0b04222315
Model: fpadovani/swe-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.4855012427506216,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"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
|
|
},
|
|
{
|
|
"entropy": 6.907085657119751,
|
|
"epoch": 0.4183927091963546,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.611,
|
|
"mean_token_accuracy": 0.10536315441131591,
|
|
"num_tokens": 931112.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.0663305759429935,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7483,
|
|
"mean_token_accuracy": 0.10146113634109497,
|
|
"num_tokens": 941120.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.127016544342041,
|
|
"epoch": 0.4266777133388567,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.6883,
|
|
"mean_token_accuracy": 0.09923007488250732,
|
|
"num_tokens": 950041.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.057574129104614,
|
|
"epoch": 0.43082021541010773,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.6703,
|
|
"mean_token_accuracy": 0.10598697513341904,
|
|
"num_tokens": 958608.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.187314367294311,
|
|
"epoch": 0.43496271748135873,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.7852,
|
|
"mean_token_accuracy": 0.09927874878048897,
|
|
"num_tokens": 966901.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.984877586364746,
|
|
"epoch": 0.4391052195526098,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.623,
|
|
"mean_token_accuracy": 0.10634999126195907,
|
|
"num_tokens": 976198.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.135362672805786,
|
|
"epoch": 0.44324772162386084,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6937,
|
|
"mean_token_accuracy": 0.09986837655305862,
|
|
"num_tokens": 985248.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.943363761901855,
|
|
"epoch": 0.44739022369511183,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.6052,
|
|
"mean_token_accuracy": 0.1056825801730156,
|
|
"num_tokens": 993289.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.099994421005249,
|
|
"epoch": 0.4515327257663629,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6611,
|
|
"mean_token_accuracy": 0.10454976856708527,
|
|
"num_tokens": 1001784.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.149687051773071,
|
|
"epoch": 0.45567522783761394,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.6452,
|
|
"mean_token_accuracy": 0.11003177165985108,
|
|
"num_tokens": 1010445.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.970145893096924,
|
|
"epoch": 0.45981772990886494,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.6866,
|
|
"mean_token_accuracy": 0.10296825990080834,
|
|
"num_tokens": 1019084.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.00351128578186,
|
|
"epoch": 0.463960231980116,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6045,
|
|
"mean_token_accuracy": 0.10343219712376595,
|
|
"num_tokens": 1027263.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.118851375579834,
|
|
"epoch": 0.46810273405136704,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7038,
|
|
"mean_token_accuracy": 0.10047925561666489,
|
|
"num_tokens": 1036275.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.024382400512695,
|
|
"epoch": 0.47224523612261804,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7695,
|
|
"mean_token_accuracy": 0.09930935353040696,
|
|
"num_tokens": 1045208.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.1062685489654545,
|
|
"epoch": 0.4763877381938691,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.692,
|
|
"mean_token_accuracy": 0.10427912548184395,
|
|
"num_tokens": 1054703.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.880273103713989,
|
|
"epoch": 0.48053024026512015,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.5487,
|
|
"mean_token_accuracy": 0.1035312443971634,
|
|
"num_tokens": 1063652.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.948973989486694,
|
|
"epoch": 0.48467274233637114,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6483,
|
|
"mean_token_accuracy": 0.10299907624721527,
|
|
"num_tokens": 1072070.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.205907678604126,
|
|
"epoch": 0.4888152444076222,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6786,
|
|
"mean_token_accuracy": 0.10379905477166176,
|
|
"num_tokens": 1079992.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.94438328742981,
|
|
"epoch": 0.49295774647887325,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6762,
|
|
"mean_token_accuracy": 0.10331593230366706,
|
|
"num_tokens": 1088862.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.976015233993531,
|
|
"epoch": 0.4971002485501243,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.4811,
|
|
"mean_token_accuracy": 0.11253077536821365,
|
|
"num_tokens": 1097614.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.891912937164307,
|
|
"epoch": 0.5012427506213754,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6811,
|
|
"mean_token_accuracy": 0.1095806010067463,
|
|
"num_tokens": 1107347.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.113644981384278,
|
|
"epoch": 0.5053852526926264,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7093,
|
|
"mean_token_accuracy": 0.10049239173531532,
|
|
"num_tokens": 1116707.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.003044986724854,
|
|
"epoch": 0.5095277547638773,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7031,
|
|
"mean_token_accuracy": 0.1056403562426567,
|
|
"num_tokens": 1126356.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.974148845672607,
|
|
"epoch": 0.5136702568351285,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6467,
|
|
"mean_token_accuracy": 0.10070049464702606,
|
|
"num_tokens": 1136281.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.98663740158081,
|
|
"epoch": 0.5178127589063795,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.6011,
|
|
"mean_token_accuracy": 0.10273146480321885,
|
|
"num_tokens": 1145466.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.933039379119873,
|
|
"epoch": 0.5219552609776305,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.6061,
|
|
"mean_token_accuracy": 0.10639488324522972,
|
|
"num_tokens": 1154225.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.023647594451904,
|
|
"epoch": 0.5260977630488816,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6404,
|
|
"mean_token_accuracy": 0.10065171793103218,
|
|
"num_tokens": 1163872.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.9506641864776615,
|
|
"epoch": 0.5302402651201326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.5717,
|
|
"mean_token_accuracy": 0.11115986928343773,
|
|
"num_tokens": 1172695.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.853932952880859,
|
|
"epoch": 0.5343827671913836,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.4515,
|
|
"mean_token_accuracy": 0.1125042974948883,
|
|
"num_tokens": 1181291.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.001178407669068,
|
|
"epoch": 0.5385252692626347,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.10361743718385696,
|
|
"num_tokens": 1189927.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.9009294509887695,
|
|
"epoch": 0.5426677713338857,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.5567,
|
|
"mean_token_accuracy": 0.11111297979950904,
|
|
"num_tokens": 1197913.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.887521409988404,
|
|
"epoch": 0.5468102734051367,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4824,
|
|
"mean_token_accuracy": 0.11478215083479881,
|
|
"num_tokens": 1207032.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.917655086517334,
|
|
"epoch": 0.5509527754763878,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5899,
|
|
"mean_token_accuracy": 0.10642103552818298,
|
|
"num_tokens": 1215710.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.945044994354248,
|
|
"epoch": 0.5550952775476388,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.5707,
|
|
"mean_token_accuracy": 0.10297949239611626,
|
|
"num_tokens": 1225619.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.85495548248291,
|
|
"epoch": 0.5592377796188898,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.5078,
|
|
"mean_token_accuracy": 0.10624194070696831,
|
|
"num_tokens": 1235000.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.8498025894165036,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5083,
|
|
"mean_token_accuracy": 0.10908133909106255,
|
|
"num_tokens": 1243594.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.964477157592773,
|
|
"epoch": 0.5675227837613919,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.7039,
|
|
"mean_token_accuracy": 0.10716133862733841,
|
|
"num_tokens": 1252279.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.074165916442871,
|
|
"epoch": 0.5716652858326429,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6102,
|
|
"mean_token_accuracy": 0.10797887593507767,
|
|
"num_tokens": 1261314.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.786513042449951,
|
|
"epoch": 0.575807787903894,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.489,
|
|
"mean_token_accuracy": 0.11091019809246064,
|
|
"num_tokens": 1270854.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.820779418945312,
|
|
"epoch": 0.579950289975145,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.459,
|
|
"mean_token_accuracy": 0.11137820258736611,
|
|
"num_tokens": 1279183.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.864804220199585,
|
|
"epoch": 0.584092792046396,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5622,
|
|
"mean_token_accuracy": 0.10689144134521485,
|
|
"num_tokens": 1288347.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.890985202789307,
|
|
"epoch": 0.5882352941176471,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5705,
|
|
"mean_token_accuracy": 0.10670721456408501,
|
|
"num_tokens": 1297449.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.991053485870362,
|
|
"epoch": 0.5923777961888981,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5221,
|
|
"mean_token_accuracy": 0.10960500165820122,
|
|
"num_tokens": 1306232.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.839966917037964,
|
|
"epoch": 0.5965202982601492,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.5397,
|
|
"mean_token_accuracy": 0.11476544663310051,
|
|
"num_tokens": 1315073.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.873430061340332,
|
|
"epoch": 0.6006628003314002,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.5763,
|
|
"mean_token_accuracy": 0.10628646165132523,
|
|
"num_tokens": 1324629.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.988832902908325,
|
|
"epoch": 0.6048053024026512,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6889,
|
|
"mean_token_accuracy": 0.10445819571614265,
|
|
"num_tokens": 1334152.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.004844760894775,
|
|
"epoch": 0.6089478044739023,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.649,
|
|
"mean_token_accuracy": 0.10557386428117752,
|
|
"num_tokens": 1343063.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.872599458694458,
|
|
"epoch": 0.6130903065451533,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.5733,
|
|
"mean_token_accuracy": 0.10616720467805862,
|
|
"num_tokens": 1352632.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.796925640106201,
|
|
"epoch": 0.6172328086164043,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.5249,
|
|
"mean_token_accuracy": 0.10955940410494805,
|
|
"num_tokens": 1361103.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.843691015243531,
|
|
"epoch": 0.6213753106876554,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.5141,
|
|
"mean_token_accuracy": 0.11272492855787278,
|
|
"num_tokens": 1370658.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.910959482192993,
|
|
"epoch": 0.6255178127589064,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5224,
|
|
"mean_token_accuracy": 0.1089279018342495,
|
|
"num_tokens": 1379817.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.854007911682129,
|
|
"epoch": 0.6296603148301574,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.4741,
|
|
"mean_token_accuracy": 0.11225551590323449,
|
|
"num_tokens": 1389329.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.684650230407715,
|
|
"epoch": 0.6338028169014085,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.5244,
|
|
"mean_token_accuracy": 0.1060152791440487,
|
|
"num_tokens": 1399279.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.8729455947875975,
|
|
"epoch": 0.6379453189726595,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.4704,
|
|
"mean_token_accuracy": 0.11294344514608383,
|
|
"num_tokens": 1408765.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.890125322341919,
|
|
"epoch": 0.6420878210439105,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.6028,
|
|
"mean_token_accuracy": 0.1117846891283989,
|
|
"num_tokens": 1418615.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.825809049606323,
|
|
"epoch": 0.6462303231151616,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5069,
|
|
"mean_token_accuracy": 0.10702720955014229,
|
|
"num_tokens": 1427359.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.951959133148193,
|
|
"epoch": 0.6503728251864126,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6603,
|
|
"mean_token_accuracy": 0.10913564860820771,
|
|
"num_tokens": 1435516.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.917226839065552,
|
|
"epoch": 0.6545153272576636,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.6098,
|
|
"mean_token_accuracy": 0.10673258900642395,
|
|
"num_tokens": 1444438.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.90554518699646,
|
|
"epoch": 0.6586578293289147,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5247,
|
|
"mean_token_accuracy": 0.10866252556443215,
|
|
"num_tokens": 1453740.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.866510057449341,
|
|
"epoch": 0.6628003314001657,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5261,
|
|
"mean_token_accuracy": 0.11231514811515808,
|
|
"num_tokens": 1462083.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.730630588531494,
|
|
"epoch": 0.6669428334714167,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4459,
|
|
"mean_token_accuracy": 0.11757130473852158,
|
|
"num_tokens": 1471459.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.8137424945831295,
|
|
"epoch": 0.6710853355426678,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.4358,
|
|
"mean_token_accuracy": 0.11511573269963264,
|
|
"num_tokens": 1480516.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.697441339492798,
|
|
"epoch": 0.6752278376139188,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.3422,
|
|
"mean_token_accuracy": 0.1162965439260006,
|
|
"num_tokens": 1489268.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.893553256988525,
|
|
"epoch": 0.6793703396851698,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5562,
|
|
"mean_token_accuracy": 0.10877497121691704,
|
|
"num_tokens": 1498819.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.8146172046661375,
|
|
"epoch": 0.6835128417564209,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5384,
|
|
"mean_token_accuracy": 0.1133031778037548,
|
|
"num_tokens": 1507852.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.823567295074463,
|
|
"epoch": 0.6876553438276719,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.5635,
|
|
"mean_token_accuracy": 0.1100037857890129,
|
|
"num_tokens": 1517538.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.773430204391479,
|
|
"epoch": 0.6917978458989229,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.4425,
|
|
"mean_token_accuracy": 0.11734201833605766,
|
|
"num_tokens": 1525389.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.780531215667724,
|
|
"epoch": 0.695940347970174,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.4279,
|
|
"mean_token_accuracy": 0.1151418298482895,
|
|
"num_tokens": 1534117.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.773794269561767,
|
|
"epoch": 0.700082850041425,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4497,
|
|
"mean_token_accuracy": 0.11154624596238136,
|
|
"num_tokens": 1542610.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.658645105361939,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.433,
|
|
"mean_token_accuracy": 0.1151619479060173,
|
|
"num_tokens": 1551771.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.913496255874634,
