b5e9bda545
Model: fpadovani/jpn-jpan-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
6101 lines
166 KiB
JSON
6101 lines
166 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 3.5087719298245617,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 2.4548874139785766,
|
|
"epoch": 0.005847953216374269,
|
|
"grad_norm": 17.375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.5381,
|
|
"mean_token_accuracy": 0.03773948270827532,
|
|
"num_tokens": 13999.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 2.515810489654541,
|
|
"epoch": 0.011695906432748537,
|
|
"grad_norm": 15.6875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.4504,
|
|
"mean_token_accuracy": 0.03857097271829844,
|
|
"num_tokens": 25888.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 2.5658308982849123,
|
|
"epoch": 0.017543859649122806,
|
|
"grad_norm": 18.75,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.3376,
|
|
"mean_token_accuracy": 0.03801379818469286,
|
|
"num_tokens": 37699.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 2.648589539527893,
|
|
"epoch": 0.023391812865497075,
|
|
"grad_norm": 18.0,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 14.4331,
|
|
"mean_token_accuracy": 0.025170913990586997,
|
|
"num_tokens": 49648.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 3.067568039894104,
|
|
"epoch": 0.029239766081871343,
|
|
"grad_norm": 21.125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 13.8106,
|
|
"mean_token_accuracy": 0.042210881039500235,
|
|
"num_tokens": 62255.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 3.5193849325180055,
|
|
"epoch": 0.03508771929824561,
|
|
"grad_norm": 25.125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 13.4282,
|
|
"mean_token_accuracy": 0.03690041042864323,
|
|
"num_tokens": 76003.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 5.110470724105835,
|
|
"epoch": 0.04093567251461988,
|
|
"grad_norm": 39.25,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 12.305,
|
|
"mean_token_accuracy": 0.032853833958506586,
|
|
"num_tokens": 87585.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 8.690666055679321,
|
|
"epoch": 0.04678362573099415,
|
|
"grad_norm": 4.90625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.8954,
|
|
"mean_token_accuracy": 0.02273947522044182,
|
|
"num_tokens": 100404.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.664226055145264,
|
|
"epoch": 0.05263157894736842,
|
|
"grad_norm": 4.375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.6246,
|
|
"mean_token_accuracy": 0.01664919788017869,
|
|
"num_tokens": 113130.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.546917343139649,
|
|
"epoch": 0.05847953216374269,
|
|
"grad_norm": 3.875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.344,
|
|
"mean_token_accuracy": 0.02105573210865259,
|
|
"num_tokens": 125430.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.102593898773193,
|
|
"epoch": 0.06432748538011696,
|
|
"grad_norm": 4.375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.2003,
|
|
"mean_token_accuracy": 0.02809331100434065,
|
|
"num_tokens": 137550.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.388475513458252,
|
|
"epoch": 0.07017543859649122,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.0452,
|
|
"mean_token_accuracy": 0.03612718116492033,
|
|
"num_tokens": 149530.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.559987545013428,
|
|
"epoch": 0.07602339181286549,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.9327,
|
|
"mean_token_accuracy": 0.04171597119420767,
|
|
"num_tokens": 161671.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.477953338623047,
|
|
"epoch": 0.08187134502923976,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.8499,
|
|
"mean_token_accuracy": 0.04532963968813419,
|
|
"num_tokens": 174160.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.54740114212036,
|
|
"epoch": 0.08771929824561403,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.7707,
|
|
"mean_token_accuracy": 0.06920528337359429,
|
|
"num_tokens": 186969.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.44933729171753,
|
|
"epoch": 0.0935672514619883,
|
|
"grad_norm": 1.7578125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.658,
|
|
"mean_token_accuracy": 0.07367655821144581,
|
|
"num_tokens": 199470.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.356026935577393,
|
|
"epoch": 0.09941520467836257,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5258,
|
|
"mean_token_accuracy": 0.07762051969766617,
|
|
"num_tokens": 211306.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.388982677459717,
|
|
"epoch": 0.10526315789473684,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.521,
|
|
"mean_token_accuracy": 0.06988351345062256,
|
|
"num_tokens": 223323.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.449160861968995,
|
|
"epoch": 0.1111111111111111,
|
|
"grad_norm": 1.8515625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.404,
|
|
"mean_token_accuracy": 0.07955777905881405,
|
|
"num_tokens": 235896.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.24062376022339,
|
|
"epoch": 0.11695906432748537,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.3366,
|
|
"mean_token_accuracy": 0.07595778591930866,
|
|
"num_tokens": 247879.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.324538803100586,
|
|
"epoch": 0.12280701754385964,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.2463,
|
|
"mean_token_accuracy": 0.07249736785888672,
|
|
"num_tokens": 259714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.299348545074462,
|
|
"epoch": 0.1286549707602339,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.1954,
|
|
"mean_token_accuracy": 0.07774566560983658,
|
|
"num_tokens": 272429.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.246688842773438,
|
|
"epoch": 0.13450292397660818,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.1908,
|
|
"mean_token_accuracy": 0.07083970755338669,
|
|
"num_tokens": 285924.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.224361610412597,
|
|
"epoch": 0.14035087719298245,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.0459,
|
|
"mean_token_accuracy": 0.08458271101117135,
|
|
"num_tokens": 299453.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.05007495880127,
|
|
"epoch": 0.14619883040935672,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.0164,
|
|
"mean_token_accuracy": 0.08130386173725128,
|
|
"num_tokens": 312295.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.085463523864746,
|
|
"epoch": 0.15204678362573099,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.8884,
|
|
"mean_token_accuracy": 0.08093869239091873,
|
|
"num_tokens": 324730.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.985839366912842,
|
|
"epoch": 0.15789473684210525,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.8096,
|
|
"mean_token_accuracy": 0.08213753812015057,
|
|
"num_tokens": 336481.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.958092880249023,
|
|
"epoch": 0.16374269005847952,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.7771,
|
|
"mean_token_accuracy": 0.08334722444415092,
|
|
"num_tokens": 348987.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.810273170471191,
|
|
"epoch": 0.1695906432748538,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.6756,
|
|
"mean_token_accuracy": 0.0778221569955349,
|
|
"num_tokens": 360390.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.81429967880249,
|
|
"epoch": 0.17543859649122806,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.6545,
|
|
"mean_token_accuracy": 0.07858845368027687,
|
|
"num_tokens": 372456.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.693943881988526,
|
|
"epoch": 0.18128654970760233,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.5776,
|
|
"mean_token_accuracy": 0.07394571527838707,
|
|
"num_tokens": 384185.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.575428581237793,
|
|
"epoch": 0.1871345029239766,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.5065,
|
|
"mean_token_accuracy": 0.07995149567723274,
|
|
"num_tokens": 395399.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.371949768066406,
|
|
"epoch": 0.19298245614035087,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.5001,
|
|
"mean_token_accuracy": 0.0805008016526699,
|
|
"num_tokens": 408676.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.294499015808105,
|
|
"epoch": 0.19883040935672514,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4325,
|
|
"mean_token_accuracy": 0.08033415265381336,
|
|
"num_tokens": 421861.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.107025909423829,
|
|
"epoch": 0.2046783625730994,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.4751,
|
|
"mean_token_accuracy": 0.08693947605788707,
|
|
"num_tokens": 434852.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.985072135925293,
|
|
"epoch": 0.21052631578947367,
|
|
"grad_norm": 0.6875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.3966,
|
|
"mean_token_accuracy": 0.0785727895796299,
|
|
"num_tokens": 447313.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.946270751953126,
|
|
"epoch": 0.21637426900584794,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3362,
|
|
"mean_token_accuracy": 0.08314507529139518,
|
|
"num_tokens": 460271.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.779074096679688,
|
|
"epoch": 0.2222222222222222,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.2913,
|
|
"mean_token_accuracy": 0.0910579614341259,
|
|
"num_tokens": 473006.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.667890739440917,
|
|
"epoch": 0.22807017543859648,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.3421,
|
|
"mean_token_accuracy": 0.08133782222867011,
|
|
"num_tokens": 486100.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.728115272521972,
|
|
"epoch": 0.23391812865497075,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.2275,
|
|
"mean_token_accuracy": 0.08423927277326584,
|
|
"num_tokens": 499460.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.724910449981689,
|
|
"epoch": 0.23976608187134502,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.2439,
|
|
"mean_token_accuracy": 0.07881896123290062,
|
|
"num_tokens": 511795.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.58439531326294,
|
|
"epoch": 0.24561403508771928,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.295,
|
|
"mean_token_accuracy": 0.08954361081123352,
|
|
"num_tokens": 524897.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.694450759887696,
|
|
"epoch": 0.25146198830409355,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.2492,
|
|
"mean_token_accuracy": 0.08542842417955399,
|
|
"num_tokens": 538325.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.672403717041016,
|
|
"epoch": 0.2573099415204678,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.294,
|
|
"mean_token_accuracy": 0.08235822916030884,
|
|
"num_tokens": 550979.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.61720962524414,
|
|
"epoch": 0.2631578947368421,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.2583,
|
|
"mean_token_accuracy": 0.08766907453536987,
|
|
"num_tokens": 563001.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.610678768157959,
|
|
"epoch": 0.26900584795321636,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.1767,
|
|
"mean_token_accuracy": 0.08766172006726265,
|
|
"num_tokens": 573600.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.663910102844238,
|
|
"epoch": 0.27485380116959063,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.1709,
|
|
"mean_token_accuracy": 0.0914827212691307,
|
|
"num_tokens": 585287.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.543773651123047,
|
|
"epoch": 0.2807017543859649,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.2154,
|
|
"mean_token_accuracy": 0.08472376465797424,
|
|
"num_tokens": 597386.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.523555374145507,
|
|
"epoch": 0.28654970760233917,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.1833,
|
|
"mean_token_accuracy": 0.09036211296916008,
|
|
"num_tokens": 609256.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.552801322937011,
|
|
"epoch": 0.29239766081871343,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.1955,
|
|
"mean_token_accuracy": 0.08631851598620414,
|
|
"num_tokens": 621506.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.489137554168702,
|
|
"epoch": 0.2982456140350877,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.0989,
|
|
"mean_token_accuracy": 0.09788421541452408,
|
|
"num_tokens": 633512.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.542137622833252,
|
|
"epoch": 0.30409356725146197,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.2385,
|
|
"mean_token_accuracy": 0.0867220088839531,
|
|
"num_tokens": 647098.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.521434879302978,
|
|
"epoch": 0.30994152046783624,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.2165,
|
|
"mean_token_accuracy": 0.08817893266677856,
|
|
"num_tokens": 659559.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.531222820281982,
|
|
"epoch": 0.3157894736842105,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.1671,
|
|
"mean_token_accuracy": 0.08971341624855995,
|
|
"num_tokens": 672621.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.470209407806397,
|
|
"epoch": 0.3216374269005848,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.1267,
|
|
"mean_token_accuracy": 0.09204860255122185,
|
|
"num_tokens": 685652.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.486028385162353,
|
|
"epoch": 0.32748538011695905,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.1752,
|
|
"mean_token_accuracy": 0.09480618685483932,
|
|
"num_tokens": 696910.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.44861536026001,
|
|
"epoch": 0.3333333333333333,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.0682,
|
|
"mean_token_accuracy": 0.09902379810810089,
|
|
"num_tokens": 709054.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.478011894226075,
|
|
"epoch": 0.3391812865497076,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.0885,
|
|
"mean_token_accuracy": 0.09357146173715591,
|
|
"num_tokens": 720902.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.469257354736328,
|
|
"epoch": 0.34502923976608185,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.153,
|
|
"mean_token_accuracy": 0.09559885188937187,
|
|
"num_tokens": 733319.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.412768745422364,
|
|
"epoch": 0.3508771929824561,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.0329,
|
|
"mean_token_accuracy": 0.09888283610343933,
|
|
"num_tokens": 744817.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.385979843139648,
|
|
"epoch": 0.3567251461988304,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.0906,
|
|
"mean_token_accuracy": 0.09885545223951339,
|
|
"num_tokens": 756582.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.444335651397704,
|
|
"epoch": 0.36257309941520466,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.0097,
|
|
"mean_token_accuracy": 0.10672620832920074,
|
|
"num_tokens": 769273.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.386385726928712,
|
|
"epoch": 0.3684210526315789,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.0753,
|
|
"mean_token_accuracy": 0.09999166950583457,
|
|
"num_tokens": 782001.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.315172958374024,
|
|
"epoch": 0.3742690058479532,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.0586,
|
|
"mean_token_accuracy": 0.10521865040063857,
|
|
"num_tokens": 794575.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.437547206878662,
|
|
"epoch": 0.38011695906432746,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.0377,
|
|
"mean_token_accuracy": 0.10137710124254226,
|
|
"num_tokens": 805583.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.299012565612793,
|
|
"epoch": 0.38596491228070173,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.0804,
|
|
"mean_token_accuracy": 0.10292363837361336,
|
|
"num_tokens": 817617.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.361987209320068,
|
|
"epoch": 0.391812865497076,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.9841,
|
|
"mean_token_accuracy": 0.10257630646228791,
|
|
"num_tokens": 830043.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.306623458862305,
|
|
"epoch": 0.39766081871345027,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.9788,
|
|
"mean_token_accuracy": 0.11217362359166146,
|
|
"num_tokens": 843200.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.317947292327881,
|
|
"epoch": 0.40350877192982454,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.9994,
|
|
"mean_token_accuracy": 0.11003749370574951,
|
|
"num_tokens": 856031.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.315475559234619,
|
|
"epoch": 0.4093567251461988,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.9637,
|
|
"mean_token_accuracy": 0.11112827882170677,
|
|
"num_tokens": 869119.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.308537578582763,
|
|
"epoch": 0.4152046783625731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.984,
|
|
"mean_token_accuracy": 0.1118388794362545,
|
|
"num_tokens": 880802.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.310247135162353,
|
|
"epoch": 0.42105263157894735,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.9752,
|
|
"mean_token_accuracy": 0.11066636517643928,
|
|
"num_tokens": 894237.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.24343318939209,
|
|
"epoch": 0.4269005847953216,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.99,
|
|
"mean_token_accuracy": 0.10824075117707252,
|
|
"num_tokens": 906695.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.380464553833008,
|
|
"epoch": 0.4327485380116959,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.9536,
|
|
"mean_token_accuracy": 0.11111580803990365,
|
|
"num_tokens": 918955.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.231604385375977,
|
|
"epoch": 0.43859649122807015,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.9735,
|
|
"mean_token_accuracy": 0.11058306768536567,
|
|
"num_tokens": 931812.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.334205436706544,
|
|
"epoch": 0.4444444444444444,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.9795,
|
|
"mean_token_accuracy": 0.10282116234302521,
|
|
"num_tokens": 945261.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.210968399047852,
|
|
"epoch": 0.4502923976608187,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.8956,
|
|
"mean_token_accuracy": 0.11960833892226219,
|
|
"num_tokens": 958376.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.18839659690857,
|
|
"epoch": 0.45614035087719296,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.9245,
|
|
"mean_token_accuracy": 0.11952447071671486,
|
|
"num_tokens": 971709.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.367930507659912,
|
|
"epoch": 0.4619883040935672,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.8919,
|
|
"mean_token_accuracy": 0.11701075136661529,
|
|
"num_tokens": 983847.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.25608139038086,
|
|
"epoch": 0.4678362573099415,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.9585,
|
|
"mean_token_accuracy": 0.10782488510012626,
|
|
"num_tokens": 997034.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.187435054779053,
|
|
"epoch": 0.47368421052631576,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.9853,
|
|
"mean_token_accuracy": 0.11163778379559516,
|
|
