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00036140682144891143,
|
|
"loss": 5.827,
|
|
"mean_token_accuracy": 0.17715798914432526,
|
|
"num_tokens": 9529430.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.289704275131226,
|
|
"epoch": 4.47953216374269,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00036097383797573205,
|
|
"loss": 5.811,
|
|
"mean_token_accuracy": 0.18603776544332504,
|
|
"num_tokens": 9542385.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.254218626022339,
|
|
"epoch": 4.485380116959064,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00036054048136993075,
|
|
"loss": 5.7748,
|
|
"mean_token_accuracy": 0.1854638710618019,
|
|
"num_tokens": 9553955.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.3724775314331055,
|
|
"epoch": 4.491228070175438,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00036010675351230616,
|
|
"loss": 5.8324,
|
|
"mean_token_accuracy": 0.17608457803726196,
|
|
"num_tokens": 9566494.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.231585121154785,
|
|
"epoch": 4.497076023391813,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003596726562852684,
|
|
"loss": 5.7514,
|
|
"mean_token_accuracy": 0.19300678074359895,
|
|
"num_tokens": 9579122.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.237770843505859,
|
|
"epoch": 4.502923976608187,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00035923819157283074,
|
|
"loss": 5.7904,
|
|
"mean_token_accuracy": 0.18340874910354615,
|
|
"num_tokens": 9592144.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.186396551132202,
|
|
"epoch": 4.508771929824562,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00035880336126060074,
|
|
"loss": 5.6768,
|
|
"mean_token_accuracy": 0.18883186429739,
|
|
"num_tokens": 9604533.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.264141511917114,
|
|
"epoch": 4.514619883040936,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00035836816723577354,
|
|
"loss": 5.857,
|
|
"mean_token_accuracy": 0.18151703029870986,
|
|
"num_tokens": 9617349.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.253802490234375,
|
|
"epoch": 4.52046783625731,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003579326113871222,
|
|
"loss": 5.8274,
|
|
"mean_token_accuracy": 0.18349558264017105,
|
|
"num_tokens": 9629599.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.258465099334717,
|
|
"epoch": 4.526315789473684,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003574966956049904,
|
|
"loss": 5.7936,
|
|
"mean_token_accuracy": 0.1862207069993019,
|
|
"num_tokens": 9642041.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.287553405761718,
|
|
"epoch": 4.5321637426900585,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00035706042178128397,
|
|
"loss": 5.8176,
|
|
"mean_token_accuracy": 0.182197205722332,
|
|
"num_tokens": 9654796.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.221886491775512,
|
|
"epoch": 4.538011695906433,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003566237918094625,
|
|
"loss": 5.8439,
|
|
"mean_token_accuracy": 0.18151349127292632,
|
|
"num_tokens": 9667211.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.230513381958008,
|
|
"epoch": 4.543859649122807,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003561868075845313,
|
|
"loss": 5.7447,
|
|
"mean_token_accuracy": 0.19109286218881608,
|
|
"num_tokens": 9679526.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.376104211807251,
|
|
"epoch": 4.549707602339181,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035574947100303327,
|
|
"loss": 5.9376,
|
|
"mean_token_accuracy": 0.17295238226652146,
|
|
"num_tokens": 9691760.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.286878347396851,
|
|
"epoch": 4.555555555555555,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00035531178396304083,
|
|
"loss": 5.8474,
|
|
"mean_token_accuracy": 0.1865091934800148,
|
|
"num_tokens": 9704154.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.184697580337525,
|
|
"epoch": 4.56140350877193,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035487374836414676,
|
|
"loss": 5.6564,
|
|
"mean_token_accuracy": 0.19328852146863937,
|
|
"num_tokens": 9715850.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.2825733661651615,
|
|
"epoch": 4.567251461988304,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00035443536610745707,
|
|
"loss": 5.8317,
|
|
"mean_token_accuracy": 0.18148200064897538,
|
|
"num_tokens": 9728202.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.3443831443786625,
|
|
"epoch": 4.573099415204679,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00035399663909558226,
|
|
"loss": 5.8295,
|
|
"mean_token_accuracy": 0.17485745698213578,
|
|
"num_tokens": 9741363.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.265570878982544,
|
|
"epoch": 4.578947368421053,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00035355756923262897,
|
|
"loss": 5.8191,
|
|
"mean_token_accuracy": 0.17813921719789505,
|
|
"num_tokens": 9755231.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.192175102233887,
|
|
"epoch": 4.584795321637427,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003531181584241919,
|
|
"loss": 5.6717,
|
|
"mean_token_accuracy": 0.19509032666683196,
|
|
"num_tokens": 9768972.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.299138116836548,
|
|
"epoch": 4.590643274853801,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003526784085773455,
|
|
"loss": 5.8644,
|
|
"mean_token_accuracy": 0.17538984268903732,
|
|
"num_tokens": 9781356.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.258856344223022,
|
|
"epoch": 4.5964912280701755,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003522383216006356,
|
|
"loss": 5.8107,
|
|
"mean_token_accuracy": 0.17778304517269133,
|
|
"num_tokens": 9793227.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.249921607971191,
|
|
"epoch": 4.60233918128655,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003517978994040713,
|
|
"loss": 5.7987,
|
|
"mean_token_accuracy": 0.1825487419962883,
|
|
"num_tokens": 9805968.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.317439937591553,
|
|
"epoch": 4.6081871345029235,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00035135714389911637,
|
|
"loss": 5.7947,
|
|
"mean_token_accuracy": 0.18034329265356064,
|
|
"num_tokens": 9817864.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.224619674682617,
|
|
"epoch": 4.614035087719298,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003509160569986814,
|
|
"loss": 5.778,
|
|
"mean_token_accuracy": 0.18772294223308564,
|
|
"num_tokens": 9831192.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.270768499374389,
|
|
"epoch": 4.619883040935672,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00035047464061711527,
|
|
"loss": 5.8119,
|
|
"mean_token_accuracy": 0.17892582714557648,
|
|
"num_tokens": 9842813.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.322039699554443,
|
|
"epoch": 4.625730994152047,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00035003289667019655,
|
|
"loss": 5.8239,
|
|
"mean_token_accuracy": 0.1842857927083969,
|
|
"num_tokens": 9854500.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.2162456035614015,
|
|
"epoch": 4.631578947368421,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003495908270751257,
|
|
"loss": 5.8019,
|
|
"mean_token_accuracy": 0.18256803452968598,
|
|
"num_tokens": 9865729.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.356060791015625,
|
|
"epoch": 4.637426900584796,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003491484337505166,
|
|
"loss": 5.834,
|
|
"mean_token_accuracy": 0.17961307466030121,
|
|
"num_tokens": 9878306.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.282822561264038,
|
|
"epoch": 4.643274853801169,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00034870571861638785,
|
|
"loss": 5.848,
|
|
"mean_token_accuracy": 0.18053650557994844,
|
|
"num_tokens": 9890384.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.213586473464966,
|
|
"epoch": 4.649122807017544,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00034826268359415487,
|
|
"loss": 5.8074,
|
|
"mean_token_accuracy": 0.18246710896492005,
|
|
"num_tokens": 9902960.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.266279315948486,
|
|
"epoch": 4.654970760233918,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003478193306066215,
|
|
"loss": 5.7831,
|
|
"mean_token_accuracy": 0.18555108308792115,
|
|
"num_tokens": 9915143.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.303533601760864,
|
|
"epoch": 4.6608187134502925,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00034737566157797147,
|
|
"loss": 5.8721,
|
|
"mean_token_accuracy": 0.1769675776362419,
|
|
"num_tokens": 9928440.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.288042449951172,
|
|
"epoch": 4.666666666666667,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034693167843376016,
|
|
"loss": 5.886,
|
|
"mean_token_accuracy": 0.17639421820640563,
|
|
"num_tokens": 9940973.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.255143022537231,
|
|
"epoch": 4.6725146198830405,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00034648738310090634,
|
|
"loss": 5.7756,
|
|
"mean_token_accuracy": 0.18952346742153167,
|
|
"num_tokens": 9953433.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.148354196548462,
|
|
"epoch": 4.678362573099415,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003460427775076835,
|
|
"loss": 5.717,
|
|
"mean_token_accuracy": 0.18960760831832885,
|
|
"num_tokens": 9964641.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 4.678362573099415,
|
|
"eval_entropy": 6.105867270523117,
|
|
"eval_loss": 6.555722713470459,
|
|
"eval_mean_token_accuracy": 0.1492111394538992,
|
|
"eval_num_tokens": 9964641.0,
|
|
"eval_runtime": 1.7342,
|
|
"eval_samples_per_second": 1560.986,
|
|
"eval_steps_per_second": 195.484,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.321864938735962,
|
|
"epoch": 4.684210526315789,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00034559786358371187,
|
|
"loss": 5.898,
|
|
"mean_token_accuracy": 0.1739409387111664,
|
|
"num_tokens": 9978625.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.236947107315063,
|
|
"epoch": 4.690058479532164,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00034515264325994975,
|
|
"loss": 5.7519,
|
|
"mean_token_accuracy": 0.18472876250743867,
|
|
"num_tokens": 9991758.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.343029975891113,
|
|
"epoch": 4.695906432748538,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00034470711846868525,
|
|
"loss": 5.8742,
|
|
"mean_token_accuracy": 0.17603392899036407,
|
|
"num_tokens": 10005496.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.253848075866699,
|
|
"epoch": 4.701754385964913,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00034426129114352807,
|
|
"loss": 5.8921,
|
|
"mean_token_accuracy": 0.18001782298088073,
|
|
"num_tokens": 10019025.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.179688501358032,
|
|
"epoch": 4.707602339181286,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003438151632194006,
|
|
"loss": 5.7609,
|
|
"mean_token_accuracy": 0.18558207005262375,
|
|
"num_tokens": 10030741.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.320484447479248,
|
|
"epoch": 4.713450292397661,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034336873663252997,
|
|
"loss": 5.8573,
|
|
"mean_token_accuracy": 0.18368003219366075,
|
|
