0b04222315
Model: fpadovani/swe-latn-10mb-after-ppt-Dp-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": 0.4142502071251036,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.316415309906006,
|
|
"epoch": 0.004142502071251036,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.9577,
|
|
"mean_token_accuracy": 0.09107546955347061,
|
|
"num_tokens": 9149.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.29176926612854,
|
|
"epoch": 0.008285004142502071,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.815,
|
|
"mean_token_accuracy": 0.09366189837455749,
|
|
"num_tokens": 16800.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.258940315246582,
|
|
"epoch": 0.012427506213753107,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.949,
|
|
"mean_token_accuracy": 0.09261607527732849,
|
|
"num_tokens": 26617.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.274951076507568,
|
|
"epoch": 0.016570008285004142,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.8768,
|
|
"mean_token_accuracy": 0.09439559504389763,
|
|
"num_tokens": 35855.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.333089065551758,
|
|
"epoch": 0.020712510356255178,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.0807,
|
|
"mean_token_accuracy": 0.08799155950546264,
|
|
"num_tokens": 45644.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.298266220092773,
|
|
"epoch": 0.024855012427506214,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.9578,
|
|
"mean_token_accuracy": 0.09257320761680603,
|
|
"num_tokens": 54641.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.2906664371490475,
|
|
"epoch": 0.02899751449875725,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.8371,
|
|
"mean_token_accuracy": 0.08952230289578438,
|
|
"num_tokens": 63924.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.365257406234742,
|
|
"epoch": 0.033140016570008285,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.9731,
|
|
"mean_token_accuracy": 0.09005696997046471,
|
|
"num_tokens": 73452.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.326370525360107,
|
|
"epoch": 0.037282518641259324,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.0211,
|
|
"mean_token_accuracy": 0.0873270682990551,
|
|
"num_tokens": 82731.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.339514112472534,
|
|
"epoch": 0.041425020712510356,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.08901808261871338,
|
|
"num_tokens": 92626.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.389622354507447,
|
|
"epoch": 0.045567522783761395,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.0312,
|
|
"mean_token_accuracy": 0.08305057361721993,
|
|
"num_tokens": 101380.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.367127227783203,
|
|
"epoch": 0.04971002485501243,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.9496,
|
|
"mean_token_accuracy": 0.09191007390618325,
|
|
"num_tokens": 109895.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.404217004776001,
|
|
"epoch": 0.053852526926263466,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.9489,
|
|
"mean_token_accuracy": 0.08632109612226486,
|
|
"num_tokens": 119203.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.383033037185669,
|
|
"epoch": 0.0579950289975145,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.8909,
|
|
"mean_token_accuracy": 0.0960758812725544,
|
|
"num_tokens": 127572.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.323318862915039,
|
|
"epoch": 0.06213753106876554,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.924,
|
|
"mean_token_accuracy": 0.08685562647879123,
|
|
"num_tokens": 137161.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.288953590393066,
|
|
"epoch": 0.06628003314001657,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 7.0298,
|
|
"mean_token_accuracy": 0.08434623554348945,
|
|
"num_tokens": 145991.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.325984477996826,
|
|
"epoch": 0.07042253521126761,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.937,
|
|
"mean_token_accuracy": 0.0885515496134758,
|
|
"num_tokens": 154927.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.31841459274292,
|
|
"epoch": 0.07456503728251865,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.8509,
|
|
"mean_token_accuracy": 0.09631360918283463,
|
|
"num_tokens": 163234.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.365633821487426,
|
|
"epoch": 0.07870753935376967,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.8134,
|
|
"mean_token_accuracy": 0.09503689482808113,
|
|
"num_tokens": 171554.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.308236169815063,
|
|
"epoch": 0.08285004142502071,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.9083,
|
|
"mean_token_accuracy": 0.09420166239142418,
|
|
"num_tokens": 180171.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.272618198394776,
|
|
"epoch": 0.08699254349627175,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.8556,
|
|
"mean_token_accuracy": 0.09322545379400253,
|
|
"num_tokens": 189216.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.345879077911377,
|
|
"epoch": 0.09113504556752279,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.0522,
|
|
"mean_token_accuracy": 0.08772843405604362,
|
|
"num_tokens": 200022.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.375020217895508,
|
|
"epoch": 0.09527754763877382,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.9495,
|
|
"mean_token_accuracy": 0.0907869316637516,
|
|
"num_tokens": 208686.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.387457752227784,
|
|
"epoch": 0.09942004971002485,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.9723,
|
|
"mean_token_accuracy": 0.09134301021695138,
|
|
"num_tokens": 218266.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.271829748153687,
|
|
"epoch": 0.1035625517812759,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.8557,
|
|
"mean_token_accuracy": 0.08564562872052192,
|
|
"num_tokens": 227658.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.235640192031861,