"num_tokens": 1010213.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.300292873382569,
|
|
"epoch": 0.47953216374269003,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.8797,
|
|
"mean_token_accuracy": 0.11556096971035004,
|
|
"num_tokens": 1021618.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.165176582336425,
|
|
"epoch": 0.4853801169590643,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.8907,
|
|
"mean_token_accuracy": 0.11225929781794548,
|
|
"num_tokens": 1033809.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.331029987335205,
|
|
"epoch": 0.49122807017543857,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.888,
|
|
"mean_token_accuracy": 0.11076272949576378,
|
|
"num_tokens": 1045291.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.092021656036376,
|
|
"epoch": 0.49707602339181284,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.8613,
|
|
"mean_token_accuracy": 0.11468090787529946,
|
|
"num_tokens": 1057516.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.167704010009766,
|
|
"epoch": 0.5029239766081871,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.8082,
|
|
"mean_token_accuracy": 0.12245556116104125,
|
|
"num_tokens": 1070675.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.19180703163147,
|
|
"epoch": 0.5087719298245614,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.888,
|
|
"mean_token_accuracy": 0.12304584905505181,
|
|
"num_tokens": 1082965.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.234239482879639,
|
|
"epoch": 0.5146198830409356,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.858,
|
|
"mean_token_accuracy": 0.112663733959198,
|
|
"num_tokens": 1095922.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.179874038696289,
|
|
"epoch": 0.52046783625731,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.8413,
|
|
"mean_token_accuracy": 0.11735650300979614,
|
|
"num_tokens": 1109002.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.141633415222168,
|
|
"epoch": 0.5263157894736842,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.8662,
|
|
"mean_token_accuracy": 0.11112897917628288,
|
|
"num_tokens": 1121199.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.179853630065917,
|
|
"epoch": 0.5321637426900585,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.8017,
|
|
"mean_token_accuracy": 0.12308088317513466,
|
|
"num_tokens": 1134721.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.180206298828125,
|
|
"epoch": 0.5380116959064327,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.8632,
|
|
"mean_token_accuracy": 0.11824943199753761,
|
|
"num_tokens": 1147954.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.030020809173584,
|
|
"epoch": 0.543859649122807,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.81,
|
|
"mean_token_accuracy": 0.1231809377670288,
|
|
"num_tokens": 1161208.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.183233261108398,
|
|
"epoch": 0.5497076023391813,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.9011,
|
|
"mean_token_accuracy": 0.11757142841815948,
|
|
"num_tokens": 1173789.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.235744667053222,
|
|
"epoch": 0.5555555555555556,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.7889,
|
|
"mean_token_accuracy": 0.12023304700851441,
|
|
"num_tokens": 1185765.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.98179259300232,
|
|
"epoch": 0.5614035087719298,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.8028,
|
|
"mean_token_accuracy": 0.1243036225438118,
|
|
"num_tokens": 1197954.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.164046669006348,
|
|
"epoch": 0.5672514619883041,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.7434,
|
|
"mean_token_accuracy": 0.12476142644882202,
|
|
"num_tokens": 1211285.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.129650735855103,
|
|
"epoch": 0.5730994152046783,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.8072,
|
|
"mean_token_accuracy": 0.1169096663594246,
|
|
"num_tokens": 1223197.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.043150568008423,
|
|
"epoch": 0.5789473684210527,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.7402,
|
|
"mean_token_accuracy": 0.12265990749001503,
|
|
"num_tokens": 1234817.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.130622863769531,
|
|
"epoch": 0.5847953216374269,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.8029,
|
|
"mean_token_accuracy": 0.12631115764379502,
|
|
"num_tokens": 1249305.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5847953216374269,
|
|
"eval_entropy": 7.920137181746221,
|
|
"eval_loss": 7.822256088256836,
|
|
"eval_mean_token_accuracy": 0.1146006511996849,
|
|
"eval_num_tokens": 1249305.0,
|
|
"eval_runtime": 1.74,
|
|
"eval_samples_per_second": 1555.772,
|
|
"eval_steps_per_second": 194.831,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.136149835586547,
|
|
"epoch": 0.5906432748538012,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.7717,
|
|
"mean_token_accuracy": 0.12191700711846351,
|
|
"num_tokens": 1262182.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.012873411178589,
|
|
"epoch": 0.5964912280701754,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.7877,
|
|
"mean_token_accuracy": 0.12733336314558982,
|
|
"num_tokens": 1274616.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.046648216247558,
|
|
"epoch": 0.6023391812865497,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.7911,
|
|
"mean_token_accuracy": 0.11969717964529991,
|
|
"num_tokens": 1287306.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.059616661071777,
|
|
"epoch": 0.6081871345029239,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.6474,
|
|
"mean_token_accuracy": 0.12940788641571999,
|
|
"num_tokens": 1299472.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.013744401931763,
|
|
"epoch": 0.6140350877192983,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.6178,
|
|
"mean_token_accuracy": 0.13267553746700286,
|
|
"num_tokens": 1312636.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.913894128799439,
|
|
"epoch": 0.6198830409356725,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.659,
|
|
"mean_token_accuracy": 0.13482385575771333,
|
|
"num_tokens": 1324058.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.036643075942994,
|
|
"epoch": 0.6257309941520468,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.7684,
|
|
"mean_token_accuracy": 0.11713232323527337,
|
|
"num_tokens": 1337605.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.078272104263306,
|
|
"epoch": 0.631578947368421,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.777,
|
|
"mean_token_accuracy": 0.1210218109190464,
|
|
"num_tokens": 1350658.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.057914018630981,
|
|
"epoch": 0.6374269005847953,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.707,
|
|
"mean_token_accuracy": 0.12556772753596307,
|
|
"num_tokens": 1362091.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.995908594131469,
|
|
"epoch": 0.6432748538011696,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.704,
|
|
"mean_token_accuracy": 0.12518641129136085,
|
|
"num_tokens": 1374055.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.98234977722168,
|
|
"epoch": 0.6491228070175439,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.58,
|
|
"mean_token_accuracy": 0.12927438765764238,
|
|
"num_tokens": 1385269.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.961950588226318,
|
|
"epoch": 0.6549707602339181,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.6932,
|
|
"mean_token_accuracy": 0.13011750653386117,
|
|
"num_tokens": 1396559.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.937076187133789,
|
|
"epoch": 0.6608187134502924,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.7119,
|
|
"mean_token_accuracy": 0.13076897338032722,
|
|
"num_tokens": 1408838.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.980747890472412,
|
|
"epoch": 0.6666666666666666,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.5961,
|
|
"mean_token_accuracy": 0.1342053160071373,
|
|
"num_tokens": 1421023.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.89539532661438,
|
|
"epoch": 0.672514619883041,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.6598,
|
|
"mean_token_accuracy": 0.13019674867391587,
|
|
"num_tokens": 1433912.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.000441455841065,
|
|
"epoch": 0.6783625730994152,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.6742,
|
|
"mean_token_accuracy": 0.12573233470320702,
|
|
"num_tokens": 1445262.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.876591634750366,
|
|
"epoch": 0.6842105263157895,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.6415,
|
|
"mean_token_accuracy": 0.12933589220046998,
|
|
"num_tokens": 1457927.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.937921953201294,
|
|
"epoch": 0.6900584795321637,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.6915,
|
|
"mean_token_accuracy": 0.12511010393500327,
|
|
"num_tokens": 1470116.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.94811863899231,
|
|
"epoch": 0.695906432748538,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.5772,
|
|
"mean_token_accuracy": 0.13277097642421723,
|
|
"num_tokens": 1482067.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.93737530708313,
|
|
"epoch": 0.7017543859649122,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.6747,
|
|
"mean_token_accuracy": 0.12135898694396019,
|
|
"num_tokens": 1493902.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.944917917251587,
|
|
"epoch": 0.7076023391812866,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.6329,
|
|
"mean_token_accuracy": 0.12311855405569076,
|
|
"num_tokens": 1505689.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.876118469238281,
|
|
"epoch": 0.7134502923976608,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.5735,
|
|
"mean_token_accuracy": 0.1439603090286255,
|
|
"num_tokens": 1518002.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.9030755996704105,
|
|
"epoch": 0.7192982456140351,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.601,
|
|
"mean_token_accuracy": 0.12741394564509392,
|
|
"num_tokens": 1529307.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.934239673614502,
|
|
"epoch": 0.7251461988304093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7544,
|
|
"mean_token_accuracy": 0.11436132788658142,
|
|
"num_tokens": 1542351.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.823674011230469,
|
|
"epoch": 0.7309941520467836,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.5697,
|
|
"mean_token_accuracy": 0.1321065440773964,
|
|
"num_tokens": 1556263.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.873884916305542,
|
|
"epoch": 0.7368421052631579,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.5818,
|
|
"mean_token_accuracy": 0.13355903923511506,
|
|
"num_tokens": 1568906.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.839196968078613,
|
|
"epoch": 0.7426900584795322,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.5782,
|
|
"mean_token_accuracy": 0.132675963640213,
|
|
"num_tokens": 1580631.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.838891458511353,
|
|
"epoch": 0.7485380116959064,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.5937,
|
|
"mean_token_accuracy": 0.12946432083845139,
|
|
"num_tokens": 1592369.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.938169670104981,
|
|
"epoch": 0.7543859649122807,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.6258,
|
|
"mean_token_accuracy": 0.13315243050456047,
|
|
"num_tokens": 1604888.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.889529895782471,
|
|
"epoch": 0.7602339181286549,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.5451,
|
|
"mean_token_accuracy": 0.13434720635414124,
|
|
"num_tokens": 1617128.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.793774127960205,
|
|
"epoch": 0.7660818713450293,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.5885,
|
|
"mean_token_accuracy": 0.1243424341082573,
|
|
"num_tokens": 1629350.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.897194147109985,
|
|
"epoch": 0.7719298245614035,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.6016,
|
|
"mean_token_accuracy": 0.1256876789033413,
|
|
"num_tokens": 1643082.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.854056262969971,
|
|
"epoch": 0.7777777777777778,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.5822,
|
|
"mean_token_accuracy": 0.13232242465019226,
|
|
"num_tokens": 1654785.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.879878759384155,
|
|
"epoch": 0.783625730994152,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.6061,
|
|
"mean_token_accuracy": 0.12869918420910836,
|
|
"num_tokens": 1667704.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.890443611145019,
|
|
"epoch": 0.7894736842105263,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.5665,
|
|
"mean_token_accuracy": 0.12847531214356422,
|
|
"num_tokens": 1680459.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.804240417480469,
|
|
"epoch": 0.7953216374269005,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.4724,
|
|
"mean_token_accuracy": 0.13116607591509818,
|
|
"num_tokens": 1692912.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.765193080902099,
|
|
"epoch": 0.8011695906432749,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.5031,
|
|
"mean_token_accuracy": 0.13217658475041388,
|
|
"num_tokens": 1705329.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.918375539779663,
|
|
"epoch": 0.8070175438596491,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.5958,
|
|
"mean_token_accuracy": 0.1265776574611664,
|
|
"num_tokens": 1717283.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.726377534866333,
|
|
"epoch": 0.8128654970760234,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.5333,
|
|
"mean_token_accuracy": 0.13292808830738068,
|
|
"num_tokens": 1730308.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.862821102142334,
|
|
"epoch": 0.8187134502923976,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.5828,
|
|
"mean_token_accuracy": 0.13102516084909438,
|
|
"num_tokens": 1742716.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.830471420288086,
|
|
"epoch": 0.8245614035087719,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.5287,
|
|
"mean_token_accuracy": 0.13678008690476418,
|
|
"num_tokens": 1754183.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.751367998123169,
|
|
"epoch": 0.8304093567251462,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.5464,
|
|
"mean_token_accuracy": 0.12743634656071662,
|
|
"num_tokens": 1766601.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.780935764312744,
|
|
"epoch": 0.8362573099415205,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.4354,
|
|
"mean_token_accuracy": 0.14405448883771896,
|
|
"num_tokens": 1778756.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.539671182632446,
|
|
"epoch": 0.8421052631578947,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.4979,
|
|
"mean_token_accuracy": 0.13751505464315414,
|
|
"num_tokens": 1791485.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.818372631072998,
|
|
"epoch": 0.847953216374269,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.518,
|
|
"mean_token_accuracy": 0.1304582789540291,
|
|
"num_tokens": 1803712.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.721041631698609,
|
|
"epoch": 0.8538011695906432,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.4694,
|
|
"mean_token_accuracy": 0.13335771411657332,
|
|
"num_tokens": 1818387.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.743247652053833,
|
|
"epoch": 0.8596491228070176,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.4671,
|
|
"mean_token_accuracy": 0.13619914576411246,
|
|
"num_tokens": 1830113.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.735281753540039,
|
|
"epoch": 0.8654970760233918,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.5761,
|
|
"mean_token_accuracy": 0.13483594059944154,
|
|
"num_tokens": 1841789.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.668903923034668,
|
|
"epoch": 0.8713450292397661,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.4507,
|
|
"mean_token_accuracy": 0.13731630146503448,
|
|
"num_tokens": 1853669.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.724067640304566,
|
|
"epoch": 0.8771929824561403,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.5109,
|
|
"mean_token_accuracy": 0.13437143638730048,
|
|
"num_tokens": 1865322.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.644190073013306,
|
|
"epoch": 0.8830409356725146,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.3942,
|
|
"mean_token_accuracy": 0.1427222914993763,
|
|
"num_tokens": 1879039.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.785293626785278,
|
|
"epoch": 0.8888888888888888,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.5554,
|
|
"mean_token_accuracy": 0.12891999259591103,
|
|
"num_tokens": 1893078.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.748642826080323,
|
|
"epoch": 0.8947368421052632,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.5767,
|
|
"mean_token_accuracy": 0.12865981981158256,
|
|
"num_tokens": 1907399.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.704698705673218,
|
|
"epoch": 0.9005847953216374,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.5235,
|
|
"mean_token_accuracy": 0.12384044975042344,
|
|
"num_tokens": 1919290.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.804563283920288,
|
|
"epoch": 0.9064327485380117,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.4842,
|
|
"mean_token_accuracy": 0.12996885776519776,
|
|
"num_tokens": 1931658.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.707001209259033,
|
|
"epoch": 0.9122807017543859,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.4773,
|
|
"mean_token_accuracy": 0.1285408228635788,
|
|
"num_tokens": 1943881.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.667708969116211,
|
|
"epoch": 0.9181286549707602,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.4053,
|
|
"mean_token_accuracy": 0.1371392272412777,
|
|
"num_tokens": 1956288.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.6015739917755125,
|
|
"epoch": 0.9239766081871345,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.3066,
|
|
"mean_token_accuracy": 0.13772179931402206,
|
|
"num_tokens": 1967265.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.597353601455689,
|
|
"epoch": 0.9298245614035088,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.447,
|
|
"mean_token_accuracy": 0.13603045642375947,
|
|
"num_tokens": 1980039.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.690238094329834,
|
|
"epoch": 0.935672514619883,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.4093,
|
|
"mean_token_accuracy": 0.13160398453474045,
|
|
"num_tokens": 1992735.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.704154777526855,
|
|
"epoch": 0.9415204678362573,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.4332,
|
|
"mean_token_accuracy": 0.13839306458830833,
|
|
"num_tokens": 2004545.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.581313180923462,
|
|
"epoch": 0.9473684210526315,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.4323,
|
|
"mean_token_accuracy": 0.13345621451735495,
|
|
"num_tokens": 2017100.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.68658595085144,
|
|
"epoch": 0.9532163742690059,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.3636,
|
|
"mean_token_accuracy": 0.14662561491131781,
|
|