"num_tokens": 10043138.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.263317823410034,
|
|
"epoch": 4.719298245614035,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00034292201332043976,
|
|
"loss": 5.8294,
|
|
"mean_token_accuracy": 0.17762552052736283,
|
|
"num_tokens": 10055927.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.26686692237854,
|
|
"epoch": 4.7251461988304095,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003424749952219411,
|
|
"loss": 5.7342,
|
|
"mean_token_accuracy": 0.18830980360507965,
|
|
"num_tokens": 10067392.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.252482795715332,
|
|
"epoch": 4.730994152046784,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00034202768427712456,
|
|
"loss": 5.7982,
|
|
"mean_token_accuracy": 0.1881715953350067,
|
|
"num_tokens": 10079245.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.187083959579468,
|
|
"epoch": 4.7368421052631575,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003415800824273518,
|
|
"loss": 5.7148,
|
|
"mean_token_accuracy": 0.18758911043405532,
|
|
"num_tokens": 10091923.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.29634461402893,
|
|
"epoch": 4.742690058479532,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003411321916152469,
|
|
"loss": 5.7849,
|
|
"mean_token_accuracy": 0.18653489053249359,
|
|
"num_tokens": 10105121.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 6.227941560745239,
|
|
"epoch": 4.748538011695906,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003406840137846881,
|
|
"loss": 5.8444,
|
|
"mean_token_accuracy": 0.1814254641532898,
|
|
"num_tokens": 10117005.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.2165430068969725,
|
|
"epoch": 4.754385964912281,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034023555088079924,
|
|
"loss": 5.7834,
|
|
"mean_token_accuracy": 0.1896864205598831,
|
|
"num_tokens": 10129958.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.24987096786499,
|
|
"epoch": 4.760233918128655,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003397868048499416,
|
|
"loss": 5.7973,
|
|
"mean_token_accuracy": 0.18600700348615645,
|
|
"num_tokens": 10141537.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 6.242813158035278,
|
|
"epoch": 4.76608187134503,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033933777763970506,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.18931635469198227,
|
|
"num_tokens": 10153882.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.286039590835571,
|
|
"epoch": 4.771929824561403,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003388884711989,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.17984680086374283,
|
|
"num_tokens": 10165147.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 6.276972150802612,
|
|
"epoch": 4.777777777777778,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003384388874775486,
|
|
"loss": 5.8602,
|
|
"mean_token_accuracy": 0.17511490881443023,
|
|
"num_tokens": 10177926.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 6.26650185585022,
|
|
"epoch": 4.783625730994152,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033798902842687627,
|
|
"loss": 5.8016,
|
|
"mean_token_accuracy": 0.18587989807128907,
|
|
"num_tokens": 10190343.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 6.333509111404419,
|
|
"epoch": 4.7894736842105265,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003375388959993037,
|
|
"loss": 5.9057,
|
|
"mean_token_accuracy": 0.17792484909296036,
|
|
"num_tokens": 10201363.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 6.1296755313873295,
|
|
"epoch": 4.7953216374269,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0003370884921484379,
|
|
"loss": 5.7349,
|
|
"mean_token_accuracy": 0.1892000839114189,
|
|
"num_tokens": 10213976.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 6.260969352722168,
|
|
"epoch": 4.8011695906432745,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003366378188290638,
|
|
"loss": 5.798,
|
|
"mean_token_accuracy": 0.18810132890939713,
|
|
"num_tokens": 10225645.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 6.325876092910766,
|
|
"epoch": 4.807017543859649,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003361868779971361,
|
|
"loss": 5.8037,
|
|
"mean_token_accuracy": 0.18900905996561052,
|
|
"num_tokens": 10237144.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 6.3107692241668705,
|
|
"epoch": 4.812865497076023,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033573567160977017,
|
|
"loss": 5.9785,
|
|
"mean_token_accuracy": 0.17349042892456054,
|
|
"num_tokens": 10249215.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 6.230036115646362,
|
|
"epoch": 4.818713450292398,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000335284201625234,
|
|
"loss": 5.7749,
|
|
"mean_token_accuracy": 0.18268027752637864,
|
|
"num_tokens": 10262428.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 6.2292500019073485,
|
|
"epoch": 4.824561403508772,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033483247000294,
|
|
"loss": 5.798,
|
|
"mean_token_accuracy": 0.1878509521484375,
|
|
"num_tokens": 10275370.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 6.234437799453735,
|
|
"epoch": 4.830409356725146,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003343804787034357,
|
|
"loss": 5.8101,
|
|
"mean_token_accuracy": 0.18721414506435394,
|
|
"num_tokens": 10287378.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 6.30766749382019,
|
|
"epoch": 4.83625730994152,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003339282296883956,
|
|
"loss": 5.8139,
|
|
"mean_token_accuracy": 0.1764067679643631,
|
|
"num_tokens": 10300063.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 6.246171474456787,
|
|
"epoch": 4.842105263157895,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000333475724920613,
|
|
"loss": 5.8115,
|
|
"mean_token_accuracy": 0.1838441386818886,
|
|
"num_tokens": 10312095.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 6.258604335784912,
|
|
"epoch": 4.847953216374269,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003330229663639909,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.18368408381938933,
|
|
"num_tokens": 10324903.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 6.365167427062988,
|
|
"epoch": 4.853801169590644,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000332569955983534,
|
|
"loss": 5.865,
|
|
"mean_token_accuracy": 0.1824457123875618,
|
|
"num_tokens": 10339253.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 6.305107307434082,
|
|
"epoch": 4.859649122807017,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003321166957453398,
|
|
"loss": 5.8729,
|
|
"mean_token_accuracy": 0.17533203661441804,
|
|
"num_tokens": 10350845.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.23463807106018,
|
|
"epoch": 4.8654970760233915,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003316631876165904,
|
|
"loss": 5.7867,
|
|
"mean_token_accuracy": 0.18521284461021423,
|
|
"num_tokens": 10363357.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.159645223617554,
|
|
"epoch": 4.871345029239766,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033120943356554334,
|
|
"loss": 5.7208,
|
|
"mean_token_accuracy": 0.18627448081970216,
|
|
"num_tokens": 10375664.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 6.279516887664795,
|
|
"epoch": 4.87719298245614,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000330755435561524,
|
|
"loss": 5.7796,
|
|
"mean_token_accuracy": 0.18429485708475113,
|
|
"num_tokens": 10387412.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.23290433883667,
|
|
"epoch": 4.883040935672515,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033030119557491604,
|
|
"loss": 5.8336,
|
|
"mean_token_accuracy": 0.18351123183965684,
|
|
"num_tokens": 10399083.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.365048789978028,
|
|
"epoch": 4.888888888888889,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003298467155771537,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.17174902707338333,
|
|
"num_tokens": 10412713.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 6.182602453231811,
|
|
"epoch": 4.894736842105263,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00032939199754071274,
|
|
"loss": 5.7348,
|
|
"mean_token_accuracy": 0.1885097712278366,
|
|
"num_tokens": 10426836.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 6.198043251037598,
|
|
"epoch": 4.900584795321637,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000328937043439102,
|
|
"loss": 5.8076,
|
|
"mean_token_accuracy": 0.1866941809654236,
|
|
"num_tokens": 10439817.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.305385255813599,
|
|
"epoch": 4.906432748538012,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032848185524685504,
|
|
"loss": 5.8168,
|
|
"mean_token_accuracy": 0.1808120146393776,
|
|
"num_tokens": 10452509.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 6.3346611022949215,
|
|
"epoch": 4.912280701754386,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00032802643493952106,
|
|
"loss": 5.8433,
|
|
"mean_token_accuracy": 0.18142112493515014,
|
|
"num_tokens": 10464593.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.226899147033691,
|
|
"epoch": 4.918128654970761,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000327570784493657,
|
|
"loss": 5.8665,
|
|
"mean_token_accuracy": 0.1784406304359436,
|
|
"num_tokens": 10476868.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 6.227827548980713,
|
|
"epoch": 4.923976608187134,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00032711490588681845,
|
|
"loss": 5.6914,
|
|
"mean_token_accuracy": 0.18843241780996323,
|
|
"num_tokens": 10489510.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 6.294796466827393,
|
|
"epoch": 4.9298245614035086,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003266588010975512,
|
|
"loss": 5.8458,
|
|
"mean_token_accuracy": 0.17990170121192933,
|
|
"num_tokens": 10501901.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.188789701461792,
|
|
"epoch": 4.935672514619883,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003262024721053827,
|
|
"loss": 5.7627,
|
|
"mean_token_accuracy": 0.1864606812596321,
|
|
"num_tokens": 10513604.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.183052349090576,
|
|
"epoch": 4.941520467836257,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00032574592089081375,
|
|
"loss": 5.7744,
|
|
"mean_token_accuracy": 0.19281402081251145,
|
|
"num_tokens": 10527882.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 6.277568340301514,
|
|
"epoch": 4.947368421052632,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032528914943530905,
|
|
"loss": 5.7528,
|
|
"mean_token_accuracy": 0.19354891330003737,
|
|
"num_tokens": 10539514.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.25462064743042,
|
|
"epoch": 4.953216374269006,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003248321597212896,
|
|
"loss": 5.812,
|
|
"mean_token_accuracy": 0.1798214465379715,
|
|
"num_tokens": 10551984.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.335310411453247,
|
|
"epoch": 4.95906432748538,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003243749537321236,
|
|
"loss": 5.8938,
|
|
"mean_token_accuracy": 0.17632542103528975,
|
|
"num_tokens": 10564915.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 6.20843539237976,
|
|
"epoch": 4.964912280701754,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032391753345211767,
|
|
"loss": 5.8563,
|
|
"mean_token_accuracy": 0.18176027089357377,