|
|
"epoch": 0.7083678541839271,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.6156,
|
|
"mean_token_accuracy": 0.10468344464898109,
|
|
"num_tokens": 1560231.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.773072385787964,
|
|
"epoch": 0.7125103562551781,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5598,
|
|
"mean_token_accuracy": 0.10812477022409439,
|
|
"num_tokens": 1569578.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.792055749893189,
|
|
"epoch": 0.7166528583264291,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4507,
|
|
"mean_token_accuracy": 0.11864269822835923,
|
|
"num_tokens": 1579308.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.788659477233887,
|
|
"epoch": 0.7207953603976802,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.10858385637402534,
|
|
"num_tokens": 1589506.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.7048583984375,
|
|
"epoch": 0.7249378624689312,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.4141,
|
|
"mean_token_accuracy": 0.11191320866346359,
|
|
"num_tokens": 1598127.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.74589695930481,
|
|
"epoch": 0.7290803645401823,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4341,
|
|
"mean_token_accuracy": 0.1158210128545761,
|
|
"num_tokens": 1608073.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.75485463142395,
|
|
"epoch": 0.7332228666114333,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.11686542183160782,
|
|
"num_tokens": 1616272.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.78275499343872,
|
|
"epoch": 0.7373653686826843,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4752,
|
|
"mean_token_accuracy": 0.11307700201869011,
|
|
"num_tokens": 1625577.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.852141284942627,
|
|
"epoch": 0.7415078707539354,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5666,
|
|
"mean_token_accuracy": 0.10551101639866829,
|
|
"num_tokens": 1634902.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.698729467391968,
|
|
"epoch": 0.7456503728251864,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.4872,
|
|
"mean_token_accuracy": 0.11352990567684174,
|
|
"num_tokens": 1644131.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.733800840377808,
|
|
"epoch": 0.7497928748964374,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3909,
|
|
"mean_token_accuracy": 0.12015661895275116,
|
|
"num_tokens": 1652272.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.715107011795044,
|
|
"epoch": 0.7539353769676885,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4837,
|
|
"mean_token_accuracy": 0.11517259702086449,
|
|
"num_tokens": 1660962.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.6873085498809814,
|
|
"epoch": 0.7580778790389395,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.419,
|
|
"mean_token_accuracy": 0.11443299949169158,
|
|
"num_tokens": 1670570.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.761066484451294,
|
|
"epoch": 0.7622203811101905,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4911,
|
|
"mean_token_accuracy": 0.11005624011158943,
|
|
"num_tokens": 1679674.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.826864004135132,
|
|
"epoch": 0.7663628831814416,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4624,
|
|
"mean_token_accuracy": 0.11269745901226998,
|
|
"num_tokens": 1688849.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.581996488571167,
|
|
"epoch": 0.7705053852526926,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.3407,
|
|
"mean_token_accuracy": 0.11457516700029373,
|
|
"num_tokens": 1698565.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.744958591461182,
|
|
"epoch": 0.7746478873239436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4315,
|
|
"mean_token_accuracy": 0.12229297310113907,
|
|
"num_tokens": 1708009.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.778655099868774,
|
|
"epoch": 0.7787903893951947,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4339,
|
|
"mean_token_accuracy": 0.1131053164601326,
|
|
"num_tokens": 1716497.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.6028002262115475,
|
|
"epoch": 0.7829328914664457,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4305,
|
|
"mean_token_accuracy": 0.11531242653727532,
|
|
"num_tokens": 1725609.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.682197093963623,
|
|
"epoch": 0.7870753935376967,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.353,
|
|
"mean_token_accuracy": 0.12001026049256325,
|
|
"num_tokens": 1734165.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.697330379486084,
|
|
"epoch": 0.7912178956089478,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.4756,
|
|
"mean_token_accuracy": 0.10846871361136437,
|
|
"num_tokens": 1744856.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.672932958602905,
|
|
"epoch": 0.7953603976801988,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.4727,
|
|
"mean_token_accuracy": 0.10662268623709678,
|
|
"num_tokens": 1753386.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.7072113990783695,
|
|
"epoch": 0.7995028997514498,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3358,
|
|
"mean_token_accuracy": 0.11681910082697869,
|
|
"num_tokens": 1762521.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.579640007019043,
|
|
"epoch": 0.8036454018227009,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3898,
|
|
"mean_token_accuracy": 0.11789416447281838,
|
|
"num_tokens": 1772403.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.674222660064697,
|
|
"epoch": 0.8077879038939519,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.3007,
|
|
"mean_token_accuracy": 0.11631953716278076,
|
|
"num_tokens": 1780924.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.698087358474732,
|
|
"epoch": 0.8119304059652029,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4432,
|
|
"mean_token_accuracy": 0.11354917660355568,
|
|
"num_tokens": 1789992.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.653201007843018,
|
|
"epoch": 0.816072908036454,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3959,
|
|
"mean_token_accuracy": 0.11834764555096626,
|
|
"num_tokens": 1799513.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.569394779205322,
|
|
"epoch": 0.820215410107705,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.261,
|
|
"mean_token_accuracy": 0.11697232872247695,
|
|
"num_tokens": 1807393.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.676162242889404,
|
|
"epoch": 0.824357912178956,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.5097,
|
|
"mean_token_accuracy": 0.10958585515618324,
|
|
"num_tokens": 1817284.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.682469367980957,
|
|
"epoch": 0.8285004142502072,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2843,
|
|
"mean_token_accuracy": 0.11956144273281097,
|
|
"num_tokens": 1825591.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8285004142502072,
|
|
"eval_entropy": 6.44553795653874,
|
|
"eval_loss": 6.471545696258545,
|
|
"eval_mean_token_accuracy": 0.11671354705845115,
|
|
"eval_num_tokens": 1825591.0,
|
|
"eval_runtime": 2.123,
|
|
"eval_samples_per_second": 1764.978,
|
|
"eval_steps_per_second": 220.917,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.6662139892578125,
|
|
"epoch": 0.8326429163214581,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.0004999998547681232,
|
|
"loss": 6.4046,
|
|
"mean_token_accuracy": 0.12007165029644966,
|
|
"num_tokens": 1834655.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.467186403274536,
|
|
"epoch": 0.8367854183927091,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999992647639447,
|
|
"loss": 6.2657,
|
|
"mean_token_accuracy": 0.12069680243730545,
|
|
"num_tokens": 1844187.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.612904119491577,
|
|
"epoch": 0.8409279204639603,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999982209116618,
|
|
"loss": 6.3449,
|
|
"mean_token_accuracy": 0.11403246521949768,
|
|
"num_tokens": 1852768.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.624643278121948,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004999967232133796,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.12195011600852013,
|
|
"num_tokens": 1862095.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.616990756988526,
|
|
"epoch": 0.8492129246064622,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999947716721196,
|
|
"loss": 6.422,
|
|
"mean_token_accuracy": 0.11276146695017815,
|
|
"num_tokens": 1871239.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.637329816818237,
|
|
"epoch": 0.8533554266777134,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999923662918179,
|
|
"loss": 6.4264,
|
|
"mean_token_accuracy": 0.11726627275347709,
|
|
"num_tokens": 1880065.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.669142055511474,
|
|
"epoch": 0.8574979287489644,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999895070773266,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.11147499307990075,
|
|
"num_tokens": 1889065.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.697521781921386,
|
|
"epoch": 0.8616404308202155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999861940344129,
|
|
"loss": 6.4571,
|
|
"mean_token_accuracy": 0.11582566872239113,
|
|
"num_tokens": 1898238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.651687860488892,
|
|
"epoch": 0.8657829328914665,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999824271697597,
|
|
"loss": 6.3975,
|
|
"mean_token_accuracy": 0.1222484327852726,
|
|
"num_tokens": 1906780.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.710152673721313,
|
|
"epoch": 0.8699254349627175,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499978206490965,
|
|
"loss": 6.4905,
|
|
"mean_token_accuracy": 0.11242052838206291,
|
|
"num_tokens": 1916183.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.546902227401733,
|
|
"epoch": 0.8740679370339686,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999735320065427,
|
|
"loss": 6.3458,
|
|
"mean_token_accuracy": 0.1232548326253891,
|
|
"num_tokens": 1925382.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.567516803741455,
|
|
"epoch": 0.8782104391052196,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999684037259216,
|
|
"loss": 6.2036,
|
|
"mean_token_accuracy": 0.12797800078988075,
|
|
"num_tokens": 1933639.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.524128103256226,
|
|
"epoch": 0.8823529411764706,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004999628216594456,
|
|
"loss": 6.4724,
|
|
"mean_token_accuracy": 0.11707659810781479,
|
|
"num_tokens": 1943044.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.655896615982056,
|
|
"epoch": 0.8864954432477217,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000499956785818375,
|
|
"loss": 6.3217,
|
|
"mean_token_accuracy": 0.11736537888646126,
|
|
"num_tokens": 1951777.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.527218246459961,
|
|
"epoch": 0.8906379453189727,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999502962148843,
|
|
"loss": 6.3884,
|
|
"mean_token_accuracy": 0.11941609233617782,
|
|
"num_tokens": 1961356.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.738359689712524,
|
|
"epoch": 0.8947804473902237,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000499943352862064,
|
|
"loss": 6.4322,
|
|
"mean_token_accuracy": 0.11732709258794785,
|
|
"num_tokens": 1970341.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.593241596221924,
|
|
"epoch": 0.8989229494614748,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999359557739194,
|
|
"loss": 6.2907,
|
|
"mean_token_accuracy": 0.11967678442597389,
|
|
"num_tokens": 1978576.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.591588544845581,
|
|
"epoch": 0.9030654515327258,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999281049653713,
|
|
"loss": 6.3461,
|
|
"mean_token_accuracy": 0.11938407868146897,
|
|
"num_tokens": 1987229.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.569654750823974,
|
|
"epoch": 0.9072079536039768,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999198004522557,
|
|
"loss": 6.26,
|
|
"mean_token_accuracy": 0.12861526757478714,
|
|
"num_tokens": 1996473.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.511166715621949,
|
|
"epoch": 0.9113504556752279,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999110422513237,
|
|
"loss": 6.2915,
|
|
"mean_token_accuracy": 0.12253913059830665,
|
|
"num_tokens": 2005551.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.5116040229797365,
|
|
"epoch": 0.9154929577464789,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999018303802414,
|
|
"loss": 6.3972,
|
|
"mean_token_accuracy": 0.12215510085225105,
|
|
"num_tokens": 2015012.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.654548215866089,
|
|
"epoch": 0.9196354598177299,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998921648575904,
|
|
"loss": 6.3983,
|
|
"mean_token_accuracy": 0.11905976086854934,
|
|
"num_tokens": 2024435.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.544571256637573,
|
|
"epoch": 0.923777961888981,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998820457028669,
|
|
"loss": 6.3065,
|
|
"mean_token_accuracy": 0.12046469524502754,
|
|
"num_tokens": 2033991.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.558928680419922,
|
|
"epoch": 0.927920463960232,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998714729364824,
|
|
"loss": 6.3193,
|
|
"mean_token_accuracy": 0.1266247771680355,
|
|
"num_tokens": 2043117.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.59407753944397,
|
|
"epoch": 0.932062966031483,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998604465797634,
|
|
"loss": 6.3396,
|
|
"mean_token_accuracy": 0.12249568551778793,
|
|
"num_tokens": 2052304.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.464563369750977,
|
|
"epoch": 0.9362054681027341,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004998489666549512,
|
|
"loss": 6.2867,
|
|
"mean_token_accuracy": 0.12436222359538078,
|
|
"num_tokens": 2061217.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.460366916656494,
|
|
"epoch": 0.9403479701739851,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998370331852021,
|
|
"loss": 6.2685,
|
|
"mean_token_accuracy": 0.1200811967253685,
|
|
"num_tokens": 2070796.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.664331769943237,
|
|
"epoch": 0.9444904722452361,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998246461945872,
|
|
"loss": 6.3871,
|
|
"mean_token_accuracy": 0.12616615518927574,
|
|
"num_tokens": 2080202.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.546797418594361,
|
|
"epoch": 0.9486329743164872,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998118057080923,
|
|
"loss": 6.3135,
|
|
"mean_token_accuracy": 0.121537134796381,
|
|
"num_tokens": 2089788.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.597958517074585,
|
|
"epoch": 0.9527754763877382,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997985117516183,
|
|
"loss": 6.3355,
|
|
"mean_token_accuracy": 0.11748650595545769,
|
|
"num_tokens": 2098727.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.42893352508545,
|
|
"epoch": 0.9569179784589892,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997847643519803,
|
|
"loss": 6.2309,
|
|
"mean_token_accuracy": 0.11966113150119781,
|
|
"num_tokens": 2107645.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.408219528198242,
|
|
"epoch": 0.9610604805302403,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997705635369085,
|
|
"loss": 6.2108,
|
|
"mean_token_accuracy": 0.12341454103589058,
|
|
"num_tokens": 2116613.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.519371175765992,
|
|
"epoch": 0.9652029826014913,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004997559093350474,
|
|
"loss": 6.2133,
|
|
"mean_token_accuracy": 0.12186720743775367,
|
|
"num_tokens": 2125585.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.568776416778564,
|
|
"epoch": 0.9693454846727423,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997408017759562,
|
|
"loss": 6.3294,
|
|
"mean_token_accuracy": 0.12122908607125282,
|
|
"num_tokens": 2134795.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.521303701400757,
|
|
"epoch": 0.9734879867439934,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004997252408901084,
|
|
"loss": 6.3297,
|
|
"mean_token_accuracy": 0.11886916756629944,
|
|
