df719f9d99
Model: fpadovani/jpn-jpan-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.7543859649122808,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 3.011503052711487,
|
|
"epoch": 0.005847953216374269,
|
|
"grad_norm": 35.75,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.1046,
|
|
"mean_token_accuracy": 0.035688506253063676,
|
|
"num_tokens": 13999.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 3.1570172786712645,
|
|
"epoch": 0.011695906432748537,
|
|
"grad_norm": 33.0,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 13.9181,
|
|
"mean_token_accuracy": 0.037344493344426154,
|
|
"num_tokens": 25888.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 3.506499719619751,
|
|
"epoch": 0.017543859649122806,
|
|
"grad_norm": 39.0,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.4816,
|
|
"mean_token_accuracy": 0.033669637702405456,
|
|
"num_tokens": 37699.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.382130026817322,
|
|
"epoch": 0.023391812865497075,
|
|
"grad_norm": 39.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 12.8513,
|
|
"mean_token_accuracy": 0.019791487697511913,
|
|
"num_tokens": 49648.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.491881942749023,
|
|
"epoch": 0.029239766081871343,
|
|
"grad_norm": 14.5625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.2853,
|
|
"mean_token_accuracy": 0.019448780827224253,
|
|
"num_tokens": 62255.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.010943794250489,
|
|
"epoch": 0.03508771929824561,
|
|
"grad_norm": 4.84375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.7751,
|
|
"mean_token_accuracy": 0.00878947232849896,
|
|
"num_tokens": 76003.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.62158660888672,
|
|
"epoch": 0.04093567251461988,
|
|
"grad_norm": 3.40625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.5825,
|
|
"mean_token_accuracy": 0.009877379098907114,
|
|
"num_tokens": 87585.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.41454849243164,
|
|
"epoch": 0.04678362573099415,
|
|
"grad_norm": 4.59375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1549,
|
|
"mean_token_accuracy": 0.019071799516677857,
|
|
"num_tokens": 100404.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.373588752746581,
|
|
"epoch": 0.05263157894736842,
|
|
"grad_norm": 2.421875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.91,
|
|
"mean_token_accuracy": 0.0510077447630465,
|
|
"num_tokens": 113130.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.290110111236572,
|
|
"epoch": 0.05847953216374269,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5708,
|
|
"mean_token_accuracy": 0.08142316564917565,
|
|
"num_tokens": 125430.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.002687644958495,
|
|
"epoch": 0.06432748538011696,
|
|
"grad_norm": 3.078125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5528,
|
|
"mean_token_accuracy": 0.0672168917953968,
|
|
"num_tokens": 137550.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.2318941116333,
|
|
"epoch": 0.07017543859649122,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4611,
|
|
"mean_token_accuracy": 0.07226917184889317,
|
|
"num_tokens": 149530.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.194271087646484,
|
|
"epoch": 0.07602339181286549,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4491,
|
|
"mean_token_accuracy": 0.07329789400100709,
|
|
"num_tokens": 161671.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.232907676696778,
|
|
"epoch": 0.08187134502923976,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4074,
|
|
"mean_token_accuracy": 0.0754720564931631,
|
|
"num_tokens": 174160.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.223019981384278,
|
|
"epoch": 0.08771929824561403,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.357,
|
|
"mean_token_accuracy": 0.06809913516044616,
|
|
"num_tokens": 186969.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 9.965007305145264,
|
|
"epoch": 0.0935672514619883,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2625,
|
|
"mean_token_accuracy": 0.0748597014695406,
|
|
"num_tokens": 199470.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.107829475402832,
|
|
"epoch": 0.09941520467836257,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1703,
|
|
"mean_token_accuracy": 0.07800811901688576,
|
|
"num_tokens": 211306.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 9.909124755859375,
|
|
"epoch": 0.10526315789473684,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.1713,
|
|
"mean_token_accuracy": 0.07304236702620984,
|
|
"num_tokens": 223323.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.078854656219482,
|
|
"epoch": 0.1111111111111111,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.0578,
|
|
"mean_token_accuracy": 0.08289388790726662,
|
|
"num_tokens": 235896.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 9.830192375183106,
|
|
"epoch": 0.11695906432748537,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.021,
|
|
"mean_token_accuracy": 0.07751639671623707,
|
|
"num_tokens": 247879.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 9.94895429611206,
|
|
"epoch": 0.12280701754385964,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.9105,
|
|
"mean_token_accuracy": 0.07603972963988781,
|
|
"num_tokens": 259714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.896305179595947,
|
|
"epoch": 0.1286549707602339,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.8763,
|
|
"mean_token_accuracy": 0.08195686042308807,
|
|
"num_tokens": 272429.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.760246086120606,
|
|
"epoch": 0.13450292397660818,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.8484,
|
|
"mean_token_accuracy": 0.07280035950243473,
|
|
"num_tokens": 285924.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.702124404907227,
|
|
"epoch": 0.14035087719298245,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.6942,
|
|
"mean_token_accuracy": 0.08684460520744323,
|
|
"num_tokens": 299453.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.505807209014893,
|
|
"epoch": 0.14619883040935672,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.6969,
|
|
"mean_token_accuracy": 0.0838764600455761,
|
|
"num_tokens": 312295.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.498956871032714,
|
|
"epoch": 0.15204678362573099,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.5488,
|
|
"mean_token_accuracy": 0.08384966105222702,
|
|
"num_tokens": 324730.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.271528625488282,
|
|
"epoch": 0.15789473684210525,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.4801,
|
|
"mean_token_accuracy": 0.08416559919714928,
|
|
"num_tokens": 336481.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.22463436126709,
|
|
"epoch": 0.16374269005847952,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4865,
|
|
"mean_token_accuracy": 0.08492137640714645,
|
|
"num_tokens": 348987.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.034030055999756,
|
|
"epoch": 0.1695906432748538,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.3974,
|
|
"mean_token_accuracy": 0.08240444362163543,
|
|
"num_tokens": 360390.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.999344062805175,
|
|
"epoch": 0.17543859649122806,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.4036,
|
|
"mean_token_accuracy": 0.08263231143355369,
|
|
"num_tokens": 372456.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.908535861968994,
|
|
"epoch": 0.18128654970760233,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.3422,
|
|
"mean_token_accuracy": 0.07766095399856568,
|
|
"num_tokens": 384185.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.835869979858398,
|
|
"epoch": 0.1871345029239766,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.3227,
|
|
"mean_token_accuracy": 0.08606852069497109,
|
|
"num_tokens": 395399.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.697229862213135,
|
|
"epoch": 0.19298245614035087,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.337,
|
|
"mean_token_accuracy": 0.08855125159025193,
|
|
"num_tokens": 408676.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.610207080841064,
|
|
"epoch": 0.19883040935672514,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.3123,
|
|
"mean_token_accuracy": 0.08628950268030167,
|
|
"num_tokens": 421861.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.750581073760987,
|
|
"epoch": 0.2046783625730994,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.353,
|
|
"mean_token_accuracy": 0.09162932708859443,
|
|
"num_tokens": 434852.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.575475406646728,
|
|
"epoch": 0.21052631578947367,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.3025,
|
|
"mean_token_accuracy": 0.08264983892440796,
|
|
"num_tokens": 447313.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.66052532196045,
|
|
"epoch": 0.21637426900584794,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.242,
|
|
"mean_token_accuracy": 0.09072699844837188,
|
|
"num_tokens": 460271.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.458691787719726,
|
|
"epoch": 0.2222222222222222,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.2117,
|
|
"mean_token_accuracy": 0.0964887373149395,
|
|
"num_tokens": 473006.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.625473880767823,
|
|