|
|
"num_tokens": 10576614.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.285629844665527,
|
|
"epoch": 4.970760233918129,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003234599008665088,
|
|
"loss": 5.8167,
|
|
"mean_token_accuracy": 0.18787658214569092,
|
|
"num_tokens": 10588746.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 6.238576459884643,
|
|
"epoch": 4.976608187134503,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003230020579614551,
|
|
"loss": 5.8373,
|
|
"mean_token_accuracy": 0.17759260088205336,
|
|
"num_tokens": 10600101.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 6.2283008098602295,
|
|
"epoch": 4.982456140350877,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003225440067240276,
|
|
"loss": 5.7935,
|
|
"mean_token_accuracy": 0.1848200023174286,
|
|
"num_tokens": 10611879.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.293163251876831,
|
|
"epoch": 4.988304093567251,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032208574914220167,
|
|
"loss": 5.8607,
|
|
"mean_token_accuracy": 0.18218080550432206,
|
|
"num_tokens": 10624201.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 6.285972690582275,
|
|
"epoch": 4.994152046783626,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003216272872048479,
|
|
"loss": 5.8023,
|
|
"mean_token_accuracy": 0.18011997491121293,
|
|
"num_tokens": 10636612.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.268482828140259,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003211686229017241,
|
|
"loss": 5.7936,
|
|
"mean_token_accuracy": 0.18320332318544388,
|
|
"num_tokens": 10647420.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 6.264548587799072,
|
|
"epoch": 5.005847953216374,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003207097582234662,
|
|
"loss": 5.64,
|
|
"mean_token_accuracy": 0.18487039357423782,
|
|
"num_tokens": 10660096.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 6.189789915084839,
|
|
"epoch": 5.011695906432749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003202506951615798,
|
|
"loss": 5.5828,
|
|
"mean_token_accuracy": 0.19722577929496765,
|
|
"num_tokens": 10672364.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 6.264693117141723,
|
|
"epoch": 5.017543859649122,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00031979143570843145,
|
|
"loss": 5.5877,
|
|
"mean_token_accuracy": 0.1913362517952919,
|
|
"num_tokens": 10685230.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 6.34747052192688,
|
|
"epoch": 5.023391812865497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00031933198185724016,
|
|
"loss": 5.6885,
|
|
"mean_token_accuracy": 0.18619947135448456,
|
|
"num_tokens": 10697837.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 6.177005624771118,
|
|
"epoch": 5.029239766081871,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00031887233560206864,
|
|
"loss": 5.5088,
|
|
"mean_token_accuracy": 0.20209251791238786,
|
|
"num_tokens": 10711549.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 6.306797170639038,
|
|
"epoch": 5.035087719298246,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031841249893781447,
|
|
"loss": 5.6745,
|
|
"mean_token_accuracy": 0.1891188442707062,
|
|
"num_tokens": 10724897.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 6.267652893066407,
|
|
"epoch": 5.04093567251462,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003179524738602019,
|
|
"loss": 5.587,
|
|
"mean_token_accuracy": 0.18976380974054335,
|
|
"num_tokens": 10735808.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 6.186126613616944,
|
|
"epoch": 5.046783625730994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00031749226236577277,
|
|
"loss": 5.5414,
|
|
"mean_token_accuracy": 0.19599257558584213,
|
|
"num_tokens": 10749083.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.247275638580322,
|
|
"epoch": 5.052631578947368,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00031703186645187776,
|
|
"loss": 5.576,
|
|
"mean_token_accuracy": 0.19630726128816606,
|
|
"num_tokens": 10761215.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 6.119256210327149,
|
|
"epoch": 5.058479532163743,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031657128811666836,
|
|
"loss": 5.4844,
|
|
"mean_token_accuracy": 0.21457024216651915,
|
|
"num_tokens": 10772505.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 6.277840518951416,
|
|
"epoch": 5.064327485380117,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00031611052935908736,
|
|
"loss": 5.6873,
|
|
"mean_token_accuracy": 0.18565389662981033,
|
|
"num_tokens": 10785205.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 6.252067565917969,
|
|
"epoch": 5.0701754385964914,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000315649592178861,
|
|
"loss": 5.5145,
|
|
"mean_token_accuracy": 0.19830695241689683,
|
|
"num_tokens": 10796632.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.279603910446167,
|
|
"epoch": 5.076023391812866,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003151884785764897,
|
|
"loss": 5.5904,
|
|
"mean_token_accuracy": 0.19124119430780412,
|
|
"num_tokens": 10809466.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.215873050689697,
|
|
"epoch": 5.081871345029239,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003147271905532394,
|
|
"loss": 5.6253,
|
|
"mean_token_accuracy": 0.19205774813890458,
|
|
"num_tokens": 10821894.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 6.213073062896728,
|
|
"epoch": 5.087719298245614,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003142657301111333,
|
|
"loss": 5.5619,
|
|
"mean_token_accuracy": 0.19304760694503784,
|
|
"num_tokens": 10833612.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 6.240942192077637,
|
|
"epoch": 5.093567251461988,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003138040992529428,
|
|
"loss": 5.6159,
|
|
"mean_token_accuracy": 0.1916877120733261,
|
|
"num_tokens": 10845146.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.222587060928345,
|
|
"epoch": 5.099415204678363,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000313342299982179,
|
|
"loss": 5.6307,
|
|
"mean_token_accuracy": 0.19599018543958663,
|
|
"num_tokens": 10857376.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 6.268917751312256,
|
|
"epoch": 5.105263157894737,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003128803343030836,
|
|
"loss": 5.6174,
|
|
"mean_token_accuracy": 0.19013683646917343,
|
|
"num_tokens": 10868778.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 6.204213094711304,
|
|
"epoch": 5.111111111111111,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00031241820422062117,
|
|
"loss": 5.6416,
|
|
"mean_token_accuracy": 0.18884945660829544,
|
|
"num_tokens": 10881050.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 6.24682583808899,
|
|
"epoch": 5.116959064327485,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031195591174046914,
|
|
"loss": 5.6582,
|
|
"mean_token_accuracy": 0.18958028107881547,
|
|
"num_tokens": 10894468.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 6.240440559387207,
|
|
"epoch": 5.12280701754386,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031149345886901,
|
|
"loss": 5.5799,
|
|
"mean_token_accuracy": 0.20274145454168319,
|
|
"num_tokens": 10907039.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 6.199896383285522,
|
|
"epoch": 5.128654970760234,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003110308476133225,
|
|
"loss": 5.5892,
|
|
"mean_token_accuracy": 0.19300804436206817,
|
|
"num_tokens": 10918772.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 6.15690598487854,
|
|
"epoch": 5.1345029239766085,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031056807998117256,
|
|
"loss": 5.5346,
|
|
"mean_token_accuracy": 0.20170521587133408,
|
|
"num_tokens": 10931397.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 6.2687255382537845,
|
|
"epoch": 5.140350877192983,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000310105157981005,
|
|
"loss": 5.6213,
|
|
"mean_token_accuracy": 0.19060459733009338,
|
|
"num_tokens": 10942728.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 6.224508428573609,
|
|
"epoch": 5.146198830409356,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00030964208362193436,
|
|
"loss": 5.5596,
|
|
"mean_token_accuracy": 0.19377126395702363,
|
|
"num_tokens": 10954277.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 6.170745706558227,
|
|
"epoch": 5.152046783625731,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00030917885891373664,
|
|
"loss": 5.5925,
|
|
"mean_token_accuracy": 0.18776555061340333,
|
|
"num_tokens": 10966894.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 6.220528650283813,
|
|
"epoch": 5.157894736842105,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003087154858668404,
|
|
"loss": 5.5809,
|
|
"mean_token_accuracy": 0.1970152586698532,
|
|
"num_tokens": 10979229.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 6.2376940727233885,
|
|
"epoch": 5.16374269005848,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003082519664923177,
|
|
"loss": 5.6413,
|
|
"mean_token_accuracy": 0.19486669898033143,
|
|
"num_tokens": 10991999.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 6.247031164169312,
|
|
"epoch": 5.169590643274854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000307788302801876,
|
|
"loss": 5.5933,
|
|
"mean_token_accuracy": 0.18969963490962982,
|
|
"num_tokens": 11004408.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 6.1813554763793945,
|
|
"epoch": 5.175438596491228,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00030732449680784886,
|
|
"loss": 5.6699,
|
|
"mean_token_accuracy": 0.18906169086694719,
|
|
"num_tokens": 11016322.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 6.253483295440674,
|
|
"epoch": 5.181286549707602,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003068605505231877,
|
|
"loss": 5.6434,
|
|
"mean_token_accuracy": 0.1938949391245842,
|
|
"num_tokens": 11028798.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 6.139987373352051,
|
|
"epoch": 5.187134502923977,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00030639646596145243,
|
|
"loss": 5.4956,
|
|
"mean_token_accuracy": 0.1991927981376648,
|
|
"num_tokens": 11039837.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 6.188649845123291,
|
|
"epoch": 5.192982456140351,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00030593224513680357,
|
|
"loss": 5.5467,
|
|
"mean_token_accuracy": 0.19465117901563644,
|
|
"num_tokens": 11052943.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 6.282036399841308,
|
|
"epoch": 5.1988304093567255,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003054678900639927,
|
|
"loss": 5.6184,
|
|
"mean_token_accuracy": 0.18871409296989441,
|
|
"num_tokens": 11065483.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 6.201414966583252,
|
|
"epoch": 5.204678362573099,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00030500340275835405,
|
|
"loss": 5.5968,
|
|
"mean_token_accuracy": 0.19642669707536697,
|
|
"num_tokens": 11077743.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 6.180878305435181,
|
|
"epoch": 5.2105263157894735,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00030453878523579596,
|
|
"loss": 5.6104,
|
|
"mean_token_accuracy": 0.19137471318244934,
|
|
"num_tokens": 11089692.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 6.121175813674927,
|
|
"epoch": 5.216374269005848,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00030407403951279174,
