b5e9bda545
Model: fpadovani/jpn-jpan-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.5847953216374269,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 268362805248000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|