"num_tokens": 2029868.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.646354150772095,
|
|
"epoch": 0.9590643274853801,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.4397,
|
|
"mean_token_accuracy": 0.13421894684433938,
|
|
"num_tokens": 2041334.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.6902111053466795,
|
|
"epoch": 0.9649122807017544,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5138,
|
|
"mean_token_accuracy": 0.13264959007501603,
|
|
"num_tokens": 2054797.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.602681684494018,
|
|
"epoch": 0.9707602339181286,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.4045,
|
|
"mean_token_accuracy": 0.13147154748439788,
|
|
"num_tokens": 2068337.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.694418144226074,
|
|
"epoch": 0.9766081871345029,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.4365,
|
|
"mean_token_accuracy": 0.13966115042567254,
|
|
"num_tokens": 2079984.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.612453842163086,
|
|
"epoch": 0.9824561403508771,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.4319,
|
|
"mean_token_accuracy": 0.13970493376255036,
|
|
"num_tokens": 2093069.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.611123704910279,
|
|
"epoch": 0.9883040935672515,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.3801,
|
|
"mean_token_accuracy": 0.13889345675706863,
|
|
"num_tokens": 2104753.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.604637861251831,
|
|
"epoch": 0.9941520467836257,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.391,
|
|
"mean_token_accuracy": 0.13473013713955878,
|
|
"num_tokens": 2116576.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.654403877258301,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.3747,
|
|
"mean_token_accuracy": 0.13166687712073327,
|
|
"num_tokens": 2129484.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.537336349487305,
|
|
"epoch": 1.0058479532163742,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.2449,
|
|
"mean_token_accuracy": 0.14783877283334732,
|
|
"num_tokens": 2142000.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.523657131195068,
|
|
"epoch": 1.0116959064327484,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.176,
|
|
"mean_token_accuracy": 0.1486393131315708,
|
|
"num_tokens": 2152965.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.559857559204102,
|
|
"epoch": 1.0175438596491229,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.2756,
|
|
"mean_token_accuracy": 0.13471193611621857,
|
|
"num_tokens": 2165078.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.430526161193848,
|
|
"epoch": 1.023391812865497,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.2031,
|
|
"mean_token_accuracy": 0.14576031267642975,
|
|
"num_tokens": 2176706.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.623335981369019,
|
|
"epoch": 1.0292397660818713,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.2517,
|
|
"mean_token_accuracy": 0.13940724432468415,
|
|
"num_tokens": 2190157.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.54054594039917,
|
|
"epoch": 1.0350877192982457,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.2133,
|
|
"mean_token_accuracy": 0.14537803903222085,
|
|
"num_tokens": 2201821.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.468587160110474,
|
|
"epoch": 1.04093567251462,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.158,
|
|
"mean_token_accuracy": 0.15055620521306992,
|
|
"num_tokens": 2214504.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.467399740219117,
|
|
"epoch": 1.0467836257309941,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.2665,
|
|
"mean_token_accuracy": 0.14105323553085328,
|
|
"num_tokens": 2226318.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.555955314636231,
|
|
"epoch": 1.0526315789473684,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.2829,
|
|
"mean_token_accuracy": 0.13729312494397164,
|
|
"num_tokens": 2237892.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.526016616821289,
|
|
"epoch": 1.0584795321637426,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.2314,
|
|
"mean_token_accuracy": 0.1435258150100708,
|
|
"num_tokens": 2250656.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.511065912246704,
|
|
"epoch": 1.064327485380117,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.2637,
|
|
"mean_token_accuracy": 0.13697190061211587,
|
|
"num_tokens": 2263685.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.494118690490723,
|
|
"epoch": 1.0701754385964912,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.232,
|
|
"mean_token_accuracy": 0.1364179164171219,
|
|
"num_tokens": 2275413.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.477536964416504,
|
|
"epoch": 1.0760233918128654,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.2195,
|
|
"mean_token_accuracy": 0.14017109125852584,
|
|
"num_tokens": 2288165.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.574978637695312,
|
|
"epoch": 1.0818713450292399,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.2053,
|
|
"mean_token_accuracy": 0.13687562569975853,
|
|
"num_tokens": 2300574.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.416238403320312,
|
|
"epoch": 1.087719298245614,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.219,
|
|
"mean_token_accuracy": 0.1380111664533615,
|
|
"num_tokens": 2311631.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.547397470474243,
|
|
"epoch": 1.0935672514619883,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.234,
|
|
"mean_token_accuracy": 0.14326555654406548,
|
|
"num_tokens": 2324312.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.512025690078735,
|
|
"epoch": 1.0994152046783625,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.2855,
|
|
"mean_token_accuracy": 0.13624050617218017,
|
|
"num_tokens": 2337747.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.413490962982178,
|
|
"epoch": 1.1052631578947367,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.2546,
|
|
"mean_token_accuracy": 0.1424066536128521,
|
|
"num_tokens": 2349830.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.565069675445557,
|
|
"epoch": 1.1111111111111112,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.1993,
|
|
"mean_token_accuracy": 0.14769526124000548,
|
|
"num_tokens": 2361825.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.369341516494751,
|
|
"epoch": 1.1169590643274854,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.1361,
|
|
"mean_token_accuracy": 0.1449691839516163,
|
|
"num_tokens": 2373963.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.453041648864746,
|
|
"epoch": 1.1228070175438596,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.2795,
|
|
"mean_token_accuracy": 0.13922639787197114,
|
|
"num_tokens": 2386348.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.489872980117798,
|
|
"epoch": 1.128654970760234,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.1985,
|
|
"mean_token_accuracy": 0.13827377706766128,
|
|
"num_tokens": 2398338.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.514949655532837,
|
|
"epoch": 1.1345029239766082,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.1533,
|
|
"mean_token_accuracy": 0.14474421963095666,
|
|
"num_tokens": 2411855.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.357582998275757,
|
|
"epoch": 1.1403508771929824,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.2126,
|
|
"mean_token_accuracy": 0.14526815563440323,
|
|
"num_tokens": 2424271.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.440734958648681,
|
|
"epoch": 1.1461988304093567,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.2256,
|
|
"mean_token_accuracy": 0.13997550830245017,
|
|
"num_tokens": 2436668.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.424071168899536,
|
|
"epoch": 1.1520467836257309,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.1748,
|
|
"mean_token_accuracy": 0.1401802718639374,
|
|
"num_tokens": 2449451.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.5277848720550535,
|
|
"epoch": 1.1578947368421053,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2966,
|
|
"mean_token_accuracy": 0.13498533368110657,
|
|
"num_tokens": 2463107.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.464729595184326,
|
|
"epoch": 1.1637426900584795,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.2098,
|
|
"mean_token_accuracy": 0.14099321886897087,
|
|
"num_tokens": 2475443.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.4347583770751955,
|
|
"epoch": 1.1695906432748537,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.1595,
|
|
"mean_token_accuracy": 0.13942319229245187,
|
|
"num_tokens": 2486884.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1695906432748537,
|
|
"eval_entropy": 7.396586781054472,
|
|
"eval_loss": 7.327540397644043,
|
|
"eval_mean_token_accuracy": 0.13058083411306143,
|
|
"eval_num_tokens": 2486884.0,
|
|
"eval_runtime": 1.7343,
|
|
"eval_samples_per_second": 1560.818,
|
|
"eval_steps_per_second": 195.463,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.410801887512207,
|
|
"epoch": 1.1754385964912282,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999996875147963,
|
|
"loss": 7.1007,
|
|
"mean_token_accuracy": 0.15155821964144706,
|
|
"num_tokens": 2499310.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.434916257858276,
|
|
"epoch": 1.1812865497076024,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999984180451436,
|
|
"loss": 7.2163,
|
|
"mean_token_accuracy": 0.1340879425406456,
|
|
"num_tokens": 2512230.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.411216306686401,
|
|
"epoch": 1.1871345029239766,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999961720662221,
|
|
"loss": 7.274,
|
|
"mean_token_accuracy": 0.13820524960756303,
|
|
"num_tokens": 2525255.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.425869369506836,
|
|
"epoch": 1.1929824561403508,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999929495877796,
|
|
"loss": 7.1618,
|
|
"mean_token_accuracy": 0.14160262942314147,
|
|
"num_tokens": 2537672.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.341161251068115,
|
|
"epoch": 1.198830409356725,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999887506238018,
|
|
"loss": 7.1885,
|
|
"mean_token_accuracy": 0.14109369292855262,
|
|
"num_tokens": 2550371.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.4244989395141605,
|
|
"epoch": 1.2046783625730995,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004999835751925124,
|
|
"loss": 7.2077,
|
|
"mean_token_accuracy": 0.13407954722642898,
|
|
"num_tokens": 2563708.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.464514541625976,
|
|
"epoch": 1.2105263157894737,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004999774233163734,
|
|
"loss": 7.1749,
|
|
"mean_token_accuracy": 0.13515729755163192,
|
|
"num_tokens": 2576529.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.382573938369751,
|
|
"epoch": 1.2163742690058479,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004999702950220841,
|
|
"loss": 7.189,
|
|
"mean_token_accuracy": 0.13476522713899614,
|
|
"num_tokens": 2588970.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.370054435729981,
|
|
"epoch": 1.2222222222222223,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999621903405819,
|
|
"loss": 7.126,
|
|
"mean_token_accuracy": 0.1400519900023937,
|
|
"num_tokens": 2601134.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.446572351455688,
|
|
"epoch": 1.2280701754385965,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499953109307042,
|
|
"loss": 7.1968,
|
|
"mean_token_accuracy": 0.13780457451939582,
|
|
"num_tokens": 2612670.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.330738973617554,
|
|
"epoch": 1.2339181286549707,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999430519608762,
|
|
"loss": 7.1517,
|
|
"mean_token_accuracy": 0.1500091567635536,
|
|
"num_tokens": 2625157.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.424869203567505,
|
|
"epoch": 1.239766081871345,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004999320183457346,
|
|
"loss": 7.1486,
|
|
"mean_token_accuracy": 0.1392211377620697,
|
|
"num_tokens": 2638318.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.385347890853882,
|
|
"epoch": 1.2456140350877192,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999200085095035,
|
|
"loss": 7.134,
|
|
"mean_token_accuracy": 0.1431312918663025,
|
|
"num_tokens": 2651024.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.197023820877075,
|
|
"epoch": 1.2514619883040936,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999070225043068,
|
|
"loss": 7.0639,
|
|
"mean_token_accuracy": 0.14909755140542985,
|
|
"num_tokens": 2665568.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.542721223831177,
|
|
"epoch": 1.2573099415204678,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998930603865042,
|
|
"loss": 7.1996,
|
|
"mean_token_accuracy": 0.13743996322155,
|
|
"num_tokens": 2678748.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.32189826965332,
|
|
"epoch": 1.263157894736842,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004998781222166929,
|
|
"loss": 7.1173,
|
|
"mean_token_accuracy": 0.14156942144036294,
|
|
"num_tokens": 2690682.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.289649391174317,
|
|
"epoch": 1.2690058479532165,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004998622080597052,
|
|
"loss": 7.1589,
|
|
"mean_token_accuracy": 0.13398671224713327,
|
|
"num_tokens": 2703077.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.3387401580810545,
|
|
"epoch": 1.2748538011695907,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998453179846098,
|
|
"loss": 7.1259,
|
|
"mean_token_accuracy": 0.13829808682203293,
|
|
"num_tokens": 2714609.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.396260023117065,
|
|
"epoch": 1.280701754385965,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998274520647109,
|
|
"loss": 7.0918,
|
|
"mean_token_accuracy": 0.1457760065793991,
|
|
"num_tokens": 2726907.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.268807315826416,
|
|
"epoch": 1.286549707602339,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998086103775478,
|
|
"loss": 7.1088,
|
|
"mean_token_accuracy": 0.13768557012081145,
|
|
"num_tokens": 2739248.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.356909608840942,
|
|
"epoch": 1.2923976608187133,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997887930048947,
|
|
"loss": 7.0384,
|
|
"mean_token_accuracy": 0.1473199874162674,
|
|
"num_tokens": 2753842.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.265995216369629,
|
|
"epoch": 1.2982456140350878,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997680000327607,
|
|
"loss": 7.1389,
|
|
"mean_token_accuracy": 0.13432746306061744,
|
|
"num_tokens": 2767249.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.400834608078003,
|
|
"epoch": 1.304093567251462,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004997462315513884,
|
|
"loss": 7.117,
|
|
"mean_token_accuracy": 0.14291927963495255,
|
|
"num_tokens": 2779657.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.343295621871948,
|
|
"epoch": 1.3099415204678362,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 0.000499723487655255,
|
|
"loss": 7.0373,
|
|
"mean_token_accuracy": 0.14398355707526206,
|
|
"num_tokens": 2792760.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.285030221939087,
|
|
"epoch": 1.3157894736842106,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004996997684430703,
|
|
"loss": 7.0528,
|
|
"mean_token_accuracy": 0.14452853202819824,
|
|
"num_tokens": 2805277.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.263354873657226,
|
|
"epoch": 1.3216374269005848,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996750740177775,
|
|
"loss": 7.157,
|
|
"mean_token_accuracy": 0.14185430854558945,
|
|
"num_tokens": 2818341.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.370803260803223,
|
|
"epoch": 1.327485380116959,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004996494044865521,
|
|
"loss": 7.118,
|
|
"mean_token_accuracy": 0.1415998138487339,
|
|
"num_tokens": 2830199.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.131627798080444,
|
|
"epoch": 1.3333333333333333,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996227599608018,
|
|
"loss": 7.0109,
|
|
"mean_token_accuracy": 0.1468503713607788,
|
|
"num_tokens": 2842141.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.376796817779541,
|
|
"epoch": 1.3391812865497075,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004995951405561658,
|
|
"loss": 7.087,
|
|
"mean_token_accuracy": 0.1350610874593258,
|
|
"num_tokens": 2854929.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.256033039093017,
|
|
"epoch": 1.345029239766082,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995665463925142,
|
|
"loss": 7.0946,
|
|
"mean_token_accuracy": 0.14368716776371002,
|
|
"num_tokens": 2867034.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.244560575485229,
|
|
"epoch": 1.3508771929824561,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004995369775939478,
|
|
"loss": 6.9939,
|
|
"mean_token_accuracy": 0.1505236506462097,
|
|
"num_tokens": 2881254.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.286165046691894,
|
|
"epoch": 1.3567251461988303,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995064342887972,
|
|
"loss": 7.102,
|
|
"mean_token_accuracy": 0.14152786880731583,
|
|
"num_tokens": 2894592.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.240753030776977,
|
|
"epoch": 1.3625730994152048,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004994749166096226,
|
|
"loss": 7.0334,
|
|
"mean_token_accuracy": 0.14414507299661636,
|
|
"num_tokens": 2907196.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.277744579315185,
|
|
"epoch": 1.368421052631579,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004994424246932129,
|
|
"loss": 7.0989,
|
|
"mean_token_accuracy": 0.1371539294719696,
|
|
"num_tokens": 2920398.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.242294979095459,
|
|
"epoch": 1.3742690058479532,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004994089586805856,
|
|
"loss": 7.0167,
|
|
"mean_token_accuracy": 0.14515721797943115,
|
|
"num_tokens": 2932052.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.358625030517578,
|
|
"epoch": 1.3801169590643274,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004993745187169852,
|
|
"loss": 7.0886,
|
|
"mean_token_accuracy": 0.1365416131913662,
|
|
"num_tokens": 2943620.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.208013010025025,
|
|
"epoch": 1.3859649122807016,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000499339104951884,
|
|
"loss": 7.0413,
|
|
"mean_token_accuracy": 0.1536354199051857,
|
|
"num_tokens": 2956159.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.2751524448394775,
|
|
"epoch": 1.391812865497076,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00049930271753898,
|
|
"loss": 7.0557,
|
|
"mean_token_accuracy": 0.1427343636751175,
|
|
"num_tokens": 2967332.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.205974292755127,
|
|
"epoch": 1.3976608187134503,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004992653566361974,
|
|
"loss": 7.0593,
|
|
"mean_token_accuracy": 0.13768181428313256,
|
|
"num_tokens": 2980168.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.358188152313232,
|
|
"epoch": 1.4035087719298245,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004992270224056849,
|
|
"loss": 7.121,
|
|