|
|
"loss": 5.4548,
|
|
"mean_token_accuracy": 0.20713042765855788,
|
|
"num_tokens": 11104220.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 6.204579830169678,
|
|
"epoch": 5.222222222222222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003036091676063711,
|
|
"loss": 5.5689,
|
|
"mean_token_accuracy": 0.1917878121137619,
|
|
"num_tokens": 11115599.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 6.169919824600219,
|
|
"epoch": 5.228070175438597,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003031441715341115,
|
|
"loss": 5.5652,
|
|
"mean_token_accuracy": 0.1940259337425232,
|
|
"num_tokens": 11128125.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 6.204525756835937,
|
|
"epoch": 5.23391812865497,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003026790533141293,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.20134678184986116,
|
|
"num_tokens": 11140105.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 6.14801173210144,
|
|
"epoch": 5.239766081871345,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003022138149650708,
|
|
"loss": 5.5508,
|
|
"mean_token_accuracy": 0.20007871240377426,
|
|
"num_tokens": 11152291.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 6.149573564529419,
|
|
"epoch": 5.245614035087719,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003017484585061039,
|
|
"loss": 5.6043,
|
|
"mean_token_accuracy": 0.20203853994607926,
|
|
"num_tokens": 11166330.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 6.270491266250611,
|
|
"epoch": 5.251461988304094,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00030128298595690905,
|
|
"loss": 5.5693,
|
|
"mean_token_accuracy": 0.19781578928232194,
|
|
"num_tokens": 11178737.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 6.1815088272094725,
|
|
"epoch": 5.257309941520468,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00030081739933767034,
|
|
"loss": 5.6075,
|
|
"mean_token_accuracy": 0.19342361837625505,
|
|
"num_tokens": 11191179.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 6.157884359359741,
|
|
"epoch": 5.2631578947368425,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003003517006690672,
|
|
"loss": 5.5795,
|
|
"mean_token_accuracy": 0.19182905405759812,
|
|
"num_tokens": 11203890.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 5.2631578947368425,
|
|
"eval_entropy": 6.126187722591524,
|
|
"eval_loss": 6.545482635498047,
|
|
"eval_mean_token_accuracy": 0.14949914047104418,
|
|
"eval_num_tokens": 11203890.0,
|
|
"eval_runtime": 1.7358,
|
|
"eval_samples_per_second": 1559.502,
|
|
"eval_steps_per_second": 195.298,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 6.203334045410156,
|
|
"epoch": 5.269005847953216,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00029988589197226515,
|
|
"loss": 5.5848,
|
|
"mean_token_accuracy": 0.19501378387212753,
|
|
"num_tokens": 11216556.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 6.209409761428833,
|
|
"epoch": 5.2748538011695905,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00029941997526890745,
|
|
"loss": 5.5742,
|
|
"mean_token_accuracy": 0.19800327569246293,
|
|
"num_tokens": 11228352.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 6.18704080581665,
|
|
"epoch": 5.280701754385965,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000298953952581106,
|
|
"loss": 5.5866,
|
|
"mean_token_accuracy": 0.19371091425418854,
|
|
"num_tokens": 11239423.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 6.182998895645142,
|
|
"epoch": 5.286549707602339,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029848782593143246,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.19329002350568772,
|
|
"num_tokens": 11251117.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 6.184751319885254,
|
|
"epoch": 5.292397660818714,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00029802159734291015,
|
|
"loss": 5.6029,
|
|
"mean_token_accuracy": 0.1949610009789467,
|
|
"num_tokens": 11264459.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 6.122611618041992,
|
|
"epoch": 5.298245614035087,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002975552688390045,
|
|
"loss": 5.4946,
|
|
"mean_token_accuracy": 0.20606748759746552,
|
|
"num_tokens": 11276020.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 6.225433206558227,
|
|
"epoch": 5.304093567251462,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029708884244361474,
|
|
"loss": 5.5673,
|
|
"mean_token_accuracy": 0.19093938171863556,
|
|
"num_tokens": 11287412.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 6.2020775318145756,
|
|
"epoch": 5.309941520467836,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00029662232018106473,
|
|
"loss": 5.6391,
|
|
"mean_token_accuracy": 0.1928391069173813,
|
|
"num_tokens": 11298607.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.257556915283203,
|
|
"epoch": 5.315789473684211,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00029615570407609467,
|
|
"loss": 5.6883,
|
|
"mean_token_accuracy": 0.187696236371994,
|
|
"num_tokens": 11310943.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 6.142689752578735,
|
|
"epoch": 5.321637426900585,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00029568899615385174,
|
|
"loss": 5.5356,
|
|
"mean_token_accuracy": 0.19334274381399155,
|
|
"num_tokens": 11323684.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 6.092800569534302,
|
|
"epoch": 5.3274853801169595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000295222198439882,
|
|
"loss": 5.5264,
|
|
"mean_token_accuracy": 0.20197111666202544,
|
|
"num_tokens": 11336817.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.266628980636597,
|
|
"epoch": 5.333333333333333,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002947553129601209,
|
|
"loss": 5.6901,
|
|
"mean_token_accuracy": 0.184974829852581,
|
|
"num_tokens": 11349246.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 6.21450777053833,
|
|
"epoch": 5.3391812865497075,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002942883417408849,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.1952193409204483,
|
|
"num_tokens": 11360995.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 6.122638463973999,
|
|
"epoch": 5.345029239766082,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002938212868088626,
|
|
"loss": 5.5252,
|
|
"mean_token_accuracy": 0.20124758630990983,
|
|
"num_tokens": 11373713.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 6.194461154937744,
|
|
"epoch": 5.350877192982456,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00029335415019110593,
|
|
"loss": 5.5915,
|
|
"mean_token_accuracy": 0.1929761439561844,
|
|
"num_tokens": 11386532.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 6.1732141971588135,
|
|
"epoch": 5.356725146198831,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002928869339150212,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.20498475432395935,
|
|
"num_tokens": 11400535.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 6.139449977874756,
|
|
"epoch": 5.362573099415204,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002924196400083607,
|
|
"loss": 5.569,
|
|
"mean_token_accuracy": 0.19922028332948685,
|
|
"num_tokens": 11412577.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 6.157861709594727,
|
|
"epoch": 5.368421052631579,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00029195227049921336,
|
|
"loss": 5.5745,
|
|
"mean_token_accuracy": 0.19396987408399582,
|
|
"num_tokens": 11427142.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 6.251806020736694,
|
|
"epoch": 5.374269005847953,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00029148482741599634,
|
|
"loss": 5.6222,
|
|
"mean_token_accuracy": 0.1967879682779312,
|
|
"num_tokens": 11441319.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 6.145890092849731,
|
|
"epoch": 5.380116959064328,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00029101731278744636,
|
|
"loss": 5.5517,
|
|
"mean_token_accuracy": 0.19568183422088622,
|
|
"num_tokens": 11453947.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 6.184706401824951,
|
|
"epoch": 5.385964912280702,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002905497286426102,
|
|
"loss": 5.5663,
|
|
"mean_token_accuracy": 0.19261671006679534,
|
|
"num_tokens": 11465463.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 6.2439820766448975,
|
|
"epoch": 5.391812865497076,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00029008207701083677,
|
|
"loss": 5.5994,
|
|
"mean_token_accuracy": 0.19196502715349198,
|
|
"num_tokens": 11477249.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 6.232153797149659,
|
|
"epoch": 5.39766081871345,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002896143599217674,
|
|
"loss": 5.648,
|
|
"mean_token_accuracy": 0.19223932474851607,
|
|
"num_tokens": 11490184.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 6.164693021774292,
|
|
"epoch": 5.4035087719298245,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002891465794053282,
|
|
"loss": 5.6627,
|
|
"mean_token_accuracy": 0.19471751302480697,
|
|
"num_tokens": 11502935.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 6.185873222351074,
|
|
"epoch": 5.409356725146199,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002886787374917198,
|
|
"loss": 5.6102,
|
|
"mean_token_accuracy": 0.19907134473323823,
|
|
"num_tokens": 11515385.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 6.232328748703003,
|
|
"epoch": 5.415204678362573,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00028821083621140986,
|
|
"loss": 5.6757,
|
|
"mean_token_accuracy": 0.19022065550088882,
|
|
"num_tokens": 11528786.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 6.18293981552124,
|
|
"epoch": 5.421052631578947,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002877428775951234,
|
|
"loss": 5.6813,
|
|
"mean_token_accuracy": 0.1846669003367424,
|
|
"num_tokens": 11540401.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 6.2166259765625,
|
|
"epoch": 5.426900584795321,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002872748636738344,
|
|
"loss": 5.6221,
|
|
"mean_token_accuracy": 0.19119976609945297,
|
|
"num_tokens": 11552920.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 6.2669923305511475,
|
|
"epoch": 5.432748538011696,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002868067964787568,
|
|
"loss": 5.6921,
|
|
"mean_token_accuracy": 0.18187858015298844,
|
|
"num_tokens": 11566237.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 6.180300045013428,
|
|
"epoch": 5.43859649122807,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00028633867804133583,
|
|
"loss": 5.5902,
|
|
"mean_token_accuracy": 0.19248692095279693,
|
|
"num_tokens": 11577380.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.179739332199096,
|
|
"epoch": 5.444444444444445,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002858705103932391,
|
|
"loss": 5.6199,
|
|
"mean_token_accuracy": 0.18746355175971985,
|
|
"num_tokens": 11590708.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 6.252267789840698,
|
|
"epoch": 5.450292397660819,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028540229556634785,
|
|
"loss": 5.659,
|
|
"mean_token_accuracy": 0.1918761685490608,
|
|
"num_tokens": 11603351.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 6.150949764251709,
|
|
"epoch": 5.456140350877193,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002849340355927477,
|
|
"loss": 5.5995,
|
|
"mean_token_accuracy": 0.19755121916532517,