"epoch": 0.22807017543859648,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2416,
|
|
"mean_token_accuracy": 0.08616446778178215,
|
|
"num_tokens": 486100.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.410247135162354,
|
|
"epoch": 0.23391812865497075,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.1352,
|
|
"mean_token_accuracy": 0.09261524304747581,
|
|
"num_tokens": 499460.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.550405311584473,
|
|
"epoch": 0.23976608187134502,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.1476,
|
|
"mean_token_accuracy": 0.091904778778553,
|
|
"num_tokens": 511795.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.412632751464844,
|
|
"epoch": 0.24561403508771928,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.204,
|
|
"mean_token_accuracy": 0.09843768626451492,
|
|
"num_tokens": 524897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.52775535583496,
|
|
"epoch": 0.25146198830409355,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.1374,
|
|
"mean_token_accuracy": 0.09786352440714836,
|
|
"num_tokens": 538325.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.386085700988769,
|
|
"epoch": 0.2573099415204678,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.1911,
|
|
"mean_token_accuracy": 0.09132648184895516,
|
|
"num_tokens": 550979.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.483522033691406,
|
|
"epoch": 0.2631578947368421,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.1602,
|
|
"mean_token_accuracy": 0.09506673067808151,
|
|
"num_tokens": 563001.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.336331844329834,
|
|
"epoch": 0.26900584795321636,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.0781,
|
|
"mean_token_accuracy": 0.09937207028269768,
|
|
"num_tokens": 573600.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.444112873077392,
|
|
"epoch": 0.27485380116959063,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.0893,
|
|
"mean_token_accuracy": 0.10078474283218383,
|
|
"num_tokens": 585287.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.33163080215454,
|
|
"epoch": 0.2807017543859649,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.1193,
|
|
"mean_token_accuracy": 0.09278298616409301,
|
|
"num_tokens": 597386.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.417845058441163,
|
|
"epoch": 0.28654970760233917,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.0979,
|
|
"mean_token_accuracy": 0.09884399548172951,
|
|
"num_tokens": 609256.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.383094787597656,
|
|
"epoch": 0.29239766081871343,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.0968,
|
|
"mean_token_accuracy": 0.10301539301872253,
|
|
"num_tokens": 621506.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.22484188079834,
|
|
"epoch": 0.2982456140350877,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.0108,
|
|
"mean_token_accuracy": 0.10739858150482177,
|
|
"num_tokens": 633512.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.35753002166748,
|
|
"epoch": 0.30409356725146197,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.1404,
|
|
"mean_token_accuracy": 0.09757127910852433,
|
|
"num_tokens": 647098.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.389013767242432,
|
|
"epoch": 0.30994152046783624,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.1411,
|
|
"mean_token_accuracy": 0.09518403038382531,
|
|
"num_tokens": 659559.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.338126182556152,
|
|
"epoch": 0.3157894736842105,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.0745,
|
|
"mean_token_accuracy": 0.09692913815379142,
|
|
"num_tokens": 672621.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.312376308441163,
|
|
"epoch": 0.3216374269005848,
|
|
"grad_norm": 2.828125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.0387,
|
|
"mean_token_accuracy": 0.10461566746234893,
|
|
"num_tokens": 685652.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.29957218170166,
|
|
"epoch": 0.32748538011695905,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.0998,
|
|
"mean_token_accuracy": 0.10161212980747222,
|
|
"num_tokens": 696910.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.248525142669678,
|
|
"epoch": 0.3333333333333333,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.9659,
|
|
"mean_token_accuracy": 0.10607651099562646,
|
|
"num_tokens": 709054.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.277126693725586,
|
|
"epoch": 0.3391812865497076,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.997,
|
|
"mean_token_accuracy": 0.1044411912560463,
|
|
"num_tokens": 720902.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.320906162261963,
|
|
"epoch": 0.34502923976608185,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.0586,
|
|
"mean_token_accuracy": 0.10378434732556344,
|
|
"num_tokens": 733319.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.274054527282715,
|
|
"epoch": 0.3508771929824561,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.9508,
|
|
"mean_token_accuracy": 0.10693862661719322,
|
|
"num_tokens": 744817.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.191329860687256,
|
|
"epoch": 0.3567251461988304,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.0044,
|
|
"mean_token_accuracy": 0.1035879172384739,
|
|
"num_tokens": 756582.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.263913822174072,
|
|
"epoch": 0.36257309941520466,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.9151,
|
|
"mean_token_accuracy": 0.1130017727613449,
|
|
"num_tokens": 769273.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.111418914794921,
|
|
"epoch": 0.3684210526315789,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.9862,
|
|
"mean_token_accuracy": 0.10989293605089187,
|
|
"num_tokens": 782001.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.221540355682373,
|
|
"epoch": 0.3742690058479532,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.9703,
|
|
"mean_token_accuracy": 0.112515377253294,
|
|
"num_tokens": 794575.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.238408851623536,
|
|
"epoch": 0.38011695906432746,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.9488,
|
|
"mean_token_accuracy": 0.10909051299095154,
|
|
"num_tokens": 805583.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.123234272003174,
|
|
"epoch": 0.38596491228070173,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.993,
|
|
"mean_token_accuracy": 0.10957283899188042,
|
|
"num_tokens": 817617.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.23239221572876,
|
|
"epoch": 0.391812865497076,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.8943,
|
|
"mean_token_accuracy": 0.11208963096141815,
|
|
"num_tokens": 830043.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.07594804763794,
|
|
"epoch": 0.39766081871345027,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.8714,
|
|
"mean_token_accuracy": 0.1205448180437088,
|
|
"num_tokens": 843200.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.210996723175048,
|
|
"epoch": 0.40350877192982454,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.918,
|
|
"mean_token_accuracy": 0.11632056310772895,
|
|
"num_tokens": 856031.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.029043245315552,
|
|
"epoch": 0.4093567251461988,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.8883,
|
|
"mean_token_accuracy": 0.11492086499929428,
|
|
"num_tokens": 869119.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.107694482803344,
|
|
"epoch": 0.4152046783625731,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.9053,
|
|
"mean_token_accuracy": 0.11929167732596398,
|
|
"num_tokens": 880802.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.145207405090332,
|
|
"epoch": 0.42105263157894735,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.8911,
|
|
"mean_token_accuracy": 0.1157754972577095,
|
|
"num_tokens": 894237.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.104444217681884,
|
|
"epoch": 0.4269005847953216,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.9097,
|
|
"mean_token_accuracy": 0.11690913587808609,
|
|
"num_tokens": 906695.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.126790809631348,
|
|
"epoch": 0.4327485380116959,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.8766,
|
|
"mean_token_accuracy": 0.11690322905778885,
|
|
"num_tokens": 918955.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.079369258880615,
|
|
"epoch": 0.43859649122807015,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.9117,
|
|
"mean_token_accuracy": 0.11631168648600579,
|
|
"num_tokens": 931812.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.124732828140258,
|
|
"epoch": 0.4444444444444444,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.9105,
|
|
"mean_token_accuracy": 0.10793866887688637,
|
|
"num_tokens": 945261.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.059471893310548,
|
|
"epoch": 0.4502923976608187,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.8211,
|
|
"mean_token_accuracy": 0.12420412302017211,
|
|
"num_tokens": 958376.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.045862627029418,
|
|
"epoch": 0.45614035087719296,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.8396,
|
|
"mean_token_accuracy": 0.12700849547982215,
|
|
"num_tokens": 971709.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.091754007339478,
|
|