"num_tokens": 2143699.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.602518701553345,
|
|
"epoch": 0.9776304888152444,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004997092267088922,
|
|
"loss": 6.2718,
|
|
"mean_token_accuracy": 0.12022883147001266,
|
|
"num_tokens": 2152187.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.49890923500061,
|
|
"epoch": 0.9817729908864954,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004996927592646097,
|
|
"loss": 6.3403,
|
|
"mean_token_accuracy": 0.12049522399902343,
|
|
"num_tokens": 2160941.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.568722772598266,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996758385904776,
|
|
"loss": 6.3268,
|
|
"mean_token_accuracy": 0.12370333299040795,
|
|
"num_tokens": 2168946.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.424992275238037,
|
|
"epoch": 0.9900579950289975,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996584647206269,
|
|
"loss": 6.234,
|
|
"mean_token_accuracy": 0.12237700819969177,
|
|
"num_tokens": 2177662.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.488782596588135,
|
|
"epoch": 0.9942004971002486,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996406376901026,
|
|
"loss": 6.3339,
|
|
"mean_token_accuracy": 0.1301935538649559,
|
|
"num_tokens": 2187461.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.459936046600342,
|
|
"epoch": 0.9983429991714996,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004996223575348635,
|
|
"loss": 6.2143,
|
|
"mean_token_accuracy": 0.1307205244898796,
|
|
"num_tokens": 2196878.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.549283266067505,
|
|
"epoch": 1.0024855012427507,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996036242917828,
|
|
"loss": 6.2226,
|
|
"mean_token_accuracy": 0.11891174912452698,
|
|
"num_tokens": 2206178.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.446871328353882,
|
|
"epoch": 1.0066280033140016,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995844379986476,
|
|
"loss": 6.0874,
|
|
"mean_token_accuracy": 0.12653569504618645,
|
|
"num_tokens": 2215330.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.492379474639892,
|
|
"epoch": 1.0107705053852527,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995647986941587,
|
|
"loss": 6.0875,
|
|
"mean_token_accuracy": 0.12935211285948753,
|
|
"num_tokens": 2224146.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.480843734741211,
|
|
"epoch": 1.0149130074565038,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995447064179308,
|
|
"loss": 6.0677,
|
|
"mean_token_accuracy": 0.13676073402166367,
|
|
"num_tokens": 2233646.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.290753316879273,
|
|
"epoch": 1.0190555095277547,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000499524161210492,
|
|
"loss": 5.9451,
|
|
"mean_token_accuracy": 0.13070207014679908,
|
|
"num_tokens": 2243663.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.452328014373779,
|
|
"epoch": 1.0231980115990058,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995031631132844,
|
|
"loss": 6.1289,
|
|
"mean_token_accuracy": 0.13538116291165353,
|
|
"num_tokens": 2252564.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.454355192184448,
|
|
"epoch": 1.027340513670257,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004994817121686633,
|
|
"loss": 6.1202,
|
|
"mean_token_accuracy": 0.12975577935576438,
|
|
"num_tokens": 2262504.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.441025686264038,
|
|
"epoch": 1.0314830157415078,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004994598084198979,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.1423887699842453,
|
|
"num_tokens": 2270553.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.357677888870239,
|
|
"epoch": 1.035625517812759,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004994374519111701,
|
|
"loss": 6.0331,
|
|
"mean_token_accuracy": 0.13042610734701157,
|
|
"num_tokens": 2279319.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.440127897262573,
|
|
"epoch": 1.03976801988401,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994146426875757,
|
|
"loss": 6.1553,
|
|
"mean_token_accuracy": 0.1311718061566353,
|
|
"num_tokens": 2288749.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.413478183746338,
|
|
"epoch": 1.043910521955261,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004993913807951232,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.13126998767256737,
|
|
"num_tokens": 2298272.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.417128038406372,
|
|
"epoch": 1.048053024026512,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993676662807345,
|
|
"loss": 6.0957,
|
|
"mean_token_accuracy": 0.1319083034992218,
|
|
"num_tokens": 2307230.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.449226379394531,
|
|
"epoch": 1.0521955260977631,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004993434991922443,
|
|
"loss": 6.1669,
|
|
"mean_token_accuracy": 0.12653778642416,
|
|
"num_tokens": 2316306.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.432471418380738,
|
|
"epoch": 1.056338028169014,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004993188795784002,
|
|
"loss": 6.0938,
|
|
"mean_token_accuracy": 0.12626147791743278,
|
|
"num_tokens": 2325672.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.298291969299316,
|
|
"epoch": 1.0604805302402651,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004992938074888628,
|
|
"loss": 6.0955,
|
|
"mean_token_accuracy": 0.13139547854661943,
|
|
"num_tokens": 2335251.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.4482035636901855,
|
|
"epoch": 1.0646230323115162,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004992682829742052,
|
|
"loss": 6.0507,
|
|
"mean_token_accuracy": 0.1331412188708782,
|
|
"num_tokens": 2344054.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.267281532287598,
|
|
"epoch": 1.0687655343827671,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004992423060859132,
|
|
"loss": 6.0521,
|
|
"mean_token_accuracy": 0.1336584560573101,
|
|
"num_tokens": 2353493.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.503845310211181,
|
|
"epoch": 1.0729080364540182,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004992158768763849,
|
|
"loss": 6.158,
|
|
"mean_token_accuracy": 0.1326499328017235,
|
|
"num_tokens": 2362687.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.422919511795044,
|
|
"epoch": 1.0770505385252693,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991889953989308,
|
|
"loss": 6.1223,
|
|
"mean_token_accuracy": 0.12865589410066605,
|
|
"num_tokens": 2372325.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.428432607650757,
|
|
"epoch": 1.0811930405965202,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004991616617077741,
|
|
"loss": 6.1441,
|
|
"mean_token_accuracy": 0.13311094492673875,
|
|
"num_tokens": 2381987.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.520003366470337,
|
|
"epoch": 1.0853355426677713,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004991338758580496,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.13536124899983407,
|
|
"num_tokens": 2390901.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.266488838195801,
|
|
"epoch": 1.0894780447390224,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004991056379058045,
|
|
"loss": 6.1041,
|
|
"mean_token_accuracy": 0.12899856343865396,
|
|
"num_tokens": 2400756.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.308699131011963,
|
|
"epoch": 1.0936205468102733,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004990769479079976,
|
|
"loss": 5.9385,
|
|
"mean_token_accuracy": 0.13213619813323021,
|
|
"num_tokens": 2409526.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.35676589012146,
|
|
"epoch": 1.0977630488815244,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004990478059225003,
|
|
"loss": 6.1008,
|
|
"mean_token_accuracy": 0.1294182427227497,
|
|
"num_tokens": 2419583.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.438393783569336,
|
|
"epoch": 1.1019055509527755,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004990182120080945,
|
|
"loss": 6.0619,
|
|
"mean_token_accuracy": 0.13824906051158906,
|
|
"num_tokens": 2429439.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.262748527526855,
|
|
"epoch": 1.1060480530240264,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989881662244748,
|
|
"loss": 6.0582,
|
|
"mean_token_accuracy": 0.1282345525920391,
|
|
"num_tokens": 2438467.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.38105845451355,
|
|
"epoch": 1.1101905550952775,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004989576686322468,
|
|
"loss": 5.9982,
|
|
"mean_token_accuracy": 0.1415070377290249,
|
|
"num_tokens": 2446619.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.294713640213013,
|
|
"epoch": 1.1143330571665286,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004989267192929272,
|
|
"loss": 6.0653,
|
|
"mean_token_accuracy": 0.13668779358267785,
|
|
"num_tokens": 2454998.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.358230304718018,
|
|
"epoch": 1.1184755592377795,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004988953182689445,
|
|
"loss": 6.0325,
|
|
"mean_token_accuracy": 0.13635167926549913,
|
|
"num_tokens": 2463497.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.3403887271881105,
|
|
"epoch": 1.1226180613090306,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988634656236378,
|
|
"loss": 5.9655,
|
|
"mean_token_accuracy": 0.13428828567266465,
|
|
"num_tokens": 2471763.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.300700998306274,
|
|
"epoch": 1.1267605633802817,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004988311614212576,
|
|
"loss": 6.0232,
|
|
"mean_token_accuracy": 0.13982509970664977,
|
|
"num_tokens": 2479949.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.378063678741455,
|
|
"epoch": 1.1309030654515326,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987984057269648,
|
|
"loss": 6.0715,
|
|
"mean_token_accuracy": 0.13535686433315278,
|
|
"num_tokens": 2489097.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.149481630325317,
|
|
"epoch": 1.1350455675227837,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004987651986068311,
|
|
"loss": 5.9023,
|
|
"mean_token_accuracy": 0.13939108699560165,
|
|
"num_tokens": 2497625.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.340058994293213,
|
|
"epoch": 1.1391880695940348,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004987315401278394,
|
|
"loss": 6.1463,
|
|
"mean_token_accuracy": 0.12731209993362427,
|
|
"num_tokens": 2507107.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.482500743865967,
|
|
"epoch": 1.143330571665286,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004986974303578819,
|
|
"loss": 6.1447,
|
|
"mean_token_accuracy": 0.12803158536553383,
|
|
"num_tokens": 2516629.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.342242622375489,
|
|
"epoch": 1.1474730737365368,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004986628693657621,
|
|
"loss": 6.0298,
|
|
"mean_token_accuracy": 0.1318236358463764,
|
|
"num_tokens": 2525924.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.442188549041748,
|
|
"epoch": 1.151615575807788,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004986278572211933,
|
|
"loss": 6.1865,
|
|
"mean_token_accuracy": 0.1275293305516243,
|
|
"num_tokens": 2535586.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.4357420921325685,
|
|
"epoch": 1.1557580778790388,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004985923939947987,
|
|
"loss": 6.2509,
|
|
"mean_token_accuracy": 0.12767949104309081,
|
|
"num_tokens": 2546685.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.281386995315552,
|
|
"epoch": 1.15990057995029,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004985564797581115,
|
|
"loss": 5.9787,
|
|
"mean_token_accuracy": 0.1356250524520874,
|
|
"num_tokens": 2556873.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.330170154571533,
|
|
"epoch": 1.164043082021541,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004985201145835747,
|
|
"loss": 6.0158,
|
|
"mean_token_accuracy": 0.13851649388670922,
|
|
"num_tokens": 2565222.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.281029844284058,
|
|
"epoch": 1.1681855840927922,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000498483298544541,
|
|
"loss": 6.0105,
|
|
"mean_token_accuracy": 0.1319947473704815,
|
|
"num_tokens": 2575467.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.319263124465943,
|
|
"epoch": 1.172328086164043,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000498446031715272,
|
|
"loss": 6.0273,
|
|
"mean_token_accuracy": 0.13361441418528558,
|
|
"num_tokens": 2583963.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.354074096679687,
|
|
"epoch": 1.1764705882352942,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984083141709391,
|
|
"loss": 6.0681,
|
|
"mean_token_accuracy": 0.1338679425418377,
|
|
"num_tokens": 2593350.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.172420501708984,
|
|
"epoch": 1.180613090306545,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000498370145987623,
|
|
"loss": 5.9719,
|
|
"mean_token_accuracy": 0.1350063219666481,
|
|
"num_tokens": 2602191.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.360668754577636,
|
|
"epoch": 1.1847555923777962,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004983315272423127,
|
|
"loss": 6.1037,
|
|
"mean_token_accuracy": 0.1312294103205204,
|
|
"num_tokens": 2613590.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.241883325576782,
|
|
"epoch": 1.1888980944490473,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004982924580129066,
|
|
"loss": 5.9199,
|
|
"mean_token_accuracy": 0.13844091519713403,
|
|
"num_tokens": 2622072.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.380551862716675,
|
|
"epoch": 1.1930405965202984,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004982529383782116,
|
|
"loss": 6.1199,
|
|
"mean_token_accuracy": 0.1306464396417141,
|
|
"num_tokens": 2630904.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.325322961807251,
|
|
"epoch": 1.1971830985915493,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004982129684179432,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.13483623936772346,
|
|
"num_tokens": 2639925.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.305817794799805,
|
|
"epoch": 1.2013256006628004,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004981725482127251,
|
|
"loss": 5.9931,
|
|
"mean_token_accuracy": 0.13657744973897934,
|
|
"num_tokens": 2649397.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.366173505783081,
|
|
"epoch": 1.2054681027340513,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004981316778440892,
|
|
"loss": 6.1291,
|
|
"mean_token_accuracy": 0.1340840056538582,
|
|
"num_tokens": 2658877.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.301034593582154,
|
|
"epoch": 1.2096106048053024,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004980903573944756,
|
|
"loss": 6.0426,
|
|
"mean_token_accuracy": 0.13118910491466523,
|
|
"num_tokens": 2668518.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.283834028244018,
|
|
"epoch": 1.2137531068765535,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004980485869472322,
|
|
"loss": 6.1132,
|
|
"mean_token_accuracy": 0.13253218531608582,
|
|
"num_tokens": 2677459.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.37120361328125,
|
|
"epoch": 1.2178956089478046,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004980063665866143,
|
|
"loss": 6.097,
|
|
"mean_token_accuracy": 0.1286916382610798,
|
|
"num_tokens": 2686668.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.197322845458984,
|
|
"epoch": 1.2220381110190555,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004979636963977853,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.13604906648397447,
|
|
"num_tokens": 2695545.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.325300121307373,
|
|
"epoch": 1.2261806130903066,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004979205764668156,
|
|
"loss": 6.0717,
|
|
"mean_token_accuracy": 0.13842247277498246,
|
|
"num_tokens": 2705514.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.290655946731567,
|
|
"epoch": 1.2303231151615575,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004978770068806825,
|
|
"loss": 5.9642,
|
|
"mean_token_accuracy": 0.13589237332344056,
|
|