|
|
"num_tokens": 11615675.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 6.211690902709961,
|
|
"epoch": 5.461988304093567,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002844657325047208,
|
|
"loss": 5.6188,
|
|
"mean_token_accuracy": 0.1901940166950226,
|
|
"num_tokens": 11628184.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 6.248294639587402,
|
|
"epoch": 5.4678362573099415,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002839973883347361,
|
|
"loss": 5.6239,
|
|
"mean_token_accuracy": 0.1938335880637169,
|
|
"num_tokens": 11641058.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 6.046961641311645,
|
|
"epoch": 5.473684210526316,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00028352900511544076,
|
|
"loss": 5.5301,
|
|
"mean_token_accuracy": 0.20218219608068466,
|
|
"num_tokens": 11653959.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 6.189428520202637,
|
|
"epoch": 5.47953216374269,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028306058487965156,
|
|
"loss": 5.6244,
|
|
"mean_token_accuracy": 0.1923079714179039,
|
|
"num_tokens": 11665241.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 6.269627046585083,
|
|
"epoch": 5.485380116959064,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002825921296603461,
|
|
"loss": 5.6436,
|
|
"mean_token_accuracy": 0.19317378997802734,
|
|
"num_tokens": 11678670.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 6.176467990875244,
|
|
"epoch": 5.491228070175438,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028212364149065347,
|
|
"loss": 5.5591,
|
|
"mean_token_accuracy": 0.19447817653417587,
|
|
"num_tokens": 11689964.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 6.212436294555664,
|
|
"epoch": 5.497076023391813,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002816551224038458,
|
|
"loss": 5.6311,
|
|
"mean_token_accuracy": 0.19270961433649064,
|
|
"num_tokens": 11701895.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 6.196808242797852,
|
|
"epoch": 5.502923976608187,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00028118657443332984,
|
|
"loss": 5.6125,
|
|
"mean_token_accuracy": 0.1931144431233406,
|
|
"num_tokens": 11714335.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 6.203803539276123,
|
|
"epoch": 5.508771929824562,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002807179996126371,
|
|
"loss": 5.586,
|
|
"mean_token_accuracy": 0.19787362366914749,
|
|
"num_tokens": 11726828.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 6.161632442474366,
|
|
"epoch": 5.514619883040936,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00028024939997541606,
|
|
"loss": 5.5205,
|
|
"mean_token_accuracy": 0.19903392493724822,
|
|
"num_tokens": 11740729.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 6.187909173965454,
|
|
"epoch": 5.52046783625731,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00027978077755542264,
|
|
"loss": 5.6438,
|
|
"mean_token_accuracy": 0.1880514517426491,
|
|
"num_tokens": 11753017.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.236352014541626,
|
|
"epoch": 5.526315789473684,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002793121343865119,
|
|
"loss": 5.5437,
|
|
"mean_token_accuracy": 0.19818583726882935,
|
|
"num_tokens": 11766139.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 6.0751176357269285,
|
|
"epoch": 5.5321637426900585,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002788434725026286,
|
|
"loss": 5.4524,
|
|
"mean_token_accuracy": 0.20898132920265197,
|
|
"num_tokens": 11780315.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 6.19849910736084,
|
|
"epoch": 5.538011695906433,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000278374793937799,
|
|
"loss": 5.5701,
|
|
"mean_token_accuracy": 0.20106644183397293,
|
|
"num_tokens": 11792342.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 6.139348077774048,
|
|
"epoch": 5.543859649122807,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00027790610072612174,
|
|
"loss": 5.5814,
|
|
"mean_token_accuracy": 0.19864000827074052,
|
|
"num_tokens": 11804587.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 6.204433631896973,
|
|
"epoch": 5.549707602339181,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00027743739490175903,
|
|
"loss": 5.6153,
|
|
"mean_token_accuracy": 0.1962747573852539,
|
|
"num_tokens": 11815884.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 6.2616087913513185,
|
|
"epoch": 5.555555555555555,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002769686784989276,
|
|
"loss": 5.6858,
|
|
"mean_token_accuracy": 0.1906959295272827,
|
|
"num_tokens": 11827279.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 6.227077436447144,
|
|
"epoch": 5.56140350877193,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00027649995355189026,
|
|
"loss": 5.5778,
|
|
"mean_token_accuracy": 0.1917523667216301,
|
|
"num_tokens": 11838646.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 6.2128472328186035,
|
|
"epoch": 5.567251461988304,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000276031222094947,
|
|
"loss": 5.6226,
|
|
"mean_token_accuracy": 0.19587834030389786,
|
|
"num_tokens": 11850565.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 6.007431507110596,
|
|
"epoch": 5.573099415204679,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002755624861624259,
|
|
"loss": 5.4694,
|
|
"mean_token_accuracy": 0.21455187499523162,
|
|
"num_tokens": 11863244.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 6.1172261238098145,
|
|
"epoch": 5.578947368421053,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002750937477886745,
|
|
"loss": 5.5394,
|
|
"mean_token_accuracy": 0.20348394066095352,
|
|
"num_tokens": 11876884.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 6.202335929870605,
|
|
"epoch": 5.584795321637427,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000274625009008051,
|
|
"loss": 5.5633,
|
|
"mean_token_accuracy": 0.1994699239730835,
|
|
"num_tokens": 11889624.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 6.199597454071045,
|
|
"epoch": 5.590643274853801,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002741562718549152,
|
|
"loss": 5.6146,
|
|
"mean_token_accuracy": 0.1903340846300125,
|
|
"num_tokens": 11901304.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 6.045202445983887,
|
|
"epoch": 5.5964912280701755,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027368753836362023,
|
|
"loss": 5.4275,
|
|
"mean_token_accuracy": 0.2090636283159256,
|
|
"num_tokens": 11913188.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 6.205039834976196,
|
|
"epoch": 5.60233918128655,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002732188105685027,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.1901662066578865,
|
|
"num_tokens": 11926771.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 6.189703035354614,
|
|
"epoch": 5.6081871345029235,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002727500905038752,
|
|
"loss": 5.6292,
|
|
"mean_token_accuracy": 0.19206828624010086,
|
|
"num_tokens": 11939162.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.265229558944702,
|
|
"epoch": 5.614035087719298,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002722813802040161,
|
|
"loss": 5.6899,
|
|
"mean_token_accuracy": 0.1824260324239731,
|
|
"num_tokens": 11952376.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 6.153215503692627,
|
|
"epoch": 5.619883040935672,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00027181268170316193,
|
|
"loss": 5.4898,
|
|
"mean_token_accuracy": 0.20416971743106843,
|
|
"num_tokens": 11965154.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 6.109413433074951,
|
|
"epoch": 5.625730994152047,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002713439970354978,
|
|
"loss": 5.5559,
|
|
"mean_token_accuracy": 0.20152464509010315,
|
|
"num_tokens": 11978492.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 6.183708906173706,
|
|
"epoch": 5.631578947368421,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002708753282351486,
|
|
"loss": 5.623,
|
|
"mean_token_accuracy": 0.19616713076829911,
|
|
"num_tokens": 11990603.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 6.206140184402466,
|
|
"epoch": 5.637426900584796,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002704066773361706,
|
|
"loss": 5.6588,
|
|
"mean_token_accuracy": 0.18772298395633696,
|
|
"num_tokens": 12002608.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 6.206177997589111,
|
|
"epoch": 5.643274853801169,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002699380463725425,
|
|
"loss": 5.6693,
|
|
"mean_token_accuracy": 0.19320938289165496,
|
|
"num_tokens": 12016112.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 6.208219861984253,
|
|
"epoch": 5.649122807017544,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000269469437378156,
|
|
"loss": 5.6782,
|
|
"mean_token_accuracy": 0.18606738448143006,
|
|
"num_tokens": 12027685.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 6.157080554962159,
|
|
"epoch": 5.654970760233918,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00026900085238680787,
|
|
"loss": 5.5265,
|
|
"mean_token_accuracy": 0.20031696707010269,
|
|
"num_tokens": 12040524.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 6.136253547668457,
|
|
"epoch": 5.6608187134502925,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002685322934321907,
|
|
"loss": 5.584,
|
|
"mean_token_accuracy": 0.19754060506820678,
|
|
"num_tokens": 12052841.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 6.207145261764526,
|
|
"epoch": 5.666666666666667,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002680637625478837,
|
|
"loss": 5.6432,
|
|
"mean_token_accuracy": 0.1860191196203232,
|
|
"num_tokens": 12065375.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 6.252925825119019,
|
|
"epoch": 5.6725146198830405,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002675952617673446,
|
|
"loss": 5.6401,
|
|
"mean_token_accuracy": 0.19413762390613556,
|
|
"num_tokens": 12078034.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 6.234569644927978,
|
|
"epoch": 5.678362573099415,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002671267931239006,
|
|
"loss": 5.6741,
|
|
"mean_token_accuracy": 0.1951745018362999,
|
|
"num_tokens": 12090218.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 6.174238300323486,
|
|
"epoch": 5.684210526315789,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026665835865073896,
|
|
"loss": 5.6091,
|
|
"mean_token_accuracy": 0.19040821194648744,
|
|
"num_tokens": 12101927.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 6.2493030548095705,
|
|
"epoch": 5.690058479532164,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002661899603808989,
|
|
"loss": 5.6542,
|
|
"mean_token_accuracy": 0.18802524656057357,
|
|
"num_tokens": 12114887.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 6.253974771499633,
|
|
"epoch": 5.695906432748538,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002657216003472626,
|
|
"loss": 5.6501,
|
|
"mean_token_accuracy": 0.18921741098165512,
|
|
"num_tokens": 12127151.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 6.217792081832886,
|
|
"epoch": 5.701754385964913,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00026525328058254604,
|
|
"loss": 5.6387,
|
|
"mean_token_accuracy": 0.20001572966575623,
|
|
"num_tokens": 12141987.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 6.12281756401062,
|
|