"epoch": 0.4619883040935672,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.8096,
|
|
"mean_token_accuracy": 0.1235598661005497,
|
|
"num_tokens": 983847.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.116325187683106,
|
|
"epoch": 0.4678362573099415,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.8836,
|
|
"mean_token_accuracy": 0.11197860836982727,
|
|
"num_tokens": 997034.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.948489475250244,
|
|
"epoch": 0.47368421052631576,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.9183,
|
|
"mean_token_accuracy": 0.11601087674498559,
|
|
"num_tokens": 1010213.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.183907890319825,
|
|
"epoch": 0.47953216374269003,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.7917,
|
|
"mean_token_accuracy": 0.11840821728110314,
|
|
"num_tokens": 1021618.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.002576017379761,
|
|
"epoch": 0.4853801169590643,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.8151,
|
|
"mean_token_accuracy": 0.11744796931743622,
|
|
"num_tokens": 1033809.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.077327156066895,
|
|
"epoch": 0.49122807017543857,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.8025,
|
|
"mean_token_accuracy": 0.11957534328103066,
|
|
"num_tokens": 1045291.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.03057599067688,
|
|
"epoch": 0.49707602339181284,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.7792,
|
|
"mean_token_accuracy": 0.11945864260196686,
|
|
"num_tokens": 1057516.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.868794345855713,
|
|
"epoch": 0.5029239766081871,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.7351,
|
|
"mean_token_accuracy": 0.12831166237592698,
|
|
"num_tokens": 1070675.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.045708703994752,
|
|
"epoch": 0.5087719298245614,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.8167,
|
|
"mean_token_accuracy": 0.12775540575385094,
|
|
"num_tokens": 1082965.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.97526912689209,
|
|
"epoch": 0.5146198830409356,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.7974,
|
|
"mean_token_accuracy": 0.11712993755936622,
|
|
"num_tokens": 1095922.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.9973499298095705,
|
|
"epoch": 0.52046783625731,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.7752,
|
|
"mean_token_accuracy": 0.12419187352061271,
|
|
"num_tokens": 1109002.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.025882577896118,
|
|
"epoch": 0.5263157894736842,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.7881,
|
|
"mean_token_accuracy": 0.11773632392287255,
|
|
"num_tokens": 1121199.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.909945821762085,
|
|
"epoch": 0.5321637426900585,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.7307,
|
|
"mean_token_accuracy": 0.12722171917557717,
|
|
"num_tokens": 1134721.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.956731367111206,
|
|
"epoch": 0.5380116959064327,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.7963,
|
|
"mean_token_accuracy": 0.12164180129766464,
|
|
"num_tokens": 1147954.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.956630373001099,
|
|
"epoch": 0.543859649122807,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.7275,
|
|
"mean_token_accuracy": 0.1270610734820366,
|
|
"num_tokens": 1161208.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.929465818405151,
|
|
"epoch": 0.5497076023391813,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.8257,
|
|
"mean_token_accuracy": 0.11997483149170876,
|
|
"num_tokens": 1173789.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.05664644241333,
|
|
"epoch": 0.5555555555555556,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.7272,
|
|
"mean_token_accuracy": 0.12666135057806968,
|
|
"num_tokens": 1185765.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.764935064315796,
|
|
"epoch": 0.5614035087719298,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.7499,
|
|
"mean_token_accuracy": 0.12476444989442825,
|
|
"num_tokens": 1197954.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.030300664901734,
|
|
"epoch": 0.5672514619883041,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.679,
|
|
"mean_token_accuracy": 0.12792463153600692,
|
|
"num_tokens": 1211285.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.9220020294189455,
|
|
"epoch": 0.5730994152046783,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.7543,
|
|
"mean_token_accuracy": 0.11961747631430626,
|
|
"num_tokens": 1223197.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.8508258819580075,
|
|
"epoch": 0.5789473684210527,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.68,
|
|
"mean_token_accuracy": 0.12632347792387008,
|
|
"num_tokens": 1234817.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.9022626876831055,
|
|
"epoch": 0.5847953216374269,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.7281,
|
|
"mean_token_accuracy": 0.1318517729640007,
|
|
"num_tokens": 1249305.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5847953216374269,
|
|
"eval_entropy": 7.841686710143863,
|
|
"eval_loss": 7.758663654327393,
|
|
"eval_mean_token_accuracy": 0.1180623706160965,
|
|
"eval_num_tokens": 1249305.0,
|
|
"eval_runtime": 2.3012,
|
|
"eval_samples_per_second": 1176.366,
|
|
"eval_steps_per_second": 147.317,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.89346866607666,
|
|
"epoch": 0.5906432748538012,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.7086,
|
|
"mean_token_accuracy": 0.12867532819509506,
|
|
"num_tokens": 1262182.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.847484970092774,
|
|
"epoch": 0.5964912280701754,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.7312,
|
|
"mean_token_accuracy": 0.1298251412808895,
|
|
"num_tokens": 1274616.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.838339996337891,
|
|
"epoch": 0.6023391812865497,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.7403,
|
|
"mean_token_accuracy": 0.11953374072909355,
|
|
"num_tokens": 1287306.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.859510087966919,
|
|
"epoch": 0.6081871345029239,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.5908,
|
|
"mean_token_accuracy": 0.13276281207799911,
|
|
"num_tokens": 1299472.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.7895465850830075,
|
|
"epoch": 0.6140350877192983,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.5534,
|
|
"mean_token_accuracy": 0.13485597670078278,
|
|
"num_tokens": 1312636.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.6986805438995365,
|
|
"epoch": 0.6198830409356725,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.5875,
|
|
"mean_token_accuracy": 0.13942663818597795,
|
|
"num_tokens": 1324058.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.906114959716797,
|
|
"epoch": 0.6257309941520468,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.7124,
|
|
"mean_token_accuracy": 0.12152797058224678,
|
|
"num_tokens": 1337605.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.845144605636596,
|
|
"epoch": 0.631578947368421,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.7079,
|
|
"mean_token_accuracy": 0.12539200559258462,
|
|
"num_tokens": 1350658.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.842404890060425,
|
|
"epoch": 0.6374269005847953,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.6214,
|
|
"mean_token_accuracy": 0.13082724660634995,
|
|
"num_tokens": 1362091.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.808346605300903,
|
|
"epoch": 0.6432748538011696,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.6453,
|
|
"mean_token_accuracy": 0.12663563415408136,
|
|
"num_tokens": 1374055.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.67389407157898,
|
|
"epoch": 0.6491228070175439,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.5291,
|
|
"mean_token_accuracy": 0.1318692184984684,
|
|
"num_tokens": 1385269.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.744857835769653,
|
|
"epoch": 0.6549707602339181,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.6424,
|
|
"mean_token_accuracy": 0.1306732676923275,
|
|
"num_tokens": 1396559.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.729473543167114,
|
|
"epoch": 0.6608187134502924,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.6538,
|
|
"mean_token_accuracy": 0.13084901124238968,
|
|
"num_tokens": 1408838.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.766721534729004,
|
|
"epoch": 0.6666666666666666,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.5368,
|
|
"mean_token_accuracy": 0.13452390804886818,
|
|
"num_tokens": 1421023.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.697804594039917,
|
|
"epoch": 0.672514619883041,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.6167,
|
|
"mean_token_accuracy": 0.13060338124632836,
|
|
"num_tokens": 1433912.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.818636512756347,
|
|
"epoch": 0.6783625730994152,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.6231,
|
|
"mean_token_accuracy": 0.1270453654229641,
|
|
"num_tokens": 1445262.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.690509176254272,
|
|
"epoch": 0.6842105263157895,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.5815,
|
|
"mean_token_accuracy": 0.13163993805646895,
|
|