"num_tokens": 2714373.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.317065334320068,
|
|
"epoch": 1.2344656172328086,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000497832987727271,
|
|
"loss": 6.0969,
|
|
"mean_token_accuracy": 0.1277540720999241,
|
|
"num_tokens": 2723887.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.267973613739014,
|
|
"epoch": 1.2386081193040597,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004977885190953724,
|
|
"loss": 5.9704,
|
|
"mean_token_accuracy": 0.1360554203391075,
|
|
"num_tokens": 2733591.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.34386134147644,
|
|
"epoch": 1.2427506213753108,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004977436010746846,
|
|
"loss": 6.0837,
|
|
"mean_token_accuracy": 0.137542824447155,
|
|
"num_tokens": 2741638.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2427506213753108,
|
|
"eval_entropy": 6.1450459260676205,
|
|
"eval_loss": 6.2417120933532715,
|
|
"eval_mean_token_accuracy": 0.13248178126143495,
|
|
"eval_num_tokens": 2741638.0,
|
|
"eval_runtime": 2.108,
|
|
"eval_samples_per_second": 1777.534,
|
|
"eval_steps_per_second": 222.488,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.328186321258545,
|
|
"epoch": 1.2468931234465617,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004976982337558124,
|
|
"loss": 5.9465,
|
|
"mean_token_accuracy": 0.1384536549448967,
|
|
"num_tokens": 2749711.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.20745882987976,
|
|
"epoch": 1.2510356255178128,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976524172302665,
|
|
"loss": 6.133,
|
|
"mean_token_accuracy": 0.13232308104634286,
|
|
"num_tokens": 2758121.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.424034690856933,
|
|
"epoch": 1.2551781275890637,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004976061515904641,
|
|
"loss": 6.1357,
|
|
"mean_token_accuracy": 0.12840573713183404,
|
|
"num_tokens": 2766764.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.323179292678833,
|
|
"epoch": 1.2593206296603148,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004975594369297277,
|
|
"loss": 6.0763,
|
|
"mean_token_accuracy": 0.1378533735871315,
|
|
"num_tokens": 2775561.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.318216180801391,
|
|
"epoch": 1.2634631317315659,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004975122733422861,
|
|
"loss": 6.0375,
|
|
"mean_token_accuracy": 0.13417807072401047,
|
|
"num_tokens": 2784691.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.251212882995605,
|
|
"epoch": 1.267605633802817,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004974646609232732,
|
|
"loss": 6.0122,
|
|
"mean_token_accuracy": 0.13721995949745178,
|
|
"num_tokens": 2793893.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.333720350265503,
|
|
"epoch": 1.2717481358740679,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004974165997687288,
|
|
"loss": 6.0845,
|
|
"mean_token_accuracy": 0.13510174974799155,
|
|
"num_tokens": 2804479.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.308886861801147,
|
|
"epoch": 1.275890637945319,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004973680899755972,
|
|
"loss": 6.1243,
|
|
"mean_token_accuracy": 0.13202088996767997,
|
|
"num_tokens": 2813966.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.304367351531982,
|
|
"epoch": 1.2800331400165699,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000497319131641728,
|
|
"loss": 5.9402,
|
|
"mean_token_accuracy": 0.14228401854634284,
|
|
"num_tokens": 2822328.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.229106187820435,
|
|
"epoch": 1.284175642087821,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004972697248658755,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.14200907796621323,
|
|
"num_tokens": 2829845.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.192464447021484,
|
|
"epoch": 1.288318144159072,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004972198697476987,
|
|
"loss": 5.892,
|
|
"mean_token_accuracy": 0.14023082777857782,
|
|
"num_tokens": 2839040.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.137626218795776,
|
|
"epoch": 1.2924606462303232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004971695663877607,
|
|
"loss": 5.9873,
|
|
"mean_token_accuracy": 0.14122241362929344,
|
|
"num_tokens": 2848574.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.311916446685791,
|
|
"epoch": 1.296603148301574,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004971188148875289,
|
|
"loss": 5.9471,
|
|
"mean_token_accuracy": 0.14324187487363815,
|
|
"num_tokens": 2856812.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.200193786621094,
|
|
"epoch": 1.3007456503728252,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004970676153493748,
|
|
"loss": 6.0356,
|
|
"mean_token_accuracy": 0.13674534633755683,
|
|
"num_tokens": 2866001.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.309698963165284,
|
|
"epoch": 1.3048881524440763,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004970159678765731,
|
|
"loss": 6.0376,
|
|
"mean_token_accuracy": 0.1356910966336727,
|
|
"num_tokens": 2875181.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.287196111679077,
|
|
"epoch": 1.3090306545153272,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004969638725733028,
|
|
"loss": 5.9585,
|
|
"mean_token_accuracy": 0.14128673523664476,
|
|
"num_tokens": 2885103.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.100416755676269,
|
|
"epoch": 1.3131731565865783,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004969113295446455,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.1386077396571636,
|
|
"num_tokens": 2894055.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.278606128692627,
|
|
"epoch": 1.3173156586578294,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004968583388965863,
|
|
"loss": 5.9981,
|
|
"mean_token_accuracy": 0.13714988753199578,
|
|
"num_tokens": 2903391.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.237105989456177,
|
|
"epoch": 1.3214581607290803,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004968049007360132,
|
|
"loss": 5.994,
|
|
"mean_token_accuracy": 0.1386870786547661,
|
|
"num_tokens": 2911760.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.175147151947021,
|
|
"epoch": 1.3256006628003314,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004967510151707169,
|
|
"loss": 5.9475,
|
|
"mean_token_accuracy": 0.13915018290281295,
|
|
"num_tokens": 2920589.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.369753313064575,
|
|
"epoch": 1.3297431648715825,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004966966823093903,
|
|
"loss": 6.0644,
|
|
"mean_token_accuracy": 0.13484889045357704,
|
|
"num_tokens": 2929635.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.208001518249512,
|
|
"epoch": 1.3338856669428334,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004966419022616287,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.13991843163967133,
|
|
"num_tokens": 2938778.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.194763469696045,
|
|
"epoch": 1.3380281690140845,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004965866751379296,
|
|
"loss": 6.0578,
|
|
"mean_token_accuracy": 0.13248574435710908,
|
|
"num_tokens": 2947836.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.303940629959106,
|
|
"epoch": 1.3421706710853356,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004965310010496922,
|
|
"loss": 6.0104,
|
|
"mean_token_accuracy": 0.13229761198163031,
|
|
"num_tokens": 2957028.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.224971008300781,
|
|
"epoch": 1.3463131731565865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004964748801092168,
|
|
"loss": 5.9812,
|
|
"mean_token_accuracy": 0.1363674595952034,
|
|
"num_tokens": 2966088.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.262715864181518,
|
|
"epoch": 1.3504556752278376,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004964183124297059,
|
|
"loss": 6.0688,
|
|
"mean_token_accuracy": 0.13448460698127745,
|
|
"num_tokens": 2975271.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.237486219406128,
|
|
"epoch": 1.3545981772990887,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004963612981252625,
|
|
"loss": 6.0718,
|
|
"mean_token_accuracy": 0.13386934548616408,
|
|
"num_tokens": 2983799.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.270224094390869,
|
|
"epoch": 1.3587406793703396,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004963038373108908,
|
|
"loss": 6.0232,
|
|
"mean_token_accuracy": 0.13873186558485032,
|
|
"num_tokens": 2992675.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.297949504852295,
|
|
"epoch": 1.3628831814415907,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004962459301024954,
|
|
"loss": 6.0931,
|
|
"mean_token_accuracy": 0.13615495786070825,
|
|
"num_tokens": 3002483.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.247576522827148,
|
|
"epoch": 1.3670256835128418,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004961875766168816,
|
|
"loss": 5.9923,
|
|
"mean_token_accuracy": 0.14012517556548118,
|
|
"num_tokens": 3010611.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.24715633392334,
|
|
"epoch": 1.371168185584093,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004961287769717547,
|
|
"loss": 5.952,
|
|
"mean_token_accuracy": 0.13905780464410783,
|
|
"num_tokens": 3019928.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.097472667694092,
|
|
"epoch": 1.3753106876553438,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004960695312857199,
|
|
"loss": 5.954,
|
|
"mean_token_accuracy": 0.1395553819835186,
|
|
"num_tokens": 3029003.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.237765502929688,
|
|
"epoch": 1.379453189726595,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004960098396782824,
|
|
"loss": 5.9547,
|
|
"mean_token_accuracy": 0.14094383865594864,
|
|
"num_tokens": 3038049.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.171917200088501,
|
|
"epoch": 1.3835956917978458,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004959497022698466,
|
|
"loss": 5.9515,
|
|
"mean_token_accuracy": 0.14299275428056718,
|
|
"num_tokens": 3046674.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.184727478027344,
|
|
"epoch": 1.387738193869097,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004958891191817163,
|
|
"loss": 5.9909,
|
|
"mean_token_accuracy": 0.13421738520264626,
|
|
"num_tokens": 3056459.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.252535200119018,
|
|
"epoch": 1.391880695940348,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000495828090536094,
|
|
"loss": 5.9889,
|
|
"mean_token_accuracy": 0.13265592753887176,
|
|
"num_tokens": 3066294.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.246398258209228,
|
|
"epoch": 1.3960231980115991,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004957666164560815,
|
|
"loss": 5.9621,
|
|
"mean_token_accuracy": 0.13426307439804078,
|
|
"num_tokens": 3075420.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.238082504272461,
|
|
"epoch": 1.40016570008285,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004957046970656784,
|
|
"loss": 6.0458,
|
|
"mean_token_accuracy": 0.13898480013012887,
|
|
"num_tokens": 3084373.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.207704544067383,
|
|
"epoch": 1.4043082021541011,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004956423324897831,
|
|
"loss": 6.028,
|
|
"mean_token_accuracy": 0.1410148784518242,
|
|
"num_tokens": 3094212.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.219083452224732,
|
|
"epoch": 1.408450704225352,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004955795228541917,
|
|
"loss": 5.9578,
|
|
"mean_token_accuracy": 0.14651229679584504,
|
|
"num_tokens": 3102769.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.253363990783692,
|
|
"epoch": 1.4125932062966031,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000495516268285598,
|
|
"loss": 5.9828,
|
|
"mean_token_accuracy": 0.14103352129459382,
|
|
"num_tokens": 3111465.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.130880308151245,
|
|
"epoch": 1.4167357083678542,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004954525689115935,
|
|
"loss": 5.9386,
|
|
"mean_token_accuracy": 0.14037666097283363,
|
|
"num_tokens": 3121106.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.199966049194336,
|
|
"epoch": 1.4208782104391053,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004953884248606669,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.14885828047990798,
|
|
"num_tokens": 3130662.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.178520822525025,
|
|
"epoch": 1.4250207125103562,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004953238362622034,
|
|
"loss": 5.9958,
|
|
"mean_token_accuracy": 0.13654247894883156,
|
|
"num_tokens": 3139538.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.213479042053223,
|
|
"epoch": 1.4291632145816073,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004952588032464855,
|
|
"loss": 6.0324,
|
|
"mean_token_accuracy": 0.1318473756313324,
|
|
"num_tokens": 3148552.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.1035576343536375,
|
|
"epoch": 1.4333057166528582,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004951933259446919,
|
|
"loss": 5.9302,
|
|
"mean_token_accuracy": 0.14550063461065293,
|
|
"num_tokens": 3156549.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.3432533740997314,
|
|
"epoch": 1.4374482187241093,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004951274044888975,
|
|
"loss": 6.0955,
|
|
"mean_token_accuracy": 0.12662078365683554,
|
|
"num_tokens": 3166821.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.133205366134644,
|
|
"epoch": 1.4415907207953604,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004950610390120731,
|
|
"loss": 5.9973,
|
|
"mean_token_accuracy": 0.1397806391119957,
|
|
"num_tokens": 3174962.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.2480703830719,
|
|
"epoch": 1.4457332228666115,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000494994229648085,
|
|
"loss": 5.9114,
|
|
"mean_token_accuracy": 0.1379687950015068,
|
|
"num_tokens": 3184027.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.124183797836304,
|
|
"epoch": 1.4498757249378624,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004949269765316948,
|
|
"loss": 6.0494,
|
|
"mean_token_accuracy": 0.13002555593848228,
|
|
"num_tokens": 3193186.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.289375114440918,
|
|
"epoch": 1.4540182270091135,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 0.0004948592797985599,
|
|
"loss": 6.0505,
|
|
"mean_token_accuracy": 0.13690062686800958,
|
|
"num_tokens": 3202676.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.324394416809082,
|
|
"epoch": 1.4581607290803644,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004947911395852317,
|
|
"loss": 6.1426,
|
|
"mean_token_accuracy": 0.13963466808199881,
|
|
"num_tokens": 3211950.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.339489889144898,
|
|
"epoch": 1.4623032311516155,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004947225560291565,
|
|
"loss": 6.0873,
|
|
"mean_token_accuracy": 0.13304879441857337,
|
|
"num_tokens": 3221135.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.130894613265991,
|
|
"epoch": 1.4664457332228666,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004946535292686746,
|
|
"loss": 5.8989,
|
|
"mean_token_accuracy": 0.14152208119630813,
|
|
"num_tokens": 3230174.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.11918535232544,
|
|
"epoch": 1.4705882352941178,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004945840594430209,
|
|
"loss": 5.9032,
|
|
"mean_token_accuracy": 0.14170208424329758,
|
|
"num_tokens": 3238898.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.205544185638428,
|
|
"epoch": 1.4747307373653686,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004945141466923235,
|
|
"loss": 5.9335,
|
|
"mean_token_accuracy": 0.14418990314006805,
|
|
"num_tokens": 3247550.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.305442523956299,
|
|
"epoch": 1.4788732394366197,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004944437911576041,
|
|
"loss": 6.0756,
|
|
"mean_token_accuracy": 0.13215108439326287,
|
|
"num_tokens": 3256810.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.062059831619263,
|
|