"epoch": 5.707602339181286,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026478500311929076,
|
|
"loss": 5.5294,
|
|
"mean_token_accuracy": 0.19793105721473694,
|
|
"num_tokens": 12154525.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 6.090994453430175,
|
|
"epoch": 5.713450292397661,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026431676998985444,
|
|
"loss": 5.4717,
|
|
"mean_token_accuracy": 0.2034291595220566,
|
|
"num_tokens": 12166599.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 6.197245168685913,
|
|
"epoch": 5.719298245614035,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002638485832264024,
|
|
"loss": 5.6933,
|
|
"mean_token_accuracy": 0.1911052107810974,
|
|
"num_tokens": 12178658.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 6.153663253784179,
|
|
"epoch": 5.7251461988304095,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026338044486089895,
|
|
"loss": 5.5369,
|
|
"mean_token_accuracy": 0.19749022871255875,
|
|
"num_tokens": 12190311.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 6.235907506942749,
|
|
"epoch": 5.730994152046784,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026291235692509795,
|
|
"loss": 5.6485,
|
|
"mean_token_accuracy": 0.1873551309108734,
|
|
"num_tokens": 12202443.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 6.212292098999024,
|
|
"epoch": 5.7368421052631575,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026244432145053467,
|
|
"loss": 5.6776,
|
|
"mean_token_accuracy": 0.18599846512079238,
|
|
"num_tokens": 12214245.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 6.146491670608521,
|
|
"epoch": 5.742690058479532,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002619763404685167,
|
|
"loss": 5.5899,
|
|
"mean_token_accuracy": 0.19894772320985793,
|
|
"num_tokens": 12226752.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 6.21858696937561,
|
|
"epoch": 5.748538011695906,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026150841601011495,
|
|
"loss": 5.6532,
|
|
"mean_token_accuracy": 0.1928119033575058,
|
|
"num_tokens": 12239043.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 6.148224353790283,
|
|
"epoch": 5.754385964912281,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002610405501061552,
|
|
"loss": 5.6374,
|
|
"mean_token_accuracy": 0.19564854651689528,
|
|
"num_tokens": 12250799.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 6.260662412643432,
|
|
"epoch": 5.760233918128655,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002605727447872091,
|
|
"loss": 5.7658,
|
|
"mean_token_accuracy": 0.17934016585350038,
|
|
"num_tokens": 12264503.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 6.212125062942505,
|
|
"epoch": 5.76608187134503,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002601050020835851,
|
|
"loss": 5.6275,
|
|
"mean_token_accuracy": 0.19414414912462236,
|
|
"num_tokens": 12276106.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 6.166960191726685,
|
|
"epoch": 5.771929824561403,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00025963732402532026,
|
|
"loss": 5.5725,
|
|
"mean_token_accuracy": 0.19277427047491075,
|
|
"num_tokens": 12288322.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 6.174505043029785,
|
|
"epoch": 5.777777777777778,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002591697126421707,
|
|
"loss": 5.6103,
|
|
"mean_token_accuracy": 0.19324022829532622,
|
|
"num_tokens": 12301059.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 6.195170164108276,
|
|
"epoch": 5.783625730994152,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002587021699636035,
|
|
"loss": 5.6547,
|
|
"mean_token_accuracy": 0.19215726405382155,
|
|
"num_tokens": 12313206.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 6.187473487854004,
|
|
"epoch": 5.7894736842105265,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00025823469801878734,
|
|
"loss": 5.5865,
|
|
"mean_token_accuracy": 0.2003424197435379,
|
|
"num_tokens": 12325426.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 6.184348058700562,
|
|
"epoch": 5.7953216374269,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00025776729883658406,
|
|
"loss": 5.6519,
|
|
"mean_token_accuracy": 0.19116644114255904,
|
|
"num_tokens": 12338683.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 6.149820327758789,
|
|
"epoch": 5.8011695906432745,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002572999744455395,
|
|
"loss": 5.6143,
|
|
"mean_token_accuracy": 0.20073840469121934,
|
|
"num_tokens": 12351139.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 6.156563854217529,
|
|
"epoch": 5.807017543859649,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00025683272687387513,
|
|
"loss": 5.5273,
|
|
"mean_token_accuracy": 0.2006761685013771,
|
|
"num_tokens": 12362804.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 6.275007486343384,
|
|
"epoch": 5.812865497076023,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002563655581494789,
|
|
"loss": 5.7075,
|
|
"mean_token_accuracy": 0.18801289349794387,
|
|
"num_tokens": 12375069.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 6.2143950939178465,
|
|
"epoch": 5.818713450292398,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00025589847029989653,
|
|
"loss": 5.5962,
|
|
"mean_token_accuracy": 0.19170916378498076,
|
|
"num_tokens": 12387434.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 6.160259151458741,
|
|
"epoch": 5.824561403508772,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00025543146535232295,
|
|
"loss": 5.6324,
|
|
"mean_token_accuracy": 0.19126828014850616,
|
|
"num_tokens": 12400141.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 6.158847045898438,
|
|
"epoch": 5.830409356725146,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00025496454533359284,
|
|
"loss": 5.5932,
|
|
"mean_token_accuracy": 0.1964272677898407,
|
|
"num_tokens": 12412619.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 6.1897941589355465,
|
|
"epoch": 5.83625730994152,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00025449771227017265,
|
|
"loss": 5.6052,
|
|
"mean_token_accuracy": 0.19122326076030732,
|
|
"num_tokens": 12425312.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 6.212467002868652,
|
|
"epoch": 5.842105263157895,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002540309681881516,
|
|
"loss": 5.6369,
|
|
"mean_token_accuracy": 0.1920468732714653,
|
|
"num_tokens": 12436972.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 6.25763521194458,
|
|
"epoch": 5.847953216374269,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002535643151132321,
|
|
"loss": 5.6689,
|
|
"mean_token_accuracy": 0.18332580476999283,
|
|
"num_tokens": 12448543.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 5.847953216374269,
|
|
"eval_entropy": 6.097200781898161,
|
|
"eval_loss": 6.5179829597473145,
|
|
"eval_mean_token_accuracy": 0.1515029760236937,
|
|
"eval_num_tokens": 12448543.0,
|
|
"eval_runtime": 1.7382,
|
|
"eval_samples_per_second": 1557.392,
|
|
"eval_steps_per_second": 195.034,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 6.1970555782318115,
|
|
"epoch": 5.853801169590644,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00025309775507072234,
|
|
"loss": 5.5995,
|
|
"mean_token_accuracy": 0.18860557228326796,
|
|
"num_tokens": 12462292.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 6.208127212524414,
|
|
"epoch": 5.859649122807017,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002526312900855261,
|
|
"loss": 5.7006,
|
|
"mean_token_accuracy": 0.18534910380840303,
|
|
"num_tokens": 12475930.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 6.248155879974365,
|
|
"epoch": 5.8654970760233915,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002521649221821349,
|
|
"loss": 5.7047,
|
|
"mean_token_accuracy": 0.1884417250752449,
|
|
"num_tokens": 12487769.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 6.188259935379028,
|
|
"epoch": 5.871345029239766,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00025169865338461894,
|
|
"loss": 5.6123,
|
|
"mean_token_accuracy": 0.2000778630375862,
|
|
"num_tokens": 12500116.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 6.114431619644165,
|
|
"epoch": 5.87719298245614,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002512324857166182,
|
|
"loss": 5.496,
|
|
"mean_token_accuracy": 0.2063222736120224,
|
|
"num_tokens": 12511896.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 6.124118614196777,
|
|
"epoch": 5.883040935672515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002507664212013337,
|
|
"loss": 5.5847,
|
|
"mean_token_accuracy": 0.19256978929042817,
|
|
"num_tokens": 12524435.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 6.219422721862793,
|
|
"epoch": 5.888888888888889,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025030046186151915,
|
|
"loss": 5.6333,
|
|
"mean_token_accuracy": 0.18958206176757814,
|
|
"num_tokens": 12536315.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 6.217531204223633,
|
|
"epoch": 5.894736842105263,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002498346097194712,
|
|
"loss": 5.668,
|
|
"mean_token_accuracy": 0.1930579051375389,
|
|
"num_tokens": 12548982.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 6.253390645980835,
|
|
"epoch": 5.900584795321637,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00024936886679702167,
|
|
"loss": 5.7614,
|
|
"mean_token_accuracy": 0.18307951092720032,
|
|
"num_tokens": 12562746.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 6.210511875152588,
|
|
"epoch": 5.906432748538012,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00024890323511552825,
|
|
"loss": 5.6686,
|
|
"mean_token_accuracy": 0.19121008962392808,
|
|
"num_tokens": 12575054.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 6.249647378921509,
|
|
"epoch": 5.912280701754386,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00024843771669586575,
|
|
"loss": 5.7036,
|
|
"mean_token_accuracy": 0.18693224489688873,
|
|
"num_tokens": 12588636.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 6.1801611423492435,
|
|
"epoch": 5.918128654970761,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002479723135584176,
|
|
"loss": 5.6321,
|
|
"mean_token_accuracy": 0.187230284512043,
|
|
"num_tokens": 12602522.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 6.212598085403442,
|
|
"epoch": 5.923976608187134,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002475070277230668,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.1926732540130615,
|
|
"num_tokens": 12616367.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 6.173742055892944,
|
|
"epoch": 5.9298245614035086,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002470418612091871,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.1956544980406761,
|
|
"num_tokens": 12628469.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 6.172438287734986,
|
|
"epoch": 5.935672514619883,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002465768160356347,
|
|
"loss": 5.5091,
|
|
"mean_token_accuracy": 0.19853661954402924,
|
|
"num_tokens": 12640506.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 6.153302335739136,
|
|
"epoch": 5.941520467836257,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002461118942207388,
|
|
"loss": 5.6734,
|
|
"mean_token_accuracy": 0.19122968018054962,
|
|
"num_tokens": 12652862.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 6.1525979995727536,
|
|
"epoch": 5.947368421052632,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024564709778229353,
|
|
"loss": 5.5751,