"num_tokens": 1457927.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.660882234573364,
|
|
"epoch": 0.6900584795321637,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.6489,
|
|
"mean_token_accuracy": 0.12540714517235757,
|
|
"num_tokens": 1470116.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.764140367507935,
|
|
"epoch": 0.695906432748538,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.5297,
|
|
"mean_token_accuracy": 0.13199650049209594,
|
|
"num_tokens": 1482067.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.73416576385498,
|
|
"epoch": 0.7017543859649122,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.6134,
|
|
"mean_token_accuracy": 0.1266568437218666,
|
|
"num_tokens": 1493902.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.750314617156983,
|
|
"epoch": 0.7076023391812866,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.5725,
|
|
"mean_token_accuracy": 0.12624214142560958,
|
|
"num_tokens": 1505689.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.611770343780518,
|
|
"epoch": 0.7134502923976608,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.5194,
|
|
"mean_token_accuracy": 0.14677974060177804,
|
|
"num_tokens": 1518002.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.710480356216431,
|
|
"epoch": 0.7192982456140351,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.5443,
|
|
"mean_token_accuracy": 0.13318197801709175,
|
|
"num_tokens": 1529307.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.769068670272827,
|
|
"epoch": 0.7251461988304093,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.6963,
|
|
"mean_token_accuracy": 0.11717625185847283,
|
|
"num_tokens": 1542351.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.606048679351806,
|
|
"epoch": 0.7309941520467836,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.5147,
|
|
"mean_token_accuracy": 0.135113126039505,
|
|
"num_tokens": 1556263.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.631821155548096,
|
|
"epoch": 0.7368421052631579,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.5509,
|
|
"mean_token_accuracy": 0.13068365454673767,
|
|
"num_tokens": 1568906.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.647361993789673,
|
|
"epoch": 0.7426900584795322,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.5375,
|
|
"mean_token_accuracy": 0.13785270750522613,
|
|
"num_tokens": 1580631.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.632867479324341,
|
|
"epoch": 0.7485380116959064,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.5521,
|
|
"mean_token_accuracy": 0.12791232094168664,
|
|
"num_tokens": 1592369.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.650920963287353,
|
|
"epoch": 0.7543859649122807,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.5862,
|
|
"mean_token_accuracy": 0.13423935249447821,
|
|
"num_tokens": 1604888.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.6349873542785645,
|
|
"epoch": 0.7602339181286549,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.4819,
|
|
"mean_token_accuracy": 0.13629580438137054,
|
|
"num_tokens": 1617128.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.645885038375854,
|
|
"epoch": 0.7660818713450293,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.5341,
|
|
"mean_token_accuracy": 0.12496847957372666,
|
|
"num_tokens": 1629350.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.682138872146607,
|
|
"epoch": 0.7719298245614035,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.5573,
|
|
"mean_token_accuracy": 0.1278813175857067,
|
|
"num_tokens": 1643082.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.628981828689575,
|
|
"epoch": 0.7777777777777778,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.5434,
|
|
"mean_token_accuracy": 0.13124902471899985,
|
|
"num_tokens": 1654785.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.671693944931031,
|
|
"epoch": 0.783625730994152,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.5755,
|
|
"mean_token_accuracy": 0.12893529310822488,
|
|
"num_tokens": 1667704.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.678587818145752,
|
|
"epoch": 0.7894736842105263,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.5077,
|
|
"mean_token_accuracy": 0.1305314414203167,
|
|
"num_tokens": 1680459.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.557713508605957,
|
|
"epoch": 0.7953216374269005,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.4271,
|
|
"mean_token_accuracy": 0.13665740936994553,
|
|
"num_tokens": 1692912.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.614948320388794,
|
|
"epoch": 0.8011695906432749,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.4574,
|
|
"mean_token_accuracy": 0.1333981230854988,
|
|
"num_tokens": 1705329.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.643642568588257,
|
|
"epoch": 0.8070175438596491,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.5618,
|
|
"mean_token_accuracy": 0.12579705193638802,
|
|
"num_tokens": 1717283.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.503328323364258,
|
|
"epoch": 0.8128654970760234,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.5046,
|
|
"mean_token_accuracy": 0.1324782818555832,
|
|
"num_tokens": 1730308.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.626899671554566,
|
|
"epoch": 0.8187134502923976,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5432,
|
|
"mean_token_accuracy": 0.12995980978012084,
|
|
"num_tokens": 1742716.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.57134051322937,
|
|
"epoch": 0.8245614035087719,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.4896,
|
|
"mean_token_accuracy": 0.13513270914554595,
|
|
"num_tokens": 1754183.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.522201442718506,
|
|
"epoch": 0.8304093567251462,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.5076,
|
|
"mean_token_accuracy": 0.12541001066565513,
|
|
"num_tokens": 1766601.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.612595176696777,
|
|
"epoch": 0.8362573099415205,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.3986,
|
|
"mean_token_accuracy": 0.14329489395022393,
|
|
"num_tokens": 1778756.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.3666143894195555,
|
|
"epoch": 0.8421052631578947,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.466,
|
|
"mean_token_accuracy": 0.13944232240319251,
|
|
"num_tokens": 1791485.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.58320369720459,
|
|
"epoch": 0.847953216374269,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.4825,
|
|
"mean_token_accuracy": 0.12936609014868736,
|
|
"num_tokens": 1803712.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.433703374862671,
|
|
"epoch": 0.8538011695906432,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.44,
|
|
"mean_token_accuracy": 0.1326867386698723,
|
|
"num_tokens": 1818387.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.604799032211304,
|
|
"epoch": 0.8596491228070176,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.4325,
|
|
"mean_token_accuracy": 0.13445507287979125,
|
|
"num_tokens": 1830113.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.533818340301513,
|
|
"epoch": 0.8654970760233918,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.5505,
|
|
"mean_token_accuracy": 0.13362800627946853,
|
|
"num_tokens": 1841789.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.506076908111572,
|
|
"epoch": 0.8713450292397661,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.3988,
|
|
"mean_token_accuracy": 0.13710973560810089,
|
|
"num_tokens": 1853669.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.449912071228027,
|
|
"epoch": 0.8771929824561403,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.469,
|
|
"mean_token_accuracy": 0.13512061759829522,
|
|
"num_tokens": 1865322.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.454483270645142,
|
|
"epoch": 0.8830409356725146,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3563,
|
|
"mean_token_accuracy": 0.14202516600489618,
|
|
"num_tokens": 1879039.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.581981468200683,
|
|
"epoch": 0.8888888888888888,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.5329,
|
|
"mean_token_accuracy": 0.1277366228401661,
|
|
"num_tokens": 1893078.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.589095783233643,
|
|
"epoch": 0.8947368421052632,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.5463,
|
|
"mean_token_accuracy": 0.1306015282869339,
|
|
"num_tokens": 1907399.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.511038160324096,
|
|
"epoch": 0.9005847953216374,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.489,
|
|
"mean_token_accuracy": 0.12949865981936454,
|
|
"num_tokens": 1919290.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.527887201309204,
|
|
"epoch": 0.9064327485380117,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.4317,
|
|
"mean_token_accuracy": 0.13319515511393548,
|
|
"num_tokens": 1931658.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.557555913925171,
|
|
"epoch": 0.9122807017543859,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.4588,
|
|
"mean_token_accuracy": 0.12898892015218735,
|
|
"num_tokens": 1943881.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.488714742660522,
|
|
"epoch": 0.9181286549707602,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.3689,
|
|
"mean_token_accuracy": 0.1345718003809452,
|
|