"epoch": 1.4830157415078706,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004943729929807772,
|
|
"loss": 5.8655,
|
|
"mean_token_accuracy": 0.14079953730106354,
|
|
"num_tokens": 3265886.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.228650426864624,
|
|
"epoch": 1.4871582435791217,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004943017523046507,
|
|
"loss": 5.9409,
|
|
"mean_token_accuracy": 0.14057195484638213,
|
|
"num_tokens": 3274522.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.238694429397583,
|
|
"epoch": 1.4913007456503728,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000494230069272925,
|
|
"loss": 6.0456,
|
|
"mean_token_accuracy": 0.14020049646496774,
|
|
"num_tokens": 3283388.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.1787903785705565,
|
|
"epoch": 1.495443247721624,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004941579440301923,
|
|
"loss": 5.9095,
|
|
"mean_token_accuracy": 0.13298032581806182,
|
|
"num_tokens": 3291887.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.147898387908936,
|
|
"epoch": 1.4995857497928748,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004940853767219372,
|
|
"loss": 5.9886,
|
|
"mean_token_accuracy": 0.13950188905000688,
|
|
"num_tokens": 3300674.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.09293565750122,
|
|
"epoch": 1.503728251864126,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004940123674945359,
|
|
"loss": 5.8255,
|
|
"mean_token_accuracy": 0.14723611921072005,
|
|
"num_tokens": 3309745.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.203557109832763,
|
|
"epoch": 1.5078707539353768,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004939389164952561,
|
|
"loss": 5.979,
|
|
"mean_token_accuracy": 0.14057357907295226,
|
|
"num_tokens": 3318888.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.158121776580811,
|
|
"epoch": 1.512013256006628,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004938650238722564,
|
|
"loss": 5.9485,
|
|
"mean_token_accuracy": 0.14149481058120728,
|
|
"num_tokens": 3328274.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.215224170684815,
|
|
"epoch": 1.516155758077879,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004937906897745861,
|
|
"loss": 5.9563,
|
|
"mean_token_accuracy": 0.14156463742256165,
|
|
"num_tokens": 3337579.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.200045490264893,
|
|
"epoch": 1.5202982601491302,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004937159143521856,
|
|
"loss": 5.937,
|
|
"mean_token_accuracy": 0.14527209475636482,
|
|
"num_tokens": 3346262.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.115064001083374,
|
|
"epoch": 1.5244407622203813,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004936406977558847,
|
|
"loss": 5.9957,
|
|
"mean_token_accuracy": 0.1344020515680313,
|
|
"num_tokens": 3354833.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.233158111572266,
|
|
"epoch": 1.5285832642916322,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004935650401374039,
|
|
"loss": 6.0827,
|
|
"mean_token_accuracy": 0.14519838392734527,
|
|
"num_tokens": 3364314.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.171499109268188,
|
|
"epoch": 1.532725766362883,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004934889416493524,
|
|
"loss": 5.908,
|
|
"mean_token_accuracy": 0.1396942026913166,
|
|
"num_tokens": 3373648.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.06867733001709,
|
|
"epoch": 1.5368682684341342,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004934124024452295,
|
|
"loss": 5.7969,
|
|
"mean_token_accuracy": 0.14711106717586517,
|
|
"num_tokens": 3382919.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.084351205825806,
|
|
"epoch": 1.5410107705053853,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004933354226794232,
|
|
"loss": 5.9393,
|
|
"mean_token_accuracy": 0.13947515338659286,
|
|
"num_tokens": 3391814.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.199561548233032,
|
|
"epoch": 1.5451532725766364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004932580025072098,
|
|
"loss": 5.9788,
|
|
"mean_token_accuracy": 0.1385117821395397,
|
|
"num_tokens": 3400392.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.148109245300293,
|
|
"epoch": 1.5492957746478875,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004931801420847545,
|
|
"loss": 6.0351,
|
|
"mean_token_accuracy": 0.13209574967622756,
|
|
"num_tokens": 3410430.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.122993516921997,
|
|
"epoch": 1.5534382767191384,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004931018415691102,
|
|
"loss": 5.8696,
|
|
"mean_token_accuracy": 0.13951531648635865,
|
|
"num_tokens": 3420654.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.151796579360962,
|
|
"epoch": 1.5575807787903893,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004930231011182176,
|
|
"loss": 6.0094,
|
|
"mean_token_accuracy": 0.13828911855816842,
|
|
"num_tokens": 3430514.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.229997634887695,
|
|
"epoch": 1.5617232808616404,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004929439208909048,
|
|
"loss": 5.9334,
|
|
"mean_token_accuracy": 0.14276001006364822,
|
|
"num_tokens": 3439551.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.119334411621094,
|
|
"epoch": 1.5658657829328915,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004928643010468869,
|
|
"loss": 5.907,
|
|
"mean_token_accuracy": 0.1400480180978775,
|
|
"num_tokens": 3448160.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.153133869171143,
|
|
"epoch": 1.5700082850041426,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004927842417467659,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.14561555981636048,
|
|
"num_tokens": 3457296.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.184989786148071,
|
|
"epoch": 1.5741507870753937,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004927037431520302,
|
|
"loss": 5.9034,
|
|
"mean_token_accuracy": 0.14709929376840591,
|
|
"num_tokens": 3466108.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.082920551300049,
|
|
"epoch": 1.5782932891466446,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004926228054250541,
|
|
"loss": 5.9882,
|
|
"mean_token_accuracy": 0.13886846452951432,
|
|
"num_tokens": 3475336.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.2809975147247314,
|
|
"epoch": 1.5824357912178955,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004925414287290979,
|
|
"loss": 5.9805,
|
|
"mean_token_accuracy": 0.13600327223539352,
|
|
"num_tokens": 3483963.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.0389931201934814,
|
|
"epoch": 1.5865782932891466,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004924596132283075,
|
|
"loss": 5.8814,
|
|
"mean_token_accuracy": 0.14751892685890197,
|
|
"num_tokens": 3492947.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.1129943370819095,
|
|
"epoch": 1.5907207953603977,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004923773590877135,
|
|
"loss": 5.9607,
|
|
"mean_token_accuracy": 0.13690620362758638,
|
|
"num_tokens": 3501877.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.226007652282715,
|
|
"epoch": 1.5948632974316488,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004922946664732314,
|
|
"loss": 6.0179,
|
|
"mean_token_accuracy": 0.13733150959014892,
|
|
"num_tokens": 3510610.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.226138401031494,
|
|
"epoch": 1.5990057995029,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004922115355516615,
|
|
"loss": 5.987,
|
|
"mean_token_accuracy": 0.1421550303697586,
|
|
"num_tokens": 3519829.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.18082628250122,
|
|
"epoch": 1.6031483015741508,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004921279664906877,
|
|
"loss": 5.9649,
|
|
"mean_token_accuracy": 0.13739833384752273,
|
|
"num_tokens": 3529408.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.077984142303467,
|
|
"epoch": 1.6072908036454017,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004920439594588781,
|
|
"loss": 5.9104,
|
|
"mean_token_accuracy": 0.14634366631507872,
|
|
"num_tokens": 3538068.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.192194318771362,
|
|
"epoch": 1.6114333057166528,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000491959514625684,
|
|
"loss": 6.019,
|
|
"mean_token_accuracy": 0.14332248717546464,
|
|
"num_tokens": 3547302.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.125579690933227,
|
|
"epoch": 1.6155758077879039,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00049187463216144,
|
|
"loss": 5.9024,
|
|
"mean_token_accuracy": 0.14861151427030564,
|
|
"num_tokens": 3556087.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.207472896575927,
|
|
"epoch": 1.619718309859155,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000491789312237363,
|
|
"loss": 5.9691,
|
|
"mean_token_accuracy": 0.14141590148210526,
|
|
"num_tokens": 3566235.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.1809961795806885,
|
|
"epoch": 1.623860811930406,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004917035550255528,
|
|
"loss": 5.9154,
|
|
"mean_token_accuracy": 0.14227381274104117,
|
|
"num_tokens": 3574762.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.051875114440918,
|
|
"epoch": 1.628003314001657,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000491617360698991,
|
|
"loss": 5.8849,
|
|
"mean_token_accuracy": 0.14124636650085448,
|
|
"num_tokens": 3583359.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.050506448745727,
|
|
"epoch": 1.6321458160729079,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004915307294315412,
|
|
"loss": 5.8328,
|
|
"mean_token_accuracy": 0.146518924087286,
|
|
"num_tokens": 3591436.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.101462888717651,
|
|
"epoch": 1.636288318144159,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000491443661397948,
|
|
"loss": 5.9486,
|
|
"mean_token_accuracy": 0.13945667892694474,
|
|
"num_tokens": 3601054.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.16776819229126,
|
|
"epoch": 1.64043082021541,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004913561567738368,
|
|
"loss": 5.9127,
|
|
"mean_token_accuracy": 0.1449202612042427,
|
|
"num_tokens": 3610641.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.071323442459106,
|
|
"epoch": 1.6445733222866612,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004912682157357146,
|
|
"loss": 5.8402,
|
|
"mean_token_accuracy": 0.14677978903055192,
|
|
"num_tokens": 3619351.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.988628768920899,
|
|
"epoch": 1.6487158243579123,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004911798384609675,
|
|
"loss": 5.8414,
|
|
"mean_token_accuracy": 0.14875895008444787,
|
|
"num_tokens": 3627341.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.1566671371459964,
|
|
"epoch": 1.6528583264291632,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004910910251278626,
|
|
"loss": 5.8825,
|
|
"mean_token_accuracy": 0.1572631388902664,
|
|
"num_tokens": 3637132.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.139908933639527,
|
|
"epoch": 1.6570008285004143,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004910017759155458,
|
|
"loss": 5.9317,
|
|
"mean_token_accuracy": 0.14228944033384322,
|
|
"num_tokens": 3646196.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6570008285004143,
|
|
"eval_entropy": 5.874023848250985,
|
|
"eval_loss": 6.080362319946289,
|
|
"eval_mean_token_accuracy": 0.14062138028871785,
|
|
"eval_num_tokens": 3646196.0,
|
|
"eval_runtime": 2.1129,
|
|
"eval_samples_per_second": 1773.361,
|
|
"eval_steps_per_second": 221.966,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.043913221359253,
|
|
"epoch": 1.6611433305716652,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004909120910040427,
|
|
"loss": 5.8727,
|
|
"mean_token_accuracy": 0.1496455654501915,
|
|
"num_tokens": 3654658.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.245485258102417,
|
|
"epoch": 1.6652858326429163,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004908219705742574,
|
|
"loss": 6.1474,
|
|
"mean_token_accuracy": 0.13809763565659522,
|
|
"num_tokens": 3664509.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.0975141525268555,
|
|
"epoch": 1.6694283347141674,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004907314148079727,
|
|
"loss": 5.8596,
|
|
"mean_token_accuracy": 0.143882005661726,
|
|
"num_tokens": 3672983.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.997087240219116,
|
|
"epoch": 1.6735708367854185,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004906404238878496,
|
|
"loss": 5.717,
|
|
"mean_token_accuracy": 0.15030423998832704,
|
|
"num_tokens": 3681196.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.160958909988404,
|
|
"epoch": 1.6777133388566694,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004905489979974265,
|
|
"loss": 5.9034,
|
|
"mean_token_accuracy": 0.14250049591064454,
|
|
"num_tokens": 3690077.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.110355710983276,
|
|
"epoch": 1.6818558409279205,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004904571373211197,
|
|
"loss": 5.9626,
|
|
"mean_token_accuracy": 0.13948154225945472,
|
|
"num_tokens": 3700121.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.085564708709716,
|
|
"epoch": 1.6859983429991714,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000490364842044222,
|
|
"loss": 5.8071,
|
|
"mean_token_accuracy": 0.15038276463747025,
|
|
"num_tokens": 3709618.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.05253415107727,
|
|
"epoch": 1.6901408450704225,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004902721123529031,
|
|
"loss": 6.0018,
|
|
"mean_token_accuracy": 0.13299599811434745,
|
|
"num_tokens": 3719570.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.275170803070068,
|
|
"epoch": 1.6942833471416736,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000490178948434209,
|
|
"loss": 5.9256,
|
|
"mean_token_accuracy": 0.14927142709493638,
|
|
"num_tokens": 3729354.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.055284786224365,
|
|
"epoch": 1.6984258492129247,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004900853504760614,
|
|
"loss": 5.8397,
|
|
"mean_token_accuracy": 0.14695540443062782,
|
|
"num_tokens": 3738453.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.981637811660766,
|
|
"epoch": 1.7025683512841756,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004899913186672577,
|
|
"loss": 5.8159,
|
|
"mean_token_accuracy": 0.14908216297626495,
|
|
"num_tokens": 3747815.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.06216344833374,
|
|
"epoch": 1.7067108533554267,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004898968531974702,
|
|
"loss": 5.8997,
|
|
"mean_token_accuracy": 0.15259757190942763,
|
|
"num_tokens": 3757298.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.112562417984009,
|
|
"epoch": 1.7108533554266776,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004898019542572462,
|
|
"loss": 5.9202,
|
|
"mean_token_accuracy": 0.14373964965343475,
|
|
"num_tokens": 3765939.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.191072845458985,
|
|
"epoch": 1.7149958574979287,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004897066220380073,
|
|
"loss": 5.9279,
|
|
"mean_token_accuracy": 0.1386391043663025,
|
|
"num_tokens": 3775160.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.114138412475586,
|
|
"epoch": 1.7191383595691798,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004896108567320489,
|
|
"loss": 5.8384,
|
|
"mean_token_accuracy": 0.14560278803110122,
|
|
"num_tokens": 3785181.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.127746057510376,
|
|
"epoch": 1.723280861640431,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048951465853254,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.14337964877486228,
|
|
"num_tokens": 3794531.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.036863708496094,
|
|
"epoch": 1.7274233637116818,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004894180276335231,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.14032400846481324,
|
|
"num_tokens": 3804602.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.191616439819336,
|
|
"epoch": 1.731565865782933,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004893209642299132,
|
|
"loss": 5.9663,
|
|
"mean_token_accuracy": 0.14373981803655625,
|
|