"mean_token_accuracy": 0.13874844163656236,
|
|
"num_tokens": 2993509.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.197741508483887,
|
|
"epoch": 1.409356725146199,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991877150138162,
|
|
"loss": 7.0636,
|
|
"mean_token_accuracy": 0.14075298309326173,
|
|
"num_tokens": 3005827.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.2247803688049315,
|
|
"epoch": 1.4152046783625731,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004991474346311879,
|
|
"loss": 7.0159,
|
|
"mean_token_accuracy": 0.1467432513833046,
|
|
"num_tokens": 3018968.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.296478223800659,
|
|
"epoch": 1.4210526315789473,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004991061814326198,
|
|
"loss": 7.0713,
|
|
"mean_token_accuracy": 0.13609698042273521,
|
|
"num_tokens": 3031892.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.236675786972046,
|
|
"epoch": 1.4269005847953216,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004990639555971539,
|
|
"loss": 6.9919,
|
|
"mean_token_accuracy": 0.1513255462050438,
|
|
"num_tokens": 3045994.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.166168117523194,
|
|
"epoch": 1.4327485380116958,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004990207573080533,
|
|
"loss": 7.0079,
|
|
"mean_token_accuracy": 0.14798654168844222,
|
|
"num_tokens": 3058304.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.150466823577881,
|
|
"epoch": 1.4385964912280702,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004989765867528014,
|
|
"loss": 7.0027,
|
|
"mean_token_accuracy": 0.14707266688346862,
|
|
"num_tokens": 3071102.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.154991912841797,
|
|
"epoch": 1.4444444444444444,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004989314441231019,
|
|
"loss": 7.0306,
|
|
"mean_token_accuracy": 0.139088436961174,
|
|
"num_tokens": 3084653.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.244064474105835,
|
|
"epoch": 1.4502923976608186,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000498885329614877,
|
|
"loss": 7.0558,
|
|
"mean_token_accuracy": 0.1452793374657631,
|
|
"num_tokens": 3097425.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.233583593368531,
|
|
"epoch": 1.456140350877193,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004988382434282668,
|
|
"loss": 7.0555,
|
|
"mean_token_accuracy": 0.14807011559605598,
|
|
"num_tokens": 3111898.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.276054811477661,
|
|
"epoch": 1.4619883040935673,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000498790185767629,
|
|
"loss": 6.9923,
|
|
"mean_token_accuracy": 0.14536793828010558,
|
|
"num_tokens": 3124271.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.214261388778686,
|
|
"epoch": 1.4678362573099415,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000498741156841537,
|
|
"loss": 7.0941,
|
|
"mean_token_accuracy": 0.14358112215995789,
|
|
"num_tokens": 3136344.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.236997652053833,
|
|
"epoch": 1.4736842105263157,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00049869115686278,
|
|
"loss": 6.983,
|
|
"mean_token_accuracy": 0.14811330884695054,
|
|
"num_tokens": 3148299.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.2414836406707765,
|
|
"epoch": 1.47953216374269,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004986401860483615,
|
|
"loss": 7.0474,
|
|
"mean_token_accuracy": 0.14126647412776946,
|
|
"num_tokens": 3161537.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.195467853546143,
|
|
"epoch": 1.4853801169590644,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004985882446194984,
|
|
"loss": 6.9968,
|
|
"mean_token_accuracy": 0.1453041210770607,
|
|
"num_tokens": 3173676.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.2394012928009035,
|
|
"epoch": 1.4912280701754386,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0004985353328016204,
|
|
"loss": 6.9508,
|
|
"mean_token_accuracy": 0.15483074486255646,
|
|
"num_tokens": 3186542.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.1377181053161625,
|
|
"epoch": 1.4970760233918128,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004984814508243685,
|
|
"loss": 6.9925,
|
|
"mean_token_accuracy": 0.1458618648350239,
|
|
"num_tokens": 3197867.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.244440937042237,
|
|
"epoch": 1.5029239766081872,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004984265989215942,
|
|
"loss": 6.9531,
|
|
"mean_token_accuracy": 0.1449991211295128,
|
|
"num_tokens": 3209671.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.145474100112915,
|
|
"epoch": 1.5087719298245614,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004983707773313589,
|
|
"loss": 6.953,
|
|
"mean_token_accuracy": 0.14984701722860336,
|
|
"num_tokens": 3220707.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.075493192672729,
|
|
"epoch": 1.5146198830409356,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004983139862959322,
|
|
"loss": 6.9205,
|
|
"mean_token_accuracy": 0.15351321399211884,
|
|
"num_tokens": 3235122.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.246700382232666,
|
|
"epoch": 1.52046783625731,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004982562260617912,
|
|
"loss": 6.91,
|
|
"mean_token_accuracy": 0.14220256134867668,
|
|
"num_tokens": 3246833.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.16511902809143,
|
|
"epoch": 1.526315789473684,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004981974968796197,
|
|
"loss": 7.0177,
|
|
"mean_token_accuracy": 0.14064179733395576,
|
|
"num_tokens": 3260493.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.1622607707977295,
|
|
"epoch": 1.5321637426900585,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004981377990043064,
|
|
"loss": 6.9473,
|
|
"mean_token_accuracy": 0.14261056780815123,
|
|
"num_tokens": 3272445.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.171639442443848,
|
|
"epoch": 1.5380116959064327,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004980771326949442,
|
|
"loss": 6.9697,
|
|
"mean_token_accuracy": 0.14227717518806457,
|
|
"num_tokens": 3284360.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.176844310760498,
|
|
"epoch": 1.543859649122807,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004980154982148296,
|
|
"loss": 6.95,
|
|
"mean_token_accuracy": 0.1433100312948227,
|
|
"num_tokens": 3296841.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.11176643371582,
|
|
"epoch": 1.5497076023391814,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004979528958314604,
|
|
"loss": 6.9468,
|
|
"mean_token_accuracy": 0.13953165858983993,
|
|
"num_tokens": 3309393.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.238583707809449,
|
|
"epoch": 1.5555555555555556,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004978893258165354,
|
|
"loss": 7.0242,
|
|
"mean_token_accuracy": 0.1429157868027687,
|
|
"num_tokens": 3322094.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.1003436088562015,
|
|
"epoch": 1.5614035087719298,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004978247884459532,
|
|
"loss": 6.9428,
|
|
"mean_token_accuracy": 0.1477564662694931,
|
|
"num_tokens": 3334468.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.157374477386474,
|
|
"epoch": 1.5672514619883042,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977592839998105,
|
|
"loss": 6.9287,
|
|
"mean_token_accuracy": 0.14433884397149085,
|
|
"num_tokens": 3346367.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.058119916915894,
|
|
"epoch": 1.5730994152046782,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976928127624013,
|
|
"loss": 6.9132,
|
|
"mean_token_accuracy": 0.15283917635679245,
|
|
"num_tokens": 3357463.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.154469633102417,
|
|
"epoch": 1.5789473684210527,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004976253750222157,
|
|
"loss": 6.9025,
|
|
"mean_token_accuracy": 0.14609844461083413,
|
|
"num_tokens": 3369465.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.102024841308594,
|
|
"epoch": 1.5847953216374269,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004975569710719382,
|
|
"loss": 7.0104,
|
|
"mean_token_accuracy": 0.13702175095677377,
|
|
"num_tokens": 3382859.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.226300191879273,
|
|
"epoch": 1.590643274853801,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004974876012084468,
|
|
"loss": 6.966,
|
|
"mean_token_accuracy": 0.1419912114739418,
|
|
"num_tokens": 3395431.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.105293655395508,
|
|
"epoch": 1.5964912280701755,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004974172657328116,
|
|
"loss": 6.8652,
|
|
"mean_token_accuracy": 0.14728413820266723,
|
|
"num_tokens": 3408388.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.1052686214447025,
|
|
"epoch": 1.6023391812865497,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004973459649502939,
|
|
"loss": 6.8683,
|
|
"mean_token_accuracy": 0.15123746544122696,
|
|
"num_tokens": 3421275.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.077537584304809,
|
|
"epoch": 1.608187134502924,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004972736991703439,
|
|
"loss": 6.9488,
|
|
"mean_token_accuracy": 0.14620793908834456,
|
|
"num_tokens": 3433537.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.12898850440979,
|
|
"epoch": 1.6140350877192984,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004972004687066005,
|
|
"loss": 6.909,
|
|
"mean_token_accuracy": 0.1481366164982319,
|
|
"num_tokens": 3445167.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.038471460342407,
|
|
"epoch": 1.6198830409356724,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000497126273876889,
|
|
"loss": 6.9066,
|
|
"mean_token_accuracy": 0.15611805990338326,
|
|
"num_tokens": 3458529.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.139416265487671,
|
|
"epoch": 1.6257309941520468,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004970511150032205,
|
|
"loss": 6.9038,
|
|
"mean_token_accuracy": 0.1489902138710022,
|
|
"num_tokens": 3470315.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.0139429569244385,
|
|
"epoch": 1.631578947368421,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004969749924117896,
|
|
"loss": 6.8256,
|
|
"mean_token_accuracy": 0.14668601006269455,
|
|
"num_tokens": 3481899.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.0898150444030765,
|
|
"epoch": 1.6374269005847952,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004968979064329741,
|
|
"loss": 6.9273,
|
|
"mean_token_accuracy": 0.15040379613637925,
|
|
"num_tokens": 3494076.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.236377954483032,
|
|
"epoch": 1.6432748538011697,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004968198574013323,
|
|
"loss": 6.9515,
|
|
"mean_token_accuracy": 0.13935640305280686,
|
|
"num_tokens": 3507670.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.092464876174927,
|
|
"epoch": 1.6491228070175439,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000496740845655603,
|
|
"loss": 6.8558,
|
|
"mean_token_accuracy": 0.14918479546904564,
|
|
"num_tokens": 3519704.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.035297775268555,
|
|
"epoch": 1.654970760233918,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004966608715387029,
|
|
"loss": 6.9247,
|
|
"mean_token_accuracy": 0.1436144933104515,
|
|
"num_tokens": 3531665.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.969947719573975,
|
|
"epoch": 1.6608187134502925,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004965799353977248,
|
|
"loss": 6.9144,
|
|
"mean_token_accuracy": 0.15216819047927857,
|
|
"num_tokens": 3543876.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.078910541534424,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004964980375839379,
|
|
"loss": 6.8397,
|
|
"mean_token_accuracy": 0.14977291375398635,
|
|
"num_tokens": 3556517.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.047800302505493,
|
|
"epoch": 1.672514619883041,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004964151784527844,
|
|
"loss": 6.8604,
|
|
"mean_token_accuracy": 0.15354330390691756,
|
|
"num_tokens": 3569378.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.156952142715454,
|
|
"epoch": 1.6783625730994152,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000496331358363879,
|
|
"loss": 6.8541,
|
|
"mean_token_accuracy": 0.14472813308238983,
|
|
"num_tokens": 3581684.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.917730379104614,
|
|
"epoch": 1.6842105263157894,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004962465776810066,
|
|
"loss": 6.7814,
|
|
"mean_token_accuracy": 0.16086088120937347,
|
|
"num_tokens": 3594961.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.072372961044311,
|
|
"epoch": 1.6900584795321638,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004961608367721217,
|
|
"loss": 6.8544,
|
|
"mean_token_accuracy": 0.14478224515914917,
|
|
"num_tokens": 3607197.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.065153741836548,
|
|
"epoch": 1.695906432748538,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004960741360093459,
|
|
"loss": 6.8769,
|
|
"mean_token_accuracy": 0.14579440355300904,
|
|
"num_tokens": 3618728.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.111606073379517,
|
|
"epoch": 1.7017543859649122,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004959864757689669,
|
|
"loss": 6.9783,
|
|
"mean_token_accuracy": 0.13519750013947487,
|
|
"num_tokens": 3630872.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.088176918029785,
|
|
"epoch": 1.7076023391812867,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004958978564314361,
|
|
"loss": 6.8708,
|
|
"mean_token_accuracy": 0.14870110005140305,
|
|
"num_tokens": 3642657.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.056329298019409,
|
|
"epoch": 1.7134502923976607,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000495808278381368,
|
|
"loss": 6.8827,
|
|
"mean_token_accuracy": 0.14525013267993928,
|
|
"num_tokens": 3654505.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.989843940734863,
|
|
"epoch": 1.719298245614035,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004957177420075377,
|
|
"loss": 6.7841,
|
|
"mean_token_accuracy": 0.16019067466259002,
|
|
"num_tokens": 3668554.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.060316467285157,
|
|
"epoch": 1.7251461988304093,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004956262477028797,
|
|
"loss": 6.9507,
|
|
"mean_token_accuracy": 0.14162230864167213,
|
|
"num_tokens": 3681448.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.1160413265228275,
|
|
"epoch": 1.7309941520467835,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004955337958644857,
|
|
"loss": 6.8709,
|
|
"mean_token_accuracy": 0.1425687327980995,
|
|
"num_tokens": 3693677.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.9832135200500485,
|
|
"epoch": 1.736842105263158,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004954403868936032,
|
|
"loss": 6.8683,
|
|
"mean_token_accuracy": 0.15228217393159865,
|
|
"num_tokens": 3706204.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.108418083190918,
|
|
"epoch": 1.7426900584795322,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000495346021195634,
|
|
"loss": 6.8412,
|
|
"mean_token_accuracy": 0.15044791102409363,
|
|
"num_tokens": 3718255.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.084663152694702,
|
|
"epoch": 1.7485380116959064,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004952506991801319,
|
|
"loss": 6.8399,
|
|
"mean_token_accuracy": 0.1522560551762581,
|
|
"num_tokens": 3730623.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.022539806365967,
|
|
"epoch": 1.7543859649122808,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004951544212608015,
|
|
"loss": 6.8765,
|
|
"mean_token_accuracy": 0.14687300175428392,
|
|
"num_tokens": 3743636.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7543859649122808,
|
|
"eval_entropy": 6.913726612529923,
|
|
"eval_loss": 6.976031303405762,
|
|
"eval_mean_token_accuracy": 0.13840462224816238,
|
|
"eval_num_tokens": 3743636.0,
|
|
"eval_runtime": 1.7356,
|
|
"eval_samples_per_second": 1559.729,
|
|
"eval_steps_per_second": 195.326,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.135053968429565,
|
|
"epoch": 1.7602339181286548,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004950571878554956,
|
|
"loss": 6.9312,
|
|
"mean_token_accuracy": 0.13546228855848313,
|
|
"num_tokens": 3755608.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.097068405151367,
|
|
"epoch": 1.7660818713450293,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004949589993862143,
|
|
"loss": 6.8152,
|
|
"mean_token_accuracy": 0.14684600085020066,
|
|
"num_tokens": 3768208.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.97324047088623,
|
|
"epoch": 1.7719298245614035,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004948598562791025,
|
|
"loss": 6.8378,
|
|
"mean_token_accuracy": 0.14705169945955276,
|
|
"num_tokens": 3780550.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.07040581703186,
|
|
"epoch": 1.7777777777777777,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004947597589644484,
|
|
"loss": 6.8259,
|
|
"mean_token_accuracy": 0.14792063385248183,
|
|
"num_tokens": 3792417.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.051655101776123,
|
|
"epoch": 1.7836257309941521,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004946587078766818,
|
|
"loss": 6.807,
|
|
"mean_token_accuracy": 0.14990664571523665,
|
|
"num_tokens": 3803887.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.055767679214478,
|
|
"epoch": 1.7894736842105263,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004945567034543712,
|
|
"loss": 6.9462,
|
|
"mean_token_accuracy": 0.13744124248623849,
|
|
"num_tokens": 3816534.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.054939889907837,
|
|
"epoch": 1.7953216374269005,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004944537461402234,
|
|
"loss": 6.8877,
|
|
"mean_token_accuracy": 0.149474073946476,
|
|
"num_tokens": 3827790.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.062603187561035,
|
|
"epoch": 1.801169590643275,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004943498363810807,
|
|
"loss": 6.8293,
|
|
"mean_token_accuracy": 0.15547692626714707,
|
|
"num_tokens": 3839800.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.857793188095092,
|
|
"epoch": 1.807017543859649,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004942449746279184,
|
|
"loss": 6.7816,
|
|
"mean_token_accuracy": 0.1539284199476242,
|
|
"num_tokens": 3852621.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.072565126419067,
|
|
"epoch": 1.8128654970760234,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004941391613358444,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.14228285551071168,
|
|
"num_tokens": 3864184.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.071480751037598,
|
|
"epoch": 1.8187134502923976,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000494032396964096,
|
|
"loss": 6.8153,
|
|
"mean_token_accuracy": 0.1449595272541046,
|
|
"num_tokens": 3876750.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.053122901916504,
|
|
"epoch": 1.8245614035087718,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000493924681976038,
|
|
"loss": 6.8419,
|
|
"mean_token_accuracy": 0.15216735750436783,