|
|
"epoch": 0.10770505385252693,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.8059,
|
|
"mean_token_accuracy": 0.0986194871366024,
|
|
"num_tokens": 236301.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.342151260375976,
|
|
"epoch": 0.11184755592377796,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.9379,
|
|
"mean_token_accuracy": 0.08733796998858452,
|
|
"num_tokens": 246355.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.415835380554199,
|
|
"epoch": 0.115990057995029,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.9199,
|
|
"mean_token_accuracy": 0.08878825083374978,
|
|
"num_tokens": 256023.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.297379016876221,
|
|
"epoch": 0.12013256006628004,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.0053,
|
|
"mean_token_accuracy": 0.08748594373464584,
|
|
"num_tokens": 266893.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.238939237594605,
|
|
"epoch": 0.12427506213753108,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.7696,
|
|
"mean_token_accuracy": 0.09012725874781609,
|
|
"num_tokens": 276075.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.302191066741943,
|
|
"epoch": 0.12841756420878211,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.8603,
|
|
"mean_token_accuracy": 0.08920295760035515,
|
|
"num_tokens": 285529.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.233504438400269,
|
|
"epoch": 0.13256006628003314,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.866,
|
|
"mean_token_accuracy": 0.09255343899130822,
|
|
"num_tokens": 294403.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.361068534851074,
|
|
"epoch": 0.13670256835128416,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.9542,
|
|
"mean_token_accuracy": 0.0920793354511261,
|
|
"num_tokens": 303507.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.357890844345093,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.864,
|
|
"mean_token_accuracy": 0.09164286777377129,
|
|
"num_tokens": 312393.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.18741021156311,
|
|
"epoch": 0.14498757249378624,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.8295,
|
|
"mean_token_accuracy": 0.09937396124005318,
|
|
"num_tokens": 321309.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.408359384536743,
|
|
"epoch": 0.1491300745650373,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.885,
|
|
"mean_token_accuracy": 0.09395725503563881,
|
|
"num_tokens": 330217.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.256880092620849,
|
|
"epoch": 0.15327257663628832,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.9273,
|
|
"mean_token_accuracy": 0.09338772147893906,
|
|
"num_tokens": 339587.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.283669328689575,
|
|
"epoch": 0.15741507870753935,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.8494,
|
|
"mean_token_accuracy": 0.09577088207006454,
|
|
"num_tokens": 349367.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.2921398162841795,
|
|
"epoch": 0.1615575807787904,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.7565,
|
|
"mean_token_accuracy": 0.09605895429849624,
|
|
"num_tokens": 358272.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.314898586273193,
|
|
"epoch": 0.16570008285004142,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.9815,
|
|
"mean_token_accuracy": 0.09187402427196503,
|
|
"num_tokens": 367709.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.341378116607666,
|
|
"epoch": 0.16984258492129245,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.8415,
|
|
"mean_token_accuracy": 0.09144494608044625,
|
|
"num_tokens": 376180.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.2672590732574465,
|
|
"epoch": 0.1739850869925435,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.9873,
|
|
"mean_token_accuracy": 0.08560611009597778,
|
|
"num_tokens": 385636.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.31936240196228,
|
|
"epoch": 0.17812758906379453,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.8665,
|
|
"mean_token_accuracy": 0.10073013603687286,
|
|
"num_tokens": 394486.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.240320158004761,
|
|
"epoch": 0.18227009113504558,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.8339,
|
|
"mean_token_accuracy": 0.09145828187465668,
|
|
"num_tokens": 403765.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.176674270629883,
|
|
"epoch": 0.1864125932062966,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.8192,
|
|
"mean_token_accuracy": 0.09741753935813904,
|
|
"num_tokens": 412988.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.243546390533448,
|
|
"epoch": 0.19055509527754763,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.7247,
|
|
"mean_token_accuracy": 0.09494704753160477,
|
|
"num_tokens": 422096.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.259120273590088,
|
|
"epoch": 0.19469759734879868,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.8913,
|
|
"mean_token_accuracy": 0.0842115230858326,
|
|
"num_tokens": 431333.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.322444343566895,
|
|
"epoch": 0.1988400994200497,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.9209,
|
|
"mean_token_accuracy": 0.09764085784554481,
|
|
"num_tokens": 440382.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.303359937667847,
|
|
"epoch": 0.20298260149130073,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.8237,
|
|
"mean_token_accuracy": 0.09803626388311386,
|
|
"num_tokens": 448469.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.2807892799377445,
|
|
"epoch": 0.2071251035625518,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.8631,
|
|
"mean_token_accuracy": 0.09314201548695564,
|
|
"num_tokens": 457356.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.395456218719483,
|
|
"epoch": 0.2112676056338028,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.8936,
|
|