"num_tokens": 3813394.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.114638471603394,
|
|
"epoch": 1.7357083678541838,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004892234685174976,
|
|
"loss": 5.967,
|
|
"mean_token_accuracy": 0.1415805034339428,
|
|
"num_tokens": 3821604.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.157201480865479,
|
|
"epoch": 1.739850869925435,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004891255406929359,
|
|
"loss": 5.9503,
|
|
"mean_token_accuracy": 0.1427985519170761,
|
|
"num_tokens": 3831977.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.134490919113159,
|
|
"epoch": 1.743993371996686,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004890271809537595,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.14459130316972732,
|
|
"num_tokens": 3840684.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.99285569190979,
|
|
"epoch": 1.7481358740679371,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004889283894983704,
|
|
"loss": 5.822,
|
|
"mean_token_accuracy": 0.14208590537309645,
|
|
"num_tokens": 3851147.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.158426713943482,
|
|
"epoch": 1.752278376139188,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000488829166526042,
|
|
"loss": 5.9451,
|
|
"mean_token_accuracy": 0.14956211298704147,
|
|
"num_tokens": 3859881.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.1649645328521725,
|
|
"epoch": 1.7564208782104391,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004887295122369176,
|
|
"loss": 6.0039,
|
|
"mean_token_accuracy": 0.14027811884880065,
|
|
"num_tokens": 3870151.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.098885107040405,
|
|
"epoch": 1.76056338028169,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000488629426832011,
|
|
"loss": 5.9401,
|
|
"mean_token_accuracy": 0.14093397855758666,
|
|
"num_tokens": 3878398.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.1110546588897705,
|
|
"epoch": 1.7647058823529411,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004885289105132054,
|
|
"loss": 5.88,
|
|
"mean_token_accuracy": 0.13898311480879783,
|
|
"num_tokens": 3888007.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.088969039916992,
|
|
"epoch": 1.7688483844241922,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004884279634832532,
|
|
"loss": 5.8674,
|
|
"mean_token_accuracy": 0.14867447167634965,
|
|
"num_tokens": 3896945.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.1467766761779785,
|
|
"epoch": 1.7729908864954433,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004883265859457754,
|
|
"loss": 5.8735,
|
|
"mean_token_accuracy": 0.15197214633226394,
|
|
"num_tokens": 3906049.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.028802394866943,
|
|
"epoch": 1.7771333885666944,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048822477810526186,
|
|
"loss": 5.8429,
|
|
"mean_token_accuracy": 0.1435732439160347,
|
|
"num_tokens": 3915050.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.134473705291748,
|
|
"epoch": 1.7812758906379453,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048812254016706994,
|
|
"loss": 5.9224,
|
|
"mean_token_accuracy": 0.1448567934334278,
|
|
"num_tokens": 3924031.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.084311532974243,
|
|
"epoch": 1.7854183927091962,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004880198723374247,
|
|
"loss": 5.9685,
|
|
"mean_token_accuracy": 0.14345090091228485,
|
|
"num_tokens": 3933756.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.121155500411987,
|
|
"epoch": 1.7895608947804473,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00048791677482341845,
|
|
"loss": 5.9416,
|
|
"mean_token_accuracy": 0.13881594836711883,
|
|
"num_tokens": 3943129.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.021676874160766,
|
|
"epoch": 1.7937033968516984,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048781324783301016,
|
|
"loss": 5.782,
|
|
"mean_token_accuracy": 0.1462274558842182,
|
|
"num_tokens": 3951494.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.067917156219482,
|
|
"epoch": 1.7978458989229495,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048770929157502505,
|
|
"loss": 5.9381,
|
|
"mean_token_accuracy": 0.14510203301906585,
|
|
"num_tokens": 3960588.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.183278799057007,
|
|
"epoch": 1.8019884009942007,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004876049062591543,
|
|
"loss": 5.927,
|
|
"mean_token_accuracy": 0.13713768348097802,
|
|
"num_tokens": 3970135.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.1064512729644775,
|
|
"epoch": 1.8061309030654515,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048750009209595443,
|
|
"loss": 5.9004,
|
|
"mean_token_accuracy": 0.1408342719078064,
|
|
"num_tokens": 3979569.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.010239458084106,
|
|
"epoch": 1.8102734051367024,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048739484929684714,
|
|
"loss": 5.8322,
|
|
"mean_token_accuracy": 0.14995085150003434,
|
|
"num_tokens": 3987885.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.10627007484436,
|
|
"epoch": 1.8144159072079535,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048728917807411856,
|
|
"loss": 6.0092,
|
|
"mean_token_accuracy": 0.1335530586540699,
|
|
"num_tokens": 3997726.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.109944820404053,
|
|
"epoch": 1.8185584092792046,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004871830786409191,
|
|
"loss": 5.8324,
|
|
"mean_token_accuracy": 0.1525810420513153,
|
|
"num_tokens": 4006437.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.003955411911011,
|
|
"epoch": 1.8227009113504558,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00048707655121126296,
|
|
"loss": 5.8011,
|
|
"mean_token_accuracy": 0.15133841037750245,
|
|
"num_tokens": 4015326.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.083161973953247,
|
|
"epoch": 1.8268434134217069,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048696959600002767,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.140908819437027,
|
|
"num_tokens": 4024294.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.029014587402344,
|
|
"epoch": 1.8309859154929577,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048686221322295347,
|
|
"loss": 5.8801,
|
|
"mean_token_accuracy": 0.15111605525016786,
|
|
"num_tokens": 4033341.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.027010774612426,
|
|
"epoch": 1.8351284175642086,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048675440309664314,
|
|
"loss": 5.832,
|
|
"mean_token_accuracy": 0.15358244329690934,
|
|
"num_tokens": 4042446.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.998649644851684,
|
|
"epoch": 1.8392709196354597,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048664616583856166,
|
|
"loss": 5.8709,
|
|
"mean_token_accuracy": 0.13802095353603364,
|
|
"num_tokens": 4051849.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.173512172698975,
|
|
"epoch": 1.8434134217067109,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048653750166703516,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.1443096585571766,
|
|
"num_tokens": 4061075.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.132181167602539,
|
|
"epoch": 1.847555923777962,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004864284108012513,
|
|
"loss": 5.9492,
|
|
"mean_token_accuracy": 0.1360410049557686,
|
|
"num_tokens": 4071022.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.038720178604126,
|
|
"epoch": 1.851698425849213,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004863188934612583,
|
|
"loss": 5.8462,
|
|
"mean_token_accuracy": 0.14651346057653428,
|
|
"num_tokens": 4080728.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.01682415008545,
|
|
"epoch": 1.855840927920464,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004862089498679645,
|
|
"loss": 5.8092,
|
|
"mean_token_accuracy": 0.15180521160364152,
|
|
"num_tokens": 4088712.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.045846223831177,
|
|
"epoch": 1.8599834299917148,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004860985802431383,
|
|
"loss": 5.8297,
|
|
"mean_token_accuracy": 0.15167509615421296,
|
|
"num_tokens": 4097488.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.994900369644165,
|
|
"epoch": 1.864125932062966,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004859877848094071,
|
|
"loss": 5.8087,
|
|
"mean_token_accuracy": 0.1456088751554489,
|
|
"num_tokens": 4106485.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.084024572372437,
|
|
"epoch": 1.868268434134217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004858765637902575,
|
|
"loss": 5.7849,
|
|
"mean_token_accuracy": 0.1513776436448097,
|
|
"num_tokens": 4115806.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.1543549537658695,
|
|
"epoch": 1.8724109362054682,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004857649174100344,
|
|
"loss": 5.9323,
|
|
"mean_token_accuracy": 0.1476415552198887,
|
|
"num_tokens": 4124636.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.017945241928101,
|
|
"epoch": 1.8765534382767193,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0004856528458939409,
|
|
"loss": 5.8882,
|
|
"mean_token_accuracy": 0.13946284651756286,
|
|
"num_tokens": 4133876.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.021767807006836,
|
|
"epoch": 1.8806959403479702,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004855403494680373,
|
|
"loss": 5.8053,
|
|
"mean_token_accuracy": 0.1419292762875557,
|
|
"num_tokens": 4143555.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.008573961257935,
|
|
"epoch": 1.884838442419221,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048542742835924133,
|
|
"loss": 5.7626,
|
|
"mean_token_accuracy": 0.14792594760656358,
|
|
"num_tokens": 4153007.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.91074275970459,
|
|
"epoch": 1.8889809444904722,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048531408279532705,
|
|
"loss": 5.8032,
|
|
"mean_token_accuracy": 0.142525102943182,
|
|
"num_tokens": 4162107.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.0140910148620605,
|
|
"epoch": 1.8931234465617233,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048520031300492507,
|
|
"loss": 5.8061,
|
|
"mean_token_accuracy": 0.1468598410487175,
|
|
"num_tokens": 4170920.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.079481410980224,
|
|
"epoch": 1.8972659486329744,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00048508611921752124,
|
|
"loss": 5.8315,
|
|
"mean_token_accuracy": 0.14988012462854386,
|
|
"num_tokens": 4180226.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.958925056457519,
|
|
"epoch": 1.9014084507042255,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048497150166345703,
|
|
"loss": 5.7554,
|
|
"mean_token_accuracy": 0.14854799956083298,
|
|
"num_tokens": 4189283.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.96260290145874,
|
|
"epoch": 1.9055509527754764,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004848564605739287,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.14435628652572632,
|
|
"num_tokens": 4199017.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.1000776290893555,
|
|
"epoch": 1.9096934548467275,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048474099618098644,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.14426328167319297,
|
|
"num_tokens": 4207756.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.022288465499878,
|
|
"epoch": 1.9138359569179784,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004846251087175348,
|
|
"loss": 5.8266,
|
|
"mean_token_accuracy": 0.1475055083632469,
|
|
"num_tokens": 4216721.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.032358646392822,
|
|
"epoch": 1.9179784589892295,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004845087984173312,
|
|
"loss": 5.8168,
|
|
"mean_token_accuracy": 0.15082903131842612,
|
|
"num_tokens": 4225596.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.084334421157837,
|
|
"epoch": 1.9221209610604806,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048439206551498644,
|
|
"loss": 5.849,
|
|
"mean_token_accuracy": 0.1496099904179573,
|
|
"num_tokens": 4234500.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.992849826812744,
|
|
"epoch": 1.9262634631317317,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004842749102459635,
|
|
"loss": 5.8144,
|
|
"mean_token_accuracy": 0.14916087314486504,
|
|
"num_tokens": 4243240.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.945455312728882,
|
|
"epoch": 1.9304059652029826,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00048415733284657727,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.15595563054084777,
|
|
"num_tokens": 4251892.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.017211198806763,
|
|
"epoch": 1.9345484672742337,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004840393335539943,
|
|
"loss": 5.7671,
|
|
"mean_token_accuracy": 0.15037013292312623,
|
|
"num_tokens": 4261447.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.039053583145142,
|
|
"epoch": 1.9386909693454846,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048392091260623195,
|
|
"loss": 5.9215,
|
|
"mean_token_accuracy": 0.14951857775449753,
|
|
"num_tokens": 4270120.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.031311130523681,
|
|
"epoch": 1.9428334714167357,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048380207024215836,
|
|
"loss": 5.7906,
|
|
"mean_token_accuracy": 0.14514822512865067,
|
|
"num_tokens": 4278300.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.057702875137329,
|
|
"epoch": 1.9469759734879868,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00048368280670149145,
|
|
"loss": 5.8306,
|
|
"mean_token_accuracy": 0.14935216754674913,
|
|
"num_tokens": 4286454.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.917616987228394,
|
|
"epoch": 1.951118475559238,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004835631222247989,
|
|
"loss": 5.71,
|
|
"mean_token_accuracy": 0.1527955412864685,
|
|
"num_tokens": 4294752.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.98473014831543,
|
|
"epoch": 1.9552609776304888,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048344301705349735,
|
|
"loss": 5.7704,
|
|
"mean_token_accuracy": 0.15101421475410462,
|
|
"num_tokens": 4303195.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.040767526626587,
|
|
"epoch": 1.95940347970174,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048332249142985195,
|
|
"loss": 5.9161,
|
|
"mean_token_accuracy": 0.1434949241578579,
|
|
"num_tokens": 4311961.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.024049615859985,
|
|
"epoch": 1.9635459817729908,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004832015455969763,
|
|
"loss": 5.7595,
|
|
"mean_token_accuracy": 0.14508079290390014,
|
|
"num_tokens": 4322267.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.939272403717041,
|
|
"epoch": 1.9676884838442419,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048308017979883114,
|
|
"loss": 5.8657,
|
|
"mean_token_accuracy": 0.1457856371998787,
|
|
"num_tokens": 4330584.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.925167560577393,
|
|
"epoch": 1.971830985915493,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048295839428022454,
|
|
"loss": 5.7161,
|
|
"mean_token_accuracy": 0.14976125806570054,
|
|
"num_tokens": 4339215.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.032935953140258,
|
|
"epoch": 1.975973487986744,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048283618928681127,
|
|
"loss": 5.7382,
|
|
"mean_token_accuracy": 0.15662530511617662,
|
|
"num_tokens": 4347711.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.927608680725098,
|
|
"epoch": 1.980115990057995,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004827135650650923,
|
|
"loss": 5.732,
|
|
"mean_token_accuracy": 0.15485673546791076,
|
|
"num_tokens": 4356982.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.8668547630310055,
|
|
"epoch": 1.984258492129246,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004825905218624139,
|
|
"loss": 5.7539,
|
|
"mean_token_accuracy": 0.1500335916876793,
|
|
"num_tokens": 4365378.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.162260103225708,
|
|
"epoch": 1.988400994200497,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048246705992696756,