|
|
"mean_token_accuracy": 0.1945019006729126,
|
|
"num_tokens": 12664390.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 6.114281463623047,
|
|
"epoch": 5.953216374269006,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00024518242873754887,
|
|
"loss": 5.5371,
|
|
"mean_token_accuracy": 0.2019401714205742,
|
|
"num_tokens": 12676489.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 6.159471130371093,
|
|
"epoch": 5.95906432748538,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024471788910320163,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.19796611368656158,
|
|
"num_tokens": 12688666.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 6.203460788726806,
|
|
"epoch": 5.964912280701754,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00024425348089538714,
|
|
"loss": 5.6324,
|
|
"mean_token_accuracy": 0.19040021449327468,
|
|
"num_tokens": 12702159.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 6.2263847351074215,
|
|
"epoch": 5.970760233918129,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002437892061296705,
|
|
"loss": 5.6613,
|
|
"mean_token_accuracy": 0.19145021289587022,
|
|
"num_tokens": 12715528.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 6.213036918640137,
|
|
"epoch": 5.976608187134503,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024332506682103722,
|
|
"loss": 5.7221,
|
|
"mean_token_accuracy": 0.18797306567430497,
|
|
"num_tokens": 12727485.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 6.119540500640869,
|
|
"epoch": 5.982456140350877,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00024286106498388533,
|
|
"loss": 5.5956,
|
|
"mean_token_accuracy": 0.19916785806417464,
|
|
"num_tokens": 12740602.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 6.2100386142730715,
|
|
"epoch": 5.988304093567251,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002423972026320162,
|
|
"loss": 5.6559,
|
|
"mean_token_accuracy": 0.19153142720460892,
|
|
"num_tokens": 12752599.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 6.248219776153564,
|
|
"epoch": 5.994152046783626,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00024193348177862551,
|
|
"loss": 5.6551,
|
|
"mean_token_accuracy": 0.19575471580028533,
|
|
"num_tokens": 12764515.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 6.193612670898437,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00024146990443629508,
|
|
"loss": 5.5819,
|
|
"mean_token_accuracy": 0.19736979752779008,
|
|
"num_tokens": 12776904.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 6.267667579650879,
|
|
"epoch": 6.005847953216374,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002410064726169838,
|
|
"loss": 5.5486,
|
|
"mean_token_accuracy": 0.20088689625263215,
|
|
"num_tokens": 12789313.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 6.12780909538269,
|
|
"epoch": 6.011695906432749,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024054318833201894,
|
|
"loss": 5.4196,
|
|
"mean_token_accuracy": 0.20639814734458922,
|
|
"num_tokens": 12801932.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 6.194341468811035,
|
|
"epoch": 6.017543859649122,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024008005359208774,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.20510816276073457,
|
|
"num_tokens": 12814530.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 6.215360927581787,
|
|
"epoch": 6.023391812865497,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00023961707040722792,
|
|
"loss": 5.5231,
|
|
"mean_token_accuracy": 0.19230549186468124,
|
|
"num_tokens": 12827053.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 6.1605018138885494,
|
|
"epoch": 6.029239766081871,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023915424078681987,
|
|
"loss": 5.3988,
|
|
"mean_token_accuracy": 0.20488255321979523,
|
|
"num_tokens": 12838875.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 6.2169623374938965,
|
|
"epoch": 6.035087719298246,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023869156673957733,
|
|
"loss": 5.4976,
|
|
"mean_token_accuracy": 0.19581529945135118,
|
|
"num_tokens": 12851434.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 6.189522981643677,
|
|
"epoch": 6.04093567251462,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00023822905027353863,
|
|
"loss": 5.4966,
|
|
"mean_token_accuracy": 0.199859419465065,
|
|
"num_tokens": 12864275.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 6.155936002731323,
|
|
"epoch": 6.046783625730994,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002377666933960586,
|
|
"loss": 5.4468,
|
|
"mean_token_accuracy": 0.20417249649763108,
|
|
"num_tokens": 12876675.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 6.077428960800171,
|
|
"epoch": 6.052631578947368,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002373044981137994,
|
|
"loss": 5.342,
|
|
"mean_token_accuracy": 0.22101518511772156,
|
|
"num_tokens": 12888482.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 6.151059532165528,
|
|
"epoch": 6.058479532163743,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023684246643272132,
|
|
"loss": 5.3777,
|
|
"mean_token_accuracy": 0.19979428499937057,
|
|
"num_tokens": 12901631.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 6.219945859909058,
|
|
"epoch": 6.064327485380117,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002363806003580752,
|
|
"loss": 5.5289,
|
|
"mean_token_accuracy": 0.19131574481725694,
|
|
"num_tokens": 12913210.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 6.1141581535339355,
|
|
"epoch": 6.0701754385964914,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023591890189439296,
|
|
"loss": 5.3585,
|
|
"mean_token_accuracy": 0.21086901128292085,
|
|
"num_tokens": 12924964.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 6.171627044677734,
|
|
"epoch": 6.076023391812866,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023545737304547893,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.2003111273050308,
|
|
"num_tokens": 12936986.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 6.284926128387451,
|
|
"epoch": 6.081871345029239,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002349960158144015,
|
|
"loss": 5.5329,
|
|
"mean_token_accuracy": 0.19234615862369536,
|
|
"num_tokens": 12949842.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 6.241711187362671,
|
|
"epoch": 6.087719298245614,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000234534832203484,
|
|
"loss": 5.5031,
|
|
"mean_token_accuracy": 0.1995372384786606,
|
|
"num_tokens": 12962112.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 6.105756568908691,
|
|
"epoch": 6.093567251461988,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002340738242142965,
|
|
"loss": 5.4059,
|
|
"mean_token_accuracy": 0.20801085680723191,
|
|
"num_tokens": 12973599.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 6.103172874450683,
|
|
"epoch": 6.099415204678363,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023361299384764677,
|
|
"loss": 5.3531,
|
|
"mean_token_accuracy": 0.21595710515975952,
|
|
"num_tokens": 12987402.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 6.087310838699341,
|
|
"epoch": 6.105263157894737,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00023315234310357147,
|
|
"loss": 5.368,
|
|
"mean_token_accuracy": 0.2067652463912964,
|
|
"num_tokens": 12999299.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 6.20887303352356,
|
|
"epoch": 6.111111111111111,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00023269187398132802,
|
|
"loss": 5.5,
|
|
"mean_token_accuracy": 0.19606976062059403,
|
|
"num_tokens": 13011911.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 6.1590488910675045,
|
|
"epoch": 6.116959064327485,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00023223158847938543,
|
|
"loss": 5.4638,
|
|
"mean_token_accuracy": 0.20242392420768737,
|
|
"num_tokens": 13023493.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 6.18142466545105,
|
|
"epoch": 6.12280701754386,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023177148859541565,
|
|
"loss": 5.4502,
|
|
"mean_token_accuracy": 0.20783536434173583,
|
|
"num_tokens": 13036364.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 6.121358346939087,
|
|
"epoch": 6.128654970760234,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023131157632628525,
|
|
"loss": 5.493,
|
|
"mean_token_accuracy": 0.2045218303799629,
|
|
"num_tokens": 13048809.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 6.136888217926026,
|
|
"epoch": 6.1345029239766085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023085185366804656,
|
|
"loss": 5.3809,
|
|
"mean_token_accuracy": 0.2071672111749649,
|
|
"num_tokens": 13061811.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 6.222945308685302,
|
|
"epoch": 6.140350877192983,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023039232261592868,
|
|
"loss": 5.5064,
|
|
"mean_token_accuracy": 0.1967059403657913,
|
|
"num_tokens": 13074331.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 6.162861776351929,
|
|
"epoch": 6.146198830409356,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022993298516432937,
|
|
"loss": 5.4459,
|
|
"mean_token_accuracy": 0.2127395898103714,
|
|
"num_tokens": 13087642.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 6.1941627025604244,
|
|
"epoch": 6.152046783625731,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022947384330680636,
|
|
"loss": 5.55,
|
|
"mean_token_accuracy": 0.19466835856437684,
|
|
"num_tokens": 13099453.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 6.198199653625489,
|
|
"epoch": 6.157894736842105,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022901489903606794,
|
|
"loss": 5.4343,
|
|
"mean_token_accuracy": 0.2046760395169258,
|
|
"num_tokens": 13112092.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 6.1285923480987545,
|
|
"epoch": 6.16374269005848,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00022855615434396527,
|
|
"loss": 5.3511,
|
|
"mean_token_accuracy": 0.2126012071967125,
|
|
"num_tokens": 13125023.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 6.136300468444825,
|
|
"epoch": 6.169590643274854,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002280976112214832,
|
|
"loss": 5.505,
|
|
"mean_token_accuracy": 0.20102546215057374,
|
|
"num_tokens": 13137902.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 6.089547443389892,
|
|
"epoch": 6.175438596491228,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022763927165873177,
|
|
"loss": 5.4134,
|
|
"mean_token_accuracy": 0.20973370522260665,
|
|
"num_tokens": 13151245.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 6.022973537445068,
|
|
"epoch": 6.181286549707602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002271811376449376,
|
|
"loss": 5.2909,
|
|
"mean_token_accuracy": 0.2159404218196869,
|
|
"num_tokens": 13163916.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 6.177674674987793,
|
|
"epoch": 6.187134502923977,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022672321116843497,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.20627960413694382,
|
|
"num_tokens": 13176265.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 6.154309225082398,
|
|
"epoch": 6.192982456140351,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00022626549421665788,
|
|
"loss": 5.4907,
|
|
"mean_token_accuracy": 0.19953753650188447,
|
|
"num_tokens": 13188916.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 6.1405637741088865,
|
|