"mean_token_accuracy": 0.09395477920770645,
|
|
"num_tokens": 465928.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.177229928970337,
|
|
"epoch": 0.21541010770505387,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.7914,
|
|
"mean_token_accuracy": 0.09916061088442803,
|
|
"num_tokens": 475269.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.238220596313477,
|
|
"epoch": 0.2195526097763049,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.7807,
|
|
"mean_token_accuracy": 0.09297533929347992,
|
|
"num_tokens": 485450.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.214696550369263,
|
|
"epoch": 0.22369511184755592,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.8391,
|
|
"mean_token_accuracy": 0.10058524161577224,
|
|
"num_tokens": 494029.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.255264091491699,
|
|
"epoch": 0.22783761391880697,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.7793,
|
|
"mean_token_accuracy": 0.10096407309174538,
|
|
"num_tokens": 503253.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.156082010269165,
|
|
"epoch": 0.231980115990058,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.7768,
|
|
"mean_token_accuracy": 0.09033024087548255,
|
|
"num_tokens": 512857.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.303338527679443,
|
|
"epoch": 0.23612261806130902,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8719,
|
|
"mean_token_accuracy": 0.09036988765001297,
|
|
"num_tokens": 522319.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.218528747558594,
|
|
"epoch": 0.24026512013256007,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8263,
|
|
"mean_token_accuracy": 0.09190768375992775,
|
|
"num_tokens": 531469.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.2514721870422365,
|
|
"epoch": 0.2444076222038111,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8537,
|
|
"mean_token_accuracy": 0.09450737237930298,
|
|
"num_tokens": 540470.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.243441724777222,
|
|
"epoch": 0.24855012427506215,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.0973442628979683,
|
|
"num_tokens": 550317.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.2743926525115965,
|
|
"epoch": 0.2526926263463132,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.8027,
|
|
"mean_token_accuracy": 0.09109159633517265,
|
|
"num_tokens": 559409.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.125630617141724,
|
|
"epoch": 0.25683512841756423,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8497,
|
|
"mean_token_accuracy": 0.10476837977766991,
|
|
"num_tokens": 568726.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.3812638282775875,
|
|
"epoch": 0.2609776304888152,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.8242,
|
|
"mean_token_accuracy": 0.0930200308561325,
|
|
"num_tokens": 577926.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.191781997680664,
|
|
"epoch": 0.2651201325600663,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.8386,
|
|
"mean_token_accuracy": 0.09706645980477333,
|
|
"num_tokens": 586695.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.304094886779785,
|
|
"epoch": 0.26926263463131733,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.8637,
|
|
"mean_token_accuracy": 0.09343150332570076,
|
|
"num_tokens": 596250.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.277995014190674,
|
|
"epoch": 0.27340513670256833,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9091,
|
|
"mean_token_accuracy": 0.09041907489299775,
|
|
"num_tokens": 605761.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.221877241134644,
|
|
"epoch": 0.2775476387738194,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8331,
|
|
"mean_token_accuracy": 0.09750878065824509,
|
|
"num_tokens": 614313.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.308935832977295,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.86,
|
|
"mean_token_accuracy": 0.09677010327577591,
|
|
"num_tokens": 624470.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.216413593292236,
|
|
"epoch": 0.28583264291632143,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7866,
|
|
"mean_token_accuracy": 0.09221929684281349,
|
|
"num_tokens": 633458.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.228837203979492,
|
|
"epoch": 0.2899751449875725,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.7796,
|
|
"mean_token_accuracy": 0.095080154389143,
|
|
"num_tokens": 643251.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.234794759750367,
|
|
"epoch": 0.29411764705882354,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8515,
|
|
"mean_token_accuracy": 0.09346813485026359,
|
|
"num_tokens": 652005.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.190905666351318,
|
|
"epoch": 0.2982601491300746,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7442,
|
|
"mean_token_accuracy": 0.0977424830198288,
|
|
"num_tokens": 661003.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.124666595458985,
|
|
"epoch": 0.3024026512013256,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.7269,
|
|
"mean_token_accuracy": 0.09757486581802369,
|
|
"num_tokens": 670221.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.42992148399353,
|
|
"epoch": 0.30654515327257664,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.0088,
|
|
"mean_token_accuracy": 0.08780332058668136,
|
|
"num_tokens": 680756.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.103591108322144,
|
|
"epoch": 0.3106876553438277,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7083,
|
|
"mean_token_accuracy": 0.09910113140940666,
|
|
"num_tokens": 689641.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.170710134506225,
|
|
"epoch": 0.3148301574150787,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.6639,
|
|
"mean_token_accuracy": 0.09918258264660836,
|
|
"num_tokens": 698209.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.087573099136352,
|
|
"epoch": 0.31897265948632975,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.6625,