|
|
"num_tokens": 3888788.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.933975648880005,
|
|
"epoch": 1.8304093567251463,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004938160168391614,
|
|
"loss": 6.8979,
|
|
"mean_token_accuracy": 0.14304276704788207,
|
|
"num_tokens": 3901816.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.026661539077759,
|
|
"epoch": 1.8362573099415205,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004937064020250804,
|
|
"loss": 6.7946,
|
|
"mean_token_accuracy": 0.15177498906850814,
|
|
"num_tokens": 3913127.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.026135396957398,
|
|
"epoch": 1.8421052631578947,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004935958380095316,
|
|
"loss": 6.8105,
|
|
"mean_token_accuracy": 0.1479490578174591,
|
|
"num_tokens": 3924878.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.993627643585205,
|
|
"epoch": 1.8479532163742691,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0004934843252723702,
|
|
"loss": 6.886,
|
|
"mean_token_accuracy": 0.14974023178219795,
|
|
"num_tokens": 3938334.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.041214179992676,
|
|
"epoch": 1.8538011695906431,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00049337186429757,
|
|
"loss": 6.8473,
|
|
"mean_token_accuracy": 0.14718355536460875,
|
|
"num_tokens": 3949733.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.096500444412231,
|
|
"epoch": 1.8596491228070176,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004932584555732192,
|
|
"loss": 6.872,
|
|
"mean_token_accuracy": 0.14800208285450936,
|
|
"num_tokens": 3963109.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.084401702880859,
|
|
"epoch": 1.8654970760233918,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004931440995915203,
|
|
"loss": 6.845,
|
|
"mean_token_accuracy": 0.14705388098955155,
|
|
"num_tokens": 3975477.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.9311549186706545,
|
|
"epoch": 1.871345029239766,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000493028796848786,
|
|
"loss": 6.8107,
|
|
"mean_token_accuracy": 0.15162300392985345,
|
|
"num_tokens": 3987757.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.065761947631836,
|
|
"epoch": 1.8771929824561404,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004929125478454389,
|
|
"loss": 6.7054,
|
|
"mean_token_accuracy": 0.15114182680845262,
|
|
"num_tokens": 3998807.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.891133451461792,
|
|
"epoch": 1.8830409356725146,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004927953530860077,
|
|
"loss": 6.7646,
|
|
"mean_token_accuracy": 0.15039825066924095,
|
|
"num_tokens": 4010320.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.068708038330078,
|
|
"epoch": 1.8888888888888888,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004926772130791261,
|
|
"loss": 6.8999,
|
|
"mean_token_accuracy": 0.1383692055940628,
|
|
"num_tokens": 4023432.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.984898090362549,
|
|
"epoch": 1.8947368421052633,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004925581283375303,
|
|
"loss": 6.8767,
|
|
"mean_token_accuracy": 0.14132775589823723,
|
|
"num_tokens": 4036282.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.9457885265350345,
|
|
"epoch": 1.9005847953216373,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004924380993780566,
|
|
"loss": 6.7499,
|
|
"mean_token_accuracy": 0.1543327748775482,
|
|
"num_tokens": 4048164.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.9637031078338625,
|
|
"epoch": 1.9064327485380117,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004923171267216392,
|
|
"loss": 6.7939,
|
|
"mean_token_accuracy": 0.14861909225583075,
|
|
"num_tokens": 4059287.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.972118234634399,
|
|
"epoch": 1.912280701754386,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004921952108933083,
|
|
"loss": 6.7804,
|
|
"mean_token_accuracy": 0.14669928476214408,
|
|
"num_tokens": 4071976.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.002675342559814,
|
|
"epoch": 1.9181286549707601,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004920723524221871,
|
|
"loss": 6.8152,
|
|
"mean_token_accuracy": 0.15132055059075356,
|
|
"num_tokens": 4085116.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.055388784408569,
|
|
"epoch": 1.9239766081871346,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004919485518414902,
|
|
"loss": 6.812,
|
|
"mean_token_accuracy": 0.14319210201501847,
|
|
"num_tokens": 4096778.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.902815294265747,
|
|
"epoch": 1.9298245614035088,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004918238096885213,
|
|
"loss": 6.7508,
|
|
"mean_token_accuracy": 0.1475529596209526,
|
|
"num_tokens": 4109490.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.984807109832763,
|
|
"epoch": 1.935672514619883,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00049169812650467,
|
|
"loss": 6.773,
|
|
"mean_token_accuracy": 0.15039625018835068,
|
|
"num_tokens": 4122832.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.0580432415008545,
|
|
"epoch": 1.9415204678362574,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004915715028354102,
|
|
"loss": 6.7956,
|
|
"mean_token_accuracy": 0.1466011457145214,
|
|
"num_tokens": 4134585.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.965006637573242,
|
|
"epoch": 1.9473684210526314,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004914439392302981,
|
|
"loss": 6.7469,
|
|
"mean_token_accuracy": 0.1464242994785309,
|
|
"num_tokens": 4145909.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.98766508102417,
|
|
"epoch": 1.9532163742690059,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004913154362429686,
|
|
"loss": 6.6539,
|
|
"mean_token_accuracy": 0.16458579450845717,
|
|
"num_tokens": 4157360.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.8865578174591064,
|
|
"epoch": 1.95906432748538,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004911859944311341,
|
|
"loss": 6.8376,
|
|
"mean_token_accuracy": 0.14514545798301698,
|
|
"num_tokens": 4169789.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.0449613571167,
|
|
"epoch": 1.9649122807017543,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004910556143565812,
|
|
"loss": 6.8308,
|
|
"mean_token_accuracy": 0.1438768610358238,
|
|
"num_tokens": 4183131.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.966651821136475,
|
|
"epoch": 1.9707602339181287,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004909242965851688,
|
|
"loss": 6.8105,
|
|
"mean_token_accuracy": 0.14589049071073532,
|
|
"num_tokens": 4194865.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.988571405410767,
|
|
"epoch": 1.976608187134503,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004907920416868253,
|
|
"loss": 6.8019,
|
|
"mean_token_accuracy": 0.14565620571374893,
|
|
"num_tokens": 4207406.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.944311332702637,
|
|
"epoch": 1.9824561403508771,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004906588502355467,
|
|
"loss": 6.7757,
|
|
"mean_token_accuracy": 0.14800650775432586,
|
|
"num_tokens": 4221623.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.886150646209717,
|
|
"epoch": 1.9883040935672516,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004905247228093933,
|
|
"loss": 6.7147,
|
|
"mean_token_accuracy": 0.16161981076002122,
|
|
"num_tokens": 4233699.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.940931224822998,
|
|
"epoch": 1.9941520467836256,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004903896599904878,
|
|
"loss": 6.7046,
|
|
"mean_token_accuracy": 0.15220564156770705,
|
|
"num_tokens": 4247669.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.927793836593628,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004902536623650125,
|
|
"loss": 6.7571,
|
|
"mean_token_accuracy": 0.14728414714336396,
|
|
"num_tokens": 4258968.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.963106155395508,
|
|
"epoch": 2.0058479532163744,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004901167305232068,
|
|
"loss": 6.5334,
|
|
"mean_token_accuracy": 0.1553110048174858,
|
|
"num_tokens": 4270684.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.848935222625732,
|
|
"epoch": 2.0116959064327484,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004899788650593648,
|
|
"loss": 6.4578,
|
|
"mean_token_accuracy": 0.1595386952161789,
|
|
"num_tokens": 4282567.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.960906648635865,
|
|
"epoch": 2.017543859649123,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004898400665718325,
|
|
"loss": 6.4928,
|
|
"mean_token_accuracy": 0.15572800636291503,
|
|
"num_tokens": 4294803.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.90195779800415,
|
|
"epoch": 2.023391812865497,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004897003356630052,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.15569688379764557,
|
|
"num_tokens": 4305898.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.879411506652832,
|
|
"epoch": 2.0292397660818713,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004895596729393252,
|
|
"loss": 6.3867,
|
|
"mean_token_accuracy": 0.1615717150270939,
|
|
"num_tokens": 4317985.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.773409175872803,
|
|
"epoch": 2.0350877192982457,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004894180790112787,
|
|
"loss": 6.4646,
|
|
"mean_token_accuracy": 0.15991110801696778,
|
|
"num_tokens": 4330482.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.953168106079102,
|
|
"epoch": 2.0409356725146197,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004892755544933934,
|
|
"loss": 6.5359,
|
|
"mean_token_accuracy": 0.14696549624204636,
|
|
"num_tokens": 4342782.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.875130081176758,
|
|
"epoch": 2.046783625730994,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004891321000042361,
|
|
"loss": 6.5404,
|
|
"mean_token_accuracy": 0.14955661445856094,
|
|
"num_tokens": 4356204.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.870871734619141,
|
|
"epoch": 2.0526315789473686,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004889877161664096,
|
|
"loss": 6.4454,
|
|
"mean_token_accuracy": 0.1566561982035637,
|
|
"num_tokens": 4370003.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.880638551712036,
|
|
"epoch": 2.0584795321637426,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048884240360655,
|
|
"loss": 6.4973,
|
|
"mean_token_accuracy": 0.15367283672094345,
|
|
"num_tokens": 4381972.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.828345870971679,
|
|
"epoch": 2.064327485380117,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004886961629553242,
|
|
"loss": 6.414,
|
|
"mean_token_accuracy": 0.16296480149030684,
|
|
"num_tokens": 4394033.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.877967786788941,
|
|
"epoch": 2.0701754385964914,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004885489948474272,
|
|
"loss": 6.5394,
|
|
"mean_token_accuracy": 0.15387267172336577,
|
|
"num_tokens": 4406799.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.8831394672393795,
|
|
"epoch": 2.0760233918128654,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004884008999215792,
|
|
"loss": 6.5346,
|
|
"mean_token_accuracy": 0.1527932584285736,
|
|
"num_tokens": 4420182.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.834596395492554,
|
|
"epoch": 2.08187134502924,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004882518788205225,
|
|
"loss": 6.5003,
|
|
"mean_token_accuracy": 0.15515169501304626,
|
|
"num_tokens": 4432014.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.9492169380187985,
|
|
"epoch": 2.087719298245614,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004881019321910198,
|
|
"loss": 6.4922,
|
|
"mean_token_accuracy": 0.15978310108184815,
|
|
"num_tokens": 4444755.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.811837768554687,
|
|
"epoch": 2.0935672514619883,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048795106068384985,
|
|
"loss": 6.5632,
|
|
"mean_token_accuracy": 0.15767932385206224,
|
|
"num_tokens": 4456390.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.876183891296387,
|
|
"epoch": 2.0994152046783627,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000487799264953806,
|
|
"loss": 6.4799,
|
|
"mean_token_accuracy": 0.1628776267170906,
|
|
"num_tokens": 4468567.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.800377607345581,
|
|
"epoch": 2.1052631578947367,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004876465456596926,
|
|
"loss": 6.4521,
|
|
"mean_token_accuracy": 0.1607213571667671,
|
|
"num_tokens": 4479553.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.800175619125366,
|
|
"epoch": 2.111111111111111,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004874929034643223,
|
|
"loss": 6.4485,
|
|
"mean_token_accuracy": 0.16596934646368028,
|
|
"num_tokens": 4491505.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.807722282409668,
|
|
"epoch": 2.116959064327485,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004873383390345132,
|
|
"loss": 6.521,
|
|
"mean_token_accuracy": 0.15679756850004195,
|
|
"num_tokens": 4504896.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.794916152954102,
|
|
"epoch": 2.1228070175438596,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00048718285304108613,
|
|
"loss": 6.513,
|
|
"mean_token_accuracy": 0.15284237414598464,
|
|
"num_tokens": 4516319.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.853626441955567,
|
|
"epoch": 2.128654970760234,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048702644615886137,
|
|
"loss": 6.5226,
|
|
"mean_token_accuracy": 0.1524695210158825,
|
|
"num_tokens": 4529132.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.864784622192383,
|
|
"epoch": 2.134502923976608,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048686911906665597,
|
|
"loss": 6.4923,
|
|
"mean_token_accuracy": 0.15571837723255158,
|
|
"num_tokens": 4541080.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.845421934127808,
|
|
"epoch": 2.1403508771929824,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00048671087244728093,
|
|
"loss": 6.4962,
|
|
"mean_token_accuracy": 0.15684320777654648,
|
|
"num_tokens": 4553442.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.731913423538208,
|
|
"epoch": 2.146198830409357,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00048655170698753774,
|
|
"loss": 6.4244,
|
|
"mean_token_accuracy": 0.16705678403377533,
|
|
"num_tokens": 4565617.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.900339508056641,
|
|
"epoch": 2.152046783625731,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000486391623378216,
|
|
"loss": 6.5217,
|
|
"mean_token_accuracy": 0.15130398869514466,
|
|
"num_tokens": 4577418.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.829915714263916,
|
|
"epoch": 2.1578947368421053,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004862306223140901,
|
|
"loss": 6.4447,
|
|
"mean_token_accuracy": 0.15913260877132415,
|
|
"num_tokens": 4589763.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.829157447814941,
|
|
"epoch": 2.1637426900584797,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048606870449391604,
|
|
"loss": 6.5366,
|
|
"mean_token_accuracy": 0.15990214347839354,
|
|
"num_tokens": 4602522.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.833024978637695,
|
|
"epoch": 2.1695906432748537,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048590587062042907,
|
|
"loss": 6.4977,
|
|
"mean_token_accuracy": 0.15934453904628754,
|
|
"num_tokens": 4614680.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.8116669178009035,
|
|
"epoch": 2.175438596491228,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048574212140033965,
|
|
"loss": 6.5341,
|
|
"mean_token_accuracy": 0.15397311598062516,
|
|
"num_tokens": 4628166.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.821261405944824,
|
|
"epoch": 2.181286549707602,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004855774575443311,
|
|
"loss": 6.5328,
|
|
"mean_token_accuracy": 0.15765905529260635,
|
|
"num_tokens": 4640200.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.8316350936889645,
|
|
"epoch": 2.1871345029239766,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00048541187976705645,
|
|
"loss": 6.4767,
|
|
"mean_token_accuracy": 0.15292651504278182,
|
|
"num_tokens": 4652968.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.770348358154297,
|
|
"epoch": 2.192982456140351,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048524538878713517,
|
|
"loss": 6.485,
|
|
"mean_token_accuracy": 0.15311889797449113,
|
|
"num_tokens": 4665741.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.82572374343872,
|
|
"epoch": 2.198830409356725,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00048507798532715005,
|
|
"loss": 6.5084,
|
|
"mean_token_accuracy": 0.155975741147995,
|
|
"num_tokens": 4680074.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.801500225067139,
|
|
"epoch": 2.2046783625730995,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048490967011364394,
|
|
"loss": 6.5232,
|
|
"mean_token_accuracy": 0.1580188348889351,
|
|
"num_tokens": 4692340.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.821602535247803,
|
|
"epoch": 2.2105263157894735,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004847404438771172,
|
|
"loss": 6.518,
|
|
"mean_token_accuracy": 0.15585452020168306,
|
|
"num_tokens": 4704470.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.843487024307251,
|
|
"epoch": 2.216374269005848,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004845703073520236,
|
|
"loss": 6.5465,
|
|
"mean_token_accuracy": 0.15025353580713272,
|
|
"num_tokens": 4716253.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.897539663314819,
|
|
"epoch": 2.2222222222222223,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048439926127676796,
|
|
"loss": 6.6592,
|
|
"mean_token_accuracy": 0.14622338116168976,
|
|
"num_tokens": 4730083.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.8224715232849125,
|
|
"epoch": 2.2280701754385963,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00048422730639370256,
|
|
"loss": 6.4221,
|
|
"mean_token_accuracy": 0.16505661010742187,
|
|
"num_tokens": 4742770.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.769926834106445,
|
|
"epoch": 2.2339181286549707,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004840544434491238,
|
|
"loss": 6.5511,
|
|
"mean_token_accuracy": 0.15389376133680344,
|
|
"num_tokens": 4754299.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.8316566944122314,
|
|
"epoch": 2.239766081871345,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048388067319326937,
|
|
"loss": 6.5164,
|
|
"mean_token_accuracy": 0.15317742228507997,
|
|
"num_tokens": 4766964.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.77725830078125,
|
|
"epoch": 2.245614035087719,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048370599638031464,
|
|
"loss": 6.5459,
|
|
"mean_token_accuracy": 0.153824120759964,
|
|
"num_tokens": 4779921.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.8391601085662845,
|
|
"epoch": 2.2514619883040936,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048353041376836937,
|
|
"loss": 6.529,
|
|
"mean_token_accuracy": 0.15380112081766129,
|
|
"num_tokens": 4792233.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.785034418106079,