"num_tokens": 1956288.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.382142972946167,
|
|
"epoch": 0.9239766081871345,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.2696,
|
|
"mean_token_accuracy": 0.1387133315205574,
|
|
"num_tokens": 1967265.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.400208759307861,
|
|
"epoch": 0.9298245614035088,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.4155,
|
|
"mean_token_accuracy": 0.13420891687273978,
|
|
"num_tokens": 1980039.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.5092360496521,
|
|
"epoch": 0.935672514619883,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.3761,
|
|
"mean_token_accuracy": 0.13293110430240632,
|
|
"num_tokens": 1992735.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.502032279968262,
|
|
"epoch": 0.9415204678362573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.4075,
|
|
"mean_token_accuracy": 0.13847337365150453,
|
|
"num_tokens": 2004545.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.40691909790039,
|
|
"epoch": 0.9473684210526315,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.3914,
|
|
"mean_token_accuracy": 0.13531816229224206,
|
|
"num_tokens": 2017100.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.43734827041626,
|
|
"epoch": 0.9532163742690059,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.3306,
|
|
"mean_token_accuracy": 0.14728882685303687,
|
|
"num_tokens": 2029868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.466191053390503,
|
|
"epoch": 0.9590643274853801,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.4035,
|
|
"mean_token_accuracy": 0.13149953931570052,
|
|
"num_tokens": 2041334.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.460909032821656,
|
|
"epoch": 0.9649122807017544,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.4782,
|
|
"mean_token_accuracy": 0.13376137018203735,
|
|
"num_tokens": 2054797.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.42348313331604,
|
|
"epoch": 0.9707602339181286,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.3561,
|
|
"mean_token_accuracy": 0.1338147610425949,
|
|
"num_tokens": 2068337.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.4611015796661375,
|
|
"epoch": 0.9766081871345029,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.3867,
|
|
"mean_token_accuracy": 0.13802094981074334,
|
|
"num_tokens": 2079984.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.447398233413696,
|
|
"epoch": 0.9824561403508771,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.4152,
|
|
"mean_token_accuracy": 0.13706867769360542,
|
|
"num_tokens": 2093069.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.325431823730469,
|
|
"epoch": 0.9883040935672515,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.3565,
|
|
"mean_token_accuracy": 0.1383918911218643,
|
|
"num_tokens": 2104753.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.395589447021484,
|
|
"epoch": 0.9941520467836257,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.3819,
|
|
"mean_token_accuracy": 0.13548673540353776,
|
|
"num_tokens": 2116576.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.470466804504395,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.3472,
|
|
"mean_token_accuracy": 0.13063811063766478,
|
|
"num_tokens": 2129484.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.311177539825439,
|
|
"epoch": 1.0058479532163742,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.1975,
|
|
"mean_token_accuracy": 0.14626543894410132,
|
|
"num_tokens": 2142000.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.366611528396606,
|
|
"epoch": 1.0116959064327484,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.123,
|
|
"mean_token_accuracy": 0.1465772069990635,
|
|
"num_tokens": 2152965.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.341231441497802,
|
|
"epoch": 1.0175438596491229,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.2225,
|
|
"mean_token_accuracy": 0.1331010527908802,
|
|
"num_tokens": 2165078.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.243547677993774,
|
|
"epoch": 1.023391812865497,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.151,
|
|
"mean_token_accuracy": 0.1457497827708721,
|
|
"num_tokens": 2176706.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.377624607086181,
|
|
"epoch": 1.0292397660818713,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.1943,
|
|
"mean_token_accuracy": 0.14025041311979294,
|
|
"num_tokens": 2190157.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.366387224197387,
|
|
"epoch": 1.0350877192982457,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.1634,
|
|
"mean_token_accuracy": 0.14333848357200624,
|
|
"num_tokens": 2201821.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.2149065971374515,
|
|
"epoch": 1.04093567251462,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.0878,
|
|
"mean_token_accuracy": 0.1505483314394951,
|
|
"num_tokens": 2214504.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.245018482208252,
|
|
"epoch": 1.0467836257309941,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.2145,
|
|
"mean_token_accuracy": 0.14023151621222496,
|
|
"num_tokens": 2226318.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.377814912796021,
|
|
"epoch": 1.0526315789473684,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.2412,
|
|
"mean_token_accuracy": 0.13482995182275773,
|
|
"num_tokens": 2237892.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.302912092208862,
|
|
"epoch": 1.0584795321637426,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.2105,
|
|
"mean_token_accuracy": 0.14221135154366493,
|
|
"num_tokens": 2250656.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.266082286834717,
|
|
"epoch": 1.064327485380117,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.2283,
|
|
"mean_token_accuracy": 0.13709186613559723,
|
|
"num_tokens": 2263685.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.261312961578369,
|
|
"epoch": 1.0701754385964912,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.1937,
|
|
"mean_token_accuracy": 0.13502099514007568,
|
|
"num_tokens": 2275413.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.305978202819825,
|
|
"epoch": 1.0760233918128654,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.1684,
|
|
"mean_token_accuracy": 0.1396795056760311,
|
|
"num_tokens": 2288165.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.334664630889892,
|
|
"epoch": 1.0818713450292399,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.1615,
|
|
"mean_token_accuracy": 0.13400667384266854,
|
|
"num_tokens": 2300574.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.243153524398804,
|
|
"epoch": 1.087719298245614,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.1727,
|
|
"mean_token_accuracy": 0.13601783886551858,
|
|
"num_tokens": 2311631.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.27063479423523,
|
|
"epoch": 1.0935672514619883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.1938,
|
|
"mean_token_accuracy": 0.13982510566711426,
|
|
"num_tokens": 2324312.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.349608087539673,
|
|
"epoch": 1.0994152046783625,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.2621,
|
|
"mean_token_accuracy": 0.133483674377203,
|
|
"num_tokens": 2337747.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.170055818557739,
|
|
"epoch": 1.1052631578947367,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.2157,
|
|
"mean_token_accuracy": 0.14102221578359603,
|
|
"num_tokens": 2349830.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.312947511672974,
|
|
"epoch": 1.1111111111111112,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.1527,
|
|
"mean_token_accuracy": 0.146473790705204,
|
|
"num_tokens": 2361825.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.233815908432007,
|
|
"epoch": 1.1169590643274854,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.084,
|
|
"mean_token_accuracy": 0.14439496994018555,
|
|
"num_tokens": 2373963.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.253248453140259,
|
|
"epoch": 1.1228070175438596,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.2355,
|
|
"mean_token_accuracy": 0.13487958535552025,
|
|
"num_tokens": 2386348.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.244376516342163,
|
|
"epoch": 1.128654970760234,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.1544,
|
|
"mean_token_accuracy": 0.13545818701386453,
|
|
"num_tokens": 2398338.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.206346225738526,
|
|
"epoch": 1.1345029239766082,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1114,
|
|
"mean_token_accuracy": 0.14215215146541596,
|
|
"num_tokens": 2411855.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.1879278182983395,
|
|
"epoch": 1.1403508771929824,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.19,
|
|
"mean_token_accuracy": 0.14433645009994506,
|
|
"num_tokens": 2424271.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.179539680480957,
|
|
"epoch": 1.1461988304093567,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.1922,
|
|
"mean_token_accuracy": 0.1384268395602703,
|
|
"num_tokens": 2436668.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.237603235244751,
|
|
"epoch": 1.1520467836257309,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1159,
|
|
"mean_token_accuracy": 0.139103065431118,
|
|
"num_tokens": 2449451.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.2184648513793945,