|
|
"mean_token_accuracy": 0.10349222794175147,
|
|
"num_tokens": 706712.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.24962100982666,
|
|
"epoch": 0.3231151615575808,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8702,
|
|
"mean_token_accuracy": 0.09640411287546158,
|
|
"num_tokens": 716087.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.140476322174072,
|
|
"epoch": 0.3272576636288318,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7452,
|
|
"mean_token_accuracy": 0.09067876264452934,
|
|
"num_tokens": 725913.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.217365264892578,
|
|
"epoch": 0.33140016570008285,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.8095,
|
|
"mean_token_accuracy": 0.10141430869698524,
|
|
"num_tokens": 735020.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.1846894264221195,
|
|
"epoch": 0.3355426677713339,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7136,
|
|
"mean_token_accuracy": 0.09213051795959473,
|
|
"num_tokens": 744299.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.125712251663208,
|
|
"epoch": 0.3396851698425849,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.6868,
|
|
"mean_token_accuracy": 0.09798395857214928,
|
|
"num_tokens": 753231.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.11104679107666,
|
|
"epoch": 0.34382767191383595,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7589,
|
|
"mean_token_accuracy": 0.09906982332468033,
|
|
"num_tokens": 763043.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.237280750274659,
|
|
"epoch": 0.347970173985087,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7865,
|
|
"mean_token_accuracy": 0.1009935662150383,
|
|
"num_tokens": 771977.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.055163717269897,
|
|
"epoch": 0.352112676056338,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7041,
|
|
"mean_token_accuracy": 0.10424543470144272,
|
|
"num_tokens": 780285.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.228680515289307,
|
|
"epoch": 0.35625517812758906,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7506,
|
|
"mean_token_accuracy": 0.09805167838931084,
|
|
"num_tokens": 789980.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.116986274719238,
|
|
"epoch": 0.3603976801988401,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7287,
|
|
"mean_token_accuracy": 0.10292106568813324,
|
|
"num_tokens": 798700.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.054797267913818,
|
|
"epoch": 0.36454018227009116,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7877,
|
|
"mean_token_accuracy": 0.09516645893454552,
|
|
"num_tokens": 808758.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.259660816192627,
|
|
"epoch": 0.36868268434134216,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.7511,
|
|
"mean_token_accuracy": 0.10131709128618241,
|
|
"num_tokens": 817692.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.174760341644287,
|
|
"epoch": 0.3728251864125932,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.8063,
|
|
"mean_token_accuracy": 0.10130528211593628,
|
|
"num_tokens": 827052.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.13604998588562,
|
|
"epoch": 0.37696768848384427,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7821,
|
|
"mean_token_accuracy": 0.10629323869943619,
|
|
"num_tokens": 836442.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.191133499145508,
|
|
"epoch": 0.38111019055509526,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6717,
|
|
"mean_token_accuracy": 0.10159996896982193,
|
|
"num_tokens": 846441.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.073972415924072,
|
|
"epoch": 0.3852526926263463,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7128,
|
|
"mean_token_accuracy": 0.10559298172593116,
|
|
"num_tokens": 856108.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.137764835357666,
|
|
"epoch": 0.38939519469759737,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.665,
|
|
"mean_token_accuracy": 0.09864805340766906,
|
|
"num_tokens": 865788.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.0983226776123045,
|
|
"epoch": 0.39353769676884837,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.7292,
|
|
"mean_token_accuracy": 0.09705496057868004,
|
|
"num_tokens": 874833.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.023185586929321,
|
|
"epoch": 0.3976801988400994,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.6816,
|
|
"mean_token_accuracy": 0.10277220755815505,
|
|
"num_tokens": 883685.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.133955621719361,
|
|
"epoch": 0.40182270091135047,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.67,
|
|
"mean_token_accuracy": 0.09730359986424446,
|
|
"num_tokens": 893908.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.086344528198242,
|
|
"epoch": 0.40596520298260147,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7808,
|
|
"mean_token_accuracy": 0.09705358296632767,
|
|
"num_tokens": 903081.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.025259399414063,
|
|
"epoch": 0.4101077050538525,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.6333,
|
|
"mean_token_accuracy": 0.10383275151252747,
|
|
"num_tokens": 912515.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.203283357620239,
|
|
"epoch": 0.4142502071251036,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6473,
|
|
"mean_token_accuracy": 0.10783689767122269,
|
|
"num_tokens": 921056.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.4142502071251036,
|
|
"eval_entropy": 6.835995750386577,
|
|
"eval_loss": 6.763827800750732,
|
|
"eval_mean_token_accuracy": 0.10202416494838211,
|
|
"eval_num_tokens": 921056.0,
|
|
"eval_runtime": 2.1276,
|
|
"eval_samples_per_second": 1761.098,
|
|
"eval_steps_per_second": 220.431,
|
|
"step": 500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12060,
|
|
"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": 204998070435840.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|