|
|
"epoch": 2.257309941520468,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004833539261194749,
|
|
"loss": 6.5129,
|
|
"mean_token_accuracy": 0.16028628200292588,
|
|
"num_tokens": 4803348.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.850829458236694,
|
|
"epoch": 2.263157894736842,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00048317653419960007,
|
|
"loss": 6.5866,
|
|
"mean_token_accuracy": 0.1491300158202648,
|
|
"num_tokens": 4816574.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.703137397766113,
|
|
"epoch": 2.2690058479532165,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048299823877863857,
|
|
"loss": 6.5065,
|
|
"mean_token_accuracy": 0.15909324884414672,
|
|
"num_tokens": 4828008.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.869985198974609,
|
|
"epoch": 2.2748538011695905,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004828190406304054,
|
|
"loss": 6.5792,
|
|
"mean_token_accuracy": 0.151179438829422,
|
|
"num_tokens": 4840621.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.806095361709595,
|
|
"epoch": 2.280701754385965,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004826389405326334,
|
|
"loss": 6.4768,
|
|
"mean_token_accuracy": 0.15327924191951753,
|
|
"num_tokens": 4853503.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.816493034362793,
|
|
"epoch": 2.2865497076023393,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.0004824579392669698,
|
|
"loss": 6.4961,
|
|
"mean_token_accuracy": 0.15774255096912385,
|
|
"num_tokens": 4865925.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.774963331222534,
|
|
"epoch": 2.2923976608187133,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048227603761897316,
|
|
"loss": 6.5814,
|
|
"mean_token_accuracy": 0.15094525516033172,
|
|
"num_tokens": 4878704.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.879746818542481,
|
|
"epoch": 2.2982456140350878,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004820932363781096,
|
|
"loss": 6.5312,
|
|
"mean_token_accuracy": 0.15396272242069245,
|
|
"num_tokens": 4891750.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.784665966033936,
|
|
"epoch": 2.3040935672514617,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004819095363377496,
|
|
"loss": 6.5187,
|
|
"mean_token_accuracy": 0.15821436196565627,
|
|
"num_tokens": 4904539.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.7290582180023195,
|
|
"epoch": 2.309941520467836,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048172493829516465,
|
|
"loss": 6.3934,
|
|
"mean_token_accuracy": 0.16299119889736174,
|
|
"num_tokens": 4916210.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.78262882232666,
|
|
"epoch": 2.3157894736842106,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004815394430515234,
|
|
"loss": 6.462,
|
|
"mean_token_accuracy": 0.15879430174827575,
|
|
"num_tokens": 4929011.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.765537166595459,
|
|
"epoch": 2.3216374269005846,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004813530514118884,
|
|
"loss": 6.4498,
|
|
"mean_token_accuracy": 0.15751578211784362,
|
|
"num_tokens": 4942484.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.7109559059143065,
|
|
"epoch": 2.327485380116959,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004811657641852129,
|
|
"loss": 6.427,
|
|
"mean_token_accuracy": 0.15696060806512832,
|
|
"num_tokens": 4955395.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.781054306030273,
|
|
"epoch": 2.3333333333333335,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048097758218433687,
|
|
"loss": 6.4703,
|
|
"mean_token_accuracy": 0.15664945989847184,
|
|
"num_tokens": 4966853.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.6470122814178465,
|
|
"epoch": 2.3391812865497075,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048078850622598355,
|
|
"loss": 6.3438,
|
|
"mean_token_accuracy": 0.16954133957624434,
|
|
"num_tokens": 4978514.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.3391812865497075,
|
|
"eval_entropy": 6.59683396190317,
|
|
"eval_loss": 6.824612617492676,
|
|
"eval_mean_token_accuracy": 0.14168384835222916,
|
|
"eval_num_tokens": 4978514.0,
|
|
"eval_runtime": 1.7431,
|
|
"eval_samples_per_second": 1552.985,
|
|
"eval_steps_per_second": 194.482,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.78842453956604,
|
|
"epoch": 2.345029239766082,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00048059853713075637,
|
|
"loss": 6.501,
|
|
"mean_token_accuracy": 0.15215353667736053,
|
|
"num_tokens": 4991331.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.730353736877442,
|
|
"epoch": 2.3508771929824563,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004804076757231347,
|
|
"loss": 6.4718,
|
|
"mean_token_accuracy": 0.15783711373806,
|
|
"num_tokens": 5004652.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.759194612503052,
|
|
"epoch": 2.3567251461988303,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048021592283147073,
|
|
"loss": 6.4715,
|
|
"mean_token_accuracy": 0.16659724712371826,
|
|
"num_tokens": 5017204.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.698295402526855,
|
|
"epoch": 2.3625730994152048,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004800232792879858,
|
|
"loss": 6.4975,
|
|
"mean_token_accuracy": 0.16038344353437423,
|
|
"num_tokens": 5029667.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.651339197158814,
|
|
"epoch": 2.3684210526315788,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047982974592876665,
|
|
"loss": 6.3622,
|
|
"mean_token_accuracy": 0.16610048711299896,
|
|
"num_tokens": 5041490.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.777406501770019,
|
|
"epoch": 2.374269005847953,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000479635323593762,
|
|
"loss": 6.4944,
|
|
"mean_token_accuracy": 0.15169765874743463,
|
|
"num_tokens": 5054353.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.768514680862427,
|
|
"epoch": 2.3801169590643276,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047944001312677867,
|
|
"loss": 6.4569,
|
|
"mean_token_accuracy": 0.15618255287408828,
|
|
"num_tokens": 5067322.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.779137372970581,
|
|
"epoch": 2.3859649122807016,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004792438153754782,
|
|
"loss": 6.5077,
|
|
"mean_token_accuracy": 0.16168376058340073,
|
|
"num_tokens": 5079957.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.691183376312256,
|
|
"epoch": 2.391812865497076,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004790467311913727,
|
|
"loss": 6.4856,
|
|
"mean_token_accuracy": 0.15492921620607375,
|
|
"num_tokens": 5091975.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.777608013153076,
|
|
"epoch": 2.39766081871345,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004788487614298219,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.1535652071237564,
|
|
"num_tokens": 5103847.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.772614765167236,
|
|
"epoch": 2.4035087719298245,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004786499069500286,
|
|
"loss": 6.442,
|
|
"mean_token_accuracy": 0.16196208149194719,
|
|
"num_tokens": 5115153.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.6315269947052,
|
|
"epoch": 2.409356725146199,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047845016861503556,
|
|
"loss": 6.3778,
|
|
"mean_token_accuracy": 0.16543841511011123,
|
|
"num_tokens": 5128008.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.6571738719940186,
|
|
"epoch": 2.415204678362573,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047824954729172136,
|
|
"loss": 6.3427,
|
|
"mean_token_accuracy": 0.1712554380297661,
|
|
"num_tokens": 5140023.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.68211727142334,
|
|
"epoch": 2.4210526315789473,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004780480438507972,
|
|
"loss": 6.3945,
|
|
"mean_token_accuracy": 0.1659012407064438,
|
|
"num_tokens": 5152660.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.691555452346802,
|
|
"epoch": 2.426900584795322,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047784565916680216,
|
|
"loss": 6.4164,
|
|
"mean_token_accuracy": 0.16120534390211105,
|
|
"num_tokens": 5164481.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.8111319065094,
|
|
"epoch": 2.4327485380116958,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004776423941181005,
|
|
"loss": 6.4971,
|
|
"mean_token_accuracy": 0.1621654786169529,
|
|
"num_tokens": 5177178.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.711190795898437,
|
|
"epoch": 2.43859649122807,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00047743824958687694,
|
|
"loss": 6.3991,
|
|
"mean_token_accuracy": 0.1595264583826065,
|
|
"num_tokens": 5189385.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.763757753372192,
|
|
"epoch": 2.4444444444444446,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047723322645913354,
|
|
"loss": 6.4655,
|
|
"mean_token_accuracy": 0.16068250834941863,
|
|
"num_tokens": 5201839.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.680694103240967,
|
|
"epoch": 2.4502923976608186,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004770273256246853,
|
|
"loss": 6.4967,
|
|
"mean_token_accuracy": 0.1545516699552536,
|
|
"num_tokens": 5214565.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.766183567047119,
|
|
"epoch": 2.456140350877193,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004768205479771566,
|
|
"loss": 6.4671,
|
|
"mean_token_accuracy": 0.16065627932548524,
|
|
"num_tokens": 5227029.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.767592906951904,
|
|
"epoch": 2.461988304093567,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00047661289441397724,
|
|
"loss": 6.4794,
|
|
"mean_token_accuracy": 0.16503392904996872,
|
|
"num_tokens": 5239208.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.667439317703247,
|
|
"epoch": 2.4678362573099415,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047640436583637865,
|
|
"loss": 6.4425,
|
|
"mean_token_accuracy": 0.1622099220752716,
|
|
"num_tokens": 5252272.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.799730634689331,
|
|
"epoch": 2.473684210526316,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00047619496314938977,
|
|
"loss": 6.5702,
|
|
"mean_token_accuracy": 0.150389401614666,
|
|
"num_tokens": 5264387.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.685615730285645,
|
|
"epoch": 2.47953216374269,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004759846872618332,
|
|
"loss": 6.5309,
|
|
"mean_token_accuracy": 0.15629530549049378,
|
|
"num_tokens": 5276662.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.752291631698609,
|
|
"epoch": 2.4853801169590644,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047577353908632156,
|
|
"loss": 6.4597,
|
|
"mean_token_accuracy": 0.16320658028125762,
|
|
"num_tokens": 5290734.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.59678897857666,
|
|
"epoch": 2.4912280701754383,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00047556151953925287,
|
|
"loss": 6.3483,
|
|
"mean_token_accuracy": 0.16701145321130753,
|
|
"num_tokens": 5302510.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.849278783798217,
|
|
"epoch": 2.497076023391813,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004753486295408073,
|
|
"loss": 6.4293,
|
|
"mean_token_accuracy": 0.15996883660554886,
|
|
"num_tokens": 5313243.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.5875578880310055,
|
|
"epoch": 2.502923976608187,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00047513487001494257,
|
|
"loss": 6.5105,
|
|
"mean_token_accuracy": 0.15729328095912934,
|
|
"num_tokens": 5325485.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.842334747314453,
|
|
"epoch": 2.5087719298245617,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00047492024188939056,
|
|
"loss": 6.5625,
|
|
"mean_token_accuracy": 0.15200666040182115,
|
|
"num_tokens": 5339335.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.765503215789795,
|
|
"epoch": 2.5146198830409356,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004747047460956525,
|
|
"loss": 6.5084,
|
|
"mean_token_accuracy": 0.15184070765972138,
|
|
"num_tokens": 5351195.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.6956579208374025,
|
|
"epoch": 2.52046783625731,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004744883835689958,
|
|
"loss": 6.4564,
|
|
"mean_token_accuracy": 0.15987571626901625,
|
|
"num_tokens": 5362884.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.669484853744507,
|
|
"epoch": 2.526315789473684,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004742711552484492,
|
|
"loss": 6.4902,
|
|
"mean_token_accuracy": 0.15629828870296478,
|
|
"num_tokens": 5375770.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.7700704574584964,
|
|
"epoch": 2.5321637426900585,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00047405306207679926,
|
|
"loss": 6.4671,
|
|
"mean_token_accuracy": 0.15315858870744706,
|
|
"num_tokens": 5388673.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.570160865783691,
|
|
"epoch": 2.538011695906433,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004738341050005859,
|
|
"loss": 6.3133,
|
|
"mean_token_accuracy": 0.1679084688425064,
|
|
"num_tokens": 5400454.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.723893356323242,
|
|
"epoch": 2.543859649122807,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.0004736142849700986,
|
|
"loss": 6.4604,
|
|
"mean_token_accuracy": 0.16196997910737992,
|
|
"num_tokens": 5413740.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.702420473098755,
|
|
"epoch": 2.5497076023391814,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004733936029393721,
|
|
"loss": 6.474,
|
|
"mean_token_accuracy": 0.1617956429719925,
|
|
"num_tokens": 5426089.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.7728574752807615,
|
|
"epoch": 2.5555555555555554,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004731720598661821,
|
|
"loss": 6.5213,
|
|
"mean_token_accuracy": 0.1508661463856697,
|
|
"num_tokens": 5438835.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.628914785385132,
|
|
"epoch": 2.56140350877193,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004729496567120415,
|
|
"loss": 6.458,
|
|
"mean_token_accuracy": 0.16138928830623628,
|
|
"num_tokens": 5452054.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.721137571334839,
|
|
"epoch": 2.5672514619883042,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.000472726394442196,
|
|
"loss": 6.4259,
|
|
"mean_token_accuracy": 0.15541900396347047,
|
|
"num_tokens": 5466455.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.7194966793060305,
|
|
"epoch": 2.573099415204678,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047250227402561976,
|
|
"loss": 6.4399,
|
|
"mean_token_accuracy": 0.1621753215789795,
|
|
"num_tokens": 5479990.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.7334068775177,
|
|
"epoch": 2.5789473684210527,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004722772964350115,
|
|
"loss": 6.4819,
|
|
"mean_token_accuracy": 0.16410714983940125,
|
|
"num_tokens": 5493035.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.627843904495239,
|
|
"epoch": 2.5847953216374266,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047205146264679015,
|
|
"loss": 6.4972,
|
|
"mean_token_accuracy": 0.15570203065872193,
|
|
"num_tokens": 5506093.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.719814825057983,
|
|
"epoch": 2.590643274853801,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00047182477364109045,
|
|
"loss": 6.4734,
|
|
"mean_token_accuracy": 0.1660175785422325,
|
|
"num_tokens": 5518552.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.664255619049072,
|
|
"epoch": 2.5964912280701755,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047159723040175896,
|
|
"loss": 6.4947,
|
|
"mean_token_accuracy": 0.16128159165382386,
|
|
"num_tokens": 5530123.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.8260692119598385,
|
|
"epoch": 2.60233918128655,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00047136883391634966,
|
|
"loss": 6.5105,
|
|
"mean_token_accuracy": 0.1525448054075241,
|
|
"num_tokens": 5542769.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.516240072250366,
|
|
"epoch": 2.608187134502924,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004711395851761198,
|
|
"loss": 6.3133,
|
|
"mean_token_accuracy": 0.17718010097742082,
|
|
"num_tokens": 5555975.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.68882999420166,
|
|
"epoch": 2.6140350877192984,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004709094851760251,
|
|
"loss": 6.422,
|
|
"mean_token_accuracy": 0.15997820794582368,
|
|
"num_tokens": 5569059.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.699761438369751,
|
|
"epoch": 2.6198830409356724,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.0004706785349147163,
|
|
"loss": 6.488,
|
|
"mean_token_accuracy": 0.15371347069740296,
|
|
"num_tokens": 5581358.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.692812108993531,
|
|
"epoch": 2.625730994152047,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000470446735394534,
|
|
"loss": 6.4376,
|
|
"mean_token_accuracy": 0.15799081325531006,
|
|
"num_tokens": 5594256.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.709306335449218,
|
|
"epoch": 2.6315789473684212,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004702140876215047,
|
|
"loss": 6.4429,
|
|
"mean_token_accuracy": 0.15952685177326204,
|
|
"num_tokens": 5607236.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.751166915893554,
|
|
"epoch": 2.6374269005847952,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004699805926053365,
|
|
"loss": 6.5489,
|
|
"mean_token_accuracy": 0.14810458123683928,
|
|
"num_tokens": 5619550.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.6508769512176515,
|
|
"epoch": 2.6432748538011697,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00046974625135941445,
|
|
"loss": 6.4041,
|
|
"mean_token_accuracy": 0.16714755594730377,
|
|
"num_tokens": 5632721.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.604614305496216,
|
|
"epoch": 2.6491228070175437,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046951106490079636,
|
|
"loss": 6.2551,
|
|
"mean_token_accuracy": 0.16622601598501205,
|
|
"num_tokens": 5643672.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.682486724853516,
|
|
"epoch": 2.654970760233918,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00046927503425020847,
|
|
"loss": 6.4072,
|
|
"mean_token_accuracy": 0.15769419074058533,
|
|
"num_tokens": 5656072.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.7076764583587645,
|
|
"epoch": 2.6608187134502925,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00046903816043204045,
|
|
"loss": 6.4216,
|
|
"mean_token_accuracy": 0.16312220990657805,
|
|
"num_tokens": 5668365.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.696519136428833,
|
|
"epoch": 2.6666666666666665,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004688004444743419,
|
|
"loss": 6.3903,
|
|
"mean_token_accuracy": 0.16126345247030258,
|
|
"num_tokens": 5680013.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.579961919784546,
|
|
"epoch": 2.672514619883041,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000468561887408817,
|
|
"loss": 6.4123,
|
|
"mean_token_accuracy": 0.16684895008802414,
|
|