"epoch": 6.1988304093567255,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00022580798877613062,
|
|
"loss": 5.4018,
|
|
"mean_token_accuracy": 0.20899286568164827,
|
|
"num_tokens": 13201444.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 6.237301540374756,
|
|
"epoch": 6.204678362573099,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022535069683245952,
|
|
"loss": 5.4603,
|
|
"mean_token_accuracy": 0.2020897939801216,
|
|
"num_tokens": 13213390.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 6.096331977844239,
|
|
"epoch": 6.2105263157894735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00022489362037032446,
|
|
"loss": 5.3637,
|
|
"mean_token_accuracy": 0.20506755858659745,
|
|
"num_tokens": 13225321.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 6.123323440551758,
|
|
"epoch": 6.216374269005848,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002244367613734702,
|
|
"loss": 5.4415,
|
|
"mean_token_accuracy": 0.20185154676437378,
|
|
"num_tokens": 13237737.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 6.231597518920898,
|
|
"epoch": 6.222222222222222,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00022398012182469734,
|
|
"loss": 5.5355,
|
|
"mean_token_accuracy": 0.2008710816502571,
|
|
"num_tokens": 13250235.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 6.204935932159424,
|
|
"epoch": 6.228070175438597,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002235237037058544,
|
|
"loss": 5.5208,
|
|
"mean_token_accuracy": 0.1974329710006714,
|
|
"num_tokens": 13263537.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 6.129282665252686,
|
|
"epoch": 6.23391812865497,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022306750899782875,
|
|
"loss": 5.4665,
|
|
"mean_token_accuracy": 0.2030663713812828,
|
|
"num_tokens": 13275436.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 6.215650606155395,
|
|
"epoch": 6.239766081871345,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000222611539680538,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.19916548430919648,
|
|
"num_tokens": 13287346.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 6.131848430633545,
|
|
"epoch": 6.245614035087719,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022215579773292188,
|
|
"loss": 5.4048,
|
|
"mean_token_accuracy": 0.21247149258852005,
|
|
"num_tokens": 13299145.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 6.16942458152771,
|
|
"epoch": 6.251461988304094,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022170028513293288,
|
|
"loss": 5.4592,
|
|
"mean_token_accuracy": 0.20493327379226683,
|
|
"num_tokens": 13311709.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 6.134395694732666,
|
|
"epoch": 6.257309941520468,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002212450038575285,
|
|
"loss": 5.4362,
|
|
"mean_token_accuracy": 0.2082882821559906,
|
|
"num_tokens": 13324441.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 6.116281175613404,
|
|
"epoch": 6.2631578947368425,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022078995588266227,
|
|
"loss": 5.3826,
|
|
"mean_token_accuracy": 0.21131782680749894,
|
|
"num_tokens": 13337276.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 6.140869855880737,
|
|
"epoch": 6.269005847953216,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002203351431832748,
|
|
"loss": 5.4287,
|
|
"mean_token_accuracy": 0.20040909945964813,
|
|
"num_tokens": 13350254.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 6.143431091308594,
|
|
"epoch": 6.2748538011695905,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002198805677332858,
|
|
"loss": 5.5451,
|
|
"mean_token_accuracy": 0.2005102440714836,
|
|
"num_tokens": 13363430.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 6.162072086334229,
|
|
"epoch": 6.280701754385965,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00021942623150558556,
|
|
"loss": 5.4356,
|
|
"mean_token_accuracy": 0.20360455811023712,
|
|
"num_tokens": 13375350.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 6.174575138092041,
|
|
"epoch": 6.286549707602339,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00021897213647202558,
|
|
"loss": 5.4921,
|
|
"mean_token_accuracy": 0.2035872146487236,
|
|
"num_tokens": 13388339.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 6.073856401443481,
|
|
"epoch": 6.292397660818714,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002185182846034111,
|
|
"loss": 5.3541,
|
|
"mean_token_accuracy": 0.22392513751983642,
|
|
"num_tokens": 13401122.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 6.140816068649292,
|
|
"epoch": 6.298245614035087,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002180646778694918,
|
|
"loss": 5.4468,
|
|
"mean_token_accuracy": 0.20988378226757048,
|
|
"num_tokens": 13413428.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 6.141312599182129,
|
|
"epoch": 6.304093567251462,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00021761131823895324,
|
|
"loss": 5.4823,
|
|
"mean_token_accuracy": 0.20487305521965027,
|
|
"num_tokens": 13427045.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 6.189494514465332,
|
|
"epoch": 6.309941520467836,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002171582076794088,
|
|
"loss": 5.4711,
|
|
"mean_token_accuracy": 0.20343320369720458,
|
|
"num_tokens": 13439524.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 6.193106365203858,
|
|
"epoch": 6.315789473684211,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002167053481573909,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.2035904824733734,
|
|
"num_tokens": 13453705.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 6.115349340438843,
|
|
"epoch": 6.321637426900585,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00021625274163834224,
|
|
"loss": 5.3933,
|
|
"mean_token_accuracy": 0.2185133144259453,
|
|
"num_tokens": 13466129.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 6.183847093582154,
|
|
"epoch": 6.3274853801169595,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00021580039008660762,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.19980748295783995,
|
|
"num_tokens": 13479522.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 6.159699726104736,
|
|
"epoch": 6.333333333333333,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00021534829546542518,
|
|
"loss": 5.3861,
|
|
"mean_token_accuracy": 0.20308229923248292,
|
|
"num_tokens": 13492620.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 6.1466528415679935,
|
|
"epoch": 6.3391812865497075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002148964597369179,
|
|
"loss": 5.3975,
|
|
"mean_token_accuracy": 0.20609309524297714,
|
|
"num_tokens": 13503573.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 6.104958200454712,
|
|
"epoch": 6.345029239766082,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021444488486208545,
|
|
"loss": 5.4367,
|
|
"mean_token_accuracy": 0.2117675170302391,
|
|
"num_tokens": 13515999.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 6.141648387908935,
|
|
"epoch": 6.350877192982456,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00021399357280079502,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.2060331627726555,
|
|
"num_tokens": 13528480.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 6.150502490997314,
|
|
"epoch": 6.356725146198831,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021354252551177333,
|
|
"loss": 5.4766,
|
|
"mean_token_accuracy": 0.20260114073753357,
|
|
"num_tokens": 13540606.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 6.182828617095947,
|
|
"epoch": 6.362573099415204,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00021309174495259814,
|
|
"loss": 5.4846,
|
|
"mean_token_accuracy": 0.2003386214375496,
|
|
"num_tokens": 13552947.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 6.137768173217774,
|
|
"epoch": 6.368421052631579,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00021264123307968913,
|
|
"loss": 5.4526,
|
|
"mean_token_accuracy": 0.20349930226802826,
|
|
"num_tokens": 13565499.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 6.02544674873352,
|
|
"epoch": 6.374269005847953,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002121909918483002,
|
|
"loss": 5.3136,
|
|
"mean_token_accuracy": 0.22207755595445633,
|
|
"num_tokens": 13578491.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 6.136955213546753,
|
|
"epoch": 6.380116959064328,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00021174102321251082,
|
|
"loss": 5.4274,
|
|
"mean_token_accuracy": 0.2047642156481743,
|
|
"num_tokens": 13590638.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 6.1598388195037845,
|
|
"epoch": 6.385964912280702,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002112913291252168,
|
|
"loss": 5.488,
|
|
"mean_token_accuracy": 0.20166852623224257,
|
|
"num_tokens": 13602198.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 6.154816341400147,
|
|
"epoch": 6.391812865497076,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000210841911538123,
|
|
"loss": 5.4379,
|
|
"mean_token_accuracy": 0.20342409759759902,
|
|
"num_tokens": 13615520.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 6.137780857086182,
|
|
"epoch": 6.39766081871345,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00021039277240173372,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.20349852442741395,
|
|
"num_tokens": 13629382.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 5.991627788543701,
|
|
"epoch": 6.4035087719298245,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002099439136653452,
|
|
"loss": 5.233,
|
|
"mean_token_accuracy": 0.2242329716682434,
|
|
"num_tokens": 13642336.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 6.106536722183227,
|
|
"epoch": 6.409356725146199,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00020949533727703657,
|
|
"loss": 5.4732,
|
|
"mean_token_accuracy": 0.20060493052005768,
|
|
"num_tokens": 13654729.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 6.155096387863159,
|
|
"epoch": 6.415204678362573,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020904704518366136,
|
|
"loss": 5.4135,
|
|
"mean_token_accuracy": 0.2106667250394821,
|
|
"num_tokens": 13667068.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 6.107669162750244,
|
|
"epoch": 6.421052631578947,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00020859903933083952,
|
|
"loss": 5.4584,
|
|
"mean_token_accuracy": 0.20781600326299668,
|
|
"num_tokens": 13680164.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 6.1750740051269535,
|
|
"epoch": 6.426900584795321,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002081513216629486,
|
|
"loss": 5.5326,
|
|
"mean_token_accuracy": 0.1941557601094246,
|
|
"num_tokens": 13693510.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 6.194888544082642,
|
|
"epoch": 6.432748538011696,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002077038941231153,
|
|
"loss": 5.4137,
|
|
"mean_token_accuracy": 0.20732810497283935,
|
|
"num_tokens": 13705763.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 6.432748538011696,
|
|
"eval_entropy": 6.020032822206661,
|
|
"eval_loss": 6.5356831550598145,
|
|
"eval_mean_token_accuracy": 0.1514763523629457,
|
|
"eval_num_tokens": 13705763.0,
|
|
"eval_runtime": 1.7326,
|
|
"eval_samples_per_second": 1562.365,
|
|
"eval_steps_per_second": 195.656,
|
|
"step": 5500
|
|
}
|
|
],
|
|
"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": 2946008458506240.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|