|
|
"epoch": 1.1578947368421053,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2528,
|
|
"mean_token_accuracy": 0.1350694477558136,
|
|
"num_tokens": 2463107.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.333126020431519,
|
|
"epoch": 1.1637426900584795,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.1829,
|
|
"mean_token_accuracy": 0.1379982963204384,
|
|
"num_tokens": 2475443.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.138613986968994,
|
|
"epoch": 1.1695906432748537,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.101,
|
|
"mean_token_accuracy": 0.14087243527173995,
|
|
"num_tokens": 2486884.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1695906432748537,
|
|
"eval_entropy": 7.2734953070108865,
|
|
"eval_loss": 7.307929992675781,
|
|
"eval_mean_token_accuracy": 0.1282176822561298,
|
|
"eval_num_tokens": 2486884.0,
|
|
"eval_runtime": 2.2594,
|
|
"eval_samples_per_second": 1198.118,
|
|
"eval_steps_per_second": 150.041,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.200505685806275,
|
|
"epoch": 1.1754385964912282,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999996875147963,
|
|
"loss": 7.0712,
|
|
"mean_token_accuracy": 0.1509634733200073,
|
|
"num_tokens": 2499310.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.196539449691772,
|
|
"epoch": 1.1812865497076024,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999984180451436,
|
|
"loss": 7.1829,
|
|
"mean_token_accuracy": 0.1330198511481285,
|
|
"num_tokens": 2512230.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.228861284255982,
|
|
"epoch": 1.1871345029239766,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999961720662221,
|
|
"loss": 7.2334,
|
|
"mean_token_accuracy": 0.134628976136446,
|
|
"num_tokens": 2525255.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.139020681381226,
|
|
"epoch": 1.1929824561403508,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999929495877796,
|
|
"loss": 7.1263,
|
|
"mean_token_accuracy": 0.14136102497577668,
|
|
"num_tokens": 2537672.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.20037808418274,
|
|
"epoch": 1.198830409356725,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004999887506238018,
|
|
"loss": 7.144,
|
|
"mean_token_accuracy": 0.13963264226913452,
|
|
"num_tokens": 2550371.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.154823732376099,
|
|
"epoch": 1.2046783625730995,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999835751925124,
|
|
"loss": 7.1484,
|
|
"mean_token_accuracy": 0.1348186582326889,
|
|
"num_tokens": 2563708.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.228437280654907,
|
|
"epoch": 1.2105263157894737,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999774233163734,
|
|
"loss": 7.1249,
|
|
"mean_token_accuracy": 0.13521802574396133,
|
|
"num_tokens": 2576529.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.180996465682983,
|
|
"epoch": 1.2163742690058479,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999702950220841,
|
|
"loss": 7.1407,
|
|
"mean_token_accuracy": 0.13425864800810813,
|
|
"num_tokens": 2588970.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.110421991348266,
|
|
"epoch": 1.2222222222222223,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999621903405819,
|
|
"loss": 7.0914,
|
|
"mean_token_accuracy": 0.1415217213332653,
|
|
"num_tokens": 2601134.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.224298429489136,
|
|
"epoch": 1.2280701754385965,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000499953109307042,
|
|
"loss": 7.151,
|
|
"mean_token_accuracy": 0.1363483726978302,
|
|
"num_tokens": 2612670.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.112039279937744,
|
|
"epoch": 1.2339181286549707,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999430519608762,
|
|
"loss": 7.1142,
|
|
"mean_token_accuracy": 0.1467585191130638,
|
|
"num_tokens": 2625157.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.188581943511963,
|
|
"epoch": 1.239766081871345,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999320183457346,
|
|
"loss": 7.1156,
|
|
"mean_token_accuracy": 0.13688333332538605,
|
|
"num_tokens": 2638318.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.182509040832519,
|
|
"epoch": 1.2456140350877192,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999200085095035,
|
|
"loss": 7.0975,
|
|
"mean_token_accuracy": 0.14040600806474685,
|
|
"num_tokens": 2651024.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.929991960525513,
|
|
"epoch": 1.2514619883040936,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004999070225043068,
|
|
"loss": 7.0076,
|
|
"mean_token_accuracy": 0.1484173998236656,
|
|
"num_tokens": 2665568.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.320855331420899,
|
|
"epoch": 1.2573099415204678,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004998930603865042,
|
|
"loss": 7.1486,
|
|
"mean_token_accuracy": 0.13886143118143082,
|
|
"num_tokens": 2678748.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.062246417999267,
|
|
"epoch": 1.263157894736842,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998781222166929,
|
|
"loss": 7.1028,
|
|
"mean_token_accuracy": 0.13795385062694548,
|
|
"num_tokens": 2690682.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.13513674736023,
|
|
"epoch": 1.2690058479532165,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998622080597052,
|
|
"loss": 7.1309,
|
|
"mean_token_accuracy": 0.13351953104138375,
|
|
"num_tokens": 2703077.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.121913051605224,
|
|
"epoch": 1.2748538011695907,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004998453179846098,
|
|
"loss": 7.1073,
|
|
"mean_token_accuracy": 0.1391266979277134,
|
|
"num_tokens": 2714609.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.14010066986084,
|
|
"epoch": 1.280701754385965,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998274520647109,
|
|
"loss": 7.0566,
|
|
"mean_token_accuracy": 0.14392231404781342,
|
|
"num_tokens": 2726907.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.031551790237427,
|
|
"epoch": 1.286549707602339,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998086103775478,
|
|
"loss": 7.0621,
|
|
"mean_token_accuracy": 0.13830177783966063,
|
|
"num_tokens": 2739248.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.132552528381348,
|
|
"epoch": 1.2923976608187133,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004997887930048947,
|
|
"loss": 7.014,
|
|
"mean_token_accuracy": 0.14461780115962028,
|
|
"num_tokens": 2753842.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.9925319194793705,
|
|
"epoch": 1.2982456140350878,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997680000327607,
|
|
"loss": 7.1127,
|
|
"mean_token_accuracy": 0.13322388678789138,
|
|
"num_tokens": 2767249.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.155430698394776,
|
|
"epoch": 1.304093567251462,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004997462315513884,
|
|
"loss": 7.0855,
|
|
"mean_token_accuracy": 0.14028178676962852,
|
|
"num_tokens": 2779657.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.046899271011353,
|
|
"epoch": 1.3099415204678362,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000499723487655255,
|
|
"loss": 6.9871,
|
|
"mean_token_accuracy": 0.14443906769156456,
|
|
"num_tokens": 2792760.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.05273756980896,
|
|
"epoch": 1.3157894736842106,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004996997684430703,
|
|
"loss": 7.0072,
|
|
"mean_token_accuracy": 0.14179261177778243,
|
|
"num_tokens": 2805277.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.0507885932922365,
|
|
"epoch": 1.3216374269005848,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996750740177775,
|
|
"loss": 7.1206,
|
|
"mean_token_accuracy": 0.13913794308900834,
|
|
"num_tokens": 2818341.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.174911212921143,
|
|
"epoch": 1.327485380116959,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004996494044865521,
|
|
"loss": 7.0709,
|
|
"mean_token_accuracy": 0.14009531438350678,
|
|
"num_tokens": 2830199.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.986618185043335,
|
|
"epoch": 1.3333333333333333,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004996227599608018,
|
|
"loss": 6.9673,
|
|
"mean_token_accuracy": 0.1467069461941719,
|
|
"num_tokens": 2842141.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.09279842376709,
|
|
"epoch": 1.3391812865497075,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995951405561658,
|
|
"loss": 7.0655,
|
|
"mean_token_accuracy": 0.13499290347099305,
|
|
"num_tokens": 2854929.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.056291580200195,
|
|
"epoch": 1.345029239766082,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995665463925142,
|
|
"loss": 7.0611,
|
|
"mean_token_accuracy": 0.14220297038555146,
|
|
"num_tokens": 2867034.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.0605995655059814,
|
|
"epoch": 1.3508771929824561,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004995369775939478,
|
|
"loss": 6.9696,
|
|
"mean_token_accuracy": 0.14844031184911727,
|
|
"num_tokens": 2881254.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.972619342803955,
|
|
"epoch": 1.3567251461988303,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004995064342887972,