"num_tokens": 5691884.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.68872218132019,
|
|
"epoch": 2.678362573099415,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004683224902708206,
|
|
"loss": 6.4742,
|
|
"mean_token_accuracy": 0.15662514567375183,
|
|
"num_tokens": 5704340.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.756664419174195,
|
|
"epoch": 2.6842105263157894,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004680822540993534,
|
|
"loss": 6.5251,
|
|
"mean_token_accuracy": 0.15745238810777665,
|
|
"num_tokens": 5717792.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.6499731063842775,
|
|
"epoch": 2.690058479532164,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046784117993705753,
|
|
"loss": 6.3996,
|
|
"mean_token_accuracy": 0.15965020656585693,
|
|
"num_tokens": 5730237.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.635748100280762,
|
|
"epoch": 2.6959064327485383,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046759926883021234,
|
|
"loss": 6.401,
|
|
"mean_token_accuracy": 0.16328736394643784,
|
|
"num_tokens": 5743953.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.630077362060547,
|
|
"epoch": 2.7017543859649122,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00046735652182872925,
|
|
"loss": 6.3658,
|
|
"mean_token_accuracy": 0.16114580780267715,
|
|
"num_tokens": 5756148.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.674561071395874,
|
|
"epoch": 2.7076023391812867,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004671129399861476,
|
|
"loss": 6.389,
|
|
"mean_token_accuracy": 0.1633088245987892,
|
|
"num_tokens": 5769163.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.665018463134766,
|
|
"epoch": 2.7134502923976607,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046686852435963016,
|
|
"loss": 6.4305,
|
|
"mean_token_accuracy": 0.15733572244644164,
|
|
"num_tokens": 5781832.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.676378059387207,
|
|
"epoch": 2.719298245614035,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004666232760099583,
|
|
"loss": 6.4262,
|
|
"mean_token_accuracy": 0.15573703944683076,
|
|
"num_tokens": 5794803.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.6417618751525875,
|
|
"epoch": 2.7251461988304095,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046637719600152737,
|
|
"loss": 6.4575,
|
|
"mean_token_accuracy": 0.15561199858784674,
|
|
"num_tokens": 5806768.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.7871111869812015,
|
|
"epoch": 2.7309941520467835,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046613028540234226,
|
|
"loss": 6.4607,
|
|
"mean_token_accuracy": 0.1528087079524994,
|
|
"num_tokens": 5819113.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.690218925476074,
|
|
"epoch": 2.736842105263158,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046588254528401264,
|
|
"loss": 6.437,
|
|
"mean_token_accuracy": 0.1578143686056137,
|
|
"num_tokens": 5830067.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.635287761688232,
|
|
"epoch": 2.742690058479532,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004656339767217485,
|
|
"loss": 6.4848,
|
|
"mean_token_accuracy": 0.16050305515527724,
|
|
"num_tokens": 5842882.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.609443235397339,
|
|
"epoch": 2.7485380116959064,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00046538458079435516,
|
|
"loss": 6.4252,
|
|
"mean_token_accuracy": 0.15916400253772736,
|
|
"num_tokens": 5854041.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.7075742244720455,
|
|
"epoch": 2.754385964912281,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004651343585842287,
|
|
"loss": 6.4508,
|
|
"mean_token_accuracy": 0.16283646672964097,
|
|
"num_tokens": 5866027.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.671390342712402,
|
|
"epoch": 2.760233918128655,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046488331117735176,
|
|
"loss": 6.4845,
|
|
"mean_token_accuracy": 0.16220867186784743,
|
|
"num_tokens": 5877064.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.720740461349488,
|
|
"epoch": 2.7660818713450293,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004646314396632878,
|
|
"loss": 6.4449,
|
|
"mean_token_accuracy": 0.1543478935956955,
|
|
"num_tokens": 5889190.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.6629119396209715,
|
|
"epoch": 2.7719298245614032,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004643787451351774,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.16133566051721573,
|
|
"num_tokens": 5901615.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.620958757400513,
|
|
"epoch": 2.7777777777777777,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046412522868973297,
|
|
"loss": 6.4197,
|
|
"mean_token_accuracy": 0.16412558257579804,
|
|
"num_tokens": 5913797.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.730884408950805,
|
|
"epoch": 2.783625730994152,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046387089142723386,
|
|
"loss": 6.5005,
|
|
"mean_token_accuracy": 0.15636750161647797,
|
|
"num_tokens": 5927607.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.529775142669678,
|
|
"epoch": 2.7894736842105265,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00046361573445152224,
|
|
"loss": 6.3267,
|
|
"mean_token_accuracy": 0.1680535688996315,
|
|
"num_tokens": 5940254.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.525655937194824,
|
|
"epoch": 2.7953216374269005,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00046335975886999754,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.17823653072118759,
|
|
"num_tokens": 5953341.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.526809024810791,
|
|
"epoch": 2.801169590643275,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046310296579361225,
|
|
"loss": 6.2108,
|
|
"mean_token_accuracy": 0.1784290373325348,
|
|
"num_tokens": 5965148.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.66150860786438,
|
|
"epoch": 2.807017543859649,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004628453563368666,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.15572031289339067,
|
|
"num_tokens": 5979143.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.686175584793091,
|
|
"epoch": 2.8128654970760234,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004625869316178043,
|
|
"loss": 6.4467,
|
|
"mean_token_accuracy": 0.15691925287246705,
|
|
"num_tokens": 5992735.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.638707208633423,
|
|
"epoch": 2.818713450292398,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000462327692758007,
|
|
"loss": 6.5145,
|
|
"mean_token_accuracy": 0.15147309452295304,
|
|
"num_tokens": 6005033.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.709354877471924,
|
|
"epoch": 2.824561403508772,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046206764088259003,
|
|
"loss": 6.3824,
|
|
"mean_token_accuracy": 0.1630718469619751,
|
|
"num_tokens": 6016746.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.573739004135132,
|
|
"epoch": 2.8304093567251463,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00046180677712019707,
|
|
"loss": 6.4565,
|
|
"mean_token_accuracy": 0.1616368681192398,
|
|
"num_tokens": 6029774.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.69282865524292,
|
|
"epoch": 2.8362573099415203,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004615451026029957,
|
|
"loss": 6.4896,
|
|
"mean_token_accuracy": 0.15256085395812988,
|
|
"num_tokens": 6043773.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.708876514434815,
|
|
"epoch": 2.8421052631578947,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.000461282618466672,
|
|
"loss": 6.4152,
|
|
"mean_token_accuracy": 0.15378952920436859,
|
|
"num_tokens": 6055840.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.5623846530914305,
|
|
"epoch": 2.847953216374269,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046101932585042597,
|
|
"loss": 6.3331,
|
|
"mean_token_accuracy": 0.16594051718711852,
|
|
"num_tokens": 6069170.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.63337721824646,
|
|
"epoch": 2.853801169590643,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00046075522589696636,
|
|
"loss": 6.3916,
|
|
"mean_token_accuracy": 0.16303064227104186,
|
|
"num_tokens": 6080888.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.606135702133178,
|
|
"epoch": 2.8596491228070176,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000460490319752506,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.16366248875856398,
|
|
"num_tokens": 6093035.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.61658935546875,
|
|
"epoch": 2.8654970760233915,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.00046022460856675663,
|
|
"loss": 6.4116,
|
|
"mean_token_accuracy": 0.15998516529798507,
|
|
"num_tokens": 6105963.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.775122499465942,
|
|
"epoch": 2.871345029239766,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004599580934929237,
|
|
"loss": 6.4301,
|
|
"mean_token_accuracy": 0.15838743150234222,
|
|
"num_tokens": 6118541.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.554420280456543,
|
|
"epoch": 2.8771929824561404,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.0004596907756877018,
|
|
"loss": 6.3637,
|
|
"mean_token_accuracy": 0.16104139834642411,
|
|
"num_tokens": 6131317.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.617832326889038,
|
|
"epoch": 2.883040935672515,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00045942265631126935,
|
|
"loss": 6.4151,
|
|
"mean_token_accuracy": 0.1558060348033905,
|
|
"num_tokens": 6143721.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.680070495605468,
|
|
"epoch": 2.888888888888889,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004591537365272837,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.1582871437072754,
|
|
"num_tokens": 6156497.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.555939054489135,
|
|
"epoch": 2.8947368421052633,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000458884017502876,
|
|
"loss": 6.3992,
|
|
"mean_token_accuracy": 0.16497224867343901,
|
|
"num_tokens": 6168140.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.706430339813233,
|
|
"epoch": 2.9005847953216373,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004586135004086461,
|
|
"loss": 6.3916,
|
|
"mean_token_accuracy": 0.15955118238925933,
|
|
"num_tokens": 6181067.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.61524453163147,
|
|
"epoch": 2.9064327485380117,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004583421864186577,
|
|
"loss": 6.3744,
|
|
"mean_token_accuracy": 0.15712310373783112,
|
|
"num_tokens": 6194072.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.65601954460144,
|
|
"epoch": 2.912280701754386,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004580700767104328,
|
|
"loss": 6.4397,
|
|
"mean_token_accuracy": 0.16010524332523346,
|
|
"num_tokens": 6206711.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.5722051620483395,
|
|
"epoch": 2.91812865497076,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045779717246494724,
|
|
"loss": 6.3593,
|
|
"mean_token_accuracy": 0.16191931813955307,
|
|
"num_tokens": 6219903.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.632825469970703,
|
|
"epoch": 2.9239766081871346,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0004575234748666249,
|
|
"loss": 6.4107,
|
|
"mean_token_accuracy": 0.15712305307388305,
|
|
"num_tokens": 6232947.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.9239766081871346,
|
|
"eval_entropy": 6.474506407712413,
|
|
"eval_loss": 6.6858367919921875,
|
|
"eval_mean_token_accuracy": 0.14477067293769577,
|
|
"eval_num_tokens": 6232947.0,
|
|
"eval_runtime": 1.7302,
|
|
"eval_samples_per_second": 1564.56,
|
|
"eval_steps_per_second": 195.931,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.636260461807251,
|
|
"epoch": 2.9298245614035086,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000457248985103333,
|
|
"loss": 6.3709,
|
|
"mean_token_accuracy": 0.16393661350011826,
|
|
"num_tokens": 6245597.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.603361558914185,
|
|
"epoch": 2.935672514619883,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004569737043663768,
|
|
"loss": 6.3252,
|
|
"mean_token_accuracy": 0.15618091374635695,
|
|
"num_tokens": 6257745.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.602143621444702,
|
|
"epoch": 2.9415204678362574,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004566976338504945,
|
|
"loss": 6.3858,
|
|
"mean_token_accuracy": 0.1644023910164833,
|
|
"num_tokens": 6268871.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.5644202709198,
|
|
"epoch": 2.9473684210526314,
|
|
"grad_norm": 0.796875,
|
|
"learning_rate": 0.00045642077475385186,
|
|
"loss": 6.3714,
|
|
"mean_token_accuracy": 0.16085665076971054,
|
|
"num_tokens": 6281487.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.543654108047486,
|
|
"epoch": 2.953216374269006,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004561431282780372,
|
|
"loss": 6.3451,
|
|
"mean_token_accuracy": 0.16275432705879211,
|
|
"num_tokens": 6292928.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.575348234176635,
|
|
"epoch": 2.95906432748538,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00045586469562805624,
|
|
"loss": 6.3098,
|
|
"mean_token_accuracy": 0.17834650427103044,
|
|
"num_tokens": 6304340.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.595432710647583,
|
|
"epoch": 2.9649122807017543,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004555854780123265,
|
|
"loss": 6.39,
|
|
"mean_token_accuracy": 0.160810686647892,
|
|
"num_tokens": 6315972.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.6777137279510494,
|
|
"epoch": 2.9707602339181287,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004553054766426725,
|
|
"loss": 6.3921,
|
|
"mean_token_accuracy": 0.16103890985250474,
|
|
"num_tokens": 6327482.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.561354255676269,
|
|
"epoch": 2.976608187134503,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004550246927343203,
|
|
"loss": 6.369,
|
|
"mean_token_accuracy": 0.16236899346113204,
|
|
"num_tokens": 6338932.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.559825801849366,
|
|
"epoch": 2.982456140350877,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00045474312750589226,
|
|
"loss": 6.4044,
|
|
"mean_token_accuracy": 0.16020567417144777,
|
|
"num_tokens": 6350770.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.638618278503418,
|
|
"epoch": 2.9883040935672516,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00045446078217940155,
|
|
"loss": 6.4193,
|
|
"mean_token_accuracy": 0.15757756531238556,
|
|
"num_tokens": 6363472.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.662160015106201,
|
|
"epoch": 2.9941520467836256,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00045417765798024716,
|
|
"loss": 6.3811,
|
|
"mean_token_accuracy": 0.15905229449272157,
|
|
"num_tokens": 6377174.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.551484727859497,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004538937561372086,
|
|
"loss": 6.3637,
|
|
"mean_token_accuracy": 0.157417456805706,
|
|
"num_tokens": 6388452.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.70260968208313,
|
|
"epoch": 3.0058479532163744,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004536090778824398,
|
|
"loss": 6.0832,
|
|
"mean_token_accuracy": 0.17075288593769072,
|
|
"num_tokens": 6401740.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.546187019348144,
|
|
"epoch": 3.0116959064327484,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.0004533236244514652,
|
|
"loss": 6.0578,
|
|
"mean_token_accuracy": 0.17957376539707184,
|
|
"num_tokens": 6414850.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.6370275020599365,
|
|
"epoch": 3.017543859649123,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00045303739708317303,
|
|
"loss": 6.2139,
|
|
"mean_token_accuracy": 0.15932946354150773,
|
|
"num_tokens": 6429250.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.571931838989258,
|
|
"epoch": 3.023391812865497,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004527503970198105,
|
|
"loss": 6.0364,
|
|
"mean_token_accuracy": 0.17207318246364595,
|
|
"num_tokens": 6440583.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.532599782943725,
|
|
"epoch": 3.0292397660818713,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004524626255069786,
|
|
"loss": 6.0535,
|
|
"mean_token_accuracy": 0.17217930853366853,
|
|
"num_tokens": 6452678.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.606011962890625,
|
|
"epoch": 3.0350877192982457,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045217408379362635,
|
|
"loss": 6.0997,
|
|
"mean_token_accuracy": 0.16845548003911973,
|
|
"num_tokens": 6464751.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.45399317741394,
|
|
"epoch": 3.0409356725146197,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045188477313204536,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.1708752527832985,
|
|
"num_tokens": 6476017.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.657937335968017,
|
|
"epoch": 3.046783625730994,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045159469477786457,
|
|
"loss": 6.0946,
|
|
"mean_token_accuracy": 0.1711534798145294,
|
|
"num_tokens": 6487404.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.580844783782959,
|
|
"epoch": 3.0526315789473686,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004513038499900451,
|
|
"loss": 6.1365,
|
|
"mean_token_accuracy": 0.16481128931045533,
|
|
"num_tokens": 6500027.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.521955156326294,
|
|
"epoch": 3.0584795321637426,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000451012240030874,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.1614599958062172,
|
|
"num_tokens": 6513067.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.583660364151001,
|
|
"epoch": 3.064327485380117,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045071986616595946,
|
|
"loss": 6.0837,
|
|
"mean_token_accuracy": 0.16450167149305345,
|
|
"num_tokens": 6524932.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.562144422531128,
|
|
"epoch": 3.0701754385964914,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.00045042672966422506,
|
|
"loss": 6.1179,
|
|
"mean_token_accuracy": 0.16554533541202546,
|
|
"num_tokens": 6536843.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.591558980941772,
|
|
"epoch": 3.0760233918128654,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004501328317979041,
|
|
"loss": 6.0298,
|
|
"mean_token_accuracy": 0.17564562559127808,
|
|
"num_tokens": 6549414.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.512995338439941,
|
|
"epoch": 3.08187134502924,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004498381738425346,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.1698380097746849,
|
|
"num_tokens": 6561492.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.513795471191406,
|
|
"epoch": 3.087719298245614,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044954275707695304,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.17263186872005462,
|
|
"num_tokens": 6572873.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.562921571731567,
|
|
"epoch": 3.0935672514619883,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004492465827832894,
|
|
"loss": 6.1596,
|
|