|
|
"loss": 5.8111,
|
|
"mean_token_accuracy": 0.140972301363945,
|
|
"num_tokens": 4374007.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.086604118347168,
|
|
"epoch": 1.992543496271748,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048234317950778975,
|
|
"loss": 5.906,
|
|
"mean_token_accuracy": 0.1491197317838669,
|
|
"num_tokens": 4383909.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.080717754364014,
|
|
"epoch": 1.9966859983429992,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004822188808547607,
|
|
"loss": 5.9774,
|
|
"mean_token_accuracy": 0.1428071916103363,
|
|
"num_tokens": 4393145.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.962521982192993,
|
|
"epoch": 2.0008285004142503,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004820941642186043,
|
|
"loss": 5.692,
|
|
"mean_token_accuracy": 0.14557612016797067,
|
|
"num_tokens": 4402024.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.005666828155517,
|
|
"epoch": 2.0049710024855014,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004819690298508879,
|
|
"loss": 5.5709,
|
|
"mean_token_accuracy": 0.1672823339700699,
|
|
"num_tokens": 4411377.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.966760778427124,
|
|
"epoch": 2.009113504556752,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048184347800402083,
|
|
"loss": 5.4526,
|
|
"mean_token_accuracy": 0.17516808956861496,
|
|
"num_tokens": 4419781.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.804680967330933,
|
|
"epoch": 2.013256006628003,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004817175089312551,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.17022604495286942,
|
|
"num_tokens": 4428427.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.961127233505249,
|
|
"epoch": 2.0173985086992543,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048159112288668403,
|
|
"loss": 5.4711,
|
|
"mean_token_accuracy": 0.1661138951778412,
|
|
"num_tokens": 4438156.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.000173330307007,
|
|
"epoch": 2.0215410107705054,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048146432012524215,
|
|
"loss": 5.4878,
|
|
"mean_token_accuracy": 0.17110148221254348,
|
|
"num_tokens": 4446342.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.898652982711792,
|
|
"epoch": 2.0256835128417565,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048133710090270446,
|
|
"loss": 5.5409,
|
|
"mean_token_accuracy": 0.16039638221263885,
|
|
"num_tokens": 4455582.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.023085260391236,
|
|
"epoch": 2.0298260149130076,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048120946547568596,
|
|
"loss": 5.5748,
|
|
"mean_token_accuracy": 0.1651100292801857,
|
|
"num_tokens": 4464518.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.046728134155273,
|
|
"epoch": 2.0339685169842583,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00048108141410164133,
|
|
"loss": 5.6752,
|
|
"mean_token_accuracy": 0.1500121273100376,
|
|
"num_tokens": 4475031.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.013257789611816,
|
|
"epoch": 2.0381110190555094,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004809529470388642,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.16635677367448806,
|
|
"num_tokens": 4484609.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.94019513130188,
|
|
"epoch": 2.0422535211267605,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004808240645464866,
|
|
"loss": 5.573,
|
|
"mean_token_accuracy": 0.1624813809990883,
|
|
"num_tokens": 4494238.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.855140542984008,
|
|
"epoch": 2.0463960231980116,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004806947668844787,
|
|
"loss": 5.4625,
|
|
"mean_token_accuracy": 0.17260214388370515,
|
|
"num_tokens": 4503258.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.941313695907593,
|
|
"epoch": 2.0505385252692627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048056505431364794,
|
|
"loss": 5.5238,
|
|
"mean_token_accuracy": 0.16605687141418457,
|
|
"num_tokens": 4512071.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.907178926467895,
|
|
"epoch": 2.054681027340514,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048043492709563877,
|
|
"loss": 5.3627,
|
|
"mean_token_accuracy": 0.17821477800607682,
|
|
"num_tokens": 4520219.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.868778133392334,
|
|
"epoch": 2.0588235294117645,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048030438549293193,
|
|
"loss": 5.5781,
|
|
"mean_token_accuracy": 0.16587240397930145,
|
|
"num_tokens": 4529386.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.954671621322632,
|
|
"epoch": 2.0629660314830156,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048017342976884424,
|
|
"loss": 5.5779,
|
|
"mean_token_accuracy": 0.1579540751874447,
|
|
"num_tokens": 4538860.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.901145601272583,
|
|
"epoch": 2.0671085335542667,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048004206018752773,
|
|
"loss": 5.5405,
|
|
"mean_token_accuracy": 0.1605120435357094,
|
|
"num_tokens": 4547842.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.86530909538269,
|
|
"epoch": 2.071251035625518,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047991027701396895,
|
|
"loss": 5.532,
|
|
"mean_token_accuracy": 0.16047001928091048,
|
|
"num_tokens": 4556238.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.071251035625518,
|
|
"eval_entropy": 5.700431426196719,
|
|
"eval_loss": 5.979622840881348,
|
|
"eval_mean_token_accuracy": 0.14706116486618767,
|
|
"eval_num_tokens": 4556238.0,
|
|
"eval_runtime": 2.0999,
|
|
"eval_samples_per_second": 1784.392,
|
|
"eval_steps_per_second": 223.347,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.91064567565918,
|
|
"epoch": 2.075393537696769,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004797780805139892,
|
|
"loss": 5.5194,
|
|
"mean_token_accuracy": 0.16701295971870422,
|
|
"num_tokens": 4565684.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.8715479373931885,
|
|
"epoch": 2.07953603976802,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004796454709542432,
|
|
"loss": 5.4743,
|
|
"mean_token_accuracy": 0.1701912671327591,
|
|
"num_tokens": 4575001.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.894387054443359,
|
|
"epoch": 2.083678541839271,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004795124486022189,
|
|
"loss": 5.5998,
|
|
"mean_token_accuracy": 0.1611030802130699,
|
|
"num_tokens": 4584128.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.919654273986817,
|
|
"epoch": 2.087821043910522,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047937901372623694,
|
|
"loss": 5.4421,
|
|
"mean_token_accuracy": 0.16908973455429077,
|
|
"num_tokens": 4593011.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.818285036087036,
|
|
"epoch": 2.091963545981773,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047924516659545005,
|
|
"loss": 5.498,
|
|
"mean_token_accuracy": 0.16083064973354338,
|
|
"num_tokens": 4602676.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.995373773574829,
|
|
"epoch": 2.096106048053024,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004791109074798425,
|
|
"loss": 5.5647,
|
|
"mean_token_accuracy": 0.16249934434890748,
|
|
"num_tokens": 4611915.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.904282283782959,
|
|
"epoch": 2.100248550124275,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004789762366502296,
|
|
"loss": 5.5341,
|
|
"mean_token_accuracy": 0.165102981030941,
|
|
"num_tokens": 4621396.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.916229724884033,
|
|
"epoch": 2.1043910521955262,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004788411543782571,
|
|
"loss": 5.5333,
|
|
"mean_token_accuracy": 0.1635112926363945,
|
|
"num_tokens": 4630533.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.833063125610352,
|
|
"epoch": 2.108533554266777,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004787056609364008,
|
|
"loss": 5.4904,
|
|
"mean_token_accuracy": 0.16558443307876586,
|
|
"num_tokens": 4639023.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.932992315292358,
|
|
"epoch": 2.112676056338028,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004785697565979657,
|
|
"loss": 5.4844,
|
|
"mean_token_accuracy": 0.16980037093162537,
|
|
"num_tokens": 4647975.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.849015378952027,
|
|
"epoch": 2.116818558409279,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004784334416370857,
|
|
"loss": 5.5125,
|
|
"mean_token_accuracy": 0.15812072902917862,
|
|
"num_tokens": 4657555.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.860912609100342,
|
|
"epoch": 2.1209610604805302,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004782967163287231,
|
|
"loss": 5.4232,
|
|
"mean_token_accuracy": 0.1693859726190567,
|
|
"num_tokens": 4667228.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.854128408432007,
|
|
"epoch": 2.1251035625517813,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047815958094866756,
|
|
"loss": 5.5068,
|
|
"mean_token_accuracy": 0.16066872626543044,
|
|
"num_tokens": 4676723.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.834451913833618,
|
|
"epoch": 2.1292460646230325,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004780220357735363,
|
|
"loss": 5.4848,
|
|
"mean_token_accuracy": 0.16301569640636443,
|
|
"num_tokens": 4686241.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.884823274612427,
|
|
"epoch": 2.1333885666942836,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047788408108077294,
|
|
"loss": 5.5521,
|
|
"mean_token_accuracy": 0.1669829785823822,
|
|
"num_tokens": 4695195.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.897109127044677,
|
|
"epoch": 2.1375310687655342,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004777457171486471,
|
|
"loss": 5.5862,
|
|
"mean_token_accuracy": 0.1627011328935623,
|
|
"num_tokens": 4704716.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.881407833099365,
|
|
"epoch": 2.1416735708367853,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004776069442562541,
|
|
"loss": 5.5701,
|
|
"mean_token_accuracy": 0.15754837542772293,
|
|
"num_tokens": 4713720.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.888014125823974,
|
|
"epoch": 2.1458160729080364,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004774677626835138,
|
|
"loss": 5.5318,
|
|
"mean_token_accuracy": 0.1549442321062088,
|
|
"num_tokens": 4722428.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.978774785995483,
|
|
"epoch": 2.1499585749792876,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047732817271117084,
|
|
"loss": 5.536,
|
|
"mean_token_accuracy": 0.15985957831144332,
|
|
"num_tokens": 4732395.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.876013517379761,
|
|
"epoch": 2.1541010770505387,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004771881746207934,
|
|
"loss": 5.529,
|
|
"mean_token_accuracy": 0.16375088393688203,
|
|
"num_tokens": 4741071.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.871488332748413,
|
|
"epoch": 2.1582435791217898,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047704776869477285,
|
|
"loss": 5.5259,
|
|
"mean_token_accuracy": 0.15932013094425201,
|
|
"num_tokens": 4749876.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.908699417114258,
|
|
"epoch": 2.1623860811930404,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047690695521632344,
|
|
"loss": 5.4755,
|
|
"mean_token_accuracy": 0.16755755990743637,
|
|
"num_tokens": 4758721.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.8012388229370115,
|
|
"epoch": 2.1665285832642915,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00047676573446948114,
|
|
"loss": 5.5038,
|
|
"mean_token_accuracy": 0.16403038650751114,
|
|
"num_tokens": 4767162.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.957990026473999,
|
|
"epoch": 2.1706710853355426,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047662410673910376,
|
|
"loss": 5.5235,
|
|
"mean_token_accuracy": 0.17056044489145278,
|
|
"num_tokens": 4776397.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.808328533172608,
|
|
"epoch": 2.1748135874067938,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047648207231087,
|
|
"loss": 5.4696,
|
|
"mean_token_accuracy": 0.16226063668727875,
|
|
"num_tokens": 4784822.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.8803400039672855,
|
|
"epoch": 2.178956089478045,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004763396314712786,
|
|
"loss": 5.5386,
|
|
"mean_token_accuracy": 0.164871746301651,
|
|
"num_tokens": 4794272.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.7970967292785645,
|
|
"epoch": 2.183098591549296,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004761967845076484,
|
|
"loss": 5.4707,
|
|
"mean_token_accuracy": 0.1630651980638504,
|
|
"num_tokens": 4804314.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.8478696823120115,
|
|
"epoch": 2.1872410936205466,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004760535317081174,
|
|
"loss": 5.4786,
|
|
"mean_token_accuracy": 0.163403283059597,
|
|
"num_tokens": 4812685.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.864224481582641,
|
|
"epoch": 2.1913835956917977,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047590987336164207,
|
|
"loss": 5.552,
|
|
"mean_token_accuracy": 0.16005686074495315,
|
|
"num_tokens": 4821761.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.890404462814331,
|
|
"epoch": 2.195526097763049,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047576580975799705,
|
|
"loss": 5.5598,
|
|
"mean_token_accuracy": 0.1633346602320671,
|
|
"num_tokens": 4830515.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.911027812957764,
|
|
"epoch": 2.1996685998343,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004756213411877743,
|
|
"loss": 5.5057,
|
|
"mean_token_accuracy": 0.16152726709842682,
|
|
"num_tokens": 4840741.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.786312103271484,
|
|
"epoch": 2.203811101905551,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047547646794238277,
|
|
"loss": 5.4645,
|
|
"mean_token_accuracy": 0.17090066373348237,
|
|
"num_tokens": 4849571.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.782705068588257,
|
|
"epoch": 2.207953603976802,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047533119031404757,
|
|
"loss": 5.4238,
|
|
"mean_token_accuracy": 0.1744443506002426,
|
|
"num_tokens": 4857194.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.780950593948364,
|
|
"epoch": 2.212096106048053,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047518550859580966,
|
|
"loss": 5.5164,
|
|
"mean_token_accuracy": 0.16455501317977905,
|
|
"num_tokens": 4865894.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.885499715805054,
|
|
"epoch": 2.216238608119304,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004750394230815249,
|
|
"loss": 5.5244,
|
|
"mean_token_accuracy": 0.16542572677135467,
|
|
"num_tokens": 4875131.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.795029067993164,
|
|
"epoch": 2.220381110190555,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047489293406586375,
|
|
"loss": 5.3891,
|
|
"mean_token_accuracy": 0.17154104858636857,
|
|
"num_tokens": 4884098.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.7773130416870115,
|
|
"epoch": 2.224523612261806,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004747460418443106,
|
|
"loss": 5.4498,
|
|
"mean_token_accuracy": 0.16123902201652526,
|
|
"num_tokens": 4892963.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.887063694000244,
|
|
"epoch": 2.2286661143330573,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047459874671316305,
|
|
"loss": 5.5169,
|
|
"mean_token_accuracy": 0.16128448247909546,
|
|
"num_tokens": 4902097.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.751895809173584,
|
|
"epoch": 2.2328086164043084,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047445104896953153,
|
|
"loss": 5.4179,
|
|
"mean_token_accuracy": 0.17486145794391633,
|
|
"num_tokens": 4910469.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.011729669570923,
|
|
"epoch": 2.236951118475559,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047430294891133846,
|
|
"loss": 5.693,
|
|
"mean_token_accuracy": 0.1579521968960762,
|
|
"num_tokens": 4921024.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.897257661819458,