|
|
"loss": 7.0747,
|
|
"mean_token_accuracy": 0.14061009883880615,
|
|
"num_tokens": 2894592.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.0729388236999515,
|
|
"epoch": 1.3625730994152048,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994749166096226,
|
|
"loss": 6.9918,
|
|
"mean_token_accuracy": 0.14350902065634727,
|
|
"num_tokens": 2907196.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.046811580657959,
|
|
"epoch": 1.368421052631579,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994424246932129,
|
|
"loss": 7.0659,
|
|
"mean_token_accuracy": 0.13506472632288932,
|
|
"num_tokens": 2920398.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.992996883392334,
|
|
"epoch": 1.3742690058479532,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004994089586805856,
|
|
"loss": 6.9787,
|
|
"mean_token_accuracy": 0.14449305981397628,
|
|
"num_tokens": 2932052.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.197365045547485,
|
|
"epoch": 1.3801169590643274,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993745187169852,
|
|
"loss": 7.0659,
|
|
"mean_token_accuracy": 0.13700218349695206,
|
|
"num_tokens": 2943620.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.975704765319824,
|
|
"epoch": 1.3859649122807016,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000499339104951884,
|
|
"loss": 7.0238,
|
|
"mean_token_accuracy": 0.15101929605007172,
|
|
"num_tokens": 2956159.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.0983617305755615,
|
|
"epoch": 1.391812865497076,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00049930271753898,
|
|
"loss": 7.0003,
|
|
"mean_token_accuracy": 0.14037386998534201,
|
|
"num_tokens": 2967332.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.987351369857788,
|
|
"epoch": 1.3976608187134503,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004992653566361974,
|
|
"loss": 7.0158,
|
|
"mean_token_accuracy": 0.13745561614632607,
|
|
"num_tokens": 2980168.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.061203384399414,
|
|
"epoch": 1.4035087719298245,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004992270224056849,
|
|
"loss": 7.0847,
|
|
"mean_token_accuracy": 0.1388035885989666,
|
|
"num_tokens": 2993509.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.067679357528687,
|
|
"epoch": 1.409356725146199,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004991877150138162,
|
|
"loss": 7.0136,
|
|
"mean_token_accuracy": 0.13967382907867432,
|
|
"num_tokens": 3005827.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.979962635040283,
|
|
"epoch": 1.4152046783625731,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004991474346311879,
|
|
"loss": 6.985,
|
|
"mean_token_accuracy": 0.14625776931643486,
|
|
"num_tokens": 3018968.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.0825128078460695,
|
|
"epoch": 1.4210526315789473,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991061814326198,
|
|
"loss": 7.0176,
|
|
"mean_token_accuracy": 0.13643715381622315,
|
|
"num_tokens": 3031892.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.010417175292969,
|
|
"epoch": 1.4269005847953216,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0004990639555971539,
|
|
"loss": 6.9633,
|
|
"mean_token_accuracy": 0.14731772094964982,
|
|
"num_tokens": 3045994.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.9524046897888185,
|
|
"epoch": 1.4327485380116958,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004990207573080533,
|
|
"loss": 6.9735,
|
|
"mean_token_accuracy": 0.14742070585489273,
|
|
"num_tokens": 3058304.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.96693868637085,
|
|
"epoch": 1.4385964912280702,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004989765867528014,
|
|
"loss": 6.9754,
|
|
"mean_token_accuracy": 0.1461901545524597,
|
|
"num_tokens": 3071102.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.980545520782471,
|
|
"epoch": 1.4444444444444444,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004989314441231019,
|
|
"loss": 7.0051,
|
|
"mean_token_accuracy": 0.13849530145525932,
|
|
"num_tokens": 3084653.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.004353618621826,
|
|
"epoch": 1.4502923976608186,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000498885329614877,
|
|
"loss": 7.0107,
|
|
"mean_token_accuracy": 0.1477586805820465,
|
|
"num_tokens": 3097425.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.01493558883667,
|
|
"epoch": 1.456140350877193,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004988382434282668,
|
|
"loss": 7.0225,
|
|
"mean_token_accuracy": 0.14672752395272254,
|
|
"num_tokens": 3111898.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.046136426925659,
|
|
"epoch": 1.4619883040935673,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000498790185767629,
|
|
"loss": 6.9569,
|
|
"mean_token_accuracy": 0.14396907165646552,
|
|
"num_tokens": 3124271.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.026466512680054,
|
|
"epoch": 1.4678362573099415,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000498741156841537,
|
|
"loss": 7.06,
|
|
"mean_token_accuracy": 0.14034834802150725,
|
|
"num_tokens": 3136344.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.956720781326294,
|
|
"epoch": 1.4736842105263157,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00049869115686278,
|
|
"loss": 6.9586,
|
|
"mean_token_accuracy": 0.14445274844765663,
|
|
"num_tokens": 3148299.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.0379047870635985,
|
|
"epoch": 1.47953216374269,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004986401860483615,
|
|
"loss": 7.0235,
|
|
"mean_token_accuracy": 0.13980134353041648,
|
|
"num_tokens": 3161537.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.993751955032349,
|
|
"epoch": 1.4853801169590644,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004985882446194984,
|
|
"loss": 6.9608,
|
|
"mean_token_accuracy": 0.14578410163521766,
|
|
"num_tokens": 3173676.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.023461723327637,
|
|
"epoch": 1.4912280701754386,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 0.0004985353328016204,
|
|
"loss": 6.9302,
|
|
"mean_token_accuracy": 0.1545111432671547,
|
|
"num_tokens": 3186542.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.9487221240997314,
|
|
"epoch": 1.4970760233918128,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984814508243685,
|
|
"loss": 6.9651,
|
|
"mean_token_accuracy": 0.14461245387792587,
|
|
"num_tokens": 3197867.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.958335924148559,
|
|
"epoch": 1.5029239766081872,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004984265989215942,
|
|
"loss": 6.9275,
|
|
"mean_token_accuracy": 0.1436944842338562,
|
|
"num_tokens": 3209671.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.9353508949279785,
|
|
"epoch": 1.5087719298245614,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004983707773313589,
|
|
"loss": 6.9134,
|
|
"mean_token_accuracy": 0.1483129307627678,
|
|
"num_tokens": 3220707.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.889391088485718,
|
|
"epoch": 1.5146198830409356,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004983139862959322,
|
|
"loss": 6.8839,
|
|
"mean_token_accuracy": 0.15239101126790047,
|
|
"num_tokens": 3235122.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.986405992507935,
|
|
"epoch": 1.52046783625731,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004982562260617912,
|
|
"loss": 6.8837,
|
|
"mean_token_accuracy": 0.14060088545083999,
|
|
"num_tokens": 3246833.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.970462274551392,
|
|
"epoch": 1.526315789473684,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004981974968796197,
|
|
"loss": 6.9859,
|
|
"mean_token_accuracy": 0.1401909612119198,
|
|
"num_tokens": 3260493.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.918372440338135,
|
|
"epoch": 1.5321637426900585,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004981377990043064,
|
|
"loss": 6.9112,
|
|
"mean_token_accuracy": 0.14301369413733483,
|
|
"num_tokens": 3272445.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.954419040679932,
|
|
"epoch": 1.5380116959064327,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004980771326949442,
|
|
"loss": 6.9368,
|
|
"mean_token_accuracy": 0.1427830085158348,
|
|
"num_tokens": 3284360.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.962142419815064,
|
|
"epoch": 1.543859649122807,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004980154982148296,
|
|
"loss": 6.9344,
|
|
"mean_token_accuracy": 0.1421595737338066,
|
|
"num_tokens": 3296841.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.923092269897461,
|
|
"epoch": 1.5497076023391814,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004979528958314604,
|
|
"loss": 6.9269,
|
|
"mean_token_accuracy": 0.14047582894563676,
|
|
"num_tokens": 3309393.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.016109418869019,
|
|
"epoch": 1.5555555555555556,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004978893258165354,
|
|
"loss": 7.0003,
|
|
"mean_token_accuracy": 0.14029522687196733,
|
|
"num_tokens": 3322094.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.91007022857666,
|
|
"epoch": 1.5614035087719298,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004978247884459532,
|
|