"mean_token_accuracy": 0.17121457010507585,
|
|
"num_tokens": 6584902.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.5826499462127686,
|
|
"epoch": 3.0994152046783627,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00044894965224696144,
|
|
"loss": 6.1754,
|
|
"mean_token_accuracy": 0.1667486160993576,
|
|
"num_tokens": 6596792.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.559206581115722,
|
|
"epoch": 3.1052631578947367,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044865196675666886,
|
|
"loss": 6.11,
|
|
"mean_token_accuracy": 0.1693141758441925,
|
|
"num_tokens": 6608559.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.5065614700317385,
|
|
"epoch": 3.111111111111111,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004483535276043881,
|
|
"loss": 6.1453,
|
|
"mean_token_accuracy": 0.16696794629096984,
|
|
"num_tokens": 6619938.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.5608642578125,
|
|
"epoch": 3.116959064327485,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004480543360853665,
|
|
"loss": 6.1206,
|
|
"mean_token_accuracy": 0.1688737764954567,
|
|
"num_tokens": 6632850.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.485915756225586,
|
|
"epoch": 3.1228070175438596,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004477543934981167,
|
|
"loss": 6.0462,
|
|
"mean_token_accuracy": 0.1824930265545845,
|
|
"num_tokens": 6645761.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.5529155254364015,
|
|
"epoch": 3.128654970760234,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044745370114441103,
|
|
"loss": 6.1162,
|
|
"mean_token_accuracy": 0.17532826513051986,
|
|
"num_tokens": 6657861.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.529551315307617,
|
|
"epoch": 3.134502923976608,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044715226032927585,
|
|
"loss": 6.0875,
|
|
"mean_token_accuracy": 0.1792955845594406,
|
|
"num_tokens": 6670495.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.374896669387818,
|
|
"epoch": 3.1403508771929824,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004468500723609859,
|
|
"loss": 6.0383,
|
|
"mean_token_accuracy": 0.174404776096344,
|
|
"num_tokens": 6683127.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.563790512084961,
|
|
"epoch": 3.146198830409357,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044654713855105864,
|
|
"loss": 6.1114,
|
|
"mean_token_accuracy": 0.16159912645816804,
|
|
"num_tokens": 6696159.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.524526977539063,
|
|
"epoch": 3.152046783625731,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004462434602142486,
|
|
"loss": 6.1785,
|
|
"mean_token_accuracy": 0.16437700539827346,
|
|
"num_tokens": 6708993.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.564835500717163,
|
|
"epoch": 3.1578947368421053,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004459390386685414,
|
|
"loss": 6.2003,
|
|
"mean_token_accuracy": 0.1650101974606514,
|
|
"num_tokens": 6722121.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.484952974319458,
|
|
"epoch": 3.1637426900584797,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044563387523514854,
|
|
"loss": 6.023,
|
|
"mean_token_accuracy": 0.17607033401727676,
|
|
"num_tokens": 6734167.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.459822940826416,
|
|
"epoch": 3.1695906432748537,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004453279712385011,
|
|
"loss": 6.1071,
|
|
"mean_token_accuracy": 0.16553861498832703,
|
|
"num_tokens": 6745720.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.503742504119873,
|
|
"epoch": 3.175438596491228,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004450213280062444,
|
|
"loss": 6.0939,
|
|
"mean_token_accuracy": 0.16853866428136827,
|
|
"num_tokens": 6757523.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.568593549728393,
|
|
"epoch": 3.181286549707602,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044471394686923206,
|
|
"loss": 6.121,
|
|
"mean_token_accuracy": 0.16629476100206375,
|
|
"num_tokens": 6768775.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.525596046447754,
|
|
"epoch": 3.1871345029239766,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004444058291615202,
|
|
"loss": 6.1133,
|
|
"mean_token_accuracy": 0.16562219709157944,
|
|
"num_tokens": 6781547.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.485271024703979,
|
|
"epoch": 3.192982456140351,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004440969762203618,
|
|
"loss": 6.1516,
|
|
"mean_token_accuracy": 0.16345081925392152,
|
|
"num_tokens": 6794708.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.508287048339843,
|
|
"epoch": 3.198830409356725,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004437873893862008,
|
|
"loss": 6.0773,
|
|
"mean_token_accuracy": 0.17581354528665544,
|
|
"num_tokens": 6807738.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.511159420013428,
|
|
"epoch": 3.2046783625730995,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044347707000266627,
|
|
"loss": 6.1269,
|
|
"mean_token_accuracy": 0.1693233847618103,
|
|
"num_tokens": 6820851.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.548748731613159,
|
|
"epoch": 3.2105263157894735,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044316601941656655,
|
|
"loss": 6.0793,
|
|
"mean_token_accuracy": 0.1794822096824646,
|
|
"num_tokens": 6833888.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.494293260574341,
|
|
"epoch": 3.216374269005848,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004428542389778834,
|
|
"loss": 6.1172,
|
|
"mean_token_accuracy": 0.17162996679544448,
|
|
"num_tokens": 6846104.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.45095477104187,
|
|
"epoch": 3.2222222222222223,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004425417300397665,
|
|
"loss": 6.1045,
|
|
"mean_token_accuracy": 0.17195063680410386,
|
|
"num_tokens": 6857458.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.5355345726013185,
|
|
"epoch": 3.2280701754385963,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00044222849395852683,
|
|
"loss": 6.074,
|
|
"mean_token_accuracy": 0.17359959185123444,
|
|
"num_tokens": 6870455.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.5262401580810545,
|
|
"epoch": 3.2339181286549707,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044191453209363165,
|
|
"loss": 6.1379,
|
|
"mean_token_accuracy": 0.1655440717935562,
|
|
"num_tokens": 6883359.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.476034498214721,
|
|
"epoch": 3.239766081871345,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004415998458076978,
|
|
"loss": 6.0727,
|
|
"mean_token_accuracy": 0.17608768194913865,
|
|
"num_tokens": 6894940.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.5159460544586185,
|
|
"epoch": 3.245614035087719,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044128443646648634,
|
|
"loss": 6.1268,
|
|
"mean_token_accuracy": 0.16884497851133345,
|
|
"num_tokens": 6905422.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.537970733642578,
|
|
"epoch": 3.2514619883040936,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044096830543889653,
|
|
"loss": 6.1286,
|
|
"mean_token_accuracy": 0.1716626450419426,
|
|
"num_tokens": 6918106.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.5358724117279055,
|
|
"epoch": 3.257309941520468,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004406514540969596,
|
|
"loss": 6.1015,
|
|
"mean_token_accuracy": 0.17332532852888108,
|
|
"num_tokens": 6930975.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.489678764343262,
|
|
"epoch": 3.263157894736842,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00044033388381583324,
|
|
"loss": 6.1935,
|
|
"mean_token_accuracy": 0.16382682770490647,
|
|
"num_tokens": 6944436.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.538244295120239,
|
|
"epoch": 3.2690058479532165,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00044001559597379503,
|
|
"loss": 6.1451,
|
|
"mean_token_accuracy": 0.16451818645000457,
|
|
"num_tokens": 6957095.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.494568872451782,
|
|
"epoch": 3.2748538011695905,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004396965919522373,
|
|
"loss": 6.0881,
|
|
"mean_token_accuracy": 0.17684150040149688,
|
|
"num_tokens": 6969792.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.4879639625549315,
|
|
"epoch": 3.280701754385965,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00043937687313565997,
|
|
"loss": 6.143,
|
|
"mean_token_accuracy": 0.16732580810785294,
|
|
"num_tokens": 6982054.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.493925428390503,
|
|
"epoch": 3.2865497076023393,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004390564409116661,
|
|
"loss": 6.1371,
|
|
"mean_token_accuracy": 0.16817447543144226,
|
|
"num_tokens": 6994057.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.474679803848266,
|
|
"epoch": 3.2923976608187133,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.00043873529667095423,
|
|
"loss": 6.1393,
|
|
"mean_token_accuracy": 0.1712675377726555,
|
|
"num_tokens": 7006417.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.477682638168335,
|
|
"epoch": 3.2982456140350878,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00043841344180731346,
|
|
"loss": 6.1547,
|
|
"mean_token_accuracy": 0.16124001741409302,
|
|
"num_tokens": 7018331.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.518821668624878,
|
|
"epoch": 3.3040935672514617,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000438090877717617,
|
|
"loss": 6.04,
|
|
"mean_token_accuracy": 0.17431365847587585,
|
|
"num_tokens": 7031654.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.5531683444976805,
|
|
"epoch": 3.309941520467836,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004377676058018161,
|
|
"loss": 6.1393,
|
|
"mean_token_accuracy": 0.1715486004948616,
|
|
"num_tokens": 7044921.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.42981367111206,
|
|
"epoch": 3.3157894736842106,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004374436274629341,
|
|
"loss": 6.0735,
|
|
"mean_token_accuracy": 0.16862040609121323,
|
|
"num_tokens": 7056700.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.495500755310059,
|
|
"epoch": 3.3216374269005846,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004371189441070602,
|
|
"loss": 6.0191,
|
|
"mean_token_accuracy": 0.16978203654289245,
|
|
"num_tokens": 7069196.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.441921043395996,
|
|
"epoch": 3.327485380116959,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004367935571433433,
|
|
"loss": 6.0614,
|
|
"mean_token_accuracy": 0.17559678256511688,
|
|
"num_tokens": 7081528.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.566071462631226,
|
|
"epoch": 3.3333333333333335,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004364674679839864,
|
|
"loss": 6.1552,
|
|
"mean_token_accuracy": 0.16209713369607925,
|
|
"num_tokens": 7093474.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.382375192642212,
|
|
"epoch": 3.3391812865497075,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004361406780442398,
|
|
"loss": 6.0816,
|
|
"mean_token_accuracy": 0.17418255656957626,
|
|
"num_tokens": 7105508.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.453981208801269,
|
|
"epoch": 3.345029239766082,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00043581318874239504,
|
|
"loss": 6.0926,
|
|
"mean_token_accuracy": 0.1744996801018715,
|
|
"num_tokens": 7117952.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.468331241607666,
|
|
"epoch": 3.3508771929824563,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043548500149977946,
|
|
"loss": 6.0771,
|
|
"mean_token_accuracy": 0.1764598608016968,
|
|
"num_tokens": 7130666.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.466050815582276,
|
|
"epoch": 3.3567251461988303,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004351561177407491,
|
|
"loss": 6.1137,
|
|
"mean_token_accuracy": 0.17108856737613679,
|
|
"num_tokens": 7142717.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.552111864089966,
|
|
"epoch": 3.3625730994152048,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004348265388926831,
|
|
"loss": 6.15,
|
|
"mean_token_accuracy": 0.1713667944073677,
|
|
"num_tokens": 7154794.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.496505641937256,
|
|
"epoch": 3.3684210526315788,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004344962663859771,
|
|
"loss": 6.1995,
|
|
"mean_token_accuracy": 0.16334100663661957,
|
|
"num_tokens": 7168907.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.445120096206665,
|
|
"epoch": 3.374269005847953,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00043416530165403777,
|
|
"loss": 6.033,
|
|
"mean_token_accuracy": 0.18498492389917373,
|
|
"num_tokens": 7180828.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.360263347625732,
|
|
"epoch": 3.3801169590643276,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004338336461332756,
|
|
"loss": 6.0388,
|
|
"mean_token_accuracy": 0.17281053215265274,
|
|
"num_tokens": 7193493.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.499506950378418,
|
|
"epoch": 3.3859649122807016,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043350130126309936,
|
|
"loss": 6.083,
|
|
"mean_token_accuracy": 0.16768689304590226,
|
|
"num_tokens": 7206033.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.505532598495483,
|
|
"epoch": 3.391812865497076,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00043316826848590955,
|
|
"loss": 6.114,
|
|
"mean_token_accuracy": 0.16994736790657045,
|
|
"num_tokens": 7218813.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.444248628616333,
|
|
"epoch": 3.39766081871345,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004328345492470924,
|
|
"loss": 6.0908,
|
|
"mean_token_accuracy": 0.17362091243267058,
|
|
"num_tokens": 7230698.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.474144983291626,
|
|
"epoch": 3.4035087719298245,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004325001449950134,
|
|
"loss": 6.1217,
|
|
"mean_token_accuracy": 0.1663724511861801,
|
|
"num_tokens": 7243354.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.392809629440308,
|
|
"epoch": 3.409356725146199,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004321650571810109,
|
|
"loss": 6.0684,
|
|
"mean_token_accuracy": 0.17750578969717026,
|
|
"num_tokens": 7255855.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.396269178390503,
|
|
"epoch": 3.415204678362573,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004318292872593899,
|
|
"loss": 6.1589,
|
|
"mean_token_accuracy": 0.16955514401197433,
|
|
"num_tokens": 7268921.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.550719833374023,
|
|
"epoch": 3.4210526315789473,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004314928366874162,
|
|
"loss": 6.1483,
|
|
"mean_token_accuracy": 0.16568040251731872,
|
|
"num_tokens": 7281467.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.365495777130127,
|
|
"epoch": 3.426900584795322,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004311557069253092,
|
|
"loss": 6.0591,
|
|
"mean_token_accuracy": 0.17555536329746246,
|
|
"num_tokens": 7294887.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.547408819198608,
|
|
"epoch": 3.4327485380116958,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004308178994362364,
|
|
"loss": 6.1935,
|
|
"mean_token_accuracy": 0.17171871960163115,
|
|
"num_tokens": 7307009.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.42487940788269,
|
|
"epoch": 3.43859649122807,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00043047941568630644,
|
|
"loss": 6.1048,
|
|
"mean_token_accuracy": 0.16843656450510025,
|
|
"num_tokens": 7319751.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.515281867980957,
|
|
"epoch": 3.4444444444444446,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043014025714456305,
|
|
"loss": 6.2338,
|
|
"mean_token_accuracy": 0.16237418949604035,
|
|
"num_tokens": 7333952.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.445792150497437,
|
|
"epoch": 3.4502923976608186,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004298004252829787,
|
|
"loss": 5.988,
|
|
"mean_token_accuracy": 0.1804724305868149,
|
|
"num_tokens": 7346834.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.475040483474731,
|
|
"epoch": 3.456140350877193,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000429459921576448,
|
|
"loss": 6.1279,
|
|
"mean_token_accuracy": 0.16938423365354538,
|
|
"num_tokens": 7359165.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.506986474990844,
|
|
"epoch": 3.461988304093567,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00042911874750278133,
|
|
"loss": 6.1996,
|
|
"mean_token_accuracy": 0.16553546041250228,
|
|
"num_tokens": 7371585.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.432463550567627,
|
|
"epoch": 3.4678362573099415,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00042877690454269884,
|
|
"loss": 6.0887,
|
|
"mean_token_accuracy": 0.17572804987430574,
|
|
"num_tokens": 7383455.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.420023965835571,
|
|
"epoch": 3.473684210526316,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00042843439417982344,
|
|
"loss": 6.1506,
|
|
"mean_token_accuracy": 0.17092279493808746,
|
|
"num_tokens": 7395918.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.524562644958496,
|
|
"epoch": 3.47953216374269,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004280912179006746,
|
|
"loss": 6.1334,
|
|
"mean_token_accuracy": 0.1703528344631195,
|
|
"num_tokens": 7408053.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.497876214981079,
|
|
"epoch": 3.4853801169590644,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.000427747377194662,
|
|
"loss": 6.1778,
|
|
"mean_token_accuracy": 0.16454286128282547,
|
|
"num_tokens": 7419947.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.438177871704101,
|
|
"epoch": 3.4912280701754383,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004274028735540789,
|
|
"loss": 6.0593,
|
|
"mean_token_accuracy": 0.16666590571403503,
|
|
"num_tokens": 7433022.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.411357355117798,
|
|
"epoch": 3.497076023391813,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042705770847409596,
|
|
"loss": 6.0595,
|
|
"mean_token_accuracy": 0.17704250365495683,
|
|
"num_tokens": 7445280.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.5110695362091064,
|
|
"epoch": 3.502923976608187,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042671188345275423,
|
|
"loss": 6.0611,
|
|
"mean_token_accuracy": 0.17194437682628633,
|
|
"num_tokens": 7457324.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.45655927658081,
|
|
"epoch": 3.5087719298245617,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004263653999909593,
|
|
"loss": 6.1125,
|
|
"mean_token_accuracy": 0.16834226250648499,
|
|
"num_tokens": 7469070.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.5087719298245617,
|
|
"eval_entropy": 6.256012293441457,
|
|
"eval_loss": 6.655336380004883,
|
|
"eval_mean_token_accuracy": 0.14697449852418495,
|
|
"eval_num_tokens": 7469070.0,
|
|
"eval_runtime": 1.7405,
|
|
"eval_samples_per_second": 1555.284,
|
|
"eval_steps_per_second": 194.77,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"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": 1600845571952640.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|