|
|
"epoch": 2.24109362054681,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004741544468373179,
|
|
"loss": 5.5352,
|
|
"mean_token_accuracy": 0.1622954562306404,
|
|
"num_tokens": 4930297.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.858204364776611,
|
|
"epoch": 2.2452361226180613,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047400554304701474,
|
|
"loss": 5.4703,
|
|
"mean_token_accuracy": 0.16952212154865265,
|
|
"num_tokens": 4938971.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.736092233657837,
|
|
"epoch": 2.2493786246893124,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047385623784078427,
|
|
"loss": 5.4983,
|
|
"mean_token_accuracy": 0.1607481509447098,
|
|
"num_tokens": 4948531.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.897338581085205,
|
|
"epoch": 2.2535211267605635,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004737065315197913,
|
|
"loss": 5.5588,
|
|
"mean_token_accuracy": 0.1676310196518898,
|
|
"num_tokens": 4957430.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.8622315406799315,
|
|
"epoch": 2.2576636288318146,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047355642438600987,
|
|
"loss": 5.5583,
|
|
"mean_token_accuracy": 0.1619584560394287,
|
|
"num_tokens": 4966681.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.813560342788696,
|
|
"epoch": 2.2618061309030653,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004734059167422225,
|
|
"loss": 5.4184,
|
|
"mean_token_accuracy": 0.17226676046848297,
|
|
"num_tokens": 4975327.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.728217124938965,
|
|
"epoch": 2.2659486329743164,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047325500889201956,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.16976486146450043,
|
|
"num_tokens": 4983741.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.841625738143921,
|
|
"epoch": 2.2700911350455675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004731037011397986,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.16324040293693542,
|
|
"num_tokens": 4993171.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.934147357940674,
|
|
"epoch": 2.2742336371168186,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047295199379076403,
|
|
"loss": 5.61,
|
|
"mean_token_accuracy": 0.1584983289241791,
|
|
"num_tokens": 5002810.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.811686086654663,
|
|
"epoch": 2.2783761391880697,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000472799887150926,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.16332145929336547,
|
|
"num_tokens": 5011875.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.856362628936767,
|
|
"epoch": 2.282518641259321,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004726473815271003,
|
|
"loss": 5.5263,
|
|
"mean_token_accuracy": 0.16980284899473191,
|
|
"num_tokens": 5020780.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.806266975402832,
|
|
"epoch": 2.286661143330572,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047249447722690747,
|
|
"loss": 5.4321,
|
|
"mean_token_accuracy": 0.1647959068417549,
|
|
"num_tokens": 5029896.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.861921787261963,
|
|
"epoch": 2.2908036454018226,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047234117455877215,
|
|
"loss": 5.5179,
|
|
"mean_token_accuracy": 0.16179582476615906,
|
|
"num_tokens": 5038956.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.813634967803955,
|
|
"epoch": 2.2949461474730737,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004721874738319225,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.17373612970113755,
|
|
"num_tokens": 5047005.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.804528903961182,
|
|
"epoch": 2.299088649544325,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004720333753563898,
|
|
"loss": 5.526,
|
|
"mean_token_accuracy": 0.15704508125782013,
|
|
"num_tokens": 5056475.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.815530300140381,
|
|
"epoch": 2.303231151615576,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047187887944300747,
|
|
"loss": 5.5013,
|
|
"mean_token_accuracy": 0.1687375158071518,
|
|
"num_tokens": 5064978.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.865201950073242,
|
|
"epoch": 2.307373653686827,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004717239864034107,
|
|
"loss": 5.5812,
|
|
"mean_token_accuracy": 0.16235170215368272,
|
|
"num_tokens": 5074235.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.82843165397644,
|
|
"epoch": 2.3115161557580777,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047156869655003554,
|
|
"loss": 5.4408,
|
|
"mean_token_accuracy": 0.16791896969079972,
|
|
"num_tokens": 5082386.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.766846418380737,
|
|
"epoch": 2.315658657829329,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047141301019611875,
|
|
"loss": 5.4334,
|
|
"mean_token_accuracy": 0.1712301880121231,
|
|
"num_tokens": 5090634.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.967849349975586,
|
|
"epoch": 2.31980115990058,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004712569276556967,
|
|
"loss": 5.5878,
|
|
"mean_token_accuracy": 0.16059968620538712,
|
|
"num_tokens": 5099003.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.787600946426392,
|
|
"epoch": 2.323943661971831,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047110044924360484,
|
|
"loss": 5.6732,
|
|
"mean_token_accuracy": 0.15896357521414756,
|
|
"num_tokens": 5108631.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.796855211257935,
|
|
"epoch": 2.328086164043082,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047094357527547725,
|
|
"loss": 5.4452,
|
|
"mean_token_accuracy": 0.166446553170681,
|
|
"num_tokens": 5117309.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.886296510696411,
|
|
"epoch": 2.332228666114333,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004707863060677459,
|
|
"loss": 5.5608,
|
|
"mean_token_accuracy": 0.1704049676656723,
|
|
"num_tokens": 5126305.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.948070526123047,
|
|
"epoch": 2.3363711681855843,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047062864193764,
|
|
"loss": 5.626,
|
|
"mean_token_accuracy": 0.1627499058842659,
|
|
"num_tokens": 5135338.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.821083450317383,
|
|
"epoch": 2.340513670256835,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047047058320318525,
|
|
"loss": 5.5402,
|
|
"mean_token_accuracy": 0.1675426945090294,
|
|
"num_tokens": 5144518.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.848140001296997,
|
|
"epoch": 2.344656172328086,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004703121301832034,
|
|
"loss": 5.6316,
|
|
"mean_token_accuracy": 0.15992611795663833,
|
|
"num_tokens": 5154102.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.895147371292114,
|
|
"epoch": 2.348798674399337,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047015328319731153,
|
|
"loss": 5.577,
|
|
"mean_token_accuracy": 0.15829345881938933,
|
|
"num_tokens": 5163719.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.83940577507019,
|
|
"epoch": 2.3529411764705883,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004699940425659215,
|
|
"loss": 5.533,
|
|
"mean_token_accuracy": 0.1642977774143219,
|
|
"num_tokens": 5173192.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.75920729637146,
|
|
"epoch": 2.3570836785418394,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046983440861023877,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.17017790228128432,
|
|
"num_tokens": 5182438.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.852911281585693,
|
|
"epoch": 2.36122618061309,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046967438165226275,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.15865150541067125,
|
|
"num_tokens": 5191595.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.7768883228302,
|
|
"epoch": 2.365368682684341,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004695139620147851,
|
|
"loss": 5.4778,
|
|
"mean_token_accuracy": 0.17112626433372496,
|
|
"num_tokens": 5200174.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.891972255706787,
|
|
"epoch": 2.3695111847555923,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046935315002138994,
|
|
"loss": 5.5743,
|
|
"mean_token_accuracy": 0.16057804375886917,
|
|
"num_tokens": 5210454.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.828851795196533,
|
|
"epoch": 2.3736536868268434,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004691919459964525,
|
|
"loss": 5.5133,
|
|
"mean_token_accuracy": 0.16855535954236983,
|
|
"num_tokens": 5219243.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.6906946182250975,
|
|
"epoch": 2.3777961888980945,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046903035026513894,
|
|
"loss": 5.4128,
|
|
"mean_token_accuracy": 0.1679743379354477,
|
|
"num_tokens": 5228656.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.81590838432312,
|
|
"epoch": 2.3819386909693456,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004688683631534056,
|
|
"loss": 5.493,
|
|
"mean_token_accuracy": 0.1606264054775238,
|
|
"num_tokens": 5237807.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.93833179473877,
|
|
"epoch": 2.3860811930405967,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004687059849879981,
|
|
"loss": 5.5967,
|
|
"mean_token_accuracy": 0.16789772659540175,
|
|
"num_tokens": 5247432.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.769515132904052,
|
|
"epoch": 2.3902236951118474,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046854321609645093,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.1679755300283432,
|
|
"num_tokens": 5256680.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.781009340286255,
|
|
"epoch": 2.3943661971830985,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 0.00046838005680708677,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.15484578013420106,
|
|
"num_tokens": 5266793.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.785918760299682,
|
|
"epoch": 2.3985086992543496,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046821650744901573,
|
|
"loss": 5.4506,
|
|
"mean_token_accuracy": 0.16751299500465394,
|
|
"num_tokens": 5275028.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.724029159545898,
|
|
"epoch": 2.4026512013256007,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004680525683521347,
|
|
"loss": 5.511,
|
|
"mean_token_accuracy": 0.16009988784790039,
|
|
"num_tokens": 5284162.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.938899421691895,
|
|
"epoch": 2.406793703396852,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004678882398471269,
|
|
"loss": 5.5904,
|
|
"mean_token_accuracy": 0.15319840013980865,
|
|
"num_tokens": 5293263.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.795382261276245,
|
|
"epoch": 2.4109362054681025,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046772352226546066,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.17464249432086945,
|
|
"num_tokens": 5301373.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.81425256729126,
|
|
"epoch": 2.4150787075393536,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004675584159393895,
|
|
"loss": 5.5178,
|
|
"mean_token_accuracy": 0.16518327295780183,
|
|
"num_tokens": 5311009.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.85831036567688,
|
|
"epoch": 2.4192212096106047,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046739292120195083,
|
|
"loss": 5.5199,
|
|
"mean_token_accuracy": 0.16533831655979156,
|
|
"num_tokens": 5320341.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.816470098495484,
|
|
"epoch": 2.423363711681856,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004672270383869657,
|
|
"loss": 5.4838,
|
|
"mean_token_accuracy": 0.16495863646268843,
|
|
"num_tokens": 5329250.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.707045841217041,
|
|
"epoch": 2.427506213753107,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004670607678290379,
|
|
"loss": 5.4653,
|
|
"mean_token_accuracy": 0.1688676357269287,
|
|
"num_tokens": 5336515.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.742906761169434,
|
|
"epoch": 2.431648715824358,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004668941098635532,
|
|
"loss": 5.5376,
|
|
"mean_token_accuracy": 0.16896287947893143,
|
|
"num_tokens": 5346043.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.862775087356567,
|
|
"epoch": 2.435791217895609,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046672706482667897,
|
|
"loss": 5.4847,
|
|
"mean_token_accuracy": 0.16607862263917922,
|
|
"num_tokens": 5355124.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.763607597351074,
|
|
"epoch": 2.43993371996686,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004665596330553633,
|
|
"loss": 5.5103,
|
|
"mean_token_accuracy": 0.16795327365398408,
|
|
"num_tokens": 5364556.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.820324087142945,
|
|
"epoch": 2.444076222038111,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046639181488733445,
|
|
"loss": 5.5533,
|
|
"mean_token_accuracy": 0.1670700043439865,
|
|
"num_tokens": 5373778.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.827516508102417,
|
|
"epoch": 2.448218724109362,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004662236106610999,
|
|
"loss": 5.5201,
|
|
"mean_token_accuracy": 0.16433579325675965,
|
|
"num_tokens": 5382633.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.73263373374939,
|
|
"epoch": 2.452361226180613,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046605502071594604,
|
|
"loss": 5.4238,
|
|
"mean_token_accuracy": 0.16572020649909974,
|
|
"num_tokens": 5391442.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.729014730453491,
|
|
"epoch": 2.4565037282518642,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004658860453919371,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.1716168373823166,
|
|
"num_tokens": 5400504.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.748004531860351,
|
|
"epoch": 2.460646230323115,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004657166850299149,
|
|
"loss": 5.4487,
|
|
"mean_token_accuracy": 0.17349344938993455,
|
|
"num_tokens": 5408939.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.76081337928772,
|
|
"epoch": 2.464788732394366,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004655469399714978,
|
|
"loss": 5.4732,
|
|
"mean_token_accuracy": 0.1633571669459343,
|
|
"num_tokens": 5418934.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.719039440155029,
|
|
"epoch": 2.468931234465617,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004653768105590799,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.16778907030820847,
|
|
"num_tokens": 5427894.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.808207082748413,
|
|
"epoch": 2.4730737365368682,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046520629713583115,
|
|
"loss": 5.4731,
|
|
"mean_token_accuracy": 0.1672350972890854,
|
|
"num_tokens": 5436830.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.713041210174561,
|
|
"epoch": 2.4772162386081193,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046503540004569555,
|
|
"loss": 5.4592,
|
|
"mean_token_accuracy": 0.1743247002363205,
|
|
"num_tokens": 5445504.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.789522886276245,
|
|
"epoch": 2.4813587406793705,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046486411963339125,
|
|
"loss": 5.5394,
|
|
"mean_token_accuracy": 0.15771088749170303,
|
|
"num_tokens": 5455378.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.846792459487915,
|
|
"epoch": 2.4855012427506216,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046469245624440955,
|
|
"loss": 5.5936,
|
|
"mean_token_accuracy": 0.16458921432495116,
|
|
"num_tokens": 5465334.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.4855012427506216,
|
|
"eval_entropy": 5.656050709519051,
|
|
"eval_loss": 5.90228796005249,
|
|
"eval_mean_token_accuracy": 0.15245503703477795,
|
|
"eval_num_tokens": 5465334.0,
|
|
"eval_runtime": 2.1213,
|
|
"eval_samples_per_second": 1766.397,
|
|
"eval_steps_per_second": 221.094,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"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": 1208016463841280.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|