"loss": 6.9053,
|
|
"mean_token_accuracy": 0.14726511985063553,
|
|
"num_tokens": 3334468.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.969208383560181,
|
|
"epoch": 1.5672514619883042,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004977592839998105,
|
|
"loss": 6.895,
|
|
"mean_token_accuracy": 0.14278344213962554,
|
|
"num_tokens": 3346367.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.812116289138794,
|
|
"epoch": 1.5730994152046782,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976928127624013,
|
|
"loss": 6.8944,
|
|
"mean_token_accuracy": 0.14796961098909378,
|
|
"num_tokens": 3357463.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.954991054534912,
|
|
"epoch": 1.5789473684210527,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004976253750222157,
|
|
"loss": 6.8847,
|
|
"mean_token_accuracy": 0.14616121351718903,
|
|
"num_tokens": 3369465.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.887133073806763,
|
|
"epoch": 1.5847953216374269,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004975569710719382,
|
|
"loss": 6.9901,
|
|
"mean_token_accuracy": 0.1367621600627899,
|
|
"num_tokens": 3382859.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.019903707504272,
|
|
"epoch": 1.590643274853801,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004974876012084468,
|
|
"loss": 6.9454,
|
|
"mean_token_accuracy": 0.14300204291939736,
|
|
"num_tokens": 3395431.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.903563165664673,
|
|
"epoch": 1.5964912280701755,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974172657328116,
|
|
"loss": 6.8525,
|
|
"mean_token_accuracy": 0.14235405698418618,
|
|
"num_tokens": 3408388.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.882439708709716,
|
|
"epoch": 1.6023391812865497,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004973459649502939,
|
|
"loss": 6.8468,
|
|
"mean_token_accuracy": 0.14951836317777634,
|
|
"num_tokens": 3421275.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.905644369125366,
|
|
"epoch": 1.608187134502924,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004972736991703439,
|
|
"loss": 6.9167,
|
|
"mean_token_accuracy": 0.14661416113376619,
|
|
"num_tokens": 3433537.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.914070510864258,
|
|
"epoch": 1.6140350877192984,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004972004687066005,
|
|
"loss": 6.8715,
|
|
"mean_token_accuracy": 0.14761748984456063,
|
|
"num_tokens": 3445167.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.824334239959716,
|
|
"epoch": 1.6198830409356724,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000497126273876889,
|
|
"loss": 6.8875,
|
|
"mean_token_accuracy": 0.15272206366062163,
|
|
"num_tokens": 3458529.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.895606422424317,
|
|
"epoch": 1.6257309941520468,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004970511150032205,
|
|
"loss": 6.8778,
|
|
"mean_token_accuracy": 0.1458844795823097,
|
|
"num_tokens": 3470315.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.840101718902588,
|
|
"epoch": 1.631578947368421,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004969749924117896,
|
|
"loss": 6.785,
|
|
"mean_token_accuracy": 0.14661094844341277,
|
|
"num_tokens": 3481899.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.906247091293335,
|
|
"epoch": 1.6374269005847952,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004968979064329741,
|
|
"loss": 6.925,
|
|
"mean_token_accuracy": 0.14958905652165413,
|
|
"num_tokens": 3494076.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.980356884002686,
|
|
"epoch": 1.6432748538011697,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004968198574013323,
|
|
"loss": 6.9196,
|
|
"mean_token_accuracy": 0.1379689686000347,
|
|
"num_tokens": 3507670.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.925516939163208,
|
|
"epoch": 1.6491228070175439,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.000496740845655603,
|
|
"loss": 6.8189,
|
|
"mean_token_accuracy": 0.14898837506771087,
|
|
"num_tokens": 3519704.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.878632068634033,
|
|
"epoch": 1.654970760233918,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004966608715387029,
|
|
"loss": 6.8935,
|
|
"mean_token_accuracy": 0.14565708339214326,
|
|
"num_tokens": 3531665.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.823861360549927,
|
|
"epoch": 1.6608187134502925,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004965799353977248,
|
|
"loss": 6.9008,
|
|
"mean_token_accuracy": 0.15192211121320726,
|
|
"num_tokens": 3543876.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.818322944641113,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004964980375839379,
|
|
"loss": 6.8106,
|
|
"mean_token_accuracy": 0.1509052187204361,
|
|
"num_tokens": 3556517.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.901790475845337,
|
|
"epoch": 1.672514619883041,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004964151784527844,
|
|
"loss": 6.8247,
|
|
"mean_token_accuracy": 0.15195192098617555,
|
|
"num_tokens": 3569378.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.8896238803863525,
|
|
"epoch": 1.6783625730994152,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000496331358363879,
|
|
"loss": 6.8214,
|
|
"mean_token_accuracy": 0.14606776535511018,
|
|
"num_tokens": 3581684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.813811302185059,
|
|
"epoch": 1.6842105263157894,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004962465776810066,
|
|
"loss": 6.7664,
|
|
"mean_token_accuracy": 0.16046139299869538,
|
|
"num_tokens": 3594961.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.862590074539185,
|
|
"epoch": 1.6900584795321638,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004961608367721217,
|
|
"loss": 6.8377,
|
|
"mean_token_accuracy": 0.14302278757095338,
|
|
"num_tokens": 3607197.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.810559988021851,
|
|
"epoch": 1.695906432748538,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004960741360093459,
|
|
"loss": 6.8444,
|
|
"mean_token_accuracy": 0.14837275072932243,
|
|
"num_tokens": 3618728.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.915595960617066,
|
|
"epoch": 1.7017543859649122,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004959864757689669,
|
|
"loss": 6.959,
|
|
"mean_token_accuracy": 0.1320843756198883,
|
|
"num_tokens": 3630872.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.886625003814697,
|
|
"epoch": 1.7076023391812867,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004958978564314361,
|
|
"loss": 6.8478,
|
|
"mean_token_accuracy": 0.14627774357795714,
|
|
"num_tokens": 3642657.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.878891372680664,
|
|
"epoch": 1.7134502923976607,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495808278381368,
|
|
"loss": 6.8575,
|
|
"mean_token_accuracy": 0.14539648741483688,
|
|
"num_tokens": 3654505.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.732045078277588,
|
|
"epoch": 1.719298245614035,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004957177420075377,
|
|
"loss": 6.7578,
|
|
"mean_token_accuracy": 0.15864465534687042,
|
|
"num_tokens": 3668554.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.8867645263671875,
|
|
"epoch": 1.7251461988304093,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004956262477028797,
|
|
"loss": 6.9037,
|
|
"mean_token_accuracy": 0.14351411610841752,
|
|
"num_tokens": 3681448.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.818768930435181,
|
|
"epoch": 1.7309941520467835,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004955337958644857,
|
|
"loss": 6.8424,
|
|
"mean_token_accuracy": 0.1386769823729992,
|
|
"num_tokens": 3693677.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.844368124008179,
|
|
"epoch": 1.736842105263158,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004954403868936032,
|
|
"loss": 6.8392,
|
|
"mean_token_accuracy": 0.15232345163822175,
|
|
"num_tokens": 3706204.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.894862413406372,
|
|
"epoch": 1.7426900584795322,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000495346021195634,
|
|
"loss": 6.8016,
|
|
"mean_token_accuracy": 0.15067129582166672,
|
|
"num_tokens": 3718255.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.843251037597656,
|
|
"epoch": 1.7485380116959064,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004952506991801319,
|
|
"loss": 6.8052,
|
|
"mean_token_accuracy": 0.15079464316368102,
|
|
"num_tokens": 3730623.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.819522666931152,
|
|
"epoch": 1.7543859649122808,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004951544212608015,
|
|
"loss": 6.8522,
|
|
"mean_token_accuracy": 0.14364330172538758,
|
|
"num_tokens": 3743636.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7543859649122808,
|
|
"eval_entropy": 6.779693004304328,
|
|
"eval_loss": 6.981348991394043,
|
|
"eval_mean_token_accuracy": 0.1366379621155714,
|
|
"eval_num_tokens": 3743636.0,
|
|
"eval_runtime": 2.2598,
|
|
"eval_samples_per_second": 1197.882,
|
|
"eval_steps_per_second": 150.012,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 8540,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 5416136